The global burden of diabetes is staggering – over 400 million people currently live with the disease – as the worldwide prevalence is estimated to be nearly 9%. However, some populations exhibit substantially higher rates, as the prevalence is disproportionate across different races and ethnicities. For example, in the United States, the highest rate of diagnosed diabetes in the adult population is among American Indians/Alaskan Natives, followed by non-Hispanic blacks, Hispanics, Asians, and lastly, non-Hispanic whites. Such racial disparities suggest that genetic makeup may be an important underlying factor contributing to the development of this chronic disease. A limitation of previous genome-based research that identified genetic loci associated with diabetes risk factors is that many of the studies were conducted in populations of European descent. New results from a large genome-wide association study by the research groups of Drs. Charles Kooperberg and Ulrike Peters in the Public Health Sciences Division at Fred Hutch were recently published in the journal Diabetologia and shed light on our understanding of genetic variants associated with risk factors for diabetes in multi-ethnic populations.
The most common form of diabetes is type 2, which is generally characterized by insulin resistance and elevated blood glucose. A state of insulin resistance develops when the body’s cells do not respond normally to insulin signaling. However, if the pancreas can produce and secrete sufficient insulin to counteract the blunted signaling, blood glucose levels can be maintained within the nondiabetic range. Individuals can progress to full-blown diabetes when insulin secretion no longer keeps up with the increased insulin demand. Thus, elevated blood glucose and insulin levels typically precede the development of type 2 diabetes, and the measurement of fasting levels of either can aid in identifying individuals who may be at risk for developing the disease.
Dr. Stephanie Bien, a scientist in the Public Health Sciences Division, led the new study that sought to determine whether genetic variants associated with diabetes risk factors previously identified in populations of European descent would also be identified in a large-scale multi-ethnic study population. The authors utilized the Population Architecture using Genetic Epidemiology (PAGE) consortium and included over 26,000 non-diabetic individuals of Hispanic/Latino, African American, Asian and Pacific Islander, or American Indian/Alaskan Native descent. The association between fasting glucose and insulin levels and genetic variants, called single nucleotide polymorphisms (SNPs), were assessed in the study population.
Using genomic data from this transethnic population, the authors replicated findings in 31 of 39 known loci associated with fasting glucose and 14 of 17 known loci associated with fasting insulin that had previously been identified in populations of European descent. These loci included genes involved in lipid metabolism, glucose metabolism, insulin signaling, gene expression, and cellular signal transduction, among other biological processes. The results suggest that many genetic variants associated with diabetes risk factors are indeed consistent across diverse populations and largely generalizable to a wide population. However, with almost 20% of the loci not replicated in the multi-ethnic population, the results also demonstrate that the inclusion of diverse populations in large genome-wide association studies is essential to better understand the genetic contribution to disease risk.
Genomic regions found to be associated with specific traits or diseases may contain tens or even hundreds of individual SNPs. A large number of gene variants at a given locus impedes the ability to identify the potentially causal variants in follow-up laboratory-based studies. In this study, the authors used a fine-mapping method in an effort to reduce the number of SNPs at 15 loci associated with diabetes risk. This method successfully reduced the number of significantly associated SNPs by 72.5%, on average (see figure for example). “Importantly, differences in genetic architecture across populations enabled us to hone in on likely functional variants across the regions that were previously associated with fasting glucose and fasting insulin. This so-called fine-mapping of associated regions is important for improving our biological understanding of these traits,” says Bien of these results.
The authors also assessed whether new genetic variants associated with diabetic risk might be discovered in this diverse population. For this analysis, a panel of nearly 200,000 SNPs were analyzed for associations with the two diabetes risk factors. This analysis identified one new locus associated with fasting insulin that had not previously been reported. This association may be at least partially mediated by adiposity, as controlling for body mass index weakened the association.
Moving forward, Bien described a recently developed research tool that comprehensively assesses genetic variation across diverse ancestries, and results from a study that used this new tool to assess over 100 traits are currently being analyzed. Say Bien, “the results from this work suggest that the assertions made above for fasting glucose and fasting insulin hold true across examined traits, and that we now have good tools and statistical methodology to perform genetic association analyses in diverse ancestries. I wholeheartedly believe an emphasis on multi-ethnic genetic research as well as other -omics studies is critical to ensure that the benefits from these efforts are equitable. To continue biasing our studies towards those of European ancestry is hindering scientific discovery.”
Also contributing to this project from the Fred Hutch were Jeffrey Haessler and Drs. Chris Carlson and Ross Prentice.
Bien SA, Pankow JS, Haessler J, Lu YN, Pankratz N, Rohde RR, Tamuno A, Carlson CS, Schumacher FR, Buzkova P, Daviglus ML, Lim U, Fornage M, Fernandez-Rhodes L, Aviles-Santa L, Buyske S, Gross MD, Graff M, Isasi CR, Kuller LH, Manson JE, Matise TC, Prentice RL, Wilkens LR, Yoneyama S, Loos RJF, Hindorff LA, Le Marchand L, North KE, Haiman CA, Peters U, Kooperberg C. 2017. Transethnic insight into the genetics of glycaemic traits: fine-mapping results from the Population Architecture using Genomics and Epidemiology (PAGE) consortium. Diabetologia. 60(12):2384-2398.
Funding for this study was provided by the National Institutes of Health.