The alternative sigma factor, sigmaE, is critically important for the virulence of Salmonella typhimurium

Abstract

In Escherichia coli, extracytoplasmic stress is partially controlled by the alternative sigma factor, RpoE (sigmaE). In response to environmental stress or alteration in the protein content of the cell envelope, sigmaE upregulates the expression of a number of genes, including htrA. It has been shown that htrA is required for intramacrophage survival and virulence in Salmonella typhimurium. To investigate whether sigmaE-regulated genes other than htrA are involved in salmonella virulence, we inactivated the rpoE gene of S. typhimurium SL1344 by allelic exchange and compared the phenotype of the mutant (GVB311) in vitro and in vivo with its parent and an isogenic htrA mutant (BRD915). Unlike E. coli, sigmaE is not required for the growth and survival of S. typhimurium at high temperatures. However, GVB311 did display a defect in its ability to utilize carbon sources other than glucose. GVB311 was more sensitive to hydrogen peroxide, superoxide, and antimicrobial peptides than SL1344 and BRD915. Although able to invade both macrophage and epithelial cell lines normally, the rpoE mutant was defective in its ability to survive and proliferate in both cell lines. The effect of the rpoE mutation on the intracellular behavior of S. typhimurium was greater than that of the htrA mutation. Both GVB311 and BRD915 were highly attenuated in mice. Neither strain was able to kill mice via the oral route, and the 50% lethal dose (LD50) for both strains via the intravenous (i.v.) route was very high. The i.v. LD50s for SL1344, BRD915, and GVB311 were <10, 5.5 x 10(5), and 1.24 x 10(7) CFU, respectively. Growth in murine tissues after oral and i.v. inoculation was impaired for both the htrA and rpoE mutant, with the latter mutant being more severely affected. Neither mutant was able to translocate successfully from the Peyer's patches to other organs after oral infection or to proliferate in the liver and spleen after i.v. inoculation. However, the htrA mutant efficiently colonized the livers and spleens of mice infected i.v., but the rpoE mutant did not. Previous studies have shown that salmonella htrA mutants are excellent live vaccines. In contrast, oral immunization of mice with GVB311 was unable to protect any of the mice from oral challenge with SL1344. Furthermore, i.v. immunization with a large dose ( approximately 10(6) CFU) of GVB311 protected less than half of the orally challenged mice. Thus, our results indicate that genes in the sigmaE regulon other than htrA play a critical role in the virulence and immunogenicity of S. typhimurium.

FIG. 1

Effect of media and temperature on the growth of S. typhimurium rpoE mutant. Overnight cultures of the S. typhimurium strains SL1344 (■), BRD915 (●), GVB311 (▴), and GVB311(pSH117) (▾) were used to inoculate fresh media, which was then incubated aerobically. Growth was followed spectrophotometrically. (A) Growth in LB-G at 42°C. (B) Growth in LB at 42°C.

FIG. 2

Effect of the rpoE and htrA mutations on the sensitivity of S. typhimurium to oxidizing agents. The salmonella strains were tested for sensitivity to 3% H₂O₂ (A) or 2% paraquat (B) by disk diffusion assay. The plates were incubated overnight at 30 or 37°C as indicated in the figure below each lane. Each bar represents the mean diameter of the zone of inhibition, and the error bar shows the standard deviation (SD) of 30 replica assays. The asterisk indicates that the mean diameter of the zones is statistically different (P < 0.05) from that of SL1344 incubated at the same temperature; the number sign (#) indicates that the mean is statistically different from BRD915 grown at the same temperature (one-way analysis of variance [ANOVA]).

FIG. 3

Effect of the rpoE and htrA mutations on the sensitivity of S. typhimurium to polymyxin B. SL1344, BRD915, and GVB311 were tested for sensitivity to polymyxin B by disk diffusion assay. Disks containing 300 U of polymyxin B (Oxoid) were placed on the plates and incubated overnight at 37°C. The diameter of the zone of inhibition was then measured in millimeters. Each bar represents the mean diameter of the zone of inhibition, and the error bar shows the SD of 30 replica assays. The asterisk indicates that the mean diameter of the zones is statistically different (P < 0.05) from that of SL1344, and the number sign (#) indicates that the mean is statistically different from BRD915 (one-way ANOVA).

FIG. 4

Effect of the rpoE and htrA mutations on the ability of S. typhimurium to invade and survive in macrophages. The ability of the different S. typhimurium strains to invade and survive in macrophages was examined by using the murine macrophage-like cell line RAW264.7. RAW264.7 cells in the wells of tissue culture plates were infected with bacteria at a multiplicity of infection of ∼1:1. The assay was performed as described in the text. The graph shows the number of viable bacteria inside macrophages at 3 and 24 h after infection. Each bar represents the mean of CFU from triplicate experiments, and the error bars indicate the SD.

FIG. 5

Colonization, growth, and survival of BRD915 and GVB311 in murine tissues after oral or i.v. inoculation. Groups of mice were infected with BRD915 or GVB311 orally (10¹⁰ CFU) or i.v. (10⁶ CFU). At the indicated periods after infection, organs were removed from groups of four mice and homogenized and the number of viable organisms present was determined. Symbols: ■, PPs; ●, mesenteric lymph nodes; ▴, spleen; and ▾, liver. Each point represents the mean CFU/organ for four mice, and the error bars indicate one standard error of the mean.