The anthrax toxin lethal factor (LF) and matrix metalloproteinase-3 (MMP-3, stromelysin-1) are popular zinc
metalloenzyme drug targets, with LF primarily responsible for anthrax-related toxicity and host death, while MMP-3 is
involved in cancer- and rheumatic disease-related tissue remodeling. A number of in silico screening techniques, most notably
docking and scoring, have proven useful for identifying new potential drug scaffolds targeting LF and MMP-3, as
well as for optimizing lead compounds and investigating mechanisms of action. However, virtual screening outcomes can
vary significantly depending on the specific docking parameters chosen, and systematic statistical significance analyses
are needed to prioritize key parameters for screening small molecules against these zinc systems. In the current work, we
present a series of chi-square statistical analyses of virtual screening outcomes for cocrystallized LF and MMP-3 inhibitors
docked into their respective targets, evaluated by predicted enzyme-inhibitor dissociation constant and root-meansquare
deviation (RMSD) between predicted and experimental bound configurations, and we present a series of preferred
parameters for use with these systems in the industry-standard Surflex-Dock screening program, for use by researchers
utilizing in silico techniques to discover and optimize new scaffolds.
Keywords: Anthrax, anthrax toxin lethal factor, docking and scoring, MMP-3, surflex-dock, virtual screening, zinc metalloproteinases.
Rights & PermissionsPrintExport