Adjunctive S100A8/A9 Immunomodulation Hinders Ciprofloxacin Resistance in Pseudomonas aeruginosa in a Murine Biofilm Wound Model

Abstract

Objective: Pseudomonas aeruginosa is known to contribute to the pathogenesis of chronic wounds by biofilm-establishment with increased tolerance to host response and antibiotics. The neutrophil-factor S100A8/A9 has a promising adjuvant effect when combined with ciprofloxacin, measured by quantitative bacteriology, and increased anti- and lowered pro-inflammatory proteins. We speculated whether a S100A8/A9 supplement could prevent ciprofloxacin resistance in infected wounds.

Method: Full-thickness 2.9cm²-necrosis was inflicted on 32 mice. On day 4, P.aeruginosa in seaweed alginate was injected sub-eschar to mimic a mono-pathogenic biofilm. Mice were randomized to receive ciprofloxacin and S100A8/A9 (n=14), ciprofloxacin (n=12) or saline (n=6). Half of the mice in each group were euthanized day 6 and the remaining day 10 post-infection. Mice were treated until sacrifice. Primary endpoint was the appearance of ciprofloxacin resistant P.aeruginosa. The study was further evaluated by genetic characterization of resistance, means of quantitative bacteriology, wound-size and cytokine-production.

Results: Three mice receiving ciprofloxacin monotherapy developed resistance after 14 days. None of the mice receiving combination therapy changed resistance pattern. Sequencing of fluoroquinolone-resistance determining regions in the ciprofloxacin resistant isolates identified two high-resistant strains mutated in gyrA C248T (MIC>32µg/ml) and a gyr B mutation was found in the sample with low level resistance (MIC=3µg/ml). Bacterial densities in wounds were lower in the dual treated group compared to the placebo group on both termination days.

Conclusion: This study supports the ciprofloxacin augmenting effect and indicates a protective effect in terms of hindered ciprofloxacin resistance of adjuvant S100A8/A9 in P.aeruginosa biofilm infected chronic wounds.

Keywords: biofilm; chronic wounds; ciprofloxacin; pseudomonas; resistance development.

Figure 1

(A) Experimental setup. 32 BALB/C 12-week-old mice were applied a full thickness necrosis by burning under analgesia/anesthesia and infected with pseudomonas aeruginosa embedded in alginate to mimic a biofilm. The mice were randomized into three treatment groups: 1) Ciprofloxacin and S100A8/A9 (n=14). 2) Ciprofloxacin monotherapy and placebo saline (n=12). 3) Saline controls (n=6). Half of the mice in the three groups were terminated day 10 post-burn (day 6 of infection) and the remaining on day 14 (day 10 of infection). All mice were treated until their sacrifice. (B) Wounds were photographed during and after sacrifice for wound size evaluation. Wounds were collected and homogenized for further quantitative bacteriology, host factor measurements by means of Luminex and resistance testing. Bacteria from three mice wounds developed Pseudomonas resistance by Etest. These were further processed with pulsed field electrophoresis and whole genome sequencing (Created with BioRender.com).

Figure 2

Quantification of colony forming units (CFU). 32 female Balb/C mice were inflicted with a burn-wound and 4 days post-infliction sub-escharly infected with P. aeruginosa. Treatment with S100A8/A9 (sub-eschar), ciprofloxacin (s.c.) or both was applied once daily. Half of the animals from each group was sacrificed on day 10 post-burn (day 6 after infection) and the other half on day 14 post-burn (day 10 after infection). Wounds were removed in toto, homogenized and quantitative bacteriology was determined. The lowest bacterial count was detected in the dual treated group on day 14 [median 95%CI: 1.84·10⁶ (9.30·10⁵; 8.10·10⁶)] and was statistically different from the level in the placebo group [1.70·10⁷(2.43·10⁶; 1.87·10⁷)]. The same difference was detected in mice sacrificed on day 10.

Figure 3

(A–C) IL-1β, TWEAK, and osteopontin. The wounds were harvested, homogenized and the supernatants were analyzed using a LUMINEX. Displayed are the individual measured cytokine concentrations in ρg/mL wound homogenate for every mouse) and median 95%CI. Statistically significant P values are displayed on the figure. (A) The pro-inflammatory IL-1β responses in the treated groups were dampened after 14 days (dual therapy: median 95%CI: 5.27·10³ (3.70·10³; 8.92·10³). Monotherapy: 6.53·10³ (3.82·10³; 9.56·10³). Placebo: 1.25·10⁴ (1.13·10⁴; 1.34·10⁴). (B) The TWEAK level was only statistically different from the placebo group on day 14 (dual therapy: 26.69 (18.65;43.51), monotherapy: 28.75(15.64 ;35.48) and placebo 56.13(54.26;64.18). (C) The level of OPN in the three groups was compared on both sacrifice-dates. Osteopontin was significantly higher in the ciprofloxacin monotherapy group on day 14 compared to the dual-treated group on day 14 [2.78·10⁴(1.90·10⁴;3.75·10⁴) and 1.50·10⁴(9.55·10³;2.76·10⁴)]. Both treatment groups had higher levels of OPN compared to the placebo group [6.18·10³(5.69·10³;8.62·10⁴)] on day 14.

Figure 4

(A–C) Immunomodulating S100A8 and S100A9and KC/CXCL-1. Illustrated are the Luminex measurements (median 95%CI) of mice inflicted with pseudomonas infected skin necrosis randomized to sacrifice on day 10 or 14 and to different treatment regimes. Statistically significant P values are displayed on the figure. (A) The level of S100A8 increased from day 10 to day 14 in the dual treated group [respectively 5.57·10⁶(4.66·10⁶; 8.02·10⁶) and 9.51·10⁶ (6.76·10⁶;1.18·10⁷)]. (B) The concentration of the alarmin S100A9 was higher in the monotherapy group (1.77·10⁴ [1.40·10⁴; 4.58·10⁴)] compared to the dual treated group (1.43·10³(9.75·10²; 1.60·10⁴) and both treatment groups, on day 14, differed from the level in the placebo mice. (C) Both treatment groups had higher levels of KC/CXCL-1 on day 14, compared to the group receiving only saline [dual treated group 1.55·10⁴(9.35·10³;1.98·10⁴)and monotherapy group 2.07·10⁴(1.13·10⁴; 3.21·10⁴)].