Pharmacology zyxwvutsrqpo & Toxicology zyxwvutsrqpo 1992, zyxwvutsrq 71, 330-334. zyxwvutsr Mechanical Effects of Some Prostanoids on Isolated Human Oesophageal Submucosal Veins G. Madsen', A. Tettrup', K.-E. Andersson' and A. Forman' 'Department of Surgical Gastroenterology L, 20bstetrics and Gynaecology, Aarhus Municipal Hospital, DK-8000 Aarhus C, Denmark, and 'Department of Clinical Pharmacology, University of Lund, Sweden (Received January 28, 1992; Accepted June 1, 1992) Abstract: The effects of prostaglandins El, El, F2=, prostacyclin, and the thromboxane A,-mimic U46619 were investigated on isolated human oesophageal submucosal veins from the oesophageal body and the oesophagogastric junction. U46619 most potently, but also PGFzn produced venocontraction without differences between preparations from the oesophageal body and the oesophagogastric junction. PGE, and prostacyclin caused relaxation of vessels precontracted with U46619 M). PGE, induced either contraction or relaxation, but biphasic effects in the same vessel segment were not seen. lndomethacin M inhibited the contractile responses to both noradrenaline and zyxw K+ (124 mM), suggesting that the agonists induced synthesis or release of vasoconstrictor prostanoids. Prostanoids exert potent mechanical effects in submucosal oesophageal veins and may be of physiological importance. Prostanoids are synthesized in the mucosa of the gastroin- testinal tract and are potent regulators of gastric and intesti- nal blood flow (Whittle & Salmon 1983; Konturek & Pawlik 1986; Whittle & Vane 1987). The role of these agents in control of venous tone is, however, less well known. Recent findings suggest that prostacyclin production is elevated in portal hypertension (Hamilton et zyxwvutsrqp al. 1982; Rimola et al. 1986), and high concentrations of prostacyclin metabolites have been demonstrated in blood from the portal vein of patients suffering from hepatic cirrhosis (Ritter et zyxwvuts al. 1986). In a recent study, haemodynamic changes induced after partial portal vein ligation in rabbits were presented by indomethacin and mimicked by prostacyclin infusion in control animals (Sitzmann et al. 1989). Thus, prostanoids could play a role in the haemodynamic changes in portal hypertension and also in the formation of oesophageal vari- ces. To our knowledge, however, the action of prostanoids on human oesophageal veins has not been investigated. In the present experiments submucosal veins were isolated from human oesophagus, and the mechanical effects of some prostanoids were investigated. To test whether prost- anoids can be generated by the vessel, preparations were exposed to indomethacin in resting state and before acti- vation with noradrenaline and potassium. Materials and Methods Oesophageal tissue was obtained from 19 patients (15 males and 4 females) operated on for gastric carcinoma (n = 6), carcinoma of the cardia (n = 8) or oesophageal cancer (n = 5). Median age was 68 years (range: 43-82). None of the patients had received antiinflam- matory drugs prior to the operation. The oesophageal tissue specimens were immediately placed in oxygenated Krebs solution (4", for composition see below). The submucosal veins from either the oesophagogastric junction or the oesophageal body, 5-8 cm above the oesophagogastric junction, were isolated using a dissection microscopy (Kyowa) and divided into approximately 1.5 mm long ring segments. Each vessel was mounted on two stainless steel wires (diameter 90 pm), one of which was connected to a force transducer (Grass FT.03). The other was attached to a movable unit, permitting adjustment of the distance between the wires. The temperature was kept constant (37.0" zy k 0.5) by a thennoregulated water circuit surrounding the organ bath. The chambers were continuously bubbled with 5% C 0 2 in O2 main- taining pH at 7.40f0.05. Isometric contractions were recorded on a 6 channel polygraph (Beckmann R 61 I). By repeated stretching, the preparations from the oesophageal body and the oesophagogastric junction were subjected to a wall tension of 0.7 mN/mm and 0.3 mN/mm, respectively, which is close to their optimum for mechanical performance. In this experimental setup, human oesophageal submucosal veins show no active basal tone (Madsen et al. 1991). After an accommodation period of one hour, during which the Krebs solution was changed at 20 min. intervals, the vessels were stimulated with a 124 mM K+ solution (for composition, see below) every 20 min. until reproducible re- sponses were obtained (less than 10% variation between two sub- sequent contractions). Subsequent contractile responses to drugs were expressed in per cent of the last two of the initial responses to K+ (124 mM). Drugs inducing contraction were added cumula- tively, and concentration-response curves were constructed. Relax- ant responses were examined in veins precontracted by U-46619 M), and a segment of the same vein was run as a control. Relaxant responses were given as per cent reduction of the initial tension induced by U-46619. When possible, the maximum contractile or relaxing response (Emax) was determined in each experiment, and the drug concen- tration producing half maximum response (EC,,) was assessed by linear interpolation on the semilogarithmic concentration-response curve as pD2 = - log(EC,,). The effect of cyclooxygenase inhibition was studied by pretreat- ment of the preparation with indomethacin M) for 30 min. prior to stimulation with 124 mM K+ or noradrenaline M). All concentrations are final bath concentrations. The experiments were terminated less than 18 hr after removal of the specimens from the patients. Only one prostanoid was tested with the same vessel segment. Calculations. Threshold concentration was defined as a mechanical response (contraction or relaxation) significantly different from the base line level. All values are meanfS.E.M. and Student's t-test was used for statistical analysis.