Pseudomonas aeruginosa small protease (PASP), a keratitis virulence factor

Abstract

Purpose: The virulence contribution of Pseudomonas aeruginosa small protease (PASP) during experimental keratitis was studied by comparing a PASP-deficient mutant with its parent and rescue strains.

Methods: The pasP gene of P. aeruginosa was replaced with the tetracycline resistance gene via allelic exchange. A plasmid carrying the pasP gene was introduced into the PASP-deficient mutant to construct a rescue strain. The PASP protein in the culture supernatants was determined by Western blot analysis. Corneal virulence was evaluated in rabbit and mouse keratitis models by slit lamp examination (SLE), bacterial enumeration, and/or histopathological analysis. Various host proteins and the rabbit tear film were analyzed for their susceptibility to PASP degradation.

Results: The PASP-deficient mutant produced a significantly lower mean SLE score when compared with the parent or rescue strain (P ≤ 0.03) at 29 hours postinfection (PI). All of the strains grew equally in the rabbit cornea (P = 0.971). Corneas infected with the PASP-deficient mutant showed moderate histopathology compared with those infected with the parent or rescue strain, which produced severe pathology inclusive of epithelial erosions, corneal edema, and neutrophil infiltration. In the mouse model, eyes inoculated with the PASP-deficient mutant had a significantly lower mean SLE score at 24 hours PI than the eyes inoculated with the parent or rescue strain (P ≤ 0.007). PASP was found to degrade complement C3, fibrinogen, antimicrobial peptide LL-37, and constituents of the tear film.

Conclusions: PASP is a commonly secreted protease of P. aeruginosa that contributes significantly to the pathogenesis of keratitis.

Figure 1

Construction of the allelic exchange construct, pEX100T-ΔPASP, and rescue plasmid pUCP20-pasP. (A) A cassette containing the flanking regions of the pasP gene and the tetracycline resistance gene Tet was assembled and cloned into the suicidal vector pEX100T. The replacement of the pasP gene in the chromosome with the Tet gene resulted from homologous recombination. sacB, levansucrase. (B) To construct the rescue plasmid, the pasP gene was PCR-amplified with addition of the restriction sites EcoRI and BamHI at the 5′ and 3′ end, respectively. Then, the pasP gene was cloned into the Pseudomonas/E. coli shuttle vector pUCP20 under the lacZ promoter. lacZ, β-galactosidase; bla, β-lactamase.

Figure 2

Analysis of the PASP gene and protein. (A) Colony PCR analysis of PA103-29, the PASP-deficient mutant, and its rescue strain for the presence of the Tet and pasP genes. A single colony of PA103-29 (parent), PA103-29ΔPASP (mutant), or PA103-29ΔPASP/pUCP20-pasP (rescue) was suspended in 30 μL dH₂O and boiled for 5 minutes. Aliquots of the supernatants of the colony suspensions (5 μL added to a 25-μL PCR reaction) were used as template DNA. (B) Identification of the PASP protein in the concentrated Pseudomonas culture supernatants. Concentrated culture supernatants of PA103-29 (20 μL, 3 mg/mL total protein), PA103-29ΔPASP (20 μL, 3 mg/mL), and PA103-29ΔPASP/pUCP20-pasP (20 μL, 3 mg/mL) were analyzed by Western blot analysis. Lane 1: parent strain; lane 2: PASP-deficient mutant; and lane 3: rescue strain.

Figure 3

Evaluation of ocular virulence of the PASP-deficient mutant and its parent and rescue strains in a rabbit intrastromal injection model of keratitis. (A) Photographs of rabbit eyes at 29 hours after intrastromal injection of the three strains. (B) SLE scores of infected rabbit eyes at 29 hours PI. The mean SLE score of the mutant group is significantly lower than that of the parent group (*P = 0.03) or the rescue group (†P = 0.011). There is no difference in SLE scores between the parent group and the rescue group (‡P = 0.109).

Figure 4

Histopathological analysis of rabbit corneas infected with the parental strain PA103-29, the PASP-deficient mutant, or the rescue strain. Corneas (n ≥ 3) were harvested at 29 hours PI. Corneal sections were stained with hematoxylin and eosin, microscopically examined, and photographed under low power (×40). The parent and rescue strains caused extensive epithelial erosion (detachment), severe neutrophil infiltration, and corneal edema. In contrast, the PASP-deficient mutant-injected corneas showed no or trace epithelial erosion, mild neutrophil infiltration, and no edema.

Figure 5

Virulence comparison of the PASP-deficient mutant, its parent strain, and the rescue strain in a mouse topical inoculation model of keratitis. (A) Photographs of infected mouse eyes at 24 hours PI. (B) SLE scores of infected mouse eyes. The mean SLE score of the mutant group is significantly lower than that of the parent group (*P = 0.009) or the rescue group (†P = 0.007). There is no difference in SLE scores between the parent group and the rescue group (‡P = 0.899).

Figure 6

Degradation of host proteins and the tear film by PASP. Heat-inactivated (lane 1) or active rPASP (1 μg; lane 2) was incubated with fibrinogen (A) or complement C3 (B) at 37°C for 4 hours. Antimicrobial peptide LL-37 (C) and rabbit tears (D) were incubated with dH₂O (lane 1), active rPASP (1 μg; lane 2), or PIV (1 μg; lane 3) for 14 hours and 5 hours, respectively. The arrows indicate where the changes in electrophoretic mobility of the host protein components occurred.