analysis of spermatogenesis using an eel model

Aqua-BioScience Monographs, Vol. 4, No. 4, pp. 105–129 (2011) www.terrapub.co.jp/onlinemonographs/absm/

Received on March 18, 2011Accepted on August 25, 2011Online published on

December 26, 2011

Keywords• fish• teleost• testis• Germ cell• in vitro culture• meiosis• gene transfer• androgen• oogenesis• reactive oxygen species

© 2011 TERRAPUB, Tokyo. All rights reserved.doi:10.5047/absm.2011.00404.0105

2. Spermatogenesis

Spermatogenesis, the formation of sperm, is a com-plex developmental process that begins with the mi-totic proliferation of spermatogonia and proceedsthrough extensive morphological changes that convertthe haploid spermatid into a mature, functional sper-matozoon. Morphologically and physiologically, theprocess of spermatogenesis can be divided into thefollowing stages: proliferation of spermatogonia, twomeiotic divisions, spermiogenesis, spermiation andsperm maturation.

2-1. Proliferation of spermatogonia

Spermatogenesis starts with the mitotic proliferationof type A spermatogonia, which are spermatogeneticstem cells. The type A spermatogonium is a relativelylarge cell (approximately 10 µm in diameter in vari-ous species) and has a clear, large homogeneous nu-cleus containing one or two nucleoli. Type A sperma-

Analysis of Spermatogenesis Using an EelModel

Chiemi Miura and Takeshi Miura*

Research Group for Reproductive Physiology, South Ehime Fisheries Research CenterEhime University1289-1, Funakoshi, Ainan, Ehime 798-4292, Japane-mail: [email protected]

AbstractSpermatogenesis is an indispensable process for the continuity of life. The process ofspermatogenesis is very complex; it begins with spermatogonial renewal, then proceedsto proliferation of spermatogonia towards meiosis, two meiotic reduction divisions andspermiogenesis, during which the haploid spermatid develops into a spermatozoa. Afterspermiogenesis, non-functional sperm pass the process of sperm maturation and thenbecome mature spermatozoa, fully capable of vigorous motility and fertilization. Theseprocesses are mainly controlled by sex steroid hormones. Spermatogonial renewal is con-trolled by estrogen; estradiol-17β (E2) through the expression of platelet-derived en-dothelial cell growth factor (PD-ECGF). The proliferation of spermatogonia toward meio-sis is initiated by androgen; 11-ketotestosterone (11-KT) produced by FSH stimulation.11-KT prevents the expression of anti-Müllerian hormone (AMH), which functions toinhibit proliferation of spermatogonia and induce expression of activin B, which func-tions in the induction of spermatogonial proliferation. Meiosis is induced by progestin;17α,20β-dihydroxy-4-pregnen-3-one (DHP) through the action of trypsin. DHP also regu-lates the sperm maturation through the regulation of seminal plasma pH.

1. Introduction

Germ cells provide the continuity of life betweengenerations. In many animals (Gilbert 1985), there isan established germ line that separates from the so-matic cells early in the developmental stage, these germcells migrate into the future gonads from other placesthrough the embryonic tissues. In the developing go-nads, germ cells will become either oogonia in ovariesor spermatogonia in testes. In the gonads they will beexposed to various hormones and cellular interactionsfollowing which, gametogenesis starts. Gametogenesis,spermatogenesis and oogenesis is a process by whichdiploid or haploid precursor cells undergo cell divi-sion and differentiation to form mature haploid gam-etes. These gametes; ovum and spermatozoon, are fer-tilized and then a new generation is started. We havebeen investigating spermatogenesis using fish modelsfor 20 years. In this monograph, we describe the proc-ess and control mechanisms of spermatogenesis basedon our investigations.

106 C. Miura and T. Miura / Aqua-BioSci. Monogr. 4: 105–129, 2011

doi:10.5047/absm.2011.00404.0105 © 2011 TERRAPUB, Tokyo. All rights reserved.

togonia occur independently, with each cell almostcompletely surrounded by Sertoli cells. In some spe-cies, early type B spermatogonia can be distinguishedfrom type A spermatogonia (Schulz et al. 2010). Al-though early type B spermatogonia are morphologi-cally similar to type A spermatogonia, they tend to forma cyst of two or four germ cells surrounded by Sertolicells. It is unclear whether early type B spermatogoniarepresent a renewal of the stem cells or a furtherprogress into spermatogenesis. In the Japanese eel(Anguilla japonica), type A and early type B sperma-togonia are undeveloped and resting spermatogonia(Miura et al. 1991c). These types of spermatogoniaproliferate rapidly by mitosis and, as a result, appearin the seminiferous lobules or tubules. The morphol-ogy of late type B spermatogonia differs from that oftheir earlier undeveloped spermatogonial counterpartsby the fact that their nucleus is denser and more het-erogeneous and their mitochondria are smaller. Afterthe proliferation of spermatogonia, the germ cells en-ter meiosis (Miura et al. 1991c).

The mitotic divisions of spermatogonial stem cellspreceding meiosis are species-specific. In teleosts, aspermatogonial stem cell of medaka (Oryzias latipes)will yield spermatocytes following 9 or 10 mitotic di-visions (Ando et al. 2000), and those of the zebra fish

(Danio rerio) after 5 or 6 (Ewing 1972) and the Japa-nese eel after 10 divisions (Miura et al. 1997). How-ever, it is not clear whether the number of mitotic di-visions is an inherent property of the type A spermato-gonial stem cell, environmentally controlled or both.

2-2. Meiosis in fish spermatogenesis

Following mitotic proliferation, type B spermatogo-nia differentiate into primary spermatocytes. Thesecells enter the first meiotic prophase and then proceedwith the first meiotic division to produce secondaryspermatocytes. These, in turn, undergo a second mei-otic division to produce haploid spermatids, cells withonly a single set of chromosomes.

In primary spermatocytes, the leptotene stages aredistinguished from the final spermatogonia by theirlarger and more homogeneous nuclei. In some species(Schulz et al. 2010), however, it is difficult to distin-guish the leptotene spermatocytes from late type Bspermatogonia, due to their morphological similarities.During the zygotene stage of prophase, spermatocytescan be identified by locating the synaptonemal com-plex in their nuclei using an electron microscope. Be-cause it is very short, the secondary spermatocyte isdifficult to observe. After two meiotic divisions, the

Fig. 1. Induction of spermatogenesis by hCG injection into the Japanese eel. Under culture conditions, male Japanese eelshave immature testes (small arrows) containing only non-proliferated type A and early type B spermatogonia. A single injec-tion of human chorionic gonadotropin (hCG) can induce testicular development (arrowheads) for 18 days. These eels have alot of spermatozoa in their testes. Electron micrographs are reprinted with permission from Zoological Science, 8, Miura etal., Induction of spermatogenesis in male Japanese eel, Anguilla japonica, by a single injection of human chorionic gonado-tropin, 63–73, Fig. 2, 1991, Zoological Society of Japan.

C. Miura and T. Miura / Aqua-BioSci. Monogr. 4: 105–129, 2011 107


germ cells develop into spermatids having small, roundand heterogeneous nuclei.

2-3. Spermiogenesis

During spermatogenesis, the round spermatids trans-form into spermatozoa. This process is characterizedby remarkable morphological changes associated withthe formation of the sperm head and its condensednucleus, with the midpiece, and with the flagellum. Thestructure of spermatozoa varies considerably in com-plexity among teleost species (Miura 1999). For ex-ample, carp (Cypinus carpio), sculpin (Aottushangiongenesis), and tilapia (Oreochromis niloticus)spermatozoa have spherical heads with a flagellum at-tached to one side. Salmonid spermatozoa have aslightly elongated and cylinder-like head. On the otherhand, the spermatozoa of guppy have an elongated andan extremely well-developed midpiece. Furthermore,eel spermatozoa possess a crescent-shaped nucleus,their flagellum has a 9 + 0 axonemal structure (gener-ally, the axonemal structure of the flagellum is 9 + 2)and a single large and spherical mitochondrion withdeveloped tubular cristae is attached to the caput endat one side of the head. An acrosome is absent in thespermatozoa of most teleosts but it is found inacipenserid fish, lamprey and shark.

2-4. Spermiation

In mammals, “spermiation” indicates that embeddedbundles of spermatozoa are released from the envel-

oping Sertoli cell and are swept into the efferent duct.In most teleosts (except in Poeciliidae), however, sper-matozoa are not associated with Sertoli cells. By com-parison, spermiation of teleosts indicates the releaseof spermatozoa from the seminal cysts into the lobularlumen or efferent duct. From the point of view of fish-eries science, however, “sperm release” or “ejacula-tion,” which occurs after milt hydration and spermmigration down the sperm duct, is more readily ob-served than spermiation. Therefore, the term spermia-tion is often used interchangeably with these otherterms in fish.

2-5. Sperm Maturation

Although the spermatozoa in the testis have alreadycompleted spermatogenesis, in some species they arestill incapable of fertilizing eggs. In salmonids, thespermatozoa in the testis and sperm duct are immotile.If sperm from the sperm duct are diluted with freshwater, they become motile, whereas testicular sperm,if diluted with fresh water, remain immotile. Thus,spermatozoa acquire their ability to become motileduring their passage through the sperm duct.

The acquisition of the motile ability of sperm is dif-ferent from the initiation of motility. The developmentfrom nonfunctional gametes to mature spermatozoafully capable of vigorous motility and fertilization isreferred to as “sperm maturation”. Sperm maturationinvolves physiological, not morphological, changes. Insalmonids, sperm maturation (acquisition of spermmotility) is induced by the high pH of the seminal

Fig. 2. Timetable of spermatogenesis of hCG injected eel testis. Germ cell development is almost synchronous throughout thetestis and the proliferation of spermatogonia, meiosis and spermiogenesis occur at definite times: 3, 12 and 15 days after hCGinjection, respectively. Reprinted with permission from Reproduction, 142, Miura et al., Gh is produced by the testis ofJapanese eel and stimulates proliferation of spermatogonia, 869–877, Fig. 1, 2011, Society for Reproduction and Fertility,and reprinted with permission from Kaiyo to Seibutsu, 24, Miura and Miura, The challenge of artificially producing spermand egg from immature gametes in vitro, 114–119, Fig. 1, 2002, Seibutsukenkyusha.



plasma (approximately pH 8.0) in the sperm duct andinvolves the elevation of intrasperm cAMP levels.Maturation of eel spermatozoa is also induced by thehigh pH of the seminal plasma and/or HCO3

–.This spermatogenesis is controlled by numerous hor-

mones and unknown factors (Steinberger 1971;Hansson et al. 1976; Callard et al. 1978; Billard et al.1982; Cooke et al. 1998). Specifically, sex steroid hor-mones, estrogen, androgen and progestin play signifi-cant roles in the control of spermatogenesis (Miura1999).

3. Regulation of spermatogenesis in eel in vivoor in vitro

Among species of teleosts, there are various repro-ductive styles and gametogenetic patterns. Teleostsconstitute the largest group (approximately 23,700 spe-cies) of living vertebrates (~48,200 species) (Nelson,1994). The Japanese eel, Anguilla japonica, one of suchspecies, has a special spermatogenetic pattern. In theJapanese eel, insufficient gonadotropin in the pituitaryis attributed to the immature testes containing only non-

proliferated type A and early type B spermatogoniaunder culture conditions (Nagahama and Yamamoto1973). However, a single injection of human chorionicgonadotropin (hCG) can induce all stages of sperma-togenesis from the proliferation of spermatogonia tospermiogenesis in vivo (Miura et al. 1991a) and thisinduction is achieved via gonadotropin stimulation ofLeydig cells to produce 11-ketotestosterone (11-KT)(Miura et al. 1991b) (Fig. 1). Germ cell developmentis almost synchronous throughout the testis and theproliferation of spermatogonia, meiosis and spermio-genesis occur at definite times: 3, 12 and 15 days afterhCG injection, respectively (Miura et al. 1991c) (Fig.2).

Furthermore, the Japanese eel is the only animal inwhich complete spermatogenesis has been induced byhormonal treatment in vitro using an organ culture sys-tem (Fig. 3) (Miura et al. 1991a), and a germ cell/so-matic cell co-culture system (Fig. 4) (Miura et al.1996). Therefore, these eel culture systems could bethe best system for analysis of the control mechanismsof spermatogenesis because their direct action on sper-matogenesis can be investigated (Fig. 5). Furthermore,

Fig. 3. The eel testicular organ culture system. Freshly removed immature eel testes were cut into small pieces, which wereplaced on floats of elder pith covered with a nitrocellulose membrane in 24-well plastic tissue-culture dishes (upper left). Byusing this system, 11-KT can induce the entire process of spermatogenesis for 36 days. Each symbol indicates: GA, type Aspermatogonia; GB, type B spermatogonia; SC, spermatocytes; ST, spermatid; SZ, spermatozoa. Bar, 10 µm. Reprinted withpermission from Handbook of Animal Cell Technology (Edited by Japanese Association for Animal Cell Technology), Miuraand Miura, 287–289, Fig. 15.6, 2000, Asakura Publishing Co., Ltd., reprinted with permission from Kaiyo to Seibutsu, 24,Miura and Miura, The challenge of artificially producing sperm and egg from immature gametes in vitro, 114–119, Fig. 2, 2002, Seibutsukenkyusha, and reprinted with permission from Zoological Science, 18, Miura and Miura, Japanese eel: amodel for analysis of spermatogenesis, 1055–1063, Fig. 2, 2001, Zoological Society of Japan.



to analyze functions of genes encoding eel sperma-togenesis related substances (eSRSs), we have devel-oped a new assay system using gene transfer techniquescombined with co-culture of the eel germ-somatic cells.The electroporation method provides a good tool inthe search for factors regulating spermatogenesis (Fig.6). Thus, the eel testis provides an excellent systemfor studying the regulation of spermatogenesis. Usingthe culture system, we analyzed the control mecha-nisms of gametogenesis.

4. The endocrine control of fish spermatogenesis

4-1. Gonadotropins and spermatogenesis

Eel spermatogenesis is also endocrinologically con-trolled, in the same way as in other vertebrates. It iswell established that in vertebrates, gonadotropins(GTHs), follicle-stimulating hormone (FSH) and lutei-nizing hormone (LH) are the primary hormones regu-lating spermatogenesis (Nagahama 1987). FSH and LHare members of the pituitary glycoprotein family, in-cluding thyroid-stimulating hormone. These hormonesare heterodimers, each consisting of a common α and

a hormone-specific β subunit (Pierce and Parsons1981). In mammals, LH and FSH have different rolesin spermatogenesis, respectively; LH regulates sex ster-oid production in Leydig cells and FSH regulates Ser-toli cell activities, such as the structural, nutritionaland regulatory support of germ cell development(Huhtaniemi and Themmen 2005). On the other hand,in fish it has also been established that two types ofGTHs, FSH and LH, exist (Swanson et al. 1991; Vander Kraak et al. 1992; Planas and Swanson 1995;Yoshiura et al. 1999). However, the definitive func-tion of each GTH has not been established. Recently,in the development of molecular biological techniques,it has become possible to analyze the differences inthe roles of FSH and LH in fish (Kamei et al. 2006;Kazeto et al. 2008; Hayakawa et al. 2008). In somesalmonids, it has been reported that FSH but not LH issecreted from the pituitary of immature fish, while LHrelease is higher during the period of sperm matura-tion (Swanson et al. 1989; Prat et al. 1996). In addi-tion, it seems that in coho salmon, FSH acts at earlystages of spermatogenesis because FSH is able tostimulate steroid hormone production, similarly to LH.However, FSH-stimulated production of steroid hor-

Fig. 4. The schema of the method of germ-somatic cells coculture system. Immature eel testes were enzymatically dissociatedand the cell suspension was filtered through meshes and centrifuged in Nycodenz gradients. Separated cell suspension wascentrifuged to make pellets and they were cultured with or without 11-KT. After 30 days culture, many spermatozoa (whitearrows) having one or two flagella were observed around the pellet of germ cells and somatic cells. Reprinted with permissionfrom Kaiyo to Seibutsu, 24, Miura and Miura, The challenge of artificially producing sperm and egg from immature gametesin vitro, 114–119, Fig. 4, 2002, Seibutsukenkyusha.



mones decreases towards the period of sperm matura-tion (Planas and Swanson 1995). Furthermore, in theJapanese eel, FSH may act on early stages of sperma-togenesis, considering that the fshβ subunit mRNA isexpressed in the pituitary of immature fish while the

lhβ subunit mRNA is not expressed until much later inthe period of sperm release (Yoshiura et al. 1999).Because of the differences in their expression pattern,the roles of FSH and LH in spermatogenesis were esti-mated in fish (Schulz et al. 2010).

Fig. 5. Using the organ culture system, germ-somatic cells coculture (pellet culture) system and cell culture system, we caninvestigate unknown factors directly added to the medium.

Fig. 6. Expression of GFP gene in germ-somatic cell pellets after electroporation. Transient expression of transfected geneswas examined two days after the electroporation of GFP cDNA into germ-somatic cell pellets. A train of eight square pulses(60 V; duration 50 msec; interval 950 msec), resulted in widespread expression of GFP fluorescence in many round germ cellsand somatic cells. CMV, cytomegarovirus promoter; GFP, green fluorescent protein. Reprinted with permission of John Wiley& Sons, Inc. from Molecular Reproduction and Development, 74, Miura et al., Transfer of spermatogenesis-related cDNAsinto eel testis germ-somatic cell coculture pellets by electroporation: methods for analysis of gene function, 420–427, Fig. 1, 2007, Wiley-Liss, Inc., a Wiley Company.



As mentioned above, in vitro systems provide directevidence of hormonal action in biological events.Therefore, we investigated and resolved the roles ofFSH in early spermatogenesis using eel testicular cul-ture systems (Ohta et al. 2007). To understand the FSHfunction in spermatogenesis, it is necessary to knowthe expression pattern of its receptor. In western blotanalysis using purified plasma membrane fraction ofimmature eel testis, eel FSH receptor (eFshr) exhib-ited two forms, each with different molecular mass:one of 41 kDa and another 72 kDa. Using extractedprotein from whole testis, however, eFshr exhibitedonly a band of 41 kDa. This result suggests that the 72kDa form is full length eel Fshr from the deduced aminoacid sequence of efshr cDNA and the 41 kDa form isextracellular domain of eFshr, whose full-lengthreceptor was cut during the process of extractingplasma membrane or testicular proteins. Moreover, toevaluate how expression of eFshr protein changes dur-ing spermatogenesis, we performed western blot analy-sis on hCG-treated eel testis. Eel Fshr was expressedbefore the initiation of spermatogenesis and continu-ously expressed during all stages. It is therefore possi-ble that FSH acts on all stages of spermatogenesis (Ohtaet al. 2007).

To determine the distribution of Fshr in the testis,we performed immunohistochemistry using an anti-eFshr antibody. The antibodies stained Leydig cells,which produce steroid hormones and Sertoli cells sur-rounding type A or early type B spermatogonia duringspermatogenesis (Ohta et al. 2007). In some teleosts,FSH protein increases at an early stage of spermatogen-esis (Swanson et al. 1989; Planas and Swanson 1995;Prat et al. 1996), and in eel, fshβ subunit mRNA isexpressed in the pituitary of immature fish (Yoshiuraet al. 1999). These results suggest that FSH acts onearly stages of spermatogenesis via Leydig and/or Ser-toli cells.

To understand whether FSH acts on spermatogen-esis, we investigated the effects of FSH on in vitro sper-matogenesis using recombinant eel FSH (r-eFSH) pro-duced from a yeast expression system. Adding r-eFSHto the culture medium induced the complete processof spermatogenesis from the proliferation of sperma-togonia to spermiogenesis. In the Japanese eel, it hasbeen reported that FSH induces the secretion of 11-KT in immature testis (Kamei et al. 2005). Therefore,it is possible that the role of FSH is to induce 11-KTsecretion, which in turn will stimulate spermatogen-esis.

Using trilostane that specifically inhibits 3β-HSDactivity, we investigated whether FSH acts on sperma-togenesis via the production and secretion of 11-KT intesticular organ culture. Adding r-eFSH and trilostaneto the culture medium reduced the percentage of cystsof late type B spermatogonia compared to treatmentwith only r-eFSH and the progress of spermatogenesis

was inhibited. These results indicate that FSH is re-lated to the regulation of spermatogenesis by trigger-ing the secretion of 11-KT (Ohta et al. 2007).

In males, androgens including 11-KT are synthesizedby Leydig cells in the testis (Payne and Youngblood1995; Dufau et al. 1997). In coho salmon, Fshr is lo-calized to Sertoli cells at all stages of spermatogen-esis, while Lhr was only found in Leydig cells inspermiating fish (Miwa et al. 1994). Nevertheless, FSHpromoted the synthesis of androgen in immature andmature testis similar to LH (Swanson et al. 1989; Planasand Swanson 1995). It is therefore possible that Ley-dig cells express Fshr or that paracrine factors secretedby Sertoli cells upon FSH stimulation promote Leydigcell’s androgen production (Lejeune et al. 1996). Fshris expressed in Leydig and Sertoli cells surroundingtype A and early type B spermatogonia in the Japaneseeel (Ohta et al. 2007). This suggests that FSH directlyacts on Leydig cells via Fshr activation and promotesthe synthesis of 11-KT.

Although Fshr localizes to Sertoli cells from fishesto mammals, including the Japanese eel, the clear func-tions of FSH in Sertoli cells via Fshr activation havenot been established. In mice, the absence of functionalfollicle-stimulating hormone beta-subunit (Fshbeta) orFshr genes leads to reduced testis size but the malesare still fertile (Kumar et al. 1997; Dierich et al. 1998;Abel et al. 2000; Krishnamurthy et al. 2000). More-over, in the Japanese eel, all stages of spermatogen-esis are induced by 11-KT alone in vitro (Miura et al.1991a; Miura et al. 1996). This suggests that FSH alsosupports testicular development and maintenance ofsperm production through the action of Sertoli cells.However, the functions of FSH via Sertoli cells arenot clear.

Thus, FSH stimulates the Leydig cells and regulatesspermatogenesis via the production and secretion ofsteroid hormones.

4-2. The regulation of spermatogonial stem cellrenewal

As mentioned above, the first step of spermatogen-esis is spermatogonial mitotic proliferation. Sperma-togonial mitosis can be categorized in slow spermato-gonial renewal and rapid proliferation of differentiatedspermatogonia toward meiosis (Clermont 1972). Bothkinds of spermatogonial mitosis are regulated by dif-ferent mechanisms by steroid hormones. We indicatedthat spermatogonial renewal is regulated by estrogen,estradiol-17β and spermatogonial proliferation towardmeiosis is regulated by androgen, 11-KT in fish (MiuraT et al. 1999).

It is widely accepted that estrogen is a female hor-mone in all animals. However, it has been reported thatestrogen exists in some male vertebrates (Schlinger andArnold 1992; Fasano and Pieratoni 1993; Betka and



Callard 1998), and that its receptors are expressed inthe male reproductive organs (Callard and Callard1987; Ciocca and Roig 1995). Estradiol-17β (E2), anatural estrogen in vertebrates, was found in Japaneseeel serum and its receptor was expressed in the Sertolicells (the only non-germinal elements within the sem-iniferous epithelium of the testes) during the wholeprocess of spermatogenesis. Thus, estrogen and itsreceptor are present in the eel testes. Using eel in vivoand in vitro experimental systems, we investigated therelationship between spermatogenesis and E2 (MiuraT et al. 1999). E2 and tamoxifen (an antagonist ofestrogen) were given intraperitoneally to eels using asilastic capsule. After 24 days of implantation, the fishwere sacrificed and their spermatogenesis wasanalyzed. E2 implantation significantly increased andtamoxifen implantation significantly decreased germcell DNA synthesis compared with control. The effectof E2 on spermatogenesis was confirmed by in vitroexperiment; E2 treatment induced DNA synthesis andmitotic division in germ cells in in vitro testicular or-gan culture. Even though E2 treatment in vivo and in

vitro induced spermatogonial mitosis, the germ cellsdid not progress further into meiosis. As mentionedabove, spermatogonial mitosis can be categorized byspermatogonial renewal and spermatogonial prolifera-tion toward meiosis. FSH or 11-KT induce spermato-gonial proliferation and the further stage of sperma-togenesis and finally produced spermatozoa in vitro.Therefore, spermatogonial mitosis induced by E2 maynot be directed toward the formation of spermatozoabut for spermatogonial renewal. In Japanese huchen(Hucho perryi), E2 also promoted spermatogonial re-newal in vitro (Amer et al. 2001). These findingsclearly indicate that estrogen is an indispensable “malehormone”, and plays an important role in spermatogo-nial renewal.

In in vitro testicular organ culture, supplementationof 10, 100 and 1000 pg/ml of E2 to the culture me-dium stimulated DNA replication and mitosis of theprimary stage of spermatogonia. The range of the ef-fective E2 concentration, 10–1000 pg/ml, conforms tothe levels found in the male serum. This shows that E2is effective at much lower concentrations than 11-KT,

Fig. 7. A schematic diagram summarizing the possible control mechanisms of spermatogenesis in the Japanese eel. FSH,follicle-stimulating hormone; LH, luteinizing hormone; 17α,20β-DHP, 17α,20β-dihydroxy-4-pregnen-3-one; PD-ECGF,platelet-derived endothelial cell growth factor; AMH, anti-Müllerian hormone; CAll, carbonic anhydrase; eSRS, sperma-togenesis related substances.



since 10 ng/ml of 11-KT are needed to induce full sper-matogenesis (Miura et al. 1991a). Thus it is indicatedthat in the Japanese eel, low concentrations of E2 (10pg/ml) act on the primary stages of spermatogoniathrough Sertoli cells, by stimulating and maintainingtheir proliferation prior to the progress of a further stageof spermatogenesis.

Generally, E2 induces the target gene expressionthrough its receptor and the factor translated from thisgene affects the biological process. We used cDNAcloning to identify a factor that regulates spermatogo-nial renewal after estrogen stimulation and subse-quently clarified its functions. To identify factors thatare regulated by E2 stimulation, we carried out geneexpression cloning in the Japanese eel (Miura et al.2003). As a result of this experiment, we found onepreviously unidentified cDNA clone that was up-regulated by E2 stimulation and named it eelspermatogenesis-related substance (eSRS) 34 cDNA.The transcription and translation of eSRS34 were de-tected in testes at every experimental stage; such ex-pression parallels spermatogonial renewal, which oc-curs continuously throughout spermatogenesis (Miuraet al. 2003). E2 constantly exists in serum during sper-matogenesis (Miura T et al. 1999). Furthermore,eSRS34 protein was expressed in Sertoli cells, in whichthe receptor for E2 is expressed. Taken together, thesefindings suggest that eSRS34 fulfills the criteria de-fined for a key factor regulating spermatogonial re-newal regulated by E2. A homology search of the pre-dicted amino acid sequence showed that eSRS34 sharescomparatively high similarity with platelet-derived en-dothelial growth factor (PD-ECGF). Furthermore,eSRS34 contains seven conserved cysteine residues aswell as the recognition motif for thymidine and pyri-midine nucleotide phosphorylase, which are key fea-tures of PD-ECGF (Miyazono et al. 1991; Furukawaet al. 1992). The function of eSRS34 was examinedusing eel in vitro systems. Recombinant eSRS34 pro-duced by Baculovirus system induced spermatogonialrenewal in testicular organ culture. Furthermore, theaddition of a specific anti-eSRS34 antibody preventedonly spermatogonial renewal induced by E2 stimula-tion in a germ cells/somatic cells co-culture system(Miura et al. 2003). These results indicate that eSRS34is a “spermatogonial renewal factor” in fish (Fig. 7).

5. The regulation of spermatogonial proliferationtoward meiosis

5-1. The role of 11-KT in spermatogenesis

The second step of spermatogenesis; spermatogonialproliferation toward meiosis, is initiated by the secre-tion of FSH and the main action of FSH on sperma-togenesis is the production and secretion of 11-KT, viastimulation of the Leydig cells, as mentioned above.

11-KT was first identified by Idler et al. (Idler et al.1961) as a major androgenic steroid in the male sockeyesalmon (Oncorhynchus nerka). In various teleost fishes,11-KT has been shown to be synthesized in the testisfollowing GTH stimulation and high levels were de-tected in the serum during spermatogenesis (Billard etal. 1982). As mentioned above, under aquaculture con-ditions, the male Japanese eel has immature testes con-taining only type A and early type B spermatogonia;primary spermatogonia, which are premitotic. A sin-gle injection of hCG can induce all the stages of eelspermatogenesis in vivo. This injection also causes anincrease in serum levels of 11-KT (Miura et al. 1991c).

Although numerous in vivo studies (Sakai 2002;Hong et al. 2004) have suggested the important role ofandrogens in vertebrate spermatogenesis (Steinberger1971; Callard et al. 1978; Billard et al. 1982), therehave been, to our knowledge, no in vitro studies to di-rectly show the involvement of androgens in this proc-ess. A testicular organ culture system has been devel-oped using eel testes, which have only undevelopedspermatogonia. Organ cultures provide a simplifiedexperimental system in which the direct effects of vari-ous factors upon the testes can be investigated. Usingan eel testicular organ culture system which we havenewly developed, we investigated the role of 11-KTon spermatogenesis (Miura et al. 1991a).

Testes removed from eels were cultured for 15 daysin a medium with or without various concentrations of11-KT (0.01, 0.1, 1, 10 and 100 ng/ml). The appear-ance of proliferated spermatogonia (late-type B sper-matogonia) in cysts was used as the criterion for mito-sis. A supplement of 11-KT into the culture mediumwas effective for the initiation of spermatogenesis.Concentrations of 10 and 100 ng/ml were almostequally effective; mitosis occurred in 50–60% of cysts.The concentration of 10 ng/ml corresponds to that inthe plasma of maturing male eels receiving a singleinjection of hCG (Miura et al. 1991c). In contrast, thelower two concentrations had no effect.

11-KT is the most effective androgen for the initia-tion of spermatogenesis. Seven different androgens (11-KT, 11β-hydroxytestosterone (11β-HT), testosterone(T), 5β-dihidrotestosterone, dehidroepiandrosterone,androsterone and androstendione) were investigated fortheir ability to induce the proliferation of spermatogo-nia in vitro. Testicular fragments were cultured in amedium containing one of these steroids at a dose of10 ng ml for 15 days. The active mitosis occurredwithin the cultivated testes only when 11-KT was addedto the medium. Although a slight stimulation was ob-served with 11β-HT and T, this may have been fromthe conversion of these steroids to 11-KT by endog-enous enzymes (Miura et al. 1991a). 11-KT can in-duce all stages of spermatogenesis, from spermatogo-nial proliferation to spermiogenesis. Sequentialchanges in germ cells were investigated in cultures for



periods up to 36 days in the presence of 11-KT at 10ng/ml. Nine days after the start of culture, spermato-gonia began mitotic proliferation. Zygotenespermatocytes of the meiotic prophase occurred in tes-ticular fragments cultured for 18 days. Afterwards,spermatids and spermatozoa were initially observed.After 36 days, all stages of germ cells, including 8.2%of spermatozoa, were present. The action of 11-KT forspermatogenesis is not limited to the Japanese eel; ithas been also recognized in goldfish (Carassiusauratus) (Kobayashi et al. 1991) and Japanese huchen(Hucho perryi) (Amer et al. 2001).

5-2. IGF-1 and spermatogenesis

As mentioned above, 11-KT is an inducing steroidof spermatogenesis in fish. However, it is believed thatthe action of 11-KT is mediated by other factors pro-duced by Sertoli cells, in which the androgen receptorexists (Ikeuchi et al. 2001). It is possible that some ofthese factors are growth factors, such as insulin-likegrowth factor-I (IGF-I). IGFs are known to be media-tors of growth hormone action in vertebrates. In therainbow trout (Oncorhynchus mykiss) testis, IGF-I isexpressed in spermatogonia and/or Sertoli cells and itbinds to type 1 IGF receptors (LeGac et al. 1996). Fur-thermore, IGF-I stimulates DNA synthesis in sperma-togonia (Loir 1994; Loir and LeGac 1994; LeGac etal. 1996). Although IGF-I is also necessary for the regu-lation of eel spermatogenesis, its role is to support theaction of 11-KT. More specifically, in the Japanese eel,11-KT is necessary for the induction of spermatogen-esis, whereas IGF-I is necessary for the continuationof the process (Nader et al. 1999).

5-3. The regulation of initiation of spermatogen-esis by two growth factors; AMH and activinB

How does 11-KT initiate spermatogonial prolifera-tion in fish? In the Japanese eel, two members of thetransforming growth factor-like (TGF-β) super fam-ily, anti-Müllerian hormone (AMH) (Miura et al. 2002)and activin B (Miura et al. 1995a), have important rolesduring the initiation of spermatogenesis induced by 11-KT.

Activin B is a dimeric growth factor belonging tothe TGF-β super family, and is composed of twoactivin-βB subunits. In the Japanese eel, activin B wasfound in the testis at the initiation of spermatogenesisafter hCG stimulation, with its expression site restrictedto Sertoli cells. Both transcription and translation ofeel activin B were induced by 11-KT stimulation invitro. Furthermore, activin B induced proliferation ofspermatogonia but its treatment could not induce meio-sis and further spermatogenesis (Miura et al. 1995a,b). It has been reported that these activin B, IGF in-

cluding IGF-binding protein and numerous othergrowth factors regulate the early stage of spermatogen-esis in teleosts and mammals (Watanabe and Onitake1995; Zhao et al. 1996; Li et al. 1997; Kim et al. 1998).Further investigation is needed to fully understand therelationship between activin B and spermatogenesis infish.

In the Japanese eel, it has become clear that a“spermatogenesis-preventing substance” is present inimmature testis and spermatogenesis is initiated by thesuppression of its expression (Miura et al. 2002). Un-der freshwater cultivation conditions, male Japaneseeels have immature testes containing only non-proliferated spermatogonia (Miura et al. 1991a). It ispossible that factors that suppress the progress of sper-matogenesis are expressed in the testis when the fishare in fresh water. In other words, eel spermatogenesismay be initiated by the downregulation of the genesencoding suppressive factors. On the basis of this hy-pothesis, we used gene expression cloning to isolatecDNA clones that show suppressed expression afterhCG treatment. As a result of this cDNA cloning, wesucceeded in identifying eel spermatogenesis relatedsubstance 21 (eSRS21) cDNA. eSRS21 shares aminoacid sequences similarity with mammalian AMH at ap-proximately 40%. Thus, we called eSRS21 eel AMH(eAMH). eAMH was expressed in Sertoli cells of im-mature testes before the initiation of spermatogenesis,but disappeared after gonadotropin stimulation. Theinitiation of spermatogonial proliferation correspondswith the disappearance of eAMH expression. Expres-sion of eAMH mRNA was also suppressed in vitro by11-KT, which is a spermatogenesis-inducing steroid.To examine the function of eAMH in spermatogenesis,recombinant eAMH produced by a CHO cell expres-sion system was added to a testicular organ culturesystem (Miura et al. 2002). Spermatogonial prolifera-tion induced by 11-KT in vitro was suppressed byrecombinant eAMH. Furthermore, the addition of aspecific anti-eAMH antibody induced spermatogonialproliferation in a germ cell/somatic cell co-culture sys-tem. These indicate that eAMH prevents the initiationof spermatogenesis and therefore, suppression ofeAMH expression is necessary to initiate spermatogen-esis. The discovery of the spermatogenesis preventingsubstance suggests that spermatogonial proliferationtoward meiosis is directly regulated by the rivalry of astimulating factor such as activin B and a preventingfactor, such as eAMH (Fig. 7).

Recently, the medaka’s AMH receptor; amhrII genewas identified, which represents the first characteriza-tion of this receptor in nonmammalian vertebrates(Kluver et al. 2007). In fish, the Wolffian duct is asso-ciated with the pronephros regardless of sex and thesperm duct and oviduct are derived from the coelomicepithelium (Suzuki and Shibata 2004), analogous tothe amniote Müllerian duct. In medaka, amh and amhrII



are expressed in the somatic cells of the developinggonads identically between the sexes during larval andjuvenile stages. These results suggest that the medakaamh and amhrII genes are involved in gonadal devel-opment, including the regulation of germ cells in bothsexes (Morinaga et al. 2007).

As mentioned above, in freshwater conditions eelspermatogenesis arrests at an immature stage prior tothe initiation of spermatogonial proliferation. Thisimmature stage seems to be maintained by the expres-sion of spermatogenesis-preventing substance in eelSertoli cells in freshwater conditions. When eels down-migrate to the ocean, spermatogenesis resumes andprogresses to maturation (Larsen and Dufour 1993).This indicates that during eel migration, spermatogen-esis is resumed through the suppression of AMH,spermatogenesis-preventing substance, expressioncaused by an increase in 11-KT, which is induced bygonadotropin stimulation.

To further elucidate the key genes involved in sper-matogonial proliferation toward meiosis, we performedscreening of stage-specific genes during eel sperma-togenesis using cDNA subtraction and differential dis-play methods (Miura et al. 1998; Miura C et al. 1999).The key genes coding factors that showed unique ex-pression during spermatogenesis were considered. Asa result of these experiments, 28 independent cDNAclones showing unique expression patterns during sper-matogenesis were obtained (Miura and Miura 2001).Among these, 16 clones are up- or down-regulated by11-KT, the spermatogenesis inducing hormone. Theprogression of eel spermatogenesis toward meiosis maybe regulated by some of these factors.

6. Initiation of meiosis

6-1. The entry of spermatogonia into meiosis

In the cultivated male Japanese eel, type A and earlytype B spermatogonia, which are primary cells, are theonly germ cells present in the testis. Exogenous gona-dotropin like hCG injection can induce complete sper-matogenesis, from proliferation of spermatogonia tospermiogenesis. In some cases, however, hCG injec-tion fails to induce complete spermatogenesis (Miuraet al. 1997). Testicular morphological observationsrevealed that hCG-injected eels could be classified intothree types, based on their testicular conditions. In Type1 eels, complete spermatogenesis, from proliferationof spermatogonia to spermiogenesis, was successfullyinduced. In type 2 eels, spermatogenesis was also in-duced by hCG injection but there were no spermato-cyte or spermatids in their testis. Type 3 eels hadthready testis, which did not develop any germ cellsduring the experimental period. These results suggestthat despite elevations of plasma 11-KT levels, hCGinjections were not successful in inducing the comple-

tion of spermatogenesis in type 2 and 3 eels. In mostspermatogonia of type 2 eels, meiosis of 23 to 26 late-type B spermatogonia was not induced in most cysts.Moreover, cysts with 27 or more spermatogonia werenot observed. This suggests that spermatogonial stemcells undergo four or five and occasionally six mitoticdivisions before the interruption of spermatogenesisin type 2 eels. It is proposed that those numbers ofmitotic divisions are related to a mediator that regu-lates the entry of spermatogonia of the Japanese eelinto meiosis (Miura et al. 1997).

Following mitotic proliferation, late-type B sperma-togonia differentiate into primary spermatocytes. Thenumber of spermatogonial generations is geneticallydetermined (Courot et al. 1970; Schulz et al. 2010).For example, 6 generations were found in Sakhalintaimen, 8 in masu salmon (Oncorhynchus masou), 6 inwhite spotted char (Salvelinus leucomaenis), 8 in gold-fish (Ando et al. 2000), 14 in the guppy (Poeciliareticulata) (Billard, 1986) and there are 10 mitotic di-visions in the Japanese eel (Miura et al. 1991a, 1997).Although the regulatory mechanisms for the initiationof meiosis are not yet clear, it has been shown that inthe Japanese eel there is a regulatory stage around thefifth mitotic division of spermatogonia, prior to thecells entering meiosis (Miura et al. 1997).

6-2. The mechanisms of initiation of meiosis inspermatogenesis

Meiosis is a special type of cell division that is re-stricted to germ cells. Meiosis produces haploid cellsand forms the basis of sexual reproduction. Many stud-ies on meiosis are directed toward chromosome dy-namics (Shinohara and Shinohara 2004; Gerton andHawley 2005; Watanabe 2005) or to oocyte matura-tion, which resumes and completes the prophase of thefirst meiotic division (Masui 2001; Thomas et al. 2002).However, the mechanism initiating the first meioticdivision is not clear. Progestins are sex steroid hor-mones that are important for reproduction. In mam-mals, the principal physiological action of progestin isto prepare the reproductive tract for pregnancy and toprovide nutritive support for the embryo during gesta-tion (Burris 1998); however, this is an evolutionarilyrecent function. In all vertebrates, progestin also playsimportant roles in gametogenesis. Progestins regulateoocyte maturation (Nagahama 1997) by binding to anoocyte plasma membrane receptor, inhibiting oocyteadenylate cyclase, followed by reduced cAMP-dependent protein kinase activity, which induces theactivation of maturation promoting factor via Cdc25,eventually triggering the resumption of division I ofmeiosis (Nagahama 1997; Thomas et al. 2002). In fishspermatogenesis, progestin also plays an important rolein spermiation and sperm maturation (Ueda et al. 1985;Miura et al. 1991b, 1992). A major progestin in teleost



fish, 17α,20β-dihydroxy-4-pregnen-3-one (DHP), in-duces sperm hydration (Ueda et al. 1985) and the ac-quisition of sperm motility in some species (Miura etal. 1991b, 1992). A related progestin, 17α,20α-dihydroxy-4-pregnen-3-one is the spermiation-induc-ing hormone in amphibia (Kobayashi et al. 1993). Tho-mas et al.’s (2005, 2006) data on progestin actions onfish gametes suggest the widespread involvement ofmembrane progestin receptor alpha (mPRα) in oocytematuration and sperm hyperactivity in spotted sea trout(Cynoscion nebulosus) and Atlantic croaker(Micropogonias undulatus). Thus, progestin is an in-dispensable hormone for gametogenesis. However,studies on progestins have been directed mainly at func-tions in late maturational stages in both sexes. In sev-eral fish species, DHP is found in blood serum at pu-berty in males (Baynes and Scott 1985; Amer et al.2001). In salmonids, there are two peaks of DHP, bigand small peak, in the blood level. One big peak is inthe spawning season and another small peak is in theprogression of spermatogonial proliferation. The bigpeak of DHP in the spawning season is related to sper-miation and sperm maturation, which will be discussedlater. The small peak of DHP was known in salmonids(Depeche and Sire 1982; Scott and Sumpter 1989) butits role had not yet been clearly described. Further-more, we demonstrated that DHP induced spermato-gonial DNA synthesis in Japanese huchen (Amer et al.2001). These findings suggest that progestin has animportant role not only in final maturation but also in

the early stages of gametogenesis. However, there isno information on the role of DHP in the early stagesof spermatogenesis. We show that DHP action in earlyspermatogenesis became clear using the eel testis tis-sue/cell culture systems (Miura et al. 2006).

To understand the possibility that DHP also acts onthe early stages of spermatogenesis, we quantified tes-ticular DHP in eels during hCG-induced spermatogen-esis and the expression of nuclear types of progester-one receptor (PR) in the eel testis. Treatment with hCGinduced spermatogenesis and a strong increase in tes-ticular DHP levels in vivo. Furthermore, types 1 and 2of nuclear PRs were expressed in the immature testisbefore initiation of spermatogenesis. These findingsopen the possibility that DHP also acts on the regula-tion of early spermatogenesis in the Japanese eel. In-terestingly, as mentioned above, DHP has been detectedin serum during the proliferation of spermatogonia insome salmonids (Baynes and Scott 1985; Amer et al.2001). Based on these findings, DHP may be involvedin regulating early spermatogenesis in salmonids andother teleosts but there is no direct physiological rolefor this steroid. Since our RT-PCR studies showed thatthe types 1 and 2 PR showed partially different cellu-lar sites of expression in the testis, DHP may not havejust one function. We found that DHP stimulated DNAthe replication of spermatogonia in testicular organ andgerm cell/somatic cell co-culture, as did 11-KT. What,however, may be the function of DHP in early sperma-togenesis and how does it differ from that of 11-KT?

Fig. 8. Effect of DHP on induction of meiosis in the testis of Japanese eels in vitro. Electron micrographs showing testicularsection from fragments cultured in basal medium without hormone (control) or with 10 ng/ml DHP for 6 days. Testicularfragment cultured with DHP showed that germ cell nuclei contained synaptonemal complexes (arrowheads) characteristic ofmeiotic cells.



To address these questions we carried out germ cell/somatic cell co-culture, using specific DHP antibodiesfor 6 (short-term culture) and 15 days (long-term cul-ture). During eel spermatogenesis in vitro, spermato-gonial proliferation starts 3 days and meiosis 15 daysafter commencing the culture with 11-KT. Therefore,the 5-bromo-2′-deoxyuridine (BrdU) index of the short-term culture reflects DNA synthesis related to sper-matogonial proliferation and that of long-term culturerelates mainly to meiosis. In both the short- and long-term culture, DHP and 11-KT induced germ cell DNAsynthesis. In the short-term culture, the DNA synthe-sis induced by 11-KT stimulation was not preventedby anti-DHP treatment. In the long-term culture, how-ever, the DNA synthesis induced by 11-KT stimula-tion was prevented by anti-DHP treatment. In the long-term culture for 15 days, DHP antibodies were onlypresent during the last 6 days, suggesting that 11-KTcan induce two kinds of DNA synthesis in germ cells,

one that is not and another, subsequent one that is me-diated by DHP. As mentioned above, the germ cellsprogress from spermatogonial proliferation to meiosison day 15 after initiation of the culture with 11-KT(Miura et al. 1991a). These findings suggest that DHPacts on a late stage of spermatogonial proliferation andor on meiosis (Miura et al. 2006).

To understand the relationship between DHP andmeiosis, we investigated the DHP-induced change ofexpression of the meiosis-specific markers Dmc1 andSpo11 by using a testicular organ culture. Spo11 is in-volved in the formation of DNA double-strand breaksduring the homologous recombination of the meioticprophase in yeast (Keeney et al. 1997; Smith andNicolas 1998) and Dmc1 is an Escherichia coli RecA-like protein involved only in meiotic recombinationin yeast (Bishop et al. 1992; Bishop 1994). The rolesof these proteins in meiosis are conserved throughouteukaryotic species (Keeney 2001). Both markers were

Fig. 9. The ovarian epithelium culture technique. To remove oocytes of previtellogenic and vitellogenic stage, ovarian frag-ments of carp were excised and treated with enzymes. The residual ovarian fragments were precultured for one month. Afterpreculture, ovarian explants were cultured in media with or without 1 ng/ml DHP for 14 days. Small arrows indicate oogonia.Photographs of 1 month culture is originally published in Miura et al., A Progestin and Estrogen Regulate Early Stages ofOogenesis in Fish, Biology of Reproduction 77, 822–828, Fig. 7, 2007.



induced by DHP in a testicular organ culture. Immu-nocytochemistry showed that the germ cells express-ing Spo11 were undifferentiated spermatogonia. Fur-thermore, synaptonemal complexes, structures specificfor the meiotic prophase, were observed in some germcells of this type in testicular fragments cultured withDHP. Collectively, these data suggest that DHP inducedearly spermatogonia to enter the meiotic prophase(Miura et al. 2006).

In conclusion, we have demonstrated a new func-tion of progestin by using an eel testis in vitro culturesystem. Progestin is an essential hormone involved inthe regulation of not only final maturation but also ofearly stages of spermatogenesis, especially the initia-tion of meiosis in fish. Hence, progestin is an initiatorof meiosis in spermatogenesis (Fig. 8).

6-3. A common point of meiosis in early oogen-esis and spermatogenesis

The control of early oogenesis, from the prolifera-tion of oogonia to the initiation of meiosis and the con-trol mechanisms of the early stages of spermatogen-esis are very similar. As mentioned above, we haveshown that DHP is an essential factor for the initiationof meiosis in spermatogenetic cells of the Japanese eel(Miura et al. 2006), thus, we investigated the involve-ment of DHP and E2 in early oogenesis (Miura et al.2007).

During early oogenesis, oogonia proliferate by mi-tosis and subsequently develop into primary oocytesthat have initiated meiosis. In general, during oogen-esis, primary oocytes arrest division at the diplotene

Fig. 10. Effect of DHP on common carp ovarian fragments cultured for 14 days. Light micrographs showing ovarian sectionfrom fragments cultured in basal medium without hormone (control) or with 1 ng/ml DHP. Cells with arrows are chromatin-nucleous stage oocytes. Electron micrograph of germ cells with synaptonemal complexes (arrowheads) in ovarian fragmentscultured with 1 ng/ml DHP. Photographs by light microscope is originally published in Miura et al., A Progestin and EstrogenRegulate Early Stages of Oogenesis in Fish, Biology of Reproduction 77, 822–828, Fig. 10, 2007.



stage in the prophase of the first meiotic division andaccumulate yolk during meiotic arrest. Thereafter, theoocytes resume the first meiotic division and differen-tiate into mature eggs through final maturation. In fish,the differentiation of primary oocytes into maturingoocytes that resume meiosis requires several steroidhormones. As in other vertebrates, oogenesis is prima-rily regulated by pituitary GTHs and ovarian endocrinefactors, including estrogens and progestins. The rolesof estrogens and progestins have been elucidatedlargely through correlating seasonal changes in circu-lating hormone levels with the different stages of theannual ovarian cycle in a variety of fish species (Crimand Idler 1978; Lamba et al. 1983; Kime et al. 1991;Cornish 1998).

Using two species of teleost fish, Japanese huchen(Hucho perryi) and common carp (Cyprinus carpio),we investigated whether sex steroids are involved inearly oogenesis in vitro (Fig. 9). Ovarian fragmentswere cultured to examine the effects of a progestin,DHP and an estrogen, E2. DHP and E2 significantlypromoted DNA synthesis in ovarian germ cells, asjudged by 5-bromo-2-deoxyuridine (BrdU) incorpora-tion into these cells. Furthermore, to detect the initia-tion of the first meiotic division of early oogenesis,we assessed ultrastructurally the occurrence of synap-tonemal complexes (SCs) and analyzed by immuno-histochemistry the expression of a meiosis-specificmarker, Spo11. In huchen, a higher percentage ofoocytes with SC was seen in DHP-treated ovarian frag-ments than in control or E2-treated ovarian fragments.Spo11 was expressed in germ cells after DHP treat-ment of carp ovarian explants. These data suggest thatthe progression of germ cells through early oogenesisinvolves two sex steroids: E2, which acts directly on

oogonial proliferation and DHP, which acts directly onthe initiation of the first meiotic division of oogenesis.Therefore, DHP is also implicated in the regulation ofearly oogenesis from oogonial proliferation to initia-tion of the first meiotic division (Fig. 10) (Miura et al.2007).

These findings from male and female gonad tissueand cell culture systems suggest that DHP, a proges-tin, is an essential factor for the initiation of meiosis;and E2, an estrogen, is essential in gonial proliferationin both spermatogenesis and oogenesis (Fig. 11 or Fig.12).

What are the molecular mechanisms by which DHPinitiates meiosis in fish? Although we attempted toclone key factors regulated by DHP stimulation (Ozakiet al. 2006), the mechanism has not yet been clarified.

6-4. The downstream factors of DHP

How does DHP initiate meiosis in spermatogenesis?In the Japanese eel, trypsin, which is a kind of serineprotease, is a key factor for the downstream regulationof DHP (Miura et al. 2009). To identify novel factorsthat are regulated by DHP, we carried out gene expres-sion cloning using eel testicular fragments that werecultured with or without DHP for six days. We screened25 up-regulated cDNA clones following DHP treat-ment. Among these, a cDNA clone of trypsinogen wasidentified (Fig. 13). Trypsinogen is a precursor oftrypsin, a member of the serine protease family that ismainly produced in the pancreas as a digestive enzyme.Although it has been reported that serine proteases arepresent in testicular tissue, their functions have not yetbeen clarified (Odet et al. 2006; Ogiwara and Takahashi2007).

Fig. 11. DHP, a progestin, is an essential factor for the initiation of meiosis in both spermatogenesis and oogenesis.



To further elucidate the relationship between testicu-lar trypsinogen expression and sex steroids such asDHP, which are associated with the regulation of sper-matogenesis, we examined the effects of 11-KT, E2and DHP upon the trypsinogen mRNA levels in theJapanese eel testis. Eel testicular fragments were cul-tured for six days with these three steroids and theirexpression of trypsinogen was examined. Northern blotanalysis showed that trypsinogen expression is inducedonly by DHP stimulation (Fig. 13). To determine thedistribution of trypsinogen in the testis, we performedimmunohistochemistry using an eel trypsinogen anti-body. Positive staining of the Sertoli cells surroundingthe late type B spermatogonia and of the spermatidsand spermatozoa was observed. These findings sug-gest that trypsinogen is related to the regulation of ini-tiation of meiosis in early stages of late type B sper-matogonial development under DHP stimulation(Miura et al. 2009).

6-5. Trypsin and meiosis

To investigate the relationship between trypsinogenand spermatogenesis, germ cell/somatic cell pelletswere cultured with or without an anti-eel trypsinogenantibody and/or DHP or 11-KT for six days. As posi-tive controls, treatment with 11-KT or DHP alone sig-nificantly stimulated DNA synthesis in spermatogonialcells. The addition of an anti-eel trypsinogen antibodysignificantly reduced DHP-induced but not 11-KT-induced spermatogonial DNA synthesis. Since trypsinis a serine protease, we therefore investigated the ef-fects of the serine protease inhibitors, 4-(2-aminoethyl)-benzenesulfonyl fluoride (AEBSF) andphenylmethylsulfonyl fluoride (PMSF) on spermato-gonial DNA synthesis stimulated by DHP. Germ cell/somatic cell pellets were cultured with or without theseinhibitors and with DHP or 11-KT for six days, butonly DHP-induced spermatogonial DNA synthesis was

Fig. 12. A schematic summary of the possible control mechanisms of spermatogenesis and oogenesis in fish. SPS,spermatogenesis-preventing substance; SSRF, spermatogonial stem cell renewal factor. Reprinted with permission from Cybium,32(2) supplement, Miura and Miura, Progestin is an essential factor for the initiation of the meiosis in spermatogenesis andearly oogenesis in fish, 130–132, 2008, Société Française d’Ichtyologie.



reduced. These findings indicate that trypsin has im-portant role in spermatogonial DNA synthesis underDHP stimulation for serine protease. The direct effectsof trypsin on spermatogenesis were monitored usingan eel germ cell/somatic cell co-culture system. Germcells/somatic cell pellets were cultured with variousconcentration of pig trypsin. At the start of cultiva-tion, all of the germ cells in the pellets were undiffer-entiated spermatogonia. The addition of trypsin to theculture medium induces DNA replication. Signifi-cantly, in the controls without any supplement, theDNA synthesis of germ cells did not change from itsinitial value. To understand the relationship betweentrypsin and meiosis, we examined the effects of trypsinon the expression of the meiosis-specific marker Spo11,in a germ cell/somatic cell co-culture system by im-munohistochemistry. Prior to cultivation, there was nodetectable Spo11 in the germ cells within the pellets.After two days in culture, however, trypsin treatmentwas found to induce Spo11 expression. These findingsindicate that trypsin has an important role in the initia-tion of meiosis in spermatogenesis.

Fig. 13. To identify factors that are regulated by DHP, we carried out gene expression cloning using eel testicular fragmentsthat were cultured with (DHP) or without (Cont) 100 ng/mL DHP for 6 days. After cultivation, poly (A)+ RNAs were ex-tracted from these testicular fragments and a subtracted cDNA library was then constructed from +DHP and –DHP cDNAenriched via the RDA procedure (Niwa et al. 1997). One thousand clones from each of these libraries were subsequentlyscreened by differential hybridization, using each enriched cDNA preparation as a probe. We thereby screened 25 non-cross-hybridizing and up-regulated cDNA fragments after DHP treatment. Southern blotting analysis showed 12 typical cDNAfragments, whose names were eSRS36, 37, 38, 39, 40, 41, 42, 43, 44, 56, 29 and 33. Among these, an eSRS56 cDNA fragmentcorresponding to trypsinogen was identified (GenBank Acc. No. AB519643). Trypsinogen expression in the Japanese eeltestis was determined by northern blot analysis of cultured testicular fragments. Pooled testicular fragments from 10 eels werecultured without (C) or with 10 ng/mL of 11-KT, E2 or DHP for 6 days. Lane IC shows the initial control before culturing.Northern blot analysis of EF1, which serves as reference, is also shown. Reprinted with permission from PNAS, 106, Miura etal., Trypsin is a multifunctional factor in spermatogenesis, 20972–20977, Fig. 1, 2009, National Academy of Sciences.

6-6. Trypsin and spermiogenesis

Trypsin has another important role in spermatogen-esis aside from initiating meiosis. Type A spermatogo-nia were cultured with various concentrations of trypsin(0.1–100 mM) for 15 days and the appearance of thecells was then evaluated. Prior to cultivation, type Aspermatogonia remained rounded but after three daysin culture with 100 mM of trypsin, these cells adopteda spindle-shaped morphology. After 15 days of treat-ment with 100 mM of trypsin, a flagelliform structurewas detectable on one side of these spindle-shapedgerm cells. In cultures without trypsin, no morphologi-cal changes were evident in the germ cells. Using flow-cytometry, the nuclear phases of these spindle-shapedgerm cells were recorded. The resulting flow-cytometric histograms showed 2C and 4C peaks, i.e.,these cells did not undergo a normal meiotic division.The morphology of the elongated germ cells exposedto trypsin was compared with normal eel spermatozoaby histological observation. The flagella of the eel sper-matozoa exhibit a 9 + 0 axonemal structure and nine



pairs of microtubules in each flagellum are divided into4 and 5 pairs, respectively, on the sperm head wherethey extend to the caput end. Both sets of microtubulesand the flagella can be detected by immunocytochem-istry using α-tubulin antibodies. In sperm-like cellstreated with trypsin, an axonemal structure could notbe detected in the flagelliform structure using electronmicroscopy, whereas flagelliform structures andmicrotubule-like lines at the cell surface were foundto be specifically stained by α-tubulin antibodies.These findings suggest that trypsin induces partial sper-miogenesis (Miura et al. 2009).

6-7. Trypsin and fertilization

Trypsin also plays an important role in fertilization.The trypsinogen expression profile in eel spermatozoawas analyzed. Using a specific antibody, trypsin wasdetected in the spermtozoa membranes and its activitywas also detected using gelatin zymography. Since ithas also been reported that membrane-type serine pro-tease exists in elongated spermatids in mammals (Wonget al. 2001; Scarman et al. 2001), there is a possibilitythat sperm head trypsin in eels is similar to this serine

protease in mammals. We also wished to investigatethe possibility that sperm head trypsin is related to fer-tilization and evaluated the relationship between ferti-lization and sperm head trypsin in the eel. Ejaculatedeel sperm were incubated in artificial seminal plasmasupplemented with the specific serine protease inhibi-tors PMSF or AEBSF or with anti-eel trypsin antibod-ies, for three hours. Inseminations were then performedusing these incubated sperm and normal eel eggs. Fourhours after insemination, fertilized eggs were countedand the rate of fertilization was thereby calculated. Theresults showed that both serine protease inhibitors andalso the trypsin antibodies significantly reduced thefertilization rate. In teleost fish, except for some spe-cies such as the sturgeon, spermatozoa do not have anacrosome (Grier 1981; Cherr and Clark 1984) and fer-tilization is facilitated via the egg’s micropyle. Hence,an acrosome reaction is not required during the fertili-zation of fish eggs. However, there is little informa-tion currently available regarding the factors or mecha-nisms that play a role in the fertilization of fish eggs(Yanagimachi et al. 1992; Morisawa 2008). Trypsin ortrypsin-like proteases in the sperm head may play acritical role in fertilization in fish.

Fig. 14. A schematic summary of the roles of trypsin on spermatogenesis in the Japanese eel.



These results demonstrate that trypsin and/or atrypsin-like protease play an important role in the regu-lation of three reproductive events in the male eel; thatis, the initiation of meiosis, spermiogenesis and ferti-lization (Fig. 14).

7. The regulation of final maturation of male fish

In mammals, the term “spermiation” indicates thatspermatozoa are released by the Sertoli cells, whichinvolves the disintegration of a junctional complexbetween Sertoli and germ cells (O’Donnell et al. 2000).Released spermatozoa can leave the testis suspendedin liquid produced by Sertoli cells, via the tubular lu-men that is connected to an efferent duct system. Inteleosts, however, such a junctional complex betweenspermatids and Sertoli cells does not exist like othervertebrates (Schulz et al. 2010) and spermiation refersto the opening of spermatogenic cysts, also terminat-ing the close relation between Sertoli and germ cells.Next to this definition sensu strictu, the term spermia-tion in fish is also used to indicate that spermatozoahydration has taken place, enabling spermatozoa mi-gration towards the sperm duct, from which it can bereleased for fertilization.

Fig. 15. An illustration showing the action of oxidative stress on eel spermatogenesis.

During the breeding season, the levels of numeroushormones show remarkable changes in male teleosts,which are initiated by an increase in LH secretion (Pratet al. 1996). LH secretion induces an increase in theproduction of the testicular steroids, such as 11-KT andDHP or 20β-S. 11-KT injections induced spermiationin goldfish and some salmonids and DHP injectionshad similar effects in several salmonids and eels. WhileLH and these sex steroids are clearly involved in regu-lating spermiation in fish, the mechanisms of action ofthese hormones on milt hydration, sperm migration tothe sperm duct or increase in milt volume, are stillunclear. In some teleost species (Miura et al. 1991b,1992), spermatozoa released from Sertoli cells aftercompletion of spermiogenesis are not yet capable offertilizing eggs. In salmonids, spermatozoa in the tes-tis and in the sperm duct are immotile. Dilution withfresh water induces motility in spermatozoa collectedfrom the sperm duct while testicular spermatozoa re-main immotile after dilution. Thus, spermatozoa ac-quire motility during their passage through the spermduct.

Sperm maturation, the phase during which non-functional gametes develop into mature spermatozoa(fully capable of vigorous motility and fertilization)



involves physiological but no morphological changes.In salmonids, sperm maturation (the acquisition ofsperm motility) has been induced by increasing theseminal plasma pH (approximately to pH 8.0) in thesperm duct, which results in the elevation of intra-sperm cAMP levels (Morisawa and Morisawa 1988;Miura et al. 1992). Similar results have been reportedfor Japanese eel spermatozoa by Miura et al. (1995b)and Ohta et al. (1997). Sperm maturation is also regu-lated by the endocrine system. In some teleosts, in-cluding the Japanese eel, it has been suggested thatDHP regulates sperm maturation (Miura et al. 1991b,1992). It seems that DHP does not act directly on thesperm; its action is rather mediated through an increasein the seminal plasma pH, which in turn increases thesperm content of cAMP, thereby allowing the acquisi-tion of sperm motility (Miura et al. 1991b, 1992,1995b). Thus, sex steroid hormones; estrogen, andro-gen and progestin, are important regulators for the pro-gression of spermatogenesis from spermatogonial stemcell renewal to sperm maturation.

8. The protection mechanisms of germ cells fromchemical-induced stress

Germ cells are highly specialized cells that are re-sponsible for the propagation of DNA which directsthe development of future generations. It is essentialfor organisms to maintain the integrity of germ cellDNA to ensure that the continuation of the species isnot compromised. Previous studies have shown thatspermatogenic cells have a lower mutation frequencythan somatic cells (Walter et al. 1998; Winn et al.

2000). Yet, germ cells are continually affected detri-mentally by endogenous and exogenous agents, suchas reactive oxygen species (ROS), which can causeDNA damage. Spermatozoa were found to be highlysensitive to ROS-induced damage (Aitken andClarkson 1987), while spermatogonia are reportedlytolerant to ROS (Aruldhas et al. 2005). Previous stud-ies revealed that in mice exposed to mild heat stress,which can consequently lead to oxidative stress (Paulet al. 2009), numerous apoptotic late-type germ cellswere found while apoptotic spermatogonia were rare(Paul et al. 2008). However, the precise reason for thisphenomenon remains unclarified. In vertebrates, a va-riety of antioxidant defense mechanisms have evolvedto protect cells and tissues against ROS. Among thewell-known antioxidant enzymes protecting cells fromROS are the superoxide dismutases (SOD).

Although the effects of ROS have been extensivelystudied in mammals, not much is known about its di-rect impact on vertebrate germ cells. It was previouslyshown that spermatogonia are highly tolerant to ROSattack while advanced-stage germ cells such as sper-matozoa are much more susceptible, however, the pre-cise reason for this variation in ROS tolerance remainsunknown (Fig. 15).

Using the Japanese eel testicular culture system,which enables a complete spermatogenesis in vitro(Miura et al. 1991a), we report that advanced-stagegerm cells undergo intense apoptosis and exhibit strongsignal for 8-hydroxy-2′-deoxyguanosine (8-OHdG), anoxidative DNA damage marker, upon exposure to hy-poxanthine (Hx)-generated ROS, while spermatogoniaremain unaltered. Activity assay of antioxidant enzyme,

Table 1. Using the Japanese eel testicular culture system, we found that advanced-stage germ cells undergo intense apoptosisand exhibit a strong signal for 8-hydroxy-2′-deoxyguanosine, an oxidative DNA damage marker, upon exposure to hypoxan-thine (Hx)-generated reactive oxygen species (ROS), while spermatogonia remain unaltered. Spermatogonia are highly toler-ant to ROS attack while advanced-stage germ cells such as spermatozoa, are much more susceptible. Low ROS, Hx treatmentat the low dose (1 µM); High ROS, Hx treatment at the highest dose (100 µM).



SOD and western blot analysis using an anti-copper/zinc (Cu/Zn) SOD antibody showed high SOD activ-ity and Cu/Zn SOD protein concentration during earlyspermatogenesis. Immunohistochemistry showedstrong expression of Cu/Zn SOD in spermatogonia butweak expression in advanced-stage germ cells. Zn de-ficiency reduced the activity of the recombinant eelCu/Zn SOD protein. Cu/Zn SOD siRNA decreased Cu/Zn SOD expression in spermatogonia and led to in-creased oxidative damage (Table 1) (Celino et al.2011).

Overall, our study demonstrated that the decrease inlevels of SOD activity, expression of Cu/Zn SOD andthe levels of Zn as spermatogenesis progresses are thereasons for the vulnerability of advanced stage germcells to ROS attack. The high levels of Cu/Zn SODand Zn in spermatogonia may render them less sus-ceptible to ROS attack. Since spermatogonia are onthe unprotected side of the testis barrier, they may beprone to more DNA damage from circulating molecu-lar insults than cells on the protected side. As stem cells,spermatogonia need to ensure the integrity of the genesrequired for development and the continuity of life.Thus, spermatogonia have evolved with a greater needfor elevated levels of protective factors against DNAdamage (Yamaguchi et al. 2009).

Fish reproduction has been considered to be a reli-able indicator of endocrine disruption in aquatic sys-tems by chemical compounds, including arsenic. Ofthe two forms, As (V) typically dominates both in oxicsea and freshwater (Smedley and Kinniburgh 2002;Duker et al. 2005); consequently, fish are likely to beexposed to As (V). However, few studies described thetoxicity of the less toxic form, As (V), on fish repro-duction. We have previously demonstrated that in vitrotreatment of As (V) inhibited spermatogenesis in theJapanese eel (Yamaguchi et al. 2007). However, themechanism involved in the direct influence of arsenicon fish spermatogenesis is not yet well clarified. Hence,using the Japanese eel testicular organ culture systemwe examined the direct effects and toxic mechanismsof arsenic on fish spermatogenesis. We found that ar-senic treatment provoked a dose-dependent inhibitionof hCG-induced germ cell proliferation, as revealedby BrdU immunohistochemistry. Time-resolved fluo-rescent immunoassay showed that arsenic suppressedhCG-induced synthesis of 11-KT in testicular frag-ments that were incubated. A 0.1 mM (7 mg/l) dose ofarsenic, which is lower than the World Health Organi-zation drinking water quality guideline of 10 mg/l, mosteffectively reduced 11-KT production. The hCG-induced synthesis of progesterone from pregnenolonewas significantly inhibited by low doses of arsenic(0.1–1 mM), implying an inhibition of 3β-hydroxysteroid dehydrogenase activity. Germ cellsundergo apoptosis at the highest dose of arsenic (100mM). An arsenic concentration-dependent increase in

oxidative DNA damage was detected by 8-OHdG im-munohistochemistry. A peak in 8-OHdG index wasobserved in testicular fragments treated with 100 mMarsenic and hCG consistent with the results of TdT-mediated dUTP nick end labeling (TUNEL) assay as amethod of detecting apoptotic cells. Thus, these datasuggest that low doses of arsenic may inhibit sperma-togenesis via steroidogenesis suppression, while highdoses of arsenic induce oxidative stress-mediated germcell apoptosis (Celino et al. 2009). Furthermore, toclarify the direct effects of ROS on germ cells, we stud-ied the effects of hypoxanthine-induced ROS on sper-matogenesis. Immunohistochemistry for BrdU showedthat Hx treatment at a low dose (1 µM) already inhib-its 11-KT-induced germ cell proliferation after six daysof culture. TUNEL assay and 8-OHdG immunohisto-chemistry revealed an intense germ cell apoptosis andhigh oxidative DNA damage in testicular fragmentscultured at the highest dose of the hypoxanthine (100µM) with 11-KT after three days of culture. Total SODactivity assay showed a decrease in SOD activity intesticular fragments after six days of culture with 11-KT. Thus, these suggest that ROS may directly inhibitspermatogenesis and that decreased SOD activityrenders proliferating spermatogonia susceptible to ROSand hence, leading to apoptosis. Our recent studies haveshown that high levels of Cu/Zn SOD are present inspermatogonia, which renders it tolerant to oxidativestress compared to advanced germ cells (Celino et al.2011), confirming these results.

9. Conclusion

Thus, fish spermatogenesis is controlled by the sexsteroid hormones. Sex steroid hormones; estrogen,androgen and progestin, are important regulators forthe progression of spermatogenesis. Mitotic divisionsof spermatogonia can be categorized by spermatogo-nial stem cell renewal and spermatogonial prolifera-tion toward meiosis. Spermatogonial renewal is regu-lated by E2 and spermatogonial proliferation towardmeiosis is promoted by 11-KT. The action of E2 and11-KT is mediated by other factors produced by Ser-toli cells; E2 is mediated by spermatogonial stem-cellrenewal factor and 11-KT is mediated by spermatogen-esis preventing substance (anti-Mullerian hormone:AMH) and activin B. Meiosis is induced by DHP, whichis progestin in teleosts. In oogenesis, DHP also initi-ates the meiotic prophase. After spermiogenesis, im-mature spermatozoa undergo sperm maturation. Spermmaturation is also regulated by DHP. DHP acts directlyon spermatozoa to activate the carbonic anhydrase thatis present in the spermatozoa. This enzymatic activa-tion causes an increase in the seminal plasma pH, ena-bling spermatozoa to become motile.

By the establishment of eel testicular organ cultureand the use of molecular biology techniques, analysis



of the control mechanisms of fish spermatogenesis hasadvanced remarkably. Recently, we tried to establishnew methods for analyzing the regulatory mechanismsof spermatogenesis, for example, the method of exog-enous gene transfer into testicular cells usingelectroporation, and the germ cell and Sertoli cell co-culture system. It is highly possible that further inves-tigations using these new methods will lead to a betterunderstanding of the general aspects of spermatogen-esis.

ReferencesAbel MH, Mootton AN, Wilkins V, Huhtaniemi I, Knight

PG, Charlton HM. The effect of a null mutation in thefollicle-stimulating hormone receptor gene on mouse re-production. Endocrinology 2000; 141(5): 1795–1803.

Aitken RJ, Clarkson JS. Cellular basis of defective spermfunction and its association with the genesis of reactiveoxygen species by human spermatozoa. J. Reprod. Fertil.1987; 81: 459–469.

Amer MA, Miura T, Miura C, Yamauchi K. Involvement ofsex steroid hormones in the early stages of spermatogen-esis in Japanese huchen (Hucho perryi). Biol. Reprod.2001; 65: 1057–1066.

Ando N, Miura T, Nader MR, Miura C, Yamauchi K. Amethod for estimating the number of mitotic divisions infish testes. Fish. Sci. 2000; 66: 299–303.

Aruldhas MM, Subramanian S, Sekar P, Vengatesh G,Chandrahasan G. Chronic chromium exposure-inducedchanges in testicular histoarchitecture are associated withoxidative stress: study in a non-human primate (Macacaradiata Geoffroy). Hum. Reprod. 2005; 20: 2801–2813.

Baynes SM, Scott AP. Seasonal variations in parameters ofmilt production and in plasma concentration of sex ster-oids of male rainbow trout (Salmo gairdneri). Gen. Comp.Endocrinol. 1985; 57: 150–160.

Betka M, Callard GV. Negative feedback control of the sper-matogenic progression by testicular oestrogen synthesis:insights from the shark testis model. APMIS 1998; 106:252–257.

Billard R. Spermatogenesis and spermatology of some teleostfish species. Reprod. Nutr. Develop. 1986; 26: 877–920.

Billard R, Fostier A, Weil C, Breton B. Endocrine control ofspermatogenesis in teleost fish. Can. J. Fish. Aquat. 1982;39: 65–79.

Bishop DK. RecA homologs Dmc1 and Rad51 interact toform multiple nuclear complexes prior to meiotic chro-mosome synapsis. Cell 1994; 79: 1081–1092.

Bishop DK, Park D, Xu L, Kleckner N. DMC1: a meiosis-specific yeast homolog of E. coli recA required for re-combination, synaptonemal complex formation and cellcycle progression. Cell 1992; 69(3): 439–456.

Burris TB. Progestins. In: Knobil E, Neil JD (eds.). Ency-clopedia of Reproduction. Academic, San Diego. 1998;23–30.

Callard IP, Callard GV. Sex steroid receptors and nonreceptorbinding proteins. In: Norris DO, Jones RE (eds.). Hor-mones and Reproduction in Fishes, Amphibians and Rep-tiles. Plenum Press, New York. 1987; 355–384.

Callard IP, Callard GV, Lance V, Bolaffi JL, Rosset JS. Tes-ticular regulation in nonmammalian vertebrates. Biol.

Reprod. 1978; 18: 16–43.Celino FT, Yamaguchi S, Miura C, Miura T. Arsenic inhib-

its in vitro spermatogenesis and induces germ cellapoptosis in the Japanese eel (Anguilla japonica). Repro-duction 2009; 138: 279–287.

Celino FT, Yamaguchi S, Miura C, Ohta T, Tozawa Y, IwaiT, Miura T. Tolerance of spermatogonia to oxidative stressis due to high levels of Zn and Cu/Zn superoxidedismutase. PLoS ONE 2011; 6(2): e16938, doi:10.1371/journal.pone.0016938.

Cherr GN, Clark WH. An acrosome reaction in sperm fromwhite sturgeon, Acipenser transmontanus. J. Exp. Zool.1984; 232: 129–139.

Ciocca DR, Roig LMV. Estrogen receptors in human non-target tissue: Biochemical and clinical implications. En-docrine Rev. 1995; 16: 35–62.

Clermont Y. Kinetics of spermatogenesis in mammals: sem-iniferous epithelium cycle and spermatogonial renewal.Physiol. Rev. 1972; 52: 198–236.

Cooke HJ, Hargreave T, Elliott DJ. Understanding the genesinvolved in spermatogenesis: a progress report. Fertil.Steril. 1998; 69: 989–995.

Cornish DA. Seasonal steroid hormone profiles in plasmaand gonads of the tilapia, Oreochromis mossambicus.Water S.A. 1998; 24: 257–263.

Courot M, Reviers MTH, Ortavant R. Spermatogenesis. In:Johnson AD (ed.). The Testis, Vol. 1. Academic Press, NewYork. 1970; 339–432.

Crim LW, Idler DR. Plasma gonadotropin, estradiol, andvitellogenin and gonad phosvitin levels in relation to theseasonal reproductive cycles of female brown trout. Ann.Biol. Anim. Biochim. Biophys. 1978; 18: 100.

Depeche J, Sire O. In vitro metabolism of progesterone and17α-hydroxyprogesterone in the testis of the rainbow trout,Salmo gairdneri Rich, at different stages of spermatogen-esis. Reprod. Nutr. Develop. 1982; 22: 427–438.

Dierich A, Sairam MR, Monaco L, Fimia GM, GansmullerA, LeMeur M, Sassone-Corsi P. Impairing follicle-stimulating hormone (FSH) signaling in vivo: Targeted dis-ruption of the FSH receptor leads to aberrant gametogen-esis and hormonal imbalance. Proc. Natl. Acad. Sci. USA1998; 95: 13612–13617.

Dufau ML, Miyagawa Y, Takada S, Khanum A, MiyagawaH, Buczko E. Regulation of androgen synthesis: The latesteroidogenic pathway. Steroids 1997; 62: 128–132.

Duker AA, Carranza EJM, Hale M. Arsenic geochemistryand health. Environ. Int. 2005; 31: 631–641.

Ewing H. Spermatogenesis in zebra fish Brachydanio rerio(Hamilton-Buchanan). Anat. Rec. 1972; 172: 308.

Fasano S, Pieratoni R. The vertebrate testis: Communica-tion between interstitial and germinal compartments. In:Facchinetti F, Henderson IW, Pierantoni R, Polzonetti-Magni A (eds.). Cellular Communication in Reproduction.Journal of Endocrinology Ltd, Bristol. 1993; 113–124.

Furukawa T, Yoshimura A, Sumizawa T, Haraguchi M,Akiyama S-I, Fukui K, Ishizawa M, Yamada Y. Angiogenicfactor. Nature 1992; 356: 668.

Gerton JL, Hawley RS. Homologous chromosome interac-tions in meiosis: diversity amidst conservation. Nat. Rev.Gene. 2005; 6: 477–487.

Gilbert SF. The saga of the germ line. In: Wigg C (ed.). De-velopmental Biology. Sinauer Associates, Sunderland, MA.



1985; 593–623.Grier HJ. Cellular organization of the testis and spermatogen-

esis in fishes. Am. Zool. 1981; 21: 345–357.Hansson D, Calandra R, Purvis K, Ritzen M, French F. Hor-

monal regulation of spermatogenesis. Vitam. Horm. 1976;34: 187–214.

Hayakawa Y, Morita T, Kitamura W, Kanda S, Banba A,Nagaya H, Hotta K, Sohn YC, Yoshizaki G, KobayashiM. Biological activities of single-chain goldfish follicle-stimulating hormone and luteinizing hormone. Aquaculture2008; 274: 408–415.

Hong Y, Liu T, Zhao H, Xu H, Wang W, Liu R, Chen T,Deng J, Gui J. Establishment of a normal medakafish sper-matogonial cell line capable of sperm production in vitro.Proc. Natl. Acad. Sci. USA 2004; 101: 8011–8016.

Huhtaniemi IT, Themmen AP. Mutations in human gonado-tropin and gonadotropin receptor genes. Endocrine 2005;26: 207–217.

Idler DR, Bitner II, Schmidt PJ. 11-Ketotestosterone: anandrogen for sockeye salmon. Can. J. Biochem. Physiol.1961; 39: 1737–1742.

Ikeuchi T, Todo T, Kobayashi T, Nagahama Y. Two subtypesof androgen and progestogen receptors in fish testes.Comp. Biochem. Physiol. 2001; 129B: 449–455.

Kamei H, Kwazoe I, Kaneko T, Aida K. Purification offollicle-stimulating hormone from immature Japanese eel,Anguilla japonica, and its biochemical properties and ster-oidogenic activities. Gen. Comp. Endocrinol. 2005; 143:257–266.

Kamei H, Kaneko T, Aida K. In vivo gonadotropic effects ofrecombinant Japanese eel follicle-stimulating hormone.Aquaculture 2006; 261: 771–775.

Kazeto Y, Kohara M, Miura T, Miura C, Yamaguchi S, TrantJM, Adachi S, Yamauchi K. Japanese eel Follicle-stimulating hormone (Fsh) and Luteinizing hormone (Lh):Production of biologically active recombinant Fsh and Lhby drosophila S2 cells and their differential actions on thereproductive biology. Biology of Reproduction 2008; 79:938–946.

Keeney S. Mechanism and control of meiotic recombina-tion initiation. Curr. Top. Dev. Biol. 2001; 52: 1–53.

Keeney S, Giroux CN, Kleckner N. Meiosis-specific DNAdouble-strand breaks are catalyzed by Spo11, a memberof a widely conserved protein family. Cell 1997; 88: 375–384.

Kim JJ, Jaffe RC, Fazleabas AT. Comparative studies on thein vitro decidualization process in the baboon (Papioanubis) and human. Biol. Reprod. 1998; 59: 160–168.

Kime DE, Lone KP, Al-Marzouk A. Seasonal changes inserum steroid hormones in a protandrous teleost, thesobaity (Sparidentex hasta Valenciennes). J. Fish. Biol.1991; 39: 745–753.

Kluver N, Pfennig F, Pala I, Storch K, Schlieder M,Froschauer A, Gutzeit HO, Schartl M. Differential expres-sion of anti-Müllerian hormone (amh) and anti-Müllerianhormone receptor type II (amhrII) in the teleost medaka.Dev. Dyn. 2007; 236: 271–281.

Kobayashi M, Aida K, Stacey NE. Induction of testis devel-opment by implantation of 11-ketotestosterone in femalegoldfish. Zool. Sci. 1991; 8: 389–393.

Kobayashi T, Sakai N, Adachi S, Asahina K, Iwasawa H,Nagahama Y. 17α,20α-Dihydroxy-4-pregnen-3-one is the

naturally occurring spermiation-inducing hormone in thetestis of a frog, Rana nigromaculata. Endocrinology 1993;133: 321–327.

Krishnamurthy H, Danilovich N, Morales CR, Sairam MR.Qualitative and quantitative decline in spermatogenesisof the follicle-stimulating hormone receptor knockout(FORKO) mouse. Biol. Reprod. 2000; 62: 1146–1159.

Kumar TR, Wang Y, Lu N, Matzuk MM. Follicle stimulat-ing hormone is required for ovarian follicle maturationbut not male fertility. Nat. Genet. 1997; 15: 202–204.

Lamba VJ, Goswami SV, Sundararaj BI. Circannual and cir-cadian variations in plasma levels of steroids (cortisol,estradiol-17β, estrone and testosterone) correlated withannual gonadal cycle in the catfish, Heteropneustus fossilis(Bloch). Gen. Comp. Endocrinol. 1983; 50: 205–225.

Larsen LO, Dufour S. Growth, reproduction and death inlampreys and eels. In: Rankin JC, Jensen FB (eds.). FishEcophsiology. Chapman & Hall, London. 1993; 72–104.

LeGac F, Loir M, LeBail PY, Ollitrault M. Insulin-likegrowth factor (IGF-I) mRNA and IGF-I receptor in trouttestis and in isolated spermatogenic and Sertoli cells. Mol.Reprod. Dev. 1996; 44: 23–35.

Lejeune H, Chuzel F, Thomas T, Avallet O, Habert R, DurandPH, Saez J. Regulation paracrine des cellules de Leydig.Ann. Endocrinol. 1996; 57: 55–63.

Li H, Papadopoulos V, Vidic B, Dym M, Culty M. Regula-tion of rat testis gonocyte proliferation by platelet-derivedgrowth factor and estradiol: identification of signalingmechanisms involved. Endocrinology 1997; 138: 1289–1298.

Loir M. In vitro approach to the control of spermatogoniaproliferation in the trout. Mol. Cell. Endocrinol. 1994; 102:141–150.

Loir M, LeGac F. Insulin-like growth factor I and II bindingand action on DNA synthesis in rainbow trout spermato-gonia and spermatocytes. Biol. Reprod. 1994; 51: 1154–1163.

Masui Y. From oocyte maturation to the in vitro cell cycle:the history of discoveries of Maturation-Promoting Fac-tor (MPF) and Cytostatic Factor (CSF). Differentiation2001; 69: 1–17.

Miura C, Miura T, Yamashita M, Yamauchi K, Nagahama Y.Hormonal induction of all stages of spermatogenesis ingerm-somatic cell coculture from immature Japanese eeltestis. Develop. Growth Differ. 1996; 38: 257–262.

Miura C, Miura T, Kudo N, Yamashita M, Yamauchi K.cDNA cloning of a stage specific gene expressed duringHCG-induced spermatogenesis in the Japanese eel. Dev.Growth. Differ. 1999; 41: 463–471.

Miura C, Higashino T, Miura T. A progestin and estrogenregulate early stages of oogenesis in fish. Biol. Reprod.2007; 77: 822–828.

Miura C, Ohta T, Ozaki Y, Tanaka H, Miura T. Trypsin is amultifunctional factor in spermatogenesis. Proc. Natl.Acad. Sci. USA 2009; 106: 20972–20977.

Miura T. Spermatogenetic cycle in fish. In: Knobil E, NeillJD (eds.). Encyclopedia of Reproduction Vol. 4. AcademicPress, San Diego. 1999; 571–578.

Miura T, Miura C. Japanese eel: a model for analysis of sper-matogenesis. Zool. Sci. 2001; 18: 1055–1063.

Miura T, Yamauchi K, Takahashi H, Nagahama Y. Hormo-nal induction of all stages of spermatogenesis in vitro in



the male Japanese eel (Anguilla japonica). Proc. Natl.Acad. Sci. USA 1991a; 88: 5774–5778.

Miura T, Yamauchi K, Takahashi H, Nagahama Y. Involve-ment of steroid hormones in gonadotropin-induced tes-ticular maturation in male eels (Anguilla japonica).Biomed. Res. 1991b; 12: 241–248.

Miura T, Yamauchi K, Nagahama Y, Takahashi H. Inductionof spermatogenesis in male Japanese eel, Anguillajaponica, by a single injection of human chorionic gona-dotropin. Zool. Sci. 1991c; 8: 63–73.

Miura T, Yamauchi K, Takahashi H, Nagahama Y. The roleof hormones in the acquisition of sperm motility insalmonid fish. J. Exp. Zool. 1992; 261: 359–363.

Miura T, Miura C, Yamauchi K, Nagahama Y. Humanrecombinant activin induces proliferation of spermatogo-nia in vitro in the Japanese eel (Anguilla japonica). Fish.Sci. 1995a; 61: 434–437.

Miura T, Kasugai T, Nagahama Y, Yamauchi K. Acquisitionof potential for sperm motility in vitro in Japanese eel(Anguilla japonica). Fish. Sci. 1995b; 61: 533–534.

Miura T, Kawamura S, Miura C, Yamauchi K. Impaired sper-matogenesis in the Japanese eel, Anguilla japonica: Pos-sibility for the existence of factors that regulate entry ofgerm cells into meiosis. Develop. Growth Differ. 1997;39: 685–691.

Miura T, Kudo N, Miura C, Yamauchi K, Nagahama Y. Twotesticular cDNA clones suppressed by gonadotropin stimu-lation exhibit ZP2- and ZP3-like structures in Japaneseeel. Mol. Reprod. Dev. 1998; 51: 235–242.

Miura T, Miura C, Ohta T, Nader MR, Todo T, Yamauchi K.Estradiol-17β stimulates the renewal of spermatogonialstem cells in males. Biochem. Biophys. Res. Com. 1999;264: 230–234.

Miura T, Miura C, Konda Y, Yamauchi K. Spermatogenesis-preventing substance in Japanese eel. Development 2002;129: 2689–2697.

Miura T, Ohta T, Miura CI, Yamauchi K. Complementarydeoxyribonucleic acid cloning of spermatogonial stem cellrenewal factor. Endocrinology 2003; 144: 5504–5510.

Miura T, Higuchi M, Ozaki Y, Ohta T, Miura C. Progestin isan essential factor for the initiation of the meiosis in sper-matogenetic cells of the eel. Proc. Natl. Acad. Sci. USA2006; 103: 7333–7338.

Miwa S, Yan L, Swanson P. Localization of two gonadotro-pin receptors in the salmon gonad by in vitro ligand auto-radiography. Biol. Reprod. 1994; 50: 629–642.

Miyazono K, Usuki K, Heldin CH. Platelet-derived endothe-lial cell growth factor. Prog. Growth Factor Res. 1991; 3:207–217.

Morinaga C, Saito D, Nakamura S, Sasaki T, Asakawa S,Shimizu N, Mitani H, Furutani-Seiki M, Tanaka M,Kondoh H. The hotei mutation of medaka in the anti-Müllerian hormone receptor causes the dysregulation ofgerm cell and sexual development. Proc. Natl. Acad. Sci.USA, 2007; 104(23): 9691–9696.

Morisawa M. Adaptation and strategy for fertilization in thesperm of teleost fish. J. Appl. Ichthyol. 2008; 24: 362–370.

Morisawa S, Morisawa M. Induction of potential for spermmotility by bicarbonate and pH in rainbow trout and chumsalmon. J. Exp. Biol. 1988; 136: 13–22.

Nader MR, Miura T, Ando N, Miura C, Yamauchi K.

Recombinant human insulin-like growth factor I stimu-lates all stages of 11-ketotestosterone-induced sperma-togenesis in the Japanese eel, Anguilla japonica, in vitro.Biol. Reprod. 1999; 61: 944–947.

Nagahama Y. Gonadotropin action on gametogenesis andsteroidogenesis in teleost gonads. Zool. Sci. 1987; 4: 209–222.

Nagahama Y. 17α,20β-dihydroxy-4-pregnen-3-one, amaturation-inducing hormone in fish oocytes: Mechanismsof synthesis and action. Steroids 1997; 62(1): 190–196.

Nagahama Y, Yamamoto K. Cytological changes in the ad-enohypophysis of fresh-water cultivated male Japanese eel,Anguilla japonica, induced to maturation by transfer tosea water and synahorin injection. Bull. Japan Soc. Sci.Fish 1973; 39: 585–594.

Nelson SS. Fishes of the World. John Wiley & Sons, NewYork. 1994.

Niwa H, Harrison LC, DeAizpura TJ, Cram DS. Identifica-tion of pancreatic b cell-related genes by representationaldifference analysis. Endocrinorogy 1997; 138: 1419–1426.

Odet F, Verot A, Le Magueresse-Battistoni B. The mousetestis is the source of various serine proteases and serineproteinase inhibitors (SERPINs): Serine Proteases andSERPINs Identified in Leydig Cells Are under Gonado-tropin Regulation. Endocrinology 2006; 147: 4374–4383.

O’Donnell L, Stanton PG, Bartles JR, Robertson DM. Ser-toli cell ectoplasmic specializations in the seminiferousepithelium of the testosterone-suppressed adult rat. Biol.Reprod. 2000; 63: 99–108.

Ogiwara K, Takahashi T. Specificity of the medakaenteropeptidase serine protease and its usefulness as a bio-technological tool for fusion-protein cleavage. Proc. Natl.Acad. Sci. USA 2007; 104(17): 7021–7026.

Ohta H, Ikeda K, Izawa T. Increase in concentrations of po-tassium and bicarbonate ions promote acquisition of mo-tility in vitro by Japanese eel spermatozoa. J. Exp. Zool.1997; 277: 171–180.

Ohta T, Miyake H, Miura C, Kamei H, Aida K, Miura T.Follicle-stimulating hormone induces spermatogenesismediated by androgen production in Japanese eel, Anguillajaponica. Biol. Reprod. 2007; 77: 970–977.

Ozaki Y, Miura C, Miura T. Molecular cloning and gene ex-pression of Spo11 during spermatogenesis in the Japaneseeel, Anguilla japonica. Comp. Biochem. Physiol. 2006;143(3): 309–314.

Paul C, Murray A, Spears N, Saunders PTK. A single, mild,transient scrotal heat stress causes DNA damage,subfertility and impairs formation of blastocysts in mice.Reproduction 2008; 136: 73–84.

Paul C, Teng S, Saunders PTK. A single, mild, transient scro-tal heat stress causes hypoxia and oxidative stress in mousetestes, which induces germ cell death. Biol. Reprod. 2009;80: 913–919.

Payne AH, Youngblood GL. Regulation of expression of ster-oidogenic enzymes in Leydig Cells. Biol. Reprod. 1995;52: 217–225.

Pierce JG, Parsons TF. Glycoprotein hormones: structure andfunction. Ann. Rev. Biochem. 1981; 50: 465–495.

Planas JV, Swanson P. Maturation-associated changes in theresponse of the salmon testis to the steroidogenic actionsof gonadotropins (GTHI and GTHII) in vitro. Biol. Reprod.1995; 52: 697–704.



Prat F, Sumpter JP, Tyler CR. Validation ofradioimmunoassays for two salmon gonadotropins (GTHI and GTH II) and their plasma concentrations throughoutthe reproductive cycle in male and female rainbow trout(Oncorhynchus mykiss). Biol. Reprod. 1996; 54: 1375–1382.

Sakai N. Transmeiotic differentiation of zebra fish germ cellsinto functional sperm in culture. Development 2002; 129,3359–3365.

Scarman AL, Hooper JD, Boucaut KJ, Sit ML, Webb GC,Normyle JF, Antalis TM. Organization and chromosomallocalization of the murine testis in gene encoding a serineprotease temporally expressed during spermatogenesis.Eur. J. Biochem. 2001; 268: 1250–1258.

Schlinger BA, Arnold AP. Circulating estrogens in a malesongbird originate in the brain. Proc. Natl. Acad. Sci. USA1992; 89: 7650–7653.

Schulz RW, França LR, Lareyre JJ, LeGac F, Chiarini-GarciaH, Nobrega RH, Miura T. Spermatogenesis in fish. Gene.Comp. Endocrinol. 2010; 165: 390–411.

Scott AP, Sumpter JP. Seasonal variations in testicular germcell stages and in plasma concentrations of sex steroids inmale rainbow trout (Salmo gairdneri) maturing at 2 yearsold. Gen. Comp. Endocrinol. 1989; 73: 46–58.

Shinohara A, Shinohara M. Roles of RecA homologuesRad51 and Dmc1 during meiotic recombination.Cytogenet. Genome Res. 2004; 107: 201–207.

Smedley PL, Kinniburgh DG. A review of the source, be-haviour and distribution of arsenic in natural waters. Appl.Geochem. 2002; 17: 6517–568.

Smith KH, Nicolas A. Recombination at work for meiosis.Curr. Opin. Genet. Dev. 1998; 8: 200–211.

Steinberger E. Hormonal control of mammalian sperma-togenesis. Physiol. Rev. 1971; 51: 1–22.

Suzuki A, Shibata N. Developmental process of genital ductsin the medaka, Oryzias latipes. Zool. Sci. 2004; 21: 397–406.

Swanson P, Bernard M, Nozaki M, Suzuki K, Kawauchi H,Dickhoff WW. Gonadotropins I and II in juvenile cohosalmon. Fish Physiol. Biochem. 1989; 7: 169–176.

Swanson P, Suzuki K, Kawauchi H, Dickhoff WW. Isola-t ion and characterization of two coho salmongonadotropins, GTHI and GTHII. Biol. Reprod. 1991; 44:29–38.

Thomas P, Zhu Y, Pace M. Progestin membrane receptorsinvolved in the meiotic maturation of teleost oocytes: areview with some new findings. Steroids 2002; 67: 511–517.

Thomas P, Tubbs C, Detweiler C, Das S, Ford L,Breckenridge-Miller D. Binding characteristics, hormo-nal regulation and identity of the sperm membrane pro-

gestin receptor in Atlantic croaker. Steroids 2005; 70: 427–433.

Thomas P, Dressing G, Pang Y, Berg H, Tubbs C,Benninghoff A, Doughty K. Progestin, estrogen and an-drogen G-protein coupled receptors in fish gonads. Ster-oids 2006; 71(4): 310–316.

Ueda H, Kanbegawa A, Nagahama Y. Involvement of gona-dotrophin and steroid hormones in spermiation in theamago salmon, Oncorhynchus rhodurus, and goldfish,Carassius auratus. Gen. Comp. Endocrinol. 1985; 59: 24–30.

Van der Kraak G, Suzuki K, Peter RE, Itoh H, Kawauchi H.Properties of common carp gonadotropin I and gonado-tropin II. Gen. Comp. Endocrinol. 1992; 85: 217–229.

Walter C, Intano G, Mccarrey J, Mcmahan C, Walter R.Mutation frequency declines during spermatogenesis inyoung mice but increases in old mice. Proc. Natl. Acad.Sci. USA 1998; 95: 10015–10019.

Watanabe A, Onitake K. Changes in the distribution offibroblast growth factor in the teleostean testis during sper-matogenesis. J. Exp. Zool. 1995; 272: 475–483.

Watanabe Y. Sister chromatid cohesion along arms and atcentromeres. Trends. Genet. 2005; 21: 405–412.

Winn RN, Norris MB, Brayer KJ, Torres C, Muller SL. De-tection of mutations in transgenic fish carrying a bacteri-ophage lambda cII transgene target. Proc. Natl. Acad. Sci.USA 2000; 97: 12655–12660.

Wong GW, Li L, Madhusudhan MS, Krilis SA, Gurish MF,Rothenberg ME, Sali A, Stevens RL. Tryptase 4, a newmember of the chromosome 17 family of mouse serineproteases. J. Biol. Chem. 2001; 276: 20648–20658.

Yamaguchi S, Miura C, Ito A, Agusa T, Iwata H, Tanabe S,Tuyen BC, Miura T. Effects of lead, molybdenum, rubid-ium, arsenic and organochlorines on spermatogenesis infish: monitoring at Mekong Delta area and in vitro ex-periment. Aqua. Toxicol. 2007; 83: 43–51.

Yamaguchi S, Miura C, Kikuchi K, Celino FT, Agusa T,Tanabe S, Miura T. Zinc is an essential trace element forspermatogenesis. Proc. Natl. Acad. Sci. USA, 2009; 106:10859–10864.

Yanagimachi R, Cherr GN, Pillai MC, Baldwin JD. Factorscontrolling sperm entry into the micropyles of salmonidand herring eggs. Develop. Growth Differ. 1992; 34(4):447–461.

Yoshiura Y, Suetake H, Aida K. Duality of gonadotropin ina primitive teleost, Japanese eel (Anguilla japonica). Gen.Comp. Endocrinol. 1999; 114: 121–131.

Zhao GQ, Deng K, Labosky PA, Liaw L, Hogan BLM. Thegene encoding bone morphogenetic protein 8B (BMP8B)is required for the initiation and maintenance of sperma-togenesis in the mouse. Genes. Dev. 1996; 10: 1657–1669.

analysis of spermatogenesis using an eel model

Documents