| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institute of Reproductive Medicine (E.V., M.Z., J.G., E.N.) and Department of Obstetrics and Gynecology (A.N.S.), University Clinics of Münster, D-48129 Münster, Germany
Address all correspondence and requests for reprints to: Eberhard Nies-chlag, Institute of Reproductive Medicine of the University, Domagkstrasse 11, D-48129 Münster, Germany. E-mail: eberhard.nieschlag{at}ukmuenster.de.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to distinguish the 46,XX male syndrome from the more frequent 47,XXY-Klinefelter syndrome in regard to clinical, hormonal, and epigenetic features.
Design: This was a case-control study.
Setting: The study was conducted at a university-based reproductive medicine and andrology institution.
Patients: Eleven SRY-positive 46,XX males were compared with age-matched controls: 101 47,XXY Klinefelter patients, 78 healthy men, and 157 healthy women [latter all heterozygous for androgen receptor (AR) alleles].
Interventions: There were no interventions.
Main Outcome Measures: There was a comparison of phenotype, endocrine profiles, and X-chromosomal inactivation patterns of AR alleles.
Results: The 46,XX males were significantly smaller than Klinefelter patients or healthy men, resembling female controls in height and weight. The incidence of maldescended testes was significantly higher than that in Klinefelter patients and controls. Gynecomastia was more frequent in comparison with controls, whereas there was a nonsignificant trend in comparison with Klinefelter patients. All XX males were infertile and most were hypogonadal. The inactivation patterns of AR alleles in XX males were significantly more skewed than in Klinefelter patients and women. Seven of 10 heterozygous XX male patients displayed an extreme skewing of more than 80% with no preference toward the shorter or longer AR allele. The length of the AR CAG repeat polymorphism was positively related to traits of hypogonadism.
Conclusions: XX males are distinctly different from Klinefelter patients in terms of clinical and epigenetic features. Nonrandom X chromosome inactivation ratios are common in XX males, possibly due to the translocated SRY gene. The existence of a Y-chromosomal, growth-related gene is discussed.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Ninety percent of these patients have Y chromosomal material including the SRY gene. Y sequences are usually located on the distal tip of the short arm of the paternal X chromosome (6). Earlier it was suggested that an unequal Y-to-X interchange occurs during paternal meiosis (7), which was corroborated by more recent studies (2, 8, 9, 10). Ten percent of XX males are SRY negative with different degrees of masculinization. Two major mechanisms of this phenomenon are discussed. A dominant autosomal or X-chromosomal inheritance of XX maleness has been described in several cases (11). It is suggested that a mutation in these unknown genes can trigger testis determination (12). As a second mechanism, a mosaicism with a prevalent XX-lineage and a hidden or scarce lineage containing a Y chromosome is discussed (12). A relationship between XX male prevalence and paternal age has not been observed (2).
To date, it remains unclear how this genetic and clinical entity has to be distinguished from the other, far more frequent disorder of males with two X chromosomes, the 47,XXY Klinefelter syndrome (2, 3, 13, 14, 15, 16, 17, 18). To this end, we present a comprehensive and systematic clinical, endocrinological, and (epi)genetic approach to the 46,XX male syndrome in a case-control setting involving large cohorts of age-matched 47,XXY Klinefelter patients, normal men, and healthy women.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
We evaluated 11 hitherto untreated XX male patients (one patient previously received two occasional injections of testosterone enanthate 250 mg, im) visiting our institute between 1981 and 2006. Patients’ ages ranged between 14 and 46 yr. These patients have not been reported previously.
The XX male patients were compared with 101 age-matched (± 2 yr) nonmosaic 47,XXY Klinefelter patients. Each individual 46,XX male patient was compared with nine or 10 untreated Klinefelter patients matching in age within a range of ± 2 yr. The second control group consisted of 78 age-matched healthy volunteers presenting for trials in male contraception. The 157 healthy women represent the heterozygous part (in terms of androgen receptor (AR) allele CAG repeat length) of a control group of another (Schüring, N., and J. Gromoll, unpublished study): the female control group consisted of women with normal ovarian function as determined by repeated hormone assessments and ultrasonographic demonstration of regular ovulation. They represented the proven healthy female partners in couples seeking advice for infertility, with husbands presenting with decreased sperm parameters. These women were studied in terms of X chromosome inactivation as controls for a trial involving women with the polycystic ovary syndrome. The age of the female control group matched our male groups (see Table 4
).
|
Parameters
The physical examination included the measurement of height, body weight, body mass index (BMI), and potential gynecomastia and the inspection of external sex organs. The anamnesis included questions concerning a history of maldescended testes. Testes and prostate size were determined by ultrasound as previously described (19). Bitesticular volume was calculated as the sum of the volume of both testes. Semen analysis was performed in our laboratory according to the World Health Organization laboratory manual (1999). In cases of azoospermia or severe oligozoospermia, analysis was performed after centrifugation of the ejaculate. The patients were requested to abstain from sexual activity for 48 h to 7 d before investigation.
Serum levels of LH, FSH, prolactin, testosterone, SHBG, estradiol, and prostate-specific antigen (PSA) were assessed as previously published (17).
Levels of free testosterone were calculated from the levels of SHBG and total serum testosterone according to the method described previously (18).
Standard laboratory procedures to assess liver functions and blood count were performed in the central laboratory of the clinic.
Cytogenetic analysis
Karyotyping was performed on peripheral blood lymphocytes in all groups of patients as previously described (18). In both 46,XX males and Klinefelter patients, 30 metaphase cells were assessed. SRY gene localization was established using fluorescent in situ hybridization with a SRY probe obtained from Vysis (Hoofddorp, The Netherlands). Cytogenetic analysis was performed in the Institute of Human Genetics of the University of Münster.
Molecular analysis
DNA was extracted from peripheral blood lymphocytes using the Nucleon kit (Amersham Life Science, Freiburg, Germany).
SRY and ZFY determination and AZF-region analysis. Diagnostic testing of deletions of the three discrete regions, AZFa, AZFb, and AZFc, located on the long arm of the Y chromosome, was performed by PCR amplification. The set of PCR primers that was used in multiplex PCRs for the diagnosis of microdeletion of the AZFa, AZFb, and AZFc region included: sY14 (SRY), ZFX/ZFY, sY84, sY86, sY127, sY134, sY254, and sY255 (20).
Assessment of the number of CAG repeats in the androgen receptor gene
DNA was extracted from peripheral blood lymphocytes and the procedure followed the method previously described (21). Patients with two detected bands of CAGn length were considered heterozygous and subjected to X chromosome inactivation analysis.
X chromosome inactivation analysis (inactivation of the AR alleles)
Analysis using leukocyte DNA of heterozygous patients was performed as previously described (17). In contrast to the method described earlier, we used a larger amount of DNA and HpaII for analysis. Equivalent 200-ng DNA aliquots were digested with 100 U HpaII. For analysis of the X-chromosome inactivation in XX male patients, we used the ABI Prism 3730 DNA analyzer (Applied Biosystems, Darmstadt, Germany). This change in method did not change the determination of the inactivation status of the alleles. Moreover, during the transition of method, we included samples with a known inactivation pattern as an internal control for the different assays to prove proper restriction of the samples. Total fluorescent peak areas for both alleles in digested and undigested samples were calculated by GeneScan program (Applied Biosystems; version 3.7). Analysis was performed only in heterozygous persons.
Statistics
Due to unequal sample size, we evaluated all data by nonparametric approaches. The XX males were compared with the other groups by Kruskal-Wallis tests followed by post hoc analyses according to Dunn and corrections according to Bonferroni. For testing prevalence of gynecomastia or maldescended testis, we used 
2 tests. Associations of CAG repeat length with phenotype were assessed by Spearman's correlation. Comparisons of inactivation of short or long AR alleles in heterozygous persons were performed by Wilcoxon’s matched pair test. For statistic analysis we used the statistical package SPSS for Windows (version 12.0; Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Phenotypically all patients were males. The clinical characteristics of the XX male group are displayed in Table 1. Table 2 demonstrates endocrinological findings of the XX male group, and Table 3 shows data of the cytogenetic and molecular investigations. A comparison with age-matched control groups is given in Table 4
. Mean height and weight correspond to the 15th percentile for men (Prader) and were significantly lower than in 47,XXY Klinefelter patients and normal men; rather, correspondence to the respective means observed in the female group was seen (Table 4).
|
|
|
). However, their height (179 ± 9 cm) was comparable with Klinefelter patients and normal men but significantly different from 46,XX males (170.6 ± 6 cm, P = 0.001).
Gynecomastia and maldescended testes in XX males (in six of 11 and five of 11 patients, respectively) exhibited higher prevalence in comparison with 47,XXY Klinefelter patients (in 39 of 101 and 17 of 101, respectively) as well as with normal men (in five of 78 and three of 78) (Figs. 1 and 2).
|
|
The evaluated serum levels of total protein, cholesterol, triglycerides, glucose, uric acid, urea, creatinine, bilirubin, glutamic-oxaloacetic transaminase (GOT) (also known as aspartate aminotransferase), glutamate-pyruvate transaminase (GPT) (alanine aminotransferase, 
-glutamyltransferase, alkaline phosphatase, and lactate dehydrogenase in the XX-males were in the normal ranges.
Mean GOT levels (17.9 ± 7 U/liter) and mean GPT levels (17.5 ± 5 U/liter) in the XX males were significantly lower than in the normal male group (29 ± 11 U/liter, P = 0.001, and 30 ± 23 U/liter, P = 0.03, respectively). No significant differences in metabolite parameters, with the exception of creatinine (0.87 vs. 0.96 mg/dl, P = 0.035), were found between the XX male group and the group of Klinefelter patients.
Blood cell count
Evaluation of leukocyte, erythrocyte, and thrombocyte counts, hemoglobin values, hematocrit, mean corpuscular hemoglobin concentration (MCHC), mean cellular volume, and mean corpuscular hemoglobin showed that the mean values of these parameters in the XX male group were in the normal range for the central laboratory of our hospital. Statistical analysis showed a significantly lower level of MCHC (33.0 ± 1.3U/liter; P = 0.028) and hemoglobin (14 ± 1 g/dl; P = 0.02) mean value in the XX males, compared with the group of normal men (33.9 ± 0.7 U/liter and 15 ± 1 g/dl).
Epigenetic analysis
Data of the AR gene polymorphism and inactivation of the AR gene in XX males are shown individually in Table 3. There were no significant differences of CAGn length between the groups (Table 4).
Differential inactivation patterns of the short and long AR-gene alleles in XX male patients in comparison with 47,XXY Klinefelter patients and healthy females (only heterozygous persons) were seen (Fig. 3). Skewing of AR allele inactivation is demonstrated in Fig. 4. Seven of 10 XX male patients exhibited extremely skewed X chromosome inactivation ratios (<20 or >80%) with no preference toward the short and/or long AR-allele. These seven patients were the four youngest and the three oldest XX males in our group. Three had highly skewed X-inactivation ratios of 99:1 and 2:98 (expressed arbitrarily as a ratio of the smaller to larger allele).
|
|
Nonparametrical correlations according to Spearman showed a relationship of the CAG repeat length of the long allele of the AR to SHBG levels (rs = 0.84, P = 0.004) and FSH concentrations (rs = 0.64, P = 0.04) (Fig. 5). LH levels did not exhibit a significant association (rs = –0.18, P = 0.62). In addition, an inverse association of CAG repeat length to concentrations of free testosterone (rs = –0.79, P = 0.01), bitesticular volume (rs = –0.64, P = 0.03), hemoglobin levels (rs = –0.72, P = 0.02), hematocrit (rs = –0.79, P = 0.007) as well as mean corpuscular volume (rs = –0.75, P = 0.01) and mean hemoglobin content (rs = –0.66, P = 0.03) of erythrocytes was seen. Other associations were not significant. The testosterone-exposed individual was excluded from this analysis.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
LH levels in XX males seem to be higher than in Klinefelter patients (Table 4). It is unclear whether Leydig cell function in these patients differs from that in Klinefelter patients, which may depend on the epigenetic (imprinting related) environment in which the SRY gene is expressed because it is abnormally located on one of the X-chromosomes. An influence of sex chromosome constitution on the genomic imprinting of germ cells has been previously demonstrated (21). This may indicate a malfunctioning signal transduction of LH, which can be due either to improper LH signaling mediated through the LH receptors or to altered Leydig cell function itself.
Usually, male hypogonadism present during the period of bone growth will result in increased body height caused by retardation of androgen-induced closure of epiphyses (19). In addition, it has been suggested that the Y chromosome contains genes that control height (2). The potential Y chromosomal growth-control gene is putatively located on the long arm of the Y chromosome. Men with short stature and infertility having deletions of the long arm of the Y chromosome were previously described (22). However, possible candidate genes are not yet identified (23). It is also known that the height of 47,XYY patients is above average (24). Our findings support the existence of a Y-linked growth gene because 46,XX males are significantly smaller than 47,XXY Klinefelter men, although they have similarly low testosterone concentrations (Table 4).
To support this concept, we extracted further information on 54 age-matched untreated men with idiopathic hypogonadotropic hypogonadism or Kallmann syndrome from our database in a post hoc approach. Their mean baseline total testosterone levels were markedly lower than in all other male groups (see Results). However, their height was comparable with Klinefelter patients and normal men but significantly different from 46,XX males (see Results). This finding further contributes to the concept of growth being strongly influenced by an Y-chromosomal factor and the lesser role of testosterone in the control of growth. Alternatively, chromosomal abnormalities in XX males may cause disturbances in expression of the SHOX (short stature homebox) gene, located in the pseudoautosomal region of the X chromosome.
Other studies showed that classical XX males have normal-for-age testosterone levels during puberty but higher frequency of hypergonadotropic hypogonadism in adulthood (20). Our results confirm this observation. The two youngest XX males are eugonadal for their age (14 and 15 yr). Klinefelter patients can have normal serum testosterone levels according to age during puberty and early adulthood (19).
We found the incidence of maldescended testes to be significantly higher in XX males than the comparison groups, also regarding Klinefelter patients. Signs of hypoandrogenism were frequent (gynecomastia, anemia) as well (see Results).
Biochemical findings
The low transaminase levels in XX males represent an interesting observation. Concentrations of GOT as well as GPT were found to be significantly lower than in normal male controls but not in Klinefelter patients. This may be due to a more frequent state of hypogonadism in both patient groups. The values rather resemble those normally found in women (GOT and GPT normal ranges for females: 10–35 U/liter; for males: 10–50 U/liter) (25). In our female cohort, liver enzymes were not assessed. Moreover, MCHC measures the concentration of hemoglobin in an average red blood cell. Low levels of hemoglobin and MCHC in hypogonadal XX males argue for anemia due to androgen deficiency.
Reproductive function
Azoospermia was found in all our XX male patients and is probably caused by the absence of the AZF region of the Y chromosome. If that is the case, it is unlikely that residual spermatogenesis would be present in the testes (26) and testicular sperm extraction attempts would remain futile. A complete Sertoli cell-only syndrome and hyperplasia of the Leydig cells have been found in other studies after histological assessment of testicular tissue of XX males (3). Histological analysis of testes in sex-reversed transgenic (XX/SRY) mice showed no spermatogenesis cells beyond the preleptotene spermatogonial stage (27). It has been demonstrated that the testicular environment of XX (SRY+) male mice is defective in supporting the later phases of spermatogenesis. The SRY gene may be insufficient to induce adequate function of Sertoli cells in XX (SRY+) mice. These studies suggest that Y chromosomal genes such as the AZF are not only crucial for spermatogenesis but also pivotal for the maintenance of normal somatic cell function.
Molecular genetic findings
XX males present a unique experiment by nature, which clearly delineates the impact of the lacking Y chromosomal regions on somatic and germ cell functions beside the SRY region present in these males. The SRY gene plays an essential role in Sertoli cell differentiation and testis development (28). All our XX male patients were SRY positive and showed various degrees of hypogonadism. Variability of sexual phenotype in 46,XX (SRY+) nonmosaic patients has been investigated in other studies (3, 13, 29, 30), which offer at least two explanations.
First, a spreading of X inactivation into the translocated Yp segment carrying the SRY gene can account for incomplete masculinization (31). Second, a disruption of normal SRY expression occurring independently of X inactivation can be responsible (29). Earlier studies (17, 32) showed the X chromosome inactivation mechanism in other individuals carrying more than one X chromosome (Klinefelter patients and women).
In most cases the distribution of X-inactivation patterns in phenotypically unaffected females is random with a 50:50 X-inactivation ratio. Thus, there is no difference of the X-weighted biallelic mean and the arithmetic mean of CAGn length of the AR gene in heterozygous females. It is suggested that the proportion of skewed X chromosome inactivation is higher in older females (32).
The distribution of X-inactivation patterns in nonmosaic 47,XXY Klinefelter patients differs from that in women. A nonrandom inactivation of the AR alleles was found in the heterozygous Klinefelter patients. A preference of the longer allele to be active has been described in these patients (17). Other studies show a random X inactivation in most Klinefelter patients (33) or significant differences in the phenotypic features between subjects with or without skewed X inactivation (34). However, these studies involve small numbers of very young Klinefelter patients and are not comparable with our previous results.
Thus, we demonstrate that XX males differ epigenetically from both other groups of subjects with two X chromosomes: Klinefelter patients and healthy women. Two thirds of XX males in our group (70%) had nonrandom X chromosome inactivation ratios. Two patients had highly skewed ratios of 2:98 and 99:1, with no preference toward the longer or shorter AR-allele. Highly skewed inactivation patterns occur more often in carriers of several X-linked disorders (32, 35). Our patients exhibiting skewed inactivation ratios were not remarkable for special diseases. A reason for the skewed X chromosome inactivation in these patients can be an X chromosome abnormality, namely the presence of the translocated SRY gene. Previous studies have shown preferential inactivation of the Y-bearing X chromosome in 46,XX (SRY+) true hermaphrodites (31, 36). X chromosome inactivation patterns in classical XX males have been shown in earlier studies as variable, but it seems to be nonrandom in most cases (20). Putatively, due to the skewed X inactivation, the SRY-bearing chromosome remains active, which can explain the male phenotype of the 46,XX patients in our study. SRY-negative 46,XX individuals may exhibit a completely different phenotype (11, 12).
Because in most cases cytogenetic and molecular genetic analyses have been performed on peripheral blood lymphocytes, the findings cannot be indicative for other tissues. Cytogenetic analysis of other tissue samples (e.g. testes) would be desirable. Whereas this is conjectural, an alternative view might be mentioned: it is possible that the abnormal genetic setting in 47,XXY Klinefelter men and 46,XX males influences the selection of clonal bone marrow cells. The significance of the relation of the longer AR allele to serum parameters (see Results), but not the X-weighted mean of CAG repeat length in XX males, points toward this hypothesis (for review also see Ref. 37).
It is known that the majority of patients with Klinefelter syndrome remain undiagnosed in the community. Presumably this could apply to a marked proportion of the 46, XX males as well. Thus, it is uncertain that this referral clinic-defined group of 11 men with 46,XX is representative of all those with this condition in the wider community. So it is possible, for example, that there is a subgroup of 46,XX men who are less hypogonadal or even possibly normal. This also leads to the possibility of a fertility potential if the germinal cell depletion is not complete until late puberty as is often seen in Klinefelter patients.
In conclusion, individuals with classical 46,XX (SRY+) male syndrome are clinically different from Klinefelter patients, especially in terms of body height and presence of maldescended testes. They tend to exhibit an extremely skewed inactivation of AR alleles, a fact not observed in Klinefelter patients or healthy women. In various target tissues, androgen activity seems to be modulated by the AR CAG repeat polymorphism.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Information: E.V., M.Z., J.G., A.N.S., and E.N. have nothing to declare.
First Published Online June 19, 2007
Abbreviations: AR, Androgen receptor; BMI, body mass index; GOT, glutamic-oxaloacetic transaminase; GPT, glutamate-pyruvate transaminase; MCHC, mean corpuscular hemoglobin concentration; PSA, prostate-specific antigen.
Received February 27, 2007.
Accepted June 11, 2007.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Foresta, D. Zuccarello, A. Garolla, and A. Ferlin Role of Hormones, Genes, and Environment in Human Cryptorchidism Endocr. Rev., August 1, 2008; 29(5): 560 - 580. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Aksglaede, R. B. Jensen, E. Carlsen, P. Kok, D. M Keenan, J. Veldhuis, N. E Skakkebaek, and A. Juul Increased basal and pulsatile secretion of FSH and LH in young men with 47,XXY or 46,XX karyotypes. Eur. J. Endocrinol., June 1, 2008; 158(6): 803 - 810. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Aksglaede, N. E. Skakkebaek, and A. Juul Abnormal Sex Chromosome Constitution and Longitudinal Growth: Serum Levels of Insulin-Like Growth Factor (IGF)-I, IGF Binding Protein-3, Luteinizing Hormone, and Testosterone in 109 Males with 47,XXY, 47,XYY, or Sex-Determining Region of the Y Chromosome (SRY)-Positive 46,XX Karyotypes J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 169 - 176. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
