acs.jmedchem.6b00875

Jun 20, 2016 - Copyright © 2016 American Chemical Society. *E-mail: [email protected]. Phone: 617-432-3335. Cite this:J. Med. Chem. 59, 13 ...
0 downloads 0 Views 273KB Size
Viewpoint pubs.acs.org/jmc

Could Killing Bacterial Subpopulations Hit Tuberculosis out of the Park? Catherine Baranowski† and Eric J. Rubin*,†,‡ †

Department of Immunology and Infectious Disease, Harvard T. H. Chan School of Public Health, Boston, Massachusetts 02115, United States ‡ Department of Microbiology and Immunobiology, Harvard Medical School, Boston, Massachusetts 02115, United States ABSTRACT: One hurdle to treating tuberculosis could be that it is so difficult to kill nonreplicating subpopulations of the causative pathogens. This work describes two new cephalosporin derivatives that specifically target this population of Mycobacterium tuberculosis.

W

hy does it take so long to treat tuberculosis? Much evidence suggests that while some causative bacteria are killed rapidly by antibiotics, many continue to survive for extended periods of time despite therapy. Some of these survivors could live in compartments where drugs penetrate poorly or are poorly active. However, others are consistently exposed to drugs. These persistent bacteria are genetically identical to those that die early.1 But clearly, they are antibiotic tolerant, at least for a time. What explains this phenotypic drug tolerance? While we do not know precisely, a prevailing theory is that these individual bacteria adopt an altered physiology that makes them less antibiotic susceptible. At least two mechanisms may be at play. Some bacteria can survive for a time but will eventually die, particularly if rechallenged with drug. This has been referred to as class I persistence, a temporary state. Other populations enter a longer-term nonreplicating state, class II persistence (Figure 1).2 Often, these bacteria can only be recovered under special conditions, such as using specific growth media. Are these mechanisms important in preventing rapid cure of tuberculosis? We simply do not know, as the appropriate tools for answering this question have been unavailable. All currently used antibiotics are most effective against rapidly dividing Mycobacterium tuberculosis (Mtb), the causative organism of tuberculosis. While some agents can also kill phenotypically tolerant cells, because they are still active against replicating organisms, we cannot use them to distinguish the importance of different physiologic states. In order to establish the roles of these various populations we need cleaner tools, specifically drugs that only target class II persisters. This is the starting point for the study by Gold et al. in this issue.3 By combining multiple stresses, the authors previously devised an assay for killing of nonreplicating Mtb that allowed them to screen in a high-throughput format. They screened a library of compounds that contained a number of novel βlactams, an unusual first choice for this biological activity (see further discussion below). Nevertheless, they found three cephalosporin esters that were able to kill nonreplicating cells at relatively low concentrations. Several derivatives, including esters and oxadiazoles, had similar activities. Strikingly, all of these compounds were able to kill nonreplicating cells but had little or no activity against replicating cells. The compounds © XXXX American Chemical Society

Figure 1. Targeting of bacterial subpopulations by antibiotics. Different subpopulations grow at different rates during Mycobacterium tuberculosis (Mtb) infection. One classification places these into phenotypic categories: rapidly growing and two slowly growing or nondividing groups. Rapidly growing and replicating cells, shown here as green rods, are targeted successfully by currently used Mtb therapy. The multicolored cells, termed class I persisters by the authors, depicted as multicolored rods, survive antibiotics, but after regrowth in the absence of drug, this subpopulation reverts to drug susceptible cells. Class II persisters, shown in gray, represent a subpopulation of cells that halts replication and growth, becoming tolerant to drugs targeting these processes. These cells are specifically targeted by the cephalosporins described by Gold et al.3

were able to kill Mtb inside macrophages, particularly after immune activation. This could be related to the finding in axenic culture that these cephalosporins act synergistically with reactive nitrogen species, substances produced by activated macrophages. The finding that β-lactams preferentially kill nonreplicating organisms is surprising. β-Lactams are certainly the most Received: June 10, 2016

A

DOI: 10.1021/acs.jmedchem.6b00875 J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

■ ■

successful class of antibiotics. But clinically useful β-lactam drugs all target synthesis of peptidoglycan (PG), a cell wall polymer. These agents block transpeptidation, the reaction that produces cross-links between the side chains off carbohydrate strands that stabilize the cell wall. But because most β-lactams target the transpeptidation that occurs during times of growth, these drugs are most effective against rapidly growing bacteria. In fact, most have no activity against nondividing cells. This suggests that the target for the cephalosporins described by Gold et al. is different from that blocked by currently available β-lactams. There are some clues about this potential target or set of targets. Most β-lactams inhibit one or more members of the penicillin-binding protein (PBP) family. These enzymes catalyze D,D-transpeptidation, a specific linkage between the fourth and third residues of two PG side chains. In mycobacteria, most stationary phase PG contains a different cross-link: the L,D linkage between the third and third amino acid of two PG side chains.4 The role of this different linkage is unknown, but it appears important for Mtb surivial and βlactam resistance in vivo.5 Interestingly, it appears that L,Dtranspeptidation does not require new PG synthesis. One class of β-lactam drugs, the carbapenems, has some activity against nonreplicating Mtb. And while these drugs do bind to several PBPs, they also are able to inhibit L,D-transpeptidases. It is certainly conceivable that the cephalosporins described here have similar activity against L,D-transpeptidases but lack the ability to block D,D-transpeptidation. Certainly the compounds described by Gold et al. have structural differences from almost all cephalosporins in current use, suggesting that they might have a different target. But like all β-lactams, these are reactive molecules that covalently bind to the active sites of their target enzymes. Thus, there is a clear path toward identifying potential targets that could greatly help in elucidating how these molecules are able to kill nonreplicating mycobacteria. Of course, the big question remains: Are the cells that are killed by these novel agents important in preventing us from rapidly curing tuberculosis? Two obstacles remain in determining the answer. Because we do not have an in vitro model that predicts the outcome of therapy, we will first require compounds that have activity in an animal model. We do not yet know if the compounds described in this report will have the appropriate pharmacologic properties to be useful. But there remains an even larger problem. Currently, it is unclear which, if any, animal model predicts treatment success in humans. It remains possible that testing these compounds would require a clinical trial, a considerable hurdle. Nevertheless, it is tempting to think a treatment strategy that incorporated compounds that, like these novel cephalosporins, only kill nonreplicating bacteria, together with more traditional drugs targeting rapidly growing bacilli, might lead to a shorter course of therapy for tuberculosis. This is not quite the same concept as drug synergy, where combining two drugs leads to increased susceptibility of individual bacteria. To make a baseball analogy (and with apologies to non North American readers), a team of fastball hitters can be easily defeated by a curveball pitcher. Mixing up the lineup with hitters who specialize in different pitches can be far more effective. Likewise, devising a regimen consisting of agents that kill different bacterial subpopulations might be the key to more rapid cure.

Viewpoint

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 617-432-3335.

REFERENCES

(1) Wakamoto, Y.; Dhar, N.; Chait, R.; Schneider, K.; SignorinoGelo, F.; Leibler, S.; McKinney, J. D. Dynamic persistence of antibiotic-stressed mycobacteria. Science 2013, 339, 91−95. (2) Nathan, C. Fresh approaches to anti-infective therapies. Sci. Transl. Med. 2012, 4, 140sr2. (3) Gold, B.; Smith, R.; Nguyen, Q.; Roberts, J.; Ling, Y.; Quezada, L. L.; Somersan, S.; Warrier, T.; Little, D.; Pingle, M.; Zhang, D.; Ballinger, E.; Zimmerman, M.; Dartois, V.; Hanson, P.; Mitscher, L. A.; Porubsky, P.; Rogers, S.; Schoenen, F. J.; Nathan, C.; Aubé, J. Novel cephalosporins selectively active on nonreplicating Mycobacterium tuberculosis. J. Med. Chem. 2016, DOI: 10.1021/acs.jmedchem.5b01833. (4) Lavollay, M.; Arthur, M.; Fourgeaud, M.; Dubost, L.; Marie, A.; Veziris, N.; Blanot, D.; Gutmann, L.; Mainardi, J.-L. The peptidoglycan of stationary-phase Mycobacterium tuberculosis predominantly contains cross-links generated by l,d-transpeptidation. J. Bacteriol. 2008, 190, 4360−4366. (5) Gupta, R.; Lavollay, M.; Mainardi, J.-L.; Arthur, M.; Bishai, W.; Lamichhane, G. The Mycobacterium tuberculosis protein LdtMt2 is a nonclassical transpeptidase required for virulence and resistance to amoxicillin. Nat. Med. 2010, 16, 466−469.

B

DOI: 10.1021/acs.jmedchem.6b00875 J. Med. Chem. XXXX, XXX, XXX−XXX