Salmonella Biofilms Tolerate Hydrogen Peroxide by a Combination of Extracellular Polymeric Substance Barrier Function and Catalase Enzymes

Abstract

The ability of Salmonella enterica subspecies enterica serovar Typhi (S. Typhi) to cause chronic gallbladder infections is dependent on biofilm growth on cholesterol gallstones. Non-typhoidal Salmonella (e.g. S. Typhimurium) also utilize the biofilm state to persist in the host and the environment. How the pathogen maintains recalcitrance to the host response, and oxidative stress in particular, during chronic infection is poorly understood. Previous experiments demonstrated that S. Typhi and S. Typhimurium biofilms are tolerant to hydrogen peroxide (H₂O₂), but that mutations in the biofilm extracellular polymeric substances (EPSs) O antigen capsule, colanic acid, or Vi antigen reduce tolerance. Here, biofilm-mediated tolerance to oxidative stress was investigated using a combination of EPS and catalase mutants, as catalases are important detoxifiers of H₂O₂. Using co-cultured biofilms of wild-type (WT) bacteria with EPS mutants, it was demonstrated that colanic acid in S. Typhimurium and Vi antigen in S. Typhi have a community function and protect all biofilm-resident bacteria rather than to only protect the individual cells producing the EPSs. However, the H₂O₂ tolerance deficiency of a O antigen capsule mutant was unable to be compensated for by co-culture with WT bacteria. For curli fimbriae, both WT and mutant strains are tolerant to H₂O₂ though unexpectedly, co-cultured WT/mutant biofilms challenged with H₂O₂ resulted in sensitization of both strains, suggesting a more nuanced oxidative resistance alteration in these co-cultures. Three catalase mutant (katE, katG and a putative catalase) biofilms were also examined, demonstrating significant reductions in biofilm H₂O₂ tolerance for the katE and katG mutants. Biofilm co-culture experiments demonstrated that catalases exhibit a community function. We further hypothesized that biofilms are tolerant to H₂O₂ because the physical barrier formed by EPSs slows penetration of H₂O₂ into the biofilm to a rate that can be mitigated by intra-biofilm catalases. Compared to WT, EPS-deficient biofilms have a heighted response even to low-dose (2.5 mM) H₂O₂ challenge, confirming that resident bacteria of EPS-deficient biofilms are under greater stress and have limited protection from H₂O₂. Thus, these data provide an explanation for how Salmonella achieves tolerance to H₂O₂ by a combination of an EPS-mediated barrier and enzymatic detoxification.

Keywords: Salmonella; biofilms; chronic infection; extracellular polymeric substances (EPSs); hydrogen peroxide; innate immunity.

Figure 1

Tolerance of WT S. Typhimurium and EPS mutants to H₂O₂. (A–H) Biofilm aggregates carrying an antibiotic resistance cassette were challenge with H₂O₂ at a known tolerable dose (2.5 mM) and a challenge dose (125 mM). Statistical significance was determined by one-way ANOVA with Dunnett correction for multiple comparisons (**p < 0.01; ****p < 0.0001). Each experiment was conducted in triplicate and the data represents the mean of three independent experiments. The error bars indicate SD.

Figure 2

Tolerance of co-cultured WT S. Typhimurium and EPS mutant biofilms to H₂O₂. (A–G) Biofilms were cultured with a 1:1 ratio of WT S. Typhimurium and an EPS mutant. Aggregates of these biofilms were challenged with H₂O₂ at a known tolerable dose (2.5 mM) and a challenge dose (125 mM) then enumerated on differential antibiotic agar. Significant differences were determined by two-way ANOVA and Tukey method for multiple comparison correction (****p < 0.0001). No significant differences were observed between WT and mutant at any one H₂O₂ concentration. Each experiment was conducted in triplicate and the data represents the mean of three independent experiments. The error bars indicate SD.

Figure 3

Tolerance of single- and co-cultured S. Typhi biofilms to H₂O₂. (A) WT S. Typhi biofilm aggregates carrying an Amp resistance cassette were challenged with H₂O₂ at a known tolerable dose (2.5 mM) and a challenge dose (25 mM). (B) S. Typhi ΔtviB biofilm aggregates were challenged in the same conditions as (A). (A, B) Statistical significance was tested for/determined by one-way ANOVA with Dunnett correction for multiple comparisons (****p < 0.0001). (C) WT S. Typhi and S. Typhi ΔtviB biofilms were cultured in a 1:1 ratio, challenged with H₂O₂, and enumerated on differential antibiotic agar. Significance was tested with two-way ANOVA and Tukey correction for multiple comparisons (***p < 0.0005). No significant differences were observed between S. Typhi and S. Typhi ΔtviB at any one H₂O₂ concentration. Each experiment was conducted in triplicate and the data represents the mean of three independent experiments. The error bars indicate SD.

Figure 4

Tolerance of S. Typhimurium catalase mutant biofilms to H₂O₂. (A–C) Biofilm aggregates were challenge with H₂O₂ at their planktonic MIC (1.25 mM) and escalating challenge concentrations of 12.5 mM, 31.25 mM, and 62.5 mM H₂O₂ to determine the limit of tolerance. Values were selected to represent a 0×, 1×, 10×, 25×, or 50× increase from the experimentally-determined MIC. Significance was determined by one-way ANOVA and Dunnett multiple comparison correction (***p < 0.0005; ****p < 0.0001). Each experiment was conducted in triplicate and the data represents the mean of three independent experiments. The error bars indicate SD.

Figure 5

Tolerance of co-cultured WT S. Typhimurium and catalase mutant biofilms to H₂O₂. (A–C) Biofilms were cultured in a 1:1 ratio of WT S. Typhimurium and one other catalase mutant. Biofilm aggregates were challenged with H₂O₂ at a known tolerable dose (2.5 mM) and two challenge doses (62.5 mM and 125 mM) representing tolerance limits for the catalase mutants and WT S. Typhimurium, respectively. Differential antibiotics were used for enumeration and significant differences were identified using two-way ANOVA and Tukey method for multiple comparison correction (****p < 0.0001). No significant differences were observed between WT and mutant at anyone H₂O₂ concentration. Each experiment was conducted in triplicate and the data represents the mean of three independent experiments. The error bars indicate SD.

Figure 6

Transcriptional response of catalase genes in WT S. Typhimurium and S. Typhimurium ΔcsgAΔwcaMΔyihOΔbcsE biofilms. qPCR measured copy number of (A) putative catalase gene (B) katE or (C) katG transcripts present in biofilm cDNA reverse transcribed from total biofilm RNA. RNA samples were collected for WT and EPS mutant at each H₂O₂ concentration from 0-2 hours. The data are expressed as fold-change in EPS mutant relative to the WT as determined by the Livak method (Livak and Schmittgen, 2001). Values >1 indicate increased transcription of the target gene by the mutant compared to WT and values <1 represent a decrease. Biofilm RNA was collected and reverse transcribed to cDNA from three independent biofilm challenge experiments. Each cDNA sample was analyzed in triplicate by qPCR to determine C_T value of each gene and the data represent the mean of three C_T values associated with each independent challenge experiment. Error bars indicate SD.

Figure 7

Catalase (translational) response in WT S. Typhimurium and S. Typhimurium ΔcsgAΔwcaMΔyihOΔbcsE biofilms. Western blot was used to analyze total catalase content present in biofilms at each H₂O₂ concentration from 0-2 hours. (A) Western blot demonstrating general catalase production from WT and EPS mutant biofilms. Densitometry analysis with ImageJ software (Schneider et al., 2012) was used to quantify Western blots of total protein collected from three independent biofilm challenge experiments for (B) WT or (C) EPS mutant biofilms. Densitometry values were normalized to the WT signal intensity at T₀, 0 mM.