Lead Optimization and Slow-Onset Inhibition of the Enoyl-ACP Reductase (InhA) from Mycobacterium Tuberculosis

Abstract

The enoyl-ACP reductase (InhA) catalyzes the last step of fatty acid biosynthesis in Mycobacterium tuberculosis, and is a validated target for antitubercular drugs. Our group previously reported diaryl ether InhA inhibitors that are highly active against M. tuberculosis (MIC = 3.15 ??g/mL). Further studies showed that introducing an ortho methyl group on the inhibitor B-ring resulted in a 430-fold increase in binding affinity. To understand the role of this methyl group, and design more potent inhibitors, analogs with various ortho groups on the B-ring were synthesized and evaluated. Enzyme inhibition was found to be highly sensitive to the size of the B-ring substituent, with electron withdrawing groups resulting in the highest binding affinity. Analogs with two B-ring substituents were also synthesized, but none resulted in improved inhibition, presumably due to steric hindrance. Pyridyl and pyrimidinyl B-rings were also incorporated, which led to an improvement in antimicrobial activity (MIC = 0.4 ??g/mL). The diaryl ethers were found to have modest in vivo efficacy, likely limited by high lipophilicity and metabolic instability. Pyranones and 4-pyridones were used to replace the phenol A-ring, which led to low ClogP values (2-4) and increased metabolic stability. The top compound had comparable MIC values to the diaryl ethers and an AUC/MIC value that had increased 5 fold. In order to reduce the number of rotatable bonds, cyclic groups were used to replace the flexible hexyl group on the A-ring. Amine and ether linkers between the cyclic groups and the A-ring resulted in lower lipophilicity without altering the binding affinity. Drug-target residence time is important for in vivo drug efficacy. Progress curve analysis was used to identify diaryl ethers that were time dependent inhibitors of InhA. Together with X-ray crystallography, MD simulations, and direct measurements of inhibitor dissociation, the residence time of the compounds on InhA were correlated with the energy changes during dissociation. These data support the importance of helix-6 and helix-7 motion in the time dependent inhibition of InhA, together with the observation that a steric clash between residues V203 and I215 is particularly important for modulating the energy of the transition state leading to the final enzyme-inhibitor complex. Significantly, replacement of V203 or I215 with Ala resulted in the loss of slow-onset inhibition by PT070. However, slow-onset inhibition was regained with PT162 and PT163, which place bulky groups between the mutated residues, thus compensating for the decrease in side chain bulk. These studies support the importance of interactions between helix-6 and 7, and suggest ways in which residence time can be rationally modulated.