Ecotin protects Salmonella Typhimurium against the microbicidal activity of host proteases

Abstract

Salmonella enterica serovar Typhimurium causes acute diarrhea upon oral infection in humans. The harsh and proteolytic environment found in the gastrointestinal tract is the first obstacle that these bacteria face after infection. However, the mechanisms that allow Salmonella to survive the hostile conditions of the gut are poorly understood. The ecotin gene is found in an extensive range of known phyla of bacteria and it encodes a protein that has been shown to inhibit serine proteases. Thus, in the present work we studied the role of ecotin of Salmonella Typhimurium in host-pathogen interactions. We found that the Salmonella Typhimurium ∆ecotin strain exhibited lower inflammation in a murine model of Salmonella induced colitis. The ∆ecotin mutant was more susceptible to the action of pancreatin and purified pancreatic elastase. In addition, the lack of ecotin led to impaired adhesion to Caco-2 and HT-29 cell lines, related to the proteolytic activity of brush border enzymes. Besides, ∆ecotin showed higher susceptibility to lysosomal proteolytic content and intracellular replication defects in macrophages. In addition, we found Ecotin to have a crucial role in Salmonella against the microbicidal action of granule contents and neutrophil extracellular traps released from human polymorphonuclear leukocytes. Thus, the work presented here highlights the importance of ecotin in Salmonella as countermeasures against the host proteolytic defense system.

Fig 1. Reduced inflammation in the streptomycin pre-treatment murine infection model induced by

∆ecotin is associated with lower bacterial counts in the cecal epithelium. Streptomycin pre-treated mice were orally infected with 10⁷ CFU of WT (n=5), Δeco (n=4), Δeco+eco (n=5) strains, or mock infected with phosphate buffered saline (PBS) (n=5) and 24h p.i. were sacrificed. (A) Cecum was weighed separately for each mouse. Bars represent the mean±SEM. Each dot represents one mouse. Kruskal-Wallis test. (B) Histopathological sections of hematoxylin and eosin stained cecum tissue. Representative images of each group are shown. Histopathological parameters were scored: (C) edema of the submucosa, (D) infiltration of PMNs, (E) amount of Goblet cells with secretory vesicles, and (F) epithelial integrity. Bars represent mean±SEM. Each dot represents one mouse. (G) Combined score. Each bar represents an individual mouse. Mann-Whitney test. Data of one representative experiment from two independent experiments. (H-I) Streptomycin pre-treated mice were infected with 10⁷ CFU of WT (n=10) or Δeco (n=10) and 24h p.i. mice were sacrificed. Cecal contents and cecal tissue were obtained. (H) A fraction of the cecal content was plated, and the other half incubated with gentamicin to kill extracellular bacteria, then bacterial counts were determined. The mean±SEM is shown. Mann-Whitney test. (I) Bacterial counts were determined in cecal tissue. The mean±SEM is shown. Pooled data from two independent experiments. Each dot represents one mouse. Mann-Whitney test. ^ns p>0.05, *p<0.05, **p< 0.01.

Fig 2. Ecotin allows STm to survive against pancreatic proteases.

WT, Δeco and Δeco+eco were incubated with (A) pepsin or buffer (pH=3), (B) pancreatin (a complex mixture of enzymes from porcine pancreas) or buffer (pH=7) and (C) pancreatic elastase or buffer (pH=8.8). Bars represent the mean±SEM. Data pooled from at least 3 independent experiments. Dots represent technical replicates. Mann-Whitney test for (A) and Kruskal-Wallis test (B-C). ^nsp>0.05, *p<0.05. (D) WT and Δeco strains were incubated with pancreatic elastase or with buffer alone, both conditions supplemented with LB and growth was monitored using a microplate reader. The growth curves were plotted and the carrying capacity was calculated and normalized by the WT strain. Bars represent mean±SEM. Dots represent independent experiments. Mann-Whitney test. ^nsp>0.05, *p<0.05.

Fig 3. Impaired adhesion of the

∆ecotin strain to Caco-2 and HT-29 epithelial cell monolayers. (A) Purified brush border membranes were co-incubated with casein-BODIPY (fluorogenic substrate) and buffer, different concentrations of recombinant Ecotin or PIC (positive control). Casein digestion was followed by fluorescence determination. The % casein digestion was calculated as percentage of the slopes of Fluorescence over time curves for each treatment compared with the buffer alone condition (100% casein digestion). Bars represent the mean±SEM. Student’s t-test. *p<0.05, **p<0.01. (B) Adhesion assay in HT-29 or Caco-2 cell line monolayers. Cells were incubated with WT, Δeco and Δeco+eco strains for 30min, washed, bacterial count determined and normalized to inoculum (adhesion index). Bars represent the mean±SEM. Dots are technical replicates. Representative of three independent experiments. Kruskal-Wallis test. ^nsp>0.05, **p< 0.01, ***p<0.001. (C-D) Adhesion assay: Caco-2 cell line monolayers were incubated for 30min with WT or ∆eco strains carrying GFP expressing plasmid (green). Actin filaments were stained with phalloidin (red) and nuclei were stained with Topro-3 (white). Bacteria in random fields were counted and normalized by the number of nuclei per field. Bars represent the mean±SEM. Dots are random fields. Pooled data from two independent experiments. Mann-Whitney test. *p<0.05. (E) Competitive adhesion assay performed in HT-29 or Caco-2 cell line monolayers. The competitive index was calculated as the ratio in bacterial count after 30min of infection corrected by the ratio in the inoculum. Bars represent mean±SEM. Dots are technical replicates. Representative of at least two independent experiments. One-sample t-test vs 1. ^nsp>0.05, **p< 0.01, ****p<0.0001. (F) Adhesion assays in Caco-2 were repeated with the pre-treatment of the indicated wells with recombinant Ecotin 1 µM or PIC for 5min before the addition of bacteria. Bars represent the mean±SEM. Dots are technical replicates. Representative of two independent experiments. One-way ANOVA. ^nsp>0.05, *p<0.05.

Fig 4. ∆ecotin has attenuated intracellular replication in murine macrophages and is more susceptible to the action of intracellular proteases.

(A) J774 macrophages were incubated with WT, ∆eco or ∆eco+eco strains for 30min. Then, cells were treated with gentamicin for 1h, washed, lysed and bacterial counts determined. Bacterial counts were normalized by the inoculum (invasion index). Bars represent the mean±SEM. Dots are technical replicates. Representative of three independent experiments. One-Way ANOVA with Bonferroni post-test. ^nsp>0.05. (B) Cells were infected as in A, but incubated for a further 3h before washing and bacterial counts determination. The bacterial counts at this time point were then normalized by the bacterial counts in the invasion time point (replication index). Bars represent the mean±SEM. Dots are technical replicates. Representative of three independent experiments. One-Way ANOVA with Bonferroni post-test. ^nsp>0.05, **p< 0.01. (C) Competitive assay performed in J774 macrophages. The competitive index was calculated as the ratio between ∆eco/WT recovered at the invasion time point or 4h time point corrected by inoculum or invasion ratio respectively. Bars represent the mean±SEM, Dots are technical replicates. Representative of three independent experiments. One-Sample t-test vs 1. ^nsp>0.05, **p< 0.01. (D) J774 macrophages were infected with WT or ∆eco strains carrying a TIMER^bac plasmid and fluorescence intensity was analyzed by confocal microscopy. The ratio of green/orange fluorescence of TIMER^bac indicates the growth rate of Salmonella, slow-growing cells appear orange/red and faster growing cells are greener. Bars represent the mean±SEM. Dots represent individual bacteria from at least 10 random fields. Pooled data from two experiments. Mann-Whitney test. ^nsp>0.05, ****p< 0.0001. (E-F) J774 macrophages were infected with WT or ∆eco strains expressing GFP, then cells were fixed and labeled with a primary anti LAMP-1 antibody and a secondary antibody coupled with AlexaFluor647. Colocalization was analyzed in random fields using Manders approach to measure LAMP-1 recruitment. Bars represent the mean±SEM. Dots are random fields pooled from two experiments. Student’s t-test. ^nsp>0.05, *p< 0.05. (G) Microsomes from J774 macrophages or buffer were co-incubated for 2h with WT or ∆eco strains. The bacterial count was normalized by WT strain. Bars represent the mean±SEM. Pooled data from two experiments. Student’s t-test. ^nsp>0.05, *p< 0.05.

Fig 5. Ecotin protects STm from the proteolytic microbicidal activity of human PMNs.

(A) Neutrophil elastase (NE) was incubated with a fluorogenic substrate and different concentrations of recombinant Ecotin or a commercial protease inhibitor cocktail (PIC). Protease activity was measured over time using a fluorescence microplate reader. The percentage of substrate digestion was calculated as the percentage of fluorescence vs. time slope regarding the condition without inhibitor (0nM) (100% substrate digestion). (B) Purified human PMNs were incubated with WT, ∆eco or ∆eco+eco strains in the presence or absence of DPI and after 90min bacterial counts were determined and normalized to bacteria incubated without neutrophils (100% Survival). Bars represent the mean±SEM. Dots are individual donors. One-way ANOVA with Bonferroni post-test. ^nsp>0.05, *p<0.05. (C) Degranulation was induced in purified human PMNs and the degranulated supernatant was incubated with WT or ∆eco strains. After 90min of incubation the bacterial counts were determined and normalized to bacteria incubated without degranulated supernatant (100% Survival). Bars represent the mean±SEM. Dots are individual donors. Student’s t-test. *p<0.05. (D) Neutrophil extracellular traps (NETs) were induced in purified human PMNs and the supernatant containing the NETs was incubated with WT, ∆eco, ∆eco+eco strains or ∆eco supplemented with recombinant Ecotin (∆eco+rEcotin). Then, 90min after incubation, the bacterial counts were determined and normalized to bacteria incubated without NETs (100% Survival). Bars represent the mean±SEM. Dots are individual donors. One-way ANOVA with Bonferroni post-test. ^nsp>0.05, *p<0.05.