Loss of meningococcal PilU delays microcolony formation and attenuates virulence in vivo

Abstract

Neisseria meningitidis is a major cause of sepsis and bacterial meningitis worldwide. This bacterium expresses type IV pili (Tfp), which mediate important virulence traits such as the formation of bacterial aggregates, host cell adhesion, twitching motility, and DNA uptake. The meningococcal PilT protein is a hexameric ATPase that mediates pilus retraction. The PilU protein is produced from the pilT-pilU operon and shares a high degree of homology with PilT. The function of PilT in Tfp biology has been studied extensively, whereas the role of PilU remains poorly understood. Here we show that pilU mutants have delayed microcolony formation on host epithelial cells compared to the wild type, indicating that bacterium-bacterium interactions are affected. In normal human serum, the pilU mutant survived at a higher rate than that for wild-type bacteria. However, in a murine model of disease, mice infected with the pilT mutant demonstrated significantly reduced bacterial blood counts and survived at a higher rate than that for mice infected with the wild type. Infection of mice with the pilU mutant resulted in a trend of lower bacteremia, and still a significant increase in survival, than that of the wild type. In conclusion, these data suggest that PilU promotes timely microcolony formation and that both PilU and PilT are required for full bacterial virulence.

Fig 1

A pilU mutant outcompetes wild-type FAM20 in a competitive adhesion assay. The assay was performed using wild-type FAM20 (WT), a pilU mutant (ΔpilU), a complemented pilU mutant (ΔpilU_C), a pilT mutant (ΔpilT), a pilU pilT double mutant (ΔpilT ΔpilU), and the double mutant complemented with pilU (ΔpilT ΔpilU_C). Mutant and wild-type bacteria were mixed at a 1:1 ratio as determined by the OD₆₀₀ and viable counts. The bacterial mixture was immediately added to FaDu cells (total MOI = 100) and allowed to adhere for 2 h in a standard adhesion assay. The competitive index denotes the relative amounts of mutants adherent to cells as determined by plating on both selective and nonselective media. The dashed line indicates a competitive index of 1, i.e., no difference in adhesion. Error bars indicate standard deviations, and asterisks denote statistically significant differences from a CI of 1 (P < 0.001; paired t test).

Fig 2

Characterization of adhesion factors in the pilU mutant. Immunoblots show the expression levels of PilC, PilT, PilE, and Opa in the wild-type FAM20 (WT), pilU mutant (ΔpilU), and complemented (ΔpilU_C) strains. Controls for the PilC immunoblot were FAM20 mutants expressing only one defined PilC variant, i.e., lanes PilC1+/PilC2− and PilC1−/PilC2+. The control for the PilT immunoblot was the PilT-deficient mutant (ΔpilT). Whole-cell lysates were analyzed in a 2-color Western blot and visualized with anti-PilC, anti-PilT, anti-pili, and anti-Opa antibodies in combination with a monoclonal anti-EF-Tu antibody. The EF-Tu band intensity was used as a loading control and used for normalization. R.I. denotes the relative EF-Tu-normalized band intensity for at least three separate experiments, set to 1.0 for the wild type. Asterisks denote statistically significant differences relative to the wild type (P < 0.05; Student's t test).

Fig 3

pilU and pilT mutants stimulate IL-6 and IL-8 similarly to the stimulation by the wild type. The graphs show the release of IL-6 (A) and IL-8 (B) from FaDu cells following infection with bacteria (MOI = 100) for 2 h (black bars) and 8 h (white bars), as quantified by ELISA of the cell culture supernatant. The following strains were used: wild-type FAM20 (WT), pilU mutant (ΔpilU), complemented pilU mutant (ΔpilU_C), pilT mutant (ΔpilT), pilU pilT double mutant (ΔpilT ΔpilU), and the double mutant complemented with pilU (ΔpilT ΔpilU_C). Values are normalized to those for uninfected controls at each time point. N.C. (negative control) indicates uninfected cells. Error bars show standard deviations.

Fig 4

The pilU mutant is delayed in microcolony formation on human epithelial cells. (A) Live-cell phase-contrast and fluorescence imaging of pharyngeal epithelial FaDu cells infected with wild-type FAM20 and the ΔpilU, ΔpilU_C, and ΔpilT mutants during a total time of 12 h. Representative time points were chosen to show when the wild type and the pilU mutant reached their respective maximal microcolony sizes. Prior to infection, bacteria were passed through a 5-μm-pore-size filter to break apart bacterial aggregates. The initial inoculum of bacteria was stained with the fluorescent DyLight488 N-hydroxysuccinimide ester, visible in red. Microcolonies are highlighted with a yellow dashed line around the periphery. The ΔpilT mutant was included as a positive control for aggregation. Bar = 10 μm. Quantification of microcolony formation (B) and microcolony dispersal (C) was done for the wild type and the ΔpilU mutant. Microcolony formation was determined as the time point when one persistent microcolony with a size of 10 μm first became visible. Microcolony dispersal was defined as the time point when tight bacterial aggregates visibly started to dissolve into single bacteria. Values are means ± standard deviations for the experiment depicted in panel A and are representative of all assays performed. Significant differences from the wild type are indicated with asterisks (P < 0.05; two-tailed Student's t test).

Fig 5

Microcolony formation is delayed in a meningococcal pilU mutant in cell-free medium. Live-cell phase-contrast microscopy was performed on bacteria growing in cell-free medium. Representative images were selected based on when the wild type and the ΔpilU mutant reached their respective maximal microcolony sizes. Microcolonies are highlighted with a yellow dashed line around the periphery. The ΔpilT mutant was included as a positive control for aggregation. Bar = 10 μm.

Fig 6

Loss of PilU increases serum resistance. (A) Survival of the wild-type strain (WT), the ΔpilU mutant, the ΔpilU_C mutant, the ΔpilT mutant, the ΔpilT ΔpilU double mutant, and the ΔpilT ΔpilU_C mutant after 60 min of incubation in 35% NHS. Bacterial survival is expressed as the percentage of CFU remaining from the initial inoculum of 10⁷ CFU in 1 ml of medium, which was set to 100%. Error bars show standard deviations. Statistically significant differences from the wild type are indicated with asterisks (P < 0.001; Tukey's range test). (B) Measurement of capsular polysaccharide by ELISA. A capsule-negative mutant (SiaD−) was used as the negative control. Purified capsule (P.C.) was used as the positive control.

Fig 7

Meningococcal ΔpilT and ΔpilU mutants are less virulent than the wild type in a mouse model of meningococcal disease. Human CD46 transgenic mice were infected i.p. with the wild-type strain, the ΔpilT mutant, or the ΔpilU mutant (10⁸ CFU/mouse). (A) Bacterial counts in blood (CFU/ml) were determined 2, 6, 24, and 48 h after the start of infection. Statistically significant values are indicated with asterisks (P < 0.05; Kruskal-Wallis nonparametric one-way ANOVA). (B) Survival of mice (n = 7 to 9 mice per group) after i.p. challenge with the wild-type strain, the ΔpilT mutant, or the ΔpilU mutant (*, P < 0.05; Kaplan-Meier survival analysis).