Reliable identification of cardiac liability in drug discovery using automated patch clamp: Benchmarking best practices and calibration standards for improved proarrhythmic assessment.

INTRODUCTION: Screening compounds for activity on the hERG channel using patch clamp is a crucial part of safety testing. Automated patch clamp (APC) is becoming widely accepted as an alternative to manual patch clamp in order to increase throughput whilst maintaining data quality. In order to standardize APC experiments, we have investigated the effects on IC50 values under different conditions using several devices across multiple sites. METHODS: APC instruments SyncroPatch 384i, SyncroPatch 384PE and Patchliner, were used to record hERG expressed in HEK or CHO cells. Up to 27 CiPA compounds were used to investigate effects of voltage protocol, incubation time, labware and time between compound preparation and experiment on IC50 values. RESULTS: All IC50 values of 21 compounds recorded on the SyncroPatch 384PE correlated well with IC50 values from the literature (Kramer et al., 2013) regardless of voltage protocol or labware, when compounds were used immediately after preparation, but potency of astemizole decreased if prepared in Teflon or polypropylene (PP) compound plates 2-3 h prior to experiments. Slow acting compounds such as dofetilide, astemizole, and terfenadine required extended incubation times of at least 6 min to reach steady state and therefore, stable IC50 values. DISCUSSION: Assessing the influence of different experimental conditions on hERG assay reliability, we conclude that either the step-ramp protocol recommended by CiPA or a standard 2-s step-pulse protocol can be used to record hERG; a minimum incubation time of 5 min should be used and although glass, Teflon, PP or polystyrene (PS) compound plates can be used for experiments, caution should be taken if using Teflon, PS or PP vessels as some adsorption can occur if experiments are not performed immediately after preparation. Our recommendations are not limited to the APC devices described in this report, but could also be extended to other APC devices.

Electrophysiological evaluation of pentamidine and 17-AAG in human stem cell-derived cardiomyocytes for safety assessment.

Human ether-a-go-go-related gene (hERG) trafficking inhibition is known to be one of the mechanisms of indirect hERG inhibition, resulting in QT prolongation and lethal arrhythmia. Pentamidine, an antiprotozoal drug, causes QT prolongation/Torsades de Pointes (TdP) via hERG trafficking inhibition, but 17-AAG, a geldanamycin derivative heat shock protein 90 (Hsp90) inhibitor, has not shown torsadogenic potential clinically, despite Hsp90 inhibitors generally being hypothesized to cause TdP by hERG trafficking inhibition. In the present study, we investigated the underlying mechanisms of both drugs' actions on hERG channels using hERG-overexpressing CHO cells (hERG-CHOs) and human embryonic stem cell-derived cardiomyocytes (hES-CMs). The effects on hERG tail current and protein levels were evaluated using population patch clamp and Western blotting in hERG-CHOs. The effects on field potential duration (FPD) were recorded by a multi-electrode array (MEA) in hES-CMs. Neither drug affected hERG tail current acutely. Chronic treatment with each drug inhibited hERG tail current and decreased the mature form of hERG protein in hERG-CHOs, whereas the immature form of hERG protein was increased by pentamidine but decreased by 17-AAG. In MEA assays using hES-CMs, pentamidine time-dependently prolonged FPD, but 17-AAG shortened it. The FPD prolongation in hES-CMs upon chronic pentamidine exposure is relevant to its clinically reported arrhythmic risk. Cav1.2 or Nav1.5 current were not reduced by chronic application of either drug at a relevant concentration to hERG trafficking inhibition in human embryonic kidney (HEK293) cells. Therefore, the reason why chronic 17-AAG shortened the FPD despite the hERG trafficking inhibition occur is still unknown.