K10 - Maturation of human induced pluripotent stem cell-derived cardiomyocytes in engineered tissue models increases doxorubicin-induced cardiotoxicity

Introduction: : Heart failure in cancer survivors is rising due to chemotherapy-induced cardiotoxicity, compromising patient quality of life. A better understanding of toxicity mechanisms is therefore needed to develop strategies for mitigating cardiotoxicity, as even commonly prescribed drugs like anthracyclines that are directly linked to heart failure incidence have poorly understood, complex mechanisms of toxicity. In particular, the cardiotoxicity of doxorubicin – the most commonly prescribed anthracycline – is hypothesized to act through multiple pathways in cardiomyocytes (CMs), including mitochondrial and oxidative stress, interference of DNA repair, and others, which confound the search for cardioprotective targets and strategies. The use of human induced pluripotent stem cell (hiPSC)-derived CMs for cardiotoxicity research is gaining momentum, yet use of these cells for predicting responses depends heavily on the structural and functional maturity of CMs. Simple 2D monolayer cultures enable high throughput assaying, while 3D engineered tissues improve muscle tissue like structure and function (often collectively called “maturity”). Further, use of fatty acids as metabolic substrates shifts mitochondrial respiration towards oxidative phosphorylation, increasing cellular oxidative stress, which is hypothesized to drive cardiotoxicity. In this study, we use hiPSC-CMs in 2D monolayers and 3D engineered microtissues as model systems for assessing doxorubicin-induced cardiotoxicity in vitro while modulating structural and metabolic maturation. To explore cardioprotection, we assessed cotreatment with an FDA-approved “cardioprotective” agent, dexrazoxane, which is an iron chelator thought to reduce oxidative stress to protect the heart but continues to be scrutinized for its marginal efficacy and ill-defined mechanism(s) of action.

Materials and

Methods: : We differentiated ventricular CMs from hiPSCs using a well-adopted small molecule biphasic Wnt pathway activation and inhibition, followed by a 4-day lactate-based metabolic purification to achieve >80% purity (cTnT+). 2D monolayer cultures of hiPSC-CMs or 3D engineered microtissues formed in non-adherent round-bottom agarose microwells (3D Petri Dish®) with 5% human ventricular cardiac fibroblasts were used. Standard culture conditions used RPMI 1640 basal culture medium supplemented with B27 and 1% pen/strep. A subset of hiPSC-CM cultures and 3D cardiac microtissues were exposed to a “maturation medium” containing low glucose, physiological calcium concentration, and fatty acid substrates for the TCA cycle to enhance mitochondrial use of (and thus development of the machinery for) oxidative phosphorylation. Exposure to a broad range of doxorubicin (DOX) concentrations for up to 72 hours provides prolonged exposure that reflects the exposure in cancer patient blood serum after IV chemotherapy infusion based on DOX half-life. Quantitative endpoint metrics included CM viability (2D MTT assay, 3D live/dead staining), 3D microtissue action potential kinetics (by optical mapping of voltage) and contractility (by strain measurements from brightfield videos), and oxygen consumption rate (measured with the Mito Stress Test in a Seahorse analyzer in 2D CMs). Dose response of the cardioprotective agent dexrazoxane was pursued with simultaneous DOX exposure for viability and metabolism.

Results, Conclusions, and Discussions:: Both concentration- and time-dependent DOX toxicity were observed in 2D hiPSC-CMs and 3D cardiac microtissues, yielding an average 80% viability at 0.1 µM DOX and 50% viability at 1 µM DOX by 2D MTT assay with 72 hrs exposure but 50% loss of contractility with 0.1 µM DOX in 3D cardiac microtissues (in standard culture conditions). Toxicity was associated with a reduction in basal oxygen consumption rate (OCR) by 33% at 0.1 µM DOX and the OCR-to-extracellular acidification rate (ECAR) ratio increased, showing reduced dependence on oxidative phosphorylation and increased glycolysis in standard culture conditions. Severe toxicity with 3 µM DOX slowed the action potential (AP) rise phase kinetics by 5-fold and reduced AP duration by 35% in 3D cardiac microtissues under standard culture conditions. DOX toxicity increased 2-fold in 3D cardiac microtissues cultured in maturation medium by live/dead staining, suggesting a critical role of metabolism in toxicity responses. We further observed that cotreatment with dexrazoxane in 3D microtissues had no statistically significant impact on preventing doxorubicin toxicity in either standard or maturation media culture conditions despite an up to 15% reduction in dead stain intensity. Surprisingly, viability by MTT assay in 2D hiPSC-CMs showed no significant effect of maturation medium vs standard medium for DOX toxicity or dexrazoxane cotreatment. In conclusion, this work supports the utility of mature hiPSC-CM-based 3d tissue models to serve as fit-for-purpose preclinical tools for understanding mechanisms of anthracycline toxicity, identifying new targets for cardioprotection, and designing novel compounds for cardioprotection against chemotherapy agents. Further, the importance of cellular and tissue maturity needs to be considered for the suitability of hiPSC-CM models to capture and accurately predict anthracycline-induced cardiotoxicity.

Acknowledgements and/or References (Optional):: We gratefully acknowledge funding from a Brown University Biomedical Innovation to Impact Award.