Abstracts

Rationale:
Variants in SCN2A have been associated with various neurological disorders, including intellectual disability, autism spectrum disorder and epileptic encephalopathies (EE). Epilepsy-associated mutations of SCN2A result in a range of disorders, spanning from self-limited infantile seizures to early-onset EE, like Ohtahara syndrome and infantile spasms. The SCN2A-p.E430A recurrent variant results in early infantile developmental EE. Seizure types include focal, myoclonic, absence and tonic as well as abnormalities in background EEG. To model SCN2A-p.E430A in mice, we introduced E430A at the corresponding position in mouse Scn2a. Following molecular validation, we systematically assessed seizure susceptibility.

Methods:
Scn2aE430A mice on the C57BL/6J strain (Jackson, #000064) were generated using CRISPR/Cas9. Founder and N1 offspring were sequenced for desired and off-target events. Male and female Scn2aE430A and WT littermates from generation ≥N2 were used in experiments. Transcript and protein expression were assessed by RT-ddPCR and immunoblotting (n=4-9/genotype, P60-81). Susceptibility to seizures induced by kainic acid (KA; 25 mg/kg, IP) was assessed using a modified Racine scale. Latency to each stage and highest stage reached within 5-minute bins were recorded (n=15/sex/genotype, P41-74). Susceptibility to seizures induced by flurothyl was assessed, including latency and end-point characterization (n=20-21/sex/genotype, P71-84). Maximal electric shock (MES) test was conducted at 60 Hz, 0.5 pulse width, 0.4 sec duration and 32 mA current with corneal electrodes (Ugo-Basile). Scn2aE430A mice were treated with 0.5% methylcellulose or 15 mg/kg phenytoin (IP) 2 hours prior to shock (n≥7/treatment, P56-72).

Results:
We confirmed the on-target E430A event and absence of off target events. There were no differences in Scn2a transcript or protein expression between WT and Scn2a^E430A (p >0.17, T-test). Scn2a^E430A mice were more susceptible to KA induced seizures. Scn2a^E430A quickly reached stage 6 (42.5±5.3 min) while average latency was slower for WT (74.6±9.7 min) (p< 0.0001, two-way ANOVA, Sidak’s). Furthermore, Scn2a^E430A were less likely to survive KA administration compared to WT (p< 0.0001, Mantel-Cox). There was no difference in latency to flurothyl induced seizures between WT and Scn2a^E430A mice (p >0.20, Mann-Whitney). However, 83% of Scn2a^E430A mice died following flurothyl exposure compared to 17% of WT littermates (p< 0.0001, Chi-square). Scn2a^E430A mice required a lower electroconvulsive stimulus compared to WT in the MES assay. Furthermore, pre-treatment with phenytoin prevented maximal seizures and death in the MES assay (p< 0.0001, Fisher’s exact).