Host Stress Drives Salmonella Recrudescence

Abstract

Host stress is well known to result in flare-ups of many bacterial, viral and parasitic infections. The mechanism by which host stress is exploited to increase pathogen loads, is poorly understood. Here we show that Salmonella enterica subspecies enterica serovar Typhimurium employs a dedicated mechanism, driven by the scsA gene, to respond to the host stress hormone cortisol. Through this mechanism, cortisol increases Salmonella proliferation inside macrophages, resulting in increased intestinal infection loads in DBA/2J mice. ScsA directs overall Salmonella virulence gene expression under conditions that mimic the intramacrophagic environment of Salmonella, and stimulates the host cytoskeletal alterations that are required for increased Salmonella proliferation inside cortisol exposed macrophages. We thus provide evidence that in a stressed host, the complex interplay between a pathogen and its host endocrine and innate immune system increases intestinal pathogen loads to facilitate pathogen dispersal.

Figure 1. Cytoskeletal rearrangements in porcine macrophages are required for increased intracellular proliferation of Salmonella.

(a) Number of intracellular Salmonella Typhimurium bacteria in porcine macrophages that were treated with unsupplemented medium, 2 μM cytochalasin D (inhibitor of F-actin polymerization), 20 μM nocodazole (inhibitor for microtubule formation) or the combination of both, for 24 hours after invasion. White bars illustrate medium without cortisol, grey bars represent medium with 1 μM cortisol and black bars indicate medium with 10 μM cortisol. An asterisk (*) refers to a significant difference compared to the respective controls without cortisol (P ≤ 0.05). Panels (b,c) represent TEM images, 6 hours after control (b) or 1 μM cortisol (c) treatment of Salmonella Typhimurium WT-infected porcine macrophages, showing SCV formation (scale bar, 1 μm).

Figure 2. ScsA mediates cortisol-induced increase in proliferation of Salmonella, both in vitro and in vivo.

(a) Shown is the effect of cortisol on the log₁₀ values + standard deviation of intracellular Salmonella Typhimurium WT, ΔscsA, ΔscsA^c and ΔscsA^pGV1106 bacteria in porcine macrophages. An asterisk (*) refers to a significant difference compared to the condition without cortisol (P ≤ 0.05). Panels (b,c) illustrate the effect of dexamethasone exposure on the recovery of Salmonella WT and ΔscsA from organs of DBA/2J mice. The log₁₀ value of the CFU/gram sample is given as the mean + standard error of the mean. Significant differences are signed with a, b, c (P ≤ 0.017) and a tendency with B (P ≤ 0.033).

Figure 3. Cortisol effect on gene expression during intracellular environment mimicking conditions.

For both WT and ΔscsA strain, the percentage of genes up or down regulated by cortisol is shown quantitatively compared to the total number of genes belonging to the (a) islands (894 genes) or (b) regulatory function (48 genes) group. Grey bars represent the percentage of genes of which the cortisol effect in WT is the opposite in ΔscsA strain. Striped bars reflect the genes that act similar after cortisol treatment. Black bars depict the genes that are up or down regulated after cortisol treatment in WT but not in ΔscsA, or vice versa. Microarray data have been deposited in the Gene Expression Omnibus at NCBI with series accession numbers GSE55430.

Figure 4. Effect of cortisol on the growth of Salmonella Typhimurium in cecal contents.

The log₁₀ values of the CFU/ml + standard deviation are given at different time points (t = 0, 4.5, 6.5, 20, 28 hours under microaerobic conditions at 37 °C) after inoculation in cecal contents. Salmonella Typhimurium (a) WT and (b) ΔscsA growth was examined in cecal contents with (1 μM or 1 mM) or without cortisol.

Figure 5. Effect of glucocorticoids on the number of Salmonella positive macrophages.

Panel (a) shows the effect of dexamethasone exposure on the colocalization of Salmonella WT and ΔscsA with F4/80+ cells in the cecal wall of DBA/2J mice. Given is the percentage of the Salmonella positive macrophages + standard error of the mean compared to the total number of counted macrophages. Significant differences are signed with a, c (P ≤ 0.017) and a tendency with B (P ≤ 0.033). Panel (b) indicates a significant (P ≤ 0.05) positive correlation between the percentage of Salmonella positive macrophages per mouse and their respective log₁₀ value of the CFU/gram cecal wall (correlation coefficient = 0.62).

Figure 6. Salmonella colocalization with F4/80+ cells in the cecum.

Cecal sections of Salmonella-infected DBA/2J mice were stained for immunofluorescence using an F4/80 antibody targeting macrophages (green) and a Salmonella O–4 antibody (red). Nuclei were stained with DAPI (blue). Shown are representative images demonstrating (a) colocalization of Salmonella and F4/80+ cells or (b) Salmonella negative F4/80+ cells (scale bar, 20 μm).