Circulating levels of lipids, glucose, and insulin result from complex and interwoven physiological mechanisms, are indicators of type-2 diabetes (T2D) and cardiovascular disease (CVD) risk, and may also be implicated in the greater risk of T2D among statin users. Although high triglyceride levels are considered a risk factor for the development of T2D, a recent and novel finding emerging from our group suggests that genetic susceptibility to elevated triglyceride levels is protective of T2D. This counter-intuitive finding, and similar findings by other groups, suggests that the link between lipid and glycemic traits is complex. Here we propose that genomic and statistical approaches that leverage the context-dependency of genetic effects can afford us greater power to identify novel loci acting at the interface of lipid and glycemic traits, improve our understanding of currently known loci, and provide insight into the specific mechanisms underlying this interface, and cardiometabolic disease in general. We thus hypothesize that there are additional loci, beyond the univariate GWAS-identified ones, which are associated with both lipid and glycemic traits, in similar and/or opposite directions. Secondly, we hypothesize that the association between genetic factors and lipid levels differs according to an individual's level of insulin resistance, and that the association between genetic factors and glycemic levels differs according to an individual's level of dyslipidemia. Thirdly, we hypothesize that there are genetic loci that could predict the extent to which someone taking lipid-lowering medications, such as statins, is likely to progress to T2D. Finally, we will follow-up our findings with further refined phenotypes, and examine associated methylation and gene expression patterns, and putative regulatory function. To test these hypotheses, we will use multiple large genomic datasets obtained through the Database of Genotypes and Phenotypes (dbGaP), and from the UK Biobank. The latter dataset notably comprises longitudinal and detailed phenotypic data, along with genome-wide data on 500,000 individuals - an unprecedented resource in genomic research. The use of prospective cohort studies from these sources will allow us to leverage 1) multiple existing sets of broad and detailed phenotypic data, including lipoprotein sub-fraction measurements, and 2) the availability of longitudinal data on glycemic and lipid measurements, along with information on medication use. It is anticipated that this project will 1) contribute to our understanding of the biological links between glycemic and lipid traits, which are major early indicators of cardiometabolic disease, 2) identify genes that may be up- or down-regulated in the context of insulin resistance and dyslipidemia, 3) identify genetic variants that may affect glycemic response to cholesterol-lowering drugs, and 4) provide insight into the underlying biological mechanisms. Our findings may lead to more fine-tuned and early assessments of T2D and CVD risk, and the development of targeted prevention and treatment strategies.
Type-2 diabetes (T2D) and cardiovascular disease are highly intertwined diseases. In this proposal, we aim to use multiple large datasets to identify genetic variants and associated pathophysiological processes that underlie the connection between lipid and glucose/insulin traits, including the higher T2D risk observed among statin users. This project will improve our understanding of the genetic and pathophysiological link between lipid and glycemic traits, and eventually our ability to incorporate genetic and clinical risk information to more precisely target prevention and treatment strategies.