IAI FigSearch
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
IAI.01098-07v1
76/3/1076    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fujii, Y.
Right arrow Articles by Okamoto, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fujii, Y.
Right arrow Articles by Okamoto, K.

 Previous Article  |  Next Article 

Infection and Immunity, March 2008, p. 1076-1082, Vol. 76, No. 3
0019-9567/08/$08.00+0     doi:10.1128/IAI.01098-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Fluid Secretion Caused by Aerolysin-Like Hemolysin of Aeromonas sobria in the Intestines Is Due to Stimulation of Production of Prostaglandin E2 via Cyclooxygenase 2 by Intestinal Cells{triangledown}

Yoshio Fujii,1* Ken Tsurumi,1 Masaaki Sato,1 Eizo Takahashi,2 and Keinosuke Okamoto2

Institute of Pharmacognosy, Faculty of Pharmaceutical Sciences, Tokushima Bunri University, Yamashiro, Tokushima, Tokushima 770-8514,1 Faculty of Pharmaceutical Sciences, Okayama University, Tsushima-naka, Okayama, Okayama 700-8530, Japan2

Received 8 August 2007/ Returned for modification 12 September 2007/ Accepted 4 December 2007


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
To clarify the mechanisms of diarrheal disease induced by Aeromonas sobria, we examined whether prostaglandin E2 (PGE2) was involved in the intestinal secretory action of A. sobria hemolysin by use of a mouse intestinal loop model. The amount of PGE2 in jejunal fluid and the fluid accumulation ratio were directly related to the dose of hemolysin. The increase over time in the level of PGE2 was similar to that of the accumulated fluid. In addition, hemolysin-induced fluid secretion and PGE2 synthesis were inhibited by the selective cyclooxygenase 2 (COX-2) inhibitor NS-398 but not the COX-1 inhibitor SC-560. Western blot analysis revealed that hemolysin increased the COX-2 protein levels but reduced the COX-1 protein levels in mouse intestinal mucosa in vivo. These results suggest that PGE2 functions as an important mediator of diarrhea caused by hemolysin and that PGE2 is produced primarily through a COX-2-dependent mechanism. Subsequently, we examined the relationship between PGE2, cyclic AMP (cAMP), and cystic fibrosis transmembrane conductance regulator (CFTR) Cl– channels in mouse intestinal mucosa exposed to hemolysin. Hemolysin increased the levels of cAMP in the intestinal mucosa. NS-398 inhibited the increase in cAMP production, but SC-560 did not. In addition, H-89, a cAMP-dependent protein kinase A (PKA) inhibitor, and glibenclamide, a CFTR inhibitor, inhibited fluid accumulation. Taken together, these results indicate that hemolysin activates PGE2 production via COX-2 and that PGE2 stimulates cAMP production. cAMP then activates PKA, which in turn stimulates CFTR Cl– channels and finally leads to fluid accumulation in the intestines.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Aeromonas hydrophila and A. sobria have been recognized as pathogens associated with acute gastroenteritis in both adults and children (5, 9). We purified and characterized the enterotoxin from the culture supernatant of A. sobria isolated from a patient with diarrhea and demonstrated that it possessed hemolytic activity in addition to enterotoxic activity (14, 26).

Analysis of the nucleotide sequence showed that the hemolysin is homologous with aerolysin (A. hydrophila hemolysin). The overall homology in amino acid sequence between the hemolysin and aerolysin was 68.5% identity (14). The two toxins have similar modes of action. Both act to form small pores in the cell membrane to which they have bound and to generate the osmotic gradient that develops as a result of cellular injury (6, 14, 41).

Both hemolysin and aerolysin possess enterotoxic activity (1, 14, 29). In a previous paper, we reported that hemolysin stimulates the production of cyclic AMP (cAMP) in T84 cells (human colon carcinoma cell line) and that the cAMP thus produced emerges into the extracellular space (15). Furthermore, we demonstrated that hemolysin activates a cAMP-dependent Cl– secretory pathway, which is presumably related to cystic fibrosis transmembrane conductance regulator (CFTR) in Caco-2 cells (a human colonic epithelial cell line) (39). From these results, we speculated that the activation of CFTR by cAMP was involved in the diarrhea caused by the hemolysin.

Chopra et al. reported that the aerolysin-related cytotoxic enterotoxin (Act) of A. hydrophila increases the production of prostaglandin E2 (PGE2) and cAMP in murine macrophage cells. Celebrex, a selective cyclooxygenase 2 (COX-2) inhibitor, significantly inhibits Act-induced PGE2 and cAMP production (8). In addition, the production of PGE2 by Act in macrophages was confirmed by Ribardo et al. (32). From these data, we thought that PGE2 might be elicited in the intestines by A. sobria hemolysin. We examined the involvement of PGE2 by use of a mouse intestinal loop assay in this study.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Hemolysin and mouse intestinal loop assay. The hemolysin was purified from a culture supernatant of A. sobria strain 357 by successive column chromatographies as described previously (14). The purified hemolysin gave a single band by sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

A mouse intestinal loop assay was carried out as described previously (14). All experiments were approved by the Institute Animal Care and Use Committee, Tokushima Bunri University.

Male ddY mice weighing 30 to 35 g were used. They were housed in plastic cages under controlled environmental conditions. The activity of the sample was assessed by the fluid accumulation ratio (weight of the intestinal loop [in grams]/length of the loop [in centimeters]). A ratio of over 0.20 was regarded as a positive response. Minimum doses resulting in a positive fluid accumulation ratio of the hemolysin and cholera toxin (CT) were 100 ng and 500 ng, respectively.

PGE2 measurement. The amounts of PGE2 released into mouse jejunal lumens after exposure to hemolysin were determined. Mouse jejunal loops were incubated for various time periods with hemolysin. The fluid accumulated in the loop was transferred to a microcentrifuge tube and then centrifuged at 15,000 x g for 1 min at 4°C. The content of PGE2 in the sample was determined as described previously (20). A commercially available radioimmunoassay kit was used (PGE2 [125I] Biotrak assay system; Amersham Biosciences, Little Chalfont, United Kingdom). The amount of PGE2 in the intestinal fluid was expressed as picograms of PGE2 per gram wet weight of the solid intestine.

Measurement of cAMP accumulation. The effect of hemolysin on the level of cAMP in the mouse jejunum was examined. Mouse jejunal loops were removed after an appropriate incubation period. The loops were immediately cut open lengthwise, and the mucosa was scraped by drawing a glass microscope slide over it. The mucosal sample obtained was immediately placed in 500 µl of 5% trichloroacetic acid and disrupted by sonication. After centrifugation of the sonicates at 15,000 x g for 10 min at 4°C, the pellet was dissolved in 1.0 N NaOH, and the solution was used for measuring the protein concentration by the method of Lowry et al. (22). The supernatant was washed five times with water-saturated diethyl ether to remove the trichloroacetic acid. The aqueous phase was evaporated dry under a vacuum centrifuge. The cAMP concentration in each sample was determined with a commercially available radioimmunoassay kit (cAMP [3H] assay system; Amersham Biosciences). The amount of cAMP was expressed as picomoles per milligram of protein.

Western blot analysis to detect COX-1 and COX-2. Samples for the Western blot analysis were prepared according to the methods of Beubler et al. (2). Rabbit polyclonal antibodies against murine COX-1 (Cayman Chemicals, Ann Arbor, MI) and rabbit polyclonal antibodies against murine COX-2 (Immuno-Biological Laboratories Co., Gunma, Japan) made to a region that was conserved between the mouse and rat were used to detect the bands corresponding to COX-1 and COX-2, respectively. Bound antibody was visualized using a goat anti-rabbit immunoglobulin G linked to horseradish peroxidase and an Immun-star horseradish peroxidase chemiluminescent kit (Nippon Bio-Rad, Tokyo, Japan) with fluorographic detection on X-ray film. The X-ray film was analyzed with the NIH image program (Wayne Rasband, U.S. National Institutes of Health, available from the Internet by anonymous ftp from http://rsb.info.nih.gov/nih-image/download.html). Fluorescence intensity was expressed as a relative percentage of the control value.

Blocking test. To confirm the specific reaction of the antibody to COX-2 with the band on the gel, a blocking test with COX-2 blocking antigenic peptide (Immuno-Biological Laboratories Co.) was performed. The COX-2 antigenic peptide was incubated with the COX-2 antibody at room temperature for 1 h. The concentrations of peptide and antibody in the mixture were 10 µg/ml and 2.5 µg/ml, respectively. The reactivity of the incubated COX-2 antibody was examined as described above.

Statistical analysis. The data were evaluated with Student's t test for unpaired samples, and P values of <0.05 were considered to be significant.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Hemolysin induced the production of PGE2 in the mouse jejunal loop. In order to evaluate the role of PGE2 in hemolysin-induced diarrhea, after hemolysin (250 ng) was injected into the mouse intestinal loop, fluid accumulation and the amount of PGE2 released into the intestinal fluid were determined between the time points of 30 and 180 min (Fig. 1A). Fluid accumulation was observed as early as 30 min after the injection of hemolysin, and the volume of accumulated fluid increased in relation to the incubation time. The amount of PGE2 in the intestinal fluid also increased in relation to the incubation time. The increase over time in the level of PGE2 was similar to that of the accumulated fluid (Fig. 1A). In further studies, the dose responses for fluid accumulation and for the amount of PGE2 in the intestinal fluid were examined (Fig. 1B). As shown in Fig. 1B, the level of PGE2 and the fluid accumulation ratio were directly related to the dose of hemolysin and reached a plateau at 250 ng. These results raise the possibility that PGE2 functions as a mediator of fluid accumulation induced by hemolysin.


Figure 1
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 1. Effect of hemolysin dose on PGE2 production and the time course of PGE2 production in a mouse intestinal loop assay. The time course (A) and dose response (B) of PGE2 release into the intestinal fluid after exposure to hemolysin were examined. (A) Two hundred fifty nanograms of hemolysin was injected into each intestinal loop and incubated for various time periods. Control loops were inoculated with phosphate-buffered saline. The fluid accumulation ratio ({circ}, loop injected with hemolysin; •, control loop) and PGE2 level in accumulated fluid ({square}, loop injected with hemolysin; {blacksquare}, control loop) were determined. (B) Hemolysin was injected into the intestinal loops of mice. After 3 h, the fluid accumulation ratio ({circ}) and PGE2 level in the accumulated fluid ({square}) were determined. Values are expressed as the means ± standard errors of five determinations.

 
Effect of carbenoxolone. PGE2 in cells is degraded into 15-keto-13,14-dihydro-PGE2 by 15-hydroxy-PG-dehydrogenase and {Delta}13-PG-reductase. Carbenoxolone has been shown to inhibit these PG-degrading enzymes, and treatment with carbenoxolone has been shown to increase the quantity of PGE2 in tissue (28). Therefore, we assumed that carbenoxolone would augment the fluid accumulation induced by hemolysin, because the degradation of PGE2 in the jejunal epithelial cells is repressed by carbenoxolone treatment. Carbenoxolone (180 µg) was injected into the intestinal loops of mice together with hemolysin (50 ng or 70 ng). The volume of fluid accumulated increased significantly in comparison with that in the absence of carbenoxolone. Carbenoxolone alone had no effect on secretion (Fig. 2).


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 2. Effect of carbenoxolone on intestinal fluid in response to hemolysin. Indicated sample solutions were injected into the intestinal loops of mice. The amount of fluid accumulated in each loop was determined 3 h after the injection of hemolysin. Values are expressed as the means ± standard errors of eight determinations.

 
Effect of COX inhibitors on fluid accumulation and production of PGE2. The above results support the possibility that PGE2 functions as an important mediator of diarrhea induced by hemolysin. Two isoforms of COX, COX-1 and COX-2, are involved in the synthesis of the relevant PGs. To elucidate the role of each isoform in hemolysin-induced fluid accumulation, the effects of inhibitors of COX (a selective inhibitor of COX-1 [SC-560; Cayman Chemicals], a selective inhibitor of COX-2 [NS-398; Cayman Chemicals], and a nonselective inhibitor [indomethacin; Sigma Chemical Co., St. Louis, MO]) on the production of PGE2 was examined. Each inhibitor was injected subcutaneously into mice. The amounts of inhibitor injected are shown in Fig. 3. Forty minutes later, an intestinal loop was created and 0.2 ml of a solution containing 125 ng of hemolysin was injected into the loop. The fluid accumulation was determined 2.5 h after the injection.


Figure 3
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 3. Effect of COX inhibitors on the intestinal fluid accumulation (A) and on luminal PGE2 release (B) in response to hemolysin (Hemo). (A) The indicated amounts of COX inhibitors were administered subcutaneously. Forty minutes later, 125 ng of hemolysin was injected into the loop. The amount of fluid accumulated was determined 2.5 h after the injection. (B) A COX inhibitor (10 mg/kg) or vehicle solution (dimethyl sulfoxide-propylenglycol solution [1:9 {vol/vol}]) was administered subcutaneously. Forty minutes later, 125 ng of hemolysin was injected into the loop. The PGE2 concentration in the fluid accumulated was determined 2.5 h after the injection. All values are expressed as the means ± standard errors of eight determinations. Significance (versus hemolysin alone). *, P < 0.02.

 
NS-398, a selective COX-2 inhibitor, and indomethacin, a nonselective inhibitor, dose-dependently inhibited the accumulation of fluid induced by hemolysin (Fig. 3A). The inhibition clearly appeared at the dose of 5 mg/kg of body weight of two compounds, indomethacin and NS-398. In contrast, SC-560, a selective COX-1 inhibitor, did not inhibit the hemolysin-induced accumulation even at a dose of 20 mg/kg. These results suggested that PGE2s generated by the actions of COX-2 play an essential role in the accumulation of fluid caused by hemolysin.

To ascertain the involvement of PGE2 in the fluid secretion due to hemolysin, the amount of PGE2 released into the intestinal lumen was determined. The fluids accumulated in the intestinal loop were used to measure the amount of PGE2 released into the intestinal lumen. Among the intestinal fluids which were obtained in the experiment shown in Fig. 3A, we selected the samples prepared with the injection of COX inhibitors at a dose of 10 mg/kg. As a control, fluids from mice not administered any inhibitors were used. The amounts of PGE2 in the fluid were determined as described in Materials and Methods. The results are shown in Fig. 3B. The selective COX-2 inhibitor NS-398 significantly reduced the release of PGE2 into the intestinal lumen; however, the selective COX-1 inhibitor SC-560 did not reduce the amount of PGE2 (Fig. 3B).

Effect of hemolysin on COX-2 expression in the mouse intestinal mucosa in vivo. First, the presence of the reactive substances with COX-2 antibody in the mouse jejunal mucosa sample was examined using the Western blotting method. The samples were prepared from untreated mice. The protein band, which was reactive with COX-2 antibody, was detected in the sample, indicating that COX-2 was expressed in the mouse intestine (data not shown). To confirm the specificity of the band, a blocking test was performed using a COX-2 blocking antigenic peptide, as described in Materials and Methods. The antigenic peptide completely blocked binding of the COX-2 antibody to the band of the gel derived from mouse jejunal mucosa. This means that the band is in fact COX-2.

Subsequently, the intestinal mucosa was recovered from intestinal loops of mice administered hemolysin (125 ng). The specimens were taken at 45 min and 150 min postadministration, and the COX-2 in the sample was examined by Western blot analysis. The level of COX-2 antigen slightly but significantly increased at 45 min. The increase became more marked following exposure to hemolysin for 150 min (Fig. 4). In contrast, the expression of COX-1 in the intestinal mucosa was suppressed by hemolysin (Fig. 4).


Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 4. Detection of COX-2 and COX-1 in the mouse intestinal mucosa by Western blot analysis. Hemolysin (125 ng) was injected into the intestinal loops of mice. After 45 min and 150 min, the loops were immediately cut open along their lengths, and the mucosa was scraped by drawing a glass microscope slide over it. The mucosa recovered was immediately placed in 600 µl of the lysis buffer and tissue homogenates were prepared by disrupting the mucosa by repeated vortexing. Western blot analyses to detect COX-1 and COX-2 were performed as described in the text. The fluorescence intensity of the corresponding band was measured. The values obtained from five experiments were leveled and expressed as values relative to those of control. Values are expressed as the means ± standard errors of five determinations. Differences between two values are presented with P values.

 
Effect of hemolysin on cAMP levels in the mouse intestinal mucosa in vivo. Since PGE2 activates intestinal mucosal adenylyl cyclase via a receptor-mediated mechanism (37), we examined the levels of cAMP in mouse intestinal epithelial cells exposed to hemolysin. As shown in Fig. 5A, 30 min after the injection of hemolysin (125 ng), the concentration of cAMP in the mouse intestinal mucosa was significantly increased. The increase became more marked following exposure to hemolysin for 180 min (Fig. 5A). Furthermore, to elucidate the relationship between PGE2 and cAMP, we examined the effect of COX inhibitors on the cAMP content of the mouse intestinal mucosa treated with hemolysin. Forty minutes after the COX inhibitors or vehicle solution was administered subcutaneously, 125 ng of hemolysin was injected into the intestinal loops. The cAMP content of the intestinal mucosa was determined 2.5 h after the injection, as described in Materials and Methods. The selective COX-2 inhibitor NS-398 significantly inhibited the hemolysin-induced increase in cAMP production, but the selective COX-1 inhibitor SC-560 did not affect the level of production (Fig. 5B).


Figure 5
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 5. Action of hemolysin to elevate the level of mucosal cAMP in mouse intestines (A) and effect of COX inhibitors on the action of hemolysin (B). (A) Hemolysin (125 ng) was injected into the intestinal loop and incubated for 30 or 180 min. (B) COX inhibitor (10 mg/kg) or vehicle solution was administered subcutaneously. Forty minutes later, 125 ng of hemolysin was injected into the loop and incubated for 2.5 h. The cAMP level in each mouse intestinal mucosa was determined using a radioimmunoassay kit. Values are expressed as the means ± standard errors of eight determinations. Significance (versus control). *, P < 0.02; **, P < 0.04.

 
Effect of H-89 and glibenclamide on fluid accumulation due to hemolysin. To elucidate whether the CFTR channel, a plasma membrane Cl– channel, is involved in fluid accumulation caused by hemolysin, the effects of H-89, a cAMP-dependent protein kinase A (PKA) inhibitor (7, 27), and glibenclamide, a selective CFTR inhibitor (34, 35), were examined. For the control experiment, CT, which causes diarrhea through the activation of CFTR (18, 23), was used. H-89 and glibenclamide dose-dependently inhibited the fluid accumulation induced by hemolysin and CT (Fig. 6A and B). Upon the administration of H-89 alone and glibenclamide alone, no fluid accumulated in the loops at all.


Figure 6
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 6. Effects of H-89 (A) and of glibenclamide (B) on the actions of hemolysin and CT to induce fluid secretion in mouse intestinal loops. The indicated amount of H-89 (A) or glibenclamide (B) was injected into the intestinal loops of mice together with hemolysin (125 ng) or CT (500 ng). After 2.5 h, the amount of fluid accumulated in each loop was determined. Values are expressed as the means ± standard errors of eight determinations. Significance (versus hemolysin alone or CT alone). *, P < 0.03; **, P < 0.001.

 

   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
To elucidate the mechanisms of the diarrheal disease induced by A. sobria, we examined the hemolysin-induced response in mouse intestinal epithelial cells in vivo. First, we found that hemolysin has the ability to activate the synthesis of PGE2 in mouse intestinal epithelial cells.

As lipopolysaccharide (LPS) has been shown to possess such activity (33), we were afraid that LPS was contaminating our hemolysin sample. We measured the content of LPS in our hemolysin sample by the Limulus amoebocyte lysate assay. The measurement showed that the amount of contaminating LPS in the 125-ng hemolysin sample (125 ng is enough hemolysin to induce a positive reaction in mouse intestinal loop assay) was 0.006 pg. The amount of LPS necessary to induce the production of PGE2 in mouse intestine has been reported to be 15 µg (33). From these data, we are convinced that the substance which activated the synthesis of PGE2 in the sample was hemolysin and not LPS.

The data showed that the dose of hemolysin necessary to induce the increase in PGE2 was similar to that necessary to induce fluid accumulation in the mouse intestinal loop (Fig. 1B). The time course experiment also showed a close relationship between the formation of PGE2 and the incidence of fluid secretion (Fig. 1A). The results from the experiment using carbenoxolone (Fig. 2) support that PGE2 is involved in the accumulation of fluids induced by hemolysin.

The actions of carbenoxolone to inhibit PG-degrading enzymes are well demonstrated and have been applied clinically to treat ulcers (28, 40). There are reports that carbenoxolone inhibits enzymes besides PG-degrading enzymes, such as 11β-hydroxysteroid dehydrogenase (10) and Na+-K+ ATPase (42). However, it is unclear whether carbenoxolone inhibits enzymes such as dehydrogenase and Na+-K+ ATPase in the mouse intestine. Judging from the results described here, it seems certain that PG is involved in the accumulation of fluid caused by hemolysin and that carbenoxolone promotes the actions of hemolysin through the inhibition of PG-degrading enzymes. We think that the contribution of other activities of carbenoxolone in promoting the actions of hemolysin is minor. Other compounds inhibiting PG-degrading enzymes might be useful for determining the role of PGE2 in the diarrhea induced by hemolysin.

The biosynthesis of PGE2 is regulated by COX. Two isoforms, COX-1 and COX-2, are known (38). To determine which COX is involved in fluid accumulation, we examined the effect of COX inhibitors on the actions of hemolysin. The hemolysin-induced fluid secretion and PGE2 formation were inhibited by the selective COX-2 inhibitor NS-398 (Fig. 3A and B). In contrast, the COX-1 inhibitor SC-560 did not affect the actions of hemolysin (Fig. 3A and B). The doses used in this experiment were the same as those needed to induce the selective inhibition of COX-2 and COX-1 (17, 36). In addition, the Western blot analysis revealed that hemolysin increased the level of COX-2 protein in a time-dependent manner (Fig. 4).

These findings strongly suggest that PGE2 functions as a mediator of diarrhea caused by hemolysin and that the PGE2 is produced via COX-2. The contribution of the system involving PGE2 has been proposed for diarrhea caused by several pathogens including Vibrio cholerae, Escherichia coli, Salmonella species, and Campylobacter jejuni. The diarrhea caused by these pathogens was suppressed by indomethacin, an inhibitor of PG synthesis (13, 19, 30). Furthermore, Beubler et al. reported that CT increases the level of COX-2 protein and that the PGE2 produced by the COX-2 mediates the secretion of fluids by CT (2). Therefore, the reaction system including PGE2 may be widely involved in diarrhea caused by infections of enteric pathogens.

COX-1 is constitutively expressed in various tissues, including the stomach, whereas COX-2 is virtually undetectable in tissues under resting conditions. COX-2 expression is induced by cytokines, endotoxins, growth factors, and tumor promoters (38). However, this simplified paradigm of the constitutive expression of COX-1 and the inducible expression of COX-2 has many exceptions. For example, COX-1 can be regulated during development or by certain hormones and growth factors (38), whereas COX-2 is constitutively expressed in the brain, reproductive tissues, kidney, and thymus (38) and in the colon of the mouse (24). In this study, the COX-2 protein band was clearly detected in the mucosal fraction of the intestine of untreated or control mice (phosphate-buffered saline injected) by Western blotting (Fig. 4). Similar findings were reported in other recent studies investigating COX-2 expression in the mouse (25) and in porcine intestinal tissue (3). Furthermore, Murmu et al. have demonstrated that not only the COX-2 protein band but also the mRNA for COX-2 was detected in mouse intestinal tissue (25). Taken together, these results lead to the conclusion that COX-2 is constitutively expressed in the mouse small intestine. Experiments on the expression of COX-2 under the influence of hemolysin indicated that PGE2 synthesized by COX-2 is deeply involved in the accumulation of intestinal fluid.

We previously showed that hemolysin increased the production of cAMP in cultured colonic epithelial cells (15, 16, 39). In the present study, we demonstrated that hemolysin also increased the levels of cAMP in mouse intestinal epithelial cells in vivo (Fig. 5). Deachapunya and O'Grady suggested that PGE2 activates adenylyl cyclase in cultured endometrial epithelial cells, resulting in an increase in intracellular cAMP (11). Other reports showed that there are four PGE2 receptors (EPs), designated EP1, EP2, EP3, and EP4, and that PGE2 stimulates the production of cAMP via EP2 and EP4 receptors linked to the Gs protein (4, 31). Therefore, we think that hemolysin acts on intestinal cells to produce PGE2s that emerge on the cell surface. These PGE2s stimulate EP2s or EP4s to produce cAMP.

It is well established that the intracellular cAMP activates CFTR Cl– channels through protein phosphorylation mediated by PKA (11). We previously found that hemolysin activates the cAMP-dependent Cl– secretory pathway, which is related to CFTR, in Caco-2 cells (a human colonic epithelial cell line) (39). In addition, we showed that an inhibitor of cAMP-dependent Cl– channels, 5-nitro-2-(3-phenylpropylamino)benzoic acid, suppressed the hemolysin-induced secretion of Cl– in Caco-2 cells and fluid accumulation in the mouse intestinal loop (39). In this study, we demonstrated that both H-89, a PKA inhibitor, and glibenclamide, a selective CFTR inhibitor, inhibit the fluid accumulation caused by hemolysin (Fig. 6). This strongly suggests that PKA and CFTR are involved in the diarrhea caused by hemolysin. From the results, it is speculated that the diarrhea caused by hemolysin is induced by the activation of the following pathway: hemolysin -> COX-2 -> PGE2 -> cAMP -> PKA -> CFTR Cl– channels -> secretion of fluid.

As shown in Fig. 6B, glibenclamide inhibited the actions not only of hemolysin but also of CT. Judging from the result obtained, it seems that glibenclamide is more effective at inhibiting the actions of hemolysin than at inhibiting the actions of CT. In contrast, there is not a major difference in the efficacies of H-89 in reducing the activities of both toxins, CT and hemolysin. We think these observed differences in the efficacies of the compounds used may be due to the difference in the modes of action of the two toxins upon intestinal cells.

CT acts upon intestinal cells to promote ADP ribosylation of the GTP regulatory subunit. The adenylyl cyclase in plasma membrane is activated and the intracellular cAMP level increases. The action of CT is irreversible (12). On the other hand, the mode of action of hemolysin in increasing the level of cAMP in intestinal cells has not been clearly demonstrated. As shown by Smith et al., PGE2 stimulates intestinal epithelial cell adenylate cyclase by a receptor-mediated mechanism (37). Our results presented in this report with COX inhibitors support that PGE2 is involved in the elevation of intracellular cAMP (Fig. 5B). Therefore, we think that PGE2 is a key substance in the elevation of intracellular cAMP by hemolysin.

However, the COX inhibitor could not completely inhibit the elevation of cAMP levels (Fig. 5B). These data indicate some other routes to elevate the level of cAMP, which are not disturbed by COX inhibitors. In previous reports, we found that the stimulation of adenosine receptor by adenosine appearing outside of the cells is deeply involved in the production of cAMP in cultured cells (T84 cells) by hemolysin (16, 21). It is likely that the route of the adenosine and adenosine receptor is involved in the production of cAMP by hemolysin in mouse intestine.

As mentioned, CT and hemolysin act upon intestinal cells in different manners. Accordingly, the condition of cells following attack by CT is different from that after hemolysin exposure. The difference in conditions of the cells may affect the reactivity of the intestine to some compounds. The difference in reactivities may also induce the difference in the efficacies of these compounds administered into intestinal lumen. We thought the efficacy of glibenclamide in the intestine which was exposed to CT only was low. Therefore, glibenclamide was less effective at inhibiting the actions of CT than at inhibiting the actions of hemolysin (Fig. 6B). In contrast, the activity of H-89 was expressed well in intestines exposed to both CT and hemolysin. Therefore, there was not any difference in the amounts of H-89 which were required to reduce the enterotoxic activities of both toxins (Fig. 6A).

In addition, it has been shown that hemolysin makes small pores in the membranes of cultured cells (14). The creation of such pores in the cell membranes induces cellular injury. We previously examined the relationship between cellular injury and fluid accumulation in the intestinal loop by hemolysin. Histopathological examination showed that intestinal cells were not injured when the secretion of fluid initiated, indicating that some signals to induce the secretion of fluid were sent in intestinal cells by the approach of hemolysin (14). However, it is possible that the cytotoxic activity of hemolysin is expressed, although the extent of the cell injury induced is slight. There may be some relationship between the injury and the cause and effect of diarrhea induced by hemolysin. Further studies are necessary to clarify the relationship.

Chopra et al. reported that Act of A. hydrophila increases the production of PGE2 and cAMP in murine macrophage cells, and the selective COX-2 inhibitor Celebrex significantly inhibits Act-induced PGE2 and cAMP production (8). Our hemolysin is homologous (68%) with aerolysin at the amino acid sequence level. We prepared the antiserum against hemolysin and confirmed the activity of the serum to neutralize the enterotoxic activity of our hemolysin. Furthermore, we found that the antihemolysin antiserum neutralizes the enterotoxic activity of living cells not only of A. sobria but also of A. hydrophila (14). These observations indicate that the hemolysin of A. sobria is immunologically related to the enterotoxin of A. hydrophila, which is regarded as homologous with Act. Though the mechanism of Act to induce diarrhea has not been investigated extensively, it is likely that two toxins, Act and our hemolysin, induce diarrhea via the same reaction pathway.


   ACKNOWLEDGMENTS
 
This work was supported in part by grants from the Program of Founding Research Centers for Emerging and Reemerging Infectious Diseases, MEXT Japan, and a grant-in-aid for scientific research from the Ministry of Education, Culture, Sports, Science and Technology of Japan.


   FOOTNOTES
 
* Corresponding author. Mailing address: Laboratory of Functional Morphology, School of Pharmacy, Yokohama College of Pharmacy, 601, Matano-cho, Totsuka-ku, Yokohama, Kanagawa 245-0066, Japan. Phone: 81-45-859-1300 (ext. 2208). Fax: 81-45-859-1301. E-mail: y.fujii{at}hamayaku.ac.jp Back

{triangledown} Published ahead of print on 17 December 2007. Back

Editor: R. P. Morrison


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
1. Asao, T., Y. Kinoshita, S. Kozaki, T. Uemura, and G. Sakaguchi. 1984. Purification and some properties of Aeromonas hydrophila hemolysin. Infect. Immun. 46:122-127.[Abstract/Free Full Text]
2. Beubler, E., R. Schuligoi, A. K. Chopra, D. A. Ribardo, and B. A. Peskar. 2001. Cholera toxin induces prostaglandin synthesis via post-transcriptional activation of cyclooxygenase-2 in the rat jejunum. J. Pharmacol. Exp. Ther. 297:940-945.[Abstract/Free Full Text]
3. Blikslager, A. T., D. N. Zimmel, K. M. Young, N. B. Campbell, D. Little, and R. A. Argenzio. 2002. Recovery of ischaemic injured porcine ileum: evidence for a contributory role of COX-1 and COX-2. Gut 50:615-623.[Abstract/Free Full Text]
4. Breyer, R. M., L. S. Davis, C. Nian, R. Redha, B. Stillman, H. R. Jacobson, and M. D. Breyer. 1996. Cloning and expression of the rabbit prostaglandin EP4 receptor. Am. J. Physiol. Renal Fluid Electrolyte Physiol. 270:F485-F493.[Abstract/Free Full Text]
5. Burke, V., J. Robinson, H. M. Atkinson, and M. Gracey. 1982. Biochemical characteristics of enterotoxigenic Aeromonas spp. J. Clin. Microbiol. 15:48-52.[Abstract/Free Full Text]
6. Chakraborty, T., A. Schmid, S. Notermans, and R. Benz. 1990. Aerolysin of Aeromonas sobria: evidence for formation of ion-permeable channels and comparison with alpha-toxin of Staphylococcus aureus. Infect. Immun. 58:2127-2132.[Abstract/Free Full Text]
7. Chijiwa, T., A. Mishima, M. Hagiwara, M. Sano, K. Hayashi, T. Inoue, K. Naito, T. Toshioka, and H. Hidaka. 1990. Inhibition of forskolin-induced neurite outgrowth and protein phospholylation by a newly synthesized selective inhibitor of cyclic AMP-dependent protein kinase, N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinoline sulfonamide (H-89), of PC12D Pheochromocytoma cells. J. Biol. Chem. 265:5267-5272.[Abstract/Free Full Text]
8. Chopra, A. K., X.-J. Xu, D. Ribardo, M. Conzalez, K. Kuhl, J. W. Peterson, and C. W. Houston. 2000. The cytotoxic enterotoxin of Aeromonas hydrophila induces proinflammatory cytokine production and activates arachidonic acid metabolism in macrophages. Infect. Immun. 68:2808-2818.[Abstract/Free Full Text]
9. Daily, O. P., S. W. Joseph, and J. C. Coolbaugh. 1981. Association of Aeromonas sobria with human infection. J. Clin. Microbiol. 13:769-777.[Abstract/Free Full Text]
10. Davani, B., A. Khan, M. Hult, E. Martensson, S. Okret, S. Efendic, H. Jornvall, and U. C. T. Oppermann. 2000. Type 11β-hydroxysteroid dehydrogenase mediates glucocorticoid activation and insulin release in pancreatic islets. J. Biol. Chem. 275:34841-34844.[Abstract/Free Full Text]
11. Deachapunya, C., and S. M. O'Grady. 1998. Regulation of chloride secretion across porcine endometrial epithelial cells by prostaglandin E2. J. Physiol. 508:31-47.[Abstract/Free Full Text]
12. Enomoto, K., and D. M. Gill. 1980. Cholera toxin activation of adenylcyclase. J. Biol. Chem. 255:1252-1258.[Free Full Text]
13. Fujii, Y., Y. Kondo, and K. Okamoto. 1995. Involvement of prostaglandin E2 synthesis in the intestinal secretory action of Escherichia coli heat-stable enterotoxin II. FEMS Microbiol. Lett. 130:259-266.[CrossRef][Medline]
14. Fujii, Y., T. Nomura, H. Kanzawa, M. Kameyama, H. Yamanaka, M. Akita, K. Setsu, and K. Okamoto. 1998. Purification and characterization of enterotoxin produced by Aeromonas sobria. Microbiol. Immunol. 42:703-714.[Medline]
15. Fujii, Y., T. Nomura, and K. Okamoto. 1999. Hemolysin of Aeromonas sobria stimulates production of cyclic AMP by cultured cells. FEMS Microbiol. Lett. 176:67-72.[CrossRef][Medline]
16. Fujii, Y., T. Nomura, R. Yokoyama, S. Shinoda, and K. Okamoto. 2003. Studies of the mechanism of action of the aerolysin-like hemolysin of Aeromonas sobria in stimulating T84 cells to produce cyclic AMP. Infect. Immun. 71:1557-1560.[Abstract/Free Full Text]
17. Futaki, N., Y. Hamasaka, S. Takahashi, S. Higuchi, and S. Otomo. 1993. Selective inhibition of NS-398 on prostanoid production in inflamed tissue in the rat carrageenan-air-pouch inflammation. J. Pharm. Pharmacol. 45:753-755.[Medline]
18. Gabriel, S. E., K. N. Brigman, B. H. Koller, R. C. Boucher, and M. J. Stutts. 1994. Cystic fibrosis heterozygote resistance to cholera toxin in the cystic fibrosis mouse model. Science 266:107-109.[Abstract/Free Full Text]
19. Hawkey, C. J., and D. S. Rampton. 1985. Prostaglandins and the gastrointestinal mucosa: are they important in its function, disease or treatment? Gastroenterology 89:1162-1188.[Medline]
20. Hitotsubashi, S., Y. Fujii, H. Yamanaka, and K. Okamoto. 1992. Some properties of purified Escherichia coli heat-stable enterotoxin II. Infect. Immun. 60:4468-4474.[Abstract/Free Full Text]
21. Li, Y., K. Okamoto, E. Takahashi, S. Miyoshi, S. Shinoda, T. Tsuji, and Y. Fujii. 2005. A hemolysin of Vibrio mimicus (VMH) stimulates cells to produce ATP and cyclic AMP which appear to be secretory mediators. Microbiol. Immunol. 49:73-78.[Medline]
22. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.[Free Full Text]
23. Ma, T., J. R. Thiagarajah, H. Yang, N. D. Sonawane, C. Folli, L. J. V. Galietta, and A. S. Verkman. 2002. Thiazolidinone CFTR inhibitor identified by high-throughput screening blocks cholera toxin-induced intestinal fluid secretion. J. Clin. Investig. 110:1651-1658.[CrossRef][Medline]
24. MacNaughton, W. K., and K. Cushing. 2000. Role of constitutive cycooxygenase-2 in prostaglandin-dependent secretion in mouse colon in vitro. J. Pharmacol. Exp. Ther. 293:539-544.[Abstract/Free Full Text]
25. Murmu, N., J. Jung, D. Mukhopadhyay, C. W. Houchen, T. E. Riehl, W. F. Stenson, A. R. Morrison, T. Arumugam, B. K. Dieckgraefe, and S. Anant. 2004. Dynamic antagonism between RNA-binding protein CUGBP2 and cyclooxygenase-2-mediated prostaglandin E2 in radiation damage. Proc. Natl. Acad. Sci. USA 101:13873-13878.[Abstract/Free Full Text]
26. Nomura, T., Y. Fujii, and K. Okamoto. 1999. Secretion of hemolysin of Aeromonas sobria as protoxin and contribution of the propeptide region removed from the protoxin to the proteolytic stability of the toxin. Microbiol. Immunol. 43:29-38.[Medline]
27. Okamoto, F., H. Kajiya, H. Fukushima, E. Jimi, and K. Okabe. 2004. Prostaglandin E2 activates outwardly rectifying Cl– channels via a cAMP-dependent pathway and reduces cell motility in rat osteoclasts. Am. J. Physiol. Cell Physiol. 287:C114-C124.[Abstract/Free Full Text]
28. Peskar, B. M. 1983. Effects of carbenoxolone on the gastric mucosal prostaglandin and thromboxane system. Acta Gastroenterol. Belg. 46:429-436.[Medline]
29. Potomski, J., V. Burke, I. Watoson, and M. Gracey. 1987. Purification of cytotoxic enterotoxin of Aeromonas sobria by use of monoclonal antibodies. J. Med. Microbiol. 23:171-177.[Abstract]
30. Rask-Madsen, J. 1986. Eicosanoids and their role in the pathogenesis of diarrhoeal diseases. Clin. Gastroenterol. 15:545-566.[Medline]
31. Reimer, R., H. K. Heim, R. Muallem, H. S. Odes, and K. F. Sewing. 1992. Effect of EP-receptor subtype specific agonists and other prostanoids on adenylate cyclase activity of duodenal epithelial cells. Prostaglandins 44:485-493.[CrossRef][Medline]
32. Ribardo, D. A., K. R. Kuhl, I. Boldogh, J. W. Peterson, C. W. Houston, and A. K. Chopra. 2002. Early cell signaling by the cytotoxic enterotoxin of Aeromonas hydrophila in macrophages. Microb. Pathog. 32:149-163.[CrossRef][Medline]
33. Riehl, T., S. Cohn, T. Tessner, S. Schloemann, and W. F. Stenson. 2000. Lipopolysaccharide is radioprotective in the mouse intestine through a prostaglandin-mediated mechanism. Gastroenterology 118:1106-1116.[CrossRef][Medline]
34. Schultz, B. D., A. D. DeRoos, C. J. Venglarik, A. K. Singh, R. A. Frizzell, and R. J. Bridges. 1996. Glibenclamide blockade of CFTR chloride channels. Am. J. Physiol. 271:L192-L200.[Medline]
35. Sheppard, D. N., and K. A. Robinson. 1997. Mechanism of glibenclamide inhibition of cystic fibrosis transmembrane conductance regulator Cl– channels expressed in a murine cell line. J. Physiol. 503:333-346.[CrossRef][Medline]
36. Smith, C. J., Y. Zhang, C. M. Koboldt, J. Muhammad, B. S. Zweifel, A. Shaffer, J. J. Talley, J. L. Masferrer, K. Seibert, and P. C. Isakson. 1998. Pharmacological analysis of cyclooxygenase-1 in inflammation. Proc. Natl. Acad. Sci. USA 95:13313-13318.[Abstract/Free Full Text]
37. Smith, G., G. Warhurst, M. Lees, and L. Turnberg. 1987. Evidence that PGE2 stimulates intestinal epithelial cell adenylate cyclase by a receptor-mediated mechanism. Dig. Dis. Sci. 32:71-75.[CrossRef][Medline]
38. Smith, W. L., and R. Langenbach. 2001. Why there are two cyclooxygenase isozymes. J. Clin. Investig. 107:1491-1495.[Medline]
39. Tanoue, N., A. Takahashi, K. Okamoto, Y. Fujii, Y. Taketani, N. Harada, M. Nakano, and Y. Nakaya. 2005. A pore-forming toxin produced by Aeromonas sobria activates cAMP-dependent Cl– secretory pathways to cause diarrhea. FEMS Microbiol. Lett. 242:195-201.[CrossRef][Medline]
40. Turpie, A. G., and T. J. Thomson. 1965. Carbenoxolone sodium in the treatment of gastric ulcer with special reference to side effects. Gut 6:591-594.[Free Full Text]
41. Wilmsen, H. U., F. Pattus, and J. T. Buckley. 1990. Aerolysin, a hemolysin from Aeromonas hydrophila, forms voltage-gated channels in planar lipid bilayers. J. Membr. Biol. 115:71-81.[CrossRef][Medline]
42. Zhou, Q. L., Z. Q. Zhang, T. Nagasawa, and S. Hiai. 1996. The structure activity relationship of saikosaponins and glycyrrhizin derivatives for Na+, K(+)-ATPase inhibiting action. Yao Xue Xue Bao 31:496-501. (In Chinese.)[Medline]

Infection and Immunity, March 2008, p. 1076-1082, Vol. 76, No. 3
0019-9567/08/$08.00+0     doi:10.1128/IAI.01098-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
IAI.01098-07v1
76/3/1076    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fujii, Y.
Right arrow Articles by Okamoto, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fujii, Y.
Right arrow Articles by Okamoto, K.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
J. Bacteriol. J. Virol. Eukaryot. Cell
Microbiol. Mol. Biol. Rev. Clin. Vaccine Immunol. All ASM Journals