Advancing Cardiovascular Drug Screening Using Human Pluripotent Stem Cell-Derived Cardiomyocytes

Abstract

Human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs) have emerged as a promising tool for studying cardiac physiology and drug responses. However, their use is largely limited by an immature phenotype and lack of high-throughput analytical methodology. In this study, we developed a high-throughput testing platform utilizing hPSC-CMs to assess the cardiotoxicity and effectiveness of drugs. Following an optimized differentiation and maturation protocol, hPSC-CMs exhibited mature CM morphology, phenotype, and functionality, making them suitable for drug testing applications. We monitored intracellular calcium dynamics using calcium imaging techniques to measure spontaneous calcium oscillations in hPSC-CMs in the presence or absence of test compounds. For the cardiotoxicity test, hPSC-CMs were treated with various compounds, and calcium flux was measured to evaluate their effects on calcium dynamics. We found that cardiotoxic drugs withdrawn due to adverse drug reactions, including encainide, mibefradil, and cetirizine, exhibited toxicity in hPSC-CMs but not in HEK293-hERG cells. Additionally, in the effectiveness test, hPSC-CMs were exposed to ATX-II, a sodium current inducer for mimicking long QT syndrome type 3, followed by exposure to test compounds. The observed changes in calcium dynamics following drug exposure demonstrated the utility of hPSC-CMs as a versatile model system for assessing both cardiotoxicity and drug efficacy. Overall, our findings highlight the potential of hPSC-CMs in advancing drug discovery and development, which offer a physiologically relevant platform for the preclinical screening of novel therapeutics.

Keywords: cardiomyocytes; cardiovascular pharmacology; drug screening; high-throughput assays; human pluripotent stem cells.

Figure 1

Optimization of culture conditions for the generation of hPSC-CMs. (a) Culture scheme for efficient differentiation and maturation of iPSCs. CHIR, CHIR99021. (b) Optimization of CHIR99021 concentration for inducing iPSC differentiation based on the expression of cTnT on D10. Upper, representative phase contrast images. Scale bar, 100 μm. Lower, representative FACS histograms. The blue area represents the isotype control, and the red area represents anti-cTnT antibody. (c) Relative mRNA levels of CM-related genes during the differentiation of iPSCs. mRNA levels of individual genes were normalized to undifferentiated cells (D0). Error bars, mean ± SEM (N = 3).

Figure 2

FFA-accelerated CM differentiation from hPSCs. (a,b) Expression levels of CM markers in iPSC-CMs on D20 and D30. The relative mRNA levels of TNNT2, MYL2, and MYL7 were quantified by qPCR and the MYL2/MYL7 ratio on D20 was calculated (a). Protein expression levels of p21, cTnI, and Connexin 43 were measured by Western blotting (b). (c) Cell populations expressing CM markers on D20. Cells positive for cTnI, MLC2V, and MLC2A were increased in the presence of FFA. The blue area represents the isotype control, and the red area represents the specific antibody. (d,e) Measurement of OCR as an indicator of cellular oxygen consumption and mitochondrial respiration. The OCR of FFA-treated iPSC-CMs was elevated on D20. Error bars, mean ± SEM (N = 3). Asterisks, *, significant differences (Student’s t-test; *, p < 0.05; **, p < 0.01; ****, p < 0.001). NS, no statistical significance.

Figure 3

Functional maturation of hPSC-CMs. (a) Relative mRNA levels of ion channels in iPSC-CMs. The mRNA levels of genes expressing Nav1.5, Cav1.2, hERG, Kv7.1, and Kir2.1 in iPSC-CMs were compared with those in undifferentiated iPSCs on D10 and D20. (b) Electrophysiological properties of iPSC-CMs in the presence of voltage-sensitive sodium channel blockers. Cells on D20 were treated with quinine (b₁), ritonavir (b₂), and propafenone (b₃). (c) Electrophysiological activity of hERG channels in iPSC-CMs on D20 compared with HEK293-hERG cells. Cardiotoxic drugs withdrawn due to adverse drug reactions, including encainide, mibefradil, and cetirizine at 5 μM, exhibited toxicity in iPSC-CMs but not in HEK293-hERG cells. Error bars, mean ± SEM (N = 3). Asterisks, *—statistically significant differences (Student’s t-test; p < 0.05). NS—no statistical significance.

Figure 4

Utilization of hPSC-CMs as a platform for high-throughput testing of the cardiotoxicity and effectiveness of drugs. Spontaneous calcium oscillations in iPSC-CMs were measured using the FLIPR system in the absence or presence of test compounds. (a,b) Cardiotoxicity test. (a) Experimental scheme. iPSC-CMs loaded with a calcium-sensitive fluorescent dye, EarlyTox^TM, were treated with a test compound, and calcium flux was observed. (b) Testing of the compounds known for their adverse effects in humans. The dye-loaded cells were exposed to ion channel blockers targeting hERG channels (E-4031 and dofetilide), sodium channels (ritonavir and flecaninide), and L-type calcium channels (diltiazem and verapamil) at designated concentrations, ranging from the highest concentration to serially diluted 1:3, while calcium oscillations were recorded. (c,d) Efficacy test. (c) iPSC-CMs (c₁) and ESC-CMs (c₂) were treated with 200 nM ATX-II, a late inward sodium current inducer, which produces atrial arrhythmias mimicking the LQT3 phenotype, followed by exposure to a test compound. (d) The elevated average peak widths of both iPSC-CMs (d₁,d₃) and ESC-CMs (d₂,d₄) by ATX-II treatment were restored to the level of untreated control cells after exposure to rotigotine (d₁,d₂) or ropivacaine (d₃,d₄). Error bars, mean ± SEM (N = 3). Asterisks, *—statistically significant differences (Student’s t-test; p < 0.05).

Figure 4

Utilization of hPSC-CMs as a platform for high-throughput testing of the cardiotoxicity and effectiveness of drugs. Spontaneous calcium oscillations in iPSC-CMs were measured using the FLIPR system in the absence or presence of test compounds. (a,b) Cardiotoxicity test. (a) Experimental scheme. iPSC-CMs loaded with a calcium-sensitive fluorescent dye, EarlyTox^TM, were treated with a test compound, and calcium flux was observed. (b) Testing of the compounds known for their adverse effects in humans. The dye-loaded cells were exposed to ion channel blockers targeting hERG channels (E-4031 and dofetilide), sodium channels (ritonavir and flecaninide), and L-type calcium channels (diltiazem and verapamil) at designated concentrations, ranging from the highest concentration to serially diluted 1:3, while calcium oscillations were recorded. (c,d) Efficacy test. (c) iPSC-CMs (c₁) and ESC-CMs (c₂) were treated with 200 nM ATX-II, a late inward sodium current inducer, which produces atrial arrhythmias mimicking the LQT3 phenotype, followed by exposure to a test compound. (d) The elevated average peak widths of both iPSC-CMs (d₁,d₃) and ESC-CMs (d₂,d₄) by ATX-II treatment were restored to the level of untreated control cells after exposure to rotigotine (d₁,d₂) or ropivacaine (d₃,d₄). Error bars, mean ± SEM (N = 3). Asterisks, *—statistically significant differences (Student’s t-test; p < 0.05).

Figure 4

Utilization of hPSC-CMs as a platform for high-throughput testing of the cardiotoxicity and effectiveness of drugs. Spontaneous calcium oscillations in iPSC-CMs were measured using the FLIPR system in the absence or presence of test compounds. (a,b) Cardiotoxicity test. (a) Experimental scheme. iPSC-CMs loaded with a calcium-sensitive fluorescent dye, EarlyTox^TM, were treated with a test compound, and calcium flux was observed. (b) Testing of the compounds known for their adverse effects in humans. The dye-loaded cells were exposed to ion channel blockers targeting hERG channels (E-4031 and dofetilide), sodium channels (ritonavir and flecaninide), and L-type calcium channels (diltiazem and verapamil) at designated concentrations, ranging from the highest concentration to serially diluted 1:3, while calcium oscillations were recorded. (c,d) Efficacy test. (c) iPSC-CMs (c₁) and ESC-CMs (c₂) were treated with 200 nM ATX-II, a late inward sodium current inducer, which produces atrial arrhythmias mimicking the LQT3 phenotype, followed by exposure to a test compound. (d) The elevated average peak widths of both iPSC-CMs (d₁,d₃) and ESC-CMs (d₂,d₄) by ATX-II treatment were restored to the level of untreated control cells after exposure to rotigotine (d₁,d₂) or ropivacaine (d₃,d₄). Error bars, mean ± SEM (N = 3). Asterisks, *—statistically significant differences (Student’s t-test; p < 0.05).