Identifying targets for bioreductive agents: using GRID to predict selective binding regions of proteins.

A computational procedure is described for investigating potential binding sites of a target macromolecule for their ability to bind both a reduced probe molecule and an oxidized probe molecule. The interaction energies are obtained using a molecular mechanics method and can be displayed as three-dimensional (3D) energy contours, indicating regions of the target molecule that may have favorable interactions with the probe molecule. Differences in the interaction energies of the oxidized and reduced probe with the target can also be plotted as contours, indicating regions that are selective for the reduced probe. These selectivity contours can be used to show whether the macromolecule is a potential target for bioreductive agents. The method has been applied to the chicken liver dihydrofolate reductase enzyme and has indicated new binding regions that may be suitable binding sites for bioreductive agents.