Fig 1.
Diagram of the oxidative and reductive TCA pathways.
Reactions and targeted genes (sdhB, fumA, fumC) of the oxidative TCA cycle are denoted by black arrows and black font. Reactions and targeted genes (fumB, frdA) of the branched, reductive TCA pathway are denoted by red arrows and red font. NADH and FADH2 are generated during reactions of the oxidative TCA cycle. Arrows indicate the direction of each reaction.
Fig 2.
FumC is required for in vivo fitness of CFT073.
In vivo Log10 competitive indices (CI) were determined following 48 h co-challenge infections in female CBA/J mice with a 1:1 ratio of (A) double mutant fumAfumB and wild-type CFT073, (B) double mutant fumAfumC and wild-type CFT073, (C) double mutant fumBfumC and wild-type CFT073, and (D) triple mutant fumAfumBfumC and wild-type CFT073. (E) In vivo Log10 CI was determined following infection with a 1:1:1 ratio of single mutants fumC, fumA and fumB. Each symbol represents bladder (closed circle) and kidneys (open square) from an individual mouse. Log10 CI>1 indicates a fitness advantage and Log10 CI<1 indicates a fitness defect of the mutant. Significant differences in colonization (**P<0.01, ***P<0.001) were determined with the Wilcoxon signed-rank test. Triple co-challenge data was analyzed using a one-way ANOVA with multiple comparisons (**P<0.01).
Fig 3.
Presence of FRD in the absence of SDH, creates a fitness defect in vivo.
In vivo Log10 competitive indices (CI) were determined following 48 h co-challenge infections in female CBA/J mice with a 1:1 ratio of (A) single mutant strain sdhB and single mutant strain frdA, (B) double mutant sdhBfrdA and wild-type CFT073, (C) double mutant strain sdhBfrdA and single mutant strain frdA and (D) double mutant strain sdhBfrdA complemented with pGEN-frdABCD and wild-type CFT073 containing pGEN-MCS. Each symbol represents bladder (closed circle) and kidneys (open square) from an individual mouse. Log10 CI>1 indicates a fitness advantage and Log10 CI<1 indicates a fitness defect of the mutant. Significant differences in colonization (*P<0.05, **P<0.01, ***P<0.001) were determined with the Wilcoxon signed-rank test.
Fig 4.
Optimal E. coli growth in vitro requires the presence of at least one fumarase.
Growth of wild-type CFT073, fumarase single mutant strains fumA, fumB, and fumC, double mutant strains fumAB, fumAC, fumBC and triple mutant strain fumABC in (A) LB medium, (B) defined medium containing 0.2% glycerol as the sole carbon source (C) defined medium containing 0.2% glucose as the sole carbon source, and (D) human urine. OD600 values were recorded each hour and the mean of three independent trails is plotted.
Fig 5.
In vitro growth for E. coli SDH and FRD mutant strains.
Growth of wild-type CFT073 and succinate dehydrogenase mutant strain sdhB, fumarate reductase mutant strain frdA, and double mutant strain sdhBfrdA in (A) LB medium, (B) defined medium containing 0.2% glycerol as the sole carbon source, (C) defined medium containing 0.2% glucose as the sole carbon source, and (D) human urine. Mean OD600 values determined at each hour time point are plotted against time.
Fig 6.
The growth defect of fumC can be rescued with addition of exogenous iron.
Wild-type CFT073 and (A) single fumarase mutant strains fumA, fumB, and fumC, (B) double and triple fumarase mutant strains fumAB, fumAC, fumBC, fumABC and (C) SDH and FRD mutant strains were examined in iron-free defined medium containing 0.2% glucose with and without 36 μM FeCl3. OD600 was recorded every hour and symbols indicate the mean values of three trials. Black symbols indicate the absence of iron and white symbols denote medium containing iron.
Fig 7.
Loss of FumC results in acidic end products.
Wild-type CFT073, single, double, and triple fumarase mutants were examined on LB agar containing phenol red (plain), LB agar containing phenol red and 0.2% glucose, LB agar containing phenol red and 0.2% glycerol, LB agar containing phenol red, 0.2% glucose, and 0.2% fumarate, and LB agar containing phenol red, 0.2% glycerol, and 0.2% fumarate. Un-inoculated phenol red agar plates with orientation of strains are located on the right and were treated under the same conditions. Alkaline pH is pink, neutral pH is orange, and acidic pH is yellow. (A) Plates were spotted with bacterial culture, incubated under anaerobic conditions at 37°C for 24 h, imaged, and then transitioned to aerobic conditions at 25°C and imaged at 6 h and 24 h. (B) Plates were spotted with bacterial culture, incubated under aerobic conditions at 37°C, imaged at 6 and 24 h, transitioned to 25°C and imaged at 48 h.
Fig 8.
Disruption of SDH results in acidic end products.
Wild-type CFT073, single mutant strains, sdhB and frdA, and double mutant strain sdhBfrdA were examined on LB agar containing phenol red (plain), LB agar containing phenol red and 0.2% glucose, LB agar containing phenol red and 0.2% glycerol, LB agar containing phenol red, 0.2% glucose, and 0.2% fumarate, and LB agar containing phenol red, 0.2% glycerol, and 0.2% fumarate. Un-inoculated phenol red agar plates with orientation of strains are located on the right and treated under the same conditions. Alkaline pH is pink, neutral pH is orange, and acidic pH is yellow. (A) Plates were spotted with bacterial culture, incubated under anaerobic conditions at 37°C for 24 h, imaged, and then transitioned to aerobic conditions at 25°C and imaged at 6 h and 24 h. (B) Plates were spotted with bacterial culture, incubated under aerobic conditions at 37°C, imaged at 6 h and 24 h, transitioned to 25°C and imaged at 48 h.
Fig 9.
Disruption of both the oxidative and reductive TCA pathway results in sensitivity to oxidative stress.
(A) Wild-type CFT073 and the oxidative and reductive TCA pathway mutant strains were examined in LB medium containing 0.3% H2O2. Samples were collected every 10 minutes, diluted, and plated on LB agar to determine CFU. Colored symbols indicate the mean of three independent biological replicates. Two-way ANOVA was use to assess statistical differences at each time point (***P<0.001, ****P<0.0001). (B) Wild-type CFT073 and the oxidative and reductive TCA pathway mutant strains were incubated in 100 mM MES buffered LB, pH 2.5, for 60 minutes, diluted, and plated on LB agar to determine CFU. The dashed line represents the intended input CFU/ml at time zero. The mean is indicated by the bar height and the error bars display SEM of three biological replicates.
Fig 10.
Absence of FumC leads to decreased susceptibility to bactericidal antibiotics.
Wild-type CFT073 and the oxidative and reductive TCA pathway mutant strains were examined in LB medium containing (A) chloramphenicol 10 μg/ml (B) trimethoprim 5 μg/ml or (C) tetracycline 10 μg/ml. Samples were collected over the course of 180 minutes, diluted and plated on LB agar to determine CFU. Colored dots signify the mean of three biological replicates. The Minimum Bacterial Concentration (MBC) of (D) ciprofloxacin 0.002–32 μg/ml (E) ampicillin 0.016–256 μg/ml or (F) streptomycin 0.064–24 μg/ml observed for wild-type CFT073 and the mutant strains were recorded following overnight incubation on Mueller-Hinton agar with antibiotic test strips. The values shown are averages from two independent trials.
Fig 11.
FumC is required for motility in CFT073.
(A) Wild-type CFT073 and each oxidative and reductive TCA pathway mutant strain was stabbed into semi-soft motility agar and incubated overnight at 30°C for 16 h. Images of representative motility agar plates following incubation with inoculated strains. Outer swim edge is indicated by a white line. (B) Bars represent the average swimming diameters (in mm) of biological triplicates. Error bars represent the standard errors of the means (SEM). Statistical significance was determined by one-way ANOVA (***P <0.001, ****P <0.0001). (C) The invertible element (IE) assay was performed on strains grown statically in LB at 37°C for 18 h and standardized to OD600 = 0.5. Bars depict percent population of bacteria with the fim promoter within the IE positioned in the on orientation (black bar) or off orientation (white bar) as calculated with pixel density values for bands of PCR products in (Fig 9A).
Table 1.
Bacterial strains used in this study.