As part of our ongoing coverage of DISC1 2010, held 3-6 September 2010, in Edinburgh, the United Kingdom, we bring you a meeting missive from Rosie Walker, a graduate student at the University of Edinburgh.
15 September 2010. The penultimate session—Genetics and Sequencing, chaired by Tom Insel of NIMH—witnessed a shift away from the biochemical and neurobiological focus of the conference to consider progress made in the identification of putative psychiatric disease-risk variants using next-generation sequencing technology.
First to speak was Elaine Green of Cardiff University, U.K., who described the identification of exon 11 sequence variants that may act as risk factors for psychiatric disease that has features of both psychosis and mania (schizoaffective spectrum). This work followed up on a previous linkage study from Cardiff University, which identified a significant linkage peak at chromosome 1q42, the DISC1 locus, in bipolar disorder and schizophrenia families (Hamshere et al., 2005). Subsequent sequencing of DISC1 in individuals with schizoaffective-bipolar disorder identified two rare variants in exon 11, prompting Green to address the question of whether exon 11 harbors rare susceptibility variants. Sequencing of a large sample of schizoaffective spectrum cases and unaffected controls resulted in the identification of four further variants, including two that would result in an amino acid change, that each occurred in only one case. Highlighting the need for large sample sizes and replication in sequencing studies, Green noted that the original two exon 11 variants that prompted this large-scale study in fact turned out to be found in several control subjects. Nevertheless, these results are in keeping with the conclusion by Song et al. (2008) that ultra-rare DISC1 variants are likely to confer risk for psychiatric illness.
The next talk was given by Denis Bauer of the University of Queensland, Australia. Bauer guided the audience through the process of identifying potential risk variants following the resequencing of a candidate region. In a familial association study of 42 Tamil Brahmin families (184 individuals, of which 99 were diagnosed with schizophrenia), Bauer and colleagues identified many significant SNPs around the DISC1 breakpoint identified in the original Scottish t(1;11) family. These SNPs were in strong linkage disequilibrium with each other, suggesting that they were probably tagging the same risk-associated variant(s). Bauer emphasized the advantages conferred by the ethnic and environmental homogeneity of the Tamil Brahmin population, and the lack of substance abuse in their culture, which increase the chance of finding risk variants associated with a “pure” schizophrenia phenotype. Resequencing of a 75Kbp region of interest in DISC1 in all 184 individuals, followed by stringent quality control measures, resulted in the identification of several DISC1 variants, the majority of which are intronic. Narrowing down these findings to a list of putative causal variants has been accomplished by functional annotation to identify those with a potential pathogenic role due to interference in processes such as protein folding and mRNA assembly. Further association analyses to elucidate the contribution of these variants to psychiatric illness susceptibility are eagerly awaited.
The final speaker in this session was Pippa Thomson of The University of Edinburgh, U.K. Prompted by the observation that many DISC1 associations are haplotypic, Thomson suggested that susceptibility variants for psychiatric illness may often be rare and of subtle effect. As such variants are most likely to be detected by sequencing, Thomson and colleagues have embarked on a large-scale study to sequence the 0.5Mbp TSNAX/DISC1 locus in 200 subjects each for bipolar disorder, unipolar depression, and schizophrenia, as well as almost 1,000 healthy controls. To date, sequencing data from all the control subjects and about 400 cases have been analyzed, resulting in the identification of just over 7,000 variants, the majority of which are in non-exonic regions. Several variants are in conserved regions, regions with predicted regulatory potential, and transcription factor binding sites, suggesting that these variants might impact on DISC1 regulation. Thus far, no obvious difference in the type of variants observed in cases and controls has been observed. Downstream association analyses and biochemical assessment of variants of interest will now be performed to determine their potential role in psychiatric illness.
A particularly interesting aspect of this study is that control subjects have been recruited from the Lothian Birth Cohort of 1936, a group of highly phenotyped individuals born in Scotland in 1936, who have undergone intelligence testing (Moray House Test) at the ages of 11 and 70, as well as a battery of other cognitive, physical, and biochemical tests, thus permitting quantitative trait analyses. Such analyses are in keeping with the suggestion that the study of the genetic underpinnings of phenotypes such as cognition, considered to be intermediate phenotypes for psychiatric illness, may confer increased power to detect genes conferring susceptibility to psychiatric illness (e.g., Tan et al., 2008). Analysis to date has focused on the general intelligence factor, “g,” which can itself be subdivided into two factors, fluid intelligence (gf) and crystallised intelligence (gc). Thomson reported that several variants have been identified as being associated with g, and that these variants fall into two clusters, one at the 5’ end and the other at the 3’ end of the gene. Variants associated with gf were enriched in conserved regions, suggesting that they may disrupt regions with regulatory function. Of note was the finding that individuals with a lower gf score carry more minor alleles. In conclusion, Thomson stated that DISC1 is a gene that’s involved in intelligence. Further analysis of the sequencing data from this rich resource will help to clarify DISC1’s contribution to intelligence in unaffected individuals, and provide clues as to how variation at this locus can confer susceptibility to psychiatric illness.
In summary, this session highlighted the value of high-throughput sequencing of genetic loci associated with psychiatric illness. The potential for this approach to identify rare variants and patterns of variation across a gene, whilst bypassing the problems created by the uncertain genetic architecture of psychiatric illness that plague the traditional genetic approaches of association and linkage, was emphasized, giving great hope that such studies will yield many as-of-yet unidentified variants for geneticists and biologists working in the field to follow up on.—Rosie Walker.