Salmonella enterica Serovar Typhimurium Exploits Inflammation to Modify Swine Intestinal Microbiota

Abstract

Salmonella enterica serovar Typhimurium is an important zoonotic gastrointestinal pathogen responsible for foodborne disease worldwide. It is a successful enteric pathogen because it has developed virulence strategies allowing it to survive in a highly inflamed intestinal environment exploiting inflammation to overcome colonization resistance provided by intestinal microbiota. In this study, we used piglets featuring an intact microbiota, which naturally develop gastroenteritis, as model for salmonellosis. We compared the effects on the intestinal microbiota induced by a wild type and an attenuated S. Typhimurium in order to evaluate whether the modifications are correlated with the virulence of the strain. This study showed that Salmonella alters microbiota in a virulence-dependent manner. We found that the wild type S. Typhimurium induced inflammation and a reduction of specific protecting microbiota species (SCFA-producing bacteria) normally involved in providing a barrier against pathogens. Both these effects could contribute to impair colonization resistance, increasing the host susceptibility to wild type S. Typhimurium colonization. In contrast, the attenuated S. Typhimurium, which is characterized by a reduced ability to colonize the intestine, and by a very mild inflammatory response, was unable to successfully sustain competition with the microbiota.

Keywords: Salmonella Typhimurium; immune response; inflammation; microbiota; pig; salmonellosis.

Figure 1

STM^ΔznuABC (group B) shows a lower virulence in piglets compared to the STM^wt (group C). (A) Mean values and standard deviation (SD) bars of body temperature of study groups in different time points. In the table on the bottom the levels of significance were reported among groups at different time points. Different letters at each time point represent significant different results (P ≤ 0.05, Dunn's test). (B) Mean values and SD bars of CFU/g of STM^ΔznuABC and STM^wt shed in feces. Results of piglets infected with STM^ΔznuABC were compared to results of STM^wt and differences were statistically significant when ^*P ≤ 0.05, Mann–Whitney test.

Figure 2

(A–D) S. Typhimurium induces an inflammatory response correlated to the virulence of the bacterial strain. Haptoglobin, TNF-α, IL1-α, and C-reactive protein levels in serum of animals were determined by ELISA. The asterisks indicate statistical significance: ^*P ≤ 0.05 and ^**P ≤ 0.01 (multiple comparisons-Dunn's test).

Figure 3

STM^wt induces a higher organs colonization than STM^ΔznuABC at 1 dpi. Piglets were orally infected with 2 × 10⁹ CFU of STM^ΔznuABC (group B) or STM^wt (group C), and bacterial burdens were determined at 1 dpi. Differences between groups B and C were estimated using non-parametric Mann–Whitney test and were considered significant when ^*P ≤ 0.05. Organ samples taken from naïve animals (group A) were negative. Error bars represent one SD from the mean.

Figure 4

(A–C) Photomicrographies showing histological changes of the cecum. (A) Naïve control piglets; (B) piglets infected with STM ^ΔznuABC: multifocal and moderate neutrophilic infiltrate (arrows), crypt abscess formation; (C) piglets infected with STM^wt: marked and diffuse neutrophilic infiltration.

Figure 5

(A–H) Cytokines expression reveals that unlike STM^wt, STM^ΔznuABC strain is not able to induce a strong host immune response. TNF-α, IL1-α, IL1-β, and INF-γ expression was measured in the cecum at 1 and 12 dpi by real time RT-PCR. Gray bars and black bars represent STM^ΔznuABC-infected (group B) and STM^wt-infected piglets (group C), respectively. The asterisk indicates statistical significance ^*P ≤ 0.05, Mann–Whitney test.

Figure 6

STM^wt and STM^ΔznuABC differently modify cecal microbiota of piglets. Piglets were sacrificed at 1 and 12 dpi (A,B). Bacterial genomic DNA was isolated from cecal content and qPCR analysis measured the abundance of specific commensal bacterial groups. White bars represent uninfected controls (group A). Gray and gray-black bars represent STM^ΔznuABC- (group B) and STM^wt- (group C) infected piglets, respectively. P-values were calculated using One-way ANOVA with Bonferroni's post-test. Significant differences between groups are indicated by ^*P ≤ 0.05, ^**P ≤ 0.01, and ^***P ≤ 0.001. Eub, all bacteria; Lacto, Lactobacillus/Lactococcus group; Clost, Eubacterium rectale/Clostridium coccoides; Bact, Bacteroides sp.; Bifido, Bifidobacterium; Prev, Prevotellaceae; Ent, Enterobacteriaceae other than Salmonella; STM, S. Typhimurium.

Figure 7

Structural comparison of fecal microbiota among groups A, B, and C. The Shannon index (A) and Fisher92s alpha (B) were used to estimate diversity of the fecal microbiota in naïve animals (group A) and in STM^ΔznuABC- (group B) and STM^wt- (group C) infected piglets. Boxes represent median, and first and third quartiles; whiskers indicate minimum and maximum values. The asterisks indicate statistical significance ^*P ≤ 0.05 and ^**P ≤ 0.01, Dunn's test.

Figure 8

Principal Coordinate analysis plot (PCoA) of unweighted UniFrac distances (A) and Bray-Curtis dissimilarity (B) for the fecal microbiota across the three study groups. PCA, principal coordinate.

Figure 9

(A,B) Heatmap indicating genus-level changes in the microbiota composition of piglets Naïve (group A), and piglets infected with STM^ΔznuABC (group B) or with STM^wt (group C) at 2 and 12 dpi. The relative abundance of the most represented genera is indicated by a gradient of color from white (low abundance) to red (high abundance). The hierarchical clustering analysis of the samples, based on the similarity of the microbiota composition, are displayed on the left. Animals 1–5: group A (Naïve), green; animals 6–10: group B (STM^ΔznuABC), blue; piglets 11–15: group C (STM^wt), orange.