ORIGINAL RESEARCH ARTICLE A genome scan and follow-up study identify a bipolar disorder susceptibility locus on chromosome 1q42 S Macgregor 1,4 , PM Visscher 1 , SA Knott 1 , P Thomson 2 , DJ Porteous 2 , JK Millar 2 , RS Devon 2 , D Blackwood 3 and WJ Muir 3 1 Institute of Cell, Animal and Population Biology, University of Edinburgh, Kings Buildings, Edinburgh, UK; 2 Molecular Medicine Centre, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh, UK; 3 Division of Molecular and Clinical Medicine, University of Edinburgh, Royal Edinburgh Hospital, Edinburgh, UK In this study, we report a genome scan for psychiatric disease susceptibility loci in 13 Scottish families. We follow up one of the linkage peaks on chromosome 1q in a substantially larger sample of 22 families affected by schizophrenia (SCZ) or bipolar affective disorder (BPAD). To minimise the effect of genetic heterogeneity, we collected mainly large extended families (average family size 418). The families collected were Scottish, carried no chromosomal abnormalities and were unrelated to the large family previously reported as segregating a balanced (1 : 11) translocation with major psychiatric disease. In the genome scan, we found linkage peaks with logarithm of odds (LOD) scores 41.5 on chromosomes 1q (BPAD), 3p (SCZ), 8p (SCZ), 8q (BPAD), 9q (BPAD) and 19q (SCZ). In the follow-up sample, we obtained most evidence for linkage to 1q42 in bipolar families, with a maximum (parametric) LOD of 2.63 at D1S103. Multipoint variance components linkage gave a maximum LOD of 2.77 (overall maximum LOD 2.47 after correction for multiple tests), 12 cM from the previously identified SCZ susceptibility locus DISC1. Interestingly, there was negligible evidence for linkage to 1q42 in the SCZ families. These results, together with results from a number of other recent studies, stress the importance of the 1q42 region in susceptibility to both BPAD and SCZ. Molecular Psychiatry (2004) 9, 1083–1090. doi:10.1038/sj.mp.4001544 Published online 13 July 2004 Keywords: schizophrenia; bipolar disorder; DISC1; genetic heterogeneity Bipolar disorder (BPAD) and schizophrenia (SCZ) are severe psychiatric illnesses, with each affecting approximately 1% of most human populations. There is strong evidence for a genetic aetiology in such disorders with high heritabilities reported in twin and adoption studies. However, the task of identifying genomic regions conferring susceptibility has yielded inconsistent results, with a large number of candidate regions identified. 1 In recent years, several studies have identified two regions of chromosome 1q (1q21 and 1q42) as important in the aetiology of SCZ. At 1q21, a study of Canadian families produced a logarithm of odds (LOD) score of 6.5, 2 a study analysing British and Icelandic families generated an LOD of 3.2 3 and a family-based association study considering Spanish origin families reported a P-value of 0.003. 4 A meta- analysis of most of the recent SCZ genome scans reported the 1q21 region as being among the most likely to harbour a SCZ susceptibility locus. 5 Interest in 1q42 began when the region was implicated by the apparent effects of a chromosomal abnormality on major psychiatric disease in a large Scottish family. 6 The family segregated a balanced t(1;11)(q42;q14.3) translocation, with the presence of the translocation appearing to be linked with disease status. A linkage analysis considering the translocation as a marker generated an LOD of 3.6 7 when individuals with SCZ were considered affected, an LOD of 4.5 when individuals with recurrent major depression or BPAD were considered affected and an LOD of 7.1 when individuals with SCZ, BPAD and recurrent major depression were treated as affected. The translocation directly disrupts two genes on chromosome 1: these have been named DISC1 (OMIM 605210) and DISC2 (OMIM 606271), respectively. 8 While this result shows a clear relationship between the presence of the translocation and psychiatric disease, it was unclear if this result was of relevance to other families in the general population. In the last 5 years however, a number of studies have reported independent evidence for the role of 1q42 in psychiatric disease susceptibility. Two studies in Finnish families affected by SCZ generated LODs of 3.82 and 3.21 9,10 for markers close to the translocation break-point. A recent study of Taiwanese families reported Received 26 February 2004; revised 18 May 2004; accepted 19 May 2004 Correspondence: Dr S Macgregor, Institute of Cell, Animal and Population Biology, University of Edinburgh, Kings Buildings, Edinburgh, UK. E-mail: MacgregorS@cf.ac.uk 4 Current address: Biostatistics and Bioinformatics Unit, Univer- sity of Wales College of Medicine, Heath Hospital, Cardiff, UK Molecular Psychiatry (2004) 9, 1083–1090 & 2004 Nature Publishing Group All rights reserved 1359-4184/04 $30.00 www.nature.com/mp