Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Soong, G. Articles by Prince, A. S. Search for Related Content PubMed PubMed Citation Articles by Soong, G. Articles by Prince, A. S.

 Previous Article  |  Next Article 

Journal of Bacteriology, April 2008, p. 2814-2821, Vol. 190, No. 8
0021-9193/08/$08.00+0     doi:10.1128/JB.01567-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

The Type III Toxins of Pseudomonas aeruginosa Disrupt Epithelial Barrier Function

Grace Soong, Dane Parker, Mariah Magargee, and Alice S. Prince^*

Department of Pediatrics and Pharmacology, College of Physicians & Surgeons, Columbia University, New York, New York 10032

Received 28 September 2007/ Accepted 20 December 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The type III secreted toxins of Pseudomonas aeruginosa are important virulence factors associated with clinically important infection. However, their effects on bacterial invasion across mucosal surfaces have not been well characterized. One of the most commonly expressed toxins, ExoS, has two domains that are predicted to affect cytoskeletal integrity, including a GTPase-activating protein (GAP) domain, which targets Rho, a major regulator of actin polymerization; and an ADP-ribosylating domain that affects the ERM proteins, which link the plasma membrane to the actin cytoskeleton. The activities of these toxins, and ExoS specifically, on the permeability properties of polarized airway epithelial cells with intact tight junctions were examined. Strains expressing type III toxins altered the distribution of the tight junction proteins ZO-1 and occludin and were able to transmigrate across polarized airway epithelial monolayers, in contrast to STY mutants. These effects on epithelial permeability were associated with the ADP-ribosylating domain of ExoS, as bacteria expressing plasmids lacking expression of the ExoS GAP activity nonetheless increased the permeation of fluorescent dextrans, as well as bacteria, across polarized airway epithelial cells. Treatment of epithelial cells with cytochalasin D depolymerized actin filaments and increased permeation across the monolayers but did not eliminate the differential effects of wild-type and toxin-negative mutants on the epithelial cells, suggesting that additional epithelial targets are involved. Confocal imaging studies demonstrated that ZO-1, occludin, and ezrin undergo substantial redistribution in human airway cells intoxicated by ExoS, -T, and -Y. These studies support the hypothesis that type III toxins enhance P. aeruginosa's invasive capabilities by interacting with multiple eukaryotic cytoskeletal components.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Pseudomonas aeruginosa produces four type III toxins that have each been demonstrated to function as a virulence factor (7, 16, 29, 32). These toxins are variably expressed in clinical settings, and they are frequently associated with invasive clinical isolates. More than 80% of P. aeruginosa strains isolated from acute infections, including ventilator-associated pneumonia and sepsis, express these toxins (13). In the setting of chronic infection, as occurs in cystic fibrosis, while the genes are usually conserved, toxin expression is significantly decreased (4, 11, 25). The activity of each of the toxins is dependent upon the host cell type; for example, epithelial cells are substantially more sensitive to ExoS intoxication than are fibroblasts or macrophages (26). The biochemistry of each of the four toxins has been studied in detail. ExoU, a phospholipase (28), is the most destructive of these toxins but is expressed by only a minority of clinical isolates, i.e., approximately one-third, according to a recent survey (29). ExoY is an adenylate cyclase which has a modest contribution to virulence in a mouse model of pneumonia (32) but may potentiate the activities of other toxins. ExoS and ExoT are related bifunctional enzymes expressed by the majority of strains (2, 6). Although ExoT does not have as dramatic effects in overall virulence as ExoS does (16, 29, 32), it has several novel activities that include inhibiting bacterial internalization into eukaryotic cells through its N-terminal Rho GTPase-activating activity and ADP-ribosylating (ADPR) activity as well as the ability to interact with the host Cbl-b ubiquitin ligase (1). ExoS has been studied extensively, and the biochemistry of its interactions with several eukaryotic targets has been defined precisely at the molecular level (2). The N-terminal domain of ExoS has GTPase-activating protein (GAP) properties and preferentially targets the Rho GTPases which maintain the actin cytoskeleton (2, 30). The C terminus has 14-3-3-dependent ADPR activity directed at itself (24, 34) as well as Ras (10, 14), Ral, Rabs (8), the ERM proteins (ezrin, radixin, and moesin) (17), and other targets.

The type III toxins were initially identified as virulence factors that facilitated the dissemination of P. aeruginosa from burn wounds (19). The specific roles of the individual type III toxins in the pathogenesis of pneumonia have been studied using bacterial mutants constructed in different genetic backgrounds and tested in animal models of infection, with some using intact hosts (16, 29) and others using neutropenic animals (32). While each of the type III toxins has some role in virulence, biological effects vary depending upon the route of bacterial delivery and the nature of the host. A consistent finding throughout these studies is the importance of ExoS in virulence (13, 32). The association of ExoS and systemic infection initiated from a site of mucosal colonization, such as pneumonia, suggests that ExoS could facilitate bacterial invasion through the epithelium. Whether the cytoskeletal effects of ExoS observed in CHO and HeLa cells, such as rounding and changes in actin filaments (9), would cause a permeability defect in polarized cells with tight junctions has not been established clearly. The ADPR domain of ExoS targets the ERM proteins, which link actin to the plasma membrane, and also causes cell rounding (17). Ezrin, the most abundant of the ERM proteins in epithelial cells, has amino and carboxy termini that can interact. Phosphorylation of ezrin is required for its maintenance in an "open" form, exposing domains that bind F-actin (33). ExoS-mediated ADPR of ezrin is predicted to impair the ezrin-F-actin interaction, which would potentially alter epithelial cytoskeletal integrity (17).

The lung is remarkably resistant to infection; airways are protected by mucociliary clearance and secreted antimicrobial peptides, and tight junctions prevent bacterial invasion between adjacent cells. As a consequence of airway inflammation, the epithelial barrier is likely to be altered both by transmigration of polymorphonuclear leukocytes moving into the airway and by cell damage due to toxins elaborated by luminal bacteria (12). Few P. aeruginosa cells attach to intact epithelial cells (27), although in the presence of trauma, glycolipid receptors for P. aeruginosa pili are exposed and provide the initial host-pathogen attachment required for type III intoxication of the host cell, which could then affect the structural integrity of the epithelial cytoskeleton (3). While actin is clearly a critical component of the cytoskeletal scaffold, many other proteins make up the epithelial tight and adherens junctions that block bacterial invasion. Several proteins, including occludin, span the paracellular junction, providing a physical barrier to organisms (5). We postulated that invasive P. aeruginosa, through the expression of type III toxins, modulates the epithelial barrier by targeting junctional as well as cytoskeletal components and thus facilitates invasion across the mucosal surface.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cell lines, bacterial strains, and reagents. 16HBE human airway epithelial cells (D. Gruenert, California Pacific Medical Center Research Institute) were grown as previously detailed (22). Strains PAK, PAKS, PAKT, and PAKSTY were provided by S. Lory (Harvard Medical School). Equivalent rates of growth in LB were documented. ExoS-HApUCP plasmid constructs with point mutations in either one of or both the GAP (R146K) and ADPR (E381D) domains (G⁺ A⁺, G⁺ A^–, G^– A⁺, or G^– A^–) (18) were provided by J. Barbieri (Medical College of Wisconsin) and used to transform PAKSTY. Equivalent levels of ExoS expression were confirmed by Western hybridization. Bacteria were grown in LB supplemented with 5 mM EDTA and 300 µg/ml carbenicillin, and supernatants were concentrated 10-fold for immunoblotting with an ExoS antiserum provided by C. Mody (University of Calgary). Green fluorescent protein (GFP)-expressing strains were also grown with 300 µg/ml carbenicillin. Mouse monoclonal antibodies used were anti-E-cadherin (BD Biosciences), anti-ZO1, anti-occludin (both from Invitrogen, Zymed), anti-ezrin (Chemicon), anti-phospho-ERM (Cell Signaling), anti-hemagglutinin (Santa Cruz Biotechnology), and anti-actin (Sigma). Rabbit polyclonal antibodies used were anti-ZO-1 and anti-occludin (Invitrogen, Zymed). All Alexa Fluor-conjugated secondary antibodies were from Invitrogen, Molecular Probes.

Dextran permeation and bacterial transmigration. Confluent monolayers of 16HBE cells, polarized at the air-liquid interface, were grown on 6.5-mm-diameter, 3-µm-pore-size Transwell-Clear filters (Corning Costar) for 5 days. Eighteen hours prior to stimulation, cells were fed basally with 0.5 ml and apically with 45 µl of minimum essential medium plus 10% fetal calf serum. After 4 h of cell stimulation with 5 µl of bacteria (5 x 10⁷ to 5 x 10⁸ CFU/ml) from an overnight culture in LB or with medium alone as a control, dextran-fluorescein (molecular weight [MW], 3,000 or 40,000) was added apically for 1 h, and fluorescence in the basal compartment was monitored at an excitation wavelength of 485 nm and emission wavelength of 535 nm on a SpectraFluor Plus fluorimeter (Tecan). Bacteria present in the basal compartment were quantified by dilution and plating on LB agar. The integrity of the monolayers was confirmed by trypan blue exclusion and lactate dehydrogenase (LDH) cytotoxicity assay. As a positive control, cells were treated for 1 h with 0.02% EGTA, an extracellular Ca²⁺ chelator which increases paracellular permeability by disrupting cadherin-mediated cell-cell adhesion. Measurements were done on sextuplicate wells (triplicate wells for transmigration studies), and experiments were performed at least three times.

Confocal microscopy and immunofluorescence. 16HBE cells, as described above, were stained as follows. Cells were fixed with 4% paraformaldehyde for 15 min at room temperature and incubated with 5% normal serum blocking solution including 0.2% Triton X-100 (TX-100) to permeabilize the cells for 15 min at room temperature. Primary antibodies diluted in blocking solution with TX-100 were added for 1 h at room temperature or overnight at 4°C, followed by three 5-min washes. Alexa Fluor 488-, 546-, and 647-conjugated secondary antibodies (Invitrogen, Molecular Probes) were added for 1 h, and cells were washed three times. Filters were removed from Transwells by use of a scalpel and mounted with Vectashield with DAPI (4',6'-diamidino-2-phenylindole; Vector Laboratories) onto glass slides. In biotin labeling experiments, 1 mg/ml EZ-Link sulfo-NHS-LC-biotin (Pierce) was added apically and incubated at 4°C for 30 min, followed by three phosphate-buffered saline (PBS) washes before fixing of the cells. Alexa Fluor 555-conjugated streptavidin (Invitrogen, Molecular Probes) was added for detection of biotin. For actin staining, Alexa Fluor 568-conjugated phalloidin (Invitrogen, Molecular Probes) was used.

Preparation of TX-100-insoluble and -soluble fractions. 16HBE cells, grown and stimulated as described above, were washed and lysed in PBS containing 1% TX-100 with protease inhibitors for 30 min on ice. Lysed cells were scraped from the plates and transferred to tubes for centrifugation at 13,000 x g for 10 min. Supernatants of TX-100-soluble fractions were saved. The pellet was washed two times with PBS and resuspended in radioimmunoprecipitation assay buffer (50 mM Tris, 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, 1 mM EDTA) with sonication to solubilize the TX-100-insoluble proteins.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Type III toxins affect the integrity of the epithelial junctions. To compare the effects of specific type III toxins on epithelial barrier function, we first documented that wild-type and mutant strains had roughly equivalent growth in vitro and that the ExoS expression from the plasmid constructs was equivalent (Fig. 1). To determine if type III toxins directly alter the integrity of the epithelial tight junctions of polarized airway epithelial cells, we monitored the ability of biotin labeled with streptavidin to intercalate between adjacent airway cells which had been exposed to PAK under routine growth conditions, compared with PAKSTY or a medium control (Fig. 2). The biotin labeling was exclusively apical and superficial in the polarized 16HBE cells exposed to medium or the STY mutant. In contrast, biotin intercalated between the cells after 4 h of exposure to PAK. An en face view of the monolayers contrasts the accumulation of biotin-streptavidin on the apical surface of the epithelial cells under control conditions or following exposure to the STY mutant-treated cells with the penetration of the label around the cells exposed to PAK (Fig. 2B).

View larger version (17K): [in this window] [in a new window]

FIG. 1. Characterization of P. aeruginosa strains. (A) Growth curves for P. aeruginosa strains in LB. (B) Immunoblot using ExoS antiserum to detect ExoS production in PAKSTY expressing pUCPExoS plasmid constructs (STY G/A) with point mutations in either (+/– or –/+), both (–/–), or neither (+/+) of the GAP (G) or ADPR (A) domains.

View larger version (30K): [in this window] [in a new window]

FIG. 2. Type III toxins increase epithelial permeability to biotin. Confluent monolayers of 16HBE cells exposed to medium, PAK, or PAKSTY for 4 h were treated with biotin, which was detected with streptavidin and imaged in a confocal microscope. (A) z sections demonstrating the entire thickness of the monolayer are shown. (B) En face (x-y) images of an apical layer.

To better characterize the effect of the type III toxins on epithelial permeability, we monitored the movement of fluorescently labeled 3,000-MW and 40,000-MW dextran across the 16HBE cell monolayers (Fig. 3A). Following 4 h of exposure to PAK or the T mutant, there was a much greater increase in the movement of both 3,000- and 40,000-MW dextran through the paracellular spaces than after exposure to either the S or STY mutant. The effect of PAK on dextran permeation was roughly 50% of the increase associated with the chelation of Ca²⁺ by EGTA, which opens epithelial tight junctions. Permeability to 40,000-MW dextran for PAK- and T mutant-stimulated cells was also significantly increased compared to that for a medium control as well as that for either the S or the STY mutant, but overall fluorescence was less than that with the 3,000-MW dextran. As controls to verify the integrity of the epithelial monolayers, wells were stained with trypan blue to document the integrity of the cells (data not shown), and LDH levels measured to document the integrity of the monolayers exposed to the medium control and to the various strains of P. aeruginosa under these conditions were comparable (Fig. 3B).

View larger version (34K): [in this window] [in a new window]

FIG. 3. Effects of type III toxins on confluent and polarized 16HBE cells exposed to medium, PAK, PAKSTY, PAKS, or PAKT for 4 h. (A) Apical-to-basal penetration of 3,000- or 40,000-MW fluorescently labeled dextran was quantified in monolayers exposed to the indicated strains. *, P < 0.001 compared to PAK or PAKT; NS, not significant compared to PAK. (B) LDH (kit from Roche) in supernatants collected from 16HBE cells exposed to bacteria for 4 h was quantified, and its activity compared to that in medium alone was plotted. Quantification of bacteria (CFU/ml) (C) and fluorescent dextran (D) found in the basal compartment of the wells was performed 4 h after inoculation of the apical chamber. The total bacterial inoculum added to the 16HBE cells was constant, but the ratio of PAKGFP to STYGFP was adjusted as labeled, and numbers of STYGFP (plated on LB plus 300 µg/ml carbenicillin) as well as total bacteria (plated on LB) in the basal compartment were enumerated. Confocal z sections (E) and x-y sections (F) of polarized 16HBE cells exposed to GFP-expressing PAK or PAKSTY (green) and stained for E-cadherin (red) are shown.

Quantification of the number of organisms able to transmigrate across the epithelial monolayer similarly indicated that production of type III toxins enabled the wild-type strain PAK to cross to the basal compartment more readily than the STY mutant (Fig. 3C). For PAK, 6% of a 5 x 10⁸-CFU inoculum was recovered in the basal compartment of Transwells after a 4-h incubation, as opposed to 0.001% of a 5 x 10⁸-CFU inoculum of PAKSTY. The addition of 10% PAK to the total 5 x 10⁸-CFU inoculum (5 x 10⁷ PAK cells plus 4.5 x 10⁸ STY mutant cells) increased the transmigration of PAKSTY 500-fold, indicating that the effects of the toxins are on the epithelium and are not limited to an individual organism. Similar effects were observed on the permeation of dextran (Fig. 3D); PAK but not PAKSTY significantly increased the permeability of the monolayers for 3,000-MW dextran, and the addition of 10% PAK to the total bacterial inoculum containing 90% PAKSTY significantly increased the amount of dextran which crossed the epithelial barrier.

The process of bacterial transmigration across the monolayer was visualized by confocal imaging. PAKGFP was seen intercalating between the epithelial cells and reached the basal portions of the wells (Fig. 3E). PAKGFP outlined the "chicken wire" distribution of E-cadherin along the cell-cell borders (Fig. 3F). In contrast, STYGFP was not seen at the basal aspects of the monolayers but remained in clumps on the apical surface (Fig. 3E and F).

The ADPR moiety of ExoS confers changes in epithelial permeability. ExoS has two domains that are independently capable of affecting components of the cytoskeleton, targeting the GTPases critical to actin polymerization and depolymerization as well as ADPR activity that causes rounding in nonpolarized cells. Some of the targets of ExoS ADPR activity, such as the small GTPases as well as the ERM proteins, may also affect Rho signaling as well as interactions with actin (15). We took advantage of mutants provided by J. Barbieri to test the effect(s) of the GAP and ADPR domains individually on the permeability properties of polarized 16HBE cells. Expression of the ADPR domain alone in a PAKSTY background was sufficient to increase permeability to 3,000-MW dextran (Fig. 4A). Corresponding changes in bacterial transmigration were observed (Fig. 4B) with the constructs expressing the ADPR domain and were associated with 2.5-fold more transmigration than that seen with constructs lacking ADPR activity. Expression of the GAP domain minimally increased permeability to dextran and bacterial transmigration, suggesting that the ADPR activity of ExoS provides the major effect on epithelial barrier function.

View larger version (10K): [in this window] [in a new window]

FIG. 4. Effects of individual GAP and ADPR domains of ExoS. The amount of fluorescent dextran found in the basal chamber (A) and bacterial transmigration (B) across Transwells containing a confluent monolayer of 16HBE cells exposed for 4 h to PAKSTY expressing both the GAP (G+) and ADPR (A+) domains of ExoS, either the GAP or the ADPR domain, or neither (G– A–) are shown.

Effects of type III toxins on actin. ExoS and ExoT target Rho GTPases, both directly through the GAP domain of ExoS and indirectly through ADP ribosylation of associated proteins (30). The Rho GTPases are critical for actin polymerization and cytoskeletal integrity (20). 16HBE cells were screened for changes in the distribution of actin following exposure to PAK or the STY mutant (Fig. 5A). The most prominent effects of the toxins on actin were observed at the basal aspects of the cells; the stress fibers that were readily apparent in the control monolayers were substantially interrupted in cells exposed to PAK but remained intact in cells exposed to the STY mutant. To determine which biological changes in barrier function can be ascribed to interruption of actin polymerization, the monolayers were treated with 20 µM cytochalasin D (cyto D), a dose expected to cap and sever the barbed ends of the actin filaments (15). Cyto D disrupted the stress fibers (Fig. 5B) and increased dextran permeation (Fig. 5C) and bacterial transmigration (Fig. 5D) across the monolayers. However, cyto D did not completely eliminate the differences in permeability attributed to type III toxin production; the increase in permeability associated with the STY, S, or T mutant even in the presence of cyto D was not equivalent to that of the wild-type PAK strain, suggesting that the type III toxins have targets other than actin that affect the epithelial permeability barrier (Fig. 5C and D). Note that the ExoT mutant behaved as would be predicted for an "anti-internalization factor" in that the T mutant was similar to PAK in its effects on dextran permeation but was associated with increased numbers of bacteria able to migrate across the epithelial barrier, an effect that was lost when actin was disrupted (Fig. 5D).

View larger version (42K): [in this window] [in a new window]

FIG. 5. Effects of type III toxins on actin in 16HBE cells. (A) Confocal images of the basal portions of polarized 16HBE cells grown on Transwells, exposed for 4 h to medium, PAK, PAKS, or PAKSTY, and stained with phalloidin. (B) Effects of cyto D on cells stimulated as described for panel A. Permeation of apically applied fluorescent 3,000-MW dextran (C) or of bacteria crossing monolayers of confluent polarized 16HBE cells pretreated with 20 µM cyto D and exposed to medium, PAK, PAKSTY, PAKS, or PAKT (D) is shown.

Effects of type III toxins on junctional proteins and ezrin. We postulated that the most apical components of the tight junction that constitute the initial barrier to bacterial invasion from the airway lumen, such as ZO-1 and the membrane-spanning protein occludin, were likely to be affected by type III toxins. Imaging 16HBE cells following exposure to PAK or PAKSTY, we found that the ZO-1 chicken wire pattern delineating the cellular borders, which was apparent in the control monolayers, was lost in cells exposed to the type III toxins (Fig. 6A). ZO-1 seemed to move from its paracellular distribution into the cytoplasm of the PAK- but not the STY mutant-exposed monolayers. The addition of cyto D did not alter the distribution of ZO-1 in these cells (data not shown), suggesting that these changes in ZO-1 were independent of alterations of the actin cytoskeleton caused by intoxication.

View larger version (50K): [in this window] [in a new window]

FIG. 6. Type III toxins alter junctional components. (A) Confocal images of confluent, polarized 16HBE cells exposed to medium, PAK, or PAKSTY for 4 h and stained for ZO-1, occludin, or ezrin. Note that in ezrin-stained PAK-treated wells, GFP-expressing bacteria are also visible. (B) TX-100-insoluble (TI) and TX-100-soluble (TS) proteins were isolated from lysates exposed to medium (M), PAK, or PAKSTY and immunoblotted (WB) as labeled. (C) 16HBE cells were stimulated with medium (M), PAK, or PAKSTY for the time indicated, lysed with 60 mM n-octyl beta-D-glucopyranoside, and immunoblotted (WB) as labeled.

The distributions of occludin and ezrin were also substantially altered in cells exposed to PAK, but much less so in cells exposed to the STY mutant (Fig. 6A). This observation was confirmed by monitoring changes in the distribution of occludin from the lipid raft (TX-100-insoluble) fraction of polarized monolayers into the TX-100-soluble or cytoplasmic fraction following exposure to PAK. PAK-exposed cells had decreased amounts of membrane-associated occludin. There were also several occludin cleavage products detected, as described for other cell types (Fig. 6B) (5). ExoS ADP ribosylates the ERM proteins, which is predicted to prevent ezrin phosphorylation and interfere with its links to actin (20). Ezrin phosphorylation was decreased in the PAK-exposed cells compared with that in cells exposed to PAKSTY (Fig. 6C). Since these monolayers retain barrier function compared to the EGTA control shown in Fig. 3, the effects of the type III toxins appear to be targeted to specific cytoskeletal and junctional proteins.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The type III toxins of P. aeruginosa are a complex group of proteins with the potential for many interactions with eukaryotic targets. Epithelial cells are especially sensitive to the effects of these toxins (26). Recent studies documented that P. aeruginosa airway infection alters the permeability properties of the epithelial junction (23). However, few studies have identified specific epithelial or junctional protein targets in polarized cells for ExoS and the other toxins of this group. The biochemistry of these toxins has been defined at the molecular level, although their contribution to the pathogenesis of pneumonia is less well defined. Gastrointestinal pathogens that have well-characterized type III toxins (such as Yersinia or Salmonella, for example) are genetically adapted for proliferation in an intracellular milieu and use type III toxins to take over the host cell (31). P. aeruginosa is primarily an extracellular opportunist, and thus it seems more likely that for P. aeruginosa, toxins such as ExoS contribute to invasion from mucosal surfaces and others, such as ExoT, facilitate persistence in the bloodstream through anti-internalization and other activities (13). This would be consistent with clinical data indicating that invasive isolates of P. aeruginosa generally have conserved expression of the type III toxins (13, 16, 29).

The data presented here suggest that type III toxins, particularly ExoS, contribute to the ability of P. aeruginosa to cross the epithelial barrier, presumably through the cell-cell junctions, as evidenced by their paracellular distribution in infected monolayers and the parallel and consistent effects of ExoS on the permeation of dextrans and bacteria through intact epithelial monolayers. We have previously shown that under similar conditions of bacterial exposure, epithelial cells in an intact monolayer do not undergo apoptosis to any significant extent (22), nor did we detect substantial amounts of necrosis by trypan blue staining or LDH assay. The actual mechanisms involved in the effects of the type III toxins on epithelial permeability remain undefined. Much of the biochemistry of ExoS has focused upon its interactions with GTPases that regulate actin polymerization (21). The GAP domain clearly affects Rho and similar GTPases, and the ADPR moiety can independently alter actin biology, possibly by preventing ERM protein phosphorylation and the subsequent ability to interact with actin (17). As expected, type III toxin-secreting PAK had a major effect on the actin cytoskeleton, particularly on the stress fibers that accumulate in the basal compartment of polarized epithelial cells. However, much of this effect was associated with the expression of ExoT and/or ExoY, as the S mutant altered this portion of the cytoskeleton as much as the wild-type strain PAK did (data not shown).

The type III toxins have a substantial effect on the epithelial barrier, as monitored by dextran permeation and bacterial movement across intact monolayers. The permeability effect of ExoS in vitro was associated with the ADPR moiety of the protein, with a modest contribution of the GAP activity, despite the expected effects of GAP activation on Rho and, subsequently, actin. The importance of the ADPR moiety to virulence was previously described by Shaver and Hauser (29). It is interesting that recent studies documented the intracellular targeting of ExoS to membranes, which would be consistent with an effect on junctional proteins (35). The type III toxin effect on permeability is at least partially due to targets other than actin. Inhibition of actin polymerization with cyto D did not completely eliminate differences in permeability attributable to the toxins, suggesting alternative mechanisms of action. While we did not define additional ExoS targets biochemically, confocal images and immunoblots of the epithelial cells demonstrated that intoxicated cells have major changes in the distributions of the tight junction proteins ZO-1 and occludin that could affect bacterial transmigration between cells. In addition, ezrin, a known ExoS target, is less phosphorylated in ExoS-exposed airway cells and moves from the membrane into the cytosol following exposure to bacteria, which would be expected to destabilize the plasma membrane-cytoskeletal links.

The interactions of the type III toxins and airway cells are important in the pathogenesis of infection and, appropriately, have been studied intensively for over a decade. Many of the molecular mechanisms of their actions on eukaryotic targets have been defined. However, it appears that additional biologically important targets of these toxins, particularly ExoS, are present in the junctional complex of polarized airway epithelial cells, which may help to explain the close association of these toxins and invasive infection.

 
We thank J. Barbieri, Medical College of Wisconsin, and S. Lory, Harvard Medical School, for generously providing strains for these studies and C. Mody at the University of Calgary for providing ExoS antiserum.

This work was supported by grant RO1 DK39693 to A.S.P.

 
* Corresponding author. Mailing address: Columbia University, 650 West 168th Street, BB4-416, New York, NY 10032. Phone: (212) 305-4193. Fax: (212) 342-5728. E-mail: asp7{at}columbia.edu

Published ahead of print on 28 December 2007.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Balachandran, P., L. Dragone, L. Garrity-Ryan, A. Lemus, A. Weiss, and J. Engel. 2007. The ubiquitin ligase Cbl-b limits Pseudomonas aeruginosa exotoxin T-mediated virulence. J. Clin. Investig. 117:419-427.[CrossRef][Medline]
2
2. Barbieri, J. T. 2000. Pseudomonas aeruginosa exoenzyme S, a bifunctional type-III secreted cytotoxin. Int. J. Med. Microbiol. 290:381-387.[Medline]
3
3. Comolli, J. C., L. L. Waite, K. E. Mostov, and J. N. Engel. 1999. Pilus binding to asialo-GM1 on epithelial cells can mediate cytotoxicity or bacterial internalization by Pseudomonas aeruginosa. Infect. Immun. 67:3207-3214.[Abstract/Free Full Text]
4
4. Dacheux, D., I. Attree, and B. Toussaint. 2001. Expression of ExsA in trans confers type III secretion system-dependent cytotoxicity on noncytotoxic Pseudomonas aeruginosa cystic fibrosis isolates. Infect. Immun. 69:538-542.[Abstract/Free Full Text]
5
5. Feldman, G. J., J. M. Mullin, and M. P. Ryan. 2005. Occludin: structure, function and regulation. Adv. Drug Deliv. Rev. 57:883-917.[CrossRef][Medline]
6
6. Feltman, H., G. Schulert, S. Khan, M. Jain, L. Peterson, and A. R. Hauser. 2001. Prevalence of type III secretion genes in clinical and environmental isolates of Pseudomonas aeruginosa. Microbiology 147:2659-2669.[Abstract/Free Full Text]
7
7. Frank, D. W. 1997. The exoenzyme S regulon of Pseudomonas aeruginosa. Mol. Microbiol. 26:621-629.[CrossRef][Medline]
8
8. Fraylick, J. E., E. A. Rucks, D. M. Greene, T. S. Vincent, and J. C. Olson. 2002. Eukaryotic cell determination of ExoS ADP-ribosyltransferase substrate specificity. Biochem. Biophys. Res. Commun. 291:91-100.[CrossRef][Medline]
9
9. Frithz-Lindsten, E., Y. Du, R. Rosqvist, and A. Forsberg. 1997. Intracellular targeting of exoenzyme S of Pseudomonas aeruginosa via type III-dependent translocation induces phagocytosis resistance, cytotoxicity and disruption of actin microfilaments. Mol. Microbiol. 25:1125-1139.[CrossRef][Medline]
10
10. Ganesan, A. K., D. W. Frank, R. P. Misra, G. Schmidt, and J. T. Barbieri. 1998. Pseudomonas aeruginosa exoenzyme S ADP-ribosylates Ras at multiple sites. J. Biol. Chem. 273:7332-7337.[Abstract/Free Full Text]
11
11. Goodman, A. L., B. Kulasekara, A. Rietsch, D. Boyd, R. S. Smith, and S. Lory. 2004. A signaling network reciprocally regulates genes associated with acute infection and chronic persistence in Pseudomonas aeruginosa. Dev. Cell 7:745-754.[CrossRef][Medline]
12
12. Guttman, J. A., F. N. Samji, Y. Li, A. W. Vogl, and B. B. Finlay. 2006. Evidence that tight junctions are disrupted due to intimate bacterial contact and not inflammation during attaching and effacing pathogen infection in vivo. Infect. Immun. 74:6075-6084.[Abstract/Free Full Text]
13
13. Hauser, A. R., E. Cobb, M. Bodi, D. Mariscal, J. Valles, J. N. Engel, and J. Rello. 2002. Type III protein secretion is associated with poor clinical outcomes in patients with ventilator-associated pneumonia caused by Pseudomonas aeruginosa. Crit. Care Med. 30:521-528.[CrossRef][Medline]
14
14. Henriksson, M. L., C. Sundin, A. L. Jansson, A. Forsberg, R. H. Palmer, and B. Hallberg. 2002. Exoenzyme S shows selective ADP-ribosylation and GTPase-activating protein (GAP) activities towards small GTPases in vivo. Biochem. J. 367:617-628.[CrossRef][Medline]
15
15. Janmey, P. A. 1998. The cytoskeleton and cell signaling: component localization and mechanical coupling. Physiol. Rev. 78:763-781.[Abstract/Free Full Text]
16
16. Lee, V. T., R. S. Smith, B. Tummler, and S. Lory. 2005. Activities of Pseudomonas aeruginosa effectors secreted by the type III secretion system in vitro and during infection. Infect. Immun. 73:1695-1705.[Abstract/Free Full Text]
17
17. Maresso, A. W., M. R. Baldwin, and J. T. Barbieri. 2004. Ezrin/radixin/moesin proteins are high affinity targets for ADP-ribosylation by Pseudomonas aeruginosa ExoS. J. Biol. Chem. 279:38402-38408.[Abstract/Free Full Text]
18
18. Maresso, A. W., M. J. Riese, and J. T. Barbieri. 2003. Molecular heterogeneity of a type III cytotoxin, Pseudomonas aeruginosa exoenzyme S. Biochemistry 42:14249-14257.[CrossRef][Medline]
19
19. Nicas, T. I., J. Bradley, J. E. Lochner, and B. H. Iglewski. 1985. The role of exoenzyme S in infections with Pseudomonas aeruginosa. J. Infect. Dis. 152:716-721.[Medline]
20
20. Nobes, C. D., and A. Hall. 1995. Rho, Rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81:53-62.[CrossRef][Medline]
21
21. Pederson, K. J., A. J. Vallis, K. Aktories, D. W. Frank, and J. T. Barbieri. 1999. The amino-terminal domain of Pseudomonas aeruginosa ExoS disrupts actin filaments via small-molecular-weight GTP-binding proteins. Mol. Microbiol. 32:393-401.[CrossRef][Medline]
22
22. Rajan, S., G. Cacalano, R. Bryan, A. J. Ratner, C. U. Sontich, A. van Heerckeren, P. Davis, and A. Prince. 2000. Pseudomonas aeruginosa induction of apoptosis in respiratory epithelial cells. Analysis of the effects of cystic fibrosis transmembrane conductance regulator dysfunction and bacterial virulence factors. Am. J. Respir. Cell Mol. Biol. 23:304-312.[Abstract/Free Full Text]
23
23. Rejman, J., S. Di Gioia, A. Bragonzi, and M. Conese. 2007. Pseudomonas aeruginosa infection destroys the barrier function of lung epithelium and enhances polyplex-mediated transfection. Hum. Gene Ther. 18:642-652.[CrossRef][Medline]
24
24. Riese, M. J., U.-M. Goehring, M. E. Ehrmantraut, J. Moss, J. T. Barbieri, K. Aktories, and G. Schmidt. 2002. Auto-ADP-ribosylation of Pseudomonas aeruginosa ExoS. J. Biol. Chem. 277:12082-12088.[Abstract/Free Full Text]
25
25. Roy-Burman, A., R. H. Savel, S. Racine, B. L. Swanson, N. S. Revadigar, J. Fujimoto, T. Sawa, D. W. Frank, and J. P. Wiener-Kronish. 2001. Type III protein secretion is associated with death in lower respiratory and systemic Pseudomonas aeruginosa infections. J. Infect. Dis. 183:1767-1774.[CrossRef][Medline]
26
26. Rucks, E. A., J. E. Fraylick, L. M. Brandt, T. S. Vincent, and J. C. Olson. 2003. Cell line differences in bacterially translocated ExoS ADP-ribosyltransferase substrate specificity. Microbiology 149:319-331.[Abstract/Free Full Text]
27
27. Saiman, L., and A. Prince. 1993. Pseudomonas aeruginosa pili bind to asialoGM1 which is increased on the surface of cystic fibrosis epithelial cells. J. Clin. Investig. 92:1875-1880.[Medline]
28
28. Sato, H., and D. W. Frank. 2004. ExoU is a potent intracellular phospholipase. Mol. Microbiol. 53:1279-1290.[CrossRef][Medline]
29
29. Shaver, C. M., and A. R. Hauser. 2004. Relative contributions of Pseudomonas aeruginosa ExoU, ExoS, and ExoT to virulence in the lung. Infect. Immun. 72:6969-6977.[Abstract/Free Full Text]
30
30. Sun, J., and J. T. Barbieri. 2004. ExoS Rho GTPase-activating protein activity stimulates reorganization of the actin cytoskeleton through Rho GTPase guanine nucleotide disassociation inhibitor. J. Biol. Chem. 279:42936-42944.[Abstract/Free Full Text]
31
31. Troisfontaines, P., and G. R. Cornelis. 2005. Type III secretion: more systems than you think. Physiology 20:326-339.[Abstract/Free Full Text]
32
32. Vance, R. E., A. Rietsch, and J. J. Mekalanos. 2005. Role of the type III secreted exoenzymes S, T, and Y in systemic spread of Pseudomonas aeruginosa PAO1 in vivo. Infect. Immun. 73:1706-1713.[Abstract/Free Full Text]
33
33. Yonemura, S., T. Matsui, S. Tsukita, and S. Tsukita. 2002. Rho-dependent and -independent activation mechanisms of ezrin/radixin/moesin proteins: an essential role for polyphosphoinositides in vivo. J. Cell Sci. 115:2569-2580.[Abstract/Free Full Text]
34
34. Zhang, L., H. Wang, S. C. Masters, B. Wang, J. T. Barbieri, and H. Fu. 1999. Residues of 14-3-3 zeta required for activation of exoenzyme S of Pseudomonas aeruginosa. Biochemistry 38:12159-12164.[CrossRef][Medline]
35
35. Zhang, Y., Q. Deng, and J. T. Barbieri. 2007. Intracellular localization of type III-delivered Pseudomonas ExoS with endosome vesicles. J. Biol. Chem. 282:13022-13032.[Abstract/Free Full Text]

---

Journal of Bacteriology, April 2008, p. 2814-2821, Vol. 190, No. 8
0021-9193/08/$08.00+0     doi:10.1128/JB.01567-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Goldberg, J. B., Hancock, R. E. W., Parales, R. E., Loper, J., Cornelis, P. (2008). Pseudomonas 2007. J. Bacteriol. 190: 2649-2662 [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Soong, G. Articles by Prince, A. S. Search for Related Content PubMed PubMed Citation Articles by Soong, G. Articles by Prince, A. S.