A common variant near the KCNJ2 gene is associated with T-peak to T-end interval.

BACKGROUND: T-peak to T-end (TPE) interval on the electrocardiogram is a measure of myocardial dispersion of repolarization and is associated with an increased risk of ventricular arrhythmias. The genetic factors affecting the TPE interval are largely unknown. OBJECTIVE: To identify common genetic variants that affect the duration of the TPE interval in the general population. METHODS: We performed a genome-wide association study on 1870 individuals of Finnish origin participating in the Health 2000 Study. The TPE interval was measured from T-peak to T-wave end in leads II, V(2), and V(5) on resting electrocardiograms, and the mean of these TPE intervals was adjusted for age, sex, and Cornell voltage-duration product. We sought replication for a genome-wide significant result in the 3745 subjects from the Framingham Heart Study. RESULTS: We identified a locus on 17q24 that was associated with the TPE interval. The minor allele of the common variant rs7219669 was associated with a 1.8-ms shortening of the TPE interval (P = 1.1 × 10(-10)). The association was replicated in the Framingham Heart Study (-1.5 ms; P = 1.3 × 10(-4)). The overall effect estimate of rs7219669 in the 2 studies was -1.7 ms (P = 5.7 × 10(-14)). The common variant rs7219669 maps downstream of the KCNJ2 gene, in which rare mutations cause congenital long and short QT syndromes. CONCLUSION: The common variant rs7219669 is associated with the TPE interval and is thus a candidate to modify repolarization-related arrhythmia susceptibility in individuals carrying the major allele of this polymorphism.

Loss-of-function mutations of the K(+) channel gene KCNJ2 constitute a rare cause of long QT syndrome.

Mutations of the KCNJ2 gene encoding the potassium channel Kir2.1 were previously shown to cause Andersen's syndrome (AS), a multisystem disease manifesting with developmental abnormalities, cardiac arrhythmias and periodic paralyses. We conducted a search for KCNJ2 mutations among 188 unrelated patients suspected to have long QT syndrome (LQTS). The screening was performed by denaturing high-performance liquid chromatography (dHPLC) and DNA sequencing. Two novel mutations of the KCNJ2 gene were detected: a missense threonine to alanine mutation (T75A) in the N-terminal region (family 1) and an in-frame deletion of two amino acids (DeltaFQ163-164) in the M2 transmembrane region (family 2). In addition, a previously described silent polymorphism C1146T was detected. In family 1, some of the affected family members had a history of periodic muscle weakness characteristic of AS, but no dysmorphic features. The mean QTc interval of the affected members were 444 +/- 24 ms (family 1, n=7) and 456 +/- 8 ms (family 2, n=2). The mutations affect functionally important regions of the KCNJ2 channel protein: upon injection of the Xenopus oocytes with the wild type and mutant KCNJ2 constructs, the channel proteins were correctly synthesized and localized to the cell surface, but no measurable inward K(+) current could be detected for the mutant KCNJ2 constructs. In conclusion, we report two novel loss-of-function mutations of the KCNJ2 channel, affecting different domains of the channel protein. Mutations of the KCNJ2 gene should be considered in genetic subclassification of LQTS patients, even in the absence of overt manifestations of AS.