Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Helicobacter pylori is regarded as an important pathogen in gastric and duodenal inflammation (Graham, 1989; Marshall, 1994). Although the pathological mechanisms involved in H. pylori-induced mucosal inflammation are not completely known, there is accumulating evidence that activated neutrophils play an important role (Blaser, 1992; Yoshida et al., 1993). Gastric epithelial cells produce and secrete interleukin (IL)-8, a potent chemotactic and activating factor for leukocytes, in response to H. pylori infection both in vivo and in vitro (Crabtree et al., 1994; Crowe et al., 1995). Direct contact with H. pylori initiates epithelial cell signaling events that result in the expression of IL-8 (Crowe et al., 1995). Prolonged IL-8 production by gastric epithelial cells during H. pylori infection could result in the recruitment of leukocytes to infected tissues and therefore may be important in the regulation of inflammatory and immune processes in response to this bacterium (Ben-Baruch et al., 1995). The expression of the IL-8 gene is primarily controlled at the transcriptional level. Nucleotide sequence analysis of the 5' regulatory region of the IL-8 gene has revealed potential binding sites for several transcription factors, including nuclear factor-B (NF-B), NF-IL6, and activator protein-1 (AP-1) (Mukaida et al., 1994). We and others have shown that the H. pylori-induced transcription of IL-8 gene requires the activation of NF-B and, to a lesser extent, AP-1 in human gastric epithelial cells (Aihara et al., 1998; Masamune et al., 1999). NF-B is known to be involved in the control of the cytokine-induced expression of many immune and inflammatory-response genes (Grilli et al., 1993). NF-B is localized in the cytoplasm in an inactive form where it is associated with the inhibitory protein IB-. IB- retains the NF-B complex in the cytoplasm and inhibits DNA binding. Various NF-B activators cause phosphorylation and degradation of the inhibitory protein IB-, allowing NF-B to be released from the complex. NF-B then moves to the nucleus where it binds to the DNA recognition site and mediates gene transcription (Grilli et al., 1993).

Sphingolipid metabolites such as ceramide (Hannun, 1994; Kolesnick and Golde, 1994) have been implicated as important signaling molecules in the regulation of diverse cellular functions. Several extracellular stimuli, including tumor necrosis factor- (TNF-), IL-1, and interferon- have been shown to activate sphingomyelinases, resulting in sphingomyelin hydrolysis and the generation of ceramide (termed the "sphingomyelin-ceramide pathway"). In turn, ceramide mediates the effects of these agonists on cell growth, differentiation, apoptosis, and inflammatory responses (Hannun, 1994; Kolesnick and Golde, 1994). We have recently shown that H. pylori-dependent ceramide production may activate NF-B and AP-1, and subsequently mediate the increased IL-8 expression in human gastric cancer cell lines (Masamune et al., 1999).

In this study, we examined the effects of rebamipide, an antigastritis and antiulcer agent, on the ceramide-mediated signaling pathway activated by H. pylori in gastric epithelial cells. We here report that rebamipide inhibited H. pylori-dependent ceramide generation and subsequent IL-8 expression in Kato III human gastric cancer cells. Our results suggest a novel mechanism by which rebamipide protects against the mucosal inflammation associated with H. pylori infection.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. C₂-ceramide (N-acetylsphingosine, a cell-permeable ceramide analog) and sn-1,2-diacylglycerol kinase were from Calbiochem (San Diego, CA). [-³²P]dCTP and [-³²P]ATP were purchased from Amersham Pharmacia Biotech (Buckinghamshire, England). Antibodies used for Western blotting were from New England Biolabs (Beverly, MA). PD98059 was purchased from BIOMOL Research Laboratories (Plymouth Meeting, PA). Rebamipide {2-(4-chlorobenzoylamino)-3-[2(1H)-quinolinon-4-yl]propionic acid} (Uchida et al., 1985), provided by Otsuka Pharmaceutical Co. (Tokyo, Japan), was dissolved and stocked at 200 mM in dimethyl sulfoxide. All other reagents were from Sigma Chemical Co. (St. Louis, MO) unless specifically described.

H. pylori and Epithelial Cell Culture. H. pylori (cagA⁺ strain 43504; American Type Culture Collection, Rockville, MD) was grown on brucella agar (Difco, Detroit, MI) plates supplemented with 7% fetal bovine serum and incubated at 37°C in a microaerophilic environment. After an overnight incubation, bacteria were harvested and transferred to brucella broth supplemented with 7% fetal bovine serum. Following another overnight incubation, the bacterial suspension was pelleted by centrifugation and resuspended in F-12 HAM medium (Life Technologies, Rockville, MD), and bacterial numbers were standardized by optical density measurement at 600 nm. The motility of H. pylori in cultures was confirmed by a phase-contrast microscopy before experimental use.

Enzyme-Linked Immunosorbent Assay. After a 24-h incubation, cell culture supernatants were harvested and stored at 80°C until the measurement. IL-8 levels in the supernatants were measured by enzyme-linked immunosorbent assay (Endogen, Woburn, MA) according to the manufacturer's instructions.

Northern Blot Analysis. Following a 4 h-incubation, total RNA was isolated using an RNeasy total RNA preparation kit (Qiagen, Chatsworth, CA). Ten micrograms of total RNA was separated on a 1% agarose-2.2 M formaldehyde gel, and transferred to a nylon membrane filter (Amersham Pharmacia Biotech). Blots were hybridized for 16 h at 42°C to the ³²P-labeled DNA probes of IL-8 and -actin generated by polymerase chain reaction as previously described (Masamune et al., 1999). After the hybridization, the filter was washed twice with 2× standard saline citrate (3 M NaCl and 0.3 M sodium citrate) and 0.1% SDS at room temperature for 10 min followed by two washes with 0.2× standard saline citrate and 0.1% SDS at 42°C for 30 min. The washed filter was subjected to the BAS 1500 system (Fuji Film, Tokyo, Japan).

Luciferase Assay. Luciferase expression vectors containing the minimally essential promoter region of the IL-8 gene (bp 133 to +44), and those containing two NF-B consensus binding sites (Masamune et al., 1999) were kindly provided by Dr. Naofumi Mukaida (Kanazawa University, Kanazawa, Japan). For the luciferase assay, 1 × 10⁶ Kato III cells were transfected with 2 µg of each luciferase vector along with 40 ng of pRL-TK vector (Promega, Madison, WI) as an internal control, using lipofectAMINE reagent (Life Technologies). After 20 h, the cells were treated with rebamipide at the indicated concentrations for 4 h, and then were stimulated with C₂-ceramide in the presence of rebamipide. After another 24-h incubation, cell lysates were prepared using a Pica Gene kit (Toyo Ink Co., Tokyo, Japan), and the light intensities were measured using a model Lumat LB9507 luminescence reader (EG&G Berthold, Bad Wildbad, Germany).

Western Blot Analysis. Kato III cells were treated with C₂-ceramide or H. pylori for 30 min and lysed in SDS buffer (62.5 mM Tris-HCl at pH 6.8, 2% SDS, 10% glycerol, 50 mM dithiothreitol, 0.1% bromophenol blue) for 15 min on ice. The samples were then sonicated, heated to 100°C for 5 min, and centrifuged at 15,000 rpm for 5 min to remove insoluble cell debris. They were fractionated on a 10 to 20% SDS-polyacrylamide gel and transferred to nitrocellulose membrane (Bio-Rad, Hercules, CA). The membranes were blocked with a 5% (w/v) solution of dried milk in Tris-buffered saline (pH 7.4). The levels of activated, phosphorylated MAP kinases in the samples were determined using phosphospecific MAP kinase antibodies [extracellular signal-regulated kinase (ERK)1/2, c-Jun NH₂-terminal kinase/stress-activated protein kinase (JNK/SAPK), and p38 kinases] in a 1:1000 dilution. After incubation with a secondary antibody (goat anti-rabbit antibody, horseradish peroxidase conjugated), proteins were visualized using an ECL kit (Amersham Pharmacia Biotech). The levels of pan-MAP kinases and IB- were determined in a similar manner.

Ceramide Measurement. Following treatment of the cells with H. Pylori, lipids were extracted as previously described (Bligh and Dyer, 1959) and the ceramide content was quantified using diacylglycerol kinase as previously described (Masamune et al., 1996). Briefly, dried lipids were solubilized in 20 µl of octyl--D-glucoside/cardiolipin solution (7.5% octyl--D-glucoside and 5 mM cardiolipin in 1 mM diethylenetriaminepentaacetic acid) by ultrasonication. The reaction was carried out in a final volume of 100 µl containing the 20-µl sample solution, 50 mM imidazole HCl, 50 mM NaCl, 12.5 mM MgCl₂, 1 mM EGTA, 2 mM dithiothreitol, 6.6 µg of diacylglycerol kinase, and 1 mM [-³²P]ATP for 30 min at 22°C. The major lipid products of their phosphorylation reaction were phosphatidic acid (from diacylglycerol) and ceramide-1-phosphate (from ceramide). They were completely resolved by thin-layer chromatography using chloroform/acetone/methanol/acetic acid/water (10:4:2:2:1, v/v) as a solvent, and visualized by autoradiography. Ceramide-1-phosphate spots were identified by the comparison with the standard sample of ceramide phosphorylated under identical conditions and developed in parallel. Quantification of ceramide-1-phosphate was carried out using the BAS 1500 system and software provided with the instrument by the manufacturer.

Statistical Analysis. Differences between experimental groups were evaluated by the two-tailed unpaired Student's t test. A p value less than 0.05 was considered statistically significant

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Rebamipide Inhibited C₂-Ceramide-Induced IL-8 Expression in Kato III Cells. We first examined the effects of rebamipide on IL-8 production in Kato III cells by enzyme-linked immunosorbent assay. As previously reported (Masamune et al., 1999), C₂-ceramide increased IL-8 production in a dose-dependent manner (Fig. 1A). C₂-ceramide at 10 µM induced IL-8 production, representing a 9.4-fold increase compared with the unstimulated control. Rebamipide inhibited C₂-ceramide-induced IL-8 expression in a dose-dependent manner (Fig. 1B). A statistically significant inhibition in IL-8 protein release was evident at as low as 10 µM rebamipide (p < 0.05). The maximal effect was observed at 1 mM, where IL-8 expression was approximately 27% of the control value. In these studies, rebamipide was added to the culture medium using dimethyl sulfoxide as a vehicle, and the amounts of dimethyl sulfoxide used (0.5% at 1 mM rebamipide) did not affect cell viability, morphology, or IL-8 production (data not shown). We examined the effect of increasing the time of exposure on IL-8 production. Rebamipide treatment was started 24, 4, 2, and 1 h before, or at 0 and 1 h after C₂-ceramide stimulation. The inhibitory effect of rebamipide was more evident in a time-dependent manner up to 4 h, but further enhancement of the effect was not observed even if the exposure time was extended to up to 24 h (Fig. 2). Rebamipide did not inhibit the C₂-ceramide-induced IL-8 expression if it was added after the C₂-ceramide treatment. Furthermore, washing out the rebamipide abolished the inhibition of C₂-ceramide-induced IL-8 production, suggesting that rebamipide must be present during the incubation period of C₂-ceramide stimulation to elicit the inhibitory effect. Rebamipide also inhibited the H. pylori-induced IL-8 expression. Rebamipide at 1 mM decreased IL-8 production to approximately 55% of the control (Fig. 1B).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Rebamipide inhibited C2-ceramide-induced IL-8 production in Kato III cells. A, Kato III cells were treated with increasing concentrations of C2-ceramide in serum-free medium. After 24 h, the IL-8 levels in the conditioned media were determined by enzyme-linked immunosorbent assay. B, Kato III cells were pretreated with rebamipide at the indicated concentrations for 4 h then treated with C2-ceramide (10 µM) or H. pylori (5 × 107 colony forming units/ml) in the presence of rebamipide. The IL-8 levels in conditioned media harvested after 24 h were determined. Data shown are expressed as means ± S.D. (n = 6 for each data point, *p < 0.05 and ** p < 0.01 versus control).

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effect of exposure time to rebamipide on C2-ceramide-induced IL-8 production. Kato III cells were treated with rebamipide for 24, 4, 2, and 1 h before, or at 0 and 1 h after C2-ceramide (at 10 µM) stimulation and then the IL-8 levels in the culture medium were determined after 24-h stimulation with C2-ceramide. In the case of the washing out procedure, cells were treated with 1 mM rebamipide for 4 h and then rebamipide was washed out with serum-free medium prior to C2-ceramide stimulation. Cells were then stimulated with C2-ceramide (at 10 µM) in the absence of rebamipide. The IL-8 levels in conditioned media harvested after 24 h were determined by enzyme-linked immunosorbent assay. Data shown are expressed as means ± S.D. (n = 4 for each data point, *p < 0.05 and **p < 0.01 versus control).

Rebamipide Decreased the Level of IL-8 mRNA. We studied the effects of rebamipide on the IL-8 gene expression. We determined the IL-8 mRNA levels by Northern blot analysis. IL-8 mRNA expression was induced by treatment with 10 µM C₂-ceramide, and it was inhibited in the presence of rebamipide in a dose-dependent manner (Fig. 3). These results were roughly consistent with the results of the enzyme-linked immunosorbent assay, and suggested that rebamipide decreased the steady-state level of IL-8 mRNA.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Effect of rebamipide on C2-ceramide-induced IL-8 gene expression. Kato III cells were pretreated with various concentrations of rebamipide for 4 h, and then treated with C2-ceramide at 10 µM in the presence of rebamipide. After 4 h of C2-ceramide-stimulation, total RNA was extracted and the levels of IL-8 and -actin mRNA were determined by Northern blot analysis. Results shown are representative of three separate experiments.

Rebamipide Inhibited Ceramide-Induced IL-8 Gene Transcription. The expression of IL-8 gene is primarily controlled at the transcriptional level (Mukaida et al., 1994). We next examined the effects of rebamipide on the C₂-ceramide-induced IL-8 gene transcription. C₂-ceramide markedly enhanced the luciferase activity in Kato III cells transfected with luciferase expression vectors containing the minimally essential promoter region of the IL-8 gene (bp 133 to +44). Pretreatment of the cells with rebamipide decreased the induced luciferase activity (Fig. 4), suggesting that rebamipide inhibited the IL-8 gene transcription induced by C₂-ceramide.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Rebamipide inhibited IL-8 gene transcription. Kato III cells were transfected with the luciferase expression vector containing the 5'-flanking region of the human IL-8 gene spanning from bp 133 to +44. After 20 h, the cells were treated with rebamipide at the indicated concentrations for 4 h, and then were stimulated with C2-ceramide (at 10 µM) in the presence of rebamipide. After another 24-h incubation, the intracellular luciferase activities were determined. The data represent mean values ± S.D. (n = 4 for each data point).

Activation of NF-B Was Inhibited by Rebamipide. We and others have shown that both ceramide- and H. pylori-induced transcription of IL-8 gene requires the activation of NF-B and, to a lesser extent, AP-1 in human gastric epithelial cells (Aihara et al., 1997; Masamune et al., 1999). Because activation of NF-B plays a central role in IL-8 gene transcription, we examined the effect of rebamipide on the activation of NF-B. As shown in Fig. 5A, rebamipide suppressed the C₂-ceramide-induced NF-B-dependent luciferase activity, suggesting that rebamipide inhibited IL-8 gene transcription at least in part through the inhibition of NF-B activation. Phosphorylation and degradation of the inhibitory protein IB-, and subsequent dissociation of this protein from NF-B are thought to be necessary for the activation (Grilli et al., 1993). We examined the effect of rebamipide on the degradation of IB- by Western blotting. Rebamipide inhibited the C₂-ceramide-induced degradation of IB-, further supporting that rebamipide inhibited the activation of NF-B (Fig. 5B).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Rebamipide inhibited the activation of NF-B. A, Kato III cells were transfected with the luciferase expression vector containing two consensus NF-B binding sites. After 20 h, the cells were treated with rebamipide at the indicated concentrations for 4 h and then were stimulated with C2-ceramide (at 10 µM) in the presence of rebamipide. After another 24-h incubation, the intracellular luciferase activities were determined. The data represent mean values ± S.D. (n = 6 for each data point). B, Kato III cells were treated with C2-ceramide (at 10 µM) or H. pylori (5 × 107 colony forming units/ml) in the presence or absence of rebamipide (1 mM). After 30 min, the cells were lysed and the cellular levels of IB- were determined by Western blotting. Results shown are representative of three separate experiments.

Rebamipide Inhibited the Ceramide-Induced Activation of MAP Kinases. MAP kinases are a family of ubiquitous, highly conserved, cell signaling molecules (Robinson and Cobb, 1997). Upon activation by upstream kinases, MAP kinases phosphorylate downstream kinases and/or mediators, including transcription factors. MAP kinases can be activated by a wide variety of extracellular stimuli such as growth factors, proinflammatory cytokines, and stresses. Three main classes of MAP kinases have been characterized: ERK1/2 (also known as p42/p44 kinases), JNK/SAPK, and p38 MAP kinases (Robinson and Cobb, 1997). H. pylori has been shown to activate these MAP kinases and the activation may be involved in the IL-8 expression (Keates et al., 1999; Naumann et al., 1999). We first examined the effects of inhibitors of MAP kinase pathways on H. pylori-induced IL-8 expression. We used PD98059, a specific inhibitor of MAP kinase kinase activation, which prevents the activation of ERK1/2, and a selective p38 inhibitor, SB203580. In agreement with the previous reports (Keates et al., 1999; Naumann et al., 1999), H. pylori-induced IL-8 expression was partially blocked by PD98059 and SB203580 (Fig. 6A). Both inhibitors did not exhibit cytotoxicity up to 25 µM. PD98059 at 25 µM did not alter the activation of NF-B, JNK/SAPK, nor p38 MAP kinases (data not shown). SB203580 at 25 µM did not alter the activation of NF-B, JNK/SAPK, or ERK1/2 (data not shown). In addition, C₂-ceramide-induced IL-8 expression was partially blocked by these inhibitors (Fig. 6B). Rebamipide enhanced inhibitory effects of these inhibitors on IL-8 expression (Fig. 6).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Ceramide-induced IL-8 expression was partially blocked by inhibitors of MAP kinase pathways. Kato III cells were pretreated with PD98059 (at 25 µM) or SB203580 (at 25 µM) in the presence or absence of rebamipide (at 1 mM) for 4 h and then treated with H. pylori (at 5 × 107 colony forming units/ml, left) or C2-ceramide (at 10 µM, right). The IL-8 levels in conditioned media harvested after 24 h were determined. Data shown are expressed as means ± S.D. (n = 5 for each data point, *p < 0.05 and **p < 0.01).

View larger version (43K): [in this window] [in a new window]   Fig. 7.   Rebamipide inhibited the activation of MAP kinases. Kato III cells were treated with C2-ceramide (10 µM) or H. pylori (5 × 107 colony forming units/ml) in the presence or absence of rebamipide (1 mM). After 30 min, the cells were lysed and the levels of activated, phosphorylated MAP kinases in the samples were determined by Western blotting using phosphospecific MAP kinases antibodies (ERK1/2, p38, or JNK/SAPK). The levels of pan-MAP kinases were also determined. Arrows denote immunoreactive bands. Results shown are representative of three separate experiments.

Rebamipide Attenuated the H. pylori-Dependent Increase in the Intracellular Ceramide Level. We previously reported that H. pylori increased the intracellular level of ceramide (Masamune et al., 1999). In Kato III cells, H. pylori infection induced statistically significant increases (approximately 2.7-fold) in the ceramide content that peaked at 60 min. Rebamipide inhibited the H. pylori-dependent increase in the intracellular level of ceramide in a dose-dependent manner (Fig. 8). Rebamipide at 1 mM decreased the H. pylori-dependent increase in the intracellular ceramide level by 59% compared with the control in the absence of rebamipide.

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Rebamipide attenuated the H. pylori-dependent increase in the intracellular ceramide level. Kato III cells were treated with rebamipide at the indicated concentrations for 4 h and then stimulated with H. pylori (5 × 107 colony forming units/ml). After a 1-h incubation, lipids were extracted, and the intracellular ceramide levels were determined by diacylglycerol kinase assay. The results represent mean values ± S.D. from three separate experiments.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

H. pylori infection induces the mucosal production of various cytokines in the host, especially IL-8, and the host-parasite interaction has an important role in the pathogenesis of gastric mucosal inflammation (Blaser, 1992; Yoshida et al., 1993). Indeed, gastric mucosal levels of IL-8 are increased and its activity correlates with the histological severity in patients with H. pylori gastritis (Ando et al., 1996). Inhibition of IL-8 production in gastric epithelial cells may reduce the inflammation of the gastric mucosa and may have a potential therapeutic application. Rebamipide, an antiulcer and antigastritis agent, has several anti-inflammatory actions: it scavenges hydroxyl radicals (Naito et al., 1996) and inhibits the production of oxygen radicals by activated neutrophils (Ogino et al., 1992), neutrophil adhesion to endothelial cells, the expression of adhesion molecules (Suzuki et al., 1994), and IL-8 production by H. pylori in human gastric cancer cell lines (Aihara et al., 1998). We have recently shown that H. pylori-dependent ceramide production may activate NF-B and subsequently mediate the increased IL-8 expression in human gastric cancer cell lines (Masamune et al., 1999). To dissect the molecular mechanisms by which rebamipide inhibits IL-8 production, we evaluated the effect of rebamipide on the ceramide signaling pathway in Kato III human gastric cancer cells. In this study, we have demonstrated that rebamipide inhibited the IL-8 production induced by C₂-ceramide as well as that by H. pylori. This is in agreement with the previous report that rebamipide suppressed H. pylori-induced IL-8 production in MKN1 and MKN45 human gastric cancer cell lines (Aihara et al., 1998). A statistically significant inhibition in IL-8 protein release was evident at as low as 10 µM rebamipide (p < 0.05). The maximal effect was observed at 1 mM. A previous pharmacokinetic study demonstrated that the rebamipide concentration in the gastric mucosa reached 10 µM to 1 mM after oral administration of a clinically effective dose (Naito et al., 1996). Hence, the in vitro suppression of H. pylori- or C₂-ceramide-induced IL-8 production by rebamipide was observed at this effective in vivo concentration in the gastric mucosa.

Rebamipide appears to exert its inhibitory effect on IL-8 expression at the transcriptional level. Northern blot analysis revealed that rebamipide suppressed the C₂-ceramide-induced increase of the steady-state level of IL-8 mRNA. Furthermore, we demonstrated the inhibitory effect of rebamipide on the IL-8 gene transcription using an IL-8 promoter-reporter gene construct. Because activation of NF-B plays a central role in the expression of IL-8 and rebamipide inhibited the ceramide-induced NF-B-dependent transcriptional activity, it seems likely that rebamipide inhibited IL-8 expression at least in part through the attenuated activation of NF-B. Phosphorylation and degradation of the inhibitory protein IB-, and subsequent dissociation of this protein from NF-B are thought to be necessary for the activation (Grilli et al., 1993). Rebamipide inhibited the degradation of IB- induced by H. pylori and C₂-ceramide, further supporting the inhibitory effects of rebamipide on NF-B activation. It has been shown that rebamipide suppressed NF-B activation and subsequent IL-8 expression induced by H. pylori, IL-1, and TNF- (Aihara et al., 1998). In addition, rebamipide inhibited C₂-ceramide-induced NF-B activation in the present study, suggesting that rebamipide may interact with the common pathways leading to NF-B activation. The importance of the oxidation state for NF-B activation has been demonstrated (Menon et al., 1993). Indeed, hydrogen peroxide, a reactive oxidant, activates NF-B and the activation is repressed by antioxidants such as N-acetylcysteine in human gastric epithelial cells (Shimada et al., 1999). The chemical basis for the inhibitory effect of N-acetylcysteine seems to lie in the oxygen radical-scavenging effect of the thiol groups. Pyrrolidine dithiocarbamate, a proven free radical scavenger, also potently inhibits the activation of NF-B and/or NF-B interaction with its upstream regulatory binding site, thereby preventing NF-B-mediated gene transcription (Munoz et al., 1996). Yoshida et al. (1996) reported that rebamipide could scavenge active oxygen species and inhibit superoxide production in human neutrophils stimulated by H. pylori extract. It has been also demonstrated that rebamipide inhibits lipid peroxidation and oxidant-mediated activation of NF-B, thereby decreasing IL-8 production by H. pylori (Kim et al., 2000). The alteration of the oxidation state by rebamipide through its oxygen radical-scavenging effect in gastric epithelial cells may, therefore, account for the inhibition of NF-B activation and the subsequent IL-8 expression. Since NF-B binding sites are found in the promoters of a variety of proinflammatory or immune-response genes, including cytokines, acute-phase reactant proteins, and adhesion molecules (Grilli et al., 1993), it would be of interest to examine whether the inhibition of NF-B activation in gastric epithelial cells by rebamipide could inhibit the expression of multiple pathophysiologically relevant agents.

In this study, we have shown that rebamipide inhibited the ceramide generation induced by H. pylori. As previously reported (Aihara et al., 1998), rebamipide at the concentrations used in this study did not affect the growth of H. pylori in vitro; thus, direct cytotoxic effects could be excluded. It has been shown that rebamipide inhibits the adhesion of H. pylori to human gastric cancer cell lines (Hayashi et al., 1998). This may at least in part account for the inhibition of the H. pylori-dependent ceramide generation by rebamipide. Another possibility is that rebamipide may directly inhibit the enzymatic activity of sphingomyelinase. Cytokine-induced sphingomyelin hydrolysis and the generation of ceramide have been shown to be redox-sensitive; antioxidants such as N-acetylcysteine and pyrrolidine dithiocarbamate potently inhibited the cytokine-induced degradation of sphingomyelin to ceramide (Singh et al., 1998). Furthermore, Liu et al. (1998) reported that gluthathione inhibited the TNF--induced activation of neutral sphingomyelinase in human acute lymphoblastic leukemic MOLT-4 and human mammary carcinoma MCF-7 cells in vitro. Rebamipide has been shown to increase the glutathione level in the colon in an acetic acid-induced colitis model (Sakurai et al., 1998). Although we did not directly measure the effect of rebamipide on H. pylori-induced sphingomyelinase activity, it is possible that rebamipide directly suppressed sphingomyelinase activity through the alterations in the level of glutathione. Experiments to test this hypothesis are underway in our laboratory.

Recent studies have shown that H. pylori activates MAP kinases in human gastric cancer cell lines (Keates et al., 1999; Naumann et al., 1999). In this study, we have shown that rebamipide inhibited the activation of three classes of MAP kinases (JNK/SAPK, ERK1/2, and p38 MAP kinases). MAP kinases are key elements in the regulation of cellular responses to external inflammatory and proliferative signals (Robinson and Cobb, 1997). Several studies have implicated MAP kinases as upstream mediators of NF-B and AP-1 activation and cytokine gene expression, including IL-8 (Robinson and Cobb 1997). Different strains (cag⁺ and cag strains) of H. pylori vary in their ability to activate MAP kinase pathways in gastric epithelial cells; H. pylori cag⁺ strains are more potent than cag strains in inducing MAP kinase activation. Activation of MAP kinases has been suggested to play a role in the H. pylori-induced expression of IL-8 in gastric epithelial cells, and the differential activation of MAP kinases is a possible mechanism for the strain-specific variations in the outcome of gastric H. pylori infection (Keates et al., 1999). It should be noted that inhibitors of MAP kinases (PD98059; MAP kinase kinase inhibitor and SB203580; p38 inhibitor) did not prevent the H. pylori-induced degradation of IB- or NF-B activation in a previous study (Keates et al., 1999). This is in disagreement with other studies demonstrating cross talk between the MAP kinase and NF-B pathways. For example, it has been reported that MAP kinase kinase kinase and NF-B-inducing kinase can each directly activate IB kinase, resulting in IB phosphorylation and subsequent activation of NF-B (Lee et al., 1998; Nemoto et al., 1998). Because MAP kinase and NF-B pathways may exert independent regulatory effects on gastric epithelial cell IL-8 production following H. pylori infection, the inhibition of the H. pylori- or C₂-ceramide-induced activation of MAP kinases by rebamipide suggests another mechanism by which rebamipide inhibits the H. pylori-induced IL-8 expression. Because activation of MAP kinases may be dependent on the production of oxygen free radicals (Adler et al., 1995), it seems likely that rebamipide inhibits the activation of MAP kinases through its oxygen radical-scavenging effect.

In summary, we have shown that rebamipide inhibited H. pylori-dependent ceramide generation and the subsequent expression of IL-8 in Kato III cells. Decreased activation of NF-B and MAP kinases may contribute to the inhibition of IL-8 expression. Our results suggest a novel mechanism by which rebamipide may protect against the mucosal inflammation associated with H. pylori infection.