Gen. Pharmac. Vol. 23, No. 4, pp. 719-727, 1992 Printed in Great Britain. All rights reserved
0306-3623/92 $5.00 + 0.00 Copyright ~, 1992 Pergamon Press Lid
MECHANISM OF EBROTIDINE PROTECTION AGAINST GASTRIC MUCOSAL INJURY INDUCED BY ETHANOL B . L . SLOMIANY,J. PIOTROWSKI,V. L. N. MURTY and A. SLOMIANY Research Center, New Jersey Dental School, University of Medicine and Dentistry of New Jersey, University Heights, II0 Bergen Street, Newark, NJ 07103-2400, U.S.A. [Tel. (201)456-7052] (Received 12 November 1991) A~tract--l. The gastroprotective properties of a new H2-receptor antagonist, ebrotidine, against ethanol-induced mucosal injury was investigated. 2. Groups of rats, with and without indomethacin pretreatment, received intragastrically either a dose of ebrotidine or vehicle only, followed by ethanol given at various intervals up to 4 hr. The gastric mucosa, 30 min after the ethanol challenge, was then subjected to macroscopic and histologic examination, and physic*chemical measurements. 3. Ebrotidine at doses of 50 mg and higher per kg body wt effectively prevented the alcohol-induced mucosal injury, even in the presence of indomethacin. The protective effect was demonstrable already at 30 min, reached maximum at I hr, and persisted up to 3 hr. 4. Physic,chemical analyses established that ebrotidine elicited 30% increase in mucus gel dimension, caused 19-20'/, increase in glycolipids and phospholipids, and evoked 21'/, increase in sulfomucin and 180 in sialomucins. As a consequence, the mucus gel viscosity increased by 1.4-fold, H + retardation capacity by 16°/,, and hydrophobicity by 65%. 5. The results demonstrate that ebrotidine is a unique H2-antagonist endowed with a remarkable mucosal strengthening capability.
INTRODUCTION
through strengthening the mucus gel physic*chemical characteristics.
Mucus coat overlying the epithelial surfaces of alimentary tract constitutes an essential element of mucosal protection. In the stomach, this viscous and slimy layer, together with cell membrane of gastric epithelium and the vasculature, are considered as major components of the defense mechanism responsible for the maintenance of gastric mucosal integrity under adverse environment of the lumen (Szabo et al., 1986; Silen, 1987; Slomiany et al., 1989; Slomiany and SIomiany, 1991). The integrity and strength o f mucus layer defense perimeter depends upon a delicate balance controlled by factors affecting the synthesis secretion and breakdown of its protein, glycoprotein, and lipid components (Yuonan et al., 1982; Bilski et al., 1987; Slomiany and SIomiany, 1991). While this balance is successfully maintained under normal physiological conditions, injury to the mucosa ensues when aggressive forces overcome factors that control mucosal defense. Indeed, the weakening o f gastric mucosal defense is intimately associated with the diminished viscoelastic qualities of mucus, decreased hydrogen ion retardation capacity and hydrophobicity, and excessive peptic erosion (Yuonan et al., 1982; Bilski et al., 1988; Piotrowski et al., 1991). Therefore, gastroprotective agents capable o f strengthening the protective qualities of mucus gel through the enhancement o f its physic*chemical properties are of value in peptic ulcer therapy. Here we present evidence that ebrotidine, a new H2-receptor antagonist (Anglada et al., 1988), in addition to its antisecretory activity also exhibits a remarkable ability o f gastric mucosal protection against ethanol-induced injury
MATERIALS AND METHODS Animals The study was conducted with male Sprague-Dawley rats weighing 160-180g. They were maintained on a regular chow diet and fasted in individual wire-bottom cages for 24 hr prior to experimentation. Water was withheld 2hr before the experiment. The animals received intragastric pretreatment with 1 ml ofebrotidine as emulsion in 5% gum arabic in a dose of 10-150mg/kg body wt, or vehicle consisting of 5% gum arabic in saline followed by 1 ml of absolute ethanol given at various time intervals up to 4 hr after the pretreatment. The animals were sacrificed 30 rain after the ethanol dose and their stomachs removed. The dissected stomachs were rinsed with tap water, opened along the greater curvature, gently washed with cold saline, and subjected to macroscopic and histologic assessment. Such dissected stomachs were also used for the measurements of the mucus coat thickness, evaluation of the adherent mucus gel mucin content and the chemical analysis of mucus components. All experiments were carried out in the presence or absence of pretreatment with indomethacin (5 mg/kg) given i.p. 30 min prior to ebrotidin¢ or vehicle (5% gum arabic in saline). Macroscopic and histological assessment For macroscopic assessment, opened stomachs were laid flat on a Teflon board and the areas of gross hemorrhagic necrosis were quantitated planimetrically. The results of analysis were expressed as a percentage of the total glandular mucosal area occupied by necrotic lesions. For mucosal histology, stomachs were cut obliquely through the entire extent of glandular mucosa into strips 5mm in width. Specimens were fixed in 10% buffered 719
720
B.L. SLOMIANYet al.
formalin and stained with hematoxylin and eosin. The disruption of the continuity of the surface epithelial layer and the extent of deep histologic necrosis were quantitated by microscopic evaluation of the stained sections of the mucosa, and expressed as a percentage of total mucosal length of each studied strip. Mucus coat thickness
Sections of gastric mucosa, luminal surface up, were mounted on a Millipore filter base and cut into strips 1.6 mm in width. Except for being sectioned, the tissue was kept immersed in 0.15 M NaCI solution. Cut strips were mounted transversely in Petri dishes and their positioning checked stereoscopically (x2.5 magnification) to ensure they were not mounted obliquely. Such prepared sections of gastric mucosa were subjected to mucus coat thickness measurement by means of an inverted microscope ( x 100 magnification). The distance between the bathing solutionmucus layer interface and mucus-glandular mucosa epithelial surface interface was recorded with the aid of eyepiece graticule (Kerss et al., 1982; Bilski et al., 1988). Mucin o f adherent mucus
The glandular segments of freshly dissected stomachs were excised, weighed and placed in 10 ml of 0. I% Alcian blue solution containing 0.16 M sucrose in 0.05 M citrate, pH 2.0 for sulfomucin measurements or 0.05 M sodium acetate buffer, pH 5.8 for sulfo- and sialomucins. After 2 hr of staining, the excess dye was removed by soaking the segments in 0.15 M sucrose (Come et al., 1974). The dye complexed with mucus adherent to the gastric walls was extracted from the mucosa with 10 ml of 0.5 MgCI 2, diethyl ether was added, the resulting emulsion was centrifuged, and the separated aqueous layer was used to determine the dye concentration by spectrometry (Bilski et al., 1988). Adherent mucus isolation
The gastric mucus gel used for physicochemical analyses was obtained by blotting the individual opened stomachs with Whatman No. 3 filter paper (Piotrowski et al., 1990). The adherent mucus, transferred on to the filter paper, was then recovered by washing the paper with 1.0 M NaCI in 0.05 M sodium phosphate buffer, pH 7.0. Such isolated mucus was filtered through a Millipore HA (0.45 gm) filter, subjected to intrinsic pepsin inactivation (pH 9.0 at 37°C for 30min), dialyzed against distilled water and lyophilized. Hydrogen ion retardation capacity
The diffusion of H ÷ through gastric mucus gel was measured in a specially constructed permeability chamber (Bilski et al., 1987). Individual samples of mucus, dissolved at 30 mg/ml in 0.15 M NaC1, were placed in the diffusion port separating the two compartments, filled on one side with 0.15 N NaC1 and on the other side with 0.15 M HCI, and the change in pH in the NaC1 compartment was recorded at 5 rain intervals for up to 2 hr with a micro-pH electrode connected to an Accumet recording ionalyzer. The amount of H ÷ diffusion through the sample was calculated in mol/sec and the permeability coefficient in cm/sec (Piotrowski et al., 1990).
Mucus gel hydrophobicity
The hydrophobicity of the adherent gastric mucus gel was evaluated using a fluorescent hydrophobic probe, bis(8anilino-l-naphthalenesulfonate) (bis-ANS). The measurements were conducted with a Perkin-Elmer model LS-5 fluorescent spectrophotometer (Gwozdzinski et al., 1988). The excitation wavelength for the probe was 365 nm and the maximum emission was observed at 530 nm. The samples, dissolved at 2mg/ml in 0.10 M NaCI/0.05 sodium phosphate buffer, pH 7.0, were reacted at 25~C with increasing amounts of bis-ANS and the induced alterations in the emission spectra of the probe were evaluated (Gwozdzinski et al., 1988). Analytical methods
The content and composition of carbohydrate in the prepared mucus samples were determined by gas-liquid chromatography following methanolysis, re-N-acetylation and derivatization with silylating reagent (Slomiany et al., 1984), and the protein was measured by the method of Lowry et al. (1951). For the analysis of lipids, individual samples of mucus powder were extracted with chloroformmethanol (Slomiany et al., 1985). The extracts were filtered through a grade F sintered glass funnel, and the lipids contained in the filtrates were separated on silicic acid columns into neutral lipids, glycolipid and phospholipid fractions (Bilski et al., 1987). The neutral lipids were separated into individual components by thin-layer chromatography, identified by comparison with chromatograms of authentic standards, and quantitated (Bilski et al., 1987, 1988). The glycolipids were quantitated by measuring their carbohydrate and lipid constituents (Slomiany et al., 1985). The phospholipids were subjected to two-dimensional thinlayer chromatography and quantitated by measuring their phosphorus content (Lowry and Tinsley, 1974). For the analysis of covalently bound fatty acids, the delipidated mucus preparations were incubated for 30 min at 37°C with 0.3 M methanolic NaOH, and the released fatty acid methyl esters were extracted with hexane and quantitated by gas-liquid chromatography (Slomiany et al., 1987b). Antiulcer drug
Ebrotidine, p-bromo-N-[[[2-[[[2-[(diaminomethylene)amino]4-thiazolyl]-methyl]-thio]-ethyll-amino]-methylene]benzene sulfonamide, lot No. D-6, was kindly donated by Ferrer International, S.A., Barcelona, Spain. The drug was stored at 4'C in the dark and was suspended in 5% gum arabic-0.9% NaC1 shortly before experimentation. The drug or vehicle (5% gum arabic in saline) were given orally in a volume of I ml through a dull metal tubing attached to a 2 ml syringe. Data analysis
All experiments were carried out in duplicate, and the results are expressed as means + SD. Student's t-test was used to determine significance, and P values of 0.05 or less were considered significant.
RESULTS
Viscosity measurements
G a s t r i c lesions a n d m u c o s a l h i s t o l o g y
Viscosity determinations were performed with a Brookfield cone/plate digital viscometer, model LVTDCP, equipped with a 1.565 ° cone and a constant (37°C) temperature bath (Slomiany et al., 1986). Shear rates were varied from 1.15 to 115 sec - t, and the sample volumes were 0.5 ml. For the measurements, samples of mucus, dissolved at 30 mg/ml in 0.10 M NaCI, 0.05 M sodium phosphate buffer, pH 6.8, were gently stirred at 4°C for 1 hr and then brought to 37°C. To calculate the specific viscosity, measurements were also taken of buffer alone.
The gross a p p e a r a n c e of gastric m u c o s a 30 min following intragastric a d m i n i s t r a t i o n o f absolute e t h a n o l in the absence a n d the presence o f pretreatm e n t with ebrotidine is depicted in Fig. 1. E t h a n o l a d m i n i s t r a t i o n to the animals given vehicle p r o d u c e d extensive necrotic h e m o r r h a g i c lesions in the glandular p o r t i o n o f the mucosa, b u t little, if any, effect was observed in the mucosal a p p e a r a n c e of the forestomach. The lesions in the glandular area were in the
Ebrotidine and mucosal protection
"1~ W
721
f
Fig. I. Gross appearance of rat gastric mucosa 30 rain after intragastfic dose of 1ml absolute ethanol.
Vehicle pretreated animals in the absence (A) and presence (B) of indomethacin. Ebrotidine pretreated (75 mg/kg) animals in the absence (C) and presence (D) of indomethacin.
form of red streaks, parallel to the long axis of the stomach and occupied over 40% of the mucosa (Fig. 2). Pretreatment with ebrotidine at doses of 50 mg/kg and above significantly reduced the extent of mucosal necrosis caused by ethanol, and this inhibitory effect of ebrotidine was not affected by indomethacin given i.p. 30 min prior to the ebrotidine (Figs 1 and 2). The histology of gastric mucosa 30 rain following intragastric administration of ethanol in the absence and the presence of pretreatment with ebrotidine is presented in Fig. 3. The results indicated that in the vehicle-treated rats, ethanol caused extensive disruption and desquamation of the surface epithelium, and produced rampant necrotic lesions penetrating deeply into the mucosa (Fig. 3, Table 1). Submucosal edema and polymorphonuclear leukocyte infiltration were also present outside the deep hemorrhagic lesions. Also, in the areas of deep hemorrhagic necrosis mucosal microvessels were often obstructed with erythrocytes or ruptured. Administration of ebrot-
idine at doses of 50 mg/kg and above markedly reduced (81%) the disruption of the surface epithelium, and the number of deep necrotic lesions decreased by 90%. Furthermore, the microvascular lesions were rare and essentially confined to the luminal surface mucous cells. Pretreatment with indomethacin prior to ebrotidine increased the extent of histologic necrotic lesions evoked by ethanol, but did not affect significantly the extent of ebrotidine elicited protection. Dose and duration The effect of the dose on the gastric mucosal protective effect of ebrotidine against ethanol injury is presented in Fig. 4. The results indicated that ebrotidine successfully prevented gastric mucosal damage by ethanol at doses as low as 75-50 mg/kg of body wt, at which concentration 89.1 and 75.8% reduction in mucosal necrosis occurred (Fig. 4). The data on the duration of gastroprotective action of ebrotidine against ethanol injury are summarized
722
B. L. SLOMIANY et al.
_
50
T
~. 4 0 30 o o
20 --~
÷
10
g Z
0
Vehicle
VQhlchl * Ethmal
"
T
1
EI~ro t Id Ine I m l ~ n e l h l ~ Ina~mlho¢~ I r . ~ , ~ " ~ ¢ m * + * + ~'tl~nol ¥~le Ethon~ Ebrotldlm + Etemr~
Fig. 2. Effect of ebrotidine on macroscopic necrosis of gastric mucosa caused by ethanol in the absence and presence of pretreatment with indomethacin. Ebrotidine (75 mg/kg) or vehicle were administered 60 min prior to ethanol dose. Values represent the means + SD of analyses performed on 10 animals in each group. The effect of ebrotidine + ethanol, and ebrotidine + indomethacin and ethanol vs vehicle + ethanol, and vehicle+ indomethacin and ethanol were significant at "P < 0.05. in Fig. 5. A significant protective effect was demonstrable already at 30 rain, reaching maximum at l hr, and was still observed at 3 hr. Adherent mucus and its mucin content
Examination of gastric mucosal surface by means of inverted microscope technique (Kerss et al., 1982) indicated the presence of three distinct regions under light-and dark-field illumination. These were identified as bathing solution, layer of mucus gel, and the glandular mucosa. By examining seven sections taken from glandular mucosa of each animal used in the study, it was found that in the group treated with vehicle, the mucus gel dimension averaged 223.5 gin, while its thickness in the animals treated with ebrotidine was 290.0/~m, which represents a 30% increase over the control value. Pretreatment with indomethacin caused a 47% reduction in the gel dimension of animals receiving the vehicle, while the animals receiving ebrotidine showed a 98% increase in mucus dimension over that of the indomethacin controls (Table 2). Table 1. Effectof ebrotidineon histologyof gastric mucosaexposed to ethanol Disruption of Deepnecrotic surface epithelium lesions Pretreatment (%) (%) Vehicle (5% gum arabic in 0.15M NaCI) 0 0 Vehicle+ ethanol 95.4 ± 4.5 46.2 + 4.3 Ebrotidine (75 mg/kg) in vehicle 0 0 Ebrotidine (75 mg/kg) + ethanol 18.4 ± 2.7* 4.3 + 0.3* Indomethacin + vehicle + ethanol
93.9 ± 5.3
48.6 + 3.8
lndomethacin+ ebrotidine + ethanol lndomethacin+ vehicle
32.3 ± 3.9* 7.3 ± 0.5* 3.5 ± 0.4 1.3 ± 0.3 Values represent the means ± SD of analyses performed on I0 animals in each group. Ebrotidineor vehicle were administered 60rainprior to ethanol dose. The effectsof ebrotidine+ ethanol, and ebrotidine + indomethacin and ethanol vs vehicle + ethanol, and vehicle + indomethacin were significant at "P < 0.05.
Evaluation of the adherent mucus gel mucin content by the Alcian blue uptake assays, conducted at pH 2.0 for sulfomucins and at pH 5.8 for sulfo- and sialomucins, indicated that in the absence of ebrotidine treatment, indomethacin evoked 37% decrease in sulfomucin and 41% decrease in sialomucin contents of the mucus gel (Table 2). In comparison to the indomethacin pretreated animals receiving vehicle, the mucus gel of animals exposed to ebrotidine showed 46% increase in the content sulfomucin and 22% increase in sialomucins. In the absence of indomethacin pretreatment, ebrotidine caused 21% increase in gastric mucus gel sulfomucin content and 18% increase in the content of sialomucin. Chemical composition
The chemical composition of the adherent mucus gel elaborated in the absence and presence of pretreatment with indomethacin, without and with the administration of ebrotidine is presented in Table 3. In the absence of pretreatment with indomethacin, the mucus of the animals receiving ebrotidine showed a significantly higher content of phospholipids (20%), glycolipids (19%) and glycoprotein (19%), and only negligible lower (8%) content of protein. Pretreatment with indomethacin in the absence of ebrotidine led to a 14% increase in mucus gel protein, 15% decrease in glycoprotein, and a 22% decrease in total lipids, reflected mainly in a marked decrease in phospholipids. The changes evoked by indomethacin in the composition of elaborated mucus were prevented by ebrotidine treatment, and the chemical characteristics of gastric mucus of ebrotidine-treated animals were essentially similar to that of the group which received vehicle in the absence of indomethacin. Hydrophobicity, viscosity and H + retardation
The data on the effect of ebrotidine administration on the hydrophobicity of the adherent gastric mucus are presented in Fig. 6. The results revealed that the amount of hydrophobic fluorescent probe, bis-ANS, required to obtain maximum fluorescence intensity with mucus gel of animals treated with ebrotidine was 65% greater than with gastric mucus gel of the group receiving the vehicle. This effect of ebrotidine, furthermore, was not thwarted by indorfiethacin pretreatment (Fig. 6). Figure 7 shows the effect of ebrotidine administration on the viscosity of the adherent gastric mucus gel. The specific viscosity of mucus from the vehicle group ranged from 3.6 at the shear rate of 1.15 sec-' to 1.9 at the shear rate of 115 sec -~. The specific viscosity of mucus gel from animals treated with ebrotidine dose of 75 mg/kg body wt ranged from 5.1 at the lowest shear rate to 3.4 at the highest shear rate, while the viscosity values for ebrotidine in vehicle alone ranged only from 0.3 to 0.1. Pretreatment with indomethacin in the absence of ebrotidine caused a 45% drop in mucus gel viscosity, while in the present of ebrotidine, the viscosity values at the shear rate of i.15 sec-' increased to 3.7 and to 2.3 at the shear rate of 115 sec-z. The effect of ebrotidine administration on the permeability of the adherent gastric mucus gel to hydrogen ion is summarized in Table 4. The H ÷
Ebrotidine and mucosal protection
123
724
B. L. SLOMIANY et al.
Table 2. Effect of ebrotidine on gastric mucus gel dimension and its content of mucin
-
5o
o
Mucin content (/~g Alcian blue/g tissue) Pretreatment Vehicle lndomethacin + vehicle Ebrotidine (75 mg/kg) Indomethacin + ebrotidine (75 mg/kg)
Mucus gel dimension (~m)
pH 2.0
pH 5.8
223.5 -+ 8.9
93.9 + 4.8
148.0 ~ 6.3
o
40
~
30
o
117.5-+9.3
59.5+3.1
91.2+_4.7
290.0 -+ 9.2*
114.2 -+ 4.6*
178.1 _+ 6.2*
232.5 + 7.5*
86.7 -+ 4.3*
140.6 + 5.5*
~- 20 o
Values represent the means_+ SD of analyses performed on l0 animals in each group. The assays were carried out 60 min following ebrotidine (75mg/kg). The effects of ebrotidine vs vehicle and ebrotidine +indomethacin vs vehicle+ indomethacin were significant at *P < 0.05.
retardation capacity of mucus samples from the animals treated with ebrotidine was 16% higher than that of the vehicle group. Indomethacin pretreatment caused a 12% decrease in mucus H ÷ retardation capacity in the absence of ebrotidine treatment, while in the presence of ebrotidine the mucus H ÷ retardation values were quite similar to those obtained with the vehicle in the absence of indomethacin.
._~ 10 b. 1
Z
0
I
l
30 Dose
J"
6O
T I J 90
t20
In~/kg)
Fig. 4. Effect of ebrotidine dose (mg/kg) on the gastroprotective potential against ethanol-induced injury. Values represent the means + SD of analyses performed on 7 animals with each dose of ebrotidine. 5o I o
*°l
o
30 o
DISCUSSION
o
Ebrotidine is a new H2-receptor antagonist with antisecretory potency comparable to that of ranitidine also showing some activity on H~-receptors (Anglada et al., 1988; Konturek et al., 1992). Structurally, although this agent shares common features with cimetidine and ranitidine, it contains N-sulfonyl formamidine group instead of N-cyanoguanidine group of cimetidine and 2-nitroethendiamine group of ranitidine, while the imidazole ring of cimetidine substituted by guanidinothiazole (Anglada et al., 1988). These modifications endow ebrotidine with diminished cytochrome P-450 binding as compared to that of ranitidine and cimetidine, and eliminate the potential for mutagenic nitros.amine formation. Furthermore, the results of a recent study suggest that ebrotidine, in contrast to ranitidine and cimetidine, exhibits remarkable gastroprotection against ethanol induced mucosal injury. This effect of ebrotidine apparently does not seem to involve prostaglandin mechanism as the gastroprotection is maintained
o
2O
m 'tO 'N
Z
0
1
I
I
I
I
30
90
150
210
270
T i m e (rain)
Fig. 5. T h e d u r a t i o n o f g a s t r o p r o t e c t i v e effect o f e b r o t i d i n e a g a i n s t e t h a n o l - i n d u c e d injury. E a c h p o i n t r e p r e s e n t s the m e a n s + S D o f a n a l y s e s p e r f o r m e d o n 7 a n i m a l s a t different time intervals following pretreatment with ebrotidine (75 m g / k g ) .
even in the presence of such potent prostaglandin synthesis inhibitors as acetylsalicylic acid and indomethacin (Palop and Marquez, 1991). The data obtained from this investigation show that the gastroprotective effect of ebrotidine against ethanol-induced injury results from the drug's ability
Table 3. Effect of ebrotidine on the chemical composition of gastric mucus gel Pretreatment Constituent (rag/100 mg mucus) Protein Carbohydrate Lipids (total) Neutral lipids Glycolipids Phospholipids Covalently bound fatty acids
Vehicle
indomethacin + vehicle
Ebrotidine
lndomethacin + ebrotidine
65.8+_6.1 11.4 + 0.9 19.6+ 1.8 11.5 -+ 1.4 4.7 -+ 0.5 3.4 -+ 0.2
74.9_+6.2 9.7_+ 1.0 15.2± 1.6 8.9 +- 1.0 4. I -+ 0.4 2.2 -+ 0.2
60.7+5.0 13.6+ 1.2" 22.1 _+2.0 12.4 4-_ 1.5 5.6 ± 0.5* 4. I + 0.4*
62.5-,-4.9* 12.0_+ 1.3' 21.6+2.1" 12.2 + [.3" 5.5 _+ 0.5* 3.8 -+ 0.5*
0.3-'-0.1
0.2+_0.1
0.4-+0.2
0.3+0.2
Values represent the means + SD of analyses performed on the individual samples of mucus obtained from 10 animals in each group subjected to pretreatment for 60 min with ebrotidine (75 mg/kg) or vehicle. The effects of ebrotidine vs vehicle and indomethacin plus ebrotidine vs indomethacin plus vehicle were significant at *P < 0.05.
725
Ebrotidine and mucosal protection
Table 4. Effect of ebrotidine on the permeability of the adherent gastric mucus gel to hydrogen ion
80 II
T
T
6O
Pretreatment
>-
Vehicle (5% gum arabic in 0.15M NaC1) Indomethacin + vehicle Ebrotidine (75 mg/kg) + vehicle Indomethacin + ebrotidine (75 mg/kg)
T
Vehicle
Ebrotldlml
|ndomethoctn I n d o m ~ h c ] e l n + +
Vehicle
Ebrotldlne
Fig. 6. Effect of intragastric ebrotidine (75 mg/kg) administration on the hydrophobicity of gastric mucus. Values are plotted as means + SD (n = 10) of the increase in relative fluorescence of hydrophobic probe (bis-ANS) due to mucus binding. Asterisk indicates a significant (P < 0.05) increase in hydrophobicity over the control value. to enhance the physicochemical qualities of mucus layer perimeter of gastric mucosal defense. Moreover, ebrotidine has the ability to maintain the physicochemical qualities of this layer even in the presence of indomethacin when prostaglandin synthesis is inhibited. Our results demonstrated that the gastroprotective effect of ebrotidine is dose dependent, and could be achieved with drug concentrations as low as 75-50 mg/kg of body weight. The time-course studies of the gastroprotective effect indicated that ebrotidine elicits maximal protective effect 30-60 rain following administration, and that the effect persists for at least up to 3 hr. The observed gastric mucosal protective action of ebrotidine is apparently exerted through enhancement of physical and chemical characteristics of mucus gel. Experiments with animals pretreated with prostaglandin-inhibitory dose of indomethacin (Robert, 1981; Ligumsky et al., 1982) revealed that 5 -
T
I!
~3 "G 2 (n
Vehicle
El=cot Idlntl
I n d o m e t h ~ Indem~.hacln Vehicle
[brotldlne
Fig. 7. Effect of intragastric ebrotidine (75 mg/kg) administration on the viscosity of gastric mucus. Values represent the means + SD of duplicate anlayses performed on the individual samples of mucus obtained from 10 animals in each group. Shear rate 11.5 sec -t. The changes in mucus viscosity evoked by the drug were significant at P < 0.05.
Permeability (tool. 10 ~0.sec-' )
Permeability coefficient (cm. 10- 6 .see- ] )
4.21 +0.31 4.73 _+ 0.38
7.05+_0.56 7.90 _+ 0.62
3.53 +_ 0.27*
5.92 _+ 0.38*
4.28 _+ 0.25*
7.17 _+ 0.34*
Values represent the means + SD of analyses performed on individual samples of mucus obtained from 10 animals in each group. The assays were carried out 60min following ebrotidine (75 mg/kg). The effects of ebrotidine vs vehicle and ebrotidine + indomethacin vs vehicle + indomethacin were significant at *P < 0.05.
ebrotidine successfully prevented the detrimental effect of indomethacin on the adherent mucus gel dimension and its content of sulfo- and sialomucins. In comparison to the indomethacin pretreated animals receiving vehicle, the mucus coat of animals treated with ebrotidine showed 98% increase in gel thickness, 46% increase in the content of sulfomucins, and a 22% increase in sialomucins. Hence, the obtained results suggest that ebrotidine-induced changes in the mucus gel characteristics resemble those evoked by such typical mucosal strengthening agents as sucralfate, colloidal bismuth subcitrate and geranylgeranylacetone (Bilski et al., 1989; Slomiany et al., 1989; Piotrowski et al., 1990). Like these mucosal strengthening agents, ebrotidine also caused significant increase in mucus gel viscosity, hydrogen ion retardation capacity, and hydrophobicity. Thus, in contrast to common H2-antagonists, ranitidine and cimetidine, ebrotidine is capable of preservation of the first line of mucosal defense by affecting these parameters of gastric mucus integrity. The improvement in physical qualities of gastric mucus gel evoked by ebrotidine apparently resulted from the elevated levels of glycoproteins, glycolipids and phospholipids as the animals pretreated with this agent showed significant increase in the content of these mucus constituents. Mucus lipids, and phospolipids in particular, in gastric lumen environment are known to form strong heterotypic complexes with mucin (Slomiany et al., 1987a, 1989) that protects the glycoprotein from excessive degradation by proteolytic enzymes, thus assuring the preservation of the polymeric structure of gastric mucin. The involvement of phospholipids in such essential functions of mucus gel as modulation of its viscosity, retardation of acid diffusion, and the maintenance of mucosal surface hydrophobicity have also been reported (Murty et al., 1984; Bilski et al., 1987; Gwozdzinski et al., 1988; Piotrowski et al., 1990). Ebrotidine, furthermore, successfully counteracted the untoward effects induced in mucus gel physicochemical characteristics by indomethacin. This clearly suggests that the mucosal strengthening action of this agent involves other than prostaglandin mechanism. Studies with other mucosal strengthening gastroprotective drugs such as sucralfate, colloidal bismuth subcitrate, and nitecapone provided evidence that their mode of action also occurs through venues
726
B.L. SLOMIANYet al.
different than the prostaglandin mediation and may involve receptor mediated processes (Waisman et al., 1988; Piotrowski et al., 1990; Slomiany et al., 1991). Sucralfate and colloidal bismuth subcitrate were shown to stimulate gastric mucosal phosphoinositide turnover even in the presence of indomethacin (SIomiany et al., 1990a,b), thus pointing towards the role of protein kinase C in the enhancement in mucus production, while the increase in mucus secretion evoked by nitecapone results from its inhibitory action on catechol-O-methyltransferase (Nissinen et al., 1988). The fact that ebrotidine could elicit such significant changes in the physicochemical qualities of gastric mucus gel so rapidly indicates its effect on mucus secretion, since the de novo synthesis requires much longer periods (Takagi et al., 1986). Indeed, studies show that in the in vitro system, gastric mucosal cells respond to a variety of mediators in rapid release of mucus glycoproteins and phospholipids (Seidler and Sewing, 1989; Sengupta et al., 1991). Acknowledgements--This work was supported by USPHS grant DK21684-15 from the National Institute of Diabetes and Digestive and Kidney Diseases and grant AA05858-10 from the National Institute of Alcohol Abuse and Alcoholism, NIH. REFERENCES
Angiada L., Marquez M., Sacristan A. and Ortiz J. A. (1988) lnhibitors of gastric acid secretion: N-sulphonyl formamidines in a series of new histamine H2-receptor antagonists. Fur. J. Med. Chem. 23, 97-100. Bilski J., Murty V. L. N., Aono M., Moriga M., Slomiany A. and Slomiany B. L. (1987) Enhancement of lipid content and physical properties of gastric mucus by geranylgeranylacetone. Biochem. Pharmac. 36, 4059-4065. Bilski J.. Murty V. L. N., Nadziejko C., Aono M., Moriga M., Slomiany A. and Slomiany B. L. (1988) Protection against alcohol-induced gastric mucosal injury by geranylgeranylacetone: effect of indomethacin. Digestion 41, 22-33. Corne S. J., Morrissey S. M. and Woods R. J. (1974) A method for the quantitative estimation of gastric barrier mucus. J. Physiol., Lond. 242, II6P. Gwozdzinski K., Slomiany A., Nishikawa H., Okazaki K. and Slomiany B. L. (1988) Gastric mucin hydrophobocity: effects of associated and covalently bound lipids, proteolysis, and reduction. Biochem. lnt. 17, 907-917. Kerss S., Allen A. and Garner A. (1982) A simple method for measuring thickness of the mucus gel layer adherent to rat, frog and human gastric mucosa: influence of feeding, prostaglandin, N-acetylcystein and other agents. Clin. Sci. 63, 187-195. Konturek S. J., Brzozowski T., Drozdowicz D. and Majka J. (1992) Ebrotidine, a novel H2-receptor antagonist with local gastroprotective activity. Fur. J. Gastroent. Hepat. 3, 941-947. Ligumsky M., Hansen D. G. and Kauffman G. L. (1982) Salicylic acid blocks indomethacin and aspirin induced cyclo-oxygenase inhibition in rat gastric mucosa. Gastroenterology 83, 1043-1046. Lowry O. H. and Tinsley I. J. (1974) A simple sensitive method for lipid phosphorus. Lipids 9, 491-492. Lowry O. H., Rosebrough N. J., Far A. A. and Randall R. J. (1951) Protein measurement with Folin phenol reagent. J. biol. Chem. 193, 265-275. Murty V. L. N., Sarosiek J., Slomiany A. and Slomiany B. L. (1984) Effect of lipids and proteins on the viscosity
of gastric mucus glycoprotein. Biochem. biophys. Res. Commun. 121, 521-529. Nissinen E., Linden I. B., Schultz E., Kaakkola S., Mannisto P. T. and Pohto P. (1988) Inhibition ofeatechol-Omethyltransferase activity by two novel disubstituted catechols in the rat. Fur. J. Pharmac. 153, 263-269. Palop D. and Marquez M. (1991) Ebrotidine Pharmacology Investigation's Manual. Grupp Ferrar, Barcelona. Piotrowski J., Bilski J,, Nishikawa H., Slomiany A. and SIomiany B. L. (1990) Enhancement in gastric mucus gel qualities with colloidal bismuth subcitrate administration. Eur. J. Pharmac. 184, 55-63. Robert A. (1981 ) Current history of cytoprotection. Prostaglandins 21, 89-96. Seidler U. and Sewing K. F. (1989) Ca2+-dependent and independent secretagogue action on gastric mucus secretion in rabbit mucosal explants. Am. J. Physiol. 19, G739-G746. Sengupta S., Piotrowski E., Slomiany A. and Slomiany B. L. (1991) Role of adrenergic and cholinergic mediators in gastric mucus phospholipid secretion. Biochem. Int. 24, 1145-1153. Silen W. 0987) Gastric mucosal defense and repair. In Physiology of the Gastrointestinal Tract (Edited by Johnson L. R.), Vol. 22, pp. 1055-1069. Raven Press, New York. Slomiany B. L. and Slomiany A. (1991) Role of mucus in gastric mucosal protection. J. Physiol. Pharmac. 42, 149--163. Slomiany B. L., Laszewicz W. and Slomiany A. (1986) Effect of sucralfate on the viscosity of gastric mucus and the permeability to hydrogen ion. Digestion 33, 146-151. Slomiany B. L., Sarosiek J. and Slomiany A. (1987a) Gastric mucus and the mucosal barrier. Digestive Dis. 5, 125-145. Slomiany B. L., Zdebska E. and Slomiany A. (1984) Structural characterization of neutral oligosaccharides of human H+L¢ b+ gastric mucin. J. biol. Chem. 259, 2863-2869. Slomiany B. L., Murty V. L. N., Piotrowski J. and Slomiany A. (1989) Effect of antiulcer agents on the physicochemical properties of gastric mucus. In Mucus and Related Topics (Edited by Chantler E. and Ratcliffe N. A.), pp. 179-191. Company Biologists Ltd., Cambridge. Slomiany B. L., Piasek A., Sarosiek J. and Slomiany A. (1985) The role of surface and intracellular mucus in gastric mucosal protection against hydrogen ion. Scand. J. Gastroent. 20, 1191-I 196. Slomiany A., Jozwiak Z., Takagi A. and Slomiany B. L. (1984) The role of covalently bound fatty acids in the degradation of human gastric mucus glycoprotein. Archs Biochem. Biophys. 229, 560-567. Slomiany A., Mizuta K., Piotrowski J., Nishikawa H. and Slomiany B. L. (1990b) Gastric mucosal protection by sucralfate involves phosphoinositides participation. Int. J. Biochem. 22, 1179-1183. Slomiany B. L., Wang X., Palecz D., Okazaki K. and Slomiany A. (1990a) Participation of phosphoinositides in gastric mucosal protection by colloidal bismuth subcitrate against ethanol-induced injury. Alcohol. elin. exp. Res. 14, 580-583. Slomiany B. L., Kosmala M., Carter S. R., Konturek S. J., Bilski J. and Slomiany A. (1987b) Intestinal release of mucin in response to HCI and taurocholate: effect of indomethacin. Comp. Biochem. Physiol. 87, 657-663. Slomiany B. L., Piotrowski J., Ismail A., Rajiah G., Tamura S., Bielanski W. and Slomiany A. (1991) Protection against alcohol-induced gastric mucosal injury by nitecapone. Gen. Pharmac. 22, 1055-1062. Szabo S., Pihan G. and Triers S. (1986) Alteration in blood vessels during gastric injury and protection. Scand. J. Gastroenterol. 21, 92-96.
Ebrotidine and mucosal protection Takagi A., Slomiany B. L., Kosmala M. and Slomiany A. (1986) Changes in mucus glycoprotein synthesized in rat gastric mucosa exposed to ethanol. Biochim. biophys. Acta
884, 1-10. Waisman Y., Zahavi I., Marcus H., Ligumsky M., Rosenbach Y. and Dianari G. (1988) Sucralfate is protec-
727
tive against indomethacin-induced intestinal ulceration in the rat. Digestion 41, 78-82. Yuonan F., Pearson J., Allen A. and Venables C. (1982) Changes in the structure of the mucus gel on the mucosal surface of the stomach in association with peptic ulcer. Gastroenterology 82, 827-831.
0306-3623/92 $5.00 + 0.00 Copyright ~, 1992 Pergamon Press Lid
MECHANISM OF EBROTIDINE PROTECTION AGAINST GASTRIC MUCOSAL INJURY INDUCED BY ETHANOL B . L . SLOMIANY,J. PIOTROWSKI,V. L. N. MURTY and A. SLOMIANY Research Center, New Jersey Dental School, University of Medicine and Dentistry of New Jersey, University Heights, II0 Bergen Street, Newark, NJ 07103-2400, U.S.A. [Tel. (201)456-7052] (Received 12 November 1991) A~tract--l. The gastroprotective properties of a new H2-receptor antagonist, ebrotidine, against ethanol-induced mucosal injury was investigated. 2. Groups of rats, with and without indomethacin pretreatment, received intragastrically either a dose of ebrotidine or vehicle only, followed by ethanol given at various intervals up to 4 hr. The gastric mucosa, 30 min after the ethanol challenge, was then subjected to macroscopic and histologic examination, and physic*chemical measurements. 3. Ebrotidine at doses of 50 mg and higher per kg body wt effectively prevented the alcohol-induced mucosal injury, even in the presence of indomethacin. The protective effect was demonstrable already at 30 min, reached maximum at I hr, and persisted up to 3 hr. 4. Physic,chemical analyses established that ebrotidine elicited 30% increase in mucus gel dimension, caused 19-20'/, increase in glycolipids and phospholipids, and evoked 21'/, increase in sulfomucin and 180 in sialomucins. As a consequence, the mucus gel viscosity increased by 1.4-fold, H + retardation capacity by 16°/,, and hydrophobicity by 65%. 5. The results demonstrate that ebrotidine is a unique H2-antagonist endowed with a remarkable mucosal strengthening capability.
INTRODUCTION
through strengthening the mucus gel physic*chemical characteristics.
Mucus coat overlying the epithelial surfaces of alimentary tract constitutes an essential element of mucosal protection. In the stomach, this viscous and slimy layer, together with cell membrane of gastric epithelium and the vasculature, are considered as major components of the defense mechanism responsible for the maintenance of gastric mucosal integrity under adverse environment of the lumen (Szabo et al., 1986; Silen, 1987; Slomiany et al., 1989; Slomiany and SIomiany, 1991). The integrity and strength o f mucus layer defense perimeter depends upon a delicate balance controlled by factors affecting the synthesis secretion and breakdown of its protein, glycoprotein, and lipid components (Yuonan et al., 1982; Bilski et al., 1987; Slomiany and SIomiany, 1991). While this balance is successfully maintained under normal physiological conditions, injury to the mucosa ensues when aggressive forces overcome factors that control mucosal defense. Indeed, the weakening o f gastric mucosal defense is intimately associated with the diminished viscoelastic qualities of mucus, decreased hydrogen ion retardation capacity and hydrophobicity, and excessive peptic erosion (Yuonan et al., 1982; Bilski et al., 1988; Piotrowski et al., 1991). Therefore, gastroprotective agents capable o f strengthening the protective qualities of mucus gel through the enhancement o f its physic*chemical properties are of value in peptic ulcer therapy. Here we present evidence that ebrotidine, a new H2-receptor antagonist (Anglada et al., 1988), in addition to its antisecretory activity also exhibits a remarkable ability o f gastric mucosal protection against ethanol-induced injury
MATERIALS AND METHODS Animals The study was conducted with male Sprague-Dawley rats weighing 160-180g. They were maintained on a regular chow diet and fasted in individual wire-bottom cages for 24 hr prior to experimentation. Water was withheld 2hr before the experiment. The animals received intragastric pretreatment with 1 ml ofebrotidine as emulsion in 5% gum arabic in a dose of 10-150mg/kg body wt, or vehicle consisting of 5% gum arabic in saline followed by 1 ml of absolute ethanol given at various time intervals up to 4 hr after the pretreatment. The animals were sacrificed 30 rain after the ethanol dose and their stomachs removed. The dissected stomachs were rinsed with tap water, opened along the greater curvature, gently washed with cold saline, and subjected to macroscopic and histologic assessment. Such dissected stomachs were also used for the measurements of the mucus coat thickness, evaluation of the adherent mucus gel mucin content and the chemical analysis of mucus components. All experiments were carried out in the presence or absence of pretreatment with indomethacin (5 mg/kg) given i.p. 30 min prior to ebrotidin¢ or vehicle (5% gum arabic in saline). Macroscopic and histological assessment For macroscopic assessment, opened stomachs were laid flat on a Teflon board and the areas of gross hemorrhagic necrosis were quantitated planimetrically. The results of analysis were expressed as a percentage of the total glandular mucosal area occupied by necrotic lesions. For mucosal histology, stomachs were cut obliquely through the entire extent of glandular mucosa into strips 5mm in width. Specimens were fixed in 10% buffered 719
720
B.L. SLOMIANYet al.
formalin and stained with hematoxylin and eosin. The disruption of the continuity of the surface epithelial layer and the extent of deep histologic necrosis were quantitated by microscopic evaluation of the stained sections of the mucosa, and expressed as a percentage of total mucosal length of each studied strip. Mucus coat thickness
Sections of gastric mucosa, luminal surface up, were mounted on a Millipore filter base and cut into strips 1.6 mm in width. Except for being sectioned, the tissue was kept immersed in 0.15 M NaCI solution. Cut strips were mounted transversely in Petri dishes and their positioning checked stereoscopically (x2.5 magnification) to ensure they were not mounted obliquely. Such prepared sections of gastric mucosa were subjected to mucus coat thickness measurement by means of an inverted microscope ( x 100 magnification). The distance between the bathing solutionmucus layer interface and mucus-glandular mucosa epithelial surface interface was recorded with the aid of eyepiece graticule (Kerss et al., 1982; Bilski et al., 1988). Mucin o f adherent mucus
The glandular segments of freshly dissected stomachs were excised, weighed and placed in 10 ml of 0. I% Alcian blue solution containing 0.16 M sucrose in 0.05 M citrate, pH 2.0 for sulfomucin measurements or 0.05 M sodium acetate buffer, pH 5.8 for sulfo- and sialomucins. After 2 hr of staining, the excess dye was removed by soaking the segments in 0.15 M sucrose (Come et al., 1974). The dye complexed with mucus adherent to the gastric walls was extracted from the mucosa with 10 ml of 0.5 MgCI 2, diethyl ether was added, the resulting emulsion was centrifuged, and the separated aqueous layer was used to determine the dye concentration by spectrometry (Bilski et al., 1988). Adherent mucus isolation
The gastric mucus gel used for physicochemical analyses was obtained by blotting the individual opened stomachs with Whatman No. 3 filter paper (Piotrowski et al., 1990). The adherent mucus, transferred on to the filter paper, was then recovered by washing the paper with 1.0 M NaCI in 0.05 M sodium phosphate buffer, pH 7.0. Such isolated mucus was filtered through a Millipore HA (0.45 gm) filter, subjected to intrinsic pepsin inactivation (pH 9.0 at 37°C for 30min), dialyzed against distilled water and lyophilized. Hydrogen ion retardation capacity
The diffusion of H ÷ through gastric mucus gel was measured in a specially constructed permeability chamber (Bilski et al., 1987). Individual samples of mucus, dissolved at 30 mg/ml in 0.15 M NaC1, were placed in the diffusion port separating the two compartments, filled on one side with 0.15 N NaC1 and on the other side with 0.15 M HCI, and the change in pH in the NaC1 compartment was recorded at 5 rain intervals for up to 2 hr with a micro-pH electrode connected to an Accumet recording ionalyzer. The amount of H ÷ diffusion through the sample was calculated in mol/sec and the permeability coefficient in cm/sec (Piotrowski et al., 1990).
Mucus gel hydrophobicity
The hydrophobicity of the adherent gastric mucus gel was evaluated using a fluorescent hydrophobic probe, bis(8anilino-l-naphthalenesulfonate) (bis-ANS). The measurements were conducted with a Perkin-Elmer model LS-5 fluorescent spectrophotometer (Gwozdzinski et al., 1988). The excitation wavelength for the probe was 365 nm and the maximum emission was observed at 530 nm. The samples, dissolved at 2mg/ml in 0.10 M NaCI/0.05 sodium phosphate buffer, pH 7.0, were reacted at 25~C with increasing amounts of bis-ANS and the induced alterations in the emission spectra of the probe were evaluated (Gwozdzinski et al., 1988). Analytical methods
The content and composition of carbohydrate in the prepared mucus samples were determined by gas-liquid chromatography following methanolysis, re-N-acetylation and derivatization with silylating reagent (Slomiany et al., 1984), and the protein was measured by the method of Lowry et al. (1951). For the analysis of lipids, individual samples of mucus powder were extracted with chloroformmethanol (Slomiany et al., 1985). The extracts were filtered through a grade F sintered glass funnel, and the lipids contained in the filtrates were separated on silicic acid columns into neutral lipids, glycolipid and phospholipid fractions (Bilski et al., 1987). The neutral lipids were separated into individual components by thin-layer chromatography, identified by comparison with chromatograms of authentic standards, and quantitated (Bilski et al., 1987, 1988). The glycolipids were quantitated by measuring their carbohydrate and lipid constituents (Slomiany et al., 1985). The phospholipids were subjected to two-dimensional thinlayer chromatography and quantitated by measuring their phosphorus content (Lowry and Tinsley, 1974). For the analysis of covalently bound fatty acids, the delipidated mucus preparations were incubated for 30 min at 37°C with 0.3 M methanolic NaOH, and the released fatty acid methyl esters were extracted with hexane and quantitated by gas-liquid chromatography (Slomiany et al., 1987b). Antiulcer drug
Ebrotidine, p-bromo-N-[[[2-[[[2-[(diaminomethylene)amino]4-thiazolyl]-methyl]-thio]-ethyll-amino]-methylene]benzene sulfonamide, lot No. D-6, was kindly donated by Ferrer International, S.A., Barcelona, Spain. The drug was stored at 4'C in the dark and was suspended in 5% gum arabic-0.9% NaC1 shortly before experimentation. The drug or vehicle (5% gum arabic in saline) were given orally in a volume of I ml through a dull metal tubing attached to a 2 ml syringe. Data analysis
All experiments were carried out in duplicate, and the results are expressed as means + SD. Student's t-test was used to determine significance, and P values of 0.05 or less were considered significant.
RESULTS
Viscosity measurements
G a s t r i c lesions a n d m u c o s a l h i s t o l o g y
Viscosity determinations were performed with a Brookfield cone/plate digital viscometer, model LVTDCP, equipped with a 1.565 ° cone and a constant (37°C) temperature bath (Slomiany et al., 1986). Shear rates were varied from 1.15 to 115 sec - t, and the sample volumes were 0.5 ml. For the measurements, samples of mucus, dissolved at 30 mg/ml in 0.10 M NaCI, 0.05 M sodium phosphate buffer, pH 6.8, were gently stirred at 4°C for 1 hr and then brought to 37°C. To calculate the specific viscosity, measurements were also taken of buffer alone.
The gross a p p e a r a n c e of gastric m u c o s a 30 min following intragastric a d m i n i s t r a t i o n o f absolute e t h a n o l in the absence a n d the presence o f pretreatm e n t with ebrotidine is depicted in Fig. 1. E t h a n o l a d m i n i s t r a t i o n to the animals given vehicle p r o d u c e d extensive necrotic h e m o r r h a g i c lesions in the glandular p o r t i o n o f the mucosa, b u t little, if any, effect was observed in the mucosal a p p e a r a n c e of the forestomach. The lesions in the glandular area were in the
Ebrotidine and mucosal protection
"1~ W
721
f
Fig. I. Gross appearance of rat gastric mucosa 30 rain after intragastfic dose of 1ml absolute ethanol.
Vehicle pretreated animals in the absence (A) and presence (B) of indomethacin. Ebrotidine pretreated (75 mg/kg) animals in the absence (C) and presence (D) of indomethacin.
form of red streaks, parallel to the long axis of the stomach and occupied over 40% of the mucosa (Fig. 2). Pretreatment with ebrotidine at doses of 50 mg/kg and above significantly reduced the extent of mucosal necrosis caused by ethanol, and this inhibitory effect of ebrotidine was not affected by indomethacin given i.p. 30 min prior to the ebrotidine (Figs 1 and 2). The histology of gastric mucosa 30 rain following intragastric administration of ethanol in the absence and the presence of pretreatment with ebrotidine is presented in Fig. 3. The results indicated that in the vehicle-treated rats, ethanol caused extensive disruption and desquamation of the surface epithelium, and produced rampant necrotic lesions penetrating deeply into the mucosa (Fig. 3, Table 1). Submucosal edema and polymorphonuclear leukocyte infiltration were also present outside the deep hemorrhagic lesions. Also, in the areas of deep hemorrhagic necrosis mucosal microvessels were often obstructed with erythrocytes or ruptured. Administration of ebrot-
idine at doses of 50 mg/kg and above markedly reduced (81%) the disruption of the surface epithelium, and the number of deep necrotic lesions decreased by 90%. Furthermore, the microvascular lesions were rare and essentially confined to the luminal surface mucous cells. Pretreatment with indomethacin prior to ebrotidine increased the extent of histologic necrotic lesions evoked by ethanol, but did not affect significantly the extent of ebrotidine elicited protection. Dose and duration The effect of the dose on the gastric mucosal protective effect of ebrotidine against ethanol injury is presented in Fig. 4. The results indicated that ebrotidine successfully prevented gastric mucosal damage by ethanol at doses as low as 75-50 mg/kg of body wt, at which concentration 89.1 and 75.8% reduction in mucosal necrosis occurred (Fig. 4). The data on the duration of gastroprotective action of ebrotidine against ethanol injury are summarized
722
B. L. SLOMIANY et al.
_
50
T
~. 4 0 30 o o
20 --~
÷
10
g Z
0
Vehicle
VQhlchl * Ethmal
"
T
1
EI~ro t Id Ine I m l ~ n e l h l ~ Ina~mlho¢~ I r . ~ , ~ " ~ ¢ m * + * + ~'tl~nol ¥~le Ethon~ Ebrotldlm + Etemr~
Fig. 2. Effect of ebrotidine on macroscopic necrosis of gastric mucosa caused by ethanol in the absence and presence of pretreatment with indomethacin. Ebrotidine (75 mg/kg) or vehicle were administered 60 min prior to ethanol dose. Values represent the means + SD of analyses performed on 10 animals in each group. The effect of ebrotidine + ethanol, and ebrotidine + indomethacin and ethanol vs vehicle + ethanol, and vehicle+ indomethacin and ethanol were significant at "P < 0.05. in Fig. 5. A significant protective effect was demonstrable already at 30 rain, reaching maximum at l hr, and was still observed at 3 hr. Adherent mucus and its mucin content
Examination of gastric mucosal surface by means of inverted microscope technique (Kerss et al., 1982) indicated the presence of three distinct regions under light-and dark-field illumination. These were identified as bathing solution, layer of mucus gel, and the glandular mucosa. By examining seven sections taken from glandular mucosa of each animal used in the study, it was found that in the group treated with vehicle, the mucus gel dimension averaged 223.5 gin, while its thickness in the animals treated with ebrotidine was 290.0/~m, which represents a 30% increase over the control value. Pretreatment with indomethacin caused a 47% reduction in the gel dimension of animals receiving the vehicle, while the animals receiving ebrotidine showed a 98% increase in mucus dimension over that of the indomethacin controls (Table 2). Table 1. Effectof ebrotidineon histologyof gastric mucosaexposed to ethanol Disruption of Deepnecrotic surface epithelium lesions Pretreatment (%) (%) Vehicle (5% gum arabic in 0.15M NaCI) 0 0 Vehicle+ ethanol 95.4 ± 4.5 46.2 + 4.3 Ebrotidine (75 mg/kg) in vehicle 0 0 Ebrotidine (75 mg/kg) + ethanol 18.4 ± 2.7* 4.3 + 0.3* Indomethacin + vehicle + ethanol
93.9 ± 5.3
48.6 + 3.8
lndomethacin+ ebrotidine + ethanol lndomethacin+ vehicle
32.3 ± 3.9* 7.3 ± 0.5* 3.5 ± 0.4 1.3 ± 0.3 Values represent the means ± SD of analyses performed on I0 animals in each group. Ebrotidineor vehicle were administered 60rainprior to ethanol dose. The effectsof ebrotidine+ ethanol, and ebrotidine + indomethacin and ethanol vs vehicle + ethanol, and vehicle + indomethacin were significant at "P < 0.05.
Evaluation of the adherent mucus gel mucin content by the Alcian blue uptake assays, conducted at pH 2.0 for sulfomucins and at pH 5.8 for sulfo- and sialomucins, indicated that in the absence of ebrotidine treatment, indomethacin evoked 37% decrease in sulfomucin and 41% decrease in sialomucin contents of the mucus gel (Table 2). In comparison to the indomethacin pretreated animals receiving vehicle, the mucus gel of animals exposed to ebrotidine showed 46% increase in the content sulfomucin and 22% increase in sialomucins. In the absence of indomethacin pretreatment, ebrotidine caused 21% increase in gastric mucus gel sulfomucin content and 18% increase in the content of sialomucin. Chemical composition
The chemical composition of the adherent mucus gel elaborated in the absence and presence of pretreatment with indomethacin, without and with the administration of ebrotidine is presented in Table 3. In the absence of pretreatment with indomethacin, the mucus of the animals receiving ebrotidine showed a significantly higher content of phospholipids (20%), glycolipids (19%) and glycoprotein (19%), and only negligible lower (8%) content of protein. Pretreatment with indomethacin in the absence of ebrotidine led to a 14% increase in mucus gel protein, 15% decrease in glycoprotein, and a 22% decrease in total lipids, reflected mainly in a marked decrease in phospholipids. The changes evoked by indomethacin in the composition of elaborated mucus were prevented by ebrotidine treatment, and the chemical characteristics of gastric mucus of ebrotidine-treated animals were essentially similar to that of the group which received vehicle in the absence of indomethacin. Hydrophobicity, viscosity and H + retardation
The data on the effect of ebrotidine administration on the hydrophobicity of the adherent gastric mucus are presented in Fig. 6. The results revealed that the amount of hydrophobic fluorescent probe, bis-ANS, required to obtain maximum fluorescence intensity with mucus gel of animals treated with ebrotidine was 65% greater than with gastric mucus gel of the group receiving the vehicle. This effect of ebrotidine, furthermore, was not thwarted by indorfiethacin pretreatment (Fig. 6). Figure 7 shows the effect of ebrotidine administration on the viscosity of the adherent gastric mucus gel. The specific viscosity of mucus from the vehicle group ranged from 3.6 at the shear rate of 1.15 sec-' to 1.9 at the shear rate of 115 sec -~. The specific viscosity of mucus gel from animals treated with ebrotidine dose of 75 mg/kg body wt ranged from 5.1 at the lowest shear rate to 3.4 at the highest shear rate, while the viscosity values for ebrotidine in vehicle alone ranged only from 0.3 to 0.1. Pretreatment with indomethacin in the absence of ebrotidine caused a 45% drop in mucus gel viscosity, while in the present of ebrotidine, the viscosity values at the shear rate of i.15 sec-' increased to 3.7 and to 2.3 at the shear rate of 115 sec-z. The effect of ebrotidine administration on the permeability of the adherent gastric mucus gel to hydrogen ion is summarized in Table 4. The H ÷
Ebrotidine and mucosal protection
123
724
B. L. SLOMIANY et al.
Table 2. Effect of ebrotidine on gastric mucus gel dimension and its content of mucin
-
5o
o
Mucin content (/~g Alcian blue/g tissue) Pretreatment Vehicle lndomethacin + vehicle Ebrotidine (75 mg/kg) Indomethacin + ebrotidine (75 mg/kg)
Mucus gel dimension (~m)
pH 2.0
pH 5.8
223.5 -+ 8.9
93.9 + 4.8
148.0 ~ 6.3
o
40
~
30
o
117.5-+9.3
59.5+3.1
91.2+_4.7
290.0 -+ 9.2*
114.2 -+ 4.6*
178.1 _+ 6.2*
232.5 + 7.5*
86.7 -+ 4.3*
140.6 + 5.5*
~- 20 o
Values represent the means_+ SD of analyses performed on l0 animals in each group. The assays were carried out 60 min following ebrotidine (75mg/kg). The effects of ebrotidine vs vehicle and ebrotidine +indomethacin vs vehicle+ indomethacin were significant at *P < 0.05.
retardation capacity of mucus samples from the animals treated with ebrotidine was 16% higher than that of the vehicle group. Indomethacin pretreatment caused a 12% decrease in mucus H ÷ retardation capacity in the absence of ebrotidine treatment, while in the presence of ebrotidine the mucus H ÷ retardation values were quite similar to those obtained with the vehicle in the absence of indomethacin.
._~ 10 b. 1
Z
0
I
l
30 Dose
J"
6O
T I J 90
t20
In~/kg)
Fig. 4. Effect of ebrotidine dose (mg/kg) on the gastroprotective potential against ethanol-induced injury. Values represent the means + SD of analyses performed on 7 animals with each dose of ebrotidine. 5o I o
*°l
o
30 o
DISCUSSION
o
Ebrotidine is a new H2-receptor antagonist with antisecretory potency comparable to that of ranitidine also showing some activity on H~-receptors (Anglada et al., 1988; Konturek et al., 1992). Structurally, although this agent shares common features with cimetidine and ranitidine, it contains N-sulfonyl formamidine group instead of N-cyanoguanidine group of cimetidine and 2-nitroethendiamine group of ranitidine, while the imidazole ring of cimetidine substituted by guanidinothiazole (Anglada et al., 1988). These modifications endow ebrotidine with diminished cytochrome P-450 binding as compared to that of ranitidine and cimetidine, and eliminate the potential for mutagenic nitros.amine formation. Furthermore, the results of a recent study suggest that ebrotidine, in contrast to ranitidine and cimetidine, exhibits remarkable gastroprotection against ethanol induced mucosal injury. This effect of ebrotidine apparently does not seem to involve prostaglandin mechanism as the gastroprotection is maintained
o
2O
m 'tO 'N
Z
0
1
I
I
I
I
30
90
150
210
270
T i m e (rain)
Fig. 5. T h e d u r a t i o n o f g a s t r o p r o t e c t i v e effect o f e b r o t i d i n e a g a i n s t e t h a n o l - i n d u c e d injury. E a c h p o i n t r e p r e s e n t s the m e a n s + S D o f a n a l y s e s p e r f o r m e d o n 7 a n i m a l s a t different time intervals following pretreatment with ebrotidine (75 m g / k g ) .
even in the presence of such potent prostaglandin synthesis inhibitors as acetylsalicylic acid and indomethacin (Palop and Marquez, 1991). The data obtained from this investigation show that the gastroprotective effect of ebrotidine against ethanol-induced injury results from the drug's ability
Table 3. Effect of ebrotidine on the chemical composition of gastric mucus gel Pretreatment Constituent (rag/100 mg mucus) Protein Carbohydrate Lipids (total) Neutral lipids Glycolipids Phospholipids Covalently bound fatty acids
Vehicle
indomethacin + vehicle
Ebrotidine
lndomethacin + ebrotidine
65.8+_6.1 11.4 + 0.9 19.6+ 1.8 11.5 -+ 1.4 4.7 -+ 0.5 3.4 -+ 0.2
74.9_+6.2 9.7_+ 1.0 15.2± 1.6 8.9 +- 1.0 4. I -+ 0.4 2.2 -+ 0.2
60.7+5.0 13.6+ 1.2" 22.1 _+2.0 12.4 4-_ 1.5 5.6 ± 0.5* 4. I + 0.4*
62.5-,-4.9* 12.0_+ 1.3' 21.6+2.1" 12.2 + [.3" 5.5 _+ 0.5* 3.8 -+ 0.5*
0.3-'-0.1
0.2+_0.1
0.4-+0.2
0.3+0.2
Values represent the means + SD of analyses performed on the individual samples of mucus obtained from 10 animals in each group subjected to pretreatment for 60 min with ebrotidine (75 mg/kg) or vehicle. The effects of ebrotidine vs vehicle and indomethacin plus ebrotidine vs indomethacin plus vehicle were significant at *P < 0.05.
725
Ebrotidine and mucosal protection
Table 4. Effect of ebrotidine on the permeability of the adherent gastric mucus gel to hydrogen ion
80 II
T
T
6O
Pretreatment
>-
Vehicle (5% gum arabic in 0.15M NaC1) Indomethacin + vehicle Ebrotidine (75 mg/kg) + vehicle Indomethacin + ebrotidine (75 mg/kg)
T
Vehicle
Ebrotldlml
|ndomethoctn I n d o m ~ h c ] e l n + +
Vehicle
Ebrotldlne
Fig. 6. Effect of intragastric ebrotidine (75 mg/kg) administration on the hydrophobicity of gastric mucus. Values are plotted as means + SD (n = 10) of the increase in relative fluorescence of hydrophobic probe (bis-ANS) due to mucus binding. Asterisk indicates a significant (P < 0.05) increase in hydrophobicity over the control value. to enhance the physicochemical qualities of mucus layer perimeter of gastric mucosal defense. Moreover, ebrotidine has the ability to maintain the physicochemical qualities of this layer even in the presence of indomethacin when prostaglandin synthesis is inhibited. Our results demonstrated that the gastroprotective effect of ebrotidine is dose dependent, and could be achieved with drug concentrations as low as 75-50 mg/kg of body weight. The time-course studies of the gastroprotective effect indicated that ebrotidine elicits maximal protective effect 30-60 rain following administration, and that the effect persists for at least up to 3 hr. The observed gastric mucosal protective action of ebrotidine is apparently exerted through enhancement of physical and chemical characteristics of mucus gel. Experiments with animals pretreated with prostaglandin-inhibitory dose of indomethacin (Robert, 1981; Ligumsky et al., 1982) revealed that 5 -
T
I!
~3 "G 2 (n
Vehicle
El=cot Idlntl
I n d o m e t h ~ Indem~.hacln Vehicle
[brotldlne
Fig. 7. Effect of intragastric ebrotidine (75 mg/kg) administration on the viscosity of gastric mucus. Values represent the means + SD of duplicate anlayses performed on the individual samples of mucus obtained from 10 animals in each group. Shear rate 11.5 sec -t. The changes in mucus viscosity evoked by the drug were significant at P < 0.05.
Permeability (tool. 10 ~0.sec-' )
Permeability coefficient (cm. 10- 6 .see- ] )
4.21 +0.31 4.73 _+ 0.38
7.05+_0.56 7.90 _+ 0.62
3.53 +_ 0.27*
5.92 _+ 0.38*
4.28 _+ 0.25*
7.17 _+ 0.34*
Values represent the means + SD of analyses performed on individual samples of mucus obtained from 10 animals in each group. The assays were carried out 60min following ebrotidine (75 mg/kg). The effects of ebrotidine vs vehicle and ebrotidine + indomethacin vs vehicle + indomethacin were significant at *P < 0.05.
ebrotidine successfully prevented the detrimental effect of indomethacin on the adherent mucus gel dimension and its content of sulfo- and sialomucins. In comparison to the indomethacin pretreated animals receiving vehicle, the mucus coat of animals treated with ebrotidine showed 98% increase in gel thickness, 46% increase in the content of sulfomucins, and a 22% increase in sialomucins. Hence, the obtained results suggest that ebrotidine-induced changes in the mucus gel characteristics resemble those evoked by such typical mucosal strengthening agents as sucralfate, colloidal bismuth subcitrate and geranylgeranylacetone (Bilski et al., 1989; Slomiany et al., 1989; Piotrowski et al., 1990). Like these mucosal strengthening agents, ebrotidine also caused significant increase in mucus gel viscosity, hydrogen ion retardation capacity, and hydrophobicity. Thus, in contrast to common H2-antagonists, ranitidine and cimetidine, ebrotidine is capable of preservation of the first line of mucosal defense by affecting these parameters of gastric mucus integrity. The improvement in physical qualities of gastric mucus gel evoked by ebrotidine apparently resulted from the elevated levels of glycoproteins, glycolipids and phospholipids as the animals pretreated with this agent showed significant increase in the content of these mucus constituents. Mucus lipids, and phospolipids in particular, in gastric lumen environment are known to form strong heterotypic complexes with mucin (Slomiany et al., 1987a, 1989) that protects the glycoprotein from excessive degradation by proteolytic enzymes, thus assuring the preservation of the polymeric structure of gastric mucin. The involvement of phospholipids in such essential functions of mucus gel as modulation of its viscosity, retardation of acid diffusion, and the maintenance of mucosal surface hydrophobicity have also been reported (Murty et al., 1984; Bilski et al., 1987; Gwozdzinski et al., 1988; Piotrowski et al., 1990). Ebrotidine, furthermore, successfully counteracted the untoward effects induced in mucus gel physicochemical characteristics by indomethacin. This clearly suggests that the mucosal strengthening action of this agent involves other than prostaglandin mechanism. Studies with other mucosal strengthening gastroprotective drugs such as sucralfate, colloidal bismuth subcitrate, and nitecapone provided evidence that their mode of action also occurs through venues
726
B.L. SLOMIANYet al.
different than the prostaglandin mediation and may involve receptor mediated processes (Waisman et al., 1988; Piotrowski et al., 1990; Slomiany et al., 1991). Sucralfate and colloidal bismuth subcitrate were shown to stimulate gastric mucosal phosphoinositide turnover even in the presence of indomethacin (SIomiany et al., 1990a,b), thus pointing towards the role of protein kinase C in the enhancement in mucus production, while the increase in mucus secretion evoked by nitecapone results from its inhibitory action on catechol-O-methyltransferase (Nissinen et al., 1988). The fact that ebrotidine could elicit such significant changes in the physicochemical qualities of gastric mucus gel so rapidly indicates its effect on mucus secretion, since the de novo synthesis requires much longer periods (Takagi et al., 1986). Indeed, studies show that in the in vitro system, gastric mucosal cells respond to a variety of mediators in rapid release of mucus glycoproteins and phospholipids (Seidler and Sewing, 1989; Sengupta et al., 1991). Acknowledgements--This work was supported by USPHS grant DK21684-15 from the National Institute of Diabetes and Digestive and Kidney Diseases and grant AA05858-10 from the National Institute of Alcohol Abuse and Alcoholism, NIH. REFERENCES
Angiada L., Marquez M., Sacristan A. and Ortiz J. A. (1988) lnhibitors of gastric acid secretion: N-sulphonyl formamidines in a series of new histamine H2-receptor antagonists. Fur. J. Med. Chem. 23, 97-100. Bilski J., Murty V. L. N., Aono M., Moriga M., Slomiany A. and Slomiany B. L. (1987) Enhancement of lipid content and physical properties of gastric mucus by geranylgeranylacetone. Biochem. Pharmac. 36, 4059-4065. Bilski J.. Murty V. L. N., Nadziejko C., Aono M., Moriga M., Slomiany A. and Slomiany B. L. (1988) Protection against alcohol-induced gastric mucosal injury by geranylgeranylacetone: effect of indomethacin. Digestion 41, 22-33. Corne S. J., Morrissey S. M. and Woods R. J. (1974) A method for the quantitative estimation of gastric barrier mucus. J. Physiol., Lond. 242, II6P. Gwozdzinski K., Slomiany A., Nishikawa H., Okazaki K. and Slomiany B. L. (1988) Gastric mucin hydrophobocity: effects of associated and covalently bound lipids, proteolysis, and reduction. Biochem. lnt. 17, 907-917. Kerss S., Allen A. and Garner A. (1982) A simple method for measuring thickness of the mucus gel layer adherent to rat, frog and human gastric mucosa: influence of feeding, prostaglandin, N-acetylcystein and other agents. Clin. Sci. 63, 187-195. Konturek S. J., Brzozowski T., Drozdowicz D. and Majka J. (1992) Ebrotidine, a novel H2-receptor antagonist with local gastroprotective activity. Fur. J. Gastroent. Hepat. 3, 941-947. Ligumsky M., Hansen D. G. and Kauffman G. L. (1982) Salicylic acid blocks indomethacin and aspirin induced cyclo-oxygenase inhibition in rat gastric mucosa. Gastroenterology 83, 1043-1046. Lowry O. H. and Tinsley I. J. (1974) A simple sensitive method for lipid phosphorus. Lipids 9, 491-492. Lowry O. H., Rosebrough N. J., Far A. A. and Randall R. J. (1951) Protein measurement with Folin phenol reagent. J. biol. Chem. 193, 265-275. Murty V. L. N., Sarosiek J., Slomiany A. and Slomiany B. L. (1984) Effect of lipids and proteins on the viscosity
of gastric mucus glycoprotein. Biochem. biophys. Res. Commun. 121, 521-529. Nissinen E., Linden I. B., Schultz E., Kaakkola S., Mannisto P. T. and Pohto P. (1988) Inhibition ofeatechol-Omethyltransferase activity by two novel disubstituted catechols in the rat. Fur. J. Pharmac. 153, 263-269. Palop D. and Marquez M. (1991) Ebrotidine Pharmacology Investigation's Manual. Grupp Ferrar, Barcelona. Piotrowski J., Bilski J,, Nishikawa H., Slomiany A. and SIomiany B. L. (1990) Enhancement in gastric mucus gel qualities with colloidal bismuth subcitrate administration. Eur. J. Pharmac. 184, 55-63. Robert A. (1981 ) Current history of cytoprotection. Prostaglandins 21, 89-96. Seidler U. and Sewing K. F. (1989) Ca2+-dependent and independent secretagogue action on gastric mucus secretion in rabbit mucosal explants. Am. J. Physiol. 19, G739-G746. Sengupta S., Piotrowski E., Slomiany A. and Slomiany B. L. (1991) Role of adrenergic and cholinergic mediators in gastric mucus phospholipid secretion. Biochem. Int. 24, 1145-1153. Silen W. 0987) Gastric mucosal defense and repair. In Physiology of the Gastrointestinal Tract (Edited by Johnson L. R.), Vol. 22, pp. 1055-1069. Raven Press, New York. Slomiany B. L. and Slomiany A. (1991) Role of mucus in gastric mucosal protection. J. Physiol. Pharmac. 42, 149--163. Slomiany B. L., Laszewicz W. and Slomiany A. (1986) Effect of sucralfate on the viscosity of gastric mucus and the permeability to hydrogen ion. Digestion 33, 146-151. Slomiany B. L., Sarosiek J. and Slomiany A. (1987a) Gastric mucus and the mucosal barrier. Digestive Dis. 5, 125-145. Slomiany B. L., Zdebska E. and Slomiany A. (1984) Structural characterization of neutral oligosaccharides of human H+L¢ b+ gastric mucin. J. biol. Chem. 259, 2863-2869. Slomiany B. L., Murty V. L. N., Piotrowski J. and Slomiany A. (1989) Effect of antiulcer agents on the physicochemical properties of gastric mucus. In Mucus and Related Topics (Edited by Chantler E. and Ratcliffe N. A.), pp. 179-191. Company Biologists Ltd., Cambridge. Slomiany B. L., Piasek A., Sarosiek J. and Slomiany A. (1985) The role of surface and intracellular mucus in gastric mucosal protection against hydrogen ion. Scand. J. Gastroent. 20, 1191-I 196. Slomiany A., Jozwiak Z., Takagi A. and Slomiany B. L. (1984) The role of covalently bound fatty acids in the degradation of human gastric mucus glycoprotein. Archs Biochem. Biophys. 229, 560-567. Slomiany A., Mizuta K., Piotrowski J., Nishikawa H. and Slomiany B. L. (1990b) Gastric mucosal protection by sucralfate involves phosphoinositides participation. Int. J. Biochem. 22, 1179-1183. Slomiany B. L., Wang X., Palecz D., Okazaki K. and Slomiany A. (1990a) Participation of phosphoinositides in gastric mucosal protection by colloidal bismuth subcitrate against ethanol-induced injury. Alcohol. elin. exp. Res. 14, 580-583. Slomiany B. L., Kosmala M., Carter S. R., Konturek S. J., Bilski J. and Slomiany A. (1987b) Intestinal release of mucin in response to HCI and taurocholate: effect of indomethacin. Comp. Biochem. Physiol. 87, 657-663. Slomiany B. L., Piotrowski J., Ismail A., Rajiah G., Tamura S., Bielanski W. and Slomiany A. (1991) Protection against alcohol-induced gastric mucosal injury by nitecapone. Gen. Pharmac. 22, 1055-1062. Szabo S., Pihan G. and Triers S. (1986) Alteration in blood vessels during gastric injury and protection. Scand. J. Gastroenterol. 21, 92-96.
Ebrotidine and mucosal protection Takagi A., Slomiany B. L., Kosmala M. and Slomiany A. (1986) Changes in mucus glycoprotein synthesized in rat gastric mucosa exposed to ethanol. Biochim. biophys. Acta
884, 1-10. Waisman Y., Zahavi I., Marcus H., Ligumsky M., Rosenbach Y. and Dianari G. (1988) Sucralfate is protec-
727
tive against indomethacin-induced intestinal ulceration in the rat. Digestion 41, 78-82. Yuonan F., Pearson J., Allen A. and Venables C. (1982) Changes in the structure of the mucus gel on the mucosal surface of the stomach in association with peptic ulcer. Gastroenterology 82, 827-831.
