In mice, the paracrine factor Wnt4 is expressed in the bipotential gonads. Wnt4 is maintained in XX gonads as they begin to form ovaries, but its expression becomes undetectable in XY gonads as they become testes. In XX mice that lack the Wnt4 gene, the ovary fails to form properly, and the cells transiently express testis-specific markers, including Sox9, testosterone-producing enzymes, and AMH (Vainio et al. 1999; Heikkilä et al. 2005). Thus, Wnt4 appears to be an important factor in ovary formation, although it is not the only determining factor.
Two other proteins, Rspo1 and β-catenin, interact with Wnt4 to promote its activity. Rspo1 is critical in ovary formation (FIGURE 1). It binds to and inactivates the intercellular proteins that specifically digest the Wnt receptor (Frizzled). Thus, Rspo1 promotes Wnt signaling, and XX humans born with RSPO1 gene mutations became phenotypic males (Parma et al. 2006; Harris et al. 2018. In the ovarian pathway, Rspo1 acts in synergy with Wnt4 to produce β-catenin, which is critical both in activating further ovarian development and in blocking synthesis of the testis-promoting transcription factor Sox9 (Maatouk et al. 2008; Jameson et al. 2012). In XY individuals with a duplication of the region on chromosome 1 that contains both the WNT4 and RSPO1 genes, the pathways that make β-catenin override the male pathway, resulting in a male-to-female sex reversal. Similarly, if XY mice are induced to overexpress β-catenin in their gonadal rudiments, they form ovaries rather than testes. Indeed, β-catenin appears to be a key “pro-ovarian/anti-testis” signaling molecule in all vertebrate groups, as it is seen in the female (but not the male) gonads of birds, mammals, and turtles, three groups having very different modes of sex determination (Maatouk et al. 2008; Cool and Capel 2009; Smith et al. 2009).
Certain transcription factors whose genes are activated by β-catenin are found exclusively in the ovaries. One possible target for β-catenin is the gene encoding TAFII105 (Freiman et al. 2002). This transcription factor subunit (which helps bind RNA polymerase II to promoters) is seen only in ovarian follicle cells. Female mice lacking this subunit have small ovaries with few, if any, mature follicles. One target for β-catenin is the gene encoding the transcription factor Foxl2, a protein that is strongly upregulated in ovaries; XX mice homozygous for mutant Foxl2 alleles develop male-like gonad structure and upregulate Sox9 gene expression and testosterone production. Both Foxl2 and β-catenin are critical for activation of the Follistatin gene (Ottolenghi et al. 2005; Kashimada et al. 2011; Pisarska et al. 2011). Follistatin, an inhibitor of TGF-β superfamily paracrine factors, is thought to be the protein responsible for organizing the epithelium into the granulosa cells of the ovary (Yao et al. 2004). XX mice lacking Follistatin in the developing gonad undergo a partial sex reversal, forming testis-like structures. Numerous other transcription factors are upregulated by the Wnt4/Rspo1 signal (Naillat et al. 2015), and we are just beginning to figure out how the components of the ovary-forming pathway are integrated.
Just as important as the construction of the ovaries is the maintenance of the ovarian structure. Gonadal organization is not stable throughout life, and without proper gene expression, female follicles can become male tubules and male tubules can become female follicles. In females, the maintainer of ovarian identity appears to be Foxl2 (Uhlenhaut et al. 2009), another target of β-catenin. When Foxl2 is deleted in adult-stage ovaries, the Sox9 gene becomes active and the ovary is transformed into a testis.