Hepatocellular Organic Anion Transporting Polypeptides (OATP1B1, OATP1B3, and OATP2B1) mediate uptake of endogenous compounds (e.g. bile salts, bilirubin) and xenobiotics into hepatocytes. They are responsible for altered pharmacokinetics of drugs, drug-drug interactions and liver toxicity. Data sparseness has significantly hampered modeling efforts done for hepatic OATPs so far.<br/>Here, we applied different modeling techniques in order to unravel the molecular basis for ligand activity and selectivity for these transporters: data mining, (sub)structural pattern analysis of ligands, predictive machine learning, ligand-based pharmacophores, and ensemble docking into distinct protein conformations.<br/>Ligand data was integrated from five different sources in a KNIME workflow (Türková J. Chem. Inf. Model. 2018). Detected ligands were analyzed with respect to enriched scaffolds, which has identified interesting SAR series to be studied further. The steroidal scaffold shows ubiquitous activity against hepatic OATPs, and differences in selectivity profiles can rather be attributed to the existence of certain side chains. The curated data sets were further used for building a general predictive OATP inhibition model as well as models for the individual hepatic OATPs. The imbalance of the data sets with respect to the distribution of actives vs inactives was counterbalanced by using techniques such as stratified bagging and conformal prediction. Balanced accuracies of the models are ranging from 0.73 to 0.79.<br/>For comparative modeling, representatives of the Major Facilitator Superfamily have been identified as useful templates since they are showing a high secondary structural similarity with hepatic OATPs. Normal mode simulations were performed on selected templates for the sake of exhaustive sampling of alternative conformations. Afterwards, homology models were constructed for selected conformations and served as a basis for large-scale docking screens of steroidal compounds into the three hepatic OATPs. Our results so-far suggest the presence of two distinct binding sites for steroid-associated compounds.<br/>In summary, combining ligand-based approaches, enhanced sampling of protein conformations, and molecular docking aids in elucidating molecular determinants contributing to the preferred activity of investigated steroidal compounds towards individual hepatic OATPs. Mutational studies to biochemically proof our hypotheses are currently under active investigation.

Linking multiscale data analyses, ligand- and structure-based modeling to explore ligand interactions with hepatic organic anion transporting polypeptides

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.