Influence of 5 major Salmonella pathogenicity islands on NK cell depletion in mice infected with Salmonella enterica serovar Enteritidis

Abstract

Background: In this study we were interested in the colonisation and early immune response of Balb/C mice to infection with Salmonella Enteritidis and isogenic pathogenicity island free mutants.

Results: The virulence of S. Enteritidis for Balb/C mice was exclusively dependent on intact SPI-2. Infections with any of the mutants harbouring SPI-2 (including the mutant in which we left only SPI-2 but removed SPI-1, SPI-3, SPI-4 and SPI-5) resulted in fatalities, liver injures and NK cell depletion from the spleen. The infection was of minimal influence on counts of splenic CD4 CD8 T lymphocytes and gammadelta T-lymphocytes although a reduced ability of splenic lymphocytes to respond to non-specific mitogens indicated general immunosuppression in mice infected with SPI-2 positive S. Enteritidis mutants. Further investigations showed that NK cells were depleted also in blood but not in the caecal lamina propria. However, NK cell depletion was not directly associated with the presence of SPI-2 and was rather an indicator of virulence or avirulence of a particular mutant because the depletion was not observed in mice infected with other attenuated mutants such as lon and rfaL.

Conclusions: The virulence of S. Enteritidis for Balb/C mice is exclusively dependent on the presence of SPI-2 in its genome, and a major hallmark of the infection in terms of early changes in lymphocyte populations is the depletion of NK cells in spleen and blood. The decrease of NK cells in circulation can be used as a marker of attenuation of S. Enteritidis mutants for Balb/C mice.

Figure 1

Death rates (panel A) and faecal shedding (panel B) in mice orally infected with S. Enteritidis and SPI mutants. Mice infected with SPI-2 positive mutants exhibited high faecal shedding and died within 3 weeks post-infection. Faecal shedding of individual mice which survived the infection with ΔSPI1, ΔSPI4 and SPI2o (i.e. SPI-2 positive mutants) beyond day 10 is not shown for clarity. Survival rates of the mice infected with ΔSPI2, ΔSPI1-5 and SPI1o, SPI3o, SPI4o and SPI5o were significantly different from those infected with the wild type S. Enteritidis as determined by Logrank test at P < 0.01.

Figure 2

Histological analysis of liver samples of mice infected with the wild-type S. Enteritidis or SPI-2 mutants. Arrows points towards necrotic areas with neutrophil infiltration. A - liver of mice infected with the wild type S. Enteritidis, B - liver of mice infected with the ΔSPI2 mutant, C - liver of mice infected with the SPI2o mutant, D - liver of mice infected with the ΔSPI1-5 mutant. Exactly the same pathology, depending on the presence or absence of SPI-2, was observed in the other mice infected with the other SPI mutants. Bar indicates 100 μm.

Figure 3

Lymphocyte proliferation assay from non-infected mice (white columns), and mice infected with SPI2-negative (light grey columns) and SPI2-positive (dark grey columns) S. Enteritidis mutants after the stimulation with different concentrations of phytohaemagglutinin (PHA), concanavalin A (ConA) or pokeweed mitogen (PWM). * - t-test different from the mice infected with the SPI-2 positive S. Enteritidis at P < 0.05.

Figure 4

CD3 CD19 double-negative lymphocytes in spleens of mice infected with S. Enteritidis SPI mutants; n.i. - non-infected mice. The Y-axis shows percentage of CD3 CD19 double-negative lymphocytes out of total CD45 positive splenic lymphocytes. Columns SPI2pos and SPI2neg show average values for all mice clustered into two groups according to being infected with either any of the SPI-2 positive or the SPI-2 negative mutants. * - t-test different from SPI2neg at P < 0.01.

Figure 5

Presence of NK cells in spleen of mice 5 days post infection with the wild type S. Enteritidis (wt) or ΔSPI2 mutant averaged from the animal infections 2, 3 and 4. n.i. - non-infected mice. * - t-test different from the non-infected or ΔSPI2 mutant infected mice at P < 0.01.

Figure 6

Presence of NK cells in spleen of mice 5 days post infection with the wild type S. Enteritidis (wt) or attenuated ΔSPI2, rfaL or lon mutants as determined in the animal infection 2. n.i. - non-infected mice. The NK cells depletion was not specific for the ΔSPI2 mutant but was a general indicator of the mutant's virulence or avirulence. Since there were only 3 animals per group and greater variation was observed among the mice infected with the wild type S. Enteritidis, none of the differences in this experiment reached statistical significance.

Figure 7

Cytokine response in caecum of mice infected with the wild type S. Enteritidis (wt) and ΔSPI2 mutant. n.i. - non-infected mice. * - t-test different from the non-infected mice at P < 0.05.

Figure 8

Distribution of NK cells in spleen, blood and caecal lamina propria of mice infected with the wild type S. Enteritidis (wt) and ΔSPI2 mutant as determined in the animal infection 4. n.i. - non-infected mice. * - t-test different from the non-infected mice at P < 0.01.