Abstracts

Rationale: Heterozygous loss-of-function mutations in SCN1A, the gene that encodes the pore-forming region of NaV 1.1, are the predominant cause of Dravet syndrome (DS). DS is a severe form of childhood-onset epilepsy marked by treatment-resistant seizures and high risk of sudden unexpected death. Studies in animal models and a recent study in humans have demonstrated that these mutations strongly impair the intrinsic excitability and network signaling of GABAergic interneurons while exerting no noticeable influence on excitatory interneurons. These findings support the conclusion that GABAergic interneurons play an important role in the development of the seizures and sudden death in DS. Other studies have shown that parvalbumin- and somatostatin-positive interneurons (two major classes of interneurons which encompass nearly 80% of all cortical interneurons), work synergistically to generate the seizure and mortality phenotypes observed in the global knock-out (KO) model of Dravet. We examined the correlation between the extent of a Scn1a mutation across all types of interneurons and the severity of seizure and mortality phenotypes in DS mice. Methods: We generated four lines of mutant mice by crossing Scn1a floxed mice with different Cre driver mice: 1) Mice with Scn1a KO in forebrain interneurons alone (using Dlx12Cre) and those with KO in both forebrain and brainstem interneurons (using Gad2Cre) to examine the aforementioned correlation.  2) Mice with Scn1a KO restricted to acetylcholine expressing neurons (using ChatCre) to examine the specificity of the correlation to interneurons. 3) Mice with Scn1a KO in all CNS excitatory and inhibitory neurons (using NestinCre which targets the mutation to all progenitor neurons) to compare the effects of above KOs and that of both excitatory and inhibitory neurons. All mice were maintained on the C57BL/6J background. Each mouse line was monitored for sporadic death, thermal seizure susceptibility, as well as seizure susceptibility in the presence of two different doses of the GABA-A agonist pentylenetetrazol (PTZ). Results: Mice with combined forebrain-brainstem interneuron KO of Scn1a had more severe phenotype than those with Scn1a KO in forebrain interneurons alone. Gad2Cre-Scn1a KO mice had frequent sporadic death, with nearly no survivor passed post-natal day 60. They exhibited thermal seizures in 50% of the mice at 40 degrees C of core body temperature and none of them was seizure free when the temperature reached 42 degrees C.  When exposed to lower dose of PTZ (non-convulsive for controls), these mice showed a short delay to the first generalized tonic-clonic seizure, higher incidence of seizures and related death.  On the other hand, Dlx12Cre-Scn1a KO mice did not experience any spontaneous death, thermal seizures, PTZ-seizures or related death. Interestingly however, these mice exhibited an elevated incidence of myoclonic seizures. ChatCre-Scn1a KO and littermate control mice for each mouse line exhibited a nearly similar response to Dlx12Cre mice, except that they did not display any PTZ induced myoclonic seizures. Finally, NestinCre-Scn1a KO mice displayed similar results as the Gad2Cre KO mice, except that they had an elevated mortality rate associated with PTZ-seizures. Conclusions: These results suggest that the extent and regions of interneuron-specific Scn1a KO in the brain may positively correlate with the severity of epilepsy and sudden death in DS mice. Therefore in the event of a mosaic mutation, the extent of the distribution and brain regions affected by SCN1A mutation may predict the severity of these DS phenotypes. Funding: Ellenbogen Chair Research Funds, UW Department of Neurological Surgery.Ryan Murphy SUDEP  Research Funds