6 July 2009
Schizophrenia is a complex psychiatric disease, with both environmental and genetic factors implicated in its development. Three large-scale international collaborative research studies by the International Schizophrenia Consortium (ISC), the SGENE-plus consortium (see previous news) and the Molecular Genetics of Schizophrenia (MGS) consortium, published in the Journal Nature, conducted genome-wide association studies (GWAS) in order to identify further genetic variants associated with schizophrenia (see previous news).
The ISC [International Schizophrenia Consortium (2009) Nature July 1 doi:10.1038/nature08185] conducted a GWAS of 3,322 cases and 3,587 controls. The most strongly identified association was on chromosome 22, located within the first intron of the MYO18B gene (P = 3.4 x 10-7), a candidate tumour suppressor gene. The second strongest association was made up by 450 SNPs within the major histocompatibility complex (MHC) on chromosome 6p (P = 3.7 x 10-7). The SGENE-plus [Stefansson et al. (2009) Nature July 1 doi:10.1038/nature08186] conducted a GWAS of 2,663 cases and 13,498 controls. This study also identified an association with the MHC region (strongest P-value = 2.3 x 10-4) and when combined with additional follow-up samples (4,999 cases and 15,555 controls), further strengthened the association identified in the MHC (strongest P-value = 4.4 x 10-9). The MGS consortium [Shi et al. (2009) Nature July 1 doi:10.1038/nature08192] conducted a GWAS of 2,681 cases and 2,653 controls of European-ancestry and 1,286 cases and 973 controls of African-American ethnicity. The strongest association identified was on chromosome 2q37.2 in the European-ancestry sample (P = 4.59 x 10-7) and chromosome 2q34 in the African-American sample (P = 2.14 x 10-6). However, when the European-ancestry data were combined with data from the ISC and SGENE-plus studies in a meta-analysis (8,008 cases and 19,077 controls), seven SNPs in the MHC region showed significant association (strongest P-value = 9.54 x 10-9).
Comment: By exchanging their GWAS summary results, these three large consortia were able to verify each other’s results, showing a statistically significant association between schizophrena and the MHC region. Due to the strong linkage disequilibrium within the MHC, as well as the gene-rich nature of this region, much work remains to identify the cause of any true association, with much larger samples and resequencing technology needed. Larger samples are also required to detect (and understand the causal nature of) further common (and potentially rare) gene variants that contribute to the underlying genetic risk of schizophrenia. The association with the MHC also supports previous hypotheses regarding the role of the immune system in schizophrenia.