We propose to develop a new class of antimicrobial drugs based on the fundamental principles of transition state analysis. Rather than structure-based drug design, which is based largely on substrate mimicry, transition-state analysis is reaction-based drug design, stemming from a rigorous chemical evaluation of the relevant catalytic chemistry to reveal the enzyme mechanism and the structural changes that stabilize a transition state. Transition state analysis will yield small molecule transition state analogs that closely mimic the transition state structure. For the proposed studies we selected Glutamate Racemase (GR), an increasingly important antimicrobial drug target. GR has been widely validated to be an attractive drug target in numerous pathogenic species. GR-catalyzed racemization is primarily achieved through extensive flexibility, which is information that is largely absent from GR crystal structures. Recent studies by our research group have strongly suggested that a more chemically diverse inhibitor space can be discovered against GR by considering its transition state structure in virtual screening campaigns. These studies have produced a potentially powerful approach for the discovery of diverse inhibitory scaffolds with high ligand efficiency, which provides a solution to the problems of meeting the relatively narrow requirements of antimicrobial drug space. However, the enormous potential of virtual screening methods are significantly hindered by the pronounced failures to relatively rapidly make quality predictions about protein-ligand affinities. This proposal directly fills two significant and related knowledge gaps that hinder discovery of true transition state inhibitors for GR: 1) determination of an experimentally validated transition state structure for GR, and the transition state pharmacophore that leads to the ultra tight binding along the reaction trajectory and 2) how to accurately rank-order hits from virtual screening against a highly flexible protein receptor. The successful completion of the proposed studies will enable the discovery and design of novel high efficiency inhibitory scaffolds for flexible enzyme drug targets, and thus yield transformative results in the field of drug discovery.
The specific aims for this proposal are as follows:
Aim 1 : An integrated computational approach to solve the rank-order problem for a flexible enzyme drug target: the Flexible Enzyme Receptor Method (FERM) for Steered MD (SMD)-Docking.
Aim 2 : A FERM Challenge: a conformational inventory of GR using a library of conformationally restricted glutamate analogs (the spiro[3.3]heptane family) Aim 3: Elucidation of the Transition State Pharmacophore for GR (B. subtilis RacE and B. anthracis RacE1 and RacE2) via Kinetic Isotope Effects.
Aim 4 : Bringing it all together: the application of FERM-SMD Docking against the GR transition state pharmacophore and ground state ensembles.
There are enormous opportunities to discover novel high potency drugs against enzyme targets by fully capturing the dynamic changes that occur in the enzyme receptor, which are related to their intrinsic ultra-tight binding to reaction intermediates. The key challenges are related to accurately capturing both small and large scale protein motions, which are not readily available from structural data, and also accurate representation of the behavior of electrostatics at the protein solvent interface. The current proposal describes a method that rapidly captures and integrates the missing information, using both experimental and computational approaches, which will eliminate the many false positives that lead to inaccurate predictions, and facilitate the discovery of novel and potent small molecule therapeutics.