Pseudomonas aeruginosa exoproducts determine antibiotic efficacy against Staphylococcus aureus

Abstract

Chronic coinfections of Staphylococcus aureus and Pseudomonas aeruginosa frequently fail to respond to antibiotic treatment, leading to significant patient morbidity and mortality. Currently, the impact of interspecies interaction on S. aureus antibiotic susceptibility remains poorly understood. In this study, we utilize a panel of P. aeruginosa burn wound and cystic fibrosis (CF) lung isolates to demonstrate that P. aeruginosa alters S. aureus susceptibility to bactericidal antibiotics in a variable, strain-dependent manner and further identify 3 independent interactions responsible for antagonizing or potentiating antibiotic activity against S. aureus. We find that P. aeruginosa LasA endopeptidase potentiates lysis of S. aureus by vancomycin, rhamnolipids facilitate proton-motive force-independent tobramycin uptake, and 2-heptyl-4-hydroxyquinoline N-oxide (HQNO) induces multidrug tolerance in S. aureus through respiratory inhibition and reduction of cellular ATP. We find that the production of each of these factors varies between clinical isolates and corresponds to the capacity of each isolate to alter S. aureus antibiotic susceptibility. Furthermore, we demonstrate that vancomycin treatment of a S. aureus mouse burn infection is potentiated by the presence of a LasA-producing P. aeruginosa population. These findings demonstrate that antibiotic susceptibility is complex and dependent not only upon the genotype of the pathogen being targeted, but also on interactions with other microorganisms in the infection environment. Consideration of these interactions will improve the treatment of polymicrobial infections.

Fig 1. P. aeruginosa supernatant alters S. aureus antibiotic susceptibility.

S. aureus strain HG003 was grown to mid-exponential phase and exposed to sterile supernatants from S. aureus HG003 (red), P. aeruginosa laboratory strains PAO1 and PA14 (grey), P. aeruginosa CF clinical isolates (blue) or P. aeruginosa burn isolates (green) for 30 min prior to addition of (A) 50 μg/ml vancomycin, (B) 58 μg/ml tobramycin or (C) 2.34 μg/ml ciprofloxacin concentrations similar to the Cmax in humans. An aliquot was removed after 24 h, washed, and plated to enumerate survivors. The dotted red line represents the number of survivors in the control culture. All experiments were performed in biological triplicate and the number of survivors following antibiotic challenge in the presence of P. aeruginosa supernatant was compared to the HG003 supernatant-treated control. Underlying data can be found in S1 Data. *p<0.05 (one-way ANOVA with Tukey’s multiple comparisons post-test analysis of surviving CFU). Error bars represent mean + sd. CF, cystic fibrosis; CFU, colony-forming units.

Fig 2. P. aeruginosa rhamnolipids potentiate aminoglycoside uptake and cell death in S. aureus.

S. aureus HG003 was grown to mid-exponential phase and exposed to (A) sterile supernatants from P. aeruginosa or S. aureus or exogenous addition of rhamnolipids (30 μg/ml) before addition of tobramycin 58 μg/ml. At indicated times, an aliquot was washed and plated to enumerate survivors. (B) Texas Red-conjugated tobramycin was added to S. aureus cultures with or without 30 μg/ml rhamnolipids. Following 1 h, Texas Red-tobramycin uptake was measured by flow cytometry. (C) Rhamnolipid production present in the supernatant of P. aeruginosa PAO1, PA14, PA14 ΔrhlA, CF isolates (blue) or burn isolates (green) were quantified by a drop-collapse assay. Experiments were performed in biological triplicate. Underlying data can be found in S1 Data. Error bars represent mean ± sd. CF, cystic fibrosis.

Fig 3. P. aeruginosa secondary metabolites inhibit S. aureus aerobic respiration resulting in a drop in intracellular ATP and protection from ciprofloxacin killing.

(A) S. aureus strain HG003 harboring plasmid PpflB∷gfp was grown to mid-exponential phase and treated with supernatant from P. aeruginosa PAO1, PA14, CF isolates (blue) or burn isolates (green), for 30 min. OD₆₀₀ and gfp expression levels were determined after 16 h using a Biotek Synergy H1 microplate reader. (B) Intracellular ATP was measured after 1.5 h incubation with supernatant. ***p < 0.0005 (one-way ANOVA with Tukey’s multiple comparison post-test). (C) S. aureus strain HG003 was grown to mid-exponential phase in MHB media and pre-treated with sterile supernatants from P. aeruginosa strains PA14 wild-type or its isogenic mutants or (D) physiologically-relevant concentrations of HQNO, PYO, or NaCN for 30 min prior to antibiotic challenge [26,27]. At indicated times, an aliquot was washed and plated to enumerate survivors. All experiments were performed in biological triplicate. Underlying data can be found in S1 Data. Error bars represent mean ± sd. CF, cystic fibrosis; CFU, colony-forming units; GFP, green fluorescent protein; HQNO, 4-hydroxyquinoline N-oxide; MHB, Mueller-Hinton broth; NaCN, sodium cyanide; OD, optical density; PYO, pyocyanin.

Fig 4. P. aeruginosa supernatant potentiates killing by vancomycin via the LasA endopeptidase.

S. aureus HG003 was grown to mid-exponential phase and exposed to sterile supernatants for 30 min prior to addition of vancomycin 50 μg/ml. Where indicated, PAO1 supernatant was heat inactivated at 95°C for 10 min. (A) At indicated times, an aliquot was removed, washed, and plated to enumerate survivors or (B) 100 μl cells were added to a 96-well plate and lysis was measured at OD₆₀₀ every hour for 16 h. (C) LasA present in the supernatant of P. aeruginosa PAO1, PA14, CF isolates (blue) or burn isolates (green) was quantified by western blot and the ability of each supernatant to lyse heat-killed S. aureus HG003 cells after 2 h. All experiments were performed in biological triplicate. Underlying data can be found in S1 Data. Error bars represent mean ± sd. CF, cystic fibrosis.

Fig 5. P. aeruginosa potentiates vancomycin killing of S. aureus in a murine model of coinfection.

Approximately 1 x 10⁵ CFU S. aureus strain HG003 was administered subcutaneously alone or in combination with approximately 1 x 10³ CFU P. aeruginosa PAO1 or PAO1 lasA∷tn 24 h after burn. Mice were left untreated or administered 110 mg/kg vancomycin subcutaneously once daily for 2 d. Mice were sacrificed 48 h post infection. (A) Tissue biopsies at the site of infection were harvested and homogenized and S. aureus burdens were enumerated on selective media. Data for each group are compiled from 2 independent experiments. (n = 6–10 mice per group) *p < 0.05, ***p < 0.005 (Mann-Whitney test). (B) Relative percentage survival for HG003 in each condition was calculated by dividing the CFU/g tissue of mice treated with vancomycin by the average CFU/g tissue of untreated mice. Maximum percentage survival is 100%. Data for each group are compiled from 2 independent experiments. (C) Expression of lasA in tissue from mono- (PAO1 alone) and coinfected (PAO1/HG003) mice relative to the starting inoculum measured by qRT-PCR. Underlying data can be found in S1 Data. *p < 0.05 (Student t test). Error bars represent mean + sd. CFU, colony-forming units; ctr, control; qRT-PCR, quantitative reverse transcription PCR; vanc, vancomycin; WT, wild-type.

Fig 6. P. aeruginosa-mediated alteration of S. aureus antibiotic susceptibility.

P. aeruginosa exoproducts PYO, HQNO, and HCN inhibit S. aureus electron transport, leading to collapse of PMF and inhibition of the F₁F₀ ATPase leading to a decrease in S. aureus antibiotic susceptibility. Conversely, P. aeruginosa RLs intercalate into the plasma membrane-forming pores that permit aminoglycoside entry into the cell in a PMF-independent manner, while P. aeruginosa endopeptidase LasA cleaves pentaglycine crosslinks between peptidoglycan molecules of the cell wall, increasing vancomycin-mediated lysis of S. aureus. HCN, hydrogen cyanide; HQNO, 2-heptyl-4-hydroxyquinoline N-oxide; NAG, N-acetylglucosamine; NAM, N-acetylmuramic acid; PMF, proton-motive force; PYO, pyocyanin; RL, rhamnolipids.