European Journal of Pharmacology, 212 (1992) 9-13
9
© 1992 Elsevier Science Publishers B.V. All rights reserved 0014-2999/92/$05.00
EJP 52298
Antiulcer mechanism of action of rebamipide, a novel antiulcer compound, on diethyldithiocarbamate-induced antral gastric ulcers in rats Keiki O g i n o a, T a t s u y a H o b a r a
a,
H i r o n o b u Ishiyama a, K a t s u y a Y a m a s a k i a, H a r u o Kobayashi ~, Yukie Izumi ~ a n d Shinji O k a b
Department of Public Health, Yamaguchi University, School of Medicine, 1144 Kogushi, Ube 755, Japan and b Kokura Memorial Hospital, Kitakyushu, Japan
Received 22 October 1991, revisedMS received25 November 1991, accepted 3 December 1991
The mechanism of the inhibitory action of rcbamipidc, a new mucosal protective drug, was studied using rats with diethyldithiocarbamate-induced gastric antral ulcers. Rebamipide reduced ulcer formation and inhibited the elevation in lipid peroxide concentration in the gastric mucosa. Rebamipide inhibited both iuminol- and lucigenin-dependent chemiluminescence of neutrophils activated by formyl-methionyl-leucyl-phenylalanine. Rebamipide did not alter the reduction of cytochrome c induced by the xanthine-xanthine oxidase system or the NADPH-dependent micirosomal lipid peroxidation in the liver. These findings suggest that rebamipide prevents diethyldithiocarbamate-induced gastric ulcer formation by inhibiting neutrophil activation. Diethyldithiocarbamate-induced antral ulcer; Neutrophil; Rebamipide
1. Introduction
It has been suggested that active oxygen species may be involved in the pathogenesis of various gastric mucosa injuries, and that some radical scavengers show a protective effect against the mucosal injury induced by active oxygen species (Perry et al., 1986; Pihan et al., 1987; Szelenyi and Brune, 1988). Of the various radical scavengers, we selected superoxide dismutase (SOD) for our study, because exogenous SOD has been reported to prevent mucosal injury caused by active oxygen species (Parks et al., 1982; Itoh and Guth, 1985; Perry et al., 1986; Pihan et al., 1987). During our investigations on SOD, we found that treatment with diethyldithiocarbamate, an inhibitor of copper, zincSOD, produced gastric mucosal injury (Ogino et al., 1987, 1988), and that the production of active oxygen species together with decreased SOD activity and ischemia caused the damage (Ogino et al., 1990). The multiple gastric lesions produced by diethyldithiocarbamate are only shallow and do not reach the muscularis mucosa histologically, and diethyldithiocarbamate treatment fails to cause ulceration in some rats. However, by performing pyloric ligation and by applying 0.1
N HCI intragastrically after treatment with diethyldithiocarbamate, we were able to produce a large ulcer that extended into the antral muscular layer (Ishiyama et al., 1989). Previous studies have indicated that active oxygen species and activated neutrophils are involved in the pathogenesis of this experimentally induced ulcer (Oka et al., 1990; Ogino et al., 1991a). Rebamipide (2-(4-chlorobenzoylamino)-3-[2-(1H)quinolinon-4-yl]propionic acid) is a novel antiulcer agent synthesized by Otsuka Pharmaceutical Co., Japan. Rebamipide increases gastric mucosa prostaglandin concentrations and protects the mucosa from damage induced by various necrotizing agents (Yamasaki et al., 1987). Wc have previously reported that rebamipide may act as a new type of mucosal protective drug by maintaining gastric mucosal SOD activity (Oka et al., 1991). In this study, the mechanism of the antiulcer effect of rebamipide on diethyldithiocarbamate-induced antral ulcers was investigated with regard to the role of active oxygen species.
2. Materials and m e t h o d s
2.1. Animals Correspondence to: K. Ogino, Department of Public Health, Yamaguchi UniversitySchool of Medicine, 1144Kogushi,Ube 755, Japan.
Male Wistar rats weighing 200-230 g were used. The animals were kept in stainless-steel cages with
ll)
wide mesh wire bottoms to prevcnt coprophagia in an air-conditioned room kept at 23 _+ 2°C with artificial lighting from 6 a.m. to 6 p.m. The animals were fasted and allowed access to water ad libitum for 24 h prior to the experiment.
2.2. Chemicals The following chemicals were purchased from the sources indicated: diethyidithiocarbamatc (Wako Pure Chemicals), xanthine oxidase (Boehringer Mannheim), N-l-naphthylethylene-diamine (Tokyo Kasei), xanthine (Sigma), cytochrome c (Sigma), Hanks balanced salt solution (HBSS) (Bioproducts), Dulbecco's phosphatebuffered saline (Bioproducts), formyl-methionyl-leucylphenylalanine (Sigma), lucigenin (Sigma), luminol (Nakalai), 2-thiobarbituric acid (Merck) and rebamipidc (Otsuka Pharmaceutical).
2.3. Production of diethyldithiocarbamate-induced antral ulcer (Ishiyarna et al., 1989) The rats were anesthetized with ether, a midline laparotomy was performed and the pylorus was ligated. Diethyldithiocarbamate dissolved in saline was given s.c. at a dose of 800 mg/kg, and a 1-ml oral dose of 0.1 N HCI was then given to each rat. The rats were killed 5 h later. The stomach was removed and filled with 10 ml of 10% formalin. The ulcer index was measured under a dissecting microscope with a square grid eyescope and is expressed as the area of the antral ulcer (mm2).
2.4. Effect of rebamipide on diethyldithiocarbamate-induced antral ulcer Rebamipide was suspended in 0.5% carboxymethylcellulose and administered i.p. 30 rain before the experiment.
2.5. Effect of rebamipide on the lipid peroxide concentration in antral mucosa The rats were killed 5 h after pylorus ligation, diethyldithiocarbamate treatment, and 0.1 N HCI administration because the lipid peroxide concentration in the antral mucosa is highest within 5 h (Ishiyama et al., 1989). The stomach was removed and the antral mucosa was scraped off. The mucosa was homogenized with 1.15% KCI and the level of thiobarbituric acid reactants was measured (Ohkawa et al., 1979).
2.6. Effect of rebamipide on chemiluminescence of activated neutrophil-rich exudate The rats were injected i.p. with 20 ml of sterile 12% (w/v) sodium caseinate in iso-osmotic (0.9%) NaCI
(Newsby, 1980). Twenty hours later, thc animals were killed by ether asphyxiation, the pcritoncal cavity was opcned and the peritoneal exudatc was collccted. After being filtered through three layers of surgical gauze, the cxudatc was centrifuged at 200 × g for 5 min, and the pcllet was washcd twicc with Dulbccco's phosphate-buffered saline. Thc pellet was suspended in 1 ml of iso-osmotic (0.9%) NaCI, 10 ml of distilled water, and 10 ml of 1.8% NaCI to lead to hypotonic lysis of erythrocyte contaminants. Thc cell suspension was centrifuged as before and the pellet was resuspended in HBSS with 1% bovine serum albumin. Thc specificity of the cell population was determined by differential counting of smears stained with Wright's stain. Viability of the neutrophils was measured with the trypan blue exclusion technique. Chemiluminescence was measured using luminol or lucigenin with the Aloka Co. luminescence rcader at 37°C. Reaction mixtures contained 1.5 × l06 granulocytes, 1 tzM FMLP ( < 0.01% DMSO), 0-1 mM rebamipide (0.001 N NaOH), and 0.07 mM luminol or lucigenin (0.05 M Tris HCi,pH 7.4) in I ml of continuously stirred HBSS. Luminol was dissolved in mildly alkaline water and then adjusted to pH 7.4 with 0.1 N HCI. Rebamipide was diluted with 0.001 N NaOH. Control studies were performed by the addition of 0.001 N NaOH alone. Thc results are exprcssed in terms of the maximal counts per min.
2. 7. Effect of rebamipide on cytochrome c reduction in the xanthine-xanthine oxidase system The standard assay for the measurement of SOD activity was used (McCord and Fridovich, 1969). The incubation mixture contained 0.01 mM cytochrome c, 0.05 mM xanthine, 0.04 U / m l xanthine oxidase, and 0-1 mM rebamipide in 3 ml of 50 mM potassium phosphate buffer (pH 7.8) containing 0.1 mM EDTA. The reduction of cytochrome c was followed at 550 nm.
2.8. Effect of rebamipide on NADPH-dependent microsomal lipid peroxidation in the rat liver Rat liver microsomes were prepared as described previously (Ogino et al., 1991b). Liver microsomes (1.5 mg p r o t e i n / m l ) were incubated with 0-1 mM rebamipide and a freshly prepared NADPH-generating system in 50 mM Tris HCI (pH 7.4). The NADPH-generating system contained 0.17 mM NADP, 3.3 mM glucose-6-phosphate, and 2 units of D-glucose-6-phosphate dehydrogenase. After a 30-rain incubation at 37°C, the lipid peroxide level was measured as the amount of malondialdehyde formed using the thiobarbituric acid method (Buege and Aust, 1975).
(A)
TABLE 1 Effect of i.p. rebamipide on the formation of diethyldithiocarbamateinduced antral ulcers. Data are m e a n s 5: S.E.
Treatment
Do~ (mg/kg)
Number of rats
Ulcer index (mm2)
Inhibition (%)
Vehicle Rebamipide Rebamipide Rebamipide
15 30 60
6 6 6 6
7.85:1.1 3.3+0.8 a 2.6 5:0.3 ~ 1.6 ± 0.6 ~
57.7 66.7 79.5
a p < 0.01 when compared with the vehicle group.
¢:: E
6Tli2
11
8
%
! 0
C
0.1 0.2 ~il~le
C
0.01
(B)
0.5 (raM)
1.0
"~20
g
2.9. Statistical analysis Statistical significance was determined with a oneway analysis of variance a m o n g multiple groups, and P < 0.05 was considered significant.
o 10 x
0
0.1
1.0
m 2.0
Rebamipide (mM)
Fig. 1. Effect of rebamipide on lucigenin-dependent (A) or luminoldependent (B) chemiluminescence in formyl-methionyl-leucyl-phenylalanine-stimulated neutrophil-rich exudate.
3. Results
3.1. Effect of rebamipide on ulcer formation Rebamipide inhibited ulcer formation in a dose-dep e n d e n t m a n n e r (table 1).
3.2. Effect of rebamipide on lipid peroxidation in antral mucosa T h e lipid peroxide level in the antral mucosa of untreated rats was 1.7 + 0.2 n m o l / m g protein. T h e diethyldithiocarbamate plus vehicle group showed an elevated lipid peroxide level, and treatment with rebamipide significantly reduced this in a dose-dependent m a n n e r (table 2).
3.3. Effect of rebamipide on chemiluminescence of neutrophil-rich exudate T h e specificity of the cell population was 90% and neutrophil viability was 99.8%. Rebamipide at 0.1-2.0 TABLE 2 Effect of rebamipide on the lipid peroxide concentration in the antral mucosa. Lipid peroxide is shown as the level of thiobarbituric acid reactants. Data are m e a n s + S.E.
Treatment
Number of rats
Thiobarbituric acid reactants (nmol/mg protein)
Normal antrum mucosa Diet hyldithiocarbamate +vehicle + rebamipide 15 mg/kg + rebamipide 30 mg/kg + rebamipide 60 mg/kg
6
1.7 + 0.2
6 6 6 6
10.15:2.6 a 4.6 + 1.3 3.1 + 1.2 c 15. :1:0.4 b
a p < 0.01 when compared with normal antrum mucosa, b p < 0.01, c p < 0.05 when compared with diethyldithiocarbamate plus vehicle.
m M inhibited l u m i n o l - d e p e n d e n t luminescence, while at 0.2-1.0 m M it inhibited lucigenin-dependent luminescence (fig. 1). Rebamipide had no effect on neutrophil viability in this study.
3.4. Effect of rebamipide on cytochrome c reduction by the xanthine-xanthine oxidase system T h e reduction rates (A O . D . / m i n ) were 0.025, 0.025, 0.024 and 0.023 at 550 nm for the control system and after the addition of 0.01, 0.1 and 1.0 m M rebamipide, respectively. T h e drug thus did not alter the reduction of cytochrome c induced by the production of superoxide radicals in the xanthine-xanthine oxidase system.
3.5. Effect of rebamipide on NADPH-dependent microsomal lipid peroxidation in the rat liver Malondialdehyde levels for controls and with 0.1 and 1.0 m M rebamipide were 14.0 + 0.5, 14.5 + 0.8 and 14.1 + 0.6 n m o i / m g protein, respectively. Each value represents the m e a n + S.E. of three determinations. Rebamipide did not inhibit hepatic N A D P H - d e p e n dent microsomal lipid peroxidation.
4. Discussion T h e mechanism underlying the antiulcer action of rebamipide was investigated in a rat diethyldithiocarbamate-induced antral ulcer model because this type of experimental ulcer is similar to h u m a n gastric ulcer (Ishiyama et al., 1989). HCI was applied to make a
12 deeper ulcer by decreasing mucosal blood flow and intramural pH through the back diffusion of luminal H* (Kivilaaksa et al., 1878; Hirose et al., 1985; Morishita and Guth, 1987), and because diethyldithiocarbamate itself inhibits acid secretion (Ogino ct al., 1990). We chose to administer rebamipide i.p. to induce ulcer formation and to study lipid peroxidation for thc following two reasons. First, intragastrie administration of a drug may induce adaptive cytoprotection. This is caused by dircct stimulation of the gastric mucosa by drugs that act as a mild irritant (Robert ct al., 1983) and is not a pharmacological effect. Second, this model involves HCI administration, so to prevent the drug from altering acidity we chose the i.p. route. Both gastric mucosa injury and the increase in lipid peroxidation were inhibited by treatment with rebamipide. Active oxygen species and lipid peroxidation are closely related to each other in the pathogenesis of diethyldithiocarbamate-induced antral ulcers because the SOD level decreases and the lipid peroxide concentration rises in thc gastric mucosa before injury (Ogino et al., 1989, 1990; Ishiyama el al., 1989), and because some radicial scavengers, e.g. SOD and catalase, inhibit ulcer formation and the decrease in mucosal SOD activity (Ishiyama et al., 1989; Oka et al., 1990, 1991). A correlation between active oxygen species and lipid peroxidation has been reported for experimental ulcers induced by ethanol, for aspirin-induced gastric mucosa injury (Pihan et al., 1987), and for ischemia reperfusion-induced injury (Yoshikawa et al., 1991). However, our study showed that rebamipide itself did not act as a superoxide radical scavenger or as an antioxidant. The source of active oxygen species in the diethyidithiocarbamate-induced antral ulcer model is not known. Xanthine oxidase is the first documented biological source of superoxide radicals and is considered to have an important role in supplying the active oxygen species involved in gastric mucosa injury (Parks et al., 1982; Perry et al., 1986). However, the reduction of cytochrome c by superoxide radicals in the xanthinexanthine oxidase system was not altered by the addition of rcbamipide. Activated neutrophils are another postulated source of active oxygen species (McCord, 1985). Neutrophil-derived free radical damage has been suggested to occur in the gastrointestinal tract (Grisham et al., 1986; Smith et al., 1987). Neutrophil depletion inhibits diethyldithiocarbamate-induced antral ulcer formation more strongly than the administration of radical scavengers, such as SOD and catalase (Ogino et al., 1991a). Luminol-dependent luminescence is said to be totally dependent on the myeloperoxidase (MPO)H 2 0 2 system (DeChatelet et al., 1982; Dahlgren and Stendahl, 1983). Lucigenin is thought to amplify the chemiluminescence response by an MPO-independent mechanism probably related to the production of su-
pcroxidc radicals by neutrophils (Williams and Cole, 1981; Dahlgren et al., 1985). Rcbamipide inhibited luminol-depcndent luminescence at concentrations of 0.1-2.0 mM and inhibited lucigcnin-dependent luminescence at concentrations of 0.2-1.0 mM. The mechanism by which rcbamipide inhibits chemiluminesccnce is unknown at present. However, our findings imply that rebamipide rcduces tissue damage caused by activated neutrophils by inhibiting the production of active oxygen species. The site of action of rebamipide may be the intravascular space for the following two reasons: (1) SOD, catalase, and neutrophil depletion inhibit ulcer formation in this model (Oka et al., 1990; Ogino et al., 1991a); and (2) activated ncutrophils attack venular endothelial cells via active oxygen species after adhering to the endothelium and produce microvascular or histological damage (Vissers et al., 1985; Sucmatu et al., 1989). However, it is not clear whether rebamipide influences neutrophil chemotaxis in the damaged mucosa and how many activated neutrophils are required to causc mucosal damage. Prostaglandin E~ and prostaglandin E 2 (10-5-10 7 M) inhibit tbrmyl-methionyl-leucyl-phenylalanine-induced superoxide radical production by human neutrophils (Fantonc et al., 1983; Bittner and Parnham, 1983). Rebamipide increases gastric mucosa prostaglandin levels (Yamasaki et al., 1987), so it is possible that superoxide radical production was influenced not only by rebamipide itself but also by an increase in mucosal prostaglandin levels. A previous study showed that rebamipide maintains gastric mucosa SOD activity (Oka et al., 1991). This activity may be decreased in our model because diethyldithiocarbamate directly inhibits SOD (Heikkila et al., 1976), or because hydrogen peroxide generated from superoxide radicals by SOD itself inhibits SOD activity (Hodgson and Friovich, 1975). The second reason appears more likely to explain the decrease in SOD activity in this experiment because administration of SOD and catalase prevent ulcer formation in our rat model (Ishiyama et al., 1989; Oka et al., 190; Ogino et al., 1991a). If activated neutrophils are a source of superoxide radicals, our finding that rebamipide inhibits the production of superoxidc radicals by activated neutrophils may support the previous report that rebamipide acts by maintaining SOD activity. This study raises the possibility that the antiulcer effect of rebamipide may be related to inhibition of the production of active oxygen species by activated neutrophils. However, further studies are required to establish the effectiveness of rebamipide in diethyldithiocarbamate-induced antral ulcers and to define further the mechanisms by which the production of active oxygen species leads to morphological and functional injury to the gastric mucosa.
13
References Bittner, C. and M.J. Parnham, 1983, Alteration by prostaglandins and indomethacin of the mouse peritoneal macrophage oxidative burst, Br. J. Pharmacol. 79, 343. Bradford, M.M., 1976, A rapid and sensitive method for the quantiration of microgram quantities utilizing the principle of proteindye binding, Anal. Biochem. 72, 248. Buege, J.A. and S.D. Aust, 1978, Microsomal lipid peroxidation, Meth. Enzymol. 52, 302. Dahlgren, C. and O. Stendahl, 1983, Role of myeloperoxidase in luminol-dependent chemiluminescence of polymorphonuclear leukocytes, Infect. Immun. 39, 736. Dahlgren, C., H. Aniansson and K.-E. Magnusson, 1985, Pattern of formyl-methionyl-leucyl-phenylalanine-induced luminol- and lucigenin-dependent chemiluminescence in human neutrophils, Infect. lmmun. 47, 326. DeChatelet, L.R., G.D. Long, P.S. Shirley, D.A. Bass, M.J. Thomas, F.W. Henderson and M.S. Cohen, 1982, Mechanism of the luminol-dependent chemiluminescence of human neutrophils, J. Immunol. 129, 1589. Fantone, J.C. and D.A. Kinnes, 1983, Prostaglandin E1 and prostaglandin 12 modulation of superoxide production by human neutrophils. Biochem. Biophys. Res. Commun. 113, 506. Grisham, M.B., L.A. Hernandez and D.N. Granger, 1986, Xanthine oxidase and neutrophil infiltration in intestinal ischemia, Am. J. Physiol. 251, G 567. Heikkila, R.E., F.S. Cabbat and G. Cohen, 1976, In vivo inhibition of superoxide dismutase in mice by diethyldithiocarbamate, J. Biol. Chem. 251, 2182. Hirose, H., T. Mizui, N. Shimono and M. Doteuchi, 1985, Gastric antral ulcers induced by a combination of acid, indomethacin and ischemia in rats, Jap. J. Pharmacol. 38, 223. Hodgson, E.K. and I. Fridovich, 1975, The interaction of bovine erythrocyte superoxide dismutase with hydrogen peroxide: Inactivation of the enzyme. Biochemistry 14, 5294. Ishiyama, H., K. Yamasaki, T. Imaizumi, T. Kanbe, K. Ogino, S. Oka, Y. Okazaki and T. Takemoto, 1989, Gastric antra[ ulcer produced by diethyldithiocarbamate in rats, J. Clin. Biochem. Nutr. 5, 155. Itoh, M. and P.H. Guth, 1985. Role of oxygenderived free radicals in hemorrhagic shock-induced gastric lesions in the rat, Gastroenterology 88, 1162. Kivilaakso, E., D. Fromm and W. Silen, 1978, Relationship between ulceration and intramural pH of gastric mucosa during hemorrhagic shock, Surgery 84, 70. McCord, J.M., 1985, Oxygen derived free radicals in postischemic tissue injury, N. Engl. J. Med. 312, 159. McCord, J.M. and I. Fridovich, 1969, Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein), J. Biol. Chem. 244, 6049. Morishita, T. and P.H. Guth, 1987, Effect of exogeneous acid on the rat gastric mucosal microcirculation in hemorrhagic shock, Gastroenterology 92, 1958. Newsby, A.C., 1980, Role of adenosine deaminase, ecto-(5'nucleotidase) and ecto-(non-specific phosphatase) in cyanide-induced adenosine monophosphate catabolis in rat polymorphonuclear leucocytes, Biochem. J. 186, 907. Ogino, K., S. Oka, S. Masuura, I. Sakaida, S. Yoshimura, K. Matsuda, K. Sasaki, K. Yamamoto, T. Yoshikawa, Y. Okazaki and T. Takemoto, 1987, Ulcer formation in rat stomach with diethyldithiocarbamate, J. Clin. Biochem. Nutr. 3, 189. Ogino, K., S. Oka, Y. Okazaki and T. Takemoto, 1988, Gastric
mucosal protection and superoxide dismutase, J. Clin. Gastroenterol. 10 (Suppl.), S129. Ogino, K., T. Hobara, T. Kawamoto, H. Kobayashi, S. lwamoto, S. Oka and Y. Okazaki, 1990, Mechanism of diethyldithiocarbamate induced gastric ulcer formation in the rat, Pharmacol. Toxicol. 66, 13. Ogino, K., T. Hobara, S. Oka, H. Ishiyama, K Yamasaki and T. Kanbe, 1991a, Reduction of the extent of dietbyldithiocarbamateinduced antral gastric ulcer by neutrophil depletion in the rat, Pharmacol. Toxicol. 68, 144. Ogino, K., T. Hobara, H. Kobayashi, H. Ishiyama, M. Gotoh, A. Imamura and N. Egami, 1991b, Lipid peroxidation induced by trichloroethylene in rat liver, Bull. Environ. Contain. Toxicol. 46, 417. Ohkawa, H., N. Ohishi and K. Yagi, 1979, Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction, Anal. Biochem. 95, 351. Oka, S., K. Ogino, T. |lobara, S. Yoshimura, H. Yanai, Y. Okazaki, T. Takemoto, II. Ishiyama, T. Imaizumi, K. Yamasaki and T. Kanbe, 1990, Role of active oxygen species in diethyldithiocarbamate-induced gastric ulcer in the rats, Experientia 46, 281. Dka, S., K. Ogino, T. Hobara, S. Yoshimura, Y. Okazaki, T. Ta.kemoto and Y. lida, 1991, Effects of various mucosal protection drugs on diethyldithiocarbamate-induced antral ulcer in rats, Eur. J. Pharmacol. 197, 99. Parks, D., G. Bulkley, D. Granger, S. Hamilton and J. McCord, 1982, lschemic injury in cat small intestine: Role of superoxide radicals, Gastroenterology 82, 9. Perry, M.A., S. Wadhwa, D.A. Parks, W. Pickard and D.N. Granger, 1986, Role of oxygen radicals in ischemic-induced lesion in the cat stomach, Gastroenterology 90, 362. Pihan, G., C. Regillo and S. Szabo, 1987, Free radical and lipid peroxidation in ethanol- or aspirin-induced gastric mucosal injury, Dig. Dis. Sci. 32, 1395. Robert, A., J.E. Nezamis, C. Lancaster, J.P. Davis, S.O. Field and A.J. Hanchar, 1983, Mild irritants prevent gastric necrosis through adaptive cytoprotection mediated by prostaglandin, Am. J. Physiol. 245, G113. Suematsu, M., I. Kurose, H. Asako, S. Miura and M. Tsuchiya, 1989, In vivo visualization of oxyradical-dependent photoemission during endothelium-granulocyte interaction in microvascular beds treated with platelet-activating factor, J. Biochem. 106, 355. Szelenyi, I. and K. Brune, 1988, Possible role of oxygen free radicals in ethanol-induced gastric mucosal damage in rats, Dig. Dis. Sci. 33, 865. Vissers, M.C.M., W.A. Day and C.C. Winterbourn, 1985, Neutrophils adherent to a non pbagocytosable surface (glomerular basement membrane) produce oxidants only at the site of attachment, Blood 66, 161. Williams, A.J. and P.J. Cole, 1981, The onset of polymorphonuclear leukocyte membrane-stimulated metabolic activity, immunology 43, 733. Yamasaki, K., T. Kanbe, T. Chijiwa, H. Ishiyama and S. Morita, 1987, Gastric mucosal protection by OPC-12759, a novel antiulcer compound, in the rat, Eur. J. Pharmacol. 142, 23. Yoshikawa, T., H. Miyagawa, N. Yoshida, S. Sugino and M. Kondo, 1989, Increase in lipid peroxidation in rat gastric mucosal lesions induced by water immersion restraint stress, J. Clin. Biochem. Nutr. 1,271. Yoshikawa, T., M. Ysuda, S. Ueda, Y. Naito, T. Tanigawa, H. Oyamada and M. Kondo, 1991, Vitamin E in gastric mucosal injury induced by ischemia-reperfusion. Am. J. Clin. Nutr. 53, 210S.
9
© 1992 Elsevier Science Publishers B.V. All rights reserved 0014-2999/92/$05.00
EJP 52298
Antiulcer mechanism of action of rebamipide, a novel antiulcer compound, on diethyldithiocarbamate-induced antral gastric ulcers in rats Keiki O g i n o a, T a t s u y a H o b a r a
a,
H i r o n o b u Ishiyama a, K a t s u y a Y a m a s a k i a, H a r u o Kobayashi ~, Yukie Izumi ~ a n d Shinji O k a b
Department of Public Health, Yamaguchi University, School of Medicine, 1144 Kogushi, Ube 755, Japan and b Kokura Memorial Hospital, Kitakyushu, Japan
Received 22 October 1991, revisedMS received25 November 1991, accepted 3 December 1991
The mechanism of the inhibitory action of rcbamipidc, a new mucosal protective drug, was studied using rats with diethyldithiocarbamate-induced gastric antral ulcers. Rebamipide reduced ulcer formation and inhibited the elevation in lipid peroxide concentration in the gastric mucosa. Rebamipide inhibited both iuminol- and lucigenin-dependent chemiluminescence of neutrophils activated by formyl-methionyl-leucyl-phenylalanine. Rebamipide did not alter the reduction of cytochrome c induced by the xanthine-xanthine oxidase system or the NADPH-dependent micirosomal lipid peroxidation in the liver. These findings suggest that rebamipide prevents diethyldithiocarbamate-induced gastric ulcer formation by inhibiting neutrophil activation. Diethyldithiocarbamate-induced antral ulcer; Neutrophil; Rebamipide
1. Introduction
It has been suggested that active oxygen species may be involved in the pathogenesis of various gastric mucosa injuries, and that some radical scavengers show a protective effect against the mucosal injury induced by active oxygen species (Perry et al., 1986; Pihan et al., 1987; Szelenyi and Brune, 1988). Of the various radical scavengers, we selected superoxide dismutase (SOD) for our study, because exogenous SOD has been reported to prevent mucosal injury caused by active oxygen species (Parks et al., 1982; Itoh and Guth, 1985; Perry et al., 1986; Pihan et al., 1987). During our investigations on SOD, we found that treatment with diethyldithiocarbamate, an inhibitor of copper, zincSOD, produced gastric mucosal injury (Ogino et al., 1987, 1988), and that the production of active oxygen species together with decreased SOD activity and ischemia caused the damage (Ogino et al., 1990). The multiple gastric lesions produced by diethyldithiocarbamate are only shallow and do not reach the muscularis mucosa histologically, and diethyldithiocarbamate treatment fails to cause ulceration in some rats. However, by performing pyloric ligation and by applying 0.1
N HCI intragastrically after treatment with diethyldithiocarbamate, we were able to produce a large ulcer that extended into the antral muscular layer (Ishiyama et al., 1989). Previous studies have indicated that active oxygen species and activated neutrophils are involved in the pathogenesis of this experimentally induced ulcer (Oka et al., 1990; Ogino et al., 1991a). Rebamipide (2-(4-chlorobenzoylamino)-3-[2-(1H)quinolinon-4-yl]propionic acid) is a novel antiulcer agent synthesized by Otsuka Pharmaceutical Co., Japan. Rebamipide increases gastric mucosa prostaglandin concentrations and protects the mucosa from damage induced by various necrotizing agents (Yamasaki et al., 1987). Wc have previously reported that rebamipide may act as a new type of mucosal protective drug by maintaining gastric mucosal SOD activity (Oka et al., 1991). In this study, the mechanism of the antiulcer effect of rebamipide on diethyldithiocarbamate-induced antral ulcers was investigated with regard to the role of active oxygen species.
2. Materials and m e t h o d s
2.1. Animals Correspondence to: K. Ogino, Department of Public Health, Yamaguchi UniversitySchool of Medicine, 1144Kogushi,Ube 755, Japan.
Male Wistar rats weighing 200-230 g were used. The animals were kept in stainless-steel cages with
ll)
wide mesh wire bottoms to prevcnt coprophagia in an air-conditioned room kept at 23 _+ 2°C with artificial lighting from 6 a.m. to 6 p.m. The animals were fasted and allowed access to water ad libitum for 24 h prior to the experiment.
2.2. Chemicals The following chemicals were purchased from the sources indicated: diethyidithiocarbamatc (Wako Pure Chemicals), xanthine oxidase (Boehringer Mannheim), N-l-naphthylethylene-diamine (Tokyo Kasei), xanthine (Sigma), cytochrome c (Sigma), Hanks balanced salt solution (HBSS) (Bioproducts), Dulbecco's phosphatebuffered saline (Bioproducts), formyl-methionyl-leucylphenylalanine (Sigma), lucigenin (Sigma), luminol (Nakalai), 2-thiobarbituric acid (Merck) and rebamipidc (Otsuka Pharmaceutical).
2.3. Production of diethyldithiocarbamate-induced antral ulcer (Ishiyarna et al., 1989) The rats were anesthetized with ether, a midline laparotomy was performed and the pylorus was ligated. Diethyldithiocarbamate dissolved in saline was given s.c. at a dose of 800 mg/kg, and a 1-ml oral dose of 0.1 N HCI was then given to each rat. The rats were killed 5 h later. The stomach was removed and filled with 10 ml of 10% formalin. The ulcer index was measured under a dissecting microscope with a square grid eyescope and is expressed as the area of the antral ulcer (mm2).
2.4. Effect of rebamipide on diethyldithiocarbamate-induced antral ulcer Rebamipide was suspended in 0.5% carboxymethylcellulose and administered i.p. 30 rain before the experiment.
2.5. Effect of rebamipide on the lipid peroxide concentration in antral mucosa The rats were killed 5 h after pylorus ligation, diethyldithiocarbamate treatment, and 0.1 N HCI administration because the lipid peroxide concentration in the antral mucosa is highest within 5 h (Ishiyama et al., 1989). The stomach was removed and the antral mucosa was scraped off. The mucosa was homogenized with 1.15% KCI and the level of thiobarbituric acid reactants was measured (Ohkawa et al., 1979).
2.6. Effect of rebamipide on chemiluminescence of activated neutrophil-rich exudate The rats were injected i.p. with 20 ml of sterile 12% (w/v) sodium caseinate in iso-osmotic (0.9%) NaCI
(Newsby, 1980). Twenty hours later, thc animals were killed by ether asphyxiation, the pcritoncal cavity was opcned and the peritoneal exudatc was collccted. After being filtered through three layers of surgical gauze, the cxudatc was centrifuged at 200 × g for 5 min, and the pcllet was washcd twicc with Dulbccco's phosphate-buffered saline. Thc pellet was suspended in 1 ml of iso-osmotic (0.9%) NaCI, 10 ml of distilled water, and 10 ml of 1.8% NaCI to lead to hypotonic lysis of erythrocyte contaminants. Thc cell suspension was centrifuged as before and the pellet was resuspended in HBSS with 1% bovine serum albumin. Thc specificity of the cell population was determined by differential counting of smears stained with Wright's stain. Viability of the neutrophils was measured with the trypan blue exclusion technique. Chemiluminescence was measured using luminol or lucigenin with the Aloka Co. luminescence rcader at 37°C. Reaction mixtures contained 1.5 × l06 granulocytes, 1 tzM FMLP ( < 0.01% DMSO), 0-1 mM rebamipide (0.001 N NaOH), and 0.07 mM luminol or lucigenin (0.05 M Tris HCi,pH 7.4) in I ml of continuously stirred HBSS. Luminol was dissolved in mildly alkaline water and then adjusted to pH 7.4 with 0.1 N HCI. Rebamipide was diluted with 0.001 N NaOH. Control studies were performed by the addition of 0.001 N NaOH alone. Thc results are exprcssed in terms of the maximal counts per min.
2. 7. Effect of rebamipide on cytochrome c reduction in the xanthine-xanthine oxidase system The standard assay for the measurement of SOD activity was used (McCord and Fridovich, 1969). The incubation mixture contained 0.01 mM cytochrome c, 0.05 mM xanthine, 0.04 U / m l xanthine oxidase, and 0-1 mM rebamipide in 3 ml of 50 mM potassium phosphate buffer (pH 7.8) containing 0.1 mM EDTA. The reduction of cytochrome c was followed at 550 nm.
2.8. Effect of rebamipide on NADPH-dependent microsomal lipid peroxidation in the rat liver Rat liver microsomes were prepared as described previously (Ogino et al., 1991b). Liver microsomes (1.5 mg p r o t e i n / m l ) were incubated with 0-1 mM rebamipide and a freshly prepared NADPH-generating system in 50 mM Tris HCI (pH 7.4). The NADPH-generating system contained 0.17 mM NADP, 3.3 mM glucose-6-phosphate, and 2 units of D-glucose-6-phosphate dehydrogenase. After a 30-rain incubation at 37°C, the lipid peroxide level was measured as the amount of malondialdehyde formed using the thiobarbituric acid method (Buege and Aust, 1975).
(A)
TABLE 1 Effect of i.p. rebamipide on the formation of diethyldithiocarbamateinduced antral ulcers. Data are m e a n s 5: S.E.
Treatment
Do~ (mg/kg)
Number of rats
Ulcer index (mm2)
Inhibition (%)
Vehicle Rebamipide Rebamipide Rebamipide
15 30 60
6 6 6 6
7.85:1.1 3.3+0.8 a 2.6 5:0.3 ~ 1.6 ± 0.6 ~
57.7 66.7 79.5
a p < 0.01 when compared with the vehicle group.
¢:: E
6Tli2
11
8
%
! 0
C
0.1 0.2 ~il~le
C
0.01
(B)
0.5 (raM)
1.0
"~20
g
2.9. Statistical analysis Statistical significance was determined with a oneway analysis of variance a m o n g multiple groups, and P < 0.05 was considered significant.
o 10 x
0
0.1
1.0
m 2.0
Rebamipide (mM)
Fig. 1. Effect of rebamipide on lucigenin-dependent (A) or luminoldependent (B) chemiluminescence in formyl-methionyl-leucyl-phenylalanine-stimulated neutrophil-rich exudate.
3. Results
3.1. Effect of rebamipide on ulcer formation Rebamipide inhibited ulcer formation in a dose-dep e n d e n t m a n n e r (table 1).
3.2. Effect of rebamipide on lipid peroxidation in antral mucosa T h e lipid peroxide level in the antral mucosa of untreated rats was 1.7 + 0.2 n m o l / m g protein. T h e diethyldithiocarbamate plus vehicle group showed an elevated lipid peroxide level, and treatment with rebamipide significantly reduced this in a dose-dependent m a n n e r (table 2).
3.3. Effect of rebamipide on chemiluminescence of neutrophil-rich exudate T h e specificity of the cell population was 90% and neutrophil viability was 99.8%. Rebamipide at 0.1-2.0 TABLE 2 Effect of rebamipide on the lipid peroxide concentration in the antral mucosa. Lipid peroxide is shown as the level of thiobarbituric acid reactants. Data are m e a n s + S.E.
Treatment
Number of rats
Thiobarbituric acid reactants (nmol/mg protein)
Normal antrum mucosa Diet hyldithiocarbamate +vehicle + rebamipide 15 mg/kg + rebamipide 30 mg/kg + rebamipide 60 mg/kg
6
1.7 + 0.2
6 6 6 6
10.15:2.6 a 4.6 + 1.3 3.1 + 1.2 c 15. :1:0.4 b
a p < 0.01 when compared with normal antrum mucosa, b p < 0.01, c p < 0.05 when compared with diethyldithiocarbamate plus vehicle.
m M inhibited l u m i n o l - d e p e n d e n t luminescence, while at 0.2-1.0 m M it inhibited lucigenin-dependent luminescence (fig. 1). Rebamipide had no effect on neutrophil viability in this study.
3.4. Effect of rebamipide on cytochrome c reduction by the xanthine-xanthine oxidase system T h e reduction rates (A O . D . / m i n ) were 0.025, 0.025, 0.024 and 0.023 at 550 nm for the control system and after the addition of 0.01, 0.1 and 1.0 m M rebamipide, respectively. T h e drug thus did not alter the reduction of cytochrome c induced by the production of superoxide radicals in the xanthine-xanthine oxidase system.
3.5. Effect of rebamipide on NADPH-dependent microsomal lipid peroxidation in the rat liver Malondialdehyde levels for controls and with 0.1 and 1.0 m M rebamipide were 14.0 + 0.5, 14.5 + 0.8 and 14.1 + 0.6 n m o i / m g protein, respectively. Each value represents the m e a n + S.E. of three determinations. Rebamipide did not inhibit hepatic N A D P H - d e p e n dent microsomal lipid peroxidation.
4. Discussion T h e mechanism underlying the antiulcer action of rebamipide was investigated in a rat diethyldithiocarbamate-induced antral ulcer model because this type of experimental ulcer is similar to h u m a n gastric ulcer (Ishiyama et al., 1989). HCI was applied to make a
12 deeper ulcer by decreasing mucosal blood flow and intramural pH through the back diffusion of luminal H* (Kivilaaksa et al., 1878; Hirose et al., 1985; Morishita and Guth, 1987), and because diethyldithiocarbamate itself inhibits acid secretion (Ogino ct al., 1990). We chose to administer rebamipide i.p. to induce ulcer formation and to study lipid peroxidation for thc following two reasons. First, intragastrie administration of a drug may induce adaptive cytoprotection. This is caused by dircct stimulation of the gastric mucosa by drugs that act as a mild irritant (Robert ct al., 1983) and is not a pharmacological effect. Second, this model involves HCI administration, so to prevent the drug from altering acidity we chose the i.p. route. Both gastric mucosa injury and the increase in lipid peroxidation were inhibited by treatment with rebamipide. Active oxygen species and lipid peroxidation are closely related to each other in the pathogenesis of diethyldithiocarbamate-induced antral ulcers because the SOD level decreases and the lipid peroxide concentration rises in thc gastric mucosa before injury (Ogino et al., 1989, 1990; Ishiyama el al., 1989), and because some radicial scavengers, e.g. SOD and catalase, inhibit ulcer formation and the decrease in mucosal SOD activity (Ishiyama et al., 1989; Oka et al., 1990, 1991). A correlation between active oxygen species and lipid peroxidation has been reported for experimental ulcers induced by ethanol, for aspirin-induced gastric mucosa injury (Pihan et al., 1987), and for ischemia reperfusion-induced injury (Yoshikawa et al., 1991). However, our study showed that rebamipide itself did not act as a superoxide radical scavenger or as an antioxidant. The source of active oxygen species in the diethyidithiocarbamate-induced antral ulcer model is not known. Xanthine oxidase is the first documented biological source of superoxide radicals and is considered to have an important role in supplying the active oxygen species involved in gastric mucosa injury (Parks et al., 1982; Perry et al., 1986). However, the reduction of cytochrome c by superoxide radicals in the xanthinexanthine oxidase system was not altered by the addition of rcbamipide. Activated neutrophils are another postulated source of active oxygen species (McCord, 1985). Neutrophil-derived free radical damage has been suggested to occur in the gastrointestinal tract (Grisham et al., 1986; Smith et al., 1987). Neutrophil depletion inhibits diethyldithiocarbamate-induced antral ulcer formation more strongly than the administration of radical scavengers, such as SOD and catalase (Ogino et al., 1991a). Luminol-dependent luminescence is said to be totally dependent on the myeloperoxidase (MPO)H 2 0 2 system (DeChatelet et al., 1982; Dahlgren and Stendahl, 1983). Lucigenin is thought to amplify the chemiluminescence response by an MPO-independent mechanism probably related to the production of su-
pcroxidc radicals by neutrophils (Williams and Cole, 1981; Dahlgren et al., 1985). Rcbamipide inhibited luminol-depcndent luminescence at concentrations of 0.1-2.0 mM and inhibited lucigcnin-dependent luminescence at concentrations of 0.2-1.0 mM. The mechanism by which rcbamipide inhibits chemiluminesccnce is unknown at present. However, our findings imply that rebamipide rcduces tissue damage caused by activated neutrophils by inhibiting the production of active oxygen species. The site of action of rebamipide may be the intravascular space for the following two reasons: (1) SOD, catalase, and neutrophil depletion inhibit ulcer formation in this model (Oka et al., 1990; Ogino et al., 1991a); and (2) activated ncutrophils attack venular endothelial cells via active oxygen species after adhering to the endothelium and produce microvascular or histological damage (Vissers et al., 1985; Sucmatu et al., 1989). However, it is not clear whether rebamipide influences neutrophil chemotaxis in the damaged mucosa and how many activated neutrophils are required to causc mucosal damage. Prostaglandin E~ and prostaglandin E 2 (10-5-10 7 M) inhibit tbrmyl-methionyl-leucyl-phenylalanine-induced superoxide radical production by human neutrophils (Fantonc et al., 1983; Bittner and Parnham, 1983). Rebamipide increases gastric mucosa prostaglandin levels (Yamasaki et al., 1987), so it is possible that superoxide radical production was influenced not only by rebamipide itself but also by an increase in mucosal prostaglandin levels. A previous study showed that rebamipide maintains gastric mucosa SOD activity (Oka et al., 1991). This activity may be decreased in our model because diethyldithiocarbamate directly inhibits SOD (Heikkila et al., 1976), or because hydrogen peroxide generated from superoxide radicals by SOD itself inhibits SOD activity (Hodgson and Friovich, 1975). The second reason appears more likely to explain the decrease in SOD activity in this experiment because administration of SOD and catalase prevent ulcer formation in our rat model (Ishiyama et al., 1989; Oka et al., 190; Ogino et al., 1991a). If activated neutrophils are a source of superoxide radicals, our finding that rebamipide inhibits the production of superoxidc radicals by activated neutrophils may support the previous report that rebamipide acts by maintaining SOD activity. This study raises the possibility that the antiulcer effect of rebamipide may be related to inhibition of the production of active oxygen species by activated neutrophils. However, further studies are required to establish the effectiveness of rebamipide in diethyldithiocarbamate-induced antral ulcers and to define further the mechanisms by which the production of active oxygen species leads to morphological and functional injury to the gastric mucosa.
13
References Bittner, C. and M.J. Parnham, 1983, Alteration by prostaglandins and indomethacin of the mouse peritoneal macrophage oxidative burst, Br. J. Pharmacol. 79, 343. Bradford, M.M., 1976, A rapid and sensitive method for the quantiration of microgram quantities utilizing the principle of proteindye binding, Anal. Biochem. 72, 248. Buege, J.A. and S.D. Aust, 1978, Microsomal lipid peroxidation, Meth. Enzymol. 52, 302. Dahlgren, C. and O. Stendahl, 1983, Role of myeloperoxidase in luminol-dependent chemiluminescence of polymorphonuclear leukocytes, Infect. Immun. 39, 736. Dahlgren, C., H. Aniansson and K.-E. Magnusson, 1985, Pattern of formyl-methionyl-leucyl-phenylalanine-induced luminol- and lucigenin-dependent chemiluminescence in human neutrophils, Infect. lmmun. 47, 326. DeChatelet, L.R., G.D. Long, P.S. Shirley, D.A. Bass, M.J. Thomas, F.W. Henderson and M.S. Cohen, 1982, Mechanism of the luminol-dependent chemiluminescence of human neutrophils, J. Immunol. 129, 1589. Fantone, J.C. and D.A. Kinnes, 1983, Prostaglandin E1 and prostaglandin 12 modulation of superoxide production by human neutrophils. Biochem. Biophys. Res. Commun. 113, 506. Grisham, M.B., L.A. Hernandez and D.N. Granger, 1986, Xanthine oxidase and neutrophil infiltration in intestinal ischemia, Am. J. Physiol. 251, G 567. Heikkila, R.E., F.S. Cabbat and G. Cohen, 1976, In vivo inhibition of superoxide dismutase in mice by diethyldithiocarbamate, J. Biol. Chem. 251, 2182. Hirose, H., T. Mizui, N. Shimono and M. Doteuchi, 1985, Gastric antral ulcers induced by a combination of acid, indomethacin and ischemia in rats, Jap. J. Pharmacol. 38, 223. Hodgson, E.K. and I. Fridovich, 1975, The interaction of bovine erythrocyte superoxide dismutase with hydrogen peroxide: Inactivation of the enzyme. Biochemistry 14, 5294. Ishiyama, H., K. Yamasaki, T. Imaizumi, T. Kanbe, K. Ogino, S. Oka, Y. Okazaki and T. Takemoto, 1989, Gastric antra[ ulcer produced by diethyldithiocarbamate in rats, J. Clin. Biochem. Nutr. 5, 155. Itoh, M. and P.H. Guth, 1985. Role of oxygenderived free radicals in hemorrhagic shock-induced gastric lesions in the rat, Gastroenterology 88, 1162. Kivilaakso, E., D. Fromm and W. Silen, 1978, Relationship between ulceration and intramural pH of gastric mucosa during hemorrhagic shock, Surgery 84, 70. McCord, J.M., 1985, Oxygen derived free radicals in postischemic tissue injury, N. Engl. J. Med. 312, 159. McCord, J.M. and I. Fridovich, 1969, Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein), J. Biol. Chem. 244, 6049. Morishita, T. and P.H. Guth, 1987, Effect of exogeneous acid on the rat gastric mucosal microcirculation in hemorrhagic shock, Gastroenterology 92, 1958. Newsby, A.C., 1980, Role of adenosine deaminase, ecto-(5'nucleotidase) and ecto-(non-specific phosphatase) in cyanide-induced adenosine monophosphate catabolis in rat polymorphonuclear leucocytes, Biochem. J. 186, 907. Ogino, K., S. Oka, S. Masuura, I. Sakaida, S. Yoshimura, K. Matsuda, K. Sasaki, K. Yamamoto, T. Yoshikawa, Y. Okazaki and T. Takemoto, 1987, Ulcer formation in rat stomach with diethyldithiocarbamate, J. Clin. Biochem. Nutr. 3, 189. Ogino, K., S. Oka, Y. Okazaki and T. Takemoto, 1988, Gastric
mucosal protection and superoxide dismutase, J. Clin. Gastroenterol. 10 (Suppl.), S129. Ogino, K., T. Hobara, T. Kawamoto, H. Kobayashi, S. lwamoto, S. Oka and Y. Okazaki, 1990, Mechanism of diethyldithiocarbamate induced gastric ulcer formation in the rat, Pharmacol. Toxicol. 66, 13. Ogino, K., T. Hobara, S. Oka, H. Ishiyama, K Yamasaki and T. Kanbe, 1991a, Reduction of the extent of dietbyldithiocarbamateinduced antral gastric ulcer by neutrophil depletion in the rat, Pharmacol. Toxicol. 68, 144. Ogino, K., T. Hobara, H. Kobayashi, H. Ishiyama, M. Gotoh, A. Imamura and N. Egami, 1991b, Lipid peroxidation induced by trichloroethylene in rat liver, Bull. Environ. Contain. Toxicol. 46, 417. Ohkawa, H., N. Ohishi and K. Yagi, 1979, Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction, Anal. Biochem. 95, 351. Oka, S., K. Ogino, T. |lobara, S. Yoshimura, H. Yanai, Y. Okazaki, T. Takemoto, II. Ishiyama, T. Imaizumi, K. Yamasaki and T. Kanbe, 1990, Role of active oxygen species in diethyldithiocarbamate-induced gastric ulcer in the rats, Experientia 46, 281. Dka, S., K. Ogino, T. Hobara, S. Yoshimura, Y. Okazaki, T. Ta.kemoto and Y. lida, 1991, Effects of various mucosal protection drugs on diethyldithiocarbamate-induced antral ulcer in rats, Eur. J. Pharmacol. 197, 99. Parks, D., G. Bulkley, D. Granger, S. Hamilton and J. McCord, 1982, lschemic injury in cat small intestine: Role of superoxide radicals, Gastroenterology 82, 9. Perry, M.A., S. Wadhwa, D.A. Parks, W. Pickard and D.N. Granger, 1986, Role of oxygen radicals in ischemic-induced lesion in the cat stomach, Gastroenterology 90, 362. Pihan, G., C. Regillo and S. Szabo, 1987, Free radical and lipid peroxidation in ethanol- or aspirin-induced gastric mucosal injury, Dig. Dis. Sci. 32, 1395. Robert, A., J.E. Nezamis, C. Lancaster, J.P. Davis, S.O. Field and A.J. Hanchar, 1983, Mild irritants prevent gastric necrosis through adaptive cytoprotection mediated by prostaglandin, Am. J. Physiol. 245, G113. Suematsu, M., I. Kurose, H. Asako, S. Miura and M. Tsuchiya, 1989, In vivo visualization of oxyradical-dependent photoemission during endothelium-granulocyte interaction in microvascular beds treated with platelet-activating factor, J. Biochem. 106, 355. Szelenyi, I. and K. Brune, 1988, Possible role of oxygen free radicals in ethanol-induced gastric mucosal damage in rats, Dig. Dis. Sci. 33, 865. Vissers, M.C.M., W.A. Day and C.C. Winterbourn, 1985, Neutrophils adherent to a non pbagocytosable surface (glomerular basement membrane) produce oxidants only at the site of attachment, Blood 66, 161. Williams, A.J. and P.J. Cole, 1981, The onset of polymorphonuclear leukocyte membrane-stimulated metabolic activity, immunology 43, 733. Yamasaki, K., T. Kanbe, T. Chijiwa, H. Ishiyama and S. Morita, 1987, Gastric mucosal protection by OPC-12759, a novel antiulcer compound, in the rat, Eur. J. Pharmacol. 142, 23. Yoshikawa, T., H. Miyagawa, N. Yoshida, S. Sugino and M. Kondo, 1989, Increase in lipid peroxidation in rat gastric mucosal lesions induced by water immersion restraint stress, J. Clin. Biochem. Nutr. 1,271. Yoshikawa, T., M. Ysuda, S. Ueda, Y. Naito, T. Tanigawa, H. Oyamada and M. Kondo, 1991, Vitamin E in gastric mucosal injury induced by ischemia-reperfusion. Am. J. Clin. Nutr. 53, 210S.
