- Email: [email protected]
Topical isoproterenol protects the rat gastric mucosa from ethanol-induced injury
.IOtlRNAt.
OF StlRGICAL
RESEARCH
46, 640-64.5 (1989)
Topical lsoproterenol Protects the Rat Gastric Mucosa from Ethanol-Induced Injury THOMAS ,J. HOWARD, M.D., F:I)WAHI) PASSAHO, tJ~., M.D., ANL) PAIJI. H. GI!TH, M.D. Research
and Surgical Seruices, West Los Angeles (‘enter for Ulcer Research and Eunfuntion
Presented at the Annual Meeting of the Association
Press.
Administrafion Los Angeles,
Medical California
Center, 9007,‘l
and the
for Academic Surgery, Salt lake City, LJtah, November
This study evaluated the effects of topical isoproterenol, a &adrenergic agonist, on the morphologic damage produced in the gastric mucosa by ethanol. The orogastric instillation of 100% ethanol in rats resulted in gross lesion formation and deep histologic injury in the gastric mucosa. Animals pretreated with oral isoproterenol (50 pg/kg, 500 bg/kg, 50 mg/kg) showed dose-dependent protection from both the gross and the histologic mucosal injury (P < 0.01, ANOVA). Pretreatment with propran0101 (2 mg/kg/sec) but not indomethacin (5 mg/kg/sec) blocked this protective effect. Isoproterenol had no effect on ethanol-induced mast cell degranulation as both mucosal and submucosal mast cell counts were significantly and equally decreased in all groups treated with 100% ethanol (P < 0.05). These findings show that topical isoproterenol protects the rat gastric mucosa from both the gross and the histologic injury caused by 100% ethanol. This protection is mediated by a &adrenergic receptor mechanism as it can be blocked by prior treatment with propranolol. but does not involve stabilization of mucosal or submucosal mast cell membranes. c 1RR9Arudemic
Vrterans (CURE),
16-19, 1988
a number of chemical ulcerogens [7, 81. Unfortunately, the potent cardiovascular effects of intravenous isoproterenol limit its clinical usefulness in preventing gastric mucosal injury. The topical application of isoproterenol, however, could potentially avoid the cardiovascular effects of intravenous administration while still providing the beneficial drug effect locally. Topical application of isoproterenol in the rat gastric mucosa has been shown previously to diminish the histologic injury produced by 40% ethanol [9]. The purpose of this study was to confirm the protective effect of topical isoproterenol using an analysis of both the gross and the histologic mucosal injury in the absolute ethanol model of gastric mucosal injury. We were also interested in determining whether the protective effect of isoproterenol involved fl-adrenergic receptors (i.e., would be blocked by fl-adrenergic receptor antagonists), stimulation of endogenous prostaglandin synthesis (i.e., would be blocked by a cyclooxygenase inhibitor), or stabilization of mucosal mast cell membranes (i.e., prevention of mast cell degranulation). METHODS
Inc.
INTRODUCTION Absolute ethanol is a noxious chemical agent widely used in the study of gastric mucosal injury [ 1J. Knowledge of the mechanisms involved in the ethanol-induced injury model are of importance in understanding the response of the gastric mucosa to injury and in developing treatment strategies designed to protect against acute gastric mucosal damage. The ethanol-induced injury produces a number of alterations in the corpus mucosa including: mast cell degranulation [2], vascular engorgement [3], increased permeability [4], and ultimately complete microvascular stasis [5]. These physiologic alterations are accompanied by the morphologic finding of vertical, dark red areas of gross mucosal injury, necrosis, edema, vascular congestion, and mucosal hemorrhage [6]. Isoproterenol, when given systemically, has been shown to prevent the gross gastric mucosal injury produced by
Kxperimental design. Thirty-six Sprague-Dawley rats (180-250 g) were fasted overnight but allowed free access to water up until 2 hr prior to the experiment. Groups of six rats each were randomly assigned to one of six groups, weighed, and appropriately marked for later identification. Groups 1 and 2 underwent pretreatment with 1 cc Krebs Solution subcutaneously (SC), groups 3 and 4 received 2 mg/kg propranolol SC,and groups 5 and 6 received 5 mg/ kg indomethacin (5 mg/kg) SC.After 30 min, group 2, 4, and 6 rats were given 50 pg/kg isoproterenol intragastrically by orogastric gavage. Group 1, 3, and 5 rats were given 1 cc Krebs solution intragastrically. Fifteen minutes later, all rats were given 100% ethanol (1 cc/200 g) intragastrically by orogastric gavage. Thirty minutes after ethanol instillation each rat was killed and the stomach removed and examined for gross mucosal lesions as described below. The dose of 50 pg/kg was chosen as this was the dose previously shown to prevent the histologic damage produced by 40% ethanol [9]. Doses of lo- and
HOWARD,
PASSARO,
AND
GUTH:
ISOPROTERENOI,
IOOO-fold increases were used to evaluate the dose-response relationship of this drug. The experiment was therefore repeated two more times using differing doses of isoproterenol (500 pg/kg and 50 mg/kg). Materials. Absolute ethanol was purchased from Midwest Grain Products of Illinois (Pekin, IL). Isoproterenol (1[3’,4’-dihydroxyphenyll-2-isopropylaminoethanol), a potent nonspecific p-adrenergic agonist, was dissolved in Krebs solution (50 mg/5 ml, 500 pg/5 ml, or 50 pg/5 ml) for intragastric application. Propranolol hydrochloride (1-(isopropylamino]-3-[1-naphthyloxyl-2-propranolol HCl), a nonselective &adrenergic antagonist, was prepared by dissolving 40 mg in 100 ml Krebs solution. Indomethacin (1-[p-chlorobenzoyll-5-methoxy-2-methyl indole-3-acetic acid) a prostaglandin synthesis inhibitor, was prepared by completely dissolving 100 mg in 5% sodium bicarbonate by heating and stirring for 5 hr at 40°C. This was then diluted with distilled water so the final solution contained 1 mg of indomethacin in 1 ml of 1% NaHC09. Each stomach was opened along Gross lesion analysis. the greater curvature, gently rinsed in saline, pinned open to expose the mucosa, and photographed using a Polaroid pack film camera and Polaroid Type 108 film (Polaroid Tech. Photo, Cambridge, MA) at 2:l magnification. The photographs were coded for blinded lesion tracing and subsequent analysis. The outline of the gastric corpus and the lesions present (characterized by dark red bands) were traced onto transparent film using a felt tip marker. The lesion area (cm*), total area of the corpus (cm*), and percentage corpus mucosa containing lesions were determined by a computerized image analyzer device (MICRO/ PDP-11, Digital Equipment Corp., Maynard, MA, equipped with imaging boards, Imaging Technology Inc., Woburn, MA). The image analyzer program for the measurement of total area, gross lesion area, and the percentage gross lesion was written by Dr. Edward Livingston. Immediately after photography of Histologic analysis. the stomach, the tissue was placed in a basic lead-acetate fixative for 24 hr and then stored in 70% ethanol. Later, a standardized strip of tissue across the entire posterior wall of the glandular corpus just distal to the limiting ridge was removed, embedded in paraffin, sectioned (4 pm thick), and stained with either hematoxylin and eosin (histologic injury analysis) or toluidine blue (mast cell analysis). Slides were coded prior to microscopic evaluation, so the examiner (T.J.H.) was unaware of the treatment group the slide came from. Histologic injury was scored using an American Optical microscope with a 45X objective and a final magnification of 450X. The mucosal injury score was determined using modified criteria based on those previously established by Lacy and Ito [lo]. Damage to the gastric mucosa was defined as follows: 0, all gastric mucosal cells intact with normal shape, location, appearance, and density; 1, superficial surface mucous cell damage; 2, extensive surface cell damage ex-
AND
GASTRIC
INJURY
641
tending into the gastric pits; 3, cellular damage extending into the gastric gland area (parietal cell area) but involving less than 50% of the total mucosa; 4, extensive damage of the mucosa extending into the lower third of the mucosa (chief cell area), accompanied by disorganization of the histologic architecture. Each section had the histologic score averaged over the length of the mucosa evaluated. Deep mucosal injury was defined histologically as mucosa with type III or IV injury. The percentage length of mucosa with type III and IV damage was quantified for each group of animals. Mast cell analysis was performed Mast cell analysis. by counting the number of mast cells in the mucosa and submucosa in 40 consecutive oil immersion fields as previously described [ 111. Mast cells were identified by their characteristic metachromatic staining with toluidine blue. Only intact, granulated, and metachromatically staining mast cells were counted in this analysis. Data analysis. Data are expressed as mean values f SEM. Statistical significance of differences was determined using the Kruskall-Wallis analysis of variance (ANOVA) or Student’s t test. RESULTS
1. Gross Injury
Study
The results of the gross injury study are presented graphically in Fig. 1. Absolute ethanol intragastrically produced mucosal lesions involving 25 f 2.0% (mean & SEM) of the gastric corpus in the control animals in all experimental groups. These lesions were characterized by long, dark red bands lying vertically primarily in the proximal corpus mucosa. Prior treatment with isoproterenol resulted in a significant reduction of the gross lesion area in all groups studied (P < 0.01). The isoproterenoltreated groups had gross lesions involving only 4 + 0.6%, 13 -t 2.6%, and 19 ? 3% of the corpus mucosa, respectively. These gross lesion areas were inversely proportional to the dose of isoproterenol given, with larger dosages of isoproterenol resulting in smaller gross lesion areas. When the data were plotted as percentage gastric mucosal protection versus the log dose of isoproterenol, a linear doseresponse function was obtained (Fig. 2). Propranolol pretreatment blocked the protective effect of 50 and 500 pg/kg of isoproterenol (P < 0.05), but did not block the protective effect of the largest dose of isoproterenol (50 mg/kg). In the propranolol-pretreated animals in experiment A (50 mg/kg isoproterenol), intragastric ethanol produced an increase in the gross lesion area when compared to control Krebs-pretreated animals (41 +_ 3.7% vs 25 f 2%, P < 0.05) as shown in Fig. 1, panel A. This increase in lesion area produced by propranolol pretreatment alone was not seen in the other experimental groups. (Fig. 1, panels B and C). In animals pretreated with indomethacin, isoproterenol treatment still showed a significant reduction in the per-
JOURNAL
642
OF SURGICAL
RESEARCH:
m
- KREBS
0
- IS0 50 MG/KG
GROSS 5o INJURY 40 (%I so 20
2. Histologic Study
80
0
- IS0 500 pG/KG
GROSS *’
INJURY j; (%)
1989
area when compared to the control Krebs-pretreated animals (36 + 3% vs 25 +- 2%, P < 0.05) (Fig. 1, panel A).
GROSS INJURY ANALYSIS C
VOL. 46, NO. 6, JUNE
20
&I
- KREBS
0
- IS0 50 /&/KG
INJURY 60 (%) j,o 20 10 0 PREITX
KREBS
PROPRANOLOL
INDOMETHACIN
FIG. 1. The effect of intragastric isoproterenol on the percentage gross lesion formation in the corpus mucosa produced by 100% ethanol. The three different doses of isoproterenol used (50 mg/kg, 500 pg/kg, and 50 fig/kg) are shown in the upper A, middle B, and lower C panels, respectively. There were six rats in each group. Three pretreatments given prior to the administration of isoproterenol are shown on the abscissa and include Krebs, propranolol, or indomethacin. The results are expressed as means + SEM. a = P < 0.01, isoproterenol treatment vs Krebs treatment; b = P < 0.05, isoproterenol treatment vs Krebs treatment; c = P < 0.05, propranolol and indomethacin pretreatment vs Krebs pretreatment. Kruskall-Wallis analysis of variance.
Intragastric isoproterenol protected the rat gastric mucosa from the deep histologic injury in a dose-dependent fashion in both the Krebs and the indomethacin-pretreated groups as shown in Fig. 3. In the propranololpretreated group, isoproterenol’s protective effect was blocked at doses of 50 and 500 pg/kg but not at a dosage of 50 mg/kg. These results confirm the findings in the gross injury study that propranolol blocked the protective effect of isoproterenol at low doses (50 and 500 l.cg/kg) but not high doses (50 mg/kg). When examined histologically, however, propranolol and indomethacin pretreatments in experiment A (Fig. 3 panel A) did not produce an increase in deep histologic injury (type III and IV) when compared to the Krebs pretreated control group (57 + 6% vs 56 f 5% vs 48 + 7%) as was shown in the gross injury study. 3. Mast Cell Counts The results in all groups these counts in the gastric
of mucosal and submucosal mast cell counts studied are presented in Table 1. Overall, are in general agreement with those found mucosa and submucosa in previously pub-
HISTOLOGIC “T
centage mucosal injury at all dosages studied (13 +- 1.6%, 17 + 4%, and 21 f 3%, P < 0.05). The mucosal protection produced by isoproterenol in indomethacin-pretreated animals also showed a dose-dependent effect analogous to that seen in the Krebs-pretreated animals. In the experiment A (isoproterenol50 mg/kg), indomethacin pretreatment alone resulted in an increase in the gross lesion
MUCOSAL
PROTECTION
III&N
5o
INJURY
4o
(N
=O 20
MPE III&N
70 60
INJURY
So
@I
‘Cl 30 10 10
PRETX’ 25 I ,,,,,..:
1o-6
- KREES
0
rlPE
80
117.5 + 10.42x
Il.,...;
ANALYSIS - 150 50 MC/KG
=
- KREES
0
- IS0 500 j&/KC
(a)
““T v=
07
INJURY
h
,,.....;
-...,i
.‘llil
,.?
10-z 1 o-3 1o-4 1o-5 LOG DOSE ISOPROTERENOL
FIG. 2. Relationship of percentage mucosal protection to the log dose of isoproterenol. This relationship is linear and defined by the equation Y = 117.5 + 10.42X. The correlation coefficient was 0.993.
KREBS
=
- KREBS
0
- IS0 50 ,&/KC
PROPRANOLOL INDOMETHACIN
FIG. 3. The effect of intragastric isoproterenol on the percentage length of type III and IV histologic injury in the gastric mucosa after 100% ethanol. The three different doses of isoproterenol used (50 mg/ kg, 500 pg/kg, and 50 rg/kg) are shown in the upper A, middle B, and lower C panels, respectively. There were six rats in each group. The three pretreatments given are shown on the abscissa and include Krebs, propranolol, or indomethacin. The results are expressed as means ? SEM. a = P < 0.01, isoproterenol treatment vs Krebs treatment; b = P < 0.05, isoproterenol treatment YS Krebs treatment.
HOWARD,
PASSARO,
TABLE
AND
GUTH:
ISOPROTERENOL
1
Effect of Ethanol with and without Isoproterenol, Propranolol, the Mast, or Indomethacin on Mast Cell Counts in the Rat Gastric Mucosa and Submucosa Treatment
N
Control VEH/VEH VEH/ISO PRO/VEH PRO/IS0 INDO/VEH INDO/ISO
6 6 6 6 6 6 6
Mucosal mast cells/400if” 79 28 28 29 33 36 34
+ 3.3 f 3.3 + 3.2 +- 3.5 rfr 2.5 + 4.1 + 3.9
P’
10.01
Submucosal mast celIs/400if 40 28 28 29 30 30 28
f + + + k f +
1.9 2.4 1.8 2.3 1.8 1.7 1.4
Pb
<0.05 <0.05 <0.05 <0.05 <0.05 10.05
Note. The results are expressed as mean values + SEM. N = 6 in the control group which did not receive ethanol, and N = 18 in all other groups which did receive ethanol. Absolute ethanol resulted in a significant decrease in both mucosal (P < 0.01) and submucosal (P < 0.05) mast cell counts. There was no effect seen with any of the treatments or pretreatments given. VEH, Krebs solution; PRO, propranolol; ISO, isoproterenol; INDO, indomethacin.
lished reports [ll, 121. Our data show that in all groups given 100% ethanol, the mucosal mast cell counts were decreased by an average of 62% compared to those of control animals (P < 0.01). This decrease in mast cell counts caused by 100% ethanol was unaltered by any of the treatments or pretreatments given. Similarly, the submucosal mast cells were decreased by an average of 28% in all groups given absolute ethanol and again there was no effect noted with any of the treatments or pretreatments given. DISCUSSION
When given intragastrically prior to the application of absolute ethanol, isoproterenol prevented both the gross and the histologic injury of the gastric mucosa in a dosedependent manner. Pretreatment of the animal with propranolol blocked the protective effect of isoproterenol at lower doses (50 and 500 pg/kg) but not at a high dose (50 mg/kg). These observations suggest that topical isoproterenol protects the rat gastric mucosa from ethanol-induced injury, and this protective effect is mediated by a P-adrenergic receptor mechanism. The inability of propranolol to block the protective effect of high dose isoproterenol is probably due to the pharmacologic properties of the propranolol blockade on the @-adrenergic receptor [ 131. Propranolol blocks the receptor by a reversible, competitive antagonist interaction. The pharmacokinetics of this type of drug-receptor interaction is such that the receptor blocking by the antagonist (propranolol) can be overcome by increasing the concentration of the agonist (isoproterenol) until ultimately achieving the same maximal effect [ 141. Intragastric isoproterenol at a dose of 50 mg/kg probably overcomes the competitive antagonism of propranolol on the
AND
GASTRIC
INJURY
643
J3-adrenergic receptor when propranolol is given at a fixed dose of 2 mg/kg subcutaneously. To conclusively establish this relationship in our model, we would need to study the dose-response characteristics of isoproterenol at several different dosages of propranolol [14]. Critical evaluation of our data reveals an apparent discrepancy between the gross and the histologic injury in animals pretreated with propranolol or indomethacin in the experiment using 50 mg/kg of isoproterenol (Figs. 1A and 3A). Both propranolol and indomethacin pretreatment resulted in a significant increase in gross lesion area compared to control animals pretreated with Krebs solution, suggesting that endogenous &adrenergic and prostaglandin mechanisms are exerting a protective effect. In contrast, histologic examination of these same stomachs revealed no difference in the percentage mucosa with deep (type III and IV) mucosal injury between any of the three pretreatment groups. The standardized tissue sampling technique which was used to obtain specimens of the gastric mucosa for histologic analysis may have introduced a sampling error into the data. In this circumstance, a decrease in gross injury high on the posterior wall of the glandular stomach (where the histologic sections were obtained) or gross injury extending further into the distal corpus with propranolol or indomethacin pretreatment would result in an underestimation of the histologic injury compared to the gross injury. This was not the case in the other two groups of experiments (50 and 500 pg/kg) where a good correlation between the gross and the histologic injury was obtained. Furthermore, in studies from both our lab and others, although gross injury analysis underestimates the types I and II histologic injury, it correlates quite well with the deep gastric mucosal injury (types III and IV) [3, lo]. Concerning the possible mechanisms of isoproterenol protection of the gastric mucosa, it has been shown in one study that mast cell degranulation with release of vasoactive mediators correlated strongly with the extent of ethanol-induced gross injury [ 21. In support of the role of mast cells in gastric mucosal injury is the finding that prior gastric mucosal mast cell depletion with dexamethasone or the inhibition of mast cell degranulation with zinc sulfate treatment both diminished the extent of hemorrhagic gastric mucosal lesions produced in a stress model of gastric mucosal injury using vagal nerve stimulation [12, 151. Mast cell degranulation and release of inflammatory mediators may play a significant pathophysiologic role in ethanol-induced injury [ 161. Mast cell degranulation results in the release of histamine, and pretreatment with an Hz histamine receptor antagonist has been shown to decrease ethanol-induced gastric injury [ 171. Isoproterenol is a potent inhibitor of antigen-induced mast cell degranulation [18]. To test the hypothesis that isoproterenol protected the gastric mucosa by stabilizing the mast cell membrane against ethanol-induced mast cell degranulation, we examined the numbers of metachromatically staining mast cells in the mucosa and sub-
644
JOURNAL
OF SURGICAL
RESEARCH:
mucosa of the tissues obtained in this study. Isoproterenol treatment had no effect on the mast cell degranulation caused by 100% ethanol. All groups given absolute ethanol had a significant reduction in mast cells in both the mucosal and the submucosal compartments. In addition, no relationship could be found between the extent of mast cell degranulation and the degree of gross and/or histologic injury produced in the gastric mucosa using this experimental technique. A second potential mechanism by which topical isoproterenol may protect the gastric mucosa is through the generation of endogenous prostaglandins. Prostaglandins are known to be cytoprotective in the gastric mucosa [ 191. The stimulation of endogenous prostaglandins has been proposed as the mechanism by which the topical application of weak irritants to the gastric mucosal surface protects against subsequent damage by more concentrated irritants [20, 211. Theoretically, topical isoproterenol might stimulate the production of endogenous prostaglandins through either a direct (pharmacologic) or an indirect (osmotic) effect on the gastric mucosa. To prevent the generation of endogenous prostaglandins, indomethacin, a cyclooxygenase inhibitor, was given in a dose (5 mg/kg) known to be effective in preventing the generation of endogenous prostaglandins [3,20]. Our data show that prostaglandin production is not the mechanism of action by which isoproterenol prevents gastric mucosal injury as complete blockage of their production does not affect the protective action of isoproterenol. On the basis of the assumption that isoproterenol protects the gastric mucosa through a local P-adrenergic receptor mechanism in the mucosa or submucosa, it is interesting to speculate on the possible methods by which this effect could be produced. Recently it has been shown that the intragastric instillation of 100% ethanol results in a rapid and marked vasoconstriction of the gastric submucosal venules [ 161. This venular constriction may play a central role in the vascular congestion, the microcirculatory stasis, and the tissue injury seen in the ethanolinduced gastric mucosal injury model [ 221. Isoproterenol may act by preventing this venoconstriction through the stimulation of adenyl cyclase, CAMP formation, and subsequent relaxation of the vascular smooth muscle resulting in vasodilatation or prevention of venoconstriction [23]. Alternatively, isoproterenol may be acting by augmenting the microvascular blood flow to the gastric mucosa through the direct dilation of submucosal arterioles and venules. Finally, the potent effects of fi-adrenergic stimulation on the prevention of microvascular permeability produced by inflammatory mediators may also play a role in this gastric mucosal protection [24]. Further investigation will be needed to delineate the precise mechanism involved in this protective efiect. In summary, this study shows that isoproterenol prevents both the gross and the histologic mucosal injury produced by intragastric 100% ethanol. This protection occurs in a dose-devendent manner and is blocked bv nre-
VOL. 46, NO. 6, JUNE
1989
treatment with propranolol but not indomethacin. These data provide evidence that the gastric mucosal protection of intragastric isoproterenol is mediated by a @adrenergic receptor mechanism and does not involve mucosal mast cell stabilization or endogenous prostaglandin production. ACKNOWLEDGMENT This work AM25891.
was supported
by National
Institute
of Health
Grant
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
Allen, A., Flemstrom, G., Garner, A., Silen, W., and Turnberg, L. A. @da). Mechanisms of Mucosal Protection in the Upper Gastrointestinal Tract. New York: Raven Press, 1984. Del, F., Borck, H., and Hosenfeld, S. Effects of somatostatin on ethanol-induced gastric erosions in the rat: Role of mast cells. Agents Actions 18: 273, 1986. Guth, P. H., Paulsen, G., and Nagata, H. Histologic and microcirculatory changes in alcohol-induced gastric lesions in the rat: Effect of prostaglandin cytoprotection. Gastroenterology 87: 1083, 1984. Ohya, Y., and Guth, P. H. Ethanol-induced gastric mucosal blood flow and vascular permeability changes in the rat. Dig. Dis. Sci. 33: 883,1988. Bou-Abboud, C. F., Wayland, H., Paulsen, G., and Guth, P. H. Microcirculatory stasis precedes tissue necrosis in ethanol-induced gastric mucosal injury. Dig. Dis. Sci. 33: 872, 1988. Ito, S., and Lacy, E. R. Morphology of rat gastric mucosal damage, defense, and restitution in the presence of luminal ethanol. Gastroenterobgy 88: 250, 1985. Ritchie, W. P., Jr., and Shearburn, E. W., III. Influence of isoproterenol and cholestyramine on acute gastric mucosa ulcerogenesis. Gastroenterology 73: 62, 1977. Moody, F. G., McGreevy, J., Zalewsky, C., Cheung, L. Y., and Simans, M. The cytoprotective effect of mucosal blood flow in experimental acute gastritis. Acta. Physiol. Stand. (Suppl); 35, 1978. Wallace,J. L., Morris, G. P., Krausse,E. J., and Greaves. S. E. Reduction by cytoprotective agents of ethanol-induced damage to the rat gastric mucosa: a correlated morphological and physiological study. Canad. J. Physiol. Pharmacol. 60: 1686, 1982. Lacy, E. R., and Ito, S. Microscopic analysis of ethanol damage to rat gastric mucosa after treatment with a prostaglandin. Gastroenterology 83: 619, 1982. Guth, P. H., and Hall, P. Microcirculatory and mast cell changes in restraint-induced gastric ulcer. Gastroenterolugy 60: 562, 1966. Cho, C. H., and Ogle, C. W. The effects of zinc sulphate on vagalinduced mast cell changes and ulcers in the rat stomach. Eur. J. Pharmacol. 43: 315, 1977. Weiner, N. Drugs that inhibit adrenergic nerves and block adrenergic receptors. In A. G. Gilman, L. S. Goodman, and A. Gilman (Eds.), The Pharmacolagical Basis of Therapeutics, 6th ed. New York: Macmillan. 1980. Pp. 190-194. Triggle, D. J. Receptor theory. In J. R. Smythies and R. J. Bradley (Eds.), Receptors in Pharmacology. New York: Dekker, 1978. Pp. l-66. Bose, R., Brown, A., and Szabo, S. Histochemical and pharmacological studies related to mast cells with ethanol- or aspirininduced gastric erosions. Fed. Proc. 44: 733, 1985. [Abstract] Oates, I’. J.. and Hakkinen. J. P. Studies on the mechanism of ethanol-induced gastric damage in rats. Castroenterology 04: 10, 1988.
HOWARD, 17.
18.
19.
20.
PASSARO,
AND
GUTH:
ISOPROTERENOL
Szabo, S., Gallagher, G. T., Horner, H., et al. Role of the adrenal cortex in gastric mucosal protection by prostaglandins, sulthydryls, and cimetidine in the rat. Gastroenterology 85: 1384, 1983. Assem, E. S. K., and Schild, H. 0. Inhibition by sympathomimetic amines of histamine release induced by antigen in passively sensitized human lung. Nafure fLoradon) 224: 1028, 1969. Robert, A., Nezamis, J. E., Lancaster, C., and Hanchar, A. J. Cytoprotection by prostaglandins in rats. Prevention of gastric necrosis produced by alcohol, HCL, NaOH, hypertonic NaCl, and thermal injury. Castroenterobgy 77: 433, 1979. Robert, A., Nezamis, J. E., Lancaster, C., Davis, J. P., Field, S. 0.. and Hanchar, A. J. Mild irritants prevent gastric necrosis through “adaptive cytoprotection” mediated by prostaglandins. Amer. J. Physd. 245: G113, 1983.
21.
22.
23.
24.
AND
GASTRIC
INJURY
645
Knapp, H. R., Oelz, O., Sweetman, B. J., and Oates, J. A. Synthesis and metabolism of prostaglandins E2, F2,, and D2 by the rat gastrointestinal tract. Stimulation by a hypertonic environment in vitro. Prostaglmdins 15: 751. 1978. Yonei, Y., Wayland, H., and Guth, P. H. Role of arachidonic acid metabolites in ethanol vasoaction in rat gastric submucosa. Amer. J. Physiol. 266: G731, 1988. Madhu, B., Srivastava, A., and Franks, D. J. Stimulation of adenylate cyclase by adenosine and other agonist in mesenteric artery smooth muscle cells in culture. Life Sci. 37: 357, 1985. Svensjo, E., Adamske, S. W., Su, K., and Grega, G. J. Quantitative physiological and morphological aspects of microvascular permeability changes induced by histamine and inhibited by terbutaline. Acta Physiol. Scand. 116: 265, 1982.
OF StlRGICAL
RESEARCH
46, 640-64.5 (1989)
Topical lsoproterenol Protects the Rat Gastric Mucosa from Ethanol-Induced Injury THOMAS ,J. HOWARD, M.D., F:I)WAHI) PASSAHO, tJ~., M.D., ANL) PAIJI. H. GI!TH, M.D. Research
and Surgical Seruices, West Los Angeles (‘enter for Ulcer Research and Eunfuntion
Presented at the Annual Meeting of the Association
Press.
Administrafion Los Angeles,
Medical California
Center, 9007,‘l
and the
for Academic Surgery, Salt lake City, LJtah, November
This study evaluated the effects of topical isoproterenol, a &adrenergic agonist, on the morphologic damage produced in the gastric mucosa by ethanol. The orogastric instillation of 100% ethanol in rats resulted in gross lesion formation and deep histologic injury in the gastric mucosa. Animals pretreated with oral isoproterenol (50 pg/kg, 500 bg/kg, 50 mg/kg) showed dose-dependent protection from both the gross and the histologic mucosal injury (P < 0.01, ANOVA). Pretreatment with propran0101 (2 mg/kg/sec) but not indomethacin (5 mg/kg/sec) blocked this protective effect. Isoproterenol had no effect on ethanol-induced mast cell degranulation as both mucosal and submucosal mast cell counts were significantly and equally decreased in all groups treated with 100% ethanol (P < 0.05). These findings show that topical isoproterenol protects the rat gastric mucosa from both the gross and the histologic injury caused by 100% ethanol. This protection is mediated by a &adrenergic receptor mechanism as it can be blocked by prior treatment with propranolol. but does not involve stabilization of mucosal or submucosal mast cell membranes. c 1RR9Arudemic
Vrterans (CURE),
16-19, 1988
a number of chemical ulcerogens [7, 81. Unfortunately, the potent cardiovascular effects of intravenous isoproterenol limit its clinical usefulness in preventing gastric mucosal injury. The topical application of isoproterenol, however, could potentially avoid the cardiovascular effects of intravenous administration while still providing the beneficial drug effect locally. Topical application of isoproterenol in the rat gastric mucosa has been shown previously to diminish the histologic injury produced by 40% ethanol [9]. The purpose of this study was to confirm the protective effect of topical isoproterenol using an analysis of both the gross and the histologic mucosal injury in the absolute ethanol model of gastric mucosal injury. We were also interested in determining whether the protective effect of isoproterenol involved fl-adrenergic receptors (i.e., would be blocked by fl-adrenergic receptor antagonists), stimulation of endogenous prostaglandin synthesis (i.e., would be blocked by a cyclooxygenase inhibitor), or stabilization of mucosal mast cell membranes (i.e., prevention of mast cell degranulation). METHODS
Inc.
INTRODUCTION Absolute ethanol is a noxious chemical agent widely used in the study of gastric mucosal injury [ 1J. Knowledge of the mechanisms involved in the ethanol-induced injury model are of importance in understanding the response of the gastric mucosa to injury and in developing treatment strategies designed to protect against acute gastric mucosal damage. The ethanol-induced injury produces a number of alterations in the corpus mucosa including: mast cell degranulation [2], vascular engorgement [3], increased permeability [4], and ultimately complete microvascular stasis [5]. These physiologic alterations are accompanied by the morphologic finding of vertical, dark red areas of gross mucosal injury, necrosis, edema, vascular congestion, and mucosal hemorrhage [6]. Isoproterenol, when given systemically, has been shown to prevent the gross gastric mucosal injury produced by
Kxperimental design. Thirty-six Sprague-Dawley rats (180-250 g) were fasted overnight but allowed free access to water up until 2 hr prior to the experiment. Groups of six rats each were randomly assigned to one of six groups, weighed, and appropriately marked for later identification. Groups 1 and 2 underwent pretreatment with 1 cc Krebs Solution subcutaneously (SC), groups 3 and 4 received 2 mg/kg propranolol SC,and groups 5 and 6 received 5 mg/ kg indomethacin (5 mg/kg) SC.After 30 min, group 2, 4, and 6 rats were given 50 pg/kg isoproterenol intragastrically by orogastric gavage. Group 1, 3, and 5 rats were given 1 cc Krebs solution intragastrically. Fifteen minutes later, all rats were given 100% ethanol (1 cc/200 g) intragastrically by orogastric gavage. Thirty minutes after ethanol instillation each rat was killed and the stomach removed and examined for gross mucosal lesions as described below. The dose of 50 pg/kg was chosen as this was the dose previously shown to prevent the histologic damage produced by 40% ethanol [9]. Doses of lo- and
HOWARD,
PASSARO,
AND
GUTH:
ISOPROTERENOI,
IOOO-fold increases were used to evaluate the dose-response relationship of this drug. The experiment was therefore repeated two more times using differing doses of isoproterenol (500 pg/kg and 50 mg/kg). Materials. Absolute ethanol was purchased from Midwest Grain Products of Illinois (Pekin, IL). Isoproterenol (1[3’,4’-dihydroxyphenyll-2-isopropylaminoethanol), a potent nonspecific p-adrenergic agonist, was dissolved in Krebs solution (50 mg/5 ml, 500 pg/5 ml, or 50 pg/5 ml) for intragastric application. Propranolol hydrochloride (1-(isopropylamino]-3-[1-naphthyloxyl-2-propranolol HCl), a nonselective &adrenergic antagonist, was prepared by dissolving 40 mg in 100 ml Krebs solution. Indomethacin (1-[p-chlorobenzoyll-5-methoxy-2-methyl indole-3-acetic acid) a prostaglandin synthesis inhibitor, was prepared by completely dissolving 100 mg in 5% sodium bicarbonate by heating and stirring for 5 hr at 40°C. This was then diluted with distilled water so the final solution contained 1 mg of indomethacin in 1 ml of 1% NaHC09. Each stomach was opened along Gross lesion analysis. the greater curvature, gently rinsed in saline, pinned open to expose the mucosa, and photographed using a Polaroid pack film camera and Polaroid Type 108 film (Polaroid Tech. Photo, Cambridge, MA) at 2:l magnification. The photographs were coded for blinded lesion tracing and subsequent analysis. The outline of the gastric corpus and the lesions present (characterized by dark red bands) were traced onto transparent film using a felt tip marker. The lesion area (cm*), total area of the corpus (cm*), and percentage corpus mucosa containing lesions were determined by a computerized image analyzer device (MICRO/ PDP-11, Digital Equipment Corp., Maynard, MA, equipped with imaging boards, Imaging Technology Inc., Woburn, MA). The image analyzer program for the measurement of total area, gross lesion area, and the percentage gross lesion was written by Dr. Edward Livingston. Immediately after photography of Histologic analysis. the stomach, the tissue was placed in a basic lead-acetate fixative for 24 hr and then stored in 70% ethanol. Later, a standardized strip of tissue across the entire posterior wall of the glandular corpus just distal to the limiting ridge was removed, embedded in paraffin, sectioned (4 pm thick), and stained with either hematoxylin and eosin (histologic injury analysis) or toluidine blue (mast cell analysis). Slides were coded prior to microscopic evaluation, so the examiner (T.J.H.) was unaware of the treatment group the slide came from. Histologic injury was scored using an American Optical microscope with a 45X objective and a final magnification of 450X. The mucosal injury score was determined using modified criteria based on those previously established by Lacy and Ito [lo]. Damage to the gastric mucosa was defined as follows: 0, all gastric mucosal cells intact with normal shape, location, appearance, and density; 1, superficial surface mucous cell damage; 2, extensive surface cell damage ex-
AND
GASTRIC
INJURY
641
tending into the gastric pits; 3, cellular damage extending into the gastric gland area (parietal cell area) but involving less than 50% of the total mucosa; 4, extensive damage of the mucosa extending into the lower third of the mucosa (chief cell area), accompanied by disorganization of the histologic architecture. Each section had the histologic score averaged over the length of the mucosa evaluated. Deep mucosal injury was defined histologically as mucosa with type III or IV injury. The percentage length of mucosa with type III and IV damage was quantified for each group of animals. Mast cell analysis was performed Mast cell analysis. by counting the number of mast cells in the mucosa and submucosa in 40 consecutive oil immersion fields as previously described [ 111. Mast cells were identified by their characteristic metachromatic staining with toluidine blue. Only intact, granulated, and metachromatically staining mast cells were counted in this analysis. Data analysis. Data are expressed as mean values f SEM. Statistical significance of differences was determined using the Kruskall-Wallis analysis of variance (ANOVA) or Student’s t test. RESULTS
1. Gross Injury
Study
The results of the gross injury study are presented graphically in Fig. 1. Absolute ethanol intragastrically produced mucosal lesions involving 25 f 2.0% (mean & SEM) of the gastric corpus in the control animals in all experimental groups. These lesions were characterized by long, dark red bands lying vertically primarily in the proximal corpus mucosa. Prior treatment with isoproterenol resulted in a significant reduction of the gross lesion area in all groups studied (P < 0.01). The isoproterenoltreated groups had gross lesions involving only 4 + 0.6%, 13 -t 2.6%, and 19 ? 3% of the corpus mucosa, respectively. These gross lesion areas were inversely proportional to the dose of isoproterenol given, with larger dosages of isoproterenol resulting in smaller gross lesion areas. When the data were plotted as percentage gastric mucosal protection versus the log dose of isoproterenol, a linear doseresponse function was obtained (Fig. 2). Propranolol pretreatment blocked the protective effect of 50 and 500 pg/kg of isoproterenol (P < 0.05), but did not block the protective effect of the largest dose of isoproterenol (50 mg/kg). In the propranolol-pretreated animals in experiment A (50 mg/kg isoproterenol), intragastric ethanol produced an increase in the gross lesion area when compared to control Krebs-pretreated animals (41 +_ 3.7% vs 25 f 2%, P < 0.05) as shown in Fig. 1, panel A. This increase in lesion area produced by propranolol pretreatment alone was not seen in the other experimental groups. (Fig. 1, panels B and C). In animals pretreated with indomethacin, isoproterenol treatment still showed a significant reduction in the per-
JOURNAL
642
OF SURGICAL
RESEARCH:
m
- KREBS
0
- IS0 50 MG/KG
GROSS 5o INJURY 40 (%I so 20
2. Histologic Study
80
0
- IS0 500 pG/KG
GROSS *’
INJURY j; (%)
1989
area when compared to the control Krebs-pretreated animals (36 + 3% vs 25 +- 2%, P < 0.05) (Fig. 1, panel A).
GROSS INJURY ANALYSIS C
VOL. 46, NO. 6, JUNE
20
&I
- KREBS
0
- IS0 50 /&/KG
INJURY 60 (%) j,o 20 10 0 PREITX
KREBS
PROPRANOLOL
INDOMETHACIN
FIG. 1. The effect of intragastric isoproterenol on the percentage gross lesion formation in the corpus mucosa produced by 100% ethanol. The three different doses of isoproterenol used (50 mg/kg, 500 pg/kg, and 50 fig/kg) are shown in the upper A, middle B, and lower C panels, respectively. There were six rats in each group. Three pretreatments given prior to the administration of isoproterenol are shown on the abscissa and include Krebs, propranolol, or indomethacin. The results are expressed as means + SEM. a = P < 0.01, isoproterenol treatment vs Krebs treatment; b = P < 0.05, isoproterenol treatment vs Krebs treatment; c = P < 0.05, propranolol and indomethacin pretreatment vs Krebs pretreatment. Kruskall-Wallis analysis of variance.
Intragastric isoproterenol protected the rat gastric mucosa from the deep histologic injury in a dose-dependent fashion in both the Krebs and the indomethacin-pretreated groups as shown in Fig. 3. In the propranololpretreated group, isoproterenol’s protective effect was blocked at doses of 50 and 500 pg/kg but not at a dosage of 50 mg/kg. These results confirm the findings in the gross injury study that propranolol blocked the protective effect of isoproterenol at low doses (50 and 500 l.cg/kg) but not high doses (50 mg/kg). When examined histologically, however, propranolol and indomethacin pretreatments in experiment A (Fig. 3 panel A) did not produce an increase in deep histologic injury (type III and IV) when compared to the Krebs pretreated control group (57 + 6% vs 56 f 5% vs 48 + 7%) as was shown in the gross injury study. 3. Mast Cell Counts The results in all groups these counts in the gastric
of mucosal and submucosal mast cell counts studied are presented in Table 1. Overall, are in general agreement with those found mucosa and submucosa in previously pub-
HISTOLOGIC “T
centage mucosal injury at all dosages studied (13 +- 1.6%, 17 + 4%, and 21 f 3%, P < 0.05). The mucosal protection produced by isoproterenol in indomethacin-pretreated animals also showed a dose-dependent effect analogous to that seen in the Krebs-pretreated animals. In the experiment A (isoproterenol50 mg/kg), indomethacin pretreatment alone resulted in an increase in the gross lesion
MUCOSAL
PROTECTION
III&N
5o
INJURY
4o
(N
=O 20
MPE III&N
70 60
INJURY
So
@I
‘Cl 30 10 10
PRETX’ 25 I ,,,,,..:
1o-6
- KREES
0
rlPE
80
117.5 + 10.42x
Il.,...;
ANALYSIS - 150 50 MC/KG
=
- KREES
0
- IS0 500 j&/KC
(a)
““T v=
07
INJURY
h
,,.....;
-...,i
.‘llil
,.?
10-z 1 o-3 1o-4 1o-5 LOG DOSE ISOPROTERENOL
FIG. 2. Relationship of percentage mucosal protection to the log dose of isoproterenol. This relationship is linear and defined by the equation Y = 117.5 + 10.42X. The correlation coefficient was 0.993.
KREBS
=
- KREBS
0
- IS0 50 ,&/KC
PROPRANOLOL INDOMETHACIN
FIG. 3. The effect of intragastric isoproterenol on the percentage length of type III and IV histologic injury in the gastric mucosa after 100% ethanol. The three different doses of isoproterenol used (50 mg/ kg, 500 pg/kg, and 50 rg/kg) are shown in the upper A, middle B, and lower C panels, respectively. There were six rats in each group. The three pretreatments given are shown on the abscissa and include Krebs, propranolol, or indomethacin. The results are expressed as means ? SEM. a = P < 0.01, isoproterenol treatment vs Krebs treatment; b = P < 0.05, isoproterenol treatment YS Krebs treatment.
HOWARD,
PASSARO,
TABLE
AND
GUTH:
ISOPROTERENOL
1
Effect of Ethanol with and without Isoproterenol, Propranolol, the Mast, or Indomethacin on Mast Cell Counts in the Rat Gastric Mucosa and Submucosa Treatment
N
Control VEH/VEH VEH/ISO PRO/VEH PRO/IS0 INDO/VEH INDO/ISO
6 6 6 6 6 6 6
Mucosal mast cells/400if” 79 28 28 29 33 36 34
+ 3.3 f 3.3 + 3.2 +- 3.5 rfr 2.5 + 4.1 + 3.9
P’
10.01
Submucosal mast celIs/400if 40 28 28 29 30 30 28
f + + + k f +
1.9 2.4 1.8 2.3 1.8 1.7 1.4
Pb
<0.05 <0.05 <0.05 <0.05 <0.05 10.05
Note. The results are expressed as mean values + SEM. N = 6 in the control group which did not receive ethanol, and N = 18 in all other groups which did receive ethanol. Absolute ethanol resulted in a significant decrease in both mucosal (P < 0.01) and submucosal (P < 0.05) mast cell counts. There was no effect seen with any of the treatments or pretreatments given. VEH, Krebs solution; PRO, propranolol; ISO, isoproterenol; INDO, indomethacin.
lished reports [ll, 121. Our data show that in all groups given 100% ethanol, the mucosal mast cell counts were decreased by an average of 62% compared to those of control animals (P < 0.01). This decrease in mast cell counts caused by 100% ethanol was unaltered by any of the treatments or pretreatments given. Similarly, the submucosal mast cells were decreased by an average of 28% in all groups given absolute ethanol and again there was no effect noted with any of the treatments or pretreatments given. DISCUSSION
When given intragastrically prior to the application of absolute ethanol, isoproterenol prevented both the gross and the histologic injury of the gastric mucosa in a dosedependent manner. Pretreatment of the animal with propranolol blocked the protective effect of isoproterenol at lower doses (50 and 500 pg/kg) but not at a high dose (50 mg/kg). These observations suggest that topical isoproterenol protects the rat gastric mucosa from ethanol-induced injury, and this protective effect is mediated by a P-adrenergic receptor mechanism. The inability of propranolol to block the protective effect of high dose isoproterenol is probably due to the pharmacologic properties of the propranolol blockade on the @-adrenergic receptor [ 131. Propranolol blocks the receptor by a reversible, competitive antagonist interaction. The pharmacokinetics of this type of drug-receptor interaction is such that the receptor blocking by the antagonist (propranolol) can be overcome by increasing the concentration of the agonist (isoproterenol) until ultimately achieving the same maximal effect [ 141. Intragastric isoproterenol at a dose of 50 mg/kg probably overcomes the competitive antagonism of propranolol on the
AND
GASTRIC
INJURY
643
J3-adrenergic receptor when propranolol is given at a fixed dose of 2 mg/kg subcutaneously. To conclusively establish this relationship in our model, we would need to study the dose-response characteristics of isoproterenol at several different dosages of propranolol [14]. Critical evaluation of our data reveals an apparent discrepancy between the gross and the histologic injury in animals pretreated with propranolol or indomethacin in the experiment using 50 mg/kg of isoproterenol (Figs. 1A and 3A). Both propranolol and indomethacin pretreatment resulted in a significant increase in gross lesion area compared to control animals pretreated with Krebs solution, suggesting that endogenous &adrenergic and prostaglandin mechanisms are exerting a protective effect. In contrast, histologic examination of these same stomachs revealed no difference in the percentage mucosa with deep (type III and IV) mucosal injury between any of the three pretreatment groups. The standardized tissue sampling technique which was used to obtain specimens of the gastric mucosa for histologic analysis may have introduced a sampling error into the data. In this circumstance, a decrease in gross injury high on the posterior wall of the glandular stomach (where the histologic sections were obtained) or gross injury extending further into the distal corpus with propranolol or indomethacin pretreatment would result in an underestimation of the histologic injury compared to the gross injury. This was not the case in the other two groups of experiments (50 and 500 pg/kg) where a good correlation between the gross and the histologic injury was obtained. Furthermore, in studies from both our lab and others, although gross injury analysis underestimates the types I and II histologic injury, it correlates quite well with the deep gastric mucosal injury (types III and IV) [3, lo]. Concerning the possible mechanisms of isoproterenol protection of the gastric mucosa, it has been shown in one study that mast cell degranulation with release of vasoactive mediators correlated strongly with the extent of ethanol-induced gross injury [ 21. In support of the role of mast cells in gastric mucosal injury is the finding that prior gastric mucosal mast cell depletion with dexamethasone or the inhibition of mast cell degranulation with zinc sulfate treatment both diminished the extent of hemorrhagic gastric mucosal lesions produced in a stress model of gastric mucosal injury using vagal nerve stimulation [12, 151. Mast cell degranulation and release of inflammatory mediators may play a significant pathophysiologic role in ethanol-induced injury [ 161. Mast cell degranulation results in the release of histamine, and pretreatment with an Hz histamine receptor antagonist has been shown to decrease ethanol-induced gastric injury [ 171. Isoproterenol is a potent inhibitor of antigen-induced mast cell degranulation [18]. To test the hypothesis that isoproterenol protected the gastric mucosa by stabilizing the mast cell membrane against ethanol-induced mast cell degranulation, we examined the numbers of metachromatically staining mast cells in the mucosa and sub-
644
JOURNAL
OF SURGICAL
RESEARCH:
mucosa of the tissues obtained in this study. Isoproterenol treatment had no effect on the mast cell degranulation caused by 100% ethanol. All groups given absolute ethanol had a significant reduction in mast cells in both the mucosal and the submucosal compartments. In addition, no relationship could be found between the extent of mast cell degranulation and the degree of gross and/or histologic injury produced in the gastric mucosa using this experimental technique. A second potential mechanism by which topical isoproterenol may protect the gastric mucosa is through the generation of endogenous prostaglandins. Prostaglandins are known to be cytoprotective in the gastric mucosa [ 191. The stimulation of endogenous prostaglandins has been proposed as the mechanism by which the topical application of weak irritants to the gastric mucosal surface protects against subsequent damage by more concentrated irritants [20, 211. Theoretically, topical isoproterenol might stimulate the production of endogenous prostaglandins through either a direct (pharmacologic) or an indirect (osmotic) effect on the gastric mucosa. To prevent the generation of endogenous prostaglandins, indomethacin, a cyclooxygenase inhibitor, was given in a dose (5 mg/kg) known to be effective in preventing the generation of endogenous prostaglandins [3,20]. Our data show that prostaglandin production is not the mechanism of action by which isoproterenol prevents gastric mucosal injury as complete blockage of their production does not affect the protective action of isoproterenol. On the basis of the assumption that isoproterenol protects the gastric mucosa through a local P-adrenergic receptor mechanism in the mucosa or submucosa, it is interesting to speculate on the possible methods by which this effect could be produced. Recently it has been shown that the intragastric instillation of 100% ethanol results in a rapid and marked vasoconstriction of the gastric submucosal venules [ 161. This venular constriction may play a central role in the vascular congestion, the microcirculatory stasis, and the tissue injury seen in the ethanolinduced gastric mucosal injury model [ 221. Isoproterenol may act by preventing this venoconstriction through the stimulation of adenyl cyclase, CAMP formation, and subsequent relaxation of the vascular smooth muscle resulting in vasodilatation or prevention of venoconstriction [23]. Alternatively, isoproterenol may be acting by augmenting the microvascular blood flow to the gastric mucosa through the direct dilation of submucosal arterioles and venules. Finally, the potent effects of fi-adrenergic stimulation on the prevention of microvascular permeability produced by inflammatory mediators may also play a role in this gastric mucosal protection [24]. Further investigation will be needed to delineate the precise mechanism involved in this protective efiect. In summary, this study shows that isoproterenol prevents both the gross and the histologic mucosal injury produced by intragastric 100% ethanol. This protection occurs in a dose-devendent manner and is blocked bv nre-
VOL. 46, NO. 6, JUNE
1989
treatment with propranolol but not indomethacin. These data provide evidence that the gastric mucosal protection of intragastric isoproterenol is mediated by a @adrenergic receptor mechanism and does not involve mucosal mast cell stabilization or endogenous prostaglandin production. ACKNOWLEDGMENT This work AM25891.
was supported
by National
Institute
of Health
Grant
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
Allen, A., Flemstrom, G., Garner, A., Silen, W., and Turnberg, L. A. @da). Mechanisms of Mucosal Protection in the Upper Gastrointestinal Tract. New York: Raven Press, 1984. Del, F., Borck, H., and Hosenfeld, S. Effects of somatostatin on ethanol-induced gastric erosions in the rat: Role of mast cells. Agents Actions 18: 273, 1986. Guth, P. H., Paulsen, G., and Nagata, H. Histologic and microcirculatory changes in alcohol-induced gastric lesions in the rat: Effect of prostaglandin cytoprotection. Gastroenterology 87: 1083, 1984. Ohya, Y., and Guth, P. H. Ethanol-induced gastric mucosal blood flow and vascular permeability changes in the rat. Dig. Dis. Sci. 33: 883,1988. Bou-Abboud, C. F., Wayland, H., Paulsen, G., and Guth, P. H. Microcirculatory stasis precedes tissue necrosis in ethanol-induced gastric mucosal injury. Dig. Dis. Sci. 33: 872, 1988. Ito, S., and Lacy, E. R. Morphology of rat gastric mucosal damage, defense, and restitution in the presence of luminal ethanol. Gastroenterobgy 88: 250, 1985. Ritchie, W. P., Jr., and Shearburn, E. W., III. Influence of isoproterenol and cholestyramine on acute gastric mucosa ulcerogenesis. Gastroenterology 73: 62, 1977. Moody, F. G., McGreevy, J., Zalewsky, C., Cheung, L. Y., and Simans, M. The cytoprotective effect of mucosal blood flow in experimental acute gastritis. Acta. Physiol. Stand. (Suppl); 35, 1978. Wallace,J. L., Morris, G. P., Krausse,E. J., and Greaves. S. E. Reduction by cytoprotective agents of ethanol-induced damage to the rat gastric mucosa: a correlated morphological and physiological study. Canad. J. Physiol. Pharmacol. 60: 1686, 1982. Lacy, E. R., and Ito, S. Microscopic analysis of ethanol damage to rat gastric mucosa after treatment with a prostaglandin. Gastroenterology 83: 619, 1982. Guth, P. H., and Hall, P. Microcirculatory and mast cell changes in restraint-induced gastric ulcer. Gastroenterolugy 60: 562, 1966. Cho, C. H., and Ogle, C. W. The effects of zinc sulphate on vagalinduced mast cell changes and ulcers in the rat stomach. Eur. J. Pharmacol. 43: 315, 1977. Weiner, N. Drugs that inhibit adrenergic nerves and block adrenergic receptors. In A. G. Gilman, L. S. Goodman, and A. Gilman (Eds.), The Pharmacolagical Basis of Therapeutics, 6th ed. New York: Macmillan. 1980. Pp. 190-194. Triggle, D. J. Receptor theory. In J. R. Smythies and R. J. Bradley (Eds.), Receptors in Pharmacology. New York: Dekker, 1978. Pp. l-66. Bose, R., Brown, A., and Szabo, S. Histochemical and pharmacological studies related to mast cells with ethanol- or aspirininduced gastric erosions. Fed. Proc. 44: 733, 1985. [Abstract] Oates, I’. J.. and Hakkinen. J. P. Studies on the mechanism of ethanol-induced gastric damage in rats. Castroenterology 04: 10, 1988.
HOWARD, 17.
18.
19.
20.
PASSARO,
AND
GUTH:
ISOPROTERENOL
Szabo, S., Gallagher, G. T., Horner, H., et al. Role of the adrenal cortex in gastric mucosal protection by prostaglandins, sulthydryls, and cimetidine in the rat. Gastroenterology 85: 1384, 1983. Assem, E. S. K., and Schild, H. 0. Inhibition by sympathomimetic amines of histamine release induced by antigen in passively sensitized human lung. Nafure fLoradon) 224: 1028, 1969. Robert, A., Nezamis, J. E., Lancaster, C., and Hanchar, A. J. Cytoprotection by prostaglandins in rats. Prevention of gastric necrosis produced by alcohol, HCL, NaOH, hypertonic NaCl, and thermal injury. Castroenterobgy 77: 433, 1979. Robert, A., Nezamis, J. E., Lancaster, C., Davis, J. P., Field, S. 0.. and Hanchar, A. J. Mild irritants prevent gastric necrosis through “adaptive cytoprotection” mediated by prostaglandins. Amer. J. Physd. 245: G113, 1983.
21.
22.
23.
24.
AND
GASTRIC
INJURY
645
Knapp, H. R., Oelz, O., Sweetman, B. J., and Oates, J. A. Synthesis and metabolism of prostaglandins E2, F2,, and D2 by the rat gastrointestinal tract. Stimulation by a hypertonic environment in vitro. Prostaglmdins 15: 751. 1978. Yonei, Y., Wayland, H., and Guth, P. H. Role of arachidonic acid metabolites in ethanol vasoaction in rat gastric submucosa. Amer. J. Physiol. 266: G731, 1988. Madhu, B., Srivastava, A., and Franks, D. J. Stimulation of adenylate cyclase by adenosine and other agonist in mesenteric artery smooth muscle cells in culture. Life Sci. 37: 357, 1985. Svensjo, E., Adamske, S. W., Su, K., and Grega, G. J. Quantitative physiological and morphological aspects of microvascular permeability changes induced by histamine and inhibited by terbutaline. Acta Physiol. Scand. 116: 265, 1982.