PMC

EbpA and EbpB expression in nonpiliated strains. Western blot analyses were performed after SDS-PAGE of the indicated bacterial strains and fractions using anti-EbpA (A and D) or anti-EbpB (B, C, and E) sera. Open arrowheads indicate the ~140-kDa EbpA and EbpB species observed in the EbpC⁻ strain (A to C) and the EbpABC⁻/p-ebpAB strain (D and E). Asterisks show the ~100-kDa EbpA monomer in EbpC⁻, EbpBC⁻, and SrtC⁻ SrtA⁻ strains (A) and in the EbpABC⁻/p-ebpAB strain (D). Hash marks show the EbpB monomer in EbpC⁻, EbpAC⁻, and SrtC⁻ SrtA⁻ strains (B and C) and in the EbpABC⁻/p-ebpAB strain (E). Brackets indicate pilus HMWLs observed in OG1RF and the EbpABC⁻/p-ebpABC strain.

The SrtC⁻ mutant is attenuated in experimental CAUTI. Mice were infected with ~2 × 10⁷ CFU of OG1RF (closed circles) or the nonpiliated SrtC⁻ mutant (open hexagons). Bacterial titers 24 h p.i. in the bladders (A) and implants (B) from 2 independent experiments are shown. Each shape corresponds to one mouse. Median titers (CFU/bladder, CFU/implant) are shown with a bar: OG1RF (2.16 × 10⁶, 4.90 × 10⁵) and SrtC⁻ strain (1.18 × 10³, 20). Dashed lines are limits of detection (40 CFU/bladder; 20 CFU/implant). Statistically significant differences between OG1RF and SrtC⁻ titers are shown (***, P < 0.001).

Transformation of EbpABC⁻ SrtC⁻ strain with p-ebpABCsrtC but not p-ebpA^AWAGABCsrtC or pGCP123 complemented its virulence defect in experimental CAUTI. Bacterial titers 24 h p.i. of the bladders (A) and implants (B) from 2 independent experiments are shown. Each shape corresponds to one mouse; open shapes represent a nonpiliated bacterial strain. Median titers (CFU/bladder, CFU/implant) are shown with a bar: OG1RF/pGCP123 (5.92 × 10⁵, 1.96 × 10⁵), EbpABC⁻ SrtC⁻/pGCP123 strain (1.94 × 10³, 1.72 × 10²), EbpABC⁻ SrtC⁻/p-ebpABCsrtC strain (2.60 × 10⁵, 1.20 × 10⁵), and EbpABC⁻ SrtC⁻/p-ebpA^AWAGABC strain (1.20 × 10³, 5.80 × 10²). Dashed lines are the limits of detection (10 CFU/bladder, 5 CFU/implant). All possible strain pairs were compared statistically. P values were adjusted for 6 comparisons. Significant differences are shown (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Mutation of EbpA’s MIDAS motif does not affect pilus biogenesis. Mouse anti-EbpC polyclonal sera were used to assess pilus biogenesis by negative-stain immunogold EM (A and B), IFM (C), and Western blot analysis (D) after SDS-PAGE of the indicated cell fractions. (A and B) Piliation of the MIDAS motif mutant strains (EbpABC⁻ SrtC⁻/p-ebpA^AWAGABCsrtC strain and EbpA^AWAGA) was similar to that of control strains (EbpABC⁻ SrtC⁻/p-ebpABCsrtC strain and OG1RF, respectively). Bars, 500 nm. (C) Comparison of the median percentages of EbpC⁺ bacterial cells in OG1RF (50%) and EbpA^AWAGA (38%) from 2 independent experiments revealed no significant differences in population piliation dynamics. Whiskers show the 10th and 90th percentiles; dots show outliers (ns, not significant). (D) Pilus HMWLs on Western blots of OG1RF and EbpA^AWAGA probed with anti-EbpC (left), anti-EbpB (middle), and anti-EbpA (right) sera were indistinguishable.

Virulence of E. faecalis OG1RF and its isogenic pilin deletion mutants in experimental CAUTI. Twenty-four-hour viable bacterial titers in the bladders (A) and those associated with implants (B) from 2 to 4 independent experiments/strain are shown. Each shape corresponds to one mouse; open shapes represent nonpiliated bacterial strains. Median titers (CFU/bladder, CFU/implant) are shown with a bar: OG1RF (3.20 × 10⁵, 6.25 × 10⁴), EbpABC⁻ strain (2.48 × 10³, 7.00 × 10²), EbpA⁻ strain (3.10 × 10³, 5.82 × 10³), EbpB⁻ strain (9.08 × 10³, 1.50 × 10⁴), EbpAB⁻ strain (4.60 × 10², 50), EbpC⁻ strain (5.80 × 10⁴, 1.04 × 10⁴), EbpBC⁻ strain (5.00 × 10², 20), and EbpAC⁻ strain (3.56 × 10³, 40). Dashed lines are the limits of detection (10 CFU/bladder, 5 CFU/implant). Results of statistical comparisons of each mutant to OG1RF for bladders and implants are shown. There were no significant differences in bladder or implant colonization between any double pilin deletion mutant and the EbpABC⁻ strain (data not shown). P values were adjusted for 10 comparisons (**, P < 0.01; ***, P < 0.001; ns, not significant).

The chromosomal MIDAS mutant (EbpA^AWAGA) was as attenuated as the EbpABC⁻ strain in experimental CAUTI. Twenty-four-hour (A and B) or 7-day (C and D) viable bacterial counts from the bladders (A and C) and implants (B and D) of mice infected with E. faecalis were pooled from 2 to 3 independent experiments for each strain. Each shape corresponds to one mouse; open shapes represent a nonpiliated bacterial strain. Median bacterial titers (CFU/bladder, CFU/implant) were determined at 24 h p.i. for OG1RF (3.16 × 10⁵, 3.24 × 10⁴), EbpA^AWAGA (2.40 × 10², 1.80 × 10²), and EbpABC⁻ (3.20 × 10², 5) strains and at 7 days for OG1RF (2.82 × 10⁴, 7.86 × 10⁴), EbpA^AWAGA (4.80 × 10², 4.60 × 10²), and EbpABC⁻ (60, 5) strains. Bars are medians; dashed lines are the limits of detection (10 CFU/bladder, 5 CFU/implant). All possible strain combinations were compared statistically; P values were adjusted for 3 comparisons. Significant differences are shown (*, P < 0.05; ***, P < 0.001).

Minor pilin deletions affect pilus biogenesis. Anti-EbpC sera were used to assess pilus assembly by Western blot analysis (A to D), to visualize pilus morphology with deep-etch EM (E), and to evaluate population piliation dynamics by IFM (F). (A) Western blot analysis of EbpA⁻ and EbpAB⁻ cell lysates showed condensed EbpC bands with reduced mobility on SDS-PAGE compared to OG1RF EbpC HMWL. (B) EbpB is expressed in OG1RF and EbpA⁻ culture supernatants. (C and D) EbpC HMWL (C) and EbpA HMWL (D) in EbpB⁻ and OG1RF strains were similar. HMWLs (brackets) indicate pilus polymerization (anti-EbpC) or minor pilin incorporation (anti-EbpA). (E) Pilus morphology was altered in EbpA⁻ and EbpAB⁻ strains. Arrowheads point to gold bead-labeled pilus fibers in OG1RF and EbpB⁻ strains. Large arrows indicate gold bead-labeled long EbpC fibers in EbpA⁻ and EbpAB⁻ strains. (F) The percentage of bacterial cells expressing EbpC (EbpC⁺ cells) was quantified in 3 independent experiments. The median percentages of EbpC⁺ cells for each strain were determined: OG1RF (69.6), EbpA⁻ strain (1.1), EbpB⁻ strain (33.9), and EbpAB⁻ strain (2.8). Whiskers show the 10th and 90th percentiles; dots show outliers. Statistically significant differences between OG1RF and each mutant strain are shown; P values were adjusted for 3 comparisons (***, P < 0.001).