Many drug candidates fail due to lack of efficacy in Phase II clinical trials. This failure occurs in all therapeutic areas and primarily stems from poo in vivo efficacy as well as lack of safety (toxicity). We hypothesize that the use of drug-target residence time (tR) measurements, together with other thermodynamic estimates of compound potency, would improve the ability to predict drug efficacy in vivo. Since much of our appreciation for the importance of tR is anecdotally based on the observation that many drugs dissociate slowly from their targets, demonstration of correlations between tR and in vivo drug activity within specific compound series would, when coupled with knowledge of drug pharmacokinetics, allow mathematical models to be created that predict drug pharmacodynamics. This goal is innovative and will create a paradigm shift in how information on the interaction of lead compounds (inhibitors, agonists, antagonists) with their targets is both gathered and used. To meet this goal, we will use a combination of X-ray crystallography, site-directed mutagenesis, chemical synthesis, and computational methods. In particular, time-independent molecular dynamics (MD) simulations will help unravel the specific atomic-level interactions that are probed by the binding kinetics measurements, and will provide dynamic information to fill in the gaps in time between the stable states observed in the crystal structure. This will be accomplished using the FabI enzymes from Mycobacterium tuberculosis (mtFabI) and Staphylococcus aureus (saFabI), both of which are clinically relevant drug targets. In addition, while both enzymes are inhibited by the diphenyl ether compound class, and a second related series based on pyridones, through the same two-step slow-onset induced-fit kinetic mechanism, the structural changes that accompany enzyme inhibition differ. Thus our goal is to determine whether we can first understand and then rationally modulate residence time in two distinct enzymes whilst keeping the compound class constant. This will provide a platform for translating our knowledge to other systems.
In Aim 1 we will elucidate the mechanism for the time-dependent inhibition of mtFabI. Time-indendent MD simulations and X-ray crystallography will be used to determine the structure of the transition state leading to the final enzyme-inhibitr complex (E-I*) and to identify key interactions critical for time-dependent inhibition. Inhibitors with increased tR values will be designed. This will provide a detailed understanding of an induced-fit binding mechanism.
In Aim 2 we will determine the structural basis for the time-dependent inhibition of saFabI. This will be accomplished using kinetic and structural approaches. Additional analogues will be synthesized to interrogate our understanding of slow-onset saFabI inhibition.
In Aim 3 we will delineate the relationship between tR, post-antibiotic effect (PAE) and in vivo activity. The PAE is the persistent suppression of microbial growth following drug exposure and removal, and is a well-known and frequently observed phenomenon in microbiology with widely stated implications for antimicrobial pharmacokinetics and the development of improved dosing regimens. The contribution of tR to PAE and, ultimately, in vivo antibacterial activity will be evaluated in S. aureus. The PAE measurements on live cells will provide a bridge between in vitro and in vivo estimates of drug activity. These studies will provide a foundation for using residence time in drug discovery. At a broader level, our studies will provide insight into the time dependence of conformational changes in proteins and how these relate directly to protein function, and will provide a platform for exploring the structural basis for time-dependent enzyme inhibition in other systems. Demonstrating the importance of tR will lead to a paradigm shift in lead compound optimization.
We hypothesize that drug-target residence time (tR) should be used in addition to measurements of thermodynamic affinity for advancing compounds along the drug discovery pipeline. This is because drug concentration is not constant in vivo. In the current proposal we will elucidate the molecular factors that modulate drug-target lifetime for a bacterial target. These studies will provide a foundation for using residence time in drug discovery. At a broader level, our studies will provide insight into the time dependence of conformational changes in proteins and how these relate directly to protein function, and will provide a platform for exploring the structural basis for time-dependent enzyme inhibition in other systems. Demonstrating the importance of drug-target residence time will lead to a paradigm shift in lead compound optimization.
Showing the most recent 10 out of 39 publications
