A Multiscale Simulation System for the Prediction of Drug-Induced Cardiotoxicity

• Published in: Journal of chemical information and modeling 2011-02, Vol.51 (2), p.483-492
• Main Authors: Obiol-Pardo, Cristian, Gomis-Tena, Julio, Sanz, Ferran, Saiz, Javier, Pastor, Manuel
• Format: Article
• Language: English
• Subjects: Cardiac arrhythmia; Cardiomyocytes; Chemistry; Computational Biology - methods; Drug-Related Side Effects and Adverse Reactions; Drugs; Electrocardiography; Electrophysiological Phenomena - drug effects; ERG1 Potassium Channel; Ether-A-Go-Go Potassium Channels - antagonists & inhibitors; Ether-A-Go-Go Potassium Channels - chemistry; Exact sciences and technology; General and physical chemistry; General. Nomenclature, chemical documentation, computer chemistry; Heart - drug effects; Heart - physiology; Humans; KCNQ1 Potassium Channel - antagonists & inhibitors; KCNQ1 Potassium Channel - chemistry; Models, Molecular; Pharmaceutical Modeling; Potassium; Potassium Channel Blockers - adverse effects; Potassium Channel Blockers - chemistry; Potassium Channel Blockers - pharmacology; Protein Conformation; Quantitative Structure-Activity Relationship; Reproducibility of Results; Simulation; Theory of reactions, general kinetics. Catalysis. Nomenclature, chemical documentation, computer chemistry; Tissues; Toxicity
• ISSN: 1549-9596; 1549-960X
• DOI: 10.1021/ci100423z

The preclinical assessment of drug-induced ventricular arrhythmia, a major concern for regulators, is typically based on experimental or computational models focused on the potassium channel hERG (human ether-a-go-go-related gene, Kv11.1). Even if the role of this ion channel in the ventricular repolarization is of critical importance, the complexity of the events involved make the cardiac safety assessment based only on hERG has a high risk of producing either false positive or negative results. We introduce a multiscale simulation system aiming to produce a better cardiotoxicity assessment. At the molecular scale, the proposed system uses a combination of docking simulations on two potassium channels, hERG and KCNQ1, plus three-dimensional quantitative structure−activity relationship modeling for predicting how the tested compound will block the potassium currents IKr and IKs. The obtained results have been introduced in electrophysiological models of the cardiomyocytes and the ventricular tissue, allowing the direct prediction of the drug effects on electrocardiogram simulations. The usefulness of the whole method is illustrated by predicting the cardiotoxic effect of several compounds, including some examples in which classic hERG-based models produce false positive or negative results, yielding correct predictions for all of them. These results can be considered a proof of concept, suggesting that multiscale prediction systems can be suitable for being used for preliminary screening in lead discovery, before the compound is physically available, or in early preclinical development when they can be fed with experimentally obtained data.