Salmonella enterica serovar Typhimurium pathogenicity island 2 is necessary for complete virulence in a mouse model of infectious enterocolitis

Abstract

Salmonella species cause a wide range of disease in multiple hosts. Salmonella enterica serovar Typhimurium causes self-limited intestinal disease in humans and systemic typhoid-like illness in susceptible mice. The prevailing dogma in murine S. enterica serovar Typhimurium pathogenesis is that distinct virulence mechanisms-Salmonella pathogenicity islands 1 and 2 (SPI1 and SPI2)-perform distinct roles in pathogenesis, the former being important for invasion and intestinal disease and the latter important for intracellular survival and systemic persistence and disease. Although evidence from bovine infections has suggested that SPI2 has a role in ileal disease, there is no evidence that SPI2 is important for inflammation in a disease that more closely recapitulates human colitis. Using S. enterica serovar Typhimurium strains that lack functional type III secretion systems, we demonstrate that SPI2 is essential for complete virulence in murine infectious enterocolitis. Using a recently characterized murine model (M. Barthel,S. Hapfelmeier, L. Quintanilla-Martinez, M. Kremer, M. Rohde, M. Hogardt, K. Pfeffer, H. Russmann, and W. D. Hardt, Infect. Immun. 71:2839-2858, 2003), we demonstrate that SPI1 mutants are unable to cause intestinal disease 48 h after infection and that SPI2-deficient bacteria also cause significantly attenuated typhlitis. We show that at the peak of inflammation in the cecum, SPI2 mutants induce diminished intercellular adhesion molecule 1 expression and neutrophil recruitment but that wild-type and mutant Salmonella are similarly distributed in the lumen of the infected organ. Finally, we demonstrate that attenuation of intestinal inflammation is accompanied by resolution of typhlitis in the mutant, but not wild-type, infections. Collectively, these results indicate that SPI2 is needed for enterocolitis, as well as for systemic disease.

FIG. 1.

Salmonella enterica serovar Typhimurium-elicited enterocolitis is most severe in the ceca of wild-type infected mice and is partially attenuated in the absence of a SPI2 TTSS. Streptomycin-treated C57Bl/6 mice were infected with 3 × 10⁸ wild-type, SPI1, or SPI2 mutant Salmonella enterica serovar Typhimurium organisms. Histopathology scores from WT-, SPI1 mutant-, and SPI2 mutant-infected ceca (A), colons (B), and ilea (C) plus representative H&E-stained ceca (D to F) are shown. Histopathology was most severe in the ceca of all infected mice and was completely attenuated in ΔinvA-infected mice. Inflammation in the ceca and colons of mice infected with SPI2 mutants was statistically intermediate between WT- and SPI1-infected mice by the Kruskall-Wallis test (P < 0.0001). P values indicated represent a Mann-Whitney U test comparison of totals between groups. (A to C) Each bar represents a single mouse. Images are shown at a magnification of ×200.

FIG. 2.

S. enterica serovar Typhimurium in the ceca of infected mice is primarily extracellular and infiltrates the mucosa as inflammation progresses. Bacterial localization of SPI1 (A to D), SPI2 (E to H), and WT (I to L) S. enterica serovar Typhimurium 48 h after infection in the ceca of streptomycin-treated C57BL/6 mice. Paraformaldehyde-fixed optimal cutting temperature compound-embedded cryosections were stained for actin (red in merge) (A, E, and I), nuclei (with DAPI [4′,6′-diamidino-2-phenylindole; blue in merge) (B, F, and J), or against Salmonella LPS (green in merge) (C, G, and K). Mucosal infiltration was more common as inflammation worsened (merges shown in panels D, H, and K). Bacteria were not seen in the mucosa of SPI1-infected mice but were occasionally present in mucosa of SPI2-infected mice and common in the mucosa of WT-infected mice. Images are a pseudocolor and are shown at ×200 magnification. Insets are shown at ×100 magnification.

FIG. 3.

SPI2 but not SPI1 mutant S. enterica serovar Typhimurium induces ICAM-1 and neutrophil recruitment in the intestines of streptomycin-treated mice 48 h after infection. Streptomycin-treated mice were infected with SPI1 (A to C), SPI2 (D to F), or WT (G to I) S. enterica serovar Typhimurium for 48 h. Ceca were retrieved, paraformaldehyde fixed, OCT embedded, and cryosectioned prior to staining for ICAM-1 (green in merge) (A, D, and G) and nuclei (DAPI; blue in merge) (B, E, and H) by immunohistochemistry. Isotype control antibody staining of WT-infected mice was not significant (insets). ICAM-1 expression was evident in the epithelium and submucosal vessels of both WT- and SPI2-infected mice but decreased in the latter. Ceca of SPI1-infected animals showed minimal ICAM-1 expression. e, epithelium; v, submucosal vessel. Pictures are representative of eight mice/group and four tissue sections per mouse.

FIG. 4.

Neutrophil infiltration is markedly reduced in S. enterica serovar Typhimurium-induced typhlitis in the absence of SPI2. Forty-eight hours after infection with SPI1 mutant (A), SPI2 mutant (B), or WT (C) S. enterica serovar Typhimurium, ceca were retrieved, fixed, and stained for the neutrophil-specific chloroacetate esterase (red staining; arrows). Neutrophils were rare in ceca infected with SPI1- or SPI2-deficient S. enterica serovar Typhimurium (A and B) but abundant in WT-infected tissues. Pictures are representative of eight mice/group and four tissue sections per mouse.

FIG. 5.

Cecal inflammation caused by S. enterica serovar Typhimurium is controlled by day 5 in the absence of a functional SPI2 TTSS in streptomycin-treated mice. (A to C) Histopathology formalin-fixed, paraffin-embedded sections of ceca were stained with hemotoxylin and eosin 5 days after oral infection with 2 × 10⁶ WT (A) or SPI2 (B) S. enterica serovar Typhimurium organisms or after mock infection (LB broth only) (C). Severe inflammatory pathology is evident in WT-infected mice. Pathology is attenuated in SPI2 infected ceca at day 5. (D and E) Evidence of bacterial invasion of the mucosa is present in WT but not SPI2 S. enterica serovar Typhimurium-infected animals. Cryopreserved ceca were stained against actin (red), nuclei (blue), and Salmonella LPS (green). (F) Histopathology scores show persistent severe pathological changes in WT-infected mice at 5 days but significant recovery in SPI2-infected mice. ***, P < 0.001; Mann-Whitney U test. Each bar represents a single mouse. Hematoxylin and eosin images are shown at a magnification of ×400; fluorescent images are shown at a magnification of ×630.

FIG. 6.

Bacterial persistence in the intestines, liver, and spleen is significantly diminished between 2 and 5 days after oral SPI2 mutant, but not WT S. enterica serovar Typhimurium infection. Mice were infected with 2 × 10⁶ bacteria orally 24 h after oral administration of streptomycin. Intestinal (A) and systemic (B and C) bacterial loads significantly decreased between day 2 and day 5 postinfection with SPI2 S. enterica serovar Typhimurium but increased in the liver, spleen, and colon during WT infection. P values were determined by Student's t test for bacterial load. Bars indicate geometric means. ***, P < 0.0001; ns, not significant.