Antibiotic resistance among Gram-negative pathogens has pushed polymyxins — drugs of last resort — into frontline use, yet resistance to polymyxins themselves is now emerging. Finding compounds that restore polymyxin's killing power without adding new toxicity could extend the lifespan of this critical antibiotic class by years, buying time while novel agents are developed.
High-throughput screening of a chemical library identified PA74, an aryloxyphenol structurally related to the antimicrobial triclosan, as a potent polymyxin adjuvant. In checkerboard assays, PA74 restored polymyxin B activity against clinical isolates of polymyxin-resistant Acinetobacter baumannii, Klebsiella pneumoniae, and Escherichia coli — three of the most dangerous drug-resistant pathogens on WHO priority lists. Beyond polymyxins, PA74 also enhanced the activity of tetracycline and chloramphenicol, suggesting a broad membrane-disruption or efflux-related mechanism. Functional studies using FabI-overexpressing strains implicated the bacterial fatty acid synthesis enzyme FabI as a partial mechanistic target; elevated FabI abundance attenuated synergy, mirroring known triclosan biology. Crucially, PA74 showed far lower resistance-selection pressure than triclosan over 15 days of serial passage, and cytotoxicity in mammalian cell lines was minimal. In vivo, PA74 co-administration improved outcomes in a murine skin infection model.
The triclosan structural lineage is scientifically interesting but also contextually loaded — triclosan's widespread consumer use contributed directly to FabI-mediated resistance and cross-resistance concerns. PA74's modified 4-amino-3-chlorophenol moiety appears to partially decouple FabI inhibition from its adjuvant activity, though the precise mechanism remains incompletely resolved. As an adjuvant rather than a standalone antibiotic, PA74 fits an increasingly attractive development paradigm: repotentiating approved drugs sidesteps the lengthy clinical pipeline for new antibiotics. Key limitations include the murine skin infection model's limited translatability to systemic infections, the absence of pharmacokinetic data, and single-laboratory origin. This is early-stage but mechanistically coherent work with a credible therapeutic rationale.