The first project area explores metabolic pathways that have been proposed based on in vitro studies to be important in non-replicating (NR)-MTb. We are exploring the importance of the biosynthesis of the cofactors biotin, coenzyme A and pyridoxal, peptidoglycan turnover, the role of putative F420-binding and genetically annotated pyridoxal-generating enzymes, beta-oxidation and iron acquisition and validating these by chemical and genetic means in non-replicating (NR)-MTb. We have shown that Rv2607 is the canonical pyridoxine phosphate oxidase of MTb and have enzymatically characterized this enzyme. In contrast, Rv1155, which is also annotated as a pyridoxine phosphate oxidase family protein has been expressed, purified, crystalized with its F420 cofactor, biophysically characterized with and without bound cofactor and we are attempting to identify the natural substrate of this protein by analyzing shared chemotypes with known metabolites from fragments identified as binders to this protein. Another F420-dependent enzyme, Rv2991, has been crystalized and fragments chemically similar to known metabolites of flavoenzymes analyzed for binding to Rv2991 with and without F420. By analyzing common pharmacophores between known metabolites and the binders identified by this fragment-based approach, we are attempting to probe the enzymatic function of this unknown protein We have also demonstrated the importance of biotin synthesis for the viability of MTb in vitro and in vivo. We have reported that conditional downregulation of pantothenate synthase makes Mtb hypersusceptible to inhibitors of coenzyme A biosynthesis and are using this approach to identify vulnerable targets in this metabolic pathway. Our studies of mycobacterial cell wall synthesis using meropenem as probe have allowed us to track the formation of the various layers of the mycobacterial cell wall during its assembly using a combination of cryo-electron, transmission and scanning electron microscopy. We have shown that the dual action of meropenem on both the D,D-carboxypeptidases as well as the transpeptidases on newly synthesized peptidoglycan leads to the observed polar lysis of cells. The second major focus area of this project starts from a different perspective and uses compounds that are in clinical development (PA-824 and SQ109) which are known to possess activity against replicating as well as NR-TB. We capitalized our recently determined crystal structure of Ddn, the nitroreductase responsible for the bioreductive activation of PA824 to understand the differences in binding of the enzyme to nitroimidazoles and the relationship of this binding to the formation of the reactive nitrogen intermediates responsible for killing of Mtb. We are attempting to understand what the natural substrate is for the Ddn, since this will allow us to probe the enzymatic processes that are important during non-replicating persistence. Preliminary studies have identified menaquinone as a substrate for this enzyme. For SQ109 we were able to demonstrate that the mechanism by which this drug kills Mtb is by inhibition of the MmpL3 protein which we identified as a trehalose monomycolate transporter. To further unravel the key events in cell wall mycolyl-arabinogalactan synthesis, we have enzymatically characterized the three mycolyl transferase enzymes (Antigens 85 A, B and C). We have found that the enzymes are kinetically distinct with Ag85C being enzymatically the most active and that certain amino acid residues residing in a secondary ligand binding site control rates of acyl transfer by affecting protein confirmation in a helix connecting the two ligand binding pockets. The third major focus of this project involves global approaches to understanding the metabolism in NR-TB. Using a chemostat model of MTb combined with metabolomic studies, we demonstrated that the NADH/NAD+ ratio changed as a function of oxygen concentration, that the direction of the TCA cycle reverses under hypoxia with concomitant extracellular succinate accumulation which is consistent with a model of oxygen-induced stasis in which an energized membrane is maintained by coupling the reductive branch of the TCA cycle to succinate secretion. An essential non-redundant step in this process is fumarase and we have initiated studies to validate the role of the forward as opposed to reverse TCA cycle in vitro as well as in vivo by using structure-based design based on the fumarase crystal structure to design inhibitors of this target. Co-crystal structures of Mtb fumarase with bound inhibitors, enzymatic as well as in situ demonstration of fumarase inhibition have corroborated our model with further inhibitor optimization being required for in vivo studies. In a fourth approach, we are identifying inhibitors of metabolism by high-throughput screening approaches performed under a variety of in vivo relevant environmental conditions. Hits from these screens have provided a useful tool to map metabolism of MTb as a function of carbon source, oxygen concentration or presence of low pH in the presence or absence of nitrosative stress and are currently being studied to identify the target. In the process of target identification, parallel studies are done to rapidly progress the hits to in vivo proof of concept studies so that the importance of the target for in vivo pathogenesis can be validated early on in the drug discovery process. We are studying some of the hits that were identified from a 35,000 compound BioFocus collection in collaboration with various researchers in South Africa. In addition, hits from a 100,000 compound library screen from a collaborator have yielded 12 different scaffolds that are being pursued. The scaffolds that gave us evidence of a specific target based on SAR studies were taken further into target identification by a combination of approaches including resistant mutant generation followed by whole genome sequencing to identify single nucleotide polymorphisms, transcriptional profiling, macromolecular incorporation assays and metabolomics studies. For 2 chemically different scaffolds, the same target in mycobacterial cell wall synthesis was identified and efficacy studies confirmed that inhibition of some cell wall biosynthetic genes in vivo, led to a mild bacteriostatic effect. The targets of eleven other scaffolds were identified. For several other scaffolds, mutations in MmpL3, a protein we previously identified as the SQ109 target, conferred resistance suggesting that this transporter is promiscuous in its ability to bind diverse ligands. For several scaffolds, generation of resistant mutants was impossible and in several of these cases, inability to generate resistant mutants was correlated with mammalian cytotoxicity suggesting a non-specific mechanism of action. One class of compounds was shown to target oxygen-dependent respiration in Mtb. We have demonstrated that the coupling of respiration to energy generation in a vulnerable point in NR-Mtb based on inhibitors identified in a screen against anaerobically persisting Mtb.The precise point in inhibition of respiration is currently being explored by analysis of respiratory knockout mutants, biochemical assays and complementation studies. Resistance to another hit mapped to an enzyme in folate metabolism. We have been able to show that this drug functions as a metabolic poison by its ability to mimic substrates and become incorporated into folate-like metabolites by a combination of metabolomics and biochemical analyses. With collaborators at Weill Cornell Medical College, we have used this inhibitor as well as other known inhibitors of folate biosynthetic enzymes to explore how perturbation of folate-dependent reactions leads to inhibition of Mtb replication.
Showing the most recent 10 out of 58 publications
