CsrA and Cra influence Shigella flexneri pathogenesis

Abstract

Shigella flexneri is a facultative intracellular pathogen that invades and disrupts the colonic epithelium. In order to thrive in the host, S. flexneri must adapt to environmental conditions in the gut and within the eukaryotic cytosol, including variability in the available carbon sources and other nutrients. We examined the roles of the carbon consumption regulators CsrA and Cra in a cell culture model of S. flexneri virulence. CsrA is an activator of glycolysis and a repressor of gluconeogenesis, and a csrA mutant had decreased attachment and invasion of cultured cells. Conversely, Cra represses glycolysis and activates gluconeogenesis, and the cra mutant had an increase in both attachment and invasion compared to the wild-type strain. Both mutants were defective in plaque formation. The importance of the glycolytic pathway in invasion and plaque formation was confirmed by testing the effect of a mutation in the glycolysis gene pfkA. The pfkA mutant was noninvasive and had cell surface alterations as indicated by decreased sensitivity to SDS and an altered lipopolysaccharide profile. The loss of invasion by the csrA and pfkA mutants was due to decreased expression of the S. flexneri virulence factor regulators virF and virB, resulting in decreased production of Shigella invasion plasmid antigens (Ipa). These data indicate that regulation of carbon metabolism and expression of the glycolysis gene pfkA are critical for synthesis of the virulence gene regulators VirF and VirB, and both the glycolytic and gluconeogenic pathways influence steps in S. flexneri invasion and plaque formation.

FIG. 1.

E. coli central carbon metabolism is regulated by CsrA and Cra. (A) The transcriptional regulator Cra inhibits expression of glycolytic genes while activating expression of genes involved in gluconeogenesis. CsrA regulates stability of mRNAs for genes involved in glycolysis and gluconeogenesis. The noncoding RNAs CsrB and CsrC sequester the CsrA protein from its targets. Expression of csrB and csrC is induced by the BarA-UvrY two-component system, which is activated by CsrA. CsrD decreases the stability of both CsrB and CsrC. Based on their genetic relatedness and the data gathered in this study, CsrA and Cra likely regulate metabolism similarly in S. flexneri and E. coli. (Illustration modified from reference with permission of the publisher). (B) Irreversible steps in the metabolic pathways allow for independent regulation of glycolysis and gluconeogenesis. The opposing effects of CsrA and Cra are due in part to their reciprocal regulation of the glycolytic gene pfkA and the gluconeogenic gene ppsA.

FIG. 2.

CsrA and Cra influence S. flexneri attachment to epithelial cells. (A) Equivalent numbers of each strain were added to subconfluent monolayers and allowed to attach for 15 min prior to washing and staining. The total number of bacteria attached to 100 Henle cells was determined for each strain. Strains tested were 2457T (WT), AGS110 (uvrY), AGS120 (csrA), and AGS190 (cra). Values were normalized to the wild-type attachment level for each experiment. Wild-type adherence was 60 to 90 bacteria/100 cells (average, 0.8 bacteria/cell). Data presented are the mean values with standard deviations for three independent experiments. *, P < 0.05, and **, P < 0.01, compared to WT (by Student's t test). (B) Representative images from bacterial invasion of Henle cells by 2457T (WT), AGS110 (uvrY), or AGS190 (cra). Intracellular bacteria were visualized by Giemsa staining and microscopy.

FIG. 3.

Effects of CsrA and PfkA on S. flexneri LPS and sensitivity to SDS. (A) LPS was harvested from 2457T (WT), AGS120 (csrA), and AGS220 (pfkA). LPS samples from equal numbers of cells grown to exponential phase were separated on a 12% tricine-SDS-PAGE gel and visualized by silver staining. “Core” refers to the migration pattern of the LPS core without attached O-antigen repeats. Mode A is the LPS core with chains consisting of approximately 11 to 17 O-antigen repeats. (B) Strains 2457T (WT), AGS120 (csrA), and AGS220 (pfkA) were grown to exponential phase and washed with 1× PBS-D. A final concentration of 0.1% SDS was added to samples at T₀. Absorbance (A₆₅₀) was determined at 0 min, 10 min, and successive 30-min intervals for 4 h. The percent relative A₆₅₀ was calculated as (A₆₅₀ at T_n)/(A₆₅₀ at T₀) × 100. Data presented are the averages of three independent experiments performed in triplicate. Error bars represent standard deviations. Detergent sensitivity of the mutant strains was compared to wild type by using an unpaired Student's t test. **, P < 0.01; *, P < 0.05.

FIG. 4.

Disruptions in glycolysis inhibit Ipa synthesis. (A) Strains tested were 2457T (WT), AGS120 (csrA), AGS220 (pfkA), AGS190 (cra), AGS120/pQCsrA, and AGS220, carrying either the empty vector (pWSK29) or the cloned pfkA gene (pWPkfA). Total cell extracts were prepared from exponential-phase cells and then analyzed by SDS-PAGE and immunoblotting with mouse monoclonal anti-IpaB or mouse monoclonal anti-IpaC antibody. Lysates from equal numbers of cells as determined by optical density were used.

FIG. 5.

Expression levels of the virulence regulators virF and virB are significantly reduced in the csrA and pfkA mutants. Real-time PCR analysis of virF and virB expression was performed using strains S. flexneri 2457T (WT), AGS120 (csrA), AGS220 (pfkA), and AGS220/pWPfkA (PfkI⁺). Values are normalized to dksA in each sample, and results shown are relative to the value obtained for 2457T. Data presented are the averages of three or more independent experiments, and error bars represent 1 standard deviation. Relative expression was compared to the wild type by using Student's t test. *, P < 0.01.

FIG. 6.

Intracellular ATP levels in wild-type, csrA, and pfkA strains of S. flexneri. Samples of bacteria grown for measurement of virulence gene expression (Fig. 5) were lysed, and the intracellular ATP level was measured in a bioluminescence assay. ATP concentrations in the wild-type cells were ∼1.2 mM, based on comparison to ATP standards. Data shown are the averages of three independent experiments. *, P < 0.05 compared to the wild type (by Student's t test).

FIG. 7.

Proposed model for regulation of S. flexneri pathogenesis by CsrA, Cra, and PfkI. Disruptions in glycolysis, caused by mutation of csrA or pfkA, reduce expression of virB and virF, leading to loss of Ipa production. Because all the virulence phenotypes of the csrA mutation are mimicked by the pfkA mutation, the effects of CsrA on attachment and invasion are likely due to its control of pfkA. Decreased expression levels of ipa and TTSS genes likely contribute to the increased resistance of the csrA and pfkA mutants to the detergent SDS. The altered LPS profile may also contribute to detergent resistance in the mutants, and these changes in the cell surface may influence the response of S. flexneri to environmental regulators of virF gene expression. Solid lines indicate known effects, and dashed lines indicate proposed effects based on the data presented.