Informing Hazard Identification and Risk Characterization of Environmental Chemicals by Combining Transcriptomic and Functional Data from Human Induced Pluripotent Stem Cell-Derived Cardiomyocytes

Cardiovascular disease remains a significant global health concern, yet the cardiotoxic effects of environmental chemicals are largely unexplored. Human induced pluripotent stem cell-derived cardiomyocytes (iPSC-Cardiomyocytes) are well-established as valuable high-throughput cardiotoxicity testing model based on the analysis of various beating parameters; however, additional adverse effects that may not be directly related to ion channel activity are difficult to ascertain from traditional assays. Transcriptomic data may provide a more comprehensive assessment for additional molecular responses and serve as a basis for dose-response analysis. We hypothesized that transcriptomic data and functional phenotypic measurements in iPSC-Cardiomyocytes could be used to effectively screen for both hazard and risk associated with chemical exposures. To test this hypothesis, we performed concentration-response analysis of 464 chemicals from 12 diverse chemical classes. Functional effects (beat frequency, QT-prolongation, and asystole) and cytotoxicity were measured. In addition, whole transcriptome RNA-seq was used to investigate global gene expression profiles. Points of departure (PODs) were derived from both phenotypic endpoints and transcriptomic data. Margin of safety (MOS) for drugs and margin of exposure (MOE) for non-drug chemicals were calculated using PODs and the exposure estimates. The phenotypic assays revealed that a varying proportion (10% to 44%) of chemicals in each class had effects on iPSC-Cardiomyocytes; positive chronotrope was the most sensitive phenotype. Overall, 244 (53%) chemicals were active in at least one functional phenotype. As expected, drugs with known adverse effects on the heart were most active. Transcriptomic data showed 69 (15%) had significant effects on the biological pathways, the pathways that were significantly affected by some tested compounds were highly relevant to cardiomyocyte function. Specifically, the enriched pathways aligned well with key characteristics of known cardiovascular toxicants, with mitochondrial dysfunction being the most affected key characteristic. Using exposure and hazard (transcriptomic and phenotypic PODs from this study), we derived MOS and MOE and found that drugs tend to have lower MOS than MOE for the environmental chemicals. Thirteen and 10 drugs had MOS < 1 using phenotypic and transcriptomic data, respectively, suggesting that further investigation may be needed for these substances to test for potential cardiovascular risk. Overall, our findings demonstrate how the integrative use of in vitro transcriptomic data and phenotypic assays in iPSC-Cardiomyocytes not only offers a unique and complementary approach for hazard and risk prioritization, but also offers comprehensive mechanistic support for in vitro test results.