VASCULAR PHARMACOLOGY OF NITRIC OXIDE SYNTHASE INHIBITORSbyYONG-XIANG WANGM.D., Anhui Medical University, 1982M.Sc., Anhui Medical University, 1985A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Pharmacology & Therapeutics, Faculty of Medicine)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIASEPTEMBER, 1993© Yong-Xiang Wang, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of________________________The University of British ColumbiaVancouver, CanadaDate________________DE.6 (2188)11AbstractThis dissertation examined the effects of nitric oxide synthase (NOS)inhibitors, NGsubstituted arginine analogues (NSAAs) including NGnitroarginine(NNA) and its methyl ester (NAME), as well as diphenyleneiodonium (DPI), onendothelium-dependent vasodilatation and pressor response in rats.All NSAAs caused complete, prolonged and “noncompetitive” inhibition ofacetyicholine (ACh)-, the calcium ionophore A 23187- and/or bradykinin-inducedrelaxations in vitro and ex vivo, with lower potencies for the D-enantiomers. TheHill coefficient (n)s for the inhibition of NSAAs, were not significantly differentfrom 1. The inhibition of NSAAs was antagonized by either pre- or post-treatmentwith L-arginine (L-Arg). L-NAME partially inhibited ACh-induced depressorresponse in vivo. NSAAs caused long-lasting pressor responses in conscious rats.The ns of the dose-pressor response curves of L-NNA, L-NAME and D-NAME were2 or more. The pressor response to L-NNA or L-NAME was not attenuated bypithing, or treatment with mecamylamine, reserpine, phentolamine, captopril, orindomethacin, while those to NSAAs were attenuated by L-Arg; L-Argcompetitively antagonized the pressor response to L-NAME at n=2.DPI caused complete,Iong-Iasting and “noncompetitive” inhibition of ACh- andA 23187-induced vasodilatation in vitro and partial inhibition of ACh-inducedvasodilatation in vivo. The inhibition was prevented by pretreatments withnicotinamide adenine dinucleotide (NADPH) and flavin adenine dinucleotide (FAD),but was slightly reversed by post-treatment with NADPH. DPI caused transient111pressor response with n of 3 or 4, as well as raised plasma catecholamines in rats.The pressor response to DPI was attenuated by pithing and spinal cordtransection, as well as treatment with tetrodotoxin, reserpine, mecamylamine,guanethidine or phentolamine.Therefore, both NSAA5 and DPI inhibit endothelium-dependent relaxations,the inhibiton of NSAAs is reversible whereas that of DPI is irreversible. The Lconfiguration is preferred but not essential for NSAA5 to inhibit endotheliumdependent relaxation. The mechanism of NSAAs involves antagonizing the NOSsubstrate L-Arg, whereas that of DPI involves interfering with the NOS cofactorsFAD and NADPH. The pressor responses to NSAAs are competitively antagonizedby L-Arg and are not dependent on the integrity of the central and autonomicnervous, angiotensin or prostaglandin system. The pressor response to DPI is dueto sympathetic activation.ivTABLE OF CONTENTSChaDter PageAbstract HTable of contents ivList of abbreviations ViiiList of tables XList of figures XiAcknowledgements XiiPreface Xiii1. Introduction 11.1. EDRF and NO 11 .1 .1. Endothelium-dependent relaxation and EDRF 11.1.2. The pharmacology of NO 31.1.3. EDRF is NO or a labile nitroso precursor 51.2. NOS 71.2.1. Constitutive NOS 91.2.2. Inducible NOS 101.3. NOS inhibitors 101.3.1. NSAAs 101.3.1.1. Structures 101.3.1.2. Competitive inhibition 111.3.1.3. Reversibility 121.3.1.4. Stereospecificity 121.3.1.5. Specificity 131.3.2. DPI and other iodonium compounds 131.3.3. Endogenous NOS inhibitors 141 .4. Actions of NOS inhibitors on the vasculature 141.4.1. Inhibition of endothelium-dependent vasodilatation 141 .4.1 .1. Discrepancy between in vitro and in vivo 141 .4.1 .2. Stereospecificity of NSAAs 1 61 .4.2. Vasoconstriction in vivo 171 .5. Mechanisms of the pressor responses to NSAAs 1 81.5.1. The endothelial L-Arg/NO pathway 191 .5.2. Endothelial membrane L-Arg transport 191 .5.3. The central nervous system 201 .5.4. The autonomic nervous system 211 .5.5. Afferent nerve transmitters 23V1 .5.6. The renin-angiotensin system 231 .5.7. The prostaglandin system 241.5.8. Vasopressin 241 .6. The aims of this study 251.6.1. Stereospecificity of NSAAs. 251 .6.2. Effects of NSAAs and DPI on ACh-induced vasodilatationsin vitro and in vivo 261 .6.3. Pharmacodynamics of NSAAs and DPI 261 .6.4. Mechanisms of the pressor responses to NSAAs and DPI 272. Materials and methods 282.1. Materials 282.1.1. Animals 282.1.2. Drugs 282.2. Methods 292.2.1. Isolated aortic rings 292.2.2. Surgery 302.2.3. Microsphere technique 312.2.4. Measurement of plasma catecholamines 312.3. Calculations and statistical analyses 322.3.1. Dose (concentration)-response curve 322.3.2. Modified Schild plot 332.3.3. Statistics 343. Results and discussion I 363.1. Results 363.1.1. Effects of L-NNA and D-NNA on contractile responses of theisolated aorta 363.1.2 Effects of NSAAs on endothelium-dependent relaxation in vitro 363.1 .2.1. Concentration-responses 363.1.2.2. Mechanisms 413.1.2.3. Time courses and reversibility 433.1.3. Effects of L-NNA and D-NNA on ACh- and SNP-inducedrelaxations ex vivo 453.1.4. Effects of L-NAME and D-NAME on the depressor responses toACh and SNP in vivo 453.1.5. Haemodynamic effects of L-NNA 503.1.6. Pressor responses to NSAAs 523.1.6.1. Time course 523.1.6.2. Dose-pressor response 543.1.6.3. Effects of L-Arg and D-Arg on the pressor responses to NSAAs 573.1 .6.4. Effects of pithing and of pharmacological antagonists on thepressor and HR responses to NSAAs 58vi3.1 .6.5. Effects of L-NNA on plasma catecholamines 613.2. Discussion 633.2.1. Effects of NSAAs on endothelium-dependent relaxation in vitro 633.2.2. Effects of NSAAs on endothelium-dependent vasodilatationin vivo 643.2.3. Stereospecificity of NSAAs in vitro and in vivo 683.2.4. Pharmacodynamic analyses of the vascular actions of NSAAs 723.2.5. Mechanisms of the pressor responses to NSAA5 733.2.5.1. Pressor responses to NSAA5 are due to vasoconstriction 733.2.5.2. Antagonism of L-Arg on the pressor responses to NSAAs 743.2.5.3. Effects of impairment of the central, ganglionic, sympathetic,angiotensin or prostanoid system on the pressor responses toNSAAs 753.2.6. Mechanisms of the HR responses to NSAAs 773.3. Summary 794. Results and discussion II 814.1. Results 814.1.1. Effects of DPI on endothelium-dependent relaxation in vitro 814.1.1.1. Concentration-responses 814.1.1.2. Mechanisms 814.1.1.3. Time courses and reversibility 824.1.2. Effects of DPI on the depressor responses to ACh and SNPin vivo 834.1.3. Pressor and tachycardiac responses to DPI 844.1.4. Mechanism of the pressor and tachycardiac responses to DPI 854.1.4.1. Effects of pharmacological antagonists on the pressor andtachycardiac responses to DPI 854.1.4.2. Effects of TTX, pithing or spinal cord (T1) transection on thepressor and tachycardiac responses to DPI 864.1.4.3. Effects of DPI on plasma catecholamines in intact, pithed orreserpinized rats. 864.1.3. Inhibitory effect of halothane on the pressor response to DPI 884.2. Discussion 894.2.1. Inhibitory effects of DPI on endothelium-dependentvasodilatations in vitro and in vivo, and their mechanisms 894.2.2. Pressor and tachycardiac responses to DPI, and theirmechanisms in vivo 924.2.3. Mechanisms of the inhibitory effect of halothane on the pressorresponse to DPI 964.3. Summary 975. General discussion and conclusions 99vii5.1. General discussion 995.2. Conclusions 1026. Bibliography 1047. Appendices 129Paper I. Possible dependence of pressor and heart rate effectsof NGnitroLarginine on autonomic nerve activity. 1 29Paper II. Pressor effects of L and D enantiomers of NGnitroLarginine in conscious rats are antagonized by L- but notD-arginine. 135Paper Ill. In vitro and ex vivo inhibitory effects of L and Denantiomers of NGnitroarginine on endotheliumdependent relaxation of rat aorta. 141Paper IV. Functional integrity of the central and sympatheticnervous systems is a prerequisite for pressor andtachycardiac effects of diphenyleneiodonium, a novelinhibitor of nitric oxide synthase. 1 50Paper V. Halothane inhibits pressor effect of diphenyleneiodonium. 161Paper VI. Selective inhibition of pressor and haenodynamic effectsof NGnitroLarginine by halothane. 168Paper VII. Inhibitory actions of diphenyleneiodonium on endotheliumdependent vasodilatations in vitro and in vivo. 177Paper VIII. Vascular pharmacodynamics of NGnitroLargininemethyl ester in vitro and in vivo. 1 85viiiAbbreviationsACh acetyicholineADMA dimethylarginineArg arginineC concentrationcAMP cyclic adenosine 3’,5’-monophosphatecGMP cyclic guanosine 3’,5’-monophosphateCO cardiac outputcpm counts per minuteD doseDPI diphenyleneiodoniumDR dose ratioEC50 half-effective concentrationED50 half-effective doseEDHF endothelium-derived hyperpolarizing factorEDRF endothelium-derived relaxing factorEGTA ethylene glycol-bis(B-amino-ethyl ether) N,N, , N’-tetraacetic acidEmax maximal effectEmin minimal effecteNOS endothelium nitric oxide synthaseFMN flavin mononucleotideGTP guanosine 5’-triphosphateHR heart rateIC50 half-inhibitory concentrationiNOS inducible nitric oxide synthaseL-NAA NGaminoLarginineLPS lipopolysaccharideixMAP mean arterial pressureMCFP mean circulatory filling pressuren Hill coefficientNAME NGnitroarginine methyl esterNANC nonadrenergic, noncholinergicNlO N-iminoethyl-ornithineNMDA N-methyl-D-aspartateNMMA NGmonomethylarginineNNA NGnitroargininenNOS neuron nitric oxide synthaseNO nitric oxideNOS nitric oxide synthaseNSAAs NGsubstituted arginine analogues6-OH-DA 6-hydroxydopaminePHE phenylephrineSNP sodium nitroprussideSOD superoxide dismutaseTPR total peripheral resistanceTTX tetrodotoxinxList of tablesTable Page1 Baseline values of MAP in conscious rats. 462 Values of n, ED50 and Emax of the dose-pressor responses ofL-NAME and D-NAME. 56xiList of figuresFigure Page1 Inhibitory effects of L-NAME and D-NAME on ACh-inducedrelaxation. 382 Analyses of the inhibition by L-NAME and D-NAME on AChinduced relaxation. 393 Effects of L-NAME and D-NAME on SNP-induced relaxation. 404 Time courses of the inhibition by L-NAME and D-NAME of AChinduced relaxation. 425 Effects of L-Arg and D-Arg on the inhibition by L-NAME andD-NAME of ACh-induced relaxation. 446 Effects of infusions of ACh and SNP on MAP. 487 Effects of bolus injections of ACh on the magnitude and durationof MAP. 498 Effects of bolus injections of SNP on the magnitude and durationof MAP. 519 Time courses of the MAP response to L-NAME and D-NAME. 5310 Dose-pressor responses of L-NAME and D-NAME. 5511 Effect of L-Arg on the MAP response to D-NAME. 591 2 Effect of indomethacin on the MAP response to L-NAME. 62xiiAcknowledgementsI thank all members of this department who have helped complete thisdissertation. I would like to acknowledge my gratitude especially to:Dr. Catherine C.Y. Pang, my research supervisor, for sharing her knowledgeand experience, for her friendship and encouragement, constructive criticism, aswell as wise guidance throughout the study, and for her generous financialsupport during the first stage of the Ph.D. program from her MRC grant.Dr. David M.J. Quastel for his invitation to work as a post-doctoral researchfellow during my first year in Canada, his help in analyzing dose-response curvesby the use of his computer program, and his valuable advice in my work.Drs. Morley C. Sutter, Michael J.A. Walker, Jack Diamond, James M. Wright,Richard A. Wall, Sultan Karim and Reza Tabrizchi for their valuable advice andcritical evaluations of my work.Ms. Su Lin Lim, Miss Christina C.l. Poon, Dr. Ting Zhou, Dr. Aly Abdelrahman,Miss Lorraine C.Y. Yau and Mr. Ken S. Poon for their technical assistance.My wife, Ji-Ping Huang, and my mother for their understanding,encouragement and help throughout my study.The Medical Research Council of Canada for its generous financial supportduring the last stage of the Ph.D. program by awarding me a Post-doctoralResearch Fellowship; the Canadian Pharmacological Society and CanadianHypertension Society for their encouragement by awarding me the Annual (1992)Graduate Student Presentation Award and Annual (1 993) Trainee Award,respectively; the University of British Columbia, American Pharmacological Societyand British Columbia Government for their financial support for me to attendscientific conferences.xliiPrefaceThis dissertation was based on the results from the following list of mypublications which were referred to by Roman numerals in the appendices, andsome unpublished data which were presented in the contents. Permission toreproduce these publications were kindly provided by the individual copyrightholders.I. Wang, Y.-X. and Pang, C.C.Y.: Possible dependence of pressor and heart rateeffects of NGnitroLarginine on autonomic nerve activity. Br. J.Pharmacol. 103: 2004-2008, 1991.II. Wang, Y.-X., Zhou, T. and Pang, C.C.Y.: Pressor effects of L and Denantiomers of NGnitroLarginine in conscious rats are antagonized by Lbut not D-arginine. Eur. J. Pharmacol. 200: 77-81, 1991.Ill. Wang, Y.-X., Poon, C.l. and Pang, C.C.Y.: In vitro and ex vivo inhibitoryeffects of L and D enantiomers of NGnitroarginine on endotheliumdependent relaxation of rat aorta. J. Pharmacol. Exp. Ther. 265: 112-119,1993.IV. Wang, Y.-X. and Pang, C.C.Y.: Functional integrity of the central andsympathetic nervous systems is a prerequisite for pressor and tachycardiceffects of diphenyleneiodonium, a novel inhibitor of nitric oxide synthase. J.Pharmacol. Exp. Ther. 265: 263-272, 1993.V. Wang, Y.-X. and Pang, C.C.Y.: Halothane inhibits pressor effect ofdiphenyleneiodonium. Br. J. Pharmacol. 109: 11 86-1 1 91, 1 993.VI. Wang, Y.-X., Abdelrahman, A. and Pang, C.C.Y.: Selective inhibition ofpressor and haemodynamic effects of NGnitroLarginine by halothane. J.Cardiovasc. Pharmacol. 22: 571-578, 1993.xivVII. Wang, Y.-X., Poon, C.l., Poon, K.S. and Pang, C.C.Y.: Inhibitory actions ofdiphenyleneiodonium on endothelium-dependent vasodilatations in vitro andin vivo. Br. J. Pharmacol. 110: 1232-1238, 1993.VIII. Wang, Y.-X., Poon, C.l. and Pang, C.C.Y.: Vascular pharmacodynamics ofNGnitroLarginine methyl ester in vitro and in vivo. J. Pharmacol. Exp.Ther. 267: 1091-1099, 1993.The following publications related to this study were also referred to in theintroduction and discussion sections, and were listed in the bibliography.Wang, Y.-X. and Pang C.C.Y.: Pressor effect of NGnitroLarginine inpentobarbital-anesthetized rats. Life Sci. 47: 221 7-2224, 1 990.Wang, Y.-X. and Pang, C.C.Y.: Effects of L and D enantiomers of NGnitroarginine on blood pressure in pentobarbital-anesthetized rats. Pharmacology(Life Sci. Adv.) 9: 723-728, 1 990.Wang, Y.-X., Zhou, T., Chua, T.C. and Pang, C.C.Y.: Effects of inhalation andintravenous anaesthetic agents on pressor response to NGnitroLarginine.Eur. J. Pharmacol. 198: 183-1 88, 1991.Abdelrahman, A., Wang, Y.-X., Chang, S.D. and Pang, C.C.Y.: Mechanism of thevasodilator action of caicitonin gene-related peptide in conscious rats. Br. J.Pharmacol. 106: 45-48, 1992.Abdelrahman, A., Wang, Y.-X., Pang, C.C.Y.: Effects of anaesthetic agents onpressor response to f-bIockers in the rat. J. Pharm. Pharmacol. 44: 34-38,1992.Wang, Y.-X., Zhou, T. and Pang, C.C.Y.: Depressor response to acetylcholine andpressor response to NGnitroLarginine in conscious Wistar, WKY and SHrats. In: Moncada, S., Marietta, M.A., Hibbs, J.B. Jr. and Higgs, E.A. (eds).xvThe Biology of Nitric Oxide. 1. Physiological and Clinical Aspects. London:Portland Press, pp 174-176, 1992.Wang, Y.-X., Zhou, T. and Pang, C.C.Y.: A comparison of the inhibitory effects ofsodium nitroprusside, pinacidil and nifedipine on pressor response to NGnitro-L-arginine. Br. J. Pharmacol. 108: 398-404, 1993.Wang, Y.-X. and Pang, C.C.Y.: Suppression by ethanol of pressor responsecaused by the inhibition of nitric oxide synthesis. Eur. J. Pharmacol. 233:275-278, 1993.Sutter, M.C. and Wang, Y.-X.: Recent cardiovascular drugs from Chinesemedicinal plants. Cardiovasc. Res. 27: 1891-1 901, 1993.Wang, Y.-X., Poon, K.S., Randall, D.J. and Pang, C.C.Y.: Endothelium-derivednitric oxide partially mediates salbutamol-induced vasodilatations. Eur. J.Pharmacol. 250: 335-340, 1 993.Wang, Y.-X. and Pang, C.C.Y.: Effects of adrenalectomy and chemicalsympathectomy on the pressor and tachycardic responses todiphenyleneiodonium. J. Pharmacol. Exp. Ther. in press, 1994.Wang, Y.-X. and Pang, C.C.Y.: NGnitroLarginine contracts vascular smoothmuscle via a non-nitric oxide synthesis mechanism. J. Cardiovasc.Pharmacol. in press, 1 994.I1. Introduction1.1. EDRF and NO1.1.1. Endothelium-dependent relaxation and EDRFThe investigation of the actions of acetylcholine (ACh) in the vasculaturewas initiated over 80 years ago. It was found that while ACh always causeddepressor responses in whole animals (Hunt, 1 91 5), the. actions of ACh in isolatedvascular preparations were rather inconsistent. Dale (1914) observed that AChdilated rabbit ear arteries whereas Hunt (1 91 5) observed that ACh had either noeffect or vasoconstriction in a similar preparation. This paradox was reported bymany authors in many vascular preparations (see review by Furchgott, 1955). Thecontroversial results were intensely studied during the subsequent more than half-century, and some hypotheses were put forward, such as excess ACh hypothesis(Burn and Robinson, 1951), adrenaline release hypothesis (Kottegoda, 1953) andtwo receptor type hypothesis (see review by Furchgott, 1955). The cause for thediffering responses, however, was not determined until the last decade.In 1 980, Furchgott and Zawadzki for the first time found that if care wastaken to preserve the intimal surface to prepare the isolated aortic rings or strips,the preparations were always relaxed by ACh. They therefore concluded that AChstimulated muscarinic receptors on vascular endothelial cells to release anonprostanoid substance which then caused relaxation of smooth muscles(Furchgott and Zawadzki, 1980). The substance was later termed endotheliumderived relaxing factor (EDRF) (Cherry et al., 1982). Since then, many othervasodilator agents were also found to require the presence of endothelial cells toproduce partial or complete relaxation of arteries, veins and microvessels. Amongthese are the calcium ionophore A 231 87, adenosine 5’-triphosphate (ATP) and2adenosine 5’-diphosphate (ADP), substance P, arachidonic acid, bradykinin (forpig, dog and human but not rabbit or cat arteries), histamine (for rat aorta andguinea pig pulmonary artery), thrombin (for canine arteries) and calcitonin gene-related peptide (for rat aorta) (see review by Furchgott, 1983; see review byMoncada et a!., 1988; see review by Lüscher and Vanhoutte, 1990a). Otherstimuli, such as hypoxia, sheer stress and electrical stimulation, also induceendothelium-dependent relaxation of vasculatures in vitro (see review by Moncadaeta!., 1988; see review by Lüscher and Vanhoutte, 1990a). Certain differencesamong species and vasculatures in the dependence of the endothelium forvasodilatation have also been identified (see review by Furchgott, 1983; seereview by Moncada et aL, 1 988; see review by Lüscher and Vanhoutte, 1 990b).There are also many compounds which cause relaxation of isolated arterialpreparations in an endothelium-independent manner. Among these are thenitrovasodilators which include sodium nitroprusside (SNP) and nitroglycerin,bovine retractor-penis inhibitor, and prostacyclin (see review by Moncada et aL,1988; see review by Lüscher and Vanhoutte, 1990a). B-Adrenoceptor agonistsare widely believed to cause endothelium-independent vasodilatation via theproduction of cyclic adenosine 3’,5’-monophosphate (cAMP) (see review byFurchgott and Vanhoutte, 1 989; see review by Lüscher and Vanhoutte, 1990; seereview by Moncada etah, 1991). However, I2-adrenoceptors have been detectedin cultured or naive (Molenaar et a!., 1988) human and bovine endothelial cells(Howell et aL, 1988; Ahmad et aL, 1990). Moreover, mechanical removal of theendothelium partially prevented salbutamol (I2-adrenoceptor agonist)-inducedrelaxation of preconstricted rat aortae, suggesting that endothelial cells contributeto2-adrenoceptor-induced vascular relaxation (Wang eta!., 1993a).The role of the endothelium in the actions of vasodilators from traditionalChinese medicinal plants has been studied, It was reported that the relaxanteffect of magnolol was dependent on the endothelium. Removal of the3endothelium, on the other hand, did not alter the vasorelaxant effects ofdicentrine, berberine, tetramethylpyrazine and norathyriol; these compounds mayhave antagonistic activities on the calcium channels or intracellular calcium release(see review by Sutter and Wang, 1993).EDRF was reported to be a labile humoral factor with a half-life of less than5 seconds in physiological preparations (Palmer et al., 1987). It has also beensuggested that EDRF is continuously released (Martin eta!., 1985). In addition torelaxing vascular preparations, EDRF also inhibits platelet aggregation (Azuma eta!., 1986; Radomski et a!., 1987a,c) and platelet adhesion (Radomski et a!.,1987d). A rise in the level of cyclic guanosine 3’,5’-monophosphate (cGMP) insmooth muscles or platelets, which is a consequence of the stimulation of solubleguanylyl cyclase (Ignarro et aL, 1986; Russe, 1987), was found to accompanyendothelium-dependent relaxation (Holzmann, 1982; Rapoport and Murad, 1983;Diamond and Chu, 1983; Ignarro et al., 1984) and the inhibition of plateletaggregation (Rapoport and Murad, 1983). Endothelium-dependent relaxation andthe inhibition of platelet aggregation are potentiated by M & B 22948 (zaprinast)and MY 5545, two selective inhibitors of cGMP phosphodiesterase (Kukovetz eta!., 1982; Martin eta!., 1986a; Radomski eta!., 1987a,c,d) and are inhibited bythe soluble guanylyl cyclase inhibitor methylene blue (Martin et aL, 1985).Moreover, endothelium-dependent relaxation, inhibition of platelet aggregation andrise in cGMP are all inhibited by haemoglobin and other reduced haemoproteinswhich bind to EDRF (Ignarro et a!., 1984; Martin et a!., 1986b; Radmoski et a!.,1 987a).1.1.2. The pharmacology of NONitric oxide (NO) is a very small lipophilic molecule that rapidly diffusesthrough biological membranes and reaches intracellular compartments of nearby4cells with diverse functions. Interestingly, much like oxygen, NO is a gas that issparingly soluble in aqueous solution and it functions biologically as a molecule insolution (see review by Ignarro, 1990). However, NO is unstable, with anultrashort half-life of probably less than 5 seconds in oxygen-containing solution orbiological tissues (Palmer et a!., 1987).Before the discovery of EDRF, authentic NO was found to stimulate guanylylcyclase (Schultz eta!., 1977; Katsuki eta!., 1977; Craven and DeRubertis, 1978;Gruetter et aL, 1979) and to relax vascular smooth muscles (Gruetter et aL,1979). Later, NO was reported to prevent platelet aggregation (Mellion et aL,1981, 1983; Azuma eta!., 1986; Radmoski eta!., 1987a) and adhesion (Radmoskiet aL, 1987d). The inhibitory effects of NO on vascular smooth muscle tone andplatelet adhesion/aggregation were inhibited by oxyhaemoglobin (Martin et a!.,1985; Radomski et aL, 1987a,b). The interaction between NO and haemoproteinof the cytosolic guanylyl cyclase results in the formation of the labile nitrosylhaeme-enzyme complex, the activated state of guanylyl cyclase, which markedlyincreases the velocity of conversion of guanosine 5’-triphosphate (GTP) to cGMP,leading to smooth muscle relaxation and the inhibition of plateletaggregation/adhesion (see review by Ignarro, 1989; 1990). In contrast to atrialnatriuretic peptide which stimulates particulate guanylyl cyclase (Winquist et a!.,1984), NO activates the soluble form of the enzyme (Gruetter et a!., 1979).Therefore, the NO-induced activation of soluble guanylyl cyclase, and subsequentrelaxation as well as inhibition of platelet aggregation/adhesion are inhibited bymethylene blue (Gruetter et a!., 1979, 1981; Mellion et a!., 1981, 1983; Martin eta!., 1985; Radomski eta!., 1987a).Organic nitrate and nitrite esters as well as nitroso compounds react withfree thiols to generate labile S-nitrosothiols which liberates NO (Ignarro et a!.,1981). The NO donors activate soluble guanylyl cyclase, elevate tissue levels ofcGMP, relax vascular smooth muscles, decrease systemic blood pressure and5inhibit platelet aggregation (Gruetter et aL, 1979; Kukovetz et aL, 1979; seereview by Ignarro and Kadowitz, 1985; Ignarro et aL, 1987). It has beenproposed that S-nitrosothiols are active intermediates which mediate thevasorelaxant effects of the nitrovasodilators; however, the final common mediatorthat stimulates cGMP and causes relaxation is NO (see review by Ignarro, 1992).1.1.3. EDRF is NO or a labile nitroso precursorSoon after the discovery of EDRF, it was postulated that EDRF might be alabile free radical formed as an intermediate from arachidonic acid, via thelipooxygenase pathway (Furchgott and Zawadzki, 1980; Cherry eta!., 1982; seereview by Furchgott, 1 983; Singer and Peach, 1 983; Forstermann and Neufang,1984), or an unknown compound with a carbonyl group near its active site(Griffith eta!., 1984), or a product of the cytochrome P-450 enzyme system (Pintoet a!., 1986; Macdonald et a!., 1986). However, these hypotheses were soondisputed (see review by Moncada eta!., 1991).In 1 986, Furchgott and Ignarro et aL independently postulated that EDRFwas NO or a closely-related derivative of NO. This was based on the similarpharmacological profiles between NO and EDRF as discussed above, i.e., bothcompounds are unstable and are inhibited by haemoglobin and stabilized bysuperoxide dismutase (SOD). Also, both agents exert biological actions bystimulating soluble guanylyl cyclase (Furchgott, 1988; Ignarro et a!., 1988).Within a year, several laboratories provided more detailed chemical evidence thatEDRF is NO. (1) Both EDRF and NO are similar in their effects on vascularrelaxation (Palmer et a!., 1987; Hutchinson et aL, 1987) as well as inhibition ofplatelet aggregation and adhesion (Radomski et a!., 1 987a). (2) Both agents havethe same chemical stability under different conditions (Palmer et a!., 1987). (3)Endothelial cells in culture release NO in amounts sufficient to account for the6activities of EDRF on vascular strips (Palmer eta!., 1987) and platelet aggregation(Radomski eta!., 1987c) as well as adhesion (Radomski eta!., 1987d). lgnarro eta!. (1 987a,b) also showed that NO or a labile nitroso species was released fromthe bovine pulmonary artery. Subsequently, the release of NO from endothelialcells was confirmed by using a spectrophotometric assay (Keim et aL, 1988). Itwas also demonstrated that the amount of NO released from the perfused aorta(Chen et a!., 1 989) or the isolated perfused heart (Amezcua et a!., 1 988; KeIm andSchrader, 1 988) was sufficient to cause the vasodilatations observed.Although NO has been identified as an EDRF, questions have arisen on thespecificity of chemical procedures employed to draw such a conclusion. It is clearthat the chemical techniques which involve chemiluminescence, diazotization, andnitrosation of haemoglobin are not selective for NO, as labile nitroso compoundsthat spontaneously decompose to liberate NO are also readily detected by all thesechemical procedures. Therefore, the chemical evidence that EDRF is identical toNO is not so definite. It has been reported that the activity and potency of EDRFmore closely resemble those of a nitrosylated compound, S-nitrocysteine, than tothose of NO (Myers et a!., 1990). EDRF is also postulated to be either a labilenitroso precursor which releases NO at the smooth muscle cell, or a mixture ofnitroso compounds plus NO (see review by Ignarro, 1992). However, it isgenerally agreed that the existence of such a precursor does not detract from thefact that the biological effects of EDRF are mediated ultimately by NO (Myers eta!., 1990; see review by Moncada eta!., 19991; see review by Ignarro, 1992).Palmer et a!. (1988a) reported that endothelial cells incubated in an Larginine (L-Arg)-free medium for 24 h have reduced capacity to release NO. Thiscapacity was restored by the addition of L-Arg to the medium prior to stimulationwith bradykinin. The effect of L-Arg was enantiomer specific as it was not sharedwith D-Arg. These findings suggest that NO formation is dependent on theavailability of free L-Arg. Furthermore, mass spectrometric analysis demonstrated7the formation of 1 5N0 and L-citrulline from L-Arg, which was previously labeledwith 1 5N at the terminal guanido nitrogen atoms. These results provideconclusive evidence that NO is derived from the terminal guanido nitrogen atom(s)of L-Arg (Palmer et aL, 1988a). Moreover, NGmonomethyILarginine (L-NMMA),an analogue of L-Arg, inhibited endothelial NO synthesis (Palmer et a!., 1988b).All these findings, which are supported by other groups (Schmidt et aL, 1988a,b;Sakuma eta!., 1988; MarIetta et aL, 1988) are very exciting, though not entirelysurprising. In fact, one year earlier Hibbs et a!. (1 987a,b) have reported that L-Argis required for the expression of the activated macrophage cytotoxic effectormechanism that inhibits mitochondrial respiration. They also demonstrated thatmurine cytotoxic activated macrophages synthesized L-citrulline and nitrite in thepresence of L-Arg but not D-Arg, and that L-citrulline and nitrite biosynthesis bycytotoxic activated macrophages was inhibited by L-NMMA, which also inhibitedthe cytotoxic effector mechanism. Furthermore, the imino nitrogen was removedfrom the guanido group of L-Arg. lyengar et al. (1987) also reported that NO2and N03 produced by cultured macrophages, were exclusively derived from theterminal guanidino nitrogens of L-Arg.It is now generally believed that the NO synthetic pathway involves theinteraction of L-Arg with °2 yielding NO and L-citrulline, with NGhydroxyLarginine as an intermediate (Stuehr eta!., 1991c). NO is then rapidly broken downin biological tissues into NO2 and N03, both of which have only weak biologicalactivities.1.2. NOSThe synthesis of NO was demonstrated in endothelial homogenates by theformation of citrulline from L-Arg via a mechanism which was dependent onreduced nicotinamide adenine dinucleotide (NADPH) and inhibited by L-NMMA8(Palmer and Moncada, 1989). The enzyme which synthesizes NO wassubsequently named NO synthase (NOS) and found to be Ca2+dependent andlocated in the soluble fraction of endothelial homogenates (Moncada and Palmer,1990). Through several years of study, NOSs are now well-known to be presentin many organs, tissues and cells, such as endothelial cells (Palmer et a!.,1988a,b; Rees eta!., 1990; Kilbourn and Belloni, 1990), platelets (Radomski eta!.,199Oa,b,d), brain neurons (Garthwaite eta!., 1988; Knowles eta!., 1989, 1990a;Bredt and Snyder, 1989, 1990; Bredt eta!., 1990; Schmidt et a!., 1989, 1992),macrophages and neutrophils (Marietta et a!., 1988; Hibbs, et aL, 1 988; Stuehr eta!., 1989a,b; McCall eta!., 1991a,b), Kupffer cells and hepatocytes (Curran eta!.,1990), smooth muscle cells (Kilbourn eta!., 1992), nonadrenergic, noncholinergic(NANC) neurons (Bredt eta!., 1990, 1991; Young et a!., 1992) and the cortex andthe medulla of the adrenal gland (Palacios eta!., 1989).Therefore, the interaction between NO and the haeme group of guanylylcyclase represents a novel and widespread signal transduction process that linksextracellular stimuli to the biosynthesis of the second messenger cGMP in adjacentcells. Due to its wide distribution, NO modulates not only the functions ofvasculatures and platelets, but also macrophage and neutrophil cytotoxicity (Hibbseta!., 1987a,b), long-term potentiation in the hippocampus (Böhme eta!., 1991),long-term synaptic depression in the cerebellum (Shibuki and Okada 1991) andnociceptive activity in the brain (Moore et al., 1991), as well as NANC relaxationof smooth muscles such as guinea pig isolated tracheal smooth muscle and ratanococcygeus (Tucker eta!., 1990; Hibbs and Gibson, 1990).Whereas many oxidative enzymes use a single electron donor, the oxidationof Arg to NO by NOS involves multiple oxidative cofactors with associated bindingsites (Bredt eta!., 1991; Xie eta!., 1992; Lowenstein eta!., 1992; Lyons et aL,1992; Lamas et a!., 1992; Marsden et aL, 1992; see review by Dinerman eta!.,1993). These cofactors are NADPH (Mayer et a!., 1989; Stuehr et a!., 1989b,91990, 1991a,b; Bredt et a!., 1991; see review by McCall and ValIance, 1992;Marsden et a!., 1992), flavin adenine dinucleotide (FAD) (Stuehr et a!., 1989b,1990, 1991a,b; Hevel. eta!., 1991; Yui eta!., 1991; Mayer eta!., 1991; Bredt eta!., 1991, 1992; Hiki et a!., 1992; White and Marietta, 1992; Lowenstein et a!.,1992; Marsden et a!., 1992), flavin mononucleotide (FMN) (Bredt et a!., 1991;Marsden eta!., 1992), iron-protoporphyrin IX haeme (McMiiIan eta!., 1992; Whiteand MarIetta, 1992), tetahydrobiopterin (Tayer and Marietta, 1989; Kwon et a!.,1989) and caicium/caimodulin (Knowles eta!., 1989; Bredt eta!., 1991; Marsdeneta!., 1992).NOS has at least two isoforms in the same or different organs, tissues orcells.1 .2.1. Constitutive NOSConstitutive NOS is present in brain neurons, platelets, endothelial cells aswell as NANC neurons. The enzyme isCa2+/calmoduiindependent (Bredt andSnyder, 1990; Bredt et a!., 1991, 1992) and not affected by L-canavanine, aguanidinooxy structural analogue of L-Arg (Palmer and Moncada, 1 989; Mayer eta!., 1989). However, cloning studies of human endothelial NOS predicted theenzyme to consist of 1 203 amino acids which were identical to those of thebovine endothelial NOS by 95%, but which shared only 60% identity with those ofthe brain NOS isoform (Bredt eta!., 1991; Marsden eta!., 1992). Moreover, thesequence of the endothelial NOS contains a site at the N-terminus formyristoylation that probably accounts for the association of endothelium NOS withmembranes, which is absent in the macrophage and cerebeilar NOS isoforms(Lamas eta!., 1992). Therefore, although brain and endotheiiai NOS share manycommon characteristics, they can be divided into two isoforms. i.e., e(endothelium) NOS and n (neuron) NOS.101 .2.2. Inducible NOSInducible NOS differs from the constitutive form in that it is not detectablein the “rest condition”, i.e., prior to activation by an inducing agent such aslipopolysaccharide (LPS) alone, or combination with interferon-y (Stuehr andMarietta, 1 985, 1 987a,b), and that it requires protein synthesis for its expression(Marietta et al., 1988). It takes hours before N02 and N03 synthesis can bedetected. The synthesis of these products then continues until either no moresubstrate is available or the cell dies (Stuehr and Marietta, 1987a,b). Therefore,the enzyme is referred to as i (inducible) NOS. iNOS is Ca2+independent andinhibited by L-canavanine (iyengar et a!., 1987; McCall et a!., 1989). Theinduction of iNOS is inhibited by glucocorticoids (Rees et oh, 1990a; Knowles etoh, 1990b; McCall et oh, 1991b). Its clone and expression have also beenreported (Xie et oh, 1992; Lowenstein et oh, 1992; Lyons et oh, 1992). iNOS ispresent in vascular smooth muscle cells, neutrophils, macrophages, Kupffer cells,hepatocytes, endothelial cells and cardiac myocytes.1 .3. NOS inhibitorsThere are at least 2 classes of NOS inhibitors, of which the NGsubstitutedarginine analogues (NSAAs) are more studied. Endogenous NOS inhibitors havealso been reported.1.3.1. NSAAs1.3.1.1. Structures11Both isoforms of NOS are reported to be inhibited in many tissues in anenantiomerically specific manner by L-NMMA (Palmer eta!., 1988b; Rees eta!.,1989a, 1990b), NGnitroLarginine (L-NNA) (lshii eta!., 1990a,b; Mülsch andBusse, 1990), NGnitroarginine methyl ester (L-NAME) (Rees eta!., 1990b), Niminoethyl-L-ornithine (L-NlO) (Rees et a!., 1 990b), and NGaminoLarginine (LNAA) (Gross et a!., 1990; Kilbourn et a!., 1992). The activities or potencies ofNSAAs on different isoforms of NOS may vary (see review by Moncada et a!.,1991). For example, L-canavanine inhibits iNOS (lyengar eta!., 1987; McCall eta!., 1989) but not eNOS (Palmer and Moncada, 1989; Mayer et a!., 1989),although it at high concentration was shown to inhibit endothelium-dependentrelaxation (Schmidt et a!., 1 988a, 1990), but probably by another mechanism (seereview by Moncada et a!., 1991). The chemical structures of NSAAs frequentlyused were shown in Fig. 1 in Appendix IV.It was also reported that NG,NGdimethylarginine (asymmetricaldimethylarginine, ADMA) inhibited NO synthesis in J744 murine macrophages,potentiated the contraction and inhibited the relaxation of preconstricted ratendothelium-intact aorta (ValIance et a!., 1992). Moreover, aminoguanidine isstructurally similar to L-Arg and NSAAs in that these compounds contain twochemically equivalent guanidino nitrogen groups. Aminoguanidine was shown tobe equipotent to L-NMMA as an inhibitor of the cytokine-induced iNOS but to be10-100 fold less potent as an inhibitor of eNOS (Corbett et a!., 1 992; Misko et a!.,1993). These results may suggest that the guanidino nitrogen group is essentialfor NSAAs to inhibit NO biosynthesis, although NSAAs and aminoguanidine inhibitdifferent isoforms of NOS with different potencies.1.3.1.2. Competitive inhibition12Since NSAAs are analogues of L-Arg and since L-Arg blocks the actions ofNSAAs, it is reasonable to assume that interactions between NSAAs and L-Arg arecompetitive. It was reported that L-NMMA, L-NNA and L-NlO inhibited brain NOSin competitive manner (Knowles et a!., 1990a), however, detailedpharmacodynamic study to elucidate the type of antagonism between L-Arg andNSAAs in the vasculature has not been performed.Interestingly, L-NMMA also inhibited endothelium-dependent relaxation inaortic rings but not in pulmonary arterial rings, and L-NMMA enhanced, rather thanreduced, NO synthesis in pulmonary artery and aortic rings. In contrast, L-NNAdid not stimulate NO synthesis and inhibited A 231 87-induced relaxation invascular rings. Therefore, it has been concluded that L-NMMA is a partial agonistfor NO synthesis (Archer and Hampl, 1992). This conclusion is confirmed byMartin et aI.’s findings (1993) which showed that L-NMMA did not block NANCrelaxation in bovine retractor penis but inhibited the blockade induced by L-NNA orL-NAME with greater potency than did the substrate L-Arg.1.3.1.3. ReversibilityIt has been reported that NSAAs inhibit iNOS for a long period of time afterwashouts (Rees et aL, 1 989a; Mülsch et aL, 1990). L-NIO has been identified asan irreversible inhibitor for iNOS since post-treatment with L-Arg did not reversethe inhibition by L-NIO (McCall eta!., 1991a). In contrast, L-NMMA, L-NNA and LNAME are reversible inhibitors of iNOS since post-treatment with L-Arg is aseffective as pretreatment to reverse their inhibitory effects on macrophage NOsynthesis (McCall eta!., 1991a).1.3.1.4. Stereospecificity13NOS has been extensively claimed to be stereospecific, since L- but not Denantiomers of NMMA (Palmer et a!., 1988b; Rees et a!., 1989a, 1990b), NNA(lshii eta!., 1990a,b; Mülsch and Busse, 1990; Rees eta!., 1990b), NAME (Reeseta!., 1990b), NlO (Rees eta!., 1990b) have been found to inhibit NO synthesis.1.3.1.5. SpecificityNSAAs are frequently used as tool drugs to elucidate the biological effectsof the inhibition of NO biosynthesis, on the basis of the assumption that NSAAsare specific inhibitors of NOS. However, it has been reported that L-NAME, butnot L-NNA or L-NMMA, has weak antimuscarinic effects (Buxton et a!., 1993).Therefore, caution must be exercised in the interpretation of results from studiesusing NSAA5.1 .3.2. DPI and other iodonium compoundsA group of iodonium compounds have been reported to be another class ofNOS inhibitors in the macrophage (Stuehr eta!., 1991b; Kwon eta!., 1991; Kelleret a!., 1992). These compounds include diphenyleneiodonium (DPI),iodoniumdiphenyl and di-2-thienyliodonium, all of which have similar chemicalstructures and are distinct from those of NSAAs. The chemical structure of DPIwas shown in Fig. 1 in Appendix IV.DPI was initially found to be a potent hypoglycaemic agent (Stewart andHanly, 1969; Gatley and Martin, 1979) which, by inhibiting gluconeogenesis fromlactate and aspartate, suppressed the oxidation of NADH-linked substances(Holland et a!., 1973). It was later shown that DPI, iodoniumdiphenyl and di-2-thienyliodonium suppressed the activities of neutrophil and macrophage NADPHdependent oxidase (Cross and Jones, 1986; Hancock and Jones, 1987; Ellis et a!.,141988, 1989). All these actions of DPI are due to its inhibition of a flavoproteinwhich is coupled to NADPH-dependent enzymes (Holland et al., 1973; Hancockand Jones, 1987; Ellis etaL, 1989; Stuehr etaL, 1991b; O’Donnell eta!., 1993).Therefore, it is not surprising that DPI and its analogues inhibit NO biosynthesis, asNADPH and FAD are cofactors of NOS (Bredt et a!., 1991; Xie et aL, 1992;Lowenstein et a!., 1992; Lyons et a!., 1992; Lamas et aL, 1992; see review byDinerman eta!., 1993). However, the inhibitory effects of DPI and its analogueson eNOS and nNOS have not been reported.1 .3.3. Endogenous NOS inhibitorsIt has been known for a long time that methylated Arg such as L-NMMA andADMA are naturally occurring agents (Kakimoto and Akazawa, 1970; Nakajima eta!., 1970; Park et a!., 1988). Valiance et a!. (1992) confirmed that ADMA wasdetected in human plasma and urine, where more than 1 0 mg is excreted in theurine over 24 hours. Moreover, in patients with chronic renal failure with little orno urine output, circulating concentration of the inhibitor rose to a level sufficientto inhibit NO synthesis de vivo. These investigators suggested that theaccumulation of endogenous ADMA may lead to an impaired NO synthesis which,in turn, may lead to the development of hypertension and immune dysfunction,conditions often associated with chronic renal failure.It was also reported that an inhibitor of endothelium-dependent relaxationexisted in the rabbit brain although its chemical structure was not known butmight be a large peptide or a protein (Moore eta!., 1990).1 .4. Actions of NOS inhibitors on the vasculature1 .4.1. Inhibition of endothelium-dependent vasodilatation15In addition to eliciting contraction and potentiating the contractile effects ofvasoconstrictor agents, NSAAs, e.g., L-NMMA (Palmer et aL, 1988b; Sakuma eta!., 1988; Rees eta!., 1989a, 1990b; Crawley eta!., 1990), L-NNA (Moore eta!.,1 990; Mülsch and Busse, 1 990; Kobayashi and Hattori, 1 990), L-NAME (Rees eta!., 1990b), L-Nl0 (Rees eta!., 1990b) and L-NAA (Fukuto eta!., 1990; Vargas eta!,, 1991), inhibit endothelium-dependent relaxation in many isolated vascularpreparations in vitro. The inhibition of endothelium-dependent relaxation byNSAAs are prevented by L-Arg but not D-Arg (Palmer et a!., 1988b; Rees et a!.,1990b; Moore eta!., 1990; Mülsch eta!., 1990) and are long-lasting. However, itis not yet known whether the sustained inhibition of relaxation by NSAAs isreversible.It was also reported that DPI effectively inhibited ACh- but not SNP-inducedrelaxation of preconstricted aortae of the rabbit (Stuehr et a!., 1991b) and rat(Rand and Li, 1993).1 .4.1 .1. Discrepancy between in vitro and in vivoWhile NSAAs effectively inhibit ACh- or bradykinin-induced vasodilatation inisolated preconstricted vascular preparations, or regional vascular beds, e.g.,coronary (Woodman and Dusting, 1991), hindquarter (Bellan et aL, 1991), renal(Gardiner et a!., 1990c, 1991; Lahera et a!., 1990), pulmonary (Fineman et a!.,1991), mesentery (Fortes et aL,, 1990; Gardiner et a!., 1990c) and carotid(Gardiner et a!., 1990c, 1991) beds, there are discrepancies in reports of theirabilities to interfere with the depressor effect of ACh in whole animals. Theinhibition of ACh- or bradykinin-induced depressor response has been shown by LNMMA (Whittle et a!., 1989; Rees et a!., 1990; Aisaka et a!., 1989b), L-NAME(Rees et a!., 1990) and L-Nl0 (Rees et a!., 1990). However, an absence of16inhibition, or even potentiation, of the depressor response to ACh has beenreported by L-NMMA (Yamazaki and Nagao, 1991), L-NNA (Treshman eta!., 1991;Wang et a!., 1992) and L-NAME (Gardiner et a!., 1990c, 1991; van Gelderen eta!., 1991).1 .4.1.2. Stereospecificity of NSAAsIt is extensibly reported that D-enantiomers of NSAA5 have no effect onendothelium-dependent relaxation in vitro. L-NMMA but not D-NMMA (Rees et a!.,1989a,b; Whittle et aL, 1989; Crawley eta!., 1990; Persson eta!., 1990; Rees eta!., 1990b), L-NAME but not D-NAME (Rees eta!., 1990b) and L-Nl0 but not DNI0 (Rees et a!., 1990b) contracted isolated arterial preparations, or reducedmicrovascular diameters in vivo. It was also reported that L-NMMA but not DNMMA, L-NAME but not D-NAME, and L-Nl0 but not D-Nl0 enhanced humanplatelet aggregation induced by ADP, arachidonic acid and thrombin (Radomski eta!., 199Cc). L-NMMA but not D-NMMA potentiated the aggregation of plateletsand white cells in rabbits (Persson et aL, 1990). Moreover, it has been reportedthat L-NNA but not D-NNA prevented EDRF release from endothelial cells andinhibited the dilator effects of ACh on rabbit femoral arteries (Mülsch and Busse,1990). L-NNA but not D-NNA inhibited NANC relaxation of guinea pig isolatedtracheal smooth muscle and rat anococcygeus (Tucker et a!., 1990; Hobbs andGibson, 1990). Therefore, it appears that all biological actions of NSAA5 areenantiomerically specific. However, it is well-known that although the Lenantiomeric form is a main configuration for biologically active drugs, many Denantiomers have less or even greater biological activities than their correspondingL-enantiomers (see review by Ariëns, 1983). Since the concentrations of Denantiomers used in all the above studies were never more than those of Lenantiomers, the activities of D-enantiomers may not be properly assessed. It is17reasonable to construct dose-response curves to elucidate the potencies andefficacies of both enantiomers of NSAAs.1 .4.2. Vasoconstriction in vivoThe administration of L-NMMA, L-NNA, L-NAME, L-Nl0 or L-NAA into intactanimals caused sustained pressor responses and bradycardia (see review byMoncada et a!., 1991). Although it was reported that L-NAME caused pressorresponse in pentobarbitone-anaesthetized rats but not in pentobarbitoneanaesthetized cats (van Gelderen et a!., 1991), the pressor responses to NSAA5were generally observed in all species tested, such as mice (Moore et a!., 1991),anaesthetized rats (Rees et a!., 1989b; Wang and Pang, 1990a,b), consciousnormotensive Sprague-Dawley (Wang et a!., 1 993b), Wistar and Wistar-Kyoto rats(Wang eta!., 1992), as well as spontaneously hypertensive rats (Yamazaki eta!.,1991; Wang eta!., 1992), guinea pigs (Aisaka eta!., 1989a,b), rabbits (Humphrieset a!., 1991), cats (Bellan et a!., 1991), dogs (Chu et a!., 1990; Woodman andDusting, 1991; Toda et al., 1993), sheep (Fineman et a!., 1991; Tresham et a!.,1991; Garcia et a!., 1992), monkeys (Peterson et a!., 1993), and human (Petros eta!., 1991). The pressor response to NSAA5 are attenuated by L-Arg but not D-Arg(Rees et a!., 1989b; Whittle et a!., 1989; Gardiner et a!., 1990b; Persson et aL,1990; Wang and Pang, 1990b). It was also reported that the pressor response toL-NMMA was accompanied by the inhibition of NO synthesis ex vivo (Rees et a!.,1989b) and in vivo (Suzuki eta!., 1992).It should be pointed out that the pressor responses to NSAAs are dependenton the conscious or anaesthetic condition of the experimental animals. Variableeffects of inhalation and intravenous anaesthetic agents on the pressor responsesto L-NNA and L-NMMA have been reported (Wang et a!., 1991; Aisaka et a!.,1991). Surgically anaesthetic doses of halothane totally and reversibly abolished18the pressor response to L-NNA (Wang eta!., 1991). Therefore, it is important toevaluate the influence of anaesthetic agents on the pressor responses tovasoactive agents (Abdelrahman et a!., 1992b), particularly to NSAAs (Wang eta!., 1991). Moreover, the effects of the vehicles for the drugs on the pressorresponses to NSAAs must also be considered. It has been reported that ethanol, afrequently used vehicle for drugs, dose-dependently and “noncompetitively”inhibited the pressor response to L-NNA (Wang and Pang, 1993). Furthermore,interactions of drugs should also be considered. It has been reported that whileSNP “noncompetitively” but selectively inhibited the pressor response to L-NNAbut not to noradrenaline or to angiotensin II, nifedipine (L type of calcium channelantagonist) nonselectively and “noncompetitively” attenuated the pressorresponses to L-NNA and noradrenaline and angiotensin. On the hand, pinacidil(ATP-sensitive K+ channel agonist) did not inhibit the pressor response to L-NNAnor noradrenaline or angiotensin II (Wang eta!., 1993b).L- but not D-enantiomers of NSAAs are generally reported to causehypertensive responses (Rees et a!., 1989b, 1990b; Humphries et aL, 1991).However, D-NNA has also been found to cause pressor responses inpentobarbitone-anaesthetized rats (Wang and Pang, 1 990a) and urethaneanaesthetized rats (Raszkiewicz et aL, 1992). The stereospecificity of NSAAs tocause vasoconstriction in vivo is not at all clear.It was also reported that aminoguanidine raised blood pressure inanaesthetized rats, but it was only approximately 1/40 as potent as L-NMMA(Corbett et aL, 1992). There are as yet no reports on the in vivo effects of DPI.1 .5. Mechanisms of the pressor responses to NSAAs19The pressor response to NSAAs are generally interpreted to be due to theinhibition of NO synthesis and subsequent endothelium-dependent vasodilatation.However, other mechanisms are possible and have been explored.1.5.1. The endothelial L-Arg/NO pathwayThe hypothesis that the pressor effects of NSAAs are attributed to theinhibition of endothelial NO biosynthesis, and subsequently that endothelial-derivedNO modulates vascular tone and blood pressure (Aisaka et a!., 1 989a; Rees et a!.,1989b, 1990b; Moncada et aL, 1991), is based on the following major evidencewhich was discussed previously in more detail. (1) NO is endogenously releasedand causes vasodilatation, therefore, the inhibition of NO biosynthesis in vivowould lead to the suppression of endothelium-dependent dilator tone and elevationof blood pressure. (2) L-Arg is the precursor of NO and the pressor responses toNSAAs are attenuated by L-Arg. However, since all the evidence obtained arefrom the results by using NSAAs, other classes of NOS inhibitors with structuresdifferent from NSAAs should be used to confirm the above hypothesis.1 .5.2. Endothelial membrane L-Arg transportIt is known that NO synthesis is absolutely dependent on the availability ofextracellular L-Arg and that L-Arg enters cells via the L-Arg transporter (Palmer etaL, 1988a; Bogle et aL, 1992a). Therefore, it is possible that NSAAs mayproduce pressor response by inhibiting the L-Arg transporter in the endothelialmembrane thereby restricting the availability of intracellular L-Arg to produce NO.However, this possibility is unlikely in the light of the report that L-Arg uptake isinhibited by L-NMMA and L-NlO but not by L-NNA or L-NAME (Bogle Ct a!.,1992b), whereas all these compounds are similarly effective in causing20hypertensive responses. Moreover, bradykinin elevated L-Arg transporter and NOsynthesis in endothelial cells (Boglé et a!., 1991). While L-NNA inhibitedbradykinin-induced NO biosynthesis, it had negligible effects on thebasal/stimulated L-Arg transporter.1 .5.3. The central nervous systemIt has been reported that intravenous injection of L-NMMA increased postganglionic renal sympathetic nerve activity both in intact rats and denervated rats.Spinal transection at C1-C2 level attenuated the pressor and elevated renalsympathetic nerve activity in response to intravenous injection of L-NMMA(Sakuma et aL, 1992). Furthermore, it was reported that intracisternal injection ofL-NMMA elicited a small pressor response accompanied by a marked increase insympathetic renal nerve activity. The increases in the sympathetic renal nerveactivity and blood pressure elicited by L-NMMA were abolished by spinaltransection at the C1-C2 level and by intravenous administration of L-Arg. Whenadministered intracisternally, L-Arg also abolished the increase in the sympatheticrenal nerve activity in response to intravenous injection of L-NMMA andsignificantly attenuated its pressor response (Togashi et a!., 1992). These resultsindicated that L-NMMA may raise blood pressure and increase sympatheticdischarge partially via the central nervous system, and that L-NMMA centrallystimulates the sympathetic nerve activity by an Arg-reversible mechanism inanaesthetized rats (Sakuma et a!., 1 992; Togashi et a!., 1992). These findings arealso supported by those of Harada et a!. (1993) in which microinjection of LNMMA into the rabbit nucleus tractus solitarius caused increases in blood pressureand renal sympathetic nerve activity, which were prevented by microinjection of LArg into the nucleus tractus solitarius. On the other hand, it was reported thatintracerebroventricular injection of L-NAME decreased blood pressure and heart21rate of LPS-treated rats but not the control rats. Furthermore, the pressorresponse to intracerebroventricular injection of N-methyl-D-aspartate (NMDA) wasenhanced by L-Arg or LPS, and in both cases the potentiation was blocked by LNAME. These results suggested that in some experimental conditions, such as theactivation of NMDA receptors or LPS pretreatment, the L-Arg/NO pathway mayinterfere with blood pressure and heart rate regulation (Mollace eta!., 1992).There is evidence that a central mechanism is not responsible for thepressor response to NSAA5 administered especially by peripheral routes. It wasreported that the pressor response to intravenous injections of L-NAME was notinfluenced by pithing (Pegoraro et a!., 1992). Moreover, intracerebroventricularinjections of L-NAME caused antinociceptive activity but not pressor responsewhile intraperitoneal injections of L-NAME caused both antinociceptive and pressorresponse in mice (Moore et a!., 1991). It was also reported that intravenousinjection of L-NMMA caused pressor response in pithed rats and that L-NAMEdose-dependently and frequency-dependently potentiated the pressor response toelectrical stimulation of the sympathetic chain in pithed rats and the potentiationwas greater in spontaneously hypertensive rats than in Wistar-Kyoto rats(Tabrizchi and Triggle, 1991, 1992). Therefore, it remains to be resolved whethera central mechanism contributes to the pressor responses elicited by peripherallyadministered NSAAs.1 .5.4. The autonomic nervous systemThe vascular endothelium has been shown to inhibit the release ofnoradrenaline from the sympathetic nerves which innervate canine pulmonaryartery and vein, by comparing the efflux of 14C-noradrenaline induced bytransmural nerve stimulation between endothelium-intact and denudedpreparations. These results suggest that endothelial cells, via EDRF/NO release,22may act as an endogenous inhibitor of transmitter release from the sympatheticnerve terminals (Greenberg et aL, 1989, 1990, 1991). Moreover, L-NMMA wasfound to increase the sympathetic nerve activity in rats (Sakuma et aL, 1992;Togashi et al., 1 992). It is therefore possible that the pressor responses to NSAAsare partially produced by activating ganglionic transmission and/or sympatheticnervous system.Indeed, it was reported that the ganglionic blockade with pentolinium orhexamethonium significantly reduced or abolished the hypertensive effect of LNMMA in urethane-anaesthetized rats (Vargas et aL, 1990) or the pressorresponse to L-NNA in pentobarbitone-anaesthetized dogs (Toda et at., 1993). Theganglion blocker chlorisondamine has also been shown to abolish L-NNA-inducedincreases in blood pressure and renal vasoconstrictor response, as well as toattenuate the increases in mesenteric and hindquarter resistances (Lacolley et al.,1991). In contrast to these findings, it was reported that pentolinium andhexamethonium potentiated the pressor response to L-NAME in urethaneanaesthetized rats (Chyu et al., 1992). The pressor responses to L-NMMA and LNAME were not influenced by treatment with hexamethonium in anaesthetized rats(Pegoraro et at., 1992). Moreover, the pressor and renal vasoconstrictorresponses to L-NNA were not impaired by chlorisondamine in pentobarbitoneanaesthetized rats (Pucci et aL, 1992). Hence, the role of ganglionic transmissionon the pressor responses to NSAAs needs to be examined in conscious animals toavoid the influence of anaesthesia and surgery on haemodynamic responses.On the other hand, it was reported that phentolamine, prazosin andIabetolol, as well as atropine and atenolol, did not inhibit the pressor response toL-NMMA, L-NNA or L-NAME (Rees et at., 1989b; Aisaka et a!., 1989a; Pucci etat., 1992; Widdop et at., 1992; Toda et a!., 1993). Moreover, L-NNA did notelevate plasma noradrenaline in conscious sheep (Tresham et at., 1991; EIsner etal., 1992). These results suggest that the pressor responses to NSAAs are not23due to the release of catecholamines or the activation of ct-adrenoceptors in thevascular smooth muscle.1 .5.5. Afferent nerve transmittersPretreatment with capsaicin did not alter the pressor response to L-NAME(Tepperman and Whittle, 1992; Wang and Pang, unpublished observation, 1993),suggesting that afferent nerve transmitters do not contribute to the pressorresponses to NSAAs.1 .5.6. The renin-angiotensin systemIt was reported that EDRF/NO inhibited the release of renin, and L-NMMAincreased renin concentration in rat renal cortical slices in vitro (Beierwaltes andCarretero, 1992). It may be reasonable to hypothesize that NSAAs elevate bloodpressure directly, by suppressing the synthesis and release of EDRF and indirectly,by elevating the activities of the renin-angiotensin system. In fact, it was reportedthat chronic treatment with L-NAME caused a sustained pressor response whichwas accompanied by elevated plasma renin activity and that the pressor responsewas prevented by the angiotensin II (AT-i) antagonist losartan (Ribeiro et aL,i992). A8i989, DuP753 and enalapril were also reported to block the pressorresponse induced by the chronic administration of L-NAME in conscious rats(Polakowski et aL, 1993). These results suggest that activation of the reninangiotensin system may contribute to the vasoconstrictor activity of L-NAMEadministered chronically. However, it was reported that the angiotensinconverting enzyme inhibitor captopril did not alter the pressor response to acuteadministration of L-NNA in rats (Pucci et aL, 1992). It appears that the roles of24the renin-angiotensin system on the pressor responses to acute or chronicadministration of NSAAs are different.1 .5.7. The prostaglandin systemAlthough it is widely-accepted in the scientific community that EDRF/NObiosynthesis does not involve the cyclooxygenase metabolism, discrepancies in theactions of the cyclooxygenase inhibitors on the pressor responses to NSAAs havebeen reported. Indomethacin did not inhibit the pressor response to L-NMMA or LNAME in anaesthetized rats (Rees etal., 1989b; Tepperman and Whittle, 1992). Itwas also reported that the combination of chlorisondamine, captopril andindomethacin did not impair pressor and renal vasoconstrictor responses to L-NNAin anaesthetized rats (Pucci et aL, 1992). In contrast, indomethacin was reportedto block the regional vasoconstrictor actions of L-NMMA in urethane-anaesthetizedmice (Rosenblum et a!., 1992). Moreover, indomethacin was also reported toabolish the pressor but not systemic vasoconstrictor and cardiac depressanteffects of L-NMMA in anaesthetized dogs (Klabunde eta!., 1991). Since all thesestudies were performed in anaesthetized animals without the construction of dose-response curves of NSAAs, the effects of cyclooxygenase inhibitors on dosepressor response curves to NSAAs should be performed in conscious animals toelucidate the role of prostanoids in the pressor responses to NSAAs.1 .5.8. VasopressinIt was reported that oral administration of L-NMMA or L-NAME causedpressor and regional vasoconstrictor responses in conscious Brattleboro rats withdiabetes insipidis which were vaspopressin deficient (Gardiner et a!., 1990a).Moreover, it was reported that the pressor and renal vasoconstrictor effects of L25NNA in anaesthetized rats were not impaired by pretreatment with a Vivasopressin antagonist (Pucci et aL, 1992), and that the combination ofpentolinium, captopril and vasopressin Vi antagonist did not inhibit the pressorresponse to L-NAME (Gardiner et aL, 1 990c). These results suggest thatvasopressin is not involved in the pressor responses to NSAAs.1 .6. The aims of this studyThis project was started in the middle of 1 990 when the in vivapharmacology of NOS inhibitors was rather novel. L-NNA was the first drug westudied due to its availability, potency and lower cost, relative to those of LNMMA. As the experiments progressed, we also studied L-NAME due to itssubstantially higher solubility in water than L-NNA. DPI was later tested becauseit is a new inhibitor of NOS with chemical structure distinct from NSAAs. Ouraims were to systematically and comprehensively investigate the vascularpharmacology of NOS inhibitors in order to elucidate the role of EDRF/N0 in bloodpressure and haemodynamic regulation as well as the pharmacodynamics of NOSinhibitors. During the course of the study, our acceptance of the hypothesis thatNSAAs cause in vivo vasoconstriction by the inhibition of endogenous endothelialNO biosynthesis has been weakened; this was reflected in our publications. Ourresearch primarily focused on the following areas.1.6.1. Stereospecificity of NSAAs.It is generally shown that only L-enantiomers of NSAAs are biologicallyactive in inhibiting NO biosynthesis and endothelium-dependent relaxation, as wellas in causing pressor responses. However, from our preliminary studies toexamine the effect of L-NNA on blood pressure, we found that D-NNA, initially26used as a negative control, also caused pressor response in pentobarbitoneanaesthetized rats (Wang and Pang, 1990a). It was also later reported that DNNA caused pressor response in urethane-anaesthetized rats (Raszkiewicz et aL,1992). Therefore, studies were conducted to examine whether stereospecificityexists for NSAAs, in their effects on the inhibition of endothelium-dependentrelaxations in vitro and ex vivo, and in their ability to raise blood pressure inconscious rats.1 .6.2. Effects of NSAAs and DPI on ACh-induced vasodilatations in vitro and invivoAlthough it is known that NOS inhibitors abolished in vitro endotheliumdependent relaxation, their in vivo effects on ACh-induced depressor response arenot clear. The information is crucial for the understanding of the pharmacology ofNOS inhibitors and the role of EDRF/NO in blood pressure regulation. In order toresolve the question of whether NOS inhibitors suppress endothelium-dependentvasodilatation in vivo, the effects of NSAAs were assessed along with those ofDPI. The effect of L-NAME on the depressor response to ACh was studied in thevehicle-treated rats as well as in phenylephrine (PHE)-treated rats. PHE wasutilized as a second control for L-NAME (and D-NAME) in order to raise bloodpressure to the same level as that produced by L-NAME (and D-NAME), sincedepressor responses are generally greater at a higher baseline blood pressure(Rees eta!., 1990b; van Gelderen eta!., 1991; Chyu eta!., 1992).1 .6.3. Pharmacodynamics of NSAAs and DPIIt is generally assumed that NSAAs are competitive inhibitors of NOS sincethey are structurally-related to L-Arg and their actions are antagonized by L-Arg.27However, there is little pharmacodynamic data to establish such a model in thevasculature, especially in vivo. There is also no report regarding thepharmacodynamics of DPI. Hence, experiments were carried out to investigate thein vitro and in vivo pharmacodynamics of L-NAME and DPI, via the use of modifieddose (concentration)-response model and modified Schild plots, by a computerprogram.1 .6.4. Mechanisms of the pressor responses to NSAAs and DPIThe inhibitory effects of NSAAs in inhibiting NO synthesis in vitro/in vivoand endothelium-dependent vasodilatation are antagonized by L-Arg. Theprolonged pressor responses to these compounds are also antagonized by L-Arg.Therefore, it is logical to assume that the pressor responses to NSAAs are due tothe inhibition of endothelial NO biosynthesis and subsequent endotheliumdependent vasodilatation. However, there are reports of other mechanisms (e.g.,inhibition of the central and ganglionic NO biosynthesis) contributing to the pressoreffects of NSAAs. Part of this dissertation examined the mechanisms by whichNSAAs cause pressor responses, e.g., whether the central and autonomic nervoussystem, renin-angiotensin system and prostanoid system, in addition to the L-Argpathway, are involved. Furthermore, the effect of NSAAs were compared withthose of DPI whose chemical structure is distinct from that of NSAAs, in order toelucidate whether the inhibition of endothelial NO biosynthesis alone necessarilylead to a rise in blood pressure in whole animals. During the coUrse of this study,it was found that DPI, unlike NSAAs which cause prolonged pressor responses,only produced transient elevation of blood pressure. Therefore, in vivoexperiments were also carried out with DPI to find out the mechanism by whichDPI causes this transient pressor response.282. Materials and methods2.1. Materials2.1.1. AnimalsMale Sprague-Dawley rats (280-420 g) were used in this study. All theanimals were from the Animal Care Center of the University of British Columbia.The recommendation from the Canada Council of Animal Care and internationallyaccepted principles in the care and use of experimental animals have been adheredto.2.1.2. DrugsThe following drugs were purchased from Sigma Chemical Co. (MO,U.S.A.): mecamylamine hydrochloride, atropine sulfate, D-L-propranololhydrochloride, N°-nitro-L-arginine (L-NNA), N°-nitro-L-arginine methyl ester (LNAME) hydrochloride, L-arginine (L-Arg) hydrochloride, guanethidine sulfate,rauwolscine hydrochloride, prazosin hydrochloride, acetylcholine (ACh) chloride,A 231 87, phenylephrine (PHE) hydrochloride, flavin adenine dinucleotide (FAD)disodium, 1-nicotinamide adenine dinucleotide phosphate (reduced form) (NADPH),indomethacin, bradykinin and tyramine hydrochloride. The following drugs werealso used: diphenyleneiodonium (DPI) sulfate (Colour Your Enzyme, Ont., Canada),phentolamine hydrochloride (CIBA Pharmaceutical Co., NJ, U.S.A.), captopril (E. R.Squibb & Sons Inc., NJ, U.S.A.), reserpine injection (CIBA Pharmaceutical Co.,Quebec., Canada) and tetrodotoxin (TTX) (Sankyo Co. Ltd., Tokyo, Japan),sodium nitroprusside (SNP) (Fisher Scientific Co., NJ, U.S.A.), halothane(Laboratories Ayerst, Montreal, Canada). All the powder drugs were solubilized in29normal saline (0.9% NaCI solution) except for DPI which was dissolved in 5%glucose solution, and prazosin, indomethacin as well as A 23187 which weredissolved in 100% 30% and 10% dimethyl sulfoxide, respectively. L-NNA and DNNA were dissolved in normal saline by 10 mm of sonication.2.2. Methods2.2.1. Isolated aortic ringsThe rats were sacrificed by a blow on the head followed by exsanguination.The thoracic aorta was removed and cleared of connective tissue. Four ringsegments of 0.5 cm length were prepared from one aorta and suspended randomlyin separate organ baths. Each ring was connected to a Grass FT-03-C force-displacement transducer (Grass Instrument Co., Quincy, MA, U.S.A.) for isometricrecording with a preload of 1 g. The rings were equilibrated for 1 h (with 3washouts) in normal Krebs solution (pH 7.4) at 37°C and bubbled with a gasmixture of 95% 02 and 5% C02. The Krebs solution had the followingcomposition (lxlO-3 mol/L): NaCI, 118; glucose, 11; KCI, 4.7; CaCI2, 2.5;NaHCO3, 25; KH2PO4, 1 .2; MgCI26H2O, 1 .2.The rings were first incubated with a vehicle or drugs followed by PHE (lxlO6 mol/L, EC90). After 15-20 mm, at the steady-state phase of the contractileresponse to PHE, a cumulative concentration-response curve of ACh, A 23187,bradykinin or SNP was constructed. Each concentration of drug was left in thebath for usually 1-3 mm until a plateau response was reached. The time taken tocomplete each concentration-response curve was approximately 20 mm. Ingroups where more than one concentration-response curve of ACh wasconstructed, the preparations were washed three times within 30 mm and givenanother 30 mm to completely recover from the effects of the previous applications30of PHE and ACh. Afterwards, PHE was again added followed by the constructionof ACh concentration-response curves.2.2.2. SurgeryThe rats were anaesthetized with sodium pentobarbitone (2.6x104 mol/kg,i.p.) or briefly with halothane, in the studies using conscious rats (4% in air forinduction and 1.2% in air for surgery). A polyethylene cannula (PE5O) wasinserted into the left iliac artery, for the measurement of mean arterial pressure(MAP) by a pressure transducer (Model P23DB) (Gould Statham, Cupertino, CA,U.S.A). Heart rate (HR) was determined electronically from the upstroke of thearterial pulse pressure using a tachograph (Model 7P44G) (Grass Instrument Co.,Quincy, MA, U.S.A.). One or two PE5O cannulae was/were also inserted into theleft or both iliac vein(s) for the administration of drugs. In some rats, anotherPE5O cannula was inserted into the right iliac artery to collect blood samples. Thecannulae were filled with heparinized (25 lU/mi) normal saline. For anaesthetizedrats, their body temperature was maintained at 370C with a heat lamp connectedto a thermostat (Model 73A) (Yellow Springs Instruments, Yellow Springs, Ohio,U.S.A.).Tetrodotoxin (TTX)-pretreated, pithed, spinal cord-transected rats and thecorresponding control rats were anaesthetized with pentobarbitone sodium.Tracheostomy was performed to allow artificial ventilation with 1 00% oxygen at54 strokes/mm and a stroke volume of 3-4 ml (1 ml/100 g body weight). Pithingwas performed through the orbit with a 3 mm-diameter stainless steel rod andspinal transection was performed at the T1 level by a pair of sharp scissors. Allexperiments were conducted 20 mm after surgery.In conscious rats, the cannulae were tunnelled subcutaneously along the backand exteriorized at the back of the neck. The rats were then put into small cages31allowing free movement and given more than 6 h’s recovery from the effects ofsurgery and halothane before use.2.2.3. Microsphere technique (Pang, 1 983)For haemodynamic experiments, additional cannulae were inserted into the leftventricle via the right carotid artery, for the injections of radioactively-labelledmicrospheres into the left iliac artery for blood withdrawal. A well-stirredsuspension of 30,000-40,000 microspheres (1 5 pm in diameter), labelled witheither 57Co or 3Sn (Du Pont Canada Inc., Ontario, Canada), was injected intothe left ventricle in the control period and after the administration of a drug or avehicle. The order of administration of the microspheres was reversed in half ofeach group of rats. At the end of the experiments, blood samples, whole organsof lungs, heart, liver, stomach, intestine, caecum and colon (presented as colon inthe table and figures), kidneys, spleen, testes and brain, as well as 30 g each ofskeletal muscle and skin, were removed for the counting of radioactivity (count perminute, cpm) using a 11 85 Series Dual Channel Automatic Gamma Counter(Nuclear-Chicago, Illinois, U.S.A.) with a 3 inch Nal crystal at energy settings of80-1 60 keV and 330-480 keV for 57Co and 11 3Sn, respectively. At these energysettings, the “spill-overs” from 57Co to the 11 3Sn channel was negligible (0.03%)and from 11 3Sn to the 57Co channel was 16%. A correction was made for the11 3Sn “spill-over”. MAP and HR were continuously recorded. Cardiac output(CC), total peripheral resistance (TPR) and organ blood flow were calculated asfollows. (1) CO (mI/min)=[rate of withdrawal of blood (mi/mm) x total injectedcpml/cpm in withdrawn blood. (2) Organ blood flow (mi/mm) =[rate of withdrawalof blood (mI/mm) x organ cpml/cpm in withdrawn blood. (3) TPR=MAP/CO.2.2.4. Measurement of plasma catecholamines (Passon and Peuier, 1973)32Plasma catecholamines were measured by a catecholamine assay kit(Amersham Canada Ltd., Ont., Canada). Blood samples (0.5 ml) were immediatelyinserted into precooled tubes containing ethylene glycol-bis(B-amino-ethyl ether)N,N,N’,N’-tetraacetic acid (EGTA) and reduced glutathione and centrifuged at1 ,200 g at 40 C. Afterwards, the plasma was removed and stored at 700 C untilassayed within a month. Duplicate assays were run for the standard, plasma ordiluted plasma samples (50 p1 each sample), with distilled water used as a blankcontrol for each run. The catechol-0-methyltransferase was used to catalyze thetransfer of a 3H-methyl group from S-adenosyl-L-[methyl-3H]-methionine to thehydroxyl group in the 3-position of noradrenaliné, adrenaline and dopamine. Theresultant products were separated by thin layer chromatography, eluted ifnecessary, and counted by a 1 600 TR liquid scintillation analyzer (PackardInstrument Co., CT, U.S.A.). The standard curves for noradrenaline, adrenalineand dopamine (0.01, 0.03, 0.1, 0.3, 1, 3 and 10 ng/ml for each standard solution)were prepared with the control rat plasma. Two standard curves wereconstructed at two separate occasions and were found to be indistinguishablefrom each other. The data were combined to formulate the following linearregression equations for noradrenaline, adrenaline and dopamine: Y=8.03x+29.5(r0.998, P<0.05), Y=2.49x-14.1 (r=0.999, P<0.05) and Y2.99x+10.2(r=0.999, P<0.05), respectively. The sensitivity of the catecholamine assay was0.005 ng/ml.2.3. Calculations and statistical analyses2.3.1. Dose (concentration)-response curve (Kenakin, 1987)33The parameters, i.e., minimal effect (Emin), maximal effect (Emax), half-effective (inhibitive) dose or concentration (ED50, EC50 or IC50) and Hillcoefficient (n) were calculated from individual dose-MAP and dose-HR curve oftesting drugs by using a program written by Dr. David M.J. Quastel in thisdepartment and executed on an IBM compatible microcomputer. To determinethese parameters, values of response (Y, rise in MAP or HR) at various doses (D)or concentrations (C) were fitted by non-linear least-squares to the relationYa+bx, where Y=response and x=[Dl’7/(E5017+[Dl) orx=[C]/(EC50+ ]) with n fixed at integral values (1, 2, 3, 4 and 5), andrepeated with n “floating” to obtain a best-fit (Quastel and Saint, 1988). Thisgave the value of ED50 or EC50 yielding a minimal residual sum of squares ofdeviations from the theoretical curve. This was preferred to the more usual fit toY=bx, in order to take into account the possible systematic underestimate oroverestimate of MAP or HR corresponding to [Dl or [C] = 0; the data set wasaugmented by 20 points with Y=0 at [Dl or [C]=0. Usually, the reduction inminimal residual sum of squares obtained by “floating” n was not significant in thesense that the reduction (from that obtained with the nearest integral value of n)was no more than expected from the reduction in degrees of freedom (by F test).With this fitting, the maximal response to [D] or [C] is given by b; values of a atthe best-fitting were never significantly different from 0.2.3.2. Modified Schild plot (Kenakin, 1987)In principle, a competitive inhibitor (I) shifts the dose-response curve to anagonist [Al in parallel to the right, with the ED50 increasing by the factor(1 +[IJ/K) when there is a one-to-one competition of I with A on the receptor.This arises from the well-known relationship: [ARI/[Rt]=[A]/{[Al+Ka(1 +[ll/Ki)}where [ARI/[Rtl is the fraction of receptors occupied by A and K8 and K, are the34dissociation constants for A and I, respectively. It can readily be shown that thesame relationship holds when A is an antagonist and I is an agonist.Using the standard assumption that equal responses, with and without I,represent equal [AR]/[Rt], it follows that Ka(l +[I1/K1) /[A]IKa/[Alo, where [A]1 isthe [A] that gives the same response in the presence of I as did [A]0 in theabsence of I. Therefore “dose-ratios” (DR) ([A]1/0,which are the same as ratiosof ED50s), follows the equation: DR=[A]1/[A]0=1+[lJ/Ki and, therefore, log (DR1)=log[l]-log K,. That is, the “Schild plot”- a graph of log (DR-i) vs log [II - has aslope of unity and Ki is given by the (extrapolated or interpolated) value of [II atDR =2.However, if the response depends upon a form of the receptor bound tomore than one (n1) molecules of agonist while more than one (p2) molecules of Iare necessary to block action of the receptor, the above equation does not hold.Instead, the above equations become, at the simplest (assuming high positivecooperativity in binding of A to R):[A1R]I[Rt] = [A]fll/{[A]fll + Ka1(1 +[I]2/K)}DR1 = (dose ratio)1 = ([A]/[A]0)fl1 = 1 + [lY2/K1DR1-i [l]fl2/KThus, one must plot log (DR1-1) vs“2 x log [I] to obtain a slope of unity.2.3.3. Statistics (Zar, 1984)The animals and aortic rings were randomly assigned into groups within eachexperimental design. Within each experimental design, all experiments wereperformed one block at a time. Six rats or 6 aortic rings (each derived from adifferent animal) were usually used except in cases where indicated.All results were expressed as mean±standard error (S.E.), or as geometricmean and 95% confidence range in cases where the results were transformed to35obtain homogeneity of variances. The results were analyzed by the analysis ofvariance followed by Duncan’s multiple range test, by Number Cruncher StatisticalSystem Program (by J. L. Hintze, Kaysville, Utah, U.S.A.), with Po 10C.,.4-C-c0-o0Bbiioi Iflui0.3 1.4 5.5 22 88Acetylcholine (1 mol/kg)Fig. 7. Effects (mean±S.E.) of i.v. bolus injections of acetylcholine on the meanarterial pressure (MAP, expressed as % baseline MAP, A) and modified half-recovery time (B) in conscious rats (N=6 each group) pretreated with L-arginine(L-Arg, 4.8x10-S mol/kg per mm), the vehicle (1 mI/kg), phenylephrine (2x10-8mol/kg per mm), NG-nitro-L-argmnine methyl ester (L-NAME, 4.8x10-5 mol/kg) or DNAME (1 .2x1 0-3 mol/kg). a denotes significant difference from the vehicle-treatedgroup; b denotes significant difference from phenylephrine-treated group.Modified half-recovery time was normalized as half-recovery time/baselineMAPx1O4.50In contrast, in the rats treated with bolus injections of SNP (1.5x109 to4.4x1 mol/kg) instead of ACh to cause transient depressions of MAP, thevasodilator response was somewhat potentiated by L-NAME or D-NAME, withsignificance at the lower two doses of SNP for L-NAME and at all the doses exceptthe third one for D-NAME, even when comparison was made with that the ratsinfused with PHE (which itself slightly and non-significantly increased SNPresponse) (Fig. BA). Decrease in the magnitude of SNP-induced depressorresponse by L-Arg was also observed, similar to those seen with ACh, but wassignificant (P040C)a200Sodium nitropruseide (1 moE/kg)Fig. 8. Effects (mean ±S.E.) of i.v. bolus injections of sodium nitroprusside on themean arterial pressure (MAP, expressed as % baseline MAP, A) and modified half-recovery time (B) in conscious rats (N6 each group) pretreated with L-arginine(L-Arg, 4.8x10-5 mol/kg per mm), the vehicle (1 mI/kg), phenylephrine (2x10-8mol/kg per mm), NG-nitro-L-arginine methyl ester (L-NAME, 4.8x10-5 mol/kg) or DNAME (1 .2x1 Q-3 mol/kg). a denotes significant difference from the vehicle-treatedgroup; b denotes significant difference from phenylephrine-treated group.Modified half-recovery time was normalized as half-recovery time/baselineMAPx1 04.A1008060ab40L—ArgVehiclePhenyiephrineL—NAMED—NAME20B1.5 6 27 110 44052L-NNA vasoconstricted all beds (Fig. 4B in Appendix VI). Changes inconductances were also expressed as % control to reflect the magnitudes ofvasoconstriction in each organ/tissue in response to L-NNA (Fig. 6 in AppendixVI). The results showed that, while L-NNA reduced vascular conductances in allbeds in conscious rats, the greatest influence was in the lungs while the least wasin the liver.3.1.6. Pressor responses to NSAA53.1.6.1. Time coursesIn the three groups of conscious rats subsequently treated with the vehicle (1mI/kg), L-NAME (4.8x105mol/kg) and D-NAME (1.2x103mol/kg), baseline MAPs(106±3, 112±3 and 117±3 mmHg) and HRs (339±8, 363±8 and 368±11beats/mm) were not significantly different from each other. l.V. bolus injection ofL-NAME and D-NAME (but not the vehicle) caused slow-developing and prolongedincreases in MAP. Plateau MAP responses to L-NAME and D-NAME were attainedapproximately 10 and 40 mm after i.v. bolus and lasted at least 2 h, with half-risephases of 2.3±0.4 and 7.3± 1.6 mm (P4 h). The prolonged duration of inhibitory effects of L-NNA is alsoobserved in the ex vivo studies, since the inhibitory effects of L-NNA on thevascular preparations were tested approximately 1 .5 h after in vivo administrationsof the drugs and after three washouts in baths. The long duration of action of LNNA is consistent with the report that L-NNA caused prolonged inhibition of NOsynthesis in cultured endothelial cells (Mülsch and Busse, 1990). Moreover, thepressor effect of L-NNA was prevented by pretreatment with L-Arg but notreversed by post-treatment with L-Arg (Wang and Pang, 1 990a; Zambetis et a!.,1991). These observations raise a possibility that the inhibitory effects of L-NNAand D-NNA are irreversible. It has been reported that L-NlO is a long-lasting andirreversible NOS inhibitor in rat peritoneal neutrophils and murine macrophages,64since its effects were not reversed by L-Arg but was prevented by concomitantincubations of L-Nl0 with L-Arg (McCall eta!., 1991a). It was also reported thatthe inhibition of NOS by L-NAA was reversible initially, but became irreversible withtime (Rouhani et a!., 1992). However, the inhibitory effects of L-NNA and LNAME on ACh-induced relaxation, unlike those of L-NlO and L-NAA, are preventedby L-Arg and reversed by L-Arg even after the preparations were incubated for1 .5 h with L-NNA and L-NAME. This also suggests that different mechanisms maybe involved in the pressor and inhibition of ACh-induced relaxation by L-NNA andL-NAME.We found that indomethacin does not affect the relaxant response to ACh.This is in accordance with studies showing that prostaglandins do not contributeto the effects of endothelium-derived relaxing factor or (EDRF)/NO (Furchgott andZawadzki, 1980; see review by Lüscher and Vanhoutte, 1990a). On the otherhand, it was recently reported that indomethacin, acetylsalicylic acid and SODblocked the effects of L-NMMA on contraction, on ACh- as well as L-Arg-inducedvasodilatations of pial arterioles and, on platelets adhesion/aggregation in mice invivo. It was suggested that L-NMMA interfered with endothelium-dependentrelaxation and it also produced constriction by activating cyclooxygenase andproducing superoxide which subsequently inactivated EDRF/NO (Rosenblum et a!.,1992). Indomethacin is frequently added to the physiological solution in order toavoid a possible contribution by prostaglandins to endothelium-dependentrelaxation (e.g., Mülsch and Busse, 1990). However, our results show thatindomethacin, at a concentration sufficient to inhibit prostaglandin synthesis, doesnot alter the inhibitory effects of L-NNA. These results suggest thatcyclooxygenase activation and subsequent superoxide production leading to theinactivation of EDRF/N0 are not involved in the inhibitory effects of L-NNA onACh-induced relaxation.653.2.2. Effects of NSAAs on endothelium-dependent vasodilatation in vivoWhile it has been shown by many laboratories that ACh-induced relaxation invitro is inhibited completely by NSAAs, there are uncertainties with respect to theefficacies of these compounds in suppressing ACh-induced vasodilatation in vivo.Inconsistent modifications (e.g., inhibition, no effect or even potentiation) of AChinduced depressor response by L-Arg analogues in vivo cast doubts upon the roleof NO in ACh-induced vasodilatation in whole animals. The discrepancies in theeffectiveness of NOS inhibitors in suppressing ACh-induced depressor responseare unlikely related to the species of animals used or the conscious oranaesthetized state. Instead, the inconsistencies are most likely due to thecomparison of ACh-elicited responses at different baseline MAPs (higher MAPcaused by NSAAs), the different modes of administrations (bolus injection vsinfusion) of ACh, as well as the way of expression of data. Vasodilator drugs areknown to cause greater hypotension at higher baseline MAPs (Rees et a!., 1990b;van Gelderen eta!., 1991; Chyu eta!., 1992).Therefore, PHE was used as a control for L-NAME (and D-NAME), and thedepressor responses were calculated as % baseline MAP and modified half-recovery time, in order to eliminate difficulties associated with the comparison ofresponses at different baseline MAPs. Indeed, the depressor responses to AChand SNP were potentiated by 140% and 95%, respectively, by PHE-inducedhypertension, as compared to those in the vehicle-treated rats. With the use ofappropriate controls, it was unequivocally clear that L-NAME (and D-NAME)interfered with the depressor response to ACh. This inhibition by L-NAME (or DNAME) of vasodepressor response has also been seen for calcitonin gene-relatedpeptide (Abdelrahman et a!., 1992a) and salbutamol (Wang et a!., 1993a).However, similar to the cases of calcitonin gene-related peptide and salbutamol,the depressor response to infused ACh was only partially (by 50%) suppressed by66L-NAME (and D-NAME). The lack of in vivo effectiveness of L-NAME (and DNAME) in eliminating response to ACh is not due to insufficient doses, since asupramaximal pressor dose (4.8x105 mol/kg, 10 times ED50 for pressorresponse) of L-NAME was used. In our preliminary studies, even at a dose as highas 3.8x10 mol/kg, L-NAME still only partially inhibited ACh-induced depressorresponse in conscious rats (N=2, data not shown). Moreover, theincompleteinhibition of ACh-induced depressor response by L-NAME is unlikely due to theactivation of nicotinic receptors by ACh, since ACh-induced activation ofganglionic nicotinic receptors are known to produce pressor response, rather thandepressor response. In addition, we found that following effective ganglionicblockade with mecamylamine, L-NAME (3.8x104mol/kg) still failed to completelyinhibit ACh-induced fall in MAP in pentobarbitone-anaesthetized rats (N =6) (Wanget a!., unpublished data, 1993). Gardiner et a!. (1990c) also reported similarfindings. These results suggest that there may be a difference between AChinduced responses in conduit vessels (e.g., the aorta) and resistance bloodvessels.Further analysis of the depressor response to bolus injections of ACh showedthat L-NAME decreased the duration, but not the magnitude, of the response toACh by 26%, which is in contrast to the depressor response to salbutamol inwhich L-NAME reduced the magnitude but not the duration (Wang et aL, 1993a).The results that L-NAME reduced the duration but not the magnitude of thedepressor response to ACh are in accordance with those of Aisaka et a!. (1 989b)using L-NMMA. Aisaka et a!. (1 989b) also showed that exogenous L-Argprolonged the duration of the depressor response to ACh and suggested that theavailability of L-Arg determined the duration of response to ACh. In contrast, wefound that L-Arg, at a dose 48-fold its half-block dose to antagonize the pressorresponse to L-NAME, did not affect the duration but reduced the magnitude of thedepressor response to the lower doses of bolus injected ACh. Van Gelderen et a!.67(1991) also found that L-Arg did not reduce the duration of ACh-induceddepressor response in anaesthetized rats. The mechanism by which L-Argsuppressed ACh-induced depressor response is not known but it is likelynonspecific, since the dose of L-Arg used was relatively high, and L-Arg alsoreduced the magnitudes of the depressor responses to the lower doses of SNP.It is unclear why L-NAME shortened but L-Arg did not prolong the duration ofthe depressor response to bolus injected ACh. It has been suggested that theendogenous concentration of L-Arg is sufficient to saturate NOS (Rees et aL,1989a). This hypothesis may explain why L-NAME but not L-Arg influenced theduration of the depressor response to ACh. An additional mechanism, besides NObiosynthesis, must be responsible for the establishment of the magnitude of thedepressor response to bolus injected ACh, which was neither reduced by L-NAMEnor prolonged by L-Arg. ACh is shown to release endothelium-derivedhyperpolarizing factor (EDHF) in addition to endothelium-derived relaxing factor(EDRF)/NO (Chen eta!., 1988; Chen and Suzuki, 1990; see review by Suzuki andChen, 1990). EDHF vasodilates some vascular preparations (Garland andMcPherson, 1 992) and may contribute to the depressor response to ACh in vivo.Consistent with this hypothesis, it was reported that the Ca2+activ ted K+channel (which leads to hyperpolarization) blocker charybdotoxin attenuated thedepressor response to ACh in rats (Watkins eta!., 1993).It is likely that L-NAME caused a redistribution of blood flow in rats. L-NNA(present study) and other NSAAs (Gardiner et al., 1990b,d) have been shown tocause pressor response by increasing total peripheral resistance via systemicvasoconstriction but their degrees of influence vary with different beds. It wasalso shown that the renal, internal carotid, common carotid and mesenteric but nothindquarter vasodilator effects of ACh were partially attenuated by L-NAME(Gardiner et a!., 1990c, 1991). The varying ability of L-NAME to inhibitvasodilator response to ACh in different beds suggests that part of the vasodilator68response to ACh is mediated via mechanism insensitive to the inhibition of NOsynthesis by L-NAME.3.2.3. Stereospecificity of NSAAs in vitro and in vivoIt has been shown that L-NMMA (Palmer et a!., 1 988b; Rees et at., 1 989a;Rees eta!., 1990b; Crawley eta!., 1990), L-NNA (Mülsch and Busse, 1990;Lamontagne eta!., 1991), L-NlO (Rees eta!., 1990b) and L-NAME (Rees etal.,1990b), but not the corresponding D-enantiomers, inhibited endotheliumdependent relaxation of isolated blood vessels and/or NO biosynthesis inendothelial cells (see review by Moncada eta!., 1991). L-enantiomeric specificityhas also been reported to exist in other tissues or cells, e.g., platelets (Radomskiet al., 1 990a,b), macrophages (McCall et at., 1 991 a), adrenal cortex (Palacioseta!., 1989) and non-vascular smooth muscles (Hobbs and Gibson, 1990; Tuckereta!., 1990). In contrast to these findings, our results indicate that both L-NNAand D-NNA, as well as L-NAME and D-NAME, efficaciously inhibit the relaxantresponse to ACh in vitro and ex vivo. Moreover, both L-NNA and D-NNA inhibitthe relaxant responses to the calcium ionophore A 231 87 and bradykinin. Theseresults suggest that both L-NNA and D-NNA as well as both L-NAME and D-NAMEinhibit endothelium-dependent relaxation induced by receptor- and non-receptor-operated mechanisms, and that the L-enantiomeric configuration is not required forthe actions of NNA and NAME. As the inhibitory effects of D-NNA and D-NAMEare also reversed by L-Arg, the results suggest that the actions of D-NNA and DNAME are also involved in the inhibition of NO synthesis.It could be argued that the effectiveness of D-NNA and D-NAME are due tocontamination with L-NNA and L-NAME, respectively. However, there is nomistake about the identity of D-NNA, as an independent analysis determined thatthe specific rotations, [ct]D, of D-NNA and L-NNA are -22.9° and +22.1°,69respectively (data shown in Appendix II). [aID of D-NNA from our independentanalysis is consistent with the information ([a]D of -23.6°) provided by thesupplier, Bachem Bioscienca Inc. (Philadelphia, PA, U.S.A.). Moreover, otherobservations also indicate that the biological activities of D-NNA or D-NAME arenot the result of contamination by L-NNA or D-NAME. (1) D-NNA from anotherdrug company (Aminotech Ltd., Ont.,. Canada) also exhibited similar biologicalactivities (data not shown). We have also examined D-NNA sent to us byinvestigators who have reported negative results and found that the drug hasactivities indistinguishable from those of our supply of D-NNA (data not shown).(2) There are differences in the biological activities between L-NNA and D-NNA aswell as L-NAME and D-NAME (see below). (3) The onsets of the pressor effectsof D-NNA and D-NAME are markedly slower than those of L-NNA and L-NAME.The reasons for the discrepancy between our results and those of others arenot apparent but may be related to differences in concentrations or doses of DNNA and D-NAME used, duration of observation, and possibly preconceived ideas.It is well-known that although the L-enantiomeric form is the main configuration ofbiologically active drugs, many D-enantiomers may have less or even greaterbiological activities than their corresponding L-enantiomers (see review by Ariëns,1983). Since the first report describing the enantiomeric specificity of L-Arg as asubstrate and L-NMMA as an inhibitor, in which the same concentrations of DNMMA and L-NMMA were used (Palmer et a!., 1988b), the concept of Lenantiomeric specificity for activating or inhibiting NOS has become widelyaccepted (see review by Moncada et a!., 1991). Due to the preconceived notionthat D-enantiomers of NSAAs are inactive, systematic studies have been notconducted with these compounds. The doses or concentrations selected for Denantiomers of NSAA5 as controls were always (without exception) the same asthose of the corresponding L-enantiomers. Moreover, conclusions were usuallydrawn without showing detailed data. Among the work cited in this dissertation,70only the experimental conditions in Mülsch and Busse’s report (1990) are similar toours. They found that L-NNA but not D-NNA (both at 3x105 mol/L) producedapproximately 80% inhibition of ACh-induced relaxation in noradrenaline (EC60)-preconstricted rabbit femoral arteries. In our study, L-NNA (3x105 mol/L,supramaximal dose) almost completely inhibited ACh-induced relaxation in rataortae. Since the in vitro potency of D-NNA is approximately 1/39 that of L-NNA,it would be expected that D-NNA (3x105 mol/L) should have caused considerablyless response in the rabbit femoral arteries. The potencies of NSAA5 are known todiffer greatly according to particular preparations and chemical structures (seereview by Moncada et al., 1991). Therefore, the potencies of D-enantiomerswould be expected to also vary with the preparations and types of NSAAs used.Indeed, we found that L-NNA is two-fold more potent than D-NNA in raising bloodpressure and 39-fold more potent than D-NNA in inhibiting endothelium-dependentrelaxation. L-NAME, on the other hand, is 52- and 337-fold more potent than DNAME in raising blood pressure and inhibiting endothelium-dependent relaxation,respectively. Moreover, the pressor responses to D-NNA and D-NAME aresubstantially slower in onset than the corresponding L-enantiomers. Thisdifference in the onsets between D- and L-enantiomers are accentuated inanaesthetized rats (Wang and Pang, 1990a). Therefore, it is reasonable to assumethat incorrect conclusions cannot be avoided when either the concentrations/dosesof NSAAs were insufficient or the observation time was not longer enough.There are differences in the vasoconstrictor effects between L-NNA and DNNA as well as L-NAME and D-NAME. Firstly, L-NNA concentration-dependentlycontracts aortic rings and potentiates PHE-induced contraction. Although D-NNAis as efficacious as L-NNA in inhibiting ACh-induced relaxation, it does not inducecontraction of aortic rings or potentiate PHE-induced contraction. It has beenreported that the concentrations of L-NNA and L-NMMA that were maximallyeffective at increasing tension in canine coronary arteries only caused submaximal71inhibition of ACh-induced relaxation (Cocks and Angus, 1991). In the presentstudy, lxi mol/L L-NNA produces maximal inhibition of ACh-induced relaxationbut does not produce maximal contractile response. The contractile effect of LNMMA was found to be endothelium-dependent and reversed by L-Arg suggestingthat this response was caused by the inhibition of basal NO formation (Palmereta!., 1988b; Rees eta!., 1989a). In contrast, Cocks and Angus (1991) showedthat the contractile response to L-NMMA in dog coronary arteries was not affectedby pretreatment with haemoglobin or FeSO4 in concentrations which inhibited therelaxations induced by SNP and NO, suggesting that the contractile response to LNMMA was independent of basal NO formation. Moreover, L-Arg was reported toreverse L-NAME-induced augmentation of contractions evoked by 5-hydroxytryptamine and histamine, but not L-NAME-induced inhibition ofendothelium-dependent vasodilatation evoked by ACh in perfused rabbit earpreparations (Randall and Griffith, 1991). We have also found that L-NNA but notD-NNA caused a slow and sustained contraction in endothelium-intact anddenuded rat aortic rings; the effect was not affected by L-Arg (Wang and Pang,i994b). In addition, L-NAME but D-NAME potentiated contraction induced byPHE. These results suggest that contraction and inhibition of the relaxantresponses by NSAA5 may be produced by different mechanisms.Another difference between L- and D- enantiomers of NNA and NAME ispotency. Although these four compounds have similar efficacies, D-NNA and DNAME are less potent than L-NNA and L-NAME in inhibiting endotheliumdependent relaxation, suggesting that the vasoconstrictor effects of NNA andNAME prefer the L-enantiomeric configuration. Moreover, the differences inpotencies between D-NNA and L-NNA as well as D-NAME and L-NAME in vitro aregreater than those in vivo. The mechanism responsible for this discrepanciesbetween the in vitro and in vivo potencies of D-NNA and L-NNA as well as DNAME and L-NAME are not known. One possible explanation is chiral conversion.72A metabolic chiral inversion has been shown to occur after the oral administrationof stereospecific drugs (Hutt and CaIdwell, 1983; Sanins eta!., 1991). Since DNNA and D-NAME are less potent and have slower onset of actions than L-NNAand L-NAME in vivo, D-NNA and D-NAME may act via metabolic conversion to LNNA and L-NAME in vivo, respectively. Metabolic conversion may account for thedifferences in the activity ratios of D- and L- enantiomers of NNA and NAME inin vivo and in vitro settings.3.2.4. Pharmacodynamic analyses of the vascular actions of NSAAsA Hill coefficient (n) of 0.9 for L-NAME-induced inhibition of ACh-inducedrelaxation was derived from our in vitro results. This value is the same as that ofL-NNA (0.9). These results suggest that one molecule of L-NAME or L-NNAcompetes with one molecule of endogenous L-Arg to inhibit NO biosynthesis. Ourin vivo results, on the other hand, show that the n for L-NAME to cause pressorresponse is 2.0, and this value is not significantly affected by L-Arg which causesa rightward displacement of the dose-response curve of L-NAME. The n for LNAME in vivo is also consistent with the n of 2.5 for L-NNA. These results implythat the presssor responses to L-NNA, L-NAME and D-NAME require the “positivecooperation!! of probably 2 molecules of the compounds (see review by Rang,1971; Pennefather and Quastel, 1982). For comparison, the n for the pressoreffects of angiotensin II is 1.0 (unpublished calculation from Wang eta!., 1993b)and DPI is 3.3 in conscious rats (Results and discussion II).The nature of the difference between n values for NSAAs obtained in vivo(pressor response) and in vitro (inhibition of vascular relaxation) is not known butmay suggest that the mechanism involve in raising MAP (in resistance bloodvessels) is different from that in inhibiting vascular relaxation (in large arteries). Itis possible that L-NAME and L-NNA raise MAP by an unknown mechanism which,73although it is reversed by the administration of L-Arg, is different from theinhibition of endogenous endothelial NO synthesis. An alternative explanation isthat there is a difference between stimulated NO synthesis (ACh-inducedrelaxation) and basal NO synthesis (endogenous dilator tone) (Chyu eta!., 1992).It was reported that the endothelium-dependent contractions elicited by L-NMMAand L-NNA in the dog coronary artery were not a consequence of the suppressionof basal NO synthesis (Cocks and Angus, 1991). It was also reported thatthimerosal, a acetyl-CoA lysolecithin acyltransferase inhibitor, blocked ACh-,substance P-, bradykinin- and A 231 87-induced relaxations but did not suppress LNNA-evoked contraction in the dog isolated coronary artery (Crack and Cocks,1992). Moreover, although both L-NNA and D-NNA inhibited ACh-inducedrelaxation in aortic rings, only L-NNA elicited contraction and potentiated PHEinduced contraction (see above).3.2.5. Mechanisms of the pressor responses to NSAAs3.2.5.1. Pressor responses to NSAAs are due to vasoconstrictionOur results show that L-NNA increased MAP in conscious rats by elevatingTPR, since both CO and HR were reduced. Reduced CO by L-NNA may be theresultant effects of reduced HR/cardiac contractility and increased flow resistance(TPR). The raised TPR is secondary to systemic vasoconstriction (reducedconductance) in all the beds. Humphries et a!. (1991) reported that intravenousinfusion of L-NNA in conscious rabbits raised MAP (by 11 mmHg) and TPR,reduced HR and CO, and caused significant vasoconstrictions in the brain, heart,kidneys, duodenum, but not in the muscle, skin, stomach, ileum or colon. Thegreater extent of vasoconstriction in response to L-NNA in our study is likely dueto the use of a higher dose of L-NNA. Other NSAAs such as L-NMMA and L74NAME were also reported to reduce CO and decrease renal, mesenteric,hindquarters or internal carotid blood flows in rats (Gardiner eta!., 1990a,b,c,d).However, the extent of vasoconstriction in response to L-NNA, as revealedby % conductance changes, is not uniform; the greatest influence is in the lungsand the least is in the liver. The lungs receive circulations from the bronchialartery, the pulmonary artery and arteriovenous anastomoses. Counts in the lungsreflect primarily circulations from the bronchial artery and arteriovenousanastomoses since microspheres are virtually completely trapped within onecirculation (Pang, unpublished observation, 1983). Likewise, due to theentrapping of microspheres in the splanchnic area, it is expected that liver bloodflow represents primarily hepatic arterial flow. It has been postulated that thehepatic arterial flow is controlled by the hepatic arterial buffer response such thatdecreases in portal venous flows are associated with increases in hepatic arterialflows, thereby maintaining the constancy of total hepatic blood flows (Lautt,1980). This hypothesis is in accordance with our findings that reduced splanchnicand consequently portal venous flows occurred concurrently with increasedhepatic arterial flow. Therefore, the lesser vasoconstrictor effect of L-NNA (lessreduced arterial conductance) in the hepatic bed may be due to the hepatic arterialbuffer response. Moreover, L-NNA caused a marked coronary vasoconstriction inconscious rats. NSAAs have been shown to cause coronary vasoconstriction inconscious rabbits (Amezcua eta!., 1989) and dogs (Chu eta!., 1990), as well as asustained increase in the rabbit coronary perfusion pressure in vitro (Palmer et a!.,1989). In contrast, Klabunde eta!. (1991) reported that L-NMMA and L-NNA didnot reduce HR or coronary flow in pentobarbitone-anaesthetized dogs. Theirinability to show coronary constrictor effect of NOS inhibitors may be due to theinfluence of pentobarbitone.3.2.5.2. Antagonism of L-Arg on the pressor responses to NSAAs75The pressor responses to L-NNA, D-NNA, L-NAME and D-NAME areprevented by L-Arg but not D-Arg. More detailed analysis shows that L-Arg dose-dependently shifted the dose-pressor response curves of L-NAME and D-NAME tothe right without changing Emax or n. A modified Schild plot demonstrates that LArg competitively antagonizes the pressor response to L-NAME, with half-blockingdose at 1x106 mol/kg per mm when n is chosen as 2. It should be pointed outthat the antagonism by L-Arg of the pressor responses to NSAAs is specific as LArg does not modify the pressor effect to noradrenaline nor angiotensin II inconscious rats. These results are consistent with reports that L-Arg did notattenuate the pressor effects of noradrenaline nor angiotensin II in pentobarbitoneanaesthetized guinea pigs (Aisaka et aL, 1989a) nor vasopressin in conscious rats(Gardiner et at., 1 990b).3.2.5.3. Effects of impairment of the central nervous, ganglionic, sympathetic,angiotensin or prostanoid systems on the pressor responses to NSAAsIt has been shown that pithing does not alter the pressor response tointravenous injection of L-NAME (Pegoraro et a!., 1992) and that intravenousinjection of L-NMMA causes a pressor response in pithed rats (Tabrizchi andTriggle, 1992). On the other hand, spinal transection has been reported toattenuate the pressor response to intravenous injection of L-NMMA (Sakuma et at.,1992; Togashi eta!., 1992). It is difficult to explain these controversial findingsbut our results, which showed that the intravenous injection of L-NAME intopithed rats caused a similar dose-pressor response to that in intact rats, suggestthat the pressor responses to peripherally administered NSAAs are not dependenton the integrity of the central nervous system. However, our results do notexclude the possibility that central NO biosynthesis modulates sympathetic nerveactivity and subsequently blood pressure and heart rate, as suggested by many76authors (Sakuma et al., 1992; Togashi eta!., 1992; Mollace et al., 1992; Haradaeta!. 1993).We have found that the ganglion blocker mecamylamine does not inhibit, butinstead potentiates, the pressor response to L-NNA in conscious rats. Moreover,mecamylamine did not inhibit the pressor response to L-NAME in conscious rats(Wang and Pang, unpublished data, 1993). These results are supported by thoseof other investigators using pentolinium or chlorisondamine (Chyu et al., 1992;Pucci et aL, 1992; Pegoraro et a!., 1992). However, there are reports whichshowed that chiorisondamine, pentolinium or hexamethonium abolished orattenuated the pressor response to L-NMMA or L-NNA in anaesthetized rats ordogs (Vargas et al., 1990; Lacolley et a!., 1991; Toda et aL, 1993), andsuggested that there were nitroxidergic vasodilator nerves innervating the arterialwall (Toda et a!., 1993). The difference in these findings also cannot beexplained. Nevertheless, the dose-response curves of our results performed inconscious animals suggest that ganglionic transmission is not a prerequisite for thepressor responses to NSAA5. It should be pointed out that potentiation of theeffects of NSAAs by ganglion blockers is not because of the lower baseline MAP,since rats treated with pentolinium and PHE to maintain similar baseline MAP asthose in the control rats also had larger pressor responses to L-NAME than thecontrols. The potentiation may suggest that the pressor response to L-NNA in thecontrol rats were limited by hypertension-induced reflex withdrawal of sympathetictone to the vasculature (Pucci eta!., 1992).Neither the blockers of the autonomic nervous system (phentolamine,reserpine, propranolol and atropine), nor the angiotensin converting enzymeinhibitor captopril attenuated the pressor effects of L-NNA or L-NAME. Theseresults are supported by many reports (Rees et aL, 1989b; Aisaka et a!., 1989a;Pucci et a!., 1 992; Widdop et a!., 1992; Toda et a!., 1 993). Also, theprostaglandin synthesis inhibitor indomethacin does not alter the dose-pressor77response curve to L-NNA or L-NAME. The results of indomethacin are inaccordance with reports which showed that the pressor response to L-NMMA or LNAME was not attenuated by indomethacin in anaesthetized rats (Rees et a!.,1989b; Tepperman and Whittle, 1992). However, indomethacin was reported toblock the pressor but not the systemic vasoconstrictor or cardiac depressionresponse to L-NMMA (Klabunde et a!., 1991). It should be pointed out that thepressor response in the latter study was small (10 mmHg). Our results suggestthat the acute pressor responses to NSAAs do not rely on the integrity of thesympathetic, parasympathetic, angiotensin or prostanoid system. Our conclusiondoes not exclude the possibility that the renin-aniotensin system may modulate thehypertension caused by chronic administration of NSAAs (Ribeiro et al., 1992;Polakowski et a!., 1993). However, it should be noted that captopril and otherangiotensin system inhibitors or antagonists also prevent the development of manykinds of hypertensions, such as in spontaneously hypertensive rats (Jonsson eta!., 1991; Chillon et aL, 1992; Wu and Berecek, 1993) and in rats harboringpheochromocytoma (Hu eta!., 1990). Therefore, the exact mechanism by whichthe inhibitors of the renin-angiotensin system prevent hypertension caused bychronic administration of NSAAs should be further studied.3.2.6. Mechanisms of the HR responses to NSAA5The bradycardiac response to L-NNA was abolished by the ganglion blockadewhile that to L-NAME was abolished by pithing. These results, which are alsosupported by another study in which mecamylamine also abolished thebradycardiac response to L-NAME (Wang and Pang, unpublished data, 1993),suggest that the bradycardiac responses to NSAAs are reflex-mediated. It is ofspecial interest that reserpine but not atropine reduced the bradycardia. The78results suggest that inhibition of the sympathetic nerve activity rather thanpotentiation of the parasympathetic nerve activity is responsible for the reflex-mediated HR response to elevation in MAP. These results are consistent with theobservation that the bradycardia caused by L-NMMA was associated with reducedrenal sympathetic nerve activity (Sakuma eta!., 1992). Our results also showedthat phentolamine increased the slope of the curve. This suggests that theenhanced reflex bradycardiac activities may be a consequence of the elevatedbackground sympathetic nerve activity in the presence of phentolamine. Indeed,HR was elevated after treatment with phentolamine. Phentolamine has beenshown to markedly elevate plasma levels of adrenaline and noradrenalinesuggesting that it increases activities of the sympathetic nervous system (Tabrizchieta!., 1988). However, it was reported that the bradycardia caused by L-NMMAin pentobarbitone-anaesthetized guinea pigs (Aisaka et at, 1989a) and thebradycardia caused by L-NAME in conscious, Long Evans rats (Widdop et at,1 992) were blocked by atropine. The difference between our results and other’smay be due to varying sympathetic and/or parasympathetic activities of theexperimental animals.L-NNA also caused a tachycardiac response following a bradycardiac phase.Biphasic HR responses to L-NNA and L-NMMA were also observed inpentobarbitone-anaesthetized rats (Wang and Pang, 1 990b), andchloralose/urethane-anaesthetized rats (Sakuma et at, 1992). The inability to seethe tachycardiac component of the HR response in rats given cumulative doses ofL-NNA may be due to the shorter observation period (15 to 20 mm). We foundthat mecamylamine also abolished the tachycardiac response to L-NNA. Sakumaet at (1 992) reported that the biphase HR responses to L-NMMA were associatedwith the biphase renal sympathetic nerve responses. Therefore, the tachycardiacresponses to NSAAs may be due to the activation of the sympathetic nervoussystem.793.3. Summary1. L-NNA and D-NNA, as well as L-NAME and D-NAME, cause efficacious, [onglasting and reversible inhibition of endothelium-dependent relaxations in vitro andex vivo. The inhibitory effects of NSAAs are antagonized by L-Arg, but not by DArg, indomethacin or NADPH. These results suggest that the inhibitory effects ofNSAAs are due to the inhibition of NO biosynthesis by antagonizing the substrateL-Arg. The L-enantiomer form of NSAAs is a preferred but not essentialconfiguration required to inhibit endothelium-dependent relaxation. Since theratios of the potencies of L-NNA vs D-NNA as well as L-NAME vs D-NAME aregreater in in vitro than in vivo settings and since D-NNA and D-NAME are slower inonset of action than L-NNA and L-NAME, respectively, it is possible that the Denantiomers of NNA and NAME are converted to the corresponding L-enantiomersin vivo.2. L-NNA and D-NNA, as well as L-NAME and D-NAME cause slow-developingand long-lasting pressor and bradycardiac responses in conscious rats, as aconsequence of generalized vasoconstriction of all beds. The bradycardiacresponses to NSAA5 are baroreflex-mediated. The pressor responses to NSAAsare not dependent on the integrity of the central/autonomic nervous, angiotensinand prostaglandin systems. L-Arg, but not D-Arg, competitively antagonizes thepressor responses to NSAAs. These results suggest that the in vivovasoconstriction elicited by NSAAs is due to a mechanism which is related to theL-Arg pathway.3. In contrast to the effectiveness in inhibiting relaxation in vitro, L-NNA and LNAME do not entirely block ACh-induced depressor responses in vivo. The ns forL-NNA and L-NAME to inhibit endothelium-dependent relaxation are 1 in vitro andthe ns for L-NNA and L-NAME to produce pressor response are 2 or more in vivo.Our results appear to be consistent with the hypothesis that the pressor responses80to NSAAs result from the inhibition of NOS and suppression of endothelial basalNO synthesis only if (1) NSAAs blocks basal NO synthesis more effectively thanstimulated NO synthesis, and (2) stimulated NO synthesis is inhibited by NSAAs ina one to one antagonistic manner (NSAAs vs L-Arg) while basal NO synthesis isinhibited by NSAAs in a two (or more) to one antagonistic manner (NSAAs vs LArg).4. Therefore, the mechanisms responsible for NSAAs to cause inhibition ofendothelium-dependent relaxation and pressor responses may be different,although both effects of NSAAs are antagonized by L-Arg. Either one of thefollowing explanations are consistent with our observations: (1) pressor responseto NSAAs is due to the activation of an “L-Arg receptor” with signal transductiondifferent from NO biosynthesis; (2) the inhibition of NO biosynthesis alone is notsufficient to cause vasoconstriction in vivo.5. It is obvious that the commonly used NOS inhibitors (I.e., NSAAs) which havestructures related to Arg cannot be used to provide further information on thequestion- whether the inhibition of NOS leads to the elevation of blood pressure.Hence, use of other NOS inhibitors (e.g, DPI) with different chemical structuresand mechanism of action would be of value.814. Results and discussion II4.1. Results4.1.1. Effects of DPI on endothelium-dependent relaxation in vitro4.1 .1 .1. Concentration-responsesAll five concentrations of DPI (3x10-8, lxlO-7, 3x10-7, ixlO-6 and 3x10-6mol/L) slightly potentiated PHE-induced contraction from the baseline value of0.99±0.10 g to 1.09±0.15, 1.27±0.11, 1.32±0.10, 1.31 ±0.10 and1.18±0.14 g, respectively. However, only the effects of the third and fourthdoses of DPI were statistically significant (P<0.05).In the vehicle-treated group, ACh (lxlO-8 to 3x10-5 mol/L) concentration-dependently relaxed the preconstricted aorta with maximal relaxation ofapproximately 60% (Fig. 1A in Appendix VII). DPI concentration-dependently and“noncompetitively” inhibited ACh-induced relaxation. At 3x10-5 mol/L ACh, then,1C50 and Emax of DPI were 1 .4, 1 .8x1 0-7 mol/L and 92%, respectively (Fig. 1 B inAppendix VII).In another two vehicle-treated groups, A 23187 (lxi 0-9 to 3x10-6 mol/L) andSNP (3x10-10 to lxlO-7 mol/L) also concentration-dependently relaxed thepreconstricted aorta, with maximal relaxations of approximately 60% and 100%,respectively (Fig. 2 in Appendix VII). DPI (3x10-6 mol/L) completely inhibitedA 231 87-induced relaxation but did not affect the relaxant response to SNP.4.1.1.2. Mechanisms82Baseline contractions elicited by PHE in the presence of the vehicle or DPI(3x10-7 mol/L) were 1.29±0.07 and 1.67±0.1.2 g, respectively. In 10 differentgroups of aortae, treatments with NADPH (1.5 and 5x10-3 mol/L), FAD (5x10-4and 5x10-6 mol/L) and L-Arg (2x10-3 mol/L) did not significantly affect PHEinduced contractions in the presence of either the vehicle (1 .04±0.06,1.04±0.11, 1.16±0.06, 1.33±0.11, 1.23±0.08 g, respectively) or DPI(1 .43 ±0.10, 1 .37±0.14, 1 .59 ± 0.06, 1 .52 ± 0.13, 1 .44±0.04 g, respectively).DPI inhibited ACh-induced relaxation (Fig. 3A in Appendix VII). Treatmentwith L-Arg did not affect either ACh-induced relaxation or the inhibitory effect ofDPI on ACh-induced relaxation (Fig. 3B in Appendix VII). Although the lowerconcentration (5x10-6 mol/L) of FAD also did not alter either ACh-inducedrelaxation or the inhibitory effect of DPI on Ach, the higher concentration (3x10-4mol/L) of FAD suppressed the relaxant effect of ACh and prevented furtherinhibition by DPI of ACh-induced relaxation (Fig. 3C in Appendix VII). Althoughboth concentrations of NADPH did not significantly affect ACh-induced relaxation,the higher concentration (5x10-3 mol/L) but not the lower concentration (1 .5x1 0-3mol/L) of NADPH completely prevented the inhibitory effect of DPI (Fig. 3D inAppendix VII). The effectiveness of pretreatment with NADPH (5x10-3 mol/L) ininhibiting the effect of DPI, expressed as the ratio of the relaxant effect of lxi 0-5mol/L ACh in the presence of NADPH (67%) to that in the absence of NADPH(32%), was 209%.4.1.1.3. Time courses and reversibilityThe PHE-induced contractions in the presence of the vehicle or DPI did notchange with passage of time (data not shown). ACh-induced maximal relaxationwas not altered until at least 4 h after washouts. Maximal relaxation at 9 h was48±7%, which was significantly less than that (69±6%) at 0 h (Fig. 5 in83Appendix VII). DPI at 3x10-7 and 3x10-6 mol/L inhibited ACh-induced relaxationby approximately 50 and 100%, respectively (Fig. 5A in Appendix VII). Theinhibitory effect of DPI remained at least 4 h after washouts (Fig. 58, 5C inAppendix VII). At 9 h after washouts, the relaxation of DPI-preconstricted ringswas still less, though insignificantly, than those of the vehicle-pretreated rings(Fig. 5D in Appendix VII).Maximal relaxation of ACh after 1 .5 h exposure to 3x1 0-7 mol/L DPI(38.3±3.1%, Fig. 6 in Appendix VII) was similar to that after a 10 mm exposureto DPI (32.2±4.8%, Fig. 3A in Appendix VII). Post-treatment (1.5 h later) withNADPH (5x10-3 mol/L) significantly but slightly suppressed the inhibitory effect ofDPI. The effectiveness of post-treatment with NADPH (5x10-3 mol/L) in inhibitingthe effect of DPI, expressed as a ratio of the relaxant effect of lxlO-5 mol/L AChin the presence of NADPH (51 %) to that in the absence of NADPH (38%), was134% (Fig. 6 in Appendix VII).4.1 .2. Effects of DPI on the depressor responses to ACh and SNP in vivoIn conscious rats, Lv. bolus injection of DPI caused an immediate and transientpressor response (see later) and did not cause additional pressor responses during2 h observation (Fig. 7 in Appendix VII). Baseline MAPs of the conscious ratsbefore and 20 mm after treatment with the vehicle were 1 09 ± 2 and 11 2 ± 3mmHg, respectively, which were similar to those (118±5 and 114±5 mmHg) ofDPI (lxlO-5 mol/kg)-treated rats (lxlO-5 mol/kg, i.v. bolus injection). l.V.infusions of ACh (6x10-8 to 1.8x10-6 mol/kg per mm, each dose for 4 mm) andSNP (3x10-8 to 4.8x10-7 mol/kg per mm, each dose for 4 mm) caused dosedependent depressor responses. Pretreatment with DPI significantly attenuatedthe depressor response to ACh but not to SNP (Fig. 8 in Appendix VII).844.1 .3. Pressor and tachycardiac responses to DPIl.V. bolus injections of the vehicle did not alter MAP or HR in pentobarbitoneanaesthetized rats (data not shown). l.V. bolus injections of DPI (1.5x10-7 to5x1 0-6 mol/kg) caused immediate and transient pressor and tachycardiacresponses as shown in a typical experimental tracing from a rat (Fig. 2 in AppendixIV). The duration of the pressor response lasted approximately 1-2 mm while thatof tachycardiac response was slightly longer (Fig. 2 in Appendix IV). At 2.5x1 06mol/kg, the half-rise time for the pressor and tachycardiac responses were2.9±0.2 s and 4.9±0.7 s (P<0.05) while the corresponding half-fall time were31 ±3 s and 60±9 s (P<0.05), respectively.Pooled (N = 12) baseline MAP and HR from the above group and the controlgroup in another protocol to be described later were 104±3 mmHg and 347± 10beats/mm, respectively. The pressor and tachycardiac response curves of DPIwere dose-dependent and notably “steep”, with negligible effect at 1.5 and 3x107 mol/kg, large increases in MAP at 6x107 and 1.2x106 mol/kg and maximaleffect at 2.5x106 mol/kg (Fig. 3 in Appendix IV). The best-fitted n for MAP andHR were 3.6±0.3 and 4.2±0.6, respectively. These two values were notsignificantly different from each other or from 4, but different from 1, 2, 3 and 5,though all values of n, best-fitted (3.6, 4.2) or selected (1, 2, 3, 4 and 5), werestatistically significant (P<0.05). The theoretical dose-response curves forintegral values of n are shown in Fig. 3 in Appendix IV. Emin and Emax calculatedfrom dose-MAP and dose-HR curves at the best-fitted n were not significantlydifferent from those observed; other values of n gave significant differences ofcalculated parameters from those observed (Emin and Emax) or derived byaveraging parameters obtained using best-fitted n for each dose-response curve(Table 1 in Appendix IV). Correlation between observed data points and85theoretical curves, expressed as l000x(1-r), was the greatest when n=3 and 4,and the least when n = 1.4.1.4. Mechanisms of the pressor and tachycardiac responses to DPI4.1 .4.1. Effects of pharmacological antagonists on the pressor and tachycardiacresponses to DPICompared to the vehicle (1 mI/kg), treatments with mecamylamine (3.8x105 mol/kg) and reserpine (7.7x106mol/kg, i.p. 26 h earlier) reduced both MAP andHR; phentolamine (3.2x105 mol/kg), propranolol (3.4x106mol/kg) and captopril(9.2x105mol/kg) reduced MAP but did not affect HR while guanethidine (2.0x105mol/kg) did not alter MAP but increased HR. On the other hand, atropine(1 .5x10 mol/kg), prazosin (2.4x106mol/kg), rauwolscine (2.6x106mol/kg) andL-Arg (1 .9x1 O mol/kg) did not alter either MAP or HR (Table 2 in Appendix IV).The dose-MAP and dose-HR response curves of DPI (1.5x1O7 to 5x106mol/kg) in the presence of the vehicle or the antagonists, and the correspondingED50s and EmaxS at the best-fitted ns were shown in Figs. 4-6 and Table 3 inAppendix IV, respectively. Reserpine, guanethidine and mecamylamine attenuatedthe MAP as well as HR responses to DPI (Fig. 4 in Appendix IV) with either adecrease in Emax or an increase in ED50 of DPI (Table 3 in Appendix IV).Phentolamine and prazosin, but not rauwolscine, reduced the MAP response bydecreasing Emax of DPI. The HR response and the corresponding Emax, on theother hand, were increased by phentolamine and rauwolscine but not prazosin(Fig. 5 and Table 3 in Appendix IV). Propranolol abolished the HR response with adecrease in Emax, but potentiated the MAP response to DPI with an increase inEmax (Fig. 5 and Table 3 in Appendix IV). Atropine potentiated both the MAP (bydecreasing ED50 and increasing in Emax) and HR (by decreasing ED50) responses86to DPI (Fig. 6 and Table 3 in Appendix IV). Captopril markedly potentiated theMAP (by increasing Emax) but not HR response while L-Arg did not affect eitherthe MAP or HR response to DPI (Fig. 6 and Table 3 in Appendix IV).4.1.4.2. Effects of TTX, pithing or spinal cord (T1) transection on the pressor andtachycardiac responses to DPIBaseline MAPs in TTX (3.lxlO-8 mol/kg)-pretreated rats (53±2 mmHg),pithed rats (45±2 mmHg) and spinal cord-transected (T1) rats (52±2 mmHg)were lower than those of the ventilated control rats (105±5 mmHg). BaselineHRs in TTX-pretreated rats (318±9 beats/mm) and pithed rats (333±10beats/mm) were similar to, while those of spinal cord-transected rats (41 2 ± 1 6beats/mm) were higher than those of the ventilated control rats (312±10beats/mm). DPI (1 .5x107 to 5x106 mol/kg) also caused dose-dependent pressorand tachycardiac responses in the ventilated control rats; pretreatment with TTXand pithing totally abolished both the MAP and HR responses to DPI (Fig. 7 inAppendix IV). However, noradrenaline (3.9x108 mol/kg) still caused increases inMAP and HR in TTX-pretreated and pithed rats; their increases were 109±1 and44±2 mmHg, and 92±6 and 57±6 beats/mm, respectively. On the other hand,spinal cord transection (T1) markedly suppressed the dose-MAP and dose-HRresponses to DPI compared to the responses in the intact rats; the suppression byspinal cord transection, however, was less than that by pithing (Fig. 7 in AppendixIV).4.1 .4.3. Effects of DPI on plasma catecholamines in intact, pithed or reserpinizedrats.87Baseline levels of plasma catecholamines were similar among the two groupsof intact rats to be treated with the vehicle and DPI (Table 4 in Appendix IV).Compared to the pooled data, pithing did not alter circulating catecholamines.Pretreatment with reserpine significantly decreased plasma noradrenaline butincreased plasma adrenaline.Compared to the vehicle, DPI (5x106 mol/kg) caused large increases inplasma noradrenaline and adrenaline (more than 1 ng/mI), and moderate increasein plasma dopamine (more than 0.1 ng/mI), as well as increases in MAP and HR(Fig. 9 in Appendix IV). The increases in MAP and HR were significantly greaterthan those caused by the same dose of DPI in the multiple dose regimen (70±2 vs53±5 mmHg and 123±7 vs 43±5 beats/mm, respectively). Reserpine markedlyreduced DPI-induced increases in plasma noradrenaline, adrenaline and dopamineby 91 %, 93% and 74%, respectively, and attenuated the pressor andtachycardiac responses by 56% and 68%, respectively (Fig. 10 in Appendix IV).In reserpinized rats, the pressor and tachycardiac responses to a single dose ofDPI were also greater than those of the multiple dose regimen (31 ±10 vs 18±1mmHg and 39±10 vs 22±1 beats/mm, respectively). Pithing totally abolished theeffects of DPI on plasma catecholamines, as well as on MAP and HR (Fig. 10 inAppendix IV).Concentration-response and linear regression models were used to examinethe relationships between plasma noradrenaline or adrenaline and MAP or HR inintact, pithed and reserpinized rats i.v. bolus injected with DPI or the vehicle(N =25). The linear regression model gave significant correlation between plasmanoradrenaline and MAP (r=0.83) or HR (r=0.87), as well as between adrenalineand MAP (r=0.78) or HR (r=0.81). The concentration-response model, however,gave better fits. Fig. 11 and 1 2 in Appendix IV showed the concentrationresponse relationships between individual plasma noradrenaline or adrenaline andMAP or HR response caused by DPI. Correlation coefficient between plasma88noradrenaline and MAP or HR were 0.97 or 0.96, respectively; correlationcoefficient between plasma adrenaline and MAP or HR were 0.94 or 0.94,respectively.4.1.3. Inhibitory effect of halothane on the pressor response to DPIIn conscious rats, DPI caused an immediate (approximately 1 5 s in onset) andtransient (1-2 mm in duration) pressor response, similar to those in pentobarbitoneanaesthetized rats. The pressor response was dose-dependent (Fig. 1 in AppendixV), with ED50 of 2.2±0.3x107 mol/kg and maximal MAP reached at 59±2mmHg, based on the best-fitted calculations (Table 1 in Appendix V). Hillcoefficient (n) of 3.3 ±0.5 was significantly different from 1, 2 and 5 but not from3 or 4. DPI also caused tachycardia at the lower doses (7.5x108 to 3x107mol/kg), bradycardia at higher doses (6x107 to 5x106 mol/kg) and movementsfollowing the onset of the pressor response (data not shown).Halothane (0.5-1 .25%) reduced baseline MAP in a dose-dependent manner(Table 1 in Appendix V). Halothane dose-dependently reduced the maximal effectof DPI and shifted the dose-pressor response curve of DPI to the right (Fig. 2 inAppendix V). ED50s were linearly correlated while EmaxS were inverselycorrelated with the doses of halothane (Fig. 3 in Appendix V). None of the dosesof halothane affected the n of the curves (Table 1 in Appendix V). Halothane alsoinhibited DPI-induced tachycardiac and bradycardiac responses, as well asmovements (data not shown).Baseline MAP in halothane (1 .25%)-anaesthetized rats was lower than that inconscious rats (Table 2 in Appendix V). Baseline plasma level of adrenaline butnot noradrenaline or dopamine in halothane-anaesthetized rats was alsosignificantly lower than that in conscious rats (Table 2 in Appendix V). Inconscious rats, i.v. bolus injection of DPI (5x106 mol/kg) caused immediate and89large increases (more than 1 ng/ml) in plasma noradrenaline and adrenaline and asmaller increase (0.1 ng/ml) in plasma dopamine (Fig. 4A in Appendix), as well asimmediate pressor response (Fig. 4B in Appendix V). Halothane markedlyattenuated DPI-induced increases in plasma catecholamines (Fig. 4A in AppendixV) and in MAP (Fig. 4B in Appendix V); the reductions of plasma noradrenaline,adrenaline and MAP were 86%, 81 % and 95%, respectively.4.2. Discussion4.2.1. Inhibitory effects of DPI on endothelium-dependent vasodilatation in vitroand in vivo, and their mechanismsOur in vitro results show that DPI selectively and completely inhibitsendothelium-dependent relaxation induced by receptor-mediated (ACh) or non-receptor-mediated (A 231 87) mechanisms. These results are consistent with thereport that DPI inhibits ACh-induced relaxation in the rabbit and rat aortae (StuehretaL, 1991b; Rand and Li, 1993). DPI also attenuates ACh- but not SNP-induceddecreases in MAP in conscious rats, and ACh- but not SNP-induced vasodilatationin the perfused rat hindquarter and mesenteric preparations (Wang et a!., 1993;unpublished data, 1993). These results suggest that DPI inhibits endotheliumdependent vasodilatation in both conductance and resistance blood vessels.Therefore, the in vitro inhibitory effects of DPI on endothelium-dependentvasodilatation are similar to those of NSAAs. These results are in accordance withthe hypothesis that the inhibition of NO synthesis suppresses endotheliumdependent vasodilatation.It has been known since 1 973 that DPI suppresses the oxidation of NADH-Iikesubstrates thereby inhibiting mitochondrial oxidation (Holland et a!., 1973). It waslater shown that DPI inhibits NADPH-dependent oxidase of neutrophils and90macrophages (Cross and Jones, 1986; Hancock and Jones, 1987; Ellis et a!.,1988, 1989), and macrophage NOS (Stuehr eta!., 1991b), by specifically bindingto and inhibiting the action of a plasma membrane polypeptide which may be acomponent of flavoprotein (Cross and John, 1986; Hancock and Jones, 1987; Elliset a!., 1989). This suggests that flavin is the site of attack by DPI and that aprotein is associated with FAD (O’Donnell et aL, 1993). lsoenzymes of NOS areknown to be flavoproteins which contain FAD as a cofactor in the endothelial cells(Marsden eta!., 1992), macrophage (Stuehr eta!., 1989b, 1990, 1991a,b; Hevel.eta!., 1991; White and MarIetta, 1992), neutrophil (Yui eta!., 1991), brain (Mayereta!.,, 1991; Lowenstein et a!., 1992; Bredt et a!., 1991, 1992; Hiki eta!., 1992)and liver (Evans eta!., 1992). There is, however, no functional documentation ofa role of FAD as a cofactor of eNOS. Our in vitro results demonstrate that FADinterferes with both ACh-induced relaxation and the inhibitory effect of DPI onACh-induced relaxation. The latter result, which is consistent with Stuehr et aL’sreport (1991b) that FAD antagonizes the inhibitory effect of DPI on macrophageNO synthesis, suggests that FAD and DPI may inhibit endothelial NO synthesis bya mechanism similar to that in macrophages. The former result is puzzling, sinceas a cofactor, FAD should facilitate rather than interfere with endotheliumdependent relaxation. FAD was indeed reported to facilitate macrophage NOsynthesis (Stuehr et a!., 1990; Hevel. et a!., 1991). The mechanism by whichFAD inhibits ACh-induced relaxation is not clear at the moment, however, theeffect may not be specific as FAD also inhibits SNP-induced relaxation (Wang andPang, unpublished observation, 1993).Our in vitro results also show that the inhibitory effect of DPI is not affectedby L-Arg, at a concentration found to reverse the inhibitory effects of L-NNA andL-NAME on endothelium-dependent relaxation in aortic rings. The results are inaccordance with those of Stuehr eta!. (1991b), in which L-Arg did not prevent theinhibitory effect of DPI in NO biosynthesis in macrophages, and with those of Rand91and Li (1993), in which L-Arg did not attenuate the inhibitory effect of DPI onendothelium-dependent relaxation in the isolated rabbit aorta. Moreover, NADPHinterferes with the inhibitory effect of DPI on ACh-induced relaxation. Theantagonism of DPI by NADPH is specific since the same concentration of NADPHdoes not alter the inhibitory effect of L-NNA. Our functional results with NADPHand DPI are consistent with those which show that both the constitutive (e.g.,brain and endothelial) and inducible (e.g., macrophage and smooth muscle) NOSare dependent on NADPH as an essential cofactor (Mayer et a!., 1989; Stuehr eta!., 1989b, 1990, 1991a,b; see review by McCall and ValIance, 1992; Marsden eta!., 1992; see review by Dinerman et aL, 1993). These results suggest that themechanism for the inhibitory effects of DPI and NSAA5 are rather different.Regarding the nature of the interaction between NADPH and FAD, it has beensuggested that NADPH suppresses the binding of DPI to the flavoprotein inneutrophil oxidase by preventing the attachment of DPI to a site in close proximityto the NADPH-binding site (Cross and Jones, 1986). It is very likely that NADPHmay interfere with the action of DPI on endothelial NOS by the same mechanism.DPI was found to inhibit ACh-induced relaxation in aortic rings for at least 4 hafter washout and to suppress ACh-induced vasodilatation for at least 2 h afterintravenous bolus injection. Therefore, our in vitro and in vivo results aresupportive of a prolonged inhibitory effect of DPI on endothelium-dependentvasodilatation. DPI has been reported to irreversibly inhibit macrophage NOS(Stuehr et a!., 1991b); the mechanism may involve the formation of a covalentbond with components of flavoprotein (Ragan and Bloxham, 1 977; O’Donnell etaL, 1993). However, our results show that post-treatment (1.5 h later) withNADPH still attenuates the effect of DPI, although the response is significantly lessthan that following pretreatment (10 mm earlier). These results imply that freshsynthesis of NO occurs in endothelial cells.924.2.2. Pressor and tachycardiac responses to DPI, and their mechanisms in vivoDetailed analyses of the dose-MAP and dose-HR response curves of DPIshow that the Hill coefficients for the MAP and HR effects of DPI are 3.6 ±0.3 and4.2±0.6, respectively, in conscious and pentobarbitone-anaesthetized rats. Theseresults suggest that the cardiovascular effects of DPI involve “positivecooperation” of probably 3 or 4 molecules of DPI (see review by Rang, 1971;Pennefather and Quastel, 1 982) and the mechanism by which DPI causes pressorresponse is different from that of NSAAs, as L-NAME cause pressor response witha Hill coefficient of 2.Moreover, the similar transient time course and pharmacodynamics suggesta common causative factor for both pressor and tachycardiac responses to DPI inpentobarbitone-anaesthetized rats. Since captopril markedly potentiated thepressor response and did not alter the tachycardiac response to DPI, it is safe toconclude that the renin-angiotensin system is not responsible for the response toDPI, although the mechanism of the potentiation by captopril is not known.By the use of sympatholytic drugs, we also investigated whether thesympathetic nervous system is responsible for the pressor and tachycardiaceffects of DPI. Reserpine markedly attenuates DPI-induced increases in MAP andHR. The results suggest that DPI causes cardiovascular effects by activating theperipheral sympathetic nerve terminals and adrenal medullae, resulting in thereleases of noradrenaline and adrenaline. This activation is dependent on thefunctional integrity of the central and autonomic nervous systems, as pithingabolishes while spinal cord transection (T1) attenuates the pressor andtachycardiac effects of DPI. The indirect activation of the sympathetic nervoussystem by DPI is further supported by the observations that TTX abolishes whileguanethidine and mecamylamine attenuate the effects of DPI. TTX has beenshown to block conductances of the central and peripheral nerves (Gage, 1971)93but not those of the vascular smooth muscle (see review by Hirst and Edwards,1989) or the myocardium (Abraham et a!., 1989), via selective blockade ofvoltage-dependent sodium channels. Guanethidine has been shown to be aspecific adrenergic neuron blocker (Shand et a!., 1 973; Kirpekar and Furchgott,1972).Furthermore, we also used bilateral adrenalectomy and chemicalsympathectomy by 6-hydroxydopamine (6-OH-DA) to examine the contributions ofthe adrenal medullae and sympathetic nerve terminals for the pressor andtachycardiac responses to DPI in pentobarbitone-anaesthetized rats. We foundthat neither bilateral adrenalectomy nor pretreatment (26 h earlier) with 6-OH-DA(4.9x10 mol/kg, i.p.) significantly affected the dose-MAP response curve of DPIalthough 6-OH-DA but not adrenalectomy slightly and significantly shifted thedose-HR curve to the right without affecting the maximum. However, thecombination of both bilateral adrenalectomy and 6-OH-DA markedly reduced themaximal MAP and maximal HR responses to DPI by 71 % and 35%, respectively.These results show that both sympathetic nerve terminals and sympathoadrenalsplay important and overlapping roles in the pressor and tachycardiac responses toDPI (Wang and Pang, 1994a).DPI causes immediate and large increases in plasma noradrenaline andadrenaline, with the same time-course as the pressor and tachycardiac responses.Pithing totally abolishes and reserpinization attenuates DPI-induced increases inplasma catecholamines as well as MAP and HR. Further analysis shows thatpositive correlations exist between DPI-induced changes in MAP and HR withplasma noradrenaline as well as adrenaline. Taken together, the above resultsindicate that DPI activates the sympathetic nerve terminals and adrenal medullaeto release noradrenaline and adrenaline in the rats with the functionally intactcentral and autonomic nervous systems. Since DPI releases large quantities ofcatecholamines, repetitive injections would lead to tachyphylaxis. Our results94indeed show that the MAP and HR responses to a single dose of DPI are greaterthan those elicited by the same dose in a multiple injection regimen. We have alsoobserved that multiple injections of high doses of DPI eventually produce negligiblepressor and tachycardiac responses (data not shown).Noradrenaline and adrenaline released by DPI would be expected to causevasoconstriction, via the activation ofx1-adrenoceptors, and tachycardia, via theactivation of1-adrenoceptors. Indeed, we found that the pressor effect of DPI issuppressed by the phentolamine and prazosin (selective x1 -adrenoceptorantagonist) but not rauwolscine (selective2-adrenoceptor antagonist). Moreover,rauwolscine and phentolamine but not prazosin enhanced the tachycardiac effectof DPI; this potentiation is likely due to the blockade of the central and/orperipheral prejunctional2-adrenoceptors which mediate inhibition of noradrenalinerelease (Berthelsen and Pettinger, 1977). This hypothesis may also explain whyrauwolscine caused a small potentiation of the pressor effect of DPI. Our resultsalso show that the tachycardiac but not pressor effect of DPI is abolished bypropranolol. The inability of propranolol to affect the MAP effect of DPI suggeststhat the pressor effect of DPI is not due to tachycardia or cardiac inotropy. Theslight potentiation of the pressor response to DPI by propranolol may be due to theblockade of vasodilator B-adrenoceptors which are prominent in skeletal musclebeds (Abdelrahman eta!., 1990).The mechanism by which DPI activates the sympathetic nervous system is notknown, however, the following possible mechanisms could be excluded. Firstly, itis logical to expect that DPI increases sympathetic discharge by inhibiting NOS.There have been extensive work done indicating that NO synthesis and releasetake place in the brain (Garthwaite et a!., 1988; Knowles et a!., 1989, 1990a;Bredt and Snyder, 1989, 1990; Bredt eta!., 1990; Schmidt et a!., 1989, 1992).NO synthesis is reported to be responsible for long-term potentiation in thehippocampus (Böhme et aL, 1991), long-term synaptic depression in the95cerebellum (Shibuki and Okada 1991) and nociceptive activity in the brain (Mooreet a!., 1991). Moreover, endothelium-derived relaxing factor (EDRF)/NO has beenshown to inhibit noradrenaline release from isolated sympathetic nervesinnervating the canine pulmonary artery and vein (Greenberg et a!., 1 989, 1 990,1991) and other preparations (see the introduction of Greenberg et a!., 1990).However, our results do not support this hypothesis. In contrast to DPI, L-NNA, ata dose which caused a maximal pressor response did not increase plasmacatecholamines. These results suggest that DPI-induced sympathetic activation isunlikely due to the inhibition of the central NO synthesis. Secondly, the effect ofDPI on the sympathetic nervous system is not due to its hypoglycemic effect,since DPI causes immediate and transient increases in blood pressure and plasmacatecholamines, while it causes hypoglycemic effect slowly and reaches theplateau at 4 h after administration (Gatley and Martin, 1979). Thirdly, the pressorresponse to DPI is not likely because of its possible activation of “pain receptor”,as we found that pretreatment with capsaicin (3.3x104 mol/kg, s.c. in 2 d)blocked DPI-induced limb kicking movements but not the pressor response inpentobarbitone-anaesthetized rats (N=6, unpublished observation). More studiesare needed to elucidate if the inhibition of other flavoproteins or NADPH-dependentenzymes account for the actions of DPI.The site(s) of the actions of DPI is not known but the central nervous systemis unlikely a primary or major site for the actions of DPI, although we can notexclude this possibility in view of the suppression of DPI’s effects by pithing andspinal cord transection. If the site of action of DPI is in the central nervoussystem, local injection of the drug then would produce greater effects thanintravenous administration. However, results from our preliminary studies showthat intracarotid and intravertebral injections of DPI caused similar pressorresponses as intravenous injections into the same rats (N=3). As well,intracerebroventricular injection of DPI into the third cerebroventricle at doses up96to 3x107 mol/kg (ED50 of 6.9x107 mol/kg by intravenous injection) did notcause any pressor or tachycardiac response (N=3). There is uncertainty aboutthe accessibility of DPI to the central nervous system. Although 1251 wasdetected in the brain 10 mm after intravenous injection of [1251] DPI (Gatley andMartin, 1979), DPI, being a charged molecule, may not adequately access thecentral nervous system within 0.5-1 mm after intravenous injection. Moreover, itis difficult to explain why pithing is more effective than spinal transection inattenuating the cardiovascular effects of DPI if the central nervous system is theonly site of action of DPI. On the other hand, we cannot rule out the possibleinvolvement of the peripheral sympathetic nervous system in the actions of DPI,since even the action of indirectly-acting sympathomimetic agents rely on afunctional amine-uptake system and therefore, sympathetic tone. Therefore,though the integrity of the central nervous system is a prerequisite for the actionsof DPI, its primary site(s) of actions is not clear. Further studies are required toidentify the site(s) of actions of DPI in the central, efferent or even afferentnervous systems.4.2.3. Mechanisms of inhibitory effect of halothane on the pressor response toDPIOur results demonstrate that halothane dose-dependently and“noncompetitively” inhibits the pressor and increases in plasma catecholaminesresponses to DPI. It should be pointed out that the inhibition by halothane is notdue to its hypotensive effects, as DPI caused greater pressor response in the ratsanaesthetized with chloralose, urethane or ethanol, where baseline blood pressureswere either similar to or lower than those in conscious rats (data shown inAppendix V). Halothane markedly inhibits DPI-induced release of catecholamine.These results suggest that the inhibitory effect of halothane on the pressor97response to DPI is primarily due to suppression of the sympathetic activation.Halothane has been shown to depress the activities of the sympathetic nervoussystem at different levels: (1) areas of the central nervous system controlling thesympathetic nerve activity (Price et al., 1 963; Miller et a!., 1 969; Larach et a!.,1987; Bazil and Minneman, 1989), (2) the sympathetic ganglia (Skovsted et al.,1969; Bosnjak et a!., 1982; Christ, 1977; Seagard et al., 1982), and (3) thesympathetic nerve endings located in the walls of blood vessels (Muldoon et a!.,1975; Lunn and Rorie, 1984; Rorie eta!., 1990).In addition, a small component of nonspecific inhibition by halothane may alsobe responsible for its effect on the pressor response to DPI. Halothane, at 1 .25%,inhibited the maximal pressor response to DPI by 95%, and inhibited the increasesin plasma noradrenaline and adrenaline by 86% and 81 %, respectively. In anotherstudy, the same concentration of halothane reduced the pressor responseproduced by exogenous noradrenaline and angiotensin II by 1 8% (data shown inAppendix VI). Therefore, the suppression by halothane of the pressor response toDPI is mainly (approximately 80%) attributable to its inhibition of sympathetictransmission, the remainder of the inhibition (approximately 20%) is due to itsnonspecific inhibition of vascular smooth muscle contraction.4.3. Summary1. DPI is an efficacious and “irreversible” inhibitor of endothelium-dependentvasodilatation in vivo and in vitro; the mechanism of the inhibition may involve theantagonism of the effects of FAD and NADPH, cofactors of NOS.2. DPI causes immediate and transient pressor and tachycardiac responses, aswell as increases in plasma catecholamines, in conscious and pentobarbitone (butnot halothane)-anaesthetized rats. These effects are inhibited by maneuvers whichinterfere with the activities of the central or sympathetic nervous systems, namely,98pithing and spinal cord transection, as well as pretreatments with TTX, reserpine,mecamylamine and guanethidine. Moreover, the pressor but not tachycardiaceffect of DPI is attenuated by phentolamine and prazosin; while the tachycardiacbut not pressor effect of DPI is inhibited by propranolol. However, the pressor andtachycardiac responses to DPI are not antagonized by L-Arg.3. These results demonstrate that DPI causes inhibition of in vitro and in vivoendothelium-dependent vasodilatation, and pressor and tachycardiac responses.However, the pressor response to DPI, unlike those of NSAAs, is via the indirectactivation of the sympathetic nervous systems, rather than the inhibition ofendothelium-dependent vasodilatation.995. General discussion and conclusions5.1. General discussionIt is well-known that all NSAAs which inhibit endothelium-dependentrelaxation in vitro and in vivo cause long-lasting pressor responses in wholeanimals. The pressor responses to NSAAs are a consequence of generalizedvasoconstriction. There are at least three existing hypotheses on the mechanismby which NSAAs cause pressor responses, namely, (1) the inhibition of endothelialNO biosynthesis (Aisaka eta!., 1989a; Rees eta!., 1989b, 1990b; see review byMoncada et a!., 1991); (2) the inhibition of brain NO synthesis (Sakuma et a!.,1992; Togashi et aL, 1992); (3) the inhibition of autonomic nerve NO synthesis(Toda et a!., 1993). Among them, the first hypothesis is best-accepted by thescientific community and it leads to the postulation that endogenous endothelialNO biosynthesis regulates vascular tone and blood pressure.Our results clearly show that the pressor response to L-NAME is not blockedby the impairment of the central nervous system by pithing, and that the pressorresponses to L-NNA and L-NAME are not attenuated by the blockade of ganglionictransmission with mecamylamine. Moreover, we also found that the pressorresponses to L-NNA are not attenuated by impairing the sympathetic nervous,angiotensin and prostanoid systems. These results, taken together, suggest thatthe pressor responses to peripherally administered NSAAs are not dependent onthe functional integrity of the central and autonomic nervous systems, and aretherefore in disagreement with hypotheses 2 and 3, which postulate that thepressor responses to NSAAs are due to the inhibition of brain and autonomic nerveNO synthesis. On the other hand, L-Arg but not D-Arg competitively antagonizesthe pressor response to L-NAME and L-NNA. Therefore, NSAAs cause pressorresponses via a novel mechanism which involves the L-Arg pathway.100The L-Arg/NO pathway is well-established in endothelial cells and otherorgans, tissues and cells (see review by Moncada et a!., 1991). It is logical toconclude that NSAAs produce pressor responses by the inhibition of endogenousendothelial NO biosynthesis. However, our results and others’ show that NSAAsonly partially inhibit endothelium-dependent vasodilatation in whole animals,whereas they completely inhibit endothelium-dependent relaxation in isolatedvascular preparations. Moreover, the pharmacodynamics of NSAAs in inhibitingendothelium-dependent relaxation and causing pressor response are different.Since NSAAs are chemically-related to L-Arg, the interaction of NSAAs and L-Argcould be simply at the membrane “receptor” level, and the signal transductionprocess may be distinct from that associated with the inhibition of NO synthesis.It is of ultimate importance to investigate the in vitro and in viva pharmacology ofother NO synthase inhibitors with structures unrelated to L-Arg.As an “irreversible” inhibitor of NOS, DPI causes prolonged inhibition ofendothelium-dependent vasodilatation in vitro and in viva, but, unlike NSAAs, itdoes not cause a prolonged pressor response. Instead, the intravenous bolusinjections of DPI only produce immediate and transient increases in MAP. Thepressor response to DPI is blocked by maneuvers which impair the activities of thecentral or sympathetic nervous systems, namely, pithing, spinal cord transectionand the administrations of TTX, reserpine, guanethidine, phentolamine andprazosin. Moreover, the pressor response to DPI but not to L-NNA is accompaniedby elevations of plasma noradrenaline and adrenaline. These results show that thetransient pressor response to DPI, unlike that of NSAAs, is solely dependent onthe activation of the sympathetic nervous system. Therefore, DPI does not elicitNO synthesis-dependent sustained rise in blood pressure as do the NSAA5.Although one may postulate that the inability of DPI to cause a sustained risein blood pressure is due to inadequate accumulation of drug in situ to inhibit NOsynthesis, this is unlikely true. DPI was shown to rapidly and adequately distribute101to all organs or tissues (Gatley and Martin, 1979). Moreover, its peakhypoglycemic effect was reached at 1 .5 h (Holland eta!., 1973) or 4 h (Gatley andMartin, 1 979) after intraperitoneal injection, suggesting a long duration of action.Our present results also show that DPI irreversibly inhibits endothelium-dependentrelaxation in vitro for more than 4 h, and partially inhibits ACh-inducedvasodilatation even at 2 h after intravenous injection.Therefore, although it is generally accepted that NSAAs produce pressorresponse by the inhibition of endothelial NO synthesis and endothelium-dependentvasodilatation in situ, and that endogenous NO modulates vascular tone and bloodpressure (Aisaka et al., 1 989a; Rëes et aL, 1 989b, 1990; see review by Moncadaet aL, 1991), our data with DPI suggest otherwise, namely, the inhibition of NOsynthesis and endothelium-dependent vasodilatation do not always causevasoconstriction in vivo. Our hypothesis is supported by other publications whichshow that methylene blue does not produce a pressor response (Loeb andLongnecker, 1 992; Pang and Wang, 1 993), although it inhibits endotheliumdependent vasodilatation in vitro (Pang and Wang, 1993) and in vivo (Loeb andLongnecker, 1992). L-NNA also caused much longer inhibition of endotheliumdependent vasodilatation than elevation of blood pressure in conscious rabbits,suggesting that the suppression of NO synthesis alone does not result inhypertension (Cocks et aL, 1992). Therefore, the hypotheses that NSAAsproduce pressor response by the inhibition of endothelial NO biosynthesis and thatendogenous NO modulates vascular tone could be challenged. Accordingly, it ispossible that NSAAs produce pressor response by activating an “L-Arg receptor”,with subsequent signal transduction distinct from the inhibition of NObiosynthesis. This hypothesis does not exclude another possibility that theinhibition of endothelial NO synthesis is a necessary but not a sufficientrequirement to cause vasoconstrictions in vivo. One requirement for the actions ofNOS inhibitors may be the presence of chemical structures similar to those of102NSAAs or Arg. Another necessary condition may be a particular type ofvasculature. It has been shown that NSAAs inhibit endothelium-dependentrelaxation in both isolated arterial and venous preparations (Gold et aL, 1990;Miller, 1 991; Nagao and Vanhoutte, 1 991; Pawloski and Chapnick, 1 991; Martineta!., 1992; Elmore et aL, 1992). Although L-NMMA has no effect on the basaltone of dorsal hand veins, it inhibits ACh-elicited venodilatation, however, LNMMA induces direct vasoconstriction and inhibits vasodilatations induced by AChand bradykinin in the brachial artery (Valiance et a!., 1989a,b). Moreover, LNAME causes a marked increase in arterial pressure but not in mean circulatoryfilling pressure (MCFP, an index of body venous tone) in intact or ganglion-blockedconscious rats (Wang and Pang, unpublished data, 1993), or a low increase inMCFP (by approximately 12%) in anaesthetized cats (Bower and Law, 1993).These results again suggest that the inhibition of endothelium-dependent relaxationalone is insufficient to cause vasoconstriction in vivo.5.2. Conclusions1. Both NSAAs and DPI inhibit endothelium-dependent vasodilatations, completelyin vitro, and partially in vivo. However, the inhibition by NSAAs is reversiblewhereas that by DPI is irreversible. The inhibitory mechanism of NSAAs involveslimiting the availability of the NOS substrate L-Arg, whereas that of DPI involvesantagonizing the actions of the NOS cofactors, FAD and NADPH.2. Although both NSAAs and DPI cause pressor responses, the time course of theresponses are different. The responses to NSAAs are slowly-developing and longlasting whereas that of DPI is immediate and transient. Moreover, the pressorresponses to NSAAs are accompanied by bradycardia and that of DPI isaccompanied by tachycardia.1033. The pressor responses to NSAAs are competitively antagonized by L-Arg andare not dependent on the integrity of the central/autonomic nervous, angiotensinand prostanoid systems. Therefore, the pressor responses to NSAAs are not dueto the inhibition of NO biosynthesis in the brain or autonomic nervous system.However, the pressor and tachycardiac responses to DPI are entirely due tosympathetic activation, rather than the inhibition of endothelium-dependentvasodilatation.4. Although the mechanism by which NSAAs produce pressor responses is noveland is antagonized by L-Arg, it may not be related to the inhibition of NObiosynthesis. This hypothesis is derived on the basis of the following evidence.(1) DPI, as an effective inhibitor of in vitro and in vivo endothelium-dependentvasodilatations, does not cause NO-mediated pressor response. (2) NSAAs andDPI do not completely inhibit ACh-induced depressor responses in vivo. (3) the nof .NSAAs for producing pressor response is different from that for inhibitingendothelium-dependent relaxation. (4) methylene blue, a putative inhibitor ofguanylyl cyclase, inhibits ACh-induced relaxation in preconstricted aortic rings butdoes not cause pressor response. It may be reasonable to postulate that NSAA5produce pressor response via interaction with “L-Arg receptors”, with subsequentsignal transduction different from that associated with the inhibition of NObiosynthesis.5. Alternatively, the pressor response produced by the NSAAs may involve theinhibition of NOS, however, this only condition may not be sufficient to causevasoconstriction in vivo. Pressor response by NOS inhibitors may require: (1) theinhibition of endogenous endothelial NO biosynthesis in arteries and arterioles, and(2) chemical structures of NOS inhibitors to be similar to those of NSAAs or Arg.1046. BibliographyAbdelrahman, A., Tabrizchi, R. and Pang, C.C.Y.: Effects of B- and B2-adrenoceptor stimulation on hemodynamics in the anesthetized rat. J.Cardiovasc. Pharmacol. 15: 720-728, 1990.Abdelrahman, A., Wang, Y.-X., Chang, S.D. and Pang, C.C.Y.: Mechanism of thevasodilator action of calcitonin gene-related peptide in conscious rats. Br. J.Pharmacol. 106: 45-48, 1 992a.Abdelrahman, A., Wang, Y.-X., Pang, C.C.Y.: Effects of anaesthetic agents onpressor response to B-blockers in the rat. J. Pharm. Pharmacol. 44: 34-38,1992b.Abraham, S., Beatch, G.N., MacLeod, B.A. and Walker, M.J.A.: Antiarrhythmicproperties of the tetrodotoxin against occusion-induced arrhythmias in therats: a novel approach to the study of the antiarrhythmic effects ofventricular sodium channel blockade. J. Pharmacol. Exp. Ther. 251: 1161-1173, 1989.Ahmad, S., Chretien, P., Daniel, E.E. and Shen, S.H.: Characterization of Badrenoceptors on cultured endothelial cells by radio ligand binding. Life Sci.47: 2365-2370, 1990.Aisaka, K., Gross, S.S., Griffith, 0.W. and Levi, R.: NGmethylarginine, an inhibitorof endothelium-derived nitric oxide synthesis, is a potent pressor agent in theguinea pig: does nitric oxide regulate blood pressure in vivo? Biochem.Biophys. Res. Commun. 160: 881-886, 1989a.Aisaka, K., Gross, S.S., Griffith, 0.W. and Levi, R.: L-arginine availabilitydetermines the duration of acetylcholine-induced systemic vasodilation invivo. Biochem. Biophys. Res. Commun. 163: 710-71 7, 1989b.Aisaka, K., Mitani, A., Kitajima, Y., Ohno, T. and Ishihara, T.: Difference in pressorresponses to NGmonomethylLarginine between conscious and anesthetizedrats. Jpn. J. Pharmacol. 56: 245-248, 1 991.Amezcua, J.L., Dusting, G.J., Palmer, R.M.J. and Moncada, S.: Acetylcholineinduces vasodilatation in the rabbit isolated heart through the release of nitricoxide, the endogenous nitrovasodilator. Br. J. Pharmacol. 95: 830-834,1988.Amezcua, J.L., Palmer, R.M.J., de Souza, B.M. and Moncada, S.: Nitric oxidesynthesized from L-arginine regulates vascular tone in the coronarycirculation of the rabbit. Br. J. Pharmacol. 97: 111 9-1 1 24, 1 989.105Archer, S.L. and Hampl, V.: NGmonomethylLarginine causes nitric oxidesynthesis in isolated arterial rings: trouble in paradise. Biochem. Biophys.Res. Commun. 188: 590-596, 1992.Ariëns, E.J.: Stereoselectivity of bioactive agents: general aspects. In: Ariëns,E.J., Soudijn, W. and Timmermans, P.B.M.W.M. (eds). Stereochemstry andbiological activity of drugs. Oxford: Blackwell Scientific Publications, pp 11-32, 1983.Azuma, H., Ishikawa, M., Gryglewgki, R.J.: Mechanism of action of someinhibitors of endothelium-derived relaxing factor. Proc. NatI. Acad. Sci.U.S.A. 83: 9164-91 68, 1986.Bazil, C.W. and Minneman, K.P.: Clinical concentrations of volatile anestheticsreduce depolarization-evoked release of [3H] norepinephrine, but not [3H]acetylcholine, from rat cerebral cortex. J. Neurochem. 53: 962-965, 1989.Beierwaltes, W.H. and Carretero, O.A.: Nonprostanoid endothelium-derived factorsinhibit renin release. Hypertension 19(2 Suppl): 1168-1173, 1992.Bellan, J.A., Minkes, R.K., McNamara, D.B. and Kadowitz, P.J.: N°-nitro-L-arginineselectively inhibits vasodilator responses to acetylcholine and bradykinin incats. Am. J. Physiol. 260: H1025-H1029, 1991.Berthelsen, S. and Pettinger, W.A.: A functional basis for classification of alphaadrenergic receptors. Life Sci. 21: 2131-2138, 1977.Bogle, R.G., Baydoun, A.R., Pearson, J.D., Moncada, S. and Mann GE.: L-argininetransport is increased in macrophages generating nitric oxide. Biochem. J.284: 15-18, 1992a.Bogle, R.G., Coade, S.B., Moncada, S., Pearson, J.D. and Mann, G.E. Bradykininand ATP stimulate L-arginine uptake and nitric oxide release in vascularendothelial cells. Biochem. Biophys. Res. Commun. 1 80: 926-932, 1 991.Bogle, R.G., Moncada, S., Pearson, J.D. and Mann, G.E.: Identification ofinhibitors of nitric oxide synthase that do not interact with the endothelialcell L-arginine transporter. Br. J. Pharmacol. 105: 768-770, 1992b.Böhme, G., Bon, C., Stutzmann, J.-M., Doble, A. and Blanchard, J.-C.: Possibleinvolvement of nitric oxide in long-term potentiation. Eur. J. Pharmacol.199: 379-381, 1991.Bosnjak, Z.J., Seagard, J.L., Wu, A. and Kampine, J.P.: The effects of halothaneon sympathetic ganglionic transmission. Anesthesiology 57: 473-479,1982.106Bower, E.A. and Law, A.C.K.: The effects of N°-nitro-L-arginine methyl ester,sodium nitroprusside and noradrenaline on venous return in the anaesthetizedcat. Br. J. Pharmacol. 108: 933-940, 1993.Bredt, D.S. and Snyder, S.H.: Isolation of nitric oxide synthetase, a calmodulinrequiring enzyme. Proc. Nati. Acad. Sci. U.S.A. 87: 682-685, 1990.Bredt, D.S. and Snyder, S.H.: Nitric oxide mediates glutamate-linked enhancementof cGMP levels in the cerebellum. Proc. Nati. Acad. Sd. U.S.A. 86: 9030-9033, 1989.Bredt, D.S., Ferris, C.D. and Snyder, S.H.: Nitric oxide synthase regulatory sites.Phosphorylation by cyclic AMP-dependent protein kinase, protein kinase Cand calcium/calmodulin protein kinase; identification of flavin and calmodulinbinding sites. J. Biol. Chem. 267: 10976-10981, 1992.Bredt, D.S., Hwang, P.M. and Snyder, S.H.: Localization of nitric oxide synthaseindicating a neural role for nitric oxide. Nature 347: 768-770, 1 990.Bredt, D.S., Hwang, P.M., Glatt, C.E., Lowenstein, C., Reed, R.R. and Snyder,S.H.: Cloned and expressed nitric oxide synthase structurally resemblescytochrome P-450 reductase. Nature 351: 714-718, 1991.Burrn, J.H. and Robinson, J.: Reversal of the vascular response to acetylcholineand adrenaline. Br. J. Pharmacol. 6: 110-119, 1951.Busse, R.: Stimulation of soluble guanylate cyclase activity by endothelium-derivedrelaxant factor, a general principle of its vasodilator and antiaggregatoryproperties. Thomb. Res. (suppl): 3, 1987.Buxton, I.L.0., Cheek, D.J., Eckman, D., Westfall, D.P., Sanders, K.M. and Keef,K.D.: NGnitro L-arginine methyl ester and other alkyl esters of arginine aremuscarinic receptor antagonists. Circ. Res. 72: 387-395, 1 993.Chen, G., Suzuki, H. and Weston, A.: Acetylcholine releases endothelium-derivedhyperpolarizing factor and EDRF from rat blood vessels. Br. J. Pharmacol.95: 1165-1174, 1988.Chen, G.E. and Suzuki, H.: Calcium dependency of the endothelium-dependenthyperpolarization in smooth muscle cells of the rabbit carotid artery. J.Physiol. (Lond). 421: 521-534, 1990.Chen, W.Z., Palmer, R.M.J. and Moncada, S.: The release of nitric oxide from therabbit aorta. J. Vasc. Med. Biol. 1: 1-6, 1989.107Cherry, P.D., Furchgott, R.F., Zawadzki, J.V., Jothianandan, D.: Role ofendothelial cells in relaxation of isolated arteries by bradykinin. Proc. Nati.Acad. Sci. U.S.A. 79: 2106-2110, 1982.Chillon, J.M., Capdeville-Atkinson, C., Lartaud, I., Guillou, J., Martes, P.M. andAtkinson, J.: Chronic antihypertensive treatment with captopril plushydrochiorothiazide improves aortic disability in the spontaneouslyhypertensive rat. Br. J. Pharmacol. 107: 710-714, 1992.Christ, D.: Effects of halothane on ganglionic discharges. J. Pharmacol. Exp.Ther. 200: 336-342, 1977.Chu, A., Chambers, D.E., Lin, C.C., Kuehl, W.D. and Cobb, F.R.: Nitric oxidemodulates epicardial coronary basal vasomotor tone in awake dogs. Am. J.Physiol. 258: H1250-H1254, 1990.Chu, A., Chambers, D.E., Lin, C.C., Kuehl, W.D., Palmer, R.M.J., Moncada, S.and Cobb, F.R.: Effects of inhibition of nitric oxide formation on basalvasomotion and endothelium-dependent responses of the coronary arteries inawake dogs. J. Clin. Invest. 87: 1964-1968, 1991.Chyu, K.Y., Guth, P.H. and Ross, G.: Effect of N-nitro-L-arginine methyl ester onarterial pressure and on vasodilator and vasoconstrictor responses: influenceof initial vascular tone. Eur. J. PharmacoI. 212: 159-1 64, 1992.Cocks, T., Ward, J., Crack, P., Stork, A. and Angus, J.: Comparison, of theeffects of the endothelium-derived relaxing factor substrate inhibitors NGnitro-L-arginine and NGmonomethylLarginine in vivo, in vitro and incultured endothelial cells. In: Moncada, S. Marietta, M.A. Hibbs, J.B. Jr. andHiggs, E.A. (eds). The Biology of Nitric Oxide. 1. Physiological and ClinicalAspects. London: Portland, 78-79, 1 992.Cocks, T.M. and Angus, J.A.: Evidence that contractions of isolated arteries by LNMMA and NOLA are not due to inhibition of basal EDRF release. J.Cardiovasc. Pharmacol. 17 (Suppl.3): S159-S164, 1991.Corbett, J.A., Tilton, R.G., Chang, K., Hasan, K.S., Ido, Y., Wang, J.L.,Sweetland, M.A., Lancaster, J.R. Jr. Williamson, J.R., and McDaniel, M.L.:Aminoguanidine, a novel inhibitor of nitric oxide formation, prevents diabeticvascular dysfunction. Diabetes 41: 552-556, 1 992.Crack, P. and Cocks, T.: Thimerosal blocks stimulated but not basal release ofendothelium-derived relaxing factor (EDRF) in dog isolated coronary artery.Br. J. Pharmacol. 107: 566-572, 1992.Craven, P.A. and DeRubertis, F.R.: Restoration of the responsiveness of purifiedguanylate cyclase to nitrosoguanidine, nitric oxide, and related activators by108heme and hemeproteins. Evidence for involvement of the paramagneticnitrosyl-heme complex in enzyme activation. J. Biol. Chem. 253: 8433-8443, 1978.Crawley, D.E., Liu, S.F., Evans, T.W. and Barnes, P.J.: Inhibitory role ofendothelium-derived relaxing factor in rat and human pulmonary arteries. Br.J. Pharmacol. 101: 166-1 70, 1990.Cross, A.R. and Jones, O.T.G.: The effect of the inhibitor diphenylene iodonium onthe superoxide-generating system of neutrophils. Biochem. J. 237: 111-11 6,1986.Dale, H.H.: The action of certain esters and esters of choline, and their relation tomuscarine. J. Pharmacol. Exp. Ther. 6: 147-1 90, 1914.Diamond, J. and Chu, E.B.: Possible role for cyclic GMP in endothelium-dependentrelaxation of rabbit aorta by acetylcholine. Comparison with nitroglycerin.Res. Commun. Chem. Pathol. Pharmacol. 41: 369-381, 1983.Dinerman, J.L., Lowenstein, C. and Snyder, S.H.: Molecular mechanism of nitricoxide regulation. potential relevance to cardiovascular disease. Circ. Res.73: 21 7-222, 1993.Ellis, J., Mayer, S.I. and Jones, O.T.G.: The effect of the NADPH oxidase inhibitordiphenyleneiodonium on aerobic and anaerobic microbicidal activities ofhuman neutrophils. Biochem. J. 251: 887-891, 1988.Ellis, J.A., Cross, A.R. and Jones, T.G.: Studies on the electron-transfermechanism of the human neutrophil NADPH oxidase. Biochem. J. 262: 575-579, 1989.Elmore, J.R., Gloviczki, P. Brockbank, K.G. and Miller, M.N.: Functional changes incanine saphenous veins after cryopreservation. mt. Angio. 11: 26-35, 1992.EIsner, D., Muntze, A., Kromer, E.P. and Riegger, G.A.: Inhibition of synthesis ofendothelium-derived nitric oxide in conscious dogs. Haemodynamic, renal,and hormonal effects. Am. J. Hypertens. 5: 288-291, 1 992.Evans, T., Carpenter, A. and Cohen, J.: Purification of a distinctive form ofendotoxin-induced nitric oxide synthase from rat liver. Proc. NatI. Acad. Sci.U.S.A. 89: 5361-5365, 1992).Fineman, J.R., Heymann, M.A. and Soifer, S.J.: N°-nitro-L-arginine attenuatesendothelium-dependent pulmonary vasodilation in lambs. Am. J. Physiol.260: H1299-H1306, 1991.109Forstermann, U. and Neufang, B.: The endothelium-dependent vasodilator effect ofacetylcholine: characterization of the endothelial relaxing factor withinhibition of arachidonic acid metabolism. Eur. J. Pharmacol. 103: 65-70,1984.Fortes, Z.B., Oliveira, M.A., Scivoletto, R., Carvalho, M.H., De Nucci, G. andNigro, D.: Nitric oxide release may be involved in the microcirculatoryresponse to acetylcholine. Eur. J. Pharmacol. 182: 143-1 47 1990.Fukuto, J.M., Wood, K.S., Byrns, R.E. and Ignarro, L.J.: NGaminoLarginine: anew potent antagonist of L-arginine-mediated endothelium-dependentrelaxation. Biochem. Biophys. Res. Commun. 168: 458-465, 1990.Furchgott R.F. and Vanhoutte P.M.: Endothelium-derived relaxing and contractingfactors. FASEB. J. 3: 2007-201 8, 1989.Furchgott, R.F. and Zawadzki, J.V.: The obligatory role of endothelial cells in therelaxation of arterial smooth muscle by acetyicholine. Nature 288: 373-376,1980.Furchgott, R.F.: Role of endothelium in responses of vascular smooth muscle.Circ. Res. 53: 557-573, 1983.Furchgott, R.F.: Studies on relaxation of rabbit aorta by sodium nitrite: the basisfor the proposal that the acid-activatable factor from bovine retractor penis isinorganic nitrite and the endothelium-derived relaxing factor. In: Vanhoutte,P.M. (ed). Mechanisms of Vasodilatation. New York: Raven Press, 401 -41 4,1988.Furchgott, R.F.: The pharmacology of vascular smooth muscle. Pharmacol. Rev.7: 183-266, 1955.Gage, P.W.: Tetrodotoxin and saxitoxin as pharmacological tools. In: Simpson,L.L. (ed). Neuropoisons: Their Pathophysiological Actions. New York:Plenum Press, p 1 87, 1 971.Gage, P.W.: Tetrodotoxin and saxitoxin as pharmacological tools. In: Simpson,L.L. (ed). Neuropoisons: Their Pathophysiological Actions. New York:Plenum Press, p 187, 1971.Garcia, J.L., Fernandez, N., Garcia-Villalon, A.L., Monge, L., Gomez, B. andDieguez, G.: Effects of nitric oxide synthesis inhibition on the goat coronarycirculation under basal conditions and after vasodilator stimulation. Br. J.Pharmacol. 106: 563-567, 1992.Gardiner, S.M., Compton, A.M., Bennett, T., Palmer, R.M.J. and Moncada, S.:Regional haemodynamic changes during oral ingestion of NGmonomethylL110arginine or NGnitroLarginine methyl ester in conscious Brattleboro rats. Br.J. Pharmacol. 101: 10-12, 1990a.Gardiner, S.M., Compton, A.M., Bennett, T., Palmer, R.M.J. and Moncada, S.:Control of regional blood flow by endothelium-derived nitric oxide.Hypertension 1 5: 486-492, 1 990b.Gardiner, S.M., Compton, A.M., Kemp, P.A. and Bennett T.: Regional and cardiachaemodynamic responses to glyceryl trinitrate, acetylcholine, bradykinin andendothelin-1 in conscious rats: effects of NGnitroLarginine methyl ester.Br. J. Pharmacol. 101: 632-639, 1990c.Gardiner, S.M., Compton, A.M., Kemp, P.A. and Bennett, T.: Regional and cardiachaemodynamic effects of NG-nitro-L-arginine methyl ester in conscious, LongEvans rats. Br. J. Pharmacol. 101: 625-631, 1990d.Gardiner, S.M., Kemp, P.A. and Bennett, T.: Effects of NGnitroLarginine methylester on vasodilator responses to acetylcholine, 5’-N-ethylcarboxamidoadenosine or salbutamol in conscious rats. Br. J.Pharmacol. 103: 1725-32, 1991.Garland, C.J. and McPherson, G.A.: Evidence that nitric oxide does not mediatethe hyperpolarization and relaxation to acetyicholine in the rat smallmesenteric artery. Br. J. Pharmacol. 105: 429-435, 1992.Garthwaite, J., Southam, E. and Anderton, M.: A kainate receptor linked to nitricoxide synthesis from L-arginine. J. Neurochem. 33: 1952-1 954, 1988.Gatley, S.J. and Martin, J.L.: Some aspects of the pharmacology ofdiphenyleionium, a bivalent iodine compound. Xenobiotica 9: 539-546,1979.Gold, M.E., Wood, K.S., Byrns, R.E., Fukuto, J. and Ignarro, L.J.: NG-methyl-Larginine causes endothelium-dependent contraction and inhibition of cyclicGMP formation in artery and vein. Proc. NatI. Acad. Sci. U.S.A. 87: 4430-4434, 1990.Greenberg, S.S., Diecke, F.P., Peevy, K. and Tanaka, T.P.: The endotheliummodulates adrenergic neurotransmission to canine pulmonary arteries andveins. Eur. J. Pharmacol. 162: 67-80, 1989.Greenberg, S.S., Diecke, F.P., Peevy, K. and Tanaka, TP.: Release ofnorepinephrine from adrenergic nerve endings of blood vessels is modulatedby endothelium-derived relaxing factor. Am. J. Hypertens. 3: 211-218,1990.111Greenberg, S.S., Peevy, K. and Tanaka, T.P.: Endothelium-derived andintraneuronal nitric oxide-dependent inhibition of norepinephrine efflux fromsympathetic nerves by bradykinin. Am. J. Hypertens. 4: 464-467, 1 991.Griffith, T.M., Edwards, D.H., Lewis, M.J., Newby, A.C. and Henderson, A.H.:The nature of endothelium-derived vascular relaxant factor. Nature 308:645-647, 1984.Gross, S.S., Stuehr, D.J., Aisaka, K., Jaffe, E.A., Levi, R. and Griffith, 0.W.:Macrophage and endothelial cell nitric oxide synthesis: cell-type selectiveinhibition by NGaminoarginine, NGnitroarginine and NGmethylarginine.Biochem. Biophys. Res. Commun. 170: 96-103, 1990.Gruetter, C.A., Barry, B.K., McNamara, D.B., Gruetter, D.Y., Kadowitz, P.J. andlgnarro, L.J.: Relaxation of bovine coronary artery and activation of coronaryguanylate cyclase by nitric oxide, nitroprusside and a carcinogenicnitrosomine. J. Cyclic Nucleotide Re. 5: 211-224, 1979.Gruetter, C.A., Cruetter, D.Y., Lyon, J.E., Kadowitz, P.L. and Ignarro, L.J.:Relationship between cyclic guanosine 3’, 5’-monophosphate formation andrelaxation of coronary arterial smooth muscle by glyceryl trinitrate,nitroprusside, nitrite and nitric oxide: effects of methylene blue andmethyhemoglobin. J. Pharmacol. Exp. Ther. 219: 181-1 86, 1981.Hancock, J.T. and Jones, T.G: The inhibition by diphenyleneionium and itsanalogues of superoxide generation by macrophages. Biochem. J. 242: 103-107, 1987.Harada, S., Tokunaga, S. Momohara, M., Masaki, H., Tagawa, T., Imaizumi, T.and Takeshita, A.: Inhibition of nitric oxide formation in the nucleus tractussolitarius increases renal sympathetic nerve activity in rabbits. Circ. Res. 72:511-516, 1993.Hevel, J.M, White, K.A. and MarIetta, M.A.: Purification of the inducible murinemacrophage nitric oxide synthase: Identification as a flavoprotein. J. Biol.Chem., 266: 22789-22791, 1991.Hibbs, J.B. Jr., Taintor, R.R. and Vavrin, Z.: Macrophage cytotoxicity: role for Larginine deiminase and imino nitrogen oxidation to nitrite. Science 235: 473-476, 1987a.Hibbs, J.B. Jr., Vavrin, Z. and Taintor, R.R.: L-arginine is required for expressionof the activated macrophage etfector mechanism causing selective metabolicinhibition in target cells. J. Immunol. 138: 550-565, 1987b.Hiki, K., Hattori, R. Kawai, C. and Yui, Y.: Purification of insoluble nitric oxidesynthase from rat cerebellum. J. Biochem. (Tokyo) 111: 556-558, 1 992.112Hirst, G.D.S. and Edwards, F.R.: Sympathetic neuroeffector transmission inarteries and arterioles. Physiol. Rev. 69: 546-604, 1 989.Hobbs, A.J. and Gibson, A.: LNGnitroLarginine and its methyl ester are potentinhibitors of non-adrenergic, non-cholinergic transmission in the ratanococcygeus. Br. J. Pharmacol. 100: 749-752, 1990.Holland, P.C., Clark, M.C., Bloxham, D.P and Lardy, H.: Mechanism of action ofthe hypoglycemic agent diphenyleneiodonium. J. Biol. Chem. 17: 6050-6056, 1973.Howell, R.E., Albelda, S.M., Daise, M.L. and Levine, E.M.: Characterization of Badrenergic receptors in cultured human and bovine endothelial cells. J. AppI.Physiol. 65: 1251-1 257, 1988.Hu, Z.-W., Billingham, M., Tuck, M. and Hoffman, B.: Captopril improveshypertension and cardiomyopathy in rats with pheochromocytoma.Hypertension 15: 210-21 5, 1990.Humphries, R.G., Carr, R.D., Nicol, A.K., Tomlinson, W. and O’Connor, S.E.:Coronary vasoconstriction in the conscious rabbit following intravenousinfusion of LNGnitroarginine. Br. J. Pharmacol. 102: 565-566, 1991.Hunt, R.: Some physiological actions of the homocholines and some of theirderivatives. J. Pharmacol. Exp. Ther. 6: 477-525, 1915.Hutchinson, P.J., Palmer, R.M.J. and Moncada, S.: Comparative pharmacology ofEDRF and nitric oxide on vascular strips. Eur. J. Pharmacol. 141: 445-451,1987.Hutt, A.J. and CaIdwell, J.: The metabolic chiral inversion of 2-arylpropionic acids:a novel route with pharmacological consequences. J. Pharm. Pharmacol. 35:693-704, 1983.Ignarro L.J.: Endothelium-derived nitric oxide: actions and properties. FASEB. J.3: 31-36, 1989.Ignarro L.J.: Nitric oxide: a novel signal transduction mechanism for transcellularcommunication. Hypertension 1 6: 477-483, 1 990.Ignarro, L.J. and Kadowitz, P.J.: The pharmacological and physiological role ofcyclic GMP in the vascular smooth muscle relaxation. Annu. Rev.Pharmacol. Toxicol. 25: 171-1 91, 1985.Ignarro, L.J., Buga, G.M., Wood, K.S., Byrns, R.E. and Chaudhuri, G.:Endothelium-derived relaxing factor produced and released from artery andvein is nitric oxide. Proc. Nail. Acad. Sci. U.S.A. 84: 9265-9269, 1987a.113Ignarro, L.J., Burke, T.M., Wood, K.S., Wolin, M.S. and Kadowitz, P.J.:Association between cyclic GMP accumulation and acetyicholine-elicitedrelaxation of bovine intrapulmonary artery. J. Pharmacol. Exp. Ther. 228:682-690, 1984.lgnarro, L.J., B.yrns, R.E., Buga, G.M. and Wood, K.S.: Endothelium-derivedrelaxing factor from pulmonary artery and vein possesses pharmacologicaland chemical properties that are identical to those for nitric oxide radical.Circ Res. 61: 866-879, 1987b.Ignarro, L.J., Byrns, R.E., Wood, K.S.: Biochemical and pharmacological propertiesof endothelium-derived relaxing factor and its similarity to nitric oxide radical.In: Vanhoutte, P.M. (ed). Mechanisms of Vasodilatation. New York: RavenPress, 427-435, 1988.Ignarro, L.J., Harbison, R.G., Wood, K.S. and Kadowitz, P.J.: Activation ofpurified soluble guanylate cyclase by endothelium-derived relaxing factorfrom intrapulmonary artery and vein: stimulation by acetylcholine, bradykininand arachidonic acid. J. Pharmacol. Exp. Ther. 237: 893-900, 1 986.Ignarro, L.J., Lippton, H., Edwards, J.C., Baricos, W.H., Hyman, A.L., Kadowitz,P.J. and Grueter, C.A.: Mechanism of vascular smooth muscle relaxation byorganic nitrates, nitrites, nitroprusside and nitric oxide: evidence for theinvolvement of S-nitrosothiols as active intermediates. J. Pharmacol. Exp.Ther. 218: 739-749, 1981.Ignarro, L.J.: Pharmacological, biochemical and chemical evidence that EDRF is NOor Labile nitro precursor. In: Ryan, U. and Rubanyi, G.M. (eds). EndothelialRegulation of Vascular Tone. New York: Marcel Dekker, Inc., 37-50, 1992.lshii, K., Chang, B., Kerwin, J.F. Jr., Huang, Z.J. and Murad, F.: N°-nitro-Larginine: a potent inhibitor of endothelium-derived relaxing factor formation.Eur. J. Pharmacol. 176: 219-223, 1990a.lshii, K., Kerwin, J.F. Jr. and Murad, F.: N°-nitro-L-arginine: a potent inhibitor ofthe L-arginine-dependent soluble guanylate cyclase activation pathway inLLC-PK1 cells. Can. J. Physiol. Pharmacol. 68: 749-751, 1990b.Iyengar, R., Stuehr, D.J., MarIetta, M.A.: Macrophage synthesis of nitrite, nitrate,and N-nitrosamines: precursor and role of the respiratory burst. Proc. NatI.Acad. Sd. U.S.A. 84: 6369-6373, 1987.Jonsson, J.R., Frewin, D.B. and Head, R.J.: Chronic captopril treatment reversesthe enhanced vascular concentration of 3 methylhistidine •in thespontaneously hypertensive rat. Blood vessels 28: 41 3-41 9, 1 991.114Kakimoto, Y. and Akazawa, S.: Isolation and identification of NG,NG and NG,N’Gdimethyl-arginine, NGmono, di, and trimethylysine, and glucosylgalactosyland galactol-c-hydroxylysine from human urine. J. Biol. Chem. 245: 5751-5758, 1970.Katsuki, S., Arnold, W., Mittal, C. and Muard, F.: Stimulation of guanylate cyclaseby sodium nitroprusside, nitroglycerin and nitric oxide in various tissuepreparations and comparison to the effect of sodium azide andhydroxylamine. J. Cyclic Nucleotide, Res. 3: 23-25, 1977.Keller, R., Bassetti, S., Keist, R. Mülsch, A. and Klauser, S.: Induction of nitricoxide synthase is a necessary precondition for expression of tumor necrosisfactor-independent tumoricidal activity by activated macrophages. Biochem.Biophys. Res. Commun. 1.4: 1364-1 371, 1992.KeIm, M. and Schrader, J.: Nitric oxide release from the isolated guinea pig heart.Eur. J. Pharmacol. 155: 317-321, 1988.Keim, M., Feelisch, M., Spahr, R., Piper, H.M., Noack, E. and Schrader, J.:Quantitative and kinetic characterization of nitric oxide and EDRF releasedfrom cultured endothelial cells. Biochem. Biophys. Res. Commun. 154: 236-244, 1988.Kenakin, T.: Pharmacologic analysis of drug-receptor interaction. New York:Raven Press, pp 129-244, 1987.Kilbourn, R.G. and Belloni, P.: Endothelial cell production of nitrogen oxides inresponse to interferon gamma in combination with tumor necrosis factor,interleukin-1, or endotoxin. J. NatI. Cancer. Inst. 82: 772-776, 1990.Kilbourn, R.G., Gross, S.S., Lodato, R.F., Adams, J., Levi, R., Miller, L.L.,Lachman, L.B. and Griffith, Q.W.: Inhibition of interleukin-1-alpha-inducednitric oxide synthase in vascular smooth muscle and full reversal ofinterleu kin-i -alpha-induced hypotension by N°-amino-L-arginine. J. NatI.Cancer Inst. 84: 1008-101 6, 1992.Kirpekar, S.M. and Furchgott, R.F.: Interaction of tyramine and guanethidine in thespleen of the cat. J. Pharmacol. Exp. Ther. 1 80: 38-46, 1 972.Klabunde, R.E., Ritger, R.C. and Helgren, M.C.: Cardiovascular actions of inhibitorsof endothelium-derived relaxing factor (nitric oxide) formation/release inanesthetized dogs. Eur. J. Pharmacol. 199: 51-59, 1991.Knowles, R.G., Palacios, M., Palmer, R.M.J. and Moncada, S.: Formation of nitricoxide from L-arginine in the central nervous system: transduction mechanismfor stimulation of the soluble guanylate cyclase. Proc. NatI. Acad. Sci.U.S.A. 86: 51 59-5162, 1989.115Knowles, R.G., Palacios, M., Palmer, R.M.J. and Moncada, S.: Kineticcharacteristics of nitric oxide synthase from rat brain. Biochem. J. 269:207-210, 1 990a.Knowles, R.G., Salter, M., Brooks, S.L. and Moncada, S.: Anti-inflammatoryglucocorticoids inhibit the induction by endotoxin of nitric oxide synthase inthe lung, liver and aorta of the rat. Biochem. Biophys. Res. Commun. 172:1042-1048, 1990b.Kobayashi, Y. and Hattori, K.: Nitroarginine inhibits endothelium-derived relaxation.Jpn. J. Pharmacol. 52: 167-1 69, 1990.Kottecoda, S.R.: The action of nicotine and acetylcholine on the vessels of therabbit’s ear. Br. J. Pharmacol. 8: 156-161, 1953.Kukovetz, W.R., Holzmann, S. and Poch, G.: Function of cyclic GMP inacetylcholine-induced contraction of coronary smooth muscle. Naunyn.Schmiedebergs Arch. Pharmacol. 31 9; 29-33, 1 982.Kwon, N.S., Nathan, C.F. and Stuehr, D.J.: Reduced biopterin as a cofactor in thegeneration of nitrogen oxides by murine macrophages. J. Biol. Chem. 264:20496-20501, 1989.Kwon, N.S., Stuehr, D.J. and Nathan, C.F.: Inhibition of tumor cell ribonucleotidereductase by macrophage-derived nitric oxide. J. Exp. Med. 174: 761-767,1991.Lacolley, P.J., Lewis, S.J. and Brody, M;J.: Role of sympathetic nerve activity inthe generation of vascular nitric oxide in urethane-anesthetized rats.Hypertension 17: 881-887, 1991.Lahera, V., Salom, M.G., Fiksen-Olsen, M.J., Raij, L. and Romero, J.C.: Effects ofNGmonomethylLarginine and L-arginine on acetyicholine renal response.Hypertension 1 5: 659-663, 1 990.Lamas, S., Marsden, P.A., Li, G.K., Tempst, P., Michel, T.: Endothelial nitric oxidesynthase: molecular cloning and characterization of a distinct constitutiveenzyme isoform. Proc. NatI. Acad. Sd. U.S.A. 89: 6348-6352, 1992.Lamontagne, D., Pohi, U. and Busse, R.: NGnitroLarginine antagonizesendothelium-dependent dilator responses by inhibiting endothelium-derivedrelaxing factor release in the isolated rabbit heart. Pflugers. Arch. 418: 266-70, 1991.Larach, D.R., Schuler, H.G., Derr, J.A., Larach, M.G., Hensley, F.A. and Zelis, R.:Halothane se’ectively attenuatesx2-adrenoceptor mediated vasoconstrictionin vivo and in vitro. Anesthesiology 66: 781-789, 1987.116Lautt, W.W.: Control of hepatic arterial blood flow: independence from livermetabolic activity. Am. J. Physiol. 239: H559-H564, 1980.Loeb, A.L. and Longnecker, D.E.: inhibition of endotheliun-derived relaxing factor-dependent circulatory control in intact rats. Am. J. Physiol. 262: Hi 1494-Hi 500, 1992.Lowenstein, C.J., Glatt, C.S., Bredt, D.S. and Snyder, S.H.: Cloned and expressedmacrophage nitric oxide synthase contrasts with the brain enzyme. Proc.NatI. Acad. Sd. U.S.A. 89: 6711-671 5, 1992.Lunn, J.J. and Rorie, D.K.: Halothane-induced changes in the release anddisposition of norepinephrine at adrenergic nerve endings in dog saphenousvein. Anesthesiology 61: 377-384, 1 984.Lüscher, T.F. and Vanhoutte, P.M.: Endothelium-derived relaxing factor. In:Lüscher, T.F. and Vanhoutte, P.M. (eds). The endothelium: modulator ofcardiovascular function. Boston: CRC Press, pp 23-42, 1 990a.Lüscher, T.F. and Vanhoutte, P.M.: Heterogeneity and chronic modulation. In:Lüscher, T.F. and Vanhoutte, P.M. (eds). The endothelium: modulator ofcardiovascular function. Boston: CRC Press, pp 91-110, 1990b.Macdonald, P.S., Dubbin, P.N., and Dusting, G.J.: Endothelium-dependentvasodilatation in bovine coronary arteries: calcium dependence and inhibitionby proadifen. Clin. Exp. Pharmacol..Pharmacol. Physiol. 13: 335-340, 1986.Marietta, M.A., Yoon, P.S., Iyengar, R., Leaf, C.D. and Wishnok, J.S.:Macrophage oxidation of L-arginine to nitrite and nitrate: nitric oxide is anintermediate. Biochemistry 27: 8706-871 1, 1 988.Marsden, P.A., Schappert, K.T., Chen, H.S., Flowers, M., Sundell, C.L., Wilcox,J.N., Lamas, S. and Michel, T.: Molecular cloning and characterization ofhuman endothelial nitric oxide synthase. FEBS. Lett. 307: 287-293, 1992.Martin, G.R., Bolofo, M.L. and Giles, H.: inhibition of endothelium-dependentvasoreiaxation by arginine analogues: a pharmacological analysis of agonistand tissue dependence. Br. J. Pharmacol. 105: 643-652, 1992.Martin, W., Furchgott, R.F., Villani, G.M. and Jothianandan, D.: Phosphodiesteraseinhibitors induce endothelium-dependent relaxation of rat and rabbit aorta by• potentiating the effects of spontaneously released endotheiium-derivedrelaxing factor. J. Pharmacol. Exp. Ther. 237: 539-547, 1 986a.Martin, W., Gillespie, J.S. and Gibson, I.E.: Actions and interactions of NGsubstituted analogues of L-arginine on NANC neurotransmission in the bovine117retractor penis and rat anococcygeus. Br. J. Pharmacol. 108: 242-247,1993.Martin, W., Smith, J.A., White, D.G.: The mechanisms by which haemoglobininhibits the relaxation of rabbit aorta induced by nitrovasodilators, nitricoxide, or bovine retractor penis inhibitory factor. Br. J. Pharmacol. 89: 563-571, 1986b.Martin, W., Villani, G.M., Jothianandan, D., Furchgott, R.F.: Selective blockade ofendothelium-dependent and glyceryl trinitrate-induced relaxation byhemoglobin and by methylene blue in the rabbit aorta. J. Pharmacol. Exp.Ther. 232: 708-716, 1985.Mayer, B., John, M. Heinzel, B., Werner, E.R., Wachter, H., Schultz, G. andBohme, E.: Brain nitric oxide synthase is a biopterin- and flavin-containingmulti-functional oxido-reductase. FEBS. Lett. 288: 1 87-1 91, 1 991.Mayer, B., Schmidt, K., Humbert, P. and Bohme, E.: Biosynthesis of endotheliumderived relaxing factor. A cytosolic enzyme in porcine aortic endothelial cellCa2+-dependently converts L-arginine into an activator of soluble guanylatecyclase. Biochem. Biophys. Res. Commun. 164: 678-687, 1989.McCall, T. and Valiance, P. Nitric oxide takes centre-stage with newly definedroles. Trends. Pharmacol. Sci. 13: 1-6, 1992.McCall, T.B., Boughton-Smith, N.K., Palmer, R.M.J., Whittle, B.J. and Moncada,S.: Synthesis of nitric oxide from L-arginine by neutrophils. Release andinteraction with superoxide anion. Biochem. J. 261: 293-296, 1 989.McCall, T.B., Feelisch, M., Palmer, R.M.J., Moncada, S.: identification of Niminoethyl-L-ornithine as an irreversible inhibitor of nitric oxide synthase inphagocytic cells. Br. J. Pharmacol. 102: 234-238, 1991a.McCall, T.B., Palmer, R.M.J. and Moncada, S.: induction of nitric oxide synthasein rat peritoneal neutrophils and its inhibition by dexamethasone. Eur. J.Immunol. 21: 2523-2527, 1991b.McMillan, K., Bredt, D.S., Hirsch, D.J., Snyder, S.H., Clark, J.E., Masters, B.S.S.:Cloned, expressed rat cerebellar nitric oxide synthase contains stoichiometricamounts of heme which binds carbon monoxide. Proc. NatI. Acad. Sci.U.S.A. 89: 11141-11145, 1992.Mellion, B.T., Ignarro, L.J., Meyers, C.B., Ohlstein, E.M, Ballot, B.A., Hyman, A.L.and Kadowitz, P.J.: inhibition of human platelet aggregation by 5-nitrosothiols. Heme-dependent activation of soluble guanylate cyciase andstimulation of cyclic GMP accumulation. Mol. Pharmacol. 23: 653-664,1983.118Mellion, B.T., Ignarro, L.J., Ohistein, E.H., pontecorvo, E.G., Hyman, A.L.,Kadowitz, P.J.: Evidence for the inhibitory role of guanosine 3’,5’-monophosphate in ADP-induced human platelet aggregation in the presenceof nitric oxide and related vasodilators. Blood 57: 946-955, 1 981.Miller, R.A., Warden, J.C., Cooperman, L.H. and Price, L.H.: Central sympatheticdischarge and mean arterial pressure during halothane anaesthesia. Br. J.Anaeth. 41: 918-928, 1969.Miller, V.M.: Selective production of endothelium-derived nitric oxide in caninefemoral veins. Am. J. Physiol. 261: H677-H682, 1991.Misko, T.P., Moore, W.M., Kasten, T.P., Nickols, G.A., Gorbett, J.A., Tilton, R.G.,McDaniel, M.L., Williamson, J.R. and Currie, M.G.: Selective inhibition of theinducible nitric oxide synthase by aminoguanidine. Eur. J. Pharmacol. 233:119-1 25, 1993.Molenaar, P., Malta, E., Johns, C.R., Buxton, B.F. and Summers, R.J.:Autoradiographic localization and function of I-adrenoceptors on the humaninternal mammary artery and saphenous. Br. J. Pharmacol. 95: 225-233,1988.Mollace, V., De Francesco, E.A. and Nistico, G.: Evidence that pharmacologicalmanipulations of central L-arginine-NO pathway influence blood pressure andheart rate in rats. Neurosci. Lett. 1 37: 87-90, 1 992.Moncada, S. and Palmer, R.M.J.: The L-arginine: nitric oxide pathway in the vesselwall. In: Moncada, S. and Higgs, E.A. (eds). Nitric Oxide from L-Arginine: ABioregulatory System. Amsterdam: Lesevier, pp 1 9-33, 1 990.Moncada, S., Palmer, R.M.J. and Higgs, E.A.: Nitric oxide: physiology,pathophysiology and pharmacology. Pharmacol. Rev. 43:109-1 42, 1991.Moncada, S., Palmer, R.M.J. and Higgs, EA.: The discovery of nitric oxide as theendogenous nitrovasodilator. Hypertension 1 2: 365-372, 1 988.Moore, P.K., al-Swayeh, O.A. and Evans, R.: Rabbit brain contains an endogenousinhibitor of endothelium-dependent relaxation. Br. J. Pharmacol. 101: 865-868, 1990.Moore, P.K., Oluyomi, A.0., Babbedge, R.C., Wallace, P. and Hart, S.L.: LNGnitro arginine methyl ester exhibits antinociceptive activity in the mouse. Br.J. Pharmacol. 102: 198-202, 1991.Muldoon, S.M., Hart, J.L., Brown, K.A. and Freas, W.: Attenuation ofendothelium-mediated vasodilation by halothane. Anesthesiology 68: 31-37,1988.119Mülsch, A. and Busse R.: NGnitroLarginine (N5-[imino (nitroamino) methyl]-Lornithine) impairs endothelium-dependent dilations by inhibiting cytosolicnitric oxide synthesis from L-arginine. Naunyn. Schmiedebergs. Arch.Pharmacol. 341: 143-147, 1990.Myers, P.R., Minor, R.L. Jr., Guerra, R. Jr., Bates, J.N. and Harrison, D.G.:Vasorelaxant properties of the endothelium-derived relaxing factor moreclosely resemble S-nitrosocysteine than nitric oxide. Nature 345: 161-163,1990.Nagao, T. and Vanhoutte, P.M. (1991). Hyperpolarization contributes toendothelium-dependent relaxations to acetylcholine in femoral veins of rats.Am. J. Physiol., 261, H1034-H1037.Nakajima, T., Matsuoka, Y., Kakirnoto, Y.: Isolation and identification of NGmonomethyl, NG,NG dimethyl and NG,N’G dimethylarginine from thehydrolysates of proteins of bovine brain. Biochem. Biophys. Acta 230: 212-222, 1970.O’Donnell, V.B., Tew, D.G., Jones, O.T.G. and England, P.J.: Studies on theinhibitory mechanism of iodonium compounds with special reference toneutrophil NADPH oxidase. Biochem. J. 290: 41-49, 1993.Palacios, M., Knowles, R.G., Palmer, R.M.J. and Moncada, S.: Nitric oxide from Larginine stimulates the soluble guanylate cyclase in adrenal glands. Biochem.Biophys. Res. Commun. 165: 802-809, 1989.Palmer, R.M.J. and Moncada, S.: A novel citrulline-forming enzyme implicated inthe formation of nitric oxide by vascular endothelial cells. Biochem. Biophys.Res. Commun. 158: 348-352, 1989.Palmer, R.M.J., Ashton, D.S. and Moncada, S.: Vascular endothelial cellssynthesize nitric oxide from L-arginine. Nature 333: 664-666, 1 988a.Palmer, R.M.J., de Souza, B.M. and Moncada, S.: Nitric oxide synthesized from Larginine regulates vascular tone in the coronary circulation of the rabbit. Br.J. Pharmacol. 97: 1119-1124, 1989.Palmer, R.M.J., Ferrige, A.G. and Moncada, S.: Nitric oxide release accounts forthe biological activity of endothelium-derived relaxing factor. Nature 327:524-526, 1987.Palmer, R.M.J., Rees, D.D., Ashton, D.S. and Moncada, S.: L-arginine is thephysiological precursor for the formation of nitric oxide in endotheliumdependent relaxation. Biochem. Biophys. Res. Commun. 153: 1251-1256,1988b.120Pang, C.C.Y. and Wang, Y.-X.: Differential in vivo and in vitro vascular effects ofmethylene blue (MB) and NGnitroLarginine (L-NNA). Pharmacologist 35:165, 1993.Pang, C.C.Y.: Vasopressin antagonist and saralasin on regional blood flowfollowing hemorrhage. Am. J. Physiol. 245: H749-H755, 1983.Park, K.S., Lee, H.W., Hong, S.Y. Shin, s., Kim, S. and Paik, W.K.: Determinationof methylated amino acids in human serum by high-performance liquidchromatagraphy. J. Chromatogr. 440: 225-230, 1988.Passon, P.G. and Peuler, J.D.: A simple radiometric assay for plasmanorepinephrine and epinephrine. Anal. Biochem. 51: 618-631, 1973.Pawloski, J.R. and Chapnick, B.M.: LTD4 and bradykinin evoke endotheliumdependent relaxation of renal vein: dissimilar mechanism. Am. J. Physiol.261: H88-H95, 1991.Pegoraro, A.A., Carretero, O.A., Sigmon, D.H. and Beierwaltes, W.H.:Sympathetic modulation of endothelium-derived relaxing factor.Hypertension 19: 643-647, 1992.Pennefather, P. and Quastel, D.M.J.: Desensitization of the nicotinic receptor atthe mouse neuromuscular junction. Br. J. Pharmacol. 77: 395-404, 1982.Persson, M.G., Gustafsson, L.E., Wikiund, N.P., Hedqvist, P. and Moncada, S.:Endogenous nitric oxide as a modulator of rabbit skeletal musclemicrocirculation in vivo. Br. J. Pharmacol. 100: 463-466, 1990.Peterson, T.V., Morley, C.D., Carter, A.B., Miller, R.A. and Tramell, M.S.: Renaleffects of nitric oxide synthase inhibition in conscious monkeys. FASEB. J.7: A7, 1993.Petros, A., Bennett, D. and ValIance, P.: Effect of nitric oxide synthase inhibitorson hypotension in patients with septic shock. Lancet 338: 1 557-1 558,1991.Pinto, A., Abraham, N.C. and Mullane, K.M.: Cytochrome P-450-dependnetmonooxygenase activity and endothelial-dependent relaxations induced byarachidonic acid. J. Pharmacol. Exp. Ther. 236: 445-451, 1986.Polakowski, J.S., Pollock, D.M., Divish, B.J. and Opgenorth, T.J.: Effect of A IIreceptor blockade during chronic inhibition of NO synthase. FASEB. J. 7:A547, 1993.Price, H.L., Price, M.L. and Morse, H.T.: Central nervous actions of halothaneaffecting systemic circulation. Anesthesiology 24: 770-778, 1 969.121Pucci, M.L., Lin, L. and Nasjletti, A.: Pressor and renal vasoconstrictor effects ofNGnitroLarginine as affected by blockade of pressor mechanisms mediatedby the sympathetic nervous system, angiotensin, prostanoids andvasopressin. J. Pharmacol. Exp. Ther. 261: 240-245, 1 992.Quastel, D.M.J. and Saint, D.A.: Transmitter release at mouse motor nerveterminals mediated by temporary accumulation of intracellular barium. J.Physiol. (London) 406: 55-73, 1988.Radomski, M.W., Palmer, R.M.J. and Moncada S.: An L-arginine/nitric oxidepathway present in human platelets regulates aggregation. Proc. NatI. Acad.Sci. U.S.A. 87: 5193-5197, 1990a.Radomski, M.W., Palmer, R.M.J. and Moncada S.: Glucocorticoids inhibit theexpression of an inducible, but not the constitutive, nitric oxide synthase invascular endothelial cells. Proc. NatI. Acad. Sci. U.S.A. 87: 10043-10047,1990b.Radomski, M.W., Palmer, R.M.J. and Moncada, S.: Characterization of the Larginine: nitric oxide pathway in human platelets. Br. J. Pharmacol. 101:32 5-328, 1 990c.Radomski, M.W., Palmer, R.M.J. and Moncada, S.: Comparative pharmacology ofendothelium-derived relaxing factor, nitric oxide and prostacyclin in platelets.Br. J. Pharmacol. 92: 181-1 87, 1987a.Radomski, M.W., Palmer, R.M.J. and Moncada, S.: Endogenous nitric oxideinhibits human platelet adhesion to vascular endothelium. Lancet 2: 1057-1058, 1987b.Radomski, M.W., Palmer, R.M.J. and Moncada, S.: The anti-aggregating propertiesof vascular endothelium: interactions between prostacyclin and nitric oxide.Br. J. Pharmacol. 92: 639-646, 1987c.Radomski, M.W., Palmer, R.M.J. and Moncada, S.: The role of nitric oxide andcGMP in platelet adhesion to vascular endothelium. Biochem. Biophys. Res.Commun. 148: 1482-1489, 1987d.Ragan, C. and Bloxham, D.P.: Specific labelling of a constituent polypeptide ofbovine heart mitochondril reduced nicotinamide-adenine dinucleotideubiquinone reductase by the inhibitor diphenyleneiodonium. Biochem. J.163: 605-61 5, 1977.Rand, M.J. and Li, C.G.: The inhibition of nitric oxide-mediated relaxations in rataorta and anococcygeus muscle by diphenylene iodonium. Clin. Exp.Pharmacol. Physiol. 20: 141-148, 1993.122Randall, M.D. and Griffith, T.M.: Differential effects of L-arginine on the inhibitionby NGnitroLarginine methyl ester of basal and agonist-stimulated EDRFactivity. Br. J. Pharmacol. 104: 743-749, 1991.Rang, H.P.: Drug receptors and their function. Nature 231: 91-96, 1972.Rapoport, R.M. and Murad, F.: Agonist-induced endothelium-dependent relaxationin rat thoracic aorta may be mediated through cGMP. Circ. Res. 52: 352-357, 1983.Raszkiewicz, J.L., Linville, D.G., Kerwin, J.F. Jr., Wagenaar, F. and Arneric, S.P.:Nitric oxide synthase is critical in mediating basal forebrain regulation ofcortical cerebral circulation. J. Neurosci. Res. 33: 129-1 35, 1992.Rees, D.D., Cellek, S., Palmer, R.M.J. and Moncada, S.: Dexamethasone preventsthe induction by endotoxin of a nitric oxide synthase and the associatedeffects on vascular tone: an insight into endotoxin shock. Biochem. Biophys.Res. Commun. 173: 541-547, 1990a.Rees, D.D., Palmer, R.M.J., Hodson, H.F. and Moncada, S.: A specific inhibitor ofnitric oxide formation from L-arginine attenuates endothelium-dependentrelaxation. Br. J. Pharmacol. 96: 41 8-424, 1989a.Rees, D.D., Palmer, R.M.J., Moncada, S.: Role of endothelium-derived nitric oxidein the regulation of blood pressure. Proc. Nati. Acad. Sci. U.S.A. 86: 3375-3378, 1989b.Rees, D.D., Palmer, R.M.J., Schulz, R., Hodson, H.F. and Moncada, S.:Characterization of three inhibitors of endothelial nitric oxide synthase in vitroand in vivo. Br. J. Pharmacol. 101: 746-752, 1990b.Ribeiro, M.O., Antunes, E., de Nucci, G., Lovisolo, S.M., and Zatz, R.: Chronicinhibition of nitric oxide synthesis. A new model of arterial hypertension.Hypertension 20: 298-303, 1992.Rorie, D.K., Hunter, L.W. and Lunn, J.J.: Halothane decreases the release ofneuropeptide Y and 3,4-dihydroxyphenyiglycol from superfused segments ofdog pulmonary artery. Anesthesiology 73: 722-730, 1 990.Rosenblum, W.l., Nishimura, H. and Nelson, G.H.: L-NMMA in brainmicrocirculation of mice is inhibited by blockade of cyclooxygenase and bysuperoxide dismutase. Am. J. Physiol. 262: Hi 343-Hi 349, 1992.Rouhani, R., Gross, S.S., Madera, A.M., Levi, R. and Griffith, 0.W.: N-methylsubstituted derivatives of N-amino-L-arginine as reversible and irreversibleinhibitors of nitric oxide synthase. FASEB. J. 6: Al 256, 1 992.123Sakuma, I., Stuehr, D.J., Gross, S.S., Nathan, C.F. and Levi, R.: Identification ofarginine as a precursor of endothelium-derived relaxing factor. Proc. Nati.Acad. Sci. U.S.A. 85: 8664-8667, 1988.Sakuma, I., Togashi, H., Yoshioka, M., Saito, H., Yanagida, M., Tamura, M.,Kobayashi, T., Yasuda, H., Gross, S.S. and Levi, R.: NGmethyILarginine,an inhibitor of L-arginine-derived nitric oxide synthesis, stimulates renalsympathetic nerve activity in vivo. A role for nitric oxide in the centralregulation of sympathetic tone? Circ. Res. 70: 607-61 1, 1 992.Sanins, S.M., Adams, W.J., Kaiser, D.G., Halstead, G.W., Hosley, J., Barnes, H.and Baillie, T.A.: Mechanistic studies on the metabolic chiral inversion of Ribuprofen in the rat. Drug Metab. Dispos. 1 9: 405-410, 1 991.Schmidt, H.H.H.W., Baeblich, S.E., Zernikow, B.C., Klein, M.M. and Bohme, E.: Larginine and arginine analogues: effects on isolated blood vessels andcultured endothelial cells. Br. J. Pharmacol. 101: 145-1 51, 1990.Schmidt, H.H.H.W., Gagne, G.D., Nakane, M., Pollock, J.S., Miller, M.F. andMurad, F.: Mapping of neural nitric oxide synthase in the rat suggestsfrequent co-localization with NADPH diaphorase but not with soluble guanylylcyclase and novel paraneural functions for nitrinergic signal transduction. J.Histochem. Cytochem. 40: 1439-1456, 1992.Schmidt, H.H.H.W., Klein, M.M., Niroomand, F. and Bohme, E.: Is arginine aphysiological precursor of endothelium-derived nitric oxide? Eur. J.Pharmacol. 148: 293-295, 1988a.Schmidt, H.H.H.W., Nau, H., Wittfoht, W., Gerlach, J., Prescher, K.E., Klein,M.M., Niroomand, F. and Bohme, E.: Arginine is a physiological precursor ofendothelium-derived nitric oxide. Eur. J. Pharmacol. 154: 21 3-216, 1988b.Schmidt, H.H.H.W., Wilke, P., Evers, B. and Bohme, E.: Enzymatic formation ofnitrogen oxides from L-arginine in bovine brain cytosol. Biochem. Biophys.Res. Commun. 165: 284-291, 1989.Schultz, K.D., Schultz, K. and Schultz, G.: Sodium nitroprusside and other smoothmuscle relaxants increase cyclic GMP levels in rat ductus deferens. Nature265: 750-751, 1977.Seagard, J.L., Hopp, F.A., Doneran, J.H., Kalbfleisch, J.H., Kampine, J.P.:Halothane and carotid sinus reflex: Evidence for multiple sites of action.Anesthesiology 57: 1 91-202, 1 982.Shand, D.G., Morgan, D.H. and Oates, D.J.A.: The release of guanethidine andbethanidine by splenic nerve stimulation: a quantitative evaluation showingdissociation from adrenergic blockade. J. Pharmacol. Exp. Ther. 184: 73-80,1973.124Shibuki, K. and Okada, D: Endogenous nitric oxide release required for long-termsynaptic depression in the cerebellum. Nature 349: 326-328, 1 991.Singer, P.J. and Peach, M.J.: Endothelium-dependent relaxation of rabbit aorta. I.Relaxation stimulated by arachidonic acid. J. Pharmacol. Exp. Ther. 226:790-795, 1983.Skovsted, P., Price, M.L. and Price, H.L.: The effects of halothane on arterialpressure, preganglionic sympathetic activity and barostatic reflexes.Anesthesiology 31: 507-514, 1969.Stewart, G.A. and Hanley, T.: Other oral hypoglycemic agents. In: Campbell, G.D.(ed). Oral Hypoglycemia Agents: Pharmacology and Therapeutics. Londonand New York: Academic Press, 398-407, 1 969.Stuehr, D.J. and Marietta, M.A.: Mammalian nitrate biosynthesis: mousemacrophages produce nitrite and nitrate in response to Escherichia coiliipopolysaccharide. Prot. NatI. Acad. Sci. U.S.A. 82: 7738-7742, 1985.Stuehr, D.J. and Marietta, M.A.: Induction of nitrite/nitrate synthesis in murinemacrophage by BCG, infection, lymphokines, or interferon-y. J. Immunol.139: 518-525, 1987a.Stuehr, D.J. and Marietta, M.A.: Synthesis of nitrite and nitrate in murinemacrophage cell lines. Cancer Re. 47: 5590-5594, 1987b.Stuehr, D.J., Cho, H.J., Kwon, N.S., Weise, M.F., Nathan, C.F.: Purification andcharacterization of the cytokine-induced macrophage nitric oxide synthase:an FAD- and FMN-containing flavoprotein. Proc. NatI. Acad. Sci. U.S.A. 88:7773-7777, 1991 a.Stuehr, D.J., Fasehum, O.A., Kwon, N.S. and Nathan, C.F.: FAD and GSHparticipate in macrophage synthesis of nitric oxide. Biochem. Biophys. Res.Commun. 168: 558-565, 1990.Stuehr, D.J., Fasehun, O.A., Kwon, N.S., Gross, S.S., Gonzalez, J.A., Levi, R. andNathan, C.F.: Inhibition of macrophage and endothelial cell nitric oxidesynthase by diphenyieneiodonium and its analogs. FASEB. J. 5: 98-103,1991b.Stuehr, D.J., Gross, S.S., Sakuma, I., Levi, R. and Nathan, C.F.: Activated murinemacrophages secrete a metabolite of arginine with the bioactivity ofendothelium-derived relaxing factor and the chemical reactivity of nitricoxide. J. Exp. Med. 169: 1011-1020, 1989a.125Stuehr, D.J., Kwon, N.S. and Nathan, C.F.: FAD and GSH participate inmacrophage synthesis of nitric oxide. Biochem. Biophys. Res. Commun.168: 558-565, 1990.Stuehr, D.J., Kwon, N.S., Gross, S.S., Thiel, B.A., Levi, R. and Nathan, C.F.:Synthesis of nitrogen oxides from L-arginine by macrophage cytosol:requirement for inducible and constitutive components. Biochem. Biophys.Res. Commun. 161: 420-426, 1989b.Stuehr, D.J., Kwon, N.S., Nathan, C.F., Griffith 0.W., Feldman, P.L. and WisemanJ.: N°-hydroxy-L-arginine is an intermediate in the biosynthesis of nitricoxide from L-arginine. J. Biol. Chem. 266: 6259-6263, 1991c.Sutter, M.C. and Wang, Y.-X.: Recent cardiovascular drugs from Chinesemedicinal plants. Cardiovasc. Res. 27: 1891-1 901, 1993.Suzuki, H. and Chen, G.: Endothelium-derived hyperpolarizing factor (EDHF): anendogenous potassium-channel activator. N.I.P.S. 5: 21 2-21 5, 1 990.Suzuki, H., Ikenaga, H., Hishikawa, K., Nakaki, T., Kato, R. and Saruta, T.:Increases in N02/N3 excretion in the urine as an indicator of the releaseof endothelium-derived relaxing factor during elevation of blood pressure.Clin. Sci. 82: 631-634, 1992.Tabrizchi, R. and Pang, C.C.Y.: Comparative effects of rauwolscine, prazosin andphentolamine on blood pressure and cardiac output in anesthetized rats.Can. J. Physiol. Pharmacol. 65: 1421-1427, 1986.Tabrizchi, R. and Triggle, C.R.: Actions of L- and D-arginine and NGmonomethylL-arginine on the blood pressure of pithed normotensive and spontaneouslyhypertensive rats. Clin. Exp. Hypertens. [A]. 14: 527-546, 1992.Tabrizchi, R. and Triggle, C.R.: Influence of N-nitro-L-arginine methyl on pressorresponses elicited by sympathetic nerve stimulation in pithed normotensiveand hypertensive rats. Life. Sci. 49: 1 989-1 995, 1 991.Tayeh, M.A. and Marietta, M.A.: Macrophage oxidation of L-arginine to nitricoxide, nitrite and nitrate. Tetrahyrobiopterin is required as a cofactor. J.Biol. Chem. 264: 19654-1 9658, 1989.Tepperman, B.L. and Whittle, B.J.: Endogenous nitric oxide and sensoryneuropeptides interact in the modulation of the rat gastric microcirculation.Br. J. Pharmacol. 105: 171-1 75, 1992.Toda, N., Kitamura, Y. and Okamura, T.: Neural mechanism of hypertension bynitric oxide synthase inhibitor in dogs. Hypertension 21: 3-8, 1 993.126Togashi, H., Sakurna, I., Yoshioka, M., Kobayashi, T., Yasuda, H., Kitabatake, A.,Saito, H., Gross, S.S. and Levi, R.: A central nervous system action of nitricoxide in blood pressure regulation. J. Pharmacol. Exp. Ther. 262: 343-347,1992.Tresham, J.J., Dusting, G.J., Coghian, J.P. and Whitworth, J.A.: Haemodynamicand hormonal effects of N-nitro-L-arginine, an inhibitor of nitric oxidebiosynthesis, in sheep. Clin. Exp. Pharmacol. Physiol. 1 8: 327-330, 1991.Tucker, J.F., Brave, S.R., Charalambus, L., Hobbs, A.J. and Gibson, A.: LNGnitroarginine inhibits non-adrenergic, non-cholingic relaxations of guinea pigisolated tracheal smooth muscle. Br. J. Pharmacol. 100: 663-664, 1990.Valiance, P., Collier, J. and Moncada, S.: Effects of endothelium-derived nitricoxide on peripheral arteriolar tone in man. Lancet 2: 997-1000, 1 989.Valiance, P., Collier, J. and Moncada, S.: Nitric oxide synthesized from L-argininemediates endothelium-dependent dilatation in human veins in vivo.Cardiovasc. Res. 23: 1053-1 057, 1989.Valiance, P., Leone, A., Calver, A., Collier, J. and Moncada, S. Accumulation ofan endogenous inhibitor of nitric oxide synthesis in chronic renal failure.Lancet 339: 572-575, 1992.van Geideren, M.E., Heiligers, J.P. and Saxena P.R.: Haemodynamic changes andacetyicholine. induced hypotensive responses after NGnitroLargininemethyl ester in rats and cats. Br. J. Pharmacol. 103: 1899-1904, 1991.Vargas, H.M., Cuevas, J.M., ignarro, L.J. and Chaudhuri, G.: Comparison of theinhibitory potencies of NGmethyl, NGnitro and NGaminoLarginine onEDRF function in the rat: evidence for continuous basal EDRF release. J.Pharmacol. Exp. Ther. 257: 1208-1215, 1991.Vargas, H.M., Ignarro, L.J. and Chaudhuri, G.: Physiological release of nitric oxideis dependent on the level of vascular tone. Eur. J. Pharmacol. 190: 393-397, 1990.Wang, Y.-X. and Pang, C.C.Y.: Effects of adrenalectomy and chemicalsympathectomy on the pressor and tachycardic responses todiphenyleneiodoniu. J. Pharmacol. Exp. Ther. in press, 1994a.Wang, Y.-X. and Pang, C.C.Y,: Effects of L and D enantiomers of NGnitroarginineon blood pressure in pentobarbital-anaesthetized rats. Pharmacology (LifeSci. Adv.) 9: 723-728, 1990a.127Wang, Y.-X. and Pang, C.C.Y.: NGnitroLarginine contracts vascular smoothmuscle via a non-nitric oxide synthesis mechanism. J. Cardiovasc.Pharmacol. in press, 1994b.Wang, Y.-X. and Pang, C.C.Y.: Pressor effect of NGnitroLarginine inpentobarbital-anesthetized rats. Life Sci. 47: 221 7-2224, 1 990b.Wang, Y.-X.. and Pang, C.C.Y.: Suppression by ethanol of pressor responsecaused by the inhibition of nitric oxide synthesis. Eur. J. Pharmacol. 233:275-278, 1 993.Wang, Y.-X., Poon, K.S., Randall, D.J. and Pang, C.C.Y.: Endothelium-derivednitric oxide partially mediates salbutamol-induced vasodilatations. Eur. J.Pharmacol. 250: 335-340, 1 993a.Wang, Y.-X., Zhou, T. and Pang, C.C.Y.: Depressor response to acetyicholine andpressor response to NGnitroLarginine in conscious Wistar, WKY and SHrats. In: Moncada, S. Marietta, M.A. Hibbs, J.B. Jr. and Higgs, E.A. (eds).The Biology of Nitric Oxide. 1. Physiological and Clinical Aspects. London:Portland, 174-176, 1992.Wang, Y.-X., Zhou, T. and Pang, C.C.Y.: A comparison of the inhibitory effects ofsodium nitroprusside, pinacidil and nifedipine on pressor response to NGnitro-L-arginine. Br. J. Pharmacol. 108: 398-404, 1993b.Wang, Y.-X., Zhou, T., Chua, T.C. and Pang, C.C.Y.: Effects of inhalation andintravenous anesthetic agents on pressor response to NGnitroLarginine.Eur. J. Pharmacol. 198: 183-188, 1991.Watkins, R., Cook, R.W. and McLeod, R.: NGmonomethylLarginine (L-NMMA)does not inhibit depressor responses to acetylcholine (ACh) in rats. FASEB.J. 7: 775, 1993.White, K.A. and Marietta, M.A.: Nitric oxide synthase is a cytochrome P-450 typehemoprotein. Biochemistry 31: 6627-6631, 1 992.Whittle, B.J., Lopez-Beimonte, J. and Rees, D.D.: Modulation of the vasodepressoractions of acetyicholine, bradykinin, substance P and endotheiin in the rat bya specific inhibitor of nitric oxide formation. Br. J. Pharmacoi. 98: 646-652,1989.Widdop, R.E., Gardiner, S.M., Kemp, P.A. and Bennett, T.: The influence ofatropine and atenolol on the cardiac haemodynamic effects of NGnitroLarginine methyl ester in conscious, Long Evans rats. Br. J. Pharmacol. 105:653-656, 1 992.128Winquist, R.J., Faison, E.P., Waldman, S.A., Schwartz, K., Murad, F. andRapoport, R.M.: Atrial natriuretic factor elicits an endothelium-independentrelaxation and activates particulate guanylate cyclase in vascular smoothmuscle. Proc. Natl. Acad. Sci., U.S.A. 81: 7661-7664, 1984.Woodman, 0.L. and Dusting, G.J.: N-nitro L-arginine causes coronaryvasoconstriction and inhibits endothelium-dependent vasodilatation inanaesthetized greyhounds. Br. J. Pharmacol. 103: 1407-1410, 1991.Wu, J.-N. and Berecek, K.H.: Prevention of genetic hypertension by earlytreatment of spontaneously hypertensive rats with the angiotensin convertingenzyme inhibitor captopril. Hypertension 22: 139-146, 1993.Xie, Q.W., Cho, H.J., Calaycay, J., Mumford, R.A., Swiderek, K.M., Lee, T.D.,Ding, A., Troso, T. and Nathan, C.: Cloning and characterization of induciblenitric oxide synthase from mouse macrophages. Science 256: 225-228,1992.Yamazaki, J., Fujita, N. and Nagao, T.: NGmonomethylLarginineinduced pressorresponse at developmental and established stages in spontaneouslyhypertensive rats. J. Pharmacol. Exp. Ther. 259: 52-57, 1991.Yui, Y., Hattori, R., Kosuga, K., Eixawa, H., Hiki, K., Ohkawas, S., Ohnishi, K.,Terao, S. and Kawai, C.: Calmodulin-independent nitric oxide synthase fromrat polymorphonuclear neutrophils. J. Biol. Chem. 266: 3369-3371, 1991.Zambetis, M., Dusting, G.J., Rajanayagam, S. and Woodman, 0.L.: Mechanism ofthe hypertension produced by inhibition of nitric oxide biosynthesis in rats.J. Cardiovasc. Pharmacol. 17 (Suppl. 3): S191-S197, 1991.Zar, J.H. (ed): Biostatistical Analysis. Englewood Cliffs: Prentice-Hall, Inc. pp 1-469, 1984.129APPEN DICESPaper IWang, Y.-X. and Pang, C.C.Y.: Possible dependence of pressor and heart rateeffects of NGnitroLarginine on autonomic nerve activity. Br. J. Pharmacol. 103:2004-2008, 1 991. The reproduction of this paper was kindly permitted by thecopyright holder, Macmillan Press Ltd., Hampshire, U.K.130!r J PharnwoL 1991), 103. 2()4—2OO8‘ .i4mLHan Press Lid. t;Possible dependence of pressor and heart rate effects ofNGnitroLarginine on autonomic nerve activityYong-Xiang Wang & LCathenne C.Y. PangDepartment of Pharmacology & Therapeutics. Faculty of Medicine. The University of British Columbia, 2176 Health SciencesMall, Vancouver, B.C. V6T 1Z3, CanadaI The effects of N°-nitro-L-arginine (L-NNA) on mean arterial pressure (MAP) and heart rate (HR) wereinvestigated in conscious rati2 Intravenous bolus cumulative doses o(L-NNA (l-32mgkg)dose.dependently increased MAP. Bothmecamylamine and phentolamine increased MAP responses to u-NNA, angiotensin LI and methoy.amine.Propranolol, reserpine atropine and captopril did not affect MAP response to L-NNA.3 A signifleant negative correlation of HR and MAP responses to t..NNA was obtained in control ratsbut not in rats pretreated with reserpine or mccainybmine. Significant negative correlations also occurredin the presence of atropine, propranolol. phentolasnine or captopril.4 A single i.v. bolus dose of L-NNA (32mgkg) raised MAP to a peak value of 53 ± 3mmHg and theeffect lasted more than 2h; the rise and recovery of MAP were accompanied by significant decrease andincrease in HR. respectively. While both pheutolamine and xnecamylansine increased peak MAP responseto t-NNA, mecamylamine abolished the biphasic HR response and phentolamine potentiated the bradycardic component of HR.5 Blockade of the autonomic nervous and renin.angiotensin systems did not attenuate the pressor effectsof t..-N’NA. However, the biphasic HR response to L-NNA is mediated via modulation of autonomic nerveactivities.Keywords: N°.nitro-L.arginine (L-NNA); vasopressor; autonornic ganglion: sympathetic and parasympathetic nervous system:renin-angiotensin systemIntroductionThere is evidence that endothelium-derived relaxing factorIEDRF) released by vascular endothelial cells is nitric oxide(NOH Palmer er aL. 1987; Ignarro et at., 1987) which is formedfrom the precursor L-arginine (L-Arg) (Palmer er at., 1988;Sakuma et at.. 19881. It has been shown that NO synthase andendochelium-dependent vascular relaxation responses in isolated arteries are inhibited by N°-substituzed L-Arg analogueswhich include N°.monomethvl-L-areinine (L.MMA) (Palmerer at.. 1988: Rees et al. 1989a: 1990). N°-nitro-L-argininemethyl ester (Moore et at., 1990: Rees et at., 1990), Nlmlnoethyl-L.ornithine (Rees ci at., 1990) and N°-nitro-Largirune (L-NNA (Moore et at.. 1990; Mülsch & Busse. 1990:Kobavashi & Hatton. 1990; Ishii e at., 1990). In nyc studiesshow that L-MMA (Rees et at.. 1989b; 1990: Aisaka et at.,1989; Whittle ci at., 1989; Gardiner er at., 1990b.c), N°-nitroL-arginine methyl ester (Gardiner ci at.. 1990a.c,d), Niminoethyl-L.ornithine Rees cc at.. 1990) and L-NNA (Wang& Pang. 1990) cause pressor responses and bradycardia.Although it is likely that the presser effects of L-Arg anaogues are caused by the inhibition of NO production fromvascular endotheljal cells (Aisaka et at., 1989: Rees er at.,1989b: Wang & Pang, 19901, other vasopressor systems maycontribute to the response. It has been shown that vascularendothelium inhibits the release of noradrenaline from sympathetic nerves which innervate canine pulmonary artery andvein suggesting that the endothelium. n part via endocheliumderived relaxing factor (EDRF) release, acted as an endoge.nous inhibitor of sympathetic transmitter retease IGreenbergcc at.. l990). Togashi cc at. 11990) showed that L-MMAincreased postaanglioriic sympathetic nerve activities in intactand bilateral sino-aortic- and vagal-denervated rats and preganglionic adrenal nerve activity in sino-aortic and vagaldenervated rats. It has also been reported that EDRFinhibited renin release Vidal er at., 1988). Therefore it islogical to postulate that the hacmodynarnic effects of L-Arganalogues are partially mediated via potentiation of activitiesof the autonomic nervous and/or renin-angiotensin system(s).The aims of this study were: (1) to assess the contribution ofthe autonomic nervous and renin.angioterisin systems onpressor response to L-NNA; (2) to examine whether thebradycardia produced in response to L-NNA is mediated viareñex activation of the autonomic nervous system.MethodsSurgical preparationSprague.Dawley rats (240—400 g) were anaestheti.zed withhajorhane (4% in air for induction, 2% in air for surgicalpreparation). A polyethylene cannula (PESO) was inserted intothe left iliac artery to allow recordings of mean arterial pressure (MAP). PE5O cannulae were also inserted into the rightor both) iliac vein(s) for the administration of drugs. The cannulae were filled with heparinized saline (25iumF) and tunnelled s.c. along the back, exteriorized at the hack of the neckand secured. The rats were given 6.- h recovery of the effectsof halorhane and surgery before further use.Experimental protocolThe indwelling arterial catheter from each rat was connectedto a pressure transducer P23DB, Gould Statham. CA. U.S.AJfor the recordings of MAP and heart rate (HR) which wasderived electronically from the upstroke of the arterial pulsepressure by a tachograph Grass, Model P4G(. The conscious rats were allowed to wander freely in a small cage forI h before the administrations of drugs. MAP and HR werecontinuously monitored. The rats were killed by an overdoseof pentobarbitone at the end of each experiment. Two mainstudies were conducted:Dose-response curves for .‘v°-nicro-L-arginine Rats, randomlydivided into seven groups n = 6 each), were pretreated with:1) normal saline (0.9% NaCl): (II) phentolamine (i.v. infusion.Author for correspondence.131ALYSIS (JF HE PRESSOR EFFECt OF L-’4NA O5Table I Baseline values of mean .irtenal pressure (MAP)and heart rate )HR) (mean ± s.e.mean) in conscious rats pnorto and 40mm alter the administration of normal saline,phentolaimne. propranolol. reserp.ne. mccamylamine. aLto-pine and captopril403 ± 10442 ± 12364 ± 10377±9370±23372 ± 19Denotes significant difference from corresponding controlvalues within the same group (P <0.05); Denotes significant difference from normal saline group (P < 00S). ii —6per group.300gkgmin’); (TI!) propranolol (Lv. bolus at 1wgkgfollowed by infusion at 1.6pgkg_tmin_t); (IV) reserpine(5mgkg, i.p.. 24h prior to the study); (V) mecamylamine(i.v. bolus at L0mgkg followed by infusion at300glcg I1Un I); (VU atropine (i.v. bolus at lomgkg followed by infusion at 8gkgmin’ and (VII) captopril(20 mg kg’ . i.v. bolus). With the exception of reserpine andcaptopril. all antagonists were continuously infused forapproximately 160 mm, i.e., to the end of the experiment.Cumulative doses of t.-NNA (l—32mgkg i.v. bolus) weregiven 40mm after the administration of the vehicle orblockers at dose-intervals of 15—20 mm. the period required toobtain steady state MAP responses. A single dose of angiotenssn 1(1 sg kg’ 1), methoxamine (20 or 30ug kg I) acetylcholine (1 #g kg - I) or isoprenaline (I ‘g kg ‘) was injected asan i.v. bolus prior to and 20mm after the start of administration of captopril. phencolamine. atropine or propranolol.respectively, and again 2 h after giving L-NNA to assess thedegrees of inhibition at the start and completion of thestudies. In rats pretreated with reserpine and vehicle. tvramine(200ugkg ) was injected as an i.v. bolus 0min prior to and2 h after giving L-NNA. Excluding the equilibration time, theduration of each study was approximately 3 Ii.Time course of responses to a single dose of N°-nizro-Larginine Another three groups of rats fl = 6 each) werepretreated with: (VIII) normal saline; (IX) rnecamylamine; iX)phentolamine. at the same doses as those described previously. In phentolamine and mecamylarnine groups. angiocensin II was injected as an i.v. bolus prior to and 20mmafter the administration of a blocker. In Groups VIII, LX andX. a single dose of t.-NNA 32mgkg) was injected as an t.v.bolus 40mm alter the start of administration of a blocker.MAP and HR were continuously monitored for 2 Ii.DrugsThe following drugs were obtained from Sigma Chemical Co.(MO. U.S.A.): N-nitro-t.-arginine (L-NNA), mecamylaminehydrochloride, atropine sulphate, Des-Asp’-angiotezisin Iacetate, angiotensin LI acetate, acetylcholine hydrochloride,(± )-propranolol hydrochloride, (— )-opropylnoradrenalinehydrochloride and tyramine hydrochloride. The followingdrugs were also used: phentolamine hydrochloride (CibaPharmaceutical Co., NJ. U.S.A.). methozamine hydrochloride(B.W. & Co. Ltd., Quebec, Canada). captopril (ER. Squibb &Sons Inc.. NJ, U.S.A.) and reserpine (Ciba PharmaceuticalCo.. Quebec. Canada). AU drugs were dissolved in normalsaline.Calculation and statistical analysisThe ED50 and ma.imum response (Em) values of LNNAwere obtained from individual dose-response curves. Correlation coefficient (r), slope and intercept were calculated fromindividual HR versus MAP curves at various doses of t-NNA.Rise phase t of t..NNA were obtained from time-coursecurves. To obtain normal distribution of rise phase t11, thedata were logarithmically-transformed prior to statisticalanalysis. All data were analyzed by the analysis of variancefollowed by Duncan’s multiple range lest with P < 0.05 selected as the criterion for statistical significance. All results areexpressed as mean standard error (s.ernean) except for risephase t1,2 which is expressed as geometric mean and 95%confidence range.ResultsEffects of antagonists on mean arterial pressure and heartrateTable I shows baseline MAP and HR protocol 1: Groups Ito VII; protocol 2: Groups VIII to X) in rats prior to and40 mm after pretreatment with normal saline. phentolamine.propranolol. reserpine. rnecamylamine. atropmne or captopril.Normal saline (protocols I and 2) affected neither MAP norHR. MAP was not affected by propranolol. atropine or captopril but significantly decreased by phentolaxnine. reserpineand mecamylamine. HR was decreased by resernine, mecamyl.amine and propranolol and increased by phentolaminc. atropine and captopril.Table 2 shows the effects of the antagonists on responses toseveral agonists. Phentolamine (Group II) completely blockedTable 2 The effects of antagonists on mean arterial pressure MAP) and heart rate HR responses to several agonists In conscious ratsChanqe n HRbeats mm —MAP (mmHg)Antagonists Before AfterHR (beats mmBefore AfterPoeoco1 INormal salinePheniolaminePropranololReserpiaeMecamylamineAtropineCaptoprilPiotocol 2Normal salineMecamylaminePhentolamine104±569±7’104±480± St88 ± 4’118 ± 3103 ± 5103 ± 5106 ± 4104±4113 ± 3116±2105 ±6106±3108 ± 4101 ± 5410 ± 8518 ± 18’334±8’295 ± Lit313 ± 13’432± 18’418 ± 16’104±3 343±13 344±1369±2’ 431±10 308±8’72±3’ 401±13 443±8’Change zn (4.4?mmHg)Blocker .4qonisr a b c a b cPhent Mist 40—6 0 0 —106— 14 0 0Reserp Tyram 463 I75 175 —108lI—16t0 —2tProp (sop —3 1l 7—S tt6t3 22 2:Mecam Mts2 55—2 101—S 43—8 —169—tv 0 0Atrop ACts — 2 0 0 47 2 0 0Capt Al 434 0 15:1 —&313 0 —8The effects mean s.c.mean of isoprenaline Isop. I Mgkg (. tyrarnine T rain. 200mgkg-’(. methosamine (Mtxl 2Oug kg: Mtx2.3Oazkg 1. acetylcholinc ACh. I mgkg ‘I and anglotensin ((Al. I gkg) on MAP and HR were obtained before a. 20mm after bithe administrations of: phentolamine Phent. propranolol (PropL reserpine Reserpi. rnecamylamine Mecam. atropine Atropi or caplopril (Capu and. 2 S after cl an Lv. bolus injection of N°-nhtro.-L-arglninc.All results in IS) and (Cl are cignificantly different from the corresponding control values al within the came group P < 0.05). The datawere from the vehmcle.treated rat group. ri = per group.201)6 Y WANG & CCY pA.’G 132Fiwe I Effects of meshoxamine (Mu) and ang,otensLn II (All) onmean arterial pressure (MAP) (a) and heart rate (HR) (b) before (oescolumns) and 20mm after the administrations of mecamylamine (filledcolumns) and phentolamine (cross batched column). Values are meanswith s.c.xnean shown by vertical bars; a 6 in each group.Represents significant difference from corresponding control valuesprior to the administration of an antagonist (P <0.05).pressor effects and bradycardia induced by methoxaminethroughout the study period. The pressor response to methoxamine was enhanced by mecamylamine (Group V) at 20mmbut not altered at 2 h following the administration of L-NNA.The reflex bradvcardia induced by methoxamine was abolished by mecamylarnine throughout the experiments. Intravenous bolus doses of tyrarnine in vehicle-treated rats (GroupI) increased MAP and decreased HR: in Group IV ratspretreated with reserpine tyramine caused markedly lesspressor and bradycardic responses than in control rats. Isoprenaline (Group III) caused depressor and tachycardicresponses both were almost totally abolished at 20mm afterthe injection of propranolol and remained markedly attenuated 2 h after the administration of L-NNA. Atropine (GroupVii completely abolished the depressor and reflex tachycardicresponse of acetylcholine throughout the study period. At0 miss after the injection of captopril Group VII), the pressure and bradycardic effects of anglotensin I were abolished.Both responses to angiotensin I remained attenuated 2 Ii afterthe injections of L-NNA.Pressor responses to angiotensin U Groups IX and X) andmethoxamine (Group V) were potenuated by rnecarnylamineor phentolamine (Figure 1). Reflex bradycardia in response toangiotensin II and rnethoxaxnine was totally blocked by mecarnylamine but reflex bradycardia to angiotensin H was unaffected by phentolamine.Dose-response curves for NG.nirroL.arginineIntravenous bolus doses of L-NNA in vehicle-treated ratsdose-dependentlY increased MAP Figure 2). Pretreatmentwith either mecamylamine or plientolamine potentiated thepressor response to L-NNA by reducing ED,0 and increasingE,, values I Figure 2. Table 3). Pretreatment with the otherantagonists used in this study did not significantly alter thedose-MAP response curves for c-NNA (Figure 2. Table 3).Figure 3 shows the relationships between HR arid MAP forrats in Groups I to VII. In vehicle-treated rats (Group I),MAP after L-NNA was negatively correlated with HR. Signifiz Dose-response cu (n ± of th effects of Lv.bolus doses of N°-rutro-L-arglnme (L-NNA) on mean arterial pressure(MAP) in groups (a 6 each) of conscious rats preueated withnormal saline (0 in a). rcscrpine ( in a). phentolamine ( in si.propranolol (A in a). aropine Q in 5). rnecamylamine (• in 6* andcaptopril in 5).cant correlations of MAP with HR were also obtained in ratspretreated with phentolamine. propranolol. atropine or captopril but not with reserpine or inecamylamine (Table 4). Theslope of the curve was not significantly altered by atropine.propranolol or captopril but was significantly increased byphentolamine. The intercept was decreased by propranoloL520702037o320_27060 ooMAP mmHg?Figure 3 Relationship of heart rate HRI al to mean .irtenal pressure 1MAP) (5) alter injection of N°-nitro-t.arginine L-NA.3zmgkg . i.v. bolus( in conscious rats pretreated with normal saline(0 in a). phento)amine in a). mecamy(axnine in a). atropine Ain ai. propranolol(Q in St. reserpine in b)and captopril • in Si.Each ooinz represents mean values from six rats ven the same doseof L-NNA.aaCaaCaQCaatCa —“aCaa100aCa-SOC 40aa20C-)6100(7’2 4 5 15 32i.-NNA 1mg kg)320— 470.• 3701.zVt.ALYS1S OF tHE PRESSOR EFFECT OF c-ATable 3 ED,0 values and maximum effects (E of N°nitro-argmine (t.-NNA) on mean arterini prmeure in conscious rats presreated with normal saline, phentolamine,propranoloL reserpine. mecamylaniine, atropine or captoprilED,,5 1mg kg’’) E (mmHg)Normal saline 4.3 ± 0.8 52±2Phentolamine 2.1 ± 0.2’ 87 ± 6’Propranolol 6.3 ± 0.6 54±3Reserpine 3.1±0.6 56±4Mecisnylasninc 1.9 ± 0.2’ 86±5’Atropine 2.7±0.3 51±5Captopril 5.0±1.1 56±4All values represent mean ± man n 6 per group.‘Deaotes ngniñcant difference from normal saline-treatedgroup)? <0.05).Table 4 Slope, intercept and correlation coecient (r) of theheart rate vs mean artery pressure curves of N°-nitro-Largimne (i-NNA, t-32mgkg’, i.v. boius) in conscious rats(ii 6 per group) pretreazcd with normal saline, phentolamine, propranolol, resapme, mamylamine, atropinc orcaptoprilInterceptGroup r SlopeNormal saline 0.85 ± 0.03 —1.17 ± 0.24 509±32Phentolamiae 0.96 ± 0.01 —2.29 ± 0.08’ 669 ± 23’Propranolol 0.80 ± 0.03’ —0.75 ± 0.20 419 ± 20’Reserpine 0.72 ± 0.06Mecamylaminc 0.46 ± 0.12 aAtropinc 0.83 ± 0.04 —1.16 ± 0.22 557 ± 51Captoptil 0.89 ± 0.02’ —1.68 0.14 570 17-—---0 20 “.0 O 30 00 20Time (miri?Figure 4 Time course of the effects of N°-nitro-c-arginine t..NNA.32mg kg”) on mean arterial pressure MAP) a and heart rate IHRJ(hI in groups (n = 6 each) of conscious rats pretreated with normalsaline (Q). mecamylamine i and phentolamine )..j. Values aremeans with s.e.nlean shown by vertical bars.increased by phentolamine but not significantly altered bycaptopril and atroprne.Time course of the effects of N°-nzcro-L-arginineIn control rats given normal saline the MAP response to asingle dose of t.-NNA started almost immediately and r2d1eda plateau 10mm after injection (Figure 4a). The rise phase t,was 4.8mm (geometric mean. 95% confidence limit: 2.0-11.6);MAP at 40mm was not different from MAP at 10mm andremained elevated 2 h after injection. Mecamylamine andphentolammne potentiated the peak MAP response to L-NNA(Figure 4a). Mecamylamine did not alter the rise phase r1,(5.5mm, 95% confidence limit: 3.2—9.4) but phentobminereduced it(to L5min,95% confidence limit: 1.0-2.3).The pressor response to L-NNA was accompanied by initialsignificant decreases of HR at 5, 10, 20mm after injection fotowed by a recovery of HR and continual significant increasesof HR at 80 and 120mm even when MAP was still above thecontrol level (Figure 4b). Mecamylamine abolished the biphasic effects of t.-NNA on HR. Phentolamine. on the other hand.potentiated and prolonged the bradycardia.DtscaeaioaOur results show that L-NNA is a potent and long-lastingpressor agent in conscious rats. Captopril and blockers of theautonomic nervous system, namely, mecamylamine. phentolamine. reserpine, propranolol and atropine. did not attenuatethe pressor responses to t.-NNA. This indicates that thepressor effect of L.N’NA does not rely on the integrity of thesetwo vasopressor systems. It has been reported that the pressoreffects of L-MMA tRees e aL, 1989b: Aisaka et at., 1989) andL.NNA (Wang & Pang, 1990) are antagonized by L-Arg suggesting that °-substituted L-Arg analogues raise MAP viainhibiting NO synzhase.Pretreatment with mecamylarnine potentiated MAPresponses to L-NNA. angiotensin II and rnethoxamirie. Phentolasnine increased pressor responses to L-NNA and angiotensin II. Phentolamine but not mecamylamine, however.reduced the rise time c,. This non-specific enhancement ofthe pressor effects of vasopressor agents after ganglionic ora-adrenoceptor blockade is consistent with well-known observations that acute pressor responses in ntact animals are bidfered by the simultaneous withdrawal of sympathetic tone tovascular smooth muscles Lum & Rashleich. 1961: Mawji &Lockett. 1963: Minson t at.. 1989).MAP response to the injection of a single dose of L-NNAwas associated with significant initial bradycardia (0 to40 minI followed y tachycardia. A biphasic HR response toL-SNA was also observed in pentobarbitone.anaesthenzedrats Wang & Pang, 1990). Biphasic HR responses to L.MMAin chloralose and urethane anaesthetized rats have also beendescribed (Togashi er at.. 1990). Mecanwlarnine abolished HRresponses to L-NNA. suggesting that the biphasic HRresponse is mediated via reflex changes in the activities of theautonomic nervous system.The tachycardic component was not seen in rats vencumulative doses of L-NNA. presumably due to the shorterobservation time given to each dose of c-NNA. The resultsfrom cumulative dose-response relationships to L-NNA in theabsence of an antagonist, show that the MAP effects oft-NNAare negatively correlated with HR. Treatment with mecaitsylarrune or reserpine abolished the reflex changes in HR following alterations in MAP. The slope of the HR-MAP curve wasslightly reduced by propranolol but unaffected by atropine.The lack of a correlation of HR to MAP after treatment withreserpine suggests that in conscious rats, inhibition of sympathetic nerve activity rather than potentlatton of parasympathetic nerve activity is involved in reflex changes in HR.These results are consistent with the observation that bradycardia induced by L-MMA was associated with reduced renalSlope (bestsmin” mmHg), intercept (beatsmin”) and ,values represent mean ± s.e.mesn. - Denotes significance of r(P <0.05): b values were not obtained due to insignificantcorrelation coefficient; ‘denotes significant difference fromrespective values in normal saline group (P <0.05)a110•, 90).7Q).5o30to).c-i30io.-aol.ii I—aol.—120(.a-t60(.-i-200L..—40 —20200S Y-X. WANG & C.C.Y PANG134sympathetic nerve activity (Togashi et al.. 1990). Our resultsalso show that phentolamine increased the slope of the curve.Phentolamine has been shown to increase markedly plasmalevels of adrenaline and rioradrenaline (Tabrizchi ci al., 1988).Therefore, enhanced reflex bradycardia in response to t.-NNAin the presence of phentolamine may have been a consequenceof elevated background sympathetic nerve activities as baseline HR was elevated by phentobmine.Referc.cesAISAKA. K.. GROSS. 5.5.. GRIFFITH. O.W. k LEVI. R. (1989). N°methytarginine, an inhibitor otendothelium-denved nitric oxidesynthesis. is a potent pressor agent in the guinea pig: Does tutricoxide regulate blood pressure in mao? Biochcnt. Biophys. Rca.Co.,umsr.. 160, 881486.GARDINER. S.M, COMPTON. AM. I BENNETT, T. (1990.). Regionaland cardiac haemodynamic responses to glyceryl trinitrate. acetylcholine, bradykinin and endothelin-I in conscious rats: effects ofN°-ntro-L-argmine methyl ester. Br. I. Pha,mcoL, 101.632-639.GARDINER. &M., COMPTON. A.M.. BENNETT. 1.. PALMER. LM.J. &MONCADA, S. (199Gb). Control of regional blood Oow byendotheium.derivcd nitric oxide. Hypertension, 15, 486-492.GARDINER. S.M.. COMPTON. AM.. BENNETT. 2.. PALMER. R.MJ. &MONCADA, 5. (19904 Regional &aemodynamic changes duringoral ingestion of N°-m000methyl-L-srginine or N°-nitro-L.arglninc methyl ester in conscious Brattleboro rats. Br. J. PharmacoL,lOt. 10-12.GARDINER. SM.. COMP’rON. AM.. KEMP. PA. & BENNETt T.(19906). Regional and cardiac haemodynasnic effects of N°-nitro.-arnine methyl ester in conscious. Long Evans rats. Br. I. Phar.,na.coL. 101. 625—631.GREENBERG. S.5 DIECKE. F.PJ.. PEEVY. K. & TANAKA. 1.?. (1990).Release of norepinephrine from adrenergic nerve endings of bloodvessels is modulated by endothelium-derived relaxing factor. (rn.J. Hvperten.. 3,211—218.IGNARRO. LL BUGA. G.M.. WOOD. KS.. BYRNS. R.E. & CIIAIJDHURI.G. (1987). Endothelium-derived relaxing factor produced and released from artery and vein is flInG oxide. Proc. Vagl. Aco.d. Sd.U.S.A., 84, 9265—9269.ISI-Ilt. K.. CHANG. B.. KERWIN. i.E Jr.. HUANG. Z.-J. Sc MIJARD. F.1990). N’.nitro-L-arginine: a potent inhibitor of endocheliuniderived relaxing factor formation. Eisr. J. PhannacoL. 176, 219—23.KOBAYASHI. Y Sc KATTORI. K. (1990). Nitroarginine inhibitsendothelium-denved relaxation. Jpn. J. Pharmacol.. 52. 167—169LUM. B.K B. Sc RASHLEIGH. PL. (1961). Potentiation of vasoactivedrugs by ganslion blocking agents. J. P’uirrncoI. Sep. Ther.. 132,13—18.MAWJI. S. Sc LOCKETT. M.F. (1963). Pressor effects of adrenaline. nor-adrenaline and reñex vasoconstriction sensitized by low concentrations of ganglion blocking drugs. J. Pharm. Ph.arrnacol... 15. 45—55.MINSON. RB.. MCR1TCHIE. RI. Sc CHALMERS. i.P. (1989). Effects ofricuropeptide Y on the heart and circulation of the consciousrabbit. J Cardiorasc. PharmacoL, 14. 699—706.MOORE. PK.. AL-SWAYEH. O.A.. CT-lONG. N.W.S.. EVkNS. R.A. ScIn conclusion, the pressor effect of L-NNA does not rely onthe integrity of the autonomic nervous system or the riminangsotensin system. The bipbasic changes of HR induced byL-NNA are attributable to reflex alterations in the activities ofthe autonomic nervous system.This work was supported by the Medical Research Coun adCanada.GIBSON. A. (1990). L-N°-nitro-arginine (i..NOARO). a no’.d. i.arginine-reversible inhibitor of endothelium-depesidcut vasodon in euro. Br. J. Pharn,acoL. 9, 408-412.MUI.SCH. A & BUSSE. L (1990). N°-aitro-s-arginine (N5-f.(nitroamino8nethyl]..-ornnhine) impairs mdothe-dopeadentdilations by inhibiting cytosolic nitric oxide synthesis fro Larginine. Nwmyn.Schntiedebergs Arch. PharwacoL 341. 143-147.PALMER. LMJ. REES. D.D.. ASHTON. OS. Sc MONCADA, S. (17).Nitric oxide release aceounts for the biological acirrity ofendothelium-derived relaxing factor. Nature, 327, 524-526.PALMER. LM.J.. tEES. O.D.. ASHTON. OS. Sc MONCADA. S. (1w).L-argminc is the physiological precursor for the formation of erscoxide in esidothclium.depcndent relaxation. Biocl,em. ophys. lea.Cowdraui., 153. 1251-1256.REES O.D. PALMER. R.MJ.. HODSON. H.F. Sc MONCADA. S. (1989.).A specific inhibitor of nitric oxide formation auenendothelium.dependent relaxation. Br. J. PharmacoL. 96. 418-424.REES. D.D.. PALMER. R.MJ. Sc MONCADA, S. (1989b). Role ofendochclium-derived nitric oxide in the regulation of blood pen-sure. P’oc. .VctL .cad. Sd. US.A.. 86. 3375—3378.REES. 0.D.. PALMER. R.MJ.. SCHULZ. R.. i-{OOSON. HF. IMONCADA. S. (1990). Characterization of three nhibitors of edothelial nitric oxide synthesis in sino and tn rico. Br. J. PharmacoL.101. 746—732.SAKUMA. L STUEHR. 0.. GROSS. S.S.. NATHAN. C Sc LEVI. R. (1988).Identification of arginine as a precursor of endotheium-derivedrelaxing factor. Proc. .VatI. Acad. Sc: U.S.A_ 85, 3664-866.TABRIZCHI. R. KING. K.A. Sc PANG. C.C.Y. (1988). Presser response :0L and fl..blockers in conscious rats treated sth phentolamme.Pharmacology. 37. 385-393.TOGASHI. 14.. SAKIJMA. 1.. YOSHIOKA. M.. SATTO. H.. YASUDA K..GROSS. H. Sc LEVI. R. (1990). Sympatho-cxcitatory effect of a soectivC inhibitor of endothelium-derived nitric oxide synthesis. L.N’methvl-arginine. in anaesthctized rats. Eur. ./. Pkarn,.acot.. 183. 5O(abstract).VIDAL Ml.. ROMERO. IC. Sc VANHOUTTE. PM !988(. Endotheliumderi’.ed relaxing factor inhibits renin release. S-sr. .1. ?hwnac,i_149. .I1)—u).WANG. Y.X. Sc PANG. CCV. (1990). Pressor effect of N°-nitro-t.-arnnne ri pentabarbital.anaesthetized rats. Life Sci_ 37. 221—2.24.WHITTLE, BJ.R.. LOPEZ-BELMONTE. I Sc REES. 0.0.41989). Moduiariot’, of the vasodepressor actions of acetvlcholine. bradykmsn. sue-stance P and endothelin in the rat by a spectñc inhibitor of nit.coxide formation. Br. J. Pharmacot.. 98, 646-652.‘Recetred .Vortimber :9. 1C)Rerised .4pr:l :..4.:epred .4pr!i .‘3. l9‘1135Paper IIWang, Y.-X., Zhou, T. and Pang, C.C.Y.: Pressor effects of L and D enantiomersof NGnitroLarginine in conscious rats are antagonized by L- but not D-arginine.Eur. J. Pharmacol. 200: 77-81, 1 991. The reproduction of this paper was kindlypermitted by the copyright holder, Elsevier Science Publishers B.V., Amsterdam,The Netherlands.European Journal ot Pharmaculoiy, 21)1)) 199!) 77—sIc2 1991 Elsevier Soence Publishers 8.V. ()014.2999/9!/503.5l)ADONIS 001329999 I0042PEJP 51934136Pressor effects of L and D enantiomers of NG.nitro.arginine in conscious ratsare antagonized by L- but not D-arginineYong-Xiang Wang, Ting Zhou and Catherine C.Y. PangDepartment of Pharmacology and Therapezaicy, Faculty of Medicine. Unwersirj of British Columbia. 2176 Health Sciences Mall.Vancour BC, Canada V6T 1Z3Received 13 December 1990. revised MS received 22 March 1991, accepted 23 April 1991The effects of N°-nitro-L-arginine (L-NNA) and N°-riitro-D.arginine (D-NNA) on mean arterial pressure (MAP) werestudied in conscious, unrestrained rats. Lv. bolus of either L-NNA (1-64 trig/kg) or D-NNA (2-64 mg/kg.) dose dependentlyincreased MAP to similar maximum values of 55 ± 7 and 52 ± 4 mm Hg and with ED0 values of 3.0 ±119 and 8.9 ± 1.2 mg/kg(P <0.05), respectively. The time course of the MAP response to a single dose (32 mg,’kg i.v. bolus) of L-NNA and D-NNA werealso obtained. The pressor effects of L-NNA and D-NNA each tasted > 2 h with the rise phase t1 of 5 and 7 mm (P <0.05).respectively. [.v. infusions ([0 mg/kg per mm) of L-arginine (L-Arg) and D.arginine (D-Arg) did not alter the pressor responseto noradrenaline nor angiotensin H. L-Arg but not D-Arg attenuated the pressor responses to both L-NNA and D-NNA.Therefore, both L-NNA and D-NNA are efficacious and long-tasting pressor agents: the pressor effects of both can beantagonized by L-Arg but not D-Arg. Our results suggest that the pressor effects of both L-NNA and D-NNA inole theL-Arg/nitric oxide pathway.N°-Nitro-L-arginine: N°Nitro-D.arginine: L.Arginine: D-Arinine: Pressor effectsI. IntroductionIt is generally accepted that endothetium-derivedrelaxing factor (EDRF) is nitric oxide (NO) or NO-containing compound(s) (Ignarro et al.. 1987: Palmer etal., 1987: MYers et aL. 1990). NO is enzvmaticallvsynthesized from L-arginine (L-Arg) (Palmer et aL1988a: Sakuma et at.. 1988: Schmidt cc at.. 1988) andspontaneously released to maintain vascular tone (Keimand Schrader. 1990). NO snthase and/or endothehum-dependent vascular relaxation response in isolated arterial preparations can be inhibited by L-Arganalogues which include N-rnonomethyl- Larginine(L-MMA) (Palmer cc at.. 1088b: Rees et at.. 1989a.1990). NGnicro.L.arginine methyl ester (L-NAME)(Moore et at.. 1990). N-iminoethyl-L-ornithine (L-NIO)(Rees et at.. 1990) and N°-nitro-L-arginine (L-NNA)(Ishii et at.. 1990: Kobayashi and Hattori. 1990: Mooreet at.. 1990: Mülsch and Busse. 1990). In vivo studiesshow that L-MMA (Aisaka et at.. 1989: Rees et a!..Correspondence to: C.C.Y. Pang. Department of Pharmacology andTherapeutics. Faculty o Medictne. Uniersity if British Columbia.21Th Health Sciences MaO. Vancouver BC. Canada VT IZ3. Tel.,o4.S22.2O39: fax O4.2.t,i)l.1989b: 1990: Whittle et a!.. 1989: Gardiner et at..1990b.c: Kitbourn et at.. 1990). L-SAME (Gardiner etat.. 1990a.c.d). L-NIO (Rees et at.. 1990) and L-NNAWang and Pang. 1990) caused pressor effects and5radcardia.It has been reported that endothetial NO synthase isinhibited in an enantiomericalty specific manner byArg analogues. L but not the D eriantiomers of MMARees et at.. 1989a: Crawley et at.. 1990: Rees et at..1990). NAME (Rees et at.. 1990) and 510 (Rees et at..1990) inhibited endothetial NO formation and1 or endothelium-dependent relaxation. As well. L but not 0enantiomers of MMA (Rees cc at.. 1989a.b: Whittle etat.. 1989: Crawtey et at.. 1990: Persson et at.. 1990:Rees et a!.. 19Q0). NAME (Rees et at.. 1990) and SIC(Rees et at.. 1990) contracted isolated arterial prepara:ioris or raised blood pressure in intact animals. It hasaiso been reported that L but not 0 stereoisomers ofMMA. NAME and SIC enhanced platelet aggregation(Radomski et at.. 1990a.b: Persson et at.. 19Q0). Moreover. L-NNA but not 0-NSA prevented EDRF reease from endothelial cells and inhibited the dilatoreffect of acetvlcholine on rabbit femorat arteries(Mülsch and Busse. 1990) and. L-NNA but not 0-NSAinhibited non-adrenergic. non-cholinergic relaxation ofguinea pig isolated tracheal smooth muscle and rat13778anococcvgeus (Hobbs and Gibson, 1990; Tucker et al.,1990). In contrast, results from our preliminary studiesshow that i.v. bolus of D-NNA into pentobarbitalarzaesthetized rats also raised arterial pressure. It is notclear if the presser response to D-NNA is related tothe NO/EDRF system. In this study, we compare theeffects of L-NNA and D-NNA on arterial pressure inconscious rats and examine the ability of L- and D-Argto antagonize the presser effects of L- and D-NNA.2. Materials and methods2.1. Measurement of optical rotation of Arg and NNAL-Arg, D-Arg, L-NNA and D-NNA were dissolvedin 2 M HG. The specific rotation of the compoundswere measured at 586 nm (D line of sodium) by apolaruneter (Optical Activity Polarimeter, model AA1000).2.2. Surgical preparationsSprague-Dawley rats (250-400 g) were anesthetizedwith halothane (4% in air for induction and 2% in airfor maintenance). A polyethylene cannula (PESO) wasinserted into the left iliac artery for continuous measurement of mean arterial pressure (MAP) by a pressure transducer (P23DB, Gould Statham, CA. U.S.A.).Heart rate (HR) was determined electronically fromthe upstroke of the arterial pulse pressure using atachograph (Grass. Model 7P4G). A cannula was inserted into the left iliac vein for the administration oftest drugs. In some experiments, a cannula was alsoinserted into the right iliac vein for the infusion of Lor D-Arg. The rats were given 6 h recovery from theeffects of surgery and anesthesia before further use.2.3. Experimental protocols2.3.1. Time course of L-NVA and D-Nr’L4The rats in Group I, II and III (n = 6 each) weregiven i.v. bolus of normal saline (0.9% NaCI. 8 mI/kg)or a single dose (32 mg/kg. S mi/kg) of L-NNA orD-NNA. respectively. MAP and HR were continuouslymonitored for approximately 3 h.2.3.2. Dose-responses of L-VVA and D-VNACumulative dose-response curve of L-NNA was constructed in Group IV (n = 6) at dose intervals of 10-15mm (time required to obtain plateau MAP response).With D-NNA, it was difficult to construct a completedose-response curve in a single rat since the timerequired to obtain peak MAP varied between 20-90mm. with generally longer onsets of action for lowdoses of D-NNA. Therefore. 30 rats (Group V) wereused to obtain a dose-response curve of D-NA withonly one dose injected into each rat (n = 6 per dose).The MAP and HR responses to each dose were followed for 2 h.2.3.3. Antagonistic effects of L-Arg and D-Arg on presserresponse to L-NNA, D-NNA, noradrenaline and angiorensin IINormal saline (0.9% NaCI), D-Arg (10 mg/kg perruin) and L-Arg (10 mg/kg per mm) were continuouslyi.v. infused into rats in Groups VI, VII and VIII,respectively. Prior to and 20 ruin after the start of theinfusion of D-Arg in Group VII, noradrenaline (15jig/kg) was i.v. bolus injected into the rats. While inGroup VIII, angiotensin 11(100 ng/kg) was Lv. bolusinjected prior to and 20 mm after the start of theinfusion of L-Arg. In all three groups (VI, VU andVIII), a single dose of L-NNA (8 mg/kg) was i.v. bolusinjected at 40 ruin after start of normal saline, D-Argand L-Arg infusions, respectively. In Groups IX. X andXI, the protocol was similar to those in VI, VII and\i1II except that angiotensin H (same dose as before)was given to Group XI. noradrenaline (5 g/kg wasgiven to Group X and D-NNA (S mg/kg) instead ofL-NNA was given to all three groups. MAP was continuously monitored for 2 h after the injection of L-4NAor D-NNA.2.4. DrugsL.Arg HCI. D-Arg HC1, L-NNA. angiotensin H acetate and noradrerialine HCI were from Sigma Chemical Co. (MO, U.S.A.). D-NNA was from Bachem Bioscience Inc. (PA. U.S.A.). L-Arg and D-Arg were dissolved in distilled water and the pH of each solutionwas adjusted to 7.0 with 3 N NaOH solution. L- andD-NNA were solubilized in normal saline by 30 trimsonication. The rest of the drugs were also dissolved innormal saline.2.5. calculations and statisticsMAP and HR readings at peak MAP responses toeach dose of a pressor agent were noted. The ED0value and maximum response (E.) of L.NNA wereobtained from individual dose-response curves. ED;0and E of D-NNA were obtained from the entiregroup of 30 rats by the method of Litchfield andWilcoxon (1949). Rise phase ti,,, values were obtainedfrom individual time course curves. To obtain normaldistribution of rise phase t1,,,. data were logarithmically transformed prior to statistical analysis. Resultswere analyzed by analysis of variance followed by Duncan’s multiple range test with P <0.05 selected as thecriterion for statistical significance. All results wereexpressed as means S.E.M. except for the rise phaset/, which was expressed as mean ± 95% confidencerange.3. Results3.1. Optical rotation of Arg and NNAThe specific rotations [a]D of L-NNA and D-NNAwere +22.1° and —22.9° while those of L-Arg andD-Arg were + 20.20 and _21.40, respectively. Thus,the compounds are optically active with D-NNA andD-Arg levorotatory and L-NNA and L.Arg dextrorotatory.3.2. Time course of L-NNA and D-NNAThe control values of MAP and HR in Groups I, IIand III were summarized in table 1. I.v. bolus ofL-NNA (32 mg/kg) into conscious rats caused a sustained increase in MAP which reached plateau response at approximately 10 mm after injection, withrise phase ti,, of 5 mm (95% confidence range of 2-12mm). MAP at 30-150 nun was lower than that at 40mm, but was still significantly higher than control MAP.I.v. bolus of D-NNA (32 mg/kg) also increased MAPto a similar plateau value, but the onset of the response was significantly slower than that of L-NNA.Plateau MAP was reached approximately 50 mm afterinjection and remained at the steady state level for> 120 mm (fig. IA). The rise phase t1,, was 27 mm(95 confidence range of 15-48 mm). which was significantly longer than that of L.NNA. Both L-NNA andD-NNA caused bradycardia during the rise in MAP. Abiphasic HR response was only observed with L-NNAwhereby during the recovery of MAP, HR continued toGroup ii MAP (mm Hg) I-JR (beats/mm)Before AfterI 6 1033— 50 9II 6 i083— 3381lIII 6 1032—39 6IV 6 II14— 4O010V 30 1O71— 386 5VI 6 1O33 102±3—VII 6 1113 1O92—VIII 6 108±3 I112—IX 6 I135 1I23—X 6 104 103±3—XI 6 1043 1042-1/c0 40 30 ‘20 •3Ttme (mitt)Fig. 1. The effects (meansS.E.M. of i.v. bolus of aormal saiine(0.9% NaCI, open triangles). N°-nitro-L.arginine open circles. 32mg/kg) and0-nitro-D-arginine (closed circles. 32 mg, kg) on meanarterial pressure (MAP) and heart rate (HR) in conscious rats. N 6per group.rise to a level above control HR even when MAP wasstill elevated above control level (fig. IB). I.v. bolus ofthe same volume of normal saline did not significantlyalter MAP nor KR throughout the entire study period.3.3. Dose-response of L-1V4 and D-V.’vAThe control MAP and HR values in Groups IV andV are summarized in table 1. Lv. bolus of both L-NNAand D-NNA dose dependently increased MAP to similar maximum values (fig. 2A). The time taken fordifferent doses of L-NNA to reach peak vIAP response ranged from 10 to 20 mm while time (mm)taken to reach peak effect in response to D-NNAwere: 56 8 (for 3 mg/kg). 49 5 (8 mg/kg). 1 = S(16 mg/kg). 53 9 (32 mg/kg) and 38 4 (64 mg ‘kg).The ED0 of L-NNA (4.0 0.9 mg/kg) was significantly less than that of D-NNA (8.9 1.2 mg/kg). TheE of L-NNA (55 7 mm Hg) and D-NNA (52mm Hg) were similar. Both L.NNA and D-NNA reduced HR at the rise phase of the pressor responses.3.4. Antagonist effects of L-Arg and D-Arg on the pressor responses to L-,VtVA and D-.V.’4Control MAP in Groups VI—Xl are shown in table1. Lv. infusions of normal saline. L-Arg (10 mg/kg permm) and D-Arg (10 mg/kg per mm) did not alter13879600,40:::rBTABLE IBaseline values (means±S.E.M.) of mean arterial pressure (MAP)and heart rate (HR) in rats from Group I to XI. MAP was alsomeasured 20 mm after start of i.v. infusion of normal saline (GroupsVI and IX). D-Arg(10 mg/kg per mm. Groups VII and X)or D-Arg(10 mg/kg per mimi. Groups VIII and XI)tncreac in MAP) mm Hg)a bL •.-(rNA(2.5.g, kg) 30a4 32iAlltIOOrlgkg) 43 4::3D.,.4rgNA5g:kg) 58±4 583.Al1(100g, kg) 39±3 404Our results show that both L-NNA and D-NNAraise MAP in conscious rats. Pharrnacokinetic differences (rise phase t11, and ED) which exist betweenthe MAP effects of L- and D-NNA show that D-NNAis a slower and less potent pressor agent than L-NNA.The results here are in contrast with those of in vitrostudies which show D-NNA to be ineffective in causingcontraction of smooth muscles (Mülsch and Busse.1990: Hobbs and Gibson, 1990; Tucker et al., 1990). Ithas also been shown that other D-Arg analogues.namely D-MMA (Rees et al., 1989a.b; Whittle et aL.1989: Crawley et aL. 1990: Persson et al.. 1990; Rees etal.. 1990). D-NAME (Rees et al.. 1990) and D-N1O(Rees et al.. 1990) are ineffective in causing contractions in both in vitro and in vivo preparations.L-Arg did not modify the pressor effect of floradrenaline not- angiotensin II in conscious rats. Theseresults are consistent with those which show that L-Argdid not attenuate the pressor effects of noradrenalinenor angiotensin II in pentobarbital-anesthetized guineapigs (Aisaka et al.. 1989) nor vasopressin in consciousrats (Oat-diner et al.. 1990b). We have previously reported that iv. holus of L-Arg attenuated the pressoreffect of L-NNA in pentobarhital-anesthetized rats(Wang and Pang. 1990). In the present study. i.v.infusion of L-Arg but not D-Arg significantly attercuated the pressor effect of L-NNA. This is consistentwith reports that L-NNA is antagonized by L-Arg butnot D-Arg in vitro (Moore et al.. 1990: Miilsch et 31..1990). The results are also consistent with observationsthat L-MMA can be antagonized by L-Arg but notD-Arg in vitro (Rees et al.. 1989a) and in vivo (Rees eta!.. 1989b: Whittle et a!.. 1989: Gardiner et al.. 1990a:Persson et a!.. 1990). In contrast to the results whichshow stereospecificfty of L-Arg to antagonize the pressor effect of L-NNA. the pressor effect of D-NNA isantagonized by L-Arg but not D-Arg. Therefore. although pharrnacokinetic differences in onset and potency of actions exist between the pressor effects of Land D-NNA. the pressor responses of both compoundsare antagonized by L-Arg. It is of great interest thatL-Arg, NO pathway can be inhibited enantiomericailynon-specifically by both L and D enantiomers of NNA.We conclude here that both L-NNA and D-NNAare efficacious pressor agents in conscious rats. D-NNAhas sloer onset and longer duration of action and lesspotency than L.NNA. The pressor responses to L-NNA13960 r baseline MAP (table 1). L-Arg and D-Arg also did notA—S-—-----.ezZ alter MAP effects of noradrenaline or angiotensin II/ I (Table 2). L-Arg attenuated the pressor effects of.!. 40 / I r/ L-NNA and D-NNA while D-Arg did not alter the/ MAP effect of L-NNA nor D-NNA (fig. 3).•204. DiscussionB00C-300U-60 -CU-9000— 202 4 8 t6 32 64N —ntro—orqinine mq/kg)Fig. 2. Dose.response curves (means ±S.E.Mi of i.v. bolus of N°nitro-L-arginine (open circles) andN0-nitro-D-arginine (closed circles) on mean arterial pressure (MAP) and heart race HR inconscious rats. N — 6 each point.TABLE 2Mean arterial pressure (MAP) responses (meansS.E.M.) to i.v.bolus of noradrenaline (NA) and angiotensin TI (All) before (a) and20 mm after (b) iv. infusion (I)) mg; kg per minI of L.arginine(L-Arg) or D.arginine D-Arg) in conscious rats in = per group)Fig. 3. Effects (means a S.E.MJ of i.v. infusions of D.arginine (ii)mg, kg per mm: filled columns), normal saline (0.9” NaCI. shadedcolumns) and L.argin,ne (10 mg, kg per mm: cross-hatched columns)on pressor responses to i.v. bolus N°nitro-L-arginine iL-NNA. Smg kg) and Nu.nitro.D.argin,ne )D-NNA. S mg,kg) n consciousrats. N 6 per group. Denotes significant difference from thenormal saline group tP <0.1)5).and D-NNA can be specifically attenuated by L-Argbut not D-Arg. Our results suggest that the pressereffects ot’ both L-NNA and D-NNA involve the L-ArgNO pathway.AcknowledgementsThe authors appreciate the assistance of Dr. Kenneth Curry(Department of Physiology. University of British Columbia) for themeasurement of optical rotations of L-NNA. D-NNA. L-Arg andD-Arg. This work was supported by the Medical Research Council ofCanada.ReferencesA.isaka. K.. S.S. Gross, O.W. Griffith and R. Levi. 1989, N°-methylarginine. an inhibitor of endothelium-derived nitric oxide synthesis. is a potent pressor agent in the guinea pig: Dose nitric oxideregulate blood pressure in viva. Biochem. Biophys. R.es. Coin.man. 160. 881.Crawlev. D.E.. S.F. Liu. T.W. Evans and P.S. Barnes. 1990. Inhibitory role of endothelium-derived relaxing factor in rat andhuman pulmonary arteries. Br. 3. Pharmacol. 101. 166.Gardiner. S.M M. Compton and T. Bennett. 1990a. Regional andcardiac haemodynamic responses to glyceryl trinitrate. acetylcholine. bradykinin and endothelin-l in conscious rats: effects ofN°.rntro-L.arginine methyl ester. Br. 3. Pharmacol. 101. 632.Gardiner. S.M M. Comptom. T. Bennett. R.MJ. Palmer and S.Moncada. 1990b. Control of regional blood flow by endotheliumderived nitric oxide. Hypertension 15. 486.Gardiner. S.M M. Compton. T. Bennett. R.M..J. Palmer and S.Moncada. 1990c. Regional haemodvnamic changes during oralingestion of N °-monomethyl.L.arginine or NGnitro.L.argininemethyl ester in eonsctous Brattleboro rats. Br. 3. Pharmacol. 101.10.Gardiner. SM M. Compton. PA. Kemp and T. Bennett. 19%d.Regional and cardiac haemodynamic effects of N°-nitro-Larginine methyl ester in conscious. Long Evans rats. Br. 3. Pharmacol. 101. 525.Hobbs. A. and A. Gibson. 1990, L-Nt’-rutro.argtnine and its methylester are potent inhibitors of tion-adrenergic. non-cholinergictransmission in the rat anococcygeus. Br. 3. Pharmacol. 100. ‘49.Ignarro. Li.. G.M. Buga. KS. Wood. RE. Byrns and G. Chaudhuri.1987. Eridothelium-derived relaxing factor produced and releasedfrom artery and vein is nitric oxide. Proc. NatI. Acad Sci. U.S.A.83. 9265.lshii. K.. B. Chang. iF. Kerwin. Jr.. Z.-J. Huang and F. Muard. 1990.N-nitro.L.arginine: a potent inhibitor of endothelium-derivedrelaxing factor formation. European J. Pharmacol. 176. 219.Kelm. M. and 3. Schrader. iqo. Control of coronary tone by nitricoxide. Cire. Res. 66. 1561.Kilbourn. R.G.. S.S. Gross. A. Jubran. J. Adams. OW. Griffith. R.Levi and R.F. Lodato. 1990. N°-methvl.L-arginine inhibits tumornecrosis factor-induced hvpotension: Implications for the involve.tnient f nitric oxide. Proc. Nati. Acad. Sci. U.S.A. 37. 3629.Kubvashi. 1. and K. Hattori. 1990. >iitroargininc inhibits :ndothe.hum-derived relaxation. Jap. 3. Pharmacol. 52. 167.Litchfield. Jr.. L.T. and F. Wilcoxon. 1949. A simplified method ofevaluating dose.effect experiments. 1. Pharmacol. Exp. liter. 96.99.Moore. P.K.. O.A. Al-Swayeh. N.W.S. (Thong, R.A. Evans and A.Gibson. 1990. L-N-nicro arginine (L-NOARG). a noveL L.arginine-reversible inhibitor of endothelium-dependen vasodilaion in vitro. Br. 3. Pharmacol. 99, 408.Mülsch. A. and R. Busse. 1990, N°.nitro-L.arginine (N5.(iinino-(nitroamino)methyl}-L-ornithine) impairs endothelium-dependentdilations by inhibiting cytosolic nitric oxide synthesis from Larginine. Naunyn-Schmiedeb. Arch. PharmacoL 341. 143.Myers. P.R.. R.L Minor. Jr.. R. Guerra., Jr.. .1.R. Bates and D.G.Harrison, 1990, Vasorelaxant properties of the endothelium-derived relaxing factor more closely resemble S-nitrosocysteine thannitric oxide, Nature 345. 161.Palmer. R.MJ.. D.S. Ashton and S. Moncada. 1988a. Vascularendothelial cells synthesize nitric oxide from L.arginine. Nature333. 664.Palmer. R.MJ.. D.D. Rees, D.S. Ashton and S. Moncada. 1987.Nitric oxide release accounts for the biological activity of endochelium-derived relaxing factor. Nature 327. 524.Palmer. R.MJ., D.D. Rees, D.S. Ashton and S. Moncada. 1988b.L-Arginine is the physiological precursor for the formation ofnitric oxide in endotheliurn-dependent relaxation. Biochem. Biophys. Res. Commun. 153. 1251.Persson. MG.. L.E. Gustafsson. NP. Wikiund. P. Hedqvist and S.Moncada. 1990. Endogenous nitric oxide as a modulator of rabbitskeletal muscle microcirculation in vivo. Br. J. Pharmacol. 100.463.Radomski. MW.. R.M.J. Palmer and S. Moncada. 1990. Characterization of the L.arginine: nitric oxide pathway in human platelets.Br. J. Pharrnucol. 101. 325.Radomski. M.W.. R.MJ. Palmer and S. Moncada. 1990b. An Larginine/nitric oxide pathway present in human platelets regulates aggregation. Proc. NatI. Acad. Sci. U.S.A. 87. 5193.Rees. D.D.. R.M.J. Palmer. H.F. Hodson and S. Moncada. 1989a. Aspecific inhibitor of nitric oxide formation attenuates endothehum-dependent relaxation. Br. J. Pharmacol. 96. 418Rees. D.D. R.M.J. Palmer and S. Moncada. 1989b. Role of enidothehium-Jerived nitric oxide in the regulation of blood pressure.Proc. NatI. Acad. Sci. U.S.A. 86. 335.Rees. D.D.. R.M.J. Palmer. R. Schulz. H.F. Hodson and S. Mon.cada. Q90. Characterization of three inhibitors of endothelialnitric jde synthesis in vitro and in v1o. Br. J. Pharmacol. 11)1.45.Sakuma. 1.. D. Stauehr. S.S. Gross. C. Nathan and R. Levi. 1Q88.Identification of arginine as a precursor of enOothelium.derivedrelaxing factor. Proc. atl. Acad. Sci. U.S.A. 85. 8663.Schmidt. H.H.. H. Nau. W. Wittfoht. 3. Gerlach. K.E. Prescher.MM. Klein. F. Niroomand and E. Bohme. 1988. Arginine 5 aphysiological precursor of endothelium.derived nitric oxide. European S. Pharmacol. i54. 213.Tucker. iF.. S.R. Brave. C. Charalambous. AJ. Hobbs and A.Gibson. 1990. L-N°.riitro arginine inhibits non.adrenergic. noncholingic relaxations of guinea-pig isolated tracheal smooth muscle. Br. 3. Pharmacol. 100. 663.Wang. Y.X. and C.C.Y. Pang. 1990. Pressor effect of N°-ttitro-L.arginine n pentobarbital-anesthetized rats. Life Sci. .57. 22.Whittle. B.J.R.. 3. Lopez-Belmonte and DO. Rees. 1989. Modula:ion of :he vasopressor actions of acetylcholinc. bradkinin. sun-stance P and eridotheliri n the rats by a specitic nhibitor of nitricoxide formation. Br. J. Pharrnaol. 92. 56.140141Paper IllWang, Y.-X., Poon, C.I. and Pang, C.C.Y.: In vitro and ex vivo inhibitory effects ofL and D enantiomers of NGnitroarginine on endothelium-dependent relaxation ofrat aorta. J. Pharmacol. Exp. Ther. 265: 112-11 9, 1 993. The reproduction of thispaper was kindly permitted by the copyright holder, Willams & Wilkins, Baltimore,U.S.A.$22-.3686/93I266.4)t12Q3.OQ/OTse JOURNAL OC PNAa)ACOLOGT *140 EZPNV4TAL TNAPEtrnCsL9G3 by The Am.ncsn Society iae Phazacoogy tad zpen.waI Th.i.peIn Vitro and ex Vivo Inhibitory Effects of i.- and D-Enantiomersof NGNitroArginine on Endothelium-Dependent Relaxation ofRat Aorta1YONG-XIANG WANG, CHRISTINA I. POON and CATHERINE C. Y. PANGDepamnent of Pharmacology & Therapetstics, Faculty of Medicine, The Un&wslty of Brt.h Columbia. Vanco,ver, B.C., CanadaAccepted for publlcabon November 30, 1992The in vitro and ex vivo inhibitory effects of NGnitroLargInme(L.NNA) and NGnitrooarginine (o-NNA) on endothellum-dependant relaxations were studied in rat aortic rings. L-NNA (3 x 1Oto 3 x 10 M) but not o-NNA (3 x 10 to 3 x 10 M) inducedcontraction of resting aortic rings and potentiated phenylephrineinduced contraction in a concentration-dependent manner. Inphenylephnne-preconstricted aortic rings. L-NNA (3 x i0 to 3x 1O M) and o-NNA (3 x 10 to 3 x 10 M) concentration-dependently inhibited the rellaxation response to acetyicholine(ACh) with simar efficacies and lC values of 10 and 3.9 x10 M, respectively. In addition, both L-NNA (3 X iO M) ando-NNA (3 x 1 0 M) almost totally inhibited the relaxation ofpreconstricted rings by the calcium ionophore A 23187. Theinhibitory effects of L- and o-NNA remained for at least 4 hr afterthe preparations were washed out. Neither the flibitory effectsof L- and o-NNA on ACh-induced relaxation nor the ACh-inducedrelaxation itself were affected by pretreatment with indomettiacm. However, pretreatment (10 mm) or post-treatment (1 hr later)with L-Arg (10 M) completely prevented or markedly reversedthe inhibitory effects of L- and D-NNA. Intravenous bolus injecdons of L-NNA (1.6 x 10’ mrnol/kg) and o-NNA (1.6 x 1 0mmol/kg) caused sustained increases in blood pressure in cnscious. unrestrained rats in vivo and inhibited ACh-induced relaxation of aortic rings ex vivo. These findings suggest that both Land o-NNA cause efficacious, long-lasting and reversible inhibidon of endothelium-dependent relaxation, for which the L-enantiorneric form is the preferred but not essential configurationrequired to inhibit endothelium-dependent relaxation.Certain analogs of N°-substituted Arg have been shown toinhibit NO synthesis (see Moncada et at., 1991). These inhibitors include L-NMMA Palmer et at.. 1988: Rees et at.. 1989a,1990), L-NAME (Rees et at., 1990), L-NIO (Rees et at., 1990),L-NNA (Ishii et at., 1990; Mülsch and Busse, 1990) and L-NAA(Vargas et at., 1991). L-NNA inhibited endothelium-dependentrelaxations of isolated arteries (Kobayashi and Hattori, 1990:Moore et at.. 1990: Mülsch and Busse, 1990) as well as nonadrenergic, noncholinergic relaxations of isolated guinea pig trachea and rat anococcygeus (Hobbs and Gibson. 1990; Tuckeret at., 1990). L-NNA also caused pressor responses and reflexbradycardia in rats (Wang and Pang, 1990a. 1991; Wang et at.,1991a) and rabbits (Humphries et at., 1991). We have foundrecently that D-NNA is as efficacious as LNNA in raisingblood pressure in pentobarbital-anesthetized (Wang and Pang,1990b and conscious (Wang et at., 1991bi rats; however, theD-enantiomer is less potent and the effect is slower in onset.Received for publication January 24. t992.This work was *upported by a gram from the Medical Research Council ofCanada MRC) and a MRC feIlowhip to Y.-X. W.The pressor response to both D- and LNNA is prevented by Lbut not by D-Arg (Wang ec at., 1991b,.Our observations are unexpected inasmuch as other investigators have reported that N°-substituted Arg analogs exhibitstereospecificity such that the L- but not the 0-enantiomersraised blood pressure (Rees et at., 1989b. 1990; Gardiner et at.,1990a.b; Humphries et at., 1991; see Moncada et at.. 1991).These compounds are believed to suppress endothelium-dependent relaxations of blood vessels by inhibiting NO synthesis(see Moncada et at.. 1991). It is not known if D-NNA inhibitsvascular relaxation uia the same mechanism as L-NNA and ifsystemic administration of o-NNA is required for its constrictar action. The aim of this study was to examine 1) whetherboth c- and D-NNA inhibit endothelium-dependent relaxationinduced by ACh and the calcium ionophore A 23187. as well asendothelium-independent relaxation induced by SN?. in preconstricted rat aortic rings in i.’itro and ex iio and 2) whetherL-Arg or D-Arg antagonizes the inhibitory effects of L- and oNNA. As both c. and o-NNA cause prolonged pressor responses in .‘iuo (Wang and Pang, 199Gb; Wang et at., 1991b).142VaL 6. No. Ia U..L4.A8STRACTABBREVIATIONS: Arg, arginine: NO. nitric Oxide: NMMA. N°-monornethyl-arginine: NAME. N°-nitro-.-arginine methyl ester L-NlO. N-oetttyl-.on’ithine; NNA, N°.nitro-argoine: L.NAA, N°-arnino-i.arginine: ACh. acetylcitoline: SNP. sodium netroprussice: PHE. pitenylepririne: MAP, meazartenal oressure: EDAF, endothelium-denved relaxing factor.1121993143L.. and o-P#4A on RlaxiUon 113and L-N1O and L-NAA have been reported to be irreversibleNO synthase inhibitors (McCall et aL, 1991; Rouhani et aL.1992), experiments were conducted to study the time courseand reversibility of the inhibition of endothelium-dependentrelaxation by t.- and o-NNA. Lastly, the effects of L- and DNNA on ACh-induced relaxation were investigated because ithas been reported recently that cyclooxygenase inhibitors prevent L-NMMA from suppressing vasodilation induced by ACh(Roaenblum et at., 1992).MethodsPreparations. Male Wistar rats (350-450 g) were sacrificed by ablow on the head followed by exsanguination. The thoracic aorta wasremoved and cleared of connective tissue. Four ring segments of 0.5 cmlength were prepared from one aorta and suspended randomly inseparate organ baths. Each ring was connected to a Grass 71’-03-Cforce-displacement transducer (Quincy, MA) for isometric recording.Before the study commenced, a preload of 1 g was applied, after whichthe rings were equilibrated for 1 hr (with three waahouts) in Krebe’solution (pH 7.4) at 37’C and bubbled with a gas mixture of 95% 02-5% CO. The composition of Krebe’ solution was as follows (miflimotar): NaCI, 118; glucose. 11; KCI, 4.7; CaCh, 2.5; NaHCO, 25; KH2PO4,1.2; and MgCI26HO, 1.2.In the in ugro studies, the rings were incubated with the vehicle ordrugs (see later). Afterwards, 10 M PHE ECqo> was added to thebathe. At the steady-state phase of the contractile response to PHE(10—20 mm later), a cumulative concentration-response curve of AChor A 23187 was constructed. Each concentration of drug was left in thebath until a plateau response was obtained, The time taken to completeeach concentration-response curve was approximately 15 mm. In somegroups in which concentration-response curves of ACh were conductedmore than once, or followed by a concentration.response curve of SNP.the preparations were washed 3 times within 30 mm and given another30 mm to recover completely from the effects of PHE and ACh.In the ex ujuc studies, the rats were anesthetized with halothane (4%in air for induction. 1.5% in air for surgery) to allow the insertion ofcannulae into the left iliac vein and artery for the injections of drugsand continuous recording of MAP by a pressure transducer (P23DB.Gould Sr.atham, Cupertino, CA), respectively. The cannulae were tunnelled s.c. and exteriorized at the back of the neck. Afterward. the ratswere given at ieast a 4-hr recovery from the effects of anesthesia beforeuse and allowed free movement. After acclimatizing the rats for 20mm. the vehicle or drug was i.v. bolus injected into the rats. Fortyminutes later, the rats were sacrificed and two :horacic aortic ringsfrom each were prepared for ex uiuo studies as described for the in uitrostudies (the time elapsed between sacrificing the rats and applicationof PHE was 1 br).Drugs. t.-Arg hydrochloride, o-Arg hydrochloride, t.-NNA, PHEhydrochloride. A 23187 arid ACh chloride were obtained from SigmaChemical Co. (St. Louis, MO). o.NNA was from Bachem BioscienceInc. Philadelphia. PA). SNP was obtained from Fisher Scientific Co.Spritigfietd, NJ. t.-Arg and D-Arg were dissolved in distilled waterand the pH of each solution was adjusted to 7.0 with NaOH solution.A 23187 and indomethacin were dissolved in 100% dimethylsulfoxideand 80% ethanol, respectively, and diluted with normal saline (0.9%NaC1). The remaining drugs were dissolved in normal saline. Thedissolution of t.. and D-NNA required 20 mm of sonication.Experimental protocols. Six to seven aortic rings, each derivedfrom a different rat, were used in each group.Effects of L- and D-NNA on the relaxations induced by ACh,A 23187 and SN?. Concentration-response curves of ACh ) 10 to 3x 10 M. as follows) followed by those of SN? 3 x 10’° to 3 x 10Ml were performed in 11 groups of PHE.preconstricted aortic rings inthe presence of ‘iehicie, five concentrations of L-NNA (3 X 10’ to 3 xlO” M) and five concentrations of o-NNA (3 X 10” to 3 x 10 M).Vehicle, t.- or o.NNA were added 10 mm before adminisuatson atPHE.Three groups of aortic rings were incubated 10 mm with vehicle. -NNA (3 x 10 M) or o-NNA )3 x 10 M). followed by the constriztisof a concentration-relaxation response curve of A 23187 (3 x 10•’’ toi0” Ml in PHE-preconstncted aortic rings using the same procedineu described for ACh.TINe course of the inhibitory effect oft- and D-NNA on ACh‘aduced reIa,xation. Three group. of aoruc rings were iaed to -amine the time course of the effects of vehicle, t.NNA (3 x 10’ M)and o-NNA (3 x 10 MI on the relaxation response evoked by ACh.After completing the first ACh concentration-response curve and sásequent washout and recovery, L-NNA, D-NNA or vehicle wasinto the baths. This was followed 10 mm later by the construction atthe second curve of ACh. At 1.5 end 4 hr after the preparations wesewashed out, the third and fourth curves of ACh, respectively, weseconstructed without further adding vehicle, L.- or n-NNA.Effects of L-Arg, o-Arg or indomethaczn on relaxation response of ACh. The effect of pretreatment with indomnethacin (10’M) on the inhibitory effects of L-NNA (10’ MI and o-NNA (3 x 10”M) on ACh-induced relaxation were investigated in six group. atpreconstncted aortic rings. The concentration-response curves of AChwere constructed in the presence of vehicle vehicle. indotnethacin +vehicle, vehicle + t.-NNA, indomethacin ÷ r..-NNA, vehicle + o-NNAand indomethacin - D-NNA. The (Ira: drug or vehicle was given 10mm before the second drug or vehicle arid this was followed 10 mmlater by the addition of PHE.The effects of pretreatment with L-Arg or o-Arg on ACh-inducedrelaxation were studied in nine groups of PHE.preconstricted aorucrings, in the presence of vehicle = vehicle. L-Arg veniCie. D-Azg —vehicle, vehicle + t.-NNA. L.Arg — L.-NNA, o-Arg — L-NNA. vehicle —o-NNA, LArg - o-NNA and o-Arg + o-NNA. The concentrations forL- and o-NNA were 10” M and 3 x 10 M, respectively, and for i.Arg and o’Arg were M. The first treatment (vehicle, L-Azg or DArg) wsa given 10 mimi before the second treatment (vehicle. L- or 0-NNA( and this was followed 10 mm later by the addition of PHE.The ability of post-treatment with L-Arg to reverse the inhibitoryeffects of L- or n-NNA was also studied in four groups of PHEpreconstrtcted aortic rings. The preparations were ;ncubated with LSNA 10 Ml or o-NNA 3 x lO M( for 1 hr before adding r.ArgMl. After 10 mm. PHE was added followed by the constructionof concentration-response curves of ACh.Effects of L- and o-NNA on MAP and relaxations induced byACh and SNP ex uivo. Three groups of conscious and unrestrainedrats ri = .3 each group) were i.v. bolus injected with vehicle. L-NNA1.6 X 10 mol/kg) Zr D.NNA 1.6 x 10 mol.kgi. respectively. MAPwas recorded before and 40 mm after the injection of a drug or vehicle.The rats were sacrificed 40 ruin after injections and two aortic ringswere prepared from each rat for ex ‘ivo concentration-relaxation response curves of ACh or single concentration of SNP I 10 MI.Calculation and statistics. Responses of ACh. A 23187 and SN?were calculated as percentage of relaxation of contractile response toPHE. tCio values were calculated by using average data from concentration.response curves of t,. and D-NNA in inhibiting ACb-inducedreaxation. All results were expressed as mean S.E. except in casesin which the error bars were smaller than the points or symbols, seefigures). The results were analyzed by the analysis of variance,cov&nance. Duncan’s multiple range test was used to compare group means,with P < .05 selected as the criterion for statistical sigriiricance.ResultsEffects of L- and D-NNA on contraction in the presenceor absence of PHE. Ten-minute incubation with L-NNA 3x 10 to 3 x iO- M) potentiated the spontaneous contractileactivity in some but not all aortae (data not shown). In addition.L-NNA caused concentration-dependent contraction in restingaortic rings (fig. IA(. o-NNA )3 x 10 to 3 x 10 Ml. on the114 W.ng.taI.144Vof. 265FIg. 1. Effects (mean ± S.E.) of L- and o-NNA on resting tone of aorticrings (A), on PHE (1 0 M)-induced contraction (B) and on ACh (3 x 1 QMHnduced relaxation in PNE-peconstricted aortic rings (C) (n = 6—7 ineach group).other hand, induces neither spontaneous nor sustained contraction in the aortae (fig. 1A). PHE (1O, EC90)-inducedcontraction reached approximately 80% maximum within 30 to60 sec followed by a slower phase which reached plateau in 10to 20 rain. Preincubation with L-, but not with D-NNA, significantly potentiated the contraction induced by PHE (fig. 1B).Effects of L- and D-NNA on relaxations induced byACh, A 23187 and SNP. ACh and SN? caused concentration-dependent relaxations of PHE.preconstricted rat aorticrings, with maximum relaxation of approximately 70 and 100%,respectively (fig. 2). Incubations with L-NNA (3 X 10 to 3 X10 M) and o-NNA (3 x 10 to 3 x 10 M) concentration-dependently and noncompetitively inhibited the relaxation responses to ACh (figs. 2A and 3M. Figure 1C illustrates thepercentage of relaxation induced by 3 x 10’ M ACh in thepresence of L- or o-NNA. Whereas L- and o-NNA were equallyefficacious (approximately 100%) in inhibiting ACh-inducedrelaxation, the IC value of L-NNA (10 M) was lower thanthat of D-NNA (3.9 x 10 M). On the other hand, neither t.nor D-NNA inhibited the relaxation response of SN? (figs. 2Band 3B).A 23187 was as equally efficacious as ACh in causing concentration-dependent relaxation which reached a maximum ofapproximately 70% at 3 x 10’ M. Incubations with both LNNA (3 x 10’ M) and D-NNA (3 x 10 M) almost inhibitedcompletely the relaxation response induced by A 23187 (fig. 4).Time course of the inhibitory effects of L- and D-NNAon ACh-evoked relaxation. In the control group, the concentration-relaxation response curves of ACh were epeated 4Sodium nitropruaeo. (Id). t..ogFig. 2. Concentration-response (mean ± S.E.) of L.NNA on ACh (A)- andsodium nutropnssde (B)-induced relaxations in PHE (10 M)-preconstncted aotic rings (n 6-7 in each group).times within 6 hr. There was time-dependent loss of relaxationresponse to ACh which became statisticaUy significant at thelast curve fig. 5A). Incubations with both t.-NNA (3 x 10M) and D-NNA 3 X 10 M) completely abolished AChinduced relaxations (fig. 5. B and C). The inhibitory effects ofL- and o-NNA were still present at 1.5 as well as 4 hr after thepreparations were washed out without further adding the drugs.even compared to the corresponding time controls (fig. 5).Effects of pretreatment with L-, D-Arg or indotnethacm and post-treatment with L-Arg on relaxation responses of ACh. Indomethacin (10 M) did not alter ACh.induced relaxation in the preconstricted sortic rings, comparedwith the vehicle group ifig. 6A). L-NNA (10 M) and o-NNA(3 x i0 M) inhibited the relaxation evoked by ACh (fig. 6, Band C). Pretreatment with indomethacin did not alter theinhibitory effects of t-NNA (fig. 6B and D-NNA (fig. 6C).Incubation with neither L-Arg (10 M) nor D-Arg (10 M)significantly altered relaxation responses to.Ch (fig. 7A). Ten-minute preincubations with both L-NNA (10 M) and D-NNA(3 x 10 M significantly inhibited the relaxation responses ofACh fig. , B and C). The inhibitory effects of L- and D-NNAwere prevented completely by 10-mm pretreatment with L-Argbut not with D-Arg ifig. 7, B and C).Figure S shows that the relaxation response of ACh was againinhibited by 1.5-hr incubations with L-NNA (3 x l0 M) ando-NNA (3 x 10 M). The inhibitory effects of L- or o-NNAwere also markedly eliminated by post-treatment (1 hr later)with L.Arg i10 v) (fig. S. A and B).A0.1-0.1OConfrolA 0—NNAAII 1.51.20.920—20-60-100-141aaa*aI0—6 —7 —6—5 —4Acstylchoiln. (Id). Logar L—NNA (Id)0—20-60—100C.AC1O83x10610—i3x105N0—nitro—orginin. (Id). Log—Z —O —9 —8 —7 —61451993 i- and o-NNA on R.iaxatlon 1150-20-40-10-10I—e —7 —6—5—4Ac.ty(choiln. (U), LogD—NNAQ4)—CICaA20—20-60-100-140•a20—20-80—icc-140QConoI•v.1.5hafterwcehoutA 4haftrwaahcut-100 -20a a — — = a aQConb1•lSoft.rwo.flout IA 4 h aft.rwaiflout00• 3x10A 3x105o io—• 3a104a-20-40-60-SO-100200-20-40-60-80I—10 —9 —e —7—6ASodium nitropnaaaida (U), Log1.3 h after woiliout4 h aft.r woahout—5 —.4—e —7 —6Acwt4choIIn. (Id). Log0—20a-40-60—80o Vehicle• L—NNAD—NNAFig. 3. ConcenUation-response (mean ± SL) of o-NNA on ACh (A)- and Pig. 5. Time course of the effects (mean ± S.E.) of veflle (A), L-NNA 3sodium nitropn.isside (BHnduced relaxations in PHE (10 M)-peecon. x 10 M. 8) and o-NNA (3 x 10’ M. C) on ACh-induced relaxationstricted aortic rings (n = 6—7 in each group). responses in PHE (1 0 M)-peconstricted rat aortic rings (n = 6-7 ineach group). Signrflcant difference from the cono4 curve (P < .05).20 by the treatments with L- and D-NNA. Maximum relaxationsin response to SNP ( 10 M) in vehicle-. L- and D-NNA-treatedaortic rings were —107 ± 4, 99 ± 3 and 99 ± 3%. respectiveiy.DiscussionIt has been shown that c-NMMA Palmer et aL, 1988; R.eeset aL, 1989a; Rees et at., 1990; Crawley et at., 1990), L-NNA(Mülsch and Busse. 1990; Lamont.agne at uL, 1991), L-N1O(Rees et aL. 1990) and L-NAME (Rees et at., 1990), but not thecorresponding 0-enantiomers, inhibited endotheium-depend100—6 ent relaxations of isolated blood vessels and/or NO biosynthesis—10 —9—8 . in endothelial cells (see Moncada et aL, 1991). L-enantiomericA 23 187 .U). Logspecificity has also been reported to exist in other tissues orFig. 4Jnhib yeffects on L-NNA (3xiO M) and o-NNA (3x10’ M)cells, e.g., platelets (Radomski et at., 1990a,b), macrophagesrings (n 6—7 in each group). ‘Significant difference ft’om tte controi (McCall et at., 1991), adrenal cortex (Palacios et aL, 1989) andcurve (P < .05). nonvascular smooth muscles (Hobbs and Gibson, 1990; Tuckeret at., 1990). In contrast to these findings, our results indicateEffects of L- and n-NNA on MAP in uiuo and ACh- that both L- and D-NNA efficaciously inhibit the relaxationinduced relaxation ex viva. Base-line MAP of conscious response of ACh in uitro and ex uwo. Moreover, both corn-and unrestrained rats which were is. bolus injected with vehi- pounds inhibit the relaxation response of the calcium ionopboredc, L-NNA (1.6 x 10 mol/kg) and D-NNA (1.6 x i0— mol/ A 23187. These results suggest that both L- and D-NNA inhibitkg) were 100 ± 2, 107 ± 1 and 112 t 2 mm Hg, respectively. endothelium.dependent relaxation induced by receptor- andVehicle did not significantly alter MAP, whereas L- and o- nonreceptor-operated mechanisms, and that the L-enantiomNNA raised MAP to similar plateau values at 40 mm after eric configuration is not required for the actions of NNA.injections (fig. 9A). The relaxation responses to ACh in PHE- Our present results are in accordance with our previous inpreconstricted aortic rings obtained from either L- or D-NNA- viva findings which show that i.v. injections of both L- and Dpretreated rats were less than in vehicle-treated rats (fig. 9B). NNA cause pressor responses in pentobarbital-anesthetizedIn contrast, the relaxation response of SN? was not affected rats (Wang and Pang, 1990b) and conscious rats (Wang et aL.—12030o•-60 Q0-N4.• Indom.thocin+O—NN.A—90—120Fig. 8. Effects (mean ± S.E.) of a 10’min pretreatment with indomethacin(1 0” M) on ACh-induced relaxations in the absence (A) and presence ofL-NNA (1Q M, B) or o-NNA (3 x 10 M. C) in PHE (10’ M)-precoristncted rat aornc rings (n = 6—7 in each group).1991b. ft could be argued that the effectiveness of D-NNA isdue to contamination with L-NNA. However, there is no mistake about the identity of o-NNA as an independent analysisdetermined that the specific rotations, [aID, of 0- and L-NNAare —22.9’ and +22.1’C, respectively (Wang et at., 1991b). [aiDof D-NNA from our independent analysis is consistent with theinformation (lain —23.6’C) provided by the supplier, Bachem(Bubendorf, Switzerland). Moreover, other observations alsoindicate that the biological activities of o-NNA are not theresult of contamination by L-NNA. 1) D-NNA from anotherdrug company (Aminotech Ltd., Ontario, Canada) also exhibitssimilar biological activities (data not shown). We have alsoexamined o-NNA sent to us by investigators who have reportednegative results and found that the drug has activities indistinguishable from those of our supply of D-NNA (data not shown).2) There are differences in the biological activities between 1.-and o-NNA see below>. 3) The onset of the pressor effect ofo-NNA is markedly slower than that of L-NNA (Wang andPang, 1990b; Wang et at., 1991b). 4) L- and 0-NAME alsoinhibit the endothelium-dependent vasodilatation evoked byACh and/or calcitonin gene-related peptide in vitro and in uiuolAbdelrahman et at., 1992: Wang et at.. 1992). In addition, botht.- and D-NAME also cause pressor responses in conscious rats(Wang et at., 1992).The reasons for discrepancies between our results and thoseof others are not apparent but may be related to differences inconcentrations or doses of o-NNA used, duration of observationand possibly preconceived ideas. It is well known that although-12030• L—ki+L—NNAD-’tg+1-NNA0• L—+0—NNAD—Arg+0—P4NA—8 —7—8 —5 —4Acet.lchol1ne (Id). LoçFig. 7. Effects (mean ± S.E.) of a 10-mm pretreatment with L-Arg (10M) or o-Arg (10 M) on ACh-induced relaxations in the presence ofvehicle (A). L-NNA (1Q” M. B) or o-NNA (3 x 1O M. C) PHE (10”M)-preconsrncted rat aortic rings (n = 6-7 in each group). ‘Significantdifference from control curie (P < .05).the L-enantiomeric form is the main configurati.’n of biologically active drugs, many 0-enantiomers may have less or evengreater biological activities than their corresponding L.enantiomets see Ariëns, 1983), Inasmuch as the first report describingthe enantiomeric specificity of L-Arg as a substrate and LNMMA as an inhibitor, in which the same concentrations ofD- and L-NMMA were used (Palmer et at., 1988), the conceptof L-enantiomeric specificity for activating or inhibiting NOsynthase has become widely accepted (Moncada et at., 1991).Due to the preconceived notion that the D-enantiomers of N -substituted Arg derivatives are inactive, systematic studies werenot conducted with these compounds. The doses selected forthe 0-enantiomers of N°-substituted Arg analogs as controlswere always (without exception) the same as those of thecorresponding L-enantiomers. Moreover, conclusions were usually drawn without showing data. Among the work cited in thispaper the experimental conditions only in MLilsch and Busse’sreport (1990) are similar to ours. They found that L- but noto-NNA (both at 3 x i0’ M) produced approximately 80%inhibition of ACh-induced relaxation in norepinephrine (EC)preconstricted rabbit femoral arteries. In the present study, LNNA (3 x i0 M, supramaximal dose almost completelyinhibits ACh-induced relaxation in rat aortae. Because the nvitro potency of o-NNA is approximately 1/39 that of L-NNA(see later), it would be expected that o-NNA (3 x 10” M)should have caused considerably less response in the rabbitfemoral arteries. The potencies of N°-substituted L-Arg analogs116 wang.taI.A0 V.flicl.4’’.êiid.A146Vol. 2653o• L—ki#yW1ldI-1203033000-30I -30-60-90—8 —7 —8 —5 —4Ac.tylchoiln. (Id), Log—120Fig. 3. Effects (mean t S.E.) of post-treatment (1 hr) with L-Arg (10 M)on the nhibitory effects of L-NNA (10’ M, A) and o-NNA (3 x 10 M,B) on ACti.nducecj relaxation in PIIE (10 M)-peconstricted rat aocticrings (n 6-7 n each group). The rings were incubated for 1 hr with t•or o-NNA followed by a 10-man treatment with L-Arg or vehicte. S4gnifi-cant difference from contml curve (P < .05).are known to differ greatly according to particular preparationsand chemical stmctures see Moncada et at., 1991). Therefore.the potencies of 0-enantiomers should also vary with the preparations and types of compounds used. We found that L-NNAis 2-fold more potent than D-NNA in raising blood pressure(Wang cc at., 1991b) and 39-fold more potent than n-NSA ininhibiting endothelium-dependent relaxation (present work).L-NAME, on the other hand, is 55- and 359-fold more potentthat 0-NAME in raising blood pressure and inhibiting endothelium-dependent relaxation, respectively Wang cC at.. 1992).Moreover, pressor responses to D-NNA (Wang and Pang,1990b; Wang et at., 1991b) and o-NA.ME (Wang et at., 1992)are substantially slower in onset than the corresponding I.enantiomers, with this difference in onset being accentuated inanesthetized rats (Wang and Pang, 1990b. Thus, it is reasonable to assume that incorrect conclusions would be derivedwhen either the concentrations (or doses) of NG.substitutedArg analogs were insufficient or the observation time was notlong enough.Palmer et L 1988’l reported that cultured endothelial cellssynthesized NO from the terminal guanido nitrogen atom of L-,but not D-Arg. They also showed that L-Arg but not D-Argproduced endothelium-dependent relaxation of vascular ringsand inhibited endothelium-dependent contractions induced byL-NMMA (Palmer cC at., 1988). More recently, it was shownthat L- but not o-Arg attenuated the inhibitory effect of I..NSA (Moore et at., 1990). These results are in accordance withours which show that L- but not n-Arg prevents the inhibitory=EEIA60Acetyichoilin. (ha). iFig. 9. Effects (mean ± SE.) of i.v. bolus nlect(oris of vehicle. L-NNA(1.6 x 10 md/kg) and o-NNA (1.6 x 10’s rnol/kg) on MAP(A) and exvivo relaxation responses to ACh (B) in PHE (10 M).preansmctedaortic rings from the treated rats (n 5 fl each group). ‘Significantdifference from vehicle-treated group (P<.05).effect of L-NNA. As the effect of n-NSA is also prevented byL-Arg but not by o-Arg, the inhibitory effect of n-NSA onACh-induced relaxation, like that of L-NNA. may also involvethe inhibition of endothelial NO synthesis.Both L- and n-NSA caused prolonged inhibition >4 hr) ofin uitro relaxation responses to ACh. The long-lasting inhibitory effects of L- and n-NSA are also seen in ex uiuo studies.because the inhibitory effects on vascular preparations weretested approximately 1.5 hr after in uwo administrations of thedrugs and after three washouts. The long duration of action ofL-NNA is consistent with the report that L-NNA causes prolonged inhibition of NO synthesis in cultured endothelial cells(Mülsch and Busse, 1990). We have reported previously thatboth L- and n-NSA are long-lasting pressor agents (Wang andPang, 1990b; Wang et at., 1991b). Therefore, the prolongedbiological effects of L- and n-NSA on endothelium-dependentrelaxation may account for, at least in part, the long-lastingpressor effects of L- and n-NSA in uivo. Moreover, the pressoreffect of L-NNA was prevented but not reversed by L-Ag(Wang and Pang, 1990b; Wang et at.. 1991b: Zambetis et at..1991). These observations raise the possibility that inhibitoryeffects of L- and n-NSA are irreversible, it has been reportedthat L-NIO is a long-lasting and irreversible NO syrithaseinhibitor in rat peritoneal neutrophils and the munne macrophage cell-line J744, inasmuch as this effect was not reversedby L-Arg but was prevented by concomitant incubations of L510 with L-Arg (McCall et at., 1991). It was also reportedrecently that the inhibition of NO synthase by L-NAA was147.- and o-NNA on R4az.on 1171993A300 L.—NNA+vehlcl• L—NNA+L-frgaC-30t —— S-go-120B20 0 D—NNA+v.r.lca.-2o—60 I—100_ ‘ ‘-7 —8 ‘——440200-20a200-20—40-60-60D-NNAC00*0 o V.iici.• L—Nt4AD—NNAAcatyicholin. (U), LogIS—8 —7 —6 —5 —4118 Wangital.reversible mitially but became irreversible with time (Rouhaniet at., 1992). However, the in vitro inhibitory effects of L-NNA(and D-NNA) on ACh-induced relaxation, unlike those of LNIO and L-NAA, are prevented by L-Arg and reversed by I.Arg even after the preparations were incubated for 1 hr with Lor D-NNA.We found that indornethacin does riot affect the relaxationresponse of ACh. This is in accordance with studies demonstrating that prostagiandins do not account for effects ofEDRF/NO (Furchgott and Zawadzki, 1980). On the other hand.it was reported recently that the cyclooxygenase inhibitorsindomethacin and acetylsalicylic acid, and superoxide dismutase, blocked the effects of L-NMMA on contraction and AChand L-Arg-induced vasodilatations of pial arterioles and platelets adhesion/aggregation in mice in vivo. It was suggested thatL-NMMA interfered with endothelium-deperident relaxationand produced constriction by activating cyclooxygenase andproducing superoxide which subsequently inactivated EDRF/NO (Rosenblum et at., 1992). Such findings are contrary toprevious results which showed that indomethacin does notinhibit the pressor response to L-NMMA (Rees et at., 1989b).Indomethacin has been frequently added to the physiologicalsolution in order to avoid a possible contribution by prostaglandins to endothelium-dependent relaxations (e.g., Mülschand Busse. 1990). However, our results show that indomethacm, at a concentration high enough to inhibit prostaglandinsynthesis, does not alter the inhibitory effects of L- and 0-NNA. These results suggest that cyclooxygenase activation andsubsequent superoxide production and inactivation of EDRF/NO does not account for the inhibitory effects of L- and DNNA on ACh-induced relaxation.There are differences in the vasoconstrjctor effects betweenc- and o-NNA. First, L-NNA concentration-dependently contracts aortic rings and potentiates PHE-induced contraction.Although D-NNA is as efficacious as L-NNA in inhibiting AChinduced relaxation, it does not induce contraction of aorticrings or potentiate PHE-induced contraction. It has been reported that the concentrations of L-NNA and L-NMvIA thatwere maximally effective at increasing tension in canine coronary arteries only caused submaxirnal innibitions of AChinduced relaxations (Cocks and Angus. 1991). In the presentstudy, 10° M L-NNA produces maximum inhibition of AChinduced relaxation but does not produce maximum contractileresponse. The contractile effect of L-NMMA was found to beendothelium-dependent and reversed by L-Arg suggesting thatthis response was caused by the inhibition of basal NO formation (Palmer et at., 1988; Rees et at., 1989a. In contrast, Cocksand Angus (1991) recently showed that the contractile responseof L-NMMA in dog coronary arteries was not affected bypretreatment with hemoglobin or FeSO4 in concentrations thatinhibited relaxations induced by SNP and NO, suggesting thatthe contractile response of L-NMMA was independent of basalNO formation. Moreover, t.-Arg was reported to reverse LNAME-induced augmentation of contractions evoked by 5-hydroxytryptamine and histamine. but not L-NAME-inducedinhibition of endothelium-dependent vasodilatation evoked byACh in perfused rabbit ear preparations (Randall and Griffith,1991>. We have also found that L- but not D-NNA caused aslow and sustained contraction in denuded rat aortic rings: theeffect is not affected by endothelium and L-Arg tWang andPang, unpublished data. 1992). These results suggest that contraction and inhibition of relaxation responses of N°-substi148Vol. 2.65tuted Arg derivatives may be produced by different mechanisms.Another difference between L- and o-NNA is potency. Asindicated above, although both compounds have similar efficacy, o-NNA is less potent than L-NNA in inhibiting endothehum-dependent relaxation, suggesting that the vasoconstrictoreffects of NNA prefer the L-enantiomeric conguzation.. Moreover, the difference in potencies between D- and L-NNA invitro is higher than those in vivo. The mechanism responslhefor this discrepancy between the in vitro and in woo potency ofD- and L-NNA is not known. One possible explanation is chiralconversion. Metabolic chiral inversion baa been shown to ourafter the p.o. administration of stereospecific drugs (Hutt andCaldweU, 1983; Sanins et at., 1991). Because D-NNA is lesspotent and has a slower onset of action than L-NNA in uwo(Wang et at., 1991b), D-NNA may act via metabolic conversionto L-NNA in vivo, thus accounting for the difference in theactivity ratios of D- and t.-NNA between in vito and in vitrosettings. It may also be speculated that the difference is atttibutable to variations in affinity ratios for o- and L-NNA withrespect to conductance and resistance arteries. More studiesare required to resolve this puzzle.In summary, both L- and o-NNA are selective, efficacious,long-lasting and reversible inhibitors of endothelium-dependent relaxation responses evoked by receptor-operated and non-receptor-operated mechanisms. However. D-NNA is less potentthan r..-NNA in inhibiting the relaxation response of ACh and.unlike L-NNA, does not produce a contractile response orpotentiate PHE-induced contraction in isolated aortic rings.Our results suggest that the L-configuratiOn of N°-substitutedArg analogs is preferred but not essential for the inhibition ofendothelium-dependent relaxation.AcknowledgmentsThe outhorn would ike to thank Ms. Su Liu E.im for her aallled techmcalosaatance.ReferencesA.. WANG. Y..X.. CHANG. S. 0. AND P.G. C. C. V.: Mechanismo the asodilazor act:on of calcztonzn gene rejated Deptide in conscious rats.Sr. J Pharnacol. 106: 45—48. 992.AR:ENS. 2. 1.: Stereoselectivity of bioactive agents: General aspects. In Stereochemistry and Biological Activity of Drugs. ed. ny by S. I. Axiens, W. Souai:nand P B.M.W. M. Timmermaris. pp. 11—32. Blackwell Scientific Publications.Otforc. 1983.CocKs. T. M. AND ANGUS. .J Evidence that contractions of :solatad arteriesby L.NMMA and NOLA are not due to inhibition of basal SDRF release. J.Cardiovasc. Pharmacol. 17: Suppl. 3. S159—S164. :991.Csawt.t. 0. 5.. Lw. S. 5.. EVANS. r. w. AND B.sNEs. P. 3.: Inhibitory role ofndochelium.derived relaxing factor in rat and human pulmonary arteries. Sr..1. PharmacoL lOt: 166—170, 1990.FURCHG0rr. II. F. AriD ZAWADZKI. J. V.: The obligatory role of endotheial cellsin the relaxation of arterial smooth muscle by acervicholine. Nature Lond.,299: 373—.36. 1280.GARDINER. S. M.. CouproN BENNETT. T.. PAL.MER. R. M.J..c.DMONCADA, S.: Control of regional 1ow by endotbeiium.derived nitric ozida.ilvpertenaion 15: 486—492. 1990a.GARDINER. S CoMvroN. A. M.. KEMP, P 4ND 3ENrigrr. T.: Regicosaand cardiac haemodnamic effects of N°. n.itro-L-arguune methyl ester nconscious. Long Evans rats. Br... Pbarm.acol. 101: 625.-,331. 990b.Hoaas. A. AND GIBSON L.N°.rutro-arginme nd its methYl ester are notitrthibitors of rion.adrenergic. non-choilnergic transmission is e rat anococcygeus. Br. .3. ?harmacol. 100: 749—752, 1990.HUMPHRIES. 2. 0.. CARE. 2. 0.. Nrcoi.. A. K.. TOML:NSON. W, .sriD O’CONNOR.S. S.: Coronary vasoconatriction in th. conscious raoDit following intravenousinfusion of L.N°.nitro-arginine. Br. J. PharmacoL 102: 565—566. 1991.KUTT. A. J. AND CALD WELL. J.: The metabolic chirai inversion ot2.aryipropioniasc:ds: A cocci route with pharmacological consequences. 3. Pharm. PhIImSCOL35: ‘593—704. 1283.ISHII. K.. CHANG. 3.. Kgaw!N. J. F. Ja.. HUANG. Z.-J. AND MUARD. F.: N-.rutro.L’srginine: A potent inhibitor of endocheiium.dertved relaxing factorformation. Eur. j. Pharmacol. 176: 219—223, 1990.1993KosayiSHI. V. AND HAT!’ORI. K.; Nitrosrr.nine inhibiti endothelium-derivedrelaxation. Jpn..3. PharmacoL 52: 167-169,1990.LAMONTAGNE. 0.. POHL, U. AND Russe. R.: N°nicro-L..rginme antagonizesendocheIium-de.ndent dilator reeponsea by inhibiting ehalium.d.eriv.drelaxing factor ielea.e in the isolated rabbit besot. PflUgers Arch. 418: 266-270. 1991MCCtU.. T. B.. Friusx, M., PALMZI. R. M. .3. urn MoNcin. S.: d.ntificatiotofN-minoethyj.L.ornjghinsaesn ibi hibitor of nitric oxide aynthae.in pbqocydc mile. Br. .1. Phasmecol. 102: 234..238. 1991.MoNcana. S., Pausu, 8. M..3. urn Hxoos, 8.. Nitric ozi& PhyiioIoev,pathopbysiolc end pharmacolor. PharmacoL Rev. 43:109-142,1991.Moose. P. K.. AL-SWATIK. 0. A., CH0NG. N. W. S., EVANS, 8. A. AND GIBSON.A.: L-N°.nitro uginine (L-NOARG), 5 novel L-srgmine-r,versjbl. inhibitorof endothelium.dep.ndera v—wSla’inn in ogre. Br. .3. PbarmacoL 99: 408-412, 1990.MOLScH. A. urn Busag, R.: N°-nitro-L-arginine (N-(imino (niboemiSO)methylJ.L.cenithina) impede endotheilum-dependent 1ationa by inhibitingcytoeolic nitric oxide aynthaeia from L-atginin.. Naunyu-Schmiadeberg’a Arch.Pbannacol. 341: 142—147. 1990.PALACIOS, M., KNOWLBS. R. G.. Pu., 8. M. aND M0NCina. S.: Nitric oxidefrom L-arginin. .timulatee the soluble guanylate cyclaae in ethenal glande.Biocham. Biophy.. Rae. Commun. 185: 802-809,1969.PALMEIt. R. M.J.. Rita, 0.0.. MHTON, 0.5. aND MONCADi, S.: L-erginine Iathe phyiiolvwica) precursor for the formation of noxic oxide in edothelium.dependent relaxation. Biochem. Biophys. Re.. Conintun. 153:1251-1.256,1968.Raooieszi. M. W., PALMER, R. M..3. AND Mo*.caoa, S.: An L-arginine to nitricoxide pathway in human platelete regulates aggregation. Ptot. NatL Aced. Sd.U.S.A. 87:5193-5197,1990..Ranoeesxi. M. W., Pax.tezi, R. M. 3. aND MONCADA, S.: Characterization ofthe L-arrinine Nitric oxide pathway in human platelet.. Br. J. Ph.rmscal.101:325-328,199Gb.RANDALL. M.D. urn Gawvrrn, T. M.: Differential effect. of L-arginie on theinhibition by N itro-L.atgznme methyl ester ofbesal end agouiat.etmmulatedEDRY activity. Br.J. Pharniacol. 104:743-749. 1991.REEx. 0. 0., Paiizji. R. M. .1., HoosoN, H. F. AND MONCADA, S.: A epeci5cinhibitor of nitric oxide formation from L.argimne attenuate. endotheliumdependent relaxation. Br..3. PhaZDII.COL 98:418—424. 1989..Rzxs. D. 0., Paijota. R. M AND MOr4CADA, S.: The role of endotheiumderived nitric oxide in the regulation of blood pressure. Poc. Nati. Aced. Sd.U.S.A. 86: 3375-3378, 1989b.Razz, 0. D.. R. M. J.. SCHULZ, R.. HODSON. H. F. AND MONCADA. S.:Characterization of three inhibitors of endotlzelial nitric oxide iynthesis inuzt,’o and in own. Br. 4. Pharmacol. 101: 46—752. 1990.. and o-NNA on Ralaxatlon149119ROseNeLUIe. W. L NIxHluuaa. H. AND NELSON. G. H.: L.NMMA in brainmicrocirculation of mice is inhibited by blockade of rycloozypna.e d bysioperotid. diamuta... Am. 3. PhyiroL 262: H1343- 1349, 1992.ROUHAHI, B.. Gaoss. S. S.. Manana. A. M.. Levi. B. aNn Geipime. 0. W.: N-methyl aubetitutad derivatives of N-amino-L-erguiin. a. reversible end irisversible inhibitors of nitric oxide synchaae (AXrst). Fad. Ani. Soc. Zap. Biol.3. 8: A1256, 1992.SueIr4s, S. M., Anusa, W.J., Luau. 0. G., HuzreAD, G. W., Hou.zy,.3.,BeaNu, H. AND R.tei. T. A.: Mechanistic suodlee an the metabodecinversion of R-ibuprofen in the rat. Drug Metab. Diepce. 19:406-410.1991.Tuczza,J. F.. Beavz. S. B., Cicae.u.AMsous, C., Hoa A.). urn GoN,A.: f.N°.nitro arginine inhibits non-edrenergic, oca-cholingic ofguinee.pig isolated tsecheal smooth mcecIa Br. .3. ?‘nsrmacoL 100 681464.1990.H. 36., Cugvu,J. 36., IGHAARO, C.). urn Cxwoauez, G.: C—i--——of the I bibitory pcej of N°.mrJyI., N°.nitro. and N°.amino-L.c—mneon EZIRF function in the rat Evidence for “—...--, besal EDRP rele.J.Pharmacol. Hap. Tb.r. 257: 1206- 1.215, 1991.WANG, Y.-X. urn Pueo, C. C. Y.: Ptes.or effect of inpentobarbitsl.anesthetized rats. Life Sd. 47: 17-2224, 1990..WANG, Y..X. aND PANG, C. C. Y.: Effects of C ad 0 enantiomers of N°-uitioL.arginineon blood prenezis in pentobarbital-anae.thadzed ret.. Life Sd. Adv.9: 723—728, 199Gb.WANG, Y..X. AND PANG. C. C. Y.: Po.aihl. dependence of preesor and heart ret.effect. of N°.nitro-L-arginine on autonotnic nerve activity. Br. J. Pharmecol.103:2004-2008,1991.WANG, Y.-X., POOH, C. L YAU. C. C. V. ariD PANG, C. C. V.: ‘The in Vice andin Viuu Vs.oconetijctor Effects of Land 0 EnantioniersofN°.N’itro-ArginbieMethyl Sisar (NAME)abeuact. p.80, The 36th Animal Meeting of CaimlisnPedention of Biological Societies, Victoria, Canada, 1992.WANG, Y.-X.. Zaou, T., CHou. 1’. C. AND PANG, C. C. V.: Effect. of inhalationand intravenous anaesthetic agents on preseor respons. to N°-uitro.L.ergmine.Hur.J. Pharmacol. 198: 183..188, 1991..WANG. Y..X., ZHOU, T. aND PANG, C. C. V.: Presaor effecta of L and 0enantiomers of ‘nitr’3gz’.e in conscioua rats are antagonized by C- butnot D-aglmne. Eur. .3. PIISOIaCOL 200: 77-81. 1991b.ZA.Marris. M., DUSTING. G. J., R.A.JANAYAGAM. S. AND WOODMAN. 0. C.:Mechanism of the hypertension produced by inhibition of nitric oxide biceynthesis in rats. 3. Cardiovasc. Pharmacol. 17: SuppL 3. 5191-5197, 1991.Send reprint request. tnt Dr. Catherine C. 1. Pang, Department of Pharma.colon & Therapeutics. Faculty of Medicine. Th. University of British Columbia.2176 Health Sciences Mall. Vancouver. B. C.. V6T 123. Canada.150Paper IVWang, Y.-X. and Pang, C.C.Y.: Functional integrity of the central and sympatheticnervous systems is a prerequisite for pressor and tachycardic effects ofdiphenyleneiodonium, a novel inhibitor of nitric oxide synthase. J. Pharmacol.Exp. Ther. 265: 263-272, 1 993. The reproduction of this paper was kindlypermitted by the copyright holder, Willams & Wilkins, Baltimore, U.S.A.TNI JOJ5N*i. 0V PNA*ACOLOGY AND EXP55IIV4?AL THUAPUT!C*© t93 ii The Simy (or Phareo. aed Expen.ncaFunctional Integrity of the Central and Sympathetic NervousSystems is a Prerequisite for Pressor and Tachycardic Effects ofDiphenyleneiodonium, a Novel Inhibitor of Nitric Oxide Synthase’YONG-XIANG WANG and CATHERINE C. Y. PANGDepattment of i°he’macclcgy & ThereutIcs, Faculty of Medicine, The Uniewity & BntAsh Columbia Vancouver, BritLsh Columbia, CanadaAccepted for publication November 23, 1992The pressor and tachycardlc effects of diphenyieneiodonium(DPI), a novel inhibitor of endothelial nitric oxide synthase withchemical structure different from those of N°-substituted Arganalogs, were studied in pentobarbital-anesthetized rats. Bolusinjections of DPI (0.05—1.6 mg/kg i.v.) caused transient (1—2 mmin duration) and dose-dependent increases in mean artenal pressure (MAP) with ED of 0.22 ± 0.02 mg/kg arid maximum effect(E,,,.0) of 58 ± 3 mm Hg, arid heart rate (HR) with ED of 0.26 ±0.03 1119/kg and E., of 60 ± 5 beats/mm. Pretreatments withtetrodotoxin, reserpine, guanethidine. mecamytamirie, but notatropine, rauwoiscine, captopnl nor L-Arg, attenuated the MAPand HR responses to DPI. Phentolamine and prazosiri attenuatedthe MAP but not HR response whereas propranolol attenuatedthe HR but not MAP response of DPI. Pithing abolished, whereasspinal cord transection reduced, the MAP and HR responses toDPI. Ptthng did not alter the presser response but blocked thereflex bradycardic response to N°-nitro-t.-arginine meth1 ester,an inhibitor of nibic oxide synthase. Bolus injection of a sinedose of DPI (1.6mg/kg i.v.) or N°-nitro-L-arginine increased MAP,but only DPI caused immediate and large ncreases (>1 ncjml)in plasma rorepineptinne. epinephrine and moderate increase mdopamirie; pretreatment with reserpine attenuated, whereas pi.tiling aboiisfled these increases. The increases in plasma norep4nepvlnne and epinephrine by DPI were positively correlated toincreases ri MAP and HR. The results demonstrate that DPI,unlike N°-substituted Arg analogs, produces presser and tachycardic effects via indirect activation of the sympathetic nervoussystem.The fIrst class of inhibitors of nitric oxide (NO) synthase arethe N°-substituted Arg analogs, which include L-NMMA, LNNA, L-NAME, L-NIO and L-NAA (see Moncada et at., 1991).These compounds suppress in uitro endothelium-dependentrelaxation and produce prolonged pressor and bradycardic responses in whole animals (Aisaka et at., 1989; Rees et at., 1989,1990; Wang and Pang, 1990a. b; Wang et at., 1991a, b; Wang etat., 1992, 1993a,b). The pressor responses induced by theseagents are antagonized by L-Arg (Aisaka et at., 1989; Rees etat., 1989; Wang et at., 1990b, 1991b, 1992) but are insensitiveto inhibitions of activities of the central nervous system (Tabrizchi and Triggle, 1992), sympathetic nervous System Wangand Pang, 1991; Aisaka et aL, 1989; Rees et at.. 1989) and reninangiotensin system (Wang and Pang, 1991). The pressor response has been attributed to the inhibition of the L.Arg/NOpathway which leads to endothelium-dependent relaxation (seeMoncada et at., 1991>; the bradycardic response is shown to beReceived for publication July 13, 1992This work was upporteci by & cant from the Medical Researcb Council ofCanada MRC) and a MRC Postdocwrsi Felloweh.ip o Y..X.W.mediated by baroreflex mechanisms (Wang and Pang, 1991:Widdop et at., 1992).DPI is a bivalent iodine compound. Its chemical strdcture isdistinct om those of N°•substituted Arg analogs g. 1. DPIwas first reported to be a potent hypoglyccmic agent Stewartand Hanly. 1969; Gatley and Martin. 1979). It was later shownto suppress activities of neutrophil and macrophage NADPHdependent oxidase Cross and Jones, 1986: Ellis et at., 1989).probably by the inhibition of a ilavoprotein (Hancock andJones, 1987; Ellis et at., 1989). Recently, DPI and its analogswere reported to inhibit macrophage NO synthase Stuebr erai.. 1991. This is not surprising because NO synthase is also aNADPH.dependent enzyme which requires FAD as a cofactor(Stuehr er at.. 1989, 1990). DPI caused long-lasting suppressionof endotnelium-dependent relaxation in rabbit Stuehr em a..1991) and rat )Poon et at., 1993) aortae and inhibition fendotheiium-dependent vasodilatation in uiuo iPoon ef at..1993).Although the in tüo cardiovascular effects of N°-substitutedArg analogs have been intensely investigated isee Moncada et151VoL 25..‘ 1a U.34.ABSTRACTABBREVIATIONS: NO, nitric ox,ie: .-NMMA. N°-monomethyl-L.arglnine; L-NNA. N -itro-i.’argiriine: L-NAME. N°-nllrc-L-&gInine methyl esterNl0, L-4mlnoethyloi’nithine: L-NAA. N°.amlno-L-arglnlne: DPI, diphenyleneioaonium: MAP, mean arteral pressure; TTX. tetrootOxIn; E.,..,,, mininuneffect: L.,, maximum effect; n,,, 111Il coefficient: -fR. heart rate; 0, dose: C, concentration.263264 wang and PangdiphenyleneiodoniumSurgical PreparationMaterials and MethodsMeasurement of Plasma Catecholamines. 265was described by Passon and Peuler (1973). Briefly, catechol-O-methyltzansferase was used to catalyze the transfer of a12N]methyl pipfrom S-adenoayl-t.-[methyl-’H].methiousne to the hydrozyl grcsp znthe 3-position of riorepinephrine, epLnephrine and dopamthe. Theresultant products were separated by thin layer chromatography, eldif necessary, end counted by a 1600 TR liquid scintillation analy(Packazd Instrument Co., Cr).Blood samples (0.5 ml) were immediately inserted into prodtubes containing EGTA and reduced glutatbione and centrif1,200 x g at 4C. Afterward, the plasma was removed and ored at—70C until essayed within a month. Duplicate assays were run for thestandatd, plasma or diluted plasma samples (50 d each sample) withdistilled water used as a blank control for each run. The “‘dcurves for norepinepbrine, epinephzine and dopamine (0.01, 0.03, 0.1,0.3, 1. 3 and 10 ag/mI for each standard solution) were prepared withcontrol rat plasma. Two standard curves were constructed at twoseparate occasions and were found to be indistinguishable from eachother. The data were combined to formulate the following lirregression equations for norepinephrine, epinephrine and dopamine Y.&03x+29.5(r-.0.998P<.0Y=2.49x-14.1(r’0.999P<.05) and Y 2.99x + 10.2 (r 0.999; P < .05), respectively. Thesensitivity of the catecholamine assay was 0.006 ng/mLDrugsThe following drugs were purchased from Sigma Chemical Ca. (St.Louis, MO): mecamylamine hydrochloride. atropine sulfate. o-t..propranolol hydrochloride, L-NNA. L-NAME hydrochloride. L.Azg hydrochloride, guanethidine hydrochloride, rauwoLscine hydrochloride andprazosin hydrochloride. The following drugs were also used: DPI sulfate(Colour Your Enzyme Ltd., Ontario. Canada), phentolamine hydrochloride (CIBA Pharmaceutical Co., NJ). captopril (E. R. Squibb &Sons Inc., NJ), reserpine injection (CIBA Pharmaceutical Co., Quebec,Canada) and TTX Sankyo Co. Ltd., Tokyo. Japan). The powder drugswere dissolved in normal saline (0.9% NaCI solution) except for DPIand t.-NNA which were solubilized in 5% glucose solution by 10 mmof 3onication and prazosin which was dissolved in 100% dimethylsulfoxide.Experimental ProtocolThe rats ri = 6 in each group except for one group with ii = ‘ asindicated) were randomly assigned into groups. The MAP and HRresponses were continuously monitored throughout the experiments.Protocol 1: Dose-responses of DPI on MAP and HR. Effects ofi.v. bolus injections of vehicle (5% glucose solution, up to 2 mI/kg) andDPI 0.05—1.6 mg/kg) on MAP and HR were examined in one group ofrats. The intervals of doses were 3 toö mm, which was required for therats to recover completely from the effects of the previous dose.Protocol 2: Effects of antagonists on the MAP and HR responses of DPI. Eleven groups of rats were pretreated with vehiclenorma.1 saline. 1 mi/kg). reserpine i5 mg/kg), mecamylamine (10 mg/kg). guanethidine (10 mg/kgj, phentolamine (10 mg/kg), prazosin ‘1mg/kgi. rauwolscine (1 mg/kg), propranolol (1 mg/kg). atropune 10mg/kgi, captopril (20 mg/kg) or L.Arg 400 mg/kg). The doses selectedfor the antagonists including TTX. see protocoL 3) were those prei.ousiv shown to block effectively the corresponding receptors, enzymesor ionic channels (Abraham er at., 1989; Tabrirchi and Pang, 1986;Wang and Pang, 1990a Wang and Pang, 1991). The drugs were L.V.bolus injected 10 mm before the construction of the dose-responsecurve of DPI except for reserpine which was i.p. injected 24 h earlier.Protocol 3: Effects of TTX, pithing and spinal cord transection )T) on the MAP and HR responses of DPI. Dose.responsecurves of i.v. bolus injections of DPI (0.05—1.6 mg/kg) were constructedin four groups of rats, namely, ventilated control. ‘fl’X-pretreated (10sg/kg). pithed and spinal cord.transected T1i rats. T’TX was i.v. bolusinjected 10 mm before the construction of the dose.response curve ofDPI. A single dose of riorepinephrine 8 ,g’kg) was also i.v. bohainjected into the T’I’X.pretreated and pithed rats .5 nun after the lastdose of DPI.1 52R1NNH2R1—H2,R2—H, R3—H: L-Arg-HCH3R2-H, R3-H: L-NMMAR,-H2,R2-N0,R3-H: L-NNAR1-H2, -N0 -CH: L-NAMER1-H2,R2-NH,R3-H: L-NAANGsubstituted arginine analoguesFIg. 1. Chemicai structures of DPI and N°-substituted argunune analogs.L-NMMA, L-NNA. L-NAME, L-NIO and L.NsSA.at., 1991), those of DPI have not been studied for the past 20years or more. As an inhibitor of NO synthase, DPI is expectedto inhibit endothelium.dependent vasodilatation thereby causing pressor response and reflex bradycardia in uiuo. The aim ofthis study was to examine the effects of DPI on blood pressureand heart rate, to elucidate its mechanism of actions and, tocompare its actions with those of N°.substituted Arg analogs.Sprague.Dawley rats (300-350 g) were anesthetized with sodiumpentobarbital (65 mg/kg i.p.). A polyethylene cannula (PESO) wasinserted into the left iliac artery for the measurement of MAP by aP23DB pressure transducer (Gould Statham, CA). HR was determinedelectronically from the upstroke of the arterial pulse pressure using atachograph (Grass, model TP4G). Another PE5O cannula was insertedinto the left iliac vein for the administration of drugs. In some rats, aPE5O eannula was also inserted into the right iliac artery to collectblood samples. The body temperature of the rats was maintained at37’C with a heating lamp connected to a thermostat (73A, YellowSprings Instruments).In I’!’X-pretreated, pithed, spinal cord-transected rats and theircontrol rats, tracheostomy was also performed to allow artificial ventilation with 100% oxygen at 54 strokes/mm and a stroke volume of 3to 4 ml (1 mi/lOG g b.w,). Pithing was performed through the orbitwith a 3-mm diameter stainless steel rod. The spinal cord was severedby a pair 0f sharp scissors at the T1 level. All experiments wereconducted 20 miii after surgery.Plasma catecbolamines were measured by a catecholamine assay kitAmersham Canada Lcd,, Ontario, Canada). The principles of the assay1993Another two groups of pithed rats were i.v. bolus injected withcumulative doses of t.-NAME (0.2—12.8 mgJkg) or equal volumes ofnormal saline at dose-intervals of 15 mm. the time required to attainplateau responses of L-NA.ME.Protocol 4: Effects of DPI on plasma catecholamine.. Threegroups of intact rats were i.v. bolus injected with vehicle (5% glucosesolution 1 mI/kg), DPI (1.6 mg/kg) or L-NNA (16 mg/kg). One groupof pithed rats and one group of reserpunuzed (5 mg/kg up., 24 hpreviously) rtta (‘ 7) were also i.v. bolua injected with DPI (LB mgikg). Blood samples (L5 ml) were withdrawn over 15 a from the iliacarterial catheter into 1-mi syringes 20 miii before and 30 a after theinjection of vehicle or DPI or 40 mm after the injection of L-NNA. AUblood samples removed were replaced with LV. injection of an equalvolume of normal saline.Calculation and Statistical AnalysisThe parameters, i.e., E,._, E,..,,, ED or BC,0 and n were calculatedfrom individual dose-MAP and dose-HR curve of DPI using a programexecuted on an IBM-compatible microcomputer. To determine theseparameters, values of response (Y, rise in MAP or HR) at various D orC were Stted by nonlinear least-squares to the relation Y a + hi,where Y response and X [DJ/(EDw + [D1”) or X(EC,.’ ÷ [C]’) with ,H fixed at integral values (1,2,3,4 and 5), andrepeated with 11H ‘floating” to obtain a best fit (Qusstel andSaint. 1988). This gave the value of ED,0 or EC,, yielding a minimumresidual sum of squares of deviations from the theoretical curve. Thiswas preferred to the more usual fit to Y = hi, in order to take intoaccount the possible systematic underestimate or overestimate of MAPor HR corresponding to [D} or [C] 0; the data set was augmented by20 points with Y 0 at [DI or [C] = 0. Usually, the reduction inminimum residual sum of squares obtained by floating riM was notsignificant in the sense that the reduction (from that obtained with thenearest integral value of it,) was no mote than expected from thereduction in degrees of freedom (by Fisher’s test). With this fitting themaximum response to [DJ or [C] is given by b; values of a at the bestfit were never significantly different from 0.For linear least-squares regression to fit response Y a + hi wherex [C], the data set was also augmented by 20 points with response Y0 at (C] = 0.All results were expressed as mean S.E. and analyzed by theanalysis of variance followed by Duncans Multiple Range test byNumber Cruncher Statistical System Program (Dr. J. L. Hintze, Kaysyule, UT), with P < .05 selected as the criterion for statistical significance.ResultsEffects of DPI on MAP and HR. The j.v. bolus injectionsof vehicle in rats did not alter MAP or HR (data not shown).Bolus injections of DPI (0.05—1.6 mg/kg i.v.) caused immediateand transient pressor and tachycardic responses as shown in atypical experimental tracing from a rat (fig. 2). The durationof the pressor response lasted approximately 1 to 2 mm,whereas that of tachycardic response was slightly longer (fig.2). At 0.8 mg/kg, half-rise time for the pressor and tachycardicresponses were 2.9 ± 0.2 and 4.9 ± 0.7 s (P < .05), whereas thecorresponding half-fall times were 31 = 3 and 60 ± 9 a (P <.05), respectively.Pooled (ri = 12) base-line MAP and HR from the abovegroup and the control group in protocol 2 were 104 ± 3 mm Hgand 347 ± 10 beats/mm, respectively. The pressor and tachycardic response curves of DPI were dose-dependent and notably“steep,” with negligible effect at 0.05 to 0.1 mg/kg. large increases in MAP at 0.2 and 0.4 mgikg and maximum effect at0.8 mg/kg fig. 3). The best-fitted ri1 for MAP and HR were3.6 ± 0.3 and 4.2 ± 0.6, respectively. These two values were notI I I02 0.4 0.8I miiiFig. 2. Typic tracigs of blood pressure (SP) 310 HR respunes afWi.v. bolos siecXns (Sflown by arrows) of DPI (0.2,0.4 d 0.8 mgjlcg) iia pentobwbital-anesthetized rat.80600—200.2 0.4DPI (mg/kg)Fig. 3. Dose-response curves (meal, ± S.E.) of MAP and HR afteri.v. bokis njections of DPI in pentobarbtaJ-anesthetized rats (n 12).The theoretical lines were calculated using V =[Dr)) with n4 1, 2, 3. 4 cr5 as shown in the figure. The correspondingED and E,, for the curves are shown in table 1.significantly different from each other or from 4, but differentfrom 1. 2, 3 and .3, though all values of tiN, best.fitted (36. 4.2)or selected (1, 2, 3. 4 and 5), were statistically significant (P <.05i. The theoretical dose-response curves for integral valuesof ‘iii are shown in figure 3. and E calculated from dose-MAP and dose-HR curves at the best-fitted fi were not significantly different from those observedi other values of flH gavesignificant differences of calculated parameters from thoseobserved (E and E) or derived by averaging parametersobtained using best-fitting rz for each dose-response curve(table 1). Correlation between observed data points and theo—DPI on Cardiovascular System153255200iiooJ—2345‘aEEIc.8aCI40200—20100806040203450.05 0.1 0.8 1.6266 Wang and PangTABLE 1154V. 25Values (mean 2 S.L) of n,b correlation ceefficlent (,) expressed as 1000(1— r), E.. E. and ED. of ti dose-MAP and doee-bresponse cur.es after i.v. bolus injections of OPt (0.05-1.8 mg/kg)) in p. srboaI-an•sthethsd rats (n 12). Th. peram.ters wecalculated froni Individual dose-response curves using the fonnul. V a + bx where a (Dr/(ED.. + CO1”) (sea tsxt)I. HRO(1-r) 6,,, E..mMAP (e?wn Hg)Obsaiveddata 0±0 59±4Best-flttad 3.8 ± 0.4 3.4 ± 1.0 0.01 ± 0.04 58 ±3 0.22 ± 0.02flw Specjfied 1 • 26.3 ± 3.6 -0.37 ± 0.09 91 ± 8 0.52 ± 0.092 9.7 ± 2.1’ —0.21 ± 0.04 63±3 0.24:0.023 8.0 ±12 -0.03 ±0.04 58±3 0.22 ± 0.024 8.7 ± 12 0.18 ± 0.04 56±3 0.22 ±0.025’ 8.9±1.5 025 ±0.05* 56±3 0.22±0.02HR (beats/mm)Otsen.’eddata 0±0 63±4Best-lIfted 4.2 ± 0.6 4.8 ± 1.6 0.03 ± 0.05 62±4 0.23±0.02nSpecified 1 31.1 ±4.1 —0.47±0.06 124±22’ 0.74±0.10’2’ 13.1 ± 2.4’ —021 ± 0.08’ 68 ± 5 0.25 ± 0.033’ 82 ± 2.0 -0.04 ± 0.07 63±4 0.22 ± 0.024 7.8 ± 2.1 0.09 ± 0.13 60 ±4 0.22 ± 0.025 8.3 ± 4.4 0.14 ± 0.07 59 ±4 021 ± 0.02Slgniftcant 5ffarence from die best-SIted data (P < .05).retical curves, expressed as 1000 (1 — r), was the greatest for n= 3 and 4, and least for is = 1.Effects of antagonists on the MAP and HR responsesof DPI. Compared to vehicle, mecamylamine and reserpinereduced both MAP and HR. phentolamixie, propranolol andcaptopril reduced MAP but did not affect HR. whereas guanethidine did not alter MAP but increased HR On the otherhand, atropine, prazosin, rauwolscine and L-Arg did not altereither MAP or HR (table 2).The dose-MAP and dose-HR curves of DPI in the presenceof vehicle or antagonists and the corresponding ED50 and E,.r...,at the best-fitted H of DPI were shown in figures 4 to 6 andtable 3, respectively. Reserpine, guanethidine and mecamylaminc attenuated the MAP as well as HR responses of DPI (fig.4) with either a decrease in Em,, or an increase in ED50 of DPI(table 3). Phentolamine and prazosin but not rauwolscine reduced the MAP responses by decreasing Em,, of DPI. The HRresponses and the corresponding Em,,, on the other hand, wereincreased by phentolamine and rauwolscine but not prazosin(fig. ö and table 3). Propranolol abolished the HR responseswith a decrease in E,,,, but potentiated the MAP responses ofTABLE 2Base-line values (mean ± S.!.) of MAP and HR at 10 mm after Lv.bolus injections of vehicle, mecamytsmine (10 mg/kg), atropine (10mg/kg), phentelamin. (10 mg/kg), prOpraflOlol (1 mg/kg). prazosin(1 mg/kg), rauwolscine (1 mg/kg), guanethidine (10 mg/kg),captopnl (20 mg/kg) and L-erginine (400 mg/kg) or 24 hr after .p.injection of reserpine (5 mg/kg) in pentobarbital-anesthet(zed rats(n = S each group)nmguan .4AP HRnsn Hg o.njnwiVehicle 1026 353±10Reserpine 66±2’ 240±30’Mecamylarnine 71 ± 2’ 283 ± 20’Quanethidwie 100±5 418 ± 14Phentolamine 82 4 329 ± 22Prazosin 97±5 362±13Rauwolscine 89±5 354±11Propranolol 86±6 322:14Atropine 91±2 351±8Captopnl 52±6 319:8.-Arginine 96 ± 3 358 ± 8Sgndlceit difference from venide-pretreated prow (P < 05).0.05 0.1 0.2 0.4 0.8 1.5DPI (mg/kg)Fig. 4. Dose-response curves (mean ± S.E.) of i.v. bolus injections ofDPI on MAP and HR in pentobarbitai-anesthetized rats (n 6 eachgroup) pretreated with vehicle, reserpine (5 mg/kg), guanettiidine (10mg/kg) and mecamylamlne (10 mg/kg). All the pretreatment drugs werei.v. bolus n1ected 10 mm before the construction of the dose-responsecurve of DPI except for reserpine which was i.p. injected 24 hr previousty.The lines represent theoretical curves using the formula V(ED’ s. (0 and best-ñtted nb,, ED50 and!,,,, shown in table 3.DPI with an increase in E, (fig. 5 and table 3). Atropinepotentiated both the MAP (by decreasing ED50 and increasingin Em.,) and HR (by decreasing ED50) responses of DPI (fig. 6and table 3). Captopril markedly potentiated the MAP (byincreasing E.,) but not HR response, whereas L-Arg did noto VehIcle• Ras.rpin.Guonathidlne100805040200—201209062C2a0.041993EE0.05 0.1 0.2 0.4 0.8 1.5DPI (mg/kg)Fig. 5. Dose-response curves (mean ± S.E.) of i.v. bolus injections ofDPI on MAP and HR in pentobartEtal-anesthetized rats (n = 6 eachgroup) pretreated with vehicle. proprandol (1 mg/kg). phentotarnine (10mg/kg), prazosin (1 mg/kg) and rauwolscine (1 mg/kg). All the pretreatment drugs were .v. bokss infected 10 mwi before the construction of adose-response curve of DPI. The lines represent theoretical curves usngthe formula V E([Dr’/(EDO + [0)”) and best-fitted n, EDw andE.,,. shown in table 3.affect either the MAP or HR response of DPI (fig. 6 and table3).Effect of TTX-pretreatment, pithing and spinal cord(T1) transection on the MAP and HR responses of DPI.Base-line MAP in VI’X-pretreated rats (53 ± 2 mm Hg). pithedrats (45 ± 2 mm Hg) and spinal cord-transected (T1) rats (52± 2 mm Hg) was lower than that of ventilated control rats (105± 5 mm Hg). Base-line HR in T1’X.pretreated rats (318 ± 9beats/mm) and pithed rats (333 ± 10 beats/mm) was similarto, whereas that of spinal cord-transected rats (412 ± 16 beats/mm) was higher than that of ventilated control rats (312 ± 10beats/mm). DPI (0.05—1.6 mg/kg) also caused dose-dependentpressor and tachycardic responses in ventilated control rats;pretreatment with TTX and pithing totally abolished both theMAP and HR responses of DPI (fig. 7). However, norepinephrine (8 pg/kg) still caused increases in MAP and HR in TTXpretreated and pithed rats; their increases were 109 ± 1 and 44± 2 mm Hg, and 92 ± 6 and 57 ± 6 beats/mm. respectively. Onthe other hand, spinal cord transection (TL) markedly suppressed the dose-MAP and dose-HR responses of DPI compared to the responses in control rats: the suppression by spinalcord transection, however, was less than that by pithing (fig.7).The base-line MAP and HR in another two groups of pithedrats i.v. bolus injected with vehicle or L-NAME were 48 ± 3and 50 ± 4 mm Hg, and 308 ± 22 and 335 t 22 beats/mm.Bolus (i.v.) injections of cumulative doses of L-NAME but not0.2 0.4DPI (mg/kg)Fig. 6. Dose-response curves (mean ± S.E.) of LV. bolus injections ofDPI on MAP and HR in pentobarbltal-anesthetized rats (n 6 eigroup) pretreated with vehicle, atropine (10 mg/kg), captopd (20 mcjlcoJand L-argrue (400 mg/kg). All the pretreatment drugs were iv. Dotsnfected 10mm before the construction of a dose-response curve of DPI.The lines represent theoretical curves using the formula V(E0’ ÷ (Or’) and best-fitted n,, EO and L shown in table 3.vehicle dose-dependently increased MAP, but not HR in pithedrats (fig. 3).Effect of DPI on plasma catecholamines in intact.pithed and reserpinized rats. Base-line !evels of plasmacatecholamines were similar among the three intact rat groupsto be treated with vehicle, DPI and L-NNA (table 4). Comparedto the pooled data, pithing did not alter circulating catecholamines. Pretreatment with reserpine significantly decreasedplasma norepinephrine but increased plasma epinepb.rine.Base-line MAP and HR of these five groups of rats were similarto those of the corresponding groups in protocol 2 and 3 I datanot shown.Compared to the vehicle, DPI (1.6 mg/kg) caused largeincreases in plasma norepinephrine and epinephrine. and moderate increase in plasma dopanaine. as well as increases in MAPand HR (fig. 9). The increases in MAP and HR were significantly greater than those caused by the same dose of DPI inthe multiple dose regimen in protocol 2 (70 t 2 us. 53 ± 5 mmHg and 123 ± 7 us. 43 ± 5 beats/mm, respectively). In contrastto DPI, L-NNA (16 mg/kg) increased MAP, decreased HR andslightly decreased plasma dopamine, but did not alter plasmanorepinepbrine or epinephrine (fig. 9). Reserpine markedlyreduced DPI-induced increases in plasma norepinephrine, epinephrine and dopatnine by 91, 93 and 74%. respectively, andattenuated the pressor and tachycardic responses by 56 and68%. respectively (fig. 10). In reserpinized rats, the pressor andtachycardic responses to a single dose of DPI were also greatero VehIcle• PropronolelPt*ntlamIn.A PrazoelnRouwdecin.1 55oct on Cardloveactiar Sy1m 267100806040200-20120o Vehicle• Atropin.CeptoprUA L—orglnin.P9060300-30C0.05 0.1 0.8 1.6268 wang and PangV. 265TASLE 3Valuea (mean ± SI.) ø fl, ED,, and E_.. of the dos.-MAP and d0514€ reaponea ir,’s of OPt (0.05-tS mg/kg) m p a1iti-aneathatizad rats peatsd with vahidi, m.mØamin. (10 mg/kg), airoplie (10 mg/kg), ph.nla,nin. (10 mg/kg), .oIo1 (1 mg(kg), peazoaln (1 mg/kg), rauwolicin (1 mg/kg), rasirpin (5 mg/kg), gu.n.IWie (10 mg/kg), c..,Ao,.iI (20 mg/kg) and L-argiwi.. (400mg/kg) (n 5 each group)MAPI’” E_.sqjIqVelilcle 42 ± 0.7 024 ± 0.03 56±6 5.0 ± 0.8 025 ± 0.03 60±5Reserpine 21±3’ 22±8’Mecarnytwnww 2.0 ± 0.4’ 0.29 ± 0.10 26±3’ 3.4 ± 0.4 0.35 ± 0.05 28 ± 8’Quanee 3.1 ± 02 0.44 ± 0.10” 39±8 4.1 ± 0.4 0.35 ± 0.04 32 ± rPhentoiwie 3.5 ± 0.4 0.19 ± 0.03 33 ± 2’ 46 ± 0.4 0.14 t 0.04 96±12”Prazosin 3.9 ± 0.7 0.23 ± 0.04 32±5’ 3.4 ± 0.5 0.17 ± 0.01 56 ± 5Rauwoiscine 3.2 ± 0.4 0.16 ± 0.02” 68 ± 4 4.0 ± 1.1 0.16 ± 0.04 100 ± 13’Propraiclol 2.5 ± 0.3’ 0.31 ± 0.07 73 ± 5 19 ± 5’Atroplie 3.1 ± 0.6 0.17 ± 0.02” 76 ± 4 3.0 ± 0.5 0.14 ± 0.02’ 58 ± 8Captopril 22±0.1’ 0.22±0.02 90±5 2.6±0.3” 0.19±0.01 68±7L-argl*e 3.0±0.3 0.19±0.02 54±8 2.8±0.1’ 0.19±0.03 76±13‘-20‘I — IF •120 505090—20-40o VehIcle•L-MA)4E/-—8 “ Q 0 00.05 0.1 0.2 0.4 0.8 1.6 0.4 0.8 1.6 3.2 6.4 12.3DPI (mg/kg) L-NAlIE (mg/kg)156Sln&it Jn.w from Sw vaties Ii vicle-pw.ated (P < .06).Obuned100 o controi• Ptth.di trana.ctlzdA T.trodotoxln8040 /4-f10080E40EECEI 60300’-30C.50C.0t40200Fig. 7. Dose-response curves (mean t SE.) of i.v. bolos injections ofDPI on MAP and HR in pentobarbital-anesthenzed control rats. TTX (10vg/kg) pretreated rats, pithed rats and spinal cord-transected (T,) rats(n 6 each group). TTX was Lv. bolos infected 10 mm before theconstruction of a dose-response curve of DPI. The lines representtheoretical curves using the formula Y = L.,([Dr/(ED90’÷ [Dj andbest-fitted n, ED,, and E.,,, (values not shown).than those of the multiple dose regimen in protocol 2 (31 ± 10us. 18 ± 1 mm Hg and 39 ± 10 us. 22 ± 1 beats/mm, respectively). Pithing totally abolished the effects of DPI on plasmacatecholamines. as well as MAP and HR (fig. 10).Concentration-response and linear regression models wereused to examine the relationships between plasma norepinephrine or epinephrine and MAP or HR in intact, pithed andreserpinized rats i.v. bolus injected with DPI or vehicle (n =Fig. 8. Cumulative dose-response curves imean ± SE.) of i.v. bolosinjections of L-NAME or equal volume of veiNcle on MAP and HR flpithed rats (n 6 each group). The lines represent theoretical curvesusing the formula Y = E,[D/(ED,,”- {D) and best-fitted nH. ED,,and E.,,. Ivalues not shown).25). The linear regression model gave significant correlationbetween plasma norepinepbrine and MAP Cr = 0.83) or HR l?= 0.87). as well as between epinephrine and MAP I r = 0.8) orHR (r = 0.81). The concentration-response model, however.gave better fits. Correlation coefficient between plasma norepinephrine and .MAP or HR were 0.97 or 0.96. respectively.correlation coefficient between plasma epinephrine and MAPor HR were 0.94 or 0.94. respectively. Figures 11 and 12 showedthe concentration-response relationships between individual1993TA8LE 4bolus lnecId with vwiicIe DPI or ,.-NNAb .d,n - S azt for iias r ai fl -7.c—.—p4.__ EØsh.qjsrIntact (v.Ilicle) 0.374 ± 0.049 0.042 ± 0.010 0.082 ± ornoInt (DPI) 0.324 : 0.053 0.054 ± 0.0 0.102 ± 0.014Intact (L..NNA) 0.410 0.048 0.050 ± 0.011 0.097 ± 0.012Pooled 0.369 t 0.030 0.049 ± 0.009 0.094 ± 0.009PIthed (DPI) 0.414 ± 0.047 0048 ± 0.009 0.095 ± 0.009Resarpflzed (OPt) 0.206 : 0.018 0.081 ±0.011 • 0.078 ± 0.007‘ Sl&...4 JITsren from th. omeol (P < .06).-50DPI Vehicle L-NNAFig. 9. Effects (mean ± SE.) of iv. bolus ineCttCrs of a single dose ofvehicle, DPI (1.6 mg/kg) cc L-NNA (16 mg/kg) on plasma catecholamines,MAP and HR in pentobartital-anesthetized rats (n 6 each group).Blood samples arid MAP and HR measurements were obtained 20 mmbefore and 30 sec after inecticn of vehicle cc DPI and 40 mu’ afteringecton of L-NNA. SignIficant difference from vehicle group (P < .05).DiscussionTo our knowledge, this study is the first to show the in vivacardiovascular effects of DPI. The similar transient time courseand pharmacodynamics suggest a common cause for both pressor and tachycardic responses of DPI. Because captopril markedly potentiated the pressor response and did not alter thetachycardic response of DPI, it is safe to conclude that therenin-angiotensin system is not responsible for the responsesof DPI although the mechanism of the potentiation by captoprilis not known.By the use of sympatholytic drugs, we investigated whetherthe sympathetic nervous system was responsible for the pressorand tachycardic effects of DPI. Reserpine markedly attenuatedthe DPI-induced increases in MAP and HR. The results suggestthat DPI causes cardiovascular ffects by activating the peripheral sympathetic nerve terminals and adrenal rnedullae causingreleases of norepinephrine and epinephrine. This activation isdependent on the functional integrity of the central and autonomic nervous systems, as pithing abolishes, whereas spinalcord transection (Ti) attenuates the pressor and tachycardiceffects of DPI. The indirect activation of the sympatheticnervous system by DPI is further supported by the observationsthat TTX abolishes, whereas guanethidine and mecamylami.neattenuate, the effects of DPI. TTX has been shown to blockconductances of the central and peripheral nerves (Gage, 1971)but not those of vascular smooth muscle (see Hirst and Edwards. 1989) or the myocardium Abraham et aL, 1989) viaselective blockade of voltage-dependent sodium channels. Gusnethidine has been shown to be a specific adrenergic neuronblocker iShand et aL, 1973; Kirpekar and Furchgott, 1972).DPI caused immediate and large increases in plasma norepinephrine and epinephrine with the same time course as thepressor and tachycardic responses. Pithing totally abolishedand reserpinization attenuated DPI.induced increases inplasma catecholamines as well as MAP and HR. Further analy5ss.-11n levels (m.an ± SL) of plasma catacholamines Inoenberbltal-anisth.zed Intact. olth.d and rserpInizd rats Lv.1 57DPI on Cardiovascular System— EphIIn.aI4.4•ul‘-048001501120800SOil— EpInsg&In.—SaI4_Q,58050E2001501100‘so04IReserpinized ntact PithedFig. 10. Effects (mean ± SE.) of iv. bolus injection of a single dose atDPI (1.6 mg/kg) on plasma catecholamines, MAP and HR in pentobarbitai-anesthetized intact. pithed and reserpwiized rats. In each group n =6 except for reserpinized group. in which ri =7. Blood samples and MAP_______arid HR measurements were obtained 20 mm before and 30 sec afterDPI intection. - Significant difference from intact rats (P < .05).Iplasma norepinephrine or epinephrine and MAP or HR response caused by DPI.0.1 1 10Nor.pan.phsin. (nc/mi)Fig. 11. Concentration-response curves of changes in plasma norepi.riephnne and MAP as well as HR after .v. bolus injections of vehicle mintact rats and DPI (1.6 mg/kg) in intact, pithed and reserp4nZed rats. Ineach group n 6 except for reserpinized group, in which n = 7. Theline represents the theoretical curie calculated using the formula V =E4Cr’/(ECQ’” + [Cr’) and best-fitted nu, EC and L.., (values notShown).sis shows that positive correlations exist between DPI-inducedchanges in MAP and HR with plasma norepinephrine as wellas epinephrine. Taken together, the above results indicate thatDPI activates the sympathetic nerve terminals and adrenalmedullae to release norepinephrine and epinephrine in ratswith functional intact central and autonomic nervous systems.Because DPI releases large quantities of catecholamines, repetitive injections should lead to tachyphylaxis. Our results indeedshow that the MAP and HR responses of a single dose of DPIare greater than those elicited by the same dose in a multipleinjection regimen. We have also observed that multiple injections of high doses of DPI eventually produce negligible pressorand tachycardic responses data not shown).Norepinephrine and epinephrine released by DPI wouldbe expected to cause vasoconstriction, via the activation ofalpha-i adrenoceptors, and tachycardia, via the activation ofbeta. 1 adrenoceptors. We found that the pressor effect of DPIis suppressed by the nonselective alpha-adrenoceptor antagonist phentolamine and selective alpha-i adrenoceptor antagorust praxosin but not the selective alphi.z-2 adrenoceptor antagonist rauwoiscine. Moreover, rauwolscine and phentolamine.but not pra.zosin, enhanced the tachycardic effect of DPI; thispotentiation is likely due to the blockade of the central and/orperipheral prejunctional alpha-2 adrenoceptors which mediateinhibition of norepinephrine release Berthelsen and Pettinger,1977). This hypothesis may also explain why rauwoiscinecaused a small potentiatioru of the pressor effect of DPI. Our0.01 0.1 1 10 100Ep4n.phrln. (nc/mi)Fig. 12. Concentration-response curves of plasma epinephilne dMAP as wed as HR after i.v. bolus injections of vehicle in intact ratsand DPI (1.6 mgJkg) in intact, pithed and reserpirzed rats. In eachgroup n 6 except for reserpinized group, in which n = 7. The erepresents the theoretIcal curve calculated using the formula V+ [Cr’) and best-fitted n,,, EC and E.,.,, (values notshown).results also show that the tachycardic but not pressor effect ofDPI is abolished by the beta ad.renoceptor antagonist propranolol. The inability of propranolol to affect the MAP effect ofDPI suggests that the pressor effect of DPI is not due totachycardia or cardiac inotropy. The slight potentiation of thepressor response to DPI by propranolol may be due to theblockade of vasodilator beta-2 adrenoceptors which are prominent in skeletal muscle beds (Abdelrahman et at., 1990).The mechanism and primary site(s) of actions for DPI areunclear at the moment. It is logical to expect that DPI increasessympathetic discharge by inhibiting NO synthase. Extensiveevidences indicate that NO synthesis and release take place inthe brain (Garthwaite et at., 1989; Knowles et at., 1989, 1990).NO synthesis is reported to be responsible for long-term potentiation in the hippocampus Böhme et at., 1991), long-termsynaptic depression in the cerebellum (Shibuki and Okada1991) and nociceptive activity in the brain (Moore et at., 1991).Moreover. endothelium-derived relaxing factor/NO has beenshown to inhibit norepinephrine release from isolated sympathetic nerves innervating the canine pulmonary artery and vein(Greenberg et at., 1990) and other preparations (see “Discussion’ of Greenberg et at., 1990). However, our results do notsupport this hypothesis. In contrast to DPI. L-NNA at a dosethat caused maximal pressor response (Wang et at., 1991b), didnot increase plasma catecholamines. These results suggest thatDPI-induced sympathetic activation is unlikely due to theinhibition of NO synthesis. More studies are needed to elucidate158vol. 2650.AVec4, intactDPI. toctDPI. rplnlz.d270 wang and Pang100 0 Vehicle, intact• DPI. intactDPI, pithedA DPI, re..rpiniz.dleo20y4 A0C- 1201804200ieoAIIII41006020—20200ieo12080400-40.••40AA0-400.01A1993whether the inhibitions of other flavoproteins or NADPHdependent enzymes account for the actions of DPI.The central nervous system is unlikely a primary or majorsite for the actions of DPI although we cannot exclude thispossibility in view of the suppression of DPI’s effects by pithingand spinal cord transection. If the site of DPI is in the centralnervous system, local injection of the drug should then producegreater effects than t.v. administration. However, results fromour preIiminry studies show that intracarotid and intravertebral injections of DPI caused similar pressor responses as Lv.injections into the same rats (n — 3). As well, intraventricularinjection of DPI into the third cerebroventncle at doses up to0.1 mg/kg (ED, of 0.22 mg/kg by Lv. injection) did not causeany pressor or tachycardic responses (n 3). There is uncertainty about the accessibility of DPI to the central nervoussystem. Although 1251 was detected in the brain 10 mm afteri.v. injection of[‘251]DPI (Gatley and Martin, 1979), DPI, beinga charged molecule, may not adequately access the centralnervous system within 0.5 to 1 mm after injection. Moreover,it is difficult to explain why pithing is more effective thanspinal transection in attenuating the cardiovascular effects ofDPI if the central nervous system is the only site of action ofDPL On the other hand, we cannot nile out the possibleinvolvement of the peripheral sympathetic nervous system inthe actions of DPI, because even the action of indirectly actingsympathomimetic agents rely on a functional amine-uptakesYstem and therefore, sympathetic tone. Therefore, althoughthe integrity of the central nervous system is a prerequisite forthe actions of DPI. its primary sites of actions is not clear.Further studies are required to identify the site(s) of actions ofDPI in the central. efferent or even afferent nervous systems.It was reported that DPI appeared to cause respiratory di!ficulties leading to deaths of rats or mice (Gatley and Martin,1979) and, chronic administration of DPI caused fatigue of theskeletal muscle (Hayes et aL, 1985: Cooper et aL, 1988). Wealso observed that higher doses (0.8—1.-’ mg/kg) of DPI occasionally caused respiratory difficulty. Therefore, it is possiblethat the cardiovascular effects of DPI are secondary to respiratory dysfunction. However, this is unlikely the situation asDPI caused similar pressor and tachycardic responses with orwithout artificial ventilation.Detailed analyses of the dose-MAP and dose-HR responsecurves of DPI show that the H for the MAP and HR effectsof DPI are 3.6 ± 0.3 and 4.2 ± 0.6, respectively. These resultssuggest that the cardiovascular effects of DPI involve “positiveco-operation” of probably 4 molecules of DPI (see Rang, 1971and “Discussion” in Pennefather and Quastel, 1982). It shouldbe noted that even if H is not exactly 4, the 4th root of eitherMAP or HR responses should be linearly correlated to thedoses of DPI.Much has been published on the effects of N°-substitutedArg analogs on the in ultra and in uiuo endothelium.dependentvasodilatation or MAP responses (Aisaka et aL, 1989; Rees etaL. 1989, 1990; Wang and Pang, 1990a. b; Wang et aL, 1991a,b: Wang et al., 1992, 1993a.b; see Moncada et aL. 1991). DPIalso inhibits endothelium-dependent vasodilatation both inuitro CStuehr et aL, 1991; Poon et at., 1993) and in viva (Poonet aL. 1993) and causes transient pressor response. However,the mechanism of the pressor response is different from thatof N°-substituted Arg analogs. The pressor responses of N°substituted Arg analogs are susceptible to inhibition by L-Argbut not the impairments of the central nervous, autonomicDPI on Cardiovascular System159271nervous and renin-angiotensin systems (Wang and Pang.1990b, 1991; Wang et at., 1991b: Wang et at., 1992; Tabrizchiand Triggle. 1992; see Moncada et aL, 1991). These drup aijocaused reflex bradycardia (Wang and Pang, 1991). The pre.and tachycardic actions of DPI, on the other band, are entirelydependent on the functional integrity of the central and sympathetic nervous systems, but are not affected by L-Azg. Thelatter observation is consistent with the report that L-Arg donot antagonize DPI-induced inhibition of NO synthase acthity(StuehretaL, 1991).In conclusion, DPI but not L-NNA concomitantly caiimmediate and transient presaor and tachycardic respons. —well as immediate increases in plasma norepinephrine andepinepbxine. These effects are inhibited by maneuvers whichinterfere with the activities of the central or sympathetic nervous systems, namely, pithing, spinal cord transection and pretreatments with ‘fl’X, reserpine, mecamylamine and guanethidine. Moreover, the pressor but not tachycardic effect of DPIis attenuated by phentolamine and prazosin. The tachycazthcbut not pressor effect of DPI is inhibited by propranoloL Ourresults suggest that DPI, unlike the NG•substituted Arg analogs.produces preseor and tachycardic effects via the indirect activation of the sympathetic nervous system.Ackmowl.dgm.utsWe would ike to express our appreciations to Dr. David M. J. Quastel forhelp in analyzing the DPI do.e-re.ponse curves by the ime of his doe.-rei.curve analysis progrsm to Ore. Morley C. Sutter, Michael J.A. Walker and RTariZChi for their critical evaluations of our work and to Dr. Gerald S. MarbQueen’s University) for his advice. We also thank Mi. So Lin Lim for her skilledtechnical assistance.Reference.A5DELSANMAN. A., TABIuzcuI. B. AND PANG. C. C. Y.: Effects of 3,- and 4adrenoceptor stimulation on hemodyn.amics in the anesthetized rat. J. Casdiovsac. Phsrmscol. 15: 720-728, 1990.ABRAHAM. S.. BEATCH. G MACLEOO. 3. A. AND WAI.J(EP., M. J. A.; Antiarrhythmic properties of the tetrodotoxin against occusion induced arrbythmi25in the rats: A novel approach to the study of the antisxthythmic effects ofentncular sodium channel blockade. J. Phartnacoi. Exp. Ther. 251: 1161—:173. 1989.AISAKA. K.. GRoss. S. S., GRIYPrrH. 0. W. .rso Lzvi. R.: N°.methytarginine. aninhibitor of endothelium-derived nitric oxide ovntbesis. is a potent pressorsent in the guinea pi Does nitric oxide regulate blood pressure In vivoBiochem. Biophys. ReL Commun. 180: 881-486. 1989.BERTHELSEN. S. AND PETrINGER. W. A.; A functional basis for classificauon ofopha.adienergic receptors. Life Sci, 21: 2131—2138. 1977.BOHME. G.. BoN. C.. STIJTZMANN. J.M DOBLE. A. AND BLANCHARD. J.-C.Possible involvement of nitric oxide in long.tertr. potenlia:ion Eur. J. Phar:nacol. 199: 379—381, 1991.COOPER. J. M.. Prrrv. R. K. H., HAYES. D. i.. CHA.uss. R. A. J., BROSNe.N. M.J., SUOUBRIOGE, E. A., RADDA. G. K., MORGAN-HUGHES. J. A.,%ND Ct.ANx.J. B.; An animal model of mitochondria.l myopathy: A biochemical and physiological investigation of rats treated in vivo with the NADH-CoQ reductss.inhibitor. diphenyleneiodonium. J. Neurol. Sci. 83: 335—347. 1988.CROSS, A. R. AND JONES, 0. T. G.: The effect of the inhibitor dlpnenyleneodonium on the superoxida-generating system of rieutropliils. Biochen. 2.237: 111—116. 1988.Eu.is. J.A.. Csoss. A. R. AND JONES. T. G: Studies on the electron-transferechanisn of the human neutrophil NADPH oxidase. Biochem. 2. 282: 575-579. 1989.GAGE. P. W.: Tetrodotoxin and sasitoxin as pharmacological tools. In Neuropotsons: Their Psthophysiologicsl Actions. ed. by L. L Simpson. p. 187. PlenumPress, New York, 1971.Gs.a’rsswArrE. J., SOU’TNAM. E. AND ANDER-rON. M.: A ksinate receptor linkedto nitric oxide synthesis from L.argireJNeurochein 33: 1952-1954. 1989.GATLEY, S. 2. AND MARTIN, J. L: Some aspects of the pharmacolo’ of diphenvieionium, trivalent iodine compound. Xenooioucs 9: 539—546. :979.GsxzNBasG. S. S.. DIECKE. F. P. 2. AND TANAKA Release of norepmephrinefrom adrenergic nerve endings of blood vessels is nodulatad by etidotheliumderived reissing factor. Am. J. Hypertens. 3: 211-21& :990.HANCOCK. 3. 7. AND JONES. 7. G.; The inhibition by diphenyleneioruum and itsanalogues of superoxide feneration by macrophages. Biochem. 3. 242: 103-107. 1987.HAYES. D. J., BYRNE. E., SHOUBRIDGE. E MORGAN-HUGHES. J A, ANDCLARK. 3. 3.: Experimentally induced defects of mitochondnal nietaboiism in272 wang and Pangrat .k.I.tal nude: Bielocil ..cta of the NADHcoanzyme Q inhibitordipbenyaodonium. Biocbem. J. 228: 109-H7. 1986.HrasT. C. D. S. AND RDw*aDa. F. R.: Sympathetic nriro.ff.ctor traxueoua.ionis arteitee and .rnaiolee. Ptq.iol. Hew. 88:546-604, 1969.S. M. ass Fu.ci80orr, R. F.: lteraction of tyranulni and guaneth.idine In the lean of the cat. J. PharmacoL E. Tbet. 188: 38—46, 1972.KN0VLZS. H. C., Pu.acios, M., Pax.ssan. R. 55. J. aND MONCADA, L Formationof nitric oxide from L.arginine in the central aervos syste. Tranedection‘—“—- far r’—’14of the soluble guanylat. cycle... Proc. Nati. Aced.Sd. U.S.A. 18:5258-5162.1969.Kowtu, H. C., PAj.Acl0I, 55.. Pas.san, S. M. J. AND Moiscana, S. Kineticcberacteitstie. atitric oxide aj’nthe.e from rat brain. Bioc. J. 288: 207—210, 1990.MoNcana, S., PALISIR. H. 55. .1. ass llzcoa, H.: Nitric ozi& Physiology,p.shysioiogy end pharmacology. PharmacoL Rev. 49:108-142.1991.Mocax, P. K., Owyotsi, A. 0., Bao1, H. C., Wu.iC1, P. ass Hair, S.L L-N°-nitrO ergmine methyl er exhibits antinociceptive activity’ in themcii.. Br. .1. PharmacoL 102:198-202. 1901.PASSON. P. C. aND Pmm.en, J. D. A aimpi. raowetric omay fot p1oor.pinephzie and .pinepbrine. AnaL Biochrm. 51:618-631.1973.PImexiamza, P. AND QUASFZZ., D. 55. J. D.eensitition of the nicodnicreceptor at the mon.. nriromuecular junction. Br. J. PharmacoL 77:396-404,1982.PooH, C. I.. WANG, Y..W., Poore. K. S. ass Pass, C. C. Y.: Inhibitory effect.of diphenyI.nadtnnjuin in ritro andin viva. FASEB J., in pue, 1983.Quasm, 1). 55. J. ass Sthrr, D. A.: Tranamitter rile... at mcii. macor nan.terminal. mediated by temporary accumulation of intracellular barium. J.PbyaioL (Load.l 408:55-73,1968.Rarso, H. P.: Drug receptor, and their functio. Nature 231:91-96,1971.Rzzs, D. D., Paj.ssxi, H. M. J. ass MONCADA, S.: Role of .ndotheljuzn-derl,.dnitric oxide in the regulation of blood prameire. Proc. NatL Acid. Sd. U.S.A.86:3375-3378. 1989.Hans. D. D., PALMER, H. 55. J., Scnuz.z, S., HoDsoN. H. F. ass MONCADA, S.:Characterization of thre, inhibitors of endothelial nitric oxide syntbeeis in0100 and in uiuo. Br. .1. Pharmacol. 101:748-752. 1990.Stt.co, D. G., MORGAN. D. H. ass Oarzs, D. J. A.: The relea.. of guanethidineand b thanidine by eplenic nerve stimulstion A quantitative ev..luation show.ing diaaociation from adrenergic blockade. J. PharmacoL Exp. That. 184:73-80,1973.SN1BUKZ, K. ass O,cana. D.: Endogenoue nitric oxide release required for long.term synaptic depression in the cerebellum. Nature 349: 326-328. 1991.STZWART. G. A. ass HaIdIxc, 1’.: Other oral hypoglycemic aenta. In OralRypoglycenura Agenta Phsrmacolo and Theripeutic.. ad. by G. D. Campbell,pp. 398-407, Academic Press, New York. 1969.Srurna. D. J., FaSRNUM. 0. A., KwoN, N. S., GROSS. S. S., FONZALEZ, J. A..Lzvr, H. AND Nartwd, C. F.: inhibition of znacrophage and endothelial cellv. 265nitric oxide ,ynthaae by diphenylenelodoniunu sad its analogs. LASER J. &98-103. 1991.Syvzxx. D. J.. ?asgwuas. 0. A., Kwow. N. S.. Gaoss. S. S.. TtsixL. B. A., Lavi,B. AIID NATRAN, C. F.: Syntheeie of nitropa oxide. from t.-e7mi. bymicrophage cytceob Requirement for incbcibh and ccumztutaBiochem. Biophya. R. Commun. 181: 420-426, 1969.Sruana. D. .1., FA$EHITh, 0. A., KWON. N. S. ass NATRAN. C. F.: FAD andGSH perticipete in maczvpbag. iyntheeie of nitric Biochem, B..Re.. Cn,nnem. 168:558—566, 1990.TAXI, S. AND PANG, C. C. Y.: Comp..rath. aafrsawolecin.,,,.,,...end on blood and casec ii k.tid —C. .1. PbleoL Pheziacol. 68: 1421—1427, 1988.T41iarxcm, S. ass Tisocr.z, C. H.: Act at t.- and n.uglnine and Na’.on the blood pze_, at pithed normacae.p....es..., hyp.rteaeiv. r. Clin. Rep. fl,p.n....e. A14: 527-646,1WANG, Y..X. AND Pass. C. C. Y.: Effect. of Lend D oftnIneonblood in pen cbazbitsj.ana.itbethed ratS. Adv. UbScL8:123-728,1910..WANG, Y..X. AND PANG. C. C. Y.. Poetible dep.,.l....c’. otprmecr and hearteffects of N.nitro.j..atnin. an autoomic mave activity. Br. J. Pharmaml.103:2004-2008,1991.WANG. Y..X.. ND PaNG, C. C. Y.: Preeeor e.ct of N°.nitro-L-erpaine inpentoh.tbit,l.aneetheti_d rats. Life Sd. 47: 2217-4. iSOOb.WANG, Y..X., POOH, C. L ass PANG. C. C. Y.. Is uo and a rumeffects of I.- and D—enentiom.rs at N°.nitro.arpnin. on andotbelium-dipandsat reiaxo at rat i.ta. J. Pheisl. Hip. Ther. 968:112—119,1963..WANG, Y..X., PooH, C. L, Yau. 1.. C. ‘1. AND PANG. C. C. T.: Th. iso andin L’wO vaeoccictor effects of 1. and D en.r”—.s of N’.aitzo.ergibla.methyl er (NAME). The 36th Annual M.enng at C-”4’ F atseina ofBiologlcal Sociede., Victoria, Caned., p. 80, 1982.Wajtt, Y.-X., Zicou, T., CHUA, T. C. AND PANG, C. C. Y.: Effects of t,h.Linand intrsvenom anaeethetlc agent. on preeeor ronee to N°.nitto-I.-unine.Eu,. ,I. Pharmacol. 198:183-188, ISels.WANG, Y.-X., Zaou, T. AND PANG, C. C. Y.; Preeeor effects of L and Denantiomer. of N°nitro.argiziine in conscious rats are antagonized by L hutnot Darglnme. Eur. J. Pharmacol. 200: 77-31, 1991b.WaNG. Y..X.. Zrsou, T. aND PANG. C. C. Y.: A compariaon of the effect. ofsodium nitropruraide, pinacidil and oifedipine on the pressor response o N°.zutro-c.-arginine. Br. J. PharmacoL 108:396.404. 1993b.Wmoop, H. S., GARDINES, S. M., KEMp, P. A. ass Bv*orrr, T.: The influenceof atropine and atenolol on the cardiac haemodynamic ifecta of N°-nito-t.arginine methyl ester in conscious rats, Long Evans rats. Br. J. PharmaceL105: 653-666. 1992.Send reprint r.qu..ts to: Dr. Catherine C. Y. Pang Depertment of Pharma.cology and Therapeutics. Faculty of Medicine, The University of British Cobmu.bia, 2176 Health Sciences Mall, Vancouvir, B.C., Canada V61’ 1Z3.160161Paper VWang, Y.-X. and Pang, C.C.Y.: Halothane inhibits pressor effect ofdiphenyleneiodonium. Br. J. Pharmacol. 109: 11 86-1 1 91, 1 993. Thereproduction of this paper was kindly permitted by the copyright holder, MacmillanPress Ltd., Hampshire, U.K.Br. J. Phoj’macol. 13993). 109. 1166-1191162© MacmilLin I’rcs’. lid. ‘J’.Halothane inhibits the pressor effect of diphenyleneiodoniumYong-Xiang Wang & ‘Catherine C.Y. PangDepartment of Pharmacology & Therapeutics. Faculty of Medicine, The University of British Columbia, 2176 Health SciencesMall, Vancouver, B.C., V6T 1Z3, CanadaI We have recently found that diphenyleneiodonium (DPI), a novel inhibitor of nitric oxide (NO)synthase, causes pressor and tachycardic responses in pentobarbitone- but not halothane-anacsthetizedrats. The present study investigated the mechanism by which halothane suppresses the pressor responseof DPI. The effects of halothane on the pressor response of DPI were also compared with those of otheranaesthetic agents.2 In conscious rats, i.v. bolus injections of DPI (0.025— 1.6mg kg) caused dose-dependent increasesin mean arterial pressure (MAP), with ED of 0.07 ± 0.01 mg kg and maximal rise of MAP (E) of59±2 mmHg. While ketasnine potentiated E_ without altering the ED and pentobarbitone increasedthe ED, without changing E of the pressor response to DPI, chloralose, urethane and ethanoldisplaced the curve to the right and potentiated E. In contrast, halothane (0.5-1.25%) dose-dependently and non-competitively reduced the preasor responses to DPI.3 Intravenous bolus injection of a single dose of DPI (1.6mg kg) caused immediate and largeincreases in plasma noradrenaline and adrenaline, as well as MAP in conscious rats. Halothane (1.25%)almost completely inhibited these increases.4 The results suggest that DPI causes a pressor response in conscious rats by activating the sympathetic nervous system and halothane abolishes this pressor response by inhibiting activities of thesympathetic nervous system. The results also show that influences of anaesthetics must be taken intoconsideration when evaluating pressor response of vasoactsve agents.Keywords: Diphenyleneiodonium (DPI); nitric oxide synthase inhibitor anaesthetics; halothane; pentobarbitonc; ketamine:urethane; ethanol; chloralose; pressor sympathetic nervous system; noradrenaline; adrenaline; endotheliurndependent vasodilatationIntroduction MethodsDiphenyleneiodonium (DPI), a bivalent iodine compound,was first reported to be a potent hypoglycaemic agent(Stewart & Hanly, 1969; Gatley & Martin, 1979). It was latershown to suppress the activities of neutrophil and macrophage NADPH-dependent oxidase (Cross & Jones, 1986), aswell as macrophage nitric oxide (NO) synthase (Stuehr et a!.,1991), probably via the inhibition of a fiavoprotein (Hancock& Jones, 1987; Ellis et at., 1989; Stuehr et a!., 1990). DPI wasalso found to inhibit endothelium-dependent vasodilatationin vitro (Stuehr e at., 1991; Poon cx a!., 1993) and in vivo(Poon er aL, 1993). Recently, we found that DPI causedpressor and tachycardic effects in pentobarbitone-anaesthetized rats; the cardiovascular effects were due to themodulation of sympathetic nerve activities (Wang & Pang,1993). In our preliminary experiments, we also found thathalothane markedly suppressed the pressor effect of DPI.Halothane has been shown to inhibit sympathetic nervoustransmission at several levels (Seagard er oh, 1982; Larach cxoh, 1987; Rorie ci a!., 1990). Halothane was also found toalter endothelium-dependent vasodilatation (Muldoon et a!.,1988; Blaise, 1991). and to abolish the pressor responses ofother NO synthase inhibitors, namely N°-nitro-L-arginine(L-NNA) and its methyl ester (L-NAME) (Wang cx oh,199la Pang cx a!., 1992). The aim of this study was to (I)investigate the characteristics of the inhibitory effect ofhalothane on the pressor response to DPI (2) to examinewhether halothane suppresses the pressor response of DPI byinhibiting activities of the sympathetic nervous system; (3) tocompare the effect of halothane with those of intravenousand inhalation anaesthetic agents on the pressor response ofDPI.Surgical preparationSprague-Dawley rats (300—360 g) were used in this study.Cannulae (PESO) were inserted into the left iliac vein for theadministration of drugs, and into the left iliac artery for therecordings of mean arterial pressure (MAP) by a P23DBpressure transducer (Gould Statham, CA, U.S.A.) and heartrate which was determined electronically from the upstrokeof the arterial pulse pressure using a tachograph (Grass.Model 7P4G). In some rats, another PESO cannula was alsoinserted into the right iliac artery tocollect blood samples.The body temperature of the anaesthetized rats was maintained at 37’C with a heating lamp connected to a thermostat(73A, Yellow Springs Instruments). In halothane-anaesthetized rats in Protocol 2 and 3 (see later). tracheostomvwas also performed with a PE164) catheter. All anacsthetizedrats were equilibrated for 20 mm at the appropriate anaesthetic doses before the commencement of studies. The conscious rats in all the protocols were first anaeschetized withhalothane (1.5% in air), to allow surgical preparation. andwere allowed to recover for at least 6 h from the effects o:’anaesthesia before further use. The catheters were tunnelledsubcutaneously and exteriorized at the back of the neck.Meawremern ofplasma catecholaininesPlasma catecholamines levels were determined by a casecholamine Assay kit (Amersham Canada Ltd.. Ont.. Canadal(Passon & Peuler. 1974). The radioactivity (3H) was detectedby a 1600 TR liquid scintillation analyzer (Packard Instrument Co.. CT, U.S.A.). Blood samples (0.5 ml) were immediately placed in prechilled tubes containing EGTA andreduced glutathione and centrifuged at 1.200 g at 4C. Afterwards, the plasma was removed and stored at — 70’C untilassayed within two weeks. Duplicate assays were run for theAuthor for correspondence.163HALOThANE AND OPt ON VASCULATURE 1187standard (in blank plasma) and plasma samples (50 I each)using distilled water as a blank control for each run. Thesensitivity of the catecholainine assay was 0.005 ng m1’ foreach catecholamine.DrugsHalothane and diphenyleneiodonium (DPI) sulphate wereobtained from Ayerst Lab. (Quebec, Canada) and ColourYour Enzyme LtcL (Oat., Canada), respectively. Urethaneethyl carbamate and ketaimne hydrochloride were fromSigma Chemical Co. (MO, USA.). The following anacsthetics were also used: -chloralose (BDH Chemical Ltd.,Poole, England). sodium pentobarbitone (M.T.C. Pharmaceuticals, Cambridge, Ontario, Canada) and anhydrous ethylalcohol (Stanchem Co., Quebec, Canada). The powder drugswere dissolved in normal saline (0.9% NaCI) except for DPIwhich was solubilized in 5% glucose solution by 10 mmsonication.Experimental protocolThe rats (n =6 each group) were randomly assigned intogroups. MAP and HR were continuously monitored throughout the experiments.(1) Dose-MAP response curves to DPI in conscious andanaesrheri:ed rats Six groups of rats were usedi consciousrats and rats anaesthetized with sodium pentobarbitone(6smgkg’), ketamine (l4Omgkg’), urethane (2gkg’).ethanol (4 g kg-’) and chioralose (150mg kg’). Dose-MAPresponse curves to i.v. bolus injections of DPI (0.025—1.6mgkg-’) were constructed at dose-intervals of 3—6 mm, the timerequired to recover from the effects of the previous dose.(2) Dose-MAP response curves of DPI in conscious andhalothane-anoesiherized rats Five groups of rats were used:conscious rat and rats anaesthetized with different concentrations of halothane (0.5. 0.75, 1 and 1.25% in air) at flowrates of 1500 ml min’. The dose-MAP response curves ofi.v. bolus injections of DPI (0.025—6.4mg kg’) were constructed at dose-intervals of 3—6 mm as above.(3) Effects of DPI on plasma carecholwnine levels in con-scious and halothane-anaesthetized rats One group of conscious rats and one group of halothane (l.25%)-anaesthetizedrats were injected i.v. with a single bolus dose of DPI(1.6mg kg’). Blood samples were collected 20 mm prior toand 30 s after the injections of DPI. MAP was continuouslymonitored.Calculation and statistical analysisMaximum effect (E, half-effective dose (ED) and Hillcoefficient (n), from individual dose-response curves werefitted by a computer programme using non-linear least-squares to the relation Y = a + bx, where Y = response andx = [Dj/(ED + [D]*) (Quastel & Saint, 1988; Wang &Pang. 1993). All results were expressed as mean ± standarderror (s.e.mean) and analyzed by the analysis of variancefollowed by Duncan’s multiple range test with P<0.05selected as the criterion for statistical significance.ResultsEffects of anaesthetics on DPI-induced MAP responseBaseline MAP in rats anaesthetized with urethane, ethanoland chloralose, but not pentobarbitone or ketamine, was lessthan that in conscious rats (Fable 1).In conscious rats, DPI caused an immediate (approximatelyl5s in onset) and transient (1—2mm in duration) pressorresponse. The pressure response was dose-dependent (Figure 1)with ED of 0.07±0.01mg kg’ and maximal MAP reachedat 59±2 mmHg, based on the best-fitted calculations (FableI). The Hill coefficient n of 3.3 ± 0.5 was significantly differentfrom 1, 2 and 5 but not from 3 or 4. DPI also causedtachycardia at lower doses (0.025—0.1 mg kg’), bradycardiaat higher doses (0.2—1.6mg kg-’) and movements followingthe onset of the pressor response (data not shown).Pentobarbitone, chioralose, urethane and ethanol but notketamine displaced the dose-MAP curve of DPI to the right(Figure 1) by increasing EDns (Fable 1). On the other hand,ketamine, chloralose, urethane and ethanol but not pentobarbitone potentiated the maximal MAP response to DPI (Figure1 and Table 2). None of the anaesthetic agents significantlyaltered the n value of the dose-response curves of DPI (Table1). Under the influence of all anaesthetics, the hindlimbs andoccasionally the forelimbs displayed kicking motionimmediately following injections of DPI.Inhibitory effect of halothane on DPI pres.sor responseHalothane (0.5—1.25%) reduced baseline MAP in a dose-dependent manner (Table I). In conscious rats. i.v. bolusinjections of DPI also caused similar pressor responses as inprotocol (I) (Figure 2 and Table 1). Halothane dose-dependently reduced the maximal effect of DPI and shifted theDPI curve to the right (Figure 2). ED values were linearlycorrelated while H,,,, values were inversely correlated with theTable I Values of baseline mean arterial pressure (MAP), as well as parameters (Hill coefficient n. EDw and H,,,,) of the dose-pressorresponse curves of diphenyleneiodonium (DPI) in conscious rats and rats anaesthetized with sodium pentobarbitone (65mgk5’).kctamine (l4omgkg-’). urethane (2gkg’), ethanol (4gkg’), chloralose (150mg kg-’) and halothanc (0.5—1.25%)Baseline MAPmmHgDose-response curve to DPIED.1, (mg kg’) E,, (mmHg)AnuestheticsProtocol IConscious 118 ± 2 3.3 ± 0.5 0.07 ± 0.01 59 ± 2Ketamine 107 ± 7 3.2 ± 0.4 0.08 ± 0.01 72 ± 2Pentobarbitone Ill ± 6 3.3 ± 0.7 0.22 ± Q.04 64 ± 4Chioralose 89±3 3.7±0.4 0.12±0.01 90±4Urethane 72 ± 7 2.9 ± 0.6 0.20 ± 0.03’ 71 ± 4’Ethanol 57 ± 6’ 2i ± 0.4 0.26 ± 0.03’ 78 ± 4’Protocol 2Conscious 113±4 2.9±0.5 0.07±0.01 59±20.5% Halothane 96 ± 10 2.6 ± 0.2 0.4 ± 0.09’ 62 ± 40.75% Halothane 83 ± 4’ 2.7 ± 0.5 0.57 ± 0.08’ 47 ± 4’1% Halothane 80 ± 7’ 3.8 ± 0.7 0.77 ± 0.15’ 34 ± 3’125% Halothane 69±2’—— 13±4”Values are mean ± s.e.mean n —6 for each group‘denotes significant diftcrence from the conscious rat groups (P<0.05). ‘represents observed data.1188 Y.-X. WANG & C.C.Y. PANG164DPI (mg kg)Flgwe 1 Dose-response (mean ± s.e.mean) curves of LV. bohis injections of diphenylemeiodonium (DPI) on mean arterial pressure(MAP) in conscious rats (0), and rats anaerthetired with pesitobarintone (6smgkg’) (•), ketsmine (140mgkg) (A), urethane(2gkg’)(A), ethanol (4gkg) (D)and chloralose (150 mgkg’)(a). it -6 each group. The lines represent theoretical curves usingthe formula Y - E, ([D]/ED, + [Dr) and best-fitted n, ED andE shown in Table 1.Figure 2 Dose-response (mean ± s.e.mean) curves of i.v. bolus injections of diphcnyleneiodonium (DPI) on mean arSenal pressure (MAP)in conscious rats and halothane (0.5-l.25%)-anaesthctized rats (n . 6each group). Ralothaner 0 (0); 0.5% (•); 0.75% (A); 1% (A);1.25% (a). The lines represent theoretical curves using the formulaY = E (fDJ/ED,i’ + (Dj9 and best-fitted n, ED and E shownin Table 1.b1.0Y= —0.68x + 0.09r= 0.993(P< 0.05)0 0.25 0.5 0.75 1 1.25Halothane 1%)Figure 3 Effects of halothane (0.5-1.25%) on the maximal effect(Em,, a) and half-effective dose (ED,0, b) of the dose-mean arterialpressure response curves of Lv. boles injections of diphenylenelodonium (0.025—6.4mg kg) in rats (n 6 each group).Table 2 Baseline values of plasma catecholanunes and meanarterial pressure (MAP) in conscious and halothane(1 .25%)-anaesthetized ratsPlasma catacholvnrne (ug mr’)Noradrenaline Adrenaline Dopomine MAP0.26 ± 0.05 0.12 ± 0.01 0.10 ± 0.01 119 ± 40.25±0.03 0.06±0.01 0.10±0.01 76±rValues are mean ± s.e.mcan. a 6 each group.*dcnotca significant difference from conscious rats (P<0.05).concentration of halothane (Figure 3). None of the doses ofhalothane affected the n value of the curves (Table 1).Halothane also inhibited DPI-induced tachycardia, bradycardia as well as movements (data not shown).Effect ofhalothane on DPI-induced co.zecholwnine releaseBaseline MAP in halothane (1.25%)-anaesthetized rats waslower than that in conscious rats (Table 2). Baseline plasmalevel of adrenaline but not noradrenaline or dopaniine inhalothane-anaesthetjzd rats was also significantly lower thanthat in conscious rats (Table 2).In conscious rats, iv. bolus injection of DPI (1.6mg kgt)caused immediate and large increases (more than 1 ng m1 ) inplasma noradressaline and adrenaline and a smaller increase(0.1 ng ml_L) in plasma dopamine (Figure 4a), as well asimmediate pressor response (Figure 4b). Halothane markedlyattenuated DPI-induced increases in plasma catecholamines(Figure 4a) and in MAP (Figure 4b); the reductions of plasmanoradrenaline, adrenaline and MAP were 86%, 81% and95%, respectively.DiscussionAs in previous experiments with pentobarbitone-anaesthetizedrats (Wang & Pang, 1993), DPI caused transient and dose-dependent increases in MAP in conscious rats with a Hillcoefficient not significantly different from 3 or 4. Moreover.none of the anaesthetica affected the Hill coefficient value. Theresults suggest that the pressor effects of DPI in conscious andanaesthetized rats are due to the positive cooperation of 3 or 4molecules of DPI with the corresponding ‘receptors’ (see Rang.1971; see discussion of Pennefather & Quastel. 1982: Wang &Pang, 1993).Although DPI inhibits endothelium-dependent vasodilatation in vitro and in vivo (Stuehr eg aL, 1991; Poon ci al.. 1993).the pressor response to DPI is not a consequence of thisinhibition. In a previous study, we found that DPI causedpressor and tachycardic responses by sympathetic stimulationas reflected by concurrent elevations of blood pressure andplasma noradrena]ine and adrenaline (Wang & Pang, 1993).Moreover, the pressor response of DPI was attenuated bymanoeuvres which impair sympathetic nerve tranemkcion.a8040Y=-36.8x+68.920 r=Q.893(P4 h to recover from the effects ofhalothane before further use. The body temperature ofhalothane-anaesthetized rats was kept at 37°C with aheating lamp connected to a 73A thermostat (YellowSprings Instruments), and the rats were used 30 ruin afteroperation.Microsphere techniqueThe reference sample microsphere technique was described previously in detail (14): 30.000—40.000 micro-spheres (15-ti.m diameter) labeled with either “Co ortt3Sn (Dii Pont Canada, Ontario, Canada) were injectedin the left ventricle in the control period and after administration of a drug or vehicle. The order of administration40,2000.4 0.5 1.6 3.2 6.4 12.5L—NMC (mJhq)FIG. 1. Dose—mean arterial pressure (MAP) response (means± SE) curves of cumulative intravenous bolus injections ofN°-nitro.L-arginine (L-NNA (A) and N°-nitro-L-arginine methyl ester (i..-NAME) (B) or equivalent volume of vehicle (0.9%NaCI) in conscious and halothane-anaesthetized rats (n = 5per group). Values represent changes from pretreatment values.of the microspheres was reversed in half of each group ofrats. At the end of the experiments, blood samples, wholeorgans of lungs, heart, liver, stomach, intestine, caecum,and colon (presented as colon in Table 1 and Figs. 1—6),kidneys, spleen, testes and brain, as well as 30 g eachskeletal muscle and skin, were removed for counting ofradioactivity with an 1185 Series Dual Channel AutomaticGamma Counter (Nuclear, Chicago, IL. U.S.A.).TABLE 1. Baseline values (means ± SE) of MAP, HR. çO. and TPR in conscious (groups 1—IV. IX—X) andhalothane-aaaesthetized (groups V—VIII, XI—XII) ratsMAP HR CO TPRGroup n (mm Hg) (beatslmin) (mllmin) (mm Hg/mm/mi)Protocol I1 5 110±5II 5 108±7UI 5 114±4IV 5 121±4Pooled 20 113±3V S 79±1VI 5 78±2VII 5 85±2VIII 5 79±1Pooled 20 80±1°Protocol 2DC 6 123±3 368±11 102±3 1.20±0.0X 6 132±3 392±16 110±8 1.20±0.0Pooled 12 128 ± 2 380 ± 10 106 ± 4 1.20 ± 0.0XI 6 73±2 319±8 84±5 0.89±0.0XII 6 79±2 298±7 82±4 0.98±0.0Pooled 12 76 .± 2° 309 ± 6° 309 ± 6° 0.94 ± 0.0MAP, mean arterial pressure; HR. heart raze; CO. cardiac output; TPR., total peripheral resistance.‘Significant difference from pooled data in conscious rats (p < 0.05).of L-NNA by halothane is caused by reduction incardiac output (CO) and/or regional vascular resistance.o Vd,d.• L—Nk& L—NME4 VdldeA —64A o,• L—NAt1 2 6 16 32L—b64A (mg/kg) -EaBJCoxdiovascPkarmacot’. VoL 22, No.4. 1993HALOTHANE INHIBITION OF L-NAME EFFECTS171573800.25 0.5 I 2 4 8Nor.pin.phrini(1glkg)S0.02 6.04 0.08 0.16 0.2 0.54.N (g)FIG. 2. Dose—mean arterial pressure (MAP) response (means± SE) curves of intravenous bolus injections of norepinephrine (A) and angiotensln 11(B) in conscious and halothaneanaesthetized rats (n = 5 per group). Values representchanges from pretreatment values.DrugsL-NNA, L-NAME hydrochloride, All acetate, and NEhydrochloride were obtained from Sigma Chemical. (St.Louis, MO, U.S.A.) and dissolved in normal saline (0.9%NaCI). Dissolution of L-NNA required 20-mm sonication.V.ide —2.5 r• 2.01.5£ 1.00.5On-sFIG. 3. Effects (means ± SE) of intravenous bolus injectionsof vehicle (0.9% NaCI) and N°-nitro-i.-arginine (i.-NNA 16 mg/kg) on mean arterial pressure (MAP), heart rate (HR). cardiacoutput (GO). and total peripheral resistance (TPR) in fourgroups of conscious and halothane-anaesthetized rats (n6 per group). Values represent changes from pretreatmentvalues. Significant difference from vehicle group (p < 0.05).bsignificant difference from conscious rat group (p < 0.05).4,eFIG. 4. Effects (means ± SE1 of intravenous bolus injectionsof vehicle (0.9% NaCI) and N-nitro-L-arginine (L-NNA, 16 mg/kg) on blood flow and vascular conductance in consciousrats (n = 6 per group). Values represent changes from pretreatment values. Significant difference from vehicle group(p <0.05).Experimental protocolIn the first study, eight groups of rats (n = 5 eachgroup) were used to construct the dose—MAP responsecurves of L-NNA, 1.-NAME, NE, All, and vehicle inconscious rats (groups I-.IV) and halothane-anaesthetizedrats (groups V—VIIl). After 30-mm equilibration to obtainstable baseline readings, vehicle was injected as an i.v.bolus in groups I and V. L-NNA (1—32 mg/kg) was cumulatively injected as an i.v. bolus in groups II and VI, and1.-NAME (0.4—12.8 mg/kg) was administered to groups IIIconsciousHOlothoneB u-4FIG. 5. Percentage of change In vascular conductance(means ± SE) after Intravenous bolus injection of N°-nitroL-arginine (L-NNA 16 mg/kg) in conscious and halothaneanaesthetized rats (n 6 per group). Values were calculatedby subtracting vehicle-induced conductance changes fromcorresponding L-NNA-induced conductance changes, dividing this difference by absolute vascular conductance valuebefore L-NNA injection (predrug control), and then multiplying the resultant number by 100%. Signlficant differencefrom conscious rats (p < 0.05).o c04).ciouw• *alothan.A80604020E-. 10080II.1004-3001.58-0.5.05 -0.5• .5.1.-2.5-2.5—CC Ezzllr1 II IHalothane was obtained from Ayerst Laboratory (Montreal, Canada).5040E2O10L—NNAHccneI.66 —20-604-6080.4oC£• 0Uc0—400S0-80— 20••-ISOJ Ca,diovasr Phatmacol”. Vol. 22. No. 4. 1993574 Y.-X. WANG ET AL.172SESSaFIG. 6. Effects (means ± SE) of intravenous bolus inlectionsof vehicle (0.9% NaCI) and NG.nitro_Larginine (L-NNA 16 mgIkg) on blood flow and vascular conductance in halothaneanaesthetized rats (n 6 per group). Values representchanges from pretreatment values. Significant differencefrom vehicle group (p < 0.05).and VII. Ten-minute dose intervals were allowed between injections for MAP to reach plateau values. Ingroups IV and VIII, i.v. bolus injections of doses of NE(0.25—8 p.gfkg) and All (0.02—0.64 i.gIkg) were given atdose intervals of 5 mm to allow recovery from the effectsof the previous dose. The sequence of the injections ofNE and All was reversed in half of the experiments ineach group.In the second study, four groups of rats (n = 6 eachgroup) were used to investigate the haemodynamic effects of L-NNA in conscious (groups IX and X) and halothane-anacsthetized rats (groups XI and XII). One minuteafter the first set of microspheres was injected, vehiclewas injected as an i.v. bolus in groups X and XII andL-NNA (16 mg/kg) was injected as an i.v. bolus in groupsIX and XI. After 20 mm more, the second set of micro-spheres was injected.Calculations and statistical analysisTotal peripheral resistance (TPR), CO. blood flow, andvascular conductance (blood flow/MAP) were calculatedas described (14). BlOOd flow and vascular conductanceare given as values per 100 g tissue. Changes in vascularconductance by L-NNA in conscious and halothaneanaesthetized rats are also determined as percentage ofconductance change to compare the magnitudes of vasoconstriction in the two states; this was obtained by subtracting the vehicle-induced organ/tissue conductancechange in the time-control group from the correspondingchange, by L-NNA, dividing this difference by the absolute conductance value before L-NNA injection (predrugcontrol), and multipLying the resultant value by 100%. Allresults are mean ± SE and were determined by analysisof variance followed by Duncan’s multiple-range test tocompare group means; p <0.05 was preselected as thecriterion for statistical significance.RESULTSEffects of halothane on MAP responses to L-NNA,L-NAME, NE, and AllThere were no significant differences in baselinevalues of MAP among conscious rats in groups I—IVand among halothane-anaesthetized rats in groupsV—VIII (Table I). Comparison of pooled valuesshows that MAP of halothane-anaesthetized ratswas lower than that of conscious rats.-Vehicle did not alter MAP in conscious rats (Fig.IA and B). Bolus injections of i.v. cumulative dosesof L-NNA (Fig. IA) and L-NAME (Fig. IB) in conscious rats caused dose-dependent increases inMAP, with maximum increases of —50 mm Hg. Inhalothane-anaesthetized rats, vehicle, L-NNA, andL-NAIVIE each caused similar slight decreases inMAP with passage of time (Fig. IA and B).Bolus i.v. injections of NE (Fig. ZA) and All (Fig.2B) caused dose-dependent increases in MAP inconscious rats with maximum increases of —50—60mm Hg. Halothane “noncompetitively” inhibitedthe pressor responses of NE and All, but the degrees of inhibition caused by halothane at the highest doses of NE and All were 17.6 and 18.0%, respectively.Haemodynamic effects of L-NNA in conscious andhalothane-anaesthetized ratsTable I shows that baseline values of MAP, HR,CO. and TPR were similar in the two groups ofconscious rats (IX and X) and the two groups ofhalothane-anaesthetized rats (Xl and XII). Pooledvalues of MAP. HR, CO, and TPR in halothaneanaesthetized rats were significantly less than thoseof conscious rats.In conscious rats, i.v. bolus injection of L-NNAsignificantly increased MAP and TPR and decreased HR and CO as compared with changes inthe vehicle-treated group (Fig. 3). In halothaneanaesthetized rats, i.v. bolus injection of L-NNAcaused slight but significant increases in MAP andTPR, significant increase in HR, and insignificantdecrease in CO (Fig. 3). However, as comparedwith the changes in conscious rats, the effects ofL-NNA on MAP, TPR, HR, and CO were significantly Less in halothane-anaesthetized rats (Fig. 3).Table 2 shows baseline values of regional bloodflow and vascular conductance in conscious andhalothane-anaesthetized rats. There were no significant differences in baseline values of blood flowand conductance in any organs/tissues between thetwo groups of conscious rats (LX and X) and between the two groups of halothane-anaesthetizedrats (XI and )Ul). Pooled data showed that halothane increased blood flow to lungs, liver, testesand brain, decreased blood flow to stomach, intestine, kidney, spleen, muscle, and skin, but did notalter blood flow to heart, caecum, and colon. Bloodflow was also normalized by MAP to yield vascular— 6040a-40.1 -IC3 —izot -160C—200— L—NNA-0.6I C.udiova5c PhannacoI, Vol. 22. No.4. 1993HALOTHANE INHIBITION OF L-NAME EFFECTS173575TABLE 2. Baseline values (means ± SE) of bloodflow and vascular conductance in conscious (groups IX—X) andhalothane-anaes:hetized (groups XI—XII) rats (n = 6 per group)Conscious HalothaneParameter Group IX Group X Pooled Group XI Group XII PooledBlood flow in each organ(ml/min/I0O g)Ltmg 72±14 111±24 91±15 267±39 184±59 225±37’Heart 269 ± 12 431 ± 63 350 ± 40 309 ± 67 201 ± 18 255 ± 38Liver 14±1 15±3 - 14±2 34±2 33±4 34±2’Stomach 71±7 72±6 71±5 41±6 32±3 36±4’Intestine 81 ± 4 79 ± 8 80 ± 4 55 ± 7 56 ± 6 56 ± 5’Colon 31±l 35±5 33±3 34±3 30±3 32±2Kidney 459 ± 54 591 ± 24 525 ± 35 422 ± 29 470 ± 23 446 ± 20’Spleen 143 ± 22 127 ± 20 135 ± 15 79 ± 10 75 ± 7 77 ± 6’Muscle 16 ± 2 16 ± 1 16 ± 1 6.9 ± 0.4 6.6 ± 0.8 6.7 ± 0.4Skin 19 ± 2 19 ± 1 19 ± 1 6.5 ± 0.4 6.9 ± 0.4 6.7 ± 0.3Testes 11±1 14±1 12±1 17±1 17±2 17±1’Brain 67 ± 6 80 ± 12 74 ± 7 163 ± 5 167 ± 12 165 ± 7’Vascular conductance in eachorgan (mllmm Hg rninJIOO g)Lung 0.57 ± 0.1 0.84 ± 0.19 0.71 ± 0.1 3.65 ± 0.56 2.28 ± 0.7 2.97 ± 0.50Heart 2.19 ± 0.1 3.27 ± 0.50 2.73 ± 0.3 4.15 ± 0.86 2.55 ± 0.2 3.35 ± 0.50Liver 0.11 ± 0.0 0.12 ± 0.03 0.11 ± 1.0 0.47 ± Ô.03 0.42 ± 0.0 0.44 ± 0.03Stomach 0.58 ± 0.0 0.54 ± 0.05 0.56 ± 0.0 0.55 ± 0.07 0.40 ± 0.0 0.47 ± 0.04Intestine 0.66 ± 0.0 0.60 ± 0.07 0.63 ± 0.0 0.75 ± 0.08 0.71 0.0 0.73 ± 0.06Colon 0.26 ± 0.0 0.26 ± 0.04 0.26 ± 0.0 0.46 ± 0.04 0.38 ± 0.0 0.42 ± 0.03Kidney 3.76 ± 0.4 4.48 ± 0.24 4.12 ± 0.2 5.73 0.30 5.98 0.3 5.86 ± 0.24Spleen 1.17 ± 0.1 0.97 ± 0.16 1.07 ± 0.1 1.10 ± 0.16 0.94 ± 0.0 1.02 ± 0.09Muscle 0.13 ± 0.0 0.12 ± 0.01 0.13 ± 0.0 0.09 ± 0.01 0.08 0.0 0.09 ± 0.01Skin 0.15 ± 0.0 0.14 ± 0.01 0.14 ± 0.0 0.09 ± 0.01 0.09 ± 0.0 0.09 ± 0.01Testes 0.09 ± 0.0 0.10 ± 0.01 0.01 ± 0.0 0.24 ± 0.01 0.21 ± 0.0 0.22 ± 0.01Brain 0.55 ± 0.0 0.61 ± 0.09 0.58 ± 0.0 2.24 ± 0.11 2.13 ± 0.1 2.18 ± 0.11‘ Significant difference from pooled data of conscious rats (p < 0.05).conductance values to reflect active changes in vascular tone. Comparison of pooled conductance values showed that halothane increased vascular conductance of lungs, liver, caecum and colon, kidney,testes, and brain, decreased conductance of muscleand skin, but did not affect conductance of otherorgans/tissues.In conscious rats, i.v. bolus injection of L-NNAas compared with vehicle decreased blood flow toall organs or tissues except liver and spleen (Fig.4A) Conductance values show that L-NNA vasoconstricted all beds (Fig. 4B). Changes in conductance were also expressed as percentages of conductance change to reflect the magnitudes of vaso..constriction in each organ/tissue response toL-NNA (Fig. 5). The results show that althoughL-NNA reduced vascular conductance in all beds inconscious rats, the greatest influence was in lungand the least was in liver.In halothane-anaesthetized rats, i.v. bolus injection of L-NNA as compared with vehicle decreasedboth blood flow and conductance in liver, intestine,spleen, and testes but not in other vascular beds(Fig. 6A and B). Comparison of percentage of conductance change in response of L-NNA in conscious and halothane-anaesthetized rats showedthat halothane reduced the vasoconstrictor effectsof L-NNA in every organ or tissue; significant reductions were obtained in all beds except liver andspleen (Fig. 5). The greatest inhibitory effect ofhalothane was in heart and lung, in which the effectof L-NNA was completely reversed from vasoconstriction to vasodilatation.DISCUSSIONOur results show that L-NNA increased MAP inconscious rats by increasing TPR, since both COand HR were reduced. Reduction in CO caused byL-NNA may be due to the effects of reduced HR.reduced cardiac contractility, and increased flowresistance (TPR). The bradycardic response toL-NNA was previously shown to be the result ofhypertension-induced activation of baroreflex activity (8).L-NNA vasoconstricted all beds, suggesting that- the endogenous EDRFINO system modulates vascular conductance in all organs/tissues in rats.Humphries and colleagues reported that intravenous infusion of L-NNA in conscious rabbits increased MAP (by 11 mm Hg) and TPR, reduced HRand CO. and caused significant vasoconstriction inbrain, heart, kidney, and duodenum, but not inmuscle, skin, stomach, ileum, or colon (15). Thegreater extent of vasoconstriction in response toL-NNA in our study may be due to use of a higherJ Cardiovaw Pharmaco!”. Vol. 22. No. 4. 1993576 Y.-X. WANG ET AL.174dose of c-NNA. Other analogues of N°-substitutedL-arglnine, N°-monomethyl-L-arginine (L-NMMA)and L-NAME were also reported to reduce CO anddecrease renal, mesenteric, hindquarters, or internal carotid blood flow in rats (16—19).The extent of vasoconstriction in response toL-NNA, as determined from percentage of conductance changes, is not uniform, however; i.e., thegreatest influence is in lung and the least is in liver.The lungs receive circulation from the bronchial artery, pulmonary artery, and arteriovenous anastomoses. Counts in the lungs primarily reflect circulation from the bronchial artery and arteriovenousanastomoses since microspheres are almost completely trapped in one circulation (unpublished observations). Likewise, owing to the entrapping ofmicrospheres in the splanchnic area, liver bloodflow is expected to represent primarily hepatic arterial flow. Hepatic arterial flow has been postulated to be controlled by the hepatic arterial bufferresponse so that decreases in portal venous flow areassociated with increases in hepatic arterial flow,thereby maintaining the constancy of total hepaticblood flow (20). This hypothesis is in accordancewith our findings that reduced splanchnic and consequently portal venous flow occurred concurrentlywith increased hepatic arterial flow. Therefore, theabsence of vasoconstrictor effect of L-NNA in thehepatic bed may be due to the hepatic arterial bufferresponse. L-NNA caused marked coronary vasoconstriction in conscious rats. NO synthase inhibitors cause coronary vasoconstriction in consciousrabbits (21) and dogs (22), as well as sustained increase in rabbit coronary perfusion pressure in vitro(21). In contrast, Klabunde and co-workers reported that L-NMMA and L-NNA did not reduceHR or coronary flow in pentobarbital-anaesthetizeddogs (23); their inability to show coronary constrictor effect of NO synthase inhibitors may have beendue to the influence of pentobarbital.Halothane reduced MAP, HR, CO, and TPR. Thehypotensive effect of halothane in this study probably is a consequence of reduced regional vascularresistances as well as myocardial depression as reflected by reduced HR. Halothane reduces HR andcardiac contractility by interfering with baroreflexactivity (24). The greatest vasodilatation effects ofhalothane were in lung, liver, and brain, wherethree- to fourfold increases in conductance valueswere obtained. Smaller degrees of vasodilatationwere also observed in caecumlcolon, kidney, andtestes. In contrast to our results, halothane was reported to decrease MAP, HR. and CO. but to havelittle effect on TPR (25).Our study shows that halothane noncompetitively and completely inhibited pressor responses ofL-NNA and L-NAME. Greenblatt and associatesreported that halothane did not affect the haemodynamic effects of L-NMMA in rats pretreated withindomethacin 5 mg/kg intravenously (i.v.) (26).However, in our experiments, halothane also abolished the pressor response of L-NNA in rats pretreated with the same dose of indomethacin (datanot shown). The results of Greenblatt and associates with halothane (26) are not characteristic because the anaesthetic nonselectively inhibits thevasoconstrictor effects of drugs [e.g., phenylephrifle (12), azepexole (11), potassium, prostanoids(27), and serotonin (13)]. Therefore, it is also important to rule out the nonspecific component of theinhibitory effects on L-NNA and L-NAME. Our results show that halothane noncompetitively inhibited pressor effects of NE and All by —18% butcompletely abolished the response of L-NNA andL-NAME. Nonselective inhibition of smooth muscle contractility by halothane may result from interference with calcium mobilization, since halothanewas shown to decrease tension developed in saponm-skinned rabbit aortic strips by depleting calciumin the sarcoplasmic reticulum (28). Because only asmall component of the abolition of the MAP response to L-NNA or L-NAME is nonselective, ourresults show that halothane selectively inhibits thevasoconstrictor effects to L-NNA and L-NAME.The mechanism by which halothane selectivelyinhibits the vasoconstrictor effects of NO synthaseinhibitors is not clear. Blaise reported that halothane potentiated acetylcholine-induced coronaryarteriolar dilatation in isolated perfused rabbithearts; the results suggested that halothane may release or potentiate the action of EDRF (29). Because L-NAME did not affect the relaxation responses of halothane and enflurane in canine middlecerebral arteries (30). release of NO may be causedby an alternative mechanism distinct from the well-known L-arginineINO synthetic pathway. We notedthat the NO donor (31,32) sodium nitroprusside andethanol, Like halothane, also selectively and non-competitively attenuated maximal pressor responseto L-NNA (33,34). Therefore, we postulate thathalothane inhibits the vasoconstrictor effects of NOsynthase inhibitors by releasing NO or potentiatingthe vasodilator effect of NO. In contrast to theabove observations, Muldoon and associates reported that halothane attenuated endotheliumdependent vasodilatation evoked by ACh andbradykinin but not endothelium-independent Vasodilatation produced by nitroglycerin in isolatedrabbit arteries, suggesting that halothane inhibitsrelease andlor action of EDRF (35). The differenteffects of halothane on endothelium-dependent reLaxation may result from different animal species,vascular preparations, and concentrations of halothane.In the haemodynarnic studies, halothane almostcompletely inhibited the effects of L-NNA on MAP,HR. CO. and TPR. Neither were the effects of halothane, as reflected by percentage of change in con-I Cardiovasc Pharmacoj’. Vol. 22. No.4. 1993HALOTHANE INHIBITION OF L-NAME EFFECTS175577ductance, uniform. Halothane significantly reducedthe vasoconstrictor effects of L-NNA in all bedsexcept the liver and spleen. The greatest inhibitoryeffect of halothane was in heart. Although halothane alone did not alter coronary conductance, itreversed the effect of L-NNA in the coronary bedfrom vasoconstriction to vasodilatation.L-NNA is an efficacious pressor agent which increases MAP by increasing TPR. TPR is increasedthrough generalized vasoconstriction, but the degree of influence is not uniform in all beds. Halothane causes selective and “noncompetitive” inhibition of vasoconstrictor and pressor responses toL-NNA and L-NAME. The results suggest that endogenous NO biosynthesis contributes to vasculartone and blood pressure; halothane may suppressthe vasoconstrictor effects of NO synthase inhibitors through release of NO or potentiation of thevasodilator action of NO.Acknowledgment: This work was supported by a grantfrom the Medical Research Council (MRC) of Canada andan MRC Fellowship to Y.-X.W.REFERENCES1. Ishii K. Chang B, Kerwin iF Jr, Huang Z-J. Muard F. N”Nitro-L-arginine: a potent inhibitor of endothelium-derivedrelaxing factor formation. Eur J Pharmacol 1990;176:219—23.2. Milsch A, Busse R. N°-Nitro-L-arginine (N5-[imino(njtroamjflo) methyl]-L-ornithine) impairs endotheliumdependent dilations by inhibiting cytosolic nitric oxide syn.thesis from L-arginine. Naunyn Schmiedebergs Arch Pharmacol 1990:341:143—7.3. Kobayashi Y, Hattori K. Nitroarginine inhibits endotheliumderived relaxation. Jpn J Pharmacol 19052:167—9.4. Moore PK, Al-Swayeh OA, Chong NWS, Evans RA, Gibson A. L-N’3-Nitro-arginine (L-NOARG), a novel, L-arginine-reversible inhibitor of endothelium-dependent vasodilation in vitro. BrJ Pharmaco! 1990;99:408—12.5. Wang Y-X, Pang CCY. Pressor effect of N°-nitro-L-argininein pentobarbital-anesthetized rats. Life Sci 1990:47:2217—24.6. Lacolley P1, Lewis Si, Brody Mi. L-/V°-Nitro arginine produces an exaggerated hypertension in anesthetized SHR.Ear J Pharmacol 1991:197:239—40.7. Lacolley P1, Lewis Si, Brody Mi. Role of sympatheticnerve activity in the generation of vascular nitric oxide inurethane-anesthetized rats. Hypertension 1991:17:881—7.8. Wang Y-X. Pang CCY. Possible dependence of pressor andheart rate effects of N°-nitro-i-arginine on autonomic nerveactivity. Br J Pharmacol 1991:103:2004—8.9. Wang Y-X, Thou T, Pang CCY. Pressor effects of L and Denantiomers of N°-nitro-L-arginine in conscious rats are antagonized by i.- but not D-arginine. Eur J Pharmacol 1991;200:77—81.10. Wang Y-X, Thou T, Chua TC, Pang CCY. Effects of inhalation and intravenous anaesthetic agents on pressor response to N°.nitro-L-arginine. Eur J Pharmacol 1991:198:1834.11. Larach DR. Schuler HG, Derr IA, Larach MG. Hensley FAJr. Zeus R. Halothane selectively attenuates alpha2 adrenoceptor mediated vasoconstriction, in vivo and in visro. Anesthesiology 1987:66:781—91.12. Kenny DK, PeIc LR. Brooks ilL. Kampinc JP, Scbmelingwr, Warftier DC. Alterations of alpha1 and alpha2 adrenoceptor-mediated pressor responses by halothane and isoflurane anesthesia. Anesthesiology 1989:71:224—31.13. Witzeling TM, Sill IC, Hughes JM, Blaise GA, Nugent M,Rorie DK. Isoflurane and halothane attenuate coronary axtery constriction evoked by serotonin in isolated porcinevessels and in intact pigs. Anesthesiology 1990;73:lOO-8.14. Pang CCY. Vasopressin and angiotensin in the control ofarterial pressure and regional blood flow in anaesthetizedsurgically-stressed rats. Can J Physiol Pharrnacol 1983:61:1494—500.15. Humphries RG, Carr RD. Nicol AK, TomlinsotiW, O’Connor SE. Coronary vasoconsu-iction in the conscious rabbitfollowing intravenous infusion of L-N°-nitro-argtnine. Br JPharmacol 1991:102:565—6.16. Gardiner SM, Compton AM, Bennett T. Regional and cardiac hanmodynamic responses to glyceryl trinitrate, acetyl.choline, bradykinin and endothelin-l in conscious rats: effects of ?‘d°-nitro-L-arginine methyl ester. Br J Pharmacol1990;lOl:632—9.17. Gardiner SM, Compton AM, Kemp PA, Bennett, T. Regional and cardiac haemodynainic effects of N°-nitro-Larginine methyl ester in conscious, Long Evans rats. Br IPharmacol 1990:101:625—31-18. Gardiner SM, Compton AM, Bennett T, Palmer RMJ, Moncads S. Control of regional blood flow by endotheiumderived nitric oxide. Hypertension 1990:15:486—92.19. Gardiner SM, Compton AM. Bennett T. Palmer RMJ, Moncada S. Regional haemodynamic changes during oral ingestion of ?J°-monomethyl-L-arginine or N°-nitro-L-argininemethyl ester in conscious Brattleboro rats. Br I Pharmacol1990:101:10—2.20. Lautt WW. Control of hepatic arterial blood flow: independence from liver metabolic activity. Am J Physiol 1980:239:H559-64.21. AniezcuaJL. Palmer RMJ. de Souza BM, Moncada S. Nitricoxide synthesized from L-arglrune regulates vascular tone inthe coronary circulation of the rabbit. Br I Pharmacol 1989;97:1119—24.22. Chu A, Chambers DE, Lin CC. Kuehl WI), Cobb FR. Nitricoxide modulates epicardial coronary basal vasomotor tonein awake dogs. Am I Physiol 1990;258:H1250—4.23. Klabunde RE, Ritger RC, Heigren MC. Cardiovascular actions of inhibitors of endotheium-derived relaxing factor (nitric oxide) formation/release in anesthetized dogs. Eur IPharmacol 1991:199:51—9.24. Marshall BE, Longnecker DE. General anesthetics. In: Gil-man AG, Rail TW, Nies AS, Taylor P. eds. Goodman andGilman’s the pharmacological basis of therapeutics. NewYork: Pergamon Press 1990:285—310.25. Seyde WC, Longnecker DE. Anesthetic influences on regional hacmodynamics in normal and hemorrhaged rats. Anesthesiology 1984:61:686—98.26. Greenblatt EP. Loeb AL, Longnecker DF. Anesthetic influences on endotheium-dependent circulatory control in vivo[abstract). FASEB I 1991;5:A399.27. Bollen BA, Tinker JH, Herrnsmeyer K. Halothane relaxespreviously constricted isolated porcine coronary artery segments more than isoflurane. Anesthesiology 1987;66:748—52.28. Su JY, Zhang CC. Intracellular mechanisms of halothane’seffect on isolated aortic strips of the rabbit. Anesthesiologyl989;71:409—l7.29. Blaise GA. Effect of volatile anaesthetic agents on endothehum-dependent relaxation. Adv E.xp Med Biol 1990301:229—35.30. Lee 1G. Coughlan MG. Bosniak V. Kampine JP. Vasodilatory effects of balothane and isoflurane in isolated caninecerebral artery segmcnts are not EDRF-mediated [Abstract]. FASE.B J 19926:A 1006.I Cardiovasc Pharnwcol’. Vol. 22. No. 4. 1993176578 Y.-X. WANG ETAL.31. Feeliscb M, Noack E, Schroder Fl. Explanation of the discrepancy between the degree of organic nitrate decornposition, nitrite formation and guanylate cyclasc stimulation.Ear Heart! t9889(suppl A)57-62.32. Ignarro LG. Endothelium-derived nitric oxide: actions andproperties. Fed Proc Soc Exp Biol 19893:31-6.33. Wang Y-X, Thou T. Pang CCY. A comparison of the inhibitory effects of sodium nitroprussidc, pinacidil and nifedipine on pressor response to M-nitro-t-arginine. Br I Pharmacal 1993;108:398—404.34. Wang Y-X, Pang CCY. Suppression by ethanol of pressorresponse caused by the inhibition of nitric oxide synthesis.Earl Pharmacól 1993,223:2754.35. Muldoon SM. Hart IL, Bowen KA, Freas W. Attenuationof endothellum-mediated vasodilatiou by halothane. Anesthesiology 1988;68:3 1—7.JCardiovascPkan,sacol’. VoL 22.No.4. 1993177Paper VIIWang, Y.-X., Poon, C.I., Poon, K.S. and Pang, C.C.Y.: Inhibitory actions ofdiphenyleneiodonium on endothelium-dependent vasodilatations in vitro and invivo. Br. J. Pharmacol. 110: 1232-1238, 1993. The reproduction of this paperwas kindly permitted by the copyright holder, Macmillan Press Ltd., Hampshire,U.K.Br. J. Pharmacol. (1993), 110, 1232— 1238178© Macmillan Press Ltd. 1993Inhibitory actions of diphenyleneiodonium on endotheliumdependent vasodilatations in vitro and in vivoYong-Xiang Wang, Christina I. Poon, Ken S. Poon & ‘Catherine C.Y. PangDepartment of Pharmacology & Therapeutics, Faculty of Medicine, The University of British Columbia, 2176 Health SciencesMall, Vancouver, B.C. V6T 1Z3, CanadaI This study examined the in vitro and in vivo inhibitory effects of diphenyleueiodonium (DPI), a novelinhibitor of nitric oxide (NO) synthase, on endothelium-dependent vasodilatations.2 DPI (3 x 10—3 x 10’ M) concentration-dependently inhibited acetyicholine (ACh)-induced relaxation in preconstricted rat thoracic aortic rings, with an IC of 1.8 x M and a maximal inhibition ofnearIy 100%. DPI (3 x l0’ M) also completely inhibited the relaxation induced by the calciumionophore, A23 187 but not by sodium nitroprusside (SNP). The inhibitory effect of DPI (3 x JO-’ M) onACh-induced relaxation was prevented by pretreatment with NADPH (5 x 10’ M) and FAD (5 xl0 M) but not L-arginine (L-Arg, 2 x 1O M). Pretreatment with NADPH did not alter the inhibitoryeffect of N°-nitro-L-arginine on ACh-induced relaxation.3 The inhibitory effect of DPI on ACh-induced relaxation in the aortae lasted >4 h after washout. Incontrast to pretreatment, post-treatment (I Ii later) with NADPH (5 x l0 M) reversed only slightly theinhibitory effect of DPI.4 In conscious rats, DPI (10’ mol kg’) inhibited the depressor response to i.v. infused ACh, but notSNP. However, it caused only a transient pressor response which was previously shown to be duecompletely to sympathetic activation.5 Thus, DPI is an efficacious and ‘irreversible’ inhibitor of endotheium-dependent vasodilatation atvivo and In vitro. The mechanism of the inhibition may involve antagonism of the effects of FAD andNADPH, co-factors of NO synthase. However, unlike the N°-substituted arginine analogues (anotherclass of NO synthase inhibitors), DPI-induced suppression of endothelium-dependent vasodilatation invivo does not lead to a sustained rise in blood pressure.Keywords: Diphenyleneiodonium (DPI); nitric oxide synthase inhibitor; endothelium-dependent relaxation; blood pressure;acetylcholine (ACh); A23 187; sodium nitroprusside (SNP); FADIntroductionA group of iodonium compounds have been reported to be anew class of nitric oxide (NO) synthase inhibitors in themacrophage (Stuehr et a!., 1991b; Kwon et al., 1991; Kelleret at., 1992). These compounds include diphenyleneiodonium(DPI), iodoniumdiphenyl and di-2-thienyliodonium, all ofwhich have chemical structures distinct from those of N°substituted arginine (Arg) analogues. DPI was initially foundto be a potent hypoglycaemic agent (Stewart & Hanley, 1969;Gatley & Martin, 1979) which, by inhibiting gluconeogenesisfrom lactate and aspartate, suppressed the oxidation ofNADH-linked substances (Holland et at., 1973). It was latershown that DPI, iodoniumdiphenyl and di-2-thienyliodoniumsuppressed the activities of neutrophil and macrophageNADPH-dependent oxidase (Cross & Jones. 1986; Hancock& Jones, 1987; Ellis et a!., 1988; 1989), probably via inhibition of a fiavoprotein (Cross & Jones, 1986; Hancock &Jones, 1987; Ellis er at., 1989; O’Donnell et a!., l993).The pharmacology of N°-substituted Avg analogues, whichinclude N°-monomethyl-L-Arg (L-NMMA), N°-nitro-L-Arg(L-NOARG), N°-nitro-L-Arg methyl ester (L-NAME), Liminoethyl-ornithine (L-NIO) and N°-amino-L-Arg (L-NAA)(see Moncada et at., 1991), has been extensively studied.These compounds cause sustained inhibition of endotheliumdependent relaxation in vitro and produce prolonged pressorresponses in whole animals (Aisaka et at., 1989; Rees el aL,1989; 1990; Wang & Pang, 1990: Wang et a!.. 1991; 1992;1993; Pang & Wang, 1993). The pressor effects of thesecompounds have been attributed to the inhibition of theL-Arg/NO pathway and endothelium-dependent vasodilatation in situ (Aisaka er at., 1989; Rees et a!., 1989; see Moncada el aL, 1991). Accordingly. DPI being an inhibitor ofNO synthase, would be expected to cause a pressor responsein whole animals. Indeed, i.v. injections of DPI producedtransient pressor responses in pentobarbitone-anaesthetizedand conscious rats (Wang & Pang, 1993a,b). However, unlikethat of the N°-substituted Arg analogues (Wang & Pang,1991), the pressor effect of DPI was attenuated or abolishedby blockers of the sympathetic nervous system (Wang &Pang, 1993a,b).The reasons for the differences in the causative factor andtime course of the pressor responses elicited by these twoclasses of NO synthase inhibitors are unclear. One possibilityis that DPI does not produce prolonged inhibition of endothelium-dependent vasodilatations in vitro and/or in vivo.Another possibility is that inhibition of endothelium-dependent vasodilatation by DPI is not the cause of the elevationof blood pressure. Hence, the first aim of this study was tofind out if DPI causes prolonged in vitro and in viva inhibition of endothelium-dependent vasodilatations, as DPI hasbeen shown to cause prolonged inhibition of NO biosynthesisin macrophages (Stuehr et at., 1991b). This was assessed bystudying the effects of DPI on relaxations induced by theendothelium-dependent vasodilators, acetylcholine (ACh) andA23 187 (calcium ionophore), and the endothelium-independent vasodilator sodium nitroprusside (SNP) in preconstricted rat aorta. In addition, the effects of DPI on depressorresponses to ACh and SNP were studied in conscious. unrestrained rats. The second aim was to determine if slowerOnset pressor responses (other than the initial transient rise inblood pressure) followed the administration of DPI. Thethird aim was to examine if the NO synthase co-factorsNADPH and FAD, as well as the NO synthase substrateL-Arg, reversed the inhibitory effect of DPI on endotheliumdependent relaxation.‘Author for correspondence.DPI AND ENDOTHELILM-DEPENDENT VASODILATATLON 1233179MethodsMale Sprague-Dawley rats (350—420 g) were used in thisstudy.Isolated aortic ringsThe rats were killed by a blow on the head followed byexsanguination. The thoracic aorta was removed and clearedof connective tissue. Four ring segments of 0.5 cm lengthwere prepared from one aorta and suspended in randomorder in separate organ baths. Each ring was connected to aGrass FT-03-C force-displacement transducer for isometricrecording with a preload of 1 g. The rings were equilibratedfor I h (with 3 washouts) in Krebs solution (pH 7.4) at 3’rCand bubbled with a gas mixture of 95% 02 and 5% Co2.The Krebs solution had the following composition (mM):NaCI 118, glucose 11, KC1 4.7, CaCI2 2.5, NaHCO3 25,KH2PO4 1.2, MgCI6HO 1.2.The rings were first incubated with vehicle or drugs (seelater) followed by phenylephrine (PE, l0_6 M, EC). After15—20 miii. at the steady-state phase of the contractile response to PE, a cumulative concentration-response curve toACts, A23 187 (calcium ionophore) or SNP was obtained.Each drug concentration was left in the bath until a plateauresponse was reached. The time taken to complete eachconcentration-response curve was approximately 20 mm. Ingroups where more than one concentration-response curve ofACh was constructed, the preparations were washed threetimes within 30 mm and given another 30 mm to recovercompletely from the effects of the previous applications of PEand ACh. Afterwards. PE was again added followed by theconstruction of ACh concentration-response curves.Conscious ratsThe rats were anaesthetized with halothane (4% in air forinduction and 1.2% in air for surgery). Polyethylene cannulae (PESO) were inserted into the left iliac artery for themeasurement of mean arterial pressure (MAP) by a pressuretransducer (P23DB. Gould Statham. CA. U.S.A.) and intothe left iliac vein for the administration of drugs. The cannulae were filled with heparinized (25 iu ml ‘) normal saline,tunnelled s.c. along the back and exteriorized at the back ofthe neck. The rats were put into small cages allowing freemovement and given >6 h recovery from the effects of surgery and halothane before use.DrugsThe following drugs were purchased from Sigma ChemicalCo. (MO, U.S.A.): acetyicholine (ACh) chloride, A23 187,phenylphnne (PE) hydrochloride, fiavin adenine dinucleotide(FAD) disodium, -nicotinamide adenine dinucleotide phosphate, reduced form (NADPH), N”-nitro-L-arginine (LNOARG) and L-arginine (L-Arg) hydrochloride. Diphenyleneiodonium (DPI) sulphate and sodium nitroprusside (SNP)were obtained from Colour Your Enzyme Ltd. (Ontario,Canada) and Fisher Scientific Co. (N.J., U.S.A.), respectively. All drugs were dissolved in normal saline (0.9% NaCI)except for DPI and A23187 which were dissolved in 5%glucose solution and 10% dimethyl sulphoxide, respectively.Experimental protocolsEach experiment included 6—8 aortic rings or 6 conscious,unrestrained rats.Protocol 1: Effects of DPI on ACh-, A23187- and SNPinduced relaxations in the aorta Six groups of aortic ringswere incubated with the vehicle or DPI (3 x l0—3 xl0_6 M) followed by PE 10 mm later. After another 10—15mm, concentration-response curves to ACh (l0—3 x l0 M)were obtained. Only one concentration of DPI was studied ineach group. Another two groups were treated with the vehicle or DPI (3 x l0 M) followed by the application of PEand construction of concentration-response curve to A23 187(l0-’—106M). The last two groups were treated in the sameway as the previous two groups except that SNP (3 x l0’iO M) was used in place of A23 187.Protocol 2: Effects of pretreatment with NADPH, FAD orL-Arg on the inhibitory effect of DPI Twelve groups ofaortic rings were treated with vehicle + vehicle, NADPH(1.5 x IO M) + vehicle, NADPH (5 x l0 M) + vehicle, FAD(5 x l0_6 M) + vehicle, FAD (5 x I0 M) + vehicle and LArg (2 x 10 M) + vehicle, vehicle + DPI, NADPH (1.5 xl0 M) + DPI, NADPH (5 x l0- M) + DPI, FAD (5 xl0_6 M) + DPI, FAD (5 x l0- M) + DPI, and L-Arg (2 xlO M) + DPI, with 3 x l0- M DPI added in all cases.Another two groups of aortic rings were treated withvehicle + L-NOARG (l06 M) and NADPH (5 x 10’M) +L-NOARG (1O_6 M). The first treatments were given 10 missprior to the second treatments. Afterward, the rings werepreconstricted with PE and relaxed with ACh as described inProtocol 1.Protocol 3: Time course and reversibility of the inhibitoryeffect of DPI on relaxation response of ACh The time courseof the inhibitory effect of DPI was studied in 3 groups ofaortic rings. After completing the first concentration-responsecurve to ACh in PE-preconstricted rings in the presence ofvehicle or DPI (3 x lO- or 3 x lO_6 M), the preparationswere washed out without further addition of drug or vehicle.Second, third and fourth ACIi curves were constructed inprecoristricted rings at 1.5, 4 and 9 h after the preparationswere washed out. In another two groups of aortic rings, theeffect of post-treatment with NADPH on the inhibitory effectof DPI was studied. These rings were incubated with DPI(3 x lO- M) or vehicle for I h. followed by the application ofNADPI-{ (5 x lO M) and the construction of ACh concentration-response curves in PE-preconstricted conditions.Protocol 4: Effect of DPI on resting MAP and depressorresponse to ACI and SiVP In one group of rats. DPI(IO— mol kg’) was injected (i.v. bolus) and blood pressurewas continuously monitored for 2 h.Two groups of rats were injected (i.v. bolus) with thevehicle or DPI (lOS mol kg-’) 20 mm prior to i.v. infusionsof ACh (6 x lO— 1.8 x l06 mol kg’ min’. each dose for4mm) and SNP (3 x lO—4.8 x 10’molkg’ min’. eachdose for 4 mm). The sequence of ACh and SNP administrations was reversed in half of the studies with 20 mm recoveryafter the completion of the first dose-response curve. Thetime taken to complete the experiment was approximately 2 hafter i.v. injection of DPI or the vehicle.Calculation and statistical analysisIC and E,,.. were calculated from individual concentration-response curves (see Wang & Pang, l993a). All results areexpressed as mean ± standard error (s.e.mean) except for thepoints where the error bars were smaller than the symbols(see figures). The results were analysed by the analysis ofvariance/co-variance followed by Duncan’s multiple rangetest with P <0.05 selected as the criterion for statisticalsignificance.ResultsEffects ofDPI on ACh, A23187- and SNP-inducedrelaxationsAll five concentrations of DPI (3 x l0-. 3 x l0. l0_6and 3 x lO_6 M) slightly potentiated PE-induced contraction1234 Y.-X.WANGeIaI.180from the baseline value of 0.99 ± 0.10 g to 1.09 ± 0.15,1.27±0.11, 1.32±0.10, 1.31±0.10 and 1.18±0.14g. respectively. However, only the effects of the third and fourthconcentrations of DPI were statistically significant.In the vehicle-treated group, ACh relaxed the preconstricted aorta concentration-dependently with maximum relaxation of approximately 60% (Figure Ia). DPI inhibitedconcentration-dependently the ACh-induced relaxation. At3 X I0 M ACh, the IC of DPI was 1.8 x lO M with amaximum inhibition of 96% (Figure Ib).In another two vehicle groups, A23187 and SNP alsorelaxed concentration-dependently the preconstricted aortae,with maximal relaxations of approximately 60 and 100%,respectively. DPI (3 x l0_6 M) completely inhibited A23187-induced relaxation (Figure 2a) but did not affect the relaxation response of SNP (Figure 2b).Influences of NADPH, FAD and L-Arg on the inhibitoryeffects of DPI and, the effect of L-NOARG on AChinduced relaxationBaseline contractions elicited by PE in the presence of vehicleor DPI (3 x 107M) were 1.29±0.07 and 1.67±0.12g. respectively. Treatment with NADPH (1.5 and S x 10 M),FAD (5 x 10-’ and 5 x 10’ M) and L-Arg (2 x 10 M) didnot significantly affect PE-induced contractions in the presence of either the vehicle (1.04 ± 0.06, 1.04 ± 0.11, 1.16 ±0.06, 1.33±0.11, 1.23±0.08g. respectively) or DPI (1.43±0.10, 1.37 ± 0.14, 1.59 ± 0.06, 1.52 ± 0.13, 1.44 ± 0.04g.respectively).DPI inhibited ACh-induced relaxations (Figure 3a). Treatment with L-Arg did not affect either the ACh-induced-100 —Figure 1 (a) Concentration-response (mean ± s.e.mean) curves ofdiphenyleneiodonium (DPI) on acetylcholine-induced relaxation inphenylephrine (10-’ M)-preConstricted aoruc rings (n 7 each group).Concentrations of DPI were as follows: vehicle (0); 3 X 10’ M (•);lO M (A); 3 X l0 ,. (A); 10’ M (C): 3 x 10’ u (a). (b)Percentage inhibition by DPI of 3 x 10 N acetylcholine-inducedrelaxation in the aorta. The data were derived from the mean valuesof (a).—9—8 —7 —6log (sodium nitroprussidel (MlFigure 2 Effects (mean ± s.e.mean) of vehicle (0) or diphenyleneiodonium (DPI, 3 x 10 N, •) on A23 187 (a) and sodium nitroprusside (b) induced relaxations in the phenylephrine (10—’ N)preconstricted aortic rings (n = 6 each group). ‘Significant differencefrom vehicle-pretreated control curve (P <0.05).—8 —7 —6 —5 —4 —8 —7 —6 —5 —4log [acetyicholine] (MlFIgure 3 Effects (mean ± s.e.mean) of vehicle, NADPH. FAD andL.-arg nine on acetylcholine-induced relaxation and on the inhibitoryinfluence of diphenyleneiodonium (DPI. 3 x iO M) on relaxation inphenylephrine (10’ M). preconstricted aortic rings (n 6 eachgroup). The first treatments were performed 10 mm before thesecond treatments. (a) Vehicle + vehicle (0); vehicle + DPI (•). (b)L-Arglnine (2 x 10’ N) + vehicle (0); L-arginine (2 x l0’ M) + DPI(•). (c) FAD (5 x l0’ M) + vehicle (0); FAD (5 x l0’ N) + vehicle(•); FAD (5 x 10-’ M) + DPI (A); FAD (5 x 10-i N) + DPI (A).(d) NADPH (1.5 x IO N) + vehicle (0); NADPH (5 x lO- N) +vehicle (•); NADPH (1.5 x 10’M) + DPI (A); NADPH (5 x l0N) + DPI (A). *Signifieant difference from control curve (P<0.05).C0‘ax‘0‘aC0‘aSi‘a‘aa10-—10 -—30-50 -—70 —b20 -—20-60--100 --140 -—10I I r I , I I—6 —5—9 —8 —7log IA 231871 (M)I I , Ia200—20-40—60—80 -I . I . I . I . IC0‘ax‘a‘a>-.0C0-0<01C0-Ox.0 c’)Cb—8 —7 —6 —5log [acetylcholinej (N)—4120100806040200a—bc -C200--20--60-—8000d—8 —7 —6log IDPII(M)181DPI AND ENDOTHELIUM-DEPENDENT VASODILATATION 1235relaxation or the inhibitory effect of DPI on ACh-inducedrelaxation (Figure 3b). Although the lower concentration(5 x lO_6 M) of FAD also did not alter either ACh-inducedrelaxation or the inhibitory effect of DPI on ACh, the higherconcentration (S x l0 M) of FAD suppressed the relaxanteffect of ACII and prevented further inhibition by DPI onACh-induced relaxation (Figure 3c). Although neither concentration of NADPH significantly affected ACh-induced relaxation, the higher (5 x lO M) but not the lower (1.5 x iOM) concentration completely prevented the inhibitory effectof DPI (Figure 3d). The effectiveness of pretreatment withNADPH (5 x l0 M) in inhibiting the effect of DPI, expressed as the ratio of the relaxation effect of iO M ACh inthe presence of NADPH (—67%) to that in the absence ofNADPH (—32%), was 209%.PE caused contractions of 1.46 ± 0.12 and 1.67 ± 0.13 g inthe presence of vehicle+ L-NOARG (10-’ M) and NADPH(5 x l0 M) + L-NOARG (10 M), respectively. Comparedto the pooled vehicle control derived from Figures Ia and 3a,L-NOARG markedly inhibited ACh-induced relaxation. Pretreatment with NADPH did not affect the inhibitory effect ofL-NOARG (Figure 4).Time course and reversiblity of the inhthitory effect ofDPI on ACh-induced relaxationThe PE-induced contractions in the presence of vehicle orDPI did not change with the passage of time (data notshown). The ACh-induced maximal relaxation was not altered until at least 4 h after washout. Maximal relaxation at 9 hwas —48 ± 7%, which was significantly less than that(—69 ± 6%) at Oh (Figure 5). DPI at 3 x 1O and 3 x 10-’ Minhibited ACh-induced relaxation by approximately 50 and100%, respectively (Figure 5a). The inhibitory effect of DPIremained at least 4 h after washout (Figure 5b,c). At 9 hafter washout, the relaxations of DPI-pretreated rings werestill less, though insignificantly, than those of vehiclepretreated rings (Figure Sd).Maximum relaxation to ACh after 1 h exposure to 3 x10-’ M DPI (—38.3 ± 3.1%, Figure 6) was similar to thatafter a 10 mm exposure to DPI (—32.2 ± 4.8%, Figure 3a).Post-treatment (I h later) with NADPH (5 x 1O M) slightlybut significantly, suppressed the inhibitory effect of DPI. Theeffectiveness of post-treatment with NADPH (5 x 1O M) ininhibiting the effect of DPI, expressed as a ratio of therelaxation effect of lO M ACh in the presence of NADPH(—51%) to that in the absence of NADPH (—38%), was134% (Figure 6).b d2O:—8 —7 —6 —5 —4 —8 —7 —6 —5 —4log [acetdcholine1 (M)Figure 5 The time course of the effects (mean ± s.e.mean) of vehicle(0), diphenyleneiodonium (DPI, 3 x I0 M, •) and DPI (3 xlO_6 M, ) on acetylcholine-induced relaxations in phenylephrine(106 M) preconstricted aortic rings (n = 6 each group). (a), (b), (c)and (d) represent responses at 0, 1.5, 4 and 9 h after washoutwithout further addition of the vehicle or DPI. Sigrnficant differencefrom vehicle-pretreated control curve (P <0.05).—TOOFigure 6 Effects (mean ± s.e.mean) of post-treatment (1 h later)with vehicle (0) or NADPH (5 x l0 M, •) on the inhibitory effectof diphenyleneiodonium (DPI, 3 x 1O M) on acetylcholine-inducedrelaxation in the phenylephrine (I0_6 M) preconstricted aortic rings(ii = 6 each group). *Significant difference from vehicle-pretreatedcontrol curve (P <0.05).I I I—8 —7 —6 —5 —4log Iacetylcholinej (MlFigure 4 Effect (mean ± s.e.mean) of pretreatments (10 mm earlier)of NADPH (5 x l0’ M) on the inhibitory effect of N°-nitro-Larginine (x.-NOARG, l0’ M) on acetylcholine.induced relaxation inthe phenylephrine (10’ M) preconstricted aortic rings (is 8 eachgroup except for the pooled control rings where is = 13). Vehicle (0);vehicle + L-NOARG (•); NADPH + L-NOARG (a). ‘Significantdifference from vehicle-pretreated control curve (P <0.05).*huh0 30 60 90 120Time (minIFIgure 7 Effect (mean ± s.e.mean) of i.v. bolus injections of diphenyleneiodonium (DPI, iO mol kg’) on mean arterial pressure(MAP) in conscious rats (a =6). Open and solid columns representpre- and post-administration with DPI. ‘Significant difference frompre-administration with DPI (P<0.05).a200—20-40-60-80C0I0xCuCu200•—20—40 --60 --80-C0IsxIsCuLg—7—6 —5 —4log facetylcholinel (M)200C0 —20Is-40C-60-80-100180150zEE 120a.9o60——30Ii1236 Y.-X. WANG ci at.182Figure 8 Dose-response curves (mean ± s.e.mean) of i.v. infusionsof acetyicholine (a) and sodium nitroprusside (b) on mean arterialpressure (MAP) in conscious rats (n = 6 each group) pretreated withiv. bolus injection of vehicle (0) or diphenyleneiodonium (DPI,lO mol kg’. •). *signifiesnt difference from vehicle-pretreatedcontrol curve (P <0.05).Effects of DPI on resting blood pressure and depressorresponses to A Ch and SN?Intravenous bolus injections of DPI (I0 mol kg’) in conscious rats caused immediate and transient increases in MAPwhich were similar to the responses in pentobarbitoneanaesthetized and conscious rats (Wang & Pang, l993a,b).MAP returned to the baseline level approximately 4 mm afterthe injection of DPI and remained there during the 2 hobservation period (Figure 7).Baseline MAPs of rats before and 20 mm after treatmentwith vehicle were 109 ± 2 and 112 ± 3 mmHg. respectively,which were similar to those of DPI-treated rats (10 molkg’, iv. bolus) (118 ± S and 114 ± 5 mmHg). Intravenousinfusions of ACh and SNP caused dose-dependent depressorresponses. Pretreatment with DPI significantly attenuated thedepressor responses to ACh but not to SNP (Figure 8).DiscussionOur in Vitro results show that DPI selectivity inhibits endothelium-dependent relaxation induced by receptor-mediated(ACh) or non-receptor-mediated (A23 187) mechanisms. Theresults are consistent with the report that DPI inhibits AChinduced relaxation in the rabbit aorta (Stuehr et a!.. 1991 b)and further suggest there is no species difference for theactions of DPI. DPI also attenuates ACh- but not SNPinduced decreases in MAP in conscious rats, and ACh- butnot SNP-induced vasodilatation in the perfused rat hindquarter preparation (unpublished observation). The resultssuggest that DPI inhibits endothelium-depeadent vasodilatations in both conductance and resistance vessels. Therefore.the in vitro inhibitory effects of DPI on endothelium-dependent vasodilatations are similar to those of the N°-substituted Arg analogues. These results are in accordance withthe hypothesis that inhibition of NO synthesis causes thesupfression of endothelium-dependent vasodilatation.It has been known since 1973 that DPI suppresses theoxidation of NADH-like substrates thereby inhibiting mitochondrial oxidation (Holland el at.. 1973). It was later shownthat DPI inhibits NADPH-dependent oxidase of neutrophilsand macrophages (Cross & Jones, 1986: Hancock & Jones,1987: Ellis ci at., 1988; 1989), and macrophage NO synthase(Stuehr et a!., l991b), by specifically binding to andinhibiting the action of a plasma membrane polypeptidewhich may be a component of flavoprotein (Cross & Jones,1986; Hancock & Jones, 1987; Ellis et at., 1989). This suggests that flavin is the Site of attack by DPI and that aprotein is associated with FAD (O’Donnell et at., 1993).Isoenzymes of NO synthase are known to be flavoproteinswhich contain FAD as a cofactor in the macrophage (Stuehret at., 1980; 1990; 199la; Hevel et a!., 1991; White &MarIetta, 1992), neutrophil (Yui et at.. 1991), brain (Mayerel at., 1991: Lowenstein et at.. 1992; Bredt er a!., 1991; 1992;Hiki et a!., 1992) and liver (Evans et at., 1992). There is.however, no functional documentation of a role for FAD asa cofactor of NO synthase in endothelial cells. Our in vitroresults demonstrate that FAD interferes with both AChinduced relaxation and the inhibitory effect of DPI on AChinduced relaxation. The latter result, which is consistent with—6Stuehr et at’s observation (l99lb) that FAD antagonizes theinhibitory effect of DPI on macrophage NO synthesis, suggests that FAD and DPI may inhibit endothelial NO synthesis by a mechanism similar to that in macrophages. Theformer result is puzzling, since as a cofactor. FAD shouldfacilitate rather than interfere with endothelium-dependentrelaxation. FAD was indeed reported to facilitate macrophage NO synthesis(Stuehr et at.. 1990: Hevel ci at.. 1991).The mechanism by which FAD inhibits ACh-induced relaxation is not clear at the moment, however, the effect may notbe specific as FAD also inhibits SNP-induced relaxation(unpublished observation).Our in vitro results also show that NADP1-{ interferes withthe inhibitory effect of DPI on ACh-induced relaxation. Theantagonism of DPI by NADPH was specific since the sameconcentration of NADPH did not alter the inhibitory effect01 L-NOARG. The inhibitory effect of DPI was also notaffected by L-Arg, at a concentration previously found toreverse the inhibitory effects of L-NOARG and L-NAME onendothelium-dependent relaxations in aortic rings (Wang cia!.. 1992: 1993). Our results with NADPH are consistent withthose which show that both the constitutive (e.g. brain andendothelial) and inducible (e.g. macrophage and smoothmuscle) NO synthases are dependent on NADPH as anessential cofactor (Mayer ci at.. 1989: Stuehr ci at.. 1989:1990: 199 Ia: see McCall & Valiance. 1992). Regarding thenature of the interaction between NADPH and FAD. it hasbeen suggested that NADPH suppresses the binding of DPIto the fiavoprotein in neutrophil oxidase by preventing theattachment of DPI to a site in close proximity to theNADPH-binding site (Cross & Jones. 1986). It is very likelythat NADPH may interfere with the action of DPI onendothelial NO synthase by the same mechanism.Pretreatment with DPI was found to inhibit ACh-inducedrelaxation in aortic rings for at least 4 h after washout and tosuppress ACh-induced vasodilatation for at least 2 h afterintravenous bolus injection. Therefore, our in vitro and invivo results are supportive of a prolonged inhibitory effect ofDPI on endothelium-dependent vasodilatations. DPI hasbeen reported to inhibit irreversibly macrophage NO synthase (Stuehr ci a!., I99lb); the mechanism may involve theformation of a covalent bond with components of a flavoprotein (Ragan & Bloxham. 1977; O’Donnell ci a!.. 1993).However, our results show that post-treatment (1 h later)with NADPH still attenuates the effect of DPI. although theresponse is significantly less than that following pretreatmenta0--20IE-40a<1—60 ‘—7 —6log acetytcholine (mol kg1 min1)OiIEEab0——20 —-40 -—60 -—80 ——8 —7log sodium nitroprusside (mol kg’ m1n1)183DPI AN[) ENDOTHELI UM-DEPENDENT VASODILATATION 1237(10 mm earlier). These results may imply that fresh synthesisof NO occurs in endothelial cells.It is well-known that all N°-substituted Arg analogueswhich inhibit endothelium-dependent relaxation in vitro causelong-lasting pressor effects in whole animals. The pressorresponse of N°-substituted Arg analogues is not blocked bythe impairment of the central nervous system (Tabrizchi &Triggle, 1992; Wang & Pang, 1993a), sympathetic nervoussystem (Wang & Pang, 1991), renin-angiotensin system(Wang & Pang, 1991), or prostaglandin system (Rees el at.,1989; Wang & Pang, unpublished data, 1993), but is inhibited by L-Arg (Aisaka et at., 1989; Rees et a!., 1989; Wang &Pang, 1990; Wang et at., 199lb; 1992). These observationshave been accepted as evidence of a role of NO in theregulation of blood pressure (Rees et at., 1989; Aisaka el at.,1989; see Moncada et a!., 1991). As an ‘irreversible’ inhibitorof NO synthase, DPI should also cause a prolonged pressorresponse. However, unlike the N°-substituted Arg analogues,intravenous bolus injections of DPI produced only immediateand transient increases in MAP. The pressor response of DPIwas blocked by procedures which impair the activities of thecentral or sympathetic nervous systems, namely, pithing,spinal cord transection and the administration of tetrodotoxin, reserpine, guanethidine, phentolamine or prazosin(Wang & Pang, 1993a). Moreover, the pressor response toDPI, but not to L-NOARG, was accompanied by elevationsof plasma noradrenaline and adrenaline (Wang & Pang,1993a,b). These results show that the transient pressor response of DPI. unlike that of the N°-substituted Arg analogues, is solely dependent on the activation of the sympatheticnervous system. i.e.. DPI does not elicit a NO-dependentsustained rise in blood pressure as do the other NO synthaseinhibitors.Although one may postulate that the lack of effect of DPIis due to inadequate accumulation of drug in situ to inhibitNO synthesis, this is unlikely. DPI was shown to distributerapidly and adequately to all organs or tissues (Gatley &Martin. 1979); moreover, its peak hypoglycaemic effect wasreached at 1.5 h (Holland et a!.. 1973) or 4 h (Gatley &Martin. 1979) after intraperitoneal injections, suggesting along duration of action. Our present results also show thatDPI inhibits irreversibly endothelium-dependent relaxation invitro for more than 4 h. and partially inhibits ACh-inducedvasodilatation in vito even at 2 h after intravenous Injection.It should be noted that a lack of complete inhibition ofReferencesAISAKA. K.. GROSS. S.S.. GRIFFITH. OW. & 1.EVI. R. (1989). N°.methylarginine. an inhibitor of endothelium-derived nitric oxidesynthesis. is a potent pressor agent in the guinea-pig: does nitricoxide regulate blood pressure in vii’o Biochem. Biophys. Res.Commun.. 160, 881—886.BREDT. D.S.. FERRIS. CD. & SNYDER. S.H. (1992). Nitric oxidesynthase regulatory sites. Phosphorylation by cyclic AMP-dependent protein kinase. protein kinase C. and calcium/calmodulinprotein kinase: identification of fiavin and calmodulin bindingsites. J. Riot. Chem.. 267, 10976—10981.BREDT. OS.. HWANG. P.M.. GLATr. C.E.. LOWENSTEIN. C... REED.R.R. & SNYDER. S.H. (1991). Cloned and expressed nitric oxidesynthase structurally resembles cytochrome P450 reductase.Nature, 351, 714—718.COCKS. T.. WARD. J.. CRACK. P.. STORK. A. & ANGUS. 3. (1992).Comparison of the effects of the endothelium-derived relaxingfactor substrate inhibitors N°-nitro-L-arginine and N°-monomethyl-L-arginine in vipo. in vitro and in cultured endothelial. cell.In The Biology of Nitric Oxide I. Phsiological and ClinicalAspects. ed. Moncada. S.. MarIetta, M.A.. Hibbs, J.B. Jr &Higgs, E.A. pp. 78—79. London: Portland.CROSS. A.R. & JONES. O.T.G. (1986). The effect of the inhibitordiphenylene iodonium on the superoxide-generating system ofneutrophils. Biochem. J., 237, 111—116.ACh-induced relaxation is a typical observation with NOsynthase inhibitors since N°-substituted Arg analogues, atmaximal pressor doses, also cause partial inhibition of AChinduced vasodilatation — this suggests that either thedepressor/vasodilatation response of ACh in vivo is onlypartially due to the release of NO (Rees et a!., 1990; Wang eta!., 1992) or it is entirely independent of the biosynthesisand/or release of NO (Pang & Wang, 1993).Although it is generally accepted that endogenous NOmodulates vascular tone and blood pressure and that N°substituted Arg analogues produce pressor response byinhibition of endothelial NO synthesis and endotheiumdependent vasodilatations in situ (see Moncada er at., 1991),our data with DPI suggest otherwise, i.e. inhibition of NOsynthesis and endothelium-dependent vasodilatations do notalways cause vasoconstriction in vivo. This hypothesis is supported by the recent publications which show that methyleneblue does not produce a pressor response (Loeb & Longnecker, 1992; Pang & Wang, 1993) although it inhibitsendothelium-dependent vasodilatation in vitro (Pang &Wang, 1993) and in vivo (Loeb & Longnecker, 1992). I.NOARG was also shown to cause much longer inhibition ofendothelium-dependent vasodilatation than elevation ofblood pressure in conscious rabbits, suggesting that the suppression of NO synthesis alone does not result in hypertension (Cocks et a!., 1992). Therefore, the hypothesis thatendogenous NO modulates vascular tone and Arg analoguesproduce pressor response by inhibition of endothelial NObiosynthesis may need re-examination.In summary, DPI efficaciously and ‘irreversibly’ inhibitsendothelium-dependent vasodilatation in vitro and in vivo bya mechanism involving the suppression of the actions ofFAD and NADPH. Unlike the N°-substituted Arg analogues. DPI does not cause NO-mediated sustained pressorresponse. Instead. DPI causes immediate and transient pressor responses which are solely due to the activation of thesympathetic nervous system (Wang & Pang, l993a,b). Theseresults suggest that inhibition of NO synthesis in situ doesnot necessarily cause a pressor response.This work was supported by a research grant from the MedicalResearch Council of Canada (MRC). an MRC Post-doctoral Fellowship to Y.-X.W. and a Heart & Stroke Foundation of B.C. & YukonResearch Traineeship to C.I.P.ELLIS. J.A.. CROSS. AR. & JONES. T.G. (1989). Studies on theelectron-transfer mechanism of the human neutrophil NADPHoxidase. Biochem. J.. 262, 575—579.ELLIS. 3.. MAYER. SI. & JONES. O.T.G. (1988). The effect of theNADPH oxidase inhibitor diphenyleneiodonium on aerobic andanaerobic microbicidal activities of human iseutrophils. Biochem.1.. 251, 887—891.EVANS. T. CARPENTER. A. & COHEN. J. (1992). Purification of adistinctive form of endotoxin-induced nitric oxide synthase fromrat liver. Proc. Vail. .4cad. Sd. U.S.A., 89, 5361—5365.GATLEY. SJ. & MARTIN. IL. (1979). Some aspects of the pharmacology of diphenyleiodonium, a bivalent iodine compound.Kenobiotica. 9. 539—546.HANCOCK. J.T. & JONES. O.T.G. (1987). The inhibition by diphenylenetodonium and its analogues of superoxide generation bymacrophages. Biochem. 1.. 242, 103—107.HEVEL J.M.. WHITE. K.A. & MARLETrA. MA. (1991). Purificationof the inducible murine macrophage nitric oxide synthase: identification as a fiavoprotein. J. Biol. (‘hem., 266, 22789—22791.HIKI. K.. HATfORI. R.. KAWAI. C. & YUI. Y. (1992). Purification ofinsoluble nitric oxide synthase from rat cerebellum. J. Biochem.(Tokyo), 111, 556—558.1238 Y.-X. WANG ci at.184HOLLAND. P.C.. CLARK, MC., BLOXHAM, OP. & LARDY, H. (1973).Mechanism of action of the hypoglycemic agent diphenyleneiodonium. J. BiOL Chem., 17, 6050—6056.KELLER. R., BASSETTI, S., KEIST. R.. MULSCH, A. & KLAUSER, S.(1992). Induction of nitric oxide synthase is a necessary precondition for expression of tumor necrosis factor-independent tumoricidal activity by activated macrophages. Bioc.hem. Biophys. Rex.Commw,., 184, 1364—1371.KWON, N.S., STUEHR, 0.3. & NATHAN, C.F. (1991). Inhibition oftumor cell nbonucleotidc reductase by macrophage-derived nitricoxide. J. Exp. Med., 174, 761—767.WEB, AL & LONGNECKER, D.E. (1992). Inhibition of endotheliumderived relaxing factor-dependent circulatory control in intactrats. Am. J. PhysioL, 262, H1494-H1500.LOWENSTEIN, CJ., GLATr. CS.. BREDT, D.S & SNYDER, S.H.(1992). Cloned and expressed macrophage nitric oxide synthasecontrasts with the brain enzyme. Proc. Nail. Acact Sc!. U.S.A.,89, 6711—6715.MAYER. B., JOHN, M.; HEINZEL. B.. WERNER. E.R.. WACHTER, H.,SCHULTZ, G. & BOHME. E. (1991). Brain nitric oxide synthase isa biopterin- and fiavin-containing multi-functional oxido-reductase. FEES Let:., 288, 187-191.MAYER. B.. SCHMIDT, K., HUMBERT, P. & BOHME, E. (1989). Biosynthesis of endothelium-derived relaxing factor. A cytosolicenzyme in porcine aortic endothelial cell Ca’-dependently converts L-arginine into an activator of soluble guanylate cyclase.Biochem. Biophys. Rex. Commun., 164, 678-687.MCCALL, T. & VALLANCE. P. (1992). Nitn oxide takes centre-stagewith newly defined roles. Trends Pharmacol. Sc!., 13, 1—6.MONCADA, S.. PALMER, R.MJ. & HIGOS. E. (1991). Nitric oxidphysiology, pathophysiology and pharmacology. Pharmacol.Rev., 43, 109-142.O’DONNELL, V.B., TEW, 0G., JONES. O.T.G. & ENGLAND, P.1.(1993). Studies on the inhibitory mechanism of iodonium compounds with special reference to neutrophil NADPH oxidase.Biochem. 1., 290, 41—49.PANG. C.C.Y. & WANG. Y.-X. (1993). Differential in vivo and in vitrovascular effects of methylene blue (MB) and N°.nitro-L-arginine(L-NNA). Pharmacologist, 35, 165.PAGAN, C. & BLOXHAM, D.P. (1977). Specific labeling of a constituent polypeptide of bovine heart mitochondril reduced nicotinamide-adenine dinucleotide-ubiquinone reductase by the inhibitordiphenylenciodonium. Biochem. .1., 163, 605—615.REES. 0.0., PALMER. R.M.J. & MONCADA. S. (1989). Role of endothelium-derived nitric oxide in the regulation of blood pressure.Proc. Nat!. Acad. Sc!. U.S.A.. 86, 3375—3378.REES. 0.0.. PALMER. R.MJ.. SCHULZ. R.. HODSON. H.F. & MON.CADA. S. (1990). Characterization of three inhibitors ofendothelial nitric oxide synthesis in vitro and in vivo. Br. J.Pharmacol., 101, 746—752.STEWART. G.A. & HANLEY. T. (1969). Other oral hypoglycemicagents. In Oral Hypoglycemia Agents: Pharmacology mid Therapeutics. ed. Campbell, G.D. pp. 398—407. London and NewYork: Academic Press.STUEHR. Di., CHO, H.J., KWON, N.S., WEISE, M.F. & NATHAN. C.F.(l991a). Purification and characterization of the cytokine-inducedmacrophage nitric oxide synthase: an FAD- and FMN-containingfiavoprotein. Proc. Nail. Acad. Sd. U.S.A., 88, 7773—7777.STUEHR, Di., FASEHUM. O.A., KWON, N.S.. GROSS, S.S.. FONZALEZ, l.A., LEVI, R. & NATHAN. C.F. (l991b). Inhibition of macrophage and endotheliaf cell nitric oxide synthase by diphenyleneiodonium and its analogs. FASEB J., 5, 98—103.STUEHR. Di., FASEHUM, O.A., KWON. N.S.. GROSS, S.S., THIEL.BA., LEVI, P. & NATHAN. C.F. (1989). Synthesis of nitrogenoxides from L-argnne by macrophage cytosol: requirement forinducible and constitutive components. Biochem. Biophys. Rex.Co,nnum., 161, 420—426.STUEHR, Di., FASEHUM, O.A.. KWON. N.S. & NATHAN. C.F. (1990).FAD and GSH participate in macrophage synthesis of nitricoxide. Biochem. Bwphys. Rex. Commun., 168, 558-565.TABRIZCH1, R. & TR.IGGLE. C.R. (1992). Actions of L- and Darginine and N°.monomethyl-L-arginine on the blood pressure ofpithed normotenaive and spontaneously hypertensive rats. Cim.Exp. Hyper., A14, 527—546.WANG, Y.-X. & PANG, C.C.Y. (1990). Pressor effect of N°-nitro-Larginine in pentobarbital-anesthetized rats. Life Sc!., 47, 2217—2224.WANG, Y.-X. & PANG. C.C.Y. (1991). Possible dependence of pressorand heart rate effects of N°-nitro-L-arginine on autonornic nerveactivity. Br. J. Pharmacol., 103, 2004-2008.WANG, Y.-X. & PANG, C.C.Y. (1993a). Functional integrity of thecentral and sympathetic nervous systems is a prerequisite forpressor and tachycardic effects of diphenyieneiodonium, a novelinhibitor of nitric oxide synthase. .1. Pha,’macol. Exp. Ther., 265,263—272.WANG. Y.-X. & PANG. C.C.Y. (1993b). Halochane inhibits preasoreffects of diphenyleneiodonium. Br. J. PharmacoL, 109, 1186-1191.WANG. Y.-X.. POON. Cl. & PANG. C.C.Y. (1993). In vitro and cx vivoinhibitory effects of L and D enantiorners of N°-nitro.arginine onendothelium-dependent relaxation of rat aorta. I. Pharmacol.Exp. Ther.. 265, 112—119.WANG, Y..X.. POON. CI.. YAU, CCV. & PANG. C.C.Y. (1992). The invitro and in vivo vasoconstrjctc,r effects of C and 0 enantiomersof N°-nitro-arginine methyl ester (NAME). Can. Fed. BioL Soc.,35, 80.WANG. Y..x.. ZHOU. T. & PANG. C.C.Y. (1991). Pressor effects of Cand 0 enantiomers of N°-nitro.arginine in conscious rats areantagonized by C- but not D-arginine. Eur. J. Pharmacol., 200,77—81.WHITE. K.A. & MARLETTA. MA. (1992). Nitric oxide synthase is acytochrome P.450 type hemoprotein. Biochemistry. 31. 6627—663 1.YUI. V.. HATTORI. R.. KOSUGA. K.. EIZAWA. H.. HIKI. K.. OHKAWAS. S.. OHNISHI. K.. TERAO. S. & KAWAI. C. (1991). Calmodulinindependent nitric oxide synthase from rat polymorphonuclearneutrophils. J. Biol. Chem., 266, 3369—3371.(Received May 11, 1993Revised July 16. 1993Accepted July _‘O. 1993)185Paper VIIIWang, Y.-X., Poon, C.l., Yau, L.C.Y. and Pang, C.C.Y.: Vascularpharmacodynamics of NGnitroLarginine methyl ester in vitro and in vivo. J.Pharmacol. Exp. Ther. 267: 1091-1099, 1993. The reproduction of this paperwas kindly permitted by the copyright holder, Willams & Wilkins, Baltimore, U.S.A.QD-36€6j93/2673- 1091503.5010rw. JouesM. , PMAa CO9.DGY AND gx,esuiaiisi.Ce,,flihe C t5 by The Acee S.e.sy fPe.e5 Zipiie.a1eVascular Pharmacodynamics of NGNitroLArginine Methyl Esterin Vitro and in Viva1ABSTRACTThe itflbtory effects of N° tro-c-arginine meth’jl ester (LNAME) on endothelium-dependent vasodllatation were studiedin conscious rats and isolated rat aoctic rings. In ptsenylephrine(PE, ED,o)-preconsthcted aortae, c-NAME caused prolonged andcomplete inhibition of acetylcholine (ACh)-induced relaxation withIC50 of 4 x 1Q’ M and HiH coefficient (n) of 1. The inhibition wasabolished by L-arginine (c-Arg), independently of whether it wasapplied 10 mm earlier or 4 hr later than c-NAME. Intravenousbolus injection of c-NAME caused prolonged increases in meanarterial pressure (MAP), with E, of 50 ± 7 mm 1-4g. ED50 of 5 ±1 x 1 O mci/kg and n of 2. Intravenous infusion of c-Arg shiftedthe dose-MAP curve of c-NAME to the right without changingE. or ti. A modified Schild plot (n 2) for the action of c-NAMEgave a slope not different from unity, suggesting that L-Arginhibits competitively the MAP response of c-NAME. Intravenousinfusion of ACh decreased MAP in rats treated with c-NAME (4.8x 1 0 mol/kg) or PE. Compared to PE-treated rats, c-NAMEinhibited the depressor response to ACh by 50%. Thus, a doseof c-NAME 10 times its ED50 in raising MAP only partially blockedthe depressor responses to ACh. Our results are consistent withthe hypothesis that pressor response to c-NAME results fromthe inhibition of NO synthase only if 1) basal NO synthesis ismore effectively blocked by c-NAME than stimulated NO synthesis and 2) stimulated NO synthesis is inhibited by c-NAME in aone to one antagonist manner whereas basal NO synthesis isinhibited by c-NAME in a two to one antagonist manner.Analogs of N°-substituted Arg have been shown to suppressNO synthesis as well as to interfere with endothelium.dependent relaxations of vascular preparations (see Moncada et at.,1991). These inhibitors, including c-NMMA (Palmer et at.,1988a,b; Rees et at., 1989a, 1990). c-NAME (Rees et at.. 1990),c-Nb (Rees et at., 1990), c-NNA (Ishii et at., 1990; Mülschand Busse, 1990) and c-NAA (Vargas et aL, 1991), cause prolonged pressor responses which are antagonized by the NOdonor sodium nitroprusside (Wang et aL, 1993c) and by L-Arg(but not by o-Arg) (Aisaka et at., 198gb; Rees et at., 1989b;Wang and Pang, 1990; Wang et at., 1991).It is generally assumed that NG.substituted Arg analogs arecompetitive inhibitors of NO synthase because they are structurally related to c-Arg and because their actions are antagonized by c-Arg. However, there is little pharmacodynamicevidence to establish such a mechanism, particularly in in vivostudies. Moreover, whereas N°-substituted Arg analogs effectively inhibit ACh-induced vasodilatations in isolated preconReceived for publication March 15, 1993.Thu work wee supported by • grant from the Medical Reseaich Council ofCanada (MRC),a MRC Poetdoctoral Feilowsbip awarded to Y.-X. Wang and aCanadian Hypertenelon Society Summer Research Studentahip awarded to C. I.Poo0.stricted vascular preparations (Palmer et at.. 1988b; Crawley etat.. 1990; Rees et at., 1990: Wang et at.. 1993b), or regionalvascular beds, e.g.. coronary (Woodman and Dusting, 1991).hindquarter (Bellan et at., 1991), renal (Gardiner et at., 1990a.1991: Lahera Ct at.. 1990), pulmonary (Fineman et aL. 1991).mesentery (Fortes et at., 1990; Gardiner et at., 1990a) andcarotid (Gardiner et aL, 1990a, 1991) beds, there are discrepancies in reports of their abilities to interfere with the depressoreffects of ACh in whole animals. Inhibition of ACh-induceddepressor responses has been shown for L-NMMA (Whittle etat., 1989; Rees et at., 1990), c-NAME (Rees et at., 1990) and LNIO (Rees et at., 1990). However, an absence of inhibition, oreven potentiation, of depressor responses to ACh has beenreported with c-NMMA (Aisaka et at., 1989a; Yamazaki andNagao, 1991), c-NNA (Wang et aL, 1992) and c-NAME (Gardiner et at., 1990a, 1991; van Gelderen et at., 1991). It istherefore unclear whether the pressor responses of these compounds are secondary to the inhibition of endogenous endothelial NO synthesis and release.To resolve the above questions, we compared the in viva andin vitro activities of c-NAME. L-NAME was selected from NG.substituted Arg analogs due to its solubility in water, low costand high potency. In vitro, the concentration-relaxant re186Vet 311. Ne. 3a U.S.A.YONG-XIANG WANG, CHRISTiNA I. POON and CATHERINE C. V. PANGDepartment of Pharmacology and Therapeutics. Faculty of Medicine, The University of British Columbia, Vancouver, B.C., CanadaAccepted for publication August 9, 1993ABBREVIATIONS: kg. arginine; NO, nitric oxide; L-NMMA, N°-monomethyt-l.-arginlfle; c-NAME, N°-nitro-c-arginine rneth1 ester c-NlO, ctfl’Inoethyc-ornithine; c-NNA,5-nitro-c-arginane; c-NAA. amino-c-arginine; ACh, acetyicholine; SNP, sodium nitroprusside; PE..phenylephrlne;MA mean arted pressure.111092 wang.tef.spouses of ACh and SNP were determined in isolated endothelium-intact aortic rings, in the presence of varied concentrations of L-NAME. In uiuo, conscious and unrestrained rats wereused to examine the effects of L-NAME on blood prewe anddepressor responses to ACh and SNP.MethodsSurgery and preparations. Sprague-Dawley rats (from The Urnvarsity of British Columbia) (300-400 g) were used for all experiments.Each group consisted of five to six rats in uwo or five to six sortie ringsin vitro, with each ring derived from a different rat.In in vitro studies, the rats were sacrificed by a blow on the headfollowed by exsanguination. The thoracic aorta was removed andcleared of connective tissues. Four ring segments of 0.5cm length wereprepared from one aorta and suspended in separate organ baths. Eachring was connected to a Grass (Quincy, MA) Fr.03-C force-displacement transducer for isometric recording with a preload 0(1 g and wasequilibrated for 1 hr (with three washouta) in normal Krebs’ solution(pH 7.4) at 37’C with a gas mixture of 95% O,-5% CO2. Krebe’ solutionhad the following composition (10’ M): NaCI, 118; glucose, 11; KC1,4.7; CaC15,2.5; NsHCO1,25; KH2PO4,L2 and MgCI,6H20 1.2.The rings were incubated with the vehicle or drugs (see later).Afterwards, 10 M PE (EdO, data not shown) was added to the bathsand 15 to 20 miii later, at the steady-state phase of the contractileresponse to PE, a cumulative concentration-response curve of ACh(10 to 3 x 10 M) or SNP (10 to 10’ M) was constructed by usingsequential doubling of drug concentration. Each concentration of AChor SNP was left in the bath for 1 to 3 mm to ensure a plateau response.The time taken to complete each concentration-response curve wasapproximately 20 miii. In some groups in which the concentration-response curve of ACh was conducted more than once, the preparationswere washed 3 times in 30 mm and given another 30 mm to recoverfrom the effects of PE and ACh, before constructing the next concentration-response curve.In the in uiuo studies, the rats were anesthetired with halothane (4%in air for induction and 1.5% in sir for surgery). A polyethylene cannula(PE-50) was inserted into the left iliac artery for the measurement ofMAP by a pressure transducer (P23DB. Gould Statham, Oxnard, CA).PE-50 cannulae were also inserted into both iliac veins for the administration of the vehicle or drugs. The rats were then put into a smallcage and allowed to move freely before as well as during the experiments. The rats were given > 6 hr to recover from the effect ofanesthesia before use.Drugs. L-NAME, ACh hydrochloride, PE hydrochloride, c. and 0-Arg hydrochloride were obtained from Sigma Chemical Co (St. Louis,MO). SN? was obtained from Fisher Scientific Co. (Springfield, NJ).All drugs were dissolved iii 0.9% saline (vehicle) except for 1.- and oArg, which were dissolved in distilled water and pH was adjusted to 7.0with NaOH solution.Experimental protocols. 1) Concentration-response curves of I.NAME on relaxation responses of ACh and SN? were studied in ninegroups of sortic rings. The first seven groups of aortae were incubatedwith vehicle or L-NAME (10 to 3.2 x 10 M), followed 10 miii laterby the administration of PE (10 M) and determination of concentration-relaxation response curves for ACh (10’ to 3 x 10’ M). Only oneconcentration of L-NAME was used in each group. In another twogroups of aortae pretreated with 1.-NAME or vehicle, SN? (10 to 10M) was used instead of ACh and L-NAME was used at only oneconcentration (32 x 10 M).2) The time course of the inhibition by t.-NAME of the relaxationresponse to ACh wu studied in two groups of aortic rings. Aftercompleting the first concentration-response curve of ACh in the presence of vehicle or t.-NAME (3.2 x iO- M), the preparations werewashed without further addition of L-NAME. The 2nd, 3rd and 4thcurves of ACh were repeated at 1.5,4 and 9 hr later.3) The effects oft.- and D-Arg on the inhibitory effects of t.-NAMEon ACh-induced relaxation were studied in six groups of aoruc rings.V& 267In the first four gzvupa, the concentration-resPonse curves for ACh(with 10’ M PE) were determined in the presence of vehicle + vehicle,vehicle + t-NAME (3.2 X 10’ M), L-AIg (3 X 10 M) + t.-NAME(3.2 x 10’ M) and o-Arg (3 x 10 M) + t.-NAME (3.2 x 10 M),with the first treatments applied 10 ruin before L-NAME (“pre-exposure”). The other two groups were first incubated with t.-NAME (3-2X 10 M) for 4 hr before adding vehicle or L-Arg (3 x 10 M)(“postexpoaure”); PE was added 10 ruin later, followed by determination of the concentration-response curve of ACh.4) The time course of the preisor response to L-NAME was studiedin two group. of conscious rita. The rats were uv. bolus injected withvehicle (1 mi/kg) or t.-NAME (4.8 X 10’ mol/kg), and MAP wasrecorded continuously for 2 hr.5) Dose-pressor response curves for i.-NAME in the abeence orpresence of 1.- or D-Arg were obtained in five groups of rats, Lv. infusedcontinuously with vehicle (0.025 mI/kg/ruin), L-Arg (1.2, 2.4 and 4.8 X10’ mol/kg/min) or D-Arg (4.8 x 10’ mol/kzJmin). Cumulative dosesof L-NAME (1.5 x 1W’ to 4.8 x 1W’ mol/kg) were Lv. bolus injectedbeginning 10 miii after starting the infusions of vehicle, L- or D-Arg atdose-intervals of 10 to 15 ruin, the time required to obtain plateaueffect.6) The effects of t-NAME on the magnitude and duration of depressor responses to LV. infusions or Lv. bolua injections of ACh orSN? were studied in seven groups of rats. The first three groups weretreated with vehicle (1 mi/kg i_v. bolus injection), PE (2 x 1W’ mol/kg/ruin, Lv. infused continuously) or t.-NAME (4.8 X 1W’ molfkg, Lv.bolus injection) followed 20 miii later by Lv. infusions of ACh (5.5 x1W’ to 8.8 X 1O mol/kg/min) or SNP (4 x 1W’ to 5.6 X 1O mol/kgJruin). Each dose of ACh or SN? was infused for 4 miii before the nexthigher dose. The sequence of administrations of ACh and SN? wasreversed in one-half of the rats in each group with an interval of 20ruin. The other four groups were treated with vehicle (1 mI/kg i.v. bolusinjection). L-Arg (4.8 x 1W’ mol/kg/mun, i.v. infusion), PE (2 x 1W’mol/kg/mun. i.v. infusion) or L-NAME (4.8 X 1W’ mol/kg, i.v. bolusinjection). At 20 mm later. ACh (3 x 1W’° to 8.8 x 1W’ mol/kg) andSN? (1.5 x 1W’ to 4.4 x 1W’ mol/kg) were i.v. bolus injected intothese rats, at intervals between injections of 2 to 5 miii. The sequenceof administrations of ACh and SNP was also reversed in one-half ofthe rats in esch group. To fsciitate the measurement of the time courseof the transient depressor responses to ACh and SNP, the chart wassped up from 10 mm/mu to 10 mm/sec.Calculation and Statistical AnalysisAnalysis of dose (concentration)-response curves. To determine the parameters (ED10 or EC10, maximum and Hill coecient, a)of dose-response or concentration-response curves, values of response(R, rise in MAP or relaxation in the aortae) were fitted (by using anIBM-compatible microcomputer) by nonlinear least-squares curves tothe relation R = a + bx, where x = [D]”I(ED,O + (D]’) or x = (C]”!(ECOO + [C]”) (see Kenakin, 1987), to give the value of ED,0 or Ed10and b (maximum response) yielding a minimum residual sum of squaresof deviations from the theoretical curve. The detailed analysis waspublished elsewhere (Wang and Pang, 1993a).Modified Schild plots. In principle, a competitive inhibitor (I)shifts the dose-response curve to an agonist (A] in parallel to the right.with the ED10 increasing by the factor (1 + [I]/Ki when there isa one-to-one competition of I with A on the receptor. This arises from thewell-known relationship: [A.R]I[P.,] (A]/)[AJ + K.(1 + [fl/Kj] where(AR]/(R,J is the fraction of receptors occupied by A and K. and R, arethe dissociation constants for A and I, respectively. It can readily beshown that the same relationship holds when A is an antagonist and Iis an agonist.By using the standard assumption that equal responses, with andwithout I, represent equal (AR]/[R,], it follows that K.(1 + [11/Kif[AJ,- K.J[AJ., where (A]1 is the (A] that gives the same response in thepresence 0(1 as did (A]. in the absence of I. Therefore “dose-ratios”(1A),/(A].. which are the same as ratios of ED10 values), followe theequation: DR = dose ratio IAIJIAI. 1 + (I]/K. and, therefore, log187;993DR — U Log [fj-1og K.. That is. the Schild plot.’ a graph of logDR— 1)u,. log(1I,haeofurntyandlCisgivezibythe(extrapolated or interpolated) valu, of [11 at DR -2.However, if the response depends upon a form of the receptor boundtO more than one (ni) molecule of agOniSt WhSres5 more than one (n,JoIecule of! ii necessary to block action of the receptor, th. aboveequation does not hold. Instead, the equations become, at the simplest(assuming high poeltive cooperation in binding of A to R)-.[A”Rl/[R. I — [AIu/l(AI + K.’1(1 + (1/fl— (dose r.tio)i = ((AJI/[AJ.)’ -1+ (1]’/K- [iI1GThus, one must plot log (DRi1) us. n1 log fI} to obtain a slope ofunity (see Kenskin, 1987).Expression of data. The relaxant response of ACh or SNP in vitrowas calculated u percentage of contractile response to PE and thedecrease in MAP response to ACh or SNP in uwo was calculated aspercentage of base-line MAP. The duration of each depressor responseto ACh or SNP in owe was expressed sa ‘modified half-recovery time’(minutes per millimeter of mercury), which was obtained by dividinghalf-recovery time by base-Line MAP end then multiplying by 10’. Allresults were expressed as mean ± S.E and analyzed by analysis ofvariance followed by Duncan’s multiple range test, with P <.05 selectedas the criterion for statistical significance.ResultsConcentration-response of i-NAME on relaxationsevoked by ACh and SNP. PE (10 M) caused an averagecontraction of 0.84 ± 0.09 g in two groups of control aorticrings to be treated with ACh or SN?. PE-induced contractionswere significantly increased by i-NAME (10 to 3.2 x 10’ M)to a pooled value of 1.05 ± 0.10 g, with potentiating effectssimilar for all doses of i-NAME. ACh caused concentration-dependent relaxation of PE-preconstricted aortae (fig. IA).Preincubation with i-NAME concentration-dependently inhibited ACh-induced relaxation, with equal effect at all concentrations of ACh (fig. 1A). Analysis of the inhibition-concentration curve for t.-NAME gave a best-fitting a of 1.28 (notsignificantly different from 1.0) with maximal inhibition of87% and 1C of 3.5 x i0 M (pD = 6.5). With n set at 1, themaximum inhibition was 96% and IC was 4.3 X iO M (pD26.4). As shown in figure 2A, the theoretical curve with rt of 1fits the observed data very much better than the theoreticalcurve with a = 2. Correspondingly, a Hill plot of the inhibitionby i-NAME of ACh-induced relaxations had a slope of 0.97 ±0.06 (fig. 2B).SN? also caused concentration-dependent relaxations of PEpreconstricted aortic rings with maximal relaxation of approximately 100%; relaxations were not significantly altered by 3.2X 10’ M i-NAME (fig. 1B).Time course of inhibitory effect of i-NAME on AChevoked relaxation of aortic rings. Repeated PE-inducedcontractions were consistent at 1.5,4 and 9 hr after a first test(0.99 ± 0.10 g), with contractions averaging 1.07 ± 0.09, L14 ±0.13 and 0.81 ± 0.07 g, respectively. On the other hand, thepotentiation of PE-induced contraction by 3.2 x 10 M LNAME (to 1.42 ± 0.15 g at the first test) slowly declined,contractions at 1.5,4 and 9 hr after the first test averaged 0.95± 0.12 (P < .05), OS? ± 0.06 (P < .05) and 0.63 ±0.06 g (P <.06), respectively. The concentration-relaxation responsecurves of ACh in aortic rings were repeated 4 times within 9 hrUt the presence of vehicle or i-NAME. In the controls, theL—NAUE (10-4 )00• 0.10.2A 0.4C 0.8• 1.8o .2—8 —7 —eSodhim Mropna.ede (Id), LoFIg. 1. Effects of L-NAME on ACh (A)- and SNP (8)-induced relaxationsin PE (10 M)-preconstricted sortie rings (n — 5-6 each group).FIg. 2. Wildon by L-NAME (10-’ to 3.2 x i0 U) of PCl (3 x 10 to3 x 10-’ M).evoked relaxation. The data were the average vakies fromlive to sIx aortic rgs. A, Ccncentrallon-tesponse CL’e Of LME. ThetheoretiCel curves we plotted with I coeflident n1 of 1 end 2 (seeMethoda’ for detads). B, I’ Plot of the concentration-response curveof L.NAME. E end E, the relaxations at 3 x 10-’ to 3 X 10-’ U ACh iithe ,eenCe or presence of dffenstt concentrations of t..NAME.188Veicular Effects of i-NAME 1093806040CII0 VehIcle• L—NAIdE (3.2xiO8U)‘-9 —8 —7 -4 —5 —4Met4chone (Id), Lag8100608040200_,f—a—5Ah (104 U03• 1030A100•C3.80.60‘C0c 400.02001.00.5r.iS—0.5—1.0-8r—O.97 (P(0.05)Stope—0J7±0.06—7—4 —5L-NAs (U), Lag1094 wang.taI.relaxation-concentration curves of ACh were slightly but time-dependently shifted to the right (fig. 3A). L-NAME (3.2 XM) almost abolished ACh-induced relaxation completely. Theinhibitory effect of the single exposure to L-NAME declinedgradually at 1.5 to 9 hr aster the preparations were washed outbetween teats (fig. 3B).Effects of pre- and postexposure of L- and D-Arg onthe inhibitory effects of t-NAME. In these groups of aorticrings, PE caused contractions of 1.07 ± 0.09 and 1.22 ± 0.10 gin the presence of vehicle + vehicle and vehicle + L-NAME(3.2 x 10 M), respectively. Pre-exposure (10 mm earlier) toL-Arg (but not o-Arg) insignificantly reduced PE-caused contraction to 0.84 ± 0.17 and 0.98 ± 0.12 g, respectively. With LArg (3 x 10’ M), but not D.Arg, the effect of L-NAE (3.2 x10 M) to abolish ACh-induced relaxation was prevented totally (fig. 4A).In another series, L-Arg was applied after L-NAME. Alter a4-hr incubation of aortic rings with 3.2 x 10 M L-NAME,PEcaused an average contraction of 1.22 ± 0.10 g. Addition of LArg (3 x 10’ M) (postexposure) reduced the mean PE-inducedcontraction to 0.85 ± 0.13 g (P < .05), and restored totally therelaxation response of ACh (fig. 4B).Pressor response to L-NAME in uivo. Base-line valuesof MAP in conscious rats treated with vehicle or L-NAME werenot significantly different from each other (106 ± 3 and 112 ±3 mm Hg, respectively). Intravenous bolus injections of LNAME (4.8 x i0 mol/kg), but not the vehicle, caused a slow-developing and prolonged increase in MAP. Plateau MAPresponse to L-NAME, which was attained approximately 10mm alter injection, with half-rise at 2.3 ± 0.4 mm, lasted atleast 2 hr (fig. 5A).—8 —7 —6Acat4cholin. (U). LogFig. 4. Effects (mean ± S.!.) of pre- O postexposure with L- (3 x 10’M) or D-Arg (3 x 10’ M) on inhibitory effect of L-NAME (3.2 x 10 M)on ACh-induced relaxation in PE (10 M)-preconstricted aortic rings (n5-6 each group). A. pre-exposure (10 mm eallier) with vehicle. L- Or 0-Arg. B. postexposure (4 hr later) with vehicle or L-Arg.n—i—s —a —7 —6 —5 —4 7—6 —5 —4Acatyldidini (‘0. Log L—NAUE (mci/kg). LogFIg. 3. Trne course of the effects (mean ± S.E.) of vehicle (A) and 1.- FIg. 5. Twne course and cumulative dose-response curves of the MAPNAME (32 x 10’ M. B) on ACh-induced relaxation in PE (10 M)- response (mean ± S.!.) to iv. bolus infections of L-NAME in consciousprOconstricted rat aortlc rings (n — 5-6 each group). The various times rats (n —5 each group). A. the time course of vehicle or L-NAME (4.8 xrepresent the times after the preparations were washed without further 10 mol/kg). B. cumi.iative dose-response of L-NAME. Theoretical resadding the vehicle or L-NAME. were drawn by using n of 1.2 and 3 (see ‘Methods for detads).189Vol. 267o Vehid.+v.hicl.• V.hld.+l..—NAMEL—kg+L—NAMEA 0—kg+L—NAMEA10080C0eo840-200—10080C0so.8§40200——gBo L—NsWE+v.hIcl.• l.—NAE+L—Arg—5 .4A100806040After wosheuts (h)00• 1.5ASC0Q0C00C0IA60 0—0 Vehici.•—• L-NAME40____200—20—40 0 40 80 120Tim. (mm)‘V100806040zEE120eo2001993In another group of rats, with base-line MAP of 106±4 mmHg (table 1), i.v. bolus injections of cumulative doses of LNAME increased MAP in a dose-dependent manner (fig. 5B).Analysis of dose-response curves with n “floating” gave averagebest-fitting n = 2.01 ± 0.3Z for in no individual curve was nsignificantly different from 2 and it is evident that the theoretical curve with n of 2 but not 1 or 3 fitted the observed databest. By using floating n for each individual curve, averagevalues of ED and E for the pressor effects of L-NAMEwere 5.0 ± 1.1 x 10’ inolfkg and 50±7 mm Hg, respectively.The calculated E. values with n 2 for each curve were notsignificantly different from observed E data (table 2).Effects of L- or n-Arg on the pressor dose-responsecurve of L-NAME. Before and 10 mm after the start ofinfusion of vehicle, L- or o-Arg, mean values of MAP were notsignificantly different from each other (table 1). intravenousinfusion of c- but not D-Arg dose-dependently shifted thepressor dose-response curve of c-NAME to the riglt, withoutsignificantly changing E,.. or n1 (fig. 6A table 2). The theoTABLE 1Values (mean ± S.E.) of MAP before (a) and 10 (b), 20 (c) or 100(d) mlii after i.v. administrations of vehicle, L.Arg (1.2 to 4.8 x 10’mol/kg/mln), o-Arg (4.8 x 10’ mol/kg/min), PE (2 x 10’ mol/kgfmm) or L-NAME, 4.8 x 10’ mci/kg) in conscious rats (n = 5—6 eachgroup)WP(nwfl)Treaimss Dnea b C dfkgyffr4t/raiVehicle 107±5 106±4L.Arg 1.2 x l0 103 ± 6 102 ± 7t-Arg 2.4x10 114±2 114±3i-Arg 4.8 x 1O 105 ± 5 106 ± 5o-Arg 4.8x105 112±6 112±6Vehicle 115±6 124±9PE 2x10’ 109±4 160±4’L-NAME 4.8 x 10 112 ± 5 160 ± 5•Vehicle 108±5 111±5 108±4t.-kg 4.8x105 113±4 115±4 110±3PE 2x10 103±4 162±5 156±3L-NAME 4.8 x 10 113 ± 4 159 ± 5 156 ± 4‘ Signthcant difference from base-line MAP (P < .05).TABLE 2Vascular Effects of L44AME1901095retical curves in figure 6A were plotted by using n1 as 2 andindeed fitted the observed data well. The modified Schild plotsin figure 6B (see “Methods”) for the apparently competitiveblock by L-Arg were drawn by using, as each dose-ratio, theED value in the presence of c-Arg divided by the EDu in theabsence of L-Arg. Plots were essentially linear whether n1 waschosen as 1, 2 or 3 when n7 was chosen as 1. However, with n1= 2 the slope was 1.17 ± 0.16, whereas with n chosen as 1 or3 the slope was 0.68 ± 0.1 or 1.71 ± 0.22, respectively, bothsignificantly different from 1.0. On the other hand, if the plotswere to use log (DR—i) (Le, n1 = 1) against n- * log (c-Arg),where n2 was chosen as 1, 2 and 3, they were also essentiallylinear, but the slopes were 0.66 ± 0.14, 0.33 ± 0.07 and 0.22 ±0.05, respectively. Thus, the data fit a model in which 1 molecule (n,) of L-Arg competes effectively with 2 molecules (is1) ofL-NAME. The calculated half-blocking infusion dose for L-Argto antagonize the pressor response to c-NAME was at a rateof 10’ aol/kg/mm (with is1 = 2).Effects of c-NAME on the depressor responses to AChand SNP. In conscious rats, i.v. infusions of ACh or SN?dose-dependently decreased MAP (expressed as percentage ofbase-line MAP); infusion of PE (2 x 10’ aol/kg/mm), whichraised MAP to the same extent as a single bolus injection of4.8 x 10 aol/kg c-NAME (table 1), enhanced the depressorresponses to various doses of ACh and SNP by average valuesof 140 ± 53 and 94 ± 31%, respectively, with greater potentiation at low doses of ACh and SNP (fig. 7, A and B). L-NAMEinhibited the depressor response to ACh markedly, by 53 ± 4%,when the ACh-responses were compared to those in rats treatedwith PE (fig. 7A). On the other hand, c-NAME potentiated thedepressor responses to all the doses of SNP, by an average of142 ± 53%; at the lowest two doses of SNP, the potentiationwas significantly (P < .05) greater than with PE infusion whichgave the same increase in base-line MAP (fig. 7B).In another four groups of rats, i.v. bolus injections of AChwere given (fig. 8). The first three doses of ACh caused significantly less reductions of MAP in rats treated with L-Arg (4.8x 10 mol/kg/rnin) than in control rats given vehicle (fig. 8A).In animals given PE (2 x lO mol/kg/min) or L-NAME (4.8x i0 mol/kg), the magnitudes of depressor responses to AChValues (mean ± S.E.) of Hil! coefficient (n), EDw and E,,,, calculated from by best fitting or by specific n, as well as observed E,,.,, of1.-NAME in the presence of vehicle, i-Arg and o-Arg in conscious rats (n = 5 each group)c-Obsewd E.,.,s ED,,1Oflfk5 I?WflH9Best-fittedVehicle 2.0±0.3 5.0± 1.1 49.7±6.7 50.0±5.8L-Arg (1 0 md/kg/mm)1.2 1.8 ± 0.3 19.4 ± 2.8 52.0 ± 2.8 52.5 ± 2.62.4 2.0 ± 02 32.0 ± 5.0’ 54.6 ± 2.2 53.0 ± 1.34.8 2.5 ± 0.3 40.9 ± 7.3 52.1 ± 1.9 52.5 ± 2.0n-kg (10 md/kg/mm)4.8 2.2 ± 0.3 4.1 ± 0.5 49.7 ± 2.9 49.5 ± 2.6Specified. n 2Vehicle 2 4.7 ± 1.1 48.4 ± 6.4 50.0 ± 5.81.-kg (10 md/kg/nm)1.2 2 1.92±2.7’ 51.1 ±2.7 52.5±2.62.4 2 311 ± 4.6’ 53.7 ± 1.3 53.0 ± 1.34.8 2 422 ± 8.4’ 52.6 ± 2.1 52.5 ± 2.0c-kg (10’ md/kg1mm)4.8 2 4.1 ± 0.5 49.4 ± 2.0 49.5 ± 2.6Significant fferenca from velicle-eeated groups (P < .05).—7 —8 —5 —4 —3L.—Arg (mel/kg/mm). LegFig. 8. A. effects (mean ± S.E.) of iv. infusions of i. and o-Arg on theMAP response to i.v. bolus injections of I.-NAME in conscious rats (n =5 each group). B. modified Schild plot (see Methods for details) for i.Arg to antagonize I-NAME (i.v. bolus injections). The tries were drawnwith ii of 1. 2 or 3.were not significantly different from those in control rats.However, the duration (expressed as modified half-recoverytime) of the (transient) responses to ACh was shortened by LNAME (fig. 8B), by 23 ± 3% when the value was compared tothat in vehicle-treated rats (P < .05 for the last 2 doses of ACh)and by 26 ± 3% when compared to that in PE-treated rats,with significance at the last 4 doses of ACh.In contrast, in the groups of rats treated in the same way butwith bolus injections of SNP to cause transient depressions ofMAP, responses were somewhat potentiated by t-NAME (fig.8, C and 0), with significance at the lower two doses of SNPeven by comparison with PE infusion (which increase responsesslightly arid nonsignificantly). A decrease in the magnitude ofMAP response to L-Arg, similar to that seen with ACh, wassignificant (P <.05) only at the third dose of SNP. The durationof depressor response to SNP was unaltered by PE, somewhatprolonged by L-Arg (P < .05 at the second dose of SNP) andsomewhat prolonged by L-NAME at the lower doses of SNP.DiscussionThere is a large body of information, which is derived primarily from in vitro studies, to support the concept that AChproduces vascular relaxation via NO biosynthesis (e.g., PalmerSt aL, 1988b; Crawley et aL, 1990 Bees et aL, 1990). Whereas itis generally shown that the ACh-induced relaxation responsein vitro is inhibited completely by N°-substituted Aig derivatives, there are uncertainties wit1 respect to the efficacies ofthese compounds in suppressing ACh-induced vasodilatationin vivo (see Introductory section). Inconsistent modifications(e.g., inhibition, no effect or even potentiation) of ACh-inducedA Ac.tyiciiofln. (10—8 mom/kg/mimi)5.5 11 22 44 88Fig. 7. Effects (mean ± S.E.) of i.V. infusions of ACh (5.5 x 10 to 8.8x 10’ mol/kg/min, 4 mm for each dose. A) and SNP (3.3 x 1O to 5.4x 10 molfkg/min. 4 mm for each dose. B) on the MAP (expressed aspercentage of base-line MAP) in conscious rats (n 6 each group)pretreated with vehicle. PE (2 x 10 mom/kg/mm) or I-NAME (4.8 x 10mom/kg). a. significant difference from vehicle-treated group: b. significantdifference from PE-treated group.depressor response by L-Arg analogs in uiuo cast doubts uponthe role of NO in ACh-induced vasodilatation in whole animals.The discrepancies in the effectiveness of NO synthase inhibitors in suppressing ACh-induced depressor responses are unlikely related to the species of animals used or the consciousoranesthetized state. Instead, the inconsistencies are most likelydue to comparison of ACh responses at different base-lineMAPs (in the absence or presence of N°-substituted L-Argderivatives), the different modes of administration (bolus injection us. infusion) of ACh as well as the method of expressionof data. Vasodilator drugs are known to cause greater hypotension at higher base-line MAPs (Bees et aL, 1990; van Gelderenet aL, 1991; Wang et aL, 1992; Chyu et at., 1992).Therefore, in this study PE was used as a base-line MAPcontrol for L-NAME and the depressor responses were calculated as percentage of base-line MAP and modified half-recovery time, in order to eliminate difficulties associated with thecomparison of responses at different base-line MAPs. Indeed,depressor responses to ACh and SNP were potentiated by 140and 95%, respectively, in PE-treated rats, as compared to thosein vehicle-treated rats. With the use of appropriate controls, itwas unequivocally clear that L-NAME interfered in part withthe depressor responses of ACh. The depressor response toinfused ACh was decreased by approximately 50%. The lack ofin uwo effectiveness of L-NAME in eliminating responses ofACh is not due to insufficient dose, since a supramaximalpressor dose (4.8 x 10 molfkg, 10 times ED for pressorresponse) of L-NAME was used. In our preliminary studies.even a dose of t.-NAME as high as 3.8 x l0 mol/kg in1096 wang .tai.EE191Vol. 267100L—NAIdE (md/kg). Log84O n—i, elop...O.68*0.10• n—2, .Iop.—1.17±0.16n—3, .lop.—i.7itO.2203= V.hicl.— Pt%.n)4.phnn.__L—NAIdEa8 SodIum nlboprus&d. (10-8 mom/kg/mm)I 93Fig. & Effects (mean ± S.E.) of .v. bolus injections of ACh (3 x 1O0 to8.8 x 10 md/kg) or SNP (1.5 x 1O to 4.4 x 1 O md/kg) on theMAP (expi-essed as percentage of base-line MAP) and modified half-recovery time , conscious rats ( 6 each group) pretreated with L-Arg(4.8 x 1O md/kg/mm). vehicle (1 mI/kg). PE (2 x 10 mci/kg/mat), LNAME (4.8 x 10’ md/kg). a. s4gniflcant difference from vehicle-eatedgroup; b. significant difference from PE-treated group. Modified half-recovery time was normalized as half-recovery time x 1O divided bybase-line MAP.conscious rats only inhibited in part ACh-induced depressorresponse (a = 2, data not shown). Moreover, the incompleteinhibition of ACh-induced depressor response by L-NAME isunlikely due to the activation of nicotinic receptor by ACh,because ACh-induced activation of ganglionic nicotinic receptors are known to produce pressor responses. In addition, wefound that after effective ganglionic blockade with mecamylamme, L-NAME (3.8 x 10 mol/kg) still failed to inhibit completely ACh-induced fail in MAP in pentobarbital-anesthetizedrats (n = 6) (Y.-X. Wang, C. J. Poon and C. C. Y. Pang,unpublished data). Gardiner et aL. (19908) also reported similarfindings. These results suggest that there may be a differencebetween ACh-induced responses in conduit vessels (e.g., theaorta) and resistance blood vessels.Further analysis of the depressor response to bolus injectionsof ACh shows that L-NAMZE decreased the duration, but notthe magnitude, of the responses to ACh by 26%. These resultsare in accordance with those of Aisaka et at. (1989a) by usingL-NMMA. Aisaka et aL (1989a) also showed that exogenous LArg prolonged the duration of the depressor response to AChand suggested that the availability of L-Arg determined theduration of response to ACh. In contrast, we found that L-Arg,at 48 times its half-block dose to antagonize the pressor response of i.-NAME, did not affect the duration but reduced themagnitude of the depressor response to low doses of bolusinjected ACh. van Gelderen et aL (1991) also found that L-Argdid not reduce the duration of the ACh-induced depressorresponse in anesthetized rats. The mechanism by which L-ArgSuppressed ACh-induced depressor response is not known butit is likely nonspecific, because the dose of L-Arg used wasVascular Effects of I.-MAME1921097relatively high and L.-Arg also reduced the magnitudes of thedepressor responses to low doses of SNP.It is unclear why L-NAME shortened but L-Arg did notprolong the duration of the depressor response to bolus injectedACh. It has been suggested that endogenous concentration ofL-Arg is sufficient to saturate NO synthase (Rees at at., 1989b).This hypothesis may explain why L-NAME but not L-Arginfluenced the duration of the depressor response to AC1L Anadditional mechanism1besides NO biosynthesis, must be responsible for the establishment of the magnitude of the depressor response to bolus injected ACh. which was neither reducedby L-NAME nor prolonged by L-Arg. ACh is shown to releaseendotheium-derived hyperpolarizing factor in addition to endothelium-derived relaxing factor/NO (Chen et aL, 1988; seeSuzuki and Chen, 1990). Endotheium-derived hyperpolarizingfactor vasodilates some vascular preparations (Garland andMcPherson, 1992) and may contribute to the depressor response to ACh in uivo. Consistent with this hypothesis, it wasreported recently that the Ca-activated K-channel (whichleads to hyperpolarization) blocker charybdotoxin attenuatedthe depressor response to ACh in rats (Watkins at aL, 1993).It is likely that t.-NAME had caused a redistribution of bloodflow in the rats. N°-substituted L-Arg analogs have been shownto cause presaor response by increasing total peripheral resistance via systemic vasoconstriction (Gardiner at aL, 1990b,c;Wang at at., 1993a), but their degree of influence vary withdifferent beds (Wang at aL, 1993a). It was also shown that therenal, internal carotid, common carotid and mesenteric, butnot hindquarter vasodilator effects of ACh, were attenuated inpart by 1.-NAME (Gardiner et aL, 1990a, 1991). The varyingability of L-NAME to inhibit vasodilator responses of ACh indifferent beds suggests that part of the vasodilator response toACh is mediated via a mechanism insensitive to the inhibitionby L-NAME.A Hill coefficient In) of 1.3 for L-NAME-induced inhibitionof ACh-induced relaxation was derived from our in vitro results.This value is similar to the n of 1.2 for L-NNA (unpublishedresult) calculated from data of a previous study (Wang at aL.1993b). These results suggest that one molecule of L-NAME orL-NNA competes with one molecule of endogenous L-Arg toinhibit NO biosynthesis. Our in uivo results, on the other hand,show that the a for L-NAME to cause pressor response is 2.0and this value is not affected by L-Arg which caused a rightwarddisplacement of the dose-response curve of L-NAME. The afor t.-NAME in vivo is also consistent with the a of 2.2 and 2.3for L-NNA (unpublished results) calculated from the data ofintact and ganglionic-blocked conscious rats, respectively(Wang and Pang, 1991). For comparison, the a for the pressoreffects of angiotensin II is 1.0 (unpublished calculation fromWang et at., 1993c) and of diphenyleneiodonium, a novel inhibitor of NO synthase (Stuehr et at., 1991) is 3.3 in conscious rats(Wang and Pang, 1993b).The nature of the difference between a values for N°-substitoted Arg analogs obtained in vivo (pressor response) and invitro (inhibition of vascular relaxation) is not known, but maysuggest that the mechanism involved in raising MAP (in resistance blood vessels) is different from that in inhibitingvascular relaxation (in large arteries). It is possible that t.NAME raises MAP by an unknown mechanism (which, nevertheless, is L-Arg-reversible) distinct from the inhibition ofendogenous endothelial NO synthesis. This hypothesis is consistent with the observation that diphenyleneiodonium pro-I.?.. BI0.3 1.4 5.3 as 1.5 5 27 110 440.t4ctdM (10.1 ‘,iol)leJ Soim nepd.(104m/hJAcknowledgmentsThe authors are deeply indebted to Dr. David M. J. Quastel in this departmentfor his help in pharmacodynamic analyses of our data and Ma. Lorraine C. Y.Yau for her technical aaaistance.Reference.AISMCA, K.. Gxoss. S. S., Giun’rrn, 0. W. AND LgVT, R.: L.aiginine availabilitydetermin the duration of acetylcholine-induced systemic vasodilation in vwn.Biocbsm. Biophy,. Re.. Common. 163: 710-717. 1989e.AISARA, K.. Gxosa. S. S., Grciypms, 0. W. AND Lrvi, R. NG.methylalginine, aninhibitor of endotbelkim.derivsd nitric oxide synthesis, Is a potent presseragent in the guinea psg Does nitric oxide regulate blood pressure in Limo?Biochem. Biophys. Rae. Common. 160:881486. 1989b.BEU.AN. J. A.. M3NKES. Il. K., MCNusrts, 1). 3. AND KADOWITZ. P. 3.: N”.nitro-L-arginine selectively inhibits vsaodjlator responses to acetyicholine andbridykinin in cata. Am. 3. Plsysiol. 260: H1025-H1029. 1991.CHEN. G.. Suzwci. H. AND WESTON. A.: Acetylcholine releases endotheliuni.WAN?derived byperpolarixing factor and EDRJ from rat blood vs.eala. Br. 3. Phar.macol. 95: 1166—1174. 1988.Cwru. IC. Y.. Gum. P. ILm Roag, C.. Effect of N.oitro-L-arginine methylester on arterial prn. and on ,s.odilator and ,asoconstrictor r—--—Influence of initial vascular tone. But. J. Pbaxmscol. 212:169-164, 1992.Cocxa, T. M. AND ANGUS, 3. A. Evi4snc. that contractions of isolated arteriesby L.NMMA and NOLA are not du, to Inhibition of basal EDRF release. 3.Cazdiovasc. Pbarmacol. 17: Suppl. 3, S159-S164, 1991.CRAWLZY. D. B., LW. S. F.. EVANS. T. W. aim BaaN. P. 3.: Inhibitosy rd. ofandothelium-dcivsd relaxing factor in rat and human pulmonary arten. Br.3. PharmacoL 101: 166—170, 1990.PINZIgAN, 1. IL, HXYMANN, M. A. aim Soi,zi. S. J. NmtroL-inmaattenust.e endotb.lium.depend.nt pulmonary vesodilation in Iambi. Am. JPhysloL 260: H1299-H1306, 1991.B. B., Ouvwia. hI. A., Savotarro. IL, CsivALNo, M. H. C.. Nucci,G. B. N. aim NIGiO. D.: Nitric oxide release may be involved in the micrccir.culatesy rsq.nae In acetylcbolix,e. But. 4. PhermacQI. 182: 143—147, 1990.Gannutz. S. hI.. C0WPToN, A. hI. AND Bgrnrrr. T. Regional and cardiachaemodynamic response. to glyceryl triniteste, scetylchollne, bradykinin and.ndothelin.1 in conscious ratst Effects o(N°.mtro-L-arginine methyl eater. Br.4. PharmacoL 101: 632-639,1990..GAIIDINzg, S. hI., COMPTON, A. hI.., BgreNzrr. T., PALMER. IL hI. 4. ANDMONCADA, S.. Control of regional blood flow by endothelium.deriv,d nitricoxide. Hypertension 16: 486-492, 199Gb.Gsanua, S. M., COMPT0N, A. M.. KEN,. P. A. AND Bversrrr. T.: Regionaland cardiac )saemodynamic effect., of NG.nitro.L.arginine methyl ester Iiconscious, Long Evans rats. Br. 4. Pliarmacol. 101: 625431. 1990c.GARDINER, S. M., Kass,. P. A. AND BZNNeFT, T.: Effects of N°.nitro-L.argininemethyl eater on vasodilator responses to acetyichoilne. 5’.N.ethyami.doadeno.jne or ialbutsmol in conscious rita. Br. 3. Pharmacol. 103: 1725-1732. 1991.GARLAND. C. 3. AND MCPHERSON. G. A.: Evidence that nitric oxide does notmediate the hyperpolartrstion and relaxation to scetyicholine in the rat smallmesenteric artery. Br. .1. PharniacoL 105: 429-435. 1992.ISHIL. K., CHANG. B.. KERWIN. 3. F., .IR.. HUANG. Z.-J. AND MuARD. F.: s”.nitro-L-arginine: A potent inhibitor of endotheliuxn-derived relaxing factorformation. Eu:. .1. Pharniacol. 176: 219-223. 1990.KENAKIN. T. P.: Pharmacologic Analyila of Drug-Receptor Interaction. pp. 129-244, Raven Press, New York, 1987.LANERA. V., SALOM. M. G., FU(SEN.OLSEN, M. 3., RALJ. L. AND R0MER0, 3. C.:Effects of N°.monomethyl.L.azginine and L.arginine on acetyicholine renalresponse. Hypertension 15: 659-663. 1990.MONCADA. S., PALMER, IL M. 3. AND RIGGS. E. A.: Nitric oxide: Physiology,pathophysiology, and pharmacology. Pharmacol. Rev. 43: 109—181. 1991.MO!.sH, A. AND SvSsEu. IL: N°.nitro.L.arginine (N’.(imino (nitroaniino)methylj.L.ornithine) impairs endothelium.dependent dilatation by inhibitingcytosolic nitric oxide synthesis from L’arginine. Naunyn.Schmiedeberg’s Arch.Pharmacol. 341: 143—147. 1990.PALMER, K. M. J., ASHTON. D. S. AND MONCADA. S.: Vascular endothelial cellssynthesize nitric oxide from L.arginine. Nature (Lond.) 333: 664—666. 1988a.PALMER. K. M. 3.. REEs. D. D.. ASHTON D. S. AND MONCADA, S.: L.arginine isthe physiological precursor for the formation of nitric oxide in endotheliumdependent relaxation. Biochem. Biophys. Re,. Common. 153: 1251—1256.1988b.PANG. C. C. Y. AND WANG, Y.-X.: Differential in rwo and in uiro vasculareffects of methylene blue (MB) and N°.nitro.L.argmixre (L.NNA). Pharnia.cologist 34: 165, 1993.POON. C. I., WANG, Y..X., POON, K. S. AND PANG, C. C. Y.: Inhibitory effectsof dlphenyleneiodonium on endotheium-dependent vasodilation in uino andin uivo. Fed. Am. So.. Lip. BioL J. 7: A261, 1993.RxzS, D. D., PALMER, IL M. 3., HODSON. H. F. AND MONCADA. S.: A specificinhibitor of nitric oxide formation &om L-argmine attenuates endothelium.dependent relaxation. Br. 3. Pharniarol. 96: 418—424. 1989..REEs. D. D.. PALMER. IL M. J. AND MONCADA. S.: The role of endotbelium.derived nitric oxide in the regulation of blood pressure. Proc. Narl. Aced. Sci.U.S.A. 86: 3375-3378, 1989..D. I).. PALMER, IL M. 3., ScHuLz, K.. HODSON, H. F. AND MONCADA. S.:Characterization of three inhibitors of eudotheilal nitric oxide synthesis inmtn, and in vivo. Br. 3. PharmacoL 101: 746—752. 1990.STUEHR. D.. FASEHUN. 0. A., KWON, F. S.. GROSS. S. S., GONZALEZ, .1. A., LEVI.IL AND NATHAN, C. F.: Inhibition of macrophage and endothelial cell nitricoxide synthase by dlpheoyleneiodonium and its analop Fed. Am. Soc. KrpBioL 3. 5: 98-103, 1991.SuzuKi, H. AND Cxzre. G.: Endothelbon-derived hyperpolarixing factor (EDHF)An endogetlod. potsaaium.cbannel ctivator. New, PhysioL Sd. 5: 212-215.1990.TODA. N., KITAMUs.s, F. AND OICAMURA, 1.: Neural mechanism of bypeitasalocby nitric oxide aysrthasc Inhibitor In dogs. Hypertension 21:3-8,1998.T0GASHI. H., SAKUUA. L. Yosmoic.t. M., KOBAYASHI. T.. YASUDA, H.. Krra.SATAKE, A., SAlTy, H.. GRoss, S. S. AND LEvj. B,: A central nervous systemaction of nitric aside in blood pressure regulation. 3. Pharmacot Exp Ther.262:343-347, 1992.VAN GELDEREN. E. M.. HEILIGERS. 3. P. C. AND SAXENA. P. IL: Haemodynamiechanges and.cetylchollne.inducrd hypotenaive responses after N°.uiuo-L.arginine methyl ester in rats and cat.,. Br. 3. Pharinacol. 103: 1899-1904.1991.1931098 Wang itl.duced pressor responses by sympathetic activation rather thanthe inhibition of NO synthesis (Wang and Pang, 1993a,b), eventhough diphenyleneiodonium caused prolonged inhibition ofendotheium-dependent vasodilatations in vitro and in uwo(Stuehz of aL, 1991; Poon of aL, 1993). Moreover, methyleneblue suppressed ACh-induced relaxations in vitro but did notcause a pressor response in vitro (Pang and Wang, 1993).Therefore, the inhibition of endogenous endothelial NO synthesis does not necessarily cause vasoconstriction in vitro. Analternative explanation is that there is a difference betweenstimulated NO synthesis (ACh-induced relaxation) and basalNO synthesis (vasodilatation) (Chyn et aL, 1992). It was reported that the endothelium-dependent contractions elicitedby t-NMMA and L-NNA in the dog coronary artery were nota consequence of the suppression of basal NO synthesis (Cocksand Angus, 1991). It was also shown that, although both LNNA and D-NNA inhibited ACh-induced relaxation in aorticrings, only L-NNA elicited contraction and potentiated PEinduced contraction (Wang of aL, 1993b). Moreover, it has beensuggested recently that N°-substituted L-Arg analogs causepressor responses by inhibiting central (Togashi of aL, 1992) organglionic (Toda et aL, 1993) NO synthesis rather than endothelial NO production. However, most authors reported thatthe pressor responses to N°.substituted Arg analogs remain (orare potentiated) after impairments of the central (Wang andPang, 1993a) or autonomic (Rees et at., 1989b; Wang and Pang,1991; Chyu et aL, 1992) nervous systems. Regardless of theexact mechanism or site of the pressor response of L-NAME,this study is the first to show that L.4&.rg antagonizes competitively the pressor response of L-NAME, with half.blocking doseat 10 mol/kg/min.In summary, L-NAME is an efficacious and long.lastinginhibitor of ACh-induced relaxation of the aortae, with a Hillcoefficient of 1 and maximal inhibition of nearly 100%. InConscious rats, L-NAM_E causes partial inhibition of the vasodilatation response to ACh. This is characterized as a shortenedduration (but not magnitude) of action of i.v. bolus injectedACh and a reduced magnitude of i.v. infused ACh. L.NAMEcauses a dose-dependent pressor response with a Hill coefficientof 2, which can be antagonized competitively by L-Arg (but notby D-Arg), without changing the Hill coefficient. Differences inHill coefficients and effectiveness of inhibition of ACh-inducedvasodilatation in vitro and in vivo for c-NAME suggest thateither one of the following hypotheses may be true: 1) that thepressor response of NO synthase inhibitors in whole animals isnot due entirely to the inhibition of endogenous endothelia]NO biosynthesis or 2) that basal NO synthesis is blocked moreeffectively by L-NAME than stimulated NO synthesis.Vascular Etlects ol L.NAME1941099VASGAS. H. CUEVAS. .1. M.. IGNAISKO. L. .1. AND CHAUDHURi, G.: Comparison,( he inhibitory potsncias of N° m.thyl, N°-nitro. and N°.amino.L-atginineon WRY function in the rat Evidence for contmuoue basal EDRY isles... J.PhermacoL Exp. That. 257: 1208-1215,1991.WANG. Y.-X.. ASneLISANMAN. A. AND PANG, C. C. Y.. Selectiv. inhibition ofpraseor and bemodynamic effects of N°.aitro.L-srginlne by balothane. J.Cardlovs.c. Pharmacol. 19 in pie.., 1993a.WANG. Y.-X. AND PANG, C. C. Y. Preesor effect of N°.nitro.L.arginin. inpentobarbital-snesthetinsd rats. UI. Sd. 41: 2217-2224, 1990.WANG. Y.-X. ANOPANG, C. C. Y.: Possible dependence of pressor and heart rateeffects of N°.oitro-L-arginlne on autonomic nerve activity. Br. J. Pharcoscol.103: 2004-2008,1991.WANG. Y.-X. aim PANG, C. C. Y. Functional integrity of the central andsympathetic neI syltema is a prerequisite for pr.or and tachycardiceffects of dlpbenyl.neiodonium. a novel inhibitor of nitric oxide synthase. J.PbarmacoL Sip. That. 265:263-372, 1993a.WANG, Y..X. AND PANG, C. C.. Y. Halothane inhibits preseor effact of diphenylenelodoniura. Br. J. Pharmacoh. 108: 1186—1191. 1993b.WANG. Y.-X.. POOH. C. L AND PANG, C. C. Y.: In uitro and ax moo inhibitoryeffects of I. and D enantiomece of l°.njOO.aepje on .ndotheliwn.dependentrelaxation of rat aorta. J. Pbarmacol. Exp. Ther. 266: 112-119. 1993b.WANG. Y.-X.. Z)sou, 1. aim PANG, C. C. Y.: Preeaot.ffecte of Land 0enantiomera of N°-oitro.arginine in conscious rats age antagonized by L. butnot D.srginizie. Eur. J. Pharmacoi. 200:77-81. 1991.WANG. Y. X., ZHOU. T. AND PANG. C. C. Y.: Depressor responses to acetylcholineand preasor responses to N°.nitro.L..rginine in conscious Wistar. WKY andSM rats. In ad. by S Moncada, M A. Marietta and J. B. Jiis. Hi),., The Biologyof Nitric Oxide. pp. 174-176. Portland Press. London, 1992.WANG, Y.-X.. Znou, T. AND PANG. C. C. Y.: A comparison of the inhibitoryeffects of sodium nitzopruaeide. pinacidal and nif.dipine on pr.eeor responseBr. J. Pharmacol. 108:396-404, 1993c.WATKINS, R., Coog, L aim McLaoD, R.: N°.inonomethyl4_.rginine(LNMMA) dose not inhibit depressor response. to acetyicholin. (ACh) in rats.Fed. Am, Soc. Sip. BioL J. 7: A775. 1993.WHrrrLz, B. J., Loraz.Bu.sso,rrs J. aim Hiss, 0. D. Modulation of thevesodeprseear actions of acecyicholine, bradykinin, subetancs P and endothelinin the rat by a specific inhibitor of nitric oxide formation. Br. .1. Pbarmacol.98: 646-862, 1980.WOODMAN. 0. L. aim OusrusG, G. J. N.nitro L.argurine causes coronaryvasoconatriction and inhibits sqdothelium.depeadent vaeodilation in ansesthe.deed greyhound. 3r 1. Pharmacol. 103: 1407-1410,1991.Yazeaxasa. 3., Pwtra, N. aim NAGAO. T.. N°.monomethy1.L.azginine.inducedpresaor response at developmental and established stages in spontaneoulyhypertensive rats. J. PharmacoL Sip. Ther. 259:52-57. 1991.Send reprint requests to: Dr. Catherine C. Y. Pang, Department of Pharma.cology & Therapeutics, Faculty of Medicine, The University of British Columbia.2176 Health Sciences Mall. Vancouver, B. C., Canada, V6T 173.