Differential methylation of the X-chromosome is a possible source of discordance for bipolar disorder female monozygotic twins.код для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 147B:459 –462 (2008) Differential Methylation of the X-Chromosome is a Possible Source of Discordance for Bipolar Disorder Female Monozygotic Twins Araceli Rosa,1,2 Marco M. Picchioni,1 Sridevi Kalidindi,1 Caroline S. Loat,2 Joanne Knight,2 Timothea Toulopoulou,1 Ronald Vonk,3 Astrid C. van der Schot,4 Willem Nolen,5 René S. Kahn,4 Peter McGuffin,2 Robin M. Murray,1 and Ian W. Craig2* 1 Division of Psychological Medicine, Institute of Psychiatry, King’s College London, London, UK MRC Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, King’s College London, London, UK 3 Reinier van Arkel groep, DZ’s-Hertogenbosch, The Netherlands 4 Department of Psychiatry, University Medical Centre Utrecht, Utrecht, The Netherlands 5 University Medical Center Groningen, Groningen, The Netherlands 2 Monozygotic (MZ) twins may be subject to epigenetic modifications that could result in different patterns of gene expression. Several lines of evidence suggest that epigenetic factors may underlie mental disorders such as bipolar disorder (BD) and schizophrenia (SZ). One important epigenetic modification, of relevance to female MZ twins, is X-chromosome inactivation. Some MZ female twin pairs are discordant for monogenic X linked disorders because of differential X inactivation. We postulated that similar mechanisms may also occur in disorders with more complex inheritance including BD and SZ. Examination of X-chromosome inactivation patterns in DNA samples from blood and/or buccal swabs in a series of 63 female MZ twin pairs concordant or discordant for BD or SZ and healthy MZ controls suggests a potential contribution from X-linked loci to discordance within twin pairs for BD but is inconclusive for SZ. Discordant female bipolar twins showed greater differences in the methylation of the maternal and paternal X alleles than concordant twin pairs and suggest that differential skewing of X-chromosome inactivation may contribute to the discordance observed for bipolar disorder in female MZ twin pairs and the potential involvement of X-linked loci in the disorder. ß 2007 Wiley-Liss, Inc. KEY WORDS: X-linkage; bipolar disorder; Xchromosome inactivation; twins; psychosis Please cite this article as follows: Rosa A, Picchioni MM, Kalidindi S, Loat CS, Knight J, Toulopoulou T, Vonk R, van der Schot AC, Nolen W, Kahn RS, McGuffin P, Murray RM, Craig IW. 2008. Differential Methylation of the X-Chromosome is a Possible Source of Discordance *Correspondence to: Ian W. Craig, SGDP Centre, King’s College London, Institute of Psychiatry, PO82, De Crespigny Park, London SE5 8AF, United Kingdom. E-mail: email@example.com Received 19 April 2007; Accepted 3 August 2007 DOI 10.1002/ajmg.b.30616 ß 2007 Wiley-Liss, Inc. for Bipolar Disorder Female Monozygotic Twins. Am J Med Genet Part B 147B:459–462. INTRODUCTION Bipolar disorder (BD) and schizophrenia (SZ) are debilitating mental disorders of largely unknown etiology. Evidence from twin, family, and adoption studies indicates a strong genetic predisposition to both. While the mode of transmission is poorly characterized, genetic epidemiology suggests that the two conditions share certain susceptibility genes because of their familial co-aggregation [Bramon and Sham, 2001; Cardno et al., 2002]. Comparable concordance rates among monozygotic (MZ) twins of between 41% and 65% have been reported in recent twin studies of both SZ and BD [Cardno and Gottesman, 2000; Cardno et al., 2002]. Since MZ twins share identical maternal and paternal chromosomes, the traditional explanation for phenotypic discordance within MZ twins is the influence of non-shared environmental factors. Recently, however, several authors have suggested the possible involvement of epigenetic mechanisms such as methylation of cytosines and histone modification in this phenomenon [e.g., Singh et al., 2002; Hardy, 2006]. Several lines of evidence suggest that such epigenetic factors, may influence susceptibility to BD and SZ [Petronis, 2001, 2006; Peedicayil, 2003]. Some empirical work has been performed on the genomes of MZ twins discordant for SZ which has identified epigenetic differences reflected in different DNA methylation patterns [e.g., Tsujita et al., 1998; Petronis, 2003]. One early epigenetic process affecting gene expression, which has the potential to create phenotypic differences within pairs of female MZ twins is X inactivation. To achieve dosage compensation, each cell in a female embryo randomly inactivates one X chromosome which is marked by hyper-methylation of CpG islands early in development. A tissue sample of a female therefore typically contains clones of cells that have inactivated the paternal X and other clones which have inactivated the maternal X. Although roughly equal proportions of cell types may be expected from this essentially random process, stochastic mechanisms operating on small numbers of cells may result in skewed X-chromosome inactivation patterns between the members of female monozygotic twin pairs (FMZ). This may allow them to express phenotypic discordance for traits that are influenced by polymorphic X-linked genes. This has already been demonstrated in FMZ twin pairs discordant for X-linked single gene disorders including Duchenne muscular dystrophy, X-linked immunodeficiencies, Lesch-Nyham disease and Haemophilia [see, Craig et al., 2004]. 460 Rosa et al. More recently, the potential contribution of X-linked quantitative trait loci (QTLs) to complex behaviors has been investigated by comparing correlations between female MZ twin pairs (FMZ) and male MZ twin pairs (MMZ). If polymorphic X-linked loci are implicated, FMZ should be more discordant than MMZ as a result of skewed X inactivation [Loat et al., 2004]. In their recent study, significant differences were observed for several behaviors including those relating to the development of early social skills. The suggestion of X-linkage for BD dates back to at least the 1930s and follows from evidence demonstrating an excess of females and a deficiency of male transmission for the disorder, further supported by evidence for linkage between BD and color blindness [Baron, 1977; Mendlewicz et al., 1979]. In SZ, a subgroup of familial cases could be due to a genetic defect on the X chromosome, a hypothesis supported by the observation of an excess of X-chromosome aneuploidies (XXX and XXY) among some patients with psychosis [DeLisi et al., 1994] and because of a number of differences between the sexes in rates and developmental courses of illness. Given the potential genetic overlap between BD and SZ together with the possible involvement of X-linked genes for both disorders, the main aim of our study was to explore whether, or not, differential X-chromosome inactivation among FMZ pairs might underlie the phenotypic discordance between such pairs for BD and SZ. MATERIALS AND METHODS Sample The study group consisted of monozygotic female twins from the Maudsley Twin Study of SZ and BD and the Dutch Twin study on BD. English patients were referred by their treating psychiatrist and performed structured clinical interviews using the Schedule for Affective Disorders and SZ-Lifetime version [Endicott and Spitzer, 1978], augmented with further clinical information, from which DSM IV [American Psychiatric Association, 1994] diagnoses were made. Control subjects were recruited from the Institute of Psychiatry Volunteer Twin Register and by advertisement in the national media. The Dutch Bipolar Twin pairs (n ¼ 14) were recruited via the Dutch Patient’s Association for Manic Depressives and Relatives (n ¼ 4), the ‘‘Lithium-Plus Working Group,’’ a collaborating group of psychiatrists in The Netherlands with a special interest in BD (n ¼ 2), by psychiatrists working in several Dutch psychiatric institutes (n ¼ 3) and by articles or advertisements in national and regional newspapers (n ¼ 5). Clinical diagnosis for axis I psychiatric disorders were confirmed via the structured clinical interview for DSM-IV (SCID) and also via available medical records. Diagnosis for BD in the two centers was based on similar criteria to achieve DSM-IV compatibility and included both bipolars I and II. The co-twins of the index twin for those pairs classified as discordant BD varied from having no symptoms through depression (NOS) and one case each of unipolar depression and SZ, paranoid type. None of the discordant co-twins reached the criteria for clinically diagnosed BD and as there was no simple quantitative means for scoring the differences in symptoms they were considered together as a discordant group. All subjects gave written informed consent before participating, after the study was approved by the ethical committee of both Institutions (i.e., Multi Centre Research Ethics Committee and Medical Ethical Review Board of the UMC Utrecht). DNA Extraction and Methylation Study Mouth swab and/or blood samples from the twins were collected. DNA was available for 63 twin pairs; these comprised 14 FMZ pairs discordant for BD, 9 FMZ pairs concordant for BD, 4 FMZ pairs discordant for SZ, 6 FMZ pairs concordant for SZ, and 30 FMZ pairs of healthy control twins with no personal or family history of a psychotic, SZ spectrum or mood disorder, this was ascertained using the Family Interview for Genetic Studies [Gershon and Guroff, 1984]. From this sample, DNA from both tissues was available for 3 pairs discordant for BD, 3 pairs concordant for BD, 3 pairs discordant for SZ, 3 pairs concordant for SZ, and 17 control pairs. The remainder had DNA available for one or other of the tissues, but for the same tissue for both members of the twin pair. DNA from buccal mucosa and from peripheral blood leukocytes was extracted following standard techniques for all the samples [Freeman et al., 1997, 2003]. Twin zygosity was determined using a standardized twin likeness questionnaire augmented with a multiplex zygosity test protocol based on the analysis of between 9 and 12 unlinked, highly polymorphic microsatellite loci to confirm zygosity status. The activation status of the X-chromosome was determined by a polymerase chain reaction amplification (PCR) of a region containing a targeted CpG site and a highly polymorphic simple sequence repeat (SSR) at the human androgen receptor locus (AR) [Allen et al., 1992]. This technique is based on the differential methylation at CpG islands of X-linked housekeeping genes on the active and inactive X-chromosomes. The cytosine residues of the CpG dinucleotides in such islands are methylated on the inactive X, which prevents digestion at CCGG sites recognized by the methylation sensitive restriction enzyme HpaII. Therefore, only CpG islands at these loci on the active X chromosome will be digested following exposure to this enzyme. Heterozygous females, distinguished by copy number at the SSR, will give PCR products that differ in size from maternal and paternal X-chromosome templates. Briefly, genomic DNA was split into three aliquots (150 ng each). One was incubated with HpaII—digesting unmethylated active DNA (informative digestion). The second aliquot was incubated similarly but without restriction enzyme enabling the amplification of both alleles (mock digestion). The third was incubated with MspI—a methylation insensitive restriction enzyme, that constitutes a control to verify there was not an SNP at the CCGG site (control). All the digests were amplified using primers, which flank the region comprising both the polymorphic and the restriction site and one of which was 6-FAM fluorescently-labeled (details on request). The fluorescently labeled products were analyzed on an automated DNA sequencer (Applied Biosystems (ABI), Warrington, Cheshire, UK). The peak heights were measured using Gene Scan software (Applied Biosystems). The ratio of the peak heights of the two alleles after the control digestion was used as a correction factor for any preferential amplification of one allele compared to the other as follows: ratio in mock digestion (Rm) ¼ Peak 1 height/Peak 2 height; ratio in HpaII digestion (Rh) ¼ Peak 1 height/Peak 2 height; normalized ratio (Rn) ¼ Rh/Rm; hence, Percentage of inactivation of allele 1 ¼ (Rn/Rn þ 1) 100 [Monteiro et al., 1998]. Differences between the inactivation ratios (i.e., differences in the percentages of inactivation for allele 1—defined as the lower molecular weight of the two alleles) were taken as measure of inactivation discordance. RESULTS Of 63 MZ female twin pairs analyzed, only two pairs were homozygous for the repeat polymorphism and hence were uninformative—one control and one discordant SZ pair. For those subjects with DNA from both peripheral blood and buccal mucosa (n ¼ 58), we compared the X-inactivation pattern between the two tissues. On the whole, a similar pattern of X-inactivation was observed within both tissues (r ¼ 0.73, Evidence for X-Linked QTLs in Bipolar Disorder P < 0.01); however, considering the possible difference between tissues for X inactivation reported in previous studies, we decided to analyze the data from both tissues separately. To estimate the degree of similarity within FMZ twin pairs, we calculated the intra-pair difference as the absolute difference in the percentages of X inactivation between both members of each pair. Then, we calculated the mean values and the standard errors of the mean for the intra-pair differences in each of the five groups studied: BD discordant (BDD), BD concordant (BDC), SZ discordant (SZD), SZ concordant (SZC), and healthy controls (C) (see Fig. 1A,B). Based on the information from cheek swabs, the twins discordant for BD appeared to be the most discordant for the methylation status of their maternal and paternal X alleles (intra-pair difference for the percentage of X inactivation 17.3 5.7; see Fig. 1A) especially compared to twins concordant for the same disorder (intra-pair difference for the percentage of X inactivation 5.1 1.7; see Fig. 1A). In pair wise analysis, discordant bipolar versus concordant bipolar showed a significant difference at the 0.05 level (BDD vs. BDC: F ¼ 5.1, df ¼ 7, P ¼ 0.05) and discordant bipolar versus controls showed a trend to significant difference: F ¼ 2.4, df ¼ 24, P ¼ 0.1. None of the other pair-wise comparisons were significant (Table I). The results for peripheral blood based on higher numbers (Fig. 1B), supported the conclusion that bipolar discordant twin pairs are significantly more skewed in inactivation (BDD vs. BDC: F ¼ 2.3, df ¼ 18, P ¼ 0.03) than concordant twin pairs and also showed a strong trend to significantly greater skewing than controls (BDD vs. C: F ¼ 1.9, df ¼ 29, P ¼ 0.06). Again, none of the other pair-wise comparisons were significant (Table I). DISCUSSION Overall, these preliminary data suggest that as a group, pairs of twins discordant for BD may be more discordant for X inactivation patterns than the other pairs of twins we studied, suggesting that the X-chromosome may be involved in BD. The discordant BD twin pairs showed a strong trend to significant difference from controls. At the molecular level, it may suggest that the expression of X chromosomal genes with alleles divergent in functions implicated in BD can be quite different between members of the twin pair, and could contribute to their phenotypic discordance for this complex behavioral phenotype. To our knowledge, modern twin studies have not generally reported sex differences in concordance for func- 461 tional psychosis. Nevertheless, according to the Maudsley Twin data [see McGuffin et al., 2003] the MZ concordances for psychosis are slightly higher in males compared to females, this would support the tentative conclusions of our preliminary findings. While the three discordant SZ pairs studied also showed greater differences than the concordant SZ pairs in the percentage of X inactivation in buccal mucosa, the discordance was less marked than for the BDD twins. This may reflect the limited numbers available to study and a larger sample will be required to establish if there any evidence for X linked QTLs in SZ. This possibility remains plausible given the observation of X-chromosome aneuploidies among psychotic patients [DeLisi et al., 1994] and the epigenetic control of some regions on the X-chromosome suggested in relation to psychosis in previous studies [Giouzeli et al., 2004]. Genomic discordance between members of FMZ twin pairs can have two possible causes: the random nature of X-inactivation patterns, and deviation in epigenetic modification such as methylation and or histone modification acting on both sex chromosomes and autosomes. This epigenetic regulation may be influenced by environmental and stochastic factors and is compatible with the epidemiological data on SZ and BD [Singh et al., 2002]. The detection of epigenetic difference at genomic sites on both sex chromosomes and autosomes between members of MZ twins has remained largely unexplored primarily because of poor understanding of the phenomenon, the difficulty in obtaining large MZ samples of patients and limitations in technology. The X-chromosome has been implicated in a number of other behavioral traits including: homosexuality, affective disorders, general cognitive ability, and antisocial behavior, as well as having a large number of loci assigned that are associated with mental retardation [Loat et al., 2004; Bocklandt et al., 2006; Ropers, 2006]. Finally, our findings should be interpreted cautiously in the light of the methodological limitations. Firstly, because of restrictions in the availability of alternative material, X-chromosome inactivation was tested in buccal mucosa, and peripheral blood, but not directly from brain, the primary organ affected in BD and SZ. There are fundamental practical difficulties in studying tissue-specific methylation patterns in brain tissue; however, from the few studies carried out, it appears that correlations of inactivation profiles between different human tissues are reasonably high [see Brown and Robinson, 2000]. Secondly, a further possible limitation of our Fig. 1. Mean value of the intra-pair differences for the percentage of X-inactivation in the five groups of monozygotic female twins (FMZ) studied (SEM)—y-axis: FMZ discordant for schizophrenia (SZD), FMZ concordant for schizophrenia (SZC), FMZ discordant for bipolar disorder (BDD), FMZ concordant for bipolar disorder (BDC), and FMZ healthy controls. Values obtained in: (A) DNA from cheek swabs, (B) DNA from blood. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] 462 Rosa et al. TABLE I. Values of the Intra-Pair Differences for the Percentage of X-Inactivation in the Five Groups Studied and Statistical Comparisons With the Control Group Intra-pair differences for X-inactivation DNA from cheek swabs DNA from blood Group SZ discordant SZ concordant BD discordant BD concordant Controls SZ discordant SZ concordant BD discordant BD concordant Controls n Mean SE Statistical comparisons 3 4 4 5 22 3 4 13 7 18 10.1 3.4 8.5 1.9 17.3 5.7 5.1 1.7 9.7 2.3 2.7 1.1 6.7 2.2 13.0 2.6 4.6 2.4 6.4 2.2 SZD vs. C: t ¼ 0.1, df ¼ 23; P ¼ 0.9 SZC vs. C: t ¼ 0.2, df ¼ 24; P ¼ 0.8 BDD vs. C: t ¼ 1.5, df ¼ 24; P ¼ 0.1 BDC vs. C: t ¼ 1.1, df ¼ 25; P ¼ 0.2 — SZD vs. C: t ¼ 0.7, df ¼ 19; P ¼ 0.5 SZC vs. C: t ¼ 0.1, df ¼ 20; P ¼ 0.9 BDD vs. C: t ¼ 1.9, df ¼ 29; P ¼ 0.06 BDC vs. C: t ¼ 0.5, df ¼ 23; P ¼ 0.6 — results and interpretations arises from the fact we do not have information on chorionicity. Whether, or not, female MZ twins have similar X inactivation is highly dependent on the timing of twinning event [Monteiro et al., 1998] and chorionicity can provide evidence of a differential prenatal environment, in which dichorionic twins may experience more extreme differences compared to monochorionic twins. This factor should be taken into account in future studies. Finally, the study’s power was limited by the small sample size of the concordant and discordant pairs, especially for SZ; however, the results for BP are replicated employing DNA extracted from both buccal cells and blood. Studies of additional MZ twins with BD and SZ are needed to explore further the findings reported here. Whether, or not, these results are replicated, it is clear that the study of MZ twins continues to provide a unique opportunity to investigate the etiology of complex disorders and future epigenetic studies may lead to a better understanding of molecular differences in genetically identical individuals. Cardno AG, Rijsdijk FV, Sham PC, Murray RM, McGuffin P. 2002. A twin-study of genetic relationships between psychotic symptoms. Am J Psychiatry 159:539–545. ACKNOWLEDGMENTS Giouzeli M, Williams NA, Lonie LJ, DeLisi LE, Crow TJ. 2004. ProtocadherinX/Y, a candidate gene-pair for schizophrenia and schizoaffective disorder: A DHPLC investigation of genomic sequence. Am J Med Genet Part B 129B:1–9. This work was supported by the Wellcome Trust (MMP Research Training Fellowship 064971) and the Stanley Medical Research Institute. We also thank AGAUR (Generalitat de Catalunya) and Fundació Seny (Barcelona) for support to A. Rosa. We are also very grateful to L. Maguire and B. Freeman who performed the zygosity tests at the SGDP and the laboratory staff of the Division of Biomedical Genetics, UMC, Utrecht. REFERENCES Allen RC, Zoghbi HY, Moseley AB, Rosenblatt HM, Belmont JW. 1992. Methylation of HpaII and HhaI sites near the polymorphic CAG repeat in the human androgen-receptor gene correlates with X chromosome inactivation. Am J Hum Genet 51:1229–1239. American Psychiatric Association. 1994. Diagnostic and statistical manual of mental disorders. Washington, DC: American Psychiatric Press. Baron M. 1977. Linkage between an X-chromosome marker (deutan color blindness) and bipolar affective illness. Occurrence in the family of a lithium carbonate-responsive schizo-affective proband. Arch Gen Psychiatry 34:721–725. Bocklandt S, Horvath S, Vilain E, Hamer DH. 2006. Extreme skewing of X chromosome inactivation in mothers of homosexual men. Hum Genet 118:691–694. Bramon E, Sham PC. 2001. The common genetic liability between schizophrenia and bipolar disorder: A review. Curr Psychiatry Rep 3: 332–337. Brown CJ, Robinson WP. 2000. The causes and consequences of random and non-random X chromosome inactivation in humans. Clin Genet 58:353– 363. Cardno AG, Gottesman II. 2000. Twin studies of schizophrenia: From bowand-arrow concordances to star wars Mx and functional genomics. Am J Med Genet 97:12–17. Craig IW, Harper E, Loat CS. 2004. The genetic basis for sex differences in human behaviour: Role of the sex chromosomes. Ann Hum Genet 68:269–284. DeLisi LE, Friedrich U, Wahlstrom J, Boccio-Smith A, Forsman A, Eklund K, Crow TJ. 1994. Schizophrenia and sex chromosome anomalies. Schizophr Bull 20:495–505. Endicott J, Spitzer RL. 1978. A diagnostic interview: The schedule for affective disorders and schizophrenia. Arch Gen Psychiatry 35:837– 844. Freeman B, Powell J, Ball D, Hill L, Craig I, Plomin R. 1997. DNA by mail: An inexpensive and noninvasive method for collecting DNA samples from widely dispersed populations. Behav Genet 27:251–257. Freeman B, Smith N, Curtis C, Huckett L, Mill J, Craig IW. 2003. DNA from buccal swabs recruited by mail: Evaluation of storage effects on longterm stability and suitability for multiplex polymerase chain reaction genotyping. Behav Genet 33:67–72. Gershon ES, Guroff JJ. 1984. Information from relatives. Diagnosis of affective disorders. Arch Gen Psychiatry 41:173–180. Hardy J. 2006. Bad luck: An unappreciated limitation in the interpretation of twin studies. Am J Med Genet Part B (Neuropsychiatr Genet) 141:681. Loat CS, Asbury K, Galsworthy MJ, Plomin R, Craig IW. 2004. X inactivation as a source of behavioural differences in monozygotic female twins. Twin Res 7:54–561. McGuffin P, Rijsdijk F, Andrew M, Sham P, Katz R, Cardno A. 2003. The heritability of bipolar affective disorder and the genetic relationship to unipolar depression. Arch Gen Psychiatry 60:497–502. Mendlewicz J, Linkowski P, Guroff JJ, Van Praag HM. 1979. Color blindness linkage to bipolar manic-depressive illness. New evidence. Arch Gen Psychiatry 36:1442–1447. Monteiro J, Derom C, Vlietinck R, Kohn N, Lesser M, Gregersen PK. 1998. Commitment to X inactivation precedes the twinning event in monochorionic MZ twins. Am J Hum Genet 63:339–346. Peedicayil J. 2003. Future strategies in psychiatric genetics. Med Hypotheses 60:215–217. Petronis A. 2001. Human morbid genetics revisited: Relevance of epigenetics. Trends Genet 17:142–146. Petronis A. 2003. Epigenetics and bipolar disorder: New opportunities and challenges. Am J Med Genet Part C (Semin Med Genet) 123:65–75. Petronis A. 2006. Epigenetics and twins: Three variations on the theme. Trends Genet 22:347–350. Ropers HH. 2006. X-linked mental retardation: Many genes for a complex disorder. Curr Opin Genet Dev 16:260–269. Singh SM, Murphy B, O’Reilly R. 2002. Epigenetic contributors to the discordance of monozygotic twins. Clin Genet 62:97–103. Tsujita T, Niikawa N, Yamashita H, Imamura A, Hamada A, Nakane Y, Okazaki Y. 1998. Genomic discordance between monozygotic twins discordant for schizophrenia. Am J Psychiatry 155:422–424.