New antibiotics with novel targets are greatly needed. Bacteria have numerous essential functions, but only a small fraction of such processes-primarily those involved in macromolecular synthesis-are inhibited by current drugs. Targeting metabolic enzymes has been the focus of recent interest, but effective inhibitors have been difficult to identify. We describe a synthetic azetidine derivative, BRD4592, that kills Mycobacterium tuberculosis (Mtb) through allosteric inhibition of tryptophan synthase (TrpAB), a previously untargeted, highly allosterically regulated enzyme. BRD4592 binds at the TrpAB α-β-subunit interface and affects multiple steps in the enzyme's overall reaction, resulting in inhibition not easily overcome by changes in metabolic environment. We show that TrpAB is required for the survival of Mtb and Mycobacterium marinum in vivo and that this requirement may be independent of an adaptive immune response. This work highlights the effectiveness of allosteric inhibition for targeting proteins that are naturally highly dynamic and that are essential in vivo, despite their apparent dispensability under in vitro conditions, and suggests a framework for the discovery of a next generation of allosteric inhibitors.

Many species of proteobacteria communicate with kin and coordinate group behaviors through a form of cell-cell signaling called acyl-homoserine lactone (AHL) quorum sensing (QS). Most AHL receptors are thought to be specific for their cognate signal, ensuring that bacteria cooperate and share resources only with closely related kin cells. Although specificity is considered fundamental to QS, there are reports of “promiscuous” receptors that respond broadly to nonself signals. These promiscuous responses expand the function of QS systems to include interspecies interactions and have been implicated in both interspecies competition and cooperation. Because bacteria are frequently members of polymicrobial communities, AHL cross talk between species could have profound impacts. To better understand the prevalence of QS promiscuity, we measured the activity of seven QS receptors in their native host organisms. To facilitate comparison of our results to previous studies, we also measured receptor activity using heterologous expression inEscherichia coli. We found that the standardE. colimethods consistently overestimate receptor promiscuity and sensitivity and that overexpression of the receptors is sufficient to account for the discrepancy between native andE. colireporters. Additionally, receptor overexpression resulted in AHL-independent activity inPseudomonas aeruginosa. Using our activation data, we developed a quantitative score of receptor selectivity. We find that the receptors display a wide range of selectivity and that most receptors respond sensitively and strongly to at least one nonself signal, suggesting a broad potential for cross talk between QS systems.IMPORTANCESpecific recognition of cognate signals is considered fundamental to cell signaling circuits as it creates fidelity in the communication system. In bacterial quorum sensing (QS), receptor specificity ensures that bacteria cooperate only with kin. There are examples, however, of QS receptors that respond promiscuously to multiple signals. “Eavesdropping” by these promiscuous receptors can be beneficial in both interspecies competition and cooperation. Despite their potential significance, we know little about the prevalence of promiscuous QS receptors. Further, many studies rely on methods requiring receptor overexpression, which is known to increase apparent promiscuity. By systematically studying QS receptors in their natural parent strains, we find that the receptors display a wide range of selectivity and that there is potential for significant cross talk between QS systems. Our results provide a basis for hypotheses about the evolution and function of promiscuous signal receptors and for predictions about interspecies interactions in complex microbial communities.