Chromosome 10q harbors a susceptibility locus for bipolar disorder in Ashkenazi Jewish families

(2008) 13, 442 450 & 2008 Nature Publishing Group All rights reserved 1359-4184/08 $30.00 www.nature.com/mp ORIGINAL ARTICLE Chromosome 10q harbors a susceptibility locus for bipolar disorder in Ashkenazi Jewish families T Venken 1,2, M Alaerts 1,2, D Souery 3, D Goossens 1,2, S Sluijs 1,2, R Navon 4, C Van Broeckhoven 1,2, J Mendlewicz 3, J Del-Favero 1,2 and S Claes 1,5 1 Department of Molecular Genetics, Flanders Institute for Biotechnology VIB, Antwerpen, Belgium; 2 University of Antwerp, Antwerpen, Belgium; 3 Department of Psychiatry, Erasme Hospital, University of Brussels, Brussels, Belgium and 4 Department of Human Molecular Genetics and Biochemistry, Sackler School of Medicine, University of Tel Aviv, Tel Aviv, Israel We report the results of a 10 cm density genome-wide scan and further fine mapping of three chromosomal candidate regions in 10 Belgian multigenerational families with bipolar (BP) disorder. This two-stage approach revealed significant evidence for linkage on chromosome 10q21.3-10q22.3, showing a maximum multipoint parametric heterogeneity logarithm of odds (HLOD) score of 3.28 and a nonparametric linkage (NPL) score of 4.00. Most of the chromosome 10q evidence was derived from a single, large Ashkenazi Jewish pedigree. Haplotype analysis in this pedigree shows that the patients share a 14-marker haplotype, defining a chromosomal candidate region of 19.2 cm. This region was reported previously as a candidate region for BP disorder in several independent linkage analysis studies and in one large meta-analysis. It was also implicated in a linkage study on schizophrenia (SZ) in Ashkenazi Jewish families. Additionally, we found suggestive evidence for linkage on chromosome 19q13.2-13.4 (HLOD 2.01, NPL 1.09) and chromosome 7q21-q22 (HLOD 1.45, NPL 2.28). Together, these observations suggest that a gene located on chromosome 10q21.3-10q22.3 is underlying the susceptibility both for SZ and for BP disorder in at least the Ashkenazi Jewish population. (2008) 13, 442 450; doi:10.1038/sj.mp.4002039; published online 19 June 2007 Keywords: linkage analysis; affective disorders; chromosome 10; psychiatric genetics Introduction Bipolar (BP) disorder (MIM 125480) is a chronic psychiatric disorder, and is characterized by disturbances in mood, ranging from mania to a severe state of depression. Patients with BP disorder type I (BPI) suffer from severe phases of mania, while in case of type II (BPII) the manic episodes are less severe (called hypomania). BP disorder is part of a spectrum of affective disorders also including recurrent unipolar depression patients only experiencing recurrent depressive episodes (UPR), major depression where only a single depressive episode has occurred (UPS), cyclothymia (CYC) alternating minor depressive and hypomanic episodes, and schizoaffective disorder of the manic (SAM) and depressive (SAD) type, which also exhibit symptoms compatible with mood disorders. Affecting 0.5 1.5% of the world Correspondence: Professor Dr J Del-Favero, Applied Molecular Genomics Group, Department of Molecular Genetics, VIB, University of Antwerp, Campus CDE, Parking P4, Building V, Room 1.14, Universiteitsplein 1, BE-2610 Antwerpen, Belgium. E-mail: jurgen.delfavero@ua.ac.be 5 Current address: Department of Psychiatry, University Hospital Gasthuisberg, Leuven, Belgium. Received 10 January 2007; revised 19 April 2007; accepted 7 May 2007; published online 19 June 2007 population, BP disorder is among the most common disabling brain diseases, constituting a major public health problem with a high rate of morbidity and mortality. 1 Although the etiology and pathophysiology is still unknown, twin, adoption and family studies support a strong genetic determination. These studies indicate that there is 60 85% heritability for BP disorder and that relatives of BP probands have an increased risk for other psychiatric disorders among the spectrum. 2 The exact pattern of transmission, the specific number of susceptibility loci, the recurrence risk ratio attributable to each locus and the degree of the interaction between loci are unknown. The pattern of inheritance is therefore complex, probably involving both multiple genes and environmental factors. This complex genetic etiology of BP disorder might explain the numerous linkage signals throughout the genome that have been reported by independent research groups. Nevertheless, some genomic regions have gained consistent support from different studies, and are therefore likely to contain BP disorder susceptibility loci. 3 5 We previously ascertained 15 Belgian families through BP probands, and multiple family members were diagnosed with phenotypes belonging to the spectrum of affective disorders. These families have

earlier been analyzed using both candidate gene and loci approaches with variable results. 6 9 One family showed suggestive linkage to chromosome 18q21.3- q22, 8,10 and further genetic and physical mapping studies have resulted in the identification of a putative candidate gene for BP disorder at 18q21.3- q22. 11 Ten of the 15 families were selected for inclusion in a genome-wide scan for BP disorder, of which two were of Ashkenazi Jewish and eight of European Caucasian descent. 6,12 Here, we present significant results from the genome-wide scan and subsequent fine-mapping analyses conducted in these 10 multigenerational families with several patients diagnosed with affective disorders. Materials and methods Samples As part of the Scientific Program on Molecular Neurobiology of Mental Illness, funded by the European Science Foundation (ESF), 10 of 15 Belgian families segregating affective disorders were selected for inclusion in a genome scan (described in detail by De bruyn et al. 6 and Mendlewicz et al. 12 ). Eight of the 10 families were of European Caucasian origin while two families (families 15 and 22) were of Ashkenazi Jewish descent (Supplementary Figure A1). The families were selected for inclusion in the ESF study based on information content, number of diagnosed patients, diagnostic instruments used and number of family members with DNA and/or cell lines available. All families were ascertained at the Department of Psychiatry, Erasme Hospital in Brussels. In each family, the proband was affected with BPI or BPII disorder. All individuals willingly to cooperate were X18 years and gave informed consent after approval of the study by the Medical Ethical Committee of the University of Brussels. Medical records were available of all patients who were examined by fully trained psychiatrists using the schedule for affective disorders and schizophrenia lifetime version modified for the study of anxiety disorders (SADS-LA). 13 Diagnoses were made based on the Research Diagnostic Criteria. 14 Two independent psychiatrists diagnosed the patients and in case of discordant diagnoses, a third experienced psychiatrist made the final diagnosis. In total, 156 persons from the 10 families were interviewed. The following affective spectrum diagnoses were obtained in 70 patients (Supplementary Table A1): BPI (15), BPII (15), recurrent depression (UPR, 25), single major depression (UPS, 9), CYC (4), schizoaffective disorder manic type (SAM, 1) and schizoaffective disorder depressed type (SAD, 1). In 83 individuals, the structured interview revealed no lifetime history of psychiatric disorders and, therefore, they were considered healthy. Three individuals obtained a diagnosis of other psychiatric disorders. In 20 individuals, no interview could be obtained. One spouse was diagnosed with single major depression. A sample of 92 unrelated healthy individuals of Ashkenazi Jewish origin were randomly ascertained at Meir Hospital, Kfar-Sabba, Israel (described in detail by Korostishevsky et al. 15 ). The gender ratio of this group was 54 females/38 males and the mean age 31.3174.9. The Medical Ethical Committees of the University of Brussels, the Meir Hospital and the University of Antwerp have approved the inclusion of the patients, relatives and unrelated healthy individuals in genetic studies. Genotyping From each participant a venous blood sample was obtained for genomic DNA extraction, using standard procedures. DNA of 176 individuals from the 10 families was sent to Généthon (Evry, France), where a 10 cm density genome-wide scan was performed using 387 fluorescent-labeled simple tandem repeat (STR) markers. Based on genetic distances (gender averaged) of the comprehensive Marshfield genetic map (http://www.research.marshfieldclinic.org/genetics), the mean marker distance was 9.8 cm. Genetic fine mapping of the chromosomal candidate regions was performed with STR markers selected from the Marshfield or the decode Icelandic genetic maps. 16 Additionally two novel STR markers on chromosome 19, identified using the Sputnik program (http://espressosoftware.com/pages/sputnik. jsp3), were genotyped: M241797, a CA-repeat marker starting at genomic position chr19:52665418 with nine alleles and 81% heterozygosity, and M1519374, a CA-repeat marker starting at genomic position chr19:53942995 with six alleles and 68% heterozygosity (genomic positions according to the Human March 2006 Assembly of the UCSC Genome Browser). In total, 23 additional STR markers were genotyped between markers D10S1791 and D10S201 on chromosome 10q, 15 STR markers between D7S2516 and D7S486 on chromosome 7q and 10 STR markers between D19S220 and D19S210 on chromosome 19q, to obtain an average marker map density between 1 and 2 cm in the candidate regions. All PCR amplifications were performed under multiplex standard reaction conditions in a total volume of 20 ml ona Primus HT (MWG AG Biotech). PCR products were sized on an ABI 3700 or ABI 3730 automated Sequencer (Applied Biosystems) and genotypes were assigned using Genotyper version 2.1 or Genemapper version 3.0 respectively. Also 27 STR markers, spanning the linked region on chromosome 10, were genotyped in 92 healthy Ashkenazi Jewish individuals to estimate allele frequencies in the Ashkenazi Jewish population. Statistical analysis Before conducting the genome-wide scan, the genetic information content of the families was estimated with the software program SLINK. 17,18 We performed four simulation studies with 10 000 replicates each under dominant and recessive models with the disease gene located at recombination fraction 443

444 y = 0.05 from a simulated marker, with eight equally frequent alleles, assuming genetic homogeneity. An average maximum logarithm of odds (LOD) score Z of 3.41 was obtained under a dominant model and 1.86 under a recessive model assuming genetic homogeneity. There was 50 and 82% power to obtain significant (LOD > 3.3) or suggestive (LOD > 1.9) evidence for linkage, respectively, according to the criteria of Lander and Kruglyak. 19 Errors in the allele calling were checked using the PedCheck program 20 and Mendelian inconsistencies were resolved; the overall error rate was 0.39%. Twopoint parametric linkage analysis was performed using the MLINK program 21 from the FASTLINK computer package (version 5.1). 22 Multipoint nonparametric allele sharing statistics (NPL pair and NPL all scores), multipoint parametric linkage analysis and haplotype reconstruction was carried out using Simwalk2 version 2.91. 23 The format of the input files for Simwalk2 was provided by Mega2 version 3.0 R9. 24 The marker map used for the multipoint analysis was based on the Marsheld genetic map. Linkage analysis was performed under an affected-only model, with patients phenotypes belonging to the affective spectrum disorders (except patients with a single major depression). Relatives without psychiatric symptoms, with other psychiatric disorders or without interview were considered unknown in the statistical analysis. The phenocopy rate was defined differently for BPI and SAM (1%), BPII or CYC (2%), and UPR or SAD (5%), in accordance with the population frequencies of these disorders. The disease allele frequency in the general population was set at 0.01 and 0.10 for the dominant and recessive models respectively. STR allele frequencies were estimated on the basis of married-in individuals in the pedigree. STR allele frequencies were also estimated on the basis of 92 healthy unrelated Ashkenazi Jewish individuals in the linked region on chromosome 10. Results Genome scan Linkage analysis resulted in two-point LOD scoresx1 with 15 markers located on nine different chromosomes: 3p, 3q, 6q, 7q, 9p, 10q, 12q, 17p, 19q and 22q (Table 1). Only on chromosome 10, three neighboring markers, D10S581, D10S537 and D10S1699 showed LOD scoresx1 under the same model. The multipoint linkage plots for nonparametric and parametric (under dominant and recessive models) affected-only analyses are shown per chromosome in Supplementary Figure A2. On 10 chromosomal regions, 3q, 6q, 7q, 9p, 10p, 10q, 13p, 15q, 17q and 19q, heterogeneity logarithm of odds (HLOD) scoresx1 were calculated, of which chromosomes 10q and 19q showed suggestive evidence for linkage (Table 2). The highest HLOD score of 2.3 was obtained on chromosome 10q under a dominant model, corresponding to a candidate region of 34 cm ranging from 69 to 103 cm from pter. In this region, the maximal NPL score was 1.77. A second locus of 20 cm was found on chromosome 19q, between 62 and 82 cm from pter, with a maximum HLOD score of 2.14 under a recessive model. In this region, a maximal NPL score of 1.23 was calculated. A third locus on chromosome 7q yielded a dominant HLOD score of 1.77, and a maximal NPL score of 2.03, within a candidate region of 40 cm (102 142 cm from pter). Fine mapping The chromosome 7q, 10q and 19q regions were further analyzed using additional STR markers selected from public genetic maps to obtain an overall Table 1 Two-point linkage results (MLINK) of the genome scan under dominant and recessive models, for markers showing a LOD scorex1 (indicated in bold) in at least one model Marker cm from pter Dominant Recessive Z max y Z max y D3S1266 53 0.10 0.40 1.01 0 D3S1282 181 1.86 0.10 0.79 0 D6S286 90 0.36 0.20 1.05 0.05 D7S635 130 0.13 0.30 2.45 0 D9S288 10 0.03 0.40 1.11 0.10 D10S581 83 2.53 0.01 1.65 0 D10S537 91 1.45 0.10 0.65 0.20 D10S1699 97 1.17 0.10 0.18 0.20 D12S85 61 1.46 0.10 0.41 0.10 D12S92 83 1.19 0.20 0.73 0.10 D12S366 133 0.10 0.40 1.20 0.05 D12S342 145 1.28 0.10 0.98 0.10 D17S1866 0 0.29 0.20 1.12 0 D19S412 70 0.16 0.30 1.16 0.10 D22S283 39 0.56 0.10 1.45 0 Abbreviation: LOD, logarithm of odds.

Table 2 Genome-wide multipoint parametric heterogeneity LOD scores (HLOD) with corresponding a-value and nonparametric maximum linkage scores (NPL pair and NPL all ). LOD scores > 1 are indicated in bold 445 Locus LOD > 1 region (cm) Parametric linkage Nonparametric linkage Peak Start end Size Dominant Recessive NPL pair NPL all HLOD a HLOD a 3q 177.8 169 182 13 1.515 0.8 0.168 0.3 0.933 1.049 6q 88.4 86 93 7 0.049 0.15 1.074 0.65 1.126 0.783 7q 112.3 102 142 40 1.769 0.6 1.346 0.75 2.027 1.983 9p 0 0 5 5 0.005 0.05 1.168 1 0.533 0.508 10p 16.9 13.5 18 4.5 0.524 0.4 1.106 0.7 1.386 1.144 10q 76.7 69 103 34 2.302 0.7 1.467 1 1.244 1.767 13p 0 0 20 20 1.091 1 < 0 0 0.698 0.779 15q 72.9 72 73 1 1.01 0.5 < 0 0 0.722 1.157 17q 62.0 56 74 18 1.455 0.4 0.529 0.35 0.748 0.702 19q 73.3 62 82 20 0.302 0.25 2.136 1 1.237 0.675 Abbreviations: HLOD, heterogeneity logarithm of odds; LOD, logarithm of odds; NPL, nonparametric linkage. The localizations of the regions, the linkage peaks within these regions and the size of these regions are given as distances in cm from pter of the chromosome (based on the Marshfield genetic map). a is the proportion of linked families. marker density between 1 and 2 cm (Figure 1). All the fine mapped regions gained further support from additional STR markers in the two-point analysis (Supplementary Table A2), with a highest LOD score of 2.89 obtained for marker D10S210 at 86.81 cm under a dominant model. Multipoint linkage analysis on chromosome 10q resulted in a maximum HLOD score of 3.28 and a maximum NPL score of 4 at 84.7 cm from pter in a candidate region of 10 cm between D10S1225 and D10S537 (79 89 cm; Figure 1a). Family 22, one of the Ashkenazi Jewish families, contributed most to the linkage peak at chromosome 10. A maximal LOD score of 2.66 and a maximal NPL score of 3.97 at 85 cm was obtained for this family alone (Table 3). Further contributions came from families 24, 45, 15, 32 and 37. On chromosome 19q, a maximum HLOD score of 2.01 and a maximum NPL score of 1.09 at 78.1 cm from pter was obtained in a candidate region of 22 cm between D19S220 and D19S921 (63 85 cm; Figure 1b). Further fine-mapping analysis on chromosome 7q showed a maximum HLOD score of 1.45 and a maximum NPL score of 2.28 at 112.3 cm from pter in a candidate region of 8 cm between D7S2782 and D7S2459 (110 118 cm; Figure 1c). Detailed analysis of chromosome 10q in Ashkenazi Jewish families The Ashkenazi Jewish family 22 contributed most to the linkage peak at chromosome 10q (Table 3). To obtain more specific allele frequencies for the markers in the chromosome 10q candidate region for the Ashkenazi Jewish population, 27 STR markers between D10S1791 and D10S201 were genotyped in 92 healthy Ashkenazi Jewish individuals. The obtained allele frequencies were subsequently used to recalculate the linkage analysis in both Ashkenazi Jewish families (families 15 and 22). The highest two-point LOD score of 2.32 was observed in family 22 at y = 0.0 with marker D10S1647 under the dominant model, while in the original analysis, a LOD score of 2.2 was obtained at the same marker under the same model (data not shown). In the multipoint linkage analysis, a maximal LOD score of 2.63 and a maximal NPL score of 3.70 was found in family 22 at 87 cm from pter (Figure 2). In this family, a second linkage peak was seen at 98 cm with an LOD score of 1.73 and an NPL score of 1.58. Segregation analysis in family 22 showed that 12 of the 14 patients share the same 14-marker haplotype (Figure 3). Meiotic recombinations were observed in patient 22.21 between D10S581 and D10S522, and in patients 22.23 and 22.28 between D10S1677 and D10S201. Additionally, two recombination events occurred between patients 22.23 and 22.17, further decreasing the candidate region to 6.7 cm between D10S522 centromeric (84.4 cm from pter) and D10S537 telomeric (91.1 cm from pter), which is in line with the location of highest linkage peak. Discussion Genome-wide linkage analysis and subsequent fine mapping of three candidate loci at 10q, 19q and 7q showed a maximum two-point or multipoint LOD scorex1.9 in 10 Belgian multigenerational BP families. After fine mapping with 1 2 cm STR density, the evidence for the chromosome 10q21.3-10q22.3 locus increased, and reached the level of genomewide significant evidence (LOD scorex3.3 or NPL scorex3.6) for linkage with a maximum HLOD score of 3.28 and a maximum NPL score of 4. The evidence for linkage to chromosomes 19q13.1-q13.4 and 7q21- q22 was confirmed but remained suggestive (LOD

446 Figure 1 Multipoint linkage analyses of the genetic fine mapping in the 10q (a), 19q (b) and 7q (c) chromosomal regions. The black curve represents heterogeneity logarithm of odds (HLOD) scores under the dominant model; the black-dashed curve under the recessive model; the gray curve represents nonparametric linkage (NPL) all scores; and the gray-dashed curve NPL pair scores. Simple tandem repeat (STR) markers included in the original genome-wide scan are indicated in bold italic. Genetic localization of the markers was based on the Marsheld genetic map.

Table 3 Fine-mapping results of chromosomal loci 10q, 19q and 7q 447 Locus Peak Model Parametric multipoint linkage Nonparametric linkage 3 15 20 22 24 30 31 32 37 45 HLOD a NPL pair NPL all 10q 84.73 D 0.69 0.42 0.88 2.66 1.00 0.52 1.37 0.28 0.25 0.78 3.28 0.55 3.35 4 19q 78.10 R 0.29 0.21 0.64 0.25 0.23 0.73 0.18 0.32 0.16 0.82 2.01 1.00 1.09 0.959 7q 112.33 D 0.62 0.72 0.59 1.41 1.05 0.83 0.48 0.86 0.53 0.83 1.45 0.50 2.23 2.28 Abbreviations: HLOD, heterogeneity logarithm of odds; LOD, logarithm of odds; NPL, nonparametric linkage. The multipoint LOD scores of each family, as well as the maximum HLOD scores of the families in total, a-values and the model under which it was obtained (D, dominant; R, recessive) are given. Also the NPL scores at the linkage peak are given. The localizations of the peaks within these regions are given as distances in cm from pter of the chromosome (based on the Marshfield genetic map). The two Ashkenazi Jewish families, families 15 and 22, are indicated in bold. Figure 2 Parametric multipoint linkage analysis (dominant model) of chromosome 10q in family 22, representing heterogeneity logarithm of odds (HLOD) scores with allele frequencies calculated from the Ashkenazi Jewish population (black curve) and allele frequencies calculated from founders and married-in individuals (gray curve). scorex1.9 or NPL scorex2.2), showing maximum HLOD scores of 2.01 and 1.45, and maximum NPL scores of 1.09 and 2.28, respectively. Relatively, few studies reported positive linkage findings with psychiatric disorders on chromosome 19q 25,26 and chromosome 7q. 27 29 Therefore, further investigation and replication studies are necessary to confirm these findings. The significant linkage findings on 10q were mainly derived from one large Ashkenazi Jewish family (family 22). This family by itself generated a multipoint LOD score of 2.66. Segregation analysis in this family confirmed co-segregation of a 14-marker haplotype between markers D10S581 and D10S201 with BP disorder. This haplotype was shared by 12 of the 14 patients studied, defining a candidate region of 19.2 cm. Additionally, two recombination events occurred between patients 22.23 and 22.17, further decreasing the candidate region to 6.7 cm between D10S522 centromeric (84.4 cm from pter) and D10S537 telomeric (91.1 cm from pter). Patient 22.22 diagnosed with CYC does not share the disease haplotype. This might be attributed to the complex diagnosis of affective disorder spectrum phenotypes with difficulties in classification and specification of spectrum boundaries, as well as to the genetic complexity of the disorder. Since CYC is a mild form of BP disorder, characterized by alternating minor depressive and hypomanic episodes, this patient could have been affected due to other factors unrelated to the genetic defect on chromosome 10q in the family. Fallin and colleagues 30 specifically studied linkage in 41 Ashkenazi Jewish families with BP disorder. In this study marker D10S185, slightly telomeric from our candidate region, reached a NPL score of 1.97 (P = 0.02). However, this score did not reach the threshold the authors set for further fine mapping. Interestingly, in a linkage study on schizophrenia (SZ) of the same research group in 29 Ashkenazi Jewish families, significant evidence for linkage was observed for a region of 12.2 Mb between markers D10S1677 and D10S1753 on 10q22. 31 This region is situated just telomeric from the 14-marker haplotype

448 Figure 3 Segregation analysis of chromosome 10q in family 22. Filled symbols represent patients. The proband is indicated with an arrow. Individuals with DNA available are indicated with an asterisk. Only the haplotypes of the affected family members are shown, of which the alleles on the disease haplotype are boxed. BPI, bipolar I disorder; CYC, cyclothymia; UPR, recurrent unipolar disorder. observed in family 22. Given the differences in methods, family numbers and family structures, these linkage signals could potentially represent the same underlying susceptibility gene. Moreover, the current Ashkenazi Jewish population, living mostly in Central and Eastern Europe and the United States, descended from a small founder population about 500 years ago. 32 Since reduced genetic variation is expected in this population, a common susceptibility gene on chromosome 10q22 for both BP disorder and SZ might be present in this isolated population. Linkage of BP disorder and SZ to 10q has also been observed in several other ethnic populations, spread over a region ranging from 10q21 to 10q26, indicating that a gene or a set of genes at 10q might attribute considerably to the risk for psychiatric disease. 29,33 43 In two of these studies, linkage was found in families, of which some were originating from Israel. 35,37 However, it was not clear if these families were from Ashkenazi Jewish descent. Additional evidence for 10q comes from a rankbased genome scan meta-analysis of 18 BP disorder genome scans. 4 Although none of the linked regions reached the criteria for genome-wide significance, the most significant loci (P-values < 0.01) were observed on chromosomes 9p22.3-p21.1, 10q11.21-q22.1, 14q24.1-q32.12 and regions on chromosome 18. The chromosome 10 candidate region in this meta-analysis ranged from 62 to 91 cm based on the Marshfield genetic map, overlapping with the region found in our study. The 10q candidate region between D10S581 and D10S201 (chr10:65,619,520-70,598,997) comprises 95 known genes, according to the RefSeq track of the UCSC Genome Browser (Supplementary Table A3). One of these genes is TACR2 that encodes the tachykinin receptor 2, a G-protein-coupled receptor, which is part of the inositol phosphatase transduction pathway. Antagonists of tachykinin receptors are reported to have antidepressant-like and anxiolytic effects in rodents. 44 46 Another interesting candidate gene is GRID1, which is located more telomeric from the 14-marker haplotype on chromosome 10q22 (chr10:87,349,292-88,116,230). Recently, a positive association was found for GRID1 with SZ as well as with BPI disorder and SZ/SA combined in an Ashkenazi Jewish population. 47 GRID1 might be involved in glutamatergic signaling, since it is highly related to ionotropic glutamate receptor subunits. In conclusion, we provided significant evidence for a risk factor on chromosome 10q21.3-10q22.3 contributing in the predisposition of BP disorder in a large multigenerational Ashkenazi Jewish family. This candidate locus is strongly supported to be involved in the susceptibility for psychiatric disorders by several independent studies. Further investigation of this susceptibility locus in both the Ashkenazi Jewish family and in population-based association studies in Ashkenazi Jews as well as in populations of other descent will enable to identify a susceptibility gene for psychiatric disorders. Acknowledgments We thank the patients and their relatives for their cooperation and participation in this research study. The genome-wide genotyping was performed at Généthon (Evry, France) within the framework of

the Scientific Program on Molecular Neurobiology of Mental Illness financed by the European Science Foundation. We acknowledge the contribution of the personnel of the VIB Genetic Service Facility (http:// www.vibgeneticservicefacility.be/) to the genetic analyses. The Interuniversity Attraction Poles program P5/19 of the Federal Science Policy Office, the Fund for Scientific Research Flanders (FWO), and the Special Research Fund of the University of Antwerp (Belgium) have funded this work. MA has a PhD fellowship of FWO. SC is a Senior Clinical Researcher of the FWO. References 1 Muller-Oerlinghausen B, Berghofer A, Bauer M. Bipolar disorder. Lancet 2002; 359: 241 247. 2 Smoller JW, Finn CT. Family, twin, and adoption studies of bipolar disorder. Am J Med Genet 2003; 123C: 48 58. 3 Badner JA, Gershon ES. Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Mol Psychiatry 2002; 7: 405 411. 4 Segurado R, Detera-Wadleigh SD, Levinson DF, Lewis CM, Gill M, Nurnberger Jr JI et al. Genome scan meta-analysis of schizophrenia and bipolar disorder, part III: bipolar disorder. Am J Hum Genet 2003; 73: 49 62. 5 McQueen MB, Devlin B, Faraone SV, Nimgaonkar VL, Sklar P, Smoller JW et al. Combined analysis from eleven linkage studies of bipolar disorder provides strong evidence of susceptibility loci on chromosomes 6q and 8q. Am J Hum Genet 2005; 77: 582 595. 6 De bruyn A, Mendelbaum K, Sandkuijl LA, Delvenne V, Hirsch D, Staner L et al. Nonlinkage of bipolar illness to tyrosine hydroxylase, tyrosinase, and D2 and D4 dopamine receptor genes on chromosome 11. Am J Psychiatry 1994; 151: 102 106. 7 De bruyn A, Raeymaekers P, Mendelbaum K, Sandkuijl LA, Raes G, Delvenne V et al. Linkage analysis of bipolar illness with X- chromosome DNA markers: a susceptibility gene in Xq27-q28 cannot be excluded. Am J Med Genet 1994; 54: 411 419. 8 De bruyn A, Souery D, Mendelbaum K, Mendlewicz J, Van Broeckhoven C. Linkage analysis of families with bipolar illness and chromosome 18 markers. Biol Psychiatry 1996; 39: 679 688. 9 De bruyn A, Souery D, Mendelbaum K, Mendlewicz J, Van Broeckhoven C. A linkage study between bipolar disorder and genes involved in dopaminergic and GABAergic neurotransmission. Psychiatr Genet 1996; 6: 67 73. 10 Verheyen GR, Villafuerte SM, Del Favero J, Souery D, Mendlewicz J, Van Broeckhoven C et al. Genetic refinement and physical mapping of a chromosome 18q candidate region for bipolar disorder. Eur J Hum Genet 1999; 7: 427 434. 11 Goossens D, Van Gestel S, Claes S, De Rijk P, Souery D, Massat I et al. A novel CpG-associated brain-expressed candidate gene for chromosome 18q-linked bipolar disorder. Mol Psychiatry 2003; 8: 83 89. 12 Mendlewicz J, Leboyer M, De bruyn A, Malafosse A, Sevy S, Hirsch D et al. Absence of linkage between chromosome 11p15 markers and manic-depressive illness in a Belgian pedigree. Am J Psychiatry 1991; 148: 1683 1687. 13 Spitzer RL, Endicott JA. The Schedule for Affective Disorders and Schizophrenia Life Time Version. New York State Psychiatric Institute: New York, 1977. 14 Spitzer RL, Endicott JA, Robins E. Research diagnostic criteria: rationale and reliability. Arch Gen Psychiatry 1978; 35: 773 782. 15 Korostishevsky M, Kaganovich M, Cholostoy A, Ashkenazi M, Ratner Y, Dahary D et al. Is the G72/G30 locus associated with schizophrenia? Single nucleotide polymorphisms, haplotypes, and gene expression analysis. Biol Psychiatry 2004; 56: 169 176. 16 Kong A, Gudbjartsson DF, Sainz J, Jonsdottir GM, Gudjonsson SA, Richardsson B et al. A high-resolution recombination map of the human genome. Nat Genet 2002; 31: 241 247. 17 Ott J. Computer-simulation methods in human linkage analysis. Proc Natl Acad Sci USA 1989; 86: 4175 4178. 18 Weeks DE, Ott J, Lathrop GM. SLINK: a general simulation program for linkage analysis. Am J Hum Genet 1990; 47: A204. 19 Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 1995; 11: 241 247. 20 O Connell JR, Weeks DE. PedCheck: a program for identification of genotype incompatibilities in linkage analysis. Am J Hum Genet 1998; 63: 259 266. 21 Lathrop GM, Lalouel JM. Easy calculations of lod scores and genetic risks on small computers. Am J Hum Genet 1984; 36: 460 465. 22 Cottingham RW, Idury RM, Schaffer AA. Faster sequential genetic linkage computations. Am J Hum Genet 1993; 53: 252 263. 23 Sobel E, Lange K. Descent graphs in pedigree analysis: applications to haplotyping, location scores, and marker-sharing statistics. Am J Hum Genet 1996; 58: 1323 1337. 24 Mukhopadhyay N, Almasy L, Schroeder M, Mulvihill WP, Weeks DE. Mega2: data-handling for facilitating genetic linkage and association analyses. Bioinformatics 2005; 21: 2556 2557. 25 Badenhop RF, Moses MJ, Scimone A, Mitchell PB, Ewen-White KR, Rosso A et al. A genome screen of 13 bipolar affective disorder pedigrees provides evidence for susceptibility loci on chromosome 3 as well as chromosomes 9, 13 and 19. Mol Psychiatry 2002; 7: 594 603. 26 Service S, Molina J, DeYoung J, Jawaheer D, Aldana I, Vu T et al. Results of a SNP genome screen in a large Costa Rican pedigree segregating for severe bipolar disorder. Am J Med Genet B Neuropsychiatr Genet 2006; 141: 367 373. 27 Hallmayer J. Chromosomes 1, 2, and 7 workshop. Am J Med Genet 1999; 88: 219 223. 28 Ekelund J, Lichtermann D, Hovatta I, Ellonen P, Suvisaari J, Terwilliger JD et al. Genome-wide scan for schizophrenia in the Finnish population: evidence for a locus on chromosome 7q22. Hum Mol Genet 2000; 9: 1049 1057. 29 Cheng R, Juo SH, Loth JE, Nee J, Iossifov I, Blumenthal R et al. Genome-wide linkage scan in a large bipolar disorder sample from the National Institute of Mental Health genetics initiative suggests putative loci for bipolar disorder, psychosis, suicide, and panic disorder. Mol Psychiatry 2006; 11: 252 260. 30 Fallin MD, Lasseter VK, Wolyniec PS, McGrath JA, Nestadt G, Valle D et al. Genomewide linkage scan for bipolar-disorder susceptibility loci among Ashkenazi Jewish families. Am J Hum Genet 2004; 75: 204 219. 31 Fallin MD, Lasseter VK, Wolyniec PS, McGrath JA, Nestadt G, Valle D et al. Genomewide linkage scan for schizophrenia susceptibility loci among Ashkenazi Jewish families shows evidence of linkage on chromosome 10q22. Am J Hum Genet 2003; 73: 601 611. 32 Ostrer H. A genetic profile of contemporary Jewish populations. Nat Rev Genet 2001; 2: 891 898. 33 Wildenauer DB, Schwab SG. Chromosomes 8 and 10 workshop. Am J Med Genet 1999; 88: 239 243. 34 Kelsoe JR, Spence MA, Loetscher E, Foguet M, Sadovnick AD, Remick RA et al. A genome survey indicates a possible susceptibility locus for bipolar disorder on chromosome 22. Proc Natl Acad Sci USA 2001; 98: 585 590. 35 Cichon S, Schmidt-Wolf G, Schumacher J, Muller DJ, Hurter M, Schulze TG et al. A possible susceptibility locus for bipolar affective disorder in chromosomal region 10q25 q26. Mol Psychiatry 2001; 6: 342 349. 36 Ewald H, Flint TJ, Jorgensen TH, Wang AG, Jensen P, Vang M et al. Search for a shared segment on chromosome 10q26 in patients with bipolar affective disorder or schizophrenia from the Faroe Islands. Am J Med Genet 2002; 114: 196 204. 37 Liu J, Juo SH, Dewan A, Grunn A, Tong X, Brito M et al. Evidence for a putative bipolar disorder locus on 2p13-16 and other potential loci on 4q31, 7q34, 8q13, 9q31, 10q21-24, 13q32, 14q21 and 17q11-12. Mol Psychiatry 2003; 8: 333 342. 38 Lerer B, Segman RH, Hamdan A, Kanyas K, Karni O, Kohn Y et al. Genome scan of Arab Israeli families maps a schizophrenia susceptibility gene to chromosome 6q23 and supports a locus at chromosome 10q24. Mol Psychiatry 2003; 8: 488 498. 449

450 39 McInnis MG, Lan TH, Willour VL, McMahon FJ, Simpson SG, Addington AM et al. Genome-wide scan of bipolar disorder in 65 pedigrees: supportive evidence for linkage at 8q24, 18q22, 4q32, 2p12, and 13q12. Mol Psychiatry 2003; 8: 288 298. 40 Williams NM, Norton N, Williams H, Ekholm B, Hamshere ML, Lindblom Y et al. A systematic genomewide linkage study in 353 sib pairs with schizophrenia. Am J Hum Genet 2003; 73: 1355 1367. 41 Etain B, Mathieu F, Rietschel M, Maier W, Albus M, McKeon P et al. Genome-wide scan for genes involved in bipolar affective disorder in 70 European families ascertained through a bipolar type I early-onset proband: supportive evidence for linkage at 3p14. Mol Psychiatry 2006; 11: 685 694. 42 Faraone SV, Hwu HG, Liu CM, Chen WJ, Tsuang MM, Liu SK et al. Genome scan of Han Chinese schizophrenia families from Taiwan: confirmation of linkage to 10q22.3. Am J Psychiatry 2006; 163: 1760 1766. 43 Marcheco-Teruel B, Flint TJ, Wikman FP, Torralbas M, Gonzalez L, Blanco L et al. A genome-wide linkage search for bipolar disorder susceptibility loci in a large and complex pedigree from the eastern part of Cuba. Am J Med Genet B Neuropsychiatr Genet 2006; 141: 833 843. 44 Griebel G, Perrault G, Soubrie P. Effects of SR48968, a selective non-peptide NK2 receptor antagonist on emotional processes in rodents. Psychopharmacology (Berl) 2001; 158: 241 251. 45 Dableh LJ, Yashpal K, Rochford J, Henry JL. Antidepressant-like effects of neurokinin receptor antagonists in the forced swim test in the rat. Eur J Pharmacol 2005; 507: 99 105. 46 Salome N, Stemmelin J, Cohen C, Griebel G. Selective blockade of NK2 or NK3 receptors produces anxiolytic- and antidepressantlike effects in gerbils. Pharmacol Biochem Behav 2006; 83: 533 539. 47 Fallin MD, Lasseter VK, Avramopoulos D, Nicodemus KK, Wolyniec PS, McGrath JA et al. Bipolar I disorder and schizophrenia: a 440-single-nucleotide polymorphism screen of 64 candidate genes among Ashkenazi Jewish case-parent trios. Am J Hum Genet 2005; 77: 918 936. Supplementary Information accompanies the paper on the website (http://www.nature.com/mp)