Redundancy and specificity of Escherichia coli iron acquisition systems during urinary tract infection - PubMed (original) (raw)
Redundancy and specificity of Escherichia coli iron acquisition systems during urinary tract infection
Erin C Garcia et al. Infect Immun. 2011 Mar.
Abstract
Uropathogenic Escherichia coli (UPEC), the predominant cause of uncomplicated urinary tract infection (UTI), utilizes an array of outer membrane iron receptors to facilitate siderophore and heme import from within the iron-limited urinary tract. While these systems are required for UPEC in vivo fitness and are assumed to be functionally redundant, the relative contributions of specific receptors to pathogenesis are unknown. To delineate the relative roles of distinct UPEC iron acquisition systems in UTI, isogenic mutants in UPEC strain CFT073 or 536 lacking individual receptors were competed against one another in vivo in a series of mixed infections. When combinations of up to four mutants were coinoculated using a CBA/J mouse model of ascending UTI, catecholate receptor mutants (ΔfepA, Δiha, and ΔiroN mutants) were equally fit, suggesting redundant function. However, noncatecholate siderophore receptor mutants, including the ΔiutA aerobactin receptor mutant and the ΔfyuA yersiniabactin receptor mutant, were frequently outcompeted by coinoculated mutants, indicating that these systems contribute more significantly to UPEC iron acquisition in vivo. A tissue-specific preference for heme acquisition was also observed, as a heme uptake-deficient Δhma ΔchuA double mutant was outcompeted by siderophore receptor mutants specifically during kidney colonization. The relative contribution of each receptor to UTI only partially correlated with in vivo levels of receptor gene expression, indicating that other factors likely contributed to the observed fitness differences. Overall, our results suggest that UPEC iron receptors provide both functional redundancy and niche specificity for this pathogen as it colonizes distinct sites within the urinary tract.
Figures
FIG. 1.
Heme and siderophore receptor expression levels by qPCR in receptor mutants. RNA was isolated from wild-type CFT073 and receptor mutants (indicated on x axis) cultured in LB containing 200 μM DPD. Transcript levels were measured in cDNA preparations from each strain and normalized to the gapA level, and results are shown as fold change relative to wild-type levels. Dashed lines represent 2-fold change and bars show the means.
FIG. 2.
Fitness of E. coli CFT073 catecholate siderophore receptor mutants during mixed infections. (A) Competitive indices from cochallenge in CBA/J mice transurethrally challenged with a mixture of equal amounts of Δ_fepA_ and Δ_iha_ mutants (n = 7). The total inoculum equaled 1 × 108 CFU/mouse. (B to D) Fitness indices calculated from the cochallenge results shown in panel A (B) or from mixed infections in mice inoculated with a mixture of Δ_fepA_, Δ_iha_, and Δ_iroN_ mutants (n = 10) (C) or wild-type CFT073 (WT) and Δ_cirA_ and Δ_fiu_ mutants (n = 9) (D). Fitness indices at 72 hpi were calculated by dividing the fraction of each mutant in the output (CFU/g tissue) by the fraction of each mutant in the input (CFU/ml inoculum). Bars represent the medians, and the dashed line indicates the theoretical fitness index of 1.0 (log101 = 0). *, P = 0.0333.
FIG. 3.
In vivo and in vitro competitive fitness of E. coli CFT073 hydroxamate receptor mutants. (A) In vivo competition in CBA/J mice transurethrally challenged with a mixture of equal amounts of Δ_fhuA_, Δ_fhuE_, and Δ_iutA_ mutants (n = 10). Fitness indices at 72 hpi were calculated by dividing the fraction of each mutant in the output (CFU/g tissue) by the fraction of each mutant in the input (CFU/ml inoculum). Bars represent the medians, and the dashed line indicates the theoretical fitness index of 1.0. The P value displayed indicates significance by the Mann-Whitney test. *, P = 0.0371, and **, P = 0.0020, by the Wilcoxon signed-rank test. (B) In vitro culture competition of Δ_fhuA_, Δ_fhuE_, and Δ_iutA_ mutants cultured in LB containing 200 μM DPD. After inoculation, cultures were passaged into fresh medium every 8 (1:100 dilution) or 16 (1:500 dilution) h, indicated by arrows. Means of triplicate cultures are plotted and error bars show standard errors of the means. Fitness indices, calculated as described for panel A, are shown.
FIG. 4.
Fitness of E. coli CFT073 siderophore and heme receptor mutants during mixed infection. In vivo competition in CBA/J mice transurethrally challenged with a mixture of Δ_iutA_, Δ_iroN_, and Δ_fepA_ mutants (n = 10) (A) or Δ_iroN_, Δ_iha_ Δ_fepA_, Δ_iutA_, and Δ_hma chuA_ mutants (n = 17) (B). Fitness indices at 72 hpi were calculated by dividing the fraction of each mutant in the output (CFU/g tissue) by the fraction of each mutant in the input (CFU/ml inoculum). Bars represent the medians, and the dashed line indicates the theoretical fitness index of 1.0. *, P < 0.02; **, P < 0.01.
FIG. 5.
Relative contribution of yersiniabactin uptake to E. coli UTI. (A) Growth of the E. coli 536 wild type carrying pGEN vector (top left) or the Δ_hma_ Δ_chuA_536 mutant carrying pGEN vector (top right), pnative_chuA_ (bottom left), or pnative_hma_ (bottom right) around wells containing (clockwise from top) 100 μM FeCl2, concentrated supernatant from the Δ_entF_536 mutant, concentrated supernatant from the Δ_ybtS_536 mutant, or 100 μM hemin. Bars indicate zones of growth. (B) Growth of the E. coli Δ_fyuA_536 mutant carrying pGEN vector (left) or pnative_fyuA_ (right) around wells containing 100 μM FeCl2 or concentrated supernatants from the Δ_ybtS_536 or the Δ_entF_536 mutant. Bars indicate zones of growth. (C and D) In vivo competition in CBA/J mice transurethrally challenged with a mixture of E. coli 536 Δ_fhuA_536, Δ_fhuE_536, Δ_fyuA_536, and Δ_hma_ Δ_chuA_536 mutants (n = 16) (C) or Δ_iroN_536, Δ_fepA_536, Δ_fyuA_536, and Δ_hma_ Δ_chuA_536 mutants (n = 13) (D). Fitness indices at 72 hpi were calculated by dividing the fraction of each mutant in the output (CFU/g tissue) by the fraction of each mutant in the input (CFU/ml inoculum). Bars represent the medians, and the dashed line indicates the theoretical fitness index of 1.0. *, P < 0.05; **, P < 0.01.
FIG. 6.
Expression of iron receptor genes by UPEC during murine experimental UTI. (A) Relative quantities of UPEC iron receptor transcripts, normalized to that of the gapA transcript, present in the urine of mice inoculated transurethrally with 1 × 108 CFU of wild-type CFT073 or 536. Bars represent the means of duplicate or triplicate samples, with each sample pooled from five mice over 6 to 72 hpi, and error bars show the standard errors of the means. (B) Mean relative transcript quantity shown in panel A plotted against the median fitness index of the corresponding mutant during bladder or kidney infection by CFT073 or 536. The FI value for each gene/mutant is the median of all mixed infections shown. For the CFT073 Δ_fepA_ mutant, the FI value is the median of the Δ_fepA_ and Δ_iha_ Δ_fepA_ mutant FIs combined. Best-fit lines, Spearman correlation coefficients (r), and corresponding P values are shown.
References
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources
Medical