A model of Salmonella colitis with features of diarrhea in SLC11A1 wild-type mice - PubMed (original) (raw)

A model of Salmonella colitis with features of diarrhea in SLC11A1 wild-type mice

Heungjeong Woo et al. PLoS One. 2008.

Abstract

Background: Mice do not get diarrhea when orally infected with S. enterica, but pre-treatment with oral aminoglycosides makes them susceptible to Salmonella colitis. However, genetically susceptible ItyS mice (Nramp1(G169D) allele) die from systemic infection before they develop diarrhea, so a new model is needed to study the pathogenesis of diarrhea. We pretreated ItyR mice (Nramp1(G169)) with oral kanamycin prior to infecting them with virulent S. Typhimurium strain 14028s in order to study Salmonella-induced diarrhea. We used both a visual scoring system and the measurement of fecal water content to measure diarrhea. BALB/c.D2(Nramp1) congenic started losing weight 5 days post-infection and they began to die from colitis 10-14 days after infection. A SPI-1 (invA) mutant caused cecal, but not colonic inflammation and did not cause diarrhea. A phoP- mutant did not cause manifestations of diarrhea in either normal or NADPH-deficient (gp91(phox)) mice. However, strain 14028s caused severe colitis and diarrhea in gp91(phox)-deficient mice on an ItyR background. pmr A and F mutants, which are less virulent in orally infected BALB/c mice, were fully virulent in this model of colitis.

Conclusions: S. enterica must be able to invade the colonic epithelium and to persist in the colon in order to cause colitis with manifestations of diarrhea. The NADPH oxidase is not required for diarrhea in Salmonella colitis. Furthermore, a Salmonella phoP mutant can be cleared from the colon by non-oxidative host defenses.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Time course of infection in BALB/c.D2Nramp1 congenic mice with Salmonella colitis.

A. Cumulative mean weight loss of mice that were infected with S. Typhimurium strain 14028s after pretreatment with oral kanamycin, compared to controls that received only oral kanamycin. (The controls were not followed beyond 8 days.) B. Numbers of Salmonella (geometric mean CFU±SEM) recovered from feces, ceca, mesenteric nodes, and spleens at indicated time points after infection. There were 5 mice at each time point except for day 14 when there were only 3 mice.

Figure 2. WT Salmonella causes extensive colitis in BALB/c.D2 mice.

A. H&E stained section from a representative mouse showing two adjacent turns of the colon, 9 days after infection, at low magnification (40X). Note the extensive inflammation throughout the colon with many mucosal ulcers (arrows), and exudation of inflammatory cells and sloughed epithelium from opposing sides of a segment of colon merging in the lumen. B. Higher power (200X) view of the colonic mucosa from the same mouse. Note the infiltration of inflammatory cells (PMN and mononuclear cells), mucosal hemorrhage (Arrow heads), and mucosal ulceration. There was no sub-mucosal edema in this segment but there was extensive edema in other areas of the section (see B).

Figure 3. A SPI-1 (invA) mutant of 14028s does not cause diarrhea despite being able to invade the colons of BALB/c.D2Nramp1 congenic mice.

A. Diarrhea scores (mean±1SEM) for mice infected with 14028s (WT) or the invA mutant, 8 or 9 days after infection. The numbers of mice in each group are shown above the bars. The difference in means (Mann Whitney U test) was significant (P<0.05). B. Mean water content of feces from uninfected (control), 14028a-infected (WT), and _invA_-infected mice 8 days after infection. There was no significant difference between the % of water in feces from _invA_-infected and from control mice. The % of water in feces from WT–infected mice was significantly greater than the % in the other two groups. C. CFU from feces and three segments of the colon 8 days after infection with either WT or _invA_- Salmonella. Each symbol represents one mouse and the horizontal bars show the geometric means. P values were calculated using the Mann-Whitney U test.

Figure 4. invA causes inflammation in the cecum but not the distal colon.

A. H&E stained section (100X) of the cecum from a representative mouse 9 days after infection with _invA_-. Note the extensive sub-mucosal edema (asterisk), inflammatory infiltrate, and an area of mucosal ulceration. B. H&E stained section of 3 turns of the colon from the same mouse, showing normal mucosal architecture with essentially no inflammation and normal numbers of mucus secreting cells.

Figure 5. A phoP mutant of 14028s invades the colon, but does not persist in the colon, and does not cause colitis in BALB/c.D2Nramp1 congenic mice.

A. Diarrhea scores for _phoP-_infected mice on days 2 and 8/9 after infection. The score is significantly higher on day 2 (P<0.05). B. Water content of the stool on those days. The water content of the stool of _phoP_-infected mice on day 2 after infection was significantly higher than that of uninfected control mice and not significantly different from the water content of mice infected with WT Salmonella (14028s). However, by day 8 the water content of feces from mice infected with WT had increased to nearly 80% while the water content of feces from _phoP_-infected mice had returned to normal. C. Comparison of the numbers of CFU of WT and phoP Salmonella on days 2 and 7 after infection. Geometric means±1SEM are shown. * = P <0.05, ** = <0.001. P values for the differences between means were calculated using the Mann-Whitney test. There were 6 mice/group. We did not calculate P values for spleen since there were <10 CFU of the phoP mutant, or for the CFU in stool on day 8 since we had only 3 stools from WT mice for culture.

Figure 6. phoP cause transient inflammation of the cecum in BALB/c.D2Nramp1 congenic mice.

A. H&E stained section of the cecum of a representative mouse 2 days after infection (100x). Note the edema, intense mixed cellular infiltrate in the mucosa and sub-mucosa, and the paucity of mucus secreting epithelial cells. B. A cecum from a representative mouse 9 days after infection with phoP (100X). The tissue is normal in appearance (100X).

Figure 7. pmrA and pmrF mutants cause severe colitis.

BALB/c.D2 mice were infected as in Methods and sacrificed 8 days later. There were no statistically significant differences between the diarrhea scores or the water content of feces from mice infected with WT or the isogenic pmrA or pmrF mutants. The numbers of WT and mutant Salmonella that we recovered from the tissues on day 8 after infection were either not significantly different or the pmr mutants were slightly more numerous.

Figure 8. Mice with CGD are more susceptible to Salmonella colitis.

Male mice with CGD and littermate controls (female carriers) were bred so that they carried one copy of an Nramp1G169 allele. Mice were infected as in Methods and sacrificed 7 days later. We infected 12 mice/group and sacrificed half of them on day 2. A. Diarrhea scores for CGD mice and littermate controls 2 and 7 days after infection. The mean diarrhea score for CGD mice was significantly higher than scores for the controls on day 2 (P<0.05), but the scores for the two strains were not significantly different on day 7, although there were only 3 surviving CGD mice on day 7. B. Geometric mean CFU±SEM for various tissues on days 2 and 7 after infection. Colony counts were similar in the two groups on day two but significantly higher in the CGD mice on day 7 after infection. * indicates a P value (Mann Whitney U test) of <0.01. On day 7 there were no solid feces in the colons of CGD mice so we could not culture their stool, and only 2/6 females had solid feces to culture.

Figure 9. Mice with CGD are not more susceptible to oral infection with a phoP mutant Salmonella.

A. Mean±SD of diarrhea scores of mice with CGD and controls (female littermates), 2 and 8 days after infection. There was no significant difference between the diarrhea scores of the two groups on either day. B. Quantitative bacteriology results for mice with CGD and control. There were no significant differences between CFU recovered from mice with CGD and littermate controls on either day. Note that by day 8 there were only ∼200 CFU recovered from the ceca of both groups, and there were <10 CFU recovered from spleen cultures. , indicating that in vivo the NADPH oxidase was not involved in the highly effective innate immune response that can resolve an oral infection with phoP mutants.

References

1. Mead PS, Slutsker L, Dietz V, McCaig LF, Bresee JS, et al. Food-related illness and death in the United States. Emerg Inf Dis. 1999;5:607–624. - PMC - PubMed
1. Roberts T. Human illness costs of foodborne bacteria. Am J Agricultural Econ. 1989;71:468–474.
1. Helms M, Ethelberg S, Molbak K DT104 Study Group. International Salmonella Typhimurium DT104 Infections, 1992–2001. Emerg Inf Dis. 2005;11:859–867. - PMC - PubMed
1. Varma JK, Greene KD, Ovitt J, Barrett TJ, Medalla F, et al. Hospitalization and antimicrobial resistance in Salmonella outbreaks,1984–2002. Emerg Inf Dis. 2005;11:943–946. - PMC - PubMed
1. Jones BD, Paterson HF, Hall A, Falkow S. Salmonella typhimurium induces membrane ruffling by a growth factor-receptor-independent mechanism. Proc Natl Acad Sci U S A. 1993;90:10390–10394. - PMC - PubMed

A model of Salmonella colitis with features of diarrhea in SLC11A1 wild-type mice - PubMed (original) (raw)