Prediction of microvesicular liver steatosis: a read-across case study with branched carboxylic acids

To date, toxicologists use in vivo animal studies to estimate a threshold below which the exposure of in vivo humans to a compounds is safe. EU-ToxRisk aims to support the use of new approach methods (NAMs), such as in vitro and in silico models, in risk assessment. Here we describe a proof of concept case study that investigates the toxicity of a group of (un)branched aliphatic carboxylic acids. Five carboxylic acids induce microvesicular liver steatosis in repeated dose toxicity studies (termed in vivo positive), whereas six structurally highly similar compounds do not (in vivo negative). Here we demonstrate the use of NAMs for a category approach comprising ethylbutyric acid and valproic acid, 4-ene valproic acid, 2-ethylhexanoic acid and pivalic acid. All compounds were tested in HepG2 cells. Exposure of HepG2 cells to test compounds resulted in lipid accumulation at subcytotoxic concentrations after 1 and 3 days of treatment. The onset of lipid accumulation was seen at lower concentration with longer exposure periods. A panel of 27 CALUX reporter gene assays and BAC GFP reporter cell lines showed activation of cellular stress pathways in acute exposures (up to 72h). All assays distinguish clearly between in vivo positive and negative compounds. The nominal EC20 values correlate to NOELs of in vivo studies. These data are compared to recent results obtained from repeated exposures of 3D HepaRG cells, PHHs and zebra fish embryos. We also used an in silico model to predict the corresponding intracellular concentration in the in vitro models. These data are included as additional information into the in vitro to in vivo correlations. This exemplary category approach demonstrates that NAMs from the EU-ToxRisk predict the toxicity of compounds after repeated exposure in a qualitative and quantitative manner. The continued development of integrated testing strategies represents the beginning of a paradigm shift, reducing, and ultimately replacing, in vivo studies in human risk assessment.