Atomistic Computational Models to Predict Drug-Mediated Cardiotoxicity

Abstract

To be biologically active, small molecule drugs must physically fit into their binding site(s) within their targets. However, in reaching their precise binding locations, drugs may interact with a variety of cellular components of various structures and functions. Recent seminal advances in computer software and hardware render in silico methods increasingly important. They can build detailed atomistic dynamical structural models to predict ligands' off-target interactions. In our lab, we focus on the potential blockade of cardiac ion channels by small molecule drugs, a critical event that can lead to acquired cardiac long QT syndrome (LQTS) and fatal cardiac arrhythmias. We build computational models for the human Ether-à-go-go-Related Gene (hERG) channel, the Nav1.5 sodium channel, the Cav1.2 calcium channel and the KCNQ1/3 and Kir2.1 potassium channels. The ultimate goal is to investigate their interactions with drugs in silico and predict the mode of binding of these drugs and their potential blockage capacity. This work will provide an overview on the different ion channels models established in our lab and their success in predicting cardiotoxicity for various drugs.