Abstract

The sex of an individual is determined by the fate of the gonad. While the expression of Sry and Sox9 is sufficient to induce male development, we here show that female differentiation requires activation of the canonical β-catenin signaling pathway. β-catenin activation is controlled by Rspo1 in XX gonads and Rspo1 knockout mice show masculinized gonads. Molecular analyses demonstrate an absence of female-specific activation of Wnt4 and as a consequence XY-like vascularization and steroidogenesis. Moreover, germ cells of XX knockout embryos show changes in cellular adhesions and a failure to enter XX specific meiosis. Sex cords develop around birth, when Sox9 becomes strongly activated. Thus, a balance between Sox9 and β-catenin activation determines the fate of the gonad, with Rspo1 acting as a crucial regulator of canonical β-catenin signaling required for female development.

INTRODUCTION

The sex of an individual is determined during early development by the fate of the gonad. Transgenic analysis has demonstrated that the expression of the sex determining gene Sry within the initially bipotential gonad is sufficient to induce the male developmental program (1). In mice, Sry is expressed for only 2 days starting from E10.5 (2,3). Shortly after the onset of Sry expression, the transcription factor Sox9 becomes upregulated, which in turn induces the differentiation of Sertoli cells (4). Sox9 stimulates expression of Pgds (5), which—together with Fgf9—is required for maintaining Sox9 expression in the XY gonad (2,6). Sertoli cells surround primordial germ cells to form the sex cords (7). The addition of further cell types completes the formation of seminiferous tubules. Sox9 activates Amh/Mis, which in turn induces the regression of the Mullerian duct, the precursor of the uterus, oviduct and part of the vagina. Testosterone secreted by Leydig cells induces Wolffian duct development into epididymis, vas deferens and seminal vesicles.

In contrast to male sex determination, very little is known about the molecular pathways governing ovary differentiation. Histologically, the first ovarian feature appears at E13.5 when oogonia enter meiosis (8). Around birth, oocytes are blocked in diplotene of prophase I and separated by cytoplasmic extensions of epithelial pregranulosa cells to form primordial follicles (9). Meiosis requires signaling through retinoic acid (RA), which stimulates the expression of Stra8 (10,11). In the XY gonad, the degradation of RA by Cyp26b1 prevents gonocytes from entering meiosis (12). Germ cells are not required for Sertoli cell differentiation, but are necessary for follicle formation. Their absence in XX gonads results in streak gonads or structures resembling sex cords (13). Adams et al. (14) showed that gonocytes can induce masculinisation of somatic cells via the paracrine signal PGD2. Moreover, after initiation of meiosis, XX germ cells antagonize sex cord formation (15) indicating the importance of germ cells for ovarian development.

Wnt4 is expressed specifically in XX gonads from E11.5 onwards and is required to suppress endothelial and steroidogenic cell migration from the mesonephros and the adrenal, respectively (16). Thus, XX Wnt4−/− mice develop ovotestes with the presence of sex cords and functional steroidogenic Leydig cells (17,18). Strikingly, oocytes degenerate after E14.5 with less than 10% remaining at birth when sex cords develop. In goats, XX sex reversal has been associated with PIS mutations (19) that result in the downregulation of FOXL2. XX FoxL2−/− mice show premature ovarian failure (20,21) and postnatal expression of male specific genes (22). Recently, female to male sex reversal in XX patients has been shown to be due to homozygous mutations in RSPO1 (23). Rspondins are secreted activators of the transcriptional activity of β-catenin (24,25) by modifying the levels of LRP6 protein available on the cell surface (26–28). In mouse gonads, Rspo1 is specifically up-regulated in XX somatic cells from E11.5 onwards (23).

RESULTS

The transcriptional function of β-catenin is specifically activated in female gonads

R-spondin genes encode secreted proteins that activate the canonical β-catenin pathway. Because of the importance of RSPO1 and Wnt4 for ovarian differentiation, it has been proposed that β-catenin may also be activated in ovary development (29,30), but until now no direct experimental evidence has been reported.

To determine whether β-catenin is involved in mammalian sex determination, we employed the Axin2+/LacZ reporter line, which carries LacZ fused to Axin2 (31). Since Axin2 is a target of β-catenin, this strain provides a readout of the canonical β-catenin signaling. While XY gonads were devoid of blue signal at all stages analyzed (E12.5, E13.5, E14.5, E16.5, E18.5), whole-mount β-galactosidase staining revealed a robust staining of XX gonads from E12.5 onwards. Additional staining was detected in the Wolffian and Mullerian ducts of both sexes (Fig. 1A).

Figure 1.

Canonical β-catenin signaling is specifically activated in female gonads. (A) X-Gal whole-mount staining of Axin2+/LacZ urogenital ridges (magnification ×20). Blue staining reflecting β-galactosidase activity was observed in XX gonads, whereas XY gonads remained unstained. Arrows and arrowhead show the Wolffian ducts and Mullerian duct staining, respectively ( G: gonad). (B) X-Gal staining and immunostaining on cryosections of XX and XY Axin2+/LacZ gonads at E14.5 (magnification ×40). Blue staining localized to somatic cells and did not overlap with the germ cell marker Mvh (red). (C) Quantitative RT–PCR expression analysis of E12.5 XX and XY gonads. For each genotype, n = 6. Gene expression levels were normalized against expression levels of Hprt1. β-catenin target genes were up-regulated in XX gonads. (D) Quantitative analysis of Lef1 and Rspo1 in XX Wnt4−/− gonads compared to XX wild-type littermates at E12.5 and E14.5 (XX Wnt4−/−, n = 8; wild type n = 4). Gene expression levels were normalized against expression levels of Hprt1. Lef1 was down-regulated in XX Wnt4−/− gonads. KO: Wnt4 knock-out; Wt: Wild type. NC: no change in gene expression between KO and Wt samples. X-Gal whole-mount staining of urogenital ridges at E13.5. XX Wnt4−/− gonads showed decreased β-galactosidase activity when compared to XX gonads. (E) X-Gal staining of gonads from XY and XX wild type, as well as Wt1:Sox9 transgenic mice (XX sex reversal) carrying the Axin2+/LacZ allele. Note the complete absence of signal in XX, Wt1:Sox9 sex-reversed gonads. Arrowheads indicate sex cords of the XY or XX Wt1:Sox9Tr/+ gonads.

Figure 1.

Canonical β-catenin signaling is specifically activated in female gonads. (A) X-Gal whole-mount staining of Axin2+/LacZ urogenital ridges (magnification ×20). Blue staining reflecting β-galactosidase activity was observed in XX gonads, whereas XY gonads remained unstained. Arrows and arrowhead show the Wolffian ducts and Mullerian duct staining, respectively ( G: gonad). (B) X-Gal staining and immunostaining on cryosections of XX and XY Axin2+/LacZ gonads at E14.5 (magnification ×40). Blue staining localized to somatic cells and did not overlap with the germ cell marker Mvh (red). (C) Quantitative RT–PCR expression analysis of E12.5 XX and XY gonads. For each genotype, n = 6. Gene expression levels were normalized against expression levels of Hprt1. β-catenin target genes were up-regulated in XX gonads. (D) Quantitative analysis of Lef1 and Rspo1 in XX Wnt4−/− gonads compared to XX wild-type littermates at E12.5 and E14.5 (XX Wnt4−/−, n = 8; wild type n = 4). Gene expression levels were normalized against expression levels of Hprt1. Lef1 was down-regulated in XX Wnt4−/− gonads. KO: Wnt4 knock-out; Wt: Wild type. NC: no change in gene expression between KO and Wt samples. X-Gal whole-mount staining of urogenital ridges at E13.5. XX Wnt4−/− gonads showed decreased β-galactosidase activity when compared to XX gonads. (E) X-Gal staining of gonads from XY and XX wild type, as well as Wt1:Sox9 transgenic mice (XX sex reversal) carrying the Axin2+/LacZ allele. Note the complete absence of signal in XX, Wt1:Sox9 sex-reversed gonads. Arrowheads indicate sex cords of the XY or XX Wt1:Sox9Tr/+ gonads.

To identify the cellular lineage in which β-catenin becomes activated, sections were examined for β-galactosidase coloration in combination with immunostaining against the germ cell marker Mvh. Overlay of the resulting images from XX gonads clearly demonstrated strong, but mutually exclusive signals indicating that Axin2 activation occurs predominantly in somatic cells (Fig. 1B). These results strongly suggest a sex-specific role for the nuclear translocation of β-catenin during ovarian development.

To confirm this female-specific activation of the canonical signaling pathway, we measured the expression of Axin2 and Lef1, two known β-catenin target genes (http://www.stanford.edu/~rnusse/wntwindow.html). Expression of Dax1, Emx2 and Wnt4 has been shown to be regulated by Wnt/β-catenin signaling (29,32,33) and were all up-regulated in E12.5 XX gonads, when compared to their XY littermates (Fig. 1C).

To test whether Wnt4 is required for activation of the canonical β-catenin pathway in vivo, we analyzed Lef1 expression levels. From E12.5, Lef1 was clearly down-regulated in XX Wnt4−/− compared to XX wild-type gonads (Fig. 1D). Moreover, X-gal staining at E13.5 showed down-regulation of the Axin2+/LacZ allele in XX Wnt4−/− gonads. However, in contrast to XY gonads weak staining persisted in XX Wnt4−/− tissues, indicating that ablation of Wnt4 is not sufficient to completely prevent β-catenin signaling. This suggests that while Wnt4 is involved in β-catenin activation, additional Wnt4 independent pathway(s) also exist.

Sox9 is a key regulator of male development and ectopic expression of this gene in XX transgenic gonads (XX Wt1:Sox9Tr/+) triggers female to male sex reversal (34). To test whether ectopic activation of Sox9 would also interfere with β-catenin signaling in vivo, we performed X-Gal staining on mice double transgenic for Axin2+/LacZ and Wt1:Sox9. Strikingly, XX gonads carrying the Wt1:Sox9 transgene remained unstained (Fig. 1E).

Taken together, these results demonstrate that the canonical β-catenin signaling pathway is activated in the somatic cells during ovarian embryonic development and is antagonized by ectopic expression of Sox9.

Targeted disruption of Rspo1

To address whether Rspo1 is a critical activator of the β-catenin pathway during normal development, we generated a knockout allele by inserting the LacZ reporter gene into exon 3 (Fig. 2A). Germ-line transmission was achieved for two targeted clones (Fig. 2B), both of which exhibited an identical phenotype in the homozygous state.

Figure 2.

Targeted disruption of Rspo1. (A) Schematic representation of the Rspo1 genomic locus, targeting vector and mutant allele. Black boxes represent protein-coding regions of exons, gray boxes represent the LacZ gene and shaded boxes the neomycine selection marker. X: XbaI; S: SpeI; E: EcoRI; N: NotI restriction enzyme recognition sites. Southern probe (white boxes) and PCR primers (P20, LacZ50, REx4, REx5, P10R) are shown. p(A): polyadenylation site. (B) Genotyping of adult mice by southern blot using a 5′ probe detecting a 9.5 kb XbaI fragment and a 2.8 kb SpeI fragment in the wild-type genomic locus, and a 6.9 kb XbaI fragment and a 7.3 kb SpeI fragment in the mutant allele. (C) RT–PCR analysis demonstrating termination of the transcript in the targeted allele. Seven hundred and fifty-five base pairs transcript joining exon 4 to the 3′-UTR (primers REx4 5′-AAGATCGAGCACTGTGAGGC/P10R 5′-TATACAGCTGGGCCACAGAA) and 605 bp transcript joining exon 5 to the 3′-UTR (primers REx5 5′-AATGTGAAATGAGCGAGTGG/P10R) were readily detected in RNA isolated from E13.5 wild-type and heterozygote embryos, but absent in RNA from Rspo1−/− embryos. GAPDH primers were used as a positive control.

Figure 2.

Targeted disruption of Rspo1. (A) Schematic representation of the Rspo1 genomic locus, targeting vector and mutant allele. Black boxes represent protein-coding regions of exons, gray boxes represent the LacZ gene and shaded boxes the neomycine selection marker. X: XbaI; S: SpeI; E: EcoRI; N: NotI restriction enzyme recognition sites. Southern probe (white boxes) and PCR primers (P20, LacZ50, REx4, REx5, P10R) are shown. p(A): polyadenylation site. (B) Genotyping of adult mice by southern blot using a 5′ probe detecting a 9.5 kb XbaI fragment and a 2.8 kb SpeI fragment in the wild-type genomic locus, and a 6.9 kb XbaI fragment and a 7.3 kb SpeI fragment in the mutant allele. (C) RT–PCR analysis demonstrating termination of the transcript in the targeted allele. Seven hundred and fifty-five base pairs transcript joining exon 4 to the 3′-UTR (primers REx4 5′-AAGATCGAGCACTGTGAGGC/P10R 5′-TATACAGCTGGGCCACAGAA) and 605 bp transcript joining exon 5 to the 3′-UTR (primers REx5 5′-AATGTGAAATGAGCGAGTGG/P10R) were readily detected in RNA isolated from E13.5 wild-type and heterozygote embryos, but absent in RNA from Rspo1−/− embryos. GAPDH primers were used as a positive control.

Rspo1+/− mice appeared phenotypically normal and fertile. RT–PCR experiments using primers located in exon 4 or 5 and a reverse primer in the 3′-UTR of Rspo1 revealed expression in Rspo1+/− but not in Rspo1−/− gonads (Fig. 2C) indicating that the insertion resulted in a complete termination of transcripts and thus loss of Rspo1 function.

RSPO1 mutations in human patients lead to palmoplantar hyperkeratosis and a predisposition to develop squamous cell carcinoma. Initial analysis of adult Rspo1−/− mice (10 weeks old) did not reveal skin defects of the footpads.

Development of ovotestes in XX Rspo1−/− mice

We next analyzed the phenotype of the genital system in Rspo1−/− mice. A comparison of the phenotype on a mixed versus pure (SV129) genetic background showed no differences. Hence, all further analyses were performed on a mixed C57Bl6/SV129 genetic background. The sex ratio, litters size and number of adult Rspo1−/− mice were normal. XY Rspo1−/− mice appeared phenotypically normal and fertile.

To determine at what stage masculinization occurs, we examined both the morphology and the histology of the XX Rspo1−/− reproductive system at various stages of development. In normal XY gonads, a coelomic vessel forms at E12.5 through endothelial cell migration from the mesonephros (16). This blood vessel also formed on XX Rspo1−/− gonads at E12.5 and became prominent at E13.5 (Fig. 3A). XX Rspo1−/− gonads were smaller than XY testes of the same age and resembled XX gonadal tissue in shape. Histological analysis revealed an absence of sex cords at this early age (data not shown).

Figure 3.

Masculinization of XX Rspo1−/− urogenital systems. (A) Mesonephroi and gonads dissected from E13.5 embryos (magnification ×20; insets ×63). In XX Rspo1−/− gonads, an ectopic coelomic vessel formed as in wild-type XY gonads (red arrowheads and insets) (G = Gonad). (B) PAS staining of sections from E18.5 gonads. Sex cords containing G1-quiescent gonocytes (blue arrowheads) were detected in XY Rspo1+/+, XY Rspo1−/− and XX Rspo1−/− gonads and absent in XX Rspo1+/+ gonads (magnification ×40 and ×63). (C) Upper panels: Macroscopic view of XX and XY urogenital systems of wild-type and Rspo1−/− mice (magnification ×5; insets at ×20). Note the persistence of Wolffian and Mullerian ducts and the presence of an ovotestis surrounded by an epididymis in XX Rspo1−/− animals (Insets and Fig. 5D). (B: bladder, E: epididymis, SV: seminal vesicles, O: ovary, Ov: oviduct T: testis, U: uterus, VD: vas deferens). Lower panels: Hematoxylin and eosin staining of the gonads from the same animals (magnification ×10 and ×100). St: seminiferous tubules; of: ovarian follicle; cl: corpus luteum; Sc: Sertoli cells; Lc: Leydig cells; oo: oocyte; af: atretic follicle.

Figure 3.

Masculinization of XX Rspo1−/− urogenital systems. (A) Mesonephroi and gonads dissected from E13.5 embryos (magnification ×20; insets ×63). In XX Rspo1−/− gonads, an ectopic coelomic vessel formed as in wild-type XY gonads (red arrowheads and insets) (G = Gonad). (B) PAS staining of sections from E18.5 gonads. Sex cords containing G1-quiescent gonocytes (blue arrowheads) were detected in XY Rspo1+/+, XY Rspo1−/− and XX Rspo1−/− gonads and absent in XX Rspo1+/+ gonads (magnification ×40 and ×63). (C) Upper panels: Macroscopic view of XX and XY urogenital systems of wild-type and Rspo1−/− mice (magnification ×5; insets at ×20). Note the persistence of Wolffian and Mullerian ducts and the presence of an ovotestis surrounded by an epididymis in XX Rspo1−/− animals (Insets and Fig. 5D). (B: bladder, E: epididymis, SV: seminal vesicles, O: ovary, Ov: oviduct T: testis, U: uterus, VD: vas deferens). Lower panels: Hematoxylin and eosin staining of the gonads from the same animals (magnification ×10 and ×100). St: seminiferous tubules; of: ovarian follicle; cl: corpus luteum; Sc: Sertoli cells; Lc: Leydig cells; oo: oocyte; af: atretic follicle.

At E18.5, normal seminiferous cords that contained gonocytes were present in XY Rspo1−/− gonads (Fig. 3B). XX Rspo1−/− gonads were smaller, but contained several clear seminiferous tubules in addition to some less-developed cord structures albeit with few gonocytes within. Interestingly, some gonocytes were also detected outside of cord structures and resembled quiescent G1 gonocytes that are typical for XY gonads of this age. However, sex reversal was incomplete and part of the gonads showed nests with pachytene oocytes usually found in E18.5 XX gonads.

Adult XX Rspo1−/− showed external masculinized genitalia with an increased distance from vagina to anus (data not shown). Macroscopic inspection of 10 week old XX Rspo1−/− animals demonstrated persistent Wolffian and Mullerian ducts with epididymis, vas deferens, small seminal vesicles and prostate being associated with oviduct, uterus and vaginal tissues, respectively (Fig. 3C). Strikingly, the gonads resembled small testes. Histological analyses revealed the presence of few ovarian follicles, as well as seminiferous tubules with Sertoli and Leydig cells (Fig. 3C). As expected, tubules were devoid of germ cells due to the incompatibility of XX germ cells with a male somatic environment (35).

These data demonstrate that Rspo1 is required for female development. Its ablation impairs the differentiation into ovaries and triggers the development of ovotestes in XX mice.

Rspo1 activates β-catenin signaling during ovarian development

As shown above, β-catenin signaling is specifically activated during female sex determination. To confirm that Rspo1 mediates this activation, we measured the levels of expression of Lef1 by QPCR and in situ hybridization experiments (Fig. 4A and B). While high levels of Lef1 were found in XX gonads at E12.5 and E14.5 (Figs 1C and 4A), expression was downregulated in the knockout tissue (Fig. 4A and B). Strikingly, using the Axin2+/LacZ reporter, we observed a complete absence of staining in E13.5 XX Rspo1−/−; Axin2+/LacZ gonads (Fig. 4A). This downregulation of Axin2 was confirmed by QPCR at E14.5 (Fig. 4A). Interestingly, in XX Wnt4−/−; Axin2+/LacZ gonads some blue staining persisted in contrast to XX Rspo1−/−; Axin2+/LacZ gonads (compare Figs 1D and 4A). These data suggest that while Rspo1 is essential for ovarian β-catenin activation, Wnt4 is only contributing to this pathway. Wnt9a has been shown to be up-regulated during ovarian differentiation and this gene may partially complement loss of Wnt4 action (36). Using QPCR, we measured the level of expression of Wnt9a in XX Rspo1−/− and wild-type gonads (Fig. 4A). Wnt9a expression was down-regulated like Wnt4 at E12.5.

Figure 4.

Rspo1 activates β-catenin signaling during ovarian development. (A) X-Gal whole-mount staining of urogenital ridges at E13.5. While strong blue staining was detected in XX gonads, XY and XX Rspo1−/− gonads remained unstained. QPCR analysis of genes involved in the β-catenin signaling pathway in XX Rspo1/ and wild-type gonads at E12.5 (wild type, n = 6; Rspo1−/−, n = 8) and E14.5 (wild type, n = 4; Rspo1−/−, n = 10). NC: no change in gene expression between KO and Wt samples. Gene expression levels were normalized against expression levels of Hprt1. β-catenin target genes were down-regulated in XX Rspo1−/− gonads. (B) Lef1 In situ hybridization at E14.5 (magnification ×20, inset ×40). XX gonads showed strong Lef1 expression, whereas no signal was detected in both XY control and XX mutant gonads. (C) Upper panels: Macroscopic view of adult XX urogenital systems of wild type, Rspo1−/− and Rspo1−/−; Sf1:CreTR; Catnbex3/+ mice (magnification ×5). The gonads are circled. Ova: Ovary; OT; Ovotestis. Lower panels: H&E staining of the gonads from the same animals (magnification ×10, insets ×20). The dotted line circles the region containing seminiferous tubules. No sex cords were present in the XX Rspo1−/−; Sf1:CreTR; Catnbex3/+ ovaries whereas they exhibit ovarian follicles (antral follicles in insets) demonstrating rescue of the Rspo1 knockout phenotype by ectopic activation of β-catenin. T: testicular part.

Figure 4.

Rspo1 activates β-catenin signaling during ovarian development. (A) X-Gal whole-mount staining of urogenital ridges at E13.5. While strong blue staining was detected in XX gonads, XY and XX Rspo1−/− gonads remained unstained. QPCR analysis of genes involved in the β-catenin signaling pathway in XX Rspo1/ and wild-type gonads at E12.5 (wild type, n = 6; Rspo1−/−, n = 8) and E14.5 (wild type, n = 4; Rspo1−/−, n = 10). NC: no change in gene expression between KO and Wt samples. Gene expression levels were normalized against expression levels of Hprt1. β-catenin target genes were down-regulated in XX Rspo1−/− gonads. (B) Lef1 In situ hybridization at E14.5 (magnification ×20, inset ×40). XX gonads showed strong Lef1 expression, whereas no signal was detected in both XY control and XX mutant gonads. (C) Upper panels: Macroscopic view of adult XX urogenital systems of wild type, Rspo1−/− and Rspo1−/−; Sf1:CreTR; Catnbex3/+ mice (magnification ×5). The gonads are circled. Ova: Ovary; OT; Ovotestis. Lower panels: H&E staining of the gonads from the same animals (magnification ×10, insets ×20). The dotted line circles the region containing seminiferous tubules. No sex cords were present in the XX Rspo1−/−; Sf1:CreTR; Catnbex3/+ ovaries whereas they exhibit ovarian follicles (antral follicles in insets) demonstrating rescue of the Rspo1 knockout phenotype by ectopic activation of β-catenin. T: testicular part.

To confirm that Rspo1 acts through the β-catenin signaling pathway, we next tested whether ectopic activation of β-catenin was able to rescue the abnormal masculinization in XX Rspo1−/− gonads. A conditional allele that allows stabilization of β-catenin by ablation of exon 3 (Catnbex3/+) has been described previously (37). To activate the floxed allele, we made use of an Sf1:cre line that in addition to strong gonad specific expression, shows widespread expression in the developing embryo (4). The majority of Catnbex3/+; Sf1:creTr embryos died during early embryonic development due to extensive activation of the canonical β-catenin signaling pathway. However, two XX Rspo1−/−; Catnbex3/+; Sf1:creTr were obtained at E18.5 and adult. Macroscopic and histological analyses of the genital system of the 11-week-old female showed the presence of primordial, primary, secondary and antral follicles and an absence of sex cords, as expected for normal ovaries (Fig. 4C). Finally, nine embryos were implanted into the uterus thus further confirming the normal function of the rescued ovaries. Taken together these data show that Rspo1 is essential for the activation of the canonical β-catenin signaling pathway.

Rspo1 controls Wnt4 expression in XX gonad

To understand the relationship between Rspo1 and Wnt4, we examined the level of Wnt4 expression in our knockout model. In XX gonads Wnt4 is activated from E11.5. At this stage, Rspo1 is also expressed in the XX gonad whereas its expression is restricted to the coelomic epithelium of the XY gonad (Fig. 5A). In situ hybridizations revealed that Wnt4 expression is down-regulated at E11.5 and E12.5 in XX Rspo1−/− gonads (Fig. 5B). In contrast, Wnt4 expression persisted in Rspo1−/− mesonephroi suggesting that activation in this tissue occurs by an Rspo1-independent pathway. At E14.5, expression differences between XX and XY gonads were less pronounced, but levels in XX Rspo1−/− gonads remained lower than in XX wild-type littermates (Fig. 5B and data not shown).

Figure 5.

Rspo1 regulates female somatic differentiation. (A) Whole-mount in situ hybridizations of Rspo1 at E11.5 (19 tail somites (ts)). Rspo1 was expressed throughout the XX gonads, whereas it was restricted to the coelomic epithelium in XY gonads. (B) Whole-mount and section in situ hybridization using Wnt4 as a probe. At E11.5 (ts19) and E12.5, XX Rspo1−/− gonads exhibited a complete absence of staining. At E14.5, Wnt4 expression was down-regulated in XX Rspo1−/− gonads. (C) QPCR analysis of female-specific and male steroidogenic-specific genes in E12.5 gonads (wild type, n = 6; Rspo1−/−, n = 8) and E14.5 (wild type, n = 4; Rspo1−/−, n = 10). NC: no change in gene expression between KO and Wt samples. Gene expression levels were normalized against expression levels of Hprt1. Note the down-regulation of Fst, a target of Wnt4, in Rspo1−/− gonads and the up-regulation of Cyp11α1 and Hsd17β3, two enzymes of the male steroidogenesis pathway. In situ hybridizations of P450Scc, a steroidogenic cell marker, demonstrated ectopic expression in XX Rspo1−/− gonads. (D) Hematoxylin and eosin staining of XX Rspo1−/− showing the presence of epididymis and oviduct and uterus. E: Epididymis; G: Gonad; K; Kidney; M: Mesonephros; Ovi: Oviduct; U: Uterus.

Figure 5.

Rspo1 regulates female somatic differentiation. (A) Whole-mount in situ hybridizations of Rspo1 at E11.5 (19 tail somites (ts)). Rspo1 was expressed throughout the XX gonads, whereas it was restricted to the coelomic epithelium in XY gonads. (B) Whole-mount and section in situ hybridization using Wnt4 as a probe. At E11.5 (ts19) and E12.5, XX Rspo1−/− gonads exhibited a complete absence of staining. At E14.5, Wnt4 expression was down-regulated in XX Rspo1−/− gonads. (C) QPCR analysis of female-specific and male steroidogenic-specific genes in E12.5 gonads (wild type, n = 6; Rspo1−/−, n = 8) and E14.5 (wild type, n = 4; Rspo1−/−, n = 10). NC: no change in gene expression between KO and Wt samples. Gene expression levels were normalized against expression levels of Hprt1. Note the down-regulation of Fst, a target of Wnt4, in Rspo1−/− gonads and the up-regulation of Cyp11α1 and Hsd17β3, two enzymes of the male steroidogenesis pathway. In situ hybridizations of P450Scc, a steroidogenic cell marker, demonstrated ectopic expression in XX Rspo1−/− gonads. (D) Hematoxylin and eosin staining of XX Rspo1−/− showing the presence of epididymis and oviduct and uterus. E: Epididymis; G: Gonad; K; Kidney; M: Mesonephros; Ovi: Oviduct; U: Uterus.

As levels of Rspo1 expression at E12.5 and E14.5 were unchanged in XX Wnt4−/− gonads (Fig. 1D), Rspo1 appears to be activated independently and upstream of Wnt4. Taken together, these data indicate that female specific expression of Wnt4 in gonads depends on Rspo1.

Rspo1 prevents XY-like steroidogenesis and vascularization

Since Rspo1 appears to act upstream of Wnt4, it was important to test whether pathways down-regulated in the Wnt4 mutant were also affected in the Rspo1 mutant. XX Wnt4−/− gonads contain steroidogenic cells that have migrated from the mesonephros indicating that Wnt4 is required to prevent this migration in XX gonads (16). These cells synthesize steroids required for the development of the Wolffian derivatives.

In XX Rspo1−/− embryos, the development of the epididymis, vas deferens and seminal vesicles (Figs 3C and 5D) suggests the presence of steroidogenic cells within the gonad at an early stage of development. Indeed, levels of Cyp11α1 and Hsd17β3, two enzymes of the steroidogenic pathway which are up-regulated in the XX Wnt4−/− gonads (18), were also increased in XX Rspo1−/− gonads at E12.5 (Fig. 5C). Similarly, in situ hybridizations with P450Scc, another marker of the steroidogenic pathway (Fig. 5C) revealed positive cells in XX Rspo1−/− and XY littermates, but not in XX gonads.

Expression of follistatin (Fst) is induced during early ovarian differentiation and is required to prevent coelomic vessel formation in the XX gonad (38). As expected from the male-specific vascularization observed in knockout mice (Fig. 3A), XX Rspo1−/− gonads showed a dramatic down-regulation of Fst as early as E12.5 (Fig. 5C).

Thus, both steroidogenic and endothelial cells show ectopic migration into XX Rspo1 knockout gonads, a phenotype similar to that found in XX Wnt4−/− embryos (16,18).

Rspo1 regulates female germ-cell differentiation

To understand the cellular changes in Rspo1 mutant gonads, we next analyzed the cellular localization of the dephosphorylated (active) form of β-catenin (nuclear, cytoplasmic and membrane-bound forms) within gonads using immunofluorescence. While in XY gonads, the strongest staining was found at the interface of germ cells (plasma membrane), in XX gonads it appeared mostly cytosolic and to some extent nuclear at E12.5 and 14.5 (Fig. 6A). This is consistent with the robust transcriptional activation of Axin2 in XX somatic cells reported in Figure 1B. In XX Rspo1−/− gonads β-catenin localized at the surface of germ cells with a proportion showing a very strong and sharp signal in an XY specific manner. Expression of E-cadherin, a previously identified germ-cell marker, decreases once XX germ cells have entered meiosis at E15.5 (39). Immunostaining for E-cadherin demonstrated a marked difference between male and female gonads at E14.5 but not at E12.5 (Fig. 6A). This result suggests that β-catenin and E-cadherin are involved in maintaining adhesion between germ cells before commitment to meiosis. Taken together these data indicate that Rspo1 participates to the loss of adherent junctions in female germ cells that normally precedes meiosis.

Figure 6.

Rspo1 ablation interferes with female germ cell differentiation. (A) Immunodetection of the dephosphorylated active forms of β-catenin (red) (upper panel), and Mvh (green) (upper panel) in gonads at E12.5 (magnification ×40). Active β-catenin was predominantly detected at the plasma membrane of XY and XX Rspo1−/− germ cells (blue arrowheads) and in the cytoplasm and nucleus of somatic XX cells (white arrowhead). Immunodetection of E-cadherin (red) and Mvh (green) in gonads at E12.5 (magnification ×20). Dapi (blue) was used to detect nuclei. Immunodetection of β-catenin (red) and Mvh (green) (upper panel) (magnification ×63), and E-cadherin (red) and Mvh (green) (lower panel) (magnification ×20) in gonads at E14.5. β-catenin was predominantly expressed at the membrane of germ cells in XY Rspo1+/+ and XX Rspo1−/− (blue arrowheads), whereas it was more cytoplasmic and nuclear in XX Rspo1+/+ gonads (white arrowhead). Similar as in XY gonads, germ cells of XX Rspo1−/− gonads showed strong β-catenin and E-cadherin staining at the membranes between germ cells. (B) Quantitative RT–PCR expression analysis of meiotic markers Oct4, Cyp26b1 and Stra8 at E14.5 (wild type, n = 4; Rspo1−/−, n = 10). Gene expression levels were normalized against expression levels of Mvh (for Oct4 and Stra8) and Hprt1 (for Cyp26b1). Oct4 and Cyp26b1 expression levels were significantly increased in XX Rspo1−/− gonads compared to their XX wild-type littermates, but did not reach those observed in XY control gonads. Stra8 expression levels were not significantly changed in Rspo1−/− gonads. Asterisks indicate levels of statistical significance: two asterisks, P < 0.01; three asterisks, P < 0.0001. In situ hybridization analysis confirmed unchanged expression levels of Stra8 in XX Rspo1−/− gonads.

Figure 6.

Rspo1 ablation interferes with female germ cell differentiation. (A) Immunodetection of the dephosphorylated active forms of β-catenin (red) (upper panel), and Mvh (green) (upper panel) in gonads at E12.5 (magnification ×40). Active β-catenin was predominantly detected at the plasma membrane of XY and XX Rspo1−/− germ cells (blue arrowheads) and in the cytoplasm and nucleus of somatic XX cells (white arrowhead). Immunodetection of E-cadherin (red) and Mvh (green) in gonads at E12.5 (magnification ×20). Dapi (blue) was used to detect nuclei. Immunodetection of β-catenin (red) and Mvh (green) (upper panel) (magnification ×63), and E-cadherin (red) and Mvh (green) (lower panel) (magnification ×20) in gonads at E14.5. β-catenin was predominantly expressed at the membrane of germ cells in XY Rspo1+/+ and XX Rspo1−/− (blue arrowheads), whereas it was more cytoplasmic and nuclear in XX Rspo1+/+ gonads (white arrowhead). Similar as in XY gonads, germ cells of XX Rspo1−/− gonads showed strong β-catenin and E-cadherin staining at the membranes between germ cells. (B) Quantitative RT–PCR expression analysis of meiotic markers Oct4, Cyp26b1 and Stra8 at E14.5 (wild type, n = 4; Rspo1−/−, n = 10). Gene expression levels were normalized against expression levels of Mvh (for Oct4 and Stra8) and Hprt1 (for Cyp26b1). Oct4 and Cyp26b1 expression levels were significantly increased in XX Rspo1−/− gonads compared to their XX wild-type littermates, but did not reach those observed in XY control gonads. Stra8 expression levels were not significantly changed in Rspo1−/− gonads. Asterisks indicate levels of statistical significance: two asterisks, P < 0.01; three asterisks, P < 0.0001. In situ hybridization analysis confirmed unchanged expression levels of Stra8 in XX Rspo1−/− gonads.

An important feature of ovarian differentiation is the onset of germ cell meiosis at E13.5 (8,15). Histological examination revealed that the majority of germ cells did not enter meiosis in XX Rspo1−/− gonads (Fig. 3B). Indeed, expression of Oct4, a pre-meiotic germ-cell marker, was increased in XX Rspo1−/− compared to XX wild-type gonads at E14.5 (Fig. 6B). However, Oct4 expression levels were lower than in XY controls confirming that only a proportion of germ cells in XX Rspo1−/− gonads failed to enter meiosis.

RA is required for the induction of meiosis. In XY gonads, induction is delayed by the expression of Cyp26b1, an enzyme that reduces local RA levels (12). In XX Rspo1−/− gonads, Cyp26b1 showed a 2.5-fold increase compared to XX gonads, but was more than 20 times weaker than in XY gonads, which is unlikely to be sufficient to diminish the RA concentration. Moreover, QPCR and in situ hybridization experiments indicated that expression of Stra8, a downstream target of RA signaling (10,11), was not significantly changed between XX and XX Rspo1−/− (Fig. 6B). These results suggest that the RA pathway is not the main cause of the blockage of meiosis in the XX Rspo1−/− gonads.

Rspo1 is a suppressor of male sex determination

To decipher the molecular pathways involved in somatic differentiation in Rspo1−/− mice, we examined expression levels of genes with a known function during female and male sex determination.

Levels of Bmp2 and FoxL2, two markers of the XX gonad, were similar between XX Rspo1−/− and wild-type gonads at E12.5 and E14.5 (Fig. 5C) suggesting that the somatic lineage has partially kept its female identity in the absence of Rspo1. This was expected since a part of the XX Rspo1−/− gonad exhibited ovarian characteristics with follicles in adults (Fig. 3C).

In XY gonads, activation of Sox9 involves a combination of nuclear import of its protein and an increase of its transcription (2,40). Both of these mechanisms seem to be enhanced by prostaglandin D2, a hormone synthesized by Pgds (2). At E12.5, Pgds, Sox9 and Amh/Mis expression were slightly increased in XX Rspo1−/− gonads (Fig. 7A and B), but insufficient to allow sex cord formation and early testis development. Insufficient levels of Amh/Mis were expected given the persistence of the Mullerian ducts (Figs 3C and 5D). Fgf9 is involved in testis-specific proliferation and the maintenance of Sox9 expression (6). At E12.5, Fgf9 was undetectable in XX wild-type and XX Rspo1−/− gonads, which is consistent with the lack of male-specific proliferation (Fig. 7A). However, at E14.5 and E18.5 both Fgf9 and Pgds levels were increased (Fig. 7A and C). Since both of these factors are able to activate Sox9, the expression of these genes in Rspo1−/− gonads is likely to contribute to the expression of Sox9 and Amh/Mis expression in the absence of Sry (Fig. 7C).

Figure 7.

Rspo1 is a suppressor of male sex determination. (A) QPCR expression analysis of male-specific genes in E12.5 and E14.5 gonads (wild type, n = 4; Rspo1−/−, n = 10). Gene expression levels were normalized against expression levels of Hprt1. Note the up-regulation of genes in XX Rspo1−/− gonads that are normally specifically expressed in XY gonads. NC: no change in gene expression between KO and Wt samples. (B) In situ hybridization of Sox9 and immunofluorescence analysis of Sox9 and Amh/Mis at E11.5 and 12.5 (magnification ×20, insets at ×40). Sox9 and Amh/Mis were undetectable in XX Rspo1+/+ and XX Rspo1−/− gonads. (C) In situ hybridization of Pgds, Sox9 and immunofluorescence analysis of Sox9 and Amh/Mis at E18.5 (magnification ×20, insets at ×40). XX Rspo1−/− gonads exhibited Pgds, Sox9 and Amh expression in sex cords. G: gonad, M: mesonephros; K: kidney.

Figure 7.

Rspo1 is a suppressor of male sex determination. (A) QPCR expression analysis of male-specific genes in E12.5 and E14.5 gonads (wild type, n = 4; Rspo1−/−, n = 10). Gene expression levels were normalized against expression levels of Hprt1. Note the up-regulation of genes in XX Rspo1−/− gonads that are normally specifically expressed in XY gonads. NC: no change in gene expression between KO and Wt samples. (B) In situ hybridization of Sox9 and immunofluorescence analysis of Sox9 and Amh/Mis at E11.5 and 12.5 (magnification ×20, insets at ×40). Sox9 and Amh/Mis were undetectable in XX Rspo1+/+ and XX Rspo1−/− gonads. (C) In situ hybridization of Pgds, Sox9 and immunofluorescence analysis of Sox9 and Amh/Mis at E18.5 (magnification ×20, insets at ×40). XX Rspo1−/− gonads exhibited Pgds, Sox9 and Amh expression in sex cords. G: gonad, M: mesonephros; K: kidney.

DISCUSSION

Sex determination is a unique process that determines the differentiation of a bipotential organ, which in turn triggers the full sexual development of an individual. Here we have shown that Rspo1 is required for Wnt4 expression in XX gonads and acts as a key regulator of β-catenin activation in female sex determination. The ablation of these genes induces differentiation of seminiferous tubules in XX gonads [present study and (17)], thus demonstrating their requirement to suppress male sexual determination. Mechanistically, RSPO1 appears to interact with Kremen, a receptor that is involved in DKK mediated internalization of LRP6. This interaction inhibits LRP6 internalization thus providing increased receptor levels at the cell surface (26). Ablation of Rspo1 has a more dramatic effect on β-catenin activation than Wnt4, which suggests the existence of functional redundancy between Wnt ligands. Alternatively, Rspo1 may activate β-catenin signaling in a Wnt-independent manner. In vitro RSPO1 binding induces LRP6 phosphorylation, which in turn favors stabilization of cytosolic β-catenin (41). Interestingly, LRP6 ablation has been described to trigger abnormal ovary development (42). To our knowledge, the gonadal phenotype in Lrp6−/− animals has not been analyzed in detail and it will be interesting to address whether aspects of the Rspo1 knockout phenotype is recapitulated in this mutant.

Once in the nucleus, β-catenin can interact with transcription factors, such as Lef/TCF, which trigger the transcriptional regulation of target genes. While Lef1 is clearly up-regulated in female gonads (this study), so far no gonadal phenotype has been reported in Lef1 knockout mice and it is likely that at least part of the transcriptional control of β-catenin is mediated through its interaction with other, yet unidentified cofactors.

We have previously shown that Rspo1 is expressed in XX somatic cells of the gonad (23). Here we show that this secreted protein activates the canonical β-catenin signaling pathway in the somatic cells. Absence of Rspo1 in XX gonads prevents up-regulation of Wnt4 and results in the presence of steroidogenic cells and the formation of a coelomic vessel. As XX Wnt4−/− gonads also show this phenotype, it is likely that ectopic migration of steroidogenic and endothelial cells is the result of Wnt4 down-regulation in the Rspo1−/− embryos.

The second striking observation is that some germ cells fail to enter meiosis. Commitment to meiosis starts at E13.5 in females and is stimulated by the RA pathway and its regulator Stra8 (10,12). In XX Rspo1−/− gonads, the RA signaling pathway does not seem to be dramatically affected. In addition, as XX Stra8−/− mice do not exhibit sex reversal (11) the partial sex reversal phenotype in Rspo1−/− mice is unlikely to be due to a simple block of meiosis.

At E12.5 immunofluorescent staining for β-catenin and at E14.5 β-catenin and E-cadherin show that cell–cell adhesion complexes between germ cells in XX Rspo1−/− animals are more abundant than in XX controls and are organized in a similar manner as in the XY gonad. These changes of adherens junctions are likely to be involved in the commitment to meiosis.

Indeed, E-cadherin has been shown to be down-regulated in XX meiotic embryonic germ cells (39). Down-regulation of cell–cell adhesions can be orchestrated by different factors. One pathway involves tyrosine kinase receptors (43). Interestingly, the receptors ErbB2 and ErbB3 are both expressed in germ cells and are down-regulated at E14.5 suggesting that they might be involved in meiosis commitment (44). Whatever the pathway, our data suggest that Rspo1 is regulating germ-cell fate. To know whether this regulation is direct, or whether it involves a secondary signal that is released from somatic cells after activation of β-catenin, warrants further investigation. Misregulation of the XX germ-cell fate in XX Rspo1−/− gonads results in the presence of G1 gonocytes that are specific to the embryonic testis.

XX meiotic germ cells have been shown to prevent sex cord formation (15) suggesting that the identity of germ cells as meiotic oogonia is required to block masculinisation of the somatic lineage. The lack of such an identity of some germ cells in the Rspo1−/− gonads can allow partial sex reversal to occur. Thus, the occurrence of these gonocytes is followed by masculinization of somatic supporting cells and formation of sex cords around birth. Gonocytes release signals that induce supporting cells to differentiate into Sertoli cells (14). The effector appears to be prostaglandin D2 that is synthetized by Pgds and is able to activate Sox9 (5). We have shown that Pgds is upregulated in early XX Rspo1−/− gonads and a strong expression is detected at E18.5. High levels of expression of Sox9 and the formation of sex cords in XX Rspo1−/- gonads appear to be secondary events, which could be triggered by the ectopic expression of Pgds. Sox9 and Pgds regulate each other in a positive feed-back loop (5,6) and the expression of these genes has to reach a critical threshold to allow sex cord formation. However, we cannot exclude that other signals such as steroids are responsible for masculinization of the somatic cells.

As shown above, Rspo1 is involved in the female specific up-regulation of Wnt4. Rspo1 knock-out animals partially recapitulate the phenotype of the Wnt4 mutants. However, Wnt4 is also required for the formation of the Mullerian duct, a process that is not under the control of Rspo1. As a result Rspo1−/− animals show hermaphroditism of the ducts. It is noteworthy that 90% of XX Wnt4−/− germ cells die before birth, whereas their number is grossly normal in XX Rspo1−/− gonads. While sexually dimorphic expression of Wnt4 in the early embryo is under control of Rspo1, levels are normal from E14.5 onwards. Wnt4 expression at this age is therefore likely to be involved in germ-cell survival.

In this study, we show that Rspo1 acts at the top of the female sex determination pathway. However, ablation of Rspo1 results in formation of an ovotestis. This is likely due to the partial and late differentiation of the Sertoli lineage. At the molecular level, this could be due to the involvement of an Rspo1-independent pathway. FoxL2 is also involved in ovarian differentiation. Although Foxl2 expression is independent of Rspo1, XX Wnt4−/−FoxL2−/− show a gonadal phenotype at birth (45) that is comparable to Rspo1−/− mutants. This raises the question whether mutations of the regulator of Rspo1 and/or FoxL2 would have a more dramatic effect. The future identification of this regulator will address this question.

Taken together, our results indicate that while the male pathway is directed through somatic events, female differentiation requires a crosstalk between somatic and germinal lineages. Moreover, our data support a model in which sex determination is controlled by a balance between two very different pathways (6). Sry is clearly a major weight that induces a cascade resulting in male differentiation. Here we have shown that Rspo1 induces β-catenin signaling to tip the balance to the female side.

MATERIALS AND METHODS

Generation of Rspo1−/− mice

A targeting vector was constructed by fusing a LacZ gene from pCH110 (Promega) followed by a neomycine resistance in frame into exon3 of Rspo1 (details on request). After electroporation clones were selected on G418 (400 µg/ml), and homologous recombination events identified by PCR P20 (5′-TTTGATGCCTGACCCCTGAG) and LacZ50 (5′-AATATCGCGGCTCATTCGAGG-3′) and confirmed by Southern blotting. Blastocyst injections allowed the generation of two independent lines using C57/Bl6J donors.

Mouse strains

Wnt4 knock-out mice were generated by (46), Axin2+/LacZ transgenic mice by (31). Mice carrying the β-catenin exon 3 floxed allele (Catnbex3/+) generated by (37) were mated with Rspo1+/− mice. Sf1:CreTR mice previously described by (4) were mated with Rspo1+/−; Catnbex3/+ mice to obtain Rspo1−/−; Sf1:CreTR; Catnbex3/+ embryos.

Genotyping of embryos and mice

The Rspo1 alleles were determined using primers P2 (5′-ATCCAGGGTCCCTCTTGATC-3′) and P12 (5′-TTGAGGCAACCGTTGACTTC-3′) for the wild type, and primers P2 and LacZ50 (see above) for the mutant. Axin2+/LacZ mice were genotyped using forward Axin2 primer 5′-AAGCTGCGTCGGATACTTGAGA-3′, reverse Axin2 primer 5′-AGTCCATCTTCATTCCGCCTAGC-3′ and reverse LacZ primer 5′-TGGTAATGCTGCAGTGGCTTG-3′ (sequences kindly provided by Michel V. Hadjihannas). Wnt4 and Catnbex3/+ alleles were identified as published (37,46). The presence of the Sf1:Cre transgene and the presence of the Y chromosome (Sry PCR) were determined as described in (4).

X-gal staining, histological and immunological analyses

X-Gal staining was performed according to (47).

Urogenital organs were dissected and fixed either in Bouin’s solution or in 4% paraformaldehyde overnight at 4°C, then dehydrated, placed in xylene and paraffin embedded. Sections of 7 µm thickness were performed and HE staining, or PAS staining performed as published (34). Immunohistochemical experiments were performed as described (4). The following dilutions of primary antibodies were used: Sox9 (provided by Michael Wegner), 1:1000, Amh (C-20, cat sc6886, Santa Cruz), 1:100, E-cadherin (cat 610182, BD Transduction Laboratory), 1:100, DDX4/Mvh (cat 13840, Abcam), 1:200, β-catenin (anti-ABC clone 8E7, cat 05-665, Upstate).

In situ hybridization

In situ hybridizations on paraformaldehyde fixed/paraffin embedded sections or on wholemount embryos were carried out essentially as described (4). Sox9 riboprobes were synthesized according to da Silva et al. (48). The plasmids used to synthetize Wnt4, Lef1, Stra8 and Pgds riboprobes were from Seppo Vainio, Jörg Hülsken, David Page and Peter Koopman, respectively.

Quantitative PCR analysis

Individual gonads without mesonephros were dissected in PBS from E12.5 and 14.5 embryos and immediately frozen at −80°C. RNA was extracted using the RNeasy Qiagen kit, and reverse transcribed using the RNA RT–PCR kit (Stratagene). Primers and probes were designed by the Roche Assay design center (https://www.roche-applied-science.com/sis/rtpcr/upl/adc.jsp). Amh: primers 5′-ggggagactggagaacagc-3′ and 5′-agagctcgggctcccata-3′ (probe 41), Axin2: 5′-gcaggagcctcacccttc-3′ and 5′-tgccagtttctttggctctt-3′ (probe 50), Bmp2: 5′-cggactgcggtctcctaa-3′ and 5′-ggggaagcagcaacactaga-3′ (probe 49), Cyp11α1: 5′-aagtatggccccatttacagg-3′ and 5′-tggggtccacgatgtaaact-3′ (probe 104), Cyp26b1: 5′-acatccaccgcaacaagc-3′ and 5′-gggcaggtagctctcaagtg-3′ (probe 41), Dax1: 5′-cgtgctctttaacccagacc-3′ and 5′-ccggatgtgctcagtaagg-3′ (probe 3), Emx2: 5′-cacgcttttgagaagaacca-3′ and 5′-gttctccggttctgaaacca-3′ (probe 38), Fgf9: 5′-tgcaggactggatttcatttag-3′ and 5′-ccaggcccactgctatactg-3′ (probe 60), Foxl2: 5′-ggcgtcgtgaactcctaca-3′ and 5′-tgcagatgatgtgcgtgag-3′ (probe 51), Fst: 5′-tggattagcctatgagggaaag-3′ and 5′-tggaatcccataggcatttt-3′ (probe 47), Hprt1: 5′-tcctcctcagaccgctttt-3′ and 5′-cctggttcatcatcgctaatc-3′ (probe 95), Hsd17β3: 5′-aatatgtcacgatcggagctg-3′ and 5′-gaagggatccggttcagaat-3′ (probe 5), Lef1: 5′-tcctgaaatccccaccttc-3′ and 5′-acccgtgatgggataaacag-3′ (probe 94), Mvh: 5′-ccaagatcaggggacacagt-3′ and 5′-ctttggtaagtgtcaccattgc-3′ (probe 77), Oct4: 5′-gttggagaaggtggaaccaa-3′ and 5′-ctccttctgcagggctttc-3′ (probe 95), Pgds: 5′-ggctcctggacactacacct-3′ and 5′-atagttggcctccaccactg-3′ (probe 89), Rspo1: 5′-cgacatgaacaaatgcatca-3′ and 5′-ctcctgacacttggtgcaga-3′ (probe 5), Sox9: 5′-cagcaagactctgggcaag-3′ and 5′-tccacgaagggtctcttctc-3′ (probe 66), Stra8: 5′-accgtggtggccttaaaga-3′ and 5′-atcatcactgggttggttgc-3′ (probe 80), Wnt4: 5′-ctggactccctccctgtctt-3′ and 5′-atgcccttgtcactgcaaa-3′ (probe 62), Wnt9a: 5′-acctcgtgggtgtgaaggt-3′ and 5′-acctcgtggaagggtgcta-3′ (probe 62). All real-time PCR assays were carried out using the LC-Faststart DNA Master kit Roche. QPCR was performed on cDNA from one gonad and compared to a standard curve. The number of gonads per genotype is given in the Figure legends. QPCR were repeated at least twice. Relative expression levels of each sample were determined in the same run and normalized by measuring the amount of Hprt1 cDNA or of Mvh cDNA (to normalize germ-cell specific genes). The results were analyzed using Sigma plot and Graphpad for statistical relevance.

FUNDING

This work was supported by Association pour la Recherche sur le Cancer (Grant no. 3912), Agence Nationale pour la Recherche (ANR-06-GANI-005, TEGOD) and European Community (EuRoGene-FP6). A.A.C. was financed by La Ligue Nationale contre le Cancer (post-doctoral fellowship).

ACKNOWLEDGEMENTS

We are grateful to Valérie Vidal and Michael Clarkson for helpful discussions and Danielle Badro and Cédric Matthews for technical help. We are thankful to Amanda Swain (London, UK) for providing the Wnt4+/− mice, Walter Birchmeier (Berlin, Germany) for the Axin2+/LacZ mice, Michael Wegner (Erlangen, Germany) for the Sox9 antibody, Thomas Willnow (Berlin, Germany) for pPol2shortneopAHSVTk and ES cells, and Seppo Vainio (University of Oulu, Finland), Jörg Hülsken (Lausanne, Switzerland), David Page (Cambridge, USA) and Peter Koopman (Brisbane, Australia) for in situ hybridization probes.

Conflict of Interest statement. None declared.

REFERENCES

1
Koopman
P.
Gubbay
J.
Vivian
N.
Goodfellow
P.
Lovell-Badge
R.
Male development of chromosomally female mice transgenic for Sry
Nature
 , 
1991
, vol. 
351
 (pg. 
117
-
121
)
2
Wilhelm
D.
Martinson
F.
Bradford
S.
Wilson
M.J.
Combes
A.N.
Beverdam
A.
Bowles
J.
Mizusaki
H.
Koopman
P.
Sertoli cell differentiation is induced both cell-autonomously and through prostaglandin signaling during mammalian sex determination
Dev. Biol.
 , 
2005
, vol. 
287
 (pg. 
111
-
124
)
3
Sekido
R.
Bar
I.
Narvaez
V.
Penny
G.
Lovell-Badge
R.
SOX9 is up-regulated by the transient expression of SRY specifically in Sertoli cell precursors
Dev. Biol.
 , 
2004
, vol. 
274
 (pg. 
271
-
279
)
4
Chaboissier
M.C.
Kobayashi
A.
Vidal
V.I.
Lutzkendorf
S.
van de Kant
H.J.
Wegner
M.
de Rooij
D.G.
Behringer
R.R.
Schedl
A.
Functional analysis of Sox8 and Sox9 during sex determination in the mouse
Development
 , 
2004
, vol. 
131
 (pg. 
1891
-
1901
)
5
Wilhelm
D.
Hiramatsu
R.
Mizusaki
H.
Widjaja
L.
Combes
A.N.
Kanai
Y.
Koopman
P.
SOX9 regulates prostaglandin D synthase gene transcription in vivo to ensure testis development
J. Biol. Chem.
 , 
2007
, vol. 
282
 (pg. 
10553
-
10560
)
6
Kim
Y.
Kobayashi
A.
Sekido
R.
Dinapoli
L.
Brennan
J.
Chaboissier
M.C.
Poulat
F.
Behringer
R.R.
Lovell-Badge
R.
Capel
B.
Fgf9 and wnt4 act as antagonistic signals to regulate Mammalian sex determination
PLoS Biol.
 , 
2006
, vol. 
4
 pg. 
e187
 
7
Tilmann
C.
Capel
B.
Mesonephric cell migration induces testis cord formation and Sertoli cell differentiation in the mammalian gonad
Development
 , 
1999
, vol. 
126
 (pg. 
2883
-
2890
)
8
McLaren
A.
Southee
D.
Entry of mouse embryonic germ cells into meiosis
Dev. Biol.
 , 
1997
, vol. 
187
 (pg. 
107
-
113
)
9
Merchant-Larios
H.
Chimal-Monroy
J.
The ontogeny of primordial follicles in the mouse ovary
Prog. Clin. Biol. Res.
 , 
1989
, vol. 
296
 (pg. 
55
-
63
)
10
Koubova
J.
Menke
D.B.
Zhou
Q.
Capel
B.
Griswold
M.D.
Page
D.C.
Retinoic acid regulates sex-specific timing of meiotic initiation in mice
Proc. Natl Acad. Sci. USA
 , 
2006
, vol. 
103
 (pg. 
2474
-
2479
)
11
Baltus
A.E.
Menke
D.B.
Hu
Y.C.
Goodheart
M.L.
Carpenter
A.E.
de Rooij
D.G.
Page
D.C.
In germ cells of mouse embryonic ovaries, the decision to enter meiosis precedes premeiotic DNA replication
Nat. Genet.
 , 
2006
, vol. 
38
 (pg. 
1430
-
1434
)
12
Bowles
J.
Knight
D.
Smith
C.
Wilhelm
D.
Richman
J.
Mamiya
S.
Yashiro
K.
Chawengsaksophak
K.
Wilson
M.J.
Rossant
J.
, et al.  . 
Retinoid signaling determines germ cell fate in mice
Science
 , 
2006
, vol. 
312
 (pg. 
596
-
600
)
13
McLaren
A.
Development of the mammalian gonad: the fate of the supporting cell lineage
Bioessays
 , 
1991
, vol. 
13
 (pg. 
151
-
156
)
14
Adams
I.R.
McLaren
A.
Sexually dimorphic development of mouse primordial germ cells: switching from oogenesis to spermatogenesis
Development
 , 
2002
, vol. 
129
 (pg. 
1155
-
1164
)
15
Yao
H.H.
DiNapoli
L.
Capel
B.
Meiotic germ cells antagonize mesonephric cell migration and testis cord formation in mouse gonads
Development
 , 
2003
, vol. 
130
 (pg. 
5895
-
5902
)
16
Jeays-Ward
K.
Hoyle
C.
Brennan
J.
Dandonneau
M.
Alldus
G.
Capel
B.
Swain
A.
Endothelial and steroidogenic cell migration are regulated by WNT4 in the developing mammalian gonad
Development
 , 
2003
, vol. 
130
 (pg. 
3663
-
3670
)
17
Vainio
S.
Heikkila
M.
Kispert
A.
Chin
N.
McMahon
A.P.
Female development in mammals is regulated by Wnt-4 signalling
Nature
 , 
1999
, vol. 
397
 (pg. 
405
-
409
)
18
Heikkila
M.
Prunskaite
R.
Naillat
F.
Itaranta
P.
Vuoristo
J.
Leppaluoto
J.
Peltoketo
H.
Vainio
S.
The partial female to male sex reversal in Wnt-4-deficient females involves induced expression of testosterone biosynthetic genes and testosterone production, and depends on androgen action
Endocrinology
 , 
2005
, vol. 
146
 (pg. 
4016
-
4023
)
19
Pailhoux
E.
Vigier
B.
Chaffaux
S.
Servel
N.
Taourit
S.
Furet
J.P.
Fellous
M.
Grosclaude
F.
Cribiu
E.P.
Cotinot
C.
, et al.  . 
A 11.7-kb deletion triggers intersexuality and polledness in goats
Nat. Genet.
 , 
2001
, vol. 
29
 (pg. 
453
-
458
)
20
Schmidt
D.
Ovitt
C.E.
Anlag
K.
Fehsenfeld
S.
Gredsted
L.
Treier
A.C.
Treier
M.
The murine winged-helix transcription factor Foxl2 is required for granulosa cell differentiation and ovary maintenance
Development
 , 
2004
, vol. 
131
 (pg. 
933
-
942
)
21
Uda
M.
Ottolenghi
C.
Crisponi
L.
Garcia
J.E.
Deiana
M.
Kimber
W.
Forabosco
A.
Cao
A.
Schlessinger
D.
Pilia
G.
Foxl2 disruption causes mouse ovarian failure by pervasive blockage of follicle development
Hum. Mol. Genet.
 , 
2004
, vol. 
13
 (pg. 
1171
-
1181
)
22
Ottolenghi
C.
Omari
S.
Garcia-Ortiz
J.E.
Uda
M.
Crisponi
L.
Forabosco
A.
Pilia
G.
Schlessinger
D.
Foxl2 is required for commitment to ovary differentiation
Hum. Mol. Genet.
 , 
2005
, vol. 
14
 (pg. 
2053
-
2062
)
23
Parma
P.
Radi
O.
Vidal
V.
Chaboissier
M.C.
Dellambra
E.
Valentini
S.
Guerra
L.
Schedl
A.
Camerino
G.
R-spondin1 is essential in sex determination, skin differentiation and malignancy
Nat. Genet.
 , 
2006
, vol. 
38
 (pg. 
1304
-
1309
)
24
Kazanskaya
O.
Glinka
A.
del Barco Barrantes
I.
Stannek
P.
Niehrs
C.
Wu
W.
R-Spondin2 is a secreted activator of Wnt/beta-catenin signaling and is required for Xenopus myogenesis
Dev. Cell.
 , 
2004
, vol. 
7
 (pg. 
525
-
534
)
25
Kim
K.A.
Kakitani
M.
Zhao
J.
Oshima
T.
Tang
T.
Binnerts
M.
Liu
Y.
Boyle
B.
Park
E.
Emtage
P.
, et al.  . 
Mitogenic influence of human R-spondin1 on the intestinal epithelium
Science
 , 
2005
, vol. 
309
 (pg. 
1256
-
1259
)
26
Binnerts
M.E.
Kim
K.A.
Bright
J.M.
Patel
S.M.
Tran
K.
Zhou
M.
Leung
J.M.
Liu
Y.
Lomas
W.E.
3rd
Dixon
M.
, et al.  . 
R-Spondin1 regulates Wnt signaling by inhibiting internalization of LRP6
Proc. Natl Acad. Sci. USA
 , 
2007
, vol. 
104
 (pg. 
14700
-
14705
)
27
Nam
J.S.
Turcotte
T.J.
Smith
P.F.
Choi
S.
Yoon
J.K.
Mouse cristin/R-spondin family proteins are novel ligands for the Frizzled 8 and LRP6 receptors and activate beta-catenin-dependent gene expression
J. Biol. Chem.
 , 
2006
, vol. 
281
 (pg. 
13247
-
13257
)
28
Wei
Q.
Yokota
C.
Semenov
M.V.
Doble
B.
Woodgett
J.
He
X.
R-spondin1 is a high affinity ligand for LRP6 and induces LRP6 phosphorylation and beta-catenin signaling
J. Biol. Chem.
 , 
2007
, vol. 
282
 (pg. 
15903
-
15911
)
29
Mizusaki
H.
Kawabe
K.
Mukai
T.
Ariyoshi
E.
Kasahara
M.
Yoshioka
H.
Swain
A.
Morohashi
K.
Dax-1 (dosage-sensitive sex reversal-adrenal hypoplasia congenita critical region on the X chromosome, gene 1) gene transcription is regulated by wnt4 in the female developing gonad
Mol. Endocrinol.
 , 
2003
, vol. 
17
 (pg. 
507
-
519
)
30
Jordan
B.K.
Shen
J.H.
Olaso
R.
Ingraham
H.A.
Vilain
E.
Wnt4 overexpression disrupts normal testicular vasculature and inhibits testosterone synthesis by repressing steroidogenic factor 1/beta-catenin synergy
Proc. Natl Acad. Sci. USA
 , 
2003
, vol. 
100
 (pg. 
10866
-
10871
)
31
Lustig
B.
Jerchow
B.
Sachs
M.
Weiler
S.
Pietsch
T.
Karsten
U.
van de Wetering
M.
Clevers
H.
Schlag
P.M.
Birchmeier
W.
, et al.  . 
Negative feedback loop of Wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors
Mol. Cell. Biol.
 , 
2002
, vol. 
22
 (pg. 
1184
-
1193
)
32
Backman
M.
Machon
O.
Mygland
L.
van den Bout
C.J.
Zhong
W.
Taketo
M.M.
Krauss
S.
Effects of canonical Wnt signaling on dorso-ventral specification of the mouse telencephalon
Dev. Biol.
 , 
2005
, vol. 
279
 (pg. 
155
-
168
)
33
Park
J.S.
Valerius
M.T.
McMahon
A.P.
Wnt/beta-catenin signaling regulates nephron induction during mouse kidney development
Development
 , 
2007
, vol. 
134
 (pg. 
2533
-
2539
)
34
Vidal
V.P.
Chaboissier
M.C.
de Rooij
D.G.
Schedl
A.
Sox9 induces testis development in XX transgenic mice
Nat. Genet.
 , 
2001
, vol. 
28
 (pg. 
216
-
217
)
35
Hunt
P.A.
Worthman
C.
Levinson
H.
Stallings
J.
LeMaire
R.
Mroz
K.
Park
C.
Handel
M.A.
Germ cell loss in the XXY male mouse: altered X-chromosome dosage affects prenatal development
Mol. Reprod. Dev.
 , 
1998
, vol. 
49
 (pg. 
101
-
111
)
36
Nef
S.
Schaad
O.
Stallings
N.R.
Cederroth
C.R.
Pitetti
J.L.
Schaer
G.
Malki
S.
Dubois-Dauphin
M.
Boizet-Bonhoure
B.
Descombes
P.
, et al.  . 
Gene expression during sex determination reveals a robust female genetic program at the onset of ovarian development
Dev. Biol.
 , 
2005
, vol. 
287
 (pg. 
361
-
377
)
37
Harada
N.
Tamai
Y.
Ishikawa
T.
Sauer
B.
Takaku
K.
Oshima
M.
Taketo
M.M.
Intestinal polyposis in mice with a dominant stable mutation of the beta-catenin gene
EMBO J.
 , 
1999
, vol. 
18
 (pg. 
5931
-
5942
)
38
Yao
H.H.
Matzuk
M.M.
Jorgez
C.J.
Menke
D.B.
Page
D.C.
Swain
A.
Capel
B.
Follistatin operates downstream of Wnt4 in mammalian ovary organogenesis
Dev. Dyn.
 , 
2004
, vol. 
230
 (pg. 
210
-
215
)
39
Di Carlo
A.
De Felici
M.
A role for E-cadherin in mouse primordial germ cell development
Dev. Biol.
 , 
2000
, vol. 
226
 (pg. 
209
-
219
)
40
Malki
S.
Nef
S.
Notarnicola
C.
Thevenet
L.
Gasca
S.
Mejean
C.
Berta
P.
Poulat
F.
Boizet-Bonhoure
B.
Prostaglandin D2 induces nuclear import of the sex-determining factor SOX9 via its cAMP-PKA phosphorylation
EMBO J.
 , 
2005
, vol. 
24
 (pg. 
1798
-
1809
)
41
Wei
Q.
Yokota
C.
Semenov
M.V.
Doble
B.
Woodgett
J.
He
X.
R-spondin1 is a high affinity ligand for LRP6 and induces LRP6 phosphorylation and beta -catenin signaling
J. Biol. Chem.
 , 
2007
, vol. 
282
 (pg. 
15903
-
15911
)
42
Pinson
K.I.
Brennan
J.
Monkley
S.
Avery
B.J.
Skarnes
W.C.
An LDL-receptor-related protein mediates Wnt signalling in mice
Nature
 , 
2000
, vol. 
407
 (pg. 
535
-
538
)
43
Nelson
W.J.
Nusse
R.
Convergence of Wnt, beta-catenin, and cadherin pathways
Science
 , 
2004
, vol. 
303
 (pg. 
1483
-
1487
)
44
Toyoda-Ohno
H.
Obinata
M.
Matsui
Y.
Members of the ErbB receptor tyrosine kinases are involved in germ cell development in fetal mouse gonads
Dev. Biol.
 , 
1999
, vol. 
215
 (pg. 
399
-
406
)
45
Ottolenghi
C.
Pelosi
E.
Tran
J.
Colombino
M.
Douglass
E.
Nedorezov
T.
Cao
A.
Forabosco
A.
Schlessinger
D.
Loss of Wnt4 and Foxl2 leads to female-to-male sex reversal extending to germ cells
Hum. Mol. Genet.
 , 
2007
, vol. 
16
 (pg. 
2795
-
2804
)
46
Stark
K.
Vainio
S.
Vassileva
G.
McMahon
A.P.
Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt-4
Nature
 , 
1994
, vol. 
372
 (pg. 
679
-
683
)
47
Moore
A.W.
Schedl
A.
McInnes
L.
Doyle
M.
Hecksher-Sorensen
J.
Hastie
N.D.
YAC transgenic analysis reveals Wilms’ tumour 1 gene activity in the proliferating coelomic epithelium, developing diaphragm and limb
Mech. Dev.
 , 
1998
, vol. 
79
 (pg. 
169
-
184
)
48
Morais da Silva
S.
Hacker
A.
Harley
V.
Goodfellow
P.
Swain
A.
Lovell-Badge
R.
Sox9 expression during gonadal development implies a conserved role for the gene in testis differentiation in mammals and birds
Nat. Genet.
 , 
1996
, vol. 
14
 (pg. 
62
-
68
)