Abstracts

Rationale: Mutations in DNM1, encoding a GTPase involved in endocytosis, have been implicated in epileptic encephalopathy (EE). Dnm1Ftfl/Ftfl mice mimic the severe intractable seizures and comorbidities observed in patients with DNM1 mutations. Due to the dominant negative nature of DNM1 and Dnm1Ftfl mutations, a possible strategy is to eliminate the toxic allele utilizing a microRNA (scAAV9-miDnm1a), hypothesizing that this will alleviate the severe seizure and developmental phenotypes observed in our mouse model and provide a possible therapeutic option for patients with DNM1 mutations. Methods: Dnm1Ftfl/Ftfl and Dnm1+/+ mice on both the inbred C57BL6/J and (FVB X C57BL6/J)F2 hybrid backgrounds were treated with varying concentrations (ranging from 1E10 to 6E11 viral particles) of scAAV9-miDnm1a, scAAV9-eGFP or saline via a single intracerebroventricular injection at birth. Knockdown of Dnm1a, and thus Dnm1Ftfl, was assessed via RT-qPCR. Survival and development were observed daily. Statistical analyses were executed using ANOVA followed by post hoc analysis using Tukey’s HSD, Students’ t-test, and the Mantel-Cox log-rank test. Results: Dnm1Ftfl/Ftfl mice on both backgrounds show significant neurodevelopmental deficits similar to those observed in patients, with some DNM1 phenotypes more exacerbated and/or specific to each background. Dnm1Ftfl/Ftfl mice show significant growth defects from postnatal day 8 (PND8) in B6J (n=15, p=0.0007) mice, and from PND 6 in F2 hybrid mice (n=11, p=0.0001) until eventual lethal seizures between PND18-20. Sensory motor delay was prominent in B6J mice from P7 (p=0.018), while hypotonia was observed in the F2 hybrid mice (0.0069). Interestingly, treatment with scAAV9-miDnm1a (n=6 to 11, depending on the experiment) reduced the growth deficits observed in Dnm1Ftfl/Ftfl on both backgrounds, albeit not to wildtype levels, mitigated the sensory motor delay observed in Dnm1Ftfl/Ftfl B6J mice at P7 and P9 (p=0.010, p= 0.0051), and rescued the hypotonia observed in the F2 hybrid background (p=0.025) compared to homozygous controls. Furthermore, neurodevelopment of Dnm1Ftfl/Ftfl treated mice did not significantly differ from wildtype controls. Notably, treatment with scAAV9-miDnm1a decreased seizure occurrence and extended the median lifespan of homozygotes by 3-10 days across increasing viral treatment concentrations and backgrounds, with as high as 72% survival past weaning of mice treated with the highest viral dose thus far. Conclusions: Although this study is still ongoing, there is enough evidence to suggest a plausible treatment approach not only for DNM1 EE but other dominant negative EE mutations that progress during the early postnatal period. This study pioneers the potential use of AAV9-mediated gene silencing therapy for EE due to gain-of-function or dominant negative mutations. Importantly, DNM1 mutational silencing is viable across strains and specifically, in the genetically diverse F2 hybrid background, boding well for potential clinical application. Funding: NIH grant R37NS31348