Stem cells are the most sensitive screening tool to identify toxicity of GATA4-targeted novel small-molecule compounds

Abstract

Safety assessment of drug candidates in numerous in vitro and experimental animal models is expensive, time consuming and animal intensive. More thorough toxicity profiling already in the early drug discovery projects using human cell models, which more closely resemble the physiological cell types, would help to decrease drug development costs. In this study we aimed to compare different cardiac and stem cell models for in vitro toxicity testing and to elucidate structure–toxicity relationships of novel compounds targeting the cardiac transcription factor GATA4. By screening the effects of eight compounds at concentrations ranging from 10 nM up to 30 µM on the viability of eight different cell types, we identified significant cell type- and structure-dependent toxicity profiles. We further characterized two compounds in more detail using high-content analysis. The results highlight the importance of cell type selection for toxicity screening and indicate that stem cells represent the most sensitive screening model, which can detect toxicity that may otherwise remain unnoticed. Furthermore, our structure–toxicity analysis reveals a characteristic dihedral angle in the GATA4-targeted compounds that causes stem cell toxicity and thus helps to direct further drug development efforts towards non-toxic derivatives.Electronic supplementary materialThe online version of this article (10.1007/s00204-018-2257-1) contains supplementary material, which is available to authorized users.

GATA4-targeted compound exhibits cardioprotective actions against doxorubicin-induced toxicity in vitro and in vivo: establishment of a chronic cardiotoxicity model using human iPSC-derived cardiomyocytes

“…The long-term exposures (up to 21 days), however, revealed toxic effects of 3i-1000 in cardiomyocytes. In our previous study (Karhu et al 2018), the toxicity of eight compounds (3i-1000 and its derivatives) at concentrations ranging from 10 nM to 30 µM on the viability of eight different cell types were studied in detail. In these short-term experiments (24 h), 3i-1000 was non-toxic to cardiomyocytes (NRVMs, hiPSC-CMs), fibroblasts and H9c2 cardiac myoblasts.…”

“…The selection of doxorubicin concentrations was based on plasma concentrations detected in patients undergoing treatment (Creasey et al 1976;Greene et al 1983;Muller et al 1993;Speth et al 1987). The selection of 3i-1000 concentrations was based on previous studies investigating the efficacy and toxicity of the compound in vitro (Karhu et al 2018;Välimäki et al 2017). For compound exposures doxorubicin, 3i-1000, and equivalent vehicle dilutions were made separately in the growth medium.…”

“…The cells were grown until 80-95% confluent. Cardiomyocytes were produced from hiPSCs using small-molecule induction, as described earlier (Burridge et al 2014;Karhu et al 2018). Differentiation was started by adding 6 µM CHIR99021 in RPMI 1640 medium supplemented with B-27 without insulin (RB-ins) to the cells (day 0).…”

Targeting GATA4 for cardiac repair

“…Furthermore, compound 3i‐1000 improved cardiac function in an experimental model of angiotensin II‐mediated hypertension in rats. Optimization of primary modulators of GATA4–NKX2‐5 interactions are ongoing to improve metabolite and safety profiles and determine the direct protein targets of transcriptional synergy inhibition associated with molecular mechanism(s) of action …”

“…Optimization of primary modulators of GATA4-NKX2-5 interactions are ongoing to improve metabolite and safety profiles and determine the direct protein targets of transcriptional synergy inhibition associated with molecular mechanism(s) of action. 103 It is also noteworthy to point out other challenges that would come from attempting to disrupt GATA4 activity. Beyond the fact that GATA4 is an essential TF in cardiac cells, it is also expressed and functional in many other tissues, prominent among these the pancreas, liver, and lung.…”