Role of xanthine oxidase-derived oxidants and leukocytes in ethanol-induced jejunal mucosal injury.

Previous reports indicate that intestinal intraluminal ethanol increases mucosal permeability (an index of mucosal injury) and histamine release by mast cells, and that the released histamine plays a role in mediating the increased permeability. In the present study, we investigated whether reactive oxygen metabolites and their major sources (xanthine oxidase and leukocytes) were involved in these ethanol effects. In rabbits, segments of the jejunum were perfused with a control solution or with 6% ethanol. In these segments, mucosal permeability was assessed by determining jejunal clearance of i.v. administered 51Cr-ethylenediaminetetraacetate (51Cr-EDTA) and 125I-bovine serum albumin (125I-BSA), and mast cell histamine release was estimated from the histamine concentration of the gut effluent. Ethanol increased 51Cr-EDTA clearance, 125I-BSA clearance, and histamine release. These ethanol effects decreased when the animals were given superoxide dismutase plus catalase (scavenger of O2- and H2O2, respectively), allopurinol, or oxypurinol (xanthine oxidase inhibitors). Administration of a monoclonal antibody (R15.7) against leukocyte adhesion molecule, CD18, inhibited completely the ethanol-induced increased 51Cr-EDTA and 125I-BSA clearances and histamine release. These and supplementary data suggest that (a) ethanol-induced mucosal injury and mast cell histamine release are mediated primarily by leukocytes, and (b) oxy radicals, especially those generated by xanthine oxidase, mediate these ethanol effects mainly by promoting leukocyte infiltration.

Adaptive cytoprotection against ethanol-induced small intestinal mucosal injury.

Exposure of the small intestinal mucose to 6% ethanol (which is found in human jejunum during alcohol consumption) causes morphological alterations, and increased permeability of the mucosa and histamine release from intestinal mast cells. The released histamine is shown to mediate a significant component of the increased mucosal permeability (i.e., mucosal injury). In the present study, we have investigated whether adaptive cytoprotection occurs against the increased mucosal permeability and histamine release induced by 6% ethanol. Rabbits were used. In each animal, three adjacent segments of upper small intestine were pre-perfused for 30 min, and then perfused for 90 min in the following order control solution followed by control solution (control segment); control solution followed by 6% ethanol (ethanol segment); 1% ethanol followed by 6% ethanol (pretreated ethanol segment). During the 90-min perfusion, mucosal permeability of each segment was measured by analyzing the effluent for intraluminal clearance of i.v. administered 51Cr-labelled ethylenediaminetetraacetic acid (51Cr-EDTA) and 125I-labelled bovine serum albumin (125I-BSA). Mast cell histamine release was assessed by determining histamine concentration of the gut effluent. All measurements were higher in the ethanol segments than in the controls. These ethanol effects were significantly lower in the pretreated ethanol segments, indicating that adaptive cytoprotection occurs against the mucosal injury induced by 6% ethanol. These findings are discussed in relation to the literature on mucosal effects of intestinal intraluminal ethanol.

Mast cells mediate acid-induced augmentation of opossum esophageal blood flow via histamine and nitric oxide.

BACKGROUND & AIMS: Increased esophageal blood flow during reflux episodes may play an important role in mucosal resistance to injury, although the mechanism remains unclear. Decreased stainable mast cells and increased luminal histamine release during acid exposure has been previously documented. Therefore, the role of mast cells, nerves, histamine, and nitric oxide in mediating increased blood flow during acid challenge of the distal esophagus was investigated. METHODS: The effects of the mast cell stabilizers disodium cromoglycate and doxantrazole, the neurotoxin tetrodotoxin, the histamine H1 receptor antagonist promethazine, and the NO synthase inhibitor N omega-nitro-L-arginine methyl ester were examined by monitoring opossum esophageal histamine release and blood flow during perfusion with 100 mmol/L HCl. RESULTS: Luminal acid challenge significantly increased both histamine release and blood flow (P < 0.05). Disodium cromoglycate, promethazine, and N omega-nitro-L-arginine methyl ester attenuated the increase in blood flow to basal (saline-perfused) levels. Tetrodotoxin did not prevent an acute increase in blood flow that rapidly returned to baseline, likely from the ensuing hypotension. CONCLUSIONS: These findings provide evidence that mast cell-derived histamine, acting through an NO-dependent mechanism, plays a central role in the response of the esophageal microcirculation to luminal acid.

Morphological evidence of mast cell degranulation in an animal model of acid-induced esophageal mucosal injury.

In previous studies we have demonstrated that microvascular permeability increases early in the course of experimental acid-induced esophageal mucosal injury. This is associated with an increase in the intraluminal appearance of histamine, suggesting a possible role for mast cells in this form of injury. In the present study, quantitative analysis of esophageal mast cells was undertaken using both light and electron microscopy in opossums undergoing intraluminal esophageal acid perfusion or normal saline control perfusion. Light microscopy showed that animals perfused with either 50 or 100 mM hydrochloric acid had an approximate 50% decrease in the number of stainable esophageal mast cells. Stereologic analysis of electron micrographs revealed that within the mucosa, the mean area of the mast cells as well as nuclear area and area of intact granules were also significantly reduced in acid perfused animals. Taken together these quantitative morphological analyses suggest that intraluminal acid exposure is associated with degranulation and/or lysis of esophageal mast cells and that released mediators from esophageal mast cells may play a role in the pathophysiology of reflux esophagitis.

Studies on ethanol-induced subepithelial fluid accumulation and jejunal villus bleb formation. An in vitro video microscopic approach.

Jejunal intraluminal ethanol causes morphological and mucosal microvascular injury. The purpose of the present study was to understand the mechanism of the morphological alterations caused by ethanol without the influence of ethanol's effect on the microcirculation. Therefore, we have investigated the ethanol-induced morphological changes in the absence of blood flow (i.e., in the jejunum in vitro) and compared these changes with those reported to occur in the presence of microcirculation (i.e., in the jejunum in vivo). The mucosa of jejunal segments was exposed to a control solution and to solutions containing 0.8, 1.6, and 4.8% (w/v) ethanol, using a specially designed apparatus. The morphological response of the mucosa to these solutions was continuously examined employing a video microscopic technique, and the changes were morphometrically evaluated on subsequent playback of videotapes. Ethanol caused a concentration-dependent increase in the number of villi with subepithelial fluid accumulation, i.e., blebs, and a decrease in the height of the villus core (i.e., lamina propria). With 0.8 and 1.6% ethanol, the contracted core remained partially attached to the epithelium and the total villus height (villus core plus epithelial layer) decreased. With 4.8% ethanol, the villus core contraction was so rapid that the stroma fully separated from the epithelium. Thus, among other factors, the rapidity of the villus core contractions appears to play a role in the subepithelial bleb formation and in the appearance of the bleb. The ethanol-induced changes in vitro are similar to those reported to occur in the jejunum in vivo. Therefore, we conclude that the effect of ethanol on morphology is independent of its action on the microcirculation.

The use of oximetry in determining intestinal blood flow.

The intraoperative evaluation of intestinal ischemia and viability is often subjective and unreliable. The results of recent reports of pulse and surface oximetry have suggested that these techniques may be useful in assessing intestinal blood flow. In the current study, we evaluated and compared the ability of intestinal tissue oxygen saturation (as measured by pulse oximetry) and intestinal surface oxygen tension (as measured by surface oximetry) to determine the actual intestinal tissue blood flow (as measured with a radiolabeled microsphere technique). In five dogs, tissue oxygen saturation, surface oxygen tension and blood flow of the proximal and distal parts of the small intestine were measured under basal conditions. A clamp placed around the root of the superior mesenteric artery was then tightened to decrease the blood flow through this artery (as measured by an ultrasonic flow probe) by 50 percent and then by 75 percent, repeating all measurements after each reduction. The two consecutive reductions in superior mesenteric artery blood flow resulted in an average 54 and 76 percent reduction in tissue blood flow, respectively. As a result of these reductions in tissue blood flow, the average intestinal tissue oxygen saturation (percentage), as determined by pulse oximetry, decreased significantly from a basal value of 93 +/- 1 to 83 +/- 1 (p < 0.05) and then to 76 +/- 1 (p < 0.05) with the two progressive blood flow reductions. Intestinal surface oxygen tension decreased more steeply, from a basal value of 97 +/- 1 to 80 +/- 6 (p < 0.05) and then to 64 +/- 7 millimeters of mercury (p < 0.05) with the same two reductions in tissue blood flow. Both techniques were capable of estimating tissue blood flow, but pulse oximetry was quicker and simpler to use. We conclude that the pulse oximeter has the potential to be of value in the intraoperative assessment of intestinal blood flow.

Ethanol-induced jejunal microvascular and morphological injury in relation to histamine release in rabbits.

BACKGROUND: To investigate the relation between ethanol-induced jejunal microvascular injury, morphological changes, and histamine release, the present study examined whether the attenuation of microvascular effect of ethanol by 16,16-dimethyl prostaglandin E2 (dmPGE2) (reported by us previously) was associated with an attenuation of epithelial damage and histamine release. METHODS: Rabbits were used. Mucosal microvascular injury was assessed by determining jejunal plasma protein loss (JPPL), histamine release by measuring histamine concentration of the gut effluent, and epithelial damage by routine histology. RESULTS: (1) During 90-minute jejunal ethanol perfusion, there was a direct relation between the time course of histamine release and that of JPPL. (2) dmPGE2 attenuated the ethanol-induced JPPL and histamine release, and the decrease in JPPL was directly proportional to the decrease in histamine release. (3) dmPGE2 did not alleviate ethanol-induced epithelial damage. (4) Ketotifen (a mast cell stabilizer), similar to dmPGE2, attenuated ethanol-induced JPPL and histamine release. (5) Ethanol caused histamine release by the jejunum in vitro; this was attenuated by dmPGE2 and also by phloretin (a mast cell stabilizer). CONCLUSIONS: It appears that (1) ethanol causes JPPL by inducing release of mediators from mucosal mast cells. (2) dmPGE2 attenuates JPPL by stabilizing mast cells. (3) The ethanol-induced mucosal microvascular injury is directly related to histamine release but not to epithelial damage.

The role of angiotensin in the intestinal vascular response to hypotension in a canine model.

It was previously shown that the vasoconstrictory response to hypotension was similar in the mucosa of the small bowel and the colon but was significantly higher in the muscularis of the latter than that of the former. To understand the mechanism of this differential response of the muscularis of the small bowel and the colon, the present study investigated the effect of an angiotensin II inhibitor (saralasin) on the hypotension-induced vasoconstriction of the mucosa and the muscularis of these two locations of the gastrointestinal tract. Dogs were used. Hypotension was induced by hemorrhage to reduce blood pressure by 40 mm Hg. Blood flow was measured by 15-microns radiolabeled microspheres. Saralasin was infused intravenously for 20 minutes at a rate of 0.05 mg.kg-1 bolus followed by 1 microgram.kg-1.min-1. Saralasin had no effect on the basal blood flow of the mucosa or the muscularis of the small bowel or on the hypotension-induced vasoconstriction of these two layers of the small bowel. In contrast, saralasin decreased blood flow to the mucosa (-28%; P less than 0.001) and increased blood flow to the muscularis (+140%; P less than 0.001) of the colon under basal conditions and also reduced the hypotension-induced vasoconstriction of the colonic muscularis (P less than 0.01). These and supplementary data indicate that there is a difference between the small bowel and the colon in local activity of vascular angiotensin system and that this system is most active in the colonic muscularis where it plays a significant role in the vasoconstrictory response to hypotension.

The role of histamine1 and histamine2 receptors in the ethanol-induced jejunal plasma protein loss.

Histamine and other mediators have been shown to be involved in the ethanol-induced jejunal plasma protein loss. In this study we have investigated whether the histamine (H)-related component of this protein loss is mediated by H1-receptors, H2-receptors or both. Four groups of dogs (n = 12 in each) were studied. They were: untreated, H1 + H2-receptor blockade, H1-receptor blockade and H2-receptor blockade. Chlorpheniramine and ranitidine were used to block H1 and H2-receptor blockade. Chlorpheniramine and ranitidine were used to block H1 and H2-receptors respectively. In all animals, jejunal protein loss was measured over 10 min periods for 90 min. Ethanol increased protein loss in all time periods (p less than 0.001). This protein loss was depressed by H1 + H2-receptors blockade throughout 90 min (p less than 0.01). H1-receptor blockade caused a similar depression of ethanol effect but only during 20 to 40 min (p less than 0.05). In contrast, H2-receptor blockade aggravated the protein losing effect of ethanol throughout 90 min (p less than 0.01). Analyses of data tend to suggest that the ethanol-induced protein loss is mediated principally by H1-receptors, and that a complete inhibition of the histamine-related ethanol-induced protein loss can be achieved only by a simultaneous blockade of both H1 and H2-receptors, and not by H1- or H2-receptor blockade alone.

The effect of pitressin on esophageal blood flow of the dog.

In a previous study on canine esophagus, we reported that intravenous infusion of isoproterenol caused mucosal (i.e., mucosal + submucosal) vasodilation only in the lower esophageal sphincter (but not in the body) and muscularis vasodilation only in the body (not in the lower esophageal sphincter). In the present study, we have investigated in dogs whether these esophageal tissues also exhibit a similar difference in their vasoconstrictory response to intravenous infusion of pitressin. All measurements were made before (basal) and after infusion of 0.02 U pitressin.min-1.kg-1 for 15 min. Pitressin significantly decreased portal venous pressure and blood flow, and increased vascular resistance of all tissues of the esophagus. This vasoconstriction of the tissues, however, was higher in the squamous mucosa of the body than in the columnar mucosa of the lower esophageal sphincter. In contrast, it was higher in the smooth muscle of the lower esophageal sphincter than in the striated muscle of the body. These data together with those of our previous report on isoproterenol demonstrate that pitressin causes a pronounced vasoconstriction in those esophageal tissues where isoproterenol had no effect. Conversely, pitressin causes least vasoconstriction in those tissues where isoproterenol produced a significant vasodilation. These differences could be the result of partial agonist actions or differences in receptor density or in receptor-effector coupling mechanism.

Microvascular permeability increases early in the course of acid-induced esophageal injury.

To test the hypothesis that microvascular injury is involved in the pathophysiology of acid-induced esophagitis, the effect of acid perfusion on intraluminal plasma protein loss was studied in relation to histological changes. Four groups of opossums (n = 6 in each) were perfused with either normal saline control) or 10, 20, or 100 mmol/L isoosmolar hydrochloric acid at 2 mL/min for 90 minutes using a midesophageal catheter. The distal esophagus was cannulated via a gastrostomy, and the effluent was collected and measured for intraluminal loss of IV injected 125I-bovine serum albumin. Plasma protein loss in the control group was constant with a total loss of 3.40 +/- 0.69 mg/g dry wt. Perfusion of 10, 20, and 100 mmol/L hydrochloric acid increased total protein loss to 8.06 +/- 2.62, 13.94 +/- 2.72, and 27.34 +/- 4.34 mg/g dry wt, respectively. The protein loss was not associated with intraluminal blood loss, as measured by previously injected 51Cr-labeled autologous red blood cells. Histological changes, scored by a blinded observer, were significant only between control animals and those perfused with 100 mmol/L hydrochloric acid. Separate studies using the vascular tracer monastral blue B demonstrated an increase in labeling of lamina propria blood vessels that varied directly with the concentration of acid perfusate, thereby providing direct morphological evidence of microvascular injury. These studies suggest that increased microvascular permeability occurs early in the course of acid-induced esophageal injury.

The microvascular anatomy of the canine stomach. A comparison between the body and the antrum.

To investigate whether there is a difference in the microvascular architecture between the body and the antrum of the canine stomach, these two locations were compared with respect to microsphere entrapment and the microvascular architecture and diameter in relation to histology by corrosion casting and by intraarterial injection of india ink. There was 63% shunting of 9-micron microspheres in the antrum, but none in the body. Corrosion cast and Indian ink studies showed that in the body there was a single microvascular network of capillaries that appeared to originate from the arterioles in the submucosa and were in close apposition to the epithelial cells of the gastric glands. The diameter of these capillaries was 8.6 +/- 0.2 microns. In contrast, there were two distinct capillary networks in the antrum: a basal and a superficial. The capillaries of the basal network of the antrum originated from the arterioles at the level of the muscularis mucosa and drained into the capillaries of the superficial mucosa. The capillaries of the superficial network had a significantly larger diameter (10.8 +/- 0.4 microns) than those of the basal network (7.3 +/- 0.2 microns). In many instances the capillaries of the superficial network originated directly from the ascending arterioles passing through the basal network. These direct arteriocapillary connections may have permitted the shunting of 9-microns spheres in the antrum.

Comparison of vascular response of different locations of the gastrointestinal tract to isoproterenol and nifedipine.

In the present study we have compared the effect of intravenous infusion of a calcium channel blocker, nifedipine (1.0 micrograms.kg-1.min-1 for 20 min), with that of isoproterenol (0.1 micrograms.kg-1.min-1 for 20 min) on the hemodynamic parameters and the vascular response of different locations and tissue layers of the gastrointestinal tract. Heart rate increased with isoproterenol but not with nifedipine. Both agents caused a similar increase in cardiac output and a similar fall in mean arterial pressure. After 20 min infusion, nifedipine increased the blood flow of the axillary artery, but isoproterenol had no such effect. Isoproterenol caused vasodilation of the mucosa in the antrum but not in the fundus and the body of the stomach or in the duodenum, jejunum, mid small intestine, ileum, and colon. The mucosal effect of nifedipine was similar, except that it also caused vasodilation in the small bowel and in the ascending colon. Nifedipine caused vasodilation of the muscularis throughout the gastrointestinal tract, but isoproterenol had no such effect. These differences are discussed in relation to the mechanism of action of these two vasodilators. It is suggested that the vascular response of different locations and tissue layers of the gastrointestinal tract to vasodilators is locally regulated by a variety of mechanisms may include beta- and alpha-receptor density and (or) sensitivity, angiotensin II activity, and metabolic need of the tissues.

Gastrointestinal blood flow in the opossum with special reference to the esophagus.

The opossum esophagus, like that of the human, is composed of striated muscle fibres proximally and smooth muscle fibres distally. Because of this similarity the opossum has been used extensively as an animal model for esophageal studies, but to date no data on esophageal blood flow have been reported in this species. The purpose of this study was to establish the basal blood flow characteristics of different regions of the opossum gastrointestinal tract with particular reference to the esophagus. Intracardiac injection of 15-microns microspheres was used to provide an estimate of blood flow (mL.min-1.g-1 dry tissue) to the whole wall, the combined layer of mucosa plus submucosa, and the muscularis propria. Basal blood flow in the whole tissue and mucosa-submucosa was significantly higher in the lower esophageal sphincter than in the proximal or distal esophagus. The muscularis propria blood flow displayed an aborally increasing gradient with flow to proximal esophagus (striated muscle) less than distal esophagus (smooth muscle) less than lower esophageal sphincter. Regional differences in blood flow to other regions of the gastrointestinal tract were similar to that described in other species. In addition, no changes in basal blood flow occurred despite repeated microsphere injections, suggesting that this species provides a good animal model for the study of gastrointestinal blood flow.

Regional differences in the vascular response of the canine esophagus to vasodilators.

The basal blood flow of the proximal and distal body of the esophagus, and that of the lower esophageal sphincter, and the vascular response of these sites to intravenous infusion of nifedipine (1.0 microgram/kg.min) and isoproterenol (0.1 microgram/kg.min) were studied. The basal blood flow of the mucosa plus submucosa, muscularis propria, and the whole wall was significantly higher in the lower esophageal sphincter than in the proximal or the distal body. Intravenous nifedipine caused a similar decrease in vascular tone of the mucosa and the muscularis propria in all sites of the esophagus. Isoproterenol infusion significantly decreased the vascular tone of the mucosa with columnar epithelium (i.e., mucosa overlying the lower esophageal sphincter), but not of the mucosa with squamous epithelium (i.e., mucosa overlying the body). In the muscularis propria the vasodilatory effect of isoproterenol and nifedipine was significant at all sites, but this effect was greater in the striated muscle of the body than in the smooth muscle of the lower esophageal sphincter (p less than 0.05). Thus, not only do different vasodilators have different effects on the vasculature of the esophagus, but also a single agent may have differing effects on the various sites and layers of this organ.

Time-related changes in microsphere entrapment in canine jejunum.

To assess the validity of repeated blood flow measurements using the microsphere technique, the apparent blood flows in the anatomic layers of the jejunum were determined from the entrapment of 9-, 11.5-, and 17-micron microspheres at 1.5, 15, 30, and 60 min after their injection. The entrapment of 17-micron spheres in the mucosa plus submucosa and in the muscularis propria remained similar at all times, but these spheres migrated (P less than 0.01) from the submucosa to the mucosa. By 1.5 min, 5 +/- 2% of 11.5-micron spheres had shunted, but no subsequent shunting was observed. No migration of 11.5-micron spheres from the mucosa, submucosa or the muscularis was observed. The shunting of 9-micron spheres from the whole wall increased from 19 +/- 4% at 1.5 min to 40 +/- 4% at 60 min (P less than 0.001). These data suggest that 17-micron spheres can only fractionate the blood flow of the whole wall into that of the mucosa plus submucosa and that of the muscularis propria, while 11.5-micron spheres may measure fractional flow to the submucosa separately. The continued washout of 9-micron microspheres precludes their use for repeated blood flow measurements.

Mechanism of ethanol-induced jejunal microvascular and morphologic changes in the dog.

To study the mechanism of morphologic and microvascular effects of intraluminal ethanol, we perfused jejunal segments of the dog with 6% (wt/vol) ethanol for 0 (control), 10, 20, 30, 60, and 90 min, and measured the time-dependent changes in (a) the prevalence of villi with epithelial damage (i.e., villi with intact blebs plus those with broken blebs) and those without epithelial damage (undamaged villi), (b) the height of the villus core and the patency of lacteals, (c) jejunal albumin loss, and (d) permeability of microvessels of the villus tip by colloidal carbon vascular labeling. We found that (a) the prevalence of villi with epithelial damage or with intact bleb increased progressively during the first 20 min of ethanol perfusion and then declined gradually; (b) the height of the villus core and the patency of lacteals in the undamaged villi and in those with intact bleb decreased during the first 20 min and then gradually increased; and (c) jejunal albumin loss and the prevalence of villi with carbon labeling increased for the first 30 min, after which the former declined gradually whereas the latter remained at a plateau. These findings suggest that contraction of the villus core and compression of the lymphatics are the primary cause of ethanol-induced epithelial damage, which is accentuated by increased microvascular permeability and consequent protein leakage. The mechanism of recovery of most parameters, in spite of continuous ethanol perfusion, remains to be investigated.

Vasculature of various locations of canine gastrointestinal tract responds differently to intravenous isoproterenol.

In this study we investigated the relative vascular response of different locations of the gastrointestinal tract to continuous intravenous infusion of isoproterenol (0.1 microgram kg-1 min-1 for 10 min). The vascular response of some nonsplanchnic organs was also examined. Blood flow of the arteries was measured by electromagnetic flowmetry and that of the tissues by 15-micron microspheres. Isoproterenol increased (P less than 0.05) blood flow of the axillary artery (+52%), and the superior mesenteric artery (+45%), but not that of the inferior mesenteric artery. In the nongastrointestinal tissues, isoproterenol increased (P less than 0.05) the blood flow of the left (+46%), and right ventricle (+85%), and the skeletal muscle (+100%). In the gastrointestinal tract, isoproterenol increased (P less than 0.05) blood flow in the esophagogastric junction (+505%) and antrum (+1511%) only, but not in the gastric body or in any location of the small or large intestine. The drug also caused a large fall in resistance in the esophagogastric junction (-74%) and antrum (-94%), and a small, but significant fall in the duodenum, jejunum, and in the mid-small intestine. It had no significant effect on vascular resistance in the gastric body, ileum, or colon. In those locations of the gastrointestinal tract where isoproterenol caused an increase in blood flow, this effect was confined to the combined mucosal plus submucosal layer, and the drug had no effect on the muscularis. These data suggest that different locations of the gastrointestinal tract respond differently to the same circulating concentration of isoproterenol. The mechanism of this difference in response merits further investigation.(ABSTRACT TRUNCATED AT 250 WORDS)

Histamine is involved in ethanol-induced jejunal microvascular injury in rabbits.

To examine for the possible involvement of histamine in the jejunal microvascular effects of ethanol, we investigated the effects of (a) intraluminal ethanol on histamine release by the jejunum and (b) simultaneous inhibition of both histamine1 and histamine2 receptors (using promethazine and cimetidine, respectively) on ethanol-induced intestinal plasma protein loss in rabbits. Ethanol increased histamine release by the jejunum both in vivo (p less than 0.01) and in vitro (p less than 0.05). To investigate the effect of antihistamines on ethanol-induced plasma protein loss, we determined the dose of blockers that would completely inhibit the histamine1 and histamine2 receptors. In the absence of antihistamines, ethanol caused a 10-fold increase in jejunal protein loss over the controls (p less than 0.001). Simultaneous inhibition of histamine1 and histamine2 receptors attenuated (p less than 0.025), but did not abolish, the ethanol-induced protein loss. These data are discussed in relation to the literature, and it is concluded that histamine may play a role in the jejunal microvascular effects of ethanol. As the ethanol-induced protein loss was not completely inhibited, other mediators or mechanisms were probably involved.

Evidence for the involvement of 5-lipoxygenase products in ethanol-induced intestinal plasma protein loss.

In this study we investigated whether the products of 5-lipoxygenase (5-LO) were involved in the jejunal microvascular injury induced by intraluminal ethanol (ETH). A group of rabbits was given orally a selective inhibitor of 5-LO (L-651,392, Merck Frosst Canada) in two 10-mg doses, 24 and 2 h before the experiments (treated group). Another group received no such treatment (untreated group). In each animal of both groups, a jejunal segment was perfused with a control solution (control segment) and an adjacent segment with an ETH-containing (6% wt/vol) solution (ETH-perfused segment). In a series of experiments, we measured 5-LO activity of the jejunal segments of both groups using the generation of leukotriene B4 (LTB4) as an index. In a second series of experiments, we determined the ETH-induced intraluminal protein loss, which was taken as a measure of mucosal microvascular damage. In the untreated group, LTB4 generation (pg/mg tissue) by the ETH-treated segment (2.49 +/- 0.70) was higher (P less than 0.005) than that by the control segment (0.68 +/- 0.14). In the treated group, the LTB4 generation decreased (P less than 0.05) both in the control (0.15 +/- 0.03) and the ETH-perfused segments (0.44 +/- 0.09), but the difference between the two segments still remained significant. ETH caused a 13-fold increase (P less than 0.01) in jejunal protein loss in the untreated group but only a 5-fold increase (P less than 0.05) in the treated group. The ETH-induced increase in protein loss was significantly lower in the treated than in the untreated group (P less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)