Beyond cytochrome P450 (CYP450)-mediated drug clearance, aldehyde oxidase (AO) metabolism of drugs is of increasing importance owing to the failure of multiple clinical candidates due to AO-mediated metabolic liabilities. AO is a cytosolic enzyme with broad substrate specificity. Previous structure-metabolism relationships studies for AO either were focused on a handful of compounds belonging to a specific chemical class (e.g. pyridine derivatives) or have analyzed datasets that do not cover a diverse chemical space. In this study, we screened nearly 1600 compounds to extract transformation rules that could guide medicinal chemists in optimizing hit compounds with potential AO liabilities. We collected in vitro half-life data for these compounds by performing in-house experiments on mouse (CD-1 male) and human cytosol (mixed gender) fractions. Matched molecular pairs analysis was performed in conjunction with sub-structural fingerprint analysis to reveal the chemical structure transformations leading to increased cytosolic stability and vice-versa. The transformation rules were prospectively validated on test datasets. Accuracies of 75% and 78.6% were achieved for rules that transform unstable compounds into stable compounds when applied on mouse and human datasets, respectively. Through exemplary cases, we highlight the significance of these rules in improving the cytosolic stability. Furthermore, the rules will be made publicly-accessible through a web-service (accessible at https://tripod.nih.gov/solmet) that facilitates browsing or prediction of chemical transformations that could be used to rationally modulate compounds and attenuate cytosol-mediated metabolic liability.

Matched molecular pairs analysis for the prediction of cytosol-mediated metabolic liabilities

2019 Spring National Meeting

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