Anthrax is an acute infectious disease caused by the sporulating, Gram-positive bacterium Bacillus anthracis.  The anthrax toxin lethal factor (LF), an 89-kDa zinc metalloenzyme, is chiefly responsible for pathogenesis and has been a popular target for rational drug design.  Although many attempts have been made to target the LF active site with small-molecule inhibitors, no such compound has yet made it to the market as a drug.  Here, we present novel support vector machine (SVM) and pharmacophore-based models that illustrate key design principles for LF inhibitors, and that can be used to rapidly identify potential LF inhibitors from compound libraries.  These models were validated using over 500 compounds with published LF biological activity data and over 800 inactive compounds deposited in PubChem. Both types of models demonstrate very high selectivity toward strongly active LF inhibitors, with the SVM model correctly identifying 95.1% of nanomolar-level compounds, and our novel, comprehensive pharmacophore hypothesis pinpointing 72.1% of nanomolar-level inhibitors in an unbiased test set.  Additionally, both models point to key LF ligand-receptor interactions that are likely to positively influence compound potency.

Support vector machine (SVM) and pharmacophore-based models for selecting, optimizing, and prioritizing anthrax toxin lethal factor (LF) inhibitors

247th National Meeting (2014)

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