Peptidoglycan Endopeptidase Spr of Uropathogenic Escherichia coli Contributes to Kidney Infections and Competitive Fitness During Bladder Colonization

Abstract

Uropathogenic E scherichia coli (UPEC) is the most common pathogen of urinary tract infections (UTIs). Antibiotic therapy is the conventional measure to manage such infections. However, the rapid emergence of antibiotic resistance has reduced the efficacy of antibiotic treatment. Given that the bacterial factors required for the full virulence of the pathogens are potential therapeutic targets, identifying such factors may facilitate the development of novel therapeutic strategies against UPEC UTIs. The peptidoglycan (PG) endopeptidase Spr (also named MepS) is required for PG biogenesis in E. coli. In the present study, we found that Spr deficiency attenuated the ability of UPEC to infect kidneys and induced a fitness defect during bladder colonization in a mouse model of UTI. Based on the liquid chromatography (LC)/mass spectrometry (MS)/MS analysis of the bacterial envelope, spr deletion changed the levels of some envelope-associated proteins, suggesting that Spr deficiency interfere with the components of the bacterial structure. Among the proteins, FliC was significantly downregulated in the spr mutant, which is resulted in reduced motility. Lack of Spr might hinder the function of the flagellar transcriptional factor FlhDC to decrease FliC expression. The motility downregulation contributed to the reduced fitness in urinary tract colonization. Additionally, spr deletion compromised the ability of UPEC to evade complement-mediated attack and to resist intracellular killing of phagocytes, consequently decreasing UPEC bloodstream survival. Spr deficiency also interfered with the UPEC morphological switch from bacillary to filamentous shapes during UTI. It is known that bacterial filamentation protects UPEC from phagocytosis by phagocytes. In conclusion, Spr deficiency was shown to compromise multiple virulence properties of UPEC, leading to attenuation of the pathogen in urinary tract colonization and bloodstream survival. These findings indicate that Spr is a potential antimicrobial target for further studies attempting to develop novel strategies in managing UPEC UTIs.

Keywords: MepS; Spr; bacteremia; flagella; motility; the complement system; urinary tract infections.

Figure 1

Role of spr in Uropathogenic Escherichia coli (UPEC)-induced urinary tract infections (UTIs). (A) The bacterial counts of ∆lacZ-UTI89 and ∆spr-UTI89 in the bladders and kidneys at 48 h after transurethral co-inoculation with equal amounts of these strains (5 × 10⁷ CFU/mouse for each strain). Animal numbers N = 10. (B) The bacterial counts of lacZ::spr∆spr-UTI89 and ∆spr-UTI89 in the bladders and kidneys at 48 h after transurethral co-inoculation with equal amounts of these strains (5 × 10⁷ CFU/mouse for each strain). N = 11. (C) The bacterial counts of WT-UTI89, ∆spr-UTI89, and lacZ::spr∆spr-UTI89 in the bladders and kidneys at 48 h after transurethral inoculation with these strains (1 × 10⁸ CFU/mouse), respectively. N = 8 for each strains. (D) The bacterial counts of Spr-C68A-UTI89 and ∆spr-UTI89 in the bladders and kidneys at 48 h after transurethral co-inoculation with equal amounts of these strains (5 × 10⁷ CFU/mouse for each strain). N = 10. The horizontal bars represent the median values. The dotted line represents the limit of detection. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Figure 2

Effects of spr deletion on the FliC level and motility. (A) Western blot analyses of the FliC levels in the bacterial lysates of UTI89 strains. The levels of OmpA served as the loading control. (B) The motility of the UTI89 strains. The quantified motility of the indicated strains was derived from experiments performed in triplicate and is presented as the means ± standard deviations. ^** p < 0.01; ^*** p < 0.001.

Figure 3

Contribution of bacterial motility to the abilities of the spr-deleted UPEC strains to cause UTIs. (A) The motilities of the UTI89 strains with or without overexpression of FlhDC. The quantification of motilities of the indicated strains. Each quantitative result was derived from experiments in triplicate and presented as the means ± standard deviations. (B) Effect of bacterial motility on the ability of the spr mutant strains to cause UTIs. N = 10. ^** p < 0.01.

Figure 4

Effect of the spr mutation on flagella synthesis. (A) The mRNA expression of the flagellar regulon in WT-UTI89 and ∆spr-UTI89. The transcript levels of the class 1 gene flhD, the class 2 genes (fliA, flgE, flhA, fliF, fliM, fliE, fliT and flgM) and the class 3 genes (fliC and motA) were determined by real-time PCR (qPCR). The transcript levels of the genes in each strain, which were normalized to those of the housekeeping gene gyrB, were presented as the relative levels compared to those of WT-UTI89. The results were derived from experiments in triplicate and are shown as the means ± standard deviations. (B) The promoter activity of the flagellar regulon in WT-UTI89 and ∆spr-UTI89. The levels of β-galactosidase activity of flhD-lacZ, fliA-lacZ, flhA-lacZ, fliM-lacZ, fliT-lacZ, and fliC-lacZ operon fusion are presented as the relative levels compared to those of WT-UTI89. (C) The levels of HA-tagged FlhD and His₆-tagged FlhC proteins in bacteria with and without spr were determined by western blot analysis with mouse anti-HA and anti-His₆ antibodies, respectively. The levels of OmpA served as the loading control. ^* p < 0.05.

Figure 5

Effect of Spr deficiency on the ability of nonmotile UPEC strains to cause UTIs. The horizontal bars and the dotted lines represent the median levels of the bacterial counts and the limit of detection, respectively. N = 10. ^* p < 0.05; ^** p < 0.01.

Figure 6

Effect of spr deletion on the sensitivity to SDS and low pH conditions. (A) The bacterial survival in LB (−) and LB with 5% SDS of the indicated strains was determined after 3 h incubation. (B) The survival in LB at pH 7.5 (pH 7.5) and LB at pH 4.5 (pH 4.5) of the indicated strains was determined after 2 h incubation. The levels of bacterial survival are presented as relative survival rates compared to the bacterial counts of WT-UTI89. The results are shown as the means ± standard deviations (SD). ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Figure 7

Effect of spr deletion on the abilities of UTI89 to survive in macrophages and serum, and the ability to recruit C3b deposition in NHS. (A) The intracellular survival of the UTI89 strains with or without spr in the macrophage cell line RAW264.7. RAW264.7 cells were independently infected with the indicated UTI89 strains and incubated for 30 min. The infected cells were incubated with 100 μg/ml gentamicin for 15 min to kill extracellular bacteria. Then, the cells harboring intracellular bacteria were incubated with 10 μg/ml gentamicin for the remainder of the experiment. After 0 and 24 h of incubation in the 10 μg/ml gentamicin, the bacterial counts were determined by plating on LB agar plates. The survival rates after 24 h incubation were calculated by the ratio of the bacterial counts after 24 h incubation to those after 0 h incubation. The data are presented as relative survival rates compared to that of WT-UTI89. (B) Survival of UTI89 strains with or without spr in 30% normal human serum (NHS) or heat-inactivated normal human serum (HI-NHS). The indicated bacteria were independently incubated with 30% NHS or HI-NHS for 2 and 4 h. The levels of bacterial survival are presented as relative survival rates compared to the bacterial counts of the original inoculums. The results are shown as the means ± standard deviations (SD). (C) The levels of C3b deposition of WT-UTI89 and Δspr-UTI89 were determined by flow cytometry analysis. WT-UTI89 and Δspr-UTI89 were incubated with 30% NHS for 1, 5, and 15 min (experimental group), and then stained with FITC-conjugated anti-C3 antibody. The bacteria in the control group of the C3b deposition were incubated in PBS instead of 30% NHS for 15 min. The percentage of C3b-labeled bacteria as presented as the mean fluorescence intensity (MFI) are shown as the means ± standard deviations (SD). (D) Flow cytometry histogram of C3b deposition on the bacteria after a 15-min incubation in NHS. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Figure 8

Role of Spr in bloodstream survival UPEC. (A) The bacterial blood counts of WT-UTI89 and Δspr-UTI89 in the coinfection mouse model of bacteremia. N = 8. (B) The bacterial blood counts of the spr mutant strain (ΔsprΔlacZ-UTI89) and the spr complement strain lacZ::Δspr-UTI89 in the coinfection mouse model of bacteremia. N = 8. The horizontal bars represent the median values. ^** p < 0.01.

Figure 9

Filamentation of UPEC strains with and without spr in the in vitro flow chamber-based infection model and the mouse model of UTI. (A) Fluorescence microscopic images of WT-UTI89 and ∆spr-UTI89 after interaction with bladder epithelial cells in the FC-based infection model of bladder epithelial cells. Both strains harbored the plasmid pFPV25.1 that constitutively expressed GFP. (B) The bacterial length of WT-UTI89 and ∆spr-UTI89 after incubation in the FC-based infection model and LB medium. For each strain, the length of 120 bacterial cells from three microscopic fields (40 cells/field) was measured. The horizontal bars indicate the median of the bacterial sizes. (C) Fluorescence microscopy analyses of UPEC strains recovered from urine at 12 h post infection. The arrows indicates bacteria. The arrow heads indicate sloughed host cells. (D) Size quantification of WT-UTI89 or ∆spr-UTI89 in the urine samples. The sizes of 120 urine bacteria of each strain from three microscopic fields (40 cells/field) were determined and plotted on the chart. The horizontal bars indicate the median of the bacterial sizes. ^**** p < 0.0001.

Figure 10

Effect of spr deletion on the association and invasion of bladder epithelial cells. The bladder epithelial cells (A) T24 cells and (B) 5637 cells were incubated with the indicated bacteria. The bars indicate relative association or invasion frequencies compared to the wild-type strain UTI89. The results are shown as the means ± standard deviations.