Supported by The Frederick J. and Theresa Dow Wallace Fund of the New York Community Trust and the Robert Dow Foundation.
Methylation-Specific PCR Allows for Fast Diagnosis of X Chromosome Disomy and Reveals Skewed Inactivation of the X Chromosome in Men With Klinefelter Syndrome
Version of Record online: 2 JAN 2013
2012 American Society of Andrology
Journal of Andrology
Volume 33, Issue 5, pages 955–962, September-October 2012
How to Cite
Mehta, A., Malek-Jones, M., Bolyakov, A., Mielnik, A., Schlegel, P. N. and Paduch, D. A. (2012), Methylation-Specific PCR Allows for Fast Diagnosis of X Chromosome Disomy and Reveals Skewed Inactivation of the X Chromosome in Men With Klinefelter Syndrome. Journal of Andrology, 33: 955–962. doi: 10.2164/jandrol.111.016030
- Issue online: 2 JAN 2013
- Version of Record online: 2 JAN 2013
- Received for publication 30 November 2011; accepted for publication 14 March 2012
- Polymerase chain reaction;
- prenatal genetic diagnosis
Abstract: Klinefelter syndrome (KS) remains the most common, yet often undiagnosed, chromosomal aberration in men. Early diagnosis and treatment can improve the health of patients with KS. The aim of this study was to evaluate the inactivation pattern of supernumerary X chromosomes. The secondary aim was to design a reliable and cost-effective molecular test for detection of X chromosome disomy. Methylation-specific polymerase chain reaction (M-PCR), with primers for familial mental retardation (FMR1) and X chromosome inactive-specific transcript (XIST) genes, was used to detect the presence of X chromosome disomy in men. Seventeen fertile males, 12 females, and 35 males with KS (28 with 47,XXY karyotype, and 7 with 47,XXY/46,XY mosaics) were included in the study. Results of the karyotype were compared with the results of semiquantitative M-PCR. Inactivation of X chromosomes was measured by XIST/FMR-1 methylation ratio. Differences in the methylation patterns of FMR1 and XIST genes between 46,XY men and men with X chromosome disomy allowed for rapid detection of the presence of an additional X chromosome, achieving 100% sensitivity and specificity using M-PCR. The methylated:unmethylated FMR1 amplicon ratio allowed the detection of 1 additional X chromosome per 100 normal XY cells (1% of XX/XY mosaicism). In our series, 50% of 47,XXY men showed skewed inactivation of the X chromosome. Men with KS can have incomplete inactivation of supernumerary X chromosomes. M-PCR is a sensitive, specific, fast, and relatively inexpensive test for the diagnosis of X chromosome disomy.
Klinefelter syndrome (KS) is the most common numerical chromosomal aberration in men, with an estimated frequency of 1:500 to 1:1000 live deliveries (Nielsen and Wohlert, 1990). It is characterized by X chromosome polysomy, with X disomy being the most common variant (47,XXY). Ninety percent of men with KS have nonmosaic X chromosome polysomy (Lanfranco et al, 2004). Nevertheless, there is considerable phenotypic variation among men with KS, with the classic findings of eunuchoid body proportions, sparse facial and pubic hair, small and firm testicles, and severe intellectual deficits being less common than previously described. Possible explanations for this phenotypic variation include differences in hormonal profiles, differences in genetic background, and abnormal inactivation of supernumerary X chromosomes.
It has been noted that men with more than 2 X chromosomes (48,XXXY; 49,XXXXY) are more severely affected than men with the classic 47,XXY karyotype (Paduch et al, 2008). The presence of 2 active X chromosomes (X ch) in animals and hybridoma models is lethal (Heard and Disteche, 2006), and inactivation of 1 X ch is critical to achieve normal development (Nguyen and Disteche, 2006). In normal females, 1 X ch randomly undergoes inactivation in embryonic tissues; it is only logical that a similar mechanism also occurs in men with X ch polysomy. The X ch bears more than 1100 genes critical for normal function of the brain and testes (Ross et al, 2005). Overexpression of X ch—linked genes could be responsible for cognitive impairment and spermatogenic failure seen in 47,XXY men (Vawter et al, 2007).
Inactivation of the supernumerary X ch is initiated within the X chromosome inactivation center (XIC), by activation of the X chromosome inactive-specific transcript (XIST) promoter (Hong et al, 2000). Transcription of XIST RNA allows for multifocal painting of the X ch and subsequent recruitment of inactivation proteins with H3 and H4 deacetylation and methylation, linking the expression of XIST to chromatin remodeling and gene silencing (Matarazzo et al, 2008). Because of this multistep mechanism of X ch inactivation and the escape of some genes from X ch inactivation, it is possible that abnormal or skewed inactivation of supernumerary X chromosomes in men with X ch polysomy leads to the cognitive and reproductive problems seen in this group.
Application of methylation-specific polymerase chain reaction (M-PCR) in the detection of X ch disomy is based on the differences in methylation of the familial mental retardation (FMR1) gene between females and males and the evolutionary principle of dose compensation, which equalizes phenotypic expression of characteristics determined by X chromosome genes. To avoid excess FMR1 protein (mental retardation), only 1 copy of the FMR1 promoter is unmethylated and transcriptionally active (FMR1-UM) at any time; in females, the second copy is inactivated through methylation of CpG islands (FMR1-M). Although the mechanism is poorly understood, it is believed that genes located in XIC are responsible for the detection of an additional X chromosome and for XIST transcription. In turn, XIST binds to its specific site on supernumerary X ch and turns off gene transcription (Escamilla-Del-Arenal et al, 2011; Heard and Turner, 2011; Werler et al, 2011). In normal males, XIST is methylated and, therefore, transcriptionally inactive; in females, 1 copy of XIST is methylated (inactive) and the other is not (active). The XIST gene, therefore, has the opposite pattern of methylation to that of FMR1. Females and KS males are expected to have similar inactivation patterns of XIST and FMR1 genes and, thus, the same amplification pattern on gel electrophoresis after M-PCR (Figure 1).
Screening for KS in target populations is limited by the cost of established tests for KS diagnosis, such as karyotype analysis and quantitative real-time polymerase chain reaction. Recently, Barr body cytology has been proposed as a cheaper test for KS screening, but although the test has 95% specificity, its sensitivity is limited to 82% (Kamischke et al, 2003).
Pena first proposed use of FMR1 gene analysis in the diagnosis of KS in a letter to the Journal of Andrology in 2003 (Pena and Sturzeneker, 2003). However, to date, there has been no previous publication evaluating this technique in the setting of KS. The objective of this study was, therefore, 2-fold. Using differences in the methylation pattern of 2 genes located on X chromosome, FMR1 and XIST, the primary aim was to evaluate the inactivation pattern of supernumerary X chromosomes in men with KS. The secondary aim was to develop a cost-effective, rapid, and reliable method of KS diagnosis based on the molecular mechanism of X ch inactivation.
Materials and Methods
This retrospective study was based on an institutional review board-approved deoxyribonucleic acid (DNA) repository. Electronic medical records of patients who had their DNA stored in the repository were reviewed, and patients with a known karyotype were identified. Seven 47,XXY/46,XY males and twenty-eight 47,XXY males were included in this study. Karyotype analysis and assays of serum testosterone, folliclestimulating hormone (FSH), luteinizing hormone (LH), and estradiol levels were performed by using a commercial clinical laboratory (ARUP Labs, Salt Lake City, Utah). Testicular volume was measured by a single attending physician using a Prader orchidometer. A DNA sample from all subjects in this project was extracted using a Promega Wizard Genomic DNA purification kit (Promega, Madison, Wisconsin) according to the manufacturer's guidelines and stored at 220°C. Cost analysis was performed using the purchase price for each kit and chemical obtained.
Y Chromosome Detection
After DNA extraction and purification, presence of the Y chromosome was confirmed by PCR with primers specific for short (SY14; UniSTS: 42547) and long arms (SY88; UniSTS: 80503) of the Y chromosome. Genomic template DNA (150 ng) was added to the PCR mixture containing 0.2 μM SY14 primer, 0.32 μM SY88 primer, 12.5 μL of Qiagen Multiplex PCR master mix (Qiagen, Valencia, California), and water for a total of 25 μL of reaction mixture. Fragments were amplified in a multiplex PCR reaction using GeneAmp PCR System 9700 (Applied Biosystems, Carlsbad, California). The presence of PCR product was verified by electrophoresis of the PCR product on 4% agarose gel stained with ethidium bromide.
DNA was deaminated using a modification of the methods described by Herman et al (1996). Human DNA (0.5 μg) was suspended in 50 μL of double-distilled water and denatured at 37°C for 17 minutes after adding 5.5 μL of 2 M NaOH. Freshly prepared 10 mM hydroquinone and 3 M bisulfite (Sigma-Aldrich, St Louis, Missouri) was added to denatured DNA, and the mixture was placed in a water bath (55°C) for 1.5 hours under PCR-grade oil (Perkin Elmer, Waltham, Massachusetts). The time of denaturation was experimentally derived to allow for a conversion rate of 99%. Deamination was terminated by adding 150 μLof96% ethanol/3 M NaOH mixture during DNA purification with Qiagen QIAquick PCR Purification Kit (Qiagen, Valencia, California). DNA was eluted with 100 μL of Tris-EDTA buffer and stored at 220°C.
Two sets of primers (methylated and unmethylated) for each DNA template sequence were used for M-PCR (Table 1; Figure 2). Four microliters of template DNA (100–200 ng) was mixed with 2 μL of primer mix and 19 μL of PCR master mix. Master mix was prepared with 2.5 μL of 10× Roche reaction buffer with magnesium (Roche, Basel, Switzerland), 0.2 μL of Stratagene 200 mM dNTP mix (Agilent, Santa Clara, California), 0.25 μL of FastStart High Fidelity Polymerase (Roche), and 16.05 μL of water per each 25-μL PCR reaction. Polymerase chain reaction amplification was performed with GeneAmp PCR system 9700 thermal cycler: 95°C for 3 minutes × 1, (95°C for 30 seconds/61°C for 40 seconds/72°C for 50 seconds) × 45 cycles, 72°C for 7 minutes × 1, and 4°C for 20 minutes. The PCR product was mixed with Blue Juice gel loading buffer (Invitrogen, Carlsbad, California) and loaded on 4% agarose gel stained with ethidium bromide. DNA molecular weight marker VIII (Roche) was used as a size standard.
|Primer||Sequence of Primer||Concentration, μM||Position||GenBank|
Sensitivity and Specificity of M-PCR in Detection of X Chromosome Disomy
After optimizing the M-PCR and primer concentrations on 2 male and 2 female controls, M-PCR was performed on 17 fertile male controls, 12 female controls, and 35 males with KS, 28 of whom were 47,XXY and 7 were 47,XXY/46,XY mosaics, verified by karyotype. The primers and PCR settings used are described earlier. The specificity of primers was evaluated with native template DNA (not deaminated) using the same M-PCR setting as for deaminated DNA. All experiments were repeated 3 times.
To determine the lowest percentage of XX/X mosaicism detectable by M-PCR, the DNA from fertile male and female was mixed to achieve XX/X ratios of 50%,25%, 12.5%,8%, 5%,3%,2%,1%, and 0.5%. Subsequently prepared samples were deaminated, and M-PCR was performed as described earlier. Each experiment was repeated 3 times to ensure reproducibility. Net optical densities of FMR1-UM and FMR1-M bands were measured using Kodak 1D Image Analysis Software v. 2.02 for Windows (Kodak-Eastman Inc, Rochester, New York).
Pattern of X Chromosome Inactivation
To test the hypothesis that men with Klinefelter syndrome have skewed inactivation of the X chromosome, the FMR1-UM/methylated XIST (XIST-M) ratio was measured in female controls and men with Klinefelter syndrome. Theoretically, the ratio of the unmethylated gene of interest and methylated XIST gene in females should be 1:1 to allow for transcription of material from active X ch. If men with KS have skewed inactivation of the X chromosome, or if the inactivation is promiscuous, then a different ratio of the unmethylated gene of interest to methylated XIST would be expected (Figure 1).
Statistical analysis was performed using GraphPad Prism software (GraphPad Software, La Jolla, California). One-way analysis of variance was used to compare the difference in means between the FMR1-UM/XIST-M inactivation ratios in men with KS and female controls. Because the inactivation ratio among men with KS had a bimodal distribution, 3 patient groups (2 groups of men with KS and 1 group of female controls) were used for the final analysis. Tukey's multiple comparison test was used to test for differences in means between groups. A P ≤ .05 was considered statistically significant.
On average, the 47,XXY and 47,XXY/46,XY men in the study population had low serum testosterone (252.22 ± 115.8 ng/dL) with elevated serum FSH (32.96 ± 12.43 IU/L) and LH (18.84 ± 9.01 IU/L) concentrations (Table 2). The average left and right testicular volume was also low, at 3.68 ± 3.11 mL and 4.12 ± 3.3 mL, respectively. The clinical and laboratory profile for these men was typical for KS. The presence of a Y chromosome was positively verified for all 35 men with KS by multiplex PCR using STS markers for both the short and long arm of the Y chromosome.
|Serum Hormone||Average Level ± SD||Reference Range|
|Total testosterone, ng/dL||252.22 ± 115.82||270–1730|
|FSH, IU/L||32.96 ± 12.43||0.4–8.0|
|LH, IU/L||18.84 ± 9.01||2.0–12.0|
|Estradiol, ng/L||23.85 ± 11.16||<60|
|T/E ratio||11.9 ± 6.1||>20|
All 46,XY men had 2 amplicon bands on gel electrophoresis: 1 unmethylated FMR1 promoter amplicon (FMR1-UM) and 1 band corresponding to the methylated XIST gene (XIST-M). All females and 47,XXY males had 4 bands on the gel, reflecting the presence of the following amplicons—methylated and unmethylated FMR1 promoter (FMR1-UM, FMR1-M) and methylated and unmethylated XIST gene (XIST-M, XIST-UM)—thus verifying the hypothesis that M-PCR can detect the presence of X chromosome polysomy (Figure 3; Table 3). The 47,XXY/46,XY males all tested positive for X ch disomy using the M-PCR assay (Table 3). To exclude nonspecific amplification of methylation-specific primers, parallel multiplex PCR with native and deaminated DNA was performed. Although the bands were always visualized using the deaminated DNA, there were no bands using native DNA from the same patients. Analysis of the inactivation pattern among females and men with KS verified that the additional X ch in men undergoes inactivation (based on the methylation pattern of FMR1 and XIST genes).
When karyotype results were compared with those of M-PCR, 100% specificity and 100% sensitivity were achieved in detecting X chromosome disomy in males with nonmosaic KS. M-PCR was able to detect 1% or more of 47,XXY/46,XY mosaicism (Figure 4).
The analysis of X ch inactivation patterns in females and KS males revealed skewed inactivation of the X ch in men with KS. In fertile females, the inactivation ratio (FMR-UM/XIST-M) was 1.0, as expected, whereas in men with the 47,XXY karyotype, inactivation was skewed toward inadequate inactivation of genetic material on the supernumerary X chromosome. The inactivation pattern in men with KS was bimodal, with a ratio of 1.8 as a cutoff point between groups. Fifty percent of men with KS had highly skewed inactivation with a ratio greater than 1.9. There was no statistically significant difference in serum hormone profiles between men with normal and highly skewed X ch inactivation (Figure 5). There was a statistically significant difference in terms of the FMR1/XIST ratio between females and KS males, with an inactivation ratio greater than 1.8, but not between females and KS males with an inactivation ration less than 1.8 (Figure 5).
Material costs to detect the presence of additional X chromosomes in males using methylation-specific PCR were estimated at $5.49 per patient blood sample. The turnaround time for M-PCR was less than 48 hours from the time of sample receipt to reporting of results.
Although KS is the most common sex chromosomal abnormality in men, most patients remain undiagnosed during their lifetime (Bojesen et al, 2003). The results of this study demonstrate that M-PCR can be used to assess the inactivation pattern of supernumerary X chromosomes in men with KS. The high sensitivity and specificity of M-PCR allow for detection of as little as 1% XX/XY mosaicism. M-PCR is fast and cost-effective, which makes it a valuable tool for the early diagnosis of KS. Our study also showed that X ch inactivation in men with KS is skewed, compared with normal women. This escape from inactivation might contribute to the phenotypic variation seen in affected individuals (Iitsuka et al, 2001).
KS is characterized by low serum testosterone, elevated serum LH and FSH, and infertility from primary testicular failure. Many patients also demonstrate some cognitive impairment and have a higher risk of developing medical comorbidities, such as autoimmune diseases, diabetes mellitus, osteoporosis, tumors, and increased mortality. This clinical phenotype is not typically present in individuals with hypogonadism due to Kallman syndrome, suggesting that abnormal function of X-linked genes, rather than low serum testosterone per se, is responsible for the phenotypic presentation of KS.
The use of M-PCR has been previously described in the diagnosis of fragile X syndrome (Coffee et al, 2009) and KS (Werler et al, 2011). In the present study, differences in the methylation patterns of FMR1 and XIST genes between 46,XY men and men with mosaic or nonmosaic X chromosome disomy allowed for fast detection of the presence of an additional X chromosome, achieving 100% sensitivity and specificity using M-PCR. These results are similar to findings of Coffee et al (2009), who also reported M-PCR as having 100% sensitivity and specificity for detecting methylation of the FMR1 gene. High sensitivity and reliability are clear advantages of the use of M-PCR as a screening tool.
Conventional cytogenetics, or karyotype analysis, is the current gold standard for the diagnosis of KS. One of the shortcomings of this approach is its low sensitivity for the detection of mosaicism (Okada et al, 2001). Our results show that M-PCR is able to detect very low levels of mosaicism. M-PCR could therefore be useful in verification of mosaicism in cases where mosaicism is clinically suspected but karyotype results are negative.
Established tests for the diagnosis of KS, such as karyotype analysis and real-time quantitative PCR, are expensive. Barr body cytology, although cheaper, is limited by sensitivity and specificity of 95% and 82%, respectively (Kamischke et al, 2003). Fluorescence in situ hybridization (FISH) has similar specificity and sensitivity to karyotype analysis but requires expensive probes, an experienced technologist, and imaging software. The average turnaround time for M-PCR in this study from the receipt of samples to availability of results was less than 48 hours, at a cost of less than $6 per sample. Molecular techniques have long been used in the preimplantation genetic diagnosis of various heritable diseases (Sermon et al, 2004). Advantages of PCR-based technology include familiarity with the technology, availability of equipment in virtually every research and clinical laboratory, low volume of blood needed for the DNA extraction, and low cost. Recent literature on prenatal screening for common aneuploidies (13, 18, 21, and sex chromosomes) confirms that molecular techniques are not only as sensitive as karyotype and FISH but also more economical and efficient (Donaghue et al, 2003; Mann et al, 2004).
Although the benefit of systematic screening for KS has been debated in the literature (Herlihy et al, 2011a), growing evidence suggests that early diagnosis and therapeutic interventions in children with KS could have a beneficial effect on their physical, academic, and social development, as well as their overall health (Simpson et al, 2003; Herlihy et al, 2011b; Samango-Sprouse et al, 2011). Unfortunately, only 10% of men affected by KS are diagnosed during preadolescence and adolescence, the time when treatment can be the most effective (Bojesen et al, 2003). Availability of a rapid and a cost-effective postnatal screening test for KS can be expected to improve significantly the diagnosis and follow-up of individuals with KS.
The results of this study confirm that the supernumerary X ch undergoes inactivation in men with KS. Nevertheless, the presence of supernumerary X chromosomes results in KS, indicating that the process of X ch inactivation is skewed, or that there is selective allelic drop-out from inactivation of certain X ch genes. The regulation of X ch inactivation in men with KS is certainly less stringent than in female controls because the ratio of FMR1-UM/XIST-M was statistically and clinically significantly different between the 2 groups (1.8 vs 1.0, respectively). A high ratio of FMR1-UM/XIST-M indicates that in approximately 50% of 47, XXY men, genetic material may escape methylation and, thus, inactivation. Some degree of skewed X ch inactivation is known to occur in normal females. Genes such as ZFK, which code for proteins involved in sperm and oocyte development and have alleles on both X and Y chromosomes, are normally “active” on both the active and inactive X ch in females. Heard and Disteche (2006) have estimated that 15% of X ch genes escape inactivation in normal females, and more often in cancer. Thus, it is possible that in men with KS, some genes on additional X chromosomes escape epigenetic regulation and inactivation. The phenotypic variation in men with KS could be explained by the extent to which genes escape inactivation.
Many genes on the X ch are highly expressed in testis, ovaries, and brain; thus, it is not surprising that abnormalities seen in men with KS affect brain and testis function (Wilda et al, 2000; Zechner et al, 2001; Khil et al, 2004). Although the mechanism of gene up-regulation in mammalian or human brain are not yet clear, it is believed that methylation of specific amino acids on H3 and H4, with subsequent chromatin changes, is critical to normal control of gene expression (Akhtar, 2003). Werler et al (2011) have studied the expression of selected X ch genes known to escape X ch inactivation in the mouse model and shown altered expression of these genes in male XXY mice compared with male XY mice and female XX controls. It is plausible that a similar phenomenon exists in humans. Thus, abnormal inactivation of supernumerary X chromosomes could help explain the cognitive and physical variations seen in individuals with KS.
One limitation of this study is that skewed inactivation of the supernumerary X ch in men with KS was demonstrated with the use of semiquantitative PCR only. It is important to confirm this finding with real-time quantitative PCR and multiple clinically significant X ch—specific genes. On the basis of the present results, a more detailed analysis using additional markers has already been initiated by the authors.
The question remains as to which genes inactivated on the X chromosome affect the phenotypic variation seen in men with KS. It is also important to consider that the inactivation statusinthisstudy wasevaluated in the patients' peripheral leukocytes and might not reflect tissue-specific inactivation patterns. Further studies using M-PCR and DNA from specific tissues, such as the testis, might be helpful in addressing this question. Although the optimal diagnostic protocol and indications of screening for KS need to be deduced from further work, M-PCR appears to be a cost-effective and accurate test to diagnose KS in men. This method could have other potential applications in prenatal genetic diagnosis and forensic medicine.
- Dosage compensation: an intertwined world of RNA and chromatin remodelling. Curr Opin Genet De. 2003; 13: 161–169.
- Prenatal and postnatal prevalence of Klinefelter syndrome: a national registry study. J Clin Endocrinol Meta. 2003; 88: 622–626. , ,
- Incidence of fragile X syndrome by newborn screening for methylated FMR1 DNA. Am J Hum Gene. 2009; 85: 503–514. , , , , , ,
- Development and targeted application of a rapid QF-PCR test for sex chromosome imbalance. Prenat Diag. 2003; 23: 201–210. , , ,
- Evolutionary diversity and developmental regulation of X-chromosome inactivation. Hum Gene. 2011; 130: 307–327. , ,
- Dosage compensation in mammals: finetuning the expression of the X chromosome. Genes De. 2006; 20: 1848–1867. ,
- Function of the sex chromosomes in mammalian fertility. Cold Spring Harbor Perspect Bio. 2011; 3: a002675. ,
- Postnatal screening for Klinefelter syndrome: is there a rationale. Acta Paediat. 2011a; 100: 923–933. , , ,
- The psychosocial impact of Klinefelter syndrome and factors influencing quality of life. Genetics in medicine. Off J Am Coll Med Gene. 2011b; 13: 632–642. , , , , ,
- Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci U S. 1996; 93: 9821–9826. , , , ,
- A revision of the human XIST gene organization and structural comparison with mouse Xist. Mammalian genome. Off J Int Mamm Genome So. 2000; 11: 220–224. , ,
- Evidence of skewed X-chromosome inactivation in 47, XXY and 48, XXYY Klinefelter patients. Am J Med Gene. 2001; 98: 25–31. , , , , ,
- Clinical and diagnostic features of patients with suspected Klinefelter syndrome. J Andro. 2003; 24: 41–48. , , ,
- The mouse X chromosome is enriched for sex-biased genes not subject to selection by meiotic sex chromosome inactivation. Nat Gene. 2004; 36: 642–646. , , ,
- Klinefelter's syndrome. Lance. 2004; 364: 273–283. , , ,
- Selective loss of imprinting in the placenta following preimplantation development in culture. Developmen. 2004; 131: 3727–3735. , , , , , ,
- Building up the inactive X chromosome. Biol Cell (Paris. 2008; 100: 63–70. , ,
- Dosage compensation of the active X chromosome in mammals. Nat Gene. 2006; 38: 47–53. ,
- Sex chromosome abnormalities found among 34,910 newborn children: results from a 13-year incidence study in Arhus, Denmark. Birth Defects Orig Artic Se. 1990; 26: 209–223. ,
- Fluorescence in situ hybridization analysis of sex-chromosome mosaicism in azoospermic men. J Andro. 2001; 22: 970–972. , , , , ,
- New concepts in Klinefelter syndrome. Curr Opin Uro. 2008; 18: 621–627. , , ,
- Molecular barr bodies: methylation-specific PCR of the human X-linked gene FMR-1 for diagnosis of Klinefelter syndrome. J Andro. 2003; 24: 809; author reply 810. ,
- The DNA sequence of the human X chromosome. Natur. 2005; 434: 325–337. , , , , , et al.
- Effects of short-course androgen therapy on the neurodevelopmental profile of infants and children with 49, XXXXY syndrome. Acta Paediat. 2011; 100: 861–865. , , , , ,
- Preimplantation genetic diagnosis. Lance. 2004; 363: 1633–1641. , ,
- Klinefelter syndrome: expanding the phenotype and identifying new research directions. Genet Me. 2003; 5: 460–468. , , , , , , , , , , , , , , , , , , ,
- Dysregulation of X-linked gene expression in Klinefelter's syndrome and association with verbal cognition. Am J Med Genet B Neuropsychiatr Gene. 2007; 144B: 728–734. , ,
- Expression of selected genes escaping from X inactivation in the 41, XX(Y)* mouse model for Klinefelter's syndrome. Acta Paediat. 2011; 100: 885–891. , , ,
- Do the constraints of human speciation cause expression of the same set of genes in brain, testis, and placenta. Cytogenet Cell Gene. 2000; 91: 300–302. , , , , ,
- A high density of X-linked genes for general cognitive ability: a run-away process shaping human evolution. Trends Gene. 2001; 17: 697–701. , , , , ,