KCNQ1, a susceptibility gene for type 2 diabetes

In 2001, the Millennium Genome Project in Japan was established to identify susceptibility genes for five diseases, including type 2 diabetes mellitus, as a national undertaking. In 2002, the consortium, composed of 11 core facilities located in various regions of Japan, began a multistage genome-wide association study to identify disease-associated single-nucleotide polymorphisms (SNPs) for type 2 diabetes among a collection of 10,000 standard Japanese polymorphisms. Three consecutive screenings resulted in the identification of 10 SNPs with a P value for association with type 2 diabetes of <0.05. Three of these 10 SNPs were located in intron 15 of KCNQ1, and one of these three, rs2237892, showed the most significant association. Further investigation of KCNQ1 revealed the significant association of rs2237892 with type 2 diabetes not only in two additional Japanese subject panels but also in Chinese, Korean, and Swedish panels. The final data set with a total of 19,930 individuals (9569 cases and 10,361 controls) yielded a P value of 1.7 · 10 and an odds ratio of 1.40 (95% confidence interval, 1.34–1.47) for rs2237892. Another Japanese group also independently identified KCNQ1 as a susceptibility gene for type 2 diabetes. To date, KCNQ1 has been confirmed as a susceptibility gene for type 2 diabetes not only in East Asians and Europeans but also in Mexican-Americans and South Asians, indicating that the disease susceptibility it confers is transethnic. KCNQ1 encodes the pore-forming subunit of a voltage-gated K channel that is essential for the repolarization phase of the action potential in cardiac muscle. Mutations of this gene are also associated with hereditary long QT syndrome and familial atrial fibrillation. In addition to heart muscle, KCNQ1 is expressed in tissues including the brain, adipose tissue, and pancreas as well as in the insulin-secreting cell line INS-1. The risk allele of KCNQ1 for type 2 diabetes is also associated with impaired insulin secretion, suggesting that this impairment might underlie the diabetes susceptibility conferred by the allele. Attenuation of KCNQ1 channel activity by the selective inhibitor chromanol 293B was found to enhance insulin secretion induced by tolbutamide in INS-1 cells. Furthermore, forced expression of KCNQ1 in the MIN6 mouse b-cell line resulted in impairment of insulin secretion induced by glucose, pyruvate, or tolbutamide. These data are consistent with the idea that increased KCNQ1 function impairs insulin secretion by inducing premature repolarization of the action membrane potential in pancreatic b cells. The risk allele of KCNQ1 may therefore increase the expression of KCNQ1 in pancreatic b cells and thereby promote the development of type 2 diabetes. KCNQ1 is also an imprinted gene. Genomic imprinting is a phenomenon whereby a subset of autosomal genes is monoallelically expressed in a manner dependent on parental origin, with KCNQ1 being expressed from only the maternally derived chromosome. Furthermore, KCNQ1 contains KvDMR1, an imprinting control region within intron 10 of the gene. KvDMR1 is methylated on the maternal chromosome but not on the paternal one, and is transcribed from the latter to yield the noncoding antisense RNA KCNQ1OT1 (Figure 1). This RNA is thought to be responsible for the repression of KCNQ1 and neighboring genes on the paternal chromosome (Figure 1). Although functional imprinting is known to be highly tissue and developmental stage specific, the rs2237892 risk allele of KCNQ1 was found to be significantly associated with diabetes when maternally transmitted but not when paternally transmitted, suggesting that KCNQ1 and ⁄or its neighboring gens are indeed imprinted in human pancreatic b cells.