NEW YORK (GenomeWeb) – Researchers from the University of Virginia and elsewhere have fine-mapped type 1 diabetes susceptibility loci and found that causal variants appear to be co-localized with enhancers for lymphoid genes.
Using a custom genotyping array dubbed the Immunochip, UVa's Stephen Rich and his colleagues searched for additional T1D risk loci and mapped these loci to their most credible SNPs, as they reported in Nature Genetics today. This then allowed the investigators to examine where these SNPs were located in relation to regulatory sequences, as well as compare Immunochip results across immune-related diseases.
"The associated SNPs localized to enhancer sequences active in thymus, T and B cells, and CD34+ stem cells," Rich and his colleagues said in their paper. "Enhancer-promoter interactions can now be analyzed in these cell types to identify which particular genes and regulatory sequences are causal."
The Immunochip, a custom Illumina Infinium high-density genotyping array, includes SNPs and indels linked to 41 known T1D-associated regions, as well as some 3,000 so-called wildcard SNPs that tag candidate genes or other genes with evidence linking them to diabetes. For this study, the researchers used the chip to examine more than 138,000 SNPs in about 12,000 cases and controls.
From this, they linked 44 regions to T1D, 38 of which were previously known and four of which were novel. The other two, the researchers said, were only marginally associated with disease.
For each of these 44 loci, Rich and his colleagues searched for additional SNPs that might also be independently associated with diabetes. Based on this, they uncovered five loci with more than one independent SNP, including a novel one near INS, a finding the researchers found surprising, as that region has been well studied.
Rich and his colleagues also compared the loci underlying genetic susceptibility to T1D to those underlying 15 other immune-related diseases. From this, they found that T1D genetic susceptibility was most similar to juvenile idiopathic arthritis, as they share some risk loci, and that T1D genetic susceptibility was least like ulcerative colitis.
Drawing on this pleiotropy, the researchers unearthed three additional T1D-associated regions, bringing the count of known diabetes-associated regions to 57. One of these new regions encodes the TNFAIP3 gene.
For each of these seven loci newly linked to T1D, Rich and his colleagues combed through the US National Human Genome Research Institute genome-wide association study catalog to see whether they had been linked to other diseases or traits before. One variant, at 17q21.31 in the MAPT gene, was in strong linkage disequilibrium, they noted, with the index SNP for a number of neurodegenerative diseases, including Parkinson's disease.
They also delved more deeply into the primary and conditional signals found near each of the 44 loci that they linked to diabetes to create a list of credible SNPs within which the causal variant was likely to be lurking.
The researchers examined 15 chromatin states of 127 tissues from the ENCODE project, finding an enrichment of credible SNPs in enhancer chromatin states in immune-related tissues. They reported that thymus, CD4 and CD8+ T cells, B cells, and CD34+ stem cells had the strongest enrichment in multiple samples of each tissue or cell type. This suggested to Rich and his colleagues that variation in enhancer sequences is important in T1D.
"Enhancer-promoter interactions can now be analyzed in these cell types to identify which particular genes and regulatory sequences are causal," the researchers added.