The Cytoprotective effect of prostaglandin E2 on Tauro-cholate-induced erosions in gastric mucosa of the rat

Path. Res. Pract. 179, 478-486 (1985)

The Cytoprotective Effect of Prostaglandin E2 on T auroCholate-induced Erosions in Gastric Mucosa of the Rat Light Microscopic, lv1orphometric and Scanning Electron Microscopic Investigations B. Fringes, D. Lorenz, W. Oehlert* Institute of Pathology - Department for Experimental Pathology and HepatoGastroenterology - University of Freiburg

SUMMARY

The morphological changes associated with the cytoprotective effect of prostaglandin E2 (PGE 2) following sodium-taurocholate (NaTC) erosive injury in the gastric mucosa were investigated in male rats. A single topic application of NaTC (80 mMol) induced multiple gastric erosions in all animals. Application of 200 [,tg PG E2/ kg body weight prior to NaTC treatment led to a significant 90% decrease in the lesion-score in PGErprotected animals. Light microscopic morphometric studies of the mucus-producing cells in the fundus mucosa were carried out. Within the PGErprotected animals a significant increase was observed in the length of zones of the mucus-producing cells at the surface and in the foveolae (both PAS-positive and alcian-blue-positive cells). Compared to the NaTCinjured animals, this increase amounted to 8.1 % for the PAS-positive, and 6.1 % for the alcian-blue-positive zone. Compared to the untreated controls, these values were 4.7% and 3.2% respectively. In the scanning electron microscope we observed a characteristic explosive release of mucus and damage of the eel/'s surface membranes in the NaTCtreated animals. The PGErprotected rats showed a predominance of exocytosis of mucus vesicles which formed a characteristic mucus network at the surface membrane. Our investigations suggest that the cytoprotective effect of PGE 2 may, in part, be due to an increase in mucus production and to a modification of mucus adherence at the cell membrane.

Introduction The cytoprotective effect of prostaglandins on the gastrointestinal mucosa was first described by Andre Robert and his coworkers23. They found that gastric mucosal lesions induced by oral administration of a variety of different necrotizing agents, like for example absolute * Dedicated to Prof. Dr. Dr. h.c. Franz Buchner in honor of his 90th birthday, Jan. 20, 1985 0344-0338/84/0179-0478$3.50/0

ethanol, aspirin, strong acid or base, and boiling water, could be prevented by pretreatment with prostaglandins22, 23, 26. Although prostaglandins are able to inhibit gastric acid secretion in animals and humans 21 ,22, the cytoprotective effect is unrelated to the inhibition of gastric acid secretion23 . The mechanism by which prostaglandins exert their cytoprotective effect is still a matter of debate. In further investigations, a stimulation of mucus secretion was influenced by prostaglandins4,22. Prostaglandins © 1985 by Gustav Fischer Verlag, Stuttgart

Gastric Cytoprotection of Prostaglandin Ez . 479

are endogenously synthesized within the gastric and intestinal mucosa where they are presumed to maintain the cellular integrity 15. Vane36 observed that aspirin and indomethacin block the activity of the prostaglandin synthetase system, both in vitro and in vivo. It was suggested that the gastrointestinal intolerance occasionally encountered in humans treated with these types of nonsteroidal anti-inflammatory agents may be due to a deficiency of endogenous prostaglandins 34, 36. Since the cytoprotective effect of prostaglandins is mainly observed in mucus-producing and excreting epithelial cells, the influence of these substances on quantity and quality of the mucus formation possibly underlies cytoprotection4,26,37. In order to investigate a cytoprotective effect upon rat gastric mucosa, we used prostaglandin E2 (PGE2 ) as the protecting, and the conjugated bile acid taurocholate as the necrotizing agent. The efficiency of the cytoprotective effect of prostaglandins upon taurocholate-induced gastric lesions has been shown by Mann 14 • In our study, special interest was directed to the mucus-producing cells in the rat gastric wall, using light microscopy, linear morphometry and scanning electron microscopy.

EXPERIMENT I

,

NaTe

animal

group

II III IV V

,

20 mM

, ,

40 mM

!

0.2 N Hel

80 mM 2X80 mM

time ( min.)

120

60

0

Fig. 1 a. Experiment I: Time-dose-scheme for determination of the effective "erosive" NaTC concentration.

EXPERIMENT II

ani rna 1

group

Material and Methods Animals: Male Wistar rats (Ivanovas, Kisslegg) with an average body weight of 210 g (± 15 g) were first fed on standard diet. The animals were fasted for 24 h, but were allowed water ad libitum, and were placed in special stainless steel cages that were constructed so as to prevent coprophagy. This procedure did not appear to be stressful. The animals were grouped according to random sampling. The substances applied were administered orally by esophagus intubation in order to avoid the known systemic side effects of prostaglandins. The intubation was tolerated by the animals and was carried out without anesthesia. Substances: 1. Prostaglandin E2 (PGE2) (Sigma Chemicals Co.), dissolved in 70% ethanol and phosphate buffer (pH 7.4). 2. Sodium-taurocholate (NaTC) , pKa 1.8 (Fluca, Buchs), solved in 0.2 N hydrochloric acid (pH 1.0).

Experimental Set-up Experiment I: In order to determine the effective ("erosive") sodium-taurocholate concentration (NaTC) , five groups were formed, each group consisting of five animals. The time-dose records (Fig. 1 a) show the different, increasing concentrations of NaTC (20mM to 80 mM NaTC). Each test concentration, diluted in 1 ml of 0.2 N HCI (pH 1), was applied topically to the animals: group I received 20 mM, group II 40 mM, group III 80 mM and group IV two successive 80 mM doses of the acidified NaTe. The controls received 1 ml 0.2 N HCI. Two hours later, the animals were killed in ether anesthesia by exsanguination and the stomachs were removed. Experiment II: Group 1: Prostaglandin Ez (200 !lg/kg body weight, solvent as above) was given orally to 20 male rats, 30 min prior to the oral administration of 80 mM acidified NaTC by two subsequent intubations.

PGE 1

2('10 fig/kg PGE 2 80 mM NaTe

NaTe

, , 2

,

2

- 30

0

60

120

time ( min.)

Fig. 1 b. Experiment II: time-dose-scheme of the experimental set up to demonstrate the cytoprotective effect of PGE z. Group 2: A single dose of 80 mM acidified NaTC (1 ml) was applied to each of the 20 animals. Animals of both groups were killed two hours after the last oesophagus intubation (see time-scheme 1 b).

Light Microscopy The stomachs were opened along the greater curvature, fixed on cork plates and immersed in buffered formalin. Following fixation during 24 h, several tissue slices taken from the fundus and antral wall were embedded in paraffin. Tissue sections were stained with hematoxilin-eosin, PAS and alcian-blue.

Quantitative Assessment of Lesions The gastric mucosa was examined with a binocular magnifier (photomacroscope 400 M, Wild) at a magnification of 60 x. The lesions were counted and their severity (number and size of the erosions) was transformed into a point scale and expressed as a single parameter, the lesion score (modified according to Carmichael et aU (see Table 1).

480 . B. Fringes, D. Lorenz and W. Oehlert Table 1. Lesion score (modified according to Carmichael et ai., 1978)

Size (mm)

Points

:s:; :s:; :s:; :s:; :s:;

1 2 3 4 5 6

>

0.5 1 2 3 4 4

Linear Morphometry Linear morphometry was carried out on light microscopic sections of the fundus mucosa, stained with PAS and alcian-blue using a digitizer (MOP Digiplan, Kontron, Munich). The zone lengths of mucus-producing cells (PAS and alcian-blue) were determined with an opto-pencil (Zeiss light microscope with a projector arm). In order to obtain reproducible results the following conditions were met: 1. tissue sclices of the gastric wall were cut perpendicular to the surface; 2. zone lengths of mucus-producing cells were only measured on gastric glands entirely visible at their long axes (i. e. cut orthograde); 3. mucosal erosions and peri-erosive areas were excluded from measurements; 4. a "fold-biased" sampling was introduced. This was necessary in order not to compare stretched (tips of gastric folds) with unstretched or compressed areas (depth of folds). After a series of trial measurements, two sites showed the criteria required of equal tickness: the intermediate area between two folds and the area between the tip and the depth of a fold. 5. At each site, the widths (lengths) of the mucosa (from the surface to the muscularis mucosae) and the lengths of the PASpositive and alcian-blue-positive zones were measured. 6. A total of two tissue blocks from 30 animals (15 in each group) was measured. For each section, a minimum of 10 measurements was performed (total number: 6900 separate measurements). The significance of the data was analyzed using Student t-test.

Scanning Electron Microscopy In Inactin®-Byk (125 mg/kg) anesthesia, 10 rats were first perfused with cacodylate buffer (0.1 M) through the left heart ventricle. The tissue was then perfusion-fixed by cacodylate buffered (0.1 M) 2% glutaraldehyde for 5-7 min. The stomachs were dissected and immediately floated in a Petri dish filled with glutaraldehyde, where they were opened. Care was taken to avoid any air-dyring. Free luminal and adhering mucus, which formed a loose layer was removed by applying a jet of glutaraldehyde from a syringe. Perfusion-fixed tissue strips (0.4 X 1 cm) were dissected from the corpus and antrum of the stomach and were immersion-fixed for 48 h. They were postfixed in 2% OS04' Following dehydration in graded series of ethanol, the tissue slices were criticalpoint dried (Polaron, Watford) and sputter-coated with gold palladium (Polaron, coating unit E5100) on a precooled tissue desk. The coated specimens were then investigated in an ISI-SEM. For photographs, Ilford PAN-F-films were used. The animal groups were as follows: three rats of each group in experiment II, and 4 controls which were only fasted over night.

Chemical Analysis Na-taurocholate (Fluca, Buchs) was analysed by DEAPSephadex-column chromatography and by separation of the frac-

tions by gas chromatography (Professor P. Back, M.D., Medical Clinics, University of Freiburg). The analysis gave the following results: - free bile acids: cholic acid desoxycholic acid - glycin-conjugated bile acids: cholic acid - taurine-conjugated bile acids: cholic acid desoxycholic acid chemodesoxycholic acid

8.5% 0.02% 0.5% 89.3% 0.9% 0.7%

Evaluation: predominantly taurocholate (89.3%); contamination caused mainly by free cholic acid (8.5%).

Results

Experiment I: Testing of the "Erosive" Dose of NaTC Animals of groups III and IV (80 mM and 2 x 80 mM NaTC) revealed numerous hemorrhagic erosions that were confined to the fundus area. The antrum was covered by masses of adhering mucus layers, but was nearly free of erosions. Table 2 shows the lesion score within the different treatment groups. Interestingly, the erosions were more numerous following a single dose of NaTC than after two successive NaTC-applications. By light microscopy, complete and incomplete erosions were seen (Fig. 2). It was observed that within the foveolae of the corpus region the zones of the mucus producing cells were smaller than in the control group.

Experiment II: Demonstration of the Cytoprotective Effect of Prostaglandin E2 Group 1: An increasing amount of free luminal and adhering mucus was found in the stomach of the prostaglandin Erpretreated animals. The mucosa was completely intact in most of the animals (Fig. 3 a). Erosions were observed only occasionally. Group 2: The stomachs of the NaTC-treated animals were extremely dilated. In the region of the fundus the mucosa showed multiple hemorrhagic erosions of different size (Fig. 3 b). The antrum was covered with dense adhering mucus layers. Table 2: Experiment I Group Number of Number of Time Concentra- Lesion animals intubation (hour) tion NaTC score (points) I II III IV V

5 5 5 5 5

1

1 1 2 1

2 2 2 2 2

20mM 0 40mM 0 80 mM 111 2x 80 mM 53 0.2 N He! 0

Gastric Cytoprotection of Prostaglandin E2 . 481

Fig. 2. Incomplete gastric erosion following application of 80 mM NaTC admitted to the animals by oesophagus intubation (Magnific. X 125).

Whereas in all NaTC-treated animals erosions were found (mean lesion score 26.6 ± 11.4 points), no erosions were seen in 45% of the prostaglandin E2-protected animals of group I following subsequent administration of an equal dose of NaTC (mean lesion score: 1.8 ± 1.7). These values correspond to a reduction of the lesion score of 93% (p < 0.0005) (Tab. 3).

Fig. 3 a. Stomach of a rat two hours after application of 80 mM NaTC. Multiple hemorrhagic erosions of different size in the mucosa of the fundus.

Table 3. Quantitative assessment of cytoprotective effect of PGE2. Experiment II group

number of animals lesion score with erosions without erosions mean ± SD

group I 10/18 (55%) 8/18 (45%) PGE2 + NaTC 20/20 (100%) 0/20 (0%) group II NaTC

1.8± 1.7 26.6 ± 11.4

Fig. 3 b. Stomach of a rat 2 hours after administration of 80 mM NaTC following a prior topic application of 200 I-tg/kg of PGE2 • No erosions visible.

482 . B. Fringes, D. Lorenz and W. Oehlert DISTRIBUTION OF THE LESION SCORE IN TAUROCHOLATE INJURED AND ANIMALS

PGEZ-PROTECTED ANIMALS

( %)

100 _

NA- TAUROCHOLAT ( 80 MM)

o

PGE 2 (200

JIG /KG)

+ NA-TAUROCHOLAT(SO MM) 50

o

5

20

10

30

40 P.OINTS LESION SCORE

Fig. 4. The most numerous gastric erosions with the greatest diameters (parameters transfered into points of the lesion score) were found to appear in the animals exclusively treated with NaTC, whereas in the PGEz-protected rats 45% of the animals were free of any erosions. Fig. 4 shows the distribution of the lesion score in both experimental groups. Light microscopic findings in the PGE2-pretreated, protected animals revealed that in the fundus mucosa, the zone of the PAS-positive surface epithelial cells as well as the alcian-blue-positive foveolar epithelial cells in the neck region of the gastric pits seemed to be broader than in group 2 animals. In addition, there was a conspicuous second, slightly PAS-positive band-like zone in the region of the mucus neck cells. In the fundus of group 2 animals similar changes were observed as described in experiment I, showing multiple incomplete and complete, hemorrhagic erosions (d. Fig. 2). The zones of the cells, producing PAS-positive and alcian-blue-positive mucus, appeared to be narrowed. Large amounts of free luminal and adhering masses of mucus were noted.

Quantitative Evaluation In order to confirm these observations of different zone widths of mucus-producing cells in the different groups, linear morphometry was applied to the PAS and the alcian-blue-stained tissue sections by use of a MOP. The data compiled revealed a significant increase in the zonelengths of mucus-producing cells in the PGEz-protected animals as compared both to the controls (4.7% for the PAS-positive zone and 3.2% for the alcian-blue-positive zone; p < 0.0005) and the NaTe-injured animals (8.1 % for the PAS-positive zone and 6.1 % for the alcian-bluepositive zone; p < 0.0005) (Table 4).

Table 4. Quantitative evaluation of the zone lengths of the mucus-producing cells in the stomach of the rat following NaTC damage and cytoprotection by PGE2

Grp. I PGE2 + NaTC n = 15 Grp. II NaTC n = 15 Contr. Grp. n =5

length of mucosa in 11m M±SD

length of alcian-blue- length of PAS-pos. pos. zone in !lm zone in 11m M±SD M±SD

relative length of relative length of alcian-blue pos. zone PAS-pos. zone in% in%

673 ±32.5

185.2 ± 25.7

150.2 ± 12.8

27.6± 3.8

22.4 ± 1.9

667.3 ± 36.8

143.3 ± 33.9

94.9 ±22.7

21.5 ± 5.1

14.3 ± 3.4

626.4 ± 35.9

152.9 ± 8.2

24.4 ±

17.7±1.7

111.1±11

1.3

Gastric Cytoprotection of Prostaglandin E2 . 483

Fig. 5. Surface cells in the antral mucosa of a control animal. The polygonal cells are flattened and sharply delimited by their cell borders. Opening of a gastric pit at left upper corner (x 5000).

Fig. 8. PGEz-protected animal: The entrances into the gastric pits are not visible due to the protruded surface cells. (Magnific. X 1000).

Fig. 6. Surface cell of a PGEz-protected animal. Note the extremely protruded surface membranes (X 5000).

Fig. 9. NaTC-treated rat: collapsed epithelial cell sacs due to explosive release of mucus (X 5000).

Fig. 7. A small erosion in the fundus mucosa of a NaTC-treated rat (arrow heads) (x 1000).

Fig. 10. PGErprotected rat: Single mucus vesicles assemble on the intact surface membrane forming a mucus-network (x 5000).

484 . B. Fringes, D. Lorenz and W. Oehlert

Scanning Electron Microscopy The gastric corpus mucosa of the control rats revealed a characteristic relief, with the longitudinal folds and a division of the mucosal surface into distinct polygonal areas; the openings of the gastric pits were clearly visible. The epithelial cells surrounding the gastric pits showed distinct intercellular borders, appearing as slightly elevated rims. A number of stump-like microvilli were observed (Fig. 5). Two main types of mucus secretion were visible: 1. small clouds of mucus leaving the openings of the gastric pits; 2. some surface epithelial cells releasing their mucus as single vesicles by exocytosis, Rarly, the entire apical part of the cell was cast off (apocrine secretion). In the NaTC-treated animals (group 2) the erosions appeared as areas covered with dense, adhering layers of mucus. Smaller erosions free of mucus were characterized as epithelial cells with defective surface membranes, devoid of cellular borders and microvilli (Fig. 7). Apart from these erosions, an explosive release of mucus by aprocrine secretion was also observed, leaving collapsed cell sacs (Fig. 9). In the prostaglandin Erpretreated animals, the gastric mucosal surface presented button-like, prominent epithelial cells (Figs. 6 and 8). The protruded surface membrane had a granular aspect; the microvilli were reduced in number, or no longer visible. There was a predominant release of mucus by exocytosis of single mucus vesicles which accumulate at the surface to form a net-like film (Fig. 10).

Discussion Bile salts break the gastric mucosal barrier and have been imglicated in the pathogenesis of gastritis and gastric u1cer12, ,33. After exposure of the gastric mucosa to bile under experimental conditions, a back diffusion of hydrogen ions occurs and trans mucosal electrial potential drops dramatically13, 18. It has been suggested that bile salts exert a detergent-like action on the gastric epitheliumS. The injurious effects of taurocholate are more pronounced at lower pH, which leads to a non-ionized lipophilic molecule able to permeate the mucosal membrane associated with solvation of membrane lipids 7,9, 10,20. Moreover, bile acids are suspected to reduce the energy metabolism of the ce1l 17• In our investigation, a single oral dose (80 mM) of sodium-taurocholate, a representative of the two major classes of conjugated bile acids proved to be able to induce erosions in a reproducible manner in the rat gastric mucosa. Both, complete and incomplete erosions were located within the fundus area where the H-ion concentration is greatest. In contrast, the mucosa of the antrum was nearly free of erosions. An explosive mucus release by apocrine secretion was observed to appear from the intact epithelium. Moreover, we made the interesting observation that the number of taurocholate induced gastric erosions decreased if 80 mM of this bile salt were given in two successive

doses. Similar effects have been described earlier by Chaudhury and Robert6, 24, 28 as "adaptive cytoprotection" which means that the injurious effects of noxinous agents may be inhibited by the prior application of a mild dose of the same substances. It is suggested that this effect might be due to a stimulation of the endogeneous prostaglandin synthesis. Our results suggest that "adaptive cytoprotection" may also be effective if two equally high doses of the same necrotizing agents are applied. The cytoprotective property of prostaglandins, mainly of the A and E types, has been described as effective against a variety of different noxious agents. This effect is time- and dose-dependent, hut is independent of the antisecretory properties of prostaglandins, since cytoprotection already appears at more than lOO-fold lower concentrations of these substances than are needed for the inhibition of gastric acid secretionS, 25. The following hypotheses as the possible mechanisms have been proposed: 1. a stimulation of mucus secretion; 2. direct stabilizing effects on the cell membrane (sodium pump, activation of adenylate cyclase) 3. a modification of the gastric mucosal circulation (for review see3, 4,16,27,30,31). The results of our study demonstrate that PGE2 significantly protects the rat gastric mucosa against the erosive influence of topical administered sodium taurocholate. A 90% decrease of the lesion score, a quantitative parameter of the number and size of mucosal erosions, was observed when the application of NaTC was preceded by a single topical dose of PGE2• By gross examination, larger amounts of mucus, distributed over the entire gastric surface were evident in the PGErprotected animals as compared to untreated controls. This is in accordance with the results obtained by Bolton et al. 4 and by Ruppin et al. 30 based on their investigations in rats, dogs and humans. Results, measuring the amount and thickness of the free luminal and adhering mucus masses are, however, quite contradictory 1, 29. We were interested in those epithelial cells still able to produce mucus substances. By measuring the zone-lenghts of neutral and acid mucus-producing cells within the gastric pits of the fundus area, a significant increase was observed in the PGE2 -pretreated animals compared to the NaTC-injured, as well as to the control group. In addition, the PAS-stained tissue sections in the PGE2 protected rats revealed a second, slightly PAS-positive band-like zone of cells that were located in the neck region of the gastric foveolae. It seems reasonable to assume that by the action of topical prostaglandins more cells could probably be induced to produce mucus substances by the activation of synthesis of secretory glycoproteins. Secretory glycoproteins have a minimal cellular prosttranslational transfer time from their site of synthesis in the RER to the cell membranes of about 20 to 30 minutes 39 • It may also be of interest that they can be cotranslationally and/or prosttranslationally modified by the addition of different oligosaccharide sidechains to the main molecule. Corresponding to our scanning electron microscopic results showing prominent surface cells in the corpus mucosa of PGErpretreated rats, transmission electron

Gastric Cytoprotection of Prostaglandin E2 . 485

microscopy detects ti~htly packed mucus vesicles within apical storage regions 1. This is in line with the observations of others32 • We also found a different preferred mode of mucus release by exocytosis of single mucus vesicles, that formed a network at the cell surfaces, thereby contributing to the development of a particularly resistent protective mucus layer. In conclusion, our studies show that the cytoprotective effect of prostaglandin E2 at the cellular level may be mediated by stimulated gastric mucus production and modification of its release. The increase in the defensive factors thus attained would result in a prevention of erosion development. In this context it would be of interest to ascertain whether modification of mucus release occurs in parallel with a modified formation of various glycoproteins of the mucopolysaccharides. This question may be elucidated by immunohistochemical studies. Acknowledgments. We would like to thank Mrs. E. Rieser and Mrs. G. Roth for excellent technical assistance. The authors are greatly indebted to Professor P. Back, M.D., Medical Clinics, University of Freiburg, for the chemical analysis of sodium-taurocholate. The assistance of Dr. S. M. de Looze and Mrs. L. Riedel in the translation of this paper is gratefully acknowledged.

References 1 Bickel M, Kauffman GL jr (1981) Gastric gel mucus thickness; effect of distention, 16, 16-dimethyl-Prostaglandin E2, and Carbenoxolone. Gastroenterol 80: 770-775 2 Black RB, Hole D, Rhodes U (1971) Bile damage to the gastric mucosal barrier: the influence of pH and bile acid concentration. Gastroenterol. 61: 178-184 3 Bolton ]P, Palmer D, Cohen MM (1976) Effect of the E2 prostaglandins on gastric mucus production in rats. Surg Forum 27: 402-403 4 Bolton ]P, Palmer D, Cohen MM (1978) Stimulation of mucus and nonparietal cell secretion by the E2 Prostaglandins. Arner] Dig Dis 23: 359-364 5 Carmichael HA, Nelson LM, Russell R] (1978) Cimetidine and Prostaglandin: Evidence for different modes of action on the rat gastric mucosa. Gastroenterol. 74: 1229-1232 6 Chaudhury TK, Robert A (1980) Prevention by mild irritants of gastric necrosis produced in rats by sodium taurocholate. Di~est Dis Sci 25: 830-836 Code ChF (1980) Defense mechanisms of the gastric mucosa. Scand ] Gastroenterol 16: 67 8 Davenpoort HW (1968) Destruction of the gastric mucosal barrier by detergents and urea. Gastroenterol 54: 175-181 9 Duane WC, Wiegand DM (1980) Mechanism by which bile salt disrupts the gastric mucosal barrier in the dog. J Clin Invest 66: 1044-1049 10 Eastwood G (1975) Effect of pH on bile salt injury of mouse gastric mucosa. A light- and electron microscopic study. Gastroenterol 68: 1456-1465 11 Fringes B, Lorenz D, Schulze-Schmeling E, Roth G, Oehlert W, Morphologische Aspekte des cytoprotektiven Effektes natiirlicher Prostaglandine an der Magenschleimhaut der Ratte. 11. Tagung der Pathologen am Oberrhein, Feldkirch 24. 4. 1982 (submitted to publication)

12 Heinkel K, Gugler R (1978) Der duodenogastrische Reflux und seine klinische Bedeutung. 10. Internat. Magen-Symposium, Berlin, 9.-12. Mai 1978, Acron Verlag 13 Ivey K] (1980) Bile salt induced back diffusion of hydrogen ions across gastric mucosa in man. Fact or fiction? Scand] Gastroenterol16: 75 14 Mann NS (1976) Bile-induced acute erosive gastritis. Its prevention by antacid, Cholestyramine and Prostaglandine E2. Amer] Dig Dis 21: 89-92 15 Miller TA, ]akobson ED (1979) Gastrointestinal cytoprotection by prostaglandins. GUT 20: 75-87 16 Nezamis ]E, Robert A (1980) Gastric mucus may mediate the cytoprotective effect of prostaglandins. Gastroenterol 78: 1228 (Abstract) 17 Ohe K, Hayashi K, Shirakawa T, Yamada K, Kawasaki T, Miyoshi A (1980) Aspirin and taurocholate-induced metabolic damage in mammalian gastric mucosa in vitro. Am] Physiol23 9: 457-462 18 Rees W, Rhodes] (1977) Bile reflux in gastro-esophageal disease. Clin Gastroenterol 6: 179-200 19 Rhodes], Barnado DE, Phillips SF (1969) Increased reflux of bile into the stomach in patients with gastric ulcer. Gastroenterol 57: 241-252 20 Ritchie WP jr (1977) Bile acids, the "barrier", and refluxrelated clinical disorders of the gastric mucosa. Surgery 82: 192-200 21 Robert A (1974) Effects of prostaglandins on the stomach and the intestine. Prostaglandins 6: 523-532 22 Robert A, Schultz ]R, Nezamis ]E, Lancaster C (1976) Gastric antisecretory and antiulcer properties of PGE2, 15-Methyl PGE2, and 16,16-dimethyl PGE2. Gastroenterol 70: 359-370 23 Robert A, Nezamis ]E, Lancaster C, Hanchar A] (1977) Gastric cytoprotective property of prostaglandins. Gastroenterol. 72: 1121 24 Robert A, Lancaster C, Hanchar Aj Nezamis ]E (1978) Mild irritants prevent gastric necrosis through Prostaglandin formation: Histological study. Gastroenterol 74: 1086 25 Robert A, Hanchar A], Lancaster C, Nezamis ]E (1978) Antisecretory and cytoprotective effects of postaglandin E2. Fed Proc 37: 46026 Robert A, Nezamis ]E, Lancaster C, Hanchar A] (1979) Cytoprotection by Prostaglandins in rats: prevention of gastric necrosis produced by alcohol, HCI, NaOH, hypertonic NaCl and thermal injury. Gastroenterol 77: 433-443 27 Robert A (1979) Cytoprotection by prostaglandins. Gastroenterol 77: 761-767 28 Robert A, Nezamis ]E, Lancaster C et al (1983) Mild irritants prevent gastric necrosis through "adaptive cytoprotection" mediated by prostaglandins. Am] Physiol 245: 113-121 29 Robert A, Bottcher W, Golanska E, Kauffman GL jr (1984) Lack of correlation between mucus gel thickness and gastric cytoprotection in rats. Gastroenterol 86: 670-674 30 Ruppin H, Person B, Robert A, Domschke W (1981) Gastric cytoprotection in man by prostaglandin E2. Scand] Gastroenterol16: 647-652 31 Schiessel R, Matthews ], Barzilai A, Merhav A, Silen W (1980) PGE2 stimulates gastric chloride transport: possible key to cytoprotection. Nature 283: 671-673 32 Schmidt KL, Henagan ]M, Hilburn P], Miller ATA (1983) Prostaglandin cytoprotection against alcohol-induced gastric injury: a histologic study. Lab Invest 48: 75 A 33 Schumpelick V, Begemann F, Bandomer G, Grossner D, Doehn M (1978) Intragastrale Gallensiiuren und Lysolezithin bei klinischer StreSulkus-Gefiihrdung. Dtsch Med Wschr 103: 735-737

486 . B. Fringes, D. Lorenz and W. Oehlert 34 Takeguchi C, Sih CJ (1972) A rapid spectrophotometric assay for prostaglandin synthetase; application to the study of non-steroid antiinflammatory agents. Prostaglandins 2: 169-184 35 Tr6tschler H, Koch HK, Oehlert W (1978) Vergleichende licht- und rasterelektronenmikroskopische Untersuchung der Magenschleimhaut im Biopsiematerial. Beitr elektronenmikroskop Direktabb Oberfl11: 335-342 3 Vane JR (1971) Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature New Bioi 231: 232-235

37 Wilson DE (1974) Prostaglandins. Their action on the gastrointestinal tract. Arch Intern Med 133: 112-118 38 Zalewski ChA, Moody FG (1979) Mechanisms of mucus release in exposed canine gastric mucosa. Gastroenterol 77: 719-729 39 Hannover JA, Lennarz WJ (1981) Transmembrane assembly of membrane and secretory glycoproteins. Arch Biochem Biophys 211: 1-19

Received October 15, 1984 . Accepted November 27, 1984

Key words: Cytoprotection - Prostaglandin E2 - Taurocholate - Gastric erosions Professor W. Oehlert, M.D., Pathologisches Institut, AlbertstralSe 19, D-7800 Freiburg