Effect of dietary alpha-linolenic acid on endotoxin-induced production of tumor necrosis factor by peritoneal macrophages in horses.

A study was conducted to determine whether dietary supplements with alpha-linolenic acid altered the ability of equine peritoneal macrophages to produce tumor necrosis factor (TNF) in response to endotoxin. Peritoneal macrophages were harvested from 6 healthy adult horses before and after the horses were fed a nutritionally balanced ration that contained 8% linseed oil as a source of alpha-linolenic acid. The macrophages were cultured in media containing no additives (control), endotoxin (0.5 to 50 ng/ml), or the calcium ionophore, A23187. Macrophage supernatants were collected after 6 and 24 hours' incubation and stored at -70 C. Tumor necrosis factor activity was estimated by a modified in vitro cytotoxicity bioassay, using the murine fibrosarcoma cell line, WEHI 164 clone 13. The TNF activity after 6 and 24 hours' incubation was greater in culture media of macrophages exposed to endotoxin than in media from control macrophages. For macrophages cultured in media that contained endotoxin, neither the concentration of endotoxin nor incubation time had any effect on TNF activity. Endotoxin-induced macrophage production of TNF, as determined by measurement of TNF activity, was significantly less after horses were fed the alpha-linolenic acid-rich ration for 8 weeks.     

Reduced endotoxin-induced production of tumor necrosis factor activity by equine peritoneal macrophages exposed to the dual inhibitor of arachidonic acid metabolism, SK & F 86002.

The purpose of this study was to determine if a structurally novel dual inhibitor of arachidonic acid metabolism, SK & F 86002, would inhibit the endotoxin-induced production of tumor necrosis factor (TNF) activity by equine peritoneal macrophages. Equine peritoneal macrophages were variously pretreated for 0, 0.5 and 2 h with SK & F 86002 at 10(-9) to 10(-4) molar final concentrations or were left untreated. Then, the macrophages were cultured in vitro in the presence of endotoxin (5 ng/mL). Supernatant media were collected after 4 h and stored at -70 degrees C until assayed for TNF activity and immunoreactive thromboxane B2 (iTxB2). Macrophage supernatant TNF activities were estimated by an in vitro cytotoxicity bioassay using the murine fibrosarcoma cell line, WEHI 164 clone 13. Concentrations of iTxB2 were quantitated by radioimmunoassay. Coincubation of macrophages with SK & F 86002 significantly decreased the subsequent supernatant TNF activity. Concentrations of SK & F 86002 from 10(-7) to 10(-4) molar effectively reduced TNF production when added to macrophages 0 and 0.5 h prior to endotoxin. After 2 h of preincubation, SK & F 86002 significantly reduced supernatant TNF activity at 10(-5) and 10(-4) M concentrations. Supernatant concentrations of iTxB2 were reduced when SK & F 86002 was added at 10(-6) to 10(-4) M concentrations, 0 and 0.5 h prior to endotoxin, and at all concentrations (10(-9) to 10(-4)) when preincubated with macrophages for 2 h.(ABSTRACT TRUNCATED AT 250 WORDS)     

Cytotoxic tumor necrosis factor activity produced by equine alveolar macrophages: preliminary characterization

Blood monocytes and alveolar macrophages (AM) were harvested from foals (aged 46 days to 6 months) and cultured in either medium alone or medium containing 10 micrograms/ml bacterial lipopolysaccharide (LPS). After 24 h, culture supernates were collected and analyzed for cytotoxic activity on sensitized L929 cells. Both monocytes and AM that had been treated with LPS produced significantly more cytotoxic activity than the same cell type exposed to medium lacking LPS. LPS-treated macrophages secreted significantly more cytotoxic activity (120 +/- 17.8 U/ml) than did LPS-treated monocytes (47.3 +/- 17.0 U/ml); however, constitutive production of cytotoxin by monocytes was higher (16.7 +/- 4.1 versus 1.2 +/- 1.2 U/ml). The identification of the cytotoxin as tumor necrosis factor (TNF) was strongly suggested by its reactivity with a rabbit antiserum directed against the N-terminal 15 amino acids of human TNF. TNF secretion by AM increased in a dose-dependent manner between LPS concentrations of 0.0001 and 1 microgram/ml, then leveled off. Most of the cytotoxic TNF activity produced by AM was secreted within the first 8 h after initial contact with LPS. Macrophage supernatant TNF was stable over a pH range of 6-11, but lost activity when kept at a pH less than 6. Equine TNF also was destroyed by exposure for 1 h to temperatures more than 60 degrees C. TNF bioactivity was recovered as a single peak after crude macrophage supernate was subjected to analysis by either anion exchange or gel filtration chromatography (molecular weight approximately 56,000).     

Endotoxin-induced production of interleukin 6 by equine peritoneal macrophages in vitro.

A study was performed to determine whether equine peritoneal macrophages produce interleukin 6 (IL-6) in vitro in response to endotoxin. Peritoneal fluid was collected from 14 clinically normal adult horses and was used as the source of peritoneal macrophages. Macrophages from each horse were isolated and cultured separately in vitro in the absence or presence of various concentrations (0.5, 5, 500 ng/ml) of endotoxin (lipopolysaccharide from Escherichia coli 055:B5). Culture medium supernatants were collected after 3, 6, 12, and 24 hours' incubation and were frozen at -70 C until assayed for IL-6 activity. Supernatant IL-6 activity was determined by use of a modified colorimetric assay and the murine hybridoma cell line B 13.29 clone B.9, which is dependent on IL-6 for survival. Results indicated that equine peritoneal macrophages produce IL-6 in vitro and that supernatant medium IL-6 activity was significantly (P less than 0.05) increased by exposure to endotoxin. Significant (P less than 0.05) time and treatment effects on macrophage IL-6 production were apparent. The IL-6 activity peaked at 6 or 12 hours' incubation, then remained high through 24 hours' incubation, regardless of endotoxin exposure. Medium IL-6 activity during 3 and 6 hours' incubation was significantly (P less than 0.05) greater in macrophages exposed to 5 or 500 ng of endotoxin/ml than in those exposed to 0.5 ng of endotoxin/ml; however peak IL-6 activity was similar among all endotoxin concentrations. Endotoxin concentration did not have an effect on medium IL-6 activity from macrophages exposed to endotoxin for 12 or 24 hours.     

Proadifen-induced production of prostacyclin by equine peritoneal macrophages

A study was performed to determine the effect of proadifen hydrochloride on prostacyclin (prostaglandin I2 [PGI2]) and thromboxane A2 (TxA2) synthesis by equine peritoneal macrophages and the effect of proadifen on endotoxin-induced synthesis of PGI2 and TxA2 by equine macrophages. Peritoneal macrophages (2.5 x 10(6)/ml) were incubated for 6 hours in tissue culture media containing 1) nothing (nontreated control), 2) proadifen hydrochloride (20, 100, 250, and 500 mumol/L, 3) endotoxin (5 ng/ml), or 4) the calcium ionophore A23187 (0.95 mumol/L). In a second series of experiments, peritoneal macrophages were incubated with endotoxin (5 ng/ml) and proadifen (250 umol/L), for 6 hours. Concentrations of 6-keto-prostaglandin F1 alpha (6-keto-PGF1 alpha) and thromboxane B2, the stable metabolites of PGI2 and TxA2, were determined in the incubation media by radioimmunoassay. Proadifen caused increased synthesis of PGI2 by equine macrophages, without affecting TxA2 production. The increased PGI2 production was similar to that induced by endotoxin and calcium ionophore; however, the latter 2 agents significantly stimulated TxA2 production as well (P less than 0.05). There were no significant differences among mean concentrations of 6-keto-PGF1 alpha in media from macrophages treated with 100, 250, or 500 mumol/L proadifen, but there was a significant curvilinear regression between their concentrations. The ratio of thromboxane B2 to 6-keto-PGF1 alpha was significantly lower than baseline in incubation media from macrophages exposed to proadifen, endotoxin, and calcium ionophore. Proadifen hydrochloride did not significantly change equine peritoneal macrophage production of PGI2 or TxA2 in response to endotoxin.     

Secretion of tumor necrosis factor by endotoxin-treated equine mammary exudate macrophages: effect of dexamethasone and pentoxifylline.

Secretion of tumor necrosis factor (TNF) by equine mammary exudate macrophages (MEM phi) exposed to bacterial lipopolysaccharide (LPS) was dose-dependent and was maximal (216.5 +/- 51.9 U/ml) at 100 micrograms LPS/ml, the highest concentration tested. All concentrations of dexamethasone tested (10(-8) to 10(-4) M) significantly (P less than or equal to 0.05) inhibited TNF production by MEM phi when the agent was added 1 hour before LPS. Pretreatment with pentoxifylline at concentrations greater than 3 micrograms/ml also significantly (P less than or equal to 0.05) reduced secretion of TNF by MEM phi. The inhibitory effect of dexamethasone (10(-4) M) was observed when the agent was added to MEM phi from 30 minutes before until 4 hours after LPS. Pentoxifylline (100 micrograms/ml) significantly (P less than or equal to 0.05) suppressed TNF when added from 2 hours before until 2 hours after LPS; however, when pentoxifylline addition was delayed until 8 hours post-LPS, TNF production was enhanced. These apparent inhibitory effects of dexamethasone and pentoxifylline were not due to reduced macrophage viability or to interfering effects of the agents at the level of the TNF bioassay.     

In vitro microbicidal activity of avian peritoneal macrophages

Peritoneal exudate macrophages collected from 4-wk-old broilers were capable of in vitro bactericidal activity against Escherichia coli and fungistatic activity against Candida tropicalis. Bactericidal activity was inhibited by treatment of macrophages with 2-deoxyglucose. These assays may be used to evaluate macrophage function in normal, diseased, and immunomodulated birds or to assess the direct effects of cytokines and toxins on macrophage microbicidal activity.     

Comparison of heartworm extract-induced shock and endotoxin-induced shock in dogs by determination of serum tumor necrosis factor concentrations

OBJECTIVE: To compare the mechanisms of heartworm (HW) extract-induced shock and endotoxin-induced shock in dogs by determination of serum tumor necrosis factor (TNF) concentrations. ANIMALS: 11 mixed-breed dogs (7 without and 4 with HW infections). PROCEDURE: Eight dogs were treated with 2 ml of HW extract IV, and 3 dogs were given endotoxin (Escherichia coli lipopolysaccharide [LPS]) at 40 or 400 microg/kg of body weight, IV. Changes in clinical and hematologic findings and serum TNF concentrations were examined from before treatment to 120 minutes after treatment in dogs given HW extract or from before treatment to 180 minutes after treatment in dogs given LPS. Tumor necrosis factor concentration was determined by cytotoxic assay, using WEHI-164 murine sarcoma cells, and plasma endotoxin concentration was determined in 2 dogs treated with HW extract, using the endotoxin-specific chromogenic test. RESULTS: Eight dogs developed shock 3 to 16 minutes after HW extract treatment. Rectal temperature did not change during examination. Serum TNF concentration was detected at a low concentration only 60 and 120 minutes after HW extract treatment, and plasma endotoxin was not detected during examination. In dogs treated with LPS, rectal temperature increased to > 40 C in 2 of 3 dogs, and serum TNF concentration began to increase 30 minutes after LPS treatment, reaching a maximum concentration by 60 minutes. CONCLUSIONS: The cause and mechanism of HW extract-induced shock may be different from those of endotoxin-induced shock, because TNF, which was a pivotal mediator in endotoxin-induced shock, increased minimally in serum of dogs treated with HW extract.     

Adenosine A2A receptor agonists inhibit lipopolysaccharide-induced production of tumor necrosis factor-alpha by equine monocytes

Adenosine is an endogenous nucleoside that regulates many physiological processes by activating one or more adenosine receptor subtypes, namely A1, A2A, A2B and A3. The results of previous studies indicate that adenosine analogues inhibit lipopolysaccharide (LPS)-induced production of reactive oxygen species (ROS) by equine neutrophils primarily through activation of A2A receptors. Because peripheral blood monocytes produce cytokines that are responsible for many of the deleterious effects of LPS, the current study was performed to evaluate the effects of an array of novel adenosine receptor agonists on LPS-induced production of tumor necrosis factor-alpha (TNF-alpha), and to assess the selectively of these agonists for equine adenosine A2A over the A1 receptor. Radioligand binding studies performed with equine tissues expressing adenosine A1 and A2A receptor subtypes yielded a rank order of affinity for the equine A2A receptor of ATL307>ATL309 approximately ATL310 approximately ATL313>ATL202 approximately ATL361 approximately ATL376>ATL372>CGS21680>NECA. Co-incubation of equine peripheral blood monocytes with LPS and these agonists resulted in inhibition of TNF-alpha production with a rank order of potency that strongly correlated with their binding affinities for equine adenosine A2A receptors. Results of experiments performed with one of the adenosine receptor agonists (ATL313) and selective adenosine receptor antagonists confirmed that inhibition of LPS-induced production of TNF-alpha occurred via stimulation of A2A receptors. Although incubation of monocytes with IB-MECA, a compound purported to act as an adenosine A3 receptor agonist, reduced LPS-induced TNF-alpha production, this effect of IB-MECA was inhibited by the A2A selective antagonist ZM241385 but not by the A3 receptor antagonist MRS1220. These results indicate that the adenosine receptor subtype responsible for regulation of LPS-induced cytokine production by equine monocytes is the A2A receptor. To address the signal transduction mechanism responsible for the anti-inflammatory effects of ATL313 in equine monocytes, production of cAMP was compared in the presence and absence of either the adenosine A2A receptor antagonist ZM241385 or the adenosine A2B receptor antagonist MRS1706. In the absence of the antagonists, ATL313 increased production of cAMP; ZM241385 inhibited this effect of ATL313, whereas MRS1706 did not. Furthermore, incubation of monocytes with either the stable analogue of cAMP, dibutyryl cAMP, or forskolin, an activator of adenylyl cyclase, also inhibited LPS-induced production of TNF-alpha production by equine monocytes. Collectively, the results of the current study indicate that adenosine analogues inhibit LPS-induced production of TNF-alpha by equine monocytes primarily via activation of adenosine A2A receptors and do so in a cAMP-dependent manner. The results of this study indicate that stable adenosine analogues that are selective for adenosine A2A receptors may be suitable for development as anti-inflammatory drugs in horses.     

The effect of immunity to core lipopolysaccharides (LPS) on the production of thromboxane and prostacyclin by equine peritoneal macrophages

An experiment was designed to determine whether a change in the ability of macrophages to respond to lipopolysaccharides (LPS) of gram-negative bacteria was involved in the development of cross-reactive immunity to endotoxemia. The endotoxin-induced production of thromboxane A2(TxA2) and prostacyclin (PGI2) by peritoneal macrophages from horses which were hyperimmunized against the common core region of LPS were compared to those in unimmunized horses. Bacterins used for induction of core LPS immunity were prepared from the J-5 mutant of Escherichia coli 0111:B4, and the R 595 mutant of Salmonella minnesota. Serum antibody titers to core LPS were determined by an indirect enzyme-linked immunosorbent assay. Immunized horses had a marked increase in titer to core LPS (p less than 0.05), while there was no change in titer in unimmunized control horses. The only significant difference in the in vitro LPS-induced production of TxA2 and PGI2 by peritoneal macrophages between immunized and control horses was a greater production of TxA2 by macrophages from immunized horses in response to 10 ng/ml LPS (p less than 0.05). Results of this experiment do not support the concept that cross-reactive immunity to LPS is attended by reduced production of TxA2 and PGI2 by equine peritoneal macrophages.     

Expression of procoagulant activity by equine lung macrophages: stimulation by blood lymphocytes

Increases in procoagulant activities (PCA) in equine lung macrophages were induced by non-adherent blood lymphocytes which were prestimulated with phytohaemagglutinin for 48 to 72 hours or by supernatants harvested from prestimulated blood lymphocyte cultures. However, prestimulated lymphocyte suspensions themselves expressed PCA which was most probably derived from contaminating monocytes. Because non-adherent cells from lymphocyte suspensions may have attached to adherent macrophages, cells within lymphocyte suspensions might have contributed to the PCAs expressed by lymphocyte-stimulated lung macrophages. Stimulation of lung macrophages for 24 hours by supernatants of phytohaemagglutinin-prestimulated blood lymphocytes induced a significantly greater PCA increase than stimulation by phytohaemagglutin alone. Thus, cytokines from lymphocyte cultures might have triggered or enhanced PCA induction. Direct stimulation of lung cell preparations with phytohaemagglutinin for 48 hours resulted in a progressive increase of PCA in only two of five specimens tested. The failure to induce PCA in three specimens could be due to the absence of sufficient numbers of T cells within the adherent lung cell preparations. In conclusion, PCA response of equine lung macrophages might be lymphocyte-stimulated in which case PCA might be a useful tool for monitoring the processes of cell-mediated immunity in horses.     

Expression of arginase by mouse myeloid leukemic cell differentiation in vitro induced with tumor necrosis factor.

Induction of arginase activity in mouse myeloid leukemic M1 cells by treatment with recombinant human tumor necrosis factor (rH-TNF) or TNF-elicited mouse serum (TNS) were examined in vitro. M1 cells differentiated into macrophage-like cells by addition of rH-TNF or TNS. The differentiated cells expressed phagocytic function and did not grow anymore. Cytolytic effect of rH-TNF or TNS was not observed. The differentiation of M1 cells into phagocytic cells by the TNS treatment was more rapidly than that by the rH-TNF treatment though TNS contained 25-time less amounts of TNF indicating species specificity of TNF action on myeloid leukemic cells or TNS containing other differentiation factor(s). 3H-ornithine formation from 3H-arginine is catalyzed by arginase (EC. 3.5.3.1). The enzyme product increased in the M1 cell culture medium by the treatment with rH-TNF or TNS. The arginase activity statistically correlated with the appearance percentage of differentiated cells with phagocytic function. These results suggest that in vitro differentiation of M1 cells is accompanied by induction of arginase activity.     

Examination of systemic tumor necrosis factor activity under physiologic and pathophysiologic conditions

Tumor necrosis factor (TNF) activity in the circulation of several animal species was determined by a bioassay, using the murine cell line L929. In healthy adult cattle, horses, pigs and dogs, species specific differences of systemic TNF activity were visible. In cattle, TNF activity in the circulation increased during growing up from calf to adult animal. In cattle suffering from various diseases, unchanged, elevated, but also reduced systemic TNF activity have proved to possess clinical relevance. Low systemic TNF activity frequently occurs during lethal inflammatory diseases and may be an indicator of generalized monozyte paralysis.     

In vitro studies into the influence of ochratoxin A on the production of tumor necrosis factor alpha by the human monocytic cell line THP-1

The influence of pure OTA and an Aspergillus-ochraceus crude toxin on the intracellular expression and the secretion of tumor necrosis factor (TNF) alpha by the monocytic cell line THP-1 was studied in vitro. After 4 hours exposure, the secretion of TNF alpha was inhibited to 50% by pure OTA in a concentration of 400 ng/ml and by crude toxin in a concentration of 100 ng/ml. The same concentrations of mycotoxins impaired the mitochondrial activity of THP-1 cells only marginally. The intracellular expression of TNF alpha was not disturbed by pure OTA in the concentrations tested, whereas crude toxin showed an inhibitory effect. The possible reasons for these findings are discussed.     

In vitro multiplication of Cryptosporidium parvum in mouse peritoneal macrophages.

Cryptosporidium parvum of bovine origin was developed in vitro in unsensitized mouse peritoneal macrophages. Macrophages growing in RPMI medium were infected with sporozoites or with oocysts, and after staining infections were studied by light microscopy. A high parasitic index was obtained with multiple infections occurring commonly. This is a simple method for the study of Cryptosporidium biology, and for in vitro assays of pharmacological activity.     

Serum tumor necrosis factor activity in horses with colic attributable to gastrointestinal tract disease.

Over a 24-month period, serum tumor necrosis factor (TNF) activity was determined in 289 horses with colic attributable to gastrointestinal tract disease. Serum TNF activity was quantitated by use of a modified in vitro cytotoxicity bioassay, using WEHI 164 clone-13 murine fibrosarcoma cells. Causes for colic, determined by clinical and laboratory evaluation, exploratory celiotomy, or necropsy included: gastrointestinal tract rupture (GTR); ileal impaction; small intestinal strangulating obstruction (SIO); proximal enteritis (PE); transient small intestinal distention; large-colon displacement; large-colon volvulus; large-colon impaction; colitis; small-colon obstruction; peritonitis; and unknown. Each diagnosis was placed into 1 of 3 lesion categories: inflammatory disorders (GTR, PE, colitis, peritonitis); strangulating intestinal obstruction (SIO, large-colon volvulus); and nonstrangulating intestinal obstruction (ileal impaction, transient small intestinal distension, large-colon displacement, large-colon impaction, small-colon obstruction, unknown). The prevalence of high serum TNF activity and/or mortality were evaluated. Differences were tested at significance level of P less than 0.05. Approximately 20% of the 289 horses has serum TNF activity greater than that found in clinically normal horses (greater than 2.5 U/ml). Twenty-three horses (8%) had marked increase in serum TNF activity (greater than or equal to 10 U/ml) which was more prevalent among horses with SIO and PE than in horses of other diagnostic groups, except those with GTR. Mortality and marked increase in serum TNF activity were greater in horses with intestinal inflammatory disorders or strangulating intestinal obstruction than in horses with nonstrangulating intestinal obstruction.(ABSTRACT TRUNCATED AT 250 WORDS)     

Sevoflurane inhibits equine myeloperoxidase release and activity in vitro

Objective To investigate the effects of the volatile anaesthetic sevoflurane on the release of total and active myeloperoxidase (MPO) by non‐stimulated and stimulated polymorphonuclear neutrophils (PMNs) in whole blood from healthy horses. Study design In vitro experimental study. Animals Adult healthy horses. Methods Samples of whole venous blood were collected and incubated in air or in air plus 2.3% or 4.6% sevoflurane for 1 hour. PMNs were stimulated with N‐formyl‐methionyl‐leucyl‐phenylalanine (fMLP), with a combination of cytochalasin B (CB) and fMLP or with phorbol myristate acetate (PMA). Total and active MPO contents released by PMNs in blood were measured by enzyme‐linked immunosorbent assay (ELISA) and specific immunological extraction followed by enzymatic detection (SIEFED) respectively. Additional experiments were performed to assess the effect of sevoflurane on the peroxidase and chlorination cycles of purified equine MPO using Amplex Red and 3’‐(p‐aminophenyl) fluorescein as fluorogenic substrates respectively. Results As compared with air alone, 1 hour exposure of whole blood to 4.6% sevoflurane in air significantly inhibited the release of total and active MPO by unstimulated and both fMLP‐ and CB + fMLP‐stimulated PMNs but not by PMA‐stimulated PMNs. Although 2.3% sevoflurane had no effect on total MPO release by unstimulated and stimulated PMNs, it significantly reduced the release of active MPO by unstimulated and fMLP‐stimulated PMNs. Additionally, sevoflurane reversibly inhibited the activity of MPO, especially the peroxidase cycle of the enzyme. Conclusions and clinical relevance Although our experimental study was not designed to assess the effects of sevoflurane in vivo, this inhibition of MPO release and activity may have relevance for anaesthetized horses and deserves further studies to examine the clinical importance of these findings.     

Effects of methylprednisolone acetate and glucosamine on proteoglycan production by equine chondrocytes in vitro

OBJECTIVE: To evaluate the effects of methylprednisolone acetate (MPA) on proteoglycan production by equine chondrocytes and to investigate whether glucosamine hydrochloride modulates these effects at clinically relevant concentrations. SAMPLE POPULATION: Articular cartilage with normal gross appearance from metacarpophalangeal and metatarsophalangeal joints of 8 horses (1 to 10 years of age). PROCEDURES: In vitro chondrocyte pellets were pretreated with glucosamine (0, 1, 10, and 100 microg/mL) for 48 hours and exposed to MPA (0, 0.05, and 0.5 mg/mL) for 24 hours. Pellets and media were assayed for proteoglycan production (Alcian blue precipitation) and proteoglycan content (dimethylmethylene blue assay), and pellets were assayed for DNA content. RESULTS: Methylprednisolone decreased production of proteoglycan by equine chondrocytes at both concentrations studied. Glucosamine protected proteoglycan production at all 3 concentrations studied. CONCLUSIONS AND CLINICAL RELEVANCE: Methylprednisolone, under noninflammatory conditions present in this study, decreased production of proteoglycan by equine chondrocytes. Glucosamine had a protective effect against inhibition of proteoglycan production at all 3 concentrations studied. This suggested that glucosamine may be useful as an adjunct treatment when an intra-articular injection of a corticosteroid is indicated and that it may be efficacious at concentrations relevant to clinical use.     

Bradykinin stimulates prostaglandin E2 production and cyclooxygenase activity in equine nonglandular and glandular gastric mucosa in vitro

REASONS FOR PERFORMING STUDY: There are few data available regarding regulation of prostaglandin (PG) generation by equine gastric mucosae and the role of the cyclooxygenase (COX) isoforms in their production. OBJECTIVES: To: 1) characterise and quantify PGE2 output in vitro; 2) examine the sensitivity of PGE2 production to exogenous bradykinin (BK) exposure; 3) determine the contribution of the COX-1 and COX-2 pathways to basal and BK-stimulated PGE2 production; and 4) measure if BK influences electrogenic ion transport in equine gastric mucosae in vitro. METHODS: Full thickness gastric sheets were obtained from horses at post mortem, stripped of muscle layers and mounted in Ussing chambers. Tissues were exposed to bradykinin (BK, 0.1 micromol/l) either alone, or following pretreatment with a selective COX-2 inhibitor (NS-398, 1 micromol/l) or a nonselective COX inhibitor (piroxicam, 1 micromol/l), or were untreated. RESULTS: BK administration increased PGE2 output from the basolateral but not the apical faces of both tissue types. Piroxicam, but not NS-398, reduced basolateral PGE2 release below control levels in both tissue types. Both piroxicam and NS-398 pretreatment inhibited BK-stimulated PGE2 release. In separate experiments, BK was without effect upon electrophysiological parameters of tissues mounted in Ussing chambers. CONCLUSIONS: PGE2 is produced by the nonglandular and glandular equine gastric mucosae in vitro. Significantly more PGE2 is released basolaterally than apically. BK stimulated the production of PGE2 from the basolateral side of both tissue types. These findings suggest that COX-1 is a significant pathway for basal PGE2 production from the basolateral faces of both nonglandular and glandular equine gastric mucosae in vitro.