Different Prostanoids Are Involved in Bradykinin-Induced Endothelium-Dependent and -Independent Vasoconstriction in Rat Mesenteric Resistance Arteries

Home , Seratrodast

J Pharmacol Sci 94, 115 – 121 (2004) Journal of Pharmacological Sciences ©2004 The Japanese Pharmacological Society Full Paper Different Prostanoids Are Involved in Bradykinin-Induced Endothelium-Dependent and -Independent Vasoconstriction in Rat Mesenteric Resistance Arteries

Hideki Nawa1, Yuji Kurosaki1, and Hiromu Kawasaki1,*

1Department of Clinical Pharmaceutical Science, Graduate School of Natural Science and Technology, Okayama University, 1-1-1 Tsushima-naka, Okayama 700-8530, Japan

Received December 3, 2002; Accepted December 3, 2003

Abstract. Mechanisms underlying bradykinin-induced vasoconstriction were investigated in rat perfused mesenteric vascular beds with active tone. In preparations with intact endothelium, bolus injections of bradykinin (1 to 1,000 pmol) dose-dependently produced three-phase vascular effects, which consisted of a first-phase vasodilation followed by a second-phase vasoconstriction and a subsequent third-phase vasodilation; these effects were abolished by FR172357 9 8 (bradykinin B2-receptor antagonist), but not by des-Arg -[Leu ]-bradykinin (bradykinin B1-receptor antagonist). In preparations with intact endothelium, indomethacin (cyclooxygenase inhibitor), seratrodast (thromboxane A2 (TXA2)-receptor antagonist), ONO-3708 (TXA2/prostaglandin H2 (PGH2)-receptor antagonist) or ozagrel (TXA2 synthesis inhibitor) markedly inhibited the bradykinin-induced vasoconstriction. In preparations without endothelium, the bradykinin-induced vasoconstriction was abolished by indomethacin and ONO-3708, while seratrodast and ozagrel had no effect. These results suggest that the endothelium-dependent vasoconstriction of bradykinin is mainly mediated by TXA2 and that prostanoids other than TXA2, probably PGH2, in mesenteric vascular smooth muscle are responsible for bradykinin-induced endothelium-independent vasoconstriction.

Keywords: bradykinin, vasoconstriction, thromboxane A2, prostaglandin H2, rat mesenteric resistance artery

Introduction We previously reported that exogenous bradykinin in the rat mesenteric resistance artery induces triphasic Bradykinin is a potent endogenous vasoactive stimu- vascular responses, which consisted of an initial phase lant, mediating both vasodilation and vasoconstriction vasodilation followed by a second-phase vasoconstric- (1 – 3). Bradykinin produces vascular action via authen- tion and a third-phase gradual vasodilation (14). We also tic receptors, the B1 and B2 subtypes, in endothelial and suggested that the arachidonic acid metabolite, probably smooth muscle cells (4 – 7). The mechanisms underly- TXA2, in the mesenteric artery was responsible for the ing the vasoconstrictions depend on the tissues and the vasoconstrictor response to bradykinin (14). However, species studied (2, 8, 9). Several studies reported that the role of prostanoids in bradykinin-induced vaso- the cyclooxygenase products, prostaglandin (PG) H2 constrictor responses of the mesenteric resistance artery (PGH2), thromboxane A2 (TXA2), and PGF2, mediate remains unknown. The present study was, therefore, the contraction evoked by the peptide (10 – 12). designed to investigate precise mechanisms underlying Although the vasorelaxant effect of bradykinin is well bradykinin-induced vasoconstriction in rat mesenteric known to be endothelium- and nitric oxide (NO)-depen- resistance arteries. dent (3, 7, 13), the nature and role of its vasoconstrictor effect is unclear. Materials and Methods

*Corresponding author. FAX: +81-86-251-7970 Animals E-mail: [email protected] Male Wistar rats, weighing 220 to 350 g, were used

115 116 H Nawa et al in the present study. Animals were given food and (model 975; Harvard Apparatus, Holliston, MA, USA). water ad libitum. They were housed in the Animal Injection volumes were 100 L for 12 s. Research Center of Okayama University at a controlled ambient temperature of 22 M 2LC with 50 M 10% relative Experimental protocols humidity and with a 12-h light/12-h dark cycle (lights Isolated mesenteric vascular beds were perfused with on at 8:00 a.m.). Krebs solution, and perfusion pressure was increased by perfusion with Krebs solution containing the 1-adreno- Perfusion of mesenteric vascular beds ceptor agonist methoxamine (7 M) in the presence of The animals were anesthetized with pentobarbital-Na guanethidine (5 M) that was added to exclude possible (50 mg/kg, intraperitoneally) and the mesenteric vas- involvement of vascular adrenergic nerves by blocking cular beds were isolated and prepared for perfusion as adrenergic neurotransmission. After the elevated perfu- described previously (14, 15). The superior mesenteric sion pressure stabilized, bradykinin at doses of 1, 3, 10, artery was cannulated and flushed gently with Krebs- 30, 100, 300, and 1,000 pmol were directly injected Ringer bicarbonate solution (Krebs solution) to elimi- into the perfusate using the injection pump. Injection nate blood in the vascular bed. After removal of the volumes were 100 l for 12 s. To assess the mechanisms entire intestine and associated vascular bed, the mesen- underlying the vascular effect of bradykinin, bolus injec- teric vascular bed was separated from the intestine by tions of bradykinin were performed in preparations cutting close to the intestinal wall. Only four main with intact endothelium during perfusion of the brady- 9 8 arterial branches from the superior mesenteric trunk kinin B1-receptor antagonist des-Arg -[Leu ]-bradykinin running to the terminal ileum were perfused. All other (1 M) (6); bradykinin B2-receptor antagonist branches of the superior mesenteric artery were tied FR172357 (0.1 M) (16); cyclooxygenase inhibitor off. The isolated mesenteric vascular bed was then indomethacin (0.05 and 0.5 M); TXA2-receptor antago- placed in a water-jacketed organ bath maintained at nist seratrodast (0.5 and 5 M) (Ki values 60 nM for 37LC and perfused with a modified (see below) Krebs TXA2 receptors) (17); TXA2 synthase inhibitor ozagrel solution at a constant flow rate of 5 mL/min with a (10 and 100 M) (18); and TXA2/PGH2-receptor antago- peristaltic pump (model AC-2120; ATTO Co., Tokyo). nist ONO-3708 (5 and 10 M) (Ki values 17.2 nM for Preparations were also superfused with the same solu- TXA2/PGH2 receptors) (19). tion at a rate of 0.5 mL/min to prevent drying. The To assess the influence of endothelium, the brady- Krebs solution was bubbled with a mixture of 95% O2 – kinin-induced vascular responses were examined in 5% CO2 before passage through a warming coil main- preparations with chemical removal of the endothelium tained at 37LC. Modified Krebs solution was of the by SD. After SD perfusion, the active tone of the following composition: 119.0 mM NaCl, 4.7 mM KCl, preparation was produced by perfusion with Krebs 2.4 mM CaCl2, 1.2 mM MgSO4, 25.0 mM NaHCO3, solution containing methoxamine and guanethidine 1.2 mM KH2PO4, 0.03 mM disodium EDTA, and (5 M). The concentration of methoxamine was reduced 11.1 mM dextrose (pH 7.4). Changes in the perfusion to 2 M from 7 M because the vasoconstrictor pressure were measured with a pressure transducer response was enhanced after endothelium removal. (model TP-400T; Nihon Kohden, Tokyo) and recorded After confirming successful removal of the endothelium using a pen recorder (model U-228; Nippon Denshi by the lack of a relaxant effect with 1 nmol ACh injec- Kagaku, Tokyo). tion, bradykinin injections were carried out. In preparations without endothelium, bradykinin injections were Chemical removal of vascular endothelium performed during perfusion of Krebs solution containing To remove the vascular endothelium, preparations indomethacin (0.5 M), seratrodast (5 M), ozagrel with resting tone were perfused with a 1.80 mg/mL (100 M), or ONO-3708 (10 M), and methoxamine solution of sodium deoxycholate (SD) in saline for 30 s. and guanethidine. This caused a transient increase (20 – 30 mmHg) in At the end of each experiment, preparations were perfusion pressure. Then, the preparations were rinsed perfused with 100 M papaverine to cause complete with SD-free Krebs solution for 40 min. After the relaxation. Vasodilator activity is expressed as percent- preparations were contracted by perfusion with Krebs age of the perfusion pressure at maximum relaxation solution containing methoxamine, chemical removal of induced by papaverine. Vasoconstrictor activity is ex- the endothelium was assessed by the lack of a relaxant pressed as percentage of the perfusion pressure before effect after a bolus injection of 1 nmol acetylcholine the bradykinin injection. (ACh), which was injected directly into the perfusate proximal to the arterial cannula with an injection pump Bradykinin-Induced Vasoconstriction 117

Statistical analyses Results Experimental results are presented as the means M S.E.M. Statistical analysis was evaluated by Tukey’s Vascular effect of bradykinin in rat perfused mesenteric multiple comparison test. Values of P0.05 were consid- arteries with endothelium ered to indicate statistical significance. As shown in Fig. 1, a bolus injection of ACh in the mesenteric vascular bed with active tone caused a Drugs sharp decrease in perfusion pressure due to vasodilation, The following drugs were used: ACh chloride (Daiichi Pharmaceutical Co., Tokyo); bradykinin and des-Arg9-[Leu8]-bradykinin (Peptide Institute, Osaka); guanethidine sulphate and indomethacin (Sigma Chemi- cal Co., St. Louis, MO, USA); FR172357 (3-bromo- 8-[2,6-dichloro-3-[N-[(E)-4-(N,N-dimethylcarbamoyl) cinnamidoacetyl]-N-methylamino]benzyloxy]-2-metyl- imidazo[1,2-]pyridine) (Fujisawa Pharmaceutical Co., Tokyo); ONO-3708 (7-[2,4-(dimethylmethano)-6- (2-cyclohexyl-2-hydroxyacetamido)-1-cyclohexyl]- 5(Z)-heptanoic acid) and ozagrel (Ono Pharmaceutical Co., Osaka); seratrodast (Takeda Chemical Industries, Ltd., Osaka); methoxamine hydrochloride (Nihon Shin- yaku Co., Kyoto); papaverine hydrochloride (Dainippon Pharmaceutical Co., Osaka); and SD (Ishizu Seiyaku Co., Tokyo). All drugs, except SD, ONO-3708, seratrodast, and indomethacin, were dissolved in distilled water and diluted with Krebs solution containing 2 to 7 M methoxamine and 5 M guanethidine when perfused and injected directly. SD was dissolved in 0.9% saline. Seratrodast and indomethacin were dissolved in ethanol and diluted with Krebs solution. ONO-3708 was dissolved in dimethylsulfoxide (DMSO) (Wako Pure Chemical Industries, Ltd., Osaka) and diluted with Krebs solution (final concentration of DMSO, 0.2 g/ml).

Fig. 2. Effects of des-Arg9-[Leu8]-bradykinin (1 M) and FR172357 (0.1 M) on vasoconstrictor responses to bolus injections of bradykinin in rat perfused mesenteric vascular beds with active tone and intact endothelium. *P0.05, **P0.01, compared with the control bradykinin response.

Fig. 1. Typical records showing triphasic vascular responses induced by bolus injections of bradykinin (BK) in rat perfused mesenteric vascular beds with active tone and with intact endothelium. Note that bradykinin induced triphasic vascular responses: the first-phase sharp vasodilation, the second-phase transient vasoconstriction and the third-phase long-lasting vasodilation. ACh: injection of acetylcholine, PPV: perfusion of 100 M papaverine. 118 H Nawa et al

Fig. 3. Effects of indomethacin (A: 0.05 and 0.5 M) and ozagrel (B: TXA2 synthetase inhibitor, 10 and 100 M) on vasoconstrictor responses to bolus injections of bradykinin in rat perfused mesenteric vascular beds with active tone and with intact endothelium. *P0.05, **P0.01, compared with the control bradykinin response.

Fig. 4. Effects of ONO-3708 (A: 5 and 10 M) and seratrodast (B: 0.5 and 5 M) on vasoconstrictor responses to bolus injections of bradykinin in rat perfused mesenteric vascular beds with active tone and with intact endothelium. *P0.05, **P0.01, compared with the control bradykinin response. Bradykinin-Induced Vasoconstriction 119 which is mediated by intact endothelium (20). In this Effect of indomethacin, seratrodast, ONO-3708, or preparation, as shown in Fig. 1, bolus injections of ozagrel treatment on vascular responses to bolus injec- bradykinin (1 to 1,000 pmol) caused triphasic vascular tions of bradykinin in rat perfused mesenteric arteries responses: the first-phase was an initial short and with endothelium sharp fall in perfusion pressure due to vasodilation As shown in Figs. 3 and 4, treatment with indo- followed by a second-phase transient increase in per- methacin (0.05 and 0.5 M) (Fig. 3A), ozagrel (10 and fusion pressure due to vasoconstriction; and this was 100 M) (Fig. 3B), ONO-3708 (5 and 10 M) then followed by the third-phase gradual decrease in (Fig. 4A), or seratrodast (0.5 and 5 M) (Fig. 4B) perfusion pressure due to vasodilation. As shown in significantly inhibited the second-phase vasoconstric- 9 Fig. 2, the bradykinin B1-receptor antagonist des-Arg - tion in response to bradykinin injection in a concentra- [Leu8]-bradykinin (1 M) had no effect on the brady- tion-dependent manner. kinin-induced second-phase vasoconstriction. However, as shown in Fig. 2, the bradykinin B2-receptor antagonist Effect of indomethacin, ONO-3708, seratrodast, or FR172357 (0.1 M) abolished vasoconstrictor responses ozagrel treatment on vascular responses to bolus injec- to bolus injections of bradykinin. tions of bradykinin in rat perfused mesenteric arteries without endothelium Treatment of the perfused mesenteric vascular bed with SD abolished the ACh-induced vasodilation, indicating successful removal of the endothelium. In preparations without endothelium, indomethacin (0.5 M) (Fig. 5) and ONO-3708 (10 M) (Fig. 5) markedly inhibited the second-phase vasoconstriction induced by bradykinin injections. However, as shown in Fig. 5, in preparations without endothelium, neither seratrodast (5 M) nor ozagrel (100 M) showed an effect on the second-phase vasoconstriction induced by bradykinin injections.

Discussion

The present study demonstrated that exogenous bradykinin in the rat mesenteric artery induced dose- dependent vascular responses, which consisted of three phases: an initial sharp vasodilation followed by a second transient vasoconstriction and a subsequent third gradual vasodilation. These vascular responses to bradykinin injections were abolished by the bradykinin B2-receptor antagonist FR172357 (16), whereas the 9 8 bradykinin B1-receptor antagonist Des-Arg -[Leu ]- bradykinin had no effect. In the previous report, we showed that the bradykinin B1-receptor agonist Des- Arg9-bradykinin induced no vascular response in the rat mesenteric artery (14). Thus, it is very likely that the triphasic response to bradykinin injection was mediated by bradykinin B2 receptors. These results are in line with those reported in the rat and rabbit arteries (12, 14, 21). In the preparation with intact endothelium, treatment with indomethacin (cyclooxygenase inhibitor), seratro- Fig. 5. Effect of treatment with indomethacin (0.5 M), ONO-3708 dast (TXA2-receptor antagonist), ONO-3708 (TXA2

(10 M), seratrodast (5 M), or ozagrel (100 M) on vasoconstrictor /PGH2-receptor antagonist), or ozagrel (TXA2 synthase responses to bolus injections of bradykinin in rat perfused mesenteric inhibitor) markedly inhibited the second-phase vaso- vascular beds with active tone and without endothelium. *P0.05, **P0.01, compared with the control bradykinin response in the constriction in response to bradykinin injection. These denuded preparations. results clearly suggest that arachidonic acid metabolites, 120 H Nawa et al probably TXA2 and related products, are responsible for Acknowledgments the second-phase vasoconstriction induced by bradykinin. However, in mesenteric arteries without endo- This study was supported in part by Grants-in-Aid thelium, neither seratrodast nor ozagrel showed an for Scientific Research (KAKENHI) from Ministry of effect on the second-phase vasoconstriction induced by Education, Culture, Sports, Science and Technology of bradykinin. TXA2 has been shown to be released from Japan (Nos. 10557244 and 13672389). Seratrodast endothelial cells (22, 23). Thus, it is likely that brady- and FR172357 were generously donated by Takeda kinin stimulates TXA2 release mainly in the presence of Chemical Industries (Osaka) and Fujisawa Pharma- intact endothelium to cause vasoconstriction. This also ceutical Co. (Tokyo), respectively. ONO-3708 and suggests that TXA2 is released from endothelium but ozagrel were generously donated by Ono Pharma- not from vascular smooth muscle cells when bradykinin ceutical Co. (Osaka). B2 receptors are stimulated. In contrast, indomethacin abolished the bradykinin- References induced second-phase vasoconstriction even in the mesenteric artery without endothelium. This finding 1 Drummond GR, Cocks TM. Endothelium-dependent relaxations suggests that prostanoids other than TXA2 are involved mediated by inducible B1 and constitutive B2 kinin receptors in in the bradykinin-induced vasoconstriction in the de- the bovine isolated coronary artery. Br J Pharmacol. 1995;116: nuded mesenteric artery. In the present study, treatment 2473–2481. 2 Bagaté K, Develioglu L, Imbs JL, Michel B, Helwig JJ, with ONO-3708, which blocks TXA2/PGH2 receptors, Barthelmebs M. Vascular kinin B1 and B2 receptor-mediated markedly inhibited the bradykinin-induced vasocon- effects in the rat isolated perfused kidney – differential regula- striction. Thus, this finding strongly suggests that PGH2 tions. Br J Pharmacol. 1999;128:1643–1650. is responsible for the bradykinin-induced endothelium- 3 Ihara E, Hirano K, Derkach DN, Nishimura J, Nawata H, independent vasoconstriction in the rat mesenteric Kanaide H. The mechanism of bradykinin-induced endothelium- artery. Bradykinin has been shown to cause vaso- dependent contraction and relaxation in the porcine interlobar constriction (10, 24, 25). The bradykinin-induced vaso- renal artery. Br J Pharmacol. 2000;129:943–952. constrictor response has been reported to be endo- 4 Regoli D, Barabe D. Pharmacology of bradykinin and related kinins. Pharmacol Rev. 1980;32:1–46. thelium-independent and indomethacin-sensitive or 5 Burch RM, Kyle DJ. Recent developments in the understanding mediated by bradykinin B2 receptors located in the of bradykinin receptors. Life Sci. 1992;50:829–838. vascular smooth-muscle layer (2, 9). In addition, other 6 Regoli D, Allogho SN, Rizzi A, Gobeil FJ. Bradykinin receptors studies have shown that the action of bradykinin was and their antagonists. Eur J Pharmacol. 1998;348:1–10. largely dependent on stimulation of the cyclooxygenase 7 Miyamoto A, Ishiguro S, Nishio A. Stimulation of bradykinin pathway to produce PGH2 and possibly TXA2, which B2-receptors on endothelial cells induces relaxation and contraction in porcine basilar artery in vitro. Br J Pharmacol. 1999; can activate smooth muscle TXA2/PGH2 receptors to elicit vasoconstriction (10, 11). However, the brady- 128:241–247. 8 Ohlmann P, Martinez MC, Schneider F, Stoclet JC, Andriant- kinin-induced vasoconstriction in the canine isolated sitohaina R. Characterization of endothelium-derived relaxing saphenous vein without endothelium was potentiated factors released by bradykinin in human resistance arteries. Br J by indomethacin (26). Therefore, involvement of pros- Pharmacol. 1997;121:657–664. tanoids in bradykinin-induced endothelium-independent 9 Persson K, Andersson RG. Biphasic response to bradykinin in vasoconstriction may be species and tissue-dependent isolated porcine iliac arteries is mediated by bradykinin B1 and (2, 9). B2 receptors. J Cardiovasc Pharmacol. 1998;31:306–313. From these results, it is concluded that bradykinin 10 Aksoy MO, Harakal C, Smith JB, Stewart GJ, Zerweck CR. Mediation of bradykinin-induced contraction in canine veins via 2 2 stimulates TXA release from endothelium and PGH thromboxane/prostaglandin endoperoxide receptor activation. release from vascular smooth muscle cells to cause Br J Pharmacol. 1990;99:461–466. endothelium-dependent and independent vasoconstric- 11 Weinberg ML, Moreira E, Weinberg J. Arachidonic acid tion in the rat mesenteric resistance artery, respectively. products-mediated contraction induced by bradykinin in relaxed The present study suggests that mechanisms underlying mesenteric arterial rings from Holtzman rats. Eur J Pharmacol. bradykinin-induced vasoconstriction are different in the 1997;320:145–150. artery with or without the endothelium and that the 12 Yu H, Carretero OA, Juncos LA, Garvin JL. Biphasic effect of bradykinin on rabbit afferent arterioles. Hypertension. 1998;32: characteristics of TXA2/PGH2 receptors in the endo- 287–292. thelium may be different from those of the receptors in 13 Lundgaard A, Aalkjaer C, Bjurholm A, Mulvany MJ, Hansen rat mesenteric vascular smooth muscle. ES. Vasorelaxation in isolated bone arteries: vasoactive intestinal peptide, substance P, calcitonin gene-related peptide, and bradykinin studied in pigs. Acta Orthop Scand. 1997;68: Bradykinin-Induced Vasoconstriction 121

481–489. Nature. 1980;288:373–376. 14 Nawa H, Kawasaki H, Nakatsuma A, Isobe S, Kurosaki Y. 21 Berguer R, Hottenstein OD, Palen TE, Stewart JM, Jacobson Triphasic vascular responses to bradykinin in the mesenteric ED. Bradykinin-induced mesenteric vasodilation is mediated by resistance artery of the rat. Eur J Pharmacol. 2001;433:105–113. B2-subtype receptors and nitric oxide. Am J Physiol. 1993;264: 15 Kawasaki H, Takasaki K, Saito A, Goto K. Calcitonin gene- G492–G496. related peptide acts as a novel vasodilator neurotransmitter in 22 Bouaziz A, de Ficquelmont-Loïzos MM, Richert A, Caprani A. mesenteric resistance vessels of the rat. Nature. 1988;335:164– Direct physical factors and PGI2 and TXA2 secretions by a 167. human endothelial cell line: in vitro investigation of pressure 16 Rizzi C, Rizzi A, Calò G, et al. Characterization of FR172357, a and shear stress applied independently in synergy. Thromb Res. new non-peptide bradykinin B2 receptor antagonist, in human, 1998;90:279–289. pig and rabbit preparations. Eur J Pharmacol. 1999;386:25–31. 23 Gao J, Zhao J, Rayner SE, Van Helden DF. Evidence that the 17 Kurokawa T, Matsumoto T, Ashida Y, Sasada R, Iwasa S. ATP-induced increase in vasomotion of guinea-pig mesenteric Antagonism of the human thromboxane A2 receptor by an anti- lymphatics involves an endothelium-dependent release of asthmatic agent AA-2414. Biol Pharm Bull. 1994;17:383–385. thromboxane A2. Br J Pharmacol. 1999;127:1597–1602. 18 O’Keefe EH, Liu ECK, Greenberg R, Ogletree ML. Effects of a 24 Fasciolo JC, Vargas L, Lama MC, Nolly H. Bradykinin-induced thromboxane synthetase inhibitor and a thromboxane antagonist vasoconstriction of rat mesenteric arteries precontracted with on release and activity of thromboxane A2 and prostacyclin in noradrenaline. Br J Pharmacol. 1990;101:344–348. vitro. Prostaglandins. 1985;29:785–797. 25 Campos AH, Calixto JB. Mechanisms involved in the contractile 19 Kondo K, Seo R, Naka M, et al. Effects of ONO-3708, an responses of kinins in rat portal vein rings: mediation by B1 antagonist of the thromboxane A2 / prostaglandin endoperoxide and B2 receptors. J Pharmacol Exp Ther. 1994;268:902–909. receptor, on platelet aggregation and thrombosis. Eur J Pharma- 26 Marsault R, Illiano S, Vanhoutte PM. Bradykinin-induced col. 1989;163:253–261. contractions of canine saphenous veins: mediation by B2 20 Furchgott RF, Zawadzki JV. The obligatory role of endothelial receptors and involvement of eicosanoids. Br J Pharmacol. cells in the relaxation of arterial smooth muscle by acetylcholine. 1997;120:215–220.