d-Serine Degradation by Proteus mirabilis Contributes to Fitness during Single-Species and Polymicrobial Catheter-Associated Urinary Tract Infection

Proteus mirabilis strain HI4320 (WT) and isogenic mutants for dsdA, dsdC, or dsdX were cultured in LB broth overnight and diluted 1:100 into fresh LB (A), PMSM (B), PMSM containing 10 mM d-serine as the sole source of carbon (C), PMSM containing 10 mM d-serine as the sole source of nitrogen (D), PMSM containing 10 mM l-serine as the sole source of carbon (E), or PMSM containing 10 mM l-serine as the sole source of nitrogen (F), and growth was assessed as described in the legend for Fig. 1. Error bars represent means ± standard deviations (SDs) from at least 6 replicates, and graphs are representative of 3 independent experiments. In PMSM with d-serine as the sole source of carbon or nitrogen, the WT exhibited significantly increased growth compared to that of the mutants by paired two-way ANOVA with Dunnett’s test for multiple comparisons (P < 0.001).

d-Serine toxicity perturbs growth of P. mirabilis in minimal medium with a rich carbon source. Proteus mirabilis strain HI4320 (WT) (A and C) and the dsdA mutant (B and D) were cultured in LB broth overnight and diluted 1:100 into the standard formulation of PMSM supplemented with 10 mM d-serine (A and B) or 10 mM d-serine and either 1 mM calcium pantothenic acid or 1 mM l-serine (C and D). Growth was assessed as described in the legend for Fig. 1. Error bars represent means ± standard deviations (SDs) with at least 6 replicates, and graphs are representative of 4 independent experiments. Visible differences in growth curves for the dsdA mutant were confirmed by paired two-way ANOVA with Dunnett’s test for multiple comparisons (P < 0.05).

d-Serine does not provide a fitness advantage during growth in urine. Proteus mirabilis strain HI4320 (WT) and isogenic mutants for dsdA, dsdC, or dsdX were cultured in LB broth overnight and diluted 1:100 into pooled filter-sterilized urine from female humans (A to E) or female CBA/J mice (F), and incubated at 37°C with aeration for 6 h. (A) Growth of the WT and each mutant was assessed at hourly intervals by sampling independent urine cultures and plating to determine bacterial CFU/ml. Error bars represent means ± SDs from three independent experiments; no significant differences were identified by two-way ANOVA with multiple-comparison test. (B) The impact of d-serine supplementation on growth was assessed for WT and dsdA strains by adding 10 mM d-serine to urine, and growth was assessed as described above. Symbols indicate bacterial CFU/ml for one independent experiment. The relative fitness of dsdA (C), dsdC (D), and dsdX (E) mutants was determined by inoculating urine with a 1:1 mixture of mutant/WT and assessing the CFU of each at hourly intervals. A competitive index (CI) was calculated at each time point by dividing the ratio of mutant/WT for a given time point by the ratio of mutant/WT at time zero. Each symbol represents the log₁₀ CI for one independent experiment. (F) The relative fitness of the dsdA mutant was also assessed in mouse urine, as above. Each symbol represents the log₁₀ CI for one independent experiment, error bars represent the medians, and the dashed lines indicate log₁₀ CI of 0 (the expected value if the ratio of mutant/WT is 1:1). No significant differences in CIs were identified by Wilcoxon signed-rank test.

Loss of d-serine utilization does not impact P. mirabilis motility or urease activity. Proteus mirabilis strain HI4320 (WT) and isogenic mutants for dsdA, dsdC, or dsdX were cultured in LB broth overnight and stab inoculated into motility agar (A), inoculated onto the surfaces of swarm agar plates (B), or subcultured in human urine for measurement of urease activity (C). Motility diameters were measured in millimeters after 16 h of incubation at 30°C (A) or 37°C (B); error bars represent means ± SDs from three independent experiments with at least 3 replicates each. *, P < 0.05, ***, P < 0.001 by two-way ANOVA with multiple-comparison test. (C) Urease activity was measured at 30-min intervals; error bars represent means ± SDs from three independent experiments. No significant differences were identified by two-way ANOVA.

d-Serine utilization contributes to P. mirabilis fitness in a murine model of CAUTI. Proteus mirabilis strain HI4320 (WT) and the dsdA mutant were cultured in LB broth overnight, washed once in PBS, and adjusted to 2 × 10⁶ CFU/ml. (A) CBA/J mice were transurethrally inoculated with 50 µl of either WT (n = 10) or dsdA (n = 11) strains at 1 × 10⁵ CFU, and a 4-mm segment of silicone catheter tubing was advanced into the bladder during inoculation. Mice were euthanized 96 h postinoculation (hpi), and bacterial burden was determined in the urine, bladders, kidneys, and spleens. Each symbol represents the log₁₀ CFU per milliliter of urine or gram of tissue from an individual mouse, gray bars represent the medians, and the dashed line indicates the limit of detection. *, P < 0.05 by nonparametric Mann-Whitney test. (B and C) CBA/J mice (n = 11) were inoculated as described above with a 1:1 mixture of WT and dsdA strains to determine fitness during cochallenge. (B) Each symbol represents the log₁₀ CFU per milliliter of urine or gram of tissue from an individual mouse, with the WT and dsdA CFU connected by a black line. Gray bars represent the means, and the dashed line indicates the limit of detection. (C) A competitive index was calculated as described above. Each symbol represents the log₁₀ CI for an individual mouse, error bars represent the medians and the dashed line indicates log₁₀ CI = 0 (the expected value if the ratio of mutant/WT is 1:1). *, P < 0.05, **, P < 0.01 by Wilcoxon signed-rank test.

Not all uropathogens can utilize d-serine as a sole source of carbon or nitrogen. Proteus mirabilis strain HI4320 (WT), Escherichia coli strain CFT073, Providencia stuartii strain BE2467, and Morganella morganii strain TA43 were cultured in LB broth overnight and diluted 1:100 into the standard formulation of PMSM (A), PMSM containing 10 mM d-serine as the sole carbon source (B), and PMSM containing 10 mM d-serine as the sole nitrogen source (C). Growth was assessed as described above. Error bars represent means ± standard deviations (SD) from 3 experiments with at least 6 replicates each. Growth of P. stuartii was significantly lower than each of the other species in panels B and C as determined by paired two-way ANOVA with Dunnett’s test for multiple comparisons (P < 0.05).

Not all uropathogens degrade d-serine during growth in human urine. Proteus mirabilis strain HI4320 (WT), E. coli CFT073 (Ec), M. morganii TA43 (Mm), and P. stuartii BE2467 (Ps) were cultured in LB broth overnight and E. faecalis 3143 (Ef) was cultured in BHI overnight. Overnight cultures were diluted 1:100 into pooled filter-sterilized urine from female donors that had been diluted 1:1 with sterile saline and incubated for 18 h at 37°C with aeration, and d- and l-serine were quantified using a fluorometric detection kit for comparison to the starting urine pool. Bars represent total serine concentration, divided into d-serine and l-serine for three independent replicates. ns, P > 0.05, ***, P < 0.001 by two-way ANOVA.

d-Serine utilization provides a fitness advantage to P. mirabilis during polymicrobial CAUTI. Proteus mirabilis strain HI4320 (WT), the dsdA mutant (dsdA), E. coli CFT073 (Ec), P. stuartii BE2467 (Ps), and M. morganii TA43 (Mm) were cultured in LB broth overnight, and E. faecalis 3143 (Ef) was cultured in BHI overnight. All cultures were washed once in PBS and adjusted to 2 × 10⁶ CFU/ml. CBA/J mice were transurethrally inoculated with 50 µl of a mixture containing 5 × 10⁴ CFU of a 1:1 mixture of dsdA and WT P. mirabilis and 5 × 10⁴ CFU of either E. coli CFT073 (A and B,; n = 8), M. morganii TA43 (C and D; n = 7), P. stuartii BE2467 (E and F; n = 8), or E. faecalis 3143 (G and H; n = 7). A 4-mm segment of silicone catheter tubing was advanced into the bladder during inoculation, and mice were euthanized 96 h postinoculation (hpi) for determination of bacterial burden in the urine, bladders, kidneys, and spleens. (A, C, E, and G) Each symbol represents the log₁₀ CFU per milliliter of urine or gram of tissue recovered from an individual mouse, connected by black lines: triangles represent the coinfecting species, black circles represent the WT P. mirabilis, and open blue circles represent the dsdA mutant. Gray bars represent the means, and the dashed lines indicate the limits of detection. (B, D, F, and H) Each symbol represents the log₁₀ CI for the dsdA mutant recovered from an individual mouse for which CFU were above the limit of detection, error bars represent the medians and the dashed lines indicate log₁₀ CI = 0 (the expected value if the ratio of mutant/WT is 1:1). *, P < 0.05, **, P < 0.01 by Wilcoxon signed-rank test.