Fig 1.
Diclofenac in combination with colistin inhibits A. baumannii growth.
Representative growth curves of ARC6851 in LB containing 256 μg/ml colistin and either the solvent control DMSO or 100 μM phenylacetic acid, 4-fluorophenylacitic acid, 4-hydroxyphenylacetic acid, mandelic acid, ibufenac, ibuprofen, or diclofenac. Data is representative of three biological replicates. ****P<0.0001 (One-way ANOVA with Tukey’s test for multiple comparisons, results were compared to the control group treated with DMSO).
Fig 2.
Diclofenac inhibits A. baumannii and K. pneumoniae growth in the presence of colistin.
(A) Representative growth curves of A. baumannii UPAB1, AB431, ARC6851 and AB347 strains in LB containing DMSO or 100 μM diclofenac (DF) (left panel) and 3 μg/ml, 16 μg/ml, 256 μg/ml, or 7 μg/ml colistin respectively (right panel). (B) Representative growth curves of K. pneumoniae AR0046, KR49, AR0125, AR0126 strains in LB containing DMSO or 100 μM diclofenac (DF) (left panel) and 1 μg/ml, 16 μg/ml, 32 μg/ml, or 2 μg/ml colistin respectively (right panel). Data is representative of three biological replicates. ****P<0.0001, unpaired t tests at 16 h for Col + DF 100 compared to DMSO control. Col (colistin), DF (diclofenac).
Fig 3.
Colistin, but not diclofenac, increases membrane permeability of ARC6851.
(A) Membrane permeability was assessed by measuring uptake of Hoescht 33342 (H33342) fluorescent dye. Bacterial cells were suspended in PBS supplemented with DMSO, 100 μM diclofenac (DF) or increasing concentrations of colistin (Col 1 μg/ml, 8 μg/ml, 16 μg/ml, 32 μg/ml, 64 μg/ml, or 128 μg/ml). Relative fluorescent units (RFU) were monitored over the course of 2 hours. (B) Representative growth curves of ARC6851 in LB containing 32 μg/ml Vancomycin or 0.01% SDS + 0.075 mM EDTA in combination with the solvent control DMSO, 100 μM Diclofenac or 32 μg/ml Vancomycin. Data is representative of three biological replicates, ****P<0.0001, (One-way ANOVA with Tukey’s test for multiple comparisons, results were compared to the control group treated with DMSO). Col (colistin), DF (diclofenac), Van (Vancomycin), SDS (sodium dodecyl sulfate), EDTA (ethylenediaminetetraacetic acid).
Fig 4.
Colistin and diclofenac inhibit growth of ARC6851 in a synergistic manner.
(A) Checkerboard microdilution assay for ARC6851 with colistin and diclofenac. Blue color represents cell density. (B) Growth curves of ARC6851 in LB containing 256 μg/ml colistin with either DMSO or increasing concentrations of diclofenac (25 μM, 40 μM, 50 μM, 60 μM, 75 μM, 90 μM, or 100 μM) (C) Growth curves of ARC6851 in LB containing 100 μM diclofenac with either DMSO or increasing concentrations of colistin (1 μg/ml, 8 μg/ml, 16 μg/ml, 32 μg/ml, 64 μg/ml, 128 μg/ml, or 256 μg/ml). Data is representative of three biological replicates ***P<0.001, ****P<0.0001 (One-way ANOVA with Tukey’s test for multiple comparisons, results were compared to the control group treated with DMSO). Col (colistin), DF (diclofenac).
Fig 5.
Oxidative stress response and type IV pili are differentially regulated in ARC6851 under colistin and diclofenac combination treatment.
(A) Number of differentially regulated genes in ARC6851 when treated with 1 μg/ml colistin, 100 μM diclofenac, or the combination. (B) Volcano plot showing differentially expressed genes in ARC6851 with colistin and diclofenac combination treatment vs DMSO control. (C) Number of differentially regulated proteins in ARC6851 when treated with 1 μg/ml colistin, 100 μM diclofenac, or the combination. (D) Volcano plot showing differentially expressed proteins in ARC6851 in colistin plus diclofenac combination treatment vs DMSO control. Col (colistin), DF (diclofenac). Five biological replicates were prepared for each condition.
Fig 6.
Diclofenac and colistin–dependent bacterial killing is mediated by ROS.
(A) ARC6851 was grown to mid-exponential phase in LB + DMSO, LB+ colistin (1 μg/ml) + DMSO, LB + diclofenac (100 μM), or LB + colistin (1 μg/ml) + diclofenac (100 μM), and treated with 0 mM or 5 mM H2O2 for 2 h. Survival was measured by serial dilution and quantification of recoverable CFU/mL. Symbols represent five biological replicates; the horizontal line represents the Median. (B) Representative growth curves of ARC6851 in LB containing 256 μg/ml colistin, 100 μM diclofenac and increasing concentrations of DMSO (0.25%, 0.5% and 1%). Data is representative of three biological replicates **P<0.01, ****P< 0.0001; One-way ANOVA, Tukey’s test for multiple comparisons. Col (colistin), DF (diclofenac).
Fig 7.
ARC6851 is attenuated in an acute murine pneumonia model under colistin and diclofenac combination treatment.
(A) C57BL/6 mice were infected with ~5 × 107 CFU of mid-exponential ARC6851. At 4 hpi, mice were IP injected with 1.25 mg/kg diclofenac. At 4, 12, and 20 hpi, mice were IP injected with 2.5 mg/kg colistin. At 24 h post-infection, lungs, kidneys, and spleens were harvested, and bacterial load was determined. Each symbol represents an individual mouse, and the horizontal bar represents the Mean with SEM. Data collected from two independent experiments. (B) Survival rate of C57BL/6 mice (n = 10 per group) following infection with ARC6851 for the different therapies shown in A. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, Kruskal-Wallis test. Col (colistin), DF (diclofenac).
Fig 8.
ARC6851 ΔpilA is sensitive to colistin treatment in an acute murine pneumonia model.
C57BL/6 mice were infected with ~5 × 107 CFU of mid-exponential ARC6851 ΔpilA. At 4 hpi, mice were IP injected with 1.25 mg/kg diclofenac or PBS. At 4, 12, and 20 hpi, mice were IP injected with 2.5 mg/kg colistin or PBS. At 24 h post-infection, lungs, kidneys, and spleens were harvested, and bacterial load was determined. Each symbol represents an individual mouse, and the horizontal bar represents the Mean with SEM. Data collected from two independent experiments. ****P<0.0001, Kruskal-Wallis test. Col (colistin), DF (diclofenac).
Fig 9.
Current model for the mechanism of colistin and diclofenac.
Created in Biorender.