Regionally Distinct Potencies of Mouse XY Genital Ridge to Initiate Testis Differentiation Dependent on Anteroposterior Axis

Transcription

1 DEVELOPMENTAL DYNAMICS 228: , 2003 ARTICLE Regionally Distinct Potencies of Mouse XY Genital Ridge to Initiate Testis Differentiation Dependent on Anteroposterior Axis Ryuji Hiramatsu, 1 Yoshiakira Kanai, 1 * Takuo Mizukami, 1 Maki Ishii, 1 Shogo Matoba, 1 Masami Kanai-Azuma, 2 Masamichi Kurohmaru, 1 Hayato Kawakami, 2 and Yoshihiro Hayashi 1 In mouse gonadal differentiation, the center-to-pole Sry expression pattern suggests the regionally distinct potencies of the genital ridge, which induce testis differentiation. In this study, we examined the anteroposterior axis-dependent differences in testis-differentiation potencies by using cultures of anterior, middle, and posterior segments of the XY genital ridge. The inducible pattern of Sertoli cell differentiation showed a center-to-pole wave similar to the initial Sry expression pattern. In contrast, the ability to induce Leydig cell differentiation emanated from the anterior segment and then spread to the posterior side. These findings suggest the presence of two distinct dynamic waves in the capacity of the genital ridge to induce Sertoli or Leydig cell differentiation at early phases of testis differentiation. Developmental Dynamics 228: , Wiley-Liss, Inc. Key words: Sry; Sox9; 3 -Hsd; Sertoli cell; Leydig cell; gonad; genital ridge; sex differentiation; anteroposterior axis; organ culture; mouse Received 15 April 2003; Accepted 8 July 2003 INTRODUCTION Sry (Sex-determining region of the Y) is essential for initiating male sex differentiation in mammals. Sry is active for a very short period in somatic cells of the gonadal ridge to initiate Sertoli cell differentiation in mice. Several recent studies have demonstrated that Sry expression is first detected in the central region of the gonadal ridge at 11.0 days post coitum (dpc, tail-somite [ts] stage), and its expression extends to both anterior and posterior ends by 11.5 dpc (approximately ts; Bullejos and Koopman, 2001; Albrecht and Eicher, 2001). These findings clearly suggest a region-dependent difference in Sry initiation or stability at the initial stages of testis differentiation, lending additional support to the hypothesis that activation of the male-specific program starts in the gonadal somatic cells located in the middle portion of the XY genital ridge. In the developing gonad, there are no appreciable morphologic differences along the anteroposterior (AP) axis. However, a possible regional difference has been implicated by the distinct expression patterns of several gonadal marker genes dependent on the AP axis. For example, the expression of Col2a1, a type II collagen gene that is expressed in the developing male gonads, is initially restricted to the anterior pole of the 11.5 dpc male gonads (McClive and Sinclair, 2003). The expression of the secreted metalloproteinase gene, Adamts19, which is predominantly expressed in female gonads, shows a higher level in the anterior region than in more posterior regions at 12.5 dpc (Menke and Page, 2002). Therefore, with regard to reports of the center-restricted initial Sry expression, these findings suggest the existence of distinct potencies that initiate gonadal development and/or sex differentiation along the AP axis of the gonadal ridge. However, their func- 1 Department of Veterinary Anatomy, The University of Tokyo, Bunkyo-ku, Tokyo, Japan 2 Department of Anatomy, Kyorin University School of Medicine, Mitaka, Tokyo, Japan *Correspondence to: Yoshiakira Kanai, D.V.M., Ph.D., Department of Veterinary Anatomy, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo , Japan. aykanai@mail.ecc.u-tokyo.ac.jp DOI /dvdy Wiley-Liss, Inc.

2 248 HIRAMATSU ET AL. tional significance remains unclear at present. In the present study, the anterior, middle, and posterior segments of the XY genital ridge isolated at various tail somite (ts) stages were cultured separately to analyze the regionally distinct potencies of the genital ridge along the AP axis. Histologic and in situ hybridization analyses of these cultured explants demonstrated the presence of two distinct dynamic waves in the capacity of the genital ridge to induce Sertoli or Leydig cell differentiation at early phases of testis differentiation. RESULTS Distinct Developmental Windows of Cord Formation and Leydig Cell Differentiation in Cultures of the Whole Genital Ridge Previous studies have demonstrated that organ cultures of the XY genital ridge isolated at stages earlier than 11.0 dpc (approximately 14 ts) failed to induce proper testicular cord organization (Taketo and Koide, 1981; Tilmann and Capel, 1999). The present culture system of whole genital ridges of mouse embryos (ICR strain) also showed that testicular cords were observed only in gonadal explants isolated after 12 ts (number of explants including welldefined cords: 23 of 34 at ts vs. 1 of 31 at 9 11 ts; Fig. 1C). These explants after 12 ts were also expressing Sox9, a Sertoli cell marker (Kent et al., 1996; Fig. 1D). In contrast, Leydig cell differentiation, as estimated by the expression of 3 - Hsd (Nordqvist and Tohonen, 1997), was efficiently detected in approximately 70% of the gonadal explants isolated at 7 11 ts (19 of 24 at 7 11 ts vs. 0 of 6 at 4 6 ts). The cell mass of 3-day cultures initiated at 7 11 ts was similar to that of 11.5 dpc gonads in size and shape (Fig. 1C), although Sox9 expression was completely missing (Fig. 1D). Moreover, the expression of 3 -Hsd, as well as Mfge8 (a marker gene for gonadal stromal cell types [Kanai et al., 2000]), was induced in these explants (Fig. 1D), suggesting Fig. 1. Organ cultures using whole genital ridges and their anterior, middle, and posterior segments (A C), and the expression patterns of several marker genes in the cultured explants of whole genital ridges (D). A: Phase microscopic photographs showing the whole XY genital ridge (left) and its separated anterior (Ant), middle (Mid), and posterior (Post) segments (right) at 16 tail-somite (ts) stage. The anterior pole is shown on the left in each plate. B: Phase microscopic photographs showing 3-day cultures of whole XY genital ridge (far left) and three segments (three at right) isolated at 13 ts (upper row) and 16 ts (lower row). C: Plastic sections showing the gonadal area in 3-day cultures of whole XY genital ridge isolated at 9 ts (left) and 12 ts (right). D: Whole-mount in situ hybridization analyses showing the expression of Sox9, 3 -Hsd, and Mfge8 in 3-day cultures of whole genital ridges isolated at 8 9 ts (upper row), 13 ts, and 16 ts (lower row). g, gonadal coelomic surface; ms, mesonephros; C, testicular cord; I, gonadal interstitial region; asterisk, 3 -Hsd positive cell mass of presumptive adrenal glands. Scale bars 100 m in A D. proper differentiation of gonadal steroidogenic and stromal cell types but not supporting cell lineage in the explants at 7 9 ts. Therefore, the present culture system can estimate the ability of the genital ridge to induce Leydig cell differentiation for genital ridge explants isolated at stages after 7 ts. The analysis of cord formation and Sertoli cell differentiation, however, may be technically limited to gen-

3 DIFFERENCES IN TESTIS DIFFERENTIATION 249 TABLE 1. Summary of Testis Cord Formation and Leydig Cell Differentiation in Cultures of Anterior, Middle, and Posterior Segments of the Genital Ridge Isolated at Various Stages ( dpc) a Stage 7 9 ts ts ts ts Region Testicular cords b Anterior Middle Posterior Leydig cells c (3 Hsd) Anterior Middle Posterior a Embryos at approximately 10.5 or 11.5 days post coitum (dpc) show 8 or 18 tail somite (ts), respectively. b Number of the explants with no cord-like structure ( ), slender cord-like structures ( ), or a well-defined testicular cord ( ). c Number of explants, including 3 -Hsd negative ( ), or weakly ( ), or strongly ( ) 3 -Hsd positive cells in their gonadal areas. ital ridge explants isolated at stages after 12 ts. Center-to-Pole Pattern of the Capacity of the Genital Ridge to Induce Cord Formation and Sox9 Expression To analyze the possibility of regionally distinct potencies of the gonad in testis differentiation, the anterior, middle, and posterior segments of XY genital ridge isolated at various stages were cultured separately (Fig. 1A). All explants of anterior, middle, and posterior segments isolated at 7 17 ts developed into a similarly shaped gonad-mesonephros structure (Fig. 1B). Moreover, no appreciable difference in size was detected among the three segment cultures initiated at the same stage, suggesting no regional difference in the outgrowth of the gonadal segment in vitro. Histologic analyses revealed a stage-dependent and regionally distinct development of testicular cords in the cultures of each segment of the XY genital ridge (upper row in Table 1). In all segments isolated before 11 ts, no cord-like structure could be induced (Fig. 2A C), which agrees with the lack of cord formation in explants of the whole genital ridge at the same stage. In cultures initiated at ts, however, approximately 70% of both anterior and posterior segments failed to form testicular cords in the gonadal area, whereas well-defined cords were observed in the middle segments at the same stages (Fig. 2D F). Figure 2D,F shows typical examples in which slender, but poorly defined, cord-like structures, were histologically visible in gonadal explants of 13 ts anterior and posterior segments. In ts explants, cord formation was induced in most anterior segments (Fig. 2G,H), whereas 50% of posterior explants still failed to develop proper cord formation (Fig. 2I). To determine whether the failure of cord formation in the gonadal cultures reflect a defect in Sertoli cell differentiation, we examined the expression pattern of Sox9, a Sertoli cell marker, in each explant by wholemount in situ hybridization (Fig. 3). The Sox9 expression profile in each gonadal explant exhibited a close correlation with the spatial and temporal patterns of testicular cord formation described above. In short, no Sox9 expression was detected in any of the three segments isolated before 11 ts (0 of 5 samples; Fig. 3A C); however, in cultures initiated at ts, Sox9 expression was highly and efficiently induced in the middle segments, compared with those in the anterior and posterior segments of the same genital ridges (Fig. 3D F). In ts gonadal explants, all three segments were expressing Sox9, although the expression level was relatively lower in the explants of posterior segments (Fig. 3G I). Therefore, these results suggest that the capacity of the genital ridge to induce both cord formation and Sertoli cell differentiation may be higher in the middle segment than in the anterior and/or posterior segments of the genital ridge at ts when Sry expression is restricted to the middle portion in vivo. Anterior-to-Posterior Pattern of the Ability of the Genital Ridge to Induce Leydig Cell Differentiation The spatiotemporal pattern of the potencies of the genital ridge to induce Leydig cell differentiation differs from the center-to-pole pattern characteristic of the initiation of cord formation and/or Sertoli cell differentiation in vitro. Whole-mount in situ hybridization using an antisense probe against 3 -Hsd, a Leydig cell marker, revealed that 3 -Hsd positive cells were detected only in explants of anterior segments but not in middle and posterior segments of the 7 9 ts genital ridge (Fig. 4A C; lower row in Table 1). In gonadal explants isolated at ts, 3 -Hsd expression was observed in most explants of the anterior segments, in a few explants of the middle segments, and in no explants of the posterior segments (Fig. 4D F). Subsequently,

4 250 HIRAMATSU ET AL. Fig. 2. Semithin sections showing testicular cord formation in 5-day cultures of anterior (A,D,G), middle (B,E,H), or posterior (C,F,I) segments of genital ridges isolated at 11 tail somite (ts; A C), 13 ts (D F), and 16 ts (G I) stages. Each inset in A and D F shows a presumptive Leydig cell (a round cell containing many lipid droplets in its cytoplasm; Pelliniemi et al., 1996) in the gonadal area. C, testicular cord; I, gonadal interstitial region. Scale bar 100 m in C (applies to A I). the ability to induce 3 -Hsd expression was detected in most segments by ts (Fig. 4G L). Histologic analysis using plastic semithin sections stained with toluidine blue also enabled us to identify Leydig cells as round cells containing many lipid droplets in their cytoplasm (Pelliniemi et al., 1996; inset plates in Fig. 2). Analysis using serial sections of the gonadal explants also confirmed a similar anterior-restricted pattern to the data obtained by using 3 -Hsd expression analysis. Presumptive Leydig cells were histologically detectable only in the cultures of anterior segments isolated before 11 ts (n 4). In explants isolated after 12 ts, however, they appeared in the gonadal interstitial region of more posterior segments. Therefore, we concluded that the capacity to induce Leydig cell differentiation may be restricted to the anterior segments of the 7 11 ts genital ridge. As the developmental stage of the genital ridge proceeds, however, its differentiation potency may extend to the posterior side by ts. In some cultures of anterior segments, 3 -Hsd positive signals were also detected in the presumptive adrenal glands which may have developed from their primordium in the isolated genital ridge (asterisks in Fig. 4G,J). Of interest, the primordium of presumptive adrenal glands was observed only in the cultures of anterior segments, which were located close to the gonadal area in the mesonephric region of some explants. This observation is also consistent with the model that two distinct steroidogenic cell lineages in the gonad and adrenal gland have a common origin near the anterior end of the mesonephros in mouse embryogenesis (Morohashi, 1997). DISCUSSION The present data on spatiotemporal patterns of cord formation and Sox9 expression in vitro showed a close correlation with Sry expression level in gonadal segments at the initiation of culture. In short, no potency to initiate cord formation and/or Sox9 expression was found in any gonadal culture initiated at stages earlier than 11 ts, when the Sry transcript level is lower than that detectable by RNase protection and in situ hybridization (Hacker et al., 1995; Bullejos and Koopman, 2001). In gonadal explants isolated at ts (approximately 11.0 dpc) when Sry expression is first detectable in the middle segment, the capacity to initiate cord formation was clearly higher in the middle segment than in the anterior or posterior segments. As the Sry expression domain extended to the anterior or posterior side, the capacity increased simultaneously in both the anterior and posterior segments. Therefore, these results show a close correlation with the centerto-pole Sry expression pattern, suggesting that the differentiation and/or maintenance of Sertoli cells probably requires a sufficient level of Sry expression in the genital ridge at culture initiation. In contrast to the center-to-pole pattern of the potencies of the genital ridge to induce cord formation and Sox9 expression, the present study has shown an anterior-to-posterior pattern of the ability of the genital ridge to induce Leydig cell differentiation during early phases of Sry initiation, as measured by expression of 3 -Hsd. This anterior-to-posterior pattern is consistent with a similar expression pattern displayed by Ptch, the receptor gene of Dhh (a positive regulator of Leydig cell differentiation), in the interstitium of XY gonads during ts (approximately dpc; Yao et al., 2002). Moreover, the ability to induce steroidogenic cell differentiation was detected in the anterior segments of genital ridges isolated at 7 11 ts (approximately dpc), when Sry expression is at a relatively low level in vivo, detectable by only reverse transcription-polymerase chain reaction (PCR; Jeske et al., 1996). These findings, therefore, sug-

5 DIFFERENCES IN TESTIS DIFFERENTIATION 251 Fig. 4. Fig. 3. Whole-mount in situ hybridization analyses showing Sox9 expression in 3-day cultured explants of anterior (A,D,G), middle (B,E,H), or posterior (C,F,I) segments of genital ridges isolated at 11 tail somite (ts; A C), ts (D F), and ts (G I) stages. The gonadal areas are shown at the upper side of each panel. g, gonadal coelomic surface; ms, mesonephros. Scale bars 100 m in A (applies to A C), in D (applies to D F), in G (applies to G I). Fig. 3. Fig. 4. Whole-mount in situ hybridization analyses showing 3 - Hsd expression in 3-day cultured explants of anterior (A,D,G,J), middle (B,E,H,K), or posterior (C,F,I,L) segments of genital ridges isolated at 8 ts (A C), 11 ts (D E), 14 ts (G I), and 16 ts (J L). The gonadal areas are shown at the upper side of each panel. g, gonadal coelomic surface; ms, mesonephros; asterisks, 3 -Hsd positive cell mass of presumptive adrenal glands. Scale bars 100 m in C (applies to A C), in F (applies to D F), in I (applies to G I), in L (applies to J L).

6 252 HIRAMATSU ET AL. gest that it is likely that the differentiation and/or development of Leydig cells is controlled by another dynamic wave, independent of Sry expression level and cord formation/sox9 expression in the XY genital ridge. Another possible explanation for this anterior-to-posterior wave is that the inducible pattern of Leydig cell differentiation may reflect the dynamic allocation of Leydig precursor cells along the AP axis of the genital ridge during early gonadogenesis. By using immunohistochemical analysis of the anti-sf1/ad4bp antibody as a steroidogenic cell marker, Hatano et al. (1996) reported some evidence that Leydig cells and steroid cells of the adrenal gland may share a common origin at 10.5 dpc near the anterior end of the mesonephros. This model is also consistent with our present observation showing that the developing primordium of presumptive adrenal glands was found only in the cultures of anterior segments, which were located close to the gonadal area in the mesonephric region of some explants. Therefore, at present, we cannot exclude the possibility that Leydig precursor cells may initially be located in the anterior region of the genital ridge and then spread to the posterior side during early stages of testis differentiation. To the best of our knowledge, the present study is the first to demonstrate the regionally distinct potencies of the genital ridge in the induction of Sertoli and Leydig cell differentiation along the AP axis of the genital ridge at early stages of testis differentiation. The present data have shown that the inducible pattern of cord formation and Sertoli cell differentiation closely resembles the center-to-pole Sry expression pattern in the XY genital ridge. The present study has also demonstrated another dynamic pattern of the capacity of the genital ridge to induce Leydig cell differentiation that emanates from the anterior region of the genital ridge. However, there is still no direct evidence whether such dynamic patterns are caused by the regionally distinct potencies of precursor cells or by their uneven or polarized localization along the AP axis of the genital ridge. Moreover, in the present study, we could not demonstrate a possible contribution of mesonephric tissue to such regionally distinct potencies due to the technical difficulty of separating the gonad and mesonephros in the genital ridge at stages earlier than 15 ts. Further understanding of the origins and the spatiotemporal behavior of each gonadal somatic cell lineage during early gonadogenesis is required to resolve these questions. EXPERIMENTAL PROCEDURES Organ Culture of Gonadal Fragments Isolated at dpc Embryos were obtained from pregnant female mice (ICR strain) at approximately dpc (7 17 ts). After counting the tail somite number and separating the head tissues for sex determination in each embryo, the genital ridges (i.e., gonad plus mesonephros) were isolated under a dissecting microscope. One of each pair of genital ridges was separated into three equal segments (i.e., anterior, middle, and posterior) by a sharp needle under a dissecting microscope (Fig. 1A). The other was used as whole gonadal explant for a control experiment. For genital ridges isolated from embryos before 9 ts, their anterior, middle, and posterior segments were used for the following cultures without separating the left and right genital ridges. The whole explant or each anterior, middle, or posterior segment of the genital ridge was placed onto an ISO- PORE membrane filter (pore size, 3.0 m; Millipore), floated on Dulbecco s Modified Eagle s Medium (Sigma) containing 10% horse serum and penicillin/streptomycin (GIBCO BRL), as described previously (Kanai et al., 1991), and cultured at 37 C for 3 to 5 days (Fig. 1B). All explants were subjected to histologic and in situ hybridization analyses as described below. In addition, genomic DNA was isolated from the head region of each embryo, and the sex of each embryo was determined by PCR using Zfy-specific primers as described previously (Bowles et al., 1999). Histologic Analysis Each explant was fixed in 2.5% glutaraldehyde-0.1 M phosphate buffer (PB) at 4 C for 4 hr. After washing with phosphate buffered saline (PBS), the samples were post-fixed in 1% OsO 4 in 0.1 M PB at 4 C for 2 hr. The explants were then dehydrated in ethanol and embedded in Araldite M. Serial semithin sections (approximately 1 m) were cut and stained with 1% toluidine blue. The presence of testicular cord formation was histologically estimated as follows: negative ( ), no cord-like structure; /, slender cord-like structure; positive ( ), well-defined testicular cords in gonadal area. Whole-Mount In Situ Hybridization Whole-mount in situ hybridization was performed mainly by using the automatic in situ hybridization system (AIH-201; Aloka, Tokyo), following the protocol described by Kanai- Azuma et al. (1999). In short, the cultured explants were fixed in 4% paraformaldehyde-pbs for 4 hr and dehydrated in methanol. By using the automatic in situ hybridization system, the samples were rehydrated, pretreated with 10 g/ml proteinase K in PBST for 60 min, and then hybridized with digoxigenin (DIG) -labeled RNA probes in a solution containing 50% formamide, 10% dextran sulfate, 5 standard saline citrate (SSC), 1% sodium dodecyl sulfate, 50 g/ml heparin, and 50 g/ml denatured yeast RNA at 68 C for 16 hr. After treatment with RNase A (100 g/ml; Sigma) at 37 C for 30 min, they were washed twice with 5 SSC/2 SSC at 65 C for 1 hr. The signals were detected by an immunologic method by using alkaliphosphatase-conjugated anti-dig antibody and nitro blue tetrazolium as the chromogen (Roche Molecular Biochemicals). RNA probes for Sox9 (Kent et al., 1996), Mfge8 (Kanai et al., 2000), and 3 -Hsd (Nordqvist and Tohonen, 1997) were used in this study. ACKNOWLEDGMENTS The authors thank Dr. Josephine Bowles for her critical reading and

7 DIFFERENCES IN TESTIS DIFFERENTIATION 253 comments on the manuscript. The authors also thank Mr. Iwao Tsugiyama for his technical and secretarial assistance. M.T. is a JSPS Research Fellow (DC3). REFERENCES Albrecht KH, Eicher EM Evidence that Sry is expressed in pre-sertoli cells and Sertoli and granulosa cells have a common precursor. Dev Biol 240: Bowles J, Cooper L, Berkman J, Koopman P Sry requires a CAG repeat domain for male sex determination in Mus musculus. Nat Genet 22: Bullejos M, Koopman P Spatially dynamic expression of Sry in mouse genital ridges. Dev Dyn 221: Hacker A, Capel B, Goodfellow P, Lovell- Badge R Expression of Sry, the mouse sex determining gene. Development 121: Hatano O, Takakusu A, Nomura M, Morohashi K Identical origin of adrenal cortex and gonad revealed by expression profiles of Ad4BP/SF-1. Genes Cells 1: Jeske YW, Mishina Y, Cohen DR, Behringer RR, Koopman P Analysis of the role of Amh and Fra1 in the Sry regulatory pathway. Mol Reprod Dev 44: Kanai Y, Hayashi Y, Kawakami H, Takata K, Kurohmaru M, Hirano H, Nishida T Effect of tunicamysin, an inhibitor of protein glycosylation, on testicular cord organization in fetal mouse gonadal explants in vitro. Anat Rec 230: Kanai Y, Kanai-Azuma M, Tajima Y, Birk OS, Hayashi Y, Sanai Y Identification of a stromal cell type characterized by the secretion of a soluble integrin-binding protein, MFG-E8, in mouse earlygonadogenesis.mechdev96: Kanai-Azuma M, Kanai Y, Okamoto M, Hayashi Y, Yonekawa H, Yazaki K Nrk: a murine X-linked NIK (Nck-interacting kinase)-related kinase gene expressed in skeletal muscle. Mech Dev 89: Kent J, Wheatley SC, Andrews JE, Sinclair AH, Koopman P A malespecific role for SOX9 in vertebrate sex determination. Development 122: McClive PJ, Sinclair AH Type II and IX collagen transcript isoforms are expressed during mouse testis development. Biol Reprod 68: Menke DB, Page DC Sexually dimorphic gene expression in the developing mouse gonad. Gene Expr Patterns 2: Morohashi K The ontogenesis of the steroidogenic tissues. Genes Cells 2: Nordqvist K, Tohonen V An mrna differential display strategy for cloning genes expressed during mouse gonad development. Int J Dev Biol 41: Pelliniemi LJ, Kuopio T, Frojdman K The cell biology and function of the fetal Leydig cell. In: Payne AH, Hardy MP, Russell LD, editors. The Leydig cell. Vienna, IL: Cache River Press. p Taketo T, Koide SS In vitro development of testis and ovary from indifferent fetal mouse gonads. Dev Biol 84: Tilmann C, Capel B Mesonephric cell migration induces testis cord formation and Sertoli cell differentiation in the mammalian gonad. Development 126: Yao HH, Whoriskey W, Capel B Desert Hedgehog/Patched 1 signaling specifies fetal Leydig cell fate in testis organogenesis. Genes Dev 16: