A dual-mechanism compound with antibacterial activity that targets both folate metabolism and membrane integrity, with a favourable resistance profile in vitro, is identified by screening a small molecule library.
The concept of combination therapies to treat bacterial infections dates back to the early days of the antibiotic era. Two antibiotics are either combined or co-formulated to broaden the spectrum for empiric therapy and achieve synergistic activity. Another option is to covalently link two antibiotics to form a dual-action hybrid compound that has the advantage of being more difficult for bacteria to evolve resistance to1. Now, Martin et al. report in Cell the identification of an antibacterial compound that kills a variety of clinically relevant bacteria, using a small molecule library of about 33,000 compounds2.
Combination therapies have long been mooted as a mean to both improve antibiotic activity and to address the global problem of antimicrobial resistance. In most cases, however, synergistic effects between compounds observed in vitro do not translate into clinical benefits. Another established way of improving efficacy is to combine an antibiotic with a potentiator, which has insufficient antibacterial activity by itself but restores — or augments — the activity of the partner antibiotic. Recently, approved combinations of cephalosporins or carbapenems with a β-lactamase inhibitor have highlighted the value of this strategy. Other attempts to combine efflux pump inhibitors or entry-enabling compounds in the form of membrane-acting agents with an antibiotic did not translate into drugs appropriate for clinical use.
Despite the hypothetical advantage of combining modalities with different mechanisms, such as synergistic activity or minimizing the risk of emergence of resistance1, antibiotic combinations have drawbacks including independent pharmacokinetics (PK), unpredictable distribution in the human body and pharmacodynamic (PD) interactions of the individual components (Fig. 1). One way to overcome PK issues is to design hybrids of two known pharmacophores into one molecule3. Dual-antibiotic hybrids with a metabolically non-cleavable bond behave as a single molecule with regards to their PK properties in the human body. If the linking bond is stable, the hybrid may behave differently in terms of the PK from either parent drug. Additionally, hybrids with links that are selectively cleaved in bacteria to achieve strain specificity have been reported4. Another example of a hybrid strategy is generating siderophore conjugates that utilise the bacterial iron import mechanism to enable an antibiotic to cross the outer membrane of Gram-negative bacteria. The only clinically successful conjugate so far is cefiderocol, which improves cephalosporin uptake through the outer membrane, increasing its antibacterial activity5. Other research and development projects with siderophore conjugates in clinical development were discontinued (such as BAL 30072 or GSK-3342830).
Many conjugates have been developed with the aim of addressing resistance and/or enabling drugs to penetrate the barrier posed by the Gram-negative cell envelope. However, despite many years of research, the few antibiotic hybrids with dual mechanisms that have been advanced to clinical development6 mainly target Gram-positive bacteria. Cefilavancin and TD-1607 are hybrid antibiotics consisting of a glycopeptide covalently bound through a stable linker to a third-generation cephalosporin. Despite the antibacterial activity of the cephalosporin against Gram-negative bacteria, the hybrid compound targets only Gram-positive bacteria and retains activity in case of resistance to the parent compounds. Fluoroquinolones are popular partners in the design of hybrid antibacterial drugs with stable linkers7. Cadazolid (which is ciprofloxacin–oxazolidinone-based and was discontinued after failing in phase III clinical development for Clostridium difficile infection), DNV-3837 (which is ciprofloxacin–oxazolidinone-based and currently in phase II clinical trials in C. difficile infection) and TNP-2092 (which is ciprofloxacin–rifampicin-based and also in phase II clinical trials in prosthetic joint infections)8 are all examples of hybrids with anti-Gram-positive activity, highlighting the great challenge to design hybrids with activity against Gram-negative rods.
The integration of two antibacterial active pharmacophores with different modes of action (MOAs) in a single molecule using rational design has emerged as a viable strategy9,10, but identifying a dual-function molecule by screening approaches has not been reported thus far. Martin et al. identified the molecule SCH-79797, which was previously described to function as a human protease-activated receptor 1 antagonist, in a screen of a small compound library against an Escherichia coli mutant with a diminished outer membrane barrier2. This defect facilitates the penetration of large hydrophobic compounds through the lipopolysaccharide of the outer membrane of Gram-negative bacteria. In their preliminary in vitro susceptibility tests, they found antibacterial activity against Gram-positive organisms, Neisseria gonorrhoeae and some strains of E. coli and Acinetobacter baumannii.
In in vitro experiments aiming to define the frequency of spontaneous resistance of a methicillin-resistant Staphylococcus aureus USA300 and A. baumannii strain, the authors were unable to identify stable SCH-79797 mutants after 5–30 serial passages. The absence of resistant mutants makes it more difficult to characterize the MOA of a new antibiotic. Instead of analysing resistant mutants, Martin et al. turned to a systems biology approach that combined quantitative imaging and proteomic, genetic, metabolomic and cell-based assays. They used multiple lines of evidence to support a mechanism by which two active moieties in SCH-79797 simultaneously target folate metabolism and membrane integrity. Preliminary in vitro and in vivo (Galleria mellonella) studies indicated activity of SCH-79797 against several important Gram-positive and some Gram-negative pathogens such as N. gonorrhoeae.
Although these preliminary data indicate the difficulty to generate resistant mutants, additional studies will be required to confirm those findings. It will be interesting to see if the anticipated benefit of two MOAs on one chemical scaffold translates to relevant clinical benefit. Characterisation of antibacterial activity and potency in vitro and in appropriate animal models, PK and PK/PD metrics and testing for toxicity in animal models are basic next steps. Developing an antibiotic for the critical priority Gram-negative pathogens is an extremely difficult feat to achieve. But finding and characterizing two distinct modalities in a single compound is an important scientific step that may provide the basis for a new, dual-function class of antibiotics.
Wang, K. K. et al. Mol. Biol. Evol. 33, 492–500 (2016).
Martin, J. K. II et al. Cell 181, 1518–1532 (2020).
Bremner, J. B., Ambrus, J. I. & Samosorn, S. Curr. Med. Chem. 14, 1459–1477 (2007).
Jubeh, B., Breijyeh, Z. & Karaman, R. Molecules 25, 1543 (2020).
Page, M. G. P. Clin. Infect. Dis. 69 (Suppl. 7), S529–S537 (2019).
Gupta, V. & Datta, P. Indian J. Med. Res. 149, 97–106 (2019).
Liu, L. et al. Molecules 24, 1641 (2019).
Ma, Z. & Lynch, A. S. J. Med. Chem. 59, 6645–6657 (2016).
Domalaon, R., Idowu, T., Zhanel, G. G. & Schweizer, F. Clin. Microbiol. Rev. 31, e00077-17 (2018).
Parkes, A. L. & Yule, I. A. Expert Opin. Drug Discov. 11, 665–680 (2016).
The author declares no competing interests.
About this article
Cite this article
Theuretzbacher, U. Dual-mechanism antibiotics. Nat Microbiol 5, 984–985 (2020). https://doi.org/10.1038/s41564-020-0767-0