Veröffentlichungen der Fakultät für Chemie und Chemische Biologie
Permanent URI for this collection
Browse
Recent Submissions
Item In vitro–in vivo extrapolation of hepatotoxicity for food-relevant compounds in the rat(2025) Gründler, Lisa; Hengstler, Jan G.; Rahnenführer, JörgHepatotoxicity of food-relevant compounds poses a significant public health risk, ranging from mild elevations in liver enzymes to severe liver failure. Particular concern arises from contaminants such as mycotoxins (e.g., aflatoxin B1), environmental pollutants including per- and polyfluoroalkyl substances (PFAS), pesticides, and pharmaceuticals found in drinking water. Chemical safety assessments still largely rely on animal studies, in which repeated-dose toxicity is evaluated based on the administered dose. To refine the assessment of organ-specific toxicity, a novel approach methodology (NAM), following Albrecht et al. (2019), was applied. This integrated in vitro and in silico methods to compare in vitro effective concentrations with the simulated peak plasma concentration in the portal vein at the oral dose corresponding to the lowest observed adverse effect level (LOAEL). Specifically, a species-internal in vitro–in vivo extrapolation (IVIVE) strategy employing physiologically based pharmacokinetic (PBPK) modeling was used to convert oral hepatotoxic LOAELs from repeated-dose animal studies into estimated portal vein plasma concentrations. Unlike in humans, for rats there is dose-dependent data on liver-specific adverse effects from repeated-dose studies, making them a suitable reference. The aim was to compare these modeled Cmax concentrations at the LOAEL with effective concentrations measured in in vitro assays. A total of 40 food-relevant compounds—including PFAS, polychlorinated biphenyls (PCBs), pharmaceuticals, pesticides, and plant constituents—were evaluated. Hepatotoxic doses were systematically extracted from published rat studies using a novel scoring system (Score of Significance, SOS), which provides a measure of study quality and enables comparison of adverse effect severity across compounds. To derive in vitro effective concentrations, three rat hepatocyte-based models were employed: two hepatoma cell lines (H4IIE, Zajdela) and cultured primary rat hepatocytes. A test battery covering three endpoints—cytotoxicity, lipid droplet accumulation, and nuclear enumeration via Hoechst staining—was used to determine effective concentrations (EC₁₀, EC₂₀, EC₅₀) for each compound, cell system, and endpoint. Comparison of in vitro effective concentrations with modeled in vivo Cmax values at the LOAEL was performed to identify the most predictive assay and cell system. This was achieved using a custom performance metric, the Toxicity Iso-concentration Index (TII), which quantifies the deviation from the iso-concentration line in either direction in a log₁₀-scaled scatterplot for each in vitro–in vivo pair. The results showed that the hepatoma cell lines outperformed primary rat hepatocytes across all assays (TII > 0.8 vs. TII < 0.8). Among the assays, cytotoxicity exhibited the best predictive performance and yielded the highest number of positive test results. Aggregating the results of all three assays into a single lowest observed effect concentration (LOEC) per cell system did not lead to an improvement in the TII. In the best-performing setup—cytotoxicity testing in H4IIE cells—50 % of the compounds matched in vivo concentrations within a 3.16-fold deviation, another 25 % within a 3.16–10-fold range, 21 % between 10–100-fold, and only 3.4 % deviated by more than 100-fold. According to PBPK modeling standards in drug development, 75 % of compounds falling within a 10-fold deviation are considered good to acceptable, and the remainder warrant further investigation. Therefore, this approach enables a quantitative estimation of the in vivo Cmax in the portal vein at the LOAEL, with a geometric mean fold error of 5.3 and a geometric standard deviation of 5.5. The resulting dataset provides species- and organ-specific concentration–effect relationships and has potential utility for regulatory databases such as ECHA, PubChem, or the CompTox Chemicals Dashboard. Furthermore, it may serve as a training set for machine learning algorithms to predict the hepatotoxicity of untested substances, thereby contributing to more efficient, animal-free chemical safety assessment strategies.Item Identification of a novel mechanism driving NAFLD progression and therapeutic strategies(2021) Myllys, Maiju Karoliina; Hengstler, Jan G.; Watzl, CarstenNon-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in Western countries with an increasing prevalence of approximately 25 %. NAFLD comprises several stages, starting as benign steatosis and progressing to non-alcoholic steatohepatitis (NASH), and in some cases to liver cirrhosis and hepatocellular carcinoma (HCC). Although several emerging therapies are currently in clinical trials, so far there are no approved drugs for treatment of NASH. The overarching goal of this thesis was to investigate the mechanisms of NAFLD stage transitions, and to establish preventive measures for the progression from benign steatosis to NASH. First, a mouse model of NAFLD progression was established by long-term feeding of male C57Bl/6N mice with western-style diet (WD) up to 54 weeks. The disease progression was evaluated time-dependently by biochemical, histopathological, and immunohistochemical analyses as well as by intravital two-photon-based imaging. This analyses revealed six stages in NAFLD progression: (1) benign steatosis, (2) macrophage crown-like structure formation, (3) macropinocytosis of bile, (4) ductular reaction, (5) dedifferentiation and functional shutdown, and (6) tumor nodule formation. The novel finding of this thesis was the identification of stage 3, where a retrograde vesicular uptake of bile from bile canaliculi to hepatocytes led to toxic accumulation of bile acids in the liver tissue, providing a link between NAFLD and cholestasis. The phenomenon was further identified as macropinocytosis by treating the mice with a macropinocytosis-specific inhibitor imipramine. As a result, a single application of imipramine efficiently blocked macropinocytosis in WD-fed mice. Interestingly, a long-term application of imipramine for 8 weeks decreased the bile acid concentrations in the liver tissue and led to significant NAFLD amelioration. Moreover, bile macropinocytosis was also found relevant in human NAFLD patients as detected by the presence of fragments of bile canaliculi within steatotic hepatocytes. In conclusion, an NAFLD mouse model recapitulating the different stages of human NAFLD progression to NASH and eventually to HCC was successfully established. Moreover, a novel mechanism possibly driving NALFD progression was identified as macropinocytosis of bile from bile canaliculi back to hepatocytes.Item Carbonic anhydrase dataset for use with Scaffold Hunter(2017) Humbeck, Lina; Koch, Oliver