05.12.2012, 17:15 - HZI-Kolloquium - Scientific

Grant Morahan "Gene interactions in type 1 diabetes: a systems genetics approach"

Aberrant regulation of immune responses can give rise to autoimmune diseases such as Type 1 Diabetes (T1D). Although many genes have been identified that affect risk of developing T1D, it is not known how they contribute to disease. We applied systems genetics analyses to explore the molecular basis for susceptibility. Over 60 loci have been reported to be associated with T1D. Many of these are defined only by anonymous SNPs: identity of the causative genes, and the molecular bases by which they mediate susceptibility, are not known. To identify these genes, we performed a systematic analysis. First, all known genes in modest linkage disequilibrium with the tag SNP for each locus were examined; we identified common non-synonymous alleles in 259 genes at 39 loci. Next, we examined the effect of SNP genotypes on regulating expression levels of nearby (<2Mb) genes (i.e. cis-regulatory effects typical of promoter/enhancer elements). To do so, we examined gene expression in a total of 442 samples of three different cell types: EBV-transformed B cell lines (resting and after PMA stimulation); and purified CD4+ and CD8+ T cells. We mapped expression quantitative trait loci (eQTL) and found 43 non-HLA loci, typically with P < 1 × 10−7. We tested for trans-regulatory effects (i.e. association with changes in expression of genes located in other chromosomal regions). We found trans-eQTL at 56 non-HLA loci that significantly (P < 1 × 10−8) affected expression of some 2,125 transcripts. Finally, we developed a new approach which allowed us to define distinct genetic subtypes of T1D, and applied this method to other diseases. In summary, our results provide novel insights into the complex genetics of T1D and identified candidate genes for further investigation. Our approach can be applied to other complex genetic diseases for which GWAS data are available.