Vinsamlegast notið þetta auðkenni þegar þið vitnið til verksins eða tengið í það: http://hdl.handle.net/1946/12063
The effects of known type 2 diabetes susceptibility loci on complications of the disease
Áhrif þekktra erfðamarka sykursýki 2 á fylgikvilla sjúkdómsins
The prevalence of type 2 diabetes (T2D) continues to grow throughout the world with the subsequent increase in burden of diabetic complications, including microvascular and macrovascular diseases of the heart, brain, kidneys and eyes. To date, between 40 and 50 genetic variants have been discovered which increase the risk of T2D. These variants may contribute to the development of T2D complications in independent ways but few studies, if any, have systematically investigated the effect of known T2D susceptibility variants on complications of the disease, which was the aim of this study.
The effects of 47 T2D susceptibility variants selected from genome-wide association studies (GWAS) on T2D and its associated phenotypes were examined in the population based AGES- Reykjavík cohort, and the REFINE-Reykjavík cohort was used for replication of observed results when possible. Association analysis was conducted for 30 metabolic, anthropometric, blood lipid, cardiovascular, renal, retinal and brain volume variables. The TCF7L2 locus was given special attention in the analysis, as it harbors the strongest risk variants for T2D identified to date, and tested for modulating factors of its effect. A genetic risk score was constructed from the total set of single nucleotide polymorphisms (SNPs), and different subscores created based on their individual associations with beta cell function (BCF), as estimated by HOMA-B, or insulin resistance (IR), as estimated by HOMA-IR, both from literature references and those observed in this analysis.
The majority of the variants (38 of 47) showed the expected direction of effect for T2D in the multiple regression analysis, and 12 showed a statistically significant association with the disease. Six variants with unknown role in T2D pathogenesis were found to possibly influence BCF and three were identified as possible IR risk variants. Previously reported single SNP associations with BMI, triglycerides, HDL and coronary calcium were corroborated in this analysis and one novel association was observed with decreased waist circumference. A number of other novel associations were observed where P < 0.05 but none survived correction for multiple testing. Additional analysis on the effects of the TCF7L2 locus showed that there was no interaction between the genotype and BMI or blood lipids on T2D or coronary heart disease (CHD) risk. The variant was associated with CHD risk in REFINE-Reykjavík, but not in the AGES-Reykjavík cohort.
Finally, a risk score consisting of 47 T2D susceptibility gene variants showed a strong association with fasting glucose, T2D and HOMA-B. Interaction analysis suggested that its effect on fasting glucose levels might be modulated by BMI. The score was associated with greater increase in fasting glucose in obese individuals compared to overweight or normal individuals. The same effect was not observed for T2D risk, possibly because of reduced statistical power. When the variants were partitioned by their proposed physiologic mechanism in the disease, a strong association was found with decreased HOMA-B for one set of SNPs and with increased HOMA-IR for another. The genetic risk score associated with IR, but not the other genetic risk scores, showed a strong association with chronic kidney disease (CKD). Mendelian randomization analysis suggested a causal effect of IR on impaired glomerular filtration rate. The association was statistically significant after the exclusion of diabetic individuals when adjusted for BMI and triglyceride levels, indicating that the effect might be independent of T2D. These results could suggest that IR is an important factor in the pathogenesis of CKD, even before the onset of T2D, and that individuals with genetic susceptibility for IR might be predisposed to renal damage. The association between the IR associated variants and CKD should be investigated in a cohort comparable to AGES-Reykjavík with regard to age and CKD prevalence, to confirm these findings.