HbA1c, an index of chronic glycemic status relevant to IR and T2D, is associated with aging, cognitive performance and mortality. To identify epistatic interactions, we assessed 65 published SNPs associated with levels of HbA1c (n=4,090) and change in HbA1c levels over time (n=2,054) among nondiabetic participants of European ancestry who had complete data from 2 exams collected 7 years apart in the LLFS, a multicenter family study with a clustering of healthy aging and exceptional survival. Both traits were adjusted for age, sex, BMI, field center and principal components, with correction for familial membership. HbA1c change over time was derived using growth curve modeling, taking into account baseline HbA1c. A knowingly more powerful Bayesian approach was used to test SNP association searching for significant (Bayes Factor, BF>3) single SNPs and best (top 1) combinations of interacting SNPs followed by MIXED regression reconfirmation (p<0.05 for main and interaction terms). HK1-rs17476364 (BF=7.8), GCK-rs4607517 (BF=5.7), HK1-rs16926246 (BF=4.8), SPTA1-rs12041363 (BF=4.5), G6PC2-rs560887 (BF=3.6) and SPTA1-rs857691 (BF=3.0) were individually associated with baseline HbA1c. Their best and significant 2-SNP association was found between HK1-rs17476364 and GCK (BF=13.5; p=0.000015, 0.0038 and 0.0093 for main and interaction terms, respectively). The emerged best 3 and 4 SNP combinations were HK1-GCK-SPTA1 (BF=18.1) and HK1-GCK-SPTA1-G6PC2 (BF=22.0), respectively. For HbA1c change over time, no single SNPs were significant; their best SNP combination involved G6PC2, ADCY5, SLC30A8, HK1, PHB2 and FN3KRP. Glucokinase (GCK product) and hexokinase (HK1 product) are isozymes that interactively and uniquely phosphorylate glucose to make G6P, an important intermediate enabling the coordination of the rate of glycolysis. Together, HK1-GCK interaction for HbA1c was observed first in a large family data with familial clustering of exceptional longevity.


P. An: None. B. Thyagarajan: None. J.H. Lee: None. N. Schupf: None. J.M. Zmuda: None. M. Province: None.


National Institutes of Health

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.