Nitrofurans: Revival of an “old” drug class in the fight against antibiotic resistance

Antibiotic resistance is one of the greatest contemporary threats to the human health, which has increasingly been undermining the effectiveness of existing antimicrobial therapies. Development of novel antibiotics is inarguably the key to combat this threat, yet this is a lengthy process (10 to 15 years) at a hefty price tag of approximately US$1.3 billion for development of an approved drug [1,2]. A new drug, once introduced into the market, also faces the risk of drug resistance emergence. It is, therefore, essential to diversify therapeutic strategies, including revival and reintroduction of “old” antibacterials for treating multidrug-resistant pathogens [3]. Nitrofuran class of synthetic molecules, introduced in the 1940s and 1950s, belongs to this category [4]. Several nitrofurans are currently on the market: nitrofurazone for topical infections and urinary catheter coating, nitrofurantoin for urinary tract infections, and furazolidone for bacterial diarrhea and Helicobacter pylori infections. Here, we highlight aspects of this drug class that have recently been unraveled, laying foundation for future improvements and judicial uses of nitrofurans against ever-expanding antibiotic-resistant bacteria.

to covalently modify the targets, while the side chain at the C 2 of the furan ring defines its selectivity, depending on how well it fits into the ligand-binding pocket of a target. Future experiments looking into the interaction between nitrofurans and individual E. coli essential proteins (approximately 300 in total) in the presence and absence of a nitroreductase may shed light on the nitrofuran cognate targets.
Existence of potential unknown nitrofuran-activating enzymes in E. coli presents another interesting question for this antibiotic class. The observations that nitrofurans still retain antibacterial activity against the triple nfsA nfsB ahpF mutant, either individually or in a synergistic manner with sodium deoxycholate, an antimicrobial bile salt [9,13], point to the existence of these enzymes, although contribution of these enzymes to the overall antibacterial activity of the nitrofurans (nitrofurantoin and furazolidone) assayed in these studies may not be remarkable. The modest effect on nitrofuran minimum inhibitory concentration (MIC) of these unknown enzymes could be due to low protein expression or low affinity for the nitrofurans tested. Identification of these nitrofuran-activating enzymes in E. coli and understanding their biology (physiological functions, regulation of protein expression/action, and enzyme-nitrofuran molecular interaction) would provide useful knowledge to facilitate development of new nitrofuran-based antibacterial therapies. For example, in the case of potential low expression of the activating enzyme, structure-based virtual screens with an ultra-large library [14] targeting the expression regulator of the candidate enzyme or targeting the enzyme allosteric site can be employed to seek the adjuvant molecules that up-regulate the enzyme expression or enhance the enzyme activity, respectively; these adjuvant molecules can be used with nitrofuran antibiotics for improved antibacterial potency. Alternatively, rational design of nitrofuran analogs such that they favorably bind to the redox active site of the activation enzyme through in silico enzyme-drug docking simulation, followed by experimental validation, may generate novel nitrofuran candidates that are different from clinically used nitrofuran drugs in the terms of prodrug activation and resistance mechanisms. A similar strategy can be used to identify novel nitrofurans that are effective against bacteria outside of the spectrum of action of current nitrofurans, e.g., Pseudomonas aeruginosa.

Nitrofuran resistance mechanisms in E. coli
Mutations of nfsA and nfsB are the major bacterial mechanism for gaining resistance to nitrofurans, in both laboratory and clinical strains of E. coli. There is one exception in which a de novo-selected nitrofurantoin-resistant E. coli strain had the wild-type nfsA and nfsB alleles but contained an in-frame deletion in the ribE gene that encodes an enzyme in the biosynthesis of flavin mononucleotide, an essential NfsA/NfsB cofactor ( Fig 1B) [15]. Perhaps due to the reduced bacterial growth rate caused by the ribE mutation, mutations in this gene have not been reported in E. coli clinical isolates and are unlikely to pose a significant threat to the efficacy of nitrofuran treatment.
The inability to prepare for the unknown often leads to disastrous consequences. In this spirit, a possible novel nitrofuran resistance mechanism(s), besides well-known nfsA/nfsB mutations, must receive due attention and resources in order to sustain the utility of this drug class. While the prevalence of nitrofuran resistance among E. coli clinical isolates in recent epidemiology surveys around the world is still low [9], nitrofuran-hyperresistant isolates with the MIC higher than 128 μg/mL have been encountered [15,16]. This high level of resistance cannot solely be explained by mutations in the nfsA, nfsB, and ahpF genes and points to unknown resistance determinants that are already circulating in pathogenic strains.
Of great concern is the reported nitrofurantoin resistance in E. coli uropathogenic clinical isolates in one hospital in North Wales (United Kingdom) in 2020 [17]. These isolates were found to have a mutated version of the extended spectrum β-lactamase CTX-M-14, which differs from the wild-type protein by 3 nonsynonymous changes (T55A, A273P, and R277C). When recombinantly overexpressed in an E. coli laboratory strain, the mutated CTX-M-14 enzyme caused hyperresistance to nitrofurantoin while retaining the ability to render β-lactam resistance [17]. The purified enzyme was able to hydrolyze nitrofurantoin in the in vitro enzymatic assay (Fig 1C). Although the exact hydrolytic products remain to be determined, we speculate that the triple CTX-M-14 mutant, by analogy to β-lactam drugs, may cut at the amide bonds in the hydantoin ring (Fig 2A, red arrows) such that the hydrolytic products have lower affinity to the activation enzymes or to the essential targets of nitrofurantoin. While the work is preliminary, this finding is very worrisome. The recent switch from trimethoprim/sulfamethoxazole combination to nitrofurantoin as the first-line therapy for urinary tract infections has majorly increased the exposure of E. coli to nitrofurantoin and will very likely select for this triple mutant of CTX-M-14. Threat of the CTX-M-14-mediated nitrofuran-resistant mutant is high, given that this is one of the 2 most predominant types of the extended spectrum β-lactamases globally [18]. Further investigation is urgently needed to understand the prevalence and expression of this CTX-M-14 variant that causes resistance to nitrofurantoin, the mechanism of nitrofurantoin hydrolysis by this variant and probably other β-lactamases, the potential of known β-lactamase inhibitors to block the hydrolyzing activity against nitrofurans. Additionally, in vitro evolution experiments with clinically important β-lactamase enzymes, directed toward nitrofurantoin hydrolysis and resistance, would predict what mutations may be selected under nitrofurantoin exposure in clinical settings. These efforts will help maintain the effectiveness of nitrofurantoin against the β-lactamase-positive E. coli.

New nitrofurans for the future of infectious disease therapies
There have been very few attempts to develop next-generation nitrofuran antibacterial drugs in several decades. Over the past few years, however, work on nitrofurans has been revived, in parallel with their resurgence as an effective treatment option in the context of widespread resistance to other antibiotics. One prominent trend is to search for nitrofuran analogs that have a broadened spectrum and increased potency in comparison to the existing clinically used nitrofurans, such as IITR06114, a novel nitrofuran recently discovered in a small molecule screen [19], and compound 17 identified in a hit-to-lead optimization effort (Fig 2A) [20]. Extensive medicinal chemistry efforts have been undertaken to design novel antimycobacterial agents from the nitrofuran scaffold, culminating in a number of candidates with the submicromolar to nanomolar in vitro MICs [21,22]. It remains to be seen whether these nitrofuran candidates will be successfully brought to clinical trials in the years to come. Development of a new nitroaromatic pharmacophore, 2-nitrothiophene-3-sulfonyl fluoride, is also worth noting (Fig 2B). This new molecule has a similar molecular structure to nitrofurans and shares the mechanism of nitroreductase-mediated activation, providing a promising starting point from which novel potent drugs can be developed, including those that may be effective against formidable Gram-negative pathogens, such as Acinetobacter baumannii and P. aeruginosa, which naturally have high nitrofuran MICs [23].
Given that development of new drugs takes years, it is also important to conserve and enhance the effectiveness of already available nitrofurans. This can be achieved by combinations with other antimicrobial agents, capitalizing on interactions between them, either via synergistic interactions or collateral sensitivity. A synergistic interaction means that the antibacterial potency of 2 antibiotics, when used in combination, is stronger than the combined effect of the individual antibiotic when used alone. Such an effect was, for example, reported in the 2-way combinations of nitrofurans with the secondary bile salt deoxycholate [13], and vancomycin [24], and further enhanced in the triple combination of nitrofurans, deoxycholate, and vancomycin [25]. Meanwhile, another beneficial interaction that can be exploited is collateral sensitivity, where a resistance mechanism to one antibiotic confers increased sensitivity to another. For example, loss-of-function mutation of the protease-encoding lon gene causing tigecycline resistance results in hypersensitivity to nitrofurantoin [26]. Collateral sensitivity leads to a strategy of switching treatment from tigecycline to nitrofurantoin in sequential therapies to eliminate the strains resistant to the former, while increasing the chance of a successful bacterial clearance with the latter [26].
In contrast to high potency against E. coli, nitrofurans have very limited inhibitory effect, if any, on growth of P. aeruginosa, an opportunistic Gram-negative pathogen that causes respiratory system infections and urinary tract and soft tissue infections. In a screen for quorum sensing inhibitors from a library of Food and Drug Administration (FDAAU : PleasenotethatFDAhasbeend )-approved drugs, nitrofurazone was found to inhibit PqsE, a regulator of P. aeruginosa quorum sensing system. Nitrofurazone suppressed pathways regulated by this protein, including biofilm formation and production of virulence factor pyocyanin [27]. This finding opens an avenue for the repurposing of nitrofurans as antivirulence drugs to attenuate the severity of P. aeruginosa infections [28].

Conclusions
The spread of antimicrobial resistance has led to a resurgence of nitrofurans, an old class of antibiotics. Recent studies have shed light on the mechanism of nitrofuran activation and raised urgent questions about emerging resistance mechanisms. Current efforts, such as search for improved nitrofuran derivatives, development of new pharmacophores, drug combinations, and repurposed uses of nitrofurans, hold a promise to make this drug class an important weapon in the combat against multidrug-resistant bacterial pathogens in the future.