CAG Repeat Sequences in Bipolar AffectiveDisorder: No Evidence for Association in a FrenchPopulation

Cecilia Zander,1 Franck Schurhoff,2 Claudine Laurent,3 Olivier Chavand,3 Frank Bellivier,2Daniele Samolyk,3 Marion Leboyer,4 Jean-Francois Allilaire,4 Howard Cann,1 Christian Neri,1 andJacques Mallet3*1CEPH, Fondation Jean Dausset, Paris, France2CNRS URA 1957, Hopital Pitie-Salpetriere, Paris, France3LGN, Batiment CERVI, Hopital Pitie-Salpetriere, Paris, France4Service de Psychiatrie Adulte, Hopital Pitie-Salpetriere, Paris, France

Anticipation has been described in bipolaraffective disorder (BPAD). However, thereare conflicting results from associationstudies screening for a link between BPADand CAG/CTG repeat expansions, the mo-lecular basis of anticipation in several he-reditary neurodegenerative disorders.Here, the repeat expansion detection (RED)method was used to screen for CAG repeatexpansion in 119 French BPAD patients.Western blotting was also used to search forpolyglutamine stretches, encoded by CAGexpansion, among proteins, extracted fromlymphoblastoid cell lines, from six selectedfamilial cases.

Maximum CAG/CTG repeat length did notdiffer significantly (P = 0.38) between the119 BPAD patients and the 88 controls in-cluded in the study. Several categories ofsubgroups were used, none of which showedsignificant association with a long repeat.Nor was a specific protein with an unusu-ally long polyglutamine stretch (lower de-tection limit, ∼33 polyglutamines) detectedin cell lysates from the familial cases stud-ied.

In conclusion, an association between along CAG/CTG repeat and BPAD in theFrench population sample studied was not

found. Nonetheless, a short repeat (<40 re-peats) might still be implicated, and thispossibility warrants further study. Am. J.Med. Genet. (Neuropsychiatr. Genet.) 81:338–341, 1998. © 1998 Wiley-Liss, Inc.

KEY WORDS: bipolar affective disorder;anticipation; genetics; tri-nucleotide repeat expansion;polyglutamine

INTRODUCTION

Little is understood about the underlying causes ofbipolar affective disorder (BPAD). A genetic etiology isstrongly supported by the familial nature of the condi-tion, by adoption studies showing an increased rate inchildren of affected parents, and by the increased con-cordance in monozygotic twins [Petronis and Kennedy,1995]. However, there is at present no consensus re-garding the genetic mechanisms that account for trans-mission of the illness within families. Positive evidencefor linkage to as many as 14 different chromosomallocations has been proposed in numerous studies since1969; none has been convincingly replicated [Risch andBotstein, 1996]. Evidence for anticipation, a clinicalphenomenon that exhibits an earlier age of onset and/or an increase in disease severity in successive genera-tions within families, has been suggested in severalstudies [McInnis et al., 1993; Nylander et al., 1994;Grigoroiu-Serbanescu et al., 1997].

The discovery of trinucleotide repeat expansion(TRE) in the early 1990s provided a convincing molecu-lar basis for anticipation. Today, an increasing numberof neurological disorders displaying anticipation havebeen identified, in which TRE plays a major role inpathogenesis, including myotonic dystrophy, fragile Xsyndrome, Huntington’s disease, and Friedreich ataxia[reviewed by Reddy and Housman, 1997].

As a consequence, the possible presence of a TRE,

Contract grant sponsor: Centre National Recherche Scienti-fique; Contract grant sponsor: Institut National de la Sante et dela Recherche Medicale; Contract grant sponsor: Association Fran-caise contre les Myopathies; Contract grant sponsor: AssistancePublique-Hopitaux de Paris; Contract grant sponsor: Rhone-Poulenc Rorer; Contract grant sponsor: European Commission;Contract grant number: ERBCHBG CT 94-0684.

*Correspondence to: Dr. J. Mallet, LGN, Batiment CERVI, Ho-pital Pitie-Salpetriere, 83 bd. de l’Hopital, 75013 Paris, France.E-mail: suorsoni@infobiogen.fr

Received 26 September 1997; Revised 19 February 1998

American Journal of Medical Genetics (Neuropsychiatric Genetics) 81:338–341 (1998)

© 1998 Wiley-Liss, Inc.

specifically CAG repeats, in the genome of BPAD pa-tients has been sought in a number of studies [Lind-blad et al., 1995; O’Donovan et al., 1995, 1996; Vincentet al., 1996; Oruc et al., 1997]. So far, no firm proof hasbeen put forward; in fact, the data are quite conflicting.We studied a French population with the repeat expan-sion detection (RED) method [Schalling et al., 1993].The RED method is a technique that permits the de-tection of long repeat sequences ù120 base pairs (bp),i.e., 40 triplet repeats, in genomic DNA without theneed for flanking sequence information. When ex-pressed, a TRE involving the triplet CAG codes for anabnormally long stretch of glutamines in the encodedpolypeptide. Therefore, a number of affected individu-als were also examined for expression of polygluta-mines with the Western blot technique and an antibodythat detects long polyglutamine stretches.

MATERIALS AND METHODSPatients and Controls

Patients suffering from bipolar affective disorder(BPAD) were recruited among consecutive admissionsto Hopital Pitie-Salpetriere. They were directly inter-viewed by an experienced psychiatrist with the Frenchversion of the Diagnostic Interview for Genetic Studies(DIGS) [Nurnberger et al., 1994]. One hundred andnineteen unrelated patients (65 women and 54 men)presented symptoms in agreement with DSM-III-R[American Psychiatric Association, 1981] and RDC[Spitzer et al., 1975] criteria for BPAD (85 BPAD typeI and 34 BPAD type II), and were subsequently in-cluded in the study. They were all Caucasians, origi-nating from France. The mean age at the interview was46.1 ± 14.7 years (SD).

Age of onset was determined, in accordance with themedical case notes and the DIGS interview, as the agewhen patients first presented symptoms defined byDSM-III-R, as either a major depressive or manic epi-sode. The patients were divided into three groups: 1)early onset, i.e., before 18 years; 2) late onset, after 40years; and 3) intermediate age of onset, between 18–40years. The sample sizes were, respectively 32, 23, and64 individuals for the three subgroups. The mean ageof onset for all patients was 28.1 ± 12.5 years (SD).

Family history of affective disorder was defined bythe presence of either BPAD I, BPAD II disorder, uni-polar depression, or major depressive episode in atleast one first-degree relative. This was assessed bothwith the proband and during the interview with onefirst-degree relative. The French version of the FamilyInterview of Genetic Studies (FIGS) [Maxwell, 1992]was used. The information was completed, whenneeded, with medical case notes of relatives. Informa-tion concerning family history of affective disorderswas obtained for 80 probands, of which 36 (45%) werefound to have a family history of affective disorder. Thefamilies were not assessed for anticipation. However,the individual with the earliest age of onset within thefamily was selected in each of the six unilineal BPADfamilies for CAG repeat screening with the REDmethod and for polyglutamine expression analysis withWestern blot.

Eighty-eight controls (54 men and 34 women) wererecruited from blood donors (Hopital Pitie-Salpetriere)after assessment with DIGS [Nurnberger et al., 1994].They were all Caucasians of French origin. To mini-mize morbid risk in the control group, only blood do-nors over 35 years old were included (mean age 43.4 ±5.4 years (SD)), as the mean age of onset for BPAD is28.1 years [Goodwin and Jamison, 1990]. Individualswith a personal or family history of affective disordersor suicide attempts revealed by interview were ex-cluded. All individuals included in this study had giveninformed consent.

Repeat Expansion Detection (RED) Analysis

Briefly, the RED technique was performed with aGeneAmp PCR System 9600 (Perkin Elmer Cetus, Nor-walk, CT), using the following conditions. Reactions (10ml) containing 1 mg of genomic DNA (extracted fromperipheral blood lymphocytes), 50 ng of 5’ end-phosphorylated (CTG)10 oligonucleotide (Genset, Evry,France), and 15 U of Ampligase (Epicentre Technolo-gies, Madison, WI) with the supplied Ampligase bufferwere incubated at 94°C for 5 min. Thereafter, sampleswere taken through 495 cycles of 80°C for 30 sec and94°C for 10 sec. Electrophoreses, capillary blotting, andhybridization were performed according to Zander etal. [1997].

All samples were analyzed at least twice. The scoringof RED products was performed by two independantinvestigators who did not know the clinical status ofthe subjects being studied. CEPH individual 134402(Centre d’Etude du Polymorphisme Humain, Paris,France) was included on all gels as a positive control.This individual has earlier been shown to have a 360-bp-long CAG/CTG ligation product (120 repeats)[Schalling et al., 1993]. Approximately 5% of samplesfailed to show a RED ligation product.

Western Blot Analysis

Lymphoblastoid cell lines were established by trans-formation with Epstein-Barr virus from 6 BPAD pa-tients, each from a family segregating BPAD. The fam-ily member with the earliest age of onset of BPAD inthe family was selected. Whole protein was extractedand Western blotting was carried out as described byTrottier et al. [1995]. The monoclonal 1C2 antibody,which recognizes long polyglutamine stretches, wasused. Protein from one patient with spinocerebellarataxia type 3 (SCA3/Machado Joseph disease) servedas a positive control (ataxin, 150 kDa). The lower de-tection limit of the 1C2 antibody is about 33 polyglu-tamines.

Statistical Methods

The Mann-Whitney U test was used for comparisonof CAG repeat length distribution between patientsand controls, as well as between different subsamplesof patients and controls. A nonparametric test was cho-sen, because CAG repeat length distributions are notknown to be normally distributed. The CAG repeatlengths were also analyzed as nominal variables, i.e.,below and above 150 base pairs (bp), using the chi-

CAG Repeat Sequences in BPAD 339

square test. To investigate the correlation between thenumber of base pairs and age at onset, the Spearmanrank correlation coeffficient was calculated. One-tailedP values were used for all tests.

RESULTS

No significant differences were observed in the CAGrepeat length distribution between the 119 patientsand the 88 control individuals (P 4 0.38). The distri-bution and frequencies for patients and controls areshown in Table I. BPAD patients with a positive familyhistory of affective disorders compared to controls didnot show any significant shift (P 4 0.34) toward longerCAG repeats. As the anticipation phenomenon relatedto TRE is known to be associated with more clinicallysevere phenotypes in succeeding generations, we triedto analyze for severe phenotypes and earlier ages ofonset. The patients were thus subgrouped into veryearly onset (onset before age 18 years), BPAD I pa-tients, and BPAD patients having psychotic symptoms,and each of the groups was compared with the controls.However, we could not find any difference between anyof these subgroups and the control group with theMann-Whitney U test (P 4 nonsignificant, range 0.20–0.46, Table I). Nor could we find any sex-specific dif-ferences. Similar results were obtained when using theCAG repeat size as a nominal variable (<150 bp andù150 bp) (data not shown). There was no negative cor-relation between CAG repeat size and age at onsetamong all probands (Spearman test, r 4 +0.16, P 40.08), nor within any of the subgroups described above.

There was no evidence for a specific protein with along polyglutamine tract, as tested with the monoclo-nal antibody 1C2 in the 6 patients, each with a positivefamily history of BPAD. The same patients tested withthe RED method showed 40 repeats (4 individuals) and60 repeats (2 individuals). However, these did not co-segregate with the disorder.

DISCUSSION

Despite the evidence for familial aggregation and ahereditary basis of bipolar affective disorder, little isknown about the underlying genetic mechanisms. Al-though the phenomenon of anticipation has been re-peatedly demonstrated in both bipolar affective disor-

der and schizophrenia, the interpretation of intergen-erational differences in disease onset and severity isdifficult. The possible relation between the presence ofanticipation and TREs has been investigated by sev-eral groups, with inconsistent results.

Our finding that the distribution of CAG repeatlength does not differ between BPAD patients and con-trols is in agreement with that of Vincent et al. [1996]but not those of O’Donovan et al. [1995, 1996], Lind-blad et al. [1995], and Oruc et al. [1997]. In theCroatian study [Oruc et al., 1997], a significant shifttowards longer CAG repeats was observed only for fa-milial BPAD I cases. However, in our study, we wereunable to find any significant differences between thecontrol group and the patient group, even when sub-grouping the patients according to categories for severedisease or early age of onset. Sixty-four percent of allcontrols showed a repeat length of 120 bp, which isslightly less than what was observed in earlier studies[Lindblad et al., 1995; O’Donovan et al., 1995, 1996;Oruc et al., 1997]. This difference could be due to varia-tion in the repeat size distribution profile between dif-ferent populations [Sirugo et al., 1997], and might ex-plain why we do not observe any significant difference.It has been demonstrated that the RED method is ca-pable of detecting even small expansions in Hunting-ton’s disease (repeat size range, 120–180 bp) against abackground of longer repeat sizes, as long as the dis-ease population is homogenous [Hofferbert et al.,1997]. However, the detection limit of the RED methodis 120 bp (40 CAG repeats). If a very short repeat isimplicated in BPAD, as reported in spinocerebellarataxia type 6 (SCA6) [Zhuchenko et al., 1997], we wouldnot be able to detect it with the RED method. The CAGrepeat expansions in patients with SCA 6 range from21–27 repeats, in the human a-1A voltage-dependentcalcium channel gene.

None of the studies published so far has demon-strated a correlation between age of onset and the CAGrepeat length found with RED, as might be expected ifTRE is the molecular basis for anticipation. If TRE isonly responsible for a minor gene effect in certain, butnot all families, that effect could not be detected in anassociation test, with the numbers of subjects analyzedin this study. In conclusion, a long CAG/CTG repeatwas not found to be associated with BPAD in the

TABLE I. Frequency Distribution of CAG Repeats (in Parentheses) and Number of Individuals in Each Group*

CAG repeat length (base pairs) among bipolar probands (n 4 119) and controls (n 4 88)

120 150 180 210 240 270 330 360 450 Total

BPAD I and II (a) 78 (65.5%) 2 (1.7%) 20 (16.8%) 15 (12.6%) 2 (1.7%) 0 1 (0.8%) 0 1 (0.8%) 119BPAD I (b) 52 (61.9%) 2 (2.4%) 15 (17.8%) 13 (15.5%) 1 (1.2%) 0 1 (1.2%) 0 0 84Family history (c) 25 (69.4%) 4 (11.1%) 4 (11.1%) 1 (2.8%) 0 0 0 1 (2.8%) 0 36Psychotic features (d) 33 (73.3%) 1 (2.2%) 6 (13.3%) 4 (8.9%) 0 0 1 (2.2%) 0 0 45Early onset (e) 22 (68.7%) 2 (6.2%) 4 (12.5%) 3 (9.4%) 1 (3.1%) 0 0 0 0 32Women only (f) 42 (64.6%) 0 12 (18.5%) 10 (15.4%) 1 (1.5%) 0 0 0 0 65Men only (g) 36 (66.7%) 2 (3.7%) 8 (14.8%) 5 (9.2%) 1 (1.8%) 0 1 (1.8%) 0 1 (1.8%) 54

Controls (h) 56 (63.6%) 2 (2.3%) 15 (17%) 8 (9.1) 4 (4.5%) 1 (1.1%) 0 1 (1.1%) 1 (1.1%) 88Women (i) 25 (73.5%) 0 4 (11.8%) 2 (5.9%) 2 (5.9%) 1 (2.9%) 0 0 0 34Men (j) 31 (57.4%) 2 (3.7%) 11 (20.4%) 6 (11.1%) 2 (3.7%) 0 0 1 (1.8%) 1 (1.8%) 54

*Mann-Whitney U test: (a) vs. (h), P 4 0.38; (b) vs. (h), P 4 0.46; (c) vs. (h), P 4 0.34; (d) vs. (h), P 4 0.16; (e) vs. (h), P 4 0.27; (f) vs. (i), P 4 0.28; (g)vs. (j), P 4 0.20.

340 Zander et al.

French population sample studied. A minor effect dueto an expanded CAG repeat (or another TRE) or a rela-tively short CAG expansion cannot be ruled out.

ACKNOWLEDGMENTS

The authors thank Y. Trottier, L. Tora, and Y. Lutzfor providing the 1C2 antibody. C.Z. is a recipient offellowship ERBCHBG CT 94-0684 from the EuropeanCommission.

REFERENCESAmerican Psychiatric Association (1981): ‘‘American Psychiatric Associa-

tion Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R).’’ Washington, DC: American Psychiatric Association.

Goodwin FK, Jamison KR (1990): ‘‘Manic Depressive Illness.’’ New York:Oxford University Press, p 938.

Grigoroiu-Serbanescu M, Wickramaratne PJ, Hodge SE, Milea S, Mihai-lescu R (1997): Genetic anticipation and imprinting in bipolar I illness.Br J Psychiatry 170:162–166.

Hofferbert S, Schanen NC, Chehab F, Francke U (1997): Trinucleotiderepeats in the human genome: Size distributions for all possible tripletsand detection of expanded disease alleles in a group of Huntingtondisease individuals by the repeat expansion detection method. HumMol Genet 6:77–83.

Lindblad K, Nylander PO, De Bruyn A, Sourey D, Zander C, Engstrom C,Holmgren G, Hudson T, Chotai J, Mendlewicz J, van Broeckhoven C,Schalling M, Adolfsson R (1995): Detection of expanded CAG repeats inbipolar affective disorder using the repeat expansion detection (RED)method. Neurobiol Dis 2:55–62.

Maxwell ME (1992): ‘‘Family Interview for Genetic Studies.’’ Clinical Neu-rogenetics Branch, Intramural Research Program, NIMH.

McInnis MG, McMahon FJ, Chase GA, Simpson SG, Ross CA, DePaulo JRJr (1993): Anticipation in bipolar affective disorder. Am J Hum Genet53:385–390.

Nurnberger JI Jr, Blehar MC, Kaufmann CA, York-Cooler C, Simpson SG,Harkavy-Friedman J, Severe JB, Malaspina D, Reich T (1994): Diag-nostic interview for genetic studies. Rationale, unique features, andtraining. NIMH Genetics Initiative. Arch Gen Psychiatry 51:849–859,863–864.

Nylander PO, Engstrom C, Chotai J, Wahlstrom J, Adolfsson R (1994):Anticipation in Swedish families with bipolar affective disorder. J MedGenet 31:686–689.

O’Donovan MC, Guy C, Craddock N, Murphy KC, Cardno AG, Jones LA,Owen MJ, McGuffin P (1995): Expanded CAG repeats in schizophreniaand bipolar disorder. Nat Genet 10:380–381.

O’Donovan MC, Guy C, Craddock N, Bowen T, McKeon P, Macedo A, MaierW, Wildenauer D, Aschauer HN, Sorbi S, Feldman E, Mynett-JohnsonL, Claffey E, Nacmias B, Valente J, Dourado A, Grassi E, Lenzinger E,Heiden AM, Moorhead S, Harrison D, Williams J, McGuffin P, OwenMJ (1996): Confirmation of association between expanded CAG/CTGrepeats and both schizophrenia and bipolar disorder. Psychol Med 26:1145–1153.

Oruc L, Lindblad K, Verheyen GR, Ahlberg S, Jakovljevic M, Ivezic S,Raeymaekers P, van Broeckhoven C, Schalling M (1997): CAG repeatexpansions in bipolar and unipolar disorders. Am J Hum Genet 60:730–732.

Petronis A, Kennedy JL (1995): Unstable genes—Unstable mind? Am JPsychiatry 152:164–172.

Reddy PS, Housman DE (1997): The complex pathology of trinucleotiderepeats. Curr Opin Cell Biol 9:364–372.

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

Schalling M, Hudson TJ, Buetow KH, Housman DE (1993): Direct detec-tion of novel expanded trinucleotide repeats in the human genome. NatGenet 4:135–139.

Sirugo G, Deinard AS, Kidd JR, Kidd KK (1997): Survey of maximumCTG/CAG repeat lengths in humans and non-human primates: Totalgenome scan in populations using the Repeat Expansion Detectionmethod. Hum Mol Genet 6:403–408.

Spitzer RL, Endicott J, Robins E (1975): Research diagnostic diagnosticcriteria. Psychopharmacol Bull 11:22–25.

Trottier Y, Lutz Y, Stevanin G, et al. (1995): Polyglutamine expansion as apathological epitope in Huntington’s disease and four dominant cer-ebellar ataxias. Nature 378:403–406.

Vincent JB, Klempan T, Parikh SS, Sasaki T, Meltzer HY, Sirugo G, ColaP, Petronis A, Kennedy JL (1996): Frequency analysis of large CAG/CTG trinucleotide repeats in schizophrenia and bipolar affective dis-order. Mol Psychiatry 1:141–148.

Zander C, Burgess CE, Johansen J, Lindblad K, Hudson T, Schalling M(1997): Repeat expansion detection. In Taylor GR (ed): ‘‘LaboratoryMethods for the Detection of Mutations and Polymorphisms in DNA.’’Boca Raton, FL: CRC Press, Inc., pp 263–271.

Zhuchenko O, Bailey J, Bonnen P, Ashizawa T, Stockton DW, Amos C,Dobyns WB, Subramony SH, Zoghbi HY, Lee CC (1997): Autosomaldominant cerebellar ataxia (SCA6) associated with small polygluta-mine expansions in the alpha 1A-voltage-dependent calcium channel.Nat Genet 15:62–69.

CAG Repeat Sequences in BPAD 341