Enterococcus faecalis Polymicrobial Interactions Facilitate Biofilm Formation, Antibiotic Recalcitrance, and Persistent Colonization of the Catheterized Urinary Tract

Abstract

Indwelling urinary catheters are common in health care settings and can lead to catheter-associated urinary tract infection (CAUTI). Long-term catheterization causes polymicrobial colonization of the catheter and urine, for which the clinical significance is poorly understood. Through prospective assessment of catheter urine colonization, we identified Enterococcus faecalis and Proteus mirabilis as the most prevalent and persistent co-colonizers. Clinical isolates of both species successfully co-colonized in a murine model of CAUTI, and they were observed to co-localize on catheter biofilms during infection. We further demonstrate that P. mirabilis preferentially adheres to E. faecalis during biofilm formation, and that contact-dependent interactions between E. faecalis and P. mirabilis facilitate establishment of a robust biofilm architecture that enhances antimicrobial resistance for both species. E. faecalis may therefore act as a pioneer species on urinary catheters, establishing an ideal surface for persistent colonization by more traditional pathogens such as P. mirabilis.

Keywords: biofilm; catheter; enterococcus faecalis; polymicrobial; proteus mirabilis; urinary tract infection.

Figure 1

Enterococcus faecalis and Proteus mirabilis persistently co-colonize during long-term catheterization. Data represent the colonization status of E. faecalis (right half, green) and P. mirabilis (left half, blue) in weekly urine cultures from baseline (0) to 30 weeks for each of five study participants (labeled A–E). White semi-circles indicate that a urine sample was collected but lacked P. mirabilis or E. faecalis, and gray circles indicate that a urine sample was not collected at that study visit.

Figure 2

E. faecalis and P. mirabilis co-colonize the catheterized urinary tract during experimental infection. The 24 h mouse model of CAUTI using E. faecalis 3143 (circles), E. faecalis OG1RF (triangles), and P. mirabilis HI4320 (squares) either in a single-infection (black symbols) or co-infection (open symbols). Each symbol represents the log₁₀ CFU/organ or catheter from an individual mouse. (A) Bacterial burden of each bacterial species during single-species infection. (B) P. mirabilis colonization during single-species infection compared to co-infection with E. faecalis 3143 or E. faecalis OG1RF. (C) E. faecalis 3143 colonization during single-species infection compared to co-infection with P. mirabilis HI4320. (D) E. faecalis OG1RF colonization during single-species infection compared to co-infection with P. mirabilis HI4320. Dashed lines indicate limit of detection, and error bars indicate the median. Mann–Whitney U test was used to compare groups. Statistical significance is represented by *P < 0.05.

Figure 3

E. faecalis co-localizes with fibrinogen and P. mirabilis co-localizes with E. faecalis in the catheterized bladder and on catheter segments. (A–C) Representative sections of bladders removed 24 hpi from mice that were catheter-implanted and infected with E. faecalis 3143 (A), P. mirabilis HI4320 (B), or both species (C). Bladders were fixed, sectioned and immunostained using antibodies to detect fibrinogen (magenta), E. faecalis 3143 (green) and P. mirabilis HI4320 (red), and DAPI was used to detect cell nuclei (blue). (D) Catheter segments were removed from a subset of mice 24 hpi with E. faecalis 3143 (top row), P. mirabilis HI4320 (middle row), or both species (bottom row). (E) Quantification of E. faecalis 3143 and P. mirabilis HI4320 co-localization with fibrinogen during single-species infections. (F) Quantification of co-localization of E. faecalis 3143 and P. mirabilis HI4320 during co-infections. Catheter segments were stained for fibrinogen (blue) and/or the respective pathogen (E. faecalis in green and P. mirabilis in red). Values represent the means ± SEM derived from co-localization of the catheter segments. The white broken line separates the bladder lumen (L) from the urothelium (U).

Figure 4

Co-culture of E. faecalis with P. mirabilis enhances biofilm biomass in a contact-dependent manner. Static biofilms were established in TSB with 10 mM glucose in 24 well tissue-culture plates (A–G), optically clear 24 glass-bottom plates (H–J), or on glass coverslips (K–P). All biofilms were established for 20 h unless otherwise indicated. (A,C–G) Biofilm biomass was quantified using crystal violet staining, and normalized to the absorbance of crystal violet at OD₅₇₀ for biofilms formed by E. faecalis. (B) Viable biofilm-associated bacteria were quantified by plating to determine log₁₀ CFU/mL of each species. (C) E. faecalis and P. mirabilis were physically separated by a transwell filter, and biofilms formed in the lower compartment were stained with crystal violet as above. Data represent the mean ± standard deviation for at least three independent experiments with at least two replicates each. ^ns = non-significant, ***P < 0.001 by Student’s t test. (H–J) Confocal microscopy of biofilms established with E. faecalis expressing GFP (H), P. mirabilis expressing DsRedDsRed (I), or both species (J). Top panels display a 3D rendering from representative z-stacks of each culture, and bottom panels display a z-slice view through the biofilm. (K–P) Representative scanning electron micrographs of biofilms formed by E. faecalis (K), P. mirabilis (L), or both species (M–P).

Figure 5

Co-culture of E. faecalis with P. mirabilis enhances biofilm 3D architecture and reduces P. mirabilis dispersal. Static biofilms were established on fibrinogen-coated glass-bottom petri dishes in human urine supplemented with 20 mg/mL BSA and imaged every 12 h for a total of 84 h. Images are representative of biofilms from each of the following inocula: E. faecalis 3143 expressing GFP (A), P. mirabilis HI4320 expressing DsRed (B), co-culture of E. faecalis 3143 GFP and P. mirabilis HI4320 DsRed (C), E. faecalis OG1RF GFP (D), and co-culture of E. faecalis OG1RF GFP and P. mirabilis HI4320 DsRed (E). Single-species biofilms are displayed at 10x, 40x, and 100x magnification, and co-culture biofilms are displayed at 10x magnification for each individual channel, and both 10x and 100x magnification for merged images. Concurrence and co-localization of E. faecalis 3143 with P. mirabilis HI4320 (F) and E. faecalis OG1RF with P. mirabilis HI4320 (G) were performed using Pearson’s correlation coefficient (r). The average of r is on the top merge panel. Values represent the means ± SD derived from co-localization at each time point.

Figure 6

E. faecalis and P. mirabilis polymicrobial biofilms provide additional antimicrobial resistance to both species. Static biofilms were established on 24 well plates in TSB-G for 20 h, then treated with the following antimicrobials for 24 h: ampicillin-clavulanate (A), ceftriaxone (B), or trimethoprim (C). Data represent the mean log₁₀ CFU/mL ± SD of P. mirabilis HI4320 (black) or E. faecalis 3143 (white) for single-species biofilms (solid bars) or co-culture biofilms (diagonal lines) from three independent experiments. ***P < 0.001 by Student’s t test.

Figure 7

E. faecalis and M. morganii polymicrobial biofilms exhibit enhanced biomass. Static biofilms were established in TSB with 10 mM glucose in 24 well tissue-culture plates for E. faecalis 3143 (Ef), P. mirabilis HI4320 (Pm), M. morganii TA43 (Mm), P. stuartii BE2467 (Ps), and E. coli CFT073 (Ec) and all co-culture combinations with E. faecalis, P. mirabilis, or M. morganii. M. morganii isolates 206 and 4108 and E. faecalis isolates OG1RF and V587 were also utilized where indicated. All biofilms were established for 20 h unless otherwise indicated. Biofilm biomass was quantified using crystal violet staining and normalized to the absorbance of crystal violet at OD₅₇₀ for single-species biofilms formed by E. faecalis. Data represent the mean ± standard deviation for at least three independent experiments with at least two replicates each. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test.