Mammalian sex determination occurs during a short period in embryonic development, with SRY (Sex determining Region on the Y chromosome) playing a key role. In the 1940s, Jost et al. demonstrated that a testis is necessary for development of the male phenotype in mammals, as implied by the fact that castrated rabbits of either chromosomal sex developed as females (Jost et al. 1973). Karyotypes of patients with Turner Syndrome (45, X) and Klinefelter syndrome (47, XXY) demonstrated that sex is chromosomally controlled, and that the Y chromosome, in particular, determines male development regardless of X chromosome number (Vilain and McCabe 1998). Based on these observations, it was hypothesized that a dominant inducer of testis formation, the so-called testis determination factor (TDF), was located on the Y chromosome and encoded a gene or genes specifying the male phenotype; in the absence of this TDF, a female phenotype ensued (Vilain and McCabe 1998).
In 1990, the master gene on the Y chromosome, Sry, was isolated by positional cloning and identified as the TDF (Polanco and Koopman 2007). In the genome of true hermaphrodites and XX males, the small fragment of the Y chromosome that had translocated to the X chromosome was found to encode Sry, where XX mice transgenic for Sry developed as males with normal testes and testicular cords, implying that Sry was the only Y-encoded gene necessary for testis formation and the resulting male phenotype (Harley et al. 2003).
Expression of SRY During Development
In mammals, the gonad initially develops in a nonspecific manner, and is morphogenically identical in XX and XY mouse embryos until Sry expression, which occurs between 10.5 and 12.5 days (d) post coitum (pc) (Sekido and Lovell-Badge 2009). At 11.5 dpc, sex-specific differences become apparent, with the differentiation of the supporting lineage into Sertoli cells and alignment into testis cord and presumptive seminiferous tubules. In the absence of SRY, the ovary forms via a default pathway 2–3 days later in female embryos forming granulosa or theca cells. After 12.5 dpc, SRY is detectable only at very low levels in the mouse (Polanco and Koopman 2007).
In humans, the gonadal ridge forms around day 33 of gestation, with Sry detectable in XY embryos at day 41, peaking at day 44 when the testis cords are visible. By 52 days gestation, the germ cells are surrounded by Sertoli cells, which continue to express Sry at low levels (Harley et al. 2003). In contrast to in the mouse, the Sry transcript is expressed widely in the human embryo, including in the brain and continues to be transcribed even in human adult testis, consistent with roles additional to those in sex determination, such as spermatogenesis (Harley et al. 2003).
Little is known of how Sry expression is regulated, with the transcription factors WT1 (Wilm’s tumor suppressor gene 1) and SF1 (Steroidogenic factor 1) playing key roles (see Fig. 1) (Polanco and Koopman 2007). Mutations in either can result in sex reversal and gonadal dysgenesis, with SF1 able to bind and transactivate the human and porcine Sry promoter in vitro. Two isoforms of WT1 have been found to play key roles in testis development, −KTS and +KTS, which exclude or include the amino acid sequence lysine-threonine-serine in the zinc finger nucleic acid binding domain, respectively. The -KTS isoform regulates Sf1 expression, with evidence that it can also bind and transactivate the human and porcine Sry promoter. Intriguingly, the +KTS isoform shows higher affinity for RNA than DNA, and has been proposed to be involved in enhancing Sry translation and/or stabilizing the Sry mRNA (Polanco and Koopman 2007). Lhx9 (LIM-homeobox gene 9) and a polycomb group protein M33 are also involved in regulating Sry expression, with both being able to upregulate Sf1 expression (Piprek 2009). CITED1 [CBP(CREB-binding protein)/p300-interacting transactivator with ED-rich tail 2], a non-DNA binding transcriptional cofactor, may also act with SF1 and WT1 to increase Sry expression (Buaas et al. 2009).
The transcription factor GATA-4, along with its cofactor FOG2 can also upregulate Sry expression, with synergism between GATA-4 and the +KTS isoform of WT1 resulting in strong transactivation of the porcine and mouse Sry promoter (Piprek 2009). The transcription factor SP1 (specificity protein 1) may also activate Sry expression, with mutations affecting its binding to the Sry promoter associated with human XY sex reversal (Polanco and Koopman 2007). The insulin receptor, insulin-related receptor, and the insulin-like growth factors are also implicated in testis development, with sex reversal resulting from a triple null mutation in mice (Sekido and Lovell-Badge 2009). A transcription factor, EMX2 (empty spiracles homeobox 2), is also believed to play a role in testis development, although little is known of the mechanism (Piprek 2009).
The expression of Sry in the embryonic gonad appears to occur as a dynamic wave, originating from the center of the gonads and diffusing out toward the anterior and posterior poles between 10.5 and 12.5 dpc in mice (Polanco and Koopman 2007). SRY positive cells are detectable at 12 tail somites (ts; ∼11.0 dpc), by 18 ts (∼11.5 dpc) one third of cells express SOX9 [SRY-related HMG (high mobility group)-box 9], half of which are SRY positive, and by 25 ts (∼12.0 dpc) 50% of cells show SOX9 expression, with 90% of all cells being SRY negative (Oh and Lau 2006). Recent studies indicate that SRY has a critical time window of 6 h to trigger Sox9 upregulation and induce testis differentiation (Hiramatsu et al. 2009). Once Sox9 is expressed, Sry expression is switched off through a negative feedback mechanism either through the actions of SOX9, since Sry expression is higher in gonads with low levels of Sox9, or through Sry itself, where mice carrying weak Sry alleles have prolonged Sry expression, presumably as a result of delayed Sox9 expression (Sekido and Lovell-Badge 2009).
Genetic studies of the variants of Y chromosome present on the C57BL/6 J (B6) mouse strain, as well as transgenic mice, indicate that threshold levels of Sry are required to be expressed at critical times during male sex development, with reduced levels and delayed expression of Sry resulting in aberrant testis development (Sekido and Lovell-Badge 2009; Polanco and Koopman 2007). These results indicate that Sry must act during a critical time window to appropriately activate Sox9 and must be expressed before the ovarian pathway is engaged for testis development to proceed (Polanco and Koopman 2007).
Targets of SRY in the Cell
There is essentially no sequence conservation outside of the HMG domain of SRY between different mammalian and reptile species, signifying its rapid divergence through evolution (Harley et al. 2003). The SRY HMG domain shows >70% amino acid sequence similarity for the SRY sequences from human, mouse, rabbit, wallaby, marsupial mouse, and sheep (Harley et al. 2003). The nuclear magnetic resonance structure of the HMG domain of SRY in complex with DNA has been determined, showing three a-helices forming an L shape. Through binding the minor groove of DNA, SRY can induce a large conformational change involving DNA helix unwinding, minor groove expansion, and DNA bending. These events collectively represent a molecular switch, allowing interactions between proteins bound at distant sites on the DNA to modulate transcription (Harley et al. 2003).
The N-terminal domain of SRY has also been shown to play an important role in DNA binding, with phosphorylation, by cAMP-dependent protein kinase (PKA), of serine residues at the N-terminal enhancing binding in the case of hSRY to DNA. Intriguingly, impaired PKA phosphorylation due to the R30I mutation of the PKA recognition site (RRSSS33) reduces DNA binding activity, resulting in XY sex reversal (Assumpcao et al. 2002). Recent data also indicates the importance of the C-terminal domain of SRY in contributing to DNA binding in vitro (Sanchez-Moreno et al. 2009).
Timing of Sry expression is clearly crucial for testis development, as is the appropriate cellular environment provided by the Sertoli cells through provision of cofactors, posttranslational modification, subcellular localization, and access to target DNA (Oh and Lau 2006). Although DNA binding, through the HMG domain, appears to be critical for sex determination, several proteins have also been identified to interact with SRY (see Fig. 2). The SRY-Interacting Protein 1 (SIP-1/NHERF-2), for example, may perhaps be involved in the modulation of SRY transcriptional activity, interacting through its protein–protein interacting (PDZ) domain, with the Thr-Lys-Leu (KTL) motif in hSRY’s C-terminus, and the bridge domain in mSRY (Sekido and Lovell-Badge 2009).
KRAB-O (Kruppel-associated box only) has also been found to interact with SRY (see Fig. 2) through the bridge domain of mSRY and a domain adjacent to the hSRY HMG-box (aa 138–155) (Oh and Lau 2006). KRAB-O is believed to be involved in modulating SRY transcriptional activity, where it interacts directly with the KRAB-association protein 1 (KAP1) and indirectly with heterochromatin protein 1 (HP1), which acts as a transcriptional repressor (Oh and Lau 2006). Intriguingly, reduced expression of Sox9 is observed in KRAB-O knockout mouse models, although no effects are evident in terms of testis development (Oh and Lau 2006).
Wilm’s tumor gene (WT1; see also above) is also essential for testis development, with mutations such as R394W and D369N impairing the interaction of WT1 with SRY (see Fig. 2) and resulting in reduced transactivation, leading to Denys-Drash syndrome (DDS), which is characterized by Wilm’s tumor, pseudohermaphroditism, and neuropathy (Matsuzawa-Watanabe et al. 2003). Other proteins that interact with SRY include PARP1 (poly(ADP-ribose) polymerase 1), which upon binding SRY’s HMG domain is thought to interfere with SRY’s DNA binding activity (Li et al. 2006).
SRY and Sex Reversal
That, as a general rule, XY embryos become males and XX embryos become females does not hold true for sex-reversed individuals where the normal process of sexual differentiation has been altered. Since gonadal differentiation in most mammals cannot be induced by environmental variations or external manipulations, these must arise from genetic abnormalities. XX female-to-male sex reversal is a rare event, observed in 1:20,000 newborns, with 80% of cases due to Sry translocation. However, XY male-to-female sex reversal occurs at a frequency of 1:3,000 in newborns (Sarafoglou and Ostrer 2000), although only 15% of all cases can be attributed to mutations within Sry, consistent with the idea that a number of other key genes are involved in sex determination, such as Sox9, Sf1, and Wt1.
Almost all patients with SRY mutations show pure or complete gonadal dysgenesis (CGD), lacking testicular development with well-developed Mullerian structures and streak gonads in place of normal ovaries (Sarafoglou and Ostrer 2000). XY sex reversal patients develop as normal females with female internal and external genitalia, but are sterile, lack ovarian function, and present clinically with primary amenorrhea (Sarafoglou and Ostrer 2000). About 50% of XY female sex reversal cases show gonadoblastomas, additional to CGD, necessitating surgical removal of gonadal tissues (Mitchell and Harley 2002). The majority of SRY sex-reversing mutations occur de novo, thus affecting only one individual in the family (Mitchell and Harley 2002). However, mutations in the HMG domain have also been described in fathers of XY females. In these so-called familial cases, it is thought that genetic background compensates for the mutation in the father, but not in the affected individual (Mitchell and Harley 2002).
Nuclear Entry of SRY
Nuclear import via the CaM-NLS is mediated in an unconventional manner through direct interaction with the calcium-binding protein calmodulin (CaM), in a Ca2+-dependent and Ran-independent manner (see Fig. 4) (Kaur and Jans 2011); precisely how CaM mediates nuclear import is not known but it seems reasonable to suggest that CaM is an adaptor binding SRY to another molecule that then transports SRY through the nuclear pore. Intriguingly, changes in intracellular Ca2+ concentrations may also play a role in modulating nuclear accumulation of SRY through switching between the nuclear import mechanisms, with impaired nuclear import via the ß-NLS pathway observed in the presence of increasing Ca2+ concentrations (Kaur and Jans 2011). What is clear, however, is that both NLSs are essential for SRY’s role in sex determination, since mutations in the ß-NLS (R133W) and CaM-NLS (M64 T and R76P) that do not affect DNA binding/bending but result in reduced binding to Impß1 and CaM, respectively, result in sex reversal (Kaur et al. 2010).
Targets of SRY Action
SRY clearly can act as a transcriptional activator, with Sox9 being a strong candidate as a direct target of SRY. Sox9 is strongly upregulated soon after Sry expression, with Sry positive cells becoming Sox9 positive cells as well in mouse testis (Sekido and Lovell-Badge 2009). More recently, transgenic mice studies have demonstrated that SRY, acting synergistically with SF1, can induce the upregulation of Sox9 transcription through a testicular specific enhancer of Sox9 core (TESCO) situated 14 kb upstream of the Sox9’s start codon, with SF1 and SRY found to directly bind multiple elements within TES in vivo (Sekido and Lovell-Badge 2009). Consistent with the idea that Sox9 is a key target of SRY action, a number of studies indicate that SOX9 can replace SRY function (Sekido and Lovell-Badge 2009). Sex reversal resulting from mutations in Sox9 has also been described in XY campomelic dysplasia (CD) patients (Sekido and Lovell-Badge 2009).
Recent evidence suggests that SRY mediates testis development by repressing Wnt/ß-catenin signaling (see Fig. 1), which is stimulated by a small secreted factor, R-spondin 1 (RSPO1; see Fig. 1). RSPO1 synergizes with Wnt4 to activate ovarian development in XX females, where mutations in Rspo1 have been found to result in female-to-male sex reversal (Lau and Li 2009). SRY’s role appears to be through binding to ß-catenin via the HMG-box in the case of hSRY, and the HMG-box and Q-rich domains in the case of mSRY (see Fig. 2) (Sekido and Lovell-Badge 2009). SRY inhibits Wnt/ß-catenin signaling in the nucleus at the level of ß-catenin, possibly through inducing degradation of ß-catenin upon SRY binding, or interfering with ß-catenin’s ability to transactivate target genes, thus, acting to repress ovarian differentiation and thereby promote male sex development (Lau and Li 2009).
SRY plays a key role in mammalian sex determination with mutations in Sry resulting in sex reversal. Since its discovery several decades ago, slow progress has been made in determining the mechanisms of SRY function. Lack of experimental systems, especially in humans, has limited the study of sex determination to the mouse, with further limitations due to the brief period of expression of Sry in a small number of embryonic cells. It is clear however, that upregulation of SOX9 by SRY is a key event in mediating male sex differentiation, with emerging evidence implicating a role for SRY in repressing ovarian development through Wnt/ß-catenin. Significantly, the dual nuclear import mechanisms of SRY play a key role in male sex determination, with mutations therein resulting in impaired SRY targeting to the nucleus and sex reversal. Intriguingly, calcium plays a role in switching between the two import mechanisms of SRY to enable efficient SRY nuclear accumulation to be maintained under various physiological conditions. A key to understanding the development of the male sex will be in unraveling the regulation of Sry expression in early gonadal development, the dynamic movements of SRY within the cell during its brief period of expression, and the interplay of key sex determining proteins leading to repressed ovarian function and favored testis development.
- Assumpcao JG, Benedetti CE, Maciel-Guerra AT, Guerra Jr G, Baptista MT, Scolfaro MR, et al. Novel mutations affecting SRY DNA-binding activity: the HMG box N65H associated with 46, XY pure gonadal dysgenesis and the familial non-HMG box R30I associated with variable phenotypes. J Mol Med. 2002;80(12):782–90.PubMedCrossRefGoogle Scholar