The past and future of “sex genes”

2.1 Sex chromosomes

From the beginning of the 20th century onwards research shifted towards genetic explanatory models of sex determination. In 1905 Nettie M. Stevens showed that inheritance of the Y chromosome produced male mealworms [17]. Subsequent experiments in the fruit fly Drosophila melanogaster revealed that changes in the ratio of sex chromosomes (gonosomes) and autosomes produced intersex flies, underlining the importance of a specific gonosome number for a specific sex development [18]. In the mid 1940s, A. H. Sturtevant described the transformation of female to male fruit flies through homozygous expression of the autosomal transformer gene tra, extending the influence of genetic sex determination outside the sex chromosomes [19]. In 1957 Mathilde Denon and Leo Sachs summarised three sets of causal factors for intersexuality: 1. single gene mutations, 2. abnormal distribution of sex chromosomes, and 3. endocrine dysfunction induced by external factors or by genetic factors affecting the endocrine system, meaning steroid hormone production and action [20]. The advancement of cytological methods made it possible to assess human sex chromosomes [20]. This made it possible to give individuals with a difference in gonosome numbers a molecularily proven diagnosis of difference in sex development (DSD). DSD defines a group of congenital conditions in which chromosomal, gonadal or anatomical sex is atypical. There are currently three main subclasses of DSD: sex chromosome DSDs, 46,XY DSDs, and 46,XX DSDs. Sex chromosome DSD is considered to comprise Turner Syndrome (45,X), Klinefelter Syndrome (47,XXY) and sex chromosome mosaicism (45,X/46,XY and 46,XX/46,XY). 46,XY DSDs and 46,XX DSDs comprise differences of gonadal development, differences of steroid hormone synthesis and differences of hormone action [21].

2.2 Genetic sex development

The identification of genes involved in sex development began in the mid-1980s due to advancements in cloning and sequencing technologies. Genetic factors leading to DSD are currently divided into genes involved in gonadal development, in steroid hormone biosynthesis and in steroid hormone action. Mutations in these genes can lead to both 46,XY and 46,XX DSD. Notably, most of these genes are autosomal [22].

The first mammalian sex determining gene, the sex determining region on the Y chromosome (SRY), was described in 1990 [23] and is still considered as a “master switch” that induces male sex development. In humans, 46,XY individuals lacking the SRY gene do not develop male gonads and appear phenotypically female at birth. In contrast, a translocation of the SRY gene on the X chromosome or autosomes in 46,XX individuals is associated with a male genital development but hypogonadism and infertility, indicating that SRY is necessary but not sufficient for testis development. Indeed, numerous other genes have now been identified as involved in testis development [24] (Figure 2).

Female sex development has long been considered as a default or passive pathway, where the absence of SRY in 46,XX individuals was considered to be sufficient to explain ovarian development. Today however, several genes involved in ovarian development have been described [25, 26] although a master switch for ovarian development remains elusive (Figure 2). The observation that 45,X women (Turner syndrome) develop streak gonads (hypoplastic ovaries without steroid hormone production and a persisting tubular structure) suggests a dosage-sensitive factor on the X chromosome that if expressed only from one allele impairs ovarian development. So far only one dosage-sensitive gene on the X chromosome has been described, NR0B1 (DAX-1), which has been shown to be involved in both ovarian and testicular development. When overexpressed it impedes testis development in 46,XY individuals [27]. No loss of function mutations in NR0B1 have been described in 46,XX individuals with ovarian dysgenesis so far.

At around 8 weeks post-conception the developing gonads start to express genes involved in steroid hormone biosynthesis. Several pathways leading to the production of androgens as well as estrogens have been identified. Steroid hormone production has been shown to occur in embryonic testis and ovaries; sex-specific hormone production in males peaks from 10 to 15 wpc for androgens and in a similar time window in females for estrogens [28]. One limitation to measuring fetal estrogen is the high estrogen production by the placenta during mammalian pregnancy. Placental estrogen production led to the hypothesis that estrogen secretion by the ovaries is negligible compared to its high placental production. In contrast to the well-established embryonic androgen action, a local paracrine and autocrine estrogen action has not been identified so far. The embryonic testis produces anti-Müllerian hormone (AMH) and testosterone. AMH causes regression of the Müllerian ducts and the onset of testicular descent. Testosterone causes the differentiation of the Wolffian ducts. In mice, NR2F2 represses the development of the Wolffian ducts during female development (Figure 2). Little is known about the role of estrogens in the development of the female genitalia. Current models are still based on the assumption that the absence of testicular testosterone and AMH production is sufficient to produce female sex differentiation – underpinning the default or passive way of female development [22, 29].

2.3 The role of estrogens in sex development

One observation suggesting that estrogens influence the differentiation of the genitalia comes from the effect of the synthetic estrogen diethylstilbestrol (DES) that was given to pregnant women in order to prevent miscarriages in the 1970s. While the treatment did not show the desired effect, boys exposed to DES during embryonic development were born with an increased incidence of hypospadias (where the urethra opens on the underside of the penis or between the anus and the scrotum), indicating that estrogens at supraphysiological doses can inhibit male genital development. Female embryos exposed to DES showed an increased incidence of vaginal cancer later in life. Interestingly, experiments performed in mice in the 2000s showed an effect on both male and female genital development when exposed to intrauterine DES [30]. Another study investigated estrogen action by knocking out the Esr1 in mice. Female Esr1 knock-out mice developed a longer clitoris with cartilage, which is typical of the penis but not for the clitoris [31]. This indicates a suppressive role for estrogens in preventing male sex development and calls into question the default model of female sex development.

Figure 2:

Human sex development from the indifferent gonad to sex differentiation is a concerted interplay of numerous genetic and hormonal factors. Shown are gene products known to be involved in human sex development that are mutated in individuals with DSD. Androgens (Testosterone and Dihydrotestosterone (DHT)) exert their role through the androgen receptor. The role of estrogens in female sex differentiation is poorly understood. Indifferent gonad with undifferentiated genitalia (left). Female (above) and male (below) gonadal determination and sex differentiation. (modified from [29]).

Steroid hormones act through steroid hormone receptors, namely the estrogen receptors 1 (ESR1) and 2 (ESR2) that are activated by estrogens and the androgen receptor that is activated by androgens. Much of the current knowledge about the role of androgens in male sex development has been derived from studying complete androgen insensitivity syndrome (CAIS). Affected individuals possess male gonosomes and testosterone-producing gonads, yet develop a female phenotype with a blind-ending vagina, and no Müllerian structures (namely no uterus) or prostate due to the absence of a functional androgen receptor. This syndrome has been used as an explanatory model for the role of androgens in the development of the male inner and outer genitalia (Figure 2). It also led to the conclusion that the absence of androgen is sufficient for female sex differentiation. This conclusion, however, is not valid. Individuals with CAIS do produce estrogens and have functional estrogen receptors and are therefore unsuitable for studying estrogen action. In addition, the development of inner female organs cannot be studied in this model as male gonads are functional in AIS and produce AMH that impedes the Müllerian structures’ development. The only available model to study estrogen action in human sex development at the organism level is complete estrogen resistance, an extremely rare condition in humans. Due to its rarity, mutations in ESR genes and the underlying molecular mechanisms of estrogen insensitivity have not been thoroughly studied. Individuals with mutations in ESR1 show a hypoplastic uterus and ovaran dysfunction (32). Loss of function mutations, in the human ESR2 are associated with ovarian dysgenesis a hypoplastic uterus and hypoplastic labia majora in females. Interestingly, in men ESR2 mutations lead to testicular dysgenesis, suggesting that this receptor is involved in gonadal development in both sexes [32]. Esr1-deficient female mice show a hypoplastic uterus and reduced fertility and an enlarged clitoris [29], indicating both an activating role in producing female sex organs and an inhibitory role in preventing male sex development.

Research on sex development in biology follows a model of causality, in which sex chromosomes are the basis of gonad formation which are the basis of steroid hormone production. This model is also based on assumptions of the two sexes, as illustrated by the idea of a master switch in male sex development. DSD contradicts this model and therefore gives space for a more complex model of sex where the concerted action of different not necessarily binary factors give rise to an individual sex (Figure 2).

3.1 Causality: Genetic determinism and essentialism

The gene SRY, at the time of its identification in 1990, seemed to confirm a pre-existing model of how genetic sex determination works: one single “master” gene on the Y chromosome directs the development of the testes and thereby determines the male sex. However, even in the late 1990s this model was heavily challenged from two sides: firstly by empirical findings, which complicated the picture considerably, and secondly by conceptual considerations about how genes work in an organism. More genes seem to be involved, some of them non-Y chromosomal genes (such as NR0B1, SOX9, DMRT1 and WNT4), which could override SRY to cause sex reversal. SRY did not offer a sufficient explanation of the phenotypes of human intersex subjects. Ken McElreavey et al. therefore suggested a “regulatory gene cascade”, in which many factors participate in pushing the balance of sex determination towards male or female [37]. This pluralistic model of determination would better explain the observed variations of intersex or DSD (differences of sex development) phenotypes. Sarah Richardson argues in her excellent review of this discussion that gender criticism played a considerable role in this transition away from genetic determinism and essentialism, and toward a contextual, process-oriented view of complex (epi-)genetic networks that emphasise the interacting factors involved in sexual development [38]. These theories are based on the idea that the role of some genes as genetic “switches” is not the cause but the result of complex interactions with other factors.

Genetic determinism, coupled with an essentialist view of the role of the genome in organismic development, has a long history going back to August Weismann’s late 19th century belief that the chromosomes contain a plan (“Architekturplan”) of the developing organism. Long before details could experimentally be elucidated, a general theoretical view promised an explanation of the role of the genome in the organism: the genome contains the information of a “genetic program” that enables the organism to develop the adult structure. Eminent biologists including Ernst Mayr, Jacques Monod and François Jacob successfully promoted this view in the early 1960s [39], because it solved the theoretical puzzle of how seemingly goal-directed developmental processes could be explained mechanistically. However, not least due to findings such as mRNA editing and alternative splicing indicating the contextual nature of genetic causality, a general uneasiness with genetic determinism and genetic essentialism arose in philosophy of biology.

“Systemic” explanations preserve the causal importance of genes while avoiding genetic determinism and essentialism. They have increasingly been recognised as more congenial to empirical findings [40, 41].

3.2 Dualism

Thinking in dualisms manifests itself in the division of autosomal and sex chromosomes, or in the distinction between X and Y chromosomes. While in public the X chromosomes are described as “motherly”, more sociable, conservative, stable [42], the Y chromosome is often seen as active, dominant, a “macho entity” [43]. These divisions often relate to a dichotomy of active and passive, since in Western culture masculinity is seen as an active agent and femininity is seen as passive. Dualism also assumes sex binarity, i. e. the view that there are only two sexes and no variants in between.

Dualism also resonates with the (now challenged) view that female sex development is a “default” pathway that occurs naturally without the effects of an (active) male-determining gene. In the view of biological femininity as absence, sex determination research has focused on questions of the genetics of male determination. As we argued in section 2, this view is inconsistent with a series of recent findings that show that the female developmental pathway is in many ways the result of “active” genetic and hormonal processes.

3.3 Androcentrism

The genetics of sex determination developed in a traditionally male-dominated scientific world. From there it has clearly inherited an androcentric bias, for instance by focusing on the Y chromosome and on testosterone and by looking for male “master” genes. The mechanisms of embryogenesis have been investigated with primary attention to the development of the male phenotype [44]. Testis formation was seen as the crucial sex-determining event, female sexual development as “default”. Researchers saw “the question of sex determination as identical to the question of genetics of male testes determination” [38]. As a consequence female sex development was under-researched and is today much less well understood than male sex development.