Antibiofilm and antibacterial effects of specific chitosan molecules on Staphylococcus aureus isolates associated with bovine mastitis

Abstract

Staphylococcus aureus is one of the major pathogens causing bovine intramammary infections (IMIs) and mastitis. Mastitis is the primary cause for the use of antibiotics in dairy farms but therapeutic failure is often observed. One of the reasons for the lack of effectiveness of antibiotic therapy despite the observed susceptibility of bacterial isolates in vitro are bacterial biofilms. In this study, we used chitosan of well-defined molecular weight (0.4-0.6, 1.3, 2.6 and 4.0 kDa) and investigated their antibiofilm and antibacterial activities in in vitro and in vivo models related to S. aureus IMIs. A chitosan of at least 6 units of glucosamine was necessary for maximum antibacterial activity. The 2.6 and 4.0 kDa forms were able to prevent biofilm production by the biofilm hyperproducer strain S. aureus 2117 and a bovine MRSA (methicillin-resistant S. aureus). The intramammary administration of the 2.6 kDa chitosan showed no adverse effects in mice or in cows, as opposed to the slight inflammatory effect observed in mammary glands with the 4.0 kDa derivative. The 2.6 kDa chitosan killed bacteria embedded in pre-established biofilms in a dose-dependent manner with a >3 log10 reduction in CFU at 4 mg/ml. Also, the 2.6 kDa chitosan could prevent the persistence of the internalized MRSA into the mammary epithelial cell line MAC-T. An in vitro checkerboard assay showed that the 2.6 kDa chitosan produced a synergy with the macrolide class of antibiotics (e.g., tilmicosin) and reduced the MIC of both molecules by 2-8 times. Finally, the intramammary administration of the 2.6 kDa chitosan alone (P<0.01) or in combination with tilmicosin (P<0.0001) reduced the colonization of mammary glands in a murine IMI model. Our results suggest that the use of chitosan alone or in combination with a low dose of a macrolide could help reduce antibiotic use in dairy farms.

Fig 1. Antibiofilm activity of the different forms of chitosan against MRSA 1158c and the biofilm hyperproducer strain 2117.

(A) CHOS, (B) 1.3 kDa, (C) 2.6 kDa, and (D) 4.0 kDa chitosan. Bars represent the means and vertical lines the standard deviation (SD). Data were obtained from three independent experiments. Significant differences in comparison to the untreated control (0 mg/ml) are shown by asterisks. Statistical analysis was performed using Kruskal-Wallis test (non-parametric one way ANOVA) with Dunn’s multiple comparison test: ns, non significant; *, P<0.05.

Fig 2. Relative innocuity of different forms of chitosan in cows.

Each quarter of cow’s udder has received 500 mg of chitosan (2.6 kDa or 4.0 kDa) or saline as the negative control. Milk samples and somatic cell counts (SCC) were determined 12 and 3 hours before the instillation of chitosan. After the intramammary instillation, milk samples were aseptically collected from cows at several points in time to evaluate inflammation by determining the SCC (A) and milk yields (B). Symbols represent the means and vertical lines the standard deviation. Data were analyzed by ANOVA using the MIXED procedure of SAS (SAS Institute Inc., Cary, NC). For the first experiment, time was used as a repeated effect and treatment (cow) was used as the subject. Orthogonal contrasts were performed to compare the effect of each treatment to control. No difference was observed between saline and the 2.6 kDa chitosan (SCC and quarter milk yield). Significant differences were observed between saline and 4.0 kDa for SCC: *, P<0.05; **, P< 0.01; ***, P<0.0001.

Fig 3. Effect of the 2.6 kDa chitosan on LDH release from MAC-T bovine mammary epithelial cells.

Cells were exposed to the 2.6 kDa chitosan at different concentrations (0.5 mg/ml, 2 mg/ml and 8 mg/ml). Triton x-100 and culture medium were used as cytotoxic positive and negative (Ctrl) controls, respectively. The amount of formazan formed was normalized to the amount formed by the non-treated cells (Ctrl). Bars represent the means and vertical lines the standard deviation (SD). Significant differences in comparison to the negative control (Ctrl) are shown. Statistical analysis was performed using the Kruskal-Wallis test (non-parametric one way ANOVA) with Dunn’s multiple comparison test: *, P<0.05; ***, P<0.0001.

Fig 4

Time-kill experiments showing the viability of S. aureus untreated (Ctrl) or in the presence of increasing concentrations of the 2.6 kDa chitosan, (A) for strain MRSA 1158c, and (B) for the biofilm hyperproducer strain 2117. The CFU detection limit is 10 CFU/ml. Data are presented as means with standard deviations from three independent experiments.

Fig 5

Viability of S. aureus 2117 and MRSA 1158c either untreated (CTRL) or in the presence of a sub-MIC of chitosan (2.6 kDa) used alone or in combination with a sub-MIC of (A) erythromycin (ERY) or (B) tilmicosin (TIL) at 10 h post-treatment. The sub-MIC concentrations are indicated on the graphs. The CFU detection limit is 10 CFU/ml. Data are presented as means with standard deviations from three independent experiments. Significant differences between the control (CTRL) and test conditions are shown (ns, non significant; *, P<0.01; ***, P<0.001; ****, P<0.0001 as determined by using one way ANOVA with Tukey’s multiple comparison test.

Fig 6. Preformed biofilms of S. aureus strains exposed to the 2.6 kDa chitosan.

(A) Reduction of a preformed biofilm on pegs following exposure to increasing concentrations of chitosan. (B) Bactericidal effect of chitosan on preformed biofilms. The CFU/peg after 24 h of biofilm formation was evaluated for control pegs (CTRL 24 h) and this represented the inoculum at the onset of treatment, which occurred at 24 h for another 24 h of incubation. The CFU/peg obtained for the untreated pegs after the total incubation period served as the reference (CTRL 48 h) for treatment efficacy. Data were obtained from three independent experiments. Significant differences in comparison to the untreated controls (0 mg/ml in A, and CTRL 48 h in B) are shown. Statistical analysis was performed using non-parametric one way ANOVA: **, P<0.005; ***, P<0.001; ****, P<0.0001.

Fig 7. Bactericidal activity of increasing concentrations of tilmicosin used alone or in combination with a fixed sub-MIC of the 2.6 kDa chitosan against S. aureus strains embedded in preformed biofilms.

The CTRL 24 h and CTRL 48 h are as defined in the legend of Fig 6. Data were obtained from three independent experiments. Significant differences between tests done with or without chitosan at each tilmicosin concentration are shown. Statistical analysis was performed using multiple t tests: ****, P<0.0001.

Fig 8

Evaluation of internalization (A) and persistence (B) of the MRSA strain 1158c into MAC-T cells after exposure to the 2.6 kDa chitosan used at different concentrations. In A, MAC-T cells were exposed to bacteria and chitosan for a period of 3h. Cell cultures were then supplemented by lysostaphin to lyse extracellular bacteria, which are then removed, together with chitosan, through washing before provoking MAC-T cell lysis for determination of the number of intracellular bacteria at 3 h. In B, instead of lysing the MAC-T cells at 3 h, the number of persisting intracellular bacteria was determined at 24 h. Data were obtained from three independent experiments. Significant differences in comparison to the untreated control (0 mg/ml) are shown. Statistical analysis was performed using one way ANOVA with Dunnett’s multiple comparison: *, P<0.05.

Fig 9. Efficacy of the 2.6 kDa chitosan used alone or in combination with tilmicosin on S. aureus in a mouse mastitis model.

(A) Chitosan was administered twice at (0 h and 4 h post-inoculation) using 0.2, 2 or 10 mg/gland for the challenge against S. aureus strain Newbould. (B) A challenge with S. aureus strain 2117 was treated with the 2.6 kDa chitosan or tilmicosin (TIL) or a combination of both. Chitosan was administered twice at (0h and 4 h post-inoculation) whereas tilmicosin was administered 4 h post-inoculation. The middle bars indicate median values for each group of glands (n = 5–8) whereas the boxes specify quartiles Q1-Q3. The detection limit was 200 CFU (dotted line), below that, the glands were considered uninfected. Statistical analysis was performed using the Kruskal-Wallis test: *, P<0.05; ***, P<0.001.