Ï. Okada & Y. Nagahama

IUBS

Edited by

ï. OKADA & Y. NAGAHAMA BIOTECHNOLOGY OF AQUATIC ANIMALS

Proceedings of an IUBS symposium organized on 25-27 November, 1991, in Toba City. Mie Prefecture, Japan

Edited by Tokindo Okada Biohistory Research haii. Murasakicho 1- 1. Takatsuki 569. Japan

Yoshitaka Nagahama National Institute for Basic Biology 38 Nishigonaka. Myodaijicho. Okasakf, 444 Japan

Special Issue-28 Biology International (C) 1993 The Intemational Union of Biological Sciences News Magazine Biology International (Special Issue No28 - 1993)

TABLE OF CONTENTS

Introduction, by T.S. Okada 5

A New Approach in Aquaculture: A Must for Feeding a Rapidly Increasing World Population and for Meeting the Ecological Demands of the 2 1st Century. by P.G.W.J. van Oordt 9

Biotechnology of Aquatic Animals: A New Frontier with Implications for Both Basic and Applied Research, by D.A. Powem. T.T. Chen. & RADunharn 17

Regulation of Oocyte Maturation in Aquatic Animais: The Comparative and Generai Aspects. by Y. Nagahama 27

Glycosphingolipids: Important Membrane Components Rather Neglected in Biotechnology, by Motonori Hoshi

Hatching Enzyme of Medaka: Molecular Aspects of Its Formation and Packaging in the Hatching Gland Cells, by K. Yamagami, S. Yasumasu, H. Shimada, & 1. Iuchi

Differential Response to Mutagenesis Arnong the Spermatogenic Stages of a Fish, the Japanese Medaka Oryzias h.flpes.by A. Shima & A. Shimada

Somatolactin, a New Member of the Growth Hormone and Prolactin Family from the Pars Intermedia of Teleost Fish, by Hiroshi Kawauchi cDNA Cloning and Structure of Teleost Growth Hormones and the Growth Promoting Activity of Recombinant Hormones, by K. Nakashima, M. Watahiki. & M. Tanaka

Genome Transfer in Teleosts, by Yan Shaoyi Biology International (Special Issue No 28 - 1993) Induced Fusion of Oocytes and Embryonic Cells, by ,S.G. Vassetzky. A.A. Bilinkis. G.G. Sekirina. & M.N. Skoblina 65

Direct Production of the Super Male 0by Androgenesis in Amago Salmon. by H. Onozato 69

Studies on Chromosome Manipulation in Cyprinid Loach (Misgurnus ang uillicaudatus), Cornmon Carp (Cyprlnus carpio), and Small Abalone (Haliotfs diversicolorj , by N.H. Chao. C.P. Yang, H.P. Tsat. W.H. Liang, &LC. Uao

The Development of Technologies for The Control and Detemination of Sex in Aquacultured Salmonids, by Edward M. Donaldson. Robert H. Deviin. Francesc Piferrer. & Igor 1. Solar

Transgenic Salmon with Enhanced Growth and Freeze Resistance, by C.L. Hew. SJ.Du. Z Gong. P.L. Davies. S.Y. Gauthier. MA. Shears. M.J. King, & G.L. Fletcher 87

Transgene Transmission and Expression in Rainbow Trout and Tilapia. by Norman Maclean, Arati Iyengar. Azfi Rahman. Zorah Sulaiman. & David Penman 95

Transgenic Technology in Fish, by Daniel Chourrout 99

Gene Transfer in Medaka, by K. Ozato 107

Summary, by C.L. Hew 111 Introduction by T.S. Okada Biohisto~yResearch Hall. Murasaki-cho 1-1. Takatsuid 569, Japan

This "Special Issue" of Biology International consists of the papers presented at the International Union of Biological Sciences' Symposium on Biotechnology in Aquatic Animals, which was held in Toba City. Mie Prefecture, Japan, on 25-27 November. 1991. Thanks to an enthusiastic collaboration of the speakers who willingly took the trouble to write the manuscripts as well as to a generous arrangement by the IUBS, this publication is very timely to promote the researches of the subject at the global level. It is particularly so, since an introduction of new technologies in this area is so swift and a request to gain knowledge of forefront is eager for the researchers working in both academic and practical fields throughout the world. In this introduction, 1 briefly summarize the background and specific aims of the Symposium. A short report on the Symposium has already appeared in Biology International NO24 in January, 1992.

NBS Programme on Reproductive Biology in Aquaculture (RBA)

Needless to Say. a number of devices in aquaculture have long benefited human life. Nevertheless, recent demand to utilize aquatic resources for biological production is more acute than before. On the scientific side, the development of technologies to be applied for aquaculture is somewhat revolutionary. Having recognized such a situation, the IUBS 23rd Generd Assembly (Canberra, Australia, 1988) accepted to look into a proposal made for an international project on reproductive biology in aquaculture as one of its Scientific Programmes. This was formally adopted at the Union's 24th General Assembly, held in Amsterdam. Netherlands, in 199 1.

The time between these two Assemblies was an important and necessary "prelude". We should like to refer to the hard work and leadership of Professor P.G.W.J. van Oordt (Utrecht, Netherlands) with great admiration. He made an extensive survey of active researchers in this subject al1 over the world, and succeeded in coordinating the international network for developing the programme. Professor van Oordt edited a "Special Issue" (N026) of Biology International entitled "Reproductive Biology in Aquaculture: A Proposal for an International Collaborative Programme of Research, which not only reported upon his energetic campaign, but also summarized the present status of basic research on the subject, including morphology, developmental biology, endocrinology, biochemistry, and others.

Before the officia1 start of the RBA programme, Our Taiwan colleagues volunteered to organize the IUBS international symposium on Reproductive Biology in Aquaculture. in Aprii. 1991, in Taipei. A volume of the proceedings from this successful meeting is now in print (Bulletin of the Institute of Zoology, Academia Sinica, Monograph 16). AU these endeavours revealed a Bwlogy Internationai (Special Issue N28 -1993) keen interest on this subject at the international level with high expectations for the present programme.

Introduction of Molecular and Cellular Techniques into RBA Studies

A recent introduction of techniques of cell and gene manipulation to the study of aquaculture looks to somehow revolutionize the studies in this field. The technical advancement to uWe these approaches in various aquatic animais is rapid. more rapid than was anticipated five or ten years ago. It promises new perspectives in the studies of reproduction of aquatic anirnals.

An introduction to these techniques was achieved earlier in the field of reproductive and developmental biology of terrestrial plants. It seems to alter the basic pattern of agriculture which human beings have continued to develop ever since the initiation of civilization. Why do we not expect a similar situation in aquaculture? Soon, we can "design" more profitable aquatic animals for human life and well reproduce them in the near future. In fact. the first step toward such a goal has been successfully started as reported in some of the papers in this volume. A rise of "transgenic fish is undoubtedly one key subject of the area. Readers may refer also to an excellent review by C.L. Hew and 2. Gong (Biology International N0Z4). A manipulation of ce11 and chromosomes is no less important. of course, in parallel with some developmental biological technologies like a production of chimerae in aquatic animais.

Several small fish like zebrafish, medaka, and others are becoming indispensable laboratory materials to serve as mode1 systems for understanding the basic mechanisms of development in vertebrates. In fact, al1 modern technologies are now easily adoptable to these aquatic organisms.

After having observed these circumstances in the present-day studies of RBA. the organizers of the IUBS Toba Symposium decided to make biotechnology as applied to fish as a core, without eliminating important topics on endocrinology, biochemistry, and others.

NBS Toba Symposium

In relation to a nature of tle subject to be discussed, we thought it more appropriate to choose a place dong the seaside, rather than cities like Tokyo or Osaka. The city of Toba is located in the central part of Japan facing the Ise Bay, and long enjoys its farne as one of the uniquely traditional centers of aquaculture in Japan, particularly the cultivation of pearls.

The Symposium, held on 25-27 November. 1991, was organized by the National Institute for Basic Biology, Okazaki. Japan, in cooperation with the Japanese IUBS National Committee of Biological Sciences of the Science Council of Japan. Professor Y. Nagaharna and 1 had the pleasure of chairing the Bwlogy Irilcrnatio~l (Spccial Is~vcN78 -1993) organizing cornmittee. Participants included 19 speakers from 9 nations and approximately 60 Japanese participants, among which were many "young scientists".

There was a half-day "General Discussion" regarding the future prospects of new technologies for both basic and applied studies of RBA. Other topics of discussion included the establishment of an effective international network on transgenic fish under the auspices of the IUBS. and an evolution of several smalî teleost fish as a mode1 system for analytical studies for the problems related to RBA.

Financial support was provided by the Japan Ministry of Education. Science, and Culture, as well as the Mie Prefecture Government and several research foundations of Japan, to which 1 am most grateful.

Finally. 1 would like to thank aii the participants. particularly Our overseas visitors. who enthusiasticaily joined this occasion.

References

HEW. C.L. & GONG. Z. 1992. Transgenic Fish: a New Technology of Fish Biology and Aquaculture. B[dogy InteniatlonaL 24-2-10. van OORDT. P.G.W.J. 1991. Reproductive Biology in Aquaculture: a Proposal for an International Collaborative Programme of Research. Btology Intemattonal, "Specfai Issue" N026. Proceedings of the International Symposium on Reproductive Biology in Aquaculture. 22- 27 Apxll, 1991. Bulletin of the InsUtute of Zodogy. Monogmph Nol 6. Academia Sinica. Nankang, Taipei, Taiwan 1 1529. China. A New Approach in Aquaculture: A Must for Feeding a Rapidly Increasing World Population and for Meeting the Ecological Demands of the 21st Century by P.G.W.J. van Oordt Department of Experimental Zoology, University of Utrecht P.O. Box 80.058. 3508 TB Utrecht. Netherlands

The rapidly expanding human population of the earth consumes increasing arnounts of food, not only food derived from agriculture, but also food from natural aquatic ecosystems and from farms. culturing algae. finfish and shellfish. Indeed, all over the world. but especially in Asian countries, including Japan and China. seaweeds. fish, moiiuscs and crustaceans form an important part of the protein rich diet.

According to recent data provided by the United Nations Food and Agriculture Organization (FAO), about 96 million tons of aquatic plants and animals are being used annually, mainly for human consumption. About 13 million tons come fiom various types of fish farms. It is expected. however, that by the turn of the century, the annual need of aquatic products will have risen to 130 million tons per year. and that in order to prevent fatal disturbance of the ecosystem in seas, oceans. rivers and lakes. total Bshing should not exceed 100 million tons per annum. That means that in about seven years' time aquaculture wiii have to be more than doubled. This can only be achieved when on a global scale aquaculture will be given a sound biotechnological basis.

This biotechnological bas& will have to serve not only the quantitative increase in the culture of aquatic animais, but at the sarne time has to safeguard it from causing ecological disorder. Indeed, on a regional basis techniques will have to be developed for culturing aquatic species that are not only of economic interest. but also leave the environment undisturbed.

Reproductive Biology in Aquaculture 0:a Key Area in Biology

The FAO, United Nations Development Programme (UNDP) and other organizations are very active in supporting aquaculture research, especially its applied aspects. Indeed, thanks to the support and guidance of national and international agricultural organizations. aquaculture is presently showing an annual increase of about 10% worldwide.

Among the important issues in aquaculture is reproduction of aquatic animals under husbandry conditions. Many aquatic species of economic interest can be kept in ponds, but fail to reproduce spontaneously under such circumstances. Thus. it is necessary to unravel their reproductive biology. and on that basis to develop techniques for artificially inducing the production of eggs and sperm ceiis, for insemination and for the early development of embryos.

A modern tool in upgrading species of plants and animals is genetic manipulation. In recent years several RBA research groups have made a good Biology Intematio~l (Special Issue N48 -1993) start in adapting these biotechnological techniques for the culture of aquatic animals.

Many countries have scientific institutes and research groups studying reproductive processes in fish. molluscs and crustaceans. Important work, immediately supporting aquaculture, is carried out in laboratories dealing wIth the practical aspects of finfish and shellfish farming. Apart from that there is much fundamental research in reproductive physiology and genetics using as models species of fish, molluscs, or crustaceans. More of this type of work is carried out in universities and other institutes of basic research.

In order to meet the rapidly growing demand for aquatic animals, the output and expertise of such basic research groups should become available for aquaculture, adding to and expanding the work of groups concentrating on appiied aspects of reproductive biology in aquaculture. In order to realize that, fundamental and applied research groups should coordinate their efforts in an international research and training programme.

Existing advanced techniques wili have to be optimized. and new techniques will have to be developed. These will have to reach scientists and farmers al1 over the world. but especiaily in developing countries.

New species will have to be introduced and to be adapted to aquaculture, foremost in regions were the production of protein rich food is most needed, i.e., in developing countries. Local scientists and farmers wiil have to be trained in culturing these new species.

Measures wili have to be developed and improved to protect the environment against pollution and biological invasions.

In summary. Borderline Research and Training in the Biotechnology of Reproduction in Aquaculture are absolutely necessary to meet the increasing demand for healthy. protein rich food of aquatic origin.

Understanding the importance of the issue, the 24th General Assembly of the International Union of Biological Sciences, meeting in Amsterdam in September 1991, adopted a Scientific Programme on Reproductive Biology in Aquaculture (RBA), aiming at the development of a "Global Network on the Biotechnology of Reproduction in Aquatic Animais".

The Main Fields of Reproductive Biotechnology in Aquaculture

The main fields of reproductive biotechnology to be included in the RBA Programme are:

1. Understanding and Manipulation of Reproductive Physiology

The control of the development, growth, differentiation and maturation of the Biology Internaiio~l (Special Issue N28 -1993) reproductive system by :

-understanding the internal organ physiological. cellular and molecular systems regulating reproduction, and

-understanding the environmental signals appropriate for successful reproduction of such species.

2. Understanding and Manipulation of Gamete Physiology

The control of gametogenesis, gamete physiology, fertilization and early embryonic development in species of aquacultural importance, and the identification of the factors regulating these processes.

3. Genetic Manipulation

The improvement of the success of aquaculture by hybridization, polyploidy, cloning, transgenic constnicts and gene transfer.

By stimulating international colla6oration in research and training in these main fields of reproductive biotechnology in aquaculture, the RBA Programme wili contribute to: a) the development and dissemination of biological containment mechanisms in aquaculture; b) the transfer from mode1 systems to practical application for control of the reproductive cycle in aquacult.ure species. important for food, commerce and basic research; c) the optimization of aquaculture production by the enhancement of the eflticiency and yield on a species-specific basis: d) the diversification of aquaculture production on regional and global levels; e) the protection of ecosystems by the prevention of the release of experimental organisms, and the prevention of biological invasions; and f) the protection of aquatic biological diversity and conservation of the gene pool of aquatic species.

ïmplernentation of the RBA Programme

Collaboration bn Research and 'hraining.

The raison d'être of the RBA Programme is the realization of the "Global Network on the Biotechnology of Reproduction". That means collaboration in explorative research and in training on a global scale, resulting in:

- the development of new techniques in reproductive biology in aquaculture and - the introduction of such techniques in the culture of aquatic animais . wherever this is needed to cope with the demands for protein rich food of a Bknogy In!er~ional(Special Issue N58 -1993) rapidly increasing human population - in a way that leaves natural resources undisturbed.

Programme Period.

In order to contribute substantialiy to the efforts of doubling aquaculture by the turn of the century, the practical work for the RBA Programme should start not later than 1993, and preferably continue for seven years, until the year 2000.

Sub-programmes in RBA Research. The first thing to be realized is international collaboration in explorative RBA research. Such research will include training, especially of RBA scientists from technically developing countries, i.e.. ail countries that lack up-to-date facilities for modern cellular, molecular, and genetic research in aquaculture. This explorative RBA research can be divided into 4 sub-programmes, ail together covering 5 projects: 1) Manipulation of Reproductive Control Mechanisms in Fish a. Understanding and Manipulation of Reproductive Physiology b. Understanding and Manipulation of Gamete Physiology; 2) Genetic Manipulation in Fish; 3) Controlled Reproduction in ~olluscs; 4) Controlled Reproduction in Crustaceans.

Collaboration in RBA Research. For each of these sub-programmes a number of outstanding RBA research teams fYom various countries and continents will join forces in a common research project. Collaboration will be brought about by an exchange of scientists and regular contacts, a.0.. during workshops and symposia.

Practical Training. The institutes. involved in the research projects, will ailow colleagues from developing countries to participate in their research by working as trainees in their laboratories for 12 months. The trainees will have the opportunity to profit from the project workshops.

Regiond Lectures. It is important to instmct RBA scientists and farmers in modern techniques of manipulating reproduction in aquatic animals and in farming transgenic finfish and shellfish. A RBA lectureship also provides a platfom to advertise the training programme.

RBA Research and Training Pmjects

1. Manipulation of Reproductive Control Mechanisms in Fish

1 a. Understanding and Manipulation of Reproductive Physiology. Modern research dealing with reproductive physiology of fish mainly focuses on molecular and cellular aspects, such as the identity of hormone producing cells and of their hormone molecules; of receptors, transducers, and genes; on the processes of hormone synthesis and release. as well as on their mode of action at the (sub-) cellular level, and the influence of the environment on these Bialogy International (Special Issue NO28 -1993) processes.

An international project has been developed named "Sexual Differentiation and Maturation of Fish . Models for this project include carp, salmonid. eel. sturgeon, and catfish species.

-Cloning of an all-female strain of the common carp (Cyprfnus carpio) makes it possible to use this species as a model in the study of sexual development for al1 kinds of fish. Thus, it will form an important basis for research on the introduction of new species in developing countries.

-Much important work on the physiology of reproduction has been carried out with a related species. the goldfish (Carassius auratus). Indeed. using the goldfish as a model, in recent years practical methods have been developed for inducing spawning in flsh cultured in developing countries.

-The black carp (Mylopharyngodon piceus) eats molluscs, and when introduced into reservoirs and aqueducts proved to be very efficient in reducing snail populations, aileviating the-problem of water flow through pipes, filters. and pumps. It could also become an environmentaliy acceptable alternative for poisonous chemicals in the fight against schistosomiasis (BiUzarziasis) in tropical and subtropical regions, provided that culture of this species can be improved by shortening the period of immaturity.

-Salmonids, such as salmon and trout species (Salmo spec. and Oncorhynchus spec.), are well known models in aquacultural research. That can best be demonstrated by pointing to the successful culture of both Atlantic and Paciflc salmon and of trout all over the world, especially in regions with specialized RBA research laboratories.

-Eels (Anguilla spec.) do not reach maturity before their catadromous migration, and induced breeding of eels has met with enormous problems. International cooperation is envisaged to solve those problems.

-Sturgeons (Acipenser spec.) are of great economic importance, but much needs to be done to fully understand their reproductive biology. Plans for international collaboration are developed for basic research leading to improved techniques of sturgeon culture.

-Catfish species, including the AMcan catfish (Clarias garieplnus) have tumed out to be very good models for studying cellular and molecular aspects of sexual development in fish. Moreover, fundamental studies on reproductive biology in the African catfish have contributed considerably to developing techniques for the culture of this species in Afi-ican countries and for the culture of an Asian catfîsh, Clarias batruchus. in Indonesia.

-There are plans for reintroducing the European catfish (Silurus glanis) in fish ponds and open water of Western Europe. The present knowledge concerning reproductive biology in other catfish species will form a scientific basis of that work. Bwlogy InternatiOMI (Special Issue NT8 -1993) 1 b. Understanding and Manipulation of Gamete Physiology in Flsh

Modern research concerning gamete physiology wiiï be carried out at the subceiiular and molecular level in an international project entitled "Garnete Physiology in Fish". The foilowing aspects receive speciai attention.

Gamete Qualitp. The main issue in understanding and manipulating gamete physiology in fish is the quality of eggs and sperm ceiis. their viabiiity and the& success in producing normal healthy larvae. The mechanisms involved in determining the quality of the gametes have never been studied in detail. Because of this lack of understanding. the technology of handling and managing gametes in aquaculture remains crude and the farmers have no possibility of evaluating the quality of fish gametes and of early embryos, and they can do next to nothing in improving that quality. This situation has to be changed, and that can only be done by supporting application directed basic research in gamete physiology.

This quaiity of eggs and sperm cells depends upon the processes that lead to the formation of the gametes, i.e., oogenesis. including vitellogenesis, and spermatogenesis. as well as to the maturation of the oocytes and sperm cells.

Spenn rnotilîty. An important aspect of sperm celi physiology is the motility of these ceiis.

Aging of gametes. The aging of gametes under in vivo and tn vitro conditions influences their quality.

Fertilization and early embryonic development. The processes concerning ferüiization and early embryonic development corne under this heading.

Cryopreservation. Last but not least needs to be mentioned the wish to conserve gametes and early embryos by freezing without loss of quality; a process known as cryopreservation. The possibility of storing high quality gametes and of purchasing deep fi-ozen gametes and embryos enable the nsh farmer to maintain, and if necessary. to împrove the quality of his livestock and the output of his farm.

2. Genetic Manipulation in Ffsh

RBA research on 'Transgenic Fish is an international project of that name. and will consist of the following coinponents.

Development of Suitable Gene Constructs for Aquaculture. Much attention fi-om the international community is given to this type of RBA research. Genetic manipulation for the improvement and enhancement of the growth rate of cultured fish is considered to be especially crucial for the success of modem aquaculture. Transgenic fish with growth hormone insert offers one potential approach. Future research will concentrate on the design of better gene constructs, the evaluation of growth performance and the regulation of transgene expression. Biology Internatio~l (Special Issue NT8 -1993) Biological Containment Requirement. The accidental release of farmed fish poses a potential ecological hazard. Until its full impact on the environment can be better assessed, it is important that these animals be contained. International cooperation is needed to develop common guidelines and approaches. One practical solution is to develop sterile transgenic fish. Research collaboration on various chromosomal manipulation techniques and novel approaches such as gene ablation should be explored and developed.

Irnprovernent in Gene Transfer and Detection Techniques. There are numerous fish species worthy of transgenic studies. The experimental conditions for many new gene transfer and analyses techniques such as electroporation of sperm and fertilized eggs, polymerase chain reaction and irnmunoblot. should be widely publicized.

3. Controïied Reproduction in MoUuscs

An international project has been developed on "Improving Cost-Effectiveness of Molluscan Shellfish Production". At present, the culture of molluscs is mainly a matter of practical experience, and is insufficiently based on biotechnological knowledge. Therefore, following a modem cellular. molecular and genetic approach, research on reproductive biology in moiluscs wiil have to be aimed first at identi@ing the neurohormones and other basic regulatory signais and mechanisms controiling reproduction and development. This can best be done in three weil characterized mode1 organisms: abdones [Hdiotis spec.), oysters (Crassostera spec.), and anrriln (Lymnaea spec.). Many of tnese signals are hormones from the nervous system. Identification of these neurohormones and their mode of action will lead to developing techni?. -s for artificial propagation of molluscs under husbandry conditions. These findings will lead to developing techniques for artificial propagation under husbandry conditions of a wide variety of locally important molluscan species in developing countries.

4. Controïied Reproduction in Crustaceans

An international project has been developed. named "Application of Biotechnology in the Aquaculture of Craistacea". Using modern cellular, molecular, and genetic techniques, the project will concentrate on the study of neuropeptides, necessary for screening of the adaptation of crustaceans to the environment of an intensive culture system. and controlling reproduction. This will lead to the production of specific markers for the development of application-kits for measuring .the neuropeptides and related metabolic components in animais growing in aquacultural systems. Such application-kits will be of great importance for irnproving the culture ofcrustaceans. especially in regions where that can best of ali be carried out, i.e.. in (sub-) tropical developing countries.

Another project proposal, easily to be combined with the above one, and developed by RBA scientists from the U.S.A. and China (Taipei), concentrates on "Gamete Physiology in Marine Shrimps". Shrimp culture is very important in these and other countries, especially in developing countries in Asia. Here Biology Internatio~l (Special Issue NI18 -1993) too, sufflcient knowledge of the signais that regulate the production of eggs and sperm cells and of the processes involved in spawning and fertilization. is necessary to irnprove and to maintain the culture of these crustaceans.

Steering Committee of the RBA Programme

The RBA Programme is one of the sclentific programmes of the IUBS. The Executive Committee and the Scientific Programme Committee of this international union regularly evaluate, and guarantee the progress and the quality of the programme.

The daily management of al1 RBA projects will be in the hands of sub- programme cornmittees of specialists, and their work will be coordinated by a Steering Committee. The members are as follows: Professor P.G.W.J. van Oordt. Utrecht, Netherlands (Chairman); Dr. J. Benzie (Townsville, Australia); Professor R. Billard (Paris, France); Professor W.H. Clark. Jr. (Bogeda Bay, U.S.A.); Professor Y.A. Fontaine (Paris, France); Dr. W.P.M. Geraerts (Amsterdam. Netherlands); Professor H.J.Th. Goos (Utrecht, Netherlands); Dr. F. van Herp (Nijmegen, Netherlands); Professor Choy L. Hew (Toronto, Canada); Professor D.E. Morse (Santa Barbara, U.S.A.); Professor Y. Nagahama (Okazaki. Japan); Professor R.E. Peter (Edmonton, Canada); Professor M.P. Schreibman (Brooklyn. U.S.A.) ; Professor N. Stacey (Edmonton, Canada); Professor Shaoyi Yan (Beijing, China); Professor 2. Yaron (Tel-Aviv, Israel); and Dr. T. Younès (IUBS Executive Director, Paris, France) ex-offlcio. References

VAN OORDT. P.G.W.J. (Ed.). 1991. Reproductive Biology in Aquaculture. A Proposai for an International Collaborative Programme of Research. Bblogy Internatbnai, "Special Issue N02ô". 46pp. Bwlogy Iniermdo~I (Special Issue NOL8 -1993) Biotechnology of Aquatic Animals: A New Frontier with Implications for Both Basic and Applied Research

by D .A powersl . T.T. chen2. RA. ~unham~ '~o~kinsMarine Station. Stanford University. Pacific Grove, CA, U.S.A. kenter of Marine Biotechnology and Department of Biologid Sciences, University of Maryland. Baltimore, MD, U.S.A, 3~epartmentof Rsheries and Allied Aquaculture. Auburn University. Auburn, AL. U.S.A

The application of recombinant DNA and hybridoma techniques to fundamental and practicai problems in the biomedical sciences has resulted in stunning victories. and the successful transfer of that information into the commercial sector has been equally impressive. Public awareness of the triumphs surrounding biotechnology has generally revolved around high profile advances in the diagnosis and treatment of diseases; however, equally important has been the production of find chemicals, including pharmaceuticals and antibiotics. In addition to accomplishments in the biomedicai field, there have been astonishing achievements in agriculture, including the generation of transgenic animals and plants, the creation of recombinant vaccines for the treatment of animal and plant diseases. and the manipulation of microbes to improve everythfng from soil nutrients to protection against frost.

The oceans cover more than 70 percent of the earth's surface and fresh water ecosystems contribute a significant portion as well. Moreover, the oceans constitute almost 95 perceiit olC the biosphere by volume. The depths of these fresh and salt water habitats contain a multitude of diverse organisms which are potential resources for both applied and basic research. These aquatic ecosystems are essentially an infinite reservoir of quality food, biomedicaliy important substances, sources for the degradation of anthropogenic waste and industrial pollutants, antibiofouling and anticorrosion substances. biosensors. biocatalysts, biopolymers, and other iiidustrially important compounds.

Even though the application of biotechnological efforts toward aquatic environments has been modest when compared to those efforts devoted to terrestriai systems. there have been some notable successes and preliminary evidence that the entire field of marine and fresh water biotechnology is on the verge of becoming a new frontier of explosive scientific discovery. 1 will provide selected examples of exciting achievements within the field and point toward a few emerging areas that have a high probability of future success. Particular attention will be given to the molecular cloning of fish genes and the use of transgenic fish for both basic and applied research.

The transfer of foreign DNA into fertilized eggs has been used to generate a wide range of transgenic animal species, including fish, with a high degree of success. Not only can this technology constitute a powerful approach for examining the fundamental mleeular mechanisms of gene expression and regulation but it also offers an effective means for the rapid development of new genetic stocks of economically important livestock or for the large scale production of novel proteins.

Although exogenous application of biosynthetic growth hormone (GH) enhances somatic growth of cultured finfish and shelifish, it may not be as cost effective as the use of genetic engineering to generate new fish strains that express high levels of GH. This would bypass many of the problems associated with exogenous GH treatment. Once the strains have been generated, they would be far more cost-effective because they would have the& own means of producing and delivering the hormone, and could transmit their enhanced growth ability to their offspring. Simila. arguments could be made for other genes. Exogenous Treatment (The use of Biosynthetic gene products)

Several studies have shown that the application of the biosynthetic growth hormone to finfish and shellfish result in a significant growth enhancement. These results suggest that the use of synthetic growth factors might be used to study the regulatory mechanisms that control hormonal function in aquatic animals. Equally important is the possibility that exogenous treatment of hormones might be used in commercial aquaculture operations. In addition to exogenous treatments, the possibility exists that gene transfer methods could be used to generate transgenic fish that could, in turn. be used to test hypotheses about the molecular basis of gene regulation and to generate strains of commercially important aquatic animals (e.g., ones with enhanced growth rates, reproductive success. disease resistance, freeze resistance. low oxygen tolerant, etc.).

For exarnple, our laboratories are conducting experiments to assess the growth enhancement effect of the biosynthetic fish growth hormone (GH) on a variety of finfish and shelifish species. We have already made substantial grogress. For exarnple, large-scale preparation of active biosynthetic trout GH polypeptide has been generated by expressing the trout GH complementary DNA (cDNA) in E. coli. Application of this protein to yearling rainbow trout by intraperitoneal injection resulted in a significant growth enhancement. After treatment with the biosynthetic GH for 4 weeks at a does of lug/g body weight/week, the weight gain in the experimental group was two times greater than the control group (see Fig. 2). A substantial increase in body length was also observed in the GH treated animals (Agellon et ai., 1988). Furthermore, the chemical composition of muscle tissues of hormone treated fish is indistinguishable from that of the control fish.

Although the injection procedure clearly demonstrated enhanced growth, there are practical limitations of this procedure, including its applicability to very small fish and the treatment of large numbers of fry. In another experiment, the growth promoting effect of the biosynthetic trout GH was assessed in rainbow trout fry by a dipping procedure (Agellon et ai., 1988). In these studies, the biosynthetic GH was administered by incubating test animals Biology Internatio~l(Speciol Issue NT8 -1993) in an isotonic saline solution containing 50 ug/l or 500 ug/l of GH polypeptide. Five weeks after hormone treatment, a total of 1.5-1.6 times increase in body weight was observed in hormone treated fish compared to controls. These results are in agreement with those reported by Sekine et al. (1985). Wagner et al. (1985). and Gill et al. (1985).

These data suggest that biosynthetic GH would be very useful to enhance the growth rates of cultured finRsh and shellfish but a series of studies needs to be done to delineate a number of practical and basic scientific issues, including the most effective and practical means of hormone delivery and the large scale production and purification of biosynthetic GH. There would also be a need to develop detailed dose regimens for each target species. including the effect of both chronic and acute GH treatment, nutrient requirements, and other rearing-related conditions. Finally, a series of studies would need to be done to assess the genetic and physiological regulation of growth hormone in mode1 aquacultured species and the potential impact of long-term exogenous GH treatment on these aquatic species.

Even though exogenous application of biosynthetic growth hormone (GH) enhances somatic growth of cultured finfish and shellfish. it may not be cost effective for routine application. As an alternative, the use of genetic engineering to generate new strains that express higher levels of GH would bypass many of the problems associated with exogenous GH treatment. Moreover, once the strains have been generated, they would be far more cost- effective because they would have their own means of producing and delivering the hormone, and could transmit their enhanced growth ability to their offspring. Similar arguments could be made for other genes. Endogenous Treatment (Transgenic fish)

Animals into which foreign genes have been introduced are termed transgenic. During the past several years. foreign DNA has been successfully introduced, by microinjection, into frogs (Rusconi & Schaffner. 1981). laboratory mice (Brinster et ai., 1981; Constantini & Lacy, 1981; Gordorn et aL. 1981). rabbits, sheep. pigs (Hammer et d,1985) and cows (Pursel et al., 1989). In each of these cases, DNA of various quantities was injected into pronuclei of the developing embryos, and the injected embryos were transplanted into the utems of pseudopregnant females for further development. Results from these studies showed that multiple copies of foreign genes integrated at random locations in the genome of transgenic individuals, in head-to-tafl tandem arrays. If the foreign gene is introduced into the developing embryos with a functional promoter, expression of the foreign gene in some transgenic individuals in expected. In some instances, the foreign gene is also transmitted through the germ line to subsequent generations.

Several laboratories have also reported the transfer of foreign genes into fish (Zhu et al., 1985: 1986; Maclean et al., 1987; Chourrout et al., 1986; Dunham et aL, 1987; Fletcher et al., 1988; McEvoy et d,1988; Ozato et al., 1986; Stuart et aL, 1988; Lu et d,1989; Rokkones et cd.. 1989, Zhang et aL, 1988; Biology Iniernutio~f (Special Issue N98 -1993)

1990; Zhiro et al.. 1989; Chen et al., 1990; Powers et d.1990). Foreign genes used in these studies are: human GH gene, bovine GH gene, chicken delta- crystalline gene, bacterial beta-galactosidase gene, hygromycin resistance gene, fish antifreeze protein gene, alpha-globin gene, new gene, and fish GH gene. For example, McEvoy et d (1988) transferred a beta-galactosidase gene into Atlantic salmon, Ozata et al. (1986) transferred a chicken delta-crystalline gene into medaka, and Zhang et d (1989; 1990) transferred a rainbow trout GH gene into common carp and catfish. In al1 three cases, the foreign genes expressed in some of the transgenic individuals.

In some cases. the gene products produced by transgenic animais can alter their phenotype. For instance, transgenic mice expressing human or rat CH gene have elevated levels of GH. consequently, they grow much faster thank their control siblings (Palmiter et ai.. 1982). In the case of transgenic common carp canying rainbow trout GH gene (Zhang et d,1990). rainbow trout GH polypeptide was detected in the fish and, consequently, they also grew substantially faster thûn their sibling controls. These results point to a significant potential for the use of the gene transfer technology to introduce desirable genetic characteristics to commerciaily important fish species.

In our gene transfer studies, we are using the common carp (Cyprinus carpio), channel catfish (Ictalurus punctatus), killifish (Fundulus heteroclitus). zebrafish (Brachydanio rerio), rainbow trous (Onchorhynchus mykiss), medaka (Oryzlas Latipes), and a variety of invertebrates as animal models and luciferase, beta- glactosidase, lactate dehydrogenase, prolactin, and growth hormone as mode1 gene systems. For example, we have created fast growing transgenic carp and channel caffish by transferring extra copies of a fish growth hormone gene into their genomes. Not only do these transgenic fish grow faster but they transmit the foreign DNA to their progeny in 3 of 4 crosses and the progeny grow approximately 40% faster than their non-transgenic full siblings. However, transmission of the foreign gene in channel catfish is more variable than that observed in transgenic common carp. Gene Transfer Methodology

Although direct microinjection. retrovirus infection, electroporation, calcium phosphate precipitation, and particle gun bombardment are methods frequently used for introducing foreign genes into somatic cells, only direct pronuclear injection and retrovirus infection are proven to be effective in transferring foreign genes into germ lines.

The most prevalent method for introducing genes into the germ line is the direct microinjection of cloned DNA into the male pronuclei of the fertilized eggs. This method has been used successfully to introduce a host of foreign genes for generating transgenic mice (Gordon et d,1980; Wagner et ai., 1981; Palmiter et al., 1982), cows, pigs, sheep. and rabbits (Hammer et al., 1985; Pursel et ai., 1989). Transgenic fish have been generated in several fish species by microinjection: rainbow trout (Chourrout et al.. 1986; Maclean et ai., 1987). salmon (McEvoy et ai.. 1988: Fletcher et al., 1988; Rokkones et al., Biology Infernalional (Special Issue NT8 -1993;) 1989). common carp (Zhang et al., 1988; 1990). loach (Zhu et aL, 1986). caffish (Dunharn et d. 1987), tilapia (Brem et d,1988). goldfish (Zhu et al.. 1985). zebrafish (Stuart et al.. 1988). and medaka (Ozato et af., 1986). Since the male pronuclei in most of the fish species studied to date are not visible, the microinjection is usually done into the egg cytoplasm, and the quantity of the foreign genes required for each injection is in the order of 106 molecules per egg or higher. In general, eggs and sperm are collected from mature females and males into separate dry containers. Fertilization is initiated by adding water and sperrn to eggs, with gentie stirring to enhance fertilization and dispersal. Eggs are water-hardened for various periods of time (depending on the fish species) and then rinsed. Microinjection is usually carried out within the first few hours after fertilization (Le., between one-ce11 to four-ce11 stages). Following microinjection, eggs are incubated until hatching.

Since fish undergo external fertilization, the injected embryos do not require complex manipulations such as in vitro culturing of embryos and transferring of embryos into foster mothers -- essential in marnmalian systems. Another added feature in the transgenic fish system is that cytoplasmic injection is much easier and less injurious to development, so that the survival rates of the injected embryos should be higher in fish than in mammalian systems. Depending on different fish species which are microinjected. the survival rates of the injected embryos range from 35% to 80% which are substantially higher than those observed in mamrnalian systems. It is interesting that even though the foreign DNA is injected into the cytoplasm, the rate of foreign gene integration in the transgenic fish system is relatively high. One exception to the cytoplasmic injection, however, is the case of medaka in which the pronuclei are visible.

The tough chorion of many fish species has frequentiy presented problems in inserting glass micro-needles into the eggs. This problem has been overcome by the following strategies. In rainbow trout and salmon eggs, an opening in the egg is made by micro-surgery before the micro-glass needle is inserted (Chourrout et d,1986; McEvoy et al., 1988). In Atlantic salmon and tilapia. a micro-needle can be inserted through the micropyle (an opening for sperm penetration at fertilization) (Fletcher et aL, 1988; Brem et al.. 1988). The chorion of zebrafish eggs can be removed manually with a pair of forceps (Ozato et aL, 1986), and that of goldfish and loach by trypsin digestion (Zhu et al., 1985; 1986). Recently, Oshiro et aL (1989) reported that hardening of rainbow trout egg chorion can be prevented if the fertilization takes place in a solution containing 1 mM glutathione solution (pH 8.0). Another approach to circumvent the problem of tough chorion is microinjection of DNA into unfertilized eggs, since the unhardened chorion can be penetrated easily by glass rnicroneedles. This method has also been shown to be very effective.

Although the microinjection method is successful in transferring foreign genes into fish embryos. it is a tedious and time consuming procedures. Hence. there is a genuine interest in developing mass gene transfer technologies for use in fish. In contrast to microinjected DNA, retroviruses integrate their genetic materials into the genome of the infected cells efficiently by a Biology International (Special Issue N58 -1993) precisely-defined mechanism. Only a single copy of the provirus is inserted at a given chromosomal site, and rearrangements of the host genome are not induced. apart from a short duplication of host sequences at the site of integration (Varmus, 1982). Therefore. the foreign gene can be inserted into the viral genome, and transferred to the host by viral infection. The prerequisite of applying this gene transfer method depends, however, on the availability of species-specific retroviruses. For instance, there are well- characterized murine and avian retroviruses available which have been used to transfer foreign genes into mice and chickens (Soriano and Jaenisch. 1986; Salter et aL, 1987). Unfortunately, this gene transfer method has not yet been tried in fish systems. because none of the reported presumptive fish retroviruses has been characterized.

Electroporation utilizes short electrical pulses to permeabilize the ce11 membrane, thereby gaining the entry of macromolecules into the cell. Utilization of this method for gene transfer has been proven successful in bacteria, cultural mamrnaiian cells, and plant protoplasts (Shigekawa & Dower, 1988). Attempts have been made by several laboratories throughout the world to adopt this technique for gene transfer in fish systems. Although initial results with electroporation were not as good as microinjection, recent attempts in Our laboratories (i.e., Powers et al.. 1992) and those of other researchers have been very prornising.

Major Breakthroughs Needed

Even though substantial progress has been made in the generation of transgenic fish. considerable research is needed in order to take full advantage of fish as experimental mode1 systems and for the development of transgenic fish for commercial aquaculture purposes.

Fish produce large volumes of eggs and sperm. Development of methods for transferring genes into large numbers of eggs could make fish even better experimental models and commercial targets than many higher vertebrates which only produce a few eggs. As mentioned previously, a number of researchers are attempting to develop mass transfer techniques such as: viral vectors, coating sperm with DNA containing particles, electroporation, precipitation methods, particle gun methods, and other novel approaches. In concert with the development of mass transfer techniques, there is a need to develop mass screening methods to identi@ presumptive transgenic individuals during very early stages of their life cycle. Efforts are underway to develop visual markers that will allow one to ident@ transgenics while they are still in the egg stage so non-transgenic individuals can be eliminated. Attempts to use biochemicai markers that can be easily visualized, such as luciferase and beta- galactosidase, have provided limited success. However, these and a variety of morphological and biochemical markers have not proven useful for universal screening at the egg stage. Thus, there is a need to develop efficient mass gene transfer methods coupled with rslpid mass screening methods that would allow the visual identification of transgenic individuals, preferably aé the embryonic stage. In addition to breakthroughs in mass gene transfer and screening technology, methods need to be developed for targeting genes to specific areas of the fish genome. Methods need to be developed for the regulation of foreign gene expression. Embryonic stem ce11 methodology, that has proven so useful in mammalian systems, needs to be introduced into fish model systems. A considerable arnount of basic research needs to be done on a variety of fish genes in order to identify useful genes and their promoters as candidates for introduction into fish lines. These include different genes in the growth process other than the GH, disease resistant control, osmotic pressure, injection enzymes, and others.

In order to assess the potential impact of foreign genes on simulated "wild type" environments, there is a need to study the physiological, nutritional. developmental, imrnunologicai, and reproductive responses of transgenic fish in outdoor fish ponds where model environmental studies can be done under controlled environmental conditions. For example, it is important to study the growth characteristics of transgenic fish. which produce elevated levels of growth hormone, under typical food saturating aquaculture conditions as opposed to simulated seasonal food variability Situations such as those that would be encountered in the natural environment. Wiil these transgenic fish starve when food is limitecl or will they out compete wild stocks? These and other important pond type experiments need to be done in order to assess the potential environmental impact of transgenic fish that might escape. As part of such a study. it would be useful to develop suicide gene constnicts that would not only prevent the animai from reproducing but would actually become lethd at an appropriate time. There are a number of possibilities for such lethal constnicts but as yet such studies have not been attempted in fish.

AGELLON, L.B., EMERY, C.J.. JONES. J.. DAVIES, S.L., DINGLE. A.D.. & CHEN, T.T. 1988. Growth Hormone Enhancement by Genetically-Engineered Rainbow Trout Growth Hormone. Can. J. Flsh Aqua. ScL 45: 146-151. BENDIG, M.M. 1981. Persistence and Expression of Histone Genes Injected into Xenopus Eggs ln Early Development. Nature. 29265-67. BREM. G.. BRENIG. G.. HORSTGEN-SCHWARK, G.. & WINNACKER, E-L. 1988. Gene Transfer in Tiiapia (Oreochrornis nUotlcus). Aquaculture. 68:209-219. BRINSTER, R.L., CHEN. H.Y., TRUMBAUER, M.E., SENEAR, A.W., WARREN, R., & PALMITER W.D. 1981. Somatic Expression of Herpes Thymidine Kinase in Mice Following Injection of a fision Gene into Eggs. CeU. 27:223-231. CHEN, T.T. & POWEEtS. DA 11990. Transgenic Flsh. îYends in Bbtechridogy. 8:U)4216. CHEN. T.T., LIN. C.M., ZHU. Z., GONZALEZ-VILLASENOR, L.I.. DUNHAM. R.A., & POWERS. D.A. 1990. Gene Transfer. Expression and Inheritance of Rainbow Trout and Human Growth Hormone Cenes in Carp and Loach. IN: Transgenic Models in Medkine and Agriculture. Robert Church (Ed.]. Wiley-Us. Inc., New York 127-139. CHEN. T.T.. ZHU, Z., DUNHAM, R.A.. & POWERS, D.A. 1990. Gene Structure, Expression and Inheritance of Rainbow Trout and Human GH genes in Carp and Loach. IN: Current Topfcs Ln Marine BbtechnoIogy. S. Miyachi. 1. Karube & Y. Ishida (Eds.). Fuji Technolog Press, Tokyo. 27 1-274. CHEN. T.T.. KIGHT. K.. LIN, C.M., POWERS, D.A.. HAYAT, D.A.. CHATAKONDI. M., RAMBOUX. A.C., DUNCHAN. P.L.. & DUNHAM, R.A. 1993. Expression and Inheritance of pRSVreGHjcDNA in the Transgenic Common Carp. Cyprinus Carpia Molecular Marine Blology and Bwtechridogy. 2(1):44-50. BwIogy International (SpeciaI Issue N28 -1993)

CONSTANTINI, F. & LACY, E. 1981. Introduction of a Rabbit-Clobin Gene into the Mouse Cern idne. Nature. 294:92-94. CHOURROUT. D.. GRYOMARD. R.. & HOUDEBINE. L-M. 1986. High Efficiency Gene Transfer in Rainbow Trout [Salmo gairdneri Rich.) by Microinjection into Egg Cytoplasm. Aquaculture. 51:143-150. DUNHAM. R.A.. CASH. J.. ASKINS, J. & TOWNES. T.M. 1987. Transfer of the Metallothionein-Human Growth Hormone Fusion Gene into Channel Catfish. 'Itnns. Am FLsh Soc. 116i87-91. DUNHAM. R.A.. RAMBOUX, A.C.. DUNCAN, P.L., HAYAT. M., CHEN. T.T.. & POWERS. D.A. 1992. Transfer. Expression, and Inheritance of Salmonid Growth Hormone Genes in Channel Catfish. Ictalurus punctatus and its Effects on Performance Traits. Mokular Marine Beyand Biotechndogy. 1(4/5):380-389. ETKIN. L.D.. PEARMAN, B.. ROBERTS, M., & BEKTESH, S.L. 1984. Replication. Integration and Expression of Exogenous DNA injected into Fertilized Eggs of Xenopus laeuis. D#-erentiatbn 26: 194-202. ETKIN. L.D. & PEARMAN. B. 1987. Distribution. Expression and Cerm Line Transmissin of Exogenous DNA Sequences Following Microinjection into Xenopus laevis Eggs. Dewlopment 99: 15-23. FEINBERG. A.P. & VOGELSTEIN. B. 1984. A Technique for Radiolabeling DNA Restriction Endonuclease Fragments to High Specific Activity. Anal. Blochem 137:266-267. FLETCHER. G.L.. SHEARS. M.A., KING. M.J.. DAVIES. P.L., & HEW, C.L. 1988. Evidence for Antifreeze Protein Cene Transfer in Atlantic Salmon (Sahj. Cm. J. Flsh Aquat. Sei 45352-357. GILL. J.A., STUMPER. J.P., DONALDSON. E.M.. & DYE. H.M. 1985. Recombinant Chicken and Bovine Crowth Hormone in Cultured Juvenile Pacific Salman (Onchorhynchus ktsutch). Biotechnology. 3W3-646. GORDON. J.W. & RUDDLE. F.H. 1985. DNA-Mediated Genetic Transformation of Mouse Ernbryos and Bone Marrow: A Review. Cene. 33121-136. HAMMER. R.E.. PURSEL. V.G.. REXROAD, C.E. JR.. WALL, R.J.. BOLT, D.J.. EBEWT, K.M.. PALMITER. R.D.. & BRINSTER, RL. 1985. Production of Transgenic Rabbits, Sheep, and Pigs by Microinjection. Nature. 315:m. HINEGARDNER. R. & ROSEN. D.E. 1972. Cellular DNA Content and the Evolution of Teleostean Ftshes. AnNatur. 106:621-6%. INOUYE, S., VLASUK, G.P.. HSIUNC, H.. & INOWE, M. 1984. Effects of Mutations at Glycine Residues in the Hydrophobie Region of the Escherfchfa coli Prolipoprotein Signal Peptide on the Secretion Across the Membrane. J. Bid Chem 259:3729-3733. LU, J.K.. ANDRISANI. O.M.. DIXON. J.E., & CHRISMAN. L. 1989. Integration and Stable Germ Line Transmission of Human Crowth Hormone Cene Via Microinjection into Early Medakau Embryos. Abstract. Second Symposium on Genetic Engineering Animais. Comell University. 25-28 June. McGRANE. M.M.. DE VENTE. J.. YUN. J., BLOOM. ., PARK, E.. WYNSHAW-BORIS. A.. WAGNER, T.. ROTïMAN. F.M.. & HANSON, R.W. 1988. Tissue-Specific Expression and Dietary Regulation of a Chimeric Phosphoenol Pyruvate Carboxykinase/Bovine Crowth Honnone Gene in Transgenic Mice. J. Bbl. Ckem 263: 11443- 11451. McKNIGHT. S. & TIJIAN. R. 1986. Transcriptional Selectiviiy of Viral Genes in Mammalian Celis. CelL 43795-805. McEVOY. T., STACK. M., KEANE. B.. BARRY. T.. SCREENAN. J.. & CANNON. F. 1988. The Expression of a Foreign Gene Salmon Embxyos. Aquaculture. 68.27-37. MACLEAN. N.D.. PENMAN. D., & ZHU. 2. 1987. Introduction of Novel Genes into Fish. Biotechnolog y. 5:257-261. MANLATIS. T., FRITSCH. E.F., & SAMBROOK, J. 1982. Molecular Cbnfng: A Laboratoq Manuai Cold Spring Harbor Laboratory, Cold Spring Harbor, New York OSHIRO, T., YOSHIZAKI. G., & TAI'ASHIMA, F. 1989. Introduction of Carp Alpha Clobin Gene in Rainbow Trout (Salmo gairdneri). Abstract. The Flrst International Marine Biotechnology Conference. pp. 4- 10, 4 1. OZATO, K.. KONDOH, H.. INOHARA, H., IWAMATSU, T.. WAWATSO, Y., & OKADA. T.S. 1986. Production of Transgenic Fish: Introduction and Expression of Chicken Cxystallin Cene in Medaka Embryos. Ce11 Duerentiation 19:237-244. PALMITER, R.D., BRINSTER, R.L.. MANNER. R.E.. TRUMBAUER, M.E.. ROSENFELD, M.G., BIRNBERC, N.C., & EVANS, R.M. 1982. Dramatic Growth of Mice that Develop from Eggs Microinjected with Metallothionein Growth Hormone Fusion Genes. Nature. 300:611-615. PALMITER. R.D., WILKIE, T.M., CHEN, H.Y., & BRINSTER. R.L. 1984. Transmission, Distortion and Mosaicism in an Unusual Transgenic Mouse Pedigree. CeIl. 36:869- 8ïï. Biology Internalional (Special Issue N"28 -1993)

POWERS. D.k 1989. FYsh as Modd Systems. Science. 246:352-358. POWERS. D.A. 1990. Marine and Freshwater Biotechnology: A New Frontier. IN: Biotechnology: Perspectives. Policles and Issues. Indra Vasil (Ed.). Uniwzadky of. Florida Press. pp.41-70. POWERS. D.A. HEREFORD. L., COLE. T.. CREECH. K.. CHEN. T.T.. LIN. C.M.. KIGHT, K., & DUNHAM. R. 1992. Electoporation: A Method for Transferring Genes into the Gametes of Zebrafish, Brachydanb reriq Channel Catfish. Ictalurus punctatus, and Common Carp, Cyprinus carpio. Molecular Marine Biology and Biotechnology. 1(4/5):301-308. PURSEL. V.G., PIMBERI', C.A., MILLER. K.F.. BLOT, D.J.. CAMPBELL. R.G.. PALMITER. R.D., BRINSTER, R.L., & HARMER. R.E. 1989. Geneîic Engineering of Livestock. Science. 244:1281-1288. ROKKONES. E.. ALESTROM. P.. SKJERVOLD. H., & GAUTVIK, KM. 1989. Microinjection and Expression of a Mouse Metallothionen Human Growth Hormone Fusion Gene in Ferülized Salmonid Eggs. J. Comp. PhysbL B158:751-758. SEKINE. S., MIIZUKAMI, T., NISHI. T.. KUWANA, Y.. SAITO. A.. SATO, M., ITOH, H.. & KAWAUCHI. H. 1985. Cloning and Expression of cDNA for Salmon Growth Hormone in Eschetichta cdL Proc. Nat'l Acad ScL, U.S.A. 82:4306-4310. STUART. G.W., McMURRAY. J.V.. & WESTERFIELD, M. 1988. Replication, Integration and Stable Germ-Line Transmission of Foreign Sequences Injected into Early Zebrafish Embryos. Dewlopment 103403-412. WAGNER, T.E.. HOPPE. P.C., JOLLICK. J.D., SCHOLL. D.R., HODINKA. R.. & GAULT. J.B. 1981. Microinjection of a Rabbit g-Globin Gene into Zygotes and its Subsequent Expression in Adult Mice and Their Offspring. Proc. NatZ Acad. ScL. U.S.A. 78:6376- 6380. WAGNER. G.F.. FARGHER, R.A.. BROWN, J-C.. & McKEOWN. B.A. 1985. Further Characterization of Growth Hormone from Chum Salmon (Oncorhynchus ketaj. Gen Comp. Endocrlnd a27-34. WILKIE. T.M. & PALMITER. R.D. 1987. Analysis of the Integrant in MyK-103 Transgenic Mfce in which Males Fail to Transmit the Integrant. Molec. CelL Bld 7:1646-1655. YAMAMOTO, T.. CROMBRUGGHE. B.D.. & PASTAN. 1. 1980. Identification of a Functional Pmmoter in the Long Terminal Repeat of Rous sarcomavinis CelL 22.787-797. ZHANG. P., HAYAT, M.. JOYCE. C., GONZALEZ-VILLASENOR. L.I.. DUNHAM. R.A.. CHEN. T.T., & POWERS, D.A. 1990. Gene Transfer. Expression and Inheritance of pRSV-Rainbow Trout-GHcDNA in the Common Carp. Cyprinus carpia Molecular Reproduction and Development 25: 13-25. ZHANG, P.. HAYAT. M.. JOYCE. C., LIN. C.M., GONZALEZ-VILLASENOR. L.J.. DUNHAM, R.. CHEN. T.T., & POWERS. D.A. 1988. Gene Transfer, Expression. and Inheritance of pRSV-Trout-GH-cDNA in Fish. Abstract. First Int'l Symp. Marine Molecular Biotechnology. Baltimore. MD. 9-1 1 October. ZHU, Z., LI. O.. HE, L., & CHEN. S. 1985. Novel Gene Transfer into the Fertilized Eggs of Goldtlsh [Carassius aumtusL 1758). Z angew. IchthyoL 1:31-34. ZHU, 2.. XU. K. LI. G.. XIE, Y.. & HE, L. 1986. Biological Effects of Human Growth Hormone Gene Microinjected into the Fertilized Eggs of Loach. Mfsgurnus anguUlfcaudatus (Canton). Xexue Tongbao, Academia Sinica, Wukan, P.R China. 31:988-990. Biology International (Special Issu N78 -1993) Regulation of Oocyte Maturation in Aquatic Animals: The Comparative and General Aspects

by Y. Nagahama Laboratory of Reproductive Biology, National Institute for Basic Biology. Okazald 444. Japan

The process of oocyte maturation (the resumption of meiosis) occurs prior to ovulation and is a prerequisite for successful fertilization; it consists of breakdown of the germinal vesicle (GVBD), chromosome condensation, and assembly of the first polar body. Although it is generally accepted that hormones play an important role in inducing oocyte maturation in both vertebrates and invertebrates, the investigation of this field has been restricted almost exclusively to aquatîc animals such as starfishes and teleost fishes. The ovaries of these animals contain large nunibers of oocytes that are relatively easy to maintai.. in vitro. The development of an in vitro system, using GVBD as a biological indicator of hormone action, was a great stimulus to work in this field. Investigations from Our laboratory and others indicate that oocyte maturation in these organisms is regulated by a series of interdependent hormonal actions. Three mediators have been described; gonadotropin (GTH) or gonad-stimulating substance (GSS), maturation-inducing hormone (MIH). and maturation-gromoting factor (MPF) (Fig. 1).

. Starflshes - Salmonld flshes 1 GTH 1

Cholesterol

t / Ma,tur:tlon-promotlng factor (ht?Fl 1 al +i % cdcZ klnase cyclln B C) 8 -Oocyte maturation -

Ffg. 1. Mechanisrns of oocyte maturation in starfishes and salmonid Bshes

In this presentation, 1 shall focus on the comparative aspects of regulatory mechanisms of oocyte maturation (starfishes versus fishes) to delineate a general pattern of this important reproductive phenomenon. Bwlogy lnicrnalio~l (Spccial Issuc NT8 -1993) Gonad-Stimulating Substance (GSS) or Gonadotropin (GTH) - Primarg Mediator

The primary hormone involved in starfish oocyte maturation has been called GSS. GSS activity is present in the granules located in the basai part of the supporting cells of the radial nerves. Purifled GSS is a heat-stable polypeptide with a molecular weight of about 2.1 Kd, consisting of 22 amino acid residues.

Two chemically distinct GTHs (GTH 1 and II), like teîrapod follicle stimulating hormone (FSH) and luteinizing hormone (LH), with apparent molecular weight of 30-50 Kd have been purified from salmon pituitaries, both of which are composed of a and P subunits. The P subunit of salmon GTH II (maturationai GTH) has approximately 30°h homology with salmon GTH IB. GTH II appears to be structurally more LH-like than GTH 1. whereas GTH 1 is more FSH-like than GTH II. It has been reported that GTH 1 and GTH II are produced in two distinctly different gonadotrope cell-types in the proximal pars distalis of salmonids at ail stages of reproductive development. An increase in plasma GTH II, but not GTH 1. occurs immediately prior to oocyte maturation, and is responsible for triggering meiotic maturation.

Maturation-Lnducing Hormone 0 - Secondary Mediator

Neither GSS in starfishes nor GTH in Bshes has apparent biological effects on oocyte maturation, but these hormones act on the follicle ceils to produce a second hormone, MIH. MIH has been isolated from the starfish, Asterias amurensis. ovary and identified as 1-methyladenine (1-MeAde) Fig. 2).

1-MeAde produced under the influence of GSS is not a breakdown product of some 1-MeAde-containing substances, but is newly synthesized. The action of GSS on 1-MeAde production by starfish ovarian foilicle cells is initiated by a receptor-mediated activation of chorela toxin-sensitive G-proteins, resulting in the activation of adenylate cyclase and CAMP formation.

Starfishes Salmonid fishes

CH3 I HO- CH

CH3-a) &--oH

N O

Fïg. 2. Maturation-inducing hormone of stariish and sahonid flsh Biology International (Special Issue Nb28 -1993)

The MIH of amago salrnon has been identified as 17a, 2OP-dihydroxy-4- pregnen-3-one (17.20-DP) (Fig. 2). Further investigations from Our laboratory and others suggest that 17.20-DP functions as the MIH common to several fish species. A two ce11 type model was proposed for the follicular production of 17.20-DP based on the results of incubation studies using various follicular preparations. In this model, the thecal ce11 produces l7a-hydroxy- progesterone (17-P) that traverses the basal lamina and is converted to 17.20- DP by the granulosa cell where GTH acts to enhance the activity of 208- hydroxysteroid dehydrogenase (20B-HSD). The first step of the simulating effect of GTH in both thecal and granulosa cells is the receptor-mediated activation of adenylate cyclase and formation of CAMP. There are progressive increases in the number of GTH receptors in both thecal and granulosa cells during follicular development. GTH- and CAMP-induced 17-P production in the thecal ce11 requires a new protein synthesis. Unlike thecal cells, granulosa cells lack the side-chain cleavage cytochrome P450 enzyme. GTH, CAMP and agents which raise intracellular CAMP al1 enhanced 2OP-HSD activity when granulosa cells obtained from fully grown foliicles are incubated with 17-P. The in vitro studies using both protein synthesis and RNA synthesis inhibitors suggest that GTH causes the de novo synthesis of 20P-HSD in the granulosa cell through a mechanism dependent on RNA synthesis.

Along with estradiol-17B which was identified as the major mediator of oocyte growth, we now have two known biologically important mediators of oocyte growth and maturation in female fishes. A dramatic switch in the steroidogenic pathway from estradiol- 17B to 17, 20-DP occurs in ovarian follicle cells immediately prior to oocyte maturation. This switch is a prerequisite step for the growing oocyte to enter the final stage of maturation, and requires a complex and integrated network of gene regulation involving cell-specificity. hormonal regulation. and developmental patterning. To investigate the molecular basis for this switch. Our current efforts center around the cloning and sequencing of the genes encoding several steroidogenic enzymes (cholesterol side-chain cleavage cytochrome P450, P450scc; 3B- hydroxysteroid dehydrogenase. 3P-HSD; 17a-hydroxylase/ 17.20 lyase cytochrome P450. P450 17a; aromatase P450, P450arom; 20B-hydroxysteroid dehydroganase, 20P-HSD) responsible for the production of estradiol-17B and 17.20-DP. We have isolated and sequenced the cDNAs encoding the rainbow trout P450scc. 3B-HSD. P45017a and P450arom. Northern blot analysis of rainbow trout ovarian follicles at different stages of development have shown that the levels of mRNA specific for aromatase are abundant in vitellogenic follicles, but undetectable in postvitellogenic follicles. follicles during oocyte maturation. and postovulatory follicles. Biology Iniernaiionol (Special Issue NT8 -1993) Maturation-Promothg Factor [MPF) - Tertiarg Mediator

1-MeAde and 17.20-DP .are ineffective in inducing oocyte maturation when microinjected into immature oocytes. but external application of these steroids are effective. Specific 1-MeAde and 17.20-DP bindings are present in oocyte cortices. (%) 1-MeAde binds to cortices isolated from full-grown propahase- arrested oocytes of the starfish. The binding of (%) to cortices was rapid and reached equilibrium in 15 min. This is in excellent agreement with the hormone-dependent period required for the induction of oocyte maturation. Analysis of Scatchard plots of the equilibrium binding of (%)1-~e~deto cortices indicates the existence of a single class of binding site with a dissociation constant of 0.3 pM and a binding capacity of 0.02 fmol per cortex. Whereas biologically active analogs did not. These results suggest that the 1- MeAde binding is necessary for the maturational action of 1-MeAde on starfish oocytes.

Specific binding of (3~)17.20-DP to plasma membranes prepared from defolliculated oocytes of rainbow trout was identified and characterized. Binding was rapid and reached equilibrium in 30 min. 17.20-DP strongly inhibited (%)17.20-DPbinding in a competitive manner. Scatchard analysis revealed two different binding sites: a high affinity binding site with a Kd of 18 nM and a Bmax of 2pmoles/mg protein; and a low affinity binding sites with a Kd of 0.5 pM and a Bmax of 10 pmols/mg protein. This binding activity was successfully solubilized with a n-heptyl-p-D- thioglucoside. (3H) 17,20-DP binding to solubilized preparations reached equilibrium in 1 hr., and was competitively inhibited with 17.20-DP. However, Scatchard analysis showed a single binding site wiîh a Kd of 0.3 FM. These results demonstrate that z specific binding site for 17.20-DP exists in the plasma membrane of rainbovr trout oocytes.

Taken together, these results suggest that a MIH receptor exists in the oocyte plasma membrane. The early step following MIH action involves the formation of the major mediator of MIH, MPF. MPF activity cycles during 1-MeAde- and 17.20-DP-induced oocyte maturation, peaking at the first and second meiotic metaphase and abruptly disappearing after fertilization. It is of particular interest that goldfish MPF can induce GVBB in immature oocytes of starfish, although 1-MeAde treatment does not exert any effect on meiotic maturation in goldfish oocytes.

Recently. MPF has been purified from mature oocytes of starfish and carp; in both species, purified MPF is heterodimeric protein corngosed of cdc2 kinase and cyclin B. A catalytic subunit of MPF is cdc2 kinase, the homolog of the serine/threonine protein kinase encoded by the fission yeast cdc2 gene. A monoclonal antibody against the PSTAIR sequence of cdc2 kinase recognizes both starfish and fish cdc2 kinase. A regulatory subunit of MPF is cyclin B, which was first discovered in the early embryos of marine invertebrates. Amino acid homology between starfish and fish cyclin B is approximately 40%. Western blot indicated that immature starfish oocytes contai. cyclin B protein. which is destroyed at the end of first meiosis, acculated at the second meiotic metaphase and.then destroyed again. In contrast, cdc2 protein is present throughout two meiotic cycles. Immunoprecipitation of oocyte extracts indicated that in immature oocytes, al1 of cylin B is already complexed with a part of cdc2 protein. After 1-MeAde addition, MPF activation is accompanied with the dephosphorylation in the cyclin-complexed cdc2 protein, while MPF activation at cyclin destruction is accompanied with partial phosphorylation of cdc2 protein. Thus, the formation of the complex is not a direct cause for MPF activation in starfish; MPF precursor is the complex of phosphorylated cdc2 protein and cyclin, and the activation of MPF occurs in this complex. Indirect immunofluorescence staining has revealed that all of the cyclin B is already associated with cdc2 in immature oocytes arrested at the G2/M border and that this inactive complex is present exclusively in the cytoplasm. After its activation, one part of MPF relocates into the germinal vesicle and accumulates in the nucleolus and condensed chromosomes. while some other part of MPF associates with meiotic asters and spindle. Thus, in addition to the intramolecular modification of cdc2-cyclin B complex, its intracellular relocation plays a key role in promoting the M phase. .- Using monoclonal antibodies against the PSTAIR peptide and E. coli-produced goldfish cyclin B, we have examined the levels of cdc2 and cyclin B during goldfish oocyte maturation induced by 17.20-DP. Protein cdc2 was found in immature oocyte extracts and did not remarkably change during oocyte maturation. Cyclin B was absent in immature oocyte extracts and appeared -nrhen oocytes underwent GVBD. Cyclin B that appeared during oocyte maturation was labelled with (35~)methionine. indicating its de nouo synthesis. The introduction of E. coli-produced cyclin B into immature oocyte extracts induced cdc2 activation, which was associated with threonine phosphorylation of cdc2 and serine phosphorylation of cyclln B. as found in oocytes matured by 17,20-DP. Clyclin B-induced cdc2 activation was not induced by inhibiting threonine phosphorylation of cdc2 by protein kinase inhibitors. These results suggest that 17,20-DP induces oocytes to synthesize cyclin B, which in turn activates pre-existing cdc2 through threonine phosphorylation of cdc2.

Conclusion

The results described above indicate that the basic mechanisms involved in oocyte maturation are the same in both starfishes and fishes, despite the differing chernical nature of the hormonal agents involved. Thus, investigations of oocyte maturation using these aquatic animals as models will grovide new infosma'taon of general selevance and insights and testahle hypotheses for verlflcatbn in sther agngnzads. Biology Infernafional (Speciol Issue N028 -1993) References

KANATANI, H., SHIRAI, H.. NAKANISHI. K. & KUROKAWA. T. 1969. Isolation and Identification of Meiosis-Inducing substance in Starfish Asterlas Amurensls. Nature. 221:273-274. NAGAHAMA. Y. & ADACHI. S. 1985, Identification of a Maturation-Inducing Steroid in a Teleost. the Amago Salmon (Oncorhynchus rhodiuus). Da>. BbL 109:428-435. NAGAHAMA. Y. 1987. Endocrine Control of Oocyte Maturation. IN: Norris. D.O. & Jones, R.E. Ws.) Hormones and Reproductfon in FLshes. Amphibtans, and Reptiles. New York, Plenum Press. pp. 171-202. KISHIMOTO. T. 1987. Regulation of Metaphase by a Maturation-Promoting Factor. Develop. Gmwth Differ. 30.105-1 15. LABBE, J-C., CAPONY. J-P., CAPUT. A., CAVADORE. L-C.. DERANCOURT. J.. KACHAD. M., LELIAS, L-M., PICARD, A.. & DOWE, M. 1989. MPF from Starfish Oocytes at First Melotic Metaphase is a Heterodimer Containing one Molecule of cdc2 and one Molecule of Cyclin B. EMBO. J. 8:3053-3058. YAMASHITA. M.. FUKADA. S.. YOSHIKUNI, M., BULET, P.. HIRAI. T.. YAMAGUCHI. A., LOU. Y- H.. ZHAO, Z., & NAGAHAMA. Y. 1992. Purification and Characterization of Maturation-Promoting Factors in Fish. Dev. Bfol. 148:&15. Biology InterMtio~l(Special Issue NT8 -1993) Glycosphingolipids: Important Membrane Components Rather Neglected in Biotechnology

by Motonori Hoshi Department of Life Science and Cene Research Center, Tokyo Institute of Technology. Yokohama 227. Japan

Recent progress in ectobiology (ce11 sociology) attracts so much of general attention to biological functions of glycoconjugates that the concept of glycobiology has been evolved. One of the most important biological functions of glycoconjugates seems to me a fine tuning of interactions between the milieu surrounding a celland the gene system in it. They smoothly harmonize a rather invariable factor, the geqe system, with a highly variable factor, the milieu. A significant chemical feature of sugars to have multiple linkage-sites renders them to construct a variety of structures even from a few building blocks of much limited kinds. Therefore it is economically reasonable to use carbohydrates to form diversity.

In this context. it is interesting that Zuckerkandle and Pauling (1965) classified molecules in living matter into three categories according to the degree to which the specific information contained in an organism is reflected; semantides. episemantides and asemantides. Arnong four major groups of biomolecules. nucleic acids and proteins constitute the semantides, i.e., the molecules that carry the genetic information or a very extensive translation thereof. Two other groups, namely carbohydrates and lipids, do not express extensively the information contained in the semantides, yet they are a product of this information. These molecules are therefore called episemantides. In other words, while the first two groups are the molecules built by using a gene (primary semantide) or m-RNA (secondary semantide) as a template, carbohydrates and lipids are those synthesized under the control of tertiary semantides (enzymes) in the absence of a template. Since the semantide is a copy of a gene, either directly or indirectly. they are favorites or more preferable targets in biotechnology, especiaily in gene manipulation.

Glycosphingolipids (or glycosylceramides) are a hybrid between two different groups of episemantides, carbohydrates and lipids. The lipid portion called ceramide consists of a long chain amino alcohol (sphingosine base or long- chain base) to which a fatty acid is linked through an amide bond. The simplest glycosphingolipid. cerebroside (galactosyl ceramide; Galf31-1Cer is the one discovered first in 1874. Glycosphingolipids containing sialic acid as a sugar component are called gangliosides. As suggested from the names of cerebroside and ganglioside, glycosphingolipids had been thought unique components of the brain and nervous tissues until similar compounds were isblated from erythrocytes by Yamakawa and Klenk independently in 1951.

Glycosphingolipids are ubiquitous membrane components particularly in animals, and the majority is assumed to be present at the outer leafiet of plasma membranes. although their exact organization in membranes remains unknown. Aniosotropic distribution of glycosphingolipids in plasma membranes Bwlogy Internaiwnal (Specùai Issue N58 -1993) suggests two possibilities (Hakomori, 1981). First, they may contribute to the structural rigidity of the surface leaflet. It is known in mode1 membrane systerns that ceramides confer greater structural rigidity than glycerides. Cepmides with sugar may confer even higher stability. Second, they are well suited to interact with exogenous ligands through their carbohydrates or to regulate functions of membrane proteins such as receptors and channels. In other words, they are suitable as cell surface interactants. transducers and regulators of transducers. Indeed, modulation of glycosphingolipids in the plasma membrane causes changes in ce11 function and rnorphology. Glycosphingolipids in the plasma membrane could be modulateci by addition of exogenous glycosphingolipid ont0 the plasma membrane, removal of intrinsic glycosphingolipids from the plasma membrane by an enzyme treatment. and specific binding of antibody (or other recognition molect!les] to them. From the results of such experiments, various functional notions of glycosphingolipids are published: cell surface markers and antigens of normal and tumor cells; differentiation inducers and markers; growth regulators and regulators of growth factor dependent phosphorylation of receptors; interactants towards cells, virus and bioactive factors like bacterial toxins, neurotransmitters, hormones, lymphokines, opiate and morphine, and interferon. Furthemore, it has been proposed that there are ganglioside- dependent, ecto-type signal transduction systems that may play a key role in cell-to-ceil interaction and recognition (Naigai & Tsuji, 1988).

We have isolateci and chemicaiiy identaed various novel glycosphingolipids in the gametes of sea urchins and starfishes (Fig. 1.). Most of them have a quite "exceptional" structure.

FIg. 1. Predicted synthetic pathways of giycosphingoiîpîds in echinoderms. Part (1)

GkNACpl-tGkgl-1Cor (7)- G.INA@4GlcNle $14lcpl-1C.r i6 tt Fuca1 Fueai GIcp14Gkgl-lCof4b oiii314*Pl.lC~(?) - wptawpi4wcgincw i6

Cer, ceramkie; 7. not identlfled: *. dfsmved in our kb Cer, ceramfde; 7, not identifled, O, discavered in our lab Since the eggs are extraordinarily large ceils. it is expected that the content of plasma membrane components such as gangliosides is relatively smd in eggs. Nevertheless, sea urchin eggs are one of the most rich sources of gangliosides. The most predominant egg ganglioside is NeuGca2-6GlcBl-1Cer (M5) in several species of sea urchins. It was surprisingly found by indirect immunofluorescent microscopy that this ganglioside is present mostly, if not all, in the cortical cytoplasm as a meshwork (Figs. 2 and 3). The immunostaining images are apparently identical to the cortical endoplasmic reticulum or to the ca2'- sequestering system. Upon fertilization, or upon exocytosis of the cortical granules, this ganglioside dislocates from the cortex. prohably coinciding with disintegration of cortical endoplasmic reticulum, to patches in the inner cytoplasm. After reconstruction of cortical endoplasmic reticulum. the meshwork of endoplasmic reticulum becomes positive to the antibody against M5 again (Kubo & Hoshi. 1990; Shogomori, Chiba & Hoshi. in preparationj. It is implied from preliminary data that a minor sperm ganglioside of the structure to be elucidated seems participating in the exocytosis of the acrosome in sea urchins (Hoshi. unpublished data). These observations suggest a role for gangliosides in the exocytosis, presumably as a modulator in the regulatoq system of cytoplasmic free ca2+.

Biotogy Internaiional (Special Issue N78 -1993)

mg. 3. Locaiization of M5 in the endoplasmic reticulum in the cortex of unfertilized eggs. Cortices isolateci as cortical lawns from unfertfllzed eggs were flxed and double stained with the aniibodies to probe M5 (BI and the carbocyanine dye. DiICl8 (3) to visuallze the endoplasmic retIculum (C). The images were obtained with a fluorescence microscope (B,C) and a differential interference contrast microscope (Al. Cortical granules adhering to the plasma membrane are shown in the image C. nie bar indicates lm.(Shogomori. Chiba, & Hoshi. unpubiished).

We anticipate that. from the potentiai importance of glycosphingolipids in ectobiologicai as well as intracellular phenomena, biology of glycosphingolipids will reveal novel features of life. probably more sophisticated and complicated ones. Together with rapid progress in the gene cloning of glycosyltransferases, glycosidases and other sugar-related proteins, in the chernical synthesis of complicated glycosphingolipids, and in some other related techniques, glycobiology wiii constitute a fertile wing of biotechnology.

References

HAKOMORI. S. 1981. Glycosphingolipids in Cellular Interaction. Differentiation. and Oncogenesis. Annu Rev. Btochem 5Q733-764. KUBO. H. & HOSHI, M. 1990. Immunocytochemical Study of the Distribution of a Ganglioside in Sea Urchin Eggs. 3. Blochem 108:193-199. NACAI. Y. & TSUJI. S. 1988. Ce11 Biological Significance of Cangliosides in Neural Differentiation and Development: Critique and Proposals. IN: Ledeen. R.W.. Hogan. E.L. Tettam-nti. G.. Yates. A.J.. & Yu. RK. (Eds.) New 'ikencis in Ganglloslde Resenrck Neumhemical and Neuroregenerattw Aspects. Fidia Research Series, Vol. 14.. LMana Press, Padova. pp. 329-350. ZUCKERKANDL. E. & PAULING. L. 1965. Molecule(i as Documents of Evolutionary History. J. nieoret BLd 8:357366. Bwlogy Inlernational (Special Issue NN58- 1993) Hatching Enzyme of Medaka Molecular Aspects of Its Formation and Packaging in the Hatching Gland Cells

by K ~ama~ami~,S. yaswu2. H. shimada2. 1. 1uchi1 '~ifeScience Institute. Sophia University. 7-1 Kioicho. Chiyoda-ku. Tokyo 102, Japan, %logical Institute, Faculty of Science. Hiroshima University. Kagarniyama. Higashi-Hiroshima 724. Japan

The hatching enzyme is an enzyme which is secreted from embryo of many animals and participates in the breakdown of the egg envelope at the tirne of hatching. This enzyme is present in the pre-hatching embryos but is absent from the post-hatched larvae of these animals. Therefore, it has been considered that this enzyme is fomed only in the embryos and is a good probe for analysis of mechanisms of formation of embryo-specific proteins and its regulation in connection with ce11 differentiation during development.

Different from al1 hatching enzymes documented so far for many animal species, the enzyme of medaka, Oryzias latipes, was found recently to be an enzyme system composed of two distinct proteolytic enzymes, high chonolytic enzyme (HCE) and low choriolytic enzyme (LCE). During embryonic development of this fish, they are synthesized in differentiating hatching gland cells, stored in the secretory granules of the cells, and secreted from there at the time of hatching.

The present report describes some biochemical and physiological features of these two constituent enzymes. cloning and structure of their cDNAs, and the pattern of their intraglandular synthesis and packaging into the secretory granules of the gland cells during development. Throughout the present experiments, outbred orange-red variety of medaka was employed as a material. The embryos were cultured at 30°C with shaking and the synchronously developed embryos hatched on day 6, if the day of fertilization was named day 1.

lMedaka Hatching Enzyme as an Enzyme System

HCE and LCE were isolated at first from the hatching liquid, and they are now found to be colocalized to the same hatching gland cells of the pre-hatchiiig embryos. This fact, together with the finding that the purified HCE and LCE can cooperate in an efficient solubilization of the egg envelope (see following), would provide a substantial cellular foundation for a view that the medaka hatching enzyme is an enzyme system. The hatching liquid of the outbred medaka embryos contains two microheterogeneous isoforms of HCE (HCE-1 and 2). but they are very similar and cannot be discriminated by any fractionation procedures except HPLC. Thus, a mixture of them is usually employed as an WCE sample. No ,isoforms of LCE have been detected.

HCE and LCE are both Zn-proteases wiaki a considerable similarity in thelr Biology International (Spccïal Issue N48- 1993) protein chemicai characteristics and some enzymological properties. Apparent molecular weights estimated by SDS-PAGE are 24 000 for HCE and 25 500 for LCE and their pIs are about 10.5 and 9.8, respectively. On account of the close resemblance in their physical chemical characteristics, they could not be segregated by the older fractionation methods described in our previous reports and the hatching enzyrne of this fish has long been regarded as a single enzyme. though there was a slight ambiguity about its singleness. However, they are now separable on repeated gel filtration chromatography using a Toyopearl HW-50column in a slightiy alkaline solution, in combination with ion-exchange chromatography. Examination using MCA-peptides as substrate shows that their substrate specificity is not so high. Both of them exert caseinolytic action with optimal pHs of about 8 or a littie more alkaline side. EDTA is a potent inhibitor of their activity.

In spite of the high similarity in their apparent nature as caseinolytic enzymes. their modes of action toward the natural substrate, the hardened egg envelope. are markedly different. Taking advantage of these different actions, HCE and LCE perform a cooperative choriolytic (egg envelope-digesting) action efficientiy: when purified HCE alone was applied to the chorion fragments, it swelled the inner layer of the chorion remarkably, releasing concurrentiy some low molecular weight peptides. This process is named choriolytic swelling. When the HCE acts on the chorion and swells it, the turnover of this enzyme remarkably slows down. The slowing down of the turnover is attributable to a tight binding of the HCE to the inner layer of chorion. Thus. the once swollen chorions remain without further degradation for a long time. On the other hand. the purified LCE alone cannot affect the intact chorion fragments at all. but solubilizes the HCE swollen inner layer of chorion very efficiently. Thus, a combined application of HCE and LCE to the intact chorion results in a quick solubilization of it without showing any sign of significant swelling of the inner layer so far as examined by light microscopy. It is conjectured that the inner layer is solubilized by LCE as soon as it was swollen by HCE (Fig. 1). In fact, we can identify a little swollen portion of the inner layer in the process of the solubilization only by electron microscopy. This is an outline of the choriolytic process by cooperation of HCE and LCE.

The two constituent enzymes have different immunogenicity and polyclonal and monoclonal antibodies specific for each of them have been raised. Polycolonal antibodies against HCE and LCE were raised in white rabbits and Balb/c mice, respectively. As each antibody was found to be slightly cross- reactive with the counterpart enzyme, the original antibody against one enzyme was absorbed by the other to make the antibody specifically reactive with the former. Monoclonal antibodies against HCE and LCE were prepared through a standard method using Balb/c mice.

cDNAs for HCE and LCE

Recently, we have cloned cDNAs for HCE and LCE. cDNA libraries were constructed in Lgt 11 from poly(~+~~~sfor HCE and LCE, extracteci from day Bwlogy Internotional (SpecLù Issu N"28-1993) 3 embryos, and the cDNAs of the positive clones. obtained by immunological screening, were sequenced. For HCE, two closely similar cDNAs with a nucleotide sequence similarily of 93% have been obtained. The cDNAs (HCE 21 and HCE 23) are 940bp long and 910bp long and contain open reading frames encoding 279 amino acids and 270 amino acids. respectively. According to the analysis of the structure of these cDNAs. it is concluded that HCE is synthesized in the form of a preproenzyme containing a signal peptide of 20 amino acids. a propeptide of 50 (or 59) amino acids, and a mature enzyme protein of 200 amino acids. As described above, there are two microheterogeneous isoforms of the secreted HCE. Whether or not these two cDNAs for HCE correspond to the two isofonns is uncertain at present. The obtained cDNA clone for LCE is 936bp long and contains an open reading frame encoding 270 amino acids. It includes a signal peptide of 20 amino acids and a propeptide of 51 arnino acids. which is presumably split on activation. Mature enzyme portion is 200 amino acids long.

Fig. 1. Choriolytic action of HCE and LCE, constituent proteases of the hafching enzyme of Oryzias iatfpes. Isolated chorions were iricubated in 50mM Ms-HCl-lOmM NaCl (pH 7.5) without (a). with 5pg of the purifieci HCE alone (b) or wlth a mixture of 4.5pg of HCE and 0.5~of LCE (c and d) at mmtemperature for 5 min (c) or 20 min (a. b, and ci). Bwlogy Internalw~l (Special Issue N78- 1993) The amino acid compositions of the mature forms of HCE and LCE, deduced from the nucleotide sequences of the cDNAs are markedly similar to those determined by amino acid analyses of the purified enzymes. NxS and Nfl, which are known as consensus amino acid sequences for possible N- glycosylation sites are found in the propeptide portions of both HCE and LCE. This fact is of considerable interest in relation to the nature of the proenzymes and the mechanisms of their intracellular packaging and activation as mentioned below. A consensus amino acid sequence. HExxH. which is considered to be a putative active site of several Zn-proteases including collagenase, stromelysin, thermolysin and sea urchin hatching enzyme, is also found in each of the mature forms of HCE and LCE. However, concerning the amino acid sequences of the regions surrounding this consensus, HCE and LCE are not so similar to human fibroblast collagenase as is the sea urchin hatching enzyme. Thus, it remains still dubious at present whether or not the 0. Latipes hatching proteases belong to the matrix metalloproteinase family as does the sea urchin hatching enzyme.

Foxmation and Packaging of HCE and LCE in Ratching Gland Cells

The fish hatching enzyme is secreted and exerts choriolytic action at the end of embryonic development, but it has been conjectured from some morphological, radiation-biological and biochemical analysis that the synthesis of this enzyme would start in the early phase of embryonic development. The appearance and accumulation of HCE and LCE proteins during the course of development of the embryos were examined by Western blotting analysis using polyclonal antibodies specific for each enzyme as probes (Fig. 2). HCE protein was discernible in the embryos at day 2 (stage of lens formation), while LCE protein could be identified for the first time in the embryos at day 3 (stage of retinal pigmentation) of development. Both proteins increased markedly at the stage of the next day and continued to increase in parallel until the time of hatching. The discrepancy in the stage of the first appearance between HCE and LCE does not seem to be very significant because it is probably due to a difference in their amounts synthesized in an embryo and/or to a different immunoreactivity of the employed antibodies against HCE and LCE. In fact, the transcription of both genes seems to start at the sarne developmental stage as described below. Besides the bands of the secreted forms of HCE (24K) and LCE (25.5a, two protein bands (36.5K and 34.5a immunoreactive with anti- HCE antibody and a band (38K) cross-reacting with anti-LCE antibody were detected on SDS-PAGE of the SDS extracts of the embryos. These higher molecular weight bands were considered to be the precursors of HCE and LCE. However, their apparent molecular weights estimated by SDS-PAGE were much higher than those predicted from the cDNA sequences of their proenzyme forms (29 470 and 28 200 for proHCE and 28 900 for proLCE). The SDS- PAGE-estimated molecular weights of the proenzyme forms, but not of mature forms. of HCE and LCE were found to be markedly reduced by a previous treatment of the proenzymes with N-glycanase (Peptide: N-glycosidase). This fact strongly suggests that both proenzymes are probably N-glycosylated at their propeptide regions. Fïg. 2. Western blotting anaiysis of formation and accumdation of HCE and LCE in dweloping Oryzias IatIpes embryos. HCE, LCE and the& proenzymes were detected by irnmunostafning ~vithanti-HCE antibody (ieft) or anti-LCE antibody (right). Armwhead and arrow refer to the mature forms of HCE and LCE, respectively. Proteins in the extracts of day 5.5 embryos were stained wlth Coomassie Briiliant Blue (middle). Abscissa, dweiopmental stages in day. F, hatched fry.

By Northem blotting analysis employing 32~-labelledcDNA fragments of HCE and LCE as probes. it was found that the transcripts of both HCE- and LCE- genes of about l.Okb and 1.2kb. respectively, appeared first in day 2 embryos. The amounts of the transcripts increased abruptiy to the maximal level in day 3 embryos, decreased thereafter until hatching. These results are compatible with those of the Western irnrnunoblotting malysisdescribed above. indicating that the transcription of -bottx HCE- and LCE-genes starts co-ncurrently in day 2 embryos and the forrned mRNAs begin to be translated soon after the transcription. The Northem blotting analysis also showed that the amount of the transcript for HCE was maintained approximately two times as much as that for LCE throughout prehatching development.

Localization and molecular forms of HCE and LCE were examined by imrnunocytochemical methods and/or, irnmunoblotting analysis for sections of the hatching gland cells and the extracts of the secretory granules of hatching gland cells (hatching enzyme granules) isolated by the procedure already estabiished. It was of interest to note that the proenzyme forms of both HCE and LCE were detected only in the secretory granules isolated in the presence of EDTA. When EDTA was absent from the medium for isolatiok of the secretory granules, no proenzymes of HCE and LCE could be detected, al1 proenzymes having been activated and changed into the mature forms. In such a preparation, putative propeptides of about 1.3K which were presumably split from the proenzymes. were found together with the mature enzymes. These results strongly suggested that the activation process of the proenzymes at the time of secretion was catalyzed by an enzyme which was inhibited by EDTA. As both HCE and LCE are known to be quite sensitive to EDTA, it is highly probable that the activation reaction of HCE and LCE is autocatalyüc.

Double immunocytochemical staining of the sections of the hatching gland ceiis and the isolated secretory granules by using an RiTC-anti-HCE (polyclonal) Bwlogy Internaiw~i (Special Issue N58- 1993) antibody system together with an FITC-anti-LCE (monoclonal) antibody system revealed that both HCE and LCE were colocalized to the same secretory granules in discrete arrangement (Fig. 3): HCE is localized to the central core of each granule and LCE is distributed at the periphery of it. Such a pattern of localization seem to be compatible with the relative amounts of HCE to LCE, i.e., the amount of HCE is more abundant than that of LCE. It has been reported that in some endocrine or neuroendocrine glands. different secretory peptides are colocalized to the same secretory granules of the gland cells. In these cases, however, the colocalizing different peptides are intermingled in the same granules without showing a discrete distribution. Thus, the colocalization of HCE and LCE in segregated distribution seems to be an unusual case. Taking into account the fact that both HCE and LCE are synthesized almost concurrently in the same hatching gland ceils. the process of segregated distribution of this type in the same granule (HCE is inside and LCE is outside) seems to be hardly explainable at the moment. The mechanism of this phenomenon is to be analyzed in the near future as it could be an exarnple of post-translational regulation of the formation and accumulation of specific proteins in a differentiating cell. Isolation and analysis of genomic DNAs for HCE and LCE are undertaken to clari@ the regulatory mechanism of the coincidental and harmonized syntheses of both the component enzymes.

Rg. 3. Copackaging of HCE and LCE in the same hatching enzyme granules in discrete arrangement. HCE and LCE were stained with polyclonal anti-HCE antibody and TRTC-conjugated anti-rabbit IgG (a], and monoclonal anti-LCE antibody, biotin-conjugated anti-mouse IgG and avidin FITC (b), respectfvely Photograph was taken separately for HCE and LCE stained in the same section.

Conclusion

HCE and LCE (high and low choriolytic enzymes), two constituent proteases of the hatching enzyme of Oryzias iatipes, are very similar in some enzymological as well as protein chemical properties but are markedly different from each Biology InterMiio~1(Special Issue N58- 1993) other in their modes of choriolytic action. The results described here indicate that they are considerably similar in the cDNA structure (or the predicted peptide structure), especially in the structure of the active center, and the length of the preproenzyme form. Transcription of the genes for HCE and LCE seems to start in day 2 embryos and increases remarkably in day 3 embryos. Concurrent accumulation of both the enzyme proteins also starts from day 2 and continues to the hatching stage (day 6). HCE and LCE are stored in proenzyme forms in the same secretory granules of the hatching gland cells: HCE is localized to the central core of the granule but LCE is situated at the periphery of the granule.

References

IUCHI, I., YAMAMOTO, M.. & YAMAGAMI. K. 1982. Presence of Active Hatching Enzyme ln the Secretoy Granule of Prehatching Medaka Embryos. Dev. Growth DjO: 24135-143. MATIRSIAN. L.M. 1990. Metalloproteinases and Their Inhibitors ln Matrix Remodelling. Ti-ends In Genetics. 6: 12 1 - 125. NOMURA. K., TANAKA. H;. KIKKAWA. Y.. YAMAGUCHI. M. & SUZUKI. N. 1991.. The Specificity of Sea Urchtn Hatching Enzyme (Envelysiln) Places it .in the Mammalian Matrtx Metallopmteinase Family. Biochernistry. 30:6115-6123. YAMAGAMI, K 1972. Isolation of a Choriolytic Enzyme (Hatching Enzyme) of the Teleost, Oryzins latlpes. Dev. Bid 29:343-348. YAMAGAMI. K 1988. Mechanisms of Hatching in Fish. IN: Hoar. W.S., Randall. D.J. (Eds.) Fish Physldogy. Vol. 1lA. Acadernlc Press, San Diego. pp. 447-499. YAMAMOTO. M. & YAMAGAMI. K. 1975. Electron Microscopie Studies on Choriolysis by the Hatching Enzyme of the Teleost. Oryzias latipes. Dev. BLd. 43313-321. YASUMASU. S.. IUCHI. 1. & YAMAGAMI, K. 1988. Medaka Hatching Enzyme Consists of 'ho Kinds of Ehteases Which Act Cooperatively. Zod ScL 5191-195. YASUMASU, S.. IUCHI, 1. & YAMAGAMI. K. 1989A. Purification and Partial Characterization of High Choriolytic Enzyme (HCE), a Component of the Hatching Enzyme of the Teleost. Oryzlas htipes. J. Bfochern 105:-211. YASUMASU, S., IUCHI, I., & YAMAGAMI, K. 1989B. Isolation and Some Properties of Low Choriolytic Enzyme of the Teleost. Oryzias latips. J. Bbchern 105212-218. YASUMASU, S., KATOW. S.. UMINO. Y.. IUCHI. I., & YAMAGAMI. K. 1989. A Unique Proteolytic Action of HCE. a Constituent Protease of a Fish Hatching Enzyme: Tight Binding to its Natural Substrate, Egg Envelope. Bfochern. Bbphys. Res. Cornrnu 16258-63. Biology Internotio~l (Special Issu N78- 1993) Differential Response to Mutagenesis Arnong the Spermatogenic Stages of a Fish, the Japanese Medaka, Oyzias latipes

by A. Shima & A. Shimada Laboratory of Radiation Biology, hlogical Institute. Faculty of Science. Untversity of Tokyo, Tokyo 1 13. Japan

It was first recognized in the 1920s that ionizing radiation damages the genetic material in reproductive cells and results in mutations that are transmittable to offspring. Since that time radiation has been used in genetic research as an important tool of obtaining new mutations in experimental organisms. A considerably large body of knowledge has already been obtained concerning the genetic effects of radiation on the germ cells, primarily spermatogonia. of the mouse. For the detection and quantitation of mutations in the mouse germ line, the mouse specific-locus test (SLT) system, which was established primarily by W.L. Russell four decades ago. has been used in combination with low linear energy transfer (Lm-radiation like X-rays or gamma-rays. The recent success by L.B. Russell in the study of the reciprocal relationship between mouse germ ce11 mutagenesis and basic genetics can be attributed to the presence of many germ-line deletion mutations derived in radiation mutagenesis studies.

About 8 years ago, we hoped to develop a new biological system for studying germline mutagenesis. For this purpose, we chose a fish, the Japanese Medaka, Oryzias latipes. In this presentation, Our 8 years' work have been on the development of a possible non-mammalian test system for environmental germ-cell mutagenesis using a fish, the Japanese Medaka, Oryzias latipes. was reviewed. The Advantages of Using the Medaka The Medaka is a freshwater teleost that is native to Asian countries. The major advantages of using the Medaka are that (1) its basic biology is well established by Yamamoto; (2) about 70 spontaneous mutations have been genetically characterized primarily by Tomita; (3) some inbred strains have been established by Hyodo-Taguchi; and (4) wild populations are well studied by Sakaizurni. Tester Menaka Stocks

To establish a multiple recessive tester stock is a prerequisite for a SLT. We first established a multiple visible recessive tester Medaka homozygous for 5 loci. However. the tester fish obtained were reduced in viability and fecundity, and we finaiiy gave up using this stock for further SLT studies. Aiternatively, we ckiose the following three autosomal visible recessive mutations as markers: b (colorless melanophores); If (leucophore-free); and gu (guanineless). These color genes were chosen because their phenotypes can be easily recognized during early embryonic development, and also because the integration of these three mutant genes did not affect the viability and fecundity of the,tester. This tester stock is currently in its 12th generation of brother-sister mating toward Bwlogy Internalio~l (Special lssut N28- 1993) development of an inbred tester strain.

Numbers of Scored Animais

By 20 September. 1991, we exarnined a total of 488 683 embryos. in which 155 451 control embryos are included.

Dose-Rcsponce Relatioanaip for Dominant Lethalo (DL) Fig. 1 sumrnarlzes DL rates as a function of gamma-ray dose delivered to Medaka sperm, spermatids. and spermatogonia A dose-dependent increase over the control value was obvious for the postmeiotics (sperm and spermatids). whereas no significant increase in DL rates could be found for spermatogonia.

Dosle-Reeponse Rclationahip for Total Mutations (Tllb) and Viable Mutations [VM) Based on our preliminary observations that many phenotypically recognized color mutant embryos eventually died during subsequent embryonic developmental stages. a unique observation first derived fmm Our Medaka SLT system. we proposed the term totai mutations for specific-locus mutations that were phenotypically detected during development and viable mutations for hatched viable mutants. Fig. 2 indicates that about 90% of the TMs are elimtnated during embryonic development. This negative selection can be explained by assuming that the induced spectfic-locus mutations were associated with deletion mutations in genomic regions surmunding the specific marker loci. Also shown in Fig. 2 L a marked differential response in terms of frequencies of TMs and VMs arnong the three spermatogenic stages.

Domuianl iethag

0.5 -

2z 0.4 -

Spermalogonia

O 2 4 6 8 10 bse(GY)

PdleMechanisma Responsible for Difîenntial Spermatogenic Response

Two possibilities can be proposed to explain the differential spermatogenic response: (1) Spermatogonia are proficient in DNA repair capability, whereas sperm and spermatids tend to be repair deficient as a result of differentiation to obtain motility: and (2) those gonially derived germ cells with gross chromosoma1 damages are ellminated at mitosis and meiosis to which the postrneiotically exposed spermatids and sperm are not subjected. These should be substantiated by future studh.

Nok This paper is also published in the Proceedlngs of the NaUonal Academy of Sciences, U.S.A. Volume 88. page^ 2545-2549, 199 1. Somatolactin, a New Member of the Growth Hormone and Prolactin Family from the Pars Intermedia of Teleost Fish

by Hiroshi Kawauchi Laboratory of Molecular Endocrlnology. School of Flsheries Sciences. Kitasab University, Sanrfku, Iwate 0220 1, Japan

The pituitary gland secretes a number of peptide hormones that control a wide spectrum of physiological processes in vertebrates. In the past 10 years, we have extensively participated in iden-g most of the fish pituitary hormones and cloning their cDNAs. Al1 pituitary hormones present in mammals and two new hormones, somatolactin and melanin-concentrating hormone. have been identified. In this presentation, 1 will discuss hormones that function in regulating some physiological aspect of reproduction. Gonadotropins

Only a single gonadotropln homologous to LS and FSH was identified from fish pituitary glands and has been thought to regulate al1 aspects of teleost reproduction. However, we have proved duaiity of fish gonadotropin by isolating two chemically distinct molecules, designated GTHI and GTHII, from salrnonids and by characterizing their biological and immunological properties.

The two GTHs are composed of a and P subunits. Sequence analyses revealed that GTHII corresponds to maturational GTH previously characterized. Sequence identity between the two $ subunits is only 31°h (Fig. 1). GTHIB is more similar to bovine FSHP than bovine LHB. and salmon GTHIIB shows higher homology to bovine LHB than to bovine FSHB. Biology Inlernaiional (Special Issue N28- 1993) Both GTHs were equipotent in stimulating estradiol production. whereas GTHII appears to be more potent in stimulating maturational steroid synthesis. GTHI is predominant during vitellogenesis and early stages of spermatogenesis in salmonids, whereas GTHII is predominant at the time of spermiation and ovulation. GTHI and GTHII are found in distinctiy separate cells. In trout, GTHI is expressed first in ontogeny. whereas GTHII cells appear coincident with the onset of spermatogenesis and vitellogenesis. and increase dramatically at the tirne of final reproductive maturation. Growth Hormone

Growth Hormone (GH) is alrnost exclusively involved in the regulation of postnatal somatic growth and the maintenance of rnetabolism in most vertebrates. GH stimulates seawater adaptation in salmonids by reducing plasma ~a+levels, while PRL stimulates freshwater adaptation by maintaining plasma ~a+levels. Recentiy, intrinsic gonadotrophic activity of salmon GH was found using recombinant salmon GH (rsGH) in hypophysectomized Fundulus. Treatment with rsGH significantly increased plasma levels of testosterone in males and estradiol in females. Direct action on gonadal steroidogenesis was also examined by incubating gonadal tissues from hypophysectomized fish in vitro. rsGH significantly stimulated the in vitro production of testosterone and Il-ketotestosterone by the testis, and estradiol by the ovary. The use of recombinant GH rules out contamination of GHTs. Evidently salmon GH has steroidogenic and activity, although physiological significance is not known. Somatolactin

We isolated a new protein, somatolactin (SL). from pituitary glands of Atlantic cod. flounder and salmon. The cellular origin of the protein was determined using immunohistochemical staining with antiserum raised in a rabbit against cod SL. The cells stained with the antiserum were found in large numbers in tile pars intermedia, bordering the neural tissue. No staining was observed in the pars distalis. The pars intermedia of higher vertebrates contains only one ceii type that synthesizes pro-opiomelanocortin (POMC) which is the common precursor of melanotropins and endorphins. However, the teleost pars intermedia contains two ceii types that can usually be distinguished by staining with Lead-Haematoxylin (PbH) and Periodic-Acid-Schiff (PAS). The PbH cells are the source of the POMC-related peptides, whereas SL localized to the PAS- positive cells. The active principle of the PAS-positive cells has been unknown previously.

Immunostaining of pituitary sections from several teleost species including both freshwater and marine fish, showed the pars intermedia of flounder, catfish, killifish, molly and eel to contain a similar protein. Even salmon and trout, which do not have PAS-staining celis but a chromophobic ce11 type in the pars intermedia, showed strong immunoreaction with the antiserum. This implies that the new protein may exist in both glycosylated and non- glycosylated forms depending on species. Biology Inîernational (Spccial Issue N"28- 1993) We determined the amino acid sequences of the proteins and the base sequences of cDNAs. These proteins are structurally similar to GH and PRL with intermediate identity between vertebrate GHs and PRLs, indicating that it probably evolved independently from the comrnon ancestral molecule (Fig. 2). Interestingly, the protein shows slightly higher identity to the tetrapod GHs and PRLs (average 29% identity) than to the teleost hormones (average 24Oh identity). The protein can therefore be regarded as a new member of the GH/PRL family and was named somatolactin. SL may be a missing link that will give a new insight into the evolution of the GH/PRL family. It is now generaiiy accepted that GH and PRL diverged fi-om a common ancestral molecule.

Growth Hormone Somato 1ac t i n Prolact in

SL is secreted from the PAS-positive ceiis as it was detected in plasma. and may therefore function as it was detected in plasma, and may therefore function as a hormone. Cytological studies have suggested that the PAS-positive cells may be involved in regulating calci6m metabolism, acid-base regulation and background adaptation. Although a definite function of SL has not been determined yet, we hypotliesized that SL may function in regulating reproduction. We measured plasma levels of coho salmon SL throughout the final year of reproductive maturation. During the period of gonadal growth. plasma SL levels increased and where highiy correlated to estradiol levels in females and Il-ketotestosterone levels in males. Peak levels of SL were observed at the tirne of finai maturation and spawning in both sexes.

The studies were supporteci by the Ministry of Edumtion, Science and Culture of Japan. the Mftsubishi Foundatbn. the Fishedes Agency of Japan, and the Kitasato Research Foundation. Biology International (Spechl Issue N018- 1993) References

KAWAUCHI. H.. SUZUKI, K., mH,H.. SWANSON. P.. NAITO. N., NAGAHAMA, Y.. NOZAKI, M.. NAKAI. Y.. & lïDH. S. 1988. The duaitty ofTe1eost Gonadoîmpins. Fkh PhysbL Biochem 7:29-38 (1989). SINCH, H., GRIFFITH, R.W.. TAKAHASHI. A., KAWAUCHI, H., THOMAS. P., & STECEMAN, J.J. 1988. Regulation of Gonadal Steroidogenesis in Fundulus heteroclftus by Recombinant Salmon Growth Hormone and Purined Salmon Prolactin. Cen. Cornp. Endocd 72144-153. ONO, M.. TAKAYAMA. Y.. RAND-=VER M., SAKATA, S., YASUNAGA, T.. NOSO. T.. & KAWAUCHI, H. 1990. cDNA Cloning of Sornatolactin, a New Pituitary Protein Related to Growth Hormone and holaetin. Proc. NdAd ScL U.S.A. 87:433û- 4334. RAND-WEAVER M.. BAKER B.I.. & KAWAUCHI, H. 1991. Cellular Locallzation of Sornatolacün in the Pars Intermedia of Some Teleost Flshes. CeU 7ïssue Res. 2a3.207-215. RAND-WEAVER M.. NOSO. T., MURAMOTO. K., & KAWAUCHI. H. 1991. IsolaUon and Characterlzatton of Sornatolactin. a New Protein Related to Growth Hormone and Prolactin hmAtlantic Cod (Cadus rnduq) Fïtuitary Glands. Btochernistry. 33 1509-1515. TAKAYAMA. Y., RAND-WEAVER M.. KAWAUCHI, H.. & ONO. M. 1991. Gene Structure of Chum Salmon Sornatolactin. a Presumed Pituitary Hormone of the Gmwth Honnone/ProlacUn Family. Md EndocrlnoL 5:778-786. TAKAYAMA, Y.. ONO. M., RAND-WEAVER,M.. & KAWAUCHI, H. 1991. Gmter Conservation of Somatoiactin. a Presumed PiNtary Hormone of the Growth Hormone/Prolactin Family, than of Growth Hormone in Teleost Flsh. Gen Comp. EndDcrInd 83:233-374. Biology Internatio~l (Specinl Issue N58- 1993) cDNA Cloning and Structure of Teleost Growth Hormones and the Growth Promoting Activity of Recombinant Hormones

by K. Nakashima, M. Watahiki, M. Tanaka Department of Biochemistry. Mie University School of Medicine. Tsu, Mie 514. Japan

Growth hormone (GH) is an essential growth-promoting hormone for the vertebrates, and composes a gene farnily together with prolactin. placenta1 lactogen and related proteins. Recent development of molecular biology enabled efficient analysis of several teleost GH cDNAs and proteins. In an effort to clarify the structure-function relationships and molecular evolution of the growth hormone gene family in vertebrates. we have isolated and analyzed GH cDNAs of yellowtail (Seriola quinqueradiataj, hard tail (Cmax delicatissirnus) , and flounder (Pardichthys oliuaceus). To ver@ the growth-promoting activities of the deduced GH molecules, recombinant yellowtail GH (ryGH) and flounder GH (rfGH) have been synthesized in Escherichia coli and refolded. Biological activities of the recombinant hormones were tested on juvenile rainbow trout (SaZrno gairdned which exhibited enhanced growth rates as compared to the control groups.

Yellowtail GH WH)cDNA cDNA library was constructed with poly (A)' RNA isolated from yellowtail pituitary gland using pSI4001 cloning vector. Full-length yGH cDNA clones were selected by colony hybridization with the DNA probe prepared from porcine GH cDNA. YGH cDNA consisted of 879 nucleotides. and encoded pre- yGH of 204 amino acid residues. Putative signal peptide contained 17 arnino acid residues, and predicted mature yGH contained 187 amino acids including 4 cysteine residues.

Hard Tai1 GH (htGH) cDNA htGH cDNA clones were isolated from hard tail pituitary cDNA library with radio-labelled DNA probe prepared from yGH cDNA. Full-length htGH cDNA contained 819 nucleotides encoding pre-htGH of 206 amino acid residues. Mature htGH was predicted to be composed of 188 arnino acid residues.

Flounder GH (fGH) cDNA

Flounder cDNA library was constructed from flounder pituitary, and fGH cDNA clone was isolated by colony hybridization with yGH cDNA probe. Full-length fGH cDNA was composed of 834 nucleotides. and coded pre-fGH of 190 amino acid residues. Putative signal peptide was composed of 17 amino acids as weii, and predicted mature fGH was found to contain only 173 amino acid residues. Molecular size of fGH is extraordinarily small and the minimum in the members of the growth hormone prolactin gene family ever reported. Homology and Molecular Evolution of GHs

Amino acid sequence of mature ht GH shows 79. 74, 72, 59, 56. 38. 37, 33, and 3Wh identity with those of yellowtail, tuna. sea bream, flounder. salmon. eel, blue shark. bullfrog, and human GHs. respectively. Nucleotide sequence homology between the coding region of htGH and those of yeliowtail. tuna. sea bream, flounder. salrnon, eel, and human are 84, 75, 72, 63, 58, 51, and 51°h respectively. Dot ma- anaiysis of nucleotide sequences of these coding regions aiso shows good correlations to the order of phylogenetic evolutians (Fig. 1). On the other hand, nucleotide sequences of the 5'- (Fig. 2) and 3'-non- coding regions of htGH (Fig. 3) exhibited homology only with those of phylogeneticaiiy closer species such as yellowtail, tuna, or sea bream. This suggested that the scores of homology in the non-coding regions would be useful to estimate the closer evolutionary distances among the vertebrates.

(Bl (Cl

(El (FI ((3) (Hl

Q Fig. 1. Dot mat* analysis of nucleotide sequences of coding regions of teleost CH cDNAs. Dot matrïx anaiyses between the coding regions of htCH and other teleost and human CH cDNAs were carrieci out. The stringency was set such that a dot denotes a match if 10 out of 15 consecutive nucleotide were identical. The horizontai axes represent the sequence for hGH, and the vertical axes are the coding sequences of yellod W. tuna (BI. red sea bream (C).flounder (D), churn salrnon (E), carp (n.eel (C). and hurnan (Hl CH cDNAs.

(Al [BI [Cl (Dl (El (FI

mg. 2. Dot matrix anaiysis of 5'-noncoding sequences of teleostean CH cDNAs. A dot donates a match of 7 of 1 1 nucleotides. The horizontai axh is hK;H cDNA and vertical axes are as in Fïg. 1 exqtthat (Fj and [Cl are eel and carp CH cDNAs, respectively. (Al (8) (Cl (0) (El (FI ((31 (Hl

FIg. 3. Dot mate anaiysts of 3non-codtng scquenccs of teleostean GH cDW. A dot donatea a match of 10 of 11 nuckotides. Axes are set ~9 tn the kgend to Fi& 1. ex-t that vntical oncs in M. (Cl. and (Hl axe coho salmn, carp. and a1 GH cDNAs. respectively.

Elucidation of structures of several teleost GHs including fGH enabled more precise analysis of the conserved domains of the GHs. 29 amino acid residues are conserved throughout the known GHs, and these residues are clustered in 5 domains. These GH speciflc domains were designated GD1, GD2. GD3, GD4, and GD5 (Fig. 4). Four domains except for GD2 correspond to 4 a-helix regions of the molecule. GD1 and GD5 are Iocalized at the amino- and carboxy- terrnini. respectively. These 5 conserved domains are considered to be structurally and functionally important for the GHs. GD 1 GD 2 GD3 Il O 20 J0 : LO 50 60 1 70 BO 90 hlGH OP IPNNO~~~~H~R~LFA~SKOSODW~-NKIFLODF~~1 1 SOIDK~~'L~ ImmEm ISSMFm-GGL- IGH OP ITENOR~SI~O~~KLF~NKPLEDQRLO-NU IASUEF~WFLW IDU~S~VPK~SVÊYRO IE~FFSRFSI-AY - yGH OP ITDW~SI~~IO~L~RLF~STLOTEDW~-NKIFLODF~~1 1 ~~DU~~L~SI~~E~FSSRF~S-GGS- iGH OP ITDWR~SI~SI~((O~L~RLFSD~SKOTEEOR~-NLIFLODFS(m1 I~IDL~~L~SI&R~E~FPSR~S-GG~- sbGH OP ITDC~R~S~~~~~L~RLFSD~SKOTEEOLK~-NKIF -PDFWW 1 IWIDK~~LK~SI&ROVE~FPSRQS-GGS- i iGH 90 IIDSO~SIwTmLNRLF SD3SYOTEEORCU-NL 1FLODFmW1 1 WIDL~~L~SI&C~VEB!?,EFPYISBS-GGS- rlGH IENO- 15WO~LmKLFNDD3GTLLPDERRd-NKIFLLDF~SOSIV~DK~U~L~~~R~IE~YPY~T~-- 1 1 SN sGH IENORnTi lmWO~LmKWN@GTLLPDERR~-NUIFLLDFENSOSIVSOVDK~~EVL~<~'~ I&ROIEWYPSOT3-- 1 1 SN CSGH IENQW15~O~LmKLFN~GTLLPDERRcù-NU IFLLDF~~SIV~DK~~LK~ I&wlEmYPSOTB-- 1 1 SN caGH SDN~~~~~~~~E~PL~KMIND~DKLPEERR~-SKIFPLYS(SOY IEA~TGKUXENSPL-lDR3 1E~FPSOTBSGTVSN gcGH SENQ~~I~O~L~KMIN~DNLLPEERR~-SK~FPLSFS(C~SIE~GK~~K~I&~IE~FPWTBSG(IVSN &H VEPIVY~~Q~L~EIYKE~RSIPPEAHR~-SKT~PLA~~SIPT~GK~I

CGH TFPALPLSNEL&L~O~LAAETYKEFERTYIPEDORYT -NKN~~~AF&SETIP~GKDDAWKSDME~RFOCV~ I~TPVOYLSKVFTN rGH LPALPLS~~LD*O~LAADTYKE~ERAYIPEGORYS- IQNAPUF3SE1 IP*JDTGKEEAOOATDYE~B~FDLB1-GPVWLYI IFTN ~GH FPT IPL-NSLDHWLAFDTYOEFEEAYIPKEOKYYLONPOT 9-3 SES IPT~Y~REETOQKSNLEXB? IBL~I~~ZAEPVQCLRSVFAN

GD& GD S 1 O0 1 110 ,120 ' 130 140 150 '160 170 180 1 h IGH AE --R--54 1S-%E~E#IO~ITTNOEGAEVF SDSSTLPLAPPFGNFFOT(X;COELLORRSYE~~T~~STE~ IGH AV--R--TOVT -SKBSE~~LK3IEANODGAGGFSESSVLOLTPY-G------NYw-TwKmFPEm yGH AL - -R- -NO 1S-PR~sE~T~ IOL$ ITANODGAEW SDVSALOLAPY-GNFYOSLGGEELL -RRNYEs-@&Km SPEAm IGH AP- -R--NO~S-PK~SE~LPPY-GNYYQNPGGDGNV-RRNYE~~T~K~SLEA~ csGH SLLVGNANO 1SEU-BSO~K@ INL~I~~~S~DGLLKDDNDSO~LPRY -GNYY~NPGGDGNV-RRNYE~~T~K~KSLEA~ caGH KTVGNPN~~TEK-~DO~(%'ISV~IKGCLDGOPNM)DNOSLPLPFE-DFYLT~~~ENNLRE---YR~~~N~SLD~ ~CGH SLTVGNPNO ITEK-DD~G I sVO IKGCLDGOPN~DNDSLPLPFE -DFYLT%E SKRE ---sFRJL-~N~~BKDSEC~Y dj~SLLFGTSDG IFDK-ED~@INE~&%VVGDGG IY IEDVRNLRYENF -DVHLRNDAGLM ----NYG&> , . K&'~K~~BFvESEC~Y

bsGH AF - -RTSDRVYDK-BIDOEEP IFA~TLEDGCSYX.FA~KFSYERFKGNL -sEEALH<----NYG~A~K~~~Kv*~~~KRF*ESN~TV blGH NOVFCN IDRVYDR-BWEaH I~IRELDDGNVRNYGVLTFTYDKFDVNLR-SEEGRAK- ---NYG% ~K~~~~KvM~R~VE SN3TF cGH NLVFGT SD .RvFE@)UD~EEI~ELEDR-SPRGPOLLRPTYDKFD IH IRNEDALLK- -- -NYG& S~~U~H~V~>UVM~F(RFGESN31 1 rCH YLFGTSD -RVYEKBKD~EE~IQA~ELEDG-SPR IGO ILKOTYDKFDAN)*ISDDALLK-- - -NYG& ~K~HB~IRvM%R~AI SS%f hGH YVYCASDSNVYDLBKE&EE IPTWRLEDG. SPRTCO IFKOTYSKFDlNSHNDDALLK - - - -NYC~Y~R~-*~~V~F'(RIV@S-VEC%T FIg. 4. Consuvd amino adds and domains in teleœt and othcr vertebratc mature GHs. Primaxy s~cturcsof hard tail. fiounder. yellowtatl. tuna. red sea bream. tilapla. raînbow hut. chum salmon. coho salmon. carp. gram carp. al. blue shark. bullhg, chicken. rat, and human CHs are comparcd. Amtno adds comcd in the teleost or al1 the vntebrates arc show in the black boxes. respectively. Armwhead indlcate uplasmid was then replaced with that of pUC19 to obtain the multi-copy expression vector pUCfGH 1. pPLyGHl was used to transform E. coli N4830-1 cells, and the ryGH expression was induced by raising the temperature (Fig. 5). On the other hand, E. coli JM109 cells were transformed by pUCfGH1. and rfGH expression was induced by the addition of IPTG (Fig. 6). ryGH and rfGH were synthesized in E. coli cells up to 15 and 8% of the total cellular protein, and were accumulated in the inclusion bodies.

Ng. 5. SDSPAGE analysis of E. wff N4830-1 Ng. 6. SDSPAGE anaiysis of E. cdlJMlû9 cells expressing recombinant yGH. The cells ucprcssing recombinant EH. The cells were induced by raising the culture celis were induced by adding 0.5 mM m. temperature hm30°C to 42%. lanes 1 : E. coll cells with pKKfGH 1. lanes 1: E. coll cells with pPLfGH1. lanes 2: E. cdcells with pUCffiH1. lanes 2: E. cdcells with pPLyCH1.

E. coli cells were disrupted and the pellets were collected. Pellets containing ryGH or rfGH were washed and dissolved in 10% sodium dodecyl sulfate to pass through an ACA 54 column. The GH fractions were precipitated by acetone and dissolved in 6 M guanidine-HCL. The solution was added to the refolding buffer containing reduced and oxidized glutathiones. The refolded ryGH and rfGH were further purified by DEAE column chromatography to homogeneity. Biology ln!crnaiional (Spccial Iss~N 028- 1993) Growth Promoting Activity of rgGH and rfGH

Activities of refolded ryGH and rfGH were verified by the growth promoting test on juvenile rainbow trout. Once a week. O.1pg or 1pg or hormone was injected intraperitoneally to each fish for 5 weeks. With 1pg of WH,Rshes exhibited about 2 times more of an increase in body weight and body length as compared to the saline treated controls (Fig. 7). Somewhat lower activity was observed with 1pg of rfGH in this assay method.

weeks weeks

wesks

Fig. 7. Biologïcal activity of recombinant yGH and CH on juvenile rainbow but A: recombinant yGH. B: recombinant WH. O : control A :O. 1pg rrCH/g body weight :lpg rCH/g body weight Arn>wheads indicate CH injections, and verthl bars indicate standard devlations of 20 flsh in each group. '9101-Z101:9S WaiiWfI Yjaloia 73SolfI 'iP3 aliIWaV3S3 Ul SaUOUJJOH ~3~0~3lapunoy pu= pgmonaA wwqu~wat~JO s=aq~uLs '~661'H WIHSVWNB "n 'V~~NOA "H 'OtIIHsn '-w'VM~~VIWA "W 6m~tr~'Y o~m~vw "n 'II'OHS '-wvana '-3'mo "MI 'MIHVLVM '6%-LPZ:L801 W3ii 'sfiiIdoiEl ul1~3ala.au=oH ~RMW PH JOJ xw3nqs PU- ma3 JO muanbaç ~PUO~I~~N-0661 n WIH!WXVN B '.A 'IH3nDIHSIN "W WNVL "W '0.LObWIWA "A 'FDIOINVX "W 'MIHVLVM "W VMVMWWA '91C -ZIC:m ~ay3 'r 'AIF~ auomo~w0~3 arl7 ul azls lmururW ayl seq muanbas vNa3 aii7 UICLIJ pnpaa auourroH wo~gJapunoId :sauouuoH tpmari) awJO smoaayl tq sanplsw ppv oqunj anbpn pue pNmo3'6861 71 WIHSWVN %' "W V:NV.L "IN VMWWNVA "W '(XLObWIWA "N 'DIIHVLVM '9ot- 1 m:OL 70U1'30PU3 '-03 '~UOUUO~w&q (onp~anbulnbWW) nWMOnaA JO anI3nrlS ~~ pue 8~013 ma3 -8861 -a'VWIHSWN B "A 'V~~NOA'-N 'VM~WNVA "N vanm '-PI 'WNVL "W 'MIHVLVM

.Waleqs 3uyaau@ua waua3 FnplArpw a~rnba~Lem y3ly~ 'vNa3 07 v~aanroq savea sllar, 1103 '3 ur s~3JO uolssaxdxa JO hual31jja 3y.L 'H3 nevonan lueUlquIO3aJ JO 7Eyl 03 AJIAI~El~3180101q ~AI)BJ[~~WO~ e palrqlyxa s~y~3 JapunoU )ueurquIo3aJ ay) pue 'suremop pauasuo;, asaq w palmol ?ou SI ley) uo13a.x e sy3q SH~UMO~ atp JO azls Fmwrnr aq 8upey auouuoq yi~o.x3Japunolj 'qugrodurr A~px~on~un~pue Içne~npnqs aq 07 pa~alpaxda= y3ry~'supnrop aArj l3unslp aq amasuoa sapaalour auouuoy yvo~3ayi 'puey Jayio aq uo .saraads a7sJqa)Jaa Jasop ayq uaamlaq sa~m~srphuonnlo~a aq alenrnsa 03 lnjasn a~esuor8a.1 3wpo3uou ayj ul L8olowoy JO saJoas aqle- qsaBns su 'sapads asop Içne3naua30lAqd ka^ 3uom L~uosno3olonroy am w~a3~3 TsoaIal JO suo@a~8ulpoa-uou-,€ pue -,s aw JO samanbas apnoapnu ay) 'suonnlo~aal)aua8olLqd JO Jap.ro aq qw uor~e~auoapoo8 ul ale wNa3 auonr.roy rp~013ajeJqalJaA ~ay)opue lsoalaj JO saauanbas ppe ou~mpue uo@ar 3urpo3 JO samanbas apnoapnu atp aIlylM Bwlogy Inter~lw~l(Special Issue N78- 1993) Genome Transfer in Teleosts by Yan Shaoyi Institute of Developmental Biology, Academia SWca. Beijing, China

The concept of nuclear transplantation (to transfer a diploid ce11 nucleus into an enucleated egg for investigating the role of nucleus in initiating embgroriic development) was proposed by Hans Spemann (Germany) early in 1938 (see Browder. 1980). This method was first accomplished by Cornrnanndon and de Fonbrune (France) in amoeba (1939) and later in amphibians by Briggs and King (U.SA.) (1952).

Using this method. Danielli (1958) made efforts in analyzing the roles of nucleus and cytoplasm in terms of inheritance in different kinds of amoeba and concluded that both the nucleus and the cytoplasm were of equal importance in inheritance of characters. Nevertheless, the simple structure. short life cycle and limited differentiation of the unicellular amoeba suggested that the mode of its inheritance might be quite different from that of higher organisms.

In amphibians. due to the natural species-specific barrier, most of the nuclei donor and the recipients cytoplasm which were used in experiments were of the sarne species or only restricted in different mutants of the same species. the researchers' interests were always attracted by and focused upon studying the developmental potential of the nucleus during embryogenesis as their investigations shown (Yan, 1989). Hence, in those cases. there is not way to distinguish the different genetic characters which might be determined either by nucleus or by cytoplasm in a nuclear transplantation system of the same species. However. a few exceptional results reported by Kawamura and Nishioka (1963) in amphibians showed that when the nuclear transplantations were performed between different species of Japanese pond frogs, their results indicated that some characteristics of their nucleocytoplasmic hybrids were similar to the species which contributed the cytoplasm or were intermediate. Those results showed that there were cytoplasmic factors also playing an important role in inheritance in higher organisms.

It was not until 1963, when this method was introduced into fish and several kinds of inter-species combinations of the nucleus and the cytoplasm were successfully made, the identification of different genetic roles of nucleus and cytoplasm in a nuclear transplantation system of fish became clear and feasible than those observed in amphibians and thereby several kinds of nucleocytoplasmic hybrid (NCH) fish with modified phenotypes of nucleus-type fish were obtained (Yan, 1989).

Since the nucleus used in nuclear transplantation experiment is a normal diploid one, this method could also be recognized as an intact genome transfer and the modified phenotypes of NCH fish obtained could be explained as a result of the interactions between the transferred genome and its surrounding cytoplasm during the development of the NCH egg. Nuclear Transplantation Procedure in FIsh

polar body gla~needle ...... 1::: ..... :i ...... ; 'i

'-y:::.:: -.+ j (dl

Fig. 1. Enucleation of a flsh egg. [a) Laterd view showing the location of polar body and nucleus. (b) A glass needle is inserted into the egg cytoplasm at the location of the nucleus. (cl The nucleus is removecl with the glass needle. (d) The nucleus is separated from the egg.

Fig. 2. Nuclear transplantation in a nsh egg. (a) A blastula embryo. (b) Isolatecl blastula cells in ~a++free medium. [c) A nucleus of blastula cell was sucked into a micro-pipette. (d) A nucleus was injected into an enucleated egg with a micro-pipette.

The Results of Experiments Obtahed through Genome Transfer

The foilowing experiments were performed in our laboratory in collaboration with other research teams in China* between 1963-1991 (NCH embryos. larval or adult fish were obtained from different combinations of blastula cell nuclei and enucleated eggs).

1. Nucleus from crucian carp (Carasslus auratus, wild type, 2N= 100) + cytoplasm from goldfish (Carassius auraius, domesti c type. 2N= 100). Adult NCH fish with essentials of crucian carp phenotype and some intermediate morphological and biochemical changes were obtained (1985).

2. Nucleus £rom goldfish + cytoplasm from crucian carp. Adult NCH fish with goldfish phenotype were obtained (paper unpublished). if. Inter-genus combinations

3. Nucleus from common carp (Cyprinus Carpio, genus Cyprinus Limaeus, 2N= 100) + cytoplasm from crucian carp (C. awntus, genus Carassius Jarocki. 2N=100). Adult NCH fish with essentials of common carp phenotype plus some intermediate and cytoplasmic influenced characters at morphological,- physiological. and biochemical levels were obtained (1980). (Fig. 3)

Fig. 3. CA) A six-month-old NCH fish obtained from the combinatton of the nucleus of cornmon carp and the cytoplasm of crucian carp; (El) A six-month-old nucleus-donor fish. common carp. cultureci in the same conditions as NCH fish.

4. Nucleus from crucian carp + cytoplasm from common carp. Adult NCH fish with crucian carp phenotype were obtained (1984).

IIL Inter-subfamily combinations

5. Nucleus from grass carp (Ctenopharyngodon idellus, subfarnily Leucinae, 2N=48) + cytoplasm from blunt-snou t bream (Megalobrama amblycephala, subfamily Abrarnidinae, 2N=48). Adult NCH fish with essentials of grass carp phenotype and some biochemical, immunoelectrophoretic changes of its blood serum were obtained (1985). (Fig. 4)

Fig. 4. A two-year-old NCH 5sh obtafned from the combination of the nucleus of grass carp and the cytoplasm of blunt-snout bream.

6. Nucleus from blunt-snout bream+cytoplasm from grass carp cytoplasm. Adult NCH fish with blunt-snout bream phenotype were obtained (paper unpublished) . Biology Inlernational (Special Issue N28- 1993)

7. Nucleus from goldfish (C. auratus, subfamily Cyprininae. 2N=100)+cytoplasm from Chinese bittering (Rhodeus sinensis, subfamily Acheilognathinae, 2N=46). Early NCH embryos with the similar developmentai shape of bittering embryo were obtained (1973).

8. Nucleus from Chinese bittering i cytoplasm from goldfish. Early NCH embryos with some intermediate characters of both kinds of embryos were obtained (1973).

N. Inter-famUy combinations

9. Nucleus from goldfish (C. auratus. subfamily Cyprininae, 2N=100) + cytoplasm from loach (Paramisgurnus dabryanus. family Cobitidae, 2N=48). Early NCH larval fish of loach developing speed were obtained (1990).

10. Nucleus from loach + cytoplasm from goldfish. Early NCH larval fish of goldfish developing speed were obtained (1990).

1 1. Nucleus from zebrafish (Brachydanio rerio. family Cyprinidae, 2N=50) + cytoplasm from loach (Paramisgurnus dabryanus. family Cobitidae, 2N=48). Late NCH larval fish did not like zebrafish type or loach type in morphology were obtained (paper unpublished).

V. Inter-order combinations

12. Nucleus from tilapia (Oreochromis nilotica, order Perciformes. 2N=44) + cytoplasm from goldfish (Carassius auraius, order Cypriniformes. 2N= 100). Early NCH embryos with some changes in its protein synthesis pattern were obtained (1990).

13. Nucleus from tilapia (order Perciformes, 2N=44) + cytoplasm from loach (order Cypriniformes, 2N=48). Late NCH larval Bsh did not like tilapia type or loach type in morphology were obtained (199 1).

VI. Inter-class combinations (mammals and fish)

14. Nucleus from white mouse (Mus musculus, class Mammalia. 2N=40) + cytoplasm from loach (class Osteichthyes, 2N=48). Early NCH blastula with mouse chromosome was obtained (paper unpublished).

Conclusion

In summary, the following tentative conclusions could be drawn regarding the results obtained from aforesaid nuclear trmsplantation experiments in fish.

(1) The nuclear transplantation could be performed in fish with less species- specific limit if it was compared with the results obtained in amphibians. So fax-, the NCH adult fish were obtained through the combinations of nuclei and Biology Iuternatioml (Spcial Issue N58- 1993) cytoplasm from fishes of different varieties. genera and subfamilies while the NCH embryos or larval fish were obtained through those combinations of different families, orders, and classes.

(2) Al1 kinds of NCH adult fish showed some modified characteristics at morphological. biochemical, physiological. and immunoelectrophoretic levels. Some of the NCH adult fish had the characteristics identicai to those of nucleus donor type; some were identical to those of cytoplasm receptor type; and some individuais had intermediate characteristics. or even exhibited some apparent novel characteristics not found in both parental species. Ail of those NCH fishes were fertile. Hence, probably they have provided good experimental models for identieing the effects or influences of the genome, cytoplasm or their interactions on the development of NCH eggs and their genetic inheritance in offsprings through expressed characteristics.

(3) Since in al1 aforesaid nuclear transplantation combinations. the NCH eggs can at least be developed up into the blastula stage regardless of how fax- the species afIinity of nucleus donor that that of receptor cytoplasm were related, it seems that the factors which can control the initiation of the cleavage of NCH egg were of non-species specific.

(4) The more distantly related is the fish species used in combination, the more serious developmental incompatibiiities are revealed in pst-NCH blastula development. However. it was observed that the same or similar number of chromosomes the species has used in combination of nuclear transplantation may reduce the incompatibilities of the nucleus and the cytoplasm in the NCH egg and naturaiiy can promote NCH eggs developing up into adult fish, such as the NCH adult fish obtained from the inter-genus combination of common carp nucleus (2N=100) and crucian carp cytoplasm (2N=100); and the inter- subfamily combination of grass carp nucleus (2N=48) and blunt-snout bream cytoplasm (2N=48).

(5) Based upon the above conclusions, it seems possible to obtain superior clones of fishes of agricultural importance from nucleo-cytoplasmic hybrids of distantly related species with distant affinities. However. since the inheritable modifications obtained in NCH fish are at random, hence to understand the mechanisms on how the genome and the cytoplasm were interacted and how those interactions could lx controlled for screening ideal NCH fish wiil be a great challenge for achieving this practical goai.

References

BROWDER LW. 1980. Developmental Biology, Saunders College. Phildelphia. U.S.A. BRIGGS. R & KING. T.J. 1952. Transplantation of Living Nuclei from Blastula Cells into Enucleated hg's 5.Proc. NdAcad. Sct U.S.A. 38:455-463. COMMANDON & DE FONBRUNE, F. 1939. Greffe nucleaire total. simple ou multiple, chez une amibe. CR SOC. BbL Paris. 130:744-748. DANIELLI, J.F. 1958. Cellular Inheritance as Studied by Nuclear Transfer in Amoebae. IN: New Approach in Cell Bblogy. Academic Ptess, London. pp. 15-22. KAWAMURA. T. & NISHIOKA, M.J. 1963. Reciprocal Diploid Nucleo-cytoplasmic Hybrids Between ïWo Biology International (Special Issue N?8- 1993)

Spedes of Japanese Pond Frogs and Their Offspring. J. Sct fjiroshima Un&. Ser. B., Diu. 1 (BOL] 2 1:65- 84. KAWAMURA. T. & NISHIOKA. MJ. 1963. Nucleocytoplasmic Hybrid Frog Between Two Species of Japanese Brown Fmgs and Their Oflspring. J. Sct Hiroshima Unlv. Ser. B., Diu. 1 [Zod). 21: 107- 134. TUNG, T.C. & YAN, SHAOYi. 1985. Nuclear Transplantation in Teleosts. V. Hybrid FLh fmm the Nucleus of Crucian and the Cytoplasm of Goldllsh. Proceedings of the Chlna-Japan Joint Meeting on CeIl Engineerin$. 20-22 May, 1985,Shanghai. Ckpp. 42-46. TUNC,T.C. et aL 1980. Nuclear Transplantation in Teleosts. 1. Hybrid Flsh from the Nucleus of Carp and the Cytoplasm df Crucian. Scfentia Sinica 23(4):517-523. TUNG. T.C. et al. 1973. Transplantation of Nuclei between Two Subfamüies of .Teleosts (Goldfish- Domesticateq Carassius aumtus. and Chinese Bittering. Rhodeus slnensfs). Acta Zod. Sh 19(3):201- 212. YAN SHAOYI et al. 1984. Nuclear Transplantation in Teleosts. II. Hybrid Fish fmm the Nucleus of Crucian and the Cytoplasm of Carp. Scienti~Sinica 27(8):729-732. YAN SHAOYI et ai. 1985. Nuclear Transplantation in Teleosts. IVa. Nuclear Transplantation Between Different Subfamilies -- Hybrid Fish from the Nucleus of Grass Carp (Ctenopharyngoden. fdeUus) and the Cytoplasm of Blunt-snout Bream (Megalobrarna amblycepha4. Chinese J. of Bbtech 1(4):15-%S. YAN SHAOYI. 1989. The Nucleo-cytoplasmic Interactions as Rwealed by Nuclear Transplantation in Fish. IN: G.M. Malacinski (Ed.)Prfmers fn Dewlopmental Bblogy. VOL 5. Cytoplasrnic Organlzatlon in Dewlopment McCraw-Hill. New York pp. 61 -81. YAN SHAOYI et ai.. 1990. Developmental Incompatibility between Cell Nucleus and Cytoplasm as Revealed by Nuclear Transplantation Experiments in Teleost of Different Familles and Orders. Int J. M.BBid 34~255-265. YAN SHAOYi et al 199 1. Further Investigation on Nuclear Transplantation in DHerent Orders of Teleost: the Combination of the Nucleus of Tilapia (Oreochrornk nüotica) and the Cytoplasm of Loach (Paramisgurnus dabryanus). Int J. Dev. BbL 35429-435.

*Research Croups in China

1. Institute of Developmental Biology, The Chinese Academy of Sciences. Beijing. 2. Guangxi Fisheries Research Institute. Nanning. 3. Pearl River Fisheries Research Institute, The Chiriese Fisheries Academy, Guangzhou. 4. Chang Jiang Flsheries Research Institute, The Chinese Fisherfes Academy. Shashi. 5. Fresh Water Fisheries Research Institute, The Chinese Fisheries Academy. Wuxian. 6. Zhuangjiang Fmsh Water Flsheries Institute, . Wuzhou. Biology lnler~twnal(Special Issue N0I8 - 1993) Induced Fusion of Oocytes and Ernbryonic Cells

by S.G. ~assetzky', AA. BWS'. G.G. ~ekirina~.M.N. skoblina1 '~nstituteof Developmental Biology, Russian Acaderny of Sciences. 26 Vavilov Sireet, Mo-w 1 17334. Russia. 'lnstitute of Cytology. Russian Acaderny of Sciences. 4 nkhorestsky Avenue. St. Petersburg, Russia

Investigations in which the method of induced cell fusion was used in studying the oocyte maturation. fertilization and early embryogenesis of animals have been reviewed on the basis of published data and the authors' experience. The problems of developing the method of induced cell fusion, its combination with the other methods (microsurgery. ce11 fragmentation, etc.) and the perspectives of its use in embryology are reviewed.

We used this approach when studying the role of interaction between the nuclear cytoplasmic material in oocyte maturation and early embryogenesis of marine invertebrates (Skoblina et ai., 1982) and mammals (Sekirina. 1985).

The oocytes of the starfish Aphelasterias japonica were denuded using 0.25% trypsin treatment. Following subsequent treatment with 1M urea and 50% polyethylene glycol (M, = 6000) they fused to form cell hybrids (up to 5-6%). easily distinguishable by the number of nuclei (germinal vesicles). The majority of fusions took place between pairs of oocytes. Neither marked morphological changes nor discernible mïxing of cytoplasm was observed in the ce11 hybrids during 4-5 h.

Further treatment of the ce11 hybrids with 1-methyladenine (~x~o-~M)resulted in nuclear maturation (germinal vesicle breakdown and extrusion of polar bodies). The matured cell hybrids were capable of cleavage after insemination. During four to five hours of observation they developed to the early morula stage.

The oocytes were divided into anuclear (cytoplasts) and nuclear (germinal vesicles) fragments by centrifugation in a Ficoll gradient with cytochalasin B. Ce11 hybrids between the oocytes and the cytoplasts were then obtained with the aid of polyethylene glycol treatment. These cell hybrids were capable of responding to 1-methyladenine by maturation. They were also capable of fertilization and subsequent early development (Vassetzky et ai., 1983). Later we studied the time course of maturation in oocytes and cell hybrids of A japonica both vitally under the light microscope using phase contrast and Hoechst 33342 (Sigma) and on fixed preparations stained fier Feulgen.

The timing of successive phases of meiotic divisions (diakinesis - Dl-2. metaphase 1 - MI, anaphase 1 - AI, telophase 1 - TI, metaphase II - MII) was determined in 50% of oocytes and ceil hybrids using probit-analysis. The duration of a phase was determined as the difference between the timing of a given and subsequent phase in 50% of cells. The data on timing of nuclear Biology Infer~tional(Spccial Issue N"28 - 1993) changes in maturing oocytes are given in Table 1.

Table 1. Duration of Meiotic Phases in Oocytes (0)and Cell Hybrids (CH) of AphelasterIas japonfca.

Phase Timing of a given phase in 5Wo Duration of phase, min of cells afkr 1-MA treatrnent 1 O CH O CH r exp 1 1 exp2 exp 1 1 exp 2 Dl-2 15 21 22 20 D3 37 36 41 5 2 2 MI 42 38 43 51 (48.6) 58 40 AI 93 96/69* 8317iP 13 (2.4) 21* 15* TI 106 90* 85* 14 (13.3) MI1 120 Total 105 (100)

Tlming and duration of phase were determined using advanced embryos. ( ) Duration of phase in O/o of the total duration of al1 phases of the 1 meiotic division.

The total duration of "active" phases of the 1 meiotic division in Aphehterias japonica at 18" was 105 min. About 50% of this period is occupied by MI, about 25% by Dl-D3 and the rest is divided almost equally between Aï and Ti (Table 1). Comparison of our data with the results of special studies on timing of meiosis in oocytes of Acipenser gueldenstaedti, Misgurnus fossilis. and Xenopus Laeuis (see Vasstez9 & Betina, 1987) has shown that. although the total duration of the 1 meiotic division in these species and A. japonica markedly differs, the relative duraffon of its phases is characterized by similar values.

The literature contains data about chronology of meiotic maturation in the starfish Asterina pectinijera: the time from the beginning of hormonal treatment to AI is 603 min. It can be seen from Table 1. that the duration of this period in A. japonica is 93 min. However, studies with these two species were carried out at different temperatures: 20" and 18". respectively. Therefore, we used for comparison of these data the value of (the duration of one rnitotic cycle during the period of synchronous cleavage divisions - Dettlaff & Dettlaff. 1960). This value equals 34 min. at 20° for Asterina pecttnifera and 58 min. at 18O for A.japonica (Davydov, 1990). When recalculated in b, the duration of the period from the beginning of hormonal treatment AI is 1.8 to for Asterina pectinifera and 1.6 to for A. japonica. i.e., it practically coincides.

A remarkable synchrony was found in phases of the 1 meiotic division between several nuclei (irrespective of their number) in the sarne celi hybrid. In cell hybrids formed from two or three oocytes. the nuclei were at the same meiotic phases in more than 70% of cases (Table 2). Biology Inîernalionai (Special Issue NT8 - 1993)

Table 2. Synchny of Meiotic Phases in Nuclei of Cell Hybrids of AphelasterIas japonica

Number of nuclei Total number Number of cell hybrids in which nuclei were at in CH of CH sarne phase adjacent phases different phases 2 158 121 (77%) 32 5 3 17 12 (71%) 5 4 2 1 1 5 1 1

The number of oocytes involved in the formation of a cell hybrid or changes in the nucleocytoplasmic ratio (as a result of enucleation) did not affect the rate of oocyte maturation. The duration of the fkst phases of the meiotic divisions in cell hybrids and in the initiai population of oocytes was practically identical (Table 1).

In Our eariier vital observations (Vasstezky et al.. 1983). it was found that the number of polar bodies extruded during maturation of ce11 hybrids of A japonica corresponds to the number of nuclei they contained. Histological analysis has shown that during maturation of ceii hybrids their nuclei do not fuse but proceed through al1 phases of the 1 meiotic division independently. A similar picture was observed in two cell hybrids after fusion of two immature (stage GV) pig oocytes: even after 26-30 h of incubation two separate nuclei were found, usuaiiy at stâge D3-MI (Fulka, 1983). In vital observations of ce11 hybrids obtained after i~~siolaof two oocytes of the starfish Murthasterias glacialls it was found that maturation was always completed by the extrusion of two II polar bodies and formation of two independent pronuclei (Guerrier & Neant, 1986). These data confirm that no fusion of nuclear material takes place during maturation of ceil hybrids.

It is noteworthy that the topographq- of the nuclear material in maturing ce11 hybrids of A. japonica markedly varied. In some cases, chromosome plates at MI were located close to each other, including the formation of one common group. Apparently. this was due to initial CO-localizationof GVs in ceii hybrids.

However, in most ce11 hybrids formed after fusion of maturing and immature (GV stage) mouse bocytes, all chromosomes formed, as a rule, a common group after 3 h of incubation (Fulka et al., 1988). A similar picture was observed in matilring cell hybrids formed after fusion of two immature mouse oocytes: already within 5-6 h of cultivation chromosomes formed a common group if 40 typical bivalents, one 1 polar body was then extruded and one metaphase plate at MI1 formed (Tarkowski & Balakier, 1980). Apparently. such union in maturing cell hybrids is due to structural and functional features of the mouse oocytes.

Fusion of the nuclear material in cell hybrids of A. japonica appears to take place already during cleavage. Further studies are needed to clari@ this Biology lnter~liond(Special Issue N028 - 1993) process and the pattern of distribution of the nuclear material at the early cleavage stages.

References

DAVYDOV, P.V. 1990. Reproduction and Development of Some Starflsh Species in Laboratory Conditions (in Russian). Author's absîract of candidate dîssertation. Institute of Developmental Biology, Moscow. DETITLAFF, T.A. & DEïTLAFF. A.A. 1960. On Dimensionless Characteristics of Development Duration in Embryology. LbU Akad Nuuk SSSR 134.199-202. FULKA. J.. Jr. 1983. Nuclear Maturation in F'ig and Rabbit Oocytes after Interspeciflc Fusion Ewp. Ceü Res. 1443212-218. FULKA. J., Jr.. FLECHON. J-E.. MOTLIK. J.. et ai. 1988. Does Autocatalytic Amplification of Maturation- Promoting Factor (MPF) Exist in Mammalian Oocytes? Gamete Res. 21: 185-192. GUERRIER P. & NEANT, 1. 1986. Metabolic Cooperation Following Fusion of Starfish Ootid and Primary Oocyte Restores Meiotic-Phase-Promoting Activity. hoc. NdAcad. SA,U.S.A. 834814-4818. GULYAS. BJ. 1986. Oocyte Fusion Dewlop. Bfol. 457-79. SEKIRINA. G.G. 1985. In vitro Studies of Products of Fusion and Ce11 Hybrids Between Mouse Oocytes. Zygotes and Blastorneres. Ontogenez. 16583-588. SEKIRINA, C.G.. SKOBLINA, M.N., VASSETZKY. S.G.. & BILINKIS. A.A. 1983. The Fusion of Oocytes of the Starfish Aphelasterfas japonlca. 1. Formation of Cell Hybrids. Ce& Dger. 127-71. SHIRAI. H., HOSOYA, N.. SAWADA, T.. et ai. 1990. Dynamics of Mitotic Apparatus Formation and Tubulin Content DurIng Oocyte Maturation in Starfish. Dewlop. Growth and Dger. 32521-529. SKOBUNA, M.N., VASSETZKY, S.G., SEKIRINA, G.G.. & BILINKIS, A.A. 1982. Hybridization of Oocytes of the Starfish Aphelasterias japonica Zh Obshch. Bid 43847-853. TARKOWSKI. A. & BALAKIER H. 1980. Nucleo-cytoplasmic Interaction in Ce11 Hybrids Between Mouse Oocytes. Blastomeres, and Somatic Cells. J. Embryd. Exp. Morphd 55319-330- VASSETZKY, S.G. & BETINA, M.I. 1987. Ktnetics of Meiosts in Oocytes of Xenopus laevls. Ontogenee. 18:36- 41. VASSETZKY, S.G. & SEKIRINA, G.G. 1984. Fusion of Germ and Embryonic Cells and Perspectives of the Method of Somatic Hybridfiation in Embryology. Ontogenez 15453-464. VASSETZKY, S.G.. SEKIRINA, G.G.. SKOBLINA, M.N.. & BILINKIS, A.A. 1983. The Fusion of Oocytes of the Starftsh Aphelasterias juponica II. The Capacity of Ce11 Hybrids for Maturation and Cleavage. Cell Dtper. 1173-76. VASSETZKY, S.G., SKOBLINA, M.N., & SEKIRINA, G.C. 1986. Induced Fusion of Echinoderm Oocytes and 5.Methods in Cell Biology. 27:36&378. Biology Inier~tional (Special Issue NT8 - 1993) Direct Production of the Super Male (YY) by Androgenesis in Arnago Salmon

by H. Onozato National Research Institute of Aquaculture, Nansei. Mie 516-01. Japan

Monosex culture in fish farms is useful in some species. Androgenesis may be a very efficient tool for all male seed production.

Eggs of amago salmon were irradiated with gamma rays and androgenesis was induced by insemination with normal sperm. First cleavage was suppressed by hydrostatic pressure (650 kg/cm2 for 6 minutes). The most efficient timing of the pressure treatrnent was 7.5 hours after insemination at 10°C.

Androgenetic male flsh matured in the first year (Fig. 1) and were mated with normal females.

Fig. 1. W super males produceci by androgenesis.

For the most part more than 90% of the fertilized eggs hatched norrnally (Table l), of the males were found in every tested case. 46% were found in the control group (Table 2).

Table 1. Developmental abiiity of eggs ferüiized with sperm of androgenetic males.

-- Androgrnetic Eggs Fertilization Noria1 eyed male, tag no. inseiizated rate eess (%)

Control 6971 631E 7651 6E4E 7356 7C13 305A Table 2. The sex ratio in progeny crosseci with androgenetic males.

Hale parent No. of tag no. progeny exaniined

Control 6971 631E 765 1 6E4E 7356 7C13 305A

It was cohedthat the androgenetic male was the super male 0in arnago salmon. Androgenesis was repeated using spermatozoa of the androgenetic male. Hatching ratio was not improved compared with first generation and a small percentage of eggs treated hatched normaily (Table 3).

Table 3. Pipduction of the second generation of the super male by repitiiion of androgenests.

Tag no. No. of Normal of super eggs eyed male used embryos X

-- Androgenesis 7651 696 O control 631E 338 O

Androgenesis 7651 5408 12 0.2 + pressure 631E 2144 85 4.0

Feminization was done in XY alevins using B-estradioL. Fry were immersed in water containing the dissolved hormone for two hours twice a week. The concentration from 50 to 400 ug/l was emcient for sex reversal. One hundred per cent feminization was produced with a treatment of 200ug/l to 60 days after hatching (Table 4). Concent- ration Hatch Feeding % Female fi 911 O 10 20 30 40 50 60 days O 50 200 Table 4. Feminization of 400 XY progeny by bestradiol. 200 200 Treatment initiation could occur from O to 30 days after hatching. They matured in the second year and eggs of good quality were obtained from 48 fish. Eggs were fertilized with normal or UV-irradiated sperm. The second maturation division (GI) or first cleavage (GII) of the gynogenetically activated eggs were suppressed by hydrostatic pressure. More than 90% of intact eggs, up to 62% of GI eggs and less than 5% of GII eggs were eyed normally vable 5) and are now under incubation.

Table 5. Cyogenests of eggs obtained fmrn sex reverseci XY fernales.

Eggs Eyed eggs used Normal X Deformed

Intact control 431 409 94.9 O Gynogenesis control 703 28 3.9 383 Gynogenesis 1 1259 781 62.0 387 Gynogenesis II 9019 361 4.0 1081

In the intact control. XX, XY, and W will separate in the ratio 1:2: 1. In GI, the number of XI fish depends on the recombination rate between both sex deterrnining genes. In GII. XY fish does not appear and the ratio XX and XY will be 1: 1. If the recombination does not occur, W super male may be produced easily by this technique. Relative distance from the centromere to the sex determining gene may be determined if recombination occurs. Biology Internafionai (Special Issue NS8 - 1993) Studies on Chromosome Manipulation in Cyprinid Loach (Misgumus anguillicaudatus), Common Carp (Cyprinus carpio), and Small Abalone (Haliotis diversicolof)

by N.H. Chao, C.P. Yang, H.P. Tsai, W.H. Liang. I.C. Liao Taiwan Fishefies Research InsUtute. Keelung, Taiwan 202. R0.C.

introduction With the aim of establishing chromosome technologies in practical work of fish farming, ploidy manipulation has been studied in various flnfish and shellfish in Taiwan. Both cyprinid loach (Misgrnus angufflicaudatus) and common carp (Cyprinus carpio) are very important in freshwater fish farming and have been the target species for triploidy manipulation. Inbred lines are traditionally produced by repeated full-sib mating. However, cyprinidae have relatively long generation intervals. For this reason. full-sib mating is not the method of choice. Instead, gynogenesis has been adopted as a quick method to produce al1 female inbred strains.

Small abalone (Haliotis diversicolor) is a unique shellfish with about 10% of its total weight being non-attractive colored gonad during its spawning season (Chao et al.. unpublished). Feasibility of ploidy manipulation to reduce the gonad somatic index for a larger edible part and better flavor needs to be verified. Triploid production has been reported to be a favorable method of obtaining sterile strains and hence, a means of increasing adult size and avoiding loss of glycogen in spent gametes.

This paper summarizes optimal conditions to induce gynogenetic diploids and triploids and practical methods to determine the presence of triploid in these species that were developed since 1985 and 1990 respectively, for finfish and shellfish in Our laboratory.

Materials and Methods

A Strategies in Retaining Polar Body or Inhmiting Cleavage

To induce triploidy in a variety of aquatic organisms. thermal shock, hydrostatic pressure shock and chemical shock have been applied to suppress polar body formation or to block mitosis. The efficiency of these shocks depends on three main parameters: (1) shock conditions (cold or heat temperature in thermal shock, pressure intensity in hydrostatic pressure shock. kind and concentration of chemical solution in chemical shock); (2) the duration of shock; and (3) its timing. Selection and determination of optimal starting time were carried out due to a very rapid chromosomal and cytoplasmic changes during early embryogenesis. Biology Inîer~twnal(Specinl Issue NT8 - 1993)

Female spawners were selected from broodstock under captivity and injected with pituitary gland or HCG. Ripened eggs were stripped out from gravid females. The testes were removed from one mature male and ground with Ringer solution for freshwater fish. Sperm was activated and mixed with eggs irnmediately. Loach eggs were exposed to 8°C for 40 min; the starting time of cold shock treatment varied from 1 to 9 min after fertiiization to find the best timing. The eggs were exposed to various temperatures of 1, 4, 6. 8, and 12°C for nine duration times of 5, 10. 20, 30. 40, 50, 60, 70, and 80 min to determine the optimal combination of cold shock conditions (Chao et al., 1986).

Breeding of golden loàch was attempted by means of induced gynogenesis. The female broodstock, golden in phenotype, were stripped for eggs after induced maturation. Milt from common loach was genetically inactivated by UV irradiation at a dosage of 1-W/cm2. or 0.05W/cm2 for 20-40 seconds. The eggs fertilized with the inactivated sperm were subjected to cold shock of 6°C for 20-50 min or 8OC for 40-60 min and pressure shock ranging from 500 to 12,000 psi for 2-5 min, beginning at 5 min after fertilization (Chao & Liao. 1990).

Male spawners with running milt and female spawners under captivity with soft belly were selected from broodstock during their propagation season, from February to April. They were kept in a small pond maintained with running water sourced from the underground for egg maturation and spawning. Artificial propagation was done. Fertilized eggs were sprayed on net or put in a pressure cylinder. The experimental treatments for inducing triploidy included cold shock of 1 and 3°C and pressure shock of 650 kg/cm2 at 1, 3. and 5 min after fertiiization.

Most recently. gynogenetic diploid induction mainly by either heat shock of 40°C for 2 min starting at 24-30 min to block mitosis or cold shock of 1°C for 30 min starting at 1-5 min to retain second polar body in fancy carp has been under preliminary field trials by using the lOOx diluted and UV-irradiated sperm during fertilization and referring the treatment parameters obtained in triploid induction (Hollebecq et al., 1986; Taniguchi et ai., 1990; Rothbard. 199 1).

Spawners with visible full whitish testis or brown greenish ovary were selected in late October and November. In separate tanks, they were treated with water heated from the original water temperature up to 4-6°C (however, never above 30°C) at a gradualiy increasing rate of 1°C/h. The heater was then turned off, allowing the temperature to drop normally. During the late phase of 1st and Biology Infcr~iional(Special Issue N928 - 1993) 2nd cycle, some male and some female spent sperrn and eggs. Milt was added into the egg tank at an optimum ratio for dcialfertilization.

For comparison, 21-42 min after fertilization. ideally before the 1st or 2nd polar body was extruded, the fertilized eggs were immersed in a solution of 0.5-lmg/l cytochalasin B dissolved in 0.1% DMSO for 20 min (Allen et al.. 1988; Beaumont & Fairbrother, 1991). The eggs were then rinsed with 0.1% DMSO solution and seawater before being stocked in outdoor ponds equipped with a series of corrugated plastic sheets for spat resting.

B. Assessrnent of Triploidy Presence

Blood smear of at least 20 day old fry or larvae were made on slides and fixed in methanol-formalin 9:l V/V for 15 min. Slides were washed twice with distiiled water, dried and stained by the Feulgen reaction procedure (Humason, 1979) or DAPI (Komen et al., 1988).

The DNA content of erythrocyte nuclei was measured (10 nuclei per fry) with Leitz MPV 3 microscope photometer (Gervai et al.. 1980; Johnstone, 1985). Each nuclei was read 32 times automatically. The readings of stains vary with DNA content in the erythrocyte nuclei and variation between and within each batch of stained slides. The mean value of readings of each sampled fry in experimental and control group were corrected against the corresponding quantitative blank smear on the same slide to exclude the variation between and within each batch of stained slides. Since the control groups produced only diploids, triploidy was acknowledged when the corrected reading on a slide of sarnple fry was about 1.5 times that of the mean reading of the control group. The histograms of fluorescence intensity representing relative DNA content were constructed for comparison. Triploid, diploid, and haploid cells show three distinct peaks.

B-2. Flow Cytometry.

Saturated EDTA solution just enough to coat a thin layer on the inner walls of a lm1 syringe was used to treat red blood ce& to avoid coagulation. PSB-BSA- NaN3 solution was added to bring the mixture up to lml. The mixture was kept at 4°C before analysis. Protocols A, B, and C were compared (see Appendix for the schematic diagrams of the protocols).

In protocol B, as an exarnple, the mixture was centrifuged at.1,100 rpm for 5 min. After the supernatant was discarded, the ceils were treated with 0.3d solution A (0.5d RNase + O. lm1 Triton X-100 + 0.5ml Propidium iodide) and incubated at 37OC for 20 min. It was then treated with 0.3 ml solution B (0.5ml Propidium iodide + O.lm1 Triton X-100+9.4ml 0.4M NaCl) and incubated at 4OC for one hour. The samples were protected from light until the flow cytomett-ic analysis using FPICS Profile II (Coulter Corporation) was processed. The Biology Inler~iwiuil(Special Issue NT8 - 1993) optimal injection flow rate of 15-30 pl/min, sheath pressure of 7.5 1- 15.OOpsi, and laser power 0,15 MV and 8.62 A were suggested to obtain better half peak coefficient of variation and histograms of fluorescence. Mean value of DNA content in triploid cells was about 1.5 times that of normal diploids and thus distinguishable.

B-3.Chromosome Preparation.

Embryo or small fish was first incubated in colchicine to arrest mitosis at metaphase for 4-8h depending on the size. Embryo, gill or other tissue of fish was then treated with hypotonie solution, either 0.005 N potassium chloride, 0.005 N sodium citrite or distilled water. The resulting material was then fixed with Carnoy solution of methanol and glacial acetic acid at 3:l V/V. The tissues were finely chopped using a scalpel. The cell suspension was spread on a warm slide to form cell rings. Proper staining was accomplished by using Giemsa solution. Observation through microscope at 600x or higher was followed by photograph taking and counting the number of metaphase chromosome. If necessary, the chromosomes may be cut individually and the cutouts pairing homologues arranged by size, shape. and length of arms. Each fish with chromosome numbers ranging from 1.2 to 1.5 ümes that of normal diploid fish was considered to be triploid. The percentage of triploid fish was. therefore, counted. Results

Chromosome manipulatioxis have been canied out in three local aquatic species in Taiwan - loach, carp, and smail abalone - using thermal, pressure and chernical shocks. Table 1 sets out the methods which have been used for induction and ploidy assessment, together with the best results obtained. The proportion of diploidy and triploidy types may change during development when mortaiity is not avoidable.

Table 1. Chromosome manipulation in cyprinid loach (Mfsgumrcs angutllloar

Best % of Induction Verification Specles Method Method G2N 3N Suggested Tlming and Duration Cyprinid Coach CS mf 80-100 5 m.a.f., I'C, 30-40 min CS cm 70 3 m.a.f., 6'C, 30 min PS cm 32 5 m.a.f., 6000 psi, 4 min Common Carp CS f c 66-86 1-3 m.a.f., l'C, 30 min PS fc 7.7 5 m.a.f., 650 kg/cm2, 10 min HS - - 1-3 rn.a.f., 40% 2 min Small Abalone CS CP O CB cp 22-50 24-39 m.a.1.. 20 min

3N = Triploid. (';ZN = gynogeneticdiploid. CB = cytoclialasin B. (3= mld shock. HS = hoat shock. i"S = pressure shock. cm = color marker. cp = chroinosnine preparation. fc = flow cytometry. rnf = microflurometry. m.a.f. = minutealter fertiliution.

When eggs of cyprinid loach and common carp were subjected to cold shock (l°C for 30-40 min in loach, and 1°C for 30 min in carp). those shocked, Biology Internaiional (Special Issue NI18 - 1993) respectively, at 5 min and 1-3 min after fertilization produced the highest proportion of triploid fry. In cyprinid loach, triploidies of 80.8. 80.0, 95.0, 100.0, and 100.0 were obtained in the groups treated at 1°C for 60. 20, 50. 40, and 30 min star-ting at 5 min after fertilization. In cornmon carp, triploidies of 66.6, 80.4, 86.2, and 86.6% were obtained in the groups treated at 1°C for 30 min starting at 1-3 min after fertilization.

When eggs of cyprinid loach, golden loach, and common carp were first fertilized with irradiated sperm and then subjected to cold shock, morphologicaily no& fry were confirmed to be diploid. In cyprinid loach, optimum cold shock of 6-8°C for 30-40 min at 3-5 min after fertilization resulted in comparatively high percentages of normal fry. However, the survival rate was low in the diploid gynogenetic group of cyprinid loach. In golden loach, the percentage of golden fry reached 70% but the survival rate was also quite low. In fancy carp. prelirninary results showed that heat shock of 40°C for 2 min to inhibit the first mitotic cleavage gave much higher hatching in most cases than cold shock of 1°C for 30 min to retain the second polar body. The induction of a desirable color pattern was far from being successful despite the adoption of properly irradiated sperms of common carp, loach. or catfish. The experiment is still undergoing to deal with this. Problems such as the high cost of pure strain spawners and the difficulty in categorizing the fhgerling of various color patterns need to be solved with the advice of experienced scientists. Various reasons in failure of inducing the desirable color pattern of fancy carp using gynogenenesis induction will be determined and minimized.

In small abalone, treatment with cytochalasin B at lmg/l for 20 min started at 24, 27, 30, and 39 min after fertilization gave a chromosome counting of triploid of 33.3, 22.2, 50.0, and 35.7%. However, no significant results were obtained with thermal shock trials.

The three methods used in triploidy identification have their own particular characteristics (Table 2). In instruments used, microfluorometry (MF) makes use of the fluorescent microscope; flow cytometry (FC). the flow cytometer; and chromosome preparation (CP), the light microscope. In the target material to be studied, MF and FC aim for the DNA content in the red blood ce11 while CP. the chromosome number in the cellular nuclei of the embryo, gill or cell line. In the possible errors when using the different methods. MF has focusing and condenser errors; FC, poor pre-treatment and inconsistent laser power; and CP, overlapping of and missing chromosome during preparation. In the number of tested cek per individual finfish and shelifish. MF has 30+ 10 cells; FC, >10,000 cells; and CP. 30-100 cells. In pre- treatment time per individual, MF requires 40 min; FC, 3 min; and CP, 3-12 h. Lastly. in either observation or sample mn time per individual, it takes about lh for MF; 2 min for FC; and 1 h for CP. Protocols A, B, and C were equally functional.

The choice of method in triploidy identification depends on the number of fish, effort involved, avaiiability or cost of instruments and the desired level of accuracy.

Table 2. Cornpariaon of three methods of polyploid determination

Mlcrofluorometry Flow Cytometry Chromosome Preparation Tool Used Fluorescence microscope Flow cytometer Common microscope Sample Taken RBC RBC, Tissue + Embryo + CuHured cell + Tissue Target Material DNA content DNA content Chromosome number

Posslble Error -, Staining difficulties Poor treatment Overlapping of and missing Inconsistent laser powerlmercury chromosomes lamp intensity Sampllng Number 30 110 cellslslide >10,000 cellslvial Tlme for PrefreatmenVIndlvldual 40 min 3 min Tlme for Obserwtlon/lndividual 1 h 2 min

Discussion

Triploids can not synapse at meiosis and. therefore, are expected to be sterile. Consequently, there are several advantages. Energy and nutrient for gamete production become avallable for somatic growth in sterile triploid. The flavor of triploid may be improved due to storage of glycogen. Sterile triploids of introduced species may be cultured or stocked in waters which might be sensitive to the accidental introduction of cornpetitor species. On the other hand. gynogenesis is believed to be an efficient method for the rapid production of highly homozygous inbred Unes. Chromosome manipulation in finfish and shellfish worth further study and progress although the progress of developing the repeatable and reliable techniques takes much time.

Cold shock was found powerful in inducing triploid of flnfish while cytochaiasin B has proved to be a reliable chernical agent for inducing chromosome doubling during the maturation division in shellfish. Heat shock to block mitotic division in inducing gynogenesis, with a better chance of obtaining high hatching rate and survival rate, in carp was noted. Both Hollebecq et ai. (1986) and Rothbard (1991) also successfully applied heat shock to induce diploid gynogenesis in carp. Further studies should be taken to determine why there are such differences in favorable methods for finfish and shellfish; for meiosis and mitosis manipulation. However, in fancy carp, how hereditary characters. such as color patterns and spot location are inherited from generation to generation has not been well studied. Inbreeding of fancy carp from the population we obtained the spawners for induction test of gynogenetic diploid using mature male and female of sarne color pattern gave non-predictable color pattern at a random composition in their offspring. Before chromosome manipulation techniques can be applied to produce a gooci number of fancy carp with desired color patter, there is much to be studied.

Ali the currently recommended artificial treatment caused some percentages of abnormaiities and increased mortality during early development. Scientlfic Biology Internaiionai (Special Issue N58 - 1993) approach to deal with the concemed mechanism and the possible alternative method with less damage is doubtlessly needed.

Commercial triploidy or gynogenetic diploidy is not yet available for either one of these three experimental species. Development of chromosome manipulation in both finfish and shellfish should continue to expand in the future to make marketing possible to explore thei.valuable feature.

After using both microfluorometry and chromosome counting for assessing ploidy, the disadvantages of either being rather intricate and tirne consuming or having much routine and tedious staining and microscopy observation job were noted. They may be useful to determine ploidy in a limited number of individuals. Nevertheless. when assessment of ploidy percentage of several populations of finfish and shellfish are undertaken within several days using traditional methods, the majority of cells examined on a slide did not often provide really countable chromosome, particularly in the fish with chromosomes more than 50.

The identification of ploidy using various pre-treatment protocols to stain DNA of suspensible cells with specific fluorescent dye and flow cytometer to determine DNA content has the advantages of being accurate to read directly the target material - DNA content of each single ce11 fi-om same individual. The capability of checking more than 10,000 cells within 3 min and providing the printed data to give convincing evidence make it an effective and recommendable instrument to meet the modem biotechnology era. Besides, a much less expensive flow cytometer (e.g., PmEC CA II) an alternative instrument equipped with mercury lamp instead of a laser one and with lirnited specific functions including DNA content measurement than the multi- functional one used for medical research is recently commercially available and is adequate for the purpose of verification of ploidy.

Acknowledgements

We are indebted to our coiieagues in the Biotechnology Research Laboratory. Taiwan Fïsheries Research Institute for research assistance. The use of flow cytometry was through the ldnd offer and advice of Ms. L.W. Wu of the Veterans' Ceneml Hospital and Mr. Y.H. Chiang of the Mithra Biotechnology Co., Ltd., both in Tatpei. Taiwan. This paper is a summary of works supported by grants from the National Science Council and Coundl of Agriculture of the Republic of China under the projects NSC-78-0409-B056-04, NSC-79-0409- B056-02. NSC-80-0409-B056-03, as well as COA-77-7.1 -Food-1 lO(G4). 78-7.1-Fïsh-06(5). and 79-7.1-Fïsh- W5). References

ALLEN. S.K., Jr., DOWNING, S.L.. & CHEW. K.K. 1988. Hatchery Manual jor Produchg 'Mpbld Oysters. Washington Sea Grant Pubitcation, University of Washington, Seattle, Washington, U.S.A. 27pp. BEAUMONT.A.R & FAIRBROTHER. J.E. 1991. Ploidy Manipulation in Moliuscan Shelifish: A Re*. J of SheUfish Research . 10(1):1-18. CHAO. N.H.. CHEN. S.J. & LIAO, I.C. 1986. Triploidy Induced by Cold Shock in Cyprinid Loach, Misgyrnus angufilfcaudatus. IN: J.L. Maclean. L.B. Dhn. L.V. Hosillos (Eds.) ïhe First Askm FLsherfes P. kun Asian Fïsheries Society. Manila. Philippines. pp. 101-104. CHAO. N.H.. LIANG. W.H.. TSAI, H.P.. & LIAO, I.C. Triploidy Induced by Cytochalasin B Shock in Srnaii Abalone. Halbtis diuersicobr. (unpubitshed). Biology Inter~iional (Special Issut N78 - 1993)

CHAO, N.H. & LIAO. I.C. 1990. Production of Golden hach, Misgumus anguiilicaudaîus by Means of Gynogenesis. IN: R Hirano & 1. Hanyu (Eds.). The Second Asian Flsherfes Forum Asian Flsheries Society. Manila. Phiiippines. pp. 535-538. GEFWAI, J.. MARVW, T., KRASZNAI. Z.. NAGY, A. & CSAYIW, V. 1980. Occurrence of Aneuploidy in Radiation Gynogenesis of Carp. Cyprinus carpio. l J. Flsh Bid 16:435439. HOLLEBECQ, M.G., CHOURROUT, D., WOHLFm, G.. & BILIARD. R 1986. Diploid Gynogenesis Induced by Heat Shocks After Activation with W-trradiated Spenn in Common Carp. Aquaculture. 5(4):69-76. HUMASON, G.L. 1979. AnLnal Tissue Technique. 4th Edition. W.H. Freernan and Co., San Frandsco. pp. 661. JOHNSTONE. R 1985. Inductlon of Mploidy in Atlantic Salmon by Heat Shock. Aquaculture. 49: 133-139. KOMEN. A.. UCHIMURA. Y.. IEYAMA, H., & WADA. K.T. 1988. Detection of Induced Triploid Scallop. Chlamys nobflfs.by DNA Microfluorometry with DAPI Staining. Aquaculture. 69:201-2û9. ROTHBARD. S. 1991. Induction..of Endornitotic Gynogenesis in the Nishiki-Coi. Japanese Ornamental Carp. Bamldgeh 434): 145-155. TANIGUCHI. N.. YAMASAKI. 1.. & SATO. M. 1990. Production of the Gynogenesis Mploids Produceci by the Cleavage Inhibition T@e in Nishikigoi and the Necessity of the Demonstration. Abstract of the Spring Conference of the Japan Flsheries Association. pp.54. Biology Internaiional (Spccial Issu NS8 - 1993)

PROTOCOL A PROTOCOL B

1 0.2 ml RBC samole in He~arinor saturated EDTA. add PBS (DH7.41 ( 1 Centriluge RBC sample. 1,100 rpm 5 mn ] .1 1 Centrluge 1.500 tpm, 10 min. dlscard sipernatanl (wash) ( Dardsupernatant \1 JI Treat wlh 0.1% Triton-X.100 0.15 M NaCI(4'C). 5 nin 1 Add solution A 30 FI, incubate, 37% 20 mn a1 Trfion-X-100 solvlion : RBC = 1 : 1 C Add soluiion B 30 pl. treat. 4'C, 1 h Wash wth PBS, 2x 1

SOLUTION A Add 100 pl 1 mgml RNase (DNA free) + 100 pl 400 @ml PropkJium iodlde 0.5 rd Propidium iodide stock solution (Keep in the dark) 0.5 rrl RNase stock solution 0.1 rd Triton-X-100 stock solution

1 Incubate, 37'C, 30 min 1 SOLUTION B 0.5 rri Prop~imiodide slock solution 0.1 ri Trdon-X-100 stocksolution 9.4 ni 0.4 M NaCl solution, ( Wash with PBS, 2x 1 adjusl to pH = 7.2

PROTOCOL C

0.2 ml RBC sample in saturated EDTA, add 0.8 ml PBS-BSA-NaN, solution [(1% BSA + 0.05% NaN.,) in PBS]

L Resuspend cells, filter through 40160 pm ne( 3/ 1 Wash with PBS, 2x. spin 1,500 rpm, 30 sec1 \1 1 Sel cell density = 8 x 18 - 1 x 18 cellslml 1 \1 1 Add 20 pl of 5% Triton Io 0.5 ml cells in PBS] \1 Add 20 pl of Propidium iodide (2.5 mglml) Biology International (Special Issue N58 - 1993)

The Development of Technologies for The Control and Deterrnination of Sex in Aquacultured Salmonids

by Edward M. Donaldson, Robert H. Devlin, Francesc Piferrer*, Igor 1. Solar CODE for Biotechnology & Cenetics in Aquaculture. West Vancouver Laboratory, Biological Sciences Branch, Fisheries & Oceans Canada. West Vancouver. B.C. VN 1N6. Canada *Current addmss - Dept Reproducttve Medicine, Univ. of CaMomia. San Diego. 9500 Gilman Drive. La Jolla. CA 920934947,USA

Introduction

The commercial aquaculture of salmonids in marine net pens has increased rapidly over the last five years on both the Atlantic and Pacific coasts of Canada. The culture of Atlantic salmon (Salmo salai) predominates on the East coast while on the West coast. chinook salmon (Oncorhynchus tshwytscha) is the main species followed by Atlantic salmon. Owing to the depletion of natural stocks, aquaculture now accounts for Canada's total commercial production of Atlantic salmon and the production of chinook salmon through aquaculture now exceeds the total commercial and recreational catch of wild and hatchery produced chinook on Canada's Pacific coast. The recent rapid growth in chinook salmon production in British Columbia is closely linked to the development and implementation of sex control biotechnologies for this species (Solar et al.. 1987; 1989). Female chinook salmon mature at 3. 4, or 5 years of age while males mature on average one year earlier at 2, 3. or 4 years. The culture of monosex female populations thus provides the aquaculturist with aogreater window within which to grow and market this species prior to the onset of the development of secondary sexual characteristics associated with sexual maturation. As the culture of mixed sex populations is not economic, virtually al1 of the approximately 15.5 thousand metric tonnes of chinook salmon grown in British Columbia in 1991 were monosex females based on monosex female sperm which was generated over a period of two generations and transferred to industry from this laboratory (Solar & Donaldson, 1991). In Atlantic salmon culture, a variable proportion of the fish mature as grilse one year earlier than full sized salmon. As grilse tend to be predominantly male, the production of monosex female Atlantic salmon would reduce the proportion of grilse (Johnstone, 1989). There is also increasing interest in the development and implementation of sterilization technologies in salmonid culture (Donaldson, 1986; Pepper, 1991). Sterile salmon maintain a high market quality at all seasons of the year and as they never become sexually mature. they have a longer life span and can grow larger than normal fish. Furthermore. there is increasing concern regarding the potential for genetic interaction between escaped farm fish and wild fish (Donaldson et ai.. 1992). This would particularly apply where the farmed fish are genetically altered through processes such as selection or transgenesis (Devlin & Donaldson. 1992). Biology IntermtionaI (Special Issue N58 - 1993) Direct Regulation of Sex Wmtiation

In Pacific salmon, the labile period during which administration of estrogen or androgen will result in the differentiation of a normal ovary or testis respectively, independent of genotype. extends from shortly before to shortly after hatching (Piferrer & Donaldson, 1989; 1993). We have investigated the influence and relative potency of a variety of natural and synthetic estrogens, aromatizable and non-aromatizable androgens and also aromatase inhibitors on sexual phenotype (Piferrer & Donaldson, 1991; 1992). The synthetic estrogen ethynylestradiol-17a is more potent than estradiol-17P for the induction of ovarian development. However. both are capable of inducing 100% feminization after a single immersion of up to 8 hours (P:!ferrer & Donaldson, 1992). Using monosex female chinook embryos as a mode1 system. we have been able to investigate the influence of natural, synthetic. aromatizable and non-aromatizable androgens and aromatase inhibitors on sex differentiation. The administration of the aromatizable androgen. 17a-methyltestosterone, at high dosages or for extended periods of time results in paradoxical feminization, while the administration of the non-aromatizable androgens. 11- ketotestosterone or 17a-methyldihydrotestosterone. results in the production of phenotypic male populations in a dose-response fashion up to 100%. Administration of a single dose of a potent non-steroidal aromatase inhibitor to monosex female chinook embryos results in the production of approximately 20% phenotypic males, and furthemore, this aromatase inhibitor also enhanced fourfold the masculinization response to the aromatizable androgen 17a-methyltestosterone(Piferrer et al.. unpublished). This data emphasizes the key role of embryonic aromatase enzyme activity in the normal sexual differentiation of female Pacific salmon. Studies have also been conducted on the direct inhibition of gonadal development by discrete or continuous immersion of embryos in a solution of exogenous androgen. However, we have not yet produced 100% sterile salrnon stocks in this manner. Indirect R-tion of Sex

The production of monosex female sperm for the generation of dl-female stocks involves masculinization of genotypic females (Hunter et al., 1982, 1983). This was initially a two generation process which involved progeny testing. However, once monosex female embryos have been obtained, a portion of them can be masculinized to produce additional monosex female sperm. We have recently investigated two methods for the generation of monosex female sperm from other salmonid stocks or species in a single generation. First, we have induced meiotic gynogenesis by fertilization of ova with W-irradiated sperm, followed by pressure shock to restore diploidy, and then masculinized the resultant gynogens during the alevin stage (Solar et al.. 1992, unpublished). Masculinized gynogens have been reared to maturity and sperm has been stripped to fertilize normal ova. The progeny from these crosses are expected to be monosex females. Biology Intwmtionul (Spcciol Issu N58-1993) Second, to facilitate the production and maintenance of monosex stocks, a DNA fragment from the Y chromosome of chinook salmon was isolated using subtractive hybridization methodologies to enrich for male-specific sequences (Fig. 1) (Devlin et al., 1991). This DNA probe can discriminate between genetic males (XY) and genetic females (XX) by detecting Y-chromosome- specific DNA sequences on Southern blots or by rapid polymerase chain reaction (PCN assays. Application of this probe aliows new monosex strains to be obtained in a single generation by dowing identification of genetic females (XX)within androgen-treated, masculinized groups of juveniles or adult salmon of mixed genetic sex (Fig. 2). It also permfts the confirmation of genetic sex in sex reversed broodstock prior to gamete collection.

Y-SPECIFIC SEQUENCE Y Randomly Shear Digest Male DNA wlth Female DNA Mbo I Restriction Enzyme

7 \ '/ / -/'' Uk \ \ 250 ug 1.0 ug

Extensively HyhrldlzeY (PERT)

CLONE INTO PLASMID

Fig. 1. Isolation of Y-specific DNA fragments from chinook salmon. A large amount of randomly sheared fernale DNA is extensively annealecl with a small amount of restriction-enzyme digested male DNA. nie resulting fragments with restriction-site sticky ends are inserted into a plasmid for cloning and subaequent analysis. ALEVINS

ANDRociEN IMMERSION +1 PHENOTYPIC MALES

Y SPEClFlC DNA PROBE

GENOTYPlC GENOTYPlC FEMALES MALES PHENOTYPIC MALES 1 DlSCARD NORMAL OVA X\ SPERM

100% FEMALE OFFSPRING

Fig. 2. Production of d-female sahnby mascullnlzatbn and DNA sex determinhg probe. In order to generate sterile salmon stocks we are utilizing the monosex female technology in combination with the induction of triploidy by temperature or pressure shock shortiy after fertilization to produce monosex femaie triploids (Fig. 3). Triploid saimon do not exhibit the rapid somatic growth which is characteristic of normal male or femaie saimon during the early stages of sexuai maturation. In order to optimize the performance characteristics of monosex female triploid salmon we have succeeded in accelerating growth in these fish by administration of a recombinant somatotropin (McLean et ai., 199 1).

Fig. 3. Production system for the generation of triploid monosex female salmonids frnodfaed fmm Donaldson & Benfey, 1987).

84 Biology Inter~twnai(Special Issue N58 - 1993)

Biotechnologies for the regulation and identification of sex have become an integral component of salmonid culture. After the treatment. parameters have been adjusted to match species-specific characteristics, these technologies offer promise for transfer to many other aquacultured species wkone sex- has better production characteristics or a higher market value, or where reproductive containment is required.

References

DEVLJN. RH.. MCNEIL, KB., GROVES. T.D.D.. & DONALDSON. E.M. 1991. Isolation of a Y- Chromosome DNA Probe Capable of Determining Genetic Sex in Chinook Salmon (Oncorhynchus tshawytscha). Can J. Flsh Aquat. Sct 48:1606-1612. DEVLIN. R.H. & DONALDSON. E.M. 1992. Containment of Genetically Altered Fish wlth Emphasis on Salrnonids. IN: C.L. Hew & G.L. Fletcher (Eds.). 7Ymgentc Flsh World Scientific . Singapore. pp. 229- 265. DONALDSON. E.M. 1986. The Integrated Development and Application of Controlled Reproduction Techniques in Pacific Salmonid Aquaculture. Flsh Physld. BIochem 29-24 DONALDSON, E.M. & BENFEY. T.J. 1987. Current Status of Induced Sex Manipulation. IN: D.R. Idler. L.W. Crhn, & J.M. Walsh (Eds.) Proc. ni6-d Int'l Symp. Repnxiuctfue Physid Flsh , St John's, Newjoundhand, 2- 7Aidg., 1987. pp. 108-119. DONALDSON. E.M.. PIFERRER F.. SOIAR, 1.1.. & DEVLIN. RH. 1991. Studies on Hormonal Sterilization and Monosex Female Technologles for Salmonids at the West Vancouver Laboratory. IN: V.A. Pepper (Ed.) Proc. of the Atlantic Canada Workshop on Methods for the Production of Non-maturing Salmonids. Dartmouth. Nova Scotla. 12-21 Feb., 1991. Can. Teck Rep. Flsh Aquat. Sct 178937-45. DONALDSON, E.M.. DEVLIN, RH., SOLAR. I.L. & PIFERRER. F. 1993. The Reproductive Containment of Cenetically Alted Salrnonids. IN: J.G. Cloud m.) Genetlc Consenxztbn of Salrnonfd Flshes. NATO AS1 Series Vol. Plenum Publishing Corp.. New York pp. 113-129. (in press). HUNTER, C.A.. DONALDSON, E.M., GOETZ. F.W., & EDGELL, P.R 1982. Production of AU Fernale and Sterile Groups of Coho Salmon (Oncorhynchus kkutch) and Experimental Evidence for Male Heterogamety. 7km.s. Am FWi. Soc. 1 1 1 567-372. HUNTER. q.k. DONALDSON. E.M., STOSS,J., & BAKER 1. 1983. Production of Monosex Female Croups of Chinook Salmon (Oncorhynchus tshawytscha) by the Fertilization of Normal Ova with Sperm from Sex Reverseci Females. Aquaculture. 33355-364. JOHNSTONE, R 1989. Maturity Control in Atlantic Salmon. A Review of the Current Status of Research in Scotland. IN: M. Carrillo. S. Zanuy. & J. Planas (Compilera). XIth InYl Symp. Cornp. Endocrino&. k.of the Satellite Symp. on Applications ojCornp. Endocrind to Flsh Culture, Ahunéc~~,Spain. 22-23 May, 1989. pp. 89-94. . MCLEAN. E.. SADAR. M.D.. DEVLIN, RH.. SOUZA. LM., & DONALDSON, E.M. 1991. Promotion of Growth in Diploid and Triploid Coho Salmon with Parenteral Delivery of a Recombinant Porcine Somatotropin. Aquat LLuing Resow. 4155-160. PEPPER. V.A. 199 1. Proceedings of the Atlantic Canada Workshop on Methods for the Production of Non- maturing Salmonids. Can Tech Rep. Fish Aquat. Sct 1789152pp. PIFERRER. F. & DONALDSON. E.M. 1989. Gonadal Dtfferentiation in Coho Salmon [Oncorhynchus kkutchJ. after a Single Treatment wlth Androgen or Estrogen at Different Stages during Ontogenesis. Aquaculture. 77:251-262. PIFERRER, F. & DONALDSON. E.M. 1991. Dosage-dependent Dinèrences in the Effect of Aromatizable and Non-aromatizable Androgens on the Resulttng Phenotype of Coho Salmon (Oncorhynchus kkutch). FLsh Physid Biochern 9145-150. PIFERRER F. & DONALDSON. E.M. 1992. The Comparative Effectiveness of the Natural and a Synthetic Estrogen for the Direct Ferninization of Chinook Salmon [Oncorhynchus tshawytscha). Aquaculture. 106:183-193. PIFERFER, F. & DONALDSON, E.M. 1993. Hormonal Sex Conhl in Pacific Salmon: the Importance of Treatment Timing. IN: J.F. Muir & RJ. Roberts (Eds.) Recent Advances in Aquaculture. Vol. 4. Blackwell Scientific Publications, London. (in press). SOUR, 1.1.. BAKER. IJ., & DONALDSON, E.M. 1987. Experimental Use of "Female Sperm" in the Production of Monosex Female Stocks of Chinook Salmon (Oncorhynchus tshawytscha) at Commercial Fish F-. Can Tech Rep. Flsh Aquat Sct 1552:14pp.

Biology Inter~twnal (Special Issue N028 - 1993) Transgenic Salmon with Enhanced Growth and Freeze Resistance

by CL~ew', S.J. Du1, 2. ~on~l,P.L. ~avies~, S.Y. ~authiep.MA. shears3. M.J. Kin$, G.L. let cher^ '~esearchInsutute, Hospital for Sick Children, Toronto, Canada. '~epartmentof Biochemistry. Queen's University, Kingston. Canada. '~emorial University of Newfoundland, St John's. Canada

Atlantic salmon cannot tolerate icy sea water due to their lack of proteineous antikeezes. The freezing threat and slow growth rates at low temperatures are the two most important obstacles for salmon farming in Atlantic Canada. The present communication describes our progress in dweloping new strains of salmon using gene transfer technology. Both the antifreeze protein and growth hormone genes were individually incorporated into the salmon genome. The incorporation, inheritance and expression of these transgenes has been demonstrated. A faster growing, GH-transgenic salmon has been produced using an "al1 fish gene cassette that will be of widespread value in aquaculture. introduction

Aquaculture, a young and fast growing industry is well suited to face the new challenges. associated with the increased demand for food production due to increased human population and depleting fish stocks. Although traditional methods such as selective breeding have been successful in raising the production and profits of the industry. there is a need to adopt new biotechnologies to further stimulate the growth of this industry. The development of fish species which are faster growing, more resistant to a variety of diseases and the acquisition of new and desirable characteristics by gene transfer technology will surely have a major impact and be very beneficial to aquaculture (Fletcher & Davies, 1991; Hew & Gong. 1992).

In Atlantic Canada. as in many other parts of the world, cold weather presents many problems to fish fanning. In addition to slow growth rates because of the low water temperatures, the presence of ice and the threat of freezing have limited the extent of salmon sea farming to only a few restricted and warmer locations. It was rationalized that if freeze-resistant or faster growing salmon stock can be developed, it will open up many new coastal locations for sea farming, stimulate local economies and provide full-time employment to regions suffering kom chronic high unemployment. Recent advances in gene transfer technology has enabled us to explore the use of transgenic fish as a solution. Experiments were initiated in 1982 and significant progress has now been achiwed. The present paper will highlight some of our recent findings in producing transgenic salmon using the growth hormone (GH) and antifreeze protein (AFP) genes. Many of these studies have been published (Fletcher et cd., 1988; Shears et d.1991; Du et al., 1992a). Biology inlu~iwnal(Spcciai Issue N78 - 1993) Antifreeze Protein Gene Transfer

Atlantic salmon (Salmo sa14 and many other marine teleosts cannot tolerate seawater temperatures below -0.6OC to -0.8OC. In contrast. several fish species such as Antarctic nototheniids (Dfssostichus mawsont Trernatomus borchgreuinkf), Atlantic cod (Gadus morhuu). winter flounder (Pleuronectes americanus). ocean put (Macrozoarces americanus), sea raven (Hernitripterus americanus), and shorthorn sculpin (Myoxocephalus scorpfus) which inhabit the icy waters in polar regions produce antifreeze glycoproteins (AFGP) or antifreeze proteins [AFP) in their sera. These proteins protect the fish from freezing in temperatures close to -2OC by inhibiting ice crystal formation. The properties of many of these proteins have been weil charactexized (Davies & Hew, 1990) (Table 1). In brief, there is one type of AFGP and three distinct types of AFP. The Type 1 AFP are alanine-rich, a helical and amphiphilic polypeptides found in righteye flounders and sculpins. Itpe II AFP is cystine- rich, rich in p sheet structure. found in sea raven, herring and smelt, while Type III AFP is neither alanine-rich nor cystine-rich. It has an average amino acid composition and is found in eel pouts. The mode of action of these proteins, which is noncoliigative. has recently been rehewed (Hew & Yang, 1992).

Table 1. Structural Characteristics of Antifreeze Proteha

AFGP AFPTweI AFP Tvue II AFP Twc iiï.-

Cxboh ydrare Yes No No No Mass 2.600 10 33.000 3300 to 4500 11,0001024,000 6500 Prirnary srnicm (Ala Ala ?hr)n alanine-rich cystine-rich generai disaccaride multiple of eleven disulfide linked aa repu expanded a helical, . B sheet srnicnJred. arnphiphrhc not distinct Temary smcture N.D. 100% helii N.D. N.D. Biosynfhesis rnultiprotein Preprn prepm AFP (?) Protein cornponenu 8 7 2-6 Tm Gene copies N.D. 3040 15 150 Fish species Antaictic nototheniids. right-eye flounden sea raven, eel pouu nonhem cods, (winter flounder) srnelt. (ccean PUS Scul~im herring. wolff~h)

The AFP from winter flounder, in particular. has been studied extensively. Its X-ray structure, mode of action, structural characteristics and biosynthesis have been elucidated (Hew & Yang, 1992). It is synthesized in the liver as a precursor polypeptide of 82 amino acids. This precursor preproAFP is further processed to yield mature AFP. While the pre sequence is involved in the vectorial transport of the nascent polypeptide across the endoplasmic reticulum. the function of proAFP, at present is unknown. The mature AFP occurs in high concentration (10-15mg/ml) in the serum during the winter months. Its synthesis is regulated both by photoperiod and growth hormone. Biology Inler~tional (Special Issue N58 - 1993) The gene structure for the AFGP and AFP have been reported. For example. AFP from the winter flounder belongs to a multigene family with 30-40 gene copies (Scott et ai.. 1985). The majority of the genes are arranged as direct tandem repeats (5-8 copies in a gene cluster). One of the representatives (2A- 7) is shown in Fig. 1. It codes for a major AFP component (HPLC-6). the gene is small, lkb in size, and contains one intron. It is interesting to note that while the 5' flankfng sequences contain the usual basal promoters, such as the TATA and CAAT boxes, putative liver-specific cis-acting sequences including LF-BI, C/EBP etc., are localized in the intron (Chan & Hew, unpublished).

Using purified AFP from the winter flounder, Fletcher et al.. (1986) demonstrated that exogenously injected AFP significantly protected the seawater-acclimated trout from freezing. While the noninjected rainbow trout was incapable of freezing resistance. the injected anirnals could tolerate freezing temperatures as low as -1.2. The ability of injected rainbow trout to tolerate ice at low temperatures correlateci closely with the amount of AFP in its circulation. These experiments suggested that AFP from other species of fish is capable of conferring freeze-tolerance. and have provided the rationale for the antifreeze protein gene transfer in Atlantic salmon.

probe

Fig. 1. A schematic diagtam of flounder AFP gene (2A-7) used in anufreeze gene transfer in Atlantic salrnon. Plasrnid 2A-7 lin- at its unique Eco RI site (E), b shown integrated into salrnon genomic DNA (broken line). The pUC-9 section of 2A-7 1s indicated by double linea and the flounder DNA insert by a single llne within which is located a rectangular box representing the 1-kb AFP gene. The open areas of the box are the exons and the hatched area ts the intenrening sequence. The cleavage sites of Barn HI @3). Sst 1 (S). and Hind III (ii)are rnarked. The 2.7-kb Sst 1 fragment used as a probe is underlined. From Davies et d,1990.

Linearized AFP gene from the winter flounder (2A-7) was microinjected into the fertilized eggs of the Atlantic salrnon via the micropyle (Fletcher et al., 1988). The salmon eggs are opaque with no distinct pronucleus. This injection procedure via the micropyle allows the delivery of foreign DNA to the proximity of the germinal vesicle. A detailed description of the protocol has been described elsewhere (Shears et ai., 1992). In general, there is no Biology Inter~tional (Speciol Issue ND28 - 1993) difference in the survival rate (-80%) between injected and noninjected samples. The incorporation frequency is approximately 2-3% with 40% of the positives expressing the transgenes. Since 1982. approximately 500-2000 eggs were injected per gene construct during the annual spawning season in October. Early experiments focused primarily in developing gene transfer techniques suitable for the salmonids and handling of the salmon eggs. The 1985 injection series, which represents our most complete data is summarized in Table 2. Earlier experiments used Southern blotting to detect transgenesis and the incorporation of the AFP gene into the salrnon genome. More recently, this has been replaced by the sensitive polymerase chain reaction (PO for routine. large scale analysis of samples.

Table 2. AFP gene transfer in Atlantic salrnon (1985 injection series)

1985 lûûû eggp infecteci. with 80% survivai rate. 198ô-1988 Transgenic salrnon identined by Southern and PCR 40%of positives produce pro AFP. 1988-1989 Transgenic founders Pd crossed with wild type and each other. Germ-iine mosaidsm indicated by the variable frequency of transgenesis in F1 individuais (17, 15. 33, and 64%). 1990 F1 tmnsgenics crossed with wild type, transmission of the AFP gene varied from 51 -5446, suggesting stable incorporation at F2 generation. 1990 Many of F1 transgenics produceci pro AFP (0.07 to 20 Md. 1992 Several F1 transgenfcs showed tissue-specific and seasonal expression of AFP transgene. Several F3 generations were produced by crossing F2 mak with dffferent F1 fernales.

Several pertinent observations can be stated. First of all, the transgenes are incorporated into the salrnon genome with a frequency of 2-3%. and many of the Po are mosaic. Genetic cross-breeding of the F1 showed that the AFP gene is stably inherited into the Fs generation.

Expression of transgenes in FI analysed by immunoblot demonstrated that the AFP transgene is producing an immunoreactive positive material with the same molecular mass of proAFP (- 6000 daltons). The Atlantic salmon, apparently lacks the processing enzymes necessary for maturation of AFP. Furthemore, there is variability in transgene expression in individual F1 offspring.

These studies are important in demonstrating the incorporation, inheritance and expression of the AFP gene and lay the foundation and justification for the transgenic approach (Shears et al., 1991). The experiments however, are only partially successful. The level of the AFP expression at present is approximately 20pg/ml. a value too low to offer any practical and meaningful protection against freezing. Although the proAFP is active (-70% as compared Biology Infernational (Special Issu N028 - 1993) to mature AFP), it is advisable that mature AFP be produced to increase its activity. In addition, more work is needed to boost the blood plasma AFP levels. Several strategies are possible and some of these experiments are now in progress.

Development of "AU Fish" Gene Cassette

For the transgenic fish to be acceptable for human consumption, the DNA injected should not appear to pose any hazard to human health, no matter how unlikely it may seem. This discourages the use. for example, of the human growth honnone gene and the use of viral promoters and promoters such as metallothionein which require heavy metals for induction. Therefore, we have developed an "al1 fish gene construct or cassette which is safe and versatile to use. While the genes of interest such as growth hormone, somatolactin, etc.. are al1 derived from fish, the promoter we use to drive the transgenic expression is from the fish antifreeze protein gene. The antifreeze promoters have the following features that offer advantages over many other fish gene promo ters (Table 3).

Table 3. Use of Antiireeze Protein Promoters for Transgenic Studies

1. The genes encode non-toxic, natural occurring îish proteins. 2 The proteins are synthesized predominantly in her, a tissue weU- sulted for the expression of secretory proteins. S The gene(s) 1s absent in most commercially imrortant fish species. Therefore transgencl aetmtion is easy with no background. 4 The absence of AFP gene In recipient fish wiU minimize the interference of homologous promoters. This. in addition, provldes flexibility in the alteration and improvement of AFP promoters. 5 Several types of AFP promoters are available.

The fidelfty of the AFP promoters have been demonstrated in fish culture cells, Japanese medaka embryos, and alsc in transgenic goldfish and Atlantic salmon (Gong et ai,, 199 1; Shears et al., 1991).

Using these AFP promoters, we have constructed several chimeric genes linked to fish GH, prolactin and somatolactin. More recently. the gene cassette has been further improved for easier manipulation (Du et al.. 1992b). Growth Hormone Gene Transfer

In the case of growth hormone gene transfer. we have made two gene constructs using the ocean pout AFP promoter fused to either the cDNA (opAFP-GHc) or the genomic clone (opAFP-GHEC) of GH from chinook salmon. The constructs were injected into salmon eggs in October 1989. We will present only the data derived from the cDNA clone, opAFP-GHc. The construct contains 2kb of 5' and 1.2kb of flanking sequences fkom the ocean pout AFP gene. While the 5' sequence includes all the promoter motifs important for transcriptional initiation, the 3' end consists of the polyadenylation signal AAT AAA and the putative transcription termination signal 'ITIXTCT. Biology Irilernatwnaî (Special Issu NT8 - 1993)

To facilitate the analysis of GH transgenic fish, the following PCR strategies were developed. Four PCR primers, A, B, C. and D were synthesized with the directions relative to the coding strand indicated in Fig 2. Primers A and B are specific for the ocean pout AFP gene and C and D for the GH gene. Four different PCR reactions are possible. These maximize the usage of the oligonucleotides and provide additionai. independent reactions to confinn the presence of transgenes. With the combination of A/B. A/D; and C/B primers, only the transgenic salrnon will be positive in generating PCR fragments of defined size (Fig. 2). With prirners C and D, two PCR fragments of 344 and 199 bp will be generated. The larger fragment is derived from the endogenous GH gene from Atlantic salmon, which encodes a short intron (intron II, 145 bp) between primers C and D while the transgenic, which is a cDNA clone. will only generate a 199 bp fragment. The 344 bp fragment, therefore acts as an internai control for the reaction. These strategies are important to ensure the accuracy of PCR

+ primers A

PREDICIION AB A/D CD C/B CONTROL -- -- 344 - TRANSGENIC 855 333 199 721

Fig. 2. A schernaîic diagram of opAFP-GHc and the positions of the PCR primexa with the predicted size of PCR fragments Ilsted below. Approximately 500 fish were analyzed by PCR using red blood cells. Nine were found to be positive. Severai of these were the largest fish in the tank. Figure 3 shows an example of the results using the combination of A and B. C and D. and A and B primers. The identity of these PCR fragments was conhed by hybridization with GH-specific probe.

It is important to point out that most of the transgenic fish exhibit larger body weight. One of the positive individuals was negative for the presence of the gene in red blood ceiis, but positive using scaie DNA, indicatlng that this fish is mosaic. The weights of these flsh were measured in October 1990 and January 199 1 (Table 4). Our results show that the GH transgenics, on the average are 4-6 fold larger and the largest transgenic fish was 13 times that of the noninjected siblings (Du et ai., 1992a). Our observaffons of dramatic growth in transgenic salmon has recently been confinned by Devlin and CO-workersin transgenic pacific coho salmon using the same opAFP-GHc construct (Devlin, personal cornmunication). 1 2 +-

855 bp -

33 bp- 344 bp- 199 bp-

Fig. 3. Screening of transgenic sabnby PCR. Samples 1 and 2 correspond to two different positive fish. (-) DNA from injected non-tmnsgenic salmon. (+) CH transgene opAFP-GHc. M molecular weight markers. oX- 174-RF DNA Hae ïïI digest. Upper panels represent the use of different primer combinations. A/D, C/D. and A/B. respectively. Lower panels represent Southem blot analysis of the PCR products using CH-specLftc probe E.

GH Weight Growth 1 (ng/ml) (e) Rate Oct 4/90 Jan 12/91 (%May) Transgenic 39.9.214-8.(5) 20.9, k5.13, (6) 47.3,kQ.s.(6) 0.836. ~0.186.(6) Siblings 28.2. k8.8, (8) 7.4. a.26. (43) 9.48,d.6. (43) 0.213,H.023. (43) P NS ~0.05 <0.01 4.001

Table 4. Results on CH gene transfer (AU values presented as means & one standard emr. Numerale in parentheses indicate the number of Bsh. NS means non-sigdicant

One can speculate on reasons for the dramatic growth enhancement observed in these transgenic salmon. First of all, the 4-6 fold increase in the present study is comparable to the 2-3 fold increase seen in salmon growth with the direct administration of human, bovine, chicken. and fish GH (Gill et al.. 1985). Our gene construct expresses the chinook salmon GH which is homologous to that of the Atlantic salmon and might be'better recognized by the endogenous GH receptors compared to other mammalian GHs. The expression of the GH transgene in liver, which is dictated by the AFP promoter, may also facilitate and speed up the interaction of GH with its liver membrane receptors. The GH in this instance. exhibits a paracrine or autocrine function. Lastly, the growth of salmonids in cold water might be suboptimal due perhaps to a reduced synthesis and/or release of endogenous GH from pituitary. The expression of GH in liver and to a lesser extent in other peripheral tissues might be adequate to initiate early growth. Many of these speculations can be experimentally tested. For example, one can examine the growth performance of these transgenic salmon at different culture temperatures, or evaluate the growth enhancement of transgenic fish from warmer waters. Biology International (Spccial Issue NO28 - 1993) Conclusion

Our experiments, together with other reports in the literature. have clearly demonstrated that transgenic fish of improved phenotypic characteristics can be accomplished routinely by gene transfer techniques. In addition to the GH and AFP gene transfer, many exciting and novel experiments can be carried out to further develop new and better strains of fish stocks. Future experiments in our laboratories will focus on the evaluation of the performance of these transgenic salmon under optimal culture conditions and the development of biological containment techniques to minimize the ecological impact from these fish in case of their accidental release. Acknowledgement

We thank Linda Gardiner for the preparation of the manuscript. This work is supported by NSERC Strategic Grant, Canada (to C.L. Hew. G.L. Fletcher, and P.L. Davies), and the Department of Nsheries and Oceans, Canada (studentship to S.J. Du).

References

DAVIES. P.L. & HEW. C.L 1990. Biochemistry of FIsh Antifreeze ProteinS. FASEB J. 42460-2466. DAVIES,P.L.. HEW, C.L.. SHEARS. MA, & FLEXCHER, C.L. 1990. Anühxm ProteIn Expression in Transgenic Salmon IN: UCIA Conference Pmœedhgs on Tmnsgenic M&Is 6-1Merifcine and Agriculture. Alan R Idss. Inc. pp. 141-161. DU. S.J.. GONG. Z.. FLETCHER G.L.. SHEARS. MA, KING, M.J.. IDLER D.R.. & HEW. C.L. 1992a. Growth Enhancement in Transgenic Atlantic Salmon: Use of Al1 Fish Chimeric Gene Constmct. Bfo/Techndogy. 1&176-181. DU, SJ.,GONG, Z.. TAN. C.H.. FLETCHER, G.L., & HEW, C.L 1992b. hiopmentof an "Al1 Fish Gene Cassette for Cene ?i-ansfer in Aquaculture. Marine MdBioL Biotech 1(415):29û-300. FLFCKHER G.L.. KAO. M.H.. & FOURNEY. RM. 1986. Antifreeze Peptides Confer Freezing Resistance to Nsh. Cari. J. Zod 641897-1901. FLETCHER G.L.. SHEARS, M.A. KING, MJ.. DAWES, P.L., & HEW. C.L. 1988. Evidenœ for Anühez Protein Gene Transfer in Atlantic Salmon (SaLno sala?. Can. J. Flsh Aquat Sei 45:352-357. FLETCHER, G.L. & DAWES, P.L. 1991. Transgenic Fish for Aquaculture. IN: J.K. Setlow m.). Geneflc Engineering, Prindples and Methods. Plenum Press, New York. pp. 331 -370. GILL, JA. SUMFiER, J.P., DONALDSON. E.M.. DYE, H.M.. SOUZA, L., BERG, T., WYWCH, J.. & LANCLEY, K. 1985. Recombinant Chicken and Bovine Growth Hormones Accelerate Crowth in Aquacultured Juvenile PaciRc Salmon. Oncorhynchus kkutch B b/Technology. 3:643-64ô. GONG, 2.. VIELKIND, J.R, & HEW, C.L. 1991. Functional Analysis and Temporal Expression of Promoter Regions fmm Flsh Antifreeze Protein Genes in Transgenic Japanese Medaka Embryos. Md Marine BLd Btotech 1:64-72. HEW. C.L. &YANG. D.S.C. 1992. Protein Interaction with Iœ. Eur. J. Bfochem 20333-42. HEW. C.L. & GONG. Z.Y. 1992. mgenicFish: a New Technology for Fish Biology and Aquacultum. Bbbgy International. 242-10. SCOïT, G.K.,HEW, CL, & DAVIES. P.L. 1985. Antifreeze Protein Genes areTandemly Linked and Clustered in the Cenome of the Winter Flounder. Proc. NdAcad Sci U.SA 82:2613-2617. SHEARS. M.A. KING, M.J.. GODDARD. S.V., & FLEXHER, G.L 1992. CeneTransfer Ln Sabnidsby Injection thmugh the Micropyle. IN: C.L. Hew & G.L. Fletcher (Eds.) Thmsgenic Fïsh World ScientiRc Publishing Co., Singapore. pp. 44-60. SHEARS, M.A., FLETCHER, G.L.. HEW. C.L.. GAUTHIER, S.. & DAWES. P.L. 1991. 'I'ransfer. Expression. and Stable Inheritance of Antifreeze Protein Genes in Atlantic Salmon (Salmo salarj. Md Marine Biol. Btotech 1:58+33. Biology Internaiiod (Special Issue N"28 - 1993) Transgene Transmission and Expression in Rainbow Trout and Tilapia

by Norman Maclean, Arati Iyengar. Aziz Rahman, Zorah Sulaiman. & David Penrnan University of Southampton. Bassett Crescent East. Southampton S09 3TU, Hampshire, U.K.

Introduction This paper will describe the production 6f transgenic rainbow trout (Oncorhynchus mykiss) and tilapia (Oreochromis niloticus) by microneedle injection of fertilized eggs with cloned copes of novel gene constructs. Two aspects'of our work with transgenic trout will be presented namely the patterns of inheritance of transgenes by the F1 generation following in vitro fertilization of gametes from transgenic and non-transgenic fish and secondly, the degree of DNA methylation of transgenes in different fish and in different tissues of the same fish. Work with transgenic tilapia has been of shorter duration, and we simply present evidence for their transgenic status, together with transient expression studies using the bacterial gene CAT (chloramphenicol acetyl transferase) when driven by fish-derived promoters. Rainbow Trout

Following the early attempts by Maclean and Talwar (1984) and Zhu et al. (1985) to microinject fish eggs with novel cloned genes, transgenic fish of many species have been produced. These include rainbow trout (references 2, 5 + 6).a species with a relatively long generation tirne but of easy culture and widespread aquacultural importance. Data on germ line transmission in this species has already been published (reference 3) and we here provide some additional information on this topic. As seen in Table 1. adult trout transgenic for a marnmalian gene construct mMtrGh (consisting of a mouse metallothionein promoter spliced to a rat growth hormone structural gene and originally kindly provided by D. Hamer), were crossed with non- transgenic adults. Transgenic status was ascertained by blood and/or fin biopsy, and also by anaiysis of milt in the male fish. It will be seen from the data that mosaic integration of the transgenes results in only half the apparent transgenics yielding transgenic progeny: even males transgenic in sperm do not reveal that multiple integration events into separate chromosomes occur. analyzed carried the transgene. As discussed in references 5, 8 + 9, Southern blot patterns of transgenic fish biopsy DNA suggest true chromosomal integration in all the fish analyzed. Further details of the analysis of these transmission patterns are now in press (reference 10). Biology Intcrnatbd (Speciol Issue NT8 - 1993) Tahle 1.

NO CROSSES TRANSGENE PROGENY TFUNSMISSION MADE PRESENT/ABSENT ANALYSED RESULTS

BLOOD SPERM 1. CONT.9X n/a n/ a 12 Al1 neqative. CONT .d (NOS.1-12) 2. CONT.9 X + + 2 1 10+5+2 (81%) TR. d (Nos.13-34) Positive. (H57 4 Neqative. 3. CONT.9 X + + 2 2 Al1 neqative. TR. d (Nos.35-56) (H61) 4. C0NT.d X + n/a 2 2 Al1 neqative. TR. Q (Nos.57-78) (H60) 5. TR.9 O)+ 9)n/a 2 2 10 (45%) (H60)X d)+ d)+ (Nos.79- Positive. TR. d 100) 12 Neqative. (NTT) 6. CONT.OX + + 22 A11 neqative. TR. cf (Nos.101- (H58) 122) 7. CONT. Q x - - 11 Al1 neqative. INJ .d (Nos. 123- .(Ha() 133) 8. CONT.QX - - 11 A11 negative. INJ .d (Nos. 134- (H79) 144) 9. C0NT.d X + n/ a 2 2 2+1 (14%) TR. O (Nos. 145- Positive. (H59) 166) 19 Negative.

Notes 1) In crosses 2 & 9. the different nurnbers of positive fish indicate dtfferent patterns of transmission obtained.. 2) Crosses 7 & 8 were made from males that had developed from injected eggs but were not positive in blood and were analyzed for transmissfon in case of extreme mosaicism where the blood appeared negative but there was a possibility that the gonads rnay be positive.

DNA methylation may be studied in vivo by the use of the restriction enzymes such as MspI, which cuts CCGG sites in DNA even if the Internai C is methylated, ,HpaII which does not cut CCGG if the internal C is methylated, Hhai which recognizes GCGC but will not cut if the internal C is methylated, and SmaI which recognizes CCCGGG but will not cut if the innermost C is methylated. By using these enzymes we have determined the degree of methylation of the mMrtGh constmct in transgenic fish. anaiyzing different tissues of the same fish and aiso comparing the pattern in different fish. The study has been complicated by the difficulty of determining Biology Inter~twnal (Special Issue ND28 - 1993) transgene mosaicism between tissues and the precise copy number of transgenes in different fish and different tissues of the sarne fish. Southern blot analysis reveals that only small differences in banding patterns of digested sarnples are apparent as between different fish, or different tissues. This supports the view that the sites methylated within the transgene are a function of the inherent sequence properties of the transgene rather than a function of the chromosomal locus of integration. The differences detected between different tissues using the same enzyme (for example MspI) suggest that the transgene is hemimethylated at certain sites in some tissues, but not in other tissues. Such hemimethylation may have important consequences for gene expression. Unfortunately we have not been able to detect expression of the mMtrGh transgene in the rainbow trout. but the work will now be continued with "al1 fish constructs where expression should occur. It wili, of course, be particularly interesting to determine differences of methylation pattern of ùansgenes. especially of promoter regions, in expressing and non-expressing tissues.

Since tilapia are important in aquaculture, have a relatively short generation time. and are easy to breed and culture, we have attempted to apply the transgenic technology to a tilapia species, Oreochromis niloticus. Previous attempts to produce transgenic tilapia by Brem et al. (1988) were successful only with multi-ce11 egg injection and thus the fish produced would be expected to be heavily =ûsaic. At that time also the evidence for genomic integration was not strong. We have now been successful in producing fish which, on southern blotting of biopsy DNA, have chromosomaily integrated copies of the transgene in both blood and fin (Rahman, A. & Maclean, N.). Original experiments involved the sequence mMtrrGh, but lately we have also used FV1 and FV2 constructs (kindly provided by Perry Hackett) which consist of bacterial CAT genes spliced to a carp beta actin promoter sequence. Individuais which proved transgenic in blood or fin have been used in breeding experiments. but analysis of progeny for transgenic status is preliminary. However, using FV1 or FV2 we have analyzed promoter activity by measuring CAT expression. A significant number of tilapia fry express CAT following injection with the FV1 construct. Since we have not as yet correlated CAT expression with transgenic status, much or ali of the measured expression may be transient but it does indicate the ability of the promoter to sustain expression when operating within Tilapia cells.

References

BREM. G., BRENIG. B., HORSTGEN-SCHWARK. G., & WNNACKER E.L. 1988. Gene Transfer in Tilapia (Oreochrornis nhticus). Aquaculture. 68:209-219. CHOURROUT. D.. GUYOMARD. R. & HOUDEBINE. L. 1980. High Eficiency Gene Transfer in Rainbow Trout by Microinjection into Egg Cytoplasm. Aquaculture. 51:143-1 50. GUYOMARD. R.. CHOURROUT, D.. & HOUDEBINE. L. 1988. Productin of Stable Transgenic Fish by Cytoplasmic Injection of Purified Genes. Proc. UCLS Symp. on Gene Pansfer and Gene Therapy, Feb., 1988, Colorado, U.S.A. Biology Inier~iionai (Special Issue NT8 - 1993)

MACLEAN. N. & PENMAN. D.J. 1990. The Application of Cene Manipulation to Aquaculture. Aquaculture. 85: 1-20. MACLEAN. N., PENMAN. D.J., & ZHU. Z. 1987a. Introduction of Novel Genes into Fïsh. Bbtechnology. 5257- 261. MACLEAN, N., PENMAN, D.J.. & TALWAR, S. 1987b. Intductlon of Novel Genes into Rafnbow Trout. EIFAC/FAO 1986 Symposium on Selection. Hybridizaion and Genetic Engineering in FLsh Vol. II. Heenernann. Berlin. MACLEAN. N. & TALWAR, S. 1984. Injection of Cloned Genes into Rainbow Trout Eggs. J. Emb& Eup. Morph 81187. PENMAN. D.J.. BEECHING. A.J.. IYENGAR A., & MACLEAN, N. 1988. Introduction of Metallothionein- somatotrophin Fusion Gene fnto Rainbow Trout: Analysis of Adult Transgenics. Bull. Aquac. Assm. Canada 88-4:137-139. PENMAN. D.J.. BEECHING. AJ.. PENN. S.. & MACLEAN, N. 1990. Factors Affecttng Survival and Integration Foiiowing Microinjection of Novel DNA into Rainbow Trout Eggs. Aquaculture. 8535-50. PENMAN. D.J., iYENGAR, A. BEECHING, A.J., RAHMAN,A. SUWMAN. Z, & MACLEAN. N. 1991. Patterns of Transgene Inheritance in Rainbow Trout (Oncorhynchus mykiss). Mdec. Reprod. Develop. 30.201-206. RAHMAN, M.h & MACLEAN, N. 1992. Production ofTransgenic Tllapta (Oreochrornis nüotkus) by One-Cell- Stage Microinjecîion. Aquaculture. 105:219-232. ZHU. 2.. id. G.. HE. L., & CHEN. S.. 1985. Novel Géne Transfer into the Fertillzed Eggs of Ooldfish (Carassius auratus). 2. Angew. Ichthyd 1:32-34. Biology Internatioml (Special Issue N78- 1993) Transgenic Technology in Fish

by Daniel Chourrout Laboratory of FIsh Cenetics, INRA. 78350 Jouy-endosas,France

Introduction

After much success in eukaryotic gene isolation and their functional test in transfected cultivated cells, strong efforts have been devoted in the last ten years to establish in vivo gene transfer methodologies. The benefits of the transgenic technology for the advancement of basic biology have been numerous, particularly in the mouse. for which the production of transformed lines is still hard bgt almost a routine operation in many laboratories.

In the mouse model, the main difficulty is due to the in vitro manipulation of eggs, which have then to be reimplanted in vivo for long term development. Genes are classically transferred as linear DNA fragments and introduced into one of the two egg pronuclei. In very peculiar cases, one may also use vectors derived from retrovirus genomes, in order to infect older embryos, but this operation does not result in animals having the foreign genes in al1 their cells.

In many fish species. the high fecundity and the extemal development make much easier any manipulation of embryos. However, research teams interested in gene transfer technology have immediately faced two problems: one is the impossibility to visualize the pronuclei in the fertilized egg, for an eventual gene injection; the other is the absence of described retrovinises in fish. which could have been used to generate infectious vectors for gene integration.

In this article. we describe the methodological efforts devoted to overcome these problems in order to obtain stable lines of transgenic fishes.

Fate of Foreign DNA injecteci into the Cytoplasm

The cytoplasmic injection has been originally tested in the Xenopus for which problems similar to those met in fish were encountered. Very large arnounts of circular or linearized conventional plasmids (1 to 1000 pg) were introduced into the egg animal pole. before its first cleavage. The foreign DNA was often considerably amplified during the segmentation. and then almost totally degraded until the end of the gastrulation. However, several copies of the injecied sequences have sometimes been detected after the metamorphosis, and in one case in the offsprings of one male (Etkin & Pearman, 1987). The major problem in the Xenopus literature is a lack of information conceming the fate of foreign DNA in adults resulting from the injections. as well as in the further generations obtained by their reproduction. This is of course fundamental for anyone who wants to know whether the cytoplasmic injection can lead to stable transgenic lines.

In fish as in amphibians. authors who have tested the cytoplasmic injection Biology Internalional (Special Issue N78- 1993) have followed very diverse protocols. DNA can be introduced in manually or enzymatically dechorionated eggs, or through the chorion kept around the egg. In the latter case. injections are performed either through a hole previously drilled in the egg shell, or through the egg micropyle. In order to facilitate such injections. a research group has set up a chemical treatment. which softens the chorion and increases its transparency (Yoshizaki et al., 1989). Two other variables may have a particular importance for the subsequent success of the transformation: the amount of DNA injected which may vary between 1 to 1O00 pg in different reports, and the time of injection. Several authors have treated pluricellular embryos. but the majority of them have performed the injections at the first cell stage. However, even in these cases, the position of the injection time in relation to the endogenous DNA replication remains unclear. because basic information of the first ce11 cycles are lacking for the fish egg.

Most contributions dealing with this method mention the detection of foreign DNA in embryos or young fishes resulting from the cytoplasmic injection (Zhu et ai.. 1985; Yoon et ai., 1989; 1990 in goldfish; Chourrout et al., 1986; Maclean et ai.. 1987; McEvoy et al., 1988; Fletcher et ai., 1988; Rokkones et ai,, 1989 in salmonids; Dunham et af.. 1987 in channel catfish; Zhang et al., 1990 in common carp; Stuart et ai., 1988 in zebrafish; Chong & Vielkind, 1989 in medaka; Brem et al., 1988 in one tilapia; Kozlov et ai.. 1988 & Benyumov et al., 1989 in one loach). These precocious examinations pose the problem of an eventual transient persistence of large amounts of foreign DNA out of the chromosomes, promised to a rapid degradation. At this stage. the yields of positive animais detected in dot blots is usually inferior to 10% when a single organ is analyzed.

Among the several studies in which older fishes have been exarnined, ours in the rainbow trout are an exception because we have obtained very high rates of positive fishes (50% and more after two years). which is for example, largely superior to those recorded by two other groups working also with salmonids (Maclean et ai., 1987; Fletcher et ai.. 1988). Penman et ai. (1990) have been able to explain partially this major difference of yields, by the very different amounts of DNA injected in different studies.

In Our case as well as for other groups, the dot blot analysis of DNA extracted from several organs unequivocally reveals a mosaic distribution of the foreign genes in most animals of the FO generation (directly issued from the treated eggs). Restriction studies indicate that multiple copies of the injected sequences are generally organized in multimers. AdditionalW. the hybridization signals are found in the high molecular weight fractions. when the genomic DNA is analyzed without digestion by restriction enzymes. However. the results of restriction analyses cannot be totally convincing for the question of integration. because the hybridizing bands interpreted as junction fragments (associating the foreign DNA and the fish genome) can be equally explained by foreign DNA rearrangements without subsequent integration. This type of analyses is also complicated by the mosaic distribution of the foreign genes. We. must mention one study which has rather clearly suggested the integration from the FO generation: Penman et cd, (1990) have observed hybridization Biology Interndio~l (Speciol Issue N78- 1993) bands at an intermediate level between that of the undigested DNA and that of the introduced plasmid. after cleavage of the fish DNA with restriction enzymes having no site in the injected plasmid.

A more interesting but longer way to address the question of gene integration consists of the examination of F1 and F2 generations of fish, obtained by the reproduction of putatively transgenic FO adults with wild-type partners. Four research groups have clearly concluded for the transmission of foreign genes through the germiine (Stuart et cd, 1988 & Culp et al., 1991 in zebrafish; Chourrout et cd, 1988; Guyomard et aL. 1989a; 1989b & Tewari et al., unpublished, in rainbow trout; Shears et ai., 1991 in Atlantic salmon). The proportions of positive F1 offsprings are usually lower than 5096, indicating that the FO gennline is also mosaic (if the foreign DNA is actually integrated). The integration hypothesis is rather weii supported by the analysis of Southern blots, which provide much simpler images than for the FO generation, containing bands likely to represent junction fragments. In these three species, F1 adults have also been reproduced with wild-type partners. and furnished Mendelian proportions of positive offsprings, usually 50% (suggesting a single foreign insertion), but also 75% (suggesting two insertions).

Although all this information tends to indicate that the cytoplasmic injection into fish eggs leads to a stable integration as does the pronuclear injection in the mouse. we must mention that some of our results in two F2 families (with 50%) transmission rate) have raised the possibility that foreign multimers might be stabilized in an extrachromosomal form (Tewari et al., unpublished). As a matter of fact, the images of Southem blots permitted to recognize bands corresponding to the free ends of the injected construct. which should be linked to each other within the multimer and linked to the genome if the multimer is integrated. A detailed examination of these two families with various restriction analyses in diverse conditions of gel electrophoresis (neutral and alkaline migration, PFGE) and with in situ hybridization on chromosomes has permitted to explain these surprising bands without rejecting the integration hypothesis. They result from the originality of the multimer organization after cytoplasmic injection. In the mouse. the multiple copies are oriented in the same direction in the multimer. while in Xenopus and in fish. the copies are oriented randomly. This latter organization creates a long palindrome at ail the junctions between copies oriented in opposite directions. and it seems that these palindromes induce the formation of secondary structures such as hairpin loops, born by intra-strand pairing. This mode1 permits interpreting the unexpected bands as the result of a cleavage of hairpin loop heads by the restriction enzymes.

Fate of Foreign DNA Introduced in Other Ways

A Japanese research group has accomplished very original work on the medaka, in testing two other routes of gene transfer. The first one is the injection of several ten thousand copies of circular plasmid (instead of several ten millions for the cytoplasmic injection), in the germinal vesicle of the immature oocytes (Ozato et ai., 1986). This nucleus has the advantage of being Biology Iniernatio~t (Special Issue N78- 199.3) much larger than the pronuclei which cannot be visualized in vivo. Copies of the injected plasmids have been subsequentiy found in the embryos resulting from the oocytes matured in vitro. However, nothing permits yet to know whether this original methods leads to stable integration. If it is the case, the plasmids have jointed the genome after the fbst round of replication. because the pictures of in situ hybridization, perfomed on histological sections of embryos. clearly suggest a mosaic distribution. The second method tested by this group consists of the electroporation of eggs, which has not only led to high rates of foreign DNA persistence in the FO generation, but also to the germline transmission (Inoue et d.1990).

Much more intriguing are the results reported on the goldfish by another team, who injected messenger RNAs into the cytoplasm of fertilized eggs. The DNA analyses suggest that the foreign RNA has been retrotranscribed and incorporated as a DNA copy into the host genome (Niu et al., 1989).

Other efforts have been concentrated on attempts to transform the sperm cells before fertilization, by incubating them with linear DNA. The only written report on the subject describes negative results with this method in the rainbow trout (Chourrout & Perrot, in press). E$rpression of Foreign DNA in Tramgenic Fish

The first convincing arguments for the expression of foreign genes has come out of the injection of chicken delta crystallin gene (absent in fish) into the germinal vesicle of medaka oocytes. The protein has been first recognized in seven-day-old embryos by immunohistology and Western blotting (Ozato et al., 1986). A more recent and complementary study (Inoue et ai., 1989) has shown an earlier expression in the lens, at the time of its formation. This is in accordance with the information previously collected with this gene in chicken and in mammalian transfected cells. and therefore illustrates the interest of transformed Bsh embryos to study the function of heterologous sequences. We have to mention here that this type of expression may be only transient, because the germinal vesicle injection is not proven to lead to stable integration.

The evaluation of sweral other expression data, obtained after cytoplasmic injection is more difficult (McEvoy et al., 1988; Yoon et ai., 1990). In addition to technical problems due to obvious "backgrounds" of expression in negative controls. the youth of the animals examined may be a difficulty, because mostiy, if not all of this expression may result from non-incorporated DNA. This has been clearly suggested in another study of salmonid embryos (Rokkones et al., 1989). after cytoplasmic injection which is known to result in a strong transient amplification of large arnounts of foreign DNA.

A large part of the expression studies are based on heterologous gene constructs. because relatively few fisli genes or cDNAs had been isolated before the last years. We have, for example, injected into rainbow trout eggs several genes and cDNAs of mammalian growth hormone coupled with several classical Bioiogy Internationai (Specbl Issue N028- 1993) heterologous promoters (Chourrout et ai., 1988; Guyomard et al., 1989a; 1989b; Chourrout et ai., unpublished). None of these constructs has been effective in the production of plasmatic hormone. However. only two fish out of one hundred harboring the bovine growth hormone cDNA (associated with the LTR of RSV and the terminator of the bGH gene) have constitutively produced the bovine growth hormone. showing that this heterologous sequence could be transcribed and that the bGH could be secreted in the trout bloodstream. However, Northem analyses suggest that this expression was accidentally due to favourable position effects in the genome. where endogenous transcription signais have replaced those probably ineffective in the construct. These studies contrast with one on the loach (Benyumov et al., 1989). where the authors mention a growth rate simulation consecutive to the introduction of similar constructs: however. few animals have been exarnined here, and it seems that the foreign hormone has not been dosed.

In two in vitro works of fish ceii transfections with such constructs associating foreign promoters and mammalian GH coding sequences (Freidereich & Schartl. 1990; Bearzotti et al.. in press), the presence of foreign hormones have not been detected in significant amounts. either in the culture medium or in cellular extracts. However, these studies as well as one by Liu et ai. (1990) have clearly shown that the expression of the CAT reporter gene can be easily observed in vitro from ,heterologous and homologous promoters. The difficulty to express marnmalian GH genes in fish cells might be due to an inability to process the corresponding immature RNAs (Rokkones et al., 1989; Freidereich & Schartl, 1990). But the fact that cDNAs are not expressed either suggests the existence of other obstacles upstream and/or downstream of the RNA maturation.

The CAT gene expression has also been detected severai times in vivo, in medakas of FO generation (Chong & Vielkind, 1989) and more interestingly in F1 transgenic zebrafish (Stuart et d,1990). Another study reports positive results of CAT expression, but at an early stage of the FO generation (Liu et al., 1990). 'ln the rainbow trout. a constitutive and ubiquitous expression of the CAT gene has been observed in both FO and F1 generations with the human CMV immediate-early enhancer/promoter (Tewari et al.. unpublished). Therefore, and at least in the case of an intronless gene coding for a non- secreted protein such as CAT, the problem of foreign gene expression in fish is not always encountered.

In fish gene transfer, there will be a need as for other animals, to manipulate the expression level and localization. Up to now, this has been achieved with heterologous promoters coupled with the CAT gene. The mammalian regulatory elements seem to keep their main properties when they are introduced into fish ceiis. The promoter of human heat shock protein gene 70 is inducible by heating transfected carp cells (Liu et ai.. 1990; Bearzotti et al., in press). Recently, we have also tested in vivo two constmcts associating the CAT gene and mouse immunoglobulin gene enhancer/promoters (Michard- Vanhée et ai., unpublished): the expression of the reporter gene was restricted to the white ceils. and predominant in the B lymphocytes. Therefore, the first attempts to target the expression in certain tissues of transgenic fish are very encouraging.

Problems to express the gene introduced are not expected when homologous fish genes are tested. The expression of a flounder antifreeze gene in the Atlantic salmon. although very low up to now (for reasons that are difficult to elucidate), illustrates the interest of this strategy (Davies et al.. 1989). A construct coupling the LTR of RSV with an incomplete cDNA of trout GH (without the signal-peptide) has been expressed in the common carp, in the FO and F1 generations (Zhang et ai.. 1990). It is not clear whether the hormone was dosed in the plasma, but the fact that the transgenic carps grew faster than their non-transgenic siblings is a priori surprising.

Finally. one can be extremely surprised by the fact that rabbit globin messengers introduced into goldfish eggs have led to an expression of the protein in the resulting fish (Niu & Tung, 1977; Niu et ai., 1981; Niu et al., 1989). This would mean that endogenous sequences. localized near the DNA copies of the retrotranscribed messengers, have driven the expression. Conclusion

The production of transgenic fish has become a reality and consists in most cases of the cytoplasmic injection of linear DNA. There are no applications yet. that can be seriously foreseen to improve fish genetically with gene transfer, as it is for example the case for cultivated crops. The success, in fish but also in al1 domestic animals, will depend upon the capacities to control and manipulate the expression levels in vivo. For fish. this may require the isolation and characterization of many homologous genes.

The system of cytoplasmic injection can already be used in many studies of basic biology, as it has been the case in mouse or drosophila with other methods of gene transfer. We have shown how it is possible to study in fish the properties of regulatory elements by associating them with a reporter gene. However, a major difficulty of the system is the management of transgenic fish lines, which cannot be obtained quickly because of the initial mosaicism. Methodological efforts should lead to the development of new tools which will not present this practical setback.

References

BEARZOTIl. M., PERROT, E., MICHARD-VANHEE, C.. JOLIVET, G., Am.J.. THERON, M.C., PUISSANT. C.. GRABOWSKI, H., DREANO. M.. KOPCHICK. J.J., POWELL, R.. GANNON. F., HOUDEBINE, L.M.. & CHOURROUT. D. The Expression of Various Gene Constructs Transfecied into Fish Cells. 3. Bwteck In press. BENYUMOV, A.O., YENIKOPOLOV. G.N.. BARMINTSEV. V.A., ZELENINA, LA., SLEFEOVA, L.A.. DORONIN, Y.K.. GOLICHENKOV, V.A., GRASCHUK. M.&. GEORGIEV, G.P.. RUBTSOV. P.M., SKRYABIN. G.. BAEV, A.A.. 1989. Integrauon and Expression of Hurnan Growth Hormone Cene in Teleostei. Genetika 25:24-35.in Tilapia (Oreochrornls niiotfcus]. Aquaculture. 69:209-219. CHONG. S.S.C. & VIELKiND, J.R. 1989. Expression and Fate of CAT Reporter Gene Microinjected into Fertilized Medaka (Oryzias &@es) Eggs in the Form of Plasmid DNA. Recombinant Phage Particles and its DNA. Theor. AppL Genet 78:369-380. Biology International (Special Issue N78-1993)

CHOURROUT, D.. GUYOMARD, R., & HOUDEBINE. L.M. 1986. High Emciency Gene Transfer in Rainbow Trout (Salmo gafrdnerf Rich) by Microinjection into Egg Cytoplasm. Aquaculture. 51: 143-150. CHOURROUT, D.. GUYOMARD, R, & HOUDEBINE, LM. 1990. Techniques for the Development of Transgenic FLsh: A Review. IN: A. Chmh (Ed.) 7Yansgenic Models in Medicine and Agrtculture. Wilgr-Liss, Inc. pp. 89-99. CHOURROUT. D., CUYOMARD. R, LEROUX C., POURRAIN. F.. & HOUDEBINE, L.M. 1988. Integration and Germline Transmission of Foreign Genes in Trout after Injection into the Egg Cytoplasm. J. CelL Biochern Suppl. 12B:188. CHOURROUT. D. & PERROT. E. No Transgenic Rainbow Trout Produced by Fertilization with Sperm Incubated with Linear DNA. Md Mar. Bid Bbtech In press. CULP. P., NUSSLEIN-VOLHARD. C.. & HOPKINS. N. 1991. High-Frequency Germ-Line Transmission of Plasmid DNA Sequences Injected into Fertilized Zebrafish Eggs. Proc. NaYL AdSci U.S.A. 88:7953- 7957. DAWES. P.L., HEW, C.L.. SHEARS, M.A., & FLETCHER C.L. 1989. AntifreQe Protein Expression in Transgenic Salmon J. CeU. Biochern SuppL 138-169. DUNHAM. R.A.. EASH. J.. ASKINS, J., & TOWNES. J.M. 1987. Transfer of Metallothionein Human Growth Hormone Fusion Cene into Channel Catfish. 'ItrYls. Amer. FLsch Soc. 116:87-91. ETKiN, L.D. & PEARMAN. B. 1987. Distribution, Expression and Germline Transmission of Exogenous DNA Sequences Following Microinjection into Xenopus Laevis Eggs. Dewloprnent 99:15-23. FLJTKHER G.L., SHEARS. M.A., KING. M.J., DAWES. P.1,. 6r HEW. C.L. 1988. Evidence for Antifreeze Protein Gene Transfer in Atlantic Salmon (Salm sdar). Can J. Fish Aqud Sci 4.5352-357. FREIDEREICH, H., & SCHARTL, M. 1990. Transient Expression Directed by Homologous and Heterologous Promoter and Enhancer Sequences in Fish Cells. Nucletc Acids Res. 18:3299-3305. GUYOMARD. R.. CHOURROUT, D.. &'HOUDEBINE, L.M. 1989. Production of Stable Transgenic Fish bj Cytoplasmic Injection of Punfied Genes. IN: 1. Verma. R Mulligan &ABeauset fEds.) Gene îYansjer anc! Gene 'Iherapy. Alan R Lis. Inc. pp. 9-18. GUYOMARD. R.. CHOURROUT. D., LEROUX C., HOUDEBINE, L.M.. & POURRAIN, F. 1989. Integration and Germline Transmission of Foreign Genes Microinjected into Fertilized Trout Eggs. Biochim. 71:857- 883. INOUE. K.. OZATO. K.. KONDOH, H.. IWAMATSU, T., WAKAMATSU. Y.. FWITA, T., & OKADA, T.S. 1989. Stage- Dependent Expression of the Chicken Delta-Crystallin Cene in Transgenic Fish Embryo. CelL Dffler. Deu. 27:574%. INOUE, K.. YAMASHITA. S.. HATA. J.I., KABENO. S.. ASADA, S., NAGAHISA. E.. & FUJITA. T. 1990. Electroporation as a New Techriique for Producing Transgenic Fish. CelL Wer.Dev. 29: 123-128. KOZLOV. A.P., RESHETNIKOV, V.L., KORZH. V.P., & NEYFAKH. A.A. 1988. The Fate of Plasrnid DNA in the Developing Embryos of Loach, Msigwnus jossilis L Molecular Btdogy (Russia). 22: 1614-1622. LIU. 2.. MOAV. B.. FARAS, A.J.. CUISE. K.S.. KAPUSCINSKI. A.R.. & HACKEïT. P.B. 1990 Development of Expression Vectors for Transgenic Fish. Bw/Technology. 8: 1268- 1272. MACLEAN. N.. PENMAN, D.. & ZHU. Z. 1987. Introduction of Novel Cenes into Fish. Bb/Technology. 5:257- 281. McEVOY, T., STACK, M., KEANE, B.. BARRY, T.. SREENAN, J. & GANNON, F. The Expression of Foreign Genes in Salmon Embryo. Aquaculture. W27-37. NIU, M.C.. XUE. C.X, NIU, L.C., & HUANG. H.Z. 1989. Transfer of Information from mRNA to Chromosomes by Reverse Transcription in Early Development of Goldfish Eggs. CeL Md. Bid 35:333-345. OZATO. K.. KONDOH, H.. INOHARA, H., IWAMATSU, T.. WAKAMATSU, Y.. & OKADA, T.S. 1986. Production of Transgenic Fish: Intductin and Expression of Chicken Delta- Crystallin Cene in Medaka Embryos. Cell. Dtferentiatfon 19:237-244. PENMAN, D.J.. BEECHINC, A.J.. PENN, S.. & MACLEAN, N. 1990. Factors Affecting Su~valand Integration Following Microinjection of Novel DNA into Rainbow Trout Egg. Aquaculture. 85:35-50. ROKKONES, E.. ALLESTROM. P.. SKJERVOLD, D.H.. & GAUIVIK. KM. 1989. Microinjection and Expression of a Mouse Metaliothionein Human Growth Hormone Gene in Fertilized Salmon Eggs. J. Comp. Physid B. 158:751-758. SHEARS, M.A., FLETCHER. G.L., HEW. C.L.. GAUTHIER, S., & DAWES. P.L. 199 1. Transfer, Expression and Stable Inheritance of Antifreeze Protein Genes in Atlantic Salmon (Salm salar). Mol. Mar. Bid Bwtech &:58-63. STUART. C.W.. McMURRAY. J.V.. & WESTERFIELD. M. 1988. Replication. Integration and Germline Transmission of Foreign Sequences Injected into Early Zebrafish Embryos. Dewlopment 103405-422. STUAFT, G.W., McMURRAY. J.V.. & WESIERFIELD. M. 1989. GedneTransformation of the Zebrafish. IN: 1. Verma, R Mulligan &A. Beauset (Eds.) Cene 7Yansjer and Gene Therapy. Alan R Liss. Inc. pp. 19-28. STUART, C.W., WELKIND. J.R., McMURRAY, J.V.. & WESTERFIELD. M. 1990. Stable Lines of Transgenic Zebrafish Exhibit Reproducible Patterns of Transgene Expression. Dewlopment 1CKk577-584.

YOON,S.J.. LIU. Z., KAPUSCINSKI. A.R., HACKEXT, P.B.. FARAS. A.. & CUISE. K.S. 1989. Successful Gene Transfer in Flsh. IN. 1. Verma, R Mulligan & A. Beauset (Eds.) Gene 7Yansfer and Gene Thempy. Alan R Lis. Inc. pp. 29-34. Biology Infernational (Specùal Issue N28- 1993)

YOON, S.J., HALLERMAN. E.M.. CROSS. M.. LIU, 2.. SCHNEIDER. J.F.. FARAS, AJ., HACKE?T, P.B., KAPUSCINSKI, AR.. & GUISE. K.S. 1990. Transfer of the Gene for Neornycin Resistance into Goldfish. Carassius auratus. Aquaculture. 8521-33. YOSHISAKI. G., OSHIRO. T., & TAKASHIMA. F. 1989. Prevention of Hardenlng of Chorion and Dechorionatton for Micminjection into Flsh Eggs. Nippon Suisan GokkaishL 55269. ZHANG, P.. HAYAT. M.. JOYCE, C.. GONZALEZ-VILLASENOR, L.I.. LIN, C.M., DUNHAM, RA,CHEN. T.T.. & POWERS. D.A. 1990. Gene Transfer Expression and Inheritance of pRSV-Rainbow Trout-GH cDNA in the Comrnon Carp. Cyprinus carpio (Unnaeus). MdRepd Develop. 253-13. ZHU, Z., LI. G.. HE. L.. & CHEN. S. 1985. Novel Gene Transfer into the Fertili2;ed Eggs of Goldfish. Z angew. IchthyoL 1:31-34. Bwlogy Inlernatwnal (Special Issue N028- 1993)

Gene Transfer in Medaka

by K. ozato Department of Biology. Faculty of Liberal Arts and Sciences. Kyoto University, Kyoto 606. Japan

Several years ago. we were discussing Our research funds with governrnental officiais of the fisheries department. They asked 'Why do you use medaka. and not trout or salmon?" Of course. the medaka is not food for human consumption. but is produced as a food for ornamental tropical fish in Japanese fish farms. But now, no one asks such a question, as it is understood that the application of transgenic technology to aquaculture is not so easy, and still requires extensive basic research for many years to corne. In such research, a model or standard system is necessary. very much as the mouse is used as a model for medical research. In this paper, I evaluate the medaka as a model in transgenic fish studies.

The SdestVertebrate

"Medaka" are two Japanese words, "me" meaning "eye" and "daka" meaning "high. thus. the eye in this species is located at a high position of the head. Until about ten years ago, the medaka had been found everywhere in small ponds and brooks in the country of Japan, but has been losing its habitat by indust-rialization of the society. The medaka has been exceedingly loved by Japanese people as a s-pbol of nature, and they are taught the "song of medaka" in primary schoois. Japanese biologists therefore are sensitive to the cultural background of this tiny animal when using it for experiments.

The medaka has been used for biological experiments from the beginning of this century. It is one of the smdest vertebrates (2-3 cm in the total length), 2-3 months in generation tirne, and has a life span of about 2 years. Females lay eggs every day throughout the year under laboratory conditions. Eggs are highly transparent and hatch within 10 days. In some laboratories, the medaka is maintained in mouse cages. In Dr. Y. Taguchi's laboratory (National Institute of Radiobiological Sciences, Chiba), the medaka is bred in thousands of mouse cages in air-conditioned rooms as well as in field ponds. Dr. C.L. Hew (Toronto) cails the medaka "mouse cage fish. These zoological characteristics make this fish an ideal model for biological experiments in various fields of biology.

The short generation tirne enables us to perform genetical experiments. Dr. H. Tomita (Laboratory of Freshwater Fish Stocks. Nagoya Univ., Nagoya) has collected and genetically analyzed about 70 mutants from natural populations. There are many color mutants including albinos and some morphological mutants including homeotic mutants. Several inbred strains have been produced in wild type populations and orange-red varieties by Dr. Taguchi. It is shown by the scale transplantation that these strains are inbred. Furthermore, each strain shows specific isozyme patterns in some enzymes. Bwlogy Iniernaîio~l (Spccial Issue N78- 1993) These strains are powerful tools in experiments in which the genetic background is important.

Embryo Manipulation

In microinjection, we use oocytes instead of fertilized eggs considering that the chorion is hard and the nucleus cannot be recognized under the microscope. In addition, when we started experiments in the early 1980s. it was said that the cytoplasmic injection was not efficient for gene transfer. The ovulation cycle of the medaka is 24 hours. Under the 14 hours light and 10 hours darkness condition. the fish ovulate at the tlme of light. Oocytes 9 hours before ovulation have large germinal vesicles, to which DNA is injected. These oocytes are cultured for several hours in a mammalian culture medium and fertilized by artificial insemination. In these experimental conditions. 70% oocytes injected are fertilized and 50% develop nomdy to the hatching stage. Thus. medaka eggs are only one example in which the nuclear microinjection can be applicable. Zkrzntly. the cytoplasmic injection and electroporation methods have been developed to introduce DNA into the cytoplasm of fertilized eggs. At the present, it is unknown which method is the most efficient for introduction and expression of foreign genes.

Stage-Dependent Expression of Mode1 Genes

When we started transgenic experiments, no fish gene or reporter genes were available for use. The first gene used was genomic chicken ii-crystallin gene containing its own promoter and enhancer. Detection of the foreign DNA and proteins was done in 7 day-old embryos just before hatching. About 50% of injected eggs developed normally to this stage. We could detect DNA in 50% of embryos examined and crystallin polypeptides in 30% of embryos examined. About tissue specific expression, the crystallin gene was expressed in many tissues, in the brain, gill, spinal cord, muscle, intestine. and retina. The problem is that expression in the lens was rather weak or absent. In order to identi@ stages where gene expression occurred, we examined gene expression at four stages. In the lens, expression was detected in a specific stage of lens formation. In other tissues, gene expression occurred according to a specific pattern. That is, in early stages, expression was detected mainly in mesodermal tissues. In 7 day-old embryos. expression occurred dramatically in many tissues. Thus. gene expression of the 3-crystallin gene is controlled developmentally. although the mechanism is stiii unknown. But it seems that this pattern of expression reflects the order of tissue differentiation, because tissue differentiation of mesodemal tissues is generaily accompiished first.

In general, gene expression is not found in animal embryos before the gastrula stage. This phenomenon has been proved by various biochemical and histological methods. Here we wiii show that this is also found in the case of exogenous genes introduced by microinjection. PmiwZ contains promoters and Biology Internatw~l(Special Issue N"28-1993)

stages is distinctive. This suggests that there is a transition phase in gene expression between morula and blastula stages. which may be similar to the mid-blastula transition in amphibian embryos. Thus, stage-dependent expression of a tissue gene and a reporter gene shows that transgenic medaka may be a good model for studying gene expression in development.

Aquaculture and Others

Besides genes mentioned above, may genes have been introduced and expressed successfully in medaka. They are fish and mammalian growth hormone genes (Inoue et al.. 1990). mouse tyrosinase gene (Matsumoto et al., 1991; Winkler et al., 1991). fish oncogene (Winkler et ai., 1991). fire fly luciferase gene (Tazawa et d. 1989; Tamiya et al., 1990). and lacZ and CAT fused to various promoters (Chong et d, 1989; Ozato et al.. 1990; Winkler et al, 1991; Inoue et ai., 1991). Although these studies are not still mature. they show that the medaka is a potent model for transgenic fish research.

Technology Transfer

Aquaculture is one of the most important fields of science and technology in developing countries. especially in Asian countries. In these past three years, 1 have participated in training courses or workshops for Asian researchers in Kyoto, as weli as other countries.

A 'Workshop on Cloning and Transfer of Gene Constructs in Fish Embryos" was held at the National Institute of Immunology in New Delhi, from 5-19 March, 1990. The workshop was organized by Dr. N. Haque, and invited lecturers also included Drs. C.L. Chrisman (Perdue University, U.S.A.) and E. Rokkones (Oslo University. Norway). 1 brought medaka of which the spawning cycle had been conditioned with light in advance in Japan. Dr. Rokkones and 1 then joined the workshop on "Gene and Chromosome Manipulation in Fish, which was held on 20 March-3 April in Madurai-Kamaraji University (southern part of India), and gave a training course on the microinjection to zebrafish. The workshop was organized by Dr. T.J. Pandian, and supported by the Department of Biotechnology of the Indian government.

In April 1991. 1 joined a one month workshop on "The basic study of Transgenic Fish which was held at the Institute of Fisheries Science, National University of Taiwan, and organized by Dr. H.J. Tsai. At this time 1 taught microinjection of gene constructs to oocyte nucleus and gene expression in embryos using medaka. This workshop was supported by the Asian Pacific .Association of Science and Technology. Through these experiences, 1 realized jhat the-medaka is a very convenient fish to use as material in training courses .>rworkshops.

In conclusion. 1 would like to emphasize that the medaka is an ideal model for , transgenic fish research in their biological and embryological characteristics, 4 109 Biology lnterma!io~I (Special Issu N98- 1993) stage-dependent patterns of transgene expression, and convenience as a teaching material. Thus. the medaka is ais0 an ideal model for basic research of aquaculture (Fig. 1).

Medaka play hide-and-seek ilc alm mon =kT-kn

Hg. 1. Medakm u a model for basic mwch in aquaculture

References

OZ4m. K. KONDOH. H.. INOHARA H.. IWAMATSU, T.. WAKAMATSU, Y.. & OKADA, T.S. 1986. Production of Transgenic Flsh: Introducün and Expression of Chtcken Delta- Crystallin Gene in Medaka Embryos. Cell DtfferentlatIon 19:237-244. INOUE, K., OZATO, K. KOMXIH, H.. IWAMATSU, T.. WAKAh4ATSU. Y.. FWïîA, T., & OKADA, T.S. 1989. Stage- Dependent Expresston of the Chicken Delta-Crystallin Cene in Transgenic Fish Embryos. Ceii DSffer. M. 27:576ô. OZATO. K., INOUE, K,& WAKAMATSU, Y. 1989. Transgenic Flsh: Biological and Technical Problems. ZooL Sc t &44-5-457. OZATO, K., INOUE, K. & WAKAMATSU. Y. 1992. Cene ïhmfer and Expression in Medaka Embtyos. IN: C.L Hew & G.L. Fletcher ws.) Tmnsgentc FLsh World SdentiBc Singapore pp. 27-43. Biology Internatio~l (Special Issue N28- 1993)

Surnrnary by C.L. Hew

A general discussion was held on the last day of the IUBS Symposium, on 27 November. 199 1. This discussion. chaired by Drs. C. Hew, P.G.W.J. van Oordt, and Y. Nagahama covered many broad topics of general interest arising from the scientific session and the proposed RBA programme. Al1 participants expresseci their gratitude to IUBS and the Japanese coileagues for organizing a high calibre and timely symposium.

Dr. van Oordt gave an overview of the historical background and progress leading to the formation of the RBA programme. Following the earlier meeting in Taipei. May 1991, with the input of many respected colieagues, three major global initiatives in reproductive biology in aquaculture were developed. These included 1. Manipulation of reproductive control mechanisms; 2. Gamete physiology; 3. Genetic manipulation. These programmes wili include fish. molluscs, and crustaceans.

Realizing a very fast progress made in the area covered by the RBA programme. a need to establish an efficient network for swift communication between the researchers at the global level has been pointed out by several participants. The RBA programme is hopeful to take an initiative in this area. Dr. Talal Younès, IUBS Executive Director. was very enthusiastic and supportive of the proposed programme. He agreed to assist Ln seeking funding support from major intemational agencies.

Several speakers pointed out that the Reproductive Biology in Aquaculture Programme should include somatic growth as an integral part of the programme. Dr. Howard Bern informed the participants that a strategic plan on aquaculture by the U.S. govemment has combined both reproduction and growth, which are interdependent. It was felt that the RBA may wish to broaden its emphasis.

Considerable time was spent on whether there is a need to identiQ a "mode1 fish". This issue arose because of the preference of different investigators in using several fish species for experimental animals. The goldfish (Carassius auratus L.) has been used extensively by many fish neuroendocrinologists to study neuroendocrine-pituitary functions and controls. The biology of Japanese medaka, Oyztas latipas, on the other hand, has been investigated by Dr. Yamamoto and other workers, Drs. Shima, Shimada. and Ozato (who attended this meeting), for mutagenesis, the development of inbred strains and transgenesis. The zebrafish (Brachydanio rerio), has gained considerable popularity for many molecular biologists and developmental biologists in North Arnerica and now intemationally. Both zebrafish and medaka are comparable and there has been considerable debate on whether one Bsh is superior and Biology International (Special Issue N58- 1993) should be developed into a "mode1 fish. For instance, in view of genetic background, medaka has preference to zebrafish. owing to a long and traditional investigation by Japanese workers. On the other hand, the zebrafish is more suitable for embryological manipulation. However, many speakers felt that no single fish species can fulfil al1 the needs and requirements. Depending on the type of research, one fish might be better than the others. There is a general wisdom that

much more can be learned from the comparative studies from any diverse fish species.

Dr. Hew described a prelirninary RBA proposal dealing in transgenic fish. The proposal was formulated after general discussion from the Baltimore meeting in October, 1991. The programme describes international collaboration and training in transgenic fish, and if successful. wiil provide funding support for biotechnology training in gene cloning and gene transfer technology, workshop and lectureship by established investigators. At the global level, it is very important to provide an opportunity for young investigators in developing countries to study in pioneering laboratories in this field. Dr. Younès informed the participants that UNESCO had initiated a short term (3-6 month) training programme in aquatic biotechnolod and several fellowships were available. The application forms for these fellowships were later distributed.

Several speakers expressed some doubt whether consumers would accept farmed transgenic fish. A marketing strategy will be crucial to overcome the perception of genetically engineered products. The issue of ecological impact of transgenic fish was raised. Many speakers felt that physical containment alone might .net be adequate and additional safety precautions such as biological containment should be included. Sterilization of the transgenic fish and other novel approaches should be explored and developed to be part of a package for successful farming. Also, it might be appropriate to consider international cooperation in forrnulating regulation to control the accidental release of transgenic aquatic animals to nature. Perhaps the RBA programme should start to consider these issues of an interaction of the scientific endeavour and the social interest.

Finally, all the participants thanked Our Japanese hosts for their hospitaiity and the many exotic seafood feasts.

Sayonara, Toba. B iology International is the News Magazine of the

BIOLOGTCAL SCIENCES President: F. di CASTRI (Italy) Past-President: J. SALANKI (Hungary) Vice-Presidents: T. OKADA (Japan) Secretary General: D. STONE (U.S.A.) V.SOKOLOV (Russia) Treasurer: D.F. ROBERTS (U.K.)

Executive Director: T. YOUNES

NBS SECRETARIAT 51. Boulevard de Montmorency, 75016 Paris. France. Tel: 33 (1)45.25.00.09 - Telefax: 33 (1) 45.25.20.29 Telex: c/o ICSU 645 554 F fieInternational Union of Biological Sciences is a non-governmental,non-profit organization, established in 1919. Its objectives are to promote the study of biological sciences, to initiate, facilitate, and coordinate research and other scientific activities that require international cooperation, to ensure the discussion and dissemination of the results of cooperaaive research, to promote the organization of international conferences and to assist in the publication of their reports.

The membership of the IUBS presently consists of 43 Ordiiary Members, adhering through Academies of Science, National Research Councils, national science associations or similar organizations, and of 75 Scientific Members, al1 of which are international scientific associations. societies or commissions in the various biological disciplines. 7National Adhering OrganizationsI ARGENTINA- Consejo Nacional de Investigaciones ITALY- Consiglio Nazionale delle Ricerche Cientificas y Technicas JAPAN- Science Council of Japan AUSTRALIA- Australian Academy of Science JORDAN-Yarmuk University, Irbid. Jordan AUSTRIA- Osterreichische Akademie der Wissenschaften MEXICO-Consejo Nacional de Ciencia y Tecnologia BEUJIUM- Royal Academy of Science, Lettas & Arts MONACO- Centre Scientifique de Monaco BRAZIL- Conselho Nacional de Desenvolvimento Cientico NETHWLANDS- Koninkiijke Nederlandse Akademie van e Tecnologico Wetenschappen BULGARIA- Bulgarian Academy of Sciences NEW ZEALAND- Royal Society of New Zealand CANADA- National Research Councii NORWAY- Det Norske Videnskaps Academi CFIILE- Sociedad de Biologia de Chile PHILIPPINES- Natioanl Science Society CHINA- Association for Science and Technology, Beijing POLAND- Academy of Sciences CHINA- Academia Sinica. Taipei ROMAMA-Academy of Sciences CUBA- Academia de Ciencias RUSSIA- Russian Academy of Sciences CZECHOSMVAKIA- Czechoslovak Academy of Sciences SAUDI ARABIA- KigAbdulaziz City for Science & DENMARK- Det Kongelige Dansk Videnskabemes Selskab Technology EGYPT- Academy of Scientific Research and Technology SOUTH AFRICA- Foundation for Research Development FINLAND- Delegation of Finnish Academies of Science SPAIN- Comision Intenninisterial de Ciencia y Tenologia & Lette.rs SWEDEN- Kunglige Vetenskapsakademien FRANCE- Académie des Sciences SWïiZERLAND- Swiss Academy of Sciences GERMANY - Deutsche Forschungsgemeineschaft UNITED KINGDOM- Royal Society HUNGARY- Academy of Sciences U.S.A.- National Academy of Sciences- National INDIA- Indian National Science Academy Research Council IRAQ- Scientific Research Councii VENEZUELA- Consejo Nacional de Investigaciones IRELAND- Royal Irish Academy Cientificas y Tecnologicas ISRAEL-Academy of Sciences and Humanities WGOSLAVIA- Union of Biological Sciences