Genetic Epidemiology of Diabetes and Obesity

National Institute Of Diabetes And Digestive And Kidney Diseases

Investigators

Linked publications, trials & patents

Abstract

In the current project, genetic determinants of type 2 diabetes mellitus and obesity are being sought using techniques of genetic linkage and association analysis. Lymphoblast cell lines have been established from informative pedigrees. DNA is available from other families in nuclear pellets extracted from blood specimens obtained in the epidemiologic studies and is amplified by whole genome amplification when needed. Genome-wide mapping studies, initially using linkage methods and more recently using association methods, are being used to identify regions containing variants conferring susceptibility to diabetes and obesity in other Indigenous populations. Exhaustive association analyses are being conducted of regions identified by these approaches and of other candidate genes in an attempt to help identify causative variants. Whole genome sequencing studies are also being pursued, as are analyses of epigenetic factors (e.g. DNA methylation). Several candidate genes that have been associated with type 2 diabetes in other populations have been evaluated for association in Indigenous Americans. In general most "established" variants associated with type 2 diabetes and obesity in other populations are also associated in Indigenous Americans, albeit with small effects which are often not individually statistically significant. Some variants identified in other populations (e.g. TCF7L2) appear to have little effect. Variants in established type 2 diabetes genes, MOB2, KLF14 and KCNQ1, are subject to parent-of-origin effects and these parent-of-origin effects replicate in American Indians. The effect of the KCNQ1 variants is particularly strong, with an odds ratio approaching 2.0 if the risk allele is inherited form the mother. An R1420H variant in ABCC8, an established MODY gene which also causes hyperinsulinemic hypoglycemia of infancy, also increases diabetes risk approximately two-fold in some indigenous populations; the risk allele frequency is ~1.5% while it is exceedingly rare in most other populations. Genome-wide association studies have also been conducted- initially using standard commercial genotyping arrays and selected samples, and, more recently, in larger population samples (N=11,000) using Amerindian-specific arrays developed from whole genome sequence data in 403 Indigenous Americans, as have exome sequencing studies in ~8,000 individuals. These studies have identified several additional potential susceptibility genes for diabetes and for obesity. Recent studies have shown that variants in the CELSR2/PSRC1/SORT1 cluster associate with LDL cholesterol and resting metabolic rate. Studies have shown that components of metabolic syndrome strongly aggregate within families, both within and across different components. Studies also showed the polygenic risk scores for type 2 diabetes strongly associated with diabetes incidence in this population and that these scores may confer potentially meaningful, albeit modest, benefits in predicting diabetes incidence; further studies of polygenic scores are ongoing. Currently fine-mapping studies with additional variants are being conducted to extract more of the genetic information in regions identified as potentially involved in diabetes susceptibility. Through collaborations, studies are being conducted to determine if any of the signals identified in the present mapping studies replicate in other populations. Variants reproducibly associated with type 2 diabetes and obesity from other populations continue to be typed to determine their role susceptibility to diabetes and obesity in American Indians. Whole genome sequencing is also being conducted in a small number of participants. Population-based linkage approaches are also being explored as a complementary mapping strategy. Additional participants are being recruited for replication studies.