Abstracts

Rationale: Genetic testing has revealed >600 variants in SCN1A, the gene encoding the voltage-gated sodium channel Nav1.1, that are causative of epilepsy. The majority of these variants arise de novo and have not previously been reported in the pediatric population. Currently, missense variants encoding amino acid substitutions in SCN1A are classified as causative using a range of bioinformatic tools including allelic frequency, conservation, structural and physiochemical properties. However, there is little correlation observed between disease severity and genotype, drastically limiting the application of SCN1A genetic testing in clinical decision making. We hypothesize that clinical severity is related to the altered structural stability of the channel during its transition from the deactivated (closed) to active (open) state, impacting channel biophysics and thus physiological function.Methods: In silico homology models of the human Nav1.1 protein in both the open (inactivated) and closed conformations were generated based on the resolved crystal structures of two different six transmembrane bacterial voltage gated sodium channels. Using Biovia (Accelrys) Discovery Studio, Nav1.1 homology models were refined using MODELLER for structural model generation, LOOPER for protein loop optimized prediction and CHARMm for model refinement and molecular dynamics simulation. In silico mutagenesis was performed and the relative change in energy caused by each amino acid substitution was calculated for SCN1A missense mutations identified across the clinical disease spectrum. Mutations were ranked to establish the extent of inherent energetic instability in each state.Results: Both stabilization and destabilization of the channel protein is observed, but this is highly dependent on the particular mutation evaluated and the specific channel conformation. The majority of mutations impact stability with the bulk destabilizing the channel conformation regardless of position within the protein. Mutation and phenotype dependent effects are observed when the stability of both the open and closed states is calculated.Conclusions: The demand for genetic diagnoses is rising rapidly as personalized medicine gains popularity. Current bioinformatics strategies have little predictive power for novel mutations in SCN1A. There is an urgent priority to develop biologically relevant in silico tools for clinical decision making to enhance accuracy of risk prediction. Mutations in ion channel genes may result in enhanced or reduced conformational stability through the dynamic structural changes that occur during the transition between the closed and open states of the channel. By better understanding how individual missense mutations impact channel stability using computational structural mutagenesis, we will further refine molecular diagnostic predictions for the purposes of diagnosis, prognosis and therapeutic decision making.