Abstracts

Rationale: SCN8A, which encodes the voltage-gated sodium channel Nav1.6, is broadly expressed throughout the central and peripheral nervous systems. An increasing number of de novo SCN8A mutations are being identified in patients with catastrophic, treatment-resistant childhood epilepsy. Interestingly, de novo SCN8A mutations have also been found in patients with autism, intellectual disability and developmental delay but with less severe epilepsy. Analyses of SCN8A mutations in heterologous systems have demonstrated varying effects on hyperexcitability. For example, Liu et al. (2019) observed a clear gain-of-function effect with the R1872W mutation, which is consistent with the severe seizure phenotype observed in patients and a conditional knock-in mouse model. However, other mutations (e.g. A1622D and R1620L) found in patients with behavioral abnormalities but without severe epilepsy exhibited both gain- and loss-of-function effects. To better understand the phenotypic spectrum associated with SCN8A dysfunction and associated disease mechanisms, we generated and characterized a knock-in mouse line expressing the human SCN8A R1620L mutation which was identified in a patient with autism and other behavioral abnormalities including ADHD, dyskinesia, aggression, and social behavior deficits. Methods: Using CRISPR/Cas9 technology, the human SCN8A R1620L mutation was knocked into the corresponding position in the mouse Scn8a gene. We previously showed that heterozygous mutants (RL/+) exhibit increased susceptibility to induced seizures and spontaneous seizures. We evaluated the ability of several anti-epileptic drugs (AEDs) to increase resistance to 6 Hz- and PTZ-induced seizures in RL/+ mutants and WT littermates. To characterize intrinsic action potential properties, we compared CA3 hippocampal slice recordings from RL/+ mutants and WT littermates. To examine neural ensemble activity, the visual cortex of RL/+ mutants and WT littermates were injected with AAV1-CaMKII-GCaMP6f and implanted with a gradient refractive index (GRIN) lens for in vivo miniscope-based calcium imaging. Baseline recordings in freely moving animals were obtained for 1 week followed by PTZ administration; miniscope imaging continued for 15 minutes after PTZ. Results: We found that Huperzine A significantly increased resistance to both 6 Hz- and PTZ-induced seizures in RL/+ mutants and WT littermates (N = 7-8/group, p < .0001); we are currently evaluating whether other AEDs can also provide seizure protection. When compared to WT littermates, CA3 pyramidal neurons from RL/+ mutants fired fewer action potentials, exhibited lower action potential firing threshold, and longer action potential duration. During baseline miniscope recordings, cortical excitatory neurons from RL/+ mutants displayed significantly higher calcium event amplitudes when compared to WT littermates (N = 200-300 cells/mouse, 2 mice/genotype, p < .0001). Calcium event response amplitudes in the RL/+ mutants were further increased and accompanied by large synchronous events after PTZ administration (p < .0001); this effect was not observed in WT littermates. Conclusions: RL/+ mutants exhibit significantly increased seizure susceptibility and neural excitability. CA3 hippocampal slice recordings suggest that the pyramidal neurons in RL/+ mutants have a lower threshold for action potential generation. These results were further extended by in vivo calcium imaging in which the RL/+ mutants exhibited significantly higher firing amplitudes compared to WT littermates. These findings provide further insight into the mechanisms of SCN8A-associated disease, and novel opportunities for the development of more efficacious treatments. Funding: This study was supported in part by a postdoctoral fellowship from the American Epilepsy Society (JCW) and the National Institutes of Health (AE and ALG, R01NS090319). Support was also provided by the Rodent Behavioral Core (RBC), which is subsidized by the Emory University School of Medicine and is one of the Emory Integrated Core Facilities. Additional support was provided by the Emory Neuroscience NINDS Core Facilities (P30NS055077).