Salmonella Invasion Is Controlled by Competition among Intestinal Chemical Signals

Abstract

The intestine is a complex, ever-changing environment replete with an array of signaling molecules. To colonize such a complex organ, pathogens have adapted to utilize specific cues from the local environment to intricately regulate the expression of their virulence determinants. Salmonella preferentially colonizes the distal ileum, a niche enriched in the metabolite formic acid. Here, we show that the relatively higher concentration of this metabolite in the distal ileum prevents other signals from repressing Salmonella invasion in that region. We show that imported and unmetabolized formic acid functions as a cytoplasmic signal that competitively binds to HilD, the master transcriptional regulator of Salmonella invasion, thus preventing repressive fatty acids from binding to the protein. This results in an increased lifetime of HilD and subsequent derepression of invasion genes. This study demonstrates an important mechanism by which Salmonella utilizes competition among signals in the gut to its advantage as a pathogen. IMPORTANCE Enteric pathogens acutely sense their environment for signals to regulate their virulence functions. We demonstrate here that the enteric pathogen Salmonella utilizes the competition among certain regional intestinal constituents to modulate its virulence determinants in that region. We show that the high concentration of formic acid in the ileum outcompetes other signals and triggers the activation of virulence genes in the ileum. This study shows a delicate spatial and temporal mechanism by which enteric pathogens may utilize the competition among environmental cues to optimize their pathogenicity.

Keywords: Salmonella; enteric pathogens; fatty acids; host-pathogen interactions; infectious disease; pathogenesis mechanisms; virulence regulation.

FIG 1

Formic acid prevents intestinal fatty acids from repressing Salmonella invasion genes. (A) Summary of the experimental setup to analyze the expression of invasion genes. (B to C) Expression of the invasion gene hilA was determined in Salmonella with different treatments (preincubation with 10 mM formic acid followed by either 500 nM c2-HDA, 40 μM oleic, 10 mM propionic acid, or an equal volume of DMSO) shown using a hilA-luxCDABE transcriptional reporter fusion. (B) Luminescence was normalized to bacterial culture density. Data show the mean ± standard deviation (SD) of five replicates. (C) The area under the curve (AUC) for each treatment was calculated. The AUC for DMSO treatment was set to 100%, and others were normalized accordingly. Bars show the normalized AUC ± SD (n = 5). Differences between the indicated treatments were calculated with the Mann-Whitney test. **, P < 0.01. (D) Expression of the invasion gene sipC was evaluated in Salmonella, preincubated with 10 mM formic acid prior to treatment with 10 μM c2-HDA or 40 μM oleic acid, using a lacZY transcriptional reporter fusion by β-galactosidase assay. Bars represent the mean ± SD (n = 4). Differences between treatments were calculated with the Mann-Whitney test. *, P < 0.05. (E) Determination of the formic acid concentration in the intestinal contents of C57BL/6 mice (n = 10). Bars represent the mean ± SD. Differences between formic acid concentrations in ileum versus cecum or colon were calculated with the Mann-Whitney test. **, P < 0.01. (F) The expression of invasion gene hilA was determined in Salmonella treated with 10 μM fatty acids or an equal volume of DMSO and 10 mM formic acid, shown using a hilA-luxCDABE transcriptional reporter fusion. Luminescence was normalized to bacterial culture density. The AUC for each treatment was calculated. The AUC for DMSO treatment was set to 100%, and others were normalized accordingly. Bars show the normalized AUC ± SD (n = 5). Differences between the indicated treatments were calculated with the Mann-Whitney test. ns, not significant.

FIG 2

Import of formic acid, but not its metabolism, is required to rescue the expression of invasion genes. (A and B) Expression of the invasion gene hilA was determined in Salmonella strains with different treatments (preincubation with 10 mM formic acid before adding 500 nM c2-HDA or an equal volume of DMSO) shown using a hilA-luxCDABE transcriptional reporter fusion. Luminescence was normalized to bacterial culture density. The AUC for each treatment was calculated. The AUC for DMSO treatment was set to 100%, and others were normalized accordingly. Bars show the normalized AUC ± SD (n = 5). Differences between the indicated treatments were calculated with the Mann-Whitney test. **, P < 0.01; ns, not significant.

FIG 3

Formic acid derepresses invasion gene expression through HilD. (A to F) The binding of different fatty acids to increasing concentrations of (A to D) wild-type or (E and F) mutant HilD proteins was assessed using ELISA. Nonlinear regression was plotted, and total one-site binding was analyzed to calculate the K_d.

FIG 4

Formic acid prevents repressive fatty acids from disrupting the DNA-binding ability of HilD. (A to C) The effects of c2-HDA, t2-HDA, and oleic acid on the binding of (A) wild-type HilD, (B) HilD^Q290A, or (C) HilD^K293A to hilA promoter DNA with or without preincubation with formic acid were assessed using EMSAs.

FIG 5

Formic acid preserves the stability of HilD in the presence of repressive fatty acids. (A to C) The stability of HilD in Salmonella carrying a chromosomal tetRA-hilD-3×FLAG tag, in the presence of either (A) DMSO, (B) c2-HDA 20 μM, (C) t2-HDA 20 μM, with or without preincubation with formic acid (10 mM) was evaluated by Western blot analysis. The half-lives after each treatment are indicated.

FIG 6

Salmonella utilizes formic acid in the murine ileum to induce invasion-genes. (A) Summary of the model used to analyze invasion gene expression in the murine ileum. (B) Salmonella strains constitutively expressing BFP were assessed for expression of the invasion gene sicA using a sicA-GFP reporter fusion in the inoculum and the ileal contents of C57BL/6 mice (n = 5), by flow cytometry. WT, wild type with PsicA-GFP; no GFP, a wild-type Salmonella strain without the GFP fusion. Bars represent the mean ± SD. The Mann-Whitney test was performed to calculate statistical significance. ***, P < 0.001. (C) Salmonella strains having different null mutations and constitutively expressing BFP were evaluated for expression of the invasion gene sicA using a sicA-GFP reporter fusion in the ileal contents of C57BL/6 mice (n = 5) by flow cytometry. Bars represent the mean ± SD. The Mann-Whitney test was performed to calculate statistical significance. **, P < 0.01; ns, not significant.