WNT4 (1p36.12) encodes a glycoprotein that plays multiple roles in ovarian differentiation and Müllerian duct formation (52) ( Table 1 ). It is modulated by RSPO1 and acts by decreasing the phosphorylation and degradation of β-catenin. Increased levels of β-catenin antagonize SOX9, leading to upregulation of DAX1, which in turn antagonizes SF1 (53) ( Figure 2 ). In mice with Wnt4 knockout, XX individuals exhibit virilization with the presence of Leydig-like cells in their gonad. While Wolffian ducts develop typically, Müllerian ducts are absent (54). In humans, heterozygous loss-of-function pathogenic variants in WNT4 have been found in virilized 46,XX women, who presented excess ovarian androgens and atypical Mullerian duct development (55–58). Additionally, a homozygous WNT4 pathogenic variant has been reported in a consanguineous family with a rare embryonic lethal syndrome known as SERKAL (SEx Reversion, Kidneys, Adrenal and Lung dysgenesis) syndrome ( Table 2 ). This syndrome is characterized by SRY-negative 46,XX testicular or ovotesticular DSD, as well as adrenal hypoplasia, renal agenesis, and severe defects in the lungs and cardiovascular structures (68).

RSPO1 gene (1p34.3) encodes a secreted agonist protein of the canonical Wnt/β-catenin signaling pathway, that is widely expressed during fetal development ( Table 1 ). RSPO1 plays a key role in gonad differentiation toward the ovary by synergizing the WNT4 to stabilize β-catenin in XX gonads (52, 59). In XX mice, the gonadal phenotype of the Rspo1 and the Wnt4 knockouts are strikingly similar: it ranges from small testes to ovotestes (26). RSPO1 is also expressed in fibroblasts and regulates the proliferation and differentiation of keratinocytes (69).

Homozygous deleterious RSOP1 variants have been identified in SRY-negative 46,XX DSD patients with atypical genitalia and palmoplantar hyperkeratosis and increased susceptibility to squamous cell carcinoma of the skin (60, 62, 69, 70) ( Table 2 ). These variants are typically located in the cysteine-rich furin domains of RSPO1, which are important for stabilizing cytosolic β-catenin. Dysregulation of β-catenin might result in the inhibition of Sox9 degradation and contribute to testis development ( Figure 2 ).

Histological examination of gonads of two affected individuals reveals testicular structures with Leydig cell hyperplasia and ovotestes with small residual ovarian tissue, respectively (63, 70). The absence of RSPO1 also affects the skin microenvironment and epidermal integrity, contributing to an elevated risk of squamous cell carcinoma in palmoplantar regions exposed to frictional stresses (71). Some patients may also present with congenital microphthalmia, cataracts, coloboma of the iris and choroid, onychodystrophy, laryngeal carcinoma, and hearing impairment (60–62, 70, 72).

The NR2F2 (Nuclear Receptor Subfamily 2 Group F Member 2) gene (15q26.2) encodes the chicken ovalbumin upstream promoter transcription factor 2 (COUP-TF2), which is an orphan nuclear receptor ( Table 1 ). COUP-TFII plays important roles during embryogenesis, particularly in cell fate determination, organogenesis, angiogenesis, and metabolism (64, 73). It also plays a role in cell regeneration or dedifferentiation. High expression of COUP-TFII is observed in the mesenchymal component of various organs, including the heart, brain, kidney, adrenal cortex, genital tubercle, otocyst, periocular mesenchyme, optic stalk, and olfactory placode, during development and organogenesis (74). Knockout and heterozygous mice lacking COUP-TFII exhibit multiple vascular abnormalities, especially in the heart and brain. These abnormalities can lead to premature death, with embryonic mortality observed in COUP-TFII knockout mice and death occurring within the first few days of life in heterozygous mice (74).

In the gonadal ridges, COUP-TF2 acts as a “pro-ovary” and “anti-testis” factor (75). Previous studies suggest that the Nr2f2 repression is necessary for fetal Leydig cell differentiation (76).

Ferreira et al. demonstrated that the human NR2F2 is highly upregulated during bipotential gonad development, being detected in early somatic cells that precede the steroidogenic cell emergence in the undifferentiated gonad. The authors propose that COUP-TFII regulates cell fate during gonad development by modulating the WNT signaling pathway, Runx2 (RUNX family transcription factor 2) activity, and the expression of Pparg (Peroxisome Proliferator Activated Receptor Gamma) and Sox9. Impairment of its function might disrupt the transcriptional plasticity of early supporting gonadal cells. This disruption during early gonad development may cause early supporting gonadal cells to commit to the testicular pathway (77).

Less than 40 individuals with heterozygous pathogenic variants in NR2F2 have been reported (78). Congenital heart defects are the most well-known phenotypes associated with its pathogenic variants, according to the expression pattern of COUP-TF2 (67, 73). However, the clinical features associated with NR2F2 variants are variable. These features include intrauterine growth restriction (IUGR), congenital heart disease (CHD), congenital diaphragmatic hernia (CDH), blepharophimosis ptosis-epicanthus inversus syndrome (BPES), developmental delays, hypotonia, feeding difficulties, failure to thrive, congenital and acquired microcephaly, dysmorphic facial features (such as up-slanted or short palpebral fissures, micrognathia or retrognathia, low-set or dysplastic ears, hypertelorism, and full cheeks), renal failure, hearing loss, strabismus, asplenia, and vascular malformations. Genital anomalies and DSD have also been described (78).

The molecular mechanisms leading to testis development in some 46,XX patients with COUP-TFII loss-of-function have yet to be defined (77, 79). NR2F2 pathogenic variants/deletion were found to be associated with five patients who had a syndromic form of SRY-negative 46,XX T/OT DSD ( Table 2 ) (65, 66, 78, 79).

These patients presented with atypical genitalia (4/5), congenital diaphragmatic hernia (CDH) (3/5), blepharophimosis ptosis-epicanthus inversus syndrome (BPES) (3/5), and congenital heart disease (CHD) (2/5). Three of the patients had frameshift variants affecting the N-terminal region of the protein, specifically, p.Gly35Argfs*75 and p.Pro33Alafs*77 and the fourth patient had a de novo nonsense variant, p.Trp8* (78, 79).

( Table 2 ). In the fifth patient, a CGH array assay identified a 3-Mb 15q26.2 [arr(GRCh37) 95127653_98146649] x1 deletion that encompassed the entire NR2F2 gene (65, 66).

Genotype-phenotype correlations cannot be identified, as individuals carrying identical NR2F2 variants may present with variable phenotypic manifestations. In the case of 46,XX patients, a single-copy genomic deletion that encompasses the entire NR2F2 gene may result in testicular tissue and atypical external genitalia in some cases, but in others, there may be no evidence of genital anomalies or DSD, despite the presence of other syndromic features (65, 66, 78, 80). These findings suggest that the phenotypic expression of NR2F2-related differences may be likely influenced by additional modifiers.

SOX9 (17q24.3) is a transcription factor that plays a significant role in various tissues, including chondrocytes and testes (84) ( Table 1 ). Studies investigating the relationship between phenotype and genotype in humans and mice have demonstrated that SOX9 expression is a critical step in testis development, occurring downstream of SRY. SOX9 is responsible for the specification of Sertoli cells, which in turn initiates testicular differentiation and triggers the production of AMH (85, 86).

Overexpression of SOX9, often caused by gene duplications or copy number variations in the upstream promoter region, has been linked to testis determination in the absence of SRY (84, 87, 88) ( Figure 2 ). In many cases of 46,XX T/OT DSD, SOX9 duplications have been identified as the most commonly observed genetic cause, second only to SRY translocation ( Table 3 ).

Many of these duplications involve a noncoding region spanning at least 24 kb, known as the RevSex region, located approximately 0.5-0.6 Mb upstream of the SOX9 gene (89–100) ( Table 3 ). This region is predicted to contain a human testis-specific enhancer, and the duplication of this enhancer drives the atypical expression of SOX9, leading to the activation of testicular differentiation (87, 92).

Ushijima et al. escribed an SRY-negative 46,XX OT DSD patient with a novel SOX9 missense variant (p.Glu50Lys) with promoter-specific gain-of-function (GoF) activity in in vitro studies. The authors demonstrated that E50K-SOX9 had (GoF) activity in the mTESCO-luciferase reporter, suggesting that it was due to change(s) in its bioactivity. GoF activity was observed in mTESCO-luc but not in mAmh-luc, thereby indicating that the acquisition of GoF activity was promoter-specific. To associate the promotor SOX9 variant with atypical expression of SOX9, and the beginning of testicular differentiation in the 46,XX OT DSD patient, mice carrying the Sox9 p.E50K were also generated and characterized. These mice, nevertheless, did not develop ovotestis (101). Such discordance of expressivity/phenotype among humans and mice are not limited to sox9/SOX9 (102) but are also described in other genes associated with DSD, including Nr5a1/NR5A1 (103) and Wt1/WT1 (104). The molecular mechanism of the promoter -specific GoF activity of E50K-SOX9 remains to be elucidated.

SOX3 (Xq27.1) is another member of the SOX gene family that is involved in gonadal development. It encodes a protein that is highly SRY-like, with an amino acid sequence similarity of 67% for the protein and 90% for the HMG DNA-binding domain (105) ( Table 1 ). Studies in transgenic mice have shown that increased ectopic expression of Sox3 in undifferentiated gonads can lead to sex reversal in XX mice, with complete virilization of external genitalia observed in 77% of animals (106). These findings suggest that Sox3 hyperexpression acts as a counterpart of Sry, leading to increased expression of Sox9. Together with Nr5a1, Sox3 binds to the enhancer region of Sox9 (106).

Like the findings in mice, when increased expression of SOX3 is observed in humans, this gene acts in conjunction with NR5A1 to promote overexpression of SOX9. This phenomenon directs the gonads toward male determination (105) ( Figure 2 ).

The duplication of the SOX3 in a patient with SRY-negative 46,XX OT DSD was initially identified by Sutton et al. (105). Several other 46,XX patients with testicular development (T and OT) and duplications of the SOX3 or in the regions located upstream of this gene have been reported (94, 107–114), supporting the importance of SOX3 in testis development ( Table 3 ).

A heterozygous deletion downstream of SOX3 was also reported in an SRY-negative 46,XX infertility male. The authors of the study speculated that this deletion may play a role in the regulation of the SOX3, potentially resulting in increased expression of SOX3 (115).

SOX10 (22q13.1) is another gene closely correlated with SOX9 in humans ( Table 1 ). Initially expressed in neural crest cells during the embryonic period, it plays a critical role in their development. SOX10 is also expressed in fetal gonads (116). In mice, the expression of Sox10 specifically in Sertoli cells strongly indicates its involvement in the testicular differentiation process and reinforces its role in the male pathway (117).

Studies in transgenic animal models have demonstrated that Sox10 overexpression causes sex reversal in XX mice (117). These studies demonstrated that the expression level of Sox10 is crucial in determining the gonadal phenotype. Complete testicular differentiation in all mice was observed in the lineage with higher levels of Sox10 expression, while the lineage expressing lower levels of the transgene showed only 30% of mice with complete sex reversal in the postnatal period. Interestingly, all fetuses from the second group (lower expression levels) were able to initiate Sertoli cell differentiation (presence of cells expressing Sox9 in XX transgenic gonads). In these gonads, cells committed to the female pathway, identified by the expression of Foxl2, were interspersed with Sox9-positive cells (117). This same pattern has been described in ovotestis in humans, as well as in mouse models of ovotestis development (118).

Similarly, gonadal, and reproductive system alterations have been reported in cases of partial duplication of chromosome 22q in 46,XX humans, a chromosomal region that contains SOX10 ( Table 4 ). Rare patients diagnosed with 46,XX T/OT DSD, both syndromic and non-syndromic, have also been described in the literature with chromosome 22 aneuploidies (116, 119–121) ( Table 4 ).

The DMRT1 gene (9p24.3) encodes a transcription factor that plays a crucial role in sex determination and gonadal development in various species ( Table 1 ). It possesses a zinc-finger-like DNA binding domain known as the DM (doublesex/MAB-3) domain. This domain allows DMRT1 to bind to specific DNA sequences and regulate the expression of genes involved in sex differentiation (122). DMRT1 expression has been observed in the undifferentiated human XY gonadal primordium. During the early fetal period (gestational weeks 8-20), it is primarily expressed in Sertoli cells, which play a crucial role in testicular development. In the second gestational trimester, childhood, and post-puberty, DMRT1 expression becomes more abundant in spermatogonia (123) This dynamic pattern of expression suggests that DMRT1 plays a significant role in both the early and later stages of male gonadal development.

DMRT1 expression has indeed been detected in oogonia and oocytes during the early stages of ovarian development, up until gestational week 20. However, it is important to note that after the onset of meiotic germ cell division, DMRT1 expression becomes absent in these cells (123).

In contrast to DMRT1 homologs in other vertebrates, mammalian DMRT1 seems to not be involved in the initial sex determination but is instead required for maintaining male gonadal fate (124, 125). Studies in mice have demonstrated that the loss of expression of certain key genes in postnatal life can lead to the reprogramming of Sertoli cells into granulosa cells and vice versa. This suggests that there is a level of plasticity in gonadal fate even after the typical formation of a testis or ovary (31, 124).

Ectopic expression of DMRT1 has been shown to reprogram differentiated female granulosa cells into male Sertoli-like cells. DMRT1 functions in collaboration with other key male sex regulators like SOX9 to maintain and reprogram sexual cell fate. It acts as a singular transcription factor, by regulating gene expression and chromatin accessibility (126).

Bertini et al. reported a three-year-old boy who presented with a typical male phenotype and an SRY-negative 46,XX karyotype (127). The genetic study conducted showed a heterozygous de novo in tandem duplication of 50,221 bp on chromosome 9p. This duplication encompassed exons 2 and 3 of the DMRT1 and was detected using MPLA, CGH-array analysis, and Sanger sequencing. The breakpoints of the duplication were in the intronic regions, and it did not disrupt the coding frame of DMRT1. To investigate other potential genetic factors contributing to the phenotype, a custom NGS panel and whole genome sequencing were performed, but no additional pathogenic or uncertain variants were found in genes known to be involved in pro-testis/anti-ovary gene cascades.

The identified duplication might have allowed DMRT1 to escape the usual transcriptional repression that occurs in 46,XX fetal gonads, leading to the activation of the testicular determination cascade. Notably, no previous cases of SRY-negative 46,XX DSD associated with alterations in DMRT1 have been reported thus far.

The FGF9 gene (13q12.1) is a signaling peptide involved in the development of various organs, including limbs, lungs, the adenohypophysis, and the gonadal ridges ( Table 1 ). FGFs are typically considered paracrine factors and play important roles in tissue patterning and organogenesis during embryogenesis. The FGF9 subfamily, which signals from epithelium to mesenchyme, stimulates mesenchymal proliferation. In Fgf9 knockout XY mice, gonadal development is severely impaired during embryonic and fetal life, leading to reproductive phenotypes ranging from different range of undervirilization to complete feminization of external genitalia (128).

In a study by Chiang et al., an SRY-negative 46,XX male with hypospadias and azoospermia was identified (129). Array-CGH analysis revealed duplicated regions on chromosomes 13q12.11 (21.143874–21.174184 Mb) and 13q31.1 (79.807500–79.813700 Mb). These duplicated regions encompassed the entire FGF9 and SPRY2 genes, respectively. The genomic gain of FGF9 was hypothesized to result in FGF9 overexpression, which could explain testicular development instead of ovarian development. Additionally, SPRY2 was previously related to a potential role in male sex organogenesis by controlling FGF9 gene-induced mesonephric cell migration to the developing testis (130). The higher amount of FGF9 would interfere with the expression of WNT4 in the embryo, thereby impeding ovarian development in SRY-negative 46,XX males.