Bipolar Affective Disorder Partially CosegregatesWith a Balanced t(9;11)(p24;q23.1) ChromosomalTranslocation in a Small Pedigree

Bora E. Baysal,1,2 Steven G. Potkin,3 Joan E. Farr,1 Michael J. Higgins,4 Jeff Korcz,5Susanne M. Gollin,2 Michael R. James,6 Glen A. Evans,5 and Charles W. Richard III1,2*1Department of Psychiatry, Western Psychiatric Institute and Clinic, The University of Pittsburgh Medical Center,Pittsburgh, Pennsylvania

2Department of Human Genetics, The University of Pittsburgh, Pittsburgh, Pennsylvania3Department of Psychiatry, The University of California at Irvine Medical Center, Irvine, California4Roswell Park Cancer Institute, Buffalo, New York5Eugene McDermott Center for Growth and Development, University of Texas Southwestern Medical School atDallas, Texas

6Wellcome Trust Centre for Human Genetics, Oxford, United Kingdom

Analysis of an extended pedigree in which abalanced t(9;11)(p24;q23.1) translocationwas found to cosegregate with bipolar affec-tive disorder revealed that five of 11 trans-location carriers had bipolar affective disor-der and one carrier had unipolar depres-sion. There were no affected individuals inthe pedigree without the balanced translo-cation. We hypothesized that gene(s) orgene regulatory regions disrupted by thetranslocation might be contributing to thebipolar affective disorder in a dominantfashion. To test this hypothesis, we isolatedthe derivative chromosome 9 and derivativechromosome 11 in somatic cell hybrids andidentified the nearest flanking markers onchromosome 9 (D9S230 and D9S2011E/HRFX3) and chromosome 11 (EST00652 andCRYA2). YAC contigs were constructed inthe region of flanking markers for bothchromosomes 9 and 11. Chromosome 11breakpoint was localized within an 8-kb re-gion in a small insert (100 kb) YAC. Chromo-some 9 breakpoint was localized within ap-proximately 2 Mb region. Several genes andESTs including EST00652, CRYA2, DRD2,5HTR3 on chromosome 11 and VLDLR andSLC1A1 on chromosome 9 were mappedwithin the vicinity of the breakpoint but

were shown not to be disrupted by thetranslocation breakpoint. Although severalpossibilities exist regarding the role of thebalanced translocation in developing bipo-lar affective disorder in this pedigree, in-cluding a chance cosegregation, identifica-tion of a disrupted gene or gene regulatoryregion with the help of physical mapping re-sources described in this study may help toidentify the presence of a susceptibilitygene for this disorder. Am. J. Med. Genet.(Neuropsychiatr. Genet.) 81:81–91, 1998.© 1998 Wiley-Liss, Inc.

KEY WORDS: bipolar disorder; transloca-tion; chromosome 11; chro-mosome 9; physical map

INTRODUCTION

Family, twin and adoption studies have long sug-gested a strong genetic contribution to the etiology ofbipolar affective disorder [Nurnberger et al., 1994]. Al-though earlier studies examining the mode of inheri-tance of bipolar disorder in families through complexsegregation analysis have yielded equivocal results,more recent segregation analysis with larger familysamples have suggested a single major gene effect[Sham et al., 1991; Spence et al., 1995]. Others haveargued against this single major locus model on theo-retical grounds [Craddock et al., 1997]. Several groups,including a collaborative multicenter, NIMH-fundedeffort, have undertaken to locate single major genesthrough a genome-wide search for linked chromosomalmarkers [Berrettini et al., 1997]. Although this genome

Contract grant sponsor: NARSAD; Contract grant sponsor:DOE; Contract grant sponsor: NIH; Contract grant number:NG00202; Contract grant sponsor: Turkish Scientific and Tech-nical Research Council.

*Correspondence to: Charles W. Richard III, M.D., Ph.D.,WPIC Room 1445, University of Pittsburgh Medical Center, 3811O’Hara St., Pittsburgh, PA 15213-2593. E-mail: richard+@pitt.edu

Received 10 June 1997; Revised 26 August 1997

American Journal of Medical Genetics (Neuropsychiatric Genetics) 81:81–91 (1998)

© 1998 Wiley-Liss, Inc.

scanning strategy has been aided by recent advances inhigh-throughput automated genotyping and the avail-ability of highly polymorphic simple-sequence repeatmarkers, no definitive linkages have yet been identi-fied and replicated [Baron et al., 1993; Kelsoe et al.,1989]. This consistent lack of linkage replication pointsto the complex genetic basis of bipolar disorder, andimplies either the presence of multiple susceptibilityloci, each conferring a low risk or the presence of con-siderable genetic heterogeneity in genes of major effect.Uncertainties over the precise mode of inheritance, thecorrect bipolar phenotype, and the presence of nonge-netic phenocopies have all hampered efforts to formu-late a single best strategy for linkage analysis. Non-parametric analysis approaches, including affected sibpair (ASP) [Blackwelder and Elston, 1985] and affectedpedigree member (APM) [Weeks and Lange, 1988] donot require specification of the genetic model of illnessinheritance. ASP and ASM methods have been appliedto some BP pedigrees with results suggesting a suscep-tibility gene in the pericentromeric region of chromo-some 18 [Berrettini et al., 1994; Freimer et al., 1996;Stine et al., 1995]. More recently, evidence for linkagehas been reported on 6p24, 13q13, 15q11, and 4p16[Freimer et al., 1996; Ginns et al., 1996], although allthese linkage results await definitive replication [Rischand Botstein, 1996]. Alternative strategies have beenexplored including direct mutation testing of favoredcandidate genes and systematic testing of associa-tion between (expressed) polymorphisms in candidategenes with disease status to identify susceptibilitygenes. Family-based association studies require poten-tially fewer numbers of affected individuals than sib-pairs analyses in identifying such low-risk genes [Rischand Merikangas, 1996]. To date, no definitive gene mu-tations in a psychiatric disease have been uncovered.

An alternative approach to bipolar gene identifica-tion involves the identification of chromosomal aberra-tions associated with bipolar disorder which may sug-gest regions in which to focus the search for diseasegenes by parametric and nonparametric statistical ap-proaches [Bassett, 1992; Craddock and Owen, 1994].The existence of isolated chromosomal translocationshas aided in the positional cloning of such single genedisorders as Duchenne’s muscular dystrophy and neu-rofibromatosis type I. Velo-cardio-facial syndrome as-sociated with bipolar illness involves cytogeneticallyvisible and submicroscopic deletions of human chromo-some 22q11 [Papolos et al., 1996] and has focused aconsortium search for predisposing genes to that chro-mosomal region [Lachman et al., 1997]. Bassett [1992]reviewed the published literature of chromosomal ab-errations in major psychiatric illness and has identifiedfour possible chromosomal regions of interest at 5q,11q, 18q, and 19p. To complement these studies, wehave attempted to clone a chromosome t(9;11)(p24,q23.1) translocation breakpoint partially segregatingwith bipolar affective disorder in a small family. Thehypothesis to be tested postulates that a coding orregulatory sequence of a gene (or genes) on either chro-mosome 9 or 11 is disrupted by the translocation break-point and acts in a dominant fashion to cause bipolarillness.

MATERIALS AND METHODSClinical and Cytogenetic Analysis

The pedigree was ascertained through structuredpsychiatric diagnostic interviews (SADS-LB) of ill andwell family members by a psychiatrist (S.G.P.) or atrained professional psychiatric diagnostician. Chro-mosome karyotypes of peripheral lymphocytes wereperformed by standard techniques.

Construction of Somatic Cell Hybrids, Isolationof Derivative 9 and Derivative 11 Chromosomes

Somatic cell hybrids segregating the derivativetranslocation chromosomes were constructed aspreviously described [Richard and Washington, 1994].Briefly, an EB-immortalized, balanced translocationcarrier cell line (1069) was fused with a hamsterHPRT-cell line (380-6) using polyethylene glycol 4000(Boehringer Mannheim, Indianapolis, IN) and colonieswere selected in the presence of hypoxanthine-aminop-terin-thymidine. DNA from a total of 141 survivingcolonies was harvested by standard techniques and as-sayed by human specific polymerase chain reaction(PCR) assays for selected chromosome 9 and chromo-some 11 markers flanking the translocation breakpoint[Richard and Washington, 1994]. Human/rodent so-matic cell hybrids containing normal human chromo-some 9 (PK87-9) or normal human chromosome 11 (J1)were obtained from the NIGMS Human Mutant CellRepository.

YAC Characterization: STS Content Mappingand Pulsed Field Gel Electrophoresis (PFGE)

Individual mega-YAC clones of Centre d’etude duPolymorhisme Humain (CEPH), ESTs, STRPs, andother custom-designed PCR primers were obtainedfrom Research Genetics (Huntsville, AL). Small insert,nonchimeric YAC clones were obtained from a chromo-some 11-specific YAC library constructed at theRoswell Park Cancer Institute (RPCI) [Qin et al.,1993]. YAC clones were grown in selective media andDNA was isolated either in liquid media for SequenceTagged Site (STS) content mapping or in agaroseblocks for PFGE analysis, using standard procedures.PCR amplification was performed using either 60 ng ofcloned template DNA or 120 ng of total genomic DNA.The final concentrations of reagents were as follows: 1× PCR reaction buffer (Boehringer-Mannheim), 1.5mM MgCl2; 1 mM PCR primers; 0.20–0.25 mM eachdNTP; and 1 U Taq Polymerase (Boehringer-Mannheim). The mixture was heated to 94°C for 3 minin a Techne PHC3 thermal cycler. The usual cyclingconditions consisted of 35 cycles of 94°C for 45 s, 55°Cfor 45 s, and 72°C for 45 s, followed by a final extensionof 72°C for 10 min. The annealing temperature wasmodified depending on the TM of a primer set. Reactionproducts were analyzed on 1.5% ethidium bromide-stained agarose gels. PCR primer sequence informationof most ESTs and STSs used in this study is available atthe Genome Database at http://gdbwww.gdb.org//. PCRprimer sequence for other STSs are listed in Table I.

The undigested and restriction enzyme-digestedhigh molecular weight DNAs of YAC clones in agarose

82 Baysal et al.

blocks were separated by PFGE using LKB PulsaphorElectrophoresis Unit (Pharmacia, Piscataway, NJ).Electrophoresis was performed in 1.5% agarose gels in1 × TBE buffer at 10°C, using various pulse times at180 V for 20–40 hours. Saccharomyces cerevisiae chro-mosomal DNA and lambda DNA digests were used assize standards. Rare cutter restriction enzyme siteswere mapped on yRP-12a6 by completely digesting thehigh molecular weight DNA and subsequently bySouthern hybridization with YAC left, right end probesand human Cot-I DNA. yRP-12a6 ends and yRP-18g8right end were isolated by inverse PCR and vector-IRSamplification, respectively. YRP-12a6 ends were con-verted to STSs by sequencing (fmol Sequencing, Pro-mega, Madison, WI), and yRP-18g8 right end was useddirectly as a hybridization probe.

YAC Subcloning and Characterizationof Cosmids

Subcloning of YAC yRP-18g8 was performed as de-scribed [Monaco and Larin, 1994], except that PFGEwas used for sizing the partial digestion products.Briefly, an Mbol partial digest of yRP-18g8 in agaroseplugs was prepared, dephosphorylated by CIAP (NEB,Beverly, MA) and ligated by T4 Ligase (NEB) into acosmid vector (SuperCos 1, Stratagene, La Jolla, CA)and packaged using Gigapack II XL packaging extracts(Stratagene).

Colonies were plated and screened by human Cot-1DNA using standard colony screening methods. A totalnumber of 148 positive clones were selected and ar-rayed in microtiter format culture plates. This numbercorresponded to ∼15-fold human YAC insert represen-tation. Next, cosmid clones were characterized by hy-bridization and restriction enzyme analysis. Hybridiza-tions were performed on filter replicas, which weremade by extracting cosmid DNAs in microtiter platesand transferring 5–50 ng of DNA to positively chargednylon membranes (Hybond N+, Amersham, ArlingtonHeights, IL) in alkali conditions (0.4 N NaOH). Eithersingle-copy probes or isolated total YAC DNAs wereused as hybridization probes. YAC DNAs were sepa-rated from yeast chromosomes by PFGE on 1.5% LMPAgarose, cut out from the gel, and purified by b-Agarase (NEB) digestion using manufacturer’s proto-col. Cosmid clones were grown overnight at 37°C in 2mL LB medium supplemented by 50 mg/mL carbenicil-lin, DNAs were extracted by alkaline lysis, 1.5–2.0 mgwas restriction enzyme digested to completion and the

products were analyzed on a 0.7% agarose gel. The re-striction enzymes were obtained from New EnglandBiolabs (Beverly, MA) and digestions were performedaccording to manufacturer’s instructions. Selected cos-mids were also partially digested by 0.01, 0.1, 0.8, and20 U/mg enzyme amounts for 2 hours, and enzyme siteswere mapped by T3 and T7 oligonucleotide hybridiza-tion.

Southern Blot Hybridization

Southern blot hybridizations were performed accord-ing to standard procedures [Sambrook et al., 1989].Briefly, 1–2.5 mg of cloned DNA or 10–20 mg of genomicDNA was restriction enzyme-digested and separated in0.7% agarose gel. DNA transfers were performed underalkaline conditions (0.4 N NaOH) to positively chargednylon membranes (Hybond N+, Amersham). Radioac-tive labeling was performed either by random oligo-nucleotide priming for complex probes or by 58-end la-beling for oligonucleotide probes. Hybridizations wereperformed overnight in a solution of 5 × SSPE, 0.5%SDS, 1× Denhardt’s solution, and 0.1 mg/ml salmonsperm DNA at 65°C for complex probes. Hybridizationtemperature for oligonucleotide probes was varied de-pending on the Tm of the probe. The final washingstringencies were 0.1 × SSPE, 0.1% SDS at 65°C forcomplex probes and 2 × SSPE, 0.1% SDS at room tem-perature for oligonucleotide probes. X-ray films (X-Omat AR, Kodak, Rochester, NY and Hyperfilm-BMax,Amersham) were exposed for 30 min to 30 days at−80°C before developing.

RESULTSClinical and Cytogenetic Analysis

We followed up on initial descriptions of a familywith a balanced t(9;11) chromosomal translocation thatwas reported to segregate with bipolar affective disor-der [Smith et al., 1989]. We initially focused on indi-viduals in the left side of the pedigree (Fig. 1, individu-als 1–37), and found that five of seven individuals withthe t(9;11) balanced translocation had bipolar I illness(6, 12, 13, 30, and 26), and one individual (29) had anearly-onset, recurrent unipolar depression. One 21-year-old female (28) was a translocation carrier andwas not affected at the time of diagnostic psychiatricevaluation. A cytogenetic reanalysis of G-bandedmetaphase chromosomes from a previously describedcarrier individual (28) defined the karyotype as

TABLE I. New STSs Derived From Chromosome 11

Locus PrimersAmpliconsize (bp)

Annealingtemperature (°C)

yRP-12a6RE 58-GAGGTGAGTCTCTGGTATAGTTTT-38 555a 6058-GATCTGACATTCACATCATCA-38

yRP-12a6LE 58-TAGTTCAGACATCATCTTCCAAGG-38 159 6058-TGAATCTAATGAGGTCACTGCCTT-38

296F7R 58-CGTTGACTAAGTCTTGCCGAC-38 250a 5558-GGAAACGAACTTCCCTAGGTC-38

272H05R 58-CCTCAGTAACTTGTTAATTCCAC-38 180a 5558-CAGATCCAGGACTGTCTGATCC-38

aAmplicon sizes are not exact, since the whole STSs were not sequenced.

Bipolar Disorder and Balanced Translocation t(9;11) 83

t(9;11)(p24;q23.1). (Fig. 2), placing the breakpoint onchromosome 9 more distally than what was reportedpreviously [Smith et al., 1989]. In the left branch of thepedigree, we found no affected individuals with normalchromosomal karyotypes. These data were consistentwith the hypothesis that the translocation might becausally related to development of bipolar affective dis-order. We began physical characterization of the trans-location breakpoint (see below), while clinical evalua-tion of the right side of the pedigree (Fig. 1, individuals38–68, initially lost to follow-up) was undertaken. Fur-ther clinical investigation revealed no affected indi-viduals in this right branch of the pedigree. However,we identified four unaffected individuals (9, 39, 52, and53) carrying the t(9;11) balanced translocation. Thesenew family data did not provide further evidence for acausative role of the translocation breakpoint in thedevelopment of bipolar affective disorder (see Discus-sion).

Isolation of the Derivative Chromosomes,Definition of the Closest Flanking Markers,

Construction of YAC Contigs, and Isolation ofYACs Spanning the Chromosome 11

Translocation Breakpoint

From a fusion of an translocation patient lympho-blastoid cell line and a recipient hamster cell line, foursomatic cell hybrids (EN311, EN308, EN315B, andEN335) were isolated which contained the human de-rivative 9 chromosome and lacked the human deriva-tive 11 and the normal human 9 and 11 chromosomes.Three somatic cell hybrids (EN328B, EN342, andEN348) were isolated which contained the human de-rivative 11 chromosome and lacked the human deriva-tive 9 and the normal human 11 and 9 chromosomes.

Using chromosome 11 PCR-based STS assays with

the isolated somatic cell hybrids [Richard et al., 1993],the translocation breakpoint on chromosome 11q23.1was initially localized to an estimated 400-KB regionbetween the gene-based markers CRYA2 andD11S966E (EST 00652). This more precise mappingplaces this 11q23.1 translocation breakpoint approxi-mately 13 Mb telomeric to a previously reportedt(1;11)(q43;q21) translocation breakpoint segregatingwith schizophrenia [St Clair et al., 1990] and manyMbs centromeric to a t(6;11)(q14.5;q25) balanced trans-location breakpoint segregating with bipolar disorder[Holland and Gosden, 1990]. STS content mapping ofregional CEPH mega-YACs identified four mega-YACscrossing the chromosome 11 translocation breakpoint(Fig. 3a). A chromosome 11 specific small-insert YAClibrary constructed at the Roswell Park Cancer Insti-tute (RPCI) was also screened by the flanking markers.This provided nine small-insert YACs positive forD11S966E and three small-insert YACs positive forCRYA2 (Fig. 3a). Four other overlapping YACs,yRP4f7, yRP4h7, yRP-12a6, and yRP2e5, were also in-cluded in the clone contig. A subset of individual small-insert YACs were used as fluorescence in situ hybrid-ization (FISH) probes on metaphase spreads of a trans-location carrier to determine if they span thebreakpoint. YACs yRP2g7, yRP3g9, and yRP-12h10mapped telomeric to the breakpoint, i.e., on derivativechromosome 9; yRP-12a6 mapped centromeric to thebreakpoint, i.e., on derivative chromosome 11; yRP-18g8, yRP-15e3, and yRP-17f11 mapped on both sidesof the breakpoint, i.e., on both derivative 9 and deriva-tive 11 chromosomes (data not shown). Hence, theselatter three YACs were identified as crossing the chro-mosome 11 translocation breakpoint.

Using PCR-based chromosome 9 STSs, the translo-cation breakpoint on chromosome 9p24 was localized

Fig. 1. Cosegregation of the balanced translocation and the bipolar disorder in the pedigree. NL, normal karyotype, 9;11, carrier of balancedtranslocation. The symbols used to define the clinical status of each individual are depicted in the box.

84 Baysal et al.

between the markers D9S230 and D9S2011E/HRFX3(Fig. 3b). The distance between these two markers isestimated to be less than 2 Mb based on WhiteheadInstitute/MIT genome center chromosome 9 radiationhybrid framework map (data release 11) and data fromthe fifth international workshop on human chromo-some 9 (http://www.gene.ucl.ac.uk). CEPH mega-YACclone contigs were constructed at both sides of the chro-mosome 9 translocation breakpoint by STS contentmapping. Although these two separate contigs couldnot be linked by any single marker, a minimal unde-tected overlap between the contigs could not be ex-cluded without further analysis.

Sixteen STSs were mapped on the proximal chromo-some 9 contig, which is comprised of 10 mega-YAC

clones. The ordering of the markers was facilitated byhigh genomic representation (1- to 7-fold) of the CEPHmega-YACs. This order is mostly consistent with thedata in a recent integrated map of the chromosome 9.The order of the mapped markers at the distal side ofthe breakpoint, however, was less certain, since theclone and marker density was poor in this region.

Mapping and Testing the Integrity of PositionalCandidate Genes

The two closest markers flanking the breakpoint onchromosome 11 are genes: an anonymous EST(EST00652, cDNA clone HFBCD37) and alpha-B crys-tallin (CRYA2), which was previously implicated in a

Fig. 2. Cytogenetic analysis of the G-banded chromosomes involved in the balanced translocation t(9;11)(p24; q23.1). Normal chromosomes 9 and 11and derivative chromosomes 9 and 11 were depicted. Four metaphase chromosomes of different sizes as well as a schematic representation were includedfor each chromosome.

Bipolar Disorder and Balanced Translocation t(9;11) 85

86 Baysal et al.

neurodegenerative disease [Iwaki et al., 1992]. PCR as-says with primers derived from 38 and 58 end of themplaced EST00652 and CRYA2 centromeric and telo-meric to the chromosome 11 translocation breakpoint,respectively. Both genes are not interrupted by thetranslocation (data not shown). Clinical and pharma-cological evidence suggests a relationship between do-paminergic neurotransmission and bipolar disorder.On the basis of a unique 210-kb Sfi1 CHEF gel frag-ment found in a t(9;11) translocation carrier not foundin nontranslocation carriers, Smith et al. [1989] sug-gested the translocation breakpoint was within 210 kbof the Dopamine Receptor D2 (DRD2) gene. Eubanks etal. [1992] reported the DRD2 gene spans 270 kb, withthe first exon separated from the second exon by about250 kb. These two results predict the t(9;11) transloca-tion would interrupt DRD2 in the first intron. Radia-tion hybrid maps of this region, however, predicts theDRD2 gene would lie several Mbs telomeric to thetranslocation breakpoint [Richard et al., 1993]. To re-solve this conflict, we used PCR amplification withprimers from the 58 untranslated region (UTR) ofDRD2, a microsatellite repeat in the second intron, andfrom the 38 UTR. We conclude all parts of the DRD2gene are centromeric to and not interrupted by thetranslocation breakpoint (data not shown). The neuro-transmitter serotonin has also been implicated in theetiology of bipolar disorder and the type 3 serotoninreceptor has been mapped to 11q23 by FISH [Weiss etal., 1995]. Using PCR primers derived from the 38 UTRof the 5HTR3 gene, we mapped 5HTR3 to a YAC contigseveral Mb telomeric to the translocation breakpoint[Baysal et al., 1997].

The SLC1A1 (previously know as HEAAC1) andVLDLR genes were previously mapped to chromosome9p24 by FISH [Gafvels et al., 1993; Sakai et al., 1994;Smith et al., 1994]. The SLC1A1 gene encodes a high-affinity glutamate transporter, a transmembrane pro-tein crucial for terminating the action of the excitatoryneurotransmitter glutamate and in maintaining extra-cellular glutamate concentrations below neurotoxiclevels [Kanai and Hediger, 1992]. VLDLR gene encodesthe receptor involved in the uptake of VLDL particlesinto the cells by binding to ApoE protein. The alleles ofa trinucleotide repeat in VLDLR may modify therisk for Alzheimer’s disease [Okuizumi et al., 1995].SLC1A1 and VLDLR genes were mapped centromericand telomeric to chromosome 9 translocation break-point, respectively, using PCR primers derived frombeginning and end of their cDNAs. Both genes are notinterrupted by the translocation (data not shown). Thegene for DNA binding protein HRFX3 and an anony-

mous gene marker, D9S2011, were mapped on the YACcontig on the telomeric side, and two anonymous genemarkers, D11S1151 and D11S1153, were mapped onthe YAC contig on the centromeric side of the chro-mosome 9 translocation breakpoint. The genes forSNF2L2 and GLDC were mapped on the telomeric andcentromeric side of the translocation breakpoint, re-spectively, but did not map to the chromosome 9 YACcontigs (Fig. 3b).

Fine Mapping the Translocation Breakpoint onChromosome 11

We subcloned the small insert YAC yRP-18g8 into acosmid vector to further locate the chromosome 11translocation breakpoint. We first isolated both ends ofan overlapping 100-kb insert YAC, yPR-12a6, and con-verted them to STSs (Table I). Next, we tested thesetwo yRP-12a6, end-derived STSs on somatic cell hy-brids and showed that this small insert YAC spannedthe translocation breakpoint (Fig. 4). Total YAC yRP-12a6 hybridization to yRP-18g8- derived cosmids iden-tified three cosmid clones, NB1H10 (30 kb), andNB1A8 (30 kb), and NB2C4 (12 kb). Two more cosmidclones, NB1G10 (35 kb) and NBRE2 (30.5 kb), weredetected by yRP-12a6 left and right end hybridizations,respectively. NB1G10 was also detected by HFBCD37(D11S966E) cDNA probe hybridization (Fig. 5). PCRanalysis of selected cosmids by yRP-12a6 end-derivedSTSs and by D11S966E confirmed the results obtainedby hybridization (data not shown). EcoRI restrictionenzyme analysis of the selected five cosmid clones didnot reveal any common bands among the five selectedcosmids, implying minimal or no overlap among theclones. Cosmids NB1H10, NB1G10, and NBRE2 werehybridized to blots containing yRP-18g8-derived cos-mids to identify any minimal overlaps. The only over-lap was detected between NB1H10 and NB2C4 by us-ing total NB1H10 DNA as a hybridization probe.NB2C4 was later placed overlapping with bothNB1H10 and NB1A8 by EcoRI end fragment hybrid-ization and cosmid end sequencing (data not shown).No other overlap was detected. The rare cutter enzymesites were used as markers to map the selected cosmidson yRP-12a6. The recognition sites for a panel of rarecutter enzyme sites were mapped on the YAC cloneyRP-12a6 and on the cosmids. The presence of uniquerestriction enzyme sites helped locate the cosmids rela-tive to yRP-12a6 ends (Fig. 5). The cosmids NB1G10and NBRE2 have minimal overlaps (<3 KB) with yRP-12a6, as tested by restriction enzyme mapping. Thereare two gaps in the contig adjacent to these cosmids,both of which are approximately 20 kb. Although thetotal screened number of yRP-18g8-derived cosmidclones corresponded to approximately 15-fold genomicrepresentation, we showed that most clones were ofvery small insert (<5 kb) and there was a considerablecloning bias (data not shown). This might explain thepresence of gaps in the cosmid contig.

After the cosmids were mapped on yRP-12a6, a sys-tematic approach was initiated to identify the junc-tional translocation fragment in the genomic DNA of a

Fig. 3. Sequence Tagged Site and gene-based physical maps of the (a)chromosome 11q23.1 and (b) chromosome 9p24 translocation breakpointregions. The translocation breakpoints were localized within the shadedregions on YAC clone contigs by STS content mapping. STSs with equiva-lent hit patterns were grouped in boxes. Light Xs represent hits detected bypolymerase chain reaction (PCR) amplification; dark Xs represent hits de-tected by PCR amplification and also confirmed by probe hybridization;dark circles represent hits detected by probe hybridization; light circlesrepresent untested but predicted hits based on evidence obtained eitherfrom STS location (for yRP-18g8) or from other databases. YAC end-derived STSs/probes were designated with a dark square at the end of theYAC from which they were derived.

Bipolar Disorder and Balanced Translocation t(9;11) 87

Fig. 4. yRP-12a6 spans the translocation breakpoint on chromosome 11q23.1. yRP-12a6RE (∼555 bp) amplified from derivative chromosome 9 (EN311)and it did not amplify from derivative chromosome 11 (EN328B). yRP-12a6LE (159 bp) amplified from derivative chromosome 11 (EN328B) and it didnot amplify from derivative chromosome 9 (EN311). Both STSs amplified from translocation carrier cell line (1069) and yRP-12a6. These results showedthat yRP-12a6 (100 kb) spans the translocation breakpoint. The hamster cell line (380-6) and water were included as negative controls.

88 Baysal et al.

carrier individual by Southern hybridization using T3and T7 EcoRI fragments of the cosmids as probes. DNAfrom a normal individual was included in the analysis,and several enzymes were used to exclude the possibil-ity of restriction enzyme-length polymorphism (RFLP).The 1.1 KB EcoRI fragment derived from the T7 end ofNB1H10 detected the junctional fragments that wasmissing in the control DNA (Fig. 6). Thus, the chromo-some 11 translocation breakpoint was mapped to ap-proximately 20 KB telomeric to a unique Sal1 site andto approximately 20 KB centromeric to the right end(telomeric end) of YAC yRP-12a6. The presence of acluster of rare cutter restriction enzyme sites approxi-mately 20 kb centromeric to the translocation break-point implies the presence of nearby gene(s).

DISCUSSION

We describe a small pedigree in which bipolaraffective disorder segregates with a balanced

t(9p24;11q23.1) translocation in six of 11 translocationcarriers. The interesting but presently unansweredquestion remains as to what specific role the translo-cation breakpoint might play in the etiology of bipolaraffective disorder in this pedigree.

One likely possibility is that there is no direct causallink between the balanced translocation and bipolaraffective disorder, and that the translocation is coseg-regating by chance in this pedigree. The presence offive individuals with the translocation breakpoint butwithout bipolar illness either supports this viewpointor alternatively argues against a fully penetrant, dom-inant effect of the translocation breakpoint. Attemptsto further expand the pedigree beyond what is pre-sented in Figure 1 by identifying additional individualswith the translocation breakpoint and bipolar illnesshave so far proven unsuccessful.

The possibility also remains that the translocationbreakpoint disrupts gene(s) or gene regulatory re-

Fig. 5. A restriction map of the translocation breakpoint region on chromosome 11q23.1. The upper part of the figure shows the location of rare-cutterrestriction enzyme sites on yRP-12a6 and the overlapping cosmids. The lower part of the figure shows the restriction enzyme sites of NB1H10 on a smallerscale map. The translocation breakpoint was confined to the interval defined by vertical dashed lines. The T7 ends of the cosmids are depicted by smallvertical arrows. The T7 EcoRI fragment of cosmid NB1H10, which is marked by a thick horizontal bar, detected junctional translocation bands on genomicDNA (see Fig. 6). There were no restriction enzyme sites for NotI, NruI, PacI, and RsrII on yRP-12a6. SmaI and XhoI sites were mapped only on cosmidNB1H10. E1, EcoRI; Ea, EagI; Bs, BssHII; Sal, SalI; M, MluI; Sc, SacI; S, SmaI; X, XhoI; and Sf, SfiI.

Bipolar Disorder and Balanced Translocation t(9;11) 89

gion(s) contributing to bipolar illness. The presence ofnonpenetrant individuals is not necessarily unexpectedas the precise mode of inheritance in this complex be-havioral disorder is far from clear (e.g. monozygotictwins are not fully concordant for bipolar illness). Thepresence of other susceptibility genes or environmentalinfluences may be necessary to trigger manic or depres-sive symptoms in individuals carrying the transloca-tion karyotype. A critical finding that would negatethis particular competing hypothesis would be thepresence of individuals in this pedigree with bipolardisorder despite normal karyotypes. No such individu-als, however, were identified. The identification of can-didate genes disrupted by the translocation breakpointwould provide a mechanistic starting point for a morein-depth analysis of the functional role of specific genesin bipolar illness. Several plausible candidates in the11q23 or 9p24 regions were investigated in this studyand were found not to be disrupted and therefore prob-ably not contributing to the bipolar phenotype, al-though we cannot entirely rule out distant chromosom-al position effects [Milot et al., 1996]. These nearbygenes also cannot be entirely eliminated since thetranslocation could merely represent a marker for amutation in a cosegregating gene. In our search for newcandidates, we have localized both chromosome 9 andchromosome 11 translocation breakpoints by definingthe closest flanking markers, and have preciselymapped the chromosome 11 breakpoint within an ap-proximately 8 kb region. These physical reagents willprovide the mapping resources for further investiga-

tion of this particular t(9:11)(p24;q23.1) translocationbreakpoint in bipolar affective disorder.

ACKNOWLEDGMENTS

This research was funded by a NARSAD Young In-vestigator Award to C.W.R. We thank Beth Fine andMichele DeAntonio for their help in family ascertain-ment, Eva Sujansky for help in karyotype analysis ofsome family members, and Moyra Smith for the gift ofimmortalized cell line HHW1069. B.E.B was partlysupported by a predoctoral fellowship from the theTurkish Scientific and Technical Research Council.GAE is supported by NIH NG00202 and DOE funding.

REFERENCES

Baron M, Freimer NF, Risch N, Lerer B, Alexander JR, Straub RE, AsokanS, Das K, Peterson A, Amos J, Endicott J, Ott J, Gilliam TC (1993):Diminished support for linkage between manic depressive illness andX-chromosome markers in three Israeli pedigrees. Nat Genet 3:49–55.

Bassett AS (1992): Chromosomal aberrations and schizophrenia. Auto-somes. Brit J Psychiatry 161:323–334.

Baysal BE, van Schothorst EM, Farr JE, James MR, Devilee P, RichardCW (1997): A high resolution of STS, EST, and gene-based physicalmap of the hereditary paraganglioma region on chromosome 11q23.Genomics 44:214–221.

Berrettini WH, Ferraro TN, Goldin LR, Weeks DE, Detera-Wadleigh S,Nurnberger JI, Jr., Gershon ES (1994): Chromosome 18 DNA markersand manic-depressive illness: Evidence for a susceptibility gene. ProcNatl Acad Sci USA 91:5918–5921.

Berrettini WH, Ferraro TN, Goldin LR, Detera-Wadleigh SD, Choi H, Mu-niec D, Guroff JJ, Kazuba DM, Nurnberger JI, Hsieh WT, Hoehe MR,

Fig. 6. Localization of the chromosome 11 breakpoint by genomic Southern hybridization. Hybridization of 1.1 kb T7 EcoRI fragment of cosmidNB1H10 detected aberrant bands (denoted by arrowheads) in the translocation carrier but not in the normal control DNA after digestion by severalrestriction enzymes. No aberrant bands were detected after PstI and BglII digestions. The small size of the normal EcoRV fragment, which was disruptedby the translocation, localized the breakpoint within ∼4.4 kb of the hybridization probe (See Fig. 5). T, translocation carrier, 1069; N, normal control. Amolecular weight marker in kb is shown on the right.

90 Baysal et al.

Gershon ES (1997): A linkage study of bipolar illness. Arch Gen Psy-chiatry 54:27–35.

Blackwelder WC, Elston RC (1985): A comparison of sib-pair linkage testsfor disease susceptibility loci. [published erratum appears in GenetEpidemiol 1986;3(5):379]. Genetic Epidemiol 2:85–97.

Craddock N, Owen M (1994): Chromosomal aberrations and bipolar affec-tive disorder. Brit J Psychiatry 164:507–512.

Craddock N, Vaneerdewegh P, Reich T (1997): Single major locus modelsfor bipolar disorder are implausible. Am J of Med Genet (Neuropsychi-atr Genet) 74:18.

Eubanks JH, Djabali M, Selleri L, Grandy DK, Civelli OX, McElligott DL,Evans GA (1992): Structure and linkage of the D2 dopamine receptorand neural cell adhesion molecule genes on human chromosome 11q23.Genomics 14:1010–1018.

Freimer NB, Reus VI, Escamilla MA, Mcinnes LA, Spesny M, Leon P,Service SK, Smith LB, Silva S, Rojas E, Gallegos A, Meza LX, FournierE, Baharloo S, Blankenship K, Tyler DJ, Batki S, Vinogradov SX, Weis-senbach J, Barondes SH, Sandkuijl LA (1996): Genetic mapping usinghaplotype, association, and linkage methods suggests a locus for severebipolar disorder (BPI) at 18q22-q23. Nat Genet 12:436–441.

Gafvels ME, Caird M, Britt D, Jackson CL, Patterson D, Strauss JF, 3d.(1993): Cloning of a cDNA encoding a putative human very low densitylipoprotein/apolipoprotein E receptor and assignment of the gene tochromosome 9pter-p23. Somatic Cell Molec Genet 19:557–569.

Ginns EI, Ott J, Egeland JA, Allen CR, Fann CS, Pauls DL, Weissen-bachoff J, Carulli JP, Falls KM, Keith TP, Paul SM (1996): A genome-wide search for chromosomal loci linked to bipolar affective disorder inthe Old Order Amish. Nat Genet 12:431–435.

Holland T, Gosden C (1990): A balanced chromosomal translocation par-tially co-segregating with psychotic illness in a family. Psychiatry Res32:1–8.

Iwaki A, Iwaki T, Goldman JE, Ogomori K, Tateishi J, Sakaki Y (1992):Accumulation of alpha B-crystallin in brains of patients with Alex-ander’s disease is not due to an abnormality of the 58-flanking andcoding sequence of the genomic DNA. Neurosci Lett 140:89–92.

Kanai Y, Hediger MA (1992): Primary structure and functional character-ization of a high-affinity glutamate transporter. Nature 360:467–471.

Kelsoe JR, Ginns EI, Egeland JA, Gerhard DS, Goldstein AM, Bale SJ,Pauls DL, Long RT, Kidd KK, Conte G, Housman DE, Paul SM (1989):Re-evaluation of the linkage relationship between chromosome 11p lociand the gene for bipolar affective disorder in the Old Order Amish.Nature 342:238–243.

Lachman HM, Kelsoe JR, Remick RA, Sadovnick AD, Rapaport MH, LinM, Pazur BA, Roe AA, Saito T, Papolos DF (1997): Linkage studiessuggest a possible locus for bipolar disorder near the velo-cardio-facialsyndrome region on chromosome 22. Am J Med Genet 74:121–128.

Milot E, Fraser P, Grosveld F (1996): Position effects and genetic disease.Trends Genet 12:123–126.

Monaco AP, Larin Z (1994): Generating subclones from large-inset genomicclones. In Dracopoli NC, Haines JL, Korf BR, Moir DT, Morton CC,Seidmann CE, Seidmann JG, Smith DR (eds): ‘‘Current Protocols inHuman Genetics.’’ New York: John Wiley & Sons, pp 5.11.1–5.11.6.

Nurnberger JI, Goldin LR, Gershon ES (1994): Genetics of PsychiatricDisorders. In Winokur G, Clayton P (eds): ‘‘The Medical Basis of Psy-chiatry.’’ Baltimore: Williams and Wilkins, pp 76–101.

Okuizumi K, Onodera O, Namba Y, Ikeda K, Yamamoto TX, Seki K, UekiA, Nanko S, Tanaka H, Takahashi H, Oyanagi K, Mizusawa H, Ka-

nazawa I, Tsuji S (1995): Genetic association of the very low densitylipoprotein (VLDL) receptor gene with sporadic Alzheimer’s disease.Nat Genet 11:207–209.

Papolos DF, Faedda GL, Veit S, Goldberg R, Morrow B, Kucherlapati RX,Shprintzen RJ (1996): Bipolar spectrum disorders in patients diag-nosed with velo-cardio-facial syndrome—does a hemizygous deletion ofchromosome 22q11 result in bipolar affective disorder. Am J Psychiatry153:1541–1547.

Qin S, Zhang J, Isaacs CM, Nagafuchi S, Jani SS, Abel KJ, Higgins MJ,Nowak NJ, Shows TB (1993): A chromosome 11 YAC library. Genomics16:580–585.

Richard CW, III, Washington SS (1994): Construction and assay of radia-tion hybrids. In Dracopoli NC, Haines JL, Korf BR, Moir DT, MortonCC, Seidmann CE, Seidmann JG, Smith DR (eds): ‘‘Current Protocolsin Human Genetics.’’ New York: John Wiley and Sons, pp 3.3.1–3.3.9.

Richard CW, III, Cox DR, Kapp L, Murnane J, Cornelis F, Julier C, La-throp M, James MR (1993): A radiation hybrid map of human chromo-some 11q22-q23 containing the ataxia-telangiectasia disease locus. Ge-nomics 17:1–5.

Risch N, Botstein D (1996): A manic depressive history. Nat Genet 12:351–353.

Risch N, Merikangas K (1996): The future of genetic studies of complexhuman diseases. Science 273:1516–1517.

Sakai J, Hoshino A, Takahashi S, Miura Y, Ishii H, Suzuki HX, Kawa-rabayasi Y, Yamamoto T (1994): Structure, chromosome location, andexpression of the human very low density lipoprotein receptor gene. JBiol Chem 269:2173–2182.

Sambrook J, Fritsch E, Maniatis T (1989): ‘‘Molecular Cloning. A Labora-tory Manual.’’ Cold Spring Harbor: CSH Laboratory Press.

Sham PC, Morton NE, Rice JP (1991): Segregation analysis of the NIMHcollaborative study. Family data on bipolar disorder. Psychiatric Genet2:175–184.

Smith CP, Weremowicz S, Kanai Y, Stelzner M, Morton CCX, Hediger MA(1994): Assignment of the gene coding for the human high-affinity glu-tamate transporter EAAC1 to 9p24: Potential role in dicarboxylic ami-noaciduria and neurodegenerative disorders. Genomics 20:335–336.

Smith M, Wasmuth JJ, McPherson JD (1989): Cosegregation of an11q22.3-9p22 translocation with affective disorder: proximity of thedopamine D2 receptor gene relative to the translocation breakpoint.Am J Hum Genet 45:A178.

Spence MA, Flodman PL, Sadovnick AD, Bailey-Wilson JEX, Ameli H,Remick RA (1995): Bipolar disorder: Evidence for a major locus. Am Jof Med Genet (Neuropsychiatr Genet) 60:370–376.

St. Clair D, Blackwood D, Muir W, Carothers A, Walker MX, Spowart G,Gosden C, Evans HJ (1990): Association within a family of a balancedautosomal translocation with major mental illness. Lancet 336:13–16.

Stine OC, Xu J, Koskela R, McMahon FJ, Gschwend MX, Friddle C, ClarkCD, McInnis MG, Simpson SG, Breschel TS, Vishio E, Riskin K,Feilotter H, Chen E, Shen S, Folstein S, Meyers DA, Botstein D, MarrTG, DePaulo JR (1995): Evidence for linkage of bipolar disorder tochromosome 18 with a parent-of-origin effect. Am J Hum Genet 57:1384–1394.

Weeks DE, Lange K (1988): The affected-pedigree-member method of link-age analysis. Am J Hum Genet 42:315–326.

Weiss B, Mertz A, Schrock E, Koenen M, Rappold G (1995): Assignment ofa human homolog of the mouse Htr3 receptor gene to chromosome11q23.1-q23.2. Genomics 29:304–305.

Bipolar Disorder and Balanced Translocation t(9;11) 91