The Cpx stress response system potentiates the fitness and virulence of uropathogenic Escherichia coli

Abstract

Strains of uropathogenic Escherichia coli (UPEC) are the primary cause of urinary tract infections, representing one of the most widespread and successful groups of pathogens on the planet. To colonize and persist within the urinary tract, UPEC must be able to sense and respond appropriately to environmental stresses, many of which can compromise the bacterial envelope. The Cpx two-component envelope stress response system is comprised of the inner membrane histidine kinase CpxA, the cytosolic response regulator CpxR, and the periplasmic auxiliary factor CpxP. Here, by using deletion mutants along with mouse and zebrafish infection models, we show that the Cpx system is critical to the fitness and virulence of two reference UPEC strains, the cystitis isolate UTI89 and the urosepsis isolate CFT073. Specifically, deletion of the cpxRA operon impaired the ability of UTI89 to colonize the murine bladder and greatly reduced the virulence of CFT073 during both systemic and localized infections within zebrafish embryos. These defects coincided with diminished host cell invasion by UTI89 and increased sensitivity of both strains to complement-mediated killing and the aminoglycoside antibiotic amikacin. Results obtained with the cpxP deletion mutants were more complicated, indicating variable strain-dependent and niche-specific requirements for this well-conserved auxiliary factor.

Fig 1

Deletion of cpxP enhances Cpx activation in both CFT073 and UTI89. Graphs indicate expression levels (± standard deviations) of the luxCDABE operon driven by the cpxP promoter in wild-type UTI89 (A) and wild-type CFT073 (B) and their mutant derivatives, following growth to early stationary phase in modified M9 medium. The pNLP1-lux plasmid carries a promoterless luxCDABE operon. Each graph shows the means ± standard errors of the means of three independent experiments performed in triplicate.

Fig 2

The Cpx system promotes UPEC fitness within the bladder. Adult female CBA/J mice were infected via catheterization with wild-type UTI89, UTI89ΔcpxP, or UTI89ΔcpxRA in noncompetitive (A) and competitive (B and C) assays. Graphs show bacterial titers present in the bladder at 3 days postinoculation. Bars denote median values for each group (n ≥ 12 mice). The data in panel B are graphed in panel C as competitive indices. P values were determined using the Mann-Whitney U test (A) or Wilcoxon-matched paired signed rank test (B). ns, no significant difference.

Fig 3

CFT073 and UTI89 mutants lacking either cpxP or cpxRA grow normally in LB broth and modified M9 medim. (A and B) Growth of wild-type UTI89 and associated ΔcpxP and ΔcpxRA mutants grown in LB broth (A) and modified M9 medium (B). Graphs are representative of at least three independent experiments performed in quadruplicate. (C to F) Results of competitive growth assays carried out in modified M9 medium with wild-type CFT073 and UTI89 versus isogenic ΔcpxRA or ΔcpxP mutants, as indicated. Data are presented as box-and-whiskers plots, with means ± the minimum and maximum values from three independent experiments.

Fig 4

Cpx effects on bladder cell invasion by UTI89. Human bladder epithelial cells were infected with the indicated strains for 2 h, followed by an additional 2-h incubation in medium containing gentamicin (100 μg/ml). Graphs show the total cell-associated bacterial titers prior to addition of gentamicin (A) and for gentamicin-protected, intracellular bacteria (B). Data are expressed relative to wild-type UTI89 (black bars) or UTI89 carrying the control plasmid pGEN-MCS (gray bars) as the means ± standard errors of the means of at least three independent experiments performed in triplicate. The indicated P values were calculated using Student's t test.

Fig 5

The Cpx system is required for full virulence of CFT073 in zebrafish embryos. The PC (A) or blood (B) of 48-hpf zebrafish embryos was inoculated with 500 to 1,000 CFU of wild-type CFT073, CFT073ΔcpxP, or CFT073ΔcpxRA, as indicated. Fish were scored for death at 0, 24, 48, and 72 h postinoculation. Data are expressed as the percent survival over time (n ≥ 17 embryos). P ≤ 0.0008 for the ΔcpxP and ΔcpxRA mutants versus control wild-type CFT073, as determined using Mantel-Cox log rank tests.

Fig 6

Cpx components have strain-dependent effects on serum resistance. About 5 × 10⁴ CFU of wild-type UTI89 or CFT073 or their mutant derivatives were incubated at 37°C with gentle shaking in modified M9 medium containing 20% human serum (A) or 20% heat-inactivated serum (C). After 2.5 h, surviving bacteria were enumerated by plating serial dilutions. (B) Similar assays with 20% serum were performed using strains carrying plasmids pJLJ41 or pJLJ42 or the control empty vector, pGEN-MCS, as indicated. Data are presented relative to the wild-type strains as the means ± standard errors of the means of at least four independent experiments. In panel B, the control wild-type strains carried pGEN-MCS. *, P < 0.007; **, P = 0.02 (determined with Student's t test).