High-Throughput Analysis of Kinase Inhibitor Drugs on Cardiac Function Using Engineered Heart Tissue Constructs

A serious concern in health care today is the unintentional long-term effects of pharmaceuticals on the heart, which are not detected during clinical trials. Tyrosine kinase inhibitors (TKIs) are commonly used as highly specific cancer targeting agents for their ability to prevent tyrosine kinase molecules from activating signaling pathways that regulate cell growth, differentiation, metabolism, migration, and programmed cell death. Due to the highly metabolic nature of cardiac tissue, cardiomyocytes are sensitive to perturbations in mitochondrial function. Clinically used TKIs have been found to have a spectrum of toxicities on the heart that were not detected during clinical trials. High-throughput preclinical screening of TKIs could offer much information as to their cardiovascular effects and suggest ideal candidates for further investigation without putting patients at risk. We hypothesized that a high-throughput, high-fidelity screening platform for evaluating the physiological effects of TKIs on human myocardium could be established using high-content screening assays, artificial neural network (ANN) modelling, and engineered heart tissue constructs termed Biowires.

High-content screening assays from Molecular Devices were used sequentially to assess cell viability and basic cardiac function of 80 TKIs on a human cardiomyocyte monolayer at 3 different concentrations. Experimental results were then analyzed using ANN modelling. A network was designed that could predict the compound with the least detrimental effects on cardiac function. This compound was then examined in the Biowire II platform, a micro-scale cardiac tissue array mimicking physiological tissue that consist of human cardiomyocytes seeded onto polymer wires and electrically stimulated in order to promote tissue maturation. Fluorescent image microscopy was used to examine tissue contraction and calcium transient profiles.

Initial monolayer testing using human induced pluripotent stem cell-derived cardiomyocytes (iPS-CM) and 2 high-content screening assays yielded four parameters for each inhibitor at each concentration: cell viability, live cell number, calcium peak magnitude, and contraction frequency. ANN modelling was used to analyze the experimental results efficiently and in an unbiased manner, to reduce the number of TKIs tested on the Biowire platform. We chose ANN modelling because of its predictive capabilities and its ability to identify complex non-linear relationships between molecular function and cell response. The effect of the least detrimental TKI was then examined on matured cardiac tissue to observe changes in force of contraction and calcium transients after exposure, as well as immunohistochemical effects. In the future, these inhibitors of interest could be studied to understand their cellular mode of action.

In conclusion, this system linked high throughput experimental screening with an empirical modelling tool to observe and measure the physiological effects of TKIs on cardiac tissue. We believe this system can be applied to screening of other compounds on cardiac tissue and should be used to minimize the cost of experimentation and maximize the impact of pre-clinical research.