Modelling and prediction of chemical-induced human cholestatic liver injury: a mechanistic approach

Cholestatic liver insults constitute a major manifestation of drug-induced liver injury. Current in vivo and in vitro approaches poorly detect drug-induced cholestatic liver injury, which is partly due to gaps in the mechanistic understanding of this type of hepatotoxicity. This doctoral project tackles this hurdle by providing a state-of-the-art scenario of the established as well as novel mechanisms that drive druginduced cholestatic liver injury. In a first study, a liver-based in vitro system was optimized to mechanistically study cholestasis, namely human hepatoma HepaRG cell cultures exposed to different cholestatic drugs and bile acids. This in vitro system has shed new light on the mechanisms underlying drug-induced cholestasis and has unveiled differences between different types of cholestasis as well as between the in vitro and in vivo situation. In a second study, the newly developed in vitro system in combination with other tools to predict cholestatic potential was evaluated for its application to chemicals outside the pharmaceutical area, in particular industrial chemicals, cosmetics ingredients, herbicides and food additives. It was found that further fit-for-purpose optimization is required for general use of the in vitro setting. A third study investigated liver samples of cholelithiasis-induced cholestasis patients and cholestatic mice by means of transcriptomic analysis in order to elucidate the mechanistic framework of different types of cholestasis and simultaneously characterize the human relevance of mouse models of cholestasis. Overall, this doctoral project has provided an important contribution to the elucidation of mechanisms underlying chemical-induced cholestasis, based on both in vitro as well as in vivo studies, and has yielded an in vitro setting fit for detecting drug-induced cholestasis.