Synergistic effect of membrane-active peptides polymyxin B and gramicidin S on multidrug-resistant strains and biofilms of Pseudomonas aeruginosa

Abstract

Multidrug-resistant Pseudomonas aeruginosa is a major cause of severe hospital-acquired infections. Currently, polymyxin B (PMB) is a last-resort antibiotic for the treatment of infections caused by Gram-negative bacteria, despite its undesirable side effects. The delivery of drug combinations has been shown to reduce the required therapeutic doses of antibacterial agents and thereby their toxicity if a synergistic effect is present. In this study, we investigated the synergy between two cyclic antimicrobial peptides, PMB and gramicidin S (GS), against different P. aeruginosa isolates, using a quantitative checkerboard assay with resazurin as a growth indicator. Among the 28 strains that we studied, 20 strains showed a distinct synergistic effect, represented by a fractional inhibitory concentration index (FICI) of ≤0.5. Remarkably, several clinical P. aeruginosa isolates that grew as small-colony variants revealed a nonsynergistic effect, as indicated by FICIs between >0.5 and ≤0.70. In addition to inhibiting the growth of planktonic bacteria, the peptide combinations significantly decreased static biofilm growth compared with treatment with the individual peptides. There was also a faster and more prolonged effect when the combination of PMB and GS was used compared with single-peptide treatments on the metabolic activity of pregrown biofilms. The results of the present study define a synergistic interaction between two cyclic membrane-active peptides toward 17 multidrug-resistant P. aeruginosa and biofilms of P. aeruginosa strain PAO1. Thus, the application of PMB and GS in combination is a promising option for a topical medication and in the prevention of acute and chronic infections caused by multidrug-resistant or biofilm-forming P. aeruginosa.

FIG 1

Primary peptide structures of lipopeptide polymyxin B (A) and cyclic decapeptide gramicidin S (B).

FIG 2

Evaluation of a checkerboard assay in the 96-well microtiter plate. The two rightmost columns contain the positive growth control (pc) and negative growth control (nc) to confirm bacterial growth without peptides and medium sterility, respectively. The green and blue circles show the peptide MIC_a. The red circle indicates the antibiotic combination that resulted in a synergistic effect on P. aeruginosa strain 55, for which the FICI calculation should be [1/4 + 4/32] = 0.375.

FIG 3

The mean FICIs for 28 different strains and clinical isolates of P. aeruginosa (Pa), with standard deviations calculated from the results from at least five independent experiments. The first 11 strains showed an FICI of <0.5 (A), and the next 9 strains revealed an FICI of 0.5 that indicated a distinct synergistic effect of the PMB-GS combination toward these 20 strains. (C) FICIs in this group of P. aeruginosa clinical isolates are >0.5 and showed a nonsynergistic effect. The standard deviations of the FICIs for the strains P. aeruginosa 56, 59, 913, 987, 5517, 5522, and 5524 as high as 0.2 indicated the high physiological variability within these strains.

FIG 4

The inhibitory effect of PMB or GS alone and in combination on the formation of P. aeruginosa PAO1 biofilms. Shown are the effect of PMB alone (A), the effect of GS alone (B), and the synergistic effect of PMB and GS in combination (C). Biofilms were quantified by crystal violet staining, and the absorbance at 595 nm was measured. The mean values obtained for negative controls were subtracted from the results for the test wells. Student's t test P values (*, P < 0.05; **, P < 0.01) were calculated for the average values from two independent experiments, each performed in triplicate.

FIG 5

PMB-and-GS combination treatment of pregrown P. aeruginosa PAO1 biofilms in comparison to treatment with these peptides alone. P. aeruginosa PAO1 biofilms were grown for 24 h in 96-well plates following the addition of PMB (8 μg/ml; closed squares), GS (32 μg/ml; closed triangles), or a combination of both peptides (open circles) and 0.1 mM resazurin, a redox indicator. Control wells (closed circles) contained 1% ethanol (vol/vol), because peptide stock solutions were prepared in 50% ethanol, resulting in final ethanol concentrations of 1% (vol/vol). At the indicated time points, resazurin reduction was quantified by measuring the absorbance at 560 nm. The figure shows means and standard deviations for the results of three independent experiments, each performed in triplicate.