Gastric cytoprotection is secondary to increased mucosal fluid secretion: a study of six cytoprotective agents in the rat.

We tested the hypothesis that rapidly developing gastric cytoprotection produced by topical application of exogenous compounds is a result of increased gastric mucosal fluid secretion. Ex vivo gastric chambers were prepared in rats which were subsequently exposed topically to one of the prostaglandin (PG) E1 analogues misoprostol or rioprostil, PGE2, nicotine, N-ethylmaleimide (NEM), 0.25 M HCl, or to their respective vehicles. All agents were added to empty chambers to avoid complications resulting from dilution by gastric contents. Effects of these agents on intraluminal volume changes, blood flow, juxtamucosal pH, histology, and on the mucosal damage resulting from necrotizing agents were studied. All six agents were cytoprotective and each increased net secretion of fluid by the chambered mucosae. Gastric blood flow was not significantly increased by NEM, by 0.25 M HCl, or by nicotine compared to controls, and the juxtamucosal pH was not significantly increased by any of the three agents for which this was studied. Vacuole formation in surface epithelial cells and subepithelial edema were seen after exposure to some agents, but none of the agents led to formation of a thick barrier of exfoliated cells and mucus. Ablation of primary afferent nerves with capsaicin abolished both protection by 0.25 M HCl and the net increase in fluid secretion by the mucosae. Capsaicin ablation did not alter either the protection afforded by NEM or the increase in volume of secretion. We conclude that increased mucosal fluid secretion is the common factor present with all six cytoprotective agents and hence may be the predominant mechanism of cytoprotection against topically applied necrotizing agents.

Prostaglandins in the treatment of reflux oesophagitis: double-blind placebo controlled clinical trial.

The efficacy of the prostaglandin analogue, rioprostil, in the treatment of reflux oesophagitis has been assessed in a double-blind, randomized, placebo-controlled trial of 25 patients with endoscopic and histological evidence of reflux oesophagitis. At the beginning and end of the study, endoscopic appearances were graded 0-4 (0 = no oesophagitis, 4 = severe oesophagitis) and the symptoms of heartburn, regurgitation, pain and dysphagia were each graded 0-3 (0 = none, 3 = severe). Fourteen patients received rioprostil, 300 micrograms twice daily, and 11 patients received identically marked placebo for a period of 12 weeks. At the end of the study there were no significant differences between the groups in mean (s.d.) endoscopic grading (rioprostil 2.4 (1.3); placebo 1.9 (0.9)) and mean (s.d.) cumulative symptom score (rioprostil 2.5 (3.1); placebo 2.6 (1.5)). Five patients in the rioprostil group reported diarrhoea. Rioprostil had no significant benefit over placebo in the treatment of reflux oesophagitis.

In vitro characterization of prostanoid EP-receptors in the non-pregnant human myometrium.

1. Prostaglandin receptors of the PGE type have been characterized in the non-pregnant human myometrium in vitro according to the scheme of Coleman et al. (1984) by use of the agonists PGE2, sulprostone, rioprostil, AY23626, butaprost, misoprostol, 16,16-dimethylprostaglandin E2, enprostil and iloprost, and, the antagonist AH6809. 2. All prostanoids tested were active in non-pregnant human myometrium either as stimulators and/or inhibitors of spontaneous activity or both. Biphasic responses to PGE2 indicate that at least two receptor types of the EP-receptor exist, one mediating relaxation and the other mediating contraction. 3. Further evidence for the EP-receptor mediating excitation and relaxation was provided by the action of the EP2-/EP3-receptor selective prostanoids rioprostil, AY23626 and misoprostol, and the EP1-/EP2-receptor selective agonist 16,16-dimethylprostaglandin E2. 4. Butaprost, an EP2-receptor selective agonist, produced potent inhibition of spontaneous activity in the tissue which was generally longer-lasting than that evoked by the natural prostanoid PGE2. 5. The EP1-/EP3-receptor selective agonist sulprostone and the EP3-receptor agonist enprostil produced potent contractile responses supporting the presence of contractile EP3-receptors in the non-pregnant human myometrium in vitro. 6. The EP1-/IP-receptor selective agonist, iloprost, produced mixed responses in non-pregnant human myometrium. The contractile response was inhibited by the EP1-receptor antagonist AH6809. However, responses to the EP1-/EP3-receptor selective agonist sulprostone were unaffected by AH6809 which may indicate that only a small population of EP1-receptors is present. 7. Therefore it would seem that a heterogeneous population of EP-receptors is present in the non-pregnant human myometrium.

BAY P 1455, a thiazolylaminobenzimidazole derivative with gastroprotective properties in the rat.

The antiulcer activity of BAY P 14551 a thiazolylaminobenzimidazole derivative, was evaluated in different experimental ulcer models and its antiulcer activity was compared to that of different reference drugs. The overall activity of the compound was equal to or more potent than that of reference antiulcer drugs, such as pirenzepine, cimetidine and carbenoxolone, but it was not as potent as rioprostil. The ED50 values (expressed as mumol/kg p.o.) were 68 (confidence limits: 51-91) for indomethacin-induced ulcers, 21 (confidence limits: 13-31) for stress-induced ulcers and 1260 mumol/kg p.o. (confidence limits: 412-3800) for ulcers induced by absolute ethanol. The compound had no activity against cysteamine-induced duodenal ulcers and lost its cytoprotective activity in adrenalectomised rats. Since inhibition of gastric acid secretion was seen, if at all, only with the higher doses, the gastro-protective action of BAY P 1455 seemed not to be due to an antisecretory effect, but more likely to a gastroprotective action as hypothesised for prostaglandins.

Cytoprotective and dose-dependent inhibitory effects of prostaglandin E1 on rat pancreas treated with ciclosporin A.

The prostaglandin E1 analogue rioprostil protects the pancreas from the noxious effect of ciclosporin A (CsA). To determine whether this cytoprotective action of rioprostil is dependent on or independent of inhibitory effects on pancreatic exocrine and endocrine secretion we studied the effect of different doses of rioprostil on pancreatic exocrine and endocrine secretions in the presence or absence of CsA. Rats received either CsA at a dose of 10 mg/kg body weight by tube feeding once in the morning, rioprostil at linearly increasing doses from 1.8 to 120 micrograms/kg body weight subcutaneously twice daily, a combination of both substances or NaCl. After 8 treatment days, the animals were operated on, and the pancreas isolated and arterially perfused. Insulin secretion was determined after stimulation with glucose, and amylase secretion after stimulation with CCK-8. Insulin and amylase secretion were significantly impaired by CsA. Rioprostil at doses of 1.8, 3.6 and 7.5 micrograms/kg body weight had no significant effect on insulin secretion in the absence of CsA but significantly improved insulin secretion in the presence of CsA. Higher doses of rioprostil significantly inhibited insulin secretion both in the presence or absence of CsA. Amylase secretion was not influenced by rioprostil at doses up to 15 micrograms/kg body weight but improved significantly amylase secretion in the presence of CsA. CsA blood and pancreatic tissue levels were not influenced by rioprostil at doses up to 120 micrograms/kb body weight. We conclude that the cytoprotective effect of the prostaglandin E1 analogue rioprostil against the noxious effect of CsA is dose dependent and is not related to its inhibitory action on endocrine and exocrine pancreatic secretion.

The synthetic prostaglandin E1 analogue rioprostil reduces the nephrotoxic potential of cyclosporine A.

Acute CyA nephrotoxicity involves alteration in the proximal tubule and leads to glomerular lesions. Administration of a vasodilatator agent such as the prostaglandin E1 analogue Rioprostil (Bayer AG, BAY 06893) might prevent preglomerular vasoconstriction and hence reduce cyclosporin nephrotoxicity. As an increased excretion of urinary enzymes as a consequence of CyA-nephrotoxicity is well known we investigated in 40 male Wistar rats the excretion of three urinary enzymes: the brush border enzyme gamma-glutamyltransferase (GGT), the leucine aminopeptidase (LAP), and the lysosomal enzyme N-acetyl-beta-glucosaminidase (NAG). Additionally we determined s-creatinine and CyA plasma level. The kidneys were studied histologically at the end of the study. Wistar rats receiving 20 or 50 mg CyA/kg/d showed a marked deterioration in renal function and an increase of all urinary enzymes determined. In the rats receiving 20 mg CyA/kg/d and Rioprostil (150 micrograms/d) renal function and the enzymes determined remained in the normal range. There was no change in the enzyme excretion and only a minor improvement of renal function in rats receiving 50 mg CyA/kg/d and Rioprostil. Histological findings showed prevention of CyA nephrotoxicity in the 20 mg/kg/d group and diminished renal damage in the 50 mg/kg/d group.

Effect of rioprostil, a methylprostaglandin E1 analog, on basal and stimulated plasma pancreatic hormone levels in man.

The effect of rioprostil, a methylprostaglandin E1 analog on circulating pancreatic hormones was evaluated in 13 healthy male subjects. Rioprostil administration, 300 micrograms twice daily resulted in a significant decrease of fasting insulin, C-peptide, glucagon, and pancreatic polypeptide. No change in fasting plasma glucose or somatostatin levels was observed. An oral glucose tolerance test induced similar increments in plasma glucose concentration before and during treatment, but a delayed rise of insulin and C-peptide levels occurred during the administration of the drug. On rioprostil, the glucose load no longer inhibited peripheral glucagon or somatostatin. Treatment with rioprostil remained without effect on mixed meal-induced changes in plasma glucose levels and concomitant increases in insulin, pancreatic polypeptide, and somatostatin levels. It is concluded that in healthy individuals rioprostil influences the basal and glucose-induced levels of glucagon, insulin, and somatostatin. In healthy men this effect did not, however, result in glucose intolerance.

Rioprostil heals pre-existing aspirin-induced gastric lesions in dogs during daily aspirin administration without altering the anti-inflammatory or analgesic efficacy of aspirin.

Rioprostil (3-100 micrograms/kg/day p.o.), but not cimetidine (30 mg/kg/day p.o.), healed pre-existing aspirin (1950 mg)-induced gastric lesions in dogs despite daily aspirin (975 mg) administration. Gastric antisecretory doses of rioprostil (10 and 100 micrograms/kg/day) decreased lesion scores by 71 to 77 and 90 to 93% after 7 and 11 days, respectively. A nonantisecretory dose of rioprostil (3 micrograms/kg/day) decreased lesion scores by 80% after 14 days. In contrast, lesion scores in dogs receiving either vehicle or a maximal gastric antisecretory dose of cimetidine (30 mg/kg/day) remained unchanged during the 28-day course of treatment. In addition, rioprostil prevented the weight loss which accompanied daily administration of aspirin (975 mg) in vehicle- or cimetidine-treated dogs. When cimetidine-treated dogs were switched to daily rioprostil (100 micrograms/kg/day) and aspirin (975 mg/day) therapy, lesions healed within 6 days. In separate studies, rioprostil had no effect on the sensitivity of the gastric mucosa to subsequent aspirin administration. In addition, discontinuation of maintenance therapy during rioprostil treatment was of no increased benefit (i.e., lesions did not heal faster). Finally, in a model of urate-induced knee-joint synovitis, rioprostil treatment did not inhibit the anti-inflammatory or analgesic efficacy of aspirin, or have any proinflammatory effect of its own.

Slow release delivery of rioprostil by an osmotic pump inhibits the formation of acute aspirin-induced gastric lesions in dogs and accelerates the healing of chronic lesions without incidence of side effects.

Rioprostil, a primary alcohol prostaglandin E1 analog, inhibits gastric acid secretion and prevents gastric lesions induced by a variety of irritants in experimental animals. Because rioprostil is relatively short-acting, it would be of significant benefit clinically if its duration of action could be extended to allow once daily dosing. This investigation demonstrates that when administered via an osmotically driven pump (Osmet, Alza Corp.), rioprostil prevents the acute effects of aspirin on the gastric mucosa of dogs, accelerates the healing of aspirin-induced gastric lesions, and heals preexisting aspirin-induced gastric lesions during chronic administration of aspiring. The potency of rioprostil against acute gastric lesion formation was greatest when delivered from a 24-hr release pump (ED50 = 0.77 micrograms/kg/24 hr) and was 37 times greater than when administered as a single oral bolus. In addition, this activity occurred at doses which had little or no gastric antisecretory activity in betazole-stimulated Heidenhain pouch dogs. When delivered from a 24-hr pump, rioprostil (100 micrograms/kg/24 hr) healed preexisting aspirin-induced gastric lesions within 8 days after removal of aspirin, or after 15 days during continued daily aspirin administration. Additional studies determined that administration of rioprostil at doses of 720, 1440, or 2160 micrograms/kg/24 hr (935-2805 times the gastroprotective ED50 in 24 hr pumps) was well tolerated, with only slight, transient increases in body temperature, softening of the stools, and mild sedation at the highest dose. Administration of rioprostil daily for 5 days at 960 micrograms/kg/24 hr from 24-hr release pumps was also well tolerated by all dogs with no evidence of any accumulation of effect of rioprostil. In summary, administration of rioprostil via an osmotic pump increases its potency and duration of action against the gastric lesion-inducing effect of aspirin, and maintains a wide ratio of safety.

Lipoxygenase metabolites in the rat gastric microvascular response to intragastric ethanol.

The role of lipoxygenase metabolites in ethanol-induced gastric mucosal injury and submucosal microvascular change in the rat was studied by in vivo microscopy. Intragastric ethanol-instillation caused marked submucosal venular constriction, dilatation of small arterioles, and corpus mucosal gross lesion formation. The venoconstriction was greater in the lesion than in the nonlesion area. Intragastric pretreatment with either BW755C, a lipoxygenase inhibitor, or Rioprostil, a synthetic prostaglandin E1 analogue, significantly inhibited both the ethanol-induced venoconstriction and the gross lesion formation. In contrast, pretreatment by the submucosal application of either BW755C (50 micrograms/ml) or Rioprostil (50-500 ng/ml) had no effect on either the submucosal venoconstriction or the mucosal lesion formation. In conclusion, leukotrienes released from the gastric mucosa appear to be involved in the ethanol-induced submucosal venoconstriction associated with gastric mucosal injury. Prostaglandin pretreatment protects against this venoconstriction and gross lesion by a mucosal effect but not by a direct effect on submucosal venules.

The interaction of the prostaglandin E derivative rioprostil with oral anticoagulant agents.

The prostaglandin E1 analogue rioprostil was tested for potential interaction with oral anticoagulant therapy in healthy male volunteers. The effect of rioprostil (0.3 mg, twice a day) was investigated on acenocoumarol (10 mg per subject, n = 7) and phenoprocoumon (0.2 mg/kg, n = 6) single-dose pharmacokinetics and pharmacodynamics. Plasma levels of thrombotest, prothrombin (factor II), and factor VII activities were assayed. Rioprostil, 7 days pretreatment, did not affect control parameters of blood coagulation activity. During the rioprostil period the effect of phenprocoumon on thrombotest and factor VII activities was significantly weaker (p less than 0.02, ANOVA) compared with the control experiment. The effect of acenocoumarol on thrombotest activity was significantly weaker at 24 and 31 hours. None of the pharmacokinetic parameters tested were affected by rioprostil medication. The findings suggest that prostaglandins, at least those of the E series, attenuate the anticoagulant action of the oral anticoagulant agents by a mechanism not related to any pharmacokinetic interaction.

Prostaglandin analogue protects pancreatic B-cells against cyclosporin A toxicity.

Cyclosporin A toxicity on pancreatic B-cells and its prevention by rioprostil, a prostaglandin E1 analogue, were studied in the model of the isolated perfused pancreas of rats treated with both compounds for 8 days. At toxic doses of cyclosporin (10 and 20 mg/kg b.wt), the B-cells showed severe hydropic degeneration of the endoplasmatic reticulum and slight degranulation of the B-cells. Accordingly, the insulin secretion was markedly impaired. Administration of rioprostil ameliorated the insulin secretion significantly, but not the ultrastructural changes. At therapeutic levels of cyclosporin (5 mg/kg b.wt), the hydropic degeneration and the drop in insulin secretion were completely prevented by rioprostil. This observation might have therapeutic implications in the treatment of patients, in particular those undergoing pancreatic transplantation.

Rioprostil and ranitidine in the treatment of prepyloric gastric ulcer. A double-blind comparative trial.

Rioprostil, a synthetic 16-methylprostaglandin E1, combines antisecretory with cytoprotective properties, the latter being active even at doses below those required for acid inhibition. To test whether rioprostil given in antisecretory doses would heal prepyloric ulcers rapidly, we assigned patients with endoscopically proved ulcers randomly to double-blind treatment with 100 micrograms rioprostil twice daily or 150 mg ranitidine twice daily for up to 8 weeks. Recruitment was terminated at the time point of planned interim analysis because the total healing rate was markedly lower than expected. Thirty patients were allocated to each treatment group. The cumulative healing rates at 4 and 8 weeks were 40% and 60%, respectively, in the rioprostil group versus 70% and 90%, respectively, in the ranitidine group (p less than 0.01). Pain relief occurred simultaneously in the two groups. No major adverse effects were noted. These findings question the clinical relevance of using 'cytoprotection' by prostaglandin analogues as treatment for prepyloric ulcer disease in the short term.

Preventive and curative effects of prostaglandins on stress ulcer in rats. Application of endoscopic observation.

The present study investigated the preventive and curative effects of prostaglandins (PGs) on gastric ulcer in rats induced by physical or psychological stresses; some rats were electrically shocked, while others were exposed to affective stimuli arising from the shocked animals. The synthetic PGs dimethyl-PGE2 and rioprostil were administered orally, and their preventive effect on gastric ulceration was evaluated by determining the incidence and the ulcer index of lesions. The curative effect of drugs on ulcer healing was evaluated by determining a time-dependent change in the mucosal surface of the stomach with an endoscopic technique. Oral administration of dimethyl-PGE2 or rioprostil (25 and 50 micrograms/kg) prevented gastric ulceration significantly. Oral administration of these drugs (50 micrograms/kg, twice per day) significantly promoted the healing process of lesions 24 and 36 hr after termination of stress loading. The present results give direct evidence of the curative effect of PGs on stress ulcers and suggest that application of the endoscopic technique to the pathology of the rat's stomach may be a substantial aid in the preclinical evaluation of antiulcer drugs.

Rioprostil, a new prostaglandin E1 analogue does not affect glucose homeostasis in healthy volunteers.

Rioprostil is a synthetic prostaglandin E1 analogue currently under study for peptic ulcer healing. Rioprostil, like prostaglandin E, caused an impairment of glucose tolerance in animal studies. We have studied the effects of single therapeutic doses of rioprostil (300 micrograms and 600 micrograms) on glucose homeostasis after oral glucose tolerance tests in 12 healthy volunteers. Plasma glucose, insulin, and C-peptide concentrations were not significantly different after rioprostil and placebo. These results indicate that a clinically significant interference of rioprostil with glucose tolerance is unlikely.

Prostaglandin E analogue inhibition of pancreatic enzyme secretion.

The effect of native prostaglandins and their methylated analogues on pancreatic enzyme secretion remains unclear, with previous studies reporting inconsistent results. To determine whether the E series prostaglandins directly influence pancreatic secretion, we studied the effect of rioprostil, a prostaglandin E1 analogue, and 16,16-dimethyl prostaglandin E2 (DMPGE2), a prostaglandin E2 analogue, on enzyme release from dispersed guinea pig pancreatic acini. Basal amylase release (4.3 +/- 0.6% of total acinar content) was not altered by either analogue (10(-10)-10(-5) M). A 50% inhibition of maximal cholecystokinin stimulation (10(-9) M; 28.8 +/- 1.2%) was seen with rioprostil (10(-7) M; 14.6 +/- 1.3%) and DMPGE2 (10(-6) M; 15.9 +/- 0.7%) (both p less than 0.005). Prostaglandin inhibition of carbachol-stimulated amylase was less pronounced. The most effective inhibitory dosage with maximal carbachol (10(-5) M; 30.2 +/- 1.9%) was 10(-6) M for both rioprostil (19.2 +/- 1.6%) and DMPGE2 (22.4 +/- 1.7%) (both p less than 0.005). Incubation of acini with A23187, phorbol ester, and 1-oleoyl-2-acetyl-glycerol resulted in a dose-dependent increase in amylase release that was not altered by maximal concentrations of either prostaglandin analogue. Our results indicate that rioprostil and DMPGE2 can directly inhibit pancreatic acinar secretion. This effect appears to occur before activation of the inositol phospholipid system.

Rioprostil prevents aspirin-induced fecal blood loss in dogs. Correlation of Hemoquant (Hb) with endoscopic findings.

Rioprostil, a primary alcohol prostaglandin E1 analog, prevents gastric lesion formation induced by a variety of irritants, including aspirin, in rats and dogs. In the present study, rioprostil reduced both gastric lesion formation and fecal blood loss in dogs caused by daily aspirin administration (1,950 mg/day) for 3 consecutive days. A gastric antisecretory dose of 100 micrograms/kg p.o. t.i.d. of rioprostil reduced fecal blood loss by 96% to a level (0.76 +/- 0.10 mg Hb/g stool) similar to basal blood loss in non-aspirin-treated vehicle dogs (0.61 +/- 0.08 mg Hb/g stool) as determined by HemoQuant assay. A nonantisecretory dose of rioprostil (1 microgram/kg p.o. t.i.d.) reduced fecal blood loss by 70% compared to vehicle-treated dogs. Gastric lesion severity scores in dogs treated with 100 or 1 micrograms/kg p.o. t.i.d. of rioprostil were 61 and 52% lower, respectively, than the mean severity score in the vehicle-treated group. There was a highly significant (p less than 0.001) correlation between gastric lesion score and fecal blood loss independent of the treatment each dog received. The efficacy of a nonantisecretory dose suggests that the gastric lesion effect of rioprostil may be independent of gastric antisecretory effects. The correlation of fecal blood loss with gastric lesion score suggests the possibility that either measurement may be used as an indication of gastric lesion occurrence or severity.