Abstracts

Rationale: Mutations in KCNQ2 gene, encoding for slowly-activating and deactivating voltage-dependent potassium current, cause neonatal epilepsies with a spectrum of phenotypic severity. KCNQ2 encephalopathy is a recently described severe form of epileptic encephalopathy typically involving neonatal seizure onset and subsequent global developmental impairment. A subgroup of individuals in some benign familial neonatal epilepsy (BFNE)-pedigrees also has been reported to show some delays in development. Y284C and Y284N, two variants at the same amino acid found in a family with BFNS and a de novo case of epileptic encephalopathy, respectively, are studied in the current project to investigate the molecular basis for this genotype-phenotype relationship to aid clinical course prediction and pharmacological treatment. Methods: Clinical information regarding occurrence of Y284C in a BFNE pedigree was reported previously1,2. A patient with epileptic encephalopathy and a de novo Y284N variant was reported to the RIKEE registry (www.RIKEE.org) at Baylor College of Medicine. Whole cell voltage-clamp electrophysiological recordings in Chinese Hamster Ovary (CHO) cells were used in the study to investigate the effect of mutations on current density, voltage-dependence of activation, and kinetics of KCNQ2 homomeric and KCNQ2/3 heteromeric channels. The variants were expressed in CHO cells to mimic the heterozygous state, i.e., as homomeric KCNQ2 tetramers in a 1:1 ratio with wild type (wt) KCNQ2, or as heteromeric KCNQ2/3 tetramers in a 1:2:1 ratio (wt KCNQ2/wt KCNQ3/mutant KCNQ2). Surface expression of channels was studied using biotinylation and western blotting. Ezogabine, a KCNQ2 opener, was tested for its effect on the rescue of channel function. Results: We found that potassium current densities were reduced by 30% in Y284C and 70% in Y284N in heterozygous heteromeric KCNQ2/3 tetramers. The voltage-dependence of activation was not significantly affected. Surface expression of channels was not significantly altered by the mutants. Experiments using potassium channel opener drugs to correct the potassium current levels are underway. We reviewed functional data on pore domain variants in KCNQ2 studied to so far to develop criteria for predicting phenotype severity. Conclusions: Our results provide further validation that, for many KCNQ2 variants, the extent of functional current reduction measured by in vitro heterologous expression is correlated with the phenotypic severity of the disease. For missense variants within the pore domain, combining early clinical phenotype information and variant location provides strong predictors of phenotype severity, which are much strengthened by heterologous functional screening. Funding: NINDS R01 NS49119 and CURE pediatric Epilepsy Grant