The idea of an XY female and an XX male may seem a contradiction in terms. Yet to those who have studied the mechanics of sexual differentiation, perhaps what is more remarkable is that most of the time there is a clear association between being XX and female, and being XY and male. The XX and XY embryos are built on a fundamentally similar outline plan, and only as development proceeds do certain modifications evolve. If at any point in this sequential process some genetic instruction is faulty, inappropriate, or cannot be acted on, the direction of sexual development may proceed imperfectly. In the extreme, the opposite path is taken. This latter state is the particular subject of this chapter, with more focus on those forms in which cytogenetic and molecular genetics comprise the key diagnostic investigations.
BIOLOGY
Somewhat simplified, the fundamental plan of the genital tract is that bilateral gonads connect with bilateral paired internal ducts (müllerian and wolffian), which enter a midline genital sinus, opening at the perineum. This opening is buttressed on each side by labioscrotal folds and capped above by a phallus. The basic plan of the genital ridge is laid down according to instruction from, among other genes, WT1 and SF1. Otherwise uninfluenced, the gonad inherently develops into an ovary, and the duct system develops into fallopian tubes and uterus. The genital sinus remains as an opening (the vagina), flanked and surmounted by labia and clitoris. The female state results.
But if a Y chromosome is present, or at least that part of the Y that contains SRY, the testis-determining gene, this inherent plan is modified. The SRY gene encodes a DNA-binding protein that enables a cascade of activation of genes elsewhere in the genome that causes the primordial gonad to develop into a testis (Sinclair et al., 1990). Included among these genes are DMRT1, DAX1, and SOX9, each of which encodes a transcription factor.
The most conserved gene in sex differentiation is DMRT1, mapping to 9p24.3. DMRT1 is expressed in greater amount in the male embryonic gonads than in the female, and expression is localized to the sex cords, which will give rise to the Sertoli cells of the testis (Smith et al., 1999a). The DAX1 gene located at Xp21.3 is a repressor of male gene expression. If, as a result of being duplicated, it is overexpressed, male-to-female sex reversal is the consequence—hence this gene's other name of DSS, for dosage-sensitive sex reversal.
The sequence whereby germ cells acquire a testicular or an ovarian destiny is exquisitely dose-sensitive: too much DAX1 product displaces WT1 from its association with SF1, and the testis-inducing influences of SRY and SOX9 are thwarted. More genes along this pathway will, no doubt, be discovered (Umehara et al., 2000). WT1 comes under the influence of WNT4, on chromosome 1p35, and this gene is another candidate (Jordan et al., 2001). Two 46,XY women having an unbalanced translocation involving the same 6p25 breakpoint, t(X;6) in one and t(6;13) in the other, may point to the existence in this region of a further sex-determining locus (Batanian et al., 2001a).
When the SRY-triggered pathway is followed normally, the testis forms and secretes hormones; androgen influences the genital tract to masculinize, and anti-müllerian hormone causes regression of the female müllerian ducts. A vas deferens forms from the duct system. The phallus enlarges. The labioscrotal folds fuse in the midline and accommodate the descending testes. The male state results.
The reader wishing a detailed treatment of the genetics of sexual differentiation is referred to the following sources: the issue of the American Journal of Medical Genetics devoted to “Sex Determination and Sex Differentiation in Humans” (Vol. 89, No. 4, 1999), Pinsky (1999), Kent-First (2000), and Koopman (2001).
XY Female
XY Pure Gonadal Dysgenesis (Swyer Syndrome)
The rare familial form provides a unique example of a Mendelian condition that can be inherited in an X-linked recessive, Y-linked, or sex-limited autosomal dominant mode. In the X-linked forms or autosomal dominant forms, the XY female has a perfectly normal Y chromosome, with a normal SRY testis-determin-ing gene; presumably, there is a mutation in a gene (whether this be X-linked or autosomal) controlling a later event in the testicular developmental pathway. In the Y-linked form, there is a mutation in the SRY gene. In some Y-hemizygotes, the mutant gene has nevertheless been able to reach a threshold of operation and to induce testis development, while in others with the same mutation it has not. Thus, for example, an XY male with a mutation in SRY may be a normal fertile man, while his XY child may be a daughter. The threshold is apparently all-or-nothing: partial expression, that is to say, an intersex state, does not result (Jäger et al., 1992; Imai et al., 1999). A man may be a gonadal mosaic for an SRY deletion, as presumably was the father in Barbosa et al. (1995). Two sisters with XY gonadal dysgenesis (one with gonadoblastoma) had a deletion of SRY, but their father showed a normal SRY result; there were three other normal sisters and six normal brothers. Similarly, Schmitt-Ney et al. (1995) describe two XY sisters and a half-sister whose father was shown to be mosaic for an SRY mutation.
Sporadic occurrence is usual, and in about 10%–15% of these the SRY gene has a de novo mutation that abolishes its function of testis determination (Hawkins, 1994; Battiloro et al., 1997). A further 10%–15% have a deletion of SRY presumably due to aberrant X/Y recombination. The remaining 70%–80% have no detectable SRY mutation (Scherer et al., 1998). XY females with an intact SRY gene presumably have a mutation (or aberrant functioning) in one of the other genes in the testis-determining pathway. On distal 9p there is a gene, possibly DMRT1, which, when deleted—that is, the patient is hemizygous at this locus— results in 46,XY gonadal dysgenesis without additional anomalies (Ottolenghi and McElreavey, 2000), and this type of mechanism may account for some of the patients without SRY mutations.
The gonad in the XY female is dysgenetic, and is seen as a “streak” gonad. The genital tract feminizes. The lack of female sex hormones causes failure of normal pubertal development. Amenorrhea and failure of pubertal development are the usual complaints that lead these girls to seek medical advice. Gonadoblastoma, a premalignant neoplastic change in the dysgenetic gonad, is common, and may progress to dysgerminoma (Verp and Simpson, 1987; Berg et al., 1989; Lukusa et al., 1991; Uehara et al., 1999a). Familial ovarian malignancy was a notable observation in a sibship of three XY women (the karyotype presumed in two who had died at ages 19 and 20 years) described in Kempe et al. (2002). An extraordinary case is reported by Cussen and MacMahon (1979) of an XY girl from whose dysgenetic gonad an XY oocyte was obtained.
Complete Androgen Insensitivity Syndrome (Testicular Feminization)
Here, the defect lies further down the developmental path. The gonad becomes a testis, and produces testicular secretions, but the genital tract, internal and external, is resistant to the effects of androgen. The inheritance is X-linked recessive, and the locus is the androgen receptor gene at Xq11–q12 (Brinkmann, 2001). One example is on record in which, in a sense, the X-linkage was directly visible to the cytogeneticist: that is to say, the X chromosome was abnormal, including the region containing the androgen receptor locus. An affected aunt and niece had the karyotype 46,Y,inv(X)(q11.2q27) and the connecting mother 46,X,inv(X) (q11.2q27) (Xu et al., 2003). The individual appears externally very much as a female, but there is amenorrhea and pubic and axillary hair is absent. Internally, the vagina is short, and the uterus and tubes are represented only by remnants; the testes may be in the inguinal canal (Boehmer et al., 2001). A unique case due to uniparental disomy X in a woman with the XXY karyotype (Uehara et al., 1999b) is noted on p. 330. Malignancy of the gonad, gonadoblastoma or dysgerminoma, is less of a concern occuring in a minority, in 9% in one study, although in two series no cancers were found. About 5% may be an average figure (Lukusa et al., 1991; Rutgers and Scully, 1991; Collins et al., 1993; Alvarez-Nava et al., 1997; Chen et al., 1999b).
XY Female with Extragonadal Defects
A number of rare conditions exist in which sex reversal coexists with physical, metabolic, and/or mental defect. By way of example, one of these is campomelic dysplasia (campomelia refers to long bone bowing) with sex reversal, a syndrome of skeletal dysplasia with female genital tract development. The usual cause is a mutation within the SOX9 gene (at 17q24.3–q25.1), one of the genes operating on the sexual differentiation pathway and which also influences limb bud mesenchymal development (Wagner et al., 1994). Another cause of campomelic dysplasia is deletion in an apparently balanced 17q translocation with haploinsufficiency at the locus (p. 275). A significant recurrence risk applies, on the order of 5%, which may include the etiologies of mildly affected transmitting parent, parental gonadal mosaicism, recessive forms, and familial rearrangement involving 17q (Mansour et al., 1995).
Other Rare forms of Gonadal Agenesis and Dysgenesis
In a presumed autosomal recessive form of XY or XX gonadal agenesis, the gonads are absent, and the external phenotype is female (Mendonça et al., 1994). An autosomal dominant form of XY gonadal dysgenesis could be assumed in the family reported in Le Caignec et al. (2003a), in which some affected XY individuals were female, some were male, and in one instance male-to-male transmission was observed. In some women with XX gonadal dysgenesis, immature ovarian follicles are seen, and one cause is homozygosity for a mutation in the follicle-stimulating hormone (FSH) receptor gene (Aittomäki et al., 1995); there may be a subtler effect on male gonadal function, expressed as a reduction in the sperm count. Homozygosity for a mutation in the luteinizing hormone receptor (LHR) gene causes varying degrees of male pseudohermaphroditism with feminized external genitalia, short blind vaginas and no uterus or tubes, and gonads that are testes but lacking Leydig cells (the cells that produce testosterone) (Kremer et al., 1995).
There are rare disorders of gonadal agenesis and extragonadal defects, for example, the syndrome of Kennerknecht et al. (1995) of agonadism in XY sisters with mental retardation, short stature, and several minor anomalies. Another is the syndrome of neurological and visual compromise with streak gonads and a female genital tract in an XY child of consanguineous parents, reported by Teebi et al. (1998).
Patients with the Denys-Drash and Frasier syndromes with gonadal dysgenesis and renal anomalies frequently have sex reversal due to mutations in WT1, one of the sexual differentiation pathway genes mentioned above (Hammes et al., 2001).
XX Male
Most XX males arise from the presence of Yp material (rarely visible cytogenetically) on one of the X chromosomes (Rigola et al., 2002), from occult XXY/XX mosaicism, or from the inappropriate activity of a gene that is normally switched on only in response to a Y-originating genetic instruction. In about three-quarters of cases the SRY gene is present, typically the consequence of an abnormal exchange between the X and Y during meiosis I in gametogenesis in the father, and thus clearly a sporadic event (Weil et al., 1994; Wang et al., 1995). These are referred to as SRY+ XX males. The phenotype in the SRY+ XX male is similar to that of Klinefelter syndrome, presumably reflecting the similar basic genotypes of active X + inactive X + SRY in the two conditions; however, the XX male differs in being of normal height and of unimpaired intelligence (Ferguson-Smith et al., 1990). Margarit et al. (1998) describe six SRY+ cases due to translocation of Yp material to Xp22.3, in whom different Y breakpoints could be identified, but whose clinical phenotypes were very similar: normal intelligence, normal stature, and testicular atrophy with azoospermia. In these SRY+ XX males, a more accurate designation would be 46,X,der(X)t(X;Y), or more fully, 46,X,der(X)t(X;Y)(p22.3;p11.2), although the exchange is not usually visible on standard cytogenetics. Thus this entity is also discussed in the section on the X;Y translocation (Chapter 5).
XX males with no SRY gene are denoted SRY-. The fact of male development being able to proceed, despite the lack of SRY product, presumably reflects an inappropriate activation of the testis-determining cascade in an otherwise normal 46,XX embryo, either as a sporadic stochastic event or because of some genetic predisposition. Jarrah et al. (2000) report an extended inbred kindred in which XX individuals with varying degrees of masculinization were present, and suggest that in this family, SRY- XX maleness and XX true hermaphroditism represented a continuum of the same disorder (see below, True Hermaphroditism). The SOX9 gene on 17q was implicated in the case of Huang et al. (1999b), a child with male development, albeit imperfect, in the whom the karyotype was 46,XX,dup(17) (q23.1q24.3)/46,XX.
Three cases are reported of males with 47,XXX chromosomes. In one well-studied example, the man was mildly intellectually disabled, with gynecomastia and hypogenitalism, and had severe testicular atrophy on biopsy (Ogata et al., 2001). One X of the three was positive for SRY. In addition to an Xp–Yp interchange in paternal gametogenesis that produced the SRY-positive X chromosome, a coincidental maternal nondisjunction was responsible for a disomic X ovum. Thus, the combination at fertilization was XX(mat) + der(X)t(X;Y)(pat), giving 47,XX,der(X)t(X;Y) and appearing karyotypically as “47,XXX.”
45,X Male
We refer to this is rare disorder on p. 118. Most 45,X males have, in fact, a molecular translocation of the SRY gene to an autosome or to the X chromosome (and might therefore be thought of as a type of Y;autosome or X;Y translocation), while a few are actually X/XY mosaics. A unique case that brings together this, the foregoing and the following sections is described in Modan-Moses et al. (2003), a patient with 46,XXSRY+/45,XSRY+/45,XSRY- mosaicism, who presented with a clinical picture of true hermaphroditism.
True Hermaphroditism
True hermaphroditism generally presents as a problem in determining the sex of a newborn infant—in other words, genital ambiguity (Hadjiathanasiou et al., 1994). The formal definition of true hermaphroditism is that the gonads comprise both ovarian and testicular elements: there may be a testis and an ovary, or one or both may be an ovotestis. The most common karyotype is 46,XX, seen in 60%. A third have mosaicism with one cell line which includes Y chromosomal sequences, mostly 46,XX/46,XY, and a few are 46,XY (Queipo et al., 2002).
Most of the 46,XX cases test negative on peripheral blood analysis for the SRY gene, and in some of these the basis of the defect may be sporadic inappropriate activation of the testicular developmental cascade in part of the gonadal tissue during its embryonic formation. XX true hermaphroditism is unusually prevalent among certain indigenous Black populations in Southern Africa, and Spurdle et al. (1995) excluded the presence of SRY and of uniparental disomy X in all of 16 individuals studied. Alternatively, an apparent XX karyotype may harbor Y material, as Margarit et al. (2000) show in a woman reared as a boy with hypospadias who went on to have gender change surgery after testing 46,XX. Several years later, reanalysis revealed a tiny segment of Yp translocated on to the X long arm, 46,X,der(X),t(X;Y)(q28;p11.31). A more common explanation in the 46,XX case may be cryptic mosaicism within the gonad itself, with an island or islands of tissue containing the SRY gene (Ortenberg et al., 2002; Queipo et al., 2002). One 46,XY case had a postzygotic mutation in SRY with SRY+/SRY- gonadal mosaicism (Braun et al., 1993). Presumably the SRY+ line was responsible for the testicular elements in the gonad and the SRY-line, for the ovarian elements.
The XX/XY state might result from the fusion of twin XX and XY embryos (XX/XY chimerism). Strain et al. (1998) reported a notable example of iatrogenic hermaphroditism, which followed IVF, presumably due to this scenario of an XX and an XY embryo fusing. Another theoretical route to the XX/XY state (in this instance, mosaicism, not chimerism) is from the postzygotic loss of the X and of the Y in separate cells of an initially 47,XXY conception (Niu et al., 2002). A single case is recorded of true hermaphroditism associated with an autosomal abnormality, and this may reflect the agency of an autosomal gene on the cascade of sexual differentiation (Aleck et al., 1999). This child had ambiguous genitalia, with one ovarian and one testicular gonad, and karyotyped 46,XX,rec(22)dup(22q)inv(22)(p13q13.1)mat. Testing for SRY was negative.
Rare familial 46,XX cases may reflect a mutation, whether autosomal or X-linked, that induces the testis developmental cascade to proceed at a post-SRY stage (Zenteno et al., 1997). An interesting example of such a possible scenario is provided by Slaney et al. (1998), who describe the case of four 46,XX cousins with abnormal sexual differentiation. Three were 46,XX true hermaphrodites, and one was a 46,XX male. The putative testis-development gene had been transmitted through two mothers.
GENETIC COUNSELING
XY Pure Gonadal Dysgenesis
Familial/Inherited Cases
XY gonadal dysgenesis, when familial, is mostly inherited as an X-linked recessive or autosomal dominant with expression limited to the XY state. The risk in this case to the known female heterozygote of having an affected child is, as for any X-linked recessive condition, 25%. She cannot be distinguished on any phenotypic basis, but only on her position as an obligate carrier in the pedigree. In the absence of knowledge of a gene mutation (and none is known at the time of writing), the risk to other female relatives is given by Bayesian analysis. Although the XY female phenotype is close to that of a normal female, but of course associated with infertility, some couples may want to consider prenatal diagnosis. The use of cytogenetics (XY) and ultrasound morphology (female external genitalia) would presumably allow detection of the condition.
The Y-linked form is recognized by the demonstration of an SRY mutation carried by the XY girl and her XY father. This circumstance would allow the counselor the rare opportunity to apply principles of Y-linked inheritance to risk estimation. Mutational analysis of the SRY gene (including deletion detection) may provide the basis for carrier detection and prenatal or preimplantation diagnosis.
Couples electing not to consider prenatal diagnosis (or to continue a pregnancy in which a positive diagnosis has been made) should know of the importance of two factors in managing these girls. First, the psychosexual orientation of these individuals is female. But with secondary sexual characteristics developing incompletely, and infertility being invariable, their self-image is vulnerable. In discussing the condition with parents, the counselor should note the importance of using language that reinforces their view of themselves as girls and women, and avoid using such terms as “genetic male.” It may be explained to them, beginning in simple terms in childhood, that a genetic factor prevented their ovaries from developing normally (Goodall, 1991). Pregnancy may be achievable with IVF using a donor ovum (Kan et al., 1997; Dirnfeld et al., 2000). Second, there is a substantial risk of neoplastic change in the dysgenetic gonad. A gonadoblastoma arises in around half of familial XY gonadal dysgenesis. The gonadoblastoma itself is noninvasive, but it is often associated with malignant elements, most commonly dysgerminoma, which do invade. Thus given that the gonad does not usefully contribute in terms of hormone production, early (first-decade) gonadectomy is advisable (Troche and Hernandez, 1986; Verp and Simpson, 1987; Lukusa et al., 1991).
Sporadic Cases
Advice on the recurrence risk in the sporadic case is less straightforward. If a de novo SRY mutation is demonstrated, only paternal testicular mosaicism—which, for the record, has been observed—could imply an increased risk for recurrence. Again, prenatal diagnosis by chromosomal/ultrasound gender discordance should be feasible.
The rare syndromes of XY female with extragonadal defects need to be judged on their individual merits.
For the XY woman herself, assisted conception is possible if a uterus is present, and a handful of successful pregnancies, using donated gametes, have been reported (Kan et al., 1997; Selvaraj et al., 2002).
Complete Androgen Insensitivity (Testicular Feminization)
This condition is inherited as an X-linked recessive trait, and the risk of recurrence follows classic Mendelian principles. The carrier may be identified and prenatal diagnosis accomplished by molecular analysis of the androgen receptor gene. While complete androgen insensitivity typically has a consistent phenotype within families, allowing for good prediction of the consequences of recurrence, incomplete androgen insensitivity can have variable phenotypes within a family (Boehmer et al., 2001). Issues relating to prenatal diagnosis are discussed in Morel et al. (1994), who also make the interesting but unsurprising point that incomplete forms can imply a worse burden than the complete form, with partially virilized males (known as Reifenstein syndrome) having “considerable psychological distress and poor function in their adult life.” Similar considerations with respect to gender orientation in the XY girl, as discussed in the preceding section, apply to complete androgen insensitivity. The risk for neoplastic change in the gonad is less, in the vicinity of 5%, in the case of testicular feminization. Thus, some propose that gonadectomy may reasonably be delayed to allow spontaneous pubertal feminization (Jones, 1978; Verp and Simpson, 1987), although regular clinical and imaging checks would be advisable.
XX Male
Many XX boys are not diagnosed until after childhood, by which time the parents are likely to have completed their family. Some cases may be recognized at amniocentesis following discordant karyotypic and ultrasonographic sex, or following the birth of a boy who was predicted to have been a girl. (We have seen one such child from an IVF pregnancy, whose parents had two more cryopreserved embryos available for potential future transfer.) The great majority occur as sporadic events in a family, and in these the likelihood of recurrence is very small. If the child is SRY+ and the father's X is SRY-, sporadic occurrence is, essentially, proven. As for the rare case of the SRY- XX male, once the postulated gene or genes have been identified (Zenteno et al., 1997), those cases that would carry a high recurrence risk will be able to be identified. If prenatal diagnosis is requested, and the fetus is 46,XX, testing for SRY along with an ultrasonic assessment of external genital morphology should enable distinction (Ginsberg et al., 1999).
True Hermaphroditism
The considerable majority of true hermaphroditism represent sporadic cases in a family. These are characterized by a 46,XX karyotype and absence (at least on peripheral blood analysis) of the SRY gene, and are presumed to reflect an “accidental” activation of the testis-determining cascade during gonadogenesis, or cryptic intragonadal mosaicism, as discussed above. In some cases, the cytogenetics (46,XY, 46,XX/46,XY or other mosaic karyotype) or molecular genetics (SRY mutation which is not present in father) may allow a more secure reassurance of nonrecurrence. Recurrence is very rare; but a positive family history would, of course, imply a high risk. In the SRY- form, a handful of families are described in which there is also a sib with XX male syndrome, and these cases may speculatively reflect “leaky mutations” in a gene operating at a point downstream in the cascade of sexual differentiation.