Sexual differentiation of gonads as a function of temperature in the turtle Emys orbicularis: endocrine function, intersexuality and growth

Ovotestes suggest cryptic genetic influence in a reptile model for temperature-dependent sex determination

Sex determination and differentiation in reptiles is complex. Temperature-dependent sex determination (TSD), genetic sex determination (GSD) and the interaction of both environmental and genetic cues (sex reversal) can drive the development of sexual phenotypes. The jacky dragon (Amphibolurus muricatus) is an attractive model species for the study of gene-environment interactions because it displays a form of Type II TSD, where female-biased sex ratios are observed at extreme incubation temperatures and approximately 50 : 50 sex ratios occur at intermediate temperatures. This response to temperature has been proposed to occur due to underlying sex determining loci, the influence of which is overridden at extreme temperatures. Thus, sex reversal at extreme temperatures is predicted to produce the female-biased sex ratios observed in A. muricatus. The occurrence of ovotestes during development is a cellular marker of temperature sex reversal in a closely related species Pogona vitticeps. Here, we present the first developmental data for A. muricatus, and show that ovotestes occur at frequencies consistent with a mode of sex determination that is intermediate between GSD and TSD. This is the first evidence suggestive of underlying unidentified sex determining loci in a species that has long been used as a model for TSD.

Chronology of Gonadal Development in the Malayan Snail-eating Turtle Malayemys macrocephala

The snail-eating turtle, Malayemys macrocephala, is a common freshwater turtle that can be used as an animal model for developmental biology. However, a thorough investigation of its development is needed before this species can be used as a model. Thus, this study aimed to examine the gonadal development of M. macrocephala. Turtle eggs were collected from rice fields in Phra Nakhon Si Ayutthaya Province, Thailand, and transported to the laboratory. Eggs were incubated in microprocessor-controlled incubators and randomly dissected on a weekly basis to reveal the developing embryos, then their developmental stage was identified according to Yntema (1968). Primordial germ cells and gonad structure were processed through the paraffin method. Moreover, the dynamics of germ cell proliferation and apoptosis were examined by immunohistochemical detection of proliferating cell nuclear antigen (PCNA) and the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL), respectively. Examination of the gonad revealed four main stages of gonadal development: (i) germ cell migration, (ii) genital ridge appearance, (iii) testicular formation, and (iv) ovarian formation. In the male turtle (incubated at 26°C), gonad developed into the testis with medullary sex cords starting at Yntema stage 17. In the female turtle (incubated at 32°C), these sex cords then degenerated, followed by cortical development into an ovarian structure starting at Yntema stage 19. Subsequently, testicular and ovarian development occurred independently, and distinct sex organs were apparent at Yntema stage 25. In addition, the presumptive testis showed germ cell proliferation in the medulla at Yntema stages 17, 19, and 25 and germ cell apoptosis in the cortex at Yntema stages 19 and 25. The presumptive ovary showed germ cell proliferation in the cortex at Yntema stages 19 and 25, and germ cell apoptosis in the medulla at Yntema stages 19 and 25.

Recent advances on the estimation of the thermal reaction norm for sex ratios

Temperature-dependent sex determination, or TSD, is a widespread phenomenon in reptiles. The shape of the relationship between constant incubation temperature and sex ratio defines the TSD pattern. The TSD pattern is considered a life-history parameter important for conservation because the wider the range of temperatures producing both sexes, the more resilient the species is to climate change impacts. We review the different published equations and methodologies that have been used to model TSD patterns. We describe a new flexible model that allows for an asymmetrical pattern around the pivotal temperature, which is the constant temperature producing both sexes in equal proportions. We show that Metropolis-Hastings with Markov chain produced by a Monte Carlo process has many advantages compared to maximum likelihood and is preferred. Finally, we apply the models to results from incubation experiments using eggs from the marine turtle Lepidochelys olivacea originating in Northeast Indian, East Pacific, and West Atlantic Regional Management Units (RMUs) and find large differences in pivotal temperatures but not in transitional ranges of temperatures.

Mechanisms related to sexual determination by temperature in reptiles

A number of strategies have emerged that appear to relate to the evolution of mechanisms for sexual determination in vertebrates, among which are genetic sex determination caused by sex chromosomes and environmental sex determination, where environmental factors influence the phenotype of the sex of an individual. Within the reptile group, some orders such as: Chelonia, Crocodylia, Squamata and Rhynchocephalia, manifest one of the most intriguing and exciting environmental sexual determination mechanisms that exists, comprising temperature-dependent sex determination (TSD), where the temperature of incubation that the embryo experiences during its development is fundamental to establishing the sex of the individual. This makes them an excellent model for the study of sexual determination at the molecular, cellular and physiological level, as well as in terms of their implications at an evolutionary and ecological level. There are different hypotheses concerning how this process is triggered and this review aims to describe any new contributions to particular TSD hypotheses, analyzing them from the "eco-evo-devo" perspective.

Insights into Emydid Turtle Cytogenetics: The European Pond Turtle as a Model Species

Our knowledge of Testudines evolution is limited by the lack of modern cytogenetic data. Compared to other reptiles, there is little information even on chromosome banding, let alone molecular cytogenetic data. Here, we provide detailed information on the karyotype of the European pond turtle Emys orbicularis, a model Emydidae, employing both chromosome banding and molecular cytogenetics. We provide a high-resolution G-banded karyotype and a map of rDNA genes and telomeric sequences using fluorescence in situ hybridization. We test hypotheses of sex-determining mechanisms in Emys by comparative genomic hybridization to determine if Emys has a cryptic sex-specific region. Our results provide valuable data to guide future efforts on genome sequencing and anchoring in Emydidae and for understanding karyotype evolution in Testudines.

Developmental asynchrony and antagonism of sex determination pathways in a lizard with temperature-induced sex reversal

Vertebrate sex differentiation follows a conserved suite of developmental events: the bipotential gonads differentiate and shortly thereafter sex specific traits become dimorphic. However, this may not apply to squamates, a diverse vertebrate lineage comprising of many species with thermosensitive sexual development. Of the three species with data on the relative timing of gonad differentiation and genital dimorphism, the females of two (Niveoscincus ocellatus and Barisia imbricata) exhibit a phase of temporary pseudohermaphroditism or TPH (gonads have differentiated well before genital dimorphism). We report a third example of TPH in Pogona vitticeps, an agamid with temperature-induced male to female sex reversal. These findings suggest that for female squamates, genital and gonad development may not be closely synchronised, so that TPH may be common. We further observed a high frequency of ovotestes, a usually rare gonadal phenotype characterised by a mix of male and female structures, exclusively associated with temperature-induced sex reversal. We propose that ovotestes are evidence of a period of antagonism between male and female sex-determining pathways during sex reversal. Female sexual development in squamates is considerably more complex than has been appreciated, providing numerous avenues for future exploration of the genetic and hormonal cues that govern sexual development.

Sex determination mode does not affect body or genital development of the central bearded dragon (Pogona vitticeps)

Background

The development of male- or female-specific phenotypes in squamates is typically controlled by either temperature-dependent sex determination (TSD) or chromosome-based genetic sex determination (GSD). However, while sex determination is a major switch in individual phenotypic development, it is unknownhow evolutionary transitions between GSD and TSD might impact on the evolution of squamate phenotypes, particularly the fast-evolving and diverse genitalia. Here, we take the unique opportunity of studying the impact of both sex determination mechanisms on the embryological development of the central bearded dragon (Pogona vitticeps). This is possible because of the transitional sex determination system of this species, in which genetically male individuals reverse sex at high incubation temperatures. This can trigger the evolutionary transition of GSD to TSD in a single generation, making P. vitticeps an ideal model organism for comparing the effects of both sex determination processes in the same species.

Results

We conducted four incubation experiments on 265 P. vitticeps eggs, covering two temperature regimes ("normal" at 28 °C and "sex reversing" at 36 °C) and the two maternal sexual genotypes (concordant ZW females or sex-reversed ZZ females). From this, we provide the first detailed staging system for the species, with a focus on genital and limb development. This was augmented by a new sex chromosome identification methodology for P. vitticeps that is non-destructive to the embryo. We found a strong correlation between embryo age and embryo stage. Aside from faster growth in 36 °C treatments, body and external genital development was entirely unperturbed by temperature, sex reversal or maternal sexual genotype. Unexpectedly, all females developed hemipenes (the genital phenotype of adult male P. vitticeps), which regress close to hatching.

Conclusions

The tight correlation between embryo age and embryo stage allows the precise targeting of specific developmental periods in the emerging field of molecular research on P. vitticeps. The stability of genital development in all treatments suggests that the two sex-determining mechanisms have little impact on genital evolution, despite their known role in triggering genital development. Hemipenis retention in developing female P. vitticeps, together with frequent occurrences of hemipenis-like structures during development in other squamate species, raises the possibility of a bias towards hemipenis formation in the ancestral developmental programme for squamate genitalia.

Developmental exposure to bisphenol A (BPA) alters sexual differentiation in painted turtles (Chrysemys picta)

Environmental chemicals can disrupt endocrine signaling and adversely impact sexual differentiation in wildlife. Bisphenol A (BPA) is an estrogenic chemical commonly found in a variety of habitats. In this study, we used painted turtles (Chrysemys picta), which have temperature-dependent sex determination (TSD), as an animal model for ontogenetic endocrine disruption by BPA. We hypothesized that BPA would override TSD and disrupt sexual development. We incubated farm-raised turtle eggs at the male-producing temperature (26°C), randomly assigned individuals to treatment groups: control, vehicle control, 17β-estradiol (E2, 20ng/g-egg) or 0.01, 1.0, 100μgBPA/g-egg and harvested tissues at hatch. Typical female gonads were present in 89% of the E2-treated "males", but in none of the control males (n=35). Gonads of BPA-exposed turtles had varying amounts of ovarian-like cortical (OLC) tissue and disorganized testicular tubules in the medulla. Although the percentage of males with OLCs increased with BPA dose (BPA-low=30%, BPA-medium=33%, BPA-high=39%), this difference was not significant (p=0.85). In all three BPA treatments, SOX9 patterns revealed disorganized medullary testicular tubules and β-catenin expression in a thickened cortex. Liver vitellogenin, a female-specific liver protein commonly used as an exposure biomarker, was not induced by any of the treatments. Notably, these results suggest that developmental exposure to BPA disrupts sexual differentiation in painted turtles. Further examination is necessary to determine the underlying mechanisms of sex reversal in reptiles and how these translate to EDC exposure in wild populations.

Conserved action of β-catenin during female fate determination in the red-eared slider turtle

In reptiles such as the red-eared slider turtle Trachemys scripta, development of an ovary from the bipotential gonad requires a coordinated expansion of the cortical domain and regression of the medulla. While estrogen, which is necessary and sufficient for ovarian development in non-mammalian vertebrates, is thought to feminize both compartments, it remains unclear whether there is a signaling relationship between the two domains that coordinates their fates. We show that aromatase, the estrogen-synthesizing enzyme, is localized to the medulla of the differentiating turtle ovary and that differentiation of the medulla precedes and is independent of cortical expansion. Coordinated feminization of the two domains may therefore rely on an estrogenic signal from the differentiating medulla. In eutherian mammals, where estrogen is dispensable for early ovary development, the canonical Wnt signaling pathway is critical for female fate determination. Whether this function is conserved among vertebrates and how it is potentially integrated with estrogen signaling are uncertain. Using a novel in vitro turtle gonad culture system, we demonstrate that ectopic activation of the canonical Wnt signaling pathway in presumptive male gonads induced a partial sex-reversal of the medulla, but inhibition of the pathway was not sufficient to sex-reverse differentiating ovaries. These patterns are similar to those previously observed in mice. Wnt signaling appears to function downstream of estrogen, as ectopic activation of the pathway rescued female development when estrogen synthesis was inhibited. Our findings therefore suggest that the ovary-promoting effects of the Wnt signaling pathway may be functionally conserved between mammals and reptiles.

The first case of intersexuality in an African dwarf crocodile (Osteolaemus tetraspis)

To the authors knowledge this is the first case of intersexuality in an African dwarf crocodile (Osteolaemus tetraspis). An adult African dwarf crocodile with a male-typical phenotype lived at Zoo Duisburg in Germany for 10 years. It died in October 2012 despite intensive treatment as a result of terminal septicemia. After a detailed pathological examination the gonads were histologically confirmed as ovotestes. Half of the 22 extant species of crocodilians have been examined for occurrence of temperature dependent sex determination (TSD). In TSD reptiles, masculinizing temperatures yield 100% or a majority of males, whereas feminizing temperatures yield 100% or a majority of females. In the transition range of temperature (TRT), a mix of males, females and sometimes intersexes are obtained. However, the molecular mechanisms behind TSD and an explanation for the occurrence of intersexuality remain elusive.

From the origin of sex-determining factors to the evolution of sex-determining systems

Sex determination is typically classified as either genotypic or environmental. However, this dichotomy obscures the developmental origin and evolutionary modification of determinants of sex, and therefore hinders an understanding of the processes that generates diversity in sex-determining systems. Recent research on reptiles and fish emphasizes that sex determination is a multifactorial regulatory process that is best understood as a threshold dichotomy rather than as the result of genetically inherited triggers of development. Here we critically assess the relationship between the developmental origin of sex-determining factors and evolutionary transitions in sex-determining systems. Our perspective emphasizes the importance of both genetic and nongenetic causes in evolution of sex determination and may help to generate predictions with respect to the evolutionary patterns of sex-determining systems and the underlying diversity of developmental and genetic regulatory networks.

Ontogeny of the oestrogen receptors ESR1 and ESR2 during gonadal development in the tammar wallaby, Macropus eugenii

Oestrogen has wide ranging effects in development mediated mainly via the two oestrogen receptors, α (ESR1, also known as ERα) and β (ESR2, also known as ERβ). Oestrogen is the key factor that directs the indifferent gonad to become an ovary in many non-mammalian vertebrates. Oestrogen is not required for early ovarian differentiation in mammals but can disrupt normal testicular development in eutherians. Surprisingly, exogenous oestrogen can cause sex reversal of an XY gonad in two marsupials, the North American opossum and the tammar wallaby. To understand the mechanism by which oestrogen induces sex reversal, we characterised the genes for ESR1 and ESR2 and examined their expression during gonadal differentiation in the tammar wallaby, Macropus eugenii. Both receptors were expressed in the somatic cells and germ cells of the indifferent gonad in both XX and XY foetuses throughout all stages of development, and persisted in these cells into adulthood. ERs were also present in many other tissues including kidney, pituitary and mammary gland. ER mRNA was not significantly altered by exogenous oestrogen in cultured XY gonads but the receptors translocated to the nucleus in its presence. These findings confirm that there is conserved expression of the ERs in the indifferent gonad despite the lack of available ligand during early gonadal development. The receptors can respond to exogenous estrogen at this early stage and are capable of transducing signals in the early mammalian gonad. However, the selective forces that maintained conserved ER expression in this tissue remain unknown.

Estrogen represses SOX9 during sex determination in the red-eared slider turtle Trachemys scripta

Production of male offspring in viviparous eutherian mammals requires a sex-determining mechanism resistant to maternal hormones. This constraint is relaxed in egg-laying species, which are sensitive to hormones during sex determination and often use an increase in aromatase, the estrogen-synthesizing enzyme, as a key feminizing signal. In the turtle Trachemys scripta, sex is normally determined by temperature, but estrogen treatment overrides this cue and leads exclusively to female development. We assessed whether the expression of SOX9, a central male sex-determining gene in mammals, or three other conserved transcription factors (WT1, GATA4, and LHX9) was regulated by estrogen signaling in the turtle. As in mice, all somatic cell types in the immature turtle gonad initially expressed WT1 and GATA4, whereas SOX9 was restricted to the Sertoli precursors and LHX9 to the coelomic epithelium and interstitium. After the bipotential period, SOX9 was abruptly down-regulated at the female temperature. Strikingly, embryos treated with beta-estradiol at the male temperature lost SOX9 expression more than two stages earlier than controls, though WT1, GATA4, and LHX9 were unaffected. Conversely, inhibition of estrogen synthesis and signaling prevented or delayed SOX9 down-regulation at the female temperature. These results suggest that endogenous estrogen feminizes the medulla of the bipotential turtle gonad by inhibiting SOX9 expression. This mechanism may be involved in the male-to-female sex reversal in wild populations exposed to environmental estrogens, and is consistent with results showing that the estrogen receptor represses Sox9 to block transdifferentiation of granulosa cells into Sertoli-like cells in the adult mouse ovary.

Lymphoid aggregates in gonads of embryos, hatchlings, and young of turtles with temperature-dependent sex determination

Cellular infiltrations forming lymphoid-like aggregates were previously observed in gonads of two turtle species exhibiting temperature-dependent sex determination (TSD): at hatching in Chelydra serpentina; at and after hatching in Emys orbicularis. We show here that such aggregates are also present in gonads of Testudo graeca by the end of embryonic development, suggesting that their occurrence is general in turtles. Since in C. serpentina, infiltrations were observed mainly in testes exhibiting remnants of the germinal epithelium, it was assumed that their occurrence was an expression of maleness leading to rejection of this epithelium. The generality of this hypothesis was tested in E. orbicularis by looking for lymphoid-like aggregates in three types of gonads (testes, ovotestes, and ovaries) and for the stages at which they occur. Gonads were from embryos, hatchlings, and young incubated at various temperatures. Ovotestes obtained by treatment with an aromatase inhibitor of eggs incubated at female-producing temperature were also examined. In these gonads, the differentiation of Sertoli cells in testicular cords/tubes was ascertained by expression of SOX9. Moreover, the cell composition of aggregates was determined on electron micrographs. Aggregates appear in ovaries and ovotestes by the end of embryonic development and are present in the majority of these gonads at hatching, and at least up to one year after hatching. They are composed mainly of lymphocytes and fibroblasts. Aggregates are not present in typical testes. Since they occur in most ovaries, they cannot be seen as an expression of maleness. Rather, lymphocytic infiltration and formation of lymphoid aggregates in turtle gonads can be seen as components of the immune system, and can be under the control of gonadal endogenous sex steroids.

Colocalization of WT1 and cell proliferation reveals conserved mechanisms in temperature-dependent sex determination

During vertebrate development the gonad has two possible fates, the testis or the ovary. The choice between these fates is made by a variety of sex-determining mechanisms, from the sex-determining gene on the Y chromosome (Sry) in mammals, to nongenetic temperature-dependent systems in many reptiles. Despite the differences in the mechanisms at the top of the sex-determining cascade, the resulting morphology and many genes involved in early testis and ovarian development are common to most vertebrates, leading to the hypothesis that the underlying processes of sex determination are conserved. In this study, we examined the early steps of gonad development in the red-eared slider turtle (Trachemys scripta), a species that uses the temperature of egg incubation to determine sex. A dramatic increase in cell proliferation was observed in the male gonad during the earliest stages of sex determination. Using the localization of Wilms' Tumor suppressor 1 (WT1), we determined that this proliferation increase occurred in a population that contained pre-Sertoli cells. The proliferation of pre-Sertoli cells has been documented during sex determination in both mice and alligators, suggesting that proliferation of this cell type has an important role in vertebrate testis organogenesis and the determination of male fate.

Temperature influences the ontogenetic expression of aromatase and oestrogen receptor mRNA in the developing tilapia (Oreochromis mossambicus) brain

Water temperature has a differential influence on the development of central neurotransmitter systems according to the developmental period in tilapia (Oreochromis mossambicus). Aromatase and oestrogen receptors (ERs) represent important components of the mechanism of brain differentiation. Gene expression of aromatase and ERs is modulated by neurotransmitters in the developing brain. In the present study, the quantitative reverse transcription-polymerase chain reaction method was used to investigate the effects of temperature on the ontogenetic expression of aromatase and ERs in the developing tilapia brain. Before day 10 posthatching, exposure to a higher temperature (32 degrees C) resulted in a significant increase in the expression of brain aromatase; conversely, a lower temperature (20 degrees C) resulted in a decrease. ERalpha expression was depressed in accordance with the decrease of temperature, but ERbeta was unaffected by temperature. Between days 10 and 20, neither brain aromatase nor ERalpha expression was altered by temperature, whereas ERbeta expression was significantly enhanced by exposure to 32 degrees C. Between days 20 and 30, brain aromatase significantly increased at the higher temperature and decreased at 20 degrees C, but neither ERalpha nor ERbeta was affected by temperature. The expression of both brain aromatase and ERs, differentially regulated according to the temperature and to the developmental period, could be related to brain-sex differentiation.

Sex reversal and aromatase in the European pond turtle: treatment with letrozole after the thermosensitive period for sex determination

In the European pond turtle (Emys orbicularis), gonadal sex differentiation is temperature-dependent. The temperature sensitive period (TSP) of gonadogenesis lies between stages 16 and 22 of embryonic development. Previous studies have shown that embryos incubated at 30 degrees C, a temperature yielding 100% phenotypic females, can be sex reversed by treatments with an aromatase inhibitor administered during TSP or even somewhat after TSP (as of stage 22+). The goal of the present study was to determine whether the ovary still retains male potential at later stages of embryonic development and whether the induced male characters persist after hatching. For this purpose, eggs of E. orbicularis were treated with letrozole, a nonsteroidal aromatase inhibitor, at or as of stages 23, 24 or 25, then gonadal aromatase activity in each individual and the related gonadal structure were studied at hatching (stage 26) and for one year after hatching. Two kinds of treatments were carried out: 1) repeated applications of 10 microg of letrozole in ethanolic solution onto the eggshell; and 2) a single injection of 10 microg of letrozole in olive oil. Similar results were obtained with either application or injection of the aromatase inhibitor. In treatments as of or at stage 23, individuals with gonadal aromatase activity lower than 20 fmoles/hour/gonad had ovotestes, i.e., 22% of the treated individuals. At hatching, the inner part of these ovotestes contained testicular cords and also mixed lacunae presenting various degrees of transdifferentiation of the epithelium into a Sertolian epithelium. The cortex was maintained, although some germ cells degenerated within it. These processes continued after hatching. However, at 12 months, gonads were still ovotestes displaying some follicles with a growing oocyte in the remaining parts of the cortex. In treatments as of or at stages 24 or 25, only a few individuals were masculinized. One had ovotestes; in others, the cortex was absent in some parts and when it was present oocytes were degenerating. These results show that in the European pond turtle, differentiation of ovotestes from ovaries can be induced by treatment with an aromatase inhibitor starting at late stages of embryonic development (between the end of TSP and hatching), although such differentiation is less frequent as embryonic development proceeds. Sex reversal persists for at least one year after hatching. J. Exp. Zool. 290:490-497, 2001.

Sexual differentiation in the spiny softshell turtle (Apalone spinifera), a species with genetic sex determination

It is hypothesized on the basis of sex determination theory that species exhibiting genetic sex determination (GSD) may undergo sexual differentiation earlier in development than species with environmental sex determination (ESD). Most turtle species exhibit a form of ESD known as temperature-dependent sex determination (TSD), and in such species the chronology of sex differentiation is well studied. Apalone spinifera is a species of softshell turtle (Trionychidae) that exhibits GSD. We studied sexual differentiation in this species in order to facilitate comparison to TSD species. Eggs were incubated at two different temperatures and embryos were harvested at various stages of mid to late development. Gonad length was measured with image analysis software, then prepared histologically. Indifferent gonads have differentiated in stage 19 embryos. Histological details of gonadogenesis follow the same pattern as described for other reptiles. Regression of the male paramesonephric duct closely follows testicular differentiation. Gonad lengths are longer at the warmer incubation temperature, and ovaries are generally longer than testes at each stage and for each temperature. Although sexual differentiation takes place at about the same stage as in other turtles with TSD (18-20), in A. spinifera this differentiation is irreversible at this stage, while in some of the TSD species sex is reversible until about stage 22. This immutable, definitive sexual differentiation may support the hypothesis of an accelerated chronology of sex differentiation for this species. We also note that sexual dichromatism at hatching is known in this species and may provide additional evidence of early differentiation. J. Exp. Zool. 290:190-200, 2001.

Temperature-dependent sex determination and gonadal differentiation in reptiles

In many reptile species, sexual differentiation of gonads is sensitive to temperature (temperature-dependent sex determination, TSD) during a critical period of embryonic development (thermosensitive period, TSP). Experiments carried out with different models including turtles, crocodilians and lizards have demonstrated the implication of estrogens and the key role played by aromatase (the enzyme complex that converts androgens to estrogens) in ovary differentiation during TSP and in maintenance of the ovarian structure after TSP. In some of these experiments, the occurrence of various degrees of gonadal intersexuality is related to weak differences in aromatase activity, suggesting subtle regulations of the aromatase gene at the transcription level. Temperature could intervene in these regulations. Studies presently under way deal with cloning (cDNAs) and expression (mRNAs) of genes that have been shown, or are expected, to be involved in gonadal formation and/or differentiation in mammals. Preliminary results show that homologues of the WT1, SF1, SOX9, DAX1 and AMH genes exist in TSD reptiles. However, the expression patterns of these genes during gonadal differentiation may be different between mammals and TSD reptiles and also between different reptile species. How these genes could interact with aromatase is being examined.

Genetics of sex determination: exceptions that prove the rule

The dogma that male and female embryos develop in identical fashion until SRY initiates Sertoli cell differentiation in the hitherto bipotential gonad is in need of reevaluation in the light of data that do not fit into this scheme. One of the exceptions that proves the rule of sex determination is true hermaphroditism, the existence of individuals with both testicular and ovarian tissue. Furthermore, the two types of tissues are asymmetrically distributed, ovaries being more common on the left side and testes and ovotestes on the right. Hermaphrodite mice also exhibit bilateral asymmetry of gonad differentiation, but in the opposite direction: ovaries on the right, testes and ovotestes on the left. To explain these asymmetries, it is necessary to consider the relationship between growth and gonadal differentiation. The idea that accelerated growth precedes histological differentiation of the testis has recently been confirmed by the finding that Sry induces cell proliferation in fetal mouse gonads, suggesting that the differentiation of Sertoli cells may be dependent on a critical cell number. Recent evidence has also shown that XY embryos develop faster than their XX counterparts at very early stages of development, and it has been reported that SRY and ZFY are expressed in early human and murine embryos. The relationship between growth and sex differentiation links the mammalian system with those of nonmammalian vertebrates with temperature-dependent sex determination. Early growth differences between male and female human embryos question the belief that all sex differences in later life are due to gonadal hormones.