• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of annsurgLink to Publisher's site
Ann Surg. Jul 2000; 232(1): 133–142.
PMCID: PMC1421122

The Key Role of Pseudomonas aeruginosa PA-I Lectin on Experimental Gut-Derived Sepsis



To examine the effect of Pseudomonas aeruginosa on intestinal barrier function and its lethal potential when introduced into the intestinal tract of mice.

Summary Background Data

The mere presence of P. aeruginosa in the intestinal tract of critically ill patients is associated with a threefold increase in death compared with matched cohorts without this pathogen. Whether this effect is a cause or a consequence of the critically ill state has not been previously addressed.


Transepithelial electrical resistance, a measure of tight junction permeability, was evaluated in Caco-2 intestinal epithelial cells cells apically inoculated with live P. aeruginosa, exotoxin A, or purified PA-I lectin, an adhesin of P. aeruginosa. Lethality studies to P. aeruginosa were carried out in mice undergoing 30% surgical hepatectomy by injecting the bacteria or its various components directly into the cecum.


Only cells exposed to P. aeruginosa or its PA-I lectin developed alterations in barrier function. P. aeruginosa or the combination of PA-I and exotoxin A was lethal to mice when injected into the cecum after partial hepatectomy. Alterations in epithelial barrier function and death in mice were prevented when Pseudomonas was pretreated with N-acetyl D-galactosamine (GalNAc), a binder of PA-I.


P. aeruginosa may act as a pathogen in the gastrointestinal tract, resulting in altered epithelial barrier function and death in a susceptible host. The PA-I lectin of P. aeruginosa may play a key role in its pathogenicity to the intestinal epithelium by inducing a permeability defect to its cytotoxic exoproducts such as exotoxin A.

Although Pseudomonas aeruginosa is a feared hospital pathogen, it is considered to be part of the normal flora in healthy adults. 1 In fact, normal volunteers who swallow live cultures of P. aeruginosa remain healthy despite carrying the organism in their feces for up to 6 days. 2 In critically ill patients, however, the mere presence of P. aeruginosa in the proximal intestinal tract is associated with a 70% death rate, a threefold increase over age-matched critically ill patients who have negative cultures for this organism. 3 Although many have proposed that intestinal bacteria or their toxins can cause a state of systemic inflammation and sepsis, no pathogen or toxin has been identified that can independently cause this hypothetical state when introduced by means of the intestinal tract. When P. aeruginosa, however, is placed in the drinking water of immunosuppressed mice decontaminated of their indigenous flora with antibiotics, they die of gut-derived bacteremia. 4 This model does not explain the clinical observation that patients can die of sepsis presumed to be due to bacteria of gut origin while their cultures remain negative.

In the intestinal tract, the invasive and cytotoxic effects of bacteria are often judged by their ability to adhere to and alter the barrier function of cultured intestinal epithelial cells. We and others have shown that bacterial adherence alone can significantly alter the permeability of the intestinal epithelium to proinflammatory luminal macromolecules. 5,6 Recent evidence suggests that the PA-I lectin of P. aeruginosa may be an important component of its virulence phenotype by acting as an adhesin, facilitating its contact to vulnerable epithelial surfaces. 7 Therefore, the extent to which P. aeruginosa adheres to and alters the permeability of intestinal epithelia, the rate-limiting barrier of the intestinal mucosa, may have clinical relevance to its virulence.

Our laboratory has also established that the combination of a 30% surgical hepatectomy and short-term starvation alters the commensal Escherichia coli population in the cecum, resulting in an upregulation of their adherence phenotype. 8 We further established in this model that bacterial adherence to the cecal mucosa induced marked alterations in mucosal electrophysiology. 9 Therefore, the aims of this study were to characterize the effects of P. aeruginosa on the intestinal mucosal epithelium in human cultured epithelia, to determine its pathogenic effects in vivo when introduced into the mouse cecum, and to define the specific pathogenic role of PA-I lectin, a cytoplasmic adhesin of P. aeruginosa, in these systems.


Cell Culture

All experiments were performed with Caco-2/C2bbe or “C2” cells (passage 51–68), which were a generous gift of Dr. Mark Mooseker (Yale University, New Haven, CT). C2 cells are a Caco-2 subclone that are highly differentiated intestinal epithelial cells. 10 C2 cells were grown in Dulbecco’s Modified Eagle Medium supplemented with 10% fetal bovine serum (Gibco BRL, Grand Island, NY), 100,000 IU/L penicillin, 100 mg/L streptomycin, and 30 mg/mL human transferrin (Gibco BRL). C2 cells were grown at 37°C at 5% CO2, but 24 hours before all experiments, cells were transferred to antibiotic-free conditions by using the same medium minus the penicillin/streptomycin. Cell viability was assessed by measuring lactate dehydrogenase release (Sigma, St. Louis, MO), as described by Tamai et al. 11

Bacterial Strains/Components

ATCC strain 27853, a clinical isolate of P. aeruginosa from a blood culture, was used for all studies. This strain has been previously assessed for the presence of exotoxin A using both polymerase chain reaction and ADP-ribosylation assays. 12,13 The strain was assayed (erythrocyte agglutination assay) for the presence of PA-I and PA-II lectins using previously published methods. 14 The strain was found to have a PA-I concentration of 1:16 hemagglutination units. Overnight cultures of bacteria were grown in tryptic soy broth (Sigma) to the late logarithmic phase of growth, washed twice in phosphate-buffered saline (PBS), and resuspended to a final concentration of 5 × 107 colony-forming units (cfu/mL). Aliquots of bacteria were then placed in sterile cryovials and frozen with an equal volume of sterile 15% glycerol. For experiments, bacteria were thawed, centrifuged (9,000 rpm for 4 minutes), washed twice in PBS (pH 7.4) and resuspended in cell culture medium to a final concentration of 5 × 107 cfu/mL. Colony counts were routinely confirmed by diluting and plating samples on Pseudomonas isolation agar. Bacteria-free supernatant suspensions were obtained by centrifugation at 9,000 rpm for 4 minutes and passing the supernatant through a sterile 0.2-μm filter. Pseudomonal slime was obtained by differential centrifugation at 3,000 rpm for 10 minutes. A thick slime layer above the bacterial pellet was removed as the slime fraction after decantation of the supernatant. Formaldehyde killing of bacteria was carried out by suspending bacteria in a 3% solution for 1 hour, followed by vigorously washing of the bacteria in PBS, centrifugation, and resuspension in PBS. Purified PA-I and exotoxin A were obtained from Sigma.

Inoculation Onto Caco-2 Cells

Fifty microliters of live or formalin-fixed Pseudomonas in nonantibiotic cell culture medium was added to the apical side of confluent cell monolayers at a final concentration of 107 cfu/mL. Fifty microliters of bacterial components was similarly added to the apical side of the Caco-2 cells at the following concentrations: PA-I, 25 μg/mL; exotoxin A, 70 μg/mL; bacterial supernatant, 1:1 dilution; bacterial slime, 1:1 dilution. To determine the effect of the PA-I lectin of P. aeruginosa on transepithelial electrical resistance (TEER) in Caco-2 cells, both cells and bacteria were pretreated with the PA-I binding sugar N-acetyl-D-galactosamine (GalNAc) (w/v 3%). 15 Dextran (MW-4,000) (3% w/v), a nonspecific blocker of Pseudomonas adherence to epithelial cells, was also used to determine its inhibitory effect in this system. 16 For experiments assessing TEER, 50 μL of either GalNAc or dextran was added to the apical side of the transwells 24 hours before bacteria or bacterial component inoculation, and an equal volume of medium was withdrawn. Bacteria or PA-I was suspended in 50 μL of either GalNAc or dextran and added to the apical side of the cell monolayers 1 hour before TEER measurements.

Bacterial Adherence Assay

Dispersed Caco-2 cells were grown in T-75 flasks to a density of approximately 5 × 105 cells. P. aeruginosa was added to the Caco-2 cells at a final concentration of 107 cfu/mL in the presence or absence of 3% GalNAc, 3% dextran, or 5% mannose (w/v). The bacteria and Caco-2 cells were incubated at 4°C for 45 minutes by end-over-end rotation and then centrifuged at 900 g for 10 minutes. 17 Bacterial adherence was expressed as the percentage of adherence of bacteria using the final colony count to the initial inoculum ratio.

Electrical Resistance Measurements and Mannitol Flux Studies

TEER was measured as previously described. 5 Transepithelial flux studies with the radioactive tracer [3H] mannitol were used to determine whether the permeability defect was paracellular in selected experiments, according to previously published methods. 18 Briefly, cell monolayers in fibrillar collagen transwells were incubated under the experimental conditions in medium containing 5 mmol/L unlabeled mannitol in both the apical and basolateral reservoirs, and 1 μC/μL [3H] mannitol (NEN, Boston, MA) was added to the apical surface after 4 hours of incubation with P. aeruginosa or PA-I in the absence or presence of GalNAc or dextran (3% w/v, 50 μL as in previous experiments). Unidirectional, mucosal-to-serosal, mannitol flux was measured during a 30-minute period.

Western Blot Analysis of Tight Junction Proteins

Levels of tight junctional proteins ZO-1, ZO-2, and occludin were analyzed by Western blotting using specific antibodies, according to previously published protocols. 19 Heat shock protein (HSP-72) was used as a positive control.

Mouse Studies

Mouse Strains/Housing

All experiments were approved by the Animal Care and Use Committee at the University of Chicago. Inbred mice of the Balb/c background weighing 20 to 25 g were used for all experiments. Mice were housed in individual wire-bottom cages to limit coprophagia during the entire experimental period.

Surgical Hepatectomy/Cecal Injection Method

For lethality experiments, mice were assigned to two surgical groups of either laparotomy only followed by direct cecal injection of P. aeruginosa or its components (control) or laparotomy + 30% surgical hepatectomy followed by direct cecal injection of P. aeruginosa or its components. Animals were allowed water ad libitum only for the remainder of the study period. Control animals were anesthetized (ketamine 100 mg/kg, xylazine 10 mg/kg, and atropine 0.04 mg/kg given intraperitoneally) and underwent a midline laparotomy incision and injection directly into the base of the cecum with 200 μL P. aeruginosa or its components at the following concentrations: live P. aeruginosa : 1 × 104 to 5 × 107 cfu/mL; PA-I: 1.33 mg/mL (undiluted 1:1 or diluted in PBS at 1:3 and 1:10); and exotoxin A: 0.33 mg/mL. After the injection and before removal of the needle, a fine silk suture was placed around the injection site and needle. As the needle was extracted, the suture material was tied, thereby sealing the puncture site. The puncture site was then swabbed with 70% isopropyl alcohol. This technique resulted in no leaks or evidence of local peritonitis after surgery. The hepatectomy was performed using an electrocautery device, excising the floppy left lobe of the liver, which resulted in minimal bleeding. Specimens were weighed and the ratio of excised liver to body weight was calculated to confirm uniformity of the resection.

To determine the systemic effect of live P. aeruginosa, PA-I, or exotoxin A on mouse death, 200 μL live P. aeruginosa (2 × 106 cfu/mL), PA-I (1.33 mg/mL), or exotoxin A (0.33 mg/mL) was injected intraperitoneally into 10 control mice and 10 mice undergoing 30% surgical hepatectomy. Mice were observed for 48 hours and death was recorded. All mice were killed at 48 hours, and the peritoneum was examined for evidence of local peritonitis. In selected animals, tissues (cecum, liver, blood) were harvested for bacterial culture and histologic examination.

Death Prevention Studies

Based on the dose-response experiments performed, animals undergoing hepatectomy and starvation were assigned to six groups of 10 animals each. Mice received 200 μL live P. aeruginosa at 2 × 106 cfu/mL or 200 μL of the combination of PA-I (1.33 mg/mL, diluted 1:2 in PBS) and exotoxin A (0.33 mg/mL) by the direct cecal injection method. Two additional groups underwent hepatectomy and had the identical doses of P. aeruginosa and PA-I and exotoxin A suspended in 13% GalNAc or 15% dextran for 20 minutes before cecal injection. To provide proliferating bacteria with a continued source of GalNAc and dextran, animals had 1 mL of each solution injected directly into the terminal ileum, which filled the small intestine proximally. To control for a possible dilutional effect alone, animals in the groups not receiving GalNAc or dextran had 0.9% NaCl similarly injected into the ileum. Animals were followed up for 48 hours and death was recorded. When the animals were killed, cultures were performed in selected cases.

Statistical Analysis

Data were loaded into the SigmaStat (Jandel Corp., San Rafael, CA) program and tested for significance using a one-way analysis of variance and Neuman-Keuls post hoc testing where appropriate. For nonparametric data involving the percentage (incidence) of death, the Fisher exact test was used. P < .05 was accepted for statistical significance.



aeruginosa and PA-I Alter the TEER of Caco-2 Monolayers in a Dose- and Time-Dependent Manner

Figure 1 demonstrates the effects of P. aeruginosa or its components on the TEER of Caco-2 monolayers measured at 2 and 4 hours after inoculation (triplicate cultures, n = 7 for each group). PA-I had an effect on TEER equal to the whole bacteria. No increase in lactate dehydrogenase release was observed in monolayers from the groups measured at 4 hours, suggesting that the monolayers were intact (data not shown).

figure 19FF1
Figure 1. Percentage change in transepithelial resistance of Caco-2 epithelial monolayers exposed to Pseudomonas aeruginosa or its components at 2 and 4 hours after exposure. Data are means ± standard error of the mean of triplicate cultures (n ...


aeruginosa Adherence to Caco-2 Cells and Its Effect on Barrier Function Are Inhibited by Dextran and GalNAc

Figure 2 demonstrates that 1% to 2% of the initial inoculating dose of P. aeruginosa adhered to Caco-2 monolayers (n = 7 per group; duplicate culture). This is in accordance with similar studies using enteric bacteria and human epithelial cells. 20 When bacteria were pretreated with mannose, an oligosaccharide that binds to the PA-II lectin of P. aeruginosa, a minimal decrease in adherence to Caco-2 cells was seen (data not shown). The effects of P. aeruginosa- and PA-I-induced responses in Caco-2 cells were studied in the absence and presence of GalNAc and dextran and are also shown in Figure 2 (n = 7 for each group; triplicate culture). Both GalNAc and dextran attenuated the decrease in TEER induced by live P. aeruginosa on Caco-2 monolayers (P < .001). GalNAc had a similar inhibitory effect on PA-I-induced decrease in TEER (P < .001). Dextran, however, had no effect on PA-I-induced decreases in TEER in this system.

figure 19FF2
Figure 2. (Small graph) Percentage adherence of Pseudomonas aeruginosa (107 cfu/mL) to dispersed Caco-2 epithelial monolayers after 45 minutes of incubation at 4°C. Data are mean ± standard error of the mean of duplicate cultures (n = ...

Results of the mannitol flux analysis are shown in Figure 2 (n = 7 for each group of triplicate culture). Both P. aeruginosa and PA-I caused a significant increase in cumulative mannitol flux across Caco-2 monolayers (P < .001). Both GalNAc and dextran completely prevented the increased mannitol flux induced by either live P. aeruginosa or PA-I after 4 hours of incubation (P < .01).


aeruginosa Alters the Tight Junction Proteins ZO-1 and Occludin via Its PA-I Lectin

Results of the Western blot analysis of the tight junctional proteins ZO-1, ZO-2, and occludin are shown in Figure 3.

figure 19FF3
Figure 3. Western blot analysis of the effect of live Pseudomonas aeruginosa or PA-I on the tight junctional proteins ZO-1, ZO-2, and occludin. Live P. aeruginosa was pretreated with 2% GalNAc. Dose and time responses of occludin after exposure ...

Results of Mouse Studies

Live Pseudomonas or the combination of PA-I and exotoxin A injected into the cecum caused lethal sepsis in mice after hepatectomy. As shown in Figure 4, mice subjected to 30% hepatectomy and receiving a direct injection into the cecum of either live P. aeruginosa or the combination of PA-I and exotoxin A had a significant death rate.

figure 19FF4
Figure 4. Death rate of mice subjected to sham laparotomy or 30% surgical hepatectomy and receiving direct injection into the cecum of live cultures of Pseudomonas aeruginosa, saline, or virulence components. Death is defined as the percentage ...

Systemic administration of exotoxin A was lethal, but PA-I had no effect. Table 1 demonstrates that P. aeruginosa adheres to the cecum of mice undergoing hepatectomy to a greater extent (P < .01) than to the cecum of control mice. Dissemination of P. aeruginosa to the liver and blood was not quantitatively different between control and hepatectomy mice. The administration of exotoxin A to mice by the intraperitoneal route resulted in 100% death at 48 hours, whereas PA-I had no effect, independent of hepatectomy. These findings suggest the possibility that exotoxin A absorption from the cecum may be necessary to induce death in this model, whereas the effect of PA-I absorption is negligible. In marked contradistinction to intracecal injection, intraperitoneal injection of P. aeruginosa resulted in no deaths in control or hepatectomy mice. Both groups of mice developed bacteremia with P. aeruginosa after intraperitoneal injection (data not shown).

Table thumbnail

Death prevention experiments are summarized in Figure 5. GalNAc completely prevented death in mice after hepatectomy and cecal injection of live P. aeruginosa. Death was also prevented with GalNAc pretreatment in mice who received the combination of PA-I and exotoxin A. Dextran only partially attenuated death in this model, and it had no effect on death prevention in mice injected with both PA-I and exotoxin A.

figure 19FF5
Figure 5. Death rate of mice after 30% hepatectomy and injection of either live P. aeruginosa or the combination of PA-I and exotoxin A in the presence and absence of 13% GalNAc or 15% dextran. Results demonstrate that although ...


P. aeruginosa is a feared hospital pathogen because of its high antibiotic resistance capacity and its lethal virulence characteristics. This pathogen tends to predominate on the mucosal surfaces of the most immunologically compromised hospital patients and is associated with a high death rate. 21,22 Among patients undergoing chemotherapy who subsequently become bacteremic with P. aeruginosa, 81% were fecal carriers of the same strain, 23 suggesting that in some instances, the intestinal tract may be a reservoir for this pathogen. Little information is available, however, on the effects of this organism on the intestinal mucosa, because it is not considered to be an intestinal pathogen.

One particularly well-studied effect of pathogenic bacteria on the intestinal mucosa is their ability to alter the tight junction permeability or barrier function of the lining mucosal epithelial cells. Bacteria-induced alterations in epithelial permeability have been demonstrated for enteroinvasive strains of E. coli and Salmonella typhimurium. 24 Recent data from these in vitro studies suggest that bacteria-induced alterations in tight junction permeability occur by means of contact-dependent signal transduction events. 25 However, the specific virulence determinants on bacteria responsible for the initial contact with epithelia are unknown. Many soluble bacterial toxins such as lipopolysaccharide (endotoxin) from various Enterobactericiae species do not alter intestinal epithelial permeability when applied locally, nor do they have other pathologic effects, such as the release of interleukin-8 or tumor necrosis factor α. 26,27 Because gram-negative bacteria constantly shed their outer membrane fragments during growth, endotoxins (lipopolysaccharide) are in abundant supply in the intestinal lumen. 28 It has been proposed, but not proved, that during critical illness, intestinally derived toxins from colonizing but otherwise apathogenic bacteria (to the intestinal mucosa) can cause systemic sepsis through histologicaly intact intestine and in the absence of an intestinal inflammatory response. 29

Results of the mice studies in this report suggest that the PA-I lectin of P. aeruginosa may be a key component necessary for lethal infection originating from the intestines of a susceptible host. A susceptible host is defined in the present study as a 30% hepatectomy—a mild, well-tolerated surgical procedure that results in high survival rates when no infectious challenge is administered. Data from the present study would suggest that live cells of P. aeruginosa, or the combination of PA-I and exotoxin A, cause an increased death rate in mice undergoing hepatectomy as a result of the immunocompromise of the host, because equal inocula of bacteria or bacterial components resulted in low death rates in control mice. The observation that systemic injection of P. aeruginosa resulted in no deaths in either group of mice, coupled with the observation of equal quantitative dissemination of bacteria to the liver and blood of both groups, despite different death rates, suggests an alternative explanation. The observation that intestinal introduction of P. aeruginosa carries a higher death rate than systemic administration was previously reported by Schook et al. 30 Given that bacteria can express or repress virulence determinants depending on their environment and tissue site, it is possible that the intestinal tract of mice undergoing hepatectomy and starvation upregulates the virulence potential of these bacteria. Subsequent work from our laboratory with this mouse model has demonstrated that hepatectomy and starvation results in significant alterations in the physical microenvironment of the cecum, with alterations in luminal pH and redox potential and elevated luminal norepinephrine concentration. 31 All three of these parameters have been demonstrated independently to induce virulence expression in bacteria in culture. 32 We have further demonstrated in this model that stock strains of P. aeruginosa express more PA-I and virulence to cultured intestinal epithelial cells when grown in a medium rich in norepinephrine (Alverdy JC, unpublished observations). Therefore, firm experimental evidence is accumulating that suggests that the enhanced death rate after hepatectomy and cecal injection of P. aeruginosa is a result of bacterial virulence phenotype transformation in response to luminal norepinephrine. This hypothesis fits nicely with current paradigms of bacterial virulence transformation in response to environmental cues, as proposed by Mekalanos. 33

It seems likely that the observed effects of P. aeruginosa on mouse death were in part mediated by both PA-I and exotoxin A, because the latter combination effectively resulted in similar death rates compared with whole bacterial cells. Further, the finding that only GalNAc completely prevented death in this model suggests a major role for PA-I in gut-induced sepsis from P. aeruginosa. The lack of effectiveness of dextran in this model may be related to the amount of freely secreted PA-I and exotoxin A that is released in the cecum during infection. PA-I lectin is reported to be present in P. aeruginosa culture medium at a concentration greater than 20 μg/mL for a bacterial inoculum of 109 bacteria per milliliter. PA-I lectin has also been detected at higher concentrations in infected sputum of cystic fibrosis patients. 34 The dose of PA-I injected intraluminally that would approximate the concentration presented to its receptors on epithelial cells from adherent P. aeruginosa is difficult to estimate, and therefore the physiologic relevance of the doses of PA-I chosen in the present study will need to be clarified.

Based on the results of the Caco-2 experiments, we would speculate that PA-I induces a permeability defect of sufficient magnitude to permit the permeation of exotoxin A systemically in mice. The finding that PA-I injected systemically (intraperitoneally) into mice does not induce death or any appearance of illness, whereas exotoxin A does, suggests that the mechanism of lethality in this model is by a PA-I-mediated permeability defect to exotoxin A. The precise mechanism by which PA-I and exotoxin A injection into the cecum of mice after hepatectomy causes death remains to be clarified.

Data from the Caco-2 experiments suggest that live or formalin-killed P. aeruginosa induces a defect in epithelial permeability at the level of the intercellular tight junctions, based on the transepithelial flux of mannitol, a paracellular permeability probe. We have previously demonstrated that live P. aeruginosa alters the tight junction regulatory protein ZO-1 in T-84 human colon epithelia. 35 Western blot analysis of ZO-1 and occludin in epithelial cells exposed to live P. aeruginosa with and without GalNAc suggests that this effect in part involves a disruption in tight junctional proteins. Data from the present study also demonstrate a rapid (<4 hour) decrease in TEER when Caco-2 cells were exposed to either whole cells of P. aeruginosa or PA-I. Previous time-response studies examining bacterial-induced alterations in barrier function report a time requirement of at least 8 to 10 hours of incubation to decrease TEER by 30%, reaching a maximal decrease in TEER at 24 hours of 50%. 36 Defined in terms of rapidity and severity of the permeability defect, P. aeruginosa is among the most pathogenic organisms to cultured intestinal epithelia reported to date.

Dextran and GalNAc were chosen for inhibition experiments because dextran has been previously shown to block P. aeruginosa adherence to epithelial cells by an unknown mechanism, and GalNAc is known to bind specifically to PA-I. 15,16 Both dextran and GalNAc inhibited the adherence of P. aeruginosa to Caco-2 cells to a similar degree of less than 105 cfu/mL. This degree of adherence was below the threshold concentration necessary for changes in permeability to occur based on dose-response experiments. This is consistent with the reported function of PA-I as a mechanism by which P. aeruginosa adheres to intestinal epithelial cells (Alverdy JC, unpublished observations). That PA-I is responsible for P. aeruginosa-induced alterations in TEER and mannitol flux is inferred from the inhibition studies with GalNAc. Because PA-I alone can induce changes in TEER and mannitol flux equal to whole bacteria, this, coupled with GalNAc inhibition studies, suggests a major role for PA-I in this response. Data from these experiments also demonstrated that although both GalNAc and dextran attenuate P. aeruginosa-induced decreases in TEER, only GalNAc inhibited PA-I-induced decreased TEER; dextran had no effect. Increased mannitol flux in Caco-2 cells exposed to either live P. aeruginosa or PA-I, in contrast, was completely prevented with GalNAc or dextran treatment. TEER to passive ion flow is an extremely sensitive measure of barrier function, whereas mannitol permeation across Caco-2 cells must conform to the Stokes radius and possibly the charge of the paracellular pathway. It is not known how dextran inhibits the adherence of P. aeruginosa to epithelial cells. Labeling studies have demonstrated that dextran neither adheres to the epithelial cells nor the bacteria. 16 Dextran, however, is known to change the surface hydrophobicity and charge of epithelial cells and may result in steric hindrance or electrostatic shielding of the cell glycocalyx. 37 Dextran has been shown to block mannitol permeability completely in cultured epithelial cells exposed to elastase, and it prevented intestinal barrier dysfunction after experimental hepatectomy in rats. 38,39 The fact that tight junctions are known to discriminate both by charge and size could explain the finding that dextran prevented PA-I-induced alterations in mannitol flux without affecting TEER.

P. aeruginosa may fulfill the requirements for inducing a state of gut-derived sepsis with or without the translocation of the whole bacterial cells. Because of its known properties of immunoevasion and biofilm production, this organism may be potentially harmful if specific phenotypes are expressed in the mammalian intestine in a susceptible host. The PA-I lectin of P. aeruginosa is of particular interest because it can switch to the “on/off” variation or to the “in/out” position on the outer surface of the cell, depending on the environmental cues it receives. 40 These adaptive features of the lectins of P. aeruginosa help regulate bacterial adaptability, whereby access to host cells for nutritional purposes, self-protection, or attack is possible. In the context of the present set of mouse experiments, it is possible that the cecum is used as an immunologically privileged site for P. aeruginosa, whereby it induces a permeability defect (by PA-I) to its cytotoxins while remaining inaccessible to the host defense tactics. Once bloodborne, however, PA-I expression may be downregulated, because this phenotype represents a liability to the organism: it is also a ligand for neutrophil attachment. Because the expression of PA-I is coregulated with exotoxin A and is located together with other cytotoxic exoproducts on the P. aeruginosa virulence island gene, bloodborne organisms have downregulated virulence and by themselves do not induce death. 41–43

In summary, we established that the PA-I lectin of P. aeruginosa is important for adherence to intestinal epithelial cells and can significantly perturb their barrier function. We speculate that in the intestinal tract, the mere presence of exotoxin A-producing strains of P. aeruginosa expressing the PA-I lectin has the potential to be of pathogenic importance in the critically ill.


The authors thanks Dr. N.C. Garber for his technical advice and assistance with the PA-I assay.


Correspondence: John C. Alverdy, MD, FACS, Dept. of Surgery, University of Chicago, 5841 S. Maryland, MC 6090, Chicago, IL 60637.

Supported in part by NIH grants DK-38510 (E.C.) and DK-47722 (E.C.), and Digestive Disease Center Grant DK-42086.

E-mail: jalverdy@surgery.bsd.uchicago.edu

Accepted for publication November 15, 1999.


1. Griffith SJ, Nathan C, Selander RK. The epidemiology of Pseudomonas aeruginosa in oncology patients in a general hospital. J Infect Dis 1989; 160:1030–1036. [PubMed]
2. Buck AC, Cooke EM. The fate of ingested Pseudomonas aeruginosa in normal persons. J Med Microbiol 1969; 2:521–525. [PubMed]
3. Marshall JC, Christou NV, Meakins JL. The gastrointestinal tract—the undrained abscess of multiple organ failure. Ann Surg 1993; 218:111–119. [PMC free article] [PubMed]
4. Furuya N, Hirakata Y, Tomono K, et al. Mortality rates amongst mice with endogenous septicaemia caused by Pseudomonas aeruginosa isolates from various clinical sources. J Med Microbiol 1993; 39:141–146. [PubMed]
5. Spitz J, Yuhan R, Koutsouris A, et al. Enteropathogenic Escherichia coli adherence to intestinal epithelial monolayers diminishes barrier function. Am J Physiol 1995; 268 (2 Pt 1):G374–379. [PubMed]
6. Spitz JC, Ghandi S, Taveras M, et al. Characteristics of the intestinal epithelial barrier during dietary manipulation and glucocorticoid stress. Crit Care Med 1996; 24:635–641. [PubMed]
7. Gilboa-Garber N. Pseudomonas aeruginosa lectins as a model for lectin production, properties, applications and functions. Zbl Bakt Hyg 1988; 270:3–15. [PubMed]
8. Hendrickson BA, Guo J, Laughlin RJ, Alverdy JC. Increased type 1 fimbrial expression among commensal Escherichia coli in the murine cecum following catabolic stress. Infect Immun 1998; 67:745–753. [PMC free article] [PubMed]
9. Alverdy JC, Hendrickson B, Guandalini SS, et al. Perturbed bioelectrical properties of the mouse cecum following hepatectomy and starvation: the role of bacterial adherence. Shock 1999; 12:235–241. [PubMed]
10. Peterson MD, Mooseker MS. Characterization of the enterocyte-like brush border cytoskeleton of the C2bbe clones of the human intestinal cell line, Caco-2. J Cell Sci 1992; 102:581–600. [PubMed]
11. Tamai H, Gaginella S, Kachur JF, et al. Ca+-mediated stimulation of Cl secretion by reactive oxygen metabolites in human colonic T84 cells. J Clin Invest 1992; 89:301–307. [PMC free article] [PubMed]
12. Khan AA, Cerniglia CE. Detection of Pseudomonas aeruginosa from clinical and environmental samples by amplification of the exotoxin A gene using PCR. Appl Environ Micro 1994; 60:3739–3745. [PMC free article] [PubMed]
13. O’Callaghan RJ, Engel LS, Hobden JA, et al. Pseudomonas keratitis. The role of an uncharacterized exoprotein, protease IV, in corneal virulence. Invest Ophthal Vis Sci 1996; 37:534–543. [PubMed]
14. Gilboa-Garber N. Pseudomonas aeruginosa lectins. Meth Enzymol 1982; 83:378–385. [PubMed]
15. Gilboa-Garber N. Lectins of Pseudomonas aeruginosa : properties, biologic effects and applications. In: Mirelman D, ed. Microbial Lectins and Agglutinins: Properties and Biologic Activity. New York: John Wiley and Sons; 1986: 255–269.
16. Barghouthi S, Guerdoud LM, Speert DP. Inhibition by dextran of Pseudomonas aeruginosa adherence to epithelial cells. Am J Respir Crit Care Med 1996; 154:1788–1793. [PubMed]
17. Letlet C, Hobden JA, Masinick SA, Hazlett LD. Environmental factors influence P. aeruginosa binding to the wounded mouse cornea. Curr Eye Res 1998; 17:231–237. [PubMed]
18. Madara JL, Dharmsathaphorn K. Occluding junction structure–function relationships in a cultured epithelial monolayer. J Cell Biol 1985; 101:2124–2133. [PMC free article] [PubMed]
19. Musch MW, Hayden D, Sugi K, et al. Cell-specific induction of hsp72-mediated protection by glutamine against oxidant injury in IEC18 cells. Proc Assoc Am Physicians 1998; 110 (2):136–139. [PubMed]
20. Gewirtz AT, McCormick B, Neish AS, et al. Pathogen-induced chemokine secretion from model intestinal epithelium is inhibited by lopoxin A4 analogs. J Clin Invest 1998; 101:1860–1869. [PMC free article] [PubMed]
21. Rello J, Rue M, Jubert P, et al. Survival in patients with nosocomial pneumonia: impact of the severity of illness and the etiologic agent. Crit Care Med 1997; 25:1862–1867. [PubMed]
22. Roberts FJ, Geere IW, Coldman A. A three-year study of positive blood cultures with emphasis on prognosis. Rev Infect Dis 1991; 13:34–46. [PubMed]
23. Tancrede CH, Andremont AO. Bacterial translocation and gram-negative bacteremia in patients with hematological malignancies. J Infect Dis 1985; 152:99–103. [PubMed]
24. Finlay BB, Falcow S. Salmonella interactions with polarized human Caco-2 epithelial cells. J Infect Dis 1990; 162:1096–1106. [PubMed]
25. Bliska JB, Galan JE, Falcow S. Signal transduction in the mammalian cell during bacterial attachment and entry. Cell 1993; 73:903–920. [PubMed]
26. Benoit R, Rowe S, Watkins SC, et al. Pure endotoxin does not pass across the intestinal epithelium in vitro. Shock 1998; 10:43–47. [PubMed]
27. Jung HC, Eckmann L, Yang SK, et al. A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion. J Clin Invest 1995; 95:55–65. [PMC free article] [PubMed]
28. Deventer SJ, Cate JW, Tytgat GN. Intestinal endotoxemia. Gastroenterology 1988; 94:825–831. [PubMed]
29. Deitch EA. The role of intestinal barrier failure and bacterial translocation in the development of systemic infection and multiple organ failure. Arch Surg 1990; 125:403–404. [PubMed]
30. Schook LB, Carrick L, Berk RS. Murine gastrointestinal tract as a portal of entry in experimental Pseudomonas aeruginosa infections. Infect Immunol 1976; 14:564–570. [PMC free article] [PubMed]
31. Laughlin RB, Rocha FM, Christian C, et al. The role of norepinephrine on bacterial virulence transformation in the intestine following stress. Surg Forum 1999; 50:248–250.
32. Guiney DG. Regulation of bacterial virulence gene expression by the host environment. J Clin Invest 1997; 99:565–569. [PMC free article] [PubMed]
33. Mekalanos JJ. Minireview: environmental signals controlling expression of virulence determinants in bacteria. J Bacteriol 1992; 174:1–7. [PMC free article] [PubMed]
34. Bajolet-Laudinat O, Girod-de Bentzmann S, Tournier JM, et al. Cytotoxicity of Pseudomonas aeruginosa internal lectin PA-I to respiratory epithelial cells in primary culture. Infect Immun 1994; 62:4481–4487. [PMC free article] [PubMed]
35. Rocha F, Stern E, Laughlin RJ, et al. Pseudomonas aeruginosa severely perturb the permselectivity of cultured human colon epithelia. Surg Forum 1998; 24:32–33.
36. Philpott DJ, McKay DM, Sherman PM, Perdue MH. Infection of T84 cells with enteropathogenic Escherichia coli alters barrier and transport functions. Am J Physiol 1996; 270:G634–645. [PubMed]
37. Baldwin AL, Wu NZ, Stein DL. Endothelial surface charge of intestinal mucosal capillaries and its modulation by dextran. Microvasc Res 1991; 42:160–178. [PubMed]
38. Peterson MW, Walter ME, Nygaard SD. Effect of neutrophil mediators on epithelial permeability. Am J Resp Cell Mol Biol 1995; 13:719–727. [PubMed]
39. Wang XO, Sun ZW, Soltesz V, et al. The role of intravenous administration of dextran 70 in enteric bacterial translocation after partial hepatectomy. Eur J Clin Invest 1997; 27:936–942. [PubMed]
40. Glick J, Garber NC. The intracellular localization of P. aeruginosa lectins. J Gen Microbiol 1983; 129:3085–3090. [PubMed]
41. Gilboa-Garber N, Avichezer D, Garber NC. Bacterial lectins. Properties, structure, effects, function, and applications. In: Gabius H-J, Gabius S, eds. Glycosciences: Status and Perspectives. Weinheim, Germany: Chapman and Hall; 1997: 369–398.
42. Garber NC, Hammer-Muntz O, Belz A. The lux autoinducer stimulates the production of the lectins of Pseudomonas aeruginosa. ISM Lett 1995; 15:164.
43. Pearson JP, Gray KM, Passador L. Structure of autoinducer required for expression of Pseudomonas aeruginosa virulence genes. Proc Natl Acad Sci USA 1994; 91:197–201. [PMC free article] [PubMed]

Articles from Annals of Surgery are provided here courtesy of Lippincott, Williams, and Wilkins
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...