iy son » Cire dime eecadgeh kl het i i ens BHM Pagegsitecs Aes ae 4 } bares & Vata resee Vath ts st mk eee: 23 i | ste | seat | eres 13 AA 4 hetero 8 tot aa Nears Mele! ye nt ; : one tLe ara eat NZ Got oyigiae Gut oA ths Ae eee’ he hat RELSCYS,, Bye: ssa hing’ the Se eNeost ks ; | : 2 Newnan : ‘ Leyva : | : | ener 7 fhe esl & ea NN Ae RES voaeken' i com Nase Seether : { Hina : Vesta Sv ine ‘ Sees ste (EEO Y Aba STS A ye Se se Be CREATAS OE adem ahs rarer en! Fae ake yepue qu arena Tepes : Gra Meet e CARS Line eae pn u tes : : Dein as eae aenee i ta : thy sue vite Apes RR EEOE Vr Size pup sen ; ‘ sf iSeiafe Ra tie tren Rae oe Hee eT Pty aay net Lae Lad ¢ cy Being aay oe ice) ‘ Fey aie : j ot : : cya : ; : 1 < (Sie tain ty Mh peas oh ei rete EE reget! peg Conc yn eis Sr erkon Do Ced ragh Tete yeas COUN UNIO 4 Bitte ath Py bor dows ¥ Seka a Say 197 “Ty ha ynecin! f, ne peter ee . | ieee a nln Pete é Tt ae LE Hearse ate ise tp tN “weit Beals ene ay, " t { een is sera ; ea tty Bt Nrive se ; i z ) Tas at i : a yey eevee wR he ap ior rye sido Seine sere sypcoaye oe Oe NEw Renee Cpa ees chyumecn yan WS nS ae open et ete sya esta” i Ue spta pee sh i - ZOOLOGICAL SCIENCE An International Journal VOLUME 9 Og, published by The Zoological Society of Japan CONTENTS OF VOLUME 9 NUMBER 1, FEBRUARY 1992 REVIEWS Morita, H.: Transduction process and im- pulse initiation in insect contact chemorecep- CO EE fe cfd Rio coh esa esol ait covnln ie gouyiviee cia os 1 Ugrumov, M. V.: Development of the hypothalamic monoaminergic system in on- togenesis. Morpho-functional aspects ...17 ORIGINAL PAPERS Physiology Okumura, T., C. Han, Y. Suzuki, K. Aida and I. Hanyu: Changes in hemolymph vitel- logenin and ecdysteroid levels during the reproductive and non-reproductive molt cy- cles in the freshwater prawn Macrobrachium UD DOMENS CIMA AA whe, CRED, Vo eh) 31) Azuma, K. and N. Iwasaki: Effects of Ba?* on the photoresponses of isolated single rods POMMUOC MEUM A: - rnin canara sd oo etes ol ees 47 Hirata, J. and H. Michibata: Electron spin resonance spectrometry of vanadium ions in the blood cells of the ascidian, Ascidia gem- mata (COMMUNICATION) Hara, T., R. Hara, A. Kishigami, Y. Koshida, S. Horiuchi, M. Yoshida, M. Yamamoto, T. Goto and U. Raj: Localization of re- tinochrome in the retina of a tetrabranchiate cephalopod, Nautilus pompilius (COM- MUNICATION) Yasuyama, K., T. Kimura and T. Yamaguchi: Proctolin-like immunoreactivity in the dorsal unpaired median neurons innervating the accessory gland of the male cricket, Gryllus bimaculatus Cell Biology Nagaishi, H. and N. Oshima: Ultrastructure of the motile iridophores of the neon tetra Sato, I.: Origin of multinucleality of muscle cells during myogenesis of the newt Yano, J.andM.Suhama: Effect of actinomy- cin D on nuclear events during conjugation in the ciliate Stylonychia pustulata Genetics Kosaka, T.: Autogamy and autogamy inheri- tance in Euplotes woodruffi syngen 1 (Ciliophora) Immunology Zhang, H., Z. Huang, K. Yamaguchi and S. Tomonaga: Granulocytes and macropha- ges in amphioxus Biochemistry Iino, T., K. Dohke and M. Tsusue: The Purification and characterization of sepiap- terin reductase from fat body of the silkworm Bombyx mori Nakauchi, U. and K. Maruyama: Immunob- lot detection of vertebrate-type of connectin (titin) in ascidian bodywall muscle and tad- pole (COMMUNICATION) Developmental Biology Satomi, D.: Developmental changes of gluta- mate decarboxylase and 2’,3-cyclic nuc- leotide 3’-phosphodiesterase in the organ- otypic culture of newborn mouse cerebellum Watanabe, M.: Egg maturation in labora- tory-reared females of the swallowtail but- terfly, Papilio xuthus L. (Lepidoptera: Papi- lionidae), feeding on different concentration solutions of sugar Endocrinology Yada, T. and T. Hirano: Influence of seawa- ter adaptation on prolactin and growth hor- mone release from organ-cultured pituitary of rainbow trout Fukuda, M., R. Hashimoto, K. Yamanouchi, Y. Arai, F. Kimura and M. Takada: ll Effects of unilateral hypothalamic lesion on serum gonadotropin in hemiovariectomized rats (COMMUNICATION) Kiriishi, S., H. Nagasawa, H. Kataoka, A. Suzuki and S. Sakurai: Comparison of the in vivo and in vitro effects of bombyxin and prothoracicotropic hormone on prothoracic glands of the silkworm, Bombyx mori... 149 Suzuki, M.,S. Hyodo and A. Urano: Cloning and sequence analyses of vasotocin and iso- tocin precursor cDNAs in the masu salmon, Oncorhynchus masou: evolution of neurohy- pophysical hormone precursors Takahashi, S., K. Okamoto, H. Sonobe, M. Kamba and E. Ohnishi: Jn vitro synthesis of ecdysteroid conjugates by tissue extracts of the silkworm, Bombyx mori Iwasawa, A., S. Tanaka, Y. Hanaoka and K. Wakabayashi: Development and applica- tion of time-resolved fiuoroimmunoassay for gonadotropin of a wide range of amphibian species Behavior Biology Chiba, Y. and K. Tomioka: Entrainability of diphasic circadian activity of the mosquito, Culex pipiens molestus to 24-hour light-dark cycles: a physiological significance of critical light-to-dark ratio Enviromental Biology Chew, S. F. and Y. K. Ip: Tolerance of the mudskipper, Boleophthalmus boddaerti, to a lack of oxygen (COMMUNICATION) Taxonomy Kubota, S.: Eucheilota intermedia Kubota is a distinct taxon and the third form of Eutima Japonica Uchida (Hydrozoa; Leptomedusae) (COMMUNICATION) Matsui, M., T. Sato, S. Tanabe and T. Hayashi: Local population differentiation in Hynobius retardatus from Hokkaido: an electrophoretic analysis (Caudata: Hyno- biidae) Nomura, K. and K. Hayashi: Rhynchocinetes striatus, a new species (Decapoda, Caridea, — Rhynchocinetidae) from southern Japan sls elite udeclaes ie. wean 199 ERRATUM >...) 002.5... Se eee 237 Instructions to, Authors” 752.85 eee eee 238 NUMBER 2, APRIL 1992 REVIEWS Kanzaki, R. and T. Shibuya: Olfactory pro- cessing pathways of the insect brain Ogawa, K.: Primary structure and function of audyneimpmotormmoleculeiay sen sass ee 265 ORIGINAL PAPERS Physiology Harmon, J. S. and M. A. Sheridan: Previous nutritional state and glucose modulate gluca- gon-mediated hepatic lipolysis in rainbow trout, Oncorhynchus mykiss Cell and Molecular Biology Kuroda, Y., Y. Shimada, B. Sakaguchi and K. Oishi: Effects of sex-ratio (SR)-spiroplas- ma infection on Drosophila primary embry- onic cultured cells and on embryogenesis Takagi, K. and S. Kawashima: Attempts to improve survival of neurons derived from neonatal rat hypothalamus-preoptic area in serum-free media Ichikawa, T. and K. Ajiki: Development of an in situ hybridization histochemistry for choline acetyltransferase mRNA with RNA probes Biochemistry Harbige, L. S., K. Ghebremeskel, G. Williams and M. A. Crawford: Hepatic fatty acids in wild rockhopper (Eudyptes crestatus) and magellanic (Spheniscus magellanicus) penguins before and after moulting Lawrence, J. M. and P. Moran: Proximate composition and allocation of energy to body components in Acanthaster planci (Linnaeus) (Echinodermata: Asteroidea) Developmental Biology Tanimura, A. and H. Iwasawa: Origin of Somatic cells in Bidder’s organ and the gonad proper in the toad, Bufo japonicus formosus (COMMUNICATION) Fujino, Y., A. Fujiwara, I. Yasumatsu and T. Fujii: Chromogranin A-like proteins in the heat-stable fraction of sea urchin eggs, embryos and the substances secreted with sperm Funakoshi, K., Y. Fukue and S. Tabata: Tooth development and replacement in the Japanese greater horseshoe bat, Rhino- lophus ferrumequinum nippon (COM- MUNICATION) Kearn, G. C., K. Ogawa and Y. Maeno: Hatching patterns of the monogenean para- sites Benedenia seriolae and Heteraxine heter- ocerca from the skin and gills, respectively of the same host fish, Seriola quinqueradiata (COMMUNICATION) Miyata, S., Y. Nishibe, M. Sendai, I. katayama and H. K. Kihara: Changes in timing and site of appearance of a protease in Xenopus embryos Reproductive Biology Takahashi, T., Y. Tsuchiya, Y. Tamanoue, T. Mori, S. Kawashima and K. Takahashi: Occurrence of a novel 350-kDa serine pro- teinase in the fluid of porcine ovarian folli- cles and its increase during their maturation Endocrinology Saad, A. H. and W. Ali: Seasonal changes in humoral immunity and blood thyroxine levels in the toad, Bufo regularis Nakazawa, T. S., T. Machida and S. Kawashi- ma: Effects of unilateral and _ bilateral orchidectomy on laterality of neurons of the preoptic area and plasma levels of gonado- tropins and testosterone in male mice ....357 Matteo, L. D., S. Minucci, M. D’Antonio, S. Fasano and R. Pierantoni: Effects of a gonadotropin-releasing hormone analog ili (HOE 766) on germinal and interstitial com- partments during the annual cycle in the green frog: Rana esculenta Amano, M., K. Aida, N. Okumoto and Y. Hasegawa: Changes in salmon GnRH and chicken GnRH-II contents in the brain and pituitary and GTH contents in the pituitary in female masu salmon, Oncorhynchus masou, from hatching through ovulation Nozaki, M. and A. Gorbman: The question of functional homology of Hatschek’s pit of amphioxus (Branchiostoma belcheri) and the vertebrate adenohypophysis Jacob, M.: In spermatogenesis in Oryctes rhinoceros (Coleoptera, Scara- baeidae): the role of ecdysone and juvenile hormone (COMMUNICATION) Kikuta, T. and H. Namiki: Identification of intracellular localization of laminin in the rat anterior pituitary (COMMUNICATION) vitro Behavior Biology Kasuya, E., T. Kumaki and T. Saito: Vocal repertoire of the Japanese treefrog, Rha- cophorus arboreus (Anura: Rhacophoridae) (COMMUNICATION) Asada, N., K. Fujiwara, H. Ikeda and F. Hihara: Mating behavior in three species of the Drosophila hypocausta subgroup Systematics and Taxonomy Hirose, E., T. Nishikawa, Y. Saito and H. Watanabe: Minute protrusions of ascidian tunic cuticle: some implications for ascidian phylogeny Kubota, S. and T. Horita: A new hydro- medusa of the genus Eirene (Leptomedusae; Eirenidae) from Toba, Japan Furuya, H., K. Tsuneki and Y. Koshida: Two new species of the genus Dicyema (Mesozoa) from octopuses of Japan with notes on D. misakiense and D. acuticephalum NUMBER 3, JUNE 1992 REVIEWS Nagai, Y.: mammals Nunomura, W.: C-reactive protein (CRP) in animals: its chemical properties and biologi- cal functions Primary sex determination in ORIGINAL PAPERS Physiology Kanzaki, R., N. Sugi and T. Shibuya: Self- generated zigzag turning of Bombyx mori males during pheromone-mediated upwind walking Matsuoka, T. and K. Taneda: Step-up and step-down photoresponses in Blepharisma Griffond, B., J. Van Minnen and C. Colard: Distribution of APGWa-immunoreactive substances in the central nervous system and reproductive apparatus of Helix aspersa Cell Biology Suzuki, T. and S. Funakoshi: Isolation of a fibronectin-like molecule from a marine bivalve, Pinctada fucata, and its secretion by amebocytes Khoo, G., V. P. E. Phang and T. M. Lim: The confocal laser scanning microscope as a tool for studying xanthophores of the sword- tail (Xiphophorus helleri) (COMMUNICA- TION) Immunology Zhang, H., T. Sawada, E. L. Cooper and S. Tomonaga: Electron microscopic analysis of tunicate (Halocynthia roretzi) hemocytes Biochemistry Roliti( Es i Suveude;santosyandpls Vande Linares: Phospholipids and fatty acids in intact and regenerating Dugesia anceps, a fresh water planaria (COMMUNICATION) Mita, M.: Diacyl choline phosphoglyceride: the endogenous substrate for energy me- tabolism in sea urchin spermatozoa Developmental Biology Makabe, K. W., S. Fujiwara, H. Nishida and N. Satoh: Failure of muscle myosin heavy- chain gene expression in quarter ascidian embryos developed from the secondary mus- cle lineage cells Kurabuchi, S. and Y. Kishida: Effect of delay in anterior or posterior amputation on regen- eration of short fragments of planaria ....575 Roudebush, W. E. and J. G. Kim: Mouse embryo biopsy: abnormal development with trophoblastic vesicle formation Iwamatsu, T.: Morphology of filaments on the chorion of oocytes and eggs in the meda- kanya: aids isk eee 589 Reproductive Biology Ishijima, S. A., M. Okuno, Y. Nakagome, H. — Odagiri, T. Mohri and H. Mohri: Separa- _ tion of x- and y-chromosome-bearing murine sperm by free-flow electrophoresis: evalu- ation of separation using PCR Endocrinology Zairin, M. Jr., K. Asahina, K. Furukawa and K. Aida: Plasma steroid hormone profiles during HCG induced ovulation in female walking catfish Clarias batrachus Tezuka, Y., H. Kobayashi and H. Uemura: Dipsogenic action of brain natriuretic pep- tide and endothelin-1 in the Japanese quail, Coturnix coturnix Japonica Shirai, M. and Y. G. Watanabe: Effect of brain on proliferative activity of adeno- hypophysial primordial cells in vitro De Jesus, E. G., T. Hirano and Y. Inui: Gonadal steroids delay spontaneous flounder metamorphosis and inhibit T3-induced fin ray shortening in vitro Gobbetti, A., M. Zerani and V. Botte: A possible involvement of prostaglandin E> in the reproduction of female crested newt, Triturus carnifex Tsutsui, K., S. Kawashima, V. L. Saxena and A. K. Saxena: Binding properties and photoperiodic influence of follicle-stimu- lating hormone receptors in the subtropical wild quail George, J. C. and T. M. John: Flight be- Vv haviour and thyroid hormone regulation in homing pigeons (COMMUNICATION) Ecology Smith, J. I. and H. Yu: The association be- tween vocal characteristics and habitat type in Taiwanese passerines NUMBER 4, AUGUST 1992 REVIEWS Matsumoto, A.: Hormonally induced synap- tic plasticity in the adult neuroendocrine AES ORE s, Py. GS ATRAE LA 2s eth ks 679 Bock, W. J.: The species concept in theory MGR RACICS heats. L3G. Wernvoug. Jered. (ae 697 ORIGINAL PAPERS Physiology Naitoh, T. and R. J. Wassersug: The emetic response of urodele amphibians Miro, J. L., S. Araneda and B. Canguilhem: Origin of serotonergic innervation of olfac- tory bulbs in the Europian hamster, Cricetus cricetus: An autoradiographic study using (7H]5-HT retrograde labelling Yamashita, S.: Effect of monochromatic illu- mination of the brain on the phototactic behavior of orb weaving spiders, Argiope amoena and Nephila clavata (COMMUNI- CATION) Developmental Biology Endo, K., S. Ueno, M. Matsufuji and Y. Kakuo: Photoperiodic control of the deter- mination of two different seasonal diphen- isms of the Asian comma butterfly, Polygo- nia c-aureum L. Ukeshima, A.: Scanning electron microscopy of differentiating chick ovaries during embryonic period Murakami, M., I. Iuchi and K. Yamagami: Isolation of intact yolk spheres of fish embryos, which contain the majority of lyso- somal acid phosphatase responsible for yolk phosphoprotein metabolism (COM- MUNICATION) Nakamura, S., R. Kagotani, H. Fujisaki and M. K. Kojima: The acid-insoluble organic matrix of spicules in the sea_ urchin LCHAUCENITOLUS SPU CICTTUIUS re. eee 741 Ohya, Y., K. Watanabe, N. Shimamoto and M. Amano: Scleral fibroblasts of the chick embryo can proliferate without transferrin in protein-free culture Inoue, C. and Y. Kakinuma: Symbiosis be- tween Cytaeis sp. (Hydrozoa) and Niotha livescens (Gastropoda) starts during their larval stage Amemiya, S. and Y. Nakajima: First electron microscopical study on the sperm morpholo- gy of the sea lily (Crinoidea, Echinodermata) (COMMUNICATION) Endocrinology Kobayashi, M., M. Amano, Y. Hasegawa, K. Okuzawa and K. Aida: Effects of olfactory tract section on brain GnRH distribution, plasma gonadotropin levels, and gonadal stage in goldfish Madsen, S. S. and H. A. Bern: Antagonism of prolactin and growth hormone: Impact on seawater adaptation in two salmonids, Salmo trutta and Oncorhynchus mykiss Kai-ya, H., J. Okuyama, T. Ishijima, Y. Sasayama, H. Yoshizawa and C. Oguro: Effects of Ca concentrations in culture medium on the release of calcitonin from incubated ultimobranchial glands of the bull- frog, Rana catesbeiana Kobayashi, Y., S. Kawashima, S. Takahashi and K. Wakabayashi: Effects of chronic Vi treatment with chlorpromazine on the aging of hypothalamo-pituitary-ovarian axis in the WA et cennta cam Pao PES HE ee a cheer 72) Sawada, K. and T. Noumura: Differential effects of testosterone and 5a-dihydrotes- tosterone on growth in mouse submandibu- lar gland Morphology Shirai, S. and K. Nakaya: Functional mor- phology of feeding apparatus of the cookie- cutter shark, Jsistius brasiliensis (Elasmo- branchii, Dalatiinae) Ando, K. and S. Arai: Neuropeptide Y in- nervation of cerebral arteries in microchiro- pteran bats Ecology Matsumoto, T.: Familial association, nym- phal development and population density in the Australian giant burrowing cockroach, Macropanesthia (Blattaria: Blaberidae) rhinoceros Taxonomy Nagatomi, A.: Notes on the phylogeny of various taxa of the orthorrhaphous Brachycera (Insecta: Diptera) Ohtsuka, S., R. Huys, G. A. Boxshall and T. It6: Misophriopsis okinawensis sp. nov. (Crustacea: Copepoda) from hyperbenthic waters off Okinawa, South Japan, with definitions of related genera Misophria boeck, 1864 and Stygomisophria gen. nov. Huys, R., S. Ohtsuka, G. A. Boxshall and T. It6: Itoitantulus misophricola gen. et sp. nov.: First record of Tantulocarida (Crus- tacea: Maxillopoda) in the North Pacific re- gion NUMBER 5, OCTOBER 1992 REVIEWS Takahashi, S.: Heterogeneity and develop- ment of somatotrophs and mammotrophs in thepratee hase aaain te Ae eee 901 Gerencser, G. A. and B. Zelezna: Chloride pumps in biological membranes .......... O25 ORIGINAL PAPERS Physiology Azuma, K., N. Iwasaki, M. Azuma, T. Shino- zawa and T. Suzuki: HPLC Analysis of retinoids extracted from the _ planarian, Dugesia japonica Fukuta, S., T. Ikata and I. Miura: Effect of disuse on muscle energy metabolism studied by in vivo 31-phosphorus magnetic reso- nance spectroscopy Ootsubo, T. and M. Sakai: Initiation of sper- matophore protrusion behavior in the male cricket Gryllus bimaculatus DeGeer Cell Biology Lee, Y. H. and C. E. Lee: | Ultrastructure of spermatozoa and spermatogenesis in Nepo- morpha (Insecta: Heteroptera) with special reference to phylogeny Immunology Rinkevich, B. and Y. Saito: Self-nonself rec- ognition in the colonial protochordate Botryllus schlosseri from Mutsu bay, Japan Rinkevich, B., M. Shapira, I. L. Weissman and Y. Saito: Allogeneic responses between three remote populations of the cosmopoli- tan ascidian Botryllus schlosseri Saad, A. H.: Sexual development of im- munocompetence in the toad, Bufo regularis (COMMUNICATION) _...... eis oleae 1081 Fabry, H. and J. L. Hedrick: Antibody pro- duction in the goat: Immunokinetics and epitope specificity using a glycoprotein im- munogen Biochemistry Asami, K.: Appearance of a nuclear histone H1 kinase at the start of DNA synthesis of regenerating rat liver ..-..-..-....-.--. 1001 Developmental Biology Kobayashi, H. and T. Hishida: Electron- microscopic observation of an ectopic PGC- like cell in the teleost Oryzias latipes (COM- MUNICATION) 1087 eeceee eee eee ee ee eee ese ee Reproductive Biology Suzuki, T., H. Yamanaka, K. Suzuki, K. Nakajima, K. Kanatani, M. Kimura and N. Otaki: Immunohistochemical demonstra- tion of metallothionein in the rat epididymis and spermatic cord 1009 Kubokawa, K., S. Ishii, K. Tanabe, K. Saitou and H. Tajima: Analysis of sex steroids in feces of giant pandas 1017 eee eee eee ee eee ee ee Endocrinology Paolucci, M., M. M. Di Fiore and G. Ciarcia: Oviduct 17-estradiol receptor in the female lizard, Podarcis s. sicula, during the sexual cycle: Relation to plasma 17(-estradiol con- centration and its binding proteins 1025 Vii trols humoral immunity in the lizard, Chal- CUACSHOCEILALUS Marne hn, eee nr 1037 Hasumi, M. and H. Iwasawa: Dependence of prolactin-stimulated tail fin growth and molt- ing on water in male salamanders (Hynobius nigrescens) (COMMUNICATION) . 1093 Kezuka, H., M. ligo, K. Furukawa, K. Aida and I. Hanyu: Effects of photoperiod, pinealectomy and opthalmectomy on circu- lating melatonin rhythms in the goldfish, COW ASSIUSTQUN AUIS) is ae oe ok ae 1047 Peter, V.S. and O. V. Oommen: _Intermedi- ary metabolism in_ castrated/thyroid- ectomized Calotes versicolor: Regulation by thyroxine and testosterone .......... 1055 Arakawa, E., T. Kaneko, K. Tsukamoto and T. Hirano: Immunocytochemical detection of prolactin and growth hormone cells in the pituitary during early development of the Japanese eel, Anguilla japonica ........ 1061 Morphology Ishikawa, Y.: Innervation of the caudal-fin muscles in the teleost fish, medaka (Oryzias Saad, A. H., M. H. Mansour, M. E. Yazji and TQUBCS))

1097 ZoologicalsSociety.of Japan 92-2 a2. = 42. D7 Gorodilov, Y. N.: Rhythmic processes in ZANNOUNCEMENM(S. chetys ace ost nae se eee 1307 lower vertebrate embryogenesis and their INEKMOWIEGCEMENS Seana cat sccies cass: eta 1308 role for developmental control ......... eT OME UTI O TaN ONG Ee err ace ea slain 1309 Suzuki, K. and D. G. Furth: What is a clas- sification? A case study in insect systema- Contents of ZOOLOGICAL SCIENCE, Vol. 9, Nos. 1-6 ish aA, ha ‘aon * anei fb einen nary ee ES * - ae p a < ‘ ae : hee oer 7 - : ar Brier = Re’ : te.” q dome, = : ‘ : | Z - . res ae ; 7 : 7 ; d i ; * : . } t j fe ¥ = j b ) - \ \ i ‘ a \ . bs ) i Development Published Bimonthly by the Japanese Society of Developmental Biologists e ° ge Distributed by Business Center for Academic Growth & Differentiation Societies Japan, Academic Press, Inc. Papers in Vol. 34, No. 6. (December 1992) 66. REVIEW: K. Yoshizato: Death and Transformation of Larval Cells during Metamorphosis of Anura 67. A. Yanagi: Nuclear Differentiatin in Paramecium caudatum: Analysis by the Monoclonal Antibody against a Macronuclear Antigen 68. L. Bosco and S. Filoni: Relationship between Presence of the Eye Cup and Maintenance of Lens-Forming Capacity in Larval Xenopus laevis 69. A. Hikosaka, T. Kusakabe, N. Satoh and K. W. Makabe: Introduction and Expression of Recombinant Genes in Ascidian Embryos 70. R. Collura and K. S. Katula: Spatial Pattern of Expression of Cyl Actin-(-Galactosidase Fusion Genes Injected into Sea Urchin Eggs 71. Z.-S. Ji, K. Kubokawa, S. Ishii and S-I. Abé: Differentiaiton of Secondary Spermatogonia to Primary Spermatocytes by Mammalian Follicle-Stimulating Hormone in Organ Culture of Testes Fragments from the Newt, Cynopus pyrrhogaster 72. K. Kinoshita, Y. Fujii, Y. Fujita, K. Yamasu, T. Suyemitsu and K. Ishihara: Maternal Exogastrula-Inducing Peptides (EGIPs) and Their Changes during Development in the Sea Urchin Anthocidaris crassispina 73. T. Sawai: Effect of Microtubular Poisons on Cleavage Furrow Formation and Induction of Furrow-like Dent in Amphibian Eggs 74. K. Mitsunaga-Nakatsubo, M. Kanda, K. Yamazaki, H. Kawashita, A. Fujiwara, K. Yamada, K. Akasaka, H. Shimada and I. Yasumasu: Expression of Na*, K*-ATPase a-Subunit in Animalized and Vegetalized Embryos of the Sea Urchin, Hemicentrotus pulcherrimus 75. T.-C. J. Wu, L. Wang and Y.-J. Y. Wan: Differential Expression of Retinoic Acid Receptor mRNA during Mouse Embryogenesis 76. A.A. Oohata: Induction of Cell Differentiation of Isolated Cells in Dictyostelium discoideum by Low Extracellular pH 77. §S. Komazaki: Ultrastructural Localization of Calcium in the Presumptive Ectodermal Cells in Gastrulae of the Newt, Cynops pyrrhogaster, by Cytochemistry and X-Ray Microanalysis 78. T. Iwamatsu, M. Kikuyama and Y. Hiramoto: Fertilization Reaction without Changes in Intracellular Ca7* in Medaka Eggs—An Experiment with Acetone-Treated Eggs 79. K. Yamada, S. Eguchi, T. Yamamoto, K. Akasaka and H. Shimada: Cis-Acting Elements for Proper Ontogenic Expression of Arylsulfatase Gene of Sea Urchin Embryo Author Index to Volume 34 Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of Developmental Biologists, Department of Developmental Biology, 1990: Volume 32. Annual subscription for Vol. 33, 1991: U.S.$ 162,00, U.S. and Canada: U.S. $ 178,00, 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. The Ultimate Name in Micromanipulation Ease of operation, and the most advanced Can be combined with New WR & MX series Model MX -2 3-D Micromanipulator Model MX-1 3-D Micromanipulator _ Model WR-87 One Dimensional Aqua Purificate Micromanipulator Model SR-6 Model PP -83 Stereotaxic Instrument for Rat Glass Microelectrode Puller * * & Send enquiries for request for MODIFICATIONS or IMPROVEMENTS * * * Physiological, Pharmacological, Zoological & Neurosciences Research Equipments TRARISHIGE) NARISHIGE SCIENTIFIC INSTRUMENT LAB. aay 9-28 KASUYA 4-CHOME SETAGAYA-KU, TOKYO 157, JAPAN PHONE (INT-L)81-3-3308-8233, FAX (INT-L) 81-3-3308-2005 CABLE: NARISHIGE LABO, TELEX, NARISHIGE J27781 XL 7 (04 NH ZOOLOGICAL SCIENCE An International Journal ZOOLOGICAL SCIENCE The Official Journal of the Zoological Society of Japan Editors-in-Chiet: The Zoological Society of Japan: Semen@ | aves (CTO) Toshin-building, Hongo 2-27-2, Bunkyo-ku, Goda | epeyesal (TP eis7o) Tokyo 113, Japan. Tel. (03) 3814-5675 Managing Editor: Ofncon. Bounce Vague n Clana) President: Hideo Mohri (Chiba) pssistaneE (Mots: Secretary: Hideo Namiki (Tokyo) wigslai’se, [Samson CO xeyeonine) Treasurer: Makoto Okuno (Tokyo) SUES oS vans) Librarian: Masatsune Takeda (Tokyo) Akiyoshi Niida (Okayama) —— 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 (Tokyo) John M. Lawrence (Tampa) Koscak Maruyama (Chiba) Roger Milkman (Iowa) Kazuo Moriwaki (Mishima) Richard S. Nishioka (Berkeley) Chitaru Oguro (Toyama) 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. Tsuneo Yamaguchi, Managing Editor, Zoological Science, Department of Biology, Faculty of Science, Okayama University, Okayama 700, 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, Godfried van Seystlaan 47, 3703 BR 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 1992, 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 9: 1-16 (1992) © 1992 Zoological Society of Japan REVIEW Transduction Process and Impulse Initiation in Insect Contact Chemoreceptor HIROMICHI Morita Emeritus Professor of Kyushu University, 6-21-3 Kasumigaoka, Higashi-ku, Fukuoka 813, Japan INTRODUCTION It was generally accepted that the local potential evoked in sense organs by adequate stimuli gener- ates afferent impulses in an adjacent region of sensory neurons (generator potential [1]), but it was after 1950 that the generator potential was actually recorded in many sensory neurons [2-9]. Since then, the word “receptor potential” gradual- ly took the place of “generator potential”. Now, the word “generator potential” is rarely found in current papers. This can be understood as we are now most interested in sensory receptor mecha- nisms. It has been suggested for these several years that cyclic AMP (cAMP) or cyclic GMP (cGMP) plays a part in sensory cells as in the target cells for some hormones [10-13]. We recently wrote a review article “chemoreception physiolo- gy” in insects [14], where we could not describe the above aspects of studies on insect chemoreception. Main purpose of the present review is therefore to describe and discuss the development of the field since then. Thus, cell biological aspects are more important here. However, generation of impulses in chemosensilla of insects is so unique in relation to their structures that this problem is also dis- cussed. There are many excellent papers describing in- sect chemoreception with behavioral responses [15-24], but studies of individual chemosensory cells have been made possible only after elec- trophysiology was introduced. Electrophysiologi- Received October 7, 1991 cal study of the insect chemosensory system had been thought difficult because of its tiny size compared with the vertebrate system. As early as 1941, Pfaffmann recorded afferent impulses from taste nerve fibers of the cat [25], while Nodai published a paper in 1953 to report electric poten- tial changes of the afferent nerve by chemical stimulation of tarsal chemosensilla of an insect, orthopteran [26]. Thereafter, many efforts were made in vain to record chemosensory impulses from the tarsal nerve of insects [27]. Once the technique was established for recording of im- pulses [28] and receptor potential [9, 29], however, it soon became clear that insects were suited for study of individual sense cells because of their small number in a single sense organ, sensillum. Now, for example, the labellar sugar receptor cell of the blowly, Phormia regina, and the fleshfly, Boettcherisca peregrina, is one of the most thor- oughly investigated chemosensory cells in the animal kingdom. SENSILLUM STRUCTURE AND IMPULSE INITIATION When we first recorded impulses from the tip of the tarsal chemosensillum of the butterfly, Vanessa inidica, we were surprised to find that the impulse represents an increase in positive potential at the tip with reference to the base of the sensillum [27]. This was true also for the labellar chemosensillum of the green bottle-fly, Lucilia caeser [30, cf. 31]. We expected that impulses would be initiated in a region near the sensillum tip and would therefore 2. H. Morita represent an increase in negative potential at the tip with reference to the base. This polarity was understood as the result of initiation of impulses somewhere near the sensillum base [30, 32]. The next question is why impulses are initiated in the basal region instead of the tip region where the first step of reception occurs. To answer the question, we have to learn morphology of insect sensilla. All the investigated hairs, bristles, pegs and cones on the body surface of insects contain primary sense cells (sensory neurons) that send their axons to the central nervous system (CNS). The number of the sense cells in a sensillum varies with types of sensilla and species of insects, but is generally one to several. The sensilla containing only one neuron are mostly mechanosensitive. The most intensively studied distance chem- osensory (olfactory) sensillum of Bombyx contains two neurons. The labellar contact “chemosensory (gustatory or taste) hair” of flies contains one mechanosensory and four chemosensory neurons (i.e., it is a mechano- and chemo-sensitive sensil- lum). All the neurons are bipolar, whose cell body (perikaryon) is located rather deeply beneath the base of the sensillum. The dendrite extends from the cell body toward the base of sensillum, on the way to which it terminates into a cilium connecting itself with the distal segment. This segment is differentiated into a sensor of the adequate stimu- li. Such is the case in the rod and cone of vertebrate retina: the cilium connects the cell body with the outer segment; this segment is differenti- ated into a photo sensor. As shown by Figure 1 [33], the distal segments of dendrites (dds) are covered together with the dendritic sheath (dsh). This sheath appears the same as cuticle as far as transmission electron microscopy shows, and sepa- rates the inside of the sensillum into two lumina: inner lumen containing the distal segments of dendrites and outer lumen containing nothing but fluid. The inner lumen proximally opens to the ciliary sinus (cs), and the outer lumen to the sensillar sinus (ss). The perikarya of sensory neurons (7) are triply wrapped by inner (i), in- termediate (m) and outer (0) sheath cells. The inner sheath cell forms the ciliary sinus (cs) around the connecting cilia (clm) and is also thought to secrete the dendritic sheath (i.e., thecogen cell). Fic. 1. Structure of a typical sensillum. bf, body fluid; clm, connecting cilium; cs, ciliary sinus; cu, cuticle, dds, dendritic distal segment; dsh, dendritic sheath; e, epidermis; g, glia cell; 7, inner sheath cell; m, intermediate sheath cell; n, nerve cell; 0, outer sheath cell; ss, sensillar sinus. Adapted with permis- sion from [33], Pergamon Press PLC. The outer and intermediate sheath cells (trichogen and tormogen cells, respectively) together form the sensillar sinus (ss). The sensillar tip is 50 to 100 mV positive with reference to the body fluid (bf) [34, 35, 36]. Thurm [35] called this potential “transepitherial potential” (TEP), and showed that TEP requires an energy dependent process, i.e., active transport of some ions to or from the sensillar sinus. In contact chemosensilla of many genera of flies, the concentric arrangement of two cylinders (den- dritic sheath and outer cuticular wall), as shown in Figure 1, changes into a very eccentric one, 1.e., Insect Contact Chemoreception 3 Fic. 2. Cross section of labellar chemosensillum at one quarter of its length from the tip (Phormia regina), < 28,000. By courtesy of Prof. Y. Tominaga, Fukuoka University. the dendritic sheath is fused with the outer wall around nearly a half of its circumference at any level of cross section (Fig. 2). Thus, the inside of the shaft of sensillum is divided into two halves with a partition wall of the remaining non-fused part of dendritic sheath. Such a sensillum appears as two-toned under an optical microscope with transmitted light: dark through the inner lumen and light through the outer lumen. The largest type of labellar chemosensilla of fleshfly or other big flies are so large that a microelectrode tip of glass pipette can easily be inserted into the outer lumen. It is somewhat difficult to insert a mi- croelectrode into the inner lumen, but not impossi- ble. Figure3 shows the records of the same impulse taken by two microelectrodes inserted into the inner (7) and outer (0) lumina, respectively, at the same level above the sensillar base. The common reference electrode was introduced into the proboscis through crushed head, and was grounded via a calibration pulse generator of low resistance. In this case, a rectangular pulse, +1 mV of 2ms duration, was added to show the fidelity of the amplifiers used. The spike height of the impulse recorded from the inner lumen is more than three times higher than that from the outer lumen: the inner and outer lumina are not equipotential. This indicates that the partition wall i a Fic. 3. Records of the same impulse obtained from the inner (z) and outer (0) lumina of the labellar chem- osensillum (Boettcherisca peregrina) [40]. 1 msecx5 Q Cie vr is highly resistive against electric currents. When an adequate stimulus such as a sugar solution is applied to the tip of the sensillum, a slow sustained potential accompanied by trains (or a train) of impulses is recorded between two microelectrodes inserted into the outer lumen, one in the distal region and the other in the proximal of the sensillum (Fig. 4). The polarity of the slow potential is negative at the distal electrode with reference to the proximal: an electric current be- gins to flow toward the tip within the outer lumen prior to initiation of impulses, and is sustained during the application of stimulus. To illustrate such a situation [42], the structure of the labellar chemosensillum is shown in Figure5, and the corresponding equivalent electric circuit in Figure 6A, the equivalent or effective resistances to the generator current for impulses being emphasized. Terminals in, out, ilt and is in Figure 6A represent the same lettered points in Figure 5, respectively. It is assumed in Figure 6A that the inner lumen is completely insulated from the outer lumen by the partition wall except at its two ends. A canal communicating the inner and outer lumina at the sensillar tip is assumed to exist because the current flows toward the tip within the outer lumen. Pre- cise location and structure of the canal need fur- ther discussion (see below). Thus, TEP (its magni- 4 H. Morita |\/64 M A 1/32 M B 1/16 M C 1/8 M D 1/4 M E 1/2 M F | i : 7 2 mV i] 0.1 Ss Fic. 4. DC records of responses of the single labellar chemosensillum (Boettcherisca). Stimulated by glucose solution of the indicated concentration [40]. out Fic. 5. Illustration for the equivalent circuit diagram in Figure 6. i/t, tip of inner lumen; in and out, inside and outside at the receptor membrane, respectively; is, impulse initiation site; Eypp, electromotive force of trans epithelial potential (TEP). Others are the same as in Figure 1 [42]. Insect Contact Chemoreception 5 B in Uy 1/ng I Ey out out (E) (0) Fic. 6. Equivalent circuit diagrams for impulse generator current (or receptor current). The terminlals represent the same lettered points in Figure 5, respectively [42]. tude, E7,p) effectively polarizes the dendrite distal segment (dds) between both ends of the partition wall (arrows in Figure 5). An adequate stimulus increases the conductance of receptor membrane (ng) and an inward current (/) flows across the receptor membrane (from out to in). Current J is divided into J; and JL, the latter flowing out from the distal segment most intensively at the proximal end of the partition wall (impulse initiation site, is) under the strong influence of TEP (the intensity of I, is the highest at is and decays exponentially along the distance from it). The canal at the tip is electron microscopically difficult to show, but was demonstrated by diffu- sion of Ag* into the inner and outer lumina when AgNOs solution was applied to the tip [14]. The process of Ag* diffusion is clearly visible under a light microscope as a growth of precipitation, most probably, of AgCl. Sometimes, dense precipita- tion makes a plug in the inner or outer lumen, beyond which no diffusion occurs. When the plug is made in the outer lumen, we can see the growth of precipitation only in the inner lumen. Such an observation does not necessarily prove the exist- ence of communication canal at the tip, but is also possible if each lumen opens outward at the tip. In fact, there is an electron microscopical study on outlets of the outer lumen at the tip [39]. If the inner lumen were completely insulated from the outer lumen by the partition wall even at both ends, i.e., if resistances R, and R> in Figure 6A were infinitely high, none of the current of inflow at the tip would flow out from the distal segment except at loci proximal to the proximal end of the partition wall, i.e., /=J,. Even though insulation at the tip and root of the distal segment is not perfect, the partition wall assists the current to flow within the distal segment from the tip to root effectively for generation of impulses. In a previous review [14], we presented the results of calculation to show how the partition wall favors transmission of signals (membrane potential changes) from the tip to root without contribution of TEP. Many investigators of insect sensilla, especially mechanosensilla, believe that TEP is the only force driving the generator current to flow. It is because K~ concentration was quite high (100 to 200 mM) in the receptor-lymph cavity (sensillar sinus, ss) [37 cited in 38], and the receptor mem- brane immersed in this lymph is considered to be already depolarized before any adequate stimulus is applied. Recently, Kijima et al. (personal com- munication) studied effects of TEP on the initia- tion of impulses in the labellar chemorecptor of the blowfly, and found a close correlation between impulse frequency and TEP which was lowered by metabolic inhibitors. However, the receptor still responded to sugar solutions with low frequency of impulses even when TEP was completely dimi- nished. Further, he found that the concentration of Na” is higher than that of K~ in the sensillar shaft. Such results suggest that the depolarization of receptor membrane on stimulation contributes to the generator current in the labellar chem- oreceptor of the blowfly. Figure 6B shows the equivalent circuit, which I have used for a model describing the relation between simulus strength and response magnitude 6 H. Morita since I published it 1969 [40]. This circuit is the same as that adopted in non-linear summation of membrane potential changes at the muscle end- plate [41]. Now, the potential at out with reference to in is defined as E when the receptor membrane current J flows for n, the number of the channels opened by the stimulus, g being the conductance per opened channel. For simplicity, n is assumed as zero at rest (without stimulus). Therefore, the potential change V by stimulus becomes as V= E— Em, Where E,, is the resting membrane potential. The theoretical maximum V,,, is written as V,,=E, —E,,, where £, is a) reversal potential of the receptor membrane. Then, we obtain: V=V,,/ (1+G/ng), and I=—GV. If we plot V against log ng, a sigmoid curve is obtained. Considering a contribution of TEP to impulse initiation, we obtain the same equation by rede- fining V,,=E,—(Em+Erep) and V=E—(E,,+ Eryep) for an ideal case where no current flows through conductance G;, [14]. Actually, however, it is most probable that a current J, flows across the receptor membrane from a source just adjacent to it. For such a case, the circuit diagram of Figure 6C is available. This is reduced from Figure 6A, where for J, to flow, R3 is negligible compared with 1/Gz, conductance 1/R; is negligible compared with 1/R3, and R> can then be included in 1/G,. Defining V and V,,, as above, we obtain an equa- tion for J, as 1h=—Vm G2/(1+ G/ng) + Eqep/(ng/G,G2+ WG Gs) where G=G,+G). The equation reminds us that a current, 5 ,—0, flows through G> even at rest (n =0). Only a current increase by stimulation is of interest for us, and it becomes Bf, p=0—= > GoxV n+ Ever Gi/G)/(1+ Ging). Defining Vi=(VintEree Gi/G)/(1+G/ng), (1) we can write the actual maximum as Vs=(Vint Erep Gi/G)/(1+G/sg), (2) where s is the total number of the channels. Asuuming that the channel is opened by sucrose making 1:1 complex with its receptor site, 1.e., introducing n=s/(1+K/C), (3) C being the sucross concentration and K the dissociation constant between sucrose and the re- ceptor site, we obtain the same equation as that published before [40]: Ve— Vel (ape (4) where K,=K/(1+sg/G). (5) The response in impulse frequency (r) in an initial steady state was ascertained to be proportional to the current response [,—I, ,— 0 [40], so that we can write equation (4) as r=r,/(1+K,/C), where r, is the actual maximum of response in impulse frequency. From this equation, we can derive so-called Beidler’s tasts equation as Clr= Clr.“ Rel te Beidler [43] obtained this relation assuming that response is proportional to the number of stimulus molecules binding to the receptor site, and, there- fore, regarded K, as the dissociation constant. TRANSDUCTION AND ADAPTATION Based on the assumption that the channel is opened directly by sucrose binding to the receptor site in 1: 1 manner, the intensity-response relation is described above. This assumption is valid for all sugars except for monosaccharide, where the bind- ing is in 2:1 manner (2 sugar molecules to 1 receptor site of the channel) [44]. Such an assump- tion should be rewritten by a model which numer- ically accounts for the maximum response and Hill coefficient [45-48], but the result may be much the same as far as we assume that the channels is directly operated by sugars. Recently, however, Amakawa et al. [49] obtained the results suggesting that cyclic GMP (cGMP) is involved in the sensory transduction as Insect Contact Chemoreception 7 the second messenger within the sugar receptor cellin Phormia. They applied a membrane perme- able cGMP analogue, dibutyryl cyclic GMP (dbcGMP), to the tip of the sensillum, and re- corded responses from the sugar receptor cell (but not from the other receptor cells). The adaptation (decline of impulse frequency) to dbcGMP sti- mulation was quite slow compared with that to sucrose stimulation in a period more than 3 s after the beginning of stimulus; so slow that the cell continued to discharge impulses of almost a con- stant frequency over 30 s. Furthermore, the cell discharged impulses for a while even after the end of stimulus, when stimulated by dbcGMP mixed with a membrane permeable inhibitor of cyclic nucleotide phosphodiesterase, theophylline or iso- butyl-methyl-xanthine (IBMX). The channel has been assumed to be directly opened by sugar binding, partly because the gener- ator current of impulses (receptor current) begins to flow without any detectable delay after the onset of stimulation. By noise analysis of the sugar receptor current in Boettcherisca, Kijima et al. [50] obtained the results: (a) The autocorrelation func- tion, or covariance function [51], C(t), could be approximately described by an exponential term. (b) The time constant of this exponential term differed with different sugars even when the recep- tor currents were the same in amplitude. General- ly, C(t) is defined as CA=((T)— , 1.e., the same amplitude of receptor current), if the channel would be brought to this equilibrium state by the same concentration of intracellular cGMP. This is quite contrary to the above results (a) and (b). Thus, it is strongly suggested that the channel is directly operated by sugars. The transduction mechanism in Phormia might be different from that in Boettcherisca. There is indeed a report in Boettcherisca that dbcGMP and its derivatives scarcely cause the sugar receptor cell to discharge impulses, but do the salt receptor cell [52] (cf. the case in Phormia: dbcGMP causes only the sugar receptor cell to discharge impulses [49]). Nevertheless, an involvement of cGMP in the sensory transduction [49] does not necessarily con- tradict a direct operation of the receptor channel by sugars [50]. It is possible that the receptor channel bound to intracellular cGMP is opened only after extracellular binding of sugars: both cGMP- and sugar-bindings are necessary for the opening, and the sugar-binding is the rate limiting step. In such a case, the time course of relaxation differs with sugars and their concentrations as shown by Kijima et al. [50]. Amakawa et al. [49] pointed out the possibility of extracellular receptor site for cGMP and dbcGMP, and there is no experiment of purely intracellular application of cGMP. The extracellular binding of stimulants may not only open the channel, but may also decrease the concentration of intracellular cGMP in some way or other (cf. rods and cones of vertebrates, [53]), resulting in a decrease of open channels. Thus, it is conceivable that the mem- brane permeable dbcGMP, whether be intracellu- larly converted to cGMP or take the part of cGMP, sustains the receptor current as long as dbcGMP stimulation continues. _ The receptor current rapidly decreases to zero after a stimulus solution is broken contact with the sensillum tip. This rapid decrease has been thought to be due to active processes within the receptor cell, because the current tends to last after the end of stimulus in deteriorated prepara- tions. Amakawa et al. [49] obtained the result suggesting that one of the above active processes is hydrolysis of cGMP by phosphodiesterase (cf. [75]). Amakawa and Ozaki [54] tried intracellular ap- plication of activators or inhibitors of protein kinase C by exposing (1 or 2 min) the sensillar tip to a solution containing one of the reagents. The response to sucrose was recorded 5 min after the treatment, and was compared with that after the treatment with the same solution lacking the rea- 8 H. Morita gent. The activator (phorbol ester; DPBA or TPA) depressed the response to sucrose, while the inhibitor (H-7) raised it. Both effects were de- tected with a delay of 0.4-0.5 s after the onset of stimulation, and were thought to be related to an adaptation to long continued stimuli. If we define adaptation simply as a decline of impulse frequen- cy in response to a continued stimulus, the adapta- tion started 0.4—0.5 s after the beginning of stimu- lus in the normal sugar receptor cell. The mem- brane impermeable agent, DPBA or H-7, was made to penetrate into the receptor cell, being mixed with the treatment solution of 0.03% sodium deoxycholate (DOC). Treatment with DOC itself accelerated the adaptation: the adapta- tion started immediately after the onset of stimula- tion. The effect of DOC treatment was antago- nized by a Ca’*-chelator EGTA mixed with DOC. Protein kinase C is activated by Ca** and di- glycerides (or instead, phorbol esters), and there- fore the kinase C was suggested to promote the adaptation to sucrose stimulation. As discussed before, the response to a long lasting stimulus seemed to depend on the in- tracellular concentration of cGMP [49]. As one of the other ways for regulation of response, the kinase C inactivates the receptor molecule in the membrane, resulting in a decrease of the mole- cules available for response. Such might be the case as a kind of desensitization, since neither activators nor inhibitors of kinase C changed the response at the beginning of stimulus. In contrast with kinase C, inhibitors of kinase A or G reduced the response at the beginning [54]. Thus, cGMP and protein kinases (C and G or A) are most probably involved in regulation of re- sponse in the sugar receptor cell of the blowfly. At present, however, we cannot describe a whole story of sensory transduction and adaptation in the sugar receptor cell of the fly. Amakawa et al. [49] ascertained in Phormia that the sugar receptor cell responds to dbcGMP, by examining the proboscis extension response to dbcGMP as well as the spike height of impulses. As noted before in Boettcherisca, the salt receptor cell responds to dbcGMP or its derivatives, but the sugar receptor cell does scarcely [52]. Thus, the same receptor site is possible to occur in a different category of receptor cell (see below). RECEPTOR MOLECULES Multiple receptor sites in a sugar receptor cell Taking into account conformations of sugars and related carbohydrates, Evans [55] proposed that the sugars which stimulate taste receptors of the blowfly combine with two or more distinct receptor sites, each with unique structural requirement, and that these sites are associated with the same recep- tor cell. This is the first definite statement of multiple receptor sites in a sugar receptor cell, referring to the glucose and fructose sites. There- after, we have been able to effectively discuss the structure-activity relationship in the blow- and flesh-flies. Shimada et al. [56] showed in the fleshfly that a 3 min treatment of a single sugar receptor cell with 0.5 mM p-chloromercuribenzoate (PCMB) almost completely depressed its response to D-glucose, but did not affect the response to D-fructose. This was the first experiment discriminating the two receptor sites in a sugar receptor, and it became possible to decide with which receptor site a sugar combines. Figure 7 shows an example: D-fructose, D-fucose and D-galactose react mainly with the fructose site, while D-arabinose, L-fucose, L- arabinose, L-sorbose, D-xylose, L-glucose, as well as D-glucose react wich the glucose site. Sucrose and maltose are judged to react with the glucose site as an a-D-glucopyranoside. Monosaccharide such as fructose has many con- formations as well as a, 8 anomers in an aqueous solution. Therefore, we had first of all to make clear what conformation and which anomer is effective. Chasing the time course of changes in stimulating effectiveness accompanied by muta- rotation, Hanamori et al. [57] concluded that D- fructose is effective as the §-D-furanose form. This and the forgoing results clearly indicated that the glucose site combines with polyols in the form of a-pyranose, while the frustose site combines with those in the form of 8-furanose. Therefore, the glucose and frustose sites are now called pyra- nose (P) and furanose (F) sites, respectively. The structural requirement of stimulants for activating Insect Contact Chemoreception 9 VILILLILLLALAA LALLA LL aR O'5 VLLLL F D-Fucose G F G Relative response after PCMB treatment D-Fructose 22222277277772777 777 PPA D- Galactose ¥2¢2/2/ 478 F D-Arabinose G F I L- Arabinose #— th RH ® ® @ @ w © e W o w S ” vo) ro) L © OuyoOulouso S,/ueoOuUZO > — = 0) x< G) G) Ww) = I ! ! J ai) (a) zy Q Fic. 7. Comparison of responses to some mono- and disaccharides after PCMB treatment. The response (with standard deviation) is normalized so that the response before the treatment is unity [56]. P site is the existence of three juxtaposed equato- rial hydroxy groups in a chair form [56, 57]. That of F site is so. strict that mehyl-@-D- fructofuranoside is ineffective [57, 58]. Shiraishi and Kuwabara [59] stimulated the sugar receptor of the fleshfly with 19 L-amino acids, out of which 6 amino acids (valine, leucine, isoleucine, methionine, phenylalanine, and tryp- tophan) were effective. Shimada and Isono [60] divided F site further into F and T sites by treat- ment with pronase, which did not affect the re- sponse to D-fructose but markedly depressed the response to L-valine. Thus, T site was responsible for reception of aliphatic amino acids. They thought that aromatic amino acids reacted with F site, because the treatment with pronase did not affect the response to phenylalanine and tryp- tophan. There was a definite stereospecificity of F site for aromatic amino acids [61]. On the other hand, there was a different rigid stereospecificity of F site for furanose and its derivatives. This was ascer- tained with 2,5-anhydro-D-hexitols of definite con- figurations [62]. The above two specificities were so different that the sites for aromatic amino acids and for furanoses must not be the same. Accord- ingly, Shimada et al. [62] proposed a new site Ar (Ar=aromatic or aryl) for aromatic amino acids, and also renamed the site for aliphatic acids R site instead of T site (R=alkyl). So far four receptor sites in total have been proposed for the receptor sites in the sugar recep- tor cell of the fleshfly. They are P site for pyranose and its derivatives, F site for furanose and its derivatives, Ar site for aromatic amino acids and its derivatives, and R site for aliphatic amino acids, small peptides, and fatty acids. In addition, the neucleotide site was proposed for the sugar recep- tor cell in Phormia as mentioned before [49], but it was for the salt receptor cell in Boettcherisca [52]. Recently, Shimada et al. [63] found in Boettcher- isca that certain 1,6-anhydro-$-D-hexopyranoses (1,6-anhydro-~-D-galactose, -altrose, -talose and -gulose) were effective in stimulating the salt re- ceptor cell: there is a sugar receptor site in the salt receptor cell. 10 H. Morita It is well known that ingestion of food in flies is released by afferent impulses of sugar receptor and water receptor cells and is blocked by those of salt receptor cells. Water is essential for the mainte- nance of life in flies, but is reyjected when con- taminated with salts of high concentrations. However, water with such salts would still be valuable if it contains nutritious sugars. The response of water receptor cell to water in itself is completely blocked by dissolved 0.05 M sodium citrate, but is recovered by addition of some sugars in it. Wieczorek and Koppl [64] discovered this phenomenon in Protophormia and called it reac- tivation by sugars. The receptor site was a kind of F site; indeed, a sugar receptor site in the water receptor cell. Search for receptor molecules—a-glucosidase Eluting the legs of flies (Phormia) with water and testing for enzyme activity in the eluate, Dethier [65] found an a-glucosidase activity. This brought about Hansen’s biochemical investigations on the receptor mechanism in Phormia [66]. He demonstrated a positive correlation between the regional a-glucosidase activity and the regional density of chemosednsilla and between the sub- strate specificity of the enzyme and the stimulating effectiveness of a-glucosides (for the above- mentioned P site). Kijima et al. [67] picked up this proposal in Phormia, separated isozymes [68], characterized them [69], and found a membrane- bound a-glucosidase in the chemosensillum [70, 71]. Over a wide spectrum of substances, they disclosed good parallelisms in substrate- and in- hibitor-specificities between the enzyme and the Sugar receptor, but only one result of a great discrepancy was enought to discard the hypothesis: the inhibition constant of nojirimycin for the a- glucosidase was in a range of 10° to 10°-° M, while that for the sugar receptor was above 4x 10-* M [72, 73]. Dethier wrote in his recent review [74]: “They (including Hansen et al.) proved that the enzymes were intimately and exclusively associated with the receptors. Exactly what part the glucosidases play in the process of transduction is still a mystery.” Whatever a mystery may be, it is certain that sucrose is split into glucose and fructose on the outside of the intact sugar receptor membrane. It amounts to 0.5 pmol per sensillum per hour, i.e., 1.4x10~'° mol s~', when the sugar receptor is stimulated by 0.1M sucrose [72]. This is just comparable with the number of monovalent cat- ions crossing the receptor membrane for 1s. The split glucose and fructose molecules might be picked up into the distal segment of the sugar receptor cell, as suggested by Hanamori [75]. He exposed the sensillar tip to '*C-D-glucose or 3-O- mehtyl-'*C-D-glucose, effective or ineffective stim- ulant for the sugar receptor cell (Phormia), respec- tively, and studied the uptake and movement of isotopic activity. When the tip was pretreated with 75 mM colchicine for 30 min, the taken up and remaining activity within the sensillum shaft in- creased for glucose, but not for 3-O-mehyl- glucose. This was explained as a result of dis- assembly of microtubules, which otherwise could transport glucose, but not 3-O-methyl-glucose, proximally to the cell body. That is, the glucosi- dase might play a part in the rapid stop of response after the end of stimlus. Affinity electrophoresis Ozaki [76] at last found a most probable candi- date for P site in Phormia, with affinity elec- trophoresis. Her strategy was based on her own finding that starch does not evoke but inhibits the response of the sugar receptor cell [77]. Starch has no electric charge and is polydisperse with huge molecular weights in an aqueous solution. There- fore, a protein molecule under electrophoresis moves only in a state free from formation of the starch-protein complex. As the mobility of protein is thus proportional to the period of this free state, we obtaine (the fraction of free state is derived from equation (3) as (s—n)/s=1—1/(1+K/C), where n represents the number of molecules mak- ing complex in 1:1 manner, s the total number of molecules): m,/m=1+[I]/Ki, (6) where m and m, is the mobility of protein with and without starch, respectively, [I] the concentration of starch, and K; the dissociation constant of the protein-starch complex [78]. Further, because a protein molecule can be assumed to have the same Insect Contact Chemoreception 11 mobility when it forms complex with a small uncharged molecule of sugar, we obtain an equa- tion as in competitive inhibition, m /(m,—m’)= Ki +[s]/Ka)/[I], (7) where m’ is the mobility of the protein in the presence of starch and the sugar, [s] the sugar concentration, and K, the dissociation constant of the protein-sugar complex [79]. In two-dimensional polyacylamide gel elec- trophoresis, starch was added to the gel of the first run. The second run was carried out through a slab gel that were the same in composition except for lacking starch. Proteins having no interaction with starch should align on a diagonal line starting at the origin of electrophoresis. In fact, a protein was detected at a spot aside from the diagonal line, at shorter distances on the first run with higher concentrations of starch (Fig. 8). The dissociation constant of the protein-starch complex obtained from equation (6) was 0.7%, which agrees well with the value, 0.6%, obtained before in her physiological study [77]. The dissociation con- -05 0 O5 10 15 20 CONCENTRATION OF STARCH, II] ‘/o B =—»> stants of the protein-sugar complex obtained from equation (7) also agreed with the values predicted by equation (5): K, is larger than K, by a factor of (1+sg/G), which was previously estimated to be about 5 in Phormia [80]. In her study, a-glucosidase did not interact with starch, suggesting that this novel protein is differ- ent from a-glucosidase mentioned above. MAINTENANCE OF RECEPTOR MEMBRANE If we examine the chemoreceptor activity in flies collected in the field, we will find that many chemoreceptor cells have lost their activity. The contact chemoreceptor cells are exposed to many dangerous substances while the fly searches for food. Therefore, it is important for the fly to recover the activity of injured chemoreceptors. Shimada [81] is the first to report the recovery from an injury resulting from contact with deter- gents in Boettcherisca. Being exposed to 0.1% sodium deoxycholate (DOC) for 5 min at the sensillar tip, the chemoreceptor cells discharged injury spikes for a while, ceased to discharge within the exposure time, were silent for 10-20 min, then began responding again, and finally recovered about 80% response at about 50 min after the detergent exposure. The destruction and reorganization of the recep- tor membrane were reinvestigated in detail in Phormia. Recording the receptor membrane cur- rent during the DOC treatment, Kashihara et al. (unpublished) follwed changes from a partial to full lysis of the membrane at the sensillar tip. The full lysis was characterized by a large continued injury current accompanied by an intense dis- charge of impulses. However, as in Boettcherisca, the response to 0.1 M sucrose (0.1 M sucrose re- sponse) recovered about 90% at 90 min after the Fic. 8. (A) Determination of dissociation constant Kg after equation (6). (B) Determination of m and m,. In the presence of starch (1st run), starch- interacted protein migrated to position p._ If starch had not existed in the first run, the protein should have migrated to position q on the diagonal line in an extension of horizontal line op. The diagonal line is defined by the protein stain (dotted area). Then, m,/m=oq/op. Adapted from [76]. 12 H. Morita DOC treatment. Ninomiya ef al. [80] found that colchicine has a unique effect on the recovery. The sugar receptor cell was treated with colchicine (5- 25mM), whose effects were not detected at all unless we examied the recovery from destruction by the succeeding DOC treatment; the recovery was deeply depressed when the receptor mem- brane was pretreated with colchicine of concentra- tions above 5 mM for 2 min. The recovery process clearly consisted of two processes, colchicine de- pendent and colchicine independent ones. The colchicine independent process was fast, approxi- mated by an exponential term with time constant of 15 min; this process is thought to be physical, i.e., desorbtion of DOC monomer from the recep- tor membrane. The colchicine dependent process was rather slow, and its time constant was 50 min wher approximated by an exponential term; this process is thought to be biological, i.e., regenera- tion of the distal segment of dendrite as well as repairs of the receptor membrane. Observing the 0.1 M sucrose response as long as 15 to 25 hr, Ozaki et al. [82] revealed a long lasting effect of the treatment with 25 mM colchicine for 2 min. As shown by Figure 9, after the colchicine- DOC treatment (25 mM colchicine treatment for 2 min succeeded by 0.3% DOC treatment for 2 min), the 0.1 M sucrose response was recovered to about 50% level through the colchicine independ- ent process. This level was kept for 12 hr thereaf- ter. The second step of recovry in response then began, whereas recovery in latency (from the onset of stimulus to the first impulse) occurred 5 hr WW Ww uJ 2 Ww) : 3 2 a. mn Mp) = WW = Ss leq lJ we Tp) (eo) 5 n > Ww (Ss) fod Zé WT colchicine tal > 0OC < = J} < J jeg TIME AFTER OOC TREATMENT Fic. 9. Changes in the response (©) and the latency (@) (from the onset of stimulus to the first impulse) after the colchicine-DOC treatment [82]. earlier. This implies that the effect of the colchi- cine terminated 7 hr after the treatment. The effect of colchicine was also detected with- out DOC treatment. The 0.1 M sucrose response decreased 12 hr after the single colchicine treat- ment, but recovered to the normal at 4 hr thereaf- ter. When the colchicine treatment was repeated every 3 hr, the response was reduced to 50% at 16 h after the first treatment. Calculation on the diffusion equation showed that the concentration of colchicine in the receptor cell was at most 5.6 10~° M at 3 hr after the treatment. The remaining effect of colchicine at such an extremely low concentration could not be explained unless we assumed involvement of microtubule system. When microtubules were disassembled into tubu- lin dimers by binding to colchicine, the resultant colchicine-tubulin dimers may cap the growing end of microtubule to prevent its reassembly even at extremely low concentrations of free colchicine, as postulated by Margolis and Wilson [83]. It was also ascertained that vinblastine had the effect but lumicolchicine did not [80]. Thus, it is most likely that microtubules is involved in the maintenance of © the receptor membrane as well as the distal seg- ment of receptor cell. As above, electrophysiology tells us living states of the receptor cell in real time. The long latency in response and lack of the initial phasic response (Fig. 10a) pointed out existence of a distance be- tween the tip of sensillum and that of distal seg- ment of the cell which had recovered the response after the colchicine-DOC treatment. This was ascertained with electron microscopy [82], and led us to calculate the actual concentration of sucrose (C’) at the tip of the distal segment at the begging of recovery. Further, assuming the time constants of the colchicine-independent and -dependent re- covery processes as 15 (7) and 50 min (7), respec- tively, and using equations (1)-(5), we can calcu- late the time course of recovery, i.e., the ratio V’,(t)/V,, where V’,(t) is the response recovered with time after the colchicine-DOC treatment. Thus, Ninomiya et al. [80] obtained the theoretical curves of recovery process as shown in Figure 11. The agreement with the experiments is excellent: after the same treatment (DOC treatment without colchicine), the recovery is seen slower in 0.01 M Insect Contact Chemoreception 13 betore 1 mV after a MTA TOYA UU LA) LLU LaLa nilbal bd san cabin 01s ot pape) x abt ee ded ae Di all Lh tt ical lh ali Fic. 10. Typical DC records of the 0.1 M sucrose responses, obtained before and after the colchicine-DOC treatment (continuous a to b). RESPONSE RELATIVE % 30 60 90 min TIME AFTER BEGINNING OF RECOVERY Fic. 11. Theoretical curves and results of experiments. (©) 0.1M sucrose response, pretreatment without colchicine; (@) 0.01 M sucrose response, without colchicine; (x) 0.1M sucrose response, pretreat- ment with 0.5mM colchicine; (A) 0.1 M sucrose response, with 5 mM colchicine [80]. sucrose response than in 0.1 M sucrose response; these recovery processes are described by the theoretical curves with the same set of parameter values (7}=15 min, »m=50 min, C’/C=0.6, and The horizontal bar represents the period of stimulus [80]. sg/G=4). Figure 12 shows the concentration-response curves before and after the DOC-colchicine treat- ment. The concentration of colchicine was so high that the colchicine-dependent recovery was com- pletely suppressed. The averaged values of K, (+ SEM) were 14+4 and 75+4 mM, respectively, before and after the treatment. The above set of the parameter values determines in turn the values of several constants: K=70+20 mM, (sg/G)’= 0.56 (the value of sg/G after the treatment), V’,/ V,=0.45 (V’, is the maximum response after the treatment). The value of 0.45 is somewhat lower than that of experiments, but is located within the standard error. In fact, if we take the value K=90 mM, the value V’,/V, becomes 0.625, which is the same as the result of experiments in Figure 12. As mentioned above, the colchicine-DOC treat- ment produces a preparation suited for reconstruc- tion of the receptor membrane in vivo. Pretreat- ment with colchicine (25 mM, 2 min) completely depresses the colchicine dependent recovery pro- cess after destruction by DOC (7.2 mM, 2 min) treatment, and retains the 0.1 M sucrose response 14 H. Morita RESPONSE RELATIVE 1073 1072 10° mM 1@! CONCENTRATION OF SUCROSE Fic. 12. Concentration-response curves for sucrose stimulation before (@) and after (©) the colchicine (25 mM)-DOC treatment. theoretical curves after equations (2), (4) and (5). the treatment [80]. to a level below a half of the intact response. Ninomiya et al. (unpublished) incubated the sensil- lar tip for 20 min, after the colchicine-DOC treat- ment, with microsomal fraction (P-fraction) which was prepared from the homogenate of labella of the blow fly. This incubation improved the recov- ery especially in the latency in response. However, when the sensillar tip was incubated with tubulin- tyrosine ligase, the recovery was significantly im- proved in the magnitude of response. In other words, P-fraction seemed to elongate the des- tructed distal segment, while tubulin-tyrosine ligase seemed to increase the number of receptor molecules. In this review, I described and discussed only a few selected subjects. For a comprehensive review before 1976, see reference [84]. Each points is the mean (+ SEM) of six preparations. The smooth curves represent the Arrows indicate K, values before (14 mM) and after (75 mM) ACKNOWLEDGMENTS I would like thank Drs. H. Kijima, I. Shimada, T. Amakawa and M. Ozaki for reading the manuscript and valuable comments. REFERENCES 1 Granit, R. (1947) Sensory Mechanisms of the Re- tina. Oxford Univ. Press, London. 2 Katz, B. (1950) J. Physiol., 111: 261-282. 3. Davis, H., Tasaki, I. and Goldstein, R. (1952) Cold Spr. Harb. Symp. Quant. Biol., 17: 143-154. 4 Alvarez-Buylla, R. and Ramirez de Arellano, J. (1953) Am. J. Phsiol., 172: 237-244. 5 Gray, J. A. B. and Sato, M. (1953) J. Physiol., 122: 610-636. 6 Eyzaguirre, C, and Kuffler, S. W. (1955) J. Gen. Physiol., 39: 87-119. 7 Eyzaguirre, C, and Kuffler, S. W. (1955) J. Gen. Physiol., 39: 121-153. 8 Edwards, C. and Ottoson, D. (1958) J. Physiol., 33 34 35) 36 37] Insect Contact Chemoreception 15 143: 138-148. Morita, H. (1959) J. Cell. Comp. Physiol., 54: 189- 204. Fesenko, E. E., Kolesnikov, S. S. and Lyubarsky, A. L. (1985) Nature, 313: 310-313. Nakamura, T., and Gold, G. H. (1987) Nature, 325: 442-444. Avenet, P., Hofmann, F. and Lindemann, B. (1988) Nature, 331: 351-354. Tonosaki, K. and Funakoshi, M. (1988) Nature, 331: 354-355. Morita, H. and Shiraishi, A. (1985) In “Compre- hensive Insect Physiology, Biochemistry and Phar- macology” Ed. by G. A. Kerkut and L. I. Gilbert. Pergamon Press, Oxford, vol. 6: 133-170. McIndoo, N. (1914) J. Exp. Zool., 16: 265-346. von Frisch, K. (1919) Zool. Jahrb., 37: 1-238. von Frisch, K. (1921) Zool. Jahrb. Abt. f. allg. Zool. u. Physiol., 38: 449-516. von Frisch, K. (1935) Z. vergl. Physiol., 21: 1-156. Mirmnich Dre. (1921) Jr Exp: Zool-. 33> 173=203. Minnich, D. E. (1926) Biol. Bull., 51: 166-178. Minnich, D. E. (1926) Anat. Rec., 34: 126. Minnich, D. E. (1929) Z. vergl. Physiol., 11: 1-55. Minnich, D. E. (1931) J. Exp. Zool., 60: 121-139. Dethier, V. G. (1955) Quart. Rev. Biol., 30: 348- SWAN. Pfaffmann, C. (1941) J. Cell. Comp. Physiol., 17: 243-258. Nodai, H. (1953) Kyushu Shigaku Z., 6: 1-3. (In Japanese) Morita, H., Doira, S., Takeda, K., and Kuwabara, M. (1957) Mem. Fac. Sci. Kyushu Univ. Ser. E., 2: 119-139. Hodgson, E. S., Lettvin, J. Y. and Roeder, K. D. (1955) Science, 122: 417-418. Morita, H. and Yamashita, S. (1959) Science, 130: OW. Tateda, H. and Morita, H. (1959) J. Cell. Comp. Physiol., 54: 171-176. Hodgson, E. S. and Roeder, K. D. (1956) J. Cell. Comp. Physiol., 48: 51-75. Dethier, V. G. (1956) Amer. Inst. BIol. Sci., Publ. No. 1: 1-30. Zacharuk, R. Y. (1985) In “Comprehensive Insect Physiology, Biochemistry and Pharmacology” Ed. by G. A. Kerkut and L. I. Gilbert. Pergamon Press, Oxford, vol. 6: 1-69. Wolbarsht, M. L. (1958) J. Gen. Physiol., 42: 413- 428. Thurm, U. (1972) Olfaction and Taste, 4: 95-101. Wieczorek, H. (1982) J. Comp. Physiol. A, 148: 303-311. Kuppers, J. (1974) In “Mechanoreception” Ed. by 38 39 40 4] 42 43 44 45 46 47 48 49 50 Si 52 33 54 SS) 56 D7) 58 59 60 61 62 J. Schwartzkopff, pp. 387-394. Abh. Rhein. Westf. Akad. Wiss. Opladen. McIver, S. B. (1985) In “Comprehensive Insect Physiology, Biochemistry and Pharmacology” Ed. by G. A. Kerkut and L. I. Gilbert. Pergamon Press, Oxford, vol. 6: 71-132. Menco, B. Ph. M. and van der Wolk, F. M. (1982) Cell & Tissue Res., 223: 1-27. Morita, H. (1969) Olfaction and Taste, 3: 370-381. Martin, A. R. (1955) J. Physiol., 151: 1-23. Morita, H. (1991) Biomed. Res., 12, Supplement 2: 225-227. Beidler, L. M. (1954) J. Gen. Physiol., 38: 133-139. Morita, H. and Shiraishi, A. (1968) J. Gen. Phy- siol., 52: 559-583. Kijima, H. and Kijima, S. (1978) J. Theor. Biol., 71: 567-585. Kijima, H. and Kijima, S. (1980) J. Theor. Biol., 82: 425-463. Kijima, H., Kijima, S., Noguchi, M. and Morita, H. (1980) J. Theor. Biol., 83: 131-161. Kijima, H. and Kijima, S. (1989) In “Biological Transduction Mechanisms” Ed. by M. Kasai, et al. Japan Sci. Soc. Press, Tokyo/Springer-Verlag, Ber- lin. pp. 17-45. Amakawa, T., Ozaki, M. and Kawata, K. (1990) J. Insect Physiol., 36: 281-286. Kijima, H., Nagata, K., Nishiyama, A. and Morita, H. (1988) J. Gen. Physiol., 91: 29-47. Neher, E. and Stevens, C. F. (1977) Ann. Rev. Biophy. Bioeng., 6: 345-381. Shimada, I. (1991) In “Molecular Mechanism of Chemosensory Transduction” (Science Report for Grant-in-Aid from Min. of Educ. Sci. and Cult. Jap., 63304063) Ed. by H. Kijima, pp. 66-67. Pugh, E. and Altman, J. (1988) Nature, 334: 16-17. Amakawa, T. and Ozaki, M. (1989) Physiol., 35: 233-237. Evans, D. R. (1963) Olfaction and Taste, 1: 165- 176. Shimada, I., Shiraishi, A., Kijima, H. and Morita, H. (1974) J. Insect Physiol., 20: 605-621. Hanamori, T., Shiraishi, A., Kijima, H. and Morita, H. (1974) Chemical Senses and Flavor, 1: 147-166. Jakinovich, W. Jr., Goldstein, I. J., Baumgarten, R. J. von and Agranoff, B. W. (1971) Brain Research., 35: 369-378. Shiraishi, A. and Kuwabara, M. (1970) J. Gen. Physiol., 56: 768-782. Shimada, I. and Isono, K. (1978) J. Insect Physiol., 24: 807-811. Shimada, I. and Tanimura, T. (1981) J. Gen. Physiol., 77: 23-39. Shimada, I., Horiki, H., Ohrui, H. and Meguro, H. (1985) J. Comp. Physiol., 157: 477-482. J. Insect 63 64 65 66 67 68 69 70 mh M2 73 74 16 Shimada, I., Ohrui, JASTS, 23: 99-102. Wieczorek, H. and Koppl, R. (1978) J. Comp. Physiol., 126: 131-136. Dethier, V. G. (1955) J. Econ. Entmol., 48: 235- 239. Hansen, K. (1969) Olfaction and Taste, 3: 383-391. Kijima, H., Koizumi, O. and Morita, H. (1973) J. Insect Physiol., 19: 1351-1362. Kawabata, K., Kijima, H., Shiraishi, A. and Mori- ta, H. (1953) J. Insect Physiol., 19: 337-348. Koizumi, O., Kijima, H. and Morita, H. (1974) J. Insect Physiol., 20: 925-934. Amakawa, T., Kijima, H. and Morita, H. (1975) J. Insect Physiol., 21: 1419-1425. Kiima, H., Amakawa, T., Nakashima, M. and Morita, H. (1977) J. Insect Physiol., 23: 467-479. Morita, H., Enomoto, K., Nakashima, M., Shima- da, I. and Kijima, H. (1977) Olfaction and Taste, 6: 39-46. Nakashima, H, Enomoto, K., Kijima, H, and Mori- ta, H. (1982) Insect Biochem., 12: 579-585. Dethier, V. G. (1990) Ann. Rev. Neurosci., 13: 1- 13). H. and Meguro, H. (1989) H. Morita 73 76 77 78 79 80 81 82 83 84 Hanamori, T. (1976) Chemical Senses and Flavor, 224229239) Ozaki, M. (1988) Zool. Sci., 5: 281-290. Hara, M. (1983) J. Insect Physiol., 29: 113-118. Takeo, T. and Nakamura, S. (1972) Arch. Biochem. Biophy., 153: 1-7. Takeo, T. and Nakamura, S. (1978) In “Affinity Chromatography” Ed. by O. Hoffmann-Ostenhof, M. Breitenbach, F. Koller, D. Kraft and O. Schneider, Pergamon Press, Oxford and New York, pp. 67-70. Ninomiya, M., Ozaki, M., Kashihara, Y. and Mori- ta, H. (1986) J. Gen. Physiol., 87: 1003-1016. Shimada, I. (1975) J. Insect Physiol., 21: 1565- 1574. Ozaki, M., Ninomiya, M., Kashihara, Y. and Mori- ta, H. (1986) J. Gen. Physiol., 87: 533-549. Margolis, R. L. and Wilson, L. (1977) Proc. Nat. Acad. Sci., 74: 3466-3470. Dethier, V. G. (1976) The Hungry Fly. Harvard University Press, Cambridge, Massachusetts and London, England. ZOOLOGICAL SCIENCE 9: 17-36 (1992) © 1992 Zoological Society of Japan REVIEW Development of the Hypothalamic Monoaminergic System in Ontogenesis. Morpho-functional Aspects MICHAEL V. UGRUMOV Institute of Developmental Biology, U.S.S.R. Academy of Sciences, Moscow 117808, U.S.S.R. CONTETS |UCROGINCTOMN 35 eo aae dancer Anarene teese On aameee ore acl rma erin ler SOE aR Ye Sc on a 5 Se a ee 17 Dov ClOpmei tol MONOAMINE SICISVStCMIS epee cs-cee- de eak ona vedas Meade deans on dep dee ese eae a oe ne 18 Pe ALCCHOLATMINEL PLC SV SLOT gai dels, lesa niiae otto. clad gout ncrec oueten hats aa Ne ns aoe woes ncidetgd ove Sota 18 AME ANTIG ITIVE CLOMICS erence pn reece erste Bohne eas neee dee so sew zh oa hc MAAC Ped asicceee a hee ge 18 bb) ee Monpho-tunctional Chanactenisttcs) s:c.9. 4-6. esas nae ea sche onecePeosadee terete een aco 21 PPPS CTOLOMMERLIC) (Sartell SV STEM Aer et Ssh eon eh Sac ca ben taseledat quel ca cneeedeateeM eee ae de eraandecsasetan: 24 AEP NTCMICE CCOMICS cd. ISSUER ERE ee Uae cece ct, com airtime WIE L 4, I PTT Te ateinniieyt 24 b) ae MonphotunctiomaluchanactenmiStt Sey. sass. ck cee ge ee seesece sn ctcce cee eeernc teach emeetoe cnclsecacee oh 26 III. Neurohumoral control of development of monoaminergic systeMS ...................0ccecee eee ee sees 26 Pa @ ALC CHOLATINITICT CIC SV StEMM i nan tea usta Seater: snes cn canie conte Ee nick Goble state acinar iemichewiee Suaiacih 26 A) MES ONUAINSTCTOINS ee ere ee anette ee het ade seek ete MMHG ch aioe on sr Aw Mya eM oh cred ect Nee 26 ) MOMS TAMORIMOMES A Fee Set ss heer Mesa s es dace cincscmiin Lutidia sens dec Se MO Ne acilte ac cnet ea eteiacronre eae Diy PRES CTOLOMIMET CICS VSUC INP eee ese EERE fence Se alts Feo te vale g TUT es « Saa aielhah seu Wee Gok bane MRE a cecal 28 Zi) SKEET SUETTOTONS, sero o cts se TE SPF cs SERIE ISIE A aN 5 28 MeO Chemhonmonese. seat saee er eternity ten sah none seca oreds uate an neh nana ee car ae 29 IW FUNC HONAlSISMICAaNnce Ol MONOAMINES 7. 2s. eke faeces snes eect es onan sect ve cae coc tue Wan cesecsceeeSeone 29 Lo (CRUSE NOLETTNNT Ec iaos8 3p ee ai ok ee Sake Sean Ae ae aCe Un Es ee Rea HTT ne eee 29 ORES CRO COMMU A eG eRe RRR TNL MOAN Lo bea Sssia chase site walelddie aaivia MERE OAN GMA ong od ecles duane see 32 Whe (COWNGIIISTIOINS “gescbecigtacaededs dAeG Seces ATOR ARON GAS 6 Se MISES ie Morr at recive ri nana ane an er ea 38) PNERERE INCE SMR ey es tere Pete eB Pe cate eer acs atic Aescial corso stores Sesoine della leas elaratinate eeceeae a te 32) I. INTRODUCTION Over the last two decades a large body of evidence have been accumulated on the key role of hypothalamic monoamines (MAs): dopamine (DA), noradrenaline (NA), adrenaline (A) and serotonin (5-HT), in the neuroendocrine regula- tion [1, 2, 3]. The MAs play the role of: a) Neurotransmitters controling the functional activ- ity of the target neurons; b) Neuromodulators regulating the release of neurohormones from adjacent neurosecretory axons; and c) Neurohor- Received November 15, 1991 mones delivered by portal blood and cerebrospinal fluid (CSF) to the target adenohypophysial cells List of addreviations AD-adrenaline, adrenergic; CA-catecholamine, catech- olaminergic; CSF-cerebrospinal fluid; DA-dopamine, dopaminergic; E-embryonic day; EM-electron micros- copy, electron microscopic; 5-HT-serotonin, seroto- ninergic; IP-immunoposivive; LM-light microscopy, light microscopic; MA-monoamine, monoaminergic; ME-median eminence; MFB-medial forebrain bundle; (M)PA-(medial) preoptic area; n-nucleus, nuclei; NA- noradrenaline, noradrenergic; P-postnatal day; pCPA- p-chlorophenylalanine; SCN-suprachiasmatic nucleus; SS-sexual steroids. 18 M. V. UGrumov / IV Mo) oo Se | ee ae 7 ! Vi Fic. 1. Schematic representation of participation of hyporthalamic monoamines in the neuroendocrine regulation. I-serotoninergic neurons (SE) of raphe nucleus; II-noradrenergic neurons (NA) of locus coeruleus; III-hypothalamic neurosecretory nuclei with dopaminergic (DA) and peptidergic (PE) neurons; IV-median eminence with capillary loops of primary portal plexus; V-portal veins; VI- adnohypophysis with the secondary portal plexus; V-ventricle. Specialized contacts of monoaminergic axons; small arrowhead-axo-ventricular; large arrowhead-synapse; middle arrow-axo-axonic; large arrow-axo-vascular. small arrow-direction of blood stream in hypophysial portal circulation. and neurons (Fig. 1) [4]. Moreover, MAs provide their influence on the development first of the whole embryos and then of the central nervous system including hypo- thalamus long before the onset of the neurotrans- mitter function [5, 6, 7, 8]. In turn, the differentia- tion of MA neurons and expression of their feno- type are under the neurohumoral control [9]. This review attempts to summarize the results mainly on the development of the hypothalamic MA system and to a lesser extent on its role in the neuroendocrine mechanisms in ontogenesis. Il. DEVELOPMENT OF MONOAMINERGIC SYSTEMS 1. CATECHOLAMINERGIC SYSTEM a) Architectonics Cell bodies. The combination of long survival [°>H]thymidine autoradiography with immunocytochemistry of tyrosine hydroxylase (TH), the key-enzyme of the catecholamine (CA) synthesis, has shown that the genesis of the hypothalamic CA neurons in rats occurs from the 11th fetal day (E11) until E17. The neurons of the dorsal accumulations (zona incerta, etc.) are mainly originated at E12, or even — earlier, while those of the ventral accumulations (arcuate region) later, at E15 [10]. TH immunocytochemistry occured to be the most useful approach in the study of the dif- ferentiation of the CA neurons, as this enzyme appears early in ontogenesis and is widely distri- buted through neurons [11]. First rare TH im- munopositive (IP) neurons were observed in the lateral hypothalamus at E13 (Fig. 2) [12] that is in agreement with the biochemical observation of TH activity in this region [13]. Over subsequent 2-3 days the THIP neurons increased in number giving rise to bilateral dorsomedial accumulations (Fig. 3) [12, 14]. From E18 onwards, THIP neurons were widely distributed concentrating in some hypothalamic nuclei (n.) (Figs. 4, 5). In the ante- rior hypothalamus, occasional neurons or their small clusters were regularly seen in the medial preoptic area (MPA), suprachiasmatic and ante- rior hypothalamic n. [12]. A considerable number of THIP neurons are concentrated dorsomedially overlapping the paraventricular n., rostrally, and zona incerta, dorsomedial and posterior hypotha- Monoaminergic System in Ontogenesis 19 Fics. 2, 3. lamic n., caudally. Ventromedially these clusters are connected with a massive periventricular accu- mulation of THIP neurons first appeared on E20 2k In the mediobasal hypothalamus two small ven- tral bilateral accumulations of THIP neurons observed only in perinatal rats (E18-P3) are local- ized lateral to the median eminence (ME) (Fig. 4). By P3 only few neurons keep their initial position while a similar neuronal accumulation appears in the arcuate n. (Fig. 5) [12, 14]. It is still uncertain whether this is a result of neuronal migration or transient expression of TH synthesis. Thus, CA neurons are originated long before birth. Over perinatal period they occupy their definitive positions concentrating mainly in zona incerta, periventricular and arcuate n., as in adults Schematic representation of catecholaminergic (THIP) neurons (black dots) in the hypothalamus and septum in fetal rats at E13 (2) and E15 (3). Each dot represents 2-3 THIP neurons. fm-foramen of Monro; lv-lateral ventricle; of-optic fissure; or-optic rudiment; p-pituitary; pa-preoptic area; v-3rd ventricle. (Modified [12]). [15]. Nerve fibers. One of the most important characteristic of the development of the CA system is the growing of the THIP fibers to the target neurosecretory nuclei and circumventricular organs. THIP fibers belong- ing to neurons of mesencephalon and pons [11, 16] first enter the hypothalamus via the medial fore- brain bundle (MFG) at E13. Then, they project into the optic chiasma and tracts as well as into the primordium of the ME from E15, the anterior comissure, the diagonal band and septum from E18, and into zona incerta, dorsomedial and para- ventricular n., etc. perinatally [17]. It is obvious that the places of the high concentrations of THIP neurons become occupied by networks of THIP 20 M. V. UGRUMOV Fics. 4, 5. neurons in the hypothalamus and septum in fetuses at E18 (4) and in neonatal rats at P9 (5). a-arcuate nucleus (n.), ac-anterior comissure; ah-anterior hypothalamic n.: db-diagonal band; dm-dorsomedial n.: It-lamina terminalis; mb-medial forebrain bundle; me-median eminence; oc-optic chiasma; of-optic fissure; on-optic nerve; ot-optic tract; pe-periventricular n.; pv-paraventricular n,; s-septum; sc-suprachiasmatic n.; so-supraoptic n.; vm-ventromedial n.; zi-zona incerta. For other abbreviations see Figs. 2, 3. (Modified from [12, 47].) fibers, however, the latter are also observed in the hypothalamic regions almost lacking the THIP neurons [17]. In the septal region and diagonal band, the THIP fibers terminate on luteinizing hormone releasing-hormone neurons in neonatal rats [18]. The number of smooth neurons innervated by THIP fibers remains at the constant level from P2 until P90, while that of neurons possessing spine- like processes and innervated by THIP fibers tri- pled [18]. This is in agreement with physiological observations of the onset of NA stimulation of the luteinizing hormone releasing-hormone release in prepubertal (P29) female rats [19]. In the SCN, though THIP fibers first appear at E18, their concentration remains insignificant until the end of fetal life. Over the first nine days of postnatal life, the concentration of THIP fibers increases abruptly, thus, highly exceeding that in the SCN of adults [20]. Further EM analysis identified many THIP fibers as axons contained synaptic and dense core vesicles. By the end of intrauterine development the THIP axons make immature synapses (presynapses) with the im- munonegative dendrites. After birth, first the symmetric (at P2) and then asymmetric (at P9) synapses appear. CA fibers might be suggested as a regulator of the presumptive oscillators, pep- Monoaminergic System in Ontogenesis pM tidergic neurons, which become functionally active perinatlly [21]. As to the paraventricular n., few THIP fibers arrive there even in fetuses (E19, 22; E17, 23), though they surround neurosecretory neurons only one-two weeks after birth [17, 23]. In addition to DA and NA, AD fibers also contribute to this innervation from Pl [24]. A number of EM double-labeling studies showed that in adults the CA fibers innervate vasopressin- [25] thyrotropin releasing-[26] and corticotropin releasing-[27] hor- mone-producing neurons in this region. The growing of the CA fibers to the ME is considered as the essential characteristic of the maturation of the tubero-infundibular system. THIP fibers first terminate there at E18. After the final settling of THIP neurons in the arcuate n., the ME receives the increased number of THIP fibers mainly from this region [14, 17] reaching maximum by puberty [22, 28, 29, 30]. From the physiological studies in adults, DA modulates the peptide release from the ME [31, 32]. Therefore, the establishment of the close relations between CA and peptidergic axons is of great importance. Double labeling technique showed first at LM [29, 33] and then at EM level [34], that the axo-axonal contacts between luteiniz- ing hormone-releasing hormone and CA fibers in the ME are established from P8 until P14, while the DA “innervation” of somatostatin axons is delayed for a week [29]. Although THIP fibers regularly spread to the periventricular region after E15, they reach the ventricular lumen only at E18-E20. This suggests the release of CA into the CSF, followed by their wide distribution through the brain. After E20, the frequency of these contacts drops and becomes practically absent in adults [17]. The same tenden- cy has been marked in phylogenesis [35]. Thus, from E13 the hypothalamus of rats is innervated by the CA fibers either belonging to the hypothalamic neurons or arriving via the MFB from the outside. Since this time, CA fibers progressively innervate such target regions as the diagonal band, septum, SCN, paraventricular n., etc. Besides, CA fibers sprout to the ME and the 3rd ventricle. b) Morpho-functional characteristics According to LM immunocytochemical observa- tions, THIP neurons undergo striking morpho- logical modifications relating to their differentia- tion. The earliest neurons (Fig. 6a) possess one or two short unbranched processes terminating with the growth cone. Two days later, in addition to uni- and bipolar, multipolar neurons first appear. From E18 onwards, THIP cell bodies increase in size and their processes in length [12]. Although the majority of THIP neurons undergo similar morphological differentiation, from E20 at least two neuronal populations could be recognized [12]. The first one consists of small uni- and bipolar neurons with short and narrow unbranched processes. These neurons are localized in ventral, mainly, the mediobasal hypothalamus (arcuate n.) (Fig. 6b, c). The second population includes large multipolar neurons with long highly ramified pro- cesses concentrated dorsally presumably in the zona incerta (Fig. 6d)[12]. Up-to-date, only THIP neurons of the SCN and arcuate n. (first population) have been studied in ontogenesis at the EM level [36]. The authors have failed to find out any principal (qualitative) differences in the ultrastructure of THIP neurons from E22 until P21. They are characterized by the relatively large nucleus and scanty cytoplasm as well as by the well-diveloped Golgi complex (Fig. 6e). As the differentiation of THIP neurons proceeds, their size increases slightly, the nuclei become indented, the dense core vesicles appear, thus, showing the secretory activity. The THIP neurons, at least from E22, are rarely innervated by immunonegative axons containing dense core vesicles and synaptic vesicles. These axons give rise mainly to axo-dendritic immature synapses (presynapses) in fetuses, transformed into the sym- metric and rarely asymmetric synapses after birth (Fig. 6f). The synaptogenesis continues at least till P21 that is manifested in the increase of both the number and complexity of synapses. Biochemical and histofluorescent techniques have provided the information on the synthesis of the final product, CA, and on their transfer via axons towards the target regions. According to biochemical data, TH activity first detected at E13 Di. M. V. UGruMov Fic. 6. THIP (a, b, d-f) and [*H]dopamine-labeled (c) catecholaminergic neurons and fibers in the hypothalamus of fetal (a) and postnatal (b, c, e-P9; f-P21) rats at light (a-d) and electron microscopic (e, f) levels. V-3rd ventricle; small arrow-axo-dendritic synapse; large arrow-randioactively labeled neuron; arrowhead-radioactively labeled fiber. a-ventral hypothalamus; b, c-arcuate n.; d-zona incerta; e, f-suprachiasmatic n. For other abbreviations see ipa 4 25. Shows a 12-fold increase prenatally and, then, [13]. 4-fold rise in four postnatal weeks. The activity of With histofluorescent technique first CA con- this enzyme is highly stimulated in embryonic taining neurons were detected in the hypothalamus tissue culture by depolarizing agent, veratridine at E12 [37]. By E18 the frequency of the fluores- Monoaminergic System in Ontogenesis 2 cent neurons increases significantly. In the anter- ior hypothalamus they are mainly concentrated in the paraventricular and periventricular n., while in the middle hypothalamus they are localized dorsal- ly, in the zona incerta and dorsomedial n. as well as in the mediobasal hypothalamus [28, 37]. From E18 to the puberty the modifications of the CA system are rather quantitative than qualita- tive. Until E20 the intensity of fluorescence in- creases progressively, followed by its drop by P9 and then it abruptly increase by P21 [28]. Accord- ing to biochemical, HPLC data, the content of DA at E21 reaches its level at P11. On the contrary, the concentration of NA increased four-fold over the same perinatal period. Aderenaline remains at the lowest level comparing with other CA both in fetus and neonates [38]. From comparison of histofluorescent and biochemical data follows that in parallel to the gradually increasing synthesis of CA their release predominates at the end of fetal life or soon after the birth. The specific uptake and K*-stimulated release of CA is the next essential functional characteristic of CA neurons. First radioautographic signs of the specific uptake of intraventricularly injected Birth = (e) fo) (e) a (e) (e} (?H)dopamine uptake (com/mg tissiuve) 45th 15th 16th 18th 20th 9th Age (days) Fic. 7. (Modified [28]). (°H)dopamine release (cpm/mg/ftaction) [H]DA by hypothalamic CA neuronal elements have been detected in fetal rats on E18 [39]. Further biochemical “isotopic” study in vitro has shown the simultaneous onset of the uptake and K*-stimulated Ca*?-dependent release of CA at E16 (Fig. 7) [28]. The uptake increased in 2,6 times for subsequent two days and, then, doubled between E20 and P9 reachng the adult level. As to K*-stimulated Ca*?-dependent release of CA, it increases considerably on E17 and remains at the same level in older fetuses and neonates, while doubled between P9 and P45 (Fig. 7) [28]. Thus, four stages in the development of the hypothalamic CA system could be distinguished. At the first stage (E12-E13), rare CA neurons and fibers of the MFB appear. The second stage (until E16) is characterized by the most intense neuro- nogenesis, followed by the onset of CA synthesis, uptake and K *-stimulated release. Over the third stage E16-E18, the final settling of CA neurons occurs simultaneously with the stimulation of synthesis, uptake and ion regulating release of CA. The fourth stage (until puberty) is mainly character- ized by completing of the CA neuron differentia- tion and the establishment of the axonal pathways Kt KiCa=O Kt . i ’ { * 15th day x 16th day o Y9thday di e e 45th day 60 N 30 | : ee me Nene” NON oo Ne e Ay, 40 50 60 70 8C 90 Time (min) Specific [7H]dopamine uptake and K *-stimulated release in the hypothalamus of fetal and postnatal rats. 24 M. V. UGRUMOvV for CA transfer to the target neurons and circum- ventricular organs. 2. SEROTONINERGIC SYSTEM a) Architectonics Cell bodies. Dorsal and median raphe n. are known to be the main sources of the hypothalamic 5-HT innerva- tion in adult mammals [40]. A number of im- munocytochemical [41, 42, 43] studies have shown that 5-HT neurons in the raphe n. of rats first appear at E12-E13. This is followed by their differentiation and axonal growing to the target regions. The question on the hypothalamic localization of the 5-HT neurons still remains opened. First radioautography has shown that some hypo- thalamic neurons have an ability for specific up- take of [PH]5-HT following its intraventricular injection [41]. These neurons are localized in the SCN of fetuses (E18) and in the dorsomedial n. of neonatal (P9) (Fig. 8a) [41] and adult [44] rats. Further 5-HT immunocytochemistry attempted to detect either 5-HT accumulating or 5-HT synth- esizing neurons in the hypothalamus in ontogene- sis. As 5-HTIP cells have not been observed in the hypothalamus of intact animals [41, 45, 46, 47], the 5-HT synthesis and its reuptake was stimulated by the pretreatment of fetuses (E18), neonatal (P9), and adult rats with L-tryptophan, precursor of 5-HT synthesis, and pargyline, inhibitor of monoamine oxidase [45, 47, 48]. In these experi- ments, two populations of 5-HTIP neurons, in the anteriolateral hypothalamus and the dorsomedial *~ % n., were observed in fetuses and neonates (Fig. 8b,c) [47] while in adults they were localized only in the dorsomedial n. [45]. In fetuses both populations, but in neonates only the first one, consist of intensely immunostained bipolar elongated neurons with long processes (Fig. 8b). In neonates [47] and adults [45] the neurons of the dorsomedial n. characterized by a fairly weak immunostaining, are oval, small in size with one or two short processes (Fig. 8c). 5-HT immunostaining provoked by pharmacological pretreatment has been prevented by the prelimin- ary infusion of the uptake inhibitor, fluoxetine. It means that immunostaining is accounted for rather by the ability for specific uptake of 5-HT from the enviroment than by its synthesis from L- tryptophan catalyzed by tryptophan hydroxylase [47, 48]. On the contrary, decarboxylation of 5- hydroxytryptophan to 5-HT with decarboxylase of the aromatic amino acids is rather probable, as both neuronal populations become 5-HT immuno- positive following pretreatment of aminals with pargyline and 5-hydroxytryptophan instead of L- | tryptophan [47]. Such characteristics of 5-HTIP neurons as an ability for the specific uptake and decarboxylation of the 5-HT precursor made them similar to the cultivated in vitro hypothalamic 5-HT-like neurons of fetal mice [49], on the one hand, and to the APUD cells [50], on the other. Over the last decade, the attention has been focused on the origin, functional significance and fate of these cells, however, this problem is far from understanding [45]. According to our sugges- tion, the 5-HTIP neurons might serve as a tempor- Fic. 8. [°H]5-HT-labeled (a, arrowhead) and 5-HTIP (b, c) neurons in the anteriolateral region (b, large arrow) and dorsomedial n. (a, c; small arrow) of the hypothalamus in rats at P9. Monoaminergic System in Ontogenesis WS al store of 5-HT either released from the 5-HT axons of the MFB, or circulated in the CSF [47]. This is in agreement with phylogenic observations of 5-HTIP neurons in the periventricular hypotha- lamic area in intact lower vertebrates. These cells are supposed to provide an intermediate link in the neurohumoral regulation of the neuroendocrine centers [35]. As to the fate of the 5-HTIP cells in the anterolateral hypothalamus of fetal and neonatal rats (see above), probably, they repre- sent the transient neuronal population partly ex- pressed fenotype of 5-HT neurons [47]. In conclusion, the hypothalamus of fetuses and neonates contains two populations of neurons which become 5-HTIP after the pargyline and L-tryptophan pretreatment, apparently due to the 5-HT absorption from the environment. More- over, these cells are characterized by the specific uptake of 5-HT and the capacity to decarboxylase 5-hydroxytryptophan into 5-HT. Thus, the 5-HT neurons of the raphe nucleus are the main source of the hypothalamic innervation, while the presence of the hypothalamic 5-HT neurons is still under question. Nerve fibers. Although 5-HTIP neurons first appear in the raphe n. of the rat as early as E12, their axons first 6150 0 ~ oO a) = =) ~ i 2 D Birth c £ £ 1000 2409 4 _ 2 D o o xs = © v Ss E s i 500 + 2 50 2 Se Te SOD 16th 18th 20th 9th 45th Age (days) reach the hypothalamus at E14-E16 [41, 42, 43, 46]. The concentration of the 5-HT fibers in- creases abruptly from E16 until E18, that is parti- cularly evident for the perichiasmatic area. Since this time, 5-HTIP fibers invade preoptic area (PA), septum and the diagonal band [42], 1.e., those regions which already contain their targets, e.g., luteinizing hormone releasing-hormone neurons [51]. The special attention has been paid to the 5-HT innervation of the SCN because of its role in the control of the circadian rythmic activity of neuroendocrine and other functions [52]. Accord- ing to LM radioautographic and immunocyto- chemical data, 5-HT fibers first invade the SCN at E18 [41, 53]. By the end of fetal life, the concen- tration of 5-HT fibers particularly in its ventro- lateral region increases considerably reaching the adult level around P9 [41, 42, 46]. Further EM analysis showed that in the oldest fetuses 5-HTIP axons make axo-dendritic immature synaptic con- tacts (presynapses) which are transformed neona- tally into symmetric, and rarely asymmetric synapses [53]. In evaluating the functional significance of 5-HT input into the SCN, it is necessary to take into account that in adults the 5-HT axons innervate vasoactive intestinal peptide neurons [54] control- Ktca=o Kt # 16th day x 17th day © 9thday e 45th day od “Se Br 40 50 60 70 80 90 Time (min) Fic. 9. Specific [PH]5-HT uptake and K*-stimulated release in the hypothalamus of fetal and postnatal rats. (Modified [57]). 26 M. V. UGRuMov ing their functional activity [55]. These cells are settled in the SCN at E19 [52], and become functionally active as spontaneous oscillators by the end of intrauterine development [56]. In the middle (mediobasal) hypothalamus, 5-HT fibers are widely distributed from E18 [41] invad- ing the arcuate and dorsomedial n. as well as the ME. In the ME, 5-HT fibers terminate in the external zone or even penetrate to the area of the primary portal plexus abutting on the portal capil- laries [41]. . The essential characteristic of the 5-HT system both in pre- and postnatal life is the progressive penetration of 5-HT fibers to the cerebral ventri- cles. 5-HT released to the CSF is believed to reach the target regions of the brain, or is transferred through the ME to the hypophysial portal circula- tion and finally to adenohypophysis [41]. Thus, 5-HT axons first arrived at the hypothala- mus at E14-E16, spread to the target regions over the perinatal period, followed by the establishment of specialized contacts with target neurons (synapses) as well as with the hypophysial portal circulation and cerebral ventricles. b) Morpho-functional characteristics According to radioautographic data, the specific uptake of intraventricularly injected [*H]5-HT by the hypothalamic nerve fibers in vivo is already evident at E18 [41]. Using biochemical “isotopic” technique the specific uptake by the hypothalamic tissue in vitro has been first detected even earlier, at E16, while the [*H]5-HT release by the depolar- izing concentrations of K* in a Ca‘?-dependent manner is evident from E17 [57] (Fig. 9). It means that Ca’*-channels become functionally active and S-HT is intravesicularly stored from E17 onwards. The most rapid increase of the [?H]5-HT uptake occurs from E16 until E18 [57] that is apparently a result of the most intense growing of 5-HT fibers to the hypothalamus as well as of the significant activation of the 5-HT-binding protein synthesis in the brain [58]. From E18 to P45, the value of the specific uptake retains at the constant level. This shows that despite the progressive increase in the number of 5-HT fibers, their number per mg of tissue does not change significantly. In contrast to uptake, the Ca‘?-dependent and K*t-stimulated relase of [*H]5-HT increases considerably over the perinatal period reaching an adult level by P9 (Fig. 9) [57]. Thus, the hypothalamic 5-HT system could be functionally active from E16-E17, i.e., shortly after first arrival of 5-HT fibers, that is manifested by the onset of the specific uptake and K*- stimulated release of 5-HT. It. NEUROHUMORAL CONTROL OF DEVELOPMENT OF MONOAMINERGIC SYSTEMS 1. CATECHOLAMINERGIC SYSTEM a) Sexual steroids Biochemistry The sexual difference in the content of CA, mainly of NA, in the hypothalamus of rats from P10-P12 onwards, suggests the role of sexual ster- oids (SS) in the development of the CA system [59, 60]. This has been checked comparing the content of CA in neonatally castrated and sham-operated male rats as well as in neonatally androgenized females. The dynamic of the NA content in androgenized females resembles that in control males [59, 60], though the maximum has been achieved earlier, by P60. Similarly, the NA level in the castrated males gradually rises until P180 as that in the control females excepted the peak at P60 (Fig. 10a) [59]. In contrast to NA, no clear sexual difference in hypothalamic DA content has been observed in neonates [60, 61]. The castration of males does not modify the DA content until P120, while at P180 its level highly exceeds that in control males and androgenized females. Similarly to NA, the DA content shows a decline in its level with age both in control males and androgenized females (Fig. 10b) [59]. The authors suggested that the initial in- crease of CA content in the control males and androgenized females is accounted for the neona- tal androgen promotion of the growing of the CA fibers. Sexual differences in the DA content were also observed in the arcuate n. and ME: the DA level of males exceeds that in females. The neonatal Monoaminergic System in Ontogenesis 27 Noradrenaline nmol/g wet weight of tissue Age In days Dopamine nmol/g wet welgnt of tissue 120 Age in days Fic. 10a, b. Mean (+SEM) concentrations of noradrenaline (a) and dopamine (b) in the hypothalamus at various postnatal ages. ook <005585 Sh 00i as hae ().001e ns-not significant. Upper panel-male; 1 sham- castrated; M castrated. Lower panel-female; 0 control; M androgenized [59]. castration of males causes the reduced DA level in adults, while neonatally androgenized females show an increased level of this neurotransmitter [62, 63]. Though the mechanism of this action is still uncertain, the effect of SS on the TH activity in ontogenesis is rather probable [13]. Morphology Further immunocytochemical study using anti- bodies to TH and DA-/-hydroxylase attempted to verify the role of SS in the development of the hypothalamic CA system [64]. The perinatal (E16- P10) androgenization of females causes decrease in both the number of THIP neurons and fibers belonged not only to the hypothalamic but also to the mesencephalic neurons. Conversely, postnatal injection of testosterone “masculinizes” only the content of cell bodies. These observations suggest that the critical period of the differentiation of hypothalamic CA neurons corresponds to the neonatal life while that of mesencephalic neurons begins earlier, i.e., in fetal life [64]. According to recent data, sexual dimorphism of DA neurons of diencephalon is first manifested without any influence of SS [65]. Thus, the DA neurons taken into tissue culture from fetuses, at E14, 1.e., before the onset of SS secretion, show more intense outgrowth of processes and higher uptake of [*H]DA in females than in males. The neurons taken into the tissue culture three days later show the only difference in specific uptake. Probably this sexual dimorphism is accounted for the earlier genesis of DA neurons in females comparing with males [65]. The role of either neurotropic factors or steroids secreted by glia cells in tissue culture in the sexual dimorphism of DA neurons [65, 66] also cannot be excluded. b) Other hormones Thyroid hormones The convincing evidence of the role of thyroid hormones in the differentiation of CA neurons has been obtained in the hypothalamic tissue culture of fetal mice (E16). Triiodthyronine results in the increased size of DA neurons and stimulates the 28 M. V. UGRUMOV growth and arborization of DA processes associ- ated with a rising of specific uptake of [7H]DA for 35%. The authors suggest either the direct or indirect, through glial cells, trophic action of thyr- oid hormones on the DA neurons [67]. To evaluate the role of thyroid hormones in the development of the hypothalamic CA system in vivo, the rats at P8-P10 were pretreated with thyroxine. This results in the increased contents of NA and DA at P16-P21, followed by their levelling or even reversion [68]. The authors hold the opinion that the thyroid hormones provide a tem- poral stimulating effect on the brain development. Adenohypophysial hormones Very recently, the stimulating effect of hor- mones of the intermediate lobe on the differenti- ation of the hypothalamic DA neurons has been suggested. In fact, the co-cultivation of the embryonic hypothalamus with the intermediate lobe results in transient increase of the specific uptake of [PH]DA [69]. More clear data were obtained on the neonatal imprinting effect of a-melanotropin in the feed- back regulation of DA neurons of the arcuate n. [70]. This regulation is established in rats between P4 and P8, that coincides with the peak in plasma melanotropin [71]. The passive immunization with antiserum to melanotropin at P5-P6, but not at P11-P12, makes the DA neurons of the arcuate n. non-sensitive to this hormone both in adult males and females [70]. In addition to the hormones of the developing organism, maternal hormones also control the maturation of the hypothalamic CA system. For example, maternal prolactin penetrated to pups in the course of suckling provides a long-term effect on the functional activity of DA neurons of the arcuate n. [72]. The depletion of maternal prolac- tin in pups at P2-P5, but not at P9-P12, by daily injections of bromcriptine to lactating females results in depression of DA turnover and in the elevation of the plasma prolactin when tested at P30-P35 [72]. Thus, the hormones control the development of the hypothalamus modifying the neuronal dif- ferentiation and metabolism, networks of nerve fibers, synaptic organization, neuroendocrine reg- ulation and behaviour. Sexual steroids seem to be the most potent factor involved in so-called “sex- ual differentiation” of the hypothalamus and its MA system. 2. SEROTONINERGIC SYSTEM a) Sexual steroids Biochemistry Up-to-date, there are controversial data on the ontogenetic sexual difference in the hypothalamic content of 5-HT. Watts and Stanley [73] failed to find out any sexual difference in the 5-HT content in period from P1.5 until P80, while other authors demonstrated its transient divergence [74]. Still, clear difference in the hypothalamic 5-HT content in adult male and female (higher level) rats [75] rises a question on the role of SS in the develop- ment of the 5-HT system. After castration of male rats at P1 the hypotha- lamic level of 5-HT is not changed comparing with control by P12. On P60 and P75, 5-HT depletes significantly, and is no longer seen at P90. Follow- ing three months, the concentration of 5-HT be-. comes higher in castrated rats than in control [76], even reaching its level in adult females [75]. In neonatally castrated adult males as in adult females the 5-HT content in the anterior hypotha- lamus exceeds that in the control males for 67%, while in the posterior hypothalamus for 46% [75]. Morphlogy The sexually dimorphic medial preoptic n. has been chosen to study the influence of SS on the distribution of 5-HT fibers. In both sexes this nucleus includes the medial part with the low concentration of 5-HT fiber, particularly in its center, and the lateral part with the extensive network of 5-HT fibers [77]. In perinatally or neonatally androgenized females, as in the control males, the area of the low fiber density (medial and central medial preoptic n.) is considerably larger, while the area of the high fiber density (lateral MPA) is proportionally smaller comparing with the control females (Fig. 11). This is more evident in perinatally androgenized females than in post- natally androgenized ones. Adult gonadectomy does not affect the distribution of 5-HT fibers in Monoaminergic System in Ontogenesis 29 Male + oil Female + oil C7 a MePO SV 7 SBS aa eee psy , y Ee is PyPO fpr ea A AS rhe f eS Qa EEE i wf ie OY BN uae Biter = TRIS RAN ARS oe nd eB m i ie NY . oe F177 \) PERE: c ee Hp as << ¥- LPO . \ anes cus WR 2 MPO 54 OS e eis ee SSeS Female+TP Fic. 11. Line drawings of the medial preoptic area at the level of the medial preoptic n. (MPN) illustrate the distribution of 5-HT-stained fibers in this nucleus and adjacent regions in oil-treated female (A), oil-treated male (B), and perinatally testosterone propionate-treated female (C) rats. m, c, l-medial, central, and lateral parts of the MPN, respectively. MPO-medial preoptic area. any area. These data show that the sexually dimorphic distribution of 5-HT fibers in the MPA is determined by the perinatal SS environment [77]. Physiology In this study, such physiological integrative char- acteristics of the hypothalamic 5-HT system as specific uptake and K*-stimulated Ca*?- dependent release of [PH]5-HT in vitro have been also compared in intact and neonatally castrated adult males, as well as in adult females [78]. In intact adult males, both parameters were signif- icantly lower than in females, while reaching the female level after castration. These data are considered as further evidence of the supressive effect of androgens on the development of the hypothalamic 5-HT system. The membrane effect of SS modifying their permeability also cannot be exculded [78]. Neuroendocrine regulation It is well-known that the neonatal androgeniza- tion of the hypothalamus results in the establish- ment of the tonic gonadotropin secretion. Con- comitantly, the mechanism of the control of gona- dotropin secretion seems to become non-sensitive to the regulative effect of 5-HT. In fact, intact For other abbreviations see [77]. females and neonatally castrated males, in contrast to intact males, show the luteinizing hormone release in response to 5-hydroxytryptophan, a 5- HT precursor. On the contrary, neonatal androgenization abolished the lutenizing hormone response to 5-HT in prepubertal female rats [79]. Thus, the sexual differences in the hypothalamic 5-HT system are manifested in the wide range diapason, from the 5-HT content to the neuroen- docrine regulations, that is the consequence of the testosterone masculinizing effect over the critical perinatal period. b) Other hormones In adition to SS, thyroid hormones, corticoster- oids, etc. contribute to the control of the dif- ferentiation of 5-HT system but this problem has been evaluated somewhere else. IV. FUNCTIONAL SIGNIFICANCE OF MONOAMINES 1. CATECHOLAMINES The role of CA in “sexual differentiation” of the hypothalamus occurred to be the most attractive problem. Their action is conveyed via pre- and postsynaptic a- and f-receptors. Presynaptic a- 30 M. V. UGRUMOov Tyrosine ay | TOH Metabolites Fie. 12. Neuron terminal Presynapse Synapse Postsynapse A proposal for the metabolism of neuronal norepinephrine in the neonatal rat. The following numbered steps refer to the numbered pathways in the figure. 1) Norepinephrine (NE) is synthesized from tyrosine by - tyrosine hydroxylase (TOH), the rate limiting enzyme. 2) Newly synthesized NE is incorporated into the synaptic storage vesicles or becomes part of the cytoplasmic pool. 3) With a neural stimulus, the synaptic vesicle fuses with the synaptic membrane, and NE 1s released into the synapse. 4) Synaptic NE will bind to postsynaptic a-receptors (step 5) or B-receptors (step 6). Synaptic NE (step 4) can also bind to a presynaptic a-receptor (step 7), which will inhibit the synthesis of TOH and the subsequent formation of new NE. 8) Synaptic NE and its physiological stimulus are terminated by rapid uptake of NE back into the neutron. The uptake mechanism is a-receptor linked. NE taken back into the neuron can be reincorporated into synaptic storage vesicles (step 9), metabolized by monoamine oxidase (MAO; step 10), or leak back out into the synapse (step 11) [81]. receptors are related to the specific reuptake of CA and to the negative feedback regulation of TH (a). Postsynaptic a; and f-receptors (/8;, 2) are responsible for the adequate reaction of target- cells (Fig. 12). Although (; and (-receptors are similar in their structure, the first ones are mainly sensitive to NA while the second to AD. Moreov- er, f)-receptors are mainly related to the post- synaptic membrane, while /5-receptors are also localized extrasynaptically [80, 81]. Morphogenesis The enlargement of the sexually dimorphic n. of the PA in the adult male and female rats were observed after the postnatal (P1-P5) stimulation of f2-receptors by agonist, salbutamol [82]. It is noteworthy that the pre- and neonatal treatment of males even with high concentrations of androgens or estrogens fails to increase the volume of this region [83]. According to the authors’ suggestion the effect of salbutamol could be accounted for by either its promotion of the neuron migration to the centre of this region, or by preventing of the neurons from the cell death [82]. The regulative influence on neuronogenesis is probably low, be- cause, as in the hypothalamic region, it practically completes by the end of fetal life [84]. Neuroendocrine regulation A number of physiological and pharmacological Monoaminergic System in Ontogenesis Sil studies have shown that SS, CA ligands and the drugs interfering with the CA metabolism, when applied neonatally, contribute to the “sexual dif- ferentiation” of the hypothalamus (Table 1) [6]. The mechanisms of their action remain uncer- tain, as substances mentioned above are multifunc- tional. Thus, SS apparently control the number of a-and #-receptors [6], on the one hand, and in- fluence the metabolism of NA, on the other [85]. TABLE 1. metabolism of catecholamines In turn, NA influences the concentration of SS receptors in the hypothalamic region [86]. Moreover, NA stimulated {-receptors are sup- posed to prevent androgenization of the hypotha- lamus by inhibition of either aromatization of testosterone or of the estradiol uptake by the neuronal nucleus. The last mechanism is closely related to the SS modulation of genome [81, 87]. Moreover, the stimulation of /-receptors inten- Effects of neonatally applied sexual steroids and drugs related to either the physiological action or Experimental Physiological Mechanism eld ast ‘ Og of action Ecce Neonatal Persistent a) phenoxybenzamine aAN { androgenization estrus in b) phentolamine aAN | of females [81] adulthood c) propranolol BAN — dja&c — Sines C — f) methyl-p-tyrosine inhibitor of a CA synthesis g) tyramine Stimulator of NA release Neonatal Female a) clonidine aA — androgenization sexual b) prazosine a, AN — of female & behavior c) yohimbine a,AN -— gonadectomy in adulthood [6]; Intraventricular Nuclear accumulation a) phenoxybenzamine aAN { injection of [*H]estradiol b) isoproterenol BA { of [*H]testo- in hypothalamus c) isoxuprine BA | sterone to d) hydroxybenxylpindolol BAN — female at P4 e)akd t [87] f)b&d t g)c &d t Neonatal LH response a) clonidine aA — treatment of to estradiol b) prazosine a, AN t females with & progesterone c) yohimbine a,AN t drugs & gonad- in adulthood d) salbutamol oA t ectomy in e) alprenolol BAN | adulthood [6, 88] f) isoprenaline BA | Neonatal female sexual a) clonidine aA | treatment of behavior b) prazosine a, AN — females with c) yohimbine a,rAN — drugs & gonadec- d) salbutamol AoA t tomy in e) alprenolol BA t adulthood [6] f) isoprenaline Neonatal female sexual a) isoprenaline aA t treatment of behavior b) salbutamol 5A — males with c) alprenolol BAN = drugs. Castration & estradiol and progesterone in adulthood [6]. a: a-receptor; A: agonist; AN: antagonist; 8: 8-receptor; f: increased |: decreased; “—”: no effect. 32 M. V. UGRuMov sifies the female sexual behavior, but it inhibits the female type of gonadotropin secretion. On the contrary, the blockade of a-receptors results in the increase of the LH-surge release, while decreasing of female sexual behavior. Activation of ap- receptors supresses the female sexual behavior, though has no effect on the LH-release (Table 1) [88]. From the physiological point of view, the in- terfering of CA, mainly DA, and SS in their action on the gonadotropin secretion is of particular interest. Thus, the gonadotropin secretion is under the DA inhibitory control in female rats, but not in males, until P20. The physiological response of both females and males are completely reversed after their neonatal androgenization and castra- tion, respectively [89]. The hypothesis that the DA neurons in the anteroventral periventricular n. of the PA mediate the SS effects on gonadotropin secretion seems to be reasonable. The failure to show a phasic gonadotropin secretion in response to estrogen and progesterone in adult females after neonatal androgenization that caused the reduc- tion of the population of DA neurons, is in favor of this idea [64]. The direct inhibiting action of DA on prolactin secretion begins in rats only after birth, at P3. First at this time, the inhibitor of DA receptors in adenohypophysis (pimozide) becomes effective in the increase of prolactin plasma level [90]. Thus, CA are believed to mediate the organizing effects of SS on the hypothalamus in the critical period of ontogenesis inducing the permanent sex- ual differences in morphogenesis of the hypothala- mic target regions, e.g., sexually dimorphic n. of the PA, the gonadotropin secretion pattern, and sexual behaviour. 2. SEROTONIN Autoregulation The good example of the autoregulation is the population of the hypothalamic 5-HT-like neurons possenssing the specific uptake of 5S-HT and synth- esizing, = 5-H from its precursor, ~ 5- hydroxytryptophan (see above). The repeated treatment of the embryonic hypothalamic tissue culture with 5-HT agonist, 8-hydroxy-2-[di(n- propyl)amino]tetralin, results in an increased num- ber of 5-HT-like neurons with the elevated level of aromatic-L-amino acid decarboxylase. Both events are highly suppressed by a specific blocker of 5-HT receptors, metergoline [49]. Neuronogenesis One of the most important evidence on the role of 5-HT in the differentiation of the target neurons in vivo has been obtained evaluating the prolifera- tive activity of cell-precursors with long-survival [(°H]thymidine radioautography following pharma- cological 5-HT depletion by pCPA in fetuses [7]. In some hypothalamic areas, zona incerta and paraventricular n., the retardation of the neuronal genesis (prolongation of cell-precursors prolifera- tion) has been detected, while in the others, the SCN and ventromedial n., the effect of 5-HT depletion is not evident. According to the authors’ suggestion these neurons are originated before the first delivery of 5-HT either via MFB or with the CSF. Sexual differentiation Although androgens are known to be the most important inductor of hypothalamic “sexual dif- ferentiation”, 5-HT is believed to play an in- termediate or even a leading role in its action. Thus, the neonatal inhibition of the 5-HT synthesis in female rats results in the increased volume of the sexually dimorphic n. to that of intact males, while it does not alter the SS level in the blood (Cay). These data have been significantly revised by further study of Jarzab and Dohler [92] who used physiological models of female rats neonatally treated with: a) Testosterone; b) Testosterone and pCPA; and c) Testosterone and L-tryptophan. According to their results, the postnatal inhibition of 5-HT synthesis does not modify the effect of testosterone either on the gonadotropin secretion or on the expression of sexual behavior. Still, postnatal stimulation of 5-HT synthesis inhibits considerably the expression of sexual behavior while it does not modify the gonadotropin release [92]. The discrepancies in the above mentioned re- sults are apparently accounted for by the interfer- Monoaminergic System in Ontogenesis ing of many biologically active substances during the critical periods. On the other hand, the pharmacological models are usually far from ideal, e.g. pCPA inhibits the 5-HT synthesis only for 50- 70%, while modifying the metabolism of CA, proteins, etc. at the same time. Neuroendocrine regulation One of the most importnat characteristic of the 5-HT is functioning in the maturation of the recep- tors on the targets. Thus, the 5-HT binding ability of the hypothalamic tissue has been estimated in postnatal female rats. The specific 5-HT-binding sites were already observed at P5; their number doubled between P16 and P27 and continued to increase at least until P37 [93]. It is of particular interest that even the fetal 5-HT receptors are capable of responding to changes in the 5-HT level: the maternal administration of the 5-HT depleter, pCPA, results in the elevated number of receptors in newborn pups, whereas 5-HT agonist, 5-methoxytryptamine, provides the opposite effect. It means that the 5-HT receptors are functional in fetuses and show the same plastic changes as in adults [94]. In other words, 5-HT may serve as a potential developmental signal even during intrauterine development [95]. Further physiological experiments have proved the functional potency of 5-HT receptors in the developing hypothalamus. Thus, 5-HT is ineffec- tive as the modulator of the SS action on LH secretion in female rats at P5. This effect first appears between P8 and P12 while increasing abruptly and remaining at the same high level until P18. Then, the response to 5-HT decreases prog- ressively until P50 [93]. This dynamic is quite different from the 5-HT induced release of prolac- tin in ontogenesis. The 5-HT regulative effect on prolactin secretion being absent over first postnatal week appears at P12 followed by its strengthening until puberty [93]. Thus, several aspects of the 5-HT action on the developing hypothalamus could be distinguished: autoregulation of 5-HT neuron differentiation, control of neuronogenesis, the regulation of “sex- ual differentiation’, etc. Vv. CONCLUSIONS 1) Hypothalamic catecholamine- and _ seroto- ninergic systems undergo synchronous develop- ment that is manifested first in the genesis of catecholamine- and serotoninergic neurons in the hypothalamus and brain stem long before birth (E12-E13). Further neuron differentiation results in the expression of the key enzymes and monoamine synthesis, the onset of specific uptake and potassium-stimulated release of monoamines. One of the most important attribute of the dif- ferentiation is the growing of axons to the target neurons, hypophysial portal circulation and III ventricle, followed by the establishment of special- ized contacts: synapses, axo-vascular and axo- ventricular. 2) The development of the monoaminergic sys- tems is under the hormonal control. Some hor- mones (thyroid hormones) provide a temporal stimulating effect on the neuron differentiation, while others, e.g., sexual steroids, have a program- ing irreversible action that occurs in the critical perinatal period. In the last case, the hormones modify considerably the neuron differentiation and metabolism, networks of fibers, synaptic orga- nization, and, finally, neuroendocrine regulation and behaviour. 3) The hypothalamic monoamines appeared ear- ly in ontogenesis are characterized by a wide range spectrum of their physiological action. First, they control the genesis and further differentiation of the target neurons. Later, over critical perinatal period, they mediate the hormonal programing influence, thus, modifying the architectonics, metabolism, receptors, etc. of the hypothalamic target regions and, finally, the adenohypophysial hormone secretion and behaviour. REFERENCES 1 Barraclough, C. A. and Wise, P. M. (1982) Endo- crine Rev., 3: 91-119. 2 Collu, R. (1989) In “Pediatric Endocrinology”. Ed. by R. Collu, J. R. Ducharme and H. J. Guyda, Raven Pres, New York, pp. 1-36. 3 Fuller, R. W., Clemens, J. A. (1981) In “Serotonin: Current Aspects of Neurochemistry and Function”, Ed. by B. Haber, S. Gabay, M. R. Issidorides and S. 13 14 15 16 7) 18 19 20 21 Upp 34 G. A. Alivesatos, Plenum Press, New York, Adv. Exp. Med. and Biol., 133, pp. 431-444. | Jackson, I. and Mueller, G. P. (1982) In “Biological Regulation and Development. Hormone Action”, Ed. by R. F. Goldberg and K. R. Yamamoto, Plenum Press, New York, pp. 127-202. Buznikov G. A. (1991) Neurotransmitters in Embryogenesis. Harwood Acad. Publ., London, 256 (in press). Jarzab, B., Kokocinska, D., Kaminski, M., Gubala, E., Achtelik, W., Wagiel, J. and Dohler, K. D. (1990) In “Hormones Brain and Behavior in Verte- brates. I. Sexual Differentiation, Neurochemical Aspects, Neurotransmitters and Neuropeptides. Comp. Physiol.” Ed. by J. Balthazart. Karger, Bas- al, Vol. 8, pp. 41-50. Lauder J. M. (1983) Hormonal and humoral in- fluences on brain development. Psychoneuroendoc- rinology, 8: 121-155. Lauder, J. M., Wallace, J. A. and Krebs, H. (1981) Adv. Exp. Med. and Biol., 133: 477-506. Jonakait, G. M. and Black, I. B. (1989) In “Hand- book of Experimental Pharmacology. Catechola- mines II”, Ed. by U. Trendelenburg and N. Weiner, Vol. 90/II: pp. 137-160. Borisova, N. A., Balan, I. and Ugrumov, M. V., unpublished. Specht, L. A., Pickel, V. M., Joh, T. H. and Reis, D. J. (1981) J. Comp. Neurol., 199: 233-253. Ugrumov, M. V., Taxi, J., Tixier-Vidal, A., Thi- bault, J. and Mitskevich, M. S. (1989) Neurosci- ences, 29. 135156: Friedman, W. J., Dreyfus, C. F., McEwen, B. S. and Black, I. B. (1989) Dev. Brain Res., 48: 177- 185. Daikoku, S., Kawano, H., Okamura, Y., Tokuzen, M. and Nagatsu, I. (1986) Dev. Brain Res., 28: 85- 98. Chan-Palay, V., Zaborszky, L., Kohler, C., Gold- stein, M. and Palay, S. L. (1984) J. Comp. Neurol. 227: 467-496. Specht, L. A., Pickel, V. M., Joh, T. H. and Reis, D. J. (1981) J. Comp. Neurol., 199: 255-276. Ugrumov, M. V., tixier-Vidal, A., Taxi, J., Thi- bault, J. and Mitskevich, M. S. (1989) Neurosci- ence, 29: 157-166. Wray, S. and Hoffman, G. (1986) Brain Res., 399: 327-331. Clough, R. W., Hoffman, G. E. and Sladek, C. D. (1988) Brain Res., 446: 121-132. Van den Pol, A. N. and Tsujimoto, K. L. (1985) Neuroscience, 15: 1049-1086. Reppert, S. M. and Uhl, G. R. (1987) Endocrinolo- gy, 120: 2483-2487. Hyyppa, M. (1969) Z. Zellforsch., 98: 550-560. 23 24 25 26 ZA 28 29 30 Sil 32 33) 34 36 37 38 39 40 M. V. UGRuMov Khachaturian, H. and Sladek, J. R. (1980) Pep- tidkeSawle 7-95. Foster, G. A., Schultzberg, M., Goldstein, M., Hokfelt, T. (1985) J. Comp. Neurol. 236: 348-381. Nakada, H. and Nakai, Y. (1985) Brain Res., 361: DAV=297 Nakai, Y., Shioda, S., Ochiai, H. and Kozasa, K. (1986) Curr. Top. Neuroendocrinol., 7: 135-160. Liposits, Zs., Sherman, D., Phelix, C. and Paull, W. K. (1986) Brain Res., 85: 95-106. Borisova, N. A., Sapronova, A. Y., Proshlyakova, E. V. and Ugrumoy, M. V. (1991) Neuroscience, 43: 223-229. Ibata, Y., Fukui, K., Obata, H. L., Tanaka, M., Hisa, Y., Sano, Y. (1982) Brain Res. Bull., 9: 407- 415. Loizou, L. A. (1971) Z. Zellforsch., 114: 234-252. Kordon, C., Enyjalbert, A., Hery, M., Joseph- Bravo, P. I., Rodsztein, W. and Ruber, M. (1980) In “Physiology of the Hypothalamus. Handbook of the Hypothalamus”, Ed. by P. J. Morgane and J. Panksepp, Dekker Inc, New York, Vol. 2, pp. 253- 306. Kuljis, R. O. and Advis, J. P. (1989) Endocrinolo- gy, 124: 1579-1581. Ibata, Y., Tani, N., Obata, H. L., Tanaka, M., Kubo, S., Fukui, K., Fujimoto, M., Knoshita, H., Watanabe, K. and Sano, Y. (1981) Cell Tiss. Res., — 216: 31-38. Ugrumov, M. V. (1991) Int. Rev. Cyt., 129: 207- AST. Parent, A., Poitras, D. and Dube, L. (1984) In “Handbook of Chemical neuroanatomy. Vol. 2. Classical transmitters in the CNS. PArt I”, Ed. by . Bjorklund and T. Hokfelt, Elsevier, Amsterdam, pp. 409-439. Ugrumov, M. V., Vladimirov, S. V., Kasmambeto- va, S., Popov, A. P. and Thibault, Neuroscience, submitted. Lichtensteiger, W. and Schlumpf, M. (1981) In “Steroid Hormone regulation of the Brain”, Eds. K. Fuxe, J. A. Gustafsson and Wetterberg. Oxford, Pergamon Press, pp. 161-172. Trembleau A., Ugrumov M., Bernabe J. and A. Calas Development of the hypothalamic catechola- minergic system and and its pharmacological deple- tion during ontogenesis. Society for Nueroscience, 21st Annual Meeting, 1991, Abstracts Vol. 17, Part Jay Do. JIVE). Ugrumoyv, M. V., Mitskevich, M. S., Halasz, B., Kiss, J. and Borisova, N. A. (1986) Int. J. Dev. Neurosci., 4: 101-111. Steinbusch, H. W. H. and Nieuwenhuys, R. (1981) In “Serotonin: Current Aspects of Neurochemistry and Function”, Ed by B. Haber, S. Gabay, M. R. 4] 42 43 44 45 46 47 48 49 50 51 2 53 54 55 56 S// 58 a9 60 61 62 63 Monoaminergic System in Ontogenesis 35 Issidorides and S. G. A. Alivisatos, Plenum Press, New York, pp. 7-35. Ugrumov, M. V., Taxi, J., Mitskevich, M. S. and Tramu, G. (1986) Dev. Brain Res., 30: 75-84. Wallace, J. A. and Lauder, J. M. (1983) Brain Res. Bull., 10: 459-479. Aitken, A. R. and Tork, I. (1988) J. Comp. Neurol., 274: 32-47. Beaudet, A. and Descarries, L. (1979) Brain Res., 160: 231-243. Frankfurt, H., Lauder, J. and Azmitia, E. (1981). Neurosci. Lett., 24: 227-232. Lidov, H. G. W. and Molliver, M. E. (1982) Brain Res. Bull., 8: 398-430. Ugrumovy, M. V., Taxi, J., Steinbusch, H. W. M., Tramu, G. and Mitskevich, M. S. (1989) Dev. Brain Res., 46: 233-241. Ugrumovy, M. V., Taxi, J., Steinbusch, H. W. M.., Tramu, G. and Mitskevich, M. S. (1988) C. R. Acad. Sci. (Paris), 307: 505-509. De Vitry, F., Hamon, M., Catelon, J., Dubois, M., Thibault, J. (1986) Proc. Nati. Acad. Sci. (USA), 83: 8629-8633. Sakumoto, T., Sakai, K., Jouvet, M., Kimura, H. and Maeda, T. (1984) Brain Res. Bull., 12: 721- 133) Horvath, J., Sétalo, G. and Flerk6, B. (1982) En- docr. Exp., 16: 229-238. Meijer, J. H. and Rietveld, W. J. (1989) Physiol. Rev., 69: 671-707. Ugrumov, M. V., Bladimirov, S. V., Kasmambeto- va, S., Popov, A. P., Novodjilova, A. P. and Tramu, G., Neurosicence, submitted Bosler, O. and Beaudet, A. (1985) J. Neurocytolo- gy, 14: 749-763. Shimatsu, A., Kato, Y., Matsushita, N., Katakami, H., Ohta, H., Yanaihara, N. and Imura H. (1983) Brain Res., 264: 148-151. Maegawa, M. (1987) Tokushima J. exp. Med., 34: 29-43. Ugrumov, M. V., Proshlyakova, E. V. and Sapro- nova, A. Y. (1989) Neuroscience, 32: 127-131. Liu, K. P., Tamir, H., Hsiung, S., Adlerberg, M. and Gershon M. D. (1987) Dev. Brain Res., 32: 31- 41. Siddiqui, A. and Gilmore, D. P. (1988) Endocr. (Copenh.) 118: 483-494. Reznikov, A. and Nosenko, N. (1990) In “Neuroen- docrinology: New Frontiers”. Ed by G. Wollmann and R. Tubingen. Wilson, W. E. and Agrawal, A. K. (1979) Dev. Neurosci., 2: 195-200. Crowley, W. R., O’Donohue, T. L. and Jacobowitz, D. M. (1978) Acta Endocr. (Copenh.), 89: 20-28. Demarest, K. T., McKay, D. W., Riegle, G. D. and Acta 64 65 66 67 68 69 70 71 2 73 74 75 76 V1 78 79 80 81 82 83 84 Moore, K. E. (1981) Neurooendocrinology, 32: 108-113. Simerly, R. B., Swanson, L. W., Handa, R. J. and Gorski, R. A. (1985) Neuroendocrinology, 40: 501- 510. Reisert, I., Engele, J. and Pilgrim, Ch. (1989) Cell Tiss. Res., 255: 411-417. Costa, E., Baulieu, E. E., Krueger, K. E., Purdy, R. H. and Puia, G. (1991) Abstracts Third IBRO Congress, Montreal, P. 8, W12. Puymirat, J., Barret, A., Picart, R., Vigny, A., Loudes, C., Faivre-Bauman, A. and Tixier-Vidal, A. (1983) Neuroscience, 10: 801-810. Lengvari, I., Branch, B. J. and Taylor, A. N. (1980) Dev. Neuroci., 3: 59-65. Charli, J. L., Faivre-Bauman, A., Loudes, C. and Kordon, C. (1991) In XXeme Colloque de la Societe de Neuroendocrinologie Experimentale, Geneve, 18-20 Septembre 1991, (in press). Lichtensteiger, W. and Schlumpf, M. (1986) Brain Res., 368: 205-210. Davis, M. D., Lichtensteiger, W., Schlumpf, M and Bruinink, A. (1984) Neuroendocrinology, 39: 1-12. Shyr, S. W., Crowley, W. R. and Grosvenor, C. E. (1986) Endocrinology, 119: 1217-1221. Watts, A. G. and Stanley, H. F. (1984) Neuroen- docrinology, 38: 461-466. Reznikov, A. G., Nosenko, N. D. and Demkiv, L. P. (1979) Endokrinologie, 73: 11-19. Borisova, N. A., Manukhin, B. N. and Selivanova, G. P. (1989) Bull. Exp. Biol. Med., 57: 135-136 (Russ). Siddique, A. and Gilmore, D. P. (1987) Neurosci- ence, Suppl. Vol. 22, p. 154. Simerly, R. B., Swanson, L. W. and Gorski, R. A. (1985) Brain Res., 340: 91-98. Borisova, N. A., Proshlyakova, E. V., Sapronova, A. Ya. and Ugrumov, M. V. Bull. Exp. Biol. Med., (in press) (Russ). Moguilevsky, J. A., Faigon, M. A., Scacchi, P. and Szwarcfarb, B. (1987) Neuroendocrinology, 45: 274-277. Jarzab, B., Sickmoller, P. M., Geerlings, H. and Dohler, K. D. (1987) Hormones and Behavior, 21: 478-492. Raum, W. J. and Swerdloff, R. S. (1981) Endocri- nology, 109: 273-278. Jarzab, B., Kaminski, M., Gubala, E., Achtelik, W., Wagiel, J. and Dohler, K. D. (1990) Brain Res., 516: 257-262. Dohler, K.-D., Coquelin, A., Davis, F., Hines, M., Shryne, J. E. and Gorski, R. A. (1984) Brain Res., 302: 291-295. Altman. J. and Bayer, S. A. (1986) Adv. Anat., Embryol. and Cell Biol., 100: 1-173. 85 86 87 88 89 90 91 36 Reznikov, A. G. and Nosenko, N. D. (1983) Exp. Clin. Endocr., 81: 91-93. Blaustein, J. D. and Turcotte, J. (1987) Brain Res., 436: 253-264. Raum, W. J., Marcano, M. and Swerdloff, R. S. (1984) Biol. Reproduction, 30: 388-396. Jarzab, B., Gubala, E., Achtelik, W., Lindner, G., Pogorzelska, E. and Dohler, K.-D. (1989) Exp. Clin. Endocrinol., 94: 61-71. Lacau de Mengido J., Becu-Villalobos D. and Libertun C. (1987) Dev. Brain Res., 35: 91-95. Ojeda, S. R. and McCann. S. M. (1974) Endocri- nology, 95: 1499-1505. Handa, R. J., Hines, M. Scoonmaker J. N., Shryne, J. E. and Gorski, R. A. (1986) Dev. Brain Res., 30: 92 93 94 95 M. V. UGRUMOV 278-282. Jarzab, B. and Dohler, K. D. (1984) In “Sex differences in the Brain”. Ed. by G. J. De Vries. J. P. C. De Bruin, H. B. M. Uylings and M. A. Corner, Elsevier, Amsterdam, Prog. Brain Res., Vol. 61, pp. 119-126. Becu-Billalobos, D. and Libertun, C. (1986) Dev. Brain Re., 25: 111-116. Hamon, M., Bourgoin, S., El Mestikawy, S. and Goetz, C. (1984) In “Handbook of Neurochemis- try”, Ed. by A. Lajtha, Plenum Publ., New York, Vol. 6, pp. 107-143. Whitaker-Azmitia, P. M., Lauder, J. M., Shemmer A. and Azmitia, E. C. (1987) Dev. Brain Res., 33: 285-289. © 1992 Zoological Society of Japan ZOOLOGICAL SCIENCE 9: 37-45 (1992) Changes in Hemolymph Vitellogenin and Ecdysteroid Levels during the Reproductive and Non-Reproductive Molt Cycles in the Freshwater Prawn Macrobrachium nipponense Takuyt OKUMURA, CHANG-HEE Han!, Yuzuru SUZUKI, Katsumi ArpA~ and Isao HANyu Laboratory of Fish Physiology, Department of Fisheries, Faculty of Agriculture, The University of Tokyo, Bunkyo, Tokyo 113, Japan ABSTRACT—Hemolymph vitellogenin and ecdysteroid levels were traced during the reproductive and non-reproductive molt cycles in adult females of the freshwater prawn Macrobrachium nipponense, in which the reproductive development coincides with the molt cycle. During the reproductive molt cycle, vitellogenin levels began to increase at stage B. High levels were maintained during stages C,-D3 in concomitance with ovarian growth, and these levels decreased following the prespawning ecdysis. Vitellogenin was at undetectable levels during the non-reproductive molt cycle, and correspondingly, Ovarian maturation did not advance. The predominant ecdysteroids in the hemolymph of M. nipponense were determined by HPLC-RIA analysis as 20-hydroxyecdysone (20E) and immunoreactive high polarity products (HPP). 20E and HPP levels peaked at stage D3 during both molt cycles. 20E levels declined rapidly to low levels after ecdysis in the reproductive molt cycle. On the other hand, 20E levels decreased slowly during stages A-C, in the non-reproductive molt cycle. HPP levels immediately declined prior to spawning at stage Ag and showed a small peak after spawning at stage A, in the reproductive molt cycle, whereas in the non-reproductive molt cycle levels were still high at stage A and declined slowly during stages B-C,. These differences in ecdysteroid levels suggest a possible relationship between ecdysteroid levels and vitellogenesis. is completed during the subsequent molt cycle. In most crabs, ovarian development, spawning, and brooding occur within a single intermolt period. Endocrinological mechanisms possibly integrat- INTRODUCTION The integration of reproduction and molting appears to be a physiological necessity in female crustaceans. Most species continue to molt and grow even after reaching sexual maturity. The timely occurrence of ecdysis ensures that the egg mass brooded abdominally is not lost with the shedding of the exoskeleton. Various manners of coordination of the molting and reproductive cy- cles exist (see [1, 2]). In a number of prawn species and in several isopods, ovarian development occurs during the intermolt period, and spawning takes place after ecdysis. Brooding of the egg mass Accepted August 13, 1991 Received June 10, 1991 " Present address: Department of Biology, College of Natural Science, Dongeui University, Busan, Korea. * To whom correspondence should be addressed. ing the processes of molting and vitellogenesis are only partially understood. A correlation between the processes of molting and reproductive activity persists after ablation of the eyestalk in the shrimps Lysmata seticaudata [3] and Palaemon serratus [4]. The eyestalk includes X-organ sinus gland complex, a secretion site of neurohormones such as vitellogenesis-inhibiting hormone and molt-inhibiting hormone. These results suggest that while the neurohormones of the eyestalk may play indirect roles, they are not indispensable to the integration of molting and reproduction. It is more likely that hormones of other sources play key roles in this integration. Removal of the Y-organ, a site of ecdysteroid production via 38 T. Okumura, C.-H. HAN et al. cholesterol modification, induces the arrest of vitellogenesis in amphipods and isopods [5-7]. These results suggest the possibility that ecdyster- oids not only control molting but also regulate vitellogenesis. In order to gain more understand- ing of the functions of ecdysteroids in vitellogene- sis, the correlation between hemolymph ecdyster- oid and vitellogenesis should be further investi- gated. However, this has been examined in only a limited number of species (see [2]). To our knowl- edge, patterns of hemolymph ecdysteroid levels in adult females have not been priorly compared between the vitellogenic and non-vitellogenic molt cycles in any species. In order to examine the relationship between vitellogenesis and molting, the freshwater prawn Macrobrachium nipponense was employed in this study. This is considered a suitable species, as the reproductive cycle of female coincides with the molt cycle and the reproductive status can be manipulated [8]. Female prawns repeatedly undergo the reproductive molt cycle during the spawning season from June to August in Japan. During the reproductive molt cycle, ovaries de- velop cyclically in concurrence with the molt cycle, and mating and spawning occur following each ecdysis. However during the non-spawning sea- son, females molt successively without the occurr- ence of ovarian development (non-reproductive or common molt cycle). The environmental factors influencing the annual reproductive cycle have also been investigated [8]; the initiation of the spawn- ing season is induced by increased temperature regardless of photoperiod, and the termination of the spawning season is caused by shortened day- length. Therefore, the reproductive molt cycle and the non-reproductive molt cycle are easily repro- ducible in this species under artificial conditions. Prawns undergo continuous cycles of reproductive molt at 13.5-day intervals when reared at 28°C under 15L9D, but successively repeat the non- reproductive molt cycle under 28°C and 12L12D at 14.8-day intervals [8]. In this study, we examined changes in hemolymph levels of vitellogenin (Vg), the precur- sor of yolk protein, during the molt cycle of M. nipponense in order to gain more insight into the coincidence of vitellogenesis and molting. Further- more, we follwed the fluctuation of hemolymph ecdysteroid during the molt cycles. Differences in ecdysteroid levels between the reproductive molt cycle (occurrence of vitellogenesis) and the non- reproductive molt cycle (absence of vitellogenesis) and relationships between ecdysteroid and Vg levels are discussed. MATERIALS AND METHODS Animals Female freshwater prawns M. nipponense origi- nating from the Kasumigaura Lake in Ibaraki Prefecture were purchased from local fishermen, and kept in outdoor tanks under ambient water temperature and photoperiod until use in experi- ments conducted the following year. For experiments, female prawns (body weight ranging from 1.4—4.6 g) were transferred to plastic aquaria (604040 cm), and reared under two separate conditions: 15L9D at 28°C from August to October in 1987 and 1988, and 12L12D at 28°C from October to November in 1988. Prawns were fed commercial pellets (Taiyo Fishery Co. Ltd.) ~ and Tubifex worms daily. In order to avoid any effects due to male prawns, only female prawns were placed in aquaria. Therefore, prawns in the reproductive molt cycle were not able to mate and brood, but could spawn continuously. After acclimation periods of more than three weeks, prawns were sacrificed at differing molt stages. Hemolymph samples were taken from the posterior aorta with a capillary tube after cutting the sixth abdominal segment. The hemolymph samples were stored at —70°C until analyses of hemolymph Vg and ecdysteroid levels. Gonado- somatic indices (GSI) were calculated as gonad weight (g) < 100/body weight (g). Molt stage Molt stages were determined according to the methods of Han [8] by observing setal develop- ment in the pleopods under a differential phase contrast microscope. As stage E is of a very short period, prawns were not sacrificed at this stage. Molt stage criteria is shown in Table 1. Vitellogenin and Ecdysteroid in Prawns 3) TABLE 1. Criteria for the determination of the molt stages in M. nipponense S Approximate R c tape duration* aaa Postmolt 2 days A Exoskeleton is very soft. Setae are filled with translucent cellular matrix. During the reproductive molt cycle, stage A is subdivided into stage Ao (before oviposition) and A, (after oviposition). B Setal matrix is constricted. Internal cones are forming inside setae. Intermolt 5 days Co Exoskeleton is fully formed. Completion of formation of internal cones is observed. C, New setae begin to appear in internal cones, but the epidermis is not yet separated from the base of setae. Premolt 7 days Do The epidermis begins to retract from the old cuticle. D, The epidermis separates from the old cuticle completely. Space between the epidermis and old cuticle widens. D, New setae form in the cellular matrix under the retracted epidermis. D; New barbules form on the new setae. Ecdysis 10 sec J8) Old cuticle is shed. *: rearing condition 28°C Enzyme-immunoassay (EIA) for Vg Vitellin was purified from mature ovaries of M. nipponense using gel-filtration and DEAE-ionex- change chromatography [8], and was used as the standard in this EIA system. Antisera were raised against the purified vitellin in rabbits [8]. The only hemolymph protein to which the antisera exhibited cross-reactivity was Vg, validating its specificity for assay use [8]. The purified vitellin was dissolved in CB (0.1 M Na,CO; buffer, pH 9.6) containing 0.01% male hemolymph and a series of 2-fold dilutions were made, ranging from 8.2 ng/ml (0.82 ng/well) to 1,050 ng/ml (105 ng/well). The hemolymph sam- ples were diluted 10,000-fold in CB, and subse- quently diluted with CB containing 0.01% male hemolymph to the concentration adequate for assay range, and subjected to EIA. The antiserum was diluted 3,000-fold with PBS-Tween (0.073 M Na,HPO, and 0.036 M KH>PO, buffer, pH 7.2, containing 0.05% polyoxyethylenesorbitan-mono- laurate, 0.04 M NaCl and 0.02% NaNs3). EIA procedures were as follows; one hundred yl of either vitellin in standard solutions or Vg in sample solutions was adsorbed to wells of 96-well EIA plates (Nunc-Immuno Plate MaxiSorp F96 Certificate, A/S Nunc) by incubation at 4°C over- night. Subsequently, solutions in wells were dis- carded, and wells were washed with PBS-Tween three times. In order to block adsorption of antibodies for the purpose of the following steps, wells were coated with 1% bovine serum albumin solution (250 1) in PBS (0.073 M NazHPO, and 0.036 M KH>PO, buffer, pH 7.2, containing 0.04 M NaCl and 0.02% NaN;), by incubation at 27°C for 2 hr. After washing, diluted antiserum (100 pl) was added to each well followed by incubation at 27°C for 2hr. After washing, 100 1 of goat anti-rabbit IgG (whole molecule) alkaline phos- phatase conjugate (Sigma Chemical Co.) in PBS- Tween (1 :300) was added to wells with incubation at 27°C for 2hr. After washing, 100 ul of 0.07% p-nitrophenylphosphate, disodium salts in 9.7% diethanolamine buffer, pH 9.8, containing 0.047% MgCl, and 0.02% sodium azide was added to each well, and incubation at room temperature was carried out for 30 min. Continuous absorbance of substrate solutions in each well was measured at 405nm by a Tosoh Co. MPR-A4 microplate- reader. The concentration of samples was calcu- lated from the standard curve, and expressed in 40 T. Oxumura, C.-H. HAN et al. milligram eugivalents of vitellin per milliliter of hemolymph. Validation of EIA Parallelism between the vitellin standard curve and the absorbance curve for serially diluted hemolymph of vitellogenic female prawns was examined. To determine the recovery rate, differ- ing doses of vitellin were added to hemolymph of non-vitellogenic female prawns, and the amounts of vitellin recovered were measured by EJA. The relationship between the amounts of vitellin added and recovered was determined by linear regression analysis. The intra-assay variabilities were also examined. Radioimmunoassay (RIA) for ecdysteroids Hemolymph ecdysteroids were assayed by re- verse-phase high performance liquid chromatogra- phy (HPLC) and RIA, as described previously [9]. Hemolymph samples at each molt stage (2-7 prawns) were pooled for analysis, as hemolymph volume of this species was too small to determine individual levels. Ecdysteroids were extracted from sample hemolymph with methanol, and partitioned be- tween 70% methanol and n-hexane (1:1). The 70% methanol phases were subjected to HPLC separation using a reverse phase column (ODS- 80TM, 4.6250 mm, Tosoh Co.). The starting solvent, methanol-water (1:1) was run for 30 min, after which the solvent system was changed to a linear gradient reaching methanol-water (3:2) in 15 min. The flow rate was 0.8 ml/min. Fractions were collected at a rate of one fraction per minute. Ecdysteroid in each fraction was measured by a double antibody RIA method. The antibody used in this study (provided by Dr. M. Nagata, The University of Tokyo) showed similar cross- reactivities toward 20-hydroxyecdysone (20E) and ecdysone. The standard curve was obtained by serial dilutions of 20E (Sigma Chemical Co.); therefore, results are expressed in nanogram equivalents of 20E per milliliter of hemolymph. Statistics Duncan’s multiple range test was used for statis- tical analysis after log-transformation of data. RESULTS Validation of EIA The EIA system used in this study was validated according to standard criteria. The absorbance curve of serially diluted hemolymph was parallel to the standard curve (Fig. 1), indicating that the adsorption of Vg in hemolymph samples to wells showed the same dose-dependence as did the standard. The amounts of standard vitellin added and recovered were significantly correlated (r= 0.984, p<0.01). The mean recovery rate was 87.4 +3.6% (mean+SE). The intra-assay coefficients of variation were under 10% in concentrations ranging from 3.8 ng/well to 90.3 ng/well. 4 Dilution of female hemolymph (10 ) e—e Gb 82 AG 8 4 2 0.5 @ 7) = o = re) a 2 = ¢ 0.1 0.05 8.2 16.4 32.8 65.6 131 263 525 1050 Vitellin (ng/ml) o—o Fic. 1. EIA system for vitellogenin. Parallelism be- tween the standard curve for purified vitellin and the absorbance curve for serial dilutions of female hemolymph. GSIs and Vg levels during the molt cycles The variations of GSIs and hemolymph Vg levels were followed during the two types of molt cycles (Fig. 2). During the reproductive molt cycle, GSIs were low during stages A;-Co after spawning, and started to increase at stage C; (p< 0.05). Values peaked at stage D3 and then de- Vitellogenin and Ecdysteroid in Prawns 4] =» Reproductive Molt Cycle o Non-Reproductive Molt Cycle e Reproductive Molt Cycle © Non-Reproductive Molt cycle 10 ue 5) Cc = ga =] : : QO. =v z : . e Sy o Y) = al O = (= oT) Vv ss 5) ” 2 S o = > O-----So7------ O------ Oils O----... (“esnsa Qauisoae Foote O------. fe) 9) O-------------- } ----- O----- O----- t}----- f}----- i----- i}----- Oo @) Reproductive MolteGycie: (2). (3). (4)_..(5) (5) (7). (6), (6). @, (2) Non-Reproductive Molt Cycle : (6) (Seer) (6) hs (S) ee (A) eG) t(5)) tS) Molt stage Fic. 2. GSI and hemolymph vitellogenin levels during the reproductive and non-reproductive molt cycles in Macrobrachium nipponense. Data for stage A during the reproductive molt cycle (divided into stages Ag and A,), and during the non-reproductive molt cycle (undivided). The data for stages A and Ay are shown twice. Values represent the mean+SE. Number of animals assayed is given in parentheses. Vitellogenin levels were calculated as vitellin equivalents. clined at spawning during stages Ag-A, (p<0.01). Hemolymph Vg levels were low during stages Ao-A,, increased at stage B (p<0.01), remained at high levels during stages C,-D3, and then declined (p<0.01) sharply after the prespawning ecdysis. During the non-reproductive molt cycle, GSIs remained constantly low (0.6-0.7%). Hemolymph Vg levels were nondetectable (less than 0.01 mg/ ml) throughout the molt cycle. Identification of ecdysteroids in hemolymph Ecdysteroids in hemolymph were separated by HPLC before RIA. The HPLC profile of hemolymph ecdysteroid at stage D3 of the repro- 42 T. Oxumura, C.-H. HAN et al. 50 HPP % Of total ecdysteroid 0 10 20 Figs: D3 Stage Reproductive Molt Cycle 60 % MeOH 50 30 40 Retention time (min) Analysis of immunoreactive ecdysteroids in the hemolymph of female prawns at stage D3 during the reproductive molt cycle by HPLC-RIA system. Hemolymph ecdysteroids were separated into fractions by reverse-phase HPLC on a methanol-water solvent system using a linear gradient, and were detected by RIA. Results are expressed as the percent of the total ecdysteroid present in the hemolymph. Arrows indicate elution . times of references; 20-hydroxyecdysone (20E), makisterone A (M), ecdysone (E), 20-hydroxyecdysone-22- acetate (A), 2-deoxy-20-hydroxyecdysone (DO20E), ponasterone A (P), 2-deoxyecdysone (DOE), respectively. HPP, high polarity products; LPP, low polarity products. ductive molt cycle is shown representatively in Fig. 3. Two peaks were identified as 20E and ecdysone, respectively, by coelution with authentic stan- dards. Several other peaks of RIA activity were also detected. However, no peaks were observed to elute at the retention times of reference com- pounds makisterone A, 20-hydroxyecdysone-22- acetate, 2-deoxy-20-hydroxyecdysone, ponaster- one A or 2-deoxyecdysone. One major unknown peak was detected prior to 20E and termed high polarity products (HPP). Small peaks which eluted after ecdysone were collectively termed low polarity products (LPP). Total ecdysteroid was calculated as the sum of all immunoreactive sub- stances eluted by HPLC. Ecdysteroid levels during the molt cycles Changes in hemolymph ecdysteroid levels dur- ing the reproductive and non-reproductive molt cycles are shown in Fig. 4. Ecdysteroid profiles differ between the two molt cycles. During the reproductive molt cycle, total levels in hemolymph declined sharply after ecdysis, and showed a small peak at stage A, after spawning. On the contrary, during the non-reproductive molt cycle, levels decreased more gently during stages A-C;. During both molt cycles, levels were low. during stages C,-Do, showing an increase during stages D,-Dp, and peaking at stage D3. 20E and HPP were the predominant im- munoreactive ecdysteroids during both molt cylces. 20E levels rapidly declined to low levels after ecdysis in the reproductive molt cycle, where- as 20E levels slowly decreased during stages A-C, in the non-reproductive molt cycle. The levels peaked at stage D3 during both molt cycles. HPP levels rapidly declined just prior to spawn- ing at stage Ap and showed a small peak at stage A, in the reproductive molt cycle. On the other hand, in the non-reproducive molt cycle, the levels Vitellogenin and Ecdysteroid in Prawns 100 Reproductive Molt Cycle — ° TOTAL ° e g ) o HPP 8 Oo m 20E o v 2 (5) 50 oF: J a 3 une 2 WwW Cc : ES Spawning o a 7) yp ™ a fe} eo © 0 eee peat Re yes eer Ag Ai B Co Ci Do Di D2 D3 Ao Molt stage Non-Reproductive Molt Cycle ” a > ° TOTAL pees Lu o HPP Vv 400 = 20E i i 4 LPP ° Ecdysteroid (ng/ml) moO as 0 o4—pa— a so4— rE ina A B Co Ci Do Dy; Molt stage D2 D3 Fic. 4. Changes in total ecdysteroid levels, and in the levels of four HPLC-separated ecdysteroids from the pooled hemolymph during the reproductive and non-reproductive molt cycles. ecdysteroid levels are shown twice. The data for stages A and Ag Ecdysteroid levels were calculated as 20-hydroxyecdysone equivalents. 20E, 20-hydroxyecdysone; E, ecdysone; HPP, high polarity products; LPP, low polarity products. were still high in stage A, and declined slowly during stages B-C,;. HPP levels reached maxima at stage D3 in both molt cycles. Ecdysone levels were low, but also peaked at stage D3 in both molt cycles. LPP was remained at trace levels throughout both molt cycles. DISCUSSION Increases in Vg levels and the advancement of ovarian development were correlated with the molt stages during the reproductive molt cycle in this species; on the other hand, vitellogenesis and ovarian development did not occur during the non-reproductive molt cycle. The observance of high Vg levels was followed by rapid GSI increases during stages C,-D3. Histological observation re- vealed that yolk globules were accumulated in the oocytes, and that oocyte diameter increased [8]. Vg levels decreased concomitanly with the cessa- tion of GSI increase at stage Ap (before spawning), suggestive of the termnation of Vg synthesis and additionally, of Vg uptake. These results concern- ing Vg levels were as expected given that the cycle of spawning is synchronous with the molt cycle in this prawn. The profile for Vg levels during the reproductive molt cycle was similar to that of hemolymph Vg levels in the giant freshwater prawn Macrobrachium rosenbergii ({10], Okumura et al., unpublished data). This correlation in vitellogenesis and molt cycle suggests that they are linked endocrinologically. During stages Ap(A)-Co, levels of 20E and HPP in the reproductive molt cycle were lower than those in the non-reproductive molt cycle. At stage B of the reproductive molt cycle, hemolymph Vg levels began to increase. These observations may suggest that lower levels of 20E and HPP are involved in the initiation of vitellogenesis. Howev- er, in this study, it is unclear whether this differ- ence in 20E and HPP levels is indicative of their participation in vitellogenesis or rather, reflective of differences in ecdysteroid metabolism between the reproductive and non-reproductive molt cy- cles. Hemolymph 20E levels peaked at stage D3 and declined just after ecdysis during the reproductive molt cycle. This rapid change was concomitant 44 T. Oxumura, C.-H. HAn et al. with the decrease of Vg levels. A similar phe- nomenon was observed in M. rosenbergii; Meusy and Payen [2] examined total ecdysteroid levels, and Okumura et al. employed HPLC-RIA analysis to determine levels of 20E, as well as of other ecdysteroid species (unpublihsed data). Further- more, 20E levels and the synthesis levels of Vg in fat body were correlated similarly in the amphipod Orchestia gammarella [11]. Such correlations are considered to suggest that hemolymph 20E partici- pates in the modulation of the termination of vitellogenesis. Ecdysteroid levels were linked to Vg levels in this study, but the physiological significance of these phenomena is not yet clear. The function of 20E in controlling vitellogenesis has been investi- gated previously. However, results in these re- ports appear inconsistent. Y-organ removal de- creased Vg synthesis in O. gammarella [5]. The arrest of oocyte growth was. observed after Y- organ ablation in the terrestrial isopod Armadilli- dium vulgare [7]. In the isopod Porcellio dilatatus, a decrease in hemolymph V¢g levels resulted after Y-organ ectomy, and this effect could be compen- sated for by 20E injection [6]. On the other hand, administration of 20E just after ecdysis inhibited the onset of vitellogenesis in L. seticaudata |12| and in O. gammarella [13]. Variances among such reports might be attributed to differing molting stages of animals used and ranges of 20E dosage utilized. It has been reported that maturing ovaries con- tain ecdysteroid in the shore crab Carcinus maenas [14], in O. gammarella [15], in the spider crab Acanthonyx lunulatus [16] and in M. rosenbergii [17], suggestive of vitellogenic ovaries as a possible site of ecdysteroid production and/or metabolism. Ovarian development may have some influence upon ecdysteroid metabolism and hemolymph ecdysteroid levels. HPP is thought to consist of substances con- verted from ecdysone via metabolic pathways, and to have no biological activity. Several types of reactions leading to high polarity ecdysteroids have been considered in ecdysteroid metabolism: hydroxylation at the C-20 and/or C-26 position, ecdysonic acid formation following hydroxylation, and conjugation (see [18]). HPP was also detected in the hemolymph of P. serratus [19], and the mysid Siriella armata [20]. Enzymatic hydrolysis has revealed that the HPP in S. armata consists mainly of conjugated forms of ecdysteroid [20]. Thus, the difference in HPP levels seen between reproductive and non-reproductive individuals appears to reflect differences in ecdysteroid me- tabolism. Hemolymph ecdysteroid levels peaked at stage D; during both molt cycles in M. nipponense. The timing of this peak was within the range of results in other crustaceans which varied from D, to D3 (see [21]). In almost all crustaceans, ecdysteroid levels decline sharply just prior to ecdysis and are maintained at low basal levels during the postmolt period. However, in M. nipponense, ecdysteroid levels showed a small peak at A, stage in the reproductive molt cycle and were relatively high during the postmolt period of the non-reproducive molt cycle. A small peak in hemolymph ecdyster- oid, in addition to a large peak prior to ecdysis, was also observed at stage B in P. serratus [19], and at the end of the intermolt period in the fiddler crab Uca pugilator [22]. The physiological implica- | tions of such patterns are not understood. In M. nipponense, as vitellogenesis cycle was in synchronization with the molt cycle, we hypothe- size that 20E may be involved in controlling both vitellogenesis and molting. However, the exis- tence of several factors for the control of vitel- logenesis are additionally postulated (see [2, 23]). It is possible that such factors also take part in the integration of vitellogenesis and molting. Further investigations are required in order to clarify the function of ecdysteroids in vitellogenesis in M. nipponense. ACKNOWLEDGMENTS We express our appreciation to Dr. Masao Nagata, The Univeristy of Tokyo for kindly providing us with antiserum. We thank Marcy N. Wilder, The University of Tokyo for reading this manuscript. This study was supported in part by Grant-in-Aid from the Ministry of Education, Science and Culture of Japan. REFERENCES 1 Adiyodi, R. G. (1985) Reproduction and its con- 10 At 12 Vitellogenin and Ecdysteroid in Prawns trol. In “The Biology of Crustacea”. Ed. by D. E. Bliss and L. H. Mantel, Academic Press, Orland FL, VOI. 9, pp. 147-215. Meusy, J.-J. and Payen, G. G. (1988) Female reproduction in Malacostracan Crustacea. Zool. Silo, 52 ZARA OS) Charniaux-Cotton, H. and Touir, A. (1973) Con- trdle de la prévitellogenése et de la vitellogenése chez la Crevette hermaprodite Lysmata seticaudata Risso. C. R. Acad. Sci. Paris, Ser D, 276: 2717- 2720. Faure, Y., Bellon-Humbert, C. and Charniaux- Cotton, H. (1981) Folliculogenése et vitellogenése secondaires chez la Crevette Palaemon serratus (Pennant); contrdle par les pédoncules ocularies et Yorgane X de la medulla externa (MEX). C. R. Acad. Sci. Paris, Ser III, 293: 461-466. Meusy, J.-J., Blanchet, M.-F. and Junera, H. (1977) Mue et vitellogenése chez le Crustacé Amphipode Orchestia gammarella Pallas. Il. Etude de la synth- ése de la vitellogénine (“fraction protéique femelle” de ’hemolymphe) aprés destruction des organes Y. Gen. Comp. Endocrinol., 33: 35-40. Souty, C., Besse, G. and Picaud, J.-L. (1982) Sti- mulation par |’ecdysone du taux hémolymphatique de la vitellogénine chez le Crustacé Isopode terres- tre Porcellio dilatatus Brandt. C. R. Acad. Sci. Paris, Ser III, 294: 1057-1059. Suzuki, S. (1986) Effect of Y-organ ablation on oocyte growth in the terrestrial isopod, Armadilli- dium vulgare. Biol. Bull., 170: 350-355. Han, C.-H. (1988) “Physiological studies on the reproductive cycle of a freshwater prawn Macro- brachium nipponense (de Haan).” Ph. D. thesis, The University of Tokyo. Okumura, T., Nakamura, K., Aida, K. and Hanyu, I. (1989) Hemolymph ecdysteroid levels during the molt cycle in the Kuruma prawn Penaeus japonicus. Nippon Suisan Gakkaishi, 55: 2091-2098. Derelle, E., Grosclaude, J., Meusy, J.-J. Junéra H. and Martin, M. (1986) ELISA titration of vitel- logenin and vitellin in the freshwater prawn, Mac- robrachium rosenbergii, with monoclonal antibody. Comp. Biochem. Physiol, 85B: 1-4. Blanchet-Tournier, M.-F. (1982) Quelques aspects des interactions hormonales entre la mue et la vitellogenése chez le Crustacé Amphipode Orchestia gammarellus (Pallas). Reprod. Nutr. Dévelop. 22: 325-344. Touir, A. and Charniaux-Cotton, H. (1974) In- fluence de l’introduction d’ecdystérone sur |’exuvia- tion et la démarrage de la vitellogenése chez la Crevette Lysmata seticaudata Risso. C. R. Acad. 13 14 IS) 16 7) 18 19 20 21 LD, yep) Sci. Paris, Ser D, 278: 119-122. Blanchet, M.-F., Junéra, H. and Meusy, J.-J. (1975) Mue et vitellogenése chez Orchestia gammarella (Crustacé Amphipode): étude de la synthése de la fraction protéique femelle aprés introduction d’ec- dystérone. Experimentia, 31: 865-867. Lachaise, F., Goudeau, M., Hetru, C., Kappler, C. and Hoffmann, J. A. (1981) Ecdysteroids and ovarian development in the shore crab, Carcinus maenas. Hoppe-Seyler’s Z. Physiol. Chem., 362: 521-529. Blanchet, M.-F., Porcheron, P. and Dray, F. (1979) Variations du taux des ecdystéroides au cours des cycles de mue et de vitellogenése chez le Crustacé Amphipode Orchestia gammarellus. Int. J. Inver- tebr. Reprod., 1: 133-139. Chaix, J.-C. and De Reggi, M. (1982) Ecdysteroid levels during ovarian development and embryogene- sis in the spider crab Acanthonyx lunulatus. Gen. Comp. Endocrinol., 47: 7-14. Wilder, M. N., Okumura, T. and Aida, K. (1991) Accumulation of ovarian ecdysteroids in synchro- nization with gonadal development in the giant freshwater prawn, Macrobrachium _ rosenbergii. Zool. Sci., 8: 919-927. Lafont, R. and Koolman, J. (1984) Ecdysone metab- olism. In “Biosynthesis, Metabolism and Mode of Action of Invertebrate Hormones”. Ed. by J. A. Hoffmann and M. Porchet, Springer-Verlag, Berlin, pp. 196-226. Baldaia, L., Porcheron, P., Coimbra. J. and Cas- sier, P. (1984) Ecdysteroids in the shrimp Palaemon serratus: Relation with molt cycle. Gen. Comp. Endocrinol., 55: 437-443. Cuzin-Roudy, J., Strambi, C., Strambi, A. and Delbecque J.-P. (1989) Hemolymph ecdysteroids and molt cycle in males and females of Siriella armata M.-Edw. (Crustacea: Mysidacea): possible control by the MI-ME X-organ of the eyestalk. Gen. Comp. Endocrinol., 74: 96-109. Skinner, D. M. (1985) Molting and regeneration. In “The Biology of Crustacea”. Ed. by D. E. Bliss and L. H. Mantel, Academic Press, Orland FL, Vol. 9, pp. 43-146. Hopkins, P. M. (1983) Patterns of serum ecdyster- oids during induced and uninduced proecdysis in the fiddler crab, Uca pugilator. Gen. Comp. Endocri- nol., 52: 350-356. Charniaux-Cotton, H., and Payen, G. (1988) Crustacean reproduction. In “Endocrinology of Selected Invertebrate Types”. Ed. by H. Laufer and R. G. H. Downer, Alan R. Liss, New York, pp. 279-303. aa ha “apa Ap Sh ase hes naps a deat es Fototald : be pfs ty yng a) bas fee iors a | Hive gibi iis a. | te ere yale SEP sia Y. Je 2 oe Mtoe Ao, hiner Di afl apres) | meee s aap i i iv ea ang an Li icoeteil pees Py tah. é (tds 2 b . i ons " f ‘ Ae he * re bo | a : Sea oe ai se “ EF (ate ee: ; 7 xs i sah} TA ; gees oe ibe e ie peyote) vee : ; ag eG RMT) ot: Bata ny ZOOLOGICAL SCIENCE 9: 47-52 (1992) © 1992 Zoological Society of Japan Effects of Ba?* on the Photoresponses of Isolated Single Rods from Frog Retina Katsu AZUMA and NAOHIKO IWASAKI Department of Biology, Osaka Medical College, Sawaragicho 2-41, Takatsuki, Osaka 569, Japan ABSTRACT—Effects of extracellular barium (Ba**) on the photoresponses of mechanically isolated single rods from the retina of the frog (Rana catesbeiana) were investigated by sucking the rod inner segments into tightly fitting pipettes. The addition of Ba** to a control solution prolonged the duration of the rod response to a flash of light (flash response), and increased its sensitivity to the flash (flash sensitivity). The response of the rod to a step of light (step response) in the control solution ascended to an initial peak, and then descended to a lower steady level within a few seconds (relaxation), showing light adaptation. Ba** abolished the relaxation of the step response. The flash sensitivity of the rod in the Ba**-solution drastically decreased with increasing intensity of adaptation light in the manner different from that in the control solution. These effects were specific to Ba**, Sr°* or Co** being ineffective. The effects were not explicable by the role of Ba** as a K*-blocker, because other K* -blockers, tetraethylammonium chloride (TEA), 4-aminopyridine (4-AP) and Cs* were ineffective. INTRODUCTION In the previous paper, we reported that lowered extacellular calcium concentration shortened the recovery phase of the photoresponse of isolated single rods from frog retinas [1]. Ca*t has an important role in light adaptation in rods and cones [2]. On the other hand, Brown and Flaming [3] showed by the intracellular recording in retinal rods of toad that Ba** induced several effects on the rod photoresponses. They described as fol- lows. Ba’t increased the rod sensitivity and markedly delayed the decay of the rod response. Ba’? also induced the oscillations of membrane potentials of the rods. Therefore, we intended to investigate whether or not Ba** would affect on the rod under no interaction with other rods or second order cells. In this study we examined the effects of Ba** on mechanically isolated single rods from the frog retina using the suction pipette method. We shall first demonstrate that Ba** prolongs the decay of the flash response and also increases the flash sensitivity. Next we shall show that Ba*t Accepted September 25, 1991 Received July 16, 1991 abolishes the descending phase of the step re- sponse and causes a significant effect on the prop- erties of the light adaptation in isolated single rods. MATERIALS AND METHODS The experiments described in this paper were very similar to those in the already published one [1]. Adult bullfrogs were dark-adapted overnight. The animals were double-pithed and then the eyeballs were enucleated. Retinas were isolated from the enucleated eyeballs. The isolated retina was placed in normal physiological saline in a Petri dish and shaken to detach the rods. The rod suspension in the saline (about 0.5 ml) was put into the measuring chamber using a disposable syringe. The chamber was set on the stage of the inverted microscope (Nikon-MAD). A single rod which had outer and inner segments was sucked at the inner segment following the methods described by Hodgkin et al. [4]. All these procedures were carried out under infrared light using an infrared image coverter (NEC-NVR2015). The physio- logical saline was used as a control superfusate which contained (in mM) NaCl, 85; KCl, 2.5; NaHCOs, 25; MgCh, 1.2; glucose, 25; HEPES, 3, and CaCl, 1, and was adjusted to pH7.6 by 48 K. AZUMA AND N. IWASAKI adding 1N-HCl. Experiments on the effects of Ba’* were carried out by adding various concen- trations of BaCl, to the control superfusate (hereafter, called Ba**-solution). The flow rate of superfusion was 2 ml min '. The temperature of the superfusate was maintained at 20+1°C. The apparatus for optical stimulation was essen- tially the same as that described by Baylor et al. [5]. The unattenuated light intensity for flash (20 ms) or step of 500-nm light was 710° photons yum *s_'. The stimulus intensity was expressed by the negative logarithm of the neutral density filter. RESULTS Effects of Ba** on the flash response The upper recordings in Figure 1 are the flash responses of a single rod superfused with the control solution. The flash intensities were indi- cated by the numerals under horizontal lines. After recording these responses, the superfusate was switched to 0.2 mM Ba’*-solution. As shown by the middle recordings in Figure 1, Ba** pro- longed the durations of the flash responses at both intensities of —4.0 and —3.0 in log unit. Ba** also markedly increased the amplitude of the flash response as seen at intensity of —4.0 log unit. It should be noticed that there is no oscillation of membrane current at any intensity of flash stimu- lus. After 30 min of superfusion with the Ba7+ -solution, the same rod was again superfused with the control solution. After 40 min of the superfu- sion (washout-period), each response almost rec- overed to its control waveform, as shown by the lower recordings in Figure 1. It was found that the prolongation of responses was induced by Ba**+ above 0.1 mM (data not shown). Figure 2 is the response-intensity plots obtained from the experiments similar to Figure 1. The peak amplitudes of flash responses at the control, 0.2mM Ba’? and the recovery were plotted against the longarithm of the relative intensitites of the flahses. The peak amplitudes of all responses were normalized against those of corresponding maximal responses. The smooth curves, 1, 2 and 3, were obtained from calculations based on the following quation [6]: control \ 0.2 mM Ba2t < BE GOVERY, Flash na | ee ere, -4.0 Fic. 1. Effects of Ba** on the flash responses. The upper recordings are control flash responses at the. intensities indicated by the numerals under horizon- tal lines. After recording these responses, the su- perfusate was changed from the control solution to 0.2mM_ Ba’*t-solution. The middle responses were recorded after 20 min of the superfusion with the Ba**-solution. The lower responses were re- corded after 40 min of recovery in control solution. R/Ro ee) Here R represents the peak amplitude of each response, with maximum (R,ax), and I represents flash intensity with half-saturating intensity (Io). The constants, Ip(n) were determined by using the log-linear, least-squared error method to fit each set of data points. When the value of n was 1.3, the values of Ip at the control, 0.2 mM Ba’* and the recovery were —3.7, —4.2 and —3.8, respec- tively. Thus 0.2mM Ba*t shifted Ig into the direction of lower intensity by 0.5 log unit, which indicated an increase in the rod sensitivity. Effects of Ba** on the step response Figure 3 shows the step responses at various Ba**-effect on isolated single rods 140 5 Control 2: 0.2 mM Bact W S recovery = Ss =>} O ne) Sj is Ons oO = See (@) = 0 -4 =. log relative intensity Fic. 2. Effects of Ba** on the response-intensity relationship. These results were obtained from the experiments similar to those used in Figure 1. The measurements were carried out consecutively with the same rod. Curve 1: control, Curve 2: 0.2 mM Ba**, Curve 3: recovery. The peak amplitudes of the flash response are plotted versus the logarithms of relative intensity, normalized against the peak amplitude of the corresponding maximal responses. -5.6 10 | pA -5.0 -4.6 -4.0 =3.0 60 Ss Fic. 3. Step responses recorded in the control solution. The intensity of step light was varied from —5.6 to —3.0 in log unit, as indicated to the upper of each response. The duration of each step light is indi- cated by a horizontal line as 60s. intensities in control solution. The step responses at the lower intensities, —5.6, —5.0 and —4.6 in log unit, rose to the initial peaks then rapidly recovered to the lower levels (relaxation) and tended to gradually recover to the dark levels. The relaxation was usually and markedly observed at the intensity of —5.0 log unit. At the stronger intensities, —4.0 and —3.0 in log unit, both re- sponses hardly indicated the relaxations and were steady at their peak levels. The upper recording in Figure 4 is the step response at the intensity of —5.0 log unit in the control solution. After the recording, the rod superfusate was changed to 0.5mM_ Ba*?- solution. The middle recording was obtained after 20 min of superfusion with the Ba’ *-solution. The recording showed that Ba** completely abolished the relaxation of the step response. The same rod was again superfused with the control solution after the recording. The step response almost recovered to its control waveform after 40 min of the superfusion, as shown by the lower recording in Figure 4. 50 K. AZUMA AND N. IWASAKI COMEGO JL -5.0 0.5 mM Ba’* 10 pA recovery 60 Ss Fic. 4. Effects of Ba*t on the step responses. The upper recording is the step response in the control solution. After recording this response, the super- fusate was changed from the control solution to 0.5 mM Ba’*-solution. The middle recording was obtained after 20 min of continuous superfusion with the Ba**t-solution. The lower recording was obtained after more than 40 min of recovery in the control solution. The intensity and duration of the step light are —5.0 log unit and 60s (indicated by a horizontal line), respectively. Light adaptation of the rod The flash responses of a single rod were re- corded at various intensities in the control solution and in the dark. Following these recordings, the responses to the flashed superimposed on the background light at the intensity of —5.5 log unit were recorded. Finally the flash responses were again recorded in the dark after turning-off the background light. The response-intensity plots obtained from those experiments are indicated in Figure 5A. The peak amplitudes of all flash re- sponses were plotted against the logarithms of the relative intensities of flashes. The peak amplitude of each flash response in the initial dark period. Smooth curves, 1, 2 and 3, were drawn after the manner of the case in Figure 2. Io(n) of curves, 1, 2 contro] normalized current =) i= (eB) = = 3 2) fo) = 0.5 ‘S = fe} Cc 0 log relative intensity Fic. 5. Response-intensity relationships obtained from - two rods. A, control solution: B, 0.5mM Ba??. Curves 1 were obtained from the flash responses at various intensities in the dark. Curves 2 were obtained from the responses to the flashés superim- posed on the background light intensity of —5.5 log unit. Curves 3 were obtained from the flash re- sponses recorded during 10 to 20 min after turning- off the background light. The peak amplitudes of all flash responses are plotted against the logarithms of the relative intensities of flashes. The peak ampli- tude of each flash response is normalized to that of a maximum flash response in the initial dark period. and 3, were —3.7 (1.1), —3.3 (0.9) and —3.7 (1.1), respectively. The background light induced a little decrease in maximum response amplitude and shifted Ip into the direction of higher intensity by 0.4 log unit. The response-intensity plots in Figure 5B were obtained from the experiments similar to those used in Figure 5A except the presence of Ba’*. Ip (n) of curves, 1, 2 and 3, were —4.1 (1.8), —3.6 (1.0) and —4.1 (1.6), respective- ly. The background light induced a significant decrease in maximum response amplitude and Ba’ *-effect on isolated single rods shifted Ip into the direction of higher intensity by 0.5 log unit. Figure 6 was obtained from the experiments similar to those described in Figure 5 (A and B). In Figure 6 the flash sensitivity (log unit) during the background light was plotted against the log relative background intensity in control solution or 0.5 mM Ba’t-solution. Flash sensitivity during the background light was defined as the logarithm of the ratio of the flash intensity needed to elicit 4 pA-photocurrent in the dark to that during the background light. We restricted the experiments presented in Figure 6 to those in which a nearly full recovery in the flash response was obtained after turning-off the background light. In control solu- tion (circles) flash sensitivity against background intensity showed the slope of approximate 1 rang- ing from —5.5 to —3.5 in log unit, where Weber’s law held. On the other hand, the slope in the case of Ba**-solution (triangles) is 2.8 ranging from — 5.5 to —4.5 in log unit, where Weber’s law did not hold. Thus Ba*t markedly modified the flash sensitivity of a single rod under background light. control flash sensitivity (log units) -/ -6 -5 -4 -3 log relative background intensity Fic. 6. Flash sensitivities (log unit) plotted against the logarithms of the intensities of background light in the control (circles) and 0.5mM Ba** (triangles) solutions. These results were obtained from the experiments similar to those described in Figure 5 (A and B). Results in the control and Ba** solu- tions were obtained from 4 cells and 3 cells, respec- tively. The data points and vertical bars indicate averaged values and standard deviations, respec- tively. DISCUSSION As already described, Ba** of 0.2 mM increased the rod sensitivity by 0.5 log unit (see Figure 2). This result was close to that shown by Brown and Flaming using the intracellular recording in retinal rods of toad [3]. They described that Ba’* of 0.6 mM increased the rod sensitivity by 0.4 log unit. Ba** also prolonged the decay of the single rod response but induced no oscillation in membrane current of the single rod. Thus this study com- pletely supported the description of Brown and Flaming [3]; i.e., Ba**-effects except the oscilla- tion were directly exerted on the rod. New findings in this study were as follows. Ba** abolished the relaxation of a step response (see Figure 4). The flash sensitivity of a single rod in the Ba**-solution drastically decreased with in- creasing intensity of background light, where We- ber’s law did not hold (see Figure 6). Thus Ba** also caused significant effects on the proporties of light adaptation in a single rod. We tested whether or not other divalent cations, Ca**, Mg’*, Sr?* and Co*+t, would induce similar effects on the rods. These ions were less effective except lower Ca** which enhanced the relaxations of step re- sponses (data not shown). It has been well known that Ba’? blocks K* -conductances of rods [3, 7] and Miiller cells [8]. Fain & Quandt [7] found that extracellular TEA (6-12 mM) and 4-AP (10 mM) blocked the K* -outward current of toad rods. Bader et al. [9] reported that extracellular TEA, Cs* and Co? blocked the current carried predominantly by K* in solitary rod inner segments from the salamander retina. However, we found that 10 mM 4-AP, 12 mM TEA and 5 mM Cs* did not indicate effects similar to those of Ba’? , whether the rod outer segments or inner segments were superfused with the physiological solution containing those K~* -blockers (data not shown). Probably the effects of Ba’? is not due to the inactivation of the KT -conductance. However, we can not thoroughly exclude the possibility, because we have not yet tested the intracellular actions of TEA, 4-AP and Csi: According to the prevailing model for the light adaptation of rod photoreceptors, a light-induced 52 K. AZUMA AND N. IWASAKI decrease in intracellular Ca** acts on guanylate cyclase and causes an increase in cyclic GMP level, which opposes the light-activated decline in the cyclic GMP level (negative feedback) [10]. The relaxations of step responses may be related to the increase in the cyclic GMP level caused by the light-induced decrease in intracellular Ca** [2]. On the other hand, it has been found that Ba** evokes an increase in cyclic GMP in the cilia of Paramecium and the Ba**-effect is specific as Mg’*, Ca** or Sr** does not elicit an increase in cyclic GMP [11]. Although such effects have not been known in the rods, the findings in Para- mecium are suggestive taking it into consideration that the outer segment of rod is phylogenetically a rudimentary ciltum. It is not unreasonable that Ba’* acts on the enzymatic process related to cyclic GMP and inhibits the negative feedback. Although the mechanism of the Ba** effect is not yet elucidated, the Ba** manipulation may be useful to study the behaviour of rods in the ab- sence of the negative feedback underlying light adaptation. REFERENCES 1 Azuma, K. (1989) Effect of low extracellular cal- cium concentration on photosensitivity of isolated rods from frog retina. J. exp. Biol., 145: 255-262. 2 Nakatani, K. and Yau, K.-W. (1988) Calcium and 10 Lh light adaptation in retinal rods and cones. Nature 334: 69-71. Brown, K. T. and Flaming, D. G. (1978) Opposing effects of calcium and barium in vertebrate rod photoreceptors. Proc. Natl. Acad. Sci. USA 75: 1587-1590. Hodgkin, A. L., McNaughton, P. A., Nunn, B. J. and Yau, K.-W. (1984) Effect of ions on retinal rods from Bufo marinus. J. Physiol., Lond., 350: 649- 680. Baylor, D. A., Lamb, T. D. and Yau, K.-W. (1979) The membrane current of single rod outer segments. J. Physiol., Lond., 288: 589-611. Naka, K. I. and Rushton, W. A. H. (1966) S- potentials from colour units in the retina of fish (Cyprinidae). J. Physiol., Lond., 185: 536-555. Fain, G. L. and Quandt, F. N. (1980) The effects of tetraethylammonium and cobalt ions on responses to extrinsic current in toad rods. J. Physiol., Lond., 303: 515-533. Bolnick, D. A., Walter, A. E and Sillman A. J. (1979) Barium suppresses slow PIII in perfused bullfrog retina. Vision Res., 19: 1117-1119. Bader, C. R., Bertrand, D. and Schwartz, E. A. (1982) Voltage-activated and calcium-activated cur- rents studied in solitary rod inner segments from the salamander retina. J. Physiol., Lond., 331: 253-284. Yau, K.-W. and Nakatani, K. (1985) Light-induced, reduction of cytoplasmic free calcium in retinal rod outer segment. Nature, 313: 579-582. | Schultz, J. E., Thomas, P. and Klumpp, S. (1986) Voltage-gated Ca** entry into Paramecium linked to intraciliary increase in cyclic GMP. Nature, 322: 271-273. ZOOLOGICAL SCIENCE 9: 53-64 (1992) Proctolin-Like Immunoreactivity in the Dorsal Unpaired Median Neurons Innervating the Accessory Gland of the Male Cricket, Gryllus bimaculatus Koust YAsuyaMa!, TETSUYA KimurA* and TSUNEO YAMAGUCHI’’* "Department of Biology, Kawasaki Medical School, Kurashiki 701-01, and *Department of Biology, Faculty of Science, Okayama University, Okayama 700, Japan ABSTRACT—The dorsal unpaired median neurons (DUMR’7s) bilaterally extend bifurcating axons to the accessory gland via the left and right branches (Br3s) arising from the seventh nerve roots of the terminal abdominal ganglion in the male cricket, Gryllus bimaculatus. Proctolin-like immunoreactivity (PLI) was demonstrated in these DUMR7 neurons by the postembedding immunogold method combined with retrograde horseradish peroxidase (HRP) labeling and ligation of one of the Br3s. Ligation of one Br3 and HRP-labeling of the opposite Br3 produced labeling of both the somata of DUMR7 neurons and the ligated axons containing numerous granular vesicles (100 nm in diameter). PLI was found within the granular vesicles (SO—200 nm in diameter) in HRP-labeled DUMR7 somata as well as within similar granular vesicles accumulated in the HRP-labeled axons. Axons containing granular vesicles (100-115 nm in diameter) with PLI were noted to associate extensively with the visceral muscles winding around the accessory gland. It is therefore evident that the HRP-labeled axons with accumulation of granular vesicles originated from the DUMR7 neurons on the basis of the characteristic bilateral morphology of these neurons. These findings indicate the possibility that DUMR7 neurons produce a proctolin-like substance and transport it through their axons to the accessory gland musculature, where this substance may serve as a neuromuscular transmitter or © 1992 Zoological Society of Japan cotransmitter. INTRODUCTION Our interest lies in investigating the neural con- trol of visceral muscles associated with the male cricket accessory gland, which produces various spermatophore-forming materials and then re- leases them at a certain period in the mating cycle according to a temporal motor program [1]. Our previous morphological and electrophysiological studies have suggested a proctolinergic innervation for the accessory gland musculature [2, 3]. Proctolin was originally isolated from the cock- roach, Periplaneta americana [4, 5], and is a pen- tapeptide with the structure Arg-Tyr-Leu-Pro-Thr- OH [6]. This peptide was initially proposed to be a neurotransmitter associated with the hindgut visceral muscle [4, 5]. Subsequently, proctolin has been detected in many arthropods by high- Accepted October 5, 1991 Received September 10, 1991 * To whom reprint requests should be addressed. performance liquid chromatography (HPLC), im- munocytochemistry, and radioimmunoassay (re- viewed in [7]), and multiple physiological functions have been proposed for this peptide. These in- clude action as a peripheral neurotransmitter or cotransmitter influencing on visceral and skeletal muscles [e.g., 8-16], action as a central neurot- ransmitter or neuromodulator [e.g., 17-23], and a role as a neurohormone [e.g., 24, 25]. With respect to the visceral musculature of the oviduct, there are a number of pieces of evidence for the myotropic action of proctolin in cock- roaches [14, 15], locusts [13], diptera [9, 26], and hemiptera [16]. By the use of HPLC and bioassay, the presence of proctolin in this tissue and its release by high potassium saline and neural sti- mulation have been demonstrated [16, 27-30]. Furthermore, immunocytochemical studies using antisera to proctolin have revealed proctolin-like immunoreactive nerve fibers in the locust oviduct and neurons with proctolin-like immunoreactivity 54 K. YASUYAMA, T. KIMURA AND T. YAMAGUCHI (PLI), which may be related to these fibers, in the penultimate (seventh) abdominal ganglion [28, 31]. The musculature of the cricket accessory gland is innervated by bilaterally paired neurons (LC neurons) and dorsal unpaired median neurons (DUMR7 neurons), with the axons running through the branches (Br3s) of the bilateral seventh nerve roots arising from the terminal abdominal ganglion (TAG) [2]. Selective antidro- mic stimulation of these DUMR7 and LC neurons evokes contraction of the accessory gland which Squeezes out spermatophore-forming material from the glandular tubules, while orthodromic stimulation of both types of neurons evokes not only contraction but also a transient reduction of the basal tonus after relaxation [3]. Synthetic proctolin produces sustained contraction of the accessory gland at a concentration as low as 107” M [3]. Using HPLC and bioassay, it has been demonstrated that proctolin is present in the inner- vated accessory gland, but not in the gland without innervation, and that this peptide is released from the accessory gland in response to the application of high potassium saline [3]. These findings sug- gest that the DUMR7 neurons are proctolinergic excitatory motoneurons and the LC neurons are inhibitory motoneurons [3]. In the present paper, we will provide additional evidence for the proctolinergic innervation of the accessory gland musculature in the male cricket, as a result of studies involving electron microscopic immunocytochemistry combined with retrograde HRP labeling. These studies have demonstrated that PLI is present in the DUMR7 neurons inner- vating the the accessory gland musculature. MATERIALS AND METHODS Animals The TAGs and accessory glands of adult male crickets (Gryllus bimaculatus) were used in this study. The animals were obtained from a labora- tory culture reared under a 12 hr light-12 hr dark Giles a ZC Antiserum preparation and testing We obtained rabbit anti-proctolin antiserum from Seikagaku Kogyo Co. (Osaka, Japan). The rabbits were immunized with glutaraldehyde- linked conjugates of synthetic proctolin and bovine serum albumin (BSA) or bovine thyroglobulin [11, 32, 33]. Proctolin antiserum was purified by affini- ty chromatography according to the method of Eckert and Ude [33]. The purified anti-proctolin- BSA serum was mainly used in light microscopic immunochemistry, and the purified anti-proctolin- thyroglobulin serum was used for the electron microscopic study. Controls were performed as follows: purified proctolin antiserum was preincu- bated with synthetic proctolin at a concentration of 0.5 mg/ml diluted antiserum for 16 hr, and then used to process sections for light and electron microscopy. Tests were also performed in which the primary antiserum was left out of the staining procedure. All immunoreactivity was eliminated by such treatment. Light microscopic immunocytochemistry Each of the left and right nerve branches (Br3s) arising from the seventh nerve roots of the TAG was tied by a single knot using a hair. After 6 hr, the ganglion was isolated together with the ligated Br3s. Both the ligation and the dissection proce- dures were undertaken in buffered cricket saline [3]. The tissue was then fixed for 15 hr at 4°C ina glutaraldehyde-picric acid mixture [34] containing 2.5% glutaraldehyde and 15% saturated picric acid in 0.1 M phosphate buffer (pH 7.3). After washing for 4hr in the same buffer, the tissue was dehy- drated, embedded in paraffin, cut into 10 um sections, and mounted on gelatin-coated slides. Immunoreactivity was visualized using the Vectas- tain avidin-biotin-peroxidase complex (ABC) method (Vector Lab.). After deparaffinization and rehydration, the sections were rinsed in 0.01 M phosphate-buffered saline (PBS, pH 7.4) for 10 min, incubated in methanol (100%) with 0.3% H,O, for 30 min to block endogenous peroxidase activity, and rinsed again in PBS. The sections were then incubated for 30 min in normal goat serum diluted to 1:50 with PBS, and were subse- quently incubated with the primary antiserum for Proctolin-Like Immunoreactivity in DUM Neurons 55 48 hr at 4°C. Proctolin antiserum was used after dilution from 1:200-1:800 with PBS containing 0.5% BSA and 0.01% Triton X-100. The sections were washed three times in PBS for 5 min each, incubated at room temperature for 40 min in biotinylated goat anti-rabbit IgG diluted with PBS (1: 200), and washed further three times in PBS for 5 min each. Thereafter, the sections were incu- bated with the PBS-diluted ABC reagent (1: 100) for 1-2 hr, washed three times in PBS for 5 min each, and incubated for 5-30 min with 0.05% 3,3'-diaminobenzidine tetrahydrochloride (DAB) and 0.01% HO; in 0.05 M Tris buffer (pH 7.2). Finally, the sections were washed in distilled water for 15min, dehydrated, and mounted in Per- mount. Electron microscopic immunocytochemistry com- bined with retrograde HRP labeling One of a pair of Br3s was tied by a single knot using a hair, and the proximal cut end of the other Br3 was plunged into a thin capillary filled with 5% HRP for 24hr at 4°C, and then the TAG plus attached Br3s were fixed in 2.5% glutaraldehyde in 0.1M cacodylate buffer (pH. 7.2) for 3hr, fol- lowed by washing overnight in the buffer. Next, DAB enzyme reaction was performed according Nassel [35]. After the enzymatic reaction, the tissues were washed in the same buffer for 5 hr, dehydrated and embedded in epoxy resin (Luveak, Nakarai Tesque). No osmium postfixation was carried out. For electron microscopic immunocy- tochemistry without HRP tracing, the accessory glands or TAG (with Br3s ligated for 6 hr) were fixed for 3 hr after dissection, washed in buffer, dehydrated, and embedded in the resin. The postembedding protein A-colloidal gold method [36] was employed as follows: ultrathin sections were collected on formvar-carbon coated nickel grids and soaked for 30 min with drops of PBS with 1% BSA. The proctolin antiserum was then applied at 1:200-1:1,000 (in PBS with 1% BSA for 24hr at 4°C). Afterwards, the grids were attached magnetically to the inner wall of a beaker and washed in PBS by furious agitation with a magnetic stirrer for 30 min. The protein A-gold complexes (Janssen Pharmaceutical; particle size: 10 or 15 nm) were applied at a concentration of 1:40 (in PBS with 1% BSA) for 1 hr at room temperature. After washing (as above) for 40 min, the sections were stained with uranyl acetate and lead citrate. For the sections of HRP labeled tissues, lead citrate staining was eliminated. Final- ly, sections were examined with a Hitachi H 500 electron microscope. RESULTS Proctolin-like immunoreactive neurons in the TAG The proctolin antiserum revealed both of bilaterally symmetrical and medial immunoreac- tive neurons in the TAG. Symmetrically arranged neurons with proctolin-like immunoreactivity (PLI) were prominent in embryonic segments 7 and 9 of the TAG, which are derived from diffe- rent segmental ganglia during embryonic develop- ment [37]. Figure 1 shows an example of dorso- laterally situated bilaterally symmetrical neurons with PLI in the region corresponding to segment 8. Other neural elements with PLI are also visualized in both the dorsal commissure and the dorsolateral neuropil. These neurons and neural elements were constantly stained in all the preparations. Medial neurons with PLI were observed in the dorsal and ventral midline of the TAG, although the number of neurons and the degree of immunoreactivity varied from preparation to preparation. Figure 2 shows that the dorsomedial neurons with PLI are located in the caudal region corresponding to segment 10 of the TAG. In male crickets, the dorsal unpaired median (DUMR7 neurons) innervating the accessory gland form three clusters along the midline in positions corres- ponding to segments 8-10 or 11 of the TAG [2, 38]. Dissection or ligation of the nerve branches (Br3s) through which the DUMR7 neurons send their axons to the accessory gland, led to inten- sified immunoreactivity of the medial neurons in the caudal TAG. In the cockroach, it has been reported that dissection of the nerve running from the sixth abdominal ganglion to oviduct results in an increase of PLI in the ganglion [39]. The intensification of PLI in the caudally located me- dial neurons following the dissection or ligation of Br3s suggests that some of these medial neurons neurons 56 K. YASUYAMA, T. KIMURA AND T. YAMAGUCHI Fic. 1. A pair of dorsolateral, proctolin-like immunoreactive neurons (arrows) in the terminal abdominal ganglion. Immunoreactive fiber profiles are also seen in the dorsal commissure (arrowheads) and in the dorsolateral neuropil (double arrowheads). may be DUMR7 neurons. The section shown in Figure 2 was made from the cricket in which the Br3s were ligated for 6hr before fixation. The medial neurons were generally stained rather weakly, but the ventral commissure and dorsome- dial neuropil constantly revealed strong PLI (Fig. 2). This ventral commissure carries the bilaterally bifurcating axons of DUMR7 neurons in the caud- al part of the ganglion (Fig. 3). This observation supports the possibility that some DUMR7 neurons are proctolinergic. Electron microscopic observation of proctolin-like immunoreactive neurons The protein A-gold method revealed the wide distribution of the axons with PLI in the muscula- ture associated with the accessory gland, 1.e., in the single-fibrillar muscles winding around most of each glandular tubule, as well as in the multifibril- lar thick muscles surrounding the opening of each tubule in the anterior portion of the ejaculatory duct. PLI was restricted to large electron-dense granular vesicles with a diameter of 100-115 nm in the axons (Fig. 4). These immunoreactive vesicles were also detected in the ligated Br3s (Fig. 5). Our previous study [2] showed that the ligation of Br3s caused a heavy accumulation of large elec- Cross-section at the level of segment 8. x 140. tron-dense granular vesicles in the axoplasm pro- ximal to the ligature. Figure 5 shows an axon containing proctolin-like immunoreactive vesicles — (ca. 100 nm in diameter) which accumulated after ligation. These findings suggest that the proctolin- like immunoreactive vesicles are produced by neurons in the TAG. It should also be noted in Figure 5 that there is a non-immunoreactive axon with smaller granular vesicles (ca. 80 nm in dia- meter) than those in the immunoreactive axons. In order to localize PLI in the DUMR7 neurons, electron microscopic immunocytochemistry com- bined with HRP labeling and ligation was per- formed. When one of the Br3s was ligated and HRP was applied to the other Br3, labeling was noted of both the somata of DUMR7 neurons and the axons at the site of ligation, where numerous large granular vesicles accumulated at the ligature. Figure 3 shows HRP-labeled DUMR7 neurons and their secondary neurites running through the later- al commissure located in the caudal region of TAG. The protein A-gold method was applied to the region of the TAG shown in Figure 3. A few proctolin-like immunoreactive, electron-dense granular vesicles were found dispersed in the HRP labeled somata (Fig. 6). The size of the vesicles with PLI ranged from 50 to 200 nm and the mean Proctolin-Like Immunoreactivity in DUM Neurons Fic. 2. Dorsomedial proctrolin-like immunoreactive neurons (arrows) in the caudal region of the terminal abdominal ganglion. Cross-section. 230. Fic. 3. HRP-labeled DUMR7 neurons in the caudal region of the terminal abdominal ganglion. Immunoreactive fiber profiles are also seen in the ventral commissure (arrowheads). Retrograde HRP lebeling through the right Br3 revealed a number of DUMR7 somata and their axons (arrowheads) running through the ventral commissure in the caudal region. 2. Cross-section. 230. diameter was 108 nm (n=88). Vesicles with PLI were seen to associate with Golgi bodies, but were not seen in the cisternae of these structure. Proc- tolin-like immunoreactive vesicles were also seen close to lysosome-like structures (Fig. 7), as pre- viously described for PLI neurons in the cockroach hypocerebral ganglion [23]. Application of the The level of this section corresponds to that shown in Fig. protein A-gold method to the ligated Br3, revealed PLI in the large electrondense granular vesicles (ca. 100 nm in diameter) accumulated in the HRP labeled axons (Fig. 8). PLI could not be localized in the somata and axons strongly labeled with HRP, as the dense HRP reaction products prevented the identifica- 58 K. YASUYAMA, T. KIMURA AND T. YAMAGUCHI Fic. 4. An axon (ax) containing proctolin-like immunoreactive, electron-dense granular vesicles and running within the proximal part of the accessory gland tubules where multilayered muscle fibers (mf) are located. HRP labeling was not performed. The section was stained with uranyl x 50,000. Part of a transverse section of a ligated Br3. Numerous vesicles are accumulated in the axons (ax1, ax2) at the labeling was restricted to the vesicles. acetate for 5 min and with lead citrate for 1 min. Fic. 5. Immunogold site of ligation. An axon (ax1) containing large electron-dense granular vesicles with proctolin-like immunoreac- tivity, is seen, together with an axon (ax2) containing smaller non-immunoreactive vesicles. HRP labeling was not performed. The section was stained with uranyl acetate for 5 min and with lead citrate for 1 min. 48,000. tion of vesicles or the cellular organelles. Howev- er, aS shown in Figures 6 and 8, PLI could be detected in the electron-dense granular vesicles of moderately HRP labeled somata and axons. Although PLI was weaker in the vesicles of HRP- labled axons when compared with unlabled axons (Fig. 5), the gold label could still be clearly seen present over the accumulated vesicles in the for- mer group of axons (Fig. 8). It seems obvious that HRP applied to severed Br3s was transported to the contralateral axons of DUMR7 neurons through their somata in the Proctolin-Like Immunoreactivity in DUM Neurons 59 Fics. 6, 7. Parts of HRP-labeled somata of DUMR7 neurons in the caudal region of the terminal abdominal ganglion. (Fig. 6). 40,000. Proctolin-like immunoreactive, electron-dense granular vesicles are found dispersed in the cytoplasm Immunoreactive vesicles are also seen close to a lysosome-like structure (ly) (Fig. 7). The sections were stained with uranyl acetate alone for 1 min. x 40,000. Fic. 8. Proctolin-like immunoreactive vesicles in the HRP-labeled axon (ax) within the ligated Br3. This section was obtained from a preparation in which one Br3 was ligated and the other Br3 was retrogradely labeled with HRP. HRP labeled the axons with accumulated vesicles in the ligated Br3. granular vesicles is seen. TAG due to the characteristic bilateral morpholo- gy of these neurons. Therefore, these results suggest that some DUMR7 neurons are procto- linergic: a proctolin-like substance may be pro- duced by these DUMR7 neurons and transported through their axons to the muscles as a neuro- muscular transmitter or cotransmitter. The section was stained with uranyl acetate alone for 1 min. Gold labeling of electron-dense x 40,000. DISCUSSION A number of studies have provided evidence that proctolin plays a role as a neuroregulatory substance in the visceral muscles of the female reproductive organs of insects [e.g., 9, 13-16, 26]. Our present and previous studies [2, 3] have concerned the proctolinergic innervation of the visceral muscles associated with the male repro- ductive organs of crickets. 60 K. YASUYAMA, T. KIMURA AND T. YAMAGUCHI The previous study demonstrated that the crick- et accessory gland muscles are innervated by dor- sal unpaired median neurons, DUMR7_ neurons, with axons running bilaterally through Br3s of the TAG seventh nerve roots [2]. The present experi- ment showed that the ligation of one Br3 and retrograde HRP labeling of the opposite Br3 caused labeling of the somata of DUMR7 neurons and of axonal granular vesicles which accumulated in the axoplasm proximal to the ligature. Thus, the HRP-labeled axons containing these vesicles were identified as originating from DUMR7 neurons and proctolin-like immunoreactive vesicles were localized in HRP-labeled DUMR7 somata and axons by the postembedding protein A-gold method (Figs. 6-8). Our present results reinforce the supposition that DUMR7 neurons are procto- linergic excitatory motoneurons that induce neurogenic contraction of the accessory gland mus- culature by releasing proctolin [3]. The ultrastructure of the proctolin-like im- munoreactive vesicles has been characterized in the ganglia and peripheral nerves of a few insects by the preembedding PAP technique [17, 19, 25, 40] and the postembedding immunogold technique [23]. The immunoreactive terminals in the dorso- caudal neuropil of the cockroach TAG were de- monstrated to contain large dense vesicles with a diameter of 140-150 nm and numerous small clear vesicles [19]. In the blowfly, immunoreactive granular vesicles with a diameter of about 100 nm were found in the immunoreactive somata of the protocerebrum and in the terminal of the pars intercerebralis [25]. The immunoreactive termin- als in the lateral abdominal nerves were reported to contain clear vesicles of about 75 nm in dia- meter and granular vesicles with a diameters of ca. 115nm [40]. The postembedding immunogold technique has demonstrated three types of procto- lin-like immunoreactive vesicles in the cockroach: small electron-dense vesicles having a mean dia- meter of 80nm in the frontal and hypocerebral ganglia and in the musculature of the oviduct and hindgut, large electron-dense vesicles (150 nm) in the frontal ganglion, and large electron-lucent vesicles (150 nm) with flocculent contents in the hypocerebral ganglion [23]. In the DUMR7 neurons of the cricket, the proctolin-like im- munoreactive vesicles in HRP-labeled somata ranged from 50 nm to 200 nm in diameter (mean diameter: 108 nm) (Figs. 6, 7). In the HRP- labeled axons within the ligated Br3, and in the axons innervating the accessory gland muscula- ture, immunoreactive vesicles with a diameter of 100-110 nm were noted (Figs. 4, 8). Ude and Eckert [23] have suggested that the different sizes of vesicles with PLI might be due to the colocaliza- tion of proctolin with various neuroactive subst- ances in the vesicles. The neurons producing these vesicles were not demonstrated in the present study, though ligation of Br3s also revealed the accumulation of non-immunoreactive vesicles (ca. 80 nm in diameter) (Fig. 5). Light microscopic observation showed a weak PLI of the medial neurons in the cricket TAG (Fig. 2). This finding was in accordance with electron microscopic abservation, which showed that the vesicles with PLI were dispersed in the DUMR7 somata (Figs. 6, 7). It seems probable that the immunoreactive vesicles produced by these neurons are rapidly transported to the peripheral organs. This is supported by the observation that — the accumulation of vesicles occurs in a rather short period of time (3 hr or less) after Br3 ligation (Yasuyama, unpublished data). Dorsal unpaired median (DUM) neurons have been described in several insects [e.g., 41-48]. Hoyle [49] first suggested that the DUMETi neuron innervating the extensor tibiae mucsle of the locust is octopaminergic, and the presence of octopamine in DUMETi soma was definitely de- monstrated by Evans and O’Shea [50]. DUM neurons of the firefly TAG have also been shown to contain significant levels of octopamine [51]. It is now established that octopamine is the substance produced and released as a neuromodulator by DUMETi neuron and probably by other DUM neurons as well. Octopamine and other biogenic amine-containing cells in invertebrates stain speci- fically with the dye neutral red [e.g., 52-54] and DUMR7 neurons are stained selectively by this dye [38]. In the cricket accessory gland, octopa- mine was effective in increasing the frequency of myogenically evoked contractions at concentra- tions as low as 10°’ M [3]. Therefore, it is possible that octopamine coexists with proctolin in the Proctolin-Like Immunoreactivity in DUM Neurons 61 DUMR7 neurons. Furthermore, as suggested by Pfluger and Wat- son [48], it is possible that there are the neurons which are morphologically similar to DUM neurons but have a different role in the innervation of visceral muscle. In the female locust, Lange and Orchard [46] reported that retrograde labeling of the branches of the oviduct nerve revealed a total of eight neurons (three pairs of bilaterally symmet- rical neurons and two DUM neurons [DUMOV neurons]) within the seventh abdominal ganglion. Subsequently, it was shown that DUMOV neurons contain octopamine by using a radioenzyme assay [55], and that bilaterally paired neurons are prob- ably proctolinergic by immunocytochemistry [28]. In addition, the presence of another anterior clus- ter of DUM neurons within the seventh ganglion of the female locust was also revealed by retrog- rade labeling of the oviduct nerve, [31, 48, 56]. Kiss et al. [56] reported two types of axon termin- als containing granular vesicles in the locust ovi- duct musculature: one of them makes conventional neuromuscular junctions but the other type con- taining larger vesicles does not. The DUMETi neuron is thought to release transmitter adjacent to the muscle fibers without forming an anatomi- cally specific neuromuscular junction [57]. On the basis of these facts, Pflliger and Watson [48] sug- gested that the anterior cluster of DUM neurons within the seventh abdominal ganglion of the female locust may be motoneurons having a diffe- rent function from the posterior octopaminergic DUMOV neurons. This possibility is also sug- gested by the detection of a number of DUM neurons following retrograde labeling of the genit- al nerve running from the male locust TAG. In the case of cricket DUMR7 neurons, the somata con- tain large electron-dense vesicles which are iden- tical with those found in the axon terminals making neuromuscular junctions with the accessory gland musculature and with those in their neurites [2]. Our previous physiological study suggested that the DUMR7 neurons are probably motoneurons mediating the squeezing of secretion from the tubules of the accessory gland [3]. Slow coxal depressor of Ds motoneuron is well known to be proctolin-containing motoneuron [11, 58]. This neuron lies bilaterally in the third thoracic ganglion of the cockroach and innervates the muscles of a proximal segment of the leg. O’Shea and Bishop [11] used a combination of intracellular dye injection and immunocytochemis- try to demonstrate that the Ds motoneuron had PLI, while the DUM neurons in the third thoracic ganglion had no PLI. In female locusts, no PLI has been located in the dorsomedial region within the seventh abdominal ganglion [28]. On the other hand, a number of studies on the mapping of PLI in the insect nervous system have shown proctolin-like immunoreactive neurons along the midline of the ganglia [e.g., 17, 18, 20, 25, 40, 59-64]. Most of these neurons appear to be or are proposed to be paired neurons, and only a few unpaired median neurons with PLI have been reported. In the cockroach TAG, one of the four ventral unpaired median (VUM) neurons inner- vating the oviduct muscle and the sphincter muscle was characterized as having PLI [64]. In lepidop- tera, Davis et al. [62] have rerpoted the dorsome- dial unpaired neuron with PLI (K neuron). This neuron occurs in all of the abdominal ganglia and appears to project bilaterally to the perivisceral organs. The K neuron was proposed to have a nuerohumor function like the midline bilateral (MB) cells in the abdominal ganglia of Manduca sexta, which have bilateral projections to the peri- visceral organs and produce cardioacceleratory peptides [65]. In addition, neurons with PLI have been found in the caudal region of the TAG or the fused abdominal ganglion in several insects (e.g., the grasshopper [20]; the cockroach [17, 18]; the Col- orado potato beetle [63]; and diptera [40, 61]. Some of these neurons were proposed to innervate the hindgut [e.g., 17, 40, 61]. In grasshoppers, the anteromedial (AM) cells with PLI in the terminal (fifth) abdominal ganglion have been reported to project to the hindgut [66]. DUMR7 neurons in male crickets form three clusters in the posterior half of the TAG, and our present study showed that the most caudal cluster of DUMR7 neurons innervating the male reproductive organs had PLI. Further experiments are needed to explore the coexistence of proctolin and octopamine in DUMR7 neurons and to explore the functional role of individual DUMR7 neurons in mediating 62 the contraction of the accessory gland according to a temporal motor program. ACKNOWLEDGMENTS This work was suported in part by a Grant-in-Aid for scientific research from Ministry of Education, Science and Culture of Japan. 10 11 12 13 REFERENCES Loher, W. (1974) Circadian control of sperma- tophore formation in the cricket Teleogryllus com- modus Walker, J. Insect Physiol., 20: 1155-1172. Yasuyama, K., Kimura, T. and Yamaguchi, T (1988) Musculature and innervation of the internal reproductive organs in the male cricket, with special reference to the projection of unpaired median neurons of the terminal abdominal ganglion. Zool. Sci., 5: 767-780. Kimura, T., Yasuyama, K. and Yamaguchi, T. (1989) Proctolinergic innervation of the accessory gland in male cricket (Gryllus bimaculatus): detec- tion of proctolin and some pharmacological prop- erties of myogenically and neurogenically evoked contractions. J. Insect Physiol., 35: 251-264. Brown, B. E. (1975) Proctolin: a peptide transmit- ter candidate in insects. Life Sci., 17: 1241-1252. Brown, B. E. and Starratt, A. N. (1975) Isolation of proctolin, a myotropic peptide, from Periplaneta americana. J. Insect Physiol., 21: 1879-1881. Starratt, A. N. and Brown, B. E. (1975) Structure of pentapeptide proctolin, a proposed neurotrans- mitter in insects. Life Sci., 17: 1253-1256. Orchard, I., Belanger, J. H. and Lange, A. B. (1989) Proctolin: a review with emphasis of insects. J. Neurobiol., 20: 470-496. Piek, T. and Mantel, P. (1977) Myogenic contrac- tions in locust muscle induced by proctolin and by wasp, Philanthus triangulum, venom. J. Insect Phy- siol., 23: 321-325. Cook, B. J. and Meola, S. (1978) The oviduct musculature of the horesefly, Tabanus sulcifrons, and its response to 5-hydroxytryptamine and procto- lin. Physiol. Entomol., 3: 273-280. O’Shea, M. and Adams, M. E. (1981) Pentapeptide (proctolin) associated with an identified neuron. Science, 213: 567-569. O’Shea, M. and Bishop, C. A. (1982) Neuropeptide proctolin associated with an identified skeletal motoneuron. J. Neurosci., 2: 1242-1251. Adams, M. E. and O’Shea, M. (1983) Peptide contrasmitter at a neuromuscular junction. Science, 221: 286-288. Lange, A. B. and Orchard, I. (1984) Some pharma- 14 5) 16 7! 18 19 20 7 22 23 24 a) 26 K. YASUYAMA, T. KIMURA AND T. YAMAGUCHI cological properties of neuromuscular transmission in the oviduct of the locust, Locusta migratoria. Arch. Insect Biochem. Physiol., 1: 231-241. Cook, B. J. and Holman, G. M. (1985) The role of proctolin and glutamate in the _ excitation- contraction coupling of insect visceral muscle. Comp. Biochem. Physiol., 80C: 65-73. Stoya, G. and Penzlin, H. (1988) The effect of different putative transmitters on spontaneous pha- sic contractions of cockroach oviduct. Zool. Jb. Physiol., 92: 301-311. Lange, A. B. (1990) The presence of proctolin in the reproductive system of Rhodnius prolixus. J. Insect Physiol., 36: 345-351. Eckert, M., Agricola, H. and Penzlin, H. (1981) Immunocytochemical identification of proctolinlike immunoreactivity in the terminal ganglion and hind- gut of cockroach Periplaneta americana (L.). Cell Tissue Res., 217: 633-645. Bishop, C. A. and O’Shea, M. (1982) Neuropeptide proctolin (H-Arg-Tyr-Leu-Pro-Thr-OH): Im- munocytochemical mapping of neurons in the cen- tral nervous system of the cockroach. J. Comp. Neurol., 207: 223-238. Agricola, H., Eckert, M., Ude, J., Birkenbeil, H. and Penzlin, H. (1985) The distribution of a procto- lin-like immunoreactive material in the terminal ganglion of the cockroach, Periplaneta americana. Cell Tissue Res., 239: 203-209. : Keshishian, H. and O’Shea, M. (1985) The distribu- tion of a peptide neurotrasmitter in the post- embryonic grasshopper central nervous system. J. Neurosci., 5: 992-1004. Marder, E., Hooper, S. L. and Siwicki, K. K. (1986) Modulatory action and distribution of the neuropep- tide proctolin in the crustacean stomatogastric ner- ous system. J. Comp. Neurol., 243: 454-467. Siwicki, K. K. and Bishop, C. A. (1986) Mapping of proctolinlike immunoreactivity in the nervous sys- tems of lobster and crayfish. J. Comp. Neurol., 243: 435-453. Ude, J. and Eckert, M. (1988) Submicroscopic characterization of proctolin-like immunoreactivity in the nervous system of the cockroach, Periplaneta americana L. Cell Tissue Res., 254: 197-202. Schwarz, IT. L., Lee, G. M -E; Siwickin Ke K-. Standaert, D. G. and Kravitz, E. A. (1984) Procto- lin in the lobster: the distribution, release, and chemical characterization of a likely neurohormone. J. Neurosci., 4: 1300-1311. Nassel, D. R. and O’Shea, M. (1987) Proctolin-like immunoreactive neurons in the blowfly central ner- vous system. J. Comp. Neurol., 265: 437-454. Cook, B. J. (1981) The action of proctolin and 5-hydroxytryptamine on the oviduct of the horesefly, Tabanus proximis. Int. J. Invert. Re- Di 28 29 30 31 32 33 34 35 36 37 38 39 Proctolin-Like Immunoreactivity in DUM Neurons 63 - prod., 3: 209-212. Holman, G. M. and Cook, B. J. (1985) Proctolin, its presence in and action on the oviduct of an insect. Comp. Biochem. Physiol., 80C: 61-64. Lange, A. B., Orchard, I. and Adams, M. E. (1986) Peptidergic innervation of insect reproductive tis- sue: the association of proctolin with oviduct visceral musculature. J. Comp. Neurol., 254: 279-286. Orchard, I. and Lange, A. B. (1987) The release of octopamine and proctolin from an insect visceral muscle: effects of high-potassium saline and neural stimulation. Brain Res., 413: 251-258. Orchard, I. and Lange, A. B. (1987) Cockroach oviducts: the presence and release of octopamine and proctolin. J. Insect Physiol., 33: 265-268. Eckert, M., Varanka, I. and Benedeczky, I. (1989) Proctolinergic innervation of the locust oviduct. Zool. Jb. Physiol., 93: 471-479. Bishop, C. A., O’Shea, M. and Miller, R. (1981) Neuropeptide proctolin (H-Arg-Tyr-Leu-Pro-Thr- OH): immunological detection and neuronal loca- lization in insect central nervous system. Proc. Natl. Acad. Sci. USA, 78: 5899-5902. Eckert, M. and Ude, J. (1983) Immunocytochemic- al techniques for the identification of peptidergic neurons. In “Functional neuroanatomy. Springer Series in Experimental Entomology”. Ed. by N. J. I. Strausfeld, Springer, Heidelberg, Berlin, New York, pp. 267-301. Newman, G. R., Jasani, B. and Williams, E. D. (1983) A simple post-embedding system for rapid demonstration of tissue antigens under the electron microscope. Histochem. J., 15: 543-555. Nassel, D. R. (1983) Horseradish peroxidase and other heme protens as neuronal markers. In “Func- tional Neuroanatomy. Springer Series in Ex- perimental Entomology”. Ed. by N. J. I. Satraus- feld, Springer, Heidelberg, Berlin, New York, pp. 44-9]. Roth, J. (1983) The colloid gold marker system for light- and electron microscopic cytochemistry. In “Techniques in Immunocytochemistry”, vol. 2. Ed. by G. R. Bullock and P. Petrusz, Academic Press, London, New York, pp. 217-284. Panov, A. A. (1964) Correlations in the ontogenetic development of the central nervous system in the house cricket Gryllus domesticus L. and the mole cricket Gryllotalpa gryllotalpa L. (Orthoptera, Gryl- loidea). Entomol. Rev., 45: 179-185. Yamaguchi, T., Kushiro, N. and Waki, T. (1985) Sexual dimorphism of the terminal abdominal gang- lion of the cricket. Naturwiss., 72: 153-154. Sobek, L., Eckert, M., Penzlin, H. and Reissmann, S. (1986) Evidence for proctolinergic innervation of the cockroach oviduct. Zool. Jb. Physiol., 90: 461- 466. 40 4] 42 43 44 45 46 47 48 49 50 51 52 33 Nassel, D. R., Holmgqvist, B. I. and Movérus, B. J. A. (1989) Vasopresin- and proctolin-like im- munoreactive efferent neurons in blowfly abdominal ganglia: development and ultrastructure. J. Comp. Neurol., 283: 450-463. Plotnikova, S. I. (1969) Effector neurones with several axons in the ventral nerve cord of the Asian grasshopper Locusta migratoria. J. Evol. Biochem. Physiol., 5: 276-278. Hoyle, G., Dangan, D., Moberly, B. and Colqu- houn, W. (1974) Dorsal unpaired median insect neurons make neurosecretory endings on skeletal muscle. J. Exp. Zool., 187: 159-165. Casaday, G. B. and Camhi, J. M. (1976) Meta- morphosis of flight motor neurones in the moth Manduca sexta. J. Comp. Physiol., 112: 143-158. Davis, N. T. and Alanis, J. (1979) Morphological and electrophysiological characteristics of a dorsal unpaired median neuron of the cricket, Acheta domesticus. Comp. Biochem. Physiol., 62A: 777- 788. Christensen, T. A. and Carlson, A. D. (1981) Symmetrically organized dorsal unpaired median (DUM) neurones and flash control in the male firefly, Photuris versicolor. J. Exp. Biol., 93: 133- 147. Lange, A. B. and Orchard, I. (1984) Dorsal un- paired median neurons, and ventral bilaterally paired neurons, project to a visceral muscle in an insect. J. Neurobiol., 15: 441-453. Watson, A. H. D. (1984) The dorsal unpaired median neurons of the locust metathoracic ganglion: neural structure and diversity, and synapse distribu- tion. J. Neurocytol., 13: 303-327. Pfluger, H.-J. and Watson, A. H. D. (1988) Struc- ture and distribution of dorsal unpaired median (DUM) neurones in the abdominal nerve cord of male and female locusts. J. Comp. Neurol., 268: 329-345. Hoyle, G. (1975) Evidence that insect dorsal un- paired median (DUM) neurons are octopaminergic. J. Exp. Zool., 193: 425-431. Evans, P. D. and O’Shea, M. (1978) The identifica- tion of an octopaminergic neurone and the modula- tion of a myogenic rhythm in the locust. J. Exp. Biol., 73: 235-260. Christensen, T. A., Sherman, T. G., McCaman, R. E. and Carlson, A. D. (1983) Presence of octopa- mine in firefly photomotor neurons. Neuroscience, 9: 183-189. Evans, P. D. and O’Shea, M. (1977) An octopa- minergic neurone modulates neuromuscular trans- mission in the locust. Nature, 270: 257-259. Wallace, B. G., Talamo, B. R., Evans, P. D. and Kravitz, E. A. (1974) Octopamine: selective asso- ciation with specific neurons in the lobster nervous 54 a5 56 i 58 59 60 61 64 system. Brain Res., 74: 349-355. Stuart, A. E., Hudspeth, A. J. and Hall, Z. W. (1974) Vital staining of specific monoamine- containing cells in the leech nervous system. Cell Tissue Res., 153: 55-61. Orchard, I. and Lange, A. B. (1985) Evidence for octopaminergic modulation of an insect visceral muscle. J. Neurobiol., 16: 171-181. Kiss, T., Waranka, I. and Benedeczky, I. (1984) Neuromuscular transmission in the visceral muscle of locust oviduct. Neuroscience, 12: 309-322. Hoyle, G., Colquhoun, W. and Williams, M. (1980) Fine structure of an octopaminergic neuron and its terminals. J. Neurobiol., 11: 103-126. Witten, J. L. and O’Shea, M. (1985) Peptidergic innervation of insect skeletal muscle: immunoche- mical observations. J. Comp. Neurol., 242: 93-101. Lange, A. B., Orchard, I. and Barrett, F. M. (1988) The presence and distribution of proctolin in the blood-feeding bug, Rhodnius prolixus. J. Insect. Physiol., 34: 379-386. Breidbach, O. and Dircksen, H. (1989) Proctolin- immunoreactive neurons persist during metamorph- osis of an insect: a developmental study of the ventral nerve cord of Tenebrio molitor (Coleoptera). Cell Tissue Res., 257: 217-225. Anderson, M. S., Halpern, M. E. and Keshishian, 62 63 64 65 66 K. YASUYAMA, T. KIMURA AND T. YAMAGUCHI H. (1988) Identification of the neuropeptide trans- mitter proctolin in Drosophila larvae: characteriza- tion of muscle fiber-specific neuromuscular endings. J. Neurosci., 8: 242-255. Davis, N. T., Velleman, S. G., Kingan, T. G. and Keshishian, H. (1989) Identification and distribu- tion of a proctolin-like neuropeptide in the nervous system of the Gypsy moth, Lymantria dispar, and in other Lepidoptera. J. Comp. Neurol., 283: 71-85. Veenstra, J. A., Romberg-Privee, H. M. and Schooneveld, H. (1985) A proctolin-like peptide and its immunocytochemical localization in the Col- orado potato beetle, Leptinotarsa decemlineata. Cell Tissue Res., 240: 535-540. Stoya, G., Agricola, H., Eckert, M. and Penzlin, H. (1989) Investigations on the innervation of the oviduct muscle of the cockroach, Periplaneta amer- icana (L.). Zool. Jb. Physiol., 93: 75-86. Tublitz, N. J. and Truman, J. W. (1985) Identifica- tion of neurones containing cardioacceleratory pep- tides (CAPs) in the ventral nerve cord of the tabacco hawkmoth, Manduca sexta. J. Exp. Biol., 116: 395- 410. Keshishian, H. and O’Shea, M. (1985) The acquisi- tion and expression of a peptidergic phenotype in the grasshopper embryo. J. Neurosci., 5: 1005- LOIS: ZOOLOGICAL SCIENCE 9: 65-75 (1992) ULTRASTRUCTURE OF THE MOTILE IRIDOPHORES | OF THE NEON TETRA Hirosut NAGAISHI and Noriko Osuma! Department of Biomolecular Science, Faculty of Science, Toho University, Miyama, Funabashi, Chiba 274, Japan ABSTRACT—As a contribution to efforts aimed at the explanation for the motile mechanism of iridophores in the neon tetra (Paracheirodon innesi), the ultrastructure of the cells was studied by electron microscopy. One or two layers of iridophores are present under the subepidermal collagenous lamella, and melanophores lie beneath these layers. Within each iridophore, two stacks of light- reflecting platelets are arranged symmetrically, overlying the slender nucleus. Each platelet was found to be surrounded by two indenpendent membranes, namely, an inner and an outer membrane. Microtubules were observed in the cortical cytoplasm, and microfilaments were also found around the stacks of platelets. The edges of the platelets in each stack appeared to be joined together by microfilaments. The possible involvement of the organelles in changing the angle of inclination of the © 1992 Zoological Society of Japan reflecting platelets is discussed. INTRODUCTION Accompanying recent progress in studies on light-absorbing chromatophores, our understand- ing of light-reflecting chromatophores has also been enhanced. In particular, unique iridophores, which exhibit cellular motility and are referred to as “motile iridophores”, have lately attracted con- siderable attention. From their studies of the iridophores responsible for the generation of the blue-green color of the lateral stripes of the neon tetra, Lythgoe and Shand [1] and Clothier and Lythgoe [2] reported that the cells are directly affected by light and that the spectrum of light reflected from the cells shifts according to changes in the ambient illumination. Actually, the color of the stripe changes in vivo from green in the daytime to violet-blue at night. The motile activity of the iridophores was also found to be controlled by the sympathetic nervous system [3]. The fluorescence-like coloration produced by the iri- dophores appears to result from the phenomenon of non-ideal thin-film interference [4, 5]. A shift in the peak wavelength of light reflected Accepted October 1, 1991 Received September 2, 1991 ' To whom correspondence should be addressed from the iridophores (i.e., a change in color) is thought to be due to a change in the distance between adjacent platelets [1, 6, 7]. In our pre- vious paper, we proposed the “theory of venetian blinds”: a change in the distance between the reflecting platelets is caused by a change in the angle of inclination of the platelets [6]. However, the mechanism controlling the inclination of the platelets has not yet been elucidated. In case of the blue damselfish, Chrysiptera cyanea, a spherical nucleus is present in the upper part of the iridophore and many stacks of small thin platelets (guanine) are distributed radially from the nucleus. It was suggested that the change in the distance between the platelets occurs simul- taneously in each stack, resulting in a change in the lengths of the stacks themselves [7, 8]. The changes may resemble the expansion and contrac- tion of the bellows of an old-fashioned camera. The tubulin-dynein system may be involved in the motile activities of damselfish iridophores since the motility is inhibited by the treatment with colchi- cine, vinblastine or podophyllotoxin, antimitotic reagents, and by _ erythro-9[3-(2-hydroxyno- nyl)|adenine, an inhibitor of ATPase activity [9]. However, such a possibility has not yet been supported by observations at the electron micros- copic level. 66 H. NAGAISHI AND N. OSHIMA In the present study, the ultrastructure of the unique motile iridophores of the neon tetra was examined mainly under the electron microscope in order to facilitate the understanding of the mechanism responsible for changes in the angle of inclination of the reflecting platelets within the cells. MATERIALS AND METHODS Individuals of both sexes of the neon tetra, Paracheirodon innesi, 21-25 mm in body length, were purchased from a local commercial source and reared in an aquarium for several days before experiments. After decapitation, the body of the fish was sliced in half along the backbone and the skins, including lateral stripes, were cut into small pieces with a clean, sharp razor blade. These pieces were then fixed in a solution of 2.5% glutaraldehyde and 1% paraformaldehyde in 0.1 M phosphate buffer Te tensile CL rs: Fic. 1. Electron micrograph of a vertical section through iridophores (1) of the neon tetra. (pH 7.2) for 2 hr at room temperature. After a rinse with the buffer for 30 min, samples were postfixed with 1% OsQO, in 0.1 M phosphate buffer (pH 7.2) for 1 hr at 4°C. Specimens were dehy- drated through a graded ethanol series, treated with methyl glycidyl ether and embedded in Epoxy resin (Quetol 812; Nisshin EM, Tokyo). Thin sections were cut with a glass or a diamond knife on a Porter-Blum MT-1 ultramicrotome (Ivan Sor- vall, Newtown), mounted on formvar-coated grids, and then stained with 3% uranyl acetate and lead citrate. Some sections were coated with carbon after the staining. Specimens were observed under a Hitachi HU-12A electron micro- scope at 75 kV. One ym sections stained with toluidine blue and the entire skin preparation were observed through a microscope equipped both with epi-illumination and usual transmission optical systems (Optiphoto, XT-BD, with CF-BD plan objective lens; Nikon, Tokyo). ie ese “he Iridophores are situated just under the subepidermal collagenous lamella (CL) in two layers and a network of melanophores (M) is beneath them. A large slender nucleus (N) is present in the lower region of the cell. The reflecting platelets (RP) are stacked regularly at a fixed distance over the nucleus, while most of them were detached during the preparation of sections. MF: Muscle fiber. Mt: Mitochondria. Bar=2pm. Ultrastructure of Motile Iridophore RESULTS General morphology The iridophores of the neon tetra are shaped like a rectangular parallelepiped of about 60-70 ym in length, 20-30 «m in breadth, and 1-2 ~m in height. As shown in Figure 1, electron micro- graphs at low power revealed that layers of iri- dophores are situated just under the subepidermal collagenous lamella, being backed by a network of melanophores, in the stripe skin. Observation of the entire skin preparation under a light-field epi-illumination microscope showed that the iri- dophores were densely distributed like paving stones in the region of the lateral stripes in two layers, though they were in a single layer in some areas. When the iridophores in the upper layer were focused, ones in the lower stratum were also visible, though out of focus. In vertical sections of one um thickness, the two layers of iridophores |m asc po F = en B ML Ficss: —~ Y 00 fan nC ; hy) i} 1 | poe ' AREY Schematic drawing showing the distribution of the motile iridophores in the stripe skin. Fic. 2. Micrograph of one pm vertical section of the stripe skin. Two layers of the iridophores (1) are recognized, but in the right part of the photograph, cells are in a single layer. CL: collagenous lamella. M: Melanophore. Bar=20 um. were also recognized (Fig. 2). The longitudinal axis (long axis) of each iridophore was found to run parallel with the dorso-ventral axis of the fish. The distribution of iridophores in the fish skin is illustrated in Figure 3. The iridophores overlap each other with thinned portions of cytoplasm to form a thin reflective stratum (Fig. 4A). The thinned portions of cyto- plasm in adjoining cells are often tangled as shown p 1 ( A: Lateral view of the fish. Part of the lateral stripe consisting of the iridophores is magnified in the middle frame and the arrangement of the reflecting platelets within the cells is drawn in the right one. Dorso-ventral axis indicated as D—V corresponds to D—V in (A) of this figure. and the arrangement of the reflecting platelets are drawn in the frames. ML: Muscle layer. Epidermis. M: Melanophore. Scale. I: Iridophore. B: Frontal view of the fish. Distribution of the iridophores CL: Collagenous lamella. E: N: Nucleus. RP: Reflecting platelet. S: 68 H. NAGAISHI Fic. 4. Parts of vertical sections through iridophores showing that the thinned portions of cytoplams in adjoining iridophores overlap one another (A) or are tangled (B). Arrowheads indicate the border of each cell. Bar=1 pum. we 4 SE am OO. Fic.5. Electron micrograph showing a desmosome (arrow) by which neighboring iridophores are con- nected. Bar=1 ym. AND N. OSHIMA in Figure 4B. Occasionally, the junctional appa- ratus of membranes can be observed (Fig. 5). In vertical sections of the stripe skin that were cut parallel to the longitudinal axes of the iridophores, a large slender nucleus is found in the lower part of each iridophore (Fig. 1). In cross sections of such cells, the triangle-shaped nucleus is located in the central region of the cell (Fig. 6). Mitochondria are present mainly in the cortical region of the cells, and in the thinned portions of cytoplasm (the edges of the cells), where the reflecting platelets are absent. Few mitochondria are seen in the region between the nucleus and the cell membrane (Figs. 1 and 6). Inside the cell membrane, many pinocytotic vesicles are found here and there (Fig. 7). Golgi apparatus (Fig. 8) and smooth endoplas- mic reticulum (Fig. 9) are sometimes observed. In addition, small electron-dense particles, which may be glycogen granules, are distributed through- out the cytoplasm, although they are scarcely found in the spaces sandwiched between the reflecting platelets. Arrangement of the light-reflecting platelets In vertical sections of the iridophores, many large reflecting platelets are observed over the nucleus (Figs. 1 and 3B). They are stacked reg- ularly at an equal distance from one another, and Fic. 6. Electron micrograph of a cross section of an iridophore. In this photograph, the iridophore is in a single layer. A triangle-shaped nucleus (N) is present in the central region of the cell. A stack of reflecting platelets (RP) is present on each side of the nucleus. CL: Collagenous lamella. M: Melanophore. Mt: Mitochondria. Bar=2 pm. Ultrastructure of Motile Iridophore 69 Sa = : : por STE Lge ernes Cua Fic. 7. A part of a horizontal section through an iri- dophore showing the pinocytotic vesicles. The section was cut through the plane near the cell membrane. Empty spaces in which the reflecting platelet has been present are seen on the left side of the photograph. The electron-dense tubular net- work (arrowheads) is thought to represent the in- vaginations of the cell membrane. Bar=1 yum. Fic. 8. Electron micrograph showing the Golgi appar- atus (arrow). The open space shows the area in which a reflecting platelet has been located. N: Nucleus. ae Sen eee ear e eRe eS aS ate’ 35 Fic. 9. Electron micrograph showing the smooth endo- plasmic reticulum (arrow). N: Nucleus. Bar=0.5 pm. they are inclined at a constant angle. Our light- microscopic observations of fixed skin under light- field epi-illumination revealed that each platelet is hexagonal in shape (guanine crystal; unpublihsed data). The long axis of most platelets is about 15- 20 um in length, and the guanine crystals become smaller at both ends of the stack. The actual reflecting platelets themselves are usually not observed in sections. As shown in Figures 1, 6, 10 etc., the platelets are represented by empty spaces since they become detached in the process of cutting or staining of sections. Moreover, the spaces are bulged by the exposure of sections to the electron beam during the observations. In cross sections of the iridophores, a pair of stacks is positioned above the nucleus or is present on both sides of the nucleus (Fig. 6). Occasionally, three stacks are seen in a cell (photographs not shown). In horizontal sections of the iridophores, each stack lies parallel to the longitudinal axis of the cell. Each platelet is inclined against the axis and the incline of the platelets in neighboring stacks is symmetrical (Figs. 10 and 3A). On the basis of these observations, three-dimensional construc- tion of iridophores is shown in Figure 11. Vesicles that include a fine guanine crystal are often observed in the proximity of both ends of the stack of the platelets (Fig. 12). Inner and outer membranes surrounding the platelet For the preservation of the reflecting platelets themselves, some sections were coated with car- bon after staining. Each platelet was found to be surrounded by two independent membranes; namely, an inner and an outer membrane, as shown in Figure 13. When the platelet is preserved completely, the inner membrane adheres closely to the guanine platelet and the outer membrane, giving the appearance of the fused membrane (Fig. 14). Therefore, the inner membrane is often barely visible. When the platelet is partly broken or detached, the inner membrane can be recognized as a very thin line (Fig. 13). A relatively electron- lucent space is present between these two mem- branes. Observations at high magnification suggest that 70 H. NAGAISHI AND N. OsHIMA Fic. 10. Horizontal section through an iridophore showing a pair of stacks of reflecting platelets (RP) parallel to the long axis of the cell. The arrangement of the platelets in both stacks is symmetrical. Mt: Mitochondria. N: Nucleus. Bar=2 yum. Reflecting Platelet Nucleus ; ee Dorsal Side Fic. 11. Illustration showing three-dimensional construction of the neon tetra iridophore. A slender nucleus is present in the lower part of the cell along the cell axis. Many reflecting platelets in hexagonal shape arrange regularly inclining at a constant angle. Long axis of each platelet is also out of perpendicular to the long axis of the cell. A pair of stacks of the platelets is positioned above the nucleus. As for the arrangement of the platelets within the cell, refer to the schematic drawings in Figure 3. the guanine platelets are not more than 5 nm thick (Fig. 14). However, guanine crystals partially re- maining in empty sacs are apparently very thick As shown in Figures 15 and 16, microtubules, (cf. Figs. 10 and 13), probably as a result of the — which run parallel or perpendicular to the longitu- deformation of the crystals. dinal axis of the iridophore, are observed in the Microtubules and microfilaments Ultrastructure of Motile Iridophore Tal ee at 7 Y é Me ARS zs ee s Oil a Pak A 2 Ee % +r A 3 i 3 Ear = = en x S LE 4 cs fy Pe “s < eg : < ‘ a, = ° Z < < ‘ = aus ca see) B = Fic. 12. Electron micrograph showing the vesicles that include a fine guanine crystal (arrowheads). Com- pare the size of the crystal with that of the reflecting platelets seen in Figures1, 4 and 8. N: Nucleus. Bar=1 um. Fic. 13. Electron micrograph showing the inner (arrow- heads) and outer (arrows) membranes that surround each reflecting platelet. Although sections were coated with carbon after electron staining in this series, the platelets themselves were partly or com- pletely broken. At the both ends of sacs, thick electron-dense lines are visible, since pieces of broken platelets adhered to the inner membrane. Bar=0.2 um. cortical cytoplasm of the iridophore. Compara- tively large numbers of microtubules are distri- buted in the lower part of the cells, and also on the edges of the cells, whereas there are only a few in the region sandwiched between the nucleus and the stacks of reflecting platelets (Fig. 15). No microtubules are seen in the spaces between the platelets. | Microfilaments of about 8 nm in diameter are found around the stacks of platelets. The edges of the platelets appear to be linked together by the microfilaments (Fig. 17). In cross sections, very long microfilaments are frequently found, as shown in Figure 18, whereas they are seldom seen showing a perfectly preserved reflecitng platelet (arrow). The platelet appears to be about 5 nm thick. When the platelet is perfectly preserved, no inner membrane can be recognized since this mem- brane adheres closely to the platelet. Each platelet is also surrounded closely by the outer membrane (arrowheads), although the membrane has become partially detached from the platelet. At the edges of the platelets, vesicular structures are visible. This section was coated with carbon after electron staining. Bar=0.2 um. Fic. 15: the microtubules (arrows) that run parallel to the Part of a cross section of an iridophore showing longitudinal axis of the cell. While microtubules are present mainly at the bottom of the iridophore, they are not abundant in the cell. Note the bundle of the microtubules within the melanophore. M: Melanophore. Mt: Mitochondria. N: Nucleus. RP: Reflecting platelets. Bar=0.5 um. in vertical sections. On the other hand, thin filamentous structures, which might connect the outer membrane that surrounds one platelet with an adjacent one, are also observed (Fig. 18). These filaments are about 6nm in diameter. The long 8-nm filaments running in parallel with one another are sometimes linked by fine fibrous struc- tures, which are very similar to those seen between the platelets (Fig. 18). The microfilaments of about 8 nm in diameter are often connected to the cell membrane by the fine filamentous structures, Fic. 16. Part of a horizontal section through an iri- dophore showing the microtubules (arrows) that run perpendicular and parallel to the longitudinal axis of the cell. They are probably present near the cell membrane, since many pinocytotic vesicles are observed around the microtubules. The large arrow indicates the longitudinal axis of the cell. Bar=0.5 um. Zo NS * Sehr Fic. 17. Cross section of an iridophore showing the microfilaments (arrows) that connect the edges of stacked reflecting platelets (RP). Fine filament- like structures (arrowheads) are derived from the microfilaments and some of them are connected to the cell membrane. The platelets themselves are no longer present (see text). Bar=0.2 um. showing microfilaments and fine filament-like struc- tures. Several long microfilaments (arrows) run along the side of the reflecting platelet (RP), and many fine filament-like structures (arrowheads) are seen between the adjacent platelets. The microfila- ments and the outer membrane that surrounds the reflecting platelet seem to be connected by the filamentous structures. The outer membranes are also mutually connected by these structures. Bar= 0.2 um. . NAGAISHI AND N. OSHIMA as shown in Figure 17. The 8-nm filaments are also distributed randomly in the cytoplasm, and some- times they form a lattice (photographs not shown). However, these filaments are not inserted deeply into the spaces between the reflecting platelets. DISCUSSION In the present study, electron microscopy clearly revealed the way in which the iridophores are distributed in the integument of the lateral stripe region of the neon tetra, and the three- dimensional arrangement of the reflecting platelets within the iridophores was clarified. Generally, within the cell, there are two stacks of more than 100 platelets and each platelet makes an acute angle of depression with respect to the median plane of the fish’s body (cf. Fig. 3B). This observa- tion accords well with our previous observation that the highest reflectance is obtained when illu- mination is applied from the dorsal side of the fish. Since the long axis of the reflecting platelets in each stack is not perpendicular to the longitudinal axis of the iridophore (cf. Figs. 10 and 11) and, furthermore, since some stacks are more or less curved, irradiation from the direction of the head or tail should also result in a low level of reflection; a phenomenon that has been observed in practice [5]. Two layers of iridophores were generally found in the stripe skin. The presence of the two layers may have the effect of purifying the interference colors reflected from the lateral stripes and may augment the reflection because of the resultant increase in the number of platelets that participate in the multilayered thin-film interference [4, 5]. Microbutules were found within the iridophores of the neon tetra. They have also been observed in iridophores of the Atlantic salmon, Salmo salar L. [10], and of the freshwater goby, Odontobutis obscura [11]. Although the role of microtubules in the cells of the latter two species has not been discussed in detail, it is assumed that the microtu- bules function as the cytoskeleton. In case of the neon tetra iridophores, the microtubules are not likely to act as the cytoskeleton since only a relatively small number of microtubules is observed and they are mainly found in the lower Ultrastructure of Motile Iridophore 73 part of the cell. Each of a pair of stacks itself may maintain the shape of the neon tetra iridophores rather than common cytoskeletal elements such as microtubules. In the dermal iridophores of the lizard, Anolis carolinensis, in the iridophores in the scales of the goldfish and Holocentrus ascensionis, and in the tapetal iridophores of the cardinal tetra, all of which are physiologically inactive iridophores, both thin (6.5 nm in diameter) and thick (10 nm) filaments are present [12, 13]. Not only thin filaments but also 10-nm filaments are considered to be cytoskeletal elements. By contrast, in phy- siologically active iridophores, such as the dermal iridophores of the leopard frog, Rana pipiens |14] and the iridophores in the scales of the cardinal tetra [13] and the freshwater goby, Odontobutis obscura [11], thin filaments are abundant and thick filaments are rarely found. In the neon tetra iridophores, thick filaments of about 8 nm in dia- meter are frequently observed in addition to thin- ner filamentous structures, although the cells are of the motile type. It seems that the 8-nm filament and the thinner fibrous structures correspond to the thick and the thin filaments present in non- motile iridophores, respectively. However, a small difference in the diameter of the two respective types of filament should be noted. Rohrlich [13] suggested that the 6.5-nm fila- ments present in physiologically active iridophores might serve to hold the crystalline sheets in their parallel array and might also participate in the movement of the platelets. These filaments were supposed to be composed of actin. Thin filament- like structures in the neon tetra iridophores, which are present in the cytoplasm between the guanine platelets, also my act as the spacers required for the parallel arrangement of the platelets. In addi- tion, the presence of such structures might keep the surface of the platelets flat. In the neon tetra iridophores, closer attention should be paid to the 8-nm filaments, which seem to surround the stacks of the platelets, linking their edges together. Thus, via the action of these filaments, the guanine crystals may be piled up evenly. Furthermore, it is proposed that the 8-nm filaments may be involved in changes in the in- clination of the platelets. In order to unequivocal- ly identify the two types of filaments and confirm the spatial distribution of them, immunohistoche- mical analyses are in progress. For researchers interested in the motile activity of pigment cells, an essential problem is the elu- cidation of the way in which the motive force is generated. The role of microtubules [15, 16, 17, 18], microtrabecular systems [19] and actin fila- ments [20] has been suggested. Above all, the tubulin-dynein system seems to be a highly prob- able candidate for the driving force behind pig- ment migration [21, 22]. Pharmacological analyses have revealed that antimitotic reagents, colchicine and vinblastine [23], and erythro-9[3-(2-hydroxy- nonyl)|adenine (EHNA), an ATPase inhibitor, block the translocation of pigment [24]. In case of the motile iridophores of the blue damselfish, Oshima and Fujii [9] also reported that antimitotic reagents and EHNA inhibit the responses of the cells. In the motile iridophores of the neon tetra, we suggest the possible involvement of the 8-nm fila- ments in changing the inclination of the reflecting platelets. If slipping between the 8-nm filaments that link the edges of platelets and the microtu- bules, or the network of filaments with which the cell membrane is backed, occurs on the upper and lower sides (epidermal and dermal sides) in oppo- site directions, the inclination of all platelets can be changed synchronously. If the microtubules participate in this phenomenon, sliding might take place only on the dermal side since the microtu- bules are present mainly in the lower part of the iridophores. In such a case, the upper end of each platelet should be fixed. However, the structural relationship between the microtubules and the filaments has not yet been fully defined. Moreov- er, it is a matter of concern that the number of microtubules might be not sufficient to allow them to discharge their duties as a “foothold”. In the physiological experiments, we have found that colchicine, vinblastine and EHNA block re- versibly the change in stripe color caused by the application of norepinephrine, implying that the motive force responsible for changing the inclina- tion of the platelets is generated by dynein-tubulin interactions. Surprisingly, cytochalasin B also in- hibited the iridophore response (unpublised data, 74 H. NAGAISHI AND N. OSHIMA Oshima and Nagaishi). This fact suggests that the motile mechanism of the neon tetra iridophores may differ from that of Chrysiptera cells, whose response to norepinephrine is never inhibited by cytochalasin B [9]. The possible role of the actin filaments in the motile activity of the neon tetra iridophores also should be considered. Therefore, we must be in a hurry for the identification of 6-nm and 8-nm filaments seen in these cells. In the present study, we found that the guanine platelets are enveloped in double bound mem- branes, which resemble the fused membrane. This finding is in accord with the results from fish and amphibian iridophores reported by Kamishima [25, 26]. However, Matsuno and Iga [11] reported that each reflecting platelet contained in the iri- dophores of the freshwater goby (Odontobutis obscura) is surrounded by a limiting membrane, and that, at its lateral margins, the platelet is supported by two membranes.. A similar observa- tion was reported by Rohrlich and Porter [12] in a study of the iridophores of Anolis carolinensis. Further comparative observations at the ultra- structural level are clearly required. ACKNOWLEDGMENTS We thank Prof. R. Fujii of our Department for his interest and encouragement. The authors are also grate- ful to Dr. K. Miyaji of Toho University for his kind technical advice. This work was supported by a Grant-in- Aid from the Ministry of Education, Science and Culture of Japan to N.O. (No. 93540590), and by the Science Research Promotion Fund from the Japan Private School Promotion Foundation. REFERENCES 1 Lythgoe, J. N. and Shand, J. (1982) Changes in spectral reflexions from the iridophores of the neon tetra. J. Physiol., 325: 23-34. 2 Clothier, J. and Lythgoe, J. N. (1987) Light- induced colour changes by the iridophores of the neon tetra, Paracheirodon innesi. J. Cell Sci., 88: 663-668. 3 Nagaishi, H. and Oshima, N. (1989) Neural control of motile activity of light-sensitive iridophores in the neon tetra. Pigment Cell Res., 2: 485-492. 4 Huxley, A. F. (1968) A theoretical treatment of the reflexion of light by multi-layer structures. J. Exp. Biol., 48: 227-245. 5 10 11 2 13 14 15 16 17 18 19 Land, M. F. (1972) The physics and biology of animal reflectors. Prog. Biophys. Molec. Biol., 24: 75-106. Nagaishi, H., Oshima, N. and Fujii, R. (1990) Light-reflecting properties of the iridophores of the neon tetra, Paracheirodon innesi. Comp. Biochem. Physiol., 95A: 337-341. Kasukawa, H., Oshima, N. and Fujii, R. (1987) Mechanism of light reflection in blue damselfish motile iridophore. Zool. Sci., 4: 243-257. 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. T. Bagnara, S. N. Klaus, E. Paul and M. Schartl. Univ. Tokyo Press, Tokyo, pp. 241-246. Oshima, N. and Fujii, R. (1987) Motile mechanism of blue damselfish (Chrysiptera cyanea) iridophores. Cell. Motil. Cytoskel., 8: 85-90. Harris, J. E. and Hunt, S. (1973) The fine structure of iridophores in the skin of the Atlantic salmon (Salmo salar L.). Tissue and Cell, 5: 479-488. Matsuno, A. and Iga, T. (1989) Ultrastructural observations of motile iridophores from the freshwa- ter goby, Odontobutis obscura. Pigment Cell Res., 2: 431-438. Rohrlich, S. T- and Porter," K- "R2"@972) Fine - structural observations relating to the production of color by the iridophores of a lizard Anolis caro- linesis. J. Cell Biol., 53: 38-52. Rohrlich, S. T. (1974) Fine structural demonstra- tion of ordered arrays of cytoplasmic filaments in vertebrate iridophores: A comparative survey. J. Cell Biol., 62: 295-304. Taylor, J. D. and Bagnara, J. T. (1972) Dermal chromatophores. Am. Zool., 12: 43-62. Bikle, D., Tilney, L. G. and Porter, K. R. (1966) Microtubules and pigment migration in the mela- nophores of Fundulus heteroclitus L. Protoplasma, 61: 322-345. Murphy, D. B. and Tilney, L. G. (1974) The role of microtubules in the movement of pigment granules in teleost melanophores. J. Cell Biol., 61: 757-779. Schliwa, M. and Euteneuer, U. (1978) Quantitative analysis of the microtubule system in isolated fish melanophores. J. Supramol. Struct., 8: 177-190. Obika, M. and Negishi, S. (1985) Effect of hexylene glycol and nocodazole on microtubules and melano- some translocation in melanophores of the medaka Oryzias latipes. J. Exp. Zool., 235: 55-63. Byers, H. R. and Porter, K. R. (1977) Transforma- tions in the structure of the cytoplasmic ground substance in erythrophores during pigment aggrega- tion and dispersion. I. A study using whole-cell preparations in stereo high voltage electron micros- 20 21 22 28 Ultrastructure of Motile Iridophore TS copy. J. Cell Biol., 75: 541-558. Akiyama, T. and Matsumoto, J. (1983) The block- ade of pigment displacement in_ swordtail erythrophores by microinjection of antiactin anti- body. J. Exp. Zool., 227: 405-411. Nakamura, N., Ikeda, Y. and Obika, M. (1987) Video and electron microscopic studies on pigment transport in Gambusia melanophores. Jpn. J. Ichthyol., 34: 351-360. Oshima, N., Inagaki, H. and Manabe, T. (1990) Evidence for involvement of dynein-tubulin system in pigment aggregation within tilapia melanophores. Comp. Biochem. Physiol., 96A: 517-523. Obika, M., Turner, W. A., Jr., Negishi, S., Menter, 24 DS 26 D. G., Tchen, T. T. and Taylor, J. D. (1978) The effects of lumicolchicine, colchicine and vinblastine on pigment migration in fish chromatophores. J. Exp. Zool., 205: 95-109. Beckerle, M. C. and Porter, K. R. (1982) Inhibitors of dynein activity block intracellular transport in erythrophores. Nature, 295: 701-703. Kamishima, Y. (1979) Electron microscopic study on two types of reflecting cells in the ventral skin of the sand eel, Ammodytes personatus. Proc. Jpn. Acad., 55B: 141-146. Kamishima, Y. (1990) Organization of reflecting platelets in vertebrate iridophores. Zool. Sci., 7: 1020. el ae hari aie See a oF 7 t 2. 4 = F , i =i Ny = 3 1 = o / t ZOOLOGICAL SCIENCE 9: 77-88 (1992) © 1992 Zoological Society of Japan Origin of Multinucleality of Muscle Cells during Myogenesis of the Newt IWANE SATO* Department of Biology, College of general Education, Osaka University, Toyonaka 560, Japan ABSTRACT—The process of multinucleality formation in muscle cells was traced in a Japanese newt. Mitotic figures of muscle cell nuclei are fairly abundant during the embryonal and larval stages in general. Cytokinesis occurs during the myoblast stages, but after a cell has become slender, and is fixed between myosepta, cytokinesis does not occur. No evidence of cell fusion between muscle cells was observed. Macerated specimens of muscle tissue were useful for the counting of the total muscle cell numbers in one myotome. The results show that the muscle cell number in one myotome does not change remarkably during the embryonal and larval stages. The invation or fusion of other kinds of cells was not observed. Hence the multinucleality of muscle cells is the result of mitosis without cytokinesis of one myoblast, and not the product of cell fusion. INTRODUCTION There are two opposing opinions as to the origin of multinucleality in muscle cells of vertebrates. One is “by cell fusion” and the other is “by mitosis without cytokinesis”. In the first case the result would be a syncytium, and in the second, a plas- modium. There have been neither decisive mic- rophotographs nor illustrations for either theory and the controversy continues. The purpose of this paper is to critique this discrepancy by the most effective material from the standopoints of embryology, histology, cytology and karyology. MATERIALS AND METHODS Myotome muscle tissues of the developing embryos and larvae of a Japanese newt, Cynops pyrrhogaster pyrrhogaster Boie, were used. The larvae of each developmental stage were fixed as a whole in Bouin’s fluid. Serial sections in 10 ~m thickness were made. Horizontal section series were most useful. Longitudinal and transverse section series of each stage were also made respec- Accepted October 19, 1991 Received August 7, 1991 * Retired in March, 1973. Author’s present address: Kamibettocho 36, Kitashir- akawa, Sakyoku, Kyoto 606, Japan tively as the occasion demanded. Double staining in Delafield’s haematoxylin and eosin was applied universally. To count the total muscle cell number in one myotome, the maceration technique was used. As larval tissue is soft and fragile, trypsin medium was unnecessary. The unilateral single certain myotome was dissected from a living larva under a dissection microscope in Holtfreter’s solution. Myotome tissue was then immersed in Ca**-free Holtfreter’s solution for about 10 minutes, and the softened unilateral myotome was transferred as a whole onto a slide with one drop of the same medium. The weight of a cover slip was just sufficient to isolate a myotome muscle into free cell groups. After counting the total cell number, the slide was immersed in 10% -formalin solution, stained and mounted for preservation. The larvae were developmentally staged according to the table by Y. K. Okada (1947) [1]. OBSERVATIONS Myogenesis “in vivo” starts as somitogenesis in the trunk region initially. When somites are formed, the cells are randomly arranged, but soon after, they differentiate into three groups. The superficial one is dermatome. The deeper one is myotome whose cells become arranged rather 78 I. Sato Fic. 1. a: somite, b: myocoel c: notochord. radially around the myocoel (Fig. 1). The third group, sclerotome, occupies a small area and is not conspicuous in later stages in this species (Fig. 1). The myoblasts are full of yolk granules with a small amount of melanin granules. They continue to divide mitotically with each following cytokinesis. The chromosome number of this species is 24 in diploid (Fig. 2). During the late neurula stage, myoblasts be- come slender and are arranged longitudinally. Such myoblasts are fixed between anterior and posterior myosepta which are the products of screlotome cells. The fixed end of a cell forms a small flat disc. In the early tail bud stage (stage 30), all the cells are mononucleated and full of yolk granules (Fig.3). Some nuclei also show the condensed chromosomes preparing for the coming mitotic activity. There is a problem in ascertaining the initial total number of myoblasts precisely, as it is dif- ficult to identify the myoblasts from other kinds of Longitudinal section through the notochord including bilateral somites of the tail bud stage. Sta2655, x60), elements. In a typical example of cell count in an embryo in stage 29, the total number of myoblasts in the fourth unilateral myotome is about 66. Among eight inidividuals examined, the number of myoblasts was not more than 75, and not less than 58. After initial cell organization in a myotome is completed, myoblasts start to divide mitotically. During mitotic activity the cell body becomes short and round. In most cases the rounding occurs on the inner surface of each posterior myoseptum. In this case both ends of the mitotic cell body are still fixed on myosepta, one end becoming slender and string-like (Fig. 4B). Sometimes such a figure might be mistaken as a mitotically dividing cell about to invade muscle tissue. Close attention must be given to the long, slender portion of the cells during mitotic activity. After mitosis cells return to the initial slender shape between the two myosepta. Cytokinesis does not follow and the cells remain binucleated retaining the myotube Multinucleality of Muscle Cells 79 ie: % Fic. 2. Mitosis of a myoblast. St. 26. Fic. 3. Mid-sagittal section of head and trunk myotomes c: myotome, d: yolk. structure. In Figure 5A, all the muscle cells are binucleated after the first mitosis. The number of yolk granules are decreased. Formation of vacuoles in the cytoplasm seems to be related to the dissolving yolk granules. Tetanucleated cells x 1,000. in early tail-bud stage. St. 28. a: brain, b: notochord, can also be seen. In this figure the inner chromatin pattern of two adjacent nuclei is quite similar. This shows that they are sister nuclei after one mitotic division. These nuclei are slightly rectangular in this specimen, which might be artifactual due 80 ca Fic. 4. The first mitosis of muscle cells. A: prophase, B: metaphase. SE 3k x500! Fic. 5. A: Binucleated stage after first mitosis. B: Pairs of sister nuclei after first mitosis. St. 32.) Aix<250) By x 600. Multinucleality of Muscle Cells 81 eh Fic. 6. primarily to heavy shrinkage along the long axis of the muscle cell during chemical fixation. In Stage 32 (Fig. 6), isolated muscle cells from the fourth myotome evidently show the presence of en- velopes (sarcolemma) around each cell. At around this stage (Stage 33) some cells are binucleated and others are tetranucleated (Fig. 7), yolk granules are diminished and large vacuoles within cyto- plasm have become more conspicuous. The de- velopment of myofibrillae starts in the early tail bud stage within the peripheral cytoplasm of a myotube. This phenomenon starts initially in the anterior trunk region, and propagates backward gradually. Large spinal ganglia are also completed corresponding and contacting with each side of respective myotomes (Fig. 8A). As a result of motor unit completion, an embryo can react against external stimuli, shaking head and trunk laterally in it’s egg capsule at this stage. Mitotic division is still going on, within the myotube structure. Figure 9 shows a cell with eight nuclei, A is in the tissue and B is an isolated one. All Isolated cell group fixed and stained after counting. St. 32. 400. « these nuclei are situated in the central cytoplasm, still retaining the myotube organization. As the cell is greatly shrunk, the transverse bands become clear under a light microscope. After hatching the yolk mass decrease rapidly, the muscle cells become longer and thicker, and the myotube structure is lost gradually. The nuc- lei, situated in the central cytoplasm in a row, gradually move into the myofibrill bundle and also to the peripheral portion of the cell. Sometimes a nucleus is situated just beneath the cell membrane, surrounded by a small amount of cytoplasm. Mi- totic division still occurs, but simultaneous nuclear division may no longer be expected, as was seen in one pair of sister nuclei in the common central cytoplasm. So such cells with nuclei of odd num- bers in one muscle cell are a common result. Examples of such mitotic division in the peripheral cytopasm are shown in Figure 10A, B and C. After hatching, the tail extends rapidly, but at the tip myotome muscle cells are still mono- nucleated, though myofibrillae are already well 82 I. SATO ee Fic. 7. Tetranucleated stage. St. 33. 400. a: myoseptum, b: vacuole. as ss Fic. 8. A: Longitudinal section of lower trunk region including bilateral myotomes. x4 ganglion. B: prophase. c: metaphase. St. 33. 400. 0. a: spinal cord, b. spinal Multinucleality of Muscle Cells 83 aw Fic. 9. Octanucleated muscle cell. St. 34. contrast microscope. Xx 600. ae: Fic. 10. Mitotic nuclear division in cells with well developed myofibrillae. developed in their cytoplasm. Soon mitotic activ- ity follows, and cells become binucleated. But in the tail region of larvae, muscle cells in the somite are rather shorter in length, and the occurrences of polynucleality are fewer. In Figure 11 one nucleus A: Trunk myotome cell. 400. B: isolated cell under a phase & 5 va PES After hatching. St. 34. is in late prophase with condensed chromosome groups. On the upper left side, one mitotic metaphase plate with mitotic apparatus is extruded together with a small amount of cytoplasm due to the remarkable shrinkage of myofibrillae during 84 I. SATO Fic. 11. The most caudal myotome in tail tip. All cells still mononucleated and with well developed myofibrillae. St. 32. 800. Upper left: metaphase with extruded nuclear plate. Fic. 12. A long muscle fiber in a cranial myotome with 16 nuclei. St. 34. 400. Multinucleality of Muscle Cells 85 Fic. 13. Longitudinal section through right forelimb. S52: chemical fixation. In this larva, it is shown that sixteen or slightly less nuclei are contained in one cell in the cranial myotome (Fig. 12). A significant comment about the myogenesis in appendages should be added (Fig. 13). The arms and legs of this urodelan amphibian are small and slender, so the total cells which constitute each appendage muscle are not abundant. Consequent- ly it is easy to count the total cell number even in the largest forelimb muscle, M. triceps brachii. The total cell number in this muscle is about 72. Both ends of the skeletal muscle cell, attaching to bones, is slender in shape, altogether different from myotome muscle cells whose cell ends form flat discs. But the process of multinucleality formation is quite similar to that seen in somito- genesis. Results of these observations are schema- tized in Figure 14. x 60. a & b: two heads of M. triceps brachii, c: humerus. DISCUSSION There are several papers and textbooks which refer to mitotic division of muscle cell nuclei in embryos and larvae of vertebrates. For instance, Weichert [2] said that “myoblasts begin to differ- entiate and become spindle-shaped and are arranged in bundle”. He also said “Mitotic divi- sion then occur, resulting increase in mass (chick embryo)”. Almost similar descriptions are seen in book by Schneider [3], Schaffer [4], Bucher [5] and so on. But recently, these observations and de- scriptions are mosly forgotten, or are neglected, and the presence of mitotic nuclear division in muscle tissue is brushed aside in the study of myogenesis. The cells of tailed amphibians are extraodianrily large and are convenient materials for the study of myogenesis. The author’s observations on newts are in complete accordance with these previous papers. 86 I. Sato myotube myoblast Fic. 14. Recently there appeared an argument worthy of notice, that is, the mitotic figures in developing muscle tissue are not the division of muscle cell nuclei itself, but are the nuclei of the so-called “satellite cells”. Concerning the nature of these cells, there are some precise and instructive discus- sions by Fawcet and Bloom [6], and Schultz [7]. It is generally known that there might exist some other kinds of dividing mesodermal elements and invading cells in muscle tissue. But in the author’s preparations the mitotically dividing cells show the presence of myofibrillae in every case. Moreover, it is clear that the dividing nuclei belong neither to the invading cells nor to some other kind of mesodermal element. It should be noted that in early developmental stages the cytoplasmic mass containing a mitotic structure sometimes protrudes from the neighboring muscle cells, and that such mass looks as if a cell with mitotic structure is invading the muscle tissue. For instance, the extrusion of the mitotic apparatus together with a = = ee t Hees ED Schema of myogenesis. small amount of cytoplasm (Fig. 11) might be of artifact, presumably due to the remarkable shrink- age and powerful contraction of the muscle tissue during chemical fixation. According to the au- thor’s experience, cells in mitotic activity, in most cases, become stationary and not motile. Even mesodermal wandering cells show a similar tendency. Consequently, invading figures of cells with mitotic structures seem somewhat ambiguous as to their origin “in vivo”. In “in vitro” specimens, on the contrary, the situation is quite different. When myoblasts are cultured “in vitro”, they show almost no mitotic activity after they start myofibrill formation in their cytoplasm. This fact was strongly emphasized by Holtzer [8]. His statement may be right as far as “in vitro” cultured cells are concerned. But in “in vivo” condition, that is to say, in normally de- veloping embryos and larvae, the mitotic division of muscle cell nuclei is common even after myo- fibriogenesis has begun. Multinucleality of Muscle Cells 87 Another serious problem is “cell fusion”. Ele- ments such as mesenchymal cells, fibroblasts and also myoblasts, tend to be highly thigmotactic, and they easily fuse to each other in “in vitro” condi- tion. In “in vivo” condition, such as within living organisms, the cells of mesodermal origin, fibro- blasts and histiocytes for example, envelop foreign substances like surgical strings or plastic tubes for medical use. In such cases, the cells fuse to each other and become multinucleated [9, 10]. The same phenomenon occurs in “in vitro” conditions between cultured cells and the inner surface of a culture bottle or a cover slip. The appearance of “Fremdk6rperriesenzellen” of Lambert [11] is a problem directly related with the fundamental nature of biological membrane. In normal physiological conditions “in vivo”, cell fusion seems to rarely occur. Between the differentiating muscle cells, fusion was never observed. Moreover, the author has never found plausible and reliable figures of cell fusion in published papers on the normal developmental course of myogenesis. The extrusion, sometimes observed, never means cell fusion, but merely means shrinkage in processing. The features of the cell body in question should be totally and careful- ly examined. To ascertain the formation of a multinucleated cell, the total cell number in the initial cell group as well as the multinucleated cell group should be analyzed. If cell fusion occurs among 130 mono- nucleated cells, the resultant binucleated cells might total 65, and the succeeding fusion might produce 33 tetranucleated cells. If the 3rd fusion occurs, 16 octanucleated cells might appear, and so on. This does not represent an actual case. According to MacCallum [12], the total number of muscle cells in the sartorius muscle in human embryos of various ages does not differ greatly from each other. Subsequently, according to Mac- Callum’s data, the increase in size of a muscle simply depends upon the growth of the individual cells. The author’s observations on myogenesis in a myotome are in agreement with this conclusion. In modern biology “in vitro” culture is conve- nient and an excellent method for cytological studies. Many findings on myofibriogenesis were made chiefly by cultured materials and under an electron microscope. On the other hand, the discrepancy between the “in vivo” and “in vitro” states on the behavior of cells as a whole should also be clarified. The author’s experience shows that no remarkable differences among those two can be seen as long as cell organellae are con- cerned, if cells are cultured properly “in vitro” and if tissue cells are physiologically normal “in vivo”. Because cell organellae, such as mitochondria, Golgi apparatus, myofibrillae and so on, are situ- ated in the cytoplasm of their own mother cell in both cases, organellae can exhibit their normal physiological behavior and show their original morphological features in both cases. But as to the behavior of cells as a whole, the condition might not always be the same in both cases. The differ- ence between culture media and normal internal milieu, and some differences in the cell environ- ment may be inevitable. These discrepancies might surely be diminished or removed in the near future by improvement of culture techniques. The simulation of physiological environmental condi- tions of muscle cells to each other during myogenesis will hopefully be achieved before long. ACKNOWLEDGMENTS The author is deeply grateful to those who have been friendly and sympathetic to this study. He also wishes to express his gratitude to his wife, Misako for her en- couragment and helpful preparation of this manuscript. REFERENCES 1 Okada, Y. K. (1947) Developmental Stage of Japanese Newt (Triturus pyrrhogaster). In “Ex- perimental Morphology, Vol. 3”. Kodansha, Tokyo. 2 Weichert, C. K. (1959) Element of Chordate Ana- tomy. McGraw Hill, New York, p. 500. 3 Schneider, K. C. (1902) Lehrbuch der Vergleichen- den Histologie. Gustav Fischer, Jena, s. 813. 4 Schaffer, J. (1920) Vorlesungen tiber Histologie und Histogenese. Wilhelm Engelmann, Leipzig, s. 209. 5 Bucher, O. (1973) Cytologie und mikroskopische Anatomie des Menschem. Stuttgart, Wien, s. 202. 6 Fawcett, D. W. and Bloom, W. (1975) A Textbook of Histology. Saunders, Philadelphia, p. 297. 7 Schultz, E. (1978) Changes in the Satellite Cells of Growing Muscle following Denervation. Anat. Rec., 190: 299-311. 8 Holtzer, H. (1970) Myogenesis. In “Cell Dif- 10 88 I. Sato ferentiation”. Ed. by O. A. Schjeide and J. de Vellis, Van Nostrand Co., New York, pp. 476-503. Kissane, J. M. (1990) Andersons’s Pathology, 9th ed. Vol. 1. The C. V. Mosby Co., St.Louis, p. 979. Martin, M. B. and Epstein, W. L. (1974) Formation of Multinucleate Giant Cells in Organized Epithe- lioid Cell Granulomas. Am. J. Pathol., 74: 263-274. 11 12 Lambert, R. A. (1912) The production of foreign body giant cell in vitro. J. exp. Med., 15: 510-515. MacCallum, J. M. (1895) On the histogenesis of striated muscle fibers and the growth of the human sartorius muscle. Johns Hopkins Hosp. Bull., 1808, cited from G. Clark (1971) The Tissue of the Body, Clarendon Press, Oxford, p. 147. ZOOLOGICAL SCIENCE 9: 89-99 (1992) Effect of Actinomycin D on Nuclear Events during Conjugation in the Ciliate Stylonychia pustulata JUNII YANO and MIkIo SUHAMA Zoolgical Institute, Faculty of Science, Hiroshima University, Kagamiyama, Higashi-hiroshima 724, Japan ABSTRACT—In conjugation of the ciliate Stylonychia pustulata, the relationship between RNA synthesis and nuclear behavior was examined by the use of actinomycin D (AmD) and *H-uridine. At different stages of conjugation, pairs were immersed in 50 ug/ml AmD. When AmD treatment was begun at the early and late stages of meiosis I, conjugation was prevented at 3 defined stages; meiosis I, postmeiotic division and formation of synkaryon. When AmD treatment was begun at meiosis II or at the early period of second postzygotic division, two nuclei produced by first postzygotic division were prevented from dividing at ana-telophase of second division (PZII), but separation of pairs into exconjugant-type cells and degradation of old macronuclei occurred. In many exconjugant-type cells, one of the 2 nuclei which did not divide in PZII swelled like a macronuclear anlage. When AmD treatment was begun at the stage of pronuclear differentiation, nuclear event of conjugation was blocked at the stage of macronuclear development. However, polytene chromosomes were visible in the macronuclear anlage. Autoradiographs of *H-uridine showed that labeled precursors were extensively incorporated into conjugants during meiotic division, but incorporation declined to zero or near the zero level until the stage of pronuclear differentiation. These results suggested that the nuclear events including meiosis, postmeiotic division, pronuclear differentiation and macronuclear develop- ment were largely controlled by RNA synthesized during meiosis. The sequential relationship between © 1992 Zoological Society of Japan RNA and protein syntheses during conjugation is discussed. INTRODUCTION In conjugation of ciliated protozoa, two mature cells of complementary mating type temporarily unite to form a pair. The conjugation includes the following nuclear events: meiosis of micronuclei (germinal nuclei), cross-fertilization, development of new micro- and macronuclei (somatic nuclei) and degradation of old macronuclei [1, 2]. The occurrence of RNA synthesis during the early period of conjugation has been known in Stylony- chia mytilus [3, 4] and Tetrahymena thermophila [5-8]. In the former the long-lived messenger RNA controls the further developmental process of conjugation. In T. thermophila, however, RNA synthesis again occurred in the late stage of con- jugation and in exconjugants [5-7]. In Para- mecium aurelia a net decrease of the cytoplasmic RNA content followed by a drastic increase occur- Accetped October 28, 1991 Received February 7, 1991 red during conjugation [9]. Thus, the initiation and duration of RNA synthesis in conjugation process are different among ciliates. In S. mytilus and 7. thermophila, however, the. relationship between RNA synthesized at the early period of conjugation and its role for nuclear changes in conjugation is unclear. How far does RNA affect nuclear changes which occur sequentially during conjugation? The aim of the present study is to analyze the pattern of RNA synthesis of S. pustulata in the whole process of conjugation and to compare it with the pattern of protein synthesis. Our previous study has shown that pair formation and initiation of meiosis in S. pustulata require RNA and protein syntheses during pair formation [10] and that the progress of furhter nuclear events of conjugation is also controlled by protein synthesized during con- jugation [11]. A detailed comparison of the patterns of RNA and protein synthesis in conjugation revealed two unique findings: synthesis of RNA corresponding 90 J. YANO AND M. SUHAMA to protein synthesis at different stages of meiosis and synthesis of RNA regulating almost all the cytological events after meiosis II. Such a com- parison has not yet been conducted in other cili- ates. The experiment was conducted by treating pairs with an inhibitor of RNA synthesis, acti- nomycin D, or labeling them with a RNA precur- sor, “H-uridine, at different stages of conjugation. MATERIALS AND METHODS Two stocks of complementary mating types, HH2 (mating type II) and TM6 (IV), of Stylony- chia pustulata were used. The stocks were cultured in a wheat infusion medium containing green algae Chlorogonium elongatum as food [12]. Cells from each mating type were separately washed twice with exhausted medium, and them mixed together. Pair formation started about 2 hr after mixing [13]. When pair formation reached near the maximum in the mixtures, A-shaped pairs were transferred within 15 min into wells of depression slides con- taining exhausted medium. This time was adopted as zero time of the formation of conjugation. At intervals of 30 min, about 100 pairs were selected at random from wells of the depression slides and fixed for 10min with acetic-alcohol (1 :3) on cover glass coated with Mayer’s albumin. The specimens were hydrolyzed for 40 min in 4N HCl at room temperature and stained for 30 min in Schiff’s reagent. The length of duration of each nuclear event in conjugation was estimated by observing the stained cells as follows. The dura- tion was defined as the time from the point that 50% of the pairs in a given time possessed the same nuclear stage to the point that 50% of the pairs in a later time had the next nuclear stage. Actinomycin D (AmD) dissolved in culture medium was added to the pairs in depression slides at known times after the onset of pairing. The last addition of AmD was made at the time when the pairs separated into exconjugants. The concentra- tion of AmD was determined as follow. Vegeta- tive cells immediately after cell division were treated at 10, 25, 50 and 100 ug/ml AmD. The replication bands which duplicated DNA in the macronuclei were not formed at 50 and 100 ug/ml AmD. At lower concentrations, the replication bands passed through the macronuclei. Thus, 50 pg/ml AmD was employed in this experiment. About 200 pairs were treated at the final concen- tration of 50 ug/ml. Every hour after the treat- ments, 20 pairs were randomly selected and stained with acetic carmine or Feulgen. The nuclear events were observed till 24 hr after the treatments. Each experiment was repeated three times at 23°C. The pairs treated with 50 ng/ml AmD showed normal morphology for 24 hr. For autoradiography, about 100 pairs were pulse-labeled with 100 ~Ci/ml °H-uridine (New England Nuclear; 26.2 Ci/ml) for 1 hr at intervals of one hour from the onset of pairing. On the other hand, the pairs were pre-treated with 50 yug/ ml AmD for 30 min, and then labeled with 100 uCi/ml *H-uridine for 1 hr in the presence of AmD. At the end of labeling procedure, the pairs were washed twice with the exhausted medium. The old macronuclei were isolated to distinguish the labeled RNA in them from the labeled RNA in the cytoplasm. Some labeled pairs were destroyed by an isolation medium containing 0.05% Triton X-100 and 0.01% spermidine phosphate (Sigma) with distilled water [15]. Pairs and isolated macro- nuclei were fixed in MFG solution (methanol: formalin: acetic acid=17:2:1) for several sec, again fixed in acetic-alcohol (1:3), spread on the subbed slide and then air-dried. Unincorporated label was eliminated by washing the slides with 5% trichloroacetic acid for 10 min at 4°C. In testing the specificity of isotope incorporation, the dupli- cated slides were treated with 0.05% ribonuclease A (RNase; Boehringer Mannheim) in 10mM phosphate buffer (pH 7.0) for 7 hr.at 37°C. Auto- radiographs were prepared with Sakura NR-M2 liquid emulsion. Exposure time was 14 days at 4°C. The labeled nuclei were stained with 2% methyl green through the emulsion after develop- ment. : Morphological changes in each nuclear event during conjugation have been described previously [13]. Vegetative cells of S. pustulata possess two micronuclei and two macronuclei. There are eight major micronuclear events: first and second maturation divisions (MI, MII), postmeiotic divi- sion (PM), pronuclear disivion (PD), formation of synkaryon (SK), first and second postzygotic di- RNA Synthesis during Conjugation 91 vison (PZI, PZII) and macronuclear development (MD). Stage MI will be described in detail later. Stage MII: two spherical nuclei rapidly undergo MII and two disk-like nuclei degenerate later. Stage PM: all four nuclei produced by MII begin to swell after a short time of interphase and then only two nuclei survive and take part in nuclear divi- sion. Four nuclei separate into anterior two and posterior two, since the mitotic spindles are oriented along the longitudinal axis of the con- jugant. The other two nuclei shrink and then degenerate. Resorption of old macronuclei into its cytoplasm begins. Stage PD: though four nuclei produced by PM swell after a short time of interph- ase, one nucleus in the anterior group differenti- ates into a migratory pronucleus and one nucleus in the posterior group into a stationary pronucleus (Fig. 3G). The remianing nuclei shrink and de- generate later. Stage SK: exchange of migratory pronuclei and subsequent fusion of migratory and stationary pronuclei occur. Stages PZI and PZII: two divisions of the synkaryon follow to yield four nuclei. Stage MD: of the four nuclei arranged in a file, the second and fourth sister nuclei from the anterior differentiate into new micronuclei, the third nucleus develops into a macronuclear anlage, and the first nucleus degenerates later (Fig. 3H). Each pair separates into two exconjugants. They possess oral apparatus (OA) with a small buccal cavity and small cytostome without paroral mem- branes (exconjugant-type OA). Polytene chromo- somes are developed in anlage. The macronuclear anlage shrinks and then amitotically divides into parts and the products become new macronuclei. The exconjugant-type OA is reorganized into a functionally normal OA at the time. RESULTS Duration of the nuclear events during conjugation Morphological changes in seven of the eight nuclear events except for MI during conjugation have already been described above. However, the exact time schedule of all the nuclear events has not yet been presented. By the method mentioned above, the micronuclear events among pairs pro- gressed synchronously to a considerable extent as shown in Figure 1. Moreover, the duration of each micronuclear events was estimated from the data l il ll IV VII Mill PM PZII MD 100 e-e-© “0-0 BEER he Ho OAKS 718) x in +-+ v » [" O V + Ww oy | vill © O © S 5 VI 2 A = % © 0) =F v Vv 5 50 & © O Ge : fo) ° oe ® io) e \ ‘ vf VW a ! O—- @-—g¢—_L_ 9 +—- 5 Ne -A- x -O-2© wey ee [KX hint it O 2 4 6 8 10 7 : 14 a 18 20 Time (h) Fic. 1. Progression of micronuclear changes during conjugation of Stylonychia pustulata. MI and MII, first and second meiotic division; I, compact nuclei; II, swollen nuclei; III, nuclei at early stage of parachute; IV, nuclei at mid stage of parachute; V, nuclei at late stage of parachute; VI, prometaphase I; VII, metaphase I; VIII, ana-telophase I; PM, postmeiotic division; PD, pronuclear differentiation; SK, formation of a synkaryon; PZI, first postzygotic division; PZII, second postzygotic division; MD, development of macronuclear anlage. 92 J. YANO AND M. SUHAMA 0 5 10 15 20 ee ee) Ae AmD / + tH _——— eee WHT —_—_—_—______$____—__+> — ———<$_————————————— _ Re a Ae eS Chx ~ es --_ =_— '— . /~ -——e. Fic. 2. Duration of each micronuclear event in conjugation (top) and effects of actinomycin D (AmD) and cycloheximide (Chx) [11] on micronuclear changes (bottom). The left end of arrow shows the onset of drug treatment, and the right end shows the end point of normal nuclear progress in the presence of drug. The length between the left and right ends shows the duration of normal nuclear progress. (Fig. 2). Especially, the micronuclear behavior from the beginning of pairing to the end of MI is newly subdivided into eight stages (stage I-VIII) on the basis of morphological features of micro- nuclei. At stage I (0-1.3 hr), two micronuclei in each conjugant remain compact after the onset of pairing. At stage II (1.3-3.8 hr), micronuclei swell from the usual size of 3 ~m in diameter to 10 um. Thin chromatin fibers first appear (Fig. 3A), fol- lowing chromatin granules in the periphery of nuclei (Fig. 3B). In each conjugant, the anterior macronucleus elongates. At stage III (3.8-6.7 hr), chromatin threads unilaterally emerge from the chromatin mass in the swollen micronuclei (Fig. 3C). The anterior macronucleus breaks into two pieces. At stage IV (6.7-8.8 hr), more chromatin threads are concentrated in a hemispherical shape, and the micronuclei assume a typical parachute- shape (Fig. 3D). At stage V (8.89.8 hr), conden- sation of chromatin threads takes place, but the small chromatin mass is still on the opposite side (Fig. 3E). At stage VI (9.8-10.2 hr), many small chromatin rods or chromosomes are dispersed in the nuclei (Fig. 3F). At stage VII (metaphase I, 10.2-10.8 hr), the chromosomes are arranged on the eugatorial plate. At stage VIII (ana-telophase I, 10.8-11.4hr), one daughter nucleus shows a spherical shape, and the other is a disk-like in shape. The duration of the other stages is measured as follows: stage MII, 11.4-12.4 hr; stage PM, 12.4- 15.3 hr; stage PD, 15.3-16.5 hr; stage SK, 16.5- 17.2 hr; stages PZI and PZII, 17.2-18.6 hr; and stage MD, 18.6-19.1 hr. Resorption of old macro- nuclei into the cytoplasm and separation of pairs occur about 14hr and 19.5 hr after the onset of pairing, respectively. Full appearance of polytene chromosomes and shrinkage of macronuclear anlange are observed about 20 and 60 hr after the separation of pairs, respectively. Effects of AmD on conjugation and nuclear events in conjugation Effect of AmD on micronuclear behavior is summarized in Figure 1. When A-shaped pairs were treated by AmD, they developed into com- plete pairs. The micronuclei proceeded from stage I to stage II, but the swollen nuclei showed unusual elongation in the presence of AmD (Fig. 4A). When pairs 2 and 3 hr after the onset of pairing RNA Synthesis during Conjugation 93 ¥ : <. : > Fic. 3. Nuclear state in different stages of conjugation. micronucleus of stage II 3 hr after pairing. Chromatin granules are located in one side of the nucleus. \H Bar=5 wm. A-F. Meiotic prophase of meiosis I. micronucleus (arrow) of stage II 2hr after pairing. The chromatin fibers appear in the nucleus. A.A B. A C.A micronucleus of stage III (early parachute) 5 hr after pairing. D.A micronucleus of stage IV (mid parachute) 7 hr after pairing. E. A micronucleus of stage V (late parachute) 9 hr after pairing. Many chromatin strands occupy in one side of the nucleus. F. A micronucleus of stage VI (prometaphase I) 10 hr after pairing. G. pronuclear differentiation 16hr after pairing. MP and SP show a migratory and a stationary nucleus, respectively. Arrows indicate degenerating nuclei. H. Differentiation of macronuclear anlage (MA) and new micronuclei (MI) 19 hr after the onset of pairing. Arrow indicates a degenerating nucleus. were treated with AmD, micronuclei proceeded from stage II to stage III, and their chromatin threads aggregated into one irregular shaped mass (Fig. 4B). Macronucleus fragmentation occurred, but resorption of the fragments scarcely took place. When pairs 4 and Shr after the onset of pairing were treated with AmD, micronuclei pro- ceeded from stage III to stage IV. However, the chromatin threads which appeared on one side of the nuclei became arranged in parallel rows about Shr after the treatment (Fig. 4C). Then, these nuclei were elongated and became spindle shaped, but the chromosomes were not arranged on the equatorial plate unlike those of the control pairs 94 J. YANO AND M. SUHAMA % aM _ 97. Fic. 4. Abnormalities of micronuclear events caused by actinomycin D (AmD). Bar=5 ym. A. Elongated micronuclei of a conjugant 5 hr after a pair was treated 15 min from the onset of pairing. B. Chromatin threads aggregating in micronuclei of a conjugant 5 hr after a 2-hr-old pair was treated. C. Chromatin threads arranged in parallel rows in two micronuclei of a conjugant 5 hr after a 4-hr-old pair was treated. D. Two micronuciei showing irregular spindle shapes. The nuclear stage is later than C. E and F. Two nuclei (arrows) blocked at anaphase (E) and telophase (F) of PM in conjugants 5 hr after a 9-hr-old pair was treated. G-H. Nuclei in conjugants treated 12 hr after the onset of pairing. G. Two nuclei (arrows) blocked at PZII remain unchanged in an exconjugant-type cell 10 hr after the onset of treatment. H. One of the 2 nuclei blocked at ana-telophase of PZII swells like macronuclear anlage (arrow) in an exconjugant-type cell 10 hr after the onset of treatment. (Fig. 4D). When AmD treatment was begun at 6 became arranged in parallel rows, but they were and 7 hr after pairing (late stage III or early step of | longer than chromosomes of metaphase I (stage stage IV), micronuclei entered stage V. However, VII) and often adhered each other. Then nuclei 3 hr after the treatment, the chromatin strands assumed a spindle shape like Figure 4D. When RNA Synthesis during Conjugation 95 AmD treatment was begun at 8 hr (stage IV), nuclear division was blocked at ana-telophase I (stage VIII). Two daughter nuclei were connected with the chromatin materials. When AmD treatment was begun at 9 and 10 hr (stage V or VI), MI and MII occurred, and then two nuclei entered into PM. However, the nuclear division was disrupted at ana- or telophase in the presence of AmD. In the former cases, two daughter neuclei stopped at anaphase, and then two daughter nuclei remained adhered (Fig. 4E). In the latter cases, two nuclei were blocked at telophase and remained connected with some chromatin materials (Fig. 4F). When AmD treat- ment was begun at 11 hr (stage VII or VIII), PM and PD occurred normally, but the formation of BIG.35;. For distribution of grains, see text. Arrows indicate macronuclei. FV, food vacuole. macronuclei at stage PZII. Autoradiographs of pairs and isolated macronuclei pulse-labeled for 1 hr with *H-uridi ee a ne at different times. Bars in A and E are 40 um, and in the others 10 um. A. A pair at stage III. B. Isolated old macronuclei at stage MII. isolated from a pair treated with AmD at stage III and labeled for 1 hr. D. Macronuclei E. A pair at stage PM. F. Isolated old 96 J. YANO AND M. SUHAMA synkaryon failed. When AmD treatment was begun at 12, 13 and 14 hr (stage MII or PM), nuclear event of conjuga- tion was blocked at ana-telophase of PZII. Each pair was separated into two cells 19 hr after pairing or slightly later. The cells possessed an excon- jugant-type OA. About 50% of these cells had neither macronuclear anlage nor new micronuc- leus (Fig. 4G), and others had one swollen nucleus like macronuclear anlage derived from one of the two nuclei which failed to divide in PZII (Fig. 4H). In such nuclei, polytene chromosomes were not found. The old macronuclei degraded. When AmD treatment was begun at 15, 16, 17, 18 and 19 hr (late PM stage or before the separation of pair), pairs separated at almost the same time as the non-treated control. Exconjugants possessed a macronuclear anlage and new _ micronuclei. Polytene chromosomes were observed in the mac- ronuclear anlage. However, unlike the non- treated control, the macronuclear anlage did not shrink. Moreover, the reorganization of excon- jugant-type OA into a functionally normal OA was inhibited by AmD treatment. RNA synthesis pattern during conjugation When pairs were pulse-labeled with *H-uridine at stages I-VIII and autoradiographed, silver grains were observed on the macronuclei and cytoplasm (Fig. 5A, B, C). Many of the grains disappeared in conjugating cells and isolated mac- ronuclei when the cells were treated with AmD (Fig. 5D) or when the preparations were treated with RNase. These suggest that the grains indicate specific incorporation of *H-uridine into RNA which is synthesized in macronuclei. RNA synth- esis actively occurred in macronuclei at stages MI and MII (Fig. 5B, C). The number of grains on micronuclei of meiotic prophases was almost simi- lar as those in the background of the cytoplasm. Macronuclear and cytoplasmic labels remark- ably decreased to zero or near the zero at stages PD, PZI, PZII and MD (Fig. 5E, F). When exconjugants with polytene chromosomes were labeled for 4 hr or more, a few grains were observed on the macronuclear anlage but not on the old macronuclei (Fig. 6A). Thus, the anlage may synthesize RNA weakly in this stage. The Fic. 6. Autoradiographs of an exconjugant and a young vegetative cell. Bar=20 um. A. An exconjugant cell 24 hr after the separation of pair and labeled for 4hr. The silver grains of macronuclear anlage (dotted line) are slightly greater in number than those of cytoplasm. No grains are located in the old macronuclei (arrows). B. A young vegetative cell having the normal oral apparatus after the reorga- nization and labeled for 1 hr. Two macronuclei (arrows) have much more silver grains than the macronuclear anlage in Fig. 6A. exconjugants had small cytostome without paroral membranes (exconjugant-type OA) and the incor- poration of precursors into the food vacuoles of exconjugants was at an extremely low level. This may influence RNA synthesis of macronuclear analge. Silver grains were observed on the new macronuclei and cytoplasm in cell with normal OA about 70 hr after separation of a pair (Fig. 6B). Therefore, most of RNA precursors may be in- corporated into exconjugant through its small cytome. DISCUSSION RNA synthesis in the macronucleus Macronuclei in the conjugating cells of Stylony- chia pustulata synthesize RNA from the onset of pairing to stage MII. When pairs were split at 10 min after pairing, the split cells underwent auto- gamy [14]. Ten min after the start of pairing, two partners of a pair loosely unite with their peris- tomes [14] and micronuclei remain compact [13]. Therefore, full conjugation is probably not neces- sary for the maintenance of this RNA synthesis. Duration of RNA synthesis in conjugation of S. RNA Synthesis during Conjugation 97 pustulata is different from the case of S. myztilus. In the latter, synthesis of stable mRNA regulating cytological events during conjugation occurs from 5 hr to 6 hr after the onset of pairing [3, 4]. Two cells of a pair loosely unite during the first 5 hours of their union. If these pairs are split, the split cells return to the vegetative stage. In the following 1 hr, the pairs become firmly attached. If the tight pairs are separated forcibly, the split cells undergo autogamy. In the Shr to 6hr following pair formation, micronuclei are slightly elongated [3]. In both species of Stylonychia, RNA synthesis of old macronuclei was inactive in conjugants after PM and in exconjugants. On the contrary, old macronuclei actively undergo RNA synthesis in Blepharisma musculus [16], Euplotes woodruffi [17], P. aurelia [18] and T. thermophila [19]. New macronuclear anlage with polytene chromosomes in S. pustulata may slightly synthesize RNA as the case of S. mytilus [3, 20]. In the latter, RNA synthesis is supported by the fact that ribonuc- leoprotein particles are present in the chromo- somes and the karyolymph of macronuclear anlagen in exconjugant [21]. However, it has been reported in Oxytricha by Spear and Lauth [23] and in S. mytilus by Ammermann [24] that polytene chromosomes scarcely incorporate RNA precur- sors. These results may be caused by the low incorporation of labeled RNA precursors from exconjugant-type cytostome into the exconjugant. In B. musculus [16], P. aurelia [25] and T. thermo- phila |6, 19], macronuclear anlagen which have no polytene chromosomes actively synthesize RNA from the early stage of development. Relationship between RNA synthesis and protein synthesis at the early period of conjugation In S. pustulata compact micronuclei swelled and the swollen micronuclei entered the middle or late parachute stage in the presence of AmD (Fig. 2). The parachute stage corresponds to the zygotene stage in meiosis of other organisms [2]. As shown in Figure 2, on the other hand, Cycloheximide (Chx) also blocked the progress of meiosis if it was added at the early stage of MI [11], suggesting that proteins which were synthesized during this period were needed for meiosis. Our present data suggest that RNA synthesized from the onset of pairing to the early parachute stage may be responsible for protein synthesis. Similar synthesis of RNA and proteins in the early period of meiosis has also been observed in T. thermophila [5]. In T. thermo- phila the synthesis of conjugation-specific mRNA reaches the maximum level just prior to crescent stage (pachytene), while conjugation-specific pro- teins are actively synthesized during meiotic prophase. In the male meiocytes of Lilium, RNA is synthesized at the meiotic prophase [26, 27]. Moreover, three meiotic-specific proteins partici- pating in crossing-over are activated at zygotene or pachytene [28]. Thus, syntheses of RNA and proteins from zygotene to pachytene may be com- mon phenomena in various organisms including ciliates. Differences of the effects of AmD and Chx at stages III to IV In S. pustulata, there is a difference between the effect of AmD and that of Chx. Micronuclei of the early parachute stage entered the late parachute stage in the presence of AmD, but the micronuclei entered prometaphase I or metaphase I in the presence of Chx. Thus, the progress of meiosis from the early parachute stage (stage III) to prom- etaphase I (stage VI) may required RNA synthesis rather than protein synthesis during this period. Since chromosomes were longer than those of metaphase I and often adhered each other, RNA may be involved in chromosomal behavior such as crossing-over or segregation of chromosomes. However, there is no evidence showing a sufficient deposition of silver grains on micronuclei of stage V or VI. Thus, that protein synthesis is not blocked by the drop of permeability of Chx into the conjugants during stage V or VI cannot be excluded. On the other hand, in the meiocytes of Lilium, the progress of meiosis immediately ceases by blockage of RNA synthesis at pachytene [26]. Small nuclear RNA which is synthesized at zygotene-pachytene regulates nuclease accessiblity in specific regions of meiotic cells [29]. RNA synthesis at the late period of conjugation Some differences were observed between the effects of AmD and Chx at the late period of conjugation (Fig. 2). If the pairs were treated with 98 J. YANO AND M. SUHAMA AmD at various stages during 9-14hr after the onset of pairing, AmD blocked either the nuclear events at ana- or telophase (segretaion of chromo- somes) of PM and PZII or the formation of a synkaryon. If AmD treatment was begun just before each stage, these nuclear events would not be affected. Thus, failure of segregation of chromosomes would not be caused directly by AmD, but by the inhibition of RNA synthesis of AmD. On the other hand Chx stopped nuclear events at nuclear differentiation or prophase of PM, PD and PZII [11]. Therefore, RNA for chromosomal segregation of PM and PZII would be synthesized later than RNA for nuclear dif- ferentiation and prophase of PM, PD and PZII. Except for nuclear division of PZII, nuclear events of PM, PD, SK and PZI may be controlled by RNA synthesized during meiosis I and II. If AmD treatment was begun at 12-14 hr, two nuclei produced by PZI were blocked at ana- or telophase of PZII. One of them was swollen like a macronuclear anlage. This result implies that one of 2 nucleus blocked at PZII can receive a signal for macronuclear development, though polytene chromosomes are not developed. On the other hand, the separation of the pair and the degenera- tion of old macronuclei safely proceeded in the presence of AmD. If pairs were treated with AmD or Chx at 15-19 hr, the development of macronuc- lear anlage was not affected by AmD, while it was stopped by Chx. These observations indicate that RNA necessary for cytological events in the late stage of conjugation is mainly synthesized until MII. Finally, nuclear events and separation of the pair into exconjugants during conjugation of S. pustulata require RNA synthesized sequentially and mainly by macronuclei from the onset of pairing to MII. RNA for meiosis is synthesized during meiotic prophase. The nuclear events in the late period of conjugation require RNA mainly during MI and MII. RNA may correspond to stable mRNA as suggested in S. mytilus by Sapra and Ammermann (3, 4]. Nuclear differentiation of PM, PD and MA, and prophase of PM, PD and PZII were blocked by Chx [11]. Thus, stable mRNA necessary for the progress of each stage may be transcribed just before or during the stage. 10 11 12 13 14 REFERENCES Raikov, I. B. (1972) Nuclear phenomena during conjugation and autogamy in ciliates. In “Research in protozoology, Vol.4”. Ed by T. T. Chen, Academic Press, New York, pp. 147-202. Raikov, I. B. (1982) The protzoan nucleus. Cell Biol. Monographs, Vol. 9. Springer-Verlag, Wien, New York. Sapra, G. R. and Ammermann, D. (1973) RNA synthesis and acquisition of actinomycin D insensi- tivity during conjugation in Stylonychia mytilus. Exp. Cell Res., 78: 168-174. Sapra, G. R. and Ammermann, D. (1974) An analysis of the developmental program in relation to RNA metabolism in ciliate Styloncychia mytilus. Dev. Biol., 36: 105-112. Martindale, D. W., Allis, C. D. and Burns, P. J. (1985) RNA and protein synthesis during meiotic prophase in Tetrahymena thermophila. J. Pro- tozool., 32: 644-649. McDonald, B. B. (1973) Nucleic acid in Tetrahyme- na during vegetative growth and conjugation. In “Biology of Tetrahymena”. Ed. by A. M. Elliot, Dowden, Hachinson and Ross, Inc., Strousdburg, Pa., pp. 287-306. : Ron, A. and Horovitz, O. (1977) Increased RNA synthesis during preconjugation and its effect on pair formation in Tetrahymena. Experimentia, 33: 1146-1149. Sugai, T. and Hiwatashi, K. (1974) Cytologic and autoradiographic study of the micronucleus at meio- tic phase in Tetrahymerna pyriformis. J. Protozool., 21: 542-548. Woodard, J., Woodard, M., Gelber, B. and Swift, H. (1966) Cytochemical studies of conjugation in Paramecium aurelia. Exp. Cell Res., 41: 55-63. Yano, J. (1986) Effects of actinomycin D and cycloheximide on pair formation and conjugant fu- sion in Stylonychia pustulata (Ciliophora). J. Sci. Hiroshima Univ., Ser. B, Div. 1, 32: 255-269. Yano, J. and Suhama, M. (1990) Effects of cyc- loheximide on meiosis and other nuclear changes during conjugation in the ciliate Stylonychia pustula- ta. J. Sci. Hiroshima Univ., Ser B, Div. 1, 34: 19-28. Yano, J. (1985) Characters of progeny clones from conjugant fusion of Stylonychia pustulata (ciliophora). J. Sci. Hiroshima Univ., Ser. B, Div. 1, 32: 209-211. Yano, J. (1985) Mating types and conjugant fusion with macronuclear union in Stylonychia pustulata (Ciliphora). J. Sci. Hiroshima Univ., Ser. B, Div. 1, 32: 157-175. Yano, J. (1985) Degeneration of the cortical organelles and nuclear changes in the split members from the early conjugating pairs in Stylonychia pus- 15 16 17 18 19 20 ay 22 RNA Synthesis during Conjugation 99 tulata (Ciliophora). J. Sci. Hiroshima Univ., Ser. B, Dived 32> 193-207. Prescott, D. M. Rao, M. V. N., Evenson, D. P., Stone, G. E. and Thrasher, J. D. (1966) Isolation of single nuclei and mass preparation of neuclei from several cell types. In “Method in cell physiology”, Vol. 2. Ed. by D. M. Prescott, Academic Press, New York, pp. 131-142. Dass, C. M., Sapra, G. R. and Kumar, R. (1982) Fine structure and nucleic acid synthesis during macronuclear development in Blepharisma muscu- lus var. seshachari. Arch. Protistenk., 126: 293-308. Rao, M. V. N. (1968) Macronucleus development in Euplotes woodruffi following conjugation. Exp. Cell Res., 49: 107-120. Kimball, R. F. and Perdue, S. W. (1964) Synthesis of RNA by fragments of the old macronucleus in Paramecium aurelia undergoing autogamy. J. Pro- tozool., 11 (suppl): 33. Wenkert, B. and Allis, C. D. (1984) Timing of the appearance of macronuclear-specific histone variant hvl and gene expression in developing new macro- nuclei of Tetrahymena thermophila. J. Cell Biol. 98: 2107-2117. Alonso, P. and Jareno, M. (1974) Incorporacion de uridina-H’ en el esbozo macronuclear de Stylony- chia mytilus. Microbiol. Esp., 27: 199-211. Gil, R. (1976) Possible nuclear-cytoplasmic trans- fers in the exconjugants of Stylonychia mytilus: ultrastructual observation. Microbiol. Esp., 29: 125-134. Rao, M. V. N. and Ammermann, D. (1970) 23 24 25 26 Di 28 29 Polytene chromosomes and nucleic acid metabolism during macronuclear development in Euplotes. Chromosoma, 29: 246-254. Spear, B. B. and Lauth, M. R. (1976). Polytene chromosomes of Oxytricha: biochemical and mor- phological changes during macronuclear develop- ment in a ciliated protozoan. Chromosoma, 54: 1- 3? Ammermann, D. (1968) Synthese und Abbau der Nucleinauren wahrend der Entwicklung des Macro- nucleus von Stylonychia mytilus (Protozoa, Ciliata). Chromosoma, 25: 107-20. Berger, J. D. (1973) Nuclear differentiation and nucleic acid synthesis in well-fed exconjugants of Paramecium. Chromosoma, 42: 247-268. Hotta, Y. (1971). Meiosis. In “Systematic cytology, Vol 5 (in Japanese)”. Ed. by K. Ogawa, Y. Oda, K. Kurozumi and S. Sugino, Asakura, Tokyo, pp. 74- NOS: Hotta, Y. and Stern, H. (1963) Syntheses of mes- senger-like ribonucleic acid and protein during meiosis in isolated cells of Trillium erectum. J. Cell Biol., 19: 45-58. Stern, H. (1981) Chromosome organization and DNA metabolism in meiotic cells. In “Chromosome today”, Vol. 7. Ed. by M. D. Bennett, M. Bobrow and G. Hewitt, George Allen and Unwin, London, pp. 94-104. Hotta, Y. and Stern, H. (1983) Small nuclear RNA molecules that regulate nuclease accessibility in spe- cific chromatin regions of meiotic cells. Cell, 27: 309-319. est , etait era pele? es i , : ~ \ - @ al ‘a ie on icaicemabinee aon oes wa) . j : 4 my « = 7 me aman ater Ame oe pelea es Pe bens.» Re Ra oe yh. aiaape + ‘- ee } i 3 pf xe at \ 15 1 i f if . ah, ng | . o . = ee yi : ‘ ee ana GE aioe bie inne Ce aan Yo ieee t- ‘ het oe sedan” { eS} gate see bray Seni: 4 ae eat, pee: in Sie) donne agate Wie oe Bes souls wanelies ahi ‘ ie aS ee e's er irs heap 7 teeny 2 soruge aan | | 2 oe fa ir oe it ‘$0 rcs at: ez oe Me, Ven? pi * ‘ x ae ‘ 0 Rient Hea ; a ee Pe oe Pa * pete ee ; - 4 =A 4 ; - a “sR hi at i ‘ ; “ane 7 has peo ioe ho a aa “We te “ % Sani 1S yh ie ' i¥ baa ae Reese Ms ees ¥ z ~ I, mee * x te . ad A es 3 ‘ i : jhe ‘ - ae ad Pos ; ; = Te, ; ites iy r ‘ Vane ‘ o i 3 ‘ i = f ¢ ane 4 3 eer ¥ . ait x = i r “ % 9 5 7 i i ig : 4 ‘ : . . 5 Fue a « ns. . ; %y 3 , i : = iar Pl 1 5 , 7 Re t K ‘ on i ” 1 4 a : Fs = 5 ) + 3 F ‘ 4 * ‘ F 2 : j “ t Nea i + nr ’ J “ one, rr By ‘ ; ZOOLOGICAL SCIENCE 9: 101-111 (1992) Autogamy and Autogamy Inheritance in Euplotes woodruffi Syngen 1 (Ciliophora) TOSHIKAZU KOSAKA Zoological Institute, Faculty of Science, Hiroshima University, Higashi-Hiroshima 724, Japan ABSTRACT— In the Euplotes woodruffi complex which consists of syngen 1, syngen 2, syngen 3 and the autogamy group, autogamy is known in all of the stocks belonging to syngen 2 and the autogamy group but not in syngens 1 and 3. However, a recent study has revealed that one exceptional stock of syngen 1 is endowed with autogamy ability. Stock KB-16 which was collected from brackish water could undergo both autogamy and conjugation. The viability of exautogamous cells was generally over 70%. The length of autogamy immaturity was between 43 and 65 cell divisions in the exautogamous clones. Through autogamy, the stock produced viable Fl, F2, and F3 exautogamous clones with autogamy ability. Since non-autogamous clones have never been obtained among them, the parental stock is considered to be homozygous for the autogamy trait. All exautogamous clones derived from the autogamy stock could also conjugate with complementary mating type stocks of syngen 1. When the stock was crossed with a doublet clone from non-autogamous stock, all of the singlet and doublet F1 exconjugants from the cross showed autogamy ability. Thus, autogamy ability is hereditary also through conjugation. Because the results showed that autogamy occurred in heterozygotes and the heterozy- gotes produced progenies with and without the autogamy trait at a ratio of 3:1 through the next autogamy, the trait is considered to be a dominant allele. To study the features of autogamy in this stock is very important in clarifying the evolution of sexual reproduction in the E. woodruffi complex as well as the evolution of sexual reproduction in ciliates. From the results, a possible evolutionary relationship © 1992 Zoological Society of Japan of the E. woodruffi complex is discussed. INTRODUCTION Sexual reproduction in ciliates occurs either as conjugation between two individuals or by auto- gamy of single individuals [1-3]. Although con- jugation is a widespread method of fertilization in ciliates, autogamy which is considered to be a specialized mode of conjugation has only been described in a relatively limited number species including the Paramecium aurelia complex [4-6], P. polycaryum [7], P. jenningsi [8], Tetrahymena rostrata [9], Frontonia leucas {10], Euplotes minuta [11-14], E. crassus [15-17], E. woodruffi [18-25], Paraurostyla weissei [26] and Aspidisca costata [27]. Of the ciliate species with autogamy ability, the Paramecium aurelia complex, P. polycaryum, Tet- rahymena rostrata, Frontonia leucas, Paraurostyla Accepted September 25, 1991 Received April 1, 1991 weissei and Aspidisca costata are freshwater dwel- lers, while Euplotes minuta and E. crassus live in sea water. In the E. woodruffi complex, however, three syngens and the autogamy group show differ- ence in habitats and form of sexual reproduction: syngen | undergoes conjugation and lives in brack- ish water to sea water [28]; syngen 2 has both conjugation and autogamy abilities and lives in freshwater, but only autogamy is effective in main- taining individuals of the syngen [20-23]; syngen 3 undergoes conjugation, especially intraclonal con- jugation or selfing, and is a freshwater dweller [29]; the autogamy group is distributed widely in freshwater rivers or lakes and has only autogamy ability [18, 19, 30]. These findings also indicate that there is a very close relationship between habitats of species (syngens and group) and their form of sexual reproduction. One stock with autogamy ability has recently been found in studies conducted during the last 15 years on sexual reproduction of many stocks of 102 T. KosSAKA syngen 1 of Euplotes woodruffi. Heredity of the autogamy trait in the E. woodruffi complex has never been analyzed because autogamy but not conjugation is the only effective way of sexual repoduction in the stocks of the autogamy group and syngen 2 in the E. woodruffi complex. In this paper, using this autogamous stock of syngen 1, several features of the stock and its descendants are reported, such as viability of exautogamous and exconjugant cells, length of autogamy im- maturity and inheritance of the autogamy trait. In the discussion, the possible evolutionary process of the E. woodruffi complex is considered based upon the form of sexual reproduction. MATERIALS AND METHODS Stocks and culture conditions Twenty-nine stocks of the hypotrichous ciliate Euplotes woodruffi GAW syngen 1, collected from Ariake Bay in Saga Prefecture in Japan, were used in the experiments. In addition to these stocks, five stocks of complementary mating types of the same syngen were used: stock MY-1 belonging to mating type I, MY-2 to II, ZN-3 to III, MJ-2 to V and AT-7 to VII. Two doublet strains which were experimentally produced by treatment with colchi- cine [31] were also employed. All stocks and descendant clones were cultured at 20-21°C in a medium containing 9 parts of a wheat grain infu- sion (40-50 wheat grains per liter of distilled water, boiled for 10 min) and 1 part of Van’t Hoff artificial sea water, and were fed with the colorless flagellate Chilomonas paramecium cultivated in the same medium. In testing the segregation of the autogamy trait, each of the F2 exautogamous clones was cultured in a plastic Petri dish 3.5 cm in diameter and was transferred into new culture medium every week for over 4 months. Tests for autogamy immaturity To measure autogamy immaturity of descendant clones in terms of the number of fissions from the beginning of a new clonal life cycle (i.e., autogamy or conjugation), all of the clones were cultured by the 2-day period isolation method [29]. To deter- mine the length of the immature period for auto- gamy, the left-over cultures of the clones were used. To induce autogamy, cells of the left-over cul- tures were transferred to 10% sea water without food organisms and were starved for one day. Then a small amount of the food organism, in- adequate to give rise to more cell divisions, was added to the culture. This method was more effective in inducing autogamy than the stimulus of starvation alone and was slightly more complicated than that used in earlier work [18-23, 25, 30], because the induction of autogamy in syngen 1 seemed to be more difficult, especially in young ages, than in the autogamy group and syngen 2 of this Euplotes species. Statistical analysis Statistical analysis was made by y°-test. RESULTS Isolation of the stock with autogamy ability Twenty-nine stocks which were collected from water samples of 5.5 to 10.5%o salinity in Ariake Bay, Japan, were cultured by the 2-day period isolation method and were examined in detail for their sexual reproduction and syngen to which they belonged. All the stocks belonged to Euplotes woodruffi syngen 1 and underwent conjugation when complementary mating type stock was added to any one culture of the stocks. In addition to conjugation, one exceptional stock, stock KB-16, showed autogamy ability. Of 600 or more stocks of E. woodruffi syngen 1 collected over a 15 year period, stock KB-16 was the first stock with auto- gamy ability found in this syngen. Autogamy trait and viability of the exautogamous cells Of 50 exautogamous cells that were isolated from the culture of stock KB-16 when cells of the stock underwent autogamy, 46 exautogamous cells survived and grew into clones. These 46 Fl clones were cultured by the isolation method for about 2 months to test whether non-autogamous clones appeared in the Fl clones. All Fl clones have Autogamy Inheritance in Euplotes 103 TABLE 1. Viability of exautogamous cells ee Exautogamous cells Eh Viability alive dead Ze) KB-16(Parent) 46 4 50 D2 Fl-1 40 10 50 80 F1-2 34 16 50 68 F1-3 42 8 50 84 F1-4 45 >) 50 90 F2-1 19 1 20 95 F2-2 4 16 20 20 F2-3 15 5 20 is F2-4 12 8 20 60 F2-5 20 0 20 100 F2-6 15 5 20 i F2-7 19 1 20 95 F2-8 19 1 20 95 F2-9 19 1 20 95 F2-10 17 3 20 85 F2-11 20 0 20 100 F2-12 12 8 20 60 F1 and F2 clones were derived from parental stock KB-16 and F1 clones, respectively, through auto- gamy. autogamy ability. F2 and F3 clones derived through autogamy from the F1 clones were also examined, as mentioned above. All 161 F2 ex- autogamous clones and all 191 F3 exautogamous clones also underwent autogamy. These results suggest that stock KB-16 and its descendant clones are homozygous for the autogamy trait. The viability of exautogamous cells is shown in Table 1. The viability was high in F1, F2 and F3 clones, with only one exceptional case. These results show that autogamy is fully effective as a mode of sexual reproduction in maintaining the autogamous strains. Autogamy immaturity and the relationship between frequency of autogamy and clonal age The length of autogamy immaturity, which im- plies immaturity for autogamy in this case, was examined. Twenty-two exautogamous cells were examined for the length of autogamy immaturity by the isolation method. The length of autogamy immaturity was between 43 and 65 cell divisions. The frequency of exautogamous cells was 1-8% when autogamy first occurred in the left-over cultures of the clones. These results show that there is a period of autogamy immaturity which varies from clone to clone, and that the frequency of exautogamous cells is low at a young age. The relationship between the frequency of ex- autogamous cells and clonal age was examined. Of the 22 clones, 12 were employed by using the isolation method for more 2 months. The reuslts (Figure 1) showed that the frequency does not increase with clonal aging. Even at the age of 110 cell divisions, no clones reached 100% of exauto- gamous cells in the left-over cultures. The results are very different from those of the autogamy group and syngen 2 of this Euplotes species. Conjugation between autogamous and_ non- autogamous stocks Cells mature for autogamy could usually conju- gate with cells of complementary mating type. Although the cause is still unknown, cells of non- autogamous (a ) stocks induced some cells of autogamous (a~) stock to undergo autogamy, and a* cells often brought about selfing in a_ stocks which usually did not undergo selfing. Heterotypic pairs between a’ and a cells were naturally formed at a frequency of 50% or less. When the viability of exconjugant cells was tested, conjugat- ing pairs were obtained from the union of one a* cell and one a cell. Pairs obtained from this method should be truly heterotypic. Three a’ clones (about the age of 100 cell divisions) and5a_ mature stocks were used in the crosses. The viability of the exconjugants were 75-100% (Table 2). The high viability of the exconjugants suggests that conjugation as well as autogamy is effective in maintaining a° strains. Cells mature for autogamy usually could conju- gate with a_ mature cells when a* and a cells were mixed together. However, it is unclear whether autogamy and conjugation abilities were expressed at the same time in the life cycle. To obtain heterotypic pairs more easily, a new method was used in which a” singlet cells and a~ doublet cells were mixed together. The doublets used here were bimicronucleates which consisted of two complete singlet cells. Any conjugating pair 104 T. KoSAKA (*/o) 0 40 50 60 70 80 90 100 110 (°lo) 40 ° e 2 20 e 20 e e ee iz e 10 oe e 0 ERIN The GEES OEY. DE SMIEVAD & BROLIN EE 0 40 50 60 70 80 90 100 110 (%e) 40 e 30 3 20 is e ee e tJ 10 ° z 3 5 0 e e e 5 e 0 40 50 60 70 80 90 100 110 (*e) 40 ‘ és e e 4 e 20 e 10 ee? C 0 Z ® eo ¥ ee ° ° 0 40 50 60 70 80 90 100 110 (*le) 40 e 30 Bes 5 56 5 ee ° 10 e e ee e 0) a 0 40 50 60 70 80 90 100 110 (*lo) 70 60 ° v 50 5 6 40 ae 30 v 20 e e d e? O 10 e e 0 e e s 10} 40 50 60 70 80 90 100 110 (°lo) 0 40 50 60 70 80 90 100 110 (Ie) 40 5 30 ° = e 20 e ote 10 e ° e é ° 0) FFF aaa SS (0) 40 50 60 70 80 90 100 110 (6) 40 50 60 70 80 90 100 110 (lo) 50 e 40 e e 1@ do : Ono e 20 ° e ee 10 .° 5 Pat 0) —————— SS 10} 40 50 60 70 80 90 100 110 (*/o) 40 3 me oe a al Oe 20 ° 10 ee ° * Y 0 Zz eielter? e 0 40 50 60 70 80 90 100 110 (“/o) 40 12 30 20 10 : e 0 5 eo S e O (p-—_—*—_*,___2._-s-2—_-s_2_, —__#_>—__#_#_— 0 40 50 60 70 80 90 100 110 Fic. 1. Change of the percentages of exautogamous cells with increase of clonal age. Ordinate: the percentages of exautogamous cells; abscissa: the total number of fissions from clonal initiation. Twelve clones were used. 1: Clones: 2223 5-5 4:50) De O 505 7 10 os 14 ONG O19 shh 203122 between a singlet and a doublet was thus a hetero- typic pair. The viability of the exconjugants from heterotypic pairs was 73% and 50% in two matings (Table 3). The viability of singlet participants was higher than that of doublet ones: 78% versus 69% in the cross between clone 7 and doublet stock AT-7 and 63% versus 38% in the cross between clone 11 and stock AT-7. Expression of autogamy ability in the exconjugant clones Expression of autogamy ability was studied in 30 Autogamy Inheritance in Euplotes 105 TABLE 2. Viability of the exconjugants (singlet with singlet) Exconjugants Viabilit Crosses : Total y ; One alive, q Both alive Ore 27:4 ‘Both dead (%) F2-1 x AT-7(VII) 14 1 1 16 91 F2-7 xMY-2(II) 10 3 1 14 82 F2-7 x AT-7(VID) DD, D 0 24 96 F2-11 x MY-1(1) 5 5 0 10 iS) F2-11 x MY-2(II) 9 0 0 9 100 F2-11 x ZN-3(III) 7 0 0 7! 100 F2-11 x MJ-2(V) 15 0 0 15 100 F2-11 x AT-7(VI) 26 2 0 28 96 TABLE 3. Viability of the exconjugants (singlet with doublet) Exconjugants Viabilit Crosses : : ; Total y Singlet alive, Singlet dead, (%) Both alive doublet dead doublet alive Both dead F2-7 x AT-7(VII) 30 15 10 3 58 73 F2-11 x AT-7(VII) 7 8 2 v) 24 50 Stock AT-7 was a doublet strain. synclones from the cross between clone 7 and stock AT-7 (doublet strain). To test both autogamy and conjugation abilities, two doublet strains originat- ing from stocks MY-2 and AT-7 and two singiet stocks (MY-2 and AT-7) were used as testers, although the number of testers was insufficient. Singlet exconjugant clones were mixed with doub- let tester stocks, while doublet exconjugant clones were mixed with singlet tester stocks. After 2 to 3 months in the new life cycle of the clones cultured by the isolation method, 27 of 30 singlet clones showed both autogamy and conjuga- tion. These clones were considered to be hetero- zygotes for the a‘ trait, because conjugation occurred between a’ homozygotes and a homozygotes. The other 3 clones died within a month after the beginning of clonal life. On the other hand, 10 of 30 doublet clones also underwent both autogamy and conjugation. The other 20 clones changed to singlet clones by cell division or died within a month. The results show that the a* trait is due to a dominant allele because autogamy occurred in the heterozygote clones. Further, since autogamy occurred in the doublet clones, it seems likely that reciprocal exchange of gamete nuclei occurred normally. Doublet clones very frequently lost the micro- nucleus in one member of the doublet during the course of the isolation method and eventually became unimicronucleate doublets. The doublet clones maintained autogamy ability after becoming unimicronucleate. Doublets often produced two singlets at cell division. When unimicronucleate doublet clones produced one amicronucleate and one micronucleate singlet, the amicronucleate singlet always died, while the micronucleate sur- vived (Figure 2). Usually, the micronucleate sing- lets from the unimicronucleate doublet clones could undergo autogamy. However, micronucle- ate singlets from the unimicronuclear doublet clone 41W completely lost autogamy ability. This shows that the macronucleus in the amicronucleate component of clone 41W doublets controlled the a’ trait; however, the viability of the amicronuc- lear component was dependent on the micronuc- lear component. 106 T. KOSAKA ee — cell A B division aN loss of one micronucleus Cc Fic. 2. Formation of singlets with or without micronuc- leus from a unimicronuclear doublet with heteroka- ryon. A: a micronucleate cell which can live and reproduce normally. B: an amicronucleate cell which always dies without cell division. C: a uni- micronucleate doublet. Segregation of the a~ trait Segregation of the a‘ trait in F2 clones was tested by using heterozygous exconjugant clones (a’/a_). Autogamy was induced in 5 exconjugant clones. Fl exconjugants showed high viability in crosses between a” cells anda cells. However, the viability of F2 exautogamous cells dropped drastically when the exconjugants underwent the next autogamy. In F2 exautogamous cells derived through autogamy from the exconjugant cells, the viability was 8-30%. For example, only 762 of 2555 and 178 of 2274 exautogamous cells isolated were viable in exconjugant clone 4W and clone 21W, respectively. The exautogamous cells obtained were cultured for 4 months. If the a* trait was due to a dominant allele and autogamy occurred in the heterozygous clones, it would be expected that approximately 3 times as many a* progeny would be expressed as a_ progeny in the phenotype of the exautogamonts. The results, shown in Table 4, indicate that the a* trait segre- gated at a ratio of about3:1tothea trait through autogamy. DISCUSSION The finding of the a* stock, KB-16, in Euplotes woodruffi syngen 1 made it possible to study the heredity of the a* trait which could not be analy- zed thus far in the E. woodruffi complex, because all stocks belonging to the autogamy group and syngen 2 of the E. woodruffi complex underwent autogamy and only autogamy was the effective sexual process for rejuvenation [18-23]. In the a* stock of E. woodruffi syngen 1, both autogamy and conjugation resulted in high viability of the prog- eny. These facts thus provided advantages for analyzing the trait. The present studies show that offspring inher- ited the a* trait from their parent through auto- gamy and conjugation. Autogamy ability was expressed in heterozygotes for the a* trait and even in heterokaryon doublets whose genotypes were a /a anda‘/a andthe gene dose of a~ in the doublets was assumed to be one fourth of the totala* anda genes. Heterozygotes from a cross between a’ and a_ stock produced a ratio of 3a*:la~ progeny following autogamy. These results suggest that a* is a dominant allele of a single locus with a pair of alleles; the mode of heredity thus is very similar to that of E. minuta [11, 12] and E. crassus [17]. In the a® stock and its descendants, there was autogamy immaturity fol- lowing the previous sexual reproduction. The TABLE 4. Segregation of autogamy ability in F2 clones Parental Autogamy ability in F2 clones Fl clones with without mae a ul 4S 74 20 94 0.695 0:3 P05 4W 89 WS 114 0.573 03<—PR0:5 21S 53 16 69 0.120 O7—P< Os 21W 135 34 169 2.148 01 P02 16W aS Dal 96 0.5 0.3 petenaie) Ang Res er : of, ate 7 a hE = > at - a i , ° c p : =) Pee 9 a oo es i -7 hb fy "i aA - = af =) 2 = t x ZOOLOGICAL SCIENCE 9: 113-118 (1992) © 1992 Zoological Society of Japan Granulocytes and Macrophages in Amphioxus HoncwE!I ZHANG)”, ZHE HUANG’, KazuHITO YAMAGUCHP and Susumu ToMoNAGA! 'School of Allied Health Sciences, Yamaguchi University, Ube 755, Japan, "Department of Biology, Shandong University, Jinan 250100, China, and *Institute of Laboratory Animals, Yamaguchi University Shool of Medicine, Ube 755, Japan ABSTRACT—Two types of mononuclear free cells were found in amphioxus. Cell type 1 with specific granules and cytoplasmic processes was usually observed in the lumen of blood vessels. The granules contained microtubule-like structures which are commonly present in insect granulocytes. Type 1 blood cell is referred to as “granulocyte”. The presence of an endothelial lining as an independent cell type was doubtful in blood vessels. Type 2 cell, the “macrophage” possessed numerous cell processes, lysosomal granules, vesicles and multivesicular bodies. Macrophages were detected not only in the coelom but also in other body spaces, the so-called lymph spaces. These two cell-types probably effect crucial immuno-defense functions in amphioxus. INTRODUCTION Amphioxus, a cephalochordate, often regarded as sharing a common ancestry with vertebrates, is an important experimental animal model for phy- logenetic analyses in the biological sciences includ- ing comparative immunology. However, little is known about the immune system. Although some workers believe that amphioxus possesses no blood cells in its vascular system [1-4], others have observed a few erythrocytes [5] and granulocytes [6, 7]. Moreover phagocytic coelomocytes occur in the body cavity of amphioxus after injection of bacteria [3]. In addition to the presence of these cells, there is a humoral factor i.e. a hemagglutinin which has been defined by Bretting and Renwrantz [8] and De Benedictis and Capalbo [9]. As one of the basic steps necessary for extending and clar- ifying information on the immuno-defense system of amphioxus, observations on the ultrastructure of blood cells, vascular system, coeiom and so- called lymph spaces [10, 11] were performed. Correspondence should be addressed to Dr. S. Tomo- naga. Accepted October 16, 1991 Received September 13, 1991 MATERIALS AND METHODS Amphioxus Adult Amphioxus (Branchiostoma _ belcheri tsingtauense), which were caught in the Yellow Sea, were supplied by The Chinese Institute of Oceanology, Oingtau, China. The animals were divided into three groups. Transmission electron microscopy (TEM) The first was used for transmission electron microscopy by fixing them in a mixture of 2% glutaraldehyde and 2% paraformaldehyde (GA/ FA) [12] for 8 hours at room temperature. The specimens were then postfixed in 1% osmium tetroxide for 1 hour and cut transversely into 1-2 mm blocks. Following dehydration with ethanol, tissue blocks were embeded in Epon 812. Ultrathin sections, stained with uranyl acetate and lead citrate, were observed under the JEM-200CX electron microscope (Japan Electron Optics Ltd.). Scanning electron microscopy (SEM) The second group was prepared for scanning electron microscopy by fixing them in GA/FA for 8 hr. Passing through tannic acid [13] the tissues were post-fixed in 1% osmium tetroxide, dehy- 114 H. ZHANG, Z. HUANG et al. drated in acetone and dried by the critical point dry method. Observations were made under the JSM T300 scanning electron microsocpe (Japan Electron Optics Ltd.). Light microscopy (LM) The third group was fixed for light microscopy in Bouin solution for 18 hours. After washing they were dehydrated in ethanol, embedded in paraffin and cut as longitudinal or transverse serial sections and stained with hematoxylin and eosin. RESULTS The coelom and lymph spaces Amphioxus has two kinds of spaces: coelom and the so-called lymph space. These spaces, in addi- tion to blood vessels, were confirmed by SEM (Figs. 1,2). Figure 1 was cut surface of pharyngeal region and Fig. 2 was between the atriopore and the anus. Granulocytes in blood vessels The dorsal aortae, the endostylar artery, the hepatic vein and the branchial vessels were ex- amined by both LM and EM. Under LM blood vessels appeared as simple structures entirely dif- ferent from those of vertebrates. One portion of the pharyngeal vascular wall consisted of only connective tissue and the other was composed essentially of epithelial cells and a basement mem- brane. The presence of endothelial cells was not confirmed in all blood vessels. A small number of mononuclear cells which stained consistently with eosin were often observed in the lumen of blood vessels. The cells were easily observed under the SEM (Fig. 3). By TEM, mononuclear cells in the vascular Space appeared to be variable in shapes, such a round, flattened or irregular (Figs. 4 and 5). Their most prominent feature was the presence of many specific cytoplasmic granules (Figs. 4 and 5). At higher magnification, the microtubule-like struc- tures, 15-20 nm in diameter were detected in the Fic. 1, 2. Scanning electron micrographs of cross sections of the amphioxus in the pharyngeal region (Fig. 1) and between the atriopore and anus (Fig. 2). intestine, ls: lymph space, n: notochord, p: pharynx, pv: postcardinal vein. a: atrium, c: coelom, cn: central nervous system, d: dorsal aorta, 1: x 100. Amphioxus Blood Cells 115 +0: NS < ae S ee lumen of blood vessel. 8,500. Fic. 4-6. Transmission electron micrographs of granulocytes (G) in the lumen of blood vessels. Fic. 4, 5. Lower power views of the granulocytes, which have pseudopods and specific granules. material in the luminal space of the blood vessel. 8,700 (Fig. 4), x 12,300 (Fig. 5). Fic. 6. A higher power view of specific granules having microtubule-like structure (arrows). 58,400. Note dense 116 H. ZHANG, Z. HUANG et al. granules (Fig.6). In addtion to granules, the | complex, free ribosomes and mitochondria, all cytoplasm usually contained an endoplasmic re- easily observed in ultrathin sections. These cells ticulum with rather dilated cisternae, the Golgi usually possessed cytoplasmic processes. We clas- = ee Se Ww OS Fic. 7. A scanning electron micrograph of macrophages (M) in the subdermal sp Cer one of the so-called lymph spaces, of amphioxus. 4,600. EE: epidermis, ct: connective tissue, ls: lymph space, M: macrophage Fic. 8. A transmission electron micrograph of macrophages (M) in the so-called lymph space. 14,400. Sis 5 aC Amphioxus Blood Cells 117 sified them as granular leukocytes because of their morphological characteristics. Although granulo- cytes were often observed near the internal wall of blood vessels, any intercellular junctions were never detected between two cells. In some sec- tions granulocytes were also found in the central area of the blood vessel lumen. Macrophages in so-called lymph spaces and the coelom Examination of longitudinal and transverse se- rial sections revealed the presence of free mono- nuclear cells in so-called lymph spaces, such as the metapleural spaces, spaces surrounding the dorsal aorta and those between muscle fasciae, the sheath of the notochord and subdermal spaces. The presence of cells in these spaces was confirmed by SEM (Fig. 7). Under SEM, macrophages appear- ed to be irregular in shape with many small cell surface pits. By TEM macrophages had numerous small vesicles which may correspond to the small pits observed by SEM. Numerous lysosomal gra- nules, phagosomes and multivesicular bodies were also found in the cytoplasm (Fig. 8). In addition to their presence in lymph spaces numerous macrophages were also found in many other specific regions of the coelom, such as the perigonadal coelom, the endostyle coelom and branchial coelom by both LM and SEM. Howev- er, macrophages were rarely observed in the supra- pharygeal or the subchordal coelom and in the perienteric coelom. Some macrophages lined the connective tissue of the coelomic wall. Such cells usually had more electron dense cytoplasmic bodies. DISCUSSION For a long time, many biologists generally be- lieved that there were no blood cells in amphioxus [1-4], although a few authors observed them [5- 7]. In our study we found mononuclear free cells having specific cytoplasmic granules. These gra- nules had specific microtubule-like structure simi- lar to those described in insect blood cells [14-17]. These findings strongly support our assigning them the name “granulocytes”. Moller and Philpott [2] investigated the ultra- structure of amphioxus blood vessels. They re- ported the absence of circulating blood cells but there were discontinuous endothelil lining cells within blood vessels. The features of endothelial cells reveals them clearly identical to granulocytes which we describe in this report. The contradic- tion between these two descriptions seems to be due only to classiflying them as endothelial cells. It should be emphasized that these cells, the granulo- cytes according to our terminology were observed not only in peripheral areas of blood vessels but also in central areas of the lumen of vessels. Granulocytes generally had cytoplasmic pro- cesses suggesting the capacity for ameboid move- ment. Moreover, cytochemical investigation by Moller and Philpott [18] demonstrated phagocytic activity and localization of acid phosphatase within cytoplasmic granules. Their findings together with our observations strongly suggest that the term “granulocyte” is entirely useful. There is one possibility that remains to be solved. Both the endothelial cells and the blood cells may originate from a common ancestor as cells which originally distribute on the vessel wall, since the presence of discontinuous endothelial cells is known in certain blood vessels of some invertebrate species [19]. The presence of coelomocytes in amphioxus was not clear until 1982 when Rhodes et al. [3] detected them after injecting bacteria. In the present study we found macrophages in non-stimulated, normal animals. The cells were observed not only in the coelom but also in the so-called lymph spaces described by Kampmeier [10] which were thought to be discontinuous to the coelom. Macrophages in various body spaces and granulocytes in blood vessels probably contribute to the immune defense mechanisms. Tunicates, (Urochordata) are another group of protochordates having many species and a univer- sal distribution. Clearly tunicates have diverse hemocyte types in their hemolymph [20]. In our recent studies using Halocynthia roretzi, about ten cell types including granulocytes and macrophages have been identified [21, 22]. We assume that these common cell-types, granulocytes and mac- rophages, present in both species, are essential cell members of the immuno-defense system of pro- tochordates. 118 ACKNOWLEDGMENTS We thank the staff of The Chinese Institute of Oceanology, Qingtau, China for kind support in sup- plying amphioxus. We express appreciation to Professor Edwin L. Cooper, Laboratory of Comparative Immunol- ogy, University of California, Los Angeles who critically read the manuscript. We also thank Mr. Akira Kuma- kura for technical assistances. 10 REFERENCES Franz, V. (1927) Morphologie der Akranier. Ergeb. Anat. Entwicklungsgesch., 27: 464-692. Moller, P. C. and Philpott, C. W. (1973) The circulatory system of amphioxus (Branchiostoma floridae) 1. Morphology of the major vessels of the pharyngeal area. J. Morphol., 139: 389-406. Rhodes, C. P., Ratcliffe, N. A. and Rowley, A. F. (1982) Presence of coelomocytes in the primitive chordate amphioxus (Branchiostoma lanceolatum). Science, 217: 263-265. Rowley, A. F., Rhodes, C. P. and Ratcliffe, N. A. (1984) Protochordate leucocytes: a review. Zool. J. Linn. Soc., 80: 283-295. Parker, T. J., Haswell, W. A. and Foster-Cooper, C. (1949) A Text-Book of Zoology. Vol II. Macmil- lan, London, 6th ed. p. 48. Hilton, W. A. (1943) The blood of Amphioxus (Branchiostoma). J. Entomol. Zool., 35: 31-32. Welsch, U. (1975) The fine structure of the pharynx, cyrtopodocytes and digestive caecum of amphioxus (Branchiostoma lanceolatum). Symp. Zool. Soc. Lon., 36: 17-41. Bretting, H. and Renwrantz, L. (1973) Unter- suchungen von Invertebraten des Mittelmeeres auf ihren Gehalt an hamagglutinierenden Substanzen. Z. Immun. -Forsch., 145: 242-249. De Benedictis, G. and Capalbo, P. (1981) An Amphioxus lanceolatus agglutinating factor for hu- man red cells. J. Immunogenetics, 8: 225-230. Kampmeier, O. F. (1969) Evolution and Compara- tive Morphology of the Lymphatic System. Charles C Thomas Publisher, Springfield, Illinois, pp. 160- At 12 113) 14 IS 16 ie, 18 19 20 21 aD H. ZHANG, Z. HUANG et al. WB: Lankester, E. R. (1889) Contributions to the know- ledge of Amphioxus lanceolatus, Yarrell. Quart. J. Micr. Sci., 29: 365-402. Karnovsky, M. J. (1967) The ultrastructural basis of capillary permeability studied with peroxidase as a tracer. J. Cell Biol., 35: 213-236. Murakami, T. (1974) A revised tannin-osmium method for non-coated scanning electron micro- scope specimens. Arch. Histol. Jap., 36: 189-193. Baerwald, R. J. and Boush, G. M. (1970) Fine structure of the hemocytes of Periplaneta americana (Orthoptera: Blattidae) with particular reference to marginal bundles. J. Ultrastruct. Res., 31: 151-161. Devauchelle, G. (1971) Etude ultrastructurale des hémocytes du Coléoptére Melolontha melolontha (L.). J. Ultrastruct. Res., 34: 492-516. Gupta, A. P. (1979) Hemocyte types: their struc- tures, synonymies, interrelationships, and taxono- mic significance. In “Insect Hemocytes. Develop- ment, Forms, Functions, and Techniques”. ed. by A. P. Gupta, Cambridge University Press, Cam- bridge, pp. 85-127. Rowley, A. F. and Ratcliffe, N. A. (1981) Insect. In “Invertebrate Blood Cells 2”. eds. by N. A. Ratcliffe and A. F. Rowley, Academic Press, London, pp. 421-488. Moller, P. C. and Philpott, C. W. (1973)” Whe circulatory system of amphioxus (Branchiostoma floridae) 11. Uptake of exogenous proteins by en- dothelial cells. Z. Zellforsch., 143: 135-141. Casley-Smith, J. R. (1980) Comparative fine struc- ture of the microvasculature and endothelium. Adv. Microcirc., 9: 1-44. Wright, R. T. (1981) Urochordate. In “Invertebrate Blood Cells 2”. eds. by N. A. Ratcliffe and A. F. Rowley, Academic Press, London, pp. 565-626. Sawada, T., Fujikura, Y., Tomonaga, S. and Fuku- moto, T. (1991) Classification and characterization of ten hemocytes types in the tunicate Halocynthia roretzi. Zool. Sci., 8: 939-950. Zhang, H., Sawada, T., Cooper, E. L. and Tomona- ga, S. (1991) Electron microscopic analysis of tuni- cate (Halocynthia roretzi) hemocytes. Zool. Sci., 9 (In press). ZOOLOGICAL SCIENCE 9: 119-125 (1992) The Purification and Characterization of Sepiaterin Reductase from Fat Body of the Silkworm Bombyx mori 'TERUHIKO IINO, 7KENJIRO DOHKE and *Motoo Tsusuf ‘Department of General Education, Nihon University, Sakurajosui, Setagayaku, Tokyo, 156. *Biological Laboratory, Kitasato University, Sagamihara, Kanagawa, 228. ABSTRACT—The enzyme sepiapterin reductase has been purified about 68,000 fold from the fat body of silkworm larvae by several column chromatographic procedures and its properties were compared with those of the same enzyme from other organisms. The molecular weight of the enzyme is 29,000 by SDS polyacrylamide gel electrophoresis and 59,000 by Ultrogel AcA 44 gelfiltration, suggesting that the native form of the enzyme consists of two identical subunits. The Km values for sepiapterin and NADPH are 10.2 uM and 1.3 uM respectively, and the V,,,, value for sepiapterin is 21.4 ~mol/min/ mg. Sepiapterin reductase from silkworm fat body and rat erythrocytes reduce several other nonpteridine carbonyl compounds in the presence of NADPH. Distribution of sepiapterin reductase activity in the larval tissues of the silkworm was examined and found in many tissues but most of the activity was detected in fat body. Viability similar to the normal strain was observed in the silkworm © 1992 Zoological Society of Japan lemon mutant which lacks sepiapterin reductase. INTRODUCTION The enzyme sepiapterin reductase (EC 1.1.1. 153) catalyzes the reversible formation of dihydro- biopterin (L-erythro-6-[1, 2-dihydroxypropyl]-7, 8- dihydropterin) and NADP from sepiapterin (6- lactyl-7, 8-dihydropterin) and NADPH [1-3]. This enzyme catalyzes the terminal biosynthesis of tet- rahydrobiopterin (L-erythro-5, 6, 7, & tetrahydrobiopterin) from 6-pyruvoyl-5, 6, 7, 8- tetrahydropterin. According to the recent reports, this enzyme catalyzes the isomerization of 6-1'- oxo-2’-hydroxypropyl tetrahydropterin to 6-1’- hydroxy-2 -oxopropyl _tetrahydropterin. This isomerization reaction may be part of the natural pathway for tetrahydrobiopterin biosynthesis [4, 5, 6]. Tetrahydrobiopterin is a natural cofactor for the pterin-dependent aromatic amino acid hyd- roxylases [7] and for the o-alkyl glycerolipid cleav- age enzyme [8], which is involved in neurotrans- mitter biosynthesis and lipid metabolism. Recent- ly, Mahmound et al. has reported that tetrahydro- biopterin is also a cofactor for N-hydroxylation of Accepted July 29, 1991 Received June 25, 1991 arginine [9]. Although sepiapterin reductase was first discovered in silkworm lavae [10], no further work was reported on the silkworm enzyme. It was however first purified from horse liver, subse- quently from rat erythrocytes and whole brain and its physicochemical and catalytic properties were studied in detail [11-13]. The medical significance of the enzyme has lead biochemists to study it in detail from mammalian sources. However pterin metabolism may have other significant roles for the insect since they recognized near ultraviolet light as a visible color. Thus, these fluorescent sub- stances effect their behavior. In this present paper, we describe a method for the purification of sepiapterin reductase from silk- worm fat body and compare its physicochemical and catalytic properties with those of the enzyme in other organisms. MATERIALS AND METHODS Experimental animals and tissue preparation Silkworm larvae were reared routinely on mul- berry leaves at 25°C. On the 5 th day of the 5 th instar, the silkworm larvae were dissected with 120 T. Inno, K. DOHKE AND M. Tsusu£ scissors and fat bodies were pooled and kept frozen at —80°C until use. : Chemicals Sepiapterin was purchased from Dr. Schircks, Switzerland and further purified and recrystallized by the method of Sugiura et al. [14]. Deoxysepiap- terin (2-amino-4-hydroxy-6-propionyl-7, 8-dihy- dropteridine) and _ 6-acetyl-dihydropterin (2- amino-4-hydroxy-6-acetyl-7, 8-dihydropteridine) were prepared from a Drosophila melanogaster mutant sepia [14]. Sepialumazin (2, 4-dihydroxy- 6-lactyl-7, 8-dihydropteridine) was prepared by enzymatic deamination of sepiapterin [15]. Erythropterin (2-amino-4, 6-dihydroxy-7-pyruvoy- Ipteridine) was a kind gift from Dr. W. L. Gyure of Seton Hall University. Butyl Toyopearl 650S and DEAE-Toyopearl 650S were from Toyo Soda Mfg. Co. (Tokyo). Cytochrome c, myoglobin, egg albumin and bovine serum albumin were from Sigma Chemical Co. (St. Louis, Mo.). All other chemicals were reagent grade from commercial sources. Enzyme assay The standard reaction mixture for sepiapterin reductase assay contained the following compo- nents (in ~ moles): potassium phosphate buffer, pH 6.5, 250; sepiapterin 0.1; NADPH 0.25; and 2 to 1200 vl of the enzyme, in a final volume of 2.5 ml. Enzyme activity was determined at 25°C by measuring the decrease in absorbance at 420 nm and/or 340 nm by a Hitachi U-3200 spectrophoto- meter. Carbonyl reductase activity was assayed by the decrease at 340 nm (16). The distribution of sepiapterin reductase in silk- worm larvae was determined as follows; The stan- dard reaction mixture of final volume 125 ul was incubated at 37°C for 10min. The reaction was stopped by adding 20 ul of 20% TCA and the precipitate was removed by centrifugation. To the supernatant 15 wl of a solution of 1 g I, dissolved in 100 ml of a solvent of 2 g KI in 100 ml water was added, the mixture was allowed to stand for 30 min at room temperature. From the mixture, 50 ul aliquots were injected into a Whatman SCX HPLC column. The mobile phase for the column system was 10mM potassium phosphate buffer pH 7.0, pumped at a flow rate of 1.0 ml/min. Biopterin was detected on elution by its fluorescence (excita- tion at 350 nm, emission at 450 nm). Substrates and inhibitors difficult to dissolve in water were first dissolved in aqueous ethyl alcohol and then added to the reaction mixture to give a final concentration of alcohol less than 0.5%, a concentration which had no detectable effect on the catalytic activity of the enzyme. One unit of the enzyme activity was defined as that amount catalyzing the conversion of 1 «mol sepiapterin per min, using an extinction coefficient for sepiapterin of 10.4x10°M~'xcm~' at 420nm [17] and NADPH of 6.210°>M~!xcm~! at 340 nm, re- spectively. Specific activity was expressed in m units per mg protein. Protein concentrations were measured with the Bio-Rad protein assay kit using bovine gamma globulin as a standard. Molecular weight determination The molecular weight of the enzyme was esti- mated by gel filtration based on the method of Andrews [18]. A column of Ultrogel AcA 44 (1.5 x 100 cm) was equilibrated with 20 mM potassium phosphate buffer pH 6.5. Polyacrylamide gel elec- © trophoresis in the presence of SDS was carried out as described by Laemmli [19]. RESULTS AND DISCUSSION Enzyme purification All procedures were carried out at 4°C, unless otherwise stated. Step 1. Preparation of crude extract. Frozen tis- sues (500 g) were homogenized with 3 vols. of 10 mM potassium phosphate buffer pH 6.5, contain- ing ammonium sulfate 20% saturation (Buffer A). The homogenate was centrifuged at 16,000 g for 40 min. Step 2. Amonium sulfate fractionation. Solid ammonium sulfate was added to the crude extract to 40% saturation and the mixture stirred for 30 min. The precipitate was removed by centrifuga- tion at 16,000g for 10min. Additional solid ammonium sulfate was added to the supernatant to 60% saturation. The mixture was stirred for 30 min and the precipitate was collected by centri- Purification and Characterization of Silkworm Sepiapterin Reductase 121 fugation at 16,000 g for 10 min, and dissolved in 0.3 vols (frozen tissue wet wt.) of Buffer A. Any insoluble materials was discarded by centrifugation at 33,000 g for 60 min. Step 3. Butyl-Toyopearl column chromatogra- phy. The enzyme solution was applied to a col- umn of Butyl-Toyopear!l 650S (2.5 x 43 cm) equili- brated with Buffer A. After washing the column with Buffer A, the adsorbed protein were eluted with 10 mM potassium phosphate buffer, pH 6.5, containing 10% saturation of ammonium sulfate. The flow rate was maintained at 40 ml/hr and fractions of 10 ml each were collected. The active fractions were pooled. Step 4. Phynyl-Sepharose column chromatogra- phy. The enzyme solution was applied to a col- umn of Phenyl-Sepharose (1.5<26cm) equili- brated with 10mM potassium phosphate buffer, pH 6.5, containing 0.1 M ammonium sulfate. Af- ter washing the column with 100 ml of the same buffer, the column was washed again with 10 mM potassium phosphate buffer, pH 6.5, containing 40 mM ammonium sulfate until the 280 nm absorb- ance reading of the eluates had decreased to the starting value. Elution was carried out with distil- led water. The flow rate was kept at 25 ml/hr and fractions of 7 ml each were collected. Step 5. DEAE-Toyopearl column chromatogra- phy. The enzyme solution from step 4 was ad- justed by adding 100mM potassium phosphate buffer, pH 7.2, containing 50% ethyleneglycol un- til a final concentration of 20 mM potassium phos- phate buffer, pH 7.2, containing 10% ethylenegly- col was reached. This solution was applied to a column of DEAE-Toyopearl 650S (1.5 x22 cm) equilibrated with 20 mM potassium phosphate buf- fer, pH 7.2, containing 10% ethyleneglycol. After washing the column with 100ml of the same buffer, elution was carried out with a linear gra- dient of 0-0.3 M KCl in the same buffer. The flow rate was maintained at 25 ml/hr and fractions of 2 ml each were collected. The activity was usually eluted in a narrow range of molarity (approximate- ly 150 mM KCl). Step 6. Hydroxylapatite column chromatogra- phy. The enzyme solution from step 5 was ad- justed to pH 6.8 by adding KH,PO,. This solution was applied to a column of hydroxylapatite (1 x 17 cm) equilibrated with 20 mM potassium phosphate buffer, pH6.8, containing 10% ethyleneglycol. After washing the column with 60 ml of the same buffer, elution was carried out with a linear gra- dient of 20-300 mM potassium phosphate buffer, pH 6.8, containing 10% ethyleneglycol. The flow rate was maintained at 7 ml/hr and fractions of 1 ml each were collected. The active fractions were combined and adjusted to 0.1 M ammonium sul- fate by solid ammonium sulfate. This solution was applied to a short column of Phenyl-Sepharose (0.55 cm) equilibrated with 10mM potassium phosphate buffer, pH6.5, containing 0.1M ammonium sulfate. Elution was carried out with distilled water. The flow rate was maintained at 3.5 ml/hr and fractions of 0.5 ml each were col- lected. A typical purification procedure is summa- rized in Table 1. Sepiapterin reductase from the silkworm fat body was purified about 68,000 fold with a 25% recovery. TABLE 1. Purification of sepiapterin reductase from silkworm fat body Total Total Specific : Step Activity Protein Activity a ecacn (unit) (mg) (mU/mg) (%) (fold) 1. Homogenate S)o45) 15800 0.33 100 1 2. A.S. Fractionation 5.10 3410 33) OF 4.5 3. Butyl-Toyopearl SoA 151 34.5 100 105 4. Phenyl-Sepharose 5.60 20.4 ZS) 107 832 5. DEAE-Toyopearl 4.23 2.6 1630 81 4930 6. Hydroxylapatite 1.30 0.058 22400 25 68,000 A.S: Ammonium sulfate 12? T. Inno, K. DOHKE AND M. TsusuE Determination of molecular weight The molecular weight of silkworm sepiapterin reductase was estimated to be 59kDa by gel filtration on an Ultrogel AcA 44 column (Fig. 1) and 29kDa by SDS polyacrylamide gel elec- trophoresis. Therefore, the enzyme consists of two identical subunits, each of which has a molecular weight of 29kDa. The molecular weight of horse liver sepiapterin reductase was reported to be 47 kDa [11], 68 kDa (monkey liver) [20] and 55 kDa (rat brain) [13]. Rat erythrocytes sepiapterin re- ductase is composed of two identical subunits of 27.5 kDa giving the enzyme an overall molecular weight of 55kDa [12]. This value is in good agreement with the silkworm sepiapterin reduc- tase. 100 90 E 2 80 EggAl SPR (59,000) a 70 60 1 Za 4 9G WC MOLECULAR WEIGHT (10°) Fic. 1. Molecular weight determination by Ultrogel AcA 44 gel filtration. Marker protein used were: Cyt-c, cytochrome c (M. W. 12,300); Myo, myoglo- bin (M. W. 18,800); Egg Al, egg albumin (M. W. 45,000); B.S.A., bovine serum albumin (M. W. 66,000); S.P.R., Sepiapterin reductase from fat body of silkworm pH optimum and enzyme kinetics The pH optimum of this enzyme was 5.2. The enzyme was assayed in 100 mM potassium phos- phate buffer between pH values of 4.8 and 7.2, while 100 mM Tris-HCl buffer was used for values from 6.5 to 8.0. The apparent Km values for sepiapterin and NADPH were determined according to the method of Lineweaver and Burk [21]. Activity tests were carried out at substrate concentrations of 5S—40 «eM sepiapterin (60 «~M NADPH) and 10- 60 «M NADPH (40 uM sepiapterin) at pH6.5. The apparent Km values for sepiapterin and NADPH were 10.2 «M and 1.3 uM, respectively, and the Vmax value for sepiapterin was 21.4 ymol/min/mg. The Km value for sepiapterin reductase from mammalian sources was reported as 14-23 uM [11, 12, 20, 22]. Biopterin synthase, an enzyme similar to sepiapterin reductase, which also catalyzes the NADPH-dependent conversion of sepiapterin to dihydrobiopterin, was found in Drosophila melanogaster |23|. The Km value for sepiapterin of this enzyme was reported to be 63 uM. A comparison of the results obtained for Km values from these several different organisms indi- cates that the silkworm sepiapterin reductase more closely resembles the mammalian enzymes than does the Drosophila melanogaster enzyme. Substrate specificity The pteridine derivatives listed in the Table 2 were tested for their susceptibility as substrates. Deoxysepiapterin and 6-acetyl-dihydropterin are 6-sidechain analogues of sepiapterin. Sepialuma- zine is a corresponding lumazine of sepiapterin. Erythropterin was completely inactive as a sub- strate, while the above described analogues were slightly affected by the enzyme (Table 2). Sepiap- terin reductase from silkworm fat body shows a strict specificity for sepiaptern in the same manner as mammalian enzyme. On the other hand, Katoh and Sueoka reported that sepiapterin reductase from rat erythrocyte reduced some vicinal dicar- bonyl nonpteridine derivatives in the presence of NADPH [16, 24]. The silkworm enzyme was examined for this property. The results show that some dicarbonyl and a quinone. (e.g. phenylpro- TABLE 2. Substrate specificity of sepiapterin reduc- tase from silkworm fat body eee Enzyme Activity (%) Substrate 420 nm 340 nm Sepiapterin 100 100 Deoxysepiaterin 4.0 6.0 6-acetyl-7,8-dihydropterin 3.6 6.0 Sepialumazin 335 520) Erythropterin — 0.0 a Purification and Characterization of Silkworm Sepiapterin Reductase 11723} TABLE 3. Km and Vmax values of silkworm sepiap- terin reductase for dicarbonyl compounds and a quinone Km Vmax Substrate (uM) (smol/min/mg) Sepiapterin 10.2 De? 1-phenyl-1,2-propanedion 3 Urrell Benzil 7.7 =~ Menadione 25.0 S)0) The aparent values of the Km and Vmax were estimated by double reciprocal plots with NADPH. panedion, benzil and menadione) are reduced in the presence of NADPH (Table 3). This property is quite similar to that of the rat erythrocytes enzyme. Inhibitors Inhibition of silkworm sepiapterin reductase us- ing sepiapterin and 1-phenyl-1, 2-propanedione as substrates was examined with various inhibitors of general aldo-keto reductase and mammalian sepiapterin reductase (Table 4). Phenobarbital and pyrazole slightly inhibited enzyme activity. Rutin and indomethacin, which are reported in- hibitors of rat erythrocytes sepiapterin reductase and human carbonyl reductase [24, 26], caused a 50% inhibition of enzyme activity at a concentra- tion of about 504M. N-acetylserotonine and 6-carboxypterin, which are reported to be potent TABLE 4. and sepiapterin inhibitors of mammalian sepiapterin reductase [24, 25], also inhibited silkworm sepiapterin reductase activity. Dicoumarol, which inhibit rat sepiapterin reductae and human carbonyl] reductase [24, 26], was the most effective inhibitor causing 50% in- hibition of the silkworm enzyme at a concentration of 0.04 uM. Sepiapterin reductase from silkworm fat body has properties similar to those reported for mammalian sources, especially for rat erythro- cytes [24]. Tissue distribution of sepiapterin reductase in silk- worm Several tissues of the silkworm larvae were removed and washed with Ringer’s solution. The tissues were homogenized 3 vol. of 10 mM potas- sium phosphate buffer, pH 7.0. The homogenate was centrifuged at 16,000 g for 30 min. and the supernatnat was brought to 70% saturation by the addition of solid ammonium sulfate. The precipi- tate was collected by centrifugation and dissolved in 0.3 vol. of 10 mM potassium phosphate buffer and dialyzed against 100 vol. of same buffer over- night. The dialysate was used as enzyme prepara- tion. Although the enzyme activities were distri- buted in many tissues of the silkworm, most of the ‘activity was detected in fat body as shown in Table 5), Sepiapterin reductase may be an indispensable enzyme for all animals, because it plays an impor- tant role in tetrahydrobiopterin biosynthesis. The Inhibition of silkworm sepiapterin reductase with phenylpropanedione Inhibition (Is) nN Sepiapterin 1-phenyl-1,2-propanedione (uM) (uM) Pyrazole 1,000 750 Phenobarbital 760 500 Indomethacin 43 Sy) Rutin 65 44 Dicoumarol 0.04 0.04 6-carboxypterin 47 40 N-acetyl-serotonin ies 1:3 The inhibiton was examined in the standard reaction mixture for 3 min in the presence of 30 uM sepiapterin and 60 ~M NADPH at pH6.5. 124 T. Inno, K. DOHKE AND M. TsusuE enzyme is widely distributed in mammalian tissues and the specific acitivity of this enzyme is uniform- ly equal in many tissues [27]. But specific acitivity of sepiapterin reductase in the larval fat body is 20-100 fold higher than found in other tissues (Table 5). TaBLE 5. Distribution of sepiapterin reductase in the larval tissues of the silkworm Enzyme activity Tissue (pmol BP/10 min/g protein) Fat body 19.4 Malpighian tube 0.258 Midgut 0.570 Posterior silk gland 0.754 Integument IIS) Body fluid 0.026 BP: Biopterin Viability study on lemon mutant Genetic background of the mutant /Jemon which lacks sepiapterin reductase was made uniform to a normal strain (C-108) by cross mating experiment. Then we compared the viability of the mutant with the normal strain. Throughout the developmental stage no significant difference was observed in body weight, mortality and number of eggs per egg-mass. Due to the lack of tetrahydrobiopterin caused by inborn or acquired deficiencies in its synthesis, many severe diseases result, such as a atypical phynylketonurea or hyperphenylalanaemia [28, 29, 30]. Lack of sepiapterin reductase, which catalyzes terminal biosynthesis of tetrahydrobiop- terin, has not yet been found in these patients, probably because a lack of this enzyme may be lethal. On the other hand, mutant /emon has the same viability as the normal strain described above. This fact may suggest that the cofactor of aromatic amino acids hydroxylase in the silkworm can be replaced by another reduced pterin (e.g., reduced sepiapterin or tetrahydroneopterin) instead of tet- rahydrobiopterin. We think that sepiapterin reductase in silkworm larvae may have functions besides tetrahydrobiop- terin biosynthesis. It may, for instance, have a role in pigment formation in insects. Further studies should clarify these points. Sepiapterin reductase from silkworm fat body has catalytic and physiological properties similar to those reported for mammalian sources, especially for rat erythrocytes. Recently, an important fea- ture in sepiapterin reductase has been discovered. It was reported that it has a domain which can bind with pterin. Anti-pterin binding site antibodies have demonstrated the structural relatedness be- tween the pterin site in sepiapterin reductase, dihydrofolate reductase and the aromatic amino acid hydroxylase, providing strong evidence for specific structures [31]. Citron et al. has reported that a pentapeptide sequence (Ala-Gly-Leu-Leu- Ser) exists in all five types of hydroxylase (human, rat, quail and rabbit), in sepiapterin reducatase and (partially) in aldose reductase [32]. They suggested that this sequence might be specific structure of pterin binding domain. Pterin binding enzymes in insects may have the same domain. Monoclonal antibodies against silkworm sepiapter- in reductase might be useful to clarify the property of the enzyme. Further studies on the enzyme are — now being undertaken and will be published else- where. ACKNOWLEDGMENTS We thank Dr. W. L. Gyure of Seton Hall University for his kind gift of erythropterin and for his correction of English. REFERENCES 1 Matsubara, M., Katoh, S., Akino, M. and Kauf- man, S., (1966) Sepiapterin reductase. Biochim. Biophys. Acta, 122: 202-212. 2 Kaufman, S., (1967) Metabolism of the phenylala- nine hydroxylation cofactor. J. Biol. Chem., 242: 3934-3943. 3. Nagai (Matsubara), M. (1968) Studies on sepiapter- in reductase; Further characterization of the reac- tion product. Arch. Biochem. Biophys., 126: 426- 435. 4 Katoh, S and Sueoka, T. (1987) Isomerization of 6-lactoyl tetrahydropterin by sepiapterin reductase. J. Biochem. (Tokyo), 101: 275-278. 5 Katoh, S and Sueoka, T. (1988) Coenzyme stimula- tion of isomerase activity of sepiapterin reductase in 10 iL 12 13 14 15 16 17 18 Purification and Characterization of Silkworm Sepiapterin Reductase the biosynthesis of tetrahydrobiopterin. J. Biochem., (Tokyo) 103: 286-289. Sueoka, T., Hikita, H and Katoh, S. (1990) Best-fit analysis of kinetic scheme for the stepwise reduction of the “diketo” group of 6-pyruvoyl tetrahydropterin by sepiapterin reductase. In “Enzymology and Molecular Biology of Carbonyl Metabolism III” Ed. by H. Weiner, et al., Plenum Press, New York. pp. 229-239. Kaufman, S. (1971) In Advances in Enzymology Ed. by A. Meister, Vol. 35, Interscience Publishers, New York, pp. 245-319. Tietz, A., Lindberg, M. and Kennedy, E. P. (1964) A new pteridine-requiring enzyme system for the oxidation of glyceryl ethers. J. Biol. Chem., 2339: 4081-4090. Tayeh M. A. and Marletta M. A. (1989) Mac- rophage oxidation of L-arginine to nitric oxide, nitrite, and nitrate. Tetrahydrobiopterin is required as a cofactor. J. Biol. Chem., 264: 19654-19658. Matsubara, M., Tsusué, M. and Akino, M. (1963) Occurrence of two different enzymes in the silk- worm, Bombyx mori, to reduce folate and sepiapter- in. Nature, 199: 908-909. Katoh, S. (1971) Sepiapterin reductase from horse liver; Purification and properties of the enzyme. Arch. Biochem. Biophys., 146: 202-214. Sueoka, T and Katoh, S. (1982) Purification and characterization of sepiapterin reductase from rat erythrocytes. Biochim. Biophys. Acta, 717: 265- Die. Katoh, S., Sueoka, T and Yamada, S. (1983) In “Chemistry and Biology of Pteridines”. Ed. by J. A. Blair, Walter de Gruyter, Berlin, pp. 789-793. Sugiura, K., Takikawa, S., Tsusué, M and Goto, M. (1973) Isolation and characterization of a yellow pteridine from Drosophila melanogaster mutant sepia. Bull. Chem. Soc. Jap., 46: 3312-3313. Tsusué, M. (1971) Studies on sepiapterin deaminase from silkworm, Bombyx mori. Purification and some properties of the enzyme. J. Biochem., (Tokyo), 69: 781-788. Katoh, S and Sueoka, T. (1984) Sepiapterin reduc- tase exhibits a NADPH-dependent dicarbonyl re- ductase activity. Biochem. Biophys. Res. Commun., 118: 859-866. Tsusué, M and Akino, M. (1965) Yellow pterin in mutant /emon of silkworm and mutant sepia of Drosophila melanogaster. Zool. Mag., 74: 91-94. Andrews, P. (1964) Estimation of the molecular weight of protein by sephadex gelfiltration. Biochem. J., 91: 222-223. 19 20 vA 22 7B) 24 25 26 27 28 2S) 30 Sil 32 125 Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacter- iophage T,. Nature, 227: 680-685. Katoh, S and Sueoka, T. (1982) Pteridine- metabolizing enzymes of Macaca _fascicularis. Comp. Biochem. Physiol., 71B: 33-39. Lineweaver, H and Burk, D. (1934) The determina- tion of enzyme dissociation constants. J. Am. Chem. Soc., 56: 658-666. Smith, G. Y. (1987) On the role of sepiapterin reductase in the biosynthesis of tetrahydropterin. Arch. Biochem. Biophys., 255: 254-266. Fan, C. L. and Brown, G. M. (1979) Partial purification and some properties of biopterin synth- ase and dihydropterin oxidase from Drosophila melanogaster. Biochem. Genet., 17: 351-369. Sueoka, T and Katoh, S. (1985) Carbonyl reductase activity of sepiapterin reductase from rat erythro- cytes. Biochim. Biophys. Acta, 843: 193-198. Katoh, S., Sueoka, T and Yamada, S. (1982) Direct inhibition of brain sepiapterin reductase by a catecholamine and an indoleamine. Biochem. Bio- phys. Res. Commun., 105: 75-81. Wermuth, B. (1981) Purification and properties of an NADPH-dependent carbonyl reductase from hu- man brain. Relationship to prostaglandin 9- ketoreductase and xenobiotic ketone reductase. J. Biol. Chem., 256: 1206-1213. Katoh, S., Nagai, M., Nagai, Y., Fukushima, T and Akino, M (1970) In “Chemistry and Biology of Pteridines” Ed. by K. Iwai, M. Akino, M. Goto and Y. Iwanami, Int. Acad. Print. Co., Tokyo, pp 225- 234. Kaufman, S. (1976) In “Advances in Neurochemis- try” VolIf Ed. by B. W. Agranoff and M. H. Aprisin, Plenum Press, New York, pp 1-132. Danks, D. M., Bartholomé, K., Clayton, B. E., Curtius, H-Ch., Grobe, H., Kaufman, S., Leeming, R., Pfleiderer, W., Rembold, H. and Rey, F. (1978) Malignant hyperphenylalaninaemia-current status (June 1977). J. Inher. Metab. Dis., 1: 49-53. Kaufman, S. (1985) | Hyperphenylalaninaemia caused by defects in biopterin metabolism. J. Inher. Metab. Dis., 8: 20-27. Jennings, I. and Cotton, R. (1990) Structural simi- larities among enzyme pterin binding sites as de- monstrated by a monoclonal anti-idiotypic antibody. J. Biol. Chem., 265: 1885-1889. Citron, B. A., Milstien, S., Gutierrez, J. C., Levine, R. A., Yanak, B. L and Kaufman, S. (1990) Isola- tion and expression of rat liver sepiapterin reductase cDNA. Proc. Natl. Acad. Sci. USA. 87: 6436-6440. Raed ae : n - Wea 5) ar: a ie ‘ 7 &) 4 oe es ere, ah as Pte ; . els vo, ey yi ee . hf batts A > ix See ae OSEL pega a ee “data bu, ye ees ae u Kf 24 venloe ee Aine ® gy Ssh ts Sig 4 one as 5m Oey ni spite Ve Pale l S| am ; : veto Dano 259 2 < ni 2 Bee Hiei a eh “4 Pos L aes my, eee AE nes bs J 5 mad ES t aie ps re er aint hit ‘nie ' - ‘ . J / ul j Yves, % “Y e208 1 eb tthe gy Pam cunt Pigg . 7 thts nehiee trees! settee rad iek, eps, c ‘ : } aie F re < te , (Rigas BY AS £ . i. : Cc be 7 ¥ > y Ly BAS te r * Z mt Ls A Seria sth 40 y + ? x } ree. Pyke f: , il y ( al : i a ttin ma rain 9 Fae ’ : } SS (75! = . “ys id “i 4 i i . Ff i : i hid + a 5 es ; v4 Lae § f H mS - € i ¢ . - 4 : I Re f ate @ PED OPO Dab ay Mee a2 hae ce = t a yi 7 ~ r a - ’ Bus pies * F = oye © ~ r 5 cs : = was Ms S72 R eats ~ af ¢ 3 fe ss ' , 4 * y i ma = =P a ; He x it = es 5 a ’ } ¢ "7 ‘ A , ep in VE ‘ ' j . a3 a dad Te t , a 3 3 q =F y ~ ae rt * i é 4 ¥ rie k ae a 4 3 5 aie pa! i" 1 j ii ag } 2 FF a y ‘ : 4 5 ahh ; £55 bal u x) ac 4 a | 7 Bos : 7 ae f3;) a, | ri¢ = ‘ * 4 a7 = c ¥ “ L ‘ = - aa ree - t * - Med Hy cle i Fe ' oy + Qe Ws i 7 Py pee i - ys . ¥ eh 3 d 4 A Si \ is i " ’ 2 es f os + a ca ; 7 « " ‘ cS , ah “ hy ( 7 = ata £ = 4 4s bs L re ) ie a ee str ss t Lee ii 4 it, F te vA ee } F 4 + i : iv! es ey . 4 ‘ ie i ; he hts fr 4 ZOOLOGICAL SCIENCE 9: 127-132 (1992) Developmental Changes of Glutamate Decarboxylase and 2’,3'-Cyclic Nucleotide 3’-Phosphodiesterase in the Organotypic Culture of Newborn Mouse Cerebellum DAISAKU SATOMI Department of Biology, College of Arts and Sciences, The University of Tokyo, Tokyo 153, Japan ABSTRACT—The quantitative changes of glutamate decarboxylase (GAD) and 2’,3’-cyclic nucleotide 3’-phosphodiesterase (CNP) during early developmental stages in the organotypic culture of newborn mouse cerebellum were examined by using the high-performance liquid chromatograph technique. Cerebellar explants were incubated for 4 weeks under standard conditions. The activity of GAD linearly increased from 20.2+12.1 nmol/mg protein/h at 2 days in vitro (2 DIV) to 111.7+15.0 nmol/ mg protein/h at 22 DIV. In the case of CNP, the activity was ketp low until 8 DIV and then increased rapidly from 16.9+4.8 ~mol/mg protein/h (8 DIV) to 62.3+12.6 ~mol/mg protein/h (15 DIV). In both enzymes, the changing patterns of activities were similar to those observed in vivo. In the presence of cytosine arabinoside, an inhibitor of DNA synthesis, GAD activity increased, whereas the levels of CNP activity were low during incubation. Morphologically, ependymal formation and myelination were © 1992 Zoological Society of Japan not observed in cytosine arabinoside-treated explants. INTRODUCTION Cerebellar organotypic culture has been used as a useful model for studying the development of central nervous system [1]. We have used this culture system for biochemical study. In previous papers [2-4], we reported that biochemical changes of neurotransmitters, such as y-aminobu- tyric acid (GABA) and myelin-characteristic galactolipids are closely related to morphological changes such as synapse formation and myelin formation. We also supposed that the increase of GABA content during the first 2 weeks of incuba- tion depends on the increase of activity of gluta- mate decarboxylase (GAD, EC 4. 1. 1. 15) in culture. Nagata et al. [5] reported the cellular localiza- tion of GAD during development of rat brain. In purified neuronal fraction GAD activity increased rapidly during the first month of postnatal life, whereas, in purified glial fraction the activity was low and increased only slightly during develop- ment. 2’,3’-cyclic nucleotide 3’-phosphodiesterase Accepted October 4, 1991 Received April 25, 1991 (CNP, EC 3. 1. 4. 37) has been used as a marker enzyme of myelin [6, 7]. Recently, Nussbaum ef al. [8] reported that the specific activity of CNP in culture of pure oligodendrocytes of newborn rat brain was much lower than that of myelin isolated from rat brain in vivo. GAD and CNP are good biochemical markers to investigate the develop- ment of central nervous system. However, there are still few reports concerning developmental changes of GAD and CNP activities in cerebellar explants [9]. Seil et al. [10] and Blank et al. [11] reported the effects of cytosine arabinoside on morphological development of cerebeller explants. In the pre- sence of cytosine arabinoside, Purkinje cells sur- vived but myelination did not occur. In this work, we examined developmental changes of GAD and CNP activities under standard conditions and in the presence of cytosine arabinoside. MATERIALS AND METHODS Mouse cerebellum culture The organotypic culture technique is based on that of Bornstein [12]. Cerebella of newborn mice 128 D. SATOMI (JCL :ICR strain) were sliced parasagitally into 7 pieces by hand and placed on collagen-coated coverslips (11x22 mm). Then, the coverslips were put into 15x150mm test tubues and 1 ml of feeding solution consisting 23% horse serum (GIBCO), 66% Eagle’s minimum essential medium with L-glutamine (GIBCO) and 0.6% glucose was added to each tube. The explants were incubated at 35°C and the feeding solution was changed every 7 days. The procedure was described in detail in the previous papers [2, 4]. Cytosine arabinoside treatment Cytosine arabinoside was dissolved in Earle’s balanced salt solution (GIBCO) and incorporated into the feeding solution at a concentration of 10 yvg/ml. Explants were incubated in the feeding solution from first day in vitro (1 DIV). After appropriate periods, cytosine arabinoside-treated and control explants were examined with a light microscope (Nikon, Diaphot-TMD) under bright- field. Then, explants were homogenized and used as enzyme preparations. Enzyme assays GAD activity was assayed by determining the amount of GABA formed from L-glutamate in the reaction mixture according to a modification of Chude & Wu [13]. The reaction was started by mixing 0.02 ml of 200 mM L-glutamate in 50 mM potassium phosphate buffer (pH 7.2) containing 0.2 mM pyridoxal phosphate and 0.18 ml of en- zyme solution (explant homogenate, about 0.2 mg protein) in 50mM _ potassium phosphate buffer (pH 7.2) containing 0.2 mM pyridoxal phosphate and 1 mM 2-aminoethyl isothiouronium bromide. The reaction was performed for 30 min at 37 C and terminated by adding 0.75 ml of cold ethanol and 0.75 ml of cold chloroform and then 250 nmol of cysteic acid in 0.05 ml of water was added as internal standard. The reaction mixture was stir- red well and centrifuged. The layers were sepa- rated. All amino acids including GABA and cysteic acid were recovered in the upper ethanolic water layer. Aliquotes of the upper layer were transferred to other test tubes and dinitropheny- lated. Dinitrophenylated samples were analyzed with the high-performance liquid chromatograph technique (HPLC) as described previously [4]. The assay of CNP was based principally on the method of Kurihara and Tsukada [6]. The reaction mixture consisted of 0.1 ml of enzyme solution (explant homogenate, about 10 yg protein), 0.1 ml of 0.2M Na,HPO,-0.1M citric acid buffer (pH 6.2) and 0.1 ml of 0.5% Triton X-100. The mix- ture was preincubated for 5 min at 37°C and then the reaction was started by adding 0.1 ml of 60 mM sodium adenosine 2’,3’-cyclic monophosphate. The reaction was performed for 20 min at 37 C and terminated by adding 0.53 ml of cold methanol and 1.07 ml of cold chloroform. After mixing, the tube was centrifuged and 2 wl of the upper layer was applied to HPLC. The chromatographic column (« Bondapack Cig, Waters) was equilibrated with 12% methanol in 16mM KH>PO, solution and isocratic elution with the same solvent was con- tinued for 10 min after the sample was injected. The flow rate was maintained at 1.0 ml/min and column temperature was ambient. The effluent was monitored by measuring the absorbance at 254 nm. The peak areas were measured with a data reduction system (Shimadzu, Chromatopac C- RIB). For comparison, the activities of GAD and CNP of cerebella from littermates were determined with the same procedures as in the case of explants. Protein determination The amounts of protein and collagen were deter- mined by the method of Lowry et al. [14] with a bovine serum albumin standard and by the method of Woessner [15] with a collagen standard, respec- tively. The protein content of the explants was calculated by subtracting the amount of collagen from the total protein value. RESULTS Changes of GAD and CNP activities during de- velopment The quantitative changes of GAD activity dur- ing early developmental stages in the explatns under standard conditions are shown in Figure 1. The activity per protein increased linearly from 2 DIV to 22 DIV. In order to compare these values Changes of GAD and CNP in Culture 129 —~ 150 ve ic 7) 9 6 100 ey E 2 e590 rat <= O 0 2 8 15 22 30 TIME( day, DIV) Fic. 1. The changes of GAD activity in organotypic culture of newborn mouse cerebellum and cerebel- lum in vivo during early development. The ex- plants cultured under standard conditions were homogenized and used as enzyme preparations. GAD activity was determined by HPLC as de- scribed in “MATERIALS AND METHODS”. For measuring GAD activity in vivo, the cerebella from littermates were used. Standard deviations were calculated from three experiments. ©, GAD activity in explants; A, GAD activity in cerebellum in Vivo. with those of cerebellum in vivo, cerebella from littermates were similarly analyzed. As a result, it was found that the changing pattern of GAD activity in explants was in good agreement with that of cerebellum in vivo (Fig. 1). Figure 2 shows CNP activity in the same sample used for the determination of GAD activity in Figure 1. In contrast to the changes of GAD activity, the levels of CNP activity remained low during first week of incubation and then increased rapidly between 8 DIV and 15 DIV. Comparing the change of CNP levels in explants with that in cerebellum in vivo, the activity in 8 DIV explants was comparable to that in vivo. However, the levels of 15 DIV and 30 DIV were approximately one-half and one-third of those in vivo, respec- tively. Effects of cytosine arabinoside Cerebellar explants were exposed to cytosine arabinoside from 1 DIV and compared morpholo- gically and biochemically with control explants. n/n) NO oO {2) CNP( pmol/mg protei >) >) 2 8 15 22 30 TIME( day, DIV) Fic. 2. The changes of CNP activity in organotypic culture of newborn mouse cerebellum and cerebel- lum in vivo during early development. The ex- plants cultured under standard condititioins were homogenized and used as enzyme preparations. CNP activity was determined by HPLC as described in “MATERIALS AND METHODS”. For measuring CNP activity in vivo, the cerebella from littermates were used. Standard deviations were calculated from three experiments. ©, CNP activ- ity in explants; A, CNP activity in cerebellum in vivo. Figure 3 shows photomicrographs of mouse cere- bellum culture (bright-field, unstained living cul- ture). In control explants, ependymal formation, which was readily recoganized by its ciliary move- ment, and myelin formation were observed after 2-3 weeks of incubation (Fig. 3-a, b). In cytosine arabinoside-treated explants, grouped large cells, approximately 25 um in diameter, were character- istically observed (Fig. 3-c), but ependymal forma- tion and myelin formation were not observed. In cytosine arabinoside-treated explants, GAD activ- ity increased, whereas CNP activity did not in- crease during incubation period (Table 1). In comparison with the explants treated with cytosine arabinoside from 1 DIV, the explants, which were incubated under standard conditions during first 2 weeks and then exposed to cytosine arabinoside, were examined. In this case, myelina- tion was detected (Fig. 3-d) and the activities of GAD and CNP were maintained at nearly the same levels as controls even at 22 DIV (data not shown). 130 D. SATOMI ation (irregular shaped cavity) in control expalnt at 15 DIV, within which separated cells (arrows) were whirled about by the activity of the ependymal cilia. (b) myelinated axons in control explant at 22 DIV. (c) grouped large cells (probably Purkinje cells) in 15 DIV explant treated with cytosine arabinoside from 1 DIV. (d) myelinated axons in 22 DIV explant treated with cytosine arabinoside from 15 DIV. Scale bars represent 50 ~m. nervous system, because synapse formation and myelin formation take place in vitro. In the case of We have used organotypic culture of newborn aggregated cell culture, Seeds [16] reported de- mouse cerebellum as a useful model of central velopmental changes of GAD activity, which in- DISCUSSION Changes of GAD and CNP in Culture 131 TABLE 1. cerebellar explants. Effects of cytosine arabinoside on the activities of GAD and CNP in Explants were incubated in the presence of cytosine arabinoside from 1 DIV. At various developmental stages (8-22 DIV), explants were homogenized and activities of GAD and CNP were assayed as described in “MATERIALS AND METHODS”. The standard deviations were calculated from three experiments. GAD activity CNP activity Eaplauts (nmol/mg protein/h) (umol/mg protein/h) 8 DIV SSE) 5.3+0.4 15 DIV 53.6+ 16.0 4.4+1.9 22 DIV 105.8 + 66.3 4.9+2.0 creased during 3 weeks of incubation and attained a level of one-third of adult mouse brain in vivo at 21 DIV (30 nmol/mg protein/h). In the present work, it was found that the activities of GAD in organotypic culture under standard conditions change in the same manner as those in vivo. As supposed in a previous paper [4], the increase of GAD activity in explants is in good agreement with the increase of GABA synthesis from [*C]glucose. It was also found that CNP activity in explants shows a similar developmental pattern to that seen in vivo and that the period of rapid increase of CNP activity corresponds to the period of rapid myelin formation observed _light- microscopically. The developmental pattern of CNP activity in vivo shown in Figure 2 is in good agreement with those reported by Sprinkle et al. [17] and Mikoshiba et al. [18]. It is important to point out further that the developmental pattern of CNP activities in explants is similar to those of cerebrosides and sulfatides reported previously [2]. This indicates that CNP as well as the galacto- lipids is a good myelin marker in vitro. Blank e¢ al. [11] described the ultrastructural alterations that occurred in the cerebellar explants exposed to cytosine arabinoside from 1 DIV to 5 DIV. In their study, Purkinje cells, Golgi cells and a few basket and stellate cells survived, but granule cells degenerated. Astrocytes and oligodendro- cytes were diminished in number. In the present study, we have observed the absence of ependymal formation and myelination in the explants exposed to cytosine arabinoside from 1 DIV. Small cells such as the granule cells were lost, and conse- quently, large cells were observed characteristical- ly as groups (Fig. 3-c). The finding that the GAD activity increased in cytosine arabinoside-treated explants implys that the cells which have the ability to synthesize GABA can differentiate to a certain degree even in the presence of the inhibitor. It would therefore be reasonable to consider that the living large cells may be Purkinje cells. On the other hand, the levels of CNP activity were low during incubation with the inhibitor, due highly probably to the prevention of normal development and differentiation of oligodendrocytes, myelin- forming cells, by cytosine arabinoside. The fact that the explants that were exposed to the inhibitor after a 2-week incubation under standard condi- tions exhibited normal developmental patterns in both the morphological and biochemical aspects appears to indicate that most of oligodendrocytes differentiate well within 2 weeks in vitro. Recent- ly, Jordan et al. [19, 20] have repored that epen- dymal cells, in primary tissue culture derived from embryonic rat cerebral cortex, proliferate in the Same manner as seen in vivo. The lack of epen- dymal formation in cytosine arabinoside-treated expltans would indicate that these cells are still immature at the time of explantation and are not able to differentiate in the presence of the in- hibitor. The present work has thus demonstrated that the developmental changes of GAD and CNP in the organotypic culture of newborn mouse cerebel- lum follow the respective patterns in vivo, as a reflection of the differentiation of cells. This culture system should facilitate the clarification of the cellular and molecular aspects of interactions between neurons and glia cells during the early development of the central nervous system in vivo. 10 132 REFERENCES Seil, F. J. (1979) Cerebellum in tissue culture. In “Reviews of Neuroscience”. Vol. 4, Ed. by D. M. Schneider, Raven Press, New York, pp. 105-177. Satomi, D. and Kishimoto, Y. (1981) Change of galactolipids and metabolism of fatty acids in the organotypic culture of myelinating mouse brain. Biochim. Biophys. Acta, 666: 446-454. Satomi, D. (1983) Uptake and metabolism of cho- line in the organotypic culture of newborn mouse cerebellum. J. Biochem., 94: 785-791. Satomi, D. (1986) Developmental changes of y- aminobutyric acid (GABA) and putative amino acid neurotransmitters in the organotypic culture of new- born mouse cerebellum. J. Biochem., 100: 285-292. Nagata, Y., Nanba T. and Ando, M. (1976) Changes in some enzymic activity of separated neuronal and glial cell-enriched fractions from rat brain during development. Neurochem. Res., 1: 299-312. Kurihara, T. and Tsukada, Y. (1967) The regional and subcellular distribution of 2’ ,3’-cyclic nucleotide 3’-phosphohydrolase in the central nervous system. J. Neurochem., 14: 1167-1174. Sprinkle, T. J. (1989) 2’,3’-cyclic nucleotide 3’- phosphodiesterase, an oligodendrocyte-Schwann cell and myelin-associated enzyme of the nervous system. CRC Crit. Rev. Neurobiol., 4: 235-301. Nussbaum, J. L., Espinosa de los Monteros, A., Pari, F. M., Doerr-Schott, J., Roussel, G. and Neskovic, N. M. (1988) A morphological and biochemical study of the myelin-like membrane structures formed in the cultures of pure oli- godendrocytes. Int. J. Dev. Neurosci., 6: 395-408. Bradbury, K. and Lumsden, C. E. (1979) The chemical composition of myelin in organ cultures of rat cerebellum. J. Neurochem., 32: 145-154. Seil, F. J., Lerman, A. L. and Woodward, W. R. (1980) Cytosine arabinoside effects on developing cerebellum in tissue culture. Brain Res., 186: 393- 408. D. SATOMI 11 12 3) 14 15 16 17 18 19 20 Blank, N. K., Seil, F. J. and Herndon, R. M. (1982) An ultrastructural study of cortical remodeling in cytosine arabinoside induced granuloprival cerebel- lum in tissue culture. Neuroscience, 7: 1509-1531. Bornstein, M. B. (1973) Organotypic mammalian central and peripheral nerve tissue. In “Tissue Cul- ture” Ed. by P. F. Kruse, Jr. and M. K. Patterson, Jr., Academic Press, New York, pp. 86-92. Chude, O. and Wu, J.-Y. (1976) A rapid method for assaying enzymes whose substrates and products differ by charge. application to brain L-glutamate decarboxylase. J. Neurochem., 27: 83-86. 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. Woessner, J. F., Jr. (1961) The determination of hydroxyproline in tissue and protein samples con- taining small proportions of this imino acid. Arch. Biochem. Biophys., 93: 440-447. Seeds, N. W. (1971) Biochemical differentiation in reaggregating brain cell culture. Proc. Natl. Acad. Sci. USA, 68: 1858-1861. Sprinkle, T. J., Zaruba, M. E. and McKhann, G. M. (1978) Activity of 2’,3’-cyclic nucleotide 3’-phos- phodiesterase in regions of rat brain during develop- ment: quantitative relationship to myelin basic pro- tein. J. Neurochem., 30: 309-314. Mikoshiba, K., Nagaike, K. and Tsukada, Y. (1980) Subcellular distribution and developmental change . of 2’,3’-cyclic nucleotide 3’-phosphohydrolase in the central nervous system of the myelin-deficient shiverer mutant mice. J. Neurochem., 35: 465-470. Jordan, F. L., Rieke, G. K. and Thomas, W. E. (1987) Presence and development of ependymal cells in primary tissue cultures derived from embryonic rat cerebral cortex. Dev. Brain Res., 35: 97-110. Jordan, F. L., Rieke, G. K., Hughes, B. W. and Thomas, W. E. (1990) Morphological diversity of ependymal cells in tissue culture. Brain Res. Bull., 25: 159-163. ZOOLOGICAL SCIENCE 9: 133-141 (1992) Egg Maturation in Laboratory-reared Females of the Swallowtail Butterfly, Papilio xuthus L. (Lepidoptera: Papilionidae), Feeding on Different Concentration Solutions of Sugar MAMORU WATANABE Department of Biology, Faculty of Education, Mie University Tsu-shi, Mie 514, Japan ABSTRACT—In the laboratory, adult females of the swallowtail butterfly, Papilio xuthus, were given sugar solutions of different concentrations. Weight loss of the unfed females was directly proportional to time since emergence. They died within 8 days. Females given 0%, 0.1% or 1% solution of sugar lost weight with ageing. Females feeding on a 10% solution of sugar maintained weight for 15 days after emergence. Conversely, females given 20% or 50% sugar solutions were found to make a gain of 20% or 60%, respectively, in weight during the same period. Newly emerged females took little quantity of 0%, 0.1% or 1% solution of sugar, but their intake increased with ageing. On the other hand, they took over 100 mg of 10%, 20% or 50% solution of sugar. The daily changes to mature eggs in the females were compared among those treated with different foods. Almost no mature eggs were added in the females on feeding less than 1% solutions of sugar. When females took more than 10% solutions of sugar, they produced significantly more mature eggs than those unfed or those who took only water. In older adults, there was a relation between the added number of mature eggs and the accumulated sugar intake. Females may allocate their available energy (fat body+sugars) for egg maturation and body © 1992 Zoological Society of Japan maintenace. INTRODUCTION Nectar intake has been shown to increase longevity and egg production in many lepidopteran insects [1-2]. Although the sugar concentration of the nectar in flowers partly depends upon weather conditions, it seems to be maintained within a range between 15% and 45% [3-4]. Three sugars (monosaccharide fructose, glucose and disacchar- ide sucrose) are generally found in the nectar of many plant species [5-6]. Their sugar concentra- tion and composition vary not only among plant species but also with the age of a single flower [4, 6]. The nectar also includes a little amino acids [7]. Sulfurs, Colias alexandra and C. meadii, prefer nectar with low sugar concentration, chiefly those including monosaccharides [5]. The energy intake efficiency for some butterfly species is maximised at about 40% sucrose solution [8-10]. In some moths, relationship between the concentration of Accepted October 14, 1991 Received June 12, 1991 sugar and fecundity or longevity has been studied [11]. However, there are few reports in butterflies on the feeding regimes related to fecundity and survival depending upon the concentration level of sugar in nectar [12]. The achievement of full reproductive potential during the adult stage may be affected by the level of sugar concentration in the nectar. Therefore, the nectar feeding habit may have come to be evolved in the lifetime reproductive strategy, as well as mating behavior. The longevity of the swallowtail butterfly, Papi- lio xuthus, has been studied under natural condi- tions [13], but no information exists on the effects of adult nutrition. Multiple matings in P. xuthus increase fecundity [14, 15]. In this paper, I present data from laboratory-reared virgin females of P. xuthus subjected to 5 different sugar solution con- centrations as a diet and compare their potential fecundity. MATERIALS AND METHODS P. xuthus in this study was collected principally 134 M. WATANABE in Mie Prefecture, in the warm-temperate zone of Japan. The females were obtained from a stock culture reared in the laboratory at room tempera- tures and under a long-day photoperiod (more than 15 hr of light) in the summers of 1986, 1988, 1989 and 1990. As soon as the females emerged, their forewing lengths and fresh weights were measured. Then, without mating, each female was assigned to one of seven groups. (1) Females were kept unfed; (2) Females were fed with distilled water: (3) Females were fed with a 0.1% solution of sugar; (4) Females were fed with a 1% solution of sugar; (5) Females were fed with a 10% solution of sugar; (6) Females were fed with a 20% solution of sugar; (7) Females were fed with a 50% solution of sugar. The sugar solution included equal amounts (weight) of fructose, glucose and sucrose. Except unfed females, they were supplied with sugar solution or water for only 3 min a day. Each female was kept in a glassine envelope and stored in a chamber at ca. 25°C. The amount of intake was determined as the odds of the weight between pre- and post-feeding, using a semi-microbalance accurate up to 0.1 mg. On death, females from each of the 7 groups had their abdomens immediately fixed in 50% ethyl alcohol, though most of them were able to survive over 15 days after emergence. Following dissec- TABLE 1. Group Forewing length (mm) (1) unfed 55.8+0.4 (n=21) (2) water feeding (3) 0.1% sugar solution (4) 1% sugar solution (5) 10% sugar solution (6) 20% sugar solution (7) 50% sugar solution n: number of females examined 56.7+40.9 (n=11) 55.9+0.8 (n=15) 56.3+0.7 (n=16) 55.8+0.5 (n=19) 56.4+40.5 (n=18) 56.4+0.7 (n=15) tion, eggs in ovaries were classified into three Stages. A detailed description of the egg at each stage is given elsewhere [14]. The numbers of mature and submature eggs were counted directly. The total number of immature eggs was estimated by multiplying the number in one ovariole by eight, that is the number of ovarioles. The dia- meter of mature eggs was also measured under a dissecting microscope. All means are shown with their standard errors. RESULTS Body size and daily changes in relative weight The total number of females used for this study was 122. It is commonly observed that the size of adult butterflies, as reflected by the length of the forewing, tends to be smaller in laboratory-reared individuals than that of adults from natural popula- tions. Such tendency was also found in P. xuthus females [15]. Accordingly, to minimize the effect of small body size, only females with a forewing length of more than 50 mm were used. As shown in Table 1, the mean forewing length in each group | was around 56mm. Consequently, they were not significantly different from the forewing length of females collected from natural populations [14]. After emergence, most females evacuated ex- crement during wing extension. Although fore- wing lengths were mostly stable (Table 1), the fresh weights of 0-day-old females were very vari- able, possibly because some females held onto excrement until they were weighed. The abdominal cavity of the females just after Forewing lengths and fresh weights of 0-day-old females used in the experiment of 7 groups Fresh weight (mg) 690.5+26.5 (n=19) 712.3+30.6 (n=11) 624.1+14.9 (n=15) 616.6+17.0 (n=16) 631.7+26.9 (n=20) 648.9+18.7 (n=19) 740.6+37.2 (n=15) Egg Maturation in Papilio xuthus 135 emergence was mostly filled with fat bodies and with relatively small air sac. Since the fat body was used for the eggs developed, the fat body gradually depleted and the volume of air sac increased with ageing so that it came to fill half the abdominal cavity in 10-day-old females. Unfed females lost weight in proportion to length of time after emergence. All died within 8 days (Fig. 1). The fresh weight of 8-day-old Weight Relative WO )ES SE RTS i see yen HS ya 3 Se eA) females just after death was only 43% of that of 0-day-old ones. Mean daily weight loss was calcu- lated at about 50 mg. The weight of each female supplied with water or sugar solution was based on the weight of pre-feeding. Females fed on distilled water also became lighter as they age (Fig. 1). There are no difference in relative weight between unfed and the water feeding females in the first 3 days. al a Awa Sy ees) Days after emergence Fic. 1. Change in the relative weight of unfed females (cross) and females feeding on water (solid circle), on sugar solutions of 0.1% (open square), 1% (open triangle), 10% (open circle), 20% (solid triangle) and 50% (solid square). 136 M. WATANABE However, the weight loss was lower than that in unfed females thereafter. Some of them survived 15 days after emergence, when the mean fresh weight was about 44% of that in 0-day-old females. Mean daily weight loss was calculated at about 27 mg. Therefore, 23 mg of distilled water was daily retained in the female body. Females fed on 0.1% or 1% solution of sugar became lighter as they age. Although the decline in their weight was similar to that in females fed 200 100 20 a SL s@ba Q 10 solution Sugar Intake — Oa distilled water up to 10-day-old, all were able to survive 15 days after emergence. Their weight was more than half of that in females just after emerg- ence. Therefore, it can be said that such sugars, whilst not maintaining the female’s weight, contri- buted to their longevities. Females fed on 10% solution of sugar kept their weight constant throughout the experimental period. Naturally, all were able to survive after the end of the experiment, 15 days after emergence. Zo —l ea . CS Se ee ae a _| OU 2 a3 ad ane, 778, 9) AOn Wl tee: siesr ees DaVS ares Smergence Fic. 2. Change in the intake of water (solid circle), sugar solutions of 0.1% (open square) and 1% (open triangle), 10% (open circle), 20% (solid triangle) and 50% (solid square) [mg, +SE]. Egg Maturation in Papilio xuthus 137 Such sugars are effective as an energy resource for maintaining body weight. Females fed on 20% or 50% solution of sugar became heavier as they age. ‘Their respective weights after 15 days was 1.2 and 1.6 times heavier than those of females just after emergence. Con- sequently, high sugar concentration enhances longevity and body weight in female adults, though crystals of sugar were found in the hindgut, some- times in the midgut of the females feeding on 50% solution of sugar. Amount of daily sugar intake Newly emerged females (=0-day-old) took little water (Fig. 2). Mean water intake by females was about 0.1 mg. Besides water, they also took a little quantity of 0.1% (ca. 1.5 mg) or 1% (ca. 4 mg) solution of sugar. The mean sugar intake by females in the 0.1% or 1% group was 0.001 mg and 0.04 mg, respectively. The water intake by 1-day-old females increased to 2mg. Females in the 0.1% and 1% group also took more sugar solution than 0-day-old ones. The water intake by females increased as they age and ultimately reaching about 40 mg. Although intake of 0.1% or 1% solution of sugar by females was somewhat higher than that of water in younger stages, daily changes in intake of such sugar solutions were similar to those in intake of water. This means that females only took 0.04mg and 0.4 mg of sugar daily from 0.1% and 1% solution, respectively. Therefore, at least the feeding response by older females to sugar solutions did not differ from that to water. Females may be unable to distinguish the sugar solution of less than 1% from water. Females of 0-day-old took about 140 mg of 10% solution of sugar, or 14mg of sugar (Fig. 2). Although the 20% sugar solution was taken in large quantity (ca. 160 mg), the intake by O-day- old females was almost the same for 10%, 20% and 50% sugar solutions. Also, daily intake by females of more than 10% sugar solutions gradual- ly decreased as they age. 14-day-old females took less than 100mg. The respective sugar intakes were calculated as ca. 8 mg, 10 mg and 25 mg for the females took 10%, 20% and 50% solutions. Daily changes in the number of eggs The diameter of mature eggs in the oviducts was about 1.27mm (min 1.183mm and max 1.400 mm). It remained constant irrespective of female age and diet. In the present study, the number of eggs in 0-day-old females was 0.3+0.3 (n=4), 52.5+1.7 (n=4) and 706.3 +34.1 (n=4) for mature, subma- ture and immature eggs, respectively. Since the unfed females or those supplied with only water were never nourished, the number of eggs in the two groups were averaged together. Table 2 shows that the number of mature eggs increased toward 20 in 3-day-old females. The number of submature eggs decreased at the 3rd day after emergence, and then the number did not increase. Therefore, few submature eggs seem to be added from immature eggs in the females. The number of immature eggs did not change throughout the tife span of adults. The daily changes in the number of eggs were compared among the females of different feeding treatments. If sugars contribute to egg maturation, the number of mature and submature eggs might be expected to increase. Although only small sample sizes were available during younger stages, almost no mature eggs were added to the females treated with 0.1% and 1% sugar solutions (Table 2). However, in females fed 10% sugar solution, the number of mature eggs was significantly higher than that of unfed females and those fed only water. Such a phenomenon was also observed among females treated with 20% and 50% sugar solutions. After 15 days, some retained more than 150 mature eggs. There seemed to be no changes in the number of subma- ture and immature eggs irrespective of sugar con- centrations (Table 2). Since many mature eggs were added from submature eggs, the females must make up for ‘the loss’ of submature eggs. Differences in sugar concentrations affects daily Sugar intake in each butterfly. If the amount of sugar was effective for females to produce excess eggs comparing to females given no sugar, the number of mature eggs would be related to the amount of accumulated sugar intake. During younger stages, there seemed to be no relationship between the number of mature eggs and the 138 M. WATANABE TABLE 2. Change in the number of eggs (three stages) in females feeding 0.1%, No. mature eggs No. submature eggs No. immature eggs Days after Unfed and 0.1% sugar emergence water feeding solution 0 03+ 0:3, *4) 1 Grose 3:0, (6) 17 (1) Z 15-0se 4:9 G) 40 (1) 3 182 O07 7) 4 (1) =) teOa 2:3" 1) 5 (1) 6~15 TScO== 7 ES (4) Jae Zee) 0 525s 1) — 1 40.8+ 9.6 (5) 42 (1) 2 O23 Ziel) 68 (1) 3 361022 860.6) 36 (1) 5 36:22:41) 35 (1) 6~15 34.3+ 12.8 (4) 3st 3e8 (7) 0 706.3+ 34.1 (4) — 1 8/2:4-=3 6919 16), 92 (1) D, 96302298185) Gia S06 (1) 3 1029.4+ 86.1 (5) 699 (1) 5 837.02 31816) asia (1) 6~15 737.3 129.1 =(4)y 494 FeaaRSSnlt Ch) (_ ): number of females examined = 005 P > 0:01 8 Ae TABLE 3. Regression coefficient (b) and determination coefficient (r*) in the log number of mature eggs on the log amount of accumulated sugar intake Days after emergence b r n t 1 0.140 0.25 fat 1.734 Mess 2 0.123 0.20 10 1.407 n.s. 3 OTS 0.72 10 4.530 P<0.01 5 0.416 0.96 12 15.692 P<0.01 10 0.358 0.32 10 18925 Ot Ps 0r05 15 0.749 O77 12 5.866 P0701 n: number of females examined accumulated sugar intake (Table 3). Egg matura- tion in younger females probably depended on the nutrition reserves carried over from the larva rather than foods taken after emergence. The fat body seems to be a major energy resource for the maturation of eggs and for maintenance of adult life. However, since the fat body decreased as they age, older females would depend more heavily on the sugar as an energy resource. Accordingly, it seems to be reasonable that the regression co- efficients (b) and the determination coefficients (r?) in the relation between the number of mature eggs and the amount of accumulated sugar intake by females increased. DISCUSSION In nature, the mean longevity of the adult swallowtail butterflies was estimated at 16.2 days for P. xuthus [13], 12.7 days for P. polytes [16] and Egg Maturation in Papilio xuthus 1%, 10%, 20% and 50% comparing with those in unfed and water feeding females 1% sugar 10% sugar 20% sugar 50% sugar solution solution solution solution 0 (1) 18.7+ 10.0 (3) ay ae 35°) 34.7+ 14.8 (3) 13 (1) eVae L7Y) (3) oooae 1S ©) Sse AVY (3) 1 (1) 44.3+ 19.9 (3) SOS 127455 16) yas “Ves =) 75 O05 @) SO.3ae ~ 30" —G) WP Vae WDD (3) OP) a AAS 2 (6) 0.8+ 0.6**(8) IWoSae 133 (G5) Die8a- 42h (4) SORSae, 1925-4 (4) WS (1) 78.0+ 24.6 (3) 44.3+ 21.9 (3) Sl) ae 22) (3) 12 (1) il ae» 1K0,7/ (3) Sse ZS (2) WEE ASS oe8) (3) Sil (1) 46.3+ 10.5 (3) Csp/ae 10,5) (3) 50.0+ 14.0 (2) AD.Sae OS" () Va isizems Te (3) S03sz, 103 (3) 652 05= 2057, (3) ISGae Gell Cs) Sie Sea lee 4) Q0sae SS)7/ (3) 84.00+ 6.9% (4) 793 (1) TSS 8120 (3) WeVMWae W340 (3) U2 ae DLO (3) 1160 (1) GQ 3a29 32.7 (B) Tse (S55) (2) 834.74 54.3 (3) 804 (1) 839.0+ 62.7 (3) 828.3+ 11.4 (3) SS0 Seem ules (2) 761.0+232.0 (2) 828.3+ 30.1 (3) S83 = (8823 (3) 883.7+ 56.0 (3) 5S. 4022), (8) 996.8 + 167.3 (4) Tease WEY (3) 764.8 + 123.0 (4) 139 16-17 days for P. helenus and P. protenor [17]. All of them relied on nectar of many flowering plants for their energy resource [18]. In this experiment, as I kept the females in the envelope all day except feeding time, so that each adult consumed the least amount of energy to maintain life. Unfed females survived for 8 days after emergence, after which no fat body was found in their body. This shows that the energy derived from the larval stage can support about 1 week of life in a motionless adult. Water intake enhanced the longevity of the checkerspot butterfly, Euphydryas editha [2]. As for P. xuthus, some females taking water survived for 15 days, and their daily weight loss was lower than that of unfed females after 4-day-old. There- fore, the water derived from the larval stage was likely to be effective during the first 3 days. The daily difference was about 28 mg, most of which would be attributable to water loss from the body. There were no significant differences in changes of body weight among females taking water and less than 1% solutions of sugar. Then, all the females in the groups survived for 15 days, suggest- ing that the sugar enhanced longevity. In order to release the cue for adult feeding, Murphy et al. [2] added a trace of sugar into the water as a diet. In the present study, however, the degree of the cue for feeding less than 1% sugar concentration was similar to that for taking water. Therefore, the trace sugar seemed to be important in maintaining the female body, though fat body was mostly exhausted at 15-day-old. Females feeding on a 10% sugar solution main- tained their body weight until 15-day-old. The amount of sugar intake seems to accord with the loss for respiratory metabolism and maintenance of body. On the other hand, females feeding a 20% and 50% sugar solutions increased their body weight. The fat body was found in the females of 15-day-old, suggesting that sugar played a role in substituting for the fat body. The fact that crystals of sugar were found in the gut of the females feeding on 50% solution of sugar suggested, however, that the 50% sugar solution was not a suitable concentration. Fat body is nutrition reservoir derived from the larval stage and is a resource for egg maturation 140 M. WATANABE during the adult stage [19]. This partly means that the number of mature eggs produced was deter- mined by the amount of fat body at emergence. In the present study, even unfed females were able to produce some mature eggs, probably using the energy derived from the fat body. Since the number of mature eggs were almost the same among 1-day-old females except those feeding on a 20% sugar solution, the energy resource for egg maturation in younger females was not sugar but either the fat body or other substances, both of which were derived from the larval stage. There was a rapid maturation of eggs in more than 2-day-old females feeding more than 10% sugar solutions, though some data were not signif- icantly different. Since unfed and water feeding females maintained a certain number of mature eggs with depletion of fat body, females must allocate their available energy (fat body-+sugars) for egg maturation and body maintenance. Furth- er oogenesis was found in females feeding on higher concentrations of sugar solutions. The available energy did not affect the size of mature egg in P. xuthus. In nature, the egg size did not change as they age [14]. Murphy et al. [2] pointed out that amino acid intake leads to heavier eggs, larvae from which are more likely to survive. However, Karlsson and Wiklund [20] showed that the egg size did not affect the survival rate of larvae. Watanabe [15] showed that multiple matings in P. xuthus increased the number of eggs deposited. This phenomenon suggested that a certain subst- ance derived from mating was used for egg matura- tion. On the other hand, flower nectar contains amino acids [4] and has been shown to be an important food resouce in lepidopteran insects [2, 7]. Therefore, the relative importance of these resources varies with the mating performance, though the primary adult food source is flower nectar [21]. The most effective sugar concentration was 40% in Thymelicus lineola [9|. In the present experi- ment, a 50% sugar solution was the most effective concentration for egg maturation. However, some Sugars were not absorbed and crystallized in the guts, probably affecting the digestive organ. It would be of interest to attempt to correlate the sugar concentration in nectar with the puddling behavior in swallowtail butterflies, though sodium was a stimulus for puddling behavior by P. glaucus [22]. ACKNOWLEDGMENTS I thank Dr. A. Kokubo for critical review of the manuscript, and anonymous referees for giving me useful comments. I am also grateful to Miss Y. Ueda and Mr. F. Todo for their assistance. This work was supported in part by Grant-in-Aid from the Ministry of Education, Science and Culture, Japan (No. 01540545). REFERENCES 1 Stern, V. M. and Smith, R. F. (1960) Factors affecting egg production and oviposition in popula- tions of Colias philodice eurytheme Boisduval (Lepi- doptera: Pieridae). Hilgardia, 29: 411-454. 2 Murphy, D. D., Launer, A. E. and Ehrlich, P. R. (1983) The role of adult feeding in egg production and population dynamics of the checkerspot but- terfly Euphydryas editha. Oecologia, 56: 257-263. 3. Roubik, D. W. and Buchmann, S. L. (1984) Nectar selection by Melipona and Apis mellifera (Hyme- noptera: Apidae) and the ecology of nectar intake . by bee colonies in a tropical forest. Oecologia, 61: 1-10. 4 Boggs, C. L. (1987) Ecology of nectar and pollen feeding in Lepidoptera. In “Nutritional Ecology of Insects, Mites, and Spiders”. Ed. by F. Slansky, Jr. & J. G. Rodriguez, John Wiley & Sons, pp. 369- SWll. 5 Watt, W. B., Hoch, P. C. and Mills, S. G. (1974) Nectar resource use by Colias butterflies. Chemical and visual aspects. Oecologia, 14: 353-374. 6 Watanabe, M., Koizumi, H., Suzuki, N. and Kirita- ni, K. (1988) Studies on ecology and behavior of Japanese black swallowtail butterflies. VII. Nectar of a glory tree, Clerodendron trichotomum, as a food resource of adults in summer. Ecol. Res., 3: 175- 180. 7 Baker, H. G. and Baker, I. (1973) Amino-acids in nectar and their evolutionary significance. Nature, 241: 543-545. 8 May, P. G. (1985) Nectar uptake rates and optimal nectar concentrations of two butterfly species. Oeco- logia, 66: 381-386. 9 Pivnick, K. A. and McNeil, J. N. (1985) Effects of nectar concentration on butterfly feeding: measured feeding rates for Thymelicus lineola (Lepidoptera: Hesperiidae) and a general feeding model for adult Lepidoptera. Oecologia, 66: 226-237. 10 Boggs, C. L. (1988) Rates of nectar feeding in 11 12 {3 14 15 16 7 Egg Maturation in Papilio xuthus butterflies: effects of sex, size, age and nectar con- centration. Functional Ecology, 2: 289-295. Wigglesworth, V. B. (1972) The principles of insect physiology. 7th ed. Chapman and Hall, London. Norris, M. J. (1936) The feeding-habits of the adult Lepidoptera Heteroneura. Trans. R. Ent. Soc. Lond., 85: 61-90. Tsubaki, Y. (1973) The natural mortality and its factors of the immature stages in the population of the swallow-tail butterfly Papilio xuthus Linne. Jap. J_ Ecol, 23: 210-217. Watanabe, M. and Nozato, K. (1986) Fecundity of the yellow swallowtail butterflies, Papilio xuthus and P. machaon hippocrates, in a wild environment. Zoological Science, 3: 509-516. Watanabe, M. (1988) Multiple matings increase the fecundity of the yellow swallowtail butterfly, Papilio xuthus L., in summer generations. J. Insect Be- havior, 1: 17-29. Watanabe, M. (1979) Population sizes and resident ratios of the swallowtail butterfly, Papilio polytes L., at a secondary bush community in Dharan, Nepal. Kontyu, 47: 291-297. Ban, Y., Kiritani, K., Miyai, S. and Nozato, K. (1990) Studies on ecology and behavior of Japanese black swallowtail butterflies. VIII. Survivorship 18 19 20 21 22 141 curves of adult male populations in Papilio helenus nicconicolens Butler and P. protenor demetrius Cramer (Lepidoptera: Papilionidae). Appl. Ent. Zool., 25: 409-414. Suzuki, N., Yamashita, K., Niizuma, A. and Kirita- ni, K. (1987) Studies on ecology and behavior of Japanese black swallowtail butterflies. 6. Nectar feeding of Papilio helenus nicconicolens Butler and P. protenor demetrius Cramer as main pollinators of glory bower, Clerodendron trichotomum Thunb. Ecol. Res., 2: 41-52. Dunlap-Pianka, H., Boggs, C. L. and Gilbert, L. E. (1977) Ovarian dynamics in Heliconiine butterflies: Programmed senescence versus eternal youth. Sci- ence, 197: 487-490. Karlsson, B. and Wiklund, C. (1984) Egg weight variation and lack of correlation between egg weight and offspring fitness in the wall brown butterfly Lasiommata megera. Oikos, 43: 376-385. Gilbert, L. E. and Singer, M. C. (1975) Butterfly ecology. Ann. Rev. Ecol. Syst., 6: 365-397. Arms, K., Feeny, P. and Lederhouse, R. C. (1974) Sodium: Stimulus for puddling behavior by tiger swallowtail butterflies, Papilio glaucus. Science, 185: 372-374. She hala ie meee 28 a nF oo cee Wok paint a his ‘ en ‘ pee ZOOLOGICAL SCIENCE 9: 143-148 (1992) Influence of Seawater Adaptation on Prolactin and Growth Hormone Release from Organ-Cultured Pituitary of Rainbow Trout TAKASHI YADA and TETSUYA HIRANO Ocean Research Institute, University of Tokyo, Nakano, Tokyo 164, Japan ABSTRACT—When immature rainbow trout (Oncorhynchus mykiss) were accliomated to 80% seawater for 2 weeks, the plasma osmolality and sodium level were slighlty but significantly higher than the levels in freshwater fish. Plasma prolactin (PRL) level of seawater-adapted fish was significantly lower than that of freshwater fish, whereas plasma growth hormone (GH) was significantly higher in seawater fish. Pituitary PRL and GH contents were significantly lower in seawater fish than in freshwater fish. When pituitaries were cultured in serum-free medium for 48 hr, an acute decrease in PRL release was seen within 12 hr in culture. Thereafter, a basal level of PRL release (0.2-0.5 ng/ pituitary-hr) was maintained, and the level was significantly lower in seawater fish than freshwater fish. On the other hand, GH release from the pituitary of both seawater and freshwater fish (200-500 ng/ pituitary-hr) was about 100 times greater than the basal PRL release throughtout the experiment, and there was no difference between the fish in seawater and those in fresh water. GH release from the trout pituitary seems to be predominantly under inhibitory control of the hypothalamus, thus resulting in an increased GH release under culture condition, and possibly masking the effect of seawater adaptation. The decreased PRL release in seawater-adapted fish may be related to the reduction in pituitary PRL © 1992 Zoological Society of Japan content. INTRODUCTION Osmoregulatory roles of prolactin (PRL) as a freshwater-adapting hormone is well established in many euryhaline teleosts [1-3]. In salmonids, a reduction in plasma level of PRL has been fre- quently observed after transfer from fresh water to seawater [4-8]. The decrease in plasma PRL in response to seawater is most likely to be releated to its sodium retaining action, which is inhibitory to maintenance of ion balance in seawater. On the other hand, recent studies suggest that growth hormone (GH) plays an important role in seawater adaptation, especially in salmonids, and the seawa- ter-adapting effect seems to be independent of growth promotion [9, 10]. An increase in plasma GH level has been observed during seawater acclimation in several salmonid species [6, 8, 11- 13]. An increased metabolic clearance rate of GH Accepted April 19, 1991 Received March 15, 1991 has also been seen when rainbow trout and coho salmon are transferred from fresh water to seawa- ter [12, 13]. Transfer of the fish to environment of different salinity generally causes chages in plasma osmolality. Reduction in osmotic pressure of culture medium is known to affect in vitro release of PRL in several teleost species [14, 15]. In salmonids, however, changes in osmotic pressure of medium within physiological range did not affect PRL and GH release in vitro [16, 17]. According to Kelley et al. [18], the basal release of newly synthesized PRL and PRL synthesis in organ-cultured pituitaries of coho salmon smolts were greater in freshwater fish than in the fish acclimated to seawater for 4 weeks; the PRL cells in vitro appear to retain at least some of their in situ characteristics. The present study was under- taken to examine the influrence of environmental salinity on spontaneous release of PRL and GH from organ-cultured pituitary of rainbow trout. 144 MATERIALS AND METHODS Immature rainbow trout (Oncorhynchus myk- iss), weighing about 100 g, were obtained from a commercial source in Tokyo. They were reared in recirculating freshwater tank at 12°C for more than a week until use. Twelve fish were transferreed directly to a recirculating tank of 80% seawater (salinity 30 ppt) at 12°C. As a control, the same number of the fish were transferred to a freshwater tank. They were fed commercially prepared pel- lets (Oriental, Chiba) during the experiment. Two weeks after the transfer, they were anesthetized in 2-phenoxyethanol (0.1 ml/liter), and blood was collected from the caudal vessels with a syringe needle treated with ammonium heparin. After centrifugation at 10,000 rpm for 5 min, the plasma was separated and kept frozen at —80°C until analyses. The pituitaries taken out upon decapita- tion were cultured in a 96-well multiple plate containing 2001 Eagle’s minimum essential medium with Earle’s salts (Gibco, New York), penicillin (100 U/ml), streptomycin (100 U/ml), and Fungizone (0.25 ug/ml, M. A. Bioproducts, Maryland). The pH of medium was adjusted to 7.3 TABLE 1. to fresh water or 80% seawater T. YADA AND T. HIRANO by sodium bicarbonate, and the osmotic pressure was 300 mOsm. Pituitaries were incubated at 12°C under an atmosphere of 95% Oy, and 5% CQO), for 48 hr. The medium was changed every 12 hr. Each pituitary was sonicated in 1 ml 0.1% Triton X-100 in 10mM phosphate buffered saline (pH 7.3). The medium and pituitary homogenates were stored at —80°C. As controls, unincubated pituitaries were also collected immediately after decapitation. Plsma sodium level was measured by an atomic absorption spectrophotometry (Hitachi, Tokyo). Osmotic pressure was measured by a vapor press- ure osmometer (Wescor, Utah). The medium and pituitary concentrations of PRL and GH were measured by homologous radioimmunoassays [19, 20]. RESULTS When rainbow trout was acclimated to 80% seawater for 2 weeks, a significant increase was seen in both plasma osmolality and sodium level (Table 1). Plasma PRL level was significantly lower in seawater fish than in freshwater fish. In © Plasma osmolality and concentrations of sodium, PRL and GH of rainbow trout acclimated ; No. Osmolality Na PRL GH SER of fish (mOsm/kg) (mEq/liter) (ng/ml) (ng/ml) fresh water i SY 323) sy2ae Il 0.8+0.2 SHOE) 80% seawater VW 334+4** IG aeze 0.4+0.1* 14.8+2.0* Data are expressed as mean+SEM. *-** Signifiantly different from the level in freshwater fish at P<0.05 and P<0.001, respectively, by Student’s f-test. TABLE 2. Pituitary contents of PRL and GH of rainbow trout acclimated to fresh water or 80% seawater No. PRL GH eo of fish (ug/ pituitary) (ug/pituitary) unincubated fresh water 6 2.67 + 0.30 19 9-5 ies pituitery 80% seawater 6 I JlseQ.1S“ S59 se ORSe pituitary incubated fresh water 6 BeI2e 093 (AN Ossil oe for 48 hr 80% seawater 6 2.07 +0.24 joa oad lis Data are expressed as mean+SEM. * Significantly different from the content in unincubated pituitary of freshwater fish at P<0.05 by Student’s t-test. PRL and GH release from trout pituitary 145 | FW SW (7) 1.0 oS NO 0.5 PRL release (ng/pituitary.hr) ow * *% * * (@) 0 0-12 42-24 24-36 36-48 hours in culture Fic. 1. Release of PRL from organ-cultured pituitaries of rainbow trout acclimated to fresh water (FW) or 80% seawater (SW). Data are expressed as mean+SEM (n=6). *°** significantly different from the corresponding level in freshwater fish at P<0.01 and P<0.001, respectively, by the Student’s t-test 600 =o ) i > FW SW = 400 | = a | 3 300 | cb) ” (40) ® 2900 | 2 1 ae © 100 i | ! | 0 | | | 0-12 12-24 24-36 36-48 hours in culture Fic. 2. Release of GH from organ-cultured pituitaries of rainbow trout acclimated to fresh water (FW) or 80% seawater (SW). Data are expressed as mean+SEM (n=6). 146 T. YADA AND T. HIRANO contrast, plasma GH level of seawater fish was significantly higher than that of freshwater fish. PRL and GH contents in the unincubated pituit- ary were significantly smaller in the fish acclimated to 80% seawater than those in freahwater fish. However, there was no significant difference in the residual pituitary contents of both PRL and GH between the fish in fresh water and 80% seawater (Table 2). As shown in Fig. 1, PRL release from the organ- cultured pituitary was 3-5 ng/pituitary-hr during the first 12 hr of incubation, and there was no difference between the fish acclimated to fresh water and those to 80% seawater. Thereafter, PRL release decreased markedly to a basal level of less than 0.5 ng/pituitary-hr. The basal release of PRL of freshwater fish (0.4-0.5 ng/pituitary-hr) was significantly greater than that of the fish in 80% seawater (about 0.2 ng/pituitary-hr). In con- trast, GH release from the cultured pituitary was about 150 ng/pituitary-hr during the first 12 hr, about 30 times greater than the PRL release during the same period, and higher levels (200-450 ng/ pituitary-hr) were maintained throughout the ex- periment (Fig. 2). There was no significant differ- ence in GH reiease between the fish in fresh water and those in seawater at any period. DISCUSSION In the present study, plasma PRL level of rain- bow trout acclimated to 80% seawater was signi- ficantly lower than in freshwater fish. This is in agreement with previous observations in several salmonid species [4-8]. The reduction in plasma PRL level seems to reflect decreased PRL relesae in vivo during seawater adaptation. PRL release from organ-cultured pituitary of seawater adapted rainbow trout was also significantly lower than that of freshwater fish, indicating that the influence of seawater adaptation on PRL release is still existent in the cultured pituitary at least for 48 hr. In coho salmon smolts, Kelley et al. [18] reported that the basal release of newly synthesized PRL, as mea- sured by incubating the pituitary with [*S]- methionine for 20 hr, was lower in seawater fish than in freshwater fish. The acute decline of PRL release within the first 12 hr in culture observed in the present study does not seem to be due to the exhaustion of PRL stored in pituitary, since a considerable amount of PRL remained in the pituitary after 48 hr in culture. Gonnet [21] re- ported that dominant hypothalamic control of PRL release in rainbow trout is stimulatory, based on their studies with perfused pituitaries. The acute decline in PRL release within the first 12 hr in culture may be caused by disappearance of the stimulatory control by hypothalamus. In teleosts, a considerable number of studies have focused on direct effect of osmotic pressure on PRL release in vitro [14, 15]. A reduction in osmotic pressure of cultrue medium directly stimu- lated PRL release in several euryhaline teleosts such as a goby, tilapia and eel [22-25]. In rainbow trout, however, the response of PRL release needed a large variation in osmotic pressure [17]. In the present study, pituitaries of seawater- adapted fish were incubated in the same culture medium as the pituitary of freshwater fish with osmolality of 300 mOsm/kg. In spite of a reduc- tion in osmotic pressure from 330 mOsm/kg in plasma to 300 mOsm/kg in culture medium, PRL release from seawater fish pituitary was still lower than in freshwater fish. In salmonids in general, changes in osmotic pressure with physiological ranges seem to have no significant influence on PRL release from organ-cultured pituitary [16- 18]. As is discussed above, osmolality of the culture medium is not the cause of the decrease in PRL release in seawater-adapted fish. It is also unlikely that the effect of hypothalamic hormones persists for more than 48 hr on PRL cells in the cultured pituitary, although hypothalamic nerve endings in the trout pituitary were found intact electron mic- roscopically after 8 days of culture [unpublished observation]. On the other hand, PRL content in the unincubated pituitary decreased significantly 2 weeks after transfer from fresh water to seawater. In rainbow trout weighing about 200 g, Prunet et al. [4] reported that pituitary PRL content in- creased within 2 days after transfer to seawater, possibly as a result of an acute decrease in PRL release. Thereafter, PRL content decreased gra- dually to the initial level 3 weeks after transfer. In coho salmon smolts, pituitary content and synth- PRL and GH release from trout pituitary esis of PRL decreased 4 weeks after transfer to seawater [18]. Although there was no significant difference in the residual PRL content between the fish in fresh water and in seawater after 48 hr in culture, the reduction in in vitro release of PRL in seawater fish as compared with that in freshwater fish seems to be related to the smaller pituitary content before the culture. In agreement with previous reports in rainbow trout and other salmonids [6, 8, 11, 13], plasma level of GH increased after adaptation of the trout to 80% seawater. However, in vitro release of GH in seawater fish was not different from that of freshwater fish. Nagahama et al. [26] suggested electron microscopically that GH cells in coho salmon pituitary were more active in seawater- adapted fish than in the fish in fresh water. Esti- mated from the metabolic clearance rate of exoge- nously administrated GH, GH released from the pituitary in vivo increased during seawater adapta- tion of rainbow trout and coho salmon [12, 13]. In salmonids, seawater adaptation seems to stimulate both release and sythesis of GH. Smaller GH content in the unincubated pituitary in seawater fish than in freshwater fish observed in the present study may reflect greater secretion rate than the rate of synthesis. In our recent study, activation of GH release and synthesis was observed in serum- free culture of rainbow trout pituitary, indicating that predominant control of GH secretion in rain- bow trout is inhibitory [27]. Thus, the influence of seawater adaptation on GH release, if any, seems to be masked by the increased GH cell activity under the culture condition. Further studies are needed with dispersed or isolated pituitary cells without nerve endings of hypothalamic neurons to clarify the influence of seawater adaptation on PRL as well as GH release from the trout pituitary. ACKNOWLEDGMENTS This study was supported in part by grants-in-aid for sciencific research from the Ministry of Education, Scien- ce and Culture and also from the Fishery Agency, Japan. REFERENCES 1 Nicoll, C. S. (1981) Role of prolactin in water and 10 fil 12 13 147 electrolyte balance in vertebrates. In “Prolactin”. Ed. by R. B. Jaffe, Elsevier, New York, pp. 127- 166. Bern, H. A. (1983) Functional evolution of prolac- tin and growth hormone in lower vertebrates. Amer. Zool., 23: 663-671. Hirano, T. (1986) The spectrum of prolactin action in teleosts. In “Comparative Endocrinology: De- velopments and Directions”. Ed. by C. L. Ralph, A. R. Liss, New York, pp. 53-74. Prunet, P., Boeuf, G. and Houdebine, L. M. (1985) Plasma and pituitary prolactin levels in rainbow trout during adaptation to different salinities. J. Exp. Zool., 235: 187-196. Prunet, P., Boeuf, G., Bolton, J. P. and Young, G. (1989) Smoltification and seawater adaptation in Atlantic salmon (Salmon salar): Plasam prolactin, growth hormone, and thyroid hormones. Gen Comp. Endocrionl., 74: 355-364. Hasegawa, S., Hirano, T., Ogasawara, T., Iwata, M., Akiyama, T. and Arai, S. (1987) Osmoregula- tory ability fo chum salmon, Oncorhynchus keta, reared in fresh water for prolonged periods. Fish Physiol. Biochem., 4: 101-110. Young, G., Bjornsson, B. Th., Prunet, P., Lin, R. J. and Bern, H. A. (1989) Smoltification and seawater adaptation in coho salmon (Oncorhynchus kisutch): Plasma prolactin, growth hormone, thyroid hormones, and cortisol. Gen. Comp. Endocrionl., 74: 335-345. Yada, T., Takahashi, K. and Hirano T. (1991) Seasonal changes in seawater adaptability and plas- ma levels of prolactin and growth hormone in land- locked sockeye salmon (Oncorhynchus nerka) and amago salmon (O. rhodurus). Gen. Comp. Endoc- rionol., 82: 33-44. . Bolton, J. P., Collie, N. L., Kawauchi, H. and Hirano, T. (1987) Osmoregulatory actions of growth hormone in rainbow trout (Salmo gairdneri). J. Endocrinol., 112: 63-68. Collie, N. L., Bolton, J. P., Kawauchi, H. and Hirano, T. (1989) Survival of salmonids in seawater and the time-frame of growth hormone action. Fish Physiol. Biochem., 7: 315-321. Sweeting, R. M., Wagner, G. F. and McKeown, B. A. (1985) Changes in plasma glucose, amino acid nitrogen and growth hormone during smoltification and seawater adaptation in coho salmon, Oncorhyn- chus kisutch. Aquaculture, 45: 185-197. Sakamoto, T., Ogasawara, T. and Hirano, T. (1990) Growth hormone kinetics during adaptation to a hyperosmotc environment in rainbow trout. J. Comp. Physiol. B, 160: 1-6. Sakamoto, T., Iwata, M. and Hirano, T. (1991) Kinetic studies of growth hormone and prolactin during adaptation of coho salmon, Oncorhynchus 14 15) 16 107) 18 19 20 148 kisutch, to different salinities. Gen. Comp. Endocri- nol., 82: 184-191. Ball, J. N. (1981) Hypothalamic control of the pars distalis in fishes, amphibians, and reptiles. Gen. Comp. Endocrino., 44: 135-170. Nishioka, R. S., Kelley, K. M. and Bern, H. A. (1988) Control of prolactin and growth hormone secretion in teleost fishes. Zool. Sci., 5: 267-280. Suzuki, R., Kishida, M., Ogasawara, T., Hasegawa, S. and Hirano, T. (1987) Prolactin and growth hormone secretion during long-term incubations of the pituitary pars distalis of mature chum salmon, Oncorhynchus keta. Gen. Comp. Endocrinol., 68: 76-81. Gonnet, F., Prunet, P., Tonon, M. C., Dubourg, P.., Kah, O. and Vaudry, H. (1988) Effect of osmotic pressure on prolactin release in rainbow trout: in vitro studies. Gen. Comp. Endocrionol., 69: 252- 261. Kelley, K. M., Nishioka, R. S. and Bern, H. A. (1990) In vitro effect of osmotic pressure and cortisol on prolactin cell physiology in the coho salmon (Oncorhynchus kisutcn) during the parr- smolt transformation. J. Exp. Zool., 254: 72-82. Hirano, T., Prunet, P., Kawauchi, H., Takahashi, A., Ogasawara, T., Kubota, J., Nishioka, R. S., Bern, H. A., Takada, K. and Ishii, S. (1985) De- velopment and validation of a salmon prolactin radioimmunoassay. Gen. Comp. Endocrinol., 59: 266-276. Bolton, J. P., Takahashi, A., Kawauchi, H., Kubo- ta, J. and Hirano, T. (1986) Development and validation of a salmon growth hormone radioimmu- noassay. Gen. Comp. Endocrionol., 62: 230-238. 72) jap) 23 24 25 26 Di T. YADA AND T. HIRANO Gonnet, F., Barret, A., Grouselle, D. and Prunet, P. (1989) Hypothalamic control of prolactin release in the rainbow trout, Salmo gairdneri: in vitro stu- dies. Fish Physiol. Biochem., 7: 301-308. Nagahama, Y., Nishida, R. S., Bern, H. A. and Gunther, R. L. (1975) Control of prolactin secre- tion in teleosts, with special reference to Gillichthys mirabilis and Tilapia mossambica. Gen. Comp. En- docrinol., 25: 166-188. Grau, E. G., Shimoda, S. K., Ford, C. A., Helms, L. M. H., Cooke, I. M. and Pang, P. K. T. (1986) The role of calcium in prolactin release from the pituitary of a teleost fish in vitro. Endocrinology, 119: 2848-2855. Grau, E. G., Ford, C. A., Helms, L. M. H., Shimoda, S. K. and Cooke, I. M. (1987) Somatosta- tin and altered medium osmotic pressure elicit rapid changes in prolactin release from the rostral pars distalis of the tilapia, Oreochromis mossambicus, in vitro. Gen. Comp. Endocrinol., 65: 12-18. Suzuki, R., Kaneko, T. and Hirano, T. (1991) Effects of osmotic pressure on prolactin and growth hormone secretion from organ-cultured eel pituit- ary. J. Comp. Physiol. B, 161: 147-153. Nagahama, Y., Clarke, W. C. and Hoar, W. S. (1977) Influence of salinity on ultrastructure of the secretory cells of the adenohypophyseal pars distalis in yearling coho salmon (Oncorhynchus kisutch). Can. J. Zool., 55: 183-198. . Yada, T., Urano, A. and Hirano, T. (1991) Growth hormone and prolactin gene expression and release in the pituitary of rainbow trout in serum-free culture. Endocrinology, 129: 1183-1192.’ ZOOLOGICAL SCIENCE 9: 149-155 (1992) © 1992 Zoological Society of Japan Comparison of the In Vivo and In Vitro Effects of Bombyxin and Prothoracicotropic Hormone on Prothoracic Glands of the Silkworm, Bombyx Mori SHONOSUKE KirtsHt’, H1roMicHt NAGASAWA~, HirosHI KATAOKA?, AKINORI SUZUKI’ and SHO SAKURAI * ‘Department of Biology, Faculty of Science, Kanazawa University, Kanazawa 920, and *Department of Agricultural Chemistry, Faculty of Agriculture, The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan. ABSTRACT— When assayed using an in vitro culture of Bombyx prothoracic glands (PGs), bombyxin stimulated ecdysteroid production by the PGs in a dose-dependent manner. However, the dose required for attaning maximum activation was as high as 800 and 1600 Samia units/ml for larval and pupal PGs, respectively. By contrast, prothoracicotropic hormone (PTTH) enhanced ecdysteroid production by larval glands 20 times as much as the control at a concentration of 16 Bombyx units/ml and that of pupal PGs by 10 times at 64 units/ml. From these results we conclude that bombyxin cannot be regarded as a physiological PITH, despite its demonstrated ability to activate the PGs in vitro at excessively large doses. INTRODUCTION Brains of certain lepidopteran insects contain two different classes of molecules which exhibit prothoracicotropic activity. The brain of Manduca sexta secretes small and big prothoracicotropic hormone (PTTHs) [1]. Activation by these PTTHs in Manduca is stage-specific: big PTTH stimulates equally the Manduca larval and pupal PGs while small PTTH activates larval PGs much more than pupal PGs. The brain of the silkworm, Bombyx mori, con- tains similarly two neurosecretory substances with prothoracicotropic activity [2]. One is PITH (30 kDa) which activates its own larval and pupal PGs of Bombyx mori. The other is bombyxin (5 kDa), once called PTTH-S [3] or 4K-PTTH [2], that is a peptide belonging to the insulin family [4, 5] and exerts strong activity to stimulate adult develop- ment in the brainless pupae of Samia cynthia ricini [3], and in the diapausing pupae of Papilio xuthus Accepted September 3, 1991 Received July 29, 1991 * To whom all correspondence should be addressed. [6]. Bombyxin failed to induce a larval moulting of the neck-ligated, 4th instar Bombyx larvae [7] and an adult development of Bombyx dormant pupae produced by brain extirpation [8]. Meanwhile, we observed that Bombyx PGs were activated in vitro by an addition of crude bombyxin sample to the culture medium (Kataoka, unpublished data). Naturally there remained a possibility that the crude bombyxin sample was contaminated with a small amount of PTTH. It has now become possible for us to use enough amount of pure, natural bombyxin to examine whether it can really activate PGs in vitro. In this paper, we reassessed the effects of bombyxin on the Bombyx PGs both in vivo by using brainless pupae and in vitro by using larval and pupal PGs and found that it stimulated both larval and pupal PGs in vitro to a certain extent though it was totally inactive in the in vivo assay. The effects of bombyxin and PTTH on PGs were also examined by qualitative and quantitative analyses of ecdysteroids produced by the PGs after stimulation by either bombyxin or PTTH. 150 S. KirusHi, H. NAGASAWA et al. MATERIALS AND METHODS | Animals Larvae of the silkworm, Bombyx mori, (Gunpo x Shuhgyoku), from which PGs were obtained for in vitro assay of PTTH and bombyxin, were reared on an artificial diet under a 12L: 12D photperiod at 25+1°C [9]. Larvae of another racial hybrid J-122 x C-115, which were used for PTTH and bombyxin bioassays, were reared on an artificial diet under an 18L:6D photoperiod at 25+1°C [8]. Hormones Ecdysone and 20-hydroxyecdysone were purch- ased from Sigma and 3-dehydroecdysone was a gift from J. T. Warren and L. I. Gilbert (The Universi- ty of North Carolina). Bombyx PTTH and bomby- xin (bombyxin-II) were obtained from Bombyx adult heads [5, 10]. A PTTH preparation after step 10 by Octyl-Sepharose chromatography and pure bombyxin-II were dissolved in Grace’s medium. The biological activity of bombyxin and PTTH is expressed in Samia unit (SU; 1 SU=0.4 ng bombyxin-II) [11] and Bombyx unit (BU; 1 BU =(0.1 ng PTTH) [8], respectively. Bioassay of bombyxin and PTTH Bioassay for bombyxin and PTTH was per- formed using brainless pupae of a racial hybrid J-122 x C-115 of Bombyx mori [8]. Each pupa received a 10 yl test solution and observed 6 days after injection for wing apolysis, an indication of initiation of adult development. Incubation of PGs PGs were dissectd out from racial hybrid Gunpo x Shuhgyoku either on day 5 of the Sth instar or on the day of pupation, and incubated in 50 ul Grace’s medium (GIBCO) for 4 h at 25°C. Activa- tion of PGs by either bombyxin or PITH was assayed according to Bollenbacher et al. [12]. One gland of a pair was used as a control while the contralateral one was incubated in the presence of bombyxin or PITH. After incubation, 10 ul of each medium diluted with Grace’s medium to an appropriate concentration of ecdysteriods were directly subjected to ecdysteroid radioimmuno- assay (RIA) to determine the ecdysteroid amount in the medium. The activation by either bombyxin or PTTH is expressed in activation ratio (Ar) accord- ing to Bollenbacher et al. [12]. To prepare the samples for reverse phase high performance liquid chromatographical (RP-HPLC) analysis of ecdys- teroids, 8-16 one-side PGs were individually incu- bated in 50 1 plain Grace’s medium while the same number of the contralateral glands were incubated in the presence of either bombyxin or PTTH. After quantification of ecdysteroids in each medium by RIA, the reamining media were combined, extracted and dissolved in 5% acetonit- rile. The solution containing ca. 5-10 ng ecdyster- oids were subjected to RP-HPLC analysis [13]. Ecdysteroids were identified by RP-HPLC and RIA. RP-HPLC was done with a Cj » Bondas- phere column (Waters, 3.9150 mm) by gradient elution of 5 to 30% acetonitrile/water in 40 min at a flow rate of 1 ml/min at 25°C, and fractions were collected every 0.5 min. Although the retention times of reference compounds varied, ecdysone and 3-dehydroecdysone were easily identified by differential RIA [14]. Two different antisera were used for the ecdysteroid RIA, H-22 with high affinity for ecdysone and 20-hydroxyecdysone, and S-3. with high affinity for ecdysone and 3- dehydroecdysone [13]. The basic procedure for RIA has been outlined by Warren and Gilbert [15], and in the present study ammonium sulfate was used to separate bound and free ligands. Since no RIA active substance other than ecdysone and 3-dehydroecdysone was detected in the incubation medium (unpublished data), data of RP-HPLC- RIA analysis are present for only the fractions including these two ecdysteroids. RESULTS Activity of bombyxin in brainless pupae To reconfirm the ineffectiveness of bombyxin on the induction of adult development of Bombyx pupae, various doses of pure bombyxin-II were injected into brainless pupae. Table 1 shows that as much as 1000 SU (400 ng) of bombyxin-II failed to induce adult development. By contrast, all the brainless pupae injected with 1 BU of PTTH Effects of Bombyxin and PTTH 151 TasLe 1. Effects of injection of natural bomyxin-II initiated adult development within 6 days after on induction of adult development in the brainless injection. Bombyx female pupae Effects of bombyxin and PTTH on PGs in vitro Response* Hormone number iil, aes A dose-response curve of activation of larval PGs was generated using both bombyxin and bombyxin, 10SU 10 10 0 PTTH (Fig. 1). When PTTH was assayed using 100 SU 10 10 0 larval PGs, the highest Ar was approximately 22 at 1000 SU 10 10 0 a concentration of 16 BU/ml. At this dose, the PTTH 1 BU 10 0 10 curve still did not reach a plateau. Bombyxin was PSA* 10 10 0 slightly effective on PGs in vitro. The activation * Number of brainless pupae that initiated adult was dose-dependent up to a 800 SU/ml concentra- development (+) or not (—) within 6 days after tion, where the maximum Ar of 6 was obtained, injection of either bombyxin or PTTH. and at a higher dose of 1600 SU/ml, the Ar was * Four micrograms of bovine serum albumin were injected as a control. SU, Samia unit; BU, Bombys unit. zs | reduced. When assayed using pupal PGs, the Ar by PTTH was 10.5 at a concentration of 64 BU/ ml, while that by bombyxin was 3.7 at 1600 SU/ml Activation ratio OFZ57 O15 0 25 Iug Sra 25 50 100 200 400 800 1600 PTTH (BU/ml) bombyxin (SU/ml) Fic. 1. Effects of PITH and bombyxin on ecdysteroid production by PGs of day 5 fifth instar larvae. One PG ofa pair was incubated in 50 wl of Grace’s culture medium for 4 h at 25°C while the contralateral PG was incubated in the presence of bombyxin-II or PTTH at varying concentrations indicated. After incubation, each medium was directly subjected to ecdysteroid RIA to determine ecdysteroid amount in the medium. Activation ratio indicates the value obtained by dividing the amount of ecdysteroids produced by PGs in the presence of bombyxin or PITH by the amount of ecdysteroids produced by the contralateral PGs in the control medium [12]. BU, Bombyx unit [8]; SU, Samia unit [11]. Each datum point represents a meant+S.D. (N=4). 152 15 10 Activation ratio ¢ o— 444 O75) al 2 64 8 PTTH (BU/ml) Fic. 2. Effects of PITH and bombyxin on ecdysteroid production by PGs of day 0 pupae. Each datum point represents a mean+S.D. (N=4). for Figure 1. 10 32 64 128 S. Kirusui, H. NAGASAWA et al. 25 50 100 200 400 8001600 bombyxin (SU/ml) Details are the same as TABLE 2. Comparison of the anount of ecdysteroids produced by larval or pupal PGs in the presence or absence of either bombyxin or PTTH at a concentration to elicit the maximum activation donor of prothoracic n noone glands (unit/ml) Sth instar 4 PTTH larvae (16BU) (day 5) 4 bombyxin (800SU) et 4 PTTH (Gay Y) (64BU) 5 bombyxin (1600SU) (Fig. 2). The actual amounts of ecdysteroid sec- reted by the PGs at the bombyxin or PTTH concentration that gave the highest activation are shown in Table 2. RP-HPLC analysis of ecdysteroids produced by PGs in the presence or absence of either bombyxin or PTTH Bombyx larval PGs in vitro secreted primarily ecdysone and 3-dehydroecdysone was secreted arate carl steal se | ecdysteroids activation (ng/giand+S.D.) ratio 0.60+0.35 ISR Sate 5-35 21.7+2.4 0.64+0.51 3.49+2.48 5 SxS Ud Qaz leo 84.70 +6.85 10.8+2.0 ae 2.725) SoM aera I3) 4.0+1.2 only in a very small amount (Fig. 3a). When stimulated by 800 SU/ml of bombyxin-II, the PGs secreted similarly only a very small amount of 3-dehydroecdysone compared to ecdysone (Fig. 3b), as in the non-stimulated control PGs. After PTTH stimulus (16 BU/ml; Fig. 3c), a consider- able amount of 3-dehydroecdysone was secreted but the proportion of 3-dehydroecdysone to ecdy- sone did not significantly differ from that of control PGs. Pupal PGs with no added hormones secreted Effects of Bombyxin and PTTH 153 C uw ~ oc > = oO Oo lJ a og iD) ss | 45 20 25 30 Retention time Fic. 3. Reverse-phase HPLC of ecdysteroids produced in vitro by the PGs of day 5, fifth instar larvae in the control medium (a) or in the presence of either bombyxin (b, 800 SU/ml) or PTTH (c, 16 BU/ml). Fractions were collected every 0.5 min and an aliquot of each fraction was subjected to ecdysteroid RIA using the H-22 antiserum (filled circles) and S-3 antiserum (open circles). E, ecdysone; 3DE, 3-dehydroecdysone. >) < > E (So) = ra < w ; Lu oO = | ra | no) 30 35 25 30 30 35 Retention time Fic. 4. _Reverse-phase HPLC of ecdysteroids produced in vitro by the PGs of day 0 pupae in the control medium (a) or in the presence of either bombyxin (b, 800 SU/ml) or PTTH (c, 64 BU/ml). Fractions were collected every 0.5 min and an aliquot of each fraction was subjected to ecdysteroid RIA using the H-22 antiserum (filled circles) and S-3 antiserum (open circles). EE, ecdysone; 3DE, 3-dehydroecdysone. 154 S. KirusHi, H. NAGASAwA et al. primarily ecdysone, accompanied by a small amount of 3-dehydroecdysone like in the larval PGs (Fig. 4a). When the PGs were stimulated by bombyxin-II (800SU/ml), the amount of 3- dehydroecdysone slightly increased (Fig. 4b). By contrast, when the PGs were stimulated by PITH (64 BU/ml), the PGs secreted a very small amount of 3-dehydroecdysone, as in the non-stimulated control PGs (Fig. 4c). DISCUSSION The present data clearly show that bombyxin does not stimulate Bombyx PGs to secrete ecdys- teroids at a physiological concentration. Although it slightly activated Bombyx larval and pupal PGs in vitro, the dose of bombyxin required to attain enough activation was extraordinarily high. In addition, an injection of a large amount of bomby- xin (1000 SU/pupa) failed to evoke adult develop- ment of brainless Bombyx pupae. The concentra- tion of haemolymph bombyxin immediately after the injection could be 2500 SU/ml at least since haemolymph occupies 30-40% of pupal body weight (unpublished data) and the average weight of a pupa was approximately 1 gr or less. This indicates that the injected bombyxin presumably stimulated pupal PGs but the activation was not yet enough to induce adult development. The contents of bombyxin in one brain throughout the Bombyx life cycle is 1-30 SU [16, 17]. Even if bombyxin secretion is very fast and only a small amount of bombyxin is stored in the secretory cells, it is unlikely that a brain can secrete a large amount of bombyxin to elevate the bombyxin content as mush as 1000 SU/pupa or more. Accor- dingly, bombyxin may not participate in the activa- tion of Bombyx PGs to induce moulting at all. When in vitro effects of bombyxin and PTTH in Bombyx are compared with those of small and big PTTHs in Manduca, the most pronounced differ- ence is that in Bombyx, only PTTH stimulates PGs at physiological concentrations while in Manduca, big PTTH stimulates the larval and pupal PGs equally whereas small PTTH preferentially stimu- lates the larval PGs. Small PTTH does not suc- cessfully stimulate Manduca pupal PGs: its dose eliciting maximal activation of pupal PGs is 6 units/25 wl, corresponding to 240 units/ml [1]. Even though small PTTH stimulates the larval PGs in vitro, its effects on larval moulting is negligible [18]. Indeed, small PTTH was not found in the course of purification of Manduca PTTH as far as using the larval assay [19]. Accordingly, Manduca small PTTH as well as bombyxin may not be egarded as a physiological PTTH in a rational consideration. We suggest, therefore, that insect brains contain a single molecular class of PITH, so called big-PTTH in Manduca and others [20, 21]. Although bombyxin is totally inactive to stimu- late adult development of Bombyx brainless pupae, it slightly activated Bombyx PGs in vitro. Bombyxin belongs to an insulin-super family with respect to amino acid sequence. If its in vitro effects were due to the insulin-like action such as stimulation of metabolic reactions, then the sti- mulation by bombyxin could be different from that by PTTH. We tried to examine this possibility by comparing the amount of each of ecdysone and 3-dehydroecdysone produced by PGs in the pre- sence of bombyxin or PTTH, but failed to find any difference to support this assumption so far ex- amined. | In spite of the inability of bombyxin to stimulate Bombyx PGs at a physiological dose both in vivo and in vitro, it exerts strong stimulatory effect on the PGs of Samia cynthia |4| and Papilio xuthus [22]. Similarly, M. brassicae ‘small PTTH’ stimu- lates Bombyx PGs and Papilio ‘small PTTH’ acti- vates PGs of M. brassicae, B. mori and P. c- aureum [21]. These observations indicate that the spectrum of the action of small PTTH is fairly wide, i.e. its species-specificity could be low. Whether the low species-specificity is due to small PTTHs belonging to insulin-super family awaits for the elucidation of the chemical structures of Samia, Papilio and other small PTTHs. ACKNOWLEDGMENTS The authors thank Drs. Jim T. Warren and Lawrence I. Gilbert (University of North Carolina at Chapel Hill) for the supply of H-22 antiecdysone antiserum and Dr. Hironori Ishizaki for helpful comments on this paper. This work was supported in part by gran-in-Aid from the Ministry of Education, Science and Culture, Japan No. 01060004 to A. S. and No. 01540592 to S. S. 10 Effects of Bombyxin and PTTH REFERENCES Bollenbacher, W. E., Katahira, E. J., O’Brien, M., Gilbert, L. I., Thomas, M. K., Agui, N. and Baumhover, A. H. (1984) Insect prothoracicotropic hormone: Evidence for two molecular forms. Si- cence, 224: 1243-1245. Ishizaki, H. and Suzuki, A. (1984) Prothoracicotro- pic hormone of Bombyx mori. In “Biosynthesis, Metabolism and Mode of Action of Invertebrate Hormones”. Ed. by J. Hoffmann and M. Prochet, Springer-Verlag, Berlin, pp. 63-77. Ishizaki, H., Mizoguchi, A., Fujishita, M., Suzuki, A., Moriya, I., O’Oka, H., Kataoka, H., Isogai, A., Nagasawa, H., Tamura, S. and Suzuki, A. (1983) Species specificity of the insect prothoracicotropic hormone (PTTH): The presence of Bombyx- and Samia-specific PTTHs in the brain of Bombyx mori. Dev. Grow. Diff., 25: 593-600. Nagasawa, H., Kataoka, H., Hori, Y., Isogai, A., Tamura, S., Suzuki, A., Guo, F., Zhong, X., Mizo- guchi, A., Fujishita, M., Takahashi, S. Y., Ohnishi, E. and Ishizaki, H. (1984) Isolation and some characterization of the prothoracicotropic hormone from Bombyx mori. Gen. Comp. Endocr., 53: 143- ISAs Nagasawa, H., Kataoka, H., Isogai, A., Tamura, S., Suzuki, A., Mizoguchi, A., Fujiwara, Y., Suzu- ki, A., Takahashi, S. Y. and Ishizaki, H. (1986) Amino acid sequence of a prothoracicotropic hor- mone of the silkworm Bombyx mori. Proc. natl. Acad. Sci. USA, 83: 5840-5843. Masaki, T., Endo, K. and Kumagai, K. (1988) Neuroendocrine regulation of the development of seasonal morphs in the Asian comma butterfly, Polygonia c-aureum L.: Is the factor producing summer morphs (SMPH) identical to the small prothoracicotropic hormone (4K-PTTH)? Zool. Sci., 5: 1051-1057. Suzuki, C. and Ishizaki, H. (1986) Prothoracicotro- pic hormone assay: Bombyx larval assay. Int. J. Invert. Reprod. Dev., 10: 256-274. Ishizaki, H., Suzuki, A., Moriya, I., Mizoguchi, A., Fujishita, M., O’Oka, H., Kataoka, H., Isogai, A., Nagasawa, H. and Suzuki, A. (1983) Prothoracicot- ropic hormone bioassay: Pupal-adult Bombyx assay. Dev. Grow. Diff., 25: 585-592. Sakurai S. (1984) Temporal organization of endoc- rine events underlying larval-pupal metamorphosis in the silkworm, Bombyx mori. J. Insect Physiol., 30: 657-664. Kataoka, H., Nagasawa, H., Isogai, A., Tamura, S., Mizoguchi, A., Fujiwara, Y., Suzuki, C., Ishiza- ki, H. and Suzuki, A. (1987) Isolation and partial characterization of a prothoracicotropic hormone of the silkworm, Bombyx mori. Agric. Biol. Chem., 11 iw 13 14 115) 16 17 18 19 20 Z| 22 155 51: 1067-1076. Ishizaki, H. and Ichikawa, M. (1967) Purification of the brain hormone of the silkworm Bombyx mori. Biol. Bull., 133: 355-368. Bollenbacher, W. E., Agui, N., Granger, N. A. and Gilbert, L. I. (1979) In vitro activation of insect prothoracic glands by the prothoracicotropic hor- mone. Proc. natl., Acad. Sci. USA, 76: 5148-5152. Kiriishi, S., Rountree, D. B., Sakurai, S. and Gil- bert, L. I. (1990) Prothoracic gland synthesis of 3-dehydroecdysone and its hemolymph 3/-reductase mediated conversion to ecdysone in representative insects. Experientia, 46: 716-721. Warren, J. T., Sakurai, S., Rountree, D. B., Gil- bert, L. I., Lee S-S. and Nakanishi, K. (1988) Regulation of the ecdysteroid titer of Manduca sexta: Reappraisal of the role of the prothoracic gland. Proc. natl. Acad. Sci. USA, 85: 958-962. Warren, J. T. and Gilbert, L. I. (1988) Radioimmu- noassay: Ecdysteroids. In “Immunological Techni- ques in Insect Biology”. Ed. by L. I. Gilbert and T. A. Miller, Springer-Verlag, New York, pp. 181- 214. Ishizaki, H. (1969) Changes in titer of the brain hormone during development of the silkworm, Bombyx mori. Dev. Grow. Diff., 11: 1-7. Mizoguchi, A., Hata, M., Sato, S., Naasawa, H., Suzuki, A. and Ishizaki, H. (1990) Developmental change of bombyxin content in the brain of the silkmoth Bombyx mori. J. Insect Physiol., 36: 655- 664. Watson, R. D., Thomas, M. K. and Bollenbacher, W. E. (1989) Regulation of ecdysteroidogenesis in prothoracic glands of the tobacco hornworm Man- duca sexta. J. Exp. Zool., 252: 255-263. Kingan, T. G. (1981) Purification of the prothoraci- cotropic hormone from the tobacco hornworm Man- duca sexta. Life Sci., 28: 2585-2594. Masler, E. P., Kelly, T. J., Thyagaraja, B. S., Woods, C. W., Bell, R. A. and Borkovec, A. B. (1986) Discovery and partial characterization of prothoracicotropic hormones of the gypsy moth, Lymantria dispar. In “Insect Neurochemistry and Neurophysiology”. Ed. by D. B. Gelman, Humana Press, Clifton, pp. 331-334. Fujimoto, Y., Endo, K., Watanabe, M. and Kuma- gai, K. (1991) Species-specificity in the action of big and small prothoracicotropic hormones of four spe- cies of lepidopteran insects, Manestra brassicae, Bombyx mori, Papilio xuthus and Polygonia c- aureum. Zool Sci., 8: 351-358. Endo, K., Fujimoto, Y. and Masaki, T. (1990) Stagedependent changes in the activity of the prothoracicotropic hormone (PTTH) in the brain of the Asian comma butterfly, Polygonia c-aureum L. Zool. Sci., 7: 697-704. ar ro pears y olsen | ie eye atl ae Le A eon a Sie Pp deamlies ; Sram ae ee = sn fina eetubnkitee | a erent een saa 7 = ye 3 i y ae = binge popluaiat veh = Png a sed ib bephikoscn eS Sia a. sR obo cag Tinian, Deadly me: ila are ARES ees eid Aon onaiaiialy Fh TOR: yaw 2 ul So RE anne oS het miwadincrdet estan coca me fi he 4 eetned:. Sat af anh 3 ; vi +e Tote ty: 4 es ah i ‘ var Es sees stay ert if e ae Lf . a = 1 ed 5 aa M z 3 ; ; +S ie tae ; pees Vata Lgl oe i 1 At na ¥ S i 7 2 ie , vhs e = } yk av: CRD ey t F 2 LN Sy ie enheis Gaponel rhe ie an ee LP eew Es tke Re SEs doa Said hi ek oe 2 eats i Tenpeiinatisthcs } ee. ida SPT ae ah : a4 Zz c ms ey: : 5 Ge TE OU Ei ae ; ; Ley te 5 7 = i r - 9 é 5 , a0 ; i : C ) a4 ce. = r ik 3 ; a ee i tall , i : , Ni be te pirat es \ i F ‘ * t » j s' = i} : a oe ; ; Noss aust Bye ene: i, i { oe) i y i : Z s =| P 4% i 4 : ci he ay Gat . i .? Sie - ms = “AEA. a , FS i , y ai ‘ , , r shy : , vdy - ' . : t : a oa ZOOLOGICAL SCIENCE 9: 157-167 (1992) © 1992 Zoological Society of Japan Cloning and Sequence Analyses of Vasotocin and Isotocin Precursor cDNAs in the Masu Salmon, Oncorhynchus masou: Evolution of Neurohypophysial Hormone Precursors MASAKAZU SUZUKI, SUSUMU Hyopo! and AKIHISA URANO Laboratory of Molecular Biology, Ocean Research Institute University of Tokyo, Minamidai, Nakano-ku, Tokyo 164, Japan ABSTRACT—We have cloned and determined the nucleotide sequences of cDNAs encoding precur- sors of neurohypophysial hormones, vasotocin (VT) and isotocin (IT), from the hypothalamus of masu salmon, Oncorhynchus masou. The deduced amino acid sequences of masu salmon VT and IT precursors (proVT-I and proIT-I) are highly homologous to those of chum salmon proVT-I and prolT-I, respectively. The VT and IT precursors are composed of a signal peptide, hormone and neurophysin (NP), the middle portion of which is highly conserved among vertebrates. Both the NPs extend about 30 amino acids at the C-terminal. The extended C-terminals have a leucin-rich segment in the carboxyl-terminal, as copeptin of vasopressin precursor. Southern bot analysis showed the presence of two types of proVT genes (proVT-I and proVT-II) and proIT genes (proIT-I and proIT-II) in an individual masu salmon, as in a chum salmon. Southern blot analysis with proVT probes further suggested that at least two different types of proVT-I genes exist in a single masu salmon. Norhtern blot analysis indicated that proVT-I and proIT-I genes are expressed in the hypothalamus, whereas proVT-II and proIT-II genes are not expressed. Evolutionary distance between proVT-I and proIT-I genes was statistically estimated based on synonymous nucleotide substitution in the coding region of the cDNAs. The magnitude of distance between masu salmon proVT-I and proIT-I genes suggested that the highly conserved central portion of NPs resulted from a gene conversion event. Between masu salmon and chum salmon, evolutionary distance for proVT-I genes is about 6-fold larger than that for proIT-I genes. INTRODUCTION Ten distinct neurohypophysial hormones have been characterized in a wide variety of vertebrates, so that many schemes have been proposed for the evolutionary pathway of amino acid substitution based on the primary structures and phyletic dis- tributions of hormones [1, 2]. Meanwhile, a recent molecular biological study clarified that neuro- hypophysial hormones were synthesized via larger precursors [3]. This fact strongly suggests that it is indispensable to reestimate molecular evolution of neurohypophysial hormones in terms of their pre- cursors. Accepted August 20, 1991 Received June 19, 1991 ' Present address: Department of Biology, College of Arts and Sciences, University of Tokyo, Komaba, Meguro-ku, Tokyo 153, Japan The nucleotide sequences of cDNAs encoding the precursors of arginine vasopression (VP) and oxytocin (OT) were first determined in bovine [4, 5]. The deduced amino acid sequence of VP precursor consists of a signal peptide, VP, neurophysin II which is a carrier protein specific to VP, and a glycoprotein named copeptin. The OT precursor is shorter and consists only of a signal peptide, OT and neurophysin I which is OT spe- cific carrier protein. Thereafter, primary struc- tures of neurohypophysial hormone precursors were revealed not only in mammals, e.g. human [6], but also in lower vertebrates such as toad [7], white sucker [8-10] and chum salmon [11, 12]. From these results, Hyodo et al. [12] estimated evolutionary distances among mRNAs for precur- sors on the basis of the nucleotide sequences, and proposed a scheme for molecular evolution of neurohpophysial hormone precursors. 158 Most salmonids are, like catostomids, con- sidered to be tetraploid, because their cell nuclei contain about 2-fold amount of DNA and chromo- somes compared to other species in the same order, Clupeiformes [13]. Furthermore, two diffe- rent cDNAs could be obtained for each of VT and IT precursors in the white sucker and the chum salmon. Chum salmon precursors were named as provasotocin-I (proVT-I) and II (proVT-II), and proisotocin-I (proIT-I) and II (proIT-II) [12]. Pairs of genes encoding certain hormones were also found in salmonid fish, e.g. two genes for melanin-concentrating hormone in the chum sal- mon [14]. The most important mechanism for generating such pairs of genes is probably genome duplication. Thereafter, the duplicate genes may diverge and individually differentiate. However, there is no detailed study at the nucleotide level about changes of duplicate genes formed by genome duplication. Investigation of the diver- gence and differentiation of salmonid proVT and prolIT genes is thus important to clarify evolution- ary changes of neurohypophysial hormone precur- sor genes in the tetraploid fish. It is also indispen- a)vasotocin amino acids M. Suzuki, S. Hyopo AND A. URANO sable for general understanding of gene evolution. In the present study, masu salmon was used as an experimental species to clarify the extent of diver- gence and differentiation within the same genus. We could clone and determine only single proVT-I and proIT-I cDNAs from the hypothalamus of masu salmon (Oncorhynchus masou). Northern blot analysis demonstrated that the genes for proVT-II and prolIT-II were not expressed, although Southern blot analysis showed the presence of genes for proVT-II and prolIT-II. MATERIALS AND METHODS Construction of cDNA library, cloning and se- quence analysis The hypothalami were collected from 27 mature female masu salmons. Total RNA was extracted with a RNA extraction kit (Amersham), and poly(A)*RNA was separated through an oli- go(dT)-cellulose column using a mRNA purifica- tion kit (Pharmacia). Complementary DNA was prepared with a cDNA synthesis system plus Cys-Tyr-Ile-Gln-Asn-Cys-Pro-Arg-Gly A mRNA 5'-UGC-UAC-AUC-CAA-AAC-UG -3' UU UG aT oligo-VT 3'-ACG-ATG-TAG-GTT-TTG-AC -5' A b)isotocin amino acids codons: csIT-I csIT-II oligo-IT Fic. 1. A A). C.. > A Cys-Tyr-Ile-Ser-Asn-Cys-Pro-Ile-Gly TGC -TAC-ATC-TCC-AAC -TGC-CCC-ATA-GG TGC -TAC-ATC-TCC-AAC-TGT-CCC-ATC-GG 5' -TGC-TAC-ATC-TCC-AAC-TGT-CCC-ATA-GG -3' G G C Synthetic oligonucleotides for the screening of VT and IT coding sequences. a) Oligo-VT is a mixture of 48 oligonucleotides (17-mer) which are complementary to all putative mRNA sequences predicted from the amino acid sequence of VT (1-6). b) Oligo-IT is a mixture of 8 oligonucleotides (26-mer) which were synthesized by considering codon usages in the IT region of chum salmon proIT cDNAs. Vasotocin and Isotocin Precursor cDNAs 159 (Amersham). A cDNA library was constructed from 50 ng of cDNA and 500 ng of Agtl0 arms using a cDNA cloning system Agt10 (Amersham). The cDNA library yielded 1.1X10° plaques of 99% recombinants. A total of 9.8 10° transformants was screened by plaque hybridizations. Probes were (1) three cDNAs encoding chum salmon provasotocin-I (cs- proVT-I), provasotocin-II (cs-proVT-II) and proisotocin-II (cs-proIT-II); and (2) two synthetic oligonucleotide mixtures designated as oligo-VT and oligo-IT (Fig. 1). Oligo-VT contained all nucleotide sequences which are complementary to putative mRNAs predicted from the amino acid sequence of VT (1-6), while oligo-IT contained consensus sequences from IT region of chum salmon proIT cDNAs. The cDNA probes were synthesized by a random priming method using a multiprime DNA labeling system and [a-*°P]dATP (Amersham). The oligonucleotide probes were 3’-end-labeled with an oligonucleotide 3’-end labeling system (DuPont/NEN) and [a-*°P]ddATP (Amersham). Hybridization solution contains 6 x SSC (1XSSC contains 150 mM NaCl and 15 mM sodium citrate), 0.1% SDS, 1xDenhardt’s rea- gent, and 100 ug/ml denatured, sonicated calf thymus DNA. Hybridization was performed at 55°C for the cDNA probes, at 35°C for the oligo- VT probe and 50°C for the oligo-IT probe. Post- hybridization washing was carried out twice in 1 x SSC containing 0.1% SDS at 50°C for the cDNA probes, while filters were washed twice in 6x SSC containing 0.1% SDS at 35°C for the oligo-VT probe and at 50°C for the oligo-IT probe. Inserts from positive clones were subcloned into Bluescript plasmid (stratagene), and then nu- cleotide sequences were determined by the dideoxy chain-termination method [15]. Nu- cleotide and amino acid sequences were compared by use of a GENETYX genetic information pro- cessing software package (Software Development Co., Ltd.). Southern and Northern blot analyses Genomic DNA of masu salmon was extracted from a single salmon liver as described by Maniatis et al. [16], and was digested separately with each of restriction enzymes, EcoR I, Hind III and Pst I. The resulting fragments electrophoresed through a 0.6% agarose gel and then transferred to a Hybond-N membrane (Amersham) according to the manufacturer’s instructions. Total RNAs (20 pg) extracted from the brain, liver, and kidney of masu salmon were electrophoresed through a 1% agarose/formaldehyde gel, and also transferred to a Hybond-N membrane. The method for labeling cDNA probes and the conditions of hybridization were the same as those in screening procedure with the cDNA probes, except that cDNA probes were generated from cDNAs encoding masu salmon proVT-I (ms-proVT-I), proIT-I (ms-proIT-I), chum salmon proVT-I (cs-proVT-I) and II (cs- proVT-II), and prolIT-I (cs-proIT-I) and II (cs- proIT-II). After hybridization, the filters were washed twice in 1 XSSC cotaining 0.1% SDS at 60°C, and were exposed to x-ray film at —80°C. Thereafter, the filters were washed twice in more stringent condition, that is, in 0.1 x SSC containing 0.1% SDS at 60°C, and again subjected to auto- radiography. Estimation of evolutionary relationship Evolutionary relationshps among _ neuro- hypophysial hormone precursor genes were statis- tically estimated by the method of Miyata et al [17] using Genetyx, genetic information processing software (Software Development Co. Ltd.). The number of substitutions per nucleotide site was calculated in terms of synonymous (Ks) and non- synonmous substitution (Kn) in the coding region. We first compared masu salmon proVT-I with proIT-I cDNAs. The presence of three exons in proVP and proOT genes of mammals [3] and in proVT genes of white sucker [10] was considered for the calculation, so that the nucleotide se- quences of masu salmon proVT-I and prolT-I cDNAs were divided into three regions: (1) region A encoding a signal peptide, hormone and amino (N)-terminal portion of neurophysin (top segment in Fig. 3); (2) region B encoding the central por- tion of neurophysin (middle segment); and (3) region C encoding the carboxyl (C)-terminal por- tion of neurophysin (bottom segment) [12]. Then, we compared proVT-I cDNAs and proIT-I cDNAs among masu salmon (ms), chum salmon (cs) and white sucker (ws). Ks was corrected (“Ks) for the 160 M. Suzuki, S. Hyopo AND A. URANO effect of multiple hits at a single site. Evolutionary distance, represented by “Ks, is directly prop- ortionate to divergence time between homologous genes and also between speices that are to be compared. Therefore, the rate of synonymous substitution per site per year (Vv) was estimated by the following formula: — (“Ks + “Ksit) /4T = (“Ksy¢ + “Ksit) / (4x 1.0.x 108) where “Ks,, is “Ks between ms- and ws-proVT-I cDNAs, ‘Ks;, is ‘Ks between ms- and ws-prolT-I cDNAs, and T is divergence time between masu salmon and white sucker. In this study, we adopted 1.010° years as T according to fossil record [18] and an isozyme study [19]. By use of the v, we estimated divergence time for proVT-I genes and for proIT-I genes, between masu salmon and chum salmon. RESULTS Cloning of cDNAs encoding VT and IT precursors We obtained 6 positive clones by screening 5.8 x 10° transformants with the **P-labeled cDNAs for cs-proVT-I and cs-proVT-II. The analyses of nuc- leotide sequences of the positive clones showed that 3 clones contained an identical VT-specific segeunce, while 1 clone contained an IT-specific sequence. These clones were designated as ms- proVT-I and ms-proIT-I cDNAs, respectively, be- cause the homology of the proVT clones to cs- proVT-I cDNA (80.5% nucleotide sequence identity) is higher than that to cs-proVT-II cDNA (48.9%), and the homology of the proIT-I clone to cs-proIT-I cDNA (97.6%) is higher than that to cs-proIT-II cDNA (62.6%). The nucleotide se- quences of ms-proVT-I and ms-proIT-I cDNAs and the deduced amino acid sequences are shown in Fig. 2. Since the presence of pairs of cDNAs for both VT and IT precursors were expected, 4.0 x 10° transformants were rescreened to obtain ms- proVT-II and ms-proIT-II clones with oligo-VT, oligo-IT, and cDNAs for cs-proVT-II and cs- proIT-II. Nevertheless, the sequence of interest was not found in any clones, although we obtained additional two ms-proVT-I and one ms-proIT-I clones. | Vasotocin precursor The ms-proVT-I is composed of 155 amino acid residues and contains a signal peptide, VT and a neurophysin (VT-NP) which is connected to the hormone by Gly-Lys-Arg sequence. The signal peptide contains a high proportion of hydrophobic amino acids. The Gly-Lys-Arg that follows VT may serve as a signal for proteolytic processing and C-terminal amidation [20]. The VT-NP is cystein- rich and contains a highly conserved portion at its center, while the C-terminal of VT-NP includes a leucin-rich core segment and shows remarkable similarity to C-terminal of amphibian VT neurophysin and copeptin of mammalian VP pre- cursors, except for a lack of glycosylation site (Fig. 3). The arginine residue at position 103 may be involved in the processing of the precursor because ° in most mammals the homologous arginine residue in vasopressin precursor is a processing signal between neurophysin and copeptin, although this is not true of amphibian VT precursors [21, 22]. Isotocin precursor The ms-proIT-I consists of 159 amino acid residues. We predicted that the initiation site for translation of ms-proIT-I is ATG at positions —66 to —64 rather than at positions —60 to —58, because CATGGCT at position —67 to —61 is the same as the consensus sequence for initiation sites of other eukaryotic genes [3, 23]. The ms-proIT-I contains a signal peptide, IT, and a neurophysin (IT-NP), as ms-proVT-I. Isotocin is also con- nected to IT-NP by Gly-Lys-Arg sequence. Although the central portion of the neurophysin is Fic. 2. Nucleotide sequences of cDNAs encoding ms-proVT-I and ms-proIT-I precursors and deduced amino acid sequences of the precursors. residue in the coding region for hormones. hormone as 1. are indicated by hyphens. Nucleotide sites are numbered in the direction from 5’ to 3’, beginning with the first The amino acid residues are numbered with the first residue (Cys) of Identical nucleotides and amino acids are indicated by colons and asterisks, respectively. Gaps The AATAAA sequence in the 3’ untranslated region is underlined. Vasotocin and Isotocin Precursor cDNAs 161 provasotocin-I GGAGTTCAGTTGTAGCCGTAGCCTACAGTACAAATTGGACGAAGCACTTTAGACGGAACAAG -61 proisotocin-I CAAATTCGCAGATCATCCTCACCAAAGCCTCAACCTCAACAC -67 Shibgriaulte PS pitesildie Se he ee ee ee ae eee Met Pro Asp Ser Thr Ile Pro Leu Leu Cys Val Leu Gly Leu Leu Ala Leu Ser Ser -2 === S55 NIG, CGA GAME AG Ie NCAT PILI C/N (Cink Cawley We, Gale WCAG: Ee, (GANG (CANE (Ce? (CAG SANG INGAR 2 PCC omoAlG iin CCCeACC ICA GIG 1h CC CC En iC1G Cre TTC Cle, CTA TCT GTA TGC ACT =2 Met Ala * Phe Gly Thr Ser Val Ser Ala * Jeet = 8) Plea a5 wm Sere Wall (Oy Ware 22 — | — HW Vasotocin —H_, ane ce Ala Cys Tyr Ile Gln Asn Cys Pro Arg Gly Gly lys Arg Ser Phe Pro Asp Leu Lys - Arg 19 GCG TGC TAC ATC CAG AAC TGT CCG CGA GGC GGG AAG CGC TCT TTT CCT GAT CTG AAA --- AGA 57 GCC TGC TAC ATC TCC AAC TGC CCC ATA GGA GGC AAG AGA TCA GCC CTA GCT TTC CCA TCC AGA 60 * * * * Ser oo “Ile * ar * Ala Leu Ala Phe Pro Ser * 20 1 —WW [sotocin —_! Pro Cys Met Ser Cys Gly Pro Gly Asn Arg Gly Leu Cys Phe Gly Pro Ser Ile Cys Cys Gly 40 CCG TGC ATG TCA TGT GGC CCT GGA AAC CGG GGC CTC TGC TTT GGC CCC AGT ATC TGC TGT GGG 120 AAG GC ATG TCA TGT GGC CCC GGG GAC AGG GGT CGC TGC TIT GGC CCC AAT ATC TGC TGT GGG 123 os ws Ws ws vs ws a= NS p * as NGS g Ws ws ws 7» Agim = vs * Ar g 41 Glu Gly Met Gly Cys Tyr Met Gly Ser Pro Glu Ala Ala Ser Cys Val Glu Glu Asn Tyr Leu 61 GAA GGG ATG GGT TGC TAC ATG GGC TCC CCA GAG GCA GCT AGT TGT GTG GAG GAG AAC TAC CTG 183 GAG GGG ATG GGC TGC TAC GTG GGC TCT CCA GAG GCA GCT GGC TGC GTG GAG GAG AAC TAC CTG 186 w * * we vw WF Val rh ¥ Ww We we ¥ Gal y WF vw v* * * We we —_ $Y. Neurrophysin 2 ATd-AWl_AWAW Thr Ser Pro Cys Glu Val Gly Gly Arg Val Cys Gly Ser Glu Glu Gly His Cys Ala Ala Pro 82 ACC TCT CCC TGT GAG GTC GGA GGA AGA GTG TGT GGG TCT GAG GAG GGA CAC TGT GCT GCA CCT 246 ccc TCC CCC TGT GAG GTC GGA GGA AGA GTG TGT GGC TCT GAG GAG GGA CGT TGT GCT GCA CCG 249 W" vs wW vs W %* Ww ws ry w vs Ws vs vw w Ar g rh We we We 83 Gly Val Cys Cys Asp Ala Glu Ser Cys Leu Leu Asp Ser Asp Cys Leu Asp Asp Ser Lys Arg 103 GGG GTG TGC TGT GAT GCT GAG AGT TGT CTA CTG GAC TCA GAC TGC CTA GAC GAC AGT AAA CGT 309 GGG ATC TGC TGT GAC GTG GAG GGT TGT AGT ATT GAC CAA TCC TGT ACT GAG GAG GAT GAA GCT 312 a7 ILIL@) as * we Wall €lgy ~ Sere Ie w Clin Sere <2) Alas Cll Gili Ws Gil Miler OVA Gln Pro Pro Ser Glu Gln Tyr Ser Ser Leu Met Glu Gly Leu Ala Gly Asp Leu Leu Gln Trp 124 CAG CCA CCC AGT GAG CAG TAC AGT TCC TTG ATG GAA GGT TTG GCA GGA GAT CTG CTG CAG TGG 372 GAA TAC ATA AGC CAA TCA GTG AGC AGT --- --- --- AGC CAT GGC CAT GAT CTG CTG ATG AAG 366 Clive <>“ GilnsSer Val * = = a) Sere bis) Gilly Weiss 2 ee Met ths 122 ee ele a ee Met Leu His Ala Thr Arg Arg Glu Arg Pro Gln end 35 ATG CTG CAT GCC ACC AGA AGG GAG AGA CCT CAG --- --- --- --- TAA CAAACCACTGGCCAGCCCT 427 CTT CTG AAC ATG ATT AGC CAC ACC CCT CCC CAC AGA GTC CAC AAA TAA AACAGCCTCTAATTCAGGG 433 Leu * Asn Met Ile Ser His Thr Pro * His Arg Val His Lys end NS CACCAAAACACACACCCAGAATAGCACCTAGATCAGTTTCACATGCACTACTACTGCAAAAAACCTACACAGCATACACACAC 510 GAGTTGAAAGACAGTGTAAAAAATCTGTCACATCTCAATGTCAGAGGCATACCTATGTACATACATTTTGTACAGAAGACAGA 516 ATCATACACATACAGCAGGACACAGGAAAGGAAGAGTGGGTTTGCTACATAAGAAAAGCAATCAGCTCTAGTACATTCATATT 593 CATAGATGTAAAGGATTAGTTTTGCTTGTAACTGTTTGTGTAATCGCTTGTGTTTCAAGAGTCAAATAAAGCTTACTGTAC SS) 7 TACTGTGCTTACAGTAGCCTTCAGTAAGTTTTAATTTCTAAGGCTTCAACTCAAATATGCATTGTACAATCCACTAGGGGTGG 676 AAGGAATGTAAATATGTAGCAAATAAATATTTTCTTGCACTATT 720 162 M. Suzuki, S. Hyopo AND A. URANO r- Signal peptide — -Hormone, —_ h-provasopressin (h-proVP) MPDT-MLPACFPGLLAFSSA CYFQNCPRG GKR AMCDL-ELRQ * ated odedede dedede deve dedededededs vedere vs ids t-provasotocin (t-proVT) TARO RCa eee CYIQNCPRG GKR SYPDT-AVRQ we deddededodeds dededededededededs dededs de dss ats cs-provasotocin-I (cs-proVT-I) uPYSTFQLINVEGLLALSSA SEE a GKR SFPDL-P- ag ms-provasotocin-I (ms-proVT-1) MPDSTIPLLCVLGLLALSSA CYTONCERG ¢ ea SFPDL-K-RP ms-proisotocin-I (ms-proIT-I1) "AMFGTSVSALCLLFLLSVCTA CYISNCPIG GKR SALAF-PSRK Gs -prolsotocin=is) (Cessprol l=) MAMEGTSVSALCLLELLSVCTA ( CYISNCPIG GKR SALAF- PSRK t-promesotocin (t-proMT) HMSYTAL- AVIFFGWLALSSA ema GKR SVIDFNDVRK h-prooxytocin (h-proOT ) MAGPSL-ACCLLGLLALTSA CYIONCPLG ¢ GKR 2 AAPDL-DVRK Neurophysin Conservative region IN=jOreO Wie CLPCGPGGKGRCLGPS ICCADELGCF VGTAEALRCQEENYLPSPCQSGQKACGS - nntitas cnc t-proVT ATBCEPENRGNCEGPNI CCURTLEC Grey LROVELT HSB CRICe ERSTE GGRCAAPGYCCSDD cs-provT-I ENS CEP GDRGRCREPNT CEGEGHGCYMGSPEMAGEVEENILEC REE AGERNG Ee MrRe rennin ns-proVT-I | CMSCGPGNRGLCFGPSICCGEGMGCYNGSPEAASCVEENYLTSPCEVGGRVCGSEEGHCAAPGVCCDAE ms-prolT-I | CMSCGPGDRGRCFGPNICCGEGMGCYVGSPEAAGCVEENYLPSPCEVGGRVCGSEEGRCAAPGICCDVE EE aCe CPR DDE DID IRE EE ET TETaE NOOR TNC ON POU TLOUR NTT NCON NC DOO TODO KPO DA eee oa esiprolel sii CHIE GIIGIING NCIC WL CEGMG CGS MAO LENE ECAGGR S222 uns aes Ok ESpToMa CLPCGPRNKGHCPCENTC EGRET EC ROTTETLECOPENRTIBS ECHS ERIE ERRCE TElteReaeen ve ke te “& sc “& sls Pid & se we se se se ss Pid sey nis se rid sje ve Pid ‘& se “& we we ss Pind Pind ss sb * se ss * se * Pid si le * ste we sk we sls h-proOT CLPCGPGGKGRCFGPNICCAEELGCFVGTAEALRCQEENY LPS PCQSGQKACG-SGGRCA-LGLCCSPD Copeptin aaa Var Sat nh te ae Cit ia ee ee: Gta hh. LiL h-proVP ct ee R ASD- Se Nr aur ae wei EEO VE TCVVDSSCLDEDSERR Rea Vanes EQNYTOMDGSASDLLLRLMEMANRQQQSKHQFY. EsiapEOVl—s “SCVEDED Cl ED SKS aR qSPSEQNAAINGCLAGDLL- “RAULEIS -ATSRGRPO ms-proVT-I SCLLDSDCLD-DSF-- R QPPSEQYSSLMEGLAGDLLQWML- ATRRERPQ ms-prolT-I GCSIDQSCTEEDE--- A EYISQSVSS~SHG--HDLLMKLLNMISHTPPHRVHK mln ata a le sc we Je sc ve sc Js se ss we ale ale nlenlanlanlantsatantants alentantas als le mls ls ls ls ls wl he ls le ls ls ts wl ls wl lc wl ls aay cay AAA AAMAS ASAI TASCA SAY PAM AMAM AMAA AMAA AM AMAIA AMAIA IAS A AIA SAY Espo ep ee A EYISQSVSS-SHG- -HDLLMKLLNMISHTPPHRVHK Esp roll, Se PAE OS -EQDSVES h-proOT GCHADPAC- -DAEATFSOR Fic. 3. Comparison of the amino acid sequences among the precursors of masu salmon VT-I and IT-I, chum salmon VT-I and IT-I, toad VT and MT, human VP and OT. Gaps indicated by hyphens have been introduced to maximize homology. Identical amino acids are indicated by asterisks. The conservative regions of neurophysins are enclosed by a frame. Note that the C-terminals of salmonid VT and IT neurophysins include a leucin-rich core segment. Vasotocin and Isotocin Precursor cDNAs 163 highly homologous to those of amphibian and mammalian neurophysins, the C-terminal, like those in the chum salmon and the white sucker, extends approximately an additional 30 amino acid residues. As C-terminal of VI-NP, this elongated terminal lacks a glycosylation site but includes a leucin-rich core segment (Fig. 3). Southern and Norhtern bolt analyses Genomic DNA obtained from a single masu salmon liver was digested separately with EcoR I, 1 EcoR:| a) Vasotocin 2: Hind Ill ms: | cs: | cs: Il 3) Pst | 203 1e2e3 1 2 8 I. on Tol gol ee ~ # Oo. 0) e e ) ® +5 ?. c kb b) Vasotocin ms: | cs: | cs: Il 2S 123s 1 23 ww @ @ % . @ = or @ @ eS a -20 @Q Oo) o =5 S kb c) Isotocin ms: | esa cs: Il 1 Bs) 128 i273 sw ad ¥ : we: "o i . F trs9320 ad @ ® —5 oo @ = = kb Fic. 4. Southern blot analysis of masu salmon proVT (a, b) and proIT (c) genes in a single masu salmon genome. Five micrograms of DNA from the liver was digested separately with restriction enzymes EcoR I (lane 1), Hind III (lane 2) and Pst I (lane 3) and electrophoresed. Probes used in hybridization are shown above the lane numbers. Filter washing was performed in 1 xSSC-0.1% SDS (a) and in 0.1 x SSC-0.1% SDS (b, c). Positions of size markers, indicated in kb, were obtained by use of Hind III and EcoR I double digests of lambda DNA. Hind Ill and Pst I, and then hybridized with ms-proVT-I, cs-proVT-I, cs-proVT-II, ms-proIT- I, cs-proIT-I and cs-proIT-II probes. In the auto- radiogram of filters washed in 1 x SSC-0.1% SDS, the band length pattern yielded by ms-proVT-I cDNA probe was obviously different from that by cs-proVT-II probe but identical with that obtained by cs-proVT-I probe. However, the intensities of hybridization signals with ms-proVT-I probe were somewhat different from those with cs-proVT-I probe (Fig. 4a). After more stringent washing of the same filters in 0.1xSSC-0.1% SDS, the band pattern by cs-proVT-I probe was different from that by ms-proVT-I probe (Fig. 4b). On the other hand, such difference was not found between the patterns yielded by cs- and ms-proIT-I probes after 1: Hypothalamus 2: Liver a) Vasotocin 3: Kidney ms: | CSHl cs: ll eens | 28 |) 28 -origin | mle kb b) Isotocin ms: | Cs: | CSz | LZ Ss 25S la2e3 ars —origin ] @ =} 0) kb Fic. 5. Northern blot analysis for masu salmon proVT mRNAs (a) and proIT mRNAs (b). Twenty micro- grams of total RNAs from the hypothalamus (lane 1), liver (lane 2) and kidney (lane 3) were elec- trophoresed and hybridized according to the proce- dure described in the text. Probes used in hybridiza- tion are shown above the lane numbers. 164 M. Suzuki, S. Hyopo AND A. URANO filter washing in 0.1xSSC-0.1% SDS, although cs-proIT-II probe detected completely different band pattern (Fig. 4c). These results strongly indicate that, in addition to genes for the analyzed proVT-I and proIT-I mRNAs, an individual masu salmon has proVT-II gene, proIT-II gene and another proVT-I gene which is considerably ho- mologous to cs-proVT-I gene. Tissue specific expression of ms-proVT and ms- proIT genes was investigated by Northern blot analysis. The ms-proVT-I and cs-proVT-I probes hybridized with a single RNA species of about 900 bases in masu salmon hypothalami, while cs- proVT-II probe did not yield any hybridization signals (Fig. 5a). The proIT probes showed the same trend, i.e. only ms-proIT-I and cs-proIT-I probes detected one band approximately 800 bases in the lane for the hypothalamus (Fig. 5b). RNAs isolated from the kidney and liver of masu salmon did not hybridize with any proVT and prolT probes. These results may explain the fact that we were able to clone ms-proVT-I and ms-prolIT-I cDNAs, but unable to obtain ms-proVT-II and ms-proIT-II clones. Statistical analysis of nucleotide substitution On the basis of nucleotide substitution, we esti- mated evolutionary relationships among genes en- coding neurohypophysial hormone precursors. Between masu salmon proVT-I and prolIT-I cDNAs, the number of substitutions per synony- mous site (Ks) and the number of substitutions per nonsynonymous site (Kn) calculated for the cen- tral portion of neurophysin (region B) are con- siderably lower than those for the other regions (Table 1). Further, we compared proVT-I and prolIT-I cDNAs of masu salmon with homologous ones of chum salmon and white sucker, respectively, as is shown in Table 2. Between masu salmon and white sucker, evolutionary distance for proVT-I genes was almost the same as that for prolIT-I genes. Therefore, based on these values, we estimated that the mean rate of synonymous sub- stitutions per site per year for proVT and proIT genes was 8.410 ” in teleosts. Between masu salmon and chum salmon, evolutionary distance for proVT-I genes was about 6-fold larger than TABLE 1. The numbers of synonymous (K,) and nonsynonymous (K,,) substitutions per nucleotide site for coding region between masu salmon (ms) proVT-I and proIT-I cDNAs. Region A en- codes a signal peptide, hormone and N-terminal part of neurophysin; region B, the central portion of neurophysin; and region C, C-terminal part of neurophysin Sequence compared K, K, ms-proVT-I vs. ms-proIT-I Region A 0.829 0.343 Region B 0.331 0.060 Region C 0.720 0.528 TABLE 2. Evolutionary distance (“K,) and diver- gence time (T). Interspecies comparison be- tween masu salmon (ms) and chum salmon (cs), and between masu salmon and white sucker (ws) Sequence compared TK T (myr) ms-proVT-I vs. cs-proVT-I 0.350 ZA ms-proIT-I vs. cs-proIT-I 0.058 35) ms-proVT-I vs. ws-proVT-I 1.75 — 100° 1.61 * This value was estimated according to fossil record [18] and an isozyme study [19]. ms-proIT-I vs. ws-proIT-I that for proIT-I genes. Accordingly, divergence time between ms-proVT-I and cs-proVT-I genes was estimated as 21 myr ago, while that between ms-proIT-I and cs-proIT-I genes as 3.5 myr ago. DISCUSSION In the present study, the primary structures of masu salmon neurohypophysial hormone precur- sors, proVT-I and proIT-I, were deduced from the nucleotide sequences of cDNAs. These precursors consist of a signal peptide, hormone and neurophysin (NP). The VT-NP and IT-NP extend about 30 amino acid residues at the C-terminal beyond those described in mammals, so that they are almost of the same length. The central por- tions of the NPs are particularly similar to each other and the extended C-terminals contain a leucine-rich core segment. Overall high homology between ms-proVT-I and ms-proIT-I supports the Vasotocin and Isotocin Precursor cDNAs 165 hypothesis that an ancestral molecule diverged into VT and IT, as is the case with neuro- hypophysial hormone precursors in the chum sal- mon and the white sucker [9, 12]. We were unable to obtain any clones of masu salmon proVT-II and proIT-II cDNAs. Northern blot analysis did not detect the presence of proVT- II and proIT-II mRNAs. However, Southern blot analysis indicated the presence of proVT-II and proIT-II genes in the masu salmon, as well as in the chum salmon [9, 12] and the white sucker [8, 10]. It is not clear why proVT-II and prolIT-II genes were not expressed in the masu salmon, but there are several plausible explanations. Expressions of proVT-II and proIT-II genes may be physiological- ly repressed in the masu salmon. Alternatively, an occurrence of mutation in 5’-upstream region and/ or the coding region may result in the lack of expression. The latter case is supported by the following reports. The gene for vasopressin pre- cursor (proVP) suffered deletion of a single G residue in exon B in the Brattleboro rat [24]. This mutant proVP gene was expressed at a markedly reduced level in the hypothalamus [25]. Furhter, the chum salmon has proVT-II mRNA in which a codon for cystein in proVT-I mRNA is altered to a stop one. The expression level of this mutant proVT-II mRNA in the hypothalamus was con- siderably low [12]. The comparison of masu salmon proVT-I and proIT-I cDNAs in terms of evolutionary rela- tionship showed that, in region B encoding the highly conserved portion of NP, the number of substitutions per nonsynonymous site (Kn) is con- siderably smaller than those in the other regions. Furthermore, the number of substitutions per synonymous site (Ks) in region B is much smaller than those in the other regions. These results suggest an occurrence of a gene conversion event encompassing region B of masu salmon proVT-I and proIT-I genes. The possibility of similar gene conversion events was pointed out in the chum salmon and mammals [12, 26, 27]. The exon encoding the central portion of NP may incline to suffer gene conversion irrespective of species. In this study, we estimated that ms-proVT-I and cs-proVT-I genes diverged about 21 myr ago, although ms-proIT-I and cs-proIT-I genes di- verged about 3.5 myr ago (Table 2). An isozyme study suggested that divergence time between masu salmon and chum salmon was about 3.0 myr ago [28]. This divergence time is consistent with that between ms-proIT-I and cs-proIT-I genes, but not with ms-proVT-I and cs-proVT-I genes. It is almost impossible to assume that the rate of molecular evolution of proVT-I genes is very rapid in salmonid, because between masu salmon and white sucker, evolutionary distance for proVT-I genes is comparable to that for proIT-I genes. A possible explanation for the above temporal dis- crepancy is that proVT-I gene duplicated in a common ancestor of masu salmon and chum salmon. Southern blot analysis showed the pre- sence of two types of proVT-I genes in the masu salmon genome, i.e. the present ms-proVT-I gene and another proVT-I gene which is highly homolo- gous to cs-proVT-I gene. The latter untranscribed proVT-I gene and cs-proVT-I gene may have diverged when the masu salmon and the chum salmon diverged. In conclusion, the present study suggested that the masu salmon has at least five genes encoding neurohypophysial hormone precursors, among which some regulatory differentiation occurred between proVT-I and II genes, and between proIT-I and II genes. These regulatory differentia- tion resulted in the lack of expression of proVT-II and proIT-II genes. ACKNOWLEDGMENTS We are indebted to Kumagaya Branch of Saitama Prefectural Fisheries Experimental Station for supplying sufficient fresh samples and to Dr. M. Sato, Kyowa Hakko Kogyo, for providing oligo-vasotocin DNA. We express our gratitude to Professor T. Hirano, Ocean Research Institutue, University of Tokyo, for his en- couragement and to Mr. S. Sakata, Nippon Oil Co., Ltd., for his advice on various aspects of this study. This study was supported in part by grant-in-aids from the Fisheries Agency, and the Ministry of Education, Cul- ture and Science, Japan. REFERENCES 1 Gorbman, A., Dickoff, W. W., Vigna, S. R., Clark, N. B. and Ralph, C. L. (1982) Comparative Endo- 10 11 12 166 crinology. Wiley, New York, pp. 95-116. Acher, R. (1985) Biosynthesis, processing, and evolution of neurohypophysial hormone precursors. In “Neurosecretion and the _ biology of neuropeptides”. Ed. by H. Kobayashi, H. A. Bern, and A. Urano, Japan Sci. Sco. Press, Toky/Sprin- ger, Berlin, pp. 11-25. Richter, D. (1987) Oxytocin and vasopressin genes: expression and structure. In “Molecular cloning of hormone genes”. Ed. by J. F. Harbener, Humana Press, New Jersey, pp. 173-206. Land, H., Schutz, G., Schmale, H. and Richter, D. (1982) Nucleotide sequence of cloned cDNA encod- ing bovine arginine vasopressin-neurophysin II pre- cursor. Nature (London), 295: 299-303. Land, H., Grez, M., Ruppert, S. Schmale, H., Rehbein, M., Richter, D. and Schutz, G. (1983) Deduced amino acid sequence from the bovine oxytocin-neurophysin I precursor cDNA. Nature (London), 302: 342-344. Sausville, E., Carney, D. and Battey, J. (1985) The human vasopressin gene is linked to the oxytocin gene and is selectively expressed in a cultured lung cancer cell line. J. Biol. Chem., 260: 10236-10241. Nojiri, H., Ishida, I., Miyashita, E., Sato, M., Urano, A. and Deguchi, T. (1987) Cloning and sequence analysis of cDNAs for neurohypophysial hormones vasotocin and mesotocin for the hypo- thalamus of toad, Bufo japonicus. Proc. Natl. Acad. Sci. USA, 84: 3043-3046. Figueroa, J., Morley, S. D., Heierhorst, J., Krent- ler, C., Lederis, K. and Richter, D. (1989) Two isotocin gene are present in the white sucker Cato- stomus commersoni both lacking introns in their protein coding regions. EMBO. J., 8: 2873-2877. Heierhorst, J., Morley, S. D., Figueroa, J., Krent- ler, C., Lederis, K. and Richter, D. (1989) Vasoto- cin and isotocin precursors from the white sucker, Catostomus commersoni: Cloning and sequence analysis of the cDNAs. Proc. Natl. Acad. Sci. USA, 86: 5242-5246. Morley, S. D:, Schonrock, C., Heierhorst, J., Figueroa, J., Lederis, K. and Richter, D. (1990) Vasotocin genes of the teleost fish Catostomus com- mersoni: gene structure, exon-intron boundary, and hormone precursor organization. Biochemistry, 29: 2506-2511. Heierhorst, J., Mahlmann, S., Morley, S. D., Coe, I. R., Sherwood, N. M. and Richter, D. (1990) Molecular cloning of two distinct vasotocin precur- sor cDNAs from chum salmon (Oncorhynchu keta) suggests an ancient gene duplication. FEBS. Lett., 260: 301-304. Hyodo, S., Kato, Y., Ono, M. and Urano, A. (1991) Cloning and sequence analyses of cDNAs encoding vasotocin and isotocin precursors of chum 3) 14 15 16 17 18 19 20 21 D2 i) 24 25 M. Suzuki, S. Hyopo AND A. URANO salmon, Oncorhynchus keta: evolutionary rela- tionships of neurohypophysial hormone precursors. J. Comp. Physiol. B., 160: 601-608. Ohno, S., Wolf, U. and Atkin, N. B. (1968) Evolu- tion from fish to mammals by gene duplication. Hereditas, 59: 169-187. Takayama, Y., Wada, C., Kawauchi, H. and Ono, M. (1989) Structures of two genes coding for mela- nin-concentrating hormone of chum salmon. Gene, 80: 65-73. Sanger, F., Nicklen, S. and Coulson, A. R. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA, 74: 5463-5467. Maniatis, T., Fritsch, E. F. and Sambrook, J. (1982) Molecular Cloning: a laboratory manual. Cold Spring Harbor laboratory, Cold Spring Harbor, New York. Miyata, T., Hayashida, H., Kikuno, R. and Yasuna- ga, T. (1985) Computer analysis of homology be- tween genes. In “Methods for gene research”. Ed. by M. Takanami, S. Nishimura, and M. Matsumura, Tokyo Kagaku Dojin, Tokyo, pp. 381-425. Harland, W. B., Holland, C. H., House, M. R., Hughes, N. F., Reynolds, A. B., Rudwick, M. J.S., Satterthwaite, G. E., Tarlo, L. B. H. and Willey, E. C. (1967) The Fossil Record: a symposium with documentation. Geological Society of London, Lon- don, pp. 655-661. Lim, S. T., Kay, R. M. and Bailey, G. S. (1975) Lactate dehyrogenase isozymes of salmonid fish. J. Bio. Chem, 250: 1790-1800. Brownstein, M. J., Russell, J. T. and Gainer, H. (1980) Synthesis, Transport, and Release of Pos- terior Pituitary Hormones. Science, 207: 373-378. Michel, G., Chauvet, J., Chauvet, M. T. and Acher, R. (1987) One-step processing of the amphibian vasotocin precursor: structure of a frog (Rana esculenta) “big” neurophysin. Biochem. Biophys. Res. Commun., 149: 538-544. Chauvet, J., Michel, G., Chauvet, M. T. and Acher, R. (1988) An amphibian two-domain ‘big’ neurophysin: conformational homology with the mammalian MSEL-neurophysin/copeptin interme- diate precursor shown by trypsin-Sepharose pro- teolysis. FEBS. Lett., 230: 77-80. Kozak, M. (1981) Possible role of flanking nu- cleotides in recognition of the AUG initiator codon by eukaryotic ribosomes. Nucleic Acids Res., 9: 5233-5252. Schmale, H. and Richter, D. (1984) Single base deletion in the vasopressin gene is the cause of diabetes insipidus in Brattleboro rats. Nature (Lon- don), 308: 705-709. Majzoub, J. A., Pappey, A., Burg, R. and Habener, J. F. (1984) Vasopression gene is expressed at low levels in the hypothalamus of the Brattleboro rat. Vasotocin and Isotocin Precursor cDNAs 167 Proc. Natl. Acad. Sci. USA, 81: 5296-5299. 26 Ivell, R. and Richter, D. (1984) Structure and comparison of the oxytocin and vasopressin genes from rat. Proc. Natl. Acad. Sci. USA, 81: 2006- 2010. 27 Ruppert, S., Scherer, G. and Schutz, G. (1984) 28 Recent gene conversion involving bovine vasopres- sin and oxytocin precursor genes suggested by nu- cleotide sequence. Nature (London), 308: 554-557. Numachi, K. (1984) A study on the divergence and phylogeny of salmonids by isozymes. The Heredity (Japan), 38: 4-11. ay id ou wade fF o eK Wie sien 1 a a ai GPS Be ¥ ‘ ‘ oe — Pk a! 5 Pi ie o t ; Sa i ) , i it Fd *, = vn = ea ; re ¢ i ” aes aN Nb) = Es a x x. 4 ~ eyteie th < ‘ fi Ro e 1 2 s i ZOOLOGICAL SCIENCE 9: 169-174 (1992) © 1992 Zoological Society of Japan In vitro synthesis of ecdysteroid conjugates by tissue extracts of the silkworm, Bombyx mori Susumu Y. TAKAHASHI’, KANAKO OKAMOTO”, HARUYUKI SONOBE’, Mari KAmBa~ and Evi OuNISsHr? ‘Biological Institute, Faculty of Liberal Arts, Yamaguchi University, Yamaguchi 753, Yamaguchi Japan, *Department of Biology, Faculty of Science, Konan University, Higashinada-ku, Kobe 658, Japan, and *Department of Natural Science, Okayama University of Sciecne, 1-1 Ridai-cho, Okayama 700, Japan. ABSTRACT— In order to obtain information about the enzyme system(s) catalyzing the formation of phosphoesters of ecdysteroids, ovarian or fat body extracts of the Bombyx silkworm were incubated with free ecdysteroids (ecdysone, 2-deoxyecdysone or 2,22-dideoxy-20-hydroxyecdysone) and [y- 2p] ATP. It was demonstrated by autoradiography that the reaction products are ecdysteroid phosphoconjugates and that the enzymic activity was localized in the 100,000 x g supernatant fraction of both tissues. Furthermore, stoichiometric analyses of the reaction products, which were synthesized by incubating the ovarian extract with [*>H]ecdysone and [y-**P]ATP, revealed that the reaction product is phosphomonoester of ecdysone. From these results, one of the enzymes involved could be defined to be ATP: ecdysone phosphotransferase. INTRODUCTION In insect eggs and ovaries, the bulk of the ecdysteroids exists as conjugates [1-3]. It has been suggested that the conjugates are the storage forms and that the free forms may exert key roles in oogenesis or embryogenesis [4, 5]. Thus, biosynth- esis and hydrolysis of the conjugates might be critical steps in the regulation of ovarian and embryonic development. However, efforts of many investigators have been directed mostly to- wards the elucidation of the chemical structures of the conjugates, and relatively little attention has been paid to the enzyme systems involved in the synthesis and degradation of the conjugates [6]. In ovaries of the silkworm Bombyx mori, six ecdysteroids accumulate in the free and conju- gated forms [7, 8]. The chemical structures of the conjugated forms have been clarified and shown to be as follows: ecdysone-22-phosphate, 20-hydroxy- Accepted September 9, 1991 Received May 17, 1991 ecdysone-22-phosphate, 2-deoxyecdysone-22-pho- sphate, 2-deoxy-20-hydroxyecdysone-22-phospha- te, 2,22-dideoxy-20-hydroxyecdysone-3-phosphate and bombycosterol-3-phosphate [7, 8]. The levels of the conjugates as well as their free forms change during ovarian development and embryogenesis [9, 10], indicating that the conjugates are actively metabolized and might play a singnificant role in the development of the silkworm. To understand the biological role of the conju- gates, we investigated the occurrence of the en- zymic activity involved in the synthesis of the conjugates using crude preparations of ovaries and fat body tissues from silkworm pupae. MATERIALS AND METHODS Animals Mature pupae of hybrid commercial strains (Kinshu-showa) were used throughout the experi- ments. They were purchased from a local farmer (Ato-cho, Yamaguchi Prefecture, Japan). The 170 S. pupal-adult ecdysis took place 12 days ayo the larval-pupal ecdysis at 25°C. Chemicals Ecdysteroid conjugates were isolated from ma- ture ovaries of Bombyx mori as described pre- viously [5]. a-[23,24-7H(N)] Ecdysone (specific activity, 89Ci/mmol) was purchased from New England Nuclear (Boston, MA, USA). [y-*’P] ATP (specific activity, 4500Ci/ mol) was the pro- duct of ICN Radiochemicals (Irvine, CA, USA). Alkaline phosphatase (calf intestine) was obtained from Boehringer Mannheim (Mannheim, Ger- many). Preparation of tissue extracts Ovaries or fat body tissues from mature pupae were homogenized with 10 volumes of 10mM sodium phosphate buffer (pH 7.5) containing 0.32 M sucrose, 1 mM EDTA, 1mM EGTA and 10 mM 2-mercaptoethanol. The homogenate was centrifuged at 100,000xg for 90min. Both the homogenate and 100,000g supernatant were used for the experiments. The 100,000 x g super- natant was used as the “tissue extract”. Assay of enzyme activity The reaction mixture for the conjugate synthesis is as follows: ecdysteroids (10 ug) or 10 yg of [>H]ecdysone (5 ~Ci), 10 nmol of [y--*P] ATP (10 pCi), 20 mM MgCh, tissue homogenate or extract (100-200 yg of protein) in 100 1 of 10 mM sodim phosphate buffer (pH 7.5). The reaction mixture was incubated at 25°C for 5 to 15 min, and the reaction was stopped by the addition of 4 volumes of ethanol. The mixture was centrifuged 10,000 x g for 10 min at 4°C, and the supernatant was col- lected. The pellet was washed twice with 0.2 ml of 80% ethanol and the supernatant and washings were combined and evaporated to dryness under vacuum. The residue was dissolved in 1 ml of distilled water and applied to Sep-Pak Cig car- tridges (Waters, MS, USA) or Bond Elut Cig cartridges (Varian, CA, USA). After washing with 10 ml of distilled water to remove ATP and any salts, the conjugates were eluted with 3 ml of 40% methanol. Unreacted ecdysteroids remaining in the cartridge could be eluted with 60% metha- Y. TAKAHASHI, K. OKAmoToO et al. nol. The conjugate fraction was concentrated under reduced pressure and subjected to high performance thin layer chromatography (HPTLC) on a precoated silica gel plate (Merck HPTLC plate, 60 Fys4, Art. 5628). The radioactive com- pounds were developed by one of the following solvent systems (chloroform: 96% ethanol=4:1, v/v or ethylacetate : ethanol : water=4:8:1, v/v). Spots were visualized under an ultraviolet lamp (254 nm). After drying, autoradiograms were pre- pared from the plates by exposing them at —70°C to Fuji X-ray RX film with an intensifying screen (DuPont Lightning Plus). In some cases, after developing the films, the located radioactive spots were scraped from the plates and materials were extracted with 80% ethanol and counted in a liquid scintillation counter. Hydrolysis of an ecdysone conjugate by alkaline phosphatase After incubation of the ovarian extract with [°H] ecdysone and [y-*P] ATP, the reaction product was partially purified by HPTLC as previously described. The conjugate was eluted from the HPTLC plate with 80% ethanol and evaporated . under the reduced pressure. The remaining matter was incubated with alkaline phosphatase (1 unit) for 15 min at 30°C. The reaction was stopped by the addition of 4 volumes of ethanol and the mixture was evaporated under reduced pressure. The residual material was analyzed by HPTLC as already described. High performance liquid chromatography (HPLC) of free ecdysteroids The HPLC system (Toyo Soda, Shin-Nanyo, Japan) consisted of two models of CCPE pumps, a model UV 8011 absorbance detector and a model Sic Chromoatocorder 12 data analyzer. Chromato- graphy of a reversed phase Cig column (ODS 120T, Tosoh, 1005 mm, 10 um particle size) was performed using a solvent system of 50% aqueous methanol at a flow rate of 1 ml per min (at 25°C). Other methods Ecdysone titer in the ovarian extract was deter- mined by radioimmunoassay as described by Borst and O’Connor [11]. Proteins were determined Ecdysteroid Phosphotransferase 171 according to the methods of Lowry et al [12]. ATP in the ovarian extract was quantified by the methods of Jaworek et al [13]. RESULTS AND DISCUSSION When the reaction mixtures consisted of the homogenate or extract of the ovaries or fat body tissues from mature female pupae, [y-*°P] ATP and ecdysteroids (2-deoxyecdysone and 2,22- dideoxy-20-hydroxyecdysone), rather conspicuous radiolabeled spots corresponding to the ecdyster- oid conjugates, were detected on the HPTLC plates (Fig. 1, A to C). Phosphate was effectively incorporated into both of the ecdysteroid conju- gates. When the ecdysteroids were omitted from the reaction mixture, the spots corresponding to the radiolabeled conjugates were not detected (Fig. 1, lane 1 in A to C). The spots disappeared when the products were treated with alkaline phosphatase prior to the application on HPTLC (data not shown). These facts seem to present strong evidence that the ovary and fat body have an enzyme system synthesizing the ecdysteroid conjugates. Since the enzymic activity was found in 100,000 x g supernatant fractions comparable to the corresponding homogenates (Fig. 1, lane 3 in A-C), the enzyme system seemed to be present mainly in cytosol. When the reaction was allowed to incubate with [°H] ecdysone, a spot corresponding to the ecdy- sone conjugate could be detected on the HPTLC plate (Fig. 2A). Furthermoer, when the reaction product was hydrolyzed with alkaline phosphatase, the only spot corresponding to free phosphate appeared on the HPTLC plate (Fig. 2B). In order to detect tritium labeled materials, silica gel was scraped from the plate in 5mm width and the radioactive materials were eluted from the gels using 80% ethanol. Radioactivity of the position corresponding to ecdysone was detected (Fig. 2C). The identity of the liberated ecdysteroid with ecdysone was further confirmed with the aid of HPLC analysis. The result is shown in Fig. 3. The retention time of the radioactive peak was com- pletely identical with that of the authentic ecdy- sone. These results show that the terminal phos- phate group of ATP was directly transferred to ecdysone. Next, the stoichiometry of the reaction was examined. In this experiment, the ovarian extract was incubated with [°H] ecdysone and [y-*’P] ATP as described in the Materials and Methods section. We assumed that the content of free ecdysteroids in the ovarian extract is not so high as to affect practically the specific activity of [*H] ecdysone; =E ee ee ieF 0>-@ @ e @ e e —=?p 1 2 3 1 2 3 1 2 3 Fic. 1. Autoradiograms showing the incorporation of phosphate into ecdysteroid conjugates. Tissue ex- tracts or homogenates were incubated in the reac- tion mixture with ecdysteroids and [y-*°P] ATP (10 Ci) for 10 min at 30°C. After termination of the reaction with an addition of ethanol, the super- natants were processed as described in Materials and Methods. The fractions eluted from Cj, cartridges with 40% methanol were applied to HPTLC plates and developed with a solvent system (ethylacetate : ethanol: water=4:8:1 v/v). After drying the plates, autoradiograms were prepared. In (A), homogenate or supernatant of ovaries was incubated with 2-deoxyecdysone. Lane 1, homogenate was incubated without 2-deoxyecdysone; lane 2, homogenate was incubated with 2-deoxyecdysone; lane 3, 100, 000 Xx g supernatant of the homogenate was incubated with 2-deoxyecdysone. (B), homogenate or supernatant of fat body tissues was incubated with 2-deoxyecdysone. Lane 1 homogen- ate was incubated without 2-deoxyecdysone; lane 2, homogenate was incubated with 2-deoxyecdysone; lane 3, 100,000 x g supernatant of the homogenate was incubated with 2-deoxyecdysone. (C), homogenate or supernatant of ovaries was incubated with 2,22-dideoxy-20-hydroxyecdysone. Lane 1, homogenate was incubated without 2,22-dideoxy- 20-hydroxyecdysone; lane 2, homogenate was incu- bated with 2,22-dideoxy-20-hydroxyecdysone; lane 3, 100,000 x g supernatant of the homogenate was incubated with 2,22-dideoxy-20-hydroxyecdysone. O and F denote origin and front, respectively. E and P represent the position of free ecdysteroids and those of ATP and phosphoric acid, respectively. 172. S. Y. TAKAHASHI, K. ww A r A ée i 3 0 F an C lo Sep? = ea oll 0 4 — 8 (cm) DISTANCE FROM ORIGIN Fic. 2. Autoradiograms showing the incorporation of [°?P] into the conjugate and the effect of alkaline phosphatase treatment on the conjugate. Ovaries (1 g) were homogenized with 10 mM phosphate buffer, pH7.5, containing 0.32M sucrose, with 1mM EDTA and 10 mM 2-mercaptoethanol, and centri- fuged at 100,000 x g for 90 min and the supernatant was incubated with [y-**P] ATP in the presence of [PH] ecdysone at 30°C for 10min. The conjugate formed was partially purified by Cis cartridges as described in the text. A part of the partially purified conjugate was applied to HPTLC (A), and the remainder was treated with alkaline phosphatase and the hydrolysate was submitted to HPTLC (B). In HPTLC, radioactive compounds were developed with a solvent system (chloroform :96% ethanol= 4:1, v/v), and located by autoradiography (A, B). After developing the autoradiogram, the HPTLC plate was sectioned in 0.5 mm width, and the mate- rials in the gel were eluted with 80% ethanol. Radioactivity in the eluate was counted in a liquid scintillation counter (C). Abscissa: distance from origin in cm. Ordinate: radioactivity, in cpm. i, cpm [°H]; =), cpm [*P]. whereas, the amount of ATP in the extract might be enough to affect the specific activity of the radioactive ATP. Therefore, we first determined the level of ATP and ecdysone in the crude ovarian extract. The amount of ecdysone was determined by radioimmunoassay after ecdysone was sepa- rated from other ecdysteroids by reverse-phase HPLC [14]. The amount of ecdysone in the ovarian extract, used for the phosphorylation reac- tion mixture, was calculated to be 0.16 pmole OKAMOTO et al. ce e de y y y 2 ise) lo etl = = & 0 i) 10 15 RETENTION TIME (MIN) Fic. 3. HPLC profiles of the free ecdysteroid liberated by enzymatic hydrolysis from the conjugate. The material localized on the HPTLC plate correspond- ing to the conjugate (Fig. 2A) was eluted from the plate and was hydrolyzed by alkaline phosphatase as already described. After evaporation, the residue was dissolved in 50% methanol and subjected to HPLC. Before application to HPLC, authentic ecdysteroids (5 wg each) were added to the sample solution as markers. One ml fractions were col- lected in counting vials and the radioactivity was counted in a liquid scintillation counter. Condi- tions for HPLC were: reverse phase Toyo Soda, ° TSKgel ODS-120T (0.4625 cm), solvent system 55% aqueous methanol, pressure 120 kg/cm, flow rate | ml/min. Elution points of marker ecdyster- oids are shown by arrows. ce, conjugated ecdyster- oids; e, 20-hydroxy ecdysone; ec, ecdysone; de, 2-deoxy-20-hydroxy ecdysone. ecdysone equivalent, and that of ATP was calcu- lated to be 0.079 umole. The results clearly show that the content of ecdysone in the extract was negligible. The specific activity of [y--°-P] ATP in the reaction mixture could be calculated from the above value. After the conjugate was separated from unreacted [°H] ecdysone and ATP by HPTLC, the amounts of ecdysone and phosphate in the conjugate were determined from their radioactivity and specific activities (7H for ecdy- sone and *’P for phosphate). The relative ratio of phosphate to ecdysone was calculated to be 1.27+ 0.06 (n=3), indicating that one mole of phosphate might be transferred to one mole of ecdysone. The enzyme involved in the formation of ecdy- steroid conjugates required ATP and Mg**. From the stoichiometry of the reaction, it was shown that Ecdysteroid Phosphotransferase WB, the enzyme catalyzes the transfer of one mole of phosphate to a mole of ecdysone and that the gamma _ phosphate group of ATP was directly transferred to ecdysone. These facts suggest that the enzyme involved in the formation of ecdyster- oid conjugates can be tentatively characterized as ATP: ecdysteroid phosphotransferase. Ecdysteroids have hydroxyl groups on the C-2, C-3, C-22 and C-25 positions which might be involved in the phosphate ester formation [15, 16]. In Bombyx silkworm, it has been demonstrated that all six conjugated ecdysteroids in the ovary are phosphorylated at C-3 or C-22 [5, 7, 8]. Thus, in order to examine whether the ATP: ecdysteroid phosphotransferase activity found in our present experiments shows a substrate specificity in the in vitro system, 2-deoxyecdysone and 2,22-dideoxy- 20-hydroxyecdysone were employed as the subs- trate, since natural conjugates of these ecdyster- oids have their phosphate at different positions (the former at C-22 and the latter at C-3). As can be seen in Figure 1, the phosphate group was transferred to both ecdysteroids quite efficiently. Thus ovarian extract has the ability to phosphory- late the hydroxyl group at both sites. Although we cannot exclude the possibility that a single enzyme can catalyze the phosphate transfer at both sites, we are inclined to think that two separate enzymes are involved in these reactions. The reasoning for this idea comes from the fact that we found in the ovaries of Bombyx only C-22 phosphomonoesters of ecdysone, 20-hydroxyecdysone, 2-deoxyecdy- sone and 2-deoxy-20-hydroxyecdysone but no C-3 phosphoesters of these ecdysteroids [8]. There- fore, an enzyme specific to the phosphate transfer at C-22 seems to be responsible for the synthesis of these conjugates. Recently, Kobbouh and Rees demonstrated in the locust, Schisocerca gregaria, that follicle cells contain ATP: 2-deoxyecdysone 22-phosphotransferase activity [17]. It is not clear at present whether ATP: ecdysteroid phosphot- ransferase found in Schisocerca gregaria and Bom- byx mori are homologous enzymes. To understand the characterization of these enzymes, including the substrate specificity and regulatory mecha- nism, isolation of these enzymes will be necessary. ACKNOWLEDGMENTS We thank Dr. K. Takimoto of the Radioisotopes Laboratory, Yamaguchi University, for his help and suggestions on radioisotope techiniques. The present work was supported in part by a Grant-In-Aid of Scien- tific Research (nos. 01540600, 02304009 and 62304018) and by a SUNBOR GRANT from the Suntory Institute for Bioorganic Research. REFERENCES 1 Mizuno, T. and Ohnishi, E. (1975) Conjugated ecdysone in the eggs of the silkworm, Bombyx mori. Devel. Growth Differ., 17: 219-225. 2 Hsiao, T. H. and Hsiao, C. (1979) Ecdysteroids in the ovary and the egg of the greater wax moth. J. Insect Physiol., 25: 45-52. 3. Dinan, L. H. and Rees, H. H. (1981) The identifica- tion and titres of conjugated and free ecdysteroids in developing ovaries and newly laid eggs of Schis- tocerca gregaria. J. Insect Physiol., 27: 51-58. 4 Hoffmann, J. A. and Lagueux, M. (1985) Endoc- rine aspects of embryonic development in insects. In “Comprehensive Insect Physiology Biochemistry and Pharmacology” Ed. by G. A. Kerkut and L. I. Gilbert, Pergamon Press, Oxford, 1: 435-460. 5 Ohnishi, E. (1986) Ovarian ecdysteroids of Bombyx mort: Reprospect and prospect. Zool. Sci., 3: 401- 407. 6 Weirich, G. F., Thompson, M. J. and Svoboda, J. A. (1986) Jn vitro ecdysteroid conjugation by en- zymes of Manduca sexta midgut cytosol. Arch. In- sect Biochem. Physiol., 3:109-126. 7 Ohnishi, E., Hiramoto, M., Fujimoto, Y., Kakinu- ma, K. and Ikekawa, N. (1989) Isolation and identification of major ecdysteroid conjugates from the ovaries of Bombyx mori. Insect Biochem., 19: 95-101. 8 Ohnishi, E. (1990) Ecdysteroids in insect ovaries. In “Molting and Metamorphosis” Ed. by E. Ohnishi and H. Ishizaki., Japan Scientific Societies Press, Tokyo, pp. 121-129. 9 Watanabe, K. and Ohnishi, E. (1984) The mode of ecdysteroid accumulation in ovaries of Bombyx mori during the pupal and pharate adult period. Zool. Sci., 1: 114-119. 10 Mizuno, T., Watanabe, K. and Ohnishie, E. (1981) Developmental changes of ecdysteroids in the eggs of the silkworm, Bombyx mori. Devel. Growth Differ., 23: 543-552. Borst, D. W. and O’Connor, J. D. (1974) Trace analysis of ecdysones by gas-liquid chromatography, radioimmunoassay and bioassay. Steroids, 24: 637- 656. 12 13 14 174 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193: 165-175. Jaworek, D., Gruber, W. and Bergmeyer, H. U. (1974) Determination of ATP with 3-phosphoglyce- rate kinase. In “Methods of Enzymatic Analysis” Ed. by H. U. Bergmeyer, Academic Press. London, pp. 2097-2101. Sonobe, H., Kamba, M., Ohta, K., Ikeda, M. and Naya, Y. (1991) In vitro secretion of ecdysteroids by Y-organs of the crayfish, Procambarus clarkii. Ex- perientia, 47: 948-9572. 15 16 7, S. Y. TAKAHASHI, K. OKAmoTO et al. Lafont, R., Baydon, P., Blais, C., Garcia, M., Lachaise, F., Riera, F., Somme, G. and Girault, J. P. (1986) Ecdysteroid metabolism: a comparative study. Insect Biochem., 16: 11-16. Koolman, J. (1990) Ecdysteroids. Zool. Sci., 7: 563-580. Kabbouh, M. and Rees, H. H. (1991) Characteriza- tion of the ATP: 2-deoxyecdysone 22-phosphotrans- ferase (2-deoxyecdysone 22-kinase in the follicle cells of Schistocerca gregaria. Insect Biochem., 21: 57-64. ZOOLOGICAL SCIENCE 9: 175-183 (1992) © 1992 Zoological Society of Japan Development and Application of Time-resolved Fluoroimmunoassay for Gonadotropin of a Wide Range of Amphibian Species ATSUSHI IwAsAwal!, SHIGEYASU TANAKA~, YOUICHI HANAOKA® and KATSUMI WAKABAYASHI ‘Hormone Assay Center, *Department of Morphology and *Department of Comparative Endocrinology, Institute of Endocrinology, Gunma University, Maebashi 371, Japan ABSTRACT—A highly sensitive time-resolved fluoroimmunoassay (TR-FIA) was developed to measure bullfrog lutropin (LH). This non-competitive (sandwich) immunoassay uses two kinds of antibodies, one anti-bullfrog LH (monoclonal) immobilized in microtiter wells, the other anti-bullfrog LH (polyclonal) labeled with europium (a non-radioisotopic element). The standard curve was almost linear from 0.031 to 64 ng/ml. The sensitivity (18 pg/ml) was about 10 times better than that of in-house radioimmunoassay (RIA). The intra- and inter-assay variations were less than 4.3 and 4.6%, respectively. The recovery of standard bullfrog LH added to bullfrog sera ranged from 101.8 to 118.6%. The correlation coefficient between LH concentrations in bullfrog sera measured by TR-FIA and RIA was 0.96 with the regression equation of Y(TR-FIA)=0.93X(RIA)—1.93. Pituitary extracts of amphibians in Rana, Bufo, Xenopus, Cynops, Hynobius and Onychodactylus groups, most of which could not be measured by the RIA for bullfrog LH with the same polyclonal antibody, cross-reacted with the TR-FIA system. This wide cross-reactivity suggests the presence of an antibody population which recognizes the gonadotropin of a wide range of amphibian species and becomes fully available by employing sandwich immunoassays. The isoelectric focusing profile of the pituitary homogenate of the adult male frogs of Xenopus laevis measured by TR-FIA closely resembled the profile obtained by Xenopus radioreceptor assay which is thought to be specific for LH-like gonadotropin. Synthetic mammalian gonadotropin-releasing hormone injected into the adult male frogs increased the serum hormone level measurable by TR-FIA more than a hundred-fold. hence not suited for measuring GTH of a wide range of amphibian species [12]. Radioreceptor assays (RRA) are known to have low species specificity in general. For example, RRAs with INTRODUCTION To measure gonadotropin (GTH) levels in blood and pituitary gland by immunoassays is indispens- able for understanding the endocrine regulation of reproduction. In amphibians, purification of GTH has been attained in four species; bullfrog (Rana catesbeiana) [1-7], leopard frog (R. pipiens) [8], Japanese toad (Bufo japonicus) [9] and tiger sala- mander (Ambystoma tigrinum) [10], for two of which, bullfrog [3, 11, 12] and Japanese toad [9], competitive radioimmunoassay (RIA) systems have been established. However, the RIA is characterized by its high species specificity, and Accepted October 3, 1991 Received August 14, 1991 ' To whom all correspondence should be addressed. testicular homogenates of Xenopus laevis (Xeno- pus RRA, [13]) and Anolis carolinensis (Anolis RRA, [14]) as the receptor preparation react with GTH in pituitaries of various amphibian species [15, 16]. RRAs are, however, not suitable for measuring GTH levels in blood, since their sensiti- vities are inadequate and they are easily affected by the presence of serum components. For the same reason, bioassays are also not suited for measuring GTH levels in blood. In this paper, we report the development of a highly sensitive immunoassay for bullfrog LH based on a non-competitive binding of antigens and antibodies. We succeeded, with this assay 176 A. IWASAWA, S. TANAKA et al. system, in measuring GTH in a wide range of amphibian species which showed no significant cross-reactions in the conventional RIA for bull- frog LH. Furthermore, we employed europium (Eu, a lanthanide element) as the tracer instead of radioisotopes. This assay method, called time- resolved fluoroimmunoassay (TR-FIA), was estab- lished in the early 1980s [17-20] and is now used mainly in laboratory diagnosis with assay kits for hormones and tumor markers. We intended to introduce this technique into the measurement of hormones of various animal species. Advantages in using non-competitive and non-radioisotopic immunoassays in the field of comparative endocri- nology will be briefly discussed. MATERIALS AND METHODS Antigens and antibodies Bullfrog LH-IV (pI 9.3) [4, 5], FSH-III (pI 6.2) [6], and their a- and f-subunits [6] used in this study were purified and characterized previously. Rat pituitary hormone preparations, NIDDK- trLH-RP-2, rFSH-RP-2, rFSH-I-7 and rTSH-RP-3 were kindly provided by Dr. A. F. Parlow and the National Hormone and Pituitary Program, U.S.A. Polyclonal anti-bullfrog LH (MOR-BF27- 01RBP83) [12] and monoclonal anti-bullfrog LH (MOR-BF27[B]-01MSM87, BL4B11 of Park et al. [21]) were previously raised and characterized. Europium-labeling procedure Eu-labeling of anti-bullfrog LH was performed by reacting antibody with Eu-labeling reagent (Eu- chelate of N'-[p-isothiocyanatobenzyl]-diethyl- enetriamine-N!, N*, N*, N°-tetraacetic acid, Phar- macia LKB, Uppsala, Sweden). A one mg IgG fraction of polyclonal anti-bullfrog LH purified on a Protein A-Sepharose CL-4B (Pharmacia) col- umn was dissolved in 0.05 M sodium carbonate- bicarbonate buffer, pH 9.8. Then, to the solution was added 0.3 ~mol Eu-labeling reagent dissolved in distilled water, and the mixture was allowed to stand for 16 hr at room temperature. The separa- tion of Eu-labeled antibody was done by gel filtra- tion on a glass column (1.6cm inner diameter) packed with 10cm Sephadex G-50 (Pharmacia) piled on 24cm Sepharose 6B (Pharmacia). The latter was used to separate aggregated reaction products from the monomer. The absorption of the effluent was recorded at 280 nm. In addition, Eu fluorescence of the labeled antibody was mea- sured with a time-resolved fluorometer (Arcus 1230, Pharmacia) and compared with the fluoresc- ence of a 1nM EuCl, solution to calculate the specific activity (Eu/IgG). The labeled antibody was stored at 4°C. Immobilization of antibody A crude IgG fraction of the monoclonal anti- body to bullfrog LH was obtained by ammonium sulfate (33% saturation) fractionation from mice ascites, followed by dialysis against saline. The fraction was immobilized by physical adsorption onto the interior surface of 96-well polystyrene microtiter plates (Microstrip, Labsystem, Hel- sinki, Finland) according to the method described in a previous report [19]. Assay procedure The assay is based on a “sandwich-binding” principle. Fifty «1 of standards or unknown sam- — ples and 150 yl of assay buffer (0.05 M Tris-HCl, pH 7.75, containing 0.5% BSA) were added to the antibody-coated wells and incubated for 1.5 hr at room temperature under shaking. Then the plates were washed with saline containing 0.02% Tween 20 to remove free antigens. For the next step, 200 wl of Eu-labeled antibody diluted to 250 ng/ml with the assay buffer was added to each well and incubated for 1.5 hr at room temperature under shaking. Then the plates were washed to remove free labeled antibodies and, to enhance Eu fluorescence, 200 ul of enhancement solution (0.1 M Acetate-phthalate buffer, pH 3.2, containing 0.1% Triton X-100, 15 ~M 2-naphthoyltrifluoro- acetone and 50 uM tri-n-octylphosphine oxide, Pharmacia) was added to each well. The plates were further shaken for 5 min, and then the fluorescence was measured with the time-resolved fluorometer. Species specificity To examine the species specificity of the assay, fresh-collected pituitaries of eight species of adult TR-FIA for Amphibian GTH 177 amphibians, Rana catesbeiana, Rana japonica, Rana porosa porosa, Bufo japonicus formosus, Xenopus laevis, Cynops pyrrhogaster, Hynobius nigrescens and Onychodactylus japonicus, were homogenized in saline and frozen at —20°C. Then the homogenates were thawed and centrifuged at 15,000 rpm for 30min. The supernatants were serially diluted with the assay buffer and assayed by TR-FIA. Isoelectric focusing (IEF) analysis Pituitary glands freshly collected from 25 adult male frogs of Xenopus laevis were homogenized with distilled water, followed by freezing and thawing treatment and centrifugation at 15,000 rpm for 30 min. The supernatant was subjected to an IEF analysis. The IEF was carried out as previously described [15] and the focused fractions were diluted with Tris-MgCl.-BSA (40mM tris[hydroxymethyl|aminomethane-HCl, 5mM MgCh, pH 7.5, containing 1% BSA) and stored at —20°C until assayed by TR-FIA and RRA. GnRH injection studies Xenopus laevis maintained in our temperature- controlled (25°C water) facility received gonado- tropin-releasing hormone (GnRH) or saline injec- tions as follows: In a dose-response study, which was performed in January, into the dorsal lymph sac of adult male frogs weighing 34.4+5.8 g (mean +SD) was injected saline or one of various doses (0.002, 0.01, 0.05, 0.25 and 1.25 ug per g body weight) of synthetic mammalian GnRH (p- Glu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH,-: 2AcOH:4H;0, Peptide Institute, Osaka) dis- solved in saline. After 15 min, blood was collected by cardiac puncture. In a time-course study, which was performed in October, into the dorsal lymph sac of adult male frogs weighing 44.9+6.9 g (mean +SD) was injected saline or synthetic mammalian GnRH (0.25 ug per g body weight) dissolved in saline. After 5, 15, 30, and 60 min, blood was collected by cardiac puncture. In both experi- ments serum samples were stored at —20°C until assayed by TR-FIA. Injections were given be- tween 2 p.m. and 4 p.m. Five animals were used in each experimental group, and the data were com- pared by one-sided Mann-Whitney’s U-tests. Radioreceptor assay and radioimmunoassay Radioreceptor assay (Xenopus RRA) was car- ried out to measure the GTH activity of the IEF fractions, with testicular homogenate of Xenopus laevis as the receptor preparation. The radioligand was prepared by '*°I radioiodination of NIDDK- tFSH-I-7 with lactoperoxidase (Boehringer Mann- heim GmbH, Mannheim, Germany) according to the method of Miyachi et al. [22]. Bullfrog LH-IV was used as the reference standard. Details of the RRA procedure have been reported by Tanaka et al. [15]. Radioimmunoassay for bullfrog LH was performed as previously described [12]. RESULTS Europium-labeling Figure 1 shows the pattern of elution of the Eu-labeled antibody from the gel filtration col- umn. Eu-labeled IgG (peak B) appeared after peak A which was considered to be aggregated reaction products. Free Eu-labeling reagent appeared as the third peak (C). The number of Eu ions incorporated into an IgG molecule of the fraction B was calculated to be 3.31. 0.2 Cc B 3 0.1 < A 0 GPa toe ez0 ee SOR a0 = oOs 60) a 70.) 60). “90 Elution volume (ml) Fic. 1. Gel filtration of the Eu-labeled antibody. Af- ter the Eu-labeling the reaction mixture was applied to a gel filtration column packed with Sephadex G-50 and Sepharose 6B, which had been equili- brated with 50mM tris-HCl, pH7.75 containing 0.1% NaCl. Elution was then performed with the same buffer at a flow rate of 0.56 ml/min. Standard curve, sensitivity and dilution test Figure 2 shows a typical standard curve, which is 178 A. IwasAwa, S. TANAKA et al. Bullfrog serum ater! 1:64 1:32 1:16 1:8 1:4 1:2 1:1 Eu intensity (cps) 01 4 1 10 100 1000 Bullfrog LH-IV (ng/ml) Fic. 2. Representative data from TR-FIA for bullfrog LH. Bullfrog LH-IV standard (open circles, the mean of 6 replicates) was compared with serum (closed circles, the mean of 3 replicates) from GnRH-stimulated bullfrog. The serum sample was diluted serially with the assay buffer. LH concentra- tion of the sample was estimated to be 51.3 ng/ml. almost linear in the range of 0.03125-64 ng/ml. The assay has a detection limit of 18 pg/ml (0.9 pg/well), defined as the hormone concentration which corresponds to the mean +2 SD of the response in the absence of hormone (standard zero) when interpolated from the standard curve. To test the parallelism of the curves obtained from the standard and bullfrog serum samples, we serially diluted GnRH-stimulated bullfrog serum with the assay buffer. As shown in Fig. 2, the slope produced by the serum samples was parallel to that of the standard. Precision, reproducibility, recovery and correlation to RIA The precision and reproducibility of the assay system were examined with control bullfrog sera at four different LH concentrations. As shown in Table 1, the intra-assay coefficients of variation (CV) for LH concentrations of 0.34, 0.95, 2.59 and 15.28 ng/ml were 4.26, 2.83, 3.16 and 3.01%, respectively. The inter-assay CVs for the same controls were 4.54, 2.77, 2.97 and 1.52%, respec- tively. To assess the recovery of the assay, analyte- supplemented serum samples were prepared as shown in Table 2. Analytical recovery of bullfrog LH ranged from 101.8 to 118.6%. Assay results for 68 serum samples by TR-FIA were compared to the results obtained with the conventional RIA, TABLE 1. Intra- and inter-assay coefficients of variation (CV) in the TR-FIA for control bullfrog sera having four different LH concentrations Mean value Sample (ng/ml) 1 0.34 2, 0.95 g Meas) 4 15.28 Intra-assay Inter-assay CV (%) CV (%) 4.26 (6)° 4.54 (5)? 2.83 (6) 2.77 (5) 3.16 (6) 2.97 (5) 3.01 (6) 1.52 (5) “ Numbers in parentheses are the number of assays or replicates. TABLE 2. Recovery test of standard bullfrog LH from pooled serum of bullfrogs Hormone, added Assay results Hormone, recovered Recovery rate (ng/ml) (%) (ng/ml) (ng/ml) 0 0.7867 + 0.0243" 0.0625 0.8608 + 0.0081 0.25 1.0440 + 0.0210 1.0 1.8048 + 0.0617 4.0 5.1046 + 0.2612 “ Mean+SD. > Mean recovery+standard deviation percentage. Number of tubes is 6 for each concentration. 0.0741 + 0.0256 Oa 780203711 1.0181 +0.0663 4.3180 +0.2623 118.56+40.92° 102.92 + 12.84 101.81+ 6.63 107 O53 1656 TR-FIA for Amphibian GTH 179 by. means of first order regression (Fig. 3). The regression equation was: Y(TR-FIA)= 0.93X(RIA)—1.93, r=0.96. 60 50 E dD) Sea0 < 25 fam = 330 > 2 =x — m 20 o S Fence fs Y=0.93X-1.93 r=-0.96 0 0 10 20 30 40 50 60 Bullfrog LH by RIA (ng/ml) Fic. 3. Correlation between RIA and TR-FIA of 68 bullfrog serum samples. Hormone and species specificity To evaluate the hormone specificity of the assay, various hormones at high concentrations were assayed by TR-FIA as shown in Table3. The TABLE 3. frog LH Hormone specificity of TR-FIA for bull- Hormone, added LH equivalent, Cross-reactivity (ng/ml) observed (ng/ml) (%) fFSH 32 0.001 0.003 128 0.029 0.022 fFSHB 25 0.154 0.616 100 0.711 0.711 fLHa as 0.044 0.176 100 0.243 0.243 fhe 25 0.003 0.012 100 0.075 0.075 nln ESE SE 128 0 0 Abbreviations: f, bullfrog; r, rat. Number of tubes is 6 for each concentration. Onychodactylus japonicus*, Xenopus laevis* 1:4096 Hynobius nigrescens* 1:4096 1:1024 1:256 1:64 1:16 1:4 1:1 x4 1:1024 1:256 1:64 1:16 1:4 1:1 x43 1:1024 1:256 1:64 1:16 1:4 1:1 x42 Cynops EOE EGR 1:1024 1:256 1:64 1:16 1:4 11 Bufo japonicus formosus* 1:256 1:64 1:16 1:4 1:1 x44 Rana japonica 1:1024 1:256 1:64 1:16 1:4 1:1 x46 Rana porosa porosa- 6 1:1024 1:256 1:64 1:16 1:4 1:1 x4 Rana catesbeiana 8 1:1024 1:256 1:64 1:16 1:4 1:1 x4 10° 10° Eu intensity (cps) 104 10° O25 10. 34.0 Bullfrog LH-IV (ng/ml) Fic. 4. Cross-reaction of the pituitary extracts of various species of amphibians to the TR-FIA. Asterisks indicate species where no significant cross-reaction was observed with the RIA for bullfrog LH. Samples were assayed in 0.063 16.0 triplicate. Each value is the mean+SE. 64.0 180 A. Iwasawa, S. TANAKA et al. cross-reactivity of these hormones was 0.711% or less. Figure 4 shows the assay results for serially diluted pituitary homogenates of eight amphibian species including bullfrog. Samples from five spe- cies, Bufo japonicus formosus, Xenopus laevis, Cynops pyrrhogaster, Hynobius nigrescens, and Onychodactylus japonicus, which could not be measured by the conventional RIA for bullfrog LH [12], were shown to be measurable by this assay, although the dilution curves were not para- llel to the standard curve and relatively flat at the high concentrations. Isoelectric focusing IEF profiles of pituitary homogenates of Xeno- pus laevis assayed by TR-FIA and Xenopus RRA are shown in Figures 5A and B. Seven compo- 100 Bullfrog LH-IV equivalent (ng/tube/gland) Bullfrog LH-IV equivalent (u1g/tube/gland) Fraction number nents were observed: the major component (A) appeared at pI 9.3 in TR-FIA and 9.7 in RRA, with the medium peak (F) at pI 5.6. Minor compo- nents were at pI values of 8.6(B), 7.8 (C), 7.0 (D), 5.8 by TR-FIA and 5.9 by RRA (E), 5.4 by RRA (G) and 5.0 by TR-FIA (H). GnRH injection studies Figure 6 shows the serum hormone levels mea- sured by TR-FIA at 15 min after injecting various doses of mammalian GnRH to Xenopus laevis. The responses were weak between 0.002 and 0.01 yg GnRH injection per g body weight but statisti- cally significant compared to the control group injected with saline. The hormone level increased markedly following injections of more than 0.01 yg and peaked following a 0.25 yg injection. This concentration was chosen for the time-course study. In the time-course study (Fig. 7), 0.25 ug Bullfrog LH-IV equivalent (ng/ml) 1.25 0 0.002 0.01 0.05 0.25 GnRH injected (ug/g body weight) Fic. 6. Serum GTH levels in adult male frogs of Xeno- pus laevis at 15 min after the injection of various doses of synthetic mammalian GnRH. N=5 for each point. Each value is the mean+SE. * p< 0.05, ** p<0.01 vs control. Fic. 5. Isoelectric focusing profiles of the pituitary ex- tract from 25 adult male frogs of Xenopus laevis assayed by TR-FIA (A) and Xenopus RRA (B). Closed circles represent the GTH levels and open circles the pH values of the fractions. The focusing and assays were carried out twice. TR-FIA for Amphibian GTH 181 3.0 N o Bullfrog LH-IV equivalent (ng/ml) ro) 0 20 40 60 Time after LHRH injection (min) Fic. 7. Serum GTH levels in adult male frogs of Xeno- pus laevis injected with synthetic mammalian GnRH (0.25 ug/g body weight, closed circles) or saline (open circles). N=5 foreach point. Each value is the mean+SE. * p<0.01 vs controls. GnRH injection per g body weight increased serum hormone levels markedly at 5 min (2.14+ 0.28 ng/ml, expressed in terms of bullfrog LH-IV, mean+SE). The level peaked at 15 min (2.7+ 0.28 ng/ml) and still remained high (2.07+0.28 ng/ml) at 60 min, while serum levels in the control groups injected with saline were low (about 0.02 ng/ml). DISCUSSION We developed a highly sensitive TR-FIA for bullfrog LH based on a non-competitive binding. TR-FIA is known to have several advantages over radioisotope-labeled immunoassays [19, 20, 23]: it is free of limitations associated with the use of radioisotopes; Eu-labeled compounds are stable for a relatively long period (more than a year in Our experience), and although the counting time per sample is one second, counting sensitivity is as good as, or better than that for radioisotopes. Compared to enzyme labeling, the labeling proce- dure is simpler, and steric hindrance due to the label seems less apparent since the molecular weight of the Eu-chelate (630) is much smaller. The sensitivity of the present TR-FIA (18 pg/ ml) is about 10 times better than that of the competitive RIA for bullfrog LH [12], though the incubation time of TR-FIA is less than 4 hr in total which is much shorter than the overnight incuba- tion in RIA. Furthermore, the present assay was about three times more sensitive than a solid-phase immunoradiometric assay for bullfrog LH, which we tried as a preliminary experiment, using the same pair of antibodies as in TR-FIA. Dilution test of bullfrog serum (Fig. 2) and recovery test (Table 2) indicated that disturbance by serum components was practically negligible in this assay system. Results of the precision and reproducibil- ity studies (Table 1) and correlation to RIA (Fig. 3) were also satisfactory. The present TR-FIA, though it uses the same polyclonal antibody as in RIA, could detect im- munoreactive GTH in pituitary homogenates of various amphibians, which was not measurable by RIA (Fig. 4). A possible explanation for this result lies in the difference between the mode of antigen-antibody reaction in competitive (RIA) and that in non-competitive (sandwich TR-FIA) immunoassays. The polyclonal antibody to bull- frog LH may contain populations which cross-react with GTH of a wide range of amphibian species. In competitive RIA, labeled antigen (bullfrog LH) and GTH in other amphibians would compete only for the “cross-reacting antibodies”. If the percen- tage of such antibody populations were very low, the apparent cross-reaction would become so low that the GTH in the animals would appear to be not measurable by RIA. On the other hand, in sandwich immunoassay, which is based on non- competitive binding, the use of an excess amount of antibody would enable us to detect the GTH even if the “cross-reacting antibody” populations are a very low percentage. The monoclonal anti- body (anti-LHf) used in the present assay may also recognize epitopes common to amphibian GTH. Actually, both monoclonal [24, unpublished for Onychodactylus japonicus] and polyclonal (unpub- lished findings) antibodies used in TR-FIA can immunostain pituitaries of the amphibian species studied herein. Just as with sandwich immunoas- say, immunohistochemistry is based on a non- 182 A. Iwasawa, S. TANAKA et al. competitive binding of antigens in tissues and antibodies applied to the tissues [25]. Thus, to find out whether an antigen in the animal species of interest can be measured by a sandwich assay with a pair of antibodies, the checking of immuno- stainability with those antibodies would be useful. To characterize the immunoreactive GTH measurable by TR-FIA, we chose Xenopus laevis for IEF and GnRH injection studies. The IEF profile of the pituitary homogenate of the adult male frogs assayed by TR-FIA closely resembled the profile obtained with Xenopus RRA (Figs. 5A and B), which is reported to be highly sensitive to bullfrog LH but not very sensitive to FSH [14, 16]. It is therefore suggested that TR-FIA recognizes bullfrog LH-like GTH in the Xenopus pituitary gland. Further studies will be necessary to deter- mine the gonadotropic nature of each IEF compo- nent in Xenopus itself. Previous immunological [26, 27] and chromatog- raphic [27] studies have suggested that Xenopus GnRH is identical to that of the mammal. The present GnRH stimulation experiment demon- strated that Xenopus GTH measurable by TR-FIA responded to extrinsic mammalian GnRH (Figs. 6 and 7). In the time-course study (Fig. 7), the GTH level reached its maximum at 15 min after the GnRH injection, which result is in accordance with that in bullfrog reported by Daniels et al. [11]. The difference between the maximal GTH levels in the two experiments (Figs. 6 and 7) is possibly due to the difference in the degree of sexual maturation of the animals used. In conclusion, the authors would like to empha- size the advantages of employing non-competitive and non-radioisotopic immunoassays in compara- tive studies in endocrinology. When there is no immunoassay system for a hormone of an animal and existing competitive immunoassays cross-react only poorly, it would be worthwhile to try sand- wich immunoassays before setting about a purifica- tion of the hormone. The present sandwich TR- FIA will contribute to the understanding of repro- ductive endocrinology in a wide range of amphi- bian species. ACKNOWLEDGMENTS The authors are much obliged to Dr. A. F. Parlow of the Pituitary Hormone and Antisera Center, Harbor- UCLA Medical Center and the National Hormone and Pituitary Program, Univ. of Maryland School of Medi- cine, for supplying NIDDK rat LH, FSH and TSH. The assistance of Drs. Y. Yamashita, M. K. Park, R. Horiuchi, Ms H. Kobayashi, Ms K. Tomizawa and Mr. F. Mizutani is also gratefully acknowledged. REFERENCES 1 Licht, P. and Papkoff, H. (1974) Separation of two distinct gonadotropins from the pituitary gland of the bullfrog Rana catesbeiana. Endocrinology, 94: 1587-1594. 2 Papkoff, H., Farmer, S. W. and Licht, P. (1976) Isolation and characterization of luteinizing hor- mone from amphibian (Rana catesbeiana) pituitar- ies. Life Sci., 18: 245-250. 3 Daniels, E. L., Licht, P., Farmer, S. W. and Pap- koff, H. (1977) Immunochemical studies on the pituitary gonadotropins (FSH and LH) from the bullfrog, Rana catesbeiana. Gen. Comp. Endocri- nol., 32: 146-157. 4 Takahashi, H. and Hanaoka, Y. (1981) Isolation and characterization of multiple components of basic gonadotropin from bullfrog (Rana catesbeiana) pituitary gland. J. Biochem., 90: 1333-1340. _ 5 Takahashi, H. and Hanaoka, Y. (1985) Character- ization of bullfrog gonadotropin molecules in com- parison with mammalian hormones. In “Current Trends in Comparative Endocrinology”. Ed. by B. Lofts and W. N. Holmes, Vol. 1, Hong Kong Univ. Press, Hong Kong, pp. 187-188. 6 Hanaoka, Y., Hayashi, H. and Takahashi, H. (1984) Isolation and characterization of bullfrog gonadotropins. In “Gunma Symposia on Endocri- nology”. Ed. by Institute of Endocrinology, Gunma Univ., Vol. 21, Center for Academic Publications Japan, Tokyo, pp. 63-77. Takada, K. and Ishii, S. (1984) Purification of bullfrog gonadotropins: presence of new subspecies of luteinizing hormone with high isoelectric points. Zool. Sci., 1: 617-629. 8 Farmer, S. W., Licht, P., Papkoff, H. and Daniels, E. L. (1977) Purification of gonadotropins in the leopard frog (Rana pipiens). Gen. Comp. Endocri- nol., 32: 158-162. 9 Takada, K., Itoh, M., Nishio, H. and Ishii, S. (1989) Purification of toad (Bufo japonicus) gonadotropins and development of their homologous radioimmu- noassays. Zool. Sci., 6: 963-974. 10 Licht, P., Farmer, S. W. and Papkoff, H. (1975) The nature of the pituitary gonadotropins and their ~J] 11 12 13 14 15 16 17 18 19 TR-FIA for Amphibian GTH role in ovulation in a urodele amphibian (Ambysto- ma tigrinum). Life Sci., 17: 1049-1054. Daniels, E. and Licht, P. (1980) Effects of gonado- tropin-releasing hormone on the levels of plasma gonadotropins (FSH and LH) in the bullfrog, Rana catesbeiana. Gen. Comp. Endocrinol., 42: 455-463. Tanaka, S., Hanaoka, Y. and Wakabayashi, K. (1983) A Homologous radioimmunoassay for bull- frog basic gonadotropin. Endocrinol. Japon., 30: 71-78. Adachi, T., Pandey, A. K. and Ishii, S. (1979) Follicle-stimulating-hormone receptors in the testis of the frog, Xenopus laevis. Gen. Comp. Endocri- Noles7s ti7—185:- Takada, K. and Ishii, S. (1985) Mammalian FSH acts as LH in the toad, Xenopus laevis. In “Current Trends in Comparative Endocrinology”. Ed. by B. Lofts and W. N. Holmes, Vol. 1, Hong Kong Univ. Press, Hong Kong, pp. 189-190. Tanaka, S., Takikawa, H. and Wakabayashi, K. (1981) Seasonal variation in pituitary gonadotropin in the adult male newt, Cynops pyrrhogaster pyrrho- gaster, revealed by isoelectric focusing technique and radioreceptor assay. Endocrinol. Japon., 28: 335-345. Tanaka, S., Park, M. K., Takikawa, H. and Waka- bayashi, K. (1985) Comparative studies on the electric nature of amphibian gonadotropin. Gen. Comp. Endocrinol., 59: 110-119. Soini, E. and Kojola, H. (1983) Time-resolved fluorometer for lanthanide chelates—a new genera- tion of nonisotopic immunoassays. Clin. Chem., 29: 65-68. Hemmila, I., Dakubu, S., Mukkala, V-M., Siitari, H. and Lévgren, T. (1984) Europium as a label in time-resolved immunofluorometric assays. Analyt. Biochem., 137: 335-343. Loveren, T., Hemmila, I., Pettersson, K., Eskola, J. U. and Bertoft, E. (1984) Determination of 20 21 22 743) 24 25 26 27 183 hormones by time-resolved fluoroimmunoassay. Talanta, 31: 909-916. Loévgren, T., Hemmila, I., Pettersson, K. and Halonen, P. (1985) Time-resolved fluorometry in immunoassay, In “Alternative Immunoassays”. Ed. by W. P. Collins, John Wiley & Sons Ltd., Chiches- ter, pp. 203-217. Park, M. K., Tanaka, S., Hayashi, H., Hanaoka, Y., Wakabayashi, K. and Kurosumi, K. (1987) Production and characterization of a monoclonal antibody against the 8-subunit of bullfrog lutropin. Gen. Comp. Endocrinol., 68: 82-90. Miyachi, Y., Vaitukaitis, J. L., Nieschlag, E. and Lipsett, M. B. (1972) Enzymatic radioiodination of gonadotropins. J. Clin. Endocrinol., 34: 23-28. Jackson, T. M. and Ekins, R. P. (1986) Theoretical limitations on immunoassay sensitivity. Current practice and potential advantages of fluorescent Eu°** chelates as non-radioisotopic tracers. J. Im- munol. Methods, 87: 13-20. Tanaka, S., Park, M. K., Kurabuchi, S., Tabuchi, M. and Kurosumi, K. (1987) Differential immunos- tainability of the subunits of pituitary glycoprotein hormones (LH, FSH and TSH) in several amphibian species. Proc. Japan Soc. Comp. Endocrinol., No. 2, pp. 105-106. Wakabayashi, K. and Tanaka, S. (1988) Assess- ment of specificity of antisera for immunohistoche- mistry. Acta Histochem. Cytochem., 21: 221-229. Deery, D. J. (1974) Determination by radioimmu- noassay of the luteinising hormone-releasing hor- mone (LHRH) content of the hypothalamus of the rat and some lower vertebrates. Gen. Comp. En- drocrinol., 24: 280-285. King, J. A. and Millar, R. P. (1980) Comparative aspects of luteinizing hormone-releasing hormone structure and function in vertebrate phylogeny. En- docrinology, 106: 707-717. a oa Aes AGIs: ih 3 rs at gait, et Ayo ; so ie ie y . “ay: is ZOOLOGICAL SCIENCE 9: 185-192 (1992) © 1992 Zoological Society of Japan Entrainability of Diphasic Circadian Activity of the Mosquito, Culex pipiens molestus to 24-hour Light-Dark Cycles: a Physiological Significance of Critical Light-to-Dark Ratio YOSHIHIKO CHIBA and KENJI TOMIOKA Environmental Laboratory, Biological Institute, Yamaguchi University, Yamaguchi, 753 Japan ABSTRACT— Diphasic circadian flight or locomotor activity of the female mosquito, Culex pipiens molestus, was recorded in 24h cycles with various ratios of light and darkness. A critical ratio for entrainment was LD 22:2, in which two rhythmic processes were dissociated. That is, the activity shows the freerunning diphasic circadian rhythm, on one hand, and is inhibited twice a day, on the other, synchronously with the LD. As the result, a unique temporal pattern is manifested. INTRODUCTION Entrainment of circadian rhythm to a 24-hour cycle of light and darkness depends on properties of the cycle including the ratio of light hours to dark ones (LD ratio), which may exert two kinds of effect, parametric and nonparametric [1, 2]. On the other hand, characteristics of the rhythm, synchronizing or freerunning, are altered by exter- nal stimuli especially by zeitgebers or by other endogenous rhythms: the alteration referred to as external or internal masking [3, 4]. Furthermore, when properties of the light and darkness cycle are out of the range of entrainment, the rhythm is often modulated for the period by the signals of zeitgeber, through which it crosses. This phe- nomenon is called relative coordination, reflecting periodically changing sensitivity of the organisms to the cyclic stimulus [5]. Many animals show the diphasic circadian rhythm comprising two peaks in one cycle, which usually correspond to the morning and the evening peaks in the synchronizing rhythm [6-14]. Howev- er, relatively prominent peak alone of the two has been observed in most of entrainment studies. Only studies dealing with the two peaks may be those done by Beeston and Morgan [12] and Jones Accepted July 29, 1991 Received June 17, 1991 [14] using the population of snail, Melanoides tuberculata and the mosquito, Culex pipiens quin- quefasciatus, respectively. The both papers in- clude a postulation that the peaks are controlled by separate circadian oscillations. The present study investigated how the diphasic acitivty of the mosquito, C. p. molestus responds to changing LD ratio in 24h cycle. Distinctions from the published studies are i) to use the photo- phile mosquito, molestus, which is active in both light and dark periods [15] and, thereby, more adequate for this kind of study than scotophile quinquefasciatus, 11) to change the LD ratio at smaller increment, and iii) to measure individual animals separately. A noteworthy finding was that, in a critical LD ratio (a limit of entrainment range), the activity showed two rhythmic compo- nents, freerunning and entrained, simultaneously. A hypothetical view will be proposed for this phenomenon. MATERIALS AND METHODS Adult female mosquitoes, Culex pipiens moles- tus, of four or five days old were used. They were provided from a laboratory colonization kept in LD 16:8 and 25°C, hereafter called the standard environmental condition. The activity recording system was of photoelectrical principle as de- scribed previously [16]. 186 C. p. molestus is autogenous; the first egg maturation is attained without blood-feeding. In 25°C, eggs are matured 3 to 4 days after emergence and are all laid at a time the next day if she is already inseminated. But virgin females oviposit at an unpredictable time or do not oviposit. Our materials were virgin and were not provided with an adequate water pool for oviposition, carrying mature eggs probably over the whole experimental span. The experiment was carried out to see effects of the light-to-darkness (LD) ratio in a 24-hour cycle (Fig. 1). Nine ratios were adopted, ranging from “zero-hour light to 24-hour darkness” (constant darkness abbreviated as DD or LD 0:24) to the reciprocal ratio (constant light abbreviated as LL or LD 24:0). Seven to fifteen mosquitoes were used for each ratio. On the first day of recording, the mosquito was held in LD 16:8, which was followed by one of the nine experimental LD ratio continuing for at least six days. The experimental ratio was made by changing the time of light-on so that the light was always turned off at the definite Y. CHIBA AND K. TOMIOKA clock hour (20) throughout the recording span. A fluorescent lamp used did not illuminate equally over the whole environment-controlled cabinet with activity chambers within, furnishing, however, on the average 170 lux in the light frac- tion of the LD cycle. Temperature was kept at about 25°C. Sugar solution (3%) was always provided by a siphon through the ceiling so that the mosquito may feed ad libitum. The solution probably made the relative humidity in the activity chamber almost saturated. Rhythmometry was based on a so-called chrono- biological window (least squares spectrum [17]) which, simply stated, involves the least Squares fit of cosine models to the time series under study and detects the best fitting cosine function. The win- dow covered trial periods from 10 to 35h. RESULTS 1. Free-running activity under DD and LL. Eleven of fourteen mosquitoes under constant darkness (DD) exhibited significant circadian —— = = Light-period per day (hr.) Fic. 1. fA A f F) — S= = =" A— pe ah ZB <7 Foe Zale : I Bal ee 12. Jnl ie Hours Circadian activity of the mosquito Culex pipiens molestus under 24-h cycles with various ratios of light (open area) to darkness (shaded area). LD 16:8 was given on the first day which is followed by the experimental LD ratios. Curves are averages from 7 to 15 females. For further explanation, see text. Entrainment of mosquito circadian rhythm 187 rhythm showing the average free-running period (tau) of 22.5+0.23(SE) h. The remaining three also were rhythmic but with ultradian periods of 18.8, 14.8 and 10.9h, respectively. In some circa- dian mosquitoes, the earliest circadian cycles after the transition from LD to DD apparently wore the diphasic character which, however, became obscure or even disappeared in the ensuing cycles. A characteristic feature of rhythm under DD is a noisiness; circadian peaks are sometimes hardly visible in an incessant activity and are not always detectable even by the statistical way. That the rhythmicity becomes quite obscure in DD with the passage of days in the mean curve is due to the individual variation of free-running period and the high noisiness (Fig. 1). Ten out of 12 mosquitoes held in LL showed a conspicuous rhythm with the average tau of 15.57 +1.20(SD)h (Figs. 1, 2, 4). Tau was on the level of 14 to 15 h in seven mosquitoes, but 16 to 17 hin the other three. The maximum value (17.9 h) could not be excluded (Smirnov’s rejection test, Ty=1.978, p>0.05). The individual activity con- verges on the peak time to yield a rhythm with a relatively low noise. This makes the grouped data (mean curve) more rhythmic than in DD. 2. Activity under the intermediate LD ratio be- tween DD and LL. Under the standard environmental condition, Time of 12 24 12 LD 16:8, where all mosquitoes were held at least on the first day of recording, activity synchronizes well with the LD, forming the diphasic pattern with the evening and morning peaks. The evening peak (diagonally striped, Fig. 1) starts rising a few hours before cresting at light-off, which is followed immediately by night phase of very low activity. The activity augments abruptly at light-on to form the morning peak (vertically striped), which is much higher than the evening one, and expands the foot over a few hours toward midday. There is a totally inactive midday phase lasting for five hours. Thus, the daily pattern is diphasic, recur- ring as long as LD 16:8 continues. It looks that LD ratio which entrains insects ranges from LD 4:20 to 21:3; the entrained diphasic rhythms keep the stable relation to LD cycles throughout the experimental days. The least squares spectrum showed a simple pattern with conspicuous peaks at trial periods of 12 h and/or 24 h as typically seen in LD 16:8 (Fig. 4). The temporal activity pattern depends on LD ratio in three ways. One pertains to the phase relation of activity rhythm to the light cycle. With reference, for example, to the onset of darkness, the evening and the morning peaks both are ad- vanced at different rates with increasing LD ratio; the former is faster. The slope coefficient in the linear regression, obtained for the activity crest day 24 12 ______ _—$<_TE——— Animal=6 Tu=14.2h LD Animal=7/ LL Ton -14.2 h aim LL Fic. 2. Double plotted activity records of two representative females held in LL following LD 16:8. 188 Y. CHIBA AND K. TOMIOKA hour (reference point: light-off) against light period per day, is significant both in the evening peak (b=—0.24, t=2.800, p<0.05) and in the morning one (b= —0.09, t=3.600, p<0.05) (Fig. 3); the slope coefficients mean the degree of peak advance in response to one-hour lenghtening (shortening) of light (dark) period. The difference between these two coefficients was demonstrated by that the regression line for the morning peak crests shows a significant slope coefficient (b= 0.16, t=7.31, p<0.01) against the regression line for the evening peak. The second way of the modification involves the size of peak. When getting out of the dark fraction in the higher LD ratio, the both peaks become much larger; this is clear when one compares, for example, the evening peak in LD 4:20 with that in LD 21:3. The third way is related to an activity peak dotted in Fig. 1, which is formed in addition to the evening and morning peaks probably as a direct response to the light steps. This exogenous peak is discernible more easily at light-on than at light-off, becoming lower as LD ratio deviates from 16:8, in which, however, the morning peak timed just at light-on probably veils the exogenous one. Light-period per day(hr) nD Oa Aiea O-—Iee In LD 22:2 which may be outside the entrain- able range, activity showed some unique tenden- cies; the linearity of peak advance in response to increasing LD ratio (Fig. 3) may not simply be extended over this particular lighting condition. The activity peak does not always occur so regular- ly as have been seen in the entrainable range and LL, weakening or almost disakppearing around the experimental hours 96 or 152. It is sometimes hard, therefore, to know how the two peaks occur in temporal relation to the LD cycle; in other words, the observable peaks look neither being fully entrained nor fully free-running. These ten- dencies were clearly observed in 13 of 15 mos- quitoes, which showed the least square spectrum peaking twice at trial periods of 12 h and 14 or 15h (Fig. 4). To see if the rhythmic components of these periods really exist in the original time series, average activities were obtained against time axes twice as long as the detected periods for each of the 13 individuals. As the results, in all indi- viduals, the activity peaked twice clearly against each of the two time axes (Fig. 5). 16 20 24 28 32 36 40 44 48 Hours Fic. 3. Timing of evening or morning peak in 24-h LD cycles with varying LD ratios. Timing of peak represented by activity crest is the average for the last three cycles (Fig. 1), in which the entrainment appears to be fully attained. Linear regression functions were obtained between the crests against the light-off line. b=22 L=24 Fic. Fic. Entrainment of mosquito circadian rhythm Trial period (hr) 30 (N=192) (N=144) 156 (N=192) 156 (N=168) (N=168) (N= 96) (N= 96) (N= 96) 165 (N= 96) 167 (N= 96) 168 (N= 96) 170 (N= 96) Ga k KETC (N=120) (N=120) 73 (N=120) (N=120) (N=120) (N=120) 188 (N=120) (N=120) LS S (N=120) (N=120) (N= 63) 166 (N=120) 164 (N=120) ee 187 (N=120) 186 (N=120) (N=120) (N=120) (N=120) sr (N=120) 10 (N=120) 47 (N=120) 8 (N=120) ees (N=120) (N=120) % 44 (N=120) 42 (N=120) 39 (N=120) 4. Least squares spectra for each female in three kinds of 24-h LD cycle with different LD ratios. Numerals on the left side (16, 22, and 24) show the length of light period (L) in hour. Those on the right side, animal number and, in parenthesis, num- ber of samples (N). Since sampling was done every hour, N shows, on the other hand, the length of span analyzed in hour. Uppermost spectrum, for exam- ple, was obtained from No. 154 mosquito recorded for the consecutive 192 hours. 5. Results of statistical analyses are shown for two mosquitoes, A and B. Uppermost panel: double- plotted activity record for 13 days while lighting condition was changed from LD 16:8 to LD 22:2. Dark period is framed. LSS: Least squares spec- trum obtained from activity record under LD 22:2. Lower two panels: Average temporal pattern of activity under LD 22:2 obtained on time axes, which are twice as long as length of periods detected significantly in the LSS. A LSS 10 12 14 24 30 TRIAL PERIOD 10 fas >e > E 0) = 0 12 24 2 Re 10 0) 0 14 28 TIME (HOURS) B 12 24 12 24 12 TIME OF DAY LSS 10 12 15 24 30 TRIAL PERIOD 10 = cE (0) > = 0 12 24 O ‘Graduate School of Human and Environmental Studies, c/o Biological Laboratory, Yoshida College, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606, Japan, ?Obihiro Centennial City Museum, Midorigaoka, Obihiro 080, Japan, and Graduate School of Science and Technology, Niigata University, Niigata 950-21, Japan, °Toji-in Minami-cho, 5040, Kita-ku, Kyoto 603, Japan, and “Department of Zoology, Faculty of Science Kyoto University, Kitashirakawa, Kyoto 606, Japan ABSTRACT— -Allozyme data are used to examine the genetic relationships among two samples of Hynobius retardatus from Obihiro, and a geographically discrete sample from Chitose. The two samples from Obihiro are morphologically distinct from each other, but are found to be electrophoretically closely related, together forming a group distinct from the Chitose sample. The degree of genetic differentiation as measured by the F statistics is also low between the two Obihiro samples. The morphological differentiation seen between these two samples does not seem to be attributable directly to their genetic differentiation, but to the ecological divergence that is yet to be surveyed. INTRODUCTION Hynobius retardatus is endemic to Hokkaido of northern Japan and is abundant, widely ranging over much of the island [1]. This species is noted for a significantly smaller number of chromosomes (2n=40) [2] than found in all other congeneric species (2n=56 or 58) [3]. Thus, many studies on this species have hitherto been made mainly from karyological analyses (e.g., [4]), and studies from other fields of biology are relatively few (see [5] for literature review). Because the geographic range of this species is larger than those of many other congeneric spe- cies, the presence of some degree of local popula- tion differentiation, both in morphology and gene- tic structure, is expected, but almost no studies have been made on intraspecific variation within this species. Accepted July 23, 1991 Received May 23, 1991 > Present Address: Tochigi Prefectural Museum, Utsu- nomiya, Tochigi 320, Japan. Sato and Nakabayashi [6] and Sato [7] found the occurrence of a unique population of H. retardatus within a narrow range of Obihiro-shi, Tokachi Plain, eastern Hokkaido. Salamanders from this population are distinguishable from individuals of neighbouring populations of H. retardatus in their unusually larger adult body size [population mean +2SE of total length (TOTL) >175 mm in males and>170mm in females (Sato, unpublished data)], and probably related to this, in the larger size of egg sac and larger egg diameter. Thus, they are tentatively named the “large” type, in contrast to the Hynobius retardatus “common” type (population mean+2SE of TOTL<170mm in males and< 165 mm in females). Although further morphological investigation is required to eluci- date the relationships of these two types, it is interesting to study them from a genetic viewpoint. In studying genetic relatiohships among local populations of caudate amphibians, the technique of electrophoresis has been widely used (e.g., [8- 10]), and successfully applied to the Japanese species [11, 12]. Herein, we present analyses of 194 M. Marsut, T. Sato et al. allozyme data to clarify the relationships of these two types, with another geographically discrete population as a reference. MATERIALS AND METHODS For electrophoretic examination, three samples were employed. These included 19 males of the large type and 11 males of the common type from Obihiro-shi (42°44’40" N, 142°5115” E, and 42°42’30” N, 142°5715” E, respectively), and for comparisons, five males of the common type from Shikotsu-ko, Chitose-shi (42°46'20” N, 141°24°30° E). The localities of the two Obihiro samples are 8.5 km apart, whereas Shikotsu-ko is 140 km away from Obihiro-shi and is clearly split from the latter locality by the Hidaka mountains and Ishikari plain. Liver tissues, removed from freshly killed anim- als and frozen ( —84°C), were used throughout the analysis. Woucher specimens were fixed in 10% formalin, later preserved in 70% ethanol and stored in the Biological Laboratory, Kyoto Uni- versity. Horizontal starch gel electrophoresis was employed using 11.5% potato starch (Connaught) to resolve allozyme products [13]. The allozymes examined, locus designations, and buffer system used are listed in Table 1. Genetic interpretations of allozyme data were based on criteria developed by Selander et al. [13]. Enzyme nomenclature and E. C. numbers followed the recommendations of the Nomenclature Committee of the International Union of Biochemistry [14], and the notation of loci, electromorphs and genotypes followed princi- pally Murphy and Crabtree [15]. Electromorphs were designated by letters with “a” representing the most rapidly migrating anodal variant. The BIOSYS-1 computer program [16] was used to calculate all statistics. Standard estimates of genetic variability, i.e., mean heterozygosity by direct count (H), proportion of polymorphic loci (P), and the mean number of electromorphs per locus (A), were computed for each sample. Vari- able loci were checked by Chi-square goodness-of- fit tests to determine whether observed genotype frequencies were in Hardy-Weinberg equilibrium in each sample. The expected numbers of each genotype were calculated using Levene’s formulae for small samples [17] and pooling. A contingency Chi-square test [18] was performed for inter- sample electromorph frequency heterogeneity. F- statistics were also calculated for all variable loci according to Wright [19]. Overall genetic differentiation among the three samples was estimated using coefficients of Nei’s unbiassed genetic distance [20] and modified Ro- gers’ genetic distance [21]. Genetic relationship among samples were estimated from the pairwise matrix of Nei’s distance, clustered according to the UPGMA algorithm [22]. This method assumes equal rates of molecular evolution along all bran- ches, but the validity of this assumption is am- biguous. We adopt this method to describe amounts of genetic divergence in this species and to allow comparison to literature accounts of variability in other caudate species. Alternatively, an unrooted Distance Wagner tree [23] was constructed using BIOSYS-1 optimized with the multiple addition criterion of Swofford and Selander [16] with metric measures of. modified Rogers’ genetic distance coefficients [21]. | RESULTS We studied the variability of 25 presumptive loci. Of these, eight were monomorphic in our sample (Table 1). These included mAat-A, sAcon-A, mMdhp-A, Est-1, Est-2, Pep-C, sSod-1, and sSod-2. Of the remaining 17 polymorphic loci, mAcon-A, Acp-A, Gpi-A, Ldh-B, and Pgm-C were the most variable loci, each with three alleles. In all the polymorphic loci, except for the mMdhp- A locus, the predominant electromorph, occurring at frequencies of 50% or more, was identical in the three samples. In the mMdhp-A locus, the Chitose sample had a predominant electromorph that was absent in the other two samples from Obihiro. In the three samples, the mean number of electromorphs per locus (A) varied from 1.24— 1.68, the percentage of polymorphic loci (P) from 24.0-44.0, and the mean heterozygosity (H) values from 0.029-0.068. The highest polymorphism and heterozygosity were found in the Obihiro-large sample, while the Obihiro-common sample ex- hibited the lowest variability. Statistically significant (p<0.05) deviation from Hardy-Weinberg equilib- Differentiation in Hynobius retardatus TABLE 1. Mitochondrial and supernatant loci are denoted by “m” and “s” prefixes, respectively Enzymes (commission number) Acid phosphatase (3.1.3.2) Acid phosphatase (3.1.3.2) Aconitate hydratase (4.2.1.3) Aconitate hydratase (4.2.1.3) Aspartate aminotransferase (2.6.1.1) Aspartate aminotransferase (2.6.1.1) Creatine kinase (2.7.3.2) Dipeptidase (3.4.13.11) Dipeptidase (3.4.13.11) Esterase (—) Esterase (—) Glucose-6-phosphate isomerase (5.3.1.9) L-iditol dehydrogenase (1.1.1.14) Isocitrate dehydrogenase (1.1.1.42) L-lactate dehydrogenase (1.1.1.27) L-lactate dehydrogenase (1.1.1.27) Malte dehydrogenase (1.1.1.37) Malate dehydrogenase (1.1.1.37) “Malic Enzyme”** (1.1.1.40) “Malic Enzyme”** (1.1.1.40) Phosphoglucomutase (5.4.2.2) Phosphoglucomutase (5.4.2.2) Phosphogluconate dehydrogenase (1.1.1.44) Superoxide dismutase (1.15.1.1) Superoxide dismutase (1.15.1.1) Locus Acp-A Acp-B mAcon-A sAcon-A mAat-A sAat-A Ck-A Pep-B Pep-C Est-1 Est-2 Gpi-A Iddh-A sIcdh-A Ldh-A Ldh-B mMdh-A sMdh-A mMdhp-A sMdhp-A Pgm-A Pgm-C Pgdh-A sSod-1 sSod-2 Buffer condition* B > Pro @ Pa es pe Oe SO Pe LP ee ee eae 4 ee ES eS D D Obihiro large 0.08 0.87 0.05 0.92 0.08 0.29 0.58 0.13 1.00 1.00 0.63 0.37 0.39 0.61 1.00 coonrionrnoaoenpenddqogrbdgosmsag ® 1.00 1.00 1.00 0.03 0.95 0.03 1.00 »aOa¢CnD D wm Pp 0.08 0.92 0.05 0.95 0.03 0.84 0.13 1.00 Toor ern Fm oy 1.00 oy 1.00 1.00 0.03 0.97 0.11 0.84 0.05 0.11 0.89 1.00 1.00 eo —) Cer i) @ Cr) Cr) [ Obihiro common a b 0.23 0.77 1.00 0.45 b 0.55 Tren o7nrorr oer ® & 1.00 1.00 0.68 0.32 0.09 0.91 0.14 0.86 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 a 1.00 * Buffer system: A=Tris-citrate, pH7.0. B=Tris-citrate, pH6.0. C=Tris-citrate, borate-EDTA, pH 8.0. ** NADP-dependent malate dehydrogenase. pH 8.0. 195 Electromorph frequencies for 25 loci resolved in three samples of Hynobius retardatus. Chitose b 0.90 c 0.10 a 1.00 b 0.80 c 0.20 a 1.00 a 1.00 a 1.00 b 1.00 0.40 0.60 1.00 1.00 1.00 1.00 ef ™) &) els a 0.70 b 0.30 b 1.00 b 1.00 b 1.00 0.40 0.60 0.40 0.60 0.80 0.20 1.00 1.00 Teor ort Fm a 0.50 b 0.50 a 0.30 b 0.70 a 1.00 a 1.00 D=Tris- 196 M. Martsut, T. Sato et al. Obihiro-large Obihiro-common Chitose 0.1 0.075 0.05 0.025 0 Nei's Unbiased Genetic Distance Fic. 1. UPGMA tree constructed from Nei’s (1978) unbiased genetic distance for three samples of Hynobius retardatus. The percent standard devia- tion=3.324 and the cophenetic correlation=0.998. The Distance-Wagner method using modified Ro- gers’ D (Wright, 1978) and rooted at the midpoint of the longest path produces the same tree topology. rium was observed for several loci in all the three samples: mAcon-A, Acp-A, Acp-B, Ck-A, sIcdh- A, Ldh-A, Ldh-B, and Pgdh-A in the Obihiro- large sample; mAcon-A, Acp-A, CK-A, and Pgm- C in the Obihiro-common sample; and mAcon-A, sMdh-A, and mMdh-A in the Chitose sample. All of these deviations were heterozygote deficiencies. TABLE 2. Summary of Fez values calculated from 17 polymorphic loci for all samples of Hynobius retardatus and for Obihiro-large and Obihiro- common types Obihiro-large Locus All samples Aad COMion Acp-A 0.050 0.028 Acp-B 0.053 0.040 sAcon-A 0.105 0.020 sAat-A 0.151 0.003 Ck-A 0.205 0.122 Pep-B 0.118 0.073 Gpi-A 0.027 0.020 Iddh-A O22 — mIcdh-A 0.054 0.041 Ldh-A 0.036 0.027 Ldh-B 0.095 0.073 mMdh-A 0.308 — sMdh-A 0.308 ~ mMdhp-A ON/27 ~ Pgm-A 0.018 0.013 Pgm-C 0.207 0.144 Pgdh-A 0.132 0.056 Mean 0.226 0.060 Contingency Chi-square tests for heterogeneity of electromorph frequencies at variable loci be- tween the two samples from Obihiro were not significant (p >0.01) except for one case (Pgm-C: y°=15.24, p<0.01). In order to quantify this spatial homogeneity, F-statistics were calculated among these two samples, and again for all three samples (Table 2). In the two samples from Obi- hiro, the Fs; varied from 0.003 for the sAat-A locus to 0.144 for the Pgm-C locus and averaged 0.060. This value was appreciably lower than the Fsy of 0.226 calculated for all three samples. Figure 1 shows genetic groups clustered by the UPGMA algorithm on the basis of Nei’s genetic distance [20]. The two samples from Obihiro formed a cluster (D+SE=0.010+0.006) that is separated from the Chitose sample with an average D of 0.072 (D+SE=0.074 +0.033 in the large and 0.070 + 0.033 in the common type). The topology of the Distance-Wagner tree constructed using modified Rogers’ D and rooted at the midpoint of the longest path is identical with the UPGMA tree. DISCUSSION The two samples of H. retardatus from Obihiro examined here have breeding sites of only 8.5 km apart and seem not to be geographically isolated by any evident barriers. However, the two sam- ples have been reported to be fairly divergent chiefly in adult and egg-sac morphology and, less significantly, in breeding season [6, 7]. Males of the large type have a mean TOTL of 180mm, while the common type has a TOTL usually less than 159mm. The common type from Chitose is morphologically indistinguishable from the Obi- hiro common type. Thus, we first expected the smaples of common-type from Obihiro and Chi- tose to be genetically more similar to each other than to the Obihiro large type. The present results, however, do not support this assumption. The values of percentage poly- morphic loci (P) and mean heterozygosity (H) found in the three samples of H. retardatus (24.0- 44.0 and 0.029-0.068) are within the intrapopula- tional range reported for many urodele species [24], and indicate that this species is almost as variable as other widely ranging Japanese urodeles Differentiation in Hynobius retardatus 197 such as Cynops pyrrhogaster (20.0-53.3 and 0.031-0.147, respectively: [11, 12]) or Hynobius nigrescens (9.1-45.5 and 0.015-0.077, respective- ly: Matsui et al., in preparation). The value of Fsr for the three samples (0.226) is even larger than that recorded in the widely spread populations of Cynops pyrrhogaster from Tohoku region (0.167: [12]), and indicates the presence of moderately great interpopulational difference in H. retardatus. However, on the basis of the contingency Chi Square test, the two samples from Obihiro were judged not to be spatially significant in electro- morph frequency differences at most of the vari- able loci. This conclusion is also supported by the low Fsy value calculated for the two populations from Obihiro (0.060). Therefore, notable inter- sample differentiation is judged to be mainly due to the differentiation between the Chitose sample and the two samples from Obihiro, and morpholo- gical and related ecological differentiation found between the latter two samples is hardly attribut- able to their genetic differentiation as far as the present results have shown. Actually, genetic groups clustered by both the UPGMA algorithm using Nei’s D and by the Distance-Wagner method using modified Rogers’ D resulted in a cluster of the two samples from Obihiro that is distinct from the Chitose sample. The distance value of 0.070, found between the two common types, is much larger than that of 0.010 obtained between the two samples from Obihiro. Thus, the two morphologically dissimilar types from Obihiro are judged to be more closely related to each other than either is to the Chitose sample, which is morphologically indistinguishable from the Obihiro common type. The obvious differences in morphology as examplified by the adult body length in the two Obihiro types might, therefore, be largely a result of the probable life history differences of the two types, including food availability, age composition, individual growth speed, time of sexual maturity, and life-span [25, 26]. Turning to overall genetic similarity, the greatest genetic distance value (0.074) obtained among the three populations of H. retardatus is much lower than the values generally reported to occur among sister species or even subspecies pairs of Japanese salamanders and newts [11, 27]. Therefore, although the three samples of H. retar- datus are judged to be genetically somewhat differ- entiated, they cannot be split at any taxonomic rank genetically. Using the correlation between D values and time of separation claimed to be unique to urodeles (1D=14 MY [28]), and the greatest dif- ference (Nei’s D [20]=0.074) between Obihiro- large sample and Chitose sample, the two were considered to have separated approximately 1x 10° years B. P. This coincides approximately with the middle of lower Pleistocene. On the other hand, the two samples from Obihiro (D=0.010) are assumed to have separated approximately 1.4 10° years B. P., during the Ris-Wurm intergla- cial period in the middle Pleistocene. ACKNOWLEDGMENTS We thank T. Hikida for computer assistance. J. A. Wilkinson corrected verbal errors. This work was sup- ported by Grants-in-Aid to MM _ (Nos. 63540599, 01304001) from Ministry of Education, Science and Culture, Japan. REFERENCES 1 Uéno, S.-I. (1981) Hynobius retardatus. In “Final report of reptiles and amphibians survey of the second national survey on the natural environment in Japan, 1978 Pt. 2”. Ed. by Nature Conserv. Soc. Jap., Tokyo, pp. 123-128. (In Japanese) 2 Azumi, J. and Sasaki, M. (1971) Karyotype of Hynobius retardatus Dunn and Hynobius nigrescens Stejneger. Chrom. Inf. Serv., 12: 31-32. 3 Seto, T. and Utsunomiya, T. (1987) Chromosome analysis of Hynobius arisanensis Maki, a salamander endemic to Taiwan. Herpetologica, 43: 117-119. 4 Kuro-o, M., Ikebe, C., Katakura, R., Izumisawa, Y. and Kohno, S. (1989) Analyses of phylogenetic relationships in the genus Hynobius by means of chromosome banding and southern blot hybridiza- tion. In “Current Herpetology in East Asia”. Ed. by M. Matsui, T. Hikida and R. C. Goris, Herpetol. Soc. Jpn, Kyoto, pp. 91-94. 5 Sato, T. (1989) Breeding environment and spawn- ing of a salamander, Hynobius retardatus, at the foot of Hidaka Mountains, Hokkaido. In “Current Her- petology in East Asia”. Ed. by M. Matsui, T. Hikida and R. C. Goris, Herpetol. Soc. Jpn, Kyoto, pp. 292-304. 10 11 12 13 14 iS 16 198 Sato, T. and Nakabayashi, S. (1987) Distribution and habitat of salamander (Hynobius retardatus) in Tokachi district, Hokkaido. I. Central Tokachi. Bull. Obihiro Centennial City Mus., 5: 17-28. (In Japanese) Sato, T. (1987) External characteristics of Hynobius retardatus from Tokachi district, Hokkaido, in the breeding season. Jpn. J. Herpetol., 12: 80. (In Japanese) Karlin, A. A. and Guttman, S. I. (1986) Systematics and geographic isozyme variation in the plethodon- tid salamander Desmognathus fuscus (Rafinesque). Herpetologica, 42: 283-301. Guttman, I. S. and Karlin, A. A. (1986) Hybridiza- tion of cryptic species of two-lined salamanders (Eurycea bislineata complex). Copeia, 1986: 96-108. Reilly, S. M. (1990) Biochemical systematics and evolution of the eastern North American newts, genus Notophtalmus (Caudata: Salamandridae). Herpetologica, 46: 51-59. Hayashi, T. and Matsui, M. (1988) Biochemical differentiation in Japanese newts, genus Cynops (Salamandridae). Zool. Sci., 5: 1121-1136. Hayashi, T. and Matsui, M. (1990) Genetic dif- ferentiations within and between two local races of the Japanese newt, Cynops pyrrhogaster, in eastern Japan. Herpetologica, 46: 423-430. Selander, R. K., Smith, M. H., Yang, S. Y., John- son, W. E. and Gentry, J. B. (1971) Biochemical polymorphism and systematics in the genus Peromy- scus. I. Variation in the _ old-field mouse (Peromyscus polionotus). Stud. Genet. 6. Univ. Texas Publ., 7103: 49-90. International Union of Biochemistry, Nomenclature Committee (1984) Enzyme Nomenclature, 1984. Academic Press, NY. Murphy, R. W. and Crabtree, C. B. (1985) Evolu- tionary aspects of isozyme patterns, number of loci, and tissue-specific gene expression in the prairie rattlesnake, Crotalus viridis viridis. Herpetologica, 41: 451-470. Swofford, D. L. and Selander, R. B. (1981) BIOSYS-1: a FORTRAN program for the compre- Ae, 18 19 20 21 22 23 24 25 26 Za]) 28 M. Matsui, T. Sato et al. hensive analysis of electrophoretic data in popula- tion genetics and systematics. J. Hered., 72: 281- 283. Levene, H. (1949) Ona matching problem arising in genetics. Ann. Math. Stat., 20: 91-94. Workman, P. L. and Niswander, J. D. (1970) Population studies on southweatern Indian tribes. II. Local genetic differentiation in the Papago. Amer. J. Hum. Genet., 22: 24-49. Wright, S. (1965) The interpretation of population structure by F-statistics with special regard to sys- tems of mating. Evolution, 19: 395-420. Nei, M. (1978) Estimation of average heterozygos- ity and genetic distance from a small number of individuals. Genetics, 89: 583-590. Wright, S. (1978) Evolution and the Genetics of Polulations. Vol. 4. Variability within and among natural populations. University of Chicago Press, Chicago. Sneath, P. H. A. and Sokal, R. R. (1973) Numerical Taxonomy. W. H. Freeman, San Francisco. Farris, J. S. (1972) Estimating phylogenetic trees from distance matrices. Am. Nat., 106: 645-668. Shaffer, H. B. and Breden, F. (1989) The rela- tionship between allozyme variation and life history: Non-transforming salamanders are less variable. Copeia, 1989: 1016-1023. Semlitsch, R. D. (1989) Geographic and local variation in population parameters of the slimy salamander Plethodon glutinosus. Herpetologica, 36: 6-16. Tilley, S. G. (1980) Life histories and comparative demography of two salamander populations. Copeia, 1980: 806-821. Matusi, M. (1987) Isozymic variation in salaman- ders of the nebulosus-lichenatus complex of the genus Hynobius from eastern Honshu, Japan, with a description of a new species. Jpn. J. Herpetol., 12: 50-64. Maxson, L. R. and Maxson, R. D. (1979) Compara- tive albumin and biochemical evolution in pletho- dontid salamanders. Evolution, 33: 1057-1062. ZOOLOGICAL SCIENCE 9: 199-206 (1992) © 1992 Zoological Society of Japan Rhynchocinetes striatus, a New Species (Decapoda, Caridea, Rhynchocinetidae) from Southern Japan Kencut Nomura and KeEn-Icut Hayasut Kushimoto Marine Park Center, Kushimoto, Wakayama 649-35 and 'Shimonoseki University of Fisheries, Nagatahonmachi, Shimonoseki, Yamaguchi 759-65, Japan ABSTRACT—A new rhynchocinetid shrimp, Rhynchocinetes striatus sp. nov., is described and illustrated. It is characterized by a unique color pattern of alternate white-and-red bands around the body, as well as by the very slender, elongate rostrum bearing numerous ventral teeth, the antennal scale bearing a very short distolateral spine never reaching the end of the lamella, and the relatively slender posterior three pereopods bearing three ventral spinules on the dactylus. Its affinities to the related species are discussed. INTRODUCTION In several recent popular publications, beautiful color photographs of shallow water decapod crustaceans, including some shrimps of the genus Rhynchocinetes, were presented [1, 2, 6]. Among these crustaceans is a rather large rhynchocinetid which is characterized by obliquely or transversely arranged, alternate white-and-red bands on the carapace and abdomen, the color pattern being apparently different from those of the known species in this genus. In addition to the material reported earlier by one of us [6], several specimens referable to this species have been collected from the Ryukyu Islands, southern Japan. Close ex- amination discloses that the species is closely re- lated to R. hiatti Holthuis and Hayashi, R. hender- soni Kemp, or R. rigens Gordon, but undoubtedly is referred to a new species described below. The holotype will be deposited in the National Science Museum, Tokyo (NSMT) and the para- types are in the Shimonoseki University of Fisher- ies (SUF) and Sabiura Marine Park Research Station (YMP). Carapace length (CL) and ros- trum length (RL) are used for measurements. Accepted July 22, 1991 Received May 5, 1991 DESCRIPTION Rhynchocinetes striatus sp. nov. (Figs. 1-4) Rhynchocinetes sp. Debelius, 1983, p. 68, with fig. [1]. Rhynchocinetes sp. Debelius, 1984, p. 68, with fig. [2]. Rhynchocinetes sp. Kamezaki, Nomura, Hama- no and Misaki, 1988, p. 74, with fig. [6]. Material.—Holotype: Ovigerous ? (NSMT, 18.1 mm in CL, 29.8mm in RL), Kadena Port, Okinawa Island, 1-10 m deep, April 5, 1989, H. Masuda leg. Paratypes: 18 (SUF 530-2-1360, 17.2 mm in CL), 12 (SUF 530-2-1361, 19.1 mm in CL), 2¢ (YMP-550, 16.0 mm in CL, YMP-553, 9.5 mm in CL): Hori, Kuroshima Island, Yaeyama Group, 3 m deep, June 27-28, 1987, K. Nomura leg; 2¢ (YMP-611A, 15.9mm in CL, YMP-613B, 12.9 mm in CL): Hori, Kuroshima Island, Aug. 2, 1987, K. Nomura leg: 1 juv. (SUF 530-3-1362, 7.3 mm in CL): Hori, Kuroshima Island, November 27, 1987, K. Nomura leg. Description.—Shell rather hard, with many transverse striae on carapace. Rostrum 1.3-2.2 times as long as carapace, overreaching antennal scale by nearly distal half of its length; longer in larger male (Figs. 1, 2a). Three teeth on carapace 200 K. NOMURA AND K. HAYASHI Sea pe uae NX, = Res Fic. 1. Rhynchocinetes striatus sp. nov. Male paratype (YMP-550), showing color pattern; dotted parts are red. directly behind rostral articulation. Dorsal margin of rostrum with 2 rather large teeth on proximal part, distal part unarmed except for 2 small sub- apical teeth. Ventral margin with 11-13 teeth, 12 teeth in 4 specimens including holoytpe, 11 teeth in 1 specimen and 13 teeth in 2 specimens. Antennal spine well developed. Pterygostomial angle largely rounded (Fig. 2a). Abdomen also with fine striae, oblique in first 2 somites, nearly longitudinal in third somite, and transverse in last 3 somites. Pleura of first 3 somites rounded. Fourth somite with strong, post- eriorly directed spine on posterior margin just above base of pleuron, posterolateral angle of pleuron with or without small spine in male. Three males (9.5, 12.9, 17.2 mm in CL) bearing this small spine on both sides but 2 males (15.9, 16.0 mm in CL) lacking it on either side. In females including holotype, pleuron produced posteriorly but not sharply pointed, lacking spine. Pleuron of fifth somite with stronger spine near base and well developed one on posterolateral agle in both sexes (Fig. 2b). Sixth somite about 1.5-1.8 times as long as fifth somite. Telson 1.1-1.3 times as long as sixth somite, with 3 pairs of dorsal spines on posterior half of its length; posterior margin end- ing in triangular median point, with 3 pairs of spines; lateral pair short, median pair longest (Fig. Dey Eyes very large. Cornea broad and rounded, with semicircular ocellus. Stylocerite sharply pointed at distal end, with anteriorly directed dorsal tooth at base, falling slightly short of end of antennular peduncle. First antennular segment with small spine at anterolateral angle. Second and third segments unarmed and subequal in length (Fig. 2d). Large male paratype, 17.2mm in CL, with nearly intact antennular and antennal flagella; upper antennular flagellum relatively long, about 4 times as long as its peduncle, basal part slightly overreaching rostral apex, more or less enlarged, bearing setae ventrally; lower flagellum longer than upper, about 5.5 times as long as peduncle. Antennal flagellum much longer, about 9 times length of carapace. A New Rhynchocinetid Shrimp 201 Fic. 2. Rhynchocinetes striatus sp. nov. Male paratypes (a, c-g: YMP-550; b: YMP-553). a, carapace, b, fourth and fifth abdominal somites, c, telson, d, antennular peduncle, e, antennal scale, f, endopod of first pleopod, g, endopod of second pleopod. Antennal scale 2.6 times as long as broad, distolateral tooth not reaching end of lamella (Fig. 2e). Basicerite with sharply pointed large lateral spine and rounded dorsal lobe, small spiniform process on membranous articulation with antennal scale. Mouthparts (Fig. 4) and branchial formula typical of genus. Mandibular palp _ three- segmented (Fig. 4a). Palp of first maxilliped also three-segmented, distal segment very small (Fig. 4d). Third maxilliped and first 3 pereopods with pleurobranch and arthrobranch, posterior 2 pereopods with pleurobranch only. Epipods dis- tinct on all maxillipeds (Fig. 4d-f) and first 3 pereopods; exopods on all maxillipeds. Third maxilliped reaching nearly to distal end of antennular peduncle in both sexes; distal segment with about 10 dark spinules distally; basal segment with slender spine on anterolateral end and 2 or 3 movable spines on anteroventral corner; exopod 202 K. NOMURA AND K. HAYASHI a-e —— Bra im i ae) | 1mm Fe, Fic. 3. Rhynchocinetes striatus sp. nov. Male paratype (YMP-550). a, first pereopod, b, second pereopod, c, third pereopod, d, fourth pereopod, e, fifth pereopod, f, dactylus of third pereopod, g, proximal part of propodus of fourth pereopod. A New Rhynchocinetid Shrimp 203 Fic. 4. Rhynchocinetes striatus sp. nov. Female paratype (SUF 530-2-1361). a, mandible, b, maxillule, c, maxilla, d, first maxilliped, e, second maxilliped, f, third maxilliped. relatively long, extending beyond distal end of basal segment (Fig. 4f). First pereopod reaching distal end of carpocerite of antennal peduncle in both sexes. Dactylus with 8-10 dark claws distally, fixed finger with 3 similar claws. Palm twice as long as dactylus in female, but more longer in male. Carpus equally long as merus or palm in female, and as long as merus but shorter than palm in male (Fig. 3a). Second pereopod reaching in female or slightly over- reaching in male distal end of first pereopod. Chela slightly longer than merus. Dactylus with 14 black claws, fixed finger with 3 similar claws. Palm 2.4—3.0 times as long as dactylus. Carpus 2.0—2.4 times as long as palm. Ischium shorter than merus (Fig. 3b). Posterior 3 pereopods slender. Third pereopod overreaching antennal scale by dactylus and part of propodus. Dactylus short, bearing 3 black spinules excluding strong terminal claw (Fig. 3f). Prop- odus 5-8 times as long as dactylus, with about 10 posterior marginal spinules. Carpus shorter than propodus, lateral surface with 2 or 3 spinules. Merus longer than propodus, with 5-8 lateral and 204 K. NOMURA AND K. HAYASHI 3-4 posterior spinules. Ischium with 2 spinules, 1 lateral and 1 posterior (Fig. 3c). Fourth pereopod overreaching antennal scale by dactylus. Dactylus short, with 3 claws as in third pereopod. Propodus 6-9 times as long as dactylus, posterior margin with 6-14 spinules; proximal 4 or 5 of them bearing several setae. Carpus about 2/3 length of propodus, with 2 or 3 lateral spinules. Propodus and carpus of fourth pereopod slightly longer than respective segments of third pereopod. Merus of fourth pereopod slightly shorter than that of third pereopod, as long as or shorter than propodus of fourth pereopod, lateral margin with 5-9 spinules, posterior margin with 3-7 spinules. Ischium with 1 lateral and 1 posterior spinule (Fig. 3d, g). Fifth pereopod reaching dista! end of antennal scale. Dactylus with same armature as in 2 preceding pereopods. Propodus as long as that of fourth pereopod, with about 10 spinules on posterior margin, lacking proximal setose spinules as in fourth pereopod. Carpus as in fourth pereopod. Merus apparently shorter than propodus, as well as those of 2 preceding pereopods, bearing 5—6 lateral and 1 or 2 posterior spinules. Ischium with 1 lateral spinule, posterior spinule occasionally absent (Fig. 3e). In male, 17.2mm in CL, endopod of first pleopod broad leaf-shaped, distally bearing two- lobed appendix interna arising from midpoint of mesial margin of endopod. Lateral and distal margins of endopd lacking setae or hairs, but mesial margin with short simple setae, mesial margin of appendix interna with rather long simple setae (Fig. 2f). In females, endopod of first pleopod similar to, but larger than exopod covered with long plumose marginal setae; appendix inter- na absent. Appendices interna and masculina on endopod of male second pleopod short rod- shaped. Appendix masculina as long as or slightly shorter than appendix interna, with numerous long setae (Fig. 2g). Color in life. —Body with transversely arranged alternate bands of red and white (Fig. 1) [1, 2, 6]. Rostrum with 3-5 red bands and white apex. Carapace with 4 red bands running obliquely back- ward on lateral surface, anterior bands short, usually disappearing dorsolaterally. Abdomen with 8 transverse red bands; anterior 2 bands narrow, situated on anterior margins of first and second somites; third band present along posterior margin of second somite; fourth band broad, placed on midtransverse portion of third somite; fifth band broad dorsally, covering posteromedial cap of third somite and entire dorsal surface of fourth somite, extending downward along post- erior margin of fourth somite; sixth band along posterior articulation of fifth somite; seventh band present on posterior margin of sixth somite, ex- tending onto base of tail fan; last band placed on posterior part of tail fan. Sixth somite with oblong red patch on central dorsal portion. Pereopods yellowish white dorsally and reddish laterally. Meri of last 3 pereopods with red band near distal articulation, chela of first pereopod with similar red band at base. Antennal flagellum reddish. The above-mentioned pattern is rather stable and nearly the same in both young and adult. Etymology.—The Latin — striatus (striated) alludes to the unique color pattern of the species. DISCUSSION The new species belongs to the species group characterized by having three teeth on the cara- pace behind the articulation with the rostrum [4], in which close relative Rhynchocinetes hiatti, R. hendersoni and R. rigens are included. They are known from nearly the same tropical area as the new species [3]. The type specimens of the new species were compared with the original and subse- quent descriptions of the related species [3-5, 7], as well as with the following reference material collected from the Japanese waters. R. hiatti: 1 ovigerous $, 13.2 mm in CL, from Iriomote Island. R. hendersoni: 1 broken @ , from Suruga Bay; 1¢, 12.9 mm in CL from Okinawa Island; 6 ¢ , 10.5- 11.8 mm in CL, 32, 6.5—9.0 mm in CL, 2 ovig. 2 | 7.3-8.8 mm in CL from Kuroshima Island. R. rigens: 19, 13.0mm in CL from Okinawa Island. The color pattern is the most distinctive charac- ter to recognize this new species, because it is constant and does not change by sex and growth. R. rigens has a spotted pattern [3, 6] and R. A New Rhynchocinetid Shrimp hendersoni has a mottled one [6]. The new species and R. hiatti show a banded pattern of alternate white-and-red bands, but their details are clearly different from each other. In R. hiatti, the ground color is red, the abdomen bears two to four narrow white transverse bands, the carapace has two or three oblique or longitudinal bands, and the ros- trum is whitish apically and reddish elsewhere [5, 6]. In the new species the white-and-red bands are conspicuous, especially in dorsal view. There are three to five oblique red bands on the carapace, eight red bands on the abdomen including the tail fan and three to five bands on the rostrum [1, 2, 6]. The differences that help to separate the new species from the three related species are given in Table 1. The pterygostomial angle is rounded in R. striatus and R. rigens, while pointed in R. hiatti and R. hendersoni. The stylocerite is much longer in R. hiatti, exceeding beyond the distal end of antennular peduncle. The three known species have a short stylocerite, just reaching or slightly overreaching the end of the antennular peduncle. TABLE 1. 205 The fifth pereopod is relatively long in R. striatus, extending as far forward as the distal end of the antennal scale, instead of reaching to the base of it as in the other three species. In the largest male of R. striatus, the third maxilliped is not extremely long as the one shown in R. kuiteri [1, 2, 8]. Unusually long first pereopods are sometimes seen in the male of R. hendersoni [6] but such examples have not been reported for the other three species including the new species. The lower rostral teeth of the new species are more numerous than those of the other species. R. striatus bears always more than 10 lower rostral teeth, this number is usually less than 11 in the other three species. The dactyli of the last three pereopods bear three spinules in R. striatus and R. hendersoni but two spinules in R. hiatti and R. rigens. The merus of the third pereopod bears three or four spinules in R. striatus and R. hender- soni, two spinules in R. rigens and only one spinule in R. hiatti. The distolateral spine of the antennal scale in R. striatus is short, never reaching the end Comparison of characters among Rhynchocinetes striatus sp. nov. and three related species Color pattern Rostrum Tail fan Pterygostomial angle Stylocerite Distolateral spine of antennal scale Third maxilliped in large male First pereopod in large male Fifth pereopod Number of lower rostral teeth Number of spinules on dactyli of last three pereopods Number of posterior spinules on merus of third pereopod R. striatus sp. nov. R. hiatti banded about 5 bands 2 transverse bands rounded reaching distal end of antennular peduncle falling short of distal end of lamella falling short of distal end of antennal scale normal overreaching antennal scale eS 3 3 or 4 banded white rostral apex without band pointed exceeding beyond distal end of antennular peduncle overreaching lamella overreaching antennal scale normal reaching base of antennal scale 8-11 2 R. rigens spotted rounded extending to or a little beyond end of antennular peduncle feebly overreaching lamella falling short of distal end of antennal scale normal overreaching antennular peduncle 8-12 2 R. hendersoni mottled pointed not reaching distal end of antennular peduncle reaching to or slightly over- reaching lamella overreaching antennal scale unusually long overreaching antennular peduncle 8-10 3 3 On 4) 206 of the lamella, while it fully overreaching in the three known species. | As mentioned above, the present species was briefly introduced as Rhynchocinetes sp. in earlier papers by living color photographs [1, 2, 6]. This is a shallow water species, usually found on coral reefs. The type-series were collected from the Ryukyu Island. Debelius’ specimen was re- ported from the Great Barrier Reef, Australia [1, 2]. ACKNOWLEDGMENTS We are grateful to Dr. Lipke B. Holthuis of the Nationaal Natuurhistorisch Museum Leiden, for reading a draft of the manuscript. REFERENCES 1 Debelius, H. (1983) Gepanzerte Meeresritter. 120 pp. Kernen Verlag, Essen. 2 K. NOMURA AND K. HAYASHI Debelius, H. (1984) Armoured knights of the sea. 120 pp. Kernen Verlag, Essen. Fujino, T. (1975) Occurrence of Rhynchocinetes rigens Gordon, 1936 (Crustacea, Decapoda, Rhyn- chocinetidae) in the Indo-Pacific region. Publ. Seto Mar. Biol. Lab., 22: 297-302. Gordon, I. (1936) On the mecruran genus Rhyn- chocinetes with the description of a new species. Proc. Zool. Soc. London, 1936: 75-88. Holthuis, L. B. and Hayashi, K. (1967) A new species of shrimp, Rhynchocinetes hiatti (Crustacea, Decapoda). Annot. Zool. Jap., 40: 161-170. Kamezaki, N., Nomura, K., Hamano, T. and Misaki, H. (1988) Crustacea. In: Marine Park Center (ed.) Illustrated Marine Organisms in Okinawa Islands, 8: 232 pp. Southern Press, Okinawa (in Japanese). Kemp, S. (1925) Notes on Crustacea Decapoda in the Indian Museum. XVII. On various Caridea. Rec. Ind. Mus., 27: 249-343. Tiefenbacher, L. (1983) A new species of Rhynchoci- netes from South-Australia (Crustacea, Decapoda, Rhynchocinetidae). Rev. fr. Aquariol., 9: 121-124. ZOOLOGICAL SCIENCE 9: 207-209 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Electron Spin Resonance Spectrometry of Vanadium Ions in the Blood Cells of the Ascidian, Ascidia gemmata Junko Hirata! and Hirosut MICHIBATA~ ‘Department of Biology, School of Dental Medicine, Tsurumi University, Tsurumi, Yokohama 230, and *Mukaishima Marine Biological Laboratory, Faculty of Science, Hiroshima University, Mukaishima-cho, Hiroshima 722, Japan ABSTRACT—Results have been obtained to refute a previous report [1] that ascidian blood cells incorporate anions of vanadium in the +5 oxidation state (vanadate) via anion channels and reduce them rapidly to vanadium in the +4 oxidation state (vanadyl), with subsequent further reduction to the +3 oxidation state. Using *8V-labeled vanadium compounds, we demonstrated previously that incorporation of vanadium into ascidian blood cells occurs slowly and with biphasic kinetics [2]. In an attempt to determine whether the reduction of vanadium in the +5 oxidation state occurs as quickly as postulated [1], we performed an electron spin resonance study and found that vanadium in the +5 oxidation state is not reduced immediately to vanadium in the +4 oxidation state after its addition to a suspension of blood cells. INTRODUCTION Ascidians accumulate large amounts of vana- dium ion from seawater in specialized blood cells that are called vanadocytes [3, 4]. Although the vanadium dissolved in seawater seems to be present as vanadate anions in the +5 oxidation state, almost all the vanadium that accumulates in ascidian blood cells is reduced to vanadyl cations in the +3 oxidation state [1, 5-12]. Dingley et al. [1] observed by ESR (electron spin resonance) spectrometry that a rapid reduction of vanadium in the +5 oxidation state (vanadate Accepted October 9, 1991 Received September 3, 1991 2 All correspondence should be addressed to Dr. Hitoshi Michibata species) to vanadium in the +4 oxidation state (vanadyl species, VO**) occurred within blood cells of Ascidia nigra after the addition of exoge- nous vanadium in the +5 oxidation state, and that subsequently the ESR signal due to the vanadyl species decreased dramatically, an indication that this chemical species was further reduced to vanadium in the +3 oxidation state, which does not generate an ESR signal. Using “*V-labeled vanadium compounds, we demonstrated previously that incorporation of vanadium into ascidian blood cells occurs slowly and with biphasic kinetics [2], in striking contrast to the results of Dingley et al. [1]. Therefore, the present experiments were designed to reexamine, by ESR spectrometry, whether the reduction of vanadium in the +5 oxidation state does actually occur very rapidly. MATERIALS AND METHODS Ascidians, Ascidia gemmata, were collected at the Asamushi Marine Biological Station of Toho- ku University, Aomori, Japan. Blood, withdrawn by cardiac puncture at 4°C, was suspended in isotonic HEPES (N-2-hydroxyethylpiperazine-N’- 2-ethanesulfonic acid)/NaCl buffer solution (50 mM HEPES/0.5M NaCl, pH8.0). So-called washed cells were obtained by centrifugation of the blood for 10 min at 130g. One ml of the suspension of washed cells contained 2.5 x 10’ to 5 x10’ cells. 208 J. HiRATA AND H. MICHIBATA Ortho-Na3VO, (Wako Pure Chemical Industries Ltd., Osaka, Japan), dissolved in double-distilled water at a concentration of 100 mM, was diluted with the above mentioned buffer solution to a concentration of 5.5mM. Ten wl of a 5.5mM solution of o-Na3VQ, solution were added to 200 ul of the suspension of washed cells of which had been placed in a ESR quartz tube for ESR measurements. Each such tube was frozen, to stop any reaction, at 77 K at regular time intervals and submitted to ESR spectrometry. The conditions for ESR measurements and the procedures for quantification of vanadium in the +4 oxidation state can be found in a previous paper [12]. RESULTS AND DISCUSSION The ESR signals due to vanadyl species (VO7T , the +4 oxidation state of vanadium) were re- 380 420 260 300 340 mT Fic. 1. The ESR spectra of vanadium at 77 K. (a) ESR spectrometry of vanadyl (+4) species de- rived from living washed blood cells (g ,, 1.935; A ,, 206.3 g). (b) ESR spectrometry, 2 min after the addition of an exogenous solution of vanadium in the +5 oxidation state to the suspension of washed blood cells. Little change in the intensity or the pattern of ESR signals was recorded. Vanadium, a multivalent metal, is present in living organisms in the +5, +4, and +3 oxidation states. Among them, vanadium in the +3 and +5 oxida- tion states are ESR silent but the chemical species of vanadium in the +4 oxidation state are detectable. Therefore, if the exogenous vanadium in the +5 oxidation state were reduced to vanadyl (+4) spe- cies, a temporary and dramatic increase in the intensity of ESR signals should be observed. corded from washed cells as shown in Figure 1a, and the results were in accord with those in the previous report [12]. Only 2.4% of the total vanadium included in the vanadocytes of A. gemmata is present as the vanadyl species and the remainder is in the +3 oxidation state [12]. ESR spectrometry, 2 min after the addition of an exogenous solution of vanadium in the +5 oxidation state to the suspension of blood cells, revealed that no increase of intensity of the signal due to the vanadyl species occurred (Fig. 1b). Subsequently, little change in the intensity or the pattern of ESR signals was recorded over the course of 5 min (Fig. 2). < 100 = 3 80 © £ = 60 2 + Lok 2) sheen pile aE ac O 1 2 3 4 5 Time (min) Fic. 2. Changes in the intensity of ESR signals due to the vanadyl species after addition of exogenous solution of vanadium in the +5 oxidation state to the suspension of washed blood cells over the course of 5 min. No increase of intensity of the signal, which was calculated based on signal height at the highest line of ESR signals, was recorded in the present ex- perimental conditions. Dingley et al. [1] reported that the intensity of ESR signals due to vanadyl species increased to about 120% of the initial value within 1 min after the addition of exogenous vanadium in the +5 oxidation state and then decreased to the initial value 5 min later. They interpreted these changes as indications that the incorporated vanadium (the +5 oxidation state) was reduced immediately to the vanadyl species (the +4 oxidation state) and, subsequently, the vanadyl species was further reduced to the +3 oxidation state. If exogenous vanadium (in the +5 oxidation state) incorporated to the blood cells is reduced to the vanadyl species within a short time, as Dingley et al. suggested [1], a temporary and dramatic ESR Spectra of Ascidian Vanadium 209 increase in the ratio of vanadyl ions to total vanadium should be observed immediately after the addition of the exogenous vanadium to the suspension of blood cells. However, in the present experiments, not only was little change observed in the intensity of ESR signals due to the vanadyl species (VOT, the +4 oxidation state), but also no temporary increase in the ratio of the vanadyl species to the total vanadium content nor any change in the pattern of ESR signals was observed (Figs. 1 and 2). What differs is the fact that Dingley et al. [1] measured the ESR signals after pouring the reaction mixture of vanadium solution and the cell Suspension into a quartz ESR tube but we re- corded the ESR signals after addition of exoge- nous vanadium to the cell suspension which had been placed in a ESR tube. We can not, however, conclude hastily that these different procedures are linked with obvious discrepancies between their and our results. Using **V-labeled vanadium, we demonstrated previously that the blood cells of A. gemmata incorporate vanadium in oxidation states +4 and +5, with biphasic kinetics and at a very low rate of bout 2 107° ng/10° cell/min [2], in striking con- trast with previously reported results [1]. We can conclude, therefore, that since exogenous vana- dium in the +5 oxidation state is not incorporated quickly to the blood cells, it is not reduced rapidly to vanadyl species but such reaction occur slowly after the addition of vanadium in the +5 oxidation state to blood cells of A. gemmata. Recently, we demonstrated that a vanadium- binding substances, extracted from the vanado- cytes of ascidians and named vanadobin, can reduce exogenous vanadate (V) [13]. Experiments with vanadobin will help us elucidate the mechan- ism of the reduction-oxidation reaction of vana- dium in vitro and in vivo. ACKNOWLEDGMENTS We express our sincere thanks to Dr. T. Numakunai and the other members of the staff of the Marine Biological Station of Tohoku University at Asamushi for supplying the materials. Thanks are also due to Dr. A. Takeuchi of Toyama University for generously making his ESR spectrometer available to us. This work was supported in part by Grants-in-Aid from the Ministry of Education, Science and Culture, Japan (401480026 and + 01304007), the Naoji Iwatani Memorial Foundation for the Promotion of Science and Technology to H. M. and was also supported partially by the fund from the Sasagawa Science Research Subsidy of the Japan Science Society to J. H. REFERENCES 1 Dingley, A. L., Kustin, K., Macara, I. G. and McLeod, G. C. (1981) Biochim. Biophys. Acta, 649: 493-502. 2 Michibata, H., Seki, Y., Hirata, J., Kawamura, M., Iwai, K., Iwata, R. and Ido, T. (1991) Zool. Sci., 8: 447-452. 3. Michibata, H. (1989) Zool. Sci., 6: 639-647. 4 Michibata, H. and Sakurai, H. (1990) In: Vanadium in Biological Systems. Ed. N. D. Chasteen, Kluwer Acad. Publ., Dordrecht, 153-171. 5 Lybing, S. (1953) Alder. Arkiv Kemi., 6: 261-269. 6 Bielig, H.-J., Bayer, E., Califano, L. and Wirth, L. (1954) Publ. Staz. Zool. Napoli, 25: 26-66. 7 Boeri, E. and Ehrenberg, A. (1954) Biochem. Biophys., 50: 404-416. 8 Webb, D. A. (1956) Publ. Staz. Zool. Napoli, 28: 273-288. 9 Tullius, T. D., Gillum, W. O., Carlson, R. M. K. and Hodgson, K. O. (1980) J. Am. Chem. Soc., 102: 5670-5676. 10 Frank, P., Carlson, R. M. K. and Hodgson, K. O. (1986) Inorg. Chem., 25: 470-478. 11 Lee, S., Kustin, K., Robinson, W. E., Frankel, R. B. and Spartalian, K. (1988) Inorg. Biochem., 33: 183-192. 12 Hirata, J. and Michibata, H. (1990) J. Exp. Zool., 257: 160-165. 13. Michibata, H., Morita, A. and Kanamori, K. (1991) Biol. Bull., 181: 189-194. Arch. Aa » et » ah x} ae FAR rt: i a bes wade 9 Lagmeeaives A. ance : is vein be eyHiqontt md ‘a ATE +o) 33 AO TON AO ERO inital ie peck peers is ee uae s ae os! Lidee ins variant aS 8 8 -e.: 16551 —— thi a Stewet Fae. AM ae oe “ pai ees i 7 De ate pas afte. a eben Hy 4a. bei 4 | RY a rhews nak: sas Oe Ds ; son URES AN i ; be det ae ix sai a a ray rat | wi “ S 2 Dn are “ai 2 hi wa ; x. waa} ve “: towhee Le dy bi vee £ Oy Bits 7 iy ie y Eo ia 5 F bai ; Leh, ° ee. Bs © ats xe pe ca h (his ae ae palttine ae ‘potas . a j ; é - € £ s. ; be : 3 oe ¢ i ~ Met! AAS bse WES oo Flite de cee Se nso = we . Ts = a ¢ ~ pe hey = 7 a are b> ATER vies ' ; { i —- OE eo ar” caine” Ty oon ee a Ba. Gaye recs, & HOt] 4 oe fi batt s ew) é * ’ uve iP 2) P x i rn ‘ We . ‘ 4 i es Pe a i ae : 4 : ? ; * ~ fs airy As Ag I f hye! } ae TE aig pe As : '2 j GEVP Ve) ee et} t 5 » if or Pe if Are} . : < F "s 1 * j ¥ rt = ~ bab : % ‘ “ ' - f 2 A ¥ : 4. i . ‘ .. 2 . fs i ree wee oes J i 4 : ‘ A » = c a2 2 : ; ; . i : Ges o ah j ¢ 4 ; , " * ~ 4 - - , u a0 : ro 3 ; ‘ r « aew ~ ar ; f f , a Wid Gt 0 % oe . =, ? i & ay PA fi f i i i ae = ee | wa ¥ ne F ‘ 4 i v 1 : eRe , 4 . 5 t 4 ¢ " a ie : i= ie Ra re i - \ 5. t ae : Z “ ; ) - - Fs ) 2 ‘ “) j es oe -; zi E : i £ = 7 j = Sea “4 Pe ws Q j 2 i os } } x “ r ? 4 Y ZOOLOGICAL SCIENCE 9: 211-217 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Localization of Retinochrome in the Retina of a Tetrabranchiate Cephalopod, Nautilus pompilius Tomiyuki Hara’, Retko Hara’, Axio KisHicAmt’, YUTAKA KosHIDA’, SHinrI Horiucui’, MASAO YOSHIDA®, MASAMICHI YAMAMOTO’, Taicuiro Goro? and U. Ras* "Department of Biology, Faculty of Science, Osaka University, Toyonaka, Osaka 560, *Department of Biology, Faculty of General Education, Osaka University, Toyonaka, Osaka 560, *Ushimado Marine Laboratory, Faculty of Science, Okayama University, Ushimado, Okayama 701-43, Japan, and “Institute of Marine Resources, the University of the South Pacific, Suva, Fiji ABSTRACT— In metacephalopods (Dibranchia) such as squid and octopus, vision is maintained by the photoreac- tion of a pair of photopigments, rhodopsin and ret- inochrome. The present work was aimed at elucidating whether or not the same process is applicable to the lensless eye of protocephalopods (Tetrabranchia) which may still preserve many ancestral functions. The retina of Nautilus pompilius was used for experiments. The nautilus retina is mainly composed of two kinkds of cells, visual and supporting cells. The former can easily be distinguished from the latter by a spherical myeloid body present in the cell body. The distal process of the visual cell is surrounded by brushes of microvilli to yield rhabdomeres, and the supporting cell has long microvilli which extend forward beyond the level of the distal end of the visual cell to form the sufrace layer of the retina. The myeloid body of this animal appears to be a compact mass of lamellated membrane structures, located in the perinuclear region far away from the rhabdomeres. According to epifluorescence microscopy, rhodopsin is contained in the rhabdomes, while retinochrome is concentrated in the myeloid bodies. Tetrabrachia retinas are thus provided with the same dual system of photopig- ments found in Dibranchia. INTRODUCTION For studies on vision in invertebrates, meta- Accepted October 11, 1991 Received May 27, 1991 cephalopods (Dibranchia) such as squid and octo- pus have been used as widely as insects, and they have provided much of the knowledge concerning fine structure and photoreceptive mechanisms of the retina. The cephalopod visual cells possess two kinds of photopigment systems, each containing rhodopsin and retinochrome. The rhodopsin sys- tem is located in the rhabdomal microvilli of the outer segments of visual cells, while the ret- inochrome system is associated with lamellated bundles of membranes (called myeloid bodies) mainly present in the inner segments. When retinochrome absorbs light, its retinal chro- mophore is isomerized from all-trans to the 11-cis form [1, 2]. As this change is just the reverse of what occurs in the photoisomerization of rhodop- sin, the 11-cis photoproduct of retinochrome (metaretinochrome) acts as a supplier of the 11-cis-retinal required for rhodopsin formation [3, 4]. As recently shown, these two photopigment systems are functionally linked by a retinal-binding protein (RALBP), which serves to transport the 11-cis-retinal to the rhodopsin system and to carry away the all-trans-retinal toward the retinochrome system [5, 6]. Consequently, in such a rhodopsin- retinochrome conjugate system, metarhodopsin and metaretinochrome can interchange their reti- 22. T. Hara, R. Hara et al. nal chromophores through the mediation of RALBP to return to their original photopigments, rhodopsin and retinochrome. This conjugate sys- tem in cephalopods is noted to be an essential mechanism for visual phtoreception and its con- tinuation [7-9]. When retinochrome was discovered in the Japanese common squid, Todarodes pacificus [10], we were deeply interested in whether or not retinochrome was also contained along with rho- dopsin in the lensless eyes of protocephalopods (Tetrabranchia) which are considered to be ances- tors of the present-day dibranchiate cephalopods. An anatomical report on the myeloid body that was observed as a compact mass by Barber and Wright [11] has strongly motivated us to investi- gate the photopigments in the nautilus retina. The 50th Anniversary Research Project of Osaka University (an expedition to the South Pacific Region) afforded us an opportunity to study the nautilus retina in cooperation with the University of the South Pacific. The present experiments were carried out with a group under the tutelage of the late Professor Masao Yoshida of Okayama University, who shared a common interest with us for many years. With the aid of earlier research [11-13], we here describe the fine structure and related photopigments of the nautilus retina with special refence to the presence of retinochrome. Some of the results were presented at the 7th International Congress of Eye Research held in Nagoya in September 1986 and at the 8th Annual Meeting of the Japan Society for General and Comparative Physiology held in Hiroshima in November 1986 [14]. MATERIALS AND METHODS Nautilus pompilius were decoyed into a baited trap left at a depth of about 500 m outside the main reef at Suva on the island of Fiji in October, 1985. Some of the animals were kept alive in cold seawater, transported directly to Osaka by air, and were used for microanatomy studies of the retina. Some of the animals were adapted overnight to darkness after capture, and isolated retinas were transferred into an embedding compound for epifluorescence microscopy, kept on dry ice and brought to Osaka from Fiji. Most of the animals were similarly dark-adapted, and their heads, kept in the dark on dry ice, were also sent to Osaka by air for biochemical analyses. For electron microscopy, pieces of the retina were fixed in 2.5% glutaraldehyde in 0.1% cacody- late buffer (pH 7.4) containing 0.4 M sucrose for 2 hr, washed with buffer, and postfixed in 1% osmium tetroxide in the same buffer for 1 hr. After fixation they were dehydrated through a graded series of ethanol, transferred into propyl- ene oxide, and embedded in epoxy resin. Thin sections were stained with alcoholic uranyl acetate followed by lead citrate, and examined in a Hitachi model H-500H electron microscope [15]. For light microscopy, the material prepared for electron microscopy was sectioned at 0.5 um, and stained with toluidine blue. Retinochrome is readily reduced by sodium borohydride (NaBH,) into an N-retinyl protein, which emits a characteristic yellow-green fluores- cence under near-ultraviolet (NUV) light [2]. Rhodopsin is also converted to a similar fluores- cent product, but only when denatured before reduction. Based on this difference, the location of retinal-bearing photopigments in the retina was determined by means of epifluorescence micros- copy [16, 17]. The dark-adapted retina embedded in Bright Cryo-M-Bed embedding compound was sectioned at 10 u~m with a Bright model 5030/ WDE rotary microtome set up in a cryostat adjusted to —25°C, mounted on glass slides, and air-dried for about 4 hr at room temperature. One specimen was immersed in a 0.2% aqueous solu- tion of NaBH, for 1 sec at 4°C, rinsed gently with water, excited by a 334-nm beam and photo- graphed to locate the yellow-green fluorescence of reduced retinochrome using an Olympus BHA- RF-A epifluorescence microscope. Another speci- men was treated with 20% formaldehyde (HCOH) for 1 min and 100% methanol (CH3OH) for 2 min at 20°C to denature rhodopsin, washed with water, immersed in 0.2% NaBH, for 5 sec, rinsed with water and photographed. In this case, the speci- men gave a micrograph of the fluorescence due to reduced products of both retinochrome and rho- dopsin. Nautilus Retinochrome 23) RESULTS Structure of nautilus retina Under the light microscope, the nautilus retina (about 500 um thick) is seen to consist of four layers, the rhabdome, black pigment, nucleated cell bodies and the nerve plexus (Fig. 1), though it consists mainly of two kinds of cells, visual and supporting cells. These microscopic features are similar to those observed in Dibranchia such as squid and octopus [2], but the cells of nautilus and dibranchiate cephalopods differ greatly in fine structure, as shown later by electron micrographs (cf. Fig. 2). Both cells are longitudinally arranged parallel to each other. The rhabdome layer is composed of processes of visual cells and long Fic. 1. Light micrograph of a longitudinal section of the central part of the Nautilus retina. A 0.5 wm epoxy resin section stained with toluidine blue. R, rhab- dome layer; B, black pigment layer; N, nucleated cell body layer; P, nerve plexus layer; «, a surface layer consisting of the tips of the long microvilli of supporting cells. Scale bar=100 um. microvilli of supporting cells, but is largely occu- pied by numerous rhabdomeric microvilli that emanate from the visual cell processes. Especially, the tips of the long microvilli of supporting cells form an additional layer to cover the rhabdome layer, as indicated by an asterisk in Figure 1. In the distal parts of both visual and supporting cell bodies, pigment granules are densely packed to form the black pigment layer. In relation to the above observations, electron micrographs of longitudinal sections of the nucle- ated cell body layer are presented in Figure 2. This layer is mainly constituted of somata of visual and supporting cells. Both cell bodies contain a great number of the black pigment granules, but the former possesses a ball-shaped organelle close to the nucleus, easily distinguishable from the latter (Fig. 2A). As previously shown by Muntz and Wentworth [13], such a spherical organelle is a myeloid body, which exhibits a variety of complex appearances depending on the plane of section (Figs. 2A and B). The superficial shape of the myeloid body is quite dissimilar to the lamellated bundles that are so familiar in other cephalopods. However, it is a compact mass which consists principally of basic piles of wavy membranes, as seen in Figure 2C. Localization of retinal phtopigments In order to detect the retinal-bearing photopig- ments (rhodopsin and retinochrome) contained in the nautilus retina, the following experiments were carried out using our routine histochemical techni- ques [16, 17]. Figure 3A presents a frozen section of the retina. It scarcely showed any fluorescence under NUV light, suggesting the absence of retinol and retinylester. When this nonfluorescent speci- men was treated with NaBH, and placed again under NUV light, many bright spots of yellow- green fluorescence were observed behind the black pigment layer, proving the presence of ret- inochrome. When the same specimen was dena- tured with HCOH and CH30OH and treated with NaBHy, the characteristic yellow-green fluores- cence of reduced photopigments appeared on both sides of the black pigment layer (Fig. 3B). Outside of this layer, a broad band of fluorescence covered the rhabdome layer, indicating the presence of 214 T. Hara, R. Hara et al. Fic. 2. Transmission electron micrographs of the nucleated cell bodies of the Nautilus retina. A) Longitudinal section showing the presence of two types of cells, a visual and a supporting cell. Vn, nucleus of visual cell: Sn, nucleus of supporting cell; Mb, myeloid body. Scale bar=5 um. B) Myeloid bodies at higher magnification. V, visual cell; S, supporing celll. Scale bar=2 ~m. C) Lamellated membranes of myeloid body. Scale bar=0.5 um. rhodopsin in the rhabdomeres. On the opposite side, many fluorescent spots were scattered in the nucleated cell body layer, whose pattern was the same as observed previously with the undenatured specimen. These spots seemed to correspond to organella probably carrying retinochrome. Further experiments were designed in order to elucidate whether the location of retinochrome is restricted to within the myeloid bodies present in the nucleated cell body layer. The retina was peeled off from the dark-adapted frozen eye, and stirred in neutral phosphate buffer to obtain tissue fragments. A small amount of the suspension was pipetted onto a slide glass so as to form a thin layer and air-dried overnight at room temperature. This specimen was immersed in a 0.2% solution of NaBH, for 1 sec at 4°C, washed with water, and photographed under NUV light. Figures 4A and B show the tissue fragment and the corresponding fluorogram, respectively. As can be readily under- stood by comparison between pictures A and B, the yellow-green fluorescence appeared only on the ball-shaped myeloid bodies, and not on the nuclei. Since the specimen in this experiment had not been subjected to any denaturation prior to reduction with NaBHgs, this fluorescence was undoubtedly due to reduced retinochrome. Even if rhodopsin were present, it could not give off any fluorescence at this stage, as rhodopsin changes to a fluorescent product only when denatured before reduction. In fact, when the same specimen was reduced again after treatment with the denatur- Nautilus Retinochrome 215 EIG.3" section. (rhodopsin and retinochrome) reduced by NaBHsg. cell body layer. Scale bar=100 um. ants, an additional yellow-green fluorescence of reduced rhodopsin appeared on the opposite side of the black pigment layer, i.e., in the rhabdomal area. It was thus revealed that in the nautilus retina, retinochrome is concentrated in the spher- ical myeloid bodies of the visual cells. DISCUSSION In the photoreceptors of various molluscs, the two photopigments, rhodopsin and retinochrome, cooperate with each other in their mutual regen- eration for maintaining visual photoreception [9]. Retinochrome is located in the somata away from Detection of photopigments in the dark-adapted Nautilus retina. ee A pe? BGS | “a . as * me. A) Light micrograph of a 10-4m frozen B) Fluorescence micrograph of the same specimen showing the location of the photopigments R, rhabdome layer; B, black pigment layer; N, nucleated the rhodopsin-carrying rhabdomes, especially associated with the myeloid bodies in squid [9] and octopus [4], and with the photic vesicles [18] in slug [16] and conch [17]. The present study has demonstrated that the nautilus visual cell also possesses both the rhodopsin system in the rhabdo- meres and the retinochrome system in the myeloid body. However, in this phylogenetically old animal, the myeloid body differs greatly not only in structure but also in abundance from those known in squid and octopus. In the nautilus, the visual cell usually contains only one spherical myeloid body near the nucleus, whereas in other cephalopods, many myeloid bodies, each of which 216 T. Hara, R. Hara et al. Fic. 4. Identification of retinochrome in the myeloid bodies. A) Light micrograph of a tissue fragment obtained by stirring an isolated retina. B) Fluorescence micrograph of the same specimen showing the localization of retinochrome in ball-shaped myeloid bodies. Rh, rhabdomes; B, black pigment; Mb, myeloid bodies. Scale bar =50 um. takes the form of lamellated bundles of mem- branes, are widely distributed in the cell body with intracellular migration [9, 19]. Consequently, nautilus retinochrome is restricted to very narrow regions in the retina, carried by the ball-shaped myeloid bodies. From the extent of the fluorescent areas shown in Figure 3B, it was also inferred that the retina contains far less retinochrome than rhodopsin. We further examined the fine structures of the retina, as previously done by Muntz and his colleagues using the same species, and have now confirmed most of their observations [12, 13]. The visual cell of the squid and octopus is usually constricted in the middle to form the inner and outer segments, the boundary of them forming the basement membrane parallel to the retinal surface. The supporting cell bodies are situated only on the distal side of this basement membrane. In the nautilus, however, the visual cell does not form such a distinct constriction, and is arranged side by side together with the supporting cell, similar to gastropods. The distal process of the visual cell is surrounded by brushes of microvilli forming the rhabdomeres different in type from those in squid and octopus. The supporting cell extends long microvilli forward beyond the front level of the rhabdomes to construct the surface layer of the retina (Fig. 1). This feature was confirmed again by the results of experiments shown in Figure 3. As can be understood when pictures A and B (Fig. 3) are placed exactly upon one another, the surface of the retina in A does not coincide with the contour of the fluorescent pattern showing the rhodopsin-carrying rhabdomes in B, leaving a nonfluorescent layer just inside the retinal surface. Nautilus Retinochrome This narrow region is occupied by the tips of the long microvilli of supporting cells, and entirely overlies the rhabdome layer, probably protecting the visual cells in the pin-hole lensless eye. Based on the present study, extraction experi- ments on rhodopsin from the microvilli and ret- inochrome from the myeloid bodies were per- formed to determine their chemical properties [20]. Like the squid, the nautilus also contained much rhodopsin and little matarhodopsin in the dark-adapted retina. According to chromatog- raphic analysis of the retinal homogenate, the molar ratio of 11-cis to all-trans retinal was about 96:4, indicating a scarcity of retinochrome which would be expected upon inspection of Figure 3. The absorption maxima of rhodopsin and ret- inochrome were at 465nm (467 nm, by Muntz [21]) and 510 nm, respectively. Further details will be described separately. ACKNOWLEDGMENTS The present work was supported in part by a Grant-in- Aid for Scientific Research to T. H. (61480022) from the Japanese Ministry of Education, Science and Culture. REFERENCES 1 Hara, T. and Hara, R. (1972) In “Handbook of Sensory Physiology. Vol. VII, Part 1, Photochemis- try of Vision”. Ed. by H. J. A. Dartnall, Springer- Verlag, Berlin, pp. 720-746. 2 Hara, T. and Hara, R. (1973) In “Biochemistry and Physiology of Visual Pigments”. Ed. by H. Langer, Springer-Verlag, Berlin, pp. 181-191. 3 Hara, T. and Hara, R. (1982) In “Method in Enzymology”. Vol.81, Biomembranes, Part H, Visual Pigments and Purple Membranes. Ed. by L. Packer, Academic Press, New York, pp. 190-197. 4 Hara, T. and Hara, R. (1987) In “Retinal Proteins”. 7A) DAG, Ed. by Yu. A. Ovchinnikov, VNU Science Press, Utrecht, pp. 456-466. Ozaki, K., Terakita, A., Hara, R. and Hara, T. (1987) Vision Res., 27: 1057-1070. Hara. Ro, Hara, \.,. Ozaki, K., -Merakita,’ A., Eguchi, G., Kodama, R. and Takeuchi, T. (1987) In “Retinal Proteins”. Ed. by Yu. A. Ovchinnikov, VNU Science Press, Utrecht, pp. 447-456. Hara, T. (1988) In “Molecular Physiology of Retinal Proteins”. Ed. by T. Hara, Yamada Science Found- ation Press, Osaka, pp. 305-310. Terakita, A., Hara, R. and Hara, T. (1989) Vision Res., 29: 639-652. Hara, T. and Hara, R. (1991) In “Progree in Retinal Research”. Ed. by N. N. Osborne and G. J. Chader, Pergamon Press, Oxford-New York, pp. 179-206. Hara, T. and Hara, R. (1965) Nature (London), 206: 1331-1334. 11 Barber, V. C. and Wright, D. E. (1969) Z. Zellforsch., 102: 293-312. Muntz, W. R. A. and Raj, U. (1984) J. exp. Biol., 109: 253-263. Muntz, W. R. A. and Wentworth, S. L. (1987) Biol. Bull., 173: 387-397. Hara, T., Hara, R., Kishigami, A., Koshida, Y., Horiuchi, S., Yoshida, M., Yamamoto, M., Goto, T. and Raj, U. (1986) Proc. Jap. Soc. Gen. Comp. Physiol., 3: 168. Goto, T., Takasu, N. and Yoshida, M. (1984) Cell Tissue Res., 235: 471-478. Ozaki, K., Hara, R. and Hara, T. (1983) Cell Tissue IRES92993005—545. Ozaki, K., Terakita, A., Hara, R. and Hara, T. (1986) Vision Res., 26: 691-705. Eakin, R. M. (1990) The Veliger, 33: 209-214. Hara, T. and Hara, R. (1976) J. gen. Physiol., 67: 791-805. Kishigami, A., Hara, R. and Hara, T. (1988) In “Molecular Physiology of Retinal Proteins”. Ed. by T. Hara, Yamada Science Foundation Press, Osaka, pp. 373-374. Muntz, W. R. A. (1987) In “Nautilus”. Ed. by W. B. Saunders and N. H. Landman, Plenum Press, New York and London, pp. 231-244. ZOOLOGICAL SCIENCE 9: 219-221 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Immunoblot Detection of Vertebrate-type of Connectin (Titin) in Ascidian Bodywall Muscle and Tadpole YUNI NAKAUCHI and KoScAK MARUYAMA Department of Biology, Faculty of Science, Chiba University, Chiba 260, Japan ABSTRACT—Immunoblot tests detected the presence of approximately 2000 and 2500 kDa connectin in asci- dian bodywall muscle and its tadpole larva, respectively, using a monoclonal antibody against chicken breast muscle connectin (3B9). The size of ascidian connectin was slightly smaller than that of chicken breast muscle one (~3000kDa), but much larger than the values reported for nematode (668 kDa) and arthropod (1200 kDa) muscle. INTRODUCTION Connectin (titin), an elastic protein of verte- brate striated muscle is the largest protein yet described: MM, 3000 kDa [1] (for a review, see [2]). Twitchin of nematode C. elegans bodywall muscle is a 668.5 kDa protein [3] and projectin of insect and crayfish muscle has a molecular mass of 1200 kDa [4]. Both twitchin and projectin share repetitive 100 amino acid sequences (motifs I and II) [3, 5] with rabbit skeletal muscle connectin [6]. Furthermore, monoclonal antibodies against chicken skeletal muscle crossreacted with twitchin [7] and projectin [4]. Thus, twitchin and projectin are regarded to be invertebrate connectin. In view of the diversity in molecular size of invertebrate connectin it is of interest to examine the size of connectin in the urochordates, one of the closest relatives among the invertebrates to the vertebrates. The present study showed that there exists connectin in ascidian bodywall muscle with MM similar to those of vertebrate proteins. Accepted August 13, 1991 Received July 20, 1991 MATERIALS AND METHODS Materials Bodywall muscle was dissected from an ascidian, Halocynthia roretzi obtained from a local market in Chiba. First, test was cut off, and bodywall muscle was taken out of mantle. Artificially fertilized eggs were allowed to develop for 48 h at 13 °C in Asamusi Marine Biological Laboratory, Tohoku University. Swimming tadpoles were collected by pipetting and sedimented by a hand centrifuge. Gel electrophoresis Approximately 0.1 g of bodywall muscle and 0.1 ml of sedimented tadpoles were separately homogenized gently in 0.5 ml of an SDS solution (10% SDS, 50 mM dithiothreitol, 5 mM EDTA, 1 mM diisopropyl fluorophosphate, 1 mM leupeptin and 0.1 M Tris-HCl buffer, pH 8.0). The suspen- sion was boiled for 1 min at 90°C, clarified by centrifugation, and aliquots of the supernatant were subjected to SDS gel electrophoresis accord- ing to Laemmli (1970) [8] using 2—6% polyacryl- amide gels. Immunoblots Electrophoretically transferred nitrocellulose sheets were treated with monoclonal antibodies against chicken connectin 3B9 and SMI [9] fol- lowed by the treatment with horseradish per- oxidase-linked anti-mouse IgG (DAKOPATTS, Copenhagen). 220 Y. NAKAUCHI AND K. MARUYAMA RESULTS SDS gel electrophoresis patterns of ascidian bodywall muscle and whole tadpoles are shown in Fig. 1B and Fig. 1C respectively in comparison with that of chicken breast muscle (Fig. 1A). There was a faint but distinct high molecular weight band in both bodywall muscle and tadpole, as shown by the arrowheads marked as C. The mobility of bodywall muscle band was comparable with that of chicken @-connectin (~2000 kDa) and the tadpole band moved slightly faster than that of A B C 123 123 123 Fic. 1. Immunoblot detection of connectin in ascidian bodywall muscle and tadpole, using monoclonal antibodies against chicken breast muscle /- connectin. A, chicken breast muscle; B, ascidian bodywall muscle; C, ascidian tadpole. lane 1, Commassie Brilliant Blue stain; lane 2, treated with 3B9; lane 3, treated with SM1. C, connectin; N, nebulin; M, myosin heavy chain. chicken a-connectin (~3000kDa) [1]. These bands of ascidian bodywall muscle and tadpole crossreacted with a monoclonal antibody against 8-connectin, 3B9 (Fig. 1B and C). However, the two bands did not react with the other monoclonal antibody SMI (Fig. 1B and Fig. 1C). Both 3B9 and SMI bound to the a- and f-connectin bands of chicken breast muscle, although SMI hardly reacted with @-connectin due to its small amount (Fig. 1A) (cf. [9]). There are several faint bands that moved faster than the nebulin band (800 kDa) of chicken muscle in bodywall muscle (Fig. 1B), but none of them reacted with anti-nebulin polyclonal antibodies (data not shown). These bands were not recog- nized in tadpole extracts (Fig. 1C). It is to be pointed that SMI strongly reacted with a band below myosin heavy chain (approximately 150 kDa) in the tadpole extract (Fig. 1C), but not at all in the bodywall muscle extract (Fig. 1B). Matsumura et al. [10] reported that SMI reacted with H subunit (~200kDa) of neurofilament protein of several vertebrate nerve tissues. There- fore, it is very probable that tadpole neurofilament subunit reacted with SMI. DISCUSSION Invertebrate connectin greatly varies in molecu- lar size: 600 kDa (scallop adductor muscle; Hu, D. H., 1990; unpublished), 668 kDa (bodywall muscle of C. elegans; {3]), and 1200 kDa (insect flight and leg muscles, and crayfish claw muscle; [4] cf. [5, 11]). These values are much smaller than those of vertebrate striated muscle connectin. Thus, the invertebrate connectin has been called “mini-titin” [11]. The present work strongly suggests that there exist connectins in the ascidian bodywall muscle and the tadpole comparable in size with vertebrate skeletal muscle connectin. On the basis of the mobility on SDS gel electrophoresis, the molecular mass of bodywall protein was estimated to be approximately 2000 kDa (= #-connectin) and that of tadpole connectin about 2500kDa slightly smaller than a-connectin (~3000kDa). An attempt to isolate connectin from ascidian body- wall muscle was unsuccessful, because the connec- Connectin in Ascidian Muscle 2A tin. was only partially solubilized with 0.2M sodium phosphate buffer (pH 7.0). Immunofluorescence observations showed that the monoclonal antibody 3B9 weakly stained muscle tissue in an ascidian tadpole tail, whereas SMI strongly bound to its notochord tissue (Nakauchi, 1990, unpublished; cf. [12]). There- fore, it is reasonable to regard that 3B9 detected muscle connectin, while SMI reacted with neurofilament protein in the present study. It is not surprising that there is a vertebrate skeletal muscle type of connectin in ascidian bodywall muscle, because troponin which is spe- cific to vertebrate striated muscle is present also in the ascidian muscle [13]. In this connection it should be pointed out that ~3000 kDa band was demonstrated in amphioxus striated muscle [14]. The slight difference in size between ascidian bodywall muscle and tadpole connectin is also not Surprising: a similar change in size of muscle connectin occurs during embryonic and neonatal development of the chicken [15]. ACKNOWLEDGMENTS We are greatly indebted to Dr. T. Numakunai of Tohoku University for handling ascidian tadpoles. We are also grateful to Dr. N. Satoh of Kyoto University for his kind suggestion on the immunofluorescence location of ascidian proteins. REFERENCES Maruyama, K., Kimura, S., Yoshidomi, H., Sawa- da, H. and Kikuchi, M. (1984) J. Biochem., 95: 1423-1433. Maruyama, K. (1986) Int. Rev. Cytol., 104: 81-114. Benian, G. M., Kiff, J. E., Neckelmann, N., Moerman, D. G. and Waterston, R. H. (1989) Nature, 342: 45-50. Hu, D. H., Matsuno, A., Terakado, K., Matsuura, T., Kimura, S. and Maruyama, K. (1990) J. Muscle Res. Cell Motil., 11: 497-511. Kakey, Aq, Ferguson. (©; Cabeit; S- Reedy; Mi; Larkins, A., Butcher, G., Leonard, K. and Bullard, B. (1990) EMBO J., 9: 3459-3467. Labeit, S., Barlow, D. P., Gautel, M., Gibson, T., Holt, J., Hsieh, C.-L., Francke, U., Leonard, K., Wardale, J., Whiting, A. and Trinick, J. (1990) Nature, 345: 273-276. Matsuno, A., Takano-Ohmuro, H., Itoh, Y., Mat- suura, T., Shibata, M., Nakae, H., Kaminuma, T. and Maruyama, K. (1989) Tissue & Cell, 21: 517- 524. Laemmli, U. K. (1970) Nature, 227: 680-685. itohhee yey SUZUKIng ie Kamunraysss Ohashin skee Higuchi, H., Sawada, H., Shimizu, T., Shibata, M. and Maruyama, K. (1988) J. Biochem., 104: 504— 508. Matsumura, K., Shimizu, T., Mannen, T. and Maruyama, K. (1989) J. Biochem., 105: 226-230. Nave, R. and Weber, K. (1990) J. Cell Sci., 95: 535-544. Mita-Miyazawa, I., Nishikata, T. and Satoh, N. (1987) Development, 99: 155-162. Endo, T. and Obinata, T. (1981) J. Biochem., 89: 1599-1608. Hu, D. H., Kimura, S. and Maruyaman, K. (1986) J. Biochem., 99: 1485-1492. Yoshidomi, H., Ohashi, K. and Maruyama, K. (1985) Biomed. Res., 6: 207-212. = eh. or ee ee 4 A ual de Heit iH Re Thomas. 3 OK 3 a este acai masa alive me raoereea ot i? bem ee Adths HELE BP UOT HRT) a nag A ina a lade ee eee ae | anil [i ee ff uae cn bac - ee A) a Ps i “i bs . a 4 af ‘ fel oe 7 ye ee sesbearys A Py ; tone Sea 3 - . ee , iis SP Tot a oA. .releetinl ie rl rj } =. t ‘ ‘4 : . a - < J ” = HOF ight 4 2 & "4 F de S a) = ‘ ¥ he WEEE P| ec T S24 7 ares 5 ESET a GS i . Lp * i my > ec oF aia ; / ‘sy ‘ e ee er epee by ot ' 2 Pixs a jy > se -_ 2. ~_ a r — a * este 3; “s : ee Maki A tye Riise Sey or, aes . " : j = ae x ee . = f Doe rh ‘. r ") a c t « - Q y i y ti tr 5 4 F x _ HL NM > . S Me . - * a - : oy i e i = P F ‘ i S ’ = * - ~ 4 = ~, ¥ el VN “ i * os . - Ly v ’ ¥ i be : -s ore. : our, é x _ 1 ‘a 7 ce i . e t. C o E 1 r } ‘ rf, Pea ~ | ~_ a ? \ - \ 5 ZOOLOGICAL SCIENCE 9: 223-226 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Effects of Unilateral Hypothalamic Lesion on Serum Gonadotropin in Hemiovariectomized rats Masaru Fukupa, RyuHE! Hasuimoto!, KoreHITo YAMANOUCHI, YASUMASA Arar’, FuKUKo Kimura! and MIcHIO TAKADA Departments of Obstetrics and Gynecology and Anatomy’, Juntendo University School of Medicine, Hongo, Bunkyo-ku, Tokyo 113, Department of Physiology’, Yokohama City University School of Medicine, Fukuura, Kanazawa-ku, Yokohama 236 and Department of Basic Human Sciences”, School of Human Sciences, Waseda University, Mikajima, Tokorozawa, Saitama 359, Japan ABSTRACT—The effect of unilateral hypothalamic lesion on serum gonadotropin during the development of ovarian compensatory hypertrophy (OCH) in hemiovar- iectomized rats was investigated. A sharp rise in serum follicle stimulating hormone (FSH) concentration was observed at 6 hr after hemiovariectomy in both controls and the animals with a left-side anterior hypothalamic lesion in which the development of OCH were recog- nized. However, this sharp rise was not found in the animals with a right-side anterior hypothalamic lesion in which OCH was suppressed. This result suggests that the right-side anterior hypothalamus plays a critical role in regulating the release of FSH at the initial stage of the development of OCH. INTRODUCTION Evidence has been accumulated suggesting the functional or constructive asymmetry in animal brains. Recently, we have reported that the unilateral lesion placed in the right-side medial anterior hypothalamus effectively suppressed ova- rian compensatory hypertrophy (OCH), whereas the lesion made in the left side failed to suppress it regardless of the side of hemiovariectomy [1, 2]. According to Gerendai et al. [3], the content of gonadotoropin-releasing hormone on the right side Accepted October 5, 1991 Received July 16, 1991 Requests for reprints should be addressed to Dr. Y. Arai. of the medial basal hypothalamus is significantly higher than on the left side in adult female rats. Furthermore, the amount of LH-RH in the right medial hypothalamus decreased significantly when compared to that in the left side by removal of both ovaries [4]. These results suggest the pre- sence of a hypothalamic laterality in regulating gonadotropin secretion. The development of OCH is generally assumed to result from an increase in gonadotropin, primarily follicle stimu- lating hormone (FSH) following a decrease of estrogen level by unilateral ovariectomy [5-7]. Therefore, a question arises as to how these gonadotropins are affected by unilateral destruc- tion of the hypothalamus. In the present study, the effect of unilateral hypothalamic lesion on serum gonadotropins during the development of OCH in hemiovariectomized rats was investigated. MATERIALS AND METHODS Sixty-five female Wistar rats (200-290 g) housed under a controlled photoperiod (14:10h, L:D) and temperature (24+1°C) were used in the experiment. Vaginal smear was taken to detect estrous cycle in each animal. In 24 female rats which had at least 3 consecutive 4-day estrous cycles, the silicon catheter was inserted through the external jugular vein and fixed in the right 224 M. FuxupbA, R. HAsuHImoTo et al. atrium and filled with saline containing heparin under ether anesthesia at a stage of estrus in order to draw blood samples. In the next day (at stage of diestrus 1), 15 females were subjected to brain surgery. A unilateral lesion was made stereotax- ically on either the right (RAHL, 10 rats) or left (LAHL, 5 rats) side of the anterior hypothalamus under ether anesthesia by means of a radiofre- quency lesion generator (Radionics Inc., Burling- ton, MA, U.S.A.). At incisor bar having been set at 5mm below the interaural line, an electrode (0.7 mm) was lowered to 9.5 mm from the bregma level at a point 1.5 mm posterior to the bregma on 1.0mm right (RAHL) or 1.0mm left (LAHL) from the midline. To produce radiofrequency lesions, current was applied and temperature at the electrode tip was kept at 57—-60°C for 1 min. In addition, 9 females with catheter received no brain surgery, as control. At the same time, the unilateral (left side) ovary was removed and weighted in all females. Blood was drawn from the catheter under light ether anesthesia prior to, at 6, 12, 24 and 30 hr after the surgery in each rat. In addition, just before autopsy, blood was taken out by needle from the heart at 2 weeks after surgery. The serum was separated by centrifugation and stored at —20°C until assayes. Two weeks after the surgery, fresh weights of the remaining ovary and uterus were recorded for each rat. All ovarian weights were expressed as mg/100 g b. wt. Each brain was removed and fixed in 10% formalin solution. Then, frozen sections stained with cresylechtviolet were made in order to determine the precise location of the lesion. Concentrations of follicle stimulating hormone (FSH) and luteininzing hormone (LH) in serum were measured by double antibody radioimmu- noassays using RIA-kit supplied by NIAMDD. Serum FSH and LH were expressed in terms of NIH-FSH-SI and NIH-LH-SI, respectively. The statistical analysis of the results was carried out by Studen’s t-test. RESULTS The mean weight of the remaining ovary was 20.2+0.6 mg/100 g b.wt and 19.8+1.4 mg/100 g b.wt in control and LAHL group, respectively. In the animals receiving RAHL, however, the mean weight of the remaining ovary (16.6+0.7 mg/100 g b.wt) was significantly low when compared to that of control (p<0.05). Thus, RAHL kproduced a suppression of the development of OCH. Serum FSH levels are shown in Figure 1. Serum FSH levels before the operation in control, LAHL and RAHL groups were 553.2+111, 568.8+191 and 501.2+143 ng/ml serum, respectively. Signi- ficant rise in serum FSH levels was observed at 6 hr after surgery in both control and LAHL groups in which the development of OCH was recognized, amount being 926.2+102.7 and 830.8+143 ng, respectively. These levels returned to pre- operative levels at 24 hr after operation. In RAHL group in which OCH was suppressed, however, the sharp rise in FSH concentration observed in control and LAHL groups was not seen. The mean content of FSH was 500.9+65.5 ng, being lower than other groups (p<.0.05). Results of LH are shown in Figure 2. Serum LH level after surgery remained unchanged in time course in each group regardless of suppression of OCH or not. In the histological examinations, RAHL and LAHL were located in the medial anterior hypothalamus including the anterior hypothalamic nucleus. In the most cases, the periventricular © Control (9) @ RAHL (10) ALAHL (5) FSH 500 ES 0 6 12 24 30hr Time after operation 2 wks Fic. 1. Changes in serum FSH levels (mean+S.E.) at various times following right- or left-side anterior hypothalamic lesion (RAHL or LAHL) and hemiovariectomy. The number in parentheses is the number of rats examined. * P<0.005 vs control and P<0.05 vs LAHL. Hypothalamic Laterality and Gonadotropin 225 ng/me serum 2 © Control (9) @ RAHL (10) A LAHL (5) Se = 0 Se Se 0 6 12 24 30hr 2 wks Time after operation LH Fic. 2. Changes in serum LH levels (mean+S.E.) at various times following right- or left-side anterior hypothalamic lesion (RAHL or LAHL) and hemiovariectomy. The number in parentheses is the number of rats examined. gray of the anterior hypothalamus was left intact, but frequently, the suprachiasmatic nucleus and/ or paraventricular nucleus were partially damaged. The lesions were found to invade the posterior part of the medial preoptic area in some RAHL and LAHL animals. No damage was seen in the ventromedial-arcuate region. The animals with the lesion extending to the contralateral side across the third ventricle were not included in this study. DISCUSSION It is generally believed that halving of estrogen titers [5-7] and/or of a non-steroidal ovarian factor, inhibin, [8] causes increased FSH release following unilateral ovariectomy. This hormonal environment is thought to be responsible for the OCH. In the present study, a sharp rise in serum FSH concentration was observed at 6hr after surgery in both control and LAHL groups in which the development of OCH was recognized. These levels returned to pre-operative values at 24 hours after surgery. There is a report indicating a significant rise in FSH levels already at 5 hr after unilateral ovariectomy regardless of whether uni- lateral ovariectomy was performed at estrus or at stage of diestrus 2 [9]. Hypothalamic hemi-island was found to suppress the serum FSH level 5 hr following unilateral ovariectomy in prepubertal female rat [10]. Similar results were also reported on hemicastrated male rats [11]. Therefore, it may be reasonable to consider that compensatory follicular growth is initiated by a rapid increase of serum FSH levels after unilateral ovariectomy. However, since estradiol benzoate-treatment start- ing after the rise of FSH has been reported to inhibit OCH in adult females [12], subsequent hormonal conditions may also be important to maintain OCH. In this experiment, the left-right difference is clearly detected, because, unlike control and LAHL groups, a sharp FSH rise following uni- lateral ovariectomy was not found in RAHL group in which OCH was suppressed. This suggest that the right-side anterior hypothalamus may play a critical role in regulating the release of FSH at the initial stage of the development of OCH. Pre- viously we reported that RAHL effectively inhi- bited the development of OCH regardless of the side of hemiovariectomy, whereas no inhibition of OCH was seen in LAHL females [1, 2]. Present result confirmed our previous report. According to Gerendai, LH-RH contents on the right side of the medial basal hypothalamus (MBH) were high- er than on the left side in female rats [3] and LH-RH release in response to ovariectomy was different between right and left side, the release of LH-RH from right MBH being higher, compared to that from left MBH [4]. These results suggest possibility that right hypothalamus is dominant in regulation of gonadotropin secretion, especially FSH. ACKNOWLEDGMENTS This study was supported by Grants-in-Aid to Y. A. from the Ministry of Education, Science and Culture of Japan. REFERENCES 1 Fukuda, M., Yamanouchi, K., Nakano, Y., Furuya, H. and Arai, Y. (1984) Neurosci. Lett., 51: 365- 370. 2 Fukuda, M., Nakano, Y., Yamanouchi, K., Arai, Y. and Furuya, H. (1987) Zool. Sci., 4: 197-199. 3 Gerendai, I. and Halasz, B. (1976) Neuroendocri- nology. 21: 331-337. 4 Gerendai, I. (1984) In “Cerebral Dominance: The Biological Foundations”. Ed. by N. Geschwind and A. M. Galaburda, Harverd Univ. Press, Cambridge, 226 pp. 167-178. Arai, Y. and Gorski, R. A. (1968) Endocrinology, 82: 871-873. Edgren, R. A., Parlow, A. F., Peterson, D. L. and Jones, R. C. (1965) Endocrinology, 76: 97-102. Flerko, B. and Bardos, V. (1961) Acta Endocrinol., 36: 180-184. Welschen, R., Dullaart, J., De Jong, F. H. (1978) Biol. Reprod., 18: 421-427. 9 10 11 12 M. Fuxkupba, R. HAsuHimoto et al. Welschen, R. and Dullaart, J. (1974) J. Endocri- nol., 63: 421-422. Nance, D. W. and Moger, W. H. (1982) Brain Res. Bull., 8: 299-302. Mizunuma, H., De Palatis, L. R. and McCann, M. (1983) Neuroendocrinol., 37: 291-296. Ramirez, V. D. and Sawyer, C. H. (1974) Endocri- nology, 94: 475-482. ZOOLOGICAL SCIENCE 9: 227-230 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Tolerance of the Mudskipper, Boleophthalmus boddaerti, to a Lack of Oxygen SHIT F. CHEW and YUEN K. Ip* Department of Zoology, National University of Singapore, 10 Kent Ridge Crescent, Singapore 0511, Republic of Singapore ABSTRACT—At 25°C, Boleophthalmus boddaerti be- came inert in the ‘closed chamber’ approximately 30 min after all the O, within the chamber had been exhausted although it exhibited a LTs9 of 1 hr 50 min when acutely exposed to anoxia. Results obtained indicated that environmental hypercapnia and the accompanied lower- ing of the ambient pH were the parameters that the mudskipper could not tolerate in the ‘closed chamber’. It excreted more CO, to the enclosed seawater (SW) than the amount of ambient O, consumed with a distinct increase in the respiratory quotient as the O, content of the SW decreased below 0.085 ~mol/ml. The tolerance of B. boddaerti to a lack of O»5 was discussed in conjuction with those of other mudskippers reported elsewhere. INTRODUCTION Mudskippers are gobioid fishes usually found in mangrove swamps in the estuaries of rivers. One species of mudskippers, B. boddaerti, inhabits the intertidal zone of the mud flats along the Pasir Ris estuary in Singapore. At low tide, they move on the mud flats and enter the water occasionally. When the tide rises, they retreat into the muddy burrows at the lower region of the mud flats and remain submerged until the tide ebbs. Another mudskipper, Periophthalmus chrysospilos, in- habits the littoral zone of the beach near the mud flat. Usually, they can be found resting on land close to the water’s edge at high and low tides. However, every year between May and June, P. Accepted August 5, 1991 Received May 1, 1991 * To whom correspondence should be addressed. chrysospilos becomes scarce; presumably staying in the burrows during this breeding period. Water within the burrows has a low dissolved O> content [1]. Working on the Chinese mudskipper, Periophthalmus cantonensis, Gordon et al. [1] reported that the mudskipper became completely inert at dissolved Oy, levels averaging 0.8 ml/1 when it was allowed to respire in ‘closed jars’ of SW. Similar results were obtained for the local mudskipper P. chrysospilos [2]. Considering the differences in behavior of B. boddaerti and P. chrysospilos in their natural habitats, one would expect the former mudskipper to have a higher tolerance to hypoxia than the latter one. Since the gills of B. boddaerti are not well-adapted for terrestrial respiration [3, 4], it has a lower affinity to land than P. chrysospilos and is naturally exposed to hypoxic water twice everyday during high tides. Hence, this study was undertaken to examine the tolerance of B. boddaerti to a lack of O>. MATERIALS AND METHODS B. boddaerti (6-22 g body weight) were col- lected along the estuarine canal at Pasir Ris, Singapore. No attempt was made to separate the sexes. They were maintained in 50% SW (15%c salinity), in small aquaria (22 cm x 12 cm x 14 cm) at 25°C. The aquaria were tilted slightly, so that the water covered only approximately half of the bottom surface of each tank. Water was changed daily and the fish were fed a manufactured preparation (Goldfish & Staple Flake, Everyday 228 S. F. CHEw AND Y. K. Ip Co., Singapore) daily. B. boddaerti weighing 11.9 g (n=7, ranged from 10.2-14.8 g) were individually put into a respira- tory chamber containing 50% SW at 25°C with continuous aeration 24hr before measurements began. The respiratory chamber was constructed according to Chew et al. [2]. At the start of the experiment, the aeration was stopped and the respiratory chamber sealed. Water within the chamber was stirred at optimal speed to minimize disturbance to the fish, which usually lay calmly at the bottom of the chamber without any noticeable fin movements, to allow for the measurement of the routine Oz consumption rate (1 O2/min per g fish). Dissolved O, content was continuously monitored with a YSI Model 53 Biological Oxygen Monitor (YSi, Ohio, U.S.A.), connected to a potentiometric recorder (Rikadenki Kogyo Co., Tokyo, Japan), with clark-type O > electrodes and standard plungers until the fish ceased all move- ment and was unable to balance itself. The concentration of O, in 50% SW equilibrated with air at 25°C was taken to be 0.26 umol/ml [5]. Total CO, contents in 10 ul of 50% SW obtained from the respiratory chamber at various intervals were immediately determined enzymatically using the Sigma Diagnostic Kit 130-A following Procedure No. 130-UV. Ethanol was determined using a Boehringer Ethanol Assay Kit No. 176290. The pH was measured with an Orion 501 digital ionalyzer and a combination pH electrode. Two groups of 10 B. boddaerti each were introduced individually into Erlenmeyer flasks containing 250ml of 50% SW pre-equilibrated with either pure N> or 5% CO, in N> for 30 min. After the introduction of the fish, the flow rate of the specific gas was adjusted to 50 1/hr. The time taken for the individual fish to become inert and unable to balance itself was recorded. The LTs values were extrapolated from plotting the results on graph paper. Results were presented as means+standard error (S.E.). Differences in means were analyzed by analyses of variance (ANOVA) followed by multiple comparisons of means by the Student- Newman-Kuele procedure. Differences with p< 0.05 were regarded as statistically significant. RESULTS At the start of the experiment (n=7), the air saturated 50% SW in the sealed respiratory cham- ber had a pH of 8.5+0.05, a dissolved O, level of 0.26 wzmol/ml and a total CO, content of 1.60+ 0.08 ~mol/ml. Changes in O, concentration were non-linear with respect to time as a result of the consumption of O; by the contained B. boddaerti. A typical trace is presented in Fig. 1. The fish became completely inert 2.94+0.17 hr after the closing of the chamber or 21.3+2.4 min after the oxygen monitor registered zero dissolved O> con- centration. At this point, the total CO, in the ambient SW increased significantly to 2.4+0.11 pmol/ml and the pH was 7.1+0.02. N fo} 1S) 100 e a t N 5 g 8 £ , had Eamets ? = $ ° a i [o) rc) s SO nS 3 oO £ = o E = C) S) x = S = y E S . = © = : 8 cs o - x Y © s 0 Y y | eee Hamann nee 0) 60 120 180 min Fic. 1. A typical record of the pattern of O, consump- tion by B. boddaerti in the ‘closed respiratory cham- ber’ containing fully aerated 50% sea water at 25°C. Total amount of O, present in the box at 100% saturation was 77 umol. At specific levels of air saturation (67%, 33%, 10%, 0%) in the enclosed SW, the amount of O, consumed were 25 mol, 50 zmol, 70 wmol and 77 pmol, respectively. The corresponding amount of CO, produced were 30.2+2.4 wmol, 65.7+4.6 pmol, 120.145.2 “mol and 185.6+7.8 umol which were significantly different from each other. The respective ratios of CO, produced to O» consumed were calculated to be 1.10+0.08, 1.25+ 0.09, 1.69+0.07 and 2.40+0.09. No ethanol was detected in the SW at the end of the experiment Hypoxia and B. boddaerti 229 (detection limit=0.19 wmol/ml). All the seven fish recovered 10 min after being transferred back to the normoxic control condition. The LTs59 for B. boddaerti exposed to SW continuously bubbled with N2 was observed to be 1 hr 52 min (Fig. 2). In SW continuously bubbled with 5% CO, in No, the dissolved CO, concentra- tion was determined to be 2.91+0.07 »mol/ml (n=5). The pH of the SW was lowered from 8.5 + 0.07 to 6.1+0.06. Under such a condition, the observed LTs) for B. boddaerti was 23 min (Fig. Dy @ Oo number of inert fish i60 Fic. 2. Survival of B. boddaerti on exposure to 50% sea water saturated with either pure N> at 25°C (0) or 5% CO, in Nz (@) at 25°C. DISCUSSION In the present studies, B. boddaerti became inert in the ‘closed chamber’ approximately 30 min after all the O, within the cahmber had been exhausted. Such a tolerance to a lack of O, exhibited by this fish in the ‘closed chamber’ ts definitely higher than those reported for P. chrysospilos and P. can- tonensis which become inert at 0.75 yl/ml [2] and 0.8 «l/ml [1] of dissolved O>, respectively, under similar conditions. The differences in the toler- ance of these mudskipper to hypoxia may be related to their differences in behavior in their natural habitat as described in the Introduction. Working on two species of mudskipper, P. can- tonensis and Boleophthalmus chinensis, Tamura et al. [6] reported that their oxygen consumption rates at 20°C were 236 and 110 ml/kg per hr, respectively, when they were able freely to select either an aquatic or terrestrial habitat. However, when confined in water, their oxygen consumption rates decreased to 196 and 72 ml/kg per hr, respectively. It was observed that P. cantonensis was more active than B. chinensis under both conditions [6]. Hence it would appear that Boleophthalmus is capable of reducing its metabo- lic rate to a greater extent than Periophthalmus, leading to a greater tolerance of the former fish to a lack of oxygen. Gordon [7] reported that Periophthalmus vul- garis, the Australian mudskipper, exposed acutely to anoxic SW collapsed within 5-10 min. How- ever, B. boddaerti was able to survive for more than one hour in anoxia with a LTs9 of 1 hr 50 min at 25°C. Such a LTs 9 value is comparable to that of Oreochromis mossambicus [8] and higher than that of Cyprinus carpio [9], but lower than those of Carassius carassius [10] and Carassius auratus [8). Hence, the general statement of Gordon et al. [1] that ‘tolerance of mudskipper for hypoxia was limited’ may not be correct. It is important to point out here that although the water deep inside the burrows in the mudskippers’ natural habitat is possibly anoxic, the fish may not orientate itself in these lethal regions of the burrows for too long a period of time compared to the less hypoxic ones. Naturally, if a fish has a choice, its first response to anoxic water is to leave [11]. Since B. boddaerti was able to survive in anoxia for more than an hour, it might have succumbed in the ‘closed chamber’ to parameters other than the low QO, tension. In the closed chamber, the decrease in the ambient dissolved O> content was accompanied by an increase in the ambient total CO, concentration. Therefore, the fish was ex- posed not only to environmental hypoxia but also to environmental hypercapnia simultaneously. The situation was further complicated by the fact that B. boddaerti excreted more CO, to the enclosed SW than the amount of ambient O, consumed. The ratio of CO, excreted to Op consumed distinctly increased as the O> content of the SW decreased below 0.085 umol/ml. The source of this extra amount of CO, was not elucidated in the present studies. Since ethanol was not detected in the SW or the blood and muscle of this fish (Chew and Ip, unpublished data), the pathway responsible for metabolic CO» production in goldfish and crucian carp [12] may not exist in this mudskipper. B. boddaerti became wa 230 S. F. CHEW AND inert after being exposed to SW continuously bubbled with 5% CQO, in N> for 15-30 min. Therefore, it would appear that environmental hypercapnia and the accompanied lowering of the ambient pH were the parameters that the muds- kipper could not tolerate in the ‘closed chamber’. REFERENCES 1 Gordon, M. S., Ng, W. W. S. and Yip, A. Y. W. (1978) JSExp, Biola 72: S715: 2 Chew, /S.k leim Ac, bow. Wb wdeee CG: L., Chan, K. M. and Ip, Y. K. (1990) Fish. Physiol. Biochem., 8: 221-227. 3 Low, W. P., Lane, D. J. W. and Ip, Y. K. (1988) Biol. Bull., 175: 434-438. 4 Low, W. P., Ip, Y. K. and Lane, D. J. W. (1990) 11 12 WK. Wp Zool, Scis7: 29=38. Umbreit, W. W., Burris, R. H. and Stauffer, J. F. (1964) In “Manometric Techniques”. Ed. by W. W. Umbreit, R. H. Burris, and J. F. Stauffer, 4th ed, Burgess Pub. Co., Minneapolis. pp. 5. Tamura, S. O., Morii, H. and Yuzuriha, M. (1976) J. Exp. Biol., 65: 97-107. Gordon, M. S. (1978) Am. Zool., 18: 606. Van den Thillart, G. and Van Waarde, A. (1985) Mol. Physiol., 8: 393-409. Smith, M. and Heath, A. G. (1980) Comp. Biochem. Physiol., 66B: 267-272. Blazka, P. (1958) Physiol. Zool., 31: 117-128. Hochachka, P. W. (1980) Living Without Oxygen Harvard University Press, Cambridge, MA. 181 pp. Shoubridge, E. A. and Hochachka, P. W. (1981) Mol. Physiol., 1: 315-338. ZOOLOGICAL SCIENCE 9: 231-235 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Eucheilota intermedia Kubota is a Distinct Taxon and the Third Form of Eutima japonica Uchida (Hydrozoa; Leptomedusae) SHIN KUBOTA Zoological Institute, Faculty of Science, Hokkaido University, Sapporo 060, Japan ABSTRACT— Eucheilota intermedia Kubota, 1984 is a distinct taxon since the mature medusa from the two known localities in Japan always possesses oval gonads, and has a potential to develop into a mature medusa that is indistinguishable from the medusa of the southern form of Eutima japonica Uchida, 1925. The wide morphological variation of the medusa of Eucheilota intermedia forms a series linking the two. Based on this knowledge, that of a high crossability between Eucheilo- ta intermedia and Eutima japonica, and the known distribution of the two forms, it is concluded that they are conspecific. Eucheilota intermedia is henceforth treated as Eutima japonica f. intermedia. INTRODUCTION Eucheilota intermedia Kubota, 1984 is a bivalve- inhabiting hydrozoan found in only two localities, both in Japan [1, 2, Kubota, unpubl.]. It seems to have unique characters in the newly-matured medusa. However, its validity is questionable when its ontogeny is compared with that of Eutima japonica [2, 3]. To settle the question, the development of the medusa was re-examined and crossing experiments between the two forms were conducted. As a result, a small number of medusae of Eucheilota intermedia appeared as more complicated than either the typical or the extraordinary ones that had been described before [1-3]. These medusae are described here, together with typical forms from the second locality. Accepted December 1, 1990 Received October 26, 1990 Ontogeny and other biological evidence so far obtained [1-3, Kubota, unpubl.] are evaluated to resolve the taxonomic question. MATERIALS AND METHODS A total of 166 mature medusae liberated from 47 host bivalve specimens of the two nominal species were obtained in the laboratory by culture. All the hosts were collected intertidally. The hosts con- sisted of four specimens of Mytilus edulis gallopro- vincialis Lamarck and four of Barbatia virescens (Reeve), and came from Takeshiki in Asou Bay, Tsushima Island, Nagasaki Prefecture, on August 9-11, 1988, and 26 specimens of M. e. galloprovin- cialis and 13 of B. virescens from Zagashima Island in Ago Bay, Mie Prefecture, collected on April 5, 1987 and February 8, 1988. Both the hosts harboring the hydroids and the released medusae from these hydroids were cultured in filtered seawater from Oshoro Bay, Hokkaido, using 60 and 80 ml polystyrene vessels at 21+1°C. They were daily fed with Artemia nauplii. The measure- ments of medusae were taken from living speci- mens with an empty stomach after narcotization in a MgCl, solution. RESULTS AND DISCUSSION Among the 142 medusae of Eucheilota interme- dia originating from seven specimens of Mytilus edulis galloprovincialis and from eight Barbatia 232. S. KUBOTA virescens, 135 possessed two tentacles, two had three tentacles and five had four. All 166 medusae monitored from release, with usually two tentacles and vestigial gonads, developed into typical dwarf mature medusae with four tentacles and oval gonads (Fig. 1A) [1-3]. The most rapidly matured medusae of both sexes achieved maturity on the third day after liberation. Among these mature medusae, at least nine (Nos. 12-20 in Table 1) conspicuously meta- morphosed after maturation, and the peduncle which is not found in Eucheilota was formed. In all nine specimens, the peduncle was longer than that of the formerly recorded [2] (* in Table 1), but it usually became shorter when the medusa was spent (Nos. 14, 18-20). Another distinct morpho- logical change was observed in the oral lips. In two specimens (Nos. 13, 16) the oral lips were not cruciform but frilled. The manubrium itself was usually short and did not protrude from the Fic. 1. The morphology of laboratory-reared Eucheilota intermedia medusae photographed in living state. A: A 5-day-old dwarf male medusa from Zagashima Island, with oval gonads, four tentacles, many cirri, and many exumbrellar nematocysts. Note the absence of peduncle. B: A 73-day-old spent female medusa from Zagashima Island (No. 18), with a peduncle and eight tentacles. umbrellar aperture and the thick apical jelly. C, D (aboral view): A 34-day-old female medusa from Tsushima Island (No. 13), with a distinct peduncle, eight tentacles, elongated gonads produced on most parts of the radial canals and many cirri. Note that the manubrium is not protruded from the Third Form of Eutima japonica 233 TABLE 1. Measurements of laboratory-reared mature and spent medusae of Eucheilota intermedia from two localities in Japan (see text) Speci- Specimen Age | Dia- Lean Ne gy Lote Neyo NO o No. Ob Wo, 9 eas : j no. of stato- of me owt days a mm 1 sec beyond the period of delivery of the olfactory stimulus. BE refers to increased firing which recovers to background firing levels <1sec after the stimulation. BE is similar to responses observed in antennal lobe neurons of M. sexta [10, 45, 48] and other species (e.g., B. mori-[49]), while LLE is apparently a property of some PC neurons but not antennal lobe neurons. The sex-pheromone blend induced LLE robustly in preparations in which other odors, including individual pheromone components, fail to do so (Fig. 8). (2) LLE responses in PC neurons Where does LLE arise within the olfactory pathway? LLE has not yet been encountered in intrinsic neurons or projection neurons of the antennal lobe in reported samples, exceeding 500 local neurons and 130 projection neurons studied in M. sexta (e.g., Fig. 4A) [9, 10, 45, 48]. Pro- longed increases of firing rate have occasionally been recorded in antennal lobe neurons of other moths (B. mori-[27], Fig. 4; Holiothis zea-[58], Fig. 6), but those responses did not remain near the peak firing level for >1 sec. Of neurons in M. sexta that appear to receive input within the lateral PC, a prominent target area for axons of antennal lobe projection neurons [10, 23, 45], none has been shown to exhibit LLE [18]. We found LLE to be associated with neurons that innervate the LALs (Fig. 7) and MBs, and as described later, LLE in neurons appeared to be synaptically down- stream from these areas (Fig. 9) [19]. Although we cannot, at present, exclude the possibility that LLE was consequence of ascending influences in our experiments, similar LLE-like responses have been recorded in preparations of B. mori in which the ventral nerve cord was severed ([59], Kanzaki- unpublished observations). Thus, it is likely that LLE is a response arising at the level of the MBs and/or the LALs in the protocerebrum. LLE responses similar to those for M. sexta have also been observed in some MB extrinsic neurons and other PC neurons of Apis mellifera [51, 60], Acheta domestica [52], and B. mori [29, 49]. Pre- vious workers have speculated that prolonged ex- citation might be produced by reciprocal synapses among intrinsic neurons of the MB [61] or by feedback loops from the lobes to the calyces of the MB by extrinsic MB neurons [52]. Among other mechanisms that must also be considered are: (1) R. KANZAKI AND T. SHIBUYA 252 Olfactory Brain Neurons of Insects sy) the possibility of plateau properties, similar to those demonstrated to cause prolonged firing in other types of neurons in invertebrates and verte- brates [e.g., 62-68], intrinsic to neurons at the source of LLE, and (2) other long-lasting synaptic or modulatory mechanism that increase excitabil- ity [e.g., 67, 68]. It is possible that LLE originates within the MB and is induced synaptically in subsequent neurons within the LALs or vice versa, or that it is independently generated at both sites. Moreover, the possibility exists that LLE is gener- ated in some other region of the CNS and transmit- ted to both the MB and the LAL. Findings to date do not permit the elimination of any of these possibilities. (3) Integration of bilateral olfactory information at LALs in PC Only stimulation of the antenna ipsilateral to the soma of a bilateral LAL neuron (Fig. 7) elicits excitation [18]. This is consistent with the observa- tion that antennal-lobe projection neurons respo- sive to pheromonal stimuli also respond only to ipsilateral stimulation (Fig. 4) [10, 45] and suggests that pheromonal information is integrated un- ilaterally within the brain and may not cross the midline before at the level of the LALs [18]. This contrasts with processing of other types of in- formation. For example, bilateral mecha- nosensory information enters the pathway at the level of certain antennal-lobe projection neurons (Fig. 6D) [10]. Bilaterally uniglomerular antennal- lobe neurons (Fig. 6D) have been described which could provide the substrate for integration of bi- lateral information [10]. One such cell was found, however, to be sensitive to olfactory stimulation of only one antenna even though it was bilaterally mechanosensory [10]. The structure of the bilateral LAL neurons (Fig. 7) supports the idea of polarized information flow from the side ipsilateral to the soma to the contra- lateral side. Ipsilateral branches typically exhi- bited smooth profiles, while contralateral branches were distinctly different in bearing numerous vari- cosities (Fig. 7) [18]. Similar differences in struc- ture have been observed in other brain neurons of M. sexta [46]. Although the notion requires verification with electron microscopy, it is possible that this polarity of structure is similar to that observed in crayfish neurons, where smooth pro- files were shown to be exclusively post-synaptic (i.e., the input side of the neuron) and the blebby, vericose profiles presynaptic (i.e., the output side of the neuron) [69]. In further support of the idea that pheromonal information from both antennal lobes is collected at the level of the LALs is the observation that descending neurons, which have branches within the LAL and are probably down- stream in the path of information flow, are sensi- tive to stimulation of either antenna and do not Fic. 7. Morphology of a bilateral LAL protocerebral neuron of M. sexta. (A) (a-f) Photomicrographs of 25-~m frontal sections exhibiting Lucifer Yellow-stained neurites of the injected neuron. Reconstructions were made from color slides of such sections. (a, c, e) Processes ipsilateral to the cell body has smooth profiles. (b, d, f) Branches contralateral to the cell body exhibited numerous varicosities. (B) Reconstruction of the stained neuron. The bilateral arbors lay anterior to but did not penetrate the calyces (Ca) of the MBs (stippled). P peduncle; VMP ventro-medial protocerebrum. Scale bars=100 wm. (Kanzaki et al. [18]). Fic. 8. (A) Intracellular recordings from a bilateral LAL protocerebral neuron that responded with LLE. Clean air (a) or odorants (b-f) were delivered to the antenna as 1-sec puff stimuli, indicated by a solid line below each trace. (a) Clean air (i.e., gentle mechanical stimulation) appeared to cause a brief inhibition of firing. (b) A puff of pheromone blend (0.20 FE) elicited LLE that persisted for over 8 sec (the two traces represent a continuous record). (c, d) Individual pheromone components delivered at rather high concentrations produced much weaker responses, while tobacco-leaf odor caused brief inhibition similar to the apparent mechanosensory component revealed by blank stimulation (e). Stimulation of the antenna even with very high concentrations of E2-6: AL failed to elicit LLE similar to that produced by pheromone blend and produced only a BE (f). Scales in (Ab) (for all traces) =1 sec, 20 mV. (B) Plots of the firing frequency, measured in 250-msec bins, of the neuron recorded in (A), including additional trials. In this and all subsequent plots, where only one concentration is specified, different symbols indicate separate trials at that concentration. Where different concentrations were tested, they are specified along with the corresponding symbol on the figure. The sequence of stimulus trials always proceeded from the lowest concentration to the highest. The time of delivery of the odor is indicated by a solid line beneath the traces or plots. (Kanzaki et al. [18]). 254 IMPULSES / SECOND R. KANZAKI AND T. SHIBUYA CLer AIR Sy FEMALE EXTRACT 0.20 FE 114 OF) AlN LIAL 1 Al may AN vA TATA A } | | WV | I WN ail A TOBACCO | | | || 11) \| Vii} PWT WT | \| 1 Hy HOE a | AIMEE NDB EDN BSIOANDO ABM VUI RE SBR ND SBO0NBNON)N NO MINI MAGINB) BOHN V RN e=)91) 00) Bh E2-6:AL 420 yg E11,213-15:AL- 100 ng FEMALE EXTRACT 0. Be FE OA A.e- eas pos fen / cree Bis “ al ARet, e fo} \/ eo, P08 ene E10,Z12-16:AL 10 ng ‘eo. ‘oe. aa e AN Ue vA \egees oR Coe Awol ik ° TOnn o O° O08 TIME (sec) Fic. 8. Olfactory Brain Neurons of Insects ASD) FEMALE EXTRACT 0.20 FE fa Z O O nt a OD Ww ap —_ = ae 2 10 TIME (sec) Fic. 9. Physiology and morphology of a descending neuron that responded to pheromonal stimuli with LLE. (A) Antennal stimulation with female extract elicited LLE lasting >20 sec (traces continuous). Scales=1 sec, 40 mV. (B) Plot of the firing frequency of the neuron shown in (A) measured in 250-msec bins. (C) Reconstruction of the Lucifer Yellow-stained DN (in frontal view) from which the recordings shown in (A) were obtained. Bilateral processes of this neuron penetrated LALs on both sides of the brain and were linked by a large neurite that crossed the midline in the LAL commissure. The appearance of sparse branching of this neuron is probably a consequence of incomplete staining. The axon left the SOG in the connective contralateral to the soma (arrow) and could not be traced to the thoracic ganglia. The axon of this neuron had a diameter of about 8 wm and traveled in the dorsomedial quadrant of the contralateral VNC. Scale=100 ~m. (Kanzaki et al. [19]). 256 R. KANZAKI AND T. SHIBUYA discriminate between the antennae if pheromonal stimuli are applied unilaterally (Figs. 9-11) [19]. (4) Excitatory responses of olfactory PC neurons The majority of responses recorded in PC neurons to olfactory stimulation are excitatory. Although brief inhibition is sometimes elicited CLEAN AIR PUFF | | FEMALE EXTRACT 0.20 FE Sj | | | | | | (MUAH I E10,Z12-16:AL 100 ng ie GIR E wv BERG BES INSLNBIR LE A a ja) z Oo O LU ep) —< dp) LU ep) = =) oO = TIME (sec) FEMALE EXTRACT 0.20 FE E10,212-16:AL o*: 10 ng before LLE [18], sustained inhibition is rare. Re- cent immunocytochemical studies using antisera to GABA have revealed that numerous PC neurons, such as optic-lobe interneurons, negative feedback fibers of the MB, and fan-shaped neurons of the lower division of the central body, are GABA- immunoreactive [47]. Because GABA is believed E11,Z13-15:AL 100 ng qT TEEN UIT VT ET A E10,Z12-16:AL (50 ng) + E11,Z13-15:AL (50 ng) e | | WI TOBACCO E10,Z12-16:AL(50 ng) + E11,Z13-15:AL(50 ng) TIME (sec) Fic. 10. LLE is elicited by pheromone blends but not by individual pheromone components. (A) LLE was elicited by stimulating with the complete wash of the female pheromone gland (b), or with blends of individual pheromone components (e), and it lasted much longer than the >15 sec shown in the two continuous traces in each trial. LLE could be elicited by stimulation of either antenna. Neither tobacco-leaf odor (f) nor individual pheromone components (c, d) activated the cell. Scales (for all traces)=1 sec, 40mV. (B) Plots of the firing frequency, measured in 250-msec bins, of the neuron recorded in (A), including additional trials. (Kanzaki et al. [19]). Olfactory Brain Neurons of Insects 257 to be the neurotransmitter at many inhibitory synapses in the insect CNS [70-72], these are likely to be inhibitory elements. None of the olfactory PC neurons characterized morphologically thus far, however, resemble neurons revealed by GABA immunoreactivity or innervate the corre- sponding neuropils. DESCENDING NEURONS As described in protocerebral neurons, many olfactory PC neurons innervate a particular neuro- pil region, the lateral accessary lobe (LAL) in PC. LALs and certain adjacent neuropil regions are also innervated by branches of neurons that re- spond to olfactory stimuli and have axons that descend in the ventral nerve cord. It appears that the LALs may be interposed in the pathway of olfactory information flow from the AL through the lateral PC to the descending neurons (DNs) (Fig. 14). The activity of these DNs is of particular interest because the information they carry repre- sents the integrated multimodal output of brain circuits that may act on thoracic motor circuitry to affect behavior. (1) Physiology of DNs Physioloigcal responses of olfactory DNs have several features in common with those of other neurons in the PC, but also have their own unique characteristics. As described for PC neurons, the responses of DNs to olfactory stimuli fall into two general classes, LLE and BE. The LLE-type responses are elicited preferentially in PC neurons by pheromonal stimuli, including individual phero- mone components (Fig. 8), whereas in DNs, only blends of pheromone components or the female extract elicit LLE (Fig. 10) [19]. Although the responses of both PC neurons and DNs fit our definition of LLE, the activity of DNs typically takes longer to reach the peak firing rate than do the responses of PC neurons (3-10 sec for DNs vs. 1-3 sec for PC neurons) (Figs. 8-11). This slower rise to peak firing rates may be due to mixed excitatory and inhibitory drive during the early part of DN responses. DNs also typically continue to fire at high rates longer (>10 sec; Figs. 9-11) than PC neurons (<10sec; Fig. 8). Some DNs show conditional responses in which identical pheromone-blend stimuli have different effects depending upon the state of firing of the DN (Fig. 11). Such conditional responses have not been encountered in olfactory neurons of either the AL or PCin M. sexta [9, 10, 18, 45, 48, 73] and B. mori [25-27, 49]. Conditional responses can be elicited in DNs by stimuli applied separately to either antenna [19]. PC neurons are sensitive only to stimuli applied to one of the antennae (Fig. 8) [18]. Integrated information about pheromone blends may be collected from circuitry on both sides of the brain by bilateral LAL neurons (Fig. 7). The arborizations of these bilateral neurons overlap with branches of DNs both within the LALs and in adjacent PC neuropils and may provide the ana- tomical substrate for synaptic interactions (Fig. 9). Neither direct structural contacts nor synaptic communications between these neurons have been demonstrated. However, we have not observed any neurons that might mediate the direct synaptic contacts between AL projection neurons and DNs, as has been suggested to occur in the lateral PC of cockroaches [31, 74]. (2) Multimodality of DNs None of the olfactory DNs showed consistent responses to mechanosensory stimuli applied to either antenna individually [19]. About 50% of pheromone-unresponsive AL projection neurons [10] and about 7% of PC neurons [18] that re- sponded to olfactory stimulation showed some responses to mechanosensory stimulation. Some DNs which carry mechanosensory information were characterized, but these neurons did not respond to any of the odorants tested (Kanzaki, unpublished data). By contrast, visual stimulation elicited excitatory or inhibitory responses in olfac- tory DNs. There were “on” or “off” responses in neurons showing LLE and reduction of spon- taneous firing with tonic increases in illumination (Fig. 12) [19]. Neurons responding to olfactory stimuli with BE also show excitatory responses to lights on or off (Fig. 13) [19]. It is suggested that there is a stronger linkage between visual and olfactory modalities than between mecha- nosensory and olfactory modalities. The fact that sustained high illumination tended to inhibit back- 258 R. KANZAKI AND T. SHIBUYA FEMALE EXTRACT 0.20 FE a FEMALE EXTRACT 0.20 FE TTT TN TTT 0 a > ARUN! LL FEMALE EXTRACT 0.20 FE PTAA TAA AAO HTH ATIIIL i © NNN NIMNNIII | a A i 2 IMPULSES/SECOND Fic. 11. HN TIME (sec) State-dependent responses of a DN exhibiting LLE. (A) Responses to puffed stimulation with pheromone blend ipsilateral to the recording site in the cervical nerve cord. When the neuron was firing at a relatively low rate near 4 Hz, stimulation of the antenna with pheromone blend elicited LLE (a). When the neuron was in a state of high background firing due to the LLE elicited in (a), a similar stimulus applied to the same antenna caused a marked reduction in the firing rate (b) rather than further excitation. This inhibition of firing lasted only several seconds, and firing gradually returned to the previous higher rate. Repeated application of pheromone blend against the higher background firing rate again caused an inhibition (c). Scales=1 sec, 40 mV. (B) Plot of the firing frequency of the neuron shown in (A). Bars above the plot indicate the segments corresponding to the traces in (A). Bars below the plot indicate times of stimulus application. (Kanzaki et al. [19]). ground firing of DNs that process pheromonal information might be relevant to the observation that M. sexta males are inactive and unresponsive to pheromone blend in the daytime. Of course, many other parameters in addition to ambient light level change in the course of a circadian rhythm, e.g., levels of octopamine in the hemolymph [75]. In Bombyx mori, visual stimuli were effective in changing the firing state of certain interneurons [59]. Moreover, some of these neurons responded to both mechanosensory stimuli applied to the antennae and directional visual stimuli [59]. Phero- monal stimuli were shown to activate or amplify visual responses in several DNs of gypsy moth Olfactory Brain Neurons of Insects JS3y) LIGHT | | : a Tes 9] ‘ f MOVEMENT B | | | a cc Nt Ara sor See Nip Peppy E10,Z12-16:AL (50 ng) + E11,Z13-15:AL (50 ng) E Wn) WU NM I AUUINLIUNNIL LIGHT D | | | 4 Y LIGHT + E10,Z12-16:AL (50 ng) + E11,Z13-1 5:AL (50 ng) E | | SaeRIORialh 4 LIGHT + E10,Z12-16:AL (50 ng) + E11,Z13-15:AL (50 ng) F | 3 Fic. 12. Multimodal influences on a DN exhibiting LLE. (A) When this neuron was in a state of low background firing, turning a bright light on (upward arrow) and off (downward arrow) elicited brief on/off responses and led to suppression of tonic firing. (B) Moving a vertical black and white striped pattern horizontally in front of the moth with only dim illumination also elicited brief spiking responses that were directional. (Movement indicator trace below intracellular recording. Upward deflection=movement R to L, downward deflection=movement L to R). (C) Stimulation of the antenna with a blend of pheromone components (50 ng each of E10,Z12-16: AL and E11,Z13-15: AL) elicited LLE. (D) When the neuron was in a state of high background firing due to such LLE, increasing illumination (upward arrow) suppressed firing. Firing was not restored to the previous high rate even when illumination was reduced (downward arrow), although a brief “off” response was elicited. When the neuron was again firing at a high rate in a similar LLE activated by puffing the blend of pheromone components on the antenna (BE), increasing illumination reduced but did not completely suppress firing of the cell. Another puff of the odor blend (50 ng each of E10,Z12-16: AL and E11,Z13-15: AL, soild bar) against this background of reduced firing caused suppression of firing lasting several seconds, whereupon the neuron gradually increased its firing rate to the pre-stimulus level. This response is consistent with the state-dependent responses elicited against the backdrop of LLE, as described above. (F) Increasing illumination (upward arrow) when the neuron was again in a state of low background firing produced an “on” response consisting of one spike and subsequently silenced the cell. Under this condition of high illumination, an odor stimulus similar to the one that previously excited the cell (e.g., c, d) (solid bar) failed to elicit LLE. Scales=1 sec, 40 mV. (Kanzaki et al. [19]). 260 R. KANZAKI AND T. SHIBUYA CLEAN AIR PUFF i MOVEMENT | | fl ee ee eee eee FEMALE EXTRACT 0.20 FE E10,Z12-16:AL 100 ng [oes E11,Z13-15:AL 100 ng £10,Z12-16:AL (50 ng) + E11,Z13-15:AL (50 ng) TOBACCO US ele ei E2-6:AL 84 ug Fic. 13. Physiology and morphology of a unilateral descending neuron responding with BE. (A) Intracellularly recorded responses to odors and movement. This neuron exhibited an extremely low background level of activity. It gave only occasional signal action potentials when unstimulated or when clean air (a), plant odor (f), E2-6: AL (g), or pheromone components (d) were puffed onto the antenna. This neuron’s failure to respond to these air-puff stimuli suggests an absence of sensitivity to gentle mechanical stimulation. Brief increases in firing at short latency, and slight increases in the occurrence of impulses at longer latencies, seemed to be elicited by female extract (b) and E10,Z12-16: AL (c, e). No responses were elicited by stimuli applied to the contralateral antenna. This neuron gave brief on/off responses consisting of one or two impulses when illumination was increased (h), and it gave a more vigorous response to movement of a pattern of vertical black and white stripes horizontally in front of the moth (i). Scales=1 sec, 40 mV. (B) Reconstruction (in frontal view) of the neuron whose responses are illustrated in (A). This neuron projected several stout branches to a region just ventral to the AMMC. The main neurite projected through the SOG ipsilaterally to the soma and sent off a fine branch there shortly before entering the ipsilateral cervical connnective (arrow). Inset: Cross section of the cervical connective showing the lateral position of the axon. Scale bars=100 um. (Kanzaki et al. [19]). Olfactory Brain Neurons of Insects 261 Lymantria dispar [76]. (3) Morphology of DNs Some DNs that show LLE or BE have branches in the LALs on either side (Fig. 9) [19]. Many pheromone-processing PC neurons innervate the LALs, and that some of these neurons provide the anatomical substrate for linking the LAL to the lateral PC [18], a region receiving axons of AL projection neurons. The LALs on the two sides of the brain are linked by bilateral PC neurons with arborizations in each LAL (Fig. 7) [18]. It appears that the LALs may be interposed in the pathway of olfactory information flow from the AL through the lateral PC to the DNs (Fig. 14). Some DNs that show BE in response to phero- monal stimuli do not have neurites in the LALs (Fig. 13) [19]. Instead, their branches innervate a variety of regions within the ventral PC of the brain. Even in the bilateral neurons with dense arborizations in the LALs, some branches project out of the LALs to adjacent neuropil regions and could, in principle, provide a linkage to DNs that reside entirely outside of the LALs [19]. Alterna- tively, olfactory information may be delivered to these DNs by pathways that do not include LAL neurons (Fig. 13). In general, LLE has been associated with neurons which contain branches within the LALs, while many cells exhibiting BE lack projections within these neuropil areas (Fig. 13) [19]. When tested with pheromonal stimuli applied to either antenna individually, the former neurons respond to stimuli applied to either side, while the latter are excited only by stimulation of the ipsilateral anten- na (Fig. 13) [19]. Thus, it could be that multiple, higher-order pathways exist by which olfactory information reaches DNs. Some pathways may carry information confined to one side of the brain and may elicit BE in DNs that do not invade the LALs with their arborizations. Other pathways may lead to the LALs, where olfactory informa- tion from both sides of the brain is collected, and may elicit LLE in DNs with neurites in this neuro- pil area (Fig. 14) [19]. (4) LLE responses in DNs Long-lasting increases in firing elicited by phero- mone components have been described in a class of axons termed “flip-flopping” interneurons in male B. mori [59]. These neurons exhibited con- ditional responses, in that stimuli applied when a neuron was in a State of low-frequency firing elicited accelerated firing, while identical stimuli applied when the neuron was in a state of high- frequency firing caused decelerated firing (hence the term flip-flop). The conditional responses of M. sexta DNs differ from the flip-flopping of DN in B. mori in several ways. In B. mori, long-lasting increases in firing could be activated in flip- flopping neurons by the primary component of the pheromone blend, bombykol, which also activates the entire sequence of zig-zag walking by which males approach signalling females [59]. In M. sexta individual pheromone components fail to elicit LLE, but blends of the major pheromone components or female extract are effective (Fig. 10) [19]. In B. mori, both high and low states of firing elicited by pheromone components were stable and lasted as long as 4min. In M. sexta, LLE elicited by pheromone blends can last several tens of seconds, but the conditional reduction of firing elicited by subsequent pheromone-blend stimuli spontaneously reverts to the state of high frequency firing after several seconds (Fig. 11) [19]. The activity states of flip-flopping neurons of B. mori were correlated with changes in antennal posture that occur in association with turns during the olfactory-mediated zig-zag walking [59]. Another group of DNs has been identified in B. mori that showed LLE responses to stimulation of antenna by bombykol but did not give conditional responses similar to flip-flopping [77]. The dose- response relationship between bombykol stimuli and prolonged firing of these DNs closely resem- bled a similar relationship between pheromonal stimuli and production of wing fluttering. Kanzaki and Shibuya [77] suggested that these neurons may have a role in initiation and maintenance of the “mating dance” of B. mori. The morphology of these neurons was shown by dye injection to be very similar to that of the M. sexta neurons illus- trated in Fig. 9C [77]. The responses recorded in DNs may play a role in central processing of olfactory information or in 262 R. KANZAKI AND T. SHIBUYA Fic. 14. Schematic summary of higher-order olfactory pathways in the brain of male moth Manduca sexta. This diagram summarizes functional pathways for flow of olfactory information through the brain derived from the morphologies of pheromone-sensitive interneurons that were revealed by Lucifer Yellow injection. Frontal view. Output neurons (AL PNs) of the antennal lobes (AL), the primary olfactory neuropil, are depicted by lines with a letter “a”. Lines with a latter “p” depict olfactory neurons intrinsic to the protocerebrum. Descending neurons are represented in lines with a letter “d”. Boxes outlined by lines depict prominent target neuropil areas in the protocerebrum defined by the projection patterns of olfactory neurons. LP; lateral protocerebrum, SP; superior protocerebrum, VP; ventral protocerebrum. generating or controlling olfactory mediated be- havior. The LLE and state dependent activity changes are particularly intriguing, because they seem to represent an essential link towards be- havior. These neurons are premotor DNs to the motor circuits of the thorax and their responses are elicited by complex and behaviorally relevant stimuli, i.e., the pheromone blend rather than individual pheromone components [19]. Analysis of upwind and casting flight behavior in several moth species by Baker’s group has led to a recent proposal that the goal-oriented upwind flight is mediated by the complementary action of a dual flight control system [78-81]. In this system, contact with pheromone-laden parcels of air in the turbulent plume would produce (1) rapidly acti- vated surges of upwind flight that would also be rapidly terminated upon the next encounter with clean air, and (2) long-lasting activation of a counterturning program which would continue to produce turns that comprise the “casting flight” that persists for seconds when the moth loses contact with the pheromone plume. Our finding that the responses of higher-order olfactory neurons, including descending neurons, fall into the general categories of phasic (BE) or tonic (LLE) activity invites the suggestion that these classes of neurons may separately drive the phasic (upwind surges) and tonic (counterturning) com- ponents of the dual flight control system [73-81]. ACKNOWLEDGMENTS The authors thank Drs. John G. Hildebrand and Edmund A. Arbas for their assitance of the research. This research has been supported by NIH grants (DC- 00348 and NS-07309) to Drs. John G. Hildebrand and Edmund A. Arbas, Arizona Research Laboratories, Division of Neurobiology (ARLDN). This research was done for the most part when the author (R. Kanzaki) was in the ARLDN. The research was also supported in part by Grant-in-Aid from the Ministry of Education, Science and Culture of Japan (Nos. 02640547, 03304010) and by a Grant-in-Aid from a Bio-Media Research Program of the Ministry of Agriculture, Forestry and Fisheries (BMP92- [-2-1). REFERENCES 1 Schneiderman, A. M., Hildebrand, J. G., Brennan, M. M. and Tumlinson, J. H. (1986) Nature, 323: 801-803. 2 Willis, M. A. and Arbas, E. A. (1992) J. Comp. 10 11 12 13 14 15 16 7 18 19 20 Za Jip a3) 24 D5 Olfactory Brain Neurons of Insects Physiol. A, in press. Sanes, J. R. and Hildebrand, J. G. (1976) Dev. Biol., 51: 282-299. Sanes, J. R. and Hildebrand, J. G. (1976) Dev. Biol., 51: 300-319. Keil i: -Ay (1989) Tiss. Cell, 21: 139=151: Kissling, K-E., Hildebrand, J. G. and Tumlinson, J. H. (1989) Arch. Insect. Biochem. Physiol., 10: 273- 279. Tumlinson, J. H., Brennan, M. M., Doolittle, R. E., Mitchell, E. R., Brabham, A., Mazomenos, B. E., Baumhover, A. H. and Jackson, D. M. (1989) Arch. Insect. Biochem. Physiol., 10: 255-271. Schweitzer, E. S., Sanes, J. R. and Hildebrand, J. G. (1976) J. Insect. Physiol., 22: 955-960. Matsumoto, S. G. and Hildebrand, J. G. (1981) Proc. R. Soc. Lond. B, 213, 249-277. Kanzaki, R., Arbas, E. A., Strausfeld, N. J. and Hildebrand, J. G. (1989) J. Comp. Physiol. A, 165: 427-453. Christensen, T. A. and Hildebrand, J. G. (1987) In “Arthropod brain: its evolution, development, structure and functions”. Ed. by A. P. Gupta, Wiley, New York, pp. 457-484. Rospars, J. P. and Hildebrand, J. G. (1991) in press. Camazine, S. M. and Hildebrand, J. G. (1979) Soc. Neurosci. Abstr., 5: 155. Christensen, T. A., Hildebrand, J. G., Tumlinson, J. H. and Doolittle, R. E. (1989) Biochem. Physiol, 10: 281-291. Erber, J., Homberg, U. and Gronenberg, W. (1987) In “Arthropod brain: its evolution, development, structure and functions”. Ed. by A. P. Gupta, Wiley, New York, pp. 485-511. Hildebrand, J. G., Matsumoto, S. G., Camazine, S. M., Tolbert, L. P., Blank, S., Ferguson, H. and Ecker, V. (1980) In “Insect neurobiology and pesticide action (Neurotox 79)”., Soc. Chem. Ind., London, pp. 375-382. Tolbert, L. P., Matsumoto, S. G. and Hildebrand, J. G. (1983) J. Neurosci., 3: 1158-1175. Kanzaki, R., Arbas, E. A. and Hildebrand, J. G. (1991) J. Comp. Physiol. A, 168: 281-298. Kanzaki, R., Arbas, E. A. and Hildebrand, J. G. (1991) J. Comp. Physiol. A, 169: 1-14. Bohm, K. K. (1911) Arb. Zool. Inst. Univ. Wien 19: 219-226. Johnston, C. (1855) Q. J. Microsc. Sci., 3: 97-102. Janet 99D) Te ReAcads Sein, 1523 110=112; Homberg, U., Montague, R. A. and Hildebrand, J. G. (1988) Cell Tissue Res., 254: 255-281. Homberg, U., Christensen, T. A. and Hildebrand, J. G. (1989) Ann. Rev. Entomol., 34: 477-501. Olberg, R. M. (1983) Physiol. Entomol., 8: 419- 26 Di 28 29 30 31 3 33 34 35 36 Sif 38 39 40 4] 42 43 44 45 46 47 48 49 50 Sl SZ 263 428. Kanzaki, R. and Shibuya, T. (1983) Appl. Ent. Zool., 18: 131-133. Kanzaki, R. and Shibuya, T. (1986) Zool. Sci., 3: 409-418. Koontz, M. A. and Schenider, D. (1987) Cell Tissue Res., 249: 39-50. Light, D. M. (1986) In “Mechanisms in Insect Olfaction”. Ed. by T. Payne, M. C. Birch and C. E. J. Kennedy. Oxford Univ. Press, London, pp. 287- 301. Boeckh, J. and Boeckh, V. (1979) J. Comp. Phy- siol., 132: 235-242. Boeckh, J. and Ernst, K. D. (1987) J. Comp. Physiol. 161: 549-565. Boeckh, J., Ernst, K. D., Sass, H. and Waldow, U. (1984) J. Insect Physiol., 30: 15-26. Rospers, J. P. (1988) Int. J. Insect Morphol. Embryol., 17: 243-249. Burrows, M., Boeckh, J. and Esslen, J. (1982) J. Comp. Physiol. 145: 447-457. Arnold, G., Masson, C. and Budharugsa, S. (1985) Cell Tissue Res., 242: 593-605. Homberg, U. (1984) J. Comp. Physiol., 154: 825- 836. Mobbs, P. G. (1982) Philos. Trans. R. Soc. London Ser. B, 198: 309-354. Goll, W. (1967) Z. Morphol. Okol. Tiere, 59: 143- 210. Masson, C. (1972) Z. Zellforsch. Mikrosk. Anat., 134: 31-64. Power, M. E. (1946) J. Comp. Neurol., 85: 487- Sli Rodriguez, V. (1988) Brain Res. 453: 299-307. Rodriguez, V. and Buchner, E. (1984) Brain Res., 324: 347-378. Stocker, R. F. and Gendre, N. (1988) Dev. Biol., 127: 12-24. Stocker, R. F., Singh, R. N., Schorderet, M. and Siddiqi, O. (1983) Cell Tissue Res., 232: 237-238. Christensen, T. A. and Hildebrand, J. G. (1987) J. Comp. Physiol., 160: 553-569. Homberg, U. and Hildebrand, J. G. (1989) Cell Tissue Res., 258: 1-24. Homberg, U., Kingan, T. G. and Hildebrand, J. G. (1987) Cell Tissue. Res., 248: 1-24. Hansson, B. S., Christensen, T. A. and Hildebrand, J. G. (1991) J. Comp. Neurol. A, in press. Kanzaki, R. and Shibuya, T. (1987) Soc. Neurosci. Abstr. 13: 139. Waldow, U. (1977) J. Comp. Physiol., 116: 1-17. Homberg, U. (1984) J. Comp. Physiol. A, 154: 825-836. Schildberger, K. (1984) J. Comp. Physiol. A, 154: 71-79. 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 264 Pearson, L. (1971) Proc. R. Soc. Lond. B, 259: 477-516. Selzer, R. (1979) J. Comp. Physiol., 134: 159-163. Selzer, L. S. (1984) J. Insect Physiol., 30: 537-546. Ernst, K. D., Boeckh, J. and Boeckh, V. (1977) Cell Tissue Res., 176: 285-308. Kent, K. S. (1985) Ph. D. Thesis, Harvard Universi- ty, Cambridge, MA. Christensen, T. A., Mustaparta, H. and Hildeb- rand, J. G. (1989) Chemical Senses, 14: 463-477. Olberg, R. M. (1983) J. Comp. Physiol. A, 152: 297-307. Erber, J. (1978) Physiol. Entomol., 3: 77-89. Schiirmann, F. W. (1974) Exp. Brain Res., 19: 406— 432. Dickinson, P. S. and Nagy, F. (1983) J. Exp. Biol., 105: 59-82. Starratt, A. M., Dahm, K. H., Allen, N., Hildeb- rands Ge abaynes wll len ROllctastlan (9 7S)s0 2: Naturforsch., 34C: 9-12. Moulins, M. and Nagym, F. (1985) In “Model Neural Networks and Behavior”. Ed. by A. I. Selverston, Plenum Press, NY and London, pp. 49- 68. Hounsgaard, J. and Mintz, I. (1988) J. Physiol., 398: 591-603. Llinas, R. R. (1988) Science, 242: 1654-1664. Katz, P. S., Eigg, M. H. and Harris-Warrick, R. M. (1989) J. Neurophysiol., 62: 558-570. Jan, Y. N., Jan, L. Y. and Kuffler, S. W. (1979) PNAS, 76: 1501-1505. Kondoh, Y. and Hisada, M. (1986) J. Comp. R. KANZAKI AND T. 70 71 72 73 74 is 76 Ta 78 TY 80 81 SHIBUYA Neurol., 251: 334-348. Usherwood, P. N. R. (1978) Adv. Comp. Physiol. Biochem. 7: 227-309. Roberts, C. J., Krogsgaard-Larsen, P. and Walker, R. J. (1981) Comp. Biochem. Physiol., 69C: 7-11. Waldrop, B., Christensen, T. A. and Hildebrand, J. G. (1987) J. Comp. Physiol. A, 161: 23-32. Kanzaki, R., Arbas, E. A. and Hildebrand, J. G. (1989) In “Neural Mechanisms of Behavior”. Ed. by J. Erber, R. Menzel, H. J. Pfliger and D. Todt, Proc. 2nd Int. Congress of Neuroethology, Georg Thieme Verlag, Stuttgart, N.Y., pp. 213. Kaissling, K. E. and Kramer, E. (1990) Arch. Insect Biochem. Physiol., 10: 273-279. Lehman, H. K. (1990) Soc. Neurosci. Abstr., 16: 1334. Olberg, R. M. and Willis, M. A. (1990) J. Comp. Physiol. A, 152: 279-307. Kanzaki, R. and Shibuya, T. (1986) Brain Res., 377: 378-382. Baker, T. C. (1986) In “Mechanisms in Insect Olfaction”. Ed. by T. L. Payne, M. C. Birch, and C. E. J. Kennedy, Oxford Univ. Press, Oxford, pp. 39- 48. Baker, T. (1990) In “Proc. 10th Int. Symp. Olfac- tion and Taste”. Ed. by K. B. Déving, Oslo, pp. 18- 35 )o Baker, T. C., Willis, M. A. and Phelan, P. L. (1984) Physiol. Entomol., 9: 365-376. ' Kanzaki, R. and Arbas, E. A. (1990) Soc. Neurosci. Abstr., 16: 758. ZOOLOGICAL SCIENCE 9: 265-274 (1992) © 1992 Zoological Society of Japan REVIEW Primary Structure and Function of a Dynein Motor Molecule KAZUO OGAWA Department of Cell Biology, National Institute for Basic Biology, Okazaki 444, Japan INTRODUCTION Gibbons and Rowe [1] were the first to describe the microtubule-associated motor protein known as dynein, which they found when they isolated an ATPase from Tetrahymena cilia. The axonemal dyneins are contained in the outer and inner arms that project from the peripheral doublet microtu- bules, and they form cross bridges between adja- cent doublet microtubules. The ATP-driven cross- bridge cycles generate the sliding between micro- tubules that gives rise to both flagellar and ciliary movements [2]. The outer and inner arm dyneins are different in terms of their peptide composi- tions. Both the dyneins are multimeric proteins and are composed of heavy chains with ATPase activity, which are assumed to be motor peptides, and several accessory peptides. The molecular mass of the heavy chains in the dyneins (~500 kDa) is 2.5-fold larger than that of the myosin heavy chain and 4.5-fold larger than that of the kinesin heavy chain, the other well-known motor molecules. Progress in analyzing the structure and function of dynein heavy chains was hindered by the lack of primary sequence information, which was due to the difficulties encountered in attempts to clone a cDNA that encodes such a large pep- tide. Nevertheless, the sequence was avidly pur- sued by many groups for quite sometime. Finally, Gibbons ef al. [3] and Ogawa [4] simultaneously determined the complete amino-acid sequence of a dynein motor molecule: the well-characterized heavy chain of sea-urchin axonemal dynein. The Received January 29, 1992 present review will be limited to a description of the “brave new world” of the sea-urchin axonemal dynein motor molecule, as revealed by molecular cloning. THE DYNEIN MOTOR MOLECULE When demembranated sea-urchin sperm are ex- tracted with high salt, the flagellar beat frequency of extracted sperm is only about half that of control sperm that have not been exposed to high salt [5]. Examination by electron microscopy revealed that extraction with high salt removes most of the outer arms from the doublet microtu- bules, leaving the inner arms apparently intact. The outer arms can be purified as ATPase- containing particles with as S value of 21 (referred to an the outer-arm dynein or-21S dynein) by centrifugation through a sucrose density gradient. SDS-polyacrylamide gel electrophoresis (SDS- PAGE) resolved the outer-arm dynein into at least nine different peptides: a and f£ heavy chains (DaHC and DfHC); three intermediate chains (IC1-3); and at least four light chains (LCs) [6]. Exposure to a low-salt medium converts 21S dy- nein into three smaller factions: one containing the DfHC/ICI1 complex; one containing aggregates of DaHC; and one containing IC2 and IC3 [7]. Sale et al. [8] were able to examine isolated outer-arm dynein by the quick-freeze, deep-etch technique. Replicas revealed that the 21S particles were composed of two globular heads jointed by two irregularly shaped stems that made contact along their length. One head was pear-shaped and the other was spherical. The stems were decorated — 266 K. OGAWA with a complex of bead-like particles. The DBHC/ IC1 complex, obtained as described above, con- tained only single-headed molecules with single stems. These heads were predominantly pear- shaped. Sale et al. concluded that each head is formed by a heavy chain, that the pear-shaped head contains the DSHC, and that the spherical head contains the DaHC. Three intermediate chains might decorate the stem that is joined to each head. The position in situ of LCs in the outer arm has not been described. Sale et al. also observed in situ the outer-arm dynein of demem- branated sptem. When frozen in reactivation buffer in the absence of ATP, each arm consists of a large globular head that is attached to the A-subfibers of doublet microtubules via distally skewed subunits and is attached to the B-subfibers by a slender stalk. In the presence of ATP, this head shifts its orientation such that it can be seen to be constructed from two globular domains. One interpretation of these observations is that these structural changes represent distinct states of a cyclic cross-bridge cycle. The subfractionated sam- ples of the outer-arm dynein were assessed by a translocation assay in vitro, in which putaitve motor protein was allowed to adsorb to a glass coverslip, and microtubules were then applied together with ATP. The “gliding” movement of microtubules under such conditions can be ex- amined by video-enhanced contrast-differential in- terference contrast (VEC-DIC) microscopy [9]. This system was originally introduced to monitor the activity of microtubule-associated motor pro- tein in cytosolic extracts of squid giant axons, with the resultant discovery of kinesin [10, 11]. The motor proteins, when properly oriented on a fragment A coverslip, can interact with a microtubule in such a way that they generate force along it, causing the microtubule to glide along the glass surface. Motors that are not properly oriented, rather than retarding the microtubule, seem unable to interact with it and have no apparent effect on the net production of force. Sale and Fox [12] observed that microtubules also glide on coverslips coated with just the DSHC/IC1 fraction. Neither the DaHC nor IC2/IC3 fractions were associated with gliding of microtubules. PROTEOLYTIC AND PHOTOLYTIC ANALYSES The functional substructure (site of hydrolysis of ATP within the molecule) of DBHC was revealed by a classical approach rather than by molecular cloning. Ogawa [13] first obtained a tryptic frag- ment with ATPase activity from a low-salt extract of dynein and named it fragment A. Fragment A is a molecule of about 360-400 kDa in its native form and it can be separated into two peptides, desig- nated f2 (190 kDa) and f3 (135 kDa), by SDS- PAGE [14]. Since f2 and f3 remain associated with each other during native PAGE, it is possible that the corresponding two regions of DGHC could be folded back on each other via intramolecular interactions. Ow et al. [15] established the princi- pal pathway for tryptic cleavage of DSHC in a low-salt buffer, as shown in Figure 1. They iso- lated fragment B (also known as f1 peptide, 130 kDa) which is detached from fragment A during digestion. DaHC did not generate a stable tryptic fragment. This result suggests that the two heavy chains that make up the outer arms are structurally {3 (135 kDa) as | (hydrolysis of ATP, DBHC —U binding to B-subfibers) f2 (190 kDa) fragment B ————~_ f1 (130 kDa) (binding to A-subfibers) Fic. slr subsequent cleavage. Principal pathway of tryptic cleavage of DBHC. T1 indicates the early cleavage and T2 indicates the A Dynein Motor Molecule 267 different from one another. The outer-arm dynein has two microtubule- binding sites. The ability of isolated dynein to rebind to the extracted axonemes was reported by Ogawa and Mohri [16]. Functional recombination of isolated outer arms revealed that the outer arms bind to the A-subfibers in a salt-dependent manner [17]. This type of binding ability of the outer arms was not associated with fragment A [13]. Frag- ment B may contain the binding site for A- subfibers [15]. However, the ability of fragment B to rebind to the extracted axonemes has not yet been demonstrated. The outer arms can associate with the adjacent B-subfibers in an ATP- dependent manner, as described above [8]. Since fragment A has ATPase activity and is slightly activated by the B-subfiber fraction [13], outer- arm dynein may interact with the B-subfibers through the fragment A moiety of DSHC in a ATP-dependent manner. Irradiation at 365 nm of D@HC in the presence of Mg-ATP and a low concentration of vanadate (V;) cleaved DBHC at a single site termed the V1 site and ATPase activity decayed in a biphasic manner [18]. Because vanadate can potently sup- press the activity of dynein ATPase, probably via occupation of a site that is normally reserved for the y-phosphate of ATP, the V1 site probably lies in the hydrolytic domain of the D@HC. Irradiation in the presence of Mn’ ions and of a higher concentration of V; resulted in cleavage of DBHC at a single site, designated V2, but this cleavage at the V2 site was not correlated with any direct effect on ATPase activity [19]. The peptides produced by sequential cleavage at the V2 site and then the V1 site indicated that the two sites are separated by a region of 100 kDa along the length of the DBHC. The ATP-hydrolysis pocket of the central domain might be composed of the y-P;- binding V1 site and the purine-binding V2 site. The D@HC can be covalently modified by reaction with the hydrolyzable photoaffinity analog of ATP, 8-azido adenosine 5-triphosphate (8- N3ATP), which is hydrolyzed by fragment A at about 10% of the rate of hydrolysis of ATP [15]. The V2 site was found to be close to the locus of attachment of 8-N3;ATP, which may correspond to the purine-binding region of the ATP-hydrolytic site on the D@HC. Mocz et al. [20] proposed a map of the sites of tryptic and photolytic cleavage on the D@HC, as shown in Figure 2. HUNTING FOR A GENUINE CLONE A much more direct approach to the analysis of the functional site is provided by the molecular cloning of the gene for D@HC. Garber et al. [22] claimed initially that they had isolated cDNAs for the dynein heavy chain from trout testis that predicted an extensive, carboxy-terminal, a-helical coiled-coil domain. Because of incomplete charac- terization, it is unknown which. of the several QQVAPLQANEVAI TITLAN?7LVGGLA?E?V fragment B fragment A Rpm eed es ee reset eri seofragmenteAie ve sr COOH Fic. 2. Tryptic (T) and photolytic (V) sites within D@HC. The original map proposed by Mocz et al. [20] has been revised [21]. Numbers below the top map represent molecular masses in kDa, as determined by SDS-PAGE. The second map shows the positions of three tryptic fragments in the molecule. The amino-acid sequences of the f2 and f3 peptides are also shown. The bottom map shows that fragment A is located on the carboxy-terminal side of the molecule adjacent to fragment B. 268 K. OGAWA heavy chains of trout dynein [23-25] these clones might encode. Mitchell [26] isolated genomic clones of a and ~ heavy chains from Chlamydomo- nas. However, none of these clones have yet been sequenced. Foltz and Asai [27] characterized a cDNA that encodes a portion of sea-urchin ciliary DGHC. Although four independent criteria sug- gest that their clone encodes a portion of DBHC, their identificaiton of immunoreactive clones from expression libraries cannot be taken as proof that the cDNA clone of interest has been isolated. Hisanaga et al. [28] isolated “cytoplasmic dy- nein” with a molecular mass and immunogenicity similar to those of axonemal D@HC from unferti- lized eggs of the sea urchin. The substructure [29] was also indistinguishable from that of the ax- onemal DBHC [8]. Ogawa et al. [30] showed that DGHCs from sperm and egg cilia may be similar to one another. There is no evidence to suggest that sea-urchin “cytoplasmic dynein” is different from ciliary or sperm DBHC. Recently, Ogawa [31] screened a cDNA library that corresponded to the poly(A)* RNA of unferilized eggs using an anti- body directed against sperm axonemal dynein heavy chains. The cDNA clones (AJ292, 4J296, AA101, AA102, AA103, and AA104) obtained may encode ciliary D@HC. Fingerprints of fusion pro- tein produced by lysogenic AJ296 were similar to those of authentic 21S dynein from sperm. The Northern blot of poly(A)* RNA revealed that 0 5 Mm WS A103 A055 Fic. 3. only two clones (AJ296 and AA103) could hybri- dize with an RNA of ~16kb in length. Since DBHC has an estimated mass of 480 kDa, it could be encoded by poly(A)* RNA of at least 14 kb in length. Thus, the two clones appear to be strong candidates. Finally, the amino-acid sequence de- duced from the nucleotide sequence of AA103 contains one of the ATP-binding motifs (GKT site, see below) and the amino-terminal sequence of the f2 peptide [21]. Thus, these two clones appear to be the first genuine partial clones of cDNA that encodes DGHC. CONSTRUCTION OF FULL-SIZE COMPLEMENTARY DNA The AJ296 and AA103 clones encode the car- boxy-terminal and central regions of D@HC, re- spectively. The missing segments of cDNA can be isolated by making mini cDNA libraries primed with oligonucleotides that are complementary to the 5’-portion of these clones, with subsequent screening with radiolabelled DNA probes. Ogawa [4] has sequenced additional three clones (AF1113, XA055, and AMO062). Full-size cDNA was con- structed by the overlapping of five clones, as shown in Figure 3. The long reading frame can encode a protein of 4,466 amino-acid residues with an unmodified molecular mass of 512 kDa. The deduced complete amino-acid sequence of DGHC 10 15 i AMO062 AJ296 Five overlapping clones that encode DBHC. The long open reading frame (thick line), flanked by non-conding sequences (thin lines), is shown at the top. Because of multiple allelic variation, the nucleotide sequences of two clones in the overlapping region are different from one another with frequency of one altered base per 100 bases. For overlapping of clones, weight was given to the nucleotide sequences of AJ296 in the case of AJ296 and AM062, AM062 in the case of AM062 and AAOSS, AA103 in the case of AANS5 and AA103, and AA103 in the case of AA103 and AFI113. is shown in Figure 4. The sequence was confirmed A Dynein Motor Molecule 269 trout and by Foltz and Asai [27] for the sea urchin. by the finding of the two amino-terminal sequences of the f2 and f3 peptides of fragment A in the The former sequence was found at amino-acid residues, 1,192-1,204 and the latter at residues 3,324-3,340. quence shows no significant similarity to the partial sequences reported by Garber et al. [22] for the deduced sequence. 1081 ala y/ak 1261 Usy5zb 1441 i5)3}IL 1621 seals 1801 1891 1981 2071 2161 BAISAL 2341 2431 2521 2611 2701 ai Sal 2881 Ziel 3061 SiIL5ib 3241 S388! 3421 SLL 3601 3691 3781 3871 3961 4051 4141 4231 4321 4411 Fic. 4. Deduced amino-acid sequence of MGDVVDARLD DNIKTNLVYG KDDSYDRSLV TERDVEAALT LEPEESLEKV LSQQVEKIHE CKYPIVLMLY EMLNLLNKYE DLTVAWYNKV KELKKESLLA LFEARLELQV SFDNYAYLYV LIKKWSFMFK TQLQELPEQW AGLFEVNMPD SLRAVSELQN ISLLKSNEEL DFKELAAEME LKFKQDDEGN TEVNISFARL KHCYANICDA YKGLAQTGAW DFELICEIML VFMGLIGDLF DLNPKAVTND SHLKTATPAT ENTPADCPKE AVLVHTNETT LVYFIDDMNM ILSQHLANIA INDQDIEAFE NRILESPRGN ASGEIPDLFA SVSKRFLDEV QQVDDLKAKL NLTELKSFGS AAGGLCSWVV LVGGLASENV EGLPSDRMST KKGRYIKIGD DNLLSRLSSA SLSIARAEPC IKSLSAMEDF EETDPATPVF GSHESYRVYM GWNRSYPFNT YHQYIDEILP YVVVAFQECE PNVVWLGGFF FIKAIPVDKQ FISEYILKSY DLSYTPLEQL HAIESVIIDW EAQDINIHLK RGALTVLKNW EFQECAKVFT DQELDQOSKET QKVFENWTKG RKTVLEVEFP LDDRQDRLKK PDMIFNPSLD DDRKEFMROF QHLIDHVTNS NNTKKIAITI YKOLKACRRE PATRERHWQQ IETLEDNOVQ KTPNVVEATN DTKLALGMYS EEGHENSMKD QOFKYSYEYLG GCFDEFNRIS VAEGFLEARL PALDVPRRRD ELFGIINPAT VSRAGILYIN LYELYFVFAS RVRF'FMDLLM PEVDTYGTVQ VSNALQKLSP KLVFEYAKKF ALLVGVGGSG DDEVENT ELLKGDIKNS ASQEVELAQK PPSAVLKVAA NIVKFYNVYC RWGEAVANFK ENATILSNCQ KEVEYNPEFR EGNFLGDTAL EDVKERVVNL RNLDRDIEGS FILSPGVDPL SAEPAGSPES GDLTISVNVL PESPYLYGLH RMNTLTSEIR NPOSFLTAIM DTRNIYECPV KLKPDKWTKC SALVDEVLVP THQIRDVLKR PLVYQIESMD RELYDEHRAK ERPYDGLDPT YDEHMRVEEA VDEVCKTNLD LIEGQLADLD RYAEITTAGE YGIADGFYDL LLYNHVLTTE LSELQEFIKV KQQVAPLOAN VRLLKGLWDL LMAATKVKFT LONLMTSKHI KARLFDRLEA KEGEYVDFDK YNKKQILOLN NTPRLVITPL VEVLSVVAVQ LARKFITLYT LDFEKVVKQS REWKDGLFSV PSDLGWNPIV IWAFGGSMFQ ERGRPVMLVG PHTLIRQ TVVSATLDLH FEDVDEEALK KOSLARLASY VRNEVKGMGL TAEFMAYVHV NEDADKLIOQV AVMVLLAPNG DVEPKRIALQ IQEKTLPGDV RWPLMIDPQL LILQTKLANP VENLETTKRT IDCITYSVFI AKRWKKFVES KDVEALGKKL HIIPQGILES YNYLEANSKV PNAEIGFLTT RSLKELDLGL QSMARKNEWP YKTKORGPTF The entire se- INVEENKILM LLANPRNHEQ DSAQPLLEGL ELEFSDLTPR LKDYFKDGKE COQEFLEDYEE NGNAPLNKNM QSLITRDDAS TRLRQAEADL KIHSLMKENL VEMLISDTYK EITEAHAEDGV GNSGLTKTVE EVAIIRRKCT IMVVRTSIED MDKETTLSDL AHFLEEVSGW IQGSLVVCEK ECECTGOVEV TLIGLLIGKL TDRCYITLTO VKCVQDATRD LCKELLSKQD TLDLKLQAED IMRDMSNITH TSWIDTREVQ DOLVDYRVEF NAGLGKSVLV @ 6060 YKHWYDRAKL KKVAQSFLPT AKPNIHCHFA ISSLEVFQIT QDTRENCWKF SVNESSKOYL VGVETEKVSK KIPKDRSWKA KANDELKAAQ LLITAFVSYI QGIKWIKOKY HYKPEMOAQT AAEISVKVEE YTTRGLFEAD ECPEKEKF'PQ GFTFDNNNFH SIKITNEPPT PWODLRYLF'G ESDNLFKVVL KGELTITPDM LDKMCLOCDV VWTFNLKSKE ATP-BINDING SITES Fragment A corresponds to amino-acid residues 1,192—4,466. In view of the ability of fragment A to hydrolyze ATP, a search was made in this region for the consensus ATP-binding motif LEFLEKADNP WPVVVSQDVL NPGPMVEINFE LAPILHTVCL VKEWEFASPL FEKKVFDLDR PDVAGOLKWS KLIMVNF'DPK NWTSDSVWEY DLFKAERASSD MASLVNRLAE PECPPTLDOF DGDYNGLVDC SFDVRQHEFR WKTTPWLEIN LALNLHNFED QKKLSTTDSV ALAEYLETKR WLNRVMDTMR TKGDROKIMT SLHLVMSGAP KKERF'NFMGE HYDWGLRAIK SFVLKVVOLE DGPKWIVLDG SERANLTILF SKWWITEFKT GDKLSNLGED TLKETHKCOY ATKFHYVFNL TGIGDPKYMP LRKGYGIPDL FIDRLRROLK TNERRYNYTT EKATVDDEEK AKVVMNKVDA DKLALIKAKT GCFTKNYRVD GDELRVIRIG TLINFTVTRD AKVTEVKINE KLIFTTOVAF EWKNKSALOK NVSLGQGOET GMFANLHKAL EIMYGGHITD ELQPRDAGGG EDLSNALFLD TKKNKEDF'SS KAAKWTLAGV QLVFTVNPAG RHVHNLKSSV WKAKCENLDC IWSNSDYYNT VETRMDNFIR RLGSILCQGF AQLRDRISKP LVSVLREVKY IQETRDQVRD IWKAYVDYVD HNGQEHYQAD KEQVDTYEKI MGHLMAVKER ERFRKEAPFI VEQMEMDCKK EVRNIVDKAV ITIWFEVORT LAFPRFYFVS STVRSOFADA ICTIDVHARD AGPACTCKTE EISLIPSVGI SVLVVAGSLK ELLAVRHSVF DIDPMWIESL DKYLPTLLDT IKFPNOGTVF SMVANVPFNY VSCMNPTSGS RDLSNVF'OGL CATWPELNKTI KLDLATVCMK TVLCFSPVGT PKSFLEQTKL KVAIINEEVS FLDSLINYDE AELDANLAEL LODRMWLPFL QRGYLDTIEN GLEDOQLLANV ARELYRPAAA QVLLMKKETA LCMMRALRAD VAEQCMDLAA YNFNODTLEM DWDRRLCRTY GGGGSSREEK QTPASWVKRA APREGSYVHG ALLLOV LITPSYEFPS YVVAGOVKGK IFQQOLRDPKV APRVIVLLQE RIETIQSLFE DDCCGLEAAF MGSLKHMEHP LOIRGEETIP LEKRVQOTKD DMVIDGFFNC LEGMDDLSDV YSEADETEPE QAATDEMF'EP FLFDGPYQCL FAKDIRSLDK KEMGMEKVLK WSHLESIFIG SADLLDILSQ VVSYEEKPRE VVAMMVLKKV TTKDLGRALG FITMNPGYAG RGDPQRPEDOQ VIGNAGTGKS NTVMD KVL LRIRFKKIIP DYY IDQESKK YTTSEMLORV FTINSRLQRH LYSGSDLLKS LVEALDTYNE AGLKNIGTVF TLRVRSRKFP YESLLAMKSK KKAKDCSEDL ENIHENCQOKA TAQFEKATSD KSQKDPIPIT AISSGDTVLI VAQERPDLEK RASLLYFILN QNELDF'LLRF RMSYAVRNF I KEGHWVILON CAREAEFKVI LEEYMAPEML IKSLLDEIVE YPSLFGLSAW LFMEGARWDT ALKNTKATYF TLLPLPVGSE RKMKELLERT ICNLLIDLCR TNVEFSKLEK KMLDCYGPLL TGVRRILESE ESAASTYEKH NVDRIKKIMA IHCTLTYLLE RNDLMDRVOT QVFDAWFRVD IKOTIELLKT DKCHSETYEM EMRAWDAYNG ELNTTWSSMD SEDIRNQLPE GNNPTQOVQRH QWLYDY PAOQV DSAQAFQWLS IMVYVFNCSE RTELPENLKA VLMRALRDFN QVLKVLNKTY TLASNERIPL IPEQSMVOML FLPWSEKVPT LEKPLEKKAG FCVFALSF PG PIDFARLWMH INAVMNLVLF LMTDAQVSDE AVVNCTSIDW ELTAKMERLE AKAEPALLAA IKEYLNDPEF KLKCQOEAEA EGLDVLSMLT ENMEEST DPV LKSDLTKQON DLNKINPLYO PIQVGLTSPV EEKLGSKYVE THLVAKWLST LFALCYFHAV DGDLYLAPGF KLPEEFNMME YADLLORIKE QTNMIADARL IKKGREPVGK KVETAAGSEE QSSYLPSFNN TFLDPSEIFK TEMGSMKGRM DRPVIRNDFE DAKVIFQOKYE ETLRKYVANL EWTKQPLFER NTDPRHCAAP IMTKAQEYRN SKPFKAALLN YDQEMSEEVH EEHMAKLQES LDATVKNMLT FDYEPHSRTG DSKRFDGIDT LSKLFDNMAK ALATTQVWWT QLRHRWADDD QMDYKSCGNI LFRPCAMVVP VPKIVSDDTP SNMKRKPVF I TPSMRLLFET CYLLECLLTP FELDPETIPMQ RNYGPPGTKK QDALSTTYNS ECQRVYGDKM EDAMQHVCRI KFLVLINDLL FHEWPQEALV NGLTKLOSTA QEALNTLNKN EPEYIKGKSL TSRTLTLANR DDADIAVWNN LDPVLGRNTI DFKIILKELE FSLKAFNTVF DFLTNSAWGA GROVEFAKSY LEKKLEQYSI VCERQKFGPQ PVPPNSDYKG IMGKVEDRTP LEQWTADFAL KELAPNMPVI sea-urchin Anthocidaris crassispina axonemal DBHC. The amino-acid residues that have been confirmed by direct sequencing are underlined and putative-ATP-binding motifs are indicated by filled circles. Amino acids marked with filled squares define the sites of trypsin cleavage. 270 K. OGAWA GXXXXGK(T/S), where X is any amino acid. This motif was found at three positions in the f2 peptide as follows, between residues 1,192 to 3,323: beginning at Gly 1,852 (termed the GKT site); at Gly 2,133 (the GKS1 site); and at Gly 2,460 (the GKS2 site). The sequence GXXXSGK is also accepted as an ATP-binding sequence of adenylate kinases [32], and this sequence was also found in the f2 region, beginning at Gly 2,805 (the SGK site). Therefore, molecular cloning of DBHC has revealed the presence of four putative ATP- binding sites in the middle region of the molecule. Since the amino-terminal sequence of the {2 peptide begins at residue 1,192, the T1 site can be identified on the carboxy side of Lys 1,191. According to the map of DSHC (Fig. 2), the y-P;-binding site (V1 site) is separated by a region of 70 kDa from the T1 site in the carboxy-terminal direction. There are 660 amino-acid residues be- tween Gln 1,192 of the T1 site and Gly 1,852 of the GKT site, and this distance is equivalent to a peptide of 73kDa. Thus, the GKT site corre- sponds to the V1 site revealed by photocleavage of DBHC and may be able to catalyze the hydrolysis of ATP. The binding site for 8-N;ATP (V2 site) is separated by a region of 170 kDa from the T1 site in the carboxy-terminal direction. There are 1,603 amino-acid residues between Gln 1,192 of the T1 site and Gly 2,805 of the SGK site, and this distance is equivalent to a peptide of 177 kDa. Thus, the SGK site corresponds to the V2 site. Since fragment A can hydrolyze 8-N3ATP at about 10% of the rate of hydrolysis of ATP, the SGK site may also be able to catalyze the hydrolysis of ATP. The presence of two GKS sites was not predicted by the photocleavage of DBHC. The sites have sequences that are very similar to one another. The ATP-dependent C/pA protease of E. coli has also two ATP-binding motifs which are very simi- lar to one another [33]. Thus, the sequence similarity between the two GKS sites in the mole- cule may not be a coincidence, but may represent proof of two functional sites for hydrolysis of ATP. Fragment B does not have any ATPase activity [15]. The sequence AXXXXGKT, beginning at Ala 154, appears to be a modified nucleotide- binding motif, as found in the GTPase superfamily [34]. At the present time, however, it is uncertain whether fragment B has the ability to bind GTP and catalyze its hydrolysis. The position of the ATP-binding motif on a motor molecule may be related to the directional- ity of movement along a microtubule. Both dynein and kinesin are microtubule-motor proteins and they move in opposite directions along a microtu- bule. The striking difference between the amino- acid sequences of both motor proteins is reflected in differences between the positions of the ATP- binding motifs on their heavy chains; the motif is located at the amino terminus of the kinesin heavy chain [35] and in the midregion of DGHC. The product of the claret (or ncd) gene belongs to the kinesin superfamily. It is noteworthy that the ATP-binding motif is located at the midregion of this gene product [36, 37] and the molecule moves . toward the microtubule’s minus end [38, 39], a direction characteristic of dynein [40]. POLYMORPHISM The nucleotide sequence shows two types of polymorphism (Fig. 5). When AF1113 was iso- lated, 14 additional shorter cDNAs were also obtained. Two clones, AF1106 and AF1114, were sequenced and their cDNAs overlapped the se- quence of AF1113. Fifteen bases common to both AF1106 and AF1113, which encode five amino acids, were absent in the sequence of AF1114 (possibly as a result of alternative splicing). Fur- thermore, the underlined nucleotide sequence in Residue number AF1113 AF 1106 AF 1114 Fic. 5. 609 OHO 61116127613) O14 1615 OL6mOie7, H Bae eoe GeV RG Re ig L GAT ECG ACG GGT GTC AGG AGA ATT Tre CAT CCG ACC GGT GTC AGG AGG ATT TTG CAT) CCG ssa pees cca tebe = ART EeETe Polymorphism of cDNA clones that encode DSHC. A Dynein Motor Molecule 24/1 Figure 5 differed between AF1106 and AF1113 (possibly as a result of multiple alleles). The latter type of polymorphism occurs at a rate of about one base per 100 bases in the two overlapping clones but, so far, no substitutions of amino acids have been found. Since the full-size cDNA was con- structed by overlapping of the present five clones, which include AF1113, the amino-acid sequence of DBHC described here is just one possible se- quence, and slightly longer and shorter versions may also be present in the sea urchin. SECONDARY STRUCTURE The secondary structure of DBHC was analyzed by Dr. Ken Nishikawa of the Protein Engineering Research Institute, Osaka, Japan (Fig. 6). There are two long a-helix-dominant regions (termed al and a2) in the sequence, suggesting that the DGHC is composed of three large (-structure-dominant domains (termed the N, M, and C domains) sepa- rated by these regions. The M domain is split by short a-helix-dominant regions into four smaller #-structure-dominant regions, and the C domain is similarly split into three smaller region. Although analysis of secondary structure predicts that the al region is rich in a@ helix, there are no long hy- drophobic heptad repeats without interruption, as Major domains N B structure dominant a a. helix dominant are found in the a-helical coiled-coil regions of filamentous motor proteins such as myosin and kinesin. The a2 region contains two heptad re- peats, which are predicted to be largely a-helical, at amino-acid residues 3,028-—3,153 and 3,234- 3,338 with interruptions, as shown in Figure 7. In particular, two leucine heptad repeats at residues 3,028-3,083 and 3,262-3,303 could favour the formation of a leucine zipper structure, with result- ant generation of a large globular structure from the M and C domains. This leucine zipper struc- ture may explain why f2 and f3 peptides remained together in fragment A during tryptic digestion of DBHC, while f1 was detached, as described above. MODEL OF THE STRUCTURE Figure 8 shows a model of structure of DBHC, as deduced from the predictions about secondary structure and the proteolytic analysis of the au- thentic protein. Quick-freze deep-etch electron microscopy of the DGHC/IC1 complex revealed that the complex is composed of a pear-shaped head and an irregularly shaped stem, while the base looks like a small globular bead [8]. Accord- ing to this structural model, the N domain may correspond to the base, the al region to the irregularly shaped stem, and the associated M and M Cc. a2 Le ee ee ee! 0 1,000 2,000 3,000 4,000 Residue number Fic. 6. Predicted secondary structure of the DGHC. The four arrowheads in the M domain indicate the GKT, GKS1, GKS2, and SGK sites of the ATP-binding motifs, from the left. The arrowheads in the al and a2 regions indicate the Tl and T2 sites, respectively. 272 C domains to the pear-shaped head. As described above, it has also been documented that the outer arm in situ attaches to the B-subfibers of adjacent outer-doublet microtubules via a slender stalk. This slender stalk is seen neither in the isolated outer arms nor in the DGHC/IC1 complex. It is possible that the slender stalk corresponds to the a2 region of the present structural model. a. eC. fh, ¢..a, b.ecd 37, O28 JERI GM SOME AAS IGG 3,049 1 Ee NGG in Ks 53,0708 INAS OMG NEG AE Se OQL V GVA eer 2 he Vi SS MLA VA le aN ee iE Sigal lgs is) AFPALLAA 37334! Y fk G UR Si A WAS WS) PB Ve INAV Sen Ce) 3 Vi 3 AO PRALINE IAAI IK AG Sy BQG A ley A tO). ke Sp Sibi 7 Wil aig te A model of the structure of DBHC. T1 and T2 indicate the principal sites of trypsin cleavage. M domain (hydrolysis of ATP) T2 A Dynein Motor Molecule 273 that have been proposed as the microtubule- binding site. These repeated sequences are not found in the N domain of the DBHC sequence. D PHC transiently associates with the B-subfibers of the adjacent doublet microtubules during the ATP-hydrolytic cycle. Members of the kinesin superfamily of microtubule motor proteins also associate transiently with microtubules during the ATP-hydrolytic cycle. They share a region of sequence homology that extends from the ATP- binding site towards the carboxy-terminal end of the molecule, and this sequence has been sug- gested to constitute the site of ATP-dependent binding to microtubules [35]. This sequence homology is not found in the C domain, which extends from the M domain (multiple ATP- binding sites) toward the carboxy-terminal end of the molecule, or anywhere else in the amino-acid sequence of DGHC. Thus, D@HC seems to be a member of a new family of microtubule-binding motor proteins with unique microtubule-binding sequences that are unlike those of MAP2, tau, or members of the kinesin superfamily. DYNEIN SUPERFAMILY It is natural to speculate that dynein may also be involved in motile functions associated with cyto- plasmic microtubules. Immunologocal studies, us- ing antibodies directed against axonemad dyneins, have shown that the segregation of chromosomes during mitosis [43-47] and meiosis [48], and the translocation of melanophore in fish [49] could involve dynein motor molecules. MAPIC is a member of a class of five extremely high- molecular-weight microtubule-associated proteins that co-purify with brain microtubules [50] and it is responsible for retrograde transport [51], a proper- ty of axonemal dynein. MAPIC has been now found to be a cytoplasmic form of ciliary and flagellar dynein [52]. Since microtubules have been shown to be responsible for the transport of membranous organelles within the cytoplasm and, thereby, to play a role in axonal transport, secre- tion, endocytosis and transcytosis, cytoplasmic dy- nein could have very general functional role in cells. Isolation of a clone for cytoplasmic dynein is now the goal of many gorups. Axonemal and cytoplasmic dyneins may consti- tute a superfamily of force-generating proteins, with each member possessing a conserved force- generating domain joined to a different “tail” that confers specific attachment properties. The outer- and inner-arm dyneins attach to different sites on axonemal doublet microtubules, while cytoplasmic dyneins interact with organelles and chromo- somes. Structural and enzymatic studies suggest that the motor domains of the dyneins are similar to one another [see ref. 53 for a review]. There- fore, it is likely that the members of the dynein superfamily share a common motor domain that is linked to a distinct tail with unique binding prop- erties in each case. ACKNOWLEDGMENTS The research described in the present review was supported in part by Grants-in-Aid for Scientific Re- search, Category C (nos. 02640574 and 03640634), from the Ministry of Education, Science and Culture of Japan. REFERENCES 1 Gibbons, I. R. and Rowe, A. J. (1965) Science 149: 424-426. 2 Gibbons, I. R. (1981) J. Cell Biol. 91: 107s—124s. 3. Gibbons, I. R., Gibbons, B. H., Mocz, G. and Asai, D. J. (1991) Nature 352: 640-643. Ogawa, K. (1991a) Nature 352: 643-645. 5 Gibbons, B. H. and Gibbons, I. R. (1973) J. Cell Sci. 13: 337-357. ; 6 Bell, C. W., Fronk, E. and Gibbons, I. R. (1979) J. Supramol. Struct. 11: 311-317. 7. shane We-le Yo, Bell, ©. W.. Sale, W.-S. and Gibbons, I. R. (1982) J. Biol. Chem. 257: 508-515. 8 Sale, W. S., Goodenough, U. W. and Heuser, J. E. (1985) J. Cell Biol. 101: 1400-1412. 9 Allen, R. D., Allen, N. S. and Travis, J. L. (1981) Cell. Motil. 1: 291-302. lORViales Re Da Schnapps Bade Reese, i. Ss: and Sheetz, M. P. (1985a) Cell 40: 449-454. Vales Rass Schnapp yy ba dey Keesey lS. and Sheetz, M. P. (1985b) Cell 40: 559-569. 12 Sale, W. S. and Fox, L. A. (1988) J. Cell Biol. 107: 1793-1797. 13 Ogawa, K. (1973) Biochim. Biophys. Acta 293: 514-525. 14 Ogawa, K. and Mohri, H. (1975) J. Biol. Chem. 250: 6476-6483. 15 Ow,R.A., Tang, W.-J. Y., Mocz, G. and Gibbons, 16 17 18 WS) 20 21 DD 23 24 25 26 27 28 29) 30 31 32 33 34 274 I. R. (1987) J. Biol. Chem. 262: 3409-3414. Ogawa, K. and Mohri, H. (1972) Biochim. Biophys. Acta 256: 142-155. Gibbons, B. H. and Gibbons, I. R. (1976) Biochem. Biophys. Res. Commun. 73: 1-6. Gibbons, I. R., Lee-Eliford, A., Mocz, G., Philip- son, C. A., Tang, W.-J. Y. and Gibbons, B. H. (1987) J. Biol. Chem. 262: 2780-2786. Tang, W.-J. Y. and Gibbons, I. R. (1987) J. Biol. Chem. 263: 17728-17734. Mocz, G., Tang, W.-J. Y. and Gibbons, I. R. (1988) J. Cell Biol. 106: 1607-1614. Ogawa, K. (1991b) Proc. Japan Acad. 67B: 27-31. Garber, A. T., Retief, J. D. and Dixon, G. H. (1989) EMBO J. 8: 1727-1734. Ogawa, K., Negishi, S. and Obika, M. (1980) Arch. Biochem. Biophys. 203: 196-203. Ogawa, K. and Izumi, S. (1983) J. Submicrosc. Cytol. 15: 371-374. Gatti, J.-L., King, S. M., Moss, A. G. and Witman, G. B. (1989) J. Biol. Chem. 264: 11450-11457. Mitchell, D. R. (1989) Cell Motil. 14: 434-445. Foltz, K. R. and Asai, D. J. (1990) Cell Motil. Cytoskel. 19: 33-46. Hisanaga, S., Tanaka, T., Masaki, T., Sakai, H., Mabuchi, I. and Hiramoto, Y. (1987) Cell Motil. Cytoskel. 7: 97-109. Hisanaga, S. and Hirokawa, N. (1987) J. Mol. Biol. 195: 919-927. Ogawa, K., Yokota, E., Hamada, Y., Wada, S., Okuno, M. and Nakajima, Y. (1990) Cell Motil. Cytoskel. 16: 58-67. Ogawa, K. (1991c) In “Biology of Echinodermata” Ed. by Yanagisawa, T., Yasumasu, I., Oguro, C., Suzuki, N., Motokawa, T. pp. 339-346. Balkema, Rotterdam. Walker, J. E., Saraste, M., Runswick, M. J. and Gay, N. J. (1982) EMBO J. 1: 945-951. Gottesman, S., Clark, W. P. and Maurizi, M. R. (1990) J. Biol. Chem. 265: 7886-7893. Bourne, H. R., Sanders, D. A. and McCormick, F. (1991) Nature 349: 117-127. K. OGAWA 35 36 21 38 39 40 4] 42 43 44 45 46 47 48 49 50 mal of 53 Yang, J. T., Laymon, R. A. and Goldstein, L. S. B. (1989) Cell 56: 879-889. Endow, S. A., Henikoff, S. and Soler-Niedziela, L. Nature 345: 81-83. McDonald, H. B. and Goldstein, L. S. B. (1990) Cell 61: 991-1000. Walker, R. A., Salmon, E. D. and Endow, S. A. (1990) Nature 347: 780-782. McDonald, H. B., Stewart, R. J. and Goldstein, L. S Bs Celli6320159>1G5s Vale, R. D. and Toyoshima, Y. Y. (1988) Cell 52: 459-469. Lewis, S. A., Wang, D. and Cowan, N. J. (1988) Science 242: 936-939. Lee, G., Cowan, N. and Kirschner, M. (1988) Science 239: 285-288. Mohri, H., Mohri, T., Mabuchi, I., Yazaki, I., Sakai, I. and Ogawa, K. (1976) Dev. Growth & Differ. 18: 391-398. Sakai, H., Mabuchi, I., Shimoda, S., Kuriyama, R., Ogawa, K. and Mohri, H. (1976) Dev. Growth & Differ. 18: 211-219. Yoshida, T. A., Ito, A. and Izutsu, K. (1985) Cell Struct. Funct. 10: 245-258. Pfarr, C. M., Coue, M., Grissom, P. M., Hays, M. E., Porter, M. E. and McIntosh, J. R. (1990) Nature 345: 263-265. Steur, E. R., Wordeman, L., Schroer, T. A. and. Sheets, M. P. (1990) Nature 345: 266-268. _ __ Ogawa, K., Yokota, E. and Shirai, H. (1988) Biol. Cell 64: 57-66. Ogawa, K., Hosoya, H., Yokota, E., Kobayashi, T., Wakamatsu, Y., Ozato, K., Negishi, S. and Obika, M. (1987) Eur. J. Cell Biol. 43: 3-9. Bloom, G. S., Schoenfeld, T. A. and Vallee, R. B. (1984) J. Cell Biol. 98: 320-330. Paschal, B. M., Shpetner, H. S. and Vallee, R. V. (1987) J. Cell Biol. 105: 1273-1282. Vallee, R. B., Wall, J. S., Paschal, B. M. and Shpetner, H. S. (1988) Nature 332: 561-563. Vallee, R. B. and Shpetner, H. S. (1990) Annu. Rev. Biochem. 59: 909-932. ZOOLOGICAL SCIENCE 9: 275-281 (1992) Previous Nutritional State and Glucose Modulate Glucagon-Mediated Hepatic Lipolysis in Rainbow Trout, Oncorhynchus mykiss JAMIE S. HARMON and Mark A. SHERIDAN! Department of Zoology, North Dakota State University Fargo, ND 58105, U.S.A. ABSTRACT— Rainbow trout were used to investigate the effects of previous nutritional state and glucose presence on glucagon-stimulated hepatic lipolysis. Experiments were conducted in vitro on liver removed from fed fish and from fish fasted for 4 or 6 weeks. Basal hepatic lipolysis was enhanced in liver removed from fasted fish compared to fed animals. Glucagon-stimulated lipolysis was more pronounced in liver removed from 4-week fasted fish than in liver isolated from fed fish. Glucagon failed to affect hepatic lipolysis in the liver of 6-week fasted animals. The effects of glucose presence and absence were also examined. Basal lipolysis in liver tissue incubated in the presence of glucose was more pronounced than rates observed in tissue cultured in the absence of glucose. Glucagon stimulated lipolysis in liver incubated in the presence of glucose to a greater extent than liver incubated in the absence of glucose. Insulin inhibited glucagon-stimulated lipolysis in liver from fed and fasted fish as well as in liver cultured in the presence and absence of glucose. These results indicate that glucagon-mediated lipolysis is modulated by previous nutritional state and by glucose. © 1992 Zoological Society of Japan INTRODUCTION Hepatic lipolysis in trout is mediated by a triacylglycerol lipase enzyme which hydrolyzes stored triacylglycerol (TG) to glycerol and fatty acids (FA) [1]. A number of hormones have been found to influence lipolysis in fish [2] and recently we have indicated a role of glucagon family pep- tide (glucagon, glucagon-like peptide)-mediated hepatic lipolysis [3]. Jn vivo administration of glucagon (GLU) or glucagon-like peptide (GLP) resulted in elevated plasma FA accompanied by enhanced hepatic TG lipase activity [3]. Recently, we have demonstrated that GLU acts directly on hepatic lipolysis. In rainbow trout liver incubated in vitro, glucagon stimulated tissue TG lipase activity as well as FA and glycerol release into culture medium [4]. Previous nutritional state has also been shown to profoundly influence the intermediary metabolism of fish. Coho salmon (Oncorhynchus kisutch) fasted for 3 weeks displayed elevated plasma FA concentrations accompanied by enhanced hepatic Accepted January 31, 1992 Received November 11, 1991 " To whom all correspondence should be addressed. TG lipase activity [5]. These fish also exhibited reduced titers of insulin (INS), GLU and GLP compared to their fed counterparts. A similar observation was found in rainbow trout fasted for 6 weeks [6]. Hepatic lipolytic activity in liver re- moved from fasted (4 weeks) trout and cultured in vitro displayed enhanced activity over liver re- moved from fed animals and similarly cultured [4]. Glucose also influences hepatic lipolysis while altering the pattern of pancreatic hormones (in- sulin, glucagon, glucagon-like peptide and somato- statin-25). Glucose injection into rainbow trout resulted in elevated plasma FA levels attended by enhanced hepatic lipolysis [7]. The extent to which nutritional state and/or glucose interacts with hor- mones to modulate lipid mobilization in fish is not known. In the present study, we used rainbow trout to examine the effects of previous nutritional state and glucose presence on _ glucagon-stimulated hepatic lipolysis. The specific questions we ad- dressed were: 1) What are the effects of glucagon on lipolysis in liver removed from animals of varying nutritional states? and 2) What are the effects of glucagon on lipolysis in liver removed from fed fish and incubated in the presence or 276 J. S. HARMON AND M. A. SHERIDAN absence of glucose? We also report the effects of insulin on lipolysis in liver removed from animals of varying nutritional states and in liver incubated in the presence or absence of glucose. MATERIALS AND METHODS Experimental animals Yearling rainbow trout (mean weight +SEM, 54 +5 g) of both sexes were obtained from Garrison National Fish Hatchery near Riverdale, ND and maintained at North Dakota State University in dechlorinated municipal water (13°C) under 12L: 12D photoperiod. Fish were fed ad libitum twice daily with Glencoe Mills Trout Chow (Glencoe, MN). Experiments were conducted in the fall of the year. Animals were divided into three groups of varying nutritional regimes: fed continuously (ex- cept 24-36 hr prior to experimentation), fasted 4 weeks, and fasted 6 weeks. Immediately prior to experimentation, fish were anesthetized in buf- fered 0.01% (w/v) tricaine methanesulfonate (MS- 222), bled from the severed caudal vessels, and the livers removed and prepared for organ culture. Organ culture and incubation conditions In vitro experiments were carried out on livers removed from several lots of fish (10-12 indi- viduals per lot) within each of the nutritional groups. Livers were carefully perfused by means of syringe and needle with cold (14°C) saline (0.75%, w/v) until cleared of blood. Livers were delicately cut into ca. 1-mm? pieces [8], placed into 35-ml plastic centrifuge tubes containing glucose- free or glucose-containing Hanks solution (137 mM NaCl, 5.4mM KCl, 4.0mM NaHCOs, 1.7 mM CaCh, 0.8 mM MgSOg,, 0.5 mM KH>PO,, 0.3 mM Na,HPO,, 10mM HEPES, pH 7.6, contain- ing 0.24% (w/v) bovine serum albumin and 5.55 mM glucose, the normal level of plasma glucose), and preincubated in darkness with a gyratory shaker (150 rpm) under 100% Oy at 14°C to stabi- lize tissue pieces. After 15 min, the pieces were centrifuged (270 x g for 5 min at 14°C) and washed three times by resuspension-centrifugation. For nutritional state experiments, liver pieces (4-6 pieces per well, ca. 15 mg fresh weight; represent- ing ca. 70 ug protein/mg fresh weight) from ani- mals of each nutritional regime were transferred to each well of 24-well plastic culture plates (Falcon 3047) containing 1ml (per well) of glucose- containing Hanks medium, into which hormone was previously dissolved. Hormone treatments consisted of either GLU (bovine/porcine, Sigma G4250), INS (bovine, Sigma 15500), or a combina- tion of GLU and INS. Hormones were added to a final concentration of 10~° M; this concentration was found to be maximally effective at stimulating trout hepatic lipolysis [4]. Mammalian hormones were chosen because of the scarcity of salmonid peptides and because of the similarity in function between mammalian and salmon hormones in sal- monid systems [cf. 3]. In hormone combination treatments, INS was added alone for 5 min prior to addition of INS and GLU together; addition of INS plus GLU-containing medium initiated the incubation (time=0). Incubation (under gyration at 100 rpm) proceeded for a period of 5 hr under 100% O, at 14°C. Preliminary studies indicated that maximum effects were observed at Shr of incubation. For glucose experiments, liver pieces (4 to 6 pieces, ca. 15mg fresh weight) from continuously fed animals were transferred to each well of multi-well plates containing 1 ml (per well) of Hanks glucose-containing or glucose-free medium. Treatments (control, single and com- bination hormone additions) and incubation were conducted as described for nutritional state experi- ments. In all experiments, tissue and medium samples were frozen and stored at —90°C for later analyses (usually within two weeks). Biochemical analyses Prior to fatty acid and glycerol analysis, samples were deproteinated by heat treatment (65 C for 10 min) followed by centrifugation (16,000 x g for 10 min). Medium fatty acid (FA) levels were meas- ured by the micromethod on Noma et al. [9]. Glycerol release was measured by the method of Worthington [10]. Protein was determined by the dye-binding method [11] using a Bio-Rad (Rich- mond, CA) microplate reader (cf., Bio-Rad Tech- nical Bulletin 1177). Lipolytic capacity was as- sessed by measuring the triacylglycerol (TG) lipase activity in partially purified preparations as de- Nutrition-Hormone Interactions in Trout Liver PH, scribed by Sheridan et al. [12]. Statistics Results are expressed as means+SEM. Statis- tical differences were estimated by analysis of variance; multiple comparisons among means were made by the Student-Newman-Keuls test; alpha was set at 0.05. Effects of nutritional state on hepatic lipolysis The effects of nutritional state on hepatic lipo- lysis were assessed by measuring the (TG) lipase activity of liver removed from fed and fasted (4 and 6 weeks) fish and cultured for 5 hr (Fig. 1). RESULTS liver from 4-week fasted animals. Glucagon stimulated a significant (P<0.05) in- crease in lipase activity in liver isolated from fed fish and in liver removed from fish fasted for 4 weeks. Glucagon-stimulated lipolysis was more pronounced in liver removed from 4-week fasted fish than in liver removed from fed animals (Fig. 1). Glucagon failed to affect hepatic lipolysis in liver isolated from 6-week fasted animals. The presence of INS within the culture medium caused a decrease (P< 0.05) in hepatic lipase activ- ity in the 4-week fasted fish, whereas there appeared to be no affect of INS on liver removed from other groups (Fig. 1). With the combination of INS plus GLU, INS inhibited glucagon-stimulated lipolysis in liver iso- lated from fed fish as well as from fish fasted 4 weeks (Fig. 1). Liver removed from fish that were fasted showed a significant (P< 0.05) increase in lipolytic activity as compared to that observed in liver removed from fed animals. Basal hepatic lipolysis in liver from 6-week fasted fish was not as pronounced as that in Fic. 1. IVity (nmol FA released/h/mg protein) Lijoase Ac Fed Effects of glucose on hepatic lipolysis The effects of glucose presence on hepatic lipolysis were determined in liver removed from fed fish and cultured for 5hr (Table 1). Basal MH Control [_] Glucagon AJ Insulin BY Insulin + Glucagon LA AA] SHH) XO ‘es SS Q OJ LA A 7 LX > SOx LY, oN 3 $< J oe, 7 x $5 SA0792 KOK KS <<) oe, OD SRK <7 6S S24 Ms > a Fasted (6 wks) Effects of glucagon and insulin on triacylglycerol lipase activity in liver removed from rainbow trout of varying nutritional states and cultured in vitro for 5 hr in glucose-containing (5.5 mM) medium. Liver pieces were cultured in medium containing no hormones, insulin (10~°M), glucagon (10~°M), or insulin (10 ° M) plus glucagon (10 °M). Data presented as means+SEM (n=12). * Significantly different from control, P<0.05. 278 J. S. HARMON AND M. A. SHERIDAN hepatic lipolysis, as assessed by measuring TG lipase activity, was enhanced (P<0.05) in liver incubated in glucose-containing medium over that in liver cultured in glucose-deficient medium. Glu- cagon significantly stimulated (P<0.05) hepatic lipolysis. Notably, glucagon-stimulated lipolysis was more pronounced in liver incubated in the presence of glucose (Table1). Insulin alone appeared to have no affect on basal hepatic lipo- lysis. In combination, INS inhibited glucagon- stimulated lipolysis in liver incubated both in the presence and absence of glucose (Table 1). Lipolysis was also reflected in the concentrations of medium metabolites. Liver was removed from fed fish and incubated for Shr in glucose- containing or glucose-deficient medium, and the concentrations of FA and glycerol were deter- mined (Fig. 2). Glucagon stimulated a significant increase (P<0.05) in FA and glycerol release. This increase was further enhanced by the pre- sence of glucose. The presence of INS or INS plus GLU indicated no significant differences in FA 100 * ern 75 > ) = LE ee 50 & ‘op * ~~ = a 06, 0 C C | Glucose + Fig. 2: I+G C G | TABLE 1. Effects of glucagon and insulin on triacylglycerol lipase activity in rainbow trout liver® Treatment Glucose + Glucose — Control 0.41 +0.07 0.24+0.05 Glucagon 1.98 +0.08° 0.84+0.11° Insulin 0.32 +0.04 0.28+0.06 Insulin + Glucagon 0.46+0.06 0.34+0.09 “ Liver tissue removed from fed fish and cultured in modified Hanks medium for 5hr. Enzyme activity expressed as nmol fatty acid released/h/mg protein. Data presented as means+SEM (n=12). »-< Significantly different from respective control (P <0.05); b significantly different from c (P<0.05). and glycerol release between the glucose- containing and glucose-deficient medium. DISCUSSION The results of the present study indicate that previous nutritional state and glucose presence MM Fatty Acid Release L] Glycerol Release EEG Glucose — Effects of glucagon and insulin on fatty acid and glycerol release from rainbow trout liver removed from fed fish and cultured in vitro in the presence or absence of glucose (5.5 mM) for 5 hr. Liver pieces were cultured in medium containing no hormones (control, C), 10~° M insulin (I), 10~° M glucagon (G), or 10° M insulin plus 10-°M glucagon (I+G). Data presented as means +SEM (n=12). * Significantly different from control, P< 0.05. Nutrition-Hormone Interactions in Trout Liver 279 influence both basal and GLU-mediated hepatic lipolysis in rainbow trout. Fasting resulted in elevated hepatic lipolytic activity. This observation is consistent with our previous finding that hepatic lipolysis is more pronounced in liver removed from fish fasted for 4 weeks than in liver removed from fed fish [4]. In our present experiments, we found for the first time that an extended fast (6 weeks) did not result in lipolytic rates as high as those seen in tissue from shorter-term (4-week) fasted animals. This may result from exhaustion of lipid reserves during fasting. Lipids have been shown to be depleted from coho salmon liver during a 4-week fast [5] and during smoltification and seawater adaptation [13]. Alterations in lipase activity during fasting of coho salmon were correlated with decreases in INS/GLU and INS/GLP ratios in plasma [5]. Fasting-associated alterations in the profile of pan- creatic hormones were also reported in trout [6]. Such hormonal patterns may influence the acti- vation state of the lipase enzyme. Schwartz and Jungas [14] suggested that fasting-associated lipolysis results from inhibition of the lipase inacti- vating system. Glucagon stimulated lipolytic activity in liver removed from fed fish and from fish fasted 4 weeks. While GLU is known to promote lipolysis in fed fish [4], we report for the first time that GLU-stimulated lipolysis is more pronounced in tissue removed from 4-week fasted fish than in tissue removed from fed animals. Fasting has also been reported to sensitize adipocytes to nor-— epinephrine-stimulated lipolysis in rats [15]. No- tably, GLU failed to enhance hepatic lipolysis in trout liver removed from animals fasted for 6 weeks. The basis of this observation is not known, although it may result from the already advanced state of lipid depletion. Similar observations were made with hepatic glycogenolysis in chinook salmon [8]. In these experiments, GLU stimulated glycogenolysis in liver from fed animals but failed to promote glycogenolysis in liver from longer- term fasted fish. Other factors may also influence reduced GLU sensitivity. “Refractoriness” to se- quential lipolytic stimulus has been reported in rat adipose tissue [16] and in various tissues of salmon after smoltification [13]. During fasting, the rela- tive abundance of GLU in the plasma of fish was higher than in fed animals [5, 6]; these altered hormonal patterns may alter receptor characteris- tics. In our present experiments we observed that glucose appeared to stimulate hepatic lipolysis. This observation is consistent with our previous findings that basal hepatic lipolysis is enhanced in the presence of glucose [4]. Glucose has been reported to increase basal lipolysis in adipose tissue isolated from mammals as well [17]. It is interesting to note that lipolysis in trout liver incubated in the presence of glucose results in the release of FA and glycerol in less than the ex- pected 3:1 ratio. This suggests that while lipolysis is proceeding at higher rates, the extent of reesteri- fication is also elevated. A similar situation was observed in mammalian adipose tissue [17]. In vivo administration of glucose to trout results in elevated plasma FA supported by increased he- patic lipolysis [7]. These metabolic alterations were associated with reduced plasma levels of somatostatin and GLU—a pattern similar to that seen in diabetic humans [18]. In normal humans, glucose infusions lowered plasma FA via insulin- stimulated uptake [19] and resulted in increased rates of FA reesterification and reduced lipid mobilization from adipose tissue [20]. Wolfe and Peters [20] concluded, however, that glucose in- creased TG-FA cycling—a contention supported by our present experiments. We also report for the first time that GLU stimulates lipolysis in liver incubated in the pre- sence of glucose to a greater extent than liver incubated in the absence of glucose. This is evidenced by alterations in tissue lipase activity as well as by increased medium FA and glycerol release. Hormone-stimulated lipolysis, as indi- cated by glycerol release, also appeared to be enhanced in epinephrine- [17] and norepinephrine- treated [21] adipose tissue isolated from the rat. Insulin inhibited glucagon-stimulated lipolysis in liver from fed fish and in liver from fish fasted for 4 weeks. Insulin appears to exert a net antilipolytic action in fish fasted for 4 weeks. The antilipolytic action of INS in mammals is well known [18]. We have also previously reported an antilipolytic ac- tion of INS in liver removed from fed trout [4]. 280 J. S. HARMON AND M. A. SHERIDAN The relationship between the antilipolytic action of INS and nutritional state is complex and has not been well addressed in fish. Emdin [22] reported that INS injection did not result in a difference in lipid metabolism between fed and fasted hagfish. Insulin also inhibited glucagon-stimulated lipolysis in trout liver incubated in the presence and absence of glucose. The inhibitory action of INS on glucagon-stimulated lipolysis in the pre- sence of glucose confirms our previous observa- tions on INS action [4]. While glucose presence enhanced FA reesterification in trout liver, INS appeared not to appreciably affect this process. In rat adipose tissue, however, INS strongly in- creased the level of reesterification [17]. These authors also have shown that the combined effects on INS and epinephrine in the presence of glucose resulted in substantially more reesterification than in the absence of glucose [17]. Dominguez and Herrera [23] found that INS altered glycerol uti- lization in adipose tissue and suggested that these effects were altered by the presence of glucose. The precise manner by which INS affects lipolysis and reesterification remains to be determined. In summary, the results of this study advance our fundamental understanding of the control of lipid metabolism in fish in two ways: 1) the lipolytic action of GLU is enhanced during shorter-term fasting, and 2) the lipolytic action of GLU is enhanced in the presence of glucose. The role of GLU in regulating lipid metabolism in fish has been somewhat confusing, since some reports indi- cate that GLU has lipolytic effects while other reports maintain that GLU has no lipolytic action [2]. Recently, Sheridan suggested that the basis of this conflict may lie in differences in fish life history patterns [24]. This contention is supported by the report that the in vivo effects of GLU in salmon were seasonally dependent [3]. The present findings suggest that nutritional state and glucose charge, underlied by hormonal status, also must be considered when evaluating GLU action in fish. ACKNOWLEDGMENTS We are grateful to the Garrison National Fish Hatch- ery (Fish and Wildlife Service, U.S. Department of the Interior) and the North Dakota State Game and Fish Department for supplying experimental fish. We also thank Carmen Eilertson, Kim Michelsen, and Pam O’Connor for their technical assistance. This work was supported by the U.S. Department of Education Ronald E. McNair Post-Baccalaureate Achievement Program and grants from the National Science Foundation, U.S.A. (RII 8610675, BBS 8704115, and DCB 8901380 to M.A.S.). REFERENCES 1 Harmon, J. S., Michelsen, K. G. and Sheridan, M. A. (1991) Purification and characterization of hepa- tic triacylglycerol lipase isolated from rainbow trout, Oncorhynchus mykiss. Fish Physiol. Biochem., 9: 361-368. 2 Sheridan, M. A. (1988) Lipid dynamics in fish: aspects of absorption, transportation, deposition and mobilization. Comp. Biochem. Physiol., 90B: 679-690. 3 Plisetskaya, E. M., Ottolenghi, C., Sheridan, M. A., Mommsen, T. P. and Gorbman, A. (1989) Metabolic effects of salmon glucagon and glucagon- like peptide in coho and chinook salmon. Gen. Comp. Endocrinol., 73: 205-216. 4 Harmon, J. S. and Sheridan, M. A. (1992) Effects of nutritional state, insulin and glucagon on lipid mobilization in rainbow trout, Oncorhynchus. mykiss. Gen. Comp. Endocrinol., in press. — 5 Sheridan, M. A. and Mommsen, T. P. (1991) Effects of nutritional state on in vivo lipid and carbohydrate metabolism of coho _ salmon, Oncorhynchus kisutch. Gen. Comp. Endocrinol., 81: 473-483. 6 Moon, T. W., Foster, G. D. and Plisetskaya, E. M. (1989) Changes in peptide hormones and liver enzymes in the rainbow trout deprived of food for six week. Can. J. Zool., 67: 2189-2193. 7 Harmon, J. S., Eilertson, C. D., Sheridan, M. A. and Plisetskaya, E. M. (1991) Insulin suppression is associated with hypersomatostatinemia and hyper- glucagonemia in glucose-injected rainbow trout. Am. J. Physiol., 261: R609-R613. 8 Klee, M., Eilertson, C. and Sheridan, M. A. (1990) Nutritional state modulates hormone-mediated hepatic glycogenolysis in chinook salmon (Oncorhyn- chus tshawytscha). J. Exp. Zool., 254: 202-206. 9 Noma, A., Okaba, H. and Kita, M. (1973) A new colorimetric micro-determination of free fatty acids in serum. Clin. Chim. Acta., 43: 317-320. 10 Worthington Biochemical Corporation (1988) Glycerol dehydrogenase. Jn “Worthington Manual: Enzymes and related biochemicals”. Ed. by C. C. Worthington, Worthington Biochemical Corp., Freehold, New Jersey, pp. 179-180. 11 Bradford, M. M. (1976) A rapid sensitive method 13 14 15 16 17 Nutrition-Hormone Interactions in Trout Liver for the quantitation of microgram quantities of protein utilizing the principle of protein-dye bind- ing. Anal. Biochem., 72: 248-254. Sheridan, M. A., Woo, N. Y. S. and Bern, H. A. (1985) Changes in the rates of glycogenesis, gly- cogenolysis, lipogenesis, and lipolysis in selected tissues of the coho salmon (Oncorhynchus kisutch) associated with parr-smolt transformation. J. Exp. Zool. 236: 35-44. Sheridan, M. A. (1989) Alterations in lipid meta- bolism accompanying smoltification and seawater adaptation of salmonid fish. Aquaculture 82: 191- 203. Schwartz, J. P. and Jungas, R. L. (1971) Studies on the hormone-sensitive lipase of adipose tissue. J. Lipid Res., 12: 553-562. Chohan, P., Carpenter, C. and Saggerson, E. D. (1984) Changes in the anti-lipolytic action and binding to plasma membranes of N®-L-Phenyliso- propyladenosine in adipocytes from starved and hypothyroid rats. Biochem. J., 223: 53-59. Schimmel, R. J. (1974) Responses of adipose tissue to sequential lipolytic stimuli. Endocrinol., 94: 1372-1380. Jungas, R. L. and Ball, E. G. (1963) Studies on the metabolism of adipose tissue. XII. The effects of insulin and epinephrine on free fatty acid and glycerol production in the presence and absence of 18 19 20 7h 23 24 281 glucose. Biochem. 2: 383-388. Hazelwood, R. L. (1989) The Endocrine Pancreas, Prentice Hall, New Jersey, p. 258. Randle, P. J., Garland, P. B., Hales, C. N. and Newsholme, E. A. (1963) The glucose fatty-acid cycle: its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 1: 785- 789. Wolfe ResRerand Peters Eada (1987) Eipolytie response to glucose infusion in human subjects. Am. J. Physiol., 252: E218-E223. Knight Be sand tities Je (1973) the eiiect of glucose, insulin and noradrenaline on lipolysis, and on the concentrations of adenosine 3° :5’-cyclic monophosphate and adenosine 5 -triphosphate in adipose tissue. Biochem. J., 132: 77-82. Emdin, S. O. (1982) Effects of hagfish insulin in the Atlantic hagfish, Myxine glutinosa. The in vivo metabolism of ['*C] glucose and ['*C] leucine and studies on starvation and glucose-loading. Gen. Comp. Endocrinol. 47: 414-425. Dominguez, M. C. and Herrera, E. (1976) The effect of glucose, insulin and adrenaline on glycerol metabolism in vitro in rat adipose tisue. Biochem. J., 158: 183-190. Sheridan, M. A. (1990) Regulation of lipid metabo- lism in heterothermic vertebrates. The Physiologist 33: A-30. oe cm clpinaet gent ys ene mirsoeinep akin Ly. j ne : ean. iy ei, ae re a , ae Ne bh iritann 10 pst ie T° abontiil, oMty rs el te oe eu tit Se bogs a : ay eH be syyeeie enn eh; prisichsc oe aa. ae : fees cae AL singh bile: (Spe Capshined saad iaah ala foramina hs-tup, aitnnas aera LagRED A ee ake q re one Sui}. Hat re 7A os " ony Le ih HOT SiH, oe eae cieae ae ‘- at ook ip Ree 4 ss LOT a eau ae a Wy oe ae a ; we ~ : it fh “de i gets Rimes Sh oe hifi hae ashe aie a ae esl rue ) “5 ae fe 4 a 5 Pare Te = i 2 is baer Suk rth ican ik Persie & i= Sesh: ft 2 ; - > 4 Ln 7 she * i> a - . te Sear ig piacere 3 Ta e P ; / y Wes yy c'ke | % am! ee hah | iB AS hi Ae a Lye od | , 4 ' re 4 = Fe thet Th) a : : 1 i , i * ui * ' j e 2; C = tea y * a3 Ty y ' ‘ t a L : . <4 , i . 5 3 oy : ” x eh ae <> ‘ 1 al i n ~ a ¢ : = ¢ ok a xX s ; 1 u * y ? m4 ; Hf is me ; =. ve ba oe Le l ens j s = = a A) 1 A a it x 4 t Y . | y 1 a fe n = 7 wt 7 F e ¢ j a 5 < & Nu F ; F . Sa i t 4 2 S { ZOOLOGICAL SCIENCE 9: 283-291 (1992) Effects of Sex-Ratio (SR)-Spiroplasma Infection on Drosophila Primary Embryonic Cultured Cells and on Embryogenesis YUKIAKI Kuropa!, YUTAKA SHIMADA~, BUNGO SAKAGUCHI? and KuGao Otsu‘ ‘Research Institute of Biosciences, Azabu University, Sagamihara, Kanagawa 229, Department of Anatomy, Chiba University School of Medicine, Chiba 260, >Laboratory of Sericultural Science, Faculty of Agriculture, Kyushu University, Fukuoka 812, and *Department of Biology, Faculty of Science, Kobe University, Kobe 657, Japan ABSTRACT—Effects of Sex-ratio(SR)-spiroplasma infection in Drosophila melanogaster were ex- amined in two experiments, in vitro and in vivo. The NSRO strain of the SR-spiroplasmas, transovarially transmitted microorganisms causing male-specific lethality in embryos, was collected from adult female hemolymph and inoculated into primary embryonic cell cultures prepared 2 or 5 days earlier. Within 1-2 days necrotic cell masses were detected in some neuroblastic and fibroblastic cells, but no detectable changes were observed in muscle cells, nerve cells, or in cellular spheres under a phase-contrast microscope. Spiroplasma-like structures were observed most abundantly in intercellular spaces of necrotic cells under the electron microscope. Similar experiments with NSRO-A, a non-male killing variant derived from NSRO, produced no such necrotic changes in neuroblastic and fibroblastic cells. Effects of NSRO on embryogenesis were examined under electron microscopy on embryos produced from NSRO-infected mothers. In about one-half the cases (presumably males) of embryos, 7.5-8.5 hr after oviposition, extensive necrotic cell masses were observed at regular intervals along the ventral side. In the remaining one-half (presumably females) of the embryos and in embryos produced from non-infected mothers, necrotic cell masses at regular intervals were also observed, but in a very © 1992 Zoological Society of Japan much reduced scale, representing most probably the normal programmed cell death. INTRODUCTION The sex-ratio organisms (SROs), transovarially transmitted spiroplasmas (Order Mycoplasma- tales) [1] (originally referred to as the sex-ratio spirochetes), which infect Drosophila and cause male-specific lethality at the embryonic stages, have been studied rather extensively [review, 1]. The SROs were found initially to infect a fraction of natural populations of the four neotropical species, D. willistoni, D. nebulosa, D. equinoxialis and D. paulistorum. Since, however, these micro- organisms are found most abundantly in the hemolymph of adult females and can be transfer- red by microinjection into other Drosophila spe- cies where they establish permanent infection and Accepted November 25, 1991 Received September 20, 1991 show male-specific lethality, most studies have been carried out using D. melanogaster as a new host where various mutations and techniques to manipulate chromosomes are available. Earlier studies have established that the infec- tion of SROs results in the lethality of only the single-X individuals regardless of their phenotypic sex [2, 3, 4, 5], and the single-X diploid cells but not polyploid or polynucleated cells in primary embryonic cell cultures prepared from SRO- infected mothers [6]. Analysis of gynandromorph survivors from the SRO-infected mothers sug- gested the primary site of the lethal action of SRO included most of the primordial nervous and mesodermal tissues [7]. Only recently, however, the transmission process of the SRO into oocytes during oogenesis was described at the ultrastruc- tural level [8], and morphological features of de- velopment in embryos infected with SRO were 284 Y. KuropA, Y. SHIMADA et al. given [9]. We report here (1) that the infection of the SRO (NSRO strain) to the primary embryonic cell cultures prepared from Oregon-R strain of D. melanogaster produced characteristic necrotic cell masses in neuroblastic and fibroblast-like cells, and (2) that during embryogenesis, in 7.5-8.5 hr- embryos derived from NSRO-infected mothers some characteristic necrotic masses which distri- buted at regular intervals along the ventral side were detected. MATERIALS AND METHODS SRO strains The NSRO (male-killing) and NSRO-A (non- male-killing) strains were used. The NSRO, de- rived originally from D. nebulosa, had been trans- ferred to and maintained in Oregon-R strain of D. melanogaster (ORNSRO). The NSRO-A is a variant which appeared spontaneously among progenies of ORNSRO [10]. Fly stocks were maintained as described previously [11]. Since NSRO and NSRO-A had not been culti- vated in vitro, samples containing each of these microorganisms were prepared as_ follows. ORNSRO or ORNSRO-A females aged 7-10 days were first injected with 0.15 M NaCl solution as much as possible, and within a few min the hemolymph (SRO is found most abundantly in the adult hemolymph) was collected by using glass microinjection pipettes. Two hundred yl each of such samples, each from 300-600 females, was then diluted with the same volume of 0.15 M NaCl solution, filtered through a Millipore filter (pore size, 0.45 wm), checked for the presence of the SRO under a dark-field microscope, and then used to infect the primary embryonic cell cultures. The hemolymph of non-infected Oregon-R simi- larly prepared and checked for the absence of the SRO was also used as a control. Embryonic cell culture The primary cultures of Drosophila embryonic cells were prepared as described previously [12, 13]. Newly-laid eggs were collected from the wildtype Oregon-R strain of D. melanogaster. The eggs were dechorionated by treatment with 3% sodium hypochloride solution for 6 min, and then washed with distilled water, and transferred to physiological salt solution (0.7 g NaCl, 0.02 g KCI, 0.002 g CaCl :2H,O, 0.01 g MgCl,-6H2O, 0.05 g NaHCOs, 0.02 g NaH,PO,:2H2O and 0.08 g glu- cose in 100 ml of distilled water) and allowed to develop at 25°C. The developmental stages of dechorionated eggs were readily determined through the transparent vitelline membrane under a binocular microscope. Embryos at the postgastrula stage were selected, sterilized in 70% ethyl alcohol for 10 min, and washed three times with the sterile physiological salt solution. One hundred embryos were transfer- red to culture medium in a hollow slide and torn into small fragments with a pair of fine needles under a binocular microscope. Then, small torn fragments of tissues and cells were transferred into culture medium on a carbon-coated coverslip in a T-5 culture flask. The flasks were closed with a stopper and incubated at 25°C. The culture medium consisted of medium K-17 of Kuroda [12], supplemented with 0.1 mg/ ml fetuin (Grand Island Biol. Co., Deutsch Method) and 15% fetal bovine serum (Microbiol. Assoc. inves UL SAE Infection of SR-spiroplasmas The primary cultures of embryonic cells pre- pared as described above were incubated at 25°C. After incubation for 24 hr, almost all tissue frag- ments and cells were adhered on a glass surface of the culture flasks. Some characteristic types of cells were observed in cultures. Upon further cultivation for a few days, muscle cells, epithelial cells, fibroblastic cells, nerve cells and cellular spheres were identified by their char- acteristic morphology and movement under a binocular microscope [12, 13]. After incubation for 2 or 5 dyas at 25°C, the primary cultures of embryonic cells were infected with 100 ul of the sample of SR-spiroplasmas. The cultures were incubated for 2 days at 25°C and observed every day under a phase-contrast micro- scope and photographed. Electron microscopy Cultured cells on cover slips were prefixed in SR Spiroplasma in Drosophila 285 2.5% glutaraldehyde in 0.1 M cacodylate buffer at pH 7.3. Whole embryos, collected at intervals and incubated for various periods of time, were de- chorionated and punctured with a fine tungsten needle in a prefix solution (3% glutaraldehyde, 1% paraformaldehyde, 1% DMSO in 0.1M cacody- late buffer at pH 7.2). Both materials were then postfixed in 1% OsO, in 0.1 M cacodylate buffer at pH 7.3, dehydrated in an ascending ethanol series, and embedded in Epon 812. Semi-thin sections cut with an LKB ultrotome were stained with Toluidine blue. Thin sections were stained with uranyl acetate and lead citrate and were examined with a Hitachi H-700 or a JEM 1200 EXII electron- microscope operated at 80-175 kV. RESULTS Effects of NSRO on embryonic cells in culture When cells obtained from Drosophila embryos at the post-gastrula stage were cultured at 25°C, various types of cells differentiated after a few days. Among them, spindle-shaped muscle cells, flat polygonal epithelial cells, fibroblastic cells, baloon-like cellular spheres and nerve cells with many stretching and branching nerve fibers were conspicuous in their specific morphology and movement [12, 13]. When cultures were infected with male-killing NSRO after cultivation for 2 or 5 days and incubated for one or two more days, some brown and black necrotic cells appeared in nerve cell masses (Figs. la and 1b). Although these necrotic cells were not identified definitely, they had some characters of neuroblastic or fibro- blastic cells in the site of their presence and their morphology. Cellular spheres, many spindle shaped muscle cells and hemocytes, which were assumed to be derived from the same embryonic tissue fragments and observed around the nerve cell masses, were not affected by infection of NSROs and showed normal morphology (Figs. 1c and 1d). In the preparation of the cultures, cells from 100 embryos were co-cultured together in the same flask. Cells from male and female embryos thus contributed 50% each in culture. Many neuroblastic and fibroblastic cells showed normal morphology in the culture. It may be assumed that the normal cells are derived from female embryos, and the necrotic cells are derived from male embryos. When of Oregon-R embryonic cells were infected with non-male- killing mutant spiroplasmas, NSRO-A, necrotic cells such as those appeared in NSRO-infected cells were only very infrequenctly detected (Figs. lc and 1d). This indicates that NSRO-A spiroplas- ma had no deleterious effects on embryonic cells of Drosophila cultured in vitro. When hemolymph obtained from non-infected Oregon-R adult females was added to the primary cultures of embryonic cells, cells and tissues showed no detectable changes in their morphology upon further cultivation. The cell cultures were processed for electron microscopic examinations. Structures identical to the SRO as previously described [8] were detected in intercellular spaces of cells in the primary cul- tures infected with NSRO often in close proximity to necrotic cells (Fig. 2a). Similar structures were also detected in cultured cells infected with NSRO- A (Fig. 2b). It was noted that the density of NSRO in such intercellular spaces of cells infected with NSRO was always higher than that in cells infected with NSRO-A. It is suggested that these observations on the presence of SRO-like struc- tures may correspond to the presence of necrotic cells. the primary cultures Effects of NSRO on cells in embryogenesis NSRO-infected females from the stock cultures kept for many generations were mated to normal males and allowed to lay eggs. Eggs were collected at 60-min intervals and allowed to develop for 3.5— 4.5, 7.5-8.5 and 12-13 hr, and processed for elec- tron microscopy. Embryos at the early stage of 3.5—4.5 hr after oviposition showed no noticeable changes in cells and tissues compared with those in embryos produced from non-infected females. Some embryos, at the stage of 12-13 hr, produced badly-preserved features, while others at this stage were quite normal: the former seemed to be the NSRO-affected dying male embryos. Embryos at the stage of 7.5—-8.5 hr, on the other hand, showed some characteristic features. Three out of 8 embryos at this stage examined showed the pre- 286 Y. Kuropa, Y. SHIMADA et al. sence of many necrotic cell masses extensively spread along the ventral side at regular intervals (Fig. 3a). Three others showed similar but less extensive necrotic cell masses (Fig. 3b). In the remaining two embryos, necrotic cell masses were detected only scarecely. Since necrotic cell masses Fic. 1. detected with similar features of the second and the last classes were also observed in normal non-infected wild type embryos (Fig. 3c), they are considered to represent normal programmed cell death. Most probably, then, the first class of the necrotic cell masses may occur in NSRO-infected Phase-contrast photomicrographs of part of embryonic cells cultured for 1 day and then infected with the male-killing NSRO (a, b), and of those infected with non-male-killing NSRO-A (c, d). Arrows indicate necrotic cells in the nerve cell mass. A cellular sphere (CS) and muscle cells (M) are not affected. Scales indicate 50 um. 287 SR Spiroplasma in Drosophila if; Electron micrographs of part of embryonic cells cultured for 1 day and then infected with NSRO (a), and of those infected with NSRO-A (b). SRO-like structures (arrowheads) are seen. Scales indicate 1 ~m. Fic. 2. Y. SHIMADA et al. ’ Y. KURODA MY 20 axe “ ES Si a Ed fe Ed Sots eons SR Spiroplasma in Drosophila 289 Be igetats: SO e ; aa a aoe oh, 2 i A % a oN $e v ee Fea Fic. 3. Electron micrographs of thin sections and photomicrographs (inset) of semi-thin sections of embryos at stages of 7.5-8.5 hr after oviposition from an NSRO-infected mother fly (a, b), and from non-infected control mother fly (c). Necrotic cell masses appear dark in both light (arrowheads) and electron micrographs. Scales indicate 1 ~m in the electron micrographs and 10 «m in the light micrographs. male embryos, while the latter two classes of the necrotic cell masses may occur in the female embryos. Based on the positions and the morpho- logical features of these necrotic cells, it is sug- gested that they are neuroblastic cells. DISCUSSION It was previously established that SROs kill selectively only male embryos of Drosophila, but do not affect female embryos (for a review see [1]). However, it is not clear how the SR spiroplasma recognizes the difference in male and female embryos, and what types of cells or tissues of male embryos are affected and killed by the SR spiro- plasma. Study on the effects of SRO infection on the viability of gynandromorphs of D. melanogas- ter suggested that the primary site of action of SRO included the primordial nervous and mesodermal tissues [7]. In the electron microscopic study [8], it was found that, during oogenesis, SROs were transmit- ted into oocytes through a tunica propria, a non- cellular membrane surrounding the egg chamber, and moved toward oocytes passing through the follicle cell layer. They were incorporated into ooplasm by pinocytosis and infolded in intracellu- lar vesicles and yolk granules [8]. In the embryogenesis of the SRO-infected strain, it was found that abnormality occurred in the ventral nervous system [9]. The occurrence of necrotic, seemingly degenerating cells in cluster, was found in the ventral nervous system. In the cell cultures of single embryos obtained from SRO-infected females, it was observed, under a phase-contrast microscope, that neurons, imaginal disc cells and plasmatocyte-like cells barely differentiated [6]. In the present experi- 290 Y; Kwuropa, Y- ment, more direct evidence that the primary target cells of SROs may be nerve cells was obtained using the primary cultures of embryonic cells in- fected in vitro with the SRO. The SRO-infected cells showed necrotic changes in fragments of the nervous tissue. Other cell types such as muscle cells, epithelial cells, hemocytes and cellular spheres derived from the same embryos, were not affected by infection with SRO. Electron micro- scopic examinations confirm these observations that necrotic cells are neuroblastic cells. In the primary culture cells infected with SRO in vitro. SRO-like structures were detected in intercellular spaces under the electron microscope. The accu- mulation of SRO-like structures in intercellular spaces suggested that the target cells of SRO may produce some attracting factor(s) to SRO. Another possibility, although not mutually exclu- sive, may be that SRO has a high activity to proliferate only in such spaces under the in vitro culture conditions. It was reported that SRO showed a transient proliferation more than 100 times the initial concentration in primary Dro- sophila embryonic cell culture [14]. Accumulation of SRO in particular intercellular spaces may be one of the initial steps of in vitro proliferation of SRO. Another interesting finding in the present study is the presence of many necrotic cell masses exten- sively spread along the ventral side at regular intervals. These necrotic cells may be neuroblastic cells, on the basis of their position in the embryo and their morphology. Tsuchiyama-Omura et al. [9] reported the presence of necrotic cells in the NSRO-infected 5-6hr-old and older embryos. However, they did not observe extensive necrotic cell masses at regular intervals along the ventral side as described in this communication. The discrepancy between these observations remains to be clarified. They found necrotic cells in the disorganized mid-ventral portion of the midgut and also, in greater concentrations, inside the yolk mass of the NSRO-infected 10 hr-old embryos. It is possible that necrotic cell masses are “removed” rapidly from their original places and are “dis- carded” into the yolk mass. Slight changes in staging the embryonic age may then give quite different figures. In any case detalied electron SHIMADA et al. microscopic examinations are required and with warrant, especially since our observations also suggested the presence of normal programmed cell death. ACKNOWLEDGMENTS The majority of this work has been carried out by the Collaborative Research Program in National Institute of Genetics in Mishima. The authors wish to thank Miss Yuko Takada and Mrs. Masako Kawahara for their technical assistance during this work. REFERENCES 1 Williamson, D. L. and Poulson, D. F. (1979) Sex ratio organisms (Spiroplasmas) of Drosophila. In: “The Mycoplasmas,” Vol. Il. Ed. by R. F. Whit- comb and J. G. Tully, Academic Press, New York, pp. 175-208. 2 Sakaguchi, B. and Poulson, D. F. (1963) Inter- specific transfer of the “sex-ratio” condition from Drosophila willistoni to D. melanogaster. Genetics, 48: 841-861. 3 Miyamoto, C..and Oishi, K. (1975) e3Eitects of SR-spirochete infection on Drosophila melanogaster carrying intersex genes. Genetics, 79: 55-61. 4 Watanabe, T. K. (1975) A new sex-transforming gene in the second chromosome of Drosophila mela- nogaster. Jpn. J. Genet., 50: 269-271. 5 Fujihara, T., Kawabe, M. and Oishi, K. (1978) A sex transformation gene in Drosophila melanogas- ter. J. Heredity, 69: 229-236. 6 Koana, T. and Miyake, T. (1983) Effects of the sex ratio organism on in vitro differentiation of Droso- phila embryonic cells. Genetics, 104: 113-122. 7 Tsuchiyama, S., Sakaguchi, B. and Oishi, K. (1978) Analysis of gynandromorph survivals in Drosophila melanogaster infected with the male-killing SR organisms. Genetics, 89: 711-721.. 8 Niki, Y. (1988) Ultrastructural study of the sex ratio organisms (SRO) transmission into oocytes during oogenesis in Drosophila melanogaster. Jpn. J. Genet., 63: 11-21. 9 Tsuchiyama-Omura, S., Sakaguchi, B., Koga, K. and Poulson, D. F. (1988) Morphological features of embryogenesis in Drosophila melanogaster in- fected with a male-killing spiroplasma. Zool. Sci., 5: 375-384. 10 Yamada, M-A., Nawa, S. and Watanabe, T. K. (1982) A mutant of SR organism (SRO) in Drosophila that does not kill the host males. Jpn. J. Genet., 57: 301-305. 11 Oishi, K. (1971) Spirochaete-mediated abnormal sex-ratio (SR) condition in Drosophila: A second SR Spiroplasma in Drosophila 291 -virus associated with spirochaetes and its use in the study of the SR condition. Genet. Res., 18: 45-56. 12 Kuroda, Y. (1974a) Studies on Drosophila embry- onic cells in vitro. I. Characteristics of cell types in culture. Develop. Growth Differ., 16: 55-66. 13. Kuroda, Y. (1974b) Jn vitro activity of cells from 14 genetically lethal embryos of Drosophila. Nature, 252: 40-41. Ueda, R., Koana, T. and Miyake, T. (1987) Tran- sient proliferation of the sex ratio organisms of Drosophila in a primary cell culture from infected embryos. Jpn. J. Genet., 62: 85-93. re ir aus s RG 4 é at 4 7; ed ic: aie oan Ly ic ier ar eRe ae AP ano Mi yal ¢ & wey a ty 7 2 | \ 4 ny - ; Phu Jee i . : 4 ay 14 oe oe pe ; ve # i, = } for t ~ ©. ) on ve * i a mare 5 2 z bee 2 J yy Cl f CH RNAs ei | i ; aN a t u 2 ; ( ZOOLOGICAL SCIENCE 9: 293-304 (1992) Attempts to Improve Survival of Neurons Derived from Neonatal Rat Hypothalamus-Preoptic Area in Serum-free Media KencHt TAKAGI and SENCHIRO KAWASHIMA Zoological Institute, Faculty of Science, University of Tokyo, Tokyo 113, Japan ABSTRACT— In primary culture of cells derived from neonatal rat hypothalamus-preoptic area, a great decrease in the survival of neurons following an increase in the number of non-neuronal cells was observed in serum-supplemented medium. However, the addition of cytosine arabinoside, a mitotic inhibitor, to serum-supplemented medium significantly enhanced the survival of neurons. In order to diminish the proliferation of non-neuronal cells, attempts were made to maintain dissociated cells in serum-free medium. The maintenance of neurons in healthy condition was very difficult in serum-free medium from the very beginning of cultivation. Culture in serum-supplemented medium followed by culture in serum-free medium was much superior to the culture in serum-free medium for the entire course of cultivation with regard to survival. In the next experiment, the effects of glass surface and poly-lysine-coated surface were compared. Cultured cells in serum-free medium on glass surface were better in morphology compared with those on poly-lysine-coated surface. In the latter, strange contraction of basal cell sheets disturbed long-term cultivation. The best culture condition established in the present study, i.e., preculture in serum-supplemented medium and then transfer to serum-free medium on glass surface substrate, was employed for the study on the effects of hydrocortisone. Survival rate of neurons was improved in cultures with hydrocortisone at the concentration of 10° ' M. © 1992 Zoological Society of Japan INTRODUCTION The serum contains various factors which gener- ally afford cultured cells a favorable condition. However, the composition of serum is variable, depending on the animals species, age and sex of its donor, and even on lots [1-3]. Moreover, in primary culture of the cells from the central nerv- ous system using serum-supplemented medium, the overgrowth of non-neuronal cells makes the interpretation of experimental results difficult [4, 5]. To solve these problems, efforts have been exerted to replace serum by certain chemically defined agents [6, 7]. In the culture of neural cells, first breakthrough was made on the culture of clonal cell lines. Mather and Sato [8] tested effects of various factors on mouse melanoma M2R cul- tured in serum-free medium, and established favorable hormone supplements. Bottenstein and Accepted December 25, 1991 Received November 17, 1991 Sato [9] succeeded in culturing rat neuroblastoma line B104 in serum-free medium with supplements called “N2”. Since then, studies have been carried out on neural cell cultures in serum-free medium and many investigators have pointed out that serum-free medium is beneficial for neuronal sur- vival, but it is difficult to culture neural cells completely in serum-free medium [10, 11]. The aim of the present study was to examine the effect of inhibition of non-neuronal cell growth by the application of cytosine-(-p-arabinofuranoside (AraC), a mitotic inhbitior, on the survival of neurons, and to afford further evidence for the effects of serum, substrates and hydrocortisone supplementation for the primary culture of cells derived from neonatal rat hypothalamus-preoptic area (Hyp-POA) in order to establish a culture condition with serum-free medium. Cells derived from neonatal rat cerebral cortices were also used in some experiments for comparison. 204 K. TAKAGI AND S. KAWASHIMA MATERIALS AND METHODS Cell dissociation Neonatal rats of the Wistar/Tw strain of both sexes were used. On the day of birth animals were decapitated, and the Hyp-POA and cerebral cor- tex were immediately taken out and pooled in a petri dish containing cold serum-supplemented medium. Tissue pieces were minced by fine scis- sors, and after discarding the medium, they were incubated with 1,500 PU/ml dispase (Gohdoh- shusei Co., Tokyo) in serum-supplemented medium for 15 min at 37°C. After incubation, they were rinsed twice with fresh cold serum- supplemented medium, followed by gentle pipet- ting in serum-supplemented medium to yield dis- sociated cell suspension. Remaining tissue pieces were then rinsed twice with Ca’*, Mg’*-free Dulbecco’s phosphate buffered saline (PBS(—)) that contained 0.5% bovine serum albumin and 0.5% glucose (PBS(—)+BG). Then, they were incubated with 2mM_ glycoletherdiaminetetra- acetic acid (EGTA) in PBS(—)—BG for 15 min at 37°C and after twice washing with PBS(—)+BG, they were subjected to gentle pipetting in PBS(— ) +BG to yield cell suspension. Both dispase and EGTA treatments were repeated for a few times. Smaller diameters of pipettes were used as the size of remaining tissue pieces became smaller. The cell suspension was passed through a nylon mesh (50 «um in pore size; NBC Industry Co., Tokyo) to remove large cell clumps, and centrifuged at 250 x g for 5 min at 4°C. Cell pellets were resuspended in a fresh serum-supplemented medium and viable cells were counted by trypan blue exclusion test using a hemocytometer. Initial cell viabilities of the Hyp-POA and cerebral cortical cell suspen- sions were about 80% and 90%, respectively. The number of viable cells obtained from one pup was 4.5-6.0 10° for the Hyp-POA and 2.0-3.0x 10° for the cerebral cortex. After dilution, the cells were plated on a round coverslip (14 mm in dia- meter) in multiwell culture plate (Sumitomo Bake- lite Co., Tokyo). In some cases, the coverslip had been coated with certain substrates. Plating den- sity was 1.5 10° cells/well for the Hyp-POA and 3.0 10° cells/well for the cerebral cortex. Unless otherwise stated, culture medium was renewed at 3-day intervals. Culture media Two types of serum-supplemented media were used. The first type consisted of 85% Eagle’s minimum essential medium supplemented with 2 mg/ml sodium bicarbonate, 9.67 mg/ml glucose, 100 IU/ml penicillin, and 100 “g/ml streptomycin, and of 15% fetal bovine serum (FBS). This medium (MEM-S) was used for culture in serum- supplemented medium. Second type of serum-supplemented medium was used only for the preculture of serum-free culture. This medium (DME/F12-S) was com- posed of 15% FBS and 85% 1:1 mixture of Dulbecco’s modified Eagle’s medium and Ham’s nutrient mixture F-12 (EME/F12) supplemented with 5.4mg/ml_ glucose, 1.2mg/ml N-2- hydoxyethylpiperazine-N’-2-ethensulfonic acid (HEPES), 1001U/ml penicillin and 100 ug/ml streptomycin. For serum-free medium, the N2-supplements of Bottenstein and Sato [9] without progesterone were used, where DME/F12 with above-— mentioned supplements without FBS was further added with 100 “g/ml human transferrin, 100 u.M putrescine, 30nM selenium (as NajSeQO3) and 5 yug/ml insulin. This serum-free medium was desig- nated as DME/F12-F. Insulin, putrescine, human transferrin and DME/F12 mixture with HEPES were purchased from Sigma. MEM was purchased from Nissui Phamaceutical Co. (Tokyo), and FBS from Hazle- ton Research Products Inc. (St. Louis). Other chemicals AraC (Sigma) at the concentration of 10°°M was dissolved in MEM-S. Hydrocortisone (Sigma) was first dissolved in ethanol and diluted in DME/F12-F at the final concentrations of 10~’—10'* M in 0.007% ethanol. Substrates For poly-t-lysine coating, the method of Pett- mann et al. [12] was used with a slight modifica- tion. Poly-L-lysine solution (500 wl, 100 ug poly-L- lysine per 1 ml boric acid buffer, pH 8.4) was Improvement of Neuronal Survival 295 added on a coverslip placed in each well of culture plate for overnight at room temperature. After discarding the solution, the coverslip was rinsed twice with PBS(—). Poly-.L-lysine (M.W. 52,000) was purchased from Sigma. For serum precoating, the method of Eccleston et al. [11] was used with a slight modification. A coverslip was incubated with DME/F12-S for overnight. Before use, it was rinsed twice with PBS(—). Cell counting Cultured cells on the coverslips were fixed for 2 days in a fixative consisting of 70% ethanol, 5% neutral formaldehyde and 5% acetic acid. After about 1 week of washing in 99% ethanol, the cells were stained by a modified Bodian’s method [13]. The cells which statisfied all the three following criteria in Bodian preparations were regarded as neurons; (i) the cell with positively stained cell body, (ii) the cell with strongly stained processes, and (iii) the cell at least one of the process length was more than 3-fold longer than the diameter of No. of Neurons / Coverslip @) 1,000 500 the cell body. The number of all the neurons on a coverslip or the number of neurons in randomly chosen 100 fields under a light microscope (one field equals to 0.19 mm”) was counted. RESULTS Survival of neurons in serum-supplemented medium and effects of AraC To examine the effects of non-neuronal cell population on neuronal survival in serum- supplemented medium, the Hyp-POA cells were cultured for the first 3 days in MEM-S on poly- lysine-coated surface, then the medium was re- placed by MEM-S supplemented with or without 10-°M AraC. After culture for 3 days with or without AraC, the cultures were washed twice with Ca**, Mg**-containing Dulbecco’s phosphate buffered saline (PBS(+)), and the medium was replaced by MEM-S without AraC and the cells were continued to be cultured in MEM-S. Phase-contrast photomicrographs of control ‘ AraC 6 Days of Culture Fic. 1. Effects of AraC on Hyp-POA cell cultures for 13 days in MEM-S. A: Photomicrograph of control culture without AraC. B: Photomicrograph of AraC-added culture. Bar=100 ~m. C: The number of neurons per coverslip at 3, 6, and 9 days of culture with (solid line) and without (broken line) AraC treatment. AraC treatment was performed between 3 to 6 days of culture. Vertical bars depict the standard errors of the means (n=2—4). 296 K. TAKAGI AND S. KAWASHIMA AraC-untreated cultures show that only few neurons were present, overlying the sheet of small- er cells on the substrate (Fig. 1A). Many glial cells, apparently dead cells (round brilliant cells) and few neurons were visible on the basal cell sheet. In AraC-treated cultures, non-neuronal cell proliferation was effectively inhibited and the bas- al surface was covered with larger flat cells. Many neurons survived on the basal sheet, and they extended prominent processes forming networks (Fig. 1B). Glial cells and apparently dead cells 60 6 Fic. 2. were less than the control AraC-untreated cul- tures. Figure 1C shows the number of neurons per coverslip. In control cultures, the number of neurons greatly decreased at 6 days, while in AraC-treated cultures, the number of neurons was almost constant for 9 days, indicating that the proliferation of non-neuronal cells greatly dis- turbed the survival of neurons in serum- supplemented medium. SS Phase-contrast photomicrographs of cultured cells. A: Cells cultured for the first 2 days in DME/F12-S then 2 days in DME/D12-F. B: Cells cultured for the first 2 days in DME/F12-F then 2 days in DME/F12-F. C: Cells cultured for the first 2 days in DME/F12-F containing 0.5% BSA then 2 days in DME/F12-F. D: Cells cultured for the first 2 days in DME/F12-F on serum precoated surface then 2 days in DME/F12-F. Bar=100 um. Improvement of Neuronal Survival Za) Survival of neurons in serum-free medium and effects of substrate To maintain neuronal cells in serum-free medium throughout the culture period, some agents for cell protective and attachment- extension functions should be added to replace the serum components [14-16]. For this purpose the supplementation of 0.5% BSA to serum-free medium and serum precoating were separately tested. The cerebral cortical cells were used as test cells, and the fiber connections among aggregates were regarded as suitable indices for neuronal viability. After centrifugation of dissociated cell suspension, cell pellets were resuspended in DME/F12-S, DME/F12-F or DME/F12-F con- taining 0.5% BSA. The resuspensions in DME/ F12-S and DME/F12-F containing 0.5% BSA were cultivated on glass surface. The cell resus- pension in DME/F12-F was cultivated on glass surface or serum precoated surface. The medium of all the four groups was replaced by DME/F12-F after twice rinses with PBS(+ ) at 2 days of culture. Figure 2 shows phase-contrast photomicro- graphs of cultured cells. When cells had been cultured in DME/F12-S before transfer to DME/ F12-F, many cells survived and numerous fiber connections were formed (Fig. 2A). When they had been cultured in DME/F12-F from the very beginning of culture, there were few healthy aggre- gates, and fiber connections were not formed (Fig. 2B). Supplementation of BSA increased the sur- vival of cells, while it had no appreciable effect on the formation of fiber connections (Fig. 2C). Serum precoating increased the formation of fiber connections, and it also had a slight beneficial effect on the survival of cells (Fig. 2D). When both supplementation of BSA and serum precoat- ing were carried out, serum-free culture from the very beginning was considerably improved. However, it was still inferior to the culture with preculture for 2 days in serum-supplemented medium. Therefore, preculture in DME/F12-S for 2 days was used routinely for later experiments. To compare culture conditions in serum-free medium, the Hyp-POA and cerebral cortical cells were cultured on glass or poly-lysine-coated sur- face in DME/F12-S for 2 days. Following the preculture, the medium was replaced by DME/ F12-F. On glass surface, cultured Hyp-POA cells first formed aggregates (40-100 “m in diameter) within 3-6h (Fig.3A). During preculture in serum- supplemented medium, the cells attached to glass surface and began to grow. Some neuron-like cells were present on extended non-neuronal cells or directly on glass surface, but most of them were observed to reside in the aggregates at 2 days of culture. Although non-neuronal cells continued to proliferate in serum-free medium, the growth was slower than that in serum-supplemented medium. As a consequence, basal sheet formation was delayed and the cells organizing basal sheet were larger and flatter than the cells in serum- supplemented medium. After reaching con- fluence, the overgrowth of non-neuronal cells was not so extensive and the translocation of glial cells from inside of the sheet to free surface was sup- pressed in serum-free medium. Neuron-like cells each possessing neurites, mostly bipolar, one clear ovoid nucleus and one or two nucleoli. Although moderate neural networks were observed before the completion of basal cell sheet formation, sub- stantial network formation proceeded after the completion of basal sheet (Figs. 3C and E). The number of neurons was counted at 5, 8, 11, 14 and 17 days (Fig. 4). Because of numerous aggregates at 2 days of culture, the number of neurons was not counted. Basal cells became confluent between 8 to 11 days of culture. The number of neurons decreased at 11 days, and after that it remained unchanged. The Hyp-POA neurons could be maintained for more than 3 weeks in DME/F12-F on glass surface. On poly-lysine-coated surface, individual cells adhered to the surface isolatedly (Fig. 3B). The developing pattern of the culture was generally similar to that of the culture on glass surface for several days. However, a drastic change began to occur during 8 to 14 days on poly-lysine-coated surface. Some basal cells began to detach from the surface and showed a change in morphology, from squamous to elongated shape. As a result of such transformation, many irregular-shaped uncovered surfaces appeared (Fig. 3D). Neural networks on this transforming basal sheet eventually dis- 298 : aw : K. TAKAGI AND S. KAWASHIMA Improvement of Neuronal Survival 299 appeared and some neuron-like cells left behind on the uncovered surface could not survive any long- er. Some basal cells remaining on the uncovered surface proliferated to cover the surface again. The cerebral cortical cells formed aggregates on glass surface, as the Hyp-POA cells did. At 2 days, fiber connections were formed among closely lo- cated aggregates. Although they had been culti- vated at 2-fold higher cell density than the Hyp- POA cells, fewer cells could survive and settle during preculture. The proliferation of basal cells was very slow in serum-free medium and basal sheet failed to reach confluence even at 11 days. Few cells spread out from the aggregates of the cerebral cortical cells. Fiber connections among the aggregates increased for several days in serum- 700 (38) 600 500 No. of Neurons / 100 Fields S 8 free medium, but they gradually degenerated (Figs. 3G and H). Most striking change in cerebral cortical cell culture was the appearance of non- neuronal cells bearing several processes with numerous branchings (Fig. 3F). This type of cells resembles differentiated astrocytes induced by glial maturation factor [26]. The appearance of differentiated astrocyte-like cells was less in Hyp- POA cultures than in cerebral cortical cell cultures in serum-free medium and was never observed neither in cerebral cortical nor Hyp-POA cultures in serum-supplemented medium. On poly-lysine-coated surface, individual cer- ebral cortical cells attached to the basal surface isolatedly like the Hyp-POA cells. Basal cell proliferation was very slow, similar to that on glass (4) | ele) (1) (2) 11 14 17 Days of Culture Fic. 4. The number of neurons in 100 microscopic fields of Hyp-POA cells cultured in serum-free medium on glass surface as a function of culture age. Vertical bars depict the standard errors of the means. The values in parentheses indicate the mean numbers of aggregates in 100 microscopic fields. Each value was obtained from triplicate cultures. Fic. 3. Phase-contrast photomicrographs (A-D, F-H) or Bodian-stained preparation (E) of cultured Hyp-POA or CC cells in serum-free medium. A: Hyp-POA cells on glass surface at 3h of culture. B: Hyp-POA cells on poly-lysine-coated surface at 3h of culture. C: Hyp-POA cells on grass surface at 14 days of culture. D: Hyp-POA cells on poly-lysine-coated surface at 14 days of culture. E: Bodian-stained preparation of Hyp-POA cells on glass surface at 17 days of culture. F: Cerebral cortical cells on glass surface at 11 days of culture, showing differentiated astrocyte-like cells. G: Cerebral cortical cell aggregates at 11 days of culture. H: Cerebral cortical cells on glass surface at 14 days of culture. Note the degeneration of fiber connections between two aggregates. Bar in H=100 «wm applies to A-D and F-H. Bar in E=50 um. 300 K. TAKAGI AND S. KAWASHIMA surface. In some regions of a coverslip, surviving neuron-like cells developed very fine networks which were similar to those observed by Romjin et al. [17] in their fetal rat cerebral cortical cell cultures in serum-free medium. These neuron-like cells began to degenerate between 8 and 11 days of culture. Differentiated astrocyte-like cells were encountered as in the cultures on glass surface. Effects of hydrocortisone on neuronal survival in serum-free medium The Hyp-POA cells were precultured for 2 days in serum-supplemented medium (DME/F12-S) on glass surface. After the preculture, the medium was replaced by DME/F12-F containing hydrocor- tisone at the concentration of 0 (control), NOW 105 Om 10m 10 or Om Mein 00072 vehicle ethanol. Cultures were maintained for 14 days and the cells on coverslips were fixed and 500 *% 400 So a ic S 300 a WY js = 200 ® zZ O re) 100 Z 0 Bassman oe ie z O21 50 WO 1 IO ao A Hydrocortisone (M) Fic. 5. stained. As the concentration of hydrocortisone increased, the density of basal cells decreased, and basal cells became more fibrous in shape (Fig. 5C). At a concentration greater than 10”? M, strange cell clumps which were brilliant with smooth out- lines under the phase-contrast microscope, were observed at 8-14 days (Fig. 5B). These clumps were in most cases constricted and drifted out into the medium. Although such clumps were also encountered in control cultures without hydrocor- tisone, they were less in number and never drifted out. The number of neurons gradually increased by hydrocortisone treatment up to the concentra- tion of 107!" M (at 107 !° M, 134.5% of the control value). At a concentration greater than 10~°’ M, the number decreased (Fig. 5A). The concentra- tion that effectively decreased the number of neurons was the same as that where the formation of cell clumps gegan to increase. Effects of hydrocortisone on the number of neurons (A) and the morphology of Hyp-POA cells (B and C) at 14 days of culture in serum-free medium. A: The number of neurons in 100 microscopic fields. Vertical bars depict the standard errors of the means (n=3). Significance of differences: (by analysis of variance) hydrocortisone-treated groups vs. control, * P<0.05, ** P<0.01, *** P<0.001. B: Phase-contrast photomicro- graph of floating clumps treated with 10’ M hydrocortisons at 11 days of culture. C: Bodian-stained preparation of cells treated with 10°’ M hydrocortisone at 14 days of culture. Note more numerous fibrous basal cells as compared to the cells in Fig. 3E. Bars=50 um. Improvement of Neuronal Survival 301 DISCUSSION The increase of non-neuronal cell proliferation in serum-supplemented medium apparently dis- turbed the survival of neurons of the Hyp-POA. Godfrey et al. [4] reported that no electrical excit- ability or synaptic activity was elicited by brain cells in culture without treatment on non-neuronal cell proliferation. Such effect of non-neuronal cells appears to occur at relatively high plating density [18]. Most of the neurons which survived for 3 days were well maintained for the rest of the culture period and formed fine networks if cul- tured in serum-supplemented medium with AraC, which acts to inhibit the proliferation of non- neuronal cells, in the present study. Recently, it was reported that AraC killed cultured neurons derived from sympathetic ganglia by evoking ac- tive death processes intrinsic to the neurons [19, 21]. In addition, Brochovsky and Bradford [21] have studied whether or not AraC has any toxicity on postmitotic cells. They found that prolonged treatment with 10° M AraC prevented the surviv- al of either neurons or astrocytes in the culture of whole brain cells from fetal rats, and stated that tetanus toxin-labeled cells, GFAP-positive cells and dopamine release all vanished after 14 days of culture. Hayashi and Patel [22] also reported that prolonged exposure to 10-° M AraC was toxic to both the neurons and glial cells grown in chemical- ly-defined medium. However, in the present study, 3-day exposure to 10° M AraC caused no remarkable death of neurons at later periods of culture. In additon to a low survival rate of neurons, serum-supplemented medium has some other problems. Honn et al. [3] reported that FBS contained 53+19ng/dl (mean+S.D.) testoster- one and 9.6+2.7 ug/ml hydrocortisone, in addi- tion to some amounts of T,, GH and insulin. To exclude these effects of FBS, serum-free culture has been attempted by a number of investigators as in the present study. At first, the culture of cells in serum-free medium from the very beginning was examined. DME/F12-F alone was unable to maintain the cells. Supplementation of BSA and/or precoating coverslips with FBS considerably improved the survival in DME/F12-F. However, the improve- ment was still inferior to the culture with serum- free culture after preculture in serum- supplemented medium. As regard to the role of BSA, Yamane [14, 15] has already indicated that it acts to protect the cells from exogenous and/or endogenous lytic factors, in addition to a role as a carrier of fatty acids and other metabolites. The present result that serum precoating helped both cell survival and fiber connection formation is in good accord with the finding of Faivre-Bauman et al. [23] that serum precoating was necessary for serum-free culture of the cells from mouse embryonic hypothalamus. Fibronectin has often been used as an attachment factor for neuronal cell cultures. However, Faivre-Bauman et al. [23] stated that so-called ‘cold-insoluble globulin’ (identical with fibronectin)-coating was less effec- tive than serum precoating, suggesting that there should be other attachment factors in FBS. Hay- man et al. [24] reported that the major attachment factor in FBS is vitronectin. Facilitation of cell attachment by serum procoating in the present study might be induced by vitronectin present in FBS. Effects of substrates were examined in the pre- sent study. On glass surface, the cells of the Hyp-POA during preculture in serum-supple- mented medium first formed aggregates. The aggregates gradually spread out in serum-free medium, as Faivre-Bauman et al. [23] found in cell culture of the fetal mouse hypothalamus. Non- neuronal cell proliferation was considerably sup- pressed and the number of neurons was almost constant during 11 to 17 days of culture. After 8 days, neural networks were observed. The cells could be maintained on glass surface in serum-free medium for at least 3 weeks. On poly-lysine- coated surface, the cells attached to the surface isolatedly, and after about 8 days of culture once- formed basal cell sheet contracted. The contrac- tion might be induced if poly-lysine is degraded as time proceeds, because cell adhesiveness to the surface might become weaker than the inter- or intracellular tension, resulting in the detachment of the cells from the surface and induction of contraction. In fact, adhesiveness of the cells on poly-lysine-coated surface was very weak at 8 days, since basal cells were easily detached by pipettings. 302 K. TAKAGI AND S. KAWASHIMA Because of such basal cell contraction, poly-lysine- coated surface does not seem to be a suitable substrate for a long-term culture of the Hyp-POA cells in serum-free medium. The morphology of cultured cerebral cortical cells was much different from that of the Hyp- POA cells. In cerebral cortical cell cultures, ageregate formation during preculture period was more apparent, and spreading out of non-neuronal cells from the aggregates was less obvious. The difference in aggregate formation was probably due to the difference in the adhesiveness of cels between the cerebral cortical and Hyp-POA cells. Less obvious spreading out of non-neuronal cells in cerebral cortical cell cultures might be due to the difference in the composition of cells of the aggre- gates. Differentiated astrocyte-like cells were more numerous in cerebral cortical cell cultures than in Hyp-POA cell cultures, suggesting Hyp- POA cell cultures contained more numerous un- differentiated cells which had higher migration activity. The regional difference in the astrocyte maturation was also reported by other investiga- tors [25]. The cerebral cortical cells were not maintained for a long period either on glass or poly-lysine- coated surface. Some factors were probably insuf- ficient or lacking in DME/F12-F for cerebral cor- tical cell cultures. Romijn et al. [17] proposed a serum-free medium for the cerebral cortical cells, consisting of 3:1 mixture of DME and F12 sup- plemented with double cocktail dose of N2 supple- ments, 0.1% BSA, 20ng/ml T3 and 0.2 ug/ml corticosterone. Some colonies of differentiated astrocyte-like cells were observed after 8 days in the present study. The shape of these cells was similar to that induced by glial maturation factor [26] or with dibutyryl-cycic AMP [27]. The beneficial role of corticoids suggested by Romijn et al. [17] was tested in the culture condi- tion established in the present study. Hydrocorti- sone diminished proliferation of basal cells, and at higher doses (10-° and 10~’M) it induced mor- phological changes of the basal cells. It is well known that corticosteroids affect glial cells. Faroo- qui et al. [28] reported that hydrocortisone induced morphological changes of both rat glioma cell line C6 and normal hamster glial cell line NN cells and activated arylsulfatase and (-galactosidase, which are concerned with myelinogenesis at the concen- tration of 7.6X10-°M. Montiel et al. [29] re- ported that 50 nM dexamethazone inhibited prolif- eration of C6 cells and stimulated activities of glycerol phosphate dehydrogenase and lactate de- hydrogenase in C6 cells. Therefore, hydrocorti- sone might have acted on some aspects of glial differentiation. Strange clumps were formed by hydrocortisone, but the cellular constituent of such clumps was not examined in the present study. The number of neurons increased as the concen- tration of hydrocortisone increased up to 107 !° M, and at still higher concentrations it steadily de- creased. Since hydrocortisone apparently affected non-neuronal cells, further study is needed for the analysis of direct effects of hydrocortisone on the survival of neurons. At any rate, the culture condition established in the present study may be useful for the investigation of other hormonal effects on the Hyp-POA. Im summary, the culture in serum-free medium of the cells derived from the Hyp-POA in the present study showed that for the better survival of neurons preculture in serum-supplemented medium was necessary and glass surface was a better substrate than poly-lysine-coated surface, and that hydrocortisone at certain concentrations was beneficial for neuronal survival. ACKNOWLEDGMENT This study was supported in part by a Grant-in-Aid from the Ministry of Education, Science and Culture, Japan to S. Kawashima (No. 02404007). REFERENCES 1 Olmsted, C. A. (1967) A physico-chemical study of fetal calf serum used as a tissue culture nutrient correlated with biological tests for toxicity. Exp. Cell. Res., 48: 283-299. 2 Esber, H. J., Payne, I. J. and Bogden, A. E. (1973) Variability of hormone concentrations and ratio in commercial sera used for tissue culture. J. Natl. Cancer Inst., 50: 559-562. 3 Honn, K. V., Singley, J. A. and Chavin, W. (1975) Fetal bovine serum: Multivariate standard. Proc. Soc. Exp. Biol. Med., 149: 344-347. 4 Godfrey, E., Nelson, P., Schrier, B., Breuer, A. 10 11 Wy 13 14 tS 16 Improvement of Neuronal Survival 303 and Ransom, B. (1975) Neurons from fetal rat brain in a new culture system: Multidisciplinary analysis. Brain Res., 90: 1-21. | Nia ©. Hertz, Eo and Hentzavice (1984) Alterations in uptake and release rates for GABA, glutamate, and glutamine during biochemical maturation of highly purified cultures of cerebral cortical neurons, a GABAergic preparation. J. Neurochem., 42: 951-960. Barnes, D. and Sato, G. (1980) Methods for growth of cultured cell in serum-free medium. Analytical Biochem., 102: 255-270. Waymouth, C. (1984) Preparation and use of serum-free culture media. In “Method for Prepara- tion of Media, Supplements and Substrata for Serum-Free Animal Cell Culture.” Ed. by D. W. Barnes, D. A. Sirbasku and G. H. Sato, Alan R. Liss, Inc., New York, pp. 23-68. Mather, J. and Sato, G. (1977) Hormones and growth facotrs in cell culture: Problems and perspec- tives. In “Cell, Tissue and Organ Culture in Neuro- biolosyea. eds by S, Eedorott and 2. Hertz, Academic Press, New York, pp. 619-630. Bottenstein, J. E. and Sato, G. H. (1979) Growth of a rat neuroblastoma cell line in serum-free sup- plemented medium. Proc. Natl. Acad. Sci. USA, 76: 514-517. Skaper, S. D., Adler, R. and Varon, S. (1979) A procedure for purifying neuron-like cells in cultures from central nervous tissue with a defined medium. Dev. Neurosci., 2: 233-237. Eccleston, P. A., Gunton, D. J. and Silberberg, D. H. (1985) Requirement for brain cell attachment, survival and growth in serum-free medium: Effects of extracellular matrix, epidermal growth factor and fibroblast growth factor. Dev. Neurosci., 7: 308- Sy, Pettmann, B., Louis, J. C. and Sensenbrenner, M. (1979) Morphological and biochemical maturation of neurones cultured in the absence of glial cells. Nature, 281: 378-380. Otsuka, N. (1976) Otsuka’s modified Bodian’s sil- ver-impregnation method. In “Histological Techni- ques—Theoretical and Applied.” Ed. by Y. Sano, Nanzando, Tokyo, pp. 358-361 (in Japanese). Yamane, I. (1975) Role of bovine serum albumin in a serum-free suspension of cell culture. Proc. Soc. Exp. Biol. Medl., 149: 439-442. Yamane, I. (1978) Role of bovine serum albumin in a serum-free culture medium and its application. Natl. Cancer. Inst. Monogr., 48: 131-133. Rogers, S. L., Letournearn, P. C., Peterson, B. A., Furcut, L. T. and McCarthy, J. B. (1987) Selective interaction of peripheral and central nervous system cells with two distinct cell-binding domains of fibronectin. J. Cell Biol., 105: 1435-1442. 7/ 18 19 20 21 Mp I} 24 a5) 26 i 28 Romijn, H. J., Habets, A. M. M. C., Mud, M. T. and Wolters, P. S. (1982) Nerve outgrowth, synap- togenesis and bioelectric activity in fetal rat cerebral cortex tissue cultured in serum-free, chemically de- fined medium. Dev. Brain Res., 2: 583-589. Dichter, M. A. (1978) Rat cortical neurons in cell culture: Culture methods, cell morphology, elec- trophysiology, and synapse formation. Brain Res., 149: 279-293. Wallace, T. L. and Johnson, Jr., E. M. (1989) Cytosine arabinoside kills postmitotic neurons: Evi- dence that deoxycytidine may have a role in neuron- al survival that is independent of DNA synthesis. J. Neurosci., 9: 115-124. Martin, D. P., Wallace, T. L. and Johnson, Jr., E. M. (1990) Cytosine arabinoside kills postmitotic neurons in a fashion resembling trophic factor de- privation: Evidence that a deoxycytidine-dependent process may be required for nerve growth factor signal transduction. J. Neurosci., 10: 184-193. Brochocovsky, O. and Bradford, H. F. (1987) De- velopment of transmitter-releasing capacity in neuron-enriched tissue cultures. J. Neurochem., 48: 787-797. Hayashi, M. and Patel, A. J. (1987) An interaction between thyroid hormone and nerve growth factor in the regulation of choline acethyltransferase activ- ity in neuronal cultures, derived from the septal- diagonal band region of the embryonic rat brain. Dev. Brain Res., 36: 109-120. Faivre-Bauman, A., Rosenbaum, E., Puymirat, J., Grouselle, D. and Tixier-Vidal, A. (1981) Dif- ferentiation of fetal mouse hypothalamic cells in serum-free medium. Dev. Neurosci., 4: 118-129. Hayman, E. G., Pierschbacher, M. D., Suzuki, S. and Ruoslahti, E. (1985) Vitronectin—a major cell attachment-promoting protein in fetal bovine serum. Exp. Cell Res., 160: 245-258. Bevery Ce Epps baskassbeng. JeaniINeIsent i) and Pilgrim, C. (1990) Region- and sex-related differ- ences in maturation of astrocytes in dissociated cell cultures of embryonic rat brain. Glia, 3: 55-64. Lit, IN. Wuvaaisi, ID. 12.5 IO, we Se ane! Ketro, 10 (1977) Differentiation of glioblasts under the in- fluence of glial maturation factor. In “Cell, Tissue and Organ Culture in Neurobiology.” Ed. by S. Fedoroff and L. Hertz, Academic Press, New York, pp. 223-235. Brunner, G., Lang, K., Wolfe, R. A., McClure, D. B. and Sato, G. H. (1982) Selective cell culture of brain cells by serum-free, hormone-supplemented media: A comparative morphological study. Dev. Brain Res., 2: 563-575. Farooqui, A. A., Elkouby, A. and Mandel, P. (1977) Effects of hydrocortisone and thyroxine on arylsulphatase A and B of cultured cells of neuronal 304 K. TAKAGI AND S. KAWASHIMA and glial orgin. J. Neurochem., 29: 365-369. and cytosolic glycerol phosphate dehydrogenase, 29 Montiel, F., Sarli¢ve, L., Pascual, A. and Aranda, lactate dehydrogenase and malic enzyme in glial cell A. (1986) Multihormonal control of proliferation in culture. Neurochem. Int., 9: 247-253. ZOOLOGICAL SCIENCE 9: 305-314 (1992) Development of an in situ Hybridization Histochemistry for Choline Acetyltransferase mRNA with RNA Probes Tomoyuki IcHIKAWA! and Kyoxo AJIKI Department of Anatomy and Embryology, Tokyo Metropolitan Institute for Neuroscience, Fuchu, Tokyo 183, Japan ABSTRACT— An mn situ hybridization histochemistry (ISHH) for choline acetyltransferase (ChAT) in paraffin sections of the central nervous system of the rat was developed using *°S-labeled RNA probes by defining the technical parameters enabling optimal detection of ChAT mRNA. Fixation by perfusion with 4% paraformaldehyde and 0.4% glutaraldehyde provided the most intense signals. Slides for mounting paraffin sections were coated with 1% BSA and fixed with 25% glutaraldehyde. Proteinase K treatment of tissue sections after fixation increased intensity of signals. Decreasing probe length increased intensity of signals; truncation to approximately 75 nucleotides by alkaline hydrolysis provided the best result. Ribonuclease A treatment after hybridization appeared to be essential to reduce nonspecific background signals. Specificity of hybridization signals was confirmed by the selective localization of signals in motoneurons in the spinal cord with an antisense probe and no signal in them with a sense probe. Distribution patterns of perikarya of neurons in the forebrain, cranial nerve motor nuclei and cervical region of the spinal cord revealed by ISHH and immunohistochemistry (IHC) were almost superimposable, giving additional evidence for specificity of the present method. However, intensity of signals for ChAT mRNA detected by ISHH and cellular content of ChAT protein revealed © 1992 Zoological Society of Japan by IHC were not always parallel. INTRODUCTION Considerable progress has been made in eluci- dating the organization of the central cholinergic system by immunohistochemistry (IHC) using monoclonal antibodies to choline acetyltransferase (ChAT), the most specific marker for cholinergic neurons [1]. Compared to the abundance of information on cholinergic neurons in the adult brain, information on the developing cholinergic system is limited [2-5]. This may be due in part to the lack of monoclonal antibodies applicable to cryostat or paraffin sections and difficulties of working with free-floating sections of delicate embryonic and early postnatal brain tissues. By introduction of molecular biological techniques, two methods are available to overcome the dis- advantage of the monoclonal antibody. One method involves the expression of protein from a cloned gene introduced into Escherichia Accepted February 19, 1992 Received October 5, 1991 ‘ To whom all correspondence should be addressed. coli and the production of antisera against the protein. We have succeeded in expressing large amounts of rat ChAT protein using rat ChAT cDNA ligated to a translation vector, and in producing an antiserum. This antiserum is highly specific to rat ChAT, immunochemically and im- munohistochemically, and stains not only peri- karya and dendrites but also axons and terminals of cholinergic neurons in cryostat sections [6]. The second method involves an in situ hybridiza- tion histochemistry (ISHH), which enables the precise localization and identification of individual cells in which a particular gene is transcribed. Among probes in current use, RNA probes offer a unique combination of advantages for ISHH: (1) antisense probes contain only the antisense strand and no sense strand to compete in solution for hybridization with target mRNAs, resulting in much higher signals, (2) control probes are easily prepared, (3) fragment length of probes can be reproducibly controlled by limited alkaline hy- drolysis [7], providing good penetration of probes into tissue, (4) RNA-RNA hybrids have higher stability compared to DNA-RNA hybrids, allow- 306 T. ICHIKAWA AND K. AJIKI ing use of more stringent washing conditions to reduce nonspecific binding of probes to tissue, and (5) nonspecific background signals, in the form of probe sticking to tissue, can be reduced by post- hybridization digestion with ribonuclease (RNase) [7, 8]. In the present study, we have defined the tech- nical parameters and established procedure for ISHH for rat ChAT mRNA in paraffin sections with RNA probes. Further, distribution and way of labeling of perikarya of neurons in the fore- brain, cranial nerve motor nuclei and cervical region of the spinal cord of the rat revealed by ISHH and IHC were compared. A preliminary study of ISHH for rat ChAT mRNA with RNA probes has been reported [9]. MATERIALS AND METHODS Preparation of RNA probes ~ The cDNA encoding rat ChAT [9] was ligated into EcoRI site of plasmid Bluescript (Stratagene). Orientation was confirmed by restriction mapping. After linearization with restriction endonucleases, templates were treated with proteinase K (Boe- hringer-Mannheim) and extracted with phenol/ chloroform/isoamyl alcohol (25:24:1) (Phenol/ CIA). To produce antisense and sense RNA probes, they were transcribed by T7 and T3 RNA polymerases (Stratagene), respectively, using an [a-*°S]JUTP (400 Ci/mmole, Du Pont-New Eng- land Nuclear) to a specific activity of 4-5 x10’ dpm/g according to the manufacturer’s protocol. Templates were removed with deoxyribonuclease I (Promega), and RNA transcripts were extracted with Phenol/CIA and precipitated with ethanol in the presence of ammonium acetate twice. Both probes were truncated to approximately 75, 150 or 300 nucleotides by limited alkaline hydrolysis [7]. Mass average size of the probe was checked by electrophoresis on a 2% agarose gel containing formaldehyde followed by autoradiography [10]. Tissure preparation Sixteen male Sprague-Dawley rats (200-300 g) were used. Two rats were decapitated without anesthesia and brains and spinal cords were rapidly dissected out, sliced at about 3 mm in thickness and immersed overnight in 4% paraformaldehyde (PA) in 0.1 M sodium phosphate (pH 7.4) (PB) at 4°C. On the next day, tissues were transferred to 70% ethanol and kept overnight at 4°C. Other rats were deeply anesthetized with sodium pentobar- bital and perfused through the aortic cone with 0.1 M PB containing 0.9% NaCl at room temperature and at a flow rate of 20 ml/min for 4-6 min, followed by either 4% PA in 0.1 M PB (2 rats) or 4% PA and 0.4% glutaraldehyde (GA) in 0.1M PB (12 rats) at 4°C and at the same flow rate for 8- 12min. Brains and spinal cords were dissected out, sliced at about 3 mm in thickness, postfixed in the same fixative at 4°C for 2 hr and kept overnight in 70% ethanol at 4°C. Tissues were dehydrated through graded ethanol series and embedded in paraffin (melting point 51-53°C, Merck). Sections were cut serially at 10 ~m and mounted on the coated slides described below. An appropriate section was hybridized with the antisense probe and the next section with the sense probe. Slide preparation Clean slides were dipped in 1% BSA for 5 min and dried at 60°C. Then, they were fixed with 25% GA for 3 min, washed in autoclaved H,O twice and dried at 60°C. They were used within 2 weeks. In situ hybridization Deparaffinized sections were either treated with proteinase K (1 and 100 ug/ml in 0.1 M Tris-HCl, pH 7.5, containing 50 mM EDTA, at 37°C for 30 min and at room temperature for 10 min, respec- tively) or only washed in 0.1 M Tris-HCl (pH 7.5), and dehydrated. Some sections treated with or without proteinase K were further subjected to acetic anhydride treatment [8], and dehydrated. °S-labeled probes were diluted to 1, 2 or 410° dpm/vl in a solution containing 20 mM Tris-HCl (pH 8.0), 0.3 M NaCl, 50% freshly deionized form- amide, 0.5mg/ml E. coli tRNA, 10 mM dithio- threitol, 2.5mM EDTA, 0.02% BSA, 0.02% Ficoll, 0.02% polyvinylpyrrolidone and 10% de- xtran sulfate, and pipetted directly onto tissue sections (5 wl/cm*). Sections were coverslipped and incubated at 50°C for 16 hr in a moist cham- ber. Hybridized slides were then washed in 2 Detection of ChAT mRNA with RNA Probes 307 SSC (1XSSC is 150mM NaCl, 15 mM sodium citrate, pH 7.0) containing 50% formamide and 0.1% -mercaptoethanol (BME) at 50°C for 1 hr, treated with RNase A (20 #g/ml in 10 mM Tris- HCl, pH 8.0, containing 0.5 M NaCl, Boehringer- Mannheim) at 37 C for 30 min, rinsed sequentially in 2XSSC containing 50% formamide and 0.1% BME at 50°C for 1 hr, in 1x SSC containing 50% formamide and 0.1% BME at 50°C for 2 hr and finally in 1xSSC containing 50% formamide at room temperature for 15 min, and dehydrated. Some slides were not treated with RNase A. Slides were dipped in NR-M2 emulsion (Konica), ex- posed at 4°C for 7-14 days and developed in Konicadol X. Appropriate sections were counter- stained with hematoxylin. Immunohistochemistry Three male Sprague-Dawley rats (200-300 g) were used. Procedure for the tissue preparation was identical to that described earlier [11]. Brains and spinal cords were sectioned on a freezing microtome at 50 um in the transverse plane. The monoclonal antibody to rat ChAT has been char- acterized in detail elsewhere [12, 13]. Procedure for IHC using ABC Kit (Vector) was identical to those described previously [11, 14]. RESULTS Specificity of hybridization signals The antisense RNA probe revealed ChAT trans- cripts exclusively in motoneurons in the medial and lateral nuclei in the lamina IX of the cervical region of the spinal cord of the rat (Fig. 1A, C). No signal was observed in the section hybridized with the sense probe (Fig. 1B). Signals for ChAT mRNA were mostly restricted to perikarya of motoneurons (Fig. 1C). These results indicate that the present ISHH is highly specific. Fic. 1. ISHH of the cervical region of the spinal cord (11 days exposure). The tissue was fixed with 4% PA and 0.4% GA followed by proteinase K treat- ment. Probes were truncated to 75 nucleotides and applied at a concentration of 210° dpm/l. A and B. Dark-field photomicrographs of adjacent 10-~m paraffin sections hybridized with the antisense (A) and sense (B) probes (x30). C. Bright-field photo- micrograph of motoneurons in the section hybri- dized with the antisense probe. No counterstaining (x80). 308 T. ICHIKAWA AND K. AJIKI Technical considerations The results obtained with the antisense probe in sections of the cervical spinal cord demonstrated the critical influence of the fixation procedure on Fic. 2. Effects of fixation procedure on ISHH (14 days exposure). All sections of the ventral horn of the cervical region of the spinal cord were hybridized with the antisense probe (truncated to 75 nuc- leotides and at a concentration of 210° dpm/yl). Dark-field photomicrographs (45). A. Fixed by perfusion with 4% PA and 0.4% GA followed by proteinase K treatment. B. Fixed by perfusion with 4% PA without proteinase K treatment. C. Fixed by immersion in 4% PA without proteinase K treatment. intensity of signals (Fig. 2). The highest signals for ChAT mRNA were obtained with the tissue fixed by perfusion with 4% PA and 0.4% GA followed by proteinase K treatment (Fig. 2A). Under these conditions, background signals were very low. Weaker signals were also observed in the tissue fixed by perfusion with 4% PA without proteinase K treatment (Fig. 2B), but no signal was observed when the tissue was fixed by immersion in 4% PA with or without proteinase K treatment (Fig. 2C). Treatment of proteinase K at a concentration of 100 ug/ml at room temperature for 10 min was better than that of 1 ~g/ml at 37°C for 30 min. When the fixative containing GA was used, pro- teinase K treatment appeared to be essential, but treatment of the enzyme to the tissue fixed only with 4% PA decreased intensity of signals signi- ficantly. Treatment with acetic anhydride did not reduce background signals. The results obtained with the antisense probe in sections of the basal forebrain indicated that de- creasing of. probe length increased intensity of signals. The highest intensity of signals was observed when the antisense probe was truncated to approximately 75 nucleotides. The probe at a concentration of 210° dpm/wl was the best in terms of signal-to-noise ratios among concentra- tions checked. The results obtained with the sense probe in sections of the neocortex and caudate-putamen exhibited that RNase A treatment after hybridiza- tion reduced nonspecific background signals dra- matically (Fig. 3). Hybridization signals were clearly detected after exposure for 7 days, and exposure for 11 days appeared to be optimal in terms of signal-to-noise ratios. Comparison of ISHH with IHC For ISHH, tissues were fixed by perfusion with 4% PA and 0.4% GA. Sections were treated with proteinase K (100 4g/ml) at room temperature for 10 min. Probes were truncated to approximately 75 nucleotides and applied to sections at a concen- tration of 2x 10° dpm/l. Autoradiographic expo- sure was 11 days. Under these optimal conditions, distribution and intensity of signals of labeled neurons in the forebrain, cranial nerve motor Detection of ChAT mRNA with RNA Probes 309 Fic. 3. Effect of RNase A treatment after hybridization on ISHH (11 days exposure). The tissue was fixed with 4% PA and 0.4% GA. Serial sections contain- ing the neocortex and caudate-putamen were hybri- dized with the sense probe (truncated to 75 nuc- leotides and at a concentration of 210° dpm/ vl), and untreated (A) or treated (B) with RNase A. Dark-field photomicrographs (x 100). nuclei and cervical region of the spinal cord did not differ appreciably in each region among 6 rats used. On the other hand, distribution of ChAT- immunoreactive neurons in these areas revealed by the present IHC was in agreement with those in previous immunohistochemical studies using mon- oclonal antibodies to rat ChAT [1, 11, 14, 15]. Immunoreactivity of labeled neurons in each re- gion was consistent among 3 rats used. These observations made it possible to compare distribu- Fic. 4. ISHH and IHC of the basal forebrain. A and B. Dark-field photomicrographs of adjacent 10-4m pa- raffin sections hybridized with the sense (A) and antisense (B) probes (x20). C. IHC of ChAT in 50-m frozen section of the similar area to A and B (x20). Note compatible distributions of labeled neurons detected by both methods. No signal was observed in the section hybridized with the sense probe. ac, anterior commisure; cp, caudate-puta- men; db, diagonal band of Broca; ms, medial septal nucleus; na, nucleus accumbens; ot, olfactory tubercle. 310 T. ICHIKAWA AND K. AJIKI tion and way of labeling of perikarya of neurons detected by ISHH and IHC in different rats. In the basal forebrain, labeled neurons by ISHH with the antisense probe were present in the olfactory tubercle (Fig. 4B), medial septal nucleus (Fig. 4B), diagonal band of Broca (Fig. 4B), sub- stantia innominata and magnocellular preoptic nucleus. No signal was observed in the adjacent section hybridized with the sense probe (Fig. 4A). In these areas, distribution pattern of ChAT- immunoreactive neurons was compatible with that of hybridization-positive neurons (Fig. 4B, C). Hybridization signals and immunoreactivity of neurons revealed by the two methods were both intense. In the neocortex, no cell was detected by ISHH, although weak ChAT-immunoreactive neurons B Fic.5. ISHH and IHC of the medial habenular nu- cleus. A. Dark-field photomicrograph of the section hybridized with the antisense probe (x20). B. IHC of ChAT (X20). were present in layers II-VI. In the caudate- putamen and nucleus accumbens, neurons de- tected by both methods showed similar topography (Fig. 4B, C). These neurons were strongly im- munoreactive, but showed weak hybridization sig- nals. In the globus pallidus, strongly ChAT- immunoreactive neurons were present, but no hybridization signal was observed. In the medial habenular nucleus, all neurons were positive by Fic. 6. ISHH and IHC of the motor nucleus of cranial nerve VII. A. Dark-field photomicrograph of the section hybridized with the antisense probe (x 30). B. IHC of ChAT (X30). Note that the smaller neurons near the nucleus (arrow) were also detected by both methods. Detection of ChAT mRNA with RNA Probes Biel both methods (Fig.5). They exhibited intense hybridization signals while their immunoreactivity was weak. : In the brainstem, motoneurons in the motor nuclei of cranial nerves III-VI, VII (Fig. 6), X (Fig. 7) and XII (Fig. 7) were labeled by both methods. Hybridization signals and immunoreac- tivity of neurons in the motor nuclei of cranial nerves III, IV and VI revealed by both methods were moderate, while those in the motor nuclei of cranial nerves V, VII and XII were intense. In neurons in the motor nucleus of cranial nerve X, hybridization signals were intense, but im- Ae. mS : . MS ¥ oy . i . Ny a B Fic. 7. ISHH and IHC of the motor nuclei of cranial nerves X (X) and XII (XII). A. Dark-field photo- micrograph of the section hybridized with the anti- sense probe (X20). B. IHC of ChAT (x20). Note that neurons in the nucleus ambiguus were not detected by ISHH, but detected by IHC (arrow). munoreactivity was weak (Fig. 7). In the nucleus ambiguus, neurons were strongly immunoreactive, but they showed no hybridization signal (Fig. 7). In the cervical region of the spinal cord, motoneurons were labeled intensely by both methods (Fig. 1). Thus, distribution patterns of perikarya of neurons revealed by ISHH and IHC were almost superimposable, but intensity of hy- bridization signals did not necessarily correlate with that of immunoreactivity. DISCUSSION ISHH has recently become a widely used proce- dure for detection and localization of mRNAs in tissue sections [16, 17]. RNA probes have several advantages over other probes as described in In- troduction. Cryostat sections are widely used for ISHH and we have experienced that cryostat sec- tions gave slightly higher hybridization signals than paraffin sections. However, some problems inher- ent to such use have become progressively appar- ent. One of these is the difficulty in obtaining good details of tissue morphology, as a result of cryostat section thickness and quality. Another major difficulty is the progressive loss of mRNA within sections after prolonged storage [18]. Par- affin sections may be stored for up to 18 months without apparent loss of hybridization signals [19]. In the present study, therefore, we established ISHH for ChAT mRNA in paraffin sections using RNA probes. Specificity of the present method was confirmed by demonstrating that the antisense probe re- vealed ChAT transcripts in motoneurons in the spinal cord, while the sense probe exhibited no hybridization signals. In addition, compatible dis- tributions of neurons in most areas in the forebrain and cranial nerve motor nuclei revealed by ISHH and IHC indicated that the present method was highly specific. So far, there are two reports on ISHH for ChAT mRNA using *°S-labeled oligo- nucleotide probes in cryostat sections of the fore- brain of the rat [20] and of the brainstem of the rat and guinea pig [21]. We demonstrated that use of full iength of the RNA probe, which was truncated to approximately 75 nucleotides after labeling, gave much higher intensity of signals and lower SZ T. ICHIKAWA AND K. AQJIKI background signals than use of the oligonucleotide probe. Higher intensity of signals may be due to higher specific activity of the RNA probe. Efficiency of the present ISHH depended on the fixation procedure. Fixation by perfusion with 4% PA and 0.4% GA in combination with proteinase K treatment appeared to be the most favorable fixative. Using this protocol, we have succeeded in detecting mRNA for Ca** /calmodulin-dependent protein kinase II in the rat brain (T. Ichikawa, S. Ohsako and T. Yamauchi, unpublished). We have also detected mRNA _ for hydroxyindole O- methyltransferase in the bovine epithalamus fixed by immersion in 4% PA and 0.1% GA in combina- tion with proteinase K treatment [22]. It should be noted, however, that mRNA for arylamine N- acetyltransferase in the chicken kidney was de- tected by our protocol only when the tissue was fixed by immersion in 4% PA [23], indicating that the appropriate fixation procedure might differ among tissues used. | Considerable loss of sections or parts of sections from slides during the hybridization procedure was encountered using slides coated by a variety of procedures, including gelatin [24], egg white, “His- tostik” [25] or polylysine [26]. Using slides coated with 1% BSA and fixed with 25% GA, we have overcome this problem. Probes whose mass average size is about 150 nucleotides are empirically used [8]. In the present study, more intense signals were observed when the probe was truncated to approximately 75 nu- cleotides. However, further truncation might cause a significant reduction in hybrid melting temperature (Tm), or much variation in Tm as a function of difference in fragment length. In addition, we have experienced that further trunca- tion decreased specificity and stability of probes. Intensity of hybridization signals for ChAT mRNA correlated with immunoreactivity for ChAT in neurons in the basal forebrain, motor nuclei of cranial nerves III-VII and XII and spinal cord. In contrast, neurons in the medial habenular nucleus and motor nucleus of cranial nerve X exhibited intense hybridization signals but weak immunoreactivity, while neurons in the caudate- putamen and nucleus accumbens showed weak hybridization signals but strong immunoreactivity. In addition, neurons in the neocortex, globus pallidus and nucleus ambiguus were not detected by ISHH, although ChAT-immunoreactive neu- rons were present there. Of necessity, comparison of intensity of hybridization signals and im- munoreactivity in neurons detected by ISHH and IHC was performed in different rats in the present study. The absence of a constant correlation between hybridization signal abundance and im- munoreactivity in the same neuron has been re- ported [27-29]. Discrepancies between amounts of mRNA and cellular contents of its translated protein could be due to a number of factors. Prominent among these is presence of endogenous factors that control translation or posttranslational events [30, 31]. Other possibilities include the relative sensitivities of ISHH and IHC. Amounts of mRNA in neurons in the neocortex, globus pallidus and nucleus ambiguus may be below the sensitivity of the present ISHH. Establishment of ISHH for ChAT mRNA in paraffin sections may be of particular use to study the embryonic and early postnatal development of cholinergic neurons. Because the time or site of synthesis of mRNA may differ from the time or site of accumulation of its translated protein [32], studies using ISHH in combination with IHC may provide more precise information on the develop- ing cholinergic neurons. ACKNOWLEDGMENTS We thank Dr. Takeo Deguchi for providing rat ChAT cDNA. This work was supported in part by Grant-in-Aid from the Ministry of Education, Science and Culture of Japan (No. 01570036) and Japan Science Foundation. REFERENCES 1 Salvaterra, P. M. and Vaughn, J. E. (1989) Regula- tion of choline acetyltransferase. Int. Rev. Neuro- biol., 31: 81-143. 2 Armstrong, D. M., Bruce, G., Hersh, L. B. and Gage, F. H. (1987) Development of cholinergic neurons in the septal/diagonal band complex of the rat. Dev. Brain Res., 36: 249-256. 3 Phelps;'P: E:, Barber, R. P.{ Houser) G) Re @raw- ford, G. D., Salvaterra, P. M. and Vaughn, J. E. (1984) Postnatal development of neurons containing 10 11 12 13 i4 15 Detection of ChAT mRNA with RNA Probes 313 ‘choline acetyltransferase in rat spinal cord: An immunocytochemical study. J. Comp. Neurol., 229: 347-361. nelpseebe see barber, Re Ps Brennante ly Ae Maines, V. M., Salvaterra, P. M. and Vaughn, J. E. (1990) Embryonic development of four different subsets of cholinergic neurons in rat cervical spinal cord. J. Comp. Neurol., 291: 9-26. Schambra, U. B., Sulik, K. K., Petrusz, P. and Lauder, J. M. (1989) Ontogeny of cholinergic neurons in the mouse forebrain. J. Comp. Neurol., 288: 101-122. Ichikawa, T., Ohsako, S. and Deguchi, T. (1991) Production of an antiserum using a fusion protein produced by a cDNA for rat choline acetyltrans- ferase. Neurosci. Lett., 131: 213-216. Coxmke lH... Deleon, Di V., Angerer, L. M. and Angerer, R. C. (1984) Detection of mRNAs in sea urchin embryos by in situ hybridization using asym- metric RNA probes. Dev. Biol., 101: 485-502. Angerer, L. M., Stoler, M. H. and Angerer, R. C. (1987) In situ hybridization with RNA probes: An annotated recipe. In “Jn Situ Hybridization. Ap- plications to Neurobiology”. Ed. by K. L. Valentino, J. H. Eberwine and J. D. Barchas, Oxford Univ. Press, Oxford, pp. 42-70. Ishii, K., Oda, Y., Ichikawa, T. and Deguchi, T. (1990) Complementary DNAs for choline acetyl- transferase from spinal cords of rat and mouse: nucleotide sequences, expression in mammalian cells, and in situ hybridization. Mol. Brain Res., 7: 151-159. Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Lab. Press, NY, 2nd ed., pp. 7.43- 7.45. Ichikawa, T. and Hirata, Y. (1990) Organization of choline acetyltransferase-containing structures in the cranial nerve motor nuclei and lamina IX of the cervical spinal cord of the rat. J. Hirnforsch., 31: 251-257. Ichikawa, T., Ishida, I. and Deguchi, T. (1983) Monoclonal antibodies to choline acetyltransferase of rat brain. FEBS Lett., 155: 306-310. Ishida, I., Ichikawa, T. and Deguchi, T. (1983) Immunochemical and immunohistochemical studies on the specificity of a monoclonal antibody to cho- line acetyltransferase of rat brain. Neurosci. Lett., 42: 267-271. Ichikawa, T. and Hirata, Y. (1986) Organization of choline acetyltransferase-containing structures in the forebrain of the rat. J. Neurosci., 6: 281-292. Houser, C. R., Crawford, G. D., Barber, R. P., Salvaterra, P. M. and Vaughn, J. E. (1983) Orga- nization and morphological characteristics of cho- linergic neurons: an immunocytochemical study with 16 7) 18 iW 20 21 2. 23 24 25 26 Zi a monoclonal antibody to choline acetyltransferase. Brain Res., 266: 97-119. Valentino, K. L., Eberwine, J. H. and Barchas, J. D. (1987) In Situ Hybridization. Applications to Neurobiology. Oxford Univ. Press, Oxford, 242 pp. Polak, J. M. and McGee, J. O’D. (1990) Jn Situ Hybridization. Principles and Practice. Oxford Univ. Press, Oxford, 247 pp. Akam, M. E. (1983) The location of Ultrabithorax transcripts in Drosophila tissue sections. EMBO J., 2: 2075-2084. Hyodo, S., Fujiwara, M., Kozono, S., Sato, M. and Urano, A. (1988) Development of an in situ hybri- dization method for neurohypophysial hormone mRNAs _ using synthetic oligonucleotide probes. Zool. Sci., 5: 397-406. Higgins, G. A., Koh, S., Chen, K. S. and Gage, F. H. (1989) NGF induction of NGF receptor gene expression and cholinergic neuronal hypertrophy within the basal forebrain of the adult rat. Neuron, 3: 247-256. Cortes; VR eAnvidssons, Us, Schalling, ME; Ceccatelli, S. and Hokfelt, T. (1990) Jn situ hybri- dization studies on mRNAs for cholecystokinin, calcitonin gene-related peptide and choline acety- Itransferase in the lower brain stem, spinal cord and dorsal root ganglia of rat and guinea pig with special reference to motoneurons. J. Chem. Neuroanat., 3: 467-485. Sato, T., Deguchi, T., Ichikawa, T., Fujieda, H. and Wake, K. (1991) Localization of hydroxyindole O-methyltransferase-synthesizing cells in bovine epithalamus: immunocytochemistry and in-situ hyb- ridization. Cell Tissue Res., 263: 413-418. Ohtomi, M., Ichikawa, T. and Deguchi, T. (1990) Arylamine N-acetyltransferase from the chicken kidney: Characterization of the enzyme and localiza- tion of the mRNA by zn situ hybridization. Biomed. Res., 11: 263-267. Gall, J. G. and Pardue, M. L. (1971) Nucleic acid hybridization in cytological preparations. Methods Enzymol., 21: 470-480. Brigati, D. J., Myerson, D., Leary, J. J., Spalholz, B., Travis, S. Z., Fong, C. K. Y., Hsiung, G. D. and Ward, D. C. (1983) Detection of viral genomes in cultured cells and paraffin-embedded tissue sections using biotin-labeled hybridization probes. Virology, 126: 32-50. McClay, D. R., Wessel, G. M. and Marchase, R. B. (1981) Intercellular recognition: Quantitation of initial binding events. Proc. Natl. Acad. Sci. USA, 78: 4975-4979. Nojiri, H., Sato, M. and Urano, A. (1986) Increase in the vasopressin MRNA levels in the magnocellu- lar neurosecretory neurons of water-deprived rats: In situ hybridization study with the use of synthetic 28 29 314 oligonucleotide probe. Zool. Sci., 3: 345-350. Guitteny, A.-F., Bohlen, P. and Bloch, B. (1988) Analysis of vasopressin gene expression by in situ hybridization and immunohistochemistry in semi- thin sections. J. Histochem. Cytochem., 36: 1373- 1378. Ichikawa, T., Ishida, I., Ohsako, S. and Deguchi, T. (1988) Jn situ hybridization demonstrating coexpres- sion of urotensins I, II-a, and II-y in the caudal neurosecretory neurons of the carp, Cyprinus car- pio. Gen. Comp. Endocrinol., 71: 493-501. T. IcHIKAWA AND K. AJIKI 30 Darnell, J. E. (1982) Variety in the level of gene 31 32 control in eukaryotic cells. Nature (Lond.), 297: 365-371. Hentze, M. W., Rouault, T. A., Harford, J. B. and Klausner, R. D. (1989) Oxidation-reduction and the molecular mechanism of a regulatory RNA-protein interaction. Science, 244: 357-359. McAllister, L. B., Scheller, R. H., Kandel, E. R. and Axel, R. (1983) In situ hybridization to study the origin and fate of identified neurons. Science, 222: 800-808. ZOOLOGICAL SCIENCE 9: 315-320 (1992) Hepatic Fatty Acids in Wild Rockhopper (Eudyptes crestatus) and Magellanic (Spheniscus magellanicus) Penguins before and after Moulting LAURENCE S. Harsice!, KEB GHEBREMESKEL~, GLYNNE WILLIAMS and MICHAEL A. CRAWFORD~ ‘Institute of Zoology, Zoological Society of London, Regents Park, London NWI 4RY, UK., and *The Institute of Brain Chemistry and Human Nutrition Hackney Hospital, Homerton High Street, London E9 6BE, UK. ABSTRACT — Intra- and inter-species hepatic differences for wild rockhopper (Eudyptes crestatus) and magellanic (Spheniscus magellanicus) penguin fatty acids were compared both pre-and post-moult. Linoleic (18:2n-6) and arachidonic (20:4n-6) acid composition were significantly higher and palmitic (16:0) acid significantly lower in pre moult rockhopper penguins than in comparable magellanics. Post-moult magellanics had significantly more palmitoleic (16:1), gadoleic (20:1n-9) and erucic, (22:1n-9) and less arachidonic and eicosapentaenoic (20:5n-3) percent fatty acids than post-moult rockhoppers. In both species moulting resulted in a significant reduction in eicosapentaenoic and docosapentaenoic (22:5n-3), and a significant increase in linoleic acid (18:2n-6) percent. In rockhoppers, post moult was associated with an increase in the proportion of palmitic (16:0) and a decrease in palmitoleic (16:1) acid. In the post-moult magellanics, however, there was a decrease in © 1992 Zoological Society of Japan stearic (18:0) and an increase in gadoleic (20: 1n-9) and erucic (22: 1n-9) fatty acid composition. INTRODUCTION Successful breeding, migration, and moulting in birds are closely linked to nutritional factors [1-4]. Scarcity of food and thus poor nutrient deposition prior to these events can result in failure to breed, poor health and mortality [1-6], especially in species such as pengiuns which abstain from feed- ing while breeding and moulting. In penguins moulting lasts between two and five weeks depend- ing on species, during which time body weight losses of 23-60% have been recorded [7-11]. Before breeding and moulting penguins build up body reserves mainly in the form of lipid. These adaptive responses result in body weight increases of 5-33% depending on species and sex [10-12]. Lipids have diverse biological roles; neutral Accepted October 29, 1991 Received July 17, 1991 ' Present Address: The Institute of Brain Chemistry and Human Nutrition Hackney Hospital, Homerton High street, London E9 6BE, UK. lipids provide important energy reserves, whilst phospholipids have membrane structural functions [13-15]. The polyunsaturated fatty acid compo- nents of phospholipids provide substrates for the cell regulatory molecules the eicosanoids, [14] and are thought to provide structural integrity to cell membranes [16-19]. In view of the importance of lipids, we have investigated the fatty acid composition of hepatic tissue from wild rockhopper (Eudyptes crestatus) and magellanic (Spheniscus magellanicus) pen- guins both pre-and post-moult. MATERIALS AND METHODS Penguins Liver tissue samples were obtained at necropsy from adult healthy free-living rockhopper and magellanic penguins inhabiting the Falkland Is- lands during February 1987 after a post-breeding feeding period. This was undertaken as a result of 316 a penguin mortality investigation in the Falkland Islands in 1986 [20]. It was not possible to differ- entiate between adults and subadults even when taking into consideration the appearence of the gonads [11]. The period between the arrival on the mouiting area and the beginning of the old feather loss was classified as pre-moult. Whereas, the penguins that had replaced their old feathers with new plumage were considered to be post-moult. Lipid extraction Total lipids were extracted from liver samples by the method of Folch et al, [21]. Tissues were homogenised in chloroform: methanol (2:1 v/v) containing 0.01% 2,6-di-t-butyl-4-methylphenol (BHT) as an antioxidant and left for 24 hr at 4°C. The homogenate-solvent mixture was filtered and transferred to a separatory funnel and left over- night at 4°C following the addition of 25% saline (0.85% NaCl) by volume. The lower organic phase was evaporated in a Rotavap-R (Buchi) under reduced pressure at 37°C. Samples were kept under nitrogen during and after the extraction TABLE 1. L. S. HarBiGeE, K. GHEBREMESKEL et al. procedures and extracts stored at —20°C until required. Fatty acid separation and identification Total lipids were transmethylated under ni- trogen at 70°C for 3 hr with 5 ml of 5% sulphuric acid in methanol as an esterifying reagent. The fatty acid methyl ester derivatives were separated and identified as previously described [22] except that the chromatograph was a Varian modei 3700, and the column a CP Sil88 (SP 2340). Statistical analysis Data are expressed as means and standard de- viations with their maximum and minimum ranges. Interspecies mean differences and pre-and post- moult means were compared by Student’s un- paired t-test. RESULTS The hepatic fatty acid composition of rockhop- per and magellanic penguins pre-and post-moult Range and mean+SD percent liver fatty acids (16:0—22:6n-3) in wild pre-and post-moult rockhopper (Eudyptes crestatus) and magellanic (Spheniscus magellanicus) penguins. Rockhopper Magellanic Fatty acid Pre-moult Post-moult Pre-moult Post-moult (n=4) (n=4) (m3) (n=4) 16:0 15.2-19.9 17.9 DARS=2 4a 22.8 DHEOEY Onl 24.8 202-24 22.9 Se 2), Il ae Ili! Se 25) +1.8 IG) 2 Al s= dll 2.8) On lee 9 Les 2,1 1.9 il J 33.2! 2.6 se i.) se (2) se), 2 ae Iti 18:0 1 ZA 20.4 18.0-23.6 20.8 18.6—22.8 20.6 1 2FAS Wine 15.8 se Il {0 +2.9 322, SEL 4! 18: 1n-9 I G=25)58) 21.4 165219" 18.1 18.0-22.4 AV.3) les 20.9 SE 28) se Ili se D2 +3.0 18 :2n-6 1 3= ZY 1.6 S.0= Cll il 0.9- 1.2 1.0 3.9- 4.4 4.0 se(0),3 aE il2 a=) 2 Se (013) 20 : 1n-9 = 2.7 7 0.9- 1.9 4 O.5= 1.0 O.7/ Ye 5). Il 4.1 +0.9 (n=) SE()).5) 32 ()).3 +1.0 20 : 4n-6 ase 8)e0) 8.6 6.8- 9.8 BS: A\ Al J/(0) 5.33 AY jl 753 6.3 +0.8 a ILO se il 4 ales 20 : 5n-3 8.6-10.6 9.3 SyW= We 4 8.5-— 8.9 8.8 2:9= 32h 3.4 se Il) se 11.5) se ()).2 se ()).3) MP2 SSIES) die 3 I DES O/= 1.33 8 MV= 2.3 Dm t= 1.2 ie se ()).7/ +0.4 se (0).2 Se ()).1I 22 :6n-3 Tay nie (io 2= MAS) 11.4 ODDS) ame) Lea i4) 3 12.9 +3.0 SE LG ae i.7/ Se ilsil 22 :1n-9 ()) 2b YES) 0.45 V-3= (0.7/ .49 02-026 0.33 0.8- 0.9 0.86 (3) + (0.007 @e=3) ae ().i17/ +0.16 +0.07 Hepatic Fatty Acids in Wild Penguins SHL7/ are shown in table 1. Linoleic (18:2n-6) and arachidonic (20:4n-6) acids were significantly (P <0.05 and P<0.025, respectively) higher and palmitic (16:0) acid significantly lower in the pre- moult rockhoppers (P<0.025) than the corres- ponding magellanics. In both species of penguins moulting resulted in reduction in eicosapentaenoic (rockhoppers P< 0.025, magellanics P<0.001) and _ doco- sapentaenoic (22:5n-3) (rockhoppers P<0.01, magellanics P<0.001) and an increase in linoleic acid percent (rockhoppers P<0.005, magellanics P< 0.001). There was lower (P<0.05) stearic (18:0), and higher gadoleic (20: 1n-9) (P<0.005) and erucic (22:1n-9) (P<0.05) acids in the post-moult magellanics compared to their pre-moult counter- parts. The post-moult rockhoppers, however, had increased palmitic (P<0.01) and decreased palmi- toleic (16:1) (P<0.05) acids compared to the corresponding pre-moult birds. Post-moult magellanics had significantly higher palmitoleic (P<0.025), gadoleic (20:1n-9) (P< 0.01) and erucic (22:1n-9) (P<0.05) acids and significantly lower arachidonic (P<0.05), and eicosapentaenoic (20 : 5n-3) (P<0.025) acids com- pared to that of the post-moult rockhoppers. DISCUSSION In both the rockhopper and magellanic penguins the major hepatic fatty acids were palmitic (16:0), stearic (18:0), oleic (18:1), arachidonic (20: 4n- 6), eicosapentaenoic (20:5n-3) and docosahex- aenoic (22:6n-3). These findings are in general agreement with the reported fatty acid profiles for total body fat in wild Adelie (Pygoscelis adeliae) penguins [23] and dermal tissue in Emperor pen- guins (Aptenodytes forsteri) [24]. Pre-moult differences in hepatic fatty acids be- tween rockhopper and magellanic penguins were likely to be due to differences in dietary habits. The rockhoppers including Eudyptes crestatus feed opportunistically on squid, crustaceans, euphau- sids, and small fish [11, 25, 26]. These foods would be rich in long chain n-3 fatty acids, with smaller amounts of the n-6 [27-30]. The magellanics’ lower hepatic n-6 fatty acids imply a greater diet- ary dependence on n-3 rich species. Diverse feeding ecologies have been reported for several penguin species [3, 31] and fatty acid compositions are known to be a reflection of both metabolism and diet [32-34]. Zar [24] suggested that the fatty acid differences between the adipose tissues of the Emperor (Aptenodytes forsteri) and of the Adelie (Pygoscelis adeliae) penguins were an effect of diet. In addition, Johnson and West [23] found that the proportions of fatty acids in Adelie pen- guin depot fat closely resembled the proportions of fatty acids in their normal diet of krill (Euphausia Sp). Both penguin species have relatively high pro- portions of liver arachidonic acid (n-6) despite living in a n-3 fatty acid rich environment. This may indicate a physiological requirement in pen- guins for arachidonic acid similar to that in mammalian species (33). Diet selection patterns or rates of desaturation and elongation could account for the relatively high arachidonic acid composition. Similarly we previously reported [35] significant proportions of n-6 fatty acids in the liver phosphoglycerides of wild dolphins feeding in n-3 fatty acid-rich environments. The post-moult disparity in hepatic fatty acids between the rockhoppers and magellanics was likely to be the result of species differences in the metabolism of lipids. It could also be that the penguins were at different stages of moulting, utilising nutrients differently. Moulting is char- acterised by three distinct phases, I and II repre- senting essentially lipid catabolism, with >90% of energy expenditure stemming from lipids in phase II; in phase III any remaining lipid reserves, are catabolised and proteins are also utilised [36, 37]. The rockhoppers were perhaps in the final phase of moulting, utilising more proteins; and the magella- nics in the earlier phases, because of their greater fatty acid mobilisation. These findings are in agreement with our earlier observations [5]; that the post-moult rockhoppers had significantly lower plasma albumin and globulin compared to their magellanic counterparts indicating that the rock- hoppers were utilising higher amounts of protein. Stearic acid (18 :0) is quantitatively a major fatty acid in animal tissues, and is not preferentially oxidised as fuel in mammals [38]. The post-moult 318 L. S. HARBIGE, K. GHEBREMESKEL et al. decrease of stearic and concomitant increase in gadoleic (20: 1n-9) and erucic (22: 1n-9) acid per- cent in the magellanics but not in the rockhoppers, is therefore interesting. Increased elongation and desaturation of stearic acid in the magellanics to compensate for the relative post-moult loss in unsaturation, may explain these findings. Physio- logical adaptation to low environmental tempera- tures result in increase of unsaturation in the tissues of many species [39-42]. The post-moult drop in unsaturation in the magellanics, may have induced this adaptive elongation and desaturation mechanism. Preferential mobilisation and utilisa- tion of the specific long chain n-3 fatty acids eicosapentaenoic and docosapentaenoic (22 : 5n-3) during moulting was a consistent biochemical finding in both the rockhopper and magellanics. Because of the dietary abundance of n-3 fatty acids in marine ecosystems, it appears that these pen- guins have evolved metabolic mechanisms to pre- ferentially utilise these fatty acids. There was a significant increase in the propor- tion of hepatic linoleic acid (18 :2n-6) after moult- ing both in the magellanics and rockhoppers, together with a reduciton in the long chain n-3 fatty acids, eicosapentaenoic (20 : Sn-3) and docosa- pentaenoic (22 :5n-3). This n-6 and n-3 interaction is consistent with the findings of Gudbjarnason and Oskarsdottir [43] and Harbige ef al. [22] in mam- mals. They found that increases in the proportion of long chain n-3 fatty acids was associated with a decrease in n-6 fatty acids, particularly linoleic acid. Linoleic acid is thought to have a role in the maintenance or formation of the epidermal water barrier [16]. It is conceivable that our observations of increased linoleic acid percent in the liver of both penguin species post-moult, may indicate specific mobilisation in relation to water barrier function during the vulnerable moulting period. Also there appears to be a differential sparing or conservation of the highly unsaturated docosahexa- enoic acid (22 :6n-3) and arachidonic acid in both species after moulting. Specific increases in mem- brane docosahexaenoic acid (22:6n-3) with cold adaptation have been reported [44, 45]; and may partly explain our findings as could the preferential retention of these fatty acids by hepatic cells. Conservation of docosahexaenoic acid, may also have contributed to the post-moult decrease in the proportions of eicosapentaenoic and docosa- pentaenoic acid composition through chain elonga- tion and or desaturation. In conclusion we suggest that pre-moult hepatic fatty acid status of the rockhopper and magellanic penguins are mainly of dietary origin. Whereas, the post-moult values are a reflection of the stage of moult, and mobilisation and utilisation of lipids which appear to be both species dependent and species independent. REFERENCES 1 Moss, R., Watson, A., and Parr R. (1975) Maternal nutrition and breeding success in red grouse (Lagopys lagopus scoticus). J. Animal. Ecol., 44: 233-244. 2 Odum, E. P. (1965) Adipose tissue in migratory birds. In “Handbook of Physiology.” Section 5, Adipose tissue. Ed. by A. E. Renold and G. F. Cahill American Physiological Society Washington D. C., pp. 37-43. 3 Heezik, Y. and Davis, L. (1990) Effects of food variability on growth rates, fledging sizes and repro- ductive success in the yellow-eyed penguin (Megady- ptes antipodes). Ibis. 132: 354—365. 4 Hornfeldt, B., Carlsson, B. G., Lofgren, O. and Eklund, U (1990) Effects of cyclic food supply on breeding performance in Tengmalm’s owl (Aegolius funereus). Can. J. Zool., 68: 522-530. | 5 Ghebremeskel, K., Williams, G., Keymer, I. F., Horsley, D. and Gardner D. A. (1989) Plasma chemistry of rockhopper (Eudytes crestatus), magel- lanic (Spheniscus magellanicus) and gentoo (Pygos- celis papua) wild penguins in relation to moult. Comp. Biochem. Physiol., 92A: 43-47. 6 Moss, R. (1972) Food selection by red grouse (Lagopus lagopus scoticus) in relation to chemical compostion. J. Animal. Ecol., 41: 411-428. 7 Cherel, Y. and Le Maho Y.(1988) Changes in body mass and plasma metabolites during short-term fast- ing in the king penguin. The Condor, 90: 257-258. 8 Stonehouse, B. (1967) The general biology and thermal balances of penguins. Adv. ecol. Res., 4: 131-196. 9 Richdale, L. E. (1957) A population study of Penguins. Clarendon Press, Oxford, p. 95. 10 Sparks, J and Soper T. (1967) Penguins. by Newton Abbott: David & Charles, p. 101. 11 Strange, I. J. (1982) Breeding ecology of the rock- hopper penguin (Eudyptes crestatus) in the Falkland Islands. Le Gerfaut. 72: 137-188. 12 Williams, A. J., Siegfried, W. E., Burger, A. E. and 13 14 15 16 ty 18 19 20 21 22 23 24 DS Hepatic Fatty Acids in Wild Penguins -Berruti A. (1977) Body composition and energy metabolism of moulting eudytid penguins. Comp. Biochem. Physiol., 56A: 27-30. Gurr, M. I. and James A. T. (1980) Lipid Bioche- mistry: an introduction. Chapman and Hall, p. 111. Crawford, M. A. (1983) Background to essential fatty acids and their prostanoid derivatives. Br. Med. Bull., 39: 210-213. Martin, D. W., Mayes, P. A. and Rodwell V. W. (1983) Harper’s Review of Biochemistry. 19th edi- tion. Lange Medical Publications, pp. 188-247. Wenz es W. “Abraham; W., Cho, E:~S: and Downing D. T. (1986) Linoleate-Rich O- Acylsphingolipids of mammalian epidermis: Struc- tures and effects of essential fatty acid deficiency. Prog. Lipid. Res., 25: 383-389. Hou, S. E., Mitra, A. K., White, S. H., Menon, G. K., Ghadially, R. and Elias P. M. (1991) Membrane structures in normal and essential fatty acid-deficient stratum corneum: Characterization by ruthenium tetroxide staining and X-ray diffraction. J. Invest. Dermatol., 96: 215-223. Budowski, P., Leighfield, M. J. and Crawford, M. A. (1987) Nutritional encephalomalacia in the chick: an exposure of the vulnerable period for cerebellar development and the possible need for both n-6 and n-3 fatty acids. Br. J. Nutr., 58: 511- SAY. Bazan, N. G. (1990) Supply of n-3 polyunsaturated fatty acids and their significance in the central ner- vous system. In “Nutrition and the brain.” Vol. 8. Ed. by R. J. Wurtman and J. J. Wurtman Raven Press Ltd., NY. pp. 1-24. Keymer, I. F. (1987) An investigation of rockhop- per penguin (Endyptes crestatus) mortality in the Falkland Islands during the 1985-1986 breeding season. Falkland Island Foundation Project Report. Folch, J. L. and Stanley, G. H. S. (1957) A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem., 226: 497- 509. Harbige, L. S., Ghebremeskel, K., Williams, G. and Summers, P. (1990) N-3 and N-6 phospho- glyceride fatty acids in relation to in vitro erythro- cyte haemolysis induced by hydrogen peroxide in captive common marmosets (Callithrix jacchus). Comp. Biochem. Physiol., 97B: 1. 167-170. Johnson, S. R. and West, G. C. (1973) Fat content, fatty acid composition and estimates of energy metabolism of adelie penguins (Pygoscelis adeliae) during the early breeding season fast. Comp. Biochem. Physiol., 45B: 709-719. Zar, J. H. (1977) Fatty acid composition of emperor penguin (Aptenodytes forsteri) lipids. Comp. Biochem. Physiol., 56B: 109-112. Brown, C. R. and Klages, N. T. (1987) Seasonal 26 “aj 28 29 30 Bill 32 38 34 30) 36 Sy) 38 Si) and annual variation in diets of Macaroni (Eudyptes chrysolophus chrysolophus) and Southern rockhop- per (E. chrysocome chrysocome) penguins at sub- Antartic Marion Island. J. Zool., Lond., 212: 7-28. Croxall, J. P., Prince, P. A., Baird, A. and Ward, P. (1985) The diet of the southern rockhopper penguin Eudyptes chrysocome chrysocome at Beauchene Is- land, Falkland Islands. J. Zool., Lond., (A)., 206: 485-496. Crawford, M. A., Doyle, W., Drury, P., Gheb- remeskel, K., Harbige, L., Leyton, J. and Williams, G. (1989) The food chain for n-6 and n-3 fatty acids with special reference to animal products. In “Di- etary n-3 and n-6 Fatty Acids biological effects and nutritional essentiality.” NATO ASI Series A. Vol 171. Ed. by C. Galli. and A. P. Simopoulos Plenum Press, pp. 5-19. Ackman, R. G. (1982) Fatty acid composition of fish oils. In “Nutritional evaluation of long-chain fatty acids in fish oil.” Ed. by S. M. Barlow and M. E. Stansby Academic Press, pp. 25-88. Sidhu, G. S., Montgomery, W. A., Holloway, G. L., Johnson, A. R. and Walker D. M. (1970) Biochemical composition and nutritive value of krill (Euphausia superba). J. Sci. Fd. Agric., 21: 293- 296. Clarke, A. (1980) The biochemical composition of krill Euphausia superba, from South Georgia. J. exp. mar. Biol. Ecol., 43: 221-236. Lishman G. S. (1985) The food and feeding ecology of adelie penguins (Pygoscelis adeliae) and Chin- strap penguins (P. antarctica) at Signy Island, South Orkney Islands. J. Zool., Lond, (A), 205: 245-263. Tinco, J. (1982) Dietary requirements and functions of alpha-linolenic acid in animals. Prog. Lipid. Res., 21: 1-45. Sanders, T. A. B. (1988) Essential and Trans-fatty acids in nutrition. Nutrition Research Reviews, 1: 57-78. Stubbs, C. D. and Smith, A. D. (1984) The modi- fication of mammalian membrane polyunsaturated fatty acid composition in relation to membrane fluidity and funciton. Biochim. Biophys. Acta., 779: 89-137. Williams, G. and Crawford M. A. (1987) Compari- son of the fatty acid component in structural lipids from dolphins, zebra and giraffe: possible evolution- ary implications. J. Zool., Lond., 213: 673-684. Cherel, Y., Robin J-P. and Le Maho Y. (1988) Physiology and biochemistry of long-term fasting in birds. Can. J. Zool., 66: 159-166. Robin, J-P., Frain, M., Sardet, C., Groscolas, R. and Le Maho Y. (1988) Protein and lipid utilization during long-term fasting in emperor penguins. Am. J. Physiol., 254: R61-R68. Leyton, J., Drury, P. J. and Crawford, M. A. (1987) 39 40 4] 42 320 In vivo incorporation of labeled fatty acids in rat liver lipids after oral adminstration. Lipids., 228: 553-558. Williams, D. D. and Platner W. S. (1967) Cold- induced changes in fatty acids of the rat and ham- ster. Am. J. Physiol., 212(1): 167-172. Aloia, R. C. and Pengelley, E. T. (1979) Lipid composition of cellular membranes of hibernating mammals. In “Chemical Zoology.” Vol XI Mamma- lia. Ed. by M. Florkin and B. T. Scheer Academic Press, pp. 1-47. Cowey, C. B. (1988) The Nutrition of fish: The developing scene. Nutrition Research Reviews, 1: 255-280. Houslay, M. D., and Stanley, K. K. (1983) Dyna- mics of Biological Membranes. John Wiley & Sons, pp. 138-141. 43 44 45 L. S. HARBIGE, K. GHEBREMESKEL et al. Gudbjarnason, S. and Oskarsdottir, G. (1977) Modification of fatty acid composition of rat heart lipids by feeding cod liver oil. Biochim. Biophys. Acta., 487: 10-15. Miller, N. G. A., Hill, M. W. and Smith M. W. (1976) Positional and species analysis of membrane phospholipids extracted from goldfish adapted to different environmental temperatures. Biochim. Biophys. Acta., 55: 644-654. Hazel, J. R. and Zerba, E. (1986) Adaption of biological membranes to temperature: molecular species compositions of phosphatidyl choline and phosphatidyl ethanolamine in mitochondrial and microsomal membranes of liver from thermally acclimated rainbow trout. J. Comp. Physiol. 156B: 665-674. ZOOLOGICAL SCIENCE 9: 321-328 (1992) © 1992 Zoological Society of Japan Proximate Composition and Allocation of Energy to Body Components in Acanthaster planci (Linnaeus) (Echinodermata: Asteroidea) Joun M. Lawrence! and PETER MoRAN Australian Institute of Marine Science, Townsville, Queensland 4810, Australia ABSTRACT—Protein was the major constituent by weight in all body components. The concentration of lipid was twice as great in the pyloric caeca as in the cardiac stomach and cardiac pouches. The body wall of an arm contained more organic material and energy (in kJ) than the pyloric caeca within the arm. The body wall of the arms is the greatest portion of the entire body, but is less important in terms of wet weight than in kJ. The ventral body-wall of the disc is massive, containind ca. 17% as many kJ as the ventral body-wall of all 16 arms. The type and amount of organic constituents allocated to the body components of Acanthaster planci indicate the functional requirements of the components. The greater amount of energy allocated to the body wall of an arm than to the pyloric caeca suggests that an increase in arm number is not adaptive unless it results in an increased capacity to obtain energy. INTRODUCTION The proximate composition of asteroids differs among the body components in a species and between the same components of different species [1-10]. The differences are particularly great for the body wall, associated with the great variation in the body-wall functional morphology [11]. The proximate composition indicates the requirements for organic classes in the body components in either gravimetric or energetic terms. The proxi- mate composition is expressed most often in rela- tive terms, but the absolute amounts of the proxi- mate constituents are of interest in considering production and allocation of material to body components. Production is best expressed in ener- gy terms [12] and knowledge of the proportional representation of the organic classes can be of value [13]. The allocation of energy should be balanced among the body components according to the principle of economization in metabolic expenditure [14]. With optimal design (sym- Accepted December 13, 1991 Received May 9, 1991 ' Permanent address: Department of Biology, Uni- versity of South Florida, Tampa 33620, U.S.A. morphosis) the allocation of resources to structural elements should meet but not exceed the require- ments of the functional system [15]. Studies of the allocation of proximate constitu- tents to the body components of asteroids have concerned five-armed species except for the mul- tiarmed Pycnopodia helianthoides [2, 5]. The relation between the relative amount of energy in the body components and body size has been established for the multiarmed Acanthaster planci [16]. The study of multiarmed species is important as a means of understanding the relationship be- tween size in terms of dimensions and biomass of a body and its components [17]. These studies are of particular interest as the multiarmed condition is relatively rare in asteroids despite its long fossil record and widespread occurrence in different families. The present study addresses this question through consideration of the allocation of proxi- mate constitutens and its energy equivalents to the body components of Acanthaster planci. MATERIALS AND METHODS Acanthaster planci were collected at Bowden Reef, Great Barrier Reef, Australia (147°56E, 19°02’S) on 3 May 1989. At the date of collection B22 J. M. LAWRENCE AND P. MorANn the perimeter of the reef had ca. 10-30% live coral cover and 1-10% dead coral cover [18]. The collected individuals were held in aquaria with running sea-water for 5—9 days before dissection. The major (R) and minor (r) radii and the disc radius were measured immediately after the indi- vidual had been removed from the aquarium. The disc was defined as that portion of the body containing the cardiac stomach [19]. The indi- viduals were dissected into their body components: dorsal and ventral body-walls of the arms and disc, cardiac stomach, cardiac pouches, and pyloric caeca. The cardiac pouches are extensions of the cardiac stomach in the proximal fused portion of the arms [20]. The arms were separated into the distal free portions and the proximal fused por- tions. Gonads were not analysed as the individuals were at the beginning of gonadal development [21]. Three arms were dissected from 6 individuals to ascertain variation in the wet weights of arm components. One arm was dissected from the remaining individuals. The entire disc was dis- sected from all individuals. Portions of each body component were weighed, lyophilized, reweighed, and homogenized. The proximate composition of the components and their energy equivalents were measured by the methods used by Lawrence [3] and the insoluble protein calculated by substraction. The amount of energy present in the components was calculated by multiplying (mg organic class/mg dry tissue) (mg dry tissue/mg wet tissue) (mg wet tissue of the body component) (kJ/mg organic class). The energy conversion factors for the organic classes were those of Kleiber [22]. These values were used to calculate the amount of energy allocated to the body components of individuals of a standard size. The mean major radius was used to designate a standard-sized individual. RESULTS The individuals varied little in size, with mean values (and SD) of 182+11, 93+13, and 45+8 mm for R, r, and disc radius, respectively. The mean arm number was 16+2 (x+SD; range, 12 to 20). The wet weights of the arm components differed among individuals and varied irregularly for different components within an individual (Table 1). The amount of variation was small. The mean weights of the components of three arms of an individual varied as much as the means of components from one arm from each of 26 indi- viduals. The proximate composition of the body wall of all parts of the body was similar (Table 2). The composition of the pyloric caeca differed from that of the cardiac stomach and pouches. The gravi- metric concentration of ash was higher in the body wall than in the viscera: Protein was the major organic constituent in all body components, and was present in highest concentration in the viscera. The concentration of lipid was twice as great in the pyloric caeca as in the cardiac stomach and pouches. The kJ per g dry weight of the body wall was less than half that of the viscera (Table 3). The kJ per g ash-free dry weight for the body wall and cardiac TABLE 1. Variation in g wet wt of arm components in individual Acanthaster planci (n=3 arms for each individual) and in 26 A. planci (one arm for each individual) from Bowden Reef in May 1989. Means + 1 SD are given. Individual 1 D 4 5 6 1-26 Free arm dorsal I4Ose0. -lSOaeZ3 IS. Aaeil.d WZ 0se ted 3 Ts I3.5se 340 ventral Qed) ae ted) SOsEz3 IS Ase IZ VAG sta0kS Dae Ne) 4.0+ Mile syassad Fused arm dorsal sarily “NO aeei(0 wae AS) ~ NODA se0.5 425209 Sol ae 02 ORs ventral O.Ose i WAAae21 9.4+0.9 Weak es Sad) ae()),4! 4.8+1.6 NO Saes40) Pyloric caeca Yael WOLOseiO 4 Saeid UO~se0.7 - 4,@se00 SWJse0S) NS as Ses) Cardiac pouches 340 a2 1 2.0+0.4 3.4+0.8 e229. ZYaesd 2.3+0.6 Zed) ae) Body Components in Acanthaster planci 323 stomach and pouches were similar and less than that for the pyloric caeca as a result of the differ- ence in lipid level. Protein was the major constituent of Acanthaster planci in either gravimetric or energetic units, but was less important in terms of the latter (76 vs 69% of the organic matter) (Table 4). The protein was equally distributed between soluble and insoluble protein. Lipid constituted 28% of the 3491 kJ in the body of a standard-sized individual, with 32% of this in the pyloric caeca. The wet or dry weights of the ventral and dorsal portions of the free and fused parts of the arm did not differ greatly (Table 4). The g organic material and kJ were slightly greater in the dorsal portion of the arm. The wet weight of the ventral body-wall TABLE 2. Per cent dry weight (in % total weight) and proximate composition (in % g dry weight and % kJ) of body components of Acanthaster planci from Bowden Reef in May 1989. A: ash, C: carbohydrate, L: lipid, SP: soluble protein, IP: insoluble protein. Means+1S.D. (n=10) are given for the % dry wt. The % kJ was calculated from the mean % dry wt values. epasotent a) pS C L SP IP % dry wt Body wall Ventral disc Dilated) 64+5 ae O.7 3.36 (0)5) 14 16+8 Dorsal disc 745) SIE 49+6 1.9+0.2 Solace) 20 te A=) Ventral free-arm Dat 3 /a2© 1.6+0.3 3.6+0.4 195: 19+6 Dorsal free-arm Ddyts 622-5 2203 7/ 3.3+0.4 Sse 18+3 Ventral fused-arm Dict 58 +6 1.6+0.3 3.4+0.5 17/ se 18+8 Dorsal fused-arm Bye 54+4 Ose) 4.1+0.5 7/ ae 24+4 Pyloric caeca 24+4 Saez VOse 10) SM) se IU 34+ 5) 22 WY) Cardiac stomach 20EE3, OFS aeles So) ae 17 15+4 33 30+4 Cardiac pouches Placed 8.7+1.4 Op leieale2 16+1 336 3444 % kJ Body wall Ventral disc 2.6 WS 38 44 Dorsal disc Mod) IS 36 45 Ventral free-arm ed) 13 42 42 Dorsal free-arm Doel) 14 38 46 Ventral fused-arm 3.0 113} 40 44 Dorsal fused-arm Dra 15 34 39 Pyloric caeca 4.9 50 34 WZ Cardiac stomach 4.5 28 4] Dif Cardiac pouches 4.4 28 33 39 TABLE 3. Bowden Reef in May 1989. Free arm Fused arm kJ per g dry weight and ash-free dry weight in body components of Acanthaster planci from DBW: dorsal body wall, VBW: ventral body wall. Disc Body Cardiac Cardiac Pyloric Component DBW VBW DBW VBW DBW VBW stomach pouches’ caeca kJ per g dry wt 9.3 ill 11 10 13 8.6 23 23 WY kJ per g ash-free dry wt We) M5) 7) 24 26 24 26 25 31 324 J. M. LAWRENCE AND P. Moran TaBLE4. Calculated amounts (g and kJ) of total material and of proximate constituents in the body components of Acanthaster planci with 16 arms and a major radius of 182mm. The values for dry weight and proximate constituents were calculated from the values for the % proximate composition given in Table 2. C: carbohydrate, L: lipid, SP: soluble protein, IP: insoluble protein, TOM: total organic material. wet wt dry wt Grams Body wall Free arms Dorsal 216 55 Ventral 181 42 Fused arms Dorsal 147 34 Ventral 168 35 Disc Dorsal DG Os 7/ Ventral >y/ 15 Total body-wall 796 188 Viscera Cardiac stomach 41 8.5 Cardiac pouches 36 7.6 Total ie, 16 Pyloric caeca 216 a2 Total viscera 293 68 Grand total 1089 256 kJ Body wall Free arms Dorsal Ventral Fused arms Dorsal Ventral Disc Dorsal Ventral Total body-wall Viscera Cardiac stomach Cardiac pouches Total Pyloric caeca Total viscera Grand total ash C L 34 0.8 1.8 24 0.6 1.4 18 0.5 1.4 20 0.6 iil 8).3) 0.1 0.3 9.8 0.2 0.5 109 2.8 6.5 1.0 0.5 1.4 0.6 0.5 2 1.6 1.0 2.6 3.8 365) 5) 5.4 4.5 18 114 WS) DS 14 70 11 S7/ 8.2 Si ili 44 DD 13} 3.4 20 50 261 8.9 55 8.1 51 7) 106 60 606 Vi! VAD 127 973 SP 8.2 7.8 5.6 DY) 13 Doe | Sil 3.4 226 6.0 7) DS 54 193 185 132 140 3) 51 133 79 60 139 411 550 1283 IP 9.8 7.8 8.0 6.4 Le 2.4 36 Ghd ied 4.9 £3 18 54 230 185 189 Syl 49 58 862 53 64 Luly 142 259 A TOM Zs 18 16 14 3.4 S77 78 7.5 7.0 14.5 49 64 142 506 437 386 346 87 USE 1893 195 183 378 1220 1598 3491 Body Components in Acanthaster planci B25 of the disc was twice that of the dorsal body-wall, and contained 131 kJ (17% of the 783 kJ of the entire ventral body-wall of all 16 arms of a stand- ard individual. The body wall was always the largest component of the arms regardless of the mode of measurement, but was more important when calculated in terms of dry weight (73% of the total) than in kJ (54% of the total). The viscera contained 1598 kJ (84% of the 1893 kJ of the entire body-wall). Almost 73% of the lipid was in the viscera. The pyloric caeca contained 85% of the viscera lipid. Protein comprised 84% of the kJ in the body wall, with slightly more insoluble than soluble protein. Protein comprised 51% of the kJ in the viscera, with more soluble than insoluble protein. The body wall contained only 66% of the total kJ due to protein as the total organic material in the viscera was so great. DISCUSSION The body wall and pyloric caeca of the arms of individual Acanthaster planci vary in size, and the amount of variation differs among individuals. Despite this, the amount of variation found is small and no greater than found with complete dissection of other asteroid species [3, 7, 9, 10]. The proximate composition of the body compo- nents of Acanthaster planci is in the range reported for other species [1-10]. The concentration of energy in the body wall and pyloric caeca in terms of kJ/dry wt is similar to those reported for other asteroid species [5-8, 10, 23, 24], and shows the great influence of the amount of ash on the concen- tration. The differences in the concentration of energy in terms of kJ/ash-free dry weight reflect better the difference in proximate organic com- position. Thus in these terms, the energy concen- tration of the body wall and stomach of A. planci are similar but less than that of the pyloric caeca. The allocation of material and energy to the components of an organism must be interpreted in terms of its biology. Acanthaster planci is disc- shaped, multiarmed, pliable, and prehensile, with a large central disc and stomach [21, 26, 27]. These features are associated with its predation on coral by extraoral feeding. Lucas [28] noted the massive development of the stomach of A. planci which is extruded over the coral in feeding. This develop- ment is so great that the disc does not contain the entire stomach, and extensions (the cardiac pouches) are found in the proximal portions of the fused arms. This may be a better solution to accomodating a large stomach than increasing the width of the disc. The great development of the ventral portion of the disc (the oral frame) sup- ports the retraction of the massive stomach. The slightly higher concentration of ash in the ventral body-wall is probably associated with require- ments for the supporting structures. Blake and Guensburg [29] listed a robust oral frame as one of a suite of characters for the “pycnopodaform” shape of multiarmed asteroids. They did not relate it to the mass of the stomach or include a massive stomach as one of the characters. The stomach has been ignored in studies of component parts of asteroids, but this may be a major error in the study of pycnopodaform species. Blake and Guensburg [29] also listed a robust, strongly articulated ambulacral column as a char- acter of pycnopodaform asteroids, although this is also true for other asteroid forms (Lawrence, unpub. obs.). However, the amount of material and energy allocated by Acanthaster planci to the ventral body-wall of the arms is similar to that allocated to the dorsal body-wall except for the disc. The dorsal body-wall is fragile, as pointed out by Kettle and Lucas [16], in keeping with the flexibility of the body noted above. Flexibility seems more important than having an armor to protect against predation as is more usual in tropi- cal asteroids [30]. The toxic dorsal spines of A. planci are few and represent a minor allocation of energy (Lawrence, unpub.). The high incidence of regenrating arms [31] indicates the susceptibility of A. planci to breakage or predation. This moderate allocation to protection would be predicted for a species with a competitive life-history strategy BZ The high amount of insoluble protein allocated to the body wall indicates the primarily structural role of the body wall, although the large amount of soluble protein shows considerable numbers of cells are present. The lack of difference in the proximate composition of the dorsal and ventral body-walls show the basic similarity in construc- 326 J. M. LAWRENCE AND P. MorAN tion of the two. An increase in strength and support seems to involve an increase in size and not difference in composition although this might occur at the histological level. McClintock [5] noted a decrease in concentration of inorganic material in the body wall of Pycnopodia helian- thoides with an increase in body size, indicating a greater reliance on organic material for strength with an increase in size. Kettle and Lucas [16] reported a decrease in the relative amount of energy allocated to the body wall with an increase in body size in A. planci, but did not indicate the absolute amount of energy involved or separate the body wall into components. Giese (2) pointed out that large amounts of organic material in the body wall of asteroids could constitute a nutrient reserve, and Lawrence and Lane [25] suggested that the material might be used during body-wall resorption during starva- tion. The concentration of organic material in the body wall of Sclerasterias mollis decreases with starvation [10]. The importance of body size in regard to a role of the body wall in nutrient reserve is seen with scaling (the proportion of body wall decreases with size in Acanthaster planci {16]) and composition (the concentration of organic material in the body wall increases with size in Pycnopodia helianthoides [5}). The amount of material and energy allocated to the cardiac pouches is nearly as much as to the cardiac stomach within the disc. The greater amount of insoluble protein may be associated with the ligaments that retract the pouches. The absolute amount of lipid in the cardiac stomach and pouches is high. The gut of echinoids stores lipid [33], and the lipids in the cardiac stomach may function as reserves also. The amount of lipid allocated to the pyloric caeca is far greater. The nutrient-reserve function of the pyloric caeca is well known, but the caecum is a combination of digestive and reserve cells [34] that makes it dif- ficult to know the exact allocation to either [25]. The multiple arms of Acanthaster planci result in a proportionally greater allocation of material and energy to the arm components than in five-armed species. The greater allocation of material has been noted for Luidia senegalensis [35] and Pycno- podia helianthoides [5]. This greater allocation is probably associated with both an increased cost of development and maintenance. If so, a positive return should result for the multiarmed condition to be adaptive [17]. Blake and Guensburg [29] pointed out that it is uncertain whether or not multiple arms are adap- tively neutral. Multiarmed asteroids can be sepa- rated into two groups: those with 6 to 12 arms that are constant in number, and those that have 8 or many more that are variable in number [36]. It is possible the functioning of genera in the first group (Luidia, Asterina, Leptasterias) is not affected sufficiently for arm number to be a selective factor. Blake and Guensburg suggested that the similar morphologies of pycnopodaform asteroids of dis- parate geological ages and ancestry strongly imply not only the benefit based on predatory feeding advantages, but that the benefit has endured. Genera in the second group (Acanthaster, Crossas- ter, Heliaster, Pycnopodia, Solaster) are all active, voracious carnivores in which the additional arms probably increase feeding capacity. Just as homeothermy is advantageous, but only if the return is worth the cost, the development of the multiarmed condition should increase the capacity to obtain energy that meets the energy require- ment for the development and maintenance of the additional arms. In this regard, Calder [37] pointed out that it is the body mass (how much tissue must be sustained and regulated) rather than the mass of the con- stituent parts, topographical layout, or history of use that determines basic support costs, opportuni- ties, and homeostatic needs. Recognizing the role of body size in the functioning of an organism, he concluded that body mass is not only an expedient measure of size but the biologically appropriate one. The amout of energy rather than weight better represents biomass. This is clear in echi- noderms where so much of the mass may be inorganic. Acanthaster planci has a much larger biomass in terms of kJ than the few other species for which values have been reported (3, 6, 38). ACKNOWLEDGMENTS We thank D. B. Blake and J. B. McClintock for their helpful comments on the manuscript. 10 11 12 13 14 Body Components in Acanthaster planci REFERENCES Giese, A. C. (1966) On the biochemical constitution of some echinoderms. In “Physiology of Echi- nodermata”. Ed. by R. A. Boolootian, Interscience Publ., New York, pp. 757-796. Giese, A. C. (1976) Physiology of the echinoderm body wall. Thalassia Jugoslav., 12: 153-163. Lawrence, J. M. (1973) Level, content, and caloric equivalents of the lipid, carbohydrate, and protein in the body components of Luidia clathrata (Echi- nodermata: Asteroidea: Platyasterida) in Tampa Bay. J. Exp. Mar. Biol. Ecol., 11: 263-274. Lawrence, J. M. and Guille, A. (1982) Organic composition of tropical, polar and temperate-water echinoderms. Comp. Biochem. Physiol., 72B, 283- 287. McClintock, J. B. (1989) The biochemical and energetic composition of somatic tissues during growth in the sea star, Pycnopodia helianthoides (Echinodermata: Asteroides). Comp. Biochem. Physiol., 93A: 695-698. McClintock, J. B. (1989) Energetic composition, reproductive output, and resource allocation of antarctic asteroids. Polar Biol., 9: 147-153. McClintock, J. B., Pearse, J. S. and Bosch, I. (1988) Population structure and energetics of the shallow- water antarctic sea star Odontaster validus in con- trasting habitats. Mar. Biol., 99: 235-246. McClintock, J. B., Hopkins, T., Watts, S. A. and Marion, K. (1990) The biochemical and energetic composition of somatic body components of echi- noderms from the northern Gulf of Mexico. Comp. Biochem. Physiol., 95A: 529-532. Scheibling, R. E. and Lawrence, J. M. (1982) Dif- ferences in reproductive strategies of morphs of the genus Echinaster (Echinodermata: Asteroidea) from the eastern Gulf of Mexico. Mar. Biol., 70: 51-62. Xu, R. A. and Barker, M. F. (1989) Laboratory experiments on the effects of diet on the gonad and pyloric caecum indices and biochemical composition of tissues of the New Zealand starfish Sclerasterias mollis (Hutton 1872) (Echinodermata: Asteroidea). J. Exp. Mar. Biol. Ecol., 136: 23-45. Blake, D. B. (1989) Asteroidea: Functional mor- phology, classification and phylogeny. Echinoderm Studies, 3: 179-223. Brody, S. (1945) Bioenergetics and growth. Hafner Publishing Company Inc., New York. Paine, R. T. (1971) The measurement and applica- tion of the calorie to ecological problems. Ann. Rev. Ecol. Syst., 2: 145-164. Calow, P. (1984) Economics of ontogeny adaptational aspects. In “Evolutionary Ecology”. Ed. by B. Shorrocks, Blackwell Scientific Publica- tions, Oxford, pp. 81-104. 15 16 17 18 19 20 Za 22 D3 24 Ds) 26 Zi) 28 29 30 B27] Taylor, C. R. and Weibel, E. R. (1981) Design of the mammalian respiratory system. I. Problem and strategy. Respir. Physiol., 44: 1-10. Kettle, B. T. and Lucas, J. S. (1987) Biometric relatiohships between organ indices, fecundity, oxy- gen consumption and body size in Acanthaster planci (L.) (Echinodermata: Asteroidea). Bull. Mar. Sci., 41: 541-551. Lawrence, J. (1988) Functional consequences of the multiarmed condition in asteroids. In. “Echinoderm Biology”. Ed. by R. D. Burke, P. V. Mladenoyv, P. Lambert and R. L. Parsley, A. A. Balkema, Rotter- dam, pp. 597-602. Bass, D. K., Davidson, J., Johnson, D. B., Miller- Smith, B. A. and Mundy, C. N. (1989) Broadscale surveys of crown-of-thorns starfish on the Great Barrier Reef 1987 to 1988. The Crown-of-thorns Study. Australian Institute of Marine Science: Townsville. Lawrence, J. (1990) The relationship between the major and minor radii and the and the internal anatomy of asteroids. Northeast Gulf Sci., 11: 90. Jangoux, M. (1982) Digestive systems: Asteroidea. In “Echinoderm Nutrition”. Ed. by M. Jangoux and J. M. Lawrence, A. A. Balkema, Rotterdam, pp. 235-272. Moran, P. J. (1986) The Acanthaster phenomenon. Ocean. Mar. Biol. Ann. Rev., 24: 379-480. Kiciber Meas @S75) malhe Firerof lite. Robert: Krieger Publishihg Company, Huintington. Dayton, P. K., Robilliard, G. A., Paine, R. T. and Dayton, L. B. (1974) Biological accomodation in the benthic community of McMurdo Sound, Antarc- tica. Ecol. Monogr., 44: 105-128. Lawrence, J. M. (1987) Echinodermata. In “Animal Energetics”. Ed. by T. J. Pandian and F. J. Vern- berg, Academic Press, San Diego, pp. 229-321. Lawrence, J. M. and Lane, J. M. (1982) The utilization of nutrients by postmetamorphic echi- noderms. In “Echinoderm Nutrition”. Ed. by M. Jangoux and J. M. Lawrence, A. A. Balkema, Rotterdam, pp. 331-371. Birkeland, C. (1989) The Faustian traits of the crown-of-thorns starfish. Am. Sci., 77: 154-163. Birkeland, C. and Lucas, J. S. (1990) Acanthaster planci: major management problems of coral reefs. CRC Press, Boca Raton. Lucas, J. S. (1984) Growth, maturation and effects of diet in Acanthaster planci (L.) (Asteroidea) and hybrids reared in the laboratory. J. Exp. Mar. Biol. Ecol., 79: 129-147. Blake, D. B. and Guensburg, T. E. (1989) Two new multiarmed Paleozoic (Mississippian) asteroids (Echinodermata) and some paleobiologic implica- tions. J. Paleont., 63: 331-340. Blake, D. B. (1983) Some biological controls on the Sl 3) 33) 34 328 distribution of shallow water sea stars (Asteroidea; Echinodermata). Bull. Mar. Sci., 33: 703-712. McCallum, H. I., Endean, R. and Cameron, A. M. (1989) Sublethal damage to Acanthaster planci as an index of predation pressure. Mar. Ecol. Prog. Ser., 56: 29-36. Lawrence, J. M. (1990) The effect of stress and disturbance on echinoderms. Zool. Sci., 7: 17-28. Lawrence, J. M., Lawrence, A. L. and Giese, A. C. (1966) Role of the gut as a nutrient-storage organ in the purple sea urchin (Strongylocentrotus purpur- atus). Physiol. Zool., 39: 281-290. Nimitz, Sister M. A. (1971) Histochemical study of gut nutrient reserves in relation to reproduction and nutrition in the sea stars, Pisaster ochraceus and 35 36 of 38 J. M. LAWRENCE AND P. MORAN Patiria miniata. Biol. Bull., 140: 461-481. Lawrence, J. (1987) Une histoire de deux etoiles: leffet du nombre des bras sur la biologie. Bull. Soc. Sc. Nat. Ouest France. Suppl. H. S., 59-61. Lawrence, J. M. and Komatsu, M. (1990) Mode of arm development in multiarmed species of aster- oids. In “Echinoderm Research”, Ed. by C. De Ridder, P. Dubois, M.-C. Lahaye and M. Jangoux, A. A. Balkema, Rotterdam, pp. 269-275. Calder, W. A. (1984) Size, Function, and Life History. Harvard Univ. Press, Cambridge. Lawrence, J. M. (1985) The energetic echinoderm. In “Echinodermata”. Ed. by B. F. Keegan and B. D. S. O’Connor, A. A. Balkema, Rotterdam, pp. 47-67. ZOOLOGICAL SCIENCE 9: 329-335 (1992) Chromogranin A-Like Proteins in the Heat-Stable Fraction of Sea Urchin Eggs, Embryos and the Substances Secreted with Sperm Yukio Funno!, AKIKO FusIwARA7, [KUO YASUMASU~ and Tomoko Fujii Department of Pharmacology, Teikyo University School of Medicine, 2-11-1, Kaga, Itabashi-ku, Tokyo 173, and *Department of Biology, School of Education, Waseda University, 1-6-1, Nishiwaseda, Shinjuku-ku, Tokyo 160, Japan. ABSTRACT—Chromogranin A, an acidic, sulfated and heat-stable glycoprotein, is a major protein component in a variety of secretory granules of neuron and para-neuron cells. Widely phylogenetic occurrence of this protein from protozoa to mammals has been reported. In the present study, several immunoreactive proteins against bovine adrenal chromogranin A anti-serum were detected by immunoblotting of the heat-stable fraction obtained from sea urchin unfertilized eggs. Apparent molecular weights of the major components were about 166, 112, 105, 36 and 34 kDa, respectively. Content of the proteins was significantly decreased after fertilization, suggesting that these are released or decomposed at fertilization. At the pluteus stage, at which nervous system appears, the content of the immunoreactive proteins was markedly increased. Apparent molecular weights of the major component were about 214, 105, 89, 84, 36 and 34 kDa. No immunoreactive band was detected in the heat-stable fraction obtained from “washed” sperm, while the fraction from “dry” sperm contained several immunoreactive proteins. This suggests that not sperm but substances secreted with sperm contain these proteins. Although the role of the immunoreactive proteins on fertilization, embryonic development and sperm functions is unknown at present, they may take a part of the role on these © 1992 Zoological Society of Japan Processes. INTRODUCTION Chromogranin A, an acidic, sulfated and heat- stable glycoprotein, is a major protein component in a variety of secretory granules in neuron and paraneuron cells [for review see 1]. It has been well known that this protein is co-stored and co-released with secretory products such as catecholamines at exocytosis of the secretory gran- ules. Widespread phylogenetic occurrence of this protein including Tetrahymena [2], Paramecium [3], snail [4], lobster, carp, frog [5], lizard [6], bird [5] and mammals [1, 5] has been also reported. In these reports, polyclonal antibodies used for the detection of chromogranin A were derived by Accepted December 20, 1991 Received December 4, 1991 ' To whom all correspondence should be addressed. immunizing rabbits with bovine adrenal chromo- granin A as an antigen and chromogranin A- related proteins from protozoa to mammals were all immunoreactive against these antibodies. Chromogranin A seems to be a very conservative protein among animal kingdom. The role of chromogranin A has been proposed to contribute to condensation of secretory pro- ducts by forming a macro-molecular complex with the secretory products and ATP resulting in lower- ing intra-granular osmolality [for review see 7]. In the previous report [8], a part of the present authors demonstrated that a macro-molecular complex of bovine adrenal chromogranin A, adrenaline and ATP was formed in a Ca’*- independent manner, when mixed together at pH 5.9, physiological pH within the bovine adrenal chromaffin granules [9-11], catecholamine storing organelles. The optimum concentration ratio of 330 Y. Fustno, A. Fusrwara et al. adrenaline to ATP for the formation was 4:1 [8], in accordance with the ratio within the granules RIS). In the present study we examined the presence of immunoreactive proteins against bovine adrenal chromogranin A anti-serum in sea urchin sperm, eggs and embryos. It has been established that a significant enrichment of chromogranin A results from boiling followed by centrifugation [14, 15], and that immunoreactivity is not abolished by boiling and in the presence of several protease inhibitors [16]. Therefore, the heat-stable fraction of sea urchin sperm, eggs and embryos, prepared in the presence of protease inhibitor, was used as the source for the detection of immunoreactive proteins. MATERIALS AND METHODS Culture of embryos: Eggs and sperm of Anthoci- daris classispina were obtained by intra-coelomic injection of 0.5 M KCl. Eggs were washed three times with artificial seawater and inseminated. Fertilized eggs were washed for three times with artificial seawater and cultured in artificial seawa- ter at 20°C with gentle agitation. Embryos were harvested at the 4-cell stage (3 hr after fertiliza- tion), the morula stage (6hr), the blastula stage (10 hr), the mesenchyme blastula stage (17 hr), the gastrula stage (24 hr) and the pluteus stage (48 hr). Preparation of the heat-stable fraction: Unferti- lized eggs and embryos were washed three times with artificial seawater and homogenized in the homogenizing medium consisted of 5 mM ethylene bis [oxyethylenenitrilo] tetraacetic acid (EGTA) and 1 mM phenylmethylsulfonyl fluoride (PMSF) using a Teflon pestle homogenizer in an ice bath. Sperm were washed twice with ice cold artificial seawater and homogenized as described above. In some experiments, “dry” sperm were used without washing. Protein concentration of the homogenate was adjusted to 5 mg protein/ml by dilution with the homogenizing medium. The homogenate was boiled in the test tube (16 mm in diameter, 100 mm in length) for 5 min, cooled to 4°C just after boiling and centrifuged at 10,000xg for 20min. The resultant supernatant was dialyzed against distilled water overnight at 4°C and lyophylized. The lyophylized sample was dissolved in a_ small amount of distilled water and used as the heat- stable fraction. Sodium dodecyl sulfate-polyacrylamide gel elec- trophoresis (SDS-PAGE): SDS-PAGE was per- formed by the method of Laemmli [17] using 10 to 20% linear concentration gradient of acrylamide. After electrophoresis, proteins in the gel were electrophoretically transferred to a nitro-cellulose membrane by the method of Burnett [18]. Total proteins on the membrane were visualized using a commercially provided kit (Blotting detection kit- for total protein, Amersham, U.K.). In this method, proteins on the membrane were unspecif- ically biotinylated with the biotinylation reagent and the biotinylated proteins were then detected using a streptavidine-alkaline phosphatase conju- gate with the detection signal generated by using a combination of nitroblue tetrazolium and 5- bromo-4-chloro-3-indolyl phosphate. In some ex- periments protein bands were also visualized by Coomassie brilliant blue staining. Immunostaining was carried out using a commercially provided kit (ProtoBlot, Promega, WI, U.S.A.), Rabbit anti- bovine adrenal chromogranin A serum, prepared as described previously [8], was used in a 1:400 dilution ratio. Protein determination: Protein concentration was measured by the method of Bradford [19] with bovine gamma globulin as a standard. Chemicals: EGTA and PMSF were purchased from Sigma Chem. Co., MO, U.S.A. Acrylamide, molecular weight markers for SDS-PAGE and the dye reagent for protein determination were from Bio-Rad, CA, U.S.A. Artificial sea water was the product of Jamarin Laboratory, Osaka, Japan. RESULTS Figure 1 indicates SDS-PAGE profiles of the heat-stable fraction prepared from the same pro- tein amount of unfertilized and fertilized egg homogenate. As shown in Figure la, lane 1, many protein bands were visualized in the fraction from Chromogranin A in Sea Urchin Gametes 331 unfertilized eggs by protein staining with the biotinylation method described in Materials and Methods. As revealed by immunostaining, some of these proteins were immunoreactive ones against bovine adrenal chromogranin A anti-serum (Fig. la, lane 3). Apparent molecular weights of the major immunoreactive proteins were about 166, 112, 105, 36 and 34 kDa. Table 1 shows the ratio of heat-stable protein in the total protein of the homogenate. The ratio decreased markedly after fertilization. As shown in Figure la, lanes 1 and 2, protein bands, detected by the biotinylation method, became slightly faint ones after fertilization. When protein bands were visualized by Coomassie brilliant blue staining, a slightly small number of protein bands in compari- son with that by the biotinylation method was developed (Fig. 1b, lane 1). These bands became significantly faint ones after fertilization (Fig. 1b, lanes 1 and 2) in accordance with the decrease in Atm 14.4- Fia. 1. SDS-PAGE of the heat-stable fraction obtained from sea urchin unfertilized and fertilized eggs. a. Protein bands, blotted to the nitro-cellulose filter, were visualized by the biotinylation method as described in Materials and Methods (lanes 1 and 2) and immunoreactive proteins against bovine adrenal chromogranin A anti-serum were detected as also described in Materials and Methods (lanes 3 and 4). Lanes 1 and 3: the fraction derived from unfertilized eggs; lanes 2 and 4: the fraction from fertilized eggs. b. Protein bands, blotted to the nitro-cellulose filter, were visualized by Coomassie brilliant blue staininig. Lane 1: the fraction derived from unfertilized eggs; lane 2: the fraction from fertilized eggs. The heat-stable proteins prepared from 280 yg protein of the homogenate were loaded. 332 Y. Fusyino, A. Fustwara et al. TABLE 1. Recovery of protein from homogenates of sperm, unfertilized eggs and embryos to the heat-stable fractions. Recovery of protein (%) “Dry” sperm 1.08* “Washed” sperm 0.15 Unfertilized eggs 28.5 Fertilized eggs [Saul Embryos at the 4-cell stage 15.4 the morula stage 1356 the blastula stage MN the mesenchyme blastula stage ‘SJoo) the gastrula stage 8.6 the pluteus stage MIL * The value indicates percentage of the heat-stable protein in the homogenate. Data are the mean of two separate experiments. the recovery of heat-stable proteins. The biotiny- lation method seems not to fully reflect the change in protein amount on the SDS-PAGE profiles. The content of the immunoreactive proteins also decreased at a significant extent after fertilization (Fig. la, lanes 3 and 4). In the heat-stable fraction derived from sea urchin “dry” sperm, many protein bands were detected and several protein bands were im- munoreactive ones (Fig. 2, lanes 1 and 3). Appar- ent molecular weights of the major immunoreac- tive proteins were about 132, 90, 51, 18.4 and 16.8 kDa. The 18.4kDa protein was the most abun- dant one in the fraction. These values did not coincide with those obtained from unfertilized eggs. Although several protein bands were visual- ized in the fraction obtained from “washed” sperm, no immunoreactive band was detected (Fig. 2, lanes 2 and 4). As indicated in Table 1, the recovery of heat-stable proteins from “washed” sperm was significantly lower than that from “dry” sperm. When non-immunized rabbit serum was used in place of the anti-serum, no band was developed in the all fractions (Data not shown). Changes of the composition and content of the immunoreactive proteins against bovine chromo- K Daa’ arate] soi Xs ee Fic. 2. SDS-PAGE of the heat-stable fraction obtained from “dry” and “washed” sperm. Protein bands were visualized by the biotinylation method as described in Materials and Methods (lanes 1 and 2) and immunoreactive proteins against bovine adrenal chromogranin A anti-serum were detected as also described in Materials and Methods (lanes 3 and 4). Lanes 1 and 3: the heat-stable fraction derived from “dry” sperm; lanes 2 and 4: the fraction from “washed” sperm. The heat-stable proteins prepared from 4mg protein of the homogenate were loaded. granin A anti-serum were examined during early development. The heat-stable fractions, prepared from the same protein amount of embryo homogenate, were loaded to SDS-PAGE in order to compare the content of the immunoreactive proteins quantitatively. As described above, the content was significantly decreased after fertiliza- tion (Fig. 3, lanes 1 and 2). The content was kept Chromogranin A in Sea Urchin Gametes 333 kay sasl Fic. 3. Pee 3 Qe CHAT B wo feorsG ss, Changes in the composition of immunoreactive proteins against chromogranin A anti-serum in the heat-stable fraction of the embryos during early development. Immunoreactive proteins against chromogranin A anti-serum in the heat-stable fraction were detected as described in Materials and Methods. Lane 1: unfertilized eggs; lane 2: fertilized eggs; lanes from 3 to 8: embryos at the 4-cell (3), the morula (4), the blastula (5), the mesenchyme blastula (6), the gastrula (7), and the pluteus stage (8), respectively. The heat-stable proteins prepared from 280 ug protein of the homogenate were loaded. at a low level from fertilization to the gastrulation (Fig. 3, lanes from 2 to 7). A 89 kDa immunoreac- tive protein was detected in the fracion from the 4-cell stage embryos (Fig. 3, lane 3). The content of this protein increased at the morula stage at a some extent and markedly at the pluteus stage (Fig. 3, lanes 4 and 8). A 84 kDa immunoreactive protein, that was first detected at the mesenchyme blastula stage, was also increased at the pluteus stage. At this stage, immunoreactive proteins, apparent molecular weights, 214, 105, 36 and 34 kDa, were also appeared, and these proteins as well as the 89 and 84kDa ones were the major immunoreactive components. DISCUSSION In the present study, immunoreactive proteins against bovine adrenal chromogranin A anti-serum were detected from the heat-stable fraction of unfertilized eggs as well as embryos at the various stages. No immunoreactive protein was detected in the fraction from “washed” sperm, while the fraction from “dry” sperm contained several im- munoreactive ones. The recovery of heat-stable protein from “washed” sperm was significantly lower than that from “dry” sperm. These suggest that not sperm but substances secreted with sperm contain the immunoreactive proteins. Apparent molecular weight of bovine adrenal 334 Y. Fusino, A. Fusiwara et al. chromogranin A, determined by SDS-PAGE, has been shown to be about 75 kDa [20, 21]. It has been established that within secretory granules chromogranin A exists in multiple forms that are yielded from proteolytic processing by endogenous peptidases [for review see 1]. The presence of the peptidases within the granules has been reported [22, 23]. On the other hand, proteoglycan form of chromogranin A has been known [24, 25]. Appar- ent molecular weight of this form, determined by SDS-PAGE, is greater than that of chromogranin A [24]. In the present study, apparent molecular weights of the major immunoreactive proteins were ranging from 16.8 (“dry” sperm) to 214 kDa (plutei). Although molecular weight of sea urchin chromogranin A is unknown at present, a greater part of the immunoreactive proteins in the heat- stable fraction, prepared from “dry” sperm, unfer- tilized eggs and embryos, seems to be modified by the post-translational processing. In the present study, both of the content of heat-stable protein and the chromogranin A im- munoreactive proteins were significantly decreased after fertilization. Chromogranin A has been found exclusively within the secretory granules [for review see 1]. If these proteins are localized within the cortical granules, these may be released at fertilization, resulting in the diminished content in fertilized eggs. Alternatively these proteins may be degraded at a significant extent at fertilization. The content of the 89 kDa protein was increased during early development, especially at the pluteus stage. The 84kDa protein, first detected at the mesenchyme blastula stage, also increased markedly at this stage. The 214, 105, 36 and 34 kDa protein was also present abundantly at this stage. It has been established that at the pluteus stage the nervous system appears [26-29]. These proteins detected at the pluteus stage may be correlated with the nervous system differentiation. This is the first report for the presence of chromogranin A immunoreactive proteins in “dry” sperm, unfertilized eggs and embryos. Although the role is unknown at present, these immunoreac- tive proteins may take a role in the condensation of secretory products within the granules such as cortical granules and neuro-secretory granules. 10 11 12 13 REFERENCES Winkler, H., Apps, D. K. and Fischer-Colbrie, R. (1986) The molecular function of adrenal chro- maffin granules: established facts and unresolved topics. Neuroscience, 18: 261-290. Berelowitz, M., Le Roith, D., Schenk, H., New- gard, C., Szabo, M., Frohman, L. A., Shiloach, Y. and Roth, Y. (1982) Somatostatin-like im- munoreactivity and biological activity is present in Tetrahymena pyriformis, a ciliated protozoan. En- docrinology, 110: 1939-1944. ; Peterson, J. B., Nelson, D. L., Ling, E. and Hogue- Angeletti, R. (1987) Chromogranin A-like proteins in the secretory granules of a protozoan, Para- mecium tetraurelia. J. Biol. Chem., 262: 17264- 17267. Schot, L. P. C., Boer, H. H., Swabb, D. F. and van Noorden, S. (1981) Immunocytochemical demon- stration of peptidergic neurons in the central nerv- ous system of the pond snail Lymnaea stagnalis with antisera raised to biologically active peptides of vertebrates. Cell Tissue Res., 216: 273-291. Rieker, S., Fischer-Colbrie, R., Eiden, L. and Winkler, H. (1988) Phylogenetic distribution of peptides related to chromogranins A and B. J. Neurochem., 50: 1066-1073. Arena, P. C., Richardson, K. C. and Yamada, J. (1990) An immunohistochemical study of endocrine cells of the alimentary tract of the Kings’ skink (Egernia kingii). J. Anat., 170: 73-85. Payne C. M. (1989) Phylogenetic considerations of neurosecretory granule contents: role of nucleotide and basic hormone/transmitter packaging mecha- nisms. Arch. Histol. Cytol., 52: suppl., 277-292. Fujino, Y. and Fujii, T. (1991) A macro-molecular complex of bovine adrenal chromogranin A, adrenal- ine and ATP in vitro. Biomed. Res., 12: 125-130. Johnson, R. G., Jr., Carlson, N. J. and Scarpa, A. (1978) ApH and catecholamine distribution in iso- lated chromaffin granules. J. Biol. Chem., 253: 1512-1521. Pollard, H. B., Shindo, H., Creutz, C. E., Pazoles, C. J. and Cohen, J. S. (1979) Internal pH and state of ATP in adrenergic chromaffin granules deter- mined by *'P nuclear magnetic resonance spectro- scopy. J. Biol. Chem., 254: 1170-1177. Bulenda, D. and Gratzl, M. (1985) Matrix free Ca’* in isolated chromaffin vesicles. Biochemistry, 24: 7760-7765. Winkler, H. and Carmichael, S. W. (1982) In “The Secretory Granule”. Ed. by A. M. Poisner and J. M. Trifaro, Elsevier Biomedical Press, Amsterdam, pp. 3-79. Winkler, H. and Smith, A. D. (1975) In “Handbook of Physiology”, American Physiological Society, 14 15 16 7) 18 19 20 Zi Chromogranin A in Sea Urchin Gametes Washington, D. C., Section 7, pp. 321-339. Rosas EullessAweltec IRE Weiss Zanini A... De Camilli, P. and Huttner, W. B. (1985) Secretogra- nins I and II: two tyrosine-sulfated secretory pro- teins common to a variety of cells secreting peptides by the regulated pathway. J. Cell Biol., 101: 1999- 2011. Schober, M., Fischer-Colbrie, R., Schmid, K. W., Busollati, G., O’Connor, D. T. and Winkler, H. (1987) Comparison of chromogranins A and B and secretogranin II in human adrenal medulla and phaeochromocytoma. Lab. Invest., 57: 385-391. O’Connor, D. T. (1983) Chromogranin: widespread immunoreactivity in polypeptide hormone produc- ing tissues and in serum. Regul. Pept., 6: 263-280. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacterio- phage. Nature, 227: 680-685. Burnett, W. W. (1981) “Western blotting”: Elec- trophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified ni- trocellulose and radiographic detection with anti- body and _ radioiodinated protein A. Anal. Biochem., 112: 195-203. Bradford, 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. Iacangelo, A., Affolter, H.-U., Eiden, L. E., Her- bert, E. and Grimes, M. (1986) Bovine chromogra- nin A sequence and distribution of its messenger RNA in endocrine tissues. Nature, 323: 82-86. Reiffen, F. U. and Gratzl, M. (1986) Chromogra- 22 23 24 25 26 Za]) 28 29 335) nins, widespread in endocrine and nervous tissue, bind Ca**. FEBS Lett., 195: 327-330. Mizuno, K., Miyata, A., Kangawa, K. and Matsuo, H. (1982) A unique proenkephalin-converting en- zyme purified from bovine adrenal chromaffin gran- ules. Biochem. Biophys. Res. Commun., 108: 1235- 1242. Hook, V. Y. and Eiden, L. E. (1984) Two pepti- dases that convert '*°I-Lys-Arg-(Met)enkephalin and '*°I-(Met)enkephalin-Arg® respectively, to '*°I- (Met)enkephalin in bovine adrenal medullary chro- maffin granules. FEBS Lett., 172: 212-218. Falkensammer, G., Fischer-Colbrie, R. and Wink- ler, H. (1985) Biogenesis of chromaffin granules: incorporation of sulfate into chromogranin B and into a proteoglycan. J. Neurochem., 45: 1475-1480. Wohlfarter, T., Fischer-Colbrie, R., Hogue- Angeletti, R., Eiden, L. E. and Winkler, H. (1988) Processing of chromogranin A within chromaffin granules starts at C- and N-terminal cleavage sites. FEBS Lett., 231: 67-70. Gustafson, T., Lundgren, B. and Treufeldt, R. (1972) Serotonin and contractile activity in the echinopluteus. A study of the cellular basis of larval behavior. Exp. Cell Res., 72: 115-139. Ryberg, E. (1973) The localization of cholinester- ases and non-specific esterases in the echinopluteus. Zool. Scripta, 2: 163-170. Ryberg, E. (1974) The localization of biogenic amines in the echinopluteus. Acta Zool., 55: 179- 189. Ryberg, E. (1977) The nervous system of the early echinopluteus. Cell Tiss. Res., 179: 157-167. | ae a 9 Raat r of? | ms ele ne aa “inane: ite in ean AT HAD RE east wee :: 2 | aha ieee tine cs pecans — k is ; a Ligon f et, nrena tf ee oe ait ed blahuerS Mice ee ah Ag vii Pet aGy . alk 3 if faerie sf shed all c , BY: =o Ce at b> sit * or RNS 9 il ae ‘Fabnotiniteb tats sche iter tore Aan Raae) i Deas fa ¢ 1 shai prcen Fj 2 rl te ay a nt ; age ae as ul ue re 8 a : el eX be f re ‘ey . j i r Re ete c - t ‘ , » at ‘ a 7 * 1 ff oi - = " ~ i ; pers % 3 S Sf aye 1 hae 7 be ° pn ) ™ = ; ~ “ ; & L ° 1 = > | : ; = e = J A ¢ helt 5 s in ‘yf ib Et ‘, ~ fa _ re 4 ; ‘< 5 i 3) be: * ! , f ee & ) “ 4 ' . - - a = »), ; ; . i ; eet is : Pairs 4 7 1 ri S re : My * ; ES ) j y >. i} ~ ’ } u “) h te = + Pi c ; \ 7 ; ) ‘ r "a s aT t i 5 Ss Li oe es 4 H 1 a x — im, i < i. i i 4 * y 2 ; a i 1 1 4 t a ~ ‘ ry i He NL i , ” ‘ t we \ A \ ~ - 3 ; 7 ZOOLOGICAL SCIENCE 9: 337-342 (1992) Changes in Timing and Site of Appearance of a Protease | in Xenopus Embryos SHOHEI Mriyata!, YAsuo NISHIBE*, MICHIKO SENDAI’, ISAO KATAYAMA‘, TERUHIKO ItNo‘* and Hirozi K. Kruara! ‘Laboratory of Research for Biosynthesis and Metabolism, Keio University, School of Medicine, Tokyo, 160, *First Department of Pathology, Saitama Medical School, Saitama, 350-04 and Department of *Chemistry, and *General Education, College of Humanities and Science, Nihon University, Tokyo 156, Japan ABSTRACT—A thiol protease was purified from embryos of Xenopus laevis. This protease has a relative mass (M,) of 43 k-44 K. Antiserum raised against the protease was used for analysis by Western blotting of proteins from embryos at various stages and from adult liver and heart. A band corresponding to a protein with an M, of approxmately 44k was detected in the unfertilized eggs and embryos with the monospecific antiserum. The size of the protein that reacted with the antiserum in a preparation of proteins from the liver of Xenopus laevis was different from that of the proteins from the eggs and embryos (M,, approximately 70k). Immunohistochemical localization with the antiserum revealed that the protease was more abundant in the animal hemisphere and in the cells derived from the animal hemisphere than in other cells of the embryo. The protease was highly enriched in the cytoplasm of ectodermal cells than the cytoplasm of endodermal cell. When proteins from Xenopus © 1992 Zoological Society of Japan embryos were used as substrate, one protein having M, of 31 K was mainly digested. INTRODUCTION Animal cells contain many different proteases and there are both lysosomal and non-lysosomal pathways of protein degradation [1-3]. We have purified an acidic protease from Xenopus embryos. The protease is sensitive to antipain, leupeptin and iodoacetic acid but it is insensitive to phenyl- methylsulfonyl fluoride and pepstatin [4]. It has an M, of 43 k-44k and can be dissociated into two subunits with M, of about 30k and 13k. The proteolytic activity is activated by nucleic acid. The effect of inhibitors, molecular weight, subunit structure, pH optimum, and phenomenon of activation by nucleic acid seem to be different from those of the proteases described thus far [1-3]. The purpose of the present study was to monitor temporal changes in the level, the distribution and the intracellular localization of the protease pro- tein in Xenopus eggs and embryos. Accepted December 24, 1991 Received October 26, 1991 MATERIALS AND METHODS Assay of proteolytic activity The proteolytic activity was determined with [SH]BSA as substrate and was assayed by measur- ing the acid-soluble radioactivity released from [H]BSA, as reported previously [4]. Preparation of embryos and purification of the protease Xenopus eggs were allowed to develop to the tail-bud stage. The embryos were immersed in cold acetone and homogenized in a. glass homogenizer. The homogenate was centrifuged at 3,000 rpm for 10 min at 4°C. The pellets were dehydrated by treatment with several changes of acetone, dried under No, and stored at —20°C. The procedure for purification of the protease has been described previously [4]. Twenty grams of the acetone-dried embryos were homogenized in 200 ml of 0.1 M acetate buffer, pH 5.0, which contained 0.1% Triton X-100 and 0.1mM EDTA, 338 S. MryaTta, Y. NISHIBE et al. in a glass homogenizer. The homogenate was centrifuged at 7,000 x g for 20 min. The proteoly- tic activity in the supernatant was concentrated by fractionation with acetone (20-50%). The result- ing precipitate was dissolved in 0.1M acetate buffer, pH 5.0, that contained 0.2 M NaCl. The extract was dialyzed against the same buffer and succesively chromatographed on columns of Sephadex G-75, CM-cellulose, and hydroxylapa- tite. The final preparation of enzyme represented a 16,500-fold purification. Production of antiserum against the protease New Zealand White rabbits were immunized with the protease. One hundred micrograms of the protease, emulsified in Freund’s complete ad- juvant (GIBCO), were administered intradermal- ly. Rabbits were given three booster injections intradermally with 100 ug protease without ad- juvant. Blood was collected 1 week later. IgG was purified from rabbit serum by chromatography on columns of DEAE-cellulose and Sephadex G-200, with subsequent precipitaion with 33% ammonium sulfate at 4°C. Immunoprecipitation of the protease Various amounts of antiserum or control serum were added to 1501 of PBS that contained purified enzyme (0.22 ug). After incubation for 48 hr at 4°C, the mixture was centrifuged in a micro- fuge for 10min. The residual activity in the supernatant was determined with [>H]BSA as sub- strate. Western Blot Analysis Proteins from embryos at various stages were fractionated by electrophoresis on 15% polyacryl- amide gels [5] and then transferred to nitrocellu- lose membranes in a buffer that contained 62.5 mM Tris, 192mM glycine (pH8.7) and 20% methanol [6], with a current of 1.2 mA/cm/? for 90 min at room temperature. The membranes were incubated with blot buffer (5% (mass/vol.) nonfat dry milk in PBS) for 2 hr and then with antiserum diluted 1:200 in blot buffer for Ihr. After washing with PBS, each membrane was exposed to anti- rabbit IgG conjugated with peroxidase in blot buffer for 1 hr, washed three times with PBS, and then developed with a solution of the substrate for peroxidase which consisted of azino-di (ethyl- benzthiazoline) sulfonic acid in 0.1 M citrate buf- fer, pH 4.2, supplemented with 0.03% hydrogen peroxide. The enzymatic reaction was terminated by washing with water. Proteins in gels were detected by silver staining [7]. Light microscopy Eggs and embryos were fixed in 2% paraform- aldehyde and 2.5% glutaraldehyde in 10 mM phos- phate buffer pH 7.5, for 4 hr at 4°C, washed thrice with PBS, dehydrated in a graded ethanol series, cleared in 0.03% Nonidet p-40 in chloroform, and embedded in Tissue Prep (Fischer Scientific). Thin sections were cut at 4 wm and deparaffinized with xylene. Immunohistochemical staining was carried out by modified version of the procedure of Wein- man et al. [8]. The sections were rinsed thrice in PBS, and treated with 0.02 M glycine in PBS and with 4% BSA in PBS. These were first labeled with antiserum against the protease diluted 1 : 20 in 1% BSA in PBS, washed 4 times with PBS, post-labeled with a biotin-labeled anti-rabbit IgG diluted 1:50 in 1% BSA in PBS, and washed thrice with PBS. The sections were reacted with avidin- biotin-peroxidase complex, washed thrice with PBS, reacted with 0.02% hydrogen peroxidase and diaminobenzidine tetrahydrochloride in PBS for visualization, and rinsed 4 times in distilled water. RESULTS Immunoprecipitation of the protease To study the characteristics of the antiserum, immunoprecipitation of the protease was carried out. The enzymatic activity remaining in the supernatant was measured with [(SH]BSA as sub- strate after addition of aliquots of antiserum or control serum (Fig. 1). The addition of increasing amounts of antiserum led to the gradual loss of proteolytic activity, demonstrating that the anti- serum reacted with the active enzyme protein. Screening with antiserum against the protease by immunoblotting Samples of the proteins from embryos at various Protease in Xenopus Embryo 339 100 50 PROTEASE ACTIVITY(%o) 0 Ome 20 40 60 ~ 120 1gG(g/0-2m1) Fic. 1. Immunoprecipitaiton of the protease. Various amounts of antiserum or control serum were addedd to 0.01 M phosphate-buffered saline that contained purified enzyme. Enzymatic activity remaining in the supernatant was measured after centrifugation, by measuring of acid-soluble radioactivity released from (PH]BSA. @, Antiserum; ©, control serum. stages and from the liver of adult frogs were fractionated by gel electrophoresis, transferred to nitrocellulose membranes, and subjected to im- munoblotting with antiserum (Fig.2). The purified enzyme gave a single band after elec- trophoresis in a 15% polyacrylamide gel (Fig. 2, lane 1). The relative mass (M,) of the enzyme was estimated to be about 44k. On screening with antiserum of proteins from unfertilized eggs and from embryos at the morula, gastrula, neurula and tail-bud stages, a band corresponding to an M, of about 44k was visualized from embryos at all stages (Fig. 2, lanes 2-6). When control serum was employed no bands were detected from embryos at any stage or from the adult liver (data not shown). The band corresponding to an M, of 44k coincided in terms of size with the purified protease detected by silver staining. The resulted indicate that the antiserum was monospecific for the protease protein from embryos. The relative intensity of the bands was almost constant among the proteins from unfertilized eggs and from lei te eGyehaee] G | | | » —67K i. @ @ —45K —25K Fic. 2. Screening with antiserum for protease in embryos by immunoblotting. Purified enzyme protein was subjected to elec- trophoresis in a 15% polyacrylamide gel and de- tected by silver staining (Lane 1). Proteins (20 ug) from unfertilized eggs, and from embryos at the morula, gastrula, neurula and tail bud stages (Lanes 2-6) and proteins (20 ug) from liver (lane 7) and heart (lane 8) of Xenopus laevis were fractionated in a 15% polyacrylamide gel and transferred to a nitrocellulose membrane. Protein was visualized by treatment with antiserum. lane 2, unfertilized egg; lane 3, embryo at morula stage; lane 4, embryo at gastrula stage; lane 5, embryo at neurula stage; lane 6; embryo at tail bud stage; lane 7, liver of Xenopus; lane 8, heart of Xenopus. Standard proteins used were bovine serum albumin (67 k), hen egg albumin (45 k) and chymotrypsinogen A (25 k). embryos at the morula, gastrula, neurula and tail-bud stages. The protease in early embryos was already present as a maternal proteins in unfertil- ized eggs. When a sample of the proteins from the liver of Xenopus laevis was immunoblotted, a protein with M, of 70k was visualized by treatment with anti- serum, but the band corresponding to an M, of 44 k was not observed (Fig. 2, lane 7). Also, the protein of 44k was not detected in the proteins 340 S. Miyata, Y. NISHIBE ef al. from the heart by immunoblotting with the anti- serum, while the protein of 70 k was detected in heart tissue (Fig. 2, lane 8). Immunohistochemical analysis of the distribution of the protease during development The distribution of the protease was examined in Fic. 3. Immunohistochemical localization of the protease in dissected eggs and embryos. All sections were cut along the animal-vegetal axis and stained with antiserum (A-D, I and J) or control serum (E-H). A and E, Unfertilized egg; B and F, embryo at morula stage; C and G, embryo at gastrula stage; D and H, embryo at neurula stage; I, higher magnification of embryo at gastrula stage; J, higer magnification of embryo at neurula stage. Abbreviations are as follows: ap, animal pole; ec, ectoderm; en, endoderm; m, mesoderm; nt, notochord; s, somites; vp, vegetal pole. Bar length: A-H, 0.25 mm; I and J, 0.025 mm. Protease in Xenopus Embryo 341 sections of eggs and of embryos at various stages by immunostaining with the antiserum. The pro- tease was detected as a dark deposit after immuno- histochemical staining. Sections were cut along the animal-vegetal axis. In the unfertilized egg, the animal region showed strong immunoreactivity for the protease antigen (Fig. 3A). The animal region at the morula stage also contained a higher concen- tration of the protease than did the vegetal region (Fig. 3B). Immunohistochemical localization of the protease antigen in the gastrula is shown in Figure 3C. The ectoderm was stained more strongly than the endoderm. In the embryos at the neurula stage, the intensity of the staining de- creased from the ectoderm, through the mesoderm, to the endoderm (Fig. 3D). The outer layer of cells of the ectoderm was most strongly stained and cells derived from the dorsal mesoderm, such as the notochord and somites, also stained with higher intensity than the en- doderm. The animal portion of the egg develops into ectodermal tissue, while an explant of the vegetal portion forms endodermal tissue. Furth- ermore, when animal pole cells are put in contact with vegetal pole cells, the fate of the animal pole cells changes and they form mesoderm tissue [9, 10]. Thus, in our experiments, the ectoderm and mesoderm stain more heavity than the endoderm as a result of the fact that the animal region stains more heavily stronger than the vegetal region. In sections treated with control serum, no differences among the regions were detected (Fig. 3E-3H). To define the intracellurar localization of the protease, magnified photographs of sections at the gastrula and neurula stages were examined (Fig. 31 and J). The cytoplasm without yolk granules or particles in the ectodermal cells at the gastrula stage was stained more darkly than the cytoplasm in the endodermal cell (Fig. 31). The cytoplasm without particles in the ectodermal cells at the neurula stage was also stained as a dark deposit (Fig. 3J). But the endodermal cells were weakly stained. Digestion of embryo proteins The activity of the protease with embryo pro- teins as substrate is shown in Figure 4. Embryo proteins were treated with the enzyme, then the Fic. 4. Digestion of embryo proteins Embryos at the neurula stage were homogenized in 0.1M acetate buffer at pH5.0. The homogenate was centrifuged in a microfuge and the supernatant was used as substrate. The embryo proteins (40 7g) were treated with the enzyme (0.5 yg), in 0.1M acetate buffer at pH 3.8. extent of digestion of the embryo proteins was assessed by SDS-polyacrylamide gel elec- trophoresis. When embryo proteins were digested in the 0.1M acetate buffer at pH 3.8, a protein with M, of 31 k was the most susceptible to diges- tion (Fig. 4, lane 2) (Fig. 4, lane 1, enzyme-free). DISCUSSION The developmental profile of a novel thiol pro- tease in Xenopus was monitored using and anti- serum raised against the protease. On screening with antiserum by immunoblotting, the protein that reacted with the antiserum was detected among the proteins from the unfertilized egg. The concentration of protease was almost the same among the proteins from unfertilized eggs as it was among proteins from embryos at the morula, gas- trula, neurula and tail-bud stages. However, we have already reported that the proteolytic activity is barely detectable in the unfertilized egg, when assays of thiol protease sensitive to antipain is performed in the presence of DTT [11]. Thus, the proteolytic activity seems to be inhibited by some unidentified mechanism in the unfertilized egg. The protease with M, of 44 k was not detected in adult tissues, and a protein of 70 k was detected in 342 S. Miyata, Y. NISHIBE et al. these tissues. The protease with M, of 44k appears to exist only in the embryo. The distribution of the protease was studied in embryos at various stages by immunohistochem- ical methods. The protease was present at high leveis in the animal region of the unfertilized egg and at the morula stage, and in cells of the ectodermal lineage in the early embryo. Further- more, the protease was mainly detected through- out the cytoplasm without yolk granules or parti- cles in ectodermal cells at the gastrula and neurula stage. It has been reported that oocytes and unfertilized eggs have lysosome-like organelles. These lysosomes appear to be located in a peripheral zone of the cytoplasm [12, 13]. Thus, this protease seems not to be localized in the lysosome. Many non-lysosomal proteases have been purified and characterized [1-3]. We reported previously that the proteolytic activity which is sensitive to the protease inhibition, antipain in- creases during early development and antipain inhibits the incorporation of [°H] uridine into RAN in Xenopus embryos [11, 14]. This study was carried out by antiserum raised against the anti- pain sensitive protease. Following fertilization, the synthesis of proteins, DNA replication and cell division begin [15, 16]. Synthesis of RNA becomes detectable at the midblastula stage [16-18]. This protease may function as one of components re- quired for maintenance of high rates of the synthe- sis of RNA, proteins or DNA in the early embryo. ACKNOWLEDGMENTS We are grateful for the skillful technical assistance in light microscopy of Mr. Kouji Ootomo of Saitama Medical School. Our thanks are also due to Mr. Hideo Kanei of Keio University for his able assistance. REFERENCES 1 Hershko, A and Ciechanover, A. (1982) Mecha- nisms of intracellular protein breakdown Ann. Rev. Biochem. 51: 335-364. 2 Pontremoli, S. and Melloni, E. (1986) Extralyso- somal protein degradation Ann. Rev. Biochem. 55: 455-481. 3 Bond, J. S. and Butler, P. E. (1987) Intracellular proteases Ann. Rev. Biochem. 56: 333-364. 4 10 11 ie 13 14 15 16 17 18 Miyata, S., Yoshida, Y and Kihara, H. K. (1989) Purification and characterizaiton of a protease from Xenopus embryos Eur. J. Biochem. 186: 49-54. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacterio- phage T 4 Nature 227: 680-685. Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacryl- amide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76: 4350-4354. Morrissey, J. H. (1981) Silver stain for proteins in polyacrylamide gels: A modified procedure with enhanced uniform sensitivity Anal. Biochem 117: 307-310. Weinman, S., Ores-Carton, C. Rainteau, D. and Puszkin, S. (1986) Immunoelectron microscopic localization of calmodulin and phospholipase A 2 in spermatozoa 1. J Histochem. Cytochem. 34: 1171 - KI). Nieuwkoop, P. D. (1969) The formation of mesoderm in urodelan amphibians. 1. Induction by the endoderm. Wilheim Roux’ Arch. 162: 341-374. Nakamura, O, Takasaki, H. and Mizohata, T. (1970) Differentiation during cleavage in Xenopus laevis. Acquision of self-differentiation capacity of the dorsal marginal zone. 1. Proc. Japan Acad. 46: 694-699. Miyata, S. and Kihara, H. K. (1988) Charactariza- tion of proteolytic activities in embryos of Xenopus laevis. Comp. Biochem. Physiol. 91B: 651-656. Busson-Mabillot, S. (1984) Endosomes transfer yolk proteins to lysosomes in the vitellogenic oocyte of the trout. Biol. Cell. 51: 53-66. Wall, D. A. and Meleka, I. (1985) An unusual lysosome compartment involved in vitellogenin en- docytosis by Xenopus oocytes. J. Cell Biol. 101: 1651-1664. Miyata, S., Shimazaki, T., Okamoto, Y., Motegi, N., Kitagawa, M. and Kihara, H. K. (1988) Inhibi- tion of RNA synthesis in embryo of Xenopus laevis by protease inhibitor. J. Exp. Zool. 246: 150-155. Ecker, R. E. and Smith, L. D. (1971) The nature and fate of Rana pipiens proteins synthesis during maturaiton and early cleavage. Dev. Biol. 24: 559- 576. Gurdon, J. B. (1974) The control of gene expression in animal development. Clarendon Press, Oxford. Shiokawa, K., Misumi, Y. and Yamana, K. (1981) Demonstration of rRNA synthesis in pre-gastular embryo of Xenopus laevis. Develop., Growth and Differ. 23: 579-587. Newport, J. and Kirschner, M. (1982) A major developmental transition in early Xenopus embryos: 1. Characterization and timing of cellular changes at the midblastura stage Cell. 30: 675-686. ZOOLOGICAL SCIENCE 9: 343-347 (1992) © 1992 Zoological Society of Japan Occurrence of a Novel 350-kDa Serine Proteinase in the Fluid of Porcine Ovarian Follicles and Its Increase during Their Maturation TAKAYUKI TAKAHASHI, YuICHI TsucHIYA!, YOSHIAKI TAMANOUE!, TaKAO Morr’, SENCHIRO KAWASHIMA’, and KENJI TAKAHASHI! Department of Biophysics and Biochemistry’ and Zoological Institute’, Faculty of Science, The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan ABSTRACT—Porcine ovary was found to contain enzyme activities hydrolyzing peptide 4- methylcoumaryl-7-amide (MCA) substrates with a preference for Arg-MCA bond. The activities were shown to be present almost exclusively in the follicular fluid and to increase several times during follicular maturation. The enzyme responsible for these activities is thought to be a serine proteinase as judged from its strong inhibition by diisopropylfluorophosphate (DFP), leupeptin and antipain. The molecular weight of the native enzyme was electrophoretically estimated to be approximately 350,000, the result indicating that the enzyme is clearly distinct from plasmin (M,=80,000) and collagenase (M,= 30,000—-65 ,000), both of which are thought to be involved in ovulatory process. The substrate specificity of the partially purified enzyme was qualitatively different from that of plasmin. These results suggest that the enzyme is a novel type of serine proteinase. INTRODUCTION The mature ovarian follicle in mammals contains fluid in the follicular space. This follicular fluid is known to consist mainly of transudates of plasma, although it also contains secretory products from follicle cells [1]. Special attention has been paid for proteolytic activities of the fluid in connection with the ovulatory process which is accompanied by drastic degradative changes leading to follicular rupture [2, 3]. At present, collagenase (3, 4] and plasmin [5-8], which is generated from plasmi- nogen by the action of tissue plasminogen activa- tor, are generally thought to be involved in this process. In an attempt to elucidate the functional role of proteolytic enzymes in mammalian reproductive organs, we happened to find a novel proteinase present in the follicular fluid of porcine ovary, preferable hydrolyzing peptide 4-methylcoumary]- 7-amide (MCA) substrates at Arg-MCA bond. Accepted December 20, 1991 Received November 20, 1991 The activity of this proteinase in the fluid was found to increase with follicular maturation. The enzyme is a proteinase with a molecular weight of about 350,000 and is clearly distinct from collage- nase and plasmin. MATERIALS AND METHODS Follicular fluid preparation Porcine ovaries were obtained from Teikoku Hormone Manufacturing Co. (Tokyo, Japan), usually within 4 hours after the animals were slaughtered. Follicular fluid specimens were obtained by aspiration from various stages of folli- cles using a 21G hypodermic needle and syringe. The samples were centrifuged at 1,000Xg for 10 min, and the supernatants were used. Distribution of follicular fluid enzyme in ovary Two ovaries were separately placed in a Petri dish containing 5 vol. of cold phosphate-buffered saline (PBS) and gently sliced with a sharp razor blade. The whole materials were collected and 344 T. TAKAHASHI, Y. centrifuged at 1,000 g for 10min. The precipi- tated tissue was gently suspended in 5 vol. of PBS, and the suspension was centrifuged again at 1,000 xg for 10 min. Both supernatants were combined, and this fraction was referred to as “follicular fluid”. The tissue fraction was minced with scissors and homogenized in Svol. of PBS, and the homogenate was centrifuged at 12,000Xg for 20 min. The resulting supernatant was referred to as “tissue extract”. Enzyme and protein assays Peptide substrates containing MCA _ were obtained from Peptide Institute (Osaka, Japan). Activities were assayed as described previously [9]. Unless otherwise stated, enzyme reactions were conducted in 0.1 M Tris-HCl (pH 8.0) containing 0.1mM_ t-butyloxycarbonyl (Boc)-Gln-Arg-Arg- MCA and an appropriate amount of sample. En- zyme activity was expressed as the amout of 7- amino-4-methylcoumarin released at 37°C per min. Protein was determined by the method of Smith et al. [10] using the BCA reagent (Pierce). Electrophoresis Polyacrylamide gel electrophoresis (PAGE) in the absence of sodium dodecyl] sulfate (SDS) was performed according to Laemmli [11] with PAG plate 4/15 gradient gel (Daiichi Pure Chemicals Co., Tokyo). Partial purification of the follicular fluid enzyme Pig ovary follicular fluid (120 wl) was mixed with an equal volume of the SDS-free sample solvent TABLE 1. Total activity TSUCHIYA et al. for PAGE, and the mixture (20 1) was applied on 12 separate wells. After the run at 4°C, the gel was cut into slices of 2.5mm width. Each slice was immersed in 2 ml of 0.1M Tris-HCl buffer (pH 8.0), crushed and left at 4°C overnight. Aliquots of extracts were assayed for enzyme activity. An extract having the highest activity was used as the partially purified enzyme preparation. The sample thus obtained had a specific acitivity of 41.2 nmol of 7-amino-4-methylcoumarin/min/mg protein (Boc-GIn-Arg-Arg-MCA as substrate), which was approximately 64 times greater than that of the crude fluid. RESULTS We have found that the crude extract of porcine Ovary contains enzyme activities hydrolyzing synthetic, arginine-containing peptide amide sub- strates. The distribution of the activity in this organ was examined using Boc-Gln-Arg-Arg-MCA as a substrate, and the results are shown in Table I. More than 90% of the total activity was recovered from the follicular fluid, and its specific activity was much higher than that of the tissue extract. The results indicate that the enzyme exists almost ex- clusively in the follicular fluid. Figure 1 shows the increase in the activity as the follicles undergo maturation. The specific activ- ities in the follicular fluid preparation derived from follicles with diameters ranging 1-2 mm and 4-5 mm were 0.28 and 0.64 nmol/min/mg protein, respectively. In parallel experiments, the activities toward Ala-MCA (a substrate for aminopeptidase M), Gly-Pro-MCA (asub-strate for dipeptidyl pep- Enzyme activity in the tissue extract and follicular fluid of porcine ovary Total protein Specific activity (nmol/min) (mg) (nmol/min/mg protein) Exp. 1 Tissue extract 4.90 45.0 0.11 Follicular fluid Sot 108 0.48 J559D, 22 Tissue extract Le 47.8 0.05 Follicular fluid 75.9 124 0.61 The weights of porcine ovaries used were 3.25 g in Exp. J and 3.10 g in Exp. 2 to obtain the tissue extract and follicular fluid fractions. Detailed procedures are given in MATERIALS AND METHODS. A Novel Follicular Fluid Proteinase 345 ~ 90 ie LAD O = 20310 > S 70) <_ E E 10 N Cc WwW 9 SZ 22S Sa) Diameter of Follicles (mm) Fic. 1. Change in enzyme activity in follicular fluid preparations during maturation. The enzyme activity toward Boc-Gln-Arg-Arg- MCA was assayed with the fluids obtained from various stages of ovarian follicles. Follicles with a diameter of 4-5 mm represent those which have almost fully matured. tidase IV) and N-succinyl-Gly-Pro-MCA (a sub- strate for prolyl endopeptidase) were also found to be detectable in the fluid. Interestingly, a gradual decline was commonly seen with these enzyme activities during follicular maturation (data not shown). The results strongly suggest that the increase in the activity toward Boc-Gln-Arg-Arg- MCA is rather specific, and is presumably due to accumulation of the corresponding enzyme(s) within ovarian follicles during their maturation. The electrophoretic analysis of fluid proteins was carried out, and a typical result with the fluid obtained from follicles with a 4-5 mm diameter is shown in Fig. 2. A gel slice extract with the highest activity constituted 82% of the total activity. Fig- ure 2 also shows that the apparent molecular weight of the enzyme is approximately 350,000. When the fluids from less-matured follicles were analyzed under the same conditions, essentially the same patterns were obtained. Thus, we tenta- tively conclude that an enzyme(s) with M,= 350,000 is solely responsible for the activity in the fluid that increases during follicular maturation. Some properties were investigated using the partially purified enzyme sample. The enzyme AB 669 — ims 440 - wa ff 140- © . 67- —@ OL WORY OQ Om Enzyme Activity (n mol/min) Fic. 2. Electrophoretic analysis of follicular fluid pro- teins and the enzyme activity. The fluid preparation from follicles (4-5 mm in diameter) was separately applied on two well posi- tions of a gradient PAGE gel. The protein amounts loaded were 13 yg and 246yg. After elec- trophoresis at 4°C, a lane loaded with 13 yg protein was stained with Coomassie Brilliant Blue R-250 (lane B), while the other lane was sliced into pieces of 2.5 mm width for overnight extraction in 0.5 ml of 0.1M Tris-HCl (pH 8.0) at 4°C. Aliquots of the extracts were assayed for the enzyme activity toward Boc-Gln-Arg-Arg-MCA. The total enzyme activity in each gel slice extract is shown. Lane A shows the Separation of molecular weight marker proteins (Pharmacia): thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), lactate dehydrogenase (140 kDa), and albumin (67 kDa). activity was completely inhibited by DFP (1 mM), leupeptin (20 “M) and antipain (20 ~M), whereas E-64 (1mM), p-chloromercuribenzoate (0.23 mM), and o-phenanthroline (1 mM) were without effect. These findings are consistent with the classification of the enzyme as a serine proteinase. Table II shows the substrate specificity toward various peptide MCA substrates. The results with porcine plasmin are also included for comparison. The follicular enzyme well hydrolyzed Boc-Gin- Arg-Arg-MCA, Boc-Gln-Gly-Arg-MCA and Boc- Leu-Lys-Arg-MCA, and to some extent Boc-Val- Pro-Arg-MCA. Boc-Glu-Lys-Lys-MCA, a typical substrate for plasmin, was a poor substrate for the enzyme. The effects of the above proteinase inhibitors on the activities toward Boc-Gln-Gly- Arg-MCA, Boc-Leu-Lys-Arg-MCA and Boc-Val- Pro-Arg-MCA were also examined in order to 346 T. TAKAHASHI, Y. TSUCHIYA et al. TABLE 2. Substrate specificity of follicular fluid enzyme Follicular Substrate fluid enzyme eo) (%) Boc-Gln-Arg-Arg-MCA 100 100 Boc-Gln-Gly-Arg-MCA 98.1 9.8 Boc-Leu-Lys-Arg-MCA 2 42.6 Boc-Val-Pro-Arg-MCA 19.6 65.9 Boc-Glu-Lys-Lys-MCA ad) SEZ Partially. purified follicular fluid enzyme was obtained as described in MATERIALS AND METHODS. For comparison, porcine plasmin (Sigma) was also tested under the same conditions. The plasmin sample had a specific activity of 357 nmol/min/mg protein when assayed with Boc-Gln- Arg-Arg-MCA. The relative values of specific activity are shown. compare with those toward Boc-Gln-Arg-Arg- MCA. The effects of these inhibitors were very similar to those on the activity toward Boc-Gln- Arg-Arg-MCA (data not shown). Therefore, we presume that these activities are probably cataly- zed by a single enzyme, although involvement of several enzymes with very similar properties can- not be ruled out completely at present. The specificity of this follicular fluid enzyme is clearly different from the plasmin specificity. DISCUSSION The generation of proteolytic activity within the follicle was suggested years ago as a possible mechanism for degrading the follicular wall [1-3]. Espey et al. [3, 4] were able to detect a collagenoly- tic enzyme in the follicle. On the other hand, Beers et al. [5, 6] demonstrated that the proteinase plasmin is capable of weakening follicle wall strips in vitro. The two enzyme activities are known to increase in the follicle and reach a peak prior to ovulation. The enzyme described in this study is evidently distinct from these two proteinases as follows: First, as judged from the strong inhibition by DFP, the enzyme is thought to be a serine proteinase, thus being different from a metalloen- zyme collagenase. Secondly, the cleavage specific- ity of the enzyme is qualitatively different from that of a serine proteinase plasmin. Finally, it also differs from plasmin in that the molecular weight of the enzyme is approximately 350,000 while that of plasmin is 80,000. Furthermore, it must be noted that the molecular weight (350,000) is much larger than those observed so far among endo- proteinases. Thus, the present enzyme is thought to be a novel serine endoproteinase. The physiological role of this follicular fluid enzyme is not clear at present. However, our finding that the enzyme activity specifically in- creases in the fluid as the follicles grow suggests its biological importance in the events relating both with follicular maturation and ovulation. To better understand the detailed molecular and enzymatic properties as well as the physiological role of this enzyme, further studies are necessary including its complete purification and characterization. ACKNOWLEDGMENTS This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan. REFERENCES 1 McNatty, K. P. (1978) Follicular fluid. In “The Vertebrate Ovary”. Ed. by R. E. Jones, Plenum Press, New York, pp. 215-259. 2 Schochet, S. S. (1916) A suggestion as to the process of ovulation and ovarian cyst formation. Anat. Rec., 10: 447-457. 3. Espey, L. L. (1975) Evaluation of proteolytic activ- ity in mammalian ovulation. In “Proteases and Biological Control”. Ed. by E. Reich, D. B. Rifkin, and E. Shaw, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, pp. 767-776. 4 Espey, L. L. and Coons, P. J. (1976) Factors which influence ovulatory degradation of rabbit ovarian follicles. Biol. Reprod., 14: 233-245. 5 Beers, W. H. (1975) Follicular plasminogen and plasminogen activator and the effect of plasmin on ovarian follicle wall. Cell, 6: 379-386. 6 Beers, W. H., Strickland, S. and Reich, E. (1975) Ovarian plasminogen activator: Relationship to ovulation and hormonal regulation. Cell, 6: 387- 394. 7 Canipari, R. and Strickland, S. (1985) Plasminogen activator in the rat ovary: Production and gonado- tropin regulation of the enzyme in granulosa and thecal cells. J. Biol. Chem., 260: 5121-5125. 8 Liu, Y. X., Peng, X. R. and Ny, T. (1991) Tissue- A Novel Follicular Fluid Proteinase 347 ‘specific and time-coordinated hormone regulation of plasminogen-activator-inhibitor type I and tissue- type plasminogen activator in the rat ovary during gonadotropin-induced ovulation. Eur. J. Biochem., 195: 549-555. Yanagida, M., Tamanoue, Y., Sutoh, K., Taka- hashi, T. and Takahashi, K. (1991) Microsomal membrane-bound serine proteinase from rat liver: Partial purification and specificity toward arginyl 10 11 peptide bonds. Biomed. Res., 12: 113-120. Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H., Provenzano, M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J. and Klenk, D. C. (1985) Measurement of protein using bicinchoninic acid. Anal. Biochem., 150: 76-85. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacte- riophage T4. Nature, 227: 680-685. “Pits Aha | sag Te ee co a yg the) af ZOOLOGICAL SCIENCE 9: 349-356 (1992) Seasonal Changes in Humoral Immunity and Blood Thyroxine Levels in the Toad, Bufo regularis ABDEL HAKIM SAAD and WAGIH ALI Zoology Department, Faculty of Science, Cairo University, Cairo 12613, Egypt ABSTRACT—The present study was designed to explore the basis of seasonal changes in the humoral immune response of the toad, Bufo regularis. The results indicated that; (1) administration of 0.4 ml of 10% RRBC suspension during the breeding season and the time of burrowing for winter elicited a high titer of antibody (Ab) and vigorous rosette-forming cell (RFC) resoponse. In contrast, during summer active life period and hibernation, immune response was slow with a low titer of Ab and minimal number of RFC; (2) cold-acclimated toads (8-10°C; 3 weeks) provided a considerable amount of serum Ab in response to primary immunization with RRBC; despite that the magnitude and the kinetic of the response were significantly different from those of control toads; (3) fluctuation in thyroxine (T4) levels was in direct correlation with the immunological indices recorded for adult toads during the different periods of the year. Our results suggest that endogenous T, levels generally trigger B. regu/aris immune system with low levels causing a significant inhibition and with high levels causing a significant © 1992 Zoological Society of Japan stimulation of humoral immune response. INTRODUCTION Ectothermic vertebrates have a variable body temperature and each has its own thermal toler- ance range, in which its life process can normally operate. As environmental temperature changes with season, each species acclimates and responds adaptively to the thermal range [1-3]. The litera- ture dealing with seasonal rhythms in amphibian immunology appears to be scant. Special emphasis was focused on seasonal changes in the thymus architecture, in relation to the annual changes in animal activity [4-6]. That is, a maximum thymic weight is recorded during the summer active months. A marked involution occurs in winter, and is followed by gradual recovery of this organ, and this in turn is terminated in the mating season. Until recently little attention has been directed towards seasonal effects on immune response of amphibians. Although, most authors reported some tempera- ture-dependent immunodepression during winter, results were contradictory, and the causative Accepted October 31, 1991 Received August 10, 1991 agents poorly defined [for review see 1]. Recently, other authors have pointed out a possible role of endocrine modulators [5, 7]. Thyroid hormones, mainly thyroxine (T4), experienced seasonal varia- tions, dramatically affecting physiological activities in amphibians [8, 9]. Little is known, however, about their effects on the immune system. There- fore, the present study aimed at contributing in- formation about some biological variables such as season and temperature on the humoral immune response of the toad, Bufo regularis. The outcom- ing results were particularly discussed from the plausible role played by T, levels in immunity of toads. MATERIALS AND METHODS Toads Adult male and female toads, Bufo regularis, were collected from Abu Rowash area (Egypt). A total of 700 toads weighing between 18-20 g and of length 8-9.5cm long from snout to vent, were used in the present study. Toads were placed in glass aquaria with tap water. Every two days granulate trout feed were given ad libitum. Ani- 350 A. H. SAAD AND W. ALI mals were maintained in a sunny animals room under natural light and ambient temperature of 10-17°C in winter, 18-27°C in spring and autumn and 30-38°C in summer. The amphibian life cycle in Egypt was demarcated into: (1) the breeding season (April-May), (2) the summer active life (June-September), (3) the time of burrowing for winter (October-November), and (4) the hiberna- tion (December-February). Preparation of spleen cells Spleen was excised from individual toads and placed in separate petri-dishes in ice-cold buffered- amphibian saline (BAS), pH=7.2. Monocellular Suspensions were prepared, cells were washed twice by centrifugation at 150g for 5 min, and pellets were resuspended in known volumes of BAS. Lymphocytes were counted and their viabil- ity was assessed by the trypan blue exclusion. Antigen and immunization Blood was collected from at least two or three healthy animals [rats (RRBC) or guinea pigs GRBC)|] by either decapitaion or by heart punc- ture. Pooled blood was mixed with an equal volume of heparinized phosphate-buffered saline (PBS), pH=7.2, and centrifuged at 150g for 15 min. Cells were washed three times with ice- chilled PBS, pH=7.2. Toads were allowed to acclimate to ambient temperature in the labora- tory for a few days before immunization. Pre- liminary experiments were carried out to deter- mine the optimum dosage and effective route of immunization. From these experiments, 0.4 ml of 10% RRBC was found to be an optimum dosage which induced maximum antibody response when given intraperitoneally (i.p.). This immunization schedule was followed to study the kinetics of antibody response. Control unimmunized toads were injected 1.p. with 0.4ml PBS, pH=7.2 and included in each experiment. Bleeding and serum preparation Individual blood samples were allowed to clot for 2 hr at room temperature then overnight at 4°C. After centrifugation at 250g for 15 min, individual sera were immediately used or stored at 20°C. Immunized and unimmunized control sera were decomplemented at 56°C for 30 min before testing to eliminate spontaneous hemolytic factors and homologous complement activity. Immunocytoadherence assay (ICA) The immunocytoadherence (ICA) assay was essentially as described by Kidder et al. [10] with minor modification. Briefly, 1-2X10° viable spleen cells were separately mixed with 10x 10° RRBC in a final volume of 100 1 culture medium in a serological glass tube. Culture medium was prepared by mixing L-15 medium (GIBCO, Grand Island, N.Y. U.S.A.) with heat-inactivated fetal calf serum in the ratio of 9:1. The medium was found to be optimal for the viability of toad cells and the maintenance of RRBC intact. Tubes were incubated for 15 min at 37°C and then overnight at 4°C. After resuspending the reaction mixture by gentle rotation and adding one drop of trypan blue, ICA-positive cells (rosettes) and lympho- cytes were counted in a haemocytometer at a magnification of 250. Each tube was counted once or twice depending on the variation among them. An ICA-positive cell (rosette-forming cell or RFC) was defined as a “rosette” consisting of a spleen cell bearing at least three adherent RRBC. Multiple layered rosettes with lymphocytes seen clearly at the center were considered as positive. No dead cells were found to form rosette. Two trails of each sample tested were run and the data © were expressed as RFC/10° spleen cells. In order to determine the specificity of “rosette” formation, RRBC was replaced by GRBC in the assay system. Haemagglutination (HA) assay Circulating antibody (Ab) titers were deter- mined according to standard procedure in micro- titration plates as described in details previously [11]. Briefly, 100 41 two-fold diluted immune or control sera in PBS, pH=7.2 and 50 ul of 1% RRBC in PBS were mixed in 96 well of round- bottom microtiter plates (Nunc, Roskid, Den- mark). After gentle aspiration, plates were incu- bated at 4°C and read 24 hr later. Titers were then recorded as the log of the last well showing micros- copic agglutination, the first well had a final dilu- tion of 1:2. Control wells contained PBS and erythrocytes only. Thyroxine and Seasonal Immunity in Toads 35 Radioimmunoassay of serum T, Determination of T, was carried out by a modi- fication of the solid phase technique described in detail by Murphy ef al. [12]. The antiserum used was raised in rabbit against L-thyroxine and sup- plied by Kallested Laboratories (Texas, U.S.A.). This antiserum showed a nearly 100% crossreactiv- ity with Ty, 2.78% with triiodothyronine, but less than 0.01% with diiodothyronine as provided with the radioassay manufacture protocol. In this sys- tem, serum samples require neither extraction nor predilution. Serum aliquots (20 1) were pipetted directly to the bottom of the corresponding assay tubes. Two hundred microliters of '°I-T, were added to each tube and agitated on vortex mixer for one min. After one hr incubation at room temperature, separation of free T, from bound was accomplished simply by decanting the supernatant. The assay tubes were allowed to drain and then were inverted for 2 min on absorbent paper to shake off all the residual droplets. The tubes were thereafter counted for one min in gamma counter apparatus (Mini. Irst. Ltd., Essex, U.K.). Repli- cate analysis of a sample of toad serum gave intra-assay coefficient of variation of 2.6% and inter-assay coefficient of variation of 4.6%. Ali- quots of serum were assayed and the smallest amounts of Ty, statistically distinguishable from zero was 1.0 «g/dl. Statistical analysis Student’s t-test was used to determine levels of significance between control and experimental groups. Differences were considered to be signi- ficant when P values <0.05 were obtained. RESULTS Season-related differences in humoral immunity The objective of this experiment was to evaluate the humoral response of adult toads during their annual life cycle. Data of two seprate experiments performed in sequence for each period, were simi- lar, and therefore pooled (Figs. 1 and 2). As depicted in Fig. 1, during hibernation period, the number of RFC rose quickly with low level on day 4, decreased sharply by day 8 and remained at this level until day 16. In active life period, immunized toads exhibited a peak at day 4, then sharply declined at day 8 and remained constant during the following eight days. However, in the time of burrowing for winter, response was signi- ficantly vigorous. High level of RFC was detected on day 4 post-immunization. Thereafter, the num- ber of RFC began to decline but remained at high level up to day 16. In breeding season, the kinetics of RFC was similar to that demonstrated during the time of burrowing for winter. Indeed RFC were detectable at high level after day 4, peak was attained on day 12 followed by gradual decline (Fig. 1). The kinetics of anti-RRBC Ab response is de- picted in Fig. 2. During hibernation, Ab titer rose 70 Number Of RFC X10 lo” Cells 0) 8 12 16 > Days post immunization Fic. 1. Kinetics of rosette-forming cell (RFC) response in the spleen of adult toads, B. regularis. Animals were immunized on day 0 with 0.4 ml of 10% RRBC suspension in the breeding season (q—pP), the active summer life (C—O), the time of burrowing for winter (@—@) and the hibernation ( Days. post Fic. 2. Serum haemagglutinin titers in adult toads, B. regularis. Animals were immunized on day 0 with 0.4 ml of 10% RRBC suspension in the breeding season (q—p), the active summer life (O—©), the time of burrowing for winter (@—@) and the hibernation (). Each point represents the mean response of 3-5 separate animals, and the vertical bars indicate standard error of the mean. immubpization rapidly with low level of serum Ab activity on day 4, and almost remained constant until day 12. The titer increased to reach the low detectable level on day 16. In summer active life period, immunized toads exhibited a slight rise in serum HA titer at day 8 post-immunization and remained constant during the following eight days. However, in the time of burrowing for winter, Ab was first detected day 8 post-immunization and titer increased sharp- ly reaching a maximum level on day 16. In the breeding season, the kinetics of Ab response was quite similar to that detected during the time of burrowing for winter (Fig. 2). In summary, the results indicated that, during hibernation and summer active life period, hu- moral and cellular response was minimal with a small amount of RFC and low level of circulating Ab. In contrast, administration of RRBC (0.4 ml; 10% RRBC suspension) during the time of bur- rowing for winter and breeding season elicited a high titer of circulating Ab and vigorous FRC response. Effect of cold acclimation on humoral immunity The toads, field-collected in mid August, were immediately used for this experiment. About 25 toads were injected i.p. with 0.4 ml of 10% RRBC and maintained in the laboratory at ambient temperature (control group). Another group of toads was transferred to the dark incubator and kept without feeding at the temperature of 8-10°C for three weeks. Then, cold-acclimated toads were injected i.p. with 0.4ml of 10% RRBC (ex- perimental group). As depicted in Fig. 3, control toads exhibited a peak at day 4, sharply declined at day 8 and remained constant during the following days. 6 Number of RFC X 10" lov cells 10) 4 8 12 16 Days post immunization Fic. 3. Kinetics of RFC response in the spleen of adult toads, B. regularis. Fresh, field-collected (O—O) and cold-acclimated (@—@) animals were immu- nized on day 0 with 0.4 ml of 10% RRBC suspen- sion. Each point represents the mean response of 3— 5 separate animals, and the vertical bars indicate standard error of the mean. *=0.05 Days post immunization Fic. 4. Serum haemagglutinin titers in adult toads, B. regularis. Fresh, field-collected (C—©) and cold- acclimated (@—@) animals were immunized on day 0 with 0.4 ml of 10% RRBC suspension. Each point represents the mean response of 3-5 separate ani- mals, and the vertical bars indicate standard error of the mean. *=0.05< P<0.01, NS=not significant. MONTHS Fic. 5. Number of viable splenic lymphocytes obtained from female () and male (q—p) adult toads, B. regularis during the different months of the year. Each point on the curve expresses mean value of 10—15 separate animals and the vertical bars indicate standard error of the mean. 354 A. H. SAAD AND W. ALI (Pg/ mi!) 1200 oo fe) {e) oy ° fe) thyroxine (T,) levels Serum Mar Apr May Jun Jul fug MONTHS Fic. 6. The levels of the serum thyroxine of adult toads, B. regularis throughout the year. Each point on the curve represents the mean level of 8-10 samples and the vertical bars indicate standard error of the mean. trypan-blue exclusion test. Determinations were performed on healthy, fresh field-collected ani- mals. Three to five toads/sex groups were sac- rificed at 10-day intervals throughout the year. As depicted in Fig. 5, no statistically significant variation in the number of splenic lymphocytes could be observed between male and female toads. However, data obtained on toads exposed to nat- ural experimental conditions suggest that a season- al rhythm in lymphocyte distribution is clearly present. Viable splenic lymphocytes were more abandant in April-June and again in October- November. Minimal levels were observed during July-September and during December-March. Seasonal rhythms in endogenous Ty level The monthly changes in circulating T, levels were assayed at 10-day intervals throughout the year. Determination was performed on serum samples from healthy fresh, field-collected toads. It is worth mentioning that these toads were from the same batches used throughout the ex- perimental immunological studies. Preliminary determination failed to show sex-related differ- ences in T, levels between male and female toads, and therefore the results were pooled. As depicted in Fig. 6, T, exhibits monthly varia- tion between 320 and 1006 pg/ml throughout the year. From April to May, serum concentration was around 320-360 pg/ml and increases up to 940 +139 pg/ml in August. From August through October, the serum levels of T, exhibited a sudden decrease. T, leveis began to rise up to 933 +33 pg/ ml in November. From December, a precipitous decrease occurred in Ty level reaching a basal value of 540+ 12 pg/ml in February, followed by a significant increase in March around 780+ 43 pg/ ml. The present data indicated that serum Ty level was greatest in July-September (the active life period) and November (the time of burrowing for winter). Minimal levels were observed in January- February (the hibernation) and April-May (the breeding season). DISCUSSION The toad, B. regularis, is terrestrial, except for a short period during breeding season. The annual life cycle of the adult toads of this species consists of several distinct phases: the breeding season, active terrestrial life after breeding season, autum- Thyroxine and Seasonal Immunity in Toads 3)5)5) nal migration and hibernation [14]. Our results indicated that, administration of 0.4ml of 10% RRBC suspension during the breeding season and the time of burrowing for winter elicited high titer of circulating antibodies and vigorous RFC re- sponse. In contrast, during both the summer active life period and hibernation, the humoral response was slow, with a low titer of circulating antibodies and limited number of RFC. Our results suggesting that seasonal rhythms strongly influenced the kinetics have invited into a study whether, in B. regularis, splenic lymphoid com- partments are also affected by seasonal rhythms. Lymphocyte density changed in the spleen throughout the year. However, no sex-related difference was observed. Various factors, principally temperature, impli- cated in seasonal variations affected the immune reactivity of amphibians and other ectotherms [1- 3]. Immunosuppressive effects of winter were repeatedly related to low temperature. However, the environmental temperature failed to explain satisfactorily the differences in the immune re- sponse in the breeding season and summer active life reported herein for B. regularis, since the natural temperature is high in both periods. In- deed, the immune response was significantly differ- ent between animals collected during the time of burrowing for winter and hibernation; although ambient temperature is relatively low in Egypt in both periods. The data also indicated that cold- acclimated toads (8-10°C; 3 weeks) produced a considerable amount of serum antibodies in re- sponse to primary immunization with RRBC. In this respect, temperature was not the sole causa- tive factor that affected seasonal rhythms in the toad’s immune response. In agreement with our observations, Bigaj and Plytycz [4] have proved that experimental changes of the external tempera- ture, which demand unphysiological behaviour of experimental frogs, cannot change significantly the season-specific morphology (and probably the function) of their thymus glands. Therefore, other factors have been suggested as potential cause of seasonal changes affecting the immune system of amphibians. We think, how- ever, that numerous physiological parameters regu- late the seasonal variations of amphibian immune system. The correlation of the latter with T, seasonal fluctuation might attributed a role of the thyroid hormones in this scenario. The present study, performed to test this hypothesis, apparent- ly gave T, a relatively important role. Serum T, levels were rather low in the breeding season and hibernation period, while the time of burrowing for winter (only during November), prebreeding and active summer life period were associated with a remarkable elevation in endogenous T, levels. The moderate or profound decrease in lymphoid density and immune response observed during the prebreeding, active summer life and hibernation are in consequence to low endogenous T,4 level during these periods. On the other hand, enrich- ment of lymphoid cells and powerful immune response observed during the breeding and the time of burrowing for winter are in consequence to high endogenous T, levels. Informations concerning the influence of T, on the immune system of amphibians is scant. It was known that in amphibians T, plays a fundamental role in triggering metamorphosis itself and many, if not all, regressive and progressive changes [15, 16]. The levels of T3 and T, in the serum increased rapidly reaching their peaks when metamorphosis was at full swing [17]. Metamorphic events could be dissected by thyroid hormone concentration. However, no absolute threshold of induction exists and any concentration of Ty could induce anuran metamorphic changes [18]. Rivier and Cooper [19] reported that injection of excessive Ty into larvae of Rana catesbeiana resulted in regression of larval lymphoid organs. However, Bovbjerg [20] and Nagata [21] reported that modulation of T, levels during metamorphosis with thiourea or exogenous T, did not affect changes in allograft rejection. Since the inhibition of allograft rejection by meta- morphosis depends on the genetic relatedness of donor and host [22], it is difficult to interpret these findings, since genetically heterogeneous popula- tions were used. Moreover, the timing of hor- monal variation might be crucial. Lastly, in young adults of Xenopus laevis, 10’ M Ty in vitro could affect antigen recognition and binding capacity of SRBC-sensitized splenic lymphicytes [18]. Our present observation might help in interpret- ing the amphibian data by suggesting that T, will 356 differentially stimulate and/or inhibit the immune system, depending on the time duration during which this hormone is released. We think that the correlation between the immune system and T, levels in our present data is not a mere coincidence in time but rather one of the arms of a dynamic immunoendocrine mechanism at work in ectother- mic animals. 10 REFERENCES Wright, R. K. and Cooper, E. L. (1981) Tempera- ture effects on ectotherm immune response. Dev. Comp. Immunol., 5: 177-122. Zeeman, M. (1986) Modulation of the immune response in fish. Vet. Immunol. Immunopathol., 12: 235-241. El-Ridi, R., Zada, S., Afifi, A., El-Deeb, S., El- Rouby, S., Farag, M. and Saad, A-H. (1988) Cyclic changes in the differentiation of lymphoid cells in reptiles. Cell Different., 24: 1-12. Plytycz, B. and Bigaj, J. (1983) Amphibian lym- phoid organs: a review. Folia Biol., 31: 225-240. Plytycz, B. and Bigaj, J. (1983) Seasonal cyclic changes in the thymus gland of the adult frog, Rana temporaria. Thymus, 5: 327-344. Plytycz, B. and Bigaj, J. (1984) Endogenous rhythm in the thymus gland of Rana temporaria (Morpholo- gical study). Thymus, 56: 369-372. Zapata, A., Garrido, E., Leceta, J. and Gomariz, R. P. (1983) Relationships between neuroendocrine and immune system in amphibians and reptiles. Dev. Comp. Immunol., 7: 771-774. Kuhn, E. R., Geverts, H., Jacobs, G. and Van- dorpe, G. (1987) Reproductive cycle, thyroxine and corticosterone in females of the giant swamp frog, Dicroglossus occipitalis at the Equator. Gen. Comp. Endocrinol., 66: 137-144. Kuhn, E. R., Darras, V. M. and Verlinden, T. M. (1985) Annual variations of thyroid activity follow- ing thyrotropin stimulation and circulating levels of thyroid hormones in the frog, Rana ridibunda. Gen. Comp. Endocrinol., 57: 266-273. Kidder, G. M., Ruben, L. N. and Stevens, J. M. 11 i 13 14 IS 16 17 18 19 20 Zl 22 A. H. SAAD AND W. ALI (1973) Cytodynamic and ontogeny of the immune response of Xenopus laevis against sheep erythro- cytes. J. Embryol. Exp. Morph., 29: 78-85. Diener, E. and Marchalonis, J. (1970) Cellular and humoral aspects of the primary immune response of the toad, Bufo marinus. Immunology, 18: 279-293. Murphy, B. E. P., Patte, C. J. and Gold, A. (1966) Clinical evaluation of new methods for determina- tion of serum thyroxine. J. Clin. Endocrinol. and Metab., 26: 247-256. Ruben, L. N. (1975) Ontogeny, phylogeny and cellular cooperation. Am. Zool., 15: 93-106. Flower, S. (1933) Notes on the recent reptiles and amphibians of Egypt, with a list of the species recorded from that kingdom. Proc. Zool. Soc., part IV; 735-772. Etkins, W. (1935) The mechanisms of anuran meta- morphosis. I. Thyroxine concentrations and the metamorphic pattern. J. Exp. Zool., 71: 317-340. Kollros, J. J. (1961) Mechanisms of amphibian metamorphosis hormones. Am. Zool., 1: 107-114. White, B. A. and Nicoll, C. J. (1981) Hormonal control of amphibian metamorphosis. In “Meta- morphosis: A problem in developmental biology”. Ed. by L. I. Gilbert and E. Freiden, Plenum Press, New York, pp. 363-393. Ruben, L. N., Clothier, R. H., Murphy, Gale. Marshall, G. D., Lee, R., Phan, P., Nobis, C. and Shiigi, S. (1989) Thyroid function and immune reactivity during metamorphosis in Xenopus laevis, the South ae Clawed toad. Gen. Comp. En- docrinol., 76: 128-135. Riviera, uF 4 and Cooper, E. L. (1973) Thyroxine- induced regression of tadpole lymph glands. Proc. Soc. Exp. Biol. Med., 143: 320-322. Bovbjerg, A. M. (1966) Rejection of skin homo- grafts in larvae of Rana pipiens. J. Exp. Zool., 161: 69-80. Nagata, S. (1976) Immune responses against skin allograft and rabbit red blood cells in metamorpho- sis-inhibited Xenopus laevis. J. Fac. Sc. Hokkaido Univ., 20: 183-191. Barlow, E. H., Dimarzo, J. S. and Cohen, N. (1981) Prolonged survival of MHC-disparate skin allografts transplanted to the metamorphosing toad, Xenopus laevis. Transplantation, 32: 51-57. ZOOLOGICAL SCIENCE 9: 357-363 (1992) Effects of Unilateral and Bilateral Orchidectomy on Laterality of Neurons of the Preoptic Area and Plasma Levels of Gonadotropins and Testosterone in Male Mice TosHIvuKI S. NAKAZAWA, TAKEO Macuipa!” and SEICHIRO KAWASHIMA~ Zoological Institute, Faculty of Science, Hiroshima University, Hiroshima 730, Japan ABSTRACT—Possible involvement of direct neural connections between the testes and the hypothala- mus in the neuroendocrine control of gonadal function was studied in male mice. Two-month-old male mice of the CD-1 strain were orchidectomized either unilaterally or bilaterally and changes in the size of neuronal nuclei and cell bodies in the medial preoptic area (POA) and plasma concentrations of luteinizing hormone (LH), follicle-stimulating hormone (FSH) and testosterone were examined one week after operation. Both nuclei and cell bodies of POA neurons were significantly larger in the right side of the brain than in the left side. Either unilateral or bilateral orchidectomy failed to abolish the left-right difference. On the other hand, bilateral orchidectomy and removal of the left testis consistently caused a significant reduction in the size of neuronal nuclei on both sides of the POA, while removal of the right testis induced a significant decrease in the size of neuronal nuclei on the right side of the POA alone. By contrast, the cell body size of both sides of the POA was significantly larger in animals removed of their right testes than in those subjected to bilateral, left-sided or sham orchidectomy. Bilateral and unilateral orchidectomy caused a significant increase in plasma LH levels, but failed to show a significant elevation in plasma FSH levels. Removal of either side of the testis invariably caused an increase in plasma levels of testosterone. The present experiments clearly exhibited the left-right difference in the morphology of POA neurons in male mice. The results further suggest that the neural connections between the testes and the hypothalamus are involved in the control © 1992 Zoological Society of Japan of hypophyseal-gonadal functions in male mice. INTRODUCTION Evidence has been accumulated on the presence of asymmetry in the brain of mammals. Diamond et al. [1] have reported that in young adult male rats specific regions in the right cerebral cortex are significantly thicker than the corresponding re- gions on the left. Asymmetry in oxygen consump- tion, glucose uptake and tissue contents of nor- epinephrine and dopamine in the cortex and other Accepted November 13, 1991 Received September 6, 1991 " Present address: Department of Regulation Biology, Faculty of Science, Saitama University, Urawa 338, Japan. * Present address: Zoological Institute, Faculty of Sci- ence, University of Tokyo, Hongo, Tokyo 113, Japan. > To whom request of reprints should be addressed. discrete areas of the brain is well known [2-4]. Several authors have also demonstrated the left- right difference in the endocrine hypothalamus in rats. Gerendai et al. [5] found that the content of gonadotropin-releasing hormone (GnRH) in the right side of the medial basal hypothalamus was significantly higher than in the left side in adult female rats. Bakalkin et al. [6] similarly reported that the content of luteinizing hormone-releasing hormone (LHRH) in the region of the arcuate nucleus (ARC) and the ventromedial nucleus (VMN) of the hypothalamus was greater in the right side than in the left side of the body of male rats. Fukuda ef al. [7] further reported that the unilateral radio-frequency lesion to the right side of the medial anterior hypothalamus was effective in suppressing the ovarian compensatory hypertro- phy in female rats. All these findings strongly 358 T. S. NAKAZAWA, T. MACHIDA AND S. KAWASHIMA suggest the existence of hypothalamic laterality in the regulation of gonadotropic functions in the rat. In this connection, Dorner and Staudt [8, 9] found that neither bilateral orchidectomy nor the replace- ment therapy with testosterone caused any changes in the morphology of neurons of the VMN and the preoptic area (POA) of the diencephalon in male rats. Although the POA is of principal importance in the control of gonadotropin secre- tion [10-12], morphological or functional laterality of this brain structure has not yet been known. On the other hand, several authors have sug- gested that the neural mediation between the peripheral endocrine glands and the hypothalamus is involved in the regulation of neuroendocrine systems in rats [6, 13-17]. Gerendai and Halasz [14] found that unilateral ovariectomy resulted in an enhanced uptake of amino acids into the contra- lateral ARC in rats. Bakalkin et al. [6] reported a decrease in LHRH contents in the hypothalamus contralateral to the side of unilateral orchidectomy in rats. They suggested that the neural signal from the gonads to the hypothalamus was transmitted through crossed afferent nerves. On the contrary, Mizunuma ef al. [17] reported that the right-sided orchidectomy caused an increase in LHRH con- tent in the ipsilateral hypothalamus in rats. In the present experiments, in order to elucidate the laterality of the POA, male mice were orchidectomized either unilaterally or bilaterally and morphometrical changes in neurons of the POA and changes in plasma concentrations of gonadotropins and testosterone were investigated. MATERIALS AND METHODS Male mice of the CD-1 strain were used in the present study. They were housed in a tempera- ture-controlled (22+1°C) room under 12-hr light (0600-1800) and 12-hr dark cycle with free access to laboratory chow and tap water. At 60 days of age, a total of 32 male mice were divided into four groups; 8 animals each of the first two groups were given removal of the testis of either the left or the right side (groups L and R) and 8 mice of the third group were orchidectomized bilaterally (group B). Eight mice of the last group served as controls receiving operations of abdominal incision but their testes were not removed (group S). All the above operations were done under ether anesthe- sia from 0900 to 1100 in the morning. One week after operation, animals were weighed and im- mobilized. Blood was collected from the jugular vein into the heparinized syringe. The plasma was separated by centrifugation at 3000 rpm for 30 min at 4°C and stored at —20°C until assayed for gonadotropins and testosterone. Immediately af- ter the blood collection, brains were quickly re- moved and fixed in 10% formalin. Prostates, preputial glands and testes, if remaining, were also taken out, weighed and fixed in 10% formalin. Samplings of blood and tissues were done from 0900 to 1100 in the morning. Blocks of brain tissues containing the POA were embedded in paraplast. Sections were cut at 6 ~m in thickness and stained with thionin. In each animal, the sizes of neuronal nuclei and cell bodies were measured in 50 cells each in both sides of the POA. The maximum diameter (a) and the diam- eter perpendicular to it (b) of individual neuronal nuclei were measured under the microscope using a micrometer, and the area of neuronal nuclei was calculated according to the formula of the ellip- soid; A= zab/4. Profiles of neurons whose nuc- lear diameters were measured were traced using a camera lucida and the sizes of cell bodies of these neurons were measured with the aid of a tablet digitizer (MGC-1000, Mutoh, Japan). Neurons © thus measured were located in the central region of the POA confined vertically between the anterior commissure and the optic chiasm, and laterally between two perpendicular lines drawn to the inner margins of the right and the left bed nuclei of the anterior commissure. Plasma gonadotropins were measured by radioimmunoassay using a double antibody method [18, 19]. Highly purified luteininzing hor- mone (LH; NIADDK- rat LH-I-5) and follicle stimulating hormone (FSH; NIADDK-rat FSH-I- 4) were radioiodinated with ‘I (Na’"'I, Radiochemical Centre, Amersham) in the pres- ence of lactoperoxidase and hydrogen peroxide using the method described previously [20] for the assay of plasma LH and FSH. Concentrations of LH in 100 pl plasma and FSH in 50 yl plasma were expressed in terms of ng NIADDK-rat LH-RP-2 Castration and Laterality of Neurons and NIADDK-rat FSH-RP-2 per ml, respectively. Gonadotropins and antisera were kindly provided by NIADDK Rat Pituitary Hormone Distribution Program, NIH, Bethesda. We have previously confirmed that the rat assay system is satisfactory for the radioimmunoassay of LH and FSH in mouse plasma samples [21]. Plasma concentra- tions of testosterone were determined by radioim- munoassay using an antiserum to testosterone (HB-31, Teikoku Hormone MFG.) and [1, 2, 6, 7--H] testosterone (Radiochemical Center, Amer- sham) and expressed in terms of ng per ml plasma. Details of the radioimmunoassay of testosterone were described elslewhere [22]. Statistical analyses were performed by ANOvA and two-tailed Student’s ¢ test, or by Kruskal- Wallis test followed by Mann-Whitney’s U test for multiple groups. RESULTS Morphometrical changes in neurons of the POA following unilateral and bilateral orchidectomy Figure 1 shows the results of morphometry of POA neurons. The neuronal nuclei were signif- icantly larger in the right POA than in the left (P< 0.001 in group S). The neuronal nuclear size of the right POA was consistently greater than that of the left POA following left or right testis removal or bilateral orchidectomy (P<0.01 for groups L and R, and P<0.001 for group B, respectively). The cell bodies of neurons in the POA were similarly larger in the right side than in the left side, the difference between the left and right POA being statistically significant in all the groups S, R and B (P<0.001, in all comparisons). On the other hand, either bilateral or unilateral orchidectomy rendered the neuronal nuclei signif- icantly smaller in both the right and left sides of the POA (P< 0.001 for either side of group B and for the right side of groups L and R, and P< 0.005 for the left side of group L, as compared to the corresponding side of group S). Although statisti- cally not significant, the nuclei of POA neurons appeared to be larger in right-testis-removed ani- mals than in left-testis-removed mice. There were no significant differences in the size of cell bodies 359 ) o Size of neuronal nucleus (jm? size of cell body (um?) Fic. 1. Changes in the size of nuclei (upper panel) and cell bodies (lower panel) of POA neurons after unilateral and bilateral orchidectomy in mice. S: sham-orchidectomized animals, L: animals sub- jected to left-sided orchidectomy, R: animals sub- jected to right-sided orchidectomy, B: bilaterally orchidectomized animals. Open columns represent the left POA neurons and hatched columns show the right POA neurons. The number of animals used is 8 in each group. Asterisks exhibit that the differ- ence between the left and the right sides of the POA is statistically significant (P< 0.001 for double aster- isks and P<0.01 for single asterisk). Triangles represent the statistically significant deviations from the corresponding side of the POA in group S (P< 0.001 for double triangles and P<0.005 for single triangle). of POA neurons among groups of sham- orchidectomy, left-sided hemiorchidectomy and bilateral orchidectomy. However, the cell bodies in either side of POA neurons of right-testis- removed animals were significantly larger than those of sham-operated animals (P<0.005 for the 360 T. S. NAKAZAWA, T. MACHIDA AND S. KAWASHIMA right and left POA). The sizes of neuronal cell bodies of the POA in right-testis-removed animals were significantly greater than those in left-testis- removed mice (P< 0.005 for the left POA and P< 0.001 for the right POA). Changes in plasma levels of gonadotropins and testosterone following unilateral and _ bilateral orchidectomy Plasma levels of LH, FSH and testosterone are fold increment in plasma LH concentrations (P< 0.001). Unilateral removal of the left or the right testis also induced a significant increase in plasma LH levels (P<0.005 for the left and right testis removal). On the other hand, plasma FSH level was 24.9+2.8ng per ml in sham-operated con- trols. No significant changes were detected in plasma FSH levels one week after unilateral or bilateral orchidectomy. Plasma testosterone level in sham-operated con- trol mice was 1.10+0.26ng per ml. It slightly reduced after bilateral orchidectomy for a week, but the difference from the level of sham-operated shown in Figure 2. Plasma LH level in sham- operated male mice was 0.69+0.10 ng per ml. Bilateral orchidectomy for a week caused a five- 8 p : 40 = E ‘ pe A) ~ oO QO O a : : E dl : B = D = c gol & iS) ~~ a6 v—) as Ww) © 4 Le W) 2 @) @) 0 i Syl Beal Riel! >: Lee ReB Sib Reee Fic. 2. Changes in plasma levels of LH (left panel), FSH (middle panel) and testosterone (right panel) one week after unilateral and bilateral orchidectomy in mice. Difference from the control value (S): P<0.001 for double asterisks and P<0.005 for single asterisk. See legend of Fig. 1 for abbreviations of S, L, R and B. TABLE 1. Effects of unilateral and bilateral orchidectomy on the weight of preputial gland, prostate and the remaining testis in mice organ weight (mg/50g B.W.) ee ae testis ; tek right preputial gland prostate S 8 (4S ae 720 143.4+7.1 15 O0}0E e252 9.8+0.9 1G 8 139.8+6.0* 146.4+8.0 134.9+21.6 bie 029 R 8 135.6+5.4 137 SEES 12830221033 10.0+1.1 B 8 148.7+8.8* 147.0+6.6* 100.4+ 13.9 0) ac Ob: S: Sham-orchidectomy, L: left-sided orchidectomy, R: right-sided orchidectomy, B: bilateral orchidectomy. * weight at orchidectomy. Castration and Laterality of Neurons controls was statistically not significant. In con- trast, unilateral orchidectomy at either side caused an increase in plasma testosterone levels. The difference between right-testis-removed animals and sham-operated animals was statistically sig- nificant (P<0.005). There were no significant differences between the weight of right testes and the left ones at the time of hemiorchidectomy (Table 1). One week after hemiorchidectomy, the weight of the remain- ing testis was not significantly different from the initial value of the contralateral testis. In the comparison of the weights of prostates and prepu- tial glands among the four groups, preputial glands of bilaterally orchidectomized animals were, although statistically not significant, smaller than those of other groups. DISCUSSION The present experiments clearly demonstrated that there is an obvious left-right difference in the morphology of neurons of the POA. The size of neuronal nuclei of the POA was significantly larger in the right side than the left in male mice. In this connection, Gerendai et al. [5] and Bakalkin et al. [6] have already reported that the content of GnRH 1s greater in the right side of the hypothala- mus than in the left side in male and female rats. Fukuda et al. [7] found that the unilateral lesion to the right side, but not to the left side, of the medial anterior hypothalamus effectively suppressed the Ovarian compensatory hypertrophy in female rats and suggested that the right side of the medial anterior hypothalamus is indispensable for induc- ing the release of gonadotropins sufficient for the compensatory growth of the remaining ovary in hemiovariectomized rats. The present results are in good harmony with these previous findings and afford additional evidence for the hypothalamic laterality in the regulation of gonadotropin release in rodents. Since both the cell bodies and the nuclei of POA neurons are larger in size in the right side than in the left side, the neurons of the right POA may be playing more important roles than those of the left POA in the neuroendocrine control of gonadal function in mice. On the other hand, Dorner and Staudt [8, 9] 361 reported that male rats 70 days after bilateral orchidectomy showed no alterations in the mor- phology of neurons of the POA and VMN. They further found that testosterone treatment was not effective in inducing enlargement of the neuronal nuclear size of the POA in castrated male rats [8]. In the present experiments, however, bilateral orchidectomy invariably caused a prominent shrinkage of neuronal nuclei of the POA in male mice. This finding seems to accord with that of Kalra and Kalra [20] who reported a decrease in LHRH levels in the medial basal hypothalamus two weeks after orchidectomy in adult male rats. A possible reason for the discrepancy between the data by Dorner and Staudt [8, 9] and the present ones may be due to the difference in the duration of orchidectomy or the difference of the species of animals. In the present experiments, removal of the right testis caused a significant enlargement of the neuronal cell bodies of both sides of the POA, while either left-sided hemiorchidectomy or bi- lateral orchidectomy failed to induce such an effect. Since there were no significant differences in plasma levels of LH, FSH and testosterone between right-testis-removed mice and left-testis- removed animals, the present results altogether suggest that the neural connections between the testes and the hypothalamus are possibly involved in the control of gonadal endocrine function in male mice. Gerendai et al. [15] have already pointed out that unilateral adrenalectomy results in an en- hanced synthesis of proteins in the VMN of the contralateral side in rats. Gerendai and Halasz [14] similarly reported an accelerated incorpora- tion of radioactive amino acids into the ARC contralateral to the side of ovariectomy. They suggested that the neural signal from the peripher- al gland to the hypothalamus was probably trans- mitted via crossed afferent nerves. Bakalkin et al. [6] found that the left-sided hemiorchidectomy caused a decrease in LHRH contents in the right side of the hypothalamus and gave support to the crossed neural mediation between the testes and the hypothalamus. Mizunuma et al. [17], in con- trast, reported that the right-sided orchidectomy caused an increase in LHRH content in the right 362 hypothalamus in male rats. However, the present result showed no particular laterality in morpholo- gy of POA neurons after right-sided or left-sided orchidectomy in male mice. All these findings seem to indicate complicated features of neural influence from the peripheral endocrine organs to the hypothalamus. It is generally believed that unilateral orchidectomy in rodents results in the attenuation in the negative feedback effect of androgens on the hypothalamo-pituitary system and consequently enhances the release of gonadotropins from the pituitary gland. However, several authors re- ported that plasma LH levels did not increase significantly but, instead, plasma FSH levels ex- hibited a substantial and prolonged increase after unilateral gonadectomy in rats [17, 24, 25]. Con- trary to these results, Howland and Skinner [26] found that serum LH levels significantly increased on the next day of hemicastration and remained elevated for 40 days without any significant changes in serum FSH levels. Our results are consistent with those of Howland and Skinner [26]. In accordance with the elevated plasma LH levels in hemiorchidectomized mice, plasma levels of testosterone in these animals were higher than in the controls. Since it has been reported that plasma testosterone levels were either unchanged or lowered in hemiorchidectomized rats [27, 28], the present results further add the contradiction to the endocrine mechanism following unilateral gonadectomy. ACKNOWLEDGMENTS We are grateful to Dr. S. Raiti and the Pituitary Hormone Distribution Program, the National Institute of Arthritis, Diabetes and Digestive and Kidney Diseases (NIADDK), U.S.A. for the supply of rat gonadotropins and antisera. This work was supported in part by Grants-in-Aid No. 62540569 to T. M. and Nos. 63304008 and 02404007 to S. K. from the Ministry of Education, Science and Culture, Japan. REFERENCES 1 Diamond, M. C., Dowling, G. A. and Johnson, R. E. (1981) Morphologic cerebral asymmetry in male and female rats. Exp. Neurol., 71: 261-268. 2 Glick, S., Ross, D. and Hough, L. (1982) Lateral T. S. NAKAZAWA, T. MACHIDA 10 sit 12 13 14 15 AND S. KAWASHIMA asymmetry of neurotransmitters in human brain. Brain Res., 234: 53-63. Oke, A., Lewis, R. and Adams, R. N. (1980) Hemispheric asymmetry of norepinephrine distribu- tion in rat thalamus. Brain Res., 188: 269-272. Ross, D. A., Glick, S. D. and Meibach, R. C. (1981) Sexually dimorphic brain and behavioral asymmetries in the neonatal rat. Proc. Natl. Acad. Sci. USA, 78: 1958-1961. Gerendai, I., Rotsytein, W., Marchetti, B. and Scapagnini, U. (1979) LH-RH content changes in the mediobasal hypothalamus after unilateral ovari- ectomy. In “Neuroendocrinology: Biological and Clinical Aspects”. Proc. Sereno Symp., Vol. 19. Ed. by A. Polleri and R. Macleod, Academic Press, New York, pp. 97-102. Bakalkin, G. Ya., Tsibezov, V. V., Sjutkin, E. A., Veselova, S. P., Novikov, I. D. and Krivosheev, O. G. (1984) Lateralization of LH-RH in rat hypo- thalamus. Brain Res., 296: 361-364. Fukuda, M., Yamanouchi, K., Nakano, Y., Furuya, H. and Arai, Y. (1984) Hypothalamic laterality in regulating gonadotropic function: Unilateral hypothalamic lesion and ovarian compensatory hypertrophy. Neurosci. Lett., 51: 365-370. Dorner, G. and Staudt, J. (1968) Structural changes in the preoptic anterior hypothalamic area of the male rat, following neonatal castration and androgen substitution. Neuroendocrinology, 3: 136- 140. . Dorner, G. and Staudt, J. (1969) Structural changes in the hypothalamic ventromedial nucleus of the male rat, following neonatal castration and androgen treatment. Neuroendocrinology, 4: 278- 281. Feder, H. H. (1981) Experimental analysis of hor- mone actions on the hypothalamus, anterior pituit- ary, and ovary. In “Neuroendocrinology of Repro- duction”. Ed. by N. T. Adler, Plenum, New York, pp. 243-278. Machida, T. (1971) Luteinization of ovaries under stressful conditions in persistent-estrous rats bearing hypothalamic lesions. Endocrinol. Japon., 18: 427- 431. Witkin, J. W., Paden, C. M. and Silverman, A. J. (1982) The luteininzing hormone-releasing hor- mone (LHRH) systems in the rat brain. Neuroen- docrinology, 35: 429-438. Engeland, W. C. and Dallman, M. F. (1975) Com- pensatory adrenal growth is neurally mediated. Neuroendocrinology, 19: 352-362. Gerendai, I. and Haldsz, B. (1976) Hemigo- nadectomy-induced unilateral changes in the rat hypothalamic arcuate nucleus. Neuroendocrinology, 7A \2 SB N=3 37), Gerendai, I., Kiss, J., Molnar, J. and Halasz, B. 16 ig 18 19 20 al Castration and Laterality of Neurons (1974) Further data on the existence of a neural pathway from the adrenal gland to the hypothala- mus. Cell Tissue Res., 153: 559-564. Halasz, B. and Szentagothai, J. (1959) Histologi- scher Beweis einer nervosen Signaltibermittlung von der Nebennierenrinde zum Hypothalamus. Z. Zell- forsch. mikrosk. Anat., 50: 297-306. Mizunuma, H., DePalatis, L. R. and McCann, S. M. (1983) Effect of unilateral orchidectomy on plasma FSH concentration: evidence for a direct neural connection between testes and CNS. Neuroendocrinology, 37: 291-296. Kovacic, N. and Parlow, A. F. (1972) Alteration in serum FSH/LH ratios in relation to the estrous cycle, pseudopregnancy, and gonadectomy in the mouse. Endocrinology, 91: 910-915. Selmanoff, M. K., Goldman, B. D. and Ginsburg, B. E. (1977) Developmental changes in serum luteinizing hormone, follicle stimulating hormone and androgen levels in males of two inbred mouse strains. Endocrinology, 100: 122-127. Tsutsui, K. and Ishii, S. (1978) Effects of follicle- stimulating hormone and testosterone on receptors of follicle-stimulating hormone in the testis of the immature Japanese quail. Gen. Comp. Endocrinol., 36: 297-305. Tsutsui, K., Shimizu, A., Kawamoto, K. and Kawashima, S. (1985) Developmental changes in the binding of follicle-stimulating hormone (FSH) to testicular preparations of mice and the effects of hypophysectomy and administration of FSH on the dep 23 24 25 26 Mall 28 363 binding. Endocrinology, 117: 2534-2543. Machida, T., Yonezawa, Y. and Noumura, T. (1981) Age-associated changes in plasma testoster- one levels in male mice and their relation to social dominance or subordinance. Horm. Behav., 15: 238-245. Kalra, P. S. and Kalra, S. P. (1980) Modulation of hypothalamic luteinizing hormone-releasing hor- mone levels by intracranial and subcutaneous im- plants of gonadal steroids in castrated rats: Effects of androgen and estrogen antagonists. Endocrinolo- gy, 106: 390-397. Gomes, W. R. and Jain, S. K. (1976) Effect of unilateral and bilateral castration and cryptorchi- dism on serum gonadotropins in the rat. J. Endocri- nol., 68: 191-196. Ramirez, V. D. and Sawyer, C. H. (1974) A sex difference in the rat pituitary FSH response to unilateral gonadectomy as revealed in plasma radioimmunoasay. Endocrinology, 94: 475-482. Howland, B. E. and Skinner, P. K. (1975) Changes in gonadotropin secretion following complete or hemicastration in the adult rat. Hormone Res., 6: 71-77. Lindgren, S., Damber, J.-E. and Caystensen, H. (1976) Compensatory testosterone secretion in uni- laterally orchidectomized rats. Life Sci., 18: 1203- 1206. Moger, W. H. (1977) Endocrine responses of the prepubertal male rat to hemiorchidectomy. Biol. Reprod., 17: 661-667. in F HS iw | ed as oat ns} hoc sv pen ve Pie nt a, coe Hy: iy Ag aed ; ci C “tio UNE sa vai cae a x soothes fe ) ete i rt Re : > apanth ete a “sesGti4onbtte ape! fore yevieneg an ; ces BS eon % aia ain ‘y., dames bogs yodatasa obit gharieg ti, Je Bf ke ‘ c i ee o pedis pORmIG. py — ie Ms mgr ee nae mga j fee ; Se y aay ie in ai Se as ee, ui NE i : +1a0 hie fe: Ae Bey ae Pig. Za) AGN sah > ey me 2 “bites stil i say Ostintosbitecd ene eS bs a. Sapp ihaid : 4 “i 5 2 at in ‘oor gaint ¢ sath ee cuewaled.¢ a Sics 2 gunk en < fi Hip sts at et toa (KT OT). aquatic £2 * ott oR, een : she i wea ant toes sasaki | ay 4 ” oe, ae ' 1; Mi ia bugga ye Ge cab es, oss AE, 3 i Sai ~ pry a ares FFE bles ty j nacre 4 ryt A af ! =) 2 © ” i ; ne " ae c 4, F “ AWS} (4% f ; : Pee ara ‘ ! i: ? if es - POPs i Jj ue oy / s : ch 1h pat Cede ee Pri z "i © : if yg ot os te bys tery ets 3 be (La) ar * 4% i RE \ 2 : . 4 ‘| ; 7 4 az E ~ ™ =" ‘ Bx = i } . s ee A ays P ba oy 1 Ee i r Lo. = fi { a . i 3 ( aes S ‘i 1 ee : = y - i , { f a me 4 4 = 3 ZOOLOGICAL SCIENCE 9: 365-373 (1992) © 1992 Zoological Society of Japan Effects of a Gonadotropin-Releasing Hormone Analog (HOE 766) on Germinal and Interstitial Compartments during the Annual Cycle in the Green Frog: Rana esculenta LOREDANA D1 MatTreo, SERGIO Minucci, MAurRo D’ ANTONIO, SILVIA FASANO and RICCARDO PIERANTONI Dipartimento di Fisiologia Umana e Funzioni Biologiche Integrate “F. Bottazzi’, via Costantinopoli 16, I Facolta di Medicina e Chirurgia, Universita di Napoli, 80138 Napoli, Italy ABSTRACT— Seasonal changes, related to treatments carried out using a GnRH-analog (HOE766, GnRHA), germinal and interstitial compartment activites were studied in the frog, Rana esculenta testis. We have investigated on: a) annual variations of mitotic index (MI) of primary spermatogonia (I SPG), b) changes of interstitial area, c) fluctuation of testicular androgen concentration, and d) variations of sperm releasing activity during different periods of the year were also studied. Our results indicate that testes of Rana esculenta show a cycle of responsiveness to GnRHA treatments with respect to androgen production, interstitial growth, I SPG multiplication and spermiation. For each phenomenon considered, the extent of the response is peculiar of the period of the year in which the experiment is carried out. INTRODUCTION Seasonal gonadal activity has been extensively studied in vertebrates [1] with respect to the regu- lation of steroidogenesis and spermatogenesis. The hypothalamus-hypophyseal control mecha- nisms appear to play a central role in driving the cascade of events which turns on the breeding seéason. temperature) act either modulating the release of GnRH from hypothalamus or directly controlling the pituitary sensitivity in term of gonadotropin discharge; this, in turn, is further regulated by the steroid milieu [2, 3]. Also factors inherent to the testis have been claimed to contribute to the phenomenon determining a cycle of gonadal sensi- tivity to gonadotropin stimulation [4]. In the frog, Rana esculenta, interstitial and ger- minal compartments show different seasonality since steroidogenesis and spermatogenic wave occur at high rate during late autumn-early spring and late spring-early autumn, respectively [2]. Indeed, environmental cues (light and Accepted November 20, 1991 Received January 17, 1991 Therefore, it is interesting to have insight in the physiology of the two different testicular compart- ments in this frog species which may constitute an useful animal model to investigate the regulation of testicular functions. Scope of the present paper is to study changes of the responsiveness of the pituitary-testis axis dur- ing the annual cycle, treating intact frogs, Rana esculenta, with a gonadotropin-releasing hormone analog (HOE 766). Indeed, a divergent response elicited by the treatment on the endocrine and germinal tissue respectively, may indicate the ex- istence of local mechanisms which activate testicu- lar compartments to be responsive to the same stimulus separately in different periods of the year. MATERIALS AND METHODS Adult frog, Rana esculenta, were collected dur- ing 1987, from January to December. Each month, 10 freshly collected males received 900 ng GnRH agonist, (GnRHA HOE 766 Hoechst, 45 ng/g BW) in Krebs-Ringer bicarbonate buffer (KRB). Injections were given into the dorsal sac on alternate days for 2 weeks. Simultaneously, 10 366 animals were used as controls and injected with vehicle (KRB) alone. During March, June and October 5 animals per experimental group were decapitated 2.5 h after the last injection and right testes were used to determine the androgen con- centration by RIA as described previously [4, 5]. Since the antiserum used was cospecific for testos- terone and 5 a-dihydrotestosterone, the results are expressed as “androgens”. Intra- and interassay coefficients of variations were 6% and 9%, respec- tively and sensitivity was 2 pg/tube. The left testes were fixed in Bouin’s fluid and examined, after staining with hemallum and eosin, for interstitial area calculation and for signs of spermiation. These periods were chosen since they are peculiar of the seasonal cycle of Rana esculenta. Indeed, during March androgens are at the highest concen- tration and spermatogenesis begins, reaching its maximum during late spring-early summer (June). In October, the spermatogenic wave is near to be extinguished and androgen levels start to increase again [2, 4, 6-8]. Interstitial area was measured in 10 randomly chosen sections per experimental group and expressed in wm? (mean+SD). Sper- 22 10 Mitotic Index ISPG J F M A M Fic. 1. J L. Di. Matreo, S. Mincct, et al. miation was calculated as the number of empty tubules per 100 total tubules observed in randomly chosen sections. The remaining animals collected all year around received 100 ug colchicine (Prolabo, Paris) into the dorsal sac and were decapitated 24h later. Testes were fixed in Bouin’s fluid and 5-6 um cross-sections were stained with hemallum and eosin. Mitotic index (MI) of the primary sperma- togonia (I SPG) was expressed as number of metaphases per total I SPG counted multiplied by 100 [9]. Significance of differences was evaluated using Student’s “rt” test for between group comparisons and Duncan’s test for multigroup comparisons. “Chi square” test was also carried out when appropriate. RESULTS a) Annual variation of MI of I SPG in GnRHA- treated frogs In control animals the MI of I SPG showed two @e Control O-OGnRHA Annual variations of ISPG mitotic index in vehicle and GnRHA treated frogs, Rana esculenta. Duncan’s test was used to evaluate significance of differences among vehicle or GnRHA-injected animals. Student’s “r” test was used to evaluate significance of differences between vehicle and GnRHA-injected animals of the same month. Values are the mean+SD. GnRH Analog Effect on Spermatogenesis 367 peaks of proliferative activity (Fig. 1). Indeed, I SPG mitosis peaked in April (P<0.001 vs Febru- ary or June) and October (P<0.001 vs July and November), although during October the peak of proliferative activity was less pronounced (P< 0.001 vs April). Similarly, GnRHA treated frogs showed two significant peaks of proliferative activ- ity of I SPG (P<0.001) simultaneously with those of control animals. Moreover, in all periods of the year GnRHA stimulated significantly I SPG multi- plication (P<0.05 at least). Interestingly, the extent of stimulation observed from June until October was greater (P<0.01) as compared with the remaining periods. Indeed, while GnRHA increased the MI of I SPG 127.32% +11.2 during June-October period, only 116.16% +6.4 was the stimulation measured in the other months. b) Measurement of testicular androgen concentra- tion and interstitial area in different periods of the year Testicular androgen content (Fig. 2) was highly GnRHA C Control ee e @ee e@ee@ eee e@e@e e Androgens ng/testis e@ee0eee ee eeeeeee¢e¢ eeoeoeeoeeee#eee8000e @ee0oeee8ee800080 @ @ e@ee#e ee e e eeeee#e#e March June October Fic. 2. Testicular androgen concentration of vehicle [_] and GnRHA f-] treated frogs, Rana esculenta. Dun- can’s test was used to evaluate significance of differ- ences among vehicle or GnRHA-injected animals. Student’s “r” test was used to evaluate significance of differences between vehicle and GnRHA-injected animals of the same month. Values are the mean+ SD: stimulated by GnRHA during March (P<0.001). No GnRHA effects were evidenced in June frogs, while in October animals GnRHA was able to increase again the androgen concentration (P< 0.001). However, the hormone levels achieved during October were significantly lower (P< 0.001) as compared with those measured in March. Conversely, interstitial area (Fig. 3) of October testes appeared unaffected by GnRHA while June testes showed greater values in peptide treated animals (P<0.001). A strong stimulation was achieved in March testes (P<0.001) in which the interstitial area increased about 3 fold and reached value significantly higher as compared with June values (P<0.001). GnRHA i|zeze [a Control Interstitial area (um) October Fic. 3. Intestitial areas calculated in testes of vehicle [| and GnRHA [fi] treated frogs, Rana esculenta. Significance of differences was evaluated as de- scribed in Fig. 2. Values are the mean+SD. c) Measurement of spermiation activity in dif- ferent periods of the year GnRHA induced spermiation in all periods ex- amined (P<0.005) although the efficiency of the phenomenon showed differences (Table 1). In- deed, March and October testes were highly sensi- tive to the peptide as compared with June testes (P <0.005). Only 40 of 100 tubules examined were empty as compared with near to 100% spermiating 368 L. Di. Matteo, S. Minccl, et al. TABLE 1. Sperm releasing activity of vehicle and GnRHA treated frogs Rana esculenta MARCH | JUNE OCTOBER a () +10 ae OS CONTROL b — 100 2) 90 (b) “LOS (c) + 95 +40 + 100 GnRHA d n hog (¢) oe (e) Ets (i AWE Gl IPOS b vs e P<0.05 GC ovs f 2 —<0105 d+f vs e P<0.005 + and — represent spermiating and non-spermiating tubules. Spermiation was calculated as the number of empty tubules per 100 total tubules observed in randomly chosen sections. Significance of differences was calculated using Y? test. tubules observed in March and October. It is interesting to note (Fig. 4) that in March testes of control animals almost exclusively spermatozoa were present. Conversely, tubules characterized by the presence of all spermatogenic stages (except spermatozoa, since the presence of spermiation) were Observed in GnRHA treated animals. Anim- als captured in June (Fig. 5) had testes fully orga- nized without differences (control vs GnRHA tre- ated animals) in respect to the development of germinal cells which were all represented. Octo- ber testes (Fig. 6) presented few spermatocytes and spermatides, while tubules full of spermato- zoa, which almost disappeared after GnRHA treatment, were observed. DISCUSSION GnRH elicits gonadotropin discharge from the pituitary in vertebrates [1]. In amphibians, the peptide elicits both LH and FSH release whose secretory pattern appears similar in frogs [10] but not in toads [11] where FSH is detectable in plasma when LH is at base-line values. In frogs, gonado- tropins act in males, through a low FSH/LH specific receptor [12], primarily stimulating androgen production by Leydig cells [3] and in- fluencing spermatogonial proliferation [13, 14]. GnRHA treatments described here indicate that the testis undergoes a cyclic seasonal responsive- ness, by both endocrine and germinal compart- ments, which occurs separately during the year. Although germinal and interstitial compartments of toad testis may be stimulated separately by FSH and LH, respectively [11], this does not seem to be the case in frogs [10]. Present data in the frog, Rana esculenta, show that maximal androgen pro- duction is available in March, while germinal com- partment appears to respond to a greater extent during spring-early autumn period. The measure- ment of the interstitial area shows that the ster- oidogenic compartment becomes highly developed during the period of maximal androgen production’ (March). Interstitial area is still stimulated during June when androgen production is not affected by GnRHA. Moreover, in October the GnRHA treatment induces androgen production, although to a less extent as compared with March, but interstitial area remains unaffected. This may indicate that the steroidogenic response to the GnRHA treatment is in a certain degree indepen- dent of the trophic response by the interstitial compartment. Substances other than androgens may be produced by the interstitium under go- nadotropin stimulation to support the gonadal activity, as suggested in mammals [15]. Of course, the knowledge of FSH and LH profiles would be helpful in elucidating the problem. Unfortunately, frog antisera currently available do not crossreact with Rana esculenta pituitary homogenate (Licht and Pierantoni, unpublished). Maximal respon- siveness of the steroidogenic tissue appears to occur concomitantly with the androgen peak which has been detected in March during the annual cycle, while lack of stimulation of androgen pro- duction has been evidenced in June when plasma GnRH Analog Effect on Spermatogenesis 369 Fic. 4. Testes of March animals treated with vehicle characterized by a) presence of numerous spermatozoa and b) normal interstitial tissue. GnRHA-treated frogs show c) absence of spermatozoa and d) abundant interstitial tissue. L. Di. Matreo, S. Minccl, et al. 370 Testes of June animals treated with vehicle (a and b) and GnRHA (c and d). Germinal compartment is fully organized and spermatozoa are still present after GnRHA treatment. Fic. 5. GnRH Analog Effect on Spermatogenesis 371 Fic. 6. Testes of October animals treated with vehicle a) characterized mostly by the presence of few spermatocytes, spermatides and spermatozoa and b) by regressed interstitial compartment. GnRHA-treated frogs show c) absence of spermatozoa and d) interstitial compartment development similar to that of vehicle treated animals. 372 L. Di. Martreo, S. Minccl, et al. and testicular androgen levels reach the nadir [4, 6, 7, 8]. This in vivo experiment confirms previous in vitro results showing that minced testes do not respond to any stimulus during summer in term of androgen production [4]. Therefore, our data strongly suggest that intratesticular mechanisms may determine an endocrine refractoriness. On the contrary, spermatogonial proliferation can be stimulated at any time, although at different ex- tent. Spermatogonial proliferation occurs, in Rana esculenta, all around the year [16] and reaches maximal values, as measured by mitotic index, during April and October. Moreover, sperma- togenic wave can be stimulated by appropriate hormonal and environmental factors (light and temperature) either during winter or during sum- mer [17]. Our results confirm the basal profile of I SPG proliferation and indicate that the germinal compartment is more responsive to GnRHA treat- ment during summer-early autumn, concomitantly with the appearance of the maximal rate of the spermatogenic wave [18]. As for the sperm-releasing activity, spermiation has been detected in all period studied, and com- parable sperm releasing activity occurs in March and October. An interesting observation to emerge is that the high rate of spermiation does not indicate the existence of a similar organization of the germinal compartment. Indeed, March testis appears to be different from October testis as far as the germinal compartment composition it concerns. Despite the diverse histological picture, spermination, after GnRHA treatment, is strong and comparable between the two periods. A cycle of responsiveness is also evident since June testes, although these are characterized by the presence of all spermatogenic stages, appear to respond with minor efficiency. In conclusion, testes of the frog, Rana esculenta show cycle of responsiveness to GnRHA treat- ment in term of androgen production, interstitial! growth, ISPG multiplication and spermiation. All these events are not concomitant but each of them is peculiar of the period of the year in which the experiment is carried out. This may be ascribed to internal testicular factors which determine how the gonads respond or utilize the gonadotropin sti- mulation. ACKNOWLEDGMENT The long acting GnRH agonist (HOE 766) was a generous gift from Dr J. Sandow (Hoecst, Frankfurt, West Germany). REFERENCES 1 Chieffi, G. (1989) New trends in the regulation of the gonadal activity in vertebrates: paracrine and autocrine control. Zool. Sci., 6: 623-637. 2 Rastogi, R. K. and Iela, L. (1980) Steroidogenesis and spermatogenesis in anuran Amphibia: A brief survey. In “Steroid and their Mechanisms of Action in Nonmammalian Vertebrates”. Ed. by G. Delrio and J. Brachet, Raven Press, New York, pp. 131- 146. 3. Licht, P. and Porter, D. A. (1987) Role of gonado- tropin-releasing hormone in regulation of gonado- tropin secretion from amphibian and reptilian pitui- taries. In “Hormones and Reproduction in Fishes, Amphibians and Reptiles”. Ed. by D. O. Norris and R. E. Jones, Plenum Press, New York, pp. 61-85. 4 Pierantoni, R., Iela, L., d’Istria, M., Fasano, S., Rastogi, R. K. and Delrio, G. (1984) Seasonal testosterone profile and testicular responsiveness to pituitary factors and gonadotropin-releasing hor- mone during two different phases of the sexual cycle . of the frog (Rana esculenta). J. Endocrinol., 102: 387-392. : e 5 Pierantoni, R., Fasano, S., Di Matteo, L., Minucci, S., Varriale, B. and Chieffi, G. (1984a) Stimulatory effect of a GnRH agonist (buserelin) in vitro and in vivo testosterone production by the frog (Rana esculenta) testis. Mol. Cell. Endocrinol., 38: 215- AAS). 6 d Istria, M., Delrio, G., Botte, V. and Chieffi, G. (1974) Radioimmunoassay of testosterone, 17/- oestradiol and oestrone in the male and female plasma of Rana esculenta during sexual cycle. Ster- oids Lipids Res., 5: 42-48. 7 Fasano, S., Minucci, S., Pierantoni, R., Fasolo, A., Di Matteo, L., Basile, C., Varriale, B. and Chieffi, G. (1988) Hypothalamus-hypophysis and testicular GnRH control of gonadal activity in the frog, Rana esculenta: seasonal GnRH profiles and annual varia- tions of in vitro androgen output by pituitary- stimulated testes. Gen. Comp. Endocrinol., 70: 31- 40. 8 Varriale, B., Pierantoni, R., Di Matteo, L., Minuc- ci, S., Fasano, S., D’Antonio, M. and Chieffi, G. (1986) Plasma and testicular estradiol and plasma androgen profile in the male frog Rana esculenta during the annual cycle. Gen. Comp. Endocrinol., 64: 401-404. 9 Rastogi, R. K., Iela, L., Di Meglio, M., Di Matteo, 10 11 1 13 GnRH Analog Effect on Spermatogenesis L., Minucci, S. and Izzo-Vitiello, I. (1983) Initiation and kinetic profiles of spermatogenesis in the frog, Rana esculents (Amphibia). J. Zool. Lond., 201: 515-525. Licht, P., McCreery, E. R., Barnes, R. and Pang, R. (1983) Seasonal and stress related changes in plasma gonadotropins, sex steroids and corticoster- one in the bullfrog, Rana catesbeiana. Gen. Comp. Endocrinol., 50: 124-145. Itho, M., Inoue, M. and Ishii, S. (1990) Annual cycle of pituitary and plasma gonadotropins and plasma sex steroids in wild population of the toad, Bufo japonicus. Gen. Comp. Endocrinol. 78: 242- 253) Takada, K., Kubokawa, K. and Ishii, S. (1986) Specific gonadotrophin sites in the bullfrog testis. Gen. Comp. Endocrinol. 61: 302-312. Rastogi, R. K., Di Matteo, L., Minucci, S., Di Meglio, M. and Iela, L. (1990) Regulation of primary spermatogonial proliferation in the frog (Rana esculenta): an experimental analysis. J. Zool. Lond., 220: 201-211. 14 15 16 17 373 Di Matteo, L., Minucci, S., Fasano, S., Pierantoni, R., Varriale, B. and Chieffi, G. (1988) A gonado- tropin-releasing hormone (GnRH) antagonist de- crease androgen production and spermatogonial multiplication in the frog (Rana esculenta): indirect evidence for the existence of GnRH or GnRH-like material receptors in the hypophysis and testis. Endocrinology, 122: 62-67. Sharpe, R. M. (1986) Paracrine control of the testis. Clin. Endocrinol. Metab., 15: 185-207. Rastogi, R. K., Di Meglio, M., Di Matteo, L., Minucci, S. and lela, L. (1985) Morphology and cell population kinetics of primary spermatogonia in the frog (Rana esculenta) (Amphibia: Anura). J. Zool. Lond., (A) 207: 319-330. Rastogi, R. K., Tela, L., Saxena, P. K. and Cheffi, G. (1976) The control of spermatogenesis in the frog, Rana esculenta. J. Exp. Zool., 196: 151-166. Rastogi, R. K. (1976) Seasonal cycle in anuran (Amphibia) testis: the endocrine and environmental controls. Boll. Zool., 43: 151-172. ‘eh ced si oe eae alae idebgnrans oa my ear O ZOOLOGICAL SCIENCE 9: 375-386 (1992) Changes in Salmon GnRH and Chicken GnRH-II Contents in the Brain and Pituitary, and GTH Contents in the Pituitary in Female Masu Salmon, Oncorhynchus masou, from Hatching through Ovulation MASAFUMI AMANO, KATSUMI AIDA, Naoto Oxumorto! and YOSHIHISA HASEGAWA~ Department of Fisheries, Faculty of Agriculture, The University of Tokyo, Bunkyo, Tokyo, Japan, ‘Nikko Branch, National Research Institute of Aquaculture, Chugushi, Nikko, Tochigi, Japan, *Department of Obstetrics and Gynecology, Gunma University, School of Medicine, Maebashi, Gunma, Japan ABSTRACT—Changes in salmon gonadotropin-releasing hormone (sGnRH) and chicken GnRH-II (cGnRH-II) contents in the brain and pituitary, and gonadotropin (GTH) subunits GTH IP and GTH IIf contents in the pituitary of female masu salmon (Oncorhynchus masou) were investigated from hatching through ovulation. Gonadosomatic index (GSI) showed a gradual increase until the second summer (two year-olds), and thereafter a rapid increase was observed in accordance with vitellogenesis. Ovulation occurred in autumn. Brain sGnRH was already measurable at hatching, whereas cGnRH-II was first detected two months later. Both GnRHs contents increased during the underyearling phase, and fluctuated thereafter. Pituitary sGnRH contents showed a stepwise increase every summer for three years. sGnRH concentrations in each discrete brain area showed seasonal changes: high during autumn-winter and low in summer. sGnRH concentrations in the olfactory bulbs and telencephalon, and pituitary contents of s6nRH, GTH If and GTH IIf significantly increased prior to ovulation. Pituitary GTH If contents showed clear seasonal changes for three years-high in autumn and low in winter-regardless of the state of ovarian maturity. Brain cGnRH-II contents were lower than sGnRH contents. Moreover, pituitary cGnRH-II contents were undetectable and no significant changes in concentration in discrete brain areas were observed during vitellogenesis and ovulation. These results suggest that sGnRH is involved in ovarian maturation via the regulation of GTH synthesis and release in © 1992 Zoological Society of Japan this species, whereas cGnRH-II has little or no involvement in reproduction. INTRODUCTION Recent studies have shown that more than one type of GnRH exists in the teleost brain [1]. In most teleost species examined including salmo- nids, salmon GnRH (sGnRH) and chicken GnRH- II (CGnRH-II) have been detected. It is generally accepted that one of the important roles of GnRH is the regulation of the synthesis and release of GTH by the GTH cells in the pituitary. Although both sGnRH and cGnRH-II molecules stimulate the release of GTH in the fish pituitary both in vivo Accepted November 25, 1991 Received October 14, 1991 and in vitro under exogenous administration [2, 3], their physiological roles in the brain and pituitary are not fully understood. These two types of GnRHs exist not only in the hypothalamus but also in the other parts of the brain, and show differen- tial distributions [4, 5]. Furthermore, distribution patterns of both GnRHs are vary according to species. Both sGnRH and cGnRH-II exist in the goldfish pituitary [4], but only sGnRH is detect- able in rainbow trout Oncorhynchus mykiss pitui- tary [5]. There are few studies on the changes in brain GnRH contents in relation to gonadal maturation, and such results are not consistent. Gentile et al. [6] measured GnRH concentrations in the tel- 376 M. AMANO, K. AIDA et al. encephalon and hypothalamus of Venezuelan freshwater fish, Pygocentrus notatus, using a mammalian GnRH RIA system and found that GnRH concentrations were high in mature fish. Yu et al. [7] measured GnRH content in discrete brain areas of female goldfish at different stages of ovarian development. However, brain GnRH content did not show clear parallel changes with seasonal ovarian development. Okuzawa et al. [5] measured sGnRH and cGnRH-II contents in dis- crete brain areas of rainbow trout sampled in September, and found that the pattern of sGnRH distribution differed with age and stage of sexual maturity. Recently, Suzuki et al. [8] reported that two structurally different GTHs, GTH I and GTH II, exist in the chum salmon pituitary. GTH I is considered to be involved in regulation of the early stages of gonadal maturation, and GTH II is considered to mainly regulate ovulation and sper- miation [9]. There is, however, no information on the changes in pituitary GTH I and GTH II contents throughout the fishes’ life span. Since the release of GTH is regulated by GnRH, it 1s speculated that changes in pituitary GTH contents are correlated with those in brain GnRH contents. Therefore, in the present study, we investigated changes in sGnRH and cGnRH-II contents in the brain and pituitary, and GTH If and IIf subunit contents in the pituitary of female masu salmon (Oncorhynchus masou) which were maintained in fresh water for three years. Changes were fol- lowed from hatching through ovulation, in order to obtain basic information on annual rhythms of these hormones. MATERIALS AND METHODS Fish For purposes of this study, eggs of masu salmon, Oncorhynchus masou, were artificially fertilized in October 1987 at the Nikko Branch, National Re- search Institute of Aquaculture, Tochigi Prefec- ture. The eggs hatched in December 1987, and the fish were reared under natural photoperiod in spring water of constant temperature (9-10°C) throughout the experiment. The fish we used for this study were offspring of wild fish which had migrated to the Shiribetsu River (Hokkaido). Wild masu salmon migrate to the sea in the spring (1.5 years-old), and return to the river in May after a one-year stay in the sea; they spawn in autumn and die. The masu salmon kept in the institute also smoltified at 1.5 years-old and matured at three years of age in fresh water, although the growth rate was not very rapid. There is a landlocked form of masu salmon called “yamame”. This variety for the most part does not smoltify. Sampling Sampling was carried out once a month (twice in May 1988) from December 1987 (month of hatch- ing) through October 1988, and at 2—4 months intervals from January 1989 through October 1990 (month of ovulation). On the day of autopsy, fish were randomly selected and were anesthetized in ethyl-p-aminobenzoate (0.05%). After measure- ments on body length and weight, brain and pitui- tary were rapidly removed and frozen on dry-ice intended for the measurement of GnRHs and GTHs. At the first two samplings (December 1987 and January 1988, mean body weight 0.16-0.39 g), the head region was cut off and used for only GnRH measurement as dissection of the brain was difficult. The pituitary was removed on sampling occasions starting from April 1988 (mean body weight 3.21 g). From September 1988 (mean body weight 17.2 g), brain tissue was dissected into five parts: the olfactory bulb, the telencephalon, the hypothalamus, the optic tectum-thalamus and the Fic. 1. Schematic diagram of sagittal section of masu salmon brain. Letters represent the following brain areas: a, olfactory bulbs and tracts; b, telencepha- lon, including preoptic area; c, hypothalamus; d, optic tectum-thalamus, including anterior part of cerebellum; e, cerebellum; f, medulla oblongata; g, pituitary. Changes of GnRH in Female Masu Salmon cerebellum-medulla oblongata. The cerebellum and medulla oblongata were further differentiated from May 1989 (mean body weight 34.5 g) as shown in Fig. 1. Brain tissues were stored at —30°C until extraction. At the first two samplings, distinction of sex was impossible. From February 1988 (mean body weight 1.35 g), ovaries were fixed with Bouin’s fluid for 24 hr and their weights were measured to calculate GSI. The ovaries were embedded in paraffin and sectioned at 5 wm. The sections were stained with hematoxylin and eosin for histological observation. The results pertaining to males will be reported separately. From September 1988, blood was also sampled in order to measure plasma GTH and steroid hormone levels. These data will also be reported elsewhere. GnRH RIAs Extraction of GnRH from the brain tissue was done according to Okuzawa et al. [5]. sGnRH and cGnRH-II contents were measured by respective RIAs established by Okuzawa et al. [5]. Prior to the measurement of sGnRH and cGnRH-II con- tents, it was confirmed that masu salmon possess both GnRHs in the brain (Fig. 2), by HPLC-RIA 150 100 50 GnRH (pg/fraction) (%) ajisjluojaoe 0 10me 2220 808240. -50 Time (min) 377 analysis according to Okuzawa et al. [5]. More- over, brain extracts of this species showed dis- placement curves which were parallel to the curves for the sGnRH and cGnRH-II standards in each RIA (Fig. 3). These sGnRH and cGnRH-II RIA were thus validated for application to masu sal- mon. GnRH was expressed in terms of both content (per region) and concentration (per g tissue). Serial dilution of brain extract US a a 100 A S GnRH ies san 50 = a 0 1 10 100 GnRH (pg/tube) Serial dilution of brain extract te 100 B se — 50 cGnRH-II fea) 0 1 10 100 GnRH (pg/tube) Fic. 3. Competition curves for sGnRH and brain ex- tract of masu salmon in sGnRH RIA system (A), and cGnRH-II and brain extract of masu salmon in cGnRH-II RIA system (B). The scale for dilution of brain extract indicates a two-fold serial dilution. Each point represents the average of duplicate de- terminations. GTH RIAs Pituitary contents of GTHs were measured by two different RIAs using the same samples used Fic. 2. Reverse-phase HPLC of masu salmon brain extract followed by sGnRH RIA (A) and cGnRH-II RIA (B). Arrows indicate the elution time of synthetic cGnRH-II and sGnRH. The mobile phase was CH;3CN (acetonitrile) containing 0.1% TFA. 378 M. AMANO, K. AIDA et al. for GnRH measurement. GTH If and GTH II? and antisera against GTH I? and GTH IIP were kindly provided by Dr. H. Kawauchi of Kitasato University. GTH If and GTH II were measured by respective RIAs. GTH If and GTH II were purified from chum salmon (Oncorhynchus keta) pituitary by Suzuki ef al. [10], and the antisera against GTH If and GTH IIf were raised by Suzuki et al. [9]. GTH If and GTH IIP were iodinated according to the method of Kobayashi et al. [11]. The procedure of each RIA was the same as that in the sGTH RIA [11]. Displacement curves for pituitary samples were parallel to the standard curves in both GTH I and GTH II RIA. The intra- and inter-assay coefficients of variation in GTH I RIA were 10.9% (n=4) and 23.0% (n=4), respectively, at about 50% binding. The sensitivity of the assay, defined as twice the standard deviation at zero dose, was 62.5 pg/tube (n=5). The antiserm against GTH If cross- reacted with GTH I, GTH II and GTH II at 1.7%, 4.0% and 4.4%, respectively, at 50% bind- ing. The intra- and inter-assay coefficients of variation in GTH II RIA were 12.0% (n=4) and 11.7% (n=4), respectively. Sensitivity was 5 pg/ tube (n=8). The antiserum against GTH II? was found to cross-react with GTH I, GTH II and GTH If at 0.22%, 3.7% and 1.0%, respectively, at 50% binding. Statistics The Student’s t-test and Cochran-Cox test were employed in statistical analysis. RESULTS GSI Changes in body weight and GSI are shown in Fig. 4. GSI gradually increased from 0.18% (February 1988) to 0.52% (May 1990), showing a small peak in September 1989. GSI rapidly in- creased from July (0.73%) through October 1990 (12.6%), in accordance with the advancement of vitellogenesis and ovulation. Ovulation was observed in 9 out of 12 individuals in October 1990. Pituitary GTH contents Changes in pituitary GTH If and GTH IIé t—— Underyearling _+——Yeaarling "2 yearrs-old__{ 200 100 10 0.1 1988 Fic. 4. Body Weight (g) GSI (%) "121234567 8 9101112123456 7 8 910111212345 6/7 8 910 1989 1990 Seasonal changes in body weight and gonadosomatic index (GSI) of masu salmon from February 1988 to October 1990. Numbers beside each symbol indicate the number of fish employed. Each value is expressed as the mean (point) and the standard error (bar). * (p<0.05), ** (p<0.01), and *** (p<0.001) indicate the levels of significant differences. Changes of GnRH in Female Masu Salmon 379 contents are shown in Fig. 5A. Pituitary GTH If contents showed clear seasonal changes: high in autumn after increases from spring, and low in winter. A distinct decrease was seen in January 1989 and a distinct increase was observed during vitellogenesis and ovulation in 1990. 10000) p 2 & 100 = = ©?) ‘Ss a= <= 10 O 1988 1000) pg ——— GTH IB = +——4 GTH IIB S 2 100 2. fo)) E o = 10 = &) ‘ 1988 Fic. 5A and B. 1989 GTH II showed a gradual increase until July 1990. Thereafter the contents increased rapidly until October 1990, in accordance with the advancement of vitellogenesis and ovulation. Changes in pituitary GTH If and GTH IIf concentrations were similar to those in contents ASO mom OmOntit2n) 23 454677, (8 StOimH2 1) 2: 345 6e7 (8s 910 1989 1990 Seasonal changes in pituitary GTH If (square) and GTH IIf (triangle) contents (A) and concentrations (B) of the masu salmon from April 1988 to October 1990 and from March 1989 to October 1990, respectively. Presentation of statistical data is as in Fig. 4. 380 M. AMANO, K. AIDA et al. (Fig. 5B). sGnRH in the brain and pituitary Brain sGnRH was already detectable at hatch- ing, and total brain content increased with growth during the underyearling stage (Fig. 6A). There- after, increases were observed from March to May 1989 and from September to January 1990. A decrease was observed from May to September 1989. Changes in brain sGnRH concentrations are shown in Fig. 6B. A significant decrease was observed from April to August 1988, and a sig- nificant increase was seen in September 1988. o1 A e © 4 — So) egy Y s1= 2 , CK oc c (©) 1 KK 7) = Thereafter, concentrations decreased until September 1989, increased again until J anuary 1990, and then decreased until July 1990. sGnRH concentrations in the discrete brain areas are shown in Figs. 7A-E. They showed a tendency to decrease from winter through summer and to increase from autumn through winter. sGnRH concentrations in the olfactory bulbs and telencephalon increased significantly during vitel- logenesis and ovulation. sGnRH concentrations in the hypothalamus also showed a similar tendency. On the contrary, no significant changes were seen in the cerebellum and medulla oblongata during vitellogenesis and ovulation. ODT 345678 9101121234567 8 910111212345678 910 1988 20 10 sGnRH (ng/g brain) 1988 Fic. 6A and B. 121 2 3 405) 678) 910d 121) 2 3, 4G 7) 89 10112 Ne 2s A SG acon 1989 1990 1989 1990 Seasonal changes in brain sGnRH contents (A) and concentrations (B) of the masu salmon from December 1987 to October 1990 and from March 1988 to October 1990, respectively. Presentation of statistical data is as in Fig. 4. Fic. 7A-E. Seasonal changes in sGnRH concentrations in olfactory bulbs (A), telencephalon (B), hypothalamus (C), optic tectum-thalamus (D) and cerebellum and medulla oblongata (E) of the masu salmon from September 1988 to October 1990. Presentation of statistical data is as in Fig. 4. sGnRH (ng/g tissue) 50) a Changes of GnRH in Female Masu Salmon Olfactory bulbs 40; D- Optic tectum-thalamus E 8 6 ek Medulla 41 Cerebellum- \ Medulla 2 Cerebellum 382 M. AMANO, K. AIDA et al. 1.5 e—e sGnRH (ng/pituitary) *—_~¢ 0 1988 Fic. 8. 200 100 (Aseyinyid 6u/6d) HYUHSsS o—o 50 4°5 678 9101112 1 DiSTAlSL6l 7 G69 101122) oi ay See eNORIOEE 1989 Seasonal changes in sGnRH contents and concentrations in the pituitary of the masu salmon from April 1988 1990 to October 1990. Presentation of statistical data is as in Fig. 4. Pituitary sGnRH contents significantly increased from July through September 1988, from May through November 1989 and from May through October 1990 as shown in Fig. 8. Pituitary sGnRH concentrations also increased significantly from May through November 1989 and from May through October 1990. cGnRH_-II in the brain and pituitary cGnRH-II was undetectable at the first two samplings (December 1987 and January 1988) and became detectable from February 1988. Brain cGnRH-II contents increased as fish grew, but were lower than those of sGnRH (Fig. 9A). Brain cGnRH-II concentrations increased signi- ficantly from August 1988 through January 1989 and from September 1989 through January 1990. On the contrary, concentrations decreased from January 1989 through September 1989 (Fig. 9B). cGnRH-II contents in the olfactory bulbs and pituitary were below the detectable limit in almost all individuals throughout the experiment. cGnRH-II concentrations in discrete brain areas are shown in Figs. 1|OA-D. They showed a tenden- cy to decrease from winter through summer and to increase from autumn through winter as those of sGnRH did, especially in optic tectum-thalamus. However, in contrast to sGnRH, no remarkable changes were observed during vitellogenesis and ovulation. 3 | DISCUSSION This is the first report which shows long term changes in two types of GnRH, sGnRH and cGnRH-II, in the brain and pituitary, and GTHs in the pituitary in fish. The observation period covers nearly the entire life cycle from hatching through ovulation. ) sGnRH was detected in the brain just after hatching, whereas cGnRH-II was first detected two months later. It is remarkable that both GnRHs appear during early developmental stages when the ovary was still in an immature stage. Both GnRH contents increased with a growth during underyearling stage. The time lag between the appearance of sGnRH and cGnRH-Il, and subsequent development of a differential distribu- tion of sGnRH and cGnRH-II suggest the differ- ence in their physiological function. sGnRH concentrations in the olfactory bulbs and the telencephalon, and pituitary sGnRH con- Changes of GnRH in Female Masu Salmon 383 cGnRH-Il (ng/brain) 1988 5) B pee ed ¢ © 2 acy 3 d) = ~ 2 cc c O (S) 1 0 1988 Fic. 9A and B. 1989 1990 ele 2hon eo) Oni CoO MOldd2 112-354 (5 647489 A0d 12 1)2)3245,6 7:8), 9) 10 1989 1990 Seasonal changes in brain cGnRH-II contents (A) and concentrations (B) of the masu salmon from December 1987 to October 1990 and from March 1988 to October 1990, respectively. Presentation of statistical data is as in Fig. 4. tents increased significantly during vitellogenesis and ovulation (Figs. 7, 8). Moreover, sGnRH concentrations in the hypothalamus showed a tendency to increase. These changes may be related to gonadal maturation, since pituitary GTH If and GTH IIf contents and GSI also increased significantly in this period (Figs. 4, 7). On the other hand, no significant changes in cGnRH-II concentration in the discrete brain areas were observed (Fig. 10). These results sug- gest that sGnRH participates mainly in vitel- logenesis and ovulation of this species possibly through the regulation of GTH synthesis and re- lease. GTH I is considered to regulate the early stages of gonadal maturation, and GTH II is considered mainly to regulate ovulation and spermiation [9]. Pituitary GTH If contents showed clear seasonal changes: contents increased from spring to autumn and decreased in winter (Fig.5A). Pituitary sGnRH contents showed stepwise increases from spring to autumn for three years (Fig. 8). Such increases in pituitary GTH If contents were likely correlated with the increase in pituitary sGnRH 384 M. AMANO, K. AIDA et al. > = ”) 2 a ro?) = r oc Cc O (S) 10 D 8 Medulla 6 4 Cerebellum- Cerebellum 2 Medulla ee —— *_ Z 1988 Fic. 10A-D. 910111212345 67 8 910111212345 67 8 910 1989 Seasonal changes in cGnRH-II concentrations in telencephalon (A), hypothalamus (B), optic tectum- 1990 thalamus (C) and cerebellum and medulla oblongata (D) of the masu salmon from September 1988 to October 1990. Presentation of statistical data is an in Fig. 4. contents. Two types of GTH I (stable and un- stable) are known to exist in the chum salmon pituitary [10]. Since GnRH extraction was under- taken under acidic conditions, unstable GTH I is considered to have dissociated to its a@ and subunits. Therefore, changes in GIH If may reflect those of unstable GTH I. In order to confirm this hypothesis, GTH I should be mea- sured by RIA; however, at present, a specific RIA for GTH I has not been established in our labora- tory. Since no significant histological changes of ovaries were observed when pituitary GITH I[f contents were high in underyearling and yearling autumn, the complete form of GTH I may not be produced or secreted during first two years. Changes in pituitary GTH II contents were simi- lar to those of GSI: after gradual increases until the second spring (two year-old fish), rapid in- creases occurred in accordance with vitellogenesis and ovulation. Changes in GTH IIf may reflect those of GTH II, since GTH II is considered to dissociate under the acidic conditions employed in GnRH extraction. sGnRH and cGnRH-II concentrations in the brain showed a tendency to increase from autumn through winter and to decrease from winter through summer (Figs. 6B, 7, 10). Since water temperature was constant throughout the experi- ment (9-10°C), photoperiod may be involved in the regulation of the synthesis and release of Changes of GnRH in Female Masu Salmon 385 sGnRH and cGnRH-II. Takashima and Yamada [12] reported that although maturation is initiated under long photoperiod, rapid vitellogenesis is induced by short photoperiod in the landlocked masu salmon “yamame.” Pituitary sGnRH contents increased significantly from July through September in underyearling, from May through November in yearling and from May through October in two year-old females, suggesting that the increase occurs under shorten- ing day length (Fig. 8). On the contrary, no significant change was observed during other periods. It may be possible to speculate from the present results that synthesis of sGnRH in the brain increases under shortening day length and sGnRH produced is transported from cell bodies to the pituitary from summer to autumn. Changes in sGnRH concentrations in the olfactory bulbs, the telencephalon and the hypothalamus support this hypothesis. cGnRH-II was mostly undetectable in the pitui- tary and no significant changes of cGnRH-II con- centrations in the discrete brain areas were observed during vitellogenesis and ovulation. cGnRH-II may not have a function of regulating GTH synthesis and release. It may function only as a neuromodulator in the brain, although cGnRH-II has the same potency to induce GTH release from the pituitary of hime salmon, Oncorhynchus nerka, in vitro (unpublished data). We have previously examined the distribution of sGnRH and cGnRH-II in the brain of masu salm- on [13]. sGnRH-immunoreactive (-ir) cell bodies were scattered in the transitional areas between the olfactory nerve and the olfactory bulb and between the olfactory bulb and the telencephalon, the ventral telencephalon, and the preoptic area, and sGnRH-ir fibers were distributed in various regions of the brain, as well as in the pituitary. On the other hand, cGnRH-II-ir cell bodies were found in the midbrain tegmentum, and cGnRH-II- ir fibers were distributed in various brain regions but not in the pituitary. The present results are in correspondence with our previous results. Gentile et al. [6] measured GnRH contents using antibody against mammalian GnRH in the tel- encephalon and hypothalamus of Venezuelan freshwater fish, “caribe colorado”, P. notatus. They found that changes of GnRH corresponded to those of GSI, especially in sexually mature female-high in May when maximal GSI is achieved. Our results nearly correspond to their results. Yu ef al. [7] reported that the sGnRH contents in the hypothalamus and pituitary were higher in sexually regressed fish compared with those in sexually mature fish under certain condi- tions of temperature acclimation (18°C). Since female masu salmon die after spawning, it is impossible to compare the data at sexually regres- sed conditions. The difference of most significance is that ConRH-II exists in the pituitary of goldfish, whereas in the pituitary of masu salmon, cGnRH- II contents are below detectable limits. Okuzawa et al. [5] measured sGnRH and cGnRH-II contents in the discrete brain areas of rainbow trout and found that the pattern of sGnRH distribution changed with age and stage of sexual maturity, but they had measured GnRH levels in 1-year-old and 3-year-old fish sampled in September. Therefore, annual changes in sGnRH levels in rainbow trout are still unknown. GnRH may also stimulate growth hormone re- lease, since sGnRH stimulates growth hormone (GH) release both in vivo and in vitro in goldfish [2, 3]. In salmonid fish, both sGnRH and cGnRH- II have a potency to stimulate GH release from the pituitary in vitro in hime salmon, Oncorhynchus nerka (unpublished data). However, cGnRH-II contents were below detectable limits by RIA and undetectable by immunocytochemistry. There- fore, if GH release is regulated by GnRH, only sGnRH_ is involved in this release in masu salmon. Future investigation on seasonal changes in the expression of GnRH and GTH genes will be required. ACKNOWLEDGMENTS We thank Dr. Hiroshi Kawauchi, Kitasato University, for providing purified GTH If, GTH IIf, and their antisera. We thank Mr. Toshio Shikama and Mr. Saburo Oda, Nikko Branch of National Research Institute of Aquaculture, for their technical assistance. We also thank Dr. Makito Kobayashi and Ms. Marcy N. Wilder, the University of Tokyo, for reading the manuscript. 386 REFERENCES Sherwood, N. M. and Lovejoy, D. A. (1989) The origin of the mammalian forms of GnRH in primi- tive fishes. Fish Physiol. Biochem. 7: 85-93. Marchant, T. A., Chang, J. P., Nahorniak, C. S. and Peter, R. E. (1989) Evidence that gonadotro- pin-releasing hormone also functions as a growth hormone-releasing factor in the goldfish. Endocri- nology 124: 2509-2518. Marchant, T. A. and Peter, R. E. (1989) Hypo- thalamic peptides influencing growth hormone secretion in the goldfish, Carassius auratus. Fish Physiol. Biochem. 7: 133-139. Yu, K. L., Sherwood, N. M. and Peter, R. E. (1988) Differential distribution of two molecular forms of gonadotropin-releasing hormone in discrete brain areas of goldfish (Carassius auratus). Peptides 9: 625-630. Okuzawa, K., Amano, M., Kobayashi, M., Aida, K., Hanyu, I., Hasegawa, Y. and Miyamoto, K. (1990) Differences in salmon GnRH and chicken GnRH-II contents in discrete brain areas of male and female rainbow trout according to age and stage of maturity. Gen. Comp. Endocrinol. 80: 116-126. Gentile, F., Lira, O. and Marcano-de Cott, D. (1986) Relationship between brain gonadotropin- releasing hormone (GnRH) and seasonal re- productive cycle of “caribe colorado”, Pygocentrus notatus. Gen. Comp. Endocrinol. 64: 239-245. 7 10 11 1 13 M. AMANO, K. AIDA et al. Yu, K. L., Nahorniak, C. S., Peter, R. E., Corri- gan, A., Rivier, J. E. and Vale, W. W. (1987) Brain distribution of radioimmunoassayable gonadot- roipin-releasing hormone in female goldfish: Sea- sonal variation and periovulatory changes. Gen. Comp. Endocrinol. 67: 234-246. Suzuki, K., Kawauchi, H. and Nagahama, Y. (1988) Isolation and characterization of two distinct go- nadotropins from chum salmon pituitary glands. Gen. Comp. Endocrinol. 71: 292-301. Suzuki, K., Kanamori, A., Nagahama, Y. and Kawauchi, H. (1988) Development of salmon GTH I and GTH II radioimmunoassays. Gen. Comp. Endocrinol. 71: 459-467. Suzuki, K., Kawauchi, H. and Nagaham, Y. (1988) Isolation and characterization of subunits from two distinct salmon gonadotropins. Gen. Comp. Endoc- rinol. 71: 302-306. Kobayashi, M., Aida, K., Sakai, H., Kaneko, T., Asahina, K., Hanyu, I. and Ishi, S. (1987) Radioimmunoassay for salmon gonadotropin. Nip- pon Suisan Gakkaishi 53: 995-1003. Takashima, F. and Yamada, Y. (1984) Control of maturation in masu salmon by manipulation of photoperiod. Aquaculture 43: 243-257. Amano, M., Oka, Y., Aida, K., Okumoto, N., Kawashima, S. and Hasegawa, Y. (1991) Im- munocytochemical demonstration of salmon GnRH and chicken GnRH-II in the brain of masu salmon, Oncorhynchus masou. J. Comp. Neurol. 314: 587- Se ZOOLOGICAL SCIENCE 9: 387-395 (1992) The Question of Functional Homology of Hatschek’s Pit of Amphioxus (Branchiostoma belcheri) and the Vertebrate Adenohypophysis Masumr Nozaki and AusBrREY GoRBMAN! Primate Research Institute, Kyoto University, Inuyama, Aichi 484, Japan and ‘Department of Zoology, University of Washington, Seattle, WA 98195, USA ABSTRACT— Using antibodies to the beta subunit of human luteinizing hormone (hLH/) and human chorionic gonadotropin, immunocytochemical evidence was obtained for gonadotropin activity in Hatschek’s pit of amphioxus, Branchiostoma belcheri. This confirms the claim by C. Y. Chang et al. [1, 2] of vertebrate-like gonadotropin in this structure, an open groove in the dorsal part of the oral cavity. If this evidence is accepted at face value, a scenario can be constructed for the evolutionary pattern of the vertebrate adenohypophysis from the protochordate Hatschek’s pit (cephalochordates) or neural gland (ascidians). Both of these structures are open to water currents in the mouth cavity. Thus, they may be able to sample thermal, chemical or pheromonal seasonally cycling clues and by gonadotropic stimulation, synchronize reproductive activity with such seasonal clues. Additional support for the idea that the early vertebrate adenohypophysis was a chemoreceptive organ comes from the fact that in cyclostomes and elasmobranchs it develops as part of the same epithelial layer and is directly contiguous with the olfactory organ. Advancement from the protochordate to vertebrate type of reproductive control involves the eventual use of sense organs and the nervous system to sample environmental changes, and the linkage of adenohypophysial function to central nervous control. The adenohypoph- © 1992 Zoological Society of Japan ysis then can be closed off from the mouth and direct environmental contact. INTRODUCTION Considerable interest by comparative endocri- nologists was ignited by the reports by C. Y. Chang et al. [1, 2] that in amphioxus immunoreac- tive responses in Hatschek’s pit to antibodies to mammalian gonadotropins could be obtained. Evidence reported from Chang’s laboratory indi- cated also that administration of ovine luteinizing hormone (LH) and prolactin to amphioxus in- creased the whole-body concentrations of sex ster- oids [2]. Furthermore, immunoreactive gonadot- ropin-releasing hormone (GnRH) and thyrotro- pin-releasing hormone (TRH) were found in Hats- chek’s pit, and saturable receptor activity for mammalian LH/human chorionic gonadotropin (hCG) and GnRH could be measured in gonads of amphioxus [2]. Accepted December 11, 1991 Received October 15, 1991 These reports, if confirmed, indicate that there are elements of a vertebrate-like mechanism for regulating reproduction in this prevertebrate pro- tochordate. A puzzling aspect of Chang’s reported data is that GnRH and gonadotropin(s) were found together in Hatschek’s pit, a shallow epithe- lial groove in the roof of the oral cavity (Fig. 1). Hatschek’s pit has long been regarded, on mor- phological grounds, to be a homologue of the vertebrate adenohypophysis [3-6]. However, although it extends toward the dorsal nerve cord, it does not contact it in the manner that the verte- brate neurohypophysis and adenohypophysis make contact, or even the neural gland and neural ganglion of ascidians. Chang’s reports stimulated efforts in other laboratories to confirm them. Among them, Fang and Wang [7] found that administration of homogenates of Branchiostoma belcheri Hats- chek’s pits stimulates testicular spermiation in young toads. Sahlin [8], in an immunohistoche- 388 M. NoZAKI AND A. GORBMAN Ss a x zh ml 4 2 \ ~ as eh ; ea Ss a Ak he A) : i 22) \ SRSA TON as Coe cade x Z 7 RAIS: Fic. 1. Transverse section through Hatschek’s pit (H) showing topographic relations of Hatschek’s pit, notochord (NO), and nerve cord (NE). OC, oral cavity. Hematoxylin and eosin stain. X75. Fic. 2. Anti-substance P immunostain. a, Transverse section through Hatschek’s pit (H). In the Hatschek’s pit, all the material reactive as substance P are evenly distributed among the cells of the pit, with possibly greater intensity in the lateral areas, whereas the nerve cord contains some darkly stained cells in the dorsal region. b, Transverse section at the level of the oesophagus and the middle part of the body, showing substance P-positive immunoreaction in cells of the nerve cord (NE, arrows). NO, notochord. a, 240; b, 310. Fic. 3. Transverse section through Hatschek’s pit (H) stained with anti-Met-enkephalin. Immunostaining is more restricted to the lateral margins of the pit, specifically limited to particular cells. NO, notochord; OC, oral cavity. UY, Gonadotropin Activity in Hatschek’s Pit of Amphioxus 389 mical study using Branchiostoma lanceolatum, found no response to antibodies to a variety of vertebrate neurohypophysial and hypophysial hor- mones (including gonadotropins) in Hatschek’s pit. However, she observed a clear reaction to an antibody to the C-terminal portion of CCK. Because of the importance of the evolutionary implications of Chang’s data, and because of the failure by others to confirm them until now, we have undertaken an immunohistochemical study of Hatschek’s pit, using the same species, Branchios- toma belcheri. MATERIALS AND METHODS Animals Specimens of Branchiostoma belcheri were col- lected during the month of April, 1987, at Tsuyazaki, a village on the northwest shore of Kyushu Island, Japan. They were collected in a large single sample of sand brought up from a depth of about 20m by dredge. They measured 2.5 to 5.9 cm and weighed 0.03 to 0.49 g. Accord- ing to Yamaguchi and Kikuchi [9], the amphioxi from various collection sites around Kyushu Island vary slightly in myotome number, but all are classified Branchiostoma belcheri, or Branchiosto- ma belcheri var. tsingtaoense, or intermediate forms between these. Chang et al. [1] stated that they used both of these forms in their experiments. Treatment The heads were removed and immersed in Bouin-Hollande sublimate for about 12 hr. They were dehydrated through a series of increasing concentrations of ethanol. After 90% ethanol, the tissue were washed in a solution containing iodine- potassium iodide in 90% ethanol for 24hr to remove deposited mercuric chloride. Tissues were embedded in Paraplast, and serial sections of 6 ~m were mounted on glass slides. Immunocytoche- mical staining was performed with a Vectastain ABC (avidin-biotin peroxidase complex) kit using a variety of polyclonal antibodies to hypothalamic, hypophysial, pancreatic and gut hormones from a number of vertebrates (Table 1). The staining procedures have been described previously [10]. Specificity of the reactions was checked by replac- ing the primary antibodies with normal sera or by using primary antibodies that were previously absorbed with corresponding antigens. RESULTS AND DISCUSSION Of the 28 antibodies tested, two yielded clear stains of cells in Hatschek’s pit (substance P and met-enkephalin; Table 1). Two yielded weaker results (hLHP and hCG; Table 1). Preabsorption of any of these four antibodies by the primary antigens blocked the staining reaction, so that in this sense, at least, the results may be considered specific reactions to the antibodies. The strength of the reaction with substance P and met- enkephalin antibodies should indicate a relatively close molecular similarity of the stained materials to the primary antigens against which the anti- bodies were produced. The weakness of the reaction to the human gonadotropin antibodies argues that molecules bearing limited structural relation to hLH# and hCG exist in Hatschek’s pit. Thus, our results confirm the report by Chang et al. [1, 2] of a vertebrate-like gonadotropin in Hats- chek’s pit. However, we could not confirm the report of the presence of immunoreactive GnRH in Hatschek’s pit. Substance P, a neurotransmitter in the central nervous system, was quite generally distributed in Hatschek’s pit, particularly in the cells of the lateral portions (Fig. 2). It also was seen in cells of the nerve cord along its entire length (Figs. 2a and b). Met-enkephalin immunoreactivity was clearly limited to cells near the lateral margins of the pit (Figs): The hLH positive cells were consistently in the deep portion of the pit, adjacent to the notochord (Fig. 4a). The anti-hCG antibody likewise reacted with cells in the deeper parts of Hatschek’s pit (Fig. 4b), but not as consistently as with the hLH? antibody. The relative weakness of the stain raises doubts concerning specificity of the gonadotropin antibody-labeling. These doubts are reinforced by the lack of immunostaining following use of two antibodies to two fish (silver carp and salmon) gonadotropins. However, arguing in favor of the significance and specificity of this gonadotropin 390 M. NozAKI AND A. GORBMAN TaBLE 1. List of antibodies used and immunoreactions to them in Hatschek’s pit Antibodies* Obtained Immunoreactivity Optimum Rererarccs to from response dilution Hypothalamic mammalian-GnRH Miles-Yeda Co. — 1000 26 hormones lamprey-GnRH J. A. King _ SRIH-14 Polysciences Co. == 500 36 AVP Raised in laboratory = 400 36 Pituitary human-LH? NIAMDD + 1000 hormones I human-FSH8 NIAMDD = 1000 human-TSHP NIAMDD = 500 human-CG Raised in laboratory +/— 1000 salmon-GTH M. Kobayashi — 500 Si silver carp-GTH M. Kobayashi — 300 38 Pituitary human-PRL NIAMDD - 1000 hormones II human-GH NIAMDD = 1000 porcine-ACTH Raised in laboratory = 400 39 a-MSH B. Baker = 300 39 salmon-PRL H. Kawauchi — 4000 40 salmon-GH H. Kawauchi = 2000 4] Pancreatic human-insulin E. Plisetskaya = 400 hormones human-glucagon Raised in laboratory _ 5000 42 porcine-PP Funakoshi Co. = 500 salmon-insulin E. Plisetskaya — 2000 10 salmon-glucagon E. Plisetskaya ~ 600 10 salmon-SRIH-25 E. Plisetskaya — 1200 10 Brain-gut CCK-8 Funakoshi Co. — 400 peptide CCK-27 N. Yanaibara = 600 porcine-VIP N. Yanaihara — 500 Substance P Raised in laboratory aE SF 800 43 Neurotensin N. Yanaihara “= 1000 Met-enkephalin Raised in Laboratory =P ar 400 39 * Abbreviations: ACTH, adrenocorticotropin; AVP, arginine vasopressin; CCK-8, cholecystokinin octapeptide; CG, chorionic gonadotropin; FSH, beta subunit of follicle-stimulating hormone; GH, growth hormone; GnRH, gonadotropin-releasing hormone; GTH, gonadotropic hormone; LH, beta subunit of luteinizing hormone; a-MSH, alpha-melanocyte-stimulating hormone; NIAMDD, U.S. National Institute of Health; PP, pancreatic polypeptide; PRL, prolactin; SRIH, somatostatin; TSH/, beta subunit of thyrotropic hormone; VIP, vasoactive intestinal polypeptide. immunostain are the following: (1) both LH and hCG are luteinizing hormones and may be pre- sumed to share antigenic determinants; (2) hFSH? antibody yielded no stain; (3) preabsorption of the two positive antibodies with their respective anti- gens blocked the immunostaining. Concerning the significance of the clear reac- tions for substance P and met-enkephalin we can say little at this time. These substances have been reported in analogous structures, the neural gland and ganglion of ascidians [11, 12], but their phys- iological significance requires additional study to define. Concerning the weak responses that we found for LH and LH-like gonadotropin, it is tempting to speculate, particularly since Chang and associates Gonadotropin Activity in Hatschek’s Pit of Amphioxus 391 te : ‘ Y? } Ms + AS GIy i Ha ay Ny aoe ye a 3 Bo Pz Ab Fic. 4. Two adjacent, but not successive, sections stained with anti-hLHf (a) and anti-hCG (b). Cells in the deepest part of Hatschek’s pit were stained consistently with anti-hLHf (arrow in a), whereas such cells (arrow in b) were stained weekly with anti-hCG. OC, oral cavity. a and b, x270. have proposed that they play a role in reproduc- tion of amphioxus. Evidence in apparent favor of this conclusion is the fact that Chang et al. [2] stimulated sex steroidogenesis in B. belcheri by injecting mammalian gonadotropins, and Fang [13] claims to have stimulated spermiation in young amphibians by injecting them with homogenates of Hatschek’s pit. Failure by Sahlin [8] to confirm in B. lanceolatum the presence of immunoreactive gonadotropin in Hatschek’s pit, might be due to seasonal factors. Species of Branchiostoma breed seasonally [14, 15], so some seasonal variation in factors regulating reproduc- tion might be anticipated. Species differences in the hormonal molecules might also play a role in producing these apparent differences. If GnRH and gonadotropin are present in Hats- chek’s pit in amphioxi, and if they have a gonad- stimulating action, as Chang et al. [1, 2] and Fang [7, 13] claim, and as partly confirmed by us, then it would appear that the protochordates had evolved a form of vertebrate-like hormonal reproductive control long before evolution of the earliest verte- brates. Amphioxus is probably a degenerate form of a more complex protochordate ancestor [16- 19], and the ascidians, likewise have evolved in a specialized direction from an ancestral form. The apparent preservation of a vertebrate-like repro- ductive regulatory mechanism in modern amphioxi, despite their apparent degeneracy, would seem to indicate that in earlier cephalochor- dates, such a mechanism may have been better developed. The fact that Hatschek’s pit is open and exposed to the environmental water should make it an appropriate organ for sampling en- vironmental thermal or chemical (pheromonal) factors that could seasonally stimulate gonadal activity. The neural gland of ascidians also retains a duct that extends directly into the stream of environmental water entering the pharynx, and therefore could involve an analogous system. In some carefully done experiments, Ruppert [20] has shown that the ciliated duct of Ascidia interrupta maintains a continues inward flow of water into the neural gland. If it were based only on the immunocytochemi- cal evidence that we have summarized here a thesis that depicts the evolution of the adenohypophysis from a chemoreceptive or olfactory structures would appear to be relatively tentative. However, some supportive evidence is available from the development of the vertebrate adenohypophysis, and also from the association of GnRH with the olfactory system. In embryos of lampreys, hagfish 392 M. NozaAkI AND A. GORBMAN a, Scanning electron micrograph; ventral view of the head of a recently hatched larval lamprey, Lampetra japonica, showing the stomodeal cleft (Sc) and nasopharyngeal opening (Np). The ciliated cells visible through the Np are the anlage of the olfactory organ; the adenohypophyseal anlage extends posteriorly from the olfactory epithelium and is hidden by an overlying fold of lip (L) tissue. b, Sagittal section of the head of a larval lamprey, Fic. 5. Lampetra japonica of about the same stage as in a. The olfactory placode is the group of very tall cells _ immediately above the label Np (nasopharyngeal opening). Extending posteriorly from it and contiguous with it at the base, is the wedge-shaped adenohypophyseal anlage (arrow). Oc, optic chiasma. a andb, X90. aandb, Reproduced with permission from Honma, Chiba and Welsch [23]. and elasmobranchs the anlage of the adenohy- pophysis is immediately contiguous with the olfac- tory placode, and it is part of the same epithelial layer [21-23] (Fig. 5). There is now a considerable literature describing the presence of immunoreactive GnRH in verte- brate embryos, as well as in adults, in the olfactory epithelium (placode), olfactory organ, olfactory tract, terminal nerves and in axons projecting from these to the hypothalamus of mammals, birds, amphibians and fishes [24-32]. Accordingly, evolution of a functional relationship between olfaction and reproduction has been an early and likely possibility. If this is a primitive form of endocrine control over reproduction via an organ that directly sam- ples pertinent environmental cues, then evolution of the more complex sense organ-hypothalamus- hypophysis form of reproductive (and other) reg- ulation can be seen as a logical further step (Fig. 6). It is of interest that the evolved vertebrate adenohypophysis which is closed off from environ- mental contact and from sampling environmental changes, still retains direct secretory sensitivity to osmotic changes. Even in vitro cultured pituitary cells respond to changes in tonicity of culture medium by changes in secretion of prolactin [33, 34]. Furthermore, Olsson [6] has proposed that prolactin cells lining and near the open duct that connects the pars distalis to the mouth in certain adult fishes, such cells may be directly responsive to environmental salinity changes in regulating prolactin secretion. Here the analogy to Hatchek’s pit and the ascidian neural gland is obvious. In considering the adaptational features that would be advantageous in the regulation of repro- ductive function, it is obvious that synchrony with seasonal environmental phenomena is highly im- portant. Synchrony of reproductive capacity of individuals within a population also is a basic Gonadotropin Activity in Hatschek’s Pit of Amphioxus 393 PROPOSED EVOLUTIONARY SCHEME FOR TROPIC FUNCTION OF ADENOHYPOPHYSIS ENVIRONMENTAL INPUTS Photoperlodic, Thermal, Chemical (Pheromonal), Tactile PROTOCHORDATES VERTEBRATES Sense organs C.N.S. Hypothalamus Hatschek's pit Neural gland Releasing factors (Mouth) Adenohypophysis | (Mouth) Gonadotropins Gonadotropins Fic. 6. Proposed evolutionary scheme from tropic func- tion of the adenohypophysis. adaptive need. Accordingly, a variety of systems have evolved in animal species to serve the pur- pose of linking reproduction to a seasonal environ- mental cue like photoperiod or temperature. Kanatani [35], for example, has explored a system in echinoderms of integumentary cells (“support- ing cells”), which secrete a gonad-stimulating hor- mone and regulate spawning. The surface location of these cells indicates a probable sensitivity to environmental factors. Hatschek’s_ pit of amphioxi, open to movement of environmental water in the mouth space, is clearly in a position appropriate for sensing pheromonal or thermal changes. Because of the small size and transparen- cy of amphioxus, it could be photo-sensitive as well. The developing vertebrate adenohypophysis is also, at first, a structure in the mouth epithelium, a fact that has invited speculations concerning its homology with the similarly situated protochor- date structures such as Hatschek’s pit and the ascidian neural gland. ACKNOWLEDGMENTS We are grateful to the staff of the Fisheries Station of Kyushu University, College of Fisheries at Tsuyazaki, for use of their research vessel and equipment used in collection of amphioxus specimens. REFERENCES 1 Chang, C.-Y., Chu Y. and Chen D. (1982) Im- munocytochemical demonstrations of luteinizing hormone (LH) in Hatschek’s pit of Amphioxus (Branchiostoma belcheri Gray). Kexue Tongbao, 27(11): 1233-1234. Pw Chane Wer Yeisen xe, Zhusp yl vandi Zhu, HH H. (1985) The reproductive endocrinology of Amphioxus. In: “Frontiers in Physiological Re- search”. Eds. by D. G. Carlick and P. I. Korner, Autstralian Academy of Science, Canberra, pp. 79- 86. 3. Tjoa, L. T. and Welsch, U. (1974) Electron micro- scopical observations on Kolliker’s and Hatschek’s pits and on the wheel organ in the head region of amphioxus (Branchiostoma lanceolatum). Cell Tis- sue Res., 153: 175-187. 4 Welsch, L. T. and Welsch, U. (1978) Histologische und elektronen-mikroskopische Untersuchungen an der praoralen Wimpergrube von Saccoglossus horsti (Hemichordata) und der Hatschekschen Grube von Branchiostoma lanceolatum. Ein Beitrag zur Phy- logenetischen Entwicklung der Adenohypophyse. Zool. Jahrb. Anat., 100: 564-578. 5 Sahlin, K. and Olsson, R. (1986) The wheel organ and Hatschek’s groove in the lancelet, Branchiosto- ma lanceolatum (Cephalochordata). Acta Zool., 67: 201-209. 6 Olsson, R. (1990) Evolution of chordate endocrine organs. In: “Progress in Comparative Endocrinolo- gy’, Eds. by A. Epple, C. Scanes and M. Stetson, Wiley-Less, New York, pp. 272-281. 7 Fang, Y.-Q. and Wang, L. (1984) The preliminary study of homogenate of the wheel organ and Hats- chek’s pit of Amphioxus on testicular development in young toad (Bufo malanostictus). Acta Biol. Exp. Sinica, 17(1): 115-117. (in Chinese with English abstract). 8 Sahlin, K. (1988) Gastrin/CCK-like immunoreac- tivity in Hatschek’s groove of Branchiostoma lan- ceolatum (Cephalochordata). Gen. Comp. Endocri- nol., 70: 436-441. 9 Yamaguchi, T. and Kikuchi, T. (1985) Representa- tive shore animals in the Amakusa Islands (2). The 10 11 12 13 14 15 16 17 18 20 Zl ij 394 amphioxus Branchiostoma belcheri. Calanus, 9: 21- SY. 3 Nozaki, M., Miyata, K., Oota, Y., Gorbman, A. and Plisetskaya, E. M. (1988) Different cellular distributions of two somatostatins in brain and pan- creas of salmonids, and their associations with in- sulin- and glucagon-secreting cells. Gen. Comp. Endocrinol., 69: 267-280. Fritsch, H. A. R., Van Noorden, S. and Pearse, A. G. E. (1979) Localization of somatostatin-, sub- stance P- and calcitonin-like immunoreactivity in the neural ganglion of Ciona intestinalis L. (Asci- diaceae). Cell Tissues Res., 202: 263-274. Georges, D. and Dubois, M. P. (1985) Methionine- enkephalin-like immunoreactivity in the nervous ganglion and the ovary of a protochordata, Ciona intestinalis. Cell Tissue Res., 236: 165-170. Fang, Y.-Q. (1985) The effects of the hypothalamic luteinizing hormone releasing hormone analogue (LHRH-A) on the spermatogenesis of amphioxus: A study of histology and ultrastructure. Acta Zool. Sinica, 31(4): 319-323. (in Chinese with English abstract). Nelson, G. E. (1968) Amphioxus in old Tampa Bay, Florida. Quart. J. Florida Acad. Sci., 31: 93-100. Wickstead, J. H. (1975) Chordata: Acrania (Cepha- lochordata). In: “Reproduction of Marine Inverte- brates”, Eds. by A. C. Giese and J. H. Pearse, Academic Press, New York, pp.283-319. Gislen, T. (1930) Affinities between the echi- nodermata, enteropneusta and chordonia. Zool. Bidrag (Uppsala), 12: 199-304. Romer, A. S. (1962) The vertebrate body. Saun- ders, Philadelphia. Barrington, E. J. W. (1965) The biology of hemichordata and protochordata. Oliver and Boyd, Edinburgh, London. Berrill, N. J. (1987) Early chordate evolution, Part 1. Amphioxus, the riddle of the sands. Int. J. Invert. Reprod. Develop., 11: 1-14. Ruppert, E. E. (1990) Structure, ultrastructure and function of the neural gland of Ascidia interrupta (Chordata, Ascidiacea): clarification of hypotheses regarding evolution of the vertebrate anterior pituit- ary. Acta Zool., 71: 135-149. Gorbman, A. (1983) Early development of the hagfish pituitary gland: evidence for the endodermal origin of the adenohypophysis. Amer. Zool., 23: 639-654. Gorbman, A. and Tamarin, A. (1985) Early de- velopment of oral, olfactory and adenohypophyseal structures of agnathans and its evolutionary implica- tions. In: “Evolutionary Biology of Primitive Fishes”, Eds. by R. E. Foreman, A. Gorbman, J. M. Dodd and R. Olsson, Plenum Press, New York and London, pp. 165-185. M. NoZAKI AND A. 23 24 25 26 21} 28 29 30 31 SZ 33) 34 35 GORBMAN Honma, Y., Chiba, A. and Welsch, U. (1990) Development of the hypophysis of the arctic lam- prey (Lampetra japonica). Fish Physiol. Biochem. ., 8: 355-364. Blahser, S. and Heinrichs, M. (1982) Immunoreac- tive neuropeptide systems in avian embryos (domes- tic mallard, domestic fowl, Japanese quial). Cell Tissue Res., 223: 287-303. Nozaki, M. (1985) Tissue distribution of hormonal peptides in primitive fishes. In: “Evolutionary Biolo- gy of Primitive Fishes”, Eds. by R. E. Foreman, A. Gorbman, J. M. Dodd and R. Olsson, Plenum Press, New York and London, pp. 433-454. Nozaki, M., Fujita, I., Saito, N., Tsukahara, T., Kobayashi, H., Ueda, K. and Oshima, K. (1985) Distribution of LHRH-like immunoreactivity in the brain of the Japanese eel (Anguilla japonica) with special reference to the nervus terminalis. Zool. Sci., 2: 537-547. Muske, L. E. and Moore, F. L. (1988) The nervus terminalis in amphibians: anatomy, chemistry and relationship with the hypothalamic gonadotropin- releasing hormone system. Brain Behav. Evol., 32: 141-150. Schwanzel-Fukuda, M. and Pfaff, D. W. (1989) Origin of luteinizing hormone-releasing hormone neurons. Nature, 338: 161-164. Kuenzel, W. J. and Blahser, S. (1991) The distribu- _ tion of gonadotropin releasing hormone (GnRH) neurons and fibers throughout the chick brain (Gal- lus domesticus). Cell Tissue Res., 264: 481-496. Norgren, R. B. and Lehman, M. N. (1991) Neurons that migrate from the olfactory epithelium in the chick express luteinizing hormone-releasing hor- mone. Endocrinology, 128: 1676-1678. Amano, A., Oka, Y., Aida, K., Okumoto, N., Kawashima, S. and Hasegawa, Y. (1991) Im- munocytochemical demonstration of salmon GnRH and chicken GnRH II in the brain of masu salmon, Oncorhynchus masou. J. Comp. Neurol. (in press). Murakami, S., Tatsunori, S., Wakabayashi, K. and Arai, Y. (1991) The ontogeny of luteinizing hor- mone-releasing hormone producing neurons in the chick embryo: possible evidence for migrating LHRH neurons from the olfactory epithelium ex- pressing a highly polysialated neural cell adhesion molecule. Neurosci. Res. (in press). Nishioka, R. S., Kelly, K. M. and Bern, H. A. (1988) Control of prolactin and growth hormone secretion in teleost fishes. Zool. Sci., 5: 267-280. Grau, F. G. and Helms, L. M. (1989) The tilapia prolactin cell: a model for stimulus-secretion cou- pling. Fish Physiol. Biochem., 7: 11-19. Kanatani, H. (1979) Hormones in echinoderms. In: “Hormones and Evolution” Ed. by E. J. W. Bar- rington, Academic Press, New York, pp. 273-307. 36 Si) 38 Sy) 40 Gonadotropin Activity in Hatschek’s Pit of Amphioxus Nozaki, M. and Gorbman, A. (1983) Immunocy- tochemical localization of somatostatin and vasoto- cin in the brain of the Pacific hagfish, Eptatretus stouti. Cell Tissue Res., 229: 541-550. Kobayashi, M., Aida, K., Sakai, H., Kaneko, T., Asahina, K., Hanyu, I. and Ishii, S. (1987) Radioimmunoassay for salmon gonadotropin. Nip- pon Suisan Gakkaishi, 53: 995-1003. Kaneko, T., Kobayashi, M., Aida, K. and Hanyu, I. (1985) Ultrastructural immunocytochemistry of gonadotrophs in the goldfish pituitary gland. Cell Tissue Res., 239: 337-342. Nozaki, M. and Gorbman, A. (1984) Immunoreac- tivity for Met-enkephalin and substance P in cells of the adenohypophysis of larval and adult sea lam- preys, Petromyzon marinus. Gen. Comp. Endocri- noleo7; 172-183. Naito, N., Takahashi, A., Nakai, Y., Kawauchi, H. 4] 42 43 595 and Hirano, T. (1983) Immunocytochemical iden- tification of the prolactin-secreting cells in the tele- ost pituitary with an antiserum to chum salmon prolactin. Gen. Comp. Endocrinol., 50: 282-291. Bolton, J. P., Takahashi, A., Kawauchi, H., Kubo- ta, J. and Hirano, T. (1986) Development and validation of a salmon growth hormone radioimmu- noassay. Gen. Comp. Endocrinol., 62: 230-238. Nozaki, M., Miyata, K., Oota, Y., Gorbman, A. and Plisetskaya, E. M. (1988) Colocalization of glucagon-like peptide and glucagon immunoreactivi- ties in pancreatic islets and intestine of salmonids. Cell Tissue Res., 253: 371-375. Nozaki, M. and Gorbman, A. (1986) Occurrence and distribution of substance P-related immunoreac- tivity in the brain of adult lampreys, Petromyzon marinus and Entosphenus tridentatus. Gen. Comp. Endocrinol., 62: 217-229. mi siemeton mona nice beebifenecstg' Peenanea his tioiiigbede: Reseed ety pA ALT Ua pas cea Slat ete RONG tidal BUUREN” cae a ee rath ry BORE sits Poe eK fi ol es Ay AL, 7 al M . a i y * . 4 * te ~ = & eS x i 7 a i Gy ieee eh? a Ps tit Les { ae i ? a 7 i ‘7 aire ‘i ; ieee aT! ‘ # ; PU ae Wee et Th Ue Ly t ort ; tt 5 ‘. iP e t : Aer; A wre k Boe Wy rw . ‘a ~~ = v 4 i B = diy z 3 ti ; = aid ; z ' = x ) i ls Ji ’ 7 % B a i i : i 5 ' 5 1 j) i % = a ; { i i ee = ra 4 nh 43 = “ _ - re 1 ~ ty Bs & . ¥ <' : _s) “s 1 re 4 * 7 bf = 3 u ; 2 : si ‘ a) . } ‘ fie 1 < } : i ‘2 LS ' hi ‘ > j A 4 i 4 t i ¥ 1 ZOOLOGICAL SCIENCE 9: 397-404 (1992) © 1992 Zoological Society of Japan Mating Behavior in Three Species of the Drosophila hypocausta Subgroup Nosuniko Asapa!, Kyoko Fustwara’, Hirosut IkKEDA>”* and Fuyuo HIHARA® ‘Biological Laboratory, Faculty of Science, Okayama University of Science, Okayama 700, Japan, *Department of Biology, Faculty of Science and Biological Institute, Faculty of General Education, Ehime University, Matsuyama 790, Japan ABSTRACT— Courtship behavior was evaluated to investigate the degree of sexual isolation in three related species, Drosophila hypocausta, D. neohypocausta and D. siamana. Patterns of male courtship behavior and the nature of courtship sounds using an oscilloscope were compared among these species. D. siamana was ethologically isolated incompletely from D. hypocausta; D. siamana females were highly receptive to D. hypocausta males, giving the average rate of insemination of 67.1%, although the reciprocal crosses gave only 13.4%. The nature of the courtship behavior and courtship sounds emitted by males showed species-specific patterns. Results of crossability tests and behavioral analyses suggested that D. siamana is a good species of the D. hypocausta subgroup of the D. immigrans species group. INTRODUCTION The Drosophila hypocausta subgroup consists of seven species including D. hypocausta and D. neohypocausta. Collecting expeditions of Dro- sophilid flies were carried out in 1979 and 1981 in southeast Asia and many specimens were collected in Malaysia and Thailand. D. hypocausta inhabits wider areas in southeast Asia including Thailand, Malaysia, Philippines and Papua New Guinea. D. neohypocausta is considered to be an endemic species of Taiwan, and D. siamana inhabits in Thailand and Malaysia. According to the morpho- logical analysis, D. siamana closely resembles D. hypocausta in general features, but differs from the latter in the shape of aedeagus and in having larger C3 fringe and two prominent bristles on the meta- tarsus of the hind legs [1]. Courtship sounds, particularly male sounds, are important for species recognition in Drosophila [2-4] as well as in other animals [5, 6]. Intrapulse Accepted February 17, 1992 Received July 7, 1991 ' To whom all correspondence should be addressed. * Deceased on April 2nd, 1987 frequency and the pulse repetition rate, or inter- pulse interval (ipi), of courtship sounds emitted by males are thought to be important factors of species discrimination for the female in acceptance of copulation [7, 8]. These findings suggest that interspecific differentiation of courtship sounds emitted by the male might act as an incipient sexual isolation mechanism during the course of speciation. In this article, the degree of reproductive isola- tion, especially sexual isolation, is analyzed and the nature of male courtship behaviors and sounds among three species is also presented. MATERIALS AND METHODS Flies The following species and strains were used: D. hypocausta, R164; collected at Palawan island of Philippines in 1979, W103; collected at Singapore in 1979, D. neohypocausta, I-Lan; collected at Chung-tou of Taiwan in 1979, D. siamana, Y110 and Y115; collected at Penang island of Malaysia in 1979, Z17 and Z28; collected at Nakhon Nayok 398 of Thailand in 1979. R164, I-Lan and Y110 strains were used for behavioral analysis. All strains were wild caught iso-female origin. Flies were reared on standard cornmeal-yeast medium at 25°C under the artificial light and dark cycle (LD=12:12). Evaluation of mating behavior and the recording of courtship sounds were performed during the light period because these species were sexually active in the morning (Asada, unpublished data). Cross experiments Mating propensity was calculated as the number of matings divided by the total number of females tested. Five females and eight males were put together in a glass vial (30 x 110 mm) for two days. To determine whether copulation had occurred or not, females were dissected and examined for the presence of sperm in the females’ internal repro- ductive organs. Approximately 100 females were tested. Evaluation of mating behavior In order to analyze male courtship behavior and repelling actions of females, two pairs of flies were put in a mating chamber (30 mm in diameter), and mating behaviors of both sexes were evaluated. The terminology of behaviors was described fol- lowing Spieth [9]. To evaluate the duration of copulation, approximately 30 pairs were placed together in a glass vial. Procedures for detecting the sound were same as TABLE 1. hypocausta subgroup N. ASADA, K. Fusrwara et al. those of Ikeda et al. [8]. The main equipment included an oscilloscope, Nihon Koden VC-7A; an amplifier, Nihon Koden AVB-9 and AVH-9; a microphone, Nihon Koden MSC-It; a camera, Nihon Koden PC-2B; and a data recorder, Sony DFR-3415. A recording cell (3x 30 mm diameter) was equipped with a microphone diaphragm, and the cell was placed inside a soundproof box. Un- anesthetized virgin flies, two females and one male, were put into the cell. Inter pulse interval (ipi) and number of cycles per pulse, and number of pulses per burst were counted in 25 samples. Male courtship sounds of D. hypocausta and D. siamana were recorded in 25 replicates for each five copulated pairs. RESULTS Crossability and productivity among three species Proportions of successful matings among three species are shown in Table 1. The average rates of females inseminated in intraspecific crosses were 84.9, 84.8 and 90.5% on average, respectively. In interspecific crosses, the rate of insemination varied from zero to 94%. No interspecific cross was found between D. neohypocausta and the other two species, showing complete ethological isolation between them. D. siamana females were highly receptive to D. hypocausta males, giving the average rate of insemination of 67.1%, whereas Percent of successful matings in intra- and interspecific crosses among three species of the D. Male D. hypocausta D. neohypocausta D. siamana — Female R164 W103 Average I-Lan YO Yis ZA ZS ee venace hypocausta RUG A WD /, I ZO) 0.0 (50) Oe AO 22209 00 12.0 (400) W103 S97 S0:0 /S5i0)h* 78225200) 0.0 (50) 240 140 17.7. 4.0 14.8 (400) Total 89.5 80.0 [84.9|(410) 0.0 (100) ID We IS 2 13).4 (UO) neohypocausta I-Lan 0.0 0.0 0.0 (100) 84.8\(125) 0.0 0.0 OL07 > VOLO 0205200) siamana SUI S40” P.O > Gil. (ZLD) 0.0 (50) 9729)| 98-0 10050, 97205 398251440) NGS TAL SAO C25 (ZAU0) 0.0 (50) 0 EA OS) 770 72.0 (445) a SD.0 S70 So U0) 0.0 (50) EO Since SRN) SS.0 » S79 (470) Z28 16:0" 6105 6825200) 0.0 (50) UB lene) SLO Wen, 82.9) (S05) Total 7/4205 KOO Stes G7 (S00) 0.0 (200) DOr Sie es 80.3 (PWS Se) Number in parenthesis: number of females dissected. Mating Behavior of Drosophila 399 the reciprocal crosses gave the lower rate of in- semination, 13.4% on average. Mating experiments were also carried out in complete dark for 48 hr using D. hypocausta and D. siamana. These two species were considered to require light to mate, because none of the 100 females tested was found to be inseminated in each intraspecific crosses. The average duration of copulation in min of D. hypocausta, D. neohypo- causta and D. siamana was 8.8 0.5, 9.7+0.33 and 14.9+0.70, respectively. No significant difference of the average duration of copulation was found between D. hypocausta and D. neohypocausta, although that of D. siamana was significantly long- er than those of the others at 0.1% level by t-test. D. hypocausta females produced fertile females and sterile males when crossed to D. siamana males. The reciprocal crosses, however, produced no viable F; flies, although a few of the 1st instar larvae was found in the cultures. Mating behavior Schematic representations of mating behaviors are shown in Figures] and 2. Male courtship sounds of D. hypocausta and D. siamana were emitted together with behaviors shown by bold- boxes in Figure 2. D. hypocausta: The male sighted a moving female approaches to the female, then repeatedly taps (Ta) the female’s body, and frequently flicks (F) both wings approximately at an angle of 80° and vibrates (V) them at the same time (A-1). The pulse sound emitted by flicking-vibration is desig- nated as the FVp sound. Then the male approaches a female much more closely behind, places his head under female’s wings, extends one wing to approximately at an angle of 20-40", vibrates both wings (A-2). The sine sound (hum- ming sound, [10]) produced during these behaviors is referred to the LVs. Finally the male attempts to C RL (3) Fic. 1. Courtship behaviors of three species belonging to the D. hypocausta subgroup. A: D. hypocausta, B: D. neohypocausta, C: D. siamana. F: flicking, f: flapping, V: vibration, S: shivering, D: drumming, R: rubbing, Ta: tapping, Th: thrusting, L: licking. 400 N. ASADA, K. Fustwara et al. A Gam) enema Stop --- in front of female? Flicking Circling Flicking Head extruding Licking Vibration Attempted copulation Copulatio Fic. 2. Attempted copulation Lopulation? Stop --- in front of female? Tapping Flicking Vibration Licking Flicking Vibration Attempted copulation Copulation A flow chart of courtship behaviors of males in three species. A: D. hypocausta, B: D. nedhypocausta, C: D. siamana. Courtship songs are emitted in movements shown by bold-boxes. mount. During wing vibration, the male con- tinuously licks or tries to lick the female’s genita- lia, and rubbs (R) the anterior lateral surface of the female’s abdomen with his forelegs. Thus, the female may receive three, at least, different kinds of stimuli at the same time from the courting male. D. neohypocausta: The male sighted a female approaches, then repeatedly taps the female’s body and moves behind her. He places his head under female wings, then drums (D) the middle dorsal surface of the female’s abdomen with his forelegs persistently for a few sec. At the same time, the male continuously licks (L) or tries to lick the female’s genitalia (B-1). The receptive female gradually spreads both wings to approximately at an angle of 100° at its maximum, then finally permits the male to mount (B-2). When the female is unreceptive, the male moves in front of the female with a crab-like motion. While moving, the male always tries to tap the female’s body. The male positioning himself in front of the female frequently thrusts (Th) at the female with his head (B-3). Thereafter, the male moves back behind the female. The male never displays wing motion during courtship. D. siamana: The male sighted the female places himself directly in front of the female’s head. He extends his hind legs so that the body is inclined to be lower side. The male raises and spreads both wings to approximately at an angle of 90° and slightly shivers for several sec. in the inclined posture; the pulse sound emitted is hardly detect- able (C-1). Thereafter the male quickly flaps (f) both wings three to four times to approximately at an angle of 180° at its maximum, emitting the pulse sound, fVp. Just after several wing-flappings, the male spreads one wing to approximately at an angle of 20-40° and vibrates it for a moment, producing the pulse sound, Vp, which is followed by the irregular sine sound, Vs (C-2). Then the male quickly moves behind the female in the crab-like manner, and displays behaviors emitting either the Vs sound or the fVp+ Vp sounds. The male tries to lick the female’s genitalia and to rub the lateral surface of the female’s abdomen with his forelegs in the head-under-wings posture, however, these are less frequent compared with the male of D. hypocausta (C-3). Figure 2 shows the flow chart of the typical case of the courtship behaviors of three species de- Mating Behavior of Drosophila 401 scribed above. Those of D. hypocausta and D. siamana are very similar each other. The flicking and vibration sounds, however, are quite different as shown in later. The flow chart of the courtship behaviors of D. neohypocausta is clearly different from the other two species; it lacks flicking and vibration of the wings and involves drumming behavior which was not observed in the other species. The nature of male courtship sounds In order to analyze the nature of male courtship sounds, an oscillogram of both D. hypocausta and D. siamana were examined. D. neohypocausta could not be used because the male of this species showed no wing movement at all. D. hypocausta: The male emitted four kinds of courtship sounds, FVs, hVs, FVp and LVs, which were clearly distinguished from each other and from those of D. siamana males by oscilloscope patterns (Fig.3A). The FVs sound was a sine sound lasting less than 50 msec. The hVs sound was a sine sound with a low amplitude. The FVp sound was a burst consisting of a train of pulses comprising 2 to 4 cycles. The average ipi was 7.8 + 0.2 msec and the average of number per burst was 15.2+1.5. This sound showed to be harmonic. A FYs ——{ih The LVs sound was a burst consisting of a train of sine sounds each of which (defined as a unit) lasted 321.07 msec on average, ranging between 227.34 and 454.26 msec. The average number of the units per burst was 19.4, the total length of a burst being approximately 6sec. The frequency was changed within a unit of the sine sound; it started with a minimum frequency, reached a maximum and then returned to the minimum. D. siamana: The male emitted three kinds of courtship sounds, f Vp, Vp and Vs (Fig. 3B). The fVp sound was followed by the Vp sound with an average interval of 68.3msec. The Vs sound always followed immediately after the Vp. Thus, a set of three sounds was produced by serial courtship behaviors including wing flapping, spreading and vibration, however it was not able to distinguish behaviors between emitting Vp and Vs. The sine sound with a low amplitude was detected when the male extended a single wing and vibrated both wings in the head-under-wings posture be- hind the female. According to the oscilloscopic pattern, the nature of this sound was essentially the same as the Vs sound, however the former lasted longer than the latter. The fVp sound consisted of 2 to 4 pulses (one pulse being emitted single wing flapping) each of 4 to 5 cycles, with hVs FVs —= FYp 50 ms LVs AANA gi ae —— anaes Vs Fic. 3. A AANA ta era t Ane fh x Them tN etme aN nore ae et a ree ——4 Osilloscope patterns of courtship sounds emitted by courting males. A: D. hypocausta, B: D. siamana. FVs: sine sound emitted by flicking-vibration, hVs: sine sound emitted by flicking-vibration, FVp: pulse sound emitted by flicking-vibration, LVs: sine sound emitted by flicking-vibration, fVp: pulse sound emitted by flapping-vibraion, Vp: pulse sound emitted by flapping, Vs: sine sound emitted by flapping. For details, see the text. 402 N. ASADA, K. FUusIWARA et al. Fic. 4. Oscilloscope patterns of courtship sounds of the F; male obtained by crossing between D. hypocausta and D. siamana. For abbreviations, see lengend to Fig. 3. TABLE 2. Courtship behaviors either of D. hypocausta- or of D. siamana-type in males of the backross generation (BC1) [(hypo ? xsia S)Fi ? Xhypo S|>BCI Ff hypocausta-type —_ siamana-type I) 0 [((hypo $ Xsia $)F, $ Xsia S|—>BC1 Sf hypocausta-type slamana-type LOmnae: 14 no Total 7 26 no Total 12 45 hypo: D. hypocausta, sia: D. siamana, no: no courtship display. average ipi of 68.3msec. The Vp sound was a burst consisting of a train of pulses with the average ipi of 12.3 msec and average number of pulses per burst of 7.2. The nature of courtship sound in F, and back- crossed hybrids The genetic basis of the courtship behaviors was examined using F, and back-crossed hybrid males (sterile) occurred between D. hypocausta and D. siamana. Results are shown in Figure 4. The courtship behavior of the F; hybrid males origi- nated from crosses between D. hypocausta females and D. siamana males was similar to that of D. hypocausta males except lacking wing display in front of females, and the courting male emitted two kinds of courtship sounds, FVp and LVs. LVs sound of hybrid males could be clearly distin- guished from that of D. hypocausta males by oscilloscope patterns showing an irregular pattern. No sound resembling to that emitted by D. siama- na males could be detected in F, hybrid males. Back-crosses (BC1) males were emerged from crosses in F, females orginated from crosses be- tween D. hypocausta females and D. siamana males and D. hypocausta (or D. siamana) males. Results of the BC1 male courtship behavior are summarized in Table 2. D. hypocausta-type be- havior involves a_ series of licking-rubbing- vibration, and that of D. siamana-type involves flapping. When BC1 males were derived from the crosses between hybrid females and D. hypocausta males, only D. hypocausta-type behavior was observed. BC1 males derived from the back crosses of F; females to D. siamana males segre- gated into statistically equal number of flies having D. hypocausta-type behaviors or D. siamana-type behavior. Thus, D. siamana-type behavior seemed to be controlled by an autosomal recessive gene(s) allelic to dominant gene(s) of D. hypocausta. DISCUSSION The male of D. hypocausta and D. siamana are readily distinguishable from each other by the degree of coloration of the body; the aged male of D. hypocausta is characterized by a black abdo- men, thorax and legs, while the male of D. siama- Mating Behavior of Drosophila 403 na has a dark brown abdomen, thorax and brown legs [1]. It is hard to distinguish females of the two species from each other on the basis of external morphology. D. hypocausta females are much more discriminatory in acceptance of males than D. siamana females. This may be closely associ- ated with facts that D. hypocausta males emit simultaneously at least three kinds of stimuli which are released through licking, rubbing and wing vibration behind the female, whereas the male of D. siamana \ess frequently displays these be- haviors. LVs sound emitted by males of D. hypocausta is species-specific song and is not observed in the other Drosophila species. For three species, it may be true that visual stimuli are important to find and/or to discrimin- ate the partner, showing that these species are completely dependent on light for copulation; no copulation occurred for 48 hr in the dark for D. hypocausta, D. siamana and possibly for D. neohy- pocausta. The importance of visual stimuli is suggested by another fact that males of the three species moved around the female in a crab-like behavior, always facing the female, during courtship. Auditory stimuli may not be included in the SMRS [11] of D. neohypocausta, since the male never showed wing displays during courtship. The male of this species persistently tries to contact physically with the female through tapping, thrust- ing and drumming motions when he is courting. Essential stimuli responsible for mating success, thus, may be chemical as well as visual [12]. It is of interest that there are significant differ- ences in courtship behaviors between D. hypo- causta and D. siamana which are thought to be closely related species on the basis of morphology and the interspecific hybridization test. Usually, differences both in the behavioral pattern and in the nature of the stimulus are not qualitative but quantitative between closely related species. For example, Spieth [9] found differences in only three elements out of 19 visible courtship behaviors tested between D. melanogaster and D. simulans. Ewing and Bennet-Clark [2] could not reveal the difference in the oscilloscopic pattern except for the ipi between these species. The ipi is thought to be one of the most important characters for the discrimination of species in related species. Results obtained from hybridization tests sug- gested that D. hypocausta-type behaviors such as emitting FVp and LVs sounds may be determined by autosomal dominant genes. D. siamana-type behaviors were found only in the back-crossed generation, suggesting that genes are autosomal recessive. Thus, it is likely that behavioral differ- ences including courtship sounds between these species are based on genetic differences. As mentioned above, D. siamana was reconfirmed as species belonging to the D. hypocausta subgroup through genetic and behavioral analyses. Howev- er, the evolutionary process to diffferentiate the genetic system controlling courtship behaviors be- tween species may be a subject for futures studies. ACKNOWLEDGMENTS This work was mainly supported by the funds of the Overseas Scientific Expedition in 1971 (No. 7114), 1979 (No. 404149) and 1980 (504344), and Grant-in-Aid (No. 548001) from the Ministry of Education, Science and Culture of Japan. REFERENCES 1 Hihara, H. and Lin, F.-J. (1984) A new species of Drosophila hypocausta subgroup of species from Malaysia and Thailand (Diptera: Drosophilidae: Drosophila). Bull. Inst. Zool. Acad. Sinica, 23: 205- 209. 2 Ewing, A. W. and Bennet-Clark, H. C. (1968) The courtship songs of Drosophila. Behaviour, 31: 288- 301. 3. Ewing, A. W. and Miyan, J. A. (1986) Sexual selection, sexual isolation and the evolution of song in the Drosophila repleta group of species. Anim. Behav., 34: 421-429. 4 Wheeler, D. A., Fields, W. L. and Hall, J. C. (1988) Spectral analysis of Drosophila courtship songs: D. melanogaster, D. simulans, and their interspecific hybrid. Behav. Genet., 18: 675-703. 5 Bentley, D. R. and Hoy, R. R. (1972) Genetic control of cricket song patterns. Anim. Behav., 20: 478-492. 6 Gerhardt, H. C. (1978) Temperature coupling in the vocal communication system of the grey frog Hyla versicolor. Science, 199: 992-994. 7 Bennet-Clark, H. C. and Ewing, A. W. (1969) Pulse interval as a critical parameter in the courtship song of Drosophila melanogaster. Anim. Behav., 17: 755-759. 8 10 404 Ikeda, H., Takabatake, I. and Sawada, N. (1980) Variation in courtship sounds among three geo- graphical strains of Drosophila mercatorum. Behav. Genet., 10: 361-375. Spieth, H. T. (1952) Mating behavior within the genus Drosophila (Diptera). Bull. Amer. Mus. Nat. Hist., 99: 399-474. Schilcher, F. v. (1976) The role of auditory stimuli in the courtship of Drosophila melanogaster. Anim. 11 12 N. AsapDA, K. Fustwara et al. Behav., 24: 18-26. Patterson, H. E. H. (1978) More evidence against speciation by reinforcement. South Afr. J. Sci., 74: 369-371. Cobb, M. and Jallon, J.-M. (1990) Pheromons, mate recognition and courtship stimulation in the Drosophila melanogaster species sub-group. Anim. Behav., 39: 1058-1067. ZOOLOGICAL SCIENCE 9: 405-412 (1992) © 1992 Zoological Society of Japan Minute Protrusions of Ascidian Tunic Cuticle: Some Implications for Ascidian Phylogeny Eurcut Hirose!, TERUAKI NISHIKAWA, YASUNORI SAITO° and Hrrosul WATANABE* '.3.4¢himoda Marine Research Center, University of Tsukuba, Shimoda, Shizuoka 415, and *Biological Laboratory, College of General Education, Nagoya University, Chikusa-ku, Nagoya 464-01, Japan ABSTRACT—Fine structure of the ascidian tunic cuticle was studied by transmission and scanning electron microscopy in 26 species belonging to 10 families. In 11 species of 8 families, the cuticular surface is ornamented with numerous minute protrusions that are papillate in shape and usually up to 100 nm in height. The present results in addition to the previous ones [1] gave us more information about morphology of protrusions and their distribution in 51 species ranging over 13 families out of the 15 ones of ascidians. It reveals a general tendency that the protrusions occur in the limited families and related species have the protrusions of similar size. Thus, the protrusions proved to have some features of certain phylogenetic significance in many cases. Brief references were made to the taxonomic position of such problematic genera as Pterygascidia in the Cionidae and Sorbera in the Hexacrobylidae on the basis of the present study. INTRODUCTION The tunic is a gelatinous or leathery integument which is peculiar to the animals belonging to the subphylum Urochordata (=Tunicata). The asci- dian tunic is composed of ground substance (tunic matrix) overlaid by a thin cuticle. The surface of the tunic cuticle is sometimes, though never al- ways, specialized to have numerous minute protru- sions, which are mostly papillate in shape and up to 100nm high. Such a structure was firstly re- ported by Katow and Watanabe [2] and Milanesi er al. [3] in two species of the family Botryllidae. Then, Hirose et al. [1] detected the protrusions in 18 species of ascidians belonging to 5 families, out of the 25 species of 9 families, and they suggested that the protrusions have some implications for Accepted November 25, 1991 Received October 2, 1991 " Present address: Department of Biology, Keio Uni- versity, Yokohama, Kanagawa 223, Japan. * Present address: Tokyo Kasei Gakuin Tsukuba Col- lege, Tsukuba, Ibaraki 305, Japan. > To whom requests for reprints should be addressed. ascidian phylogeny. To examine this suggestion closely, we should have more information. In this study, fine struc- ture of tunic cuticle was newly described in 26 species of 10 families. Thus, we now get informa- tion about the protrusions from 51 species ranging over 13 families, although not yet examined speci- mens of the two remaining families (Octacnemidae and Plurellidae). On the basis of the information, we will give a comment on the phylogenetic sig- nificance of the stated structure. MATERIALS AND METHODS Animals and prefixation Specimens of Leptoclinides echinatus, Clavelina viola, Ascidia ahodori, Ascidia zara, Ascidia sp. (cf. tapni) and Pyura mirabilis were collected in Shimoda, Shizuoka Pref., and fixed in a solution containing 2.5% glutaraldehyde, 0.45 M sucrose and 0.1 M cacodylate (pH 7.4). Ascidia gemmata and Chelyosoma siboja were collected in Asa- mushi, Aomori Pref., and fixed in 2.5% glutar- 406 E. Hirose, T. NISHIKAWA et al. aldehyde-seawater. Specimens of the following species were fixed in 10% formol-seawater or 10% formol-water: Syndiazona grandis collected off Minabe, Wakayama Pref.; Adagnesia_vesicu- liphora off Fukushima Pref.; Corella sp. (cf. japo- nica) in Kuroshima, Okinawa Pref.; Eugyrioides glutinans off Oga Pen.; Molgula manhattensis in the Nagoya Harbor, Aichi Pref.; Molgula tectifor- mis in the Mutsu Bay; Chelyosoma yezoense and Cnemidocarpa clara from Otsuchi, Iwate Pref.; Cnemidocarpa irene, Polycarpa cryptocarpa kro- boja and Polycarpa maculata taken off the Oki Island, Shimane Pref.; Ciona edwardsi and Ciona roulei [cf. 4] collected around Banyuls-sur-Mer (France); Ciona intestinalis in Napoli (Italy); and Sorbera unigonas during the Iucal cruise from N. Atlantic (47°30°N, 9°35°W; 4217-4366 m deep). Pterygascidia longa was collected off Zamami, Okinawa, and fixed in 70% ethanol. The paratype specimens of Polyandrocarpa stolonifera, fixed in 70% ethanol, are deposited in the Shimoda Marine Research Center. Electron microscopy Samples of tunic, several millimeters square in size, were cut off from the above-listed specimens. Samples from the specimens in the fixative con- taining cacodylate were washed with 0.45 M su- crose solution buffered with 0.1 M cacodylate (pH 7.4), while samples from the specimens in the fixatives without cacodylate were washed with filtrated seawater. Then, the samples were post- fixed with 0.1% osmium tetroxide solution buf- fered with 0.1M cacodylate (pH7.4) for 1.5 hours, and afterward, dehydrated through an etha- nol series. For scanning electron microscopy (SEM), they were dried in a critical point dryer, sputter-coated with Au-Pd, and examined by Hitachi S-570 scanning electron microscope at 15 to 20kV. For transmission electron microscopy (TEM), they were cleared in n-butyl glycidyl ether, and embedded in a low viscosity resin [5]. Thin sections were doubly stained, and examined by Hitachi HS-9 transmission electron microscope at 75 kV. RESULTS Tunic cuticle is an electron dense layer covering tunic matrix, and its structure varies from species to species. In some species, the cuticle forms minute protrusions. Sometimes there is a layer of moderate electron density or a zone of the mixture of tunic matrix and cuticular material under the cuticle, and it is called “subcuticle” by De Leo et al. [6]. Subcuticle is almost indistinguishable from the cuticle, when they are both of high electron density and thick, as seen in some solitary species in the Corellidae and Styelidae. Thickness of the cuticle and subcuticle is revealed variable in differ- ent species, and in different parts of tunic in a specimen. When the cuticle and/or subcuticle are thick, the tunic is usually hard, and leathery or cartilaginous. The protrusions shown in the pre- sent study are papillate in shape without excep- tion. In the suborder Aplousobranchia, two species are studied. In Leptoclinides echinatus the cuticle is flat and thin, about 10 nm thick, and its surface shows a fuzzy appearance (Fig. 1A). In Clavelina viola the cuticular surface has minute protrusions, about 30-40 nm high (Fig. 1B). In the suborder Phlebobranchia, only 2 species out of the 13 studied species have minute protru- sions. In the family Cionidae, Syndiazona grandis and three Ciona species have a flat cuticle accom- panied with a subcuticle, but no protrusions (Fig. 1C). Thickness of the subcuticle ranges from about 30 nm (in C. edwardsi) to more than 300 nm (in C. roulei). On the other hand, Pterygascidia longa of the same family has minute cuticular protrusions, about 60 nm in height (Fig. 1D). In the Ascidiidae, the four species of Ascidia have a flat cuticle without any protrusions; its thickness is about 20-30 nm (in A. ahodori, A. sp. (cf. tapni) and A. zara) or more than 100 nm (in A. gemma- ta). In the Agnesiidae, Adagnesia vesiculiphora has minute cuticular protrusions, only about 20 nm high (Fig. 1E). In the Corellidae, all the three species examined here have no minute protru- sions. Corella sp. (cf. japonica) has a thin cuticle, only about 30 nm, while the two species of Chely- osoma have a thick one, indistinguishable from the underlying subcuticle (Fig. 1F). Minute Cuticular Protrusions of Ascidian Tunic 407 i Fic. 1. Fine structure of the tunic cuticle in the order Enterogona. Tunic matrix positions in lower part of each figure in TEM. (A) Leptoclinides echinatus (TEM). (B) Clavelina viola (TEM). (C) Ciona roulei (TEM). (D) Pterygascidia longa (SEM). (E) Adagnesia vesiculiphora (TEM). (F) Chelyosoma siboja (TEM). Arrowheads indicate minute protrusions. Scale bars, 0.2 um. In the suborder Stolidobranchia, the minute cuticular protrusions are found in all the examined species excluding some solitary ones in the Styeli- dae. A strange styelid, Polyandrocarpa stolo- nifera, which propagates by stolonic budding but without any functional connection among the zooids [see 7], has the protrusions, 100 nm or less in height. Among the 6 examined solitary styelids, Cnemidocarpa clara has the protrusions about 100 nm (Fig. 2A, B), and Styela clava has smaller ones (about 30 nm high) and a thick subcuticle under- lying the cuticle. The other four solitary styelids lack the protrusions; the tunic is leathery or cartila- ginous, with the thick cuticle and subcuticle (Fig. 2C). In Pyura mirabilis of the Pyuridae, minute protrusions are about 100 nm high, and the sub- E. Hirose, T. NISHIKAWA et al. 408 Minute Cuticular Protrusions of Ascidian Tunic 409 cuticle has a complex structure (Fig. 2D). In the Molgulidae, the protrusions are found in all of the 3 examined species. They are about 40-50 nm high (in Molgula manhattensis and M. tectiformis) or 30 nm (in Eugyrioides glutinans) (Fig. 2E, F), and they are rather lower than those seen in the other species in the Stolidobranchia. In the suborder Aspiraculata, we could examine only a signle species, Sorbera unigonas. The cuticle has minute protrusions, about 80 nm in TABLE 1. Species Order Enterogona Suborder Aplousobranchia Family Polyclinidae Aplidium pliciferum A. yamazil Family Didemnidae Diplosoma mitsukurii Didemnum moseleyi *Leptoclinides echinatus Family Polycitoridae Clavelina miniata *C. viola Polycitor proliferus Suborder Phlebobranchia Family Cionidae *Syndizona grandis Ciona savignyi *C. edwardsi *C. intestinalis *C. roulei *Pterygascidia longa Family Perophoridae Perophora japonica P. multiclathrata Family Ascidiidae Ascidia sydneiensis *A. ahodori *A. gemmata *A. zara Solitary (S) or Colonial (C) height (Fig. 2G, H). The protrusions are similar in size and shape to those usually seen in the Stoli- dobranchia. DISCUSSION Table 1 represents a summation of the present results and the previous ones [1] in terms of the presence or absence and height of the minute cuticular protrusions. So far as these results are Presence and height of minute cuticular protrusions of ascidian tunic Approx. height of protrusions (nm) C 30 C 60 C absent C absent C absent C 30 C 30-40 € 40 C absent S absent S absent S absent S absent S 60 C absent C absent S absent S absent S absent S absent Fic. 2. Fine structure of the tunic cuticle in the order Pleurogona. (A, B) Cnemidocarpa clara (A: TEM, B: SEM). (C) Cnemidocarpa irene (TEM). (D) Pyura mirabilis (TEM). (E) Molgula manhattensis (TEM). (F) Molgula tectiformis (SEM). (G, H) Sorbera unigonas (G: TEM, H: SEM). Arrowheads indicate minute protrusions, asterisks debris. Scale bars, 0.2 um. 410 E. Hirose, T. NISHIKAWA et al. TABLE 1. (continued) Species Solitary (S) or Colonial (C) Approx. height of protrusions (nm) *A. sp. (cf. tapni) Family Agnesiidae *Adagnesia vesiculiphora Family Corellidae *Chelyosoma siboja *C. yezoense *Corella sp. (cf. japonica) Order Pleurogona Suborder Stolidobranchia Family Botryllidae Botryllus primigenus B. scalaris B. sexiens B. schlosseri Botrylloides fuscus B. lentus B. simodensis B. violaceus Family Styelidae Metandrocarpa uedai Polyandrocarpa misakiensis *Polyandrocarpa stolonifera Polyzoa vesiculiphora Symplegma reptans *Cnemidocarpa clara *C. irene *Polycarpa_ cryptocarpa_ kro- boja *P. maculata Styela plicata *S. clava Family Pyuridae Halocynthia roretzi (Type A) Herdmania momus *Pyura mirabilis Family Molgulidae *Eugyrioides glutinans *Molgula manhattensis *M. tectiformis Suborder Aspiraculata Family Hexacrobylidae *Sorbera unigonas “ protrusions are not papillate. > subcuticle has complex structure. S AAA DAaAAA © A AAA OA oO ea AnD S species firstly examined for the protrusions in this study. absent 20 absent absent absent 100 100 100 100 100 100 100 100 100 100 100 100 100 100 absent absent absent absent 30 50? 100 100° 30 40-50 40-50 80 Minute Cuticular Protrusions of Ascidian Tunic 4i1 concerned, the shape of the protrusions is always papillate only except in Halocynthia roretzi, which has irregular-shaped ones [1]. This table shows clearly such a general tendency that the protru- sions occur in the limited families and the related species have the protrusions of similar size. Thus, we may reasonably conclude that the mentioned features are of certain phylogenetic significance. On the other hand, we cannot make any references to the function of the protuberances. In the suborder Aplousobranchia, the protru- sions are found in the Polyclinidae and Polycitor- idae. The protrusions are about 50 nm or less in height. In contrast, no protrusions are found in the Didemnidae. In the suborder Phlebobranchia, most species lack the protrusions. Exceptionally, however, we could find them in Pterygascidia longa in the Cionidae and Adagnesia vesiculiphora in the Agnesiidae. This fact reminds us of Kott’s long- held opinion that “Ciallusiinae” represented by the genus Pterygascidia should be included as a sub- family in the family Agnesiidae, which contains Adagnesia as a member of the other subfamily Agnesiinae [ref., 8]. Kott [9] paid attention to the similarity in the mantle musculature between these two subfamilies delimited by herself (for discus- sions on the taxonomic position of Prerygascidia in terms of macroscopical morphology, see Tokioka [10]). Further examinations of Kott’s opinion are expected on the basis of more information about the protrusions from the related species. Among the 25 examined species of the suborder Stolidobranchia, only the four solitary species in the Styelidae lack the protrusions. It seems to be curious that Cnemidocarpa clara and Styela clava have the protrusions, while the respective conge- ners C. irene and S. plicata lack them. It should be noted that such inconsistencies are limited to the subfamily Styelinae in the Styelidae. The lack of the protrusions in the four species may be related to the fact that the cuticle is rather thicker in these species than the others in this suborder. In the other stolidobranchians, the protrusions are papil- late and about 100 nm high. Exceptionally, they are of smaller size in Styela clava and the three molgulid species, or are not papillate in Halocy- nthia roretzi. It is unclear about the phylogenetic significance of these exceptions at present. There is a hot controversy on the systematic position of the family Hexacrobylidae in the sub- order Aspiraculata. Monniot et al. [11] separated this family from the class Ascidiacea and estab- lished for it a new class Sorberacea of the subphy- lum Urochordata (=Tunicata). The reason for this treatment was that they considered such fea- tures of this family to be quite unique among tunicates, as the virtual lack of the branchial sac, the existence of the “cordon nerveux dorsal”, and the histological peculiarities of the gut [also see 12]. On the contrary, Kott has been holding the opinion that the family Hexacrobylidae should be regarded as very closely related (or possibly even joinable) to the family Molgulidae of the suborder Stolidobranchia and therefore never mertis a differ- ent suborder. Kott [13] regarded the unique features of this group mentioned by Monniot et al. as due merely to a high adaptation to its deep-sea life, and claimed that its basic morphological fea- tures were shared with the Molgulidae, such as the presence of a kidney, the arrangement of gonads, the thin but tough and fibrous tunic, etc. On the other hand, Nishikawa [14] supported the tradi- tional treatment of the Hexacrobylidae as the single constituent of the suborder Aspiraculata, although acknowledging the uniqueness of this group strongly claimed by Monniot eft al. The present study revealed that the protrusions found in Sorbera unigonas were similar in shape and size to those of the stolidobranchian species rather than of the aplousobranchian and phlebobranchian ones. This fact may favor Kott’s view, but it does not necessarily exclude the other two taxonomic treatments if we regard the mentioned similarity as a convergent nature. On the other hand, a closer examination shows that the protrusions of S. un- igonas are much higher than those of the examined species in the Molgulidae (about 80 nm in the former, instead of 30-50 nm in the latter). This might suggest the somewhat distance of phylo- genetic position between Sorbera and the Molgu- lidae. Accordingly, we prefer the tranditional systematic treatment of this group at present. More information should be gained as to the fine structure of tunic in some other species of the Molgulidae and Hexacrobylidae. 412 As shown above, some features seen in the fine structure of tunic cuticle have certain phylogenetic significance. We expect that more attentions will be paid to the fine structures in terms of the plylogeny of not only ascidians but also the other tunicates. ACKNOWLEDGMENTS We wish to thank Drs. T. Hirata, K. Ito, C. & F. Monniot, F. Lafargue, T. Numakunai, T. Ohtsuka and H. Uchida for kindly providing specimens. This study is supported by Grant-in-Aid for Co-operative Research (A) 401304007 from the Ministry of Education, Science and Culture of Japan. This report includes contributions from the Shimoda Marine Research Center, No. 538. REFERENCES 1 Hirose, E., Saito, Y. and Watanabe, H. (1990) Minute protrusions of the cuticle: fine surface struc- tures of the tunic in ascidians. J. Morph., 204: 67- 73s 2 Katow, H. and Watanabe, H. (1978) Fine structure and possible role of ampullae on tunic supply and attachment in a compound ascidian, Botryllus pri- migenus Oka. J. Ultrastruct. Res., 64: 23-34. 3 Milanesi, C., Burighel, P., Zaniolo, G. and Sabba- din, A. (1978) The structure and the fate of the test cuticle during the fusion-nonfusion reaction in colo- nies of Botryllus schlosseri (Tunicata). Boll. Zool., 45: 83-86. 4 Lambert, C., Lafargue, F. and Lambert, G. (1990) Preliminary note on the genetic isolation of Ciona species (Ascidiacea, Urochordata). Vie Milieu, 40: 10 11 1 13 14 E. Hirose, T. NISHIKAWA et al. 293-295. Kushida, H. (1980) An improved embedding method using ERL4206 and Quetol 653. J. Electron Microsc., 29: 193-194. De Leo, G., Patricolo, E. and Frittitta, G. (1981) Fine structure of the tunic of Ciona intestinalis L. II. Tunic morphology, cell distribution and their func- tional importance. Acta Zool. (Stockh.), 62: 259- Die Kawamura, K. and Watanabe, H. (1981) Studies of Japanese compound styelid ascidians. III. A new, possibly asexual Polyandrocarpa from Shimoda Bay. Publ. Seto mar. Biol. Lab., 26: 425-436. Kott, P. (1985) The Australian Ascidiacea Part I, Phlebobranchia and Stolidobranchia. Mem. Qd Mus., 23: 1-440. Kott, P. (1969) Antarctic Ascidiacea. Antarct. Res. Ser. Washington, 13: 1-239. Tokioka, T. (1971) The ascidian genera Pterygasci- dia Sluiter, 1904 and Ciallusia Van Name, 1918. Zoologische Mededelingen, 45: 119-125. Monniot, C., Monniot, F. and Gaill, F. (1975) Les Sorberacea: Une nouvelle classes de tuniciers. Arch. Zool. exp. gén., 116: 77-122. Monniot, C. and Monniot, F. (1990) Revision of the class Sorberacea (benthic tunicates) with descrip- tions of seven new species. Zool. J. Linn. Soc., 99: 239-290. Kott, P. (1989) The family Hexacrobylidae Seelin- ger, 1906 (Ascidiacea, Tunicata). Mem. Qd. Mus., 27: 517-534. Nishikawa, T. (1986) Classification and phylogeny of the Ascidiacea. In “Systematic Zoology” (Dobut- su Keitobunruigaku), Nakayama Shoten, Tokyo, Vol. 8 (3), pp. 244-264 (In Japanese), ZOOLOGICAL SCIENCE 9: 413-421 (1992) © 1992 Zoological Society of Japan A New Hydromedusa of the Genus Eirene (Leptomedusae; EKirenidae) from Toba, Japan SHiIn Kusota! and TAKUSHI Horita~ 'Zoological Institute, Faculty of Science, Hokkaido University, Sapporo 060, and *Toba Aquarium, Toba, Mie 517, Japan ABSTRACT—A new hydromedusa Eirene lacteoides (Leptomedusae; Eirenidae) from Toba, Mie Prefecture, Japan is described based on the laboratory-reared 18 male medusae and 33 immature ones. The new species resembles Eirene lactea (Mayer, 1900) in having distal projections of peduncle, but can be distinguished by the presence of an adaxial papilla on every well-developed tentacular bulb and even on some small tentacular bulbs bearing very short tentacles. The development of medusa, the nematocyst equipment, and the distributional records of E. lacteoides are also described. INTRODUCTION Many mature and immature medusae of an undescribed form of the family Eirenidae were obtained by culture. They were at first found in seawater tanks of the Toba Aquarium as small immature leptomedusae. In Japanese waters six species representing five genera of the family Eire- nidae [1] have hitherto been found [2-11]: Eirene hexanemalis (Goette, 1886) [4], Eirene menoni Kramp, 1953 [8], Eutima japonica Uchida, 1925 [2, 9], Eutonina indicans (Romanes, 1876) [3, 5, 6], Tima formosa L. Agassiz, 1862 [2, 5], and Eugym- nanthea japonica Kubota, 1979 [10, 11]. Except for the last species, their description is originally based on mature medusae collected from the natu- ral seas in Japan. The hydromedusa in question, though obtained by culture, is apparently different from all known species of the Eirenidae. In the present paper characteristics of this medusan spe- cies are evaluated based on 51 mature and imma- ture specimens. Accepted December 26, 1991 Received August 6, 1991 ' Present address: Japan. Seto Marine Biological Laboratory, Kyoto University, Shirahama, Wakayama 649-22, MATERIALSAND METHODS Very young medusae, about 1 mm in diameter, provided with four tentacles, occurred in seawater tanks of the Toba Aquarium, Mie Prefecture, Japan on August 12-26, 1989. Their occurrence was also detected during the period between February 4 and April 21, 1990. In these tanks the natural seawater taken just in front of the Toba Aquarium was stored at room temperature. Many such immature medusae without any trace of gonads were picked up and reared to maturity in either 2800 ml, 900 ml, or 200 ml vessels filled with natural seawater from Toba or occasionally in artificial seawater (Jamarin U) at 25-28°C, being fed with Artemia nauplii for up to 69 days. Eight- een mature medusae selected and examined were all males (Nos. 1-18). The 13 specimens were measured immediately after being narcotized with MgCl, solution, while the others were measured after preservation in formalin-seawater (see Tables 1, 5). The nematocyst equipment was examined on one living 45-46 day old specimen (No. 5) under a phase-contrast microscope, and Figs. 7-10 were drawn on this occasion. The development of medusa was observed not only on these specimens. The other 33 immature ones were also reared for 6-11 days and then preserved in formalin- seawater. Among them the development was examined in detail on two medusae (Nos. 11, 12) 414 S. KuBoTA AND T. Horita as shown in Table 5. They were reared in 900 ml vessels filled with seawater from Toba at 26-27 C, fed sufficiently with Artemia nauplu, and seawater was changed every day. Figures 1-6 were drawn from specimens preserved in formalin-seawater solution. Figures 1-10 were made, with a drawing apparatus, Nikon SMZ 10 and Olympus BH-DA. Eirene lacteoides n. sp. (Figs. 1-11) [Japanese name: Kobu-eirene-kurage, new] Type-series The type-series 1s deposited in the collection of the Zoological Institute, Faculty of Science, Hok- kaido University, Sapporo, Japan [ZIHU-498 (holotype: specimen number 1); ZIHU-499 (para- types: Nos. 2-4, 6-8)], Seto Marine Biological Laboratory, Kyoto University, Japan [SMBL Type Nos. 370-372 (paratypes: Nos. 9, 12, and 33 im- FA [ é C3 \ " A CERI LMS Fe mature medusae)], Institut des Sciences Naturelles de Belgique, Bruxelles, Belgium [U.L.B.Z.J.B.C. 2E (paratypes: Nos. 10, 11)], British Museum (Natural History), London, UK [1991. 3. 1. 1-2 (paratypes: Nos. 13, 14)], Royal Ontario Museum, Toronto, Canada [ROMIZ B1152-1153 (para- types: Nos. 15, 16)], National Science Museum (Nat. Hist.), Tokyo, Japan [NSMT-Co 562 (para- type: No. 17)], and Toba Aquarium, Toba, Mie Prefecture, Japan [TAMBL C1 (paratype: No. 18)]. Description of holotype The holotype was examined when it was 20 days old and reexamined after preserved in formalin- seawater on the 24th day. The umbrella is wider than high, measuring 24.7 mm in diameter and 10.7 mm in height (Fig. 1; Dables 93), 7 the peduncle is not wide even at its base and protrudes from the velar opening for a short distance when it is well-extended (Table 3). Four projections are 2-4 10 mm 1mm Fics. 1-4. The morphology of the holotype (Specimen no. 1) of Eirene lacteoides n. sp. 1: Side view. The peduncle broadens due to slight contraction of body. 2: A marginal portion of the umbrella, showing two tentacular bulbs, each with an adaxial papilla (one in oral view, the other in oblique view), two statocysts and a marginal wart, and a distal portion of the male gonad produced along the radial canal. well-developed tentacular bulb, viewed from aborally (tentacles are not drawn). 3: A papilla on the adaxial side of a 4: A marginal portion of umbrella, showing statocysts and a marginal wart between two successive tentacular bulbs, viewed from abaxially. Note no papillae on this side, and statoliths disappeared due to preservation. TABLE 1. Diameter and the number of marginal swellings of the mature medusa of Eirene lacteoides n. sp. New Eirenid Hydromedusa Total number of Specimen Days of rearing Umbrellar number (=age) diameter in mm tentacular marginal bulbs warts i 20 24.7 105 17 1D* 24 21.6 119 ils 2 12 Heh. 7/ 61 25 3 12 14.0 45 oD 4 367, 19.8 75 28 5 44?) 19.8 65 10 6 43?) 19.8 68 26 ie 19 16.7 90 17 Sy 19 15.4 75 7 g1.3) 19 13,2, 65 16 10” IY) Is 63 18 13 30 305 7/ 141 12 14 35 28.5 157 8 15 35 293 149 7 16 36 30.7 148 14 ia? 38 24.0 153 9 18” 69 IS)2) 55) *: The holotype. Nos. 2-18 are paratypes and for measurements of Nos. 11 and 12, see Table 5) " Measured after preserved in formalin-seawater. > Reared in an artificial seawater for 2-3 weeks at 25+2°C after keeping them in natural seawater from Toba at 27+1°C (the others were reared in natural seawater from Toba at 27+ Mey *) Umbrella is contracted when preserved. * A specimen with five radial canals. TABLE 2. Specimen number Number of statocysts and statoliths of Eirene lacteoides n. sp. Total number of Relative abundance of the number of statoliths per statocyst statocysts statoliths 0 1 7) 3 ie 146 147 0 145 1 0 2 118 125 0 Ath 7 0 3 94 100 0 88 6 0 11 293 Sie 0 275 18 0 12 205 226 2 181 21 1 13 267 285 0 251 14 2 14 254 264 0 244 10 0 tS 263 YS) 0 248 14 1 16 Dill 274 0 241 15 1 7 289 — — — — ane 18 235 — — — — = *: The holotype. —: Unavailable due to disappearance while preservation (see Table 1). 415 416 S. KUBOTA AND T. Horita TABLE 3. Measurements of various body portions of Eirene lacteoides n. sp., in mm taken from living specimens | Specimen Umbrellar Tienes mia as ees number’? height Ae soox peduncle manubrium oral lips stomach gonad Il" 10.7 6.7 Oy) od) De Doi) 0% 2 8.0 4.7 4.0 18 1k) es) 0.2 3 6.0 SG 4.0 0.9 ies real 0.2 4 WES) J) So// 1.6 0.8 eZ 0.3 5 10.3 5.6 2 13 eal 1.4 0.1 6 10.3 Sof) 4.9 3 1.4 1.6 OW 11 21 6.0 Se) Jos) 2.0 == 0.3 1 8.7 3.6 — 2.0 eZ ae 0.3 13 1a, 6.5 7.3 — Hod — 0.2 14 WaT 6.1 6.0 3 1.4 _ 0.2 15 10.1 6.0 4.8 0.8 1 —= 0.1 16 WS_7/ 8.0 6.0 0.8 1.0 — Wait " For ages of medusae, see Tables 1 and 5. *: The holotype. —: Not measured. found interradially at the distal end of the pe- duncle. Even when the peduncle broadens due to the contraction of the body, the projections are still apparent, being conical in shape. However, extreme body contraction may lead to disappear- ance of the projections. The jelly at the umbrellar apex is as thick as the length of the peduncle (Table 3). The manubrium is short, measuring 1.7 mm in length, and provided with four well- developed oral lips which are crenulated and folded many times (cf. Fig. 5). The oral lips are longer than the length of the manubrium; their tips are pointed (Table 3; cf. Fig. 6). The stomach is small, being cruciform in section (cf. Fig. 5). The four gonads are linear, extending from the base of the peduncle close to the umbrellar margin along the four radial canals (Figs.1, 2), but never reaching the ring canal.. In a quadrant 25-27 tentacles are present (27-34 ones, 4 days thereaf- ter) and totally 105 in number (119, 4 days thereaf- ter). The tentacular bulbs are swollen and well- demarcated from the tentacles (Fig. 4). A papilla exists on the adaxial side of every well-developed tentacular bulb (Figs. 2, 3) and also on some small tentacular bulbs bearing very short tentacles. The number of marginal warts in a quadrant varies between 2-7 (2-4, 4 days thereafter), and 17 in all (11, 4 days thereafter) (Table 1). Neither lateral nor marginal cirri are found. In a quadrant 30-40. statocysts are produced, and totally 146 in number (Table 2). The number of statocysts is slightly greater than that of the marginal swellings (= tentacular bulbs+ marginal warts) in every quad- rant since two statocysts are frequently found between two successive marginal swellings. Most of the statocysts contain one statolith, and only one statocyst contains two (Table 2). Variation In a quadrant, up to 44 tentacles, 28 marginal warts, and 84 statocysts were found. The maxi- mum number of tentacles, marginal swellings, and statocysts per specimen was 159, 170, and 304, respectively (Tables 1, 2, 5). Two statocysts are sometimes adjoined together in aged medusae, and in 69-day-old specimen (No. 18) up to five statocysts were found between two succesive mar- ginal swellings. A statocyst contained maximally three statoliths as a very rare case (Table 2). The number of crenulations per lip was more than ten (Fig. 5). An aperture which may function as an excretory pore (excretion of particles was observed New Eirenid Hydromedusa 417 Fics. 5, 6. A paratype (Specimen no. 9) of Eirene lacteoides n. sp. 5: Oral lips, cruciform stomach, and conical projections of the peduncle, viewed from orally. 6: Side view of the manubrium and the projections of the peduncle. at higher magnification), seems to be present on the adaxial papillae. The gonads, radial canals, the projections of the peduncle, and the oral lips were rarely five in number (No. 10). Such abnormal conditions took place in the development of a medusa, though normal in its early developmental stages. Nematocysts Two types of basitrichous isorhizas were found on tentacles and oral lips of the mature medusa (Figs. 7-10). The dimensions of these nematocysts are shown in Table 4. The exumbrellar nemato- cysts, found in early developmental stages of the medusa, disappeared in due time. Development of medusa The growth of medusae was rapid (Table 5, see also Table 1). When the umbrella became over 4.5 mm in diameter, the gonads appeared (on the 6th to 8th day). At this developmental stage the peduncle was also visible. After 11 to 13th day, when the umbrella was over 14.0 mm in diameter, the gonads attained the maximum width and fully matured, the projections of the peduncle were already distinct, and the number of statocysts was 10 pm 2 10 Fics. 7-10. Undischarged and discharged capsules of two types of basitrichous isorhizas in the mature medusa (Specimen no. 5) of Eirene lacteoides n. sp. 418 S. KUBOTA AND T. Horita TaBLE 4. Dimensions of undischarged capsules of nematocysts of mature medusa of Eirene lacteoides n. sp., in um Length X maximum width of two one of basitrichous Body isorhizas: Mean+SD (Range), sample size portions Large type Small type Tentacles Sets OPS3 1 te OD OFS O23 2, dats Oa (14.8-16.8) (3.2-4.0), 20 (9.2-10.4) (2.2-2.6), 20 Oral lips 173058 )<3. 3083 10.2+0.4X2.2+0.1 (15.6-18.8) (2.8-3.8), 24 (9.6-11.2) (2.0-2.2), 20 TABLE5. Development of two medusae of Eirene lacteoides n. sp. Age D H J ite Mw St Stl/St Stl St/Ms_ St/Tb Gl Gw Specimen number 11 1 0.95 ORS5 = ONI3 4 — 8 1 8 — 2D 0 0 3) 1.8 1.4 0.48 6 4 8 I 8 I Z 0 0 6 4.5 2S 1.4 18 9 29 1 29 1 1-2 ODOR 0105 8 7.9 4.1 Ded Di) iW 50 1 50 1-2 1-3 1A ORO 9 17 5.0 Vee) 31 20 62 1-2 63 1-2 1-3 A Ont() 11 14.4 af) 3.3 53 26 88 lan Vil 1-2 1-3 4.5 0.23 13 17.6 1s) 3.6 fs 14 AVY 1-2 123 1-2 1-3 6.30 m025 15 1983 Te, ef) 81 Za 122 1-2 WS) 1-2 1-3 geal 0.20 i7/ Uso) S07 43 97 11 169 1-2 180 1-2 1-3 8.5 0.25 19 Dee 8.7 Jed) 105 11 Ja) JE 1-3 234 1-3: 11-4 9.5 0.30 23 30.6 — — Hits ay) 238 1-2 261 1-2 1-4 12-3 0.25 26 31.4 10.1 5.6 135 14 251 0-2 269 1-2 1-3 Lil Os i027. 3) SB Well 6.0 159 11 293 1-2 SU 1-2 1-4 1323 0.30 Specimen number 12 1 1.0 O87 O85 + 4 8 1 8 1 Z 0 0 3 2.0 1S 0.55 8 1 8 1 8 0-1 1 0 0 6 4.3 Dee) 1.6 17 4 26 1 26 1-2 1-2 0 0 8 Wed) AO 22 Wp 3 +4 1 44 1-2 1-3 1.4 0.08 9 9.3) 4.3 Me) 30 18 56 1 56 -2 1-3 1.4 0.10 13 16.8 6.0 3) 67 y/ 1113) 1-2 115 1-2 1-3 OKO =e WZ8) IS 18.9 EDS ED 79 16 124 1-2 129 1-2 1-3 6.4 0.23 7) 20.8 Joo 4.7 93 3 129 0-2 136 1-2 0-3 8.1 0.25 19 24.8 8.7 3.6 96 4 205 0-3 226 1-3 1-5 9:64 4h0:30 AQ” 21.4 — — 141 7a) 304 a= — 0-3 0-8 8.7 0.23 —: Not measured. '. Measured after preserved in formalin-seawater. Age: days of rearing. D: umbrellar diameter, in mm; H: umbrellar height, in mm; J: thickness of jelly at the umbrellar apex, in mm; Te: total number of tentacles; Mw: total number of marginal warts; St: total number of statocysts; Stl: total number of statoliths; Stl/St: range of number of statoliths per statocyst; St/Ms: range of number of statocysts between two neighboring marginal swellings; St/Tb: range of number of statocysts between two neighboring tentacular bulbs; Gl: length of gonads, in mm, measured from aboral side; Gw: maximum width of gonads, in mm. New Eirenid Hydromedusa 419 Ej medusa with papilla on most tentacular bulbs medusa with papilla on some tentacular bulbs [] medusa without papilla on all tentacular bulbs No. of specimens 2.0 | 3.0 3.5 4. 4.5 Umbrellar diameter, in mm Fic. 11. Frequency distribution of the number of immature medusa in Ejirene lacteoides n. sp. according to the presence (plentiful or scarce) or absence of adaxial papilla on tentacular bulbs. TABLE 6. Distinguishing characters of mature medusa between Eirene lacteoides n. sp. and the closest congener Eirene lactea Species Eirene lacteoides Eirene lactea Mayer, 1900 Tortugas, Florida Toba, Mie Pref., sek ; SE USA? Locality Japan* (present study) en oie (after Brinckmann-Voss 1973) Distinguishing characters Peduncle distal projections distal projections distal projections present; extends absent; extends present; not extend ing to for a short for a short the velar opening distance beyond distance beyond (occupies 2/3 of the height the velar the velar of subumbrellar opening opening cavity) Oral lips fairly crenulated simple slightly crenulated Diameter of 24.7 mm 5 mm”? 6.0-20.0 mm umbrella (6.0-33.2 mm)” Number of 105 (45-159)? 18222) 24-68 tentacles Number of 146 (94-304)? below 44” below 91”? statocysts Adaxial present on well-developed absent”? absent papilla tentacular bulbs *: Mature medusae (male in E. lacteoides and female in E. lactea) obtained by culture in the laboratory. The origin of the hydroid colony is not known. **: Mature medusae collected from the natural sea. Sex of these medusae was not described. . Undescribed precisely, therefore caliculated based on description. >) After Brinckmann-Voss 1973, pp. 65-66. ») Following measurements of the holotype, those of paratypes are shown in parentheses. ” The holotype is not designated. 420 S. KuBoTA AND T. Horita always more than that of the marginal swellings (= tentacular bulbs+ marginal warts). The adaxial papillae were produced in immature and very small medusae, minimally 2.4mm in diameter; young specimens, over 3.0mm in diameter, had always these papillae on almost all tentacular bulbs (Fig. 11). Continuous enlargement of diameter of the umbrella lasted for about a month. In all developmental stages the umbrella always re- mained wider than high (Table 5); no cirri were produced. Remarks In the genus Eirene and other known genera of the family Eirenidae [12-17] such a conical projec- tion of the peduncle as observed on Eirene lac- teoides n. sp. has never been described in literature except for an illustration made by Brinck- mann-Voss [16] of the laboratory-reared female medusae of Eirene lactea (Mayer, 1900). How- ever, Brinckmann-Voss (per. comm.) did not think the projections might provide a disting- uishing character for a new species. The original material of E. lactea is small, 5 mm in umbrellar diameter (see Table 6), and it had no projections on peduncle. Accordingly it seems that the projec- tions are not produced in E. /actea if the medusa of this species matures in a small size. On the other hand, as was described above, the adaxial papillae were already produced at im- mature stages (the diameter of the umbrella over 2.4mm) in E. lacteoides, whereas the papillae were not found in E. lactea even though the species was smaller in size or developed fully to the umbrellar size of 20.0 mm [16]. Further, FE. lac- teoides tends to have more statocysts, more tenta- cles, and more crenulated oral lips than those of E. lactea (Tables 5, 6). When the peduncle of the present new species is contracted, it may become wider at its base and looks like that of Eirene pyramidalis (L. Agassiz, 1862) that also has an excretory pore [12, 13]. However, E. pyramidalis does not possess any distal projection on the peduncle despite attaining a similar size or much more, up to 35mm in diameter [12, 13]. Further, FE. pyramidalis has maximally 100 statocysts and the same number of tentacles [17], showing thus much smaller numbers in these two meristic characters than those of E. lacteoides (see Table 6). Consequently, by unique characteristic states of the projections of peduncle, the number of stato- cysts as well as the tentacles, and of the adaxial papillae of tentacular bulbs of medusa, we treat the present material as a new species. Although all specimens of E. lacteoides are males, there is seemingly no taxonomic difficulty since the sexual dimorphism is very rare in hydromedusae; so far only in Sphaerocoryne multitentaculata (Warren, 1908) [19] and Australomedusa baylii Russell, 1970 [1, 20, 21]. It should be mentioned here that the kind of nematocysts of the new species is identified as basitrichous isorhizas as described above, but for such a small nematocyst Ostman [22] proposed a new category, i.e., pseudo-microbasic b- mastigophore, through her SEM investigations. Distribution Besides the specimens taken in the Toba Aquarium, several laboratory-reared male me- dusae of E. lacteoides were also obtained by Mr. S. Takayama in the Uozu Aquarium, Toyama Prefec- ture, in 1991. Further, one well-developed, laboratory-reared medusa photographed by Dr. Y. Hirano, which was originally found in the Oarai Aquarium, Ibaraki Prefecture in 1983, could also belong to the present species, but this needs fur- ther examination. No specimens referable to the present new species have been collected from the natural sea. ACKNOWLEDGMENTS We wish to express our gratitude to Drs. Jean Bouil- lon, Anita Brinckmann-Voss, Mayumi Yamada, and anonymous reviewers of the journal for their critical reading of the manuscript. We also thank to the staff of the Toba Aquarium for the use of facilities to rear the medusae. Sincere thanks are extended to Dr. Yayoi Hirano and Mr. Shigeki Takayama for their kindness to inform us of the occurrence of the present species at two different sites from the type locality. Drs. Jean Bouillon, Dale R. Calder, Pual F. S. Cornelius, Eiji Harada, Minoru Imajima, and Masatune Takeda kindly gave us a chance to deposit the paratypes. This study was partly supported by Grant-in-Aid for Scientific Reserach No. 01740442 from the Ministry of Education, Science and Culture of Japan to S. Kubota. 10 11 New Eirenid Hydromedusa REFERENCES Bouillon, J. (1985) Essai de classification des hydro- polypes-hydroméduses (Hydrozoa-Cnidaria). Indo- Malayan Zool., 2: 29-243. Uchida, T. (1925) Some hydromedusae from north- er Japan. Jap. J. Zool., 1: 77-100. Uchida, T. (1933) Medusae from the vicinity of Kamchatka. J. Fac. Sci. Hokkaido Univ. Ser. VI. Zoolx, 2(3)2 125-133. Uchida, T. (1938) Medusae in the vicinity of the Amakusa Marine Biological Station. Bull. biogeogr. Soc. Japn., 8: 143-149. Uchida, T. (1938) Medusae in Onagawa Bay and its vicinity. Sci. Rep. Tohoku Imp. Univ., 4th Ser., Biol., 13(1): 47-58. Uchida, T. (1940) The fauna of Akkeshi Bay XI. Medusae. J. Fac. Sci. Hokkaido Univ. Ser. VI. AGI): 277-297. Yamazi, I. (1958) Preliminary check-list of plankton organisms found in Tanabe Bay and its environs. Publ. Seto Mar. Biol. Lab., 7(1): 111-163. Sugiura, Y. (1979) Ona hydromedusa Eirene meno- ni Kramp from Amakusa, Japan. Proc. Jap. Soc. Syst. Zool., 16: 5-8. Kubota, S. (1983) Studies on life history and sys- tematics of the Japanese commensal hydroids living in bivalves, with some reference to their evolution. J. Fac. Sci. Hokkaido Univ. Ser. VI, Zool., 23(3): 296-402, pl. X. Kubota, S. (1985) Systematic study on a bivalve- inhabiting hydroid Eugymnanthea inquilina japonica Kubota from central Japan. J. Fac. Sci. Hokkaido Univ. Ser. VI, Zool., 24(1): 70-85. Kubota, S. (1991) The stability of diagnostic charac- ters of the medusa of a bivalve-inhabiting hydrozoan Eugymnanthea japonica Kubota Proc. Japan. Soc. Syst. Zool., 44: 1-7. 12 13) 14 15 16 17 18 19 20 Zk Vij 421 Kramp, P. L. (1959) The hydromedusae of the Atlantic Ocean and adjacent waters. Dana Rep., 46: 1-283, 2 pls. Kramp, P. L. (1961) Synopsis of the medusae of the world. J. mar. biol. Ass. U. K., 40: 1-469. Kramp, P. L. (1968) The hydromedusae of the Pacific and Indian Oceans. Sections II and III. Dana Rep., 72: 1-200. Bouillon, J. (1984) Hydroméduses de la Mer de Bismarck (Papouasie Nouvelle Guinée). Partie IV: Leptomedusae (Hydroza-Cnidaria). Indo-Malayan LOO N25 — le Brinckmann-Voss, A. (1973) The life-cycle of Eirene lactea (Mayer, 1900) and Helgicirrha schulzei Hartlaub, 1909 (Phylum Cnidaria, Class Hydrozoa, Order Leptomedusae, Family Eirenidae). Publ. Seto Mar. Biol. Lab., 20 (Proc. 2nd Int. Symp. Cnidaria): 63-72. Mayer, A. G. (1900) Some medusae from the Tortugas, Florida, Bull. Mus. Comp. Zool. Har- vard., 37(2): 13-82, pls. 1-44. Mayer, A. G. (1910) Medusae of the world. Vol. II. Hydromedusae, pp. 231-498, pls. 30-55. Yamada, M. and K. Konno (1973) Polyp and medusa of the hydroid Sphaerocoryne multitentacu- lata (Warren) from Japan. Publ. Seto Mar. Biol. Lab. 20 (Proc. 2nd Int. Symp. Cnidaria): 103-109. Russell, F. S. (1970) On a new species of medusa from an inland salt lake in south Australia. J. Zool. Lond. 162: 449-4572. Russell, F. S. (1971) On the female of the medusa Australomedusa baylii. J. Zool. Lond. 164: 133-135. Ostman, C. (1988) Nematocysts as taxonomic criteria within the family Campanulariidae, Hydro- zoa. In “The Biology of Nematocysts”. Ed. by D. A. Hessinger and H. M. Lenhoff. Academic Press Inc. San Diego, California, USA. pp. 501-517. er Es plowed ig ae she ie ait, eyes Ae coo nchckinhe * ? cf sh i whenipdesHh bag (ODO! ave} Ras are ee Mien byH’ ab we Ao ieltre 2 traded) es Basler ah idattth goubiniasial ‘nglieigbi Saath orem atlistie Ne, nals fips paueats * a bath ly le , is 2 fe aeelae aOR ae agabiscys, 28 aout She 2 sit \omiog ie ie ae AS a ae, wie DY) AST Ve trae Atk oe 2eentior - 2 ey MOE. | ay Ht etal eH HARE PE, Baginre Lebo bianw bift tel ober de pada Bore eed anit Fait iets amin: nby és ; - 4 A <— a a mice ss ae hide 9 BRh bib, el HR, “a: AM nvstiahseve BIO a ald Orae Shih y ERE fiakb iD eeakent sta ohies PD jankes epiteitens ea ins imei giasitnl fica ii ne (cet teat ts 5 J as on f Were ‘ ae 4 ie ~. - . , ‘ a Pe it d ‘ raat {-\ 4 Lely TAY tak. ay] ters | * . b wy re : ee | : we Jim beviy wires ’ ‘? AE: : ai a Nees? ' oT od Pei th Fai? What d A 30, r ed rvaltire Mili < - nite WP 4 = te S ie ct ont | ol =) j ¥ f 5 a7 , ge > Bes = Sy is-2 Le By ba ber it ON ee a ; 5 % f vay , : . ff a : Li — r re < y 4 4 { 4 Ne 5 4 —) 4 > r / 3 1 oh v c x. } e ea er a ee é a, ‘- mks ae apres Soe Eaves - Sth x i is : fu PRONG % Bash Sig EYES | JOGR AV ite : t yr ’ 4 “ r ‘ ‘ er Lot I ‘ bh { kv 44s - ¥ * rie : dee enh 3 Lh Set i ey 2s fi 4 x t nl ie , ¢ , ZOOLOGICAL SCIENCE 9: 423-437 (1992) Two New Species of the Genus Dicyema (Mesozoa) from Octopuses of Japan with Notes on D. misakiense and D. acuticephalum HIDETAKA FURUYA, KAZUHIKO TSUNEKI and YUTAKA KOSHIDA Department of Biology, College of General Education, Osaka University, Toyonaka 560, Japan ABSTRACT—We examined dicyemid mesozoans from the renal sacs of both Octopus vulgaris and Octopus minor, obtained off the coast of Japan, and found two new species that belong to the genus Dicyema. Dicyema japonicum sp. nov. from O. vulgaris, is a medium sized dicyemid, rarely exceeding 1500 um in length. The number of peripheral cells in the vermiform phases is usually 22. The disc-shaped calotte and parapolar cells form the cephalic enlargement. The axial cell is cylindrical but is rounded anteriorly, and it extends forward to the base of propolar cells. Infusoriform embryos consist of 37 cells. In each of the four urn cells, there are the cell’s own nucleus and one germinal cell with its own nucleus. Dicyema clavatum sp. nov. is a relatively small sized dicyemid, infrequently reaching 1000 um in length, and it is the first mesozoan species described from O. minor. The number of peripheral cells in the vermiform phases is usually 22. The calotte is cap-shaped and smoothly rounded. The axial cell is enlarged and rounded in the calotte region, and it extends anteriorly to the base of the propolar cells. Uropolar cells occasionally become verruciform. Infusoriform embryos are composed of 39 cells. Each of the urn cells contains two nuclei of its own and one germinal cell with its own uncleus. Further details relevant to the description of infusoriform embryos of Dicyema misakiense Nouvel et Nakao are provided and a note to Dicyema acuticephalum Nouvel is given. The dicyemid fauna in the © 1992 Zoological Society of Japan two species of octopuses is briefly discussed. INTRODUCTION The first record of dicyemid mesozoans in Japan was published in 1938 by Nouvel and Nakao [1]. They found Dicyema misakiense Nouvel et Nakao, 1938, in the renal sac of Octopus vulgaris, and also Dicyema orientale Nouvel et Nakao, 1938, in Sepioteuthis lessoniana. Later, Nouvel [2] de- scribed Dicyema acuticephalum Nouvel, 1947, which was also obtained from Octopus vulgaris, and identified another dicyemid from Sepia esculenta as Pseudicyema truncatum Whitman, 1883, which had been already described in Europe. All these host cephalopods were collected in the waters close to the Misaki Marine Biological Sta- tion of the University of Tokyo. Since then no reports on the dicyemid species from Japan have been published. We examined the dicyemids in the renal sacs of Accepted January 6, 1992 Received November 9, 1991 octopuses caught off the coast of Japan, and we found at least two new dicyemids that are distinctly different from each of the four species mentioned above and from the other species so far described in various regions outside Japan. The present paper deals with these two new dicyemid species: one obtained from Octopus vul- garis and the other from Octopus minor. In addition, we give a detailed description of the infusoriform embryos of Dicyema_ misakiense Nouvel et Nakao and further give an account of Dicyema acuticephalum Nouvel. MATERIALS AND METHODS From April of 1989 to May of 1991, 23 indi- viduals of Octopus vulgaris and 19 of Octopus minor were obtained for detection of dicyemids. Most of them were obtained from fish markets and fishermen, but two were caught by the authors. The size, sex, and source of each of these octo- puses are given in Tables 1 and 2. 424 H. FuruyA, K. TSUNEKI AND Y. KOSHIDA From every octopus, which was brought alive to the laboratory, the head and tentacles were cut off just behind the eyes, without anesthetic. Then the visceral hump was placed, ventral side up, in a tray, and the mantle was opened to expose the TABLE 1. Octopus vulgaris paired renal sacs. Pieces of renal tissues were smeared with the fluid from the renal sac on slide glasses. Some preparations were used to confirm the occurrence of living dicyemids under the phase-contrast microscope, while others were and parasite dicyemids Hosts Dorsal mantle Body xe Date of Dicyemids No. length (cm) weight (g) SOs source examination D. acuticephalum VU12 WZ 350 M 1 06.21.1989 D. japonicum D. misakiense D. japonicum VU98 9.5 865 Ia i 07.14.1990 D. misakiense D. sp. (A) VU5 7.4 270 F 1 06.14.1989 VU16 5.4 78 M 2 07.06.1985 VU17 1 123 M 2 07.06.1989 VU18 6.5 - 148 M 2 07.06.1989 VU20 ES — Is 3) 07.24.1989 VU28 Dod — M 3) 08.04.1989 VU32 {3.2 — M 4 08.10.1989 D. japonicum VvU41 W225) 740 F 2 10.20.1989 D. misakiense VU43 8.2 528 M 6 10.27.1989 VU44 Qe) 738 M 6 10.27.1989 VU45 6.9 460 M 1 12.01.1989 VU83 1325 465 M 1 05.11.1990 VU101 Dd 96 M 2) O21 LOD VU102 6.6 218 F Z OZ NOS VU103 V9 — F 5 07.30.1990 VU104 Tete) 508 M 2 05.08.1991 VU3 8.2 260 M 1 04.26.1989 VU4 11.4 770 F 1 05.30.1989 D. acuticephalum VU40 OF 750 F 2 09.06.1989 VU105 OF 550 M 1 5-27-1991 D. misakiense VUI5 4.0 70 F 2 07.06.1989 None * Source numbers indicate the following: (1) Commercially supplied from a fish market at Akashi (Hyogo Pref.) where fish dealers receive octopuses caught mainly inside Osaka Bay and/or in the Sea of Harima (the eastern area of the Inland Sea). (2) Commercially supplied from a fish market at Shounai (Toyonaka, Osaka Pref.). Locations of these octopuses are presumably similar to those of octopuses bought at Akashi. (3) Commercially supplied from a fish market at Shirahama (Wakayama Pref.). The octopus was probably caught inside Tanabe Bay where it faces the Pacific Ocean. (4) Commercially supplied from a fish market at Sakaiminato (Tottori Pref.). The octopus was possibly obtained from the coast of the Sea of Japan near Sakaiminato. (5) Caught by fishermen in Dozen, Oki Islands (Shimane Pref.), located in the Sea of Japan about 50 km to the north of Sakaiminato. (6) Collected by the authors in Maizuru Bay (Kyoto Pref.) where it is open to the Sea of Japan. Dicyemid Mesozoans from Japan 425 TABLE 2. Octopus minor and parasite dicyemids Hosts Dorsal mantle Body x Date of Dicyemids IN@e length (cm) weight (g) Bex Source examination MI89 6.5 78 M 2 07.11.1990 MI90 7.9 168 M D, 07.14.1990 D. clavatum MI91 8.3 204 M 2 07.14.1990 D. sp. (B) MI92 8.8 239 M 2 07.14.1990 MI97 DZ 68 F 2 07.14.1990 MI93 8.2 126 F 2, 07.14.1990 D. clavatum MI94 8.3 207 le 2 07.14.1990 MI95 8.3 194 F 2 07.14.1990 MI96 WES) 145 F D 07.14.1990 MI63 Wot 105 F 1 04.10.1990 MI64 On 175 M 1 04.10.1990 MI65 Vo? 142 F 1 04.10.1990 D. sp. (B) MI69 W3 110 F 1 04.10.1990 MI71 6.5 231 M 1 04.10.1990 MI84 6.4 124 M 2 07.11.1990 MI85 7.4 155 M Z, 07.11.1990 MI86 6.5 192 M Z 07.11.1990 D. sp. (C) MI87 6.9 111 M yy 07.11.1990 MI88 Oy 118 M 2, 07.11.1990 * See explanations for Table 1. promptly fixed in Carnoy’s fluid or in alcoholic Bouin’s fluid (a 15:5: 1 mixture of saturated picric acid in absolute ethanol, formalin, and acetic acid). Carnoy-fixed preparations were subjected to Feulgen or McMannus’s periodic-acid Schiff (PAS) procedures, and then they were stained with Ehrlich’s acid hematoxylin and light green. Alcoholic Bouin-fixed preparations were subjected to the last two staining procedures only. After the staining, the preparations were dehydrated and mounted in the usual fashion for observation of dicyemids under the light microscope. Measurements and drawings were made with the aid of a micrometer and an image tracer, respec- tively. Dicyemidae van Beneden, 1882 Dicyema von Kolliker, 1849 Dicyema japonicum sp. nov. Furuya et Tsuneki [New Japanese name: Yamato-nihaityu] (Figs. 1-7, Tables 1, 3, and 4) Host: Octopus vulgaris Lamarck, Octopodidae. Locality: Western Honshu, Japan. See “Source” in Table 1. Syntypes: A slide registered as NSMT-Me-1 was deposited at the National Science Museum, Tokyo, Japan. This slide was prepared from No. VU44 octopus (Table 1) and contains both nematogens and rhombogens. It includes D. misakiense as well, but D. japonicum can be clearly distinguished from D. misakiense in the shape of the head as described below. Other slides were numbered according to the host number in Table 1 and are in the authors’ collection. Etymology: The specific name “japonicum” was given, because this species is common in Octo- pus vulgaris caught off the coast of Japan. DESCRIPTION Diagnosis: Body length up to 1800 um. Peripheral cell number of vermiform phases (ver- 426 H. FuruyA, K. TSUNEKI AND Y. KOSHIDA miform embryo, nematogen, and rhombogen) usually 22:4 propolars, 4 metapolars, 14 trunk cells. Propolar and metapolar cells form a disc-like head together with parapolar cells. Infusoriform embryos consist of 37 cells. Nucleus number of urn cell, 1. Host, Octopus vulgaris. Nematogens (Figs. 1-3): Body slender, 300 to 1800 um long; 40-75 um wide. Peripheral cell number usually 22 (Table3): 4 propolars, 4 metapolars, 2 parapolars, 10 diapolars, 2 uropo- lars. Calotte and parapolar cells form cephalic enlargement. Calotte becomes disc-shaped as in- dividuals grow. Cilia covering calotte about 4.7 ym long, oriented forward, slightly shorter but denser than cilia of trunk cells. Cytoplasm of both propolar and metapolar cells, stained by hema- toxylin more conspicuously than that of other cells; Fic. 1. Dicyema japonicum sp. nov. Four entire nema- togens of various sizes are shown on the left and three rhombogens on the right. Note the character- istic shape of the head. Bar represents 100 um. Drawn from specimens prepared from No. VU44 octopus. PAS-positive, but negative after saliva test. Cells and nuclei of propolars, smaller, respectively, than those of metapolars. Trunk mostly uniform in width. Trunk cells, arranged in opposed pairs. No verruciform cells. Axial cell, cylindrical and rounded anteriorly, extends forward to base of propolar cells. Usually two sizes of axoblasts, standard and large; the latter often twice the size of the former. In large individuals, about 80 vermiform embryos at most found in the axial cell, calotte occasionally flower-like in shape, and a few accessory nuclei evident in both peripheral and axial cell. In Carnoy-fixed large individuals, many fine granules found in peripheral cells. Transitional individuals from nematogens to trhombogens enclose degenerating vermiform embryos, proliferating infusorigens, and develop- ing or full-grown infusoriform embryos, simul- taneously, in the axial cell. Standard axoblasts only, no large ones in these individuals. Vermiform embryos (Fig. 4): Full-grown vermi- form embryos, 40 to 70 um long, 8 to 12 um wide; peripheral cell number fixed at constantly 22 (Table 3). The ratio of total body length to calotte length, 1:0.24 to 0.28. Anterior end of calotte tapering slightly and pointed bluntly. Trunk cells, arranged in opposed pairs. Axial cell tapering anteriorly, sometimes pointed, extending forward to base of propolar cells, as seen in nematogens. Axial cell nucleus, usually located in center or in anterior half of axial cell, always anterior to one or two standard-sized axoblasts. Stem nematogens (Fig. 5): Two stem nema- togens were obtained. One of them, 221 um long, with three axial cells. Peripheral cell number: 23 (4 propolars, 4 metapolars, 2 parapolars, 11 diapo- lars, 2 uropolars). Wermiform embryos found in both second and third axial cell but not in the first. The other stem nematogen, 223 um long, also with three axial cells. Peripheral cell number: 26 (4 propolars, 4 metapolars, 3 parapolars, 14 diapo- lars, 1 uropolar). Vermiform embryos found in all three axial cells. Only standard-sized axoblasts in both stem nematogens. Rhombogens (Figs. 1-3): Slightly shorter and stockier than nematogens, otherwise generally similar in shape and proportion; body 300 to 1000 yum long; 40 to 75 wm wide. Peripheral cell num- Dicyemid Mesozoans from Japan 427 Fic. 2. Anterior part of a nematogen (a) and a rhombogen (b) of D. japonicum. Photographs were taken after alcoholic Bouin fixation, which was followed by Ehrlich’s acid hematoxylin and light-green staining. Note that the calotte (C) is conspicuously stained and is covered with a dense array of cilia. AX, axoblast; AC, axial cell; D, diapolar cell; IF, infusoriform embryo; PA, parapolar cell; V, vermiform embryo. Bars beueecut 20 wm. (a) Prepared from No. VU103 octopus, (b) prepared from No. VU44 octopus. ber, usually 22, sometimes lower (Table 3). Cephalic enlargement, composed of calotte and parapolar cells as in nematogens. Calotte becomes disc-shaped as individuals grow. Shape and tip position of axial cell, similar to those in nema- togens. Axial cell sometimes expanded at the region where infusorigens are included. One, sometimes two, and very rarely three infusorigens in the axial cell. Some accessory nuclei observed occasionally in both peripheral and axial cell. No verruciform cells. Infusorigens (Fig. 6): Medium-sized, sometimes relatively large. Axial cell usually irregular in shape. In 25 mature infusorigens examined: num- ber of external cells including oocytes, 8 to 58 (mode, 15); number of internal cells including spermatocytes, 3 to 19 (mode, 3). Fertilized eggs, 12.3 um in diameter. Infusoriform embryos (Fig. 7): Ovoid, rounded bluntly to pointed posteriorly. Based on 100 full-grown embryos, length (excluding cilia), 23.68 +1.98 um (mean+S.D.); length-width-height ratio, 1 :0.80:0.73. Cilia at the posterior end, 6 to 7 um long. Refringent bodies, smaller than total mass of four urn cells, occupy anterior one-third or so of embryo, when viewed from lateral side. Nuclei of second ventral cells, small and pycnotic. Ventral internal cells project cilia to urn cavity. Capsule cells with many large granules on side adjacent to urn. Granules, intensely stained by PAS procedure and staining-resistant in saliva test. Full-grown infusoriform embryos consist of 37 428 H. FuruyA, K. TSUNEKI AND Y. KOSHIDA TABLE 3. Number of peripheral cells of Dicyema Japonicum sp. nov. Number of individuals No. of cells a Nematogens Rhombogens 20 0 2 21 0 2 8 22 54 52 54 Q 1a SioRomts Fic. 3. Anterior part of three nematogens (top) and two rhombogens (bottom) of D. japonicum. Cilia are shown in optical section on the nematogen depicted on the far right. Bar represents 50 um. Drawn from specimens prepared from No. VU103 octopus. Fic. 5. A stem nematogen of D. japonicum with three axial cells. This individual has vermiform embryos in both the second and the third axial cell; in the first axial cell its nucleus is visible but no embryos are seen. Photograph (a) was taken after alcoholic Bouin fixation, Ehrlich’s acid hematoxylin and light- green staining. The tracing (b) was drawn from the photograph. AC1, first axial cell; AC2, second axial cell; AC3, third axial cell. Bar represents 20 um. Prepared from No. VU103 octopus. Fic. 4. A vermiform embryo within the axial cell (the cell outline is omitted) of a nematogen (on the left) and a free-living vermiform larva (on the right) of D. japonicum. Bars represent 104m. Drawn from specimens prepared from no. VU103 octopus. Dicyemid Mesozoans from Japan 429 cells: 33 somatic and 4 germinal cells (Table 4). Cell terminology used here is that of Nouvel [3] and Short and Damian [4]. Somatic cells are composed of peripheral cells that cover a large part of anterior and lateral surfaces of embryo (2 enveloping ceils), peripheral cells with cilia on external surfaces (2 paired dorsal cells, 1 median dorsal cell, 2 dorsal caudal cells, 2 lateral caudal cells, 1 ventral caudal cell, 2 lateral cells, 2 post- eroventral lateral cells), peripheral cells with re- fringent bodies (2 apical cells - cilia not clearly Fic. 6. Infusorigen of D. japonicum. Around the in- fusorigen, a two-pronuclei stage egg (E), a fertilized egg (F), and an axial cell nucleus (N) are found. Bar represents 10 ~m. Drawn from a specimen prepared from No. VU103 octopus. Fic. 7. Infusoriform embryo of .D. japonicum. Dorsal view (a), ventral view (b), sagittal section (c), and horizontal section (d, cilia oimtted) are shown. Enveloping cells are ommitted. A, apical cell; AL, anterior lateral cell; C, couvercle cell; CA, capsule cell; DC, dorsal caudal cell; G, germinal cell; L, lateral cell; LC, lateral caudal cell; MD, median dorsal cell; PD, paired dorsal cell; PVL, posteroventral lateral cell; R, refringent body; U, urn cell; UC, urn cavity; VC, ventral caudal cell; VI, ventral internal cell, V1, first ventral cell; V2, second ventral cell. Bar represents 10 7m. Drawn from specimens prepared from No. VU103 octopus. 430 H. Furuya, K. TSUNEKI AND Y. KOSHIDA TABLE 4. Number of cells in infusoriform embryos and the nucleus number of urn cells in four dicyemid species : No. of No. of nucleus Species cells* of urn cells* D. japonicum of 1 D. misakiense 3 it D. acuticephalum 37 Z D. clavatum 39 D, * These numbers were very constant in four species examined. - visible on these cells), peripheral cells without cilia (2 first ventral cells, 2 second ventral cells, 2 anterior lateral cells, 1 couvercle cell), internal cells with cilia (2 ventral internal cells), internal cells without cilia (2 dorsal internal cells, 2 capsule cells, 4 urn cells). Each of the four urn cells encloses its own nucleus and one germinal cell with its one nucleus (Table 4). All somatic nuclei tend to become pycnotic as infusoriform embryos ma- ture. Geographical variations: Not found either in vermiform stages or infusoriforms. bee Dicyema misakiense Nouvel et Nakao, 1938 [Japanese name: Misaki-nihaityu] (Fig. 8a, Tables 4 and 5) Materials examined: Slides were numbered according to the host number in Table 1 and afe in the authors’ collection. Diagnosis: Body length up to 1700 um. Peripheral cell number 22. Calotte consists of two tiers; the tier of propolars slightly smaller than that of metapolars and faintly constricted from the tier of metapolars. Infusoriform embryos, 37 cells. Nucleus number of urn cell, 1. Host, Octopus vulgaris. TABLE 5. Number of peripheral cells of Dicyema misakiense Number of individuals No. of Vermif galls embryos Nematogens Rhombogens 21 0 1 (0). 22 51 51 a) Fic. 8. Anterior part of two rhombogens of D. misakiense (a) and a nematogen of D. acuticephalum (b). Compare the head shape of these species with that of D. japonicum (shown in Fig. 2). Photographs were taken after alcoholic Bouin fixation, Ehrlich’s acid hematoxylin and light-green staining. AC, axial cell; AX, axoblast; D, diapolar cell; IF, infusoriform embryo; M, metapolar cell; N, axial cell nucleus; P, propolar cell; PA, parapolar cell. Bar represents 20 um. (a) Prepared from No. VU44 octopus, (b) prepared from No. VU40 octopus. Dicyemid Mesozoans from Japan 431 Note: Nouvel and Nakao [1] reported that the number of peripheral cells in vermiform phases was usually 22, but was sometimes lower. How- ever, we found this number to be almost constant; 157 out of 158 vermiform individuals examined had 22 peripheral cells, but only one nematogen had 21 cells (Table 5). The original description of the infusoriform embryos is brief: the embryos (excluding cilia) are 25 um long, 20 um wide, and 18 ~m high; and each of the urn cells has its own nucleus and one germinal cell [1]. We confirmed in our materials that the nucleus number of urn cells is one (Table 4). Additional details of the infusoriform embryos are given, based on our observations. Infusoriform embryos: Ovoid and rounded bluntly to pointed posteriorly. Based on 100 full-grown embryos, length (excluding cilia), 24.67 +1.28 um (mean+S.D.); length-width-height ratio, 1:0.86:0.81. Cilia at the posterior end, 6 to 7 wm long. Refringent bodies, smaller than total mass of four urn cells, occupy about 40% of anterior part of embryo, when viewed from lateral side. Nuclei of second ventral cells, small and pycnotic. Ventral internal cells with cilia project- ing into urn cavity. Capsule cells contain many large granules, which are located on side adjacent to urn. Full-grown infusoriform embryos com- posed of 37 cells (33 somatic and 4 germinal). Somatic cell composition same as in D. japonicum (Table 4). Somatic nuclei tend to show pycnosis as embryos mature. Dicyema acuticephalum Nouvel, 1947 [New Japanese name: Togari-nihaityu] (Fig. 8b, Tables 4 and 6) Materials examined: Slides were numbered according to the host number in Table 1 and are in the authors’ collection. Diagnosis: Body length up to 800 um. Peripher- al cell number constantly 18. Calotte bell-like in shape. Infusoriform embryos consisting of 37 cells. Nucleus number of urn cell, 2. Host, Octopus vulgaris. Note: Nouvel [2] reported that the number of peripheral cells in this species ranged from 16 to 19 (generally 18), and the nucleus number of urn cells was two. We confirmed his findings in our mate- rials, but the number of peripheral cells in the vermiform embryos was constant and equal to 18 and no variation was detected (Table 6). We also revealed that fully mature infusoriform embryos consist of 37 cells (Table 4). TABLE 6. Number of peripheral cells of Dicyema acuticephalum Number of individuals No. of oeus yeaa Nematogens Rhombogens 16 0 0 2 17 0 i 0 18 62 56 58 19 0 1 2 REMARKS The distinction between Dicyema japonicum and Dicyema acuticephalum is clear because of differences in typical number of peripheral cells in full-grown vermiform phases (22 vs. 18) as shown in Tables 3 and 6, in calotte shape (disc type vs. conical type), in position of the axial cell tip (extending to the base of propolar cells vs. metapolar cells), and in the number of somatic nuclei in the urn cells of infusoriform embryos (one vs. two) as shown in Table 4. _ Both Dicyema japonicum and Dicyema mis- akiense are similar in the average length of the body and have, typically, 22 peripheral cells (Tables 3 and 5). In both species, the axial cell extends to the base of propolar cells in vermiform phases, and infusoriform embryos consist of 37 cells, which include urn cells that contain their own single nuclei (Table 4). Nevertheless, we can point out the following differences between D. japoni- (1) The calotte of D. japonicum is a disc-shaped and forms the cephalic enlargement with parapolar cells (Figs. 2 and 3), whereas D. misakiense has a slender calotte (Fig. 8a). No individuals with a calotte that was in- termediate in shape between those of the two species were found. (2) Vermiform embryos of D. japonicum often have two axoblasts within the axial cell, while those of D. misakiense consistently cum and D. misakiense. 432 H. FuruyA, K. TSUNEKI AND Y. KOSHIDA have only one axoblast [1]. (3) The length-width- height ratio of infusoriform embryos of D. japoni- cum is different from that of D. misakiense (1: 0.80:0.73 vs. 1:0.86:0.81). These differences should be sufficient for establishing species in the dicyemids and therefore we have identified D. Japonicum as a new species. In the paper by Nouvel and Nakao [1], the text in lines 33-36 on p. 74, in lines 1-5 on p. 75, and Figures 6 and 7 on p. 76 could be regarded as referring to D. japonicum and not to D. misakiense. Dicyema aegira McConnaughey and Kritzler, 1952 [5], from Octopus vulgaris, has 22 peripheral cells as equal as D. japonicum, but the two species are easily distinguishable from the following differ- ences. D. aegira has a calotte that is slightly longer than the breadth of the anterior end of the trunk, and the axial cell ends at the base of the metapolar cells, unlike the arrangement in D. japonicum. Dicyema orientale Nouvel et Nakao, 1938 [1] has a calotte that is very similar in shape to that of D. Japonicum, and it also has 22 peripheral cells. However, at full-grown vermiform stages, D. orientale is much larger than D. japonicum; the number of axoblasts within vermiform embryos in D. orientale is also larger than in D. japonicum (8 vs. 1 or 2). The host cephalopod of D. orientale is Sepioteuthis lessoniana, a decapod, not an octo- pod. It seems reasonable, therefore, that these two dicyemids should be considered members of different species. Dicyema benthoctopi Hochberg et Short, 1970 [6] is very similar to D. japonicum in the calotte shape and in the size of vermiform stages, but the number of peripheral cells in D. benthoctopi is very variable. The host of D. benthoctopi is Benthoctopus magellanicus, a deep-sea octopod distributed in the Atlantic Ocean near the Falk- land Island, at about 500 km east of the Strait of Magellan, the southern extremity of the South American Continent. It is hard to conclude, therefore, that D. benthoctopi and D. japonicum are the same species, even though infusoriforms of D. benthoctopi have not yet been described. Dicyemidae van Beneden, 1882 Dicyema von Kolliker, 1849 Dicyema clavatum sp. nov. Furuya et Koshida [New Japanese name: Konbou-nihaityu] (Figs. 9-14, Tables 2, 4. and 7) Host: Octopus minor (Sasaki), Octopodidae. Locality: Western Honshu, Japan. See “Soruce” in Table 2. Syntypes: A slide registered as NSMT-Me-2 was deposited at the National Science Museum, Tokyo, Japan. This slide was prepared from No. MI95 octopus (Table 2) and contains nema- togens and rhombogens of D. clavatum exclu- sively. Other slides were numbered according to the host number in Table 2 and are in the authors’ collection. Etymology: The specific name “clavatum” 1s given, because of the shape of the body is characteristi- cally clavate. DESCRIPTION Diagnosis: Body length up to 1000 um. Peripheral cell number of vermiform phases usual- ly 22:4 propolars, 4 metapolars, 14 trunk cells. Propolar and metapolar cells flat and covering the enlarged end of the axial cell. The infusoriform embryos consist of 39 cells. Nucleus number of urn cell, 2. Host, Octopus minor. Nematogens (Figs. 9-11): Body, pestle-like; 200 to 1000 “zm long; 60 to 120 ~m wide. Peripher- al cell number usually 22; 4 propolars, 4 metapo- lars, 2 parapolars, 10 diapolars, 2 uropolars, diapo- lars sometimes lower (Table 7). Cephalic swelling distinct. Calotte, smoothly rounded and cap- shaped in most individuals, flattened and disc- shaped in a few individuals. Cilia covering calotte about 6 um long, oriented forward, more densely distributed than those on trunk cells. Trunk cells arranged in opposed pairs. Uropolar cells occa- sionally verruciform; their nuclei, often enlarged; their cytoplasm, stained intensely by hematoxylin. Axial cell extends anteriorly to base of propolar cells, typically enlarged and rounded in calotte regions. Rounded tip of axial cell is covered by a cap of thin propolar cells and by metapolar cells often arranged as a band. Large individuals en- Dicyemid Mesozoans from Japan 433 close about 80 vermiform embryos, and peripheral cells often decrease to 21, sometimes to 20, in number. Vermiform embryos (Fig. 12): Full-grown ver- miform embryos, 70 to 100 um long; 12 to 18 um wide; peripheral cell number fixed at 22 (Table 7). Ratio of total body length to calotte length, 1:0.17 to 1:0.19. Anterior end of calotte, rounded. TABLE 7. Number of peripheral cells of Dicyema clavatum sp. nov. Number of individuals No. of ; cells Vermiform embryos Nematogens Rhombogens 22 59 AA 20 Fic. 9. D. clavatum sp. nov. Four entire nematogens of various sizes are shown on the left and five rhom- bogens on the right. Note the characteristic balloon- ing of anterior part of the axial cell. Bar represents 100 ~m. Drawn from specimens prepared from No. MI95 octopus. Fic. 10. Photographs of a nematogen (a) and a rhombogen (b) of D. clavatum, after the same procedure as used in the case of D. japonicum shown in Fig. 2. AX, axoblast; IG, infusorigen; M, metapolar cell; PA, parapolar cell; UR, uropolar cell. See the legend to Fig. 2 for other abbreviations. Bars represent 50 ~m. Prepared from No. MI95 octopus. 434 H. FuruyA, K. TSUNEKI AND Y. KOSHIDA Fic. 11. Anterior part of three nematogens (top) and three rhombogens (bottom) of D. clavatum. The fat left nematogen and the far right rhombogen are depicted in optical section. Note the band-like appearance of the metapolar cells around the axial cell. Bars represent 50 ~m. Drawn from specimens prepared from No. MI95 octopus. Fic. 12. Vermiform embryos of D. clavatum. The left Fic. 13. Infusorigen of D. clavatum. Around the in- embryo is depicted in the optical section to show one fusorigen, a fertilized egg at the two-pronuclei stage axoblast (AX) adjacent to the axial cell nucleus (N). (E), a developing infusoriform embryo (DI), and an The drawing of the right embryo shows an opposed- embryo at the two-cell stage (T) are found. Bar type arrangement of diapolar peripheral cells. Bar represents 10 4m. Drawn from a specimen prepared represents 10 zm. Drawn from specimens prepared from No. MI95 octopus. from No. MI95 octopus. Dicyemid Mesozoans from Japan 435 Fic. 14. Infusoriform embryo of D. clavatum. The dorsal view (a), the ventral view (b), a sagittal section (c), anda horizontal section (d, cilia omitted) are shown. Enveloping cells are omitted. SC, short cilia; V3, third ventral cell. See the legend to Fig. 7 for other abbreviations. Bar represents 10 ~m. Drawn from specimens prepared from No. MI95 octopus. Trunk cells, arranged in opposed pairs. Axial cell nucleus, usually located in the center of the cell, one or two axoblasts anteriorly. Rhombogens (Figs. 9, 10 and 11): Slightly smal- ler than nematogens, 200 to 600 «m long; 60 to 140 ym wide. Shape of calotte similar to that in nematogens, although cephalic swelling more mas- sive and broader as individuals grow. Cilla cover- ing calotte, similar in length and in orientation to those of nematogens. Two uropolar cells, some- times one additional cell adjacent to uropolars, occasionally verruciform; these cells often with large nuclei, with cytoplasm staining intensely with hematoxylin. Shape of axial cell and arrangment of both propolar and metapolar cells with respect to axial cell, similar to nematogens. Infusorigens (Fig. 13): Relatively small. Axial cell usually ovoid, about 12 to 17 um in long axis. About one-third of axial cell surface, always ex- posed. In 25 infusorigens examined: number of external cells including oocytes, 5 to 12 (mode, 6); number of internal cells including spermatocytes, 2 to 5 (mode, 3). Fertilized eggs, about 12.1 ~m in diameter. Infusoriform embryos (Fig. 14): Ovoid and rounded bluntly to pointed posteriorly. Based on 100 full-grown embryos, length (excluding cilia), 24.10+1.35 wm (mean+S.D.); length-width- height ratio, 1:0.86:0.80. Cilia at posterior end about 7 to 8 wm long. Refringent bodies, about equal in size to total mass of four urn cells, occupy about 40% of anterior part of embryo, when viewed from lateral side. Nuclei of anterior lateral cell, small and pycnotic. Ventral internal cells project cilia into urn cavity. Short cilia growing from apical cells through dorsal fenestrae of en- 436 H. Furuya, K. TSUNEKI AND Y. KOSHIDA veloping cells. Full-grown infusoriform embryos consist of 39 cells (35 somatic and 4 germinal) and 43 nuclei in total (Table 4). Somatic cells com- posed of peripheral cells that cover anterior and lateral surfaces of embryo (2 enveloping cells), peripheral cells with cilia (2 apical cells, 2 paired dorsal cells, 1 median dorsal cell, 2 dorsal caudal cells, 2 lateral caudal cells, 1 ventral cell, 2 lateral cells, 2 posteroventral lateral cells), peripheral cells without cilia (2 first ventral cells, 2 second ventral cells, 2 third ventral cells, 2 anterior lateral cells, 1 convercle cell), internal cells with cilia (2 ventral internal cells), and internal cells without cilia (2 capsule cells, 2 dorsal internal cells, 4 urn cells). Each of four urn cells has two nuclei of its own and one germinal cell with its own nucleus (Table 4). REMARKS D. clavatum is very similar’ to D. robsonellae Short, 1971 [7], which was isolated from Rob- sonella australis in New Zealand, in the enlarged head and cap-shaped calotte. However, differ- ences between D. clavatum and D. robsonellae are distinct in terms of the peripheral cell number in vermiform phases (22 vs. 20), in the cell number of infusoriform embryos (39 vs. 37), and in the occurrence of cilia on ventral internal cells (pre- sent vs. absent), respectively. D. benthoctopi [6], D. orientale [1], and D. TABLE 8. specie atte Dicyema acuticephalum 16-19 Dicyema aegira DD) Dicyema bilobum 16-18 Dicyema megalocephalum 16 Dicyema misakiense MD Dicyema monodi* 16 Dicyema paradoxum 28 Dicyema _typoides 18 Dicyema typus 18-19 Dicyemennea lameerei 23 Conocyema polymorpha i Japonicum have 22 peripheral cells and an enlarged head, but each of these species has a conspicuously swollen and disc-shaped calotte, instead of a cap- shaped one. The hosts of these three species and D. clavatum are also different from one another. O. minor is now regarded as a member of the O. macropus species complex [8]. D. paradoxum von Kolliker, 1849, [2, 9] was isolated from O. macro- pus in Europe, but this dicyemid species could be easily distinguished from D. clavatum by the num- ber of peripheral cells (25-28 vs. 22). D. clavatum is the first dicyemid species obtained from O. minor to be described. DICYEMID FAUNA IN Octopus vulgaris AND Octopus minor In 22 out of 23 individuals of O. vulgaris, we found a total of four kinds of dicyemid, i.e., D. misakiense, D. acuticephalum, D. japonicum, and one more dicyemid not yet identified to species (Table 1). This unidentified dicyemid certainly belongs to the genus Dicyema because its calotte consists of 4 propolar and 4 metapolar cells that are not twisted in their arrangements; therefore this dicyemid is tentatively termed Dicyema sp. (A). As shown in Table 1, three species of dicyemids, including Dicyema sp. (A), were de- tected together in two hosts; two species of dicyemids, D. misakiense and D. japonicum in all cases, were found in 16 hosts; and a single species of dicyemid was found in four hosts. Only in one Dicyemid species recorded ..° parasites of Octopus vulgaris in the literature Distribution Reference Japan [2] Florida, U.S.A. [5] Florida, U.S.A. [10] West Africa [12] Japan [1] West Africa [12] Europe and West Africa [9] Florida, U.S.A. [11] Europe [2] Europe [13] Europe [14] * According to Bogolepova-Dobrokchotova [15], this species name is a synonym of D. megalocephalum. Dicyemid Mesozoans from Japan 437 individual, listed as Host no. VU15, the smallest in size among the octopuses examined, no dicyemids at all were detected. In another small octopus, Host no. VU101, dicyemids were scarce, although two species were found in one of the renal sacs, while none were found in the other. In Table 8 we have listed all the dicyemid species from Octopus vulgaris that we were able to find in the literature. It is apparent that a considerable number of dicyemid species have been recorded in association with this cephalopod. In Octopus minor, nine out of nineteen indi- viduals were infected by D. clavatum (Table 2). In five out of these nine individuals, in addition to D. clavatum, another species was also found. This species clearly belongs to the genus Dicyema be- cause of the cell composition and arrangement of the calotte, but the species is unidentified as yet and is named tentatively Dicyema sp. (B) (Table 2). In the octopus listed as Host no. MI92, this unidentified species was found in one renal sac, but D. clavatum was found in other one. Yet another species belonging to the genus Dicyema, tentative- ly named Dicyema sp. (C), was found in O. minor (Table 2), but it too has not yet been fully char- acterized. ACKNOWLEDGMENTS We wish to express our sincere gratitude to Dr. Takasi Tokioka, the former Professor and Director of the Seto Marine Biological Laboratory, Faculty of Science, Kyoto University, for his active interest and valuable advice, and also to Ms. Kikuko Kurihara for her constant assistance in the course of this work. We are also indebted so much to Dr. Minoru Imajima, Head of the Zoology Division, National Science Museum, for kindly allowing us to deposit syntypes of two new dicyemid species at the Museum. REFERENCES 1 Nouvel, H. and Nakao, Y. (1938) Dicyémides du 10 iLL WZ, 3) 14 15 Japon. Bull. Soc. Zool. Fr., 63: 72-80. Nouvel, H. (1947) Les Dicyémides. le partie: systé- matique, générations vermiformes, infusorigéne et sexualité. Arch. Biol., 58: 59-220. Nouvel, H. (1948) Les Dicyémides. 2e partie: infusoriforme, tératologie, spécificité du para- sitisme, affinités. Arch. Biol., 59: 147-223. Short, R. B. and Damian, R. T. (1966) Morphology of the infusoriform larva of Dicyema aegira (Meso- zoa: Dicyemidae). J. Parasit., 2: 746-751. McConnaughey, B. H. and Kritzler, H. (1952) Mesozoan parasites of Octopus vulgaris Lam. from Florida. J. Parasit., 38: 59-64. Hochberg, F. G. and Short, R. B. (1970) Dicyemen- nea littlei sp. n. and Dicyema benthoctopi sp. n.: dicyemid Mesozoa from Benthoctopus magellanicus. Trans. Amer. Microsc. Soc., 89(2): 216-224. Short, R. B. (1971) Three new species of Dicyema (Mesozoa: Dicyemidae) from New Zealand. Biolo- gy of the Antarctic Seas, IV: 231-249. Taki, I. (1981) A catalogue of the Cephalopod a of Wakayama Prefecture. Publications of the Seto Marine Biological Laboratory, Special Publication Series, Vol. 7: 233-264. Stunkard, H. W. (1948) Dicyema paradoxum von Kolliker, 1849. Science, 108: 565-566. Couch, J. A. and Short, R. B. (1964) Dicyema bilobum sp. n. (Mesozoa: Dicyemidae) from the northern Gulf of Mexico. J. Parasit., 50: 641-645. Short, R. B. (1964) Dicyema typoides sp. n. (Meso- zoa: Dicyemidae) from the northern Gulf of Mex- ico. J. Parasit., 50: 646-651. Nouvel, H. (1934) Observations sur les Dicyémides provenant d’un poulpe de Mauritanie, description de deux espéces nouvelles. Bull. Soc. Zool. Fr., 59: 176-186. Nouvel, H. (1932) Un dicyémide nouveau du poulpe Dicyemennea lameerei n. sp. Bull. Soc. Zool. lB, Se AiG@—2). Beneden, E. van (1882) Contribution a Vhistoire des Dicyémides. Arch. Biol., Paris, 3: 195-228. Bogolepova-Dobrokchotova, I. I. (1963) The mod- ern system of dicyemides. Parazit. Sbor., 21: 259- ile ee a ee et oe ; ie apiey “Lariat: “ois Mert ate ds: a eicaneriedibeWe a pa ine, Lf Yotriss Lic: Ce) ORR aR By elighh,’ tte aie oe oc Se OTR) EE, ne bee. a ome a eS ena Gi IAD apatinnd or ee ie fe aa as, ee ey na <- at sig se api AN Big ne Bost fi bansioe, x RS o : “FBT Gh Ibe SA Sek Saat -- ee di i: Base! pote ake Fag hiey® fe Pree} er Br Hagel we Celia abv: fe to ees Peitee te toehi ers Toit] ‘gnxny of} ; he se - PRR ie One £ Wi ts shale ahd tee ‘ E lise 3 hooks sti y nya fod {PAE ay Ee as. - aie eee ~ ont te arrests Baies ae eraneeay. eile Sigurt fsrone- sie, dfioeniel issigdlnit-sniisM vt we se we 4 _ : te eo a ae ot) at, Aha ae Ad i wy be a A lie A or @ EV E y te ate 7 AALS. 9h Sat Rg tN soit sat nrg ee CBA viet +h! Risse fal a eer ee ee BOF 48nd BL EAN fe agaist Pind a a (Gap habes. cantante coil Laborde aude! OF sttife me bet if “he 1iK> Las, oat. a) 7 - TT eyenese FOES OU 9 FER FX Wasi , at Ph m Loe See OM, | RES ret . 2 ~~ as ea ‘ : - re.ce riy ~ oar 7 ~~ ¢ { st —— . ts 5 ° F pf Me * a 9 : De Ee = 12 - = =. ; = ~ & : ; : : x % ; = ay s Y bryy Arte. pity, : g e . = EES Pia ? My 7 shay L4 ; $ ep eae? che | ‘ 7 - Peg chy hs | / } iy i rs Oe) Ae a SveaTE EVE) ig) MELTED y f i ry + ] . TOLAT oe ee a* ' 2 = ave 6 r? “| ‘ tae s} Ache | y] ' 5 wey. : ey eee ays plat " £3 Ch 24M AEP SSES any ; " x) . + Da fa S Ose are tei hess + 7 - ¢ v 7 1 4 \ 5 yn te i ; 7 are t ae ry - A ar 3 ai v i we ~~ 2 te n ~ p % - F ' he. ak r he , a eee ee ; . ¥ i a . + » ; . — S n rs bet an P ¥ ares : a) 4 +54 2 € , = y ’ p. ; Es = ; re . ‘ ud ra € . ¥ 4 , ij " 4 : j J 4 F gee! Z — . , oe he a * \ ee ) ; * ) : ree 3 ri Hi } ‘ “ J i ) » t 53 yf i = a : = 2 i ; ; I P t A I ana 2 : = 4 j ae ey \ ZOOLOGICAL SCIENCE 9: 439-444 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Origin of Somatic Cells in Bidder’s Organ and the Gonad Proper in the Toad, Bufo japonicus formosus AKIHIKO TANIMURA! and HISAAKI _IwaASAwa2 ‘Department of Dental Pharmacology, School of Dentistry, Higashi-Nippon-Gakuen University, Ishikari-tobetsu, Hokkaido 061-02, and * Biological Institute, Faculty of Science, Niigata University, Niigata 950-21, Japan ABSTRACT—The development of Bidder’s organ was observed ultrastructurally and made a comparison with the gonad proper in Bufo japonicus formosus. The primordia of Bidder’s organ and the gonad proper consisted of coelomic epithelial cells and primordial germ cells. The gonad proper showed a typical cortico- medullary structure. After the gonadal sex differentia- tion, the medulla in the primordial testis developed together with migration of germ cells from the cortex, and the medulla in the primordial ovary developed as a somatic cell mass. Bidder’s organ showed no cortico- medullary structure and no sexual difference. Numerous gonial cells enlarged without meiotic nuclear change, and were surrounded by a layer of epithelial cells. The origin of somatic cells in Bidder’s organ was identical with that in the gonad proper, so that the peculiar development of Bidder’s organ is not attributed to the cellular origin. We discuss the possible cause of the development of Bidder’s organ. INTRODUCTION Bidder’s organ, which is peculiar to Bufonidae, develops into an ovary-like organ irrespective of the genetic sex [1, 2]. In this organ, follicles appear at earlier stages than in the ovary [2, 3-8], but remain immature [9, 10]. The ovary-like development of Bidder’s organ has been attributed to the agenesis of the medulla [2, 8], which derived from mesonephric or interrenal blastemal cells [{11, 12]. In the last decade, however, ultrastructural studies on anuran gonads have shown that the gonadal medulla is formed with coelomic epithelial Accepted November 21, 1991 Received September 18, 1991 cells [13-18]. These findings have cast doubt on the theory presented by Witschi [19] that the gonadal sex differentiation is attributed to the antagonistic development of the cortex and medulla. In the present study, the embryological origin of somatic cells in Bidder’s organ and the gonad proper in the toad Bufo japonicus formosus was examined first by electron microscopy. MATERIALS AND METHODS Freshly spawned egg strigns of Bufo japonicus formosus were collected in Niigata City, in the beginning of April. Hatched larvae were kept in dechlorinated tap water at 18+1°C and were fed commercial fish meal. Under these conditions, larvae metamorphosed 45-50 days after fertiliza- tion. To determine the developmental stages of the larval specimens, the normal table of develop- ment of this species [20] was used. Gonads and surrounding tissue obtained from 10-15 animals were fixed every 3-6 days in Karnovsky fixative (1.3% glutaraldehyde and 1.3% paraformaldehyde in 0.08M _ cacodylate buffer) for 1.5-2 hr at room temperature, then postfixed in 1% OsO, in the same buffer for 2 hr at 4°C. After dehydration through a graded series of ethanol, specimens were embedded in Epok 812. Thin and semithin serial sections were cut using a Porter-Blum MT-1 ultramicrotome. Thin sections were stained with uranyl acetate and lead citrate, and examined using a Hitachi H-300 electron 440 A. TANIMURA AND H. IwASAWA microscope. Semithin sections were stained with a mixture of 1% toluidine blue and 1% borax for light microscopic observations. RESULTS The genital ridge, the primordium of the gonad proper and Bidder’s organ, first appeared at stage 30 (external gill stage) at the ventral medial region of the mesonephros, and consisted of coelomic epithelial cells and primordial germ cells (PGCs) (Fig. 1). The primordium projected into the coelom in stages 32-33, and then the cortico- medullary structure was observed clearly at the middle to posterior region of the gonad proper at Fic. 1. Cross sections of the urogenital region at stage 30. a: The anterior region of the genital ridge, the primordium of Bidder’s organ. Scale bar: 50 um. b: High magnification at the open arrow in Fig. la. Scale bar: 5m. c: Middle region of the genital ridge, the primordium of the gonad proper. Scale bar: 50 wm. CE: coelomic epithelial cells, G: pri- mordial germ cell, Md: mesonephric duct, Ms: mesentery, arrow: lipid droplet: asterisk: yolk gran- ule. stage 40 (toe development stage) (Fig. 2a). Mesenchymal tissue invaded into the gonad proper simultaneously with the appearance of the cortico- medullary structure. The cortex and medulla were connected directly to one another and the two regions were covered with a continuous basal lamina (Fig. 2b), which suggests that both the cortical and medullary somatic cells are derived from coelomic epithelial cells. Sexual differentia- tion occurred at stage 41 (hindlimb completion) in the gonad proper. In the primordial testis, the germ cells and somatic cells migrated to the medulla through the connections between the cortex and medulla in stages 42-45 (climactic metamorphosis) (Fig. 4). In the primordial ovary during these stages, germ cells were situated in the cortex, and the medulla developed as a somatic cell mass (Fig. 5). At stage 40, Bidder’s organ consisted of epi- thelial cells, filled with lipid droplets, and germ cells (Fig. 3). At stage 42, lipid droplets within the epithelial cells decreased rapidly and numerous gonial cells which had a lobular nucleus enlarged without meiotic nuclear change. At stage 45, the. enlarged germ cells were surrounded by follicular cells which originally consisted of the epithelial tissue of Bidder’s organ (Fig. 7a). The follicular cells were covered with the basal lamina, and were connected with an enlarged germ cell with micro- villi (Fig. 7b). Mesenchymal cells and blood vessels invaded to the periphery at stage 42 (Fig. 6), and then were distributed around the follicles (Fig. 7). These extragonadal cells were separated from the epithelial cells by the basal lamina (Fig. 6b and 7b). Neither the cortico-medullary struc- ture nor sexual difference was observed in Bidder’s organ. DISCUSSION The present study is the first to conclude with electron microscopic findings, that Bidder’s organ consists of the coelomic epithelium-derived cells and germ cells, and that the medullary structure is not developed. Light microscopic studies have reported the absence of a medulla in Bidder’s organ [2, 8], while the presence of slight medullary elements have also been claimed [5, 6]. The Origin of Somatic Cell in Bidder’s Organ 44] Fic. 2. Sexually indifferent gonad at stage 40. a: Cortex (C) and medulla (M) can be distinguished. G: germ cell, Scale bar: 20 um. b: Higher magnification of the priomordial gonad indicated by the open arrow in Fig. 2a. Cortex and medulla are tightly connected and are covered with a common basal lamina (arrow heads). Scale bar: 2 wm. Fic. 3. Buidder’s organ at stage 40. Germ cells (G) and somatic cells are filled with lipid droplets (arrows). Scale bar: 20 um. Fic. 4. Cross section of testis of larva at stage 43. Germ cells (arrows) are migrating from the cortex (C) to the medulla (M) through their connection. Scale bar: 30 um. Fic. 5. Cross section of ovary of larva at stage 45. The medulla (M) develops as a somatic cell mass. All germ cells are situated in the cortex (C). Scale bar. 30 wm. A. TANIMURA AND H. IWASAWA ion of mesenchymal cells (small arrows) and blood vessels Invas a . s organ of female larva at stage 42 b] idder B is seen 6. FIG : 50 um. b: Higher Scale bar issue (large arrows) Mesenchymal cells (Me) are separated from the ep ithelial t the ep In ing 6a 1 cells are enlargi ia Some gon (B) ] tissue (E) thelia i 1g G F In he open arrow ina (arrow heads) s organ of female larva at stage 45 I cells fication at t by the basal lam Te magni 2 um icles (large arrow) cons scale bar: Foll in the enlarged germ cell ’ : germ cell t of enlarged germ cells and a layer 1S a >) dder i it B FIG Normal gonial cells (small arrows) are seen ial cells (E) is covere 1S seen 50 um A lobular nucleus of epithelia he per basal 2 um. th a continuous i dw mesenchymal cell, scale bar thel A layer of epi b th germ cells (G) via m Scale bar: 10n heral regi Ip in t lli. Me icrovl in contact wi 1S ina (arrow head) and lam Our observations, lack of male inductors [2, 8] 1C- medulla has been regarded as a mesonephr however, indicate that cells migrated from the mesonephric region were separated from the derived male inductor [11, 19], and the ovarian d to show a S Organ 1s interprete nature of Bidder’ Origin of Somatic Cell in Bidder’s Organ 443 epithelial tissue within the gonad proper by the basal lamina and were not included in the medulla. The cortex and medulla were directly connected and were covered with a common basal lamina, which strongly suggests the coelomic epithelial origin of the medulla. This is consistent with findings obtained in previous ultrastructural studies on several anuran species [13-18]. It is indicated that the origin of the somatic cells in Bidder’s organ is identical to that of the cells in the gonad proper, and there is no contribution of the cellular origin to the ovarian nature of Bidder’s organ. Lepori deduced that the most medial part of the coelomic epithelium has masculinizing capabilities, and that the absence of this part in Bidder’s organ causes the ovary-like development [4]. Although this hypothesis fits with current opinions concerning the origin of gonadal somatic cells, the presence of masculinizing capabilities of such parts is difficult to demonstrate. Our observations indicate that the invasion of the mesenchyme into the gonad proper occurred at the hindlimb stage simultaneous with the develop- ment of the medulla, whereas the invasion into Bidder’s organ occurred with the formation of follicles at the metamorphosing stage. It is pos- sible that the delay in the invasion of mesenchymal tissue can be attributed to the agenesis of the medulla in Bidder’s organ. In B. japonicus formosus, the medulla differentiated into semi- niferous cords and the inner epithelium of the ovarian cavity in the testis and ovary, respectively. Therefore, Bidder’s organ should have no semi- niferous cord or ovarian cavity, since the medulla was not developed. In the early testicular develop- ment, germ cells were incorporated within the seminiferous cord and did not enter into meiosis. In Bidder’s organ, oogenesis occurs at the peripheral region [7], and no particular structure was formed around germ cells, like in the ovary. Thus, the agenesis of the seminiferous cords which is likely due to the lack of a medulla contributes to the female type differentiation of germ cells in Bidder’s organ. It is reported that follicles observed in Bidder’s organ at the metamorphosing stage are formed without meiosis [4-7, 21]. We show clearly that these follicles were formed by a direct enlargement of gonial cells and subsequent enclosure of the germ cell within a layer of epithelial cells. Sur- prisingly, contacts via microvilli were observed among the enlarging gonial and follicular cells, which appear in normal oocytes at the diplotene stage in the meiotic prophase [17, 18, 22]. This is an important phenomenon in understanding the changes in the interaction between germ cells and somatic cells during oogenesis. REFERENCES 1 Martha, K. P. R., Hobart, M. S. and Chiszar, D. (1990) Amphibia-Reptilia, 11: 225-235. 2 Witch, E. (4933) Am. J- Anat., 52: 461-515. 3. Asayama, S. (1959) J. Inst. Polytech. Osaka City Univ. Ser. D, 10: 129-141. 4 Lepori, N. G. (1980) Sex Differentiation, Hermap- roditism and Intersexuality in Vertebrates including Man, Piccin Medical Books, Padua. 5 Takahashi, H. (1956) J. Fac. Sci. Hokkaido Univ., Ser. VI, 12: 297-308. 6 Tanimura, A. and Iwasawa, H. (1986) Sci. Rep. Niigata Univ., Ser. D, 23: 11-21. 7 Tanimura, A. and Iwasawa, H. (1987) Zool. Sci., 4: 657-664. 8 Witschi, E. (1956) Development of Vertebrates. W. B. Saunders Co., Philadelphia. 9 Moriguchi, Y., Tanimura, A. and Iwasawa, H. (1991) Sci. Rep. Niigata Univ., Ser. D, 28: 11-17. 10 Wachtel, S. S., Wachtel, G. and Nakamura, D. (1990) In “Vertebrate Endocrinology: Fun- damentals and Biomedical Implications”. Ed. By P. K. T. Pang and N. P. Schreibman, Academic Press, Inc., San Diego, pp. 149-180. 11 Lofts, B. (1984) In “Marshall’s Physiology of Reproduction”. Ed. by G. E. Lamming, Churchill Livingstone, Edinburgh, pp. 127-205. 12 Vannini, E. and Sabbadin, A. (1954) J. Embryol. Exp. Morph., 2: 275-289. 13. Iwasawa, H. and Yamaguchi, K. (1984) Zool. Sci., 1: 591-600. 14 Merchant-Larios, H. (1978) In “The Vertebrate ovary”. Ed. by R. E. Jones, Plenum Press, New York, pp. 47-81. 15 Merchant-Larios, H. and Villalpando, I. (1981) Anat. Rec., 199: 349-360. 16 Tanimura, A. and Iwasawa, H. (1988) Develop. Growth and Differ., 30: 681-691. 17 Tanimura, A. and Iwasawa, H. (1989) Embryol., 180: 165-173. 18 Tanimura, A. and Iwasawa, H. (1991) J. Exp. Zool., 259: 365-370. Anat. 444 A. TANIMURA AND H. IWASAWA 19 Witschi, E. (1957) J. Fac. Sci. Hokkaido Univ., Ser. 256-265. VI, 13: 428-439. : 21 King, H. D. (1908) J. Morph., 19: 439-465. 20 Iwasawa, H. (1987) In “Biology of Toads”. Ed. by 22, Dumont, J. N. (1972) J. Morph., 136: 153-180. A. Urano and K. Ishihara, Shokabo, Tokyo, pp. ZOOLOGICAL SCIENCE 9: 445-450 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Tooth Development and Replacement in the Japanese Greater Horseshoe Bat, Rhinolophus ferrumequinum nippon KIMITAKE FuNAKosHI!, YUKO FUKUE” and SHON TABATA? ‘Biological Laboratory, Kagoshima Keizai University, Kagoshima 891-01, "Department of Biology, Faculty of Science, Kagoshima University 890, and °Department of Oral Anatomy, Kagoshima University Dental School, Kagoshima 890, Japan ABSTRACT—The dental formula of the deciduous teeth in Rhinolophus ferrumequinum nippon was i1/2 cl/1 p2/3X2=20. All the germs of deciduous teeth were present in the fetuses of 7-8 weeks. The deciduous teeth, exclusive of the p3, attained their maximal size at 9-10 weeks, and then odontoclasts appeared in the dental pulp and began to resorb the dentin. The resorption and exfoliation of the teeth have completed at birth. The typical sequence of resorption of the deciduous teeth was iy, Ib, C1, © , P-> Po, 1-, P, Pa. On the other hand, the p; appeared late, developed slowly, and was remained in a small size, while the tooth germ of the P3 was degener- ating. The permanent teeth began to erupt at birth, and the usual sequence of their eruption was I,, In, C’, Cy, Pye (25M, M>, M*, M”), Mz, M°,.(?; P2, p;). The incidence of lack of the p3 was 13.2, 5.3 and 5.3% for the left, right and both sides of the jaw, respectively. Such a congenital abnormality suggests that the p3 is genetically undergoing a process of degeneration. INTRODUCTION Unlike deciduous teeth of most mammals, those in the Chiroptera are not functional in feeding. In most bat species, however, the deciduous teeth are highly specialized, strikingly different from the permanent ones, and used by the young in clinging to the maternal nipple [1-11]. On the other hand, in at least four genera, Rhinolophus (Rhinolophi- dae), Lavia (Megadermatidae), Hipposideros and Accepted December 26, 1991 Received October 11, 1991 * Present address: Department of Biology, Faculty of Science, Kanazawa University, Kanazawa 920, Japan. Triaenops (Hipposideridae), all the deciduous teeth are resorbed or exfoliated prior to birth [12- 14], and in Tadarida (Molossidae) some of them are also resorbed prenatally [15]. Reasons for such differences are not clear, but it has been suggested that the young in such genera are for some reasons in less need of the deciduous teeth after birth [3]. From an evolutionary or phylogenetic point of view, such deciduous dentition and its replacement would be of interest to study. However, there have been few reports which dealt with the sequential resorption and histological changes of the deciduous teeth in bats [14]. The present study was therefore carried out to examine histological changes of the deciduous teeth in the Japanese greater horseshoe bat, Rhinolophus ferrum- equinum nippon. MATERIALS AND METHODS A total of 34 (14 pregnant females and 20 young males) were collected at the tuffaceous cave of Katano-d6 in Kagoshima Prefecture during the period from May to August in 1988. On the day of capture, the fetuses were taken out from the mothers under ether anesthesia, weighed, meas- ured of the lengths of skull and forearm, and sexed (see Table 1). The age of each fetus or young was determined on the basis of the assumptions that in R. f. nippon, fertilization would have occurred on ~3 April [16], and the greatest length of the skull would have increased at a rate of 0.32 mm per day 446 TABLE 1. K. FUNAKOSHI, Y. FUKUE AND S. TABATA External measurements, body weights and estimated age of fetus in R. f. nippon. Greatest length Forearm Body Estimated age Date captured Fetus Os of skull (mm) length (mm) weight (g) — of fetus in weeks** 29 May SPE & 10.1 dod 0.89 7 29 May SPY & tee 7.4 1.03 8 29 May 5JJ di 1S 8.5 18 8 29 May SPB a SS) 8.8 L25) 8 29 May SPC © 11.6 8.3 126 8 11 June 6PD é 14.3 11.3 2.40 9 11 June 6PZ @ 14.5 12 2.66 9 Iie une 6PU = So5) (Bef BalG 10 11 June 6PL é 15.8 14.6 3.61 10 11 June 6PW é' 16.2 14.5 3.90 10 26 June 6PT & 18.5 18.0 4.75 il 26 June 6PC e 18.6 20 Seo”) 11 26 June 6PB o 19,5) 24.2 6.72 tal 26 June 6PH* Bf 21.0 28.5 7.58 1 * Newborn young shortly after birth. *“ Estimated from the presumed day (~3 April) of fertilization [16]. on and after the 3rd week [17]. RESULTS Of 14 fetuses, 8 were fixed in 10% formalin and their skulls were decalcified in Plank-Rychlo’s fluid, embedded in Sorvall embedding medium PN 45582 (DuPont Comapny, Connecticut). Trans- verse serial sections (3 um) for light microscopy were stained with Mayer’s hematoxylin and eosin. Six fetuses were fixed in 80% ethanol and their skulls were stained with alucian blue 8GS and alizarin red S. Cartilaginous tissues are stained deeply blue with the former, while bone and teeth are stained deeply red with the latter. The deciduous teeth were observed using a binocular microscope. Adult females or young males rang- ing from newborn to volant sizes were fixed in 10% formalin and transferred to 70% ethanol; their teeth were observed in situ using a binocular microscope. The tooth nomenclature used here was followed by that of Miller [2]. Uppercase letters signify permanent teeth and lowercase letters, deciduous ones; numbers indicate tooth positions in the upper or lower jaw (e.g. P* or P> respectively). The dental formula of permanent teeth in R. f. nippon has been known as follows [aS WL 2ACiGERASNB3)—=32. Development and resorption of deciduous teeth The dental formula of the deciduous dentition in the Japanese greater horseshoe bat was found as, 11/2 cl/1 p2/3 X2=20. In the fetuses of the 7th or 8th week, all the germs of the deciduous teeth were observed in the upper and lower jaws (Figs 1A, B and 2A-E). In these deciduous tooth germs, predentin or dentin has been formed with the exception of p3. The p3 appeared later in the development and was still in the cap or bell stage at the 8th week (Fig. 2E). The dental laminas of permanent teeth were already present in the alveoli lingually to the deciduous teeth. Espe- cially, the P, developed just below the pz and was in the cap stage at the 8th week (Fig. 2D). At the 9-10th week, the deciduous teeth, exclusive of the p3, attained their greatest size (0.05—0.34 mm long), moved upward, and located in the tissue just over the permanent teeth. The deciduous incisors ((ige i; and i5) were simple in shape, that is, ovoid in the longitudinal section of the teeth (Figs 1C, D and 2F). The deciduous canines (c! and c,) were relatively small, with the Dental Ontogeny in Greater Horseshoe Bat 447 crowns being only slightly larger than those of the D and 2H-J). The p3 (0.19 mm long) was rooted at deciduous incisors (Figs 1C, D and 2G). The the upper-labial edge of the alveolus, and many deciduous premolars (p’, p*, p2, p3 and ps) had odontoblasts were observed along the margin of roughly triangular and recurved crowns (Figs 1C, the dental pulp (Fig. 21). Calcification of the Fic. 1. Diagrams showing the occlusal view of deciduous and permanent teeth in R. f. nippon after staining with alucian blue 8GS and alizarin red S. A-B, C-D and E-F, based on specimen No. 5PY, 6PD and 6PT, respectively. A, C, E, lower jaws; B, D, F, upper jaws. Abbreviations in Figs 1 and 2: i, c and p, deciduous incisor, canine and premolar, respectively. I, C, P and M, permanent incisor, canine, premolar and molar, respectively. K. FUNAKOSHI, Y. FUKUE AND S. TABATA 448 OE iy wo Fatt I we Mam nO” CNTY Fee Yaw NAN NG One. * Na » eM apf) Ce ¥ py ¢ < & \ é rR Ge LA ee ee SOD Se = a oh ole a , ‘ 4 won eee 2 “ Fix e = . ee -* ee ; . bi Fate tas oN Ree ants : Dy PUN. v2 Pec 4 We ae SESS 1s oni’ ) BeOS) AALS ” When » Dental Ontogeny in Greater Horseshoe Bat 449 dentin and enamel in the p3 was also in progress because of their darker staining with hematoxylin and eosin. On the other hand, the tooth-germ of the P3; became degenerated. In other deciduous teeth, calcification did not occur at this stage. Larger and deeply-staining odontoclasts, having several nuclei, appeared in the dental pulp and destroyed and resorbed the root dentin when the permanent tooth germ began to develop (Fig. 2F-H, J). Degrees of the resorption were variable in different individuals. For example, in fetuses 6PD and 6PZ (9th week), the left i;, i, and right i, had been resorbed in the former, while 1,, 1) on both sides and left c; had been resorbed in the latter. All the upper deciduous teeth, however, were still intact at this stage (Fig. 1D). At the 11th week, only the deciduous lower premolars (p> or p4) were located at the alveolar crest on the labial side. Resorptions of these teeth progressed in considerable part of the dentin (Fig. 2K, M). On the other hand, the p3 grew and located in the upper, labial and slightly posterior region of the alveoli in the developing permanent plemolar (P,4) (Fig. 2L). No deciduous teeth were observed in the newborn young (6PH) at birth (12th week) and other young males. Namely, all the deciduous teeth, except the p3, were resorbed and exfoliated prior to birth. Tooth replacement The sequence of resorption of the deciduous teeth was usually as follows: i;, ir, c;, ¢’, p*, po, 1, p', pa. However, the third deciduous premolar (p3) was not resorbed but remained. Parturition occurred from late June to early July. At birth, permanent teeth began to erupt. I,, L, C' and C, appeared within 1-2 weeks after birth, and P, and P* followed. At 3-4 weeks of age, P>, M;, M>, M! and M* erupted simultaneously, and M3 and M°* followed. I°, P* and p; appeared simultaneously during the weanling period of 5 weeks of age. The usual sequence of eruption of the permanent teeth was shown as follows (groups of teeth shown here in parentheses were considered to erupt simul- taneously): I,, Ib, C'; Cy, Ps, P*, (Ps, Mi; Mo, M?, M7’), M3, M?, (I*, P?, ps). Congenital abnormalities in the dentition of the bat were found. Of 38 specimens, five (13.2%) lacked the left p3, two (5.3%) lacked the right ps, and two (5.3%) had no p3 on both sides. DISCUSSION The present study revealed that complete num- ber of the deciduous teeth in R. f. nippon (Rhinolophidae) is expressed by the dental for- mula of 11/2 cl/1 p2/3X2=20. These teeth are greatly reduced in size and less complex in form as compared with those of the Vespertilionidae [2, 5, 6, 8, 12, 19, 20] and Phyllostomidae [2, 11]. Furhter, in R. f. nippon, the deciduous teeth, exclusive of p3, are resorbed and exfoliated prior to birth. The development of deciduous teeth in R. f. nippon is similar to that in Hipposideros ruber or Triaenops persicus (Hipposideridae) [14], though in the latter two species, the resorption process in each deciduous tooth is as yet unknown. In a newborn young of H. ruber, two small deciduous teeth are still retained [14]. Thus, it is assumed that the degeneration of deciduous teeth in R. f. nippon is more marked as compared with that in H. ruber. The formation of p3 begins later and it develops slowly in comparison with other deciduous teeth in R. f. nippon. The p3 is not replaced by the permanent premolar (P3) which is degrading, and the p3 erupts last. The congenital abnormality, that is, loss of p3 is found in R. f. nippon. The rate of loss (23.7%) is similar to that (20.4%) in the Same species examined in Nagano Prefecture [21]. These facts seem to provide a evidence that the p3 is undergoing a regression, from the evolutional point of view. In Hipposideros the p3 or P3 is missing [2]. Hipposideros seems to be genetically more advanced and specialized form than Rhino- lophus [22, 23]. Similarly, the small P* in R. f. nippon developed Fic. 2. Cross sections through the lower jaw in R. f. nippon, showing deciduous and developing permanent teeth after staining with Mayer’s haematoxylin and eosin. A-E, F-J and K-M, obtained from specimens No. 5JJ, 6PU and 6PB, respectively. d, dentin; e, enamel; ob, odontoblast; oc, odontoclast; pg, permanent tooth germ; pl, permanent dental lamina. Scale bar, 0.1 mm. 450 slowly and erupted late, and was missing in 5 out of 54 specimens (9.3%) [21], while none of the specimens examined here lost the P’. The deciduous teeth of the Phyllostomidae are generally smaller in size and morphologically less complex than those of the Vespertilionidae and Molossidae [2, 11, 24]. The size and shape of deciduous teeth seem to correlate with the mater- nal care pattern in the bat. It has been suggested that the increased complexity in the deciduous dentition of vespertilionids may facilitate for the young to grasp the mother rather than to maintain a hold on the nipple after clinging to the mother [25]. On the other hand, in rhinolophids and hipposiderids, the young are not left alone but are attaching to the mother at the roost all through the day. Such young behavior and parental care in rhinolophids and hipposiderids may not so strongly require the deciduous teeth for the young to grasp the mother as do those in vespertilionids and molossids. Eruption of the permanent teeth in R. f. nippon begins after the exfoliation of the deciduous teeth excluding the p3 has ended. The eruption, how- ever, is not necessarily associated with the exfolia- tion of their deciduous counterparts, though this is true for the lower incisors. The development of the permanent teeth in R. f. nippon, particularly I,, L, C’ and C,, is rapid as well as in H. ruber [14], and they erupt earlier than those in most bat families. Therefore, the erupted permanent teeth may permit, to some extent, for the young to hold on their mother’s nipples or on the false ones existing on the lower part of the abdomen, in place of the deciduous teeth. ACKNOWLEDGMENTS We wish to thank Prof. T. Semba of Kagoshima University Dental School, and Prof. M. Hotta and Dr. S. Yamane of Kagoshima University for their warm en- couragement and the use of their facilities, and to Drs. K. wada and T. Nakama of Kagoshima University Dental School for their valuable advice and discussions during the course of this study. REFERENCES 1 Miller, G. S. (1896) Proc. Biol. Soc. Washington, Oo CO ND 10 JU 12 tS 14 15 16 17 18 19 o) a 72) 22 ap) 24 25 K. FUNAKOSHI, Y. FUKUE AND S. TABATA 10: 113-114. Miller, G. S. (1907) Bull. U. S. Nat. Mus., 57: 1- 282. Allen, G. M. (1939) Bats. Harvard Univ. Press, Cambridge, pp. 1-368. Matthews, L. H. (1950) Mammalia, 14: 11-13. Reeder, W. G. (1953) Mus. Zool. Univ. Mich., 545: 1-3. Stegman, L. C. (1956) J. Mamm., 37: 58-63. Friant, M. (1963) Acta Anat., 52: 90-101. Fenton, M. B. (1970) Can. J. Zool., 48: 817-820. Orr, R. T. (1970) In “Biology of Bats, I”. Ed. by W. A. Wimsatt, Academic Press, New York, pp. 217- Bile Slaughter, B. H. (1970) In “About Bats”. Ed. by B. H. Slaughter and D. W. Walton, Southern Method- ist Univ. Prss, Dallas, pp. 51-83. Phillips, C. J. (1971) Misc. Publ. Mus. Nat. Hist., Univ. Kansas, 54: 1-138. Spillmann, Fr. (1927) Abhandl. Senckenbergischen Naturforschenden Gesellschaft, Frankfurt A. M., 40: 249-255. Dorst, J. (1953) Mammalia, 17: 83-84. Hermanson, J. W., Woods, C. A. and Howeil, K. M. (1982) J. Mamm., 63: 527-529. Dorst, J. (1957) Mammalia, 21: 133-135. Oh, Y. K., Mori, T. and Uchida, T. A. (1985) J. Reprod. Fert., 73: 121-126. Fukue, Y. (1989) Unpublished graduation thesis. Fac. Sci. Kagoshima Univ., pp. 1-39 (in Japanese). Imaizumi, Y. (1970) The Handbook of Japanese Land Mammals. Shin-shicho-sha, Tokyo, pp. 1-350 (in Japanese with English description). Dorst, J. (1949) Mammalia, 13: 45-48. Morii, R. (1976) J. Mamm. Soc. Japan, 6: 248-258 (in Japanese with English abstract). Miyao, T. (1973) J. Mamm. Soc. Japan, 5: 230-233 (in Japanese with English abstract). Ando, K. (1982) Unpublished Ph. D. thesis. Fac. Agr. Kyushu Univ., pp. 1-359 (in Japanese with English summary). Yoon, M. H. and Uchida, T. A. (1983) J. Fac. Agr., Kyushu Univ., 28: 135-146. Phillips, C. J., Grimes, G. W. and Forman, G. L. (1977) In “Biology of Bats of the New World Family Phyllostomatidae. Part II”. Ed. by R. J. Baker, J. K. Jones, Jr. and D. C. Carter, Spec. Publ. Mus. Texas Tech Univ., Lubbock, pp. 121-246. Kleiman, D. G. and Davis, T. M. (1979) In “Biology of Bats of the New World Family Phyllo- stomatidae. Part III.” Ed. by R. J. Baker, J. K. Jones, Jr. and D. C. Carter, Spec. Publ. Mus. Texas Tech Univ., Lubbock, pp. 367-402. ZOOLOGICAL SCIENCE 9: 451-455 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Hatching Patterns of the Monogenean Parasites Benedenia seriolae and Heteraxine heterocerca from the Skin and Gills, Respectively, of the Same Host Fish, Seriola quinqueradiata GRAHAM C. KeEarNn!, Kazuo OGawa’ and YuKIO MAENO* ‘School of Biological Sciences, University of East Anglia, Norwich, NR4 7TJ, UK, *Department of Fisheries, Faculty of Agriculture, The University of Tokyo, Tokyo 113, and *National Research Institute of Aquaculture, Nansei, Mie 516-01, Japan ABSTRACT—Eggs of the capsalid monogenean skin parasite Benedenia seriolae and the polyopisthocotylean gill parasite Heteraxine heterocerca, from the same host, the yellowtail Seriola quinqueradiata, have strikingly different hatching patterns when incubated under iden- tical environmental conditions with natural illumination and constant temperature. B. seriolae eggs hatch throughout the hours of daylight, with an indication of an early and a late hatching peak during the day, while H. heterocerca has a strong hatching peak at dusk and emergence declines in frequency during the night. The significance of this difference in hatching pattern is discussed. INTRODUCTION It has been found by one of us (YM) that the eggs of two unrelated monogenean parasites, namely Benedenia seriolae (Yamaguti, 1934) Price, 1939 and Heteraxine heterocerca (Goto, 1894) Yamaguti, 1938, from the skin and gills respective- ly of the yellowtail, Seriola quinqueradiata, readily become entangled in nylon netting suspended in the tank with the fishes. This provides an excellent opportunity to incubate and hatch the eggs of the two parasites under identical environmental condi- tions and, since the larvae of the two parasites can readily be distinguished with a dissecting micro- Accepted January 21, 1992 Received August 29, 1991 scope, to compare their daily hatching patterns. This is important because there has been only one comparative study of hatching in monogeneans sharing the same host fish, namely that of Whit- tington [1] and although S. quinqueradiata is an important food fish in Japan, cultured extensively in marine fish farms, the hatching patterns of B. seriolae and H. heterocerca are unknown. MATERIALS AND METHODS Eggs of Benedenia seriolae and Heteraxine heter- ocerca became entangled by their appendages in nylon netting with a mesh size of about 1.5 mm, suspended in tanks containing infested fishes at the National Research Institute of Aquaculture at Nansei. The netting was left in situ for about 24 hr and the following day transported to the Seto Marine Biological Laboratory at Shirahama. Two pieces of the netting about 40 cm? were cut out and each piece was placed in a crystallizing vessel with sloping sides and a bottom diameter of about 6 cm containing about 2 cm of filtered sea water. The vessels were incubated at 23°C in a glass-fronted constant temperature cabinet facing a north win- dow. The approximate official times of dawn and dusk in the Shirahama area at the time of year (October) when the experiments were conducted were 06.00 hr and 17.30 hr respectively. During 452 the period of incubation the sea water was changed twice a day by gently transferring each piece of netting with forceps to a new vessel. When free-swimming larvae were detected, the netting was transferred in the same way to fresh sea water at 2-hourly intervals throughout the day and the night. During the night a dim background light permitted detection and transfer of the white netting. After each transfer, a few drops of formaldehyde were added to the vessel from which the netting had been removed. This treatment killed rapidly any free-swimming larvae and since dead oncomiracidia of B. seriolae and H. heter- ocerca are easy to distinguish from each other with a stereomicroscope by their size and shape (Fig. 1), it was possible to determine the number of oncomiracidia of each species which had hatched during each 2 hr period. Counting was made easier by inscribing a grid on the bottom of each glass vessel and, since larvae occasionally become trap- ped at the water/air interface, it was also necessary to scan the surface of the sea water in each vessel. RESULTS Oncomiracidia were first discovered in the ves- sels on the fifth day after the beginning of the egg collection period. Regular sampling at intervals of TABLE 1. 2 hr intervals beginning on Day 1. G. C. KEARN, K. OGAWA AND Y. MAENO Fic. 1. The shapes and main features of preserved oncomiracidia of A, Benedenia seriolae and B, Heteraxine heterocerca. e, Eye. Scale bar=50 um. 2 hr was begun immediately and continued without interruption until no more larvae were available for hatching (on the fourth day of recording in B. seriolae and on the second day in H. heterocerca). The hatching patterns for the two parasites from Numbers of larvae of Benedenia seriolae and Heteraxine heterocerca collected at B. seriolae H. heterocerca Time Day 1 Day 2 Day 3 Day 4 Day 1 Day 2 Day 3 04.00—06.00 — 81 94 12 2 0 (dawn 06.00) . 06.00—08.00 — 467 214 16 — 0 0 08.00—10.00 — 214 173 5 — 0 0 10.00—12.00 — 209 22 3 — 1 0 12.00-14.00 2 282 79 1 9 0 0 14.00-16.00 14 943 48 2 2 OF 0 16.00-18.00 38 408 24 0 187 107 0 (dusk 17.30) 18.00—20.00 ful 58 1 0 109 13} 0 20.00—22.00 1 12 0 —- 18 0 0 22.00-24.00 0 3 0 -— 14 0 — 24.00—02.00 0 0 0 8 0 — 02.00—04.00 0 0 0 — 5 0 — The figures are the sums of larvae from two separate experiments. —, no observation made. Hatching Patterns in Two Monogeneans 453 the two separate experiments were similar and the were 3437 and 475, respectively. It can be seen results from the two experiments have been added from Table1 that B. seriolae larvae hatched together (see Table 1). The total numbers of throughout the hours of daylight but after dusk larvae of B. seriolae and H. heterocerca collected hatching was curtailed and few larvae emerged 30 20 Percentage hatch 06.00 08,00 10.00 12.00 14.0016.00 1800 2000 2200 2400 0200 0400 70 Time 60 50 <= 2 £ ® d) i] = = 0 Oo ben ® a 20 10 0600 0800 10.00 1200 14.00 16.00 18.00 20.00 22.00 24.00 0200 04.00 Sl se Time Fic. 2. The daily hatching pattern of A, Benedenia seriolae and B, Heteraxine heterocerca. The histogram value for each 2 hr period is the sum of all larvae hatching during the same period on all days when larvae were collected, expressed as a percentage of the total number of larvae collected. Night-time is indicated by the black bar beneath the histogram and the official times of dawn and dusk are shown by the vertical arrows. 454 G. C. KEARN, K. OGAWA AND Y. MAENO during the night. On the second and third days of recording there was an initial period of extensive hatching during the period 04.00 to 08.00 hr, around the time of dawn. Hatching then continued but declined in frequency until about 12.00 hr when it again showed a significant increase until dusk. In H. heterocerca, few oncomiracidia were observed during daylight but there was a pro- nounced hatching peak on the first and second days of recording at dusk; hatching continued with decreasing frequency during the night. These contrasting patterns can be seen more clearly in Fig.2, in which the daily hatching patterns for each species are compared by sum- ming all the larvae of each species hatching during the same 2 hr period on each day of observation, and expressing this as a percentage of the total number of larvae collected. This also emphasizes the apparently bimodal nature of the diurnal hatching of B. seriolae with peaks at 06.00-08.00 hr and 14.00—-16.00 hr. DISCUSSION The hatching patterns of the monogeneans Benedenia seriolae and Heteraxine heterocerca, from the same host, Seriola quinqueradiata, have been determined under identical environmental conditions. Nylon netting with the entangled eggs of both parasites was incubated at 23°C with natural illumination, conditions which are not too different from those in the natural environment. If it is assumed that monogeneans hatch at a time of day or night when the host is most vulnerable to infection, then it would be expected that monoge- neans such as B. seriolae and H. heterocerca, which share the same host, would exhibit identical daily hatching patterns. Although there is some degree of convergence between the two parasites in the nature of their egg appendages and the fate of the eggs, their hatching patterns are surprisingly differ- ent (Fig. 2); in the skin parasite B. seriolae hatching continues throughout the hours of day- light, while the gill parasite H. heterocerca has a strong hatching peak at dusk and emerges with declining frequency during the night. A possible explanation for this paradox is that the two parasites have different invasion sites and that access to these different sites may be optimal at different times of day. There is evidence indicating that their invasion sites are different because post-oncomiracidia of B. seriolae were collected from the skin of experimentally infected Seriola aureovittata 24 hr after exposure to oncomiracidia (Kearn, Ogawa and Maeno, in preparation), while Ogawa and Egusa [2] found very young specimens of H. heterocerca without clamps on the gills, indicating that the oncomiracidia of the latter parasite are drawn in with the gill-ventilating current. There is evidence that other polyopisthocoty- lean gill parasites invade the gills directly, for example Rajonchocotyle emarginata and Plectano- cotyle gurnardi (see [3] and [4] respectively). Whittington [1] has made a comparison of hatching in two monogeneans, which, like B. seriolae and H. heterocerca, inhabit the skin and gills respec- tively of the same host. These are parasites of the dogfish, Scyliorhinus canicula, with the skin in- fected by the microbothriid monogenean Lepto- cotyle minor and the gills by the polyopisthocoty- lean Hexabothrium appendiculatum. He found a. striking convergence between the two parasites, with similarities between them in egg shapes and in their requirement for a chemical hatching stimulus from the host. Whittington [5] pointed out that similarities between the two parasites in their oncomiracidial behaviour indicate that the larvae of both parasites establish themselves on the skin, those of H. appendiculatum migrating later to the gills. Therefore, although these two parasites occupy different microhabitats when adult, their larvae may share the same invasion site, namely the skin. | The best time for invading the skin is not necessarily the best time for invading the gills and some behavioural features of yellowtail, as yet unknown, may create a situation where establish- ment of oncomiracidia on the skin is more likely to be successful during the day while access of oncomiracidia to the gills via the gill-ventilating current may be more favoured at dusk or early in the night. There may be a difference between the oncomiracidia of B. seriolae and H. heterocerca in their responses to light, gravity and water currents related to their essentially diurnal and nocturnal Hatching Patterns in Two Monogeneans 455 hatching patterns and different invasion sites. Apart from a comment by Hoshina [6] on the positive phototaxis of the larva of B. seriolae, the behaviour of these oncomiracidia is unknown and deserves further study. The arguments above assume that S. quinquera- diata is the natural host for B. seriolae and H. heterocerca. There is little doubt that S$. quin- queradiata is the natural host of H. heterocerca since the parasite was reported from this host by Goto [7], by Yamaguti [8] and, as Axine seriola, by Ishii [9]. However, there is no record of B. seriolae from S. quinqueradiata in the wild. Yamaguti [10] collected his original specimens of B. seriolae from the skin of wild S. aureovittata and no further host records were reported by Kamegai and Ichihara [11]. Thus, although the parasites share the same host, S. quinqueradiata, in fish farms, it is not certain that they do so in the wild and since the behaviour of even closely-related hosts may differ, the hatching patterns of B. seriolae may be attuned to a host other than S. quinqueradiata. ACKNOWLEDGMENTS The senior author would like to thank the Royal Society (U. K.) for the award of an Overseas Study Visit Grant which enabled him to work at the Seto Marine Biological Laboratory (Kyoto University) at Shirahama, Japan in 1990 and to visit the National Research Institute of Aquaculture at Nansei. We are most grateful to the Director of the Seto Laboratory, Professor E. Harada, and to his staff, especially Mr Y. Yusa and Dr. S. Yamato, for their hospitality and assistance. We are also grateful to the Director of the National Research Institute of Aquaculture, Dr. S. Sakaguchi, to the head of the Fish Pathology Division, Dr. Y. Inui, and to the chief of the Pathogen Section, Dr. M. Sorimachi, for their kindness in providing facilities and hospitality during the visit of G K and K O to their laboratory. REFERENCES 1 Whittington, I. D. (1987) J. mar. biol. Ass. UK., 67: 729-756. 2 Ogawa, K. and Egusa, S. (1981) Bull. Jpn. Soc. scient. Fish., 47: 1-7. 3. Whittington, I. D. and Kearn, G. C. (1986) J. mar. biol. Ass. UK., 66: 91-111. 4 Whittington, I. D. and Kearn, G. C. (1989) J. mar. biol. Ass. UK., 69: 609-624. 5 Whittington, I. D. (1987) J. mar. biol. Ass. UK.., 67: 773-784. 6 Hoshina, T. (1968) Bull. Off. int. Epizoot., 69: 1179-1191. 7 Goto, S. (1894) J. Coll. Sci. imp. Unv. Tokyo, 8: 1- DBs 8 Yamaguti, S. (1938) Jpn. J. Zool., 8: 15-74. 9 Ishii, N. (1936) Zool. Mag., Tokyo, 48: 781-790. 10 Yamaguti, S. (1934) Jpn. J. Zool., 5: 241-541. 11 Kamegai, S. and Ichihara, A. (1972) Res. Bull. Meguro Parasit. Mus., 6: 1-43. ee aa i ecrtihe pitts Pountieatd Sa, ae ce be gapsisiies 2.0 bigeye Ay eee oo i 4 ‘et CHa SH, eer pee Deramnnswie en a pnb re Fig. ie By one asa sara ‘pect ae a te Bh 2g ti Mts 5 [2 ate inith it a the fe ii My ee eet ke 08D) nt ies aS i : i . W = fio. ‘ae Pee 24 “ord ats et LS ros) | eae aa ; it TP ae eae 20g is ak f iol She e queer § ia Reni ty fae ae Foie’ * cite ie: 2 i ay cle 7 cibie *eests ee = —" ai ate Pete way ns ‘t signi i) A ceviodae- with peaks seers + savin: i rer sd (omy Tne oe Tee ee ae (Vry. ay a : matt oem eae As ee an, te ee Bs aes. nt AO ee MoD 1 (BBL) .2.thore < ah iN ! 1 NS ge » MOCET) i¢ igp iene Re Wi “aN \ hie if : e Tenia riety aa : , fol hn aa abta he TT. a! ST aes Sr ieeniie: ae uie@ ict CH2ctHy Crater Bd TE ar che Ura hs, ot es : i heel Hivizies ix a. TT ie Picadas a oa Y to 5 | es Toa = seit 4 ’ draut e i uy int ee : | Pe oe ae: yee I= y : : ~ PY , j “a 7 itt ; uy: rfid j * a va Pie ‘ j ne i ° : ee lo : ‘ ‘ tan 2 \ = \\ f j ZOOLOGICAL SCIENCE 9: 457-461 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan In vitro Spermatogenesis in Oryctes rhinoceros (Coleoptera, Scarabaeidae): The Role of Ecdysone and Juvenile Hormone MARIAMMA JACOB Department of Zoology, University of Kerala, Kariavattom, P. O. Trivandrum Kerala, India, PIN-695 581 ABSTRACT—The hormonal factors which regulate spermatogenesis in Oryctes rhinoceros were investigated in vitro. Spermatogonial multiplication in cultures occur- red in the presence of testis sheath which is the proposed site of ecdysteroid production. Meiosis of spermatocytes occurred when exogenous ecdysone (20-hydroxy ecdy- sone) was added to the culture. The secondary spermato- cytes thus developed underwent spermiogenesis only when active corpus allatum was introduced. INTRODUCTION The endocrine control of spermatogenesis in insects differs according to species. Ecdysone stimulates spermatogonial multiplication in Rhod- nius prolixus [1] and in Locusta migratoria [2], and promotes meiosis and spermiogenesis in most of the lepidopteran insects [3-7]. According to Shimizu et al. [6], Loeb and Woods [7] and Giebultowicz et al. [8] ecdysone is produced in testis. In Samia cynthia spermatogenesis continues in the culture medium in the presence of ecdysone and a macromolecular factor [9, 10]. In Diptera, ecdysone apparently has no role during sperma- togenesis [11, 12]. Juvenile hormone (JH) and its analogues have in general been found to inhibit spermatogenesis [5, 13, 14]. However, in some insects, topically applied JH analogues promote spermatogenesis [15, 16]. Investigations on the regulatory mechanisms of insect spermatogenesis have been mainly confined to Lepidoptera with relatively little attention being Accepted December 25, 1991 Received June 26, 1991 given to other groups. /n vivo studies of sperma- togenesis in the coleopteran insect Oryctes rhi- noceros, a major pest of the coconut palm, have revealed that even though secondary spermato- cytes appear in the pupal stage, spermiogenesis occurs only after adult emergence [17]. It has also been demonstrated that high doses of methoprene topically applied on 0 day-old pupa of this insect inhibit meiosis and spermiogenesis [18]. However, very little is known about the hormonal mecha- nisms involved in spermatogenesis, and hence the present in vitro studies were undertaken in order to elucidate the endocrine regulation of sperma- togenesis in this insect. MATERIALS AND METHODS The third instar larvae of Oryctes rhinoceros were reared in the laboratory on sterilized cow dung as reported earlier [17]. One hr-old male pupae were sterilized superficially in a mixture of mercuric chloride and ethanol (10 mg HgCl, dis- solved in 50% ethanol) for two minutes. They were washed repeatedly in sterile water and then swabbed in isopropyl alcohol. The testes were dissected under a Laminar flow hood and were placed in Rinaldini’s solution [19]. Grace’s insect medium and fetal calf serum (GIBCO) compound- ed in a ratio 100 ml:20 ml, supplemented with 0.2 ml antibiotic mixture (Penicillin-sodium - salt 100,000 units and Streptomycin sulphate 100 mg, each dissolved in 5 ml triple distilled water, along with Gentamycin 1 mg/ml) served as the basic medium (BM) for the present study. 20-Hydroxy 458 M. JAcosB ecdysone (SIGMA) was dissolved in Rinaldini’s solution 100 ug/ml, and from this DONO, SX 10-° and 1X10~° molar solutions (M) were prepared to study the dose-effect. The addition of prothoracic gland, with or without the brain, corpora cardiaca and corpora allata complex (BR, CC and CA, respectively) of male pupae also constituted a culture treatment as described in the experimental protocol. The excised testes were teased into smaller pieces after removing the remaining surrounding tissues. Fragments of testis with or without testis sheath were introduced into tissue culture vials with screw caps (12 cm length x1cm diameter, Borosil) containing 1 ml basic medium (BM). The culture thus prepared was incubated under aseptic conditions at 28°C for 16 to 21 days. The medium was exchanged after 16 days. The percentage of spermatocytes in meiotic and post-meiotic stages were determined in five (50 wl) samples of the medium, after 16 days of incubation, in each of the respective vial (n=8), corresponding to the respective dose. The data were Statistically analysed employing Single Factor ANOVAR. The experimental protocol is as follows: Group 1. Testis fragments incubated in BM without testis sheath. Testis fragments and BR CC CA of male pupae (n=5) incubated in BM without testis sheath. Testis fragments and 20-hydroxy ecdysone (1X10 -°M) incubated in BM without testis sheath. Testis fragments incubated in BM along with testis sheath. Testis fragments, testis sheath and BR CC CA of male pupae (n=5) incubated in BM. Testis fragments, testis sheath, prothoracic gland and BR CC CA of male pupae (n=5) incubated in BM. Testis fragments, testis sheath and 20-hydroxy ecdysone (2 x 107° M) in- cubated in BM. Testis fragments, testis sheath and 20-hydroxy ecdysone (5x 107° M) in- cubated in BM. Group 2. Group 3. Group 4. Group 5. Group 6. Group 7. Group 8. Group 9. ‘Testis fragments, testis sheath and 20-hydroxy ecdysone (1 x 10~° M) in- cubated in BM. Testis fragments, testis sheath and 20-hydroxy ecdysone (1 x 107° M) in- cubated for 10 days followed by addi- tion of CA of newly emerged adult males (n=5). Testis fragments, testis sheath and 20-hydroxy ecdysone (1 x 107° M) in- cubated for 10 days, followed by addition of BR CC CA of newly emerged adult males (n=S). Testis fragments, testis sheath and 20-hydroxy ecdysone (1 x 10~° M) in- cubated for 10 days, followed by addition of CA of one week-old adult males (n=5S). Testis fragments, testis sheath and 20-hydroxy ecdysone (1 10° M) in- ‘cubated for 10 days, followed by addition of CA of one month-old adult males (n=5). Group 10. Group 11. Group 12. Group 13. RESULTS Results of various culture treatments are pre- sented in Table 1. The testis fragments, when cultured without testis sheath, degenerated after 10 days, although a few spermatogonia had multi- plied in the initial stages (Groups 1, 2, 3). On the other hand, testis under culture containing the testis sheath (Group 4) showed abundant sperma- togonial multiplication (Fig. 1). However, no further development was observed even with maintenance of the cultures for up to 21 days. The brain, corpora cardiaca, and corpora allata com- plex of male pupa did not induce any development (Group 5). However, when prothoracic gland was introduced along with the brain complex (Group 6), meiotic stages and a few spermatids appeared. 20-Hydroxy ecdysone accelerated meiosis. In testis cultured with 20-hydroxy ecdysone, primary spermatocytes appeared on the second day and secondary spermatocytes were observed from the 10th day onwards (Groups 7, 8, 9). Furthermore, a dose-dependent increase in meiotic division was observed with the addition of 20-hydroxy ecdysone Spermatogenesis in Oryctes rhinoceros 459 TABLE 1. Results of testis culture of 1 hr-old pupae of O. rhinoceros Group a Additives to Basic Medium ceeaaa Results 1 8 Testis fragments only 16 A few spermatogonia; de- generation after 10 days 8 Z, + and BR CC CA of pupae (n=5S) 16 Z, Z, Z, 8 7. + and 20-hydroxy ecdysone (1x 10~° M) 16 . L L 8 Z; » and testis sheath 16 Mitoses of spermatogonia only 5 8 Z, » testis sheath and BR CC CA of # pupae 16 Z, Z, 7, (n=S) 6 8 Z, 7 testis sheath, prothoracic gland and BR 16 A few spermatids and sper- CC CA of & pupae (n=5) matozoa a 8 Z % testis sheath and 20-hydroxy ecdysone (2 16 Primary spermatocytes on x10~°M) 2" day, secondary sper- matocytes on 10" day cyst degenerated after 21 days 8 8 Z, ” testis sheath and 20-hydroxy ecdysone (5 16 2 2 2 x10~°M) 9 8 Z % testis sheath and 20-hydroxy ecdysone (1 16 2; 2, Z, x 107° M) 10 8 Testis fragments, testis sheath and 20-hydroxy 10+6 Secondary spermatocytes ecdysone (1X10 °M) incubated for 10 days fol- did not undergo any de- lowed by addition of CA of newly emerged # velopment adult (n=5S) fi 8 Testis fragments, testis sheath and 20-hydroxy 10+6 Secondary spermatocytes ecdysone (1X10 °M) incubated for 10 days fol- underwent spermiogenesis lowed by addition of BR CC CA of newly emerged J adult (n=5) 12 8 Testis fragments, testis sheath and 20-hydroxy 10+6 Z, Z, 2, ecdysone (1X10 °M) incubated for 10 days fol- lowed by addition of CA of one week old & adult (n=5) 13 8 Testis fragments, testis sheath and 20-hydroxy 10+6 Z, Z, Z, ecdysone (1X10 °M) incubated for 10 days fol- lowed by addition of CA of one month old & adult (n=S) Fig. 4). The secondary spermatocytes survived u a oe : E DISCUSSION to 21 days without undergoing spermiogenesis, but the cyst began to degenerate (Fig. 2). The corpora allata of newly emerged adult males were intro- duced into 10 days incubated culture (Group 10); this did not stimulate the development of secon- dary spermatocytes. However, when the same experimental model was repeated by adding BR CC CA of newly emerged adult males (Group 11), CA of one week-old adult males (Group 12) or CA of one month-old adult males (Group 13) respec- tively, the secondary spermatocytes underwent spermiogenesis. As a result, spermatids and Spermatozoa appeared in the culture vials (Fig. 3). In vivo studies of spermatogenesis in Oryctes rhinoceros {17] have shown that spermatogonia occupy the germinal zone of 1 hr-old pupa. Primary spermatocytes normally appear on the 10th hr and meiotic stages are seen in two day-old pupa. However, spermiogenesis is observed only after adult emergence (2-4 days). It has been observed in the present study that in the absence of testis sheath, spermatogonia degenerate. It may be considered that testis sheath secretes factor(s) which sustain spermatogonia in the culture and promote their multiplication. Loeb and Woods [7] and Giebultowicz et al. [8] suggested that testis 460 M. JAcos Fic. 1. Spermatogonial multiplication after culturing 1 hr-old pupal testis fragments in BM and testis sheath x 600. Fic. 2. Secondary spermatocytes in a degenerating condition developed after culturing testis fragments in 20- hydroxy ecdysone and kept beyond 21 days without corpora allata 600. Fic. 3. Sperm bundles and released spermatozoa developed after culturing 10 days incubated culture in 20-hydroxy % of meiosis ecdysone, with active corpora allata of 1 month old-adult < 600. Abbreviations for Fics. 1-3. PSG, primary spermatogonia; SB, sperm bundles; SSC, secondary spermatocytes; SSG, secondary spermatogonia; SZ, spermatozoa. 80 60 Fic. 4. Showing the effect of three doses of 20-hydroxy ie ecdysone on the meiotic division of spermatocytes of O. rhinoceros. Analysis of variance has shown the dose response to be highly significant (P<0.05). 0) represents SD of meiosis. 2x 10 oy 5 x 10°eM iL 3 10) =I Spermatogenesis in Oryctes rhinoceros sheath produces ecdysteroids. The present study has revealed that exogenous ecdysone promotes meiosis. In Rhodnius [1] and in Locusta [2] spermatogonia multiply in the presence of exogenous ecdysone, whereas in Lepi- doptera, meiosis and spermiogenesis are promoted by ecdysone [3-7]. In Oryctes rhinoceros, the percentage of meiotic and post-meiotic spermato- cytes increased with the increase in the dose of 20-hydroxy ecdysone. In the present culture treatments, spermiogenesis occurred only after the introduction of active corpora allata into the medium. It appears that the corpora allata of newly emerged adult males have not attained activity, but brain and corpora cardiaca seem to have induced their activity (Groups 10, 11). In Oryctes rhinoceros, the corpus allatum has a stimulatory effect on spermiogenesis. Promotion by juvenile hormone (JH) of any processes related to spermatogenesis as reported here for Oryctes rhinoceros, seems unusual. JH inhibits the elonga- tion of spermatids in the testis culture of Bombyx mori [5]. JH and its analogues inhibit sperma- togenesis in the several other insects [13, 14]. However, JH analogues applied topically have been shown to accelerate spermatogenesis in Draeculacephala crassicornis [15] and in Dysdercus cingulatus [16]. On the whole it appears that in Oryctes rhinoceros, ecdysone as well as juvenile hormone are equally important during sperma- togenesis. In the normal insect, spermiogenesis occurs after adult emergence because the corpora allata become active at that time. ACKNOWLEDGMENTS The author is grateful to Prof. V. K. K. Prabhu, University of Kerala, for his constant encouragement, to CSIR., New Delhi, for the financial assistance and to National Institute of Virology, Poona, India, for the training undergone in tissue culture. 13 14 15 16 17 18 19 461 REFERENCES Dumser, J. B. and Davey, K. G. (1975) Can. J. Zool., 53: 1682-1689. Dumser, J. B. (1980a) Int. J. Invertebr. Reprod., 2: 165-174. Takeda, N. (1972) J. Insect Physiol., 18: 571-580. Fukushima, T. and Yagi, S. (1975) Appl. Entomol. Zool., 10: 220-225. Yagi, S. and Fukushima, T. (1975) Appl. Entomol. Zool., 10: 77-83. Shimizu, T. Moriabayashi, A. and Augi, N. (1985) Appl. Entomol. Zool., 20: 56-61. Loeb, M. J. and Woods, C. W. (1989) Arch. Insect Biochem. Physiol., 10: 83-92. Giebultowicz, J. M., Loeb, M. J. and Borkovec, A. B. (1987) Int. J. Invertebr. Reprod. Develop., 11: 211-226. Kambysellis, M. and Williams, C. M. (1971a) Biol. Bull., 141: 527-540. Kambysellis, M. and Williams, C. M. (1971b) Biol. Bull., 141: 541-552. Kuroda, Y. (1974) J. Insect. Physiol., 20: 637-640. Leloup, A. M. (1976) In “Invertebrate tissue culture”. Ed. by E. Kurstak and K. Maramorosch, Academic Press, London, pp. 179-183. Dumser, J. B. (1980b) Annu. Rev. Entomol. 25: 341-369. Koeppen Jee ken nuchs, IME. Chenu ile luntessle- M., Kovalick, G. E., and Briers, T. (1985) In “Comprehensive Insect Physiology, Biochemistry and Pharmacology”. Ed. by G. A. Kerkut and L. I. Gilbert, Pergamon Press, New York, Vol. 8, pp. 165-203. Reissig, W. H. and Kamm, J. A. (1975) Anal. Entomol. Soc. Amer., 68: 353-354. Ambika, B. and Prabhu, V. K. K. (1978) Entomon, 3: 165-175. Jacob, M. (1989a) Proc. Indian Acad. Sci., 98: 233- 242. Jacob, M. (1989b) Current Science, 58: 469-471. Rinaldini, L. M. (1954) Nature, 173: 1134-1135. ro ZOOLOGICAL SCIENCE 9: 463-467 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Identification of Intracellular Localization of Laminin in the Rat Anterior Pituitary TOSHITERU KikuUTA and HipEo Namiki! Department of Biology, School of Education, Waseda University, Shinjuku-ku, Tokyo 169-50, Japan ABSTRACT—The intracellular localization of laminin of endocrine cells in the rat anterior pituitary was examined by biochemical methods and _ electron- microscopical immunocytochemistry. The secretory gra- nules of the anterior pituitary cells were separated. And the granules were immunostained by the post-embedding method using antiserum against laminin. Laminin was detected on the surface of the membrane of separated granules. On the other hand, the granules were immuno- blotted using antiserum and monoclonal antibody against laminin. The antiserum reactive with both A-chain and B-chains of the antigen (mouse EHS sarcoma laminin) cross-reacted with only the B-chain(s) of the secretory granules, but the monoclonal antibody showed no cross-reactivity. These results suggest that laminin in the anterior pituitary cells may exist as B-chain singly in the secretory granules and it may be different from mouse EHS sarcoma laminin in structure. INTRODUCTION Laminin [1, 2], one of the major components of the basement lamina, has been immunocytochemi- cally observed in various organs, and it has also been proved to play a role in cell attachment, migration, proliferation and differentiation. In the pituitary gland, Tougard et al. have detected immunocytochemically that laminin exists on the basement lamina and the parenchymal cells, espe- cially gonadotrophs of the anterior pituitary gland [3]. Thereafter, similar evidence has been re- ported by others [4-9], but obtained results including the identification of the positive cells and Accepted January 30, 1992 Received July 6, 1991 * To whom all correspondence should be addressed. the intracellular site of laminin are still debatable. Besides, in these reports examinations have been conducted only by immunocytochemical methods for the identification of laminin, and there is no information about its intracellular localization using other methods. It is important to clarify the intracellular localization of laminin in the immuno- reactive pituitary cells (laminin positive cells). In the present study, we separated the secretory granules from the anterior pituitary of rats and identified the localization of laminin in the secre- tory granules by both the method of immuno- blotting and that of electron-microscopical im- munocytochemistry. MATERIALS AND METHODS Animals Wistar-Imamichi male rats (60 day-old) were obtained from the Imamichi Institute for Animal Reproduction. One hundred rats for granule separation and immunoblotting, and six for light- microscopical immunocytochemistry were used. Antibodies The following antibodies were used for detecting laminin; antiserum against mouse EHS sarcoma laminin (Advance, Tokyo, x 1000-32000 in dilu- tion) and monoclonal antibody against human placental laminin (Iwaki Glass, Tokyo, x 1000). Antiserum against rat LH-8 (NIDDK, x 8000) was used for immunocytochemical identification of LH cells. Antiserum against bovine type IV collagen 464 T. KikuTA AND H. NAmIKI (Advance, 4000) was used for checking the contamination of basement lamina in the granule solution. The ABC kit (Vector Lab.) was used for the further process of immunocytochemistry and immunoblotting, after using the first antibodies. And the gold-colloid (15 nm diameter) conjugated antibody (Ig-G) against rabbit Ig-G (E. Y. Lab.) was used as the second antibody for electron- microscopical immunocytochemistry. The specific- ity of the antisera for laminin and type IV collagen was reported by Aihara et al. [10]. The antisera for mouse laminin and bovine type IV collagen suf- ficiently cross-react with rat laminin or type IV collagen respectively. And the adsorption-test of the antisera for mouse laminin and rat LH-f had done by Kusaka et al. [9]. Enzyme-linked immuno- sorbent assay (ELISA) [11, 12] showed that the monoclonal antibody strongly reacted with human placental laminin, and it reacted also with both subunits of mouse EHS sarcoma laminin but only slightly. | Immunocytochemistry After decapitation, the pituitary glands were removed and fixed in formolsublimate solution and embedded in Paraplast. Three mum thick serial sections were then prepared, deparaffinized, and treated with pepsin [13]. Two serial sections were stained by immunoperoxidase method (ABC method) [14] with the antisera either against laminin (X32,000) or LH-f respectively. The stained sections were counterstained with Hema- toxylin solution. Separation A modified method reported by Costoff et al. [15] was used for separation of the secretory granules from the rat anterior pituitary. After decapitation, the anterior glands were removed and homogenized. The nuclei were excluded by a centrifuge and the supernatant was filtrated, loaded on 5-50% Ficoll 400 (Pharmacia), 0.25 M saccharose, 0.5 mM EDTA (pH 7.2), and centri- fuged (100,000g, 2hr.). After the centrifuga- tion, all the zones containing the hormone gra- nules were collected and centrifuged (100,000 x g, lhr.). The precipitate was then suspended with PBS and sonicated. The solution thus obtained was used as the secretory granule solution. No practical contamination of basement lamine was observed in the granule solution since the solution immobilized on the nitrocellulose membrane did not react immunocytochemically with the anti- serum against type IV collagen, a major compo- nent of basement lamina. Electron microscopy The purity and immunoreactivity of the sepa- rated secretory granules were checked by the post-embedding method [16, 17]. The granules were fixed in picric acid-paraformaldehyde solu- tion [18] and osmium tetroxide solution [19], and embedded in Quetol 651 (Nisshin EM, Tokyo) by a modified method reported by Kushida [20]. The thin sections were reacted with the antiserum against laminin (x 1000) or normal rabbit serum (NRS), followed by the gold-colloid conjugated 2nd antibody. They were then refixed in the same fixative. No electron staining was employed. Again, no practical contamination of other cellular components was visually observed (Fig. 2). Immunoblotting The granule solution was separated by SDS- PAGE [21] (gel concentration: 6%) and silver- stained (Silver Stain KANTO, Kanto Chemical, Tokyo) or immunoblotted [22], by the use of ABC method. Antiserum against mouse laminin (xX 4000) and monoclonal antibody against human laminin were used as the first antibody. And the silver-stained gel was analyzed by a chromatoscan- ner (Shimadzu CS-9000). RESULTS The resuit of light-microscopical immuno- cytochemistry is shown in Figure 1. The antiserum against EHS laminin reacted with the capillary basement lamina and some of parenchymal cells. The laminin positive cells (Fig. la) were corre- sponded with LH cells (Fig. 1b). Electron-microscopical observation revealed that the separated granule fraction contained practically only the secretory granules (Fig. 2), and the antiserum against EHS laminin more or less reacted with the granules despite of their densities Laminin in the Pituitary 465 vO Fic. 1. The adjacent sections of the anterior pituitary were immunostained by antiserum against mouse lamin in (a) or rat LH- (b). arrowhead: laminin positive cell, arrow: capillary, Bar: 50 um. Fic. 2. Electron-microscopical immunocytochemistry of the secretory granule solution. Antiserum against mouse laminin reacted with the granules. a: antiserum against laminin (low density), b: antiserum against laminin (high density), c: NRS, Bar: 200 nm. a b ecd (Fig. 2a, b). The results of silver-staining and immuno- blotting using EHS laminin antiserum are shown in Figure 3. The B-chains were not separated in this electrophoresis condition; therefore two bands of mouse laminin (the A-chain and B-chains bands) were stained (Fig. 3a, c). The lane of electro- phoresed granule solution was_ silver-stained oe almost entirely (Fig. 3b), in which immunoblotting detected only B-chains but not A-chain (Fig. 3d). In addition, some bands of lower molecular weight were stained. Fic. 3. Mouse laminin (a, c: 0.18 “g) and the secretory granules (b, d: 85.6 ug) were silver-stained (a, b) and immunostained by antiserum against mouse EHS sarcoma laminin (c, d). A: A-chain, B: B-chains. , 466 T. KIkKUTA AND H. NAmIkI Fic. 4. Mouse laminin (a: 1.5 wg) and the secretory granules (b: 85.6 4g) were immunostained by monoclonal antibody against human placental lami- nin. A: A-chain, B: B-chains. The monoclonal antibody was able to detect both A-chains and B-chains of mouse EHS laminin (Fig. 4a), but neither A- nor B-chains of the secretory granules were detected (Fig. 4b). Again, some bands with lower molecular weight were stained. DISCUSSION The granule preparation seemed to include almost all kinds judging from their specific grav- ities and sizes [15]. Almost all the granules were laminin positive as far as examined, although stainabilities varied from granule to granule. This result agreed with the demonstration of Vila- Porcile et al. [5] in which all the secretory cells embedded in an electron-microscopical thin sec- tion, were entirely stained. Light-microscopical observation, however, showed that only LH cells were laminin positive that accorded with the result obtained by Tougard et al. [3]. The reason of the discrepancy may be explained as follows; LH granules contains more laminin than others, thus the light-microscopical method is not sensitive enough for the detection of intracellular laminin other than LH cells, nevertheless electron- microscopical method can detect such the slight amount of the antigen because of intracellular localization. On the molecular study by immunoblotting, a polyclonal antiserum against mouse EHS sarcoma laminin reacted with the electrophoretically B- chain equivalent portion of laminin and several smaller molecules of the secretory granules but not with the A-chain, although it reacted with both the A-chain and the B-chains of the antigen. On the other hand, no laminin equivalent bands were stained with a monoclonal antibody against the human laminin that could stain both chains of EHS laminin. Many smaller molecules were again stained with the antibody. A-chain equivalent molecules were not detected at any time in the present experiment. These results suggest that there exist laminin-like mol- ecules different in structure from the laminin of the basement lamina in the secretory granules of the rat anterior pituitary. We now tentatively assume that these consist of a single or double subunit of B-chain-like molecules. Many immunopositive molecules smaller than B-chain molecules were detected by the two antibodies used presently. It is however still unknown if they are fragments of the molecule or they are physiologically synthesized in the pituitary cells. Recently, Hunter et al. reported that a type of laminin (s-laminin) different from that of the basement lamina exists in normal rat tissues (e.g. synaptic cleft, glomerular) [23]. The secretory granule laminin (g-laminin) presently reported may also have been different from that of the basement lamina. ACKNOWLEDGMENTS This research was supported by a grant from Waseda University to H. Namiki. We thank Dr. B. de Chene of Waseda University of critical reading of the manuscript. The antiserum against rat LH-f was received from the National Hormone and Pituitary Program (NHPP, Univ. of Maryland Sch. of Med.). We also thank to the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). 10 Laminin in the Pituitary REFERENCES Timpl, R., Rohde, H., Robey, P. G., Rennard, S. I., Foidart, J. M. and Martin, G. R. (1979) J. Biol. Chem., 254: 9933-9937. Beck, K., Hunter, I. and Engel, J. (1990) FASEB J., 4: 148-160. Tougard, C., Louvard, D., Picart, R. and Tixier- Vidal, A. (1985) In vitro Cell Develop. Biol., 21: 57-61. Holck, S., Albrechtsen, R. and Wewer, U. M. (1987) Lab. Invest., 56: 481-488. Vila-Porcile, E., Picart, R., Tixier-Vidal, A. and Tougard, C. (1987) J. Histochem. Cytochem., 35: 287-299. Vila-Porcile, E., Picart, R., Olivier, L., Tixier- _ Vidal, A. and Tougard, C. (1988) Cell Tissue Res., 254: 617-627. Leardkamolkarn, V., Heck, L. W. and Abraham- son, D. R. (1989) Cell Tissue Res., 257: 587-596. Nunez, E. A., Pomeranz, H. D., Gershon, M. D. and Payette, R. F. (1990) Anat. Rec., 226: 471-480. Kusaka, S., Yamashita, S., Yamaguchi, S. and Kusunoki, S. (1987) Zool. Sci., 4: 991 (Proceeding). Aihara, M., Izawa, H., Kumagai, N., Ishida, H., Ogino, Y., Kimura, Y., Yashiro, T. and Suzuki, T. (1988) J. Jpn. Plastic Reconstruct. Surgery, 8: 410- 418. 11 7 13 14 iS 16 17 18 19 20 21 22 23 467 Engvall, E. and Perlmann, P. Im- munochem., 8: 871-874. Miles, L. E. M. and Hales, C. N. (1968) Nature, 219: 186-189. Burns, J., Dixon, A. J. and Woods, J. C. (1980) Histochem., 67: 73-80. Hsu, S-M., Raine, L. and Fanger, H. (1981) J. Histochem. Cytochem., 29: 577-580. Costoff, A. and McShan, W. H. (1969) J. Cell Biol., 43: 564-574. Faulk, W. P. and Taylor, G. M. (1971) munochem., 8: 1081-1083. Roth, J., Ravazzola, M., Bendayan, M. and Orci, L. (1981) Endocrinol., 108: 247-253. Zamboni, L. and De Marrino, C. (1967) J. Cell Biol., 35: 148A (Abstract) Millonig, G. (1962) In “Electron Microscopy. Vol. 2 (Fifth international congress for electron micros- copy)” Ed. by SS Breese Jr., Academic Press, New York, pp. 8. Kushida, H. (1974) J. Electron Microscopy, 23: 197. Laemmli, U. K. (1970) Nature, 227: 680-685. Burnette, W. N. (1981) Anal. Biochem., 112: 195- 203. Hunter, D. D., Shah, V., Merlie, J. P. and Sanes, J. R. (1989) Nature, 338: 229-234. (1971) Im- pak te si fines vi i ay he A ts ry na spite oer: Lae ort nha 3 ZOOLOGICAL SCIENCE 9: 469-473 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Vocal Repertoire of the Japanese Treefrog, Rhacophorus arboreus (Anura: Rhacophoridae) Entir Kasuya’, Toru Kumaki and TAKAYOSHI SAITO” Laboratory of Biology, Faculty of Education, Niigata University, 2-8050 Ikarasi, Niigata 950-21, Japan ABSTRACT— The vocal repertoire of males and females of the treefrog, Rhacophorus arboreus was described with audiospectrogram and oscillogram. The repertoire of males included the advertisement call, the courtship call, four types of aggressive calls and compound calls. The repertoire of females included three types of release calls. INTRODUCTION Acoustic communication is one of the well documented features in the mating of anuran amphibians [1]. Male anurans use different kinds of calls in response to particular situations [1-3]. Description of the vocal repertoire is one of the prerequisites to study the mating behavior and acoustic communication in anurans. The vocal repertoire in the foam-making rhacophorid frogs has not been described. The advertisement call has been described in several species [4, 5] and Coe [6] reported the change in Chiromantis male calls occurred in response to the approach of females. In the present paper, we describe the physical properties of vocalizations, their functions and the situations where they were emitted in the Japanese treefrog, Rhacophorus arboreus. Accepted November 14, 1991 Received September 20, 1991 ‘ To whom offprint requests should be sent. * Alphabetical order. MATERIALS AND METHODS The observations were made from May to July, in 1984 and 1985 at the Hyoutan pond in Iwamuro, Niigata, Japan (altitude about 180m) [7]. Frogs were individually marked with colored waist bands. Observations were made with a 6V battery head lamp that appeared not to disturb the behavior of frogs. Vocal activities of frogs were recorded with SONY TC-D5M tape-recorder and SONY ECM Z-300 microphone. Metal cassette tapes were used as the recording media. All the recordings were made when the ambient temperature was from 18 to 23°C. Soundspectrogram analysis was _per- formed on MacReacorder Sound analyzing system and Kay 7800 Digital Sonagraph. RESULTS AND DISCUSSION We were able to distinguish the following nine types of vocalizations. They were typical ones and intermediates between call types were not de- scribed in this study except the variation in the advertisement call in the section of the compound call. Of the nine types, six were produced by males and three by females. We followed the classifica- tion of vocalizations by Wells (cited in [3]) except for compound calls. In this classification, vocaliza- tions were classifed by their function as other terminologies of anuran vocalizations. In statisti- cal analysis of calls, the sample size was the number of individuals. We presented mean+SD 470 E. Kasuya, T. KUMAKI AND T. Saito for physical features of calls. The statistical test used was Mann-Whitney U-test if not otherwise mentioned. Figure 1 shows audiospectrograms (sonograms) and oscillograms of these calls. Advertisement call (A, hereafter) This call is a rattle that consists of 2 to 6 notes. The harmonics were not clear. The dominant frequency ranged from 900 to 1900 Hz. This call sometimes has a preceding and/or a following weak notes (Fig. 2). The spectral feature of this weak additional note was different from notes of A. The spectrum of this additional note was near the pure tone and had a sharp peak at about 900 Hz Courtship call (Ac, hereafter). This call was emitted by males when females were near (within about 50 cm from males). The aa F1 temporal structure of Ac was similar to A except a shorter interval between notes. The spectral feature of individual notes was similar to A. The interval between notes in a call was significantly shorter in Ac (39.6+4.2 ms, n=10) than in A . (Of. 5226.4 ms, n=26) (Z—4.6-r <0. 00 Aggressive calls multi-note aggressive call (B, hereafter): This call is also a rattle consisting of several notes similar to A. The dominant frequency was similar to A. But, B had a flatter spectrum than A. As shown in Figure 1, the temporal feature of B was similar to A except that a note of B often had two pulses. The duration of a whole note of this call (29.10+5.9ms, n=14) was significantly longer than A (z=3.52, P<0.001) though the duration of the major pulse (20.72+2.6ms, n=14) was not significantly longer than that of A (21.2+7.0 ms, n F3 Vocal Repertoire in Treefrog 471 Ac Eee KHe A __.._ A Narrow Band K _Narrow Band FO Narrow Band Fic. 1. Oscillographs (p. 470) and sonograms (p. 471) of vocalizations. horizontal bar shows 0.1s. For sonograms, the grids of 2 KHz interval were also shown. Oscillograms and sonograms were made by MacRecorder and Sound Edit. Sonogram setting was transform size=64 (frequency resolution was 344 Hz) if not mentioned, and transform size=512 (frequency resolution was 43 Hz) if shown as “narrow band”. A: advertiserment call, B: multi-note aggressive call, Ac: courtship call, C: one note call (encounter call), G and K: aggressive call, Fl, F2 and F3: female release call. Fic. 2. Weak note which preceded and/or followed advertisement call. horizontal bar shows 0.1s. Left: additional note preceding an advertisement call, and right: following an advertisement call. Oscillograms were made by MacRecorder and Sound Edit. = 472 E. KasuyA, T. KUMAKI AND T. SAITO =26) (z=0.60, P>0.05). This call was usually emitted when a calling male was in contact physically with another male or grasped by another male during oviposition. This call probably has the function similar to release calls of other anurans as well. single note call (C, hereafter): This call con- sisted of a single note of short duration. The sound intensity of this call was lower than the advertise- ment call. This call was emitted by males when another male was nearby. Therefore, this is probably an encounter call. The duration of note of C (6.8+0.9 ms, n=15) is significantly shorter than A (z=5.28, P<0.001). Other aggressive calls (G and K): There were 2 types of other aggressive calls, which were given in active choruses and aggressive male-male interac- tions. G_ This call was a longer trill-like note. A note had pulses with short intervals. Individual pulses were not audible. The number of pulses in a note ranged from 5 to 18. The interval between pulses (the interval between the points of maximum amplitude, 22.7+6.3 ms, n=18) was significantly shorter than the note interval in A (Mann-Whitney U-test, z=5.59, P<0.001) and Ac. (Mann- Whitney U-test, U=4, P<0.001). K_ This is a one note call. Though C is also a one note call, the duration of a note of K (30.0+ 5.2ms, n=18) was significantly longer than C (Mann-Whitney U-test, U=0, P<0.001) and A (Mann-Whitney U-test, z=4.23, P<0.001). This call was not emitted alone but accompanied with A or G. This call often has 2 different parts in one note (see, oscillogram of Fig. 1). In this case, the first part with the larger amplitude had the spectral feature similar to A, and the second part with the small amplitude had the clear harmonics (fun- damental frequency was about 700Hz). Female release calls (F) These calls had different temporal and spectral features from the vocalizations by males. These calls included three types of vocalizations (Fig. 1). The first one (Fl) had harmonics with a weak frequency modulation (the freqeuncy was falling). The fundamental frequency was about 1100 Hz. The second one (F2) had two parts in a note. The first part consisted of two kinds of harmonics with the fundamental frequencies of about 300 and 500 Hz. The second part had clear harmonics with a freqeuncy modulation where the fundamental fre- quency was rising from about 900 to 1200 Hz. The third type (F3) had clear harmonics with frequency modulation where the fundamental frequency was falling from about 1300 Hz. These were emitted by females when grasped by silent males. All the three sub-types (Fl, F2 and F3) were sometimes given during a single event of grasping by a silent male. The occurrence of the calls by females were much less than those of males because grasping by a silent male was rare except in the event of a high density of females in and around a pond. Compound calls The compound call is a complex of advertise- ment and aggressive calls [1] (A, G and K in R. arboreus). Compound calls were A+G, A+K or A+G-+K (Fig. 3). These were emitted often in the escalated choruses. A bout of chorus was usually initiated by the advertisement calls of one A+G Fic. 3. Compound calls. horizontal bar shows 0.1 s. Oscillograms were made by MacRecorder and Sound Edit. Vocal Repertoire in Treefrog 473 or a few males and was escalated into a frequent exchange of compound calls by a large number of males. In compound calls, a note of A sometimes had two pulses (like B) and intermediates between A and G or K were observed (Fig. 3). The vocal repertoire of R. arboreus males included the advertisement call, the courtship call, the call used in encounter with other males (C) and calls used in escalated chorus (G and K). In addition to these calls that were reported in other anurans, the call usually given during oviposition (B) was observed in R. arboreus. Because R. arboreus is a foam-nesting species and males not in amplexus try to release sperm into the foam during oviposition (sneaky joining males), male-male competition during oviposition seems to be intense [7]. The call-type B would have aggressive mes- sages to other males around a female during Oviposition. Probably this type of call is unique one for the foam-making species with joing males. The difference among three sub-types of the female release call in the function and context is not clear. The observations that all of the three sub-types were emitted when a silent male grasped a female suggest that there is no difference in their function or context. In the vocal repertoire of R. arboreus, only A was described by Kuramoto [4] and Maeda and Matsui [5]. Kuramoto reported the dominant frequency of the advertisement calls ranged from 1800 to 1900 Hz, whereas Maeda and Matsui noted this was about 800 Hz. The present study reports frequency values higher than those of Maeda and Matsui and lower than those of Kuramoto. Both Kuramoto [4] and Maeda and Matsui [5] described “after call” that often followed the advertisement calls. Kuramoto showed this “after call” lacked the component of a higher frequency in the spectrum. These “after-calls” were not observed in the present study. These differences suggest the geographical variation in the spectral and temporal features of the advertisement calls in R. arboreus. ACKNOWLEDGMENTS We are grateful to K. Kinefuchi for his advice and K. Fukuyama, M. Hirota, H. Shigehara and anonymous reviewers for useful comments. REFERENCES 1 Wells, K. D. (1977) In “The Reproductive Biology of Amphibians”. Ed. by D. H. Taylor and S. I. Guttman, Plenum, pp. 233-262. NT kcu Neus (1983) ee linens Mate choices ee dasby 9B. Bateson, Cambridge University Press, Cambridge, pp. 181-210. 3. Rand, A. S. (1988) In “The Evolution of the Amphibian Auditory System”. Ed. by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington and W. Walkowiak, Wiley and sons, New York, pp. 415-431. 4 Kuramoto, M. (1974) Bull. Fukuoka Univ. Educa- tion, 23(3): 67-77. 5 Maeda, N. and M. Matsui (1989) Frogs and toads of Japan. Bunichi Sogo Shuppann, Tokyo. 6 Coe, M. J. (1974) J. Zool., Lond. 172: 13-34. 7 Kasuya, E., H. Shigehara and M. Hirota (1987) Zool. Sci., 4: 693-697. 7 | i Sik: iain Hy os ee Beet $ ee her oe yee mas | 5 A iis Ei! WE etait by o I oe ° p “0 a. Bir viene Ovid eR sexing aoa’ ve he iigwemcyir age rad Iau re 1k RIOT = Me Br hare re ao Be eS As =e pas us shisied “ae Erte i. Vata, pene re A ig J ie peers — Mito Ch arises 4 i a Nae m - ' Seite enn 3 a af 5 a cea, o SaITE ChOnaes ance) ein RAR ie te eam Siig ov vee ati ehiae ny ee = ate ns, F ME ud ak hg ; San Lroloney Sato eee Abts ce! A. sia i a, : Dy s E : : a € aa oT m4 ee oe 2 f Pe ve 7 tl et oh { ¥ a Ys Wee Ns Seasianeae ; . fiat puluit york stulgpkeseeeale, etter eel ade a fees = ida Ra et CLA Abeuate Se Lith BR puegbsivuny 2 eS Ramer PY Ae PRORAIERD 78S 3 y Pea bar eas ag Ae eee me re) Pe Ss pp tae totsbh : : ie Detter AS IF TRS 3 Sl eae ‘ait iri toe . ah to: Bonulee: YT od ool AU Eo 1s a: miaetl ists rieaelsa pent # : begac eet os SH eee Srie Gak ui Ray = ~ ee. is Ts = ‘ ; Ls re CJ6S gh: - 9 Te “sd fhe ats 4 ; hey i cay f A. wi tr bd ~'S P S rath . : ¥ , eas Seer Se? vate: © 7 By) : at i ; £ ‘ 1 % 4 ries wiaoint if : a iy F ie . ; i ; \ Sieik: By: mahee Wiis ip re Pea , 8 ve i Lee ANNOUNCEMENT THE MELANOTROPIC PEPTIDES A CONFERENCE OF THE NEW YORK ACADEMY OF SCIENCES A Conference of the New York Academy of Sciences on “The Melanotropic Peptides” will be held in Rouen, France, from September 6 to 9, 1992. The conference will include 30 invited lectures and contributed posters. Each poster presentation will entitle the authors to publish a short article in the volume of the Annals of the New York Academy of Sciences which will result from the Conference. The deadline for submission of abstracts is June 22, 1992) For further information, please contact: Dr Hubert VAUDRY European Institute for Peptide Research Laboratory of Molecular Endocrinology CNRS URA 650 University of Rouen POB 118 76134 Mont-Saint-Aignan France D evel O im ent Published Bimonthly by the Japanese Society of Developmental Biologists Distributed by Business Center for Academic Growth & Differentiation Societies Japan, Academic Press, Inc. Papers in Vol. 34, No. 2. (April 1992) 14. REVIEW: P. Nick and M. Furuya: Induction and Fixation of Polarity—Early Steps in Plant Morphogenesis 15. R. Gualtieri, C. Campanella and P. Andreuccetti: Cytochemical Ca** Distribution in Activated Discoglossus pictus Eggs: A Gradient in the Predetermined Site of Fertilization 16. M. H. Fuhrman, J. P. Suhan, and C. A. Ettensohn: Developmental Expression of Echinonectin, an Endogenous Lectin of the Sea Urchin Embryo 17. T. Harumi, M. Kurita and N. Suzuki: Purification and Characterization of Sperm Creatine Kinase and Guanylate Cyclase of the Sea Urchin Hemicentrotus pulcherrimus 18. T. Harumi, K. Hoshino and N. Suzuki: Effects of Sperm-Activating Peptide I on Hemicentro- tus pulcherrimus Spermatozoa in High Potassium Sea Water 19. T. Iwamatsu, S. Y. Takahashi, M. Oh-ishi, T. Yokochi and H. Maeda: Changes in Electrophoretic Patterns of Oocyte Proteins during Oocyte Maturation in Oryzias latipes 20. S. Aronsson and A. Enemar: On the Development of the Eminentia Mediana of the Hypophysis in Rana temporaria, Studied in Normal, Hypophysectomized, and Thyroidecto- mized Tadpoles . 3 21. L. L. Lindsay and W. Clark Jr.: Protease-Induced Formation of the Sperm Acrosomal Filament 22. N. Osumi-Yamashita, S. Iseki, S. Noji, T. Nohno, E. Koyama, S. Taniguchi, H. Doi and K. Eto: Retinoic acid Treatment Induces the Ectopic Expression of Retinoic Acid Receptor £ Gene and Excessive Cell Death in the Embryonic Mouse Face 23. Y. Kamata, S. Furuya, K. Takei-Mikami, A. Fujiwara and I. Yasumasu: Identification of GTP-Binding Proteins by ADP-Ribosylation in the Presence of Cholera Toxin, Pertussis Toxin and Botulinum Toxin D in Plasma Membrane Isolated from Eggs and Embryos of Sea Urchin 24. M. Watanabe, K. Itoh, K. Abe, T. Akizawa, K. Ikenishi and M. Furusawa: Immuno- Localization of DEAD Family Proteins in Germ Line Cells of Xenopus Embryos 25. K. Hashimoto, M. Noguchi and N. Nakatsuji: Mouse Offspring Derived from Fetal Ovaries or Reaggregates Which were Cultured and Transplanted into Adult Females 26. S. Yamada, M. Ikeda, K. Eto: Differential Expression of c-myc and N-myc during Oral Organogenesis of the Mouse Embryo Development, Growth and Differentiation (ISSN 0012-1592) is published bimonthly by The Japanese Society of Developmental Biologists, Department of Developmental Biology, 1990: Volume 32. Annual subscription for Vol. 33, 1991: U.S.$ 162,00, U.S. and Canada: U.S. $ 178,00, 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 MOOS ICES Ae The Ultimate Name tn Micromanipulation Ease of operation, and the most advanced Parente § Can be combined with New WR & MX series Model Mx -2 3-D Micromanipulator Model MX-1 3-D Micromanipulator Model WR-87 One Dimensional Aqua Purificate Micromanipulator Model SR-6 Model PP -83 Stereotaxic Instrument for Rat Glass Microelectrode Puller * * & Send enquiries for request for MODIFICATIONS or IMPROVEMENTS * * * Physiological, Pharmacological, Zoological & Neurosciences Research Equipments NARISHIGE SCIENTIFIC INSTRUMENT LAB. \Je25N7, 9-28 KASUYA 4-CHOME SETAGAYA-KU, TOKYO 157, JAPAN PHONE (INT-L)81-3-3308-8233, FAX (INT-L)81-3-3308-2005 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 Yy Yj Ly Y0 Design Corrosion Resistance HIGH SPEED REFRIGERATED MICRO CENTRIFUGE move. MR-150 TOMY CORPORATION SOLE AGENT 1001 SOLEIL NARIMASU BLDG., 31-8, NARIMASU 1-CHOME, ITABASHI-KU, TOKYO 175 JAPAN TEL:(0313976-3411 FAX:(03)3976- 3421 TOMY SEIKO CO.,LTD. MANUFACTURER 2-2-12. ASAHICHO NERIMA-KU, TOKYO 179 JAPAN TEL:(03)3976-3111 (Contents continued from back cover) humoral immunity and blood thyroxine levels in the toad, Bufo regularis ......... 349 Nakazawa, T. S., T. Machida and S. Kawashi- ma: Effects of unilateral and bilateral orchidectomy on laterality of neurons of the preoptic area and plasma levels of gonado- tropins and testosterone in male mice ....357 Matteo, L. D., S. Minucci, M. D’Antonio, S. Fasano and R. Pierantoni: Effects of a gonadotropin-releasing hormone analog (HOE 766) on germinal and interstitial com- partments during the annual cycle in the green frog: Rana esculenta Amano, M., K. Aida, N. Okumoto and Y. Hasegawa: Changes in salmon GnRH and chicken GnRH-II contents in the brain and pituitary and GTH contents in the pituitary in female masu salmon, Oncorhynchus masou, from hatching through ovulation Nozaki, M. and A. Gorbman: The question of functional homology of Hatschek’s pit of amphioxus (Branchiostoma belcheri) and the vertebrate adenohypophysis_ .............. 387 Jacob, M.: In vitro spermatogenesis in Oryctes rhinoceros (Coleoptera, Scara- baeidae): the role of ecdysone and juvenile hormone (COMMUNICATION) Kikuta, T. and H. Namiki: Identification of intracellular localization of laminin in the rat anterior pituitary (COMMUNICATION) Behavior Biology Kasuya, E., T. Kumaki and T. Saito: Vocal repertoire of the Japanese treefrog, Rha- cophorus arboreus (Anura: Rhacophoridae) (COMMUNICATION) Asada, N., K. Fujiwara, H. Ikeda and F. Hihara: Mating behavior in three species of the Drosophila hypocausta subgroup ..... ST) Systematics and Taxonomy Hirose, E., T. Nishikawa, Y. Saito and H. Watanabe: Minute protrusions of ascidian tunic cuticle: some implications for ascidian PIVIO PEMA eee ees hres Weve Lana 405 Kubota, S. and T. Horita: medusa of the genus Eirene (Leptomedusae; Eirenidae) from Toba, Japan Furuya, H., K. Tsuneki and Y. Koshida: Two new species of the genus Dicyema (Mesozoa) from octopuses of Japan with notes on D. misakiense and D. acuticephalum A new hydro- ZOOLOGICAL SCIENCE VOLUME 9 NUMBER 2 APRIL 1992 CONTENTS REVIEWS Kanzaki, R. and T. Shibuya: cessing pathways of the insect brain Olfactory pro- Ogawa, K.: Primary structure and function of adynein motor molecule? {s453ssceeo ae 265 ORIGINAL PAPERS Physiology Harmon, J. S. and M. A. Sheridan: Previous nutritional state and glucose modulate gluca- gon-mediated hepatic lipolysis in rainbow trout, Oncorhynchus mykiss Cell and Molecular Biology Kuroda, Y., Y. Shimada, B. Sakaguchi and K. Oishi: Effects of sex-ratio (SR)-spiroplas- ma infection on Drosophila primary embry- onic cultured cells and on embryogenesis Takagi, K. and S. Kawashima: Attempts to improve survival of neurons derived from neonatal rat hypothalamus-preoptic area in serum-free media Ichikawa, T. and K. Ajiki: Development of an in situ hybridization histochemistry for choline acetyltransferase mRNA with RNA probes Biochemistry Harbige, L. S., K. Ghebremeskel, G. Williams and M. A. Crawford: Hepatic fatty acids in wild rockhopper (Eudyptes crestatus) and magellanic (Spheniscus magellanicus) penguins before and after moulting Lawrence, J. M. and P. Moran: Proximate composition and allocation of energy to body components in Acanthaster planci (Linnaeus) (Echinodermata: Asteroidea) Developmental Biology Tanimura, A. and H. Iwasawa: Origin of Somatic cells in Bidder’s organ and the gonad proper in the toad, Bufo japonicus formosus (COMMUNICATION) Fujino, Y., A. Fujiwara, I. Yasumatsu and T. Fujii: Chromogranin A-like proteins in the heat-stable fraction of sea urchin eggs, embryos and the substances secreted with sperm Funakoshi, K., Y. Fukue and S. Tabata: Tooth development and replacement in the Japanese greater horseshoe bat, Rhino- lophus ferrumequinum (COM- MUNICATION) Kearn, G. C., K. Ogawa and Y. Maeno: Hatching patterns of the monogenean para- sites Benedenia seriolae and Heteraxine heter- ocerca from the skin and gills, respectively of the same host fish, Seriola quinqueradiata (COMMUNICATION) , Miyata, S., Y. Nishibe, M. Sendai, I. katayama and H. K. Kihara: site of appearance of a protease in Xenopus embryos nippon Changes in timing and Reproductive Biology Takahashi, T., Y. Tsuchiya, Y. Tamanoue, T. Mori, S. Kawashima and K. Takahashi: Occurrence of a novel 350-kDa serine pro- teinase in the fluid of porcine ovarian folli- cles and its increase during their maturation Endocrinology Saad, A. H. and W. Ali: Seasonal changes in (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 june 1992 An International Journal PHYSIOLOGY CELL and MOLECULAR BIOLOGY GENETICS IMMUNOLOGY BIOCHEMISTRY DEVELOPMENTAL BIOLOGY REPRODUCTIVE BIOLOGY ENDOCRINOLOGY BEHAVIOR BIOLOGY ENVIRONMENTAL BIOLOGY and ECOLOGY SYSTEMATICS and TAXONOMY published by Zoological Society of Japan © arated 6 Business Center for Academic Societies Japan es VSP, Zeist, The Netherlands 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. Phone (03) 3814-5675 Managing Editor: Officers: (oures Yamaguchi (Okayama) President: Hideo Mohri (Chiba) Assistant Editors: Secretary: Hideo Namiki (Tokyo) Yoshihisa Kamishima (Okayama) Treasurer: Makoto Okuno (Tokyo) Nilasns Saal (CIS vamne) Librarian: Masatsune Takeda (Tokyo) Akiyoshi Niida (Okayama) a... Editorial Board: Howard A. Bern (Berkeley) Walter Bock (New York) Aubrey Gorbman (Seattle) Horst Grunz (Essen) Robert B. Hill (Kingston) Yukio Hiramoto (Chiba) Susumu Ishii (Tokyo) Yukiaki Kuroda (Tokyo) John M. Lawrence (Tampa) Koscak Maruyama (Chiba) Roger Milkman (Iowa) Kazuo Moriwaki (Mishima) Richard S. Nishioka (Berkeley) Chitaru Oguro (Toyama) Tokindo S. Okada (Okazaki) Andreas Oksche (Giessen) Hidemi Sato (Nagano) Mayumi Yamada (Sapporo) Ryuzo Yanagimachi (Honolulu) Hiroshi Watanabe (Tokyo) ZOOLOGICAL SCIENCE is devoted to publication of original articles, reviews and communications in the broad field of Zoology. The journal appears bimonthly. An annual volume consists of six numbers of more than 1200 pages including an issue containing abstracts of papers Pies ie at the annual weeline of the Zoological Society of Japan. MANUSCRIPTS OFFERED FOR CONSIDERATION AND CORRESPONDENCE CONCERN- ING EDITORIAL MATTERS should be sent to: | Dr. Tsuneo Yamaguchi, Managing Editor, Zoological Science, Department of Biology, Faculty of Science, Okayama University, Okayama 700, 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, Godfried van Seystlaan 47, 3703 BR 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. © Copyright 1992 by the Zoological Society of Japan. In the U.S.A., authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by [copyright owner’s name], provided that designated fees are paid directory to Copyright Clearance Center. For those organizations that have been granted a photocopy license by CCC, a separate system of payment has been arranged. Copyright Clearance Center, Inc. 27 Congress St., Salem, MA, U.S.A. (Phone 508-744-3350; Fax 508-741-2318). 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 9: 475-498 (1992) © 1992 Zoological Society of Japan REVIEW Primary Sex Determination in Mammals | YUKIFUMI NAGAI “ Biological Laboratory, Fukui Medical School, Shimoaizuki 23, Matsuoka-cho, Yoshida-gun, Fukui 920-11, Japan ABSTRACT—The basic embryonic program in mammalian sexual differentiation is inherently femi- nine. The presence of a gene on the Y chromosome diverts the basic program into male pathway. Recently, a candidate for the gene on the Y chromosome has been almost certainly identified. This review summarizes the H-Y antigen hypothesis, zinc finger protein and current state of knowledge on the nature and function of TDF and Tdy. INTRODUCTION Sexual differentiation in mammals can _ be viewed as a sequential and ordered process as formulated by Jost [1] (Fig. 1). Each step in this process is dependent on the preceeding one. It begins with the establishment of genetic (chromo- somal) sex and gradually proceeds during long period covering from fetal life to puberty. GENETIC SEX | GONADAL SEX PHENOTYPIC SEX Fic. 1. Jost formulation of sequential event in mamma- lian sexual differentiation. In mammals, the female is called as the homo- gametic sex because she has two X chromosomes, and produces the ovum which possesses a single X chromosome. On the other hand, the male is refered to as the heterogametic sex because he has an X and Y chromosome and produces two Received February 12, 1992 populations of sperm, one X-bearing and the other Y-bearing. Genetic sex is therefore determined at the moment of fertilization, depending on whether an X or Y-bearing sperm fuses with the ovum. Gonadal (primary) sex is determined by the gene- tic sex. Once the gonads have completed dif- ferentiation, the gonadal sex in turn dictates phe- notypic (somatic) sex. Our understanding of phe- notypic sex differentiation comes largely from the classic experiments of Jost et al. [2]. They demons- trated that male phenotype was induced by secre- tions from the fetal testes while their absence resulted in female phenotype. Now two hormones essential for male development are well characte- rized: an androgenic steroid produced by the Leydig cells, testosterone, which is responsible for virilization of the internal and external genitalia, and a glycoprotein produced by the Sertoli cells, anti-Mullerian hormone (AMH), which causes re- gression of Mullerian ducts [3]. Sexual differentia- tion of the brain also occurs in fetal or neonatal life under influence of testosterone. Sexual dimorph- ism culminating at puberty is elaborated by the action of gonadal sex hormones. It has been perceived that the basic program of mammalian sexual differentiation is imherently feminine. The gonadal primordium is program- med to develop into the ovary. In the presence of the Y chromosome, however, the gonadal primor- dium organizes the testis instead of the ovary. In 476 Y. NAGAI mammals, the only genetic difference between the two sexes is represented by the Y chromosome, which is male-specific. Although only the female has two X chromosome, the disparity in the num- ber of X chromosome between the two sexes is compensated by the X inactivation mechanism [4]. Occasionally individuals with wrong number of sex chromosomes are born mostly owing to the nondis- junction of sex chromosomes during the meiotic division of the germ cell. In humans, individuals with sex chromosome constitution XY, XXY, XXXY and even XXXXY develop as men and those with XO, XX and XXX develop as women. That is, no matter how many X chromosome are present, the testis will develop in the presence of at least one Y chromosome [5]. It thus became apparent that Y chromosome contains a determi- nant that is essential for the development of the testis, the testis-determining gene. If the gonadal primordium develops into the testis, the program of phenotypic sex differentiation is diverted to that of male development not by direct gene action but by hormones produced by the testis. The development of the gonadal primordia into the testes (testis determination) is a crucial event in mammalian sex differentiation, and thus the testis-determining gene is thought to be a major regulatory gene that are responsible for diverting the basic feminine program into that of male development. Although analyses of structural abnormalities in the human Y chromosome have suggested that the testis-determining gene is lo- cated on the short arm, virtually nothing was known about either the nature or the function of this gene until recently. Very recently, candidates for this gene in humans and mice have been almost certainly identified [6, 7]. It has been also demon- strated that this gene alone can induce testis dif- ferentiation and subsequent male development when introduced in XX female mouse embryos [8]. This review will summarize current state of knowl- edge on the primary sex determination in mam- mals including historical aspect of the research. MORPHOLOGICAL ASPECTS OF GONADAL DIFFERENTIATION The gonadal primordia of the male and female, the gonadal ridges, have a common origin and the bipotentiality that can develop as either the testis or the ovary. They appear as thickening on the ventral cranical surface of each mesonephros, the second of the three consecutive nephroic struc- tures, early in fetal life, consisting of the coelomic epithelium and the underlying mesenchymal tissue supported by the developing mesonephric tissue. They are morphologically indistinguishable be- tween the two sexes and thus called the indifferent gonad. The primordial germ cells (PGCs), the progenitors of the germ cells, are of extra-gonadal origin and later in development migrate in the gonadal ridges [9]. The gonadal differentiation begins when the migration of PGCs from the yolk sac through the tissues of the hindgut and the dorsal mesentery has completed. In general, histo- logical differentiation of the fetal testis precedes that of the fetal ovary, depending on the species, by days to weeks. The early gonadal primordia consists of the migrant germ cells and somatic cells of three different tissues: coelomic epithelium, mesenchyme and mesonephric tissues. Although there is still a controversy about mesonephric origin of somatic cells in the gonadal primordia [10], histological and ultrastructural observations with high resolution techniques of early stage of the developing gonad favor the view that substan- tial population of gonadal somatic cells is derived from the mesonephros [11]. During gonadal growth the migrant germ cells divide mitotically and move into the underlying gonadal blastema where cells stream in from the coelomic epithelium and the mesonephric tissue. The developmental fate of gonads seems to be determined by the - well-controlled interaction of the germ cells with the different cell type of somatic blastema in the gonad [12]. The first morphological sign of sexual diffentiation in the fetal gonad is the formation of the testicular cords in the male. Primordial Sertoli cells emerged out of the intermingling blastema aggregate and enclose the germ cells into the cords [13]. Once incorporated into the cords, prolifera- tion of germ cells is suppressed and differentiation beyond the spermatogonial stage is arrested. The testicular cords are separated from the surround- ing blastema by a distinct basal lamina, thus creat- ing intracordal germ cell compartment. Through- Primary Sex Determination in Mammals 477 out differentiation the cords retain connection with mesonephric cell mass which gradually form the rete testis. The coelomic epithelium transforms into an epithlium with the basal lamina and simul- taneuously the testicular cords retract from the testicular surface. An area of loose mesenchymal tissue left between them develops into the tunica albuginea. The Leydig cells differentiate in the extracordal compartment shortly after the testicu- lar cords have formed. In contrast to the early development of the fetal testis, the internal organization of the female gonad remains indifferent and the sexual dif- ferentiation starts at a later stage with species- dependent variation. After a period of rapid proliferation, female germ cells enter the meiosis. The presence of the germ cells initiating meiosis is the characterstic feature of a differentiating female gonad in most mammals. The male germ cells do not start meiosis at this time. However, initiation of meiosis is not the first sign of ovarian dif- ferentiation. Two types of early ovarian dif- ferentiation is recognized, depending on the period separating gonadal sex differentiation and onset of meiosis. In female of some species, such as mouse and rat, the germ cells enter the first meiotic prophase simultaneously with or shortly after morphological sex differentiation. Their Ovaries appear compact with the germ cells uni- formly distributed throughout the ovarian tissues. On the other hand, the initiation of meiosis is more or less delayed with respect to the ovarian dif- ferentiation in females of other species, such as sheep and pig. In such ovaries the germ cells become enclosed in cell cords almost simul- taneously with the formation of testicular cords in the testis, but the cell cords begin to break up in the central part of the ovary by the end of the delayed period. The invading mesonephric cells push the germ cells from the central area of the developing ovary toward the periphery, thereby forming an ovarian cortex populated with germ cells and a medulla consisting mainly of mesonephric cells. When each oocyte reaches the diplotene stage of meiosis, it becomes surrounded by pregranulosa cells supplied by the somatic blastema, and a primodial follicle is formed. The follicle is equivalent to the testicular cords in that both structures enclose and separate the germ cells from the surrounding environment, providing a unique compartment for the germ cells to differ- entiate and maturate. Different experiments indicate that the germ cells may be not necessary for gonadal develop- ment and differentiation. Selective elimination of PGCs with busulphan in the rat does not interfere with all morphogenetic events occuring in a normal gonad [14]. Mutant mice homozygous for a genetic mutation (steel) are deficient in germ cells because of a failure of the germ cells to proliferate. They can form well-developed testicular cords consisting of Sertoli cells only [15]. Hence, it appears that the gonadal somatic cells can organize into a testis or an Ovary irrespective of the presence or absence of the germ cells. GANADAL SEX DETERMINATION AND Y CHROMOSOME As described above, gonadal sex in mammals is normally determined by the genetic sex. The ganadal sex of fishes, amphibians and birds is known to be changeable by applying sex steroid hormones during early stages of differentiation, although becoming increasingly resistant as the vertebrate evolution has advanced [16-18]. The gonadal sex of mammals is, however, highly stable, once it has been determined by the genetic sex, and the sex steroid hormones have essentially no effect on the gonads of eutherian mammals, although there are reports which indicate that estrogen can induce a sex reversal of the testis in the newborn male of the marsupials. Mammalian embryos of both sexes are destined to develop in the mother’s womb dominated by female sex hor- mones. It is appropriate, therefore, that the embryonic gonads are designed to be independent of these hormones. In 1959, the Y chromosome was for the first time shown to carry a male-determining factor by the finding in both human [19, 20] and mouse [21] that XXY individual was male while XO individual was female. Cytogenetic studies have shown that the human Y chromosome consists of an euchromatic and of a heterochromatic portion. The former comprises the entire short and the proximal long 478 Y. NAGAI arm; the latter the centromeric region and the distal long arm [22]. Many structural abnormality of the human Y chromosome have been detected by cytogenetic studies. Correlations of these abnormal karyotype with the phenotype have re- vealed that male-determinng factor is located in the pericentric region of the short arm of the Y [5, 23]. It has been widely assumed that male- determining factor in the mouse is located on the proximal region of the long arm of the Y, but very recently its location on the minute short arm of the Y has been confirmed by in situ hybridization with a specific DNA probe [24, 25]. As described earlier, the basic embryonic plan of mammals is inherently feminine. The diversion of this plan is carried out by the male-determining factor on the Y chromosome which directs the embryonic in- different gonad to organize the testis instead of the ovary. The testis then secretes testosterone, which induces male accessory glands and ducts, and all other masculine secondary sex characteristics. Therefore, the male-determining factor has been regarded as a master regulatory gene (genes) posi- tioned at the top of a hierarchy consisting of a number of the regulatory genes involved in mammalian sex differentiation. Since all that the Y-encoded regulatory element do is to direct the indifferent gonad to develop as a testis, they have been named TDF (testis-determining factor) in humans and Tdy (testis-determining gene-Y chromosome) in mice, respectively. In what fol- lows I shall refer to them as Tdy unless in the specificed case. It would be reasonalbe to assume that mamma- lian sex determination and differentiation involve either directly or indirectly a large number of genes, but they should be controlled by a series of genetic regulatory systems that consitute an order- ly hierarchy. Based on the above thinking, Ohno has put forward an attractive theory concenrning the genetic basis of mammalian sexual develop- ment [26, 27]. The mammalian embryo has an inherent tendency to develop a female. The male development is due to two-step interventions by two major regulatory genes: the first for the gonad- al sex determination, which is on the Y chromo- some (Tdy) and the second for the phenotypic sex differentiation, which is on the X chromosome. As described above, development of male extra- gonadal characters is induced mostly by testoster- one secreted by the testis. Evidence for direct involvement of the X-linked gene in the develop- ment of male phenotype has came from a mutation of the relevant gene in the mouse [28]. The affected XY mice, carrying Tdy, have organized normal testosterone-producing testes, but do not show further masculine development, thus exhibit- ing externally female phenotype known as testicu- lar feminization (Tfm). The Tfm mutation renders all the target cells completely nonresponsive to androgens [29]. According to the current concept of how androgen acts within target cells, the androgens entered into the cells bind to a specific cytoplasmic receptor proteins. Subsequently androgen-receptor protein complexes move into the nucleus, where they associate with specific binding sites (hormone responsive elements) on the chromosome and promote the transcription of tissue-specific genes [30]. Further studies on the Tfm mice have revealed that the nonresponsive- ness is due to a mutational deficiency of the androgen receptor proteins specified by the Tfm locus on the X chromosome [31]. In general, all genetic defects in the process essential to life like morphogenesis are expected to cause lethal abnormalities followed by abortion. However, the defects in sexual development are only an exception, because normal sexual develop- © ment is essential only to the survival of the species and not to the life of individuals. It follows then that individuals with various abnormalities in sex- ual development are found in different species. The analysis of these individuals, particularly those resulting from a single gene mutation, has pro- vided us with important clue in defining the genes and gene products involved in normal sexual de- velopment. Indeed, the Tfm mutation in mice has greatly contributed to assign the primary regula- tory gene responsible for extragonadal sexual dif- ferentiation to the X-linked Tfm locus and also to define the product of the Tfm gene as an androgen receptor protein. On the other hand, as noted above, Tdy represents a master regulatory gene which switchs on the program toward male de- velopment. Therefore, any mutational defects rendering Tdy nonfunctional are equivalent to its Primary Sex Determination in Mammals 479 absence, resulting in development of externally normal female without clue as to either the nature or mode of function of the product specified by Tdy. Although the existence of Tdy on the Y chromosome was widely accepted, no relevant mutation has been found until very recently and the nature and product of the gene has been the speculative subject. It was H-Y antigen that was introducted as the first candidate of the protein specified by Tdy. H-Y ANTIGEN HYPOTHESIS In 1955, Eichwald and Silmser found an unex- pected phenomenon that in a highly inbred strain of mice skin grafts from males to females were rejected whereas the grafts exchanged between all other sex combination survived indefinitely [32, 33]. These male skin grafts were rejected slowly compared with grafts from major histocompatibil- ity complex (MHC)-incompatible donors. The only genetic difference between males and females within a highly inbred strain is confined to the presence of the Y chromosome in males. For this reason, the rejection was attributed to a ‘weak’ transplantation antigen specified by a ‘minor’ his- tocompatibility locus on the Y chromosome. This antigen became known as the H-Y (histocompati- bility-Y) antigen [34]. It is ubiquitously expressed in various cell type of all the males, but absent from those of normal females [35]. Since such a weak antigen can be demonstrated only within inbred strain, subsequent studies of H-Y antigen have been confined largely to the mouse. When antigen enters the body, two different types of immunological reaction occur generally: one is the synthesis and release of antibody (humoral immunity); the other is the generation of specifically sensitized T cells, e.g., helper T cells (T;,) and cytotoxic T cells (Tc) (cell-mediated immunity). Mismatch for the minor histocompat- ibility antigen elicits graft rejection concomitanly with the generation of the sensitized T cells, but it has been very difficult to raise antibodies to them. However, in the case of H-Y antigen, antibody response was found in the first of the two re- sponses. Goldberg et al. [36] demonstrated female mice which had rejected several male skin grafts produced H-Y antibodies and the potency of the antiserum could be determined by the cytotoxicity (killing) against mouse spermatozoa in the pre- sence of complement. Antimale (anti-H-Y) speci- ficity of the antibodies was ascertained by the absorption test: when either male or female cells were mixed with the antisera, male and not female cells could specifically remove the cyotoxicity for spermatozoa. Shortly thereafter, Scheid et al. [37] developed a cytotoxicity test using male and female epidermal cells prepared from mouse tail skin as target cells. H-Y antibody specifically kills male but not female epidermal cells in the pre- sence of complement. Subsequently, for reasons of technical convenience, H-Y antibodies were raised by repeated injections of female inbred mice or rats with syngeneic male cells at weekly inter- vals. Except spermatozoa, male epidermal cells and 8-cell XY embryos [38], all other types of male cells were not susceptible to lysis by H-Y antibody, but the presence of H-Y antigen on their cell membranes was proved by the absorption test. Once the cytotoxicity test was developed it became possible to examine the expression of H-Y antigen on cells of other animal species by the absorption test. Wachtel et al. [39] demonstrated that male cells from several mammalian species (rat, guinea pig, rabbit and man) had H-Y components anti- genically related to H-Y antigen of the mouse whereas female of these species possessed no cross-reactive components. In subsequent studies extending their survery to classes other than mam- mals, they found the occurrence of a cross reactive component in the chicken and in two amphibian species [40]. Interestingly, in chickens and in species of frogs in which the female is the hetero- gametic sex (male and female are refered to as ZZ and ZW, respectively), the situation was reversed in that female cells now absorbed murine H-Y antibody. These results have indicated that the cell surface component which evokes production of H-Y antibody in the female mouse ts highly con- served in vertebrate evolution. An important lesson which we have learned from the primary structure of functional protein is that genes of fundamental importance are seldom permitted to undergo evolutionary change in the active sites of the proteins for which they code. Accordingly, the 480 Y. NAGAI evolutionary conservation of the H-Y antigen indi- cates that this antigen is not a minor histocompat- ibility antigen, but rather a molecule that has been performing the invariant and important function during the past few hundred million years. The H-Y antigen is invariantly associated with the heterogametic sex, that is, either the Y or the W chromosome, signifying that the function is sex- related. Ubiquitous expression of the antigen on male cells can be attributed to the constitutive expression of the H-Y gene that will not be con- trolled by other regulatory genes. In 1957, Mosco- na showed that a suspension of individual cells dissociated from embryonic organs in a rotation culture could autonomously reconstitute the orga- nized structure characteristic for the organ from which those cells were derived [41]. The similar type of experiments subsequently performed with a variety of organs have supported the view that the information necessary for organogenesis is carried by the plasma membrane protein or gly- coprotein of individual component cells. Inter- action between the surface of somatic cells and germ cells also seems essential for the organization of testicular structure. When intact viable cells are injected into animals, those proteins (or the sugar moieties of glycoprotein) can be recognized as the cell surface antigen, resulting in production of the antibodies directed for them. Based on these consideration, Wachtel et al. [42] have proposed that H-Y antigen is the long sought-after product of the Tdy of mammals. In the recent review [43], Ohno, one of proposers, describes in retrospect that, in addition to the above consideration, they has also the following realization: The testis and the ovary are really two sides of the same coin. Thus, it appeared that a single species of male- specific plasma membrane antigen should suffice to divert the inherent inclination of embryonic in- different gonads toward ovarian differentiation and cause their testicular development. The proposed identity of H-Y antigen and the product of the Tdy was tested on a considerable variety of exceptional individuals whose gonadal sex do not agree with their chromosomal sex [44, 45]. The proposal predicts the following: any individual who possesses testes in spite of the apparent absence of the Y chromosome should express H-Y antigen on his cells; on the contrary, any individual who has ovaries in spite of the apparent presence of the Y chromosome should be H-Y antigen negative on her cells. A series of absorption tests performed on such exceptional individuals yielded no exception from the above expectation. How H-Y antigen works in testicular organization was examined by two types of in vitro culture systems: one by Moscona-type aggregation systems of dissociated gonadal cells from murine neonates [46, 47] and the other by organ culture system of bovine fetal indifferent gonads. Dissoci- ated individual gonadal cells of murine neonates reorganized histotypic aggregates in the rotation culture: testicular cells formed tubules-like struc- tures; ovarian cells reorganized follicle-like struc- tures. It has been found that in the presence of their antibodies in excess, most, if not all, of the plasma membrane antigens gather over one pole of the cell, namely, cause “capping”. Since capped antigens are engulfed and digested by the cell, its plasma membrane becomes temporally void of particular antigen (lysostripping phenomenon) [48]. Testicular cells, lysostripped of H-Y antigen by this method, reorganized follicle-like aggre: gates while ovarian cells in the presence of pre- sumptive H-Y antigen partially formed tubule- like aggregates. Daudi cells, a human male Burkitt lymphoma cell line, have been found to excrete a component capable of absorbing the H-Y antibody of mice into the culture medium due to the de- ficiency of HLA antigen expression [49]. Organ culture of bovine XX embryonic indifferent gonads in the presence of the H-Y component from Daudi cells induced very precocious testicu- lar organogenesis [50]. The proposal that the Tdy specifies H-Y antigen was supported by these experimental evidences and initially seemed to be plausible. However, appearance of a mutant in mice delivered a fatal blow to the proposal. Before dealing with the fatal blow against the proposal, it will be better to touch on the problems concerning H-Y serology, and the identity of H-Y transplanta- tion antigen and the antigen detected by serologi- cal assay. Primary Sex Determination in Mammals 481 H-Y TRANSPLANTATION ANTIGEN AND SDM ANTIGEN There are technical difficulties associated with cytotoxicity test for H-Y antibody. Especially it requires considerable skill and experience to mas- ter the sperm cytotoxicity test. Thus, other sim- plified serological methods for detection of H-Y antigen have been developed [51, 52]. I have been attempting to purify and characterize the H-Y antigen for more than ten years. Now [ realize that the most serious problem in H-Y serology is lack of H-Y antibody of good quality. My routine method to prepare the antibody is as follows. Antisera are raised by immunizing female mice at least six-times with syngeneic male spleen cells at weekly inter- vals. Antiserum collected from individual female mice is selected by two-step procedure using a refined cytotoxicity test against epidermal cells [53]: First, for potency by the cytotoxicity against male epidermal cells and second, for specificity by the absorption test with male and female spleen cells, respectively. Male-specific antisera are pooled and used for experiments. Although being tedious and time-consuming, this procedure is essential to get reliable H-Y antibodies. Accord- ing to my experience, the antisera, so obtained, are always of low titer and also contain very often autoantibodies being cytotoxic to both male and female epidermal cells. Moreover, only a few of immunized female mice produce male-specific antibodies, resulted in the limited supply of the antiserum of the same quality. This has hampered the biochemical study of H-Y antigen necessary for understanding of its function. The hybridoma technology introduced by Kohler and Milstein opened a novel way to generate a continuous cell line producing a monoclonal antibody [54]. In several laboratories including us, this technique has been hopefully applied to produce monoclonal antibodies which have the advantages of specific- ity, potency and ample supply of homogeneous antibody [55-58]. They succeeded independently in producing different types of monoclonal H-Y antibodies. But these antibodies do not seem to improve substantially the above defects of poly- clonal antibodies as evidenced in little progress of studies on the biochemical nature and function of H-Y antigen. When inbred female mice were grafted with male skin of the same strain, not all could reject grafts: in the A/Jax strain (H-2* haplotype), male skin grafts were rejected by some recipients; in the C57BL/6 strain (responder, H-2° hapolotype), females rejected consistently male skin grafts; in C3H_ strain (nonresponder, H-2* haplotype) females accepted usually male skin grafts indefi- nitely [32]. In contrast, antibody responses were found in all strains of female mice grafted with syngeneic male skin irrespective of responder or nonresponder [59]. At that time, this difference was not seriously considered, for expression of H-Y antigen in the male of nonresponder strains was clear from the experiments that female F, hybrids between a responder strain and a nonres- ponder strain could reject male skin of either parental strain [60]. In the 1970’s it had been assumed that the antigen detected by serological assays and the male-specific transplantation anti- gen discovered by Eichwald and Silmser were one and the same. Later, two instances of individuals were reported that lacked H-Y transplantation antigen but possessed H-Y antigen detected by serological assays [61, 62]. Regarding these findings as an indication that the two H-Y antigens were separate molecules, Silvers et al. [63] have recommended that the term H-Y antigen must reserve for the H-Y transplantation antigen while the H-Y antigen detected by serological assays had better refer to as serologically detectable male (SDM) antigen. If so, the H-Y antigen hypothesis should be refered to as SDM antigen hypothesis, for it is mostly based on evidences obtained by serological assays. H-Y transplantation antigen is the most studied of all minor histocompatibility antigens. As with all histocompatibility antigens, the skin graft rejection by females is mediated by T lymphocytes: T, cells which can_kill the target cells; T;, cells which help to generate T, cells and to cooperate with B cells in production of antibody. After the first demonstration of occurrence of T. cells in the spleen of H-2° female mice previously grafted with syngeneic male skin [64], a refined in vitro assay has been developed for typing the presence or absence of the antigen [65]. Subse- quently, H-Y specific T. and T;, cell clones were 482 Y. NAGAI isolated and have been used for typing [66]. Minor histocompatibility antigens are generally recongnized by T cells in an MHC-restricted man- ner like virus-induced cell surface antigens [67]. The current view of MHC-restriction in the T cell response is as follows [68]: Both T, and Ty, cells carry individually a unique T cell antigen receptor on their cell surface, which enables them to recog- nize specific antigens. Both T cells do not see free antigen, but rather recognize antigenes in associa- tion with MHC antigens. T, cells usually recognize antigen in association with class I antigens, which are expressed on all nucleated cells, while T), cells recognize antigen in association with class II anti- gens, which are expressed mostly on antigen- presenting cells and also some lymphocytes. Anti- gems are presented on the cell surface via two different route depending on their origin, either external antigen like secreted bacterial toxin or endogenous proteins, after undergoing processings [69]. In the case of minor histocompatibility antigen, the processed antigens are peptide pro- duced from cellular proteins by proteases. They are transported into endoplasmic reticulum where they are mounted in a groove formed by the restricting MHC molecule, and then to the cell surface where they can be recognized by specific T cells [70]. In mice, T cell responce to the H-Y antigen is H-2 restricted [71]. Thus, T, cells from a female immunized with syngeneic male spleen cells can recognize the H-Y antigen only on cells of the same H-2 haplotype. Very recently, R6tzschke er al. [72] have succeeded in defining naturally pro- cessed H-Y peptide by the following approach: peptide fractions extracted from male mouse spleen of two strains are separated by a reverse HPLC; The fractions are screened by loading them externally on culture cells of the same H-2 haplo- type that lack the antigen and testing the suscep- tibility of these cells to lysis by a H-Y specific T, cell line. Although the entire amino acid sequence of the peptide is not determined, available evidences suggest that the H-Y antigen is short peptide, less than 16 amino acid residues. Studies of tissue distiribution have revealed that the H-Y peptide is present in thymus, spleen and lung, and less in skin, whereas it is undetectable in testis, brain, skeletal muscle and heart [73]. They have not identified parent molecule (H-Y protein) from which the H-Y peptide derives. This approach is expected to lead more refined understanding of the following issues: biochemical nature of the H-Y peptide and protein; their involvement in sex- related function; cloning of the gene controlling H-Y antigen expression on the short arm of the mouse Y chromosome. In contrast, as described above, virtually nothing is known about the mole- cule of SDM antigen. Although Goldberg [74] argued that H-Y antigen and SDM antigen were one and the same, it seems unlikely that the peptide mounted in the groove formed by the restricting H-2 molecule is recognized by B cells without H-2 restriction. But this question remains unsolved. SEX REVERSAL IN MICE Sxr (sex-reversed) mutation of the mouse first reported by Cattanach et al. [75] is a dominant inherited trait, by which the sex reversal of carrier XX (XXsxr) female mice is caused. It had been thought to be a dominant autosomal gene, prob- ably derived by translocation from the Y chromo- some. In the normal male mouse, Tdy is located on the short arm of the Y chromosome. Sxr mutation has now been found that the segment containing Tdy is duplicated and transposed to the distal end of long arm [76, 77]. The homologous segments between the X and Y chromosome are located at the distal end of their long arm. The Y chromosome pairs with the X chromosome during the first meiotic division at this region, where an obligatory crossover take place [78, 79] (Fig. 2a). Being attached at the tip of the homologous re- gion, the extra copy of Sxr region is invariably transferred to the long-arm tip of one of the two X sister chromatids after the first meiosis of the XYsxr males (Fig.2b). Thus half of the XX progeny inherits Xsxr from father and usually develops as phenotypic males, but they are sterile. The presence of two X chromosomes is assumed to be unfavorable to male germ cell development since XX germ cells degenerate soon after birth. XXsxr males have testes and are positive for H-Y antigen as determined by skin graft techniques [80], by T. cells response [81] and by serological Primary Sex Determination in Mammals 483 PAR Xx Y Xx XXsxr4 XXsxr'h H-Y(+) H-Y(-) Ysxr Fic. 2. Sex reversal in mice. (a) crossing over between the X and Y chromosome in normal male during meiosis. (b) crossing over in XYsxr male. (c) XXsxr male. (d) XXsxr’ male. PAR: pseudoautosomal region. The dotted and hatched areas at the long-arm distal end of the X and Y chromosome represents Sxr and Sxr’ fragments, respectively. Tdy: testis-determining gene. Hya: the gene controlling expression of the H-Y antigen. assays [80, 82], so they have been regarded as an evidence in support of the H-Y antigen hypothesis that Tdy and Hya are an identical gene. Subse- quent studies on Sxr mutation, however, have indicated that the two genes are tightly linked but two separate entities. In somatic cells of females, one of the two X chromosomes is randomly inactivated early in embryonic life. McLaren and Monk [83] examined whether the Sxr region attached to the X chromo- some was subject to the inactivation by making use of Searle’s X-autosome translocation T(X;16)H (T16H) (Fig. 3a). The normal X chromosome is known to undergo preferentially inactivation in T16H/X individuals. By mating T16H/X female with male carrying Sxr (X/Ysxr), they produced progeny in which the paternally derived Xsxr chromosome is invariably inactivated (Fig. 3b). All XXsxr individuals inheriting the normal X chromosome from mother were normal sterile males as expected, but when T16H was combined with Xsxr, some proportions of T16H/Xsxr indi- viduals developed as females. They explained this xX 16 (a) (b) Searle’s X-autosome translocation and the X inactivation. (a) Searle’s X-autosome translocation (T16H) that splits the X chromosome into two nearly halves is Fic. 3. schematically illustrated. The X chromosome is drawn as in Fig.2 whereas chromosome 16 is marked by dots. (b) somatic cell of T16H/Sxr is schematically illustrated. Under this condition in- tact X chromosome is preferentially inactivated. Shadow on the Xsxr chromosome represents ex- pected inactivated zone. observation by the analogy of the autosomal genes translocated on the inactive X chromosome [84] as follows. Inactivation may spread to a variable extent beyond the inactive X chromosome into the attached Sxr region. Thus both T16H/Xsxr and X/Xsxr individuals shows a mosaicism, that is, Tdy may be inactivated in some cells but expressed in others. This is a situation closely analogous to XX (interval 1) was further divided into four intervals; PAR soil 1A2. 16. 1C@u PAR 1A1 1A2 35 kb Molecular map of human Y chromosome Fic. 5. Primary Sex Determination in Mammals 487 1A1 adjacent to the pseudoautosomal boundary to 1A2, 1B and 1C in the order (Fig. 5, middle). Two patients were especially useful to narrow the search for TDF down to 1A2 intervals: a XX males carrying the smallest portion (0.5%) of the Y chromosome (1A1 and 1A2 intervals) that most other XX males had in common; a XY girl with a reciprocal Y;22 translocation who had 99.8% of the Y chromosome and only a deletion of intervals of 1A2 and 1B that were missing in other XY females. This led Page and his colleagues to propose that at least part o TDF must lie within 1A2 interval. They have cloned a 230-kilobase (kb) segment of the human Y chromosome that spans the interval (140 kb), the lowest common denominator in the two patients. In general, genes of vital function tend to be evolutionally con- served. The DNA sequences of TDF are expected to be conserved in all mammals adopting a Y chro- mosomal sex-determining mechanism. Accord- ingly, TDF is likely to be identified by a search of cloned 1A2 interval for DNA sequences conserved in mammalian species. Upon this reasoning, Page and his colleagues have screened the clones collec- tively representing 1A2 interval by hybridization to a “Noah’s ark blots” containing DNA from male and female pairs of eutherian mammals. Four highly conserved DNA fragments were found that cross-hybridize with sequences from DNA of all mammals examined. DNA sequencing of one highly conserved fragment has shown that it appears to encode a protein with “zinc finger”, a nucleic acid binding motif first described in a frog transcription factor [108], and has suggested that the encoded protein acts as a transcription factor in the first step of mammalian sex determination. Curiously, in addition to homologous DNA sequ- ences on the Y chromosome, this fragment also detected a very similar DNA sequences on the X chromosome of all mammals tested. Each of the three other conserved segments also detected their related sequences on the human X chromosome. The high degree of evolutionary conservation in both DNA sequences suggested that neither the Y locus nor the X locus was a pseudogene. The Y-encoded and the X-encoded zinc finger gene were designated ZFY and ZFX, respectively. Based on these findings Page and his colleagues formulated the four possible models assuming that ZFY is TDF whereas ZFX encodes a structurally similar protein which is probably involved in sex determination. Of the four models, three fit in with the currently prevailing notion that ZFY is a dominant male determiner. But one model is unique in postulating gene dosage as the basis of sex determination like those depending on the ratio of X chromosome to autosomes in Drosophi- la and Caenorhabditis [109-111]: ZFY and ZFX produce functionally interchangeable proteins, both are testis determining and ZFX is subject to X-chromosome inactivation. Hence, gonadal sex is determined by the total number of expressed X and Y loci; one active copy for a female and two active copies for a male. Subsequent studies have demonstrated that ZFX indeed shows the exten- sive structural similarity to the ZFY (99%, the degree of homology of the zinc finger region at amino acid level). However, transcription analysis of human-rodent hybrid cell lines containing inac- tive human X chromosome has revealed that ZFX escapes X inactivation [112], indicating that there is essentially no difference between male with one copy each of ZFX and ZFY and females with two copies of ZFX. The model is therefore incorrect. Northern analysis of the expression of the ZFY and ZFX showed that the four highly conserved sequences within the interval 1A2 and the corres- ponding sequences on the X chromosome consti- tuted exons of the ZFY and ZFX, and that both genes were transcribed in primary culture fibro- blast, in transformed cells and in all tissues ex- amined. The situation of ZFY-related gene in mice is rather complex, with a total four genes [113, 114]: two on the Y chromosome (Zfy-1 and Zfy-2), one on the X chromosome (Zfx) and an autosomal homologue on chromosome 10 (Zfa). Hybridiza- tion studies have revealed differences among Sxr, Sxr and Sxr”: while Zfy-1 is present in all three, Zfy-2 is present in Sxr, absent in Sxr’ and regained in Sxr’. Of the two Y chromosomal homologues only Zfy-1 was thought to be sufficient for testis determination since XXsxr and XXsxr mice have testes. The zinc finger motif is one of the highly con- served motifs found in the transcription factor that 488 Y. NAGAI bind to DNA in a sequence specific manner [115]. The DNA binding region has unique repeats that contain two invariant pairs of cysteine (Cys) and histidine (His) residues which coordinate single atom of zinc, resulting in a finger-like structure. Since ZFY was considered the best candidate for TDF, this gene and its homologues have been intensively studied particularly in humans and mice [116, 117]. The ZFY-related genes-encoded proteins predicted from open reading frames in their DNA sequence form a distinct subfamily of zinc finger proteins with the following characteris- tics. All have 13 zinc fingers of Cys-Cys/Hlis-His type encoded by a single exon and showing an unique two-finger repeat pattern of primary struc- ture, and consists of three domains in the order from the amino termini: an acidic domain that seems to mediate transcriptional activation func- tion; a short basic domain that is similar to the nuclear localization signal of SV40 large T antigen; zinc finger domain. As already described, most placental mammals except mice have two ZFY- related genes: one on the Y chromosome (ZFY) and one on the X chromosome (ZFX). Using ZFY probe DNA from male and female pairs of other vertebrate species were also surveyed by the hybrid- ization to determine whether genes homologous ZFY were present and if so whether they were present in the sex-limited pattern. Marsupials [118], birds [107], reptiles [119] and amphibians [117] also have genes homologous to ZFY but these are not present on the sex chromosomes. The unexpected finding was that the ZFY homo- logues in marsupials were not on either the X or the Y chromosome, but mapped to the autosomes [118]. In spite of the fact that the Y chromosome in marsupials was male-determining as in placental mammals, homologous sequences were not local- ized to the Y chromosome even after hybridization at reasonable low stringency. In addition to this finding two recent reports have questioned the role of ZFY and Zfys in male sex determination. Amplification analysis of adult testes mRNA by reverse transcription and polymerase chain reac- tion (RT-PCR) has shown that both Zfy-1 and Zfy-2 loci are transcribed, suggesting that both loci are functional. Male and female mouse embryos are indistinguishable until about 11.5 days post coitum (d.p.c.). The first visible sign of male development, the formation of testes with the alignment of Sertoli cells into cords, occurs within 24 hours. Accordingly, if being Tdy, Zfy-1 should be expressed in male gonadal ridge at or just before this stage of development. By applying a RT-PCR technique to mRNA derived from either pooled embryonic gonadal tissues or individual embryos, Koopman et al. [120] have shown that in male, but not female, gonadal ridge, the Zfy-1 transcripts appear just before testicular differentia- tion begins (at 10.5 d.p.c.), maintain the increasing level as the differentiation ensues (from 11.5 to 14.5 d.p.c.), and then decrease once testis forma- tion completes. By contrast, Zfy-2 transcripts were not detected in both embryonic mouse gonads at any of the stages. This finding seemed at the first sight to support to testis-determining function of Zfy-1. However, when the expression of Zfy-1 was analysed in W°/W* mutant male mice embryos that develop testes lacking germ cells, any Zfy-1 transcripts could not be detected, demon- strating that Zfy-1 expression in normal testis was due to the presence of germ cells. It was argued that neither Zfy-1 nor Zfy-2 was Tdy and Zfy-1. expression might instead have a role in the de- velopment of male germ cells. The presence of abnormally interchanged ZFY could explain majority but not all instances of XX males [121]. If ZFY is TDF, it should be predicted that all X-Y interchanged males carry ZFY, whereas all XX males lacking ZFY are due to mutations elsewhere in their genomes. According- ly, in the absence of the X-Y interchange the latter should also lack the Y-chromosome derived pseudoautosomal boundary. Using a PCR assay Palmer et al. [122] have searched the presence of DNA sequences derived from the X and Y chro- mosomal pseudoautosomal boundary in 14 XX males or hermaphrodites who lacked ZFY: three XX males and one of hermaphrodite were found to have the sequence from Y pseudoautosomal boundary. Their phenotype vary from normal male with testes lacking germ cells to hermaphro- dite with bilateral ovotestes and a uterus. The position of the exchanges in these individuals mapped to the region within 60kb of the pseudoautosomal boundary that did not include Primary Sex Determination in Mammals 489 the 1A2 intervals previously defined as the sex- determining region. They have suggested that TDF gene lies close to or even spans the pseudoautosomal boundary. The possibility that ZFY is TDF has been completely eliminated by their finding. A more detailed study for the breakpoints of the four individuals further narrowed the search for TDF down to a 35kb segment from the pseudoautosomal boundary (Fig.5, bottom). Sinclair et al. [6] found a 2.1 kb Y-specific sequ- ence from genomic clones covering this region. When a “Noah’s ark blot” was hybridized with a probe from this Y-specific sequence, this probe detected conserved and male-specific sequences in a wide spectrum of mammals. Analysis of this Y-specific sequence revealed the presence of two long reading frames. The longer one was found to encode a region of 120 amino acids probably corresponding to the last exon of a novel gene. The predicted amino-acid sequence showed a striking similarity to 80 amino acids of the mating- type protein Mc required for mating in the fission yeast [123]. This 80-residues conserved motif also showed homology with the nuclear non-histone high mobility group (HMG) proteins thought to play a role in chromosomal structure and gene activity [124]. Northern blot analysis of poly(A)* RNA from human tissues revealed that the novel gene encoded a testis-specific transcript. Scinclair and colleagues have termed this new human Y- located gene SRY (sex-determining region of Y) and proposed to be a candidate for TDF. Their results conflict with the result of Page et al. [107] on the woman with a reciprocal Y ;22 transloca- tion. She was shown to have only a deletion of 1A2 and 1B intervals. However, in the same issue of the journal, Page et al. [125] have reported that she has also a deletion of a second position of 1A1 corresponding closely to the region found in the ZFY-negative males and hermaphrodite. Very recently two reports have provided evidences sup- porting for SRY being TDF. By the single-strand conformation polymorphism assay and subsequent sequencing, Berta et al. [126] have shown that a de novo point mutation in the conserved motif of SRY is associated with sex reversal in an XY female. Jager et al. [127] have identified a frame shift mutation with a fournucleotide deletion in a sequence of the conserved motif of SRY in one out of 12 XY females with gonadal dysgenesis. In humans, translocations and deletions of the Y chromosome in XX males and XY females have provided an invaluable information for localizing and cloning a candidate for TDF. It is Sxr muta- tion in mice that has helped to define the position of Tdy. In addition to Tdy, Zfy-1 and Zfy-2, two genes and a DNA sequence are also located on the Sxr fragment and consequently to the short arm: the gene controlling expression of the H-Y antigen (Hya), the gene involved in spermatogenesis (Spy) that acts cell-autonomously in the germ cell line and Bkm DNA repeated sequence [128]. The gene controlling the expression of SDM antigen is also assumed to map to this fragment. It was suggested that Hya and Spy are an identical gene [129], but very recently a candidate for Spy (termed Sby) mapping to this fragment has been isolated, which is expressed in the testis and has extensive homolo- gy to X-linked human ubiquitin-activating enzyme El involved in DNA replication [130, 131]. As previously described, McLaren et al. [85] found a heritable variant of Sxr (Sxr ) that retained Tdy, but was deleted for Hya and Spy. The Sxr fragment is a minimum portion of the mouse Y chromosome known to contain Tdy. Later, they found an another variant arose from Sxr’ that regained Hya and termed Sxr’. Subsequent molecular and cytogenetical studies [24, 25] have showed that the generation of the Sxr’ mouse from Sxr is not due to a point mutation affecting Hya but involves a partial event within the Sxr fragment and also that the event leading to the generation of the Sxr” mouse is restricted to the Sxr fragment. Based on these findings, the two groups have proposed the models for the origin of Sxr and Sxr’: the Sxr’ and Sxr” mutants are generated during male meiosis by an unequal recombination event between the two Sxr segments and by in- trachromosomal recombination between the Y short arm and Sxr fragment, respectively. Re- cently, Lovell-Badge and Robertson [132] have generated a heritable mutation in mice by injecting XY embryonic stem cells multiply infected with a retroviral vector into host blastocysts. A resulting chimaeric male mouse gave rise to a low propor- 490 Y. NAGAI tion of XY females among his offspring. This mutation has been found to segregate exclusively with the Y chromosome among the offspring of these females and to be complemented not only by Sxr but also by Sxr’. Based on the phenotype and deduced location of the mutation, they concluded that it had occurred in Tdy itself. This mutated gene was refered to as Tdy™ and for simplicity expressed by the symbol ¥. Karyotypic analysis have revealed that about half of the offspring are sex-chromosome aneuploids (XY¥ males, XX¥ and XO females), implying that there is essentially no pairing between the X and ¥ chromosomes in female meiosis, with the X and ¥ segregating at random. Gubbay et al. [133] have shown that the Zfy genes did not undergo any detectable structu- ral alterations in the X¥ female mice and are transcribed normally from the Y chromosome in botn adult XY-¥ testis and X¥ female embryonic gonads. These finding also provides evidence that Zfy genes are not directly involved in testis deter- mination. Using the human Y-DNA sequence (SRY) as probe, Gubbay et al. [7| have cloned a homologous sequence from the mouse Y chromosome, which maps to Sxr’ fragment. The sequence contains an open reading frame homologous to those of both SRY and Y-linked sequence from rabbit DNA, strongly indicating that it is an exon of a functional gene. They have termed this gene Sry (a gene from the sex-determinign region of the Y chromo- some). The amino acid sequence encoded by the open reading frame also included a conserved motif showing homology both to the C-terminal 80 amino acid residues of Mc protein from the fission yeast and to known or putative DNA-binding proteins. Detailed hybridization studies showed that Tdy™ mutation was due to a deletion of part of Sry including the highly conserved exon. An RT-PCR assay with primers specific to the se- quences in the conserved exon has demonstrated that Sry is transcribed in testes, but not in liver of adult mice. Sry was found to be indeed expressed at the predicted time in male, but not female, uro- genital ridges. However, they failed to detect cDNA corrsponding to Sry in two cDNA libraries prepared from 11.5-d.p.c. male genital ridge. The screening extended to 14.5-d.p.c. testes and an 8.5-d.p.c. whole-mouse embryo library also failed to detect the cDNA. Although the testis libraries also failed to yield any recombinants with homolo- gy to the Sry probe, the 8.5-d.p.c. library produced four positive clones presumably derived from auto- somal loci. Sequence analysis of these autosomal Sry-related genes has shown that Sry is a member of a new family of at least five mouse genes which are related by the presence of a conserved amino- acid domain found either in a gene involved in mating type of the fission yeast or in the DNA- binding protein. Subsequently, Koopman et al. [134] examined in more detail Sry expression dur- ing testis development and in adult testis. Trans- cripts were first detected in male gonadal ridge at 10.5 d.p.c. just before onset of testis differentia- tion, were present at similar levels in the 11.5- d.p.c. urogenital ridge and decreased in 12.5- d.p.c. testis, followed by rapid cessation of its transcription. No Sry transcripts were also de- tected in 7.5-, 8.5- and 9.5-d.p.c. embryos before gonadal ridge formation. These observations have indicated that fetal expression of Sry is limited to the period in which testicular cord formation be- gins. Using in situ hybridization they have also shown that the expression of Sry in 11.5-d.p.c. embryos is confined to gonadal tissues. This was substantiated by the RT-PCR analysis of RNA from various parts of 11.5-d.p.c. embryos: there was no evidence for expression in any tissue other | than urogenital ridges. However, they also failed to detect the Sry transcripts present in 11.5-d.p.c. gonadal ridge or 12.5-d.p.c. testes by Northern blotting. This failure was ascribed to the low level of expression. In contrast to Zfy-1 expression, W‘/W* mouse fetuses with testes lacking germ cells have proved indistinguishable from wild type in RT-PCR analyses of Sry expression in gonadal ridge at 11.5 d.p.c. On the other hand, Sry expression in adult testis was found to be depend- ent on germ cells. The best way to test the function of Sry is to inject it into XX embryos, and to see if they develop as males. Very recently, Koopman et al. [8] have demonstrated that Sry gives rise to normal testis development in chromosomally female trans- genic mice. Transgenic mice were produced by microinjecting 14 kb genomic fragment containing Primary Sex Determination in Mammals 491 the conserved motif of Sry into fertilized eggs and then transferring to pseudopregnant foster mothers. Phenotypic sex was assayed at both embryonic and adult stage. Testicular cord forma- tion gives a characteristic stripped appearance to the developing testis, distinguishing it from the fetal ovary. Embryonic phenotypic sex was assayed by examining the appearance of the gonads in fetuses about 14 days after oviduct transfer. Chromosomal sex was determined by staining for sex chromatin in amnion cells and by Southern blot analysis using the Zfy probe. Of 158 embryos obtained, testes were found in two embryos whose sex chromatin indicated an XX sex chromosome constitution. Southern blot analysis showed that both of these males lacked Zfy sequ- ence and were transgenic, with many copies of Sry. Histological examination also revealed that their testicular cord formation was normal and that their gonads were indistinguishable from testes of nor- mal XY sib embryos. The examination of all the embryos scored as females by PCR identified two more mice as transgenic for Spy, indicating that not all XX transgenics showed sex reversal. The adult phenotype of Sry transgenic mice was ex- amined after some of embryos were allowed to develop to term. Of 93 animals that were born, three were identified by Southern blotting to be XX transgenics. One of them was a transgenic XX male: he had no Y chromosome and was externally male. he had a normal male reproductive tract with no signs of hermaphroditism, but was sterile with rather smaller size of testes than an XY control littermate as expected from the failure of germ cells in XX males to proceed beyond pro- spermatogonia. His copulation behavior was also normal. However, other two XX _ transgenics showed an external female phenotype, but carried many copies of Sry. They have produced offspring and so have functional reproductive tracts and ovaries. One of them transmitted the transgene to female offspring. This finding also provides evi- dence that the transgene does not always cause sex reversal. From these experiments, they conclude that a 14kb genomic fragment carrying Sry sequ- ence is sufficient to direct the formation of testes in XX transgenic embryos and subsequently to give rise to complete phenotypic sex reversal in a chromosomally XX transgenic adult. They have also produced three transgenic mice by injection of Y-specific DNA fragment carrying the SRY con- served domain. Two of them were XY founders that transmitted SRY to their offspring. Trans- gene expression was clearly demonstrated at the expected time in the developing gonads, but any sex reversal XX embryos was not observed. They suggested that differences in the sequence resulted in the human SRY protein failing to interact with other regulatory proteins or target genes in mouse cells. Tiersch et al. [135] investigated the phylogenetic conservation of the SRY gene by Southern blot analysis of restiction-digested DNA from 23 spe- cies representing five classes of vertebrate using a probe from the conserved motif. They found the presence of the SRY homologous sequences in all species tested, with and without sex chromosome and with temperature sex determination. But sex-specific signals were observed only in mam- mals. As described above, SRY gene seems to encode a DNA-binding protein (SRY protein) containing a high mobility group (HMG) box. Among known sequence-specific DNA-binding proteins, SRY protein is closely related to T cell-specific DNA-binding protein (TCF-1). Har- ley et al. [136] produced a recombinant SRY protein in both E. coli and insect cells, and tested for their binding to the TCF-1 target sequence AACAAAG and variants of this sequence. SRY protein bound to the target sequence in a sequence dependent manner. SRY protein of XY females with point mutation within the region encoding the HGM box showed negligible or reduced binding activities. These results support the view that the DNA-binding activity of SRY protein is required for testis determination. FUNCTION OF Tdy How Tdy acts to bring about testis determina- tion has been investigated independently of the cloning of Tdy. The gonad is composed of cells derived from the four cell lineages: the supporting cell lineage (Sertoli cells in the testis and granulosa cells in the ovary), the steroidogenic cell lineage (Leydig cells in the testis and theca cells in the 492 Y. NaGal ovary), the germ cell lineage and the connective tissue cell lineage. Of them the connective cell lineage contributes the development of both embryonic gonads although it adopts different architecture. The remaining three cell lineages form sex-specific cell types. There are several evidences indicating that differentiation of sup- porting cell lineage beyond pregranulosa cells de- pends on interaction with the germ cell population in the developing female gonad [137]. In contrast, it is known that germ cells do not contribute for testicular cord formation [14, 15]. The earliest sign of the differentiation in the embryonic indifferent gonad is the testicular cord formation. Studies on fetal rat testes by light and electron microscopy have revealed that the differentiation of testicular cell types occurs earlier than that of ovarian cell types, that is, a new cell type characterized by a large and clear cytoplasm appears in the gonads prior to testicular cords formation [13]. These cells (pre-Sertoli cells) aggregate; enclose germ cells and form the testicular cords. Accordingly, it seems reasonable to assume that Tdy is first ex- pressed in the ancestral supporting cell lineage. The production of chimaeric mice has been employed as one of the most useful techniques to analyse the interaction of cells of two or more genotypes in the development of a single organism in many areas of biology. One of the first ques- tions examined with chimaeras produced by ran- dom fusions between two blastocysts was that of sex determination [138]. In a balanced combina- tion, although on average half of the chimaeras should be XX@ XY combination, true hermaphro- ditism is rare and most XX@XY chimaeras (70- 80%) develop as phenotypic male with testes [139]. These observations and two reports of analysis on somatic cells in testes of XX@XY chimaeric mice in which XX cells contributed to both the Sertoli cells and Leydig cells [140, 141] lent support to the view that the initial stages of testes organization are orchestrated by a locally diffusible testis-organizing molecule which is con- trolled by Tdy and to which both XX and XY cells can respond. However, recent studies on XX“ XY chimaeric mice have provided evidences against the involvement of diffusible _ testis- organizing molecule. Burgoyne et al. [142] showed that when embryonic XX gonadal tissues were cocultured or cografted under the kidney capsule with develop- ing testis, the germ cells and somatic cells of the XX gonad did not organize into testicular cords. Bradbury [143] observed that in most KK@XY mouse chimaeras fetal gonads initially developed as ovotestes following regression of the ovarian portion in the more advanced fetuses. He has argued that in XX@XY chimaeras the initial ovotestes are converted to testes through the re- gression of the ovarian tissue. The in situ hybrid- ization analysis with mouse Y-specific DNA probes provided direct means to estimate the proportion of XX and XY cells contributing to the major cell lineages of the gonads from sectioned and air-dried material. Recently, the correlation of gonadal sex with sex chromosome constitution of the major somatic cell lineages of the gonad was examined in more detail applying this new means. Burgoyne et al. [144] reported that in prepuberal and adult XX XY chimaeric mice the Sertoli cells and germ cells are exclusively XY cells while XX cells can contribute to the Leydig cells, the peri- tubular cells and the vasculized connective tissue of the tunica albuginea. They proposed that Tdy is expressed at the level of a single cell (cell- autonomously) in an initially bipotential support- ing cell lineage to bring about Sertoli cell dif- ferentiation, and that the commitment of the other . components of the testis to the male pathway is directed by the Sertoli cells without further Tdy involvement (cell-autonomous Y-action model) [145]. According to this proporsal, fetal XX XY gonads would be expected to consist of two patch- es of both XY Sertoli cells and XX granulosa cells, so that the gonads should initially differentiate as ovotestes. This is in agreement with the report of Bradbury. The cell-autonomous Y-action model also pre- dicts that granulosa cells should be exclusively XX in XX@XY female chimaeras. However, there is a report of two XX@XY female chimaeras in which XY cells could contribute to the granulosa cells [146]. Recently, Burgoyne et al. [147], and Palmer and Burgoyne [148] also found that XY cells could contribute granulosa cells in the ovary of XX@XY female chimaeras. Further, Palmer Primary Sex Determination in Mammals 493 and Burgoyne [149] have shown that a minor proportion of XX Sertoli cells is also present in testes of fetal, prepuberal and adult XX@XY chimaeras. Although the Sertoli cells were not exclusively XY cells, there was a strong XY bias that was already established at fetal stage. On the other hand, there was no XY bias in any of other cell types at all three stages. Patek et al. [150] also demonstrated in the gonads of a series of adult XX He Boyse; Es Aq; Bennet; Db”, Scheid, M. and Carswell, E. A. (1971) Nature, 232: 478-480. Scheid, M., Boyse, E. A., Carswell, E. A. and Old, L. J. (1972) J. Exp. Med., 135: 938-955. Krco, C. J. and Goldberg, E. H. (1976) Science, 193: 1134-1135. Wachtel, S. S., Koo, G. C., Zuckermann, E. E., Hammerling, U., Scheid, M. and Boyse, E. A. (1974) Proc. Natl. Acad. Sci. (USA), 71: 1215- 1218. Wachtel, S. S., Koo, G. C. and Boyse, E. A. (1975) Nature, 254: 270-272. Moscona, A. (1957) Proc. Natl. Acad. Sci. (USA), 43: 184-193. Wachtel, S. S., Ohno, S., Koo, G. C. and Boyse, Ee Aca 1975) Natures 2575 235—230; Ohno, S. (1985) Endocrine Rev., 6: 421-431. Nagai, Y. (1981) Cong. Anom., 21: 465-476 (in Japanese). Wachtel, S. S. (1983) H-Y antigen and the Biology of Sex Determination. Grune & Stratton, Inc. New York, pp. 82-98. Ohno, S., Nagai, Y. and Ciccarese, S. (1978) Cytogenet. Cell Genet., 20: 351-364. Zenzes, M. T., Wolf, U., Gunther, E. and Engel, W. (1978) Cytogenet. Cell Genet., 20: 365-372. Taylor, R. B., Duffs, W. P. H., Raff, M. C. and de Petris, S. (1971) Nature New Biology, 233: 225- 229) Beutler, B., Nagai, Y. 13: 509-513. Nagai, Y., Ciccarese, S. and Ohno, S. (1979) Differentiation, 13: 155-164. Koo, G. C. (1981) Hum. Genet., 58: 18-20. Bradley, M. P., Ebensperger, C. and Wiberg, U. H. (1987) Hum. Genet., 76: 352-356. Nagai, Y. (1984) In “Experimental Biology Series 13 ; Genetic Biology”. Ed. by K. Moriwaki Maruzen, Tokyo, pp. 34-48 (in Japanese). Kohler, G. and Milstein, C. (1975) Nature, 256: 495-497. and Ohno, S. (1978) Cell, 496 30) 56 S// 58 59 60 61 62 63 64 65 66 67 68 69 70 ial 72 18 74 Ts 76 77 78 79 80 Koo, G. C., Tada, N., Chaganti, R. and Hammer- ling, U. (1981) Hum. Genet. 57: 64-67. Farber, C-M., Wachtel, S. S. and Cunningham- Rundles, C. (1982) Clin. Exp. Immunol. 50: 450- 453. Meck, J. M. and Goldberg, E. H. (1984) J. Immunol. Meth., 73: 293-299. Booman, P., Kruijt, L., Tieman, M., Piedrahita, J. A., Veerhuis, R., de Boer, P. and Ruch, F. E. (1989) J. Repro. Immunol., 15: 195-205. Goldberg, E., Boyse, E. A., Scheid, M. and Ben- nett, D. (1972) Nature New Biology, 238: 55-57. Silvers, W. K. and Billingham, R. E. (1967) Scien- ce, 158: 118-119. Melvold, R. W., Kohn, H. I., Yerganian, G. and Fawcett, D. W. (1977) Immunogenetics, 5: 33-41. Engel, W., Klemme, B. and Erbrecht, A. (1981) Hum. Genet., 57: 68-70. Silvers, W. K., Gasser, D. L. and Eicher, E. M. (1982) Cell, 28: 439-440. Goldberg, E. H., Shen, F. and Tokuda, S. (1973) Transplantation, 15: 334-336. Simpson, E. and Gordon, R. T. (1977) Immunol. Rev., 35: 59-75. Tomonari, K. (1983) J. Immunol., 131: 1641- 1645. Zinkernagel, R. M. and Doherty, P. C. (1974) Nature, 251: 547-548. Von Boehmer, H. (1990) Eur. J. Biochem., 194: 693-698. Braciale, T. J. and Braciale, V. L. (1991) Im- munol. Today, 12: 124-129. Lindahl, K. F. (1991) Trends in Genetics (TIG), 7: 219-224. Simpson, E., Chandler, P. and Pole, D. (1981) Cell. Immunol., 62: 251-257. R6tzschke, O., Falk, K., Wallny, H-J., Faath, S. and Rammensee, H-G. (1990) Science, 249: 283- 287. Griem, P., Wallny, H-J., Falk, K., R6tzschke, O., Arnold, B., Schonrich, G., Hammerling, G. and Rammensee, H-G. (1991) Cell, 65: 633-640. Goldberg, E. H. (1988) Phil, Trans. R. Soc. Lond. Brs22275— ole Cattanach, B. M., Pollard, C. E. and Hawkes, S. G. (1971) Cytogenetics, 10: 318-337. Singh, L. and Jones, K. W. (1982) Cell, 28: 208- 216. Evans, E. P. and Burtenshaw, M. D. and Catta- nach, B. M. (1982) Nature, 300: 443-445. Burgoyne, P. S. (1982) Hum. Genet., 61: 85-90. Rouyer, F., Simmler, M-C., Johnsson, C., Verg- naud, G., Cooke, H. J. and Weissenbach, J. (1986) Nature, 319: 291-295. Bennett, D., Mathieson, B. J., Scheid, M. Yanagi- Y. NAGAI 81 82 83 84 85 86 87 88 89 90 il 2 93 94 95 96 D7 98 99 100 101 102 103 104 sawa, K., Boyse, E. A., Wachtel, S. and Catta- nach, B. M. (1977) Nature, 265: 255-257. Simpson, E., Edwards, P., Wachtel, McLaren, A. and Chandler, P. (1981) munogenetics, 13: 355-358. Shapiro, M., Erickson, R. P. Lewis. S. and Tres, L. L. (1982) J. Repro. Immunol., 4: 191-206. McLaren, A. and Monk, M. (1982) Nature, 300: 446-448. Cattanach, B. M. (1974) Genet. Res., 23: 291- 306. Simpson, E., McLaren, A., Chandler, P. and Tomonari, K. (1984) Transplantation, 37: 17-21. McLaren, A., Simpson, E., Tomonari, K., Chand- ler, P. and Hogg, H. (1984) Nature, 312: 552-555. Simpson, E., Chandler, P., Goulmy, E., Disteche, C. M., Ferguson-Smith, M. A. and Page, C. D. (1987) Nature, 326: 876-878. Mandel, J-L., Willard, H. F., Nussbaum, R. L., Romeo, G., Puck, J. M. and Davies, K. E. (1989) Cytogenet. Cell Genet., 51: 384-437. Ohno, S. (1967) Sex Chromosomes and Sex-linked Genes. Springer-Verlag. Berlin, pp. 5-24. Burgoyne, P. S. (1986) Nature, 319: 258-259. Ellis, N. and Goodfellow, P. N. (1989) TIG, 5: 406-410. Tiepolo, L. and Zuffardi, O. (1976) Hum. Genet., 34: 119-124. Kunkel, L. M., Smith, K. D. and Boyer, S. H. (1976) Science, 191: 1189-1190. Cooke, H. (1976) Nature, 262: 182-186. Bostock, C. J., Gosden, J. R. and Mitchell, A. R. (1978) Nature, 272: 324-328. Baron, B., Métézeau, P., Hatat, D., Roberts, C., Goldberg, M. E. and Bishop, C. (1986) Somatic Cell Mol. Genet., 12: 289-295. Bishop, C. E., Guellaen, G., Geldwerth, D., Voss, R., Fellous, M. and Weissenbach, J. (1983) Na- ture, 303: 831-832. Singh, L., Purdom, I. F. and Jones, K. W. (1976) Chromosoma, 59: 43-62. Singh, L., Purdom, I. F. and Jones, K. W. (1981) Cold Spring Harb. Symp. Quant. Biol., 45: 805- 813. Epplen, J. T., McCarrey, J. R., Sutou, S. and Ohno, S. (1982) Proc. Natl. Acad. Sci. (USA), 79: 3798-3802. Singh, L., Phillips, C. and Jones, K. W. (1984) Cell, 36: 111-120. Singh, L. and Jones, K. W. (1986) Hum. Genet., 73: 304-308. Kiel-Metzger, K., Warren, G., Wilson, G. N. and Erickson, R. P. (1985) New Eng. J. Med., 313: 242-245. Ferguson-Smith, M. A. (1966) Lancet, If: 475- SyeS3: Im- 105 106 107 108 109 110 id 112 ILS) 114 HIS) 116 iv) 118 119 120 121 122 123 124 125 126 127 Primary Sex Determination in Mammals 476. Page, D. C. and de la Chapelle, A. (1984) Am. J. Hum. Genet., 36: 565-575. Werenaud: G- Page. D: C.;) Simmier”M-C., Brown, L., Rouyer, F., Noel, B., Botstein, D., de la Chapelle, A. and Weissenbach, J. (1986) Am. J. Hum. Genet., 38: 109-124. Page, D. C., Mosher, R., Simpson, E. M., Fisher M. C., Mardon, G., Pollack, J., McGillivray, B., de la Chapelle, A. and Brown, L. G. (1987) Cell, 51: 1091-1104. Miller, J., McLachlan, A. D. and Klug, A. (1985) EMBO J., 4: 1609-1614. Nothiger, R. and Steinmann-Zwicky, M. (1987) Development, 101 (Suppl.): 17-24. Hodgkin, J. (1987) Development, 101 (Suppl.): 5— 16. Hodgkin, J. (1990) Nature, 344: 721-728. Schneider-Gadicke, A., Beer-Romero, P., Brown, L. G., Nussbaum, R. and Page, D. C. (1989) Cell, 57: 1247-1258. Mardon, G., Mosher, R., Disteche, C. M. Nishioka, Y., McLaren, A. and Page, D. C. (1989) Science, 243: 78-80. Nagamine, C. M., Chan, K., Kozak, C. A. and Lau, Y-F. (1989) Science, 243: 80-83. Klug, A. and Rhodes, D. (1987) TIBS, 12: 464- 469. Erickson. R. H. and Verga, V. (1989) Am. J. Hum. Genet., 45: 671-674. Koopman, P., Ashworth, A. and Lovell-Badge, R. (1991) TIG, 7: 132-136. ScmelaingAGcH.;)-Foster, J. W-,.Spencer, J. A.; Page D. C., Palmer, M., Goodfellow, P. N. and Graves, J. A. M. (1988) Nature, 336: 780-783. Bull, J. J., Hillis, D. M. and O’Steen, S. (1988) Science, 242: 567-569. Koopman, P., Gubbay, J., Collignon, J. and Lovell-Badge, R. (1989) Nature, 342: 940-942. de la Chapelle, A. (1988) Am. J. Hum. Genet., 43: 1-3. Palmer, M. S., Sinclair, A. H. Berta, P., Ellis, N. A., Goodfellow, P. N., Abbas, N. E. and Fellous, M. (1989) Nature, 342: 937-939. Kelly, M., Burke, J., Smith, M., Klar, A. and Beach, D. (1988) EMBO J., 7: 1537-1547. Einck, L. and Bustin, M. (1985) Exp. Cell Res., 156: 295-310. Page, D. C., Fisher, E. M. C., McGillivray, B. and Brown, L. G. (1990) Nature, 346: 279-281. Berta, P., Hawkins, J. R., Sinclair, A. H., Taylor, A., Griffiths, B. L., Goodfellow, P. N. and Fel- lous, M. (1990) Nature, 348: 448-449. Jager, R. J., Anvret, M., Hall, K. and Scherer, G. (1990) Nature, 348: 452-454. 128 129 130 131 132 133 134 135 136 7) 138 139 140 141 142 143 144 145 146 147 148 497 Sutcliffe, M. J. and Burgoyne, P. S. (1989) De- velopment, 107: 373-380. Burgoyne, P. S., Levy, E. R. and McLaren, A. (1986) Nature, 320: 170-172. Mitchell, M. J., Woods, D. R., Tucker, P. K., Opp. J..S. and Bishop, C. E. (1991) Nature, 354: 483-486. Kay, G. F., Ashworth, A., Penny, G. D., Dunlop, M., Swift, S., Brockdorff, N. and Rastan, S. (1991) Nature, 354: 486-489. Lovell-Badge, R. and Robertson, E. (1990) De- velopment, 109: 635-646. Gubbay, J., Koopman, P., Collignon, J., Bur- goyne, P. and Lovell-Badge, R. (1990) Develop- ment, 109: 647-653. Koopman, P., Miinsterberg, A., Capel, B., Vivian, N. and Lovell-Badge, R. (1990) Nature, 348: 450- 452. Tiersch, T. C., Mitchell, M. J. and Wachiel, S. S. (1991) Hum. Genet., 87: 571-573. lanley. Ve) Re Jackson) D> I Hextall- Pee, Hawkins, J. R., Berkovitz, G. D., Sockanathan, S., Lovell-Badge, R. and Goodfellow, P. N. (1992) Science, 255: 453-456. McLaren, A. (1991) BioEssay, 13: 151-156. Tarkowski, A. T. (1964) J. Embryol. Exp. Mor- phol., 12: 735-757. McLaren, A. (1984) In “Chimeras in Developmen- tal Biology”. Ed. by N. Le Douarin and A. McLaren, Academic Press, London, pp. 381-399. Mintz, B. (1969) In “First Conferences on the Clinical Delineation of Birth Defects” series S., Ed. by D. Bergsma and V. McKusick, National Foundation, New York, pp. 11-22. Ohno, S., Ciccarese, S., Nagai, Y. and Wachtel, S. S. (1978) Arch. Androl., 1: 103-109. Burgoyne, P. N., Ansell, J. D. and Tournay, A. (1986) In “Development and Function of the Reproductive Organs”. Ed. by A. Eshkol, B. Eck- stein, N. Dekel, H. Peters and A. Tsafriri, Ares- Serono Symposia, Rome, pp. 23-39. Bradbury, M. W. (1987) Develop. Genet., 8: 207- DNS: Burgoyne, P. © , Buehr, M., Koopman, P., Ros- sant, J. and McLaren, A. (1988) Development, 102: 443-450. Burgoyne, P. S. (1988) Phil. Trans. R. Soc. Lond. Bey 32220372. ROKGh Cres Vans-ee ob. burtenshaws Me Dr Clegg, H., Barnes, R. D. and Tuffrey, M. (1974) Differentiation, 2: 321-333. Burgoyne, P. S., Buehr, M. and McLaren, A. (1988) Development, 104: 683-688. Palmer, S. J. and Burgoyne, P. S. (1991) Develop- ment, 111: 1017-1019. 498 149 150 151 152 153 154 1595) Palmer, S. J. and Burgoyne, P. S. (1991) Develop- ment, 112: 265-268. Patek, C. E., Kerr, J. B., Gosden, R. G., Jones, K. W., Hardy, K., Muggleton-Harris, A. L., Handyside, A. H., Whittingham, D. G. and Hoop- er, M. L. (1991) Development, 113: 311-325. Ciccarese, S. and Ohno, S. (1978) Cell, 13: 643- 650. Vigier, B.; Picard) J:°¥:5 Gran, Di Leveai, IL. and Josso, N. (1984) Endocrionology, 114: 1315-1320. Byskov, A. G. and H@gyer, P. E. (1988) In “The physiology of Reproduction”. Ed. by E. Knobil, J. Neill, L. Ewing, L. Markert, S. Greenwald and D. Pfaff, Raven Press, New York, pp. 265-302. Vigier, B., Watrin, F., Magre, S., Tran, D. and Josso, N. (1987) Development, 100: 43-55. Charpentier, G. and Magre, S. (1990) Develop- Y. NAGAI 156 157, 158 159 160 161 162 ment, 110: 839-849. Behringer, R. R.; Cate; RivL., Froelick iG, ge Palmiter, R. D. and Brinster, R. L. (1990) Nature, 345: 167-170. Crumeyrolle-Arias, M. and Ascheim, P. (1981) Gerontology, 27: 58-71. Taketo-Hosotani, T., Merchant-Larios, H., Thau, R. B. and Koide, S. S. (1985) J. Exp. Zool., 236: DIY Taketo-Hosotani, T. and Scinclair-Thompson, E. (1987) Develop. Biol., 124: 423-430. Hashimoto, N., Kubokawa, R., Yamazaki, K. Noguchi, M. and Kato, Y. (1990) J. Exp. Zool. 253: 61-70. Eicher, E. M. and Washburn, L. L. (1986) Ann. Rev. Genet., 20: 327-360. McLaren, A. (1990) Nature, 346: 216-217. ZOOLOGICAL SCIENCE 9: 499-513 (1992) © 1992 Zoological Society of Japan REVIEW C-reactive Protein (CRP) in Animals: Its Chemical Properties and Biological Functions WATARU NUNOMURA Nippon Bio-Test Laboratories. Inc. and Tumour Laboratory Kokubunji, Tokyo 185, Japan ABSTRACT—C-reactive protein (CRP), which was found in the sera of pneumococci in 1930 and has been used as a marker protein for inflammation in clinical laboratories, has been isolated as well from the sera of fishes and body fluids of invertebrates. Taking it into consideration that the evolutionary origin of this protein may be early enough, the CRP is supposed to act an important role. Recent studies on CRP have revealed the biological functions of this protein. macrophages make up an amplification mechanism in primary stage of immune response. For example, CRP and On the other hand, it was revealed that the serum level of CRP was controlled by not only interleukin(s) but also sex hormone(s). INTRODUCTION The animals, in their serum, have les jumellaux that function in the first stage of defence system, being neither immunoglobulins nor proteases such as complements. They are C-reactive protein (CRP) and serum amyloid P component (SAP). Both have similar physicochemical properties and electronmicroscopically a cyclic pentamer form. Although CRP has been used as a marker protein for inflammation in clinical laboratories, its evolu- tionary conservation suggests an important role of this protein. Recent studies have been revealed immunological functions of CRP, for example activation of marcophages and tumoricidal activ- ity. Furthermore, CRP solubilizes endogenous substances derived from damaged cells such as chromatin cooperating with complements. In this article, I would like to summerize the chemistry of CRP and discuss its biological and immunological functions, focusing the ex- perimental results with rats and japanese eel (Anguilla japonica) in our laboratory. Received March 5, 1992 SHORT HISTORY OF STUDIES ON CRP In 1930, Tillett and Francis observed that sera of patients during acute febrile illnesses produced a precipitation with a component in the extract of pneumococcus, first designated Fraction C and later C-polysaccharide [1]. The substance re- sponsible for this reaction was termed “C- precipitin”. In 1941, Abernathy and Avery [2] showed that the substance bound to C- polysaccharide in the membrane of Streptococcus pneumoniae in the presence of calcium ion, and the reaction was inhibited by phosphorylcholine (PC). The authors designated the substance as “C-reactive protein”. In 1947, McCarty succeeded in crystalization of CRP and raising specific anti- body in rabbit [3]. Since 1947, the serum level of CRP has been measured by the immunoassay, for example single radial immuodiffusion and/or en- zyme-immunoassay using specific antibody [4-6]. Gotshclich and Edelman observed that CRP con- sisted of 24 kDa of five identical subunits associ- ated hydrophobically [7]. In 1977, an electron- microscopic observation by Osmand et al. demon- strated the shape of CRP being cyclic pentamer [8]. The authors proposed in their report that CRP and 500 W. NUNOMURA SAP were termed “pentraxin”. The interaction of CRP and complements has been studied since 1974. Once CRP has com- plexed with PC residues of phospholipids such as lecithin and sphingomyelin, the complex activates the classical pathway [9]. In 1980, DiCamelli et al. reported the binding of CRP to polycations [10]. The in vitro experiments clarified that the CRP bound to nuclear proteins, such as histone and protamine, more strongly than PC [11-13]. CRP seems to relate to the solubilization of chromatin following complement activation by CRP- chromatin complex [14]. The tumoricidal effect of human CRP has been studied since 1982. It was observed that human CRP inhibited metastases of melanoma to lung in mouse [15-17]. Several reports described that tumor cells were killed by human CRP, having mouse macrophages to produce superoxide-anion in vitro [18-21]. The CRPs are found in the sera of some verte- brates including fishes, for example rabbit [22, 23], bovine [24], dog [25], goat [26], plaice (Pleuro- nectes platessa) a marine teleost [27], rainbow trout (Salmo gairdneri) [28, 29], and dogfish (Mustelus canis) [30]. Almost phylogenic studies on CRP have dealed with its purification and physicochemi- cal analysis, but there are few studies on its biological activity. However, in 1981, a CRP was isolated from the body fluid of horseshoe crab (Limulus polyphemus) and its lectin activity was studied by Robey and Liu [31]. In our laboratory, a CRP was isolated from the serum of japanese eel (Anguilla japonica) {32] and the CRP analogue from the serum of female whiteedged rockfish (Sebastes taczanowskii), a typical viviparous marine teleost [33]. It was demonstrated that these molecules have also lectin activity. A topological analysis of CRP from hoseshoe crab, human, rabbit and human SAP indicated that these proteins may originate from the same ancestral gene [34]. Since 1982, the epitopes of human CRP have been analysed by using mouse monoclonal anti- body by Kilpatrick et al. [35] and Roux et al. [36]. CRP has at least two different epitopes; the one is locted near or at the PC binding site, the con- formation of which is changed by chelation with calcium ion. In 1987, Potempa et al. [37] observed that the subunit of CRP, termed “neo-CRP”, had a different antigenecity from the native CRP mol- ecule. However, there is few reports on the analysis of epitope(s) and biological significance of “neo-CRP” in other animals [38]. CHEMICAL APPROACH TO CRP Purification of CRP The ligand specificity of CRP and SAP in the presence of calcium ion differes completely [39]. The CRP specifically binds to PC and the SAP to agarose, respectively. The CRP can be specifically precipitated by being mixed with PC or C- polysaccharide in the presence of calcium ion. CRP in the serum of rat or eel was purified by the lecithin precipitation method described by Hoka- ma et al. [40]. In brief, serum was mixed with soy-bean lecithin and then the mixture was di- alyzed against 0.01 M CaCl,. The precipitate was dissolved in sodium citrate buffer, pH 7.0 and then chloroform was add to the mixture. After cen- trifugation, the upper aqueous layer was dialyzed - against CaCly. The precipitate was again dissolved in sodium citrate buffer pH 7.0. Further purifi- cation was carried out chromatography on DEAE- Sephacel and gel filtration on S-300. The purifi- cation procedure was previously reported in detail [32, 41]. Recently, a new method for purification of CRP in rat serum has been developed in our laboratory. Rat sera were brought to between 209 g/l and 409 g/l of ammonium sulfate at pH 7.2. The precipitate was extensively dialysed against 10 mM Tris-HCl buffer, pH 8.0 containing 0.14 M NaCl and 10mM CaCl, (Buffer A) and then applied to a column of Sepharose 4B loaded with 100 mg of protamine (derived from salmon sperm, grade X, Sigma). Washing the column with 10 mM Tris-HCI buffer, pH 8.0 containing 1.14 M NaCl, bound CRP was eluted with 10mM Tris-HCl buffer, pH 8.0 containing 1.14 M NaCl and 0.5 M arginine-HCl. Eluted fraction was dialysed against Buffer A and then applied to a column of Argi- nine-Sepharose 4B (Pharmacia). The CRP was eluted with 50mM sodium citrate buffer pH 7.0 containing 0.2 M NaCl. Further purification was CRP in Animals 501 carried out by gel filtration on S-300 [42]. The recovery by the new method was higher than that by lecithin-precipitation. On the other hand, PC- or C-polysaccharide-Sepharose affinity column chromatographies are widely used for the purifi- cation of CRP [43, 44]. However, not only the coupling of PC with Sepharose is rather difficult, but also a sufficient amount C-polysaccharide for an affinity column is not easy to obtain [45]. We tried to purify CRP from rat serum using the PC-Sepharose, however, many contaminating pro- teins of a similar characteristics as CRP appeared. De Beer et al. succeeded in purification of rat CRP by C-polysaccharide-Sepharose affinity chroma- tography [46]. Using the PC-Sepharose, the CRPs were actually purified from sera of human, rabbit, goat [47], white-edged rockfish [33] and eel in our laboratory. Figure 1 shows a cellulose-acetate membrane electrogram of purified human and eel CRP. In addition, using Sepharose coupled with mouse monoclonal antibody, human CRP was highly purified without destroying its hydrophobic association [48]. Perse hCRP ES eCRP NHS Fic. 1. Cellulose-acetate membrane electrophoresis of human and eel CRP. NHS, normal human serum; hCRP, human CRP; ES, eel serum; eCRP, eel CRP. Anode is top. Physicochemical chracteristics of CRP The molecular weight of rat CRP was 165 kDa determined by gel filtration and that of its subunits - H R 4H 94> : 94> ors 67 43> 485 eo 30> ee = alia BESS 20:1> 201> , 144> 144> Fic. 2. SDS-PAGE of CRP under reducing (A) with 2-mercaptoethanol and non-reducing (B) condi- tions. R and H represent rat and human CRP, respectively. Marker proteins are as follows: rabbit muscle phosphorylase b (94kDa), bovine serum albumin (67kDa), ovalbumin (43 kDa), bovine erythrocyte carbonic anhydrase (30 kDa), soybean trypsin inhibitor (20.1 kDa) and bovine milk a- lactalbumin (14.4 kDa).O, origin. Each CRP sam- ple (20 ug) were run in plural lanes. was 29 kDa estimated by SDS-PAGE (Fig. 2). On SDS-PAGE, without reducing by 2-mercap- toethanol (2ME), CRP appeared as two bands, 27 kDa and 53.5kDa. The results indicate that the rat CRP has a disulfide bond between in its sub- units [41, 46]. By gel filtration and SDS-PAGE, eel CRP was found to consist of 24kDa five identical subunits [32]. No disulfide bonds among the subunits was detected in the eel CRP molecule as well as CRPs of human, rabbit and goat. The electrophoretical mobilities of CRPs dif- fered each other among the animal species. In electrophoresis at pH 8.6, rat CRP [41] and eel CRP moved near the region of albumin. In contrast human CRP moved to the region of gamma globulin as shown in Fig. 1. The CRPs of rabbit and goat moved as slowly as the latter CRP. The isoelectric points of rat CRP were 4.29, 4.22, 4.21 and 4.16 determined by the thin layer polyacrylamide isoelectricfocusing. In O’Farrell’s two dimension electrophoresis [49], the first dimension being isoelectricfocusing in polyacryl- amide gel containing 9.6 M urea and the second dimension being SDS-PAGE, rat CRP gave two spots of pH 5.3 and 5.4, each molecular weight 502 W. NUNOMURA IEF pH 65 6.0 5.55.0 4.5 a A © O. D Fic. 3. O’Farrell’s two dimension electrophoresis of rat CRP. The first dimension is isoelectricfocusing containing 9.6M urea in polyacrylamide and the second is SDS-PAGE in gradient gel. being 27 kDa (Fig. 3). Since rat CRP could move in a gel containing high concentration of urea, hydrophobic binding of CRP would disociate. The results suggest that the isoelectric points of the subunits of rat CRP were different from the native pentameric molecule. AGGLUTINATING ACTIVITY OF CRP CRPs of rat, eel and white-edged rockfish strongly agglutinate Streptococcus pneumoniae in the presence of calcium ion and the agglutinting activity is inhibited by PC or chelation of calcium ion (EDTA addition), as in the case of human CRP. However, bacteria having “Type 4” polysac- charide which does not contain PC residue were agglutinated by human CRP [50]. The CRP of horseshoe crab agglutinated horse red blood cells and the activity was inhibited sialic acid [31]. The result suggests that the CRPs of animals in lower classes may have a lectin activity. Actually, eel CRP strongly agglutinated rabbit red blood cells in the presence of calcium ion (Fig. 4). This agglutinating activity was inhibited by D- glucosamine and D-mannose [32]. A lectin purified from eel serum [51] agglutinated human H(O) type red blood cells via fucose residue [52]. The physicochemical and biological characteristics A BO zeae int Gp Ne Ge Fic. 4. Agglutinating activity of eel and human CRP in the presence of calcium ion (GVB**). Lanes 1-3 are eel serum, eel CRP and human CRP, respective- ly. The final concentration of each CRP was 20 yg/ ml. Abbreviations A, B, O, AB, human blood cell types; Rb, rabbit; Rt, rat; Gp, gunia pig; Ho, horse; Sh, sheep; Gs, goose red blood cells. Fresh blood cells were used for this experiment. of the lectin completely differ from those of eel CRP. Furthermore, CRP purified from the serum of female white-edged rockfish also agglutinated rabbit red blood cells and the activity was inhibited by D-glucosamine and _ N-acethyl-D-galacto- samine. In contrast, rat and human CRP could not agglutinate any red blood cells. Uhlenbruck et al. reported that human CRP formed a precipitin line with galactan in haemolymph of Helix pomatia and Octopus vulgaris in agarose gel containing calcium ion [53, 54]. Soelter and Uhlenbruck described that the anti-galctan activity of the human CRP was attributed to its anti-PC activity and not to the lectin activity [55]. However, it is possible to consider that the lectin-like activity remains in the human CRP. Fish CRPs belong to a “lectin family”, but rat and other mammals’ ones do not. The animals in low classes do not have active immunoglobulins, such as immunoglobulin G (IgG). Fish, for example chum salmon (Oncorhynchus keta) have only one class of immunoglobulin M (IgM) [56]. The results suggest that the CRP may possibly be one of the major defence substances in animals of lower classes. The role of CRP has been changed along the development of the defence system through the evolution of animals. Two lectins were purified from the plasma of ascidia Didemnum candidum, termed DCL-I and DCL-II. The N-terminal amino acid sequence of DCL-I showed up to six identities in a stretch of 19 residues with human CRP [57]. In addition, DCL-I cross-reacted in CRP in Animals 503 TABLE 1. Biochemical Properties of CRPs human rat eel Molecular Weight 110 165 120 Gel filtration (S-300) (kDa) 24 29 24 SDS-PAGE (+2ME)” 24 i AS oo) 24 SDS-PAGE (—2ME) Shape pentamer pentamer pentamer Mobility Y a Albumin pH 8.6, ~=0.05 Isoelectricpoint 6.4” 4.16—4.29 N.D. N.D. 5.3+5.4 N.D. O’Farrell’s method” Binding activity phosphorylcholine Yes Yes Yes polycation Yes Yes Yes nuclear protein Yes BYIES Yes sugars galactan no D-glucosamine N.D. not determiend. *) 2-mercaptoethanol >) Potempa et al. [91] ° O'Farrell [49] enzyme-linked immunosorbent assays (ELISA) with antibodies made against human CRP. Although Sir M. Burnet long ago suggested that invertebrates lectins might be related to the early precursors of immunoglobulins found in verte- brates, little subsequent information has been obtained to support this hypothesis. Recent evi- dences of amino acid sequence, immunochemical cross-reaction and agglutination activity suggest that CRP seems to be ascribed to invertebrates lectins. Both eel CRP and human CRP reacted with polycations in agarose gel [32]. Robey et al. [14] reported a precipitation reaction of rabbit CRP with nucleosome core particles in agarose gel, and suggested a possibility of CRP for acting as a scavenger of chromatin fragments from damaged cells. As eel has only undevelopoed immune net work, it has not known whether the autoimmune response seen in mammals actually occur in eel as well. That is a very interesting problem of CRP in phylogeny. The biochemical properties of CRPs are summerized in Table 1. ANTIGENECITY OF CRP Rat CRP does not react with antiserum raised D-mannose Fic.5. Immunochemical cross-reactoin among the CRPs in agarose gel. A, rat CRP; B, human CRP; C, rabbit CRP; a, goat antiserum to rat CRP; b, goat antiserum to human CRP. against CRP of the other species (Oucterlony’s method, Fig. 5). In addition, mouse monoclonal antibodies to human CRP recognizing two differ- ent epitopes [58] also did not react with rat CRP in ELISA system. However, the immunochemical cross-reactions were observed among the human, rabbit and goat CRPs. Maudsley and Pepys re- ported that immunochemical cross-reactions among CRPs of 30 kinds of animals including lower classes [59]. They observed no immuno- chemically cross-reaction between rat CRP 504 W. NUNOMURA Le 1.2 3.4. 5-6 -7.8..a -b © d 2c. Samples Standards B 1.2%* r=0.99 y=-69 .64+143.1x Height (cm) w r=0.99 y=-60.4+124.2x CRP (mg/ml) Fic. 6. Rocket immunoelectrophoresis of rat CRP (A) and the standard curve of rat CRP for rocket immunoelec- trophoresis (B). The numbers (1-8) are samples of rat sera and a-f are standards of rat CRP. a, 0.402 mg/ml; b, 0.268 mg/ml; c, 0.134 mg/ml; d, 0.067 mg/ml; e; 0.05 mg/ml. In B, 1.2% and 1.8% represent the concentration of rabbit antiserum to rat CRP in agarose gel. and CRPs of other 9 kinds of mammals. Although the chemical characteristics of rat CRP are differ- ent from those of human, Taylor et al. reported that 45 residues of carboxyl terminal of rat CRP had 71.7% similarity to that of human CRP [60, 61]. Three dimensional conformation of both CRPs seem to be so different each other as to yield a common antigenecity. On the other hand, an antibody to eel CRP formed a precipitin line with the CRP analogue of white-edged rockfish [32, 33]. The precipitin line is spur, indicating a partial identical antigenecity between them. However, there are no reports on the cross-reaction among neo-CRP (subunit of CRP) of various animals. CRP AS A TYPICAL ACUTE PHASE REACTANT Serum level of CRP The serum level of rat CRP was measured by so-called “rocket immunoelectrophoresis” [62] as shown in Fig. 6, and by sandwich ELISA. CRP level of normal adult rat is approximately 0.5 mg/ ml, being 100-5000 times higher than other ani- mals, for example 100 ng/ml in human [4-6], 1.5 pg/ml in rabbit [63], 60 ug/ml in dog [25], 30 ug/ ml in rainbow trout [29], 1 ~g/ml in eel [32]. The concentration of CRP in the body fluid of the 4 & D 2 Days Fic. 7. Serum CRP levels in rats with chemically- induced inflammation. Four rats were injected intramuscularly with 1 ml of turpentine-oil on Day 0 (indicated by the arrow) and the serum levels were followed up to Day 5 by rocket immuno- electrophoresis. CRP in horseshoe crab was 40% of total protein [31]. The differences of serum level of CRP in various species may be caused by their adptation to the environment. CRP in chemical inflammation The serum level of CRP immediately increases after destruction of tissues, for example bacterial infection or intoxication of turpentine-oil injec- tion. Although this increase in inflammation was 1000-times in human and rabbit [39, 63], rat CRP elevated approximately 3-4 times after the injec- tion of turpentine-oil (Fig. 7). Rats have also plural acute phase reactants in the serum as well as human. In rat, the a,-acid glycoprotein (AGP) is a major acute phase reactant [64]. By immunohisto- chemical method, CRP was demonstrated in the cytoplasm of hepatocytes but not in macrophages in liver, named specially “Kupffer cells” (Fig. ac! @ Fic. 8. = Animals 505 8A) and thymus and spleen. It has been evi- denced by in vitro experiment that CRP localizes in hepatocytes but not lymphoid tissues. In rat, CRP was detected in the medium of primary culture of hepatocytes but not in that of lympho- cytes [65]. CRP was immunohistochemically detected on the surface of injured tissues (Fig. 8C) and the surface of white blood cells infiltrated to the site of the inflammation (Fig.8D). It is interesting whether lymphocytes have a specific receptor for CRP or not. Zeller et al. reported that human monocytes had CRP binding site distinct from IgG receptor [66, 67]. According to Tseng and Morten- sen, CRP binds to fibronectin in the presence of calcium [68]. Therefore, it is considered that CRP binds to white blood cells through the fibronectin on the cells. It is interesting to investigate the mechanism of binding of CRP and lymphocytes. D Immunohistochemical localization of CRP. (A) in hepatocytes of a rat injected with turpentine-oil 2 days in advance (X660), (B) in hepatocytes of a rat injected with CCl, (8 hr after injection), x 1000, (C, D) muscle tissues 2 days after the turpentine-oil injection, C; x 330 and D; x500. CRP deposits were found injured muscle and in intravenous space between mucle fibers (C) and in cytoplasm of infiltrated white blood cells (D). Specificity of antibody in rat CRP was examined by immunoelectrophoresis [41]. The avidin-biothin complex method was employed for the immunostaining. macrophages (Kupffer cells) or red blood cells. Note CRP was not detected in the normal muscle, liver 506 W. NUNOMURA A 0.6 2 04 E Sy ! 5 a2 0 B % 2000 : = 1500 cc) 5 S 1000 Oo 8 500 2 3 = a Days Fic. 9. Changes in serum levels of CRP (A), GOT (B) and AFP (C) in CCl, injected rats. CRP was determined by rocket immunoelectrophoresis, GOT by the method of Karmen [92] and AFP level by enzyme-immunoassay [77]. Figure 9 shows the changes of the serum level of CRP, GOT and a-fetoprotein (AFP) in CCl- intoxication. CRP markedly decreased to 0.03- 0.05 mg/ml within 2 to 3 days and returned to the initial level on day 7. GOT rapidly elevated on the Ist day after injection. The increasing serum AFP level indicates the appearence of regenerating cells in liver [69]. Immunohistochemical staining of CRP revealed that CRP appeared in the nuclei of CCly-intoxicated hepatocytes (Fig. 8B). These results confirm that rat CRP actually bind to the nuclei in vivo. Furthermore, the CRP in rat plays a role as a scavenger of endogenous material such as nuclei derived from damaged cells [41]. The changes in serum levels of CRP during chemical carcinogenesis were observed. The rats (Donryu strain, male) were fed on a diet contain- ing 0.06% 3°-methyl-4-dimethylaminoazobenzene (3'-MeDAB) for 10 weeks and then the normal diet until sacrifice [70]. CRP level did not change remarkably through the first 8 weeks, but signif- icantly decreased to about half level of normal (0.28 mg/ml) in the 10th week. Then it gradually increased until the 15th week (Fig. 10). Immuno- histochemically, CRP was strongly stained in the non-cancerous area but not in the cancer cells [41]. CRP is able to bind to damaged cells but not intact cells including malignant cells except some kinds of lymphocytes. It seems unlikely that CRP itself plays an important role in carcinogenesis by affect- ing the metabolism of the azo-dye. Onoe et al. [70] observed that normal heptocytes gradually dis- CRP (mg/ml) AFP (ln ng/ml) Weeks Fic. 10. Changes in serum levels of CRP (A) and AFP (B) during hepatocarcinogenesis. Sixteen rats were fed with diet containing 3;Me-DAB for 10 weeks, followed by normal diet. The CRP was determined by rocket immunoelectrophoresis and AFP by en- zymeimmunoassay [77]. Results are expressed as mean+S.E. of 16 rats. Note that AFP level is represented In ng/ml. CRP in Animals 507 appeared in the first 10 weeks after 3-MeDAB ingestion, being replaced by regenerating hepato- cytes and a new type cells (named “oval cells”). In the present study, the decrease of serum CRP was found after the 9th to 10th week when the liver was occupied by the regenerating cells. Such regener- ating cells do not produce CRP. Summerizing these observations, CRP is thought to be produced by normal adult hepatocytes, but neither by re- generating hepatocytes nor cancer cells. INTERACTION OF CRP WITH MACROPHAGES Immunohistochemical observation of rat CRP indicates that there may be some interactions between CRP and lymphocytes. Several reports demonstrated the relationship between CRP and macrophages [16-21]. Rat CRP is somewhat differ- ent in structure and the serum level from human as described above. In rat, the interaction of CRP and lymphocytes was investigated in vitro and in vivo. Although CRP was detected in the culture medium of hepatocytes, the concentration of CRP gradually decreased with time. By adding the culture medium of macrophages obtained from rats eli- cited by an intraperitoneal injection of glycogen, the sysnthesis of CRP was sustained. However, the culture of thymocytes or splenocytes obtained from rat injected with turpentine-oil, yielded no such effect (Fig. 11). CRP was detected neither in the culture medium of each lymphocyte nor macro- phages culture. The effect of the culture medium of macrophages on the serum level of CRP in 10-day-old rats was observed [65]. The serum level of CRP was significantly increased by injecting the culture medium of macrophages com- pared with rats injected with plane culture medium or no treatment [65]. The synthesis of CRP in heptocytes is initiated by interleukin (IL)-6 pro- duced by macrophages. In preliminary examina- tion, a recombinant human IL-6 strongly enhanced a production of CRP by rat hepatocytes in vitro, but not recombinant human IL-1. When !*I labeled recombinant human IL-6 was injected into rat, the radioactivity was localized in hepatocyts. [71]. It was demonstrated that the acute phase 50 a0 30] 20 CRP (ng/m1/10°cells/24hr) 10 N T Effect of culture media of lymphocytes on the production of CRP by the primary culture of rat hepatocytes. N, none (control); T, thymocytes; S, splenocytes and M, macrophages. The hepatocytes were cultured for 54 hr. The concentration of CRP in culture media were determined by enzyme- Fic. 11. immunoassay [77]. Thymocytes and splenocytes were obtained from rats 24hr after njection of turpentine-oil, and macrophages obtained from rats 5 days after an intraperitoneally injection of 5% glycogen. Vertical bars represent S.D. reactants, for example AGP, a -macroglobulin and albumin in rat, were controled by recombinant human IL-6 [72, 73]. However, it has not been known whether CRP in fishes are also controlled by soluble factor(s) derived from macrophages such as IL-6 or not. Release of superoxide anion from rat macro- phages increased by adding CRP to the culture of macrophages [65]. The findings, including immunohistochemically localization of CRP, may suggest that CRP serves as a physiolosical macro- phage activator, contributing to the acceleration of a nonspecific host resistance in inflammatory response. In its subunits, human CRP has an analogous structure to “Tuftsin” which is a peptide consisted of four amino acid, THR-LYS-PRO- ARG, and strongly activates macrophages [74, 508 W. NUNOMURA 75]. Although the complete amino acid sequences of rat CRP has not been known, 45 carboxyl terminal residues were determined [61]. The rat CRP also has a Tuftsin analogue sequence, ILE- LYS-PRO-GLN. The results would indicate that CRP may be a source of the macrophage- activating peptide. I propose that CRP locally gathering around the site of inflammation is bound to macrophages, and then by digestive protease(s) of the cells a small peptide such as Tuftsin is derived from subunits of CRP to activate the macrophages. Actually, Robey et al. reported that the peptide(s) produced by the proteolysis of hu- man CRP have an immunomodulating activity [76]. There may be a system for amplification by CRP and macrophages in the primary stage of defence. SEX HORMONES AND CRP The mean concentration of rat serum CRP was 1000 100 a = ~ 'o)) = 2 10 1] f= Birth Fic. 12. 3.6+0.8 g/ml (n=5) at the birth and no apparent change was observed during the first 15 days. Thereafter it increased rapidly until day 30 reaching 0.1 mg/ml and further increased gradual- ly to the adult level of 0.4-0.8 mg/ml. No differ- ences were observed between male and female [77]. The change of serum AFP, which is a typical onco-fetal protein in mammals [78], in newborn rats is also shown in Figure 12, decreasing with age in contrast to CRP. Serum CRP level after conception showed two peaks before deliverly. The first peak was seen on the day of fertilization and the second peak was observed on day 15 of gestation. The mean levels at the peaks were 0.70+0.06 and 0.77+0.10 mg/ ml, respectively. Then the CRP decreased markedly to 0.42+0.05 mg/ml on the day of par- turition (day 20 of gestation) and elevated again to the normal level within 2 days. The CRP levels on day 0, 15 and 20 significantly were high, as com- pared to the control (P<0.05), examined by the 10 1 C o Female ® Male tgs | Unsexed los ss 10.01 Age (days) The changes in serum levels of CRP and AFP during development of rats. Both levels were determiend by rocket immunoelectrophoresis, and represented as the value of 5-10 rats. CRP in Animals 509 ° o Serum Level (mg/ml) (2) oO fe) ESS -10 -5 O 5 10 15 20 25 30 34 a og = Fertilization Parturition Days Fic. 13. immunoelectrophoresis. 2-tailed Student t-test (Fig. 13). It is interesting that the CRP level in rats changed with its development, especially, during 10 days after 15-day-old, increasing 60-times. Zouaghi et al. [79] reported that in neonatal rats, 20 Relative Values = nots . L ecite eee ce Eas se see Eas baa NO The changes in serum CRP levels of pregnant rats. The levels of CRP were determined by rocket injection of turpentine-oil caused the decline of serum level of AFP, a well known carrier of estrogen [80], with a marked increase of hapto- globin being one of the acute phase reactants. Although, it is not clear whether the production of CRP relates to that of AFP or not, the serum levels of both proteins may be controlled by estradiol. According to Dohler and Wuttke, the serum level of estrogen in newborn rat is 10 times higher than that of adult [81]. The serum level of estradiol increases towared the parturition [82]. These results suggest that the production of CRP in rat is controled by estradiol. In fact, when a rat was injected with estradiol-178, the serum level of CRP was decreased. However, no significant effect of other steroid hormones was observed (Fig. 14). In contrast, the serum level of AGP, a Fic. 14. Effect of steroid hormones on the serum level of CRP in rats. E>; estradiol-17, PR; progestrone, TS, testosterone; CS, corticosterone; CT, control. Five rats were dialy (2 days) injected s.c. with a dose of 1 mg/kg in sesame oil or oil alone as contorl. The serum level of CRP was determined by rocket immunoelectrophoresis. The values were repre- sented as the ratio to the initial level (0.57+0.12 mg/ml). Open and meshly horizontal bars indicate intact and ovariectomized rats, respectively. * Sta- tistically significant over the control: P<0.05. 510 W. NUNOMURA major acute phase protein in rat, was remarkably increased by administration of estradiol [83]. These results indicate that the changes of serum level of CRP might be a hormonal effect but not acute injuries of hepatocytes. In both syrian hamster and white-edged rockfish, CRPs are so- called “female protein”. In syrian hamster, the serum level of CRP in female is as high as 1.5-3 mg/ml and that of male is 4-20 ng/ml [84, 85]. The serum level of CRP in male syrian hamster was increased by removing the testis. By injection to female animals of testosterone, the serum level of CRP in female was decreased. The production of CRP was suppressed by testosterone in syrian hamster. On the other hand, the serum level of the CRP analogue in white-edged rockfish was en- hanced by estradiol-178 but not testosterone and cortisol. In the male fish, the level was remarkably elevated from the initial level of 25.0+19.0 ug/ml to 1533.3 +484.0 ug/ml (n=5) during 10 days af- ter injection with estradiol-17@ (1 mg/kg) [33]. Although it is well known that human chorionic gonadotropin (hCG) has an immunosuppressive activity during pregnancy [86], there is no report describing the effect of hCG on the serum level of acute phase reactants. Injection of hCG to normal male rats significantly elevated the serum level of CRP as well as AGP (Table 2). This elevation was independent of the dose commonly used for rat experiments. At present, the mechanism of hor- monal control of CRP production in hepatocytes has not been known yet. It is important to TABLE 2. Serum levels of acute phase reactants in rats injected with hCG G CRP AGP te eS (mg/ml) (mg/dl) Control 0.41+0.01 Galea hCG treated I 20 IU 57 ae 0Bs S) ae" II 100 IU O263—5 2038205 9.8+0.9* Ill 300 IU Ol63 ee OL03 Fa 1056220285" IV 1000I1U OSs8ee OOS ss Boar Lege IMIGANSES IE TP AO ee OI OT Serum was collected 24 hr after injection. Serum CRP and AGP levels were determiened by rocket- immunoelectrophoresis and single radial immuno- diffusion [93], respectively. investigate carefully the relationship between the sex hormones and CRP synthesis in hepatocytes. CRP IN STRESS The topic in recent studies on CRP is a rela- tionship with stress. The serum level of CRP in rainbow trout markedly elevated when fish reared in water of high tempareture [87]. A preliminary experiment in our laboratory, when 8 rats reared in a cage of 29X34 16 cm for 16 days (crwoding stress), the serum level of CRP was slightly in- creased at Day 3 and then decreased to the initial level on Day 16 (unpublished data). The rela- tionship between CRP and stress has not been studied in detail. CONCLUSIONS Since 1930, CRP has been used only as a marker for inflammation in the clinical laboratories, how- ever, a number of studies let us know that this protein has active functions related to the defence system. The biological functions of CRP are summerized in Fig. 15. CRP is synthesized in - hepatocytes, stimulated by IL-6 derived from macrophages in inflammatory sites. CRP activates macrophages and inhibit the action of T-cell [88- 90] to block production of antibody against en- dogenous materials. CRP acts as a scavenger of endogenous materials, cooperating with comple- ments. In rat, CRP does not activate complement, C3 [46]. On the other hand, production of CRP is controlled by sex hormone(s). The studies on CRP have remained a lot of problems to be solved, for example the evolution- ary origin of this protein, to act really as a scaven- ger of endogenous substances in lower classes of animals having undeveloped immune net works, functional relationship with lectin, the activity as a modulator of immune response and relationship with stress. It is neccesary to investigate what role CRP plays in the reproduction. Unfortunately, studies on this protein have been carried out in the field of medicine and veterinary, and few reports have been brought from zoological field. I hope that more investigators will be intersted in CRP and the mechanism of defence system in animals CRP in Animals Sli COMPLEMENT IL-6 QO: BACTERIA *™ ayy NUCLEUS 3a ey, O2 <== HEE PEPTI DE { | M¢@ INFLAMMATORY DAMAGE Fic. 15. will be soon clarified. ACKNOWLEDGMENTS I thank late Dr. H. Hirai, former President of Tumour Laboratory for his continuous guidance, and Dr. T. Higashi, President of Tumour Laboratory, for his critical reading of this manuscript and encouragement. I also thank Dr. Y. Takakuwa, professor of Tokyo Women’s Medical College, Dr. K. Takano, professor of University of Ryukus, Dr. K. Mukhopadhyay, Tumour Laboratory and Dr. M. Mikuni, National Institute of Neuro Science, National Center of Neurology and Psychiatry for their valuable advices. I am grateful to Dr. A. Takemura, University of Ryukus for providing the samples and the stafs of Tumour Laboratory for their kindest help. REFERENCES 1 Tillett, W. S. and Francis, T. (1930) J. Exp. Med., 52: 561-571. 2 Abernethy, T. J. and Avery., O. T. (1941) J. Exp. Med., 73: 173-182. 3 McCarty, M. (1947) J. Exp. Med., 85: 491-498. 4 Claus, D. R., Osmand, A. P. and Gewurz, H. (1976) J. Lab. Clin. Med., 87: 120-128. 5 Salonen, E. M. (1982) J. Immunol. Methods, 48: 45-50. 6 Siboo, R. and Kulisek, E. (1978) J. Immunol. Methods, 23: 59-67. 7 Gotshclich, E. C. and Edelman, G. M. (1965) Proc. Natl. Acad. USA, 54: 558-566. LIVER T-cell The mechanism of sysnthesis and the biological functions of CRP. 10 iLil 12 13 14 iI) 16 i, 18 19 20 aN SEX HORMONE HY is = Get — F880 009) REPRODUCTION cre O; ACTIVATE €: INHIBIT Osmand, A. P., Friedenson, B., Gweurz, H., Pain- ter, R° A Hofmann, “i 'and Shelton; "22 (1977) Proc. Natl. Acad. Sci. USA, 74: 739-743. Kaplan, M. H. and Volanakis, J. E. (1974) J. Immunol., 112: 2135-2147. DiGamelli= WR Lotempa, ls AGxSiegels JE Suyehira, L., Petras, K. and Gewurz, H. (1980) J. Immunol., 125: 1933-1938. Robey, F. A., Jones, K. D., Tanaka, T. and Liu, T. Y. (1984) J. Biol. Chem., 259: 7311-7316. Du Close. W.. -Zlocke si) de and Rubing Rs I. (1988) J. Immunol., 141: 4266-4270. Shephard, E. G., van Helden, P. D., Strauss, M., Bohem, L. and De Beer, F. C. (1986) Immunology, 58: 489-494. Robey, F. A., Jones, K. D. and Steinberg, A. D. (1985) J. Exp. Med., 161: 1344-1356. Deodhar, S. D., James, K., Chiang, T., Edinger, M. and Barna, B. P. (1982) Cancer Res., 42: 5084- 5088. Barna, B. P., Deodhar, S. D., Gautam, S., Yen- Lieberman, B. and Roberts, D. (1984) Cancer Res., 44: 305-310. Thombre, P. S. and Deodhar, S. D. (1984) Cancer Immunol. Immunother., 16: 145-150. Zahedi, K. and Mortensen, R. F. (1986) Cancer Res., 46: 5077-5083. Barna, B. P., James, K. and Deodhar, S. D. (1987) Cancer Res., 47: 3959-3963. Barna, B. P., Thomassen, M. J., Wiedmann, H. P., Ahmad, M. and Deodhar, S. D. (1988) J. Biol. Response Mod., 7: 1-5. Gautam, S. and Deodhar, S. D. (1989) J. Biol. 22 23 24 DS 26 27 28 29 30 31 Sy) 33 34 35 36 Si] 38 3) 40 41 42 43 512 Response Mod., 8: 560-569. Anderson, H. C. and McCarty, M. (1951) J. Exp. Med., 93: 25-36. Croxatto, H. D., Nishimura, E. T., Yamada, K. and Hokama, Y. (1968) J. Immunol., 100: 563-568. Maudsley, S., Rowe, I. F., De Beer, F. C., Munn, E. A., Herbert, J., Feinstein, A. and Pepys, M. B. (1987) Clin. Exp. Immunol., 67: 662-673. Caspi, D:, Baltz, M_ySnel, F., Gruys, E.; Niv, D:, Batt, R. M., Munn, E. A., Buttress, N. and Pepys, M. B. (1984) Immunology, 53: 307-313. Maudsley, S., Baltz, M. L., Munn, E. A., Buttress, N., De Beer, F. C., Herbert, J., Feinstein, A. and Pepys, M. B. (1987) Biochim. Biophys. Acta, 924: 75-80. Pepys, Ma Bi Der Beers Ee (C= Milstein, «©: Ps March, J. F., Feinstein, A., Buttress, N. Clamp, J. R., Taylor, J., Bruton, C. and Fletcher, T. C. (1982) Biochim. Biophys. Acta, 704: 123-133. Winkelhake, J. L. and Chang, R. J. (1982) Dev. Comp. Immunol., 6: 481-489. Murai, T., Kodama, H., Naiki, M., Mikami, T. and Izawa, H. (1990) Dev. Comp. Immunol., 14: 49-58. Robey, F. A., Tanaka, T. and Liu, T. Y. (1983) J. Biol. Chem., 258: 3889-3894. Robey, F. A. and Liu, T. Y. (1981) J. Biol. Chem., 256: 969-975. Nunomura, W. (1991) Biochim. Biophys. Acta, 1076: 191-196. Takemura, A., Nunomura, W., Takano, K. and Hirai, H. Proceedings of The 3rd Meeting of Japanese Association for Developmental and Com- parative Immunology. Dev. Comp. Immunol. (in press). Nguyen, N. Y., Suzuki, A., Boykins, R. A. and Liu, T. Y. (1986) J. Biol. Chem., 261: 10456-10465. Kilpatrick, J. M., Kearney J. F. and Volanakis, J. E. (1982) Molecul. Immunol., 19: 1159-1165. Roux, K. H., Kilpatrick, J. M., Volanakis, J. E. and Kearney, J. F. (1983) J. Immunol., 131: 2411-2415. Potempa, L. A., Siegel J. N., Fiedel, B. A., Potem- pa, R. T. and Gewurz, H. (1987) Molecul. Im- munol., 24: 531-541. Ress, R. F., Gewurz, H., Siegel, J. N., Coon, J. and Potempa, L. A. (1988) Clin. Immunol. Immuno- pathol., 48: 95-107. Pepys, M. B. and Baltz, M. L. (1983) Adv. Im- munol., 34: 141-212. Hokama, Y., Tam, R., Hirano, W. and Kimura, L. (1974) Clin. Chim. Acta, 50: 53-62. Nunomura, W. (1991) Zool. Sci., 8: 277-286. Nunomura, W. and Hatakeyama, M. (1990) Hok- kaido J. Med. Sci., 65: 474-480. Pepys, M. B., Dash, A. C. and Ashley, M. J. (1977) Clin. Exp. Immunol., 30: 32-37. 44 45 46 47 48 49 50 51 a2 53 54 35) 56 i 58 59 60 61 62 63 64 65 66 67 68 69 W. NUNOMURA Volanakis, J. E., Clements, W. L. and Schrohenloh- er, R. E. (1978) J. Immunol. Methods, 23: 285-295. Gotschlich, E:"C,and Liu, TY. (i967) J Biel: Chem., 242: 463-470. De Beer, Fr C., “Baltz, Mea.” Minne Feinstein, A., Taylor, J., Bruton, C., Clamp, J. R. and Pepys, M. B. (1982) Immunology, 45: 55-70. Nunomura, W. and Higashi, T. (1990) Zool. Sci., 7: 1053. Nunomura, W., Hatakeyama, M. and Hirai, H. (1990) J. Biochem. Biophy. Methods, 21: 75-80. O'Farrell, P. H. (1975) J. Biol. Chem., 250: 4007- 4021. Mold, C., Rodgers, C. P., Kaplan, R. L. and Gewurz, H. (1982) Infect. Immunity, 38: 392-395. Bezkorovainy, A., Springer, G. F. and Desai, P. R. (1971) Biochemistry, 10: 3761-3764. Matsumoto, I. and Osawa, T. (1974) Biochemistry, 13: 582-588. Uhlenbruck, G., Sdlter, J., Janssen, E. and Haupt, H. (1982) Ann. N. Y. Acad. Aci., 389: 476-479. Uhlenbruck, G., Karduck, D., Haupt, H. and Schwick, H. G. (1979) Z. Immunol. Forsch., 155: 262-266. Soelter, J. and Uhlenbruck, G. (1986) Immunology, 58: 139-144. Kobayashi, K., Hara, A., Takano, K. and Hirai, H. (1982) Molecul. Immunol., 19: 95-103. Vasta, G. R., Hunt, J. C., Marchalonis, J. J. and Fish, W. W. (1986) J. Biol. Chem., 261: 9174-9181. Nunomura, W., Hatakeyama, M. and Hirai, H. (1987) Proc. Jap. Soc. Immunol., 17: 594. Maudsley, S. and Pepys, M. B. (1987) Immunology, 62:22? Woo, P., Korenberg, J. R. and Whitehead, A. S. (1985) J. Biol. Chem., 260: 13384-13388. Taylor, J. A., Bruton, C. J., Anderson, J. K., Mole, J. E., De Beer, F. C., Baltz, M. L. and Pepys, M. B. (1984) Biochem. J., 221: 903-906. Laurell, C. B. (1966) Anal. Biochem., 15: 45-52. Kushner, I. and Feldmann, G. (1978) J. Exp. Med., 131: 466-477. Jamieson, J. C., Ashton, F. E., Friesen, A. D. and Chou, B. (1972) Can. J. Biochem., 50: 871-880. Nunomura, W., Watanabe, H. K. and Hirai, H. (1990). Zool. Sci., 7: 767-770: Zeller, Jv M.," Landay, A: i) Einte aise and Gewurz, H. (1986) J. Lab. Clin. Med., 108: 567- 576. Zeller, J. M., Kubak, B. M. and Gewurz, H. (1989) Immunology, 67: 51-55. Tseng, J. and Mortensen, R. F. (1988) Molecul. Immunol., 25: 679-686. Watanabe, M., Miyazaki, M. and Taketa, T. (1976) Cancer Res., 36: 2171-2175. 70 7A U2 73 74 1S 76 Td 78 79 80 CRP in Animals Onoe, T., Dempo, K., Kaneko, K. and Watabe, H. (1973) In “Gann Monograph on Cancer Research 14” Ed. by H. Hirai and T. Miyaji. University of Tokyo Press, Tokyo. pp. 233-247. Castells. V.. Geiger, T.; Andus; @.; Walter, E., Hirano, T., Kishimoto, T. and Heinrich, P. C. (1988) Eur. J. Biochem., 177: 357-361. Geiger, T., Andus, T., Klapproth, J., Hirano, T., Kishimoto, T. and Heinrich, P. C. (1988) Eur. J. Immunol., 18: 717-721. Marinkovic, S., Jahreis, G. P, Wong, G, G. and Baumann, H. (1989) J. Immunol., 142: 808-812. Tzehoval, E., Segal, S., Stabinsky, Y., Fridkin, M., Spirer, Z. and Feldman, M. (1978) Proc. Natl. Acad. Sci. USA, 75: 3400-3404. Buchta, R., Fridkin, M., Pontet, M. and Romeo, D. (1986) Peptides, 7: 961-968. Robey, F. A., Ohura, K., Futaki, S., Fujii, N., Yajima, H., Goldman, N., Jones, K. D. and Whal, S. (1987) J. Biol. Chem., 262: 7053-7057. Nunomura, W. (1990) Comp. Biochem. Physiol., 96A: 489-493. Watabe, H., Hirai, H. and Sato, H. (1972) Gann, 63: 189-199. Zouaghi, H., Savu, L., Vranckx, R., Martin, M. E. and Nunez, E. A. (1984) Marker Proteins in In- flammation, 2: 549-552. Nunez, E. A., Benassayag, C., Vallette, G., Savu, L., Clerc-Hofmann, F., Delorme, J. and Christeff, N. (1983) Ann. N. Y. Acad. Sci., 417: 137-148. 81 82 83 84 85 86 87 88 89 90 91 92 93 S13 Dohler, K. D. and Wuttke, W. (1975) Endocrinol., 97: 898-907. Taya, K. and Greenwald, G. S. (1981) Biol. Repro- duction, 25: 683-691. Diarra-Mehrpour, M., Bourguignon, J., Leroux- Nicollet, I., Marko-Vercaigne, D., Biou, D., Hiron, M. and Lebreton, J. P. (1985) Biochem. J., 225: 681-687. Coe; J. E-, Margossian; S- S:, Slyater, H. S. and Sogn, J. A. (1981) J. Exp. Med., 153: 977-991. Coc; =e] andsRoss. MM: I= (1983) oe Exp, Med, 157: 1421-1433. Adcock, E. W., Teasdale, F., August, C. S. Cox, S., Meschia, G., Battaglia, F. C. and Naughton, M. A. (1973) Science, 181: 845-847. Kodama, H., Nakanishi, Y., Mikami, T. and Izawa, H. (1988) Proceedings of the 105th Meeting of The Japanese Society of Veterinary Science. p. 221. Whisler, R. L., Newhouse, Y. G. and Mortensen, R. F. (1986) Cell. Immunol., 102: 287-298. Mortensen, R. F., Osmand, A. P. and Gewurz, H. (1975) J. Exp. Med., 141: 821-839. Mortensen, R. F. and Gewurz, H. (1976) J. Im- munol., 116: 1244-1250. Potempa, L. A., Maldonado, B. N., Laurent, P., Zeuel, E. S. and Gewurz, H. (1983) Molecul. Immunol., 20: 1165-1175. Karmen, A. (1955) J. Clin. Invest., 34: 131-133. Mancini, G., Carbonara, A. O. and Heremans, J. F. (1965) Immunochem., 2: 235-254. BO Pere bud ai ante a oe : ‘e Mee ase (DEG idl de aannasogT. (deer Sabha = ae ee Stag, Re eeeice tae at: doitaebe sg aoe Asha tOM 28 “S098” Sara 7. Reb! IRE el NEE “aid a Limit da) ; ‘eae Seema. be ster er aOR le Ee eee E Sito Ri Ais) (es eee at sort: saan Se. Aras ait: mist ae Haiweblunt A uns sd erie at: el x ‘Ae zhi AB) an kag Iie oS a Tihs a2 ral 2 a " Hpk nue + af iff es Serie c by eet isiiemnb vete Lane) art i. ‘ he ie rg, : oa leat ue ea! : Ine = a re . . wee Gl , re y = “J o F - - “¥ 2 = 4 oe 2 - y ol ae : ie = ‘ H oes 2 Fi = = ee ‘ Py oe * + am é ; ra ‘ I -) bk 7 : ‘pe i. 2 ¢ ) be a I> *e . i , 4 ¥ 4 x a3 i a ko = F 1 Noe y, B 2 ZOOLOGICAL SCIENCE 9: 515-527 (1992) Self-generated Zigzag Turning of Bombyx mori Males during Pheromone-mediated Upwind Walking Ryoue! Kanzaki', NAoKo Suci and TATSUAKI SHIBUYA Institute of Biological Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba-shi, Ibaraki 305, Japan ABSTRACT— In order to further understand the pheromone-oriented walking of the male silkworm moth Bombyx mori, we have analyzed the movement of free walking males in a laboratory wind tunnel using a continuous pulsed stimulation of various frequencies with pheromones. Single and brief pulsed stimulation (100 msec) with pheromones to either antenna elicited a long-lasting zigzag turing pattern (<4 sec) which consisted of left and right turns or vice versa. The direction of the first turn of the zigzag turning was strongly related to which antenna was stimulated. While continuous pulsed stimulation was applied, similar zigzag turnings to those triggered by a single pulsed stimulation were repeatedly observed with each stimulation. The tempo and the angle of the zigzag turning gradually increased turn by turn and were successively reduced with each stimulation. We exposed two control mechanisms during upwind walking to the pheromone odor source: (1) a self-generated zigzag turning program triggered by a pulsed pheromonal stimulation, and (2) reset mechanisms of this program. We propose that walking B. mori males use similar control mechanisms © 1992 Zoological Society of Japan for movements toward a pheromone source to those of flying moths. INTRODUCTION Male silkworm moths Bombyx mori respond to the sex-attractant pheromone released by conspe- cific females with a characteristic behavioral reper- toire called the “mating dance” including wing vibrations, walking, and abdominal curvature [1- 3]. Neurons in the brain of the male moths process information about the pheromone, integrating this information with sensory information of other modalities, then initiate and possibly regulate the motor outflow that results in oriented behavior [4— 7|. Recent findings using several species of moths have indicated that, in addition to optomotor anemotaxis (steering with respect to the wind), flying males employ a self-steered program of counterturns in the process of locating the phero- mone source [8-13]. It has been observed that B. mori [14] and Periplaneta americana [15] males, which exclusive- ly or primarily walk to sources of their pheromone Accepted February 26, 1992 Received December 2, 1991 * To whom offprint request should be addressed [14-17], execute frequent and spontaneous turns while walking in an airstream uniformly permeated with pheromones. In contrast, Willis and Baker [18] reported that Grapholita molesta males do not use a counterturning program while walking up- wind to pheromones, whereas they do while flying upwind. We now know that the odor plume formed downwind from the female has a highly variable structure [19, 20], and that an intermittent phero- monal stimulus greatly improves a male moth’s ability to orient upwind to a pheromone source [8, 21-23]. Kramer [23] analyzed the B. mori walking behavior on a “Kramer’s sphere” using pulsed pheromonal stimulation, and showed general fea- tures of the behavior. However, the detailed walking pattern of individual B. mori males has not yet been analyzed well. In order to further understand the walking pattern, in this study we analyzed the movement of free walking B. mori males in a laboratory wind tunnel using a con- tinuous pulsed stimulation of various frequencies with pheromones. We exposed two control mechanisms during upwind walking to the pheromone odor source: (1) 516 R. KANZAKI, N. SuGI AND T. SHIBUYA a self-generated zigzag turning program triggered by a pulsed pheromonal stimulation, and (2) reset mechanisms of this program. We propose that walking B. mori males use similar control mecha- nisms for movements towards a pheromone source to those of flying moths expected from Baker’s model [22], rather than just walking directly up- wind to the source. MATERIALS AND METHODS Insects Bombyx mori (Lepidoptera, Bombycidae) were reared in the laboratory on an artificial diet at 25°C and 50-60% relative humidity under a 12:12 LD cycle, and were used within 3 days after eclosion. Pheromonal stimulation Two pheromonal stimuli were used in this ex- periment: (1) Crude pheromone extract was pre- pared by washing the tip of a virgin-female moth’s abdomen, which contains the pheromone gland, in 100 ul of n-hexane (Wako reagent-grade) for 15 min, yielding “1 female equivalent (FE)” of “female extract”, and (2) synthetic (£, Z)-10, 12-hexadecadien-1-ol, bombykol, the principal pheromone component of Bombyx mori, kindly provided by Shin-Etsu Chemical Co. Ltd. Each pheromonal stimulant was applied to a piece of filter paper (12cm), which was then inserted into a plastic cartridge (1 ml). The car- tridge was placed in the wind tunnel 1 cm above the floor and upwind. Pulsed olfactory stimulation was produced by switching the clean air stream to the stimulant cartridge, which was controlled by a 3-way solenoid valve [24]. The frequency (0.25-2 Hz) and duration (100 msec) of the stimulation was controlled by an electronic stimulator. Odor- ants were thus introduced into the wind tunnel (1.67 ml/100 msec). The stimulation was moni- tored by later video observation of a synchronized green light-emitting diode (LED) connected to the electronic stimulator, which was on the floor of the wind tunnel. Wind tunnel and data recording The wind tunnel was constructed using clear polycarbonate plastic which had a working section 49 cm long, 49 cm wide at floor level, and 32 cm high. Air flow was introduced into the tunnel by negative pressure generated by a voltage regulated fan. The odors used were removed from the room by an exhaust duct attached to the fan at the downwind end of the tunnel. The wind velocity was ().5 m/sec, which was determined by measur- ing the flow rate of the exhausted air with a flow meter. The shape and position of the pheromone plume was simulated by applying TiCl4, which makes a smoke plume, to the same size filter paper as used for pheromone sources, and then placing it in the same location in the wind tunnel as the pheromone would be during an experiment. To prevent pheromonal contamination on the floor of the wind tunnel, the brown paper on the floor was replaced for every experiment. Indi- vidual males were placed on the center of the plume 20-25 cm away from the tip of the stimulant cartridge. A continuous clean air stream did not arouse males on the floor of the wind tunnel. The room temperature was approx. 26°C. Walking tracks of individual males were re- corded on a Sony SLO-333 video recorder using a Sony HVC-80 video camera positioned above the wind tunnel. All recordings were played back frame-by-frame through a 53cm (21 inch) NEC video monitor. The computer (NEC PC-9801VX) screen was superimposed on the video monitor by a video-converter system and the consecutive loca- tions of the head and tail of each moth were digitized and input every 0.1 sec into the computer by pointing the position by moving a ‘mouse’ cursor on the screen and pushing the ‘mouse’ button. The position, walking speed, and body axis with respect to the upwind direction of each moth were calculated by the computer. RESULTS Upwind walking to various frequencies of con- tinuous pulsed pheromonal stimulation Continuous pulsed pheromonal stimulation was applied to free walking Bombyx mori males to determine the affect of the frequency of the stimu- lation on the upwind walking movement. In Behavior of Walking Moth to Pheromones S17 these experiments the duration of the stimulation was constant (100 msec). Female extract (1 FE) was used as pheromonal stimulant. In some cases 0.1-10 ug synthetic bombykol was used. Approxi- mately 1 zg of bombykol was found to be present in 1 FE of solution using gas chromatographic analysis (data not shown). Both stimulants affected the initiating walking (197 out of 200 trials (98.5%) in 35 males, Table 1), of which >85% (in average) located the source (Table 1). All males TABLE 1. Percent of Bombyx mori males responding to different frequencies of continuous pulsed stimulation with pheromone oil bhetscte of those of those | 2 ee itage banoiy VANE xc ltieting wallinen a PMN MINING. CRS Par % locating upwind the pattern constant*° 17 100.0 88.2 0.0 nt 2.00 66 97.0 93.8 0.0 0.12 1.00 40 Dies 94.9 3)3)ae) 0.71 0.50 34 100.0 Oie2 94.1 1.28 0.33 22 100.0 86.4 WI nt 0.25 Pash 100.0 162 90.5 2.36 *! Duration of the stimulation was constant (100 msec). *2 Total number of turns were divided by the number of stimulation. *3 Constnant stimulation was introduced into the air stream. nt: not tested. Sip\ 2 take CENTIMETERS o) 255, 20 15 10 5 CENTIMETERS (P) Fic. 1. Plots of the tracks, as viewed from above, B. mori males walking upwind to various frequencies (Hz) of continuous pulsed pheromonal stimulation (1 FE). Duration of the stimulation was 100 msec. Wind was from the right (0.5 m/sec). Pheromonal stimulation was introduced into the air stream of the wind tunnel as (A) 2 Hz, (B) 2 Hz, (C) 1 Hz, (D) 0.5 Hz, (E) 0.25 Hz pulses. Filled circles indicate the head position when the olfactory stimulation was introduced into the air stream. Plots of B-E are of the same moth. Lines indicate the average boundaries of the pheromone plumes simulated by TiCl, smoke. Outer borders are the dimensions of the wind Ee OrZoutlz (9) 20 1 10 5 0) CENTIMETERS (P) tunnel and the diagrams are of linear scale from the pheromone source (P). 518 R. KANZAKI, N. SuGi AND T. SHIBUYA walking upwind did so while fluttering their wings. No significant differences in the walking move- ment were observed while using female extract or synthetic bombykol. In a few cases, males did not walk all the way to the source but stopped and sat in the plume area or turned away from the plume and sat outside it (e.g., Fig. 1C). Figure 1 shows plots of tracks of a male moth walking upwind in response to various freqeuncies of continuous pulsed stimuli with pheromones (1 FE). A clean air pulse introduced into the wind tunnel did not arouse males (data not shown). When male moths received a constant or a high frequency (e.g., 2 Hz) of pheromonal stimulation, most walked in nearly a straight line, steering directly upwind (Fig. 1A). In a few cases they walked at angles between 20° and 50°, and some- times more than 70° oblique to the direction of upwind (Fig. 1B), as described by Kramer [14]. Although their side to side movement was less than their wing-span (approximately 4cm, Fig. ANGLE OF THE BODY AXIS TO UPWIND (degrees) @) ro) 10 15 1A), the angle of the body axis with respect to upwind (0°) changed dramatically as the frequency of pulsed stimulation was reduced (Figs. 1, 2). A turn was defined as when the body axis changes from clockwise to anticloskwise (or vice versa) of more than 10° and the time interval (tempo) of the turn was more than 0.2 sec, in order to eliminate angular error during digitizing. Figure 2 shows the angles of the body axis with respect to upwind plotted against times. In 0.25 Hz stimulation (Fig. 2E), when the male received the stimulation puff while the angle of the body axis was between 0° and 90°, the male attempted to turn to the right and flipped to the left. On the other hand, males showed symmetrical movements when the angle was between 270° and 360’; i.e., they turned to the left then flipped to the right. The tempo of the first turn elicited just after the stimulation was 1.1+0.1 sec (mean+SE, n=48, Table 3). The angle of the turn was 68+5° (n=48, Table 3). Each time a male received the stimu- 20 29 30 39 TIME (sec) Fic. 2. The angles of the body axis with respect to upwind (0°) plotted against time when the male was in the pheromone plume. Replotted from the same data shown in Fig. 1. (A) The angle of the body axis of the male is measured with respect to upwind. The angle of the male in (A) is 45°. The arrow indicates the wind direction. (B-E) Plots of the angle against time when a various frequency of continuous pulsed stimulation was applied. Solid lines beneath plots represent the period of olfactory stimulation. The region of plot (E) indicated by the bar is expanded in Fig. 4B. Behavior of Walking Moth to Pheromones 519 lation at any phase of the turn, the switching of the movement occurred. More than 85% (in average) of males which were locating upwind showed this zigzag turning pattern when the frequency was between 0.5 and 0.25 Hz (Table 1). Thus, the direction of the first turn elicited by the stimulation was strongly related to the angle of the body axis to upwind. The zigzag turning pattern was repeatedly observed when the frequency was reduced (Fig. 2C-E). We then, applied continuous pulsed stimulation to males for which an antenna had been removed completely from the basal segment; such males could receive the pheromones only by one intact antenna. In 0.25 Hz stimulation, most of the treated males could not orient to the pheromone source. Although they showed a similar zigzag turning pattern when the antenna had been stimu- lated, they turned away from the plume. It may be because that the treated males usually began to turn in the same direction as the intact antenna and the possibility to turn away from the plume be- came high. In some cases, as shown in Fig. 3B, the male succeeded to orient to the source with show- ing a typical zigzag pattern similar to the intact males. By contrast, when a treated moth received 2 Hz stimulation, >80% of treated males could 180 0 360 180 (degrees) 180 0 360 ANGLE OF THE BODY AXIS TO UPWIND BIG 3: basal segment, are plotted against time. orient to the source as shown in Fig. 3A. In this case, at first the male continued to turn in the same direction as the intact antenna, and then walked in nearly a straight line, steering directly upwind. thus, such treated males showed a similar zigzag turning pattern to intact ones. The average number of turns per stimulation is shown in Table 1. Males showed 2.36 turns on the average per stimulation of 0.25 Hz. The number of turns decreased with increases in frequency. With a stimulation frequency of 2 Hz, turning was rarely observed (0.12/stimulation). Thus, the zig- zag turning pattern gradually disappeared with increased frequency of pulsed olfactory stimula- tion. Figure 4 illustrates the walking tracks and the angles of the body axis with respect to upwind against time: Plots of the same data are shown in Fig. 2E (solid line above the plots). In some cases, before showing the turning pattern, males walked virtually straight along the body axis (i.e., their body axis angle changed only slightly compared with the zigzag turns). This straight line walking was also elicited in response to higher frequencies of stimulation (0.5—1 Hz) and lasted <0.5 sec, and then the zigzag turns followed this straight line walking (Figs. 2-4). In many other cases, how- TIME (sec) The angles of the body axis of the treated males for which the left antenna was completely removed from the (A) 2 Hz pulsed pheromonal stimulation. (B) 0.25 Hz pulsed stimulation. Plots of (A) and (B) are of the different moth. Even treated males showed zigzag turnings in response to the stimulation. Solid lines beneath plots represent the period of stimulation (1 FE). 520 R. KANZAKI, N. SuGI AND T. SHIBUYA A ep) ae Lu = Lu = = FZA Lu O B ANGLE OF THE BODY AXIS TO UPWIND(degrees) TIME (sec) Fic. 4. Plots of the tracks (A) and the angles of the body axis with respect to upwind, are plotted against time (B): Same data as in Figs. 1E and 2E. In some cases, B. mori males walked virtually straight along the body axis (S) before typical zigzag turnings (T). This straight line walking lasted <0.5sec. Filled circles in (A) indicate the head position when the olfactory stimulation was introduced into the air stream. Solid lines from the filled circles indicate the body axis. Solid lines beneath plots in (B) represent the stimulation. ever, it was difficult to distinguish any brief straight line walking from the turns. It is sometimes observed that the walking speed became relatively higher when the moths showed the straight line walking (data not shown). Males whose eyes were pained black to prevent visual affects on the movement could also orien- tate towards the goal with a similar zigzag walking pattern (data not shown) [25]. Single pulsed pheromonal stimulation to either antenna In order to know the detail of this zigzag turning pattern against time, single pulsed pheromonal stimulation (100 msec of duration) was applied to either antenna of a resting male on the open field without wind (Fig. 5). The stimulant cartridge was placed in front of either antenna approximately 1 cm apart. In 170 out of 172 trials (98.8%) using 29 males, single pulsed stimulation elicited initiation of walking (Table 2). 87.3% (in average) of those initiating walking showed identical zigzag turning patterns to those triggered by a low frequency of pulsed stimulation; that is, the male showed back and forth turning by applying the pulsed stimu: lation to the antenna (e.g:, Fig. 5Aa,b). We predict that with single pulsed stimulation there is a tem- poral difference in the pheromone concentration and/or a delay in signal reception between one antenna and the other. To prevent contamination of the stimulant to the other antenna, we also used treated males for which one antenna had been removed from the basal segment as described previously (Figs. 3, 5Ac, d). Figure 5B illustrates relative changes of the angle of the body axis against time with single TABLE 2. Percent of Bombyx mori males responding to single pulsed stimulation with pheromone of those stimulation*’ n Yoinitiating walking initiating walking Yoshowing the pattern intact left 71 98.6 84.3 right 68 98.5 77.6 treated left 13 100.0 95.0 right 20 100.0 9273 *! Treated males for which other antenna was removed completely from the basal segment. Behavior of Walking Moth to Pheromones 521 REL.ANGLE OF THE BODY AXIS (degrees) @) 5 10 1S 20 TIME (sec) Fic. 5. @) 5 10 1S 20 TIME (sec) Single pulsed stimulation (100 msec) with pheromones (1 FE) was applied to the antenna. (A) Stimulation was applied independently to the left antenna (a, c) and right antenna (b, d) of the intact moths (a, b) and treated moths (c, d) for which one antenna had been removed completely from the basal segment. Arrows indicate the direction of the stimulation. (B) Plots of the relative angles of the body axis against time when single pulsed stimulation was applied to the antenna as shown in (A). Plots of (a, b) and (d) are of the same moth. Solid lines beneath the plots in (B) indicate the stimulation period. pulsed stimulation (1 FE). For approximately 4 sec after the stimulation, both treated males (Fig. 5Ac, d) and intact males (Fig. 5Aa, b) showed a turning pattern which was identical to the pattern shown in Fig. 4B. A symmetrical zigzag turning pattern was elicited by applying the pheromones to one antenna or the other (Fig. 5B). When the stimulation was applied to the left antenna (Fig. 5Aa, c), the male turned to the left first, then flipped to the right (Fig. 5Ba, c), while the male turned to the right first, then flipped to the left, when the stimulation was applied to the right antenna (Fig. SAb, d, 5Bb, d). More than 80% (in average) of intact maies and >90% (in average) of treated males showed similar turning patterns (Table 2). The tempo of the first turn was 1.2+0.1 sec (mean+SE, n=16, Table 3). The angle of the first turn was 91+14° (n=16, Table 3). Other males stopped walking before complete turning and a few males turned in the opposite direction from the stimulated antenna. In general, the direction of the first turn depends strongly on the antenna stimulated with phero- mones. A brief straight line walking was some- times observed in this set of experiments. After males showed the zigzag turning pattern two or three times, the pattern was shifted to a “looping” (turns of more than 360° or greater). Seven out of 522 TABLE 3. stimulation and 0.25 Hz pulsed stimulation Single Pulsed Stimulation tempo (sec) angle (degrees) n 0.25 Hz Pulsed Stimulation tempo (sec) angle (degrees) n R. KANZzAKI, N. SuGI AND T. SHIBUYA Ist 2nd turn turn abs*! 1.2+0.13 1.9+0.3° rel 12 1.5+0.2° abs 91+14? 168+31° rel 12 1.8+0.2° 16 16 1st 2nd turn turn abs 1.1+0.1° 1.6+0.1° rel te 1.5+0.1° abs 68 +5? 121+10° rel 12 2.0+0.2° 48 48 Mean and SE of tempo and angle of the zigzag turn in response to single pulsed 3rd turn 2.1+0.4° 2.4+0.4° 161s=35" 3.20%" 8 3rd turn 1.7+0.1° 159 Olas 132 ==22 De." 29 Means in the same row having no letters in row are significantly different according to t-test (P<0.05). *! abs: absolute time or angle; rel: relative to the Ist turn A SINGLE PULSE PZ —| D> i= toe oO W fa \ => Ww uy ae a TURNS B SINGLE PULSE 4 z -& 3 T= om aoe —| .— c Ww ee | 1st 2nd TURNS 3rd TEMPO REL.TO ANGLE REL.TO THE 1ST TURN THE 1ST TURN 1st 0.25Hz CONTINUOUS PULSES 2nd 3rd TURNS 1st 0.25Hz CONTINUOUS PULSES 2nd 3rd TURNS Fic. 6. Histograms of the relative tempo (A, C) and angles (B, D) of each turn of the zigzag turns, which were elicited in response to single pulsed stimulation (A, B) and 0.25 Hz pulsed stimulation (C, D). All the values were calculated by the computer and the mean and standard error are indicated. Behavior of Walking Moth to Pheromones 323 10 (70%) males tested looped in the same direction as the antenna stimulated. One way looping continued for more than 10sec (Fig. 5B). Then, the direction of the looping was irregularly changed (e.g., Fig. 5Ba). In many cases the total period of looping lasted more than 30 sec. Changes of tempo and angle of the zigzag pattern against time Changes of tempo and angle of the zigzag turn- ing pattern elicited by 0.25 Hz (e.g., Fig. 2E) and single pulsed stimulation (e.g., Fig. 5B) with phero- mones were measured against time (Table 3). These two parameters (i.e., tempo and angle) of each turn were compared with those of the first turn which occurred after each stimulation. The results are represented as histograms in Fig. 6. Figures 6A and 6B illustrate the relative changes of the tempos and angles turn by turn in response to a single pulsed stimulation to either antenna. The average tempos relative to the first turn were 1.5+0.2 (mean+SE, n=16) in the second turn and 2.4+0.4 (n=8) in the third turn. The average angles were 1.8+0.2 (n=16) in the second turn and 3.2+0.4 (n=8) in the third turn. The values of both tempo and angle were significantly differ- ent turn by turn (f-test, P<0.05). Thus, the values of both parameters in each turn were increased turn by turn (Table 3). Figures 6C and 6D illustrate the relative changes of the tempos and angles turn by turn in response to 0.25 Hz pulsed stimulation. The average tem- pos relative to the first turn were 1.5+0.1 (n=48) in the second turn and 1.9+0.1 (n=29) in the third turn. The average angles were 2.0+0.2 (n=48) in the second turn and 2.7+0.3 (n=29) in the third turn. The values of the tempo were singificantly different turn by turn (P<0.05). The values of the angles between the first turn and the second turn were significantly different (P<0.05). Thus, the values of each parameter were increased turn by turn between the stimuli as in the case of single pulsed stimulation. However, each value was successively reduced by the new pulsed stimu- lation. Then, each value again, gradually increased turn by turn until the next stimulation (Fig. 6C, D). Similar characteristics were observed on the treated males for which one antenna was com- pletely removed from the basal segment (Fig. 3B). These results imply that the zigzag turning pattern, which was triggered by a brief pheromonal stimu- lation, is “reset” every time the new pheromon- al stimulation is applied to either antenna. In addition, the reset mechanisms occurred at any phase of the turning (Figs. 2, 3). DISCUSSION Kramer [23] previously investigated similar be- havioral experiments as this study using his “Kramer’s sphere”. In his study, he averaged a great number of runs on the sphere and showed the general features of the pheromone-oriented walking of B. mori males. In this study, we have analyzed a detailed walking pattern of individual male walking freely in a laboratory wind tunnel using a continuous pulsed stimulation of various frequencies with pheromones. Self-generated zigzag turning program Single pulsed stimulation (100 msec) with phero- mones to either antenna exposed a zigzag turning pattern which consisted of left and right turns or vice versa, as shown in Fig. 5. Even the treated males for which one antenna was removed showed a similar zigzag turning pattern in response to the single pulsed stimulation (Fig.5Bc, d). Brief pulsed stimulation elicited the zigzag turning pat- tern repeatedly for a longer period (approximately 4 sec) than the stimulation period (100 msec): We did not have to keep applying the constant stimu- lation to the male during the zigzag turning pattern. The tempo and angle of the zigzag turning gradual- ly increased turn by turn when the male was stimulated with pulsed pheromones (Figs. 5, 6A, B, Table 3). It has been reported in flying moths that the tempo at which turns occur becomes progressively longer during “casting” flight and that turns persist for seconds when the moth loses contact with the pheromone plume [11, 13]. We propose that walking moths also have a self-generated zigzag turning program with characteristics similar to the counterturning program of the “casting” flight. In this study, we have not examined the affect of the wind speed and pheromone concentrations on 524 R. KANZAKI, N. SuGi AND T. SHIBUYA the zigzag walking pattern. In order to further understand the program, we must clarify the affect of these factors on the program [23]. Although Kramer [23] showed a rate of change of direction during the walk and the cosine of the walking direction with respect to the upwind, such a zigzag turning found in this study, was not detected by his sphere experiments. This differ- ence might have any of several explanations: (1) We have analyzed an movement of an individual moth, whereas in the Kramer’s experiments, a great number of runs were averaged. (2) Kramer averaged a rate of change of direction and the cosine of the walking direction, whereas, we have analyzed the actual changes of the body axis of an individual moth (i.e., the tempo and the angle). (3) Discrepancies might be due to the differences in the experimental methods, i.e., we have used a freely walking moth in the wind tunnel, whereas a moth was on the Kramer’s sphere. As illustrated in Fig. 5 the direction of the first turn of this zigzag turning was strongly related to which antenna was stimulated with pheromones. It is possible that the direction of the first turn may be related to the concentration differences and/or delay between one antenna and the other. How- ever, Kramer [23] suggested it was unlikely that the B. mori males detected and utilized small differences between the time of stimulation of the two antennae. He did not report the direction of the first turn just after the stimulation. Again, these differences may reflect differences in experi- mental methods and so on described above. It is reported by Kramer [23] that the pulsed stimulation increased the walking speed of B. mori rapidly. We have measured the changes of the walking speed of an individual B. mori. It is sometimes observed that the walking speed be- came relatively higher when the moths showed the straight line walking just after the stimulation. “Reset mechanisms” of the zigzag turning program As illustrated in Fig. 6C, D. and Table 3, when continuous pulsed stimulation (0.25 Hz) was ap- plied during upwind walking, the tempo and the angle of the turn just before the stimulation were successively reduced with each stimulation, and then the tempo and angle gradually increased turn by turn until the next stimulation. Similar charac- teristics were observed on the treated males for which one antenna was completely removed from the basal segment (Fig. 3B). These results imply that the zigzag turning program, which was trig- gered by a brief pheromonal stimulation, is “reset” every time the new pheromonal stimulation is applied to either antenna. In addition, the reset mechanisms occurred at any phase of the turning (Figs. 25:3): Straight line walking during higher frequencies of pulsed stimuli with pheromones We found that some males walked in almost straight lines along the body axis for <0.5 sec just after each of the continuous pulsed stimuli, and the zigzag turning pattern followed to the brief straight line walking (Fig. 4). The results of continuous pulsed stimulation indicated that the turning pat- tern gradually disappeared with increases in the stimulation frequency and that males walked in almost straight paths, steering directly upwind (Figs. 1A, 2B, 3A, Table 1). Kramer [23] reported similar results in a male B. mori with the “Kram- er’s sphere” experiments. We predict that as the frequency of the stimu- lation is increased, the “reset mechanisms” as de- scribed above and a “brief straight line walking” occur more frequectly. Under this condition, changes in the angle of the body axis may become smaller, until finally the axis is almost directly upwind. This might be part of the reason why the males may walk in almost a straight line when the frequency is higher (i.e., 2 Hz). Willis and Baker [18] reported that male Grapholita molesta do not use a counterturning program while walking upwind to pheromone, whereas they do while flying upwind. Constant olfactory stimulation, or in other words, high frequency of pulsed stimulation, was applied in their experiments. We are very interested in whether G. molesta males show similar zigzag walking patterns to those observed in B. mori males when continuous low frequency pulsed stimulation is applied. On the other hand, when the frequency of the pulsed stimulation is lower (e.g., <0.5 Hz), we propose that following things happen on the male Behavior of Walking Moth to Pheromones as a part of the pheromone-oriented mechanisms of B. mori males. The pulsed pheromonal stimu- lation triggers the zigzag turning program includ- ing the straight line walking. Then, the male begins to display a zigzag turning pattern. The first turn of the zigzag turns is strongly related to which antenna is stimulated. The tempo and the angle of the turn are gradually increased turn by turn until the next stimulation is applied to the antenna. When the next new pulse stimulates the antenna, the zigzag turning program is “reset” and the male begins to turn to the same direction as the antenna stimulated, and then repeats the procedure. The upwind walking of B. mori males mediated by pheromonal stimulation may contain several control mechanisms. In this study we found that the upwind walking contains at least following two control mechanisms: (1) a self-generated zigzag turning program, and (2) reset mechanisms of this program. And the direction of the first turn of this zigzag turning might be strongly related to which antenna is stimulated with pheromones. As described above, it is possible that the direc- tion of the first turn may be related to the concen- tration differences and/or delay between one antenna and the other. While, treated males, for which one antenna had been removed, also showed similar characteristics of zigzag turnings to intact ones (Figs. 3, 5). For example, some of the treated males walked in nearly a straight line, steering directly upwind in 2 Hz stimulation (Fig. 3A) and in 0.25 Hz stimulation, they showed a typical zigzag turning and oriented toward the odor source (Fig. 3B). Though the treated males could not detect the concentration differences or delay between one antenna and the other, they could detect the upwind and orient toward the pheromone source. This suggests that the male moths may have some other upwind detecting mechanisms while orienting toward the odor source. Walking moths vs. flying moths Baker [11] proposed a model to explain the generation of zigzagging upwind flight in moths. This model invokes the complementary action of a dual flight control system in which contact with pheromone filaments in the turbulent plume would 525 produce: (1) longer-lasting activation of a counter- turning program, which would continue to produce the turns that constitute “casting” flight, and (2) rapidly activated surges of upwind flight that would also rapidly cease upon the next encounter with clean air. These two control mechanisms seem similar to those we found in B. mori males in this study; 1.e., (1) a self-generated zigzag turning program and (2) “reset” mechanisms of the program and a brief straight line walking. We still do not know whether this brief straight line walking occurs every time stimulation is applied. However, re- sults of this study basically support Baker’s model; that is, it is possible that a similar control system exists in both walking B. mori males and flying moths while they are orienting to the pheromone source. The zigzag turning pattern was sustained for approximately 4 sec after the pheromonal stimu- lation was applied to either antenna (Fig. 5B). This turning was elicited even by a single and brief (100 msec) pulsed stimulation. Therefore, it is sug- gested that some neurons in the central nervous system maintain this neural information. Long- lasting increases in firing elicited by pheromones have been described by Olberg [4] in a class of descending neurons (DNs), which project from the brain toward the thoracic motor center, termed “flip-flopping” interneurons, in male B. mori. These neurons exhibited conditional or state de- pendent responses, in which stimuli applied when a neuron was in a state of low-frequency firing elicited accelerated firing, while identical stimuli applied when the neuron was in a state of high- frequency firing caused decelerated firing (hence the term flip-flop). The activity states correlated with changes in turnings during the olfactory- mediated walking. Olberg [4] did not report the initial activity changes of the flip-flop activity. We are interested in the activity changes of these DNs when a short pulsed olfactory stimulation is ap- plied to either antenna. While, in flying male moths Manduca sexta, Kanzaki et al. [24] have reported DNs which show state-dependent activ- ities to stimulation of the antenna by their phero- mone blend. Another group of DNs has been physiologically and morphologically identified by 526 Kanzaki et al. in B. mori {7| and M. sexta [24] that showed long-lasting excitation (LLE) responses to stimulation of the antenna by pheromones, but did not give conditional responses similar to flip- flopping. The morphology of these DNs are simi- lar in both species [7, 24]. These LLE responses were elicited only by pheromonal stimulation in both B. mori [7| and M. sexta [24]. Stimulation of other modalities (e.g. visual and mechanosensory stimuli to the antennae) did not elicit such re- sponses [7, 24]. Thus, similar neural activities exist in pre-motor DNs of both walking and flying moths. This also supports that the orientation of walking and flying moths may be controlled by similar neural systems. B. mori males usually walk with fluttering their wings and they really try to fly, but their bodies are obviously too heavy. We know that they use the wing system similar to flying moths Manduca sexta [26-28] after the analysis of the wing movements (especially wing remote) using a high-speed video camera system (1000 frames/sec), and they show similar zigzag turning only by their wings when their legs have been completely removed [Kanzaki et al. unpublished observation]. It seems that the degeneration of the flight system of B. mori males may not have gone. We predict, therefore, that the zigzag turning program also affects the wing system of the B. mori. Using intracellular recording and _ staining methods, we hope to clarify the role(s) of LLE and state-dependent activity changes of olfactory DNs in the generation of motor activity (1.e., the zigzag turning program) underlying pheromone-mediated upwind walking with fluttering. ACKNOWLEDGMENTS This research was partially supported by Grant-in-Aid from the Ministry of Education, Science and Culture (Nos. 02640547, 03304010), Grant-in-Aid from Bio- Media Program of the Ministry of Agriculture, Forestry and Fisheries of Japan (BMP 91-I-2-4) and also for Insect Pheromone Research from the Shin-Etsu Chemical Co. Ltd. REFERENCES 1 Schneider, D. (1969) Insect olfaction: deciphering system for chemical messages. Science, 163: 1031- 10 11 12 13 14 15 16 R. KANZAKI, N. SuGi AND T. SHIBUYA 1037. Kaissling, K. E. (1971) Insect olfaction. In “Hand- book of Sensory Physiology IV, Chemical Senses, Olfaction”. Ed. by L. M. Beidler, Springer-Verlag, Berlin, pp. 351-431. Obara, Y. (1979) Bombyx mori mating dance: an essential in locating the female. Appl. Ent. Zool., 14(1): 130-132. Olberg, R. M. (1983) Pheromone-triggered flip- flopping interneurons in the ventral nerve cord of the silkworm moth, Bombyx mori. J. Comp. Phy- siol., 152: 297-307. Olberg, R. M. (1983) Interneurons sensitive to female pheromone in the deutocerebrum of the male silkworm moth, Bombyx mori. Physiol. En- tomol., 8: 419-428. Kanzaki, R. and Shibuya, T. (1986) Identification of the deutocerebral neurons responding to the sexual pheromone in the male silkworm moth brain. Zool. Sci., 3(3): 409-418. Kanzaki, R. and Shibuya, T. (1986) Descending protocerebral neurons related to the mating dance of the male silkworm moth. Brain Res., 377: 378- Sho), Kennedy, J. S., Ludlow, A. R. and Sanders, C. J. (1981) Guidance of flying male moths by wind- borne sex pheromone. Physiol. Entomol., 6: 395- 412. Kuenen, L. P. S. and Baker, T. C. (1983) A: non-anemotactic mechanism used in pheromone source location by flying moths. Physiol. Entomol., 8: 277-289. Baker, T. C., Willis, M. A. and Phelan, P. L. (1984) Optomotor anemotaxis polarizes self-steered zigzag- ging in flying moths. Physiol. Entomol., 9: 365-376. Baker, T. C. (1990) Upwind flight and casting flight: Complementary phasic and tonic system used for location of sex pheromone sources by male moths. In “Proc. 10th Int. Symp. Olfaction and Taste”. Ed. by K. B. D@ving, Oslo, pp. 18-25. Kanzaki, R. and Arbas, E. A. (1990) Olfactory and visual influences on fictive flight in Manduca sexta. Soc. Neurosci. Abstr., 16: 758. Willis, M. A. and Arbas, E. A. (1991) Odor- modulated upwind flight of the sphinx moth, Man- duca sexta L. J. Comp. Physiol. A 169: 427-440. Kramer, E. (1975) Orientation of the male silkmoth to sex attractant bombykol. In “Olfaction and Taste V”. Ed. by D. A. Denton and J. P. Coghlan, Academic Press, New York, pp. 329-335. Tobin, T. R. (1981) Pheromone orientation: role of internal control mechanisms. Science, 214: 1147- 1149. Tobin, T. R. and Bell, W. J. (1982) Guidance system for pheromone orientation in moths. Nature, 2957203) 17 18 19 20 21 jul Behavior of Walking Moth to Pheromones Preiss, R. and Kramer, E. (1986) Anemotactic orientation of gypsy moth males and its modification by the attractant pheromone (+ )-disparlure during walking. Physiol. Entomol., 11: 185-198. Willis, M. A. and Baker, T. C. (1987) Comparison of manoeuvres used by walking versus flying Grapholita molesta males during pheromone- mediated upwind movement. J. Insect Physiol., 33(11): 875-883. Mutrlis, J. (1986) The structure of odour plumes. In “Mechanisms in Insect Olfaction”. Ed. by T. L. Payne, M. C. Birch and C. E. J. Kennedy, Oxford University Press, Oxford, pp. 27-38. Murlis, J. Willis, M. A. and Cardé, R. T. (1990) Odour signals: patterns in time and space. In “Proc. 10th Int. Symp. Olfaction and Taste”. Ed. by K. B. Deving, Oslo, pp. 6-17. Willis, M. A. and Baker, T. C. (1984) Effects of intermittent and continuous pheromone stimulation on the flight behaviour of the oriental fruit moth, Grapholita molesta. Physiol. Entomol., 9: 341-358. Baker, T. C. (1986) Pheromone-modulated move- ments of flying moths. In “Mechanisms in Insect Olfaction”. Ed. by T. L. Payne, M. C. Birch, C. E. J. Kennedy, Oxford University Press, Oxford, pp. 39-48. 23 24 25 26 Pai) 28 SZ Kramer, E. (1986) Turbulent diffusion and phero- mone-triggered anemotaxis. In “Mechanisms in In- sect Olfaction”. Ed. by T. L. Payne, M. C. Birch and C. E. J. Kennedy, Oxford University Press, Oxford, pp. 59-67. Kanzaki, R., Arbas, E. A. and Hildebrand, J. G. (1991) Physiology and morphology of descending neurons in pheromone-processing olfactory path- ways in the male moth Manduca sexta. J. Comp. Physiol. A, 169: 1-14. Charlton, R. E. and Cardé, R. T. (1990) Orienta- tion of male gypsy moths, Lymantria dispar (L.), to pheromone sources: the role of olfactory and visual cues. J. Insect Behav., 3(4): 443-469. Kammer, A. E. (1967) Muscle activity during flight in some large Lepidoptera. J. Exp. Biol. 47: 277- 295» Kammer, A. E. (1971) The motor output during turning flight in a hawkmoth, Manduca sexta. J. Insect Physiol. 17: 1073-1086. Rheuben, M. B. and Kammer, A. E. (1987) Struc- ture and innervation of the third axillary muscles of Manduca relative to its role in turning flight. J. Exp. Biol. 131: 373-402. 3 LPL 4 x aifai Ad ye “el —¥) sbeik, , _ aes “angtog, dWatnossuter, sit epee, As eee i “aL (ay fmuneoety cp awed 110 alin BS lente a sar He W Pe fe ae" By erieere aR anne Bea RN ant Re Pinas ¥ ea AG eras Ceara ra aan ae [apa a utes phar isie wives eH ghovinkien wis eeebh se SOP RETE ide to diay ee ee Se as a fa <"« e Wt i, ~o ELE eg , ti a vet - eye ve = F oe oe hy, { eer ase 3 & Le er ej 4 Ad oh he Sax ere wis os Te “5 faci csse rin al aut Bc Paes : ’ r ie ies hae ryt ; eye ie ey ; 5 ; i ae ae = She Tint | 0 ee : ' whine hu i ; : ' oy t ’ > — . re F is A oe 1). 4 ke ee t; sitet mts, ~~ i _ ie = i ea rey FTL ect: eh et fl oie t t hee ai by AS eae . ih J «Sepa BG pram ot he i. E f t Hagey x ray * eh i . Sa! Gj ; 4 2 ; Z ae a i { ‘ ‘ wv a % - (3 ‘er 5 =! . a h) ee es F ? =) 7 x 7 1 t : i ‘ i 7 ; 0 =) E a P of ~~ i f ri R =] ‘ fs 2 = * EL - t t Pp ‘ ; ’ ~S = ZOOLOGICAL SCIENCE 9: 529-532 (1992) © 1992 Zoological Society of Japan Step-Up and Step-Down Photoresponses in Blepharisma TATSUOMI MATSUOKA and Kojl TANEDA Department of Biology, Faculty of Science, Kochi University, Kochi 780, Japan ABSTRACT—The cells of Blepharisma responded to a step-up in light intensity (step-up response) by ciliary reversal, while the cells showed a temporal repression of ciliary reversal which was accompanied by swimming acceleration when light intensity was suddenly decreased (step-down response). The step-up response occurred only in anterior fragments obtained by bisection of the cells, while the step-down response occurred in both of fragments. The topographical difference in photosensitivity indicates that the photoreceptor systems responsible for the step-up photophobic response might be different from those for the step-down photoresponse (repression of ciliary reversal accompanied by acceleration of swimming velocity). INTRODUCTION The cells of Blepharisma show photodispersal that is caused by step-up photophobic response and changes in swimming velocity (photokinetic response) which depends on absolute light inten- sities [1]. The distribution of photoreceptor sys- tems inducing the step-up photophobic response is different from that responsible for the photokinet- ic response; the photoreceptor systems responsible for the step-up photophobic response localize on anterior portion of a cell, while those for photo- kinetic response occur over the entire cell [2]. In the present study, we found that the cells of Blepharisma showed a repression of ciliary rever- sal accompanied by swimming acceleration in re- sponse to a step-down in light intensity. The present paper reports on the distribution of photo- receptor systems that cause these response. MATERIALS AND METHODS Blepharisma japonicum was cultured in 100-fold diluted lettuce juice at 23°C. The cells were collected by low-speed centrifugation, and washed in a standard saline solution containing 1mM CaCl, 1 mM KCI and 5 mM Tris-HCl buffer (pH Accepted March 2, 1992 Received October 21, 1991 7.2). Bisection of the cell was performed with a fine glass needle. Prior to analysis of response, the motility of cells was recorded on a video tape through a dissecting microscope. White light intensity was varied by using neutral density filters, and was determined by a silicon photodiode (S1226-SBK, Hamamatsu Photonics Co.) photometer. To eliminate effects of heat rays, infrared-absorbing filters were placed in front of a light source. Temperature-controlled water (23°C) was circulated beneath the chamber to keep the temperature of the cell suspension HV \ cell cover slip hy paint Fic. 1. A side-view of an apparatus for local photosti- mulation. A small circular region (diameter; 0.5 mm) of lower cover slip was painted black. The lower slip was fixed to a stage of microscope which could be manipulated. Upper cover slip (bottom of a chamber) on which the cell suspension was placed was fixed not to move. Double rays of white light (1.3 W/m? each) was applied to the cells for adapta- tion. 530 T. MATSUOKA AND K. TANEDA constant, except for an experiment of local sti- mulation (Figs. 1, 3). To locally shade a cell, an apparatus shown in Fig. 1 was employed. Double beams of white light (1.3 W/m? each) were applied for at least 30 sec to adapt the cells. One of the beams was applied from the bottom, the other from the upper direction at an angle of 45°C against a horizontal line. In order to shade a certain portion of a cell, one of beams was ob- stracted with a small black-painted area (diameter; 0.5 mm) of a cover slip (Fig. 1, lower slip) which was placed just beneath a fixed chamber. This slip was fixed to a stage which could be manipulated in two dimensional directions. RESULTS AND DISCUSSION When light intensity was suddenly raised, most of intact cells of Blepharisma responded by ciliary £ 100 Bg, ak Q o % 50 aS > gee leat ee ee, hs ona 5 6 88. @. e O o38Sa5 400 B Velocity (m/sec) WwW o o —@- Ae lee ae ee ee 2 e 4 2c 0 mie pf c~ Cc () = 2} ah Z 0 oF (0) 30 ~@=6©60 90 120 150 180 210 240 Time (sec) reversal which continued for 15-20 sec (Fig. 2A), followed by a continuous increase in swimming velocity (Fig. 2B). On the other hand, the intact cells displayed a temporal repression of ciliary reversal (Figs. 2A, 2D) accompanied by a tempor- al increase in swimming velocity (Figs. 2B, 2E) when light intensity was suddenly decreased. The rate of the response gradually decreased to a constant level. anterior fragments responded by ciliary reversal in response to a step-up in light intensity, whereas posterior fragments did not (Fig. 2A). In intact cells, all of the cilia (even in posterior portion) can reverse in response to light irradiation. Therefore, the fact that the posterior Ciliary reversal (%) 350 m/sec) is) Ww ol ro) ro) ro) No (2) (o) Velocity ( rS) a ro) ro) ———_-9————| o— O wp fF ee Time (sec) Light intensity (W/m2) Fic. 2. Photoresponses of anterior fragments (open circles), posterior fragments (closed circles) and intact cells (open squares) of Blepharisma to a step-up and step-down in light intensity. (A) Changes of degree of ciliary reversal to stepwise changes in light intensity. The degree of the response was expressed as the percentage of the total number (n=30-40 cells) showing ciliary reversal in 5 sec. (B) Changes of forward swimming velocity to stepwise changes in light intensity. Bars correspond to the means and SE (determinations of 25-30 cells). (C) Changes in light intensity with time (sec). Light intensities were changed between 0.27 and 2.7 W/m’. The swimming velocity was not measured in a period (dashed line) in which cells rerponded by ciliary reversal. The swimming velocity was determined by tracing the forward swimming cells for 0.5 sec. (D), (E) Scaled-up figures of the step-down responses showed in (A) and (B). (F) Changes in light intensity with time (sec). Photoresponses in Blepharisma 3)! fragments can not respond indicates that photore- ceptor systems for the step-up photophobic re- sponse do not exist in the posterior portion. However, strong light irradiation of the posterior fragments often induced a slight increase in degree of ciliary reversal. This may be attributed to existence of a few photoreceptor systems around middle portion of a cell where the cell is bisected. When light intensity was decreased, both of the fragments responded by a temporal repression of ciliary reversal (Figs.2A, 2D) and swimming acceleration (Figs. 2B, 2E). In bisected fragments, swimming acceleration occurred in 4—5 sec after light intensity was decreased, although the intact cells responded just after stimulation (Fig. 2E). The delay of the response in bisected fragments may be due to some damages by bisection. Uusing intact cells, a local stimulus (a step-down in light intensity) was applied. When either the anterior or posterior was shaded, swimming veloc- ity of the cells increased (Fig. 3). The results also suggest that the photoreceptor systems inducing the step-down response distribute over the entire cell. Degree of temporal acceleration of swimming velocity of the cells which were locally shaded (Fig. 3) was different from that of the cells entirely shaded (Fig. 2). This may be attributed to reasons as follows: (1) Differences in excitability of diffe- rent cell populations. (2) Temperature effect on the response by light irradiation (In Fig. 3, we failed to control the temperature of cell suspen- sion, because gap between two sheets of cover slips reduced the conduction of heat). The topographical difference in photosensitivies of the cells of Blepharisma to a step-up and step-down in light intensity indicates that photore- ceptor systems mediating these responses show in different distribution. In Blepharisma, ciliary re- versal is caused by an increase in intracellular Ca** concentration [3] as well as Paramecium [4]. In Paramecium bursaria, depolarizing photorecep- tor potential which is probably mediated by local- ized photoreceptor systems [5] is caused by a transient increase in the membrane conductance to Ca** [6]. It is presumed that step-up photophobic response (ciliary reversal caused by the step-up in light intensity) of the cells of Blepharisma may be mediated by an activation of photoreceptor Ca 18 1.6 1.4 | PF e72 Velocity (relative unit) 1 Osris | ; | ORR Ohor 1: Oral 5520 Time (sec) Fic. 3. Acceleration of forward swimming velocity of the intact cells of Blepharisma which were locally shaded. The forward swimming velocity (relative unit) was expressed as the rate of the velocity of the cell in 1 sec before the step-down stimulation was applied. Before shading, the cells were adapted to the double beams of white light at least 30sec. Open and closed circles indicate the forward swim- ming velocities of the cells shaded in anterior and posterior portions, respectively. Points and bars correspond to the means and SE (determinations of 20-30 cells). The velocity was determined by trac- ing the swimming cells for 0.5 sec. The cells were shaded in about 1/3 portions from the anterior or posterior end. Duration of the shading was 0.2-0.5 sec. Abscissa; time course (sec) after cells were locally shaded. ; channels. Localization of photosensitivity of the cells of Blepharisma to the step-up in light intensity is possibly attributed to that of photoreceptor Ca channels associated with photoreceptor pigments. In Paramecium, mechanical stimulation of the anterior portion induces a trasient depolarization of the membrane triggered by an activation of mecahnoreceptor Ca channels that is responsible for ciliary reversal, whereas stimulation of the posterior portion leads a temporal hyperpolariza- tion of the membrane triggered by an activation of mechanoreceptor K channels which might be in- volved in a temporal acceleration of swimming velocity [4, 7, 8]. In Blepharisma, a temporal acceleration of swimming velocity induced by a 532 T. MATSUOKA AND K. TANEDA step-down in light intensity might be responsible for photoreceptor K channels as well as the mecha- noreceptor K channels in Paramecium. If so, distribution of photosensitivity of the Blepharisma cells to the step-down in light intensity may be due to the distribution of photoreceptor K channels associated with some photoreceptor pigments. The acceleration of swimming velocity induced by a step-down in light intensity is a temporal photoresponse. In contrast, an increase in swim- ming velocity of Blepharisma caused by a step-up in light intensity, which is not temporal response, is possibly related to continuous hyperpolarized membrane potential, because continuous irradia- tion of strong light reduce frequency of spon- taneous ciliary reversal (unpublished). Preparation works showed that action spectrum for the step-up photophobic response (ciliary re- versal) had three peaks at 580, 540 and 480 nm in visible range. Further examinations involve, by determining action spectrum for the swimming acceleration caused by a step-down in light inten- sity, to compare the action spectrum for the step- up photophobic response with that for the step- down swimming response. Photoreceptor systems responsible for the step- up photophobic response (ciliary reversal) localize anterior portion of the cell; especially anterior end of the cell is most sensitive to light [2]. The cells of Blepharisma avoid from light by step-up photo- phobic response [1]. Therefore, such a localization of photoreceptor systems might be significant, because anterior portion of cell faces light source when the cell is swimming toward light source. On the other hand, swimming acceleration induced by a step-down in light intensity might contribute for the cell to rapidly escape for light when the cell is swimming toward darker region. Hence, photore- ceptor systems responsible or the step-down re- sponse (swimming acceleration) would distribute over the entire cell. REFERENCES 1 Matsuoka, T. (1983) Negative phototaxis in Blephar- isma japonicum. J. Protozool., 30: 409-414. 2 Matsuoka, T. (1983) Distribution of photoreceptors inducing ciliary reversal and swimming acceleration in Blepharisma japonicum. J. Exp. Zool., 225: 337- 340. 3 Matsuoka, T., Watanabe, Y., Kuriu, T., Arita, T. Taneda, K., Ishida, M. Suzuki, T. and Shigenaka, Y. Cell models of Blepharisma: Ca**-dependent modi- fication of ciliary movement and cell elongation. Europ. J. Protistol. in press. 4 Naitoh, Y. and Eckert, R. (1969) Ionic mechanisms controlling behavioral responses of Paramecium to mechanical stimulation. Science, 164: 963-965. 5 Nakaoka, Y. (1989) Localization of photosensitivity in Paramecium bursaria. J. Comp. Physiol., 165: 637-641. : 6 Nakaoka, Y., Kinugawa, K. and Kurotani, T. (1987) Ca’*+t-dependent photoreceptor potential in Para- mecium bursaria. Exp. Biol., 131: 107-115. 7 Naitoh, Y. and Eckert, R. (1973) Sensory mecha- nism in Paramecium. II. Ionic basis of the hyperpo- larizing mechanoreceptor potential. J. Exp. Biol., 59: _ 53-65. 8 Ogura, A. and Machemer, H. (1980) Distribution of mechanoreceptor channels in the Paramecium sur- face membrane. J. Comp. Physiol., 135: 233-242. ZOOLOGICAL SCIENCE 9: 533-539 (1992) © 1992 Zoological Society of Japan Distribution of APGWa-immunoreactive Substances in the Central Nervous System and Reproductive Apparatus of Helix aspersa BERNADETTE GRIFFOND!, JAN VAN MINNEN? and CLAUDE COLARD! ‘Laboratoire de Zoologie et Embryologie, SDI CNRS 6319, Faculté des Sciences, place Maréchal Leclerc, 25030 Besancon Cedex, France, and "Department of Biology, Vrije Universiteit, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands ABSTRACT—The distribution of substances related to the tetrapeptide APGWa was investigated in the central nervous system (CNS) and the reproductive apparatus of Helix aspersa by immunocytoche- mistry. In the CNS, APGWa immunoreactive neurons were detected in all ganglia except the pedal ganglia. Concerning the mesocerebrum of the cerebral ganglia, only neurons of the right mesocerebral lobe reacted positively to the antiserum. In the genital apparatus, positive neurons fibres were seen in the muscular layer of the penis and, in the gonad, an immunoreactive material occurred on the heads of some spermatozoa. On the basis of these observations and of previous electrophysiological studies, an implication of AGPWa-like peptides in the control of mating behaviour is proposed. The significance of the positive reaction of the spermatozoa remains unclear. INTRODUCTION During the last ten years, many neuropeptides have been isolated and sequenced from inverte- brates. Among these, the tetrapeptide APGWa was isolated and purified from the ganglia of the mollusc Fusinus ferrugineus [1]. At the same time, Smit et al. [2] were able to clone a gene from the anterior lobe of the cerebral ganglia of Lymnaea stagnalis which codes for APGWa. This peptide exerts a biological action on various molluscan muscles [3]; in Lymnaea stagnalis for instance, it is supposed to play a role in the control of the penial complex, especially during copulation [2, 4]. In order to collect some informations on the control of reproduction and sexual behaviour in Helix aspersa, we have undertaken immuno- cytochemical studies using antibodies directed against biologically active neuropeptides such as a-caudo-dorsal cell peptide (eaCDCP), a peptide encoded by the genes of the egg-laying hormone of Accepted March 10, 1992 Received February 14, 1992 Lymnaea stagnalis (unpublished results). In the present work, we investigated for the occurrence of APGWa-like substances in the central nervous system and the genital apparatus. The nature of the factors which control the male copulatory behaviour in Helix is not known as yet but some morphological and physiological observations sug- gest that the right mesocerebrum is associated with a sexual function(s). Thus a special attention was paid to this part of the brain. MATERIALS AND METHODS Animals Adult snails (Helix aspersa), bred in the Centre Universitaire d’Héliciculture* under controlled conditions (photoperiod: 18 h light 6 h dark, tem- perature: 20°C, humidity: 100%) were used. Three were sacrificed during copulation, three other were in an active non-reproductive state. * Centre Universitaire d’ Héliciculture—5, rue Ron- chaux—25000 Besancon. 534 B. GRIFFOND, J. V. MINNEN AND C. COLARD Methods Central nervous systems (CNS) and different parts of genital apparatus (ovotestis, sperm- oviduct, prostate, penis, dart pouch and multifid glands) were fixed in Bouin-Hollande fixative (+10% mercuric chloride) for 24hr at room temperature. They were embedded in paraffin and serially sectioned at 6 um. Antibodies to APGWa were raised in rabbits. They were prepared and tested for specificity in the Department of Biology of the free University of Amsterdam according to a previously described procedure [6]. For immunocytochemical observa- tions, the sections were processed with the PAP method [7]. After suppression of possible endoge- nous peroxidase activity by a rinse in phosphate buffer saline (PBS) containing 0.0125% H,Od;, sections were incubated overnight with the pri- mary antiserum (anti-APGWa, dilution 1/1000; 4°C). They were then treated with the second antiserum (goat anti-rabbit, dilution 1/200, 1 hr at room temperature) followed by PAP solution (dilution 1/100; 1 hr at room temperature). The peroxidase was visualized with 0.05% 3,3° diamino- benzidine tetrahydrochloride in PBS containing 0.01% H2Od>. RESULTS Central nervous system Immunoreactive cells were found in all ganglia except the pedal ganglia (Fig. 1). Depending on the animals, weak differences were observed for example in the intensity of staining or in the Fic. 1. Diagram of the central nervous system of Helix aspersa showing APGWa-immunoreactive perikarya (irregular circles) and fibres (short lines). CC, cerebral commissure; CG, cerebral ganglia; PA, parietal ganglia; PD, pedal ganglia; PL, pleural ganglia; V, visceral ganglion; MSC, mesocerebrum; MTC, metacerebrum; PC, procerebrum. number of immunoreactive cells but they do not seem to be correlated with the mating or non- FiG. 2. Immunoreactive perikarya in the left cerebral ganglion. Two groups of small neurons (arrows) and one bigger cell (arrowhead) are located at the external margin. Note the presence of a large positive fibre in the cerebral commissure (CC). FiGe 3: positive reaction (arrow) whereas those of the left lobe (arrowhead) are not stained. commissure; DB, dorsal body area. Fic. 4. < 120. MT, metacerebrum. Section through the cerebral ganglia (CG). The neurons located in the right mesocerebral lobe exhibit a x80. CC, cerebral Right cerebral ganglion. The axons (arrowheads) of the immunoreactive neurons (arrow) of the mesocereb- ral lobe project toward the right interganglionic connectives. 100. MS, mesocerebrum; MT, metacerebrum; PC, procerebrum. Gao) close to the pleural ganglion (PL). Right parietal ganglion (PA). Note the presence of a group of positive cells (arrow) at the posterior margin, 120. n, neuropil. Fic. 6. Small strongly labelled perikarya (arrows) in the left pleural ganglion (PL). Immunoreactive fibres are visible in the cerebro-pleural connective (arrowhead). x 100. 535 AGPWa-immunoreactive Substances in Helix 536 B. GRIFFOND, J. V. MINNEN AND C. COLARD AGPWa-immunoreactive Substances in Helix 537 mating state. Cerebral ganglia: Three major groups of posi- tive cells could be distinguished in each cerebral ganglion. Two cell groups were located at the lateral external margin of the metacerebrum (Fig. 1). The first group consisted of about 20 small cells (10 wm in diameter) and 1 larger cell (diameter= 25-30 wm) (Fig. 2). The second group, located posteriorly, consisted of about 15 cells measuring 20 wm in diameter (Fig. 2). The third group was exclusively found in the anterior lobe of the right mesocerebrum only (Figs.3, 4). Among the population of pear-shaped cells (40-45 mm in dia- meter) characteristic of this area, 40 to 60 showed a positive reaction; their perikarya exhibited a less intense staining than those of the other groups; their axons formed a bundle running across the right cerebral ganglion to the origin of the cerebro- pleural and pedal connectives (Fig. 4); from this area, it was difficult to follow the axonal projec- tions, most of them apparently ran through the right cerebro-pedal connective to the pedal gan- glia. The right cerebro-pedal and cerebro-pleural connectives contained many more fibres than those on the left side. Positive fibres were present in the neuropil of each cerebral ganglion and large axons crossing the commissure were found (Fig. 2). Furthermore, immunoreactive fibres were found in some nerve roots such as the labial, penial and peritentacular nerves. Parietal ganglia: About 10 cells of 25-30 ~#m in diameter in the anterior part of the ganglia, close to the pleural ganglia showed a positive reaction. At the posterior part, 7 elongated cells (20 10 wm) and 2 larger cells (30 4m) were found, occupying similar positions near the pleural gan- glia (Fig. 5). In addition, a cluster of 7 to 10 darkly stained cells (15 um in diameter) were located in the left parietal ganglion close to the anterior margin of the visceral ganglion. Several other positive perikarya were occasionally dispersed in the anterior half of the left parietal ganglion. A network of immunoreactive varicose fibres could be observed in the neuropils. Pleural ganglia: Most of the positive cells were located at the lateral external side of the ganglia, 16 cells (12-15 xm) at the anterior part and about 15 cells (diameter 25 um) posteriorly. In each ganglion a cluster of 40 small, strongly labelled, neurons (10-12 ~m) occurred at the basis of the cerebro-pleural connective (Fig.6). Another group consisting of about 40 intensely stained cells (10-15 ~m) was observed medially on the anterior face of the ganglia (Fig. 7). About 12 additional cells of 15 ~m in diameter were scattered in each ganglion and 2 neurons (25-30 um) were visible posteriorly, near the parietal ganglia. Prominent bundles of immunoreactive fibres were observed in the neuropils (Fig. 7). Visceral ganglion: 2 or 3 large neurons (140 ym) located at the posterior face stained moder- ately (Fig. 8). Positive fibres were revealed in the neuropil and in the nerve roots. Pedal ganglia: Although no immunoreactive perikarya were found, many positive fibres were present in the neuropils, running in different direc- tions towards the connectives and the nerves leav- ing these ganglia. Many more fibres were present in the right ganglion than in the left. Genital apparatus Among the investigated organs, immunoreac- tivity was only found in the penis and the gonad. In the penis, positive fibres were relatively numer- ous, especially in the proximal half. Thick tracts of varicose and non-varicose fibres, usually oriented in a longitudinal direction, ran in the well- developed muscular layer (Fig. 9). Small lightly stained peripheral neurons were seen occasionally (Fig. 9). In some gonads, a clear label occurred on Fic. 7. Cluster of strongly stained neurons (arrow) in the medial part of the right pleural ganglion. Arrowheads: positive fibres in the neuropil. x 290. Fic. 8. Visceral ganglion (VG). Note the presence of a large neuron whose perikaryon and axon are moderately stained (arrow). 120. n, neuropil. Fic. 9. Longitudinal section of the penis. Positive fibres (arrows) are observed in the muscular layer (M). A small lightly stained cell is visible (arrowhead). 120. E, epithelial layer. Fic. 10. Gonadal acini. Note the strong reaction (arrows) on the heads of bundled spermatozoa. 290. 538 B. GRIFFOND, J. V. MINNEN AND C. COLARD the heads of bundles of spermatozoa (Fig. 10) whereas the heads on other bundles: remained negative. DISCUSSION Our study demonstrates the presence of APGWa-immunoreactive substances in the CNS, the penis and the gonad of Helix aspersa. Whether the immunoreactive material is APGWa or cross- reacting substances cannot be determined by im- munocytochemistry and should be determined by means of biochemical and/or molecular biological techniques. However, since APGWa had been demonstrated in distantly related molluscan spe- cies such as Lymnaea stagnalis (pulmonates) and Fusinus ferrugineus (prosobranchs), it may be assumed that APGWa occurs in Helix aspersa. Except the pedal ganglia, all ganglia contain im- munoreactive neurons and most of the nerves possess immunoreactive fibres. This large distribu- tion is consistent with the proposed role of APGWa in the muscular physiology: it has mod- ulatory effects on the contractions of various mol- luscan muscles; depending on the muscles, it shows a stimulatory or an inhibitory action [1]. Concern- ing the penis, the peptide is probably involved in the contraction or the relaxation of the muscular layer during copulation. A comparable action of APGWa was also demonstrated in Lymnaea stag- nalis, where APGWa antagonizes dopamine- and serotonin-induced contractions of the penis retrac- tor muscle [2]. The origin of the positive fibres in the penis of Helix is unknown; according to pre- vious descriptions, the axons of the penial nerve originate from the right pedal ganglion [8, 9]. As we never observed immunoreactive neurons in this ganglion, we can hypothesize that 1) a number of processes travel directly from the cerebral ganglia to the penis, 2) axons from different ganglia follow a lengthy way via the pedal ganglia to the penis. The presence of APGWaz-like substances in the right mesocerebral lobe is of great interest. Other peptides were immunocytochemically revealed in the mesocerebrum but they always show a bilateral distribution; this is the case for example for methionine-enkephalin [10], FMRFa [10] and somatostatin [11]. The asymmetrical occurrence of APGWa-like substance suggests that they might be responsible for the control of asymmetrical organs, in particular the reproductive organs which are located on the right side of the body. According to Chase [5], a mean number of 138 cells was counted in the right mesocerebrum of Helix aspersa. In our investigations, a high proportion of these cells (more than a quarter to a third) were APGWa- positive; their axons were traced to the right cerebro-pedal connective. This observation is in agreement with the results of dye-injections in mesocerebral neurons of the right side, demon- strating that the axons project almost without exception in the ipsilateral cerebro-pedal connec- tive to the right pedal ganglion [5]. From elec- trophysiological studies [5, 12], we know that the mesocerebrum is a centre for coordinating the execution of mating. Firstly, electrical stimula- tions of the right mesocerebrum provoke large movements of the penis and the dart sac; these effects are mediated by axons that travel directly to the right pedal ganglion in which the motor- neurons lie [5]; secondly, the right mesocerebrum is responsible for the suppression of withdrawal responses during mating, by acting on the parietal command [12]; and thirdly, it can inhibit simul- taneously a group of serotonergic neurons located in the pedal ganglia and capable of sensitizing the afferent excitation of the avoidance command cells [12]. Interestingly, also in Lymnaea stagnalis a - group of serotonergic neurons in the right pedal ganglia, the so called pedal Ib cluster, are inner- vated by APGWa containing neurons in the right anterior lobus of the cerebral ganglia [4]. Our results suggest that APGWa-immunoreactive sub- stances of the right mesocerebral participate at least at one of these functions but their mechanism of action remains to elucidate. The presence of APGWa-immunoreactive sub- stances on the head of spermatozoa is surprising. It is unlikely that the immunoreactive material could be the neuropeptide because we did not observe positive nervous fibres around the gonad; it consists more probably of immunologically re- lated substances whose nature and significance must be investigated. AGPWa-immunoreactive Substances in Helix 539 ACKNOWLEDGMENTS The authors wish to thank Brigitte Jolibois and Bruno Régent for the organization and illustrations. REFERENCES Kuroki, Y., Kanda, T., Kubota, I., Fujisawa, Y., Ikeda, T., Muira, A., Minamitake, Y. and Muneoka, Y. (1990) A molluscan neuropeptide related to the crustacean hormone, RPCH. Biochem. Biophys. Res. Commun., 167: 273-279. Smit, A. B., Li, K. W., Van der Schors, R. C., Van Minnen, J., Kits, K. S., Geraerts, W. P. M. and Joosse, J. (1990) A novel neuropeptide involved in male reproductive behaviour of Lymnaea stagnalis. Proceedings of SYMON III (Symposium on Mollus- can Neurobiology), Amsterdam, p. 125. Kobayashi, M. and Muneoka, Y. (1990) Structure and action of molluscan neuropeptides. Zool. Sci., 7: 801-814. Croll, R. P., Van Minnen, J. and Smit, A. B. (1990) Distribution of the neuropeptide APGW-NH,; in the central nervous system and male reproductive organs of Lymnaea stagnalis. Proceedings of SYMON III (Symposium on Molluscan Neurobio- logy), Amsterdam, p. 91. Chase, R. (1986) Brain cells that command sexual behavior in the snail Helix aspersa. J. Neurobiol., 17: 669-679. Van Minnen, J., Van den Haar, J., Raap, A. K. and 10 11 12 Vreugdenhil, E. (1988) Localization of ovulation hormone-like neuropeptide in the central nervous system of the snail Lymnaea stagnalis by means of immunocytochemistry and in situ hybridization. Cell Tissue Res., 251: 477-484. Sternberger, L. A. (1979) Immunocytochemistry. John Wiley and Sons, New York, 2nd ed. Van Mol, J. J. (1967) Etude morphologique et phylogénétique du ganglion cérébroide des Gastér- opodes Pulmonés (Mollusques). Acad. Roy. Belg., 37: 1-168. Franc, A. (1968) Sous-classe des Pulmonés. In “Traité de Zoologie. Anatomie, systématique, biologie”. Ed. by P. P. Grassé, Masson et Cie, Paris, 5: pp. 325-607. Marchand, C. R., Griffond, B., Mounzih, K. and Colard, C. (1991) Distribution of methionine- enkephalin-like and FMRFamide-like immuno- reactivities in the central nervous system (including dorsal bodies) of the snail Helix aspersa Miller. Zool. Sci., 8: 905-913. Baud, C., Colard, C. and Marchand, C. R. (1991) Mise en é€vidence de la présence et des lieux de synthése d’une substance apparentée a la somato- statine par immunocytochimie et hybridation in situ dans le cerveau d’ Helix aspersa. Cell. Molec. Biol., 37: 205-212. Balaban, P. and Chase, R. (1990) Stimulation of mesocerebrum in Helix aspersa inhibits the neural network underlying avoidance behavior. J. Comp. Physiol. A, 166: 421-427. wine: sided AR ite wR vallaicet Isi@b ba Pen ai T sbabest ai bhe SOL banked | Paanl. 4e, b- gage. gf “ab sababryy ah, Se . z S 4 ; ¥ Bel v7 X. . f ; i 12 290! A” : lose a b. = eS Set »§ 1 ¥ ‘ “74 pas ; ’ “7 z 2 Pe ‘ i! = f. but} st a » é 5 2: j i i 1 ie me 4 wa ty = Te 1 l A =4 » by # x ee = id 7 By “— ‘ = © & + ¥ > = ? 5 solution (3 mg 4-chloro 2-naphthol in 1 ml ethanol, 5 ml Tris-buffered saline, pH 7.4, and 1 yl 3% H,O>) was used as the HRP substrate. Fibronectin-Like Molecule of Pinctada 543 Immunohistochemistry For the immunohistochemical detection of GBP in the pearl oyster, specimen was dissected and fixed in Bouin’s fixative at 4°C for 6hr. After washing in PBS three times for 1 hr each, the samples were embedded in paraffin and sliced into 5 wm sections. After removing paraffin, the sec- tions were incubated with 5% BSA, then with anti-GBP or normal rabbit serum (1: 100 dilution with PBS) for 20min. The sections were then washed with PBS and incubated with fluorescein isothiocyanate (FITC)-conjugated swine anti- rabbit Ig (FITC anti-rabbit Ig; Dakopatts) at a dilution of 1:100. To detect GBP and type-I like collagen in amebocytes, blood was dropped onto glass slides which were then maintained for 20 min at room temperature to help amebocytes adhere. Cells were fixed with 5% paraformaldehyde in PBS for 20 min and treated with acetone. Samples were treated with anti-GBP or anti-collagen and then FITC anti-rabbit Ig as described. To observe GBP in the ECM synthesized in vitro, amebocytes were cultured for 6 days and fixed as previously reported [5]. The samples were incubated with anti-GBP (1:100) for 20min, washed and incubated with HRP anti-rabbit Ig (1:100) for 20min. Diaminobenzidine (DAB)- HO; solution (1 mg 3,3’-diaminobenzidine in 5 ml of 50 mM Tris/HCl buffer, pH 7.6, and 5 wl 5% H,0O>) was used as the substrate for HRP. In the wound experiment, a shell bead, 7mm diameter, was implanted into the gonad of a pearl oyster by means of an incision as previously re- ported [5]. The gonad was dissected from the specimen 7 days after the operation and processed for indirect immunofluorescence as described. RESULTS Purification of GBP Pearl oyster hemolymph was passed through a column of Sepharose 4B, and the flow through fractions were applied to an affinity column with gelatin-Sepharose using the batch method. The retained proteins were recovered from the gel with a small volume of buffered 4M urea. The eluate gave a homogeneous protein band at the position of 220 kDa in SDS-PAGE (Fig. la, b). A subse- quent elution of the column with 8 M urea did not yield any detectable proteins. When Sepharose coupled with bovine serum albumin was used for affinity chromatography instead of gelatin- Sepharose, the 220kDa polypeptide was not obtained. Therefore, the adsorption of the mol- ecule on the gelatin-Sepharose in probably due to specific binding with gelatin. Thus, we refer to the substance as gelatin binding protein (GBP). Approximately 20 ug of GBP was isolated from 10 ml of hemolymph containing 0.9-1.2 mg protein/ ml. Molecular structure Affinity-purified GBP gave a molecular mass of 220 kDa in the SDS-PAGE under reducing and non-reducing conditions (Fig. la, b). These results suggest that GBP is a 220kDa monomer of polypeptide in the hemolymph. oe 2b © a ¢ ©! 200 SS * Mm. * | 16 = | 335— : 200=) "mq 66 = | | | 16 =| 42 = | | | | Silver—staining Immunoblot Immunoblot Fic. 1. SDS-PAGE and immunoblot analyses of gela- tin-binding protein (GBP) purified from pearl oyster hemolymph. Affinity-purified GBP using gelatin- Sepharose was electrophoresed using 7.5% (a-d) or 5.0% (e-f) polyacrylamide gel and subjected to protein-staining or western blotting. a and D: silver- stained gel. c-f: nitrocellulose sheet immunostained using anti-GBP. In each lane, 0.14 wg of protein was run. R, reduced by 2-mercaptoethanol; N, non-reduced. 544 T. SUZUKI AND S. FUNAKOSHI Cell-spreading activity The cell-spreading activity of GBP was studied using BHK cells. Some of the cells elongated on the GBP-coated dish (Fig. 2b). However, the degree of spreading was incomplete as compared with the cells on the bovine fibronectin-coated dish which were fully flattened (Fig. 2c). When partial- ly spread cells were included, the ratio of spread- ing cells was 16% higher on the GBP-coated than on the non-coated dish (Table 1). Fic. 2. Effect of GBP on the spreading of BHK cells. BHK cells were incubated for 1 hr on non-coated (a), GBP-coated (b) and bovine fibronectin-coated (c) dishes. TABLE 1. Ratio of BHK cells spread on non- coated, GBP-coated and bovine fibronectin- coated dish. Cells spread on dish Coating x 100 Total cells None 8* GBP 24* Bovine fibronectin 94 * Cells as arrowed in Fig. 2a, b were included. Localization In the immunoblot analysis using 7.5% poly- acrylamide gel, anti-GBP recognized a 220 kDa polypeptide under non-reducing conditions, but the staining intensity was weak under reducing conditions (Fig. lc, d). By using a 5.0% gel a protein was recognized by anti-GBP under both conditions (Fig. le, f). Tissue localization of GBP was investigated by indirect immunofluorescence. Fluorescent staining was noted at the following connective tissues; those under the epithelium in the byssus gland, body wall and kidneys (Fig. 3b, d, f); endomysium of the muscle (Fig. 3h); and loose connective tissue distributed between the digestive diverticula and the body wall (Fig. 3j). In particular, intense staining was observed adjacent to the basement membrane of the epithelium in the byssus gland and body wall. No fluorescence was detected in tissues when normal rabbit serum was used as a control. The 4M urea extract (adductor muscles tissue) was immunoblotted using 5% polyacrylamide gel to determine the structure of tissue GBP. The adductor muscle was chosen for study because of its immunohistochemical reactivity with anti-GBP and the availability of large quantities. Under reducing conditions, anti-GBP produced an in- tense band at the 220kDa position (Fig. 4c). Under non-reducing conditions, a weakly stained ~ band was detected at the position of approximately 450 kDa (Fig. 4d), suggesting that the 220kDa polypeptide was dimerized or more highly poly- merized in the tissue. These bands were not visualized when using Coomassie Brilliant Blue for protein staining (Fig. 4a, b), probably because the concentration was below a detectable level. Immunological properties Immunological cross-reactivity of GBP with mammalian fibronectins was tested by immunoblot analysis. Anti-GBP did not react with either human or bovine fibronectin. Neither did anti- — human and anti-bovine fibronectin react with GBP. Thus, GBP and mammalian fibronectins lack mutual immunological cross-reactivity. In addition, since no hemolymph proteins reacted with anti-human and anti-bovine fibronectin, the pearl oyster has no detectable level of immuno- logically relevant substances to mammalian fibronectins in the hemolymph. Production of GBP. by amebocytes GBP in pearl oyster amebocytes was observed by indirect immunofluorescence. The agranular amebocytes exhibited a reaction to anti-GBP in the cytoplasm, as granular staining (Fig. 5a). On the other hand, the cells did not react with anti- collagen. Next, we examined the location of GBP in amebocytes cultured for 6 days to determine Fibronectin-Like Molecule of Pinctada 545 TESaN SAE Zr ae See eee PE AL SRB he Kk I graye ian paseo Bread egy ie ticene Ps Seer SUS ; et om Se Pee gis fe bie See we ~en - os th ee 4 Boel we: : g See RTT e eNO Fic. 3. Immunolocalization of GBP in the pearl oyster. The photographs on the right are of indirect im- munofluorescence using anti-GBP. Those on the left are of phase contrast micrography of the same field as that on the right. a and b: byssus gland. The arrows indicate the basement membrane. c and d: body wall. The arrows show the basement membrane. e and f: kidney. g and h: muscle. i and j: loose connective tissue between digestive diverticula and body wall. 546 T. SUZUKI AND S. FUNAKOSHI 4M urea extract of adductor muscle RRS ana ee a b. ©¢ d. Eo 2a. tC kDa § 669= | = th ¥ 200= » 16 = Coomassie RN oes ny mmunoblot whether amebocytes secrete GBP into ECM. An ECM started to be deposited inside the aggregates of amebocytes after 3-4 days of culture, as re- ported by Suzuki et al. [5]. When enzyme-labeled staining was applied to 6-day cultured aggregates, the gel-like matrix in the aggregates displayed intense reaction with anti-GBP (Fig. 5d). Fibrillar staining was also detected around it. Thus, GBP was deposited de novo in the ECM secreted by amebocytes. In the control, using pre-immune serum, neither amebocytes nor the in vitro matrix Fic. 4. Detection of GBP in pearl oyster adductor muscle of western blotting. 4M urea extract of adductor muscle was electrophoresed in 5% poly- acrylamide gel and subjected to protein staining or immunoblotting. a and b: Coomassie-stained gel. c and d: nitrocellulose sheet immunostained using anit-GBP. In each lane, 20 ug of protein was run. R and N, same as in Fig. 1. Fibronectin-Like Molecule of Pinctada 547 Fic. 6. Immunolocalization of GBP and type I-like collagen in the amebocyte sheath formed around an abiotic implant (shell bead) inserted into gonad (on day-7 after wounding). a: hematoxylin-eosin staining. b and c: indirect immunofluorescence using anti-GBP and anti-type I like collagen, respectively. G, gonad; I, abiotic implant; S, amebocyte sheath. were stained. Finally, GBP was localized at the experimental wound site. In this experiment, an abiotic implant (shell bead) was inserted into the gonad via an incision. At 7-day post wounding, the agranular amebocytes formed a cellular sheath, covering the implant (Fig.6a). Reactions to anti-GBP and anti-collagen were detected in the amebocyte sheath (Fig. 6b, c). DISCUSSION Mammalian fibronectin exists in a soluble form in plasma and as an insoluble form in tissues and cell surfaces [28]. Plasma fibronectin has an approximate 450 kDa molecular mass and consists of two similar subunits of 220-240 kDa covalently bound by two disulfide bonds [29]. Plasma fibronectin is synthesized and secreted mainly by hepatocytes [30]. On the other hand, cellular fibronectin is synthesized by various cells such as fibroblasts, myoblasts and endothelial cells, and is highly polymerized as an insoluble form with other ECM components [31]. One major biological function of fibronectin is mediation of cell- adhesion. In invertebrates, fibronectin(-like) molecules have been isolated and some of their biochemical properties have been described in coelenterates [19], sea-urchin Pseudocentrotus depressus [16, 32], snail Helix aspersa [17] and Drosophila {18}. Each type has the ability to bind to denatured mammalian collagen (i.e., gelatin). Tissue fibronectins of these animals structurally coincide well with mammalian fibronectin in both molecular weight and the possession of interchain disulfide bonds. However, the subunit structure of the soluble form of invertebrate fibronectins has not been described. Sea-urchin fibronectin promotes the spreading of BHK cells [32], so that the mediation of cell-adhesion seems to be a universal property of both vertebrate and invertebrate fibronectin(-like) molecules. In this study, we fractioned hemolymph of the i ee ee ee ee ae Fic. 5. Immunolocalization of GBP in agranular amebocytes (a and b) and in an amebocyte aggregate depositing extracellular matrix (c and d). a: indirect immunofluorescence using anti-GBP. The cells were stained soon after blood collection. b: phase contrast micrography of the same field as that on a. c: phase contrast micrography of amebocyte aggregate cultured for 6 days. The extracellular matrix deposited inside the aggregate is shown by the arrows. d: enzyme-labeled antibody staining using anti-GBP showing that the matrix (arrows) is reactive. 548 T. SUZUKI AND S. FUNAKOSHI pearl oyster, Pinctada fucata, by gelatin-affinity chromatography under conditions used for purify- ing mammalian plasma fibronectin [24]. This pro- cedure yielded a monomeric gelatin-binding pro- tein (GBP). GBP was similar in size to fibronectins with regard to the single polypeptide chain, but differed from it in subunit structure. Proteolysis of fibronectin yields a fragment almost as large as the original subunits when the first proteolytic cleav- age occurs at the C-terminal end [33]. Among bivalves which have not developed a humoral clotting system [34], it is inconceivable that hemolymph proteins had rapidly degenerated under the purification conditions used _ here. However, it remains possible that the 220 kDa polypeptide is a product of proteolytic cleavage which may occur in vivo. GBP was localized at the connective tissues of various organs using immunohistochemical techni- ques. Immunoblot analysis suggested the exis- tence of an insoluble form of GBP in the tissues, which is polymerized by disulfide linkages. Thus, GBP seems to be present in the connective tissues as an insoluble component of the ECM. GBP exhibited a low level of cell-spreading activity toward BHK cells compared with that of bovine fibronectin. One hypothesis is that the affinity of GBP for plasma membrane receptors of BHK cells is low due to the wide evolutionary distance between mammals and molluscs. For critical evaluation, GBP cell-spreading activity should be examined with cells of the pearl oyster. Therefore, we prepared primary cells (fibroblast- like) from several tissues of the pearl oyster, but, unfortunately, all cells rapidly spread in non- coated (control) plastic wells even in the physio- logical saline (data not shown). In conclusion, GBP may be a cell-adhesive protein of the pearl oyster which exists in blood as a single 220 kDa polypeptide and also in tissue as polymerized insoluble form. GBP of the pearl oyster is a probable homologue (fibronectin-like molecule) to mammalian fibronectin, with which it shares some properties. Immunologically, GBP had no_ cross-reactivity with mammalian fibronectins. Since sea urchin and _ bovine fibronectins lack mutual cross-reactivity [16], im- munological properties do not always serve as criteria for identifying fibronectin. The molluscs have a wound healing system which widely differs from that of mammals [1-5]. In these animals, the wound site is healed via following four cellular reactions; the removai of tissue debris, cellular sheath formation, ECM pro- duction and epithelial regeneration. The cellular sheath formation is thought to be a hemostatic reaction in molluscs, which lack a humoral clotting system. In the pearl oyster, the agranular amebo- cytes, which are macrophage-like cells, are re- sponsible for not only phagocytosis of debris but also the two successive healing reactions [5]. The epithelial cells migrate along the ECM newly sec- reted by the sheath amebocytes [5]. The agranular amebocytes have ability to se- crete type I-like collagen and proteoglycans as components of the ECM, in vitro [5]. The present study demonstrated that GBP exists in the ECM produced by the amebocytes, suggesting that it is secreted with other ECM components by the cells. At the experimental wound site in the gonad, where an abiotic implant (shell bead) was inserted via an incision, the agranular amebocytes formed a sheath of 10-20 cell layers to cover the implant, after which the ECM began to be deposited in the spaces between sheath cells [5]. It was ascertained that GBP is deposited with type I-like collagen in the amebocyte sheath. These data suggest that GBP is synthesized and secreted by the agranular — amebocytes with other ECM components during wound healing. It is hypothesized that GBP acts as a mediator of cell-adhesion for migrating epithelial cells at wound sites in the pearl oyster. ACKNOWLEDGMENT This study was supported by a grant in aid (Bio-Media Program) from the Ministry of Agriculture, Forestry and Fisheries (BMP-91-II-2-4). We are grateful to Dr. J. Scarpa for linguistic improvements of the manuscript. REFERENCES 1 DesVoigne, D. M. and Spark, A. K. (1968) The process of wound healing in the Pacific oyster, Crassostrea gigas. J. Invert. Pathol., 12: 53-65. 2 Pauley, G. B. and Heaton, L. H. (1969) Ex- perimental wound repair in the freshwater mussel 10 11 12 13 14 15 16 Fibronectin-Like Molecule of Pinctada Anodonta oregonensis. J. Invert. Pathol., 13: 241- 249. Ruddell, C. L. (1971) The fine structure of oyster agranular amebocytes from regenerating mantle wounds in the Pacific oyster, Crassostrea gigas.J. Invert. Pathol., 18: 260-268. Sminia, T., Pietersma, K. and Scheerboom, J. E. M. (1973) Histological and ultrastructural observations on wound healing in the freshwater pulmonate Lym- naea stagnalis. Z. Zellforsh., 141: 561-573. Suzuki, T., Yoshinaka, R., Mizuta, S., Funakoshi, S. and Wada, K. (1991) Extracellular matrix pro- duction by amebocytes during epithelial regenera- tion in the marine bivalve Pinctada fucata. Cell Tissue Res., 266: 75-82. Bohn, H. (1975) Growth-promoting effect of haemocytes on insect epidermis in vitro. J. Insect Physiol., 21: 1283-1293. Bohn, H. (1977a) Conditioning of a glass surface for the outgrowth of insect epidermis (Leucophaea maderae, Blattaria). In Vitro, 13: 100-107. Bohn, H. (1977b) Enzymatic and immunological characterization of the conditioning factor for epidermal outgrowth in the cockroach Leucophaea maderae. J. Insect Physiol., 23: 1063-1073. Clark, R. A. F., Lanigan, J. M., DellaPelle, P., Manseau, E., Dvorak, H. F. and Colvin, R. B. (1982) Fibronectin and fibrin provide a provisional matrix for epidermal cell migration during wound reepithelialization. J. Invest. Dermatol., 79: 264- 269. Fujikawa, L. S., Foster, C. S., Gipson, I. K. and Colvin, R. B. (1984) Basement membrane compo- nents in healing rabbit corneal epithelial wounds: immunofluorescence and ultrastructural studies. J. Cell. Biol., 98: 128-138. Yamada, K. M. (1988) Fibronectin domains and receptors. In “Fibronectin.” Ed. by D. F. Mosher, Academic Press, San Diego, pp. 48-121. Bronner-Fraser, M. (1986) An antibody to a recep- tor for fibronectin and laminin perturbs cranial neural crest development. Dev. Biol., 117: 528-536. Boucaut, J. C., Darribere, T., Boulekbache, H. and Thiery, J. P. (1984) Prevention of gastrulation but not neurulation by antibodies to fibronectin in amphibian embryos. Nature, 307: 364-367. Thiery, J. P., Duband, J. L. and Delouvee, A. (1982) Pathways and mechanism of avian trunk neural crest cell migration and localization. Dev. Biol., 93: 324-343. Akiyama, S. K. and Johnson, M. D. (1983) Fib- ronectin in evolution: presence in invertebrates and isolation from Microciona prolifera. Comp. Biochem. Physiol., 76B: 687-694. Iwata, M. and Nakano, E. (1981). Fibronectin from the ovary of the sea urchin, Pseudocentrotus depress- 17 18 19 20 7a) Ms 23 24 75) 26 M2) 28 29 30 549 us. Wilhelm Roux’s Archives., 190: 83-86. Bride, M., Barre, P., Griffond, B. and Bride, J. (1988) Localization of a fibronectin-like molecule in the ovotestis of the snail Helix aspersa. Cell Tissue Res. 254: 421-428. Gratecos, D., Naidet, C., Astier, M., Thiery, J. P. and Semeriva, M. (1988) Drosophila fibronectin: a protein that shares properties similar to those of its mammalian homologue. EMBO J., 7: 215-223. Schlage, W. K. (1988) Isolation and characteriza- tion of a fibronectin from marine coelenterates. Eur. J. Cell Biol., 47: 395-403. Labat-Rabert, J., Robert, L., Auger, C., Lethias, C. and Garrone, R. (1981) Fibronectin-like protein in Porifera: its role in cell aggregation. Proc. Natl. Acad. Sci. USA., 78: 6261-6265. Katow, H., Yamada, K. M. and Solursh, M. (1982) Occurrence of fibronectin on the primary mesen- chyme cell surface during migration in the sea urchin embryo. Differentiation, 22: 120-124. Katow, H., Yazawa, S. and Sofuku, S. (1990) A fibronectin-related synthetic peptide, Pro-Ala-Ser- Ser, inhibits fibronectin binding to the cell surface, fibronectin-promoted cell migration in vitro, and cell migration in vivo. Exp. Cell Res., 190: 17-24. Suzuki, T. and Mori, K. (1989) A galactose-specific lectin from the hemolymph of the pearl oyster, Pinctada fucata martensii.Comp. Biochem. Physiol., 92B: 455-462. Ruoslahti, E., Hayman, E. G., Pierschbacher, M. and Engvall, E. (1982) Fibronectin: purification, immunochemical properties, and biological activi- ties. Methods. Enzymol., 82: 803-831. Isemura, M., Yamaguchi, Y., Munakata, H., Aika- wa, J., Kan, M., Yamane, I. and Yosizawa, Z. (1984) Isolation and characterization of human placenta fibronectin. J. Biochem., 96: 163-169. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacter- iophage T4. Nature, 227: 680-685. Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacryla- mide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA., 76: 4350-4354. Yamada, K. M., Hayashi, M., Hirano, H. and Akiyama, S. K. (1984) Fibronectin and cell surface interactions. In “The role of extracellular matrix in development.” Ed, by D. F. Mosher, Alan R. Liss, New York, pp. 89-121. Kornblihtt, A. R. and Gutman, A. (1988) Molecu- lar biology of the extracellular matrix proteins. Biol. Rev., 63: 465-507. Tamkun, J. W. and Hynes, R. O. (1983) Plasma fibronectin is synthesized and secreted by hepato- cytes. J. Biol. Chem., 258: 4641-4647. Si 32 38) 550 Hynes, R. O. and Yamada, K. M. (1982) Fi- bronectin: multifunctional modular glycoproteins. J. Cell. Biol., 95: 369-377. Iwata, M. and Nakano, E. (1983) Characterization of sea-urchin fibronectin. Biochem. J., 215: 205- 208. Mosesson, M. W., Chen, A. B. and Huseby, R. M. 34 T. SUZUKI AND S. FUNAKOSHI (1975) The cold-insoluble globulin of human plas- ma: studies of its essential structural features. Biochim. Biophys. Acta, 386: 509-524. Needham, A. E. (1970) Haemostatic mechanisms in the invertebrates. Symp. Zool. Soc. Lond., 27: 19- 44. ZOOLOGICAL SCIENCE 9: 551-562 (1992) © 1992 Zoological Society of Japan Electron Microscopic Analysis of Tunicate (Halocynthia roretzi) Hemocytes HoncwE!I ZHANG!, TomMoo SAWADA?, EDWIN L. Cooper’ and Susumu TomonaGa! ‘School of Allied Health Sciences, Yamaguchi University, Ube 755, Japan, "Department of Biology, Shandong University, Jinan, P. R. China, >Department of Anatomy, Yamaguchi University School of Medicine, Ube 755, Japan and “Department of Anatomy and Cell Biology, University of California School of Medicine, Los Angeles, California 90024, U.S.A. ABSTRACT—Hemocytes from hemolymph of the tunicate, Halocynthia roretzi were investigated by transmission electron microscopy (TEM). Nine types were identified according to their ultrastructural characteristics: phagocytes or macrophages (PH cells), granulocytes with small granules (GS cells), granulocytes with large granules (GL cells), vesicle-containing cells (VC cells), fibrous material- containing cells (FM cells), vacuolated cells type 1 (VA1 cells), vacuolated cells type 2 (VA2 cells), basophilic cells (BA cells) and lymphoid cells (LY cells). Among these hemocyte types VC cells and FM cells were unique and novel. One functional assay, i.e. phagocytic activity against sheep red blood cells (SRBC) and rat red blood cells (RRBC) was developed. This investigation resulting from improved fixation has served as the basis for standardizing hemocyte types and for defining future analyses that can be used in functional assays. INTRODUCTION Tunicate hemolymph contains many hemocyte types often referred to as coelomocytes or coelo- mic cells. They have long been investigated by light and phase contrast microscopy in diverse species [1-9]. In addition, more recent studies have correlated fine structure with certain func- tional assays [10-13]. Despite attempts to define certain hemocyte functions such as coagulation, excretion, nutrition, immune responses and tunic formation [14-19], as summarized by Wright (1981), most of these functional analyses require more extended experimentation [20]. The ultrastructural characterization of hemo- cytes has been established, in some species [21- 23], leaving us with confusion in terminology with respect to common features of hemocytes from different species. Recently we classified hemo- Accepted March 4, 1992 Received October 21, 1991 cytes of the tunicate, Halocynthia roretzi, into ten groups according to morphology by light, phase and fluorescence microscopy [24]. The present study was undertaken for several reasons. First, confusion in reaching a consensus concerning hemocyte types may result from inadequate fixa- tion. Second, to rectify this, improved fixation revealing fine structural features of hemocytes was then investigated. Third, structure was viewed together with phagocytic activity as one assay for hemocyte function. MATERIALS AND METHODS Tunicates Tunicate, Halocynthia roretzi, a Urochordate was collected in the Mutsu Bay, Aomori prefec- ture, Japan. Harvesting hemocytes Hemocytes were harvested in tubes from live 552 H. ZHANG, T. SAWADA et al. tunicates by cutting the tunic near the point of attachment. Hemolymph was centrifuged at 250 g for 5 min and the pellets were fixed for electron microscopy. Preparation for transmission electron microscopy (TEM) Since any single method of fixation was not always suitable for all cell types (generally one fixative was reasonably good for only a few), we tried several approaches changing composition and concentration of fixatives and buffers. After sever- al trials, we found that a mixture of 3% glutaral- dehyde and 3% paraformaldehyde preserved many cell types reasonably well. In most cases hemocytes were fixed in a mixture of 3% glutaral- dehyde and 3% paraformaldehyde in 0.2M sodium cacodylate, pH 7.4. To examine the effect of pH on preservation of hemocyte structure, fixatives at four different pH, 6.4, 7.4, 8.0 and 8.5 were often used. Hemocytes were post-fixed in 2% osmium tetroxide in cacodylate buffer for 1 hr, and dehydrated in alcohol or acetone. The speci- mens were embedded in Epon 812, ultrathin sec- tions cut with an LKB Ultrotome Nova, stained with uranyl acetate and lead citrate and then examined with a JEM-200CX electron microscope (Japan Electron Optics Ltd.). Phagocytosis Fresh tunicate hemolymph was mixed with SRBC or RRBC immediately after harvesting and incubated at room temperature for 20 min. Hemo- cytes were thent= collected3 fixed’ Gintt93% glutaraldehyde/ paraformaldehyde for 2 hr, post- fixed in 2% osmium tetroxide and embedded in Epon. In other experiments 0.35-3.5 ml of 10% RRBC, which were first fixed with 2% glutaral- dehyde and paraformaldehyde, then injected into the coelomic cavity via papillae and hemocytes prepared for TEM by the usual method. Young tunicate (stage of organogenesis) Specimens of young tunicates 8-10 days after hatching were supplied by Drs. Yasuo Sugino and Yu Ishikawa. These were also examined as de- scribed above to ascertain the ontogeny of hemo- cytes. RESULTS CHARACTERIZATION CYVGRE TYE General features Many different hemocyte types were observed in tunicate hemolymph. By TEM we classfied them into nine groups: phagocytes (or macrophages; PH cells), granulocytes with small granules (GS cells), granulocytes with large granules (GL cells), vesi- cle-containing cells (VC cells), fibrous material- containing cells (FM cells), vacuolated cells type 1 (VAI cells), vacuolated cells type 2 (VA2 cells), basophilic cells (BA cells) and lymphoid cells (LY cells). This classification was based on observa- tions of hemocytes fixed in a mixute of 3% glutaraldehyde/paraformaldehyde in 0.2M sodium cacodylate, pH 7.4. Correlation of TEM with LM classification together with TEM studies in different species is summarized in Table 1. Phagocytes or macrophages Phagocytes had numerous cell shapes and often pseudopodia. We found lysosomal granules and many vesicles with contents of variable electron: density in their cytoplasm (Figs. 1-3). Some phagocytes had a tube-like anastomosing structure which contained high electron-dense substances (Fig. 1). Phagosome-like large vesicles containing amorphous substances or myelinated figures were - often observed. Active phagocytic activity was observed when hemolymph was incubated with SRBC or RRBC. Their phagocytic activity was also demonstrated by experiments using an in vivo system when fixed RRBC were injected into the coelomic cavity (Fig.3). Phagocytized RRBC were found in phagocytic vacuoles and electron dense tubular or globular structures were often encountered around the engulfed RRBC (Fig. 3). Granulocytes Granulocytes with small granules (GS cells) were generally spherical and contained many gran- ules which varied in electron density and size (Fig. 4). The granules were usually smaller than 0.5 um. GS cells had several mitochondria, rough- endoplasmic reticulum (RER) with long slender cisternae and numerous small vesicles. Golgi complex was also observed. Granulocytes with OF EACH HEMO- . TABLE l. with other reports Cell types in this report Major H. roretzi characteristics by TEM Phagocytes lysosome fe Pee pseudopodia PH) tubular structure phagocytic Granulocytes small granules with (0.5 um) small granules (GS) Granulocytes with large granules GL) Vesicle- containin cells (VC Fibrous material- containing cells (FM) Vacuolated cells type 1 (VA1) large granules (1-2 um) vesicles (0.5 ~m) SER/Golgi complex fibrous material in vesicles RER/Golgi complex Vacuolated cells type 2 (VA2) Basophilic cells (BA) one large vacuole RER several vacuoles central nucleus vesicular RER Golgi complex Fine Structure of Tunicate Hemocytes Sawada et. al. Milanesi and Overton [21] 553 Characteristics of nine types of hemocytes in the tunicates, Halocynthia roretzi and correlation Lymphoid cells poor organella (LY) (ly) large granules (GL cells) contained numerous large granules whose diameter was approximately 1-2 um (Fig. 5), but the volume which they occu- pied was not as large as it was in vacuolated cells. Granulocytes also contained small vesicles, slender RER and their nuclei were located near the cyto- plasm’s center. Vesicle-containing cells Vesicle-containing cells (VC cells) possessed numerous spherical vesicles (about 0.5 um in dia- meter) filled with material of low electron density (Figs. 6, 7). Many of the spherical vesicles seemed to lose their contents during fixation and dehydra- [24] Fuke [32] Burishel [231 “Perophora Wright (20) H. roretzi “TEM hl ee viridis Review LM schlossori TEM TEM Phagocytic cells Fine granular Macrophages Phagocytes Hyaline type 1 (pl) amoeboid cells leucocytes? and type 2 (p2) Granular cells Minute Microgranular type 1 (gl) granular amoebocytes amoeboid cells Granular Granular amebocytes leucocytes ? Macrogranular amoebocytes Granular cells type 3 (g3)? Granular cells type 2 (g2)? Vacuolated cells Signet ring type 1 (vl) cells and type 3 (v3)? Vacuolated Vacuolated cells OSH ai GELS Morula cells? Compartment cells type 2 (v2) cells and/or type 4 (v4)? Granular cells type 2 (g2)? Lymphoid cells Lympho- cytes tion. The cytoplasm had long, flexible and rod shaped mitochondria, abundant smooth-surfaced endoplasmic reticulum (SER) but a small amount of RER. The Golgi complex was well developed. Immature VC cells (figure not shown) had numer- ous globular RER and a prominent Golgi complex as well as the specific vesicles. Fibrous material-containing cells Fibrous material-containing cells (FM cells) were Oval or spherical. The nucleus of high electron-density occupied a central position and it often appeared to be compressed by tightly packed cytoplasmic vesicles. These tightly-packed vesicles 554 H. ZHANG, T. SAWADA et al. Fics. 1, 2. Phagocytes (or macrophages; PH cells) with small vesicles (arrow heads) and granules (thick arrows). Note electron dense anastomosing canalicular structure (thin arrows) in a phagocyte of Fig. 1. n, nucleus. scale bar, 1 um. Fic. 3. A phagocyte engulfed three RRBC (R) in its phagosomes. Three hours after intracoelomic injection of RRBC. Arrow heads, phagosome membrane; Arrows, electron dense substances; n, nucleus. scale bar, 1 um. Fine Structure of Tunicate Hemocytes 555 ey Fics. 4, 5. granules; n, nucleus. scale bar, 1 “m. filled with fibrous material were the most promi- nent feature (Figs. 8, 9, 10). Rather immature cell types contained large amounts of vesicular or spherical RER (Fig. 8). High electron-density of their cytoplasmic matrix was a characteristic fea- ture of mature cells (Figs. 9, 10). Vesicles varied in size, shape and were often fused together. Certain vesicular structures in the cytoplasmic periphery appeared to be open. A prominent Golgi complex with dense contents, electron dense bodies and small vesicles were also observed in the cytoplasm (Figs. 9, 10). Vacuolated cells Vacuolated cells type 1 (VA1 cells) had an eccentric nucleus and a large vacuole which usually contained electron dense material (Fig. 11), while the vacuolar content often appeared to be lost during preparation (Fig. 12). Small vesicles with or without electron-dense substances were observed. RER was well developed in some of these cell type (Fig. 12). Vacuolated cells type 2 (VA2 cells) contained variable numbers of large vacuoles which occupied most of the cell’s volume (Fig. 13). The content of vacuoles appeared homogeneous in certain cells but heterogeneous in others. The nucleus occupied a central position. x ee Granulocyte with small granules (GS; Fig. 4) and granulocyte with large granules (GL; Fig. 5). g, Basophilic cells Basophilic cells (BA cells) has numerous dis- tended RER (Fig. 14) and a centrally located nucleus. Spherical or ellipsoidal mitochondria and well developed Golgi complex were usually pre- sent. Lymphoid cells Lymphoid cells (LY cells) were small round or oval (figure not shown). We observed only a small quantity of cytoplasmic organelles without a char- acteristic component. Sparse organelles consisted of a few spherical mitochondria, a few short and slender RER, and clusters of free ribosomes. HEMOCYTES IN YOUNG TUNICATE All adult hemocyte types were also found in the coelom of young tunicates, 8-10 days after hatch- ing. The fibrous material-containing cells (FM cells) seemed to be more abundant in younger tunicates than in adults. Vacuolated cells type 2 (VA 2 cells) were observed in the larval tunic. EFFECT OF FIXATIVE’S pH ON PRESERVA- TION OF HEMOCYTE FINE STRUCTURE Four different pHs, 6.4, 7.4, 8.0 and 8.5 were examined. Generally the preservation of hemo- 556 Vesic le-conta H. ZHANG, T. SAWADA et al. 0.5 wm in d jameter; ve). S)// Fine Structure of Tunicate Hemocytes her electron in rat iculum (re) i complex (gc) ic ret endoplasm Well developed rough- Fibrous-material containing cells (FM) ture FM cell (Fig. 8). Note fibrous-material (arrows) dense hyaloplasm and nucleoplasm. n, nucleus. scale bar, 1 ~m Fics. 8-10. b] , prominent Golg icles . ‘In ves Imma 558 H. ZHANG, T. SAWADA et al. Fine Structure of Tunicate Hemocytes Sy Dd f os omy, & wai a4 athe Re: f “ sy ¥ A yt cS ak ae Sie ee Fics. 11, 12. Vacuolated cells type 1 (VA1) with a single large vacuole (va). Vacuolar content of the cell in Fig. 11 preserved well with high pH fixative. The cell in Fig. 12 contains well developed rough-endoplasmic reticulum (re). scale bar, 1 um. Fic. 13. Vacuolated cell type 2 (VA2) with several large vacuoles (va) and central nucleus (n). scale bar, 1 um. Fic. 14. Basophilic cell (BA) with numerous dilated rough-endoplasmic reticulum (re). scale bar, 1 um. 560 H. ZHANG, T. SAWADA et al. cytes was suboptimal when a lower pH, such as 6.4 was used. Contents of vesicular structures found in phagocytes and the contents of vacuolated cells were well preserved at higher pHs (Fig. 11). DISCUSSION Recently much attention has been devoted to the immune system of tunicates, since they seem to be one of the key animals necessary for under- standing evolution of sophisticated immune recog- nition mechanisms [25]. There are numerous tunicate species, and for several reasons, Halocy- nthia roretzi is particularly an ideal experimental animal for immunological investigations. First, a single individual has large quantities of hemolymph which in turn contains numerous hemocytes. Second, and perhaps most importantly this species can be obtained throughout the year. Third, certain humoral factors, which play a role in immuno-defense mechanisms, have been isolated and characterized [26-30]. Still there is a need to classify and characterize hemocytes of H. roretzi as one basic study essential to reveal the cellular components of the immune system and to mini- mize confusion concerning functional cells. To begin this approach, we investigated hemocytes in a previous study by light microscopy [24] and here we describe their ultrastructure. Correlation with the previously-reported classifica- tion In the previous LM study, we classified hemo- cytes into ten groups: Phagocytes type 1 and 2 (p1- and p2-cells), granular cells type 1, 2, 3 (gl-, g2- and g3-cells), vacuolated cells type 1, 2, 3 and 4 (vl-, v2-, v3- and v4-cells) and lymphoid cells (ly-cells) [24]. Due to limited power of resolution by LM and limited information from thin slices of TEM materials, we are only able to provide a partial correlation. Nevertheless, considering our own results and those of others we propose four or five major cell groups; phagocytes, granulocytes, vesicle- and fibrous material-containing cells, and vacuolated cells (Table 1). Phagocytes or macrophages In our previous LM study [24] we described the presence of two types of phagocytic cells with different characteristics (pl and p2 cells). Two forms of phagocytic cells, one with a tubular structure containing highly electron-dense sub- stances (Fig. 1) and the other without the tubular structure (Fig. 2), were also identified in the pre- sent TEM observation. The biological significance of this ultrastructural difference is obscure and thus it is difficult to present a definite correlation with the results of LM classification [24]. The most important finding is that both cell forms have pseudopodia, lysosomal granules and _ exhibit strong phagocytic activity, characteristics which strengthen their important role in front line, cellu- lar immuno-defense mechanisms. Granulocytes Milanesi and Burighel [23] identified two gran- ulocytes in the tunicate, Botryllus schlossori which agrees with our classification (Table 1). Although the GS cell clearly corresponds to gl-cells (Table 1), a cell viewed in LM which could correspond to the GL cell is not conclusive. Differences between the chemical components and functional analysis of the two types of granules remain to be solved.’ Vesicle- and fibrous material-containing cells We observed two types of unique cells whose structural details have not been adequately de- scribed in the past. One is the 0.5 um vesicle containing cell (VC cells). We assume that VC cells correspond to granular cells type 3 (g3-cells) of our LM classification [24]. The fibrous material- containing cell (FM cell) is another novel cell which we describe for the first time in H. roretzi. Glomerulocytes, which contained intracytoplasmic fibrous material, were described in the hemocoel of a styelid ascidian [31]. Since the cell shape of the glomerulocyte is different from that of the FM cell, disk-like in the glomerulocyte but oval or spherical in the FM cell and distibution patterns of fibrous material in two cells are entirely different, the glomerulocyte seems to be another unique cell type in certain ascidians. Although structural differences between these two cells are empha- sized above, we cannot rule out the possibility that contents of fibrous materials from these two cells are the same nor are they similar in their charac- Fine Structure of Tunicate Hemocytes 561 ters. By LM the FM cell is probably equivalent to granular cell type 2 (g2-cell) because of the size of its vesicles and cytoplasmic basophilia [24]. Vacuolated cells Vacuolated cells were classified into two types, one had a single, large vacuole with an eccentric nucleus (vacuolated cells type 1, VA1 cells) and the other had several vacuoles with central nuclei (vacuolated cells type 2, VA2 cells). VA1 cells are apparently the same type as signet ring cells and VA2 cells are the compartment cells according to Overton [21]. For these two types Wright [20] recommended the term vacuolated cells (Table 1). VAI cells also seem to correspond to the vesicular- cells described by Fuke (1979), which reach with allogeneic cells during contact reactions [17, 32]. As shown in Fig. 12 some of the VA1 cells had well developed RER suggesting active protein synthesis and storage of synthesized protein in the large vacuole. Perhaps the protein contents of the vacuole are excreted during contact reactions. For clarity, preservation of vacuolar contents largely depends on pH of fixatives. When cells were fixed in a solution of higher pH, such as pH 8-8.5, the contents were well preserved reflecting the stabil- ity of their chemical components. Differentiation of hemocytes We have classified the hemocytes of H. roretzi into nine groups and each of them had ultrastruc- turally distinguishable features. However, we are unable to dismiss the possibility that certain cell types may merely represent differentiation stages of the other hemocytes. For example, certain basophilic cells (BA cells) may be considered as immature because of transitional forms which con- tain large amounts of RER together with 0.5 u~m vesicles or vacuoles. Now we have only insufficient evidence to delineate cell differentiation pathways. Thus further experimentation should be conducted to solve this important question. One approach would be to do extensive investigations of hemo- poietic tissues, homolymph, and hemocytes in young tunicates. These may provide information necessary for understanding developmental cell lineages especially if compared with adult stages that have been combined with successful in vivo and in vitro assays [33-35]. ACKNOWLEDGMENTS We thank the staff of Asamushi Marine Biological Station of Tohoku University, especially Dr. R. Kuraishi, for kind support in supplying tunicates. We also thank Drs. Yasuo Sugino and Yu Ishikawa for their supply of young tunicates. REFERENCES 1 Guenot, L. (1891) Etudes sur le sang et les glandes lymphatiques dans la serie animale. Archs. Zool. Exp. Gen., 9: 13-90. 2 Ohue, T. (1936) On the coelomic corpuscles in the body fluid of some invertebrates III. The histology of the blood of some Japanese ascidians. Sci. Rep. Tohoku Univ., 11: 191-206. 3 Geoge, W. C. (1939) A comparative study of the blood of the tunicate. Quart. J. Micr. Sci., 81: 391- 431. 4 Endean, R. (1960) The blood-cells of the ascidian, Phallusia mammilata. Quart. J. Micr. Sci., 101: 177- IST, 5 Andrew, W. (1961) Phase microscope studies of living blood-cells of the tunicates under normal and experimental conditions, with a description of a new type of motile cell appendage. Quart. J. Micr. Sci., 102: 89-105. 6 Andrew, W. (1962) Cells of the blood and coelomic fluids of tunicates and echinoderms. Amer. Zool., 2: 285-297. 7 Freeman, G. (1964) The role of blood cells in the process of asexual reproduction in the tunicate Perophora viridis. J. Exp. Zool., 156: 157-183. 8 Smith, M. J. (1970) The blood cells and tunic of the ascidian Halocynthia aurantium (Pallas). I. Hema- tology, tunic morphology and partition of cells be- tween blood and tunic. Biol. Bull., 138: 354-378. 9 Ermak, T. H. (1975) An autoradiographic demon- stration of blood cell renewal in Styela clava (Urochordata: Ascidiacea). Experientia, 31: 837- 839. 10 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: 91-200. 11 Rowley, A. F. (1982a) Ultrastructural and cytochem- ical studies on the blood cells of the sea squirt, Ciona intestinalis. I. Stem cells and amoebocytes. Cell Tissue Res., 223: 403-414. 12 Rowley, A. F. (1982b) The blood cells of Ciona intestinalis: an electron probe X-ray microanalytical study. J. Mar. Biol. Ass. U.K., 62: 607-620. 13 14 15 16 17 18 19 20 Pa 22 23 24 M5) 26 562 Rowley, A. F. (1983) Preliminary investigations on the possible antimicrobial properties of tunicate blood cell vanadium. J. Exp. Zool., 27: 319-323. Kalk, M. (1963) Intracellular sites of activity in the histogenesis of tunicate vanadocytes. Quart. J. Micr. Sci., 104: 483-493. Smith, M. J. (1970) The blood cells and tunic of the ascidian Halocynthia aurantium (Pallas). 1. Hema- tology, tunic morphology and partition of cells be- tween blood and tunic. Biol. Bull., 138: 354-378. Fuke, M. T. (1979) Studies on the coelomic cells of some Japanese ascidians. Bull. Mar. Biol. Stn. Asa- mushi, Tohoku Univ., 16: 143-159. Fuke, M. T. (1980) “Contact reactions” between xenogeneic or allogeneic coelomic cells of solitary ascidians. Biol. Bull., 158: 304-315. Fujimoto, H. and Watanabe, H. (1976) The charac- terization of granular amoebocytes and their possi- ble roles in the asexual reproduction of the poly- styelid ascidian, Polyzoa vesiculiphora. J. Morphol., 150: 623-638. Burighel, P., Brunetti, R. and Zaniolo, G. (1976) Hibernation of the colonial ascidian Botrylloides leachi (Savigny): histological observations. Boll. Zool., 43: 293-301. Wright, R. K. (1981) Urochordates. In “Inverte- brate Blood Cells 2”. Eds. by N. A. Ratcliffe and A. F. Rowley, Academic Press, London, pp. 565-626. Overton, J. (1966) The fine structure of blood cells in the ascidian Perophora viridis. J. Morph., 119: 305-326. Botte, L. and Scippa, S. (1977) Ultrastructural study of vanadocytes in Asidia malaca. Experientia, 33: 80-81. Milanesi, C. and Burighel, P. (1978) Blood cell ultrastructure of the ascidian Botryllus schlosseri. 1. Hemoblast, granulocytes, macrophage, morula cell and nephrocyte. Acta Zool., 59: 135-147. Sawada, T., Fujikura, Y., S. Tomonaga and T. Fukumoto (1991) Classification and characteriza- tion of ten types of hemocytes in tunicate Halocy- nthia roretzi. Zool. Sci., 8: 939-950. Cooper, E. L. (1985) Comparative immunology. Amer. Zool., 25: 649-664. Yokozawa, H., Harada, K., Igarashi, K., Abe, Y., Zi 28 Dy 30 31 32 318) 34 35 H. ZHANG, T. SAWADA et al. Takahashi, K. and Ishii, S. (1986) Galactose- specific lectin in the hemolymph of solitary ascidian, Halocynthia roretzi. Molecular, binding and func- tional properties. Biochim. Biophys. Acta, 870: 242-247. Harada-Azumi, K., Yokozawa, H. and Ishii, S. (1987) N-acetyl-galactosamine-specific lectin, a novel lectin in the hemolymph of the ascidian Halocynthia roretzi: Isolation, characterization and comparison with galactose-specific lectin. Com. Biochem. Physiol., 88B: 375-381. Azumi, K., Yokozawa, H. and Ishii, S. (1990) Halocyamines: Novel antimicrobial tetrapeptide- like substances isolated from the hemocytes of the solitary ascidian, Halocynthia roretzi. Biochemistry, 29: 159-165. Azumi, K., Yokozawa, H. and Ishii, S. (1991) Lipopolysaccharide induces release of a metallo- protease from hemocytes of the ascidian, Halocy- nthia roretzi. Dev. Comp. Immunol., 15: 1-7. Azumi, K., Ozeki, S., Yokosawa, H. and Ishii, S. (1991) A novel lipopolysaccharide-binding hemag- glutinin isolated from hemocytes of the solitary ascidian, Halocynthia roretzi: it can agglutinate bac- teria. Dev. Comp. Immunol., 15: 9-16. Mukai, H., Hashimoto, K. and Watanabe, H. (1990) Tunic cords, glomerulocytes, and eosinophi- lic bodies in a styelid ascidian, Polyandrocarpa misakiensis. J. Morphol., 206: 197-210. Fuke, M. (1990) Self and nonself recognition in the solitary ascidian, Halocynthia roretzi. In “Defense Molecules”. Eds. by J. J. Marchalonis and C. Reinisch, Alan R. Liss, New York, pp. 107-117. Raftos, D. A., Stillman, D. L. and Cooper, E. L. (1990) Jn vitro culture of tissue from the tunicate Styela clava. In Vitro, 26: 962-970. Raftos, D. A. and Cooper, E. L. (1991) Prolifera- tion of lymphocyte-like cells from the solitary tuni- cate, Styela clava in response to allogeneic stimuli. J. Exp. Zool., 260: 391-400. Kelly, K. L.,- Cooper, E- sand Raftoss3D> A: (1992) In vitro allogeneic cytotoxicity in the solitary urochordate Styela clava. J. Exp. Zool. 262: 202- 208. ZOOLOGICAL SCIENCE 9: 563-568 (1992) © 1992 Zoological Society of Japan Diacyl Choline Phosphoglyceride: The Endogenous Substrate for Energy Metabolism in Sea Urchin Spermatozoa MASATOSHI MITA Department of Biochemistry, Teikyo University School of Medicine, Itabashi-ku, Tokyo 173, Japan ABSTRACT— Endogenous choline phosphoglycerides (CPG) are substrates for energy metabolism in the spermatozoa of the sea urchin, Hemicentrotus pulcherrimus. It has also been reported that alkenylacyl, alkylacyl and diacyl phosphoglycerides are distributed in sea urchin spermatozoa. This study was undertaken to determine whether CPG available for utilization in energy metabolism is a diacyl and/or ether-containing compound. After incubation of spermatozoa in seawater, only the diacyl choline CPG content was found to have decreased significantly, and no changes were detectable in the other phospholipids. Analysis by gas-liquid chromatography indicated that 16:0, 18:0, 20:1, 20:4 and 20:5 at the 1-position and 20:4 and 20:5 at the 2-position of diacyl CPG had decreased during incubation. Phospholipase A, activity also had high substrate specificity for diacyl CPG. Thus it seems likely that sea urchin spermatozoa obtain energy throught the oxidation of diacyl CPG. INTRODUCTION Flagellar movement in sea urchin spermatozoa occurs partly through reactions catalyzed by dyne- in ATPase [1-4]. Thus, energy metabolism for production of ATP is indispensable for swimming. Sea urchin spermatozoa have been shown to obtain energy for movement from oxidation of endogenous phospholipid [5-7]. It has also been reported that sea urchin spermatozoa contain va- rious phospholipids and cholesterol [8, 9]. Triacylglycerol and glycogen are also present in trace amounts [7-9]. Choline, ethanolamine and serine phosphoglycerides (CPG, EPG and SPG) are the predominant components. After incuba- tion of spermatozoa in seawater, the content of endogenous CPG has been shown to be decreased significantly, with no change in the levels of other phospholipids [8, 10]. CPG thus appears to be a substrate for energy metabolism in sea urchin spermatozoa. The preferential hydrolysis of CPG, among the phospholipids available for energy metabolism, is related to the properties of phos- pholipase A>, which in sea urchin spermatozoa has high substrate specificity for CPG [10]. Accepted April 9, 1992 Received February 6, 1992 In addition to diacyl phospholipids, ether- containing derivatives, such as alkenylacyl EPG, plasmalogen, have been shown to be present in sea urchin spermatozoa [11, 12]. It was shown recently that CPG contained alkylacyl (19%) and diacyl (81%) components [13]. However, previous studies [8, 10] on energy metabolism have been done using mixtures of diacyl and ether-containing phosphoglycerides. Thus, whether the CPG avail- able for utilization in energy metabolism is a diacyl and/or ether containing compound is still unclear. For further clarification of energy metabolism us- ing phospholipids, alkenylacyl, alkylacyl and di- acyl phospholipids in spermatozoa of Hemicentro- tus pulcherrimus were analyzed in this study. MATERIALS AND METHODS Materials Spermatozoa of the sea urchin, H. pulcherrimus, were obtained by forced spawning induced by injection of 0.5 M KCI into the coelomic cavity. Semen was always collected freshly as ‘dry sperm’ and kept undiluted on ice. The number of sperma- tozoa was calculated on the basis of protein con- centration, as determined by the method of Lowry et al. [14], using bovine serum albumin as the 564 M. Mita standard. The protein content per 10’ spermato- zoa was 0).50 mg. | Extraction of lipids Dry sperm were diluted 100-fold in artificial seawater (ASW) containing 458mM NaCl, 9.6 mM KCl, 10mM CaCl, 49mM MgSO, and 10 mM Tris-HCl at pH 8.2, and incubated for 1 hr at 20°C. Each sample was centrifuged at 3,000 x g for 5 min at 0°C. Total lipids were extracted from the precipitate by the method of Bligh and Dyer [15]. Individual phospholipids were separated by thin- layer chromatography (TLC) as described pre- viously [8, 9]. Separation of alkenylacyl, alkylacyl and diacyl phospholipids Alkenylacyl, alkylacyl and diacyl choline, etha- nolamine and serine phosphoglycerides were sepa- rated as 1,2-diradyl-3-acetylglycerol derivatives as described previously [16, 17]. The contents of each type of 1,2-diradyl-3-acetylglycerol were estimated by gas-liquid chromatography (GLC) assays of amounts of fatty acyl moieties in the lipid class, using 17:0 methyl ester as the internal standard [16]. Determination of fatty acid composition The fatty acyl residues of diacyl CPG were analyzed as the methyl esters by GLC [8, 9]. To investigate the positional distribution of fatty acids in 1,2-diacyl-3-acetylglycerol, fatty acids at the 1-position were liberated by Rhizopus delemain lipase (Sigma) and the resulting monoglycerides were separated by TLC and transmethylated as described previously [16, 18]. The fatty acid methyl-esters were extracted with n-hexane, fol- lowed by N>-blow evaporation. The residues were dissolved in a small amount of n-hexane and analyzed using a GC-R1A gas-liquid chromato- graph (Shimadzu, Kyoto) equipped with a coiled column packed with 15% EGSS-X. Estimation of phopholipase A> activity Dry sperm were homogenized with 10mM MgCl, 10 mM CaCl, 1 mM dithiothreitol and 50 mM Tris-HCl at pH 7.5. The homogenate was in- cubated with either 4.6 kBq 1-palmitoyl-2-[1-'*C]- arachidonyl-CPG(2.18 GBq/mmol)(Du Pont-New England Nuclear) or 18.5 kBq 1-O-hexadecyl-2-[5, 6,8,11,12,14,15--H(N)]-arachidonyl-CPG (2749.1 GBq/mmol) (Du Pont-New England Nuclear) for 1 hr at 20°C in a total volume of 0.4 ml, followed by extraction of total lipids. Radioactivity in the free fatty acid fraction separated by TLC was measured by liquid scintillation spectrometry. RESULTS CPG, EPG and SPG extracted from sea urchin spermatozoa have been shown to have an ether- containing component in addition to the diacyl 20 — je) NO Oo O — oO i) Oo O Concentration (g/10° sperm) 10 > O iy O Alkenylacyl Alkylacyl Fic. 1. Changes in levels of CPG (a), EPG (b) and SPG (c) in sea urchin spermatozoa following incubation with seawater. Dry sperm were diluted 100-fold in seawater and incubated for 1 hr at 20°C. Before (clear) and after (dotted) incubation, total lipids were extracted from spermatozoa. CPG, EPG and SPG separated by a thin-layer chromatography were used for analysis of alkenylacyl, alkylacyl and diacyl derivatives. Each value is the mean of three sepa- rate experiments. Vertical bars show S.E.M. Energy Metabolism in Sea Urchin Sperm 565 component [13]. An experiment was carried out to confirm this, and to determine whether the con- tents of the alkenyl-acyl, alkylacyl and diacyl com- ponents of CPG, EPG and SPG decreased during incubation in seawater. Before incubation, CPG contained alkylacyl ether (19%) and diacyl (81%) components (Fig. 1). A trace amount of alkenyl- acyl analog was also present. EPG consisted of 47% alkenyl ether, 2% alkyl ether and 51% diacyl compounds. The SPG fraction contained a con- siderable amount of the diacyl compound (91%). The content of the alkyl ether compound was only 9%, and the alkenyl ether analog was present only at a trace level. After incubation of sea urchin spermatozoa for 1 hr at 20°C, the preparation of alkylacyl CPG had increased from 19% to 27%, and diacyl CPG had decreased from 81% to 73%. The level of the CPG mixture in dry sperm (25 yg/ 10° spermatozoa) thus decreased (20 yg/10” sper- matozoa) following incubation. Similarly, the net content of diacyl CPG decreased from 20+1 yg/ 10° spermatozoa to 15+1 4g/10’ spermatozoa (Fig. 1). However, the alkyl ether CPG content was almost constant (5+1 ~g/10’ spermatozoa) during incubation. Previous studies have shown that CPG, com- posed partly of unsaturated fatty acids, is con- sumed preferentially during incubation [8, 9, 19]. The fatty chain moieties at the 1- and 2-positions in diacyl CPG metabolized during incubation were also examined. Fatty acids at the 1-position of diacyl CPG were mostly of the saturated and monoenoic type, such as 16:0 (22%) and 20:1 (19%) (Table 1). 20:4 (13%) was also present at the 1-position. By contrast, significant amounts of 20:4 (47%) and 20:5 (23%) were found primarily among fatty acids at the 2-position. During dilu- tion and incubation in ASW for 1 hr at 20°C, the relative percentages of fatty acids at the 1- and 2-positions of diacyl CPG remained almost con- stant (Table 1). Based on the net content of diacyl CPG shown in Fig. 1, changes in the levels of fatty acid moieties at the 1- and 2-positions in diacyl CPG TABLE 1. Fatty acid composition of diacyl choline phosphoglyceride in sea urchin spermato- zoa before and after incubation in seawater Dry sperm Incubation for 1 hr Fatty acid 1-position 2-position 1-position 2-position 14:0 1.8+0.2 th 2.4+0.4 tr. 5730) 0.7+0.1 tie 0.6+0.1 n.d. 16:0 DD lests OFZ Deda Oil 723) 7zic\), J) Dp Sista 16:1 3).5) a0) di 2eSac\)), it ac il 2,.9)ack)3 18:0 3-08 Zac), Il 4.1+0.1 O-Aae(s it Ihe} ¢ 1 11.6+0.4 3Qac())Qil (il Zae Oil 3022052 ilte} 32 eitseW. tt 1.0+0.1 1.0+0.1 1.0+0.1 SES V.daeQ, Il n.d. O.d 220) il n.d. 18:4 @,3)a20).3) NOLAacO) 6.4+1.0 LO%3 22188 20:1 19.1+0.8 2e2a2|0 11 17.4+1.0 pita Ol A a7 3.8+0.6 n.d. 3), 5802 n.d. 203 2.4+0.4 5. 7ac U6 2.0+0.1 Dest 028 20:4 (n-6) 13.0+0.8 46.9+1.2 132022056 46.8+1.4 20:5 (n-3) Y),9220,6 rigs) a2 (V9) 9.2+0.8 rp faa Wess 22:4 0.8+0.3 10 = 053 1.0+0.1 0.7+0.2 2255 0.4+0.1 tr. OBEEOW 0.2+0.1 DRO 0.20, 1.4+0.4 0.6+0.1 1.4+0.3 Dry sperm were diluted 100-fold in seawater and incubated for 1 hr at 20°C. Each value is the percentage of the total and the mean+S.E.M. of three separate experiments. tr., trace amount (less than 0.1%); n.d., not detectable. 566 M. MITA during incubation for | hr were calculated. 16:0, 18:1, 20:1, 20:4 and 20:5 at the 1-position and 20:4 and 20:5 at the 2-position were decreased (Fig. 2). During incubation for 1 hr, about 7 nmol of fatty acids per 10’ spermatozoa was liberated from the 1- and 2-positions of diacyl CPG, respec- tively. In contrast to CPG, no change could be detected in the compositions of EPG and SPG in each lipid class. Relative fatty acid content in CPG (nmol/10° sperm) QO © = Tea Ke) @ GC). Co). “C) © TS - — —E WN WN WA Fatty chain Fic. 2. Changes in levels of fatty acid moiety at the 1- (a) and 2-positions (b) of diacyl CPG following incubation with seawater. Content of fatty acids in diacyl CPG was calculated from the absolute value and the relative percentage of its fatty acid moiety. Each value is the mean of three separate experi- ments. Vertical bars show S.E.M. Since the hydrolysis of CPG was shown pre- viously to occur via the action of phospholipase A> [8, 10], an examination was conducted to deter- mine whether the phospholipase A> is capable of catalyzing alkylacyl CPG in addition to diacyl CPG. The homogenate from dry sperm was incu- bated with 1-O-hexadecyl-2-[5,6,8,11,12,14,15- °H(N)]-arachidonyl-CPG and 1-palmitoyl-2-[1- 'C]-arachidonyl-CPG for 1 hr, followed by extrac- tion and separation of free fatty acid by TLC. The radioactivities of *H- and '‘C-free fatty acids were calculated after hydrolysis of alkyl ether and diacyl CPG, respectively. The hydrolysis of alkylacyl CPG was only one tenth of that of diacyl CPG (Table 2). Phospholipase A, thus appears to have greater substrate specificity for diacyl CPG. TaBLE 2. Phospholipase A, activity in sea urchin spermatozoa Specific activity Substrate (nmol CPG hydrolyzed /hr per mg protein) 1-O-Alkyl-2-acyl-CPG 0925022 1,2-Diacyl-CPG 8.8+1.4 The homogenate of dry sperm was incubated with 1- O-hexadecyl-2-[5,6,8,11,12,14,15-°H(N)] arachidonyl- CPG or 1-palmitoyl-2-[1-'*C]arachidonyl-CPG for 1 hr at 20°C. Each value is the mean+S.E.M. obtained in four separate experiments. DISCUSSION The results presented above are further evi- dence of our previous proposal [8] that CPG is consumed during incubation to provide energy for flagellar movement in sea urchin spermatozoa. CPG available for utilization in energy metabolism was found here to be composed of diacyl com- pounds (Fig.1). This preferential hydrolysis of diacyl CPG is due to the particular properties of phospholipase A>. It was also found that phospho- lipase A» in sea urchin spermatozoa possesses greater substrate specificity for diacyl CPG (Table 2). This appears to confirm the specific use of diacyl CPG for energy metabolism. In this study, about 5 ug of diacyl CPG was consumed in 10’ spermatozoa following incubation for 1 hr (Fig. 1). The fatty acids of the consumed diacyl CPG increased 16:0, 18:1, 20:1, 20:4 and 20:5 at the 1-position and 20:4 and 20:5 at the 2-position (Fig. 2). This observation is consistent with data obtained using a mixture of diacyl and ether-containing CPG in previous reports [8, 19, 20]. Previous studies also indicated that 'C- labeled diacyl CPG and fatty acid are oxidized to SCO, in a cell-free system of sea urchin spermato- zoa [8, 10]. Thus possibly, the fatty acid obtained by hydrolysis of diacyl CPG is metabolized for ATP production through (-oxidaton. Energy Metabolism in Sea Urchin Sperm 567 In ram spermatozoa, the choline-containing plasmalogen, alkenylacyl CPG, has been shown to be a predominant lipid [21], which is metabolized during aerobic incubation [22]. However, the plasmalogen content of sea urchin spermatozoa remains constant during incubation [11], as con- firmed by the present data. Alkenylacyl CPG in sea urchin spermatozoa has been shown to be present in a trace amount [13]. Although a con- siderable amount of EPG as an alkenylacyl deriva- tive was observed, the levels of alkenylacyl EPG were certainly unchanged after incubation (Fig. 1). Thus, it is unlikely that sea urchin spermatozoa use ether-containing phospholipids as a substrate for energy metabolism. Alkylacyl CPG is well known to be related to the production of a platelet- activating factor (PAF) in leukocytes [23, 24]. Therefore it is of interest that CPG in sea urchin spermatozoa contains an alkylacyl component in addition to the diacyl component [13] (Fig. 1). However, it is unclear whether alkylacyl CPG serves as a PAF precursor in sea urchin sperma- tozoa. ACKNOWLEDGMENTS The author is grateful to Dr. N. Ueta, Teikyo Uni- versity School of Medicine, for encouragement and valu- able advice, and to Dr. S. Nemoto and the staff of the Tateyama Marine Laboratory, Ochanomizu University, for help in collecting the sea urchins. This study was supported in part by Grant-in-Aid (03740396) from the Ministry of Education, Science and Culture of Japan. REFERENCES 1 Gibbons, B. H. and Gibbons, I. R. (1972) Flagellar movement and adenosine triphosphatase activity in sea urchin sperm extracted with triton X-100. J. Cell Biol., 54: 75-97. 2 Christen, R., Schackmann, R. W. and Shapiro, B. M. (1982) Elevation of intracellular pH activates sperm respiration and motility of sperm of the sea urchin Strongylocentrotus purpuratus. J. Biol. Chem., 257: 14881-14890. 3 Christen, R., Schackmann, R. W. and Shapiro, B. M. (1983) Metabolism of sea urchin sperm. Inter- relationships between intracellular pH, ATPase activity, and mitochondrial respiration. J. Biol. Chem., 258: 5392-5399. 4 Evans, J. A. and Gibbons, I. R. (1986) Activation of dynein | adenosine triphosphatase by organic 10 11 12 1g) 14 15 16 17 18 19 solvents and by triton X-100. J. Biol. Chem., 261: 14044-14048. Rothschild, Lord and Cleland, K. W. (1952) The physiology of sea-urchin spermatozoa. The nature and location of the endogenous substrate. J. Exp. Biol., 29: 66-71. Mohri, H. (1957) Endogenous substrates of respira- tion in sea-urchin spermatozoa. J. Fac. Sci. Univ. Tokyo IV, 8: 51-63. Mita, M. and Yasumasu, I. (1983) Metabolism of lipid and carbohydrate in sea urchin spermatozoa. Gamete Res., 7: 133-144. Mita, M. and Ueta, N. (1988) Energy metabolism of sea urchin spermatozoa, with phosphatidylcholine as the preferred substrate. Biochim. Biophys. Acta, 959: 361-369. Mita, M. and Veta, N. (1989) Fatty chain composi- tion of phospholipids in sea urchin spermatozoa. Comp. Biochem. Physiol., 92B: 319-322. Mita, M. and Ueta, N. (1990) Phosphatidylcholine metabolism for energy production in sea urchin spermatozoa. Biochim. Biophys. Acta, 1047: 175- U7. Mohri, H. (1959) Plasmalogen content in sea-urchin gametes. Sci. Pap. Coll. Gen. Educ. Univ. Tokyo, 9: 263-267. Mohri, H. (1961) Column chromatographic separa- tion of phospholipids in sea urchin spermatozoa. Sci. Pap. Coll. Gen. Educ. Univ. Tokyo, 11: 109-118. Mita, M. and Veta, N. (1992) Fatty chains of alkenylacyl, alkylacyl and diacyl phospholipids in sea urchin spermatozoa. Comp. Biochem. Physiol., 102 B: 15-18. Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193: 265-275. Bligh, E. G. and Dyer, W. J. (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol., 37: 911-917. Sugiura, T., Masuzawa, Y. and Waku, K. (1980) Alkenyl and alkyl ether phospholipids in pig mesen- teric lymph node lymphocytes. Lipids, 15: 475-478. Sugiura, T., Nakajima, M., Sekiguchi, N., Nakaza- wa, Y. and Waku, K. (1983) Different fatty chain compositions of alkenylacyl, alkylacyl and diacyl phospholipids in rabbit alveolar macrophages: High amounts of arachidonic acid in ether phospholipids. Lipids, 18: 125-129. Waku, K. and Nakazawa, Y. (1978) Incorporation rates of [1-'*C]glycerol into the molecular species of alkyl ether phospholipids of Ehrlich ascites tumor cells in vivo. Eur. J. Biochem., 88: 489-494. Mita, M., Harumi, T., Suzuki, N. and Ueta, N. (1991) Localization and characterization of phos- phatidylcholine in sea urchin spermatozoa. J. Biochem., 109: 238-242. 20 Zi 22 568 Mita, M., Ueta, N., Harumi, T. and Suzuki, N. (1990) The influence of an egg-associated peptide on energy metabolism in sea-urchin spermatozoa: the peptide stimulates preferential hydrolysis of phosphatidylcholine and oxidation of fatty acid. Biochim. Biophys. Acta, 1035: 175-181. Lovern, J. A., Olley, J., Hartree, E. F. and Mann, T. (1957) The lipids of ram spermatozoa. Biochem. J., 67: 630-643. Hartree, E. F. and Mann, T. (1959) Plasmalogen in ram semen, and its role in sperm metabolism. Biochem. J., 71: 423-434. M. MITA 28 24 Demopoulos, C. A., Pinckard, R. N. and Hanahan, D. J. (1979) Platelet-activating factor. Evidence for 1-O-alkyl-2-acetyl-sn-glyceryl-3-phosphorylcho- line as the active component (a new class of lipid chemical mediators). J. Biol. Chem., 254: 9355- 9358. Hanahan, D. J., Demopoulos, C. A., Liehr, J. and Pinckard, R. N. (1980) Identification of platelet activating factor isolated from rabbit basophils as acetyl glyceryl ether phosphorylcholine. J. Biol. Ghent... 255; 5514—55 10; ZOOLOGICAL SCIENCE 9: 569-573 (1992) © 1992 Zoological Society of Japan Failure of Muscle Myosin Heavy-Chain Gene Expression in Quarter Ascidian Embryos Developed from the Secondary Muscle Lineage Cells Kazuuiro W. Makase!, SHIGEKI Fustwara!*, Hiroki NISHIDA> and Noriyuki SATOH! ‘Department of Zoology, Kyoto University, Kyoto 606, *Department of Biology, Faculty of Science, Kochi University, Kochi 780, and *Department of Life Science, Tokyo Institute of Technology, Midori-ku, Yokohama 227, Japan ABSTRACT—Muscle cells of the ascidian tadpole larva originate from two different lineages, the primary (B4.1 line) and secondary (A4.1 and b4.2 lines) lineages. Experiments with 8-cell embryos have indicated that isolated blastomeres of the primary lineage show autonomous muscle development, whereas blastomeres of the secondary lineage rarely develop the differentiation markers (muscle- specific antigens and specific enzyme activity) in isolation. However, there is the possibility that A4.1 and b4.2 quarter embryos might express a muscle-specific gene but the transcripts might not be translated into proteins, thus we would not be able to detect the muscle differentiation. In order to examine the possibility, four blastomere-pairs (a4.2, b4.2, A4.1, and B4.1 pairs), isolated from the 8-cell embryo of Halocynthia roretzi, were allowed to develop into quarter embryos, and the occurrence of transcripts of myosin heavy-chain gene was determined by in situ hybridization of whole-mount specimens. The transcripts were evident only in B4.1 quarter embryos and not in A4.1, b4.2 and a4.2 quarter embryos. Thus, the proportion of A4.1 and b4.2 quarter embryos that develop muscle cells does not increase even when examined at the transcriptional level. INTRODUCTION A tadpole larva of the ascidian H. roretzi con- tains forty-two unicellular, striated muscle cells (twenty-one cells on each of the right and left sides of the tail). The lineage of the muscle cells is well documented [1, 2]. Twenty-eight muscle cells in the anterior and middle part of the tail originate from a pair made up of the right and left B4.1 cells (posterior-vegetal blastomeres) of the 8-cell embryo, while four cells in the posterior region and ten in the caudal tip region are derived from the A4.1 (anterior-vegetal) and b4.2 (posterior- animal) pairs, respectively. A pair of a4.2 cells (anterior-animal) in the 8-cell embryo does not contribute to formation of muscle. Many inves- tigations have been carried out to elucidate the cellular and molecular mechanisms involved in the Accepted February 9, 1992 Received January 4, 1992 specification of muscle cells in ascidian embryos [for recent reviews see 3-8]. The extensive poten- tial for self-differentiation of muscle cells from isolated B4.1 cells has been demonstrated by the assessment of the occurrence of several markers of muscle differentiation [9-14]. Muscle differentia- tion was also evident in partial embryos developed from isolated a4.2+b4.2+A4.1 cells [12, 15]. By contrast, few of A4.1 and b4.2 quarter embryos developed muscle cells [9, 13-16], suggesting a difference in mechanisms for specification between primary and secondary muscle cells. Recently, cDNA probes for an ascidian gene for muscle-specific actin [17, 18] and a similar gene for myosin heavy chain [19, 20] have been prepared. Analysis by Northern blotting and in situ hybridi- zation with the aid of these probes has shown that transcripts of the two genes are detectable at stages as early as the gastrula stage [17-20]. It is highly probable that the genes for muscle-specific actin and myosin heavy chain become transcriptionally 570 K. W. MAKABE, S. FUJIWARA et al. activated at around the time of the initiation of gastrulation. Therefore, it remains a possibility that the differentiation of muscle cells in A4.1 and b4.2 quarter embryos might be evident at the transcriptional level but, for unknown reasons, the transcripts might not be _ translated into polypeptides. Thus, we would not be able to detect the development of muscle by assessing the occurrence of muscle-specific antigens and the histochemically detectable activity of acetyl- cholinesterase. The present study was, therefore, designed to determine whether the quarter asci- dian embryos that originate from isolated A4.1 and b4.2 pairs differentiate as muscle cells at the transcriptional level. Using the technique of in situ hybridization, we examined the occurrence of transcripts of the gene for myosin heavy chain in quarter embryos of H. roretzi. The transcripts were detected only in B4.1 quarter embryos and not in A4.1, b4.2 and a4.2 quarter embryos. MATERIALS AND METHODS Embryos Naturally spawned eggs of Halocynthia roretzi were fertilized with a suspension of sperm from another individual. Fertilized eggs were raised in filtered seawater at about 13°C. At this tempera- ture, they developed to gastrulae about 9 hr after insemination and to early-tailbud embryos after about 15 hr of development; they hatched about 35 hr after fertilization. Isolation of blastomeres and production of quarter embryos Eggs were dechorionated with sharpened tung- sten needles about 20min after fertilization. Naked eggs were cultured to the 8-cell stage in 1.0% agar-coated petri dishes. Only 8-cell embryos with a normal appearance were used for the isolation of blastomeres. The four blastomere- pairs of the 8-cell embryo (a4.2, b4.2, A4.1, and B4.1 pairs) were separated with a glass needle under a dissecting microscope. They were reared to quarter partial embryos. Features, such as the location of polar bodies, the configurations of the blastomeres, and the distribution of pigments were used to assess orientation of the embryos. Isolated blastomere-pairs were cultured separately in 24- well multiwells plates (Falcon) coated with 1% agar. Millipore-filtered (pore size, 0.2 ~m) sea- water containing 50 ug/ml streptomycin sulfate was used for culture of dechorionated eggs and isolated blastomere-pairs. Quarter embryos were cultured until normal embryos hatched and then they were fixed for in situ hybridization. In situ hybridization of whole-mount specimens In situ hybridization with a digoxigenin-dUTP- labeled DNA probe was carried out on whole- mount specimens basically according to the method described by Makabe et al. [20]. The original cDNA probe used for in situ hybridization was a 1.6-kb EcoRI fragment of cDNA that en- codes a part of myosin heavy chain of H. roretzi [19]. Labeling of the 1.6-kb fragment with digox- igenin-dUTP (Boehringer Mannheim, Germany) was carried out by the random primer method according to the protocol from Boehringer. Quarter embryos, as well as middle-tailbud embryos (as controls), were fixed for 30 min in ice-cold ethanol : acetic acid (3:1, v/v). The fixed specimens were washed extensively with PBT (phosphate-buffered saline that contained 0.1% Tween 20) and treated with 10 “g/ml proteinase K in PBT for 30 min at 37°C. After washing with PBT, the specimens were post-fixed with 4% para- — formaldehyde in PBT for 20 min at room tempera- ture, and then they were washed again with PBT. Specimens were then treated with PBT: hybridi- zation buffer (1:1, v/v) for 10 min at room tempera- ture and then with hybridization buffer alone for 10 min at room temperature. After a 1-hr pre- hybridization at 42°C, the specimens were hybridi- zed with the digoxigenin-labeled DNA probe for 18 hr at 42°C. The hybridization buffer consisted of 5XSSC (1XSSC comprises 0.15 M NaCl and 0.015 M Naz citrate), 100 “g/ml sonicated salmon sperm DNA, 50 ug/ml heparin, 50% formamide, and 0.1% Tween 20. After hybridization, the solution was gradually exchanged for PBT. The samples were then incubated for 1 hr with 500 yl of Dig-AP conjugate (polyclonal antibodies raised in sheep against digoxigenin-Fab fragments, conju- gated to alkaline phosphatase; Boehringer Mann- Muscle Gene Expression in Quarter Embryo 571 heim) diluted in PBT (1:2000). After washing with PBT and then with colour-developing buffer (100 mM Tris-HCl, 100 mM NaCl, 50 mM MgCh, pH 9.5), the samples were transferred into 1 ml buffer contained 4.5 ul NBT (nitroblue tetra- zolium salt) and 3.5 ul X-phosphat (5-bromo-4- chloro-3-indol phosphate) solutions. Colour was allowed to develop for about 1 hr, and the reaction was stopped by addition of stop solution (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). Samples were treated with a mixture of benzyl alcohol : benzyl- benzoate (1:2, v/v) [21] so that the embryos became transparent and the reaction products could be easily distinguished. RESULTS AND DISCUSSION The four blastomere-pairs of the 8-cell embryo of H. roretzi were separated and allowed to de- velop into quarter embryos. They were fixed at the time when control, intact embryos hatched, and the occurrence of transcripts of the gene for myosin heavy chain in the quarter embryos was examined by whole-mount hybridization in situ. In | faa 2 each of hybridization experiments, the quarter embryos were processed with control middle- tailbud embryos. After all hybridizations, the control middle-tailbud embryos showed distinct positive signals in the differentiating muscle cells on the right and left sides of the tail (Fig. 1A). Isolated a4.2 and b4.2 cells gave rise to partial embryos that looked like permanent blastulae co- vered with transparent tunic (Fig. 1B, C) while A4.1 and B4.1 quarter embryos consisted of a tail-like cluster of smaller cells to which larger cells were attached (Fig. 1D, E). Results of the experiment can be summarized by reference to Table 1 and Figure 1. In the first series of experiments, all of the twenty-five B4.1 quarter embryos examined showed distinct ex- pression of the gene for myosin heavy chains (Fig. 1E, F). The positive staining was detected only in the larger cells (Fig. 1E, F). The total number of positive cells was around 25, but the exact number could not be determined. By contrast, none of the A4.1 quarter embryos (0/22; Fig. 1D) and none of the b4.2 quarter embryos (0/27; Fig. 1C) showed positive signals for the occurrence of the trans- E F Fic. 1. Detection of transcripts of the gene for muscle-specific myosin heavy chain by in situ hybridization using a digoxigenin-labeled DNA probe and whole-mount preparations. (A) Control early-tailbud embryos showing distinct expression of the gene for the muscle-specific protein in differentiating muscle cells on the right and left sides of the tail (arrows). (B) a4.2 quarter embryo, (C) b4.2 quarter embryos, (D) A4.1 quarter embryo, and (E, F) B4.1 quarter embryos. The presence of transcripts of the gene for the muscle-specific protein is evident only in the B4.1 quarter embryos (arrows). Scale bars represent 50 um in all photographs. 572 K. W. MAKABE, S. FusIwAra et al. TABLE 1. Halocynthia roretzi Occurrence of myosin heavy chain mRNAs in quarter embryos of the ascidian Number of embryos expressing the transcripts Experiment a4.2 b4.2 A4.1 B4.1 I 0/18 0/27 0/22 25/25 II 0/32 0/38 0/47 — 43/43 total 0/50 0/65 0/69 68/68 cripts of the gene for the muscle-specific protein (Table 1). The transcript was undetectable in the a4.2 quarter embryos (0/18; Fig. 1B, Table 1). A year later, we carried out the second series of experiments, and the results were the same as those of the first one. The transcripts for myosin heavy-chain gene were detected only in all of the forty-three B4.1 quarter embryos (Table 1). The gene expression was not observed in A4.1 (0/47), b4.2 (0/38) and a4.2 (0/32) quarter embryos (Table 1). The present results demonstrated that, even when muscle differentiation is assessed by use of a specific cDNA probe, the gene transcripts can be detected only in quarter embryos that originated from the primary-lineage presumptive muscle cells and not in quarter embryos that developed from the secondary-lineage cells. Thus, this study confirms the results of previous experiments in which differentiation of muscle cells in B4.1 quar- ter embryos was assessed by morphology [9], by histochemical detection of the enzymatic activity of acetylcholinesterase [10, 12, 14-16], by ultra- structural observations of the development of myofibrils [11], and by immunocytochemical de- tection of the muscle-specific antigens [13, 14]. Very few A4.1 and b4.2 quarter embryos of H. roretzi develop acetylcholinesterase activity (only about 3%) [16] or showed evidence of the muscle- specific antigen [13]. A recent experiment by Nishida [14] has demonstrated that the secondary- lineage presumptive muscle cells do not show evidence of muscle differentiation even if isolated from 64-cell embryos. The present study has confirmed such a tendency of non-autonomous development of the secondary-lineage presump- tive muscle cells at the transcriptional level. Cellu- lar mechanisms for the determination of the fate of muscle cells in the ascidian embryo, may differ between the primary and secondary lineages. As recently discussed by Davidson [8], the differentia- tion of the primary-lineage presumptive muscle cells takes place autonomously and is controlled by intrinsic factors, while the specification of progeni- tors of the secondary-lineage b4.2 and A4.1 cells occurs conditionally (subject to regulation by ex- trinsic factors). However, in some experiments about 15% of b4.2 quarter embryos of H. roretzi developed the muscle-specific antigen (myosin heavy chain protein) [13]. In addition, in the case of Ascidia ceratodes, Meedel et al. [15] reported that all of the A4.1 quarter embryos develop acetylcholinesterase, although less than 5% of the - animal-half embryos (a4.2+b4.2) showed the en- zyme activity. These positive partial-embryos might express muscle-specific genes. Therefore, the proportion of A4.1 or b4.2 quarter embryos that develop muscle cells is not always the same, depend on species and sometimes on batches of eggs. In summary, quarter ascidian embryos that origi- nated from the secondary-lineage presumptive muscle cells did not express the gene for muscle- specific myosin heavy chain, even though they have the developmental potential to form muscle during normal embryogenesis. ACKNOWLEDGMENTS K.W.M. is supported by a Postdoctoral Fellowship from the Japan Society for the Promotion of Science for Japanese Junior Scientists. This study was supported by Grants-in-Aid from the Ministry of Education, Science and Culture, Japan to K.W.M. (No. 02954052) and to N.S. (Nos. 01480027, 01044077, 02236101). 10 11 Muscle Gene Expression in Quarter Embryo REFERENCES Nishida, H. and Satoh, N. (1985) Cell lineage analysis in ascidian embryos by intracellular injec- tion of a tracer enzyme. II. The 16- and 32-cell stages. Dev. Biol., 110: 440-454. Nishida, H. (1987) Cell lineage analysis in ascidian embryos by intracellular injection of a tracer en- zyme. III. Up to the tissue-restricted stage. Dev. Biol., 121: 526-541. Uzman, J. A. and Jeffery, W. R. (1986) Cyto- plasmic determinants for cell lineage specification in ascidian embryos. Cell Differ., 18: 215-224. Satoh, N. (1987) Towards a molecular understand- ing of differentiation mechanisms in _ ascidian embryos. BioEssays, 7: 51-56. Whittaker, J. R. (1987) Cell lineages and determi- nants of cell fate in development. Amer. Zool., 27: 607-622. Venuti, J. M. and Jeffery, W. R. (1989) Cell lineage and determination of cell fate in ascidian embryos. Int. J. Dev. Biol., 33: 197-212. Satoh, N., Deno, T., Nishida, H., Nishikata, T. and Makabe, K. W. (1990) Cellular and molecular mechanisms of muscle cell differentiation in ascidian embryos. Int. Rev. Cytol., 122: 221-258. Davidson, E. H. (1990) How embryos work: a comparative view of diverse modes of cell fate specification. Development, 108: 365-389. Reverberi, G. and Minganti, A. (1946) Fenomeni di evocazione nello sviluppo dell’uovo di Ascidie. Risultati dell’indagine sperimentale sull’uovo di Ascidiella aspersa e di Ascidia malaca allo stadio di otto blastomeri. Pubbl. Stn. Zool. Napoli, 20: 199- DD: Whittaker, J. R., Ortolani, G. and Farinella- Ferruzza, N. (1977) Autonomy of acetylcholinester- ase differentiation in muscle lineage cells of ascidian embryos. Dev. Biol., 55: 196-200. Crowther, R. J. and Whittaker, J. R. (1983) De- velopmental autonomy of muscle fine structure in muscle lineage cells of ascidian embryos. Dev. Biol., 96: 1-10. 2 13 14 15 16 7/ 18 19 20 21 WS Deno, T., Nishida, H. and Satoh, N. (1984) Autono- mous muscle cell differentiation in partial ascidian embryos according to the newly verified cell lineages. Dev. Biol., 104: 322-328. Nishikata, T., Mita-Miyazawa, I., Deno, T. and Satoh, N. (1987) Muscle cell differentiation in ascidian embryos analysed with a tissue-specific monoclonal antibody. Development, 99: 163-171. Nishida, H. (1990) Determinative mechanisms in secondary muscle lineages of ascidian embryos: de- velopment of muscle-specific features in isolated muscle progenitor cells. Development, 108: 559- 568. Meedel, T. H., Crowther, R. J. and Whittaker, J. R. (1987) Determinative properties of muscle lineages in ascidian embryos. Development, 100: 245-260. Deno, T., Nishida, H. and Satoh, N. (1985) Histo- specific acetylcholinesterase development in quarter ascidian embryos derived from each blastomere pair of the eight-cell stage. Biol. Bull., 168: 239-248. Tomlinson, C. R., Beach, R. L. and Jeffery, W. R. (1987) Differential expression of a muscle actin gene in muscle cell lineages of ascidian embryos. Development, 101: 751-765. Kusakabe, T., Suzuki, J., Saiga, H., Jeffery, W. R., Makabe, K. W. and Satoh, N. (1991) Temporal and spatial expression of a muscle actin gene during embryogenesis of the ascidian Halocynthia roretzi. Dev. Growth & Differ., 34: 227-234. Makabe, K. W. and Satoh, N. (1989) Temporal expression of myosin heavy chain gene during asci- dian embryogenesis. Dev. Growth & Differ., 31: 71-77. Makabe, K. W., Fujiwara, S., Saiga, H. and Satoh, N. (1990) Specific expression of a myosin heavy- chain gene in muscle lineage cells of the ascidian embryo. Roux’s Arch. Dev. Biol., 199: 307-313. Dent, J. A., Polson, A. G. and Klymkowsky, M. W. (1989) A whole-mount immunocytochemical analy- sis of the expression of the intermediate filament protein vimentin in Xenopus. Development, 105: 61-74. aa eee i nary! oe ee | REE oa cele Saket “Ti ia dso ner earner itt mse tease ‘. diiats lean eM, ‘AvinpirtiyatoGhy || FRU: | _ i rawlougtl gL, spthrei A ae tite ve a 2 salir : Be be ages om Bd fe sink F : f . qi ees oy es oar eee te ae Ci ot Ee & ULE thls Stee ee Re eh engl a dh pe ; i ageses +tu She a : Weg Sal, ee aly + "> eu: te Gel) torte Dor iret ryt qmeged es i : 92 sii IER Sh: ey : pale She. aly te) ange Aa Sec eis. ne, fat adhe pC ie ie ete 5) Fest yar eseTaat gui aiarenatect ae ae nuiee BOR | ginatier rash ee Ser: a < bres Tesi ry ere = ite: ii ‘iii i pe ened pat 4 hia AR avon = i 3 — \ e ‘ ss = « ‘ Paes, ‘ 4 ‘ay arn . ; 5 wee ie ee et ee oe ene 0 S)° dapiaie ng ee ae DS 4 hy D : ah : Essay. ai, ey 4 Let Asa ayy onan ave (O5 oe Ne. 4 RA ge = Ps , . r an ~ A i 5 £: J ri { ha PR mT - a \ ~ - - = * = af a 5 Ss uP " > ou rbes - £ ¢ ¥ r ees as , oS j pet 7 va ; ¥ . } } id — aa , | f i rig ESS (9 wey Y 4 O74 4 tes * fi 4 t bt Y a - s 2 3 A , , { y F ¢ - >t 4 : wel 3 iy Y; a ¢. SL tea vee : i} i « : : ’ tinh é N 5% tT: pie ieee hs - ry = 45 ' vf Mh a ea be ( ay : ; a Ayr “ Bi ; 7 fc , = s 4 > : = =f : : H ; ‘ ie t = + \> as ‘ +s y . q ¥ v , mes * ‘ 3 - 5 { 2 , pat s Rc at r 4 a bs wee a y al v =" = -] a E : if \ > > . AS i ne Vaan — a 2 i = » \ i { ‘ ZOOLOGICAL SCIENCE 9: 575-581 (1992) Effect of Delay in Anterior or Posterior Amputation on Regeneration of Short Fragments of Planaria SHINGO KURABUCHI and YoOsHIKAzU KisHIDA! Department of Histology, School of Dentistry, Nippon Dental University, Tokyo 102, and ‘Department of Biology, School of Education, Okayama University, Okayama 700, Japan ABSTRACT— Regeneration of fragments obtained by two transverse amputations, with the second amputation being delayed for various periods of time after the first, was examined in the freshwater planarian Dugesia japonica. When the two amputations were performed simultaneously, all of the fragments isolated from pre- and postpharyngeal regions regenerated as normal worms. However, when the second amputation was performed later than the first, a delay from 12 hours to 3 days occasionally caused the reversal of axial polarity and the regeneration of bipolar heteromorphs. In the first group, which consisted of fragments in which the anterior amputation was performed later than the posterior one, bipolar tails were obtained predominantly and the incidence of such regenerates increased considerably with a delay of 1 to 2 days, with the exception of cases in which bipolar heads were formed on the fragments isolated from the prepharyngeal region. In the second group, which consisted of fragments in which the posterior amputation was performed later than the anterior one, bipolar heads were obtained, and their incidence also increased considerably with a delay of 1 to 2 days betweer amputations. However, when the second amputaiton was delayed for more than four days, no bipolar heteromorphs were obtained in either group. Based on these results, a discussion is presented © 1992 Zoological Society of Japan on the role of the anterior and posterior cut ends of a regenerating fragment in body patterning. INTRODUCTION Every section of the body of a freshwater pla- narian can normally regenerate a head from the anterior cut end and a tail from the posterior cut end and, thus, it develops into a new individual with normal proportion. The mechanism that allows the sections to maintain the information related to the original axial polarity has been investigated and discussed for a long time. According to up-to-date model of biological pat- tern formation of planaria [1], it is suggested that the planarian body and regenerating sections are a system with an organizing center at each end, as in the case of the body of hydra [2]. In other words, the tail and regenerating tail are a second organiz- ing area, in addition to the head and regenerating head. Chandebois [3] showed that a tail part was not remolded by a head piece in contact with it but, Accepted February 9, 1992 Received January 6, 1992 rather, the tail part brought about morphallaxis of the head piece. Our previous transplantation experiments [4, 5, 6] also indicated that a tail has a considerable capacity for determining polarity, as a head does. These experimental results provide strong support for Meinhardt’s suggestion [1]. On the other hand, it was ascertained by several groups of reserchers that determination of whether cut surfaces after amputation will regenerate a head or a tail occurs soon after amputation, when a blastema has not yet formed [7-13]. The ingenious experiment of Sengel [14] showed that the early blastema, separated from the stump and cultured in vitro, can differentiate into a head or a tail according to the original position of the fragments. These findings suggest that new positional in- formation within the body section has been already established prior to the formation of the blastema and, possibly, that the stump tissue plays an impor- tant role to the determination of the blastema. So, such dynamic actions should be occurred not only at the anterior cut end but also at the posterior 576 S. KURABUCHI AND Y. KISHIDA one. Very short fragments of the planarian body tend to become bipolar heteromorphs, namely, bipolar heads, referred to as Janus heads, or bipolar tails, Janus tails, as reported in Dugesia lugubris and Dugesia maculata by Morgan [15, 16], in Dugesia tigrina (maculata) by Child [17] and in Polycelis sapporo by Watanabe [18]. However, if the second amputation that engenders the fragments is delayed for some periods of time after the first amputation, even short fragments that often de- velop into bipolar-head regenerates never produce such heteromorphs [18]. If the tail is an organizing area, as mentioned above, and if it plays some role in determining the blastema, does a similar event occur in the case of the bipolar tail? To examine this question, we monitored the regeneration of short fragments, making anterior and posterior transverse amputations at different times. MATERIALS AND METHODS Specimens of the freshwater planarian Dugesia Japonica were collected from a stream in Kanaza- wa City. Asexual worms, 15 mm in length, were selected and fasted for at least for one week before sectioning. The operations on the worms were performed as follows. After the worms ceased to move as a result of being placed on a piece of wet filter paper on an ice plate, each was cut in a relaxed condition to generate sections of as uni- form length as possible. The first transverse amputation was made through the mid portion of the prepharyngeal or postpharyngeal region to divide the worm into two parts. After various intervals the second amputation was made at a level 0.5mm away from the first cut surface. Consequently, two groups of 0.5-mm-long frag- ments were prepared from the prepharyngeal and postpharyngeal regions; the first series consisted of fragments whose anterior cut surfaces after amputation were generated later than the post- erior ones, referred to as anterior later than post- erior amputation. The second series consisted of fragments whose posterior cut surfaces after amputation were generated later than the anterior ones, referred to as posterior later than anterior amputation. In Dugesia japonica, the organism used in this study, it was confirmed by preliminary experiments that a length of 0.5 mm for the frag- ments was suitable. The survival ratio of the fragments declined markedly if they were made shorter than 0.5mm in length, and longer frag- ments, for example, those of 1.0 mm or 1.5 mm in length, did not give rise to any bipolar hetero- morphs. The fragments, prepared as mentioned above, were kept in separate Petri dishes with aged tap water, maintained at 14+1°C. The water was changed every three days for one month. After the regeneration from the anterior and posterior cut surfaces was completed, the newly regenerated individuals were carefully examined under the binocular microscope and photographed. RESULTS Under the present condition (14+1°C), a blas- tema occurred from the cut surface was observed about the foruth day and the eye in the blastema about the fifth day after amputation. The 0.5 mm long fragments, as showing in Tables, frequently produced bipolar heteromorphs. However, any bipolar heteromorphs were not obtained ‘in the fragments of 1.0mm and 1.5 mm in length made by the same way as those in the present experi- ment, of which particular data were not presented here. Five types of the regenerate, namely, normal, no-head, no-tail, bipolar-head and bipolar-tail re- generates, were obtained. The normal type with a head at the anterior end and a tail at the posterior end was obtained predominantly in each ex- perimental group. The no-tailed and no-headed types (Fig. la, b), in which regeneration of head or tail occurred only at one end of the body piece, were also found in each groups. However, these individuals were not considered to be a type of axial heteromorphs because they developed main- ly as a result of simple fusion of the right and the left margins of the cut surface, preventing the formation of a blastema in one-sided amputation stump. The bipolar-head and bipolar-tail regener- ates (Fig. 1c, d) are axial heteromorphs in the true sense, because the characteristics of one end of each was reversed with respect to the original body Axial Polarity in Planarian Regeneration Hi, Fic. 1. head; d, bipolar tail. x30. polarity. As summarized in Table 1 and 2, the 0.5mm long fragments, which were isolated by two simul- taneous transverse amputations from the pre- and postpharyngeal region of the worms, regenerated as normal worms in all cases examined. No-head and no-tail specimens occasionally occurred in every case of delayed second amputation, but it Four types of heteromorph that developed from isolated short fragments. a, No-head; b, no-tail; c, bipolar appears that there is scarcely any relationship between the appearance of such worms and the interval between the two amputations. The appearance of bipolar heteromorphs, whether with bipolar heads or bipolar tails, was closely related to the interval between the first to the second amputation. TABLE 1. Rogeneration of fragments, of which posterior cut ends were made earlier than the anterior cut ends NOMBCEKGE Type of regenerate Eaation Heme N N Bipol : praguicdbesg =Nommalt sa A Mrz NG in ody MM anle RA Isolated from the prepharyngeal region 0 18 18(100) 0(0) 0(0) 0(0) 0(0) 0.5 D5 22(88) 0(0) 0(0) 3(12) 0(0) 1 32 21(66) 3(9) 5(16) 2(6) 1(3) 2 30 19(63) 5(17) 3(10) 0(0) 3(10) 3) 25 25(100) 0(0) 0(0) 0(0) 0(0) 4 20 15(75) 2(10) 3(15) 0(0) 0(0) 6 58 55(95) 1(2) 2(3) 0(0) 0(0) Isolated from the postpharyngeal region 0 Zl 21(100) 0(0) 0(0) 0(0) 0(0) 0.5 48 40(84) 5(10) 1(2) 0(0) 2(4) 1 45 32(71) 7(16) 0(0) 0(0) 6(13) 2 43 33(77) 2(5) 4(9) 0(0) 4(9) 3 26 20(76) 1(4) 3(12) 0(0) 2(8) 4 25 22(88) 3(12) 0(0) 0(0) 0(0) 6 35 33(86) 2(14) 0(0) 0(0) 0(0) The numbers in brackets are the percentages. reversal of axial polarity. Values indicated in boldface correspond to the cases of 578 S. KURABUCHI AND Y. KISHIDA TABLE 2. Regeneration of fragments, of which posterior cut ends were made earlier than the posterior cut ends : Numiberor Type of regenerate = eee fragments No No Bipolar Bipoalr 5 Ses Nein head tail head tail Isolated from the prepharyngeal region 0 15 15(100) 0(0) 0(0) 0(0) 0(0) 0.5 48 41(86) 3(6) 2(4) 2(4) 0(0) 1 38 19(50) 0(0) 9(24) 10(26) 0(0) 2 46 26(57) 2(4) 9(20) 9(20) 0(0) 3 Sy 18(56) 1(3) 11(34) 2(6) 0(0) 4 73) 19(83) 0(0) 4(17) 0(0) 0(0) 6 33 25(76) 0(0) 8(14) 0(0) 0(0) Isolated from the postpharyngeal region 0 18 18(100) 0(0) 0(0) 0(0) 0(0) 0.5 42 37(88) 1(2) 2(5) 2(5) 0(0) 1 46 40(87) 0(0) 1(2) 5(11) 0(0) 2 49 33(67) 0(0) 10(20) 6(12) 0(0) 3 De 19(83) 0(0) 3(13) 1(4) 0(0) 4 Dig) 21(95) 0(0) 1(5) 0(0) 0(0) 6 46 43(94) 1(2) 2(4) 0(0) 0(0) The numbers in brackets are the percentages. Values indicated in boldface correspond to the cases of reversal of axial polarity. Anterior later than posterior amputation The results are summarized in Table 1. In the fragments isolated from the prepharyngeal region, when the second amputation was delayed for 12 hours (0.5 day), bipolar-head regenerates were found in 3 (12%) out of 25 cases examined. With the second amputation delayed for one day, two (6%) bipolar-head and one (3%) bipolar-tail re- generate out of 32 cases were obtained. Further- more, with the second amputation delayed for two days, bipolar-tail regenerate developed in three (10%) out of 30 cases. However, no bipolar heteromorphs were obtained from the fragments in which the second amputation was delayed for more than three days. In the fragments isolated from the postpharyngeal region, the bipolar-tail regenerates were obtained exclusively as hetero- morphs as a result of the delayed second amputa- tion. When the second amputation was delayed for 12 hours, two regenerates (4%) out of 48 were bipolar-tail. With the second amputation delayed for one day, 6 cases (13%) of bipolar-tail rege- nrates out of 45 were obtained, which was the maximum percentage obtained. With amputations delayed still further, the incidence of these re- generates gradually decreased as the second amputation was delayed for three days. When the second amputation was delayed for more than four days no bipolar-tail regenerates were formed. Posterior later than anterior amputation The results are summarized in Table 2. When the posterior amputation was performed from 12 hours to three days after the first amputation, the bipolar-headed type of regenerate appeared. In the case of regeneration of fragments isolated from the prepharyngeal region, the rate of appearance of bipolar heads, brought about by the delay in the second amputation, was 2 (4%) out of 48 cases with a delay of 12 hours, and then the rate reached a maximum of 26% with a delay of one day, thereafter decreasing gradually to zero in the case of a delay of 4 days or more. In the case of regeneration of the fragments isolated from the postpharyngeal region, the pattern of occurrence Axial Polarity in Planarian Regeneration 579 of the bipolar-headed type of regenerate was almost the same as that with the fragments isolated from the prepharyngeal region, that is, a delay in the second amputation of 12 hours gave 2 (5%) bipolar heads out of 42 cases. Subsequently, the incidence increased as the delay in the second amputations was prolonged, and the incidence reached a maximum of 12% with a delay of two days. Then the appearance of the bipolar heads decreased as the duration of the delay increased, and, when the second amputation was delayed for more than four days, no bipolar heads were obtained. DISCUSSION In regeneration of Dugesia japonica, within less than 24 hours of amputation needs the determina- tion of the blastema to be a head or a tail under condition of about 20°C [10, 12], on which condi- tion a blastema forms approximately at the second day after amputation. However, in the present experiments, the regeneration went on at a slow pace under low temperature. Therefore, the periods requiring for the determination of a blas- tema seem to be delayed. Such periods shall be seriously related to the production of bipolar heteromorphs from the short fragments made with the second amputation being delayed for some intervals after the first, as follows. The 0.5mm long body pieces that were long enough to be expected to maintain the original antero-posterior polarity when isolated as a result of two simultaneous amputations. It is noteworthy that the bipolar heteromorphs were developed when they were prepared with a limited interval of time between the two amputations. Such interval of time ranged from 12 hours to three days, when a blastema was not formed yet from the first cut surface. They appeared at the highest frequency with an interval of one or two days between amputations. Furthermore, the characteristics of the bipolar form, either bipolar-head or bipolar- tail, could be almost controlled at will by varying the order of the two amputations; the fragments for which anterior amputation was performed later than the posterior one tended to become bipolar- tail regenerates, while the fragments for which posterior amputation was performed later than the anterior one tended to become _ bipolar-head regnerates. The formation of bipolar hetero- morphs is clearly due to the interval between the two amputations. So, what occurred in the body pieces which were prepared with an interval? It is conceivable that an area organized previously for differentiation of either a head or a tail is estab- lished near the cut surface of the stump made by the first amputation prior to formation of a blas- tema there. That is, if the cut surface is situated at the anterior end of a section, a covert area with head-forming potential is established, and if the cut surface is situated at the posterior end of the sections, a covert area for a tail is established. The potential for regeneration of a head or tail in such covert areas increases with the passage of time. If the second amputation is made within the confines of the covert area, regeneration from the new cut surface will conform to preexisting developmental potential and, thus, some of the fragments isolated should show the mirror-image regeneration. In the case of short fragments made by two simultaneous amputations, covert areas with potential for head and tail coexist in the respective areas near the anterior and posterior cut surfaces at the same time. The effective range of the covert area seem not to extend far away. It is because almost of all the 1.0mm and 1.5 mm long fragments normally regenerated, even though they were made in the same way as the 0.5 mm long fragments, as shown in our preliminary experiments. Several fragments regenerated as the bipolar- headed type, in spite of belonging to the group in which posterior amputation was performed first. Such exceptional regenerates appeared in the case of pieces isolated from the prepharyngeal region, only if the second amputation was done within 24 hours, and not in the case of pieces from the postpharyngeal region. The occurrence of the bipolar head in this experimental group may be due to the fact that the capacity for producing a head is greater in the anterior region and decrease in the posterior direction along the body axis, as pointed out by Kanatani [10]. Accordingly, the covert area for forming a head initiated by the second amputation seems to have priority, because the second amputation was done before th tail- 580 S. KURABUCHI AND Y. KISHIDA forming covert area prepared by the first amputa- tion was not fully established. With an interval of more than four days between two amputations in the present experiments, any bipolar heteromorphs, either bipolar heads or bipolar tails, were not produced. The effect of such a delay in the second amputation clearly varied with the intervals between two amputa- tions. It appears that, after a blastema to be a head or a tail is determined by the covert area, the potential as the covert area seems to disappear gradually, and the regenerating head or tail acts as a organizing center, one of which actions is to inhibit regeneration of identical characteristics in a feedback system, as shown in our previous experi- ments [6]. Child [19] lately found that the fre- quency of bipolar heads decreased with delays in amputation intervals, in which isolation of the fragments is almostly the same as in the present experiment, and he considered that the anterior cut end of the fragment, once determined as the position at which a head regenerates, dominates the posterior regeneration. However, as above mentioned, both cut ends coordinate the body patterning. Conceivably, the posterior cut end, once determined as the position as which a tail regenerates, is a second organizing area, in addi- tion to the anterior cut end, both of which are determined by the covert areas, as suggested by the present experiment. It is probably that, for the establishment of such covert areas with regenerative capacity, some dynamic changes in the cells included therein should occur near the cut surface. Indeed, remark- able increases in rates of synthesis of DNA, RNA and protein, and cell division are found in tissues near the cut surface [11, 20-22]. Although the determinative factors associated with the develop- ment of potential for formation of a head or a tail in the covert areas are known so far, it is clear that they are affected by the exposure of fragments to various inhibitors of protein synthesis and to anti- mitotic drugs [8, 10, 22-24], since treatment with such drugs causes the development of bipolar-head regenerates. Thus, at least one of the factors needed for establishment of the regenerating area is proteinaceous in nature and may be involved in cell division. By contrast, the nervous system is known to play an important role in controlling the normal proportions of the body during regenera- tion; the fragments deprived of ventral nerve cords regenerated to be bipolar heads in high frequency [25-27] and the strips of nerve cord and cordless strips could not regenerate to be normal worms [28-30]. Although specific factors have not yet been found from the nervous system, it is quite possible that the patterning in planarian regenera- tion is closely related to the nervous system. REFERENCES 1 Meinhardt, H. (1982) Models of Biological Pattern Formation. Academic Press, London. 2 Wolpert, L., Hornbruch, A. and Clarke, M. R. B. (1971) Positional information and positional signall- _ ing in hydra. Amer. Zool., 14: 647-663. 3 Chandebois, R. (1984) Intercalary regeneration and level interactions in the fresh-water planarian Duge- sia lugubris 1. The anteroposterior system. Roux’ Arch. Dev. Biol., 193: 149-157. 4 Kurabuchi, S. (1979) The influence of a tail graft on regeneration in the planarian, Dugesia japonica. Zool. Mag., 88: 8-16. 5 Kurabuchi, S. and Kishida, Y. (1990a) Influence of the tail graft on axial polarity in planarian regenera- tion. J. Exp. Zool., 253: 334-339. 6 Kurabuchi, S. and Kishida, Y. (1990b) Comparative study of the influence of head and tail grafts on axial polarity in regeneration of the freshwater planarian. Zool. Sci., 7: 841-847. 7 Ansevin, K. D. (1969) The influence of a head graft on regeneration of the isolated postpharyngeal body section of Dugesia tigrina. J. Exp. Zool., 171: 235- 248. 8 Ansevin, K. D. and Wimberly, M. A. (1969) Mod- ification of regeneration in Dugesia tigrina by acti- nomycin D. J. Exp. Zool., 172: 349-362. 9 Child, C. M. (1914) Studies on the dynamics of morphogenesis and inheritance in experimental re- production. VIII. Dynamic factors in _ head- determination in planaria. J. Exp. Zool., 17: 61-79. 10 Kanatani, H. (1958) Formation of bipolar heads induced by demecolcine in the planarian, Dugesia genocephala. J. Fac. Sci., Univ. Tokyo, IV, 8: 253- 270. 11 Kohl, D. M. and Flickinger, R. A. (1966) The role of DNA synthesis in the determination of axial polarity of regenerating planaria. Biol. Bull., 131: 323-330. 12 Kurabuchi, S. and Kishida, Y. (1979) The role of the nervous system in the planarian regeneration. III. The influence of the head and the nerve cords on 13 14 15) 16 My 18 19 20 AN Maps Axial Polarity in Planarian Regeneration the blastema regeneration. Ann. Zool. Jpn., 52: 179-187. Teshirogi, W. (1963) Transplantation experiments of two short pieces of a freshwater planarian, Bdel- locephala brunnea. Jap. J. Zool., 14: 21-48. Sengel, C. (1960) Culture in vitro de blastémes de régénération de Planaires. J. Embryol. Exp. Mor- phol., 8: 468-476. Morgan, T. H. (1902) The internal influences that determine the relative size of double structures in Planaria lugubris. Biol. Bull., 3: 132-139. Morgan, T. H. (1904) Regeneration of heteromor- phic tails in posterior pieces of Planaria simplicissi- ma. J. Exp. Zool., 1: 385-393. Child, C. M. (1941) Patterns and Problems of Development. University of Chicago Press, Chi- cago. Watanabe, Y. (1948) Physiological studies on a freshwater triclad, Polycelis sapporo. 1. Character and rate of reconstitution of transverse pieces in relation to level of body. J. Exp. Zool., 109: 291- 330. Child, C. M. (1911) Studies on the dynamics of morphogenesis and inheritance in experimental re- production. I. The axial gradient in Planaria doro- tocephala as a limiting factor in regulation. J. Exp. Zool., 10: 265-320. Coward, S. J. and Flickinger, R. A. (1965) Axial patterns of protein and nucleic acid synthesis in intact and regenerating planaria. Growth, 29: 151- 163. Lindh, N. O. (1957) The nucleic acid composition and nucleotide content during regeneration in the flatworm Euplanaria polychroa. Ark. F. Zool., 11: 153-166. Flickinger, R. A. (1959) A gradient of protein 73) 24 25 26 Zi) 28 29 30 581 synthesis in planaria and reversal of axial polarity of regenerates. Growth, 23: 251-271. Flickinger, R. A. and Coward, S. J. (1962) The induction of cephalic differentiation in regenerating Dugesia dorotocephala in the presence of the normal head and in unwounded tails. Dev. Biol., 5: 179- 204. McWhinnie, M. (1955) The effects of colchicine on reconstitutional development in Dugesia_ doro- tocephala. Biol. Bull., 108: 54-65. Kishida, Y. and Kurabuchi, S. (1978) The role of nervous system in the planarian regeneration. I. Regeneration of body fragments deprived of ventral nerve cords. Ann. Zool. Jpn., 51: 90-99. Kurabuchi, S. and Kishida, Y. (1978) The role of the nervous system in the planarian regeneration. II. Regeneration of body fragments with the reversed nerve cords. Sci. Rep. Kanazawa Univ., 23: 57-63. Kurabuchi, S. and Kishida, Y. (1988) The role of nervous system in planarian regeneration. In: Re- generation and Development. S. Inoue, ed. Proc. 6th M. Singer Symp., Okada Printing and Pub- lishing, Maebashi, pp. 99-110. Sperry, P. J., Ansevin, K. D. and Tittel, F. K. (1973) The inductive role of the nerve cord in regeneration of isolated postpharyngeal body sec- tions of Dugesia dorotocephala. J. Exp. Zool., 186: 159-174. Sperry, P. J. and Ansevin, K. D. (1975) Determina- tion in regenerating tissues of Dugesia dorotocephala: the influence of nerve cord grafts. J. Embryol. Exp. Morphol., 33: 85-93. Sugino, H. (1964) On the role of the ventral nerve cords in the regeneration in the common Japanese freshwater planarian. Mem. Osaka Gakugei Univ., 13: 109-115. packs sted: te ag to eb tS eee : 2 te prima ERA ihe aia bicatheidiamcenacniancede if % is | Fie hp apa esa ae Bit a ing Srna, } NOME ass 2 ee ap e mate nena" ior a ae te er oft: (er pa eet te) pie % oudapt r eee” — es ACRE a s Hbea Ria it ae bho! > . eT EA ORO Rh a vate get nA lhe Fe ae she, de sa saambatiathe Absa srneaz veal: “the AIRS Ecyit eryerincn ak Re bas Lee h aii heya, sii ot Ye cit 4h £3 ha ie) mi ops 5 sa ye ii be ea epee 1 Une ® Tithe ak mi ie Bias ss ay be AY - vi aie ie ie a i Ona ‘Bete erg poe; cnet oe ON: ise te ie a i 7s 72h i { Bet jaa} a bs Pts ae Bess ‘ “ é ee ere PPE ye ! thers Ceres : + eT : ; 4 er 3 iS pe > 4 ¥ * 8 pie ‘ in ; - B ; a - - « ~ by ‘ ' =< =: Bn ; | vy a - xf ae ae es rir ot F Das rs + i) 7 . 5 ‘ z Z - =: ~ a 1 a f 2 7 "i i a i} | is . ¥ ry. ag / 3 “h r 4 , h j 4 to ‘ j . A = i f = c= wv a * \ i iy é 1 2 ZOOLOGICAL SCIENCE 9: 583-587 (1992) © 1992 Zoological Society of Japan Mouse Embryo Biopsy: Abnormal Development with Trophoblastic Vesicle Formation W. E. RoupesusH!” and J. G. Kim Department of Obstetrics and Gynecology, James H. Quillen College of Medicine, East Tennessee State University Johnson City, Tennessee 37614, USA ABSTRACT— Preimplantation mouse embryos at the four- and eight-cell stages were subjected to biopsy and evaluated for abnormal (trophoblastic vesicle) development in vitro. A maximum of two blastomeres can be removed from four-cell embryo, whereas four blastomeres can be taken at biopsy from an eight-cell mouse embryo, without significantly increasing trophoblastic vesicle formation. A significant increase in trophoblastic vesicle formation was observed only when the cellular mass per four-or eight-cell embryo was reduced to less than 50%. INTRODUCTION Removal of one or more blastomeres from the preimplantation embryo has been proposed for the early diagnosis of genetic disease, detection of chromosomal abnormalities, and the pretransfer detection of transgenic incorporation [1]. The ability to successfully biopsy embryos is dependent upon the cell-stage, number of cells removed and the biopsy technique [1, 2]. Blastomere biopsy has been accomplished at various preimplatation stages of embryonic development from the two- cell to the blastocyst cell [2-6]. Handyside et al. [3] have reported the establishment of pregnancy fol- lowing the uterine transfer of biopsied-sexed hu- man embryos. A barrier to successful utilization of this technique could be abnormal growth patterns. Abnormal growth patterns include trophoblastic vesicles (no inner cell mass, ICM) or multiple blastocoel cavities. While the abnormal growth patterns may be seen in nonmanipulated embryos [7], they are more common in manipulated Accepted March 7, 1992 Received December 9, 1991 ' To whom reprint requests should be addressed. > Present Address: Division of Reproductive Endocri- nology and Infertility, Department of Obstetrics and Gynecology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, SC 29425-2233 U.S.A. embryos [5]. Single blastomeres removed from four- or eight-cell mouse embryos uniformly de- velop into trophoblastic vesicles [8]. It is impera- tive that the largest possible number of cells be obtained for assay while protecting the develop- mental integrity of the conception. Whether re- moval of a single blastomere from a four-cell embryo is detrimental is controversial. Removal of one blastomere from a four-cell mouse embryo does not significantly reduce in vitro or in vivo development [2, 9]. However, Krzyminska et al. [6] reported that biopsy of four-cell embryos significantly impaired develop- ment in vitro. Biopsied four-cell embryos often fail to undergo subsequent division or compaction and it has been suggested that the biopsy technique may interfere with development [6]. The dissimi- larity between in vitro development rates may be attributed to differences in mouse strains, culture media and/or biopsy technique used [2, 10]. Re- ports on the efficiency of mouse embryo biopsy have suggested that the ideal developmental stage is the eight-cell embryo [2, 6]. The present study was performed to determine the effect of cell-loss (via biopsy) and cell-stage on trophoblastic vesicle (no ICM present) formation in the mouse nondifferentiated embryo. 584 W. E. RouDEBUSH AND J. G. Kim MATERIALS AND METHODS Embryos at the one-cell stage were flushed from the excised oviducts of female B,D.F,/J (The Jackson Laboratory, Bar Harbor MA) mice primed with pregnant mare serum gonadotropin (7 IU, Sigma Chemical Co., St. Louis) followed 48 hours later with human chorionic gonadotropin (7 IU, Sigma). Females were bred with fertile B.D2F,/J males immediately after administration of human chorionic gonadotropin. Embryos were collected, manipulated and cultured in Ham’s F-10 (Sigma) medium supplemented with 0.3% bovine serum albumin (BSA; Sigma) in an atmosphere of 5% CO> in air, 95% relative humidity at 37°C [2]. Mouse embryos were biopsied by the displace- ment technique at the four- and eight-cell stages as previously described [2]. Briefly, embryos were held in place with a fire polished holding pipette and gentle suction while an opening was made in the zona pellucida with a beveled pipette. The beveled pipette was withdrawn and then reinserted throught a second site, a gentle flow (1-2 psi) of medium injected through the pipette was used to dislodge and displace the blastomeres; subsequent- ly pushing them out of the zona pellucida through the first puncture site. Biopsied embryos were transferred to microdrops (25 yl) of Ham’s F-10 medium, under filtered (0.22 ~m; Corning, Corn- ing N.Y.) sterilized light mineral oil (Fisher; Fair 7o 25 20 Lawn, N.J.) in 35 mm tissue culture dishes (Corn- ing). Embryonic development was evaluated after 72 hours in culture. Formation of trophoblastic vesicles, no ICM within a single cellular wall were judged abnormal. Data were analyzed by the Chi square two-by- two contingency table. The care and use of the animals were approved by the Animal Care Committee of East Tennessee State University. RESULTS A total of 527 nondifferentiated mouse embryos were collected from 30 mice; 339 were subjected to biopsy and 188 served as controls. There were no significant differences in trophoblastic vesicle formation at the four-cell stage between controls (0%) and removal of one or two blastomeres (1% and 7%, respectively) by biopsy (Fig. 1). Biopsy of three (21%) blastomeres from the four-cell mouse embryo resulted in a statistically significant increase in formation of trophoblastic vesicles (P< 0.001; Fig. 1). There were no significant differences in tropho- - blastic vesicle formation between controls (0%; no blastomeres biopsied) and one, two, three, or four blastomeres (0%, 1%, 2%, and 9%, respectively) biopsied from eight-cell mouse embryos (Fig. 2). The biopsy of five (75%) blastomeres from eight- blastomeres/embryo a:significant difference (P<0.001) Fic. 1. Trophoblastic vesicle formation in biopsied 4-cell mouse embryos. Abnormal Development of Biopsied Embryos 585 5/8 4/8 al blastomeres/embryo a: significant difference (P<0.001) Fic. 2. cell mouse embryos was found to significantly increase trophoblastic vesicles formation when compared with controls (P<0.001; Fig. 2). DISCUSSION The present study confirms our earlier report that mouse embryos can be biopsied and have 50% of the cells removed without a significant increase in the formation of abnormal blastocysts [2]. We have previously reported that a maximum of one (25% of embryonic mass) or three (37.5% of embryonic mass) blastomeres can be biopsied from four- or eight-cell mouse embryos without signif- icantly affecting normal development in vitro or in vivo [2]. Kelly [11] reported that blastomeres from four- to early eight-cell mouse embryos maintain totipotentiality. However, cells from compacted eight-cell mouse embryos will develop into trophoblastic vesicles [8]. This suggests a loss of cellular totipotency in the late eight-cell mouse embryo. It may be that blastomeres from early embryos require a minimal cellular mass (e.g. 50%) to maintain developmental totipotentiality. Fetal and term development has been estab- lished in a number of species, including humans, following biopsy and oviductal or uterine transfer ae ao lls pa lowever manipulated embryos do not develop in vivo as well as non- manipulated embryos [13]. Early _ post- Trophoblastic vesicle formation in biopsied 8-cell mouse embryos. implantation half embryos have significantly more trophoectodermal cells than ICM [15]. The poor post-transfer development of biopsied pre- embryos can not be solely due to abnormal de- velopment (trophoblastic vesicle formation) of the developing embryos. The poor post-transfer development success of biopsied preembryos may be attributed to bacte- rial, viral, or immune attack through the violated zona pellucida. Nichols and Gardner [14] reported that damage to the zona pellucida will impair embryo development in vivo. They suggested that the intact zona pellucida protects the developing embryo from damage by oviductal compression. However, Cohen [16] reported that improved im- plantation and clinical pregnancy rates can be achieved following the uterine transfer of partial zona dissection (PZD) of in vitro fertilized human embryos. Further studies are required to under- stand why biopsied embryos do not develop as well as non-manipulated or PZD-manipulated embryos following transfer. Cell differentiation is first grossly apparent at the blastocyst stage when two distinct cell types are found, the ICM and trophectoderm. Radially polarized blastomeres in 8-cell embryos is the first identifiable stage of cell differentiation in the mouse [17]. Prior to differentiation the embryonic cells must become determined. The process of cell determination is not fully understood as to how or 586 what influences it to occur. However, several theories exist, these include: (1) epigenetic: depend- ent upon blastomere position at time of determina- tion-differentiation [11, 18-20]; (2) cell apposi- tion: based on the epigenetic hypothesis, where the differentiation signal arises from the sensitivity of the cell’s metabolic activity to the percentage of the cell’s surface that is apposed vs. free to the external environment [21]; (3) cell division asyn- chrony: first division cells are more likely to de- velop as ICM [11]; (4) cytoplasmic determinants [17]; and (5) cellular interactions [22]. Cosby et al. [23] suggested that cell determination occurs at or prior to the eight-cell stage in the developing mouse embryo. The removal of trophoec- todermal-detemined but non-differentiated ceils may result in the reduction of sufficient quantities of cells required for implantation and subsequent placental formation. Krzyminska et al. [6] sug- gested that the biopsy of embryos might result in fewer cells to form the ICM. However, several studies have reported fetal development following the transfer of trophoectodermal biopsied embryos [12, 13]. There may be a critical number of trophoectodermal cells required for sufficient placental development. REFERENCES 1 Roudebush, W. E., and Dukelow, W. R. (1991) Embryonic biopsy by cell displacement maintains an intact blastomere without disrupting development. Zool. Sci. $3 323—328. 2 Roudebush, W. E., Kim, J. G., Minhas, B. S., and Dodson, M. G. (1990) Survival and cell acquisition rates after preimplantation embryo biopsy: Use of two mechanical techniques and two mouse Strains. Am. J. Obstet. Gynecol., 162: 1084-1090. 3 Handyside, A. H., Kontogianni, E. H., Hardy, K., and Winston, R. M. L. (1990) Pregnancies from biopsied human preimplantation embryos sexed by Y-specific DNA amplication. Nature, 344: 768-770. 4 Monk, M., Handyside, A., Hardy, K., and Whit- tingham, D. (1987) Preimplantation diagnosis of deficiency of hypoxanthine phosphoribosyl trans- ferase in a mouse model for Lesch-Nyhan syndrome. Lancet, 2: 423-425. 5 Nijs, M. M., Camus, M., and Van Steirteghem, A. C. (1989) Evaluation of different biopsy methods of blastomeres from 2 cell mouse embryos. Hum. Reprod., 3: 999-1003. 6 10 11 72 13 14 15 16 17 18 19 20 W. E. ROUDEBUSH AND J. G. Kim Krzyminska, U. B., Lutjen, J., and O’Neill, C. O. (1990) Assessement of the viability and pregnancy potential of mouse embryos biopsied at different preimplantation stages of development. Hum. Re- prod., 5: 203-208. Fiser, P. S., and Macpherson, J. W. (1976) De- velopment of embryonic structures from isolated mouse blastomeres. Can. J. Anim. Sie 3055-350. Rossant, J. (1976) Postimplantation development of blastomeres isolated from 4- and 8-cell stage mouse embryos. J. Embryol. exp. Morph., 36: 283-290. Wilton, L. J. and Trounson, A. O. (1989) Biopsy of preimplantation mouse embryos: Development of manipulated embryos and proliferation of single blastomeres in vitro. Biol. Reprod., 40: 145-152. Dandekar, P. V. and Glass, R. H. (1987) Develop- ment of mouse embryos in vitro is affected by strain and culture medium. Gamete. Res., 17: 279-285. Kelly, S. J. (1975) Studies of the potency of the early cleavage blastomeres of the mouse. In “The Early Development of Mammals”. Ed. by M. Balls and A. E. Wild, Cambridge University Press, Lon- don, pp. 97-105. Monk, M., Muggleton-Harris, A. L., Rawings, E. and Whittingham, D. G. (1988) Pre-implantation diagnosis of HPRT-deficient male and carrier female mouse embryos by trophectoderm biopsy. Hum. Reprod., 3: 377-381. Summers, P. M., Campbell, J. M. and Miller, M. W. (1988) Normal in vivo development of marmoset monkey embryos after trophectoderm biopsy. Hum. Repord., 3: 389-393. Nichols, J. and Gardner, R. L. (1989) Effect of damage to the zona pellucida on development of preimplantation embryos in the mouse. Hum. Re- — prod., 4: 180-187. Rands, G. F. (1986) Size regulation in the mouse embryo. II. The development of half embryos. J. Embryol. exp. Morph., 98: 209-217. Cohen, J. (1991) Assisted hatching of human embryos. J. In Vitro Fert. Transfer, 8: 179-190. Johnson, M. H., Chisholm, J. C., Fleming, T. P. and Houliston, E. (1986) A role for cytoplasmic determinants in the development of the mouse early embryo? J. Embryol. exp. Morph., 99: 97-121. 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. Morph., 18: 155-180. Hillman, N., Sherman, M. I. and Graham, C. (1972) The effect of spatial arrangement on cell determina- tion during mouse development. J. Embryol. exp. Morph., 28: 263-278. Fleming, T. P. (1987) A quantitative analysis of cell allocation to trophoectoderm and inner cell mass in the mouse blastocyst. Develop. Biol., 119: 520-531. Abnormal Development of Biopsied Embryos 21 Wiley, C. M. (1984) The cell surface of the mamma- Mp lian embryo during early development. In “The Ultrastructure of Reproduction”. Ed. by Blerkom J. Motta, Martinus Nijhoff, Boston, pp. 190-204. Dyce, J., George, M., Goodall, H. and Fleming, T. P. (1987) Do trophectoderm and inner cell mass Jip) 587 cells in the mouse maintain discrete lineages? De- velopment, 100: 685-698. Cosby, N. C., Roudebush, W. E., Ye Lian and Dukelow, W. R. (1988) Cell differentiation as determined by micromanipulation of mouse embryos. Biol. Reprod., 38 (suppl. 1): 128. es i oe jae weil’ nirks ca hel ¥ hae pois sae ve” toy Me wary ; , Aas 2 ©) eee ee Oe y aj " Sytelphasntie seep ; 2 TEE Sar: tines “i RE BARE: aa 4 on - , ah \ ' F iy i a. “ ? aren . r me See 21 ee el) ae Ser thes ae Wah pe : cal x abe! Borg a 5 f x ce : eae’ | t ir a ie i acini: | ' ZOOLOGICAL SCIENCE 9: 589-599 (1992) Morphology of Filaments on the Chorion of Oocytes | and Eggs in the Medaka TAKASHI IWAMATSU Department of Biology, Aichi University of Education, Kariya 448, Japan ABSTRACT— The morphology of filaments on the chorion of oocytes and eggs of the medaka, Oryzias latipes, was investigated as a step in the clarification of the mechanism for the determination of egg polarity. In mature oocytes, long attaching filaments arising from the vegetal pole area (VPA) of the egg membrane (chorion) were wound spirally on the axis between the animal and the vegetal poles of the egg, and covered the chorionic surface in the vegetal hemisphere. Short non-attaching filaments which were distributed on the chorionic surface in the area other than the VPA were also bent in a unilateral direction on the egg axis. In the previtellogenic stage of oogenesis, both attaching and non-attaching filaments appeared as verruciform structures on the oocyte surface before appearance of the chorion. The VPA at the one end of the animal-vegetal axis was first recognizable morphologically by the appearance of a cluster of verruciform structures which were destined to develop into attaching filaments. As these primitive filaments grew to form horn-shaped structures, their distal parts bent and pointed either to the right or the left around the axis. Six inbreedings of fish selected for right-handed filament pattern failed to demonstrate a hereditory determination of the pattern. In addition, it was ascertained that eggs of several other species of Oryzias also possessed both attaching and non-attaching filaments on the chorionic surface which showed an unilateral spiral pattern on the animal-vegetal axis. The results of this study suggest that in the medaka, egg polarity is established by localization of the © 1992 Zoological Society of Japan VPA, followed by the unilateral bending of non-attaching filaments on the chorion. INTRODUCTION In teleostean fishes, the oocyte surrounded by developing follicular cells differentiates and grows to form a large telolecithal egg. This specialized cell has a thick extracellular envelope (chorion) resembling the cell wall of a plant cell. Young oocytes prior to formation of the chorion are driven to differentiate into eggs as a result of their interaction with the surrounding granulosa cells. The heterogeneous differentiation of the follicular cells and the formation of unequal structures in the chorion determine egg polarity, i.e. the regional differences in intracellular structures and reactivity of the egg itself. The micropylar cell which forms a pore (micropyle) in the chorion differentiates from granulosa cells [1]. The side of the oocyte where the micropylar cell is located becomes the animal pole of the egg. On the other hand, hair-like structures are located on the chorionic surface at Accepted March 10, 1992 Received January 20, 1992 the pole (vegetal pole) opposite to that of the micropylar cell. These structures are termed “attaching filaments” in Oryzias [2-5] and cho- rionic fibrils in Fundulus [6]. The attaching fila- ments differentiate in the previtellogenic stage of oogenesis in Oryzias [7]. The egg axis between the animal and the vegetal poles is easily distinguish- able due to these morphological characteristics of the chorion, as well as to the unequal distribution of ooplasmic inclusions. The Oryzias egg seems to be a suitable model for clarification of the mecha- nism by which egg polarity is determined during oogenesis in the teleost. The main purpose of the present study was to obtain basic information on the dynamic rela- tionship between the oocyte and its surrounding follicular cells, because this information is most likely to clarify the mechanism of determination of egg polarity. It has previously been reported that attaching filaments extend in a spiral pattern from the animal-vegetal axis and cover the chorionic surface in the vegetal hemisphere of the Oryzias ege [8]. In a single batch of eggs, eggs bearing 590 T. IWAMATSU filaments wiht both left- and right-handed spirals can be found. In connection with the spiral pattern of the attaching filaments, the present author is interested in (a) the process by which the egg axis is determined and the spiral pattern of the cho- rionic filaments is formed during oogenesis, (b) whether or not the spiral pattern of attaching filaments is inherited, and (c) whether or not the spiral pattern is common in this genus. The present study was performed to clarify these points. The results indicate that (a) the spiral pattern around the animal-vegetal axis of the oocyte is first recognized in non-attaching fila ments on the oocyte in the early previtellogenic stage of oogenesis, (b) the pattern is probably not inherited, and (c) in all species of Oryzias ex- amined, the attaching filaments are arranged in a left- or right-handed spiral pattern. MATERIALS AND METHODS Most of the medakas used in the present study were Oryzias latipes (Yamato-koriyama, Nara Pref.). In order to compare egg of O. latipes with those of other species of Oryzias, spawned eggs of O. celebensis, O. curvinotus, O. javanicus, O. mekongensis, O. melastigma and O. minutillus, which had been breeding in our laboratory, were also examined to determine the spiral pattern or the distribution of attaching filaments. A light-microscopic survey of the spiral patterns of the attaching filaments and non-attaching fila- ments was conducted on living eggs (oocytes) in each of ovaries using both a binocular dissecting microscope (X20) and an ordinary light micro- scope (100, Olympus). In cases in which it was difficult to determine the spiral pattern of the attaching filaments, the direction in which the tops of the non-attaching filaments bent was used to make a determination. Fertilized eggs with the right-handed spiral pattern were selected and allowed to develop to fry. Newly hatched fry were raised to adults in a glass bowl (30 cm in diameter, ca. 12cm in water-depth, 28°-30°C) with con- tinuous lighting. Both the fry and the adults were fed a powdered diet [9]. They were inbred by sister-brother matings for six generations during a period of two years from 1989 to 1991. For observations by transmission electron micro- scopy, previtellogenic follicles in ovarian tissue were fixed with modified Karnovsky’s fixative for more than 6hr at 4°C. After the samples were rinsed, post-fixed for 60 min in phosphate-buffered (pH 7.2) 1% OsOy, at 0-4°C, and dehydrated in an alcohol and acetone series, they were embedded in Epon 812. Ultrathin sections were stained with uranyl acetate and lead citrate and examined with a JOEL JEM-100C electron microscope. The data were statistically analysed by the Stu- dent’s t-test. RESULTS The spiral pattern of attaching and non-attaching filaments in various sized oocytes during oogenesis Primitive attaching and non-attaching filaments were first recognized as verruciform structures in the space between the surface of transparent small oocytes (about 120 ~m in diameter: Stage IV in 7) and the surrounding follicular cells just after the formation of very thin rudiments of chorion (Fig. 1). Most of the verruciform structures were distri- ° buted at intervals of about 12.5 ~m on the ‘cho- rionic surface in St. IV oocytes [7]. The primitive attaching filaments, which were distributed at in- tervals of about 8 “m, were clustered in the vegetal pole area (VPA). The non-attaching filaments first became horn-shaped (Fig. 1), then candle-like (Fig. 2), which caused them to bend in the same direction in St. IV oocytes (130-140 ~m in dia- meter). They bent at right angles to the oocyte axis, one end of which was determined by the VPA with its verruciform structures (primitive attaching filaments). Attaching filaments situated at the VPA of the oocyte became unilaterally curved candle-like structures in St. [IV oocytes 140-150 ym in diameter (Fig. 2, Table 1). When the oocyte diameter reached about 350 «zm (St. VI), the mean distance between adjacent attaching filaments was about 14 um (the diameter of the VPA, ca. 160 ym) which represented an increase of only 5 um per 100 um increase in oocyte diameter. In con- trast, the interval between non-attaching filaments was about 23 um and increased about 8 um per 100 em increase in oocyte diameter (Fig. 3). As the Filaments on the Chorion in Medaka Eggs 591 Ko Pe, BA Toe ae Fic. 1. Early previtellogenic oocytes showing verruciform or horn-shaped non-attaching filaments on the oocyte surface in Oryzias latipes. A: Electron micrograph of a part of an oocyte (a) with rudiments non-attaching filaments present as verruciform structures (asterisks). 2,100. B: Electron micrograph of a part of an oocyte (b) with horn-shaped nonattaching filaments (asterisks). 2,700 C: Micrograph of intact intrafollicular previtellogenic oocytes showing verruciform filaments as dots (asterisk in oocyte a) and horn- or stickle-shaped non-attaching filaments (asterisk in oocyte b). These oocytes correspond to those in Figs. A and B, respectively. Arrows indicate fibrous structures connecting with the other ovarian tissues. G, granulosa cell layer; N, oocyte nucleus; T, thecal cell layer. 415. vitellogenic oocytes grew rapidly, the surface of the chorion expanded accordingly except for the VPA which expanded only slightly and occupied an area 410-450 4m in diameter in mature oocytes. Attaching and non-attaching filaments arose from the chorion and elongated among the granu- losa cells within the basement membrane (Fig. 4A). These filaments were composed of bundles of tubular structures (18-19 nm outside diameter), as reported by Hart and Donovan [10]. They arose from and morphologically resembled the outer layer of the chorion, which also consisted of elec- tron-dense and microtubular structures (16-18 nm outer diameter) (Fig.4B). The numbers of attaching and non-attaching filaments on an oocyte of O. latipes (25-35 and 190, respectively) did not change regardless of the size of the growing oocytes. The attaching filaments clustered at the vegetal pole of the oocyte axis elongated in a unilateral direction around the oocyte axis. Long attaching filaments, which began to form in pre- vitellogenic oocytes eventually formed a special cap over the vegetal hemisphere as they elongated progressively in the vitellogenic oocyte (Figs. 2 and 5). As shown in Table1, both left- and right-handed spiral patterns of attaching filaments (Fig. 6) were present. Inheritance of the spiral pattern of attaching fila- ments of O. latipes eggs Fifteen fertilized eggs (Fo) showing the right- handed spiral pattern of attaching filaments were selected from a batch of eggs spawned by a single pair (sister-brother mating) of orange-red type O. latipes and developed to adulthood. Three pairs 592 T. IWAMATSU Mi A Ips I] ff ai aN Fic. 2. A diagram of the changes in attaching and non-attaching filaments during oocyte growth in Oryzias latipes. This diagram displays the successive changes in non-attaching (Non-AF) and attaching (AF) filaments (the left-handed spiral type) of growing oocytes. The filaments bend in a unilateral direction around the animal (AP)-vegetal (VP) axis (oblique bars). Attaching filaments (AF) are spun over the chorion in the vegetal hemisphere of a large oocyte (400 ~m). A presumptive direction (the right-handed) of the rotation of oocytes is indicated by an arrow. TABLE 1. Differentiation of filaments on the chorion and the left- or right-handed spiral pattern of the filaments in Oryzias latipes eggs Sao ret glace Speaoattenneiia eine MA ee oocytes oocytes filaments (um) (Females) Right Left (%)’ Vv H B <110 56 — — ND ND ND — 110-119 3 — — ND 100 0 0 120-139 44 — — ND 54 39 7 140-149 41 36 64 UY 0 17 83 150-199 116 45 35 100 0 0 100 200-299 121 46 54 100 0 0 100 300-399 51 63 3m 100 0 0 100 400-499 33 70 30 100 0 0 100 500-600 16 44 56 100 0 0 100 > 600 65 58 42 100 0 0 100 * Percentages of the number of oocytes showing attaching filaments or non-attaching filaments among the total number of oocytes in each size. Abbreviations for non-attaching filaments: B, unilaterally bended; H, short horn-shaped; V, verruciform. Filaments on the Chorion in Medaka Eggs 593 Distance (ym) between filaments Fic. 3. 7 8 9 10 11 12 Oocyte diameter (x10? um) Change in the distance between filaments during oocyte growth. The change in the distance between adjacent filaments on the chorion in proportion to the increase in the oocyte diameter (um) is diagramed at the upper left. The mean distances between filaments in the VPA (open squares) and in the remainer of the chorion surface (AE: the animal pole and the equatorial area, closed circles) was spotted against oocyte diameter. selected at random from among the sexually ma- ture fish were mated yielding 218 fertilized eggs in 16 ovipositions. Among these spawned eggs (F1), about 64% (a right-handed to left-handed type (R/ L) ratio, 1.6) showed the right-handed spiral pat- tern of attaching filaments. Six pairs that de- veloped from these right-handed type eggs were selected again for mating (each pair was mated in a separate aquarium). A total of 623 eggs was obtained in 41 ovipositions. The total R/L ratio among these F, eggs was 1.1, although the number of right-handed type eggs among these eggs of two pairs (FR-F,_,, FR-F2_>) was significantly higher. The F3; (FR-F3_ 15) eggs spawned by the FR-F)_ pair in Table 2 were reared and again pair-mated. The right-handed spiral pattern among the Fy, (FR-F4_19) eggs (1625) spawned by these pairs (FR-F3_4) was not significantly more prevelant than the left-handed pattern. Two pairs (FR- F;_ 43) of the off-spring of FR-F4,_5; spawned 604 eggs (F.), which showd a R/L ratio of 1.0. In the right-handed type pedigrees, the R/L ratio among 5,002 eggs in six generations was about 1.1. The spiral pattern of attaching filaments in several species of Oryzias Table 3 lists the R/L ratios of the spiral pattern of attaching filaments on the chorions of O. celebensis, O. curvinotus, O. javanicus, O. mekongensis, O. melastigma and O. minutilus eggs. The right-handed pattern tended to occur more frequently than the left-handed pattern, but the difference was not statistically significant. DISCUSSION Both the left- and the right-handed spiral pat- terns are found in the attaching filaments on the chorion of the medaka egg. A similar observation on the spiral pattern of filamentous structures on the chorion was reported early by Kurakami [11] for the pacific saury Cololabais saira. In a pre- 594 T. IWAMATSU Fic. 4. Micrographs of previtellogenic oocytes showing non-attaching filaments among granulosa cells. A: An oocyte at the late previtellogenic phase of oogenesis exhibits non-attacing filaments (asterisks) among the granulosa cells (G) between the chorion (E) and the basement membrane (B). T, thecal cell. 13,000. B: An oocyte at an early vitellogenic phase exhibits non-attaching filaments (asterisks) originating from the chorion, which consists of an outer layer (E) and an inner layer (L). G, granulosa cells. 9,700. Inset: the outer layer of the chorion which is composed of electron dense amorphous and tubular structures. 92,000. Filaments on the Chorion in Medaka Eggs 595 Fic. 5. Microphotographs of vitellogenic oocytes of Oryzias latipes showing attaching filaments on the chorion in the vegetal hemisphere. These oocytes at the early stage (A: the right-handed spiral type) and at the late stage (B: the left-handed spiral type) of vitellogenesis have different spiral patterns. Non-attaching filaments in the animal hemisphere bend and the thread-like attaching filaments elongating from the VPA are wound on the vegetal hemisphere of the oocyte in a unilateral direction. The proximal part of the attaching filaments bends in the direction of the animal pole of the oocyte. In Fig. 5B, early previtellogenic oocytes (ca. 115 ~m in diameter, arrow) are present showing verruciform primitive filaments. 210. liminary note [8], It has been reported that eggs showing the right-handed spiral of attaching fila- ments might be counted more frequently than those showing the left-handed spiral. This tenden- cy was also recognized in eggs of other Oryzias species in the present study, although the higher frequency was not statistically significant, and eggs spawned by only a single female of each species were studied. In addition, we tried to select genetically for the eggs showing the right-handed spiral pattern of attaching filaments. The results seem to indicate that the spiral pattern is not hereditory. That is, the spiral pattern may be randomly determined during oogenesis. The tendency for the right- handed type to somewhat predominant in frequen- cy is possibly caused by some unknown factor(s) such as the revolution of the earth. The spiral of attaching filaments and the un- ilateral bending of non-attaching filaments on the chorion are both at right angles to the oocyte axis and appear to be involved in the formation of egg polarity. All filaments must elongate in haphazard and random directions among the granulosa cells if no force is exerted on them. However, since all elongating filaments arise from the chorion and elongate among the granulosa cells, they should bend in the same direction if the oocyte itself rotates within the non-motile follicular consti- tuents connected to the other tissue of ovary, or if all the granulosa cells translocate in a unilateral direction in response to unknown stimulation. Granulosa cells directly contact to the oocyte surface by inserting their cytoplasmic extensions into the chorion, and the cytoplasmic extensions of the oocyte also closely attach to the granulosa cells through numerous pores in the chorion until ovula- tion [12]. For this reason, it is unlikely that either translocation of the granulosa cells or the rotation of the oocyte within the non-motile granulosa cell 596 T. IWAMATSU Fic. 6. Diagrams illustrating the patterns of attaching and non-attaching filaments on the chorion of mature Oryzias eggs. a: A view of the animal pole side (Arrow indicates a micropyle: non-attaching filaments, the right-handed pattern). b: A lateral view from the equatorial side. c and c’: In these views of the vegetal pole side, attaching filaments (A) exhibit either the left-handed (c) or the right-handed (c’) spiral pattern. NA, non-attaching filaments. layer can occur because both would have to be accompanied by mutual separation of the oocyte and granulosa cells before ovulation. It is plausible that the oocyte intimately connected with its sur- rouding granulosa cells can translocate within the basement membrane. In young previtellogenic oocytes, the granulosa cell layer is extremely thin, so the tops of the elongating filaments are com- pressed by the basement membrane and the thecal layer. Consequently, unilateral rotation of the oocyte and the surrounding granulosa cell layer within the basement membrane may cause un- ilateral bending at the distal part of the elongating filaments on the chorion. Rotation of the oocyte must result in bending of only the attaching and non-attaching filaments that are directly in contact with the basement membrane, and the bending must be in the opposite direction to that of the rotation. The basal regions of attaching filaments, which are surrounded by tall granulosa cells, and the long basal parts of attaching filaments in the vegetal hemisphere are not affected by the oocyte rotation and are radially straight from the VPA toward the animal pole after slightly bending dur- Filaments on the Chorion in Medaka Eggs 7) TaBLE2. Inheritance of the right-handed spiral pattern of attaching filaments in Oryzias eggs ee eee No. of NOME Spiel pattern of attaching filaments | Ovipositions eggs Right-handed (L) Left-handed (R) ratio FR-Fo May 790 1 DS 13 12 1.0 FR-F, Aug. 790 16 218* 140 78 1.6* FR-F>_; Oct. 790 5 54# 35 19 LG FR-F,_ > 2, 7 95 54 4] 13" FR-F,_ 3 Z, 6 113 57 56 1.0 FR-F,_4 2 16 209 98 111 0.9 FR-F3_ 5 2 4 73 40 33) LZ FR-F>_6 2 3) 79 38 4] 0.9 FR-F3_; Rebae.9il 16 439 UD) 210 1.1 FR-F3_ 5 2 22 471 Dail 220 lea FR-F3_ 3 2 14 364(145) 185(72) S735) EO FR-F3_ 4 2 16 417(117)# 224(66) 193(51) eZ FR-F3_ 5; 2 11 ASD) 121(74) 94(98) 13 FR-F,_; Aug. 791 10 DS) IZ 125 1.1 FR-F,4_ > y 10 12) 91 104 0.9 FR-F,_ 3 2 10 159(103) 74(47) 85(56) 0.9 FR-F,4_4 2; 11 121(87) 63(47) 58(40) loll FR-F,_ 5 2, 18 Das 126 98 3 FR-F,4_ 6 2 16 284 1377 52 0.9 FR-F,_7 2 1) 163 96 67 1.4 FR-F,_ 2 13 139 2 67 1.1 FR-F,4_9 7; 2 84 41 43 1.0 FR-Fs_; Oct. °91 8 244 130 114 ted FR-F;_ 3 7 8 142 66 76 0.9 FR-F;_ 3 2 iit 218 fe: 106 lel P/L ratio: the ratio of the number of eggs with the right-handed spiral pattern of attaching filaments to the nubmer with the left-handed spiral pattern. F3 were obtained from a single pair of FR-F,_3 adults. *Significant in Student’s t-Test. * Mature fish derived from these eggs were used for mating. The number in parentheses indicates the number of oocytes in ovaries. TABLE 3. The spiral pattern of attaching filaments in eggs of several species of Oryzias No. of Natiot Spiral pattern of attaching filaments STSCI Ovipositions eggs Right-handed (R) Left-handed (L) ae O. celebensis 13 340 188 152 2 O. curvinotus 2 82 48 34 1.4 O. javanicus 10 55 35 20 1.8 O. mekongensis 5 93 53 40 ed) O. melastigma A) 480 287 193 15 O. minutilus 6 69 39 30 13 ing the early stage of oogenesis (Fig. 3). Thus, itis | rotate within the basement membrane before or inferred that in medaka follicles, oocytes sur- when the oocyte axis is established in the very rounded by a granulosa cell layer may begin to early stage of oogenesis, although the transloca- 598 T. IWAMATSU tion of the granulosa cell layer within the basement membrane remains to be investigated. | Associated with the rotation of the oocyte and granulosa cells on the oocyte axis, the first mor- phological evidence of Oryzias egg polarity is the appearance of the VPA. The VPA appears as cluster of verruciform structures, which are pri- mordia for the attaching filaments, prior to forma- tion of the chorion. As these filaments become progressively horn-shaped, they bend in a unilater- al direction with the rotation of the oocyte around its axis. This axis corresponds to the future animal- vegetal axis of the egg prior to differentiation of the long, slim attaching filaments at the VPA and the micropylar cell at the animal pole [1]. The oocyte axis may be established by an unknown factor(s) that determines the position of the VPA. Investigations on determination of the origin and the formation process of these filamentous struc- tures on the chorion may contribute to clarifying the determination of egg polarity in some fishes. However, it remains to be determined whether these filaments, as well as the chorion, are derived from the oocyte itself [13], the follicle cells or both [10, 14]. As described above, the animal-vegetal axis of Oryzias eggs is first recognized by the appearance of the VPA and the unilateral bending of non- attaching filaments prior to the initiation of vitel- logenesis. Therefore, the egg axis is not induced by an unequal distribution of the yolk that accumu- lates in the ooplasm during vitellogenesis. These observations with Oryzias oocytes are consistent with those with Xenopus oocytes, which may have an animal-vegetal polarity that is further differenti- ated by the endocytotic process for vitellogenin (15). In addition to yolk accumulation, a change in the distribution of attaching and non-attaching filaments on the chorion is recognized as the vitellogenic oocyte grows rapidly. The interfila- ment distance between non-attaching filaments on the chorion increases in proportion to the increase in diameter of vitellogenic oocytes, in contrast to the distance between attaching filaments on the chorion in the VPA. This suggests that the outer layer of the chorion is stretched except in the VPA where the attaching filaments elongate extremely during vitellogenesis. Attaching filaments consist of the same components as the chorion [10] or the outer layer of the chorion as shown in the present study. Therefore, in the chorion of the VPA the components which are supplied for formation of the outer layer probably by follicular cells [4, 10] should be limited due to their use in formation of the elongating attaching filaments. Consequently, less chorion is formed in the VPA than in the remaining area of the chorion. That is, since there is no change in the total number of filaments on the chorion, the lesser distance between attaching filaments must result from a lesser expansion of the chorion in the VPA than in the remaining area of the chorion. Further experimentation is being conducted currently in order to analyse the process of formation of the VPA and the mechanism responsible for the spiral pattern of attaching filaments in this fish. ACKNOWLEDGEMENTS The author thank Emiko Oshima for her help with electron microscopy, and Dr. Cherrie A. Brown, Califor- nia Primate Research Center, University of California, for assistance in preparing the manuscript. REFERENCES 1 Nakashima, S. and Iwamatsu, T. (1989) Ultra- stractural changes in micropylar cells and formation of the micropyle during oogenesis in the medaka Oryzias latipes. J. Morph., 202: 1-11. 2 Yamamoto, T. (1961) Physiology of fertilization in fish eggs. Intern. Rev. Cytol., 12: 361-405. 3 Nakano, E. (1969) Fishes. In “Fertilization: Com- parative morphology, biochemistry and immunolo- gy”. Ed. by C. Metz and A. Monroy, Vol. 2, Academic Press Inc., pp. 295-324. 4 Tsukahara, J. (1971) Ultrastructural study on the attaching filaments and villi of the oocyte of Oryzias latipes during oogenesis. Develop. Growth Differ., 13: 173-180. 5 Hart, N. H., Pietri, R. and Donovan, M. (1984) The structure of the chorion and associated surface filaments in Oryzias. Evidence for the presence of extracellular tubules. J. Exp. Zool., 230: 273-296. 6 Dumont, J. N. and Brummett, A. R. (1980) The vitelline envelope, chorion, and micropyle of Fun- dulus heteroclitus eggs. Gamete Res., 3: 25-44. 7 Iwamatsu, T., Ohta, T., Oshima, E. and Sakai, N. (1988) Oogenesis in the medaka Oryzias latipes. — Stages of oocyte development. Zool. Sci., 5: 353- 10 11 2 Filaments on the Chorion in Medaka Eggs BIBS Iwamatsu, T. (1974) Studies on oocyte maturation of the medaka, Oryzias latipes. 11. Effects of several steroids and calcium ions and the role of follicle cells on in vitro maturation. Annot. Zool. Japon., 47: 401-408. Yamamoto, T. (1958) Artificial induction of func- tional sex-reversal in genotypic females of the me- daka (Oryzias latipes). J. Exp. Zool., 137: 227-264. Hart, N. H. and Donovan, M. (1983) Fine structure of the chorion and site of sperm entry in the egg of Branchydanio. J. Exp. Zool., 227: 277-296. Kurakami, M. (1914) On eggs and fry of the pacific saury Cololabaia saira Brevoort. Rep. Hokkaido- suishi suisan, 3: 47-52. (in Japanese) TIwamatsu, T. and Ohta, T. (1981) On a relationship 13 14 15 599 between oocyte and follicle cells around the time of ovulation in the medaka, Oryzias latipes. Annot. Zool. Japon., 54: 1-29. Tesoriero, J. V. (1978) Formation of the chorion (zona pellucida) in the teleost, Oryzias latipes. II. Autoradiography of *H-proline incorporation, J. Ultrastruct. Res., 64: 315-326. Riehl, R. (1984) The oocytes of the goby Pomatos- chistus minutus IV. Electron microscopic auto- radiography of °*H-proline incorporation. Cytologia (Tokyo), 49: 127-142. Danilchik, M. and Gerhart, J. C. (1987) Dieffern- tiation of the animal-vegetal axis in Xenopus leavis oocytes. I. Polarized intracellular translocation of platelets establishes the yolk gradient. Dev. Biol., 122: 101-112. Se: ee : fs Si Alec elon deta brit hee as eal wtp Sv robin: sie iecréatsitonahe tia os oe ALICE ree shh sat od: | ht: “The ore tected) as drootly ites mn Set tel ibkaowine? of ‘tawil: ise ar ie . 3. “ = € = » | ‘ 2 Wore x ‘ i { <7 ZOOLOGICAL SCIENCE 9: 601-606 (1992) © 1992 Zoological Society of Japan Separation of X- and Y-Chromosome-Bearing Murine Sperm by Free-Flow Electrophoresis: Evaluation of Separation Using PCR SANAE A. IsHiIMA, MAKOTO OxuNo, HIDEHO Opacirt', TosHiko Mourr’, and HipEo Monri® Department of Biology, College of Arts and Sciences, University of Tokyo, Tokyo 153, 'Kamiyabe Public High School, Kanagawa 245, and University of the Air, Chiba 260, Japan ABSTRACT—The effectiveness of separation of murine X- and Y-bearing sperm by free-flow electrophoresis was evaluated by the polymerase chain reaction (PCR). The ratio of X- and Y-bearing sperm from cauda epididymis was analyzed before and after free-flow electrophoresis. AL Ye chromosome-specific sequence (pY353/B) and an autosomal sequence (myogenin) were used to estimate the ratio between X- and Y-sperm in the separated fractions. Cauda epididymal mice sperm were separated into two peak fractions under the electric field. Each peak fraction contained sperm of normal shape, however, the motility of the sperm was extremely diminished after separation by electrophoresis. DNA was extracted from 10* sperm from each fraction and from the unseparated sperm, and Y-chromosome specific PCR was performed. thes BCR experiment revealed that fraction No. 16 (the peak near the cathode) was a Y-sperm rich fraction, whereas fraction No. 22 (the peak near the anode) was a Y-sperm poor one. These results suggested that murine X- and Y-sperm could be successfully separated by free-flow electrophoresis. Analysis of the chromosome-specific sequence by PCR was demonstrated to be a direct and adequate method to evaluate the separation of X- and Y-sperm. INTRODUCTION For the purpose of preselecting the sex of offspring in domestic and laboratory animals as well as in humans, many methods have been developed for separating X- and Y-sperm [1, 2]. Recently, Johnson and collaborators [3-5] suc- ceeded in separating X- and Y-sperm from bull, boar, chinchilla, rabbit, and ram by flow cytometry. However, to distinguish the two frac- tions, they used the DNA dyes DAPI or H33342 and a high energy laser beam, so the risk of carcinogenesis, DNA damage etc., could not be eliminated. Other methods have been reported to be successful in a limited number of animal spe- cies, especially in humans [1]. Attempts to separate sperm by means of galva- Accepted February 28, 1992 Received January 24, 1992 nization have been made by many workers using various animals [6-9]. We succeeded in separating human sperm by free-flow electorphoresis when examined for the presence or absence of the F-body, a marker for Y-bearing sperm after stain- ing with quinacrine mustard [10-12]. Preliminary results have indicated that various animal sperm also show profiles indicating separation in the electric field. In these animals, however, the F-body test by quinacrine staining can not be used successfully at present. Thus, the evaluation must depend on either chromosome analysis using the zona-free hamster ova [13] or the sex check of offspring. In recent years, DNA sequences of some specific genes on either X- or Y-chromosome have been determined. These provide useful probes for dis- tiguishing the two sperm populations. In the case of the mouse, Zfy [14], pY353/B [15, 16], and Sry [17] are examples of specific probes for the Y- 602 S. A. IsHuima, M. Oxuno, H. Opaairl et al. chromosome: Of these pY353/B is highly sensitive because it recognizes a repeating sequence. In this paper, we shall describe the separation of murine X- and Y-sperm by free-flow elec- trophoresis and evaluation of purities of the sepa- rated fractions by a PCR method for the Y- chromosome-specific repeating sequence. MATERIALS AND METHODS Sperm preparation Sperm were obtained from the cauda epididymis of 4- to 5-month-old ICR male mice and were suspended in a culture dish (35X10 mm, Nunc) containing 0.5 ml of Hanks’ balanced salt solution (BSS) and 1% bovine serum albumin (BSA) on a hot plate, followed by preincubation for 15 min at 37°C. The sperm were washed by centrifugation (600 x g, 10 min) with the chamber buffer, which contained 0.226M sucrose, 30mM MgSQg, 0.5 mM MgCl, 1mM CaCl, and 10mM HEPES (N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid)-NaOH buffer, pH 7.2. The supernatant was then discarded, and 0.5 ml of fresh chamber buffer was added to the loosely packed sperm. Separation of X- and Y-bearing sperm Separation of murine X- and Y-sperm was car- ried out using a free-flow electrophoretic appar- atus (ACE 710, Hirschmann, FRG) as previously described [10-12]. The separation chamber was composed of two glass plates (4 x 13 cm) separated by a narrow gap (0.3 mm) in which chamber buffer streamed down with a laminar flow from the top of the chamber to the bottom. While an electric field of 75-V/cm was applied perpendicularly to the buffer stream, the actively motile sperm suspen- sion in a total volume of 0.5 ml was injected continuously into one spot at the top of the cham- ber. Sperm having different electrophoretic mobi- lities descended along different paths and were collected into 35 microtubes at the bottom of the chamber. The cell distribution was monitored by measurement of the intensity of scattering light at 520 nm by use of a video camera equipped with a vidicon tube. The internal computer system calcu- lated the cell mass from a series of densitometric tracings, and the results were recorded both on a monitor and on chart paper. Electrophoresis was done at 25°C. Electrode buffer consisted of 100 mM HEPES and 80 mM NaCl, pH 7.0. DNA extraction from the murine sperm DNA was extracted from the sperm of each fraction and from the unseparated sperm in capped 1.5-ml polypropylene microcentrifuge tubes. The sperm were suspended in 300 ul of a lysis buffer consisting of 100mM NaCL, 9mM EDTA (ethylenediaminetetraacetic acid), 1.5% SDS (sodium dodecyl sulfate), 40mM DTT (dithio- threitol), 187 ug proteinase K, and 10 mM Tris (tris-hydroxymethyla-minomethane)-HClI buffer, pH 8.0. The mixture was incubated for 90 min at 37°C. Organic extraction was performed as fol- lows; 300 wl of water-saturated phenol/chloroform (1:1) was added to each polypropylene tube. Then the mixture was vortexed 4 times for 20s each time and centrifuged at 12,000 rpm for 10 min using a TOMY MC-15A centrifuge. The phenol extraction was repeated once and was followed by extraction with Sevag solution (24:1, chloroform: isoamyl alcohol). After the organic extraction, the - DNA in the aliquot layer was precipitated over- night at —20°C by the addition of two volumes of 99.9% ethanol. The DNA recovered by centri- fugation was resuspended in 70% ethanol for desalting, and then suspended in 60 yl of distilled water. Standard PCR About 50 ng of sperm DNA from each fraction obtained by free-flow electrophoresis was am- plified enzymatically in 100 wl of a reaction mix- ture, which include a 1 ~M concentration of each primer, 200 ~M deoxynucleotide (Pharmacia), 1 unit of Thermus aquaticus DNA polymerase (Taq polymerase) (Perkin-Elmer Cetus), 50mM KCl, 1.5mM MgCl, 10 mM Tris-HCl (pH 8.4), and 20 yvg/ml gelatin, and was overlaid with mineral oil. PCR was performed in a programmable thermal cycler (Perkin-Elmer Cetus) for 25 or 50 cycles, each consisting of 1 min of denaturation at 94°C, 2 min of annealing at 55°C, and 2 min of polymeriza- tion at 72°C. At the end of the cycles, the samples were held at 72°C for 7 min and then cooled. Pre-Sexing of Murine Sperm 603 TaBLE 1. List of primers for PCR Locus Primer sequence "eae Ref. myogenin 5’-TTACGTCCATCGTGGACAGC-3’ 746 [14] 5’-TGGGCTGGGTGTTAGTCTTA-3’ pY353/B 5’-GAATTCATATATATGACAGACGC-3’ 102 [15, 16] 5’-CCATTCCCTTCAAATATCATACT-3’ : Samples were then either analyzed immediately by cathode anode agarose gel electrophoresis or frozen overnight at = ; —20°C and analyzed on the following day. Prim- = ers used in this experiment are summarized in > Table 1. a =e Gel Electrophoresis = , Eighteen wl of the PCR product was mixed with 2 wl of a loading buffer containing 40% sucrose, 0.25% bromphenol blue, and 0.25% xylene cyanol and run at 90 V for 2 h in an agarose gel containing x105 2.4% SeaKem LE and 1.2% NuSieve in TBE- E buffer consisting 45mM_ Tris-borate and 1mM = : EDTA, pH 8.0. — 3 Densitometric analysis of the PCR products on z ; agarose gel was performed with a DMC-33C de- = nsitometer (Toyo Sci. Co. Ltd., Japan). = i : 14 15 16 17 18 19 20 21 22 23 RESULTS Fraction Number Fic. 1. A typical profile of electrophoretically sepa- Separation of sperm by free-flow electrophoresis Sperm from the cauda epididymis of ICR mice were separated into two distinct peaks by free-flow elctrophoresis. Both peaks migrated toward the anode at pH7.2. Fig. 1 shows at typical elec- trophoretic profile obtained with the ACE 710 apparatus and the cell count of each fraction. The two peaks contained virtually the same number of sperm. The separated sperm in both peaks were normally shaped and uncontaminated with other cells but their motility was essentially zero. To evaluate what type of sperm were contained in these two peaks, PCR was performed with the DNA isolated from sperm in each fraction. PCR of unseparated sperm As a control experiment, the PCR was per- formed with DNA extracted from male and female rated murine sperm from cauda epididymis. The upper shows the densitometric trancing by ACE710 and the lower shows the sperm count of each fraction obtained. mouse blood and from unseparated sperm. As shown in Fig. 2, both male blood DNA and sperm DNA showed a strong band corresponding to amplified pY353/B. The density of pY353/B was much higher than that of myogenin (an autosomal sequence; lanes 2 and 3 from the left). On the other hand, pY353/B was not detected in female blood DNA, although a good amplification of myogenin was observed (lanes 4 and 5 from the left). PCR of electrophoretically separated sperm Sperm fractionated by free-flow electrophoresis 604 S. A. ISHIJIMA, Blood Male | Female Sperm MAYAYAY Fic. 2. Ethidium-bromide stained agarose gels of con- trol PCR experiments of pY353/B (Y) and myoge- nin (A) loci in male blood DNA, female blood DNA, and sperm DNA. M represents molecular size markers of 4,870, 2,016, 1,360, 1,107, 926, 658, 489, 267, 80 bp (pHY marker: Takara Shuzo Co., Ltd., Kyoto, Japan). @: pY353/B, —«: myogenin gene. were counted under the microscope and DNA from 10° sperm from each fraction was extracted for PCR experiments. A portion of each extracted DNA sample was used for a control experiment for the myogenin gene. Fig. 3 compares the results of the PCR experiments on fractionated and unfrac- tionated sperm. All fractions and control sperm exhibited an equivalent density of the amplified myogenin gene. In other words, all fractions and unfractionated sample contained the same amount of DNA under the present extraction conditions (Fig. 3b). On the other hand, when the Y- chromosome specific sequence pY353/B was am- plified by PCR, the fluorescence of etidium- bromide was intense in fractions nearer the M. Oxuno, H. Opbaairi et al. Meh eIGGAS Salma) i7eoai8 1D 20K 2. Fic. 3. PCR experiment of unseparated (C) and elec- trophoretically separated murine sperm. Each num- ber represents the fraction number of free-flow electrophoresis as shown in Fig. 1. For each frac- tion, DNA was extracted from 10* sperm. M represents the same molecular size markers as in Fig. 2. @: pY353/B, —a: myogenin gene. cathode. Peak fraction No. 16 exhibited the high- est fluorescence of pY353/B, and the fluorescence of the bands gradually diminished toward the anode. Densitometric analysis of PCR products The relative value of the density of the pY353/B band in fraction No. 16 was 1.6 times higher than that in the unfractionated sperm, but little differ- ence was seen in the myogenin density. On the other hand, peak fraction No. 22 showed a relative value of 0.2 for the density of the pY353/B band; Pre-Sexing of Murine Sperm 605 and again there was little difference in myogenin band. These results suggested that the cathodic peak fraction No. 16 contained Y-bearing sperm in a high ratio, while the other peak fraction No. 22 did X-bearing sperm by means of the Y- chromosome specific sequence pY353/B. DISCUSSION In the present study, the efficiency of sperm separation by means of free-flow electrophoresis was tested by use of the PCR methodology rather than quinacrine staining. The results indicated a successful separation of murine X- and Y-bearing sperm. We clearly demonstrate that DNA analysis is anew and direct method to test the separation of animal sperm. The nonisotopic PCR method used here is more rapid than Southern’s technique using specific labeled probes [18]. The former takes about 8 hours, whereas 6 days or more are needed for the latter. The separation of murine X- and Y-sperm under an electric field was also observed with a zeta- potentialmeter Photal ELS-800 (Otsuka Electro- nics Co., Tokyo; data not shown). The same was true with human sperm [12]. Furthermore, sperm of other animals such as hamster and mongolian gerbil exhibited similar two-peak profiles (data not shown). This paper is the first description of the electrophoretic separation of X and Y-sperm in a species other than human by means of free-flow electrophoresis. For human sperm, the F-body test by quinacrine staining is the conventional method to detect Y-sperm; but for other animal sperm, the F-body test is not useful and no practic- al methods, except the measurement of DNA content by flow cytometry has hitherto been avail- able. The reproducibility of the PCR method was confirmed in a calibration experiment using blood DNA from 50 ng to 1 ug (data not shown). When the amplification of DNA by PCR is performed up to the saturation level, quantitative analysis to estimate the amount of DNA is not appropriate. The present study, however, was carried out at low amplification levels so that the amount of DNA amplified by PCR was roughly proportional to that of applied pY353/B. To eliminate the possibility of artifactral amplification in experiments where only sex chromosome-speicfic sequence were used, the murine autosomal myogenin gene was used to standardize the amplification procedure and the obtained results. Equal portions of each sperm fraction was subjected to pY353/B and myogenin PCR. These reactions were carried out indepen- dently to eliminate the possibility that reactions initiated by the two sets of primers would interfere with one another. Attempts at separating X- and Y-bearing sperm by means of flow cytometry and cell sorting have been made by several workers using various anim- als, [3-5, 19]. Those have used activation the dyes DAPI or H33342 for staining DNA and a high energy laser beam. Even though the wavelength was relatively long, the risk of hereditary defects such as carcinogenesis could not be excluded. Percoll density gradient centrifugation was effec- tive in obtaining active X-bearing human sperm as the sediment, although the proportion of Y- bearing sperm did not increase in the upper layers [20]. When this techinique was applied to dairy cattle, two thirds of the delivered calves were female, but this result was not statistically signi- ficant [1]. Following free-flow electrophoresis, sperm motility was much reduced. Improvements in the Separation conditions are needed to maintain sperm activity. However, even in their present condition, separated sperm have a practical ap- plication by microinjection into ova. Most impor- tantly, free-flow electrophoresis is performed with intact sperm cells and is applicable not only to human [10-12] but also to murine sperm as de- monstrated in this paper. This method therefore to be widely applicable for separating X- and Y-sperm. ACKNOWLEDGMENTS We thank Dr. Y. Nakagome and Dr. Y. Nakahori for their valuable discussion, and Ms. S. Seki for her valu- able advice on experiments. Finally, we are grateful to Dr. J. S. Hyams for his critical reading of the manuscript. This work was supported in part by Grants-in-Aid from the Ministry of Education, Science and Culture of Japan for Research on Priority Areas (Nos. 63640001, 01640001, 02222101 and 03207101 to H. M. and No. 03207101 to M. O.). 10 11 606 REFERENCES Mohri, H., Oshio, S., Kaneko, S., Kobayashi, T., and lizuka, R. (1987) Separation and characteriza- tion of mammalian X- and Y- bearing sperm. In “New Horinzons in Sperm Cell Research”. Ed. by H. Mohri, pp. 469-481. Japan Sci. Soc. Press, Tokyo/Gordon and Breach Sci. Publ., New York. Gledhill, B. L. (1988) Selection and separation of X- and Y-chromosome-bearing mammalian sperm. Gamete. Res., 20: 377-395. Johnson, L. A., Flook, J. P., Look, M. V., and Pinkel, D. (1987) Flow sorting of X and Y chromo- some-bearing spermatozoa into two populations. Gamete. Res., 16: 1-9. Johnson, L. A. and Clarke, R. N. (1988) Flow sorting of X and Y chromosome-bearing mammalian sperm: Activation and pronuclear development of sorted bull, boar, and ram sperm microinjectioned into hamster oocytes. Gamete. Res., 21: 335-343. Johnson, L. A., Flook, J. P., and Hawk, H. W. (1989) Sex preselection in rabbits: Live births from X and Y sperm separated by DNA and cell sorting. Biol. Reprod., 41: 199-203. . Mudd, E. B. H., Mudd, S. and Keltch, A. K. (1929) Effect of echinoid egg-waters on the surface poten- tial difference of the sperm. Proc. Soc. Exp. Med., 26: 392-394. Hafs, H. D. and Boyd, L. J. (1974) Sex ratios of calves from insemination after electrophoresis of sperm. J. Animal. Sci., 38: 603-604. Shishito, S., Shirai, M., and Matuda, S. (1974) Galvanic separation of X- and Y-bearing human spermatozoa. Andrologia 6: 17-24. Engelmann, U., Krassnigg, F., Schatz, H., and Schill, W.-B. (1988) Separation of human X and Y spermatozoa by free-flow electrophoresis. Gamete Res., 19: 151-159. Kaneko, S., lizuka, R., Oshiro, S., Nakajima, H., and Mohri, H. (1983) Separation of human X- bearing and Y-bearing sperm using free-flow elec- trophoresis. Proc. Japan Acad. (ser. B), 59: 276- DID: Kaneko, S., Oshio, S., Kobayashi, T., lizuka, R., and Mohri, H. (1984a) Human X- and Y-bearing 2 13 14 15 16 17 18 19 20 S. A. IsHvimA, M. Oxuno, H. OpDaairi et al. sperm differ in cell surface sialic acid content. Biochem. Biophys. Res. Commun., 124: 950-955. Ishijima, S., Okuno, M., and Mohri, H. (1991) Zeta potential of human X- and Y-bearing spermatozoa. Int. J. Androl., 14: 340-347. Ueda, K. U., Yanagimachi, R. (1987) Sperm chromosome analysis as a new system to test human X- and Y-sperm separation. Gamete. Res., 17: 221- 224. Koopman, P., Gubbay, J., Collignon, J., and Lovell-Badge, R. (1989) Zfy gene expression pat- terns are not compatible with a primary role in mouse sex determination. Nature (London), 342: 940-942. Bishop, C. E., Boursot, P., Baron, B., Bonhomme, F., and Hatat, D. (1985) Most classical Mus muscu- lus domesticus laboratory mouse strains carry a Mus musculus musculus Y chromosome. Nature (Lon- don), 315: 70-72. Bradbury, M. W., Isola, L. M., and Gordon, J. W. (1990) Enzymatic amplification of a Y chromosome repeat in a single blastomere allows identification of the sex of preimplantation mouse embryos. Proc. Natl. Acad. Sci. USA, 87: 4053-4057. Gubbay, J., Collignon, J., Koopman, P., Capel, B., Economou, A., Munsterberg, A., Vivian, N., Goodfellow, P., and Lovell-Badge, R. (1990) A gene mapping to the sex-determining region of the mouse Y chromosome is a member of a novel family - of embryonically expressed genes. Nature (Lon- don), 346: 245-250. Southern, E. M. (1975) Detection of specific sequ- ences among DNA fragments separated by gei elec- trophoresis. J. Mol. Biol. 98: 503-517. Metezeau, P., Cotinot, C., Colas, G., Azoulay, M., Kiefer, H., Goldberg, M. E., and Kirszenbaum, M. (1991) Improvement of flow cytometry analysis and sorting of bull spermatozoa by optical monitoring of cell orientation as evaluated by DNA specific prob- ing. Mol. Reprod. Dev. 30: 250-257. Kaneko, S., Oshio, S., Kobayashi T., Mohri H., and lizuka R. (1984b) Selective isolation of human X-beargin sperm by differ ential velocity sedimenta- tion in Percoll density gradients. Biomed. Res. 5: 187-194. ZOOLOGICAL SCIENCE 9: 607-617 (1992) Plasma Steroid Hormone Profiles during HCG Induced Ovulation in Female Walking Catfish Clarias batrachus MUHAMMAD ZAIRIN, Jr.', KryosH! ASAHINA~, KtyosH1 FURUKAWA! and KaTsuMI AIDA*! ‘Department of Fisheries, Faculty of Agriculture, The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan *College of Agriculture and Veterinary Medicine, Nihon University, Setagaya-ku, Tokyo 154 ABSTRACT—The present experiment was performed in order to investigate the response of walking catfish to human chorionic gonadotropin (HCG) and the resultant hormonal changes during ovulation. Mature female walking catfish were given a single intramuscular injection of 0.8 IU HCG/g body weight. Of 14 fish treated with a single injection of HCG, 5 fish ovulated at 20 hr and 9 fish at 24 hr following treatment. Germinal vesicle breakdown occurred after an elapse of 12-16 hr. After HCG injection, plasma testosterone peaked at 4hr, and then gradually decreased to initial levels at 24 hr. Progesterone levels started to increase at 4hr, and exhibited a small peak at 12hr. Plasma 17a-hydroxyprogesterone levels began to increase at 8 hr, peaked at 12 hr, and returned to basal levels at 20 hr. Plasma 17a,20-dihydroxy-4-pregnen-3-one (17a,208-P)levels suddenly increased and peaked at 12 hr following treatment. 17a,208,21-trihydroxy-4-pregnen-3-one (20/-S) also increased at 12 hr and peaked at 16 hr. Meanwhile, 17a,20a-dihydroxy-4-pregnen-3-one, a stereoisomer of 17a,208-P, started to increase at 4 hr, reaching a peak at 12 hr, with maximum levels lower than those of 17a,206-P or 206-S. These peaks were concomitant to the first occurrence of GVBD. Plasma estradiol-17/ levels in HCG-treated fish remained constant throughout the experiment, whereas levels in the control group were seen to decrease. These results indicate that HCG is effective in inducing ovulation in walking catfish and suggest that 17a,208-P and/or 208-S are the maturation inducing steroid(s) in this species. © 1992 Zoological Society of Japan INTRODUCTION Changes in plasma gonadotropin (GtH) and/or steroid hormone levels during ovulation have been investigated intensively in several teleost species in order to understand the endocrine control of ovulation in fish. These investigations showed that the process of ovulation occurred following an increase of internal GtH secretion from the pitui- tary gland. This GtH mediates the process of final oocyte maturation and ovulation by inducing the synthesis of the maturation inducing steroid (MIS), 17a,20f-dihydroxy-4-pregnen-3-one (17a, 208-P), in ovarian follicles [1, 2]. It has been proposed recently that 17a,208,21-trihydroxy-4- pregnen-3-one (208-S) is also an MIS in the Atlan- Accepted February 25, 1992 Received January 16, 1992 * To whom reprint requests should be addressed tic croaker [3] and the spotted seatrout [4]. 17a,20a-dihydroxy-4-pregnen-3-one (17a,20a-P) is reported to be produced during ovulation in flatfish [5]. At present, however, available in- formation concerns a limited number of species, and information on tropical fishes is insufficient. The walking catfish, Clarias batrachus, is a tro- pical freshwater fish belonging to the order Siluri- formes. In the natural habitat of this species, spawning occurs during the rainy season and can be induced by manipulating the water level of the culture pond for practical purposes. Normal ovulation in walking catfish can be induced not only by environmental means, but also by artificial stimulation such as HCG injection [6]. However, equivalent information on hormonal changes dur- ing ovulation in Clarias batrachus is not yet avail- able. Therefore, in this investigation hormonal changes during ovulation were studied using fish treated with HCG. Changes in plasma testoster- 608 M. ZAIRIN JrR., K. ASAHINA et al. one, progesterone, 17a-hydroxyprogesterone (17a- P), 17a,208-P, 208-S, 17a,20a-P and estradiol-17 were monitored during ovulation, and simul- taneously, the timing of germinal vesicle break- down (GVBD) was examined. MATERIALS AND METHODS Fish stock Two-months old walking catfish were trans- ported from Bogor, Indonesia, to Japan in 1988, and reared in an indoor concrete pond (width x length x depth=1.5 <3 X0.5 m) supplied with run- ning freshwater (23-25°C) of deep-well origin at the Fisheries Laboratory, the University of Tokyo, Maisaka, Shizuoka Prefecture. During this rearing period, fish were subjected to a photoperiodic cycle of 12L12D. Fish were fed twice per day with commercially available trout pellets at a daily ration of 2-3% of body weight. Under the above stocking conditions, fish began to mature at an age of 9 months and thereafter, conditions of maturity were maintained without undergoing natural spawning (Zairin et al., unpublished data). Experimental fish Twenty four 18-months old mature female walk- ing catfish (325-675 g in weight and 34-40 cm in total length) were selected from the stock pond, and then randomly assigned to two groups, HCG- injected and control groups. Each group was further divided into two sub-groups. Each sub- group was kept separately in an indoor concrete tank (width x length x depth=1x3X0.5 m)_ sup- plied with running fresh water of deep-well origin 23-25°C. Photoperiod was maintained 12L12D. Fish did not receive feed during the experiment. Treatment HCG was purchased from Teikoku Zoki Phar- maceutical Company, Japan. In order to meet the required dosage, the original hormone was diluted with 0.6% saline solution into a 1 [U/yl solution. The HCG-treated group received an intramuscu- lar injection of 0.8IUHCG/g body weight, whereas the control group received an equal volume of saline injection, just below the front edge of the dorsal fin. A single injection was given to each fish at 0800 hr. Blood and oocyte sampling Initial blood and oocyte samples were taken from all fish before administering the hormone solution or saline. Thereafter, sampling was car- ried out as follows. Up until 24hr of post- treatment, sampling was performed alternately between the two HCG sub-groups as well as between the control sub-groups. The first sub- groups were sampled at 4, 12, 20, 24, 28 and 52 hr, and the second sub-groups were sampled at 8, 16, 24, 28 and 54 hr. Data from two sub-groups were combined in order to obtain 4 hour interval data for each type of treatment. Approximately 0.8 ml of blood was drawn from the caudal vasculature with a heparinized syringe fitted with a 24-gauge, 1-inch needle after anesthe- tizing fish with 600 ppm of 2-phenoxyethanol (Wako, Japan). Blood samples were centrifuged at 3000 rpm, and plasma was stored in 1.5 ml polypropylene centrifuge tubes at —20°C until analysis by enzyme immunoassay (EIA, for 20/-S) or radioimmunoassay (RIA, for other steroids). Following blood sampling, a small amount of oocytes was drawn by using a polyethylene cannula (2.0 mm in inner diameter, 2.5 mm in outer dia- meter). Oocytes were treated with clearing solu- tion (ethanol: formalin: acetic acid=6:3:1) for ascertaining whether GVBD had occurred. RIA Steroid extraction from 0.25 ml plasma was car- ried out twice using 2 ml diethylether. The ether was evaporated using a centrifugal evaporator at room temperature. Samples were reconstituted with 0.5 ml of PBS containing 0.1% gelatin, 10 mM phosphate buffer and 140 mM NaCl (pH 7.5). In this experiment, plasma testosterone, prog- esterone, 17a-P, 17a,208-P, 17a,20a-P and estra- diol-172 were determined by RIA. Details for RIAs for each steroid have been described pre- viously [7-10]. Testosterone was determined using [1,2,6,7-°H] testosterone (Amersham, England) and an anti- serum against testosterone-1la-succinate-BSA. The antiserum was kindly provided by Prof. M. Catfish Steroid Changes during Ovulation 609 Honma, Laboratory of Veterinary Physiology, the University of Tokyo, Japan. This antiserum against testosterone -cross-reacted with 11- ketotestosterone, 5a-dihydrotestosterone, andros- tenedione, and androstenediol at 1.5, 30, 1.0, and 0.25%, respectively. Progesterone was determined using [1,2,6,7-°H] progesterone purchased from New England Nuc- lear, England, and an antiserum against progester- one-3-carboxy-methyl-oxime-BSA (Teikoku Zoki Pharm. Co., Tokyo). This antiserum cross-reacted with 17a-P, 20a-hydroxyprogesterone, pregneno- lone, 11-deoxycorticosterone at 0.89, 6.73, 1.46, and 6.60%, respectively. 17a-P was determined using [1,2,6,7-*H] 17a-P (New England Nuclear) and an antiserum against 17a-hydroxyprogesterone-3-oxime BSA (Teikoku Zoki Pharm. Co.). The antiserum cross-reacted with progesterone, 20a-hydroxyprogesterone, and pregnenolone at 7.85, 3.23, and 0.52%, respec- tively. 17a,208-P was determined using [22657 17a,208-P and an antiserum against 17a,20/- dihydroxy-4-pregnen-3-oxime-BSA which was kindly provided by Dr. Y. Nagahama, National Institute for Basic Biology, Okazaki, Japan. The antiserum to 17a,208-P cross-reacted with 17a, 208-P, 17a,20a-P, and 5f-pregnane-3,17a,20/- triol at 2.54, 1.55, and 0.82%, respectively. 17a,20e-P was determined using [1,2,6,7-*H] 17a,20a-P and an antiserum against 17a,20a- dihydroxy-4-pregnen-3-oxime-BSA_ which was kindly provided by Dr. A. Kambegawa, Depart- ment of Obstetrics and Gynecology, Teikyo Uni- TABLE 1. binding rate, respectively versity School of Medicine, Tokyo, Japan. The antiserum to 17a,20a-P cross-reacted with 17a,20/- P, progesterone, and deoxycortisol at 0.48, 0.17, and 0.10%, respectively. Plasma levels of estradiol-17@ were determined using [2,4,6,7--H] estradiol-17@ (New England Nuclear) and an antiserum against estradiol-173-6- CMO-BSA (Teikoku Zoki Pharm. Co.). The antiserum against estradiol-17 cross-reacted with estrone, estriol, and testosterone at 3.2, 1.77, and 0.29%, respectively. Validation of the system for use in walking catfish plasma was achieved by obtaining parallel curves for serial dilutions of plasma samples col- lected from several fishes. Intraassay and interas- say coefficients of variation at binding rates of 25% , 50% and 75% are presented in Table 1. EIA Plasma 208-S levels were measured using a specific EIA. The method for steroid extraction for EIA was the same as that for RIA. Samples were reconstituted with 0.05% borate buffer containing 0.5% BSA (pH7.8). An antiserum was raised against 17a,20,21-trihydroxy-4- pregnen-3-CMO-BSA. This antiserum cross- reacted with progesterone, 17a-P, 17a,208-P, and 17a,20a-P at 1.0, 0.01, 0.01 and 0.8%, respective- ly. Horseradish peroxidase (Sigma, USA) was used for labeling the antigen. Absorbance at 492 nm was measured using an EIA reader (Bio Rad, England) for microtiter plates. Intraassay and interassay coefficients of variation at binding rates of 25%, 50% and 75% are presented in Table 1. Intraassay and interassay coefficients of variation measured at 25, 50, and 75% of Intraassay Interassay Steroids 25% 50% 75% 25% 50% 75% Testosterone 19.6 6.6 6.0 22.6 leg) 18.0 Progesterone 4.0 Sil 8.3 4.7 Sail 4.3 17a-P 55 6.0 Teal 15.4 jOZ2) 1320 208-S 13.0 2S EZ 19.5 17.4 12.6 17a, 20a-P 9.7 6.8 4.0 OZ 8.7 11.0 17a, 208-P 353 4.7 De) 320 533 9.0 Estradiol-17 10.9 4.8 3.0 ee 5) 8.6 610 M. ZAIRIN JR., K. ASAHINA et al. Details for this EIA will be published separately (Asahina et al., unpublished data). Statistics The Student-t test was used to compare means between treated and control groups. The multiple range test of Duncan and the Kruskal-Wallis test were used to analyze the time course changes in each group. RESULTS Ovulation occurred in all fish in the HCG- treated group. Of 14 fish treated with HCG, 5 fish ovulated at 20hr and 9 fish ovulated at 24 hr following treatment. GVBD was observed at 12 and 16 hr. However, all fish in the control group failed to ovulate. Changes in plasma testosterone levels both in the HCG-treated and control group are shown in Fig. 1. In the HCG-treated fish, plasma testoster- one levels showed a rapid increase (P< 0.01; 0 hr vs 4 hr), peaking at 4 hr (38.3 ng/ml), and gradual- ly returned to initial levels at 24 hr (P<0.01; 4 hr vs 24 hr). Plasma testosterone levels in the control group started to decrease at 16 hr (P<0.01; 0 hr vs 50 40 30 20 TESTOSTERONE (ng/ml) 10 0 4 8 12 16 hr) without returning to initial levels at the end of the experimental period (P<0.01; Ohr vs 52 hr). Changes in plasma progesterone levels both in the HCG-treated and control groups are presented in Fig. 2. The change of amplitude in levels of this hormone was small but could be considered signi- ficant (P<0.01; 0 hr vs 12 hr). Progesterone levels in HCG-treated fish started to increase at 4 hr after the treatment (P<0.01; 0 hr vs 4 hr) and reached a peak (1.7 ng/ml) at 12 hr, and subsequently de- creased below the detectable limit at 20 hr. With the exception of the beginning and at the end of the experimental period, plasma progesterone levels in the control group were always below the detectable limit. Changes in plasma 17a-P levels both in the HCG-treated and control groups are shown in Fig. 3. Plasma 17a-P levels in the HCG-treated group increased at 8 hr (P<0.01; 0 hr vs 8 hr), peaked at 12 hr (20.0 ng/ml) (P<0.01; 8hr vs 12 hr), and then rapidly returned to initial levels after 16 hr (P <0.01; 12hr vs 16hr). Thereafter, hormone levels showed small fluctuations. No significant change was observed in the control group. Changes in plasma 17a,20-P levels both in the e——e HCG-Treated o——o Control GVBD MUA OVULATION Wy 0 O O O tne l6aieh 20: Vi) 24" 28, eee HOURS AFTER TREATMENT Fic. 1. Changes in plasma testosterone levels during HCG-induced ovulation in walking catfish. Data from 4-20 hr represent 7 and 5 fish for treated and control groups, respectively. Subsequent data represent 14 and 10 fish for treated and control groups, respectively. Each point is represented as mean+SEM. Catfish Steroid Changes during Ovulation 611 PROGESTERONE (ng/ml) 0 4 8 12 GVB MUA e——e HCG-Treated o——o Control OVULATION Wan 16 20 24 28 52 HOURS AFTER TREATMENT Fic. 2. Changes in plasma progesterone levels during HCG-induced ovulation in walking catfish. Arrow indicates the detectable limit of the assay. See Fig. 1 for experimental detail. 25 20 17a-P (ng/ml) an —_ So 0 4 8 12 v——v HCG-Treated GVBD Wan v——v Control OVULATION MUM 16 20 24 28 52 HOURS AFTER TREATMENT Fi. 3. experimental detail. HCG treated and control group are shown in Fig. 4. In the HCG-treated fish, plasma 17a,20,-P levels were below the detectable limit of 0.24 ng/ ml until 8 hr after injection followed by a sudden increase (P<0.01; 8hr vs 12 hr) until reaching a peak (8.3 ng/ml) at 12 hr in association with the initiation of GVBD. Thereafter, the levels drop- Changes in plasma 17a-P levels during HCG-induced ovulation in walking catfish. See Fig. 1 for ped below the detectable limit at 24 hr (P<0.01; 12 hr vs 24hr). Plasma 17a,208-P levels of the control group were always below the detectable limit during the course of the experiment. Changes in plasma 20£-S levels both in the HCG treated and control group are shown in Fig. 5. As just observed in 17a,208-P levels, plasma 20{-S 612 M. ZAIRIN JR., K. ASAHINA et al. 10 ~ 170,20 6-P (ng/ml) Oo 0 4 8 12 GVBD Wa =—s HCG-Treated o—1a Control OVULATION MM 16 20 24 28 52 HOURS AFTER TREATMENT Fic. 4. Changes in plasma 17a, 208-P levels during HCG-induced ovulation in walking catfish. Arrow indicates the detectable limit of the assay. See Fig. 1 for experimental detail. GVBD Mu“ 20£-S (ng/ml) 0 4 8 12 u—an HCG-Treated o——o Control OVULATION We 16 20 24 28 52 HOURS AFTER TREATMENT Fic. 5. experimental detail. levels stayed low during 8 hr after injection, then increased rapidly at 12 hr (5.6 ng/ml; P<0.01; 8 hr vs 12 hr). Levels ramained high until 16 hr and then decreased returning to the initial levels threafter. On the other hand, no significant changes occurred in the control group during the course of the experiments. Changes in plasma 17a@,20a-P levels both in the Changes in plasma 206-S levels during HCG-induced ovulation in walking catfish. See Fipsl tor HCG treated and control group are shown in Fig. 6. The hormone started to increase at 4 hr (P< 0.01; 4hr vs 12 hr), peaked at 12 hr (2.2 ng/ml), and then decreased. The magnitude of its peak was lower than those of 17a,208-P or 208-S. Plas- ma 17a,20a-P in the control group remained under detectable limits throughout the experiments. Changes in plasma estradiol-17? levels both in Catfish Steroid Changes during Ovulation 613 3 #——_ HCG - Treated GVBD O— Control MU“ = 2 oS = QO. 3 oO N 2 = 1 OVULATION Mla ~—+—— 0 4 8 12 16 20 24 28 52 HOURS AFTER TREATMENT Fic. 6. Changes in plasma 17a, 20a-P levels during HCG-induced ovulation in walking catfish. Arrow indicates the detectable limit of the assay. See Fig. 1 for experimental detail. 15 GVBD yy a—a HCG-Treated 4——“ Control OVULATION ue VV ESTRADIOL -17/3 (ng/ml) 1) 4 8 12 16 20 24 28 52 HOURS AFTER TREATMENT Fic. 7. Changes in plasma estradiol-17f levels during HCG-induced ovulation in walking catfish. See Fig. 1 for experimental detail. the HCG-treated and control groups are presented Changes in the seven sex steroid hormone levels in Fig. 7. In the HCG-treated group, no singificant —_ during ovulation in the HCG-treated female walk- change was observed, and plasma levels were ing catfish are summarized in Fig. 8. maintained at relatively high levels. In the control group, however, a gradual decrease occurred dur- ing the experiment (P<0.01; 0 hr vs 20 hr). 614 M. ZatIRIN JR., K. ASAHINA et al. 50 i) (2%) = (=) (=) (<>) HORMONE LEVELS (ng/ml) —_ (=) 0 4 8 12 GVBD Wy e——+ Testosterone «—- Estradiol-17/3 O----0 171-P s—=# 171,20/}-P 4 ----A Progesterone o—© 170,200-P O—O 208-S OVULATION We = 16 20 24 28 52 HOURS AFTER TREATMENT Fic. 8. Changes in the seven sex steroid hormone levels during HCG-induced ovulation in walking catfish. Arrow indicates the detectable limit of the assay. See Fig. 1 for experimental detail. DISCUSSION HCG has been used effectively in Clarias mac- rocephalus [11] and Clarias lazera [12] in the induction of ovulation. Recently, HCG has been successfully employed in inducing normal ovula- tion of Clarias batrachus [6]. In the present experiment, of 14 fish treated with HCG, 5 fish ovulated at 20 hr and 9 fish ovulated at 24 hr. These results indicate that the function of internal GtH in catfish can be replaced by HCG. Various doses of HCG (0.2, 0.4, 0.8, and 1.6 IU/g body weight) have been tried in inducing ovulation of walking catfish (Zairin et al., unpub- lished data). In all the trials, HCG treatment succeeded in inducing ovulation at dosages of 0.4 IU/g body weight or higher. At 0.41TU/g body weight, the quantity of ovulated eggs was small and practically negligible, whereas a large number of ovulated eggs were observed in the group treated with 0.8 and 1.6 IU HCG/g body weight. Detail about the experiment above will be re- ported in the next paper. Considering that there is no significant difference between dosages of 0.8 and 1.6 IU/g body weight, dose of 0.8 IU HCG/g body weight has been chosen in the present experi- ment. Testosterone levels peaked at 4hr. Among the sex steroids monitored, this peak is the earliest and highest. This suggests that the injection of HCG activates steroidogenesis in the follicular layer. It ” is commonly accepted at present that GtH works on the theca cells of the follicle and stimulates the synthesis of testosterone which in turn is converted into estradiol-172 in granulosa layer [13]. In contrast, testosterone levels in the control groups decreased during the experiment. This decrease is probably due to some unknown problems in the sampling protocol, such as repeated blood sam- pling, stress caused by handling or an effect of anesthetics. HCG injection caused testosterone to peak at 4 hr which was followed by the increase in 17a-P, a precursor of 17a,208-P. This suggests that the shift in the steroidogenic pathway, from androgen to progestin synthesis, is induced by HCG injection as iS proposed according to in vitro studies in Clarias macrocephalus [14]. 17a-P secreted from thecal layer is likely converted into 17a,208-P in the granulosa layer where HCG, an external GtH, acts to enhance the activity of 208-HSD, which is a key enzyme in this conversion [1]. As a result, a peak of 17a,208-P occurs and propels the oocytes to final maturation. Catfish Steroid Changes during Ovulation 615 Progesterone levels in this experiment increased over the basal level. It is not known whether this hormone possesses any role in oocyte maturation and ovulation in fishes. Jn vitro studies in Clarias lazera [15], however, showed that no changes in progesterone levels occurred at and after ovula- tion. Most likely, progesterone is not involved in either final maturation or ovulation. In this experiment, 17a-P levels, a precursor of 17a,208-P, increased prior to those of 17a,20/-P, showing their precursor-product relationship. Both of these hormones showed characteristic changes: very high levels during the oocyte maturation process, declining levels at ovulation. In unovulated fish of the same species, 17a-P was detected in low levels throughout the year (Zairin et al., unpublished data), whereas 17a,208-P was under detectable limit of our assay. Accordingly, these changes strongly suggest that both steroids play important roles in oocyte maturation in this species. The occurrence of prominent peak in both hormones around the time of oocyte maturation and ovulation has been reported in goldfish [16], bitterling [7], carp [17], and salmonids [18-19]. The pattern of increase in both 17a,208-P and 17a-P seems to differ according to species. In bitterling, 17a,208-P increased over 17a-P at peak [7]. However, in the present investigation, 17a-P increased over 17a,208-P. Similar results were obtained in the winter flounder, Pseudo- pleuronectes americanus [20]. In the present study, we found that the plasma levels of 206-S also increased during several hours prior to ovulation. Since 20{-S is as effective as 17a,208-P in inducing final oocyte maturation in some teleost species [2], it is probable that both 17a,208-P and 206-S act as MIS in this species. Some in vitro experiments are being undertaken to ascertain this. The status of the maturation inducing steroid in catfish has been a source of controversy for quite a while. Formerly, corticosteroid was regarded as an oocyte maturation inducing substance in an Indian catfish, Heteropneustes fossilis [21-23]. However, subsequent results from in vitro studies on the same species [24] did not support this hypothesis. It was later reported that MIS in another Indian catfish, Mystus vitiatus is 17a,208-P [25]. As men- tioned previously, both 17a,206-P and 20/-S levels increased prior to ovulation in walking catfish. This result seems to lead to the so-called “multiple MIS” theory in teleosts. This also suggests the existence of a correlation between the ovary and interrenal kidney as shown in old maturation theories of catfish, because 208-S is a form of corticoid. On the other hand, although 17@,20a-P also increased prior to ovulation, the role of this steroid, the isomer of 17a,20{8-P, is of yet uncer- tain, since the activity of 20a-HSD is very low compared to that of 206-HSD both in vitro and in VIVO. Patterns of fluctuation in estradiol-17f in this experiment are particularly interesting, as this hormone did not peak following a peak of testos- terone. HCG injection did not change the plasma estradiol-17 levels. This is in contrast to goldfish [26], where a peak of estradiol-17£ is subsequent to a peak of testosterone. Estradiol-17/ levels in the control groups decreased during the experi- ment, and did not return to the basal levels as expected, likely due to some unknown problems in the sampling protocol, such as repeated blood sampling, stress caused by handling or an effect of anesthetics. Considering this possibility, it could be assumed that estradiol-178 production in the HCG-treated group is actually stimulated during ovulation. In another experiment, when spawning is induced experimentally by water and tempera- ture level manipulation, the occurrence of high levels of estradiol-17@ in ovulated- and _ just- spawned females were observed (Zairin et al., unpublished data). In the present experiment, higher estradiol-17@ levels in the HCG-treated group than in the control group may be due to the action of HCG on the vitellogenic oocytes, as this fish possesses various sizes of oocytes throughout the year (Zairin et al., unpublished data). ACKNOWLEDGMENTS We express our thanks to colleagues at the Faculty of Fisheries, Bogor Agricultural University, Indonesia, for their kind help in supplying catfish fry stock. This study was partially funded by a Grant-in Aid for Scientific Research from the Ministry of Education, Science and Culture. 10 11 12 616 REFERENCES Nagahama, Y. (1987) Gonadotropin action on gametogenesis and_ steroidogenesis in_ teleost gonads. Zool. Sci., 4: 209-222. Scott, A. P. and Canario, A. V. M. (1987) Status of oocyte maturation-inducing steroid in teleost. Proc. of the Third Int. Symp. on the Reprod. Physiol. of Fish, St. John’s, Newfoundland, Canada, pp. 224- 234. Trant, J. M. and Thomas, P. (1989) Isolation of a novel maturation-inducing steroid produced in vitro by ovaries of Atlantic croaker. Gen. Comp. Endoc- rinol., 75: 397-404. Trant, J. M. and Thomas, P. (1989) Evidence that 17a,208,21-trihydroxy-4-pregnen-3-one is a matura- tion inducing steroid in spotted seatrout. Fish Phy- siol. Biochem., 7: 185-191. Canario, A. V. M. and Scott, A. P. (1989) Synthesis of 20a-hydroxylated steroids by ovaries of the dab (Limanda). Gen. Comp. Endocrinol., 76: 147-158. Zonneveld, N., Wilbrink, A. C., Soeprijanto, A., Viveen, W. J. A. R. and Nursalam, Y. (1989) Induced spawning of the Asian catfish (Clarias bat- rachus) by means of HCG. The Second Asian Fisheries Forum, Tokyo, pp. 587-590. Shimizu, A., Aida, K. and Hanyu, I. (1985) Endoc- rine profiles during the short reproductive cycle of an autumn-spawning bitterling, Archeilognathus rhombea. Gen. Comp. Endocrinol., 60: 361-371. Aida, K., Kato, T. and Awaji, M. (1984) Effects of castration on the smoltification of precocious male masu salmon Oncorhynchus masou. Bull. Japan. Soc. Sci. Fish., 50: 565-571. oust SaaWeaeAidal Kee ranyulee SakaryakKe: Nomura, M., Tanaka, M. and Tazaki, S. (1984) Endocrine profiles in the female of a twice-annually spawning strain of rainbow trout. Aquaculture, 43: 13-22. Kobayashi, M., Aida, K. and Hanyu, I. (1986) Hormone changes during the ovulation process in the goldfish. In “Pars Distalis of the Pituitary Gland: Structure, Function and Regulation” (F. Yoshimura and A. Gorbman, Eds.), pp. 477-479. Elsevier, Amsterdam. Carreon, J. A., Estocapio, F. A. and Enderez, F. M. (1976) Recommended procedures for induced spawning and fingerling production of Clarias mac- rocephalus Gunther. Aquaculture, 8: 269-281. Eding, E. H., Janssen, J. A. L., Kleine Staarman, Grebe] andsRichter.@s Je. (1982) Effects of human chorionic gonadotropin (HCG) on matura- tion and ovulation of oocytes in the catfish Clarias lazera (C & V). In “Proceedings of the International Symposium on Reproductive Physiology of Fish, Wageningen, The Netherlands,” pp. 99-102. i) 14 15 16 7, 18 19 20 21 LD, M. ZAIRIN JR., K. ASAHINA et al. Kagawa, H., Young, G., Adachi, S. and Nagahama, Y. (1982) Estradiol-172 production in amago sal- mon (Oncorhynchus rhodurus) ovarian follicles: Role of the thecal and granulosa cells. Gen. Comp. Endocrinol., 47: 440-448. Suzuki, K., Tan, E. S. P. and Tamaoki, B. (1989) Change of steroidogenic pathways in the ovary of a tropical catfish, Clarias macrocephalus, Gunther, after HCG treatment. Gen. Comp. Endocrinol., 76: 223-229. Lambert, J. G. D. and van den Hurk, R. (1982) Steroidogenesis in ovaries of African catfish, Clarias lazera, before and after an HCG induced ovulation. In “Proceedings of the International Symposium on Reproductive Physiology of Fish, Wageningen, The Netherlands,”, pp. 99-102. Kobayashi, M., Aida, K. and Hanyu, I. (1987) Hormonal changes during ovulation and effects of steroid hormones on plasma gonadotropin levels and ovulation in goldfish. Gen. Comp. Endocrinol., tle DAD Santos, A. J. G., Furukawa, K., Kobayashi, M., Bando, K., Aida, K. and Hanyu, I. (1986) Plasma gonadotropin and steroid hormone profiles during ovulation in the carp Cyprinus carpio. Bull. Japan. Soc. Sci. Fish., 52: 1159-1166. Scott, A. P., Sheldrick, E. L. and Flint, A. P. F. (1982) Measurement of 17a,20$-dihydroxy-4- pregnen-3-one in plasma of trout (Salmo gairdneni . Richardson): Seasonal changes and response to sal- mon pituitary extract. Gen. Comp. Endocrinol., 46: 444_451. Scott, A. P., Sumpter, J. P. and Hardiman, P. A. (1983) Hormone changes during ovulation in the rainbow trout (Salmo gairdneri Richardson). Gen. Comp. Endocrinol., 49: 128-134. Campbell, C. M., Walsh, J. M. and Idler, D. R. (1976) Steroids in the plasma of the winter flounder (Pseudopleuronectes americanus Walbaum). A sea- sonal study and investigation of steroid involvement in oocyte maturation. Gen. Comp. Endocrinol., 29: 14-20. Goswami, S. V. and Sundararaj, B. I. (1971) Tem- poral effects of ovine lutenizing hormone and deox- ycorticosterone acetate on maturation and ovulation of oocytes of the catfish, Heteropneustes fossilis (Bloch): An in vivo and in vitro study. J. Exp. Zool., 178: 457-466. Goswami, S. V. and Sundararaj, B. I. (1976) In vitro maturation and ovulation of oocytes of the catfish, Heteropneustes fossilis (Bloch): Effects of mammalian hypophyseal hormones, catfish pituitary homogenates, steroid precursors and metabolites, and gonadal and adrenocortical steroids. J. Exp. Zool., 178: 467-478. Sundararaj, B. I. and Goswami, S. V. (1977) Hor- 24 25 Catfish Steroid Changes during Ovulation _ monal regulation of in vivo and in vitro oocyte maturation in the catfish, Heteropneustes fossilis (Bloch). Gen. Comp. Endocrinol., 32: 17-28. Ungar, F., Gunville, R., Sundararaj, B. I. and Goswami, S. V. (1977) Formation of 3a-hydroxy- 5-pregnen-20-one in the ovaries of catfish, Heterop- neustes fossilis (Bloch). Gen. Comp. Endocrinol., 31: 53-59. Upadhyaya, N. and Haider, S. (1986) Germinal vesicle breakdown in oocytes of catfish, Mystus vittatus (Bloch): Relative in vitro effectiveness of 26 617 estradiol-17£, androgens, corticosteroids, progester- one, and other pregnen derivatives. Gen. Comp. Endocrinol., 63: 70-76. Stacey, N. E., Peter, R. E., Cook, A. F., Truscott, B., Walsh, J. M. and Idler, D. R. (1983) Endocrine changes in plasma concentration of gonadotropin, _ 17f-estradiol, testosterone, and 17a,208-hydroxy- 20-dihydroprogesterone during spontaneous and brain lesion induced ovulation in goldfish. Can. J. Zool., 61: 2646-2652. Ce) 5 > dk ‘a “> = } sty ptraved uy a es ihe fn Tey Tey Poy ee - ZOOLOGICAL SCIENCE 9: 619-624 (1992) © 1992 Zoological Society of Japan Dipsogenic Action of Brain Natriuretic Peptide and Endothelin-1 in the Japanese Quail, Coturnix coturnix Japonica Yoko TEZUKA, HipEsHI Kopayasut’ and HARUKo UEMURA Biological Laboratory, Kanagawa Dental College, Yokosuka, Kanagawa 238, and ‘Research Laboratory, Zenyaku Kogyo Co., Ltd., Nerima, Tokyo 178, Japan ABSTRACT— Porcine brain natriuretic peptide (pBNP) elicited a significant increase in water intake, when administered intraperitoneally (5 and 10 “g/bird) or intracerebroventricularly (0.3 ~g/bird), in the water-replete Japanese quail, Coturnix coturnix japonica. Intraperitoneal injection of endothelin-1 (ET-1; 0.3, 1 and 3 ug/bird) did not affect water intake, but intracerebroventricular administration of ET-1 (10 ng/bird) slightly enhanced water intake in the water-replete Japanese quail. The possible involvement of these peptides in thirst mechanisms is discussed. INTRODUCTION Porcine brain natriuretic peptide (pBNP) is composed of 26 amino acid residues [1, 2], and exhibits high sequence homology to a-human atrial natriuretic peptide (a-hANP). Injections of pBNP have natriuretic, diuretic and hypotensive actions similar to those induced by a-hANP in rats {1}. Furthermore, intracerebroventricular (i.c.v.) in- jections of pBNP [3] or rat BNP (rBNP) [4, 5] suppress the water intake that is stimulated by angiotensin II (ANG IJ) or dehydration in rats, as seen also with i.c.v. injections of a-rANP (5-28) [6, 7], a-rANP [7], and a-hANP [7-9]. However, pBNP [3] and rBNP [5] appear not to suppress water intake in water-replete rats. In the water- replete Japanese quail, by contrast, both intraperi- toneal (i.p.) and i.c.v. injections of e-hANP stimu- lated water intake [10]. One of the purposes of this study was to examine whether or not pBNP is antidipsogenic or dipsogenic in the Japanese quail. Endothelin (ET) is a potent pressor/vasocon- strictor peptide, and there are at least three iso- forms of this peptide, ET-1, ET-2 and ET-3, in mammals [11, 12]. Samson et al. [13] reported that Accepted February 27, 1992 Received February 5, 1992 ET-3 administered into the third ventricle inhi- bited drinking that was stimulated by dehydration, hyperosmotic challenge or ANG II in the rat. A second purpose of this study was to test whether or not ET is involved in thirst mechanisms in the water-replete Japanese quail. MATERIALS AND METHODS Male Japanese quail, Coturnix coturnix japonica (8 weeks old) were purchased from a commercial source. They were housed in a room maintained at 21-25°C under a 12L photoperiod (07 : 00-19 : 00 h). Birds were kept individually in bird cages [30 (D)x15.5 (W)x22 (H)cm*] and screened one from another by pieces of cardboard inserted between the cages. Water and food were available ad libitum before and during the experimental period. Food consisted of crushed corn, kaoliang, wheat and crushed dry fish meat (Showasangyo Co., Ibaragi). The average body weight was 100.4 g, ranging from 84 to 121g, at 20 weeks after hatching. Synthetic porcine BNP, ANG II (Val?-ANG II) and ET-1 were obtained from the Peptide Insti- tute, Inc., Osaka. Each was dissolved in saline solution, and injections were given between 12 : 00 and 13:30h. During the observation period (10: 620 Y. TEzUKA, H. KoBAYASHI AND H. UEMURA 00-17:00h), the drinking rate was usually almost constant in the Japanese quail [14]. Measurement of water intake Birds were trained to drink water from a small hole at the end of a glass tube attached to an up-ended 20-ml cylinder for about 2 weeks before the first injection. The amount of water ingested was estimated by reading the scale on the cylinder to the nearest 0.1 ml. Measurements of water intake started from 2hr before injections, and were recorded every 30 min. I.c.v. injection Birds were anesthetized with Nembutal, and a stainless-steel guide cannula (o.d., 0.7 mm; length, 13mm) was implanted stereotaxically into the brain with the aid of X-rays, with the tip being guided until it was located in the third ventricle. The cannula was fixed securely to the skull with dental resin and two anchoring screws. The details of the cannulation technique have been described in an earlier paper [15]. The implanted cannula was closed off with nylon thread when it was not in use to prevent blood coagulation in the cannula. Birds were allowed to recover from the operation for at least one week before the i.c.v. injections. Injections were performed with a stainless-steel injector (o.d., 0.4mm), which was | mm longer than the guide cannula. The injector was con- nected to a 10-1 microsyringe with a 10-cm piece of polyethylene tubing (o.d., 0.4mm). To verify that the tip of the guide cannula had been success- fully located in the third ventricle, ANG II (30 or 50 ng/bird) was injected into the brain through the cannula, and the dipsogenic response of the birds was verified. Only those birds that responded to ANG II by drinking, as shown in a previous paper [10], were used for subsequent experiments. Experiment I: i.p. injection of pBNP and water intake Thirty-seven birds (10-19 weeks old) received a single i.p. injection of pBNP at a dose of 0 (saline, n=11), 1 (@=9), 5 M@=8), 10 (n=6) or 25 “xg / bird (n=3). The injection volume was 0.1 ml/bird. Water intake after the injection was measured at intervals of 30 min for 2 hr. Dose-response curve was depicted from the data obtained 30 min after the injection. Experiment IT: i.c.v. injection of pBNP and water intake Seven birds (24-27 weeks old) were given a single i.c.v. injection of pBNP at a dose of 0 (saline), 0.03, 0.1 and 0.3 ug/bird. Each bird received each dose once on different days in random order. Injection volume was 1 yl/bird. Water intake after the injection was measured at intervals of 30 min for 2 hr. Experiment III: i.p. injection of ET-I and water intake Thirty-four birds (10-19 weeks old) received a single i.p. injection of ET-1 at a dose of 0 (saline, n—9), 0:3 @—7), | @—8)s 3 — 7) orld bind (n=3). The injection volume was 0.1 ml/bird. Water intake after the injection was measured at intervals of 30 min for 4 hr. Experiment IV: i.c.v. injection of ET-1 and-water intake Seven birds (25-28 weeks old) were given a-° single i.c.v. injection of ET-1 at a dose of 0 (saline), 10 and 30 ng/bird. Birds were injected in random order, each bird receiving each dose once. Injection volume was 1 ywl/bird. Water intake after the injection was measured at intervals of 30 min for 3 hr. Statistical analysis Data from Experiments I and III were analyzed by the Kruskal-Wallis test. When differences were significant, Mann-Whitney’s U-test was also em- ployed. Data from Experiment II were analyzed by Friedman’s test. When differences were signi- ficant, Wilcoxon’s test was also employed. Wilcox- on’s test was also employed for data from Experi- ment IV. | RESULTS Experiment I: i.p. injection of pBNP and water intake Water intake for 30 min after an injection of BNP, ET-1 and Drinking in the Quail pBNP (1 to 10 yg/bird) increased in a dose- dependent manner (Y =0.220X + 0.147, P<0.001; Fig. 1). Copious water intake was induced by a single i.p. injection of pBNP (5 and 10 »g/bird), starting at 12.4+3.3 min (5 wg/bird) and 15.8+ 2.7 min (10 “g/bird) after the injection. The effect of pBNP on water intake persisted for 60 min (5 Water intake (ml/30 min/ bird) O41 5 10 Dose of BNP (yg/bird ) Fic. 1. Dose-response curve for the effects of single i.p. injections of porcine brain natriuretic peptide (pBNP) on water intake over a 30-min period in the Japanese quail. Each point shows the mean with SE. Numbers of birds are shown in parentheses. ©, Saline control; @, experimental. ** P<0.01 com- pared with saline control. 2 a 4 S = ek 10 yg (6) ® #K 3 3 = aos ® ee 5 ug (8) 3 #0 1 yg (9) i) = x3 © Saline (11) oO 31 T £ Js oO O 'e, (0) (0) 30 60 90 120 (min) Time after i.p. injection of BNP Fic. 2. Cumulative water intake after i.p. injection of pBNP in the Japanese quail. Injection volume was 0.1 ml. Each point shows the mean with SE. Num- bers of birds are shown in parentheses. ** P<0.01 compared with saline control. 621 yg/bird) and 90 min (10 ug/bird) (Fig. 2). No behavioral changes were observed after the injec- tion of pBNP at any dosage used, except when the dose was 25 ug/bird (n=3). Since behavioral depression was observed at this dosage, the data of water intake of these birds were excluded. Experiment II: t.c.v. injection of pBNP and water intake I.c.v. administration of pBNP (0.3 “g/bird) stimulated water intake (Fig. 3), and the latency was 9.7+2.2 min (n=7). Significant increases in cumulative water intake were observed 60 (P< 0.05) and 90 (P<0.05) min after injection. I.c.v. injections of 0.03 and 0.1 u~g pBNP had no effect on drinking for 120 min (Fig. 3). No overt changes were observed in the behavior of birds injected at any dosage. Water intake in control birds injected intracerebroventricularly appeared to be less than that of those injected intraperitoneally. This may be due to some unknown effects of implantation of a cannula. 1.5 2 = Sy E 0.3 yg ® * 0.1 yg x S&S 09 “= £ ry 6 Saline 3 0.6 0.03 yg 2 i“ : a 3 03 fer =) O oO 0 . 0 30 60 90 120. (min) Time after i.c.v. injection of BNP Fic. 3. Cumulative water intake after i.c.v. injection of pBNP in the Japanese quail. For each dosage, seven birds were injected with the peptide dissolved in 1 yl of saline once on different days in random order. * P<0.05 compared with saline control. Experiment III: i.p. injection of ET-I and water intake I.p. administration of ET-1 (0.3, 1, 3 ~g/bird) did not significantly affect water intake, as com- pared with saline controls, throughout the 622 Y. TEZUKA, H. KoBAyASHI AND H. UEMURA (mi/ bird ) Cumulative water intake O 30 60 90 120 3 yg (7) Saline (9) 1 yg (8) 0.3 yg (7) 150 180 210 240 (min) Time after i.p. injection of ET-1 Fic. 4. Cumulative water intake after 1.p. administration of ET-1 in the Japanese quail. Injection volume was 0.1 ml. Each point shows the mean with SE. Numbers of birds are shown in parentheses. 1.8 1.5 (mI/ bird ) 0.9 0.6 0.3 Cumulative water intake O 30 60 “10 ng Saline 120 150 180 (min) Time after i.c.v. injection of ET-1 Fic. 5. Cumulative water intake after i.c.v. administration of ET-1 in the Japanese quail. Seven birds received saline or 10 ng ET-1/bird dissolved in 1 yl saline once on different days in random order. * P<0.05 compared with saline control. observation period (Fig. 4). Severe behavioral depression occurred at a dosage of 10 “g/bird (n= 3). Therefore the data of water intake of these birds were excluded. Experiment IV: i.c.v. injection of ET-1 and water intake A significant increase in water intake was observed 3 hr after a single 1.c.v. injection of ET-1 (10 ng/bird, P<0.05) (Fig.5). No behavioral changes were observed at this dosage. At a dosage of 30 ng (n=3), however, depressed behavior was observed and hence the data of water intake were excluded. Water intake in control birds injected intracerebroventricularly was apparently less than that of those injected intraperitoneally. This may be due to some unknown effects of implantation of a cannula. DISCUSSION The present study demonstrates that i.p. (5, 10 ug/bird) and i.c.v. (0.3 ug/bird) injections of pBNP significantly increase water intake in the water-replete Japanese quail. These results sug- gest that i.p. pBNP may reach the receptive site for BNP, ET-1 and Drinking in the Quail 623 BNP in the brain and that pBNP or pBNP-like peptide may act as a dipsogen in the Japanese quail. In rats, by contrast, i.c.v. administration of pBNP [3] or rBNP [5] had no effect on water intake in water-replete rats, although these pep- tides prevented any increase in water intake as a result of dipsogenic treatments, such as injecion of ANG II [3, 4] or water deprivation [4, 5]. The difference in the response to BNP between the water-replete rat and the water-replete Japanese quail may be due to the following factors: (1) species-related differences in thirst mechanisms, for example, it is known that, in the Japanese quail, injection of carbachol does not evoke drink- ing, whereas in rats it induces drinking [8, 16]; (2) differences in doses of BNP used, namely, 0.3 to 2.0 nmol in rats and 0.03 to 0.3 ug (=0.01 to 0.1 nmol) in the Japanese quail; and (3) the use of a heterologous peptide, pBNP, in the Japanese quail, for example, it is known that in the water- replete rats, both low (200-800 ng) [17] and high (S ug) [8] doses of a-hANP do not alter water intake, whereas low dosage of a-rANP (200-800 ng) [17] stimulates drinking in water-replete rats. Discrepancies between the effects of some neuropeptides on water intake in birds and mam- mals have also been reported: physalaemine, ele- doisin and bombesin stimulate spontaneous drink- ing in the pigeon and the duck, but they inhibit drinking induced by dipsogenic treatments in the rat [18, 19]. Furthermore, a-hANP (about 1.5 nmol) stimulates spontaneous drinking in the Japanese quail [10], but the same dosage of a- hANP does not alter water intake in the water- replete rats [8] and inhibits drinking in rats sub- jected to dipsogenic treatments [7-9]. Although the conditions were different for the experiments in the rats from those with quail, these observa- tions present an important problem with respect to thirst mechanisms in terms of comparative physiol- ogy and neuroendocrinology. In the Japanese quail, the minimum effective dose of pBNP was 1 to 5 ug (+0.3 to 1.7 nmol) for 1.p. administration and 0.1 to 0.3 ug (=0.03 to 0.1 nmol) for i.c.v. administration. Okawara et al. [10] reported that a-hANP, with a structure similar to BNP, stimulated water intake at the dose of 1 nmol i.p. and at the dose of 0.03 nmol i.c.v.. These results indicate that the dipsogenic effect of pBNP is almost similar to that of an equimolar dose of a-hANP. Samson et al. [13] reported that i.c.v. adminis- tration of ET-3 (30 to 60 ng) inhibited water intake in rats subjected to dipsogenic treatments, such as injection of ANG II, hyperosmotic challenge and dehydration. These effects appeared within 15 min after the injection. There were no behavioral changes after the injection (30 to 60 ng ET-3). In the present study, however, 1.c.v. injection of 10 ng ET-1 had a dipsogenic effect 3hr later in water-replete Japanese quail. These differences might be explained by the differences in species and in the peptides used, namely, ET-1 in the quail and ET-3 in the rat. Furthermore, i.p. administra- tion of ET-1, even at 3 ug, had no effect on water intake, while i.c.v. injection had a stimulatory effect at 10 ng of ET-1. These observations sug- gest that, in the Japanese quail, ET may be degraded rapidly in the blood and, furthermore, that ET injected into the brain may secondarily stimulate water intake, since an increase in cumulative water intake appeared 3 hr after the injection. In avian species, more than ten kinds of peptide are known to be dipsogenic (ANG II, parathyroid hormone, a-hANP, and urotensin II in the quail, and ANG II, eledoisin, physalaemin, bombesin, and substance P in the pigeon) or antidipsogenic (substance P, leucine-enkephalin, thyrotropin- releasing hormone, arginine vasotocin, and soma- tostatin in the quail) [18, 19, 20]. Among these peptides, only ANG II has been investigated in detail in terms of mechanisms that affect water intake, and it has become clear that ANG II is involved in physiological mechanisms associated with drinking in birds and mammals [16]. In the present studies, pBNP and ET-1 were dipsogenic in the Japanese quail. However, it is not known whether they are physiologically involved in thirst mechanisms. Further investigations are needed to clarify the problem not only with regard to pBNP and ET-1, but also with regard to the other pep- tides mentioned above. 624 ACKNOWLEDGMENTS This investigation was supported in part by a Grant-in- Aid for Scientific Research from the Ministry of Educa- tion, Science and Culture of Japan (no. 02640587) to Prof. H. Uemura. 10 REFERENCES Sudoh, T., Kangawa, K., Minamino, N. and Mat- suo, H. (1988) A new natriuretic peptide in porcine brain. Nature, 332: 78-81. Kojima, M., Minamino, N., Kangawa, K. and Mat- suo, H. (1989) Cloning and sequence analysis of cDNA encoding a precursor for rat brain natriuretic peptide. Biochem. Biophys. Res. Commun., 159: 1420-1426. Itoh, H., Nakao, K., Yamada, T., Shirakami, G., Kangawa, K., Minamino, N., Matsuo, H. and Im- ura, H. (1988) Antidipsogenic action of a novel peptide, ‘brain natriuretic peptide’, in rats. Euro- pean J. Pharmacol., 150: 193-196. Fregoneze, J. B., Gutkowska, J., St.-Pierre, S. and Antunes-Rodrigues, J. (1989). Effect of brain nat- riuretic peptide on dehydration- or angiotensin II- induced water intake in rats. Brazilian J. Med. Biol. Res., 22: 765-768. Uehara, Y., Shimizu, H., Shimomura, Y., Kobayashi, I. and Kobayashi, S. (1990) Rat brain natriuretic peptide (BNP) reduces water intake fol- lowing dehydration. Neuropeptides, 17: 107-110. Antunes-Rodrigues, J., McCann, S. M., Rogers, L. C. and Samson, W. K. (1985) Atrial natriuretic factor inhibits dehydration- and angiotensin II- induced water intake in the conscious, unrestrained rat. Proc. Natl. Acad. Sci. USA, 82: 8720-8723. Itoh, H., Nakao, K., Morii, N., Sugawara, A., Yamada, T., Shiono, S., Saito, Y., Mukoyama, M., Arai, H., Sakamoto, M. and Imura, H. (1987) Central actions of atrial natriuretic polypeptides in spontaneously hypertensive and normotensive rats. jon, Cre, Veg Sie IAS INS), Nakamura, M., Katsuura, G., Nakao, K. and Im- ura, H. (1985) Antidipsogenic action of a-human atrial natriuretic polypeptide administered in- tracerebroventricularly in rats. Neurosci. Lett., 58: 1-6. Katsuura, G., Nakamura, M., Inouye, K., Kono, M., Nakao, K. and Imura, H. (1986) Regulatory role of atrial natriuretic polypeptide in water drink- ing in rats. European J. Pharmacol., 121: 285-287. Okawara, Y., Seki, K. and Kobayashi, H. (1986) Biologically active peptides and water intake II. Increases of water intake and excretion following injection of atrial natriuretic peptide in the Japanese 11 wy 13 14 15 16 7) 18 19 20 Y. TEzuKA, H. KoBayAsHI AND H. UEMuRA quail, Coturnix coturnix japonica. Zool. Sci., 3: 831-836. Yanagisawa, M., Inoue, A., Ishikawa, T., Kasuya, Y., Kimura, S., Kumagaye, S.-I., Nakajima, K., Watanabe, T. X., Sakakibara, S., Goto, K. and Masaki, T. (1988) Primary structure, synthesis, and biological activity of rat endothelin, an endothelium- derived vasoconstrictor peptide. Proc. Natl. Acad. Sci. USA, 85: 6964-6967. Masaki, T. (1991) Tissue specificity of the endothe- lin-induced responses. J. Cardiovasc. Pharmacol., 17 (Suppl. 7): S1-S4. Samson, W. K., Skala, K., Huang, S. F.-L., Gluntz, S., Alexander, B. and Gémez-Sanchez, C. E. (1991) Central nervous system action of endothelin-3 to inhibit water drinking in the rat. Brain Res., 539: 347-351. Kasuya, Y., Karakida, T., Okawara, Y., Yama- guchi, K. and Kobayashi, H. (1985) Daily drinking patterns and plasma angiotensin II in the budgerigar (Melopsittacus undulatus) and the Japanese quail (Coturnix coturnix japonica). J. Yamashina Inst. Ornith., 17: 32-43. Uemura, H. and Kobayashi, H. (1971) Effects of dopamine implanted in the median eminence on the estrous cycle of the rat. Endocrinol. Japon., 18: 91- 100. Kobayashi, H. and Takei, Y. (1982) Mechanisms for induction of drinking with special reference to - angiotensin II. Comp. Biochem. Physiol., 71A: 485- 494. Squadrito, F., Frisina, N., Buemi, M., Sturniolo, R., Autolitano, A., Magri’, V., Squadrito, G. and Caputi, A. P. (1989) A comparison of synthetic human and rat ANP administered intracereb- roventricularly in freely moving normotensive and hypertensive rats. J. Cardiovasc. Pharmacol., 13 (Suppl. 6): S27-S30. De Caro, G., Perfumi, M. and Massi, M. (1988) Tachykinins and body fluid regulation. Prog. Psychobiol. Physiol. Psychol.,.13: 31-66. Takei, Y. and Kobayashi, H. (1990) Hormonal regulation of water and sodium intake in birds. In “Endocrinology of Birds: Molecular to Behavioral”. Ed. by M. Wada, S. Ishii and C. G. Scanes, Japan Sci. Soc. Press, Tokyo/Springer-Verlag, Berlin, pp. 171-183. Kobayashi, H., Okawara, Y., Uemura, H. and Kasuya, Y. (1991) Effects of biologically active peptides on drinking behavior in the Japanese quail, Coturnix coturnix japonica. In “Current Themes in Comparative Endocrinology”. Ed. by R. N. Saxena, K. Muralidhar, L. Bhagat, N. Sehgal, T. Saxena and P. Kaushal, Delhi University Press, Delhi, p. 160. ZOOLOGICAL SCIENCE 9: 625-632 (1992) Effect of Brain on Proliferative Activity of Adenohypophysial | Primordial Cells in vitro MANABU SHIRAI and YUICHI G. WATANABE Department of Biology, Faculty of Science, Niigata University, Niigata 950-21, Japan ABSTRACT—The pattern of cell proliferation in the fetal rat adenohypophysial primordium was investigated both in vivo and in vitro with special reference to the diencephalic floor. Proliferating cells were labelled by bromodeoxyuridine (BrDU) followed by its immunohistochemical detection with a monoclonal antibody. In fetal rats of 12.5 days of age, BrDU-labelled cells were distributed almost evenly throughout the adenohypophysial primordium. On days 13.5 and 14.5, on the other hand, labelling was mostly confined to the dorsal wall of the adenohypophysial primordium that was in contact with the diencephalic floor. To examine the in vitro effect of the developing diencephalic floor on cell proliferation of the adenohypophysial primordium, Rathke’s pouches were isolated from fetal rats on days 12.5 and 13.5 of gestation and kept in organ culture for 1-2 days with or without the brain. The incidence of BrDU-labelled cells was markedly high if the diencephalic floor was left intact. Moreover, proliferating adenohypophysial primordial cells were concentrated in the region adjacent to brain tissue. In the absence of brain, only a small number of cells were labelled. Such a low incidence of labelling was also observed when brain was replaced by liver. From these results we conclude that the developing diencephalon is essential for proliferation of adenohypophysial primordial cells. It remains to be settled © 1992 Zoological Society of Japan if the adoral part of primordial cells can also respond to this neural agent. INTRODUCTION The hypophysis consists of two different compo- nents, i.e., neuro- and adenohypophysis. From the early stage of development, the adenohy- pophysial primordium makes close contact with the diencephalic floor. Such a close relationship between the two tissues has lead many investiga- tors to assume that the differentiation of the adenohypophysis is under the influence of the brain. In amphibians, the pars intermedia fails to form after removal of the posterior hypothalamus [1-4]. The reports are contradictory as to whether brain is essential for the development of the pars distalis: some workers accept the inductive in- fluence of the brain [5, 6], whereas others believe that neural tissue plays little role [1, 3, 4]. In mammals, we have previously shown the involve- ment of the diencephalic floor in the cytodif- ferentiation of the pars distalis in vitro [7-9]. Accepted March 2, 1992 Received November 21, 1991 Moreover, cell proliferation of adenohypophysial primordium also appeared to be stimulated by brain since its removal caused a marked reduction in the size of culture explants. Daikoku et al. [10] have provided quantitative data that the size of the adenohypophysial primordium markedly increased when it was co-cultivated with brain. Jn vivo, mitotic figures are more frequently observed in the dorsal region of the adenohypophysial primordium that was in contact with the diencephalic floor [11, 12]. Recently, Ikeda and Yoshimoto [13] have studied proliferative activity of the fetal rat hypophysis by use of bromodeoxyuridine (BrDU). According to their report, BrDU-labelled cells were concentrated in the dorsal portion of the adenohypophysial primordium that faces the de- veloping neural lobe. To date, however, there is little direct information on the role of the neural element. In this study we examined the in vitro effect of diencephalic floor on cell proliferation of the adenohypophysial primordium. In addition, we re-investigated the pattern of cell proliferation in the rat adenohypophysial primordium in vivo. 626 M. SHIRAI AND Y. G. WATANABE MATERIALS AND METHODS Adult rats of the Sprague-Dawley strain were mated at night. If spermatozoa were found in the vaginal smears the next morning, the noon of that day was designated at day 0.5 of gestation. In vivo experiments Pregnant rats from day 12.5 to 16.5 of gestation received an intraperitoneal injection of bromo- deoxyuridine solution (30mg/kg, Amersham, UK). Three hours after injection, mothers were anesthetized with ketamine hydrochloride and fetuses were removed for fixation. Heads were immersed in Bouin’s solution and trimmed under a dissecting microscope to obtain good penetration of fixative into the adenohypophysial primordium. In vitro experiments Pregnant rats of days 12.5 and 13.5 were anes- thetized and fetuses were removed one by one by Caesarian section. The hypophysial primordium was separated in Ca- and Mg-free Hanks solution as described previously [14]. In some primordia mesenchymal tissue was removed as much as possi- ble with the diencephalic floor left intact. These were cultivated so as to adenohypophysial and neural tissue as well as their boundary were clearly OD x Fic. 1. Diagram showing how the adenohypophysial primordium was placed on a piece of cellulose acetate membrane. The primordium was oriented with its anterior wall (A) at the bottom. Asterisk indicates a mark which was made in order to facili- tate the identification of brain tissue at the time of fixation. Fic. 2. Schematic drawing showing how a section of explant was compartmentalized for quantitative analysis. Lines were drawn at 50 um intervals. shown during the subsequent histological pro- cesses (Fig. 1). In other primordia the diencephalic floor and mesenchyme were removed by the use of 0.15% collagenase (Sigma, type V) in Ca- and Mg-free Hanks solution. During this period, the upper half of the primordium was in contact with the developing brain, whereas its lower half was surrounded by a rich amount of mesenchyme. A small number of mesenchymal cells were also observed between the adenohypophysial primor- ° dium and future infundibulum. Removal of the latter mesenchyme was impossible without sepa- rating the brain. Those primordia from which only mesenchyme was removed in the enzyme solution served as a control. After enzymatic removal of the brain from a few adenohypophysial primordia, they were combined with a piece of the hepatic rudiment. In this study enzyme treatment was completed as quickly as possible, generally within 10min. Tissues were then placed on pieces of cellulose acetate membrane and maintained in Falcon dishes (no. 3037) for organ culture. The culture medium was e-MEM to which 30mM glucose was added. Fetal bovine serum (Gibco) was added at a concentration of 0.1%; this low concentration was employed to minimize the effects of growth factors, if any, contained in the blood serum. From | to 2 days after cultivation, BrDU solution was added at a dose of 6 ng/ml medium. Explants were fixed in Bouin’s solution 3 hr later. After overnight fixation, they were embedded in paraffin and cut at 2 ~m using glass knives. Deparaffinized sections were first incu- Cell Proliferation of Adenohypophysis 627 bated with monoclonal antibody to BrDU (Amer- sham) for 60 min and then with peroxidase- conjugated anti-mouse IgG. The reaction product was visualized with 3,3 -diaminobenzidine solution containing HjO>. In all explatns, different levels of sections at 8 um intervals were mouted on a slide and stained. After confirming that the pat- tern of BrDU-labelling was consistent irrespective of the section levels, a section containing a maxim- al number of labelled cells was selected for quan- titative measurement. In those cultures where the brain was co-cultivated, lines were drawn on each photomicrograph of adenohypophysial tissue at 50 ym intervals along the boundary of the attached brain (see, Fig. 2). The entire area of adenohy- pophysial tissue was measured by use of an image analyzer (IBAS-2000, Germany) whereas the area of each compartment was calculated by the paper weight method. Then the respective areas of the compartments were calculated based on their weight rate. In case of explants without the brain, only the entire area was measured. The frequency of labelled cells was expressed as the number of labelled cells per area. RESULTS In vivo observations On day 12.5 of gestation, the epithelial walls of the adenohypophysial primordium were found to contain many BrDU-labelled cells. There was no topographical difference in the distribution of labelled cells particularly in terms of the presump- tive neural lobe (Fig. 3a). On the other hand, a marked regional difference in labelling was observed in the adenohypophysial primordium on days 13.5 and 14.5. During this period, labelling was mostly confined to the upper half of the adenohypophysial tissue that faced the neural ele- ment (Fig. 3b). On day 16.5 BrDU-labelled cells were sparsely distributed throughout the adenohy- pophysis (Fig. 3c). In vitro experiments When maintained in organ culture for 1-2 days, the Rathke’s lumen was found to be narrowed to varying degrees. In some cases, the lumen was obliterated completely. Identification of brain tissue was easy because great care was taken as to the orientation of explants throughout the course of experiment as already described in Materials and Methods. Most explants had no typical mass of mesenchymal tissue. Some cultures, however, were found to possess a small amount of mesen- chyme at a confined area usually toward the oppo- site side of brain tissue (Fig. 4d). In explants that were separated on day 12.5 and cultivated for one day with the diencephalic floor, adenohypophysial tissue was crowded with many BrDU-labelled cells. The majority of labelled cells were included in the half of explant to which brain tissue attached (Fig. 4a). Fig. Sa shows the result of quantitative data on the distribution of prolifer- Fic. 3. Photomicrographs showing BrDU-labelled proliferating cells in the pituitary primordia of fetal rats. Asterisk indicates the presumptive neural lobe. 105. a. On day 12.5 labelled cells are observed homogeneously. b. On day 14.5 only a few cells are labelled in the lower half of the adenohypophysial primordium. Labelling is completely lacking in the lateral lobe (arrow). c. On day 16.5 labelled cells are distributed homogeneously. 628 M. SHIRAI AND Y. G. WATANABE Fic. 4. Photomicrographs showing BrDU-labelled cells in culture explants of adenohypophysis removed on days 12.5 (a, b, c, e, f) and 13.5 (d) of gestation. Asterisk indicates co-cultivated brain. 113. a. One-day explant. Labelled cells are sparsely distributed toward the opposite side of the brain. b. Two-day explant. The distribution pattern of labelled cells is essentially similar to that in a. c. One-day explant treated with collagenase before culture. d. One-day explant. Labelled cells are seen toward the brain. e. One-day explant without the brain. f. One-day explant explant co-cultivated with liver tissue after removal of the brain. ating cells with special reference to brain tissue. The density of proliferating cells was highest to- ward the brain. After removal of the diencephalic floor, the number of BrDU-labelled cells was markedly re- duced (Figs. 4e, 6a). The distribution of labelled cells was inconsistent. This was also the case if the brain was replaced by liver tissue (Figs. 4f, 6a). Collagenase treatment did not affect proliferative activity of the adenohypophysial primordium (Figs. 4c, 5a). The overall incidence of labelling became lower after 2 days in culture (Figs. 5a, 6a). Owing to such lowered activity of cell proliferation, the Cell Proliferation of Adenohypophysis 629 80 al 80 b aE £ a Oo < 60 60 L rT) (3) ro) £ G 40 40 a NS ~ oO Control z A Estradiol uw @ Testosterone 0 1 2 DURATION OF TREATMENT (weeks) Fic. 2. Effects of immersion in 100 ng/ml estradiol or testosterone on the length of the second fin ray. Each point represent the means of 20-30 measure- ments. Vertical bars indicate the standard errors of the means. *, Significantly different (P<0.01) from the control. Gonadal Steroids Inhibit Flounder Metamorphosis 635 rays, and a significant effect became apparent starting 3 days after culture. Treatment of the fine rays simultaneously with T3 and estradiol did not cause significant shortening even after 7 days, indicating that estradiol blocked the effects of T3. Testosterone likewise inhibited the effects of T3. When testosterone (100 ng/ml) was added to the medium together with T3, the stimulatory effect of T3 was only observed on day 3. Figures 2—4 show the changes in fin ray length, CE2T 100 a 25 2} We ire ro) jag Lu > = 50 = ,), and sex steroid hormones (progesterone, androgens, and 17/-estradiol) were determined in the female crested newt, Triturus carnifex, during the annual reproductive cycle. /n vivo experiments were carried out to study the effects of PGE, and PGF;, on plasma sex steroid hormones during prereproduction, reproduction, and postreproduction; simultaneously, in vitro experiments were performed to study the effects of these two prostaglandins on sex steroid hormones ovarian release. The effects of one week’s captivity on PGE>, PGF;, and sex steroid hormones were also evaluated. The PGE; plasma level was low from October to December, then it rapidly increased to peak in March, after which it soon fell to reach its minimum value in May. Plasma PGF>, and sex steroid hormones showed similar trends to those found in previous studies. From December to April, the PGE, plasma values were negatively correlated to those of PGF>, and positively to those of androgens; PGF>, plasma values were positively correlated to those of estradiol. PGE, in vivo treatment increased plasma progesterone and decreased 17/-estradiol in April, while PGF;, induced the opposite effects in the same month; PGE; increased androgens in January and March, and PGF;, increased androgens in April. The in vitro experiments were in agreement. These results suggest that PGE, and PGF,, play opposite role(s) in the reproductive processes of the female Triturus carnifex. PGE, could be involved in the reproduction processes through androgen secretion, © 1992 Zoological Society of Japan while PGF,, in the ending of reproduction through estradiol increase. INTRODUCTION In mammals, prostaglandins (PGs) of both F and E series intervene in the regulation of the ovary functions, including steroidogenesis [1-4]. PGs are involved in ovulation of the chicken follicle [5] and in inducing the relaxation of the avian oviduct [6]. PGs are also implied in the regulation of reproduction in reptiles [7-10]. There is little information, however, on the role of PGs in amphibian reproductive processes. In the anuran, Rana esculenta [11, 12], and in the urodele, Triturus carnifex {13, 14], plasma prosta- glandin F5, (PGF>,) levels change in relation to the various periods of the annual reproductive cycle, and in both species PGF), could be implied in breeding period termination. A recent in vitro study, carried out on the male Accepted March 17, 1992 Received January 22, 1992 Triturus carnifex abdominal gland, suggested for PGE), a role in inducing pheromonal activity [15]; these data indicated an involvement of this pro- staglandin in reproductive processes. It seemed, therefore, of interest to determine the annual profile of PGEs, in relation to that of PGF>,, and of sex steroid hormones in the female crested newt, Triturus carnifex, and, in addition, to evaluate the PGE, and PGF, effects on plasma levels and ovarian output of sex steroid hormones. MATERIALS AND METHODS Animals The reproductive cycle of the Triturus carnifex population comprises two major phases: during the summer period the animals disappear under- ground (August-September), during the other months the newts are in the pond and undertake reproduction during the cold months (January to 640 A. GosBeTTI, M. ZERANI AND V. Botte March). Adult female newts, Triturus carnifex (Laur.), were collected (15 each month) in a small moun- tain pond (Colfiorito, Umbria, Italy; 870 m above sea level), from October to July during 1990/91. In that place, newts hide on land during August and September and therefore, in these two months, it was not possible to obtain enough animals for our study. At once, after capture, newts were anesthetized in the field with 3- aminobenzoic acid ethyl ester (Tricaine, Sigma, USA) and bled through a heparinized microtube inserted in the heart. Individual blood samples were collected into chilled tubes containing acetyl- salicylic acid (Aspirin, Sigma) and EDTA (5 pg and 7 ug/ml of blood, respectively) [11]. After centrifugation, plasma samples were kept at —20°C until use. Captivity In December (prereproduction), January (be- ginning of reproduction), March (ending of repro- duction), and April (postreproduction), 15 female newts were captured and bled in the field as reported above (control I), while simultaneously 15 other animals were transferred to our labora- tory aquaria, kept under natural photothermal conditions and fed on larvae of Chironomidae, Tubifex and Daphnia, ad libitum to study the effects of captivity on plasma hormone levels. A week later, the animals in captivity were bled, and simultaneously yet another 15 were sacrificed in the field (control II). In vivo experiment In December, January, March, and April, 84 female newts (for each month) were captured, transferred to our laboratory and kept under nat- ural photothermal conditions. One week later, the animals were divided into 4 groups: a) 21 newts received a single s.c. injection of 350 ng PGE, (Sigma; ca 35 ng/g body weight) dissolved in 100 yl amphibian saline (0.64% w/v NaCl solution); b) 21 newts received a single s.c. injection of 300 ng PGF,, (Sigma; ca 30 ng/g body weight) dissolved in 100 wl amphibian saline; c) 21 newts received a single s.c. injection of 100 ul amphibian saline only; d) 21 newts were untreated. Seven of the 21 animals of each of the groups were bled, as de- seribed above, 6, 24, and 48 hr after treatment. After centrifugation, plasma samples were kept at —20°C until use. A further batch of seven un- treated animals were bled at the beginning of the experiments (controls, time 0). The times of treatment and the minimum effective doses of PGE, and PGF,, were chosen after preliminary tests (data not shown). In vitro experiment The method used for the in vitro experiment followed Gobbetti and Zerani [15, 16]. In Decem- ber, January, March, and April, 7 female newts (for each month) were captured, transferred to laboratory aquaria, and maintained as reported above. One week later, the animals were decapi- tated, the ovaries rapidly removed and placed in cold Dulbecco’s modified Eagle medium (DME; Sigma) containing 10 mM Hepes, 1 mg Penicillin G/ml, and 2mg streptomycin/ml. For each month, the ovary from one animal was divided into equally sized fragments, pooled and equally distri- buted over 12 incubation wells. Multiwell tissue culture plates (Becton Dickinson, USA) were util- ’ ized. Each incubation set of wells was divided into 3 experimental groups (each consisting of 4 wells). The experimental groups were: a) ovarian tissue incubated with DME alone; b) ovarian tissue incubated with DME plus PGE, (50 ng); c) ova- rian tissue incubated with DME plus PGF,, (50 ng). The final volume of each well was 1 ml. Culture plates were wrapped with aluminium foil and incubated in a shaking water bath (19°C), set at 30 revolutions/min. One well of each ex- perimental group was removed, respectively, after 30, 60, 120 and 240 min of incubation. The incuba- tion medium samples were immediately stored at —20°C for later hormone determination; ovarian tissues were soon homogenized in amphibian saline, and protein content was determined by the Protein Assay method (Bio-Rad, USA). In addi- tion, the whole experiment was repeated with incubation sets without ovarian tissue. The experi- ment was replicated 7 times for each month. Preliminary evidence led to our choice for the incubation conditions and the minimum effective doses of PGE, and PGF>,, utilized in the present in PGE, and Reproduction in 7. carnifex 641 vitro study (data not shown). PGE>, PGF>,, progesterone, androgens, and 17/:- estradiol determination PGE, was determined following the radioimmu- nological method (RIA) for plasma [13, 14] and incubation media [15], PGF,, determinations used for this species are here briefly described. PGE, determinations were carried out on duplicate plas- ma samples (100 1) and incubation media (500 pl) that were extracted with 10 volumes of diethyl ether for 5 min. The mixtures were briefly centri- fuged, organic fractions were transferred into glass tubes and evaporated to dryness under a nitrogen stream. The extracts were resuspended with 100 wl of assay buffer and assayed. ‘The recovery of added labelled PGE) was 85.9+0.84%. The para- llelism among the standard curve in buffer, a standard curve in incubation medium (then ex- tracted), and a serial dilution of a single incubation medium sample (extracted) were constant. The progesterone, androgens, and 17/-estradiol content of the plasma and incubation media was determined by RIA according to the previously reported methods [13, 15, 17]. The following sensitivities were recorded: PGE), 18 pg (intra-assay variability: 6.5%; inter- assay variability: 12%); PGF>,, 17 pg (intra-assay variability: 5%; inter-assay variability: 11.5%); progesterone, 7.0 pg (intra-assay variability: 5%; inter-assay variability: 11.5%); androgens, 9.0 pg (intra-assay variability: 6%; inter-assay variability: 9%); 17@-estradiol, 8.5 pg (intra-assay variability: 4%; inter-assay variability: 9%). The PGF,,, progesterone, testosterone, and 17/-estradiol anti- sera were provided by Dr. G. F. Bolelli and Dr. F. Franceschetti (CNR-Physiopathology of Repro- duction Service, University of Bologna, Italy), the PGE, antiserum was purchased from Cayman Chemical (USA). Testosterone was not separated from 5a-dihydrotestosterone and therefore, since the antiserum used is not specific, the data are expressed as androgens. Tritiated PGE>, PGF>,, progesterone, testosterone, and 17/-estradiol were purchased from Amersham International (Eng- land), non-radioactive PGE, progesterone, tes- tosterone, and 17/-estradiol from Sigma. Statistics Data relative to each hormone were submitted to analysis of variance (ANOVA) followed by Duncan’s multiple range test [18, 19]. Correlation coefficents followed Scossiroli and Palenzona [20]. RESULTS Hormone annual reproductive cycles The PGE, plasma level was low from October to December, then it rapidly increased to peak in March (P<0.01 vs all months) and soon fell to reach its minimum value in May (P<0.01 vs January, February, March, April, June, and July), then PGE, level increased during the summer and peaked in July (P<0.01 vs all months except February) (Fig. 1). The PGF , plasma level in- 4000 PGE> 3000 2000 1000 ) 5000 PGF. 3750 2500 a Ab, ne 1250 = ™ ie) a 0 zs 3000) PROGESTERONE =] Lu > Lu SJ} < > 5000) ANDROGENS < 3750 —| Re 2ake 1250 0 12000) 175-ESTRADIOL 9000 6000 3000 0 Oct Nou Bee | Jan Feb Mar ee ee Jun Jul PREREPRODUCT ION REPRODUCTION POSTREPRODUCT ION Fic. 1. Plasma levels of PGE;, PGF;,, and sex steroid hormones during the annual reproductive cycle in female crested newt, Triturus carnifex. Each mean refers to 15 determinations+S.D.. 642 A. GosBETTI, M. ZERANI AND V. BoTTE creased in autumn to peak in December (P<0.01 vs all months except April), decreased from Janu- ary to March and peaked again in April (P<0.01 vs all months except December), and was low from May to July (Fig. 1). The progesterone plasma level exhibited minor changes, reaching the mini- mum value in April (P<0.01 vs all months) (Fig. 1). The plasma androgens increased from October to December and more from January to March (P <0.01 vs October, November, December, April, May, June, and July), they started to drop in April and reached the minimum value in July (P<0.01 vs all months except June) (Fig. 1). The 17/- 4400 PGE, 3300 2200 1100 ANDROGENS FPILASIMGA IkEWELS aoc” ian i) December March January April Fic. 2. Effects of one week’s captivity on PGE, PGF, and androgens plasma levels in female crested newt, Triturus carnifex, during December (prereproduc- tion), January (reproduction beginning), March (re- production ending), and April (postreproduction). Experimental groups: (C1): newts captured in the field and bled at once (control I); (MM): newts bled after one week’s captivity in laboratory; (@): newts captured and bled at once in the field, one week after control I (control II). Each mean refers to 15 determinations+S.D.. * P<0.01 vs control I and II (Duncan’s multiple range test). estradiol plasma level peaked in December (P< 0.01 vs all months except April, May, and July) and in April (P<0.01 vs all months) (Fig. 1). From December to April the following correla- tions were found: PGE, plasma values were nega- tively correlated to those of PGF), (r= —0.523; df =73; P<0.001) and estradiol (r= —0.469; df=73; P<0.001), and positively correlated to those of androgens (r=0.501; df=73; P<0.001), PGF), plasma values were negatively correlated to those of androgens (r= —0.545; df=73; P<0.001), and positively to those of estradiol (r=0.434; df=73; P <0.001), androgens plasma values were negatively correlated to those of estradiol (r= —0.408; df= 133 Je, injection on progesterone and 17/-estradiol plasma levels in female crested newt, TJvriturus carnifex, during April (postreproduction). | Experimental groups: (MM): untreated newts; (11): amphibian- saline-only injected newts; (@): PGE, injected newts; (4): PGF>, injected newts. Each mean refers to 7 determinations+S.D.. * P<0.01 vs untreated and amphibian-saline-only injected newts (Duncan’s multiple range test). PGE, and Reproduction in T. carnifex In vivo experiment PGE, treatment increased progesterone level in April at 6 and 24 hr (P<0.01) (Fig. 3), androgens in January and March at 6, 24 and 48 hr (P<0.01) (Fig. 4), and decreased 17/-estradiol in April at 24 and 48 hr (P<0.01) (Fig. 3). PGF>, treatment decreased progesterone level in April at 6, 24 and 48hr (P<0.01) (Fig. 3), increased androgens in April at 24 and 48 hr (P< 0.01) (Fig. 4), and 17(-estradiol in April at 24 and 48 hr (P<0.01) (Fig. 3). 7000 7 DECEMBER = ce * Le) ‘ Y Q Lo = SNe (©) Qo YO ea Lu a) () fad = z <— 7000 7 APRIL 5250 3500 * * | eZ cw oy : ) 24 48 (controls) HOURS Fic. 4. In vivo effects of 350 ng PGE; or 300 ng PGF3, injection on androgens plasma levels in female crested newt, Triturus carnifex, during December (prereproduction), January (reproduction begin- ning), March (reproduction ending), and April (postreproduction). Experimental groups: (M): un- treated newts; (C1): amphibian-saline-only injected newts; (Z): PGE; injected newts; (@): PGF>, in- jected newts. Each mean refers to 7 determinations +$.D.. * P<0.01 vs untreated and amphibian- saline-only injected newts (Duncan’s multiple range test). 643 In vitro experiment PGE, basal release was higher in January and March (P<0.01) compared with December and April; March values were higher (P<0.01) than those of January (Fig. 5). PGF, basal release was higher in April (P<0.01) compared with the other months (Fig. 5). Progesterone basal release was lower in April (P<0.01) compared with the other months (Fig. 6). Basal release of androgens was higher in January and March (P<0.01) compared with the other months (Fig. 7). Estradiol basal level was higher in December and April (P<0.01); April values were higher (P<0.01) than those of December (Fig. 8). At all incubation times, the 3000 7 DECEMBER 2250 1500 750 ee ee JANUARY (Keehn) fir ote 1h) a a Jim a MARCH 20 (PGF ho ho ul o a,b 1500 a,b S i = 3S c (> 30007 APRIL = 2250 a c 1500 Cc te BRA spe 0) Sel 30 60 120 240 NGO AM Te OMN etal ee Gmakigmne Fic. 5. In vitro basal release of PGE, (™) and PGF,, (C1) from ovary of female crested newt, Triturus carnifex, incubated in December (prereproduction), January (reproduction beginning), March (repro- duction ending), and April (postreproduction). Each mean refers to 7 determinations +S.D. a, P< 0.01 vs same time December and April; b, P<0.01 vs same time January; c, P<0.01 vs same time December, January, and March (Duncan’s multiple renge test). PROGES TER CONE (pic indap omernnn) Fic. DECEMBER WY, anwbto LL x os eo MARCH YY APRIL 120 INCUBATION TIME eee 6. In vitro effects of 50 ng PGE; or 50 ng PGE,, on progesterone release from ovary of female crested newt, Triturus carnifex, incubated in December (prereproduction), January (reproduction begin- ning), March (reproduction ending), and April (postreproduction). Experimental groups: (™): ova- ry incubated with medium alone; (1): ovary incu- bated with PGE,; (&): ovary incubated with PGF,,. Each mean refers to 7 determinations +S.D.. a, P< 0.01 vs same time December, January, and March; * P<0.01 vs same time medium-alone (Duncan’s mul- tiple range test). TABLE 1. A. Gossetti, M. ZERANI ANDROGENS ( pg/mg protein) Fic. AND V. BoTTE 1000 7 DECEMBER 750 500 250 Y oe ey |G 1000 7 JANUARY 750 500 : | a ay 1000 7 MARCH 750 500 “len a | 7 1000 7 APRIL 750 i OO 250 0 =a 120 Wopsieeee Tl ine Lares 7. Invitro effects of 50 ng PGE; or 50 ng PGF>, on androgens release from ovary of female crested newt, Triturus carnifex, incubated in December (prereproduction), January (reproduction begin- ning), March (reproduction ending), and April (postreproduction). Experimental groups: (™): ova- ry incubated with medium alone; ((): ovary incu- bated with PGE,; (@): ovary incubated with PGF,,. Each mean refers to 7 determinations+S.D.. a, P< 0.01 vs same time December and April; * P<0.01 vs same time medium-alone (Duncan’s multiple range test). Correlation coefficients among PGE,, PGF>,, androgens (A), and 17/-estradiol (E) levels released by female Triturus carnifex ovary incubated at various times during December, January, March, and April Incubation times 30 min 60 min 120 min 240 min PGE, vs PGF2, —0.768 —().657 — 0.743 =i lid)! PGF, vs A 0.724 0.745 0.774 0.761 RGED evs & —0.650 =((),037) —(0.672 —().621 PGF,, vs A —0.733 =()). 758 —0.731 —0.749 PGF,, vs E 0.715 0.748 0.686 0.668 Avs E —0.620 =(H70 — 0.646 —0.645 All correlations show the same level of significance (P<0.001). df=26. PGE, and Reproduction in T. carnifex 645 DECEMBER a . a ay, WZ ae ey JANUARY MARCH Ss oo SY) Oo nw oo Lb © (S-= 0 520 O* "Ol Oo 1 O [Ln SL Lt tL (Loe tn tn Ls dt LC ax Wl 2400 7 APRIL 1800 it 1200 * a,b a,b °° a i | ive = 0 30 60 120 INCUBATION TIME et Fic. 8. In vitro effects of 50 ng PGE; or 50 ng PGF;, on 17£-estradiol release from ovary of female crested newt, Triturus carnifex, incubated in December (prereproduction), January (reproduction begin- ning), March (reproduction ending), and April (postreproduction). Experimental grousp: (MM): ova- ry incubated with medium alone; (1): ovary incu- bated with PGE;; (@): ovary incubated with PGF,,. Each mean refers to 7 determinations+S.D.. a, P< 0.01 vs same time January and March; b, P<0.01 vs same time December; * P<0.01 vs same time medium-alone (Duncan’s multiple range test). Ge 1S ey EN DI T@ME UC Yor tel ee [antic] foi tereine te iT) | PGE; values were negatively correlated to those of PGF,, (P<0.01) and estradiol (P<0.001), and positively correlated to those of androgens (P< 0.001); PGF, values were negatively correlated to those of androgens (P<0.001), and positively to those of estradiol (P<0.001); androgens values were negatively correlated to those of estradiol (P <0.001) (Tab. 1). PGE, treatment increased progesterone release in April at 30, 60 and 120 min (P<0.01) (Fig. 6), androgens in January and March at 60, 120 and 240 min (P<0.01) (Fig. 7), and decreased estradiol in April at 60, 120 and 240 min (P<0.01) (Fig. 8). PGF, treatment decreased progesterone levels in April at 30, 60 and 120 min (P<0.01) (Fig. 6), increased androgens in April at 60 and 120 min (P <0.01) (Fig. 7), and estradiol in April at 60, 120 and 240 min (P<0.01) (Fig. 8). DISCUSSION The mountain population of female Triturus carnifex, utilized in this study, shows a discon- tinuous reproductive cycle similar to those de- scribed in several other newts living in temperate zones [21-23]. In this female newt, ovulations and egg depositions occur from January to the end of March, the oogenetic cycle in spring and summer, and vitellogenesis in autumn and winter [24]. This is the first work to evidence in vivo and in vitro the presence of PGE, in female Triturus carnifex. The PGE, plasma titers showed peculiar changes during the newt annual reproductive cy- cle; during the reproductive period (January- March), this prostaglandin had high values which rapidly fell during postreproduction (April). The high plasma levels of PGE, during reproduction suggest the involvement of this prostaglandin in the breeding processes. PGF >,, progesterone, androgens and estradiol plasma levels showed similar annual trends to those reported in previous studies [14, 24]. Briefly, the April increase of plasma PGF;, was simultaneous with an estradiol rise and a progesterone drop. High plasma values of androgens characterized the reproductive period in female Triturus carnifex, confirming pre- vious studies carried out in the same species [25, 26] and in the same population [24], which estab- lished the role of androgens in inducing the repro- ductive processes was addressed in the female crested newt. In this context we recall that, in another amphibian species, Rana esculenta, the highest values of male and female brain androgen receptors were found during the reproductive period [27]. As regards estradiol plasma pattern, low levels were found in autumn and winter, when vitellogenesis occurs. It was well estrablished in amphibians that estradiol induces vitellogenin synthesis [28], therefore the low estradiol plasma values found during vitellogenesis are difficult to 646 A. GosBeTT!I, M. ZERANI AND V. BotTrTE explain, but, in this context, we recall that similar data were obtained in other amphibian species [29]. In accordance with the plasma values, the in vitro studies demonstrated that ovarian tissue re- leased high amounts of PGE, and androgens dur- ing reproduction and high levels of PGF,, and estradiol during postreproduction, while in this latter period progesterone was low. The in vivo and in vitro PGE, treatment in- creased androgens levels during reproduction sug- gesting a causal relationship between these two hormones. This hypothesis is also supported by the positive correlation between PGE, and androgens found in the plasma of the animals sacrificed in the field. The same positive correla- tion was also found between the basal values of PGE, and androgens obtained in the in vitro experiment. These results seem further to indicate that PGE, is involved in the reproductive proces- ses, by the regulation of the androgens synthesis, even if other mechanisms cannot be excluded. The involvement of this prostaglandin in reproduction is also suggested by previous studies which attri- bute to PGE, a role in inducing pheromonal activity by the abdominal gland of Triturus carnifex [15]. Also in another amphibian species, Xenopus laevis, PGE, was found to be involved in the breeding processes, since this prostaglandin in- duced the sexual behavior [30]. The in vivo and in vitro results indicated that PGE, treatment in animals captured in April in- duced opposite effects to those determined by PGF>, administered in the same month. In fact, PGE, in vivo and in vitro increased progesterone and decreased estradiol, while PGF>, decreased progesterone and increased estradiol during this period, in agreement with the plasmatic trends here and previously reported [14, 24]. The causal relationship which binds the PGF;, and estradiol plasma patterns supported the hypothesis that PGF,, contributed to reproductive period termina- tion. In fact, in several temperate-zone living amphibians and reptiles, a significant estradiol plasma rise occurred near the end of reproduction [31-33]. This steroid is believed responsible for the breeding interruption probably through a neg- ative feedbach mechanism at local [34] and central [35] levels. Taken together, these results suggest that PGE, and PGF,, play opposite role(s) in the reproduc- tive processes of the female Triturus carnifex. Summarizing, PGE, is involved in the reproduc- tion through androgens secretion, while PGF, in the ending of reproduction through estradiol in- crease. This possible antagonistic role between PGE, and PGF,, is supported by their negative correlation found in the plasma. This hypothesis is in agreement with previous studies which assign an inhibitory role to PGF, in sexual behavior in the reptiles, Anolis carolinensis and Thamnophis sirta- lis parietalis [36, 37] and in the amphibian, Bufo americanus [38], but a stimulatory role to PGE, in receptivity behavior in another amphibian, Xeno- pus laevis [30]. The antagonistic effect of PGE, and PGF, in the regulation of in vivo and in vitro progesterone release found during postreproduction is still un- known. We recall that a positive 3-hydroxy- steroid dehydrogenase response was reported in the postovulatory structures of Triturus carnifex [39], Rana esculenta [40] and Rana cyanophlyctis [41], but evidence of ultrastructural analysis is lacking [22]. Nevertheless, in mammals, PGE, ° counteracts the luteolytic effect of PGF,, and might be one of several factors prolonging the life span of the corpus luteum [42, 43]. Finally, this work confirms that captivity caused a decrease of circulating PGs and androgens. These results are in agreement with those reported for this [14] and other amphibian species [44—47]. ACKNOWLEDGMENTS The authors would like to thank James Burge, of the Camerino University Institute of Linguistics, for help with English, and Elio Marchegiani for his assistance. This work was supported by a grant to Prof. Alberta Polzonetti from Italian Ministry of Education and C. N. R. REFERENCES 1 Espey, L. 'L., Norris, ©. and Saphire aD =@.986) Effect of time and dose of indomethacin on follicular prostaglandins and ovulation in the rabbit. Endocri- nology, 119: 746-754. 2 Sedrani, S. H. and El-Banna, A. A. (1987) Effect of indomethacin on plasma level of vitamin D 10 11 12 13 PGE, and Reproduction in T. carnifex metabolites, oestradiol and progesterone in rabbit during early pregnancy. Comp. Biochem. Physiol., 87: 635-639. ; Greenhalgh, E. A. (1990) Luteal steroidogenesis and regression in the rat: progesterone secretion and lipid peroxidation induced in luteal cells by human chorionic gonadotrophin, phospholipase A, and prostaglandin F,,. J. Endocrinol., 125: 397-402. Dharmarajan, A. M., Sueoka, K., Miyazaki, S., Atlas, S. J., Ghodgaonkar, R. B., Dubin, N. H., Zirkin, B. R. and Wallach, E. E. (1989) Prostaglan- dins and progesterone secretion in the in vitro perfused pseudopregnant rabbit ovary. Endocrinol- ogy, 124: 1198-1203. Shimada, K., Olson, D. M. and Etches, R. J. (1984) Follicular and uterine prostaglandin levels in rela- tion to uterine contraction and the first ovulation of a sequence in the hen. Biol. Reprod., 31: 76-82. Asem, E. K. and Todd, H. and Hertelendy, F. (1987) In vitro effect of prostaglandins on the accumulation of cyclic AMP in the avian oviduct. Gen. Comp. Endocrinol., 66: 244-247. Guillette, L. J., Jr, Lavia, L. A., Walker, N. J. and Roberts, D. K. (1984) Luteolysis induced by pros- taglandin F>, in the lizard, Anolis carolinensis. Gen. Comp. Endocrinol., 56: 271-277. Guillette, L. J., Jr, Gross, T. S., Matter, J. M. and Palmer, B. D. (1990) Arginine vasotocin-induced prostaglandin synthesis in vitro by the reproductive tract of the viviparous lizard Sceloropus jarrovi. Prostaglandins, 39: 39-51. Jones, R. E., Orlicky, D. J., Austin, H. B., Rand, M. S. and Lopez, K. H. (1990) Indomethacin inhibits ovarian PGE secretion and gonadotropin- induced ovulation in a reptile (Anolis carolinensis). JE xpeZools2255: 57-62: Guillette, L. J., Jr, Bjorndal, K. A., Bolten, A. B., Gross, TI. S., Palmer, B. D:, Witherington, B. E. and Matter, J. M. (1991) Plasma estradiol-17{, progesterone, prostaglandin F, and prostaglandin E, concentration during natural oviposition in the log- gerhead turtle (Caretta caretta). Gen. Comp. Endoc- HNO S22 — 1350) Z Gobbetti, A., Zerani, M., Carnevali, O. and Botte, V. (1990) Prostaglandin F>, in female water frog, Rana esculenta: piasma levels during the annual cycle and effects of exogenous PGF;, on circulating sex hormones. Gen. Comp. Endocrinol., 80: 175- 180. Gobbetti, A., Zerani, M., Bolelli, G. F. and Botte, V. (1991) Seasonal changes in plasma prostaglandin F,, and sex hormones in the male water frog, Rana esculenta. Gen. Comp. Endocrinol., 82: 331-336. Gobbetti, A., Zerani, M. and Botte, V. (1991) Plasma prostaglandin F,, in the male Triturus car- nifex (Laur.) during the reproductive annual cycle 14 15 16 17 18 19 20 21 22 23 24 25 26 il 647 and effects of exogenous prostaglandin on sex hor- mones. Prostaglandins, 41: 67-74. Gobbetti, A., Zerani, M., Bellini-Cardellini, L. and Botte, V. (1991) Plasma prostaglandin F,, and reproduction in the female newt Triturus carnifex (Laur.) Prostaglandins, 42: 269-277. Gobbetti, A. and Zerani, M. (1992) PGF,,, PGE, and sex steroids from the abdominal gland of the male crested newt Triturus carnifex (Laur.). Pros- taglandins 43: 101-109. Gobbetti, A. and Zerani, M. (1991) Gonadotropin- releasing hormone stimulates biosynthesis of pros- taglandin F,, by the interrenal gland of the water frog, Rana esculenta, in vitro. Gen. Comp. Endocri- nol., 84: 434-439. d’Istria, M., Delrio, G., Botte, V. and Chieffi, G. (1974) Radioimmunoassay of testosterone, 17/- oestradiol and estrone in the male and female plas- ma of Rana esculenta during sexual cycle. Ster. and Lip. Res., 5: 42-48. Duncan, D. B. (1955) Multiple range and multiple F test. Biometrics, 11: 1-42. Sokal, R. R. and Rohlf, F. J. (1981) Biometry 2nd Ed., Freeman, San Francisco. Scossiroli, R. E. and Palenzona, D. L. (1979) Manuale di biometria. Zanichelli, Bologna. Galgano, A. (1944) II ciclo sessuale annuale in Triturus cristatus carnifex (Laur.). I. Il ciclo naturale nei due sessi. Arch. Anat. Embriol., 50: 1-44. Lofts, B. (1984) Amphibians. In “Marshall’s Phy- siology of reproduction” Ed. by Lamming, G. E., Churchill, London, Vol. I, pp. 127-205. Verrell, P. A., Hallyday, T. R. and Griffiths, M. L. (1986) The annual reproductive cycle of the smooth newt (Triturus vulgaris) in England. J. Zool. Lon- don, 210: 101-119. Zerani, M., Vellano, C., Amabili, F., Carnevali, O., Andreoletti, G. E. and Polzonetti-Magni A. (1991a) Sex steroids profile and plasma vitellogenin during annual cycle of crested newt (T7viturus car- nifex Laur.). Gen. Comp. Endocrinol. , 82: 337-344. Andreoletti, G. E., Colucci, D., Vellano, C. and Mazzi, V. (1984) Annual cycle of testosterone and oestradiol in the crested newt (Triturus cristatus carnifex Laur.). Atti Accad. Sci. Torino, 118: 157- 164. Andreoletti, G. E., Colucci, D., Vellano, C. and Mazzi, V. (1985) Ovine prolactin lowers serum testosterone levels in summer and winter newts (Triturus cristatus carnifex). Monit. Zool. Ital. (N. Sap 19 eI292135: 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). Monit. Zool. Ital. 28 29 30 31 32 33 34 35 36 3 38 648 (N. S.), 22: 133-144. Wallace, R. A. (1978) Oocyte growth: non mamma- lian vertebrates. In “Evolution of the vertebrate ovary”. Ed. by Jones R. E., Plenum Press, New York, pp. 469-502. Polzonetti-Magni, A. M., Curini, R., Carnevali, O., Novara, C., Zerani, M. and Gobbetti A. (1990) Ovarian development and sex steroid hormones during the reproductive cycle of Rana esculenta complex. Zool. Sci., 7: 265-271. Weintraub, A. S., Kelley, D. B. and Bockman, R. S. (1985) Prostaglandin E, induces receptivity behaviors in the female Xenopus laevis. Horm. Behav., 19: 386-399. Botte, V. and Angelini, F. (1980) Endocrine control of reproduction in reptiles: the refractory period. In “Steroids and their mechanism of action in non- mammalian vertebrates”. Ed. by Delrio, G. and Brachet, J., Raven Press, New York., pp. 201-211. Polzonetti-Magni, A., Bellini-Cardellini, L., Gob- betti, A. and Crasto, A. (1984) Plasma sex hor- mones and post-reproductive period in the green frog, Rana esculenta complex. Gen. Comp. Endocri- NO S40) 2—51e Varriale, B., Pierantoni, R., Di Matteo, L., Fasano, S., D’Antonio, M. and Chieffi, G. (1986) Plasma and testicular estradiol and plasma androgen profile in the male frog, Rana esculenta during the annual cycle. Gen. Comp. Endocrinol., 64: 401-404. Fasano, S., D’Antonio, M. and Pierantoni, R. (1991) Sites of action of local oestradiol feedback mechanism in the frog (Rana esculenta) testis. Gen. Comp. Endocrinol., 81: 492-499. Pavois BSayvand=:bicht-w Ph Gi989) ny Effects of gonadectomy and steroids on pituitary gonadotropin secretion in a frog, Rana pipiens. Biol. Reprod., 40: 40-48. Tokarz, R. R. and Crews, D. (1981) Effects of prostaglandins on sexual receptivity in the female lizard, Anolis carolinensis. Endocrinology, 109: 451-457. Whittier, J. M. and Crews, D. (1986) Effects of prostaglandin F,, on sexual behaviour and ovarian function in female garter snakes (Thamnophis sirta- lis parietalis). Endocrinology, 119: 787-792. Schmidt, R. S. and Kemnitz, C. P. (1989) Anuran 39 40 41 42 43 44 45 46 47 A. GosBetti, M. ZERANI AND V. BoTTE mating calling circuits: inhibition by prostaglandin. Horm. Behav., 23: 361-367. Botte, V. and Cottino, E. (1964) Ricerche istochi- miche sulla distribuzione del colesterolo e di alcuni enzimi della steroidogenesi nei follicoli ovarici e post-ovulatori di Rana esculenta e Triturus cristatus. Boll. Zool., 31: 491-500. Chieffi, G. and Botte, V. (1963) Osservazioni istochimiche sull’attivita della steroide-3/-olo- deidrogenasi nell’interrenale e nelle gonadi d girini adulti di Rana esculenta. Riv. Istochim. Norm. et Patol., 9: 172-174. Saidapur, S. K. and Nadkami, V. B. (1972) D°-3/- hydroxysteroid dehydrogenase, glucose-6- phosphate dehydrogenase and DPH-diaphorase activities in the post-ovulatory follicles of Rana cyanophlyctis (Schneider). Current Science, 41: 531-532. Patwardhan, V. V. and Lanthier, A. (1980) Con- centration of prostaglandins PGE and PGF, estrone, estradiol, and progesterone in human corpora lutea. Prostaglandins, 20: 963-969. McCraken, J. A. and Schramm, W. (1988) Prosta- glandin and corpus luteum regression. In “Prosta- glandins: biology and chemistry of prostaglandins and related eicosanoids” Ed. by Curtis-Prior, P. B., Churchill Livingstone, pp. 425-462. Moore, F. L. and Zoeller, R. T. (1985) Stress- induced inhibition of reproduction: evidence of sup- * pressed seecretion of LHRH in an amphibian. Gen. Comp. Endocrinol., 60: 252-258. Licht, P., McGreery, B. R., Barnes, R. and Pang, R. (1983) Seasonal and stress-related changes in plasma gonadotropins, sex steroids, and corticoster- one in the bullfrog, Rana catesbeiana. Gen. Comp. Endocrinol., 50: 124-145. Paolucci, M., Esposito, V., Di Fiore M. M. and Botte, V. (1991) Effects of short postcapture con- finement on plasma reproductive hormone and cor- ticosterone profiles in Rana esculenta during the sexual cycle. Boll. Zool., 57: 253-259. Zerani, M., Amabili, F., Mosconi, G. and Gobbetti, A. (1991) Effects of captivity stress on plasma steroid levels in the green frog, Rana esculenta, during the annual reproductive cycle. Comp. Biochem. Physiol., 98A: 491-496. ZOOLOGICAL SCIENCE 9: 649-657 (1992) Binding Properties and Photoperiodic Influence of Follicle- Stimulating Hormone Receptors in the Subtropical Wild Quail KazuyosHi Tsutsut!, SENCHIRO KAWASHIMA’, V. L. SAXENA” and A. K. SAXENA‘ Department of Radiation Biophysics, Kobe University School of Medicine, Kobe 650, Zoological Institute, Faculty of Science, University of Tokyo, Tokyo 113, Japan, *Laboratory of Reproductive Biology, D. G. College, Kanpur-208001, “Laboratory of Reproductive Biology, D. A. V. College, Kanpur-208001, India ABSTRACT—The knowledge on gonadotropin receptors in the wild avian species inhabiting in the subtropical zone is scanty. Basic properties and photoperiodism of follicle-stimulating hormone (FSH) binding to the testis or ovary of adult rain quails found abundantly in fields of northern India were studied. The binding of radioiodinated rat FSH to the particulate fraction of testicular homogenates of rain quails was competitively inhibited by mammalian FSHs but not by prolactin (PRL). Mammalian luteinizing hormones (LHs) showed some competition only at high concentrations. The Scatchard plot analysis of the binding of FSH showed a straight line, suggesting the presence of a single class of FSH-binding sites. The mean equilibrium constant of dissociation (Kd) of the specific binding and the number of binding sites were 1.16 (0.88-1.71, 95% confidence interval) nM and 3.06 (2.52—4.11) fmol/ mg tissue. Adult males transferred from short-day (SD) to long-day (LD) photoperiods showed marked increases in testicular weight and total FSH binding per testis. FSH binding per unit weight tended to increase at the initial phase of photostimulation. In contrast to the male, photostimulated females showed no significant effect of LD exposure on the total FSH binding per ovary. The changes in ovarian weight after LD exposure were much smaller than those in testicular weight and the changes from the control (SD) value were statistically not significant. These results suggest that 1) specific FSH receptors are present in the gonad of the subtropical wild quail, and their binding properties were basically the same to those of FSH receptors previously reported in several temperate birds including domestic quails, 2) photoperiod is an effective environmental factor in the regulation of FSH binding to the testis but the ovarian FSH binding is not altered by LD photoperiods in this quail, and 3) photoperiodic effects on testicular FSH binding are accompanied by pronounced changes in the testicular weight. © 1992 Zoological Society of Japan INTRODUCTION The activity of gonadal function in most species of wild birds shows a seasonal variation, and is regulated by external environmental and internal hormonal factors. Gonadotropins are essential hormones for the gonadal function, and the initial event of gonadotropin action is the binding to its specific membrane receptors in the gonads [1-3]. Photoperiod is an important environmental factor in the regulation of annual changes in the gonadal Accepted April 6, 1992 Received February 8, 1992 activity in most temperate birds, and the gonadal growth takes place under long-day (LD) photo- periods [4]. In the subtropical zone, the reproduc- tive seasons are relatively scattered throughout the year, and several environmental factors may be at work [5]. According to Chandola and Thapliyal [6], the gonadal growth in the spotted munia (Lonchura punctulata) began with the first mon- soon showers and the regression coincided with the end of the monsoon. They [6] further demon- strated that exposing the munia to constant photo- periods had no influence on the reproductive cycle and that short-day (SD) photoperiods caused the gonadal growth. In contrast, the situation in the 650 K. Tsutsul, S. KAWASHIMA et al. Baya weaver finch (Ploceus philippinus) approxi- mated much more to temperate birds. Pavgi [7] indicated that the seasonal variation of gonadal activity in this finch was photoperiodically reg- ulated. Thus, information on photoperiodic in- fluences on gonads is still confusing in subtropical birds. Therefore, study of the effects of photo- period on the gonadal activity and gonadotropin receptors certainly provides useful information on endocrine control mechanisms of the changes in gonadal function in wild avian species inhabiting in the subtropical zone. Previous studies have extensively demonstrated FSH receptors in the testis of several species of temperate birds, such as white-crowned sparrow [1], domestic fowl [8, 9], turkey [9] and domestic quail, i.e., Japanese quail [3, 10, 11]. Similarly, the knowledge of ovarian FSH receptors has been accumulated in the temperate birds, such as domestic hen [12, 13] and turkey [9]. Photo- periodic responsiveness of FSH receptors has been reported in two male avian species (white-crowned sparrow and Japanese quail) among the temperate species [1, 10]. However, to the best of our knowledge the study on FSH receptors in the subtropical wild species has not yet been reported. In the present study, we used the wild quail population, rain quail, inhabiting in the subtropi- cal zone in northern India. First purpose of this study is to characterize the basic properties of FSH binding to the gonad. The other purpose is to determine the photoperiodic influence on gonadal FSH binding in this subtropical birds. We will present evidences suggesting the presence of spe- cific FSH receptors and the sex difference in their photoperiodic response. MATERIALS AND METHODS Animals Adult rain quails, Coturnix coromandelica, found abundantly in fields of northern India were used in the present study. All of the subtropical wild quails were obtained from local suppliers in June, 1987 and kept in wiremesh cages (50 x 30 x 23 cm) in the aviary under SD photoperiods (6-hr light, 18-hr dark) at natural ambient temperature (maximum monthly average, 35°C in May; mini- mum monthly average, 14°C in December) with supply of commercial food and water ad libitum. Birds were divided into five groups of five birds each. The first group was transferred to LD photoperiods (18-hr light, 6-hr dark) on 30th Au- gust. The second, third and fourth groups were transferred to LD on 15ths September, October and November, respectively. The last group main- tained on SD throughout served as controls. The day of transfer to LD photoperiods was designated as day 0. Birds in all groups were simultaneously sacrificed by decapitation on 25th November, 1987. Receptor preparations Immediately after blood collection, the testes and ovary were removed and weighed on a torsion balance to the nearest 0.5mg. They were snap- frozen on dry ice-ethanol and stored at —80°C or on dry ice (a few days during sample- transportation from India to Japan) until the bind- ing assay for FSH was performed in January, 1988. The frozen samples were rapidly thawed and homogenized in cold Tris-HCl buffer (0.04 M; pH: 7.4) containing MgSO, (5 mM) and 0.1% BSA. The homogenates were centrifuged at 11,000 xg for 20min at 4°C. The resulting pellets were resuspended in cold buffer and adjusted to contain 4 mg equivalent wet tissue/100 wi as the receptor preparation. A part of receptor preparations in the first group was used to characterize basic properties of FSH receptors. Hormone preparations Highly purified rat FSH (NIDDK-rFSH-I-6) was radioiodinated for the assay of FSH receptors. Unlabeled NIDDK-rFSH-I-6, NIDDK-ovine (0)FSH-17, | NIH-FSH-P-2, ©. NIDDK-rLH-I-5, NIDDK-oLH-25 and NIDDK-rPRL-I-4 were used as competitors for competition-binding experi- ments. Unlabeled NIH-FSH-P-2 was used to cor- rect for nonspecific binding throughout the assay of FSH receptors. Binding assay For the assay of FSH receptors NIDDK-rFSH-I- 6 was radioiodinated with '*!I (Na‘*'I, Radio- FSH Receptors in the Subtropical Wild Quail 651 chemical Centre, Amersham, United Kingdom) in the presence of lactoperoxidase and hydrogen peroxide using the method described previously [10, 11]. The specific activity of labeled FSH was 52 uw Ci/ug calculated by our previous method [14]. Due to the small amount of membranes available in the gonads of rain quails, most of the binding experiments were performed using a mic- ro-radioreceptor assay (RRA) described previous- ly [14-16]. The volume of the reaction mixture in the micro-RRA (90 1) was approximately a half of the standard assay system (200 wl; [17]). The precision index (A) was less than 0.17 in the micro-RRA and 0.10 in the standard RRA. For the FSH-binding assay, receptor preparation (4 mgeq original tissue/100 wl in the standard RRA; 2mgeq tissue/50 ul in the micro-RRA) and [‘*'I]iodo-rFSH (0.98 ng/50 #1 in the standard RRA; 0.39 ng/20 ul in the micro-RRA) were incu- bated at 37°C for 3 hr with or without unlabeled NIH-FSH-P-2 (20 ug/50 yl in the standard RRA; 8 wg/20 wl in the micro-RRA). In the saturation binding experiments, different amounts of ['*'Iiodo-rFSH (0.15-4.9 ng; 20 wl) and receptor preparations (2 mgeq original tissue; 50 ul) were incubated with or without an excess amount of cold NIH-FSH-P-2 (2.5-80 yg; 20 wl). At the end of incubation, 1 ml cold Tris-HCI buffer (0.04 M; pH 7.4) containing 5mM MgSO, and 0.1% BSA was added to each tube, and the tubes were centrifuged at 11,000g for 3 min at 4°C. The pellets were washed twice with cold buffer, and the radioactivity of resultant pellets was counted in an autowell y-counter. Before the experiments, all reaction tubes had been coated with BSA. Scatch- ard plots were constructed from the data obtained from the saturation binding experiment. The equilibrium constant of dissociation (Kd) and the number of binding sites were determined from the Scatchard plots. A straight line was fitted to the plots by the method of least squares. Statistical analysis To compare the patterns of changes in gonadal weight and binding capacity after transfer from SD to LD between the male and female, rates of changes in these parameters from each control value were employed to normalize indices. The gonadal weight and binding capacity after photo- stimulation were expressed as the ratio to the mean value of each parameter of the control. Results were expressed as the mean+SEM and were analyzed for significance of difference by Bartlett test, followed by Duncan’s multiple range test [18]. Statistics for linearity, precision and 95% confidence interval were computed according to the method of Bliss [19]. RESULTS Binding properties of FSH receptors in the subtro- pical wild quail Figure 1 shows the effect of incubation time on FSH binding to the testis of the subtropical wild quail. When 0.98 ng of ['°'IJiodo-rFSH and the receptor preparation derived from 4mgeq wet tissue were incubated by the standard RRA system (200 wl per assay tube), specific binding of ['*'T] iodo-rFSH increased rapidly during the first 1 hr of incubation at 37°C and tended to reach a plateau after 3 hr. Specific Binding (cpm x 10-2) : 1 2 3 Incubation time (hr) Fic. 1. Binding of ['*'I]iodo-rFSH to the particulate fraction of testicular homogenates of the rain quail as a function of incubation time. Incubation of standard RRA at 37°C. Solid and open circles represent specific and nonspecific bindings of dupli- cate determinations. In order to examine the ligand specificity of FSH binding to the testis, competition experiments were performed by the micro-RRA system (0.39 ng labeled NIDDK-rFSH-I-6 and 2 mg testicular NIDDK-rPRL-I-4 NIDDK-oLH-25 ~~ v . . . 4. NIH-FSH-P-2 4 08 NIDDK-oFSH-17 Binding (cpm x 10-2) -2 -i 0 Competitor in log (ng) Fic. 2. Competition of specific binding of ['*'Iiodo- tFSH to the particulate fraction of testicular homogenates of the rain quail by various gonadotro- pin preparations (NIDDK-rFSH-I-6, NIDDK- oFSH-17, NIH-FSH-P-2, NIDDK-rLH-I-5 and NIDDK-oLH-25) and NIDDK-rPRL-I-4. Incuba- tion of micro-RRA for 3 hr at 37°C. tissue; 90 zl per assay tube) using NIDDK-rFSH-I- 6, NIDDK-oFSH-17, NIH-FSH-P-2, NIDDK- rLH-I-5, NIDDK-oLH-25, and NIDDK-rPRL-I-4 as competitors. When three kinds of FSH prepara- tions were used as the competitors, the binding of ['°'T]iodo-rFSH was inhibited as a function of the concentration of the competitor (Fig. 2). In con- trast, NIDDK-rPRL-I-4 failed to inhibit FSH bind- K. Tsutsu1, S. KAWASHIMA et al. ing, and NIDDK-oLH-25 tended to slightly inhibit only at high concentrations (>2.0 ug). Although NIDDK-rLH-I-5 showed some competition at high concentrations, the inhibitory potency was clearly less than any FSH preparations (e.g., rLH-I-5 vs. FSH-P-2, ca. 1:32). The FSH contamination of NIDDK-rLH-I-5 was less than 0.04 x NIH-FSH-S1 (data from NIDDK, U.S.A.). The biological potency of NIH-FSH-P-2 was 0.69 x NIH-FSH-S1 (data from NIDDK). Therefore, it may be consid- ered that the FSH contamination of NIDDK-rLH- I-5 is less than 0.058 x NIH-FSH-P-2 (e.g., rLH-I-5 vs. FSH-P-2, ca. 1: >17). The inhibitory potency of NIDDK-rLH-I-5 might be due to the con- tamination of FSH in this preparation. The effect of receptor concentration on the binding level was examined in this experiment. The receptor preparation of various concentra- tions ranging from 0.5 to 8 mgeq wet tissue and 0.98 ng of ['*'I]iodo-rFSH were incubated by the standard RRA system (200 ul per tube). Specific binding of labeled FSH was increased as a function of receptor concentration (Fig. 3). To examine the saturability of FSH binding to the testis, satura- tion-binding experiments were conducted by the - micro-RRA system (2 mg testicular tissue and 0.15-4.9 ng labeled FSH; 90 wl per tube). As shown in Figure 4, specific binding of ['*'I]iodo- 8 4 a =a EB Specific oO. oO 2 e @ & 2 42) A=} aia on . 3 Dil aig oesssweceeteceer coe re e000 esse - Tissue concentration (mg) Fic. 3. Binding of ['°'I]iodo-rFSH to the particulate fraction of testicular homogenates of the rain quail as a function of receptor tissue concentration. Incubation of standard RRA for 3 hr at 37°C. Solid and open circles represent specific and nonspecific bindings. FSH Receptors in the Subtropical Wild Quail 653 rFSH tended to saturate with respect to the con- centration of labeled FSH. In contrast, nonspecific binding increased linearly. Scatchard plots were constructed from the data of this experiment (Fig. 4). A straight line could be fitted to the plots (P< 0.05). However, there was a rather large variation in Scatchard plots (Fig. 4). The mean apparent dissociation constant (KD) and the number of FSH-binding sites calculated from the slope of the fitted straight line were 1.16nM (0.88-1.71 nM, 95% confidence interval) and 3.06 (2.52-4.11) fmol/mg tissue, respectively. Photoperiodisms of testicular and ovarian FSH bindings in the subtropical wild quail As shown in Figure 5, upper panel, transfer from SD to LD induced a marked increase in testicular weight, though the testis in the SD (control) group 0.06 Binding (cpm x 10 ~3) 0.02 0 20 40 B (pM) remained small (P<0.05, SD vs. LD-day 71; P< 0.01, SD vs. LD-day 87). The specific binding of ['°'Ijiodo-rFSH per unit testicular weight (density of FSH binding) tended to increase up to 41 days of photostimulation (SD vs. LD-day 41, ca. 1: 1.9), but the alteration was not significant (Fig. 5, middle panel). In contrast, FSH binding per testis (total FSH binding) markedly increased during photostimulation (P<0.05, SD vs. LD-day 71; P< 0.01, SD vs. LD-day 87; Fig. 5, lower panel). On the other hand, the changes in ovarian weight after LD exposure were much smaller than those in testicular weight and the changes from the control SD value were statistically not significant (Fig.5, upper panel). As shown in Figure 5, middle panel, the density of FSH binding was almost constant during photostimulation. Unlike in the testis, there was no significant effect of LD 131]-FSH (ng) 60 Fic. 4. Scatchard plots of the binding of ['*'I]iodo-rFSH to the particulate fraction of testicular homogenates of the rain quail. B, Concentration of bound hormone at apparent equilibrium; F, concentration of free hormone at apparent equilibrium. Inset, specific (solid circle) and nonspecific (open circle) bindings in the saturation binding experiment. Incubation of micro-RRA for 3 hr at 37°C. 654 K. Tsutsur, S. KAWASHIMA et al. (Ee Testis wl Ovary 30 + ae cere io wn E Ww : IMM Relative gonadal weight ( Ye EC Aaa de converter PDE sles tatiiata ig Pea rae Netra graye tie hese! st a oe Minvateht a a beets Sp TIL “ial 2 rT, HeIao 1 is)” hk o> TROT wile a (fcHtt ey Sty Fee “anton? : ee o.oo le vo Be - hati ie i. al v2 Joi errant ; eae 7 Chie. STIR, ee es ae ‘Rol if vy Kod Was Shorr gst Siskel Te: mate . ‘ feta S sve ly : ; a ne he} att mbites SIDA laa bess bere | ~ é \ = * s% Ve +a. : a ; . Bis OFF te Ser ned Sed, de Rhea de ues dae el 2 De 9 Ste ROOREHE I 3S st ee 4 ae Age = | + (iat aah) ve a2 | eae if ee Ey Se ie ve deve Oe el ia ee. Sea ae = cs Wteshyrlbouyd i peyes att AE PRN: ae EY | Pee : Eo 4 1 2a, hs De UIE TIE LB days Witt aD" Eats yiRes Wis Sipe Sey _ j fe oa IL 2s) Macana No Me - hab) = Peale itt) dary. = : yf 5 Atreg aera yr ee et wee! o y : a eS es PREPAY = Teed ee Gee Oe eS ee = Thy To tare ere. i ait it — ; i oat “ te Pros a ; ise et | S3ED? pera - h; LU Po A + P J ei iJ Ef i, Jibz 7 t : set he i ~ = 3 a 5 Nien i ies = bead ; ’ ‘ Af 7{ bs J =a | ae é f 6 i UL f riko iee| Bet Erect ie” 7 ' F z 02 ies i | ye. 8: eo bis FTE YS % — ‘ i : ibis i Ape Ary ais z 4 p Pasi 3 : naam = vila eee ep ’ (SHE efit: 7 . i f aes 3 vs Pe iy ‘ee F ae J ‘KOH by mr ues, at, agi *? fe- ia a iu : Se x i ‘ 4 > 4 ca + ’ 4 ays Eee . J , 7 r x " re ¥ > i 4 ¢ ' . S 4 : 5 = i > 3 > 4 ] y 4 } vr ‘J i 3 Fa i , ) a, # 5 i . i b as ZOOLOGICAL SCIENCE 9: 659-664 (1992) The Association between Vocal Characteristics and Habitat Type in Taiwanese Passerines JULIA I. SmitH and Hon-TSEN Yu Museum of Vertebrate Zoology and Department of Integrative Biology University of California, Berkeley, California 94720, U.S.A. ABSTRACT— We analyzed the vocalizations of twelve Taiwanese passerines to examine whether vocal characteristics are associated with habitat type: edge, forest canopy, and forest floor. Morton [1] and Richards and Wiley [2] argue that because sound wave propagation is dependent upon the characteris- tics of the environment, selection should favor the use of song features which will carry information over the required distance [3]. For each vocalization we calculated the frequency emphasized, the range of frequency modulation, and the duration. We found relatively little variation among the members of each species. Vocal comparisons among habitat types revealed that bird sounds of Taiwanese habitats differ. However, in contrast to the pattern found in the neotropics [1, 2, 4], Taiwanese edge species do not emphasize higher frequencies, over a wider range, and with more rapid repetition, than forest dwellers. Most of the vocal differences among habitat types result from forest floor dwelling Taiwanese passerines vocalizing, on average, at frequencies 2000 Hz lower than other species. This supports the hypothesis that, due to the rapid attenuation of high frequency sound near the ground, bird sounds of © 1992 Zoological Society of Japan the forest floor will be of low frequency. INTRODUCTION Many researchers [1-4] have argued that the association of certain vocal characteristics with a specific habitat is the result of selection acting to produce a salient signal. Studies of neotropical passerines [1, 2] have demonstrated that forest dwelling species vocalize at lower frequencies, utilize a more restricted range of frequencies, and employ less rapid repetition than birds in open habitats. However, these patterns of vocalization have been documented for some, but not all, North American temperate habitats [4, 5]. Wiley and Richards [4] suggest that behavior may account for vocal differences, between temperate and tropical regions. Specifically, they [4] propose that neotropical forest species, unlike temperate forest passerines, will sing directly from the ground where the effects of attenuation make high fre- quency sound useless. In addition, Richards and Wily [2] argue that forest dwelling birds, because of the greater influence of reverberation, may be Accepted March 4, 1992 Received August 10, 1991 more likely to avoid high repetition rates than birds of open habitats. This association of vocal frequency and repetition rate with habitat type has never been examined in Old World tropical Taiwanese passerines. In fact, the vocalizations of the species in our study have never before been analyzed. Therefore, we will describe the basic structure and pattern of vocalization for these twelve Taiwanese passerines and test two hypoth- eses: 1) whether birds of forested habitats vocalize at lower frequencies and with less variance and slower repetition rates than do birds of open edge areas; and 2) whether birds of the forest floor- subject to the most stringent effects of attenuation- vocalize at lower frequencies than canopy dwel- lers. MATERIALS AND METHODS Recordings were made with a Sony TC-D5M cassette recorder and Sony C-74 microphone. Birds were recorded, in June of 1987, at Wuling (elevation 1800 m) and Chitou (elevation 1200 m) in central Taiwan. The natural vegetation of these two localities is broad-leaf forest dominated by 660 J. I. SmirH AND H.-T. Yu members of the Fagaceae and Lauraceae [6]. A large portion of the natural forest has been logged and replaced by conifer plantations, primarily Cryptomeria japonica. Some birds were recorded along the northeast coast at Yenliao (elevation 50 m); here, Phyllostachys makinoi and Acacia con- fusa make-up the secondary growth. Birds were recorded in two general habitat types: edge and forest. Forest habitat is characte- rized by two strata. The lower stratum averages 2 m to 8m high and consists of herbaceous and shrubby plants. The upper stratum, the canopy with many lianas, is about 10 m to 20 m high. The edge areas remain open with grasses (Miscanthus spp.) and herbaceous plants. We studied twelve species of Taiwanese passer- ines, representing three taxonomic subfamilies- Sylviinae, Timaliinae, and Muscicapinae (Table 1). Based on behavioral observations (H-TY) and behavioral descriptions by DeSchauensee [7], and Hachisuka and Udagwa [8], the species were categorized according to preferred habitat type. For forest dwellers the position they occupy— canopy or floor—was also recorded (Table 1). Vocalizations were analyzed with a Kay Elemetrics 6061B sonagraph. Following Morton [1], the fre- quency emphasized (in Hz) for each sonogram was determined by locating the midpoint of the darkest region of the vocal trace. The extent of frequency modulation in each vocal trace was described by subtracting the minimum from the maximum fre- quency for each species. For each species, the duration of a typical vocalization was measured to the nearest 0.1 second. Finally, by visual inspec- tion, the occurrence of rapid repetition of similar notes within a vocal trace was noted. RESULTS Description of Vocalizations Edge Species Liocichla steeri has a vocalization consisting of two loud piercing introductory notes followed by a less emphasized slur, “Dee, deeee-yer” (Fig. 1a). This vocalization is comprised of very high fre- quencies—the highest in our study—with exten- sive frequency modulation (Table 1). Interesting- ly, the male will duet with the female. The female enters the song during the last syllable of the male’s vocalization with a buzzy, “Gee, gee, gee ...” (Fig. 1a). The alteration of voices is precisely timed. It is often difficult to discern that the voices have changed. Garrulax canorus taewanus, often kept as a cage bird, is renowned for its musical vocalizations. The vocalization consists of clear, rich whistles with an accent on the first note of each doublet or triplet (Fig. 1b). Relative to the other species we examined this vocalization is long and is also highly modulated (Table 1). Cettia fortipes has a vocalization consisting of a soft hum that crescendos and explodes into a sharp series of notes, “Hmmmmmmm, switch you!” Ex- tensive frequency modulation (almost 4000 Hz; Table 1, Fig. 1c) occurs in the final elements of the vocalization. Bradypterus seebohmi vocalizes with a high, thin, rhythmically repeating whistle, “Da, da, deeee, tick.” Despite the fact that the entire vocalization covers a wide range of frequencies (almost 3500 Hz; Table 1), no song notes were rapidly modulated; each note of the song is dis- crete, without any slurring (Fig. 1d). The vocaliza- tion is very syncopated. é Forest Canopy Species Abroscopus albogularis has a vocalization con- sisting of a high-pitched, extemely rapid repetition of bell-like tinkling notes (Fig. le). Of the species investigated, A. albogularis exhibits the fastest repetition of notes. It uses very little frequency modulation; within almost one second of vocaliza- tion there is less than 1000 Hz of frequency modula- tion (Table 1). Niltava vivida produces a lilting series of thin whistled notes (Fig. 1f). It is a relatively long, musical vocalization consisting of much slurred freqeuncy modulation (Table 1). Heterophasia auricularis uses a loud strident song initiated with several sharply accented notes and ending with a descending slur, “Wheep, wheep, wheep, weeee-ooo0.” Each note in the vocalization is clear and distinct (Fig. 2a). Each note exhibits extensive frequency modulation, at Vocalization and Habitat Type Association 661 kHz L kHz b FE OMR RTE kHz a 0.5 1.0 15 SEC Fic. 1. 0.5 1.0 1.5 SEC Sonograms representing the vocalizations of: a) Liocichla steeri, the female’s vocalization is indicated by an arrow; b) Garrulax canorus taewanus, c) Cettia fortipes, d) Bradypterus seebohmi, e) Abroscopus albogularis, and f) Niltava vivida. times as much as 4000 Hz (Table 1). Yuhina brunneiceps uses a thin, lilting whistle with a descending slur at the end (Fig. 2b). This is musical vocalization. There is a great deal of frequency modulation within and between notes (Table 1). Forest Floor Species Pomatorhinus ruficollis musicus vocalizes using an accelerating series of repetitive notes, initiated with an accent, “Duh, doo, doo, doo...Breep.” It exhibits very rapid repetition of relatively unmodu- lated notes (Fig. 2c). Of the species we investi- gated P. r. musicus utilizes the lowest frequencies (Table 1). Stachyris ruficeps has a vocalization characte- rized by a very rapid series of clear notes, “Dee, dee, dee ....” All notes are of the same frequency and about the same sonographic shape (Fig. 2d). The frequency utilized is relatively low and unmodulated (Table 1). Garrulax poecilorhynchus uses a very slow, melodic whistle, “Wooo-whip.” The frequency of this vocalization is quite low; it never reaches over 2500 Hz (Table 1). The notes are distinctly sepa- rated, and the frequency modulation is quite slow (Fig. 2e). Alcippe brunnea uses a variable series of flute- like notes. It is very musical vocalization. Each note is extensively modulated (Fig. 2f) but within a relatively narrow range of frequencies (typically, not more than 2400 Hz; Table 1). Habitat Association Striking vocal differences among species are found within each habitat type (edge, Fig. la-d; forest canopy, Fig. le-f and Fig. 2a-b; forest floor, Fig. 2c-f). Across all habitat types there are songs which exhibit extensive frequency modulation: edge—L. steeri (Fig. 1a), forest canopy—H. au- 662 J. I. SmitH AND H.-T. Yu TABLE 1. The mean vocal frequency (+SD), frequency range, and duration of a typical vocalization for each species. Also noted are: habitat type, habitat position, number of individuals analyzed (n), and taxonomic subfamily. aie) Subtamily Megp Remieney °” Fieaueay ae Edge Liocichla steeri 8 Timaliinae 4524 + 302 2186-6470 1.0-1.5 Garrulax canorus taewanus 1 Timaliinae S377 1661-4022 3.0-6.0 Cettia fortipes D, Sylviinae 4459 + 124 1574-5902 1.2-1.5 Bradypterus seebohmi 2 Sylviinae 3847 +0 2623-5508 0.7-0.8 Mean 4038 +491 Canopy Abroscopus albogularis 3 Sylviinae 4750 + 182 4459-5333 0.6-1.0 Niltava vivida 1 Muscicapinae 4327 3235-5246 1.4 Heterophasia auricularis V Timaliinae 3485 + 239 1486-5333 0.9-1.1 Yuhina brunneiceps + Timaliinae 4502 + 497 2098-5945 0.7-0.8 Mean _ 4277 +477 Forest Floor Pomatorhinus ruficollis musicus 3 Timalinae 787 +87 262-3322 0.8-1.6 Stachyris ruficeps 5 Timaliinae 22035259 1749-2448 0.6-1.4 Garrulax poecilorhynchus Timaliinae 1661 1573-2361 0.6 Alcippe brunnea HL Timaliinae 3091 + 280 1574-4197 0.8-1.2 Mean 1936 + 837 ricularis (Fig. 2a), and forest floor—A. brunnea (Fig. 2f). Likewise rather unmodulated vocaliza- tions also occur in each habitat: edge—B. seebohmi (Fig. 1d), forest canopy—A. albogularis Fig. le and forest floor—S. ruficeps (Fig. 2d). Vocalizations with rapid repetition are found only in forest dwelling species (forest canopy—A. albo- gularis—Fig. le and forest floor dwellers—P. ruficollis musicus and S. ruficeps—Fig. 2c and d). In edge habitat we discovered no rapidly repeating vocalizations (Fig. la-d). No apparent structural feature of the vocal traces correlates with habitat type. Within each species we found relatively little variation in frequency emphasized; standard de- viations were never greater than 500 Hz (Table 1). We calculated the mean frequency emphasized, standard deviation, and coefficient of variation (x —mean+sd, and CV) in each habitat type by averaging the mean frequencies of the species found in that area: edge (n=4, x=4038 Hz+491; CV=12.5), forest canopy (n=4, x=4277 Hz+ 477; CV=11.2), and forest floor (n=4, x=1936 Hz+837; CV=43.2). Mean frequenices empha- sized in each of these habitats differ significantly (Kruskal-Wallis Test H=7.53, dfi=2; P<0.01; [9]). Frequeinces used in the forest canopy are the highest and those of the forest floor are the lowest. Additionally, the mean frequencies used in the three habitats were compared in a pairwise fashion with Dunn’s Test [10]. Species from the forest floor use significantly lower frequencies than those of the forest canopy (Z=2.55, 0.05< P<0.10) as well as those of the edge habitat (Z=2.16, 0.10

0.20). DISCUSSION In a comparison of edge and forest habitats we found no support for the hypothesis that edge species utilize higher frequenices over a wider range, with more repetition, than all forest dwel- Vocalization and Habitat Type Association 663 kHz a kHz Ae So w maceet kHz 0.5 SEC Fic. 2. = 2a eee = SEC Sonograms representing the vocalizations of: a) Heterophasia auricularis,b) Yuhina brunneiceps c) Poma- torhinus ruficollis musicus, the vocal trace at 4 kHz is an insect; d) Stachyris ruficeps, e) Garrulax poecilorhyn- chus, and f) Alcippe brunnea. lers. Specifically, mean frequencies do differ among the habitats but unlike neotropical passer- ines the frequencies in edge habitat are only higher than those used by forest floor dwellers not those of the forest canopy. Edge species of the neotro- pics [1, 2, 4] emphasize a wider range of frequen- cies than forest dwellers, again a pattern not seen in Old World Taiwanese passerines. The coef- ficients of variation of mean frequencies empha- sized in edge and forest canopy are similar, where- as forest floor dwellers exhibit the greatest extent of variation—the CV is over three times that of the edge species. Contrary to predictions concern- ing the increased effects of reverberation in the forest on repeated notes [2], we found that Taiwanese edge species do not use more rapid repetition rates than forest dwellers. In fact, of the edge species we investigated (Fig. la-d) none util- ized a repetition rate greater than the forest canopy dweller A. albogularis (Fig. le). In addi- tion, two of the four forest floor dwellers (P. r. musicus and S. ruficeps) utilize extremely rapid repetition (Fig. 2c and 2d). Again the pattern exhibited by Taiwanese (Old World tropical) pas- serines is not that of their neotropical counter- parts. Interestingly, Handford [11] and Notte- bohm [12] both report, from analyses of within species variation in Zonotrichia capensis, that some differences among dialects do not match Morton’s [1] predictions. They [11, 12] speculate that there are features of a habitat, other than the physical milieu, that can influence the structure of song and thus vocal patterns can contradict Mor- ton’s [1] predictions. Perhaps in the Old World tropics some aspects of avian vocalizations are not only modified by the physical environment but also by the vocalizations of other species. The most striking vocal difference among Taiwanese passerines is found in the frequencies used by forest floor dwellers. These frequencies 664 are, on average, 2000 Hz lower than those of the canopy and edge dwellers. This pattern cannot be explained by phylogeny whereby phylogenetic constraints result in the most closely related spe- cies occupying the same habitat and utilizing the same vocal frequences. Within each habitat type more than one subfamily is represented (Table 1). Indeed, the two most closely related species in the study—the Garrulax congeners—occupy different habitat types and vocalize at very different fre- quencies (Table 1). Thus, there is support for the hypothesis that, due to the rapid attenuation of high frequency sound near the ground, bird voca- lizations of the forest floor tend to be of low frequency. In conclusion, our data do not support the hypothesis that tropical edge species vocalize at higher frequencies, with more rapid repetition, and with a wider range of frequencies than forest dwellers. However, our data do support the hypothesis that vocalizations of forest floor dwell- ing passerines may be influenced by the acoustic properties of their evnironment. ACKNOWLEDGMENTS We wish to thank the Curator of Birds, Ned K. Johnson, in the Museum of Verebrate Zoology, Uni- versity of California at Berkeley, for access to sonog- traphic equipment. For recording equipment we would like to thank Hai-Yin Wu of the National Taiwan Uni- versity. For technical support and expert advice we would like to thank: D. A. Bell, Y. L. Chien, C. Cicero, M. N. F. da Silva, J. G. Groth, R. E. Jones, N. K. Johnson, K. Klitz, M. Kuramoto, J. L. Patton, C. J. Schneider, B. R. Stein, and F. X. Villablanca. Com- ments from two anonymous reviewers were extremely valuable. 10 12 J. I. SmitH AND H.-T. Yu REFERENCES Morton, E. S. (1975) Ecological sources of selection on avian sounds. Amer. Natur. 109: 17-34. Richards, D. G. and Wiley, R. H. (1980) Rever- berations and amplitude fluctuations in the propaga- tion of sound in a forest: implications for animal communication. Amer. Natur. 115: 381-399. Nottebohm, F. (1985) Sound transmission, signal salience, and song dialects. Behav. Brain Sciences 8: 112-113. Wiley, R. H. and Richards, D. G. (1982) Adapta- tion for acoustic communication in birds: Sound transmission and signal detection. In “Acoustic com- munication in birds”. Ed. by D. E. Kroodsma and E. H. Miller, Academic Press, Inc. New York, Vol. 1, pp. 131-181. Wasserman, F. E. (1979) The relationship between habitat and song in the white-throated sparrow. Condor 81: 424-426. Li, H.L., Liu, 1. S.; Huang) i © Skovamasieond DeVol, C.E. (1975) Flora of Taiwan. Vol. 1. Taipei, Epoch Publishing Corp. DeSchauensee, R. M. (1984) The birds of China. Washington, D. C., Smithsonian Institution Press. Hachisuka, M. and Udagwa, T. (1951) Contribu- tions to the ornithology of Formosa. Quart. J. Taiwan Museum 4: 1-180. Sokal, R. R. and Rohlf, F. J. (1981) Biometry. 2nd - ed. San Francisco, W. H: Freeman and Company. Dunn, O. J. (1964) Multiple comparisons using rank sums. Technometrics 6: 241-252. Handford, P. (1981) Vegetational correlates of variation in the song of Zonotrichia capensis. Behav. Ecol. and Sociobiol. 8: 203-206. Nottebohm, F. (1975) Continental patterns of song variability in Zonotrichia capensis: some possible ecological correlates. Amer. Natur. 109: 605-624. ZOOLOGICAL SCIENCE 9: 665-669 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan The Confocal Laser Scanning Microscope as a Tool for Studying Xanthophores of the Swordtail (Xiphophorus helleri) G. Kuoo, V. P. E. PHanc! and T. M. Lim Department of Zoology, National University of Singapore, Kent Ridge, Singapore 0511, Republic of Singapore ABSTRACT—The morphology of xanthophores on scales of the swordtail, Xiphophorus helleri, was studied using bright-field light, differential interference contrast (DIC) and confocal laser scanning microscopies (CLSM). Under light microscopy, xanthophores were indistinct with diffused cell outline. Improved imaging, showing gradient shaded relief-like images of xantho- phores, was obtained under DIC. The CLSM produced the clearest images of xanthophores and also intracellular details as these cells were autofluorescent when scanned with the argon ion laser beam. The cell nuclei were non-autofluorescent and thus appeared dark. Selected portions of a cell can be highly magnified to show intracellular structures using the CLSM computer soft- ware. Also, serial optical tomographic sections of xanthophores with the CLSM enabled studies of intracellular structures present at different depths of the cells, thus permitting reconstruction of _ three- dimensional cell models. Therefore, use of the CLSM is highly recommened to complement bright-field light and DIC microscopies for morphological studies of xantho- phores and other autofluorescent cells. INTRODUCTION Confocal laser scanning microscopy is an emrging technology with numerous applications in cell biology and cytochemistry. The confocal laser scanning microscope (CLSM), used primarily for visualization of fluorescent molecules (which can be either autofluorescent or fluorochromelabeled antibodies and ion sensitive dyes), permits more Accepted March 6, 1992 Received November 27, 1991 ' To whom all correspondence should be addressed. precise observations of sub-cellular structures, such as microtubules, centrosomes, chromosomes and nuclei, than could be achieved with conven- tional epifluorescent microscopy [1-3]. There is also improved spatial and lateral imaging of fluorescent specimens with overlapping structures such as eggs and embryos [3]. During confocal laser scanning, a specimen is illuminated with a small diffraction limited spot from a focused laser beam. Fluorescent signals from the illuminated spot, viewed with spatially restricted imaging optics, are descanned, spatially filtered and detected by a photomultiplier prior to image assembly in an integral image processor and display on a high-resolution video monitor {1]. A confocal image is then formed with improved rejection of out-of-focus information and greater resolution than conventional light microscopy [1- SIE The CLSM can contribute greatly to the field of cellular tomography. Thick fluorescent stained specimens, for example; sea urchin eggs, micro- tubules of HeLa cells, plasmacytoma cells, Dro- sophila salivary glands, nematodes and chick embryos can be optically sectioned to generate high resolution images of serial tomographic sec- tions without the need for laborious microtome sectioning of fixed or frozen specimens [1, 4, 5]. Specimens are sectioned along the X-Y plane by the laser beam at varying preset thickness to produce Z-Series optical sections at different levels. A smoothly continuous three-dimensional model can be reconstructed from these serial 666 G. Kuoo, V. P. E. PHANG AND T. M. Lim sections [5]. Brightly colored vertebrates, especially fishes, possess various chromatophore types such as mel- anophores (black), erythrophores (red), xantho- phores (yellow), xantho-erythrophores (yellow with red periphery), leucophores (white) and iridophores (reflecting) [6-9]. _Chromatophore morphology is usually studied with conventional light microscopy and ultrastructural details with electron microscopy [9-11]. In this study, three different microscopy techni- ques; bright-field light, differential interference contrast (DIC) and confocal laser scanning were employed to examine the morphological character- istics of xanthophores of fancy color varieties of the swordtail, Xiphophorus helleri. Studies of xanthophores with light and DIC microscopes have always produced indistinct images due to the dense carotenoid pigment content [6, 8, 9]. There- fore, the main focus of this study was the applica- tion of the special functions of the CLSM. Since xanthophores possess fluorescent pteridines in pterinosomes [10-13], they are ideal specimens for CLSM investigations which have several distinct advantages over bright-field light and DIC tech- niques. MATERIALS AND METHODS Scales from the dorsal and ventral regions of the Golden and Neon varieties of swordtail were detached with fine forceps and_ individually mounted in teleost physiological saline, TPS (6.5 g NaCl, 0.4g KCl, 0.15 g CaCl,-2H,O and 0.15 ¢ MgS0O,:7H20 in 1 liter of distilled water, pH 7.3) on glass slides. An Olympus BH2 binocular light microscope, at 200 to 1000 magnifications, was used for conventional bright-field light microscopy studies. The BH-2 microscope was also fitted with a Nomarski attachment (BH2-NIC) for DIC studies. The BIO-RAD Lasersharp MRC-500 Confocal Laser Scanning Microscope (BIO-RAD Micro- science Div., UK), consisting of a computer controlled laser scanner assembly attached to a Nikon Optiphot microscope, was used for in-depth studies of autofluorescent granules and intracellu- lar structures in xanthophores present on the scales. The CLSM had an argon ion laser scanning in multiline mode but a special interference filter assembly, comprising 488 nm (blue) excitation and 515 nm (yellow) emission filters, was used to select specific lines. The CLSM laser beam, of 488 nm wavelength, was used at different magnifications for scanning the xanthophores. Xanthophores were optically sectioned along the X-Y plane by the laser beam at various preset thickness (1-2 wm) to generate Z-Series optical tomographic sections at different depths of the cells. Photomicrographs of xanthophores and - images collected by the CLSM were taken using the Olympus BH2 microscope with PM-10AD Exposure Unit, and Polaroid Freeze-frame Photo- micrographic System, respectively. Kodak TMX- 100 print film was used for the photographs. RESULTS AND DISCUSSION Xanthophores of swordtails were difficult to distinguish as discrete cells with distinct cell outline Fic. 1. High magnification light microscope photomicrograph of a xanthophore on a dorsal scale of a swordtail (Golden variety), showing indistinct cell margin and pterinosomes (arrowheads) in the cell processes (P). Scale bar, 10 um. Fic. 2. Low magnification differential interference contrast photomicrograph of a dorsal scale from a swordtail (Golden variety), depicting interlinking processes (arrowheads) between a xanthophore (X) and adjacent xantho-erythrophores (XE). Scale bar, 50 um. Fic. 3a. Low magnification confocal laser scanning image of three xanthophores on a dorsal scale of a swordtail (Neon variety), showing autofluorescent pterinosomes (arrows) and non-autofluorescent nucleus (N). Scale bar, 50 um. Fic. 3b. High magnification view of the outlined region in Fig. 3a, illustrating uniformly intense autofluorescent pterinosomes concentrated in high densities around the cell nuclei (N) and scattered ones in lower densities (arrows) in the dendritic cell processes. Scale bar, 25 um. Fic. 4. Confocal laser scanning images of Z-Series optical tomographic sections (a-h) at 1.5 ~m depth intervals of a xanthophore on a ventral scale of a swordtail (Golden variety). Scale bar, 50 um. 667 Confocal Laser Scanning of Xanthophores 668 G. Kuoo, V. P. E. PHANG AND T. M. Lim under bright-field light microscopy due to a homogeneously distributed dense central yellow pigment content and diffused cell margin. At high magnifications (Fig. 1) and under DIC (Fig. 2), circular carotenoid vesicles and pterinosomes were observed to be scattered in the extended cell processes. These pterinosomes were observed by Matsumoto [10] to be organelles possessing a three-layered limiting membrane and inner lamel- lae which appeared whorl-like due to a concentric arrangement of parallel membranes. The dense yellow pigment in xanthophores was described by Goodrich et al. [8] and Matsumoto [10] as carotenoid pigments. The CLSM proved immensely useful as xantho- phores were autofluorescent (Fig.3a) when scanned with the argon ion laser beam [14, 15]. The capacity for autofluorescence confirmed Mat- sumoto’s [10], Matsumoto and Obika’s [11], and Valenti’s [12] findings that visible colored and colorless fluorescent pteridines were present in these cells. According to Matsumoto [10], xantho- phores lack brightly-colored pteridines while their colorless pteridine content varies both in quality and quantity. Consequently, colorless pteridines such as xanthopterin, isoxanthopterin and biopterin [9, 10-12], and colored ones like yellow sepiapterin [13] might be the primary source of autofluorescence in xanthophores. Con- versely, carotenoids such as lutein, zeaxanthin and B-carotene, studied by Goodrich et al. [8], Valenti [12] and Rempeters et al. [16] in X. helleri and X. maculatus, were not thought to contribute to autofluorescence in these cells. The actual shapes of xanthophores can be elucidated from the distribution and arrangement of autofluorescent pterinosomes in the cells (Fig. 3a). High densities of pterinosomes caused intense autofluorescence at the cell center as opposed to the scattered distribution of autofluorescent pterinosomes in the cell processes [10, 11]. The CLSM micrograph showed xanthophores having numerous fine dendritic processes unlike the dif- fused cell outline produced by light microscopy (Fig. 1). Intracellular organelles such as nuclei were non-autofluorescent, thus appearing dark and oval-shaped in contrast to the autofluorescent pterinosomes in the cytoplasm (Figs. 3a, b). Sec- tions of xanthophores could be selected and highly magnified with the CLSM computer software for precise and detailed intracellular studies (Fig. 3b). Furthermore, dimensions such as length and width of cells or sub-cellular structures could be easily measured using this software. Swordtail xanthophores were commonly observed to form interlinkages through the fusion of their long dendritic processes to adjacent xanthophores and xantho-erythrophores. These tip-to-tip interconnections between the two chro- matophore types were best observed with the DIC attachment (Fig. 2) as the CLSM depicted overlap- ping and interconnecting autofluorescent processes as being similar. Interlinking xanthophores were not shown with light microscopy. Z-Series optical tomographic sectioning of xanthophores on fresh scales with the CLSM was performed without the laborious tasks of proces- sing and microtome sectioning of scales. Serial optical sectioning at a preset thickness (1-2 um) permitted the observation of cells at consecutive intracellular levels. Fig. 4 depicted a xanthophore scanned at eight successive levels. From these images, xanthophores could be made out as flattened cells with long, fine dendritic processes projecting in all directions from the cell body. Hence, three-dimensional cell models could be reconstructed from the serial sections produced by the CLSM. In conclusion, we recommend the use of the CLSM to complement conventional bright-field light and DIC microscopy techniques for fast and comprehensive morphological. studies of fish xanthophores and other autofluorescent cells. ACKNOWLEDGMENTS This study was supported by a grant from the National Science and Technology Board of Singapore. Sincere thanks to Mr. H. K. Yip for developing and printing the photographs. REFERENCES 1 White, J.G., Amos, W. B. and Fordham, M. (1987) J. Cell Biol., 105: 41-48. 2 Robinson, J. M. (1990) J. Histochem. Cytochem., Confocal Laser Scanning of Xanthophores 669 38: 1080. Wright, S. J., Schatten, H. and Schatten, G. (1989) J. Cell Biol., 109: 308a. Shotton, D. and White, N. (1989) Trends Biochem. Sci., 14: 435-439. Takamatsu, T. and Fujita, S. (1988) J. Microsc., 149: 167-174. Fujii, R. (1969) In “Fish Physiology”. Ed. by W. S. Hoar and D. J. Randall, Academic Press, New York, Vol. III, pp. 307-353. Fujii, R. and Oshima, N. (1986) Zool. Sci., 3: 13- 47. Goodrich, H. B., Hill, G. A. and Arrick M: S. (1941) Genetics, 26: 573-586. Bagnara, J. T. and Hadley M. E. (1973) In “Chromatophores and Color Change”. Prentice Hall, New Jersey, pp. 6-26. 10 11 2 113) 14 5 16 Matsumoto, J. (1965) J. Cell Biol., 27: 493-504. Matsumoto, J. and Obika, M. (1968) J. Cell Biol., 39: 233-250. Valenti, R. J. (1972) Ph. D Dissertation, New York University, New York, USA. Hama, T. (1963) Ann. N. Y. Acad. Sci., 100: 977- 986. Chanky Seen Lhangoey. weave. ang cim= aleve (1990a) Res. Trends, 2: 7-8. Chan, S.-Y, Phang V. P. E. and lam, T. M. (1990b) In “Essays in Zoology: Papers Commemor- ating the 40th Anniversary of the Department of Zoology, National University of Singapore”. Ed. by L. M. Chou and P. K. L. Ng, National University of Singapore, Singapore, pp. 255-263. Rempeters, G., Henze, M. and Anders, F. (1981) Comp. Biochem. Physiol. 69B: 91-98. OS de x aera sy S a aN ae | a # > 7

4 5 Sr ? ‘ i +> f z . 7 J = 7 w ‘ ae ie t 3 ty F =~ at of A> 4 eS > #= « - a in ot of a” ' ry ~ é tl UM ‘ F ‘ 5 i xe _ t ) oe ] ct Bae cd ‘ % a 4 1 S y ) 7 i . » Eid z rl a % ¢ _ 4 = ' r ~ fae ety ee a 7 a ? 2 Pe oe y < # e ee a iw " ~* BY + ~ iz i < ss a 7 “ ’ me f . a = . e " E . Le - c. Tod i: — 4 > Bs ~ i “ 2 x ¥ i _ = f ~ j ye e : é. P - f a Ei ie + iv 2 ni 3 ns 4 x | Pirnenisr. = erate: x4 ad ak as 4 aes i ae car a Sa * ° ets yi a wy fomet aa] eS oa ot eS $. J aaah ene dit OF sah api you mn J ane oh ed Reigns, = ae 2) wre tue, i. inv pei a3 ii ro Sp se” ZOOLOGICAL SCIENCE 9: 671-674 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Phospholipids and Fatty Acids in Intact and Regenerating Dugesia anceps, a Fresh Water Planaria Luis E. Pouiti, SILVIA VALENTINUZzI de SANTos! and LILIANA VERGARA de LINARES* Instituto de Investigaciones Bioquimicas; ‘*Departmento de Biologia y Bioquimica; Universidad Nacional del Sur, 8000 Bahia Blanca, Bs. As., Argentina ABSTRACT—Phospholipids and their fatty acid com- position in intact and regenerating Dugesia anceps, a fresh water planaria, were determined. Phospholipids and their percentual distribution in whole animals, in regenerating trunks and heads were similar. A decrease in the percentual contribution of several unsaturated fatty acids in phospholipids of regenerating animals was observed. INTRODUCTION The regenerative power of fresh water planar- ians has been known for more than 150 years [1]. Some morphological and cytological aspects of regeneration in these animals are well known. The process has been studied morphologically at the ultrastructural level [2]. From a biochemical point of view, the role of dopamine, serotonin and cAMP in regeneration has been determined [3], but no similar studies were conducted on lipids. This work is centered on the phospholipids content and their fatty acid composition in intact and regenerating D. anceps, a fresh water planaria; to our knowledge, this is the first report on this subject. Accepted March 6, 1992 Received December 9, 1991 ' To whom all correspondence should be addressed. MATERIALS AND METHODS Animals The animals used were collected in Arroyo Naposta, a stream in the vicinity of Bahia Blanca. The determination of the species was reported in [4]. The specimens were kept in glass containers filled with filtered water from the same stream at room temperature [5]. They were fed fresh raw bovine liver and subsequently fasted for six days before use. Regeneration For regeneration experiments, the heads of a set of about 120: animals were amputated just before the auricles; several subsets of heads and trunks were kept in separate contain- ers filled with food-free filtered stream water. For analytical purposes they were retrieved after varying intervals (from four to five days in the case of heads and four to seven days in the case of trunks) but always before completion of regenera- tion. Intact planarians were simultaneously kept to be used as control. The control was performed either with whole animals or with heads and trunks sectioned just before homogenizing. Lipid extraction After several washings with distilled water, planarians, whether control or regenerating, were transferred to a mortar. Lipids were extracted by homogenizing the tissue with 3 ml chloroform/methanol (2:1, v/v) [6]. The crude 672 L. E. Pouiti, S. V. de SANTOS AND L. V. de LINARES total lipid was washed with CaCl, 0.05%, dried under N> and resuspended in a small volume of solvent. Samples were stored at —80°C under Np. Phosphoglyceride separation Phospholipids were isolated by two dimensional chromatography in 250 wm thick silica gel H plates (Merck-Sharp and Domme) following the procedure of Rouser et al. [7]. After development with either I, vapor or dichlorofluorescein (0.2%) in methanol, phospho- lipids were scraped into screw-capped tubes. Phospholipid phosphorus determination Phos- phorus was measured by the procedure of Rouser et al. [7| after perchloric acid digestion. Methanolysis and gas liquid chromatography It was done with BF3 in methanol according to Morrison and Smith [8]. Fatty acid methyl esters were analyzed in a model 3700 Varian gas liquid chromatographer, equipped with a hydrogen flame detector; a stainless steel column (2 mX2 mm ID) packed with OV 275 on N> carrier was used. Samples were run with temperature programmed between 160°C and 220°, 5°C/min, with the injec- tor and the detector maintained at 220°C and 240°C, respectively. The measurement of peak areas was done with an automatic integrator (CDS-111) attached to the GLC. Fatty acids were identified by comparing their retention times with those of standard methy] esters. Quantitative Analysis Quantitative determina- tions were made in triplicate samples from at least 30 trunks, heads or whole planarians and are expressed as mean + standard deviations. Signi- ficance was determined by using the two tailed Student’s f-test. Unsaturation index This index is defined as the percentage of the sum of the product of the number of double bonds in each fatty acid methyl ester times its percentage. RESULTS AND DISCUSSION Phospholipid composition of intact and regenerat- ing planarian tissues The phospholipid composition of regenerating heads and trunks and of control animals is shown in Table 1. No significant differences could be found between them. PC and PE amount to more than 80% of the total, acidic phospholipids (PS and PI) to about 10%, the precursor PA to less than 1% and phosphonolipids to approximately UY. Although some mechanisms controlling regen- eration in planaria have been reported [3], the possible, if any, role of lipids during that process is unknown. The results shown in Table 1 indicate that no differences in lipid composition exist between the regenerate and the old tissue. TABLE 1. Phospholipid composition of intact and regenerating planarian tissues Phospho Heads lipids C R EG SN).jsae2,1I 52.0+3.8 PE 34.9+1.5 35.0+0.8 PS TeSieraley Tac A PI s).oac I) San lke7 PA OBE0 0.4+0.0 Cl Osa 0 O22 0% L 0.0+0.0 OnEOst PNL 2.4+0.8 1.8+1.4 Trunks eee G R ti@ 55.8+8.0 53.4+9.2 49.0+3.5 33202279 32.0+6.6 36.6+4.5 4.54+3.1 Dae LO) 6.6+0.7 4.2+1.4 4.4+1.1 4.5+1.6 0.4+0.1 0.8+0.6 0.3+0.2 0). 2ac Oo 0.6+0.6 0.7+0.4 0.0+0.0 0) ar. Il OEEOD 1ESEE IEG S10 aievlem DASEOM PC, Phosphatidylcholine; PE, Phosphatidylethanolamine; PS, phosphatidylserine; PI, Phosphatidyl- inositol; PA, Phosphatidic acid; CL, Cardiolipin; L, Lysophospholipids; PNL, phosphonolipids; C, Control; R, Regenerating. Mean percentages +SD of total phospholipid phosphorous from 3 samples of 30-120 pooled planarians. Phospholipids in Fresh Water Planaria 673 TABLE 2. Fatty acid distribution in phospholipids of control and 4-day regenerating planarian trunks PA € ihe OR Goma SR Casio R euvenhudn 14:0 D220) S200 02202" OS02 —One00 sO Moker 15:0 13402 O2201 092028 “US208) 00400 02202 Otis “eee 15:1 OSeOot HO4OUL Mos s05en Os01e O1e0l O2201 O1201. O1201 PO Gots 180209 113407 8124407 26406. 3.8405 1.0405 2640.6" 16:1 EOD 10 0S To 0O | 46206 120s 14406 0.6203 07203 PET-H08 TILSA00) toe 19523 — 4010430) 442409 425405 45.243.1 igeeeneS 4-0 Ne 35164-0008 14010143421. 1220 150415: 190226 21.7230 18:2 ABLDO AOLDS “AVEO 92 6R0G — 20206) 29206 > S602 Oe eee 20:1 (AOU SOLO 4UN 2GELOMNe 08203" 22404 06403" 15404 06-03" COrenomD 0M 13406 foeoo 16402 12404 11208 15404 09403 MPO OG 03114 952406 62405 98412 115209 7540.6 10641.5** POPnee 0-203. 1.903% 140.4 12403 923405 44204 25403 1.1403" esas 4 113408 26.0409 26.9425 182409 105+0.8** 43403* 20205" PEC GEOG 6 224046 «6020549 T4412 5.6202 52405 21403 ..5.440.7% ene 02) 0st Oe LOO 14+05 O7404 07403 05402 0.6403 The data represent the results of three samples of 30-120 pooled trunks. Other abbreviations as in Table 1. the methyl end of the acid chain. The n indicates the double bonds from * | statistically significant difference (P< 0.05); **, statistically very significant difference (P<0.01). The presence of phosphonolipids has been reported in ciliates, Cnidaria [9] and Mollusca [10]. In all these groups. as in planaria, epidermal cells are not protected by a cuticula and they are directly exposed to the medium [11]. The chemical properties of phosphonolipids could help these animals to cope with environmental changes [11, 12]. Conversely, phosphonolipids are absent in Taenia {13] which although being also a member of the phylum Platyhelmintes, is protected from external damage by a very active syncyctial epidermis. SPH was not present although large amounts of it have been reported in several invertebrate species [14, 15]. Fatty acid composition and changes during regen- eration Information about fatty acid composition of phospholipids in invertebrates has been growing in the last decade [16]. The activity of protein kinase C, apparently involved in planarian regeneration, as well as some functional properties, like for instance membrane fluidity, depend on that com- position [3, 17, 18]. Fatty acid composition of PC, PE and acidic phospholipids was compared in control trunks and in regenerating trunks in the fourth day of the process (Table 2). Saturated fatty acids amount approximately to 30% in PC and PE and to 44% in PS and PI. At that stage of the regeneration process, phospholipids undergo changes in their percentual composition. The table is basically a 15 x8 matrix showing the 60 possible differences between con- trol and regenerating animals. Four of the differ- ences were found to be statistically significant (P< 0.05) and ten statistically very significant (P< 0.01). Two of them are increases (both occurring in PI) and twelve are decreases. These changes caused the decrease in the unsaturation index in PS and PI from 2% to 1.6% and from 1.6% to 1.3%, respectively. The more important variations are thouse of 20:4n6 in PI and 22:5n3 in PS, since these fatty acids are major constituents of these phospholipids. 20:5 has been found in most phospholipids in invertebrates [15, 16, 19]. In agreement with these reports, it was consistently found here in all phospolipids analyzed, although it represented only a small fraction. Curiously, this fatty acid 674 along with other unsaturated fatty acids decreased its percentual contribution in several phospho- lipids during regeneration. The polyene 22 :5n3 in PI also follows the general pattern although its decrease is partially compensated by an increase in 22:5n6. The percentrual participation of 16:0 and 22:5n6, which increased in PI during regen- eration, is at odds with the pattern found in the rest of the fatty acids. The fact that many variations were statistically non significant does not allow to conjecture about the fate of the fatty acids whose contribution diminished during regen- eration. ACKNOWLEDGMENTS The authors are grateful to Dr. F. Barrantes and to Dr. M. I. Aveldano for helpful suggestions. REFERENCES 1 Brondsted, H. V. (1955) Biol. Rev., 30: 65-126. Hori, I. (1989) J. Submicrosc. Pathol., 21(2): 307- 315). 3. Martelly, I. and Franquinet R. (1984) Trends Biochem. Sci., 9: 468-471. 4 Cazzaniga, N. and Curino, A. (1987) Boll. Zool., 54: 141-146. 5 Lutz, F., Wich P., Galtsoff, P. and Needham, J. (1959) In “Culture Methods for Invertebrate Anim- als”. Dover Publications Inc., N. York., pp. 154- 156: 6 10 11 12 13 14 15 16 IW, 18 19 L. E. Pouiti, S. V. de SANTOS AND L. V. de LINARES Folch, J., Lees, M. B. and Sloane-Stanley, G. H. (1957) J. Biol. Chem., 193: 497-509. Rouser, G., Fleischer, S. and Yamamoto, A. (1970) Lipids, 5: 494-496. Morrison, W. R. and Smith, L. M. (1964) J. Lipid Res., 5: 600-608. Rosenberg, H. (1973) In “Phosphonolipids. Form and Function of Phosphonolipids”. Ed. by G. B. Ansell, R. M. Dawson and J. N. Hawthorn, Elservier, BBA Library, Vol. 3, pp. 333-344. Itasaka, O., Hori, T. and Sugita M. (1969) Biochem. Biophys. Acta, 176: 783-788. Florin-Christensen, J., Florin-Christensen, M., Knudsen, J. and Rasmussen, L. (1986a) trends Biochem. Sci., 129: 354-355. Florin-Christensen, J., Florin-Christensen, M., Ras- mussen, L. and Knudsen, J. (1986b) Comp. Biochem. Physiol., 85B: 143-148. Politi, L. E., Santos, S. V., Linares, L. V. (1984) Res. VII Jorn. Arg. Zool., 21-26 Oct. 1984, Mar del Plata, pp. 220. Kaneda, Y., Nagakura, K. and Goutsu, T. (1986) Comp. Biochem., Physiol., 83B: 533-536. Chapelle, S. (1986) Comp. Biochem. Physiol., 84B: 423-439. Isay, S. V. and Busarova N. G. (1984) Comp. Biochem. Physiol., 77B, (4): 803-810. Martelly, I., Moraczewski, J., Franquinet, R. and Castagna, M. (1987) Comp. Biochem. Physiol., 86B: 405-409. Bell, M. V. and Sargent, J. R., (1987) Comp. Biochem., Physiol., 86B (2): 227. Takagi, T., Kaneniwa, M. and Itabashi, Y. (1986) Lipids : 430-43. ZOOLOGICAL SCIENCE 9: 675-677 (1992) [COMMUNICATION] © 1992 Zoological Society of Japan Flight Behaviour and Thyroid Hormone Regulation in Homing Pigeons JOHN C. GEORGE and T. MATHEW JOHN Department of Zoology, University of Guelph, Guelph, Ontario, NIG 2W1, Canada ABSTRACT—Homing pigeons which were not given flight training for 3 months prior to the experiment, were flown the same distance of 48 km from the usual release site as reported in our earlier studies using pigeons which were in regular training. In these pigeons the flight lasted 90-160 min instead of the usual 60-80 min taken by pigeons which had regular training. This flight produced a more than two-fold increase in plasma levels of reverse triiodothyronine (rT3), concomitant with reductions in thyroxine (T4) and triiodothyronine (T3) levels and also in T3/T, ratio. The increase in plasma levels of rT3 and the concomitant decrease in levels of T, and T3 with no change in plasma osmolality, suggest inhibition of T, secretion and 5 -monodeiodination, and conversion of T, to rIl3. The conversion of T, to rT3, the inactive form of T3, represents a mechanism of autoregulation of thyroid hormone function during strenuous and extended flight. INTRODUCTION Our recent studies with homing pigeons before and after natural homing flight, have shown significant post-flight (a distance of 48 km from the usual release site covered in 60-80 min) increases in plasma levels of glucose, free fatty acids (FFA), lactate, adrenaline, noradrenaline and growth hormone (GH), but not in the levels of corticoster- one and the thyroid hormones, thyroxine (T4) and triiodothyronine (T3) [1, 2]. The increase in plasma adrenaline and noradrenaline was indica- tive of increased sympathetic activity and it was Suggested that the flight-induced increase in plasma adrenaline could have stimulated the re- lease of glucagon which in turn would have Accepted April 8, 1992 Received November 30, 1991 brought about the increase in plasma glucose. The increase in plasma FFA was attributed to the increase in at least one adipokinetic hormone, GH, and the lack of any increase in plasma corticosterone to the stress-free nature of the flight. In a subsequent study [3], the homing pigeons used, unlike those in the previous studies, did not receive the regular flight-training for a period of 3 months prior to the experiment, and so they took 90-160 min to fly the same distance. Significant increases in the levels of plasma glucagon (gluca- gonlike immunoreactivity) and presumably of glu- cagon-stimulated FFA, were observed. In marked contrast to the observations in the previous study [2], plasma levels of Ty, T3; and T3/T, ratio were significantly reduced. It occurred to us that this was due to the possible inhibition of T,4 secretion and 5’monodeiodination with the conversion of T, to reverse T; (3, 3’, 5’-triiodothyronine or rT3), the inactive form of T3, as a mechanism for the regulation of thyroid hormone metabolism during a more strenuous and extended flight. In the present study, we have ventured to test this postulation by measuring levels of rT; in plasma samples obtained from the same birds used in the previous experiment [3]. MATERIALS AND METHODS Pigeons (Columba livia) used in our studies were from a colony of homing pigeons which were raised and maintained out-doors in lofts under natural photoperiod and temperature. They were 676 J. C. GEORGE AND T. M. JoHN fed a daily ration of corn, wheat and barley. All studies [1-3] were conducted in the forenoon of a typical sunny mid-autumn day (6C). The release site and distance flown (48 km) were the same in all studies. The control birds were given the usual 40 min car ride in order to simulate the 40 min car ride received by the experimental birds while being taken to the release site. Pigeons of both sexes weighing 350-400 g were used and blood samples (5 ml) were drawn from the brachial vein of each bird into heparizined syringes within 1-3 min of their arrival after the car ride (control birds) or flight (experimental birds). Blood samples were kept on ice and transported to the laboratory within 15 min and centrifuged (3000 X g for 10 min at 4°C). The separated plasma was stored at —20°C in separate vials in duplicate. The present investigation is an extension of our previous study [3] and parts of plasma sampels obtained then are now used for the estimation of plasma rT3. In contrast to the previous studies [1, 2], these pigeons [3] had not received the regular flight training for a 3-month period between end of the racing season and time of the experiment. The levels of rT3 were measured in freshly thawed frozen plasma samples using a radioimmu- noassay kit (code 10834) manufactured by BIODATA S.p.A., Italy. Osmolality was meas- ured using a vapor pressure osmometer (Wescor, Utah). Analysis of variance (ANOVA) was initially employed to test for sex differences and flight effect. Since sex did not prove to be a significant variable from the ANOVA results obtained, values from both sexes were subsequently pooled and subjected to unpaired Student’s f-test. RESULTS These pigeons, unlike the pigeons which re- ceived flight training prior to the experiment, took 90-160 min to fly “home” instead of the 60-80 min taken by the trained pigeons. Flight induced significant increases in plasma levels of rT3 as opposed to decreases in Ty, T3 levels, and T3/T, ratio. The post-flight plasma rT; amounted to more than a two-fold increase over control values. Flight caused no significant change in plasma osmolality (Table 1). DISCUSSION The marked increase in plamsa levels of rT3 with no significant change in plasma osmolality, observed in the present study is indicative of increased conversion of T, to rT3. Similar increase in plasma rT; as a possible regulatory mechanism to limit T3 activity by inhibiting T, conversion to T; has been observed in humans under increased physical exercise [4]. In flown homing pigeons the increase in rI3 and the concomitant reduction of plasma levels of T,, T3 and T3/T, ratio (Table 1) indicate inhibition of peripheral deiodination of T, in order to limit the continued production of T3. The peripheral conversion of T, to T3 has been shown to be stimulated by GH in chickens [5]. In TABLE 1. Plasma osmolality and levels of thyroid hormones in resting and flown homing pigeons Control pigeons Osmolality (mmol/kg) SB.aae iil CO) Reverse triiodothyronine (rT3) JA Dae WG.) (CO) (pg/ml) Triiodothyronine (T3)' 2.29+0.14 (10) (ng/ml) Thyroxine (T,)' 19.07+1.54 (10) (ng/ml) dt de ratio: 0.12+0.10 (10) Flown pigeons 306.1+ 3.8 (7) 493.0+90.8 (7)** 0.89+0.14 (8)* 12.54+2.36 (8)* 0.08+0.01 (8)** Values are mean+SEM. Figures in parentheses denote number of birds. Je Re JEON ' Data from previous study (George ef al., 1989) Homing Pigeons: Thyroid and Flight 677 an earlier study using trained homing pigeons, post-flight circulating levels of GH were found to be significantly increased [2]. In more strenuous and extended flight such as was involved in the present investigation, an initial stimulation of Ty release followed by an increase in rT3 levels should be expected. If so, the production of rT3 could not be concomitant but should follow the release of T, so that excess T, could be eliminated by convertion to rI3. That this is so, has been indicated in experiments with tilapia [6] in which it was observed that rT3 levels were the same as the low levels present in the control one hour after injection of T, despite the high concentrations of T, in the plasma. Since plasma Ty, and rT; increased following injection of Ty, it was sug- gested that conversion of T, into rT3 is indepen- dent of pituitary control. In light of these observa- tions, it may be stated that the post-flight increase in plasma levels of GH observed in homing pigeons [2] could stimulate T, release in addition to releasing FFA from the fat depots. It is also possible that the inhibition of peripheral deiodina- tion of T, to T3 could have been brought about by the increased plasma levels of glucagon since it has been shown in the domestic fowl that glucagon inhibits 5’monodeiodination and may also cause initial reduction of Ty secretion [7]. Rudas and Pethes [8] observed that rT3 appears in the serum of chickens after warm exposure and suggested that cold exposure stimulates T3 forma- tion whereas heat exposure inactivates the T, secreted to produce rT3. During flight there is significant increase in body temperature of pigeons [9]. As flight becomes more strenuous and ex- tended as observed in the present study, thermo- regulation becomes crucial. Conversion of T4 to rT3 instead of T3; would be a useful mechanism of thermoregulation and conservation of thyroid me- tabolism. Since familiarity with the release site has been shown to reduce “release site bias”, a behaviour characterized by deviation in the direction and better orientation of homeward flight [10-13], the longer time (90-160 min) taken by these pigeons could be attributed to the lack of the flight training prior to the experiment. ACKNOWLEDGMENTS Financial support for this work was obtained from the Natural Sciences and Engineering Research Council of Canada awarded to J. C. George. This work would not have been possible had it not been for Bert DeRidder who very generously made his homing pigeons available to us and also assisted us in transporting them to their release site. Thanks are also due to Mike Vanderjagt for his unstinted help in the field, to J. R. Geraci for kindly providing access to his laboratory where the hormonal assay was done, to M. Vijayan for his valuable coopera- tion and to M. Patterson for technical assistance. REFERENCES 1 Viswanathan, M., John, T. M., George, J. C., and Etches, R. J. (1987) Horm. Metab. Res. 19: 400- 402. 2 John, T. M., Viswanathan, M., George, J. C., and Scanes, C. G. (1988) Horm. Metabol. Res. 20: 271- DIS 3 George, J. C., John, T. M. and Mitchell, M. A. (1989) Horm. Metabol. Res. 21: 542-545. 4 Premachandra, B. N., Winder, W. W., Hickson, R., Lang, S., and Holloszy, J. O. (1981) Eur. J. Appl. Physiol. 47: 281-288. 5 Kihn, E. R., Verheyen, G., Chiasson, R. B., Huts, C., Huybrechts, L., Van den Steen, P., and Decuy- pere, E. (1987) Horm. Metab. Res. 19: 304-308. 6 Byamungu, N., Corneillie, S., Mol, K., Darras, V., and Kuhn, E. R. (1990) Gen. Comp. Endocrinol. 80: 33-40. 7 Mitchell, M. A., and Raza, A. (1986) Comp. Biochem. Physiol. 85A(2): 217-223. 8 Rudas, P., and Pethes, G. (1984) Comp. Biochem. Physiol. 77A: 567-571. 9 Aulie, A. (1971) Comp. Biochem. Physiol. 38: 173- 176. 10 Keeton, W. T. (1973) J. Comp. Physiol. 86: 1-16. 11 Keeton, W. T. (1974) Adv. Study Behav. 5: 47-132. 12 Walraff, H. G. (1978) In “Animal Migration, Navigation and Homing”, Ed. by K. Schmidt- Koenig and W. T. Keeton, Springer-Verlag, Berlin, Heidelberg, New York. pp. 171-183. 13. Kowalski, U., and Wiltschko, R. (1987) J. Exp. Biol. 133: 457-462. si ay = , tay * ee " ea My ¥ hy = f oe les | f peasy suodhtcn 4 iv sep! sunny mid-aupsan ‘ey . “pite aad: baer Rewri 4s a ey ei studies. | Tee eee | | edema eee ‘wwe me x i a a?” Jon bhtow sow’ aa suis a) Babe ne ft > Nob bi Matt eer 101 ne a Py i. we “Gas cee on | Te HAE, arse ane ‘Sart Binet ay Pay ite Mbnorte! nfiyegri (egies AP Bers saci’ Set ‘al =i . Wa, } 4g eh } Hn ae’ aah hes acai vet! r é ee settee “suai iteeoortaithen cis 1a oF i; yh a de esl Roy aes a bene Misia Cones aes i fs ial aad bas | Bad thapen-crieisulgie on : E ve specsgsmpeey: 5 itt 5 : - ; Cees = ? Ns > * ; ss : f oie as PRs 5 Del, Chiehs DeapeRE eee iy cist a : F : fa ¥ ial oe t= - H : cy ‘ ' : a = b ¥. ey Oe “a . hs rr ; i] La ” howe * : a r - % N Va j * " a ee / j 4 ea si ' i { FI : = oie « BY i ¢ Ct ey aye. et von 3 ri cone Fy Fart As = * ‘ 4 rat bs Bane a 7 * i oe 5 = y | AR y Se, 4 é i j a oe t 7 ve 5 : ‘ ss co att P vibe ips bey ae 7 i ; : Pubs ToD nd AE Litt. (ope i : , ‘ t i ; xa ’ a, 4 é s- bg Te ‘ i ; i i We 4 ‘ , tt ~ ? s/h in , = c : Sr ye vi ey. he 4 oe . ; we = Ki ORAS Be ' Sin a é pis } ria) } I Rs ; i Ve Mee ' ' x ~ i Te _ st 1 met ’ rs { 7 ahs SPAS al ‘ / . E 1 . svn roe | ee: Y ~ \ a en ; i PANE ; A er nT Rae ie * f y : : Sor rite La 4 , ap Sy ~ : i $5 é i" Bey Yee te ra EE ee. patos 2 4 « » = ‘ i 6 ie ee rs 1 r im 7 ys i i "I \ — <4 7 z i . — Ai te t a : ye na E \ Bt i t : A } . 3 ( : in the reproduction of female crested newt, Triturus carnifex Tsutsui, K., S. Kawashima, V. L. Saxena and A. K. Saxena: Binding properties and photoperiodic influence of follicle-stimu- lating hormone receptors in the subtropical wild quail George, J. C. and T. M. John: Flight be- haviour and thyroid hormone regulation in homing pigeons (COMMUNICATION) Ecology Smith, J. I. and H. Yu: The association be- tween vocal characteristics and habitat type Ie ANES EHD ASSEMIMESS soniye gee ase 659 ZOOLOGICAL SCIENCE VOLUME 9 NUMBER 3 JUNE 1992 CONTENTS REVIEWS Nagai, Y.: mammals Primary sex determination in Nunomura, W.: C-reactive protein (CRP) in animals: its chemical properties and biologi- cal functions ORIGINAL PAPERS Physiology Kanzaki, R., N. Sugi and T. Shibuya: Self- generated zigzag turning of Bombyx mori males during pheromone-mediated upwind walking Matsuoka, T. and K. Taneda: Step-up and step-down photoresponses in Blepharisma Griffond, B., J. Van Minnen and C. Colard: Distribution of APGWa-immunoreactive substances in the central nervous system and reproductive apparatus of Helix aspersa Cell Biology Suzuki, T. and S. Funakoshi: fibronectin-like molecule from a marine Isolation of a bivalve, Pinctada fucata, and its secretion by amebocytes Khoo, G., V. P. E. Phang and T. M. Lim: The confocal laser scanning microscope as a tool for studying xanthophores of the sword- tail (Xiphophorus helleri) (COMMUNICA- TION) Immunology Zhang, H., T. Sawada, E. L. Cooper and S. Tomonaga: Electron microscopic analysis of tunicate (Halocynthia roretzi) hemocytes Biochemistry Pokti, E. L., S. V. de Santos-and IL. V>@e Linares: Phospholipids and fatty acids in intact and regenerating Dugesia anceps, a fresh water planaria (COMMUNICATION) Mita, M.: Diacyl choline phosphoglyceride: the endogenous substrate for energy me- tabolism in sea urchin spermatozoa Developmental Biology Makabe, K. W., S. Fujiwara, H. Nishida and N. Satoh: Failure of muscle myosin heavy- chain gene expression in quarter ascidian embryos developed from the secondary mus- cle lineage cells Kurabuchi, S. and Y. Kishida: Effect of delay in anterior or posterior amputation on regen- eration of short fragments of planaria ....575 © Roudebush, W. E. and J. G. Kim: Mouse embryo biopsy: abnormal development with trophoblastic vesicle formation : Iwamatsu, T.: Morphology of filaments on the chorion of oocytes and eggs in the meda- Ka 2208). cee heel.) eee 589 Reproductive Biology Ishijima, S. A., M. Okuno, Y. Nakagome, H. Odagiri, T. Mohri and H. Mohri: Separa- tion of x- and y-chromosome-bearing murine sperm by free-flow electrophoresis: evalu- ation of separation using PCR Endocrinology Zairin, M. Jr., K. Asahina, K. Furukawa and K. Aida: Plasma steroid hormone profiles (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 June 15 Printed by Daigaku Letterpress Co., Ltd., Hiroshima, Japan re \' sm ; eS ‘ i L. ‘ i Ss? é \ ¥ {