Objective: Pathogenic variants of KCNQ2, which encode a potassium channel subunit, cause either benign (familial) neonatal epilepsy-B(F)NE)-or KCNQ2 encephalopathy (KCNQ2 DEE). We examined the characteristics of KCNQ2 variants.
Methods: KCNQ2 pathogenic variants were collected from in-house data and two large disease databases with their clinical phenotypes. Nonpathogenic KCNQ2 variants were collected from the Genome Aggregation Database (gnomAD). Pathogenicity of all variants was reevaluated with clinical information to exclude irrelevant variants. The cumulative distribution plots of B(F)NE, KCNQ2 DEE, and gnomAD KCNQ2 variants were compared. Several algorithms predicting genetic variant pathogenicity were evaluated.
Results: A total of 259 individuals or pedigrees with 216 different pathogenic KCNQ2 variants and 2967 individuals with 247 different nonpathogenic variants were deemed eligible for the study. Compared to the distribution of nonpathogenic variants, B(F)NE and KCNQ2 DEE missense variants occurred in five and three specific KCNQ2 regions, respectively. Comparison between B(F)NE and KCNQ2 DEE sets showed that B(F)NE missense variants frequently localized to the intracellular domain between S2 and S3, whereas those of KCNQ2 DEE were more frequent in S6, and its adjacent pore domain, as well as in the intracellular domain between S6 and helix A. The scores of Protein Variation Effect Analyzer (PROVEAN) and Percent Accepted Mutation (PAM) 30 prediction algorithms were associated with phenotypes of the variant loci.
Significance: Missense variants in the intracellular domain between S2 and S3 are likely to cause B(F)NE, whereas those in S6 and its adjacent regions are more likely to cause KCNQ2 DEE. With such regional specificities of variants, PAM30 is a helpful tool to examine the possibility that a novel KCNQ2 variant is a B(F)NE or KCNQ2 DEE variant in genetic analysis.
Keywords: KCNQ3; Kv7.2; genotype; infant; self-limiting epilepsy.
Wiley Periodicals, Inc. © 2019 International League Against Epilepsy.