Coordination of swarming motility, biosurfactant synthesis, and biofilm matrix exopolysaccharide production in Pseudomonas aeruginosa

Abstract

Biofilm formation is a complex process in which many factors are involved. Bacterial swarming motility and exopolysaccharides both contribute to biofilm formation, yet it is unclear how bacteria coordinate swarming motility and exopolysaccharide production. Psl and Pel are two key biofilm matrix exopolysaccharides in Pseudomonas aeruginosa. This opportunistic pathogen has three types of motility, swimming, twitching, and swarming. In this study, we found that elevated Psl and/or Pel production reduced the swarming motility of P. aeruginosa but had little effect on swimming and twitching. The reduction was due to decreased rhamnolipid production with no relation to the transcription of rhlAB, two key genes involved in the biosynthesis of rhamnolipids. Rhamnolipid-negative rhlR and rhlAB mutants synthesized more Psl, whereas exopolysaccharide-deficient strains exhibited a hyperswarming phenotype. These results suggest that competition for common sugar precursors catalyzed by AlgC could be a tactic for P. aeruginosa to balance the synthesis of exopolysaccharides and rhamnolipids and to control bacterial motility and biofilm formation inversely because the biosynthesis of rhamnolipids, Psl, and Pel requires AlgC to provide the sugar precursors and an additional algC gene enhances the biosynthesis of Psl and rhamnolipids. In addition, our data indicate that the increase in RhlI/RhlR expression attenuated Psl production. This implied that the quorum-sensing signals could regulate exopolysaccharide biosynthesis indirectly in bacterial communities. In summary, this study represents a mechanism that bacteria utilize to coordinate swarming motility, biosurfactant synthesis, and biofilm matrix exopolysaccharide production, which is critical for biofilm formation and bacterial survival in the environment.

FIG 1

Motility of Psl/Pel-inducible strains WFPA801, WFPA831, and WFPA833 at various concentrations of arabinose (inducer). (A) Swarming patterns at 0 to 0.5% arabinose. (B) Swimming zones at 0 or 1% arabinose. (C) Twitching zones at 0 or 1% arabinose. The values in the images are the diameters of the corresponding swimming or twitching zones. PAO1 ΔfliC and PAO ΔpilA were used as negative controls for swimming and twitching motility, respectively.

FIG 2

Overproduction of Psl and/or Pel polysaccharide decreases RL production. (A) The drop collapse assay shows a reduction of RL in Psl/Pel-induced strains. The image shows the bead formation of the culture supernatants of the Psl/Pel-inducible strains grown with and without arabinose. CS, culture supernatants; DW, distilled water; +, 1% arabinose; −, no arabinose. (B) Swarming patterns, swarming zones, total RL production, and corresponding rhlAB transcription of PAO1, WFPA801, WFPA831, and WFPA833 with 0 or 1% arabinose. Shown under each image are the corresponding quantitative values of swarming (average zone diameters in centimeters), total RL content, and rhlAB-lacZ transcription. The RL extracts were normalized to the level of PAO1 in the absence of arabinose (1 = 4.9 g/liter total RL by dry-weight calculation or 284.3 mg/liter RL by the anthrone colorimetric assay). Transcription of rhlAB was determined by the β-galactosidase activity of a P_rhlAB-lacZ fusion integrated into the chromosomes of the corresponding strains. The values were also normalized to those of PAO1 without arabinose (1 = 116 Miller units). 0, no swarming zone.

FIG 3

TLC and LC-MS analyses of total RL extracts from the PAO1 and Psl/Pel-inducible strains. (A) TLC analysis of RL extracted from cultures of the corresponding strains grown in the absence (−) or presence (+) of 1% arabinose. (B) LC-MS peak of Rha-C₁₀-C₁₀ (m/z 503) (mono-RL). (C) Relative abundances of Rha-C₁₀-C₁₀ in corresponding RL extracts analyzed by LC-MS. (D) LC-MS peak of Rha-Rha-C₁₀-C₁₀ (m/z 649) (di-RL). (E) Relative abundances of Rha-Rha-C₁₀-C₁₀ in corresponding RL extracts analyzed by LC-MS. The relative abundances of Rha-C₁₀-C₁₀ and Rha-Rha-C₁₀-C₁₀ were normalized to the levels in PAO1 (respectively, 3,646,111 and 1,653,883 without arabinose and 2,722,080 and 2,223,691 with 1% arabinose). The corresponding peaks of mono-RL and di-RL are indicated in panels B and C. ESI, electrospray ionization.

FIG 4

Swarming ability and total RL and/or Psl production of RL-deficient mutants and psl and pel mutant strains. (A) Swarming pattern, corresponding swarming zones, Psl levels, and total RL production of PAO1 and the PAO1 isogenic rhlR, rhlAB, pslA, and galU mutants. (B) Swarming and RL production of MJK8, the isogenic psl and/or pel mutant, and native Psl-negative strain PA14. The mean swarming zone diameter and total RL production are presented under each corresponding image. The RL extracts were normalized to the level of PAO1 (214 mg/liter). Psl was detected by immunoblotting with Psl antiserum. ++++, Psl level 4-fold that of PAO1; −, Psl negative; 0, no swarming zone; ND, not determined.

FIG 5

The biosynthesis pathways of RL, Psl, and Pel have the sugar precursor Glc-1-P (catalyzed by AlgC) in common. (A) Schematic representation of the P. aeruginosa PAO1 metabolic routes for the biosynthesis of Psl, Pel, and RL. The solid lines represent known metabolic routes, and the dashed line represents a hypothetical route. (B) Swarming zone of PAO1/algC (algC was constitutively expressed by plasmid pLPS188.) and control strain PAO1/vector. (C) Comparison of RL production in PAO1/algC and PAO1/vector. The amount of RL was normalized to the PAO1/vector level (261 mg/liter). (D) Comparison of PAO1/algC and PAO1/vector Psl levels. *, P < 0.05.

FIG 6

Effect of RhlR/RhlI QS system on Psl polysaccharide biosynthesis. (A) Swarming zones of rhlRI-inducible PAO/rhlRI and ΔrhlR/rhlRI mutant strains grown with or without arabinose. (B) Psl and RL production of PAO/rhlRI and the ΔrhlR/rhlRI mutant at various concentrations of arabinose. The amount of RL was normalized to the PAO1/vector level (214 ± 3 mg/liter). *, P < 0.05; ND, not determined.