Type 2 diabetes risk allele near CENTD2 is associated with decreased glucose-stimulated insulin release

By combining multiple genome-wide association (GWA) studies and comprehensive replication efforts, 12 novel type 2 diabetes associated loci have recently been discovered. Here we evaluate the effect of lead variants of these loci on estimates of insulin release and insulin resistance derived from an oral glucose tolerance test. We examined 12 lead variants in or near HMGA2, CENTD2 (also known as ARAP1), KLF14, PRC1, TP53INP1, ZBED3, ZFAND6, CHCHD9, DUSP9, KCNQ1, BCL11A and HNF1A in 5,722 middle-aged people from the population-based Inter99 sample. Carriers of the major diabetogenic allele of rs1552224 in CENTD2 had increased 30-min plasma glucose values (2.0%, p = 2 x 10(-5)) as well as 4.2% reduced insulin release 30 min after an oral glucose load (p = 0.001). Risk allele carriers also had decreased BIGTT-acute insulin release (AIR), which is a surrogate measure of insulin release where sex, BMI, plasma glucose and serum insulin are integrated (5.3%, p = 8 x 10(-7)). In addition, a decreased corrected insulin response (CIR; 9.9%, p = 3 x 10(-8)) was observed. For rs5945326 near DUSP9 on the X-chromosome we stratified according to sex. Male carriers of the risk allele showed nominally decreased BIGTT-AIR (2.6%, p = 0.01). No associations with intermediate metabolic traits were found in women. For the remaining ten lead variants no consistent associations were demonstrated. Of the lead variants from 12 novel type 2 diabetes associated loci, CENTD2 significantly associated with increased plasma glucose values and decreased glucose-stimulated insulin release, suggesting that the diabetogenic effect of this locus is mediated through an impaired pancreatic beta cell function.