Computational Studies of Molecular Medicine for Voltage-Gated Sodium Channels - Project Summary A variety of mutations in voltage-gated sodium (Nav) channels are associated with a broad spectrum of channelopathies, including epilepsy, autism, intellectual disability, migraines, pain syndromes, myopathies, and cardiac arrhythmias. The genotype-phenotype relationship in Nav channels is profoundly complex, with identical mutations sometimes leading to variable or even opposing effects. The molecular mechanisms underlying these mutational disruptions remain largely unexplored, hindering structure-based drug design of targeted and mutation-specific therapeutics essential for precision medicine. Moreover, the functional impacts of thousands of Nav natural variants are still uncertain, challenging disease diagnosis and personalized treatment. We hypothesize that the variable or opposing effects of a specific Nav channel variant, such as both amplification (gain-of-function, or GoF) and diminution (loss-of-function, or LoF) of Na+ current, could arise from its impacts on multiple functional transitions or diverse functional aspects. Alternatively, these variable effects might be affected by the cross-talk between post-translational modifications (PTMs) and mutations. Given mutations’ roles are intrinsically rooted in their structural dynamics during functional transitions or their dynamic interactions with PTMs, molecular dynamics (MD) simulations are ideal to investigate the structural and dynamic effects of these mutations. In this study, we employ customized MD and conformational free-energy calculation strategies for several representative mutations to gain mechanistic insights within a practical simulation timescale. A comprehensive investigation of thousands of uncertain variants via MD simulations or electrophysiology is unfeasible due to resource and time constraints. While AI-based prediction offers a more efficient alternative, current AI tools only provide binary prediction (benign/pathogenic or GoF/LoF). More precise phenotype predictions are crucial for risk assessment, personalized treatment, and therapeutic innovation. We propose developing a machine-learning model that integrates protein sequence, structure, dynamics, and function data to accurately predict the gating properties of variants. This innovative approach could revolutionize the diagnosis and treatment strategies of channelopathies. This proposal outlines our ongoing efforts on three critical aspects of Nav channels using two computational approaches: (i) MD simulations to investigate 1) how mutations affect structural transitions and channel gating and 2) how mutation-PTM cross-talk leads to differential effects; and (ii) ML modeling to predict 3) the functional impacts of new variants. As an Early-Stage Investigator (ESI) nearing the end of my eligibility period (until June 2025), this MIRA ESI proposal represents a critical juncture in my career. The timing of this application aligns perfectly with my career goals and research trajectory. It provides a springboard for launching an innovative and long-term computational program bridging the molecular pathophysiology of channelopathies and rational drug design, promising lasting scientific impact in biophysics, pharmacology, and biomedical research of ion channels. 1
