Interrelationships between the development of the gastric cytoprotective effects of prostacyclin, atropine, cimetidine and the gastric mucosal superoxide dismutase activity in rats

Gastric mucosal damage was produced by the intragastric administration of 96% ethanol or 0.6 M HCl. The cytoprotective doses of prostacyclin (PGI2) (5 micrograms/kg), atropine (0.025 mg/kg) or cimetidine (2.5 mg/kg) were given intraperitoneally 30 min before the administration of the necrotizing agents. The animals were killed 1 hr later. The number and severity of gastric mucosal lesions (ulcer) were recorded. At the time of the sacrifice of the animals, superoxide dismutase (SOD) was prepared from the gastric fundic mucosa and its activity was measured. It was found that PGI2 (5 micrograms/kg), atropine (0.025 mg/kg) and cimetidine (2.5 mg/kg) significantly decreased the number and severity of gastric mucosal lesions (ulcers) produced by the intragastric administration of 96% ethanol a 0.6 M HCl, PGI2, atropine, cimetidine, given in cytoprotective doses, significantly mounted the ethanol-induced increase of gastric mucosal SOD activity; PGI2, atropine, cimetidine, given them in cytoprotective doses significantly shunted the HCl-induced decrease of gastric mucosal SOD activity. It has been concluded that; chemically different cytoprotective agents (PGI2, atropine, cimetidine) give rise to similar tendencies in the changes of gastric mucosal SOD activity; both the significant decrease (in the ethanol-model) and the significant increase (in the HCl-model) of this enzyme seem to be involved in the development of gastric mucosal protection by PGI2, atropine and cimetidine.     

The antiarrhythmic action of prostacyclin (PGI2) on aconitine induced arrhythmias in rats.

Prostacyclin (PGI2) produces an antiarrhythmic effect on aconitine induced arrhythmias in rats. The ED50 of PGI2 was 0.7 microgram/kg and the maximum antiarrhythmic effect 54 per cent. The equi-effective doses of PGE2 and PGF2alpha were higher (ED50 of PGF2alpha = 1.2 microgram/kg, ED50 of PGE2 = 2.7 microgram/kg). However, PGF2alpha and PGE2 had a maximum antiarrhythmic effect of 80 per cent in this model.     

Effects of intracerebroventricular administration of prostaglandins I2, E2, F2 alpha and indomethacin on blood pressure in the rat.

The effects of intraventricularly administered prostaglandins I2 (PGI2), E2 (PGE2), F2alpha (PGF2 alpha) and indomethacin on systemic blood pressure were investigated in conscious rats. PGI2 (1.25--10 micrograms/kg) decreased blood pressure in a dose-related manner, whereas PGE2 (100--1000 mg/kg) dose-dependently increased blood pressure. Both PGF2 alpha (0.31--20 micrograms/kg) and indomethacin (0.625--40 micrograms/kg) had no effects on blood pressure. These results indicate that intraventricular injection of PGI2 or PGE2 can induce significant changes in blood pressure, while endogenous prostaglandins synthesized in the brain seem to play a minor role in direct regulation of systemic blood pressure in the rat.     

Effect of indomethacin, tiaprofenic acid and dicrofenac on rat gastric mucosal damage and content of prostacyclin and prostaglandin E2

Gastric ulcerogenicity and depletion of endogenous prostaglandins (PGs) content induced by tiaprofenic acid, dicrofenac and indomethacin were examined using the same antiinflammatory effective doses. Male Wistar rats were given each of these drugs intragastrically 24, 18, and 3 hrs before sacrifice in the following doses (mg/kg): indomethacin (0.8, 4 and 20); tiaprofenic acid (1.2, 6 and 30); dicrofenac (0.8, 4 and 20). Endogenous prostacyclin (PGI2) and PGE2 in fundic mucosa were determined by radioimmunoassay. The three compounds produced fundic mucosal lesions in a dose-dependent manner. However, tiaprofenic acid and dicrofenac were both less potent than indomethacin in producing gastric mucosal lesions at similar antiinflammatory doses. Mucosal PGE2 content was abolished by the three compounds in the following doses (mg/kg): indomethacin (4 and 20); tiaprofenic acid (6 and 30); dicrofenac (20). Mucosal PGI2 was maintained around 50% of the control value in rats given tiaprofenic acid in a dose of 6 mg/kg or dicrofenac in a dose of 4 mg/kg, while indomethacin in a dose of 4 mg/kg markedly reduced mucosal PGI2 to 17% of the control value. In larger doses, tiaprofenic acid and dicrofenac were also significantly less potent in reducing mucosal PGI2 than indomethacin. These results suggest that the difference in ulcerogenicity between indomethacin and the other two compounds was closely related to their potency in decreasing PGI2 in the gastric (fundic) mucosa.     

Effects of inflammatory mediators on canine airway neuromuscular function.

Inflammatory mediators can both enhance or inhibit canine airway reactivity. PGE2 and PGI2 in general are inhibitory, interfering with release of acetylcholine and with responses to bronchoconstrictors. These prostaglandins may be more effective against agonists that open voltage-dependent Ca2+ channels to induce Ca2+ influx and contraction compared with those agonists that release internal Ca2+. Other mediators are excitatory: histamine, PGD2, thromboxane A2 (TxA2), and leukotrienes (LT) C4, D4, and E4. In canine airway only histamine and TxA2 have effects in the absence of indomethacin, i.e., in the presence of the large amounts of PGE2 and PGI2 produced in vitro LTs are ineffective. Effects of TxA2 and histamine may be potentiated if the synthesis of these inhibitory PGs is inhibited. Whether histamine or TxA2 normally promote synthesis and release of PGE2 and PGI2 in a kind of homeostasis remains to be explored. It is also unclear whether pre- as well as post-junctional TxA2 receptors exist and have different pharmacological sensitivities to antagonists. LTC4 and LTD4 also constrict canine bronchi but only when PGE2 and PGI2 synthesis is blocked and, again, whether this is a result of LT-induced release of inhibitory mediators is unknown. The concept that airway responsiveness can be caused by turning off PGE2 and PGI2 production and turning on TxA2 or LT production (or unmasking their actions) needs further exploration. Our recent data suggest that such a mechanism may explain ozone-induced responsiveness in dogs.     

Influence of 15-keto-metabolites and 13,14-dihydro-15-keto-metabolites of PGE2 and PGF2 alpha as well as of 6-keto-PGF1 alpha as a metabolite of PGI2 on aconitine induced cardiac arrhythmia in rats

The 15-keto-metabolites of PGE2 and PGF2 alpha produced an antiarrhythmic effect on aconitine induced arrhythmias in rats. The ED50 values of these metabolites were approximately 2.0 micrograms/kg. The 13,14-dihydro-15-keto-metabolites of PGE2 and PGF2 alpha had no statistically significant antiarrhythmic effect. PGI2 (0.25-1.00 micrograms/kg) produced an antiarrhythmic effect between 15-54% (ED50 0.75 micrograms/kg), whereas 6-keto-PGF1 alpha, a metabolite of PGI2, showed no significant antiarrhythmic effect. The results suggest a participation of 15-keto-metabolites in the antiarrhythmic effects of PGE2 and PGF2 alpha.     

Cytoprotective and antisecretory properties of a non-diarrheogenic and non-uterotonic prostacyclin analog: U-68,215

U-68,215 [15-Cyclohexyl-9-deoxo-13,14-dihydro-2',9 alpha-methano-4,5,6,16,17,18,19,20-octanor-3-oxa-3,7-(1', 3'-interphenylene)-PGE1] is a stable prostacyclin analog. When given orally to rats, it is cytoprotective for the stomach (ED50: 0.8 micrograms/kg) and the intestine (ED50: 22 micrograms/kg), is gastric antisecretory (ED50: 35 micrograms/kg) and antiulcer (aspirin) (ED50: 5 micrograms/kg). The oral antisecretory ED50 in dogs is 50 micrograms/kg. It has a long duration of gastric cytoprotection: 8-10 hours compared to 3 hours for 16,16-dimethyl PGE2. Unlike most prostaglandins of the E type, it is not diarrheogenic (not enteropooling), it does not induce cellular proliferation of the gastrointestinal mucosa, when given twice a day for eight days, it is not uterotonic (in monkeys), and it does not prevent embryo implantation in hamsters. It inhibits ex vivo platelet aggregation (ED50: 300 micrograms/kg), but does not promote bleeding from cut vessels nor from gastric ulcers. U-68,215 lowers blood pressure at an oral dose corresponding to 1-5 times the antisecretory ED50 in rats and dogs, and to 150 times the cytoprotective ED50 in rats. It may be of therapeutic value in the treatment of conditions where inhibition of gastric acid secretion is desirable, e.g., gastric and duodenal ulcer, and in conditions responding to cytoprotection, e.g., stress ulcers, hemorrhagic gastritis and gastric erosions associated with nonsteroidal antiinflammatory drugs.     

Characterization of prostaglandin (PG)-binding sites expressed on human basophils. Evidence for a prostaglandin E1, I2, and a D2 receptor.

Recent data suggest that prostaglandins (PGs) are involved in the regulation of basophil activation. The aim of this study was to characterize the basophil PG-binding sites by means of radioreceptor assays using 3H-labeled PGs. Scatchard analysis for pure (greater than 95%) chronic myeloid leukemia (CML) basophils revealed two classes of PGE1-binding sites differing in their affinity for the natural ligand (Bmax1 = 217 +/- 65 fmol/10(8) cells; Kd1 = 0.5 +/- 0.2 nM; Bmax2 = 2462 +/- 381 fmol/10(8) cells; Kd2 = 47 +/- 20 nM; IC50 = PGE1 less than PGI2 less than PGD2 less than PGE2 less than PGF2 alpha) as well as two classes of PGI2 (iloprost)-binding sites (Bmax1 = 324 +/- 145 fmol/10(8) cells; Kd1 = 0.5 +/- 0.3 nM; Bmax2 = 2541 +/- 381; Kd2 = 27 +/- 6 nM; IC50 = PGI2 less than PGE1 less than PGD2 less than PGE2 less than PGF2 alpha. In addition, CML basophils exhibited a single class of PGD2-binding sites (Bmax = 378 +/- 98 fmol/10(8) cells; Kd = 13 +/- 4 nM; IC50: PGD2 less than PGI2 less than PGE1 less than PGE2 less than PGF2 alpha). In contrast, we were unable to detect specific saturable PGE2-binding sites. Primary and immortalized (KU812) CML basophils revealed an identical pattern of PG receptor expression. Basophils (KU812) expressed significantly (p less than 0.001) lower number of PGE1 (PGI2)-binding sites (Bmax1: 9% (20%) of control; Bmax2: 36% (50%) of control) when cultured with recombinant interleukin 3 (rhIL-3), a basophil-activating cytokine, whereas rhIL-2 had no effect on PG receptor expression. Functional significance of binding of PGs to basophils was provided by the demonstration of a dose-dependent increase in cellular cAMP upon agonist activation, with PGE1 (ED50 = 1.7 +/- 1.1 nM) and PGI2 (ED50 = 2.8 +/- 2.3 nM) being the most potent compounds. These findings suggest that human basophils express specific receptors for PGE1, PGI2 as well as for PGD2.     

Prevention by aluminium phosphate of gastric lesions induced by ethanol in the rat: role of endogenous prostaglandins and sulfhydryls

This study was designed to demonstrate the cytoprotective effect of an antacid containing aluminium phosphate (Phosphalugel) against ethanol-induced gastric injury in the rat and to determine whether this cytoprotective effect is mediated by endogenous prostaglandins and sulfhydryls. We have quantitatively evaluated gastric mucosal lesions using macroscopic and histological techniques one hour after ethanol administration. Two ml of aluminium phosphate given orally one hour before administration of 2 ml of 100% ethanol significantly (p less than 0.01) reduced the area of macroscopic lesions induced by ethanol (3.3 +/- 0.9%) when compared to distilled water (20 +/- 4.8%). The histological study showed that aluminium phosphate prevented deep tissue necrosis. However, it did not protect surface epithelial cells against ethanol injury. Pretreatment with indomethacin, 5 mg/kg sc one hour before aluminium phosphate, slightly but significantly (p less than 0.05) reduced the cytoprotective effect of aluminium phosphate. Macroscopic lesions occupied 4.3 +/- 0.94% and 1.88 +/- 0.41% of total mucosal area in indomethacin group and in vehicle group, respectively. On the other hand, the sulfhydryl blocker, N-ethyl-maleimide, 10 mg/kg sc, given one hour before aluminium phosphate, completely abolished the cytoprotective effect of aluminium phosphate (32.92 +/- 4.85% in N-ethyl-maleimide group versus 3.78 +/- 1.41% in vehicle group; p less than 0.01). These results show that aluminium phosphate has a cytoprotective effect against ethanol injury in the rat. This property appears to be mediated by both endogenous prostaglandins and sulfhydryls.     

Hypotensive response to prostacyclin and 6-keto-PGE1 following hepatectomy in the rat

Prostacyclin (PGI2) is metabolized to 6-keto-prostaglandin E1 (6-keto-PGE1) which is more stable yet equipotent to PGI2 in lowering systemic arterial blood pressure in the dog. In this study, partial hepatectomy was performed to determine the role of the liver in the vasodepressor response to both intravenously administered PGI2 and 6-keto-PGE1. The magnitude and the duration of systemic hypotensive responses were measured in hepatectomized and sham-operated male Wistar rats following less than maximal, equidepressor doses of PGI2 (0.3 microgram/kg), 6-keto-PGE1 (1.0 microgram/kg), and also PGE1 (3.0 micrograms/kg) and PGE2 (3.0 micrograms/kg). Hepatectomy did not significantly alter the magnitude of the systemic hypotensive response to any of the prostaglandins tested. This indicates that the liver and hepatic circulation do not contribute significantly to the hypotensive effect of these prostaglandins by alterations of systemic vascular resistance, venous pooling of blood, or the generation of additional vasoactive metabolites as may be expected following administration of these prostaglandins. However, hepatectomy did significantly increase the duration of the hypotensive response to PGI2 and 6-keto-PGE1 but not PGE1 or PGE2. We conclude that in vivo, the liver has a more significant role in PGI2 and 6-keto-PGE1 inactivation than in the inactivation of PGE1 and PGE2 when administered intravenously. These results also support the relatively greater significance of the lung in the inactivation of PGE1 and PGE2 in vivo.     

Prostaglandin synthesis by the cochlea of the guinea pig. Influence of aspirin, gentamicin, and acoustic stimulation

This study describes the synthesis of prostaglandins (PGs) by the vascular structures of the inner ear (lateral wall = stria vascularis and spiral ligament) in vitro. The main PGs produced were PGI2, PGF2 alpha and PGE2. PGI2 and PGF2 alpha were also found in the perilymph. A 350 mg/kg ip injection of aspirin decreased PG synthesis by the lateral wall and PG levels in perilymph. This effect was reversed after 3 days. Gentamicin (10(-9) to 10(-5) M) decreased significantly and reversibly PG synthesis in vitro, as did 100 mg/kg ip injection. Acoustic stimulation increased ex vivo PGI2 and PGE2 synthesis without modifying PG levels in perilymph. Results suggest that PGs could be one humoral mediator of the cochlear microcirculation homeostasis, and, possibly, of the circulatory disturbances reported after acoustic stimulation. The decreased PG synthesis after gentamicin treatment could account for the angiotoxic component observed in aminoglycoside ototoxicity.     

Metabolism of prostaglandin endoperoxide by microsomes from human lung parenchyma and comparison with metabolites produced by pig, bovine, rat, mouse and guinea-pig

The metabolism of PGH2 by human lung parenchymal microsomes was characterized by radiometric high performance liquid chromatography and compared with metabolism by pig, bovine, rat, mouse, and guinea pig lung microsomes. Microsomes from human lung synthesized 0.74 nmoles/mg protein and 0.72 nmoles/mg protein, PGI2 (6-Keto-PGF1 alpha) and TxA2 (TxB2) respectively, upon incubation with 4.0 nmoles of PGH2. Pig, bovine, rat, mouse, and guinea pig microsomes respectively synthesized 1.0, 1.0, 0.9, 0.4, and 0.1 nmoles of PGI2/mg protein, and 0.9, 1.0, 0.7, 0.3, 1.8 nmoles of TxA2/mg protein, and preparations formed some PGE2, PGF2 alpha, and PGD2. Mouse lung microsomes were unique in synthesizing PGE2 as the major prostaglandin. The thromboxane synthetase inhibitor 1-benzylimidazole was a specific inhibitor in these six species.     

Influence of some prostaglandins and prostaglandin analogues on PAF-induced shock in mice.

Prostaglandins and Prostaglandin-analogues were investigated for their ability to protect mice from platelet-activating factor (PAF) induced shock. 75% mortality in female NMRI mice was induced by i.v. injection of 75 micrograms/kg PAF. Nileprost and PGE1, the most potent substances, produced a dose dependent protection against PAF. Iloprost and PGI2 were less effective. PGE2, nalador, flunoprost and U 46619 were neither protective nor deleterious. The strong difference in the effectiveness between the two prostaglandins of the E-series and the poor effect of PGI2 and the PGI2 analogue is remarkable. Flunoprost and U 46619 that increased the TXB2 synthesis or release in two experimental models did not enhance the PAF mortality; TXA2 seems to be only a secondary mediator of the acute PAF-induced death.     

Anti-ulcer compound from Voacanga africana with possible histamine H2 receptor blocking activity.

Voacanga africana is used in Cameroonian ethnomedicine for the treatment of peptic ulcers. We have tested the cytoprotective, anti-secretory and ulcer healing actions of an alkaloid (TN) obtained from the fruit extract. Oral administration of TN (50-100 mg/kg) dose-dependently prevented ulcer formation by HCl/ethanol (36-75%), absolute ethanol (43-75%), HCl-ethanol/indomethacin (58-84%), Pylorus ligation (31-100%), cold restraint stress (68-100%) and histamine (49-100%). The inhibitory effect at 50 and 100 mg/kg against HCl/ethanol was not suppressed by pre-treatment with indomethacin (20 mg/kg, i.p.). TN reduced Shay-ligated gastric acid secretion from 77 mEq/l in the controls to 46 and 25 mEq/l for the 50 and 100 mg/kg doses. Augmented histamine-induced gastric acid secretion was reduced from 84 mEq/l in the controls to 45 and 21 mEq/l for the two doses of TN, with total inhibition of gastric and duodenal ulcers by the 50 mg/kg dose. Healing rate of chronic acetic acid-induced ulcers was 62 and 83%, respectively, for the dose of 50 and 100 mg/kg of TN compared with the controls. TN has gastric anti-secretory effects similar to histamine receptor blockers. Its cytoprotective and ulcer healing properties are related to its ability to strengthen gastric mucosal defenses through enhanced gastric mucus production.     

Contents of prostaglandins F2alpha, I2 (prostacyclin) and thromboxane A2 in the dog liver during development of hemorrhagic shock]

Contents of prostaglandins (PGs) F2 alpha, I2 and thromboxane (TH) A2 in the liver of 6 dogs has been studied in dynamics of hemorrhagic shock development. The results show that acute haemorrhage (31.9 +/- 1.6 ml/kg) stimulates the synthesis and release of all the eicosanoids studied. Nevertheless, in the progression of shock a significant difference was noticed between the high TXA2 and PGF2 alpha rates and comparatively lower rise of PGI2 production. The late stage of shock (201 +/- 44 minutes after haemorrhage) was characterized by a further rise in TXA2 and PGF2 alpha rates with simultaneous lowering of PGI2 as a result of vascular endothelium affection which is the main source for prostacyclin synthesis. In view of the antagonism of PGI2 to TXA2 and PGF2 alpha and its cytoprotective effects, the results of present investigation give the experimental background for using prostacyclin or its synthetic analogues in a complex therapy of acute liver failure in clinic.     

Prostaglandins in enhanced pressor response in renal prehypertensive rabbits

The role of prostaglandins (PGs) in the pressor response to norepinephrine (NE) was examined in one-kidney, one clip rabbits with renal artery stenosis for 3-day's duration (3-day clipped rabbits) and in sham operated rabbits with one-kidney without renal artery stenosis. An exaggerated pressor response to NE, 800 ng/kg/min, was observed in the 3-day clipped rabbits, and it was abolished by angiotensin II antagonist, [Sar1, Ile8] angiotensin II (AIIA). Treatment with indomethacin, 10 mg/kg, induced hyperresponsiveness to NE in the sham operated rabbits and also produced a further increase in the response in the 3-day clipped rabbits: the enhanced responses with similar levels were not attenuated by AIIA in both groups. A subdepressor dose of PGE2, 800 ng/kg/min, abolished the hyperresponsiveness in the 3-day clipped rabbits, while subdepressor or depressor dose of PGI2, 10 or 20 ng/kg/min did not, but the concurrent infusion of AIIA with PGI2 attenuated it. These results indicate that PGs, in particular PGE2 might be involved in the enhanced pressor response to NE in the 3-day clipped rabbits in addition to the altered renin-angiotensin system.     

The actions of prostaglandins I2 and E2 on arrhythmias produced by coronary occlusion in the rat and dog.

Prostaglandins E2 and I2 were compared with known antiarrhythmics for their actions against arrhythmias produced by occlusion of the left anterior descending coronary artery in the anaesthetised rat while PGI2 was also examined in the dog. PGI2 in the dog suppressed early arrhythmias produced during occlusion but did not influence those produced by occlusion-release or those occurring 24 hours after a permanent occlusion; none of the A,B,C or D series prostaglandins tested markedly reduced 24 hour arrhythmias. In the rat PGE2 was antiarrhythmic against early occlusion arrhythmias (30 minutes occlusion) in a dose related manner (infusions of 1-4 microgram/kg/min) whereas PGI2 infusions potentiated the arrhythmogenic effect of occlusion. PGE2 was as effective an antiarrhythmic as 10mg/kg Org. 6001 which was more effective in this test situtation than dl-propranolol. No obvious mechanisms for the actions of PGE2 or PGI2 were apparent although both agents lowered blood pressure and reduced the size of the occluded zone produced by ligation.     

Interleukin-1 is cytoprotective, antisecretory, stimulates PGE2 synthesis by the stomach, and retards gastric emptying

Human recombinant interleukin 1 beta (IL-1) administered intraperitoneally to rats produced the following gastric effects: 1. It was cytoprotective, preventing gastric mucosal necrosis produced by oral administration of one ml of absolute ethanol to fasted animals. The ED50 was 1200 units/kg (110 ng per animal). IL-1 was 125 times more potent than prostaglandin E2 (on a weight basis), and 6,000 times more potent (on a molar basis). 2. The cytoprotective effect of IL-1 was blocked by indomethacin (inhibitor of prostaglandin synthesis) and by IRAP (a specific interleukin-1 receptor antagonist protein). IRAP did not inhibit cytoprotection induced by PGE2. 3. IL-1 prevented the formation of gastric erosions induced by aspirin. 4. IL-1 inhibited gastric secretion (volume, acid concentration and output), in the pylorus-ligated rat, with an ED50 of 300 units/kg (3.2 ng per animal). 5. Indomethacin and IRAP blocked the antisecretory effect of IL-1. 6. IL-1 retarded gastric emptying, an effect blocked by IRAP, but not by indomethacin. 7. IL-1 increased synthesis of prostaglandin E2 by the gastric mucosa by 111%. IL-1 is the most potent of known agents that are gastric cytoprotective, antiulcer, antisecretory, and delay gastric emptying. It appears to act mostly by stimulating the synthesis of prostaglandins by the stomach. These studies suggest that the stomach possesses IL-1 receptors. These are probably located on parietal cells (that produce acid), on prostaglandin-producing cells, on smooth muscle cells (responsible for gastric emptying), and on as yet unidentified cells involved in gastric cytoprotection. Both IL-1 and IRAP, being natural substances, may play a physiological role in the maintenance of gastric mucosal integrity, and in the regulation of acid secretion and gastric motility.     

Effects of intracerebroventricular administration of prostaglandins I2, E2, F2α and indomethacin on blood pressure in the rat

The effects of intraventricularly administered prostaglandins I₂ (PGI2), E₂ (PGE2), F_2α (PGF2α) and indomethacin on systemic blood pressure were investigated in conscious rats. PGI2 (1.25 – 10 g/kg) decreased blood pressure in a dose-related manner, whereas PGE2 (100 – 1000 ng/kg) dose-dependently increased blood pressure. Both PGF2α (0.31 – 20 μg/kg) and indomethacin (0.625 – 40 μg/kg) had no effects on blood pressure. These results indicate that intraventricular injection of PGI2 or PGE2 can induce significant changes in blood pressure, while endogenous prostaglandins synthesized in the brain seem to play a minor role in direct regulation of systemic blood pressure in the rat.     

Effect of H. pylori infection on the expression of cyclooxygenase-2 in human gastric mucosa

Cyclooxygenase-1 is the primary isoform responsible for the production of cytoprotective prostaglandins (PGE(2) and PGI(2)) in the stomach. In contrast COX-2 is induced at the sites of inflammation. Using Helicobacter pylori infection as a model of inflammation, this study was designed to evaluate the effects of H. pylori infection on prostanoid synthesis and expression of COX-2 in human gastric mucosa.Prostaglandin (PGE(2)) and prostacyclin (PGI(2)) synthesis in gastric biopsies obtained from 21 patients undergoing diagnostic endoscopy, were determined. H. pylori was detected by CLO test, histology and culture. Biopsy samples were incubated either with NS-398, selective COX-2 inhibitor or aspirin. Samples were also treated with endotoxin (LPS) in order to induce COX-2 expression. Tissue was also analysed for COX-2 expression in vivo by immunohistochemistry.In 15 out of 21 patients, H. pylori was detected by at least two of the three methods. Higher levels of PGE(2) and PGI(2) were seen in patients infected with H. pylori (191+/-30 and 245+/-88ng/mg protein, respectively) compared with non-infected patients (77+/-17 and 120+/-36ng/mg protein, respectively). There was significant inhibition of PGE(2) and PGI(2) with aspirin in both H. pylori infected (28+/-6.6 and 53+/-43ng/mg, respectively) and in non-infected patients (16+/-7 and 12.5+/-3.5ng/mg protein, respectively). However, NS-398 and LPS did not alter prostaglandin function significantly. Immunohistochemistry in all patients irrespective of Hp status demonstrated expression of COX-2.Lower concentration of constitutive expression of COX-2 was detected in human gastric mucosa by immunohistochemistry, however, H. pylori infection failed to induce COX-2 protein. In addition, increased prostaglandin synthesis in Hp-infected patients appears to be COX-1 mediated rather than COX-2. Furthermore, failure of endotoxaemia-treated sample to produce more PGE(2) in the face of enhanced COX-2 expression in gastric mucosa further suggests that increased prostanoids in human gastric stomach are COX-1 mediated.