Presence or Absence of Lipopolysaccharide O Antigens Affects Type III Secretion by Pseudomonas aeruginosa

Bacterial strains and growth conditions.The bacterial strains used in this study and their previously published O-antigen characteristics are listed in Table 1. P. aeruginosa strains were routinely grown in Luria-Bertani (LB) broth (Sigma-Aldrich, St. Louis, MO) with shaking at 250 rpm at 37°C for 18 h. For quantitative PCR (Q-PCR) experiments, laboratory strains were cultured overnight in inducing medium (16), and the cultures were diluted (optical density at 600 nm [OD₆₀₀] = 0.1) in fresh inducing medium and reincubated to mid-exponential phase (OD₆₀₀ = 0.7). As a control, strains were also cultured under noninducing conditions (inducing medium lacking EGTA). For analysis of TTSS protein production and secretion by P. aeruginosa laboratory strains, cultures were grown overnight in inducing medium (19), diluted to an OD₆₀₀ of 0.1, and reincubated for a further 24 h in the same medium. For comparative assessment of gene expression and protein production by both planktonic and biofilm cultures of the same strains, each strain was cultured in 5 ml of TTSS-inducing MINS medium (19) overnight and diluted to an OD₆₀₀ of 0.1 in the same medium. Half a milliliter of this diluted inoculum was plated on MINS agar plates and incubated statically at 37°C for 24 h (biofilm culture). The remaining 4.5 ml was incubated at 37°C for 24 h with vigorous shaking (planktonic culture).

RNA isolation and quantitative PCR.Total RNA was isolated from bacterial strains grown under the various conditions described above using Trizol (Invitrogen, California) according to the manufacturer's instructions. Specific primer sets for target genes (Table 2) were generated using Beacon Designer v.3.0 (Premier Biosoft International, California). Total RNA was checked for integrity and absence of DNA by PCR prior to cDNA synthesis using 1 μg of total RNA and the GeneAmp RNA PCR Kit (Applied Biosystems, California) according to the manufacturer's instructions. mRNA transcript concentrations were measured using SYBR Green PCR Master Mix (Applied Biosystems, California) according to the manufacturer's instructions. Reactions were carried out using the Mx3000P Real-Time PCR System (Stratagene, California) as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 30 s, 58°C for 1 min, and 72°C for 30 s. The data acquisition step was set at 58°C with a final melting-curve analysis to ensure amplification of a single product. The transcript levels of each gene were normalized to that of the internal control, rpoD (23), and expression was calculated using the 2^−ΔΔCT method (13).

Analysis of type III effector protein production and secretion. P. aeruginosa laboratory and clinical strains were cultured as described above for biofilm and planktonic cultures. Biofilms were scraped from agar plates into microcentrifuge tubes and centrifuged to form a cell pellet. Planktonic cultures were harvested by centrifugation, and the supernatants were carefully removed and concentrated using Amicon Ultra centricon tubes (molecular weight cutoff, 10,000; Millipore, Massachusetts). The planktonic and biofilm cell pellets were resuspended in lysis buffer (1% sodium dodecyl sulfate and 50 mM Tris-HCl, pH 8.0) and boiled for 3 min. The protein concentrations of all samples were determined using the DC protein assay kit (Bio-Rad, California). Equal concentrations (10 μg) of each sample were separated on a sodium dodecyl sulfate (12.5% Tris-HCl)-polyacrylamide gel (Bio-Rad, California). The gels were transferred onto a polyvinylidene difluoride membrane (Bio-Rad, California) using a semidry electrotransfer unit (Bio-Rad, California). Immunoblotting was carried out using antibodies against ExoU, ExoS, and PcrV as previously described (22).

Cytotoxicity assay using human BEAS-2B lung epithelia.A human bronchiolar epithelial cell line (BEAS-2B; ATCC CRL9609) was cultured in Dulbecco's modified Eagle medium with low glucose (DMEM H-16; Invitrogen, California) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen, California). The cells were seeded into 96-well flat-bottom tissue culture plates at a density of 2.5 × 10⁵ cells per well and incubated to confluence in a 5% CO₂ chamber at 37°C for 18 h. The wells were rinsed twice with sterile phosphate-buffered saline (PBS), followed by the addition of 180 μl of RPMI 1640 containing l-glutamine (without phenol red, supplemented with 10% fetal bovine serum) to each well. Overnight P. aeruginosa 5-ml LB cultures were centrifuged, and the pellets were rinsed twice with Lactated Ringers (Baxter Healthcare Corporation, Illinois) to ensure the removal of exoproducts. The pellets were resuspended in sterile Lactated Ringers-saline (1:1) solution. Twenty microliters of Lactated Ringers-saline solution containing a standardized concentration of each P. aeruginosa strain (10⁷ CFU ml⁻¹) was added to triplicate wells. Negative controls received 20 μl of cell-free Lactated Ringers-saline solution. At various time intervals, 100 μl of supernatant from triplicate wells was removed, placed in a sterile 96-well flat-bottom tissue culture plate, and centrifuged at low speed (3,000 rpm). Lactate dehydrogenase activity assays were carried out on 50 μl of cell-free supernatant using the CytoTox96 Non-Radioactive Cytotoxicity Assay Kit (Promega, California) according to the manufacturer's instructions.

In vivo murine virulence.Pathogen-free male BALB/c mice (12 weeks old) were obtained (Charles River Laboratories, Massachusetts). After inhaling anesthesia with sevoflurane, each mouse received an airspace instillate containing 2% bovine serum albumin, 2 mg of Evans blue dye, 0.12 μCi of ¹³¹I-labeled albumin, and P. aeruginosa (PAO1 or LPS mutants; 5 × 10⁷ CFU ml⁻¹; four mice per strain) in sterile PBS. A sample of the instillate was saved for radioactivity measurement (counts/min/g) using a γ-ray counter (Packard, Illinois) and quantitative bacterial culture on LB agar plates to confirm the inoculation CFU. The mice were returned to their cages and allowed access to food and water. Four hours after bacterial instillation, the mice were euthanized and exsanguinated. Lung injury was quantified by radioactivity counts (epithelial permeability) and lung weight (lung edema) as previously described (7, 24).

LPS B-band O-antigen serotyping.All laboratory and clinical strains used in this study were grown as both biofilm and planktonic cultures as described above, and their B-band O-antigen serotypes were characterized in each mode of growth using a P. aeruginosa Serotyping Kit (ERFA Biotech, Canada) according to the manufacturer's instructions. Since the serotypes of PAO1 and isogenic mutants are known, these strains served as controls to ensure the accuracy of the kit. Three or four loopfuls of biofilm biomass per strain was resuspended in 100 μl PBS. Planktonic cultures were harvested by centrifugation and resuspended in 1 ml of PBS. Ten microliters of each cell suspension per strain was then mixed with 10 μl of each antibody solution (1× concentration) prior to being vortexed and incubated at room temperature for 20 to 30 min. Serotype-positive strains exhibited aggregation and settling to the bottom of the tube. Negative controls remained uniformly suspended in solution.

Quantitative PCR of exsA, exoS, exoT, and pcrV.To determine whether the LPS structure affected TTSS gene expression, quantitative-PCR (Q-PCR) analysis was carried out on PAO1 and isogenic mutants (containing specific O-antigen mutations) (Table 1) to measure mRNA expression levels for genes involved in the TTSS (Table 2) under inducing and noninducing conditions. Figure 1A to D illustrates the relative expression levels observed for four key genes of the TTSS of PAO1; the positive regulator (exsA), two genes encoding secreted proteins (exoS and exoT) (cytotoxins), and pcrV (which facilitates cytotoxin injection). Under noninducing conditions, gene expression levels were identical for all four strains (data not shown). However, under TTSS-inducing conditions, wbpM (A⁺ B⁻) and rmlC (A⁻ B⁻) mutant strains lacking the highly structured B-band O antigen consistently exhibited significantly increased (P = 0 for all for genes examined) TTSS gene expression relative to PAO1 (Fig. 1A to D). In addition, the two strains also demonstrated increased expression compared to wbpM, which exhibited TTSS gene expression levels similar to those of PAO1.

• Open in new tab
• Download powerpoint

FIG. 1.

Relative expression levels of exsA (A), exoS (B), exoT (C), and pcrV (D). The results are representative of three independent experiments carried out in triplicate. The error bars indicate standard deviations.

Analysis of TTSS cytotoxin production and secretion.Since measurement of mRNA expression may not directly correlate with concentrations of secreted effector proteins and therefore cytotoxicity, Western immunoblotting analysis was performed on PAO1 and isogenic LPS mutants to determine the production (intracellular) and secretion (extracellular fraction) of ExoS (the most potent of the cytotoxins of PAO1) and PcrV (crucial for effective cytotoxin secretion [17]). While all strains produced and secreted the ExoS cytotoxin, the three LPS mutant strains produced and secreted two- to fourfold more than the wild type (Fig. 2A). Wild-type PAO1 produced barely detectable concentrations of PcrV intracellularly and secreted only a low concentration of the protein. The LPS mutant strains, however, both produced and secreted approximately twofold more PcrV than PAO1 (Fig. 2B), suggesting that these strains would exhibit increased cytotoxicity compared to the wild type.

• Open in new tab
• Download powerpoint

FIG. 2.

Analysis of ExoS (A) and PcrV (B) production (white bars) and secretion (black bars) from PAO1 and isogenic LPS mutant strains. The results are representative of three independent experiments. The error bars indicate standard deviations.

Cytotoxicity of PA01 and LPS mutants toward BEAS-2B epithelial cells.Delivery of effector proteins into target cells via the TTSS is central to acute infection by P. aeruginosa. To determine if observed differences in TTSS mRNA expression and protein production and secretion between PAO1 and isogenic mutants correlated with an increase in mammalian-cell death, cytotoxicity was determined as described in Materials and Methods. Increased cell death (compared to PAO1) was observed when the mutant strains were incubated with lung epithelial cells for 6 h. Consistent with Q-PCR and immunoblot analyses, the mutant wbpM and rmlC strains, which both lack the B-band O antigen, produced the greatest increases in cytotoxicity in vitro. Exposure of the BEAS-2B cell line to the gmd, wbpM, or rmlC mutant strain increased cytotoxicity by 8, 14, and 46%, respectively, compared to the wild-type PAO1 (Fig. 3).

• Open in new tab
• Download powerpoint

FIG. 3.

Cytotoxicity assay using BEAS-2B lung epithelial cells and standardized inocula of PAO1 and isogenic LPS O-antigen mutant strains. Black bars, 4 h postinfection; white bars, 6 h postinfection. The results are representative of at least two independent experiments carried out in triplicate. The error bars indicate standard deviations.

In vivo murine study.To assess if this dramatic difference in cytotoxicity between the strains was maintained in vivo, a murine study using BALB/c mice was carried out to evaluate lung injury. Two indicators of lung injury were measured, lung edema and epithelium permeability, as well as low body temperature (an indicator of bacteremia). Figure 4A demonstrates that the mice infected with LPS mutant strains demonstrated increased lung weight compared to those instilled with PAO1, suggesting that the mutant strains induced greater lung injury than the wild type. Consistent with these observations, measurement of epithelial permeability also demonstrated that the LPS mutants caused significantly more lung epithelial permeability than was elicited by PAO1 (Fig. 4B). All LPS mutant strains exhibited an approximately twofold increase in both lung edema and epithelial damage, suggesting that the alteration in the LPS O antigen resulted in increased bacterially induced lung injury. Mice instilled with the mutant strains also demonstrated considerably lower body temperatures than those infected with the wild-type strain, suggesting that these mice were also severely bacteremic (Fig. 4C).

• Open in new tab
• Download powerpoint

FIG. 4.

In vivo evaluation of lung injury in BALB/c mice infected with PAO1 and an isogenic LPS mutant. (A) Measure of lung edema. (B) Epithelial permeability measurement. (C) Body temperature. The results represent the average measurement for four mice in each group. The error bars indicate standard deviations.

LPS characterization of P. aeruginosa clinical isolates.From our initial studies, the B-band O antigen appeared to be a critical determinant affecting type III gene expression and protein secretion. To determine if loss of this O antigen also affected cytotoxin secretion by clinical isolates, three strains encoding the potent type III cytotoxin ExoU were isolated from patient respiratory samples and examined, in addition to the virulent laboratory strain PA103. As a control, the original laboratory strains used in this study were also analyzed. When the strains were cultured as biofilms and analyzed in a blind test, PAO1 and its iosogenic gmd mutant strain agglutinated with anti-O5 monoclonal antibody, while the other two LPS mutant strains lacking the B-band O antigen exhibited nontypeable phenotypes, confirming the ability of the test to differentiate between serotypes. The clinical isolates CI1 to -3, cultured as biofilms, each agglutinated with anti-O11 monoclonal antibody, similar to the virulent laboratory strain PA103 (Table 3). When each of these clinical isolates was cultured planktonically, it exhibited a nontypeable B-band serotype.

Analysis of TTSS cytotoxin expression, production, and secretion by P. aeruginosa clinical isolates.Since Western immunoblotting analysis had previously provided an excellent indication of cytotoxic potential, this approach was used to examine type III secretion by biofilm and planktonic cultures of each of the clinical strains. For planktonic cultures, intracellular and secreted fractions were separated, and analysis of the separated fractions was then performed. Planktonic strains (containing no B-band O antigen) produced (data not shown) and secreted (Fig. 5) both ExoU and PcrV proteins. Using samples normalized for total protein, immunoblot analysis was also performed on the clinical isolates cultured as biofilms (containing the 011 B-band O antigen). Under these conditions, all strains exhibited barely detectable or undetectable ExoU and PcrV proteins (Fig. 5).

• Open in new tab
• Download powerpoint

FIG. 5.

Immunoblot analysis of ExoU (A) and PcrV (B) production and secretion by clinical isolates cultured as either planktonic or biofilm cultures. The results are representative of at least three independent experiments. The error bars indicate standard deviations.