Evaluation of chronic drug-induced electrophysiological and cytotoxic effects using human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs)

Abstract

Introduction: Cardiotoxicity is one of the leading causes of compound attrition during drug development. Most in vitro screening platforms aim at detecting acute cardio-electrophysiological changes and drug-induced chronic functional alterations are often not studied in the early stage of drug development. Therefore, we developed an assay using human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) that evaluates both drug-induced acute and delayed electrophysiological and cytotoxic effects of reference compounds with clinically known cardiac outcomes. Methods: hiPSC-CMs were seeded in 48-well multielectrode array (MEA) plates and were treated with four doses of reference compounds (covering and exceeding clinical free plasma peak concentrations -fC_max values) and MEA recordings were conducted for 4 days. Functional-electrophysiological (field-potentials) and viability (impedance) parameters were recorded with a MEA machine. Results: To assess this platform, we tested tyrosine-kinase inhibitors with high-cardiac risk profile (sunitinib, vandetanib and nilotinib) and low-cardiac risk (erlotinib), as well as known classic cardiac toxic drugs (doxorubicin and BMS-986094), ion-channel trafficking inhibitors (pentamidine, probucol and arsenic trioxide) and compounds without known clinical cardiotoxicity (amoxicillin, cetirizine, captopril and aspirin). By evaluating the effects of these compounds on MEA parameters, the assay was mostly able to recapitulate different drug-induced cardiotoxicities, represented by a prolongation of the field potential, changes in beating rate and presence of arrhythmic events in acute (<2 h) or delayed phase ≥24 h, and/or reduction of impedance during the delayed phase (≥24 h). Furthermore, a few reference compounds were tested in hiPSC-CMs using fluorescence- and luminescence-based plate reader assays, confirming the presence or absence of cytotoxic effects, linked to changes of the impedance parameters measured in the MEA assay. Of note, some cardiotoxic effects could not be identified at acute time points (<2 h) but were clearly detected after 24 h, reinforcing the importance of chronic drug evaluation. Discussion: In conclusion, the evaluation of chronic drug-induced cardiotoxicity using a hiPSC-CMs in vitro assay can contribute to the early de-risking of compounds and help optimize the drug development process.

Keywords: arrhythmias; cytotoxicity; drug-induced cardiotoxicity; hiPSC-CMs; in vitro assay; multielectrode arrays (MEA); risk prediction.

FIGURE 1

Determination and characterization of the main MEA parameters. (A) A typical field potential (FP) recording highlighting the depolarization spike, the T-wave and beating rate. (B) Determination of the FP duration (FPD), from the spike to the peak of the T-wave. (C) Descriptive parameters for FP changes, including quiescence (i.e., cessation of beating), beating below quantification level (when the monolayer is beating but the field potential amplitude is lower than 0.3 mV), Flat T-wave (absence of a clear repolarization wave), and arrhythmic events (EADs-like events). (D) Representative MEA recordings showing the time-dependent changes in the morphology of the FP in DMSO 0.1% (vehicle control), cetirizine 1 µM (negative control) and Pentamidine 3 µM (positive control). The vertical bar is located at the top of the T-wave to mark the FPD; note that at 96 h, pentamidine does not show a repolarization wave (Flat T-wave).

FIGURE 2

Distribution and tolerance intervals for the main MEA parameters. Density plots depicting the distribution of DMSO 0.1% experiments (expressed as % change from baseline) and the relative tolerance intervals (blue dotted lines) at 24 h.

FIGURE 3

Effect of two negative controls on FP parameters. (A) Representative traces of FP from hiPSC-CMs treated with aspirin 100 µM and captopril 30 µM over time (from 24 to 96 h). Notably, no effects on FP morphology were observed. The vertical lines mark the top of the T-wave. Electrophysiological changes in FPD, FPDc and impedance caused by (B) aspirin and (C) captopril. In the figure legend, the doses indicated with a black outline are the closest to fC_max.

FIGURE 4

Effect of doxorubicin and BMS-986094 on FP. (A) Representative FP trace from baseline and after 96 h-treatment with doxorubicin 0.3 µM or BMS-986094 10 μM, in which the disappearance of the T-wave is evident. Effect of (B) doxorubicin and (C) BMS-986094 on FPDc and impedance. In the figure legend, the dose of doxorubicin indicated with a black outline is the closest to fC_max, while that for BMS-986094 is the estimated dose that produced cardiotoxicity in monkeys (Gill et al., 2017). *: p < 0.05 and outside the tolerance intervals of vehicle controls.

FIGURE 5

Ion-channels trafficking inhibition effect on FP. (A) Representative FP traces from baseline and after 72 h-treatment with pentamidine 3 µM. The light-blue shading indicates the FPD, significantly prolonged after treatment. In the center and right, representative FP traces at different time points of FP after treatment with As₂O₃ 1 µM and probucol 10 µM. The vertical lines indicate the top of the T-wave. Effect of (B) pentamidine, (C) As₂O₃ and (D) probucol on of FPD, FPDc and impedance. Q: quiescent (i.e., cessation of beating). In the figure legend, the doses indicated with a black outline are the closest to fC_max. *: p < 0.05 and outside the tolerance intervals of vehicle controls.

FIGURE 6

Effect of TKi on FP and impedance. (A) Representative traces depicting the time course of sunitinib 0.3 µM on FP. (B) Effect of sunitinib 0.3 µM on the morphology of the T-wave (the red arrow indicates an EADs-like event). (C) Representative traces depicting the time course of erlotinib 10 µM on FP. Effect of (D) sunitinib, (E) vandetanib and (F) erlotinib on FPD, FPDc. Q: quiescent (i.e., cessation of beating). The doses indicated in the figure legend with a black outline are the closest to fC_max. *: p < 0.05 and outside the tolerance intervals of vehicle controls.

FIGURE 7

Cytotoxicity and cell-death mechanisms assays. Assays measuring the impact of the drug treatments on (A) cell viability, (B) apoptosis as reflected by caspases 3,7 or (C) Annexin V activation, and (D) necrosis. Data are normalized by vehicle control (DMSO 0.1%) and expressed as mean ± SD. “*” indicates p < 0.05.