Autosomal linkage analysis of a Japanese single multiplex schizophrenia pedigree reveals two candidate loci on chromosomes 4q and 3q.код для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 144B:735 –742 (2007) Autosomal Linkage Analysis of a Japanese Single Multiplex Schizophrenia Pedigree Reveals Two Candidate Loci on Chromosomes 4q and 3q Naoshi Kaneko,1 Tatsuyuki Muratake,1,2* Hideki Kuwabara,1 Takanori Kurosaki,3 Mitsuru Takei,4 Tsuyuka Ohtsuki,2,5 Tadao Arinami,2,5 Shoji Tsuji,6 and Toshiyuki Someya1 1 Department of Psychiatry, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan CREST, Japan Science and Technology Agency, Kawaguchi-shi, Saitama, Japan 3 Honda Hospital, Uonuma, Niigata, Japan 4 Gunma Prefectural Psychiatric Medical Center, Isesaki, Gunma, Japan 5 Department of Medical Genetics, Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba, Ibaraki, Japan 6 Department of Neurology, Graduate School of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo, Japan 2 We analyzed a large multiplex schizophrenia pedigree collected in mid-eastern Japan using 322 microsatellite markers distributed throughout the whole autosome. Under an autosomaldominant inheritance model, the highest pairwise LOD score (LOD ¼ 1.69) was found at 4q (D4S2431: theta ¼ 0.0), and LOD scores at two other loci 3q (ATA34G06) and 8q (D8S1128) were 1.62 and 1.46, respectively. In multipoint analysis, LOD scores of the regions on 4q and 3q remained at a similar level; however, the LOD score of the region on 8q apparently decreased. Additional dense map analysis revealed haplotypes on 4q and 3q regions shared by affected individuals. On chromosome 4q, the haplotype spanning about 8 centiMorgans (cM) was shared by four of six genotyped individuals with schizophrenia and one affected individual whose haplotype was estimated. On 3q, the haplotype spanning about 20 cM was shared by five genotyped individuals with schizophrenia. We obtained two candidate regions of major susceptibility loci for schizophrenia on chromosomes 3q and 4q. ß 2007 Wiley-Liss, Inc. KEY WORDS: genome scan; linkage analysis; multiply affected family; schizophrenia; microsatellite marker Please cite this article as follows: Kaneko N, Muratake T, Kuwabara H, Kurosaki T, Takei M, Ohtsuki T, Arinami T, Tsuji S, Someya T. 2007. Autosomal Linkage Analysis of a Japanese Single Multiplex Schizophrenia Pedigree Reveals Two Candidate Loci on Chromosomes 4q and 3q. Am J Med Genet Part B 144B:735–742. *Correspondence to: Dr. Tatsuyuki Muratake, Department of Psychiatry, Niigata University Graduate School of Medical and Dental Sciences, 1-757 Asahimachi, Niigata 951-8510, Japan. E-mail: email@example.com Received 27 August 2005; Accepted 21 November 2006 DOI 10.1002/ajmg.b.30488 ß 2007 Wiley-Liss, Inc. INTRODUCTION Genetic epidemiological studies, such as family, twin, and adoption studies, have shown that genetic factors play an important role in the pathogenesis of schizophrenia (MIM 181500) [Lichtermann et al., 2000; Tsuang et al., 2001]; however, its genetic factors are complex, and conclusive results have not yet been determined. Like other complex diseases, it is difficult to map the disease loci for several reasons, including its non-Mendelian inheritance, diagnostic or phenotypic definition, and disease and allelic heterogeneity. To overcome those difficulties in the detection of susceptibility loci, several approaches are possible. One method is to include more case samples, such as a large number of affected pedigrees or pairs of affected siblings, or more data for linkage disequilibrium mapping with an extreme dense marker set. Over 20 genomewide linkage studies for schizophrenia have been reported and two meta-analyses were performed. Badner and Gershon  applied the multiple scan probability method to published genome scan data, and found evidence of schizophrenia susceptibility loci on chromosomes 8p, 13q, and 22q. Lewis et al.  applied the Rank-based genome scan metaanalysis method to published and unpublished data. They reported that most significant evidence of schizophrenia was found on chromosome 2q and evidence was also found in nine other regions (5q, 3p, 11q, 6p, 1q, 22q, 8p, 20q, and 14p). Most genome-wide linkage studies for schizophrenia analyzed a relatively large number of pedigrees consisting of few members. Though a large number of pedigrees can increase the power, it may cause pedigree-to-pedigree genetic heterogeneity. Another method is to analyze multiply affected pedigrees with the same etiological background. Although the majority of cases conceivably result from the interaction of several genes and environment factors, it was proposed that the causes of schizophrenia were diverse and that single rare genes or the most common specific genes caused schizophrenia in a minority of cases [Gottesman, 1991]. Compared with the nonparametric approach with a large number of pedigrees with relatively few members such as affected sib-pair families, a gene or genes with a major effect may be segregated in large multiplex families. Small sets of large multiplex pedigrees are powerful tools to study complex diseases as in the case of earlyonset Alzheimer’s disease [Schellenberg et al., 1992]. Therefore, we performed a genome scan on a multiply affected family in mid-eastern Japan to identify the gene with the major effect in schizophrenia. Here, we report linkage analysis with 322 736 Kaneko et al. microsatellite markers distributed throughout the whole autosome to identify an autosomal susceptibility locus. MATERIALS AND METHODS Pedigree Sample A moderately large multiplex schizophrenia pedigree collected in mid-eastern Japan was analyzed (Fig. 1). All individuals had Japanese ethnicity. The subjects participating in this study gave written consent after receiving an explanation of the study procedures and their implications. This study was approved by the ethics committee on genetics of the Niigata University School of Medicine. We obtained peripheral blood from 15 family members including 6 affected cases (1 male and 5 females). Diagnoses were made using criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) on the basis of structured interviews, review of clinical records, and information from reliable relatives and mental health professionals who had treated at least one of the pedigree members. Direct interviews were conducted using the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) and Axis II Disorders (SCID-II). Final diagnoses were made by the consensus of two trained psychiatrists (T.M. and H.K.) based on structured interview records, medical charts, and other information. All affected cases were diagnosed with schizophrenia according to DSM-IV criteria. Eight family members were confirmed as having no disorders on the Axis I Diagnosis of DSM-IV or any personality disorders including schizotypal, schizoid, and paranoid personality disorders that are regarded as being within the schizophrenia spectrum [Kendler et al., 1995; Tienari et al., 2003]. Those eight members were defined as ‘‘unaffected’’ in statistical analyses. One member had severe mental retardation and was defined as ‘‘unknown.’’ Four deceased individuals were defined as ‘‘affected’’ according to the medical records and interviews with mental health professionals. Other deceased members were defined as ‘‘unknown.’’ penetrance of the disease gene homozygote and heterozygote ¼ 0.8. Other estimated model parameters were as follows: disease gene frequency ¼ 0.01, penetrance of the normal gene homozygote ¼ 0.001. Age-dependant penetrance was not used because the youngest unaffected man was 28 years old and woman was over 50, which exceeded the peak age-at-onset of schizophrenia for each gender in the Japanese population [Nakane et al., 1992]. Marker allele frequencies in the Japanese population were obtained from the JSSLG web site (http://www.md. tsukuba.ac.jp/public/basic-med/m-genetics/JSSLGhp/1stScr/ WeberJSSLGAlleleFreq.htm). For marker allele frequencies not available on this site, we used frequencies in 25 newly genotyped Japanese healthy individuals. After pairwise linkage analysis, we obtained three promising regions at chromosomes 3, 4, and 8. Multipoint LOD score linkage analyses of these chromosomes were performed with the LINKMAP program in the FASTLINK program. Marker order and genetic distances between each marker were obtained from the Marshfield Medical Clinic website [Broman et al., 1998] (http://research.marshfieldclinic.org/genetics/). Simulation analysis was also carried out with the same data set. We used the same pedigree structure and the markers used in the actual analysis to generate 1,000 unlinked replicates with the SIMULATE program (ftp://linkage.cpmc.columbia. edu/software/simulate) and then conducted analyses with the MLINK used in the actual analysis. We obtained the LOD score of 1 random occurrence per genome scan and 1 occurrence per 20 genome scans. We also performed non-parametric linkage analysis using the same data set by SIMWALK2, version 2.91 [Sobel and Lange, 1996]. Non-parametric linkage analysis was performed based on identity by descent measurements at the marker loci and reports of empirical P-values for five non-parametric statistics, BLOCKS, MAX-TREE, ENTROPY, NPL_PAIR, and NPL_ALL. BLOCKS tended to be the most powerful for a recessive trait and MAX-TREE for a dominant trait. The remaining statistics were designed to be the most powerful for additive traits. The level of significance was 0.05 in this analysis. Genotyping Genomic DNA was extracted from peripheral blood by the standard phenol/chloroform method. Microsatellite markers were used (MapPairs human screening set ver. 9; Research Genetics, Inc., Huntsville, AL), which are distributed throughout the genome with an average interval about 10 cM. Polymerase chain reaction (PCR) was set up with 5 ng genomic DNA, and PCR cycling consisted of denaturation at 958C for 12 min, followed by 10 cycles at 948C for 15 sec, at 53–598C for 15 sec and at 728C for 30 sec, 20 cycles at 898C for 15 sec, at 53–598C for 15 sec and 728C for 30 sec, and concluded with final extension at 728C for 10 min. All subjects were genotyped using an ABI 377 genetic analyzer (Applied Biosystems, Foster City, CA) with GeneScan program ver.2.1 (Applied Biosystems). Genotyping was performed under a blind-to-diagnosis condition. Discrepancy from Mendelian inheritance was checked using Checkfam software (http:// www.genstat.net/checkfam/). When a non-Mendelian pattern of inheritance was observed, genotyping was repeated. Additional Data Two promising regions on chromosomes 4 and 3 were obtained by pairwise and multipoint analyses, so we genotyped 24 and 34 SNPs spread over these regions, respectively (Figs. 1, 2). These SNP markers were typed using the Illumina BeadStation 500G (Illumina, San Diego, CA) according to the manufacturer’s standard recommendation. We re-analyzed the multipoint LOD score by LINKMAP program using SNPs and microsatellite markers on these regions. Allele frequencies of SNPs in Asian populations were obtained from the Illumina website (http://www.illumina.com/General/support/ excel/11186808_Linkage_IV_B_Genetic_Map_v2_RevD.xls). For dense analyses, marker order and genetic distances between SNPs and microsatellite markers were based on the deCODE map [Kong et al., 2002]. Haplotypes of these regions were estimated using the MERLIN program [Abecasis et al., 2002]. We employed a haplotype estimation mode which provides haplotypes corresponding to the most likely pattern of gene flow. Statistical Analyses Pairwise LOD score linkage analyses were performed with the MLINK program in the FASTLINK package version 4.1 [Cottingham et al., 1993]. Based on the observation of disease distribution in the pedigree, an autosomal-dominant inheritance model was employed. Consequently, we estimated the RESULTS Affected individuals were aggregated into two branches in the pedigree. We found an affected married-in individual (II-2) in the pedigree and surveyed these relatives as possible; however, we could not identify more affected individuals. Linkage Study of Single Multiplex Pedigree for Schizophrenia 737 Fig. 1. Structure of the pedigree with individual IDs and a shared haplotype on chromosome 4 are shown. Black rectangles represent affected cases. Unenclosed IDs represent unaffected individuals. White rectangles are defined as unknown. Genotyped individuals are identified by (*) symbols. Deceased individuals are identified by (þ) symbols. To protect personal information, gender, age, and order of siblings have been disguised. On the left of the figure, genotyped markers are listed with genetic distances of microsatellite markers according to the Marshfield Medical Clinic website. A common haplotype was shared by four genotyped cases (III-2, 8, 9, and IV-1), one estimated case (III-4), and one genotyped individual whose affected status is unknown (IV-4). There were two pairs of transmission from a possible-carrier father to an affected son in the pedigree; therefore, the mode of inheritance was presumed to be not X-linked. A total of 322 microsatellite markers were used in this analysis (average interval 10.8 cM, range ¼ 0.6–26.4 cM). In 1,000 simulation analyses, we obtained on average one highest LOD score, 1.30 per genome scan, in the absence of linkage. The highest LOD score of 1.66 appeared only once in every 20 genome scans in the absence of linkage (Fig. 3). The result of pairwise LOD score analysis is shown in Figure 4 and the Table I. D4S2431 generated the highest pairwise LOD score (LOD score ¼ 1.69: theta ¼ 0.0). This score exceeded the one random occurrence per 20 genome scans in simulation analysis. ATA34G06 and D8S1128 generated LOD 738 Kaneko et al. Fig. 2. Haplotype on chromosome 3. A common haplotype was shared by five genotyped cases (III-1, 2, 3, 8, and 9). SNP markers unrelated to haplotype or not informative were omitted. Linkage Study of Single Multiplex Pedigree for Schizophrenia Fig. 3. The distribution of the highest lod scores from the permutations. Frequencies of the highest lod scores from the permutations were shown. The frequencies were evaluated at intervals of a LOD score of 0.1. 739 analyses were conducted on these three chromosomes using LINKMAP program. For chromosome 4, the highest LOD score of 1.69 was obtained at 174.8 cM from p-ter (nearest marker: D4S2431). The highest LOD score for chromosome 3 was 1.66 at 139.5 cM (nearest marker: ATA34G06). For chromosome 8, the highest LOD score decreased to 1.04 at 138.3 cM. Additional analysis using a dense SNP map was carried out on promising regions on chromosomes 4 and 3 as they maintained a relatively high multipoint LOD score. The results of multipoint LOD score analysis using SNP and microsatellite data are shown in Figure 5. On chromosome 4, the multipoint LOD score was still high, and the estimated haplotype from rs1824347 to rs2044868 (spanning about 8 cM) was shared by five affected individuals (four genotyped and one estimated) and one unknown individual (Fig. 1). Also on chromosome 3, the multipoint LOD score was high over 20 cM, and a haplotype from rs326361 to rs1355782 was shared by five affected individuals (all genotyped) and not shared by all unaffected and unknown individuals (Fig. 2). DISCUSSION scores of 1.62 (theta ¼ 0.0) and 1.46 (theta ¼ 0.0), respectively, exceeding the score of one occurrence per genome scan; other markers did not exceed this LOD score. On the other hand, by non-parametric linkage analysis, significant linkage was observed at GAAT1A4 on chromosome 8 (P-value for NPL_ALL statistic ¼ 0.0428), D9S1118 on chromosome 9 (P-value for ENTROPY statistic ¼ 0.0336 and P-value for NPL_PAIR statistic ¼ 0.0394), and D10S212 on chromosome 10 (P-value for ENTROPY ¼ 0.0368, P-value for NPL-PAIR ¼ 0.0474, and P-value for NPL_ALL ¼ 0.0403); however, these values were nominal and quite modest under the conditions of genome wide analysis. Consequently, additional analyses were not carried out on these regions. As three promising regions on chromosomes 3, 4, and 8 were found on pairwise parametric linkage analysis, multipoint This study attempted to identify linkage to schizophrenia susceptibility loci by analyzing a moderately large multiplex schizophrenia pedigree in the Japanese population. The pedigree is highly multiplex with schizophrenia and would be useful for identifying a ‘‘pedigree-specific’’ Mendelian gene, if it exists. It is considered that multiple genes and environments are associated with developing schizophrenia in most cases; however, it is helpful to identify rare single or major genes associated with schizophrenia to realize the molecular pathogenesis of schizophrenia, as in other common diseases (e.g., Alzheimer’s disease, diabetes, or hyperlipidemia) [Peltonen et al., 2006]. To avoid Type II error due to limited information from the moderate extent of the pedigree, we undertook simulation analysis to obtain the score of linkage support. Fig. 4. Pairwise LOD score throughout the whole autosome (theta ¼ 0.0). Chromosomes are arranged by number from p-ter to q-ter with genetic distances on a linear scale. Horizontal broken lines are LOD scores 1.66 and 1.30, respectively. b 124.2 134.6 138 152.6 161.0 158.0 167.6 176.2 181.9 195.1 125.3 135.1 139.5 148.1 164.5 110.2 58.3 170.9 cM 0.1 0.278750 0.722962 1.205113 0.682373 0.159391 0.305645 0.186203 1.217937 0.339390 0.054438 0.336288 0.704249 1.065193 0.096582 0.534285 0.667207 0.017026 0.208246 Theta ¼ 0.0 0.515631 0.923861 1.621723a 1.685463 0.100501 0.576538 0.229854 1.692203b 0.619805 0.371202 1.498008 0.883282 1.461049a 0.640357 2.372965 0.842408 0.467966 0.398779 0.4 0.042886 0.106277 0.134704 0.033017 0.048992 0.038391 0.044953 0.098516 0.041830 0.035219 0.040123 0.051849 0.094896 0.008530 0.017932 0.078233 0.055912 0.022703 0.3 0.091987 0.278214 0.405821 0.137655 0.106145 0.128451 0.091258 0.351513 0.142952 0.073772 0.086083 0.224380 0.327973 0.001397 0.070606 0.248104 0.046218 0.001753 0.2 0.160810 0.496608 0.784987 0.334330 0.151595 0.253256 0.139004 0.750962 0.272287 0.098514 0.161746 0.465290 0.672140 0.011664 0.196872 0.460766 0.000892 0.077187 The LOD scores 1.62 and 1.46 of ATA34G06 and D8S1128 exceed the highest score of random occurrence of a genome scan. The LOD score 1.69 of D4S2431 exceeds the highest score of random occurrence of 20 genome scans. 3 3 3 3 3 4 4 4 4 4 8 8 8 8 8 8 9 10 D3S3045 D3S2460 ATA34G06 D3S1764 D3S1744 D4S1629 D4S2368 D4S2431 D4S2417 D4S408 D8S592 D8S1179 D8S1128 D8S256 D8S373 GAAT1A4 D9S1118 D10S212 a Chr Marker Pairwise LOD score obtained by parametric linkage analysis 0.9382 0.6601 0.6566 0.4182 0.2437 0.6313 0.7743 0.6573 0.6548 0.9350 0.9463 0.6590 0.2109 0.3837 1.0000 0.3065 0.2145 0.1931 BLOCKS 0.5849 0.4847 0.4853 0.9476 0.4760 0.7892 0.6961 0.4843 0.9534 0.9578 0.9557 0.4839 0.4866 0.7899 0.9498 0.1560 0.1154 0.1641 MAX-TREE 0.7673 0.2636 0.2601 0.3741 0.1601 0.5174 0.5345 0.3643 0.4627 0.8991 0.8858 0.2191 0.0811 0.3487 0.9923 0.0827 0.0336 0.0368 ENTROPY 0.6894 0.2177 0.2101 0.5246 0.2561 0.5808 0.5175 0.3072 0.5455 0.9209 0.8925 0.1787 0.1396 0.4661 0.9923 0.0603 0.0394 0.0474 NPL_PAIR P-value obtained by non-parametric analysis TABLE I. The parametric and non-parametric pairwise LOD score of the promising markers and adjacent markers 0.6431 0.1601 0.1523 0.5485 0.2335 0.5644 0.4370 0.1543 0.5198 0.8928 0.8071 0.1364 0.1362 0.4416 0.9720 0.0428 0.0634 0.0403 NPL_ALL 740 Kaneko et al. Fig. 5. The results of multipoint linkage analysis of SNPs and microsatellite markers on chromosomes 4 and 3. a: Chromosome 4, (b) Chromosome 3. Marker order and genetic distances between markers are based on the deCODE map in those analyses. As mentioned above, pairwise LOD score analysis provided three promising loci with linkage to schizophrenia at marker D4S2431 on chromosome 4q31, marker ATA34G06 on 3q13, and marker D8S1128 on 8q24. In multipoint analysis, LOD scores on two loci on 4q31 and 3q13 remained at a similar level, whereas the LOD score of the locus on 8q24 apparently decreased. These signals were very weak compared with the thresholds for significant linkage and suggestive linkage by Lander and Kruglyak . The thresholds of Lander and Kruglyak were derived from the assumption of densemarker genotyping and complete extraction of inheritance information; therefore, these thresholds were strict and there was a risk of failure to detect significant linkage. Wiltshire et al.  conducted simulation analyses with complete autosomal genomes for sets of sib-pair families under the null hypothesis of no linkage and suggested that an independent region showing evidence of linkage with a LOD score of 1.51–1.55 is expected to occur once by chance per autosomal genome scan under experimental and data conditions encountered during a typical primary autosomal genome scan (a 10 cM marker map and 15% missing genotypes). Our results of two-point linkage analysis on markers D4S2431 and ATA34G06 exceeded this score. Non-parametric linkage analysis did not show sufficient results in this study. This might be due to our pedigree structure and/or the number of genotyped individuals not having adequate information to determine allele sharing or identical descent between affected pedigree members. Linkage Study of Single Multiplex Pedigree for Schizophrenia Subsequent dense map analyses on chromosomes 4q and 3q revealed two haplotypes shared by pedigree members with schizophrenia. One haplotype spanned about an 8 cM around marker D4S2431 on chromosome 4q. Straub et al.  reported linkage support at 4q24-32 including the D4S2431 locus in 270 Irish multiplex schizophrenia families. Both Japanese linkage analysis with schizophrenia using sib pair families [Japanese Schizophrenia Sib-Pair Linkage Group (JSSLG), 2003] and using two multiply affected families in Japan [Rees et al., 1999] did not provide linkage support at this region. Another haplotype spanned about 20 cM including markers D3S2460 and ATA34G06 on 3q. Paunio et al.  reported Zmax ¼ 2.17 for marker D3S2460 that was located 134.6 cM from p-ter of chromosome 3 using 53 pedigrees that were internally isolated in Finnish schizophrenia pedigrees. They subsequently failed to maintain a high LOD score using 238 Finnish pedigrees on this locus. Initial sib-pair analysis in Japanese pedigrees reported nominal significance with MLS 1.11 at 40 cM distal from the region [Japanese Schizophrenia Sib-Pair Linkage Group (JSSLG), 2003]. Chromosome 8p is one of the most promising regions linked to schizophrenia [Badner and Gershon, 2002; Berry et al., 2003; Owen et al., 2004], whereas there are few reports of linkage support at chromosome 8q, especially around D8S1128. Since our multipoint linkage analysis on 8q24 showed that the LOD score decreased from the value of pairwise linkage analysis, this region may not have a major role in the pedigree. There have been several parametric linkage analyses for schizophrenia in relatively large multiplex pedigrees. Coon et al.  reported the results of genome-wide parametric linkage analysis using a large multiplex pedigree from the Micronesian nation of Palau. Nine strictly defined cases of schizophrenia and 17 unaffected relatives were genotyped and the highest LOD score of 2.17 was found on 2p13-14. Rees et al.  analyzed 13 families, 11 UK families and 2 Japanese families multiply affected with schizophrenia and related disorders. Bailer et al.  conducted genome-wide linkage scan and the highest parametric LOD score of 1.33 was observed for marker D3S1279 with five schizophrenia families. We found two promising haplotypes on chromosomes 4q and 3q shared by most of the individuals with schizophrenia but detailed examination is required to determine the schizophrenia susceptibility gene. Among several hundreds of genes located in the two regions, the Dopamine receptor D3 gene (DRD3) is located on 3q13. Association between the DRD3 gene and schizophrenia was investigated in many populations as well as in the Japanese population [Tanaka et al., 1996; Ishiguro et al., 2000]; nevertheless, the results were conflicting. Recently, meta-analysis of association studies reported that the DRD3 gene was genetically associated with schizophrenia [Lohmueller et al., 2003]. This study has some limitations, including an affected married-in spouse (II-2), with four of six affected individuals genotyped in this study being her offspring. Although we could not find any affected individuals in her relatives, it is possible that the spouse introduced susceptibility genes into the family. This may produce a false assumption of the inheritance model or reduce genetic homogeneity in the pedigree. Contrary to this possibility, affected individuals in the left branch of the pedigree (III-1, 2, 3, and IV-1) and those in the middle branch (III-8 and 9) shared haplotypes on chromosomes 3q and 4q. We, therefore, consider that the affected individuals of both branches have a common genetic background. Separate analysis of the left branch might reveal new candidate loci if each branch of the pedigree has distinct genetic factors. This analysis, however, could introduce type II error due to the reduced observable number of meiosis. In this study, we carried out an autosomal genome scan to identify the major susceptibility locus for schizophrenia and we 741 located two candidate regions on chromosomes 3q and 4q. Further linkage mapping of isolated populations might narrow the candidate regions, or linkage disequilibrium tests of familial schizophrenia on these regions might reveal the susceptibility gene for schizophrenia. ACKNOWLEDGMENTS We express our deep gratitude to all the subjects who took part in the study. We are grateful to Dr. Chizuko Kishi at Kishi Hospital, Mr. Mamoru Fukushima at Ohiradai Gakuen, and many other mental health professionals for their help in recruiting the subjects. We thank Dr. Kenju Hara, Dr. Takeshi Ikeuchi, Dr. Hisashi Kobayashi, and Dr. Takashi Suzuki at Department of Neurology, Brain Research Institute, Niigata University for their helpful advice. Hiroshi Kusano, Miki Inomata, Hiromi Tanabe, and Tomoko Yamada provided excellent technical assistance. REFERENCES Abecasis GR, Cherny SS, Cookson WO, Cardon LR. 2002. Merlin—Rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet 30:97–101. Badner JA, Gershon ES. 2002. Meta-analysis of whole-genome linkage scans of bipolar disorder and schizophrenia. Mol Psychiatry 7:405–411. Bailer U, Leisch F, Meszaros K, Lenzinger E, Willinger U, Strobl R, Heiden A, Gebhardt C, Doge E, Fuchs K, Sieghart W, Kasper S, Hornik K, Aschauer HN. 2002. Genome scan for susceptibility loci for schizophrenia and bipolar disorder. Biol Psychiatry 52:40–52. Berry N, Jobanputra V, Pal H. 2003. Molecular genetics of schizophrenia: A critical review. J Psychiatry Neurosci 28:415–429. Broman KW, Murray JC, Sheffield VC, White RL, Weber JL. 1998. Comprehensive human genetic maps: Individual and sex-specific variation in recombination. Am J Hum Genet 63:861–869. Coon H, Myles-Worsley M, Tiobech J, Hoff M, Rosenthal J, Bennett P, Reimherr F, Wender P, Dale P, Polloi A, Byerley W. 1998. Evidence for a chromosome 2p13-14 schizophrenia susceptibility locus in families from Palau, Micronesia. Mol Psychiatry 3:521–527. Cottingham RW, Idury RM, Schaffer AA. 1993. Faster sequential genetic linkage computations. Am J Hum Genet 53:252–263. Gottesman II. 1991. Schizophrenia genetics: The origin of madness. New York: Freeman and Company. pp 230–232. Ishiguro H, Okuyama Y, Toru M, Arinami T. 2000. Mutation and association analysis of the 50 region of the dopamine D3 receptor gene in schizophrenia patients: Identification of the Ala38Thr polymorphism and suggested association between DRD3 haplotypes and schizophrenia. Mol Psychiatry 5:433–438. Japanese Schizophrenia Sib-Pair Linkage Group (JSSLG). 2003. Initial genome-wide scan for linkage with schizophrenia in the Japanese Schizophrenia Sib-Pair Linkage Group (JSSLG) families. Am J Med Genet Part B 120B:22–28. Kendler KS, Neale MC, Walsh D. 1995. Evaluating the spectrum concept of schizophrenia in the Roscommon family study. Am J Psychiatry 152: 749–754. Kong A, Gudbjartsson DF, Sainz J, Jonsdottir GM, Gudjonsson SA, Richardsson B, Sigurdardottir S, Barnard J, Hallbeck B, Masson G, Shlien A, Palsson ST, Frigge ML, Thorgeirsson TE, Gulcher JR, Stefansson K. 2002. A high-resolution recombination map of the human genome. Nat Genet 31:241–247. Lander E, Kruglyak L. 1995. Genetic dissection of complex traits: Guidelines for interpreting and reporting linkage results. Nat Genet 11:241–247. Lewis CM, Levinson DF, Wise LH, DeLisi LE, Straub RE, Hovatta I, Williams NM, Schwab SG, Pulver AE, Faraone SV, Brzustowicz LM, Kaufmann CA, Garver DL, Gurling HM, Lindholm E, Coon H, Moises HW, Byerley W, Shaw SH, Mesen A, Sherrington R, O’Neill FA, Walsh D, Kendler KS, Ekelund J, Paunio T, Lonnqvist J, Peltonen L, O’Donovan MC, Owen MJ, Wildenauer DB, Maier W, Nestadt G, Blouin JL, Antonarakis SE, Mowry BJ, Silverman JM, Crowe RR, Cloninger CR, Tsuang MT, Malaspina D, Harkavy-Friedman JM, Svrakic DM, Bassett AS, Holcomb J, Kalsi G, McQuillin A, Brynjolfson J, Sigmundsson T, Petursson H, Jazin E, Zoega T, Helgason T. 2003. Genome scan metaanalysis of schizophrenia and bipolar disorder, part II: Schizophrenia. Am J Hum Genet 73:34–48. 742 Kaneko et al. Lichtermann D, Karbe E, Maier W. 2000. The genetic epidemiology of schizophrenia and of schizophrenia spectrum disorders. Eur Arch Psychiatry Clin Neurosci 250:304–310. Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN. 2003. Metaanalysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet 33: 177–182. Nakane Y, Ohta Y, Radford MH. 1992. Epidemiological studies of schizophrenia in Japan. Schizophr Bull 18:75–84. Owen MJ, Williams NM, O’Donovan MC. 2004. The molecular genetics of schizophrenia: New findings promise new insights. Mol Psychiatry 9:14–27. Paunio T, Ekelund J, Varilo T, Parker A, Hovatta I, Turunen JA, Rinard K, Foti A, Terwilliger JD, Juvonen H, Suvisaari J, Arajarvi R, Suokas J, Partonen T, Lonnqvist J, Meyer J, Peltonen L. 2001. Genome-wide scan in a nationwide study sample of schizophrenia families in Finland reveals susceptibility loci on chromosomes 2q and 5q. Hum Mol Genet 10:3037–3048. Peltonen L, Perola M, Naukkarinen J, Palotie A. 2006. Lessons from studying monogenic disease for common disease. Hum Mol Genet 15: R67–R74. Rees MI, Fenton I, Williams NM, Holmans P, Norton N, Cardno A, Asherson P, Spurlock G, Roberts E, Parfitt E, Mant R, Vallada H, Dawson E, Li MW, Collier DA, Powell JF, Nanko S, Gill M, McGuffin P, Owen MJ. 1999. Autosome search for schizophrenia susceptibility genes in multiply affected families. Mol Psychiatry 4:353–359. Schellenberg GD, Bird TD, Wijsman EM, Orr HT, Anderson L, Nemens E, White JA, Bonnycastle L, Weber JL, Alonso ME, Potter H, Heston LL, Martin GM. 1992. Genetic linkage evidence for a familial Alzheimer’s disease locus on chromosome 14. Science 23:668–671. Sobel E, Lange K. 1996. Descent graphs in pedigree analysis: Applications to haplotyping, location scores, and marker-sharing statistics. Am J Hum Genet 58:1323–1337. Straub RE, MacLean CJ, Ma Y, Webb BT, Myakishev MV, Harris-Kerr C, Wormley B, Sadek H, Kadambi B, O’Neill FA, Walsh D, Kendler KS. 2002. Genome-wide scans of three independent sets of 90 Irish multiplex schizophrenia families and follow-up of selected regions in all families provides evidence for multiple susceptibility genes. Mol Psychiatry 7:542–559. Tanaka T, Igarashi S, Onodera O, Tanaka H, Takahashi M, Maeda M, Kameda K, Tsuji S, Ihda S. 1996. Association study between schizophrenia and dopamine D3 receptor gene polymorphism. Am J Med Genet 67:366–368. Tienari P, Wynne LC, Laksy K, Moring J, Nieminen P, Sorri A, Lahti I, Wahlberg KE. 2003. Genetic boundaries of the schizophrenia spectrum: Evidence from the Finnish adoptive family study of schizophrenia. Am J Psychiatry 160:1587–1594. Tsuang MT, Stone WS, Faraone SV. 2001. Genes, environment and schizophrenia. Br J Psychiatry Suppl 40:s18–s24. Wiltshire S, Cardon LR, McCarthy MI. 2002. Evaluating the results of genomewide linkage scans of complex traits by locus counting. Am J Hum Genet 71:1175–1182.