Using rosetta and alphafold to assess Jingzhaotoxin-III binding to hNav1.5 and hKv2.1

Ion channels are pore-forming transmembrane proteins regulating ion flux across cell membranes and regulators of the ventricular action potential. The human voltage-gated sodium channel 1.5 (hNav1.5) is predominantly expressed in cardiac myocytes and is solely responsible for the ventricular action potential's initial upstroke phase. A series of mutations in hNav1.5 are classified as Long QT Syndrome subtype 3 (LQT3) and are linked to hNav1.5 gain-of-function, resulting in an enhanced inward current prolonging repolarization. Alternatively, improper hNav1.5 inactivation results in a prolonged, late sodium current during the action potential plateau, causing a similar LQT3 gain-of function phenotype. Current drug therapies for hNav1.5-associated arrhythmia involve combinations of beta-adrenergic blockers and hNav1.5-targeting small-molecules such as flecainide. Alternatively, hNav1.5-targeting peptide toxins have the potential as novel therapeutics and investigational tools for ion channel disorders. One such peptide, Jingzhaotoxin-III (JzTx-III) is a 36-residue peptide that has been found to inhibit hNav1.5 and human voltage-gated potassium channel isoform 2.1 (hKv2.1) but does not inhibit neuronal and skeletal hNav channels. Mutational studies of JzTx-III identified unique residues that modulate hNav1.5 and hKv2.1 activity. We hypothesize that prior single-point mutational data will demonstrate differing sets of binding free energies and peptide-channel contacts for hNav1.5 and hKv2.1. In this study, we used AlphaFold and Rosetta to generate models of hNav1.5 and hKv2.1 in the deactivated state, docked JzTx-III to each channel, and assessed in silico binding free energies of wild-type and single-point mutations. This structural and energetic information can then be used as a guideline for multi-point mutational design to predict improved hNav1.5 binding specificity for further investigation of LQTS and pro-arrhythmic ventricular action potentials.