Exploring mechanisms of inhibition and inactivation in voltage-gated sodium channels using molecular simulations

Abstract

Voltage-gated sodium (Nav) channels play a pivotal role in the conduction of electrical signals across the human body, with different subtypes expressed across excitable tissue in the brain, heart and muscle. Nav channels transiently form transmembrane pores by cycling through the closed, open and fast-inactivated states to control the sodium ion flux across the membrane and initiate the firing of nerve and muscle cells. Dysfunction in Nav channels is implicated in numerous diseases, including epilepsy, cardiac arrhythmias, paralysis and genetic pain conditions, which are commonly treated using small molecule drugs that inhibit their activity by blocking the pore. This thesis explores the use of molecular simulations to understand how the Nav channel interacts with modulatory molecules of interest, such as pore-inhibiting drugs and endogenous lipids, as well as how its dynamics are altered by specific mutations. Since majority of Nav channel inhibitors bind in the highly conserved pore, they typically cannot selectively target just one subtype, thus leading to adverse side effects. Chapter two investigates the dynamic nature of the membrane-facing pore fenestrations, showing that three out of the four fenestrations are wide enough for drug passage. However, there is a lack differences between the subtypes, suggesting the infeasibility for subtype-selective drug access routes via the fenestrations. Building on this, Chapter three examines where a structurally diverse range of compounds bind in the Nav channel. Enhanced simulations suggest that each drug can adopt a variety of favourable binding sites in the pore, with certain drugs occupying the central cavity, whereas some smaller drugs preferred to bind within the fenestrations. Aside from exogenous drugs, phosphoinositides, a class of negatively charged phospholipids in the inner leaflet of the membrane, have been experimentally shown to modulate Nav channel activity, however the molecular basis of such modulation is unclear. In Chapter four, multi-scale simulations reveal the phosphoinositide binding site and provide mechanistic insight into how this could enhance the channel's ability to inactivate. Conversely, certain disease mutations are known to attenuate or abolish inactivation, leading to excessive neuronal activity. In Chapter five, a de novo mutation causing severe epilepsy, is characterised experimentally and computationally. Electrophysiology data show that the mutant Nav channels exhibit persistent, non-inactivating sodium current. Additionally, in both unbiased and enhanced simulations, mutations to this residue interaction lead to destabilisation of the region responsible for fast inactivation. Finally, Chapter six examines the dynamic range of structures across the voltage-gated ion channel superfamily and the potential of using AlphaFold2 to predict different conformational states. Some voltage-sensing domains in the superfamily are consistently generated in one state, whereas others produced a variety of conformations. Notably, one of the Nav channel's four heterologous voltage sensor is predicted in a range of intermediate states with further conformational heterogeneity evident in the pore and fast inactivation gate. Overall, this work provides insights into the molecular mechanisms underlying Nav channel modulation and dysfunction. By integrating various computational approaches, the findings pave the way for more effective drug design and a deeper understanding of Nav channel-related diseases.