Abstracts

Rationale: SUDEP prevention is hindered by a lack of available pharmacological treatment options. Here, we examine the therapeutic efficacy of the FDA-approved KCNQ2/3 opener retigabine (RTG) in preventing seizures and sudden death in two models of SUDEP: Kcna1 knockout and Kcnq1A340E knockin mice. Mice lacking the Kcna1 gene encoding voltage-gated Kv1.1 potassium channels exhibit severe early-onset seizures, brain-driven cardiac arrhythmias, and premature death. The Kcnq1 gene encodes voltage-gated Kv7.1 potassium channels. Mice carrying the Kcnq1A340E mutation exhibit spontaneous seizures and cardiac abnormalities that occur concomitantly with cortical discharges. The homologous mutation in humans is associated with sudden death due to ventricular tachyarrhythmias related to long QT syndrome. Given the ability of KCNQ openers to reduce epileptiform activity, we hypothesized that administration of RTG would prevent seizures and protect Kcna1 and Kcnq1 mutant mice from SUDEP. Methods: Behavioral responses to RTG were monitored in mice (4-6 weeks old; n?-3/genotype) for 40 minutes post-injection. Seizure thresholds were determined in mice (P29-31; n?-6/genotype) by measuring the latency to seizure following exposure to the convulsant flurothyl. The effect of RTG on survival was monitored in Kcna1 knockout mice (10 mg/kg/day, ip; n?-20/treatment) between the ages of P15 to P36. Quantitative RT-PCR was performed to measure Kcnq2 and Kcnq3 mRNA levels in hemi-brains, cortices, and hippocampi of mice (P29-31; n?-3/genotype). Results: RTG administration caused wildtype mice to exhibit motionless, inactive behavior. In contrast, RTG caused behavioral hyperexcitability in Kcna1 knockouts, characterized by dose-dependent hopping, vocalization, and tremoring phenotypes (P=0.01), which were not observed in Kcnq1A340E mutants. RTG was ineffective at reducing the latency to flurothyl-induced seizure in Kcna1 knockouts suggesting no modification of seizure susceptibility. However, RTG significantly increased the latency to seizure in Kcnq1A340E mutants (P=0.05) suggesting the drug may reduce seizure susceptibility in patients who have long QT-associated mutations in KCNQ1. Daily administration of RTG had no significant effect on lifespan in Kcna1 knockout mice, further supporting a lack of effectiveness in that genetic model. qRT-PCR analyses revealed a potential increase in Kcnq2 gene expression, but not Kcnq3, in the hemi-brains of Kcna1 knockouts. Analyses of regional dissections revealed a significant 37% increase in Kcnq2 levels in Kcna1 knockout hippocampus (P=0.01) suggesting a potential molecular cause for the differential drug effects. In contrast, Kcnq2 and Kcnq3 mRNA levels were not significantly altered in Kcnq1A340E brains. Conclusions: The differing pharmacogenetic interactions between RTG and the Kcna1 and Kcnq1 mutations demonstrates that genetic profile may significantly alter the drug's therapeutic efficacy, emphasizing the need for personalized genome-based therapy in epilepsy. Funding: Citizens United for Research in Epilepsy, Epilepsy Foundation, and Ike Muslow Predoctoral Fellowship.