In vivo and in vitro analyses of the role of the Prc protease in inducing mucoidy in Pseudomonas aeruginosa

Abstract

In Pseudomonas aeruginosa, alginate biosynthesis gene expression is inhibited by the transmembrane anti-sigma factor MucA, which sequesters the AlgU sigma factor. Cell envelope stress initiates cleavage of the MucA periplasmic domain by site-1 protease AlgW, followed by further MucA degradation to release AlgU. However, after colonizing the lungs of people with cystic fibrosis, P. aeruginosa converts to a mucoid form that produces alginate constitutively. Mucoid isolates often have mucA mutations, with the most common being mucA22, which truncates the periplasmic domain. MucA22 is degraded constitutively, and genetic studies suggested that the Prc protease is responsible. Some studies also suggested that Prc contributes to induction in strains with wild-type MucA, whereas others suggested the opposite. However, missing from all previous studies is a demonstration that Prc cleaves any protein directly, which leaves open the possibility that the effect of a prc null mutation is indirect. To address the ambiguities and shortfalls, we reevaluated the roles of AlgW and Prc as MucA and MucA22 site-1 proteases. In vivo analyses using three different assays and two different inducing conditions all suggested that AlgW is the only site-1 protease for wild-type MucA in any condition. In contrast, genetics suggested that AlgW or Prc act as MucA22 site-1 proteases in inducing conditions, whereas Prc is the only MucA22 site-1 protease in non-inducing conditions. For the first time, we also show that Prc is unable to degrade the periplasmic domain of wild-type MucA but does degrade the mutated periplasmic domain of MucA22 directly.

Importance: After colonizing the lungs of individuals with cystic fibrosis, Pseudomonas aeruginosa undergoes mutagenic conversion to a mucoid form, worsening the prognosis. Most mucoid isolates have a truncated negative regulatory protein MucA, which leads to constitutive production of the extracellular polysaccharide alginate. The protease Prc has been implicated, but not shown, to degrade the most common MucA variant, MucA22, to trigger alginate production. This work provides the first demonstration that the molecular mechanism of Prc involvement is direct degradation of the MucA22 periplasmic domain and perhaps other truncated MucA variants as well. MucA truncation and degradation by Prc might be the predominant mechanism of mucoid conversion in cystic fibrosis infections, suggesting that Prc activity could be a useful therapeutic target.

Keywords: Pseudomonas aeruginosa; alginate; proteases; regulation of gene expression.

Fig 1

AlgU regulation in wild-type and mucA22 strains. In wild-type strains, two inducing signals must be present to initiate sequential proteolysis of MucA, release of AlgU, and expression of the algD operon and other targets. Mislocalized OMPs interact with AlgW to active it, and mislocalized LPS interact with MucB to disengage it from MucA. Only then can AlgW cleave the C-terminus of MucA, which makes truncated MucA a substrate for MucP, which cleaves the MucA transmembrane region. Finally, the ClpXP complex degrades the cytoplasmic domain of MucA to release AlgU. The mucA22 mutation truncates MucA, removing the MucB binding site but leaving the AlgW cleavage site present. AlgW still cannot initiate MucA22 proteolysis unless environmental conditions trigger OMP mislocalization. However, previous genetic studies suggested that mucA22 strains might be mucoid in normally non-inducing conditions because the C-terminus of MucA22 is cleaved by the constitutively active Prc protease. Whether or not Prc could cleave wild-type MucA was unclear from previous studies, although most suggested that it might not.

Fig 2

Effects of ∆prc and ∆ctpA mutations on regulation of the algD operon and mucoidy. (a) Φ(algDp-lacZ) operon fusion expression. Strains with the wild-type mucA gene (wild type) or mucA22 were grown with or without exposure to D-cycloserine. Protease genotypes are shown below each bar (WT = wild-type prc⁺ ctpA⁺). Error bars indicate the positive standard deviations from the means. *P  <  0.05 and **P  <  0.01; n.s., not statistically significant, P > 0.05 (unpaired t test). Significance is measured comparing each protease mutant to the corresponding WT strain grown in the same condition. (b) Anti-AlgD immunoblot analysis of one representative set of the triplicate cultures used to generate the data in panel a. (c) Phenotypes on Pseudomonas isolation agar (PIA) with or without ammonium metavanadate (AMV). Immunoblots and agar plates are single representatives of several replicate experiments.

Fig 3

Comparison of the effects of ∆prc and ∆algW mutations on regulation of the algD operon and mucoidy. (a) Φ(algDp-lacZ) operon fusion expression. Strains with wild-type muc genes (wild type), a mucA22 mutation, or a ∆mucB mutation were grown with or without exposure to D-cycloserine. Protease genotypes are shown below each bar (WT = wild-type prc⁺ algW⁺). Error bars indicate the positive standard deviations from the means. *P  <  0.05, **P  <  0.01, ***P  <  0.001, and ****P  <  0.0001; n.s., not statistically significant, P > 0.05 (unpaired t test). Significance is measured comparing each protease mutant to the corresponding WT strain grown in the same condition. (b) Anti-AlgD immunoblot analysis of one representative set of the triplicate cultures used to generate the data in panel a. (c) Phenotypes on PIA with or without AMV. Immunoblots and agar plates are single representatives of several replicate experiments.

Fig 4

Analysis of MucA periplasmic domain cleavage in vitro. (a) Coomassie blue-stained SDS-PAGE gel analysis of the purified Prc-His₆ or Prc-S456-His₆ proteins, with Mw markers sizes to the left (kDa). For proteolysis assays, MBP-′MucA22 (b) or MBP-′MucA (c) proteins were incubated with Prc-His₆ or Prc-S456-His₆ at 37°C. Single-reaction tubes were used, with an aliquot removed from it at the indicated number of hours. For each panel, the top image shows a Coomassie blue-stained SDS-PAGE gel with Mw markers sizes to the left (kDa), and the bottom image is anti-MBP immunoblot, with the bracket showing the corresponding regions of the two images. The relative density of the MBP-′MucA22 and MBP-′MucA bands was determined relative to the 1-h timepoint, which was set to 1.0. The 1-h timepoint was chosen as the reference point for rigor, because the subsequent samples were all taken from the same tube. For wild-type MucA, the MBP-′MucA protein purified as a mixture of mostly MBP-′MucA and MBP, along with some likely MBP-′MucA-truncated intermediates visible as faint bands between MBP-′MucA and MBP on both the Coomassie blue-stained gel and anti-MBP immunoblot. Data presented are single representatives of several replicate experiments.