CUO NANOPARTICLES: MEROPENEM POTENTIATION AND EFFLUX-ASSOCIATED RESPONSES IN KLEBSIELLA PNEUMONIAE AND PSEUDOMONAS AERUGINOS

Background: Efflux systems and carbapenemases contribute to antibiotic resistance in Klebsiella pneumoniae and Pseudomonas aeruginosa. Sub-inhibitory CuO nanoparticles may modify efflux-associated responses and antibiotic susceptibility. Objectives: To assess antibiotic potentiation and efflux-associated responses to sub-inhibitory CuO nanoparticles in clinical isolates of both species, and to compare the effects of the antibiotic partner and bacterial species. Methods: One hundred twenty clinical isolates (60 per species) underwent susceptibility, carbapenemase, EtBr retention, transcriptional, biofilm and cytotoxicity analyses. Mechanistic CuO exposure was 0.25 × the isolate-specific MIC. Antibiotic interaction analyses used the recorded point-specific combination endpoints and separate 4 × 4 response matrices, with explicit FICI, Bliss and highest-single-agent (HSA) calculations. Correlations were assessed for the pooled cohort and separately within each species. Twenty-four-hour additional killing was recalculated from isolate-level log10 CFU/mL endpoints. Continuous outcomes are median (IQR), unless specified as means; categorical outcomes are n/N (%). Results: Meropenem, ciprofloxacin and CuO minimum inhibitory concentrations did not differ significantly between species (P = 0.920, 0.409 and 0.545, respectively). blaKPC and blaOXA-48 occurred only in K. pneumoniae, in 26/60 (43.3%) and 33/60 (55.0%) isolates, respectively, and blaVIM occurred only in P. aeruginosa, in 30/60 (50.0%) isolates (all P < 0.001). Recorded point-specific FICI values met the synergy threshold with meropenem in 56/60 (93.3%) K. pneumoniae and 59/60 (98.3%) P. aeruginosa isolates, compared with 39/60 (65.0%) and 41/60 (68.3%), respectively, with ciprofloxacin (pooled exact McNemar P < 0.001). The median FICI for both species was 0.375 (IQR 0.375–0.500) for meropenem. Median ethidium bromide accumulation increases were 242.2% (IQR 176.6–300.0%) and 265.7% (IQR 218.4–328.7%) in K. pneumoniae and P. aeruginosa, respectively. All eight efflux-associated targets had lower cohort median expression; gene-specific median fold changes ranged from 0.331 to 0.431 (Holm-adjusted P < 0.001 for each target), with discordant individual isolates. Median biofilm inhibition was 66.92% (IQR 61.39–72.04%) and 73.62% (IQR 67.36–77.99%), respectively. The meropenem combination resulted in median additional reductions at 24 h relative to the more active single agent of 2.306 (IQR 2.286–2.335) and 2.295 (IQR 2.270–2.328) log10 CFU/mL, respectively. Median curve-specific CC50 estimates were 365.96 (IQR 186.48–476.22) and 383.22 (IQR 215.20–492.90) µg/mL, with median MIC-normalised CC50/(0.25 × CuO MIC) ratios of 6.30 (IQR 5.57–7.09) and 6.18 (IQR 5.46–7.17), respectively. Mean cell viability was 94% at 0.25 × MIC, and 119/120 (99.2%) isolates had a ratio above 1. Within-species EtBr–gene-expression correlations were −0.890 and −0.900, and EtBr–biofilm-inhibition correlations were 0.861 and 0.947, respectively (all Holm-adjusted P < 0.001). Conclusions: More recorded meropenem FICI values met the synergy threshold than ciprofloxacin values, while median additional killing at 24 h was similar. The recorded FICI points were not independently validated as whole-checkerboard classifications. Associations of EtBr retention with gene expression and biofilm inhibition remained strong within both species. Meropenem FICI associations were significant within K. pneumoniae but not P. aeruginosa; ciprofloxacin associations had the opposite direction in the pooled heatmap. These findings do not establish a uniform efflux-mediated mechanism. MIC-normalised cytotoxicity ratios do not establish a clinical safety margin.