Improved in vitro systems for prediction of hepatotoxicity
Liver toxicity is a major societal burden and hepatotoxicity is difficult to foresee due to species differences and the difficulties to keep hepatocyte cultures functionally appropriate because of rapid de-differentiation of the cells. In order to mimic true in vivo human hepatotoxicity it is also evident that diseased models would be of importance for the in vitro models to better comply with different situations in vivo in man.
Within the NOTOX project we have further developed and applied 3D models based on HepaRG and human primary hepatocytes to improve predictability of primary human hepatocyte (PHH) based in vitro systems for measuring long-term toxicity. Within the project we monitor the changes induced by different compounds at the transcriptomic, epigenomic, proteomic and metabonomic levels. We have characterized the spheroids morphologically and functionally and initially found that 3D HepaRG spheroids can be kept active for at least 30 days even without addition of serum. In many cases they can better predict hepatotoxicity than the corresponding 2D models. We have also found that the PHH spheroids are of good functionality with respect to urea and albumin production and expression of CYP enzymes for at least 35 days and are suitable for determining chronic drug toxicity during at least 28 days. In addition we are developing modified diseased spheroid models mimicking inflammation, viral infection, cholestasis and steatosis. In particular the HepaRG and PHH derived cholestatic models have been found to be effective and to be able to discriminate between cholestatic and non-cholestatic toxicity of drugs in the presence of bile acids.
In the current presentation recent data from the hepatocyte models with respect to effects of drugs on cytotoxicity, transcriptional and metabonomic pathways under acute and chronic conditions will be presented.