Benzoxaboroles are a class of boron-containing compounds with a broad range of biological activities. A subset of benzoxaboroles have antimicrobial activity due primarily to their ability to inhibit leucyl-tRNA synthetase (LeuRS) via the oxaborole tRNA trapping mechanism, which involves formation of a stable tRNA Leu-benzoxaborole adduct in which the boron atom interacts with the 2′-and 3′-oxygen atoms of the 3′-terminal tRNA adenosine. We sought to identify other antibacterial targets for this promising class of compounds by means of mode of action studies, and we selected a nitrophenyl sulfonamide-based oxaborole (PT638) as a probe molecule because it had potent antibacterial activity (MIC of 0.4 μg/mL against methicillin-resistant Staphylococcus aureus) but did not inhibit LeuRS (IC 50 > 100 μM). Analogues of PT638 were synthesized to explore the importance of the sulfonamide linker and the impact of altering the functionalization of the phenyl ring. These structure-activity relationship studies revealed that the nitro substituent was essential for activity. To identify the target for PT638, we raised resistant strains of S. aureus and whole genome sequencing revealed mutations in leuRS, suggesting that the target for this compound was indeed LeuRS, despite the lack of enzyme inhibition. Subsequent analysis of PT638 metabolism demonstrated that bacterial nitroreductases readily converted this compound into the amino analogue, which inhibited LeuRS with an IC 50 of 3.0 ± 1.2 μM demonstrating that PT638 is thus a prodrug.

The translation of time-dependent drug-target occupancy to extended pharmacological activity at low drug concentration depends on factors such as target vulnerability and the rate of target turnover. Previously, we demonstrated that the postantibiotic effect (PAE) caused by inhibitors of bacterial drug targets could be used to assess target vulnerability, and that high levels of target vulnerability coupled with relatively low rates of target resynthesis resulted in a strong correlation between drug-target residence time and the PAE following compound washout. Although the residence time of inhibitors on UDP-3-O-acyl-N-acetylglucosamine deacetylase (LpxC) in Pseudomonas aeruginosa (paLpxC) results in significant PAE, inhibitors of the equivalent enzyme in Escherichia coli (ecLpxC) do not cause a PAE. Hyperactivity of the fatty acid biosynthesis enzyme FabZ or the inclusion of sub-MIC levels of azithromycin lead to the observation of a PAE for three inhibitors of ecLpxC. FabZ hyperactivity has been shown to stabilize ecLpxC, and using mass spectrometry, we demonstrate that the appearance of a PAE can be directly linked to a 3-fold increase in the stability of ecLpxC. These studies substantiate the importance of target turnover in time-dependent drug activity.

No abstract

The enoyl-acyl carrier protein (ACP) reductase (ENR) is a key enzyme within the bacterial fatty-acid synthesis pathway. It has been demonstrated that small-molecule inhibitors carrying the diphenylether (DPE) scaffold bear a great potential for the development of highly specific and effective drugs against this enzyme class. Interestingly, different substitution patterns of the DPE scaffold have been shown to lead to varying effects on the kinetic and thermodynamic behavior toward ENRs from different organisms. Here, we investigated the effect of a 4′-pyridone substituent in the context of the slow tight-binding inhibitor SKTS1 on the inhibition of the Staphylococcus aureus enoyl-ACP-reductase saFabI and the closely related isoenzyme from Mycobacterium tuberculosis, InhA, and explored a new interaction site of DPE inhibitors within the substrate-binding pocket. Using high-resolution crystal structures of both complexes in combination with molecular dynamics (MD) simulations, kinetic measurements, and quantum mechanical (QM) calculations, we provide evidence that the 4′-pyridone substituent adopts different tautomeric forms when bound to the two ENRs. We furthermore elucidate the structural determinants leading to significant differences in the residence time of SKTS1 on both enzymes.

UDP-3-O-(R-3-hydroxymyristoyl)-N-acetylglucosamine deacetylase (LpxC) is a promising drug target in Gram-negative bacteria. Previously, we described a correlation between the residence time of inhibitors on Pseudomonas aeruginosa LpxC (paLpxC) and the post-antibiotic effect (PAE) caused by the inhibitors on the growth of P. aeruginosa. Given that drugs with prolonged activity following compound removal may have advantages in dosing regimens, we have explored the structure–kinetic relationship for paLpxC inhibition by analogues of the pyridone methylsulfone PF5081090 (1) originally developed by Pfizer. Several analogues have longer residence times on paLpxC than 1 (41 min) including PT913, which has a residence time of 124 min. PT913 also has a PAE of 4 h, extending the original correlation observed between residence time and PAE. Collectively, the studies provide a platform for the rational modulation of paLpxC inhibitor residence time and the potential development of antibacterial agents that cause prolonged suppression of bacterial growth.