Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Reactive oxygen species are known to be produced by inflammatory cells such as neutrophils and macrophages in the case of bacterial infection and they function to attack invading bacterial pathogens. In addition to this beneficial function, reactive oxygen species are associated with tissue damage because the infiltrating cells cannot distinguish between bacterial pathogens and host tissues and they produce a large amount of oxygen radicals around them.

Rebamipide {(±)-2-(4-chlorobenzoylamino)-3-[2(1H)-quinolinon-4-yl] propionic acid}, an antigastritis and antigastric ulcer drug has anti-inflammatory action, inhibiting oxygen radical production by neutrophils induced by fMLP, opsonized zymosan, or Helicobacter pylori, and it thereby prevents injury of neighboring cells caused by oxygen radicals (Ogino et al., 1992; Yoshikawa et al., 1993b; Suzuki et al., 1994; Yoshida et al., 1996; Danielsson and Jurstrand, 1998). In addition, this drug inhibits release of granulocytes elastase, exocytosis of secretory granules, and chemotaxis by fMLP-stimulated human neutrophils (Murakami et al., 1998; Kobayashi et al., 2000). Rebamipide also has the ability to scavenge hydroxy radicals and it protects tissues from lipid peroxidation (Yoshikawa et al., 1993a,b).

Two intracellular signal transduction pathways are reported to participate in the receptor agonist-induced O₂ production by activated neutrophils (Dewald et al., 1988). One is the classical pathway involving an increase in the intracellular concentration of free calcium ([Ca²⁺]_i) as a result of activation of phospholipase C and accumulation of inositol 1,4,5-trisphosphate. The depletion of extracellular Ca²⁺ results in the attenuation of O₂ production (Dewald et al., 1988). The other is a pathway involving PI 3-kinase, which phosphorylates the D-3 position of the inositol ring of phosphoinositides (Whitman et al., 1988) and it is activated by heterotrimeric G protein -subunits in the case of activation of G protein-coupled receptors such as the fMLP-receptor (Okada et al., 1996; Kurosu et al., 1997; Stephens et al., 1997) and PIP₃ is mainly produced as a lipid second messenger (Traynor-Kaplan et al., 1988). The participation of PI 3-kinase in O₂ production was demonstrated in studies using wortmannin, a specific inhibitor of PI 3-kinase that is effective to completely abolish O₂ production by activated neutrophils (Baggiolini et al., 1987; Dewald et al., 1988; Okada et al., 1994).

Rebamipide has been shown to inhibit the increase in [Ca²⁺]_i when human neutrophils are stimulated by fMLP (Murakami et al., 1998). The effect of rebamipide on the PI 3-kinase pathway, however, has not been examined. In this study, therefore, we investigated the effect of rebamipide on the production of PIP₃ in fMLP-stimulated human neutrophils and its effect on the enzyme activity of PI 3-kinase fractions immunoprecipitated by anti-PI 3-kinase p85 subunit antibodies in a cell-free system.

Also, rebamipide has been shown to inhibit fMLP binding to its receptor, using intact rabbit neutrophils (Kim and Hong, 1997). They indicated that the number of binding sites for fMLP in intact neutrophils decreased as a result of rebamipide treatment in vitro and ex vivo. In the present study, we used a crude membrane fraction derived from human neutrophils to further examine the direct action of rebamipide on the interaction between fMLP and membrane receptors.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Materials were obtained from the following sources: rebamipide from Otsuka Pharmaceutical Co., Ltd. (Tokushima, Japan); wortmannin, fMLP, N-t-butoxycarbonyl-methionyl-leucyl-phenylalanine (Boc-MLP), and 2-methyl-6-[p-methoxyphenyl]-3,7-dihydroimidazo[1,2-a]pyrazine-3-one (MCLA) from Sigma Chemical Co. (St. Louis, MO); L--phosphatidylinositol (PI) and L--phosphatidylserine from Avanti Polar Lipids, Inc. (Alabaster, AL); and ³²Pi (150 mCi/ml), [-³²P]ATP (10 mCi/ml), and formyl-L-methionyl-L-leucyl-L-phenylalanine, N-[phenylalanine-ring-3,4,5-³H(N)]- (1 mCi/ml) from NEN Life Science Products, Inc. (Boston, MA). All other reagents from commercial sources were of analytical grade.

Neutrophils. Neutrophils were isolated from 30 ml of heparinized venous blood of healthy volunteers. Blood was mixed with 20 ml of 2% dextran saline containing 2.5 mM EGTA (pH 7.4) in a plastic syringe, and left to sit vertically for 30 min at room temperature. The resultant upper phase was taken and placed on 7.5 ml of Ficoll-Paque Plus containing 2.5 mM EGTA (pH 7.4). By centrifugation at 400g for 30 min at 20°C, neutrophils were sedimented at the bottom of the tube. After removal of the upper phase and interface, contaminating erythrocytes were lysed by mixing the sample with 25 ml of 0.2% NaCl for 30 s, and then 25 ml of 1.6% NaCl was added. After centrifugation at 200g for 5 min, the cells were washed twice with Ca²⁺-free HEPES-buffered medium [10 mM HEPES/NaOH (pH 7.4), 136 mM NaCl, 4.9 mM KCl, 5.5 mM glucose].

Superoxide Production. Neutrophils were suspended in Ca²⁺-free HEPES-buffered medium. Unless otherwise indicated, aliquots (5 × 10⁵ cells) of the cells were incubated with 0.5 µM MCLA in the presence or absence of rebamipide in a final volume of 2.0 ml of regular Krebs-Ringer-HEPES medium [134 mM NaCl, 4.7 mM KCl, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 2.5 mM CaCl₂, 5 mM glucose, 20 mM HEPES/NaOH (pH 7.4)] at 37°C for 10 min. Rebamipide was directly dissolved in the incubation medium. The reaction was started by addition of fMLP at the indicated concentration and the peak value of MCLA-derived chemiluminescence was regarded as indicative of the level of O₂ production. The intensity of control (0.1 µM fMLP) was 363 ± 111 (mean ± S.E.) × 10³ counts/min.

PIP₃ Production in Intact Neutrophils. PIP₃ production in neutrophils was estimated by the method described by Traynor-Kaplan et al. (1989). Neutrophils were suspended at a density of 10⁸ cells/ml in Ca²⁺-free HEPES-buffered medium. After incubation with ³²Pi (150 µCi/ml) at 30°C for 30 min, the cells were washed twice and resuspended at 2 × 10⁸ cells/ml. Aliquots (8 × 10⁶ cells) of the cells were incubated with or without rebamipide in a final volume of 400 µl of regular Krebs-Ringer-HEPES medium at 37°C for 10 min before stimulation with 0.1 µM fMLP for 30 s. Rebamipide was directly dissolved in the incubation medium. The reaction was terminated by the addition of 1.77 ml of chloroform/methanol/8% HClO₄ (50:100:5), followed by vigorous stirring. Thereafter, 570 µl of chloroform and 570 µl of 8% HClO₄ were added to the mixture, the organic phase was subsequently recovered, washed with chloroform-saturated 1% HClO₄ containing 0.5 M NaCl, and dried. The lipids were dissolved in 40 µl of chloroform/methanol (95:5). After brief centrifugation, aliquots were spotted on a thin layer plate (silica gel 60; Merck KGaA, Darmstadt, Germany), which had been impregnated with potassium oxalate, through a procedure involving development in a solvent system of 1.2% potassium oxalate-containing methanol/water (2:3), and activated by heating at 110°C for 20 min before spotting. The plate was developed in chloroform/acetone/methanol/acetic acid/H₂O (70:50:20:20:20), dried, and radioactivity was visualized by means of a Fuji BAS2500 bioimaging analyzer.

PI 3-Kinase Activity Using p85-Immunoprecipitated Fractions. Neutrophils (8 × 10⁶ cells) were incubated with 0.1 µM fMLP at 37°C for 30 s and lysed by addition of an equal volume of 2-fold concentrated RIPA buffer [final 50 mM Tris/HCl (pH 7.4), 1% Nonidet P-40, 0.25% sodium deoxycholate, 5 mM EDTA, 1 mM sodium orthovanadate, 30 mM NaF, 0.1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride, 1 mM EGTA, 25 kallikrein inhibitor units/ml aprotinin, 10 µg/ml leupeptin, 0.5% bovine serum albumin, and 1 mM dithiothreitol]. After centrifugation (15,000g, 20 min, 4°C), the supernatant was cleared with protein G Sepharose at 4°C for 60 min. The cleared supernatant was incubated with anti-PI 3-kinase p85 subunit rabbit antiserum (Upstate Biotechnology, Lake Placid, NY) at 4°C overnight. The immune complex was precipitated by addition of protein G Sepharose, and washed twice with RIPA buffer, twice with buffer A [40 mM Tris/HCl (pH 7.4), 1 mM dithiothreitol, and 0.5 M LiCl], twice with buffer B [40 mM Tris/HCl (pH 7.4), 1 mM dithiothreitol, and 100 mM NaCl], and once with buffer C [40 mM Tris/HCl (pH 7.4), 0.5 mM EGTA]. Immunoprecipitates were suspended in buffer C and used as the enzyme source in the PI 3-kinase assay. The reaction mixture in the PI 3-kinase assay consisted of the enzyme source, 40 mM Tris/HCl (pH 7.4), 0.5 mM EGTA, 0.8 mM PI, 0.8 mM L--phosphatidylserine, 10 mM MgCl₂, 800 µM ATP, 5 µCi of [-³²P]ATP, and rebamipide in a final volume of 100 µl. Rebamipide was dissolved in dimethyl sulfoxide and added to assay mixtures. The final concentration of dimethyl sulfoxide was 0.5% in all treatments. The reaction was allowed to proceed at 37°C for 5 min before termination by the addition of 20 µl of 8% HClO₄ and 450 µl of chloroform/methanol (1:2). After vigorous stirring, 150 µl of chloroform and 150 µl of 8% HClO₄ were added to the mixture, the organic phase was subsequently recovered, washed with chloroform-saturated 1% HClO₄ containing 0.5 M NaCl, and dried. The lipids were dissolved in 40 µl of chloroform/methanol (95:5). After brief centrifugation, aliquots were spotted on a thin layer plate (silica gel 60; Merck KGaA) that had been activated by heating at 110°C for 20 min before spotting. The plate was developed in chloroform/methanol/28% NH₄OH/H₂O (70:100:25:15), dried, and radioactivity was visualized using a Fuji BAS2500 bioimaging analyzer.

Preparation of Crude Neutrophil Membrane. Neutrophils were suspended in sucrose buffer [340 mM sucrose, 5 mM HEPES/NaOH (pH 7.5), 1 mM MgCl₂, 200 µM 4-(2-aminoethyl)benzenesulfonyl fluoride, 20 µM leupeptin, 50 kallikrein inhibitor units/ml aprotinin, 2.5 mM EDTA] at a concentration of 2 × 10⁷ cells/ml and disrupted with a Bioruptor (Olympus Optical, Co., Ltd., Tokyo, Japan). Following disruption, the suspension was centrifuged at 1000g for 10 min at 4°C to pellet the nuclei and unbroken cells. The resultant supernatant was further centrifuged at 20,000g for 30 min at 4°C to obtain the crude membrane fraction. The crude membrane fraction, including plasma membrane and all granules, was suspended in PBS() and stored at 80°C. Membrane protein was determined by the method of Bradford using bovine albumin as the standard (Bradford, 1976).

[³H]fMLP Binding to Neutrophil Membrane. Membrane protein (50-60 µg) was incubated at 25°C for 60 min in 200 µl of PBS() containing [³H]fMLP in the presence or absence of rebamipide at the indicated concentrations. Rebamipide was dissolved in dimethyl sulfoxide and added to assay mixtures. The final concentration of dimethyl sulfoxide was 0.5% in all treatments. The reaction was terminated by filtration of the mixture through a Whatman GF/B glass filter followed by four washes with PBS(). Radioactivity retained on the filter was measured using a liquid scintillation counter. Specific binding was calculated by subtracting the amount of [³H]fMLP bound in the presence of 1 mM fMLP from the total [³H]fMLP bound.

Statistical Analysis. Data were expressed as means ± S.D. Statistical analyses, including ANOVA (Figs. 1, 2, 4-6, 8, and 9), the two-tailed t test (Figs. 1 and 2), and regression analysis (Figs. 1, 2, 5, and 7), were performed using SAS (R6.12; SAS Institute Japan, Ltd., Tokyo, Japan). P < 0.05 was considered significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Rebamipide Inhibits fMLP-Induced O₂ Production in Human Neutrophils. Stimulation of human neutrophils with 0.1 µM fMLP induced rapid production of O₂ within 60 s. Treatment of the cells with rebamipide attenuated the fMLP-induced O₂ production in a dose-dependent manner. The IC₅₀ value [95% confidence interval (CI)] for the inhibitory action of rebamipide was 0.20 (0.17-0.23) mM. A PI 3-kinase inhibitor, wortmannin, completely abolished the fMLP-induced O₂ production in human neutrophils (Fig. 1). The inhibitory effect of rebamipide on O₂ production was also seen in fMLP-stimulated human monocytes (C. Nagano, unpublished data).

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Inhibitory effect of rebamipide on fMLP-induced O2 production in human neutrophils. Neutrophils were pretreated with the indicated concentrations of rebamipide or 1 µM wortmannin (Wort) at 37°C for 10 min and stimulated with 0.1 µM fMLP. The peak value of MCLA-derived chemiluminescence was regarded as indicative of the level of O2 production and is expressed as percentage of control. Data are shown as means ± S.D., N = 3. **P < 0.01 versus no treatment by one-way ANOVA followed by two-tailed Dunnett's test. P < 0.01 versus no treatment by two-tailed t test for unpaired data.

Rebamipide Inhibits fMLP-Stimulated PIP₃ Production in Human Neutrophils. Activation of PI 3-kinase is responsible for fMLP-induced O₂ production as demonstrated by the result obtained using wortmannin (Fig. 1). When ³²P-labeled neutrophils were stimulated with 0.1 µM fMLP, production of labeled PIP₃, the product of PI 3-kinase was observed within 30 s. Treatment of the cells with rebamipide prevented the accumulation of PIP₃ in a dose-dependent manner (Fig. 2, A and B). The IC₅₀ value (95% CI) for the inhibitory action of rebamipide on PIP₃ production was 0.31 (0.20-0.48) mM.

View larger version (53K): [in this window] [in a new window]   Fig. 2.   Inhibitory effect of rebamipide on fMLP-induced PIP3 production in human neutrophils. 32Pi-labeled neutrophils were treated with the indicated concentrations of rebamipide or 1 µM wortmannin (Wort) at 37°C for 10 min and stimulated with 0.1 µM fMLP for 30 s. Phospholipids were extracted and separated on a thin layer chromatography plate. A, autoradiogram of a thin layer chromatography plate is shown and the position of PIP3 is indicated by an arrow. (), no stimulation. B, PIP3 production calculated from autoradiograms is expressed as percentage of control. Data are shown as means ± S.D., N = 3. *P < 0.05, **P < 0.01 versus no treatment by one-way ANOVA followed by two-tailed Dunnett's test. P < 0.01 versus no treatment by two-tailed t test for unpaired data.

Rebamipide Does Not Inhibit PI 3-Kinase Activity in Anti-p85 Immunoprecipitates. To investigate whether the effect of rebamipide was due to a direct inhibitory effect on PI 3-kinase activity or its effect on a factor upstream in the signal transduction pathway, the activity of PI 3-kinase was examined in the presence of rebamipide. The anti-PI 3-kinase p85 subunit immunoprecipitates were used as an enzyme source in the PI 3-kinase assay in a cell-free system. ³²P-labeled phosphatidylinositol 3-phosphate [PI(3)P] accumulated when PI was used as the substrate. The PI 3-kinase inhibitor wortmannin completely abolished the accumulation of PI(3)P, whereas rebamipide showed no effect on the activity of PI 3-kinase (Fig. 3).

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Effect of rebamipide on the PI 3-kinase activity of anti-p85 immunoprecipitates. Neutrophils were stimulated with 0.1 µM fMLP and lysed in RIPA buffer. The lysates of the cells were subjected to immunoprecipitation with anti-p85 antibody. PI 3-kinase activity associated with the precipitates was measured in the absence or presence of the indicated concentrations of rebamipide or 1 µM wortmannin (Wort) as described under Experimental Procedures. An autoradiogram of a thin layer chromatography plate is shown and the position of PI(3)P is indicated by an arrow. The results were reproduced in three separate experiments.

Rebamipide Does Not Inhibit NaF-Stimulated O₂ Production in Human Neutrophils. A G protein activator, NaF, induced a moderate level of production of O₂ by human neutrophils. Wortmannin treatment completely abolished the NaF-induced production of O₂, whereas rebamipide showed no inhibitory effect (Fig. 4). This result indicates that the inhibition of fMLP-induced O₂ production by rebamipide is due to its inhibitory action on a component of the signal transduction pathway upstream of G protein.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Effect of rebamipide on NaF-induced O2 production in human neutrophils. Neutrophils were pretreated with 1 mM rebamipide or 1 µM wortmannin at 37°C for 4.5 min and stimulated with 10 mM NaF. The peak value of MCLA-derived chemiluminescence was regarded as indicative of the level of O2 production and is expressed as percentage of control. Data are shown as means ± S.D., N = 3. **P < 0.01 by one-way ANOVA followed by two-tailed Dunnett's test. N.S., not significant.

Rebamipide Inhibits fMLP-Receptor Binding. A crude membrane fraction of human neutrophils was used to investigate the effect of rebamipide on the interaction between fMLP and its receptor. [³H]fMLP binding to the neutrophil membrane was inhibited by rebamipide in a dose-dependent manner. The IC₅₀ value (95% CI) for the inhibitory action of rebamipide was 0.18 (0.15-0.20) mM (Fig. 5).

View larger version (44K): [in this window] [in a new window]   Fig. 5.   Inhibitory effect of rebamipide on [3H]fMLP binding. Crude neutrophil membrane was incubated with 100 nM [3H]fMLP in the presence or absence of the indicated concentrations of rebamipide at 25°C for 60 min. Radioactivity specifically bound is shown as the mean percentage of control ± S.D., N = 3. **P < 0.01 versus no rebamipide by one-way ANOVA followed by two-tailed Dunnett's test.

Rebamipide Increases the K_D Value of [³H]fMLP in Scatchard Analysis. To investigate the mechanism of action of rebamipide in inhibiting fMLP binding to its receptor, Scatchard analysis was performed. As shown in Fig. 6, rebamipide increased the K_D value of [³H]fMLP without altering the number of [³H]fMLP binding sites. The K_D values of [³H]fMLP in the presence of 0.1 and 0.3 mM rebamipide were 49.7 ± 3.3 and 105.8 ± 13.4 nM, respectively, versus the control value of 27.8 ± 4.9 nM. The corresponding densities of [³H]fMLP binding sites were 0.74 ± 0.04, 0.66 ± 0.07, and 0.81 ± 0.08 pmol/mg of protein.

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Scatchard plots for [3H]fMLP binding. Crude neutrophil membrane was incubated with [3H]fMLP (1.56-100 nM) in the presence or absence of the indicated concentrations of rebamipide at 25°C for 60 min. The bound-to-free ratio was plotted against the bound ligand concentration. Data are shown as means ± S.D., N = 3. **P < 0.01 versus control by one-way ANOVA followed by two-tailed Dunnett's test.

Rebamipide Shifts the Dose-Response Curve for fMLP-Induced O₂ Production to the Right. To confirm the mode of action of rebamipide, the effect of rebamipide on the dose-response curve for fMLP-induced O₂ production was determined. fMLP induced O₂ production in a dose-dependent manner and the estimated EC₅₀ value (95% CI) was 0.032 (0.027-0.038) µM. Rebamipide caused a parallel shift in the fMLP dose-response curve to the right. Additionally, the inhibition by rebamipide of the maximal response to fMLP was surmountable by applying a higher concentration of fMLP. The EC₅₀ value (95% CI) of fMLP in the presence of 0.3 mM rebamipide was 0.36 (0.33-0.40) µM (Fig. 7).

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Effect of rebamipide on the dose-response curve for fMLP-induced O2 production. Neutrophils were pretreated with vehicle or 0.3 mM rebamipide at 37°C for 10 min and stimulated with fMLP at the indicated concentrations. The peak value of MCLA-derived chemiluminescence was regarded as indicative of the level of O2 production and is expressed as percentage of the maximal response of the control. Data are shown as means ± S.D., N = 3.

Washing Out of Rebamipide Abolishes the Inhibition of fMLP-Induced O₂ Production. To examine the reversibility of the inhibitory action of rebamipide, neutrophils were treated with rebamipide for 10 min and then rebamipide was washed out before stimulation with fMLP. Figure 8 shows that the fMLP-induced O₂ production recovered after rebamipide had been washed out. This result indicates that the inhibitory action of rebamipide is reversible.

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Washing out of rebamipide abolishes the inhibition of fMLP-induced O2 production. Neutrophils were pretreated with 1 mM rebamipide for 10 min, and then rebamipide was washed out with regular Krebs-Ringer-HEPES medium before stimulation with 0.1 µM fMLP. In the case of the control and rebamipide groups, the cells were pretreated, washed, and resuspended in the control and rebamipide medium, respectively. The peak value of MCLA-derived chemiluminescence was regarded as indicative of the level of O2 production and is expressed as percentage of control. Data are shown as means ± S.D., N = 3. *P < 0.05 by one-way ANOVA followed by Tukey's test.

Rebamipide Action Is Similar to the Action of a fMLP Antagonist. A further experiment was performed to compare the effect of rebamipide with that of a fMLP antagonist, Boc-MLP, on fMLP-induced O₂ production, with additional time of exposure to the drugs. Treatment of the cells with Boc-MLP or rebamipide was started 30 min before, 0, 5, or 10 s after fMLP stimulation, and then O₂ production was measured. As shown in Fig. 9, similar inhibitory effects were observed with rebamipide and Boc-MLP. When treatment was started 30 min before fMLP stimulation, no greater effect was observed in comparison with simultaneous addition of the drug. When treatment was started 5 s after fMLP stimulation, O₂ production was partially inhibited, and when treatment was started 10 s after, there was almost no inhibitory effect. This result indicates that the inhibitory action of rebamipide is similar to that of an fMLP antagonist.

View larger version (44K): [in this window] [in a new window]   Fig. 9.   Effects of additional time of exposure to an fMLP antagonist and rebamipide on fMLP-induced O2 production. Treatment of neutrophils with vehicle, 7 µM Boc-MLP, or 0.3 mM rebamipide was started 30 min before (30 m), 0, 5, or 10 s after (+5 s, +10 s) the initiation of fMLP stimulation (0.1 µM), and then O2 production was measured and it is expressed as percentage of control. Data are shown as means ± S.D., N = 3. **P < 0.01 versus time 0 of the rebamipide group, P < 0.01 versus time 0 of the Boc-MLP group by one-way ANOVA followed by two-tailed Dunnett's test. N.S.1, not significant versus time 0 of the rebamipide group. N.S.2, not significant versus time 0 of the Boc-MLP group.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

In this study, we investigated the mechanism of action of rebamipide in inhibiting the fMLP-induced O₂ production of human neutrophils.

Two intracellular signal transduction pathways are reported to participate in agonist-induced O₂ production. One is a Ca²⁺-dependent, wortmannin-insensitive pathway that leads to an increase in [Ca²⁺]_i, and the other is a Ca²⁺-independent and wortmannin-sensitive pathway that leads to activation of PI 3-kinase. The observation that O₂ production is attenuated in the absence of extracellular Ca²⁺ or in the presence of PI 3-kinase inhibitors indicates that both pathways are functional in agonist-induced O₂ production (Dewald et al., 1988).

Rebamipide is reported to inhibit the increase in [Ca²⁺]_i when human neutrophils are stimulated by fMLP (Murakami et al., 1998). The effects of rebamipide on the PI 3-kinase pathway remain to be investigated. We therefore examined the effect of rebamipide on the level of PIP₃ produced upon activation of PI 3-kinase when human neutrophils were stimulated by fMLP. In parallel with the inhibition of O₂ production, rebamipide showed an inhibitory effect on PIP₃ production induced by fMLP. We next examined the effect of rebamipide on the activity of PI 3-kinase using anti-PI 3-kinase p85 immunoprecipitates as the enzyme source, to investigate whether rebamipide directly affects PI 3-kinase activity or whether it inhibits a component further upstream in the signal transduction pathway. Figure 3 shows that rebamipide does not have a direct inhibitory effect on the activity of PI 3-kinase.

In the signal transduction pathway for fMLP-induced O₂ production, it is reported that G protein is located upstream of PI 3-kinase and that the G protein -subunits activate PI 3-kinase (Okada et al., 1996; Kurosu et al., 1997; Stephens et al., 1997). We next investigated the effect of rebamipide on O₂ production caused by direct activation of G protein. NaF is reported to directly activate G protein without affecting receptors and it induces O₂ production in neutrophils (Ross and Gilman, 1980; Strnad and Wong, 1985). The PI 3-kinase inhibitor wortmannin abolished the NaF-induced O₂ production, whereas rebamipide showed no inhibitory effect. These results suggest that the target of rebamipide may be located upstream of G protein in the signal transduction pathway.

We therefore investigated the effect of rebamipide on fMLP binding to its receptor. Figure 5 shows that rebamipide inhibited [³H]fMLP binding to human neutrophil membrane in a dose-dependent manner. The IC₅₀ value for the inhibitory action of rebamipide on fMLP binding was 0.18 mM and this is consistent with the value for its inhibitory action on O₂ production or PIP₃ production. The results obtained in a study focusing on O₂ production using an fMLP antagonist, Boc-MLP, support the view that rebamipide has a similar antagonistic action (Fig. 9).

Scatchard analysis was performed to investigate the mechanism of action of rebamipide in inhibiting fMLP-receptor binding. In the presence of rebamipide, the K_D value was increased in a dose-dependent manner, with little change in the B_max value. This result indicates that rebamipide decreased the apparent affinity of [³H]fMLP to its receptor without altering the number of [³H]fMLP binding sites. From this result, it is suggested that rebamipide acts as a competitive antagonist of the fMLP-receptor.

The competitive antagonistic action of rebamipide was also demonstrated in the O₂ production study. In the fMLP concentration-dependent response curve, rebamipide at 0.3 mM caused a parallel shift, and the inhibition of maximal fMLP-induced O₂ production was surmountable by applying a higher concentration of fMLP. A preliminary study of O₂ production using various concentrations of rebamipide further confirmed the competitive antagonistic action of rebamipide, with the slope of the Schild plot being 1. The inhibitory effects of competitive antagonists are generally known to be reversible. The inhibitory action of rebamipide was suggested to be reversible because washing out of rebamipide abolished the inhibition of O₂ production. From the results of the Scatchard analysis and the results concerning O₂ production, we confirmed the competitive antagonistic action of rebamipide on the fMLP-receptor.

Kim and Hong (1997) have also reported that rebamipide inhibits [³H]fMLP binding to rabbit neutrophils. There, however, are some differences between the results of their study and ours. They indicated that the number of binding sites for fMLP decreased as a result of rebamipide treatment in vitro and ex vivo, whereas our present study indicates that the number of binding sites is not affected by rebamipide. These contradictory results seem to be due to the difference in receptor samples used in obtaining the data for the Scatchard analysis; they used intact neutrophils, whereas we used its crude membrane. In the case of intact neutrophils, many physiological events that affect receptor binding may occur at 37°C, and, especially in the case of fMLP-receptors, it is reported that the receptors are stored on the membrane of intracellular granules called secretory vesicles and other granules, and are translocated to the cell surface upon activation of neutrophils (Sengeløv et al., 1994; Borregaard and Cowland, 1997). In the presence of rebamipide, however, the translocation of the fMLP-receptors apparently does not occur as a result of the antagonistic action of rebamipide on the fMLP-receptors and the apparent number of binding sites is assumed to decrease compared with the number on control neutrophils. On the other hand, an increase in the K_D value was observed in both our study and theirs although the increase reported by them was small. In the present study, we were able to demonstrate the competitive antagonistic action of rebamipide on the fMLP-receptor by using a simpler sample than intact cells.

Sengeløv et al. (1994) have reported that a 25-kDa fMLP-binding protein (neutrophil gelatinase-associated lipocalin, NGAL) was identified in a specific granule of human neutrophils. Because our receptor sample was crude membrane fraction that contains all the granules, it was considered that our data of binding assay may reflect fMLP binding to NGAL. To confirm whether rebamipide binds upon fMLP-receptor, we performed the preliminary experiment of receptor binding assay using a human cloned fMLP-receptor sample (purchased from NEN Life Science Products, Inc.). Rebamipide inhibited [³H]fMLP binding to the receptor in a dose-dependent manner as well as the present result using crude membrane. This result confirms that rebamipide acts on the fMLP-receptor, although we couldn't completely rule out the effects of NGAL.

In the case of H. pylori infection, lots of neutrophils are known to infiltrate into the gastric mucosa and injure the host tissue. Mooney et al. (1991) reported that a substance that reacted with fMLP antibody was produced by H. pylori and it activated human neutrophils. Rebamipide has been shown to inhibit H. pylori-induced and H. pylori extract-induced production of oxygen radicals in human neutrophils and thereby prevent injury of neighboring cells (Suzuki et al., 1994; Yoshida et al., 1996; Danielsson and Jurstrand, 1998). Considering the results obtained in the present study, it seems possible that rebamipide may inhibit the binding of H. pylori-derived fMLP to the fMLP-receptor and suppress the activation of neutrophils by H. pylori.

In conclusion, the present study demonstrates that the competitive inhibitory action of rebamipide on fMLP binding to its receptor is involved in the inhibitory mechanisms of fMLP-induced neutrophil activation.