Type IV pilus assembly in Pseudomonas aeruginosa over a broad range of cyclic di-GMP concentrations

Abstract

Pseudomonas aeruginosa is a Gram-negative, opportunistic pathogen that utilizes polar type IV pili (T4P) for twitching motility and adhesion in the environment and during infection. Pilus assembly requires FimX, a GGDEF/EAL domain protein that binds and hydrolyzes cyclic di-GMP (c-di-GMP). Bacteria lacking FimX are deficient in twitching motility and microcolony formation. We carried out an extragenic suppressor screen in PA103ΔfimX bacteria to identify additional regulators of pilus assembly. Multiple suppressor mutations were mapped to PA0171, PA1121 (yfiR), and PA3703 (wspF), three genes previously associated with small-colony-variant phenotypes. Multiple independent techniques confirmed that suppressors assembled functional surface pili, though at both polar and nonpolar sites. Whole-cell c-di-GMP levels were elevated in suppressor strains, in agreement with previous studies that had shown that the disrupted genes encoded negative regulators of diguanylate cyclases. Overexpression of the regulated diguanylate cyclases was sufficient to suppress the ΔfimX pilus assembly defect, as was overexpression of an unrelated diguanylate cyclase from Caulobacter crescentus. Furthermore, under natural conditions of high c-di-GMP, PA103ΔfimX formed robust biofilms that showed T4P staining and were structurally distinct from those formed by nonpiliated bacteria. These results are the first demonstration that P. aeruginosa assembles a surface organelle, type IV pili, over a broad range of c-di-GMP concentrations. Assembly of pili at low c-di-GMP concentrations requires a polarly localized c-di-GMP binding protein and phosphodiesterase, FimX; this requirement for FimX is bypassed at high c-di-GMP concentrations. Thus, P. aeruginosa can assemble the same surface organelle in distinct ways for motility or adhesion under very different environmental conditions.

Fig 1

Extragenic suppressors restore surface pili. (A) Twitching zone diameter as measured by the subsurface stab assay. Strains were assayed 3 to 5 times, and means ± standard deviations (SD) (n = 4) from a representative experiment are shown. All of the strains were compared to ΔfimX using one-way ANOVA followed by Dunnett's multiple-comparison test. (B) Total surface pili as determined by surface pilin ELISA. Two representative Tn insertions were assayed for each gene. Bars indicate means ± SD from three independent experiments. All of the samples were compared to ΔfimX using one-way ANOVA followed by Dunnett's multiple-comparison test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. (C) Visualization of T4P by transmission electron microscopy. Bacteria from early exponential growth phase were stained directly with 1% phosphotungstate. Panels show higher magnification pictures of representative cells for each strain, and insets show lower magnification pictures of the same cells. Black arrows indicate the nonpolar site of attachment of the pili with the cell body. White arrow indicates T4P on a WT surface. Scale bar, 500 nm (main panel) or 2 μm (inset).

Fig 2

Extragenic suppressor phenotypes can be reverted by expressing PA0171, PA1121, or PA3703 in trans. Extragenic suppressor strains carrying either empty vector (VC) or PA0171, PA1121, or PA3703 cloned under the pBAD promoter were grown on LB plates plus 0.2% arabinose. Strains were assayed for surface pilin expression by ELISA (A) and twitching motility by subsurface stab assay (B). Bars represent means ± SD of triplicate samples from four independent experiments. P values were determined by ANOVA with Bonferroni's multiple-comparison test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig 3

c-di-GMP levels are elevated in the extragenic suppressor strains. Bacteria were grown on LB agar plates overnight supplemented with antibiotic and 0.2% arabinose as appropriate. Each sample was prepared in biological duplicates and measured in 2 to 3 experiments; bars represent means ± SD. Values over each bar represent the c-di-GMP fold increase in the respective strains compared to the c-di-GMP level in ΔfimX or ΔfimX VC, respectively.

Fig 4

Overexpression of diguanylate cyclases restores surface pilus assembly in ΔfimX. PA103ΔfimX carrying pMQ72 (VC) or the indicated DGCs cloned under the pBAD promoter were grown on plates with or without 0.2% arabinose. (A) Total surface pili as determined by surface pilin ELISA. Bars indicate means ± SD of triplicate samples from three independent experiments. Two-way ANOVA followed by Bonferroni's multiple comparison test was used to analyze the data. All induced (+Arabinose) samples were compared to ΔfimX/pMQ72 plus arabinose. *, P < 0.05; **, P < 0.01; ***, P < 0.001. (B) T4P visualized by transmission electron microscopy after staining with 1% phosphotungstate. Panels show a higher magnification image of a representative cell. Scale bar, 500 nm. ns, not significant.

Fig 5

Inactive DGC does not suppress ΔfimX pilus assembly defects. ΔfimX expressing PA1120 or the enzymatically inactive allele PA1120m was harvested from plates supplemented with 0.2% arabinose. Total surface pilin was measured by ELISA. Bars indicate means ± SD of triplicate samples from three independent experiments. One-way ANOVA followed by Dunnett's comparison test was used to compare values to the vector control, ΔfimX/pMQ72 (VC). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig 6

ΔfimX bacteria form biofilms and stain for surface pilin. (A) PAO1, PAO1ΔfimX, and PAO1ΔpilA were grown as static biofilms for 24 h at 30°C in triplicate wells and then stained with crystal violet. Bars represent means ± SD from three independent experiments; values are normalized to PAO1. *, P < 0.05; **, P < 0.01; ***, P < 0.001. (B) Three-dimensional plot of day 4 flow biofilms formed by PAO1, PAO1ΔfimX, and PAO1ΔpilA. (C) Scatter plots showing average thickness and roughness of day 4 flow biofilms of PAO1, PAO1ΔfimX, and PAO1ΔpilA, as calculated by COMSTAT. (D) PAO1, PAO1ΔfimX, and PAO1ΔpilA were grown on polylysine-coated coverslips for 24 h under static biofilm conditions and then stained for surface pilin. All images were acquired under identical conditions. Scale bar, 10 μm.

Fig 7

DGC overexpression in PA103 or PAO1 does not inhibit surface pilus assembly. (A) WT PA103 or PAO1 overexpressing indicated DGCs (PA1120 and PA3702) was grown on plates with 0.2% arabinose. Bars represent the twitching zone diameter as measured by the subsurface stab assay. (B) WT PA103 or PAO1 overexpressing indicated DGCs (PA0169, PA1120, and PA3702) were grown on plates with or without 0.2% arabinose. Surface pilin was measured by ELISA. Bars indicate means ± SD of triplicate samples from three independent experiments.