Deep learning detects cardiotoxicity in a high-content screen with induced pluripotent stem cell-derived cardiomyocytes

Abstract

Drug-induced cardiotoxicity and hepatotoxicity are major causes of drug attrition. To decrease late-stage drug attrition, pharmaceutical and biotechnology industries need to establish biologically relevant models that use phenotypic screening to detect drug-induced toxicity in vitro. In this study, we sought to rapidly detect patterns of cardiotoxicity using high-content image analysis with deep learning and induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs). We screened a library of 1280 bioactive compounds and identified those with potential cardiotoxic liabilities in iPSC-CMs using a single-parameter score based on deep learning. Compounds demonstrating cardiotoxicity in iPSC-CMs included DNA intercalators, ion channel blockers, epidermal growth factor receptor, cyclin-dependent kinase, and multi-kinase inhibitors. We also screened a diverse library of molecules with unknown targets and identified chemical frameworks that show cardiotoxic signal in iPSC-CMs. By using this screening approach during target discovery and lead optimization, we can de-risk early-stage drug discovery. We show that the broad applicability of combining deep learning with iPSC technology is an effective way to interrogate cellular phenotypes and identify drugs that may protect against diseased phenotypes and deleterious mutations.

Phenotypic screening with deep learning identifies HDAC6 inhibitors as cardioprotective in a BAG3 mouse model of dilated cardiomyopathy

“…To efficiently and reproducibly quantify sarcomere damage in BAG3 KD iPSC-CMs, we adopted an imaging analysis method that uses deep learning ( 25 ). As we previously described ( 21 ), we used a two-class deep learning model based on healthy (SCR siRNA–treated) and diseased (BAG3 siRNA–treated) iPSC-CMs. We labeled about 1300 images from each class of iPSC-CMs and fed them into the neural network (Fig.…”

“…Using CRISPR-Cas9, we generated HDAC6 knockout (HDAC6 KO ) iPSCs. We successfully differentiated these cells to cardiomyocytes, as described previously ( 21 ), which showed expression of sarcomeric markers [cardiac troponin T (TNNT2) and MYBPC3] and hyperacetylation of tubulin (fig. S5, C and D).…”

“…In this study, we aimed to identify small-molecule therapeutic targets for BAG3-related DCM that could enable drug development for genetically defined cardiomyopathies that disrupt protein quality control. Previously, we showed the feasibility of high-content phenotypic screening using deep learning with iPSC-CMs ( 21 ). We hypothesized that we could use a similar approach to identify cardioprotective small-molecule targets in BAG3-deficient iPSC-CMs that would translate to therapeutic benefits in a mouse model of DCM.…”

Artificial intelligence opportunities in cardio-oncology: Overview with spotlight on electrocardiography

“…In a recent study, biologically relevant models capable of detecting drug-induced toxicity via phenotypic screening were developed [ 128 ]. Deep learning, high-content image analysis, and cardiomyocytes derived from induced pluripotent stem cells (iPSC-CMs) were used to rapidly screen for instigators of cardiotoxicity.…”

Application of the Pluripotent Stem Cells and Genomics in Cardiovascular Research—What We Have Learnt and Not Learnt until Now

“…Novel high-throughput, high-content images, and automated platforms that utilize human iPSC-derived 3D-engineered cardiac tissue constructs to better recapitulate heart functions and drug responses are being developed and are becoming sophisticated, with comprehensive profiling of the cellular responses to drugs across multidimensional parameter spaces. Artificial intelligence's machine learning and deep learning approaches have been shown to handle multidimensional datasets in an automated fashion to accurately predict the contractile behavior of hPSC-CMs exposed to cardioactive drugs and have proven to be very powerful tools for more reliable predictions of cardioactive drug-mediated cellular responses [113,115].…”