COLD SPRING HARBOR, NY (GenomeWeb) – Researchers from the US and Finland have identified a small subset of regulatory signatures in skeletal muscle that coincide with the presence of type 2 diabetes traits.
At the Biology of Genomes meeting here yesterday, National Human Genome Research Institute researcher Stephen Parker presented findings from a regulatory analysis of type 2 diabetes traits done for the Finland United States Investigation of NIDDM Genetics (FUSION) study.
For that arm of the study, the team started with more than 300 Finnish individuals with or without type 2 diabetes or related glucose profiles and focused on skeletal muscle, where some 80 percent of insulin stimulated glucose uptake takes place.
For 278 of the participants, the researchers did both array-based genotyping and strand-specific, paired-end RNA sequencing on muscle biopsy samples to glean genotyping and gene expression profiles, respectively, in that tissue type. That was further complemented by information on individuals' fasting glucose levels, glucose tolerance, medical history, and the like.
In their subsequent search for variants influencing the splicing or expression of nearby genes, the researchers combined RNA sequence data with genotyping information at nearly 7.7 million directly genotyped or imputed SNPs — an investigation that revealed expression or splicing quantitative trait loci associated with more than 8,000 genes.
The eQTL side of the analysis centered on cis-eQTLs influencing the expression of local genes, Parker explained, since additional power would be needed to investigate more distant trans-eQTL effects.
When they looked at the distribution of apparent cis-eQTLS with respect to protein-coding sequences, pseudogenes, long intergenic non-coding RNAs, and antisense sequences, the researchers found that most fell within transcription start sites, consistent with their proposed regulatory roles.
And while most of the eQTLs appeared to be active in skeletal muscle regardless of type 2 diabetes status, around 1 percent of the eQTLs exhibited more state-specific regulatory roles.
The cis-eQTL sites in skeletal muscle seemed to be enriched at or near SNPs detected in past genome-wide association studies of type 2 diabetes or related traits such as body mass index and fasting glucose levels. The extent of that overlap was statistically significant in the case of the BMI and fasting glucose GWAS, Parker noted, but was slightly more tenuous for GWAS of type 2 diabetes itself.
Nevertheless, the analysis pointed to overlap between a SNP detected by GWAS for type 2 diabetes and a cis-eQTL regulating ANK1, a gene that's differentially expressed between individuals with type 2 diabetes and those with normal glucose tolerance.
Another variant on chromosome 11 affected the expression of a gene called TTNT3 in muscle tissue from individuals with normal glucose tolerance, Parker noted, but did not seem to have a cis-eQTL role in individuals with impaired glucose tolerance or type 2 diabetes.
The team also saw a SNP that specifically decreases the expression of a specific gene in individuals with impaired glucose tolerance or type 2 diabetes. Still other cis-eQTLs were active in the normal and impaired glucose tolerance groups but not in the type 2 diabetes cases.
The team plans to continue mining the Finnish muscle samples and other data sources for additional information on the regulatory processes at play in individuals with diabetes or related traits using both additional cell types and experiments focused on other regulatory features such as chromatin state.
"This rich data resource enables identification of diverse molecular processes involved in muscle-based insulin resistance, changes in transcription with progression towards [type 2 diabetes], and reveals mechanistic insights about disease predisposition," Parker and his co-authors wrote in the abstract accompanying the Biology of Genomes presentation.