Activating Mutations in Kir6.2 and Neonatal Diabetes

New Clinical Syndromes, New Scientific Insights, and New Therapy

Andrew T. Hattersley; Frances M. Ashcroft


Diabetes. 2005;54(9):2503-2513. 

In This Article


The high prevalence of Kir6.2 mutations in permanent neonatal diabetes means that all children <6 months of age diagnosed with diabetes should be tested for Kir6.2 mutations at diagnosis and assessed for neurological features. Details of genetic testing are available at many centers, including Finding a mutation offers the possibility of discontinuing insulin and implementing sulfonylurea therapy while maintaining good glycemic control.[2,37,38,40,43] Early treatment with sulfonylureas may also limit neurological damage, even when it is unable to control the diabetes. Long-term surveillance and reporting of both short-term and long-term outcomes of all patients with Kir6.2 mutations receiving sulfonylureas is vital, as insulin injections remain a good treatment for diabetes and an alternative therapy must be clearly superior or help alleviate extrapancreatic symptoms.

The demonstration that Kir6.2 mutations cause neonatal diabetes has implications for future genetic studies. It shows that genes in which common polymorphisms predispose to polygenic type 2 diabetes are excellent candidates for severe mutations resulting in monogenic diabetes and vice versa. It provides another example of activating and inactivating mutations in the same gene producing opposing phenotypes, in this case hypoglycemia and hyperglycemia. This supports the idea that genes associated with hyperinsulinism are good candidates for diabetes and vice versa. It emphasizes the value of selecting proteins known to play important roles in β-cell function as candidate genes, especially when genetic manipulation in experimental animals results in diabetes or hyperglycemia.

The demonstration that Kir6.2 mutations can cause neonatal diabetes illustrates the value of genetic studies of children with diabetes diagnosed before 6 months of age and indicates that such patients who lack Kir6.2 mutations provide an excellent resource for new genetic studies. Potentially, these may identify novel genes involved in insulin secretion and/or provide information about critical stages in β-cell development and function. One obvious candidate gene for such studies would be SUR1. There is also the interesting possibility that mutations that cause even less increase in the KATP current may be associated with diabetes in later life, as observed for the C42R mutation.[44] Thus, other diabetic phenotypes, including maturity-onset diabetes of the young and atypical diabetes, should be examined for mutations in Kir6.2 to assess if there is a milder monogenic phenotype than neonatal diabetes. In this respect, it is interesting that the common E23K variant in Kir6.2 is consistently associated with type 2 diabetes in large-scale association studies (see below).[31,32,33]

In the past, monogenic studies have concentrated on large families with multiple affected members, as typically found in maturity-onset diabetes of the young. However, Kir6.2 mutations causing neonatal diabetes are predominantly de novo. Although it is not possible to map de novo mutations using reverse genetics, it is possible to rapidly establish that these mutations are likely to be causal when an affected proband born to unaffected parents has a mutation not present in the parents. The probability of such a result occurring by chance is very low, as the spontaneous mutation rate for any given nucleotide in an individual is estimated to be 1 × 109.

Multiple large genetic studies consistently show a moderate association of the E23K variant in Kir6.2 with type 2 diabetes, with an ~20% increased risk associated with inheriting the E23 allele.[31,32,33] Some of the smaller studies (<500 cases) did not detect an association,[63] probably because they were underpowered for an effect of this size. However, the meta-analyses of all studies performed to date is highly significant (P < 10-6),[32,33] and because the frequency of K allele is ~40% in Caucasians,[63] it constitutes a significant population risk. The genetic studies do not, in themselves, conclusively establish that the E23K poymorphism is the causal variant because of strong linkage disequilibrium across the gene. For example, European Caucasians who inherit the less common K allele also inherit the rarer allele at the A1369S polymorphism in the adjacent SUR1 gene.[32]

Functional studies suggest that overactivity of the KATP channel is likely to underlie the association of the E23K variant in Kir6.2 with type 2 diabetes. However, the mechanism by which this is achieved is controversial. A twofold reduction in the ATP sensitivity of Kir6.2/SUR1 channels[64] and Kir6.2/SUR2A channels[65] has been reported when E is mutated to K, yet a third study found little effect on ATP block.[66] Minor variations in experimental protocol or DNA species (human versus mouse) might account for this difference. It has also been shown that the E23K mutation enhances activation of Kir6.2/SUR1 currents by Mg-nucleoside diphosphates[67] and by long-chain acyl CoAs,[66] both of which lead to a small reduction in apparent inhibition by MgATP. Consequently, it is speculated that the K allele is associated with a small increase in the β-cell KATP current. In addition, recent studies suggest the E23K allele substantially reduced the ATP sensitivity of skeletal muscle KATP (Kir6.2/SUR2A) channels at acid pH,[65] implying the KATP current will be enhanced during muscle ischemia and exercise. This could contribute to impaired glucose homeostasis by reducing glucose uptake in muscle.

The functional effects of the E23K polymorphism on glucose homeostasis in humans are equally controversial. Although some studies support the idea that the E23K variant is associated with reduced insulin secretion,[32,68] others have failed to detect such an association.[69,70,71] It has also been suggested that the E23K variant may mediate its effect via changes in glucagon secretion.[70] It is likely that the different findings of these studies relate to the relatively small numbers of the patients examined and the differences in age of the subjects, as the positive studies are in older patients. More studies are clearly needed to clarify these issues.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.