Atomistic Modeling Toward Predictive Cardiotoxicity

Current methods for assessing cardiac arrhythmia risk are insufficient for distinguishing between cardiotoxic and benign drugs. A blockade of the cardiac potassium channel encoded by the human Ether-à-go-go-Related Gene (hERG) channel and resulting prolongation of the QT interval on the surface electrocardiogram (ECG) are not selective indicators for drug-induced arrhythmogenesis. The chapter will illustrate the use of molecular dynamics (MD) simulations in a prototype study focusing on the interactions of the hERG channels with dofetilide, a drug having a high pro-arrhythmia risk. The simulation forms an essential component of a multi-scale computational framework to distinguish between safe and unsafe hERG blockers. The atomistic models examine an open conducting hERG channel interacting with a neutral and charged (cationic) dofetilide. Multi-microsecond drug “flooding” MD simulations reveal spontaneous drug binding to the channel pore through the intracellular gate. Umbrella sampling MD simulations have computed dofetilide affinity to hERG, in good agreement with experiment, as well as ingress and egress rate constants. These quantities derived from MD simulations provide a novel linkage between the molecular structural and functional scales and represent the first necessary components of a computational framework in the drug screening for virtual cardiac safety pharmacology assessment.