Amh) Relative to Sox9a, Sox9b, and Cyp19a1a, During Gonad Development

Gene Expression Patterns 5 (2005) 655–667 www.elsevier.com/locate/modgep

Characterization and expression pattern of zebraﬁsh anti-Mu¨llerian hormone (amh) relative to sox9a, sox9b, and cyp19a1a, during gonad development

Adriana Rodrı´guez-Marı´, Yi-Lin Yan, Ruth A. BreMiller, Catherine Wilson, Cristian Can˜estro, John H. Postlethwait*

Institute of Neuroscience, University of Oregon, 1425 E. 13th Avenue, Eugene, OR 97403, USA

Received 28 January 2005; received in revised form 28 February 2005; accepted 28 February 2005 Available online 19 April 2005

Abstract

The role of Anti-Mu¨llerian hormone (Amh) during gonad development has been studied extensively in mammals, but is less well understood in other vertebrates. In male mammalian embryos, Sox9 activates expression of Amh, which initiates the regression of the Mu¨llerian ducts and inhibits the expression of aromatase (Cyp19a1), the enzyme that converts androgens to estrogens. To better understand shared features of vertebrate gonadogenesis, we cloned amh cDNA from zebrafish, characterized its genomic structure, mapped it, analyzed conserved syntenies, studied its expression pattern in embryos, larvae, juveniles, and adults, and compared it to the expression patterns of sox9a, sox9b and cyp19a1a. We found that the onset of amh expression occurred while gonads were still undifferentiated and sox9a and cyp19a1a were already expressed. In differentiated gonads of juveniles, amh showed a sexually dimorphic expression pattern. In 31 days post-fertilization juveniles, testes expressed amh and sox9a, but not cyp19a1a, while ovaries expressed cyp19a1a and sox9b, but not amh.In adult testes, amh and sox9a were expressed in presumptive Sertoli cells. In adult ovaries, amh and cyp19a1a were expressed in granulosa cells surrounding the oocytes, and sox9b was expressed in a complementary fashion in the ooplasm of oocytes. The observed expression patterns of amh, sox9a, sox9b, and cyp19a1a in zebrafish correspond to the patterns expected if their regulatory interactions have been conserved with mammals. The finding that zebrafish sox9b and sox8 were not co-expressed with amh in oocytes excludes the possibility that amh expression in zebrafish granulosa cells is directly regulated by either of these two genes. q 2005 Elsevier B.V. All rights reserved.

Keywords: Anti-mullerian hormone; Amh; Mullerian inhibiting substance; Mis; TGF-beta; Teleost; Zebraﬁsh; Gonad development; Gonad differentiation; Sox9; Sox9a; Sox9b; Cyp19a1; Cyp19a1a; Aromatase; Cytochrome P450 aromatase; Sox8; Testis; Testes; Sertoli cells; Ovary; Ovaries; Oocyte; Follicle cells; Granulosa cells; Theca cells; Synteny; Sex determination

1. Results and discussion In concert with other factors, including Sf1, Wt1, and Gata4 (Shen et al., 1994; de Santa Barbara et al., 1998; Mammalian embryos initially develop an indifferent Nachtigal et al., 1998; Viger et al., 1998; de Santa gonad that has the potential of becoming an ovary or a testis Barbara et al., 2000), Sox9 activates the Anti- (see for recent reviews (Brennan and Capel, 2004; Park and Mu¨llerian Hormone gene (Amh)(di Clemente et al., 1992; Jameson, 2005)). The Y-chromosome gene Sry stimulates Kent et al., 1996; Morais da Silva et al., 1996; de Santa early testis development (Gubbay et al., 1990; Sinclair et al., Barbara et al., 1998; Rouiller-Fabre et al., 1998; de Santa 1990) and activates the expression of the transcription factor Barbara et al., 2000; Lasala et al., 2004), also known as Sox9 inSertolicellprecursors(Sekido et al., 2004). Mu¨llerian-inhibiting substance (Mis), which encodes a glycoprotein member of the transforming growth factor- beta (TGF-beta) superfamily (Cate et al., 1986; Picard et al., * Corresponding author. Tel.: C1 541 346 4538; fax: C1 541 346 4548. 1986). Amh is produced by Sertoli cells in fetal testes and E-mail address: [email protected] (J.H. Postlethwait). causes the regression of the Mu¨llerian ducts, which in 1567-133X/$ - see front matter q 2005 Elsevier B.V. All rights reserved. females differentiate into Fallopian tubes and uterus doi:10.1016/j.modgep.2005.02.008 (Munsterberg and Lovell-Badge, 1991; Josso et al., 1993). 656 A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667

Amh also blocks the ability of cAMP and FSH to cause to that of the zebrafish orthologs of mammalian genes known rat fetal testis to express Cyp19a1, the P450 aromatase to be upstream (Sox9) and downstream (Cyp19a1) of Amh. enzyme that converts androgens to estrogens (Rouiller-Fabre et al., 1998). In accord with these regulatory interactions 1.1. Isolation, characterization, and mapping (Sox9/Amh-jjCyp19a1), the onset of Sox9 expression of zebrafish amh precedes Amh expression in mammalian testes. In contrast to mammals, Amh expression precedes Sox9 expression in To identify a zebrafish ortholog of the mammalian Amh the indifferent gonad of chicken and alligator embryos, gene, we performed a tBLASTn (Altschul et al., 1997)search suggesting that different classes of vertebrates differ in the of the zebrafish genome database (Sanger Institute) using the regulatory interactions of these genes (Smith et al., 1999; mouse AMH protein (NP_031471) as query. The best Western et al., 1999; Oreal et al., 2002). zebrafish hit (ENSDART00000013803) returned Amh as the In addition to the early role in the mammalian bipotential most similar protein among mammals in a BLASTx search of gonad, Amh plays a later role in negatively regulating the NCBI nr database. We used the ENSDART00000013803 gonadal development in both sexes. In females, Amh plays sequence to design RACE primers for amplifying the cDNA an important role in folliculogenesis (Durlinger et al., 2002). sequence of the zebrafish amh gene and submitted the After Mu¨llerian ducts have lost sensitivity to Amh action in sequence to GenBank (AY881649). This sequence provides late fetal life, ovarian granulosa cells produce Amh, thereby a 2275 bp mRNA which includes the complete coding inhibiting the transcription of aromatase (Cyp19a1) and LH sequence for the zebrafish Amh protein. Comparing the receptor (Vigier et al., 1989; Josso et al., 1998). Females mRNA to its genomic sequence (Sanger Institute) revealed maintain high levels of Amh throughout reproductive life that the zebrafish amh gene consists of seven exons and spans a until menopause, so serum levels of Amh provide a good genomic region of 13.6 kb (Fig. 1A). Human AMH marker of ovarian follicle reserve and fertility potential contains just five exons (Cate et al., 1986), and the difference (Fanchin et al., 2005). In males, Amh also influences the in exon number is due to the presence of two additional introns development of adult testes by blocking the differentiation (intron 1 and intron 6) in the zebrafish gene (Fig. 1A). of mesenchymal into Leydig cells and by decreasing the Comparative analysis of the Amh promoter among expression of steroidogenic enzymes (Josso et al., 1998). mammals revealed the presence of conserved binding sites In contrast to mammals, the molecular genetic regulation for Sox9, SF-1 and GATA proteins (Arango et al., 1999; of gonad differentiation in other vertebrates remains largely Watanabe et al., 2000). Our analysis of the zebrafish amh unknown. In fishes, only basally diverging ray-fin fish, such promoter showed the presence of a putative TATA box at as sturgeon, have Mu¨llerian ducts, and these do not undergo -67 nt relative to the ATG start codon, and in accordance regression in males (Wrobel, 2003). Teleost fish, like eel, with consensus sequences described in mammals (Arango zebrafish, salmon, and flounder, do not have Mu¨llerian ducts et al., 1999; Watanabe et al., 2000), we found several (Lasala et al., 2004). Despite the absence of Mu¨llerian ducts, putative transcription factor binding sites: Sox9 binding a gene coding for a protein with high similarity to Amh, sites (TTTGAG) at K865, K3588, and K3832 nt; GATA originally named spermatogenesis preventing substance, has binding sites (TGATAG) at K1752 and K3515 nt; and been recently sequenced in three teleosts: Japanese eel SF-1 binding site (CCAAGGACAC) at K123 nt, and SF-1 (BAB93107), Japanese flounder (BAD37138) and Atlantic binding sites (AGGTCA) at K2074 and K3399 nt. Future salmon (AAU85130). In Japanese eel, expression of this gene experiments are required to test whether these putative was reported in Sertoli cells of immature testes, where it transcription factor binding sites shown to be functional in prevents the initiation of spermatogenesis, but its expression mammals are active in zebrafish. was not investigated in ovaries (Miura et al., 2002). In To determine the genomic location of amh, we mapped Japanese flounder, Amh expression was detected in Sertoli the sequence on both the HS meiotic mapping panel cells in testes but not in ovaries (Yoshinaga et al., 2004). (Woods et al., 2000) and the radiation hybrid mapping In this work, we investigate the poorly understood process panel LN54 (Hukriede et al., 1999). These experiments of gonadal differentiation in zebrafish. Zebrafish, like many revealed that zebrafish amh maps to LG22. There were no other fish species, do not have sex chromosomes (Schreeb et recombinants between amh and markers AI396840, oaz1, al., 1993; Pijnacker and Ferwerda, 1995; Amores and and 34 other markers previously located at 38.9 cM on Postlethwait, 1999). Juvenile zebrafish first develop an LG22 on the HS mapping panel (Fig. 1B). In the current undifferentiated ovary-like gonad containing immature Sanger database, however, zebrafish amh and oaz1 are oocytes. In females, these oocytes become mature ova, assembled separately in LG7 and LG22, respectively. while in males, they disappear by apoptosis by 30 days post- Because our two independent mapping experiments on hatching as gonads differentiate into testes (Takahashi, 1977; two very different mapping panels agreed that amh and oaz1 Uchida et al., 2002). To help better understand the genetic are located very near each other on LG22, we conclude that mechanisms underlying gonad development, we isolated, the LG22 location is correct. characterized and studied the expression pattern of The zebrafish amh gene is on the genomic contig the zebrafish amh gene, and compared its expression pattern Zv4_scaffold578 from Sanger Institute, which contains A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667 657

Fig. 1. Genomic analysis of zebrafish amh. (A) Comparison of exon-intron organizations of zebrafish amh (Dre) and human AMH (Hsa) genes. Exons are represented by numbered boxes, and grey areas indicate untranslated regions. The correspondence of intron-exon boundaries between zebrafish and human genes is indicated by dashed lines. Scale bar is 1 kb. (B) Schematic representation of the conserved synteny in the amh genomic regions of human, fugu and zebrafish. The zebrafish amh gene is on genomic scaffold Zv4_Scaf578, which contains the four neighbor genes (dot1l, ell, fkbp8 and ssbp4), and maps to LG22 together with the oaz1 gene. All these genes are contained in a single genomic scaffold of Fugu (FruV3_Scaf303), and all of them map to human chromosome 19p13.3-1.

ENSDART00000013803, ENSDART00000013892, zgc: cleavage sites (R/K–XX–R/K) (arrowheads in Fig. 2A) 63836, zgc:77672, and ENSDARESTT00000021285, (Duckert et al., 2004). Protein alignment shows that the which are respectively zebrafish orthologs of the human cleavage site demonstrated for human and bovine forms genes AMH, DOT1L, ELL, FKBP8, and SSBP4 (using the (arrow in Fig. 2A) (Cate et al., 1986) aligns with the first of reciprocal best hit criterion (Wall et al., 2003)), and all these the two putative cleavage sites in zebrafish. Finally, analysis human genes map to human chromosome 19p13.3K1 of cysteine distribution throughout the zebrafish Amh protein (Fig. 1B). Comparative analysis of zebrafish and Takifugu shows that 9 out of the 13 cysteines are in conserved positions rubripes genomic regions containing Amh revealed con- among most vertebrates (asterisks in Fig. 2A). The finding of served syntenies including four neighboring genes, Dot1L, this overall high preservation of functional motifs in Ell, Fkbp8 and Ssbp4 (Fig. 1B). Overall, our genomic zebrafish Amh is consistent with conserved molecular comparative analysis among zebrafish, fugu, and human properties with other vertebrate Amh proteins. suggests that the chromosome segment containing amh has Phylogenetic analysis of related members of the TGF- been conserved for 450 million years since the last common beta superfamily shows that zebrafish Amh clusters within ancestor of ray-fin fishes and mammals. the clade of other fish Amh proteins with high bootstrap support, well separated from other members of the TGF- beta superfamily (Fig. 2B). This analysis shows that the 1.2. Sequence and phylogenetic analysis of the Amh protein gene we have identified has the phylogenetic properties and structural features expected for the zebrafish ortholog of The zebrafish amh gene encodes a protein of 549 amino human AMH. acids that contains the two domains characteristic of the Amh protein family: the TGF-beta domain (TGFB) at the C-terminus and the Amh domain (AMH_N) in the middle 1.3. Expression of amh during gonad development of the protein (Fig. 2A); the N-terminus of the protein is much in zebrafish less well conserved. Zebrafish AMH_N and TGFB domains are more similar to the orthologous regions in other fishes To determine the expression pattern of amh during (42 and 58% identical, respectively) than to the correspond- development, we conducted in situ hybridization exper- ing regions of tetrapod Amh genes (20 and 41%), as expected iments on whole-mount embryos (1 and 2 days post- from organism phylogenies. Sequence analysis of the region fertilization (dpf)) and on sections of larvae (3, 5, 7, 9, 11, between the TGFB and AMH_N domains of zebrafish Amh, 17, 18, 20 and 21 dpf), juveniles (31, 34 and 38 dpf), and reveals the presence of two consecutive putative plasmin adults (6, 8, and 12 months old). 658 A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667

Fig. 2. Comparative analysis of zebrafish Amh. (A) Protein alignment of zebrafish Amh with other related members of the Amh family. Conserved positions with the same amino acid (black) or similar residues (grey) in more than 70% of the sequences are shadowed. The TGF-beta domain at the C-terminus (TGFB in black under bar) and the AMH domain (AMH_N in grey under bar) that characterize this protein family are labeled. Putative consensus cleaving target sites (arrowheads in R/K-XX-R/K) and the cleavage site in the human AMH (arrow) by which the active TGF- beta domain is released are indicated. Each of the eight positions with conserved cysteines is labeled with an asterisk. (B) Protein sequence relationships between the Amh family and other protein families belonging to the TGF-beta superfamily are shown in a phylogenetic tree. The branch lengths are drawn to scale to the evolutionary distance based on Poison-corrected Neighbor-Joining method, and the numbers are the percentage of bootstrap values supporting each node from 500 replicas. Scale bar, 0.2 substitutions per site. Aja, Anguilla japonica (Japanese eel); Dre, Danio rerio (zebrafish); Gga, Gallus gallus (chicken); Hsa, Homo sapiens (human); Mmu, Mus musculus (mouse); Pol, Paralichthys olivaceus (Japanese flounder); Ssa, Salmo salar (Atlantic salmon). A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667 659

Expression of amh was not detected in embryos of 1 and decreased and was detected only in a reduced number of 2 dpf or in larvae from 3 to 11 dpf (data not shown). We granulosa cells in early stage III oocytes (purple arrowhead detected low levels of amh expression in undifferentiated in Fig. 3H), characterized by little yolk accumulation. In late gonads for the first time in 17 dpf larvae (Fig. 3A), and it stage III oocytes or later stages, expression of amh in gradually increased in 18 dpf, 20 dpf (data not shown), and granulosa cells was not further detected (green arrowhead in 21 dpf animals (Fig. 3B). Fig. 3H and data not shown). Gonads were obviously differentiated into testes or Our results in zebrafish differ from the expression of amh ovaries in 31 dpf juveniles, and in these animals, amh in Japanese flounder, which was not detected in the ovary expression showed a clear sexually dimorphic pattern by RT-PCR or by Northern blot analysis, and was (Fig. 3C,D). At this age, while high levels of amh expression described as ‘barely detectable’ by in situ hybridization were detected in male gonads (Fig. 3C), amh expression was (Yoshinaga et al., 2004). In contrast, the zebrafish data are not detected over background in female gonads (Fig. 3D). consistent with the expression of AMH in human ovarian This sexually dimorphic amh expression pattern was granulosa cells from the fetal stage to the end of maintained during gonad development in 34 dpf juveniles reproductive life (Lasala et al., 2004). Despite histological (data not shown), and 38 dpf juveniles (Fig. 3E,F). differences between the ovaries of zebrafish and humans, In contrast to juveniles, adults showed expression of amh their common expression pattern of amh in granulosa cells in both male and female gonads (Fig. 3G–J). In adult male is expected if the amh gene already had this expression testes, amh expression was high in cells surrounding the pattern in the last common ancestor of zebrafish and human. cysts in positions expected for Sertoli cells (Fig. 3G,I). This localization agrees with the expression pattern of amh in 1.4. Expression of amh relative to other gonadal genes Sertoli cells described in eel and flounder (Miura et al., 2002; Yoshinaga et al., 2004) and it is consistent with the In mammalian testes, Sox9 is an upstream positive expression of AMH in human Sertoli cells from early fetal regulator of Amh, which in turn is an upstream stage until puberty, when AMH expression starts to decrease negative regulator of Cyp19a1 (di Clemente et al., 1992; due to the action of testosterone (see for review (Lasala Kent et al., 1996; Morais da Silva et al., 1996; de Santa et al., 2004)). Barbara et al., 1998; Rouiller-Fabre et al., 1998; de Santa In zebrafish ovaries, oocytes pass through several Barbara et al., 2000). Although the mammalian ovary does developmental stages (Selman et al., 1993). Briefly, in the not express Sox9, and the regulation of Amh in mammalian initial pre-follicle phase, developing oocytes reside in a nest ovary is not yet fully understood, Amh is nevertheless an of oocytes surrounded by a single layer of pre-follicle cells upstream negative regulator of Cyp19a1 in ovary, as it is in (stage IA). As development proceeds, individual oocytes testis. (Picard et al., 1986; Vigier et al., 1989; di Clemente leave the nest, the germinal vesicle becomes visible, and et al., 1992). follicle (granulosa) cells and a vascularized connective In previous work, we examined the expression of P450 tissue layer containing theca cells surround the oocyte (stage aromatase (cyp19a1)(Chiang et al., 2001b) and sox9 genes IB). As the oocyte grows, it accumulates cortical alveoli (Chiang et al., 2001a) in adult zebrafish gonads, but there (stage II), which are displaced to the cortex as yolk has as yet been no analysis of the expression of these genes accumulates (stage III). Finally, the oocyte enlarges and across many developmental stages. Teleosts have dupli- matures (stage IV), and becomes a mature oocyte, the egg, cated copies of both of these tetrapod genes due to a genome ready for ovulation (stage V). In histological sections, it is duplication event in teleost phylogeny (Amores et al., 1998; difficult to distinguish late stage III oocytes from early stage Postlethwait et al., 1998; Vogel, 1998; Wittbrodt et al., IV oocytes because there are no structural markers that 1998; Meyer and Schartl, 1999; Postlethwait et al., 2000; indicate this transition (Selman et al., 1993). Postlethwait et al., 2002; Taylor et al., 2003; Amores et al., In adult female zebrafish, amh expression in the ovary 2004; Jaillon et al., 2004; Koopman et al., 2004). In addition was confined to granulosa cells of specifc stages of to the expression of sox9a and sox9b in multiple tissues oogenesis (Fig. 3H,J). Stage IA oocytes and most of the during embryonic and larval development in zebrafish, in oocytes of stage IB did not express amh. Some oocytes of adults, sox9a is expressed in testes, and sox9b is expressed stage IB, presumably late stage IB because they were large in ovaries (Chiang et al., 2001a; Li et al., 2002; Yan et al., but still lacked observable alveoli in DIC microscopy, 2002; Liu et al., 2003). Except for the ovarian expression of showed few granulosa cells expressing amh (orange arrow- sox9b, the sum of the sox9a and sox9b expression patterns head in Fig. 3H). As oocytes entered stage II with obvious generally matches the mouse pattern (Wright et al., 1995; cortical alveoli, they were surrounded with a monolayer of Akiyama et al., 2002). As with sox9, zebrafish has two genes granulosa cells that had accumulated substantial quantities for aromatase. The cyp19a1a gene is expressed primarily in of amh transcript (red arrowhead in Fig. 3H), but we the zebrafish ovary, and cyp19a1b is expressed primarily in detected no amh expression in the layer of theca cells (tc) the brain (Callard et al., 2001; Chiang et al., 2001b; Chiang surrounding the granulosa cell layer (gc)(Fig. 3J). As et al., 2001c; Kazeto et al., 2001; Kishida and Callard, 2001; oocytes transitioned to stage III, expression of amh Kishida et al., 2001; Tchoudakova et al., 2001; Tong et al., 660 A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667

Fig. 3. Expression pattern of amh during zebraﬁsh gonad development. In situ hybridization was performed on sections of (A,B) undifferentiated gonads; (C,E,G,I) male gonads; (D,F,H,J) female gonads. (A) The earliest amh expression we observed was weakly detected transcript in 17 days post-fertilization (dpf) larvae. (B) Amh expression became more intense in the still undifferentiated gonads of 21 dpf larvae. (C,D) In 31 dpf juveniles, gonads were already A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667 661

2001; Trant et al., 2001; McAllister and Kime, 2003; Tong (Picard et al., 1986; Vigier et al., 1989; di Clemente et al., and Chung, 2003; Fenske and Segner, 2004; Goto-Kazeto 1992; Rouiller-Fabre et al., 1998). These results are et al., 2004; Kazeto et al., 2004; Uchida et al., 2004); the consistent with the experiments of Fenske and Segner single CYP19A1 gene of humans is expressed in both the (2004) and Uchida et al. (2004) in zebraﬁsh, which ovary and the brain (Simpson et al., 1993; Simpson et al., demonstrate that blocking aromatase activity with fadrozole 1994). causes testis development and results in cyp19a1a down- To investigate the expression patterns of sox9a, sox9b, regulation in gonads. Together with our results, this and cyp19a1a relative to amh in zebraﬁsh, we performed in supports the hypothesis that the inhibition of aromatase is situ hybridization on adjacent sections with different probes involved in the transformation of undifferentiated gonads using 17 dpf larvae, 31 dpf and 38 dpf juveniles, and adults. into testes (Vigier et al., 1989).

1.4.1. Expression of amh, sox9a and cyp19a1a 1.4.3. Expression of amh, sox9b and cyp19a1a in ovaries in undifferentiated gonads In ovaries of 31 and 38 dpf juveniles, amh expression The earliest age at which we detected distinct, but low was not detected, sox9b expression was detected at low levels, of amh expression in undifferentiated gonads was levels in the ooplasm of some oocytes, and cyp19a1a 17 dpf (Fig. 4A). In contrast, by this age the expression of expression was detected in cells surrounding oocytes sox9a (Fig. 4B) and cyp19a1a (Fig. 4C) was strong in (Fig. 4M–R). restricted regions of these undifferentiated gonads, and both In adult ovaries, amh, sox9b and cyp19a1a were markers were expressed in the same gonad in the same expressed in a highly dynamic pattern, showing variations individual. The expression of cyp19a1a would be expected among different cell types and stages of oocyte maturation if the undifferentiated gonad is ovary-like, as suggested (Fig. 4S–W). During oocyte maturation, amh and cyp19a1a previously (Takahashi, 1977; Uchida et al., 2002), but the shared a pattern of increasing expression levels in granulosa strong expression of sox9a and initial expression of amh at cells from (i) no expression in early stage IB (yellow 17 dpf, would be compatible with the expression of Sox9 arrowheads in Fig. 4S,U), (ii) a small number of expressing and Amh in early testes development in mammals (see for cells in late stage IB (orange arrowheads in Fig. 3H and recent review (Lasala et al., 2004)). From these consider- Fig. 4U), and (iii) expression in most of the granulosa cells ations, we conclude that 17 dpf represents a transitional surrounding stage II oocytes (red arrowheads in Fig. 4S,U). stage in zebrafish gonad differentiation. By 31 dpf, gonads By early stage III, however, amh and cyp19a1a began had already differentiated into testes and ovaries. opposite expression dynamics, and while amh expression was down-regulated, cyp19a1a expression was up-regu- 1.4.2. Expression of amh, sox9a and cyp19a1a in testes lated. Consequently, we observed in late stage III oocytes In testes of 31 dpf juveniles, expression of amh and that the absence of amh transcripts in granulosa cells sox9a was detected at high levels, but cyp19a1a was not coincided with the maximum accumulation of cyp19a1a (Fig. 4D–F). Juveniles at 38 dpf maintained the same transcripts (green arrowheads in Fig. 4S,U). The high expression pattern as at 31 dpf (Fig. 4G–I). In adult testes, resolution of our in situ hybridization analysis provides amh and sox9a were both expressed in cells surrounding the detail at the cellular level supporting previous results using cysts, in positions expected for Sertoli cells (Fig. 4J,K). This autoradiographic expression analysis of cyp19a1a in adult tissue expression pattern is in agreement with the interaction ovaries (Goto-Kazeto et al., 2004). In late stage IV oocytes, described in human testes where SOX9 activates AMH in amh transcripts continued to be undetectable, and cyp19a1a Sertoli cells (Kent et al., 1996; Morais da Silva et al., 1996; expression was down-regulated (data not shown). Sekido et al., 2004). In contrast to amh and sox9a,no Our observation that cyp19a1a expression increases expression of cyp19a1a was detected in adult zebrafish while amh expression decreases in granulosa cells during testes (Fig. 4L). These results reveal that during the process the maturation of stage III oocytes, and the fact that the in which undifferentiated gonads develop into testes, maximum amount of cyp19a1a transcripts coincides with cyp19a1a (aromatase) expression decreases, while amh the absence of amh expression in late stage III, is expression increases, in accord to the mammalian regulatory compatible with previous observations in tetrapods in scheme, in which Amh negatively regulates Cyp19a1 which amh inhibits Cyp19a1 in ovaries (Vigier et al., 3 differentiated into testes and ovaries. At this time, amh expression was detected only in testes (C) and was not detected in ovaries (D). (E,F) In 38 dpf larvae, as gonad development proceeded, amh maintained the same sexually dimorphic expression pattern. (G–J) In contrast to larval and juvenile stages, adults showed amh expression in both testes and ovaries. In males, amh expression was detected only in presumptive Sertoli cells (G). A higher magnification of a region expressing amh (dashed box) is shown in (I). In adult ovaries, amh expression differed among oocytes of different stages (IB, II, III) (H). Expression of amh was mostly detected in the monolayer of granulosa cells (gc) of Stage II oocytes (red arrowhead in H), but no expression of amh was detected in Theca cells (tc). Only reduced number of amh expressing cells could be detected in late stage IB oocytes (orange arrowhead) or early stage III oocytes (purple arrowhead), and no expression was detected in late stage III oocytes (green arrowhead). A higher magnification of a region expressing amh (dashed box) is shown in (J). Scale bars are indicated in each panel. ca, cortical alveoli; gc, granulosa cells; gv, germinal vesicle; o, ooplasm; tc, theca cells; y, yolk. Colored arrowheads label different stages of oocytes: orange, late stage IB; red, stage II; purple, early stage III; green, late stage III. 662 A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667

Fig. 4. Comparative analysis of the expression pattern of amh, sox9a, sox9b, cyp19a1a and sox8 during zebraﬁsh gonad development. (A–C) undifferentiated gonads; (D–L) male gonads; (M–X) female gonads. To facilitate comparison among the expression patterns of amh (A,D,G,J,M,P,S,V), sox9a (B,E,H,K), sox9b (N,Q,T,W), cyp19a1a (C,F,I,L,O,R,U), and sox8 (X), in situ hybridization experiments were performed on adjacent sections. (A–C) The undifferentiated A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667 663

1989; di Clemente et al., 1992). However, this opposite sequenced the full-length cDNA of the zebrafish sox8 gene expression dynamics of cyp19a1a and amh observed (AY883015). during stage III contrasts with the simultaneous early In situ hybridization analyses on sections of adult female increase of cyp19a1a and amh expression during stage IB zebrafish showed that in the ovary, sox8 was detected in the and II. This could indicate that rather high levels of Amh ooplasm of stage IB and stage II oocytes, but not in granulosa are necessary to inhibit cyp19a1a expression in adult cells (Fig. 4X). These results reveal that Sox8 is not likely to granulosa cells. The down-regulation of cyp19a1a occurs be directly regulating amh in zebrafish ovarian granulosa in the absence of transcription of amh at late stage IV, cells, and this finding is in agreement with recently published suggesting that some factor other than Amh must be studies in mouse, where Sox8 and Amh are not co-expressed effecting the down-regulation of cyp19a1a in these late- in the granulosa cells of pre-antral follicles (Salmon et al., stage oocytes. These observations reveal a non-linear 2005). relationship between the transcription of amh and Since our results showed that neither sox9a, sox9b, cyp19a1a and will require additional experiments to nor sox8 is co-expressed with amh in granulosa cells, we understand fully the mechanisms regulating the conclude that some other factor, perhaps one or more of expression of cyp19a1a in zebrafish ovary. the transcription factors that have been described to act In contrast to the expression of amh and cyp19a1a in in concert with Sox9 in mammals (e.g. Sf1, Gata4, Wt1), granulosa cells, sox9b expression was restricted to the is responsible for activating amh in zebrafish granulosa ooplasm of oocytes (Fig. 4T,W). While high expression cells. of sox9b was detected in the ooplasm of stages IB and II In conclusion, we cloned the zebrafish anti-Mu¨llerian oocytes (yellow and red arrowheads in Fig. 4T, hormone gene, and characterized its cDNA and genomic respectively), its expression was down-regulated by structure. Comparative analysis revealed that essential stage III (green arrowhead in Fig. 4T). The finding that structural features of amh and its genomic neighborhood zebrafish amh and sox9b are not co-expressed in have been maintained for 450 million years since the granulosa cells (Fig. 4V,W) excludes the possibility divergence of lineages leading to zebrafish and mammals. that Sox9b is directly regulating amh expression in this At the onset of amh expression, sox9a and cyp19a1a cell type. already showed strong expression in undifferentiated gonads. As testes developed, expression of amh and sox9a was strong, and expression of cyp19a1a was not 1.4.4. Is Sox8 regulating amh expression in zebrafish detectable. These results are consistent with the gene granulosa cells? interactions shown in mammals in which Sox9 positively Mouse SOX8, a gene closely related to SOX9, has been regulates Amh, which negatively regulates Cyp19a1. In described to reinforce SOX9 function in testis development adult ovaries, amh and cyp19a1a were expressed in (Chaboissier et al., 2004), and it has been shown to activate granulosa cells, while sox9b was expressed in the the Amh promoter in vitro, although less efficiently than ooplasm of oocytes. The expression in different cell- SOX9 (Schepers et al., 2003). Since our results showed types of sox9a, sox9b, and sox8, compared to amh ruled that zebrafish Sox9b was not directly regulating amh in out the hypothesis that sox9a, sox9b, or sox8 directly granulosa cells of ovaries, we wondered whether the regulates amh expression in granulosa cells of adult zebrafish Sox8 ortholog was playing this role. Our previous zebrafish ovaries. These results show a striking similarity work identified zebrafish sox8 sequences (Chiang et al., in the expression dynamics of gonad differentiation 2001a; Cresko et al., 2003) and for the present work, we regulators in zebrafish and mammals.

3 gonads of 17 day dpf larvae showed amh expression at low levels but already showed high expression of sox9a and cyp19a1a. (D–F) In 31 dpf juveniles, gonads had already differentiated. At this time in males, expression of amh and sox9a, but no cyp19a1a, was detected in testes. (G–L) As development of the gonads proceeded, the same expression pattern as in 31 dpf was maintained in later juvenile stages and adults, in which amh and sox9a expression could be observed in presumptive Sertoli cells, and cyp19a1a expression was not detected. (M–R) In gonads of 31 and 38 dpf female juveniles, no amh expression could be detected (M,P), sox9b was expressed at low levels in the ooplasm of some oocytes (N,Q) and cyp19a1a was expressed in cells surrounding the oocytes (O,R). (S–U) In adult female ovaries, amh expression was detected in granulosa cells of stage II oocytes (red arrowhead in S), but no expression was detected in early stage IB oocytes (yellow arrowhead in S) or late stage III oocytes (green arrowhead in S). Expression of sox9b was detected in the ooplasm of stage IB oocytes (yellow arrowhead in T) and stage II oocytes (red arrowhead in T) but no expression was detected in late stage III oocytes (green arrowhead in T). Regions in dashed boxes in (S) and (T) are shown in higher magniﬁcation in (V) and (W), respectively. Expression of cyp19a1a was not detected in early stage IB oocytes (yellow arrowhead in U), some granulosa cells were expressing cyp19a1a in late stage IB oocytes (orange arrowhead in U), most of the granulosa cells expressed cyp19a1a in stage II oocytes (red arrowhead in U), and maximum cyp19a1a expression was observed in late stage III oocytes (green arrowhead in U). (V–X) Comparison of amh, sox9b and sox8 expressions in stage II oocytes revealed that while amh was expressed in granulosa cells (gc) (V), sox9b and sox8 were expressed in the ooplasm (o) of oocytes (W,X). The main image in (X) is an enlargement of the boxed region of the inset, which shows a lower magniﬁcation of a complete view of a stage II oocyte. Roman numbers and letters indicate the stage of the oocytes according to (Selman et al., 1993). Scale bars: 20 mm (A–C as in A), 50 mm (D–L as in D), 50 mm (M–U as in M), 20 mm (V–X as in V). Colored arrowheads label different stages of oocytes: yellow, early stage IB; orange, late stage IB; red, stage II; green, late stage III. 664 A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667

2. Experimental procedures 2.3. Protein alignment and phylogenetic analysis

Protein alignments were generated with ClustalX 2.1. Cloning amh (Thompson et al., 1997). Alignment of the TGF-beta domain was used to construct phylogenetic trees with the To determine whether zebrafish possesses a gene MEGA package (Kumar et al., 2001) based on Poisson- orthologous to mammalian Amh, we searched the zebrafish corrected Neighbor-Joining amino acid distances and genome database (Sanger Institute, Ensembl: http://www. considering gap pairwise deletion. Protein accession ensembl.org/Danio_rerio) by tBLASTn comparison numbers used for phylogenetic analysis: AMH: Zebrafish (Altschul et al., 1997) with the mouse AMH (NP_031471). Danio rerio (Dre_Amh, AY881649), Japanese eel Anguilla The best zebrafish hit (ENSDART00000013803) resulting japonica (Aja_Amh, BAB93107), Japanese flounder from this search, was then compared with GenBank (http:// Paralichthys olivaceus (Pol_Amh, BAD37138), Atlantic www.ncbi.nlm.nih.gov/blast/) to infer its nature, returning as salmon, Salmo salar (Ssa_Amh, AAU85130), chicken best hit the Japanese eel spermatogenesis-preventing sub- Gallus gallus (Gga_Amh, NP_990361), mouse Mus stance sequence BAB93107 (Miura et al., 2002), followed by musculus (Mmu_AMH, NP_031471), human Homo mammalian Amh proteins. This finding provided the first sapiens (Hsa_AMH, NP_000470); BMP2: Homo sapiens evidence that zebrafish possesses an Amh ortholog. (Hsa_BMP2, NP_001191), Danio rerio (Dre_Bmp2a, We deduced the mRNA containing the complete coding NP_571434), Danio rerio (Dre_Bmp2b, NP_571435); sequence of the zebrafish amh gene (GenBank Accession GDF7: Homo sapiens (Hsa_GDF7, NP_878248), Danio number AY881649) after performing multiple SMART rerio (Dre_Gdf7, AAD20829); GDF1: Homo sapiens RACE-PCR experiments (ClonTech) with specific (Hsa_GDF1, NP_001483), Danio rerio (Dre_Dvr1, primers (F1: 50-AGGCTCAGTACCGTTCAGTGTTGC-30; NP_571023); TGFbeta2: Homo sapiens (Hsa_TGFbeta2, F2: 50-CCACCAACATCTCCTACAAGACCAACG-30; F3: NP_003229), Danio rerio (Dre_TGFbeta2, NP_919366). 50-TACCCACTGTTACTCCAAGGACACGC-30;R1:50- CGTTGGTCTTGTAGGAGATGTTGG-30; R2: 50-GTCT 2.4. In situ hybridization TCATCAGCTCTCGCTGCTCGC-30;R3:50-GAGCAA- CACTGAACGGTACTGAGCC-30;R4:50-GCTCTGAG For gene expression analysis, zebrafish AB wild-type GCAGGATGTCCCTTAAGAAC-30). We inferred embryos were reared and collected under standard con- zebrafish amh gene intron-exon structure by comparing ditions (Westerfield, 1995). Zebrafish were staged accord- an amh mRNA sequence to the genomic sequence ing to criteria in Kimmel et al. (1995); and ZFIN (http://zfin. ENSDARG00000014357 (Zv4_Scaf578). During the prep- org): embryos (0–2 dpf), larvae (3–29 dpf), juveniles (30– aration of this manuscript, additional sequences for zebrafish 89 dpf), adults (O90 dpf). Whole-mount in situ hybridiz- amh appeared in GenBank (AY677080, NM_001007779). ation experiments were conducted as described (Jowett and Yan, 1996). In situ hybridization experiments on zebrafish 2.2. Mapping and conserved synteny cryosections were performed following the protocol described by (Strahle et al., 1994), using modifications The zebrafish amh gene was mapped on the heat shock from (Jowett et al., 1995), except that hybridization and (HS) doubled haploid mapping panel (Postlethwait et al., following washes were performed at 70 8C, and anti- 2000; Woods et al., 2000) and on the LN54 radiation hybrid digoxigenin antibody dilution was 1:5000. The zebrafish mapping panel (Hukriede et al., 1999) as described sox8 full-length cDNA (AY883015) (Chiang et al., 2001a; (Postlethwait et al., 2000) (mapping primers F4: 50- Cresko et al., 2003), and amh cDNA fragment amplified by GAAGGGCACCCGTTGTCTAAAA-30,R5:50-ATC- primers F1–R1 were cloned into Bluescript SKII (Strata- TACCCCTTCTATCTGTCTGTTGCTT-30). The HS panel gene) and TOPO (Invitrogen) vectors, respectively, and consists of 42 meioses, and has been scored for 4074 used to synthesize DIG-labeled riboprobes (Boehringer markers, including 139 on LG22. The Meiotic maps were Mannheim). Probes for sox9a and sox9b (Yan et al., 2005) constructed using MapManager (Manley et al., 2001). and, cyp19a1a (Chiang et al., 2001b) were made as For comparative mapping, a zebrafish gene was con- described. sidered orthologous to a human or fugu gene if the two The zebrafish aromatase genes were initially called genes were symmetrical best hits in reciprocal tBLASTn cyp19a and cyp19b, but these have been changed to searches against the other genome (Wall et al., 2003). cyp19a1a and cyp19a1b respectively, to follow the Human AMH genomic data were obtained from NCBI nomenclature conventions established for zebrafish, and (http://www.ncbi.nlm.nih.gov/genome/guide/human/)and this has been implemented in the Zebrafish Information the fugu amh genomic region (scaffold 303) was obtained Network (ZFIN) (http://zfin.org/). from the Takifugu rubripes genome database v3.0 available All work on animals was conducted according to the at the Joint Genome Institute (http://aluminum.jgi-psf.org/ stipulations of the University of Oregon Institutional prod/bin/runBlast.pl?dbZfugu6). Animal Care and Use Committee. A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667 665

Acknowledgements Cresko, W.A., Yan, Y.L., Baltrus, D.A., Amores, A., Singer, A., Rodriguez- Mari, A., et al., 2003. Genome duplication, subfunction partitioning, We thank Joy Murphy, Amber Starks, and the University and lineage divergence: Sox9 in stickleback and zebrafish. Dev. Dyn. 228, 480–489. of Oregon Zebrafish Facility for providing animals, and Poh de Santa Barbara, P., Bonneaud, N., Boizet, B., Desclozeaux, M., Kheng Loi and Amber Selix of the Histology Facility for Moniot, B., Sudbeck, P., et al., 1998. Direct interaction of SRY-related sectioning. We also thank Ricard Albalat at the Department protein SOX9 and steroidogenic factor 1 regulates transcription of the of Genetics at the University of Barcelona, for helpful human anti-Mullerian hormone gene. Mol. Cell. Biol. 18, 6653–6665. comments on the manuscript. The work was supported by de Santa Barbara, P., Moniot, B., Poulat, F., Berta, P., 2000. Expression and subcellular localization of SF-1, SOX9, WT1, and AMH proteins NIH grants R01RR10715 and HD22486 and NSF grant during early human testicular development. Dev. Dyn. 217, 293–298. IBN-9728587. We thank the Spanish Ministry of Education, di Clemente, N., Ghaffari, S., Pepinsky, R.B., Pieau, C., Josso, N., Culture and Sports for support for CC (EX2002–0059). Cate, R.L., et al., 1992. A quantitative and interspecific test for biological activity of anti-mullerian hormone: the fetal ovary aromatase assay. Development 114, 721–727. Duckert, P., Brunak, S., Blom, N., 2004. Prediction of proprotein convertase cleavage sites. Protein Eng. Des. Sel. 17, 107–112. References Durlinger, A.L., Visser, J.A., Themmen, A.P., 2002. Regulation of ovarian function: the role of anti-Mullerian hormone. Reproduction 124, Akiyama, H., Chaboissier, M.C., Martin, J.F., de Schedl, A., 601–609. Crombrugghe, B., 2002. The transcription factor Sox9 has essential Fanchin, R., Taieb, J., Mendez Lozano, D.H., Ducot, B., Frydman, R., roles in successive steps of the chondrocyte differentiation pathway and Bouyer, J., 2005. High reproducibility of serum anti-Mullerian hormone is required for expression of Sox5 and Sox6. Genes Dev. 16, measurements suggests a multi-staged follicular secretion and strength- 2813–2828. ens its role in the assessment of ovarian follicular status. Hum. Reprod. Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., 20, 923–927. Miller, W., et al., 1997. Gapped BLAST and PSI-BLAST: a new Fenske, M., Segner, H., 2004. Aromatase modulation alters gonadal generation of protein database search programs. Nucleic Acids Res. 25, differentiation in developing zebrafish (Danio rerio). Aquat. Toxicol. 3389–3402. 67, 105–126. Amores, A., Postlethwait, J.H., 1999. Banded chromosomes and the Goto-Kazeto, R., Kight, K.E., Zohar, Y., Place, A.R., Trant, J.M., 2004. zebrafish karyotype. In: Detrich III., H.W. et al. (Ed.), The Zebrafish: Localization and expression of aromatase mRNA in adult zebrafish. Genetics and Genomics Meth. Cell Biol., vol. 60. Academic Press, San Gen. Comp. Endocrinol. 139, 72–84. Diego, CA, pp. 323–338. Gubbay, J., Collignon, J., Koopman, P., Capel, B., Economou, A., Amores, A., Force, A., Yan, Y.-L., Joly, L., Amemiya, C., Fritz, A., et al., Munsterberg, A., et al., 1990. A gene mapping to the sex-determining 1998. Zebrafish hox clusters and vertebrate genome evolution. Science region of the mouse Y chromosome is a member of a novel family of 282, 1711–1714. embryonically expressed genes. Nature 346, 245–250. Amores, A., Suzuki, T., Yan, Y.L., Pomeroy, J., Singer, A., Amemiya, C., Hukriede, N.A., Joly, L., Tsang, M., Miles, J., Tellis, P., Epstein, J.A., et al., et al., 2004. Developmental roles of pufferfish Hox clusters and genome 1999. Radiation hybrid mapping of the zebrafish genome. Proc. Natl evolution in ray-fin fish. Genome Res. 14, 1–10. Acad. Sci. USA 96, 9745–9750. Arango, N.A., Lovell-Badge, R., Behringer, R.R., 1999. Targeted Jaillon, O., Aury, J.M., Brunet, F., Petit, J.L., Stange-Thomann, N., mutagenesis of the endogenous mouse Mis gene promoter: in vivo Mauceli, E., et al., 2004. Genome duplication in the teleost fish definition of genetic pathways of vertebrate sexual development. Cell Tetraodon nigroviridis reveals the early vertebrate proto-karyotype. 99, 409–419. Nature 431, 946–957. Brennan, J., Capel, B., 2004. One tissue, two fates: molecular genetic Josso, N., Lamarre, I., Picard, J.Y., Berta, P., Davies, N., Morichon, N., events that underlie testis versus ovary development. Nat. Rev. Genet. et al., 1993. Anti-mullerian hormone in early human development. 5, 509–521. Early Hum. Dev. 33, 91–99. Callard, G.V., Tchoudakova, A.V., Kishida, M., Wood, E., 2001. Josso, N., Racine, C., di Clemente, N., Rey, R., Xavier, F., 1998. The role of Differential tissue distribution, developmental programming, estrogen anti-Mullerian hormone in gonadal development. Mol. Cell. Endocri- regulation and promoter characteristics of cyp19 genes in teleost fish. nol. 145, 3–7. J. Steroid. Biochem. Mol. Biol. 79, 305–314. Jowett, T., Yan, Y.L., 1996. Double fluorescent in situ hybridization to Cate, R.L., Mattaliano, R.J., Hession, C., Tizard, R., Farber, N.M., zebrafish embryos. Trends Genet. 12, 387–389. Cheung, A., et al., 1986. Isolation of the bovine and human genes for Jowett, T., Ingham, P.W., Henrique, D., Lettice, L., Wilkinson, D., Mullerian inhibiting substance and expression of the human gene in Yan, Y.L., et al., 1995. An EMBO Practical Course. University of animal cells. Cell 45, 685–698. Newcastle, Newcastle p. 50. Chaboissier, M.C., Kobayashi, A., Vidal, V.I., Lutzkendorf, S., van de Kazeto, Y., Ijiri, S., Place, A.R., Zohar, Y., Trant, J.M., 2001. The 50- Kant, H.J., Wegner, M., et al., 2004. Functional analysis of Sox8 and flanking regions of CYP19A1 and CYP19A2 in zebrafish. Biochem. Sox9 during sex determination in the mouse. Development 131, Biophys. Res. Commun. 288, 503–508. 1891–1901. Kazeto, Y., Place, A.R., Trant, J.M., 2004. Effects of endocrine disrupting Chiang, E.F., Pai, C.I., Wyatt, M., Yan, Y.L., Postlethwait, J., Chung, B., chemicals on the expression of CYP19 genes in zebrafish (Danio rerio) 2001a. Two sox9 genes on duplicated zebrafish chromosomes: juveniles. Aquat. Toxicol. 69, 25–34. expression of similar transcription activators in distinct sites. Dev. Kent, J., Wheatley, S.C., Andrews, J.E., Sinclair, A.H., Koopman, P., 1996. Biol. 231, 149–163. A male-specific role for SOX9 in vertebrate sex determination. Chiang, E.F., Yan, Y.L., Guiguen, Y., Postlethwait, J., Chung, B.C., 2001b. Development 122, 2813–2822. Two Cyp19 (P450 aromatase) genes on duplicated zebrafish chromo- Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., somes are expressed in ovary or brain. Mol. Biol. Evol. 18, 542–550. 1995. Stages of embryonic development of the zebrafish. Dev. Dyn. Chiang, E.F., Yan, Y.L., Tong, S.K., Hsiao, P.H., Guiguen, J., 203, 253–310. Postlethwait, J., et al., 2001c. Characterization of duplicated zebrafish Kishida, M., Callard, G.V., 2001. Distinct cytochrome P450 aromatase cyp19 genes. J. Exp. Zool. 290, 709–714. isoforms in zebrafish (Danio rerio) brain and ovary are differentially 666 A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667

programmed and estrogen regulated during early development. (Top1) and Snail genes provides evidence on genome expansion and the Endocrinology 142, 740–750. radiation of teleost fish. In: Shimizu, N. et al. (Ed.), Aquatic. Genomics: Kishida, M., McLellan, M., Miranda, J.A., Callard, G.V., 2001. Estrogen Steps Toward a Great Future. Springer, Tokyo, pp. 20–31. and xenoestrogens upregulate the brain aromatase isoform Rouiller-Fabre, V., Carmona, S., Merhi, R.A., Cate, R., Habert, R., (P450aromB) and perturb markers of early development in zebrafish Vigier, B., 1998. Effect of anti-Mullerian hormone on Sertoli and (Danio rerio). Comp. Biochem. Physiol. B Biochem. Mol. Biol. 129, Leydig cell functions in fetal and immature rats. Endocrinology 139, 261–268. 1213–1220. Koopman, P., Schepers, G., Brenner, S., Venkatesh, B., 2004. Origin and Salmon, N.A., Handyside, A.H., Joyce, I.M., 2005. Expression of Sox8, diversity of the Sox transcription factor gene family: genome-wide Sf1, Gata4, Wt1, Dax1, and Fog2 in the mouse ovarian follicle: analysis in Fugu rubripes. Gene 328, 177–186. Implications for the regulation of Amh expression. Mol. Reprod. Dev. Kumar, S., Tamura, K., Jakobsen, I.B., Nei, M., 2001. MEGA2: molecular 70, 271–277. evolutionary genetics analysis software. Bioinformatics 17, Schepers, G., Wilson, M., Wilhelm, D., Koopman, P., 2003. SOX8 is 1244–1245. expressed during testis differentiation in mice and synergizes with SF1 Lasala, C., Carre-Eusebe, D., Picard, J.Y., Rey, R., 2004. Subcellular and to activate the Amh promoter in vitro. J. Biol. Chem. 278, molecular mechanisms regulating anti-Mullerian hormone gene 28101–28108. expression in mammalian and nonmammalian species. DNA Cell Schreeb, K.H., Broth, G., Sachsse, W., Freundt, K.J., 1993. The karyotype Biol. 23, 572–585. of the zebrafish (Brachydanio rerio). J. Exp. Anim. Sci. 36, 27–31. Li, M., Zhao, C., Wang, Y., Zhao, Z., Meng, A., 2002. Zebrafish sox9b is an Sekido, R., Bar, I., Narvaez, V., Penny, G., Lovell-Badge, R., 2004. SOX9 early neural crest marker. Dev. Genes Evol. 212, 203–206. is up-regulated by the transient expression of SRY specifically in Sertoli Liu, D., Chu, H., Maves, L., Yan, Y.L., Morcos, P.A., Postlethwait, J.H., cell precursors. Dev. Biol. 274, 271–279. et al., 2003. Fgf3 and Fgf8 dependent and independent transcription factors are required for otic placode specification. Development 130, Selman, K., Wallace, R.A., Sarka, A., Qi, X., 1993. Stages of oocyte 2213–2224. development in the zebrafish, Brachydanio rerio. J. Morphol. 218, Manley, K.F., Cudmore Jr., R.H., Meer, J.M., 2001. Map Manager QTX, 203–224. cross-platform software for genetic mapping. Mamm. Genome 12, Shen, W.H., Moore, C.C., Ikeda, Y., Parker, K.L., Ingraham, H.A., 1994. 930–932. Nuclear receptor steroidogenic factor 1 regulates the mullerian McAllister, B.G., Kime, D.E., 2003. Early life exposure to environmental inhibiting substance gene: a link to the sex determination cascade. levels of the aromatase inhibitor tributyltin causes masculinisation and Cell 77, 651–661. irreversible sperm damage in zebrafish (Danio rerio). Aquat. Toxicol. Simpson, E.R., Mahendroo, M.S., Means, G.D., Kilgore, M.W., 65, 309–316. Corbin, C.J., Mendelson, C.R., 1993. Tissue-specific promoters regulate Meyer, A., Schartl, M., 1999. Gene and genome duplications in vertebrates: aromatase cytochrome P450 expression. J. Steroid Biochem. Mol. Biol. the one-to-four (-to-eight in fish) rule and the evolution of novel gene 44, 321–330. functions. Curr. Opin. Cell Biol. 11, 699–704. Simpson, E.R., Mahendroo, M.S., Means, G.D., Kilgore, M.W., Miura, T., Miura, C., Konda, Y., Yamauchi, K., 2002. Spermatogen- Hinshelwood, M.M., Graham-Lorence, S., et al., 1994. Aromatase esis-preventing substance in Japanese eel. Development 129, cytochrome P450, the enzyme responsible for estrogen biosynthesis. 2689–2697. Endocr. Rev. 15, 342–355. Morais da Silva, S., Hacker, A., Harley, V., Goodfellow, P.N., Swain, A., Sinclair, A.H., Berta, P., Palmer, M.S., Hawkins, J.R., Griffiths, B.L., Lovell-Badge, R., 1996. Sox9 expression during gonadal development Smith, M.J., et al., 1990. A gene from the human sex-determining implies a conserved role for the gene in testis differentiation in region encodes a protein with homology to a conserved DNA-binding mammals and birds. Nat. Genet. 14, 62–68. motif. Nature 346, 240–244. Munsterberg, A., Lovell-Badge, R., 1991. Expression of the mouse anti- Smith, C.A., Smith, M.J., Sinclair, A.H., 1999. Gene expression during mullerian hormone gene suggests a role in both male and female sexual gonadogenesis in the chicken embryo. Gene 234, 395–402. differentiation. Development 113, 613–624. Strahle, U., Blader, P., Adam, J., Ingham, P.W., 1994. A simple and Nachtigal, M.W., Hirokawa, Y., Enyeart-VanHouten, D.L., Flanagan, J.N., efficient procedure for non-isotopic in situ hybridization to sectioned Hammer, G.D., Ingraham, H.A., 1998. Wilms’ tumor 1 and Dax-1 material. Trends Genet. 10, 75–76. modulate the orphan nuclear receptor SF-1 in sex-specific gene Takahashi, H., 1977. Juvenile hermaphroditism in the zebrafish, Brachy- expression. Cell 93, 445–454. danio rerio. Bull. Fac. Fish Hokkaido Univ. 28, 57–65. Oreal, E., Mazaud, S., Picard, J.Y., Magre, S., Carre-Eusebe, D., 2002. Taylor, J., Braasch, I., Frickey, T., Meyer, A., Van De Peer, Y., 2003. Different patterns of anti-Mullerian hormone expression, as related to Genome duplication, a trait shared by 22,000 species of ray-finned fish. DMRT1, SF-1, WT1, GATA-4, Wnt-4, and Lhx9 expression, in the Genome. Res. 13, 382–390. chick differentiating gonads. Dev. Dyn. 225, 221–232. Tchoudakova, A., Kishida, M., Wood, E., Callard, G.V., 2001. Promoter Park, S.Y., Jameson, J.L., 2005. Minireview: transcriptional regulation of characteristics of two cyp19 genes differentially expressed in the brain gonadal development and differentiation. Endocrinology 146, and ovary of teleost fish. J. Steroid Biochem. Mol. Biol. 78, 427–439. 1035–1042. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G., Picard, J.Y., Goulut, C., Bourrillon, R., Josso, N., 1986. Biochemical analysis of bovine testicular anti-Mullerian hormone. Fed. Eur. 1997. The CLUSTAL_X windows interface: flexible strategies for Biochem. Soc. Lett. 195, 73–76. multiple sequence alignment aided by quality analysis tools. Nucleic Pijnacker, L.P., Ferwerda, M.A., 1995. Zebrafish chromosome banding. Acids Res. 25, 4876–4882. Genome 38, 1052–1055. Tong, S.K., Chung, B.C., 2003. Analysis of zebrafish cyp19 promoters. Postlethwait, J.H., Yan, Y.-L., Gates, M., Horne, S., Amores, A., J. Steroid Biochem. Mol. Biol. 86, 381–386. Brownlie, A., et al., 1998. Vertebrate genome evolution and the Tong, S.K., Chiang, E.F., Hsiao, P.H., Chung, B., 2001. Phylogeny, zebrafish gene map. Nat. Genet. 18, 345–349. expression and enzyme activity of zebrafish cyp19 (P450 aromatase) Postlethwait, J.H., Woods, I.G., Ngo-Hazelett, P., Yan, Y.-L., Kelly, P.D., genes. J. Steroid Biochem. Mol. Biol. 79, 299–303. Chu, F., et al., 2000. Zebrafish comparative genomics and the origins of Trant, J.M., Gavasso, S., Ackers, J., Chung, B.-C., Place, A.R., 2001. vertebrate chromosomes. Genome Res. 10, 1890–1902. Developmental expression of cytochrome P450 aromatase genes Postlethwait, J.H., Amores, A., Yan, Y.L., Austin, C.A., 2002. Duplication (CYP19a and CYP19b) in zebrafish fry (Danio rerio). J. Exp. Zool. of a portion of human chromosome 20q containing Topoisomerase 290, 475–483. A. Rodrı´guez-Marı´ et al. / Gene Expression Patterns 5 (2005) 655–667 667

Uchida, D., Yamashita, M., Kitano, T., Iguchi, T., 2002. Oocyte apoptosis Western, P.S., Harry, J.L., Graves, J.A., Sinclair, A.H., 1999. Temperature- during the transition from ovary-like tissue to testes during sex dependent sex determination: upregulation of SOX9 expression after differentiation of juvenile zebrafish. J. Exp. Biol. 205, 711–718. commitment to male development. Dev. Dyn. 214, 171–177. Uchida, D., Yamashita, M., Kitano, T., Iguchi, T., 2004. An aromatase Wittbrodt, J., Meyer, A., Schartl, M., 1998. More genes in fish? BioEssays inhibitor or high water temperature induce oocyte apoptosis and 20, 511–515. depletion of P450 aromatase activity in the gonads of genetic female Woods, I.G., Kelly, P.D., Chu, F., Ngo-Hazelett, P., Yan, Y.L., Huang, H., zebrafish during sex-reversal. Comp. Biochem. Physiol. A Mol. Integr. et al., 2000. A comparative map of the zebrafish genome. Genome Res. Physiol. 137, 11–20. 10, 1903–1914. Viger, R.S., Mertineit, C., Trasler, J.M., Nemer, M., 1998. Transcription Wright, E., Hargrave, M.R., Christiansen, J., Cooper, L., Kun, J., Evans, T., factor GATA-4 is expressed in a sexually dimorphic pattern during et al., 1995. The SRY-related gene Sox9 is expressed during mouse gonadal development and is a potent activator of the Mullerian chondrogenesis in mouse embryos. Nat. Genet. 9, 15–20. inhibiting substance promoter. Development 125, 2665–2675. Wrobel, K.H., 2003. The genus Acipenser as a model for vertebrate Vigier, B., Forest, M.G., Eychenne, B., Bezard, J., Garrigou, O., Robel, P., urogenital development: the Mullerian duct. Anat. Embryol. (Berl) 206, et al., 1989. Anti-Mullerian hormone produces endocrine sex reversal of 255–271. fetal ovaries. Proc. Natl Acad. Sci. USA 86, 3684–3688. Yan, Y.L., Miller, C.T., Nissen, R., Singer, A., Liu, D., Kirn, A., et al., Vogel, G., 1998. Doubled genes may explain fish diversity. Science 281, 2002. A zebrafish sox9 gene required for cartilage morphogenesis. 1119–1121. Development 129, 5065–5079. Wall, D.P., Fraser, H.B., Hirsh, A.E., 2003. Detecting putative orthologs. Yan, Y.L., Willoughby, J., Liu, D., Crump, G.C., Wilson, C., Miller, C.T., Bioinformatics 19, 1710–1711. et al., 2005. A pair of Sox: distinct and overlapping functions of Watanabe, K., Clarke, T.R., Lane, A.H., Wang, X., Donahoe, P.K., zebrafish sox9 co-orthologs in craniofacial and pectoral fin develop- 2000. Endogenous expression of Mullerian inhibiting substance in ment. Development 132, 1069–1083. early postnatal rat sertoli cells requires multiple steroidogenic factor- Yoshinaga, N., Shiraishi, E., Yamamoto, T., Iguchi, T., Abe, S., Kitano, T., 1 and GATA-4-binding sites. Proc. Natl Acad. Sci. USA 97, 2004. Sexually dimorphic expression of a teleost homologue of 1624–1629. Mullerian inhibiting substance during gonadal sex differentiation in Westerfield, M., 1995. The zebrafish book: a guide for the laboratory use of Japanese flounder, Paralichthys olivaceus. Biochem. Biophys. Res. zebrafish (Danio rerio). University of Oregon Press, Eugene, OR. Commun. 322, 508–513.