American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 141B:544–548 (2006)

Brief Research CommunicationMethylation of Two Homo sapiens-Specific X-YHomologous Genes in Klinefelter’s Syndrome (XXY)Norman L.J. Ross,1* Rekha Wadekar,1 Alexandra Lopes,2 Adam Dagnall,1

James Close,1 Lynn E. DeLisi,3 and Timothy J. Crow1

1SANE POWIC, University of Oxford, Warneford Hospital, Oxford, United Kingdom2IPATIMUP, Instituto de Patologiae Imunologia Molecular da Universidade do Porto, Portugal3The Department of Psychiatry New York University, NewYork, New York

An increased incidence of psychiatric and struc-tural brain abnormalities in individuals withKlinefelter syndrome (KS, 47 XXY) could be dueto the presence of extra copies of X-Y homologousgenes that escape X inactivation. Of particularinterest are the two brain-expressed genes Proto-cadherin11XY (PCDH11XY) and the Synaptobre-vin-like gene (SYBL1) which have been duplicatedfrom the X chromosome to the Y chromosome togive X-Y homologous gene pairs that are specific tomodern humans. We examined the DNA of KSindividuals reported recently by DeLisi et al.[2005] and determined the parental origin of theX alleles, the degree of skewed X inactivation andinvestigated the CpG island methylation statusof PCDH11XY and SYBL1 by bisulphite sequen-cing and quantification of methylated HpaII sites.We used a novel method for quantification ofunmethylated CpGs with the restriction enzymeMcrBC which cuts methylated but not unmethy-lated CpGs. The results showed that KS indivi-duals have two methylated and one unmethylatedSYBL1allele whereasPCDH11XY is unmethylatedand escapes X inactivation on the extra X chromo-some. Overexpression of PCDH11XY in KS isprobable and variable escape from inactivationof this Homo sapiens-specific gene could accountfor some abnormalities in KS. The origin of theparental alleles or their preferential X inactiva-tion was not associated with psychotic symp-toms. � 2006 Wiley-Liss, Inc.

KEY WORDS: KS; DNA; SYBL1; PCDH11XY; X-inactivation

Please cite this article as follows: Ross NLJ, Wadekar R,Lopes A, Dagnall A, Close J, DeLisi LE, Crow TJ. 2006.Methylation of Two Homo sapiens-Specific X-Y Homo-logous Genes in Klinefelter’s Syndrome (XXY). Am JMed Genet Part B 141B:544–548.

Several studies have reported verbal abnormalities andpsychiatric conditions in KS males [reviewed in DeLisi et al.,1994; van Rijn et al., 2005]. XXY men are found in about 0.8–

1% of males with schizophrenia, approximately 4- to 5-foldexcess compared to normal male live births. Similarlyapproximately 1% of individuals with mental retardation havean XXY karyotype [Rosen et al., 1970]. In addition, structuralabnormalities in the brain including enlarged ventricles andreversed or diminished cerebral asymmetries have beenreported [Warwick et al., 1999, 2003; Rezaie et al., 2004].

Recently DeLisi et al. [2005] reported that KS individualshad abnormalities in both gray and white matter volumesof frontal and temporal lobes and white matter tracts leading tothese areas. DeLisi et al. [2005] and others [Crow, 1994;Geschwind et al., 1998] have postulated that excessiveexpression of one or more genes on the X chromosome,particularly those that escape X inactivation (that includethe genes that have homologues on the Y) could be responsiblefor the abnormalities. Crow [2002a,b] has further suggestedthat the gene PCDH11X, and its epigenetic regulation, is theleading candidate. This gene is part of the Xq21.3 block thatwas transposed to Yp11.2 6 million years ago in the hominidlineage, and at that point PCDH11X acquired a homologuePCDH11Y. Thus humans have a gene pair PCDH11X andPCDH11Y whereas other mammals have only the gene pcdhX.Both copies are strongly expressed in the human brain andretain undisrupted open reading frames [Blanco et al., 2000].The encoded protein is a member of the protocadherin family ofcell surface adhesion molecules which have the potential toplay a role in cell–cell recognition. The precise function of the Xand Y copies has not been determined. While males have the Xand Y copies of the gene and females have copies on each oftheir two X chromosomes, it is not known if the copy on theinactive X escapes inactivation.

In this study we explore the possibility that there might be anassociation between an escape from inactivation by methyla-tion of PCDH11XY (and thus an excess production of thisgene’s product) and the abnormal phenotypes reported byDeLisi et al. [2005]. Specifically we have examined themethylation status of the CpG islands of PCDH11XY in bloodleukocytes of the KS individuals using bisulfite sequencing aswell as the restriction enzyme McrBC which allowed quanti-fication of unmethylated CpGs. In general, genes havingunmethylated CpGs in their CpG islands are transcriptionallyactive whereas those containing methylated CpGs are silent.Exceptions have been reported by Anderson and Brown [2005]for cultured cell lines. We hypothesize that if PCDH11Xescapes inactivation, in the case of XXY individuals, all threealleles ofPCDH11XYwill be unmethylated whereas males andfemales of normal karyotype will have two unmethylatedalleles.

We have included SYBL1 in our study because associationswith bipolar disease [Saito et al., 2000; Muller et al., 2002] havebeen reported. This gene like the PCDH11X and Y gene pair ispresent on both X and Y chromosomes in humans but only onthe X chromosome in other primates. Moreover, the gene isunusual in that one allele (X or Y) is methylated and inactive in

*Correspondence to: Norman L.J. Ross, SANE POWIC, Uni-versity of Oxford, Warneford Hospital, Oxford OX3 7JX, UnitedKingdom. E-mail: norman.ross@psych.ox.ac.uk

Received 21 September 2005; Accepted 20 March 2006

DOI 10.1002/ajmg.b.30339

� 2006 Wiley-Liss, Inc.

individuals with normal karyotype [Matarazzo et al., 2002]. Itis not known how many SYBL1 alleles are active in KSindividuals. In addition to determining the methylation statusof the gene in the probands we have also determined theparental origin of the X chromosomes and the level of skewed Xinactivation, in view of the earlier report of Iitsuka et al. [2001]who observed skewed X inactivation in 5 of 16 KS cases andsuggested that this phenomenon could explain the wide rangeof mental and phenotypic abnormalities observed among KSindividuals.

The KS individuals, their parents and normal (XY) maleshave been described [DeLisi et al., 2005]. The results ofstructured psychiatric interviews summarized in Table Iindicated that 4 of the 11 KS were psychotic (KS-PSY) andhad hallucinations and delusions. A group of unrelated malesand females from Oxford, UK whose Diagnostic interview forGenetic Studies failed to reveal any psychiatric diagnosis byDSM-IV category I or II criteria served as normal controls.DNA was extracted commercially (Tepnel Life Sciences plc,Manchester, UK) from blood leukocytes of these individualsusing the Nucleon kit.

We carried out the human androgen receptor (HUMARA)methylation assay [Allen et al., 1992] to determine whether theparental origin of the extra X or skewed inactivation couldexplain the abnormalities reported in these KS individuals.One hundred nanograms of genomic DNA were digestedovernight with 20 U of HpaII (Roche Diagnostics Ltd, EastSussex, UK) at 378C in 50 ml volumes. The isoschizomer MspIfrom the same manufacturer was routinely used as a control tocheck that digestion was complete. The digests were inacti-vated at 708C before amplification by PCR using CY5-labelledprimers. Usually 10 ng of the digested DNA were amplified in athermocycler (Eppendorf Mastercycler, Hamburg, Germany).Primers and conditions for amplification are summarized inTable II. Five microliters of the amplified products weredenatured and separated by electrophoresis in polyacrylamidegels (Sequagel XR: National Diagnostics, East Riding, York-shire, UK) in an Alf Express sequencer (Amersham Bios-ciences, Amersham, UK). Peak areas were computed using theAllelix software (Amersham Biosciences). Skewing of Xinactivation was calculated as described [Iitsuka et al., 2001].Our results (Table I) are partly in agreement with Iitsuka et al.[2001] in that approximately 60% (5/8) of the KS had onematernal and one paternal X whereas in 40% (3/8) of cases bothalleles were maternally derived. In contrast to Iitsuka et al.[2001], none of the 8 KS who had two distinguishable X alleles

showed preferential inactivation of one allele (�80% skewed)but two mothers had marked skewed inactivation. Their KSoffsprings inherited the active maternal allele and the paternalallele. In view of the small number of informative samples, wewere unable to identify an association between the origin orselective inactivation of parental alleles and psychosis.

To determine how many SYBL1 alleles are methylated inKS, we quantified the relative number of molecules containingmethylated CmCGG at three HpaII sites in the SYBL1promoter. These correspond to CpGs 6, 19, and 29 [Matarazzoet al., 2002] and are known to be unmethylated on active X butmethylated on inactive X and Y. Digests containing 10 ng ofgenomic DNA were amplified by PCR using fluorescentlylabeled primers (Table II). The digests were also amplified withprimers for MSX2, an autosomal gene, which does not containHpaII sites in the amplified region, used previously as aninternal control for normalization [Ross et al., 2003]. All PCRassays were done separately in triplicate as described aboveand 5 ml amounts of amplified products were quantified asdescribed for the HUMARA assay. Results were expressed asratios of SYBL1/MSX2. When the data for KS and KS withpsychosis (KS-PSY) were pooled for analysis, the mean ratios ofSYBL1/MSX2 for the pooled KS group, and normal males andfemales were 0.36, 0.17, and 0.18, respectively. The twofolddifference between the KS and normal controls was highlysignificant (One way ANOVA, P< 0.001) indicating that thereare approximately twice as many methylated CmCGG in theKS group compared to normal males and females. SYBL1/MSX2 ratios of mothers and fathers were lower than those ofthe KS group (P< 0.001) but did not differ significantly fromthose of normal males and females.

We conclude from these results that KS have two methylatedand one unmethylated SYBL1 allele.

Preliminary experiments (not shown) using the methyla-tion-sensitive restriction enzyme HpaII indicated that CpGislands 1 and 2 of PCDH11XY (Fig. 1) are not methylated.Bisulfite sequencing of genomic DNA from 4 KS and 2 KS-PSYindividuals carried out essentially as described by Lopes et al.[2006] showed that the percentage of methylated CpGs in allclones examined ranged from 0 to 7% for fragment A (CpGs 1–16) and 0–3% for fragment B (CpGs 17–22). Data wereobtained by sequencing a total of 20 clones (360 CpGs)from fragment A and 22 clones (132 CpGs) from fragmentB. Similarly none of the CpGs examined from the 50 end ofCpG island 2 was methylated. In view of the sparse, randomdistribution of methylated CpGs, we concluded that the

TABLE I. Genotypes and X Inactivation Status of the KS Families

Sample Alleles % Skewed Remarks Clinical diagnosis

01 XXY 282, 285 59 No parental DNA02 M 248, 269 5802 XXY 269 (M), 285 (F) 61 Delusions and hallucinations03 M 267, 273 9003 XXY 273 (M), 283 (F) 6204 XXY 264, 268 55 No parental DNA Delusions05 M 280, 286 6605 XXY 286 (M) X 2 50 Two copies of same allele06 XXY 271 48 No parental DNA Delusions and hallucinations07 M 264, 277 8007 XXY 264 (M), 274 (F) 5508 XXY 264 (M), 290 (M) 69 Paternal DNA only12 M 267, 283 6812 XXY 283 (M), 264 (F) 6614 XXY 266 (M) X 2 49* Paternal DNA only Delusions and hallucinations15 XXY 266 (M) X 2 51* Paternal DNA only Delusions and hallucinations

Bold numbers indicate the size in bp of the most prominent allele (i.e., methylated) in HpaII digests. (*) indicates that the maternal alleles wereindistinguishable in size. It is therefore not known whether both alleles were digested equally by HpaII or whether X inactivation was skewed in these cases.(M) and (F) indicate mother and father, respectively.

DNA Methylation in Klinefelter’s Syndrome 545

CpG islands of PCDH11XY are unmethylated and that noneof the alleles are silenced. Our results are consistent withthose of Lopes et al. [2006] who concluded that both CpGislands are unmethylated in males and females of normalkaryotype.

Since it was not possible to quantify methylated CpGs ofPCDH11XY, we used a novel approach to determine whetherKS had more unmethylated CpG alleles than controls ofnormal karyotype. The method consists essentially in digest-ing genomic DNA with the restriction enzyme McrBC (NewEngland Biolabs, Hitchin, Hertfordshire, UK) which cutsmethylated CpGs but fails to cut unmethylated CpGs [Yamadaet al., 2004], followed by PCR amplification and quantificationof the amplified products. We used PGK1 for normalizationsince it is established that this gene is subject to X inactiva-tion [Pfeifer et al., 1990] and consequently both KS and normalkaryotypes are expected to have one unmethylated allele(Fig. 2).

To test the validity of this assumption, we quantifiedmethylated CmCGG in the CpG islands and promoter regionofPGK1andSYBL1using primers and conditions summarizedin Table II. The results showed that the ratios of methylatedPGK1 relative to MSX2 for KS and normal female were0.34� 0.05 (2�SE) and 0.38� 0.04 (2�SE), respectively,whereas the corresponding ratios for SYBL1/MSX2 were0.34� 0.06 and 0.16� 0.02, respectively. Thus, KS containapproximately twofold more methylated CmCGG in SYBL1compared to normal females (t-test, P¼ 0.002) whereas therelative number of molecules containing methylated sites inPGK1 were the same. These findings are consistent with thefollowing: (1) both KS and XX females have one active and oneinactive PGK1 allele, (2) KS have two inactive and one activeSYBL1 allele whereas normal females have one active and oneinactiveSYBL1 allele. Accordingly we usedPGK1 to normalizequantitation of unmethylated PCDH11XY and SYBL1 by theMcrBc method.

One hundred nanograms amounts of genomic DNA weredigested with 20 U of McrBC (New England Biolabs) and 10 ngof the digested DNA were amplified using primers spanning 17CpGs located at the 50 end of the CpG island 1 (Fig. 1 andTable II). For each experiment, aliquots of 10 ng of the digestswere also amplified separately with the PGK1 primersdescribed above (Table II). All assays were done in triplicateand the products quantified as above. Results were expressedas ratios of unmethylated PCDH11XY/unmethylated PGK1(Fig. 3A). The mean ratio of unmethylated PCDH11XY/PGK1of the pooled KS group was 1.4 times (0.57/0.4) greater thanpooled normal controls (t-test;P¼ 0.001) and is consistent withour hypothesis outlined in Figure 2 that PCDH11XY escapesinactivation and that all three alleles are unmethylated in KS.Similar results were obtained when unmethylated CpGs(CpGs 1–14) in CpG island 2 were quantified. The ratio ofunmethylated PCDH11XY/PGK1 for the pooled KS relative tocontrols of normal karyotype was 1.3 and the difference wassignificant (t-test, P< 0.05). In contrast to PCDH11XY, weexpected that ratios for unmethylated SYBL1/PGK1 would beconstant across genotypes. The results (Fig. 3B) are consistentwith this hypothesis. The mean ratios were 1.27 and 1.35 forthe pooled KS and normal controls respectively and were notsignificantly different (t-test, P> 0.05).

The main finding reported here is that KS individuals havethree unmethylated alleles ofPCDH11XY compared to the twounmethylated alleles that are in females and males of normalkaryotype. Consistent results were obtained for both CpGislands using bisulfite sequencing on the one hand and thenovel method for quantification of unmethylated CpGs inMcrBC digests of genomic DNA on the other. The latter methodhas the advantage that it examines CpGs in both strands ofDNA in a more representative sample of the genome (estimated

TA

BL

EII

.S

um

mary

ofP

CR

Pri

mer

san

dC

ond

itio

ns

for

Am

pli

fica

tion

For

ward

pri

mer

Rev

erse

pri

mer

An

nea

lte

mp

eratu

re(8

C)

HU

MA

RA

50 C

Y5T

AG

GG

CT

GG

GA

AG

GG

TC

TA

C50 G

AA

CC

AT

CC

TC

AC

CC

TG

CT

59.8

SYBL1

50 C

Y5T

CG

AA

CG

AA

CG

TG

AA

AC

AC

T50 C

AC

CC

CA

CT

CC

AT

TT

AC

CT

G60

PGK1

50 C

Y5G

CA

TT

TG

CA

TT

GA

TC

CT

GG

T50 A

GG

AA

CC

TT

CC

CG

AC

TT

AG

G60

PCDH11X

YC

pG

isla

nd

150 C

Y5G

TG

AG

CA

GA

GG

AC

CG

AG

GA

GA

AC

50 C

AG

CA

CC

AA

CT

GC

TG

AG

TA

CC

60

PCDH11X

YC

pG

isla

nd

25

0 CY

5A

AG

CA

AG

TC

GA

AG

TA

AT

AG

TG

TA

TG

TA

50 G

GA

CT

CA

GC

TT

TT

GG

GA

AC

T55

PC

Rs

wer

eca

rrie

dou

tin

25ml

react

ion

su

sin

gR

edT

aq

Rea

dy

Mix

PC

Rre

act

ion

kit

(Sig

ma-A

ldri

ch,D

orse

t,U

K).

Aft

eran

init

iald

enatu

rati

onst

epat

948C

for

5m

in,sa

mp

les

wer

esu

bje

cted

to20

–25

cycl

es(p

red

eter

min

edto

ach

ieve

lin

ear

am

pli

fica

tion

)fo

r40

sec

at

948C

,40

sec

at

the

ap

pro

pri

ate

an

nea

lin

gte

mp

eratu

re,e

xte

nsi

onfo

r1

min

at

728C

,fol

low

edby

afi

nals

tep

for

10

min

at

728C

.Not

eth

at

the

pri

mer

sfo

rPCDH11X

Yd

on

otd

iffe

ren

tiate

bet

wee

nX

an

dY

all

eles

.

546 Ross et al.

to be approximately 1,800 molecules) based on the amount ofDNA used for quantification.

The implications of our results are that X-Y homologousgenes such as PCDH11XY that escape X inactivation have thepotential to be expressed at higher levels in KS individualscompared to individuals with normal karyotype; these genescould therefore contribute to the abnormalities reported in KS.

We realize that DNA methylation may not always beconcordant with expression [Anderson and Brown, 2005].However, it is of interest that microarray hybridization hasshown elevated expression of PCDH11XY mRNA in lympho-cytes from individuals with four or five X chromosomes[Sudbrak et al., 2001]. Moreover, using PCDH11X-specificprimers for RT-PCR Lopes et al. [2006] have shown thatexpression of the X allele is twofold greater in XX females thanXY males and concluded that the gene escapes X inactivation.Expression of PCDH11Y was not examined by Lopes et al.[2006], but there is evidence from other work [Blanco et al.,2000] that PCDH11Y is also expressed in the brain. Ouranalysis lacked power to demonstrate an association betweenpsychosis among the KS and levels ofPCDH11XY methylation.Clearly a larger number of cases should be examined in thefuture.

In view of the evidence that sex chromosomal aneuploidiesare associated with reciprocal (verbal with an extra X or Y vs.non-verbal with absence of one X or Y) deficits in cognitivedevelopment, the concept that a determinant of the relativedevelopment of the two hemispheres is represented on the Xand Y chromosomes [Crow, 1994, 1998] is plausible. Furtherinvestigation of the identity and epigenetic control of dosage ofsuch genes is indicated. The arguments that directionalasymmetry on a population basis is Homo sapiens-specificand that the cerebral torque is the defining feature of thehuman brain [Crow, 2004] make such genes relevant to thehuman capacity for language, while the evidence that theXq21.3/Yp region of homology was established at 6MYA, that

Fig. 1. (A) and (B) represent CpG islands 1 and 2 of PCDH11XY, respectively. Vertical lines indicate CpG sites and thick lines represent exons. Arrowindicates the approximate position of initiation of transcription [Blanco-Arias et al., 2004].

Fig. 2. Schematic illustration (not to scale) of X and Y chromosomesshowing gene dosage in KS (XXY) and normal males and females. Empty andfilled symbols represent unmethylated and methylated alleles, respectively.If PCDH11X escapes inactivation (not methylated), the ratio of unmethy-lated PCD11XY/PGK1 is expected to be 2 in XX and XY individuals and 3 inXXY. Hence the ratio for XXY individuals relative to normal males andfemales should be 1.5. Ratios of unmethylatedSYBL1/PGK1 are expected tobe 1 in each case. (?) indicates that the methylation status of the markedgenes was unknown when this study was undertaken.

Fig. 3. Quantification of unmethylated PCDH11XY (A) and unmethylated SYBL1 (B). Graphs show ratios (means and 2 X SE) of peak areas relative toPGK1. All-KS (pooled KS and KS-PSY, n-9, excluding two outliers). Controls (n¼ 9, pooled 4 males and 5 females).

DNA Methylation in Klinefelter’s Syndrome 547

is, at about the time of separation of chimpanzee and hominidlineages and subsequently modified in the course of hominidevolution identifies this block of homology and the brainexpressed gene pair PCDH11X/Y as prime candidates forfurther investigation. Epigenetic control of this small class ofgenes may have played a critical role in the evolution of thehuman brain.

ACKNOWLEDGMENTS

We are grateful to SANE and the T J Crow Psychosis Trustfor support. Alexandra Lopes has a Ph.D. grant from the FCTPortugal.

REFERENCES

Allen RC, Zoghbi H, Moseley AB, Rosenblatt HM, Belmont JW. 1992.Methylation of HpaII and HhaI sites near the polymorphic CAG repeatin the human androgen-receptor gene correlates with X chromosomeinactivation. Am J Hum Genet 51:1229–1239.

Anderson CL, Brown CJ. 2005. Epigenetic predisposition to expression ofTIMPI from the human inactive X chromosome. BMC Genet 6:48.

Blanco P, Sargent CA, Boucher CA, Mitchell M, Affara NA. 2000.Conservation of PCDHX in mammals; expression of human X/Y genespredominantly in brain. Mamm Genome 11:906–914.

Blanco-Arias P, Sargent CA, Affara NA. 2004. A comparative analysis of thepig, mouse, and human PCDHX genes. Mamm Genome 15:296–306.

Crow TJ. 1994. The case for an X-Y homologous determinant of cerebralasymmetry. Cytogenet Cell Genet 67:393–394.

Crow TJ. 1998. Why cerebral asymmetry is the key to the origin of Homosapiens; how to find the gene or eliminate the theory. CurrentPsychology of Cognition/Cahiers de Psychologie Cognitiv 11:1237–1277.

Crow TJ. 2002a. In: Wray A, editor. The Transition to language. Oxford, UK:Oxford University Press. pp 93–112.

Crow TJ. 2002b. Handedness, language lateralisation and anatomicalasymmetry: Relevance of protocadherin XY to hominid speciation andthe aetiology of psychosis. Br J Psychiatry 181:295–297.

Crow TJ. 2004. Directional asymmetry is the key to the origin of modernHomo sapiens (the Broca-Annett axiom). Response to Lesley Rogers.Laterality, 9:233–242.

DeLisi LE, Freidrich U, Wahlstrom J, Boccio-Smith A, Forsman A, EklundK, Crow TJ. 1994. Schizophrenia and sex chromosome anomalies.Schizophr Bull 20:495–505.

DeLisi LE, Maurizio AM, Svetina C, Ardekani B, Szulc K, Nierenberg J,Leonard J, Harvey PD. 2005. Klinefelter’s syndrome (XXY) as a geneticmodel for psychotic disorders. Am J Med Genet B Neuropsychiatr Genet135:15–23.

Geschwind DH, Gregg J, Boone K, Karrim J, Pawlikowska-Haddal A, Rao E,Ellison J, Ciccodicola A, D-Urso M, Woods R, Rappold GA, Swerdloff R,Nelson SF. 1998. Klinefelter’s syndrome as a model of anomalouscerebral laterality: Testing gene dosage in the X chromosome pseudoau-tosomal region using a DNA microarray. Dev Genet 23:215–229.

Iitsuka Y, Bock A, Nguyen D, Samango-Sprouse CA, Simpson JL, BischoffFZ. 2001. Evidence of skewed X-chromosome inactivation in 47, XXY and48,XXYY Klinefelter patients. Am J Hum Genet Part B NeuropsychiatrGenet 98:25–31.

Lopes AM, Ross N, Close J, Dagnall A, Amorim A, Crow TJ. 2006.Inactivation status of PCDH11X: sexual dimorphisms in gene expres-sion levels in brain. Hum Genet Jan 20, 1–9 (Epub ahead of print).

Matarazzo MR, De Bonis ML, Gregory RI, Vacca M, Hansen RS, MercadanteG, D’Urso M, Feil R, D’Esposito M. 2002. Allelic inactivation ofthe pseudoautosomal gene SYBL1 is controlled by epigenetic mechan-isms common to the X and Y chromosomes. Hum Mol Genet 11:3191–3198.

Muller DJ, Schulze TG, Jahnes E, Cichon S, Krauss H, Kesper K, Held T,Maier W, Propping P, Nothen MM, Rietschel M. 2002. Associationbetween a polymorphism in the pseudoautosomal X-linked gene SYBL1and bipolar affective disorder. Amer J Med Genet Part B NeuropsychiatrGenet 114:74–78.

Pfeifer GP, Steigerwald SD, Hansen RS, Gartler SM, Riggs AD. 1990.Polymerase Chain Reaction-Aided Genomic Sequencing of an XChromosome- linked CpG island: Methylation patterns suggest clonalinheritance, CpG site autonomy, and an explanation of activity statestability. Proc Natl Acad Sci USA 87:8252–8256.

Rezaie R, Roberts N, Cutter WJ, Murphy DGM, Robertson DMW, Daly EM,Maurizio A, DeLisi LE, Crow TJ. 2004. Anomalous asymmetry inTurner’s and Klinefelter’s syndromes-Further evidence for X-Y linkageof the cerebral dominance gene. Am J Med Genet Part B NeuropsychiatrGenet 130B:102–103.

Rosen M, Bergman M, Telfer MA. 1970. MMPI profiles of males withabnormal sex chromosome complements. J Ment Defic Res 15:169–176.

Ross NL, Mavrogiannis LA, Sargent CA, Knight SJ, Wadekar R, DeLisi LE,Crow TJ. 2003. Quantification of X-Y homologous genes in patients withschizophrenia by multiplex polymerase chain reaction. Psychiatr Genet13:115–119.

Saito T, Parsia S, Papolos DF, Lachman HM. 2000. Analysis of thepseudoautosomal X-linked gene SYBL1in bipolar affective disorder:Description of a new candidate allele for psychiatric disorders. Am J MedGenet Part B Neuropsychiatr Genet 96:317–323.

Sudbrak R, Wieczorek G, Nuber UA, Mann W, Kirchner R, Erdogan F,Brown CJ, Wohrle D, Sterk P, Kalscheuer VM, Berger W, Lehrach H,Ropers HH. 2001. X-Chromosome-specific cDNA arrays: Identification ofgenes that escape from X-inactivation and other applications. Hum MolGenet 10:77–83.

Van Rijn S, Swaab H, Aleman A, Kahn RS. 2005. Klinefelter Syndrome (47XXY): Biological-Genetic vulnerability to schizophrenia and impair-ments in social-emotional processing. Schizophr Bull 31:280.

Warwick MM, Doody GA, Lawrie SM, Kestelman JN, Best JJ, JohnstoneEC. 1999. Volumetric magnetic resonance imaging study of the brain insubjects with sex chromosome aneuploidies. J Neurol NeurosurgPsychiatry 66(5):628–632.

Warwick MM, Lawrie SM, Beveridge A, Johnstone EC. 2003. Abnormalcerebral asymmetry and schizophrenia in a subject with Klinefelter’ssyndrome (XXY). Biological Psychiatry 53:627–629.

Yamada Y, Watanabe H, Miura F, Soejima H, Uchiyama M, Iwasaka T,Mukai T, Sakaki Y, Ito T. 2004. A comprehensive analysis of allelicmethylation status of CpG islands on human chromosome 21q. GenomeRes 14:247–266.

548 Ross et al.