Edhf 2002

EDHF 2002

EDHF 2002

Edited by

Paul M. Vanhoutte Department of Pharmacology The University of Hong Kong Hong Kong

First published 2003 by Taylor & Francis 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Taylor & Francis Inc, 29 West 35th Street, New York, NY 10001 Taylor & Francis is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2003 Taylor & Francis All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Every effort has been made to ensure that the advice and information in this book is true and accurate at the time of going to press. However, neither the publisher nor the authors can accept any legal responsibility or liability for any errors or omissions that may be made. In the case of drug administration, any medical procedure or the use of technical equipment mentioned within this book, you are strongly advised to consult the manufacturer’s guidelines. This publication has been made possible through an educational grant from Servier. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0-203-98736-5 Master e-book ISBN

ISBN 0–415–30853–4 (Print Edition)

Contents

Preface Acknowledgements Contributors 1

Potassium channels and membrane potential in vascular endothelial and smooth muscle cells

xi xii xiii

1

KATHRYN M. GAUTHIER AND NANCY J. RUSCH

2

Possible contribution of CLCA1 to calcium-activated chloride channels in murine smooth muscle cells

13

I.A. GREENWOOD, F. BRITTON, S. OHYA AND B. HOROWITZ

3

Trafficking and transduction functions of the Na pump in vascular smooth muscle cells

20

JULIUS C. ALLEN, ASLIHAN AYDEMIR-KOKSOY AND JOEL ABRAMOWITZ

4

Isoforms of the Na,K-ATPase

27

KRISTAN LANSBERY, MARGARETTA L. MENDENHALL, LAUREN C. VEHIGE, JAMES A. TAYLOR, GADIS SANCHEZ, GUSTAVO BLANCO AND ROBERT W. MERCER

5

Calcium sparks and membrane potential

35

DELRAE M. ECKMAN, LUIS FERNANDO SANTANA, THOMAS J. HEPPNER, ADRIAN D. BONEV AND MARK T. NELSON

6

Proteinase-activated receptor-2: release of an endothelium-derived hyperpolarizing factor distinct from that released by acetylcholine

47

J.J. M C GUIRE, Y. GUI, X. WANG, R.D. LOUTZENHISER, C.R. TRIGGLE AND M.D. HOLLENBERG

7

Mechanical stimulation increases the activity and expression of cytochrome P450 2C in porcine coronary artery endothelial cells B. FISSLTHALER, U.R. MICHAELIS, R. BUSSE AND I. FLEMING

56

vi

Contents

8 Important role of hydrogen peroxide as an endothelium-derived hyperpolarizing factor in animals and humans

63

HIROAKI SHIMOKAWA, TETSUYA MATOBA, KEIKO MORIKAWA, TOYOTAKA YADA, HIROSHI KUBOTA, YOJI HIRAKAWA AND AKIRA TAKESHITA

9 Altered calcium dynamics do not account for the attenuation of EDHF-mediated dilatations in the middle cerebral artery of female rats

70

ELKE M. GOLDING, DOROTA M. FERENS, SEAN P. MARRELLI AND ROBERT M. BRYAN JR

10 Connexin-mimetic peptides: influence on nitric oxide synthase- and cyclooxygenase-independent renal vasodilatation, basal renal blood flow and blood pressure in the rat

78

AN S. DE VRIESE, JOHAN VAN DE VOORDE AND NORBERT H. LAMEIRE

11 Urocortin-induced relaxations of the rat coronary artery

87

YU HUANG, FRANKY LEUNG CHAN, CHI-WAI LAU, ZHEN-YU CHEN, GUO-WEI HE, SUK-YING TSANG AND XIAOQIANG YAO

12 Nitric oxide is the only EDHF released by the endothelium in lymphatic vessels of the guinea-pig mesentery

93

PIERRE-YVES VON DER WEID AND ALICE K. CHAN

13 Role of EDHF in vascular tone in vivo

101

HELENA C. PARKINGTON, HAROLD A. COLEMAN AND MARIANNE TARE

14 Endothelium-derived hyperpolarizing factor, myoendothelial gap junctions and hypertension

108

SHAUN L. SANDOW, NARELLE J. BRAMICH, HARI PRIYA BANDI, NICOLE M. RUMMERY AND CARYL E. HILL

15 Improvement of age-related impairment of endothelium-dependent hyperpolarization by renin-angiotensin system blockade

117

YASUO KANSUI, KOJI FUJII, KENICHI GOTO AND MITSUO IIDA

16 Characterization of endothelium-derived hyperpolarizing factor-mediated relaxation of small mesenteric arteries from diabetic (db/db⫺/⫺) mice MALARVANNAN PANNIRSELVAM, TODD J. ANDERSON AND CHRISTOPHER R. TRIGGLE

124

Contents 17 Endothelium-dependent responses in small arteries isolated from normal and pre-eclamptic pregnant women

vii 132

WILLIAM R. DUNN, LOUISE C. KENNY, DAVID A. KENDALL, PHILIP N. BAKER AND MICHAEL D. RANDALL

18 Free radical species and endothelium dysfunction during deoxycorticosterone-salt induced hypertension

139

AYOTUNDE S.O. ADEAGBO, IRVING G. JOSHUA, SUNDAY O. AWE, RUSSELL A. PROUGH AND K. CAMERON FALKNER

19 EDHF involvement in skin pressure-induced vasodilatation

151

AMBROISE GARRY, SANDRA MERZEAU, BÉRENGÈRE FROMY AND JEAN LOUIS SAUMET

20 N-acetylcysteine and immobilization stress attenuate dysregulation of the endothelium-dependent coronary vascular tone induced by acute hemorrhage

156

L.EU. BELYAEVA, V.I. SHEBEKO AND A.P. SOLODKOV

21 Red wine polyphenolic compounds induce EDHF-mediated relaxation and hyperpolarization in the porcine coronary artery: involvement of redox-sensitive mechanisms

165

T. CHATAIGNEAU, M. NDIAYE, J.C. STOCLET AND V.B. SCHINI-KERTH

22 Estrogen substitution restores the basal influence of nitric oxide and endothelium-derived hyperpolarizing factor on vascular tone in isolated mesenteric arteries from ovariectomized rats

174

M. ZERR, T. CHATAIGNEAU, F. HUDLETT, F. PERNOT AND V.B. SCHINI-KERTH

23 Ascorbate inhibits EDHF in the bovine eye but not in the porcine coronary artery

181

ALISTER J. M C NEISH, SILVIA NELLI, WILLIAM S. WILSON AND WILLIAM MARTIN

24 Gabexate mesilate inhibits endothelium-dependent relaxation, but causes endothelium-independent relaxation of rat blood vessels

188

MIKIO NAKASHIMA, TOMOKO HAMADA, SHINJI MITSUMIZO AND TADAHIDE TOTOKI

25 Mechanisms underlying basal vascular tone in the guinea-pig mesenteric arterioles YOSHIMICHI YAMAMOTO AND HIKARU SUZUKI

193

viii

Contents

26 Endothelium-dependent depolarization and its implications for endothelium-derived hyperpolarizing factor

199

HAROLD A. COLEMAN, MARIANNE TARE AND HELENA C. PARKINGTON

27 Role of gap junctions in EDHF-mediated relaxation response in human subcutaneous resistance arteries

205

P. COATS AND C. HILLIER

28 Permissive role of cAMP in the mediation of relaxations initiated by endothelial hyperpolarization

211

TUDOR M. GRIFFITH, ANDREW T. CHAYTOR AND DAVID H. EDWARDS

29 Myoendothelial gap junctions – the critical link for endothelium-derived hyperpolarizing factor

223

MARIANNE TARE, SHAUN L. SANDOW, HAROLD A. COLEMAN, SUSAN J. WIGG, HELENA C. PARKINGTON AND CARYL E. HILL

30 Longitudinal spread of agonist-evoked hyperpolarization in the rat mesenteric artery

234

H. TAKANO, K.A. DORA AND C.J. GARLAND

31 Effect of HEPES on EDHF responses in porcine coronary and rat mesenteric arteries

239

M.J. GARDENER, G. EDWARDS, M FÉLÉTOU, P.M. VANHOUTTE AND A.H. WESTON

32 Quantification of the amount of potassium released by cultured porcine coronary endothelial cells, stimulated by bradykinin

248

JEAN-LOUIS BÉNY, ALEXANDER SCHUSTER, MAUD FRIEDEN, MONICA SOLLINI AND ANNE BARON

33 The intensity of agonist-stimulation influences the mechanism for relaxation in rat mesenteric arteries

256

K.A. DORA, L . M C EVOY, M. KING, N. INGS AND C.J. GARLAND

34 Small and intermediate conductance Ca2⫹-activated K⫹ channels (SKCa and IKCa) in porcine coronary endothelium: relevance to EDHF

261

R. BYCHKOV, M.P. BURNHAM, G .R . RICH A R D S, C. TH O LLO N, G. EDWARDS, A.H. WESTON, P.M. VANHOUTTE AND M. FÉLÉTOU

35 The role of KCa in endothelial cell hyperpolarization and endothelium-dependent relaxation in the rabbit aorta X. KUANG, L. CHAN, W. LIANG, I. LAHER, C. VAN BREEMEN AND X. WANG

274

Contents 36 The contribution of D-tubocurarine and apamin-sensitive potassium channels to endothelium-derived hyperpolarizing factor-mediated relaxation of small mesenteric arteries from eNOS⫺/⫺ mice

ix

283

HONG DING, YANFEN JIANG AND CHRIS R. TRIGGLE

37 Ouabain blocks EDNO-mediated relaxation in mesenteric veins and EDHF-mediated relaxation in mesenteric arteries of the guinea-pig

297

SIMON ROIZES AND PIERRE-YVES VON DER WEID

38 Inhibitors of EDHF-evoked responses and the calcium signal in endothelial cell of mesenteric artery

304

PHILIPPE GHISDAL AND NICOLE MOREL

39 Roles of the inward-rectifier K⫹ channel and Na⫹/K⫹-ATPase in the hyperpolarization to K⫹ in rat mesenteric arteries

309

GILLIAN EDWARDS, GILLIAN R. RICHARDS, MATTHEW J. GARDENER, MICHEL FÉLÉTOU, PAUL M. VANHOUTTE AND ARTHUR H. WESTON

40 Importance of intracellular concentration of sodium in the relaxation of rat isolated mesenteric arteries by potassium

318

DIDIER X.P. BROCHET AND PHILIP D. LANGTON

41 Inhibition of bradykinin-induced relaxations by an epoxyeicosatrienoic acid antagonist: 14,15-epoxyeicosa-5Z-monoenoic acid

325

KATHRYN M. GAUTHIER, PHILLIP F. PRATT, J.R. FALCK AND WILLIAM B. CAMPBELL

42 Local release of EDHF initiates a conducted dilatation, but is not the upstream mediator in arterioles of the hamster

332

COR DE WIT, BERND HOEPFL, STEFFEN-SEBASTIAN BOLZ AND ULRICH POHL

43 Interaction of astrocytes and cerebral endothelial cells: function of astrocytic epoxyeicosatrienoic acids in the differentiation of endothelial cells

341

DAVID R. HARDER, CHENYANG ZHANG, JAYASHREE NARAYANAN AND MEETHA MEDHORA

44 11,12-EETs hyperpolarize human platelets FLORIAN KRÖTZ, TOBIAS RIEXINGER, MATTHIAS KELLER, HAE-YOUNG SOHN AND ULRICH POHL

349

x

Contents

45 Epoxyeicosatrienoic acid activates cloned BKCa channel ␣-subunit through ADP-ribosylation of the G-protein G␣s

356

MITSUHIRO FUKAO, HELEN S. MASON, SATOSHI NAWATE, TAKAMITSU SOMA, ICHIRO SAKUMA, SOICHI MIWA, JAMES L. KENYON, BURTON HOROWITZ AND KATHLEEN D. KEEF

46 Different role of epoxyeicosatrienoic acids (EET11,12) in EDHF-mediated relaxation in small porcine coronary and pulmonary arteries

366

GUO-WEI HE, WEI ZOU, YU HUANG, APC YIM AND QIN YANG

47 EDHF 2002: the take home message

371

MICHEL FÉLÉTOU AND PAUL M. VANHOUTTE

References Index

376 417

Preface

This monograph is the fourth of a series devoted to endothelium-dependent hyperpolarizations. It consists of the Proceedings of the Fourth International Symposium devoted to this topic, which was held in Vaux de Cernay (France) from June 5–7, 2002. The first two meetings established the existence of endothelium-derived hyperpolarizing factor (EDHF) but also revealed the multiplicity of chemical factors potentially involved in endothelium-dependent hyperpolarizations, while the third symposium continued to expand the worldwide interest in knowledge in the matter, and in particular, highlighted the emerging role of gap junctional communication in the genesis of the phenomenon. This fourth monograph illustrates the fact that the EDHF field has truly become the most exciting topic in the current understanding of vasomotor control, particularly in smaller blood vessels. A major emphasis has been to understand better the role of ionic channels, both in endothelial and vascular smooth muscle cells, in mediating endothelium-dependent hyperpolarizations, and this constitutes the first part of the monograph. A second group of chapters addresses in particular the role of EDHF in physiological responses. The third section focuses on the growing evidence that EDHF-mediated responses are modulated by aging and diseases, in particular hypertension and diabetes. Finally, three groups of chapters continue the debate as to the involvement of potassium ions, EETs and gap junctions, respectively in endothelium-dependent hyperpolarizations. As in the other monographs, the last chapter attempts to summarize two days of most exciting science and to provide a reasonable state of the art. As always with multiauthored texts, the responsibility for the scientific content rests with the individual authors. Hence, all the statements made are not necessarily endorsed by the editor. His task has been mainly to select the authors (with the help of the Scientific Committee), to streamline their texts, and to achieve as much uniformity of presentation as possible. EDHF 2002 will be of interest not only to physiologists and pharmacologists puzzled by the complexity of the interactions between the endothelium and the underlying vascular smooth muscle cells, but also to clinical researchers and to the physicians who treat patients with cardiovascular diseases. Indeed, the understanding of the nature and role of EDHF already appears to be crucial in the quest for an improvement in the treatment of hypertension, diabetes, ischemia-reperfusion and other vascular disorders.

Acknowledgements

The participants of the Fourth International Symposium on Endothelium-Derived Hyperpolarizing Factor will not forget the warm hospitality of Mrs Aufort and Mrs Bosio. This monograph would not have been possible without the total dedication of Mr Robert R. Lorenz who took responsibility for the illustrations. The editor would like to thank most sincerely Dr Jacques Servier, who supported the endeavor to try to make this monograph into a reasonably uniform text, despite the many contributors. He would also like to thank the authors most sincerely for their collaboration and understanding when faced with extensive editing of their manuscript. Last, but not least, the staff at the publishers, in particular Mr Grant Soanes and Mr Vincent Antony, should be complimented for their efficient handling of the manuscript.

Contributors

Abramowitz, J. Section of Cardiovascular Sciences Department of Medicine Baylor College of Medicine Houston Texas, 77030 USA Adeagbo, A.S.O. Department of Physiology and Biophysics University of Louisville Louisville, KY 40292 USA e-mail: [email protected] Allen, J.C. Baylor College of Medicine One Baylor Plaza Houston TX, 77030 USA e-mail: [email protected] Anderson, T.J. Division of Cardiology Foothill Hospital Calgary, Alberta Canada Awe, S.O. Departments of Physiology/Biophysics Health Sciences Center School of Medicine University of Louisville Louisville, KY 40292

Aydemir-Koksoy, A. Section of Cardiovascular Sciences Department of Medicine Baylor College of Medicine Houston Texas, 77030 USA Baker, P.N. Maternal & Fetal Health Research Centre University of Manchester Manchester, M13 0JH UK Bandi, H.P. Division of Neuroscience John Curtin School of Medical Research Australian National University Canberra, ACT, 0200 Australia Baron, A. Institut de Pharmacologie Moléculaire et Cellulaire 660 route des Lucioles 06560 Valbonne Sophia-Antipolis France e-mail: [email protected] Belyaeva, L.Eu. Pathophysiology Department Medical University Frunze Street, 27 Vitebsk Belarus, 210602 e-mail: [email protected]

xiv

Contributors

Bény, J.-L. Department of Zoology and Animal Biology University of Geneva, Sciences III Quai E.-Ansermet 30 CH-1211 GENEVA 4 Switzerland e-mail: [email protected] Blanco, G. Department of Cell Biology and Physiology Washington University School of Medicine 660 S. Euclid Avenue St Louis, MO 63110 USA Bolz, S.-S. Physiologisches Institut Ludwig-Maximilians-Universität München 80336 München Bonev, A.D. Department of Pharmacology Given Building, College of Medicine University of Vermont 89 Beaumont Avenue Burlington, Vermont 05405 USA

Brochet, D.X.P. Laboratory of Cardiovascular Sciences GRC, NIA, NIH 5600 Nathan Shock Drive Baltimore, MD 21224 USA Bryan, R.M. Jr Department of Anesthesiology Baylor College of Medicine, Houston TX, 77030 USA Burnham, M.P. School of Biological Sciences University of Manchester G38 Stopford Building Oxford Road Manchester, M13 9PT, UK e-mail: [email protected] Busse, R. Klinikum der J.W.G.-Universität Institut für Kardiovaskuläre Physiologie Theodor Stern Kai 7 60590 Frankfurt/Main Germany e-mail: [email protected] Bychkov, R. Institute of Research Servier 11 rue des Moulineaux 92150 Suresnes, France e-mail: [email protected]

Bramich, N.J. Division of Neuroscience John Curtin School of Medical Research Australian National University Canberra, ACT, 0200 Australia

Campbell, W.B. Department of Pharmacology and Toxicology Medical College of Wisconsin 8701 Watertown Plank Road Milwaukee, WI 53226 USA e-mail: [email protected]

Britton, F. Department of Physiology and Cell Biology and COBRE Program University of Nevada School of Medicine Reno, Nevada 89557-0046 USA

Chan, A.K. Mucosal Inflammation and Smooth Muscle Research Groups Department of Physiology & Biophysics Faculty of Medicine University of Calgary Canada

Contributors xv Chan, L.F. Department of Pharmacology & Therapeutics University of British Columbia Vancouver Canada Chataigneau, T. Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires UMR CNRS 7034 Faculté de Pharmacie Université Louis Pasteur F-67401 Illkirch France e-mail: [email protected] Chaytor, A.T. Department of Diagnostic Radiology Wales Heart Research Institute University of Wales College of Medicine Heath Park, Cardiff CF14 4XN UK Chen, Zhen-Yu Department of Biochemistry Faculty of Medicine The Chinese University of Hong Kong Hong Kong China Coats, P. Department of Physiology & Pharmacology University of Strathclyde 27 Taylor Street Glasgow Scotland e-mail: [email protected] Coleman, H.A. Department of Physiology Monash University Victoria, 3800 Australia De Vriese, An S. Renal Unit Ghent University De Pintelaan 185 OK12 Belgium

de Wit, C. Physiologisches Institut Ludwig-Maximilians-Universität Schillerstr. 44 80336 München Germany e-mail: [email protected] Ding, H. Microvascular Biology Group School of Medical Sciences RMIT University, Bundoora Victoria, 3083 Australia e-mail: [email protected] Dora, K.A. Department of Pharmacy and Pharmacology Claverton Down, 5W Building University of Bath Bath BA2 7AY UK e-mail: [email protected] Dunn, W.R. School of Biomedical Sciences Queen’s Medical Centre University of Nottingham Nottingham, NG7 2UH e-mail: [email protected] Eckman, D.M. Wake Forest University Health Sciences Departments of Pediatrics and Physiology/Pharmacology 2nd floor, Watlington Hall Medical Center Boulevard Winston-Salem, North Carolina 27157 USA e-mail: [email protected] Edwards, D.H. Department of Diagnostic Radiology Wales Heart Research Institute University of Wales College of Medicine Heath Park, Cardiff CF14 4XN UK

xvi

Contributors

Edwards, G. School of Biological Sciences G38 Stopford Building University of Manchester Manchester M13 9PT e-mail: [email protected] Falck, J.R. Department of Biochemistry University of Texas Southwestern Medical School 5323 Harry Hines Blvd Dallas, Texas 75235 Falkner, K.C. Department of Biochemistry/Molecular Biology Health Sciences Center, School of Medicine University of Louisville Louisville, KY 40292 Félétou, M. Département Diabète et Maladies Métaboliques Institut de Recherches Servier 11 rue des Moulineaux 92150 Suresnes France e-mail: [email protected] Ferens, D.M. Department of Anesthesiology Baylor College of Medicine Houston, Texas, 77030 USA

Frieden, M. Département de Pharmacologie et Physiologie Centre Médicale Universitaire 1211 Genève 4 Switzerland e-mail: [email protected] Fromy, B. Laboratoire de Physiologie UPRES EA 2170 UER Médecine 49045 Angers cedex France e-mail: [email protected] Fujii, K. Department of Medicine and Clinical Science Graduate School of Medical Sciences Kyushu University Fukuoka 812-8582 Japan e-mail: [email protected] Fukao, M. Department of Pharmacology Hokkaido University School of Medicine Sapporo 060-8638 Japan e-mail: [email protected]

Fisslthaler, B. Klinikum der J. W. Goethe Universität Institut für Kardiovaskuläre Physiologie Theodor Stern Kai 7 60590 Frankfurt / Main Germany e-mail: [email protected]

Gardener, M.J. School of Biological Sciences University of Manchester G.38 Stopford Building Oxford Road Manchester M13 9PT e-mail: [email protected]

Fleming, I. Klinikum der J.W.G.-Universität Institut für Kardiovaskuläre Physiologie Theodor Stern Kai 7 60590 Frankfurt/Main Germany e-mail: [email protected]

Garland, C.J. Department of Pharmacy and Pharmacology University of Bath Bath BA2 7AY UK

Contributors xvii Garry, A. Laboratoire de Physiologie UPRES EA 2170, UER Médecine 49045Angers cedex France e-mail: [email protected] Gauthier, K.M. Department of Pharmacology and Toxicology 8701 Watertown Plank Road Medical College of Wisconsin Milwaukee, WI 53226 USA Ghisdal, P. Laboratoire de Pharmacologie Université Catholique de Louvain – UCL 5410 – Avenue Hippocrate 54 – B 1200 Bruxelles Belgique Golding, E.M. Department of Anesthesiology One Baylor Plaza, Suite 434D Houston, Texas, 77030 USA e-mail: [email protected] Goto, K. Department of Medicine and Clinical Science Graduate School of Medical Sciences Kyushu University Fukuoka 812-8582 Japan Greenwood, I.A. Department of Pharmacology & Clinical Pharmacology St George’s Hospital Medical School London. SW17 0RE e-mail: [email protected] Griffith, T.M. Department of Diagnostic Radiology Wales Heart Research Institute University of Wales College of Medicine Heath Park Cardiff CF14 4XN e-mail: [email protected]

Gui, Y. Group on Regulation of Vascular Contractility Faculty of Medicine University of Calgary Calgary Canada T2N 4N1 Hamada, T. Surgical Center and Department of Anesthesiology and Critical Care Medicine Saga Medical School Nabeshima, Saga. 849-8501 Japan Harder, D.R. Cardiovascular Center MEB M4870 Medical College of Wisconsin 8701 Watertown Plank Road Milwaukee, WI 53226 USA e-mail: [email protected] He, G.-W. Department of Surgery The Chinese University of Hong Kong Hong Kong, China & Providence Heart Institute Department of Surgery Oregon Health & Science University Portland, OR, USA e-mail: [email protected] Heppner, T.J. Department of Pharmacology Given Building, College of Medicine University of Vermont 89 Beaumont Avenue Burlington, Vermont 05405 USA Hill, C.E. Division of Neuroscience John Curtin School of Medical Research Australian National University Canberra, ACT, 0200 Australia

xviii Contributors Hillier, C. School of Biological and Biomedical Sciences Glasgow Caledonian University Cowcaddens Road, Glasgow Scotland, UK

Hirakawa, Y. Department of Cardiovascular Medicine Kyushu University Graduate School of Medical Sciences Fukuoka Japan

Hoepfl, B. Physiologisches Institut Ludwig-Maximilians-Universität München 80336 München

Hollenberg, M.D. Group on Regulation of Vascular Contractility Faculty of Medicine University of Calgary Calgary Canada T2N 4N1

Horowitz, B. Department of Physiology and Cell Biology and COBRE Program University of Nevada School of Medicine Reno, Nevada 89557-0046 USA

Huang, Y. Department of Physiology Faculty of Medicine The Chinese University of Hong Kong Hong Kong China e-mail: [email protected]

Hudlett, F. Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires UMR CNRS 7034, Faculté de Pharmacie Université Louis Pasteur, F-67401 Illkirch France Iida, M. Department of Medicine and Clinical Science Graduate School of Medical Sciences Kyushu University Fukuoka 812-8582 Japan Ings, N. Department of Pharmacy and Pharmacology University of Bath BATH BA2 7AY UK Jiang, Y. Smooth Muscle Research Group Faculty of Medicine University of Calgary Calgary Canada T2N 4N1 e-mail: [email protected] Joshua, I.G. Department of Physiology/Biophysics, Health Sciences Center School of Medicine University of Louisville Louisville, KY 40292 Kansui, Y. Department of Medicine and Clinical Science Graduate School of Medical Sciences Kyushu University Fukuoka 812-8582 Japan

Contributors xix Keef, K.D. Department of Physiology & Cell Biology University of Nevada School of Medicine Reno, Nevada USA

Kuang, X. Department of Occupational Disease Yang-Pu District Central Hospital Shanghai China

Keller, M. Institute of Physiology Ludwigs-Maximilians-Universität Schillerstr. 44 80336 München

Kubota, H. Department of Cardiovascular Medicine Kyushu University Graduate School of Medical Sciences Fukuoka Japan

Kendall, D.A. School of Biomedical Sciences Queen’s Medical Centre University of Nottingham Nottingham, NG7 2UH UK Kenny, L.C. Maternal & Fetal Health Research Centre University of Manchester Manchester, M13 0JH UK Kenyon, J.L. Department Physiology & Cell Biology University of Nevada School of Medicine Reno, Nevada USA King, M. Department of Pharmacy and Pharmacology University of Bath BATH BA2 7AY UK Krötz, F. Institute of Physiology Ludwig-Maximilians-Universität Schillerstr. 44 80336 München Germany e-mail: [email protected]

Laher, I. Department of Pharmacology & Therapeutics University of British Columbia Vancouver Canada Lameire, N.H. Renal Unit Ghent University De Pintelaan, 185 OK12 Belgium e-mail: [email protected] Langton, P.D. Department of Physiology University of Bristol Bristol, BS8 1TD UK Lansbery, K. Department of Cell Biology and Physiology Washington University School of Medicine 660 S. Euclid Avenue St. Louis, MO 63110 USA Lau, C.-W. Department of Physiology Faculty of Medicine The Chinese University of Hong Kong Hong Kong China Leung Chan, F. Department of Anatomy Faculty of Medicine The Chinese University of Hong Kong Hong Kong China

xx

Contributors

Liang, W. Department of Pharmacology & Therapeutics University of British Columbia Vancouver Canada

Mason, H.S. Physiology & Cell Biology University of Nevada School of Medicine Reno, Nevada USA

Loutzenhiser, R.D. Smooth Muscle Research Group Faculty of Medicine University of Calgary Calgary, Alberta Canada T2N 4N1

Matoba, T. Department of Cardiovascular Medicine Kyushu University Graduate School of Medical Sciences Fukuoka Japan e-mail: [email protected]

McEvoy, L. Department of Pharmacy and Pharmacology University of Bath BATH BA2 7AY UK McGuire, J.J. Smooth Muscle Research Group Faculty of Medicine University of Calgary Calgary Canada T2N 4N1 e-mail: [email protected] McNeish, A.J. Division of Neuroscience & Biomedical systems Institute of Biomedical & Life Sciences West Medical Building University of Glasgow Glasgow G12 8QQ Scotland, UK Marrelli, S.P. Department of Anesthesiology Baylor College of Medicine Houston, Texas, 77030 USA Martin, W. Division of Neuroscience & Biomedical systems Institute of Biomedical & Life Sciences West Medical Building University of Glasgow Glasgow G12 8QQ Scotland e-mail: [email protected]

Medhora, M. Cardiovascular Center Medical College of Wisconsin Milwaukee, Wisconsin USA Mendenhall, M.L. Department of Cell Biology and Physiology Washington University School of Medicine 660 S. Euclid Avenue St Louis, MO 63110 USA Mercer, R.W. Washington University School of Medicine 660 S Euclid Ave Box 8228 St Louis, MO 63110 USA e-mail: [email protected] Merzeau, S. Laboratoire de Physiologie UPRES EA 2170, UER Médecine, 49045 Angers cedex France Michaelis, U.R. Klinikum der J.W.G.-Universität Institut für Kardiovaskuläre Physiologie Theodor Stern Kai 7 60590 Frankfurt/Main Germany e-mail: [email protected]

Contributors xxi Mitsumizo, S. Surgical Center and Department of Anesthesiology and Critical Care Medicine Saga Medical School Nabeshima, Saga. 849-8501 Japan Miwa, S. Department of Pharmacology Hokkaido University School of Medicine Sapporo Japan Morel, N. Laboratoire de Pharmacologie UCL 5410 Avenue Hippocrate 54B-1200 Bruxelles Belgium e-mail: [email protected] Morikawa, K. Department of Cardiovascular Medicine Kyushu University Graduate School of Medical Sciences Fukuoka Japan Nakashima, M. Surgical Center Saga Medical School Hospital 5-1-1 Nabeshima Saga, 849-8501 Japan e-mail: [email protected] Narayanan, J. Cardiovascular Center Medical College of Wisconsin Milwaukee, Wisconsin USA Nawate, S. Department Cardiovascular Medicine Hokkaido University School of Medicine Sapporo Japan

Ndiaye, M. Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires UMR CNRS 7034, Faculté de Pharmacie Université Louis Pasteur, F-67401 Illkirch France Nelli, S. Division of Neuroscience & Biomedical systems Institute of Biomedical & Life Sciences West Medical Building University of Glasgow Glasgow G12 8QQ Scotland, UK Nelson, M.T. Department of Pharmacology Given Building, College of Medicine University of Vermont 89 Beaumont Avenue Burlington, Vermont 05405 USA e-mail: [email protected] Ohya, S. Department of Physiology and Cell Biology and COBRE Program University of Nevada School of Medicine Reno, Nevada 89557-0046 USA Pannirselvam, M. Department of Pharmacology & Therapeutics & Smooth muscle research group Faculty of Medicine, University of Calgary Calgary, Alberta, T2N 4N1 Canada Parkington, H.C. Department of Physiology PO Box 13F Monash University Victoria, 3800 Australia e-mail: [email protected]

xxii

Contributors

Pernot, F. Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires UMR CNRS 7034, Faculté de Pharmacie Université Louis Pasteur, F-67401 Illkirch France Pierre-Yves von der Weid Department of Physiology & Biophysics Faculty of Medicine University of Calgary 3330 Hospital Drive N.W Calgary, Alberta Canada T2N 4N1 e-mail: [email protected] Pohl, U. Physiologisches Institut Ludwig-Maximilians-Universität München 80336, München

Riexinger, T. Institute of Physiology Ludwigs-Maximilians-Universität Schillerstr. 44 80336 München Roizes, S. Mucosal Inflammation and Smooth Muscle Research Groups Department of Physiology & Biophysics Faculty of Medicine University of Calgary Canada Rummery, N.M. Division of Neuroscience John Curtin School of Medical Research Australian National University Canberra, ACT, 0200 Australia

Pratt, P.F. Department of Pharmacology and Toxicology Medical College of Wisconsin 8701 Watertown Plank Road Milwaukee, WI 53226

Rusch, N.J. Department of Pharmacology Medical College of Wisconsin 8701 Watertown Plank Road Milwaukee, WI 53226 USA e-mail: [email protected]

Prough, R.A. Department of Biochemistry/Molecular Biology Health Sciences Center School of Medicine University of Louisville Louisville, KY 40292

Sakuma, I. Department of Cardiovascular Medicine Hokkaido University School of Medicine Sapporo Japan

Randall, M.D. School of Biomedical Sciences Queen’s Medical Centre University of Nottingham Nottingham, NG7 2UH Richards, G.R. Department of Biochemistry and Molecular Biology Merck Sharp & Dohme Eastwick Road Harlow Essex CM20 2QR UK

Sanchez, G. Department of Cell Biology and Physiology Washington University School of Medicine 660 S. Euclid Avenue St Louis, MO 63110 USA Sandow, S.L. Division of Neuroscience John Curtin School of Medical Research Australian National University Canberra, ACT, 0200 Australia e-mail: [email protected]

Contributors xxiii Santana, L.F. University of Washington Department of Physiology and Biophysics Box 357290 Seattle, Washington 98195 USA

Sohn H.-Y. Medizinische Poliklinik Ludwigs-Maximilians-Universität Ziemssenstr. 1 80336 München

Saumet, J.L. Laboratoire de Physiologie UPRES EA 2170, UER Médecine 49045 Angers cedex France

Sollini, M. Département de Pharmacologie et Physiologie Centre Médicale Universitaire 1211 Genève 4 Switzerland e-mail: [email protected]

Schini-Kerth, V.B. Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires UMR CNRS 7034, Faculté de Pharmacie Université Louis Pasteur, F-67401 Illkirch France e-mail: [email protected]

Solodkov, A.P. Central Scientific Research Laboratory Vitebsk State Medical University Prospect Frunze, 27 210023, Vitebsk Belarus Republic e-mail: [email protected]

Schuster, A. Whitaker Institute of Biomedical Engineering University of California, San Diego 9500 Gilman Drive La Jolla, CA 92093-0412 e-mail: [email protected] Shebeko, V.I. Pathophysiology Department Vitebsk Medical University Frunze street, 27 Vitebsk, Belaru, 210602 Shimokawa, H. Department of Cardiovascular Medicine Kyushu University Graduate School of Medical Sciences Fukuoka 812-8582 Japan e-mail: [email protected]

Soma, T. Department of Cardiovascular Medicine Hokkaido University School of Medicine Sapporo Japan Stoclet, J.C. Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires UMR CNRS 7034, Faculté de Pharmacie Université Louis Pasteur, F-67401 Illkirch France Suzuki, H. Department of Regulatory Cell Physiology Bioregulatory Medicine Nagoya City University Graduate School of Medical Sciences Nagoya 467-8601 Japan

xxiv

Contributors

Takano, H. Department of Pharmacy and Pharmacology University of Bath Bath BA2 7AY UK Takeshita, A. Department of Cardiovascular Medicine Kyushu University Graduate School of Medical Sciences Fukuoka Japan Tare, M. Department of Physiology Monash University Clayton, Victoria 3800 Australia e-mail: [email protected] Taylor, J.A. Department of Cell Biology and Physiology Washington University School of Medicine 660 S. Euclid Avenue St Louis, MO 63110 USA Thollon, C. Institute of Research Servier 11 rue des Moulineaux 92150 Suresnes, France e-mail: [email protected] Totoki, T. Surgical Center and Department of Anesthesiology and Critical Care Medicine Saga Medical School Nabeshima, Saga. 849-8501 Japan Triggle, C.R. Department of Pharmacology & Therapeutics 3330, Hospital Drive (N.W) Calgary, Alberta T2N 4N1 Canada e-mail: [email protected]

Tsang, S.-Y. Department of Physiology Faculty of Medicine The Chinese University of Hong Kong Hong Kong China van Breemen, C. Department of Pharmacology & Therapeutics University of British Columbia Vancouver Canada Vanhoutte, P.M. Department of Pharmacology, Faculty of Medicine, The University of Hong Kong, 2/F, Laboratory Block, 21 Sassoon Road, Hong Kong e-mail: [email protected] Van de Voorde, J. Department of Physiology and Pathophysiology De Pintelaan 185 – Blok K 9000 Gent Belgium e-mail: [email protected] Vehige, L.C. Department of Cell Biology and Physiology Washington University School of Medicine 660 S. Euclid Avenue St Louis, MO 63110 USA Wang, X. Department of Pharmacology & Therapeutics Faculty of Medicine The University of British Columbia 2176 Health Sciences Mall Vancouver BC Canada V6T 1Z3 e-mail: [email protected]

Contributors xxv Weston, A.H. School of Biological Sciences University of Manchester Manchester M13 9PT UK Wigg, S.J. Department of Physiology PO Box 13F Monash University Victoria, 3800 Australia Wilson, W.S. Division of Neuroscience & Biomedical systems Institute of Biomedical & Life Sciences West Medical Building University of Glasgow Glasgow G12 8QQ Scotland, UK Yada, T. Department of Medical Engineering Kawasaki Medical School Kurashiki Japan Yamamoto, Y. Department of Regulatory Cell Physiology Bioregulatory Medicine Nagoya City University Graduate School of Medical Sciences Nagoya 467-8601 Japan e-mail: [email protected] Yang, Q. Department of Surgery The Chinese University of Hong Kong Hong Kong China

Yao, X. Department of Physiology Faculty of Medicine The Chinese University of Hong Kong Hong Kong China Yim, APC Department of Surgery The Chinese University of Hong Kong Hong Kong China Yu Huang Department of Physiology Chinese University of Hong Kong Hong Kong China e-mail: [email protected] Zerr, M. Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires UMR CNRS 7034, Faculté de Pharmacie Université Louis Pasteur, F-67401 Illkirch France Zhang, C. Cardiovascular Center Medical College of Wisconsin Milwaukee, Wisconsin USA Zou, W. Department of Surgery The Chinese University of Hong Kong Hong Kong China

1

Potassium channels and membrane potential in vascular endothelial and smooth muscle cells Kathryn M. Gauthier and Nancy J. Rusch

Potassium channels regulate the resting membrane potential of endothelial and vascular smooth muscle cells. In both types of cells, these ion channels are key elements in cellular pathways that promote vasodilatation. In endothelial cells, the activation of K⫹ channels results in membrane hyperpolarization and an enhanced electrical gradient for Ca2⫹ influx through nonselective cation channels. The subsequent rise in intracellular Ca2⫹ in the endothelial cells stimulates the synthesis and release of relaxing factors that act on the adjacent smooth muscle cells to reduce arterial tone. In contrast, the activation of K⫹ channels in vascular smooth muscle cells directly mediates vasodilatation by reducing Ca2⫹ influx through voltage-gated Ca2⫹ channels. Recent evidence suggests that the endothelial and vascular smooth muscle cells of the same artery express unique and cell-specific populations of K⫹ channels that emanate from at least five different gene families. The pattern and level of expression of these K⫹ channel genes may be altered during cardiovascular disease states, resulting in abnormal K⫹ channel profiles and abnormal vasodilator responses.

1. INTRODUCTION The plasma membranes of endothelial and vascular smooth muscle cells express unique populations of K⫹ channels, and the K⫹ currents generated by these channels regulate electromechanical coupling and contraction of the blood vessel wall. In both endothelial and vascular smooth muscle cells, the opening of K⫹ channels establishes the resting membrane potential and promotes vasodilatation. In endothelial cells, the activation of K⫹ channels enhances the Ca2⫹-dependent synthesis and release of relaxing factors. In contrast, the opening of K⫹ channels in vascular smooth muscle cells directly promotes vasodilatation at the level of the effector cell. Thus, K⫹ channels in endothelial and smooth muscle cells work in concert to establish an optimal level of contraction in the blood vessel wall that, under normal circumstances, permits organs and tissues to be perfused commensurate with their metabolic needs. Knowledge related to the structure–function relationship of K⫹ channels in different tissues has been revolutionized in the past 15 years by molecular techniques, protein crystallography, and by the direct study of ion channel properties using patch-clamp techniques. Many of the genes encoding the pore-forming subunits of K⫹ channels have been cloned, and the function of these subunits characterized alone or in combination with regulatory proteins in heterologous expression systems. Unfortunately, the molecular characterization of K⫹ channels in endothelial and vascular smooth muscle cells has lagged behind those in other tissues, due in large part, to the difficulty of isolating mRNA and protein from these small cells and studying them by patch-clamp methods. Regardless, the emerging picture provided by a number of laboratories suggests that endothelial and vascular smooth muscle

2

Kathryn M. Gauthier and Nancy J. Rusch

cells maintain unique K⫹ channel populations within the arterial wall that minimally include five K⫹ channel gene families, including inwardly rectifying K⫹ channels, voltage-gated K⫹ channels and three types of Ca2⫹-sensitive K⫹ channels. Furthermore, it appears that the expression of these gene families is altered in endothelial and vascular smooth muscle cells during cardiovascular diseases, resulting in abnormal K⫹ channel profiles that mediate vasodilator responses.

2. REGULATION OF VASCULAR TONE BY K⫹ CHANNELS Under resting conditions, K⫹ efflux across the plasma membrane is the primary driving force that confers a negative level of membrane potential to the endothelial and smooth muscle cells of the arterial wall. There are several unique characteristics that distinguish the electrical properties of these cells from those of cardiac myocytes and neurons. First, endothelial cells and vascular smooth muscle cells rely predominantly on the inhibition of resting K⫹ efflux for depolarization because their plasma membranes appear to lack expression of fast sodium channels to carry excitatory inward current. Second, endothelial cell membranes apparently lack significant numbers of voltage-gated Ca2⫹ channels, and rely instead on voltage-insensitive cation channels as pathways for Ca2⫹ influx. Third, the range of resting membrane potentials of ⫺35 mV to ⫺60 mV recorded in endothelial and vascular smooth muscle cells is considerably less negative than the values of ⫺70 mV to ⫺90 mV observed in cardiac and neuronal cells. Finally, endothelial and vascular smooth muscle cells appear to be electrically coupled by gap junction proteins, permitting ionic cross-talk between these cell types to coordinate electrical events and the level of excitability within the blood vessel wall. Membrane hyperpolarization mediated by the opening of K⫹ channels promotes Ca2⫹related vasodilator pathways in endothelial and vascular smooth muscle cells. In view of this commonality, it initially seems surprising that K⫹ channel opening enhances Ca2⫹ influx in endothelial cells but reduces Ca2⫹ influx in vascular smooth muscle cells to mediate vasodilator events. This apparent discrepancy can be explained by the different properties of the Ca2⫹-permeable channels that are expressed in the plasma membranes of the two cell types. Endothelial cells primarily express voltage-insensitive cation channels as Ca2⫹ influx pathways (Nilius et al., 1997), and the level of Ca2⫹ influx (and the movement of other cations) through these channels is a function of the transmembrane potential (Figure 1.1).

Endothelial cells

A Cation channel

Ca

2+

(+) – –– [Ca2+] – – – – – K+ – – EDRF synthesis

B

Smooth muscle cells Voltage-gated Ca2+ Ca2+ channel (–) – –– [Ca2+] –– – – K+ – Contraction

Figure 1.1 (A) The opening of K⫹ channels in endothelial cells establishes the electrical gradient that enhances Ca2⫹ influx through cation channels, and increases the release of endotheliumderived relaxing factors (EDRFs). (B) The opening of K⫹ channels in vascular smooth muscle cells inhibits Ca2⫹ influx through voltage-gated Ca2⫹ channels and attenuates contraction.

Vascular K⫹ channels

3

Thus, hyperpolarization of the endothelial cell membrane mediated by the opening of K⫹ channels enhances the electrical gradient for Ca2⫹ influx through the cation channels and ultimately increases the cytosolic Ca2⫹ concentration to permit the synthesis and release of vasoactive factors. Under physiological circumstances, the release of endothelium-derived relaxing factors appears to predominate over the release of constrictor factors so that the opening of K⫹ channels in endothelial cell membranes culminates in the relaxation of the adjacent smooth muscle cells. In contrast, whereas endothelial cells rely on voltage-insensitive cation channels for Ca2⫹ influx, the vascular smooth muscle cells primarily rely on voltage-gated Ca2⫹ channels to provide extracellular Ca2⫹ for vascular contraction (Figure 1.1). Under these conditions, the membrane hyperpolarization resulting from the activation of K⫹ channels closes the voltagesensitive Ca2⫹ channels, reduces Ca2⫹ influx, and triggers vasodilatation. Thus, due to the cell-specific expression of different types of Ca2⫹ channels in endothelial and vascular smooth muscle cells, the hyperpolarization mediated by K⫹ channel opening initiates pathways in both cell types that promote vasodilatation. Membrane hyperpolarization enhances Ca2⫹ influx into endothelial cells to increase the synthesis and release of vasodilator substances, and in parallel, reduces Ca2⫹ influx into vascular smooth muscle cells to mediate vasodilatation directly at the level of the effector cell. The loss of the K⫹ channels that mediate hyperpolarization in endothelial or vascular smooth muscle cells may contribute to the anomalous levels of vascular tone that exist in some cardiovascular disease states including hypertension, diabetes, and atherosclerosis.

3. POTASSIUM CHANNELS IMPLICATED IN THE RELEASE AND ACTION OF EDHFs Potassium channels emanate from many gene families, and their expression is highly heterogeneous between different arterial sites. More than 70 K⫹ channel genes have been cloned from human tissues, and further diversity is generated by the presence of gene subfamilies, alternative splicing, and channel association with regulatory subunits and proteins (Coetzee et al., 1999). Due to space constraints, this chapter will focus only on those K⫹ channel families that have been clearly implicated in endothelium-dependent dilator responses. In addition, this chapter will primarily report the findings of studies in freshly isolated rather than cultured cell preparations, due to concerns regarding the phenotypic modulation of K⫹ channels under culture conditions.

3.1. Inwardly rectifying K⫹ channels There is ample evidence from vascular reactivity, patch-clamp and gene deletion studies that inwardly rectifying K⫹ (Kir) channels are expressed and functionally active in endothelial and vascular smooth muscle cells. These channels are named for their common property of “inward rectification,” indicating that they conduct inward K⫹ current more readily than outward K⫹ current and, therefore, can efficiently function at the negative membrane potential levels at which endothelial and smooth muscle cells are physiologically poised. The Kir channels emanate from many gene families (i.e. Kir1 to Kir7) and subfamilies (i.e. Kir2.1 to Kir2.4), and their variable distribution in different vascular preparations implies a high level of heterogeneity (Quayle et al., 1997). The pore-forming ␣-subunit has a unique membrane topology of only two transmembrane regions (M1 and M2) leading to cytoplasmic amino and carboxy termini, and a central pore-forming loop (Figure 1.2). Four ␣-subunit

4

Kathryn M. Gauthier and Nancy J. Rusch Inward rectifier (Kir)

Voltage-gated (Kv)

α NH3+

α COO–

+

NH3

COO–

β

Calcium-sensitive (KCa) +

β

BKCa

NH3

IKCa

α

α

+

NH3 COO–

SKCa

α CBD COO–

NH3+

CBD COO–

Figure 1.2 Proposed topology of K⫹ channels expressed in vascular endothelial cell or smooth muscle cell membranes. CBD ⫽ calmodulin binding domain.

proteins from the same gene family assemble to form a tetrameric structure that possesses the fundamental channel properties of Kir channels including K⫹ permeability and inward rectification. The inward rectification is at least partially conferred by channel block from intracellular Mg2⫹ and polyamines at specific residues located in the M2 domain and carboxyl-terminal regions of the Kir channels (Yang et al., 1995). The absence of selective pharmacological blockers for the Kir channels has hampered efforts to define their precise contribution to the membrane potential of endothelial and vascular smooth muscle cells. The Kir channels are blocked by micromolar concentrations of barium (Ba2⫹). Although this inorganic cation is used as a “selective” inhibitor of the Kir channel, it also interferes with other ionic processes at somewhat higher concentrations. With this caveat in mind, Ba2⫹-sensitive Kir channels have been described in endothelial cells from a number of arterial sites (Nilius et al., 1997). For example, in freshly isolated guinea-pig coronary endothelial cells, the expression of 50–60 strongly rectifying Kir channels with a unitary conductance of 26 pS determines the resting membrane potential to a large extent (von Beckerath et al., 1996). Similarly, Kir channels contribute importantly to the resting membrane potential of vascular smooth muscle cells from autoregulatory beds such as the coronary and cerebral circulations (Quayle et al., 1993; Robertson et al., 1996). The Kir channels in vascular smooth muscle cells are regarded as potential metabolic sensors, because their single-channel conductance increases as a function of the extracellular K⫹ concentration. Thus, small elevations in extracellular K⫹ concentration may enhance K⫹ efflux through Kir channels resulting in a Ba2⫹-sensitive hyperpolarization and relaxation in some types of vascular smooth muscle cells (Zaritsky et al., 2000). Although the molecular composition of Kir channels in endothelial cells is unknown, the Kir2.1 gene subfamily appears to provide native Kir current in vascular smooth muscle cells. Transcript for Kir2.1 but not Kir2.2 or Kir2.3 channels has been detected in rat mesenteric, cerebral, and coronary arteries, and the properties of cloned Kir2.1 channels closely resemble those of the native Kir channels (Bradley et al., 1999). Furthermore, a recent study in cerebral

Vascular K⫹ channels

5

smooth muscle cells from Kir2.1 null mice revealed a lack of Kir current and K⫹-induced dilatations, indicating that Kir2.1 gene expression is necessary for normal Kir channel function in these arterial cells (Zaritsky et al., 2000). Considering that the accumulation of endothelium-derived K⫹ in the myoendothelial space has been proposed as an endotheliumdependent hyperpolarizing factor, the Kir2.1 channels in vascular smooth muscle may represent the ion channels that mediate arterial relaxation in response to elevated levels of extracellular K⫹.

3.2. Voltage-gated K⫹ channels The voltage-gated K⫹ (Kv) channels represent a diverse superfamily of K⫹ channels that share the common properties of K⫹ selectivity, voltage-dependent activation, and sensitivity to block by 4-aminopyridine. Structural homology has classified Kv channels into nine gene families (Kv1–Kv9) that share a common ␣-subunit topology (Coetzee et al., 1999). Structurally, the Kv ␣-subunits contain six hydrophobic transmembrane segments (S1–S6) flanked by hydrophilic amino- and carboxyl-terminal sequences located in the cell interior (Figure 1.2). Transmembrane domain S4 possesses an intrinsic voltage sensor that is associated with channel gating. Because the channel pore represents a heterotetrameric ␣-subunit complex composed of similar or different ␣-subunits originating from the same gene family (see related inset), the “mixing and matching” of ␣-subunits provides for a heterogeneous population of Kv channels. Voltage-sensitivity and channel inhibition by 4-aminopyridine, the classical Kv channel blocker, are inherent to the tetramer complex. An additional level of complexity is introduced by the Kv ␤-subunits, which may interact with the cytoplasmic domains of the ␣-subunit to enhance channel inactivation or modify cell-surface expression (Hulme et al., 2001). Significantly, some Kv channels are sensitive to blockade by drugs that are sometimes regarded as specific blockers of other K⫹ channel gene families. For example, low millimolar concentrations of tetraethylammonium are often used to preferentially block high-conductance Ca2⫹-sensitive K⫹ (BKCa) channels, but these concentrations of tetraethylammonium also effectively block Kv1.3 channels. Similarly, at least several subtypes of the Kv1 gene family, including Kv1.2, 1.3, and 1.5, are blocked by nanomolar concentrations of charybdotoxin, a drug that is often used to block intermediate-conductance Ca2⫹-sensitive K⫹ (IKCa) channels (Coetzee et al., 1999). As the pharmacological properties of the Kv channels are highly diverse, pharmacological studies must carefully evaluate if blocking drugs directed against other K⫹ channel gene families also have a joint effect on functions mediated by Kv channel subtypes. The Kv channels necessarily contribute to the resting membrane potential and tone in the vascular smooth muscle cells of resistance arteries including those of the cerebral, coronary, mesenteric, and pulmonary vascular beds (Berger and Rusch, 1999). However, because the biophysical and pharmacological properties of the Kv channels reflect their hybrid ␣-subunit composition, patch-clamp studies have struggled to identify the ␣-subunits that compose the major Kv channel types in vascular smooth muscle cells. Some Kv channels in vascular smooth muscle cells appear to emanate from the Kv1 (Shaker-related) gene family, but transcripts for the Kv2, Kv3, and Kv4 family members also have been detected (Xu et al., 1999; Cox et al., 2001). Studies in other tissues indicate that the Kv2 ␣-subunits may form tetramers with Kv5–Kv9 family members to modify Kv2 channel properties (Coetzee et al., 1999). Considering this complexity, its not surprising that variable unitary conductances have been reported for Kv channels in vascular smooth muscle cells, and that a complex

6

Kathryn M. Gauthier and Nancy J. Rusch

pattern of site-specific expression and coassembly of Kv channels is thought to confer electrical heterogeneity within and between different circulatory beds. Kv channels sensitive to block by 4-aminopyridine also have been reported in endothelial cells from several vascular beds, although the molecular identity of these endothelial Kv channels remains unclear. For example, freshly isolated endothelial cells from resistancesized rat pulmonary arteries show high densities of Kv current and express Kv1.5 ␣-subunits (Hogg et al., 1999). Whole-cell and single-channel Kv currents also have been recorded from freshly isolated coronary capillary endothelial cells from the guinea-pig heart and represent the prominent K⫹ current in these cells (Dittrich and Daut, 1999). Although a functional role for Kv channels in regulating the resting membrane potential of endothelial cells has not been convincingly demonstrated, the observation that blockade of Kv channels by 4-aminopyridine partially inhibits acetylcholine-induced hyperpolarizations in intact endothelial cells of guinea-pig carotid arteries infers that Kv channels may modulate membrane potential in some endothelial cells (Quignard et al., 2000). Thus, Kv channels appear to be expressed in the plasma membranes of endothelial as well as vascular smooth muscle cells, and potentially may provide a hyperpolarizing influence in both cell types.

3.3. Calcium-sensitive K⫹ channels The Ca2⫹-sensitive K⫹ (KCa) channels represent a superfamily of channels that have the common property of Ca2⫹-sensitive activation, but exhibit different amino acid sequences, kinetics, single-channel conductances, and pharmacological properties. These channels are broadly divided into three main classes based on their single-channel conductances. High-conductance KCa channels, often called “BKCa channels” because of their big amplitude currents, exhibit unitary conductances of 200–300pS in symmetrical K⫹ solutions and are densely expressed in vascular smooth muscle cell membranes. In the same symmetrical K⫹ solutions, smallconductance KCa channels (SKCa) exhibit low unitary conductances of 5–20pS, whereas intermediate-conductance KCa channels (IKCa) exhibit unitary conductance values between 20–80pS. In the blood vessel wall, the IKCa and SKCa channels appear to be highly expressed in the endothelial cell membranes (Sage and Marchenko, 2001; Triggle, 2001). The membrane topology, primary amino acid sequences and the molecular mechanisms for Ca2⫹-sensitivity differ between BKCa versus IKCa and SKCa channels (Figure 1.2). The pore-forming ␣-subunit of the BKCa channel shows partial homology with the Kv channel ␣-subunit in six (S1–S6) of its seven (S0–S6) transmembrane domains including the S4 voltage-sensing region that mediates voltage-sensitivity. However, the Ca2⫹-sensitivity of the BKCa channel appears to be conferred by four additional hydrophobic domains (S7–S10) in the C-terminal region of the ␣-subunit, and by close association with ␤-subunits that may interact with the unique extracellular N-terminus of the S0 transmembrane domain (Tanaka et al., 1997). Unlike the Kv channels that originate from multiple gene families, BKCa channels appear to arise from only one or two gene families although phenotypic diversity may be generated by a high level of alternative splicing of the common primary transcript. Iberiotoxin, a scorpion toxin, selectivity blocks BKCa channels when applied to the external membrane surface in nanomolar concentrations (Galvez et al., 1990). In contrast, the ␣-subunits of the IKCa and SKCa channels exhibit only six transmembrane domains flanked by intracellular amino- and carboxyl termini; it is likely that the channels also assemble as tetramers (Figure 1.2). In this respect their general topology is similar to the Kv channels, but their primary amino acid sequences are very different except for common residues in the pore-forming region between S5 and S6 (Vergara et al., 1998).

Vascular K⫹ channels

7

Three subtypes (SK1, SK2, SK3) of the SKCa channel family have been cloned which show variable sensitivity to block by the bee venom apamin and to D-tubocurarine. All three SKCa channel subtypes differ from the BKCa channels in a number of key phenotypic features: (a) Their single-channel conductances are 10- to 50-fold less than the BKCa channels; (b) they lack significant voltage-sensitivity; (c) SKCa channels are sensitive to block by pico- to nanomolar concentrations of apamin whereas BKCa channels are inhibited by iberiotoxin; and (d) SKCa channels are highly sensitive to activation by intracellular Ca2⫹ (concentration for half-maximal activation ≈ 400nM), whereas BKCa channels are primarily gated by voltage at low intracellular Ca2⫹ concentrations. The structural features responsible for Ca2⫹-sensitivity appear to differ between BKCa and SKCa channels. Whereas the Ca2⫹-sensitivity of the BKCa channel is conferred by four hydrophobic domains (S7–S10) in the C-terminus of the ␣-subunit and by close association with ␤-subunits, the SKCa channel ␣-subunits appear to possess a calmodulin-binding domain in the proximal C-terminus that constitutively interacts with calmodulin to permit Ca2⫹-activation of the SKCa channel (Bond et al., 1999). Finally, IKCa channels show a similar membrane topology as SKCa channels, and share many phenotypic features including voltage-insensitivity and constitutive calmodulin-mediated Ca2⫹-sensitivity. However, their higher single-channel conductance range of 20–80 pS and their sensitivity to block by charybdotoxin and clotrimazole distinguishes them from the SKCa channel family (Vergara et al., 1998). To judge from studies in freshly isolated or intact endothelial and vascular smooth muscle cells, the KCa channel families appear to be expressed in a cell-specific, or at least a cellpreferential manner, in the blood vessel wall. The BKCa channel is densely expressed in vascular smooth muscle cells, and contributes to the resting membrane potential and tone of cerebral, coronary, and pulmonary small arteries (Berger and Rusch, 1999). At normal intraluminal pressures, membrane depolarization and intracellular calcium act synergistically to activate BKCa channels, and the small size and high input resistance of the vascular smooth muscle cells coupled to the large unitary amplitudes of BKCa currents provide a powerful hyperpolarizing influence. In contrast, there is little evidence for the expression of BKCa channels in freshly isolated endothelial cells. For example, freshly isolated vascular smooth muscle cells from bovine coronary arteries express BKCa ␣-subunits and show iberiotoxinsensitive K⫹ currents (Figure 1.3). However, BKCa ␣-subunits are not detected in freshly isolated endothelial cells from the same arteries and these cells also lack iberiotoxin-sensitive currents (Figure 1.3). Reports in cultured endothelial cells indicating the presence of BKCa currents may relate to the induction of this channel by culture conditions (Jow et al., 1999), although in intact endothelial cells of porcine aorta, an iberioxotin-sensitive K⫹ current has been identified (Papassotiriou et al., 2000). In contrast to the preferential expression of BKCa channels in vascular smooth muscle cell membranes, the expression of IKCa and SKCa channels appears to be primarily limited to endothelial cells. For example, mRNA and protein for SKCa ␣-subunits have been detected in the intact endothelial cell layer, but not the smooth muscle-containing medial layer, of porcine coronary artery (Burnham et al., 2002). Ample evidence also exists for apaminsensitive SKCa currents and charybdotoxin-sensitive IKCa currents in patch-clamped endothelial cell membranes but not in vascular smooth muscle cell membranes (Sage and Marchenko, 2001; Triggle, 2001). Indeed, the properties of SKCa and IKCa channels indicate that these two types of K⫹ channels are well suited to promote the release of endotheliumderived factors from endothelial cells, a process that occurs optimally at negative levels of membrane potential. First, because SKCa and IKCa channels are insensitive to voltage, they can effectively open at negative membrane potentials to promote hyperpolarization and the

8

Kathryn M. Gauthier and Nancy J. Rusch A

Smooth muscle cells α-actin

B

BKCa

Smooth muscle cells

600

Iberiotoxin (10–7 M) 600

300

300

Control

0 (pA)

0 (pA) C

Endothelial cells Pecam

Endothelial cells

D BKCa

Iberiotoxin (10–7 M)

Control 60

60

30

30

0 (pA)

0 (pA)

(200 msec)

Figure 1.3 (A) Single smooth muscle cells isolated from bovine coronary arteries express the ␣-subunit of the BKCa channel. Smooth muscle cell identity was confirmed using an antibody directed against smooth muscle-specific ␣-actin. In similar cells, BKCa ␣-subunit expression was detected using an antibody directed against the S9–S10 linker of the BKCa channel (Liu et al., 1998). Nuclei are indicated in blue using 4,6-diamidino-2-phenylindole. (B) Patch-clamped smooth muscle cells show iberiotoxin (IBTX)-sensitive K⫹ currents. Whole-cell K⫹ currents were generated by depolarizing pulses (8mV increments) from⫺70mV to ⫹58 mV. (C) The identity of endothelial cells isolated from bovine coronary arteries was confirmed by the expression of the endothelial marker, PECAM. The expression of the BKCa ␣-subunit was not detected in similar endothelial cells. (D) Iberiotoxin-sensitive K⫹ currents were not observed in patch-clamped endothelial cells. Currents were generated as described in (B) (see Color Plate 1). Source: Adapted from Gauthier et al. (2002).

electrical gradient for Ca2⫹ influx in endothelial cells that enhances factor release. Second, because the opening of SKCa and IKCa channels is highly sensitive to intracellular calcium, Ca2⫹ influx in response to the activation of endothelial cells will enhance the open-state probabilities of these channels to further promote hyperpolarization and the Ca2⫹-dependent synthesis and release of endothelium-derived factors.

4. REMODELING OF VASCULAR K⫹ CHANNELS IN DISEASE In vivo, the distinct populations of K⫹ channels expressed in plasma membranes of endothelial and vascular smooth muscle cells act in concert to regulate the diameter of small arteries and arterioles. However, findings from several laboratories indicate that the molecular composition of these vascular K⫹ channels is regulated dynamically by the local environment, resulting in a “remodeling” of K⫹ channels in response to environmental abnormalities. From a therapeutic perspective, “disease-specific” profiles of K⫹ channels have been identified in endothelial or vascular smooth muscle cells during pathologies such as pulmonary and systemic hypertension, and during hyperproliferative diseases. Under these conditions, the

Vascular K⫹ channels

9

expression level of K⫹ channels is altered, and in some instances, K⫹ channel genes are expressed that are normally quiescent. These findings infer that designing new channelbased therapies based on the K⫹ channel profiles expressed in normal endothelial or smooth muscle cell membranes is of questionable value. Instead, the development of new therapeutic agents may depend on identifying the K⫹ channels in the endothelial and vascular smooth muscle cells that are highly expressed and functionally important in a given disease state, and then designing drugs to enhance or inhibit their activity.

4.1. Evidence for K⫹ channel remodeling in endothelial cells Several recent studies have reported changes in the levels of membrane potential or K⫹ channel expression in vascular endothelial cells during cardiovascular diseases, providing initial evidence that K⫹ channel remodeling occurs in endothelial cells during pathological conditions. For example, an altered expression of KCa channels has been reported in mesenteric endothelial cells of patients with colonic adenocarcinoma, inferring altered K⫹ channel profiles in endothelial cells associated with tumor angiogenesis (Kohler et al., 2000). In this study, membrane patches from endothelial cells of cancer patients showed more IKCa currents than similar cells from control patients (Figure 1.4). Furthermore, BKCa currents were absent in mesenteric endothelial cell membranes of control patients, but were detected in those of cancer patients (Figure 1.4), suggesting a disease-related induction of the BKCa channel gene in this abnormal environment. In rat carotid arteries after balloon angioplasty, a loss of SKCa and IKCa channels in the regenerated endothelium may account for the impaired acetylcholine-induced hyperpolarization and dilatation observed in the ballooned arteries (Kohler et al., 2001). In this study, whole-cell K⫹ currents associated with the SKCa and IKCa channels were attenuated in the regenerated endothelial cells, and the levels of mRNA encoding the SKCa and IKCa channels also were reduced. Other evidence that endothelium-derived hyperpolarizing factors promote proliferation of endothelial and vascular smooth muscle cells by activating growth-related kinase pathways strengthens the concept that endothelial K⫹ channels may be linked to cell proliferation (Fleming et al., 2001). Electrophysiological studies in coronary endothelial cells also have implicated altered K⫹ channels in hypertension. For example, perforated patch-clamp techniques have demonstrated that the resting Em of endothelial cells in small coronary arteries from hypertensive rats is depolarized compared to cells of normotensive animals, suggesting a down-regulation of the expression or activity of K⫹ channels (Gauthier and Rusch, 2001). However, at present, its unclear if this abnormality attenuates the release of relaxing factors in the coronary arteries of the hypertensive animals as would be predicted if the electrical gradient for Ca2⫹ influx was reduced in the affected endothelial cells.

4.2. Evidence for K⫹ channel remodeling in vascular smooth muscle cells Abnormal K⫹ channel expression in vascular smooth muscle cells also appears to be involved in the pathogenesis of vascular disorders. The best characterized of these “diseasespecific” K⫹ channel profiles is the reduced expression of Kv channels in the pulmonary smooth muscle cells of rats exposed to chronic hypoxia to induce pulmonary hypertension. In the pulmonary smooth muscle cells of these animals, mRNA and protein for the Kv1.2 and Kv1.5 channel subtypes are rapidly down-regulated in response to hypoxic conditions, resulting in depolarization of the vascular smooth muscle cells, vascular hyperreactivity, and

10

Kathryn M. Gauthier and Nancy J. Rusch A

B

IKCa currents 1 × 10–7 4 × 10–7 1 × 10–6

%IKCa – positive endothelial cell

1 × 10–8

2 × 10–3 3 pA 250 ms

30 20 10 0

Cancer

D

BKCa currents 1 × 10

20

–8

4 × 10–7 3 × 10–6 6 × 10–6 2 × 10–5

15 10 5 n.d. 0

5 pA 250 ms

Control

[Ca2+], M

%BKCa – positive endothelial cell

C

40

[Ca2+], M

Control

Cancer

Figure 1.4 (A) IKCa currents showing Ca2⫹-dependent activation recorded in inside-out patches of endothelial cells from human mesenteric arteries at a patch potential of ⫺80 mV. (B) IKCa currents were observed in a higher percentage of endothelial cells from patients with adenocarcinoma than in endothelial cells from control patients. (C) BKCa currents showing Ca2⫹-dependent activation recorded in inside-out patches of endothelial cells from human mesenteric arteries at a patch potential of ⫹80 mV. (D) BKCa currents were only observed in mesenteric endothelial cells from cancer patients and were not observed in the endothelial cells from control patients. Values are mean ⫾ SE. *⫽P⬍0.05, Mann-Whitney U test. n.d.⫽not detected. Source: Reproduced with permission from Kohler et al., 2000.

pulmonary hypertension (Wang et al., 1997). An abnormally low expression level of the same Kv channels has been reported in patients with primary pulmonary hypertension (Yuan et al., 1998). From the therapeutic perspective, the loss of Kv channels in these vascular smooth muscle cells during pulmonary hypertension implies that endotheliumderived dilator factors that act through Kv channels will be less effective in mediating vasodilatation in the diseased arteries. Furthermore, the depolarized state of the diseased pulmonary smooth muscle cells may affect the efficacy of endothelium-derived hyperpolarizing factor (EDHFs) with regard to their dilator action. Several reports also indicate that Kv channels are down-regulated in vascular smooth muscle cells during systemic hypertension, but in this disease, an upregulation of the BKCa channels appears to partially compensate for this loss. The physiological importance of the compensatory overexpression of BKCa channels in this disease has been demonstrated in the cerebral circulation of spontaneously hypertensive rats. In this preparation, the pore-forming ␣-subunits of the BKCa channel are upregulated in the cerebrovascular smooth muscle cells of the spontaneously hypertensive rats compared to normotensive Wistar Kyoto (WKY) rats,

Vascular K⫹ channels A

B

11

Control Iberiotoxin

SHR

WKY 10 pA

WKY

BKCa 125 kDa

Control

100 ms

Iberiotoxin SHR

C

WKY

SHR

Figure 1.5 (A) Western blot showing the upregulation of the BKCa ␣-subunit (125 kD) in cerebral smooth muscle cells of spontaneously hypertensive rats (SHR) compared to normotensive Wistar Kyoto (WKY) rats. (B) Cerebrovascular smooth muscle cells of SHR rats showed a larger component of iberitoxin-sensitive current than similar cells from normotensive WKY rats. (C) The effect of topical application of 10⫺7 M iberiotoxin on the diameters of WKY (left frames) and SHR (right frames) pial arterioles. The selective blockade of BKCa channels by iberiotoxin only mildly constricted the WKY arterioles but severely constricted the SHR arterioles. Source: Reproduced with permission from Liu.

and this results in a higher component of iberiotoxin-sensitive BKCa current in the vascular smooth muscle cells of the hypertensive animals (Figure 1.5) (Liu et al., 1998). The cerebral microcirculation of the spontaneously hypertensive rat appears to profoundly rely on abundant BKCa channels to prevent anomalous vasoconstriction during chronic hypertension, because pharmacological blockade of BKCa channels by iberiotoxin triggers vasospasm in cerebral arterioles of these animals whereas the arterioles from normotensive WKY rats only mildly constrict (Figure 1.5). These findings support the concept that different K⫹ channel types regulate the resting membrane potential of vascular smooth muscle cells in normal and disease states. Similarly, the types of K⫹ channels in the vascular smooth muscle cells that are available to mediate the vasodilator actions of endothelial factors may be dramatically altered during cardiovascular pathologies, including common diseases such as systemic hypertension.

5. Summary In vivo, the level of arterial tone relies on the resting membrane potential of the vascular smooth muscle cells that, in turn, are dynamically modulated by vasoactive factors released

12

Kathryn M. Gauthier and Nancy J. Rusch Endothelial cells SKCa

Smooth muscle cells

IKCa

Kir

Kv

BKCa

EDHF Kir

Kv

Figure 1.6 Distinct profiles of K⫹ channels may mediate the release of endothelium-derived hyperpolarizing factor (EDHF) by endothelial cells, and also represent the effector targets for EDHF on vascular smooth muscle cells.

from the endothelial cells. In this regard, distinct profiles of K⫹ channels appear to have evolved in endothelial and vascular smooth muscle cells to mediate their vasodilator pathways. Endothelial cells minimally express populations of Kir, Kv, SKCa, and IKCa channels (Figure 1.6). The activation of the SKCa and IKCa channel families may be particularly involved in the release of dilator factors from endothelial cells. In contrast, vascular smooth muscle cells appear to rely on Kv and BKCa channels to regulate the resting membrane potential, whereas Kir channels may contribute to the resting membrane potential and mediate K⫹-induced dilatations in the vascular smooth muscle cells of smaller arteries. The relationship between membrane potential and cell excitability is probably steep in most excitable cells, which indicates a tight relationship between K⫹ channel opening, membrane potential, and the influx of activator Ca2⫹. Thus, small changes in K⫹ channel activity may profoundly affect the Ca2⫹-dependent synthesis and release of vasodilator factors in endothelial cells, and also may precisely modulate voltage-gated Ca2⫹ influx and the level of contraction in vascular smooth muscle cells. New findings indicating that K⫹ channel expression is altered in endothelial and vascular smooth muscle cells during hypertension and proliferative diseases suggest that identifying the K⫹ channel profiles expressed in the affected cells may lead to the development of new vasoactive therapies appropriately targeted in a cell-specific manner to highly expressed and functionally relevant K⫹ channel types. ACKNOWLEDGMENTS The authors appreciate the administrative and graphics support of Mr. Miodrag Pesic. Research in the author’s laboratories was supported in part by R01 HL-59238 and R01 HL-68406 from the USA National Institutes of Health.

2

Possible contribution of CLCA1 to calcium-activated chloride channels in murine smooth muscle cells I.A. Greenwood, F. Britton, S. Ohya and B. Horowitz

Smooth muscle cells actively accumulate chloride ions (Cl⫺), therefore activation of chloride channels produces Cl⫺ efflux and membrane depolarization. Ca2⫹-activated Cl⫺ currents (ICl(Ca)) have been recorded in numerous smooth muscle cells and have been shown to exhibit various consistent characteristics that include time- and voltage-dependent kinetics. In comparison to the information available on native ICl(Ca) there is little known about the molecular composition of the channel underlying this current although a family of candidate genes (termed CLCA) has been identified. The present study investigated whether or not CLCA genes were responsible for ICl(Ca) in murine smooth muscle cells. Voltage-clamp experiments on acutely dispersed smooth muscle cells showed that these cells exhibited ICl(Ca) with characteristics similar to ICl(Ca) recorded in other smooth muscle cells. RT-PCR of RNA isolated from single smooth muscle cells revealed that only mCLCA1 was expressed. Comparison of the native ICl(Ca) with mCLCA1 expressed in HEK cells showed that there were marked differences in Ca2⫹ sensitivity and kinetics but the relative permeability of various anions through the respective channels was comparable. These data show that CLCA1 alone does not constitute the native ICl(Ca) in murine portal vein myocytes.

1. INTRODUCTION Ca2⫹-activated Cl⫺ currents (ICl(Ca)) are found in a variety of smooth muscle cells from a number of different species (Large and Wang, 1996). As the equilibrium potential for chloride (ECl) is more positive than the resting membrane potential in smooth muscle cells the physiological consequence of the activation of ICl(Ca) in smooth muscle is membrane depolarization and subsequent contraction in response to excitatory agents (Large and Wang, 1996). Ca2⫹-activated Cl⫺ currents in smooth muscle tissues exhibit a number of common features, namely a sensitivity to pharmacological blockers such as niflumic acid and stilbene derivatives (Large and Wang, 1996), low threshold for activation by Ca2⫹ and the existence of voltage-dependent kinetics. This latter point is highlighted by studies whereby the ICl(Ca) in smooth muscle cells has been evoked by pipette solutions containing a fixed concentration of free Ca2⫹ (Greenwood et al., 2001; Piper et al., 2002). Using this technique a sustained ICl(Ca) is activated at the holding potential of ⫺50 mV and depolarization produces an initial outward current followed by a slowly developing outward relaxation. Subsequent repolarization leads to a slowly declining inward current the time course of which is voltagedependent. These voltage-dependent kinetics were identical to those exhibited by ICl(Ca) evoked by this technique in secretory and endothelial cells (Arreola et al., 1996; Nilius et al., 1997b). Although ICl(Ca) has been recorded in many smooth muscle cells there has been no information as to the molecular identity of the underlying channel. Several members of the

14

I.A. Greenwood et al.

CLCA gene family, including the murine isoform mCLCA1, have been expressed functionally in mammalian cells and elicit outwardly rectifying membrane currents that are anion-selective (Fuller and Benos, 2000; Fuller et al., 2001) and mCLCA4 exists in tissues rich in smooth muscle cells (Elble et al., 2002). The aim of the present study was to determine if murine smooth muscle cells that exhibit ICl(Ca) express mCLCA genes and whether the products of these genes could constitute the native channel.

2. METHODS Native ICl(Ca) and RNA extraction were performed on acutely dispersed smooth muscle cells isolated from the portal veins of adult Balb-c mice. After dissection and removal of fat and connective tissue segments of the vein were incubated in physiological salt solution containing 3 mg ml⫺1 collagenase (type 1A), 2 mg ml⫺1 trypsin inhibitor, 2 mg ml⫺1 bovine serum albumin and 0.1 mg ml⫺1 protease (type XIV) for 10 min at 36 ⬚C. Following digestion isolated smooth muscle cells were liberated by passing the tissue through a wide-bore Pasteur pipette and stored at 4 ⬚C in 50 ␮M Ca2⫹ containing physiological salt solution. Cells were transferred to the stage of a phase-contrast microscope and either allowed to adhere for 15 min for electrophysiological recording or were collected for RNA extraction by aspirating them into a wide-bore borosilicate pipette and the pipette contents ejected into sterile 0.5 ml tubes. Approximately 60 muscle cells were collected, flash-frozen in liquid nitrogen, and stored at ⫺80 ⬚C until use.

2.1. Total RNA isolation and reverse transcriptase polymerase chain reaction Total RNA was isolated from murine tissue and enzymatically dispersed smooth muscle and cardiac cells using the SNAP Total RNA isolation kit (Invitrogen, Carlsbad, CA), following the manufacturers instructions, including the use of polyinosinic acid (20 ␮g) as an RNA carrier. First-strand cDNA was prepared from the RNA using the Superscript IITM Reverse Transcriptase kit (Life Technologies Inc., Rockville, MD). One ␮g of total RNA was reverse transcribed with 200 units reverse transcriptase in a 20 ␮l reaction containing 25 ng Oligo dT(12–18) primer, 500 ␮M each dNTP, 50 mM Tris–HCl, pH 8.3, 75 mM KCl, 3 mM MgCl2 and 10 mM dithiothreitol (DTT). Reverse transcriptase polymerase chain reaction (RT-PCR) amplification was performed with gene specific primers for mCLCA1-4 and ␤-actin on 10␮l cDNA using AmpliTaq Gold reagents (PE Applied Biosystems, Foster City, CA). RT-PCR with ␤-actin primers controlled for genomic DNA contamination in the source RNA, since these primers were designed to span two exons and an intron. The no-template control, a PCR reaction in which no template was added, controlled for primer contamination and non-specific amplification. PCR products generated from each pair of primers were gelextracted and sequenced to confirm the specificity of the primers. The mCLCA1 channel was cloned from murine portal vein tissue by PCR amplification with specific primers designed against the published sequence for mCLCA1 (GenBank accession # AF047838; Gandhi et al., 1998). mCLCA1 was first amplified as two fragments which span the entire coding region and contained an overlapping region of 301 bp. These fragments were then used as the template in an overlap extension PCR reaction to obtain the complete coding region of mCLCA1 (Britton et al., 2002).

Contribution of CLCA1 to calcium-activated chloride channels

15

2.2. Electrophysiological experiments CLCA1 isolated from murine portal vein was stably transfected into HEK293 cells by the calcium phosphate co-precipitation method followed by Geneticine (Life Technologies Inc., Rockville, MD) antibiotic selection. Transfected and untransfected HEK293 cells were seeded on to shards of glass 24 h before recording. In experiments on native cells, IClCawas evoked by pipette solutions containing 100 nM, 500 nM or 1 ␮M Ca2⫹ using a pipette solution of the following composition (mM): tetraethyl ammonium-Cl (20), CsCl (106), 4-(2-hydroxy-ethyl)-1-piperazine ethanesulphonic acid (5), 1,2-bis(2-aminophenoxy)ethane-tetra-acetic acid (10), MgATP (3), GTP.diNa (0.2), MgCl2 (0.42) and pH was set to 7.2 by addition of CsOH. [Ca2⫹] was buffered by adding the appropriate amount of CaCl2 determined by a calcium buffer program (EQCAL by Biosoft, Ferguson, MO). Currents were evoked in transfected HEK cells with either a pipette solution containing 500 nM Ca2⫹ (described above) or 2 mM Ca2⫹ used in the studies by Gandhi et al. (1998) and Gruber et al. (1999). The composition of this pipette solution was (mM): N-methyl-D-glucamine-Cl (126), sucrose (30), 4-(2-hydroxy-ethyl)-1-piperazine ethanesulphonic acid (5), mgCl2 (2) and CaCl2 (2). In all experiments the external solution used to bathe the cells had the following composition (mM): NaCl (126), 4-(2-hydroxyethyl)-1-piperazine ethanesulphonic acid (10), glucose (20), CaCl2 (1.8), MgCl2 (1.2), tetraethyl ammonium-Cl (10) and pH was set to 7.2 with 10 M NaOH.

3. RESULTS The expressions of members of the murine CLCA family were examined by performing RT-PCR on isolated muscle cells from several murine tissues. PCR products were generated using gene-specific primers for mCLCA1, mCLCA3 and mCLCA4 that were designed to be non-homologous to other members of the CLCA family of genes. The mCLCA2 isoform; GenBank # NM017474 (Komiya et al., 1999) is 97% identical to mCLCA1 at the nucleotide level; therefore the mCLCA1 primers will not differentiate between these two isoforms. RT-PCR with ␤-actin primers was performed on all isolated smooth muscle cell preparations as both a control for the integrity of the cDNA and as a control for genomic DNA contamination, since these primers were designed to span an intron as well as two exons. The resulting PCR experiment revealed a 498 bp amplicon indicative of non-genomic DNA. RT-PCR with primers specific for mCLCA1 resulted in detectable amplicons at 105 bp. However, the expression of mCLCA3 or mCLCA4 was not detected in any of the isolated muscle cell preparations tested, even with 30 additional cycles of PCR whereas amplicons to mCLCA3 were detected when RT-PCR was performed on RNA isolated from whole tissue. These data show that murine portal vein smooth muscle cells express mCLCA1 only.

3.1. Ca2⫹-activated Cl⫺ currents in murine portal vein myocytes and HEK cells expressing mCLCA1 Few experiments have been performed on murine smooth muscle cells and therefore it is not known whether smooth muscle cells from murine portal vein exhibit Cl⫺ currents with characteristics consistent with ICl(Ca) recorded in other species. To allow the comparison of data from murine smooth muscle cells with that obtained in other species (e.g. rabbit portal vein, Greenwood et al., 2001) currents were activated by pipette solutions where the Ca2⫹ was buffered at concentrations between 100 nM and 1 ␮M. Whilst 100 nM failed to evoke

16

I.A. Greenwood et al. A-1

B

A-2

Figure 2.1 Characteristics of ICl(Ca) evoked in murine portal vein myocytes (A-1), an example of a family of currents evoked by a pipette solution containing 500 nM Ca2⫹ at voltages between ⫺100 mV and ⫹120 mV (voltage protocol is shown in (A-2)). Superimposed lines show the exponential fit for the outward relaxation at ⫹120 mV and the corresponding tail current recorded at ⫺60 mV. The time constants (␶) for these fits are included. Dashed line represents the zero current level. Panel B shows the mean of the late current densities respectively evoked by 100 nM (open triangle), 500 nM (filled circle) and 1 ␮M Ca2⫹ (open circle). Each point is the mean of between 6 and 10 cells with error bars representing s.e.m. Source: Reproduced and modified with permission by The Journal of Physiology from Britton et al. (2002).

any currents that could be distinguished from leak currents pipette solutions containing 500 nM or 1 ␮M Ca2⫹ generated currents with characteristics identical to ICl(Ca) recorded previously in rabbit smooth muscle cells (Figure 2.1). Upon rupture of the membrane to achieve whole cell with a pipette solution containing 500 nM Ca2⫹ elicited a sustained current at ⫺60 mV of ⫺20 ⫾ 3 pA (n ⫽ 11) and step depolarization to positive potentials produced a characteristic outward relaxation that could be fitted by a single exponential. Upon repolarization to the holding potential a slowly declining inward current was evoked. In normal Cl⫺ containing external solutions, the direction of the tail current reversed close to the theoretical Cl⫺ equilibrium potential (mean Erev was ⫹5 ⫾ 1 mV, n ⫽ 9) and replacement of the external solution for one containing NaSCN instead of NaCl shifted Erev to ⫺44 ⫾ 1.5 mV (n ⫽ 9). Pipette solutions containing 500 nM Ca2⫹ did not evoke a current in HEK cells stably transfected with mCLCA1 cloned from murine portal vein (Figure 2.2). Cl⫺ currents were evoked by pipette solutions used by Gandhi et al. (1998) and Gruber et al. (1999) that contained 2 mM Ca2⫹ in transfected but not untransfected HEK cells (Figure 2.2). The currents evoked in transfected cells by pipette solution containing 2 mM Ca2⫹ reversed close to the theoretical Cl⫺ equilibrium potential (mean ⫽ ⫺2.8 ⫾ 1.2 mV, n ⫽ 8; Figure 2.3) and the reversal potential was shifted by replacement of the external anion by the more permeable SCN⫺ (mean ⫽ ⫺39 ⫾ 1.5 mV, n ⫽ 7). In comparison to native ICl(Ca) Cl⫺ currents generated by the expression of mCLCA1 did not exhibit any time-dependent kinetics upon stepping to depolarizations. Consequently, there were no inward tail currents upon repolarization. A further

Contribution of CLCA1 to calcium-activated chloride channels A

17

C

B-1

B-2

D

Figure 2.2 Characteristics of ICl(Ca) recorded in HEK293 cells stably transfected with mCLCA1. (A) the lack of mCLCA1 PCR products in untransfected HEK293 cells (lane 1) but the presence of mCLCA1 amplicons in stably transfected HEK 293 cells (lane 2). ␤-actin PCR products were detected in both cell types. Panels (B-1) and (B-2) show families of currents evoked by pipette solutions containing 500 nM Ca2⫹ (B-1) or 2 mM Ca2⫹ (B-2) recorded at potentials between ⫺100 mV and ⫹100 mV from HEK293 cells transfected with mCLCA1 cloned from portal vein. Dashed line represents the zero current level. (C) the mean current density immediately before the end of the test step, evoked by pipette solutions containing 500 nM Ca2⫹ (open diamonds; n ⫽ 6) and 2 mM Ca2⫹ (open triangles, n ⫽ 14) in transfected HEK293 cells. The density of currents evoked by 2 mM Ca2⫹ in untransfected cells is represented by the filled triangles (labelled as UT, n ⫽ 7). (D) the mean current-voltage relationship for currents evoked by 2 mM Ca2⫹ in HEK293 cells transfected with mCLCA1 in external solutions containing 126 mM NaCl (open triangles), 126 mM NaSCN (open circles) or 126 mM Na-isethionate (filled triangles). Each point is the mean of between 5 and 8 cells with error bars representing the s.e.m. Source: Reproduced and modified with permission by The Journal of Physiology from Britton et al. (2002).

difference between the native and expressed currents was the sensitivity to the reducing agent DTT that inhibited hCLCA gene products (Fuller et al., 2001). Application of 10 ⫺3 M DTT for 5 min had no effect on either the amplitude or kinetics of native ICl(Ca) recorded from murine portal vein myocytes whereas the amplitude of currents produced by the heterologous expression of mCLCA1 was inhibited markedly by DTT (Figure 2.3).

4. DISCUSSION Smooth muscle cells isolated from murine portal veins exhibit Ca2⫹-activated Cl⫺currents with distinctive characteristics similar to those reported previously in rabbit vascular myocytes (Greenwood et al., 2001; Piper et al., 2002). RT-PCR of RNA isolated from smooth muscle cells with primers specific for mCLCA 1-4 genes revealed that this cell type expressed only the mCLCA1 isoform. RT-PCR of RNA derived from whole smooth muscle

18

I.A. Greenwood et al. A

B

C

Figure 2.3 DTT inhibits ICl(Ca) recorded in HEK293 cells stably transfected with mCLCA1 but not native ICl(Ca). (A) the amplitude of the current in HEK cells expressing mCLCA1 recorded at ⫹100 mV in the absence and presence of 10⫺3 M DTT. Insets show ensembles of currents recorded at potentials between ⫺100 mV and ⫹120 mV in the absence (at the point denoted by *) and presence of DTT (denoted by **). (B) the amplitude of the instantaneous current (filled circle) and late current (open circle) recorded at ⫹70 mV from murine portal vein myocytes in the absence and presence of 1 mM DTT. Insets show ensembles of currents recorded at potentials between ⫺100 mV and ⫹120 mV in the absence and presence of DTT. (C) the mean normalized amplitude of currents recorded from HEK cells (filled box) or portal vein myocytes (hatched box) after 5 min application of 1 mM DTT (n ⫽ 4). Source: Reproduced and modified with permission by The Journal of Physiology from Britton et al. (2002).

preparations showed that mCLCA3 was also expressed suggesting that this gene is present in a contaminating cell type (Britton et al., 2002). These observations support the putative role of CLCA genes products as chloride channels. Comparison of the characteristics of native ICl(Ca) and heterologously expressed mCLCA1 cloned from murine portal vein myocytes showed that there were a number of differences notably in Ca2⫹-sensitivity and voltage-dependent kinetics between the two currents. However, the relative permeability of thiocyanate and isethionate through the two channels was similar (Britton et al., 2002). These data suggest that mCLCA1 alone does not constitute the native channel protein in vascular smooth muscle cells. CLCA4 has also been considered as a candidate for the native smooth muscle cell Ca2⫹-activated Cl⫺channel (Elble et al., 2002). However, mCLCA4 amplicons were not detected in murine portal vein myocytes by RT-PCR. Moreover, heterologous expression of this gene in HEK cells yielded a Cl⫺ current that did not exhibit the distinctive voltage-dependent kinetics exhibited by the native ICl(Ca) (Elble et al., 2002). Recent studies have investigated whether a combination of mCLCA1 with an auxiliary, nonpore forming protein could underlie the native Ca2⫹-activated Cl⫺ channel in smooth muscle cells. Co-expression of mCLCA1 with the ␤-subunit encoded by KCNMB1 that normally associates with large conductance Ca2⫹-activated K⫹ channels increased the Ca2⫹-sensitivity

Contribution of CLCA1 to calcium-activated chloride channels

19

of mCLCA1 compared to expression of mCLCA1 alone (Greenwood et al., 2002) and a physical protein–protein interaction was established by the use of a mammalian two hybrid assay system. However, the co-expression of mCLCA1 with KCNMB1 did not confer consistently time- and voltage-dependent kinetics on the current (Greenwood et al., 2002). Consequently, the molecular nature of the channel underlying ICl(Ca) in smooth muscle cells is far from established. ACKNOWLEDGEMENTS This work was made possible by the support of all members of Prof B. Horowitz’s laboratory at the Department of Physiology and Cell Biology at the University of Nevada, Reno, NV USA.

3

Trafficking and transduction functions of the Na pump in vascular smooth muscle cells Julius C. Allen, Aslihan Aydemir-Koksoy and Joel Abramowitz

The various effects of the endothelium derived hyperpolarizing functions (EDHFs) include possible interaction with the membrane Na K pump of vascular smooth muscle cells. While the fundamental ionic control mechanisms of the pump appear to be similar to its function in other cells, the actual number of functional pumps in the membrane of a given cell may vary. Thus when cells require an increase in membrane pumps, the initial response is the translocation of preformed pumps from a cytoplasmic pool. Drugs can also activate movement both to and from the membrane and pump content in the membrane is quite dynamic. The pump complex may also act as a transducing protein, initiating a signaling cascade from the sarcolemma to various other cellular compartments, entirely separate from its ionic regulatory role. This signaling function is activated by low concentrations of ouabain at which cellular ionic changes cannot be detected. In vascular smooth muscle cells, this signaling pathway is: Src→EGFR→ERK1/2→proliferation. Smooth muscle cell membranes contain fewer pump sites than other excitable cells, but they still contribute significantly to resting membrane potential. This may be due to multiple population clusters of pumps in the membrane. Caveolae are specialized lipid rafts in the cell membranes, and are identified by their content of caveolin. A variety of proteins demonstrate signaling functions when localized to these specific structures, since caveolin acts as a bridge, bringing proteins into close proximity, to effect “interaction.” Thus the “signaling” function of the Na pump occurs by interaction with Src, via caveolin binding. The “transport” function occurs when the pumps are localized to the bulk membrane compartment. Studies to assess these hypotheses are underway as are a determination of the trafficking between these two membrane phases of pump localization and the cytoplasm, which may be important for the action of EDHF on the pump.

1. INTRODUCTION The various effects of EDHF (endothelium derived hyperpolarizing “function”) involve at least three different “activities”: K⫹, cytochrome P450 arachidonic acid metabolites, or the formation of myoendothelial gap junctions. (See American Journal of Physiology, Heart and Circulation, 280: H2413–H2477, 2001, for details.) While any of the three functions above could potentially induce membrane hyperpolarization, and subsequent relaxation of the vascular smooth muscle cell (vsmc) by action on the Na Pump, it should be recognized that the pump complex in the membrane is a distinctly dynamic entity, constantly turning over, and as such may present a variable profile in different vascular beds. Indeed pump content may well be altered in a given bed depending on a variety of circumstances. Thus, not only may the pump present different characteristics in different beds, its specific function may differ in the same tissue, depending on the specific circumstances in question.

Trafficking and transduction functions of the Na pump

21

The functions of the Na pump can be divided into two general categories: (a) a “responsive” function, in which the pump’s major role is the maintenance of ionic gradients as it responds to various physiologic and pharmacologic perturbations; and (b) a “regulatory” function, in which the pump can act as a transducing protein complex, initiating a signaling cascade which is entirely separate from any ionic changes. The purpose of this brief review is to discuss situations in which both the localization and the function of the pump sites may vary, and therefore present different situations within which the EDHF in question may modulate pump function. It will address three separate topics. The first will be a discussion of the regulation of pump site number in the sarcolemma, as a part of the “responsive” function of the pump. The second will address the pump as a transducing protein complex, a newly identified function totally distinct from its well-known ion regulating role; the “regulatory” function. Finally a unifying model will be proposed, linking these seemingly disparate roles, and relating them to the functions of EDHF.

2. CONTROL OF Na PUMP SITE NUMBER (RESPONSIVE FUNCTION) The number of functional pump sites in vsmc sarcolemma is considerably less than that in other excitable tissues. Thus the capacity of this tissue to regulate ions could be readily compromised by inhibitors of the pump or by drugs that activate a Na influx pathway. In order to respond to these perturbations, and maintain appropriate cellular ionic content, the cell has developed a continuum of response mechanisms involving the pump that can be classified as short term, intermediate term, and long term. This series of controls comprise the “responsive” functions referred to above. Short-term regulation occurs as pump turnover is increased as necessary by PKA or PKC phosphorylation of the ␣ subunit. However, since there are relatively fewer pump sites in the membrane, even with this activation paradigm, the cell’s pumping capacity could be quickly exceeded, and thus would have to recruit additional pumps to the membrane. The mechanism whereby this is accomplished has generally been thought to be the activation of transcription and translation of the ␣ and ␤ subunit genes (long term). However, in vascular smooth muscle cells containing fewer pump sites, it appears that the first line of defense as the cellular pumping capacity approaches its limit is the translocation of preformed pump sites from the cytoplasm to the cell membrane (intermediate term). When canine vascular smooth muscle cells are incubated in a low K medium, which inhibits the pump resulting in an increase in Nai, pump site number in the cell membrane increases as expected (Koksoy and Allen, 2001). However, this increase occurs in the absence of any increase in total ␣ or ␤ subunit protein. In addition, this increase in pump site number, as measured by 3H ouabain binding, was inhibited by the PI3 Kinase inhibitor LY 294002, suggesting cytoplasmic translocation. This same compound inhibited the increase in pump sites in the sarcolemma of rat aortic cells when activated by cyclic stretching (Liu et al., 1998b). Thus the cytoplasmic translocation of preformed pump sites may well be the initial response as the vascular smooth muscle cell increases the number of pumps in the cell membrane, as part of the overall responsive function. Similar cytoplasmic translocation events occur in a variety of epithelial cells, but these translocations have generally been drug induced. Indeed in preliminary experiments, it was also shown that ␤-adrenergic agonists activate a cytoplasmic translocation in canine vsmc (Allen and Bertorello, unpublished). Similar but far more extensive results have been demonstrated with these same drugs in lung epithelial cells (Ridge et al., 2002).

22

Julius C. Allen et al.

Thus, in vsmc, cytoplasmic translocation of preformed pump sites is the first mechanism whereby the cell increases the number of pump sites in the membrane. In other cells grown in a medium which inhibits the pump, and thus increasing Nai, gene transcription of the pump subunits probably occurs more readily (Taniguchi and Kaya, 2000). The identity of a specific Na sensitive site on the ␣ subunit gene has thus far been elusive (Herrera et al., 1998). Nevertheless, it can be speculated that cells contain specific mechanisms for replenishing Na pumps in the cell membrane. First would be the translocation of preformed pumps residing in the cytoplasm, followed by gene transcription/translation of the pump subunits. The specific control mechanisms of this continual process have not yet been clearly delineated for any one cell type, but while the control systems between cells may be similar, there most certainly will be cell-specific components. Thus there probably is feedback from the cell membrane to the pool (endoplasmic reticulum?), and in turn from the pool to the nucleus. The specifics of these interactions are unknown, especially regarding feedback from the cytoplasmic pool to the nucleus. As the pool becomes devoid of its pumps, as they move to the membrane, the nucleus must sense the need for specific gene activation, and more pumps then are synthesized, and moved first to the cytoplasmic pool to await the summons to be moved to the cell membrane. While the important area of basal Na pump recycling has not been extensively studied, Doné et al. (2002) have shown that the dopamine activated endocytosis of Na pump in rat kidney epithelial cells occurs through specific sites on the ␣1 subunit that interact with activation sites for clathrin coated vesicular transport. This is clearly of considerable importance if one is to suggest that EDHF-mediated responses involve stimulation of the Na pump. Indeed, variations may well limit the effect of K or any agent, depending on the numbers of functional pumps residing within the cell membrane. Thus a knowledge of drug pretreatment of the vessel in question would be important to understand any specific effect of K on pump function. Thus “dormant” smooth muscle cells may have fewer pumps in the membrane than cells that have been stretched, or treated with drugs for example. The same level of myoendothelial K obviously could have quite different effects, if there were varying numbers of functional pumps within the cell membrane upon which K could act.

3. THE Na PUMP AS A TRANSDUCING PROTEIN COMPLEX, INITIATING A SIGNALING CASCADE – (REGULATORY FUNCTION) The pump may have functions in addition to its well-known control of ionic gradients. Indeed in rat cardiomyocytes lower concentrations of ouabain, a well-known inhibitor of the pump and can activate a signaling pathway via Src→EGFR→ERK1/2 (Peng et al., 1996). While the pathway of activation was clearly identified in these cells, the functional consequences of such activation could only be presumed to be an hypertrophic response, since these cells cannot proliferate. Nevertheless, this cascade is entirely separate from any inhibitory effect that ouabain had on the pump (Haas et al., 2002). However, when this highly novel ouabain effect was studied in canine vsmc referred to above, the functional cellular response to the addition of very low concentrations of ouabain was a proliferation of vsmc, as measured by both BRDU uptake (a measure of DNA synthesis) and actual cell counting (Koksoy et al., 2001). Effects of low concentrations of ouabain have now been shown to occur in two additional cell types; prostate smooth muscle cells (proliferation) (Chueh et al., 2001) and rat kidney

Trafficking and transduction functions of the Na pump

23

epithelial cells (Ca oscillations) (Aizman et al., 2001). Thus this nonionic effect of ouabain interaction with the Na pump has been demonstrated in four different cell types, inducing three different responses. Two of these effects have been demonstrated in rat tissue (cardiomyocytes and kidney epithelial cells), known to be highly resistant to the effects of ouabain. Because of this resistance, the concentration of the drug necessary to cause these responses was between 10⫺5 M and 10⫺6 M. Nevertheless, the signaling functions without doubt occurred in the absence of any significant inhibition of the Na pump. Using canine vsmc, much lower concentrations of ouabain can be used, since cells from this species are known to be three orders of magnitude more sensitive to ouabain as well as other glycosides. A concentration of 10⫺10 M–10⫺9 M ouabain activated the same pathway in canine cells as did the higher concentrations in rat cardiomyocytes: Src→EGFR→ERK1/2. However, most importantly, the functional consequence of the activation of this pathway, namely cellular proliferation was shown. In addition, because of proliferation at such low levels of the drug, concentration response curves were possible. At higher concentrations ouabain inhibited the proliferation effect, that is, the effect was biphasic. This was similar to the data shown for the prostate smooth muscle cells, proliferation at lower concentrations, and inhibition of proliferation at the higher levels. One explanation for these effects is as follows. At these very low ouabain concentrations, ~1–2% of the pumps are actually inhibited. Thus, the very small increase in Nai would immediately be eliminated by the presence of the uninhibited pumps in the cell membrane. Hence not only is the effect on Nai extremely small, it is also very transitory, and hence very difficult to detect. This does not rule out a local increase in Na, but since we have shown that the Na /Ca exchanger in these cells is not a major regulator of Cai, a Ca amplifying effect is highly unlikely. Because of the very low concentrations of ouabain at which these effects occur, the usual ligand binding studies for determination of affinity are difficult to perform. As a result, the question could be raised as to whether ouabain is binding to the ␣1 subunit of the pump, or to another as yet undetermined protein. Because ouabain has never been shown to consistently bind to other proteins with the extremely high affinity necessary to initiate these proliferative effects, it was assumed that the ␣1 subunit is the receptor, and ouabain concentrations response curves were performed on rat vsmc (A7r5 cells) known to contain only the ␣1 subunit which is three orders of magnitude less sensitive than the subunit in “sensitive” species (Figure 3.1).

Figure 3.1 Ouabain concentration response effect on proliferation of dog and rat vsmc. Open bars are canine cells, hatched bars are rat cells.

24

Julius C. Allen et al.

Figure 3.2 A drug–receptor interaction model, explaining the biphasic ouabain effects on vsmc proliferation.

It can be seen in canine cells that the effect starts at 10⫺10 M, peaks at 10⫺9 M, and is then inhibited at 10⫺8 M. In the rat A7r5 cells, the proliferative effect begins at 10⫺7 M, peaks at 10⫺6 M, continuing with inhibition, but at three orders of magnitude higher than the effective concentrations with the canine cells. Thus despite the lack of direct binding data, these data are quite consistent with the hypothesis that ouabain is binding to the ␣1 subunit of the pump and activating the signaling pathway through specific interaction with this protein component of the Na pump. One proposal for the mechanism of these newly identified effects of ouabain is based on standard drug–receptor interactions (Figure 3.2). Thus the drug (ouabain) interacts with the receptor, ␣1 subunit (D ⫹ R → 1DR) activating the proliferative pathway. While the pumps to which these low concentrations of ouabain bind, are by definition inhibited, not enough of them are inhibited sufficiently to detect a change in Nai. In addition, any increase in Nai that does occur is immediately reduced by the many remaining uninhibited pumps. Thus at low concentrations of ouabain with minimal increases in Nai, the proliferative pathway is predominant, and occurs in the absence of detectable pump inhibition. However, as the ouabain concentration is increased, (⫹D →→ 2DR), and more pumps are inhibited, the standard increase in Nai occurs and is now detectable, and which inhibits the proliferative pathway. The biphasic ouabain concentration response curves that are observed with both canine cells and a rat vsmc line are entirely consistent with the model (Figure 3.2).

4. FUNCTIONAL INTEGRATION OF THE Na PUMP IN VSMC It might seem that the responsive and regulatory roles of the Na pump are unrelated. The first role is directly related to the physiological function of the pump that is part of the homeostatic mechanism regulating cellular ion transport. This role of the pump is “responsive” in that it responds to a variety of perturbations, as it seeks to maintain the ionic equilibrium of the cell. First its ionic turnover can be activated by PKA (or PKC) phosphorylation of the ␣1 subunit. If this is not sufficient, then cytoplasmic translocation occurs to increase the numbers of pumps residing in the membrane. Lastly specific gene transcription occurs and the resulting synthesis of new pump sites refills the cytoplasmic pool for relocation to the cell membrane as needed.

Trafficking and transduction functions of the Na pump

25

How can the transduction cascade component of the pump be placed within the context of the physiological paradigms that have been discussed? In order to relate these two important functions, it is necessary to define at least two locations of Na pumps within the membrane, one defined by localization within the caveolae, and the other within the bulk membrane. Such specific localization of a variety of proteins including channel proteins, and the sarcoplasmic reticulum Ca pump, occur within these specialized lipid rafts (Thyberg, 2000; Carpenter, 2000; Ushio-Fukai et al., 2001; Daniel, 2002). A major function of these caveolae can be defined by their content of caveolin, of which there are at least three tissue specific isoforms. The role of caveolin is to act as a scaffolding protein, forming clusters of proteins which are not usually in close proximity when limited to the bulk cell membrane. Thus, for example, in rat cardiomyocytes the Na pump complex within the caveolae may interact with caveolin forming protein clusters within the caveolar structure, bringing two or more proteins within close proximity, and allowing possible interaction (Mohammed et al., 2002). The responsive function of the pump occurs as it responds to physiologic or pharmacologic incidents, and is represented by the right side of Figure 3.3, and the pumps are designated 1, 2, and 3. These three pumps may not be identical in pumping function or capacity even though they reside in the bulk membrane component. They clearly could have different ionic capacities, as well as turnover functions, since pump 1 has just been moved to the membrane, pump 2 is “fully” operational, and pump 3 is in the process of endocytosis. As pump 1 is moved into the membrane, (top of Figure 3.3) it may proceed to become a functional pump 2, or it may move toward the caveolae. While the mechanism of activation of protein movement into caveolae is largely unknown, it has been suggested that the regulatory protein may be epidermal growth factor, as it interacts with its receptor (EGFR). Carpenter (2000) has suggested that the EGFR is a “nexus” for both trafficking and signaling. Such a concept has not yet been applied to the Na pump, but it is important to note that the transactivation of the EGFR is essential for the activation of the cascade in both cardiac cells (6, 12) and vsmc (10). Thus this receptor might well mediate the transduction pathways activated by the pump complex, as well as the trafficking as the pump moves through the

Figure 3.3 A model of Na pump compartmentation within the membrane, and the interaction between regulatory and responsive functions. See text for explanation.

26

Julius C. Allen et al.

cytoplasm to the membrane, or indeed as the pump moves between different membrane localizations. Thus pump 1, on the left may become attached to caveolin “C” and continue to be further localized to the caveolae, where it then can be brought into close proximity with the signaling proteins, Src, and EGFR and would then be able to respond to low concentrations of ouabain activating a signaling cascade. The caveolar pumps must have normal pumping potential, since they are activated by low concentrations of ouabain, but because of their sequestration or interaction with other proteins, are prevented from maximal function, but can become “activated” when moved out of the caveolae. Such a model suggests that specific localization of pump proteins can significantly modulate cellular function. Indeed, the concept of responsive and regulatory functions almost necessitates the selective localization of the pump proteins. Most of the regulatory components of these two functions are unknown. One of the possible EDHF “compounds” is the low level of myoendothelial K, generated by endothelial cells. Obviously, in order for K to effectively stimulate the pump it must be in an operating and accessible modality, and by this hypothesis be localized to the bulk membrane compartment. The functional pump content localized within the membrane will determine its response to K. Thus, depending on the status of the cell/ blood vessel in question, K may or may not have a stimulatory effect. For example, what would the effect of K be on a pump that was already maximally stimulated by phosphorylation of the ␣ subunit? Would such a situation require the movement of preformed pumps to the membrane? A recent paper (Schubert et al., 2002) suggested that there also may be a complex relationship between both K and gap junctions as EDHFs. It was shown that there was a direct interaction between connexin 43 the major gap junction protein and caveolin-1, and that the two proteins co-localized to specific caveolae. The significance of this observation is currently unknown.

5. SUMMARY AND CONCLUSIONS The purpose of this short review on the cell biology of the Na pump was specifically not to determine how or whether K ( or any other presumed EDHF) might activate the pump to induce hyperpolarization of vsmc. That these events occur in selected vascular beds is understood. The identities of these compounds is not addressed here. However, the concept was introduced that the pumps as they exist in the sarcolemma cannot be considered to be stable, consistent monotonic regulators of ion transport. While indeed, their major function may well be to maintain proper transmembrane ionic gradients, and as such be activated by modulating levels of myoendothelial K (EDHF), their response to any such ion or ligand may well depend on the physiological and pharmacologic state of the cell. Thus the numbers of functional pump sites may vary considerably in the same vascular bed, such that at one time increases in myoendothelial K may have little or no effect, while at another time, under different circumstances even small increases in K may have a profound effect on cell membrane potential. Thus, not all pumps located within the membrane may be equally accessible for transport, or to modifying agents. This answers no questions about EDHF, but rather presents new factors about one possible receptor for the EDHFs, which may complicate data interpretation involving stimulation of the Na pump, resulting in the hyperpolarization of vsmc. Whether any of the EDHFs have any interaction with caveolae or their contents is an area that has not yet been studied.

4

Isoforms of the Na,K-ATPase Kristan Lansbery, Margaretta L. Mendenhall, Lauren C. Vehige, James A. Taylor, Gadis Sanchez, Gustavo Blanco and Robert W. Mercer

The Na,K-ATPase, or Na pump, is a ubiquitous, multi-spanning membrane protein that actively maintains the high internal K⫹ and low internal Na⫹ concentrations characteristic of animal cells. The Na,K-ATPase consists of at least two noncovalently linked subunits: a multispanning membrane protein termed the ␣ subunit, and the ␤ subunit, a smaller glycosylated membrane protein. At present, four different ␣ polypeptides (␣1–␣4) and three ␤ isoforms (␤1–␤3) have been identified in mammals. All ␣ isoforms can stably assemble with all of the ␤ isoforms to form catalytically competent Na,K-ATPase molecules with unique kinetic properties. In addition to the ␣ and ␤ subunits, other proteins interact with the Na pump to modify its activity. Members of a gene family of small membrane proteins that contain an invariant FXYD motif also influence the activity of the enzyme. Thus, the expression and regulation of specific Na pump polypeptides and accessory proteins allow cells to precisely coordinate Na,K-ATPase activity to their physiological requirements. To better understand the electrophysiological properties of endothelial and smooth muscle cells the expression of the Na,K-ATPase subunits and FXYD family members were cataloged. In addition, the ouabainsensitivity of the Na,K-ATPase isozymes when associated with some members of the FXYD gene family was determined. These results may provide important insights into the role of the enzyme in the establishment, maintenance and regulation of cellular transmembrane ion gradients.

1. INTRODUCTION The Na,K-ATPase establishes and maintains the high internal potassium and low internal sodium concentrations characteristic of most animal cells. These ion gradients are fundamental to such diverse cellular functions as the regulation of cell volume and pH, nutrient uptake and membrane excitability (Blanco and Mercer, 1998). The Na,K-ATPase consists of two noncovalently linked subunits: a 100 kDa multi-spanning membrane protein termed the ␣ subunit, and the ␤ subunit, a smaller glycosylated membrane protein. A small peptide, termed the ␥ subunit, has also been identified in purified preparations of the enzyme. The ␥ subunit was shown to be part of the enzyme complex when it was demonstrated that the protein, along with the ␣ and ␤ subunits, could be covalently labeled by photoaffinitylabeled derivatives of the specific inhibitor, ouabain (Forbush et al., 1978). All enzymatic functions of the enzyme have been assigned to the ␣ subunit. The ␣ subunit contains the binding sites for ATP and ouabain; it is phosphorylated by ATP and undergoes liganddependent conformational changes accompanying the binding, occlusion and translocation of ions. The ␤ subunit appears to serve a chaperone function facilitating the correct membrane insertion of the newly synthesized ␣ subunit (Hasler et al., 1998). In addition, the ␤ subunit influences the cation affinities of the Na,K-ATPase and is essential in stabilizing, or may

28

Kristan Lansbery et al.

take part in forming, the cation occluding complex of the enzyme (Blanco and Mercer, 1998). While the ␥ subunit is not required for Na,K-ATPase function, the subunit does influence the cation affinity of the enzyme (Arystarkhova et al., 1999; Therien et al., 1999). In mammalian cells, multiple isoforms for both the ␣ (␣1–␣4) and ␤ (␤1–␤3) subunits have been identified. The ␣ and ␤ isoforms exhibit a tissue-specific and developmental pattern of expression that may be important in the maintenance and regulation of Na,K-ATPase activity. The ␣ subunit isoforms, which are products of different genes, are over 85% homologous and have identical predicted secondary structures. The ␤ subunit isoforms display more divergence than the ␣ isoforms. In rat, the ␤2 subunit exhibits a 42% and 49% amino acid sequence similarity with the ␤1 and ␤3 subunits, respectively. The ␤3 and ␤1 isoforms share 37% amino acid similarity. It appears that all the ␣ isoforms can stably assemble with all of the ␤ isoforms to form catalytically competent Na,K-ATPase molecules with unique kinetic properties. The major kinetic differences occur among Na,K-ATPases isozymes differing in ␣ subunit composition. However, the variation in the ␤ subunit composition can significantly affect the properties of a particular ␣ polypeptide. The most conspicuous kinetic difference among the Na,K-ATPase isozymes corresponds to their reactivity towards ouabain, with ␣3 and ␣4 displaying a high, ␣2 an intermediate, and ␣1 a low sensitivity to the cardiotonic steroid. The physiological significance of this difference is not known. However, clinically, this difference has been exploited in the treatment of congestive heart failure. For over 200 years cardiac glycosides or steroids, such as ouabain, digoxin or digitalis, have been administered for congestive heart failure. In fact, it has been suggested that ouabain is actually an adrenal cortical hormone that plays a role in cellular Na⫹ regulation and therefore whole body salt and water balance (reviewed in Blaustein, 1993). Although the exact mechanism of action of the cardiac glycosides on the heart is controversial, the conventional view of the positive inotropic mechanism proposes that the cardiac glycosides inhibit myocardial ␣2 or ␣3 Na,K-ATPases, leading to an increase in intracellular Na⫹. This increase in Na⫹ stimulates the uptake of Ca2⫹ through the plasma membrane Na⫹/Ca2⫹ exchanger. The small change in intracellular Ca2⫹ is amplified several thousandfold in the sarcoplasmic reticulum by the Ca-ATPase. The increased Ca2⫹ augments the contractility of the myocardial cell. Thus, inhibition of a few pump sites can result in relatively large changes in the contractility of the heart (Blaustein, 1993). As mentioned, a third protein, termed the ␥ subunit, also has been identified in purified preparations of the enzyme. The ␥ subunit is a small, single-membrane spanning protein that is stoichiometrically associated with the ␣ and ␤ subunits. It is clear from expression studies that the ␣ and ␤ subunits are sufficient for the functional expression of Na,K-ATPase activity. However, it appears that the ␥ subunit can influence the characteristics of the ␣␤ heteromer. For example, the subunit can modify the voltage dependence of K⫹ activation (Beguin et al., 1997) and influence the apparent affinity of the enzyme for Na⫹, K⫹ and ATP (Arystarkhova et al., 1999; Therien et al., 1999). Two isoforms of the subunit (termed ␥a and ␥b) are present in the kidney (Figure 4.1). These two splice variants differ only in their extracellular N-termini. Both variants affect the enzymatic properties of the Na,K-ATPase to a similar extent (Pu et al., 2001). Although an exact functional role for the ␥ subunit is not known, it may be important in renal function. A mutation in the gene coding the subunit results in renal hypomagnesemia associated with hypocalciuria (Meij et al., 2000). The conversion of a conserved glycine within the transmembrane domain to arginine (G41R) leads to misrouting of the ␥ subunit from the plasma membrane to intracellular compartments. In the kidney, the misrouting of ␥ may result in the diminution of Na,K-ATPase activity at the plasma membrane, causing hypomagnesemia (Meij et al., 2000).

Isoforms of the Na,K-ATPase

29

The ␥ subunit is a member of a gene family of structurally related, small membrane proteins containing a single transmembrane domain and an invariant motif of FXYD (phenylalanine-X-tyrosine-aspartate). The FXYD family has a characteristic 35-amino-acid stretch that includes seven invariant amino acids before, in, and just after the membrane span. These proteins are present mainly in tissues that perform transepithelial fluid and solute transport or that are electrically excitable (Sweadner and Rael, 2000). Currently, the family consists of seven members numbered according to the year of their initial sequencing (Figure 4.1): phospholemman (PLM; FXYD1), Na,K-ATPase ␥ subunit (FXYD2), mammary tumor protein of 8-kDa (Mat-8; FXYD3), corticosteroid hormone-induced factor (CHIF; FXYD4), related to ion channel (RIC; FXYD5), phosphohippolin (PHP; FXYD6) and one new family member (FXYD7) identified only in the databases. Phospholemman is a membrane protein that in myocardium is the major plasma membrane substrate for protein kinase A (PKA) and protein kinase C (PKC). CHIF is an epithelial cell-enriched protein that is aldosterone-induced in a tissue specific manner. CHIF mRNA is abundantly expressed in the distal colon and the medullary and papillary kidney. Similarities in the expression patterns of CHIF and the ␣- subunit of the colonic H,K-ATPase in response to aldosterone, have suggested that CHIF may act as a modifier enzyme (Grishin et al., 1999). However, CHIF can modify the activity of the Na,K-ATPase (Beguin et al., 2001). When expressed in Xenopus oocytes with the Na,K-ATPase ␣ and ␤ subunits, CHIF increases the Na⫹ affinity and decreases the apparent K⫹ affinity of the enzyme. It is still not clear if CHIF interacts with the Na,K-ATPase in native tissue. Mat-8 and RIC were originally identified as factors up-regulated by oncogenes. Mat-8 RNA is expressed in primary human breast tumors and in breast tumor cell lines (Morrison et al., 1995), and murine Mat-8 is expressed at high levels in the uterus, stomach and colon. The protein appears to be a marker of cells transformed by Neu or Ras oncoproteins. RIC gene expression has also been detected in the heart, spleen, lung, testis and skeletal muscle of normal adult rats (Fu and Kamps, 1997). It is not clear if

Figure 4.1 Alignment of FXYD family members. Invariant residues are shaded.

30

Kristan Lansbery et al.

these up-regulated genes are required for the transformed state or if these proteins are characteristic features of secretory epithelial cells and are simply maintained in the transformed state. PHP was originally isolated from a rat hippocampus library and found to be abundantly expressed in rat brain (mainly the hippocampus and cerebellum) and kidney. The function of PHP is unknown. FXYD7 has been identified only in the databases. To better understand the role of the Na,K-ATPase in maintaining cellular ion gradients we have characterized the expression of the Na,K-ATPase subunits and FXYD family members. In addition, the ouabain-sensitivity of the Na,K-ATPase isozymes when associated with some members of the FXYD gene family was determined. These results may provide important insights into the role of the enzyme in the establishment, maintenance and regulation of cellular transmembrane ion gradients. 2. METHODS

2.1. Reverse transcription-polymerase chain reaction (PCR) Total RNA was isolated from cultured human umbilical vein endothelial cells using guanidinium isothiocyanate and ␤-mercaptoethanol (Chirgwin et al., 1979). Reverse transcription-PCR was performed using the AccessQuick RT-PCR system (Promega Corp., Madison, WI). PCR reactions were carried out using gene-specific primers to the human FXYD family members. Amplification consisted of 40 cycles at 94 ⬚C for 30 s, 65 ⬚C for 1 min and 68 ⬚C at 2 min.

2.2. Viral constructions and infections The rat ␣1, ␣2, ␣3, ␤1, CHIF, mouse phospholemman and human and rat ␥ Na,K-ATPase cDNAs were subcloned into the baculovirus expression vector pBlueBac 4.5. Recombinant baculovirus preparation, selection and amplification were performed (O’Reilly et al., 1992). Uninfected and infected Sf-9 cells were grown in 150-mm petri dishes in TNM: FH medium (JRH Biosciences, Lenexa, KS), supplemented with 10% vol/vol fetal bovine serum, 100 units/ml of penicillin, 1.72 ⫻ 10⫺4 M of streptomycin and 2.71 ⫻ 10⫺7 M of Fungizone. Viral infections were done at a viral multiplicity of infection ranging from 5 to 10. At 72 h after infection, cells were scraped from the plates in the incubating media, centrifuged at 1500 ⫻ g for 10 min and then washed three times in 10 mM imidazole hydrochloride (pH 7.5), 1 mM EGTA. The final pellet was resuspended in the same solution. For the determination of enzymatic activity, the intact cells were used after permeabilization with alamethicin (10 ␮g/mg protein) for 10 min at 25 ⬚C as described previously (Blanco et al., 1995).

2.3. Biochemical assays Protein assays were performed using the bicinchninic acid/copper sulfate solution as described by the supplier (Pierce Chemical Co, Rockford, IL). Na,K-ATPase activity was assayed at 37 ⬚C through determination of the initial rate of release of 32Pi from ␥[32P]-ATP. The maximal Na,K-ATPase activity of samples (50 ␮g total protein) was measured in a final volume of 0.25 ml of medium containing 120 mM NaCl, 30 mM KCl, 3 mM MgCl2, 0.2 mM EGTA, 30 mM Tris–HCl (pH 7.4) ⫾ 1 mM ouabain. Sodium azide (2 mM final concentration) was included in the mixture to inhibit mitochondrial ATPase. The assay was started by the addition of ATP with 0.2 ␮C [␥-32P]ATP (3 mM final concentration). Following a 30-min incubation at 37 ⬚C, the tubes were placed on ice and the reaction terminated by the addition

Isoforms of the Na,K-ATPase

31

of 25 ␮l of 55% trichloroacetic acid. Released 32Pi–Pi was converted to phosphomolybdate and extracted with isobutanol. The organic phase of 0.15 ml containing the phosphomolybdate complex was removed and radioactivity measured by liquid scintillation counting. The Na,K-ATPase activity was determined as the difference in ATP hydrolysis in the absence and presence of 1 mM ouabain. The ATP hydrolyzed never exceeded 15% of the total ATP present in the sample and hydrolysis was linear over the incubation time. To determine the effect of different ouabain concentrations on Na,K-ATPase activity, samples were incubated in the reaction medium with the indicated concentrations of ouabain for 30 min at 37 ⬚C prior to the addition of ATP.

2.4. Data analysis Curve fitting of the experimental data was carried out using a Marquardt least-squares nonlinear regression computing program (Sigma Plot, Jandel Scientific, San Rafael, CA). Doseresponse relations for the ouabain inhibition of Na,K-ATPase activity showed a single homogeneous population and were fitted by the equation: v ⫽ 100 (1/(1 ⫹ [I]/Ki)) where v is the Na,K-ATPase activity corresponding to a certain concentration of the inhibitor ouabain [I], expressed as a fraction of activity in the absence of ouabain, and Ki is the concentration of ouabain that gives the half-maximal inhibition. Statistical analysis of the ouabain dose-response curves was done applying a F test to compare the relative goodness of fit between the curves describing the individual kinetics of the Na,K-ATPase isozymes in the absence or presence of the ␥ subunit with the one obtained combining the data of the ␣␤ and ␣␤␥ enzymes. The F value was calculated as reported previously (Blanco et al., 1995) using the following equation: F ⫽ (SSc ⫺ SSi/dfc ⫺ dfi)/(SSi/dfi) where SSc is the total sum of squared deviations for the simultaneous fitting of the combined data for the ␣␤ and ␣␤␥ enzymes, SSi is the sum of squared deviations for the fits of the individual curves, and dfc and dfi, are the degrees of freedom (number of data points – number of parameters) for the simultaneous and both individual fits respectively. Differences were considered significant when the F value exceeded the theoretical F value for the corresponding degrees of freedom with p ⫽ 0.01.

3. RESULTS 3.1. RT-PCR/ Tissue distribution of ESTs RT-PCR was used to detect the expression of mRNAs coding for the Na,K-ATPase ␣ isoforms, PLM, Na,K-ATPase ␥ subunit, Mat-8, CHIF, RIC, PHP, and FXYD7 in total RNA obtained from cultured human umbilical vein endothelial cells. In addition, the tissues of origin of the expressed sequence tags (ESTs) representing the above cDNAs were identified from the individual GenBank report annotations and by using the UniGene resources offered by the National Center for Biotechnology Information at the NIH (Table 4.1). In human umbilical vein endothelial cells, RT-PCR detected mRNA coding for RIC and the Na,K-ATPase

32

Kristan Lansbery et al. Table 4.1 Relative tissue distribution of Na,K-ATPase ␣ subunits and FXYD family members

␣1 ␣2 ␣3 ␣4 PLM (FXYD1) Gamma (FXYD2) Mat-8 (FXYD3) CHIF (FXYD4) RIC (FXYD5) PHP (FXYD6) FXYD7

Muscle

Nervous tissue

Transport epithelia

Endothelial cells*

⫹ ⫹⫹ ⫹ ⫺ ⫹⫹ ⫹ ⫹ ⫺ ⫹ ⫹⫹ ⫺

⫹ ⫹⫹ ⫹⫹ ⫺ ⫹ ⫺ ⫹⫹ ⫺ ⫹ ⫹⫹ ⫹⫹⫹

⫹⫹⫹ ⫺ ⫺ ⫺ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹ ⫺

⫹ ⫺ ⫹ ⫺ ⫺ ⫹/⫺ ⫹/⫺ ⫹/⫺ ⫹ ⫹/⫺ ⫺

Notes Distribution of Na,K-ATPase ␣ subunits and FXYD family members in cDNA libraries from mouse and human expressed sequence tags database. ⫺ not found; ⫹⫹⫹ over 60% of identified cDNAs; ⫹⫹ 25–60%; ⫹ ⬍ 25%. * Expression of cDNAs in cultured human endothelial cells determined using RTPCR. ⫺ not detected; ⫹/⫺ trace; ⫹ detected. PLM, phospholemman; Mat-8, mammary tumor protein of 8-kDa; CHIF, corticosteroid hormone-induced factor; RIC, related to ion channel; PHP, phosphohippolin.

␣1 and ␣3 subunits. There was also weak expression of ␥, CHIF, Mat-8 and PHP mRNA. The Na,K-ATPase ␣ isoform ESTs are distributed in several tissues with the exception of the ␣4 isoform which is found exclusively in the testes (Table 4.1). With the exception of PHP and FXYD7, the FXYD family members are found predominately in libraries derived from transport epithelia. PHP and FXYD7 were mostly found in libraries derived from the nervous system. ESTs coding for the Na,K-ATPase ␣1, ␣2, and ␣3 isoforms, PLM, ␥, Mat-8, RIC and PHP were also found in libraries derived from smooth muscle.

3.2. Ouabain-sensitivity of the Na,K-ATPase when associated with the ␥ subunit or CHIF To determine if the ␥ subunit influences the ouabain-sensitivity of the Na pump, the kinetics of inhibition by the cardiotonic steroid was studied in Na,K-ATPases containing or lacking the ␥ subunit. Coexpression of ␣ and ␤ isoforms in Sf-9 cells results in catalytically competent Na,K-ATPase molecules (Blanco et al., 1995a,b, 1999). Dose-response curves for the ouabain inhibition of Na,K-ATPase activity for the ␣1␤1 and ␣1␤1␥ complexes were determined under non-limiting ligand concentrations (120 mM NaCl, 30 mM KCl, 3 mM MgCl2 and 3 mM ATP). When the rat and human ␥ subunits are expressed with the ␣1 and ␤1 subunits, the resulting Na,K-ATPase isozyme is more sensitive to the cardiotonic steroid (Table 4.2). Thus, ␣1␤1␥ consisting of the rat and human ␥ subunits are approximately 10-fold more sensitive to the cardiotonic steroid than the corresponding ␥ free isozyme. To determine if the human ␥ subunits influence the ouabain-sensitivity of other Na,K-ATPase isozymes, the ouabain-sensitivity of the ␣2␤1␥ and ␣3␤1␥ isozymes were characterized. The expression of the human ␥ subunit with the ␣2␤1 and ␣3␤1 isozymes did not significantly change the ouabain-sensitivity of the enzymes (Table 4.2).

Isoforms of the Na,K-ATPase

33

Table 4.2 Apparent inhibition constants (Ki) of the Na,KATPase ␣␤ isoforms in the absence or presence of different ␥ subunits and the CHIF polypeptide. The enzymes expressed in insect cells as well as the native ␣1␤1␥ from rat kidney are shown Isozyme

Ki (M)

⌬

␣1␤1 ␣1␤1␥ (native) ␣1␤1␥ human ␣1␤1␥ rat A ␣1␤1␥ rat B ␣1␤1 CHIF ␣1␤1 PLM ␣2␤1 ␣2␤1␥ human ␣3␤1 ␣3␤1␥ human

1.5 ⫾ 0.3 ⫻ 10⫺5 9.8 ⫾ 0.9 ⫻ 10⫺5 1.3 ⫾ 0.1 ⫻ 10⫺6 1.6 ⫾ 0.2 ⫻ 10⫺6 1.1 ⫾ 0.2 ⫻ 10⫺6 7.5 ⫾ 1.3 ⫻ 10⫺7 1.0 ⫾ 0.4 ⫻ 10⫺5 2.2 ⫾ 0.1 ⫻ 10⫺7 2.3 ⫾ 0.3 ⫻ 10⫺7 3.9 ⫾ 0.4 ⫻ 10⫺8 2.2 ⫾ 1.0 ⫻ 10⫺8

— 6.5 ⫻ res 11.5 ⫻ sen 9.4 ⫻ sen 13.6 ⫻ sen 20.0 ⫻ sen * — * — *

Note * Not significantly different from the respective isozyme in the absence of the ␥ subunit.

To determine if other members of the FXYD gene family can influence the ouabain sensitivity of the Na,K-ATPase, CHIF and PLM were expressed with the ␣1␤1 isozyme (Table 4.2). As shown, CHIF also associates with and influences the ouabain affinity of the enzyme. The ␣1␤1CHIF enzyme is approximately 20-fold more sensitive to inhibition by ouabain than ␣1␤1. PLM did not influence the ouabain-sensitivity of the ␣1␤1 isozyme.

4. DISCUSSION The possibility that twelve functionally unique Na,K-ATPase isozymes exist suggests that Na⫹ and K⫹ transport can be modified through the expression of specific Na,K-ATPase isozymes. Moreover, the ability of accessory proteins to modify the functional properties of the Na,K-ATPase adds to this complexity. To better understand the cellular mechanisms of cation transport, the presence of the individual Na,K-ATPase isoforms and FXYD proteins were cataloged. Although highly expressed in epithelia, the majority of Na,K-ATPase isoforms and FXYD proteins exhibit a broad pattern of tissue distribution. Thus, multiple Na,K-ATPase isozyme complexes are probably formed in most tissues. Because the Na,K-ATPase isozymes have kinetic properties that are unique, isozyme-specific expression may be important in adapting Na,K-ATPase function to the requirements of each cell. In addition, the present results demonstrate that some members of the FXYD gene family can interact with the Na,K-ATPase to influence its functional properties. While both the ␥ subunit and CHIF can influence the ouabain-sensitivity of the ␣1␤1 isozyme, PLM does not. The ability of ␥ to influence the ouabain-sensitivity of the Na,K-ATPase is limited to the ␣1␤1 isozyme as ␥ does not significantly influence the ouabain-sensitivity of the ␣2 or ␣3 isozymes. It is possible that the human ␥ subunit does not associate with the ␣2␤1 or ␣3␤1 isozymes, however, this possibility seems unlikely since a chimeric human ␥ subunit with an extra 31 residues at the N-terminus, reduces the ouabain sensitivity of these isozymes

34

Kristan Lansbery et al.

(data not shown). Nevertheless, the possibility that the human ␥ subunit may not be assembling with the ␣2 or ␣3 isozymes cannot be excluded. CHIF and the Na,K-ATPase are highly expressed in both the kidney and colon, where they may have the opportunity to interact. The exact physiological significance of the effect of CHIF and ␥ on the ouabain-sensitivity of the Na,K-ATPase is unknown. However, if endogenous ouabain-like molecules regulate the Na,K-ATPase, then the ␥ subunit and CHIF may play a role in establishing the ouabain sensitivity of the enzyme.

5

Calcium sparks and membrane potential Delrae M. Eckman, Luis Fernando Santana, Thomas J. Heppner, Adrian D. Bonev and Mark T. Nelson

Calcium sparks are local transient elevations of intracellular calcium caused by the opening of ryanodine-sensitive, calcium-release channels (“ryanodine receptors”) in the sarcoplasmic reticulum membrane. In pressurized arteries, calcium sparks in smooth muscle cells occur at a frequency of about 1 per cell per second. This frequency depends on calcium entry through voltage-dependent calcium channels, and increases with pressure-induced depolarization and decreases with voltage-dependent calcium channel inhibition. Calcium sparks in pressurized arteries, even at the highest observed frequency, contribute little to overall average (“global”) intracellular calcium. Calcium sparks activate large-conductance, calcium-sensitive potassium (BKCa) channels in the surface membrane to cause a transient membrane potential hyperpolarization. Simultaneous measurements of calcium sparks and BKCa currents indicate that virtually every calcium spark increases the activity of about 20 nearby BKCa channels 106-fold. Disabling the communication of calcium sparks to BKCa channels by deleting the ␤1-subunit of the BKCa channel leads to depolarization and constriction of pressurized arteries, as well as to an elevation in blood pressure. These results support the concept that an elementary calcium signal (calcium spark) can regulate a global smooth muscle cell parameter (membrane potential) by providing high local calcium to BKCa channels, and that detuning this local communication, for example, ␤1 gene deletion, can elevate vascular tone and blood pressure.

1. INTRODUCTION Intracellular calcium ([Ca2⫹]i) ions play a central role in the regulation of arterial function through effects on membrane potential, contraction and gene transcription. Three distinct Ca2⫹ signaling modalities have been identified in arterial smooth muscle. Ca2⫹ sparks are stationary, local Ca2⫹ transients caused by the opening of ryanodine-sensitive Ca2⫹ release channels (“ryanodine receptors”) in the sarcoplasmic reticulum membrane (Figure 5.1). Ca2⫹ waves are local elevations of Ca2⫹ which propagate across a cell and are likely caused by opening of ryanodine receptors and/or IP3 receptors (IP3Rs). Global Ca2⫹ changes are graded alterations in average [Ca2⫹]i concentration throughout the entire cytoplasm, that are regulated by Ca2⫹ influx through dihydropyridine-sensitive, voltage-dependent Ca2⫹ channels. The spatial, frequency and amplitude components of these Ca2⫹ signals encode different cellular outcomes (e.g. membrane hyperpolarization, contraction, gene transcription) that are interlinked at multiple levels. Ca2⫹ sparks were first identified in cardiac, and subsequently in skeletal muscle and smooth muscle (Cheng et al., 1993; Nelson et al., 1995; Tsugorka et al., 1995; Klein et al., 1996). Ca2⫹ sparks arise from the opening of a group of ryanodine receptors in the sarcoplasmic reticulum membrane. In cardiac muscle, the action potential activates voltagedependent calcium channels that stimulate nearby ryanodine receptors to release calcium

36

Delrae M. Eckman et al.

Diameter (µm)

250

[Ca2+]cyt (nM)

Ca2+ channel inhibitor: Nisoldipine, 10 nM

300

200 150

200 100 1 min

Figure 5.1 Voltage-dependent Ca2⫹ channels regulate intracellular Ca2⫹ and diameter of intact pressurized cerebral arteries (see Color Plate 2).

from the sarcoplasmic reticulum, referred to as calcium-induced calcium release. The elementary calcium release event in cardiac muscle appears to be a Ca2⫹ spark. Thus, the calcium transient caused by calcium-induced calcium release reflects the summation of many calcium sparks. In arterial smooth muscle, Ca2⫹ sparks activate nearby BKCa channels in the surface membrane to cause a hyperpolarizing K⫹ current, which leads to closure of voltage-dependent calcium channels, a decrease in Ca2⫹ entry and a corresponding decrease in global [Ca2⫹]i. Thus, a local Ca2⫹ signal can oppose changes in global Ca2⫹.

2. RESULTS AND DISCUSSION Global elevations in [Ca2⫹]i are critical for the contractile response of smooth muscle cells (Figure 5.1), but the once monolithic view of uniformly distributed Ca2⫹ regulating smooth muscle physiology has given way to a new paradigm (Figure 5.2). In this new view, subtle differences in Ca2⫹ signals are important and there is an increasing appreciation of the exquisitely tuned molecular mechanisms that exist to discriminate among them. For example, as will be addressed in this chapter, the transient release of [Ca2⫹]i through ryanodine receptors in the form of Ca2⫹ sparks regulates smooth muscle membrane potential and contractility through activation of physically juxtaposed BKCa channels (Figure 5.3), an action that highlights the importance of Ca2⫹ signal localization and amplitude in determining physiological responses. This chapter focuses on some key aspects of calcium sparks in arterial smooth muscle.

2.1. Calcium sparks: basic properties in smooth muscle Ca2⫹ sparks were first identified in smooth muscle cells from rat cerebral arteries (Nelson et al., 1995) (Figure 5.2), using a laser scanning confocal microscope and the calcium-sensitive

5 µm F/F0

2.0 1.0 400 ms 50 pA

400 ms

Figure 5.2 Spatial–temporal characteristics of a Ca2⫹ spark in an isolated posterior cerebral artery cell. Confocal line-scan of a fluo-3 loaded cerebral artery smooth muscle cell (top). The time course of the Ca2⫹ spark was averaged over the region indicated by the bar (middle). The bottom trace demonstrates the time course of a typical spontaneous transient outward current recorded from a different cell (⫺40 mV). The calcium spark image is a series of 6 ms line scans of a single region of the cell. The cell was scanned for a total of 1536 ms [From Nelson et al., Science, 1995, 270: 633–637.] (see Color Plate 3).

SPARK Global Ca2+

SR Ca2+ load Ca2+ Contraction

Figure 5.3 Proposed functional roles of Ca2⫹ sparks in smooth muscle cells. Ca2⫹ sparks activate BKCa channels to complete a negative-feedback loop (green arrow) through membrane potential hyperpolarization to decrease Ca2⫹ entry. Also illustrated is a positive-feedback contribution (red arrow) of Ca2⫹ release from Ryanodine receptor channels to contraction. In most smooth muscle types, the negative-feedback pathway appears to dominate [Modified from Jaggar et al., Am. J. Physiol., 2000, 278: C235–C256.] (see Color Plate 4).

38

Delrae M. Eckman et al.

fluorescent dye, fluo-3. Ca2⫹ sparks occur at a rate of ~1 spark/cell/s. Ca2⫹ sparks are rapid events, having a rise time of approximately about 20 ms with a decay rate (␶1/2) of about 60 ms. Ca2⫹ sparks occur close to the cell membrane (Nelson et al., 1995; Gollasch et al., 1998), and cover a surface area of about 13 ␮m2. A single smooth muscle appears to have a small number (⬍ 5) of spark sites per cell. In cardiac muscle, Ca2⫹ sparks occur due to the opening of ryanodine receptors located in the sarcoplasmic reticulum membrane (Cheng et al., 1993; Cannell et al., 1995; LopezLopez et al., 1994). This seems to be the case in smooth muscle as well. Ryanodine, an inhibitor of ryanodine receptors, inhibited calcium sparks in smooth muscle (Nelson et al., 1995; Jaggar et al., 1998b). Depletion of sarcoplasmic reticulum calcium by inhibition of the sarcoplasmic reticulum Ca2⫹ ATPase abolished calcium sparks in smooth muscle (Nelson et al., 1995). Increasing sarcoplasmic reticulum calcium leads to an elevation in calcium spark frequency and amplitude (ZhuGe et al., 1999; Wellman et al., 2001). For example, sarcoplasmic reticulum calcium load and calcium spark frequency were elevated in myocytes from cerebral arteries of phospholamban (PLB) knock-out mice (Wellman et al., 2001). PLB inhibits the SR Ca2⫹ ATPase, and its phosphorylation or removal increases sarcoplasmic reticulum Ca2⫹ ATPase activity, and consequently sarcoplasmic reticulum calcium load. These data indicate that Ca2⫹ sparks reflect local calcium release through ryanodine receptors in the sarcoplasmic reticulum membrane.

2.2. Communication between voltage-dependent calcium channels and Ca2⫹ sparks Resistance arteries exist in a partially constricted state (Figure 5.1). Elevation of intravascular pressure causes graded cell membrane depolarization and increases in global [Ca2⫹]i. The elevation of [Ca2⫹]i is due to activation of voltage-dependent calcium channels, since voltagedependent calcium channel inhibitors or membrane hyperpolarization decreases [Ca2⫹]i (Knot and Nelson, 1998). An increase in [Ca2⫹]i would be predicted to elevate Ca2⫹ spark frequency, since ryanodine receptors are activated by both cytoplasmic and sarcoplasmic reticulum Ca2⫹. Indeed, membrane depolarization by either elevating external K⫹ or intravascular pressure increased global Ca2⫹ and Ca2⫹ frequency in intact arteries (Jaggar et al., 1998b). For example, depolarizing smooth muscle cells from ⫺60mV to ⫺40mV increased global Ca2⫹ from 100 to 245nM (Knot and Nelson, 1998) and Ca2⫹ spark frequency from 0.2 to 1.0Hz (Jaggar et al., 1998b) (Figure 5.4). Inhibiting voltage-dependent calcium channels decreased [Ca2⫹]i and Ca2⫹ spark frequency and prevented membrane depolarization (Jaggar et al., 1998b; Knot and Nelson, 1998; Knot et al., 1998). These results indicate that Ca2⫹ entry through voltage-dependent calcium channels regulates [Ca2⫹]i and Ca2⫹ spark frequency. In cardiac muscle, local entry through single voltage-dependent calcium channels rapidly and effectively activates Ca2⫹ sparks. In contrast, the communication from voltage-dependent calcium channels to Ca2⫹ sparks appears to be markedly different in smooth muscle. In cardiac muscle, the calcium current activates, within milliseconds, many sparks to cause a global calcium transient (Cannell et al., 1995; Lopez-Lopez et al., 1995). In contrast, the calcium current in smooth muscle (urinary bladder) activates a small number (1–3) of calcium sparks with a relatively long delay (50 ms)(Collier et al., 2000; see also Herrera et al., 2002). In cardiac muscle, the activation of calcium sparks depends on the flux of calcium ions through single voltage-dependent calcium channels (local communication) (Niggli and Lederer, 1990; Cannell et al., 1995; Lopez-Lopez et al., 1995; Lipp and Niggli, 1996; Collier et al., 1999). In smooth muscle, this is not the case: activation of calcium

Calcium sparks and membrane potential

39

A

B

Figure 5.4 Depolarization increases frequency of Ca2⫹ sparks in intact rat cerebral arteries. (A) Average fluorescence over 10 s (100 images averaged) of 56.3 ⫻ 52.8-␮m areas from same artery bathed in 6 mM (left) and 30 mM (right) extracellular K⫹ (K⫹o). Elevating external K⫹ increased global [Ca2⫹]i fluorescence (F/Fo) 1.62-fold in this artery. Local [Ca2⫹]i transients were detected by eye and are indicated by labeled boxes (1.54⫻1.54 ␮m). (B) Local F/Fo changes with time for corresponding boxes. In 6 mM K⫹o (left) over a 10 s period, 1 Ca2⫹ spark occurred (a). Changing bath solution to 30 mM K⫹o solution (right) increased frequency of Ca2⫹ sparks. [Taken from Jaggar et al., Am. J. Physiol, 1998b, 274: C1755–C1761.]

sparks depends on a rise in global cytoplasmic and sarcoplasmic reticulum calcium (Collier et al., 2000; Wellman et al., 2001). These results are consistent with the idea of spatial separation of voltage-dependent calcium channels in the surface membrane and ryanodine receptors in calcium spark sites in smooth muscle.

2.3. Activation of BKCa channels by Ca2⫹ sparks Calcium sparks in arterial smooth muscle are asynchronous and occur at a frequency (1–3 sparks/s/cell) which would have little impact on average cytoplasmic calcium. Calcium sparks cause a substantial activation of large conductance, BKCa channels. BKCa channels brake pressure-induced membrane depolarization of arterial smooth muscle cells, and thus, act as a negative-feedback element to oppose vasoconstriction (Brayden and Nelson, 1992; Rusch et al., 1992; Knot et al., 1998). These channels, which are activated by [Ca2⫹]i and membrane potential depolarization, have an exceedingly low Po of approximately 10⫺6 at physiological membrane potentials (~⫺40 mV) and [Ca2⫹]i (~200 nM), characteristic of intact pressurized arteries. This low Po under physiological conditions is at odds with the

40

Delrae M. Eckman et al.

clear contribution made by BKCa channels to the regulation of the membrane potential of smooth muscle cells in intact arteries (Brayden and Nelson, 1992; Nelson and Quayle, 1995; Knot et al., 1998). The discovery of Ca2⫹ sparks in smooth muscle (Nelson et al., 1995) provided a resolution to this dilemma, that is, a means to deliver sufficient Ca2⫹ to elevate BKCa channel activity to a level that would contribute to the regulation of membrane potential. In smooth muscle simultaneous measurements of calcium sparks and BKCa channel currents in arterial myocytes indicate that most calcium spark sites are in close apposition to BKCa channels in the cell membrane and that a single calcium spark delivers 10–100 ␮M Ca2⫹ to elevate the open probability of about 20–30 nearby BKCa channels about 104- to 106-fold (Perez et al., 1999, 2001; Figure 5.5).

2.4. Regulation of arterial membrane potential and diameter by Ca2⫹ sparks A single calcium spark causes an approximate 20 mV membrane hyperpolarization of isolated arterial myocytes (Jaggar et al., 2000) (Figure 5.6). In intact pressurized cerebral arteries, blocking BKCa channels with iberiotoxin or calcium sparks with ryanodine causes a non-additive membrane depolarization of about 10 mV (Nelson et al., 1995; Knot et al., 1998) (Figure 5.7). These results are consistent with the idea that calcium sparks can regulate the membrane potential of arterial smooth muscle through activation of BKCa channels. Membrane potential regulates arterial diameter by controlling calcium entry through voltage-dependent calcium channels, and thereby [Ca2⫹]i (Nelson et al., 1990). For example, a membrane potential depolarization of 10 mV, induced by elevating external potassium, blocking BKCa channels or inhibiting Ca2⫹ sparks, elevated [Ca2⫹]i by 45 nM and constricted pressurized cerebral arteries by 50 ␮m (Nelson et al., 1995; Knot and Nelson, 1998; Knot et al., 1998).

2.5. Vasodilators increase calcium spark and transient BKCa channel currents Calcium sparks, through activation of BKCa channels, regulate arterial smooth muscle membrane potential (Brayden and Nelson, 1992; Nelson et al., 1995; Knot et al., 1998) (Figure 5.8). Therefore, activation of calcium sparks should cause membrane hyperpolarization through stimulation of BKCa channels. Indeed, activation of BKCa channels contributes to the mechanism of action of many vasodilators which elevate cAMP or cGMP (Nelson and Quayle, 1995; Jaggar et al., 2000). Vasodilators can activate BKCa channels in part through increasing calcium spark frequency. In coronary artery smooth muscle cells, elevation of cGMP by sodium nitroprusside and nicorandil increased Ca2⫹ spark frequency (Jaggar et al., 1998; Porter et al., 1998) as well as transient BKCa currents (Robertson et al., 1993; Wellman et al., 1996). Similarly, in cerebral artery myocytes, forskolin and cAMP analogs increased Ca2⫹ spark frequency, an effect which was reversed by using the PKA-inhibitor, H-89 (Porter et al., 1998; Wellman et al., 2001). One logical target of cAMP- or cGMPdependent protein kinase is PLB, which upon phosphorylation by these kinases dissociates from the SR Ca-ATPase, leading to increased calcium uptake into the sarcoplasmic reticulum, and hence elevated sarcoplasmic reticulum calcium load. Forskolin increased sarcoplasmic reticulum Ca2⫹ load and spark frequency in control (Porter et al., 1998; Wellman et al., 2001),

Calcium sparks and membrane potential

41

A

B

Figure 5.5 Ca2⫹ sparks activate BKCa channel currents in smooth muscle cells from cerebral arteries. (A) Original sequence of two-dimensional confocal images obtained every 8.33 ms of an entire smooth muscle cell (top), followed by subsequent images of region of interest (dotted box) illustrating time course of fractional increase in fluorescence (F/Fo) and decay of a typical Ca2⫹ spark. Images are color coded as indicated by color bar. (B) Simultaneous BKCa channel current and Ca2⫹ spark (F/Fo) measurements at 40 mV illustrating temporal association. Current (blue) is indicated above F/Fo average (red and green) of red and green boxes (2.2 ␮m/side) indicated in (A), respectively. Purple bar, segment of trace illustrated in panel A [Taken from Perez et al., The Journal of General Physiology, 1999, vol. 113, pp. 229–238] (see Color Plate 5).

but not in PLB knockout mice (Wellman et al., 2001), suggesting a central role for PLB. The observations are consistent with the idea that modulation of calcium spark frequency is an effective means to regulate the membrane potential of arterial smooth muscle through alterations in BKCa channel activity.

A

0.5 F/Fo 500 ms B

Voltage-clamp 2 F/Fo

HP = –40 mV

C

Current-clamp

2 F/F0 – 40

30 pA

mV – 60 1s

Figure 5.6 (A) 3D Ca2⫹ spark form an arterial smooth muscle cell and the corresponding fluorescent ratio showing three Ca2⫹ sparks recorded from the same site. (B) Simultaneous recordings from an arterial smooth muscle cell of Ca2⫹ sparks and STOCs (BKCa currents) under voltage clamp (holding potential ⫺40 mV), and (C) Ca2⫹ sparks and STHs (spontaneous transient membrane hyperpolarizations) under current clamp.

A

B

Figure 5.7 Inhibitors of ryanodine-sensitive calcium release channels (ryanodine) and BKCa channels (iberiotoxin, IbTx) depolarize pressurized cerebral arteries. (A) Ryanodine (1⫻10⫺5) depolarizes the membrane potential of a pressurized (to 60mmHg) cerebral artery, and IbTx (1⫻10⫺7) has no effect in the presence of ryanodine (1⫻10⫺5). (B) IbTx (1⫻10⫺7) depolarizes the membrane of a pressurized (to 60mmHg) cerebral artery, and ryanodine has no effect in the presence of IbTx. [Modified from Knot et al., J. Physiol., 1998, 508: 211–221.]

Calcium sparks and membrane potential

43

A

B

Figure 5.8 Ryanodine increases arterial wall Ca2⫹ and constricts pressurized cerebral arteries. Original recordings of the simultaneous measurement of arterial wall [Ca2⫹] and diameter in pressurized (to 60 mm Hg) posterior cerebral arteries loaded with fura-2. Elevation of intravascular pressure from 10 to 60 mm Hg caused a sustained elevation in arterial wall calcium and a sustained constriction. Ryanodine and high potassium (61 mm) increased on arterial wall [Ca2⫹] (A) and decreased arterial diameter (B). IbTx was without effect in the presence of ryanodine. The dihydropyridine calcium channel inhibitor nimodipine (Nimod) decreased arterial wall calcium in the presence of ryanodine and IbTx. The level of arterial wall calcium at 60 mm Hg in the absence of drugs or high potassium is indicated by the horizontal dotted line. [Modified from Knot et al., J. Physiol., 1998, 508: 211–221.]

2.6. Tuning Ca2⫹ sparks to the BKCa channel The activity of BKCa channels appears to be finely calibrated to the calcium delivered by calcium sparks. Thus, alteration in the calcium-sensitivity of the BKCa channel should affect their coupling to calcium sparks, and thereby destabilize regulation of membrane potential by BKCa channels. A molecular calibrator or tuner of the calcium sensitivity of BKCa channels is its accessory protein, the ␤1-subunit (Knaus et al., 1994). BKCa channels consist of a pore forming ␣-subunit and an accessory ␤1-subunit. The ␤1subunit elevates the apparent calcium- and voltage-sensitivity of the ␣-subunit (Cox and Aldrich, 2000). The ␤1-subunit appears to be exclusively expressed in smooth muscle (Brenner et al., 2000). To probe the role of the BKCa channel in arterial function, the effects of deletion of the ␤1-subunit gene on BKCa channels, their activation by calcium and calcium sparks, on arterial membrane potential and diameter, and on blood pressure was explored (Brenner et al., 2000; Pluger et al., 2000). Single BKCa channels in excised patches from cerebral artery myocytes had greatly reduced calcium- and voltage-sensitivity when compared to control. The coupling of calcium sparks to BKCa channels was reduced markedly (Figure 5.9).

BKCa currents (pA)

A

10 µ

µm

F/Fo 2.0 1.5 1.0

20

β1-KO

10

β1-KO uncouples Ca2+ sparks from BKCa currents

20 pA

Ca2+ sparks

40

3 1 2 Ca2+ spark amplitue (F/Fo)

β1-KO BKCa currents

Control

0

m

WT

60

1 F/Fo

B

%

100 80 60 40 20 0 C

on tro l β1 -K O

1s

Figure 5.9 Decreased coupling of calcium sparks to (BKCa) channels in ␤1-KO myocytes. Consecutive pseudocolor three-dimensional images of (A) ␤1-KO cell obtained every 8.33 ms. (B) Simultaneous BKCa current (blue) and Ca2⫹ spark measurements (fractional fluorescence, F /Fo) from a control cell with one spark site and a ␤1-KO cell with two spark sites (red and green) (⫺40 mV). Inset top: relationships between BKCa current and Ca2⫹ spark amplitudes in control cells (blue, 94 sparks, 6 cells) and ␤1-KO cells (red, 71 sparks, 7 cells). Lines represent linear regression fit (slope control ⫽ 37.8 1.6 versus slope ␤1-KO ⫽ 6.7 0.5, P ⬍ 0.001). Inset bottom: percentage of Ca2⫹ sparks causing transient BKCa currents in control (blue) and 1-KO (red) cells [Modified from Brenner et al., Nature, 2000, 407: 870–876.] (see Color Plate 6).

A

0

75 0

% Constriction

C

10 Time (min)

60

105

40

95

20

85

20

D

20

0

10

* 0

0 0

Em (mV)

Diameter (µm)

20

115

10 20 Time (min)

30

Pressure (mm Hg) 10 60 10 60

–20 –40 *

–60 20 40 60 Pressure (mm Hg)

Pressure (mm Hg)

40

85

Pressure (mm Hg)

60

Diameter (µm)

B 95

*

Figure 5.10 Removal of the ␤1-subunit of the BKCa channel depolarizes and constricts mouse cerebral arteries. Examples of pressure induced vasoconstrictions in control (A, blue) and ␤1subunit gene ablated (B, red) mice (gray line indicates pressure step). Isolated cerebral arteries from ␤1-subunit gene ablated mice (red) constrict more to pressure than controls (blue) (C). Pressurized (10 mm Hg, solid bar or 60 mm Hg, hatched color) cerebral arteries from ␤1-KO mice (red) are depolarized compared to controls (blue) (D) [Modified from Brenner et al., Nature, 2000, 407: 870–876.] (see Color Plate 7).

Calcium sparks and membrane potential

45

Smooth muscle cells in pressurized arteries from ␤1-knockout mice were depolarized by about 10 mV when compared to control (Figure 5.10), and the arteries exhibited significantly elevated pressure-induced constrictions. The membrane potential and diameter of pressurized arteries from ␤1-knockout mice were insensitive to the BKCa channel blocker, iberiotoxin, supporting the idea that the effects of ␤1-gene deletion were manifested through diminished BKCa channel activity. Consistent with an important role of the BKCa channel and its ␤1-subunit in the regulation of vascular tone, ␤1-KO mice had elevated blood pressure and cardiac hypertrophy (Brenner et al., 2000).

3. CONCLUSIONS These results support the concept that a local calcium signal – the calcium spark – through its activation of BKCa channels can regulate global [Ca2⫹]i, and hence arterial tone through

Ca2+

Ca2+

Figure 5.11 Possible mechanisms of action of protein kinase A (PKA), protein kinase G (PKG) and protein kinase C (PKC) on Ca2⫹ sparks, BKCa channels, and sarcoplasmic reticulum Ca2⫹-ATPase in arterial smooth muscle cells. Activation of PKA or PKG increases Ca2⫹ spark frequency and increases Ca2⫹ load of the sarcoplasmic reticulum (probably through activation of the sarcoplasmic reticulum Ca2⫹-ATPase via disinhibition of PLB). Increased Ca2⫹ spark frequency could occur due to a direct effect on the ryanodine receptor channel and/or by a secondary effect from increased sarcoplasmic reticulum Ca2⫹ load. PKA/PKG activation also increases the activity of BKCa channels, which could manifest itself by an elevation in BKCa channel current amplitude and steady-state BKCa channel activity. Synergistic effect of increased Ca2⫹ spark frequency and a direct effect on BKCa channels will result in a significant increase of BKCa channel activity. Membrane potential hyperpolarization closes L-type Ca2⫹ channels, which reduces Ca2⫹ influx, lowering the cytoplasmic Ca2⫹ concentration, and leads to vasodilation. Activation of PKC reduces frequency of Ca2⫹ sparks and amplitude of BKCa channel currents, due to a direct inhibitory effect on ryanodine receptor channels and BKCa channels, respectively. Additive effect results in decreased BKCa channel activity, a depolarization of the smooth muscle cell membrane potential, and activation of L-type Ca2⫹ channels [Taken from Jaggar et al., Am. J. Physiol., 2000, 278: C235–C256.] (see Color Plate 8).

46

Delrae M. Eckman et al.

control of the membrane potential. Regulation of calcium spark frequency through changes in calcium load of the sarcoplasmic reticulum appears to be an important mechanism of action of vasodilators. Matching the calcium signal to calcium sensitivity of the target protein appears to be critical for a physiological outcome. Detuning this communication by deletion of the ␤1-subunit of the BKCa channel has a significant impact on membrane potential and tone of arterial smooth muscle and on blood pressure (see Figure 5.11). ACKNOWLEDGEMENTS The authors would like to thank our collaborators on the different aspects of this project as well as support from the NIH, NSF, American Heart Association, and the Totman Trust for Medical Research.

6

Proteinase-activated receptor-2: release of an endothelium-derived hyperpolarizing factor distinct from that released by acetylcholine J.J. McGuire, Y. Gui, X. Wang, R.D. Loutzenhiser, C.R. Triggle and M.D. Hollenberg Proteinase-activated receptor-2 (PAR2) is a G protein-coupled receptor activated by serine proteinases like trypsin, which unmasks an N-terminal cryptic tethered-ligand, or by synthetic peptides that mimic the tethered-ligand. In capacitance vessels (aorta, femoral), PAR2 causes an endothelium-dependent, NO-mediated relaxation. It was hypothesized that PAR2 would also release an endothelium-derived hyperpolarizing factor (EDHF) from resistance vessels like mesenteric or renal afferent arterioles. Using SLIGRL-NH2 or trypsin as agonists in either isolated murine mesenteric rings or in perfused rat renal preparations in vitro, the relaxant effects of activating PAR2 were evaluated in the absence or presence of inhibitors of cyclooxygenases, NO synthase and soluble guanylyl cyclase, of elevated extracellular K⫹ and of the combination charybdotoxin plus apamin. The effects of PAR2 agonists were compared with the actions of acetylcholine. In contrast with the relaxant effects of SLIGRL-NH2 and acetylcholine in murine aorta or femoral artery, that were both blocked by combined NO synthase/cyclooxygenase/soluble guanylyl cyclase inhibition, in small mesenteric arteries the endothelium-dependent relaxant actions of SLIGRL-NH2 essentially were unaffected by these inhibitors whilst vasodilatation by acetylcholine was blocked. In the rat renal preparations, trypsin and SLIGRL-NH2 caused an initial transient vasodilatation. The transient vasodilatation was resistant to NO synthase/cyclooxygenase/soluble guanylyl cyclase inhibition, but was absent in the presence of 25 mM K⫹, pointing to an endothelium-dependent hyperpolarizing activity. However, PAR2-mediated dilatation was not blocked by charybdotoxin plus apamin as was the response to acetylcholine. In conclusion, PAR2 causes an endothelium-dependent hyperpolarization differing from that mediated by acetylcholine. In addition, different G proteincoupled receptors acting in the same tissue can produce distinct endothelium-dependent hyperpolarizing factors.

1. INTRODUCTION G protein-coupled receptors on endothelial cells including muscarinic, neurokinin and bradykinin receptors, are known to produce both nitric oxide-dependent and independent vasodilatation of various blood vessels (Furchgott and Zawadzki, 1980; Cherry et al., 1982; De Mey et al., 1982; Regoli et al., 1984). PAR2 (Hollenberg and Compton, 2002) differs fundamentally from these other endothelial receptors by the fact that the specific proteolytic activity of several extracellular serine proteases, rather than a classic soluble extracellular ligand, reveals a self-contained amino-terminus tethered-ligand that then changes the activation state of this receptor (Vu et al., 1991; Nystedt et al., 1994). However, serine protease-mediated activation of PAR2 such as that caused by trypsin, factor X or tryptase can be mimicked by using synthetic peptides that correspond to the revealed tethered-ligand activating sequence, that is, SLIGRL-amide for PAR2 (Hollenberg et al., 1997).

48

J.J. McGuire et al.

The endothelium-dependent vasodilator mechanisms elicited by Acetylcholine, Substance P and bradykinin cause the endothelium-dependent vasodilator responses in many isolated blood vessel preparations in vitro. Studies regarding the PAR2-mediated actions on blood vessels in vitro have been limited to the aorta, pulmonary, renal and coronary arteries of humans, pigs and rats, renal afferent arterioles of rats, human umbilical veins, and the small mesenteric arteries and renal artery of mice (reviewed by Cicala, 2002; McGuire et al., 2002). In large arteries, PAR2 stimulation primarily utilizes a NO synthase-dependent mechanism in vitro. Coronary, renal afferent and mesenteric blood vessel preparations are resistance vascular preparations and EDHFs with distinct properties were proposed to mediate the responses of PAR2 activation in these preparations (McGuire et al., 2002; McLean et al., 2002; Trottier et al., 2002). This chapter presents a comparison of the endothelium-dependent responses of large conductance (aorta and femoral) and small (100 ␮m) murine blood vessels (mesenteric) upon activation of PAR2. Further, the ability of PAR2 activation to cause both NO-dependent and NO-independent increases in renal perfusion flow in an intact perfused kidney preparation is described. Finally, the distinct EDHF mechanisms are described for the vasodilator responses elicited by PAR2, and muscarinic activation in renal afferent arterioles of an intact perfused hydronephrotic rat kidney preparation.

2. METHODS 2.1. Materials Acetylcholine, angiotensin II, apamin, barium chloride, bradykinin, catalase, charybdotoxin, cinnamyl-2,4-dihydroxy-a-cyanocinnamide, cirazoline, N␻-nitro-L-arginine methylester hydrochloride (L-NAME), nordihydroguaiaretic acid, 1H-[1,2,4]-oxadiazolo[4,5–1] quinoxalin-1-one (ODQ) and ouabain were supplied by Sigma (St Louis, MO). Baiclein (5,6,7-trihydroflavone), cinnamyl-2,4-dihydroxy-a-cyanocinnamide, and MK886 (same as L-663, 536) were purchased from Biomol (Plymouth Meeting, PA). The peptide SLIGRLNH2 was synthesized by Denis McMaster (University of Calgary, AB) and purified by HPLC to ⬎ 95% purity.

2.2. Bioassay using murine aorta, mesenteric and femoral arterial rings Vascular tissues, obtained from male mice (8–14 weeks of age; 20–40 g), were prepared for bioassay (McGuire et al., 2002). NOS3(⫺/⫺) and C57BL/6J background mice were obtained from the Jackson Laboratory (Bar Harbor, ME). All animals were cared for in accordance with the guidelines of the Canadian Council on Animal Care. In brief, after sacrifice, appropriate arterial segments were isolated using a dissection light microscope and four to eight rings (1–2 mm length) were cut from the isolated blood vessels to be used for isometric tension measurements. Rings were suspended between a micropositioner and force transducer by either hooks (aorta; 200-␮m diameter) or gold-plated tungsten wires (mesenteric and femoral arteries; 20-␮m diameter) in a Mulvany–Halpern style organ bath (5 ml volume). Isometric tension was recorded on-line via serial connection to a computer hard drive at a rate of 1 Hz. Resting tensions of 1 mN (mesenteric), 1.5 mN (femoral) or 5 mN (aorta) were fixed for an initial equilibration period of 1 h. Pharmacological reagents were added directly to the tissue bath containing a modified Kreb’s-bicarbonate solution

Proteinase-activated receptor-2 49 (118 mM NaCl, 4.7 mM KCl, 0.87 mM MgSO4, 0.86 mM K2HPO4, 2.5 mM CaCl2, 10 mM D-glucose, and 25 mM NaHCO3) gassed with 95% O2/5% CO2 (pH 7.4 at 37 ⬚C). Inhibitors were added 20 min prior to evoking a contraction. Tissue viability was monitored routinely by measuring a contractile response to 120 mM KCl.

2.3. Perfused normal rat kidney preparation Male albino Sprague–Dawley rats (250–300 g) were used in accordance with the Canadian Council on Animal Care. Animals were anaesthetized with halothane and methoxyflurane. The perfused kidney preparation (Loutzenhiser et al., 1989), was employed to examine the effects of PAR2 activation on renal perfusion flow. Kidneys were perfused continuously with oxygenated (95% O2, 5% CO2), modified Kreb’s-Ringer buffer, containing 30 mM sodium bicarbonate, 5 mM D-glucose, and 5 mM HEPES. To avoid the use of albumin, which would preclude testing the system with trypsin and that might sequester the peptide, SLIGRL-NH2, 4.5% (w/v) dextran (average molecular weight 6500; Sigma) was used as an oncotic agent in lieu of albumin. Renal arterial pressure was maintained constant at 100 mm Hg to ensure adequate glomerular filtration. The total volume of the recirculating perfusion system was 80 ml. The renal perfusion flow rate was monitored by a transit-time flowmeter (model T106, Transonic Systems, Ithaca, NY). Kidneys were allowed to recover during an initial 20 min equilibration period, before any measurements were obtained. Tone was established with angiotensin II, which was added as a single bolus plus constant infusion at 0.1 nM for 10 min, followed by a single bolus and constant infusion of either the PAR2-activating peptide, SLIGRL-NH2 or trypsin (⬍ 2U/ml) for 10 or 20 min at the indicated concentrations.

2.4. Isolated perfused rat hydronephrotic kidney The in vitro perfused hydronephrotic rat kidney preparation (Trottier et al., 1998; Trottier et al., 2002) was used to assess the effects of agonists (SLIGRL-NH2, acetylcholine, bradykinin) on the angiotensin II-constricted (0.1 nM) renal afferent arteriole. Unilateral hydronephrosis was induced in male Sprague–Dawley rats and kidneys were harvested at 6–8 weeks post-ligation at which time tubular atrophy allows for direct visualization of the microvasculature. After cannulation of the renal artery in anaesthetized animals, the organ was excised and the isolated kidney was perfused using a single-pass perfusion system. Perfusion pressure was maintained at 80 mm Hg using Dulbecco’s modified Eagles medium (Sigma) containing 30 mM sodium bicarbonate, 5 mM glucose and 5 mM HEPES, pH 7.4. The perfusion medium was equilibrated with 95% air–5% CO2 and was maintained at 37 ⬚C. Pharmacological agents were added directly to the perfusate and afferent arteriolar diameter (␮m) was monitored microscopically.

2.5. Statistical analysis Values reported in the text and graphs (symbols or bars) represent the mean ⫾ standard error of the mean (error bars on graphs) from n animals. Student’s t-test for unpaired data was used for comparisons of the means between two groups. One-way Analysis of Variance (ANOVA) followed by Student–Newman Keuls post-hoc test was used for comparison between the means of more than two groups. A P less than 0.05 was considered to indicate a statistically significant difference.

50

J.J. McGuire et al.

3. RESULTS 3.1. Murine aorta and femoral artery The PAR2-activating peptide, SLIGRL-NH2 (10⫺5 M), induced relaxation of cirazolinecontracted aortic rings with endothelium from wild-type mice, but not those aortae from NOS3 (⫺/⫺) mice (Figure 6.1). In complete accord with the data obtained with aortic preparations from the NOS3(⫺/⫺) mice, the SLIGRL-NH2-induced relaxation of cirazoline contracted murine femoral arteries with endothelium from wild-type animals was inhibited completely by treatment of these arteries with a combination of 10-4 M L-NAME plus 10⫺5 M ODQ (Figure 6.1). After removal of the endothelium, SLIGRL-NH2 failed to relax aorta or femoral preparations from wild-type mice (data not shown).

3.2. Murine mesenteric arterioles SLIGRL-NH2 (10⫺7–10⫺4 M) caused a concentration-dependent relaxation of cirazolinecontracted small mesenteric arteries with endothelium from both wild-type mice (C57BL/6J) A

B 20 Tension (mN)

Cirazoline

Wash

15 10 5 0 0

20 Tension (mN)

Time (min)

NOS3 (–/–) Aorta

C57BL/6J Femoral arteries

10 µM SLIGRL-NH2

75 Percent relaxation

C57BL/6J Aorta

30

50

25

0

* Control L-NAME + ODQ 10 µM SLIGRL-NH2

10 µM SLIGRL-NH2 Wash

Cirazoline

15 10 5 0 0

Time (min)

30

Figure 6.1 PAR2-induced relaxation of murine aortae and femoral arteries is solely endothelial NOS3-dependent. (A) Representative data illustrate the typical isometric tension responses of contracted rings of aorta from wild-type (upper panel A) and NOS3-deficient mice (lower panel A) after exposure to PAR2-activating peptide, SLIGRL-NH2. (B) A summary of data describing the maximal relaxation of contracted murine femoral arteries induced by SLIGRL-NH2 (10⫺7–10⫺4 M) in the absence and presence of inhibitors of NO synthase and soluble guanylyl cyclase (n ⫽ 8; the asterisk indicates P less than 0.05, two-tailed Student’s t-test for unpaired data).

Proteinase-activated receptor-2 51 A

B

C57BL/6J small mesenteric arteries

10 SLIGRL-NH2 (0.1 to 100 µM)

9

L-NAME + ODQ + indomethacin

9

8

8

7

7

Tension (mN)

6 5 4 3

SLIGRL-NH2 (0.1 to 100 µM)

6 5 4 3

2

2 Cirazoline

1

0 0

0

20

Cirazoline 0

15

Time (min)

Time (min)

100

8 Tension (mN)

7 6

Wash

5 4 3 2 1

Cirazoline

* 75

* 50

25

0

0 0

20 Time (min)

L-NAME

*

* BaCl2 + ouabain

Maximum relaxation by SLIGRL-NH2 (%)

SLIGRL-NH2 (0.1 to 100 µM)

9

C57BL/6J small mesenteric arteries

apamin + ChTX

D

10

30 mM KCl

NOS3 (–/–) small mesenteric arteries

C

Control

1

Control

Tension (mN)

C57BL/6J small mesenteric arteries

10

+ ODQ + indomethacin

Figure 6.2 NO/prostacyclin-independent EDHF mechanisms mediate the PAR2-agonist induced vasodilatation of murine small mesenteric arteries. Representative data illustrate changes in isometric tension of contracted mesenteric arterioles from wild-type mice in either the (A) absence of or (B) presence of inhibitors of NO synthase, soluble guanylyl cyclase plus cyclooxygenases and from (C) NOS3 (⫺/⫺) deficient mice. (D) Summary of the effects of various pretreatment protocols on the maximal PAR2-mediated relaxations of small mesenteric arteries from wild-type mice. The asterisk indicates P less than 0.05, one-way ANOVA compared to control. ChTX ⫽ charybdotoxin. (Adapted from McGuire et al., 2002.)

and NOS3 (⫺/⫺) mice, even in the presence of a combination of 10⫺4 M L-NAME, 10⫺5 M ODQ plus 10⫺5 M indomethacin (Figure 6.2; McGuire et al., 2002). Treatment of small mesenteric arteries from wild-type mice with 10⫺4 M L-NAME, 10⫺5 M ODQ plus 10⫺5 M indomethacin reduced the maximal relaxation to SLIGRL-NH2 by 15% (Figure 6.2; (McGuire et al., 2002) ). However, combinations of either 10⫺6 M apamin plus 10⫺7 M

52

J.J. McGuire et al.

charybdotoxin or 3 ⫻ 10⫺5 M BaCl2 plus 10⫺5 M ouabain in the presence of L-NAME, ODQ and indomethacin caused greater inhibition (80% and 30%, respectively). Similar inhibition (80%) was seen when tissues were contracted with 30 mM KCl rather than cirazoline (Figure 6.2; P ⬍ 0.05, one-way ANOVA compared to controls; (McGuire et al., 2002) ).

3.3. Whole kidney glomerular perfusion rate Perfusion of intact isolated rat kidneys with SLIGRL-NH2 caused an increase in renal perfusion flow rate in preparations that had been pre-constricted with 10⫺10 angiotensin II. Upon exposure of the preparation to 10⫺10 angiotensin II, renal perfusion flow was reduced from 23 ⫾ 3 ml/min/g tissue (n ⫽ 6) to 14 ⫾ 3 ml/min/g (n ⫽ 6). In the continued presence of angiotensin II, the addition of 10 ␮M SLIGRL-NH2 to the perfusate increased the renal perfusion flow rate to 21 ⫾ 3 ml/min/g (n ⫽ 4). In the angiotensin II-constricted preparations, treatment of the preparations with 10⫺4 M L-NAME did not prevent a rapid initial transient vasodilatation caused by SLIGRL-NH2 (P ⬎ 0.05, Student’s t-test for unpaired data). The infusion of trypsin (2 U/ml, a concentration sufficient to activate PAR2) caused a L-NAME/ODQ-insensitive increase in renal perfusion flow that was comparable to that caused by the PAR2-activating peptide, SLIGRL-NH2 in the presence of L-NAME and ODQ (n ⫽ 6, P ⬎ 0.05, Student’s t-test for unpaired data).

3.4. Renal afferent arteriolar vasodilatation In the intact perfused hydronephrotic rat kidney preparation, SLIGRL-NH2 caused an L-NAME-insensitive reversal of angiotensin II induced vasoconstriction that was also unaffected by charybdotoxin plus apamin (n⫽3, controls, 6⫾1␮m vasodilatation versus 5⫾1␮m vasodilatation after treatment with charybdotoxin plus apamin; also see Wang and Loutzenhiser, 2002a). In contrast, treatment with L-NAME in combination with charybdotoxin plus apamin was sufficient to inhibit vasodilatation caused by acetylcholine (100% inhibition (n ⫽ 6); also see Wang and Loutzenhiser, 2002a). Thus, the inhibitory profiles for the actions of PAR2 and muscarinic receptor-mediated EDHF release were distinct.

3.5. Lipoxygenases, vanilloid receptors and hydrogen peroxide SLIGRL-NH2 (10⫺5 M)-induced vasodilatation was not inhibited by pretreatment of murine small mesenteric arteries with L-NAME plus ODQ plus indomethacin in combination with either (a) inhibitors of lipoxygenases (10⫺5 M nordihydroguarectic acid, 10⫺5 M baiclein, 3 ⫻ 10⫺6 M cinnamyl-2,4-dihydroxy-a-cyanocinnamide or 10⫺7 M MK886), (b) vanilloid receptor antagonist (3 ⫻ 10⫺6 M capsazepine) or (c) catalase (1300 Units/ml) (Figure 6.3).

4. DISCUSSION The vasodilatation responses of vascular tissues in vitro to PAR2 stimulation are mediated by endothelium-derived factors. These vasodilator factors are produced selectively by the endothelium of different vascular beds upon activation of PAR2, as well as, by other G proteincoupled receptors such as those used by muscarinic agonists. Endothelium-derived nitric oxide is the sole mediator of the PAR2 elicited vasodilatation of the large conduit-type vessels (aorta and femoral artery) whereas non-NO EDHFs are the primary mediators of the

Proteinase-activated receptor-2 53

Relaxation (%)

100

75

50

25

0

Control

NDHGA

Baiclein

MK886

CDC

Capsazepine Catalase

Treatment

Figure 6.3 Lack of involvement by lipoxygenases, vanilloid receptor and hydrogen peroxide in the NO/prostacyclin-independent PAR2-mediated relaxation of murine small mesenteric arteries. A summary of data describing the relaxations of contracted murine small mesenteric arteries induced by 10⫺5 M SLIGRL-NH2 in the presence of L-NAME plus ODQ plus indomethacin with the addition of inhibitors of lipoxygenases (10⫺5 M nordihydroguaiaretic acid (NDHGA), 10⫺4 M baiclein, 10⫺7 M MK886 or 3⫻10⫺6 M cinnamyl-2,4-dihydroxy-acyanocinnamide (CDC)), vanilloid receptor antagonist (3⫻10⫺6 M capsazepine) or catalase (1300 U/ml). Differences between means were not statistically significant.

PAR2 elicited vasodilatation of the small resistance-type blood vessels such as the small mesenteric arteries and renal afferent arterioles (McGuire et al., 2002; Trottier et al., 2002). The expression of endothelial NO synthase was obligatory for the endothelium-dependent relaxation of murine aorta mediated by both muscarinic (Huang et al., 1995) and PAR2 stimulation in vitro. The complete inhibition of PAR2-mediated vasodilatation by L-NAME and ODQ in murine femoral arteries also demonstrates that the primary signal transduction mechanism involves endothelium-derived nitric oxide and stimulation of soluble guanylyl cyclase (Figure 6.1B). These results are in keeping with the vascular responses produced by acetylcholine-triggered muscarinic receptors in wild-type mice (Waldron et al., 1999) and by SLIGRL-NH2 activation of PAR2 in the aorta of rats (Saifeddine et al., 1996). In contrast with muscarinic receptor activation in the endothelium-intact femoral arteries of NOS3 (⫺/⫺) mice (Waldron et al., 1999), PAR2 activation in these same blood vessels did not cause vasodilatation (McGuire et al., 2000). Thus, the signal transduction mechanisms linking PAR2 and muscarinic receptors to EDHF production in the femoral artery of NOS3(⫺/⫺) mice must differ. These results indicate that: (a) there are selective compensatory mechanisms for endothelial G protein-coupled receptor-mediated EDHF release that are induced by NOS3 deficiency; and (b) the activation of distinct G protein-coupled receptors (e.g. muscarinic versus endothelial PAR2 in the femoral artery of NOS3(⫺/⫺) mice) can selectively release a distinct EDHF. The PAR2-induced vasodilatation in small mesenteric arteries of both wild-type and NOS3(⫺/⫺) mice indicates that non-endothelial NO synthase endothelium-dependent mechanisms (i.e. EDHFs), are constitutive in this tissue (McGuire et al., 2002). These data differ from those results obtained with muscarinic activation of these same blood vessels, wherein the acetylcholine-induced relaxation was entirely inhibited by treatment with

54

J.J. McGuire et al.

inhibitors of NO synthase/soluble guanylyl cyclase/cyclooxygenases in wild-type mice; and yet, there is a compensatory up-regulation of cyclooxygenases or an EDHF in these mesenteric vessels from NOS3(⫺/⫺) mice (Chataigneau et al., 1998; Waldron et al., 1999). Treatment of the mesenteric tissues with 30 mM KCl, apamin plus charybdotoxin or BaCl2 plus ouabain inhibited to various degrees the PAR2-mediated vasodilatation. Based on the similar pharmacological profile of the PAR2 and muscarinic receptor mediated responses in small mesenteric arteries of NOS3(⫺/⫺) mice (Waldron et al., 1999), this compensatory mechanism for muscarinic-induced vasodilatation appears to be similar to the constitutive PAR2-mediated mechanism (McGuire et al., 2002). Therefore, the ability of endothelial G protein-coupled receptors (e.g. muscarinic) to elicit vasodilatation can be selectively altered by changes in the gene expression of NOS3 (Waldron et al., 1999). And yet endothelial NO synthase deficiency may not affect other G protein-coupled mechanisms for EDHF release, for example, for PAR2 activation in the murine small mesenteric arteries. These interpretations must be taken in light of the differences between the pharmacological characteristics of the EDHF-mediated responses reported for these same blood vessels by different investigators using muscarinic agonists, including the lack of sensitivity to either catalase or indomethacin (Chataigneau et al., 1998; Matoba et al., 2000). These data differ from the results observed in response to PAR2 activation (McGuire et al., 2002). Although in murine blood vessels the EDHFs produced by PAR2 and muscarinic receptors are both inhibited by a selective combination of inhibitors of the calcium-activated K⫹ channels, apamin plus charybdotoxin, those EDHFs resulting from activation of muscarinic, PAR2 and bradykinin receptors are different in renal afferent arterioles of hydronephrotic rats (Wang and Loutzenhiser, 2002b). Activation of PAR2 produced a transient vasodilatation of rat renal afferent arterioles that was partially resistant to inhibitors of NO synthase and cyclcooxygenases, but insensitive to apamin plus charybdotoxin, whereas the L-NAME plus ibuprofen-resistant muscarinic receptor responses were blocked by the combination of apamin plus charybdotoxin (Wang and Loutzenhiser, 2002a). And yet, a bradykinin-induced EDHF response appears to differ from that to both muscarinic receptors and PAR2 in this renal afferent arteriole preparation. A combination of apamin plus charybdotoxin failed to inhibit vasodilatation by PAR2 activation, but when added in combination with 17-octadecynoic acid, an inhibitor of cytochrome P450s that metabolizes arachidonic acid to epoxyeicosatrienoic acids, inhibited the vasodilatation elicited by bradykinin, but not that by PAR2 activation (Wang and Loutzenhiser, 2002a). Clearly, endothelial G protein-coupled receptors, including muscarinic, PAR2 and bradykinin receptors, within the same tissues are capable of producing distinct EDHFs that differ also from those produced by these same receptors in blood vessels from other species. Multiple EDHFs are responsible for the vasodilator responses of various tissues across species. The EDHFs elicited by activation of murine endothelial PAR2 have a pharmacological profile that suggests the involvement of endothelial-derived K⫹, the activation of inwardly rectifying K⫹ channels and Na⫹-K⫹-ATPases, and both the small- and intermediateconductance Ca2⫹-activated K⫹ channels (McGuire et al., 2002). Contrary to these results, the EDHFs released by PAR2 activation in renal afferent arterioles from hydronephrotic rats are different, since they are not sensitive to the combination of apamin plus charybdotoxin. Whether the EDHF profiles are different because of species or blood vessel differences remains to be determined. An EDHF produced by PAR2 in coronary arteries of rats (McLean et al., 2002) differs from that of PAR2-stimulated EDHF described here for murine small mesenteric arteries. Unlike the results obtained using the coronary preparation of the rat,

Proteinase-activated receptor-2 55 a vanilloid receptor antagonist (capsazepine) and inhibitors of lipoxygenases (nordihydroguarectic acid, baiclein, cinnamyl-2,4-dihydroxy-a-cyanocinnamide and MK886) failed to inhibit the PAR2-mediated EDHF response in murine small mesenteric arteries. Ultimately, the characterization of the membrane potential hyperpolarization that occurs in response to PAR2 activation in murine small mesenteric arteries (unpublished results) may be used for comparison with those putative changes by EDHFs in other preparations, and thus, clarify the possible differences in EDHF-mediated mechanisms. In conclusion, there are complexities with respect to endothelium-derived relaxing factors/EDHF properties both in terms of tissue specificity and in terms of G proteincoupled receptor signaling mechanisms. Thus, (a) different agonists acting via distinct G protein-coupled receptors in the same tissue can cause the release of the same endothelium-derived relaxing factors/EDHF; (b) The same agonist, activating ostensibly the same G protein-coupled receptor in different tissues can cause the release of distinct EDHFs; and (c) distinct agonists acting in the same tissue via different G protein-coupled receptors can cause the release of distinct EDHFs. The loss of endothelium-mediated vasodilatation has been identified as a vascular change that occurs in a number of cardiovascular and metabolic diseases such as diabetes (Pannirselvam et al., 2002). Changes in endothelial NO synthase expression may be associated with this endothelial dysfunction. Therapeutic strategies that utilize constitutive and disease-induced vasodilator (i.e. EDRF/EDHF-mediated) signal transduction mechanisms elicited selectively by G protein-coupled receptors may provide treatments for the vascular complications of many diseases.

7

Mechanical stimulation increases the activity and expression of cytochrome P450 2C in porcine coronary artery endothelial cells B. Fisslthaler, U.R. Michaelis, R. Busse and I. Fleming Epoxyeicosatrienoic acids (EETs), metabolites of arachidonic acid generated by cytochrome P450 enzymes, play a crucial role in the nitric oxide/prostacyclin-independent relaxation in the coronary circulation. Since coronary arteries are subjected to pronounced rhythmic variations in diameter and blood flow, stimuli known to modulate the expression of a variety of genes, the purpose of this study was to determine the effect of cyclic stretch and fluid shear stress on the activity and expression of cytochrome P450 2C in coronary endothelial cells. Acute (5 min) stimulation with cyclic stretch or fluid shear stress, significantly increased the production of 11,12-, and 14,15-EET by cultured porcine coronary endothelial cells. Prolonged exposure (24 h) to cyclic stretch or shear stress increased the expression of cytochrome P450 2C mRNA and protein by five to ten-fold, effects which were accompanied by comparable increases in EET generation. Inhibitors of tyrosine kinases and the p38 mitogenactivated protein kinase, which are activated in endothelial cells by cyclic stretch, had no influence on the stretch-induced increase in cytochrome P450 2C expression. An increase in cytochrome P450 2C mRNA and protein expression was also observed in native endothelial cells, in isolated segments of porcine coronary artery perfused under steady flow or pulsatile conditions for 6 h. These results have identified the coronary cytochrome P450 2C as a novel mechano-sensitive gene product in native and cultured endothelial cells.

1. INTRODUCTION In a variety of vascular beds, metabolites of arachidonic acid synthesised by cytochrome P450 2C enzymes have been shown to play a central role in the initiation of EDHF-mediated responses. The hyperpolarizing metabolites in question are epoxyeicosatrienoic acids (EETs), and their generation is particularly well documented in coronary arteries from dogs, pigs and humans (Campbell et al., 1996; Oltman et al., 1998; Widman et al., 1998; Fisslthaler et al., 1999). Most of the studies performed to characterise EDHF-mediated responses have employed receptor-dependent agonists to increase intracellular Ca2⫹ levels and liberate arachidonic acid from membrane phospholipids, the rate limiting step in epoxide production. Mechanical stimuli are thought to be the physiologically most important signals for the generation of vasodilator autacoids but there are only a few studies in which the generation of EDHF has been assessed following mechanical stimulation. In pig coronary arteries, for example, flowdependent dilatation is independent of nitric oxide and depends solely on the generation of an EDHF-like factor (Dube and Canty, 2001) and pulsatile stretch elicits the release of a cytochrome P450-like EDHF (Popp et al., 1998). In human coronary arteries, flow-induced dilatation is also insensitive to the inhibition of endothelial nitric oxide synthase and cyclooxygenases but markedly attenuated by cytochrome P450 inhibitors (Miura et al., 2001).

Activity and expression of cytochrome P450 2C

57

Although cytochrome P450 2C protein is expressed in isolated coronary arteries which demonstrate a prominent cytochrome P450-dependent EDHF, cytochrome P450 proteins levels decrease rapidly under organ culture and cell culture conditions in which mechanical stimulation no longer takes place. As the activity and expression of numerous enzymes and gene products in endothelial cells are known to be modulated by mechanical stimulation (Chien et al., 1998; Frangos et al., 2001), the aim of the present study was to determine whether or not the cytochrome P450 2C enzymes expressed in coronary endothelial cells are affected by fluid shear stress and cyclic stretch. 2. METHODS

2.1. Cell culture Endothelial cells were isolated from the left coronary artery of freshly slaughtered pigs as described (Popp et al., 1996) and cultured on fibronectin coated tissue culture dishes in M199/MCDB (1:1 containing 14% fetal calf serum, L-glutamine 10 mmol/l, fibroblast growth factor 1 ng/ml; epidermal growth factor 0.1 ng/ml; endothelial cell stimulating growth factor 0.4%; penicillin 50 U/ml and streptomycin 50 ␮g/ml). Confluent cultures of cells from the second or third passage were either exposed to fluid shear stress in a cone plate viscosimeter or to cyclic stretch. For exposure to cyclic stretch the cells were seeded on flexible-bottomed six-well culture plates coated with pronectin (BioFlex, Flexcell International Corporation, McKeesport, PA). When the cells were confluent the plates cells were mounted onto loading plates in a FlexerCell FX-3000 strain unit (Flexcell) and placed in an incubator. Cells were stretched with an average strain of 6% at a rate of 1 Hz and static control experiments were performed on cells on stretch plates not exposed to cyclic strain.

2.2. Cyclic stretch of porcine coronary artery segments Epicardial arteries were excised and cut into two segments of equal length. The segments were cannulated at both ends and side branches of the segment were sealed with surgical clips. Arteries were placed into vessel chambers in which the perfusion pressure was continuously monitored. Blood vessels were pressurized to 40–60 mmHg and perfused (1 ml/min) with minimum essential medium. After equilibration, sinusoidal pressure changes (30–40mmHg, 1Hz), were applied to one of the coronary artery pairs. The diameter changes induced corresponded to a calculated strain of between 6% and 8%.

2.3. RNA isolation and RT-PCR RNA from cultured and native endothelial cells was isolated as described (Chomczynski and Sacchi, 1987). The RNA concentration was determined by measuring the absorption at 260 nm and equal amounts of RNA used for the reverse transcriptase reaction. Random hexanucleotide primers were used for reverse transcription and the oligonucleotides used for PCR were derived from the porcine cytochrome P450 2C34 sequence (genebank accession no.: U35843), upstream primer: AGACAACGAGCACCACTCTG, downstream primer: CTTGGGGATGAGGTAGTTT) which exhibited a high homology to the human cytochrome P450 2C8 sequence (genebank accession no.: Y00498) and elongation factor (EF) upstream primer: GACATCACCAAGGGTGTGCAG, downstream primer: GCGGTCAGCACACTGGCATA.

58

B. Fisslthaler et al.

PCR products were separated on a 1.5% TAE agarose gel and visualised by staining with ethidium bromide. For the verification of the DNA fragment, the PCR-products were transferred to nylon membranes and hybridised with 32P-labelled DNA fragments derived from a plasmid containing the coding sequence of cytochrome P450 2C8.

2.4. Immunohistochemistry Cultured cells or arterial segments were fixed in paraformaldehyde (2%, 1 h). The cells were permeabilised with Triton X-100 (0.2%, v/v) and after blocking with phosphate buffered saline containing 10% horse serum, were incubated with antibodies directed against either cytochrome P450 2C or (␤-actin). The signal from the FITC (cytochrome P450 2C) or Texas-red conjugated secondary antibodies (␤-actin) was detected by confocal microscopy.

2.5. Determination of EET levels For the detection of the intracellular eicosanoids, endothelial cells were cultured and either left statically or subjected to fluid shear stress or cyclic stretch. After the stimulation the cells were collected and frozen in liquid nitrogen. The cells were lysed and cytochrome P450-derived eicosanoids were extracted three times by addition of an equal volume of diethyl ether. Dried samples were dissolved in acetonitrile/methanol (80/20, v/v) and subjected to isocratic reversed-phase high-performance liquid chromatography with online photodiode array detection as described (Kiss et al., 1998). Cytochrome P450-derived metabolites were detected at 204 nm, identified by spectra plot on the peak maximum and co-elution with commercial standards.

2.6. Statistical analysis Data are expressed as mean ⫾ SEM, and statistical evaluation was performed using the Student’s t test for unpaired data. Values of P less than 0.05 were considered statistically significant. 3. RESULTS Cultured porcine coronary artery endothelial cells were subjected to either fluid shear stress (12 dynes/cm2) or cyclic stretch (6%) for 5 min or 24 h. Both stimuli elicited a significant increase in the generation of 11,12- and 14,15-EET release compared to that detected under static conditions (Figure 7.1). Short-term exposure to cyclic strain resulted in a significant increase in arachidonic acid levels (static 30.50 ⫾ 4.96 ng/plate vs strain 45.59 ⫾ 4.10 ng/plate). Similarly, after 24 h stimulation with either cyclic strain or fluid shear stress the intracellular release of arachidonic acid was increased by approximately 50%. To assess whether or not the increase in EET-formation after 24 h was solely due to the increase in substrate availability or also to an increase in the expression of arachidonic acid epoxygenases, RT-PCR as well as immunohistochemistry was performed to evaluate cytochrome P450 2C mRNA and protein levels. Using both methods a significant increase in cytochrome P450 2C expression was detectable in cells exposed for 24 h to either cyclic stretch (6%) or shear stress (12 dynes/cm2, Figure 7.2). mRNA levels of other endothelial cytochrome P450 isoforms, among them other arachidonic acid epoxygenases (cytochrome P450 2B, cytochrome P450 3A and cytochrome P450 2J) were not affected by either fluid shear stress (data not shown) or cyclic strain (Fisslthaler et al., 2001).

5 min 20

15

*

*

24 h Static * Shear Stretch

*

EET (ng)

15 10 10 5 5 0

0 Static

Shear

Stretch

Static

Shear

Stretch

Figure 7.1 Fluid shear stress and cyclic stretch increase the generation of 11,12-EET in porcine coronary artery endothelial cells. EET formation was assessed in cultured coronary artery endothelial cells under control conditions (static) and following exposure to either fluid shear stress (12 dynes/cm2) or cyclic strain (6% mean elongation, 1 Hz) for 5 min (left panel) or 24 h (right panel). The bar graphs summarize the results from four independent experiments, the asterisks indicate statistically significant differences between static and shear or static and stretch (P ⬍ 0.05).

Control

Control

Shear

Stretch CYP 2C

mRNA expression (CYP2C vs EF-2)

EF-2 0.75

4

*

*

3

0.50

2 0.25

1 0 Static

Shear

Control

Stretch

Figure 7.2 Fluid shear stress and cyclic stretch increase CYP 2C mRNA expression in endothelial cells. Total RNA from cultured porcine coronary artery endothelial cells either maintained under static conditions or following exposure to fluid shear stress (12 dynes/cm2, 24 h) or cyclic strain (6% mean elongation, 1 Hz, 24 h) was used for RT-PCR. The upper panel shows the original southern blot (CYP 2C) or ethidium bromide stained gels (elongation factor-2, EF-2) of the PCR products. The bar graphs summarize the densitometrically evaluated data from four independent experiments. Cytochrome P450 2C levels were normalised relative to EF-2. The asterisks indicate statistically significant differences between static and shear or static and stretch (P ⬍ 0.05).

60

B. Fisslthaler et al.

CYP 2C/EF-2 (% control)

1000

750

Basal PP-2 PP-3 SB 220025 SB 203580

500

250

0

Figure 7.3 Inhibitors of p38-MAP kinase and Src kinase increase cytochrome P450 2C mRNA in porcine coronary artery endothelial cells. Porcine coronary artery endothelial cells were treated for 24 h with solvent (Basal), SB220025 (60 nmol/l), SB 203580 (1 ␮mol/1), PP-2 (30 nmol/l) or PP-3 (30 nmol/l). RT-PCR was performed using total RNA for cytochrome P450 2C and EF-2. CYP 2C levels were normalised by comparison with the amounts of elongation factor-2. The data shown were obtained in three to four separate experiments.

The application of a mechanical stimulus to endothelial cells cultured and maintained under static conditions is reported to activate a number of intracellular signal transduction cascades, such as the mitogen-activated protein kinases (ERK1/2, p38 MAP kinase) and the tyrosine kinase c-Src. Specific inhibitors for these pathways were used and the effect on the shear stress or stretch-induced expression increase of cytochrome P450 2C was investigated by RT-PCR. It was however impossible to determine the effects of some specific pathways on the induction of cytochrome P450 expression by mechanical stimuli since most of the inhibitors used (PP-2, PP-3 to target Src and SB220025, SB 203580 to inhibit the p38 MAP kinase), induced the expression of cytochrome P450 2C (Figure 7.3). In freshly excised porcine coronary arteries, immunohistochemistry studies demonstrated the expression of CYP 2C protein exclusively in the endothelial cell layer. Exposure of the arterial segments to cyclic strain (6%, 6 h, 1 Hz) increased the expression of cytochrome P450 2C mRNA and protein in the endothelium five to seven-fold above levels detected in blood vessels exposed to flow at a constant pressure (Figure 7.4). 4. DISCUSSION The mechanical stimulation of cultured and native coronary artery endothelial cells by either fluid shear stress or cyclic stretch increases the activity and expression of cytochrome P450 2C. As reported previously (Fisslthaler et al., 2001), mechanical stimulation did not influence the expression of other cytochrome P450 isozymes expressed in endothelial cells (cytochrome P450 3A and cytochrome P450 2J). As a consequence of the continuous contraction and relaxation of the heart, coronary arteries are constantly exposed to pronounced rhythmic changes in transmural pressure and blood flow. These mechanical forces can elicit the synthesis and/or release of a variety of endothelial factors, including nitric oxide and eicosanoids, which in turn regulate local vascular tone (Busse and Fleming, 1998). In the case of cytochrome P450 2C, cyclic stretch was expected to be a better stimulus for an increase in activity than fluid shear stress because

Activity and expression of cytochrome P450 2C A

61

B Flow + stretch

Flow

Flow

nc

pc CYP 2C

Flow + stretch

CYP 2C/PECAM-1

PECAM-1 1000

* 750 500 250 0 Flow

Flow + stretch

Figure 7.4 Effect of cyclic stretch on the expression of cytochrome P450 2C in native coronary artery endothelial cells. (A) Immunohistochemical staining of cytochrome P450 2C and ␤-actin protein in sections of the same porcine coronary arteries after 6 h incubation under either non-pulsatile conditions (upper picture) or cyclic stretch (6–8% strain, 1 Hz, lower picture). The results are representative of data obtained in two additional experiments. The bar represents 100 ␮m. (B) Paired segments of porcine coronary artery were perfused for 6 h under non-pusaltile (Flow) or pulsatile conditions, (Flow ⫹ stretch). Thereafter, total RNA was extracted for RT-PCR and a positive control (p.c., porcine liver) and a negative control (n.c., no reverse transcriptase in the RT reaction) were included in each PCR experiment. PECAM-1 indicates platelet and endothelial adhesion molecule-1 as a control for equal amounts of endothelial RNA. The bar graph provides a statistical summary of the data obtained in 4 separate experiments normalised relative to the control (flow) group; the asterisk indicates a statistically significant difference between flow and flow ⫹ stretch (P ⬍ 0.05). Source: Reprinted from (Fisslthaler et al., 2001) with permission.

cyclic stretch induces a more sustained increase in endothelial Ca2⫹ compared to laminar shear stress (Naruse and Sokabe, 1993; Helmlinger et al., 1995). Therefore it would be expected to be a better activator of phospholipase A2 and thus liberate more arachidonic acid from membrane phospholipids. The observation that similar amounts of arachidonic acid and EET were generated by both stimuli suggests that mechanisms other than an increase in intracellular Ca2⫹ regulate the activity of cytochrome P450 epoxygenases in endothelial cells. While the acute increase in EET levels elicited by shear stress or cyclic strain can be attributed to an increase in intracellular level of arachidonic acid, the long-term effects appear to be related to a combination of increased substrate availability as well as increased enzyme expression. Relatively little is known about the regulation of cytochrome P450 expression in endothelial cells but the expression of cytochrome P450 enzymes is in general determined by the transcription rate (Honkakoski and Negishi, 2000). In coronary artery endothelial cells a variety of additional stimuli, such as ␤-naphthoflavone, HMG CoA reductase inhibitors, cortisol and the calcium-antagonist nifedipine also enhance EET production and EDHF-mediated responses by increasing cytochrome P450 2C expression rather than the activity of phospholipase A2 (Fisslthaler et al., 1999; Castilla et al., 1999;

62

B. Fisslthaler et al.

Bauersachs et al., 2001; Bauersachs et al., 2002; Fisslthaler et al., 2003). Given that the attachment of cells to the extracellular matrix can influence signalling cascades activated by mechanical stimuli especially those involving integrins, it was necessary to demonstrate that cyclic strain was able to enhance cytochrome P450 expression in native endothelial cells and not only in cultured endothelial cells seeded onto fibronectin-coated culture plates. In native coronary artery endothelial cells, pulsatile stretch significantly enhanced cytochrome P450 2C expression. The signal transduction steps that link mechanical stimulation into an increase in cytochrome P450 2C expression remain to be elucidated as nothing is currently known about the promoter of the isozymes(s) expressed in porcine coronary arteries. Fluid shear stress or cyclic stretch are known to activate numerous signalling cascades in endothelial cells including protein kinase B/Akt, tyrosine kinases and mitogen-activated protein kinases (Chien et al., 1998; Frangos et al., 2001). Whether the activation of these kinases is involved in the increase in cytochrome P450 2C expression in coronary endothelial cells could not be assessed since the pharmacological tools used to interfere with specific signalling pathways, themselves increased cytochrome P450 2C expression. Indeed, the major role for hepatic cytochrome P450 enzymes is the metabolism of xenobiotics and pharmacological compounds, and such substances often act as transcriptional inducers for cytochrome P450 enzymes, a phenomenon referred to as substrate induction. It is likely that such a mechanism may account for the effects of kinase inhibitors on cytochrome P450 2C expression by coronary artery endothelial cells. The data provided in this chapter demonstrate that the activity as well as the expression of cytochrome P450 2C, an enzyme crucially involved in the generation of EDHF in the porcine coronary artery, is increased by fluid shear stress as well as cyclic strain. Although the exact mechanisms involved in mediating this effect remain to be identified, cytochrome P450 2C has been identified as a novel mechano-sensitive gene in coronary endothelial cells. ACKNOWLEDGEMENTS The authors are indebted to Isabel Winter and Stergiani Hauk for expert technical assistance.

8

Important role of hydrogen peroxide as an endothelium-derived hyperpolarizing factor in animals and humans Hiroaki Shimokawa, Tetsuya Matoba, Keiko Morikawa, Toyotaka Yada, Hiroshi Kubota, Yoji Hirakawa and Akira Takeshita The endothelium plays an important role in maintaining vascular homeostasis by synthesizing and releasing several vasodilator factors, including prostacyclin, nitric oxide and the yet unidentified endothelium-derived hyperpolarizing factor (EDHF). Several mechanisms have been proposed to account for EDHF-mediated vascular responses, including cytochrome P450-dependent metabolites, K⫹, and gap junctional electrical communications. Endotheliumderived hydrogen peroxide (H2O2) has been identified as an EDHF in mouse mesenteric arteries. The present study was designed to examine whether or not H2O2 is an EDHF in other blood vessels including human arteries. In human mesenteric arteries, bradykinin elicited endothelium-dependent relaxations and hyperpolarizations in the presence of indomethacin and N␻-nitro-L-arginine, which were attributed to EDHF. The EDHF-mediated responses were inhibited by catalase, an enzyme that decomposes H2O2, whereas it did not affect endotheliumindependent hyperpolarizations to levcromakalim. Exogenous H2O2 elicited relaxations and hyperpolarizations in arteries without endothelium. A gap junction inhibitor 18␣glycyrrhetinic acid partially inhibited, whereas inhibitors of cytochrome P450 did not affect the EDHF-mediated relaxations. These results indicate that H2O2 is a primary EDHF in human mesenteric arteries, and that gap junctions may play a role in those arteries. Catalase also inhibited bradykinin-induced relaxations and hyperpolarizations in the presence of indomethacin and N␻-nitro-L-arginine. Similarly, endothelium-derived H2O2 has also been identified as an EDHF in porcine coronary microvessels. These results suggest that H2O2 may be an EDHF not only in animals but also in humans.

1. INTRODUCTION The endothelium plays an important role in maintaining vascular homeostasis by synthesizing and releasing several vasodilator factors, including prostacyclin, nitric oxide (NO), and yet unidentified endothelium-derived hyperpolarizing factor (EDHF). Since the first reports on the existence of EDHF (Chen et al., 1988; Félétou and Vanhoutte, 1988), several candidates for EDHF have been proposed, including cytochrome P450 metabolites (Rosolowsky and Campbell, 1993; Fisslthaler et al., 1999), endothelium-derived K⫹ (Edwards et al., 1998; Beny and Schaad, 2000). Electrical communications between endothelial cells and smooth muscle cells through gap junctions may also play a role in EDHF-mediated responses (Taylor, et al., 1998; Yamamoto et al., 1999). Endothelium-derived hydrogen peroxide (H2O2) is a primary EDHF in mouse mesenteric arteries (Matoba et al., 2000). Although it is possible that the nature of EDHF varies among different species and vascular beds (Mombouli and Vanhoutte, 1997; Shimokawa, 1999; Shimokawa et al., 1999), H2O2

64

Hiroaki Shimokawa et al.

could be a reasonable candidate of EDHF in general since it is produced from several oxidases in endothelial cells including endothelial NO synthase (Matoba et al., 2000). In human arteries, the existence and importance of EDHF have been demonstrated in various vascular beds both in vivo and in vitro (Urakami-Harasawa et al., 1997; Miura et al., 1999; Tagawa et al., 1999; Honing et al., 2000). However, the nature of EDHF still remains to be identified in human arteries (Urakami-Harasawa et al., 1997). In addition, the nature of EDHF in the coronary vasculature especially in microvessels is also an important issue to be examined. The present study was thus designed to examine whether H2O2 is an EDHF in different vascular beds including human mesenteric arteries and porcine coronary microvessels.

2. METHODS 2.1. Human mesenteric arteries This study was approved by the local human research committee. Since the arteries used in this study were classified as surgical specimens, their use was exempted from requiring each patient’s consent. Human mesenteric arteries (250–300 ␮m in diameter) were obtained from the nets during gastrectomy operations. Mesenteric arteries were removed of adherent tissues and cut into rings for organ chamber experiments and electrophysiological experiments (Urakami-Harasawa et al., 1997; Matoba et al., 2002).

2.2. Porcine coronary arteries Domestic male pigs (Nihon Crea, Tokyo, 2–3 months old, and weighing 25–30 kg) were used. The animals were killed and the right ventricular free walls were carefully removed. Coronary microvessels (250–300 ␮m in diameter) were excised from right ventricular free walls and carefully cleaned of adherent perivascular connective tissue under a microscope. Microvessels were cut into 1.2-mm rings for organ chamber experiments.

2.3. Organ chamber experiments Experiments were performed in 37 ⬚C Krebs solution containing indomethacin (10⫺5 M) and N␻-nitro-L-arginine (L-NNA, 10⫺4 M) bubbled with 95% O2 and 5% CO2. Isometric tension was recorded in isolated arterial rings contracted with prostaglandin F2␣ (3–10 ⫻ 10⫺6 M) or KCl (40–60 mM). The extent of contraction was adjusted to 50–70% of the response to 62 mM KCl (Shimokawa et al., 1996; Urakami-Harasawa et al., 1997). Endotheliumdependent relaxations to BK and endothelium-independent relaxations to H2O2 were analyzed referring a maximal relaxation to sodium nitroprusside as 100% relaxation (Shimokawa et al., 1996; Urakami-Harasawa et al., 1997; Matoba et al., 2000). In some patients, EDHF-mediated responses were impaired, which might be related to underlying risk factors (Urakami-Harasawa et al., 1997). Irrespective of its cause, data were abandoned from patients in whom control relaxation to bradykinin was less than 70% in order to clarify the effect of each inhibitor on the EDHF-mediated relaxations. To examine the nature of EDHF, the following inhibitors were used; catalase (6250 U/ml), sulfaphenazole (10⫺5 M, a specific inhibitor of cytochrome P450 epoxygenase) (Fisslthaler et al., 1999), 17-octadecynoic acid (17-ODYA, 3 ⫻ 10⫺6 M, an inhibitor of cytochrome P450) (Zygmunt et al., 1996; Vanheel and Van de Voorde, 1997), and 18␣-glycyrrhetinic acid (18␣-GA, 10⫺4 M, an inhibitor of gap junctions) (Taylor et al., 1998).

H2O2 as an EDHF in animals and humans 65

2.4. Electrophysiological experiments Rings of human mesenteric arteries were placed in experimental chambers perfused with 37 ⬚C Krebs solution containing indomethacin (10⫺5 M) and L-NNA (10⫺4 M) bubbled with 95% O2 and 5% CO2. A fine glass capillary microelectrode filled with 3 M KCl was impaled into the vascular smooth muscle. Changes in membrane potentials of a vascular smooth muscle in response to bradykinin, levcromakalim (an opener of ATP-sensitive K⫹ channels), or H2O2 were recorded continuously (Shimokawa et al., 1996; Urakami-Harasawa et al., 1997; Matoba et al., 2000).

2.5. Drugs and solution The ionic mM composition of Krebs solution was as follows; Na⫹ 144, K⫹ 5.9, Mg2⫹ 1.2, ⫺ ⫺ Ca2⫹ 2.5, H2PO⫺ 4 1.2, HCO3 24, Cl 129.7, and glucose 5.5. Bradykinin, indomethacin, L-NNA, catalase, 18␣-GA, 17-ODYA and sulfaphenazole were obtained from Sigma Chemical Co. (St Louis, MO). Levcromakalim was a kind gift from SmithKline Beecham Pharmaceuticals. Catalase was directly dissolved in Krebs solution. Indomethacin was dissolved in 10⫺2 M Na2CO3. 18␣-GA, 17-ODYA and sulfaphenazole were dissolved in dimethyl sulfoxide. Levcromakalim was dissolved in 90% ethanol and other drugs were dissolved in distilled water. The solvent used did not affect the mechanical or electrical responses at their final bath concentrations.

2.6. Statistical analysis Data are shown as mean ⫾ SEM. Concentration-response curves were analyzed by two-way ANOVA followed by Scheffé’s post-hoc test for multiple comparisons. Other values were analyzed by paired t-test or one-way ANOVA according to propriety. P values less than 0.05 were considered to be statistically significant.

3. RESULTS AND DISCUSSION 3.1. Existence of EDHF-mediated vascular responses Endothelium-dependent relaxations to bradykinin of human mesenteric arteries were observed in the presence of indomethacin and L-NNA (Figure 8.1). These relaxations could be attributed to endothelium-dependent hyperpolarization of the membrane since they were abolished by removal of the endothelium and markedly inhibited by increased extracellular K⫹ concentrations (data not shown). Indeed, bradykinin also caused endothelium-dependent hyperpolarizations in the presence of indomethacin and L-NNA (Figure 8.1). Thus, EDHFmediated relaxations and hyperpolarizations were observed in human mesenteric arteries, a finding consistent with the previous results (Urakami-Harasawa et al., 1997).

3.2. H2O2 as an EDHF H2O2 can be a candidate of EDHF for the following reasons: (a) vascular endothelial cells have a capacity to produce superoxide anions and H2O2 from several intracellular sources, including eNOS, cyclooxygenase, lipoxygenase, cytochrome P450 enzymes and NAD(P)H oxidases (Katusic, 1996; Matoba et al., 2000) and (b) H2O2 is known to activate KCa channels, the targets of EDHF and to cause hyperpolarizations in vascular smooth muscle cells (Barlow and White, 1998; Hayabuchi et al., 1998a). To examine the role of H2O2

66

Hiroaki Shimokawa et al. Control

Relaxations (%)

A

B

50

* n =5 100

Hyperpolarization (∆ mV)

BK 0

9

8 7 BK (–log M)

Catalase LK 10–5 M

M

0

0

–10

–10

–20

–20

–30 10

10–7

†

n = 3 –30

n =3

6

Figure 8.1 Inhibitory effects of catalase on EDHF-mediated relaxations and hyperpolarizations of human mesenteric arteries. (A) Bradykinin (BK) concentration-dependently elicited relaxations in the presence of indomethacin (10⫺5 M) plus L-NNA (10⫺4 M). The relaxations were inhibited by catalase (6250 U/mL). (B) Bradykinin (10⫺7 M) and levcromakalim (LK, 10⫺5 M) caused hyperpolarizations of human mesenteric arteries. Catalase significantly inhibited EDHF-mediated hyperpolarizations, whereas it did not affect levcromakaliminduced endothelium-independent hyperpolarizations. The asterisk and cross indicate statistically significant difference (P ⬍ 0.05) by ANOVA and by paired t-test, respectively. Source: Reproduced with permission from Matoba et al., 2002.

as an EDHF, catalase was used that selectively decomposes H2O2 into oxygen and water (Matoba et al., 2000). Treatment with catalase (6250 U/mL) did not significantly affect resting tension (data not shown) or resting membrane potentials (data not shown). Importantly, treatment with catalase markedly inhibited the bradykinin-induced EDHF-mediated relaxations and hyperpolarizations of human mesenteric arteries (Figure 8.1). By contrast, catalase did not affect endothelium-independent hyperpolarizations to levcromakalim, a direct K⫹ channel opener. These results suggest that catalase inhibits EDHF-mediated responses without affecting vascular smooth muscle and that H2O2 plays a crucial role in the EDHF-mediated responses. This was also the case in porcine coronary microvessels (Figure 8.2). Subsequently, the effects of exogenous H2O2 was examined on the force and membrane potential of mesenteric arteries without endothelium. Exogenous H2O2 (10⫺8–10⫺4 M) elicited concentration-dependent relaxations of human mesenteric arteries, which were inhibited by increased extracellular K⫹ (Figure 8.3). H2O2 (10⫺5–10⫺4 M) also elicited hyperpolarizations (Figure 8.3). These results indicate that exogenous H2O2 elicits direct relaxations through membrane hyperpolarization of smooth muscle cells. It is conceivable that H2O2 activates KCa channels on the smooth muscle to cause hyperpolarization (Barlow and White, 1998; Hayabuchi et al., 1998a; Barlow et al., 2000; Matoba et al., 2000), although the mechanism of H2O2-induced hyperpolarization remains to be examined. Vascular endothelial cells are capable of producing reactive oxygen species including H2O2 although it remains to be examined whether the amount of H2O2 produced from the endothelium is sufficient to elicit responses of the vascular smooth muscle (Cosentino et al., 1998; Matoba et al., 2000). The present findings suggest that H2O2 is a primary EDHF in human mesenteric arteries (Matoba et al., 2002). Subjects studied in this study had some risk factors for atherosclerosis, which are associated with increased production of reactive oxygen species in blood vessels (Somers and Harrison, 1999; Katusic, 2001). Although the importance of H2O2

H2O2 as an EDHF in animals and humans 67 A Membrane potential (mV) Membrane potential (mV)

–40

Membrane potential (mV)

Indomethacin + L-NNA –40

–40

B

BK

LK

–60

–80

Indo + L-NNA + Catalase

C

BK

LK

–60

–80 Indo + L-NNA + Catalase + Aminotriazole BK

LK

–60

–80 5 min

Figure 8.2 Bradykinin (BK, 10⫺7 M) and levcromakalim (LK, 10⫺5 M) elicited hyperpolarizations of porcine coronary microvessels (A). In the presence of catalase (6250 U/mL), bradykinin-induced hyperpolarizations were abolished, whereas levcromakaliminduced hyperpolarizations were unaffected (B). When inactivated by aminotriazole (50 mM), catalase lost its inhibitory effect on bradykinin-induced hyperpolarizations (C). Experiments were performed in the presence of indomethacin (10⫺5 M) and L-NNA (10⫺4 M).

as a signaling molecule might be enhanced in the present study because of the underlying risk factors, that endothelium-derived H2O2 acts as a primary EDHF has been demonstrated even under physiological conditions (Matoba et al., 2000). Furthermore, this hypothesis explains well the redundancy of NO and EDHF, as well as the compensatory interactions between the two factors under the above-mentioned disease states (Shimokawa, 1999).

3.3. Role of cytochrome P450 enzymes in EDHF-mediated responses Cytochrome P450 enzymes may play a role in EDHF-mediated vascular responses of human arteries to judge from the effects of an inhibitor of cytochrome P450, miconazole (Coats et al., 2001; Halcox et al., 2001). The major products of cytochrome P450 that are capable of hyperpolarizing smooth muscle cells are epoxyeicosatrienoic acids (EETs), derivatives of arachidonic acids formed by cytochrome P450 epoxygenase (Rosolowsky

Hiroaki Shimokawa et al.

Control

A

KCl

B

Relaxations (%)

0

50

* n = 5–6

Hyperpolarization (∆ mV)

68

0

–5

–10

† –15

100

8

7 6 5 H2O2 (–log M)

H2O2 (–log M) 5 4.5 4

n = 3–4

4

Figure 8.3 Effects of exogenous H2O2 in human mesenteric arteries. (A) Exogenously applied H2O2 elicited relaxations of endothelium-stripped arteries. H2O2 caused concentration-dependent relaxations that were inhibited by increasing extracellular K⫹ concentrations by KCl (40–60 mM). (B) Exogenous H2O2 caused concentration-dependent membrane hyperpolarizations of vascular smooth muscle. All experiments were performed in the presence of indomethacin (10⫺5 M) plus L-NNA (10⫺4 M). The asterisk and cross indicate statistically significant difference (P ⬍ 0.05) by ANOVA and by paired t-test, respectively. Source: Reproduced with permission from Matoba et al., 2002.

and Campbell, 1993; Fisslthaler et al., 1999). Cytochrome P450 epoxygenase is present in human endothelial cells (Node et al., 1999). Thus, the role of cytochrome P450 enzymes was examined in EDHF-mediated responses using sulfaphenazole, a specific inhibitor of cytochrome P450 epoxygenase (Fisslthaler et al., 1999), and 17-ODYA, a nonspecific inhibitor of cytochrome P450 enzymes (Zygmunt et al., 1996; Vanheel and Van de Voorde, 1997). Importantly, neither inhibitor affected the EDHF-mediated relaxations of human mesenteric arteries (Figure 8.4), suggesting that the cytochrome P450 metabolism may not account for the EDHF-mediated vascular responses. Since previous studies have suggested the involvement of cytochrome P450 in EDHF-mediated vasodilatation in humans (Coats et al., 2001; Halcox et al., 2001), this discrepancy might be due to the difference in vascular beds studied and/or experimental methods used including the pharmacological agents tested.

3.4. Gap junctions and EDHF-mediated relaxations Gap junction-mediated intercellular communications between endothelial and vascular smooth muscle cells may account in part for EDHF-mediated responses (Taylor et al., 1998; Yamamoto et al., 1999). To examine the role of gap junctions in the EDHF-mediated responses, 18␣-GA was used that disrupts gap junction structure and effectively inhibits intercellular electrical communications (Taylor et al., 1998). 18␣-GA partially inhibited EDHF-mediated relaxations of human mesenteric arteries, suggesting that gap junctions may play a role in the response (Figure 8.4). There are at least three hypotheses regarding the involvement of gap junctions in the EDHF-mediated responses. First, hyperpolarizations of the endothelium may simply spread to the underlying smooth muscle through gap junctions (Yamamoto et al., 1999). Second, another substance may move from the endothelium to the smooth muscle through gap junctions as an EDHF (Hutcheson et al., 1999). Third, a diffusible EDHF may elicit hyperpolarizations in the smooth muscle at subendothelial layer, where hyperpolarization may subsequently spread through gap junctions within smooth muscle layers (Edwards et al., 2000). This issue remains to be clarified in future studies.

H2O2 as an EDHF in animals and humans 69

Relaxations (%)

A

B

0

0

50

50

n =4

*

n =5

100

100 10

9 8 7 BK (–log M)

6

Control 17-ODYA

10

9 8 7 BK (–log M)

6

18α-GA Sulfaphenazole

Figure 8.4 Effect of cytochrome P450 inhibitors and a gap junction inhibitor on the EDHF-mediated relaxations in human mesenteric arteries. (A) Neither sulfaphenazole nor 17-ODYA significantly affected the EDHF-mediated relaxations of human mesenteric arteries. (B) 18␣-glycyrrhetinic acid (18␣-GA, 10⫺4 M) partially but significantly inhibited the EDHF-mediated relaxations. Experiments were performed in the presence of indomethacin (10⫺5 M) plus L-NNA (10⫺4 M). The asterisk indicates statistically significant difference (P ⬍ 0.05) by ANOVA. Source: Reproduced with permission from Matoba et al., 2002.

3.5. Importance of EDHF in physiological regulation of vascular tone Regarding the vascular tone in vivo, mechanical stimuli such as shear stress, pulsatile stretch and changes in perfusion pressure are important. Shear stress causes the release of NO by calcium-dependent and -independent mechanism (Fleming et al., 1997). Shear stress has also been shown to cause endothelium-dependent relaxations that are resistant to indomethacin and L-NNA (presumably mediated by EDHF) in rat mesenteric arterioles (Takamura et al., 1999). Pulsatile stretch is also an important stimulus to vascular endothelial cells in vivo to cause a release of EDHF in isolated porcine coronary arteries (Popp et al., 1998). The exact nature of EDHF released in response to those mechanical stimuli remains to be identified. The coronary autoregulatory vasodilator response to decreases in coronary perfusion pressure is inhibited by catalase or a potassium channel inhibitor in the working canine heart, suggesting that endogenous H2O2 plays an important role in coronary autoregulation and thus in the physiological modulation of vascular tone in vivo (Yada et al., 2003). ACKNOWLEDGMENTS We thank Dr Saku, Dr Ikejiri and Ms Tanaka (The Department of Surgery, National Kyushu Medical Center Hospital, Fukuoka, Japan) for cooperation in the human study. This work was supported in part by the grants from the Japanese Ministry of Education, Science, Sports, Culture and Technology, Tokyo, Japan (10357006, 12032215, 12470158, 12877114, 13307024, 13557068, 13670724) and the Mitsui Life Social Welfare Foundation, Tokyo, Japan.

9

Altered calcium dynamics do not account for the attenuation of EDHF-mediated dilatations in the middle cerebral artery of female rats Elke M. Golding, Dorota M. Ferens, Sean P. Marrelli and Robert M. Bryan Jr The authors tested the hypothesis that the contribution of endothelium-derived hyperpolarizing factor (EDHF) to ATP-mediated dilatations is significantly attenuated in the middle cerebral artery of intact and estrogen-treated ovariectomized female rats compared to males and vehicletreated ovariectomized females. Since an increase in endothelial calcium is a critical prerequisite in the EDHF-mediated response, the authors also tested the hypothesis that endothelial cell intracellular calcium [Ca2⫹]i fails to reach sufficient levels in order to elicit robust EDHFmediated dilatations in females and that this effect is mediated by estrogen. Vascular diameter and [Ca2⫹]i were measured concomitantly in perfused middle cerebral artery segments using videomicroscopy and fura 2 fluorescence, respectively. In the presence of N␻-nitro-L-arginine (L-NAME) and indomethacin, the dilatation to luminal ATP was reduced in intact females and estrogen-treated ovariectomized females compared to intact males and vehicle-treated ovariectomized females. Contrary to the initial hypothesis, endothelial cell [Ca2⫹]i increased to comparable levels in intact females, estrogen-treated ovariectomized females, intact males and vehicle-treated ovariectomized females. In response to luminal ATP, the [Ca2⫹]i in the smooth muscle cells decreased to a greater degree in males compared to females and to a greater degree in vehicle-treated ovariectomized females compared to estrogentreated ovariectomized females. The data suggest that loss of a factor coupling EDHF to reduction of smooth muscle cell [Ca2⫹]i accounts for the attenuated EDHF-mediated dilatations in the female middle cerebral artery.

1. INTRODUCTION Gender specific differences in vascular reactivity have been described both in the periphery (Li and Duckles, 1994) and in the cerebral circulation (Geary et al., 1998). For instance, nitric oxide (Skarsgard et al., 1997; McNeill et al., 1999) and prostacyclin (Bolego et al., 1997) are upregulated in females compared to their male counterparts. In peripheral blood vessels, EDHF-mediated dilatations are enhanced in females (McCulloch and Randall, 1998; White et al., 2000) and estrogen-treated ovariectomized females (Liu et al., 2001). The identity of EDHF in the cerebral circulation appears to be different from that in the periphery (Dong et al., 2000). While non-sequestered metabolites of cytochrome P450 mediate the EDHF response in some peripheral vessels (Fisslthaler et al., 1999; Campbell et al., 2001), this does not appear to be the case in cerebral vessels (Bryan et al., unpublished observations). Second, unlike in the periphery (Bauersachs et al., 1996), EDHF does not appear to be a “redundant” mechanism for nitric oxide in the cerebral vasculature since it persists in the presence of a nitric oxide donor (Schildmeyer and Bryan, 2002).

Attenuation of EDHF-mediated dilatations

71

The primary stimulus for the production and/or release of EDHF is a rise in intracellular calcium [Ca2⫹]i in the endothelial cell (Chen and Suzuki, 1990; Fukao et al., 1995). Moreover, the magnitude of the [Ca2⫹]i increase appears to distinguish a nitric oxide-mediated dilatation from an EDHF-mediated response, with the latter requiring a greater increase in [Ca2⫹]i (Marrelli, 2001). Given that EDHF is different in the cerebral vasculature than in the periphery, the authors determined the contribution of EDHF to endothelial stimulation in males and females, and the role of intracellular calcium in mediating this effect.

2. MATERIALS AND METHODS Experiments were carried out in strict accordance with NIH guidelines for the care and use of laboratory animals and were approved by the Animal Protocol Review Committee at Baylor College of Medicine. Rats were housed under a 12 h light/12 h dark cycle with unrestricted access to food and water. Experiments were performed on age-matched (70–90 days old) male (275–324 g) and female (200–224 g) Long–Evans rats. Four groups of rats were used: (a) intact males (n ⫽ 14); (b) intact females (n ⫽ 13); (c) vehicle-treated ovariectomized females (n ⫽ 14); and (d) estrogen-treated ovariectomized females (n ⫽ 18).

2.1. Estrogen depletion and repletion All surgical procedures were undertaken under aseptic conditions. Animals were secured to a nose cone and allowed to spontaneously breathe (2% isoflurane). Rectal temperature was maintained at 37 ⬚C using a heating pad and a temperature controller (Harvard Apparatus Inc, MS). A bilateral ovariectomy was performed by accessing the ovaries through two lateral abdominal incisions. A dorsal incision was made and an osmotic minipump (Model 2004, Alza Corporation, Palo Alto, CA) was implanted subcutaneously to deliver either 17␤-estradiol or 50%DMSO/0.045%NaCl (vehicle). 17␤-estradiol was administered at a physiological dose of 4 ␮g/kg/day. Validation of estrogen depletion and repletion comprised of monitoring body weight and measuring plasma estradiol concentration. For the latter, trunk blood (3 ml) was obtained from all animals immediately following decapitation. The blood was centrifuged for 3 min at 8,000 rpm and the resulting plasma was frozen at ⫺20 ⬚C. 17␤-estradiol was measured at a later time using an ultra-sensitive radioimmunoassay (Diagnostic Systems Laboratory, Webster, TX). Following completion of surgery, wounds were sutured and the animals returned to the holding facility for two weeks.

2.2. Harvesting and mounting cerebral vessels Animals were placed in an anesthetic chamber, allowed to spontaneously breathe isoflurane and then decapitated. The brain was removed from the cranium and placed in ice-cold physiological salt solution. The middle cerebral artery was excised, cleaned of surrounding connective tissue and cannulated with micropipettes in a custom-made vessel chamber. Physiological salt solution was circulated abluminally and perfused luminally. Monitoring of intraluminal pressure was performed with in-line transducers, which were connected to two strain gage panel meters (Omega, Stamford, CT). The vessel chamber was mounted on the stage of an inverted fluorescence microscope. Transmural pressure was set at 85 mmHg with a flow of 100 ␮l/min through the lumen, and the vessels allowed to equilibrate for one hour. During this time they developed spontaneous tone by constricting from their fully dilated diameter at initial pressurization. After the development of tone, the experiment was initiated.

72

Elke M. Golding et al.

2.3. Measurement of vascular diameter and calcium Vascular diameter and intracellular calcium were simultaneously measured using a charge coupled device video system and photomultiplier tube, respectively (Intracellular Imaging, Cincinnati, OH) (Marrelli, 2000). Frequency of acquisition was 8 Hz for vascular diameter and 4 Hz and 5 Hz for [Ca2⫹]i in endothelial and smooth muscle cells, respectively. For the measurement of [Ca2⫹]i the Ca2⫹-sensitive indicator fura 2 was selectively loaded into either the endothelial cells or smooth muscle cells, respectively. Fura 2 was excited at 340 nm and 380nm, and the emission light was sampled at 510nm. The fluorescence ratio (340nm/380nm) was calculated after subtraction of the background fluorescence. Selective loading of the endothelium was achieved by perfusing fura 2-AM (0.67⫻10⫺6 M) through the lumen for a period of 4 min. The fura 2-AM containing buffer was subsequently washed out and an additional period of 15min was allowed for complete deesterification of the dye. Loading of smooth muscle was achieved from the adventitial side by replacing the abluminal physiological salt solution with MOPS buffer containing fura 2-AM (10⫺6 M). Loading was continued for 5 min at room temperature followed by a washout period with fresh physiological salt solution at 37 ⬚C. The blood vessel was allowed 30 min for complete de-esterification of fura 2-AM before [Ca2⫹]i measurements were acquired. Previous studies have confirmed the selectivity of loading the endothelial cells and smooth muscle cells using this paradigm (Marrelli, 2000).

2.4. EDHF-mediated dilatations EDHF-mediated dilatations were assessed by luminal application of ATP, a P2Y2 purinoceptor agonist, in the presence of L-NAME and indomethacin (You et al., 1999b). Following the development of spontaneous tone, L-NAME (3 ⫻ 10⫺5 M) and indomethacin (10⫺5 M) were added to the luminal and abluminal baths in order to remove the nitric oxide synthase and cyclooxygenase contributions, respectively. A concentration response curve to luminal application of ATP (10⫺7–10⫺5 M) was determined in the four groups. Vascular diameter and either endothelial or smooth muscle cell [Ca2⫹]i were assessed in parallel. In order to assess whether or not the smooth muscle female middle cerebral arteries hyperpolarize, 15 mM KCl was added to the abluminal bath (McPherson and Keily, 1995). Experiments were terminated by replacing the physiological salt solution with calcium-free physiological salt solution containing 10⫺3 M EGTA in order to obtain maximal dilatation.

2.5. Reagents and buffers All chemicals were purchased from Sigma (St Louis, MO, USA) with the exception of fura 2-AM and Pluronic F-127 (TefLabs; Austin, TX). The physiological salt solution contained the following (mM): NaCl 119, NaHCO3 21, KCl 4.7, KH2PO4 1.18, MgSO4 1.17, CaCl2 1.6, glucose 5.5 and EDTA 0.026. MOPS buffer consisted of (mM): NaCl 145, NaH2PO4 1.2, KCl 4.7, MgSO4 1.17, CaCl2 1.6, glucose 5, pyruvate 2, EDTA 0.02 and MOPS 3. The MOPS buffer was adjusted to pH 7.4 at room temperature. Stock solutions of ATP (10⫺2 M) and L-NAME (3 ⫻ 10⫺2 M) were prepared in distilled water, aliquotted and then frozen. A stock solution of indomethacin (10⫺2 M) was prepared in a solution of Na2CO3 and distilled water. Fura 2-AM was mixed with 50 ␮l of DMSO and 25 ␮l of Pluronic F-127 in dimethyl sulphoxide (DMSO).

Attenuation of EDHF-mediated dilatations

73

2.6. Data analysis and calculations All data are presented as mean ⫾ SEM. Both diameter and [Ca2⫹]i measurements were averaged over a two-minute period immediately following luminal exposure to ATP. Changes in vascular diameter are presented as a percentage of the maximal diameter of the middle cerebral arteries, as calculated according to Equation (1): % Maximal Diameter ⫽

(DATP ⫺ Dbase) ⫻ 100, (Dmax ⫺ Dbase)

(1)

where DATP is the diameter of the middle cerebral artery after luminal administration of ATP, Dbase is the baseline diameter of the middle cerebral artery before addition of ATP and Dmax is the maximal diameter of the middle cerebral artery in the presence of calcium-free physiological salt solution. Endothelial cell and smooth muscle cell [Ca2⫹]i were determined according to Equation (2): [Ca2⫹]i ⫽ ␤·Kd

(R ⫺ Rmin) , (Rmax ⫺ R)

(2)

where [Ca2⫹]i is the calcium concentration in the endothelial or smooth muscle cells, ␤ is the ratio of 380 unbound/380 bound, Kd is the dissociation constant for fura 2 to Ca2⫹, R is the ratio of the 340/380 emission, Rmin is the ratio in calcium-free conditions, and Rmax is the ratio in calcium-saturating conditions. In situ calibration (Marrelli, 2000) was performed on both intact male (n⫽8) and intact female middle cerebral arteries (n⫽5). The calibration curves did not differ statistically and were therefore averaged to yield the following values: 2.017 (Rmax), 0.1365 (Rmin) and 5.218 (␤). The in situ Kd was 282nM (Knot and Nelson, 1998). Statistical comparisons of body weight changes and plasma estradiol concentration were performed using a one-way analysis of variance followed by a Bonferroni t-test for multiple comparisons. Statistical comparisons of the concentration-response curves to ATP were performed using a two-way analysis of variance with repeated measures and multiple comparisons were made using a Student–Newman–Keuls test. Comparisons of baseline [Ca2⫹]i were made using a one-way analysis of variance followed by a Bonferroni t-test. Differences were considered statistically significant at error probabilities less than 0.05 (P⬍0.05).

3. RESULTS 3.1. Ovariectomized rats At two weeks following ovariectomy and pump insertion, plasma estradiol levels were significantly elevated in estrogen-treated ovariectomized females (36 ⫾ 7 pg/ml) compared to vehicle-treated ovariectomized females (11 ⫾ 2 pg/ml; P ⬍ 0.05; one-way analysis of variance). Estrogen-treated ovariectomized females had also gained significantly less weight compared to vehicle-treated ovariectomized females (1.9 ⫾ 5% vs 47 ⫾ 3%; P ⬍ 0.05; one-way analysis of variance).

3.2. EDHF-mediated dilatations and endothelial changes in calcium Following the development of tone, resting diameters were similar between groups: 241 ⫾ 9␮m (intact males), 230⫾3␮m (intact females), 231⫾15␮m (vehicle-treated ovariectomized

74

Elke M. Golding et al. A Diameter change (%)

100 80

Male Female

60 40 20 0

B

600

[Ca2+]i (nM)

500 400 300 200 –5

100 0

* P < 0.05 compared to males at 10 M ATP # P < 0.05 compared to males at rest Rest

–7 –6 ATP (log M)

–5

Figure 9.1 Bar graphs summarizing (A) the percent diameter change and (B) endothelial [Ca2⫹]i of male (n⫽6) and female (n⫽6) middle cerebral arteries in response to luminal delivery of ATP (3⫻10⫺5 M L-NAME and 10⫺5 M indomethacin present). The EDHFmediated dilatation was significantly reduced in female compared to male middle cerebral arteries. However the concomitant increase in endothelial cell [Ca2⫹]i was comparable in both groups.

females) and 212 ⫾ 3 ␮m (estrogen-treated ovariectomized females). Exposure to L-NAME and indomethacin resulted in similar constrictions in the four groups [17 ⫾ 3% (intact males), 25⫾3% (intact females), 24⫾2% (vehicle-treated ovariectomized females) and 19⫾ 3% (estrogen-treated ovariectomized females)]. Changes in diameter and endothelial cell [Ca2⫹]i in response to luminal delivery of ATP were assessed in arteries isolated from intact males, intact females (Figure 9.1), vehicletreated and estrogen-treated ovariectomized females (Figure 9.2). ATP-induced dilatations were significantly reduced in intact females and estrogen-treated ovariectomized females compared to intact males and vehicle-treated ovariectomized females (P ⬍ 0.05, repeated measures analysis of variance). Resting endothelial [Ca2⫹]i was significantly greater in intact females (174 ⫾ 13 nM) compared to intact males (126 ⫾ 10 nM), and in estrogentreated ovariectomized females (113 ⫾ 15 nM) compared to vehicle-treated ovariectomized females (62⫾5nM) (P⬍0.05; one-way analysis of variance). However, in response to 10⫺5 M ATP, endothelial cell [Ca2⫹]i increased to similar levels in intact males (421 ⫾ 77 nM), intact females (461 ⫾ 116 nM), vehicle-treated ovariectomized females (530 ⫾ 92 nM), and estrogen-treated ovariectomized females (417 ⫾ 50 nM).

3.3. Smooth muscle changes in calcium In the presence of L-NAME and indomethacin, the resting [Ca2⫹]i in smooth muscle cells was comparable in intact males (264 ⫾ 6 nM), intact females (237 ⫾ 12 nM), vehicle-treated

Attenuation of EDHF-mediated dilatations A Diameter change (%)

100 80

75

OVX / Vehicle OVX/ E2

60 40 20 0 700

B [Ca2+]i (nM)

600 500 400 300 200 –5

* P < 0.05 compared to OVX/Vehicle at 10 M ATP # P < 0.05 compared to OVX/Vehicle at rest

100 0

Rest

–7 –6 ATP (log M)

–5

Figure 9.2 Bar graphs summarizing (A) the percent diameter change and (B) endothelial cell [Ca2⫹]i of vehicle-treated ovariectomized females (n ⫽ 7) and estrogen-treated ovariectomized females (n ⫽ 9) middle cerebral arteries in response to luminal delivery of ATP (3 ⫻ 10⫺5 M L-NAME and 10⫺5 M indomethacin present). The EDHF-mediated dilatation was significantly reduced in estrogen-treated ovariectomized female compared to vehicle-treated ovariectomized female middle cerebral arteries. However the concomitant increase in endothelial [Ca2⫹]i was comparable in both groups.

ovariectomized females (221 ⫾ 10 nM) and estrogen-treated ovariectomized females (237 ⫾ 11 nM) (Figures 9.3 and 9.4). In response to luminal application of 10⫺5 M ATP, [Ca2⫹]i decreased in the smooth muscle cells of both intact males (170 ⫾ 9 nM) and vehicle-treated ovariectomized females (176 ⫾ 17 nM), while increasing in intact females (244 ⫾ 15 nM) and estrogen-treated ovariectomized females (264 ⫾ 17 nM). The increase in [Ca2⫹]i reflected the fact that these blood vessels dilated transiently, followed by a constriction. The authors calculated the minimal value that the [Ca2⫹]i reached at 10⫺5 M ATP and verified that the [Ca2⫹]i decreased to a greater degree in males (37 ⫾ 4%) compared to females (21 ⫾ 5%) and vehicle-treated ovariectomized females (18 ⫾ 7%) compared to estrogen-treated ovariectomized females (3 ⫾ 5%) (P ⬍ 0.05, one-way analysis of variance). Hyperpolarization of smooth muscle cells induced by 15 mM KCl caused a comparable dilatation in intact males (71⫾9%) and intact females (64⫾4%) with a similar corresponding decrease in [Ca2⫹]i in the smooth muscle cells [22 ⫾ 6% (intact males) and 24 ⫾ 3% (intact females)]. Comparable dilatations were also observed in estrogen-treated females (69 ⫾ 5%) and vehicle-treated ovariectomized females (73 ⫾ 6%). The concomitant decrease in [Ca2⫹]i of the smooth muscle cells was also comparable between estrogen-treated ovariectomized females (29 ⫾ 2%) and vehicle-treated ovariectomized females (30 ⫾ 4%).

A Diameter change (%)

100 80

Male Female

60 40 20 0

B

300

[Ca2+]i (nM)

250 200 150 100 50 0

–5 * P < 0.05 compared to males at 10 M ATP

Rest

–7 –6 ATP (log M)

–5

Figure 9.3 Bar graphs summarizing (A) the percent diameter change and (B) smooth muscle cell [Ca2⫹]i of male (n ⫽ 8) and female (n ⫽ 7) middle cerebral arteries in response to luminal delivery of ATP (3 ⫻ 10⫺5 M L-NAME and 10⫺5 M indomethacin present). The EDHF-mediated dilatation was significantly reduced in female compared to male middle cerebral arteries. The [Ca2⫹]i in smooth muscle cells decreased to a greater degree in intact male middle cerebral arteries compared to intact females, ruling out the possibility of an alteration of the Ca2⫹ sensitivity of the vascular smooth muscle contractile apparatus in females. A Diameter change (%)

100 80

OVX / Vehicle OVX / E2

60 40 20 0

B

300

[Ca2+]i (nM)

250 200 150 100 50 0

–5 * P < 0.05 compared to males at 10 M ATP

Rest

–7 –6 ATP (log M)

–5

Figure 9.4 Bar graphs summarizing (A) the percent diameter change and (B) smooth muscle cell [Ca2⫹]i of vehicle-treated ovariectomized female (n ⫽ 7) and estrogen-treated ovariectomized female (n ⫽ 9) middle cerebral arteries in response to luminal delivery of ATP (3 ⫻ 10⫺5 M L-NAME and 10⫺5 M indomethacin present). The EDHF-mediated dilatation was significantly reduced in estrogen-treated ovariectomized female compared to vehicle-treated ovariectomized female middle cerebral arteries. The [Ca2⫹]i in the smooth muscle cells decreased to a greater degree in vehicle-treated ovariectomized female compared to estrogen-treated ovariectomized female middle cerebral arteries, ruling out the possibility of an alteration of the Ca2⫹ sensitivity of the vascular smooth muscle contractile apparatus by estrogen.

Attenuation of EDHF-mediated dilatations

77

4. DISCUSSION The results of the present study suggest that EDHF-mediated dilatations are attenuated in the rat middle cerebral artery isolated from intact females and estrogen-treated ovariectomized females. Furthermore, a factor or mechanism coupling EDHF to a reduction of [Ca2⫹]i in the smooth muscle cells appears to account for the attenuated EDHF-mediated dilatations in the female middle cerebral artery. This conclusion is supported by the following findings. First, endothelial cell [Ca2⫹]i reached sufficient levels to potentially elicit an EDHF-mediated response in all experimental groups. However, this was not accompanied by a robust dilatation in intact females and estrogen-treated ovariectomized females. Second, the [Ca2⫹]i in the smooth muscle cells decreased to a greater degree in intact males and vehicletreated ovariectomized females compared to intact females and estrogen-treated ovariectomized females. In the presence of L-NAME and indomethacin, resting but not ATP-stimulated endothelial [Ca2⫹]i was elevated in middle cerebral arteries isolated from intact females and estrogentreated ovariectomized females compared to intact males and vehicle-treated ovariectomized females. These findings agree with the observation in endothelial cells of coronary arteries where the basal but not the acetylcholine-stimulated [Ca2⫹]i was significantly elevated in females compared with their male counterparts (Knot et al., 1999). The finding that basal endothelial [Ca2⫹]i was elevated in the middle cerebral artery could suggest a greater driving force for calcium, perhaps instigated by greater hyperpolarization (Luckhoff and Busse, 1990). The notion that female endothelial cells are more hyperpolarized may offer an explanation for the attenuated EDHF-mediated dilatations since the amplitude of the ATP-induced hyperpolarization depends on the resting membrane potential (Luckhoff and Busse, 1990). The resting [Ca2⫹]i in smooth muscle cells was comparable in middle cerebral arteries isolated from intact males, intact females, vehicle-treated ovariectomized and estrogentreated ovariectomized females. However, in response to luminal ATP, [Ca2⫹]i decreased to a greater degree in smooth muscle cells of males compared to females and vehicle-treated ovariectomized females compared to estrogen-treated ovariectomized females. Since dilatation to EDHF in the male middle cerebral artery is mediated by hyperpolarization of the smooth muscle cells (Chen et al., 1988; You et al., 1999a), the attenuation of the EDHF-mediated dilatation in females could possibly be attributed to the inability of the smooth muscle to hyperpolarize. However, exposure to 15 mM KCl elicited decreases in [Ca2⫹]i and a concomitant dilatation, suggesting that in females the smooth muscle cells can hyperpolarize. The present evidence, albeit indirect, suggests that the hyperpolarization of the smooth muscle is inadequate in the EDHF-mediated pathway in intact female and estrogen-treated ovariectomized female middle cerebral arteries. In conclusion, the present study suggests that a factor or mechanism coupling EDHF to a reduction of ionized [Ca2⫹]i in the smooth muscle cells accounts for the attenuated EDHFmediated dilatations in the female middle cerebral artery. The results support the idea that there is inadequate hyperpolarization of the smooth muscle in the female middle cerebral artery.

10 Connexin-mimetic peptides: influence on nitric oxide synthaseand cyclooxygenase-independent renal vasodilatation, basal renal blood flow and blood pressure in the rat An S. De Vriese, Johan Van de Voorde and Norbert H. Lameire Research on the physiological role of EDHF is hampered by the lack of specific inhibitors that are suitable for in vivo use. In vitro studies support a role for gap junctions in EDHF-mediated signal transmission. This study examines the contribution of gap junctional communication to the EDHF-mediated responses in the renal microcirculation of the rat in vivo and addresses the physiological role of EDHF. The effects of intrarenal administration of connexin-mimetic peptides on the L-NAME- and indomethacin-resistant renal blood flow response to acetylcholine on basal renal blood flow and on systemic blood pressure were examined. 43Gap 27, a peptide homologous to the second extracellular loop of connexin 43, partially inhibited the L-NAME- and indomethacin-resistant renal blood flow response to acetylcholine, whereas 40 Gap 27, homologous to the second extracellular loop of connexin 40, abolished the response. A control peptide, with a replacement of two amino acids in the motif SRPTEK present in the second extracellular loop of connexins 40 and 43, was without effect. None of the peptides affected the response to detaNONOate, pinacidil or papaverine. Intrarenal infusion of 43Gap 27 or 40Gap 27 decreased basal renal blood flow and increased mean arterial blood pressure, both in the presence and absence of systemic infusion of L-NAME and indomethacin. It is concluded that inhibition of gap junctional communication with connexinmimetic peptides blocks EDHF-mediated signal transmission in vivo. The peptides also decrease basal renal blood flow and increase blood pressure, supporting a role for tonic EDHF release in the control of tissue perfusion and vascular resistance.

1. INTRODUCTION The nature of EDHF is as yet one of the most contentious questions in vascular physiology. Current evidence suggests that EDHF is more than one substance and that the identity, as well as the mechanisms of action of EDHF show substantial tissue and species heterogeneity. Several candidate mediators have been proposed, including epoxyeicosatrienoic acids (EETs), which are cytochrome P450 mono-oxygenase derived metabolites of arachidonic acid (Fisslthaler et al., 1999; Quilley and McGiff, 2000), the endogenous cannabinoid anandamide (Randall and Kendall, 1998) and potassium ions (Edwards et al., 1998). In some blood vessels, however, neither a cytochrome P450 metabolite nor a cannabinoid or potassium meets the pharmacological criteria for an EDHF. Except for the involvement of the opening of K⫹-channels on endothelial and/or smooth muscle cells, the mechanisms of the EDHFpathway are still a matter of debate. The persistent controversy hampers the development of

Connexin-mimetic peptides 79 specific inhibitors that are suitable to elucidate the physiological role of EDHF. Although both cascade bioassay (Popp et al., 1996) and sandwich preparations (Chen et al., 1991) have suggested that EDHF can diffuse freely in the extracellular space, several lines of evidence support the direct transfer of EDHF through myoendothelial gap junctions. Gap junctions are formed when two connexons contributed by neighboring cells dock through interactions between their extracellular loops. A connexon consists of six connexins arranged around an aqueous central pore, that allows the transfer of electrical current and small molecules, ⬍1 kDa in size. Despite the large number of connexins identified, only connexin 43, 40 and 37 are expressed in the mammalian vasculature (Christ et al., 1996). Several pharmacological agents interfere with gap junction formation, including heptanol, the 18␣ and 18␤ isoforms of glycyrrhetinic acid and carbenoxolone, but these molecules may have unwanted aspecific effects when used in vivo. The use of connexin knockout mice (de Wit et al., 2000) or of connexin-mimetic peptides (Chaytor et al., 1998; Griffith and Taylor, 1999; Hutcheson et al., 1999; Dora et al., 1999; Edwards et al., 1999, 2000) may represent more specific means to interfere with gap junction function. Connexin-mimetic peptides possess sequence homology to specific regions of connexin proteins. The interactions between the extracellular loops of connexins are not fixed and immobile, but involve a dynamic docking–undocking mechanism. The connexin-mimetic peptides bind to the essential components of the connexin docking sites and thus interfere with the docking of two connexons contributed by neighboring cells. Little research concerning EDHF has been conducted in vivo, reflecting the toxicity of potassium channel blocking agents and the inability to directly measure endotheliumdependent hyperpolarization, as the hallmark of EDHF activity. Nevertheless, in vivo measurement of the contribution of EDHF to endothelium-dependent vasodilatation may be important, as the magnitude of the response is determined by the resting membrane potential, which is quite different in arteries under physiological pressure in vivo than in those suspended in vitro (Campbell and Harder, 1999). Furthermore, in vivo research allows investigation of the physiological relevance of EDHF, which is currently unknown. The aim of the present study was to characterize the NO synthase- and cyclooxygenaseindependent endothelium-mediated vasodilatation in response to acetylcholine in the renal microcirculation in vivo. More in particular, the potential contribution of gap junctional communication to the EDHF-reponse was evaluated by the use of connexin-mimetic peptides. In addition, the physiological role of EDHF was addressed by evaluating the effects of blockade of EDHF-mediated responses on basal renal blood flow and on systemic arterial blood pressure.

2. MATERIALS AND METHODS Female Wistar rats of ⫾ 250 g (Iffa Credo, Brussels, Belgium) were anesthetized with thiobutabarbital (Inactin, RBI, Natick, USA, 100 mg/kg i.p.). The trachea was intubated, a jugular vein was cannulated for continuous infusion of isotonic saline (3 ml/h) and administration of drugs, and a carotid artery was cannulated for continuous monitoring of arterial blood pressure and recording of heart rate. The right renal and suprarenal arteries were exposed via a small abdominal incision. The suprarenal artery was cannulated for intrarenal administration of drugs. A blood flow sensor with an inner diameter of 0.6–0.8 mm was placed on the renal artery, allowing continuous renal blood flow monitoring by an electromagnetic square wave flow meter (Skalar Medical, Delft, The Netherlands) (De Vriese et al., 2000a).

80

An S. De Vriese et al.

2.1. Experimental protocols Series 1 All studies were performed in the continuous and combined presence of systemic NO synthase and cyclooxygenase blockade: L-NG-nitroarginine methylester HCl (L-NAME, Sigma Chemical Co, St Louis, MO, 10 mg/kg bolus followed by 20 mg/kg/h) and indomethacin (Sigma, 4 mg/kg bolus followed by 8 mg/kg/h). The renal blood flow response to intrarenal acetylcholine (Sigma, 1–50 ng in bolus), to the NO donor detaNONOate (Alexis, Grünberg, Germany, 16–80 ␮g in bolus), to the K⫹channel opener pinacidil (Sigma, 25–125 ␮g in bolus) and to papaverine (Federa, Brussels, Belgium, 10–50 ␮g in bolus) was examined before and 5 and 30 min after infusion of 43gap 27 peptide (3.91 mg, sequence SRPTEKTIFII, synthesized by Sigma-Genosys, Cambridge, UK) (n ⫽ 8), 40gap 27 peptide (3.87 mg, sequence SRPTEKNVFIV) (n ⫽ 6), a control peptide (3.62 mg, sequence SRGGEKNVFIV) (n ⫽ 6) or solvent (1 ml 1% bovine serum albumin in saline) (n ⫽ 6). Before administration of the next dose of acetylcholine, detaNONOate, papaverine and pinacidil, renal blood flow was allowed to return to baseline values. The doses and timing of administration of the pharmacological agents were determined in pilot experiments. The upper limit of the dose-response curve to acetylcholine, detaNONOate, papaverine and pinacidil was chosen as the highest dose that was devoid of systemic blood pressure effects. The dose of the connexin-mimetic peptides was selected as the lowest dose that achieved the maximal effect. Time-response curves showed that the effect of the peptides was maximal after 5 min and generally disappeared after 30 min. Series 2 The renal blood flow response to intrarenal acetylcholine, detaNONOate, pinacidil and papaverine was examined before and 5 and 30 min after infusion of 40gap 27 peptide (n ⫽ 6) or control peptide (n ⫽ 6). The experiments were performed in the absence of systemic NO synthase and cyclooxygenase blockade.

2.2. Statistical analysis The data are presented as mean ⫾ SEM. The renal blood flow response to the different agonists is expressed as the area under the curve of the change in renal blood flow (ml/min ⫻ min) (De Vriese et al., 2000). Analysis of variance, paired and unpaired t-tests were used as appropriate. The significance level was set at P less than 0.05.

3. RESULTS 3.1. NO synthase- and cyclooxygenase-independent renal vasodilatation to acetylcholine Intrarenal infusion of 43Gap 27, a short peptide possessing conserved sequence homology to part of the second extracellular loop of connexin 43, partially inhibited the L-NAME- and indomethacin-resistant renal blood flow response to acetylcholine, when it was evaluated 5 min after administration of the peptide (Figures 10.1 and 10.2). The renal blood flow response recovered to baseline values, when the dose-response curve to acetylcholine was repeated 30 min after infusion of the peptide.

Connexin-mimetic peptides 81

Figure 10.1 Representative trace of a dose-response curve to acetylcholine (ACh) before and after administration of 40Gap 27 peptide. The experiment was performed in the continuous and combined presence of systemic NO synthase and cyclooxygenase blockade. Source: Reproduced with permission from De Vriese et al., 2002.

Figure 10.2 The renal blood flow increase in response to intrarenal acetylcholine after intravenous L-NAME and indomethacin before (squares), 5 min after (circles) and 30 min after (triangles) intrarenal infusion of a connexin-mimetic peptide. (A) 43Gap 27 peptide (n⫽8), *P ⬍ 0.05 vs baseline; (B) 40Gap 27 peptide (n ⫽ 6), *P ⬍ 0.01 vs baseline; (C) control peptide (n ⫽ 6); (D) solvents (n ⫽ 6). The area under the curve (AUC) of the change from baseline values was calculated for each bolus acetylcholine and the data are expressed as mean ⫾ SEM. Source: Reproduced with permission from De Vriese et al., 2002.

Infusion of 40Gap 27, which is homologous to the second extracellular loop of connexin 40, abolished the L-NAME- and indomethacin-resistant renal vasodilatation to acetylcholine 5 min after infusion (Figure 10.1). The response almost completely recovered 30 min after infusion of the peptide. A control peptide, with a replacement of two animo acids in the motif SRPTEK present in the second extracellular loop of connexins 40 and 43, was without effect on the vasodilatation to acetylcholine. The L-NAME- and indomethacin-resistant vasodilatation to acetylcholine remained stable over time after infusion of the solvent without connexin-mimetic peptide (Figure 10.2).

82

An S. De Vriese et al.

Figure 10.3 The renal blood flow increase in response to intrarenal acetylcholine in the absence of L-NAME and indomethacin before (squares), 5 min after (circles) and 30 min after (triangles) intrarenal infusion of a connexin-mimetic peptide. (A) 40Gap 27 peptide (n⫽6), *P⬍0.05 vs baseline; (B) control peptide (n ⫽ 6). The area under the curve (AUC) of the change from baseline values was calculated for each bolus acetylcholine and the data are expressed as mean ⫾ SEM. Source: Reproduced with permission from De Vriese et al., 2002.

3.2. Renal vasodilatation to acetylcholine in the absence of systemic NO synthase- and cyclooxygenase-blockade In the absence of L-NAME and indomethacin, intrarenal infusion of 40Gap 27 decreased the renal vasodilatation to acetylcholine 5 min after infusion (Figure 10.3). The response to acetylcholine recovered to baseline values 30 min after infusion of the peptide. The control peptide did not affect the renal blood flow response to acetylcholine (Figure 10.3).

3.3. Renal vasodilatation to detaNONOate, pinacidil and papaverine 43

Gap 27, 40Gap 27 and the control peptide did not affect the renal vasodilatation to detaNONOate, pinacidil and papaverine when tested under conditions of systemic NO synthase and cyclooxygenase blockade (Figure 10.4). The vasodilatation to detaNONOate, pinacidil and papaverine remained stable over time after infusion of solvent without connexin-mimetic peptide (data not shown). 40Gap 27 and the control peptide did not alter the renal vasodilatation to detaNONOate, pinacidil and papaverine, measured in the absence of L-NAME and indomethacin (data not shown).

3.4. Basal renal blood flow, mean arterial blood pressure and heart rate Basal renal blood flow was not different between the treatment groups. Renal blood flow decreased significantly after systemic L-NAME and indomethacin infusion and to a similar extent in all treatment groups. Infusion of 43Gap 27 caused a mild but significant decrease of basal renal blood flow, in addition to the fall caused by combined NO synthase- and cyclooxygenase-blockade. After intrarenal administration of 40Gap 27, a more pronounced decrease of basal renal blood flow was observed. Thirty minutes after infusion of the peptides, renal blood flow recovered to control values. In contrast, infusion of the control peptide or of solvent did not affect renal blood flow (Figure 10.5). Baseline mean blood pressure was not different between the treatment groups. Concomitant systemic infusion of L-NAME and indomethacin increased mean arterial blood pressure significantly and to a similar extent in the different treatment groups. Infusion of

Figure 10.4 The renal blood flow increase in response to intrarenal detaNONOate, pinacidil and papaverine after intravenous L-NAME and indomethacin before (squares), 5 min after (circles) and 30 min after (triangles) intrarenal infusion of a connexin-mimetic peptide. (left) 43Gap 27 peptide (n ⫽ 8); (right) 40Gap 27 peptide (n ⫽ 6). The area under the curve (AUC) of the change from baseline values was calculated for each bolus detaNONOate and the data are expressed as mean ⫾ SEM.

Figure 10.5 Renal blood flow in baseline conditions (open bars), after intravenous administration of L-NAME with indomethacin (hatched bars), 5 min after intrarenal infusion of a connexinmimetic peptide (closed bars) and 30 min after the infusion of the peptide (squares). (A) 43Gap 27 peptide (n ⫽ 8); (B) 40Gap 27 peptide (n ⫽ 6); (C) control peptide (n ⫽ 6); (D) solvents (n ⫽ 6). *P ⬍ 0.001 vs baseline values, #P ⬍ 0.001 vs after L-NAME with indomethacin, §P ⬍ 0.02 vs 5 min after the peptide. Source: Reproduced with permission from De Vriese et al., 2002.

84

An S. De Vriese et al.

43

Gap 27 caused a mild but significant increase of systemic blood pressure, in addition to the rise caused by combined NO synthase- and cyclooxygenase-blockade. After intrarenal administration of 40Gap 27, a significant rise of systemic blood pressure was observed. Thirty minutes after infusion of the peptides, blood pressure fell to values similar to those before administration of the peptides. Infusion of the control peptide or of the solvent did not affect blood pressure (Figure 10.6). When the experiments were conducted in the absence of L-NAME and indomethacin, intrarenal infusion of 40Gap 27 significantly decreased basal renal blood flow after 5 min, with a recovery to baseline values after 30 min. Administration of the control peptide did not alter basal renal blood flow (Figure 10.7). After infusion of 40Gap 27 without L-NAME and indomethacin, a significant increase of mean arterial blood pressure was observed, whereas blood pressure remained stable after infusion of the control peptide (Figure 10.8). The administration of connexin-mimetic peptides did not induce alterations in heart rate: heart rate was 318 ⫾ 4.9/min and 328 ⫾ 3.7/min before and after 40Gap 27, respectively, and 300 ⫾ 13.7 /min and 293 ⫾ 11.2 /min before and after the control peptide, respectively.

Figure 10.6 Mean arterial blood pressure (BP) in baseline conditions (open bars), after intravenous administration of L-NAME with indomethacin (hatched bars), 5 min after intrarenal infusion of a connexin-mimetic peptide (closed bars) and 30 min after the infusion of the peptide (squares). (A) 43Gap 27 peptide (n ⫽ 8); (B) 40Gap 27 peptide (n ⫽ 6); (C) control peptide (n ⫽ 6); (D) solvents (n ⫽ 6). *P ⬍ 0.01 vs baseline values, #P ⬍ 0.01 vs after L-NAME with indomethacin, §P ⬍ 0.01 vs 5 min after the peptide. Source: Reproduced with permission from De Vriese et al., 2002.

Figure 10.7 Renal blood flow (RBF) baseline conditions (open bars), 5 min after intrarenal infusion of a connexin-mimetic peptide (closed bars) and 30 min after the infusion of the peptide (squares). (A) 40Gap 27 peptide (n ⫽ 6); (B) control peptide (n ⫽ 6). *P ⬍ 0.01 vs baseline values, #P ⬍ 0.05 vs 5 min after the peptide.

Connexin-mimetic peptides 85

Figure 10.8 Mean blood pressure (BP) baseline conditions (open bars), 5 min after intrarenal infusion of a connexin-mimetic peptide (closed bars) and 30 min after the infusion of the peptide (squares). (A) 40Gap 27 peptide (n ⫽ 6); (B) control peptide (n ⫽ 6). *P ⬍ 0.01 vs baseline values, #P ⬍ 0.05 vs 5 min after the peptide.

4. DISCUSSION In the renal microcirculation of the rat, a large residual response to acetylcholine is present after combined high-dose NO synthase- and cyclooxygenase inhibition (De Vriese et al., 2000a). A short peptide homologous to a sequence in the second extracellular loop of connexin 40 is a potent inhibitor of the NO synthase- and cyclooxygenase-independent acetylcholine-induced vasodilatation in the kidney. A peptide homologous to the extracellular loop of connexin 43 also inhibits the NO synthase- and cyclooxygenase-independent acetylcholine-induced vasodilatation, but to a lesser extent. The salient observation made in the present study is that the administration of these connexin-mimetic peptides in vivo substantially decreases basal renal blood flow and increases systemic blood pressure. The results thus suggest that gap junctional communication is essential for EDHF-mediated signal transmission in the kidney in vivo and provide compelling evidence to support a role for EDHF in the control of blood pressure and tissue perfusion. Previous studies have shown that connexin-mimetic peptides are capable of interfering with EDHF-mediated signal transmission in thoracic aorta of the rabbit (Chaytor et al., 1998), superior mesenteric artery of the rabbit (Chaytor et al., 1998; Hutcheson et al., 1999; Dora et al., 1999), jugular vein of the rabbit (Griffith et al., 1999), carotid artery of the guinea-pig (Edwards et al., 1999), hepatic and mesenteric artery of the rat (Edwards et al., 1999) and coronary artery of the pig (Edwards et al., 2000). In all these studies 43Gap 27 peptide was used, thus suggesting that connexin 43 is the most important connexin for EDHF-mediated responses in these blood vessels. In the present study in the renal microcirculation of the rat, 43Gap 27 peptide only partially inhibited the L-NAME- and indomethacin-resistant renal vasodilatation in response to acetylcholine. However, 40Gap 27 peptide abolished the response, supporting a more important role for connexin 40 in this vascular bed. Both 40Gap 27 and 43Gap 27 contain the motif SRPTEK present in the second extracellular loop of connexin 40 and 43. A control peptide, with a replacement of two amino acids in this motif, was inactive, underlining the specificity of the results. The acetylcholine response recovered to almost normal values 30 min after infusion of the 40Gap 27 and 43Gap 27 peptides, suggesting that binding is reversible and that the peptides are removed from the circulation, possibly by glomerular filtration or degradation by proteases. None of the peptides interfered with the responses to detaNONOate, pinacidil and papaverine, indicating specificity for endothelium-dependent responses. None of the gap junctional blocking agents studied to date can discriminate between inhibition of myoendothelial gap junctions and inhibition of gap junctional coupling within the endothelial or smooth muscle layers. It is usually assumed that gap junction blockers affect transfer of an EDHF from the endothelium to smooth muscle cells via myoendothelial

86

An S. De Vriese et al.

gap junctions. An equally tenable assumption is that they inhibit the propagation of hyperpolarization within the endothelial or the smooth muscle cell layer. The exploration of the physiological role of EDHF has been hampered by the absence of selective inhibitors of EDHF that are suitable for in vivo use. Pulsatile changes in intraluminal pressure induce the synthesis of EDHF, the inhibition of which leads to a marked reduction in vascular compliance (Popp et al., 1998). EDHF plays a role in the maintenance of basal perfusion pressure in the perfused mesenteric arterial bed of the rat (Adeagbo and Triggle, 1993). In other studies, however, the removal of endothelium did not affect the resting membrane potentials, suggesting no important tonic release of EDHF under basal conditions, at least not in vitro (Shimokawa et al., 1996). Charybdotoxin and apamin induced only a minor increase in resistance in the isolated perfused hindlimb of eNOS knockout mice, arguing against a significant basal release of EDHF in that model (Brandes et al., 2000). The results of the present study strongly support a tonic release of EDHF in vivo, with a modulation of basal vascular tone. Administration of 40Gap 27 peptide, which abolished the L-NAME- and indomethacin-resistant vasodilatation induced by acetylcholine, also substantially decreased basal renal blood flow and increased systemic blood pressure. As suggested by the absence of changes in heart rate, no major interference with cardiac connexins occurred that could explain the blood pressure effects. Several groups have proposed that EDHF may be a back-up vasodilator mechanism in conditions with compromised bioavailability of NO (Kilpatrick and Cocks, 1994; Brandes et al., 2000). It can, however, be questioned whether such a complex and powerful mechanism would have developed, only to serve as a reserve vasodilator system. However, a similar decrease in renal blood flow and increase in blood pressure was observed in the absence of L-NAME and indomethacin, underlining that the effects are not merely uncovered by NO synthaseand cyclooxygenase-blockade, and arguing against a back-up role for EDHF. The findings are fully in line with the observation that connexin 40-deficient mice are hypertensive (de Wit et al., 2000). If EDHF has a role in the control of vascular resistance and tissue perfusion, the finding of impaired EDHF-mediated relaxation in hypertension (Fujii et al., 1992; Van de Voorde et al., 1992), hypercholesterolemia (Eizawa et al., 1995), diabetes mellitus (De Vriese et al., 2000b) and aging (Fujii et al., 1993) has important implications for the development of vascular complications associated with these risk factors. Conversely, strategies that improve EDHF-release, as demonstrated for folate (De Vriese et al., 2000a) or nifidepine (Fisslthaler et al., 2000), or mimic it, including the administration of K⫹-channel openers, may have potential beneficial therapeutic impact. ACKNOWLEDGMENTS The authors thank Tommy Dheuvaert and Julien Dupont for their expert technical assistance.

11 Urocortin-induced relaxations of the rat coronary artery Yu Huang, Franky Leung Chan, Chi-Wai Lau, Zhen-Yu Chen, Guo-Wei He, Suk-Ying Tsang and Xiaoqiang Yao

The present study was aimed to determine the role of K⫹ channels in the urocortin-induced endothelium-dependent and -independent relaxation in the rat left anterior descending coronary artery. Urocortin induced both endothelium-dependent and -independent relaxations. Removal of the endothelium reduced the relaxant effect of urocortin. In endotheliumintact rings pretreated with NG-nitro-L-arginine methyl ester or ODQ, the urocortin-induced relaxation was similar to that observed without endothelium. Ba2⫹ inhibited the response to urocortin. Combined treatment with NG-nitro-L-arginine methyl ester did not cause further inhibition. In urocortin-relaxed rings, Ba2⫹ or tetraethylammonium ions induced concentrationdependent contractions. Urocortin produced greater relaxations in rings without endothelium contracted by U46619 than by elevated extracellular K⫹. Both tetraethylammonium ions and iberiotoxin inhibited the urocortin-induced endothelium-independent relaxation. In contrast, apamin, Ba2⫹, and glibenclamide were ineffective. The present results indicate that endothelial nitric oxide and subsequent activation of Ba2⫹-sensitive K⫹ channels is likely to be responsible for urocortin-induced endothelium-dependent relaxations, while activation of Ca2⫹-activated K⫹ channels may partly mediate the endothelium-independent relaxations to urocortin.

Corticotropin-releasing factor (CRF) produced mainly in the hypothalamus, is the major neuromediator of the hypothalamic-pituitary-adrenal stress axis in mammals (Vale et al., 1981). Several structurally related peptides including fish neuropeptide urotensin, sauvagine, and urocortin belong to the CRF family. Urocortin or CRF exerts positive inotropic effects, elevates cardiac output and causes coronary vasodilatation in sheep and rats (Parkes et al., 1997; Terui et al., 2001). This effect may partly be caused by an increased production of cardiac cyclic AMP (Heldwein et al., 1996). Urocortin and CRF-related peptides decrease coronary perfusion pressure in isolated rat hearts because they cause coronary vasodilatation (Terui et al., 2001). The coronary vasodilating and the positive ionotropic effects of urocortin suggest a potential cardiac protective action. However, the exact mechanisms underlying the urocortin-induced coronary vasodilatation remain unclear. In view of the important role of K⫹ channels in the regulation of vascular tone, the aim of the present study was to investigate effects of various putative K⫹ channel blockers on relaxations to urocortin in isolated rat coronary artery rings.

88 Yu Huang et al.

1. METHODS 1.1. Tissue preparation Male Sprague–Dawley rats (~300 g) were sacrificed. The heart was removed rapidly and placed in a dissecting plate containing ice-cold Krebs solution. The left anterior descending arteries were dissected out carefully and cut into two rings (~1.5 mm in length). Each ring was mounted in a Multi Myograph System (Danish Myo Technology A/S, Denmark). The chamber bath was filled with 5 ml of Krebs solution. The Krebs solution contained (in mM): NaCl 119, KCl 4.7, CaCl2 2.5, MgCl2 1, NaHCO3 25, KH2PO4 1.2, and D-glucose 11. Each ring was bathed in Krebs solution that was continually oxygenated with a gas mixture of 95% O2 plus 5% CO2 at 37 ⬚C. In some rings the endothelium was removed mechanically by rubbing the lumen of the ring several times with a small stainless steel wire. The functional removal was confirmed if the ring failed to relax in response to 3 ⫻ 10⫺6 M acetylcholine. In experiments using high K⫹-containing solution, an equimolar concentration of K⫹ replaced Na⫹ in order to maintain a constant ionic strength.

1.2. Experimental protocols Thirty minutes after mounting, each ring was initially contracted by 3 ⫻ 10⫺8 M U46619 to test its contractility. The ring was then rinsed three times with an interval of 20 min and baseline tone (0.5 mN) was readjusted when necessary. In the first set of experiments, U46619 was used to generate steady tone; the relaxations to cumulative concentrations of urocortin were subsequently studied in control and in the presence of 10⫺4 M NG-nitro-L-arginine methyl ester (L-NAME), 10⫺4 M Ba2⫹, or L-NAME plus BaCl2. In some experiments the effects of Ba2⫹ or tetraethylammonium were tested in rings that had been previously relaxed by urocortin. In the second group of experiments using rings without endothelium, the effects of tetraethylammonium and iberiotoxin were examined on the relaxant responses to urocortin. The relaxant effect to urocortin was also tested in rings contracted by high K⫹ (3.5 ⫻ 10⫺2 or 5 ⫻ 10⫺2 M) as compared with the rings constricted by U46619. In all experiments, a similar initial tone was induced by adjusting the concentration of U46619 in different rings.

1.3. Drugs Phenylephrine, acetylcholine, urocortin (human), indomethacin, NG-nitro-L-arginine methyl ester (L-NAME), charybdotoxin, apamin, U46619, tetraethylammonium, iberiotoxin, sodium nitroprusside (Sigma, St Louis., MO, USA). 1H-[1,2,4]oxadiazolo[4,2-␣]quinoxalin-1-one (ODQ) (Tocris, UK). U46619 and ODQ were dissolved in dimethyl sulfoxide (DMSO), which at 0.2% (v/v) did not affect U46619-induced tension.

1.4. Statistics Data are means ⫾ SEM of n rings prepared from different rats. Concentration–relaxation curves were constructed based on responses to cumulative concentrations of urocortin and analyzed by non-linear curve fitting using Graphpad software (Version 3.0). pD2 is the negative logarithm of the concentration required to produce 50% the maximal relaxation (Emax). Statistical analysis was performed by using Student’s t test or one-way analysis of variance followed by Newman–Keuls test. A P-value less than 0.05 was taken to indicate a statistically significant difference.

Urocortin-induced relaxations of the rat coronary artery

89

2. RESULTS 2.1. Endothelium-dependent relaxation induced by urocortin The cumulative application of urocortin induced concentration-dependent relaxations in rings with endothelium (pD2: 8.71⫾0.03, n⫽6, Figure 11.1). Urocortin produced significantly less relaxation in rings without endothelium without an effect on the maximal relaxation. Treatment with 10⫺4 M L-NAME or 3⫻10⫺6 M ODQ attenuated urocortin-induced relaxation to the same as that obtained in rings without endothelium. Urocortin-induced concentrationdependent relaxations were inhibited significantly following treatment with 10⫺4 M Ba2⫹ (pD2: 8.06 ⫾ 0.09, P ⬍ 0.05 as compared with control, Figure 11.1). Combined treatment with 10⫺4 M Ba2⫹ plus 10⫺4 M L-NAME did not further inhibit urocortin-induced relaxation (pD2: 7.93 ⫾ 0.12). During exposure to urocortin (10⫺8 M), Ba2⫹ (10⫺6 ⫺ 10⫺4 M) or tetraethylammonium (5 ⫻ 10⫺5 ⫺ 3 ⫻ 10⫺3 M) concentration dependently reversed the urocortin-induced relaxation (Figure 11.1). In rings without endothelium, treatment with 10⫺4 M Ba2⫹ had no effect on the urocortin-induced relaxation (pD2: 7.67 ⫾ 0.21, n ⫽ 5, P ⬎ 0.05 compared with control) but inhibited the response to sodium nitroprusside (pD2: 8.03 ⫾ 0.03 in control and 7.49 ⫾ 0.09 in Ba2⫹, P ⬍ 0.05). In rings with endothelium, urocortin-induced relaxation was unaffected by indomethacin (10⫺5 M), ouabain (10⫺4 M), glibenclamide (10⫺6 M), or charybdotoxin plus apamin (each at 10⫺7 M).

Figure 11.1 Representative traces showing relaxations to urocortin in isolated rat coronary artery rings in control (A), in the presence of 10⫺4 M Ba2⫹ (B), or of 10⫺4 M L-NAME plus 10⫺4 M Ba2⫹ (C). Scale bars apply to all traces. Concentration-response curves for the urocortininduced relaxations under three conditions (D). Ba2⫹ – or tetraethylammonium-induced reversal of vessel tones in 10⫺8 M urocortin-relaxed rings (E). All experiments were conducted in rings with endothelium. Data are means ⫾ SEM of six experiments.

90

Yu Huang et al.

Figure 11.2 Representative traces showing the urocortin-induced relaxation of coronary artery rings contracted by U46619 in control (A) and in the presence of 10⫺3 M tetraethylammonium (B) or 10⫺7 M iberiotoxin (C). All experiments were performed in rings without endothelium. Data are means ⫾ SEM of six experiments.

2.2. Endothelium-independent relaxation induced by urocortin In U46619-contracted rings without endothelium, urocortin induced relaxations (Figure 11.2) with a pD2 of 8.31 ⫾ 0.04 (n ⫽ 6) and the relaxations were attenuated by treatment with 10⫺3 M tetraethylammonium or 10⫺7 M iberiotoxin (Figure 11.2). Urocortin produced significantly less relaxant effect in K⫹ (5 ⫻ 10⫺2 M)-contracted rings (Figure 11.3). Urocortin-induced partial relaxation in rings contracted with 3.5 ⫻ 10⫺2 M K⫹ and this relaxation was inhibited by 10⫺3 M tetraethylammonium (Figure 11.3). The maximal relaxation was reduced by increasing the extracellular K⫹ concentration (Figure 11.3). The maximal response to urocortin was similar in rings contracted with 5 ⫻ 10⫺2 M K⫹ or with 3.5 ⫻ 10⫺2 M K⫹ in the presence of tetraethylammonium (n ⫽ 5, Figure 11.3). In contrast, apamin at 10⫺7 M had no effect. Both tetraethylammonium and iberiotoxin induced a steady increase in tone (Figure 11.2).

3. DISCUSSION The present study shows that urocortin induced both endothelium-dependent and independent relaxations of the coronary artery of the rat. The endothelium-dependent response was likely caused by endothelial nitric oxide and inhibited by Ba2⫹, whilst the endotheliumindependent response was mediated partially through activation of large-conductance Ca2⫹activated K⫹ channels. CRF and urocortin reduce coronary perfusion pressure in isolated rat hearts, indicating that both peptides possess a coronary vasodilator effect (Terui et al., 2001). The present results demonstrate a potent vasorelaxant effect of urocortin with an IC50 of 2.9 ⫻ 10⫺9 M in the isolated rat coronary arteries with intact endothelium but this effect was significantly

Urocortin-induced relaxations of the rat coronary artery

91

Figure 11.3 Representative traces showing urocortin-induced relaxations of rings contracted with 3.5⫻10⫺2 M K⫹ in the absence (A) and presence of 10⫺3 M tetraethylammonium (B). The maximal relaxant relation to urocortin under four conditions (C). All experiments were performed in rings without endothelium. Data are means⫾SEM of five to six experiments. The asterisk (*) indicates a statistically significant difference from control (P⬍0.05).

less following treatment with L-NAME or ODQ and in rings without endothelium, although the maximal response to urocortin was unchanged. This indicates that both nitric oxide and cyclic GMP are involved. The CRF-induced relaxation is blunted markedly by removal of the endothelium in the rat aorta (Jain et al., 1997). Similarly, NG-nitro-L-arginine, another inhibitor of nitric oxide synthase, reduces the vasodilator effect of CRF in the isolated rat heart (Grunt et al., 1993). The urocortin-mediated relaxant effect was significantly inhibited by Ba2⫹ at concentrations used to selectively block the activity of inward rectifier K⫹ (KIR) channels in coronary arterial smooth muscle cells (Robertson et al., 1996). In the urocortin-relaxed rings, Ba2⫹ induces concentration-dependent contractions. Ba2⫹ failed to inhibit urocortin-induced relaxation in the presence of L-NAME or in rings without endothelium. Ba2⫹ also inhibited the relaxant response to sodium nitroprusside, an exogenous nitric oxide donor in rings without endothelium. It is likely that urocortin stimulates release of endothelial nitric oxide that in turn activates the Ba2⫹-sensitive K⫹ channels probably via a cyclic GMP-dependent mechanism in the arterial smooth muscle cells. Both Ba2⫹ and inhibition of nitric oxide synthase reduce flow-induced relaxations only in the rabbit middle cerebral arteries with endothelium (Wellman and Bevan, 1995). The other novel finding of the present investigation is that activation of the Ca2⫹-activated ⫹ K channels is likely to be involved in the urocortin-induced endothelium-independent relaxations of the coronary arteries. The urocortin-induced relaxation was impaired in rings constricted with K⫹. One major consequence of raising extracellular K⫹ concentration is to narrow the electrochemical driving force for K⫹ efflux, so that the effect of K⫹ channel

92

Yu Huang et al.

Figure 11.4 Possible mechanisms underlying the endothelium-dependent and -independent relaxation to urocortin in the rat coronary artery. Urocortin stimulates the endothelial cells to release nitric oxide. Nitric oxide stimulates soluble guanylate cyclase and cyclic GMP-dependent pathway that in turn activates Ba2⫹-sensitive K⫹ channels in vascular smooth muscle cells. Urocortin binds to CRF receptors to induce relaxation by activating Ca2⫹-activated K⫹ channels via unknown intracellular pathways in vascular smooth muscle cells. BK channel, large-conductance Ca2⫹-activated K⫹ channel; CRF-R, Corticotropin-releasing factor receptor; GC, guanylate cyclase; NO, nitric oxide; NOS, nitric oxide synthase; PKG, cyclic GMP-dependent protein kinase.

activation would be neglected. The urocortin-induced coronary relaxations were inhibited to the same extent by tetraethylammonium or iberiotoxin, putative blockers of Ca2⫹-activated K⫹ channels in U46619-contracted rings. Tetraethylammonium also reduced the urocortin response in K⫹-contracted rings. In non-vascular smooth muscle cells, urocortin produces hyperpolarization via stimulation of Ca2⫹-activated K⫹ currents (Petkova-Kirova et al., 2000). In contrast, apamin, Ba2⫹ or glibenclamide were without effect, suggesting that other types of K⫹ channels play a very minor role. In summary, both endothelial nitric oxide and subsequent activation of Ba2⫹-sensitive K⫹ channel in arterial smooth muscle cells mediate the endothelium-dependent component of urocortin-induced coronary relaxation. A cyclic GMP-dependent mechanism is involved. Urocortin may activate Ca2⫹-activated K⫹ channels to mediate part of the urocortin-induced endothelium-independent relaxation. Immunoreactivity of urocortin has been detected in the endothelial layer of the rat coronary arteries (Huang et al., 2002). This suggests that urocortin may be locally produced in blood vessels. If this were to occur in the endothelial cells, urocortin may act as a new candidate as an endothelium-derived relaxing factor (Figure 11.4). ACKNOWLEDGMENTS The research was supported by CUHK Direct Grant and CUHK Mainline Research Grant.

12 Nitric oxide is the only EDHF released by the endothelium in lymphatic vessels of the guinea-pig mesentery Pierre-Yves von der Weid and Alice K. Chan

The endothelium of lymphatic vessels plays an important role in the regulation of spontaneous rhythmical constrictions of the smooth muscle, which underlies the propulsion of lymph. In lymphatic vessels of the guinea-pig mesentery, acetylcholine decreases lymphatic pumping activity and hyperpolarizes the lymphatic smooth muscle membrane potential. These responses are caused by the endothelial release of nitric oxide which is, under the used experimental conditions, the only endothelium-derived hyperpolarizing factor (EDHF) in lymphatic vessels. As a further evaluation of the mechanisms of endothelium-dependent hyperpolarization in lymphatic smooth muscle, both electrophysiological and contractile responses were investigated in vitro in lymphatic vessels isolated from the mesentery of young guinea-pigs using cyclopiazonic acid and the proteinase-activated receptor-activating peptide SLIGRL-NH2. These two compounds induce an endothelium-dependent hyperpolarization of the smooth muscle in vascular beds. The hyperpolarization to both compounds is abolished by NG-nitro L-arginine and is reverted in some preparations to a depolarization suggesting to result from a direct action on the smooth muscle. These findings suggest that the hyperpolarization of lymphatic smooth muscle by cyclopiazonic acid and SLIGRL-NH2 results from the endothelial release of nitric oxide. They further confirm nitric oxide as the only factor released by the stimulated lymphatic endothelium able to cause a hyperpolarization in lymphatic smooth muscle.

1. INTRODUCTION The lymphatic circulation maintains tissue fluid homeostasis through removing fluid and proteins that accumulate in the interstitium and return them to the systemic circulation. Lymph transport depends mainly upon the active constriction of the successive chambers that comprise the collecting lymphatic vessels. The contractile mechanism is intrinsic to the smooth muscle present in the vessel wall and consequent to L-type Ca2⫹ channel-mediated action potential (McHale and Allen, 1983). In lymphatic vessels found in the mesentery of the guinea-pig, the pacemaker mechanism underlying the generation of action potential is due to the summation of spontaneous electrical events termed spontaneous transient depolarizations (Van Helden, 1993; Van Helden et al., 1996). As in blood vessels, the endothelial cell layer covering the luminal side of lymphatic vessels plays a key role in the regulation of smooth muscle tone (Ohhashi and Takahashi, 1991; Ferguson, 1992). Most importantly, the endothelium is involved in the modulation of the frequency of spontaneous constrictions (Yokoyama and Ohhashi, 1993; Reeder et al., 1994; von der Weid et al., 1996; Rayner and Van Helden, 1997; Gao et al., 1999), through the release of vasoactive substances. Lymphatic pumping increases following stimulation with substance P or ATP via the endothelial release of thromboxane A2 (Rayner and Van Helden, 1997; Gao et al., 1999). An endothelium-dependent decrease in lymphatic pumping activity was demonstrated after superfusion with acetylcholine and shown to be due to the release of

94

Pierre-Yves von der Weid and Alice K. Chan

nitric oxide (Yokoyama and Ohhashi, 1993; Reeder et al., 1994; von der Weid et al., 1996). In lymphatic vessels of the guinea-pig mesentery, the nitric oxide-induced inhibition of pumping is mediated by two distinct actions on the smooth muscle membrane potential: (a) a nitric oxide-induced decrease in the frequency and amplitude of the spontaneous transient depolarizations, and (b) a nitric oxide-induced hyperpolarization (von der Weid et al., 1996; 2001). To date, no other EDHFs are known to cause an endothelium-dependent hyperpolarization in lymphatic smooth muscle. As well, no electrical coupling between endothelial and smooth muscle cells has been demonstrated to account for endothelium-dependent hyperpolarization in lymphatic vessels (von der Weid and Van Helden, 1997). Depending on the vascular bed, endothelium-dependent hyperpolarizations may participate in the endothelium-dependent relaxations caused by agents such as acetylcholine, ATP, bradykinin or substance P (Bény and von der Weid, 1993; Triggle et al., 1999; Félétou and Vanhoutte, 1999a). The selective inhibitor of the endoplasmic reticulum Ca2⫹-ATPase, cyclopiazonic acid, activates an EDHF-dependent mechanism in vascular beds (Fukao et al., 1995; Lagaud et al., 1999; Tomioka et al., 2001). Activation of the proteinase-activated receptor 2 with the activating peptide, SLIGRL-NH2, causes endothelium-dependent relaxation in blood vessels, accompanied by an endothelium-dependent hyperpolarization (Al-Ani et al., 1995; Saiffedine et al., 1996; Emilsson et al., 1997; Moffatt and Cocks, 1998; McGuire et al., 2002; McLean et al., 2002). The present study aimed to determine whether cyclopiazonic acid and SLIGRL-NH2, induce an endothelium-dependent hyperpolarization in guinea-pig mesenteric lymphatic vessels and to evaluate the importance of this hyperpolarization in the inhibition of spontaneous lymphatic contractile activity. 2. METHODS

2.1. Tissue preparation Guinea-pigs (five to fifteen days of age) of either sex were killed by decapitation during deep anesthesia consequent to inhalation of halothane (5–10%). This procedure has been approved by the University of Calgary Animal Care and Ethics Committee and conforms to the guidelines established by the Canadian Council on Animal Care. The small intestine and attached mesentery were removed rapidly and placed in a physiological saline solution of the following composition (10⫺3 M): CaCl2 2.5; KCl 5; MgCl2 2; NaCl 120; NaHCO3 25; NaH2PO4 1; glucose 11. The pH was maintained at 7.4 by constant bubbling with 95% O2/5% CO2.

2.2. Measurement of vessel contractile activity Small collecting lymphatic vessels (diameter less than 230 ␮m) supplying the jejunum and ileum were dissected together with their associated artery and vein and left intact within the surrounding mesentery. The mesentery was used to pin out the tissues on the Sylgard-coated base of a 2 ml organ bath, mounted on the stage of an inverted microscope (Olympus, CK40) and continuously superfused at a flow rate of 3 ml/min with the physiological saline solution heated to 36 ⬚C. Contractile activity is dependent on vessel filling (Florey, 1927; Smith, 1949), the lumen of the vessel was thus perfused to induce a consistent rate of vessel constrictions. Perfusion was performed by inserting a fine glass micropipette cannula into the lumen of the vessel after it had been cut. The cannula was connected to an infusion pump via a Teflon tubing allowing the lumen to be perfused in the direction of the valves at a flow

Nitric oxide, the only EDHF released by the endothelium

95

rate of 2.5 ␮l/min. This flow rate was selected from preliminary experiments as the most consistent in inducing a regular rhythmical contractile activity in lymphatic vessels in the range of diameter used and for the duration of the experiment (typically 3–4 h). As normal Ca2⫹-physiological saline solution tended to block the cannula, a low-calcium solution, in which 3 ⫻ 10⫺4 M CaCl2 was substituted for 2.5 ⫻ 10⫺3 M, was used. Perfusion with this solution did not affect vessel contractile activity, nor endothelial responsiveness (von der Weid et al., 1996). The lymphatic vessels were monitored using a video camera attached to the microscope and connected to a monitor. Changes in vessel diameter and constriction frequency of the lymphatic chambers were analyzed in real time, or off-line from images recorded on videotape, using a video-dimension analyzer (model V94, Living System Instrumentation, Burlington, VT). Data recorded on a computer via an analog-to-digital converter (PowerLab/4SP, ADInstruments, Mountain View, CA). Preparations were allowed a 30-min equilibration period prior to the first application of agonist. Agonists were applied for 4 min into the superfusion solution and effect on constriction frequency was assessed during the minutes of maximal effect within a 10-min period. In experiments where the effects of inhibitors were investigated, agonists were applied to the superfusion solution after the inhibitor was present for at least 15 min. Data from different vessels were averaged and expressed as percentage of the mean of the constriction/min values obtained during a 5-min period immediately preceding the agonist application.

2.3. Electrophysiology Lymphatic vessels attached to the mesentery were pinned onto the sylgard-coated bottom of a small organ bath (volume 100 ␮l), mounted on the stage of an inverted microscope (TMS, Nikon) and superfused (flow rate of 3 ml/min) with the physiological saline solution heated to 36 ⬚C. The resting membrane potential was measured using conventional glass intracellular microelectrodes with resistances of 150–250 M⍀ when filled with 0.5 M KCl. Electrodes were connected to an amplifier (Intra 767, World Precision Instruments, Sarasota, FL) through an Ag-AgCl half-cell. The resting membrane potential was monitored on a digital oscilloscope (VC6525, Hitachi) and recorded simultaneously on a computer via a PowerLab/4SP. Impalements of smooth muscle cells were obtained from the adventitial side of lymphatic vessels, cut into short segments (length 125–350 ␮m) with fine dissecting scissors. Short segments were used in order to ensure simplified electrical properties of the smooth muscle such that electrical activity, even if generated at localized foci within the smooth muscle, produced similar potential changes in all the smooth muscle cells of the segment (Van Helden, 1993). Lymphatic smooth muscle impalements were characterized by a sharp drop in potential that settled after 10–15 s to a value typically more negative than ⫺45 mV. Impalements were maintained for more than 5 min in more than 90% of the cases and up to 3 h in the rest of them. In experiments where the effect of inhibitors on the agonist response was studied, agonists were applied first as a control and then, at least 20 min later in the presence of the inhibitor that had been superfused for at least 10 min. This protocol was performed optimally during the same impalement. However in some instances, successive impalements were obtained from neighboring cells in the same segment. No significant difference in the response, induced by the same agonist applied at the same concentration 20 min apart in the absence of an inhibitor, was observed. Spontaneous transient depolarization activity was assessed by measuring the frequency and amplitude of events greater than 1 mV. Frequency and amplitude of spontaneous transient

96

Pierre-Yves von der Weid and Alice K. Chan

depolarizations occurring during an interval of 15 s to 1 min (depending on the stability of the recording, but typically 30 s) before agonist application were compared with that occurring during a period of the same duration while the maximal response to the agonist was observed.

2.4. Lysis of endothelium The lymphatic endothelium was damaged in vitro by repeatedly passing brief streams of air through the lumen of the vessel (five to six times for 5–10 s) at a rate of about 3 ␮l/min. The success of the endothelial destruction was confirmed by applying acetylcholine (10⫺5 M) followed by sodium nitroprusside (10⫺4 M) into the superfusion solution, during vessel lumen perfusion. A negative response to acetylcholine and a positive response to sodium nitroprusside were used as confirmation of the success of the procedure. Endothelial destruction based on this testing procedure proved successful in about 50% of treated vessels. The use of sodium nitroprusside was necessary, as it has been shown that 40% of guinea-pig mesenteric lymphatic vessels that had an intact endothelium did not respond in any way to either acetylcholine or sodium nitroprusside. The main reason for that was shown to be due to a high basal production of nitric oxide (von der Weid et al., 1996).

2.5. Chemicals and drugs Acetylcholine, cyclopiazonic acid and NG-nitro L-arginine were purchased from Sigma/Aldrich (Oakville, ONT). SLIGRL-NH2 was synthesized by the Peptide Synthesis Core Facility (University of Calgary) and was a generous gift from Dr M.D. Hollenberg (University of Calgary). All drugs were dissolved in distilled water (with the exception of NG-nitro L-arginine that was dissolved in 0.1 M HCl) to give 10⫺2 M stock solutions.

2.6. Statistical analysis Experimental data have been expressed as the mean ⫾ standard error of the mean (SEM). Statistical significance was assessed using paired or unpaired Student’s t-test with P less than 0.05 being considered statistically significant. 3. RESULTS Intraluminal perfusion of lymphatic vessels allows the establishment of a rhythmical contractile activity of 4 to 15 constriction/min. Application of cyclopiazonic acid (10⫺5 M, 4 min) modified this activity, by inducing an initial increase in constriction frequency, followed by a decrease, sometimes marginal, in constriction frequency (Figure 12.1). This response was not significantly attenuated by NG-nitro L-arginine (10⫺4 M). Cyclopiazonic acid caused a complex response of the smooth muscle. The response was usually characterized by an increase in activity of spontaneous transient depolarizations (leading to action potentials in some cases), followed by a decrease in activity. The latter was correlated with a hyperpolarization of 11 ⫾ 4 mV, developed from a resting potential of ⫺51 ⫾ 1 mV (n ⫽ 8; Figure 12.2). In the presence of NG-nitro L-arginine (10⫺4 M), the decrease in spontaneous transient depolarization activity persisted whereas the hyperpolarization was inhibited (Figure 12.2). In some preparations, the hyperpolarization was replaced by a long-lasting depolarization (Figure 12.2).

Figure 12.1 Effect of cyclopiazonic acid on the contractile activity of lymphatic vessel in the guinea-pig mesentery. (A) Original trace of vessel diameter changes in an actively constricting lymphatic chamber where downward deflections represent constrictions. (B) Histogram comparing the mean response (⫾ SEM) to cyclopiazonic acid (CPA, 10⫺5 M) applied for 4 min under control conditions (open bar) and in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M, closed bar).

Figure 12.2 Effects of cyclopiazonic acid on membrane potential in guinea-pig mesenteric lymphatic smooth muscle. (A) Intracellular microelectrode recordings in response to cyclopiazonic acid (CPA, 10⫺5 M) applied for the duration of the horizontal bars under control conditions (left trace) and in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M, right trace). Mean data (⫾ SEM) are summarized in (B). Resting membrane potential values are indicated on the left side of the traces.

98

Pierre-Yves von der Weid and Alice K. Chan

Figure 12.3 Effect of the proteinase-activated receptor 2-activating peptide SLIGRL-NH2 on the contractile activity of lymphatic vessel in the guinea-pig mesentery. (A) Original trace of vessel diameter changes in an actively constricting lymphatic chamber where downward deflections represent constrictions. (B) Histogram comparing the mean response (⫾ SEM) to the inactive reverse peptide LRGILS-NH2 (5⫻10⫺5 M) and to SLIGRL-NH2 (10⫺5 M) applied for 4 min under control conditions, in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M) and after destruction of the endothelium.

Application of SLIGRL-NH2 (10⫺6–10⫺5 M) caused a decrease in the frequency of constriction of perfused lymphatic vessels that was often preceded by a brief increase in rate (Figure 12.3). The frequency of constriction was not significantly affected in the presence of the reverse peptide LRGILS-NH2 (5 ⫻ 10⫺5 M). SLIGRL-NH2 decrease in vessel pumping was significantly inhibited in the presence of NG-nitro L-arginine (10⫺4 M), and following destruction of the endothelium (Figure 12.3). In non-perfused vessels, intracellular microelectrode recordings showed that SLIGRLNH2 usually induced an initial increase followed by a decrease in spontaneous transient depolarization frequency and amplitude. The latter was accompanied with a hyperpolarization of 8 ⫾ 1 mV from a resting potential value of ⫺51 ⫾ 1 mV (n ⫽ 10, Figure 12.4). The hyperpolarization was abolished by NG-nitro L-arginine (10⫺4 M; Figure 12.4) and in two preparations the initial depolarization was prolonged together with the increase in spontaneous transient depolarization activity. 4. DISCUSSION The present observations show that cyclopiazonic acid and SLIGRL-NH2 modulate contractile and electrophysiological activities of lymphatic smooth muscle.

Nitric oxide, the only EDHF released by the endothelium

99

Figure 12.4 Effects of SLIGRL-NH2 on the membrane potential in guinea-pig mesenteric lymphatic smooth muscle. (A) Intracellular microelectrode recordings in response to SLIGRLNH2 (10⫺5 M) applied for 1min (horizontal bars) under control conditions (left trace) and in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M, right trace). Mean data (⫾ SEM) are summarized in (B). Resting membrane potential values are indicated on the left side of the traces.

Proteinase-activated receptor 2-activating peptides caused endothelium-dependent relaxations of blood vessels (Al-Ani et al., 1995; Hamilton et al., 1998; Hwa et al., 1996; Roy et al., 1997; Sobey and Cocks, 1998; Sobey et al., 1999). The relaxation is mediated by the endothelial release of nitric oxide, consequent to activation of the constitutive form of nitric oxide-synthase (Al-Ani et al., 1995; Saiffedine et al., 1996; Emilsson et al., 1997; Moffatt and Cocks, 1998). The present result that a SLIGRL-NH2-induced decrease in lymphatic pumping was abolished after endothelial destruction and nitric oxide-synthase inhibition by NG-nitro L-arginine suggests the involvement of endothelium-derived nitric oxide, with no contribution of other relaxing factors, at least under the used experimental conditions. However, intracellular membrane potential recordings revealed that the nitric oxide-dependent action of SLIGRL-NH2 caused the lymphatic smooth muscle to hyperpolarize. This observation confirms that nitric oxide either produced by the endothelium after stimulation with acetylcholine or released by nitric oxide-donors hyperpolarizes lymphatic smooth muscle (von der Weid et al., 1996; 2001). This finding contrasts with reports indicating that SLIGRL-NH2-induced relaxation involves an endothelium-dependent hyperpolarization independent from nitric oxide-synthase, soluble guanylate-cyclase and cyclooxygenase activations in mouse mesenteric arterioles (McGuire et al., 2002) and in the rat coronary vasculature (McLean et al., 2002). In these two studies the hyperpolarization appeared to be mediated by apamin/charybotoxin-sensitive K⫹ channel(s) in the mesenteric artery and by a lipoxygenase-derived eicosanoid in the coronary artery. In addition to the contractile inhibition and hyperpolarization, SLIGRL-NH2 also caused a short increase in pumping and in the activity of spontaneous transient depolarizations, events involved in the initiation of lymphatic pumping (Van Helden, 1993). This increase in activity persisted and was sometimes enhanced after NG-nitro L-arginine treatment or endothelial destruction (Chan and von der Weid, 2002), suggesting a direct action of SLIGRL-NH2 on the lymphatic smooth muscle, as described for arteries in the mouse renal vasculature (Moffatt and Cocks, 1998).

100

Pierre-Yves von der Weid and Alice K. Chan

Cyclopiazonic acid decreased lymphatic pumping and hyperpolarized the smooth muscle. The hyperpolarization was strongly inhibited by NG-nitro L-arginine, further suggesting the predominant role of nitric oxide. However, cyclopiazonic acid still reduced the contractile activity in the presence of NG-nitro L-arginine. This discrepancy in the ability of NG-nitro L-arginine to block the effects of cyclopiazonic acid could be explained if cyclopiazonic acid inhibits Ca2⫹-ATPase on the smooth muscle, leading to the decrease in cytosolic Ca2⫹ available for initiation of pacemaking and subsequent vessel pumping. This possibility is supported by the observation that the decrease in spontaneous transient depolarization activity caused by cyclopiazonic acid persisted in the presence of NG-nitro L-arginine. The present study further substantiates the important role played by the endothelium in the modulation of lymphatic pumping by vasoactive agents. In addition, these results show that endothelium-dependent hyperpolarization and associated pumping inhibition caused by cyclopiazonic acid and SLIGRL-NH2, like those caused by acetylcholine, are mediated by nitric oxide. ACKNOWLEDGMENTS This study was supported by grants from the Alberta Heritage Foundation for Medical Research (AHFMR) and the Heart and Stroke Foundation of Canada. Pierre-Yves von der Weid is an AHFMR Scholar. The authors wish to thank S. Roizes for excellent technical assistance and Dr M.D. Hollenberg for the gift of the peptides.

13 Role of EDHF in vascular tone in vivo Helena C. Parkington, Harold A. Coleman and Marianne Tare

The importance of the endothelium in the regulation of vascular tone is well established. Blockade of endothelium-derived nitric oxide (NO) synthesis in vivo results in an increase in mean arterial pressure of about 50 mmHg and a reduction in basal blood flow in every bed studied. Endothelium-derived eicosanoids or EDHF appear to have little influence on mean arterial pressure or basal blood flow. Thus NO is the dominant player in determining basal tone. In contrast, although NO synthase inhibitors all but abolish chemically induced increases in blood flow in large vessels, they leave intact a substantial blood flow increase in smaller vessels. This residual flow can be accounted for by endothelium-derived hyperpolarizing factor (EDHF). Pharmacological and molecular biological approaches have identified EDHF as an eicosanoid in the coronary arterial bed in many, but not all, species. In rat mesenteric, hindlimb and renal arteries, EDHF is likely mediated by electrotonic spread of the hyperpolarization generated in the endothelial cells spreading to the underlying smooth muscle cells via myoendothelial gap junctions.

1. INTRODUCTION Vascular tone is a pivotal determinant of tissue perfusion. The level of tone represents the balance between contraction and relaxation influences on the smooth muscle in the blood vessel wall. The pioneering work of Furchgott and Zawadzki in 1980 revealed that the endothelium contributes significantly to the relaxation component of the balance dictating vessel tone. Subsequently, nitric oxide (NO), an eicosanoid and endothelium-derived hyperpolarizing factor (EDHF) were identified as the major relaxing elements involved. Understandably, the emphasis has been on isolated vessels, in which an understanding of mechanism has been a major focus. The complicated system that is the intact animal is reflected in a smaller number of studies of events occurring in vivo, but an understanding of the role of the endothelium in health and in disease relies on such knowledge. The objective of this chapter is to attempt to draw together in comparison in vitro and in vivo information as much as possible. 2. ENDOTHELIUM-DEPENDENT RELAXATION IN VIVO Of the three major endothelium-derived relaxing factors, NO is the most thoroughly studied. The development of blockers of NO synthase made possible detailed assessment of the role of NO in determining vascular tone, initially in vitro (Palmer et al., 1987) and subsequently in vivo in which the blockers promptly evoked a substantial increase in arterial pressure (Rees et al., 1989; Vallance et al., 1989; Gardiner et al., 1990) (Figure 13.1). From those experiments it is evident that NO is continuously released from the endothelium in vivo, stimulated either by circulating agents or as a result of shear stress on the endothelial cells. Thus, NO is clearly important in ensuring adequate basal tissue perfusion. Administration of the amino acid L-arginine, the precursor of NO, may restore impaired endothelium-dependent

102

Helena C. Parkington et al.

Figure 13.1 (A) Nitric oxide and prostanoid production were blocked by intrajugular administration of 10 mg/Kg N␻-nitro-L-arginine methyl ester (L-NAME) and 3 mg/Kg indomethacin (indo), respectively. The ion channels mediating the EDHF response were blocked by close intra-arterial infusion of 3 ⫻ 10⫺8 mol/l charybdotoxin (ChTx) and 2.5 ⫻ 10⫺7 mol/l apamin. Mean arterial pressure (MAP) and basal blood flow were recorded. (B) The endothelium was stimulated by close intra-arterial infusion of acetylcholine (ACh) (10 s bolus) in the absence and in the presence of blockers. (C) Pooled data of basal vascular conductance in 22 and 5 rats are summarized (from Parkington et al., 2002).

vasodilatation (humans with atherosclerosis, Creager et al., 1990; rats treated with L-NMMA, Gardiner et al., 1990) confirming the pivotal role of NO in vivo. Neither eicosanoid nor EDHF appear to contribute to basal flow/conductance in a variety of vascular beds (Jackson and Blair, 1998; Nishikawa et al., 1999; Parkington et al., 2002) (Figure 13.1). The contribution of NO to agonist-induced vasodilatation in vivo is heterogeneous, depending on the vascular bed (Gardiner et al., 1991) and on the calibre of the vessel (Widmann et al., 1998; Nishikawa et al., 1999). NO contributes almost exclusively to chemically induced (usually with acetylcholine) vasodilatation in human large coronary arteries, while a substantial NO- and prostanoid-independent entity, likely EDHF, is important in coronary arteries/arterioles of small diameter (Lefroy et al., 1993). Similar observations have been reported in the coronary bed in dogs (Widmann et al., 1998; Nishikawa et al., 1999). These observations in intact animals confirm the observed greater role for EDHF in vessels of small diameter in vitro (Hwa et al., 1994; Garland et al., 1995; Shimokawa et al., 1996; Hill et al., 2001). Relatively scant attention has been paid to determining the contribution of EDHF to peripheral resistance, blood flow and tissue perfusion in vivo. Such studies had to wait until

EDHF in vascular tone in vivo

103

Figure 13.2 The effects of L-NAME and indo, iberiotoxin (IbTx, 5 ⫻ 10⫺8 mol/l), and charybdotoxin (ChTx) plus apamin on the peak increase in mesenteric vascular conductance evoked by acetylcholine, bradykinin, or sodium nitroprusside (SNP) in 5 rats are summarized (from Parkington et al., 2002).

blockers for the various endothelium-dependent vasodilators became available, which has been particularly tardy in relation to EDHF. This situation has been hindered by the lack of consensus as to the nature of EDHF, even from more easily controlled in vitro experiments. In addition, the issue has been complicated by observations in isolated vessels which suggest that NO may suppress the availability or effectiveness of EDHF, at least in some arteries (Kilpatrick and Cocks, 1994; Bauersachs et al., 1996). Thus, impaired NO bioavailability, as has been implicated in many diseases involving vascular dysfunction, might be expected to facilitate the prominence of EDHF. From these in vitro studies it has been suggested that a major role for EDHF is to compensate for the reduced impact of NO in these instances (e.g. Taddei et al., 1999). Suppression of EDHF by NO may depend on the nature of EDHF and hence may not occur in all vessels. For example, in in vivo studies of rat mesenteric and hindlimb beds, EDHF-mediated increases in blood flow are clearly present prior to administration of NO synthesis blocker, and are unchanged in amplitude in the presence of the blocker (Figures 13.1 and 13.2). An understanding of the role of EDHF in vivo in its own right is important in view of the reports of depressed EDHF hyperpolarization and relaxation in vessels isolated in disease (Félétou and Vanhoutte, 1996): in humans (Taddei et al., 1995; Knock and Poston, 1996; Ashworth et al., 1997; Pascoal et al., 1998) and in experimental animals with age (Fujii et al., 1993) and in models of hypertension (Fujii et al., 1992) and diabetes mellitus (Fukao et al., 1997c; Wigg et al., 2001). Evidence is now emerging in support of a role for EDHF in vivo (Figures 13.1 and 13.2) including in disease (Taddei et al., 1995; De Vriese et al., 2000a). 3. HYPERPOLARIZATION OF SMOOTH MUSCLE AND VASCULAR TONE A hallmark of EDHF is the hyperpolarization that occurs in the vascular smooth muscle cells and which mediates the relaxation. Membrane potential plays an important role in determining vascular tone. Pivotal in the mechanism implicated is Ca2⫹ influx through voltageoperated Ca2⫹ channels. Depolarization increases, while hyperpolarization decreases the

104

Helena C. Parkington et al.

probability of the opening of these ion channels to facilitate contraction or relaxation, respectively. The importance of this mechanism in vivo is indicated by the success in reducing vascular resistance of drugs that block these channels. Membrane potential may additionally influence vascular tone by modulating the activity of transmembrane proteins (e.g. phospholipase C, adenylyl cyclase) that are accessed by vasoactive hormones and drugs that determine contraction and relaxation of smooth muscle. In 1984, Bolton and colleagues first demonstrated that stimulation of the vascular endothelium evoked hyperpolarization and relaxation in guinea-pig mesenteric arterial smooth muscle. At that time the full complexity of endothelium-dependent vasodilatation had not been appreciated. This, and the lack of pharmacological tools, hampered identification of the agent or agents responsible for these responses. The ability of eicosanoids released from the endothelium, especially prostacyclin, to effect vasodilation had been established (Moncada et al., 1976; Singer et al., 1984) and the substantial role of NO was recognized subsequently (Furchgott, 1988; Ignarro et al., 1988). However, whether or not vascular smooth muscle hyperpolarization was amongst the mechanisms by which these autacoids caused vasodilatation was unknown. Endothelium-derived NO can elicit hyperpolarization of the smooth muscle in some vessels, as initially shown by Tare et al. (1990). This conclusion was made possible only following the development of blockers of NO synthase. Prostacyclin was later found to evoke similar hyperpolarization (Parkington et al., 1993). However, it became apparent that stimulation of the endothelium in many vascular beds was still capable of eliciting hyperpolarization of vascular smooth muscle when NO and prostaglandin synthesis had been blocked. The name EDHF was coined to describe this event (Taylor and Weston, 1988), since the nature of the causative agent was unknown. That this hyperpolarization was followed by substantial relaxation of vascular smooth muscle spawned an intense experimental effort. Although its identity remains controversial, it is emerging that EDHF is unlikely to be attributable to one single “factor” but may be a different entity depending on the vascular bed and on the other prevailing relaxing factors. 4. PRODUCTS OF CYTOCHROME P450 The possibility that EDHF might be a product(s) of the cytochrome P450 (CYP) enzyme, the epoxyeicosatrienoic acids (EETs), was initially suggested as a result of effective blockade by the imidazole antimycotic drugs (Harder et al., 1994; Hecker et al., 1994). However, these agents were later found to block the Ca2⫹-activated K⫹ channels responsible for the hyperpolarization mediating EDHF (Alvarez et al., 1992; Vanheel and van de Voorde, 1997), thus limiting their usefulness in the identification of EDHF. Nonetheless, powerful support for the initial proposal of EETs as EDHF, at least in some arteries, was provided when agents that induce CYP augmented, and antisense oligonucleotides against the enzyme blocked, EDHF responses in porcine coronary artery (Fisslthaler et al., 1999). EET-induced hyperpolarization is blocked by iberiotoxin, a selective blocker of large-conductance Ca2⫹-activated K⫹ channels (Campbell et al., 1996; Baron et al., 1997; Eckman et al., 1998). Complicating the situation is a report of in vivo blockade of NO vasodilation by iberiotoxin in porcine arteries (Zanzinger et al., 1996). The suggestion that this EDHF is released from the endothelial cells to act on the subjacent smooth muscle is not entirely without question. EET-attributable EDHF responses could not be detected in endothelium-denuded recipient bovine or porcine coronary tissues even when the endothelium of an intact donor bovine or porcine coronary arteries was in physical contact (Hecker et al., 1994), an observation which has to stand alongside the

EDHF in vascular tone in vivo

105

established stability of EETs in physiological solutions. The same investigators subsequently reported that the effluent resulting from perfusion of donor porcine coronaries through the lumen was capable of hyperpolarizing cultured smooth muscle cell recipients (Popp et al., 1996). Transferance of EET-attributed EDHF has been shown in rat mesenteric artery (Chen and Cheung, 1996). The most recent development of agents that selectively block the membrane receptors for EETs (Gauthier et al., 2002) will undoubtedly progress the understanding of the role of EETs in determining vascular tone. The in vivo responses attributed to EDHF evoked in dog coronary arteries (Widmann et al., 1998; Nishikawa et al., 1999) and in mouse (Hungerford et al., 2000) and hamster (de Wit et al., 1999) skeletal muscle arterioles are mediated by large-conductance Ca2⫹-activated K⫹ channels, and are reduced by inhibitors of CYP, implicating EETs. EDHF and NO appeared to make similar contributions (about 50% each) to acetylcholine-induced dilatation of hamster skeletal muscle arterioles (de Wit et al., 1999) and in dog small coronaries (Widmann et al., 1998; Nishikawa et al., 1999). Either EDHF or NO on their own were capable of eliciting maximal dilation in response to acetylcholine in vivo. Despite the evidence for a role of an EET in EDHF evoked by endothelial stimulation in vivo, basal release of EETs in vivo, and a contribution to mean arterial pressure is lacking (Jackson and Blair, 1998; Nishikawa et al., 1999). This is contrary to expectations from observations in vitro (see Section 7). 5. POTASSIUM RELEASE FROM ENDOTHELIAL CELLS In many vascular beds EDHF is insensitive to iberiotoxin, which blocks the EET-attributable EDHF, and to agents that suppress CYP activity. Most commonly observed in vitro, EDHF is abolished by a cocktail of the toxins charybdotoxin plus apamin, which block intermediateand small-conductance, Ca2⫹-activated K⫹ channels, respectively. These ion channels occur predominantly on the endothelial rather than on the vascular smooth muscle cells (Zygmunt et al., 1996; Plane et al., 1997; Vanheel and Van de Voorde, 1997; Eckman et al., 1998; Edwards et al., 1998; Coleman et al., 2001a). The K⫹ exiting the endothelial cells through these channels may act to open inwardly rectifying K⫹ channels, which are blocked by low concentrations of barium, and to activate Na⫹-K⫹ ATPase pumps, blocked by ouabain, in the cell membrane of the smooth muscle and/or endothelial cells, giving rise to EDHFattributed hyperpolarization and relaxation (Edwards et al., 1998; Bény and Schaad, 2000). While the EDHF in a variety of isolated arterial preparations, including some human vessels, is resistant to the effects of barium and ouabain (Quignard et al., 1999; Vanheel and van de Voorde, 1999; Buus et al., 2000; Drummond et al., 2000; Lacy et al., 2000; Coleman et al., 2001a,b), cerebral arteries appear to be exquisitely sensitive to and relax when extracellular K⫹ is elevated (Knot et al., 1996). However, the principal source of the K⫹ in this instance is likely to be the nervous tissue of the brain rather than the endothelium. So far, this K⫹ hypothesis of EDHF has received limited testing in vivo using definitive blockade with barium and ouabain. The increase in blood flow evoked by acetylcholine (Taddei et al., 1995) or bradykinin (Taddei et al., 1999) in the forearm of normotensive individuals was not affected by ouabain but was mediated largely by NO. In contrast, ouabain markedly reduced the response to bradykinin, but not to acetylcholine, in hypertensive individuals. An involvement of the charybdotoxin- plus apamin-sensitive EDHF in basal flow/ conductance could not be demonstrated in the mesenteric and hindlimb beds or on mean arterial pressure in healthy normotensive rats (Parkington et al. 2002) (Figures 13.1

106

Helena C. Parkington et al.

and 13.2). However, a significant, if small, ouabain-sensitive contribution to basal flow has been suggested in the forearm bed of both normotensive and hypertensive humans (Taddei et al., 1995; Taddei et al., 1999). 6. ELECTROTONIC SPREAD OF HYPERPOLARIZATION FROM ENDOTHELIAL TO SMOOTH MUSCLE CELLS An alternative view suggests that the charybdotoxin- plus apamin-sensitive hyperpolarization arising in the endothelial cells spreads to the underlying smooth muscle via gap junctions to give rise to EDHF. The difficulty in observing reliable transfer of hyperpolarization via extracellular pathways leaves open the possibility of such direct intercellular spread of the hyperpolarization between the two cell types. In rat mesenteric artery, amongst the most thoroughly studied vessels in vitro, an EET-independent EDHF contributes prominently to endothelium-dependent vasodilatation; on its own it can achieve complete relaxation and is sensitive to the cocktail of charybdotoxin plus apamin (Zygmunt et al., 1995; Plane et al., 1997; Edwards et al., 1998; Doughty et al., 1999; Wigg et al., 2001). The possibility of direct intercellular spread of current has substantial support in the literature in terms of functional evidence, that is, direct spread of electrical events (von der Weid and Bény, 1993; Edwards et al., 1998; Yamamoto et al., 1998; Coleman et al., 2001a,b; Sandow et al., 2002), and also in the confirmed existence of pentalaminar myoendothelial gap junctions (Sandow and Hill, 2000; Sandow et al., 2002). Electrophysiological and electronmicroscopical studies show that the charybdotoxin- and apamin-sensitive EDHF in rat arteries is likely to spread via myoendothelial gap junctions; where these junctions occur EDHF is observed and where there are no myoendothelial gap junctions there is no EDHF (Sandow et al., 2002). Furthermore, peptide inhibitors of gap junctions all but abolished EDHF in the smooth muscle. Direct support for the involvement of myoendothelial gap junctions in EDHF in vivo has been presented using the peptide gap junction inhibitors in the rat renal bed (De Vriese et al., 2002). The study showed that these peptides not only block acetylcholine-induced EDHFattributed increases in blood flow but they also decreased basal blood flow in that bed. 7. FLOW, SHEAR AND THE CO-ORDINATION OF VASORELAXATION Optimal perfusion dictates that the tissue requirement for blood be communicated in a co-ordinated manner such that vasodilatation occurs at all levels of the vascular bed. Flow, shear stress and the propagation of dilation originating at focal points of need are candidates for such co-ordination. Flow/shear stress induces release of NO (Rubanyi et al., 1986), eicosanoid (Koller et al., 1993) and EDHF (Popp et al., 1998) in isolated arterial preparations. Similarly, an increase in flow/shear stress in vivo evokes the release of different autacoids from the endothelium of different vascular beds. Vasodilatation has been attributed to NO in rat spinotrapezius (Friebel et al., 1995), canine coronary artery (Recchia et al., 1996), hamster cremaster arterioles (de Wit et al., 1997), rat mesenteric and hindlimb beds (Parkington et al., 2002), or prostanoid in rat cremaster (Koller and Kaley, 1990). Blockade of EDHF with charybdotoxin plus apamin had no effect on basal flow/conductance in rat mesenteric and hindlimb vascular beds in vivo, which might be interpreted in terms of a lack of EDHF in flow-mediated vasodilatation in these beds (Parkington et al., 2002). However, in the rat renal bed, peptide blockers of gap junctions reduced basal flow (De Vriese et al., 2002) and ouabain reduced basal flow in human forearm (Taddei et al., 1999).

EDHF in vascular tone in vivo

107

The mechanism by which point-initiated relaxation of smooth muscle might propagate has been investigated in vivo in the feed arteries and arterioles of the hamster cheek pouch. In this bed the hyperpolarization and relaxation were both markedly reduced by miconazole or 17-octadecynoic acids implicating an EET in this EDHF-mediated response (Welsh and Segal, 2000b). Relaxation similarly spreads from the focus of stimulation in the arterioles of cremaster muscles in normal mice. In animals in which the gene for the gap junction protein connexin 40 was deleted, propagation of vasodilatation was restricted markedly (de Wit et al., 2000). Connexin 40 forms the gap junctions between endothelial cells, while connexin 43 links vascular smooth muscle cells. Thus, the spread of hyperpolarization via electrotonic spread of hyperpolarization between endothelial cells in both directions away from the focus of stimulation may provide a mechanism of co-ordinating relaxation throughout the branches of the vascular bed. 8. CONCLUSION Studies of endothelium-dependent relaxation in vitro are essential when attempting to identify the factors involved, to elucidate detailed mechanisms of action and to eliminate complex reflexes that would occur in vivo and confound interpretation of results. However, it is important to then move on and to test the hypotheses thus formed in the complex milieu that is the intact animal. Even then, the presence of anaesthetic agents can present a barrier to the complete final clarification of events. These issues are important, not only for an understanding of the physiological role of the endothelium, but importantly, in coming to grips with events in disease. Genetic manipulations teach that deletions and additions, and the resulting long-term changes in elements of physiological control can lead to compensatory up or down regulation of other unaffected elements. Such compensations may also occur in chronic diseases (e.g. diabetes mellitus, atherosclerosis) and in vivo approaches are important experimental strategies to explore such changes. ACKNOWLEDGEMENTS This work was supported by Monash University Small Grants, the National Heart Foundation of Australia and the National Health and Medical Research Council (Australia).

14 Endothelium-derived hyperpolarizing factor, myoendothelial gap junctions and hypertension Shaun L. Sandow, Narelle J. Bramich, Hari Priya Bandi, Nicole M. Rummery and Caryl E. Hill Electrical coupling via myoendothelial gap junctions has been implicated in the activity of endothelium-derived hyperpolarizing factor (EDHF). The number of layers of smooth muscle cells in the media may thus exert a limit on the efficacy of EDHF in different vessels. The present study has correlated the incidence of myoendothelial gap junctions and EDHF with the number of smooth muscle cell layers in arteries from normotensive and spontaneously hypertensive rats. Structural characteristics were quantified using serial section electron microscopy and immunohistochemistry. Functional characteristics were determined using electrophysiology and tension myography. Data showed that the incidence of myoendothelial gap junctions and EDHF was inversely correlated with the number of smooth muscle cell layers in the media of vessels from normotensive rats. In hypertensive rats, hypertrophic and eutrophic remodeling were found. The number of myoendothelial gap junctions was significantly greater, as was the number of smooth muscle cell layers, in the caudal artery of hypertensive rats, compared to normotensive rats. Acetylcholine-induced hyperpolarization and relaxation occurred in this vessel in both hypertensive and normotensive rats and these were abolished by apamin and charybdotoxin. In contrast, no myoendothelial gap junctions were found in the femoral artery of hypertensive rats, although the number of smooth muscle cell layers was decreased compared to normotensive animals. These findings suggest that EDHF activity in normotensive rats is dependent on the presence of myoendothelial gap junctions and inversely related to the number of smooth muscle cell layers in the media. In hypertension, however, these relationships may be disrupted.

1. INTRODUCTION The coordination of vasomotor activity requires both the homocellular coupling, as well as in many cases, the heterocellular coupling, of smooth muscle and endothelial cells. This coupling occurs via gap junctions which enable the transfer of electrical signals and/or small (⬍1 kD) signaling molecules between each respective cell type (Beny, 1999; Welsh and Nelson, 2000; Hill et al., 2001). Differences in the incidence and characteristics of gap junctions and their constituent connexin proteins reflect functional diversity within the vasculature under both normal and pathological conditions (Christ et al., 1996; Severs, 1999; Welsh and Nelson, 2000). Gap junctions connecting smooth muscle and endothelial cells, the myoendothelial gap junctions, play a role in this diversity and evidence suggests that they are important for the activity of endotheium-derived hyperpolarizing factor (EDHF), one of three vasodilator factors derived from the endothelium (Beny, 1999; Hill et al., 2001; McGuire et al., 2001; Busse et al., 2002; Campbell and Gauthier, 2002). The role of EDHF as a vasodilatory substance appears to be more important in smaller vessels (Beny, 1999; Hill et al., 2001; McGuire et al., 2001), although EDHF responses have also been recorded in a number of larger arteries

EDHF, myoendothelial gap junctions and hypertension

109

(Chaytor et al., 2001; Griffith et al., 2002). To this end, it is considered that, if myoendothelial gap junctions mediate EDHF activity, then the increased number of smooth muscle cell layers in larger vessels places a limit on the ability of the endothelium to electrically drive the membrane potential of the smooth muscle (Beny, 1999). In disease, such as hypertension, the normal cellular relationships that are present in the arterial wall are altered (Mulvany et al., 1996; Intengan and Schiffrin, 2000). Whilst hypertension is characterized by an increase in arterial blood pressure, changes in the media are heterogeneous amongst different vessels and different models of hypertension. Thus, increases may occur in the number of smooth muscle cell layers and the cross sectional area of the media in some vessels, while in others, changes may occur due to the rearrangement of the medial smooth muscle cells (Mulvany et al., 1996; Intengan and Schiffrin, 2000). Changes in the expression of arterial connexins, the protein constituents of gap junctions, have also been demonstrated in hypertensive animals. For example, connexin43 has been reported to increase in the media of the aorta of two kidney, one clip and deoxycorticosterone acetate (DOCA) hypertensive rats (Haefliger et al., 1997), while the expression of connexin40 has been reported to be reduced in the endothelium of the caudal artery of the spontaneously hypertensive rat (SHR; Rummery et al., 2002). However, it has yet to be determined whether cellular coupling between the two cell layers through myoendothelial gap junctions is altered in hypertension. Endothelial dysfunction is common in vascular disease and is characterized by alterations in the production and activity of nitric oxide, prostaglandins and EDHF (Vanhoutte, 1996). Studies examining EDHF under hypertensive conditions have been limited to the mesenteric arteries of spontaneously hypertensive rat (SHR) (Fujii et al., 1992) and stroke-prone SHR (Sunano et al., 1999), where EDHF-mediated responses are reduced. Increases in the number of smooth muscle cell layers in the media of vessels under hypertensive conditions could contribute to the reduced ability of an endothelium-derived factor, such as EDHF, to hyperpolarize and relax the smooth muscle. To date no studies have correlated the activity of EDHF in hypertension with the incidence of myoendothelial gap junctions and the number of smooth muscle cell layers. The present study has tested the hypothesis that the incidence of myoendothelial gap junctions and EDHF is inversely correlated with the number of smooth muscle cell layers in different vascular beds. Included in the study are the femoral artery, in which EDHF-mediated relaxation is not present (Wigg et al., 2001), and the tertiary branches of the mesenteric resistance arteries in which relaxation to acetylcholine is predominantly due to EDHF (Hill et al., 2000). The corollary of this hypothesis is that, in hypertensive vessels, in which the number of medial smooth muscle cell layers is altered, the incidence of myoendothelial gap junctions may show corresponding changes in line with any changed importance of EDHF as a vasodilatory factor. 2. METHODS

2.1. Animals and tissues For all studies, rats were deeply anaesthetized with intraperitoneal ketamine and rompun (44 and 8 mg/kg, respectively). In the comparative studies of vessels from normotensive rats, segments of the femoral artery, distal caudal artery, primary mesenteric arteries and tertiary mesenteric arteries supplying the ileum were removed. For the studies of the impact of hypertension on structure and function, 12-week-old inbred SHR rats were chosen and studies conducted on the femoral and distal caudal arteries.

110

Shaun L. Sandow et al.

Previous studies have shown that the maximal increase in blood pressure was attained by this age (Rummery et al., 2002).

2.2. Electron microscopy, myoendothelial gap junctions and arterial morphology Rats were perfused with 3% glutaraldehyde and 1% paraformaldehyde (ProSciTech, Thuringowa, Qld., Australia) in 10⫺7 M sodium cacodylate (ProSciTech, Thuringowa, Qld., Australia) with 2 ⫻ 10⫺3 M CaCl2 . 6H2O, 10⫺4 M sucrose, 10⫺2 M betaine, pH 7.35 at 25 ⬚C. Sodium nitrite (1.5 ⫻ 10⫺3 M) was present in the pre-perfusion clearance solution to induce maximal vasodilation. Small segments of arteries were processed for electron microscopy using standard procedures (Sandow et al., 2002). Arterial morphology was determined from digitized electron microscopic montages (⫻2,500) of individual transverse arterial sections from different rats. The vessel circumference and diameter were determined at the level of the internal elastic lamina. The number of medial smooth muscle cell layers was determined by averaging the number of smooth muscle cell profiles ⱖ 5 ␮m in length, from the outer edge of the internal elastic lamina to the inner edge of the external elastic lamina along four linear plots 90⬚ apart. Serial, dark silver to gold transverse arterial sections were collected over a 5 ␮m length of each vessel and the sections placed on Formvar and carbon coated slot grids. All myoendothelial gap junctions were identified and photographed through each series of sections at ⫻ 20,000 to ⫻ 40,000 (Sandow et al., 2002). The circumference for each vessel was used to calculate an intimal surface area and the number of myoendothelial gap junctions was determined in an intimal area of 1000 ␮m2.

2.3. Immunohistochemistry and endothelial cell morphology The surface area of endothelial cells was determined using immunohistochemistry and affinity purified antibodies raised in sheep against amino acids 254 to 270 of rat connexin40 (Institute for Medical and Veterinary Science, Adelaide, SA., Australia and Mimotopes, Clayton, Vic., Australia). Arteries were removed from perfusion fixed animals (2% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4) and whole mounts of the vessels were incubated with connexin40 antibodies (1:100) for 2 h at 37 ⬚C. Labeling was detected following incubation in Cy3-conjugated donkey anti-goat immunoglobulins (1:100, Jackson ImmunoResearch) for 1 h at 25 ⬚C. Preparations were mounted in buffered glycerol and photographed using a Nikon Coolpix 950 digital camera on a compound microscope at a resolution of 1600 ⫻ 1200 pixels. Morphometric measurements of randomly selected endothelial cells were made using MCID software (Imaging Research, Canada).

2.4. Electrophysiology and EDHF-mediated hyperpolarization Intracellular recordings were made from caudal artery segments pinned flat in a shallow recording chamber, with preparations being continuously superfused with physiological saline, gassed with 95% O2, 5% CO2. Membrane potential recordings were made using conventional techniques with fine glass microelectrodes (resistance: 120–240 M⍀) filled with 0.5 M KCl (Hill et al., 1999). Successful impalements of cells were characterized by a sharp decrease in potential of 20–40 mV, followed by a progressively slower decrease to a resting value of around ⫺60 mV. Intracellular recordings were made by advancing electrodes from the adventitial surface. Dye filled electrodes were used to confirm the nature of the recorded cells. Dyes used were 10% fluorescein-conjugated dextran (FITC-dextran, 3000 MW; Molecular Probes, Eugene,

EDHF, myoendothelial gap junctions and hypertension

111

OR, USA), or 3% lucifer yellow (Sigma-Aldrich, Sydney, Australia), backfilled with 0.5 M KCl or LiCl. EDHF-mediated hyperpolarization was recorded after administration of acetylcholine, following incubation of the preparations for at least 30 min in the presence of the nitric oxide synthase inhibitor, N␻-nitro-L-arginine methyl ester hydrochloride (L-NAME; 10⫺4 M) and the cyclooxygenase inhibitor, indomethacin (10⫺5 M), to block responses due to nitric oxide and prostaglandins, respectively. In order to confirm the involvement of EDHF, the effect of acetylcholine was tested in the presence of charybdotoxin (6 ⫻ 10⫺8 M) and apamin (5 ⫻ 10⫺7 M), blockers of large/intermediate and small conductance calcium activated potassium channels, respectively, and in preparations in which the endothelium had been removed.

2.5. Tension recordings and EDHF-mediated relaxation EDHF-mediated relaxation was assessed using a Mulvany–Halpern style myograph. Arterial segments, 1 mm in length, were mounted between two 40 ␮m wires and allowed to equilibrate for 30 min prior to incremental stretching to an equivalent tension of 80 mmHg. Preparations, continuously superfused with physiological saline, gassed with 95% O2, 5% CO2, were constricted with phenylephrine (2.5 ⫻ 10⫺6 to 4 ⫻ 10⫺6 M) to approximately 60% of maximal constriction to phenylephrine. Tissues were equilibrated for 15 min prior to subsequent addition of drugs. The integrity of the endothelium of each preparation was assessed by an initial application of acetylcholine (10⫺6 M), with preparations showing a relaxation of ⬍ 40% being discarded. EDHF-mediated relaxation was assessed following the application of acetylcholine in the presence of L-NAME (10⫺4 M) and indomethacin (10⫺5 M), which were present for 30 min prior to acetylcholine. Tension was recorded using a MacLab chart recorder, with results being expressed as % relaxation of phenylephrine-induced constriction. The effects of charybdotoxin (6 ⫻ 10⫺8 M) and apamin (5 ⫻ 10⫺7 M) on the relaxation to acetylcholine were examined in the presence of L-NAME and indomethacin.

2.6. Drugs Acetylcholine, apamin, betaine, charybdotoxin and indomethacin were from Sigma-Aldrich (Sydney, Australia); ketamine from Parnell Laboratories (Alexandria, NSW, Australia), L-NAME from Sapphire Bioscience (Crows Nest, NSW, Australia), phenylephrine from RBI (Natick, MA, USA), rompun from Bayer (Pymble, NSW, Australia) and sodium nitrite from BDH Chemicals (Poole, UK).

2.7. Statistics Results are expressed as mean ⫾ standard error of the mean (SEM). Statistical significance was tested using a paired or unpaired t-test. A P value of less than 0.05 was considered to be statistically significant. Graph presentation and regression analysis was performed using GraphPad Prism. 3. RESULTS

3.1. Arterial morphology and incidence of myoendothelial gap junctions in arteries of normotensive rats The caudal, femoral, primary and tertiary mesenteric arteries were investigated as they represented vessels of decreasing diameter (Table 14.1). In general the number of smooth muscle cell layers in the media decreased with decreasing arterial diameter although this

112

Shaun L. Sandow et al.

Table 14.1 Morphological characteristics of rat arteries from normotensive rats Vessel characteristics

Femoral

Caudal

1⬚ Branch mesenteric

1⬚ Branch mesenteric

3⬚ Branch mesenteric

Age Vessel diameter (␮m) No. of SMC layers Area sampled at IEL (␮m2) No. of MEGJs per 1000 ␮m2 of intimal surface

16 wk 306 ⫾ 12 (3)

12 wk 346 ⫾ 16 (3)

16 wk 206 ⫾ 18 (3)

5 wk 159 ⫾ 21 (3)

5 wk 115 ⫾ 6 (3)

9.8 ⫾ 0.62 (3)

7.0 ⫾ 0.30 (3)

4.8 ⫾ 0.27 (3)

5.1 ⫾ 0.74 (3)

3.7 ⫾ 0.18 (3)

4810 ⫾ 191 (3)

5442 ⫾ 358 (3)

3240 ⫾ 287 (3)

2493 ⫾ 325 (3)

1813 ⫾ 105 (3)

0 ⫾ 0 (3)

0.92 ⫾ 0.06 (3)

2.25 ⫾ 0.4 (3)

2.91 ⫾ 0.42 (3)

7.43 ⫾ 1.36 (3)

Note Numbers in parentheses represent the numbers of preparations, each from a different animal. SMC, smooth muscle cell; IEL, internal elastic lamina; MEGJ, myoendothelial gap junction.

was not absolute (Table 14.1). In each case the vessels were maximally dilated. The surface area of the endothelial cells was measured in the femoral, caudal and primary mesenteric arteries and found not to be significantly different amongst the three vessels (P ⬎ 0.05, 363 ⫾ 6, n ⫽ 154 cells from 4 rats for femoral artery; 384 ⫾ 16, n ⫽ 83 cells from 4 rats for caudal artery, 351 ⫾ 5, n ⫽ 182 cells from 4 rats for primary mesenteric artery; Rummery et al., 2002; Sandow et al., 2002). Myoendothelial gap junctions were found in the caudal, primary and tertiary mesenteric arteries, but not in the femoral artery (Table 14.1; Sandow et al., 2002). There was no significant difference in the incidence of myoendothelial gap junctions between five and sixteen weeks of age, in the case of the primary mesenteric arteries. In the mesenteric arteries, both primary and tertiary, all myoendothelial gap junctions were present between the ends of thin projections of endothelial cells which penetrated the internal elastic lamina to contact the surface of smooth muscle cells. In the caudal artery, 40% of myoendothelial gap junctions were of this type (Figure 14.1) while the remainder occurred within the internal elastic lamina between projections from both endothelial cells and smooth muscle cells. When the number of myoendothelial gap junctions in each vessel was measured in a constant area of intimal surface, it was found that the incidence of these junctions was inversely related to the diameter of the arteries and to the number of smooth muscle cell layers in the media (Table 14.1, Figure 14.4).

3.2. Arterial morphology and incidence of myoendothelial gap junctions in arteries from hypertensive rats There were significantly more layers of smooth muscle cells in the caudal artery of hypertensive than of normotensive rats (Tables 14.1, 14.2, P ⬍ 0.05). On the other hand, the number of smooth muscle cell layers in the media of the femoral artery of hypertensive rats was decreased relative to that in normotensive rats (Tables 14.1, 14.2). In both arteries the vessel diameter was significantly less in the hypertensive than in the normotensive rat (Tables 14.1, 14.2, P ⬍ 0.05). The number of myoendothelial gap junctions was significantly greater in the caudal artery of SHR compared to the caudal artery from normotensive rats (P ⬍ 0.05). In contrast,

Figure 14.1 Myoendothelial gap junction in the caudal artery of the rat. Myoendothelial gap junctions were found at the end of projections from endothelial cells (ec). Pentalaminar regions of contacts between endothelial and smooth muscle cells (smc) characteristic of gap junctions (inset) are present at discrete points between the two cell contacts (arrowed). iel, internal elastic lamina.

Figure 14.2 Morphology of the endothelial (A,B) and smooth muscle cells (C,D,E,F) of the caudal artery from normotensive (A,C,D) and hypertensive (B,E,F) rats. Endothelial cell borders were highlighted using immunohistochemistry and antibodies to connexin40. Smooth muscle cells were filled with either FITC-dextran (C,E) or lucifer yellow (D,F). Arrows (D,F) indicate the impaled smooth muscle cell from which dye was found to move in a radial, but not longitudinal direction. Longitudinal vessel axis is left to right for all panels (see Color Plate 9).

114

Shaun L. Sandow et al. Table 14.2 Morphological characteristics of rat arteries from hypertensive rats Vessel characteristics

Femoral

Caudal

Age Vessel diameter (␮m) No. of SMC layers Area sampled at IEL (␮m2) No. of MEGJs per 1000 ␮m2 of intimal surface

16 wk 199 ⫾ 21 (2) 8.2 ⫾ 0.70 (2) 3130 ⫾ 325 (2) 0 ⫾ 0 (2)

12 wk 282 ⫾ 6 (3) 8.8 ⫾ 0.50 (3) 4443 ⫾ 124 (3) 1.51 ⫾ 0.13 (3)

Note Number in parenthesis represents the number of preparations, each from a different animal. SMC, smooth muscle cell; IEL, internal elastic lamina; MEGJ, myoendothelial gap junction.

myoendothelial gap junctions were not found in the femoral arteries of the hypertensive rat. When endothelial cells were measured in the caudal artery of the hypertensive rat, they were found to be significantly smaller in area than those from normotensive rats (265 ⫾ 11, 87 cells from 4 rats, Figure 14.2(A) normotensive, Figure 14.2(B), hypertensive; Rummery et al., 2002). Even taking this reduction in area into account, there were still significantly more myoendothelial gap junctions per endothelial cell in the caudal artery of the hypertensive than of the normotensive rats. The two morphological types of myoendothelial gap junctions, which were found in the caudal artery of the normotensive rats, were also found in similar proportions in the hypertensive rats, that is, they were found between projections of endothelial cells and smooth muscle cells, as well as between the projections of endothelial cells and the surface of smooth muscle cells.

3.3. EDHF-activity in caudal arteries of hypertensive rats All intracellular recordings in arteries from both normotensive and hypertensive rats were made from smooth muscle cells. Cells were identified through the use of electrodes filled with either FITC-conjugated dextran, which was not observed to spread between adjacent cells (Figure 14.2(C) normotensive, Figure 14.2(E), hypertensive), or lucifer yellow, which was commonly seen to spread amongst adjacent smooth muscle cells in a radial, but not longitudinal, direction (Figure 14.2(D), normotensive, Figure 14.2(F), hypertensive). The resting membrane potential of smooth muscle cells did not differ significantly between normotensive and hypertensive rats (P⬎0.05; ⫺59⫾1mV, n⫽23 from 7 normotensive rats and ⫺59⫾1mV, n ⫽ 19, from 7 hypertensive rats). Addition of acetylcholine hyperpolarized the cell membrane of smooth muscle cells in caudal arteries from both normotensive and hypertensive rats (Figure 14.3). These hyperpolarizations occurred in the presence of L-NAME (10⫺4 M) and indomethacin (10⫺5 M). L-NAME did not affect resting membrane potential in either strain, although indomethacin caused a 5–10 mV depolarization. In both normotensive and hypertensive rats, charybdotoxin (6 ⫻ 10⫺8 M) and apamin (5 ⫻ 10⫺7 M; n ⫽ 8, for each) prevented the hyperpolarizations following application of acetylcholine. Removal of the endothelium from the arteries also prevented the acetylcholine-induced hyperpolarization (n ⫽ 3).

EDHF, myoendothelial gap junctions and hypertension

115

Figure 14.3 EDHF-mediated hyperpolarization in caudal arteries of normotensive and hypertensive rats elicited in response to the application of acetylcholine (ACh, 3⫻10⫺6 M). Experiments were undertaken in the presence of L-NAME (10⫺4 M) and indomethacin (10⫺5 M) to block the effects of nitric oxide and prostaglandins, respectively (from Sandow et al., 2003. Arterio. Thromb. Vasc. Biol., 23: 822–828).

Figure 14.4 Relationship between myoendothelial gap junctions and arterial morphology. The number of myoendothelial gap junctions per 1000␮m2 of intimal surface was inversely related to the number of smooth muscle layers in the media (A) and to arterial diameter (B). Data was fitted with a one phase exponential curve where the R2 value was 0.99 in A and 0.82 in B.

Acetylcholine produced a concentration-dependent relaxation in caudal arteries from both normotensive and hypertensive rats in the presence of L-NAME and indomethacin. Like the effect on the hyperpolarization, the combination of charybdotoxin and apamin abolished the EDHF relaxation in both normotensive and hypertensive rats (n ⫽ 4, for each).

4. DISCUSSION The present study has shown that the incidence of myoendothelial gap junctions is inversely correlated with the number of layers of smooth muscle cells in the media of several arteries of normotensive rats. In the vessels chosen, a positive correlation exists between the

116

Shaun L. Sandow et al.

incidence of myoendothelial gap junctions and the importance of EDHF as a vasodilator. Thus, there are more myoendothelial gap junctions in the distal segments of the mesenteric vascular bed and EDHF plays a greater role in vasodilation in these vessels than in the proximal segments (Shimokawa et al., 1996; Hill et al., 2000; Sandow and Hill, 2000). On the other hand, in the femoral artery of the rat, which lacks EDHF, myoendothelial gap junctions are absent (Sandow et al., 2002). These data support the hypothesis that electrical coupling through myoendothelial gap junctions mediates the action of EDHF. The role of EDHF is limited in larger arteries due to the increased number of smooth muscle cell layers, which might act as an electrical sink, and the decreased incidence of myoendothelial gap junctions. In the caudal artery of spontaneously hypertensive rats, hypertrophic remodeling has been found with the number of smooth muscle cells layers in the media being significantly increased. In contrast, in the femoral artery of the same hypertensive strain, data from the present study demonstrates that changes are more consistent with eutrophic or hypotrophic remodeling. These results support the concept that the type of remodeling associated with hypertension may vary according to the model examined and the vascular bed under investigation (Mulvany et al., 1996; Intengan and Schiffrin, 2000). In spite of the increased number of smooth muscle cell layers in the caudal artery of the hypertensive rats, serial section electron microscopy showed that the incidence of myoendothelial gap junctions was significantly greater than in the same arteries of normotensive animals. In these vessels, application of acetylcholine caused a hyperpolarization and relaxation as it did in the normotensive vessels. That the hyperpolarizations and relaxations in both normotensive and hypertensive vessels were due to EDHF was confirmed by demonstrating their abolition after application of charybdotoxin and apamin, universally accepted blockers of EDHF-mediated responses (Hill et al., 2001; McGuire et al., 2001; Campbell and Gauthier, 2002), or by removal of the endothelium. Thus, the increased heterocellular coupling via myoendothelial gap junctions found in the hypertensive rats may have served to offset the increased number of smooth muscle cell layers and thereby maintain EDHF-mediated hyperpolarization and relaxation. A different situation was found in the femoral artery of hypertensive rats. In spite of a reduction in the number of smooth muscle layers, no myoendothelial gap junctions were found. Preliminary studies from our laboratory suggest that in these vessels, acetylcholine does not cause hyperpolarization or relaxation of the smooth muscle cells. Taken together with the results in the caudal artery, the data suggest that in pathological states the simple relationship between the presence of myoendothelial gap junctions, the role of EDHF and the number of layers of smooth muscle in the media may break down. Thus, the role of heterocellular coupling and EDHF is heterogenous within vascular beds and in disease states. ACKNOWLEDGEMENTS We thank Dr Marianne Tare for invaluable advice regarding the use of the myograph, the National Heart Foundation and National Health and Medical Research Council of Australia for support.

15 Improvement of age-related impairment of endothelium-dependent hyperpolarization by renin-angiotensin system blockade Yasuo Kansui, Koji Fujii, Kenichi Goto and Mitsuo Iida The hyperpolarization and relaxation mediated by endothelium-derived hyperpolarizing factor (EDHF) are impaired with aging. The present study tested whether inhibition of the renin-angiotensin system would improve age-related impairment of the EDHF-mediated hyperpolarization and relaxation. Normotensive Wistar Kyoto rats (WKY) were treated for 3 months with the converting enzyme inhibitor enalapril, the AT1 receptor antagonist candesartan, or a combination of hydralazine and hydrochlorothiazide from 9 to 12 months of age. The three treatments lowered systolic blood pressure comparably. EDHF-mediated hyperpolarization to acetylcholine in mesenteric arteries was improved similarly in both enalapril-treated WKY and candesartan-treated WKY, but not in WKY treated with the combination of hydralazine and hydrochlorothiazide, compared with age-matched WKY, and the responses obtained were similar to those in 3-month-old young WKY. EDHF-mediated relaxation was also similarly improved in enalapril-treated WKY and candesartan-treated WKY, but not in WKY treated with the combination of hydralazine and hydrochlorothiazide, compared with age-matched WKY. These findings suggest that the AT1 receptor antagonist and the converting enzyme inhibitor similarly improve age-related impairment of the EDHF-mediated hyperpolarization and relaxation in normotensive rats, presumably through inhibition of the renin-angiotensin system itself but not due to a blood pressure lowering. The present findings raise the possibility that the inhibitors of the renin-angiotensin system might serve as novel tools with which to prevent the endothelial dysfunction associated with aging.

1. INTRODUCTION Endothelium-derived hyperpolarizing factor (EDHF) may play an important role in the control of vascular tone both in animals and humans (Félétou and Vanhoutte, 1988; Chen et al., 1988; Chen and Suzuki, 1989; Cohen and Vanhoutte, 1995; Urakami-Harasawa et al., 1997), although its exact nature is still controversial (Campbell et al., 1996; Chaytor et al., 1998; Edwards et al., 1998; Fisslthaler et al., 1999; Yamamoto et al., 1999). EDHF-mediated hyperpolarization and relaxation are impaired with advancing age as well as by hypertension (Fujii et al., 1992, 1993; Nakashima and Vanhoutte, 1996). Endothelial dysfunction is now recognized as an aggravating factor for atherosclerosis (Vanhoutte, 1997), and in this context it may be important to prevent or reverse endothelial dysfunction. Antihypertensive treatments improved endothelial dysfunction in hypertension (Tschudi et al., 1994; Onaka et al., 1998; Goto et al., 2000a). Furthermore, inhibitors of the renin-angiotensin system tended to be more effective in improving endothelial function compared with other classes of antihypertensive agents despite achieving similar reductions in blood pressure (Schiffrin and Deng, 1995; Onaka et al., 1998; Schiffrin et al., 2000), implying that the beneficial effects of the renin-angiotensin system blockade may partially

118

Yasuo Kansui et al.

be independent of its blood pressure lowering effects. These findings led to hypothesize that the renin-angiotensin system blockade would also improve age-related impairment of EDHF-mediated hyperpolarization. The studies summarized in this chapter tested whether chronic blockade of the reninangiotensin system with either the angiotensin II type 1 (AT1) receptor antagonist candesartan cilexetil or the converting enzyme inhibitor enalapril improves age-related endothelial dysfunction (Goto et al., 2000b; Kansui et al., 2002). 2. METHODS

2.1. Animals Nine-month-old male Wistar Kyoto rats (WKY) were assigned to either a control group or to one of three treatment groups. The WKY were treated with either a combination of hydralazine 50 mg/kg/day and hydrochlorothiazide 7.5 mg/kg/day, enalapril 20 mg/kg/day, or candesartan cilexetil 3.5 mg/kg/day for three months until the age of 12 months. All drugs were given in drinking water. Untreated, three-month-old WKY also served as young controls. There were 7–12 rats in each of the five groups. Systolic blood pressure was measured in conscious rats by the tail-cuff method. The drugs were withdrawn two days prior to the experiments. The rats were anesthetized with ether and killed by decapitation. The main branch of the mesenteric artery was excised and bathed in cold Krebs solution having the following composition (in mmol/L): Na⫹ 137.4; K⫹ 5.9; ⫺ ⫺ Mg2⫹ 1.2; Ca2⫹ 2.5; HCO⫺ 3 15.5; H2PO4 1.2; Cl 134; and glucose 11.5. The artery was cut into rings of 3 mm and 1.2 mm for the electrophysiological and tension experiments, respectively.

2.2. Membrane potential recording Transverse strips cut along the longitudinal axis of the rings were placed in an experimental chamber with the endothelial layer up. Tissues were carefully pinned to the rubber base attached to the bottom of the 2-ml chamber and then superfused with 36 ⬚C Krebs solution aerated with 95% O2–5% CO2 (pH 7.3–7.4) at a rate of 3 ml/min. After equilibration for at least 60 min, the membrane potentials of vascular smooth muscle cells were recorded using conventional microelectrode technique. Briefly, conventional glass capillary microelectrodes filled with 3 mol/L KCl and with tip resistance of 50–80 M⍀ were inserted into the smooth muscle cells from the endothelial side. Electrical signals were amplified through an amplifier (MEZ-7200, Nihon Kohden, Tokyo, Japan), monitored on an oscilloscope (VC-11, Nihon Kohden), and recorded with a pen recorder (RJG-4002, Nihon Kohden).

2.3. Isometric tension recording Rings with endothelium were placed in 5-ml organ chambers filled with 36 ⬚C Krebs solution aerated with 93% O2–7% CO2 (pH 7.4). Two fine, stainless steel wires were placed through the lumen of the ring; one was anchored, and the other was attached to the mechanotransducer (UM-203, Kishimoto, Kyoto, Japan). After the rings were allowed to equilibrate for 60 min at an optimal resting tension of 1.0 g, they were exposed to indomethacin, a cyclooxygenase-inhibitor, and NG-nitro-L-arginine, a NO synthase inhibitor. The rings were contracted with 10⫺5 mol/L norepinephrine, and the relaxant effects of acetylcholine were studied by adding the drug in increasing concentrations, from 10⫺9 to 10⫺5 mol/L.

RAS blockers prevent age-related decline in EDHF 119 In some preparations, the rings were contracted with 77mmol/L KCl solution in the presence of 10⫺5 mol/L indomethacin, and a relaxation to acetylcholine was observed. Relaxation in response to levcromakalim, a direct activator of ATP-sensitive K⫹ channels, and sodium nitroprusside was studied in rings contracted with 10⫺5 mol/L norepinephrine in the presence of 10⫺5 mol/L indomethacin. The extent of the relaxation was expressed as the percentage of the initial contraction evoked by the contractile agonist.

2.4. Drugs and solutions Drugs used in this study were acetylcholine chloride, norepinephrine, indomethacin, NGnitro-L-arginine, enalapril, hydralazine, hydrochlorothiazide, sodium nitroprusside (Sigma, St Louis, USA), candesartan cilexetil (a gift from Takeda Pharmaceuticals, Osaka, Japan), and levcromakalim (a gift from SmithKline Beecham Pharmaceuticals, Worthing, UK). Indomethacin was dissolved in 10 mmol/L Na2CO3, NG-nitro-L-arginine in 0.2 mol/L HCl, and levcromakalim in 90% ethanol. All other drugs were dissolved in distilled water. The solutions containing 20 mmol/L or 77 mmol/L KCl were obtained by equimolar replacement of NaCl by KCl in Krebs solution.

2.5. Statistics Results are given as mean ⫾ SEM. The concentration-response curves for hyperpolarizations and relaxations were analyzed by two-way analysis of variance followed by Scheffé’s test for multiple comparisons. Other variables were analyzed by one-way analysis of variance followed by Scheffé’s test for multiple comparisons or paired Student’s t-test. A level of P less than 0.05 was considered statistically significant. 3. RESULTS

3.1. EDHF-mediated hyperpolarization and relaxation Enalapril, candesartan and a combination of hydralazine and hydrochlorothiazide lowered arterial blood pressure to a comparable extent (Table 15.1). The resting membrane potential of smooth muscle cells of the mesenteric artery did not differ among the study groups. Table 15.1 Systolic blood pressure before and after 3 months of treatment, and the resting membrane potential of the mesenteric arteries in the five study groups SBP (mmHg)

WKY-OLD WKY-H WKY-ENA WKY-CAN WKY-Y

Membrane potential (mV)

Before

After

150 ⫾ 4 158 ⫾ 4 157 ⫾ 3 153 ⫾ 3 148 ⫾ 4

156 ⫾ 5 124 ⫾ 4*† 123 ⫾ 6*† 125 ⫾ 2*† ND

⫺46.8 ⫾ 0.8 ⫺50.3 ⫾ 2.2 ⫺49.8 ⫾ 1.1 ⫺51.9 ⫾ 1.1 ⫺48.8 ⫾ 1.2

Notes Values are mean ⫾ SEM. n ⫽ 7–12 rats in each group. SBP, systolic blood pressure; WKY-OLD, 12-month-old WKY; WKY-H, WKY treated with the combination of hydralazine and hydrochlorthiazide, WKY-ENA, enalapril-treated WKY; WKY-CAN, candesartan-treated WKY; WKY-Y, 3-month-old WKY; ND, not determined. * P ⬍ 0.05 vs WKY-OLD; †P ⬍ 0.05 vs before treatment.

120

Yasuo Kansui et al.

Enalapril treatment and candesartan treatment, but not the combination of hydralazine and hydrochlorothiazide, improved hyperpolarization to acetylcholine, applied either at the resting state of the membrane (Figure 15.1) or in depolarized condition with norepinephrine (Figure 15.2), compared with age-matched WKY (p ⬍ 0.05), and the response attained in enalapril-treated WKY and candesartan-treated WKY was comparable to that in young WKY (Figures 15.1, 15.2). EDHF-mediated relaxation to acetylcholine, as assessed in rings contracted with norepinephrine in the presence of indomethacin and NG-nitro-L-arginine, was significantly smaller in 12-month-old WKY than in 3-month-old WKY (Figure 15.3). Enalapril treatment and candesartan treatment, but not the combination of hydralazine and hydrochlorothiazide, significantly improved EDHF-mediated relaxations to acetylcholine to a level comparable to that in 3-month-old WKY (Figure 15.3).

3.2. Endothelium-dependent, nitric oxide-mediated relaxation When rings treated with indomethacin were contracted with 77 mmol/L KCl, which eliminates EDHF-mediated hyperpolarization, no difference was found in acetylcholine-induced relaxations among the five groups, suggesting that nitric oxide-mediated relaxation may not be impaired at the age of 12 months (Figure 15.4).

3.3. Endothelium-independent hyperpolarization and relaxation Endothelium-independent relaxation to sodium nitroprusside, a NO donor did not differ among the five study groups (data not shown).

Hyperpolarization (mV)

0

–5

–10

–15

–20

WKY-OLD WKY-H WKY-ENA WKY-CAN WKY-Y 8

*† *† *† 7

6

5

Acetylcholine (-log mol/L)

Figure 15.1 Concentration-response curves of hyperpolarization to acetylcholine in mesenteric arteries with endothelium of 12-month-old, untreated Wistar Kyoto rats (WKYOLD), WKY treated with a combination of hydralazine and hydrochlorothiazide (WKY-H), enalapril-treated WKY (WKY-ENA), candesartan-treated WKY (WKYCAN), and 3-month-old, untreated WKY (WKY-Y). Acetylcholine was applied in the resting state of the cell membrane. Values are mean ⫾ SEM. There were 7 to 12 rats in each group. The asterisk indicates P ⬍ 0.05 vs WKY-OLD; The dagger indicates P ⬍ 0.05 vs WKY-H, by two-way analysis of variance (modified from Kansui et al., 2002).

A

B

ACh 10–5 mol/L WKY-OLD

0

Acetylcholine 10–7 mol/L 10–5 mol/L

Hyperpolarization (mV)

WKY-H

WKY-ENA

20 mV

WKY-CAN

WKY-Y

–10

–20

WKY-OLD WKY-H WKY-ENA WKY-CAN WKY-Y

–30

* * *†

3 min

Figure 15.2 (A) Representative tracings showing hyperpolarization to 10⫺5 mol/L acetylcholine (ACh) under conditions of depolarization with norepinephrine (10⫺5 mol/L) in the presence of indomethacin (10⫺5 mol/L) in mesenteric arteries with endothelium of 12-monthold, untreated WKY (WKY-OLD), WKY treated with a combination of hydralazine and hydrochlorothiazide (WKY-H), enalapril-treated WKY (WKY-ENA), candesartan-treated WKY (WKY-CAN), and 3-month-old, untreated WKY (WKY-Y). (B) Amplitudes of hyperpolarizations to 10⫺7 mol/L and 10⫺5 mol/L ACh based on panel A. Values are mean ⫾ SEM. There were 7 to 12 rats in each group. The asterisk indicates P ⬍ 0.05 vs WKYOLD; The dagger indicates P ⬍ 0.05 vs WKY-H, by two-way analysis of variance (modified from Kansui et al., 2002).

0

Relaxation (%)

–20

–40

–60

–80

WKY-OLD WKY-H WKY-ENA WKY-CAN WKY-Y

*† *† *†

–100 9

8 7 6 Acetylcholine (-log M)

5

Figure 15.3 Concentration-relaxation curves to acetylcholine in mesenteric arterial rings with endothelium contracted with norepinephrine (10⫺5 mol/L) in the presence of indomethacin (10⫺5 mol/L) and NG-nitro-L-arginine (10⫺4 mol/L) of WKY-OLD, WKY-H, WKY-ENA, WKY-CAN and WKY-Y. There were 7 to 12 rats in each group. The asterisk indicates P ⬍ 0.05 vs WKY-OLD; The dagger indicates P ⬍ 0.05 vs WKY-H, by two-way analysis of variance (modified from Kansui et al., 2002).

122

Yasuo Kansui et al. 0

Relaxation (%)

–20

–40

WKY-OLD WKY-H WKY-ENA WKY-CAN WKY-Y

–60

–80

P = N.S. for all

–100 9

8

7

6

5

Acetylcholine (-log M)

Figure 15.4 Concentration-relaxation curves to acetylcholine in rings contracted with 77 mmol/L KCl in the presence of 10⫺5 mol/L indomethacin in mesenteric arterial rings with endothelium of WKY-OLD, WKY-H, WKY-ENA, WKY-CAN and WKY-Y. There were 7 to 12 rats in each group. NS, not significant.

Endothelium-independent hyperpolarization and relaxation to levcromakalim, a direct activator of ATP-sensitive K⫹-channels were also similar among the five study groups (data not shown). 4. DISCUSSION The studies summarized in this chapter have demonstrated that the inhibitors of the reninangiotensin system, that is, the AT1 receptor antagonist candesartan and the converting enzyme inhibitor enalapril, but not a traditional combination of antihypertensive agents, improved EDHF-mediated hyperpolarization and relaxation in the mesenteric arteries of aged (12-month-old) WKY compared with age-matched WKY (Goto et al., 2000b; Kansui et al., 2002). EDHF-mediated hyperpolarization and relaxation are impaired not only by hypertension but also with increasing age even in normotensive rats, which becomes evident at the age of about 12 months (Fujii et al., 1993, 2001; Nakashima and Vanhoutte, 1996). Inhibitors of the renin-angiotensin system improved endothelial dysfunction in hypertension (Tschudi et al., 1994; Onaka et al., 1998; Goto et al., 2000a), partially independently of their blood pressure lowering effects (Schiffrin and Deng, 1995; Onaka et al., 1998; Schiffrin et al., 2000). In addition, such improvement was in part attributed to the restoration of EDHFmediated hyperpolarization (Onaka et al., 1998; Goto et al., 2000a). In the present study, the inhibitors of the renin-angiotensin system, but not the traditional combination of antihypertensive agents, improved the EDHF-mediated responses in 12-month-old, aged rats despite similar reductions in blood pressure in all treated animals. These findings suggest that the renin-angiotensin system blockade itself may play a pivotal role in the improvement of age-related impairment of EDHF-mediated responses, and that a lowering of arterial blood pressure alone is not sufficient to account for this improvement.

RAS blockers prevent age-related decline in EDHF 123 AT1 receptor antagonists and converting enzyme inhibitors exhibit some ancillary effects: for example, AT1 receptor antagonists block the action of angiotensin II regardless of its formation-pathway (Urata et al., 1996); under blockade of AT1 receptors, angiotensin II may stimulate angiotensin II type 2 (AT2) receptors; and converting enzyme inhibitors inhibit bradykinin degradation (Mombouli et al., 1992; Nakashima et al., 1993). However, because in the present study AT1 receptor antagonists and converting enzyme inhibitors improved EDHF-mediated responses comparably, their beneficial effects on endothelial function may be mainly attributed to the inhibition of the action of angiotensin II rather than to the other effects of each agent (Goto et al., 2000a; Kansui et al., 2002). The underlying mechanisms of the improvement of EDHF-mediated responses by the inhibitors of the renin-angiotensin system in aged rats remain unclear, partially because of the elusive nature of EDHF (Cohen and Vanhoutte, 1995). However, in the rat mesenteric artery, there is some evidence that EDHF-mediated hyperpolarization is at least partially mediated by myoendothelial gap junctions (Edwards et al., 1999; Hill et al., 2000; Goto et al., 2002). It is now well documented that the renin-angiotensin system blockade efficiently corrects cardiovascular structural remodelings. If gap junctional communication is indeed involved in the EDHF-mediated response (Chaytor et al., 1998; Edwards et al., 1999; Yamamoto et al., 1999), it is possible that vascular structural changes, involving gap junctions, might in part account for the impaired EDHF-mediated responses with aging and its improvement by the renin-angiotensin system inhibitors. Endothelium-dependent relaxation is impaired in humans (Taddei et al., 1995), and this impairment may partially be accounted for by a defective EDHF system (UrakamiHarasawa et al., 1997). Only limited information is available as to whether or not certain therapeutic interventions can alleviate the age-related endothelial dysfunction (Atkinson et al., 1994). The studies summarized in this chapter provide unequivocal evidence that the inhibitors of the renin-angiotensin system can improve the age-related endothelial dysfunction attributable to a defective EDHF system. These findings raise the possibility that the inhibitors of the renin-angiotensin system might work as novel tools with which to prevent age-related endothelial dysfunction. In conclusion, the studies summarized in this chapter demonstrate that the treatment of normotensive rats with either an AT1 receptor antagonist or an inhibitor of converting enzyme prevents age-related impairment of EDHF-mediated hyperpolarization and relaxation, presumably through inhibition of the renin-angiotensin system itself but not because of the lowering of arterial pressure alone.

16 Characterization of endotheliumderived hyperpolarizing factor-mediated relaxation of small mesenteric arteries from diabetic (db/db⫺/⫺) mice Malarvannan Pannirselvam, Todd J. Anderson and Christopher R. Triggle Wire myograph techniques were used to investigate acetylcholine- and bradykinin- mediated relaxation of contracted small mesenteric arteries from control (db/db⫹/?) and diabetic (db/db⫺/⫺) mice. The maximal relaxations to acetylcholine were reduced in db/db⫺/⫺ compared to db/db⫹/?, whereas bradykinin-induced maximal relaxations were similar. The combination of indomethacin, L-nitroarginine and ODQ reduced maximal relaxations to acetylcholine and bradykinin in db/db⫹/?, but not in db/db⫺/⫺ mice. The EDHF-mediated component of the relaxation to acetylcholine and bradykinin of arteries from db/db⫺/⫺, was completely inhibited by high potassium and tetraethylammonium at 10⫺2 M, but not by tetraethylammonium at 10⫺3 M. Charybdotoxin, iberiotoxin and a combination of ouabain and barium significantly reduced the maximal relaxation to acetylcholine in db/db⫺/⫺. Charybdotoxin or iberiotoxin either alone or in combination with apamin reduced the sensitivity to the EDHF-mediated component of the response to bradykinin. 17-ODYA and catalase had no effect on either sensitivity or maximal relaxation to acetylcholine. In contrast, 17-ODYA, but not catalase, significantly reduced the sensitivity, but not the maximal relaxation, to bradykinin in arteries from db/db⫺/⫺, but not in db/db⫹/? mice. Together, these data suggest that the bradykinin-induced, EDHF-dependent relaxation of arteries from db/db⫺/⫺ is mediated via a cytochrome P450 product of arachidonic acid that activates large-conductance, calcium-activated potassium channels, whereas the acetylcholine-induced, EDHF-mediated relaxation of arteries from db/db⫺/⫺ involves neither a cytochrome P450 product nor hydrogen peroxide.

1. INTRODUCTION Vascular complications are the principal cause of morbidity and mortality in patients with type II diabetes. Endothelial dysfunction probably plays an important role in the pathogenesis of diabetic vascular complications (De Vriese et al., 2000b). The hyperglycemia, a characteristic feature of type I and type II diabetes, through different biochemical mechanisms including the polyol pathway (Dvornik et al., 1973), abnormal advanced glycation end products (Brownlee et al., 1986), protein kinase C pathway (King et al., 1997) and oxidative stress (Curcio and Ceriello, 1992) contributes to the endothelial dysfunction which leads to macroand microvascular complications. The endothelial cells modulate vascular tone by releasing various vasoactive substances including prostacyclin (Moncada and Vane, 1979a), nitric oxide (Furchgott and Zawadzki, 1980; Palmer et al., 1987) and endothelium-derived hyperpolarizing factor (EDHF, Félétou and Vanhoutte, 2000; McGuire et al., 2001). Impaired endothelium-dependent relaxation

EDHF-mediated relaxation of small mesenteric arteries

125

has been consistently demonstrated in different vascular beds from streptozotocin (STZ)induced diabetic rats (De Vriese et al., 2000b) The endothelium-dependent relaxation in response to acetylcholine and bradykinin is largely mediated by the release of nitric oxide. However, acetylcholine-induced, endothelium-dependent vasodilatation may also be associated with hyperpolarization and involves a non-prostanoid non-nitric oxide factor, termed EDHF. The role of EDHF and endothelium-dependent hyperpolarization may be secondary to the effects of nitric oxide (NO) in large conduit vessels, but is the major determinant of vascular tone in resistance vessels (Garland et al., 1995; Félétou and Vanhoutte, 2000; McGuire et al., 2001). In small mesenteric arteries from mice deficient in endothelial nitric oxide synthase, the contribution of EDHF is up-regulated (Waldron et al., 1999; Ding et al., 2000). By contrast, endothelium-dependent hyperpolarization is depressed in mesenteric arteries from spontaneously hypertensive rats (Fujii et al., 1992). The impaired endothelium-dependent relaxation of mesenteric arteries from STZ-induced diabetic rat results in part from a diminished contribution from EDHF (Fukao et al., 1997c; Makino et al., 2000; Wigg et al., 2001). However, the data from animal models of diabetes has been primarily derived from the streptozotocin-induced diabetic rat, which is a model of type I diabetes. Since more than 90% of the diabetic population exhibit type II diabetes, it seems important to study endothelial function in animal models of type II diabetes. Thus endothelium-dependent relaxations of small mesenteric arteries from spontaneously diabetic (db/db⫺/⫺) mice are impaired (Pannirselvam et al., 2002). The present study was designed to investigate the pharmacological characteristics of EDHF-mediated relaxation to acetylcholine and bradykinin, in arteries of that model.

2. METHODS

2.1. Animals Twelve- to sixteen-week-old male C57BL/KsJ db/db mice and non-diabetic controls were purchased from Jackson Laboratories (Bar Harbour, ME, USA). In accordance with a protocol approved by the University of Calgary Animal Care Committee, mice were killed by cervical dislocation. The mesenteric arcade was removed and first order branches of the mesenteric artery were dissected out into cold Krebs solution of the following composition (in mM): NaCl 120, NaHCO3 25, KCl 4.8, NaH2PO4 1.2, MgSO4 1.2, Dextrose 11.0, CaCl2 1.8, aerated with 95% O2 and 5% CO2.

2.2. Myograph studies Small mesenteric arteries were cut into 2 mm rings and mounted on a Mulvany–Halpern myograph. The passive tension-internal circumference was determined by stretching to achieve an internal circumference equivalent to 90% of that of the blood vessel under a transmural pressure of 100 mmHg (Mulvany and Halpern, 1977). All experiments were performed at 37 ⬚C.

2.3. Statistics Data are expressed as pEC50 values, defined as the negative logarithm to base ten of the EC50 values and maximal relaxation, as the maximal response obtained at the highest concentration

126

Malarvannan Pannirselvam et al.

tried. In all experiments, n equals the number of animals used in the protocol. Relaxation is expressed as mean percentage (⫾ SEM) of phenylephrine-induced tone. Statistical significance of difference between means of different groups was evaluated using one-way ANOVA. Multiple comparisons of the paired groups were performed using Student–Newman–Keuls method. A P value less than 0.05 was considered to be statistically significant.

2.4. Drugs Acetylcholine, apamin, barium, bradykinin, catalase, charybdotoxin, iberiotoxin, indomethacin, N␻;-nitro-L-arginine (L-NNA), 17-octadecynoic acid (17-ODYA), ouabain, 1H-(1,2,4) oxadiazolo (4,3-a)quinoxalin-1-one (ODQ), phenylephrine, tetraethylammonium, were obtained from sigma (St Louis, MO, USA). Drugs were dissolved in distilled water except indomethacin (dissolved in 95% ethanol), ODQ and 17-ODYA (dissolved in dimethyl sulfoxide). 3. RESULTS

3.1. Acetylcholine and bradykinin The acetylcholine-induced maximal relaxation during phenylephrine (10⫺6 M)-induced contractions of small mesenteric arteries from db/db⫹/? mice was significantly (P ⬍ 0.05) reduced by a combination of L-NNA (10⫺4 M) and indomethacin (10⫺5 M) (Figure 16.1). The addition of ODQ (10⫺5 M) to the combination of L-NNA and indomethacin had no further

Figure 16.1 Acetylcholine-induced (A) and bradykinin-induced (B) maximal relaxation in small mesenteric arteries from db/db ⫹/? and db/db⫺/⫺ mice, in the absence and presence of combinations of inhibitors of cyclooxygenase (indomethacin, 10⫺5 M) plus nitric oxide synthase (N␻;-nitro-L-arginine; L-NNA, 10⫺4 M) or inhibitors of cyclooxygenase, nitric oxide synthase and soluble guanylyl cyclase [1H-(1,2,4) oxadiazolo(4,3-a)quinoxalin-1-one; ODQ, 10⫺5 M]. Tissues were contracted with phenylephrine (10⫺6 M). Data are shown as means⫾SEM (n⫽5–16). The asterisk indicates P⬍0.05 compared to respective control group, The dagger indicate P ⬍ 0.05 compared to db/db⫹/? control group.

EDHF-mediated relaxation of small mesenteric arteries

127

effect. The acetylcholine-induced maximal relaxation of small mesenteric arteries from db/db⫺/⫺ was significantly reduced when compared to db/db⫹/? However, the combination of L-NNA and indomethacin did not significantly reduce the acetylcholine-induced maximal relaxation. The bradykinin-induced maximal relaxations of small mesenteric arteries from db/db⫹/? and db/db⫺/⫺ mice were similar, without any difference in the sensitivity (Figure 16.1). The bradykinin-induced relaxation was inhibited significantly by the combination of L-NNA plus indomethacin in db/db⫹/? but not in db/db⫺/⫺ mice (Figure 16.1).

3.2. Role of potassium channels in EDHF-mediated relaxation in small mesenteric arteries from db/db⫺/⫺ mice A depolarizing concentration of potassium (60 mM) abolished EDHF-mediated relaxation to acetylcholine and bradykinin in small mesenteric arteries from db/db⫹/? and db/db⫺/⫺ mice (Figure 16.2). Tetraethylammonium (10⫺2 M) significantly reduced the EDHFmediated maximal relaxations to acetylcholine and bradykinin in both db/db⫹/? and db/db⫺/⫺ mice whereas tetraethylammonium (10⫺3 M) had no effect (Figure 16.2). Apamin (10⫺6 M), alone, had no effect on acetylcholine-induced EDHF-mediated relaxation in small mesenteric arteries from db/db⫹/? and db/db⫺/⫺ mice (Figure 16.3).

Figure 16.2 Effect of depolarizing concentration of potassium chloride (KCl; 60 mM), tetraethylammonium (TEA 10⫺3 M and 10⫺2 M) in the presence of inhibitors of cyclooxygenase (indomethacin, 10⫺5 M), nitric oxide synthase (N␻;-nitro-L-arginine; L-NNA, 10⫺4 M) and soluble guanylyl cyclase [1H-(1,2,4)oxadiazolo(4,3-a)quinoxalin-1-one; ODQ, 10⫺5 M] on acetylcholine-induced (A) and bradykinin-induced (B) maximal relaxations of small mesenteric arteries from db/db ⫹/? and db/db ⫺/⫺ mice. Tissues were contracted with phenylephrine (10⫺6 M). Data are shown as means ⫾ SEM (n ⫽ 3–5). The asterisk indicates P ⬍ 0.05 compared to control group, the dagger indicates P ⬍ 0.05 compared to tetraethylammonium 10⫺3 M group.

128

Malarvannan Pannirselvam et al.

Figure 16.3 Effect of apamin (10⫺6 M), charybdotoxin (ChTX, 10⫺7 M), iberiotoxin (IbTX, 10⫺7 M), ouabain (10⫺5 M), barium (Ba, 3⫻10⫺5 M), in the presence of inhibitors of cyclooxygenase (indomethacin, 10⫺5 M) and nitric oxide synthase (N␻;-nitro-Larginine; L-NNA, 10⫺4 M), on acetylcholine-induced maximal relaxation of small mesenteric arteries from db/db⫹/? and db/db⫺/⫺ mice. Tissues were contracted with phenylephrine (10⫺6 M). Data are shown as mean ⫾ SEM (n ⫽ 3–17). The asterisk indicates P ⬍ 0.05 compared to control group, the dagger indicates P ⬍ 0.05 compared to all other groups.

Charybdotoxin (10⫺7 M) or iberiotoxin (10⫺7 M), alone, significantly inhibited the EDHFmediated maximal relaxations in small mesenteric arteries from db/db⫺/⫺ whereas only a combination of charybdotoxin plus apamin inhibited it in arteries from db/db⫹/? mice. However, iberiotoxin had no effect on acetylcholine-induced EDHF-mediated relaxations of small mesenteric arteries from db/db⫹/? mice (Figure 16.3). The combination of ouabain (10⫺5 M) and barium (3 ⫻ 10⫺5 M) significantly inhibited the EDHF-mediated maximal relaxations in both db/db⫹/? and db/db⫺/⫺ mice (Figure 16.3). The bradykinin-induced EDHF-mediated relaxation was not affected by apamin in either db/db⫹/? or db/db⫺/⫺ mice. Charybdotoxin or iberiotoxin, alone, significantly reduced the sensitivity to bradykinin in small mesenteric arteries from db/db⫺/⫺ mice, without a significant change in maximal relaxation. Ouabain and barium, either alone or in combination, had no effect on bradykinin-induced EDHF-mediated relaxation in db/db⫺/⫺ mice (Figure 16.4). However, in small mesenteric arteries from db/db⫹/?, ouabain, barium, or a combination of charybdotoxin and apamin, but not iberiotoxin, significantly reduced the sensitivity to bradykinin without affecting the maximal relaxation to the peptide (Figure 16.4).

3.3. Cytochrome P450 17-ODYA (10⫺5 M) had no effect on acetylcholine-induced EDHF-mediated relaxations in small mesenteric arteries from db/db⫹/? or db/db⫺/⫺ mice. However, 17-ODYA significantly reduced the sensitivity to bradykinin in small mesenteric arteries from db/db⫺/⫺ mice, but not in small mesenteric arteries from db/db⫹/?. Catalase (1250 U/ml) had no effect on either acetylcholine- or bradykinin-induced EDHF-mediated relaxation of small mesenteric arteries from db/db⫺/⫺ and db/db⫹/? mice (Figure 16.5). 4. DISCUSSION Spontaneously diabetic (db/db⫺/⫺) mice used in this study were obese, showed significantly high glucose, triglycerides and cholesterol levels compared to non-diabetic controls.

EDHF-mediated relaxation of small mesenteric arteries

129

Figure 16.4 Effect of apamin (10⫺6 M), charybdotoxin (ChTX, 10⫺7 M), iberiotoxin (IbTX, 10⫺7 M), ouabain (10⫺5 M), barium (Ba, 3⫻10⫺5 M), in the presence of inhibitors of cyclooxygenase (indomethacin, 10⫺5 M) and nitric oxide synthase (N␻-nitro-Larginine; L-NNA, 10⫺4 M), on the sensitivity to bradykinin of small mesenteric arteries from db/db⫹/? and db/db⫺/⫺ mice. Tissues were contracted with phenylephrine (10⫺6 M). Data are shown as mean⫾SEM (n⫽5–8). The asterisk indicates P⬍0.05 compared to control group, the dagger indicates P⬍0.05 compared to apamin group.

Figure 16.5 Effect of 17-ODYA (10⫺5 M) and catalase (1250 U/ml) on the sensitivity to acetylcholine (A) and bradykinin (B) in small mesenteric arteries from db/db⫹/? and db/db⫺/⫺ mice in the presence of a combination of inhibitors of cyclooxygenase (indomethacin, 10⫺5 M) plus nitric oxide synthase (N␻-nitro-L-arginine; L-NNA, 10⫺4 M). Tissues were contracted with phenylephrine (10⫺6 M). Data are shown as means ⫾ SEM (n ⫽ 5–7). The asterisk indicates P ⬍ 0.05 compared to respective control group.

130

Malarvannan Pannirselvam et al.

The small mesenteric arteries from db/db⫺/⫺ showed enhanced vascular reactivity to phenylephrine and impaired relaxation to acetylcholine (Pannirselvam et al., 2002). The present study demonstrates a selective impaired endothelium-dependent relaxation to acetylcholine but not to bradykinin in small mesenteric arteries from db/db⫺/⫺. Similar results with an impaired endothelium-dependent relaxation to acetylcholine, but not to bradykinin were obtained in mesenteric and popliteal arteries of STZ-induced diabetic rats (Taylor et al., 1995). Furthermore, in the present study, acetylcholine- and bradykinininduced relaxation of small mesenteric arteries was resistant to a combination of inhibitors of cyclooxygenase, nitric oxide synthase and soluble guanylyl cyclase. This observation suggests that the EDHF-mediated relaxation of small mesenteric arteries is not affected by the diabetic state. This is in contrast to earlier reports from STZ-induced diabetic rats concluding that the impaired endothelium-dependent relaxation of the mesenteric artery results from a decreased EDHF-mediated contribution (Fukao et al., 1997c). This earlier study showed that the amplitude and duration of the hyperpolarization produced by acetylcholine was decreased in mesenteric artery from diabetic rats (Fukao et al., 1997c). Inhibition of nitric oxide synthase did not affect the hyperpolarization to acetylcholine in mesenteric artery from control and diabetic rats, thus prompting the conclusion that the EDHF-mediated response is dysfunctional in the mesenteric artery of diabetic rats. The selective impairment of endothelium-dependent relaxation to acetylcholine is reported in mesenteric arteries, but not in femoral arteries. Further research confirms that the impairment of the response in the former can be attributed to a reduced EDHF-dependent response (Wigg et al., 2001). A depolarizing concentration of potassium and tetraethylammonium (10⫺2 M), a nonselective blocker of potassium channels, reduced the maximal EDHF-mediated relaxation to acetylcholine- and bradykinin, suggesting an involvement of potassium channels. Charybdotoxin, an inhibitor of large/intermediate-conductance, calcium-activated potassium channel, and apamin, an inhibitor of small mesenteric arteries-conductance potassium channel, abolished the acetylcholine-induced EDHF-mediated relaxation of rat mesenteric arteries when selectively applied intraluminally, but not extraluminally (Doughty et al., 1999). These results suggest the presence of charybdotoxin-sensitive and apamin-sensitive calcium-activated potassium channels in the endothelium. The requirement for the combination of charybdotoxin and apamin has become one of the characteristic features of EDHF-mediated relaxations. Iberiotoxin, a selective inhibitor of large-conductance, calcium-activated potassium channels, cannot substitute for charybdotoxin (McGuire et al., 2001). Activation of charybdotoxin and apamin sensitive potassium channels in the endothelium will lead to the efflux of potassium, which in turn can activate barium-sensitive, inward rectifying potassium channels and ouabain-sensitive, sodium/potassium ATPase in the vascular smooth muscle, leading to its hyperpolarization and relaxation. Under these conditions potassium ions can act as EDHF (Edwards et al., 1998). In the present study, the combination of charybdotoxin with apamin and barium with ouabain significantly reduced the EDHF-mediated relaxations to acetylcholine-induced and bradykinin-induced relaxation in small mesenteric arteries from db/db⫹/? mice. However, iberiotoxin had no effect on EDHF-mediated relaxation, as already described in the mice mesenteric artery (Ding et al., 2000). In contrast, charybdotoxin or iberiotoxin alone inhibited the acetylcholine-induced and bradykinin-induced EDHF-mediated relaxation of small mesenteric arteries from db/db⫺/⫺ mice whereas apamin, either alone or in combination with charybdotoxin, had no effect. The combination of ouabain and barium reduced the EDHF-mediated maximal relaxation to acetylcholine. These data suggest that the pharmacological characteristics of the acetylcholine-induced and bradykinin-induced EDHF-mediated

EDHF-mediated relaxation of small mesenteric arteries

131

relaxations differ in db/db⫺/⫺ mice. The acetylcholine-induced EDHF-mediated relaxation seems to involve large-conductance, calcium-activated potassium channels, inward rectifying potassium channels and sodium/potassium ATPase whereas the bradykinin-induced EDHF-mediated relaxation involves mainly large-conductance, calcium-activated potassium channels. Two candidate molecules for EDHF that can activate large-conductance calcium-activated potassium channels are epoxyeicosatetranoeic acid (EETs), a product of cytochrome P450 metabolite of arachidonic acid (Campbell et al., 1996) and reactive oxygen species such as hydrogen peroxide (Barlow and White, 1998; Matoba et al., 2000). In the current study, 17-ODYA, a specific inhibitor of cytochrome P450, reduced the sensitivity to bradykinin of small mesenteric arteries from db/db⫺/⫺ but had no effect on acetylcholine-induced EDHF-mediated relaxations. These data suggest that the EETs play a role in the bradykinininduced, but not the acetylcholine-induced EDHF-mediated relaxation. Catalase, which dismutases hydrogen peroxide, had no effect on either acetylcholine-induced or bradykinininduced EDHF-mediated relaxations. This observation suggests that hydrogen peroxide is unlikely to be involved in EDHF-mediated responses of small mesenteric arteries from db/db⫺/⫺ and db/db⫹/? mice. In summary, the present study suggests that the pharmacological characteristics of EDHF vary depending on both the endothelium-dependent agonist as tested and the metabolic state of the darn animal. Thus bradykinin-induced EDHF-mediated relaxations involve EETs, which activate large-conductance, calcium-activated potassium channel leading to hyperpolarization and relaxation. Furthermore, it is likely that cytochrome P450 is upregulated in the diabetic state. In contrast, acetylcholine-induced EDHF-mediated relaxations of small mesenteric arteries from db/db⫺/⫺ mice involve large-conductance, calcium-activated potassium channel, inward rectifying potassium channels and the sodium/potassium ATPase, but not EETs. Hydrogen peroxide is not involved in the activation of largeconductance, calcium-activated potassium channel in this spontaneous model of diabetes.

17 Endothelium-dependent responses in small arteries isolated from normal and pre-eclamptic pregnant women William R. Dunn, Louise C. Kenny, David A. Kendall, Philip N. Baker and Michael D. Randall

Pre-eclampsia is a pregnancy-specific disorder associated with hypertension and altered endothelial function. This study investigated the mechanisms of endothelium-dependent vasodilatation in small arteries isolated from normal pregnant women and those with pregnancies complicated by pre-eclampsia. Small arteries (⬍ 500 ␮m) were obtained from uterine biopsies obtained at the time of surgery and dilator responses to bradykinin were determined in contracted arteries mounted on a pressure myograph at 60 mmHg. Responses to bradykinin were assessed in the absence and presence of N-nitro-L-arginine methyl ester or raised extracellular K⫹ alone or in combination. Further experiments, using vessels isolated from normal pregnant women, determined the effects of 18-␣-glycyrrhetinic acid, carbenoxolone, palmitoleic acid and SR141716A on responses to bradykinin. N-nitro-L-arginine methyl ester attenuated responses to bradykinin in vessels isolated from women with pre-eclampsia but was without effect in blood vessels isolated from normal pregnant women. In the latter arteries, raising extracellular potassium alone had no effect on relaxations to bradykinin, but, the combination of N-nitro-L-arginine methyl ester and extracellular K⫹ abolished the responses. In blood vessels isolated from normal pregnancies, responses to bradykinin were attenuated by 18-␣-glycyrrhetinic acid, carbenoxolone, palmitoleic acid and SR141716A. These results show that EDHF-type responses are evident in blood vessels isolated from normal pregnant women, but are largely absent in vessels from women with pre-eclampsia. The mechanism underlying endothelium-dependent vasodilatation in normal pregnancy involves functional gap junctions.

1. INTRODUCTION Pre-eclampsia is a multisystemic disorder of pregnancy characterised by hypertension and proteinuria. It is a leading cause of maternal mortality and is responsible for considerable perinatal mortality and morbidity (Hibbard and Milner, 1994), with additional implications in later life, including an increased risk of hypertension, heart disease and diabetes (Barker et al., 1990). While the precise aetiology of pre-eclampsia is poorly defined, a current model for the progression of the disease implicates inappropriate adaptation of the interface between the maternal vasculature and the developing placenta. In normal pregnancy, the process of placentation is believed to involve invasion of the maternal vasculature by cytotrophoblasts, which modify the structural properties of the uterine arteries, causing the loss of smooth muscle and hence contractility (Van Wijk et al., 2000). This leads to the uterine vasculature becoming a low resistance system that allows an increase in blood supply to the placenta and the developing fetus. In pre-eclampsia, the processes involved in placentation may be abnormal, such that the maternal vasculature is not adapted appropriately and there is a reduced delivery of blood to the placenta. In this model, poor perfusion of the placenta

Endothelium-dependent responses in small arteries

133

is proposed to result in the secretion of a factor(s) into the maternal circulation that, by some mechanism, results in the clinical manifestations of the disease. The nature of the circulating factor, and the site upon which it acts to produce its effects, is not fully established, but there is strong evidence that the endothelium is a target (Roberts et al., 1991). Indeed, a number of studies have reported abnormal endothelium-dependent responses in isolated systemic arteries obtained from women with pre-eclampsia compared to normal pregnancies (McCarthy et al., 1993, Knock and Poston, 1996; Pascoal et al., 1998). The aim of the present study was to compare endothelium-dependent responses in myometrial arteries isolated from normal and compromised pregnancies, since the blood supply to this vascular bed will determine blood flow delivery to the fetus, and also, to determine the nature of the factor released from the endothelium from blood vessels isolated from each patient group. 2. METHODS This investigation conformed to the principles outlined in the Declaration of Helsinki (1989). Ethical permission for the study was obtained from the Nottingham City Hospital Trust Ethical Committtee and all subjects gave written informed consent to participation. Samples were obtained at Caesarean section following the delivery of the baby and the placenta. A full thickness biopsy of myometrium (approximately 2 ⫻ 1 ⫻ 1 cm) was obtained from the upper margin of the transverse lower uterine incision. Specimens were placed in oxygenated physiological salt solution at 4⬚C. With the use of a dissecting microscope, arteries (⬍500␮m) were identified, dissected from the sample and utilised immediately.

2.1. Pressure myography Each artery was mounted on a pressure myograph as described in detail elsewhere (Wallis et al., 1996). In brief, vessels were secured between two glass cannulae using single strands (20 ␮m diameter) of surgical braided nylon suture. One cannula was closed and the other connected to a system containing physiological salt solution, which in turn was linked to a pressure-servo unit. This allowed the intraluminal pressure to be continuously monitored and precisely controlled. The arteriograph in which the vessel was secured was connected to a reservoir of physiological salt solution that was bubbled continuously with a 5% CO2–95% O2 gas mixture and circulated with a Masterflex pump. The vessel was imaged using a video camera and analysed with an appropriate dimension analyser (Living Systems Instrumentation, Burlington VT), which was linked to a MacLab data acquisition system. The intraluminal pressure was set at 60 mmHg and the temperature of the organ bath was maintained at 37 ⬚C. The blood vessels were allowed to equilibrate under these conditions for 90 min, during which time they developed a variable degree of myogenic tone. Subsequently, the arteries were further constricted to approximately 50% of their original diameter with incremental amounts of the thromboxane mimetic U46619 (10⫺9–10⫺6 M). All experimental drugs were applied extraluminally. Blood vessels were then exposed to increasing concentrations of the endothelium-dependent vasodilator, bradykinin in three-fold increments (10⫺9–10⫺6 M). Following washing, the concentration response curves to U46619 and bradykinin were repeated in the presence and absence of various inhibitors. The inhibitors assessed were N-nitro-L-arginine methyl ester (10⫺4 M), raised extracellular potassium (25 mM), the combination of N-nitro-L-arginine methyl ester (10⫺4 M) and raised extracellular potassium (25 mM), 18-␣-glycyrrhetinic acid (10⫺4 M), carbenoxolone (10⫺4 M), palmitoleic acid (5 ⫻ 10⫺6 M) and SR141716A (10⫺5 M). At the end of each

134

William R. Dunn et al.

experiment the vessels were washed for approximately 60 min in calcium free physiological salt solution in order to obtain the maximal passive diameter of each artery.

2.2. Drugs and solutions The following drugs and chemicals were used; 9,11-Dideoxy-11␣,9␣-epoxymethanoprostaglandin F2␣ (U46619), bradykinin acetate, indomethacin, N-nitro-L-arginine methyl ester (L-NAME), clotrimazole, Ethylene glycol-bis (␤-aminoethyl ether)-N,N,N⬘,N⬘tetraacetic acid (EGTA), Ethylenediaminetetraacetic acid (EDTA), 18-alpha-glycyrrhetinic acid, 3␤-Hydroxy-11-oxoolean-12-en-30-oic acid 3-hemisuccinate (carbenoxolone), cis-9Hexadecenoic acid (palmitoleic acid) and R(⫹)-arachidonyl-1⬘-hydroxy-2⬘-propylamide (methanandamide) (all obtained from Sigma, Poole, UK), and N-(piperidin-1-yl)-5(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide hydrochloride (SR141716A), supplied by Sanofi-Winthrop. All drugs were made up fresh from stock solutions on the day of the experiment and dissolved in physiological salt solution with the exception of SR141716A and clotrimazole which were dissolved in ethanol and palmitoleic acid and 18-␣-glycyrrhetinic acid, which were dissolved in dimethyl sulphoxide (DMSO). Extracellular potassium was increased by equimolar exchange of NaCl with KCl. All drugs were added to the extraluminal side of the vessel. Control experiments confirmed that the vehicles had no significant effect on constrictor tone or bradykinin-induced relaxations at the final concentration employed. The composition of the physiological salt solution was as follows (in mM): CaCl2.2H2O 1.6, NaCl 119, NaHCO3 25, KCl 4.7, KH2PO4 1.18, MgSO4.7H2O 1.17, Na2EDTA 0.023, Glucose 5.5 (pH 7.4 when gassed with 5% CO2–95% O2). Calcium-free physiological salt solution was prepared by omitting calcium and adding 0.5 mM EGTA. High K⫹ PSS was prepared by equimolar exchange of KCl with NaCl. Calcium-free physiological salt solution was prepared by omitting calcium and adding 0.5 mM EGTA.

2.3. Calculations and statistical analysis Vasodilator responses to bradykinin were calculated as percentages of the maximal possible vasodilator range, that is, the diameter in calcium-free physiological salt solution minus the diameter in the presence of myogenic and U46619-induced tone. Results are shown graphically as means (⫾ SEM), with n ⫽ the number of patients. Rmax refers to the maximal relaxation achieved in the presence of the highest concentration of bradykinin used (10⫺4 M). Statistical evaluation of concentration response curves within each group was performed using repeated measures analysis of variance (ANOVA) with post-hoc testing (Bonferroni/Dunn). Between group comparisons were analysed by the Mann Whitney U Test. A value for P ⬍ 0.05 was considered statistically significant. 3. RESULTS Samples were obtained from normal pregnant women (31.8 ⫾ 0.8 years, n ⫽ 25) and women with pre-eclampsia (27.7 ⫾ 1.2 years, n ⫽ 18). Mean arterial pressure was significantly greater (P ⬍ 0.05) in pre-eclamptic women (122 ⫾ 1.1 mmHg) compared with normal pregnant women (86.8 ⫾ 1.7 mmHg). Bradykinin produced concentration-dependent vasodilator responses that did not differ between groups with regard to potency or Rmax (Table 17.1).

Endothelium-dependent responses in small arteries

135

Table 17.1 Relaxation responses to bradykinin in myometrial resistance arteries isolated from women with normal pregnancies and with pregnancies complicated by pre-eclampsia

⫺logIC50 Rmax (%)

Pre-eclampsia (n ⫽ 6)

6.75 ⫾ 0.15 81.3 ⫾ 5.4

6.90 ⫾ 0.23 77.7 ⫾ 5.2

Normal pregnant

B

Pre-eclampsia

0

0

20

20

Relaxation (%)

Relaxation (%)

A

Normal pregnancy (n ⫽ 6)

40 60 80

* 60 80

100 –10

40

–9

–8

–7

Log [bradykinin] (mol/L)

–6

100 –10

–9

–8

–7

–6

Log [bradykinin] (mol/L)

Figure 17.1 Effect of N-nitro-L-arginine methyl ester (10⫺4 M) (closed circles) on responses to bradykinin (open circles) in myometrial small arteries isolated from women with normal pregnancies (A) and women with pregnancies complicated by pre-eclampsia (B). Each point represents the mean⫾SEM (n⫽6). The asterisk represents a statistically significant difference between means (P ⬍ 0.05).

Responses to bradykinin were reproducible over two consecutive concentration response curves in each of the patient groups (data not shown). L-NAME (10⫺4 M) did not significantly affect responses to bradykinin in myometrial small arteries isolated from normal pregnancies, but nearly abolished the response to bradykinin in blood vessels isolated from women with pre-eclampsia (control Rmax ⫽ 78.8 ⫾ 2.1%; L-NAME Rmax ⫽ 19.6 ⫾ 10.4%) (Figure 17.1). Raising extracellular potassium by 25 mM did not affect responses to bradykinin in myometrial small arteries isolated from women with either normal pregnancies or pregnancies complicated by pre-eclampsia (Figure 17.2). The combination of L-NAME (10⫺4 M) and raising extracellular potassium (25 mM) essentially abolished responses to bradykinin in myometrial small arteries from both normal (control Rmax ⫽ 83.0 ⫾ 5.6%; L-NAME plus raised extracellular potassium Rmax ⫽ 6.6 ⫾ 3.3%) and pre-eclamptic pregnancies (control Rmax ⫽ 78.8 ⫾ 2.1%; L-NAME plus raised extracellular potassium Rmax ⫽ 8.2 ⫾ 3.4%) (Figure 17.3). The putative gap junction inhibitors 18-␣-glycyrrhetinic acid (10⫺4 M) (Taylor et al., 1998), carbenoxelone (10⫺4 M) (Edwards et al., 1999), palmitoleic acid (5 ⫻ 10⫺6 M) (Harris et al., 2000) and SR1417116A (10⫺5 M) (Harris et al., 2000, 2002), virtually abolished responses to bradykinin in myometrial small arteries isolated from women with normal pregnancies (Figure 17.4).

136

William R. Dunn et al. Normal pregnant

B

Pre-eclampsia

0

0

20

20

Relaxation (%)

Relaxation (%)

A

40 60

40 60

80 100 –10

80

–9

–8

–7

100 –10

–6

Log [bradykinin] (mol/L)

–9

–8

–7

–6

Log [bradykinin] (mol/L)

Figure 17.2 Effect of raising extracellular potassium (by 25 mM) (closed circles) on responses to bradykinin (open circles) in myometrial small arteries isolated from women with normal pregnancies (A) and women with pregnancies complicated by pre-eclampsia (B). Each point represents the mean ⫾ SEM (n ⫽ 6).

Normal pregnant

Pre-eclampsia B

0

0

20

20

*

40 60

*

40 60 80

80 100 –10

Relaxation (%)

Relaxation (%)

A

–9

–8

–7

Log [bradykinin] (mol/L)

–6

100 –10

–9

–8

–7

–6

Log [bradykinin] (mol/L)

Figure 17.3 Effect of raising extracellular potassium (by 25 mM) alone (closed circles) and in combination with N-nitro-L-arginine methyl ester (10⫺4 M) (open squares) on responses to bradykinin (open circles) in myometrial small arteries isolated from women with normal pregnancies (A) and women with pregnancies complicated by pre-eclampsia (B). Each point represents the mean ⫾ SEM (n ⫽ 6). The asterisk represents a statistically significant difference between control responses and those in the presence of the combination of raised extracellular potassium and N-nitro-L-arginine methyl ester (P ⬍ 0.05).

4. DISCUSSION The present study shows that there is no difference in the overall sensitivity of responses to bradykinin in myometrial arteries isolated from normal pregnancies and those complicated by pre-eclampsia. This observations differs from observations made in isolated systemic small arteries studied under isometric conditions, where endothelium-dependent responses are impaired by pre-eclampsia in subcutaneous (McCarthy et al., 1993; Knock and Poston,

Endothelium-dependent responses in small arteries

A

B 0

Relaxation (%)

0 25 50

25

*

50

Control

75 100 –9

18-α-GA

–8 –7 –6 log [bradykinin] (M)

(10–4 M)

–5

C Relaxation (%)

100 –9

Control Carbenoxolone (10–4 M)

75

–8 –7 –6 log [bradykinin] (M)

–5

25

25

*

75 100 –9

*

D 0

0

50

137

–8 –7 –6 log [bradykinin] (M)

Control

50

Palmitoleic acid (5 × 10–6 M)

75

–5

100 –9

*

–8 –7 –6 log [bradykinin] (M)

Control SR141716A (10–5 M) –5

Figure 17.4 Effect of (A) 18-␣-glycyrrhetinic acid (18-␣-GA), (B) carbenoxolone, (C) palmitoleic acid and (D) SR1417116A on responses to bradykinin in myometrial small aretries isolated from women with normal pregnancies. Each point represents the mean ⫾ SEM (n ⫽ 6). The asterisk represents a statistically significant difference between means (P ⬍ 0.05).

1996) and in omental (Pascoal et al., 1998) arteries. This discrepancy could reflect differences in the susceptibility to endothelial dysfunction of systemic and myometrial small arteries in pre-eclampsia. Alternatively, they may reflect methodological differences, since myometrial small arteries isolated from pre-eclamptic women, mounted in an isometric myograph, show impaired responses to bradykinin (Ashworth et al., 1996). In contrast to the pressure myograph system used in the present study, it is relatively difficult to produce a consistent degree of vasoconstriction upon which to measure relaxation. Despite the fact that there was no difference in the overall response to bradykinin in pressurised vessels, a more detailed examination revealed major differences in the mechanism underlying endothelium-dependent responses in myometrial small arteries isolated from normal pregnancies and those compromised by pre-eclampsia. In women who had preeclampsia, responses to bradykinin were unaffected by raising extracellular potassium, but almost completely abolished by L-NAME, indicating that nitric oxide (NO) was the sole mediator of endothelium-dependent responses in these blood vessels. By contrast, responses to bradykinin in vessels isolated from normal pregnant women, were resistant to both L-NAME alone and raised extracellular potassium alone. This could be taken as evidence for the lack of involvement of either NO, or EDHF, in these vessels. However, the combination of L-NAME and raised extracellular potassium abolished responses to bradykinin. These observations indicate that, in normal pregnancy, both NO and EDHF can mediate responses in myometrial small arteries independently. Thus, in normal pregnancy, in the absence of the production of EDHF, NO acts as the endothelium-dependent vasodilator, and vice versa. In pre-eclamptic women, the absence of EDHF-mediated responses in myometrial small arteries could render women susceptible to endothelial dysfunction under conditions where

138

William R. Dunn et al.

NO production is impaired. This can happen when there is an increase in free radical production as may occur during both normal and compromised pregnancies (Hubel, 1999). The importance of EDHF in mediating responses in myometrial small arteries further highlights the importance of this endothelium-dependent mechanism in the resistance vasculature. Indeed, non-nitric oxide, non-prostanoid functional responses have been reported in several small arteries isolated from a number of vascular beds in humans (Pascoal et al., 1998; Coates et al., 2001), and these responses are associated with endothelium-dependent hyperpolarisation (Buus et al., 2000, Coleman et al., 2001). The nature of the EDHF mechanism in human blood vessels however, is less clear. In human myometrial small arteries, no relaxation response was apparent when the extracellular concentration of potassium was increased over the range 1–15 mM (Kenny et al., 2002), ruling out a role for potassium as an EDHF, as previously proposed in vessels isolated from animals (e.g. Edwards et al., 1998). Furthermore, neither anandamide nor methanandamide caused relaxation (Kenny et al., 2002) discounting a role for endocannabinoids as an EDHF (Randall and Kendall, 1998). The potential for other mechanisms in mediating EDHF-type responses, such as cytochrome-P450 products (Campbell et al., 1996) and functional gap junctions (Chaytor et al., 1998, Harris et al., 2002) has been limited to some extent, by the lack of freely available, selective inhibitors. With regard to the latter, some agents commonly used to interfere with gap junctions possess unrelated, but pertinent, actions (Tare et al., 2002). In the present study, several chemically distinct inhibitors of gap junction communication. The findings that, 18-␣-glycyrrhetinic acid (Taylor et al., 1998), carbenoxelone (Edwards et al., 1999), palmitoleic acid (Harris et al., 2000) and SR1417116A (Harris et al., 2000, 2002), abolished responses to bradykinin in human myometrial small arteries, provides strong evidence that functional gap junctions are involved in mediating the response to EDHF in these blood vessels. In summary, myometrial small arteries isolated from women whose pregnancies were complicated by pre-eclampsia, released only NO in response to bradykinin. By contrast, myometrial arteries isolated from women with normal pregnancies released both NO and EDHF in response to the peptide. The additional production of EDHF in these women may confer a compensatory mechanism to ensure appropriate endothelium-dependent responses in circumstances where NO production is compromised. The lack of such a redundant mechanism may leave vulnerable women susceptible to pre-eclampsia and associated endothelial dysfunction. The mechanism underlying EDHF-mediated responses in women with normal pregnancies involves functional gap junctions.

18 Free radical species and endothelium dysfunction during deoxycorticosterone-salt induced hypertension Ayotunde S.O. Adeagbo, Irving G. Joshua, Sunday O. Awe, Russell A. Prough and K. Cameron Falkner Acetylcholine elicits dilatation of rat mesenteric arteries by releasing endothelium-derived hyperpolarizing factor (EDHF) identified as a P450 metabolite of arachidonic acid, epoxyeicosatrienoic acid (EET). This study hypothesizes that free radical species generated as by-products of arachidonic acid metabolism contribute to impaired EDHF-mediated dilatation in deoxycorticosterone acetate (DOCA)-salt induced hypertension. EDHF dilatation was studied in arterial beds isolated from sham and DOCA-salt rats, and perfused (5 ml/min) with physiological salt solutions (37 ⬚C; bubbled with carbogen) and containing blockers of nitric oxide and prostanoid biosyntheses. EDHF-mediated dilatation to acetylcholine decreased significantly: 28.3 ⫾ 0.5% (DOCA-salt) vs 88.1 ⫾ 2.4% (sham); blood pressures were 180 ⫾ 5/166 ⫾ 5 (DOCA-salt) vs 128 ⫾ 3/116 ⫾ 6 (sham), respectively. Arterial microsomes convert 14 [C]-arachidonic acid to EETs in the presence of NADPH and generate free radical species as by-products; microsomes of DOCA-salt treated arteries form less (P ⬍ 0.05) EETs, but significantly more free radicals vs microsomes from sham rats. Treatment of DOCA-salt rats with an antioxidant, tempol, alleviates hypertension and improves EDHF-mediated dilatation to acetylcholine. It is concluded that: (1) generations of free radical species accompany metabolic conversion of arachidonic acid, and (2) free radicals modulate vascular reactivity to EDHF in DOCA-salt-induced hypertension.

1. INTRODUCTION The endothelium of perfused vascular beds release an endothelium-derived hyperpolarizing factor (EDHF) in response to agonists such as acetylcholine, bradykinin or histamine (Cowan and Cohen, 1991; Adeagbo and Triggle, 1993; Fulton et al., 1995). EDHF has been identified as a cytochrome P450 (CYP)-linked metabolite of arachidonic acid epoxyeicosatrienoic acid (EETs) in the porcine coronary arteries (Campbell et al., 1996), and in perfused heart (Fulton et al., 1998) and mesenteric arteries (Adeagbo et al., 2001) of the rat. Microsomes from mesenteric arteries metabolize arachidonic acid to epoxides and acetylcholine-induced EDHF-like dilatation of rat mesenteric arterial bed is accompanied by the release of EETs. These studies suggest that an EET is EDHF. Besides EETs, another candidate with considerable evidential support is K⫹ (Edwards et al., 1998). Thus, K⫹ released from endothelium of hepatic and mesenteric arteries activates vascular smooth muscle Na⫹/K⫹-ATPase and the inward rectifying potassium channels (KIR) to induce hyperpolarization. Endothelium-dependent vasodilatation is impaired in human (Miyoshi et al., 1997; Shimokawa, 1998; Lind et al., 2000) and animal forms of hypertension (Fujii et al., 1992; Kimura and Nishio, 1999), including DOCA-salt induced hypertension (Millette et al., 2000). The mechanisms proposed include: impaired nitric oxide synthase activity; increased

140 Ayotunde S.O. Adeagbo et al. production of endothelium-derived prostanoid contracting factors and endothelins (Luscher, 1990); increased generation or availability of superoxide anions (Somers et al., 2000), and decreased influence of EDHF (Fujii et al., 1992). These mechanisms vary in importance with different types of blood vessels and with different forms of hypertension. In the spontaneously hypertensive rats (SHR), for example, blunted endothelium-dependent relaxation found in the aorta is due to increased formation of contractile endoperoxides (Luscher, 1990; Cordellini, 1999) and/or altered NO3 distribution (Sullivan et al., 2002). In rat mesenteric arteries on the other hand, blunted hyperpolarization accounts for impaired endothelium-dependent vasodilatation (Fujii et al., 1992). CYP450 enzymes are present in endothelial cells. Biochemical reactions catalysed by CYPs including CYP epoxygenases, generate free radical species as by products (Puntarulo and Cederbaum, 1998). In fact, CYP2C9, identified as the EDHF synthase of coronary arteries, may be a physiologically relevant generator of reactive oxygen species in endothelial cells and modulates both vascular tone and homeostasis (Fleming et al., 2001). Thus, free radical species generated as by-products of CYP-mediated metabolism of arachidonic acid may contribute to the pathophysiology of both DOCA-salt induced hypertension and its impaired EDHFmediated vasodilatation. This hypothesis was tested by (a) investigating the effects of an antioxidant, tempol, on blood pressure and on EDHF-mediated vasodilatation of DOCA-salt treated rats, and (b) using a fluorescent probe, dichlorofluorescein diacetate, to demonstrate the generation of free radical species from arterial microsomes.

2. METHODS

2.1. Experimental hypertension Experiments were performed on age-matched groups of male Sprague-Dawley rats, which were divided into groups as follows: Group I (control): rats fed with normal rat chow and water drink. Group II (sham-salt): s.c. placebo pellet implants; high NaCl (8%)/K diet. Group III (DOCA-salt): s.c DOCA (100 mg, 21-day release) pellet implants; high NaCl/K diet. Group IV (DOCA-salt/Tempol): s.c DOCA (100mg, 21-day release) pellet implants; high NaCl/K diet. Groups II, III and IV rats were uninephrectomized surgically under pentobarbital (50mg/kg; i.p.) anesthesia before the treatments as stated before and all received 1% saline drink instead of regular water. Arterial blood pressures of all groups of rats were measured indirectly by a digital tail cuff plethysmography (Letica; Model 5001).

2.2. Perfusion studies Mesenteric vascular beds of rats were excised under pentobarbitone (60mg/kg, i.p.) anesthesia and perfused (5 ml/min; Masterflex peristaltic pump) in vitro with physiological salt solution at 37 ⬚C. The physiological salt solution is composed of the following (in mM): NaCl 118; KCl 4.7; CaCl2 2.5; KH2PO4 1.2; MgSO4 1.2; NaHCO3 12.5 and glucose 11.1. K⫹-free physiological salt solution was made by omitting KCl, and substituting an equimolar amount of NaH2PO4 for KH2PO4; the pH of all physiological salt solution after bubbling with 95% O2, 5% CO2 gas mixture was 7.4. Vascular beds were routinely perfused with physiological salt solution (PSS) containing L-NAME (1 ⫻ 10⫺4 M) plus indomethacin

Free radical species and endothelium dysfunction

141

(5 ⫻ 10⫺6 M) to block endogenous syntheses of nitric oxide and prostanoids, respectively. Perfusion pressures were recorded with Statham pressure transducers coupled to a Grass polygraph (Model 7H). In all relaxation studies, vascular tone was generated by an infusion of ␣1-adrenoceptor selective agonist, cirazoline (2–5 ⫻ 10⫺7 M). Acetylcholine, or KCl was injected into the perfused vascular beds in bolus doses and the dilator responses (expressed as % of prevailing tonus) were compared in the groups of rats used. The effects of acetylcholine were also examined in the absence, or during infusion of apamin (5 ⫻ 10⫺7 M) plus charybdotoxin (1 ⫻10⫺7 M) in order to ascertain the dependence or otherwise of the responses on the opening of KCa channels. Vasodilatation to KCl in normal physiological salt solution medium was examined during the continuous presence of 4-aminopyridine (3 ⫻ 10⫺4 M) plus penitrem A (1 ⫻ 10⫺7 M) to block voltage-gated (KV) and maxi-K potassium channels, respectively.

2.3. Preparation of arterial microsomes Mesenteric vessels were minced with a sharp pair of scissors in a beaker containing 20 ml 0.1 M phosphate plus 0.15% KCl buffer (homogenization buffer; pH 7.4). The buffer contained phenylmethylsulfonyl fluoride and a cocktail of six other protease inhibitors with broad specificity for the inhibition of aspartic, cysteine, and serine proteases as well as aminopeptidases. The homogenates were centrifuged for 15 min at 20,000 g; the supernatant layer was retained while the tissue debris (bottom) and floating fat (top) layers were discarded. The supernatant was centrifuged further for 1 h at 100,000 g. The cytosolic (top) layer was discarded while the microsomal pellets (bottom portion) were reconstituted in 0.5 ml homogenization buffer containing 15% glycerol (to preserve their functional integrity). This was eventually dispersed evenly with a glass pestle and aliquots made according to the desired experiments, or stored at ⫺80 ⬚C. The total protein content of the microsomal fractions was estimated (Bradford, 1976).

2.4. Microsomal metabolism of arachidonic acid The vessel microsomes were used to study the conversion of 14[C]-arachidonic acid to EETs. Three reaction tubes, each containing KHPO4 buffer (0.01mM, pH 7.4), were set up as follows: (a) 1mg microsomal protein plus 0.5␮Ci 14[C]-arachidonic acid plus 0mM reduced nicotinamide adenine dinucleotide phosphate (NADPH; negative control); (b) 1mg microsomal protein plus 0.5 ␮Ci 14[C]-arachidonic acid plus 1 mM NADPH; and (c) 1 mg microsomal protein plus 0.5 ␮Ci 14[C]-arachidonic acid plus 1 ⫻ 10⫺3 M NADPH plus 5 ⫻ 10⫺6 M clotrimazole. Indomethacin (1 ⫻ 10⫺6 M) was added to each of the tubes. The tubes were incubated in a shaking water bath (200 r.p.m) at 37 ⬚C under a stream of 95% O2–5% CO2 gas mixture to provide adequate oxygenation. At the end of 30min incubation, the reactions were terminated by acidification with 0.1 M formic acid (pH 3.5). Thereafter, metabolites of arachidonic acid were extracted with two volumes of ethyl acetate and the organic phase evaporated to dryness with N2 gas. The extract was reconstituted with ethanol followed by product separation by reverse phase high performance liquid chromatographic (HPLC) using a linear gradient of 1.25% per min from acetonitrile:water:acetic acid (50:50:0.1, v/v) to acetonitrile:acetic acid (100:0.1, v/v) solvent phase over a 25-min period at a flow rate of 1.0ml/min. The elution profile of the radioactive metabolites of arachidonic acid was detected and counted using an online radioactive detector (Radiomatic Instruments & Chem. Co., Meriden, CT). The identity of each metabolite was confirmed by the migration characteristics of authentic standards.

142

Ayotunde S.O. Adeagbo et al.

2.5. Microsomal production and spectrofluorimetric determination of free radicals Production of free radical species from arachidonic acid was investigated with microsomes obtained from control, sham-salt or DOCA-salt arteries, which were set up in reaction cuvettes containing 40 mM potassium phosphate buffer (pH 7.4) plus 1 ⫻ 10⫺3 M sodium azide. The reaction mixtures were set up as follows: (1) 2 mg boiled microsomal protein plus 5 ⫻ 10⫺4 M NADPH plus 1 ⫻ 10⫺3 M arachidonic acid [negative control]; (2) 2 mg microsomal protein plus 5 ⫻ 10⫺4 M NADPH plus 1 ⫻ 10⫺3 M arachidonic acid [control microsomes]; (3) 2 mg microsomal protein plus 5 ⫻ 10⫺4 M NADPH plus 1 ⫻ 10⫺3 M arachidonic acid [DOCA-salt microsomes]. Each reaction cuvette was incubated for 10 min at 37 ⬚C in a final volume of 2 ml and in the presence of 2⬘,7⬘-dichlorofluorescein diacetate (DCFDA; 5 ⫻ 10⫺6 M). The fluorescence generated by the reaction mixture was monitored in a Hitachi spectrofluorometer at excitation and emission wavelengths of 488 and 525 nm, respectively. Corrections for autofluorescence were made by inclusion of parallel blanks (i.e. assay mixture without microsomes, or without arachidonic acid) in each experiment.

2.6. Drugs Acetylcholine hydrobromide, arachidonic acid, clotrimazole, NADPH, N-nitro-L-arginine methyl ester, indomethacin, penitrem A, and 4-aminopyridine were purchased from Sigma Chem. Co. (St Louis, MO, USA). Cirazoline hydrochloride was purchased from Research Biochem (Natick, MA., USA). The Protease Inhibitor Cocktail Set III and phenylmethylsulfonyl fluoride were purchased from Calbiochem-Novabiochem Corp. (La Jolla, CA., USA); 2⬘,7⬘DCFDA was purchased from Molecular Probes, Eugene, OR.

2.7. Data analysis Changes in perfusion pressure were expressed as a percentage of the arterial perfusion pressure before the administration of a vasodilator agent. Values are expressed as mean ⫾ SEM, and differences between the mean values were compared using either Student’s t-test for paired observations, or ANOVA (multiple comparisons). The difference between means was considered significant when P was less than 0.05. 3. RESULTS The systolic, as well as diastolic blood pressures were significantly (P⬍0.05) higher in uninephrectomized rats that received subcutaneous implants of DOCA (100mg; 21-day release) pellets and fed with potassium-supplemented high (8%) sodium chloride diet, and 1% saline drink. The age-matched control and sham-salt treated rats did not develop high blood pressure (Table 18.1). The body mass of DOCA-salt rats was not different from control and sham-salt rats. However, heart mass, and heart mass to body mass ratios were significantly (P⬍0.05) higher in DOCA-salt treated rats vs control and sham-salt rats (see Table 18.1).

Free radical species and endothelium dysfunction

143

Table 18.1 Blood pressure, heart mass, body mass and heart-to-body mass ratio in DOCA-salt rats treated for three weeks and age-matched salt-treated sham, and control untreated rats Parameters

Control

Sham-salt

DOCA-salt

Systolic/Diastolic pressures (mmHg) Heart mass (mg) Body mass (g) Heart/Body mass ratio (mg/g) n

125 ⫾ 7/108 ⫾ 3

137 ⫾ 3/106 ⫾ 6

194 ⫾ 5/166 ⫾ 4*

1398 ⫾ 28 350 ⫾ 6 3.99 ⫾ 0.50

1440 ⫾ 40 358 ⫾ 10 4.02 ⫾ 0.48

1860 ⫾ 32* 301 ⫾ 9* 6.18 ⫾ 0.22*

37

30

32

* P ⬍ 0.05.

Figure 18.1 EDHF-mediated dilator responses elicited by acetylcholine in mesenteric vascular beds. EDHF-mediated responses were recorded during combined blockade of nitric oxide synthase and cyclooxygenases with L-NAME (1 ⫻ 10⫺4 M) plus indomethacin (5 ⫻ 10⫺6 M), respectively. Each data point on the graphs represents the mean ⫾ SEM, (n ⫽ 8). The asterisks (*) denote statistically significant differences (P ⬍ 0.05). Note blockade of EDHF-mediated vasodilatation by apamin (5 ⫻ 10⫺7 M) plus charybdotoxin (ChTx; 1 ⫻ 10⫺7 M).

3.1. Responses to acetylcholine and KCl Acetylcholine (1⫻10⫺9–1⫻10⫺5 M) elicited dose-dependent decreases in perfusion pressure (index of vasodilatation) during continuous blockade of nitric oxide synthase and cyclooxygenases with L-NAME (1⫻10⫺4 M) and indomethacin (5⫻10⫺6 M), respectively. The dilator responses to the compound were abolished by a combination of apamin (5 ⫻ 10⫺7 M) plus penitrem A (1 ⫻ 10⫺7 M) (Figure 18.1). Acetylcholine-induced EDHF dilatations were significantly reduced in vascular beds of DOCA-salt induced hypertensive compared to sham-salt and control groups (Figure 18.2). However, pD2 values (index of tissue

144

Ayotunde S.O. Adeagbo et al.

Figure 18.2 Acetylcholine-induced EDHF responses in perfused mesenteric beds of control (normal feed and water drink), sham-salt (high-salt diet plus 1% salt drink) and DOCA-salt (high salt diet, 1% salt drink plus DOCA implants) rats. Each data point on the graphs represents the mean⫾SEM (n⫽9). The asterisks (*) denote statistically significant differences (P ⬍ 0.05).

sensitivity) for the agonist were statistically similar for vessels of all groups: 9.87 ⫾ 0.13 (control), 9.86 ⫾ 0.13 (sham-salt) and 9.56 ⫾ 0.15 (DOCA-salt) rats, respectively. KCl (1 ⫻ 10⫺4–1 ⫻ 10⫺1 M) failed to elicit dilatation of perfused mesenteric vascular beds during perfusion with normal physiological salt solution, or with physiological salt solution containing 4-aminopyridine (3 ⫻ 10⫺4 M) and penitrem A (1 ⫻ 10⫺7 M) to block KV and maxi-K potassium channels, respectively. However, in vascular beds perfused with K⫹-free physiological salt solution, KCl (1 ⫻ 10⫺4–1 ⫻ 10⫺1 M) initiated dose-dependent dilatations, which were abolished by ouabain (5 ⫻ 10⫺5 M). This concentration of ouabain attenuated acetylcholine-induced EDHF responses by 38.0 ⫾ 0.5%. Vascular beds of DOCA-salt treated rats exhibited reduced dilatation to KCl compared to control arteries in K⫹-free media (Figure 18.3).

3.2. Modulation by anti- and pro-oxidant compounds Treatment of DOCA-salt rats with an antioxidant, tempol (15 mg/kg, i.p.) for 21 days significantly lowered elevated blood pressure and also restored EDHF-mediated vasodilatation (Figure 18.4). Treatment with the compound did not alter the reactivity of control rats. Conversely, infusion of tert-butyl hydroperoxide (t-BOOH; 1 ⫻ 10⫺6 M), a membrane permeant lipid peroxidant did not alter cirazoline-induced tone of perfused vascular beds. However, the compound caused marked attenuation of acetylcholine-induced EDHF dilatation of normotensive vascular beds; the dilator response initiated by sodium nitroprusside (1 ⫻ 10⫺5 M), a nitric oxide donor, was not altered (Figure 18.5).

Figure 18.3 Original recordings of the effects of KCl in mesenteric vascular beds during perfusion with normal physiological salt solution (top panel), normal physiological salt solution plus ouabain (5 ⫻ 10⫺5 M; middle panel), or during perfusion with K⫹-free physiological salt solution (bottom panel). KCl effects were tested in arterial beds constricted with cirazoline (0.3–1 ⫻ 10⫺6 M). The perfusion salt solution (normal or K⫹-free) also contained 4-aminopyridine (3 ⫻ 10⫺4 M) and penitrem A (1 ⫻ 10⫺7 M) to block KV and maxi-K potassium channels, respectively.

Figure 18.4 Effects of treatment with an antioxidant, tempol (15 mg/kg, i.p., 21 days) on systolic hypertension (bar graphs) and on EDHF-mediated dilation of mesenteric arterial bed elicited by acetylcholine (line graph). Each data point on the bar graph represents the mean ⫾ SEM, (n ⫽ 13); each data point on the line graph represents the mean ⫾ SEM (n ⫽ 7). The asterisks (*) denote statistically significant differences (P ⬍ 0.05).

146

Ayotunde S.O. Adeagbo et al.

Figure 18.5 Dose-responses to acetylcholine and their attenuation by t-BOOH; 1 ⫻ 10⫺6 M infusion into control normotensive perfused rat mesenteric arterial bed. The bar graph (insert) indicates vasodilator responses to sodium nitroprusside (SNP, 1 ⫻ 10⫺6 mole) before (open bar) and during infusion (solid bar) of 1 ⫻ 10⫺6 M t-BOOH. Each data point on the graph represents the mean ⫾ SEM, (n ⫽ 5); each bar column represents the mean ⫾ SEM (n ⫽ 5). The asterisks (*) denote statistically significant differences (P ⬍ 0.05).

3.3. Microsomal metabolism of

14

[C]-arachidonic acid

In the presence of NADPH (1 mM), molecular oxygen and blockade of cyclooxygenases with indomethacin (1 ⫻ 10⫺5 M), microsomes isolated from mesenteric blood vessels convert 14[C]-arachidonic acid to products with reverse-phase HPLC retention times (RTs) of 5–7, 8–9 and 12–14 minutes, respectively. These RTs correspond to those of authentic dihydroxyeicosatrienoic acids (DHETs), 20-hydroxyeicosatetraenoic acid (20-HETE) and EETs, respectively. Epoxygenase activity, taken as a summation of the % total radioactivity that make up the DHET plus EET peaks, was blocked by clotrimazole (5 ⫻ 10⫺5 M), an epoxygenase blocker and significantly reduced in microsomes from DOCA-salt hypertensive vs control normotensive rats (Figure 18.6). 3.4. Microsomal production of free radical species There was no measurable fluorescence in the reaction mixtures of DCFDA and phosphate buffer over many hours, or following the addition to arachidonic acid and NADPH. It took the presence of viable microsomes of control, sham-salt or DOCA-salt arteries, with arachidonic acid and NADPH to emit fluorescence from the addition of DCFDA to reaction cuvettes; microsomes denatured by boiling did not emit fluorescence. Mesenteric arterial microsomes of DOCA-salt rats produced significantly stronger fluorescence compared to those of control and sham-salt rats (Figure 18.7(A)). Treatment of rats with ␤-naphthoflavone (␤-NF; 60 mg/kg, i.p., 4 days), an inducer of CYP1/2 families, significantly increased fluorescence emission in DOCA-salt rat microsomes, but not in control microsomes (Figure 18.7(B)).

Figure 18.6 NADPH-dependent metabolism of 14[C]-arachidonic acid in control (open bar column), DOCA-salt (hatched bar column) microsomes of rat mesenteric blood vessels, and microsomes treated with clotrimazole (5 ⫻ 10⫺6 M, solid bar column). Each bar column represents the mean ⫾ SEM, (n ⫽ 12 rats). The asterisks (*) denote statistically significant differences (P ⬍ 0.05) from control.

Figure 18.7 Arterial microsomal generation of free radical species from arachidonic acid. (A) Control (C), sham-salt (Sham), DOCA-salt (DOCA) and boiled microsomes, respectively. (B) Control (C), DOCA-salt (D), and microsomes from ␤-naphthoflavone (C ⫹ ␤-NF or D ⫹ ␤-NF) treated rats. Each bar column represents the mean ⫾ SEM, (n ⫽ 6 rats). The asterisks (*) denote statistically significant differences (P ⬍ 0.05) from control, sham or boiled (A), or from control or C ⫹ ␤-NF (B).

148

Ayotunde S.O. Adeagbo et al.

4. DISCUSSION The present study shows that: (a) EDHF-mediated dilatation in response to acetylcholine is impaired in DOCA-salt induced hypertension; the impairment is apparently unrelated to muscarinic receptor function since the pD2 for acetylcholine is not altered; (b) treatment of DOCA-salt rats with an anti-oxidant, tempol, alleviates hypertension and improves EDHFmediated vasodilatation to acetylcholine. Conversely, infusion of a lipid pro-oxidant compound, t-BOOH to constricted vascular beds of control rats attenuates EDHF, but not nitric oxide-mediated vasodilatation; and (c) microsomes from mesenteric vessels form EETs, as identified by HPLC profiles, and fluorescence, index of free radical species that oxidizes fluorescein product of DCFDA, from arachidonic acid. Microsomes of DOCA-salt hypertensive rat arteries form significantly less EETs vs those of control and sham-salt normotensive rat arteries. Conversely, DOCA-salt rat microsomes produced significantly higher amounts of free radical species compared to control and sham-salt rats. This study thus indicates a dual role for arachidonic acid as the putative precursor for EDHF (EETs), and as a source of free radical species that modulates acetylcholine-induced EDHF vasodilatation in rat mesenteric arteries. Acetylcholine elicits dilatation of perfused rat mesenteric arterial bed mainly by releasing an EDHF, which is insensitive to nitric oxide synthase inhibitors, and relaxes vascular smooth muscle by opening KCa channels (Adeagbo and Triggle, 1993). The acetylcholineinduced, EDHF vasodilatation is attenuated in vascular beds of DOCA-salt hypertensive rats compared to control and sham-salt rat vessels. The pD2 values, an index of vessel sensitivity to the agonist, did not change in the three groups of rats. Thus, it is unlikely that the dysfunction in EDHF-mediated vasodilatation occurred at the level of muscarinic cholinoceptors. An alternative hypothesis that vasodilatation induced by agonists releasing EDHF is reduced in DOCA-salt hypertension because the formation of the factor is impaired was tested. This hypothesis is founded on the postulate that EDHF released by rat mesenteric arteries is a cytochrome P450 derived metabolite of arachidonic acid, an epoxyeicosatrienoic acid, (Adeagbo et al., 2001). Arterial microsomal pellets metabolise 14[C]-arachidonic acid to products with similar reverse-phase HPLC profiles as dihydroxyeicosatrienoic acid (DHET; retention time 5–6 min), 20-hydroxyeicosatetraenoic acid (20-HETE; retention time 8–9 min), and epoxyeicosatrienoic acids (EETs; retention time 14–16 min). Epoxygenase activity, determined as the sum of EET and DHET formation, was reduced in microsomes of DOCA-salt rat mesenteric arteries compared to control arterial microsomes. Thus, the formation of epoxyeicosatrienoic acids is reduced, and this reduction probably accounts for decreased EDHF-mediated responses to acetylcholine in DOCA-salt hypertension. Endothelium-dependent relaxation is impaired in many forms of experimental (Luscher, 1990; Fujii et al., 1992; Kimura and Nishio, 1999; Mombouli and Vanhoutte, 1999), as well as in human (Panza et al., 1990; Treasure et al., 1992; Miyoshi et al., 1997; Houghton et al., 1998) hypertension. The endothelium dysfunction has been attributed to decreased endotheliumderived nitric oxide (EDNO) levels in Dahl salt-sensitive rats exposed to high salt intake, due to a deficiency in substrate for nitric oxide synthesis (Chen and Sanders, 1991). However, using vasoconstrictor responses to intra-arterial infusion of L-NMMA as an index of endothelial function does not reveal an impairment of EDNO in salt-sensitive and saltresistant patients (Miyoshi et al., 1997). In the present study, no impairment in the EDNOmediated component of the dilatation elicited by acetylcholine was obvious during DOCA-salt induced hypertension. However, the EDHF-mediated component of vasodilatation elicited by this agonist is attenuated in DOCA-salt induced hypertension. This finding

Free radical species and endothelium dysfunction

149

agrees with those reported for mesenteric arteries of the SHR (Fujii et al., 1992) and in saltsensitive hypertensive patients (Miyoshi et al., 1997). The elevated production of free radical species contributes to the onset or progression of spontaneous (Kerr et al., 1999), Dahl salt-sensitive (Swei et al., 1999) and mineralocorticoidsalt induced (Somers et al., 2000; Beswick et al., 2001) hypertension in rats, as well as development of hypertension in humans (Kumar and Das 1993; Lacy et al., 2000). Treatment of DOCA-salt rats with tempol, a hydrophilic nitroxide antioxidant (Mitchell et al., 1990), alleviates hypertension and restores acetylcholine-induced EDHF-mediated vasodilatation. This is perhaps the most intriguing finding of the present study, as it suggests that free radical species modulate EDHF-mediated dilatation of rat mesenteric arteries. This assertion is corroborated by the finding that infusion of t-BOOH, a membrane permeant lipid pro-oxidant (Garcia-Cohen et al., 2000) to perfused vascular beds of control normotensive rats attenuates EDHF dilatation elicited by acetylcholine, but not that caused by sodium nitroprusside, a nitric oxide donor. t-BOOH elicits oxidative stress via the generation of lipid (L⫺), lipid peroxyl (LOO⫺) radicals and superoxide anions (Schilling and Elliot, 1992). Thus, oxidative stress elicited by DOCA-salt induced hypertension or by the infusion of a pro-oxidant alters EDHF vasodilatation of perfused rat mesenteric arteries. The alleviation by tempol, of DOCA-salt induced hypertension as observed in this study agrees conceptually with a role for free radical species in the pathogenesis of the disease. Treatment with tempol normalizes arterial blood pressure and renal vascular resistance in the SHR (Schnackenberg et al., 1998, 1999), attenuates systolic blood pressure, and suppresses renal nuclear factor-␬B (NF-␬B) binding activity and aortic superoxide accumulation in mineralocorticoid-induced hypertension (Beswick et al., 2001). Chronic treatment with tempol prevents vascular remodelling and progression of hypertension in salt-loaded stroke prone SHRs (Park et al., 2002). The blood pressure lowering effects of tempol have been attributed to a decreased generation of superoxide anions (O⫺ 2 ) (Park et al., 2002). Endothelium-dependent vasodilatation of rat mesenteric arteries is accompanied by the release of arachidonic acid (Adeagbo et al., 2001), and enzymatic metabolism of this fatty acid generates free radical species including lipid (L⫺), lipid alkoxyl (LOO⫺), lipid peroxides and superoxide anions. These radical species are also all produced during lipid peroxidation. In order to establish that arachidonic acid was the source of free radical species, the fluorescence generated by the oxidation product of a non-fluorescent dye, DCFDA was monitored. This dye enters cells or microsomes, becomes de-esterified by intramicrosomal esterases to dichlorofluorescein (DCF), which is then oxidized by free radical species to give measurable fluorescence (LeBel and Bondy, 1990; Bondy and Naderi, 1994). Viable microsomes and arachidonic acid are obligatory for measurable fluorescence to occur. No fluorescence was observed with the dye DCFDA in phosphate buffer without microsomes, or with boiled microsomes. Thus, the source of free radical species that is modulated by the antioxidant tempol to bring about alleviation of high blood pressure and restoration of EDHF vasodilatation is the metabolism of arachidonic acid. This fatty acid is metabolized by three enzyme systems: cyclooxygenases, lipoxygenases and cytochrome P450 (CYPs). Although cyclooxygenases and CYPs are microsomal enzymes, only the CYPs use NADPH as a co-factor. The requirement of NADPH, as well as the enhancement by the CYP1/2 inducer, ␤-napthoflavone, argue strongly in favour of CYPs as the enzyme system that catalyses the production of free radical species from arachidonic acid in microsomes of the rat mesenteric artery. An elevated free radical status in DOCA-salt hypertension (Somers et al., 2000) may drive EDHF precursor (arachidonic acid), toward production of vasoconstrictors. The low

150

Ayotunde S.O. Adeagbo et al.

EDHF biosynthetic activity, with the elevated vasoconstrictor formation, together translates to the increased total peripheral resistance and hypertension. Tempol possibly reacts with free radical by-products of the metabolism of arachidonic acid to terminate the chain reactions associated with lipid peroxidation (Zhang et al., 1998), or with O⫺ 2 to prevent non-enzymatic oxidation of arachidonic acid to constrictor products such as isoprostanes (Banerjee et al., 1992; Takahashi et al., 1992). The present study shows that EDHF-mediated dilatation of rat mesenteric arteries is impaired in DOCA-salt induced hypertension. The impairment is due to decreased formation of EETs rather than to a dysfunction at the level of muscarinic cholinoceptors. Second, infusion of a pro-oxidant lipid tert-butyl hydroperoxide to normotensive control arteries mimics impairment of endothelial function. Third, treatment with an antioxidant, tempol, alleviates DOCA-salt hypertension and significantly improves EDHF-mediated vasodilatation. These data suggest that free radical species, probably emanating from lipid metabolism and/or peroxidation modulates EDHF-mediated vascular regulation in DOCA-salt induced hypertension. ACKNOWLEDGMENTS The work was supported by an American Heart Association (Ohio Valley Affiliate) Grant-in-Aid, and in part to the Jewish Hospital Foundation Grant awarded to ASOA.

19 EDHF involvement in skin pressure-induced vasodilatation Ambroise Garry, Sandra Merzeau, Bérengère Fromy and Jean Louis Saumet

At least three different vasodilator agents are synthesised by the endothelium upon exposure to mechanical forces or to receptor-dependent agonists: nitric oxide (NO), prostaglandins and endothelium-derived hyperpolarising factor (EDHF). Cutaneous pressure-induced vasodilatation is a neuronal response to locally applied pressure discovered in humans and in rats. This new mechanism results from a complex response originating from capsaicin-sensitive skin sensory fibres and local secretion NO and prostaglandins. The aim of the present study was to examine the potential role of the EDHF in the pressure-induced vasodilatation development in treated rats with a combined infusion of charybdotoxin plus apamin and in controls. Cutaneous blood flow was measured by laser-Doppler flowmetry in response to a progressive local pressure applied to the skin. In treated rats as well as in controls, the skin vascular conductance increased with increments of local pressure. The vasodilator capacity was not altered in rats treated with charybdotoxin plus apamin compared to controls. In conclusion, the present study indicates that when the NO pathway is intact, there is no or little implication of EDHF in the development of cutaneous pressure-induced vasodilatation in rats.

1. INTRODUCTION The endothelium plays an important role in the control of vascular tone. At least three different vasodilator agents are synthesised by endothelial cells upon exposure to mechanical forces, such as shear stress, or to receptor-dependent agonists, such as acetylcholine and bradykinin. Whereas the mechanism of vasodilatation induced by nitric oxide (NO) and prostaglandins is well known, the nature and mechanism of action of the third vasodilator, the endothelium-derived hyperpolarising factor (EDHF), is still controversial (Waldron and Cole, 1999; Brandes et al., 2000). However, it is largely admitted that EDHF release effects are abolished by charybdotoxin, apamin or a combination of charybdotoxin and apamin (Corriu et al., 1996; Zygmunt and Hogestatt, 1996; Plane et al., 1997; Edwards and Weston, 1998; Edwards et al., 1998; Waldron and Cole, 1999). Cutaneous pressure-induced vasodilatation is a neuronal response to locally applied pressure discovered in humans (Fromy et al., 1998) and in rats (Fromy et al., 2000b). This new mechanism results from a complex response originating from capsaicin-sensitive skin sensory fibres and the local secretion of prostaglandins and NO. No information has been published on EDHF in the development of pressure-induced vasodilatation. Therefore the aim of the present study was to determine the possible involvement of EDHF in that response.

152

Ambroise Garry et al.

2. METHODS

2.1. Animal Instrumentation The studies were performed in 14 female and 7 male Wistar rats (200–280 g). To present an hairless area for the laser-Doppler flow measurements, the hair was removed from the skulls of the animals with a depilatory lotion. This was performed at least two days prior to the experiment to prevent skin irritation from confounding the results. Animals were housed in a regulated environment with a constant ambient temperature of 24 ⬚C. Procedures for the maintenance and use of the experimental animals were carried out in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-23, revised 1996). For the experiments, rats were anaesthetised by intraperitoneal injection of thiopental (50 mg/kg body weight). The level of anaesthesia was determined by testing eye reflexes. A catheter was inserted into the femoral artery to measure the arterial blood pressure. The left femoral vein was cannuled for the administration of antagonists or vehicle and the right vein was cannuled for the injection of the agonist. The rats were placed in an incubator (Médipréma, Chambray-Les-Tours, France) kept at 30 ⬚C to allow for thermal stability, which was controlled by a rectal thermometer throughout the experiment. They were studied in the prone position and their head was fixed on a frame. The local pressure application was performed on the skull through a laser-Doppler probe (12.6 mm2 circular contact surface), which measured simultaneously the skin blood flow at the pressure application site. The rats were separated into two groups.

2.2. Experimental procedure for assessment of the pressure-induced vasodilatation A weighbridge was designed to hold a laser-Doppler probe (PF408, Periflux, Perimed, Sweden), connected to a laser-Doppler flowmeter (PF4001 Master, Periflux, Perimed, Sweden). The weighbridge was carefully equilibrated and the probe was positioned in the middle of the hairless area of the rat’s skull to apply the progressive local pressure and to measure simultaneously the cutaneous blood flow at the pressure application site (Fromy et al., 2000a). Data collection began with a one-minute control period prior to the onset of increasing pressure application at a rate of 11.1 Pa/sec. The signals were continuously recorded at a 20 Hz sample frequency using a computerised data acquisition system (MP100 WPI, Biopac, Santa Barbara, California, USA) and averaged every 30 s to reduce the instantaneous variability of the signals. At the end of the experiment, the rat was sacrificed by an intravenous lethal dose of thiopental. The skin blood flow and arterial blood pressure values were recorded for 5 min following the sacrifice of the rat to assess the minimum value of the measurements. This minimum value was subsequently subtracted from corresponding measured values.

2.3. EDHF involvement in the development of pressure-induced vasodilatation The role of the EDHF was examined in the development of pressure-induced vasodilatation by an intravenous infusion of a combination of charybdotoxin and apamin. Charybdotoxin is a peptide produced by the scorpion Leirus quinquestriatus, which is selective for

EDHF involvement in skin pressure-induced vasodilatation

153

Ca2⫹-activated K⫹ (K-Ca2⫹) channels (Nelson and Quayle, 1995; Beech, 1997). Charybdotoxin inhibits mainly high but also small conductance K-Ca2⫹ channels (Beech, 1997). It was used at a dose of 7.5 ␮g/kg/min (Tominaga et al., 1988; Berg and Koteng, 1997; Yonehara and Takiuchi, 1997). Apamin, a bee venom peptide, is a specific inhibitor of small conductance K-Ca2⫹ channels (Nelson and Quayle, 1995; Yonehara and Takiuchi, 1997). Apamin was used at a dose of 25 ␮g/kg/min (Yonehara and Takiuchi, 1997; Jeremy and McCarron, 2000). The infusion began 10 min before the start of the experiment. To verify the effectiveness of the combined administration of charybdotoxin plus apamin in rats, a bolus of 0.4 mg/kg of hydralazine was injected (El-Mas and Abdel-Rahman, 1999) in the right femoral vein at the end of experiment. Hydralazine is an agonist of K-Ca2⫹ channels (Bang et al., 1998). The injection of hydralazine produces a hypotension resulting from relaxation of the smooth muscle of arteries and arterioles.

2.4. Data analysis Baseline values were calculated as the average over the one-minute control period prior to the onset of local pressure application. One-way ANOVA with repeated measures was performed on mean arterial blood pressure and rectal temperature and the coefficient of variation was calculated to determine the relative measure of variability in each data set. The coefficient of variation is the ratio of the standard deviation to the mean, expressed as a percentage. These two parameters were compared at baseline between rats treated with charybdotoxin plus apamin and controls using a Mann–Whitney test. All laser-Doppler flow values were expressed in arbitrary units of flow since the flowmeter signal is not calibrated in conventional flow units. Cutaneous vascular conductance was calculated as the ratio of cutaneous blood flow to mean arterial blood pressure (expressed in arbitrary units/mmHg). Baseline conductance, maximal conductance, its corresponding percent increase from baseline and the locally applied pressure for which maximal conductance occurred were expressed as mean ⫾ SEM. These results were compared between groups using a Mann–Whitney test. Within each group, maximal conductance was compared to baseline conductance by a Wilcoxon test. A P value less than 0.05 was considered statistically significant. 3. RESULTS Mean arterial blood pressure and rectal temperature did not change significantly in any group throughout the experiments (Table 19.1). Table 19.1 Mean arterial blood pressure and rectal temperature in the rat. Data shown as mean ⫾ SEM. The minimal and maximal values of the coefficient of variation (CV) are expressed in percent (%). No significant difference (ns) in arterial blood pressure and rectal temperature was observed between the two groups

Charybdotoxin plus apamin Controls

Mean arterial blood pressure (mmHg)

Rectal temperature (⬚C)

123.2 ⫾ 4.4 (9.8 ⬍ CV ⬍ 12.1%) ns 116.2 ⫾ 3.9 (10.2 ⬍ CV ⬍ 14.3%)

37.6 ⫾ 0.2 (1.3 ⬍ CV ⬍ 2.0%) ns 37.3 ⫾ 0.2 (1.4 ⬍ CV ⬍ 1.5%)

154

Ambroise Garry et al. 3.0 Baseline

2.5 Cutaneous conductance (arbitrary units/mmHg)

Controls (n = 11) Charybdotoxin plus apamin(n = 10)

2.0 1.5 1.0 0.5 0.0

-0 .0 -0 .9 -0 .8 -0 .7 -0 .6 -0 .5 .4 .3 .2 0

-1.2 -1.1 -1 -0 -0

1

2 3 4 5 Locally applied pressure (kPa)

6

7

Figure 19.1 Cutaneous vascular conductance obtained with 11.1 Pa/sec local external pressure application in the rats treated with charybdotoxin plus apamin compared with control rats. Baseline values over the one-minute control period prior to the local pressure application are represented outside the scale of pressure. Data shown as mean ⫾ SEM.

Hydralazine-induced hypotension was significantly attenuated by administration of charybdotoxin plus apamin. The injection of hydralazine induced a decrease in mean arterial blood pressure of 14.9 ⫾ 1.5% in rats treated with charybdotoxin plus apamin and of 26.7 ⫾ 4.0% in controls (P ⬍ 0.05). In rats treated with charybdotoxin plus apamin, the cutaneous vascular conductance increased with increments of local pressure from baseline to 0.6 ⫾ 0.1 kPa (P ⬍ 0.01), representing a maximal percent increase from baseline of 56.5 ⫾ 11.1% (Figure 19.1). In control rats, the cutaneous vascular conductance increased from baseline to 0.8 ⫾ 0.1 kPa (P ⬍ 0.001), representing a maximal percent increase from baseline of 59.5 ⫾ 8.4%. The baseline conductance was not different between groups (P ⬎ 0.05) but the maximal conductance was lower in rats treated with charybdotoxin plus apamin compared to controls (P ⬍ 0.05). The maximal percent increase from baseline was not different between the two groups (P ⬎ 0.05). With further increasing pressure, the cutaneous conductance decreased progressively in the two groups. 4. DISCUSSION The present study shows that although the maximal conductance was lower in rats treated with charybdotoxin plus apamin compared to controls, their vasodilator capacity was not altered. Thus, pressure-induced vasodilatation in the rat cutaneous microcirculation is not or little influenced by EDHF. Although EDHF appears to be of primary importance in the regulation of vascular resistance in small resistance arteries (Garland et al., 1995; Quilley et al., 1997), the pressure-induced vasodilatation, which is the result of small vessels vasomotricity, is EDHF-independent. This vasodilator response to local pressure is totally inhibited by an acute administration of N␻-nitro-L-arginine, indicating that NO plays a major role in the

EDHF involvement in skin pressure-induced vasodilatation

155

development of pressure-induced vasodilatation (Fromy et al., 2000b). The contribution of EDHF to endothelium-dependent vasodilatation under physiological condition is difficult to assess, because NO exerts an inhibitory influence on the EDHF-mediated component (Bauersachs et al., 1996; Nishikawa et al., 2000). Indeed, small doses of the NO donor were able to inhibit in vivo EDHF-mediated vasodilatation (Bauersachs et al., 1996; Nishikawa et al., 2000). Since EDHF-mediated responses can be unmasked when NO pathway is compromised, the lack of EDHF influence in the development of pressure-induced vasodilatation in the present study can be explained by the presence of NO under physiological conditions in the microcirculation of the rat skin. EDHF is able to completely compensate for the lack of NO (Brandes et al., 2000), showing that EDHF serves as a back-up mechanism during conditions of compromised NO (Bauersachs et al., 1996; McCulloch et al., 1997; Nishikawa et al., 2000; Golding et al., 2001; Ruiz-Marcos et al., 2001). The pressure-induced vasodilatation occurs in endothelial NO synthase knockout mice (unpublished preliminary data). However the maximal percent increase in skin blood flow tends to be reduced following a combined administration of charybdotoxin plus apamin in these knockout mice. This is in accordance with a compensatory role for EDHF in the development of pressure-induced vasodilatation when the NO pathway is compromised. In conclusion, the present study indicates that when the NO pathway is intact, there is no or little implication of EDHF in the development of pressure-induced vasodilatation in the rat skin.

20 N-acetylcysteine and immobilization stress attenuate dysregulation of the endothelium-dependent coronary vascular tone induced by acute hemorrhage L.Eu. Belyaeva, V.I. Shebeko and A.P. Solodkov A disturbance of coronary vessels tone following severe acute hemorrhage which may potentiate cardiac contractility abnormalities in particular is characterized by a decrease in coronary myogenic tone and an increase in shear–stress induced coronary vasodilatation. The purpose of this study was to assess the effects caused by shifts of the myo-cardial redox-state on endothelium-dependent regulation of coronary tone after a two-hour posthemorrhagic arterial hypotension (mean arterial pressure, 40–50 mmHg). Experiments were performed on isolated rat hearts perfused at constant perfusion pressure and at constant coronary flow in the absence ⫺ ⫺ or presence of the inhibitor of NO-synthase L-NAME. The serum concentration of NO3 /NO2 was measured. The degree of oxidative stress in myocardium was evaluated by determining the malondialdehyde concentration and the reduced glutathione content. In order to alter myocardial redox-state the rats were pretreated with antioxidant N-acetylcysteine, a precursor of reduced glutathione (40 mg/kg, intraperitoneally), or they underwent a preliminary 6-h immobilization stress. N-acetylcysteine pretreatment resulted in the diminishing of the excessive volume velocity of coronary flow, attenuating shear-stress induced coronary ⫺ ⫺ vasodilatation and preventing the elevation of NO3 /NO2 serum concentration after a 2-h posthemorrhagic arterial hypotension. A previous 6-h immobilization which increased malondialdehyde concentration and decreased reduced glutathione in myocardium in part ⫺ ⫺ prevented the disturbances of coronary vascular tone and diminished the NO3 /NO2 serum content after hemorrhage. The alteration in myocardial redox-state may affect the endotheliumdependent regulation of coronary vessel tone in severe acute hemorrhage.

1. INTRODUCTION The integrity of the mechanisms regulating coronary vascular tone and myocardial contractility is a main condition which determines the survival rate during severe acute hemorrhage (Golden and Jane, 1970). However, after a 2-h posthemorrhagic arterial hypotension myogenic tone of coronary vessels, coronary autoregulation and myocardial contractility decreased. Such disturbances of the regulation of coronary vessels tone led to inadequate coronary flow and hyperperfusion of the isolated rat heart after hemorrhage (Solodkov et al., 2001). Endothelium-derived relaxing factors play an important role in the local regulation of coronary blood flow (Furchgott and Zawadzki, 1980; Vanhoutte et al., 1995; Corrio et al., 1998; Komaru et al., 2000). Because of the impact of the redox-state on different cellular functions (Kamata and Hirata, 1999; Schafer and Buettner, 2001), changes of the myocardial redox-state to the more oxidized or reduced conditions may influence on the endotheliumdependent regulation of coronary vascular tone during acute hemorrhage. The aims of the

Dysregulation of endothelium-dependent coronary vascular tone

157

present study were: (a) to determine the role of endothelium-derived relaxing factors in the coronary vascular tone disturbances after acute hemorrhage and (b) to determine possible changes in the endothelium-dependent regulation of coronary vascular tone caused by shifts in the myocardial redox-state.

2. METHODS The experiments were performed on 143 hearts isolated from female Wistar rats. The animals were divided into the following groups: “control,” “hemorrhage,” “immobilization stress,” “stress plus hemorrhage.” Rats from the “hemorrhage” group were anesthetized with pentobarbital sodium (60 mg/kg, intraperitoneally). The right jugular vein and left common carotid artery were cannulated. Hemorrhage was induced by withdrawing arterial blood from the carotid artery to sustain mean arterial pressure at the level of 40–50 mmHg during 2 h. The animals were not resuscitated after that. Stress was caused by the strong immobilization of all extremities of the non-anesthetized rats during 6 h (always from 8 a.m. to 2 p.m.). In the “stress plus hemorrhage” group, rats under stress were anesthetized and blood loss was induced after a 6-h immobilization. A thiol-containing antioxidant N-acetylcysteine (“Sigma,” USA, 40 mg/kg) was injected intraperitoneally (Table 20.1). After the experimental protocol was carried out, cardiectomy was performed under anesthesia. Isolated hearts were perfused by Langendorff method with Krebs–Henseleit solution (NaCl 118.0 mM, KCl 4.8 mM, NaHCO3 25.0 mM, CaCl2 2.5 mM, KH2PO4 1.2 mM, MgSO4 1.18 mM, glucose 5.5 mM), aerated with a mixture of 95% O2 and 5% CO2, to achieve pH 7.3–7.4. In the first series of experiments isometrically working isolated hearts were perfused with constant coronary perfusion pressure, which was elevated gradually from 40–120 mm Hg (each time by 20 mm Hg increments). The volume velocity of coronary flow and developed left ventricular pressure (LVP) were measured. Autoregulation index (AI), which indicates the ability of the coronary vessels, namely arteries, to constrict in response to the perfusion pressure elevation (Novikova, 1972), was calculated: AI ⫽ (⌬Q1 – ⌬Q2)/⌬Q1, where ⌬Q1 is the coronary flow increase in response to perfusion pressure elevation, ⌬Q2 is the difference between initial coronary flow and coronary flow fixed at the new step under perfusion pressure rise. The AI higher than 0.5 characterized an adequate coronary autoregulation. Table 20.1 Experimental groups and protocol Experimental groups

Without N-acetylcysteine

With N-acetylcysteine supplementation. N-acetylcysteine was given:

Control Hemorrhage

n ⫽ 22 n ⫽ 21

Immobilization stress

n ⫽ 16

Stress plus hemorrhage

n ⫽ 14

1 h before cardiectomy, n ⫽ 17 1 h following posthemorrhagic hypotension period, n ⫽ 22 1 h before the immobilization, n ⫽ 15 1 h before the immobilization, n ⫽ 16

158

L.Eu. Belyaeva et al.

The cardiac contractility was characterized by the “intensity of hearts’ structures functioning” (ISF, Meerson and Alekhina, 1968): ISF ⫽ (HR ⫻ LVP)/m, where HR is the rate of the beating isolated heart per minute, LVP is the developed left ventricular pressure, m is the dry weight of the left ventricular myocardium. NO-synthase was inhibited with L-NAME (“Sigma,” USA, 60 ␮M/L) added to the perfusion solution. In the second series of experiments, isotonically contracting hearts were perfused at constant coronary flow. The initial coronary flow volume produced by peristaltic pump (PP-2B15, “Zalimp,” Poland) was 4 ml/min. When the coronary perfusion pressure had stabilized, coronary flow was elevated rapidly to 20 ml/min. Thus, increase of the coronary flow resulted in the rapid rise of the coronary perfusion pressure. After 45–90 s, the coronary perfusion pressure decreased. The degree of the shear–stress induced endothelium-dependent coronary vasodilatation (coronary vasodilatation, CV) was calculated from the formula: CV ⫽ ⌬CPP2 ⫻ 100%/⌬CPP12, where ⌬CPP1 – is the difference between the initial coronary perfusion pressure (CPP) and maximal coronary perfusion pressure; ⌬CPP2 is the difference between the maximal CPP and minimal CPP after 45–90 s. ⫺ ⫺ The concentration of the stable products of nitric oxide degradation – (NO3 /NO2 ) was determined spectrophotometrically with Greiss reagent (Green et al., 1982). Cyclic guanosine monophosphate (cGMP) concentration in the blood serum was detected with radioimmune assay. The degree of oxidative stress in the myocardium was assessed by measuring the concentration of the malondialdehyde (Placer et al., 1966). Reduced glutathione content was determined with Ellmann reagent (Ellmann, 1959). Student’s t-test was used to determine statistical differences between groups. Values of P less than 0.05. were considered statistically significant. 3. RESULTS The volume velocity of coronary flow after hemorrhage, immobilization stress and combined influence of the stress and hemorrhage were significantly higher than in the control hearts (Figure 20.1). The AI in the rat hearts from the experimental groups was less compared to the control groups (Table 20.2). Developed left ventricular pressure measured in the isolated hearts in all experimental models was also lower than in the control (Table 20.3). Moreover, the total volume of perfusate measured per unit of the beating hearts from experimental groups was increased by two to three times compared with the control value. The inhibitor of NO-synthase added to the perfusion solution significantly decreased the volume velocity of coronary flow in the isolated hearts of the control and experimental rats. However, the degree of the coronary flow reduction differed among the groups: L-NAME was capable of completely preventing the rise in coronary flow in the hearts from “stress” and “stress plus hemorrhage” groups. The L-NAME only partially diminished coronary flow in the rat hearts after a 2 h posthemorrhagic arterial hypotension (Figure 20.2). In the second series of experiments where isolated hearts were perfused at constant coronary flow, the average of initial CPP was similar in the control and experimental hearts. Sudden

Coronary flow (ml/m)

Dysregulation of endothelium-dependent coronary vascular tone 300

Control

Hemorrhage

250

Stress

Stress + hemorrhage

159

200 150 100 50 0 40

60 80 100 Perfusion pressure (mmHg)

120

Figure 20.1 Influence of stress, hemorrhage and their combination on volume velocity of coronary flow in isolated rat hearts. * Significance difference (P ⬍ 0.05) between control and experimental groups.

coronary flow elevation from 4–20 ml/min resulted in an abrupt rise in coronary perfusion pressure. Nevertheless, maximal CPP in the “hemorrhage” group was lower by 33.1% than in the controls (85.7 ⫾ 5.2 mmHg against 128.0 ⫾ 19.0 mmHg in the controls, P ⬍ 0.05). The previous 6-h immobilization prevented the reduction in maximal coronary perfusion pressure after hemorrhage, which was at the level 112.7 ⫾ 14.8 mmHg. After increase of CPP for 45–90 s, it began to decrease. The coefficient of the coronary vasodilatation after stress, hemorrhage and their combination was higher than in the controls by 2.7, 3.7 and 3.1 times, respectively. The inhibitor of NO-synthase abolished the decrease in coronary perfusion pressure in response to shear stress. ⫺ ⫺ The concentration of the stable nitric oxide degradation products (NO3 /NO2 ) in the serum after hemorrhage, immobilization stress and their combination was higher than in the control by 97.8%, 101.5% and 47.4%, respectively. However, the amount in cGMP increased only after immobilization stress (by 19.2%) and in the group “stress plus hemorrhage” (by 21.4% compared with control). After 6 h of immobilization, 2 h of posthemorrhagic arterial hypotension and their combination the malondialdehyde concentration in the myocardium was elevated by 2.4, 2.6 and 3.9 times, respectively, compared to control value. The reduced glutathione concentration in the myocardium decreased by 40.6% after immobilization stress compared with control values. Previous treatment with N-acetylcysteine in part prevented the increase in volume velocity of coronary flow after stress, hemorrhage and their combination (Table 20.2). In addition, supplementation with N-acetylcysteine decreased the AI of coronary flow: in the control group – by 35.3–30.7% in perfusion pressure range from 80 to 120 mm Hg, compared with mean values in the same group without NAC injection; in the group “hemorrhage plus N-acetylcysteine” – by 30.6–50.7% in the range of perfusion pressure 80–120 mm Hg than in the controls, after combined action of stress and hemorrhage by 46.2% when perfusion pressure was 80 and 120 mmHg (Table 20.2). N-acetylcysteine also did not restore the AI, which was significantly diminished after 6-h immobilization. The inability of N-acetylcysteine to restore the AI combined with increase in heart rate in the isolated beating rat hearts in the controls and after hemorrhage (compared with heart rate in the same group without N-acetylcysteine, Table 20.3). After stress N-acetylcysteine did not cause any significant changes in the heart rate as in the group “stress plus hemorrhage.” Despite the fact that

Control (n ⫽ 10) Hemorrhage (n ⫽ 10) Hemorrhage with NAC (n ⫽ 11) Immobilization stress (n ⫽ 8) Immobilization stress with NAC (n ⫽ 8) Stress plus hemorrhage (n ⫽ 6) Stress plus hemorrhage with NAC (n ⫽ 8) Control (n ⫽ 10) Hemorrhage (n ⫽ 10) Hemorrhage with NAC (n ⫽ 11) Immobilization stress (n ⫽ 8) Immobilization stress with NAC (n ⫽ 8) Stress plus hemorrhage (n ⫽ 6) Stress plus hemorrhage with NAC (n ⫽ 8)

Experimental groups

— — —

59.4 ⫾ 7.0 80.3 ⫾ 8.6* 45.4 ⫾ 3.9 75.5 ⫾ 6.8* 46.4 ⫾ 5.9 81.6 ⫾ 6.9* 40.8 ⫾ 3.5 — — —

40 97.5 ⫾ 11.7 140.0 ⫾ 13.2* 94.3 ⫾ 9.6 134.0 ⫾ 12.4* 103.6 ⫾ 10.3 138.7 ⫾ 18.9* 91.4 ⫾ 11.8 0.36 ⫾ 0.06 0.24 ⫾ 0.03* 0.25 ⫾ 0.05 0.26 ⫾ 0.06 0.23 ⫾ 0.04* 0.29 ⫾ 0.10 0.24 ⫾ 0.04

60

Perfusion pressure (mmHg)

Notes All values are presented as means ⫾ SE. NAC – N-acetylcysteine. * Significant difference between experimental groups and control at P ⬍ 0.05.

Autoregulation index

Volume velocity of coronary flow (ml ⫻ g/min)

Variable

129.0 ⫾ 14.2 165.5 ⫾ 13.0* 144.3 ⫾ 12.1 173.7 ⫾ 15.3* 141.2 ⫾ 12.5 160.7 ⫾ 20.6* 139.1 ⫾ 10.4 0.51 ⫾ 0.06 0.50 ⫾ 0.06 0.26 ⫾ 0.03* 0.29 ⫾ 0.03* 0.35 ⫾ 0.04* 0.55 ⫾ 0.05 0.27 ⫾ 0.02*

80

159.8 ⫾ 14.1 199.5 ⫾ 15.1* 193.1 ⫾ 13.8* 216.1 ⫾ 16.6* 175.4 ⫾ 15.8 197.5 ⫾ 21.2* 186.0 ⫾ 14.8* 0.53 ⫾ 0.06 0.42 ⫾ 0.07 0.28 ⫾ 0.03* 0.35 ⫾ 0.04* 0.39 ⫾ 0.06 0.42 ⫾ 0.04 0.43 ⫾ 0.07

100

178.7 ⫾ 13.5 213.4 ⫾ 15.2* 227.6 ⫾ 15.0* 242.5 ⫾ 21.2* 204.3 ⫾ 15.6 224.5 ⫾ 21.6* 218.7 ⫾ 9.3* 0.75 ⫾ 0.08 0.56 ⫾ 0.06* 0.37 ⫾ 0.04* 0.48 ⫾ 0.06* 0.49 ⫾ 0.05* 0.52 ⫾ 0.07* 0.41 ⫾ 0.09*

120

Table 20.2 Coronary flow and coronary autoregulation after immobilization stress, hemorrhage and their combination with N-acetylcysteine treatment

152 ⫾ 25 151 ⫾ 33 183 ⫾ 15 139 ⫾ 36 178 ⫾ 10 99.0 ⫾ 7.8 77.6 ⫾ 19.5 65.0 ⫾ 12.0* 68.8 ⫾ 18.1* 67.6 ⫾ 6.7*

Control (n ⫽ 10) Hemorrhage (n ⫽ 10) Hemorrhage with NAC (n ⫽ 11) Stress plus hemorrhage (n ⫽ 6) Stress plus hemorrhage with NAC (n ⫽ 8) Control (n ⫽ 10) Hemorrhage (n ⫽ 10) Hemorrhage with NAC (n ⫽ 11) Stress plus hemorrhage (n ⫽ 6) Stress plus hemorrhage with NAC (n ⫽ 8)

Heart rate (min⫺1)

225 ⫾ 20 197 ⫾ 25 247 ⫾ 8 176 ⫾ 42 215 ⫾ 14 191.2 ⫾ 17.2 127.2 ⫾ 21.5* 142.6 ⫾ 15.8* 103.5 ⫾ 27.2* 152.6 ⫾ 15.0

73.2 ⫾ 5.5 56.0 ⫾ 6.0* 53.7 ⫾ 5.0* 47.3 ⫾ 4.8* 64.7 ⫾ 5.3⫹

60

All values are means ⫾ SE. NAC – N-acetylcysteine. * Significant difference between experimental groups and control at P⬍0.05. ⫹ Significant difference between experimental groups without and with N-acetylcysteine at P (⬍ 0.05).

Intensity of hearts’ structure functioning (mmHg ⫻ g/min)

61.3 ⫾ 6.6 45.5 ⫾ 6.0* 34.0 ⫾ 4.9* 41.8 ⫾ 6.3* 35.0 ⫾ 3.0*

Control (n ⫽ 10) Hemorrhage (n ⫽ 10) Hemorrhage with NAC (n ⫽ 11) Stress plus hemorrhage (n ⫽ 6) Stress plus hemorrhage with NAC (n ⫽ 8)

Developed left ventricular pressure (mm Hg)

40

Perfusion pressure (mmHg)

Experimental groups

Variable

229 ⫾ 26 199 ⫾ 22 258 ⫾ 10⫹ 169 ⫾ 34 222 ⫾ 12 273.4 ⫾ 36.9 150.5 ⫾ 20.6* 186.9 ⫾ 20.4* 109.0 ⫾ 18.6* 187.1 ⫾ 20.7*,⫹

103.0 ⫾ 6.5 65.6 ⫾ 5.4* 66.8 ⫾ 5.5* 54.3 ⫾ 4.3* 76.1 ⫾ 5.3*,⫹

80

105.8 ⫾ 6,0 76.3 ⫾ 5,6* 78.0 ⫾ 7,3* 66.7 ⫾ 5,4* 79.2 ⫾ 7,0* 208 ⫾ 23 196 ⫾ 21 266 ⫾ 9⫹ 180 ⫾ 32 240 ⫾ 12 264.9 ⫾ 36.0 182.2 ⫾ 23.3* 223.7 ⫾ 24.1 151.3 ⫾ 33.9* 209.6 ⫾ 26.6

100

Table 20.3 Cardiac contractility after hemorrhage and in the “stress plus hemorrhage” group under N-acetylcysteine treatment

104.9 ⫾ 6.1 76.8 ⫾ 5.8* 81.6 ⫾ 7.0* 83.3 ⫾ 5.3* 86.5 ⫾ 7.6* 225 ⫾ 30 212 ⫾ 23 270 ⫾ 7⫹ 287 ⫾ 35 253 ⫾ 11 284.2 ⫾ 44.6 196.9 ⫾ 25.3* 235.7 ⫾ 22.9 190.1 ⫾ 34.9 241.7 ⫾ 30.0

120

162

L.Eu. Belyaeva et al.

Coronary flow (ml/m)

250

Control

Control + L-NAME

Hemorrhage

Hemorrhage + L-NAME

200 150 100 50 0 40

60

80

100

120

Perfusion pressure (mmHg)

Figure 20.2 Volume velocity of coronary flow in the isolated rat hearts after NO-synthase inhibition. * Significance difference (P ⬍ 0.05) between groups “hemorrhage” and “control,” ⫹ significance difference (P ⬍ 0.05) between groups “hemorrhage plus L-NAME” and “control plus L-NAME.” *

*

* 0.8 Degree of 0.6 coronary 0.4 vasodilatation 0.2 0 1

2 With NAC

3

4

Without NAC

Figure 20.3 Influence of N-acetylcysteine on the degree of shear-stress induced coronary vasodilatation following 2 h posthemorrhagic arterial hypotension, 6 h immobilization stress and combined action of the stress and hemorrhage. 1: “control”, 2: “stress,” 3: “hemorrhage,” 4: “stress plus hemorrhage.” * Significance difference (P ⬍ 0.05) between experimental groups and “control.”

N-acetylcysteine did not blunt decline in the developed LVP in all experimental groups, it caused significant attenuation in “intensity of the hearts’ structures functioning” after hemorrhage and in the “stress plus hemorrhage” group. When hearts were perfused at constant coronary flow N-acetylcysteine did not influence the initial CPP means both in the control and experimental groups. Maximal CPP in the group “hemorrhage plus N-acetylcysteine” did not differ from the mean value in the controls. N-acetylcysteine supplementation also led to a decrease in the degree of the coronary shear–stress induced vasodilatation in the “hemorrhage,” “stress” and “stress plus hemorrhage” groups (Figure 20.3). ⫺ N-acetylcysteine supplementation also prevented the elevation of NO⫺ 3 /NO2 concentration in the serum after stress, hemorrhage and their combination. In addition, N-acetylcysteine increased the concentration of cGMP in the control serum by 1.7 times. The concentration of cGMP in the serum after 6-h immobilization returned to control value after N-acetylcysteine supplementation. However after hemorrhage and combined action of stress plus hemorrhage, N-acetylcysteine did not cause significant changes in cGMP concentration.

Dysregulation of endothelium-dependent coronary vascular tone

163

N-acetylcysteine significantly increased survival rate of animals belonging to the “hemorrhage” group – from 77.6% to 92.2%; in the “stress plus hemorrhage” group – from 60.1% to 86.0%. 4. DISCUSSION The coronary tone was reduced in the rat hearts isolated after 6-h immobilization, 2 h of posthemorrhagic arterial hypotension or their combination. These data, together with the deterioration in the ability to constrict in response to an increase in perfusion pressure after stresses demonstrate a decrease in the basal coronary tone under such conditions. The simultaneous lack of increase in developed LVP in response to a gradual rise in perfusion pressure in all experimental groups illustrates ineffective hyperperfusion phenomenon in these isolated hearts, and thus a mismatch between the regulations of coronary flow and cardiac contractility. The inhibitor of NO-synthase L-NAME decreased coronary flow in the hearts isolated both from control and experimental rats. In the hearts from animals belonging to the “stress” and “stress plus hemorrhage” groups coronary flow was not different compared to those in control hearts with NO-synthase inhibition. By contrast, the coronary flow after L-NAME supplementation was higher in hearts from the “hemorrhage” group than in the “control plus L-NAME” group. Thus, nitric oxide hyperproduction in the coronary vessels endothelial cells was not the only reason for the decrease in basal coronary vascular tone after posthemorrhagic arterial hypotension. Endothelium-derived hyperpolarizing factor (EDHF) may also play a significant role in the relaxation of the coronary vasculature following severe acute hemorrhage, as may hyperproduction of prostacyclin or abnormal behavior of coronary smooth muscle cells. The late possibility is supported by the findings indicating that the maximal CPP in response to elevating coronary flow after hemorrhage was less than in the controls. Not only basal but stimulated shear–stress induced production of nitric oxide was increased in coronary vessels after immobilization, hemorrhage and in the “stress plus hemorrhage” group, as indicated by the larger shear–stress induced coronary vasodilatation in all experimental groups. Shear–stress results in an increase of EDHF formation and liberation from coronary arteries (Popp et al., 1998; Hammarstrom et al., 1999). Whether EDHF plays any role in the shear–stress induced coronary vasodilatation in the present experiments is unknown. The observed elevation of the stable products of nitric oxide degradation in the serum after hemorrhage without an increase in cGMP suggests either accelerated inactivation of nitric oxide or actions of nitric oxide unrelated to cGMP. The previous 6-h immobilization altered the characteristics of the regulation of coronary vascular tone after acute hemorrhage. First, the changes of the AI in the “stress plus hemorrhage” group were less noticeable than after hemorrhage alone. Second, previous immobilization resulted in the increase in the pressure response to L-NAME in the coronary vasculature after acute hemorrhage. Third, immobilization prevented the decrease in the maximal CPP in the “hemorrhage” group. Fourth, the previous 6-h immobilization attenuated the rise in the volume velocity of coronary flow after 2-h posthemorrhagic arterial hypotension in the isotonically contracted isolated hearts (Solodkov et al., 2001). These data suggest that immobilization modulates either production of the endothelium-derived relaxing factor or the contractility of coronary vascular smooth muscle cells during posthemorrhagic arterial hypotension. Change of the redox-state of the myocardium and coronary vessels cells could play a significant role. Indeed, cellular redox-state is one of the main

164

L.Eu. Belyaeva et al.

factors, which determine cellular response to different stimuli acting on cells. The ratio between reduced and oxidized glutathione is thought to be a “switcher,” determining redox-state (Kamata and Hirata, 1999; Schafer and Buettner, 2001). Based on the data of the strong oxidative stress in the myocardium and reduced glutathione concentration decrease following immobilization stress, it seems logical to speculate that changes in redox-state in the myocardium and coronary vascular cells contribute to the dysregulation of coronary vascular tone. N-acetylcysteine, a precursor of reduced glutathione, has different chemical properties, such as nucleophilicity, antioxidant action and ability to undergo transhydrogenation or thiol-disulfide exchange reactions with other thiol redox couples (Cotgreave, 1997). N-acetylcysteine affected coronary flow after stress, hemorrhage and their combination. Indeed, the volume velocity of coronary flow was lower in all experimental groups with N-acetylcysteine supplementation than in those without N-acetylcysteine treatment. This means that N-acetylcysteine prevented excessive coronary vasodilatation after stress, hemorrhage and their combination. Paradoxically, a previous 6-h immobilization and N-acetylcysteine had similar effects in coronary vascular tone, although more animals died in the “stress plus hemorrhage” group while N-acetylcysteine had a protective action. N-acetylcysteine also diminished the shear–stress induced coronary vasodilatation in all experimental groups. There may be different possible explanations for the effects of N-acetylcysteine. On the one hand, N-acetylcysteine may change the balance between nitric oxide and EDHF release and their action. N-acetylcysteine may influence NO-synthase activity in the coronary vessels, because this agent almost completely prevented NO3⫺/NO2⫺ elevation in the blood serum from the experimental groups. N-acetylcysteine can influence both the constitutive and inducible NO-synthase isoforms (Bergamini et al., 2001; Kampf and Roomans, 2001). N-acetylcysteine may change bioavailability of nitric oxide and causes formation of S-nitrosothiols (Cotgreave, 1997). On the other hand, N-acetylcysteine may affect the redox-state of myocardium and coronary vessels. Besides these mechanisms, N-acetylcysteine may directly influence the contractility of coronary smooth muscle cells following posthemorrhagic arterial hypotension. This speculation is based on the observation that the decline in maximal coronary perfusion pressure is prevented by N-acetylcysteine after hemorrhage in response to sudden coronary flow elevation from 4 to 20 ml/min. Hence, N-acetylcysteine may hold advantages in the treatment of acute hemorrhage because it permits a more adequate adjusting coronary flow to cardiac contractility and because it improves survival during posthemorrhagic arterial hypotension.

21 Red wine polyphenolic compounds induce EDHF-mediated relaxation and hyperpolarization in the porcine coronary artery: involvement of redox-sensitive mechanisms T. Chataigneau, M. Ndiaye, J.C. Stoclet and V.B. Schini-Kerth The cardiovascular protective effects of red wine polyphenolic compounds have been attributed, at least in part, to an enhanced endothelial formation of nitric oxide (NO), a factor with potent anti-atherosclerotic properties. This study examines whether red wine polyphenolic compounds could also induce endothelium-derived hyperpolarizing factor (EDHF)-mediated responses in arteries and, if so, to characterize the underlying mechanisms. Porcine coronary artery rings were suspended in organ chambers for measurement of changes in isometric tension and segments were used to determine membrane potential with intracellular microelectrodes. All experiments were performed in the presence of N␻-nitro-L-arginine and indomethacin to inhibit NO and prostanoid synthesis, respectively. In rings contracted with the thromboxane A2 analog U46619, red wine polyphenolic compounds caused concentration-dependent relaxations in the presence of endothelium whereas only negligible relaxations were observed in those without endothelium. In addition, red wine polyphenolic compounds induced slowly developing and transient hyperpolarizations of smooth muscle cells in intact vessels. Both endothelium-dependent relaxation and hyperpolarization to red wine polyphenolic compounds were abolished by the combination of apamin and charybdotoxin. Red wine polyphenolic compound-induced relaxations were abolished by calmidazolium, an inhibitor of calmodulin, and by the membrane-permeant superoxide dismutase, Mn [III] tetrakis [1-methyl-4-pyridyl] porphyrin (MnTMPyP) and were significantly reduced by N-acetylcysteine, an antioxidant, and by catalase whereas superoxide dismutase had no effect. In addition, MnTMPyP also abolished red wine polyphenolic compounds-induced hyperpolarization. In conclusion, red wine polyphenolic compounds cause EDHF-mediated endothelium-dependent relaxation and hyperpolarization in the porcine coronary artery. These effects are likely to involve reactive oxygen species-dependent mechanism(s).

1. INTRODUCTION Dietary intake of polyphenolic derivatives such as flavonoids, from either beverages or fruits and vegetables, is associated with a decreased risk of coronary events (Renaud and de Lorgeril, 1992; Knekt et al., 1996). The polyphenolic component of red wine has been shown to inhibit platelet aggregation both in vivo and in vitro (Wang et al., 2002) and to increase plasma antioxidant capacity in humans (Serafini et al., 1998). In addition, red wine polyphenolic compounds induce the endothelial formation of nitric oxide (NO) and NOmediated endothelium-dependent relaxation of arterial rings in vitro (Fitzpatrick et al., 1993; Andriambeloson et al., 1997).

166

T. Chataigneau et al.

During the last two decades, the crucial role of vascular endothelium in the regulation of vascular homeostasis has emerged progressively. Indeed, the endothelial production of vasoactive substances such as NO (Furchgott and Zawadzki, 1980) and prostacyclin (Moncada and Vane, 1979b) and endothelium-dependent hyperpolarization of the vascular smooth muscle cells contribute to the local regulation of vascular tone. The endotheliumdependent hyperpolarization which is resistant to inhibitors of cyclooxygenase and nitric oxide synthase, has been attributed to endothelium-derived hyperpolarizing factor (EDHF; Chen et al., 1988; Félétou and Vanhoutte, 1988; Huang et al., 1988; Taylor and Weston, 1988). However, in most of the vascular beds studied so far, the nature of EDHF remains elusive. As endothelial dysfunction is probably one of the earliest manifestations of atherosclerosis (Mano et al., 1996), improvement of the protective endothelial function might represent a potential mechanism by which ingestion of polyphenolic compounds may help reduce cardiovascular risk. Therefore, the purpose of this study was to assess whether red wine polyphenolic compounds induce endothelium-dependent responses involving EDHF in the isolated porcine coronary artery and if so, to investigate the underlying mechanism(s).

2. METHODS 2.1. Preparation of red wine polyphenolic compounds The polyphenolic extract dry powder from red French wine (Corbières A.O.C.) was provided by Dr M. Moutounet (Institut National de la Recherche Agronomique, Montpellier, France). Phenolic compounds were adsorbed on a preparative column; the alcoholic-eluent was gently evaporated; the concentrated residue was lyophilized and finely sprayed to obtain red wine polyphenolic compounds dry powder. One liter of red wine produced 2.7 g of red wine polyphenolic compounds, which contained 471 mg/g of total phenolic compounds expressed as gallic acid. Phenolic levels in red wine polyphenolic compounds were measured by HPLC. The extract contained 8.6 mg/g catechin, 8.7 mg/g epicatechin, dimers (B1: 6.9 mg/g, B2: 8.0mg/g, B3: 20.7mg/g and B4: 0.7mg/g), anthocyanins (malvidin-3-glucoside: 11.7mg/g, peonidin-3-glucoside: 0.66 mg/g, and cyanidin-3-glucoside: 0.06 mg/g) and phenolic acids (gallic acid: 5.0 mg/g, caffeic acid: 2.5 mg/g, and caftaric acid: 12.5 mg/g).

2.2. Preparation of porcine coronary artery Porcine hearts were obtained from a local slaughterhouse and placed into ice-cold Krebs bicarbonate solution of the following composition (in mM): NaCl 119, KCl 4.7, CaCl2 1.25, MgSO4 1.17, KH2PO4 1.18, NaHCO3 25 and glucose 11. Left anterior descending coronary arteries were excised and cleaned of adherent fat and connective tissues.

2.3. Microelectrode studies Segments of arteries were slit open longitudinally, pinned to the sylgard base of a heated organ chamber with the intimal side upward and superfused (5 ml/min) with heated (37 ⬚C) Krebs bicarbonate solution (buffered with 95% O2/5% CO2, pH 7.4). The membrane potential of smooth muscle cells was recorded with glass capillary microelectrodes (tip resistance of 80 to 100 M⍀) filled with KCl (3 M) and connected to the headstage of a high impedance amplifier (intra 767, WPI). The reference electrode (Dri-Ref reference electrodes, WPI), also filled with KCl but exhibiting low electrolyte leakage, was immersed into the bathing

EDHF and reactive oxygen species

167

solution and directly connected to the amplifier. The signal was continuously monitored on an oscilloscope and simultaneously recorded on paper. Impalement of smooth muscle cells was performed from the intimal side. Successful impalements were signalled by a sudden negative drop in potential from the baseline (zero potential reference) followed by a stable negative potential for at least two minutes. The preparations were superfused for at least 45 min prior to any recording. All experiments were performed in the presence of N␻-nitro-L-arginine (10⫺4 M), indomethacin (10⫺5 M) and U46619 (10⫺8 M), the stable thromboxane A2 analog. All drugs were applied by continuous superfusion.

2.4. Vascular reactivity Rings (4–5 mm in length) were suspended in 10 mL organ chambers filled with Krebs bicarbonate solution (buffered with 95% O2/5% CO2, pH 7.4, 37 ⬚C) for the measurement of changes in isometric force. They were equilibrated for 90 min under an optimal passive tension of 5g. Thereafter, the rings were repeatedly exposed to a 80mM KCl-containing Krebs bicarbonate solution until stable contractions were obtained. After washing, the rings were contracted with U46619 (10⫺8–10⫺7 M) to approximately 80% of the maximal response to KCl. At the steady state of the contraction, concentration–response curves were obtained in response to cumulative concentrations of red wine polyphenolic compounds (10⫺4–10⫺1 g/l). Indomethacin (10⫺5 M) was present throughout all experiments. In some cases, the endothelium was removed by rubbing the intima with forceps.

2.5. Drugs Levcromakalim (BRL 38227) was a generous gift from GlaxoSmithKline (USA). U46619 (9,11-dideoxy-11␣, 9␣-epoxymethano-prostaglandin F2␣) was from Cayman Chemical Company (USA). Allopurinol, apamin, calmidazolium, catalase, charybdotoxin, diphenylene iodonium, hydrogen peroxide, indomethacin, myxothiazol, N␻-nitro-L-arginine, sulfaphenazole, superoxide dismutase and tiron (4,5-dihydroxy-1,3-benzene-disulfonic acid) were obtained from Sigma (La Verpillère, France). MnTMPyP (Mn [III] tetrakis [1-methyl-4-pyridyl] porphyrin) and BH4 ([6R]-5,6,7,8-tetrahydrobiopterin) were purchased from Alexis Biochemicals (Lausen, Switzerland). N-acetylcysteine (Fluimucil®) is a product of Zambon (Issy-les-Moulineaux, France). Red wine polyphenolic compounds solutions were prepared extemporaneously in a stock solution of ethanol and deionized water (50% v/v) at a concentration of 100 g/l.

2.6. Statistical analysis All values are reported as means ⫾ SEM; n indicates the number of vascular segments (animals) examined. Statistical analysis was performed using Student’s t-test for paired or unpaired observations. A probability value of P less than 0.05 was considered significant.

3. RESULTS 3.1. Characterization of red wine polyphenolic compound-induced relaxation and hyperpolarization Red wine polyphenolic compounds (10⫺4 to 10⫺1 g/l) induced concentration-dependent relaxations of isolated rings of porcine coronary arteries with endothelium contracted with U46619 (Figure 21.1). The maximal effect was achieved at a concentration of 10⫺1 g/l of red

168

T. Chataigneau et al.

Figure 21.1 Original recordings illustrating the relaxant effects of red wine polyphenolic compounds (RWPC, 10⫺4 to 10⫺1 g/l) in isolated rings of porcine coronary artery with (A) or without endothelium (B) contracted with the thromboxane A2 analog U46619 (6 ⫻ 10⫺8 M), in the presence of indomethacin (10⫺5 M).

Figure 21.2 Original recordings showing the effects of a combination of charybdotoxin (10⫺7 M) and apamin (10⫺7 M) on red wine polyphenolic compound (RWPC)-induced endothelium-dependent relaxation and hyperpolarization in isolated porcine coronary artery in the presence of N␻-nitro-L-arginine (10⫺4 M), indomethacin (10⫺5 M) and U46619. The effect of levcromakalim (LK, 10⫺5 M), a selective opener of ATPsensitive potassium channels, on membrane potential is also shown.

wine polyphenolic compounds and amounted to 98.1 ⫾ 0.9% at 10⫺1 g/l (n ⫽ 5). In rings without endothelium, only modest relaxations were observed in response to high concentrations of red wine polyphenolic compounds (31.4 ⫾ 4.5% at 10⫺1 g/l red wine polyphenolic compounds, n ⫽ 5; Figure 21.1). The endothelium-dependent relaxations to red wine polyphenolic compounds were partially inhibited by either N␻-nitro-L-arginine (10⫺4 M; Figure 21.1) or the combination of charybdotoxin (10⫺7 M) and apamin (10⫺7 M; data not shown) alone whereas they were abolished by the combination N␻-nitro-L-arginine and charybdotoxin plus apamin (Figure 21.2). The red wine polyphenolic compounds (10⫺1 g/l)induced, EDHF-mediated relaxation was markedly reduced by calmidazolium, a calmodulin inhibitor (10⫺5 M; from 77.7 ⫾ 3.1% to 10.4 ⫾ 5.1%, n ⫽ 5).

EDHF and reactive oxygen species

169

In isolated segments exposed to U46619 (10⫺8 M) and in the presence of N␻-nitro-L-arginine, red wine polyphenolic compounds (10⫺1 g/l) evoked slowly developing and transient endothelium-dependent hyperpolarizations of smooth muscle cells (⫺10.4 ⫾ 1.5 mV from a resting membrane potential of ⫺47.6 ⫾ 4.4 mV, n ⫽ 5; Figure 21.2, right panel). These responses were abolished by the combination of the two toxins, charybdotoxin and apamin (Figure 21.2, right panel). Under these conditions, levcromakalim, a selective opener of ATP-sensitive potassium channels, could still induce a hyperpolarization of smooth muscle cells (Figure 21.2, right panel).

3.2. Involvement of reactive oxygen species Red wine polyphenolic compound (10⫺1 g/l)-induced EDHF-mediated relaxations were significantly inhibited by tiron and abolished by N-acetylcysteine, two antioxidants (Figure 21.3). They were also markedly reduced by the membrane-permeant superoxide dismutase, MnTMPyP, and slightly but significantly reduced by catalase whereas superoxide dismutase had only minor effects (Figure 21.4). In the presence of MnTMPyP, the resting membrane potential of smooth muscle cells was not affected significantly (⫺45.0 ⫾ 3.5 mV in control vs ⫺46.8 ⫾ 3.6 mV in the presence of MnTMPyP, n ⫽ 6). Under these conditions, the endothelium-dependent EDHF-mediated hyperpolarization induced by red wine polyphenolic compounds (10⫺1 g/l) was abolished (Figure 21.4). In the presence of N␻-nitro-L-arginine, hydrogen peroxide relaxed both endotheliumintact and endothelium-denuded rings of porcine coronary artery only at the highest concentrations. Relaxations to H2O2 (10⫺3 M) amounted to 89.9 ⫾ 2.3% and 73.6 ⫾ 5.9%, in the presence and absence of endothelium, respectively (n ⫽ 4–6). The effects of diphenylene iodonium, an inhibitor of flavin-dependent enzymes such as the NAD(P)H oxidase, allopurinol, a xanthine oxidase inhibitor, sulfaphenazole, a selective cytochrome P450 2C9 inhibitor and myxothiazol, an inhibitor of the mitochondrial site III electron transport, were investigated on red wine polyphenolic compound-induced

Figure 21.3 Effects of N-acetylcysteine (10⫺2 M) and tiron (10⫺2 M), two antioxidants, on the endothelium-dependent relaxation evoked by red wine polyphenolic compounds (RWPC, 10⫺1 g/l) in isolated porcine coronary artery in the presence of N␻-nitro-Larginine (10⫺4 M) and indomethacin (10⫺5 M). Data are shown as means ⫾ SEM. The asterisk indicates a statistically significant difference between a treatment and its respective control. The numbers into parenthesis refer to the number of vascular segments (animals) examined.

170

T. Chataigneau et al.

Figure 21.4 Effects of catalase (1250 U/ml), superoxide dismutase (SOD, 100 U/ml) and a membrane-permeant superoxide dismutase (MnTMPyP, 10⫺4 M) on the endotheliumdependent responses induced by red wine polyphenolic compounds (RWPC, 10⫺1 g/l) in isolated porcine coronary artery in the presence of N␻-nitro-L-arginine (10⫺4 M), indomethacin (10⫺5 M) and U46619. Data are shown as means ⫾ SEM. The asterisk indicates a statistically significant difference between a treatment and its respective control. The numbers into parenthesis refer to the number of vascular segments (animals) examined.

Figure 21.5 Effects of diphenylene iodonium (DPI, 10⫺5 M), allopurinol (10⫺5 M), sulfaphenazole (10⫺5 M) and myxothiazol (5⫻10⫺7 M) on the endothelium-dependent relaxation evoked by red wine polyphenolic compounds (RWPC, 10⫺1 g/l) in isolated porcine coronary artery in the presence of N␻-nitro-L-arginine (10⫺4 M) and indomethacin (10⫺5 M). Data are shown as means ⫾ SEM. The numbers into parenthesis refer to the number of vascular segments (animals) examined.

EDHF-mediated relaxations in order to determine the potential enzymatic source of reactive oxygen species. None of these inhibitors significantly modified the red wine polyphenolic compound-induced EDHF-mediated relaxation (Figure 21.5). In contrast, BH4, an important cofactor of NO synthase, induced a time-dependent reduction of the red wine polyphenolic compound-induced EDHF-mediated relaxation (Figure 21.6).

EDHF and reactive oxygen species

171

Figure 21.6 Time-dependent effects of tetrahydrobiopterin (BH4, 3 ⫻ 10⫺4 M), an essential cofactor of NO synthase, on the endothelium-dependent relaxation evoked by red wine polyphenolic compounds (RWPC, 10⫺1 g/l) in isolated porcine coronary artery in the presence of N␻-nitro-L-arginine (10⫺4 M) and indomethacin (10⫺5 M). Data are shown as means ⫾ SEM. The asterisk indicates a statistically significant difference between a treatment and its respective control. The numbers into parenthesis refer to the number of vascular segments (animals) examined.

4. DISCUSSION Previous studies have shown that red wine polyphenolic compounds can induce the endothelial formation of NO and NO-mediated endothelium-dependent relaxations of rat aortic rings in vitro (Fitzpatrick et al., 1993; Andriambeloson et al., 1997). The present findings demonstrate that red wine polyphenolic compounds can, beside NO, also induce endotheliumdependent EDHF-mediated relaxations and hyperpolarizations in the porcine coronary artery. Moreover they suggest that a mechanism(s) that is(are) critically dependent on the generation of reactive oxygen species appears to be important for the generation of EDHFmediated responses to red wine polyphenolic compounds. Red wine polyphenolic compounds caused concentration-dependent relaxations of isolated porcine coronary arteries, which were more pronounced in intact than in denuded preparations. Since these endothelium-dependent relaxations were only partially inhibited by N␻-nitro-L-arginine, NO is unlikely to be the sole endothelium-derived relaxing factor involved. The N␻-nitro-L-arginine-resistant endothelium-dependent relaxations to red wine polyphenolic compounds were abolished by the combination of charybdotoxin and apamin, two well-known inhibitors of EDHF-mediated responses probably acting on intermediate (IKCa)- and small (SKCa)-conductance calcium-activated potassium channels (Waldron and Garland, 1994; Corriu et al., 1996a; Zygmunt and Hogestatt, 1996; Edwards et al., 1998, 1999b; Andersson et al., 2000). Moreover, red wine polyphenolic compounds induced endotheliumdependent hyperpolarizations of smooth muscle cells in the porcine coronary artery, which were abolished by the combination of charybdotoxin and apamin. Altogether these findings indicate that endothelium-dependent relaxations to red wine polyphenolic compounds involve two components, one due to the release of NO and one mediated by EDHF. The sensitivity of the EDHF-mediated relaxation induced by red wine polyphenolic compounds to calmidazolium suggests that calmodulin plays a major role in the generation

172 T. Chataigneau et al. of this response in the porcine coronary artery. This is in agreement with previous studies demonstrating that the formation of a Ca2⫹-calmodulin complex following an increase in [Ca2⫹]i in endothelial cells is a key step in the generation of EDHF-mediated responses in canine coronary arteries (Illiano et al.; 1992; Nagao et al., 1992). Indeed, it seems most likely that an increase in endothelial [Ca2⫹]i is responsible for the activation of intermediate (IKCa)- and small (SKCa)-conductance calcium-activated potassium channels leading to the hyperpolarization of endothelial cells which is then transmitted to smooth muscle cells (Marchenko and Sage, 1996; Edwards et al., 1998, 1999b; Ohashi et al., 1999; Quignard et al., 2000). The present findings indicate that the red wine polyphenolic compound-induced EDHFmediated relaxation is inhibited by the antioxidants N-acetylcysteine and tiron. In addition, MnTMPyP, a membrane-permeant superoxide dismutase mimetic (Gardner et al., 1996), prevented both EDHF-mediated relaxation and hyperpolarization whereas catalase had only a minor effect. These results indicate that reactive oxygen species and more particularly superoxide anions are involved in a crucial step of red wine polyphenolic compoundinduced EDHF-mediated responses in the porcine coronary artery and that hydrogen peroxide could play a role but to a lesser extent. The fact that superoxide dismutase itself did not affect red wine polyphenolic compound-induced EDHF-mediated relaxation suggests that the participation of reactive oxygen species concerns most probably the endothelial signaling pathway. Indeed, superoxide anions affect endothelial calcium signaling by stimulating an increase in [Ca2⫹]i and enhancing agonist-stimulated calcium responses (Graier et al., 1998; Lounsbury et al., 2000). Red wine polyphenolic compounds induce an increase in [Ca2⫹]i leading to the synthesis of NO in bovine aortic endothelial cells (Stoclet et al., 1999; Martin et al., 2002). Furthermore, these effects do not seem to be related to the antioxidant and free radical scavenging properties of polyphenolic compounds (Andriambeloson et al., 1997). Therefore, the production of reactive oxygen species in endothelial cells is most likely a key step of red wine polyphenolic compound-induced EDHF-mediated responses in the porcine coronary artery. Hydrogen peroxide may represent an EDHF in mouse and human mesenteric arteries (Matoba et al., 2000, 2002). In the present study, hydrogen peroxide relaxed the smooth muscle of porcine coronary artery independently of the endothelium and only at high concentrations, confirming previous studies (Beny and von der Weid, 1991; Barlow and White, 1998; Hayabuchi et al., 1998a). Moreover since catalase had only a weak inhibitory effect, hydrogen peroxide appears to play only a minor role in the red wine polyphenolic compound-induced EDHF-mediated relaxation. In an attempt to characterize the enzymatic source responsible for the production of superoxide anions in response to red wine polyphenolic compounds, EDHF-mediated relaxations were challenged with several inhibitors. None of the main enzymatic systems generating reactive oxygen species in endothelial cells, such as NAD(P)H oxidase, xanthine oxidase, cytochrome P450 2C or even the mitochondrial respiratory chain (Wolin, 2000; Fleming et al., 2001b), appeared to be involved in this process. Furthermore, a role for cyclooxygenase in the production of superoxide anions can be excluded as all experiments were performed in the presence of indomethacin. Studies on both neuronal and endothelial NO synthases have shown that in the absence of the substrate L-arginine or the cofactor BH4, heme reduction in the enzyme uncouples NO synthase resulting in the formation of superoxide anions rather than NO (Cai and Harrison, 2000). In the present study BH4 was used in order to inhibit superoxide generation from endothelial NO synthase (Xia et al., 1998; Vasquez-Vivar et al., 1999). This cofactor induced a time-dependent reduction of the red wine polyphenolic compound-induced

EDHF and reactive oxygen species

173

EDHF-mediated relaxation indicating that the endothelial NO synthase itself is most likely the main enzymatic source responsible for the production of superoxide anions in response to red wine polyphenolic compounds. An uncoupling of endothelial NO synthase leading to a production of superoxide anions is generally associated to an impairment of endothelial function as seen in vascular diseases such as atherosclerosis and diabetes. However, it seems in the light of the present results that it may also constitute a molecular mechanism leading to the production of EDHF-mediated responses by red wine polyphenolic compounds in the porcine coronary artery. In support of these findings is the demonstration that in a model of mice deficient in BH4 the endothelium-dependent relaxations are in part mediated by the generation of reactive oxygen species catalyzed by endothelial NO synthase (Cosentino et al., 2001).

5. CONCLUSION The capacity of red wine polyphenolic compounds to stimulate EDHF-mediated responses might represent a potential novel mechanism by which moderate consumption of red wine and ingestion of polyphenolic compounds may reduce the risk of coronary events.

22 Estrogen substitution restores the basal influence of nitric oxide and endothelium-derived hyperpolarizing factor on vascular tone in isolated mesenteric arteries from ovariectomized rats M. Zerr, T. Chataigneau, F. Hudlett, F. Pernot, and V.B. Schini-Kerth Estrogens have protective effects on the cardiovascular system in part by improving the endothelial function. Chronic estrogen treatment potentiates endothelium-dependent relaxation by stimulating the endothelial formation of nitric oxide (NO) and prostacyclin. The present study examines whether administration of 17␤-estradiol affects also endotheliumderived hyperpolarizing factor (EDHF)-induced suppression of vascular tone. Rings of mesenteric artery from either sham-operated rats, dosed with solvent (sesame oil), control ovariectomized rats, dosed with sesame oil, and ovariectomized rats, dosed with 17␤-estradiol, were suspended in organ chambers for the measurement of changes in isometric tension in the presence of indomethacin. In the sham-operated group (one and four weeks treatments), N␻nitro-L-arginine (an inhibitor of nitric oxide synthase) and the combination of charybdotoxin (an inhibitor of large and intermediate conductance calcium-activated potassium channels) plus apamin (an inhibitor of small conductance calcium-activated potassium channels) significantly shifted to the left the concentration–contraction curve to phenylephrine. N␻-nitro-Larginine also increased contractile responses to phenylephrine in the control group of ovariectomized rats after 1 week of treatment whereas no such effect was obtained after 4 weeks of treatment. Charybdotoxin and apamin affected contractile responses to phenylephrine neither after 1 week nor 4 weeks of treatment in the control group of ovariectomized rats. Estrogen-substitution restored the potentiation by N␻-nitro-L-arginine of contractile responses to phenylephrine and that by the combination of charybdotoxin and apamin after 4 weeks of treatment. These experiments suggest that restoration of physiological estrogen levels by estrogen-replacement therapy is able to prevent the loss of the basal influence of both NO and EDHF in the mesenteric artery of ovariectomized rats.

1. INTRODUCTION Clinical studies have indicated a lower rate of coronary artery disease in premenopausal women in comparison to age-matched men or postmenopausal women (Kalin and Zumoff, 1990; Stampfer et al., 1991). This difference has been attributed to the protective effects of estrogens in the female cardiovascular system. Estrogens alter the lipid profile by reducing plasma LDL cholesterol and increasing HDL cholesterol levels (Bush et al., 1987; Gruchow et al., 1988). However, this beneficial change of the lipid profile is unlikely to account solely for the cardioprotective effect of estrogens in women. It has also been shown that estrogens can depress vascular contractility by changing the responsiveness of the smooth muscle to

Estrogen regulation of basal influence of NO and EDHF

175

[Ca2⫹]i (Murphy and Khalil, 2000), by reducing the release of endothelium-derived contracting factors (Bilsel et al., 2000) and by potentiating that of endothelium-derived relaxing factors (Kauser and Rubanyi, 1997). Moreover, increasing evidence suggests that a sustained upregulation of endothelial NO synthesis by estrogens is one of the major mechanisms involved in vascular protection (Kauser and Rubanyi, 1997). Indeed, endothelial nitric oxide (NO) production is stimulated by hormonal changes during the menstrual cycle particularly when estradiol levels are the highest (Kharitonov et al., 1994). Furthermore, after menopause, the risk of coronary artery disease rises to levels close to those of men (Stampfer et al., 1991), and estrogen replacement therapy reduces the risk of cardiovascular disease in postmenopausal women (Stampfer et al., 1991) possibly by potentiating endothelium-dependent vasodilatation (Gilligan et al., 1994). Finally, a positive relationship between plasma levels of estrogens and basal NO has been demonstrated in women (Sudhir et al., 1996). The endothelial NO synthase appears to be an important target of estrogens for potentiating endothelial vasodilator function but other mechanisms, particularly those related to another major endothelium-dependent relaxing factor, endothelium-derived hyperpolarizing factor (EDHF), might also contribute. Such a possibility is particularly attractive in the coronary circulation and in small arteries where EDHF is a major mediator of the endothelial control of vascular tone and blood flow. Therefore, the aim of this study is to examine the possibility that estrogens can also modulate the basal influence of EDHF on vascular tone in isolated rat mesenteric arteries.

2. METHODS 2.1. Animal preparation Sexually mature female Wistar rats (12–13 weeks of age, approximately 220 g) were used in these studies. All rats were housed individually in temperature-controlled (23 ⬚C) on a 10 h dark/14 h light-cycled with ad libitum access to standard rat chow and water. Rats were anesthetized with sodium pentobarbital (60 mg/kg, i.p.) under aseptic conditions. Sham-operated (control) group received only laparotomy. Ovariectomy was carried out by making a small incision in the lower abdomen and removing both ovaries (Nekooeian et al., 1998). After a resting period of 1 week, rats were divided into groups receiving daily a sub-cutaneous injection of drug or vehicle (sesame oil, vehicle) for either 1 or 4 weeks. The following groups were obtained: sham-operated group, dosed with vehicle; control ovariectomized group, dosed with vehicle; ovariectomized group, dosed with 17␤-estradiol (22 ␮g/kg/day).

2.2. Preparation Rats were anesthetized with sodium pentobarbital (60 mg/kg, i.p.), and blood samples (4 ml) were collected from the abdominal aorta for measurement of 17␤-estradiol. After centrifugation of the blood for 10 min at 6000 rpm, plasma was collected and frozen at ⫺27 ⬚C for later measurement of 17␤-estradiol level. The mesenteric bed was removed and placed into ice-cold Krebs bicarbonate solution of the following composition (in mM): NaCl 119, KCl 4.7, CaCl2 1.25, MgSO4 1.17, KH2PO4 1.18, NaHCO3 25 and glucose 11. The main mesenteric artery was excised and cleaned of adherent fat and connective tissues, and cut into rings of 2 mm length.

176

M. Zerr et al.

2.3. Vascular reactivity Each ring was suspended by a pair of stainless triangles in a water-jacketed bath filled with 10 ml Krebs bicarbonate solution (buffered with 95% O2 / 5% CO2, pH 7.4, 37 ⬚C) for the measurement of changes in isometric tension. The rings were progressively stretched until an optimal resting tension of 1 g was loaded and then allowed to equilibrate for at least 45 min before the start of experiments. Thereafter, rings were repeatedly challenged with KCl (80 mM) until stable contractions were obtained. After washing followed by a 30-min equilibration period, the rings were contracted again with phenylephrine (10⫺6 M). Once a stable contraction was obtained, acetylcholine (10⫺6 M) was added to test the presence of a functional endothelium. All experiments were performed in the presence of indomethacin (10⫺5 M). Rings were exposed to either solvent, N␻-nitro-L-arginine (10⫺4 M, an inhibitor of nitric oxide synthase), or the combination charybdotoxin (10⫺7 M) and apamin (10⫺7 M), inhibitors of intermediate and small conductance calcium-activated potassium channels, respectively, for at least 30 min before a concentration–contraction curve to phenylephrine was performed. In some experiments, rings were contracted sub-maximally (80%) with phenylephrine before a concentration-relaxation curve to either sodium nitroprusside or 1-ethyl-2-benzimidazolinone (1-EBIO, an activator of intermediate-conductance calciumactivated potassium channels) was performed.

2.4. Measurement of blood plasma 17␤-estradiol concentration Blood plasma concentrations of estradiol were determined using the Immunotech Estradiol assay kit.

2.5. Drugs 1-EBIO was from Tocris. All other drugs were obtained from Sigma (La Verpillère, France).

2.6. Statistical analysis All data are expressed as means ⫾ SEM; n indicates the number of different experiments. Statistical analysis was performed using Student’s t-test for paired observations. A probability value less than 0.05 was considered statistically significant.

3. RESULTS 3.1. Hormonal impregnation of rats As expected plasma levels of 17␤-estradiol were heterogeneous in the sham-operated group of rats depending on the phase of the oestrus cycle. The range of plasma levels of 17␤-estradiol was between 0 to 100 pg/ml (n ⫽ 16). Following ovariectomy, plasma levels of 17␤-estradiol were low or even below the detection level of the immunoassay (10 pg/ml). Estrogenic substitution of ovariectomized rats restored plasma 17␤-estradiol levels to approximately 80 pg/ml. In the sham-operated group of rats, uterus wet weight amounted between 100 and 400 mg reflecting the presence of rats in the different phase of the oestrus cycle. Following ovariectomy, the uterus wet weight was below 100 mg, and estrogenic substitution restored uterus wet weight to 400–450 mg.

Estrogen regulation of basal influence of NO and EDHF

177

3.2. Effect of ovariectomy on the basal influence of NO and EDHF on vascular tone In the sham-operated group of rats (1 and 4 weeks of treatment), phenylephrine elicited concentration-dependent contractile responses in mesenteric arteries, which were significantly increased in the presence of N␻-nitro-L-arginine or the combination charybdotoxin and apamin (Figures 22.1–22.4).

Figure 22.1 Effect of N␻-nitro-L-arginine on the concentration–contraction curve to phenylephrine in the isolated mesenteric artery of sham-operated rats, dosed with solvent (sesame oil) (A), ovariectomized rats, dosed with sesame oil (B) and ovariectomized rats substituted with 17␤-estradiol (22 ␮g/kg/day) (C) for 1 week. Rings of arteries were suspended in organ chamber for the measurement of changes in isometric tension. They were exposed to N␻nitro-L-arginine (10⫺4 M) 30 min before addition of increasing concentrations of phenylephrine. All experiments were performed in the presence of indomethacin (10⫺5 M). Results are shown as means ⫾ SEM of 6–16 experiments. The asterisks indicate statistically significant differences (P ⬍ 0.05).

Figure 22.2 Effect of N␻-nitro-L-arginine on the concentration–contraction curve to phenylephrine in the isolated mesenteric artery of sham-operated rats, dosed with solvent (sesame oil) (A), ovariectomized rats, dosed with sesame oil (B) and ovariectomized rats substituted with 17␤-estradiol (22 ␮g/kg/day) (C) for 4 weeks. Rings of arteries were suspended in organ chamber for the measurement of changes in isometric tension. Rings were exposed to N␻nitro-L-arginine (10⫺4 M) 30 min before addition of increasing concentrations of phenylephrine. All experiments were performed in the presence of indomethacin (10⫺5 M). Results are shown as means ⫾ SEM of 5–6 experiments. The asterisks indicate statistically significant differences (P ⬍ 0.05).

178

M. Zerr et al.

Figure 22.3 Effect of the combination charybdotoxin and apamin on the concentration–contraction curve to phenylephrine in the isolated mesenteric artery of sham-operated rats, dosed with solvent (sesame oil) (A), ovariectomized rats, dosed with sesame oil (B) and ovariectomized rats substituted with 17␤-estradiol (22 ␮g/kg/day) (C) for 1 week. Rings of arteries were suspended in organ chamber for the measurement of changes in isometric tension. They were exposed to charybdotoxin (10⫺7 M) and apamin (10⫺7 M) 30 min before addition of increasing concentrations of phenylephrine. All experiments were performed in the presence of indomethacin (10⫺5 M). Results are shown as means ⫾ SEM of 6–8 experiments. The asterisks indicate statistically significant differences (P ⬍ 0.05).

Figure 22.4 Effect of charybdotoxin and apamin on the concentration–contraction curve to phenylephrine in the isolated mesenteric artery of sham-operated rats, dosed with solvent (sesame oil) (A), ovariectomized rats, dosed with sesame oil (B) and ovariectomized rats substituted with 17␤-estradiol (22 ␮g/kg/day) (C) for 4 weeks. Rings of arteries were suspended in organ chamber for the measurement of changes in isometric tension. They were exposed to charybdotoxin (10⫺7 M) and apamin (10⫺7 M) 30 min before addition of increasing concentrations of phenylephrine. All experiments were performed in the presence of indomethacin (10⫺5 M). Results are shown as means ⫾ SEM of 5–6 experiments. The asterisks indicate statistically significant differences (P ⬍ 0.05).

In the ovariectomized control group of rats, the potentiating effect of N␻-nitro-L-arginine on phenylephrine-induced contractile responses was also obtained after 1 week of treatment whereas no such effect was observed after 4 weeks of treatment (Figures 22.1 and 22.2). In addition, the combination charybdotoxin and apamin did not significantly affect contractile responses to phenylephrine in control ovariectomized rats treated for either 1 or 4 weeks with sesame oil (Figures 22.3 and 22.4).

Estrogen regulation of basal influence of NO and EDHF

179

3.3. Effect of short- and long-term treatment with 17␤-estradiol on the basal influence of NO and EDHF on vascular tone Substitution of ovariectomized rats with 17␤-estradiol was associated with a potentiation of the contractile responses to phenylephrine by N␻-nitro-L-arginine both after 1 and 4 weeks of treatment (Figures 22.1 and 22.2). Estrogenic substitution restored also the potentiating effect of the combination charybdotoxin and apamin on phenylephrine-induced contractions after 4 weeks of treatment whereas no effect was obtained after 1 week of treatment (Figures 22.3 and 22.4).

3.4. Relaxations to sodium nitroprusside and 1-EBIO In the presence of N␻-nitro-L-arginine, concentration–relaxation curves to sodium nitroprusside were significantly shifted to the right in arteries from the ovariectomized group compared to the sham-operated group (4 weeks of treatment, data not shown). Estrogenic substitution did not restore relaxations to sodium nitroprusside in the ovariectomized group of rats (data not shown). In the presence of N␻-nitro-L-arginine, 1-EBIO elicited similar concentration–relaxation curves in the 3 groups of rats investigated after 4 weeks of treatment (data not shown).

4. DISCUSSION In the mesenteric artery, as well as in many other types of arteries, the presence of a functional endothelium blunts contractile responses to vasoconstrictors. This influence of endothelial cells is due to the generation of NO as indicated by the leftward shift of the concentration–contraction curve to phenylephrine by the inhibitor of NO synthase, N␻nitro-L-arginine. In addition, it also involves EDHF-dependent mechanisms as suggested by the leftward shift of the concentration-contraction curve to phenylephrine by the combination charybdotoxin and apamin, two toxins known to inhibit EDHF-mediated responses. Since the endothelial control of vascular tone in the rat mesenteric artery is abolished by the combination N␻-nitro-L-arginine, charybdotoxin and apamin (M. Zerr, personal communication), NO and EDHF appear to be the two major endothelial factors involved in the control of vascular tone. The present findings indicate that the basal inhibitory effect of endothelium-derived NO and EDHF on the tone of the isolated mesenteric artery is controlled by the hormonal impregnation of mature female rats. Indeed, both the NO and EDHF components of the endothelial control of vascular tone vanish progressively following ovariectomy of rats. However the kinetic of the disappearance of the NO and EDHF components appears to be different. The endothelial NO-mediated inhibition of vascular tone persisted 2 weeks after ovariectomy but was abolished after 5 weeks. In contrast, the EDHF-mediated control of vascular tone was already abolished 2 weeks after ovariectomy. Thus, the basal EDHF-mediated control of vascular tone involves mechanism(s) that appear to be particularly sensitive to hormonal changes in rats. The present study indicates also that a daily subcutaneous injection of 17␤-estradiol (22 ␮g/kg) to ovariectomized rats restored to near physiological levels the estrogenic impregnation as indicated by the plasma level of 17␤-estradiol and the uterus wet weight. Moreover the estrogenic substitution restored both the basal NO and EDHF-dependent suppression of vascular tone after 4 weeks of treatment. Although the present study did

180

M. Zerr et al.

not address precisely the mechanisms involved in the protective effects of estrogens on NO and EDHF-dependent mechanisms, the data indicate that ovariectomy was associated with a significant decrease of the sensitivity of the vascular smooth muscle to the NO donor sodium nitroprusside whereas that to 1-EBIO, an activator of intermediate conductance calcium-activated potassium channel, was unaffected. Altogether, the restoration of the NO component by estrogen substitution in ovariectomized rats must be accompanied by a pronounced upregulation of the biological activity of nitric oxide in order to compensate for the decreased sensitivity of the vascular smooth muscle to the endothelial mediator.

5. CONCLUSION The hormonal impregnation of mature female rats is a crucial physiological parameter controlling the basal inhibitory effect of endothelium-derived NO and EDHF on the tone of the mesenteric artery.

23 Ascorbate inhibits EDHF in the bovine eye but not in the porcine coronary artery Alister J. McNeish, Silvia Nelli, William S. Wilson and William Martin

Endothelium-derived hyperpolarizing factor (EDHF)-mediated vasodilatation in the isolated perfused bovine eye is selectively inhibited by ascorbate (McNeish et al., 2002). The present study further investigated the selectivity of blockade in the eye and whether a similar blockade could be observed in rings of a large conduit artery, the porcine left anterior descending coronary artery. In the bovine perfused eye, acetylcholine-induced EDHF-mediated vasodilatation was powerfully inhibited by infusion of ascorbate. In contrast, vasodilatation induced by the KATP opener, levcromakalim, was unaffected by the infusion of ascorbate. Thus, the blockade of EDHF-mediated vasodilatation by ascorbate in the bovine eye appears to be highly selective. In rings of porcine coronary artery, the EDHF-mediated vasodilatation induced by bradykinin was unaffected by charybdotoxin, slightly inhibited by apamin, but virtually abolished by the combination of the two blockers. In the presence of ascorbate this EDHF-mediated vasodilatation induced by bradykinin was unaffected. Furthermore, ascorbate failed to cause further inhibition of vasodilatation when combined with either charybdotoxin or apamin alone. Thus, ascorbate does not appear to block calcium-sensitive potassium channels, at least in the porcine coronary artery. The data show that ascorbate inhibits EDHF-mediated vasodilatation in the bovine eye, but not in the porcine coronary artery. The possibility that this apparent anomaly is related to vessel size or to some other factor remains to be determined.

1. INTRODUCTION The ciliary body of the eye actively concentrates ascorbate from blood plasma, supplied through the ciliary vascular bed, into the aqueous humour of the eye (Millar and Kaufman, 1995). The concentration of ascorbate in human blood plasma is 46 ⫾ 8 ⫻ 10⫺6 M (range 30–150 ⫻ 10⫺6 M) (Keaney and Vita, 1995; Levine et al., 1996), whereas that in the aqueous humour may exceed 1 ⫻ 10⫺3 M (Davson, 1980; Halliwell and Gutteridge, 1989). As ascorbate scavenges superoxide anions (Som et al., 1983) the concentration of ascorbate may compensate for low levels of superoxide dismutase in the eye (Halliwell and Gutteridge, 1989). Ascorbate also has well-documented antioxidant effects in the vascular system, especially on nitric oxide-induced vasodilator responses. For example, ascorbate can recover impaired nitric oxide-dependent vasodilatation in isolated arterial rings which have been subjected to oxidant stress (Dudgeon et al., 1998; Fontana et al., 1999). Moreover, the ability of ascorbate to protect nitric oxide-dependent vasodilatation has been used therapeutically; treatment with ascorbate has resulted in recovery of impaired nitric oxide-dependent vasodilatation in patients with essential hypertension, chronic heart failure, atherosclerosis, hypercholesterolaemia and diabetes (for reviews see Carr et al., 2000; May, 2000).

182

Alister J. McNeish et al.

Vasodilatations attributed to endothelium-derived hyperpolarizing factor (EDHF) in the bovine isolated arterially perfused eye and the rat perfused mesenteric vascular bed are inhibited by ascorbate at concentrations which are within the physiological plasma range (McNeish et al., 2002). The ability of ascorbate to inhibit EDHF-mediated vasodilatation in the eye appeared to be selective, as that produced by the nitric oxide donor, glyceryl trinitrate, was unaffected by ascorbate (McNeish et al., 2002). The purpose of the present study was to establish further the selectivity of ascorbatemediated blockade of EDHF-dependent vasodilatation in the bovine isolated perfused eye, by assessing the effect of ascorbate on the response to an opener of KATP channels, levcromakalim. Furthermore, the inhibition of EDHF-mediated vasodilatation by ascorbate has, to date, only been observed in perfused vascular beds where the blood vessels controlling tone are resistance arterioles. Therefore, this study also investigated if ascorbate can inhibit EDHF in a large conduit blood vessel, the porcine left anterior descending coronary artery. 2. METHODS

2.1. Bovine isolated arterially perfused eye The ciliary vascular bed of the bovine eye was perfused using the constant flow method as previously described (McNeish et al., 2001). In brief, bovine eyes obtained from a local abattoir within 1 h of killing were cannulated through a long posterior ciliary artery and perfused at 37 ⬚C with Krebs solution containing (⫻ 10⫺3 M): NaCl 118; KCl 4.7; CaCl2 2.5; KH2PO4 1.2; MgSO4 1.2; NaHCO3 25; glucose 11.5; and gassed with O2 containing 5% CO2. Flow was commenced at ~0.2–0.5 ml/min and was raised in 5–10 increments to a final constant rate of 2.5 ml/min, over a 50-min period. When this final flow rate was achieved, eyes were perfused for a further equilibration period of at least 30 min. Perfusion pressure was measured using Gould Statham P32 ID transducers via a side arm located immediately proximal to the inflow cannula. Only eyes that had a basal perfusion pressure of 20–60 mmHg after the equilibration period were used for further study. In order to observe vasodilator responses in the bovine eye the perfusion pressure was first raised to ~130 mmHg using the thromboxane A2-mimetic, U46619 (3 ⫻ 10⫺7 M). In all experiments, responses to acetylcholine (1 ⫻ 10⫺10–1 ⫻ 10⫺7 mol) or levcromakalim (1 ⫻ 10⫺10–3 ⫻ 10⫺8 mol) were elicited by adding 10 ␮l of the appropriate concentration with a Hamilton micro-syringe. It has been previously demonstrated that vasodilator responses elicited by acetylcholine are mediated solely by an EDHF-like substance and do not involve a contribution by nitric oxide or a cyclooxygenase product (McNeish et al., 2001). Consequently, inhibitors of nitric oxide synthase or cyclo-oxygenase were not required to study EDHF-like responses in this preparation. Experiments were conducted to assess the effect of ascorbate (5 ⫻ 10⫺5 M) on acetylcholine- or levcromakalim-induced vasodilatation. In control experiments, full doseresponse curves to acetylcholine or levcromakalim were constructed and compared to those obtained following the infusion of ascorbate (50 ␮M, ⬎ 120 min).

2.2. Porcine left anterior descending coronary artery Porcine hearts were obtained from a local abattoir and transported to the laboratory in Krebs solution. The left anterior descending coronary artery was dissected out, cut into 2.5 mm ring segments, suspended between two steel hooks within 10 ml organ baths, stretched to 2 g tension and maintained at 37 ⬚C in Krebs solution containing L-NAME (1 ⫻ 10⫺4 M) and

Ascorbate inhibits EDHF in the bovine eye

183

indomethacin (3 ⫻ 10⫺6 M) gassed with 95% O2 and 5% CO2. Tension was recorded isometrically with Grass FTO3C transducers. Tissues were allowed to equilibrate for 60 min before experiments were carried out, during which time the resting tension was adjusted to 2 g, as required. In order to observe vasodilator responses, endothelium-containing rings of porcine coronary artery were contracted to about 60% (6.6 ⫾ 0.5 g) of the maximal U46619-induced tone: this level of tone was achieved with U46619 at 0.1–1⫻10⫺6 M. In order to observe EDHF-mediated vasodilatations to bradykinin, all rings of porcine coronary artery were incubated with the nitric oxide synthase inhibitor, L-NAME (1 ⫻ 10⫺4 M) and the cyclo-oxygenase inhibitor, indomethacin (3 ⫻ 10⫺6 M). These drugs were present throughout all experiments. In control rings, EDHF-dependent vasodilatations were elicited by the addition of bradykinin (1 ⫻ 10⫺10–3 ⫻ 10⫺7 M). The effects of the small conductance calcium-sensitive potassium channel (SKCa) inhibitor, apamin (1 ⫻ 10⫺7 M) and the large and intermediate conductance calcium-sensitive potassium channel (BKCa/IKCa) inhibitor, charybdotoxin (1 ⫻ 10⫺7 M), each alone and in combination, were assessed. Apamin and charybdotoxin were present for at least 20 min before effects on vasodilator responses were tested. In some rings charybdotoxin alone caused vasoconstriction. In such preparations the concentration of U46619 was adjusted so that the final level of tone was equal to that in control experiments. The effects of ascorbate were also assessed on EDHF-mediated, bradykinin-induced vasodilatation in the presence and absence of the SKCa and BKCa/IKCa blocking drugs apamin and charybdotoxin, each alone and in combination. In these experiments ascorbate (1.5 ⫻ 10⫺4 M) was included in the Krebs solution used to bathe the arterial rings and was present for at least 2 h before vasodilator responses were elicited.

2.3. Drugs and chemicals Acetylcholine chloride, apamin (from bee venom), ascorbic acid, indomethacin (free base), NG-nitro-L-arginine methyl ester (L-NAME) and U46619 (9,11-dideoxy-11␣,9␣-epoxymethanoprostaglandin F2␣) were purchased from Sigma (Poole, UK). Charybdotoxin (synthetic) was purchased from Latoxan (Valence, France). Levcromakalim was a gift from SmithKline Beecham Pharmaceuticals (Loughborough, UK). All drugs were dissolved in 0.9% saline except indomethacin (1 ⫻ 10⫺2 M stock), which was dissolved in sodium carbonate (1 M) and levcromakalim (1 ⫻ 10⫺2 M stock), which was dissolved in pure ethanol.

2.4. Statistical analysis Results are expressed as the mean ⫾ SEM of n separate observations, each from a separate eye or arterial ring. Vasodilator responses are expressed as percentage reduction of U46619induced perfusion pressure or tone. Graphs were drawn and statistical comparisons made using one-way analysis of variance with Bonferroni’s post-test, with the aid of a computer package, Prism (GraphPad, San Diego, USA). A probability (P) less than or equal to 0.05 was considered significant. 3. RESULTS

3.1. Ciliary vascular bed of the eye When bovine isolated eyes were perfused with normal Krebs solution, acetylcholine (0.1–100 ⫻ 10⫺9 mol) produced powerful, dose-dependent falls (max 59.5 ⫾ 4.5%, n ⫽ 8,

184 Alister J. McNeish et al. B 90

Control Ascorbate 50 µm

60

LK-induced vasodilatation (%)

ACh-induced vasodilatation (%)

A 70

50 40 30 20 10 0

Control Ascorbate 50 µm

75 60 45 30 15 0

–12

–11

–10 –9 –8 ACh (log mol)

–7

–6

–10

–9

–8 LK (log mol)

–7

Figure 23.1 Dose-response curves showing the effect of infusing ascorbate (50 ␮M, 120 min) on (A) acetylcholine (ACh)-induced, EDHF-mediated vasodilatation and (B) levcromakalim (LK)-induced vasodilatation in the bovine isolated perfused eye. Data represent the mean ⫾ SEM of 5–8 observations. **P ⬍ 0.005 and ***P ⬍ 0.001 indicate a difference from control.

Figure 23.1) in the U46619-induced perfusion pressure. These vasodilator responses are mediated entirely by an EDHF-like factor (McNeish et al., 2001). In eyes perfused with Krebs solution containing ascorbate (5 ⫻ 10⫺5 M, ⬎ 120 min) acetylcholine (0.1–100 ⫻ 10⫺9 mol)-induced, EDHF-mediated vasodilator responses were significantly attenuated (max 28.1 ⫾ 2.4%, n ⫽ 6, P ⬍ 0.001, Figure 23.1). Ascorbate also uncovered a normally masked vasoconstrictor response to acetylcholine (McNeish et al., 2002), but for ease of comparison, only the attenuation of the vasodilator response is shown here. Experiments were conducted to determine if vasodilator responses mediated by mechanisms other than EDHF could be inhibited by ascorbate. In control eyes perfused with normal Krebs solution the ATP-sensitive potassium channel (KATP) opener, levcromakalim (0.1–30⫻ 10⫺9 mol), produced, dose-dependent vasodilatation (max 80.3 ⫾ 3.4%, n ⫽ 6, Figure 23.1). Levcromakalim (0.1–30 ⫻ 10⫺9 mol)-induced vasodilatations were unaffected (max 80.3 ⫾ 3.2%, n ⫽ 7, Figure 23.1) by infusion of ascorbate (5 ⫻ 10⫺5 M, ⬎ 120 min).

3.2. Left anterior descending coronary artery EDHF-dependent vasodilatation was revealed by inclusion of the nitric oxide synthase inhibitor, L-NAME (1⫻10⫺4 M), and the cyclooxygenase inhibitor, indomethacin (3⫻10⫺6 M), in the bathing medium. In rings of porcine coronary artery bathed in normal Krebs solution the EDHF component of bradykinin (0.1–300 ⫻ 10⫺9 mol)-induced vasodilatation was powerful and concentration-dependent (max 80.3 ⫾ 6.5%, n ⫽ 10, Figure 23.2). In the presence of the BKCa/IKCa inhibitor, charybdotoxin (1 ⫻ 10⫺7 M), the bradykinin-induced, EDHF-mediated vasodilatation was unaffected (max 71.3 ⫾ 9.2%, n ⫽ 8, Figure 23.2). The SKCa inhibitor, apamin (1 ⫻ 10⫺7 M), appeared to slightly attenuate the bradykinin-induced EDHF-mediated vasodilatation (max 62.2 ⫾ 5.2%, n ⫽ 8, Figure 23.2), but this was not

B

Control Apamin ChTx ChTx+apamin

75

Bradykinin-induced relaxation (%)

A

Bradykinin-induced relaxation (%)

Ascorbate inhibits EDHF in the bovine eye

50

25

0

185

Control Apamin ChTx ChTx+apamin

75

50

25

0 –10

–9 –8 –7 Bradykinin (log M)

–6

–10

–9 –8 –7 Bradykinin (log M)

–6

Figure 23.2 Concentration-response curves showing bradykinin (100 pM–300 nM)-induced, EDHFmediated vasodilatation in rings of porcine left anterior descending coronary artery. These show the effects of the inhibitor of small conductance calcium-sensitive potassium channels (SKCa), apamin (100 nM), and of the intermediate conductance calcium sensitive potassium channels (IKCa), charybdotoxin (ChTx, 100 nM), each alone and in combination. Experiments were conducted in the absence (A) and presence (B) of ascorbate (150 ␮M). All responses were obtained in the presence of the nitric oxide synthase inhibitor, L-NAME, (100 ␮M), and the cyclo-oxygenase inhibitor, indomethacin (3 ␮M). Data represent the mean ⫾ SEM of 8–11 observations. *P ⬍ 0.05, **P ⬍ 0.005 and ***P ⬍ 0.001 indicate a difference from control.

statistically significant. In contrast, the combination of charybdotoxin plus apamin virtually abolished EDHF-mediated vasodilatation (max 11.3 ⫾ 3.3%, n ⫽ 10, P ⬍ 0.001, Figure 23.2). In rings of porcine coronary artery, bradykinin-induced, EDHF-dependent vasodilatation was unaffected (max 76.7 ⫾ 4.7%, n ⫽ 11, Figure 23.2) by ascorbate (1.5 ⫻ 10⫺4 M, ⬎ 120 min). Ascorbate also failed to affect bradykinin-induced EDHF-mediated vasodilatation in the presence of charybdotoxin (max 76.1 ⫾ 5.9%, n ⫽ 8, Figure 23.2). Furthermore, ascorbate failed to enhance the small blockade of EDHF-dependent vasodilatation produced by apamin.

4. DISCUSSION The results of the current study confirm that ascorbate (5 ⫻ 10⫺5 M) inhibits the EDHFmediated vasodilator responses in the ciliary vascular bed of the bovine isolated perfused eye (McNeish et al., 2002). In the bovine eye, vasodilatation mediated by the nitric oxide donor, glyceryl trinitrate, was unaffected suggesting that the blockade of EDHF-dependent vasodilator responses by ascorbate was selective (McNeish et al., 2002). The data from the current study support this conclusion. Specifically, vasodilatation induced by the KATP opener, levcromakalim, was unaffected by the infusion of ascorbate (5 ⫻ 10⫺5 M, ⬎ 120 min). Levcromakalim causes vasodilatation by opening of KATP channels; this leads to hyperpolarization of smooth muscle cells. EDHF-dependent vasodilatations are thought not to involve the opening of these glibenclamide-sensitive KATP channels (Corriu et al., 1996);

186

Alister J. McNeish et al.

indeed in the bovine eye EDHF-dependent vasodilatation is not affected by the infusion of glibenclamide (McNeish et al., 2001). Ascorbate inhibits EDHF-mediated vasodilator responses not just in the perfused bovine eye but also in the perfused rat mesentery (McNeish et al., 2002). Both these preparations are perfused vascular beds, so the vessels controlling the perfusion pressure are likely to be small resistance arterioles. The possibility that EDHF-mediated vasodilatation might also be inhibited by ascorbate in a large conduit artery was investigated using rings of the porcine left anterior descending coronary artery. EDHF-dependent vasodilatation and hyperpolarization have been studied extensively in this tissue (Beny and Schaad, 2000; Edwards et al., 2000; Edwards et al., 2001). In agreement with these reports, we found that EDHFmediated vasodilatation was uncovered by inhibiting nitric oxide synthase and cyclooxygenase. Inhibiting SKCa with apamin, or IKCa with charybdotoxin had no significant effect on bradykinin-induced, EDHF-mediated vasodilatation. The combination of charybdotoxin and apamin, however, virtually abolished the EDHF-mediated response. In porcine coronary arteries that had been treated with ascorbate (1.5 ⫻ 10⫺4 M, ⬎120min) the bradykinin-induced, EDHF-mediated vasodilatation was completely unaffected. This is in stark contrast to the powerful inhibitory effect of a lower concentration of ascorbate (5 ⫻ 10⫺5 M) in the perfused eye. It is unclear why physiological concentrations of ascorbate inhibit EDHF-mediated vasodilatation in the bovine eye but not in the porcine coronary artery, but several explanations are possible. For example, in the bovine eye, EDHF-mediated vasodilatation is powerfully inhibited by charybdotoxin alone and the blockade of EDHF by ascorbate was similar in character: both caused an inhibition of vasodilatation and unmasked a vasoconstrictor response to acetylcholine (McNeish et al., 2001). On this basis, it was possible that ascorbate could be inhibiting EDHF-mediated vasodilatation by blocking IKCa channels. It might therefore be expected that ascorbate alone would have no effect in the porcine coronary artery where blockade of both SKCa and IKCa channels is required to inhibit EDHF-mediated vasodilatation. However, in the porcine coronary artery, ascorbate failed to inhibit the EDHF-mediated vasodilatation in combination with charybdotoxin and failed to enhance the small blocking effect of apamin. Thus it appears that ascorbate does not block either SKCa or IKCa, at least not in the porcine coronary artery. Whether there is heterogeneity between KCa channels in the bovine eye and porcine coronary artery is unknown at present. Another factor that could potentially account for the difference in the ability of ascorbate to inhibit EDHF-mediated vasodilatation in the bovine eye and porcine coronary artery is vessel size. Specifically, the porcine coronary artery is a large conduit artery, whereas the blood vessels studied in the bovine eye are likely to be small resistance arterioles. In the rat mesenteric arterial bed, as vessel size decreases from the superior mesenteric artery to the third-order arteries, myoendothelial gap junctional plaque expression increases (Sandow and Hill, 2000). Moreover, the effectiveness of inhibitors of gap junctional communication to block EDHF responses (Chaytor et al., 1998; Chaytor et al., 2001) increases with decreasing vessel size (Berman et al., 2002). Therefore, if ascorbate were inhibiting EDHF-mediated vasodilatation by interfering with gap-junctional communication it may explain why it is effective in the perfused bovine eye but not in the porcine coronary artery. Another difference between the perfused bovine eye and porcine coronary artery is that the ciliary body of the eye has the ability to concentrate ascorbate from blood plasma into the aqueous humour (Millar and Kaufman, 1995). This ability to concentrate ascorbate may result in high local concentrations of ascorbate in the eye and this could potentially underlie the inhibition of EDHF. This argument seems unlikely, however, since EDHF-mediated

Ascorbate inhibits EDHF in the bovine eye

187

vasodilatation is inhibited by ascorbate in the rat perfused mesenteric arterial bed (McNeish et al., 2002), which lacks the ability to concentrate ascorbate. In summary, ascorbate at concentrations normally found in plasma (5 ⫻ 10⫺5 M) can inhibit EDHF-mediated vasodilatation in the ciliary vascular bed of the bovine isolated perfused eye. This inhibition of vasodilatation appears to be selective as vasodilatation induced by levcromakalim was completely unaffected. In contrast to the bovine eye, EDHFmediated vasodilatation in the porcine coronary artery was completely unaffected by ascorbate (up to 1.5 ⫻ 10⫺4 M). Furthermore, ascorbate failed to increase the blockade produced by the SKCa inhibitor, apamin, and the IKCa inhibitor, charybdotoxin, either alone or in combination, indicating that ascorbate is unable to inhibit either of these potassium channels. The blockade of EDHF by ascorbate in the bovine eye but not in the porcine coronary artery may indicate that vessel size or some other factor is a major determinant of the blocking action of ascorbate.

24 Gabexate mesilate inhibits endothelium-dependent relaxation, but causes endothelium-independent relaxation of rat blood vessels Mikio Nakashima, Tomoko Hamada, Shinji Mitsumizo and Tadahide Totoki Gabexate mesilate is an anti-inflammatory serine protease inhibitor used to treat pancreatitis and disseminated intravascular coagulation. Drip infusion of this agent into peripheral veins can cause side effects such as thrombophlebitis or panniculitis. The mechanism underlying these side effects is unknown. The protease inhibitor inhibits the production of H2O2, one of the candidates as endothelium-derived hyperpolarizing factor (EDHF). The present study was designed to elucidate the effects of gabexate mesilate on vascular smooth muscle cells and the endothelium. Mesenteric arteries and saphenous veins of the rat were used for the measurement of isometric tension recordings. Gabexate mesilate relaxed the arterial rings contracted with either phenylephrine or high potassium in a concentration-dependent manner. The relaxations were observed regardless of the presence or absence of both nitro-L-arginine and indomethacin. The protease inhibitor relaxed venous rings, but high concentration of the drug caused transient but significant contraction. After intra-arterial infusion of gabexate mesilate, acetylcholine induced, endothelium-dependent relaxations were inhibited. The present study demonstrated that gabexate mesilate causes endothelium-independent relaxation of the rat blood vessels and that high concentration of the drug produces transient contraction of the saphenous vein, which might be related to its venous side effects. The present data also demonstrate that the drug inhibits endothelium-dependent relaxations of the mesenteric artery.

1. INTRODUCTION Gabexate mesilate, [ethyl p-(6-guanidinohexanoyloxy) benzoate methanesulfonate (FOY, Ono Pharmaceutical, Osaka, Japan)] is an anti-inflammatory serine protease inhibitor (Muramatsu and Fujii, 1968) which is widely used as a therapeutic agent for acute pancreatitis (Freise et al., 1985) and disseminated intravascular coagulation (Taenaka et al., 1982). After drip infusion of this agent into peripheral veins, complications such as panniculitis, skin ulcer, and superficial thrombophlebitis have been reported (Nakayama, 1997). Although it has been suggested that these complications are caused by injury of blood vessels due to the administration of a high concentration of gabexate mesilate, the precise mechanism underlying them is unknown. Moreover, the synthetic protease inhibitor inhibits the production of various activated oxygen radicals (Tamura et al., 1992), such as superoxide, hydroxyl radical, and hydrogen peroxide – one of the candidates of endothelium-derived hyperpolarizing factor EDHF (Matoba et al., 2000; Matoba et al., 2002). The present study was designed to study the effects of gabexate mesilate on the vasomotor function. 2. METHODS Male Sprague–Dawley rats aged 7–9 weeks and weighing 200–270 g were used (Kyudo, Fukuoka, Japan). The main branches of the superior mesenteric arteries and saphenous vein

Effects of gabexate mesilate on rat blood vessels

189

were excised under anesthesia with diethyl ether inhalation and placed on a plate containing cold modified Krebs-Ringer bicarbonate solution [millimolar concentration: NaCl 118.3, KCl 4.7, CaCl2 2.5, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25.0, edetate calcium disodium (CaEDTA) 0.026, and glucose 11.1] (control solution) aerated with 95% O2 and 5% CO2. In some experiments, mesenteric artery was cannulated with a 24-gauge catheter (SurfloTM, Terumo Co., Ltd., Tokyo, Japan), and either gabexate mesilate (10⫺3 M) or saline was perfused through the catheter for 5 min. The blood vessels were cleaned of adherent connective tissue and cut into rings (1 mm long). Some experiments were performed in the presence of indomethacin (10⫺5 M) or L-nitroarginene methyl ester (L-NAME; 10⫺4 M) to inhibit the formation of vasoactive prostanoids or nitric oxide, respectively.

2.1. Organ chamber studies The arterial or venous rings were suspended in organ chambers (5 ml) filled with control solution (37 ⬚C) aerated with 95% O2 and 5% CO2, by means of two stainless-steel stirrups. One of the stirrups was anchored to the bottom of the chambers and the other was connected to an isometric force transducer (Statham Universal UC2, Nihon Kohden, Tokyo, Japan). After 45 min of incubation, the blood vessels were stretched in a stepwise manner to the optimal point of their length–tension relationship (approximately 1g; determined in preliminary experiments with 45 mM KCl). After 45 min of equilibration, the preparations were exposed to increasing concentrations of gabexate mesilate (10⫺6 – 3 ⫻ 10⫺3 M) during contraction to phenylephrine (10⫺6 – 3 ⫻ 10⫺6M) or KCl (60 mM). Relaxations were expressed as a percentage of the active contractions. The following drugs were used: acetylcholine, indomethacin, phenylephrine, nitro-L-arginine (Sigma, St Louis, MO, USA), and gabexate mesilate (Nippon Baruku, Osaka, Japan). All the other drugs were dissolved in distilled water. Stock solutions of 10⫺5 M indomethacin were prepared in an equal molar concentration of Na2CO3. The results are expressed as means ⫾ SEM. In each experiment, N indicates the number of tissues from different rats tested. Statistical analysis was performed with the paired or unpaired Student-t test (two tailed). When P was less than 0.05, differences were considered to be statistically significant.

3. RESULTS In the mesenteric arteries, gabexate mesilate (10⫺6 – 3 ⫻ 10⫺3 M) induced concentrationdependent relaxations in tissues contracted with phenylephrine (n ⫽ 4, Figure 24.1). The threshold concentration of gabexate mesilate to induce relaxation was 3 ⫻ 10⫺5 M. There was no significant difference in relaxation to gabexate mesilate after incubation with indomethacin or L-NAME, alone or in combination (Figure 24.2). Arterial rings contracted with KCl also relaxed on exposure to gabexate mesilate; L-NAME or indomethacin did not modify the relaxation. In saphenous vein contracted with phenylephrine, gabexate mesilate (10⫺6 to 3 ⫻ 10⫺4 M) produced relaxation in a concentration-dependent manner (Figure 24.3). At 10⫺3 M, gabexate mesilate produced a biphasic response with a transient contraction and a subsequent relaxation. Incubation with indomethacin and L-NAME augmented the relaxation and contraction with gabexate mesilate (10⫺6 – 3 ⫻ 10⫺4 M) (Figure 24.3). In arterial rings perfused with gabexate mesilate, acetylcholine-induced relaxations were attenuated significantly compared with controls (Figure 24.4).

Figure 24.1 Effects of gabexate mesilate on rat mesenteric arteries contracted with phenylephrine. Results were expressed as percentage of contraction to the latter, and are shown as means ⫾ SEM (n ⫽ 4). Gabexate mesilate was applied in a cumulative manner. There was no significant difference in relaxation to gabexate mesilate in the presence or absence of indomethacin (10⫺5 M) or L-nitroarginene methyl ester (L-NAME; 10⫺4 M), alone or in combination.

Figure 24.2 Effects of gabexate mesilate on rat mesenteric arteries contracted with KCl. Results were expressed as percentage of contraction to KCl (60 mM), and are shown as means ⫾ SEM (n ⫽ 4). Gabexate mesilate was applied in a cumulative manner. Arterial rings contracted with KCl also relaxed to gabexate mesilate, in the presence or absence of L-nitroarginene methyl ester (L-NAME; 10⫺4 M) plus indomethacin (10⫺5 M).

Figure 24.3 Effects of gabexate mesilate on a rat saphenous vein contracted with phenylephrine. Results were expressed as percentage of contraction to ␣1-adrenergic agonist, and are shown as means ⫾ SEM (n ⫽ 4). Gabexate mesilate was applied in a cumulative manner. 10⫺3 M of gabexate mesilate produced biphasic response with transient contraction and a subsequent relaxation. Incubation with indomethacin (10⫺5 M) plus L-nitroarginene methyl ester (L-NAME; 10⫺4 M) augmented the relaxation and contraction with gabexate mesilate. The asterisks indicate statistically significant differences (P ⬍ 0.05).

A

B

Figure 24.4 Effect of acetylcholine on rat mesenteric arteries contracted with phenylephrine. The arteries were perfused by either gabexate mesilate (10⫺3 M; A) or saline (B) through a 24-gauge catheter for 5 min prior to the application of acetylcholine.

192

Mikio Nakashima et al.

4. DISCUSSION The present experiments demonstrate that gabexate mesilate causes relaxation in rat mesenteric arteries and saphenous veins. The relaxation to gabexate mesilate is not due to nitric oxide, vasoactive prostanoids or EDHF, since L-NAME, indomethacin or KCl did not modify it. These findings suggest that gabexate mesilate acts as a vasodilator in the rat blood vessels through endothelium-independent mechanisms. Several possible mechanisms could underlie the vasorelaxation to gabexate mesilate. First, kallistatin, a serine protease inhibitor like gabexate mesilate induces endotheliumindependent relaxation of the aorta and the renal artery of the rat (Chao et al., 1997). Moreover, nafamostat mesilate, another serine protease inhibitor, also produces endotheliumindependent relaxations (unpublished observation). These results suggest that serine protease inhibitors in general act as vasodilators. The present study also demonstrated that gabexate mesilate produced a transient, but significant contraction in saphenous veins. The involvement of thromboxane A2 is unlikely, since the cyclooxygenase inhibitor indomethacin did not inhibit the responses. Gabexate mesilate is generally used in patients with acute pancreatitis and disseminated intravascular coagulation. Thrombophlebitis and other venous complications are reported when this agent is infused in peripheral veins at the high concentration. The mechanism underling this venous complication is unclear. The present data indicate that venous constriction might be one of the possible mechanisms for the venous damage. The present findings suggest that gabexate mesilate inhibits endothelium-dependent relaxations. Gabexate mesilate inhibits the production of various activated oxygen radicals including hydrogen peroxide, which are thought to be endothelium-derived hyperpolarizing factor in mice (Matoba et al., 2000) and human (Matoba et al., 2002). Such an effect could contribute to the blunting of the endothelium-dependent relaxation to acetylcholine. However, in the present study, the effects of gabexate mesilate on hydrogen peroxide were not examined further.

25 Mechanisms underlying basal vascular tone in the guinea-pig mesenteric arterioles Yoshimichi Yamamoto and Hikaru Suzuki

The electrical and mechanical activities of everted segments, or segments without adventitia of mesenteric arterioles dissected from guinea-pigs were observed using conventional or perforated whole-cell clamp method and video-edge detection. The arteriolar segments maintained some basal tone under control conditions, since they dilated when the external Ca2⫹ was removed. The arteriolar diameter was randomly fluctuating due to small and local contractions. The basal tone persisted after removing the adventitial layer, which contained the perivascular nerves. An involvement of the endothelium appeared unlikely as disruption of the endothelium failed to abolish the fluctuating basal tone. When the membrane potential of a smooth muscle cell was recorded simultaneously with the arteriolar diameter, there were no correlations between them. A gap junction blocker, 18␤-glycyrrhetinic acid reduced the arteriolar basal tone. As high K⫹ solution induced comparable contractions in the absence and the presence of 18␤-glycyrrhetinic acid, this substance did not seem to affect the voltage-dependent Ca2⫹ channels and the contractility of the smooth muscle. A blocker of IP3-induced Ca2⫹ release from the endoplasmic reticulum, 2-aminoethoxydiphenyl borate dose-dependently decreased the vascular tone. These results indicate that movement of either IP3 or Ca2⫹ through the gap junctions between smooth muscle cells may be important to generate basal tone.

Peripheral vascular resistance is important to maintain arterial blood pressure and a normal blood circulation. Smooth muscle cells in the peripheral resistance arteries and arterioles contract partially under physiological conditions and play a key role in producing the basal tone of these blood vessels. The mechanisms underlying this basal tone are still not fully understood. As most of the blood vessels are innervated by the autonomic nerves, neural activity may be involved. The endothelium may also be playing a role in the production of the basal tone, because the endothelial cells release not only relaxing factors but also constricting substances such as endothelin (Yanagisawa et al., 1988). In the irideal arterioles of the young rats, rhythmical contractions are superimposed on the basal tone and this mechanical activity is independent of the autonomic nerves and the endothelium (Hill et al., 1999; Haddock et al., 2002). Inositol 1,4,5-trisphosphate (IP3) and intercellular communication via gap junctions may be involved in this rhythmic behavior (Haddock et al., 2002), but the details are still not well documented. In the present experiments, electrical and mechanical activities were recorded simultaneously in guineapig mesenteric arterioles, where fluctuating basal tone was also observed. The results obtained indicate that movement of either IP3 or Ca2⫹ through gap junctions between the smooth muscle cells may be important to produce the basal tone.

194

Yoshimichi Yamamoto and Hikaru Suzuki

1. METHODS

1.1. Preparations Male guinea-pigs, weighing 250–400 g, were anesthetized by sevoflurane inhalation and killed with an animal guillotine. The ileum was excised and the first-order arterioles were dissected out (diameter; 50–100 ␮m). For the experiments in which electrical recordings were made from the smooth muscle cells, the arteriole was incubated in physiological saline containing 0.5 mg/ml collagenase (Wako Pure Chemical Industries, Ltd., Osaka, Japan) for 15 min at 35 ⬚C, and then the adventitial layer was removed mechanically to reveal the smooth muscle layer. When electrical recordings were made from the endothelial cells, the arteriole segment was everted; no enzymatic treatment was used. The everted preparations were also used in most of the experiments where only the mechanical responses were observed. One-millimeter segments of these preparations were pinned out in a small recording chamber (0.8 mm wide ⫻ 1.5 mm long), which was placed on the stage of an inverted microscope (Nikon Diaphot, Tokyo, Japan). Preparations were superfused with preheated (35 ⬚C) and aerated (100% O2) bath solution at a constant rate (1 ml/min).

1.2. Mechanical activity The image through an objective lens (⫻ 40) was captured by a CCD camera (FTM800, Philips Electronics, New York, NY, USA) and the distance between both edges of the preparation was detected with a video-edge motion detector (VED-104, Crescent Electronics, Sandy, UT, USA).

1.3. Membrane potential The tight-seal patch-clamp method was applied in conventional or perforated whole-cell clamp configuration using gramicidin (Ebihara et al., 1995) mostly under current clamp conditions. Voltage or current signals were acquired using an EPC-7 patch-clamp system (List-Medical, Darmstadt, Germany), then digitized with a Digidata 1322A data acquisition system (Axon Instruments, Union City, CA, USA). The sampling rate was always five times higher than the cut-off frequency (-3 dB) of a low-pass Bessel filter. Current or voltage clamp protocols were controlled with pCLAMP 8 software (Axon Instruments).

1.4. Solutions and drugs The composition of control bath solution was (mM): NaCl 141.5; KCl 5.4; CaCl2 1.8; MgCl2 1; Hepes 10; glucose 5. The pH was adjusted to 7.3 with NaOH. High-K⫹ solution was prepared by equimolar replacement of NaCl with KCl. The pipette solution for conventional whole-cell clamp experiments contained (mM): KCl 143; MgCl2 1; EGTA 10; Hepes 10; glucose 5. The pH was adjusted to 7.3 with KOH. The pipette solution for perforated whole-cell clamp experiments contained: KCl 150 mM; Hepes 10 mM; gramicidin D 20–50 ␮g/ml. Acetylcholine, 18␤-glycyrrhetinic acid, gramicidin D and nifedipine were obtained from Sigma (St Louis, MO, USA), 2-aminoethoxydiphenyl borate (2APB) from Calbiochem (San Diego, CA, USA) and triton X-100 from Katayama Chemical (Osaka, Japan). 2APB, 18␤-glycyrrhetinic acid, gramicidin D and nifedipine were dissolved in dimethyl sulfoxide to make stock solutions.

Basal tone in arteriole

195

2. RESULTS

2.1. Basal tone In most of the experiments where only the mechanical activity was observed, the everted preparations were used because their smooth edges were suitable for the video-edge detection. After a stabilizing period, small random fluctuations in diameter could be observed (Figure 25.1). Such fluctuations were also observed in the intact arterioles and in arterioles without adventitia. When Ca2⫹ was removed from the superfusate, the arteriole dilated indicating that the vessel exhibited certain magnitude of tone under control conditions (basal tone, Figure 25.1). Using everted preparations, the membrane potential of the endothelial cell could be measured with the perforated whole-cell clamp method. When acetylcholine (3 ⫻ 10⫺6 M) was applied, the membrane of the endothelial cell hyperpolarized and the arteriole dilated even though the blood vessel was not contracted (Figure 25.1). As acetylcholine-induced hyperpolarization of the endothelial cells conducts to the smooth muscle cells with the efficiency of more than 80% (Yamamoto et al., 2001), this dilatation was most likely to be induced by the hyperpolarization of the smooth muscle cells. To evaluate the involvement of the endothelium in the fluctuating mechanical activity, the preparation was exposed to a solution containing triton X-100 (1%) for 5 s. This procedure effectively disrupted the endothelium without significant damage to the smooth muscle cells. Even after the endothelium had been disrupted, the arteriole still exhibited a fluctuating basal tone (Figure 25.1).

2.2. Membrane potential and fluctuation of the diameter Fluctuations in arteriolar diameter might be related to changes in the membrane potential of the smooth muscle cells. When the membrane potential of a smooth muscle cell was

Figure 25.1 Basal tone observed in the guinea-pig mesenteric arteriole. (A) Effect of Ca2⫹ removal on the diameter of an everted arteriole. In all figures, an upward deflection indicates decrease in the diameter (muscular contraction). (B) Effect of acetylcholine (ACh) on the endothelial membrane potential and the arteriolar diameter. Simultaneous recording in an everted arteriole. (C) Fluctuating contraction of an arteriole in which the endothelium had been disrupted. All recordings were from different preparations.

196

Yoshimichi Yamamoto and Hikaru Suzuki

Figure 25.2 Relationship between the membrane potential of a smooth muscle cell (A) or an endothelial cell (B) and the fluctuating contractions. The membrane potential (perforated whole-cell clamp method, a) and the arteriolar diameter (b) were measured simultaneously in a preparation without adventitia (A) or in an everted preparation (B). A and B were from different preparations.

Figure 25.3 Effect of 18␤-glycyrrhetinic acid (18␤-GA). (A) Membrane potential of a smooth muscle cell (perforated whole-cell clamp method, a) and the arteriolar diameter (b) were measured simultaneously in a preparation without adventitia. (B) High K⫹ (40 mM) solution was applied in the absence and the presence of 18␤-glycyrrhetinic acid while the arteriolar diameter was observed. A and B were from different preparations.

recorded simultaneously with the arteriolar diameter in a preparation without adventitia, however, changes in membrane potential were not correlated to the fluctuations in the diameter (Figure 25.2). This was also the case when the membrane potential of an endothelial cell was recorded simultaneously with the arteriolar diameter of an everted preparation (Figure 25.2).

2.3. Gap junctions and basal tone 18␤-glycyrrhetinic acid is a potent blocker of the gap junctions in the studied preparation (Yamamoto et al., 1998, 1999). When 18␤-glycyrrhetinic acid was applied in a preparation without adventitia, the arteriole dilated and the membrane of the smooth muscle cell depolarized (Figure 25.3). This 18␤-glycyrrhetinic acid-induced depolarization was variable among preparations ranging from a few to some 20 mV. As 18␤-glycyrrhetinic acid might have some other effects than blocking gap junctions and dilate the arteriole, constriction induced by a high concentration of K⫹ was compared in the absence and the presence of the gap junction blocker (Figure 25.3). The 40 mM-K⫹ solution induced similar constrictions before and after the arteriole was dilated by the gap junction blocker.

2.4. Voltage-dependent Ca2⫹ channels The effect of nifedipine was examined to evaluate the involvement of L-type voltagedependent Ca2⫹ channels. Nifedipine, at 10-5 M, did not affect the basal tone in most

Basal tone in arteriole

197

Figure 25.4 Effect of nifedipine (A) or 2APB (B) on arteriolar diameter. A and B were from two different everted arterioles.

preparations (Figure 25.4). In some arterioles, however, this concentration of nifedipine reduced vascular tone slightly (results not shown).

2.5. IP3 2APB is known to block IP3-induced Ca2⫹ release from the endoplasmic reticulum (Cui and Kanno, 1997; Maruyama et al., 1997). 2APB was applied at concentrations between 10⫺5 and 10⫺4 M in everted arterioles. 2APB decreased vascular tone dose-dependently and the maximal effect was observed at the concentration of 10⫺4 M (Figure 25.4). 3. DISCUSSION In the present experiments, dissected segments of guinea-pig mesenteric arterioles exhibited some basal tone under control conditions, and Ca2⫹ removal or application of acetylcholine increased the arteriolar diameter. Involvement of the perivascular nerves is unlikely because the arteriolar tone persisted after removal of the adventitial layer where the nerve fibers run. An involvement of the endothelium is also unlikely because the fluctuating arteriolar tone could be recorded in preparations in which the endothelium had been disrupted. The magnitude of the basal tone was not constant but always fluctuating randomly. In the irideal arterioles of young rats, rhythmical contractions can be observed and they are always preceded by large, slow depolarizations which are thought to be due to an unknown Ca2⫹activated channel (Hill et al., 1999; Haddock et al., 2002). In the present experiments, on the other hand, the fluctuations of the arteriolar tone were smaller than those in the rat irideal arterioles and not accompanied by electrical events at the cell membrane. More smooth muscle cells seem to be contracting synchronously in the rat irideal arterioles than in the guinea-pig mesenteric arterioles. In the latter, the fluctuation seemed to be a non-conducting random change in the basal tone of the individual smooth muscle cells. Morphological observations indicate that the endothelial and smooth muscle layers are separated electrically in the rat irideal arterioles (Hirst et al., 1997), while the two layers are electrically coupled to each other in the guinea-pig mesenteric arterioles (Yamamoto et al., 2001). This difference in intercellular electrical coupling might be a reason for the difference in the number of coordinating smooth muscle cells. As an electrotonic potential produced in a smooth muscle cell by current injection diminishes steeply with distance as it spreads along the muscle layer in the guinea-pig mesenteric arterioles (Yamamoto et al.,

198

Yoshimichi Yamamoto and Hikaru Suzuki

2001), change in membrane potential of one smooth muscle cell could hardly be observed with an electrode placed on another cell which was more than several cells away from the origin cell. In the present experiments, because of technical difficulties, the portion where the diameter was measured was not exactly the same as the portion where the patched cell resided. So the membrane potential of the fluctuating cell itself could exhibit the coordinating change in the membrane potential even though it did not conduct effectively to the neighboring smooth muscle cells. If the coupling resistance between the smooth muscle cells is relatively low in the rat irideal arterioles, many smooth muscle cells would contract coordinately and the electrical event could be detected from any one of these cells. Blockade of gap junctions with 18␣-glycyrrhetinic acid abolishes the rhythmic electrical and mechanical activities and reduces vascular tone in the rat irideal arterioles (Hill et al., 1999). Another gap junction blocker, 18␤-glycyrrhetinic acid also reduced the vascular tone in the guinea-pig mesenteric arterioles in the present experiments, and this effect was accompanied by membrane depolarization. Nifedipine did not affect the vascular tone in most of the preparations used in the present experiments. This does not necessarily mean that the smooth muscle cells do not have voltage-dependent Ca2⫹ channels, because nifedipine-insensitive, rapidly inactivating, voltage-dependent Ca2⫹ channels have been found in the terminal branches of guinea pig mesenteric artery (Morita et al., 1999). The voltage-dependent Ca2⫹ channels and the contractile apparatus seemed to be intact in the presence of 18␤-glycyrrhetinic acid, because high concentration of K⫹ could produce comparable contractions in the presence and the absence of the gap junction blocker. So, the voltage-dependent Ca2⫹ channels were likely to be opened by the membrane depolarization in the presence of the gap junction blocker. Yet the blood vessel dilated indicating that the overall [Ca2⫹]i should be decreased by 18␤-glycyrrhetinic acid in spite of the Ca2⫹ influx through the voltage-dependent Ca2⫹ channels. If the ionic currents passing through the gap junctions make the membrane potential of the smooth muscle cell somewhat depolarized and produce the basal tone, the membrane would be hyperpolarized by the gap junction blocker. It was not the case in the present experiments, and substances other than ionic currents traveling through the gap junctions seem to be responsible in the generation of the basal tone. 2APB dose-dependently reduced vascular tone, suggesting an involvement of IP3 in the production of the basal tone. Under control conditions, IP3 seems to be produced constitutively, releasing a certain amount of Ca2⫹ from the sarcoplasmic reticulum, which would keep the smooth muscle cells partially contracted. If the constitutive production of IP3 occurs not in every but in some special smooth muscle cells, IP3 might have to spread into the neighboring ordinary smooth muscle cells to cause the contraction of these cells. As the intercellular Ca2⫹ waves are mediated by a passive diffusion of IP3, but not Ca2⫹, through gap junctions (Churchill et al., 1998; Sneyd et al., 1998; Fry et al., 2001), IP3 seems to be able to travel from one smooth muscle cell to another via the gap junctions. On the other hand, the possibility that Ca2⫹ travels through the gap junctions remains because this phenomenon can mediate the propagation of the intercellular calcium waves (Höfer et al., 2001). In conclusion, there seems to be smooth muscle cells where IP3 is constitutively produced in the guinea-pig mesenteric arterioles. Movement of either IP3 itself or Ca2⫹ through the gap junctions from such special cells to the surrounding ordinary smooth muscle cells may be important to produce the basal tone of the whole arteriolar segment.

26 Endothelium-dependent depolarization and its implications for endotheliumderived hyperpolarizing factor Harold A. Coleman, Marianne Tare and Helena C. Parkington

Stimulation of the endothelium characteristically evokes hyperpolarization and relaxation of the underlying smooth muscle. While the release of vasodilating factors such as nitric oxide and prostacyclin may be accompanied by hyperpolarization of the smooth muscle, hyperpolarization is a defining feature of, and may be obligatory for the relaxation evoked by endothelium-derived hyperpolarizing factor (EDHF). This chapter discusses evidence that indicates that the same endothelial stimulants which evoke hyperpolarization and relaxation can also invoke an endothelium-dependent depolarizing and constricting influence on vascular smooth muscle. Under normal conditions the hyperpolarization dominates, but if the hyperpolarization is blocked with K⫹ channel blockers or reduced by disease, then depolarization can be revealed. This depolarization has important implications for the interpretation of the effects of disease on EDHF, and also presents therapeutic potential.

1. INTRODUCTION In some blood vessels, the hyperpolarization evoked by acetylcholine or bradykinin and attributed to EDHF is followed by a depolarizing phase. This is evident in recordings of membrane potential made from rabbit mesenteric arteries (Murphy and Brayden, 1995b), canine coronary arteries (Mombouli et al., 1996), porcine coronary/rat aorta bioassay system (Popp et al., 1996), guinea-pig submucosal arterioles (Hashitani and Suzuki, 1997; Yamamoto et al., 1999; Coleman et al., 2001a), guinea-pig coronary arteries (Parkington et al., 1995; Nishiyama et al., 1998) and cheek pouch arterioles of anaesthetized hamsters (Welsh and Segal, 1998). Furthermore, in some studies in which the EDHF-mediated hyperpolarization was blocked with charybdotoxin and/or apamin, endothelial stimulation resulted in depolarization in mesenteric arteries of rabbits (Murphy and Brayden, 1995b) and rats (Tare et al., unpublished observations), guinea-pig carotid arteries (Corriu et al., 1996a) and submucosal arterioles (Figure 26.1) (Hashitani and Suzuki, 1997; Coleman et al., 2001a). Although a direct effect of agonists on the smooth muscle can occur in some tissues, particularly at higher concentrations (Bolton et al., 1984; Inoue and Kuriyama, 1993; Large, 2002) (Figure 26.2), an endothelium dependence has been demonstrated in some studies (Murphy and Brayden, 1995b; Corriu et al., 1996a; Coleman et al., 2001c). Several studies indicated the involvement of a diffusible factor (Mombouli et al., 1995; Popp et al., 1996) that was referred to as endothelium-derived depolarizing factor (EDDF) (Mombouli et al., 1995). The nature of the putative factor(s) and processes underlying the depolarization were not elucidated. The existence of an EDDF raises the possibility that the hyperpolarizing actions of EDHF are counteracted and diminished by this/these influence(s). This article considers some of the endothelium-dependent factors that can evoke depolarization of blood vessels and their

Figure 26.1 Endothelium-dependent depolarization revealed by block of EDHF. Submucosal arterioles were depolarized and constricted with Ba2⫹ (10⫺4 M) and the endothelium was stimulated with acetylcholine (10⫺6 M) to evoke hyperpolarization (upper trace) and relaxation (lower trace). EDHF-attributed ion channels were blocked using apamin (2.5 ⫻ 10⫺7 M) plus charybdotoxin (ChTx, 3–5 ⫻ 10⫺8 M). The results are from a continuous recording from the same cell. (Reproduced with permission from Coleman et al., 2001a.)

Figure 26.2 Schematic summary of endothelium-dependent depolarizing mechanisms in blood vessels. Depolarization can result from the activation of endothelin receptors (ET), muscarinic receptors (M), thromboxane A2 receptors (TP), and bradykinin receptors (B), and may involve the generation of inositol trisphosphate (IP3), reactive oxygen species (ROS), and unknown mechanisms resulting from emptying of the internal Ca2⫹ stores. Due to electrical coupling via myoendothelial gap junctions (MEGJ) between endothelial and smooth muscle cells in some but not all vessels, a depolarizing influence in one cell type will spread electrically to the other cell type. The main ionic mechanisms underlying depolarization are likely to involve the activation of poorly selective cation channels (PSCC), which include NSCC1, NSCC2 and SOCC. At least some of the PSCC are likely to belong to the transient receptor potential (TRP) channel family. Additional depolarizing mechanisms include Ca2⫹-activated Cl⫺ channels, and the inhibition of various types of K⫹ channels, including voltage-activated (KV), inward rectifier (KIR), and large conductance Ca2⫹-activated (BKCa) K⫹ channels.

Endothelium-dependent depolarization and its implications

201

underlying mechanisms. Since endothelial and smooth muscle cells are electrically coupled in a range of vessels (reviewed by Bény, 1999; Coleman et al., 2002), the effects of agents that depolarize endothelial cells also need to be considered.

2. ENDOTHELIN Endothelin-1, one of the most potent constricting factors released from the vascular endothelium, evokes changes in membrane potential in both smooth muscle and endothelial cells.

2.1. Smooth muscle Endothelin-1 activates an inward current that is carried by poorly selective cation channels (Chen and Wagoner, 1991; Nakajima et al., 1997; Guibert and Beech, 1999). The inward current activated by endothelin ETA receptors involves three different types of cation channels, all permeated by Ca2⫹, and termed NSCC-1, NSCC-2 (non-selective cation channels), and SOCC (a store-operated channel) (Iwamuro et al., 1999; Zhang et al., 1999; Kawanabe et al., 2002) (Figure 26.2). NSCC-1 was activated via a G12 protein-dependent pathway, blocked by LOE-908 but not by SKF96365, which are agents that can block some cation channels, and contributed about 20% to the increase in cytoplasmic free Ca2⫹ evoked by endothelin. NSCC-2 was activated via G12 and Gq protein-dependent pathways, blocked by both LOE-908 and by SKF96365, and contributed about 40% to the increase in cytoplasmic free Ca2⫹. SOCC was activated via a Gq/phospholipase C-dependent pathway, blocked by SKF96365 but not by LOE-908, and contributed about 40% to the increase in cytoplasmic free Ca2⫹ (Iwamuro et al., 1999; Kawanabe et al., 2002). Since endothelin ETA receptors are coupled to the Gq/phospholipase C pathway, the resulting inositol trisphosphate-induced Ca2⫹ release from intracellular stores could be expected to activate Ca2⫹-dependent ion channels. Activation by endothelin of Ca2⫹dependent Cl⫺ channels, which would evoke depolarization, has been reported for blood vessels such as the porcine coronary and human mesenteric arteries (Klockner and Isenberg, 1991), rat cerebral arterioles (Yamazaki and Kitamura, 2001), rat pulmonary smooth muscle cells (Salter and Kozlowski, 1996) and rabbit choroidal arterioles (Curtis and Scholfield, 2001). Inhibition of K⫹ channels is another means of evoking depolarization, and such an effect by endothelin-1 has been reported for voltage-activated K⫹ (KV) channels in smooth muscle cells from rat renal arteries (Betts and Kozlowski, 2000), and pulmonary arteries from rats (Salter and Kozlowski, 1996; Li et al., 1999a) and humans (Shimoda et al., 2001), the latter via pathways involving protein kinase C.

2.2. Endothelium Endothelin-1, as well as being released by endothelial cells, also activates receptors on the endothelial cells to evoke depolarizing currents. These currents arise from the activation of poorly selective cation channels that are likely to be permeated by Ca2⫹ (Amano et al., 1994; Zhang, et al., 1994), and from the inhibition of inward rectifier potassium channels (KIR ) (Zhang, et al., 1994). These effects of endothelin-1 are not universal since it had no effects on the membrane potential of the endothelium isolated from the rat aorta (Marchenko and Sage, 1994b).

202

Harold A. Coleman et al.

3. PRODUCTS OF CYCLOOXYGENASE A variety of studies have reported on endothelium-derived contracting factors (EDCFs) that were not endothelin but likely to involve products of the cyclooxygenase pathway such as prostaglandin H2 and/or thromboxane A2 (Lüscher and Vanhoutte, 1986; Nishimura et al., 1995; Shirahase et al., 1995; Pomposiello et al., 1997; Sunano et al., 1999; Zhou et al., 1999). The involvement of membrane depolarization in the contractile effects was not determined. However, activation of thromboxane A2 (TP) receptors by agonists such as U46619 is used routinely in many studies to depolarize and contract blood vessels. The ionic mechanisms by which U46619 evokes depolarization of vascular smooth muscle are not clear, but may involve a decrease in K⫹ conductance (Scornik and Toro, 1992) and/or an increase in conductance to Ca2⫹-permeable cation channels (Tosun et al., 1998) (Figure 26.2).

4. PRODUCTS OF CYTOCHROME P450 Products of the cytochrome P450 pathway, particularly 20-hydroxyeicosatetraenoic acid (20-HETE) may be endothelium-dependent contracting factors (Escalante et al., 1993; Ma et al., 1993), with the production of 20-HETE being increased by the activation of endothelin receptors (Oyekan and McGiff, 1998). 20-HETE evokes depolarization of vascular smooth muscle (Ma et al., 1993) and this is likely to be due, at least in part, to the inhibition of large conductance Ca2⫹-activated K⫹ channels (Sun et al., 1998) (Figure 26.2).

5. REACTIVE OXYGEN SPECIES Reactive oxygen species can be generated in endothelial cells by various enzymatic pathways which include xanthine oxidase, NAD(P)H oxidase, nitric oxide synthase, cyclooxygenase, prostaglandin H2 synthase, and cytochrome P450. Superoxide anions generated by the endothelium evoked contraction of canine basilar arteries through unknown mechanisms and it was therefore suggested that superoxide anion is an EDCF (Katusic and Vanhoutte, 1989). Prostaglandin H2 synthase was thought to be involved in the process (Katusic and Vanhoutte, 1989). In the aorta from spontaneously hypertensive rats, acetylcholine evokes endothelium-dependent contractions that involve oxygen-derived free radicals, likely to be hydroxyl radicals and/or hydrogen peroxide. These reactive oxygen species probably activate phospholipase A2 which results in the activation of cyclooxygenase-1 and the production of prostaglandin H2, evoking contraction by activation of TP receptors (Yang et al., 2002) (Figure 26.2). Apart from stimulation of endothelial cells, superoxide anions have also been implicated in contractions induced by flow (Liu et al., 1998a). The involvement of ionic mechanisms in these contracting actions of the endothelium-dependent reactive oxygen species have not been determined. Endothelial cells depolarize in response to oxidant stress as a result of the activation of cation channels (Elliott and Koliwad, 1997; Balzer et al., 1999). The channels involved are thought to belong to the transient receptor potential (TRP) family (Balzer et al., 1999). Oxidant stress also inhibits KIR channels, thus contributing an additional depolarizing effect (Elliott and Koliwad, 1997; Bychkov et al., 1999) (Figure 26.2) and this may occur through its inhibition of the production of phosphatidylinositol bisphosphate (PIP2) (Karschin, 1999; Mesaeli et al., 2000; Ruppersberg, 2000).

Endothelium-dependent depolarization and its implications

203

6. ENDOTHELIAL STIMULATION AND ELECTRICAL COUPLING Stimulation of endothelial cells with acetylcholine or bradykinin results in biphasic changes in membrane potential during which the membrane initially hyperpolarizes and this phase is followed by depolarization. This biphasic response has been recorded from cultured endothelial cells of the guinea-pig coronary artery (Mehrke and Daut, 1990), bovine aorta (Vaca et al., 1996) and human umbilical vein (Groschner et al., 1994), from intact endothelium of the guinea-pig coronary artery (Chen and Cheung, 1992a) and rat aorta (Marchenko and Sage, 1994a) that were mechanically removed from the smooth muscle, from rabbit aortic valve endothelial cells that are separate from smooth muscle cells (Ohashi et al., 1999), and from guinea-pig lymphatic endothelial cells which are not electrically coupled to the smooth muscle (von der Weid and Van Helden, 1997). The hyperpolarization is attributed to Ca2⫹-activated K⫹ channels (Nilius and Droogmans, 2001); it is blocked by charybdotoxin with/without apamin (von der Weid and van Helden, 1997; Ohashi et al., 1999) which is consistent with the identity of the channels being of intermediate conductance (IKCa) with/without channels of small conductance (SKCa). The depolarization has been attributed to the activation of poorly selective cation channels (Nilius and Droogmans, 2001) (Figure 26.2). It is likely that one or more members of the TRP channel family are involved since at least TRPC1,3,4,5 are expressed in endothelial cells (Balzer et al., 1999; Kamouchi et al., 1999b). The agonist-induced depolarization may also be associated with a decrease in membrane conductance (Marchenko and Sage, 1993), suggested to be due to the inhibition of K⫹ channels (Mehrke and Daut, 1990). In view of the myoendothelial electrical coupling that occurs in some, but not all blood vessels, agonist-induced depolarization in the endothelial cells would be expected to spread via myoendothelial gap junctions to exert a depolarizing influence on the smooth muscle (Figure 26.2). In the femoral artery of the rat, in which myoendothelial electrical coupling, myoendothelial gap junctions, and EDHF do not occur (Sandow et al., 2002), endotheliumdependent depolarization of the smooth muscle also does not occur (Tare et al., unpublished observations). 7. IMPLICATIONS AND CONCLUSIONS Agents that are capable of releasing vasodilators from the endothelium also evoke endotheliumdependent depolarization of the smooth muscle. The nature of the endothelium-dependent depolarization has not been determined. There are a number of diffusible factors that are released from the endothelium that can depolarize the smooth muscle. Which one(s) if any, are involved physiologically or pathophysiologically, is far from clear. An alternative explanation is that the depolarization is initiated in the endothelial cells and spreads electrotonically to the smooth muscle, analogously to the charybdotoxin and apamin sensitive EDHF-mediated responses (Sandow et al., 2002). However, the lack of effective and specific blockers of myoendothelial gap junctions (Edwards and Weston, 2001; Busse et al., 2002; Tare et al., 2002) hampers rigorous testing of this possibility. The ionic mechanisms that underlie the endothelium-dependent depolarization have received little attention. Based on the effects of known endothelium-dependent factors, the most likely contender would be the activation of a poorly selective cation channel, while the additional or alternative involvement of Ca2⫹-activated Cl⫺ channels is a strong possibility.

204

Harold A. Coleman et al.

Irrespective of the nature of the depolarizing effects, the overall result of endothelial stimulation is a balance between the hyperpolarizing effects of the vasodilators, and these opposing depolarizing effects. Under normal conditions in healthy vessels the hyperpolarizing effects are dominant. However, if the hyperpolarizing currents are diminished, such as with K⫹ channel blockers, then the depolarizing effects can become prominent and have a range of effects. Endothelial depolarization can spread via myoendothelial gap junctions to depolarize the smooth muscle and this can initiate contraction of the vessel. Endothelial depolarization can also stimulate the formation of superoxide anions (Sohn et al., 2000), which in turn can inactivate NO, thereby contributing a contracting influence on blood vessels. Depolarization of the endothelium has several important implications for the actions of EDHF in health and disease. First, since the actions of EDHF are counteracted by depolarizing effects, EDHF is not as effective as it could be in evoking hyperpolarization and relaxation. Second, in view of the opposing depolarization, the actions of EDHF could be enhanced by blockade of the depolarizing effects. This could be of considerable value therapeutically. Third, some studies have shown that the effects of EDHF are diminished with age or in some disease states (hypertension, diabetes mellitus), and concluded that this was due to a diminished EDHF. An alternative (or additional) explanation could centre on an enhancement of the opposing endothelium-dependent depolarization. 8. ACKNOWLEDGEMENTS This work was supported by the National Health and Medical Research Council (Australia) and the National Heart Foundation (Australia).

27 Role of gap junctions in EDHF-mediated relaxation response in human subcutaneous resistance arteries P. Coats and C. Hillier

The aim of the present study was to determine a role for gap junctions in the EDHF-mediated relaxation response in human subcutaneous resistance arteries. Human subcutaneous arteries isolated from gluteal biopsies taken from volunteers with no history of hypertension or diabetes were cannulated and pressurized. The specific EDHF-mediated component of relaxation to acetylcholine (ACh) was identified and compared with the relaxation response to extracellular potassium in the presence or absence of the gap junction inhibitor 18␣-glycyrrhetinic acid. Acetylcholine relaxation was significantly reduced from 69 ⫾ 2% to 15 ⫾ 4% following incubation with 18␣-glycyrrhetinic acid (P ⬍ 0.05) whereas potassium-induced relaxation was insensitive to 18␣-glycyrrhetinic acid. These results demonstrate that the relaxation mediated by EDHF is dependent on gap junction communication in human subcutaneous resistance arteries. Moreover, if potassium were EDHF in these arteries the potassium-mediated response should have been sensitive to 18␣-glycyrrhetinic acid. These results indicate that the EDHF-mediated response is via gap junctions, and supports previously reported observations that EDHF and potassium act via different mechanisms and are thus unlikely to be similar entities in human subcutaneous resistance arteries.

1. INTRODUCTION The endothelium modulates local blood flow via the dynamic release of the vasoactive factors nitric oxide, prostacyclin and endothelium-derived hyperpolarizing factor (EDHF) (Moncada and Vane, 1979; Furchgott and Zawadzki, 1980; Félétou and Vanhoutte, 1988; Taylor and Weston, 1988). The few studies that have examined EDHF in human small resistance arteries suggest that an endothelium-dependent non-nitric oxide, non-prostanoid mechanism predominates in these important regulatory vessels (Coats et al., 2001). Based on experimental findings, the most likely EDHF candidates are the P450 metabolite of arachidonic acid metabolism, 11,12-epoxyeicosatrienoic acid and potassium ions (Hecker et al., 1994; Campbell et al., 1996; Edwards et al., 1998; Fisslthaler, 1999; Coats et al., 2001). However, despite focused and intensive research there remains, presently, no clear consensus on the identity of EDHF. The one consensual observation that has emerged from EDHF studies is that the EDHF-mediated response involves calcium-sensitive potassium channels that can be blocked by the actions of apamin plus charybdotoxin (Zygmunt and Hogestatt, 1996; Edwards and Weston, 1998). Additionally, EDHF-mediated relaxation is probably due to electrical coupling with hyperpolarization spreading from the endothelium to the underlying vascular smooth muscle via myo-endothelial gap junctions (Chaytor et al., 1998; Yamamoto et al., 1998; Edwards et al., 1999; Chaytor et al., 2001; Ungvari et al., 2002). Whether EDHF is a diffusible factor such as 11,12-epoxyeicosatrienoic or endotheliumderived potassium that initiates a wave of hyperpolarization to smooth muscle cells a role for

206

P. Coats and C. Hillier

gap junction communication is plausible (diffusion or conduction). In view of this, the aim of the present study was to determine a role for gap junctions in the EDHF-mediated response in human subcutaneous resistance arteries. 2. METHODS

2.1. Patients and vessel preparation This study has been performed in accordance with the Declaration of Helsinki (Declaration of Helsinki, 1999) and with the approval of the local Hospital Ethics Committee. Each subject gave informed written consent. Clinical characteristics of the patients are given in Table 27.1.

2.2. Biopsy procedure Healthy volunteers with no history of vascular disease, diabetes, hypertension or renal impairment attended the Clinical Investigations Research Unit at the Western Infirmary, Glasgow, UK. Subcutaneous gluteal fat biopsies (1.5 ⫻ 1.5 ⫻ 0.5 cm) were excised under local anaesthesia with 1% lidocaine and immediately transferred to cold physiological saline solution. Twenty-six subcutaneous resistance-size arteries (lumen diameter 117 ⫾ 4 ␮m) were isolated from fourteen biopsies and cleaned of adherent tissue under a dissection microscope (Zeiss Stemi 2000, Mag X6–X45).

2.3. Apparatus and experimental procedure Isolated arteries were mounted on a pressure myograph (Danish MyoTech P110 Aarhus, DK) and studied at an intraluminal pressure of 40 mmHg in a no-flow state (Coats and Hillier, 1999). The pressurized artery was imaged via a digital video camera. The video image was analysed using MyoView software (Danish MyoTech, Aarhus, DK). MyoView permitted the measurement of multiple parameters; edge detection of the viewed image provided measurement of external diameter, internal diameter, wall thickness and lumen diameter. Additionally, time, right and left in-flow pressure, mean pressure, longitudinal force, temperature and manual intervention were recorded. Data was acquired at time intervals of one-second and recorded on a personal computer. All arteries were studied using physiological saline solution gassed with 95% O2/5% CO2 maintaining pH 7.4 at 37 ⬚C. Functional viability was assessed by maximum constriction to 60 mM potassium solution and norepinephrine (10⫺5 M) and vasorelaxation (⬎ 80%) to ACh. Table 27.1 Basic clinical data on the fourteen volunteers who provided subcutaneous biopsies Age Creatinine Urea Cholesterol Glucose Systolic BP Diastolic BP Note BP Blood pressure.

64 ⫾ 1 years 92 ⫾ 3.1 ␮mol/L 5.4 ⫾ 0.3 mmol/L 5.2 ⫾ 0.2 mmol/L 5.4 ⫾ 0.2 mmol/L 143 ⫾ 7 mmHg 78 ⫾ 5 mmHg

Role of gap junctions in EDHF-mediated relaxation response

207

2.4. Pharmacological protocols Acetylcholine: the arteries were constricted with norepinephrine to approximately 75% of the maximal response. In all cases endothelium-dependent relaxations to acetylcholine were measured in the continual presence of 10⫺3 M NG-nitro L-arginine (L-NOARG) and 3⫻10⫺5 M indomethacin. Cumulative concentration response curves were measured following intraluminal incubation with the potassium channel inhibitors apamin (10⫺7 M), charybdotoxin (ChTX) (10⫺7 M), apamin⫹ChTX and apamin⫹iberiotoxin (10⫺7 M). Responses were also measured following incubation with the gap junction inhibitor 18␣-glycyrrhetinic acid (10⫺4 M). Potassium-dependent relaxation: In normal physiological saline, arteries were constricted to approximately 50% of the maximal response. Then extracellular potassium concentration was elevated from 4.6–20 mM in 2 mM steps.

2.5. Drugs and solutions Acetylcholine: L-NOARG, and norepinephrine were purchased from Sigma (Poole, Dorset, UK); apamin from Calbiochem (Nottingham, UK); ChTX from Bachem (Saffron Walden, UK). The concentration of physiological saline solution was (in mmol) – NaCl 119, KCl 4.5, NaHCO3 25, KH2PO4 1.0, MgSO47H2O 1.0, glucose 11.0 and CaCl2 2.5. High potassium solutions (4.6–20mM) were generated by equimolar substitution of NaCl with KCl.

2.6. Data and statistical analysis Relaxations are represented as percent relative to the constricted diameter of the artery. Values are presented as mean ⫾ standard error of the mean. Comparison of the responses was by one-way ANOVA for repeated measures. Statistical significance was assumed if P were less than 0.05.

3. RESULTS

3.1. EDHF-dependent relaxations In the presence of L-NOARG and indomethacin cumulative addition of ACh resulted in a maximum relaxation of 68 ⫾ 3% (Figure 27.1, n ⫽ 6,). This response was insensitive to apamin or ChTX alone (Figure 27.1, n ⫽ 3). However, incubation with the combination of apamin and ChTX abolished the relaxation to ACh (Figure 27.1, n ⫽ 6). Substitution of ChTX with iberiotoxin failed to inhibit the relaxation (data not shown).

3.2. Gap junctions Following incubation with the gap junction inhibitor 18␣-glycyrrhetinic acid the maximal relaxation to ACh was reduced to from 69 ⫾ 2% to 15 ⫾ 4% (Figure 27.2, n ⫽ 6, P ⬍ 0.05). Incubation with 18␣-glycyrrhetinic acid had no effect on basal vascular tone or the sensitivity to norepinephrine (data not shown).

3.3. Potassium Elevation of extracellular potassium ion concentration induced a maximal relaxation of 20 ⫾ 7% (Figure 27.3, n ⫽ 6). This potassium-dependent relaxation was unaffected by the gap junction inhibitor 18␣-glycyrrhetinic acid (Figure 27.3, n ⫽ 6).

0 10 20

Relaxation (%)

30 40 50 60 70 ACh, L-NOARG + INDO apamin

80

ChTX apamin + ChTX

90 100 –10

–9

–8

–7

–6

–5

ACh (M)

Figure 27.1 Relaxation to ACh, in the presence of L-NOARG ⫹ indomethacin before and following luminal incubation with either apamin or ChTX and finally apamin plus ChTX. Data shows mean ⫾ SEM (n ⫽ 6).

0 10 20

*

Relaxation (%)

30 40 50 60 70 80 90 100 –10

ACh, L-NOARG + INDO 18 a GA

–9

–8

–7

–6

–5

ACh (M)

Figure 27.2 Relaxation to ACh in the presence of L-NOARG⫹indomethacin following incubation with 18␣-glycyrrhetinic acid (18␣GA). Data are mean ⫾ SEM (n ⫽ 6). ANOVA for repeated measures, comparison of curves. The asterisk indicates a P ⬍ 0.05.

Role of gap junctions in EDHF-mediated relaxation response

209

30 25 20 15 Lumen diameter (uM)

10 5 0 –5 –10 –15 K+ 18 a GA

–20 –25 –30 4.6

6

8

10 12 14 [K+]o (mM)

16

18

20

Figure 27.3 The effect of increasing the extracellular potassium concentration ([K⫹]o) from 4.6 to 20mM before and after incubation with 18␣-glycyrrhetinic acid (18 ␣ GA, n ⫽ 6, ------- indicates baseline diameter).

4. DISCUSSION This study demonstrates that the EDHF-mediated response in human subcutaneous resistance arteries uses gap junctions. Furthermore, it supports previously reported observations that EDHF and potassium are unlikely to be similar entities in human subcutaneous resistance arteries (Coats et al., 2001). There is growing evidence showing the importance of gap junction communication between the endothelium and smooth muscle. With respect to EDHF, potassium-dependent endothelial cell hyperpolarization may be transmitted via gap junctions or EDHF, as an endothelium-derived factor, may diffuse via gap junctions to the underlying vascular smooth muscle (Chaytor et al., 1998; Yamamoto et al., 1998; Edwards et al., 1999; Chaytor et al., 2001; Coleman et al., 2001; Ungvari et al., 2002). Results obtained with a number of structurally unrelated inhibitors of gap junctions support this hypothesis. In the present study, 18␣-glycyrrhetinic acid, which is reported to be one of the more specific inhibitors of gap junction communication (Chaytor et al., 2000), had profound effects on the non-nitric oxide, non-prostanoid, apamin-ChTX sensitive endothelium-dependent relaxation attributable to EDHF. Presently there are concerns over the specificity of commonly used inhibitors of gap junctions including 18␣-glycyrrhetinic acid (Davidson and Baumgarten, 1988; Chaytor et al., 2000). However, in this study 18␣-glycyrrhetinic acid had no effect on basal or agonist-induced tone, nor did it affect endothelium-independent mechanisms of relaxations (data not shown).

210

P. Coats and C. Hillier

Elevation of the extracellular potassium concentration dilates isolated arteries from a number of vascular beds (Knot et al., 1996; Dong et al., 1998; Edwards et al., 1998; Coats et al., 2001). In this study, elevation of the extracellular potassium concentration resulted in a small relaxation of constricted arteries. The maximal relaxation occurred at around 14–16mM potassium, increasing extracellular potassium beyond 16 mM resulted in constriction. The potassium-dependent relaxation observed in these arteries has been previously reported (Coats et al., 2001) to be sensitive to the combination of Ba2⫹ and ouabain but not Ba2⫹ or ouabain alone and is in agreement with part of the work published by Edwards and colleagues (1998). In the present study, 18␣-glycyrrhetinic acid had excellent inhibitory effects on the EDHF-mediated relaxation to ACh. Since, however, the inhibition of gap junctions had no effect on the potassium-mediated relaxation, this observation provides further evidence that the EDHF-dependent response is mediated via different mechanisms than those through which potassium operates and is thus unlikely to be EDHF in human subcutaneous resistance arteries.

28 Permissive role of cAMP in the mediation of relaxations initiated by endothelial hyperpolarization Tudor M. Griffith, Andrew T. Chaytor and David H. Edwards

An endothelium-derived hyperpolarizing factor (EDHF) is widely hypothesized to underpin vascular relaxations that are independent of nitric oxide (NO) and prostanoids. Although a diverse variety of agents, ranging from ions (specifically K⫹) to products of arachidonic acid metabolism, have been proposed as the active mediator, bioassay techniques have failed to provide unequivocal evidence for the existence of a freely diffusible EDHF. The present chapter summarizes evidence that the EDHF-type response to acetylcholine is mediated by electrotonic conduction of endothelial hyperpolarization via myoendothelial and homocellular smooth muscle gap junctions whose conductance and permeability are regulated by an associated synthesis of cAMP. In rabbit iliac arteries relaxation is accompanied by transient elevations in cAMP levels of the smooth muscle and mechanical responses, nucleotide accumulation and subintimal hyperpolarization are abolished by the connexin-mimetic peptide Gap 27 and by endothelial denudation. Subintimal hyperpolarization is also abolished by inhibition of adenylyl cyclase, whereas inhibition of cAMP phosphodiesterase enhances transmission of endothelial hyperpolarization through the media and potentiates relaxation. Parallel studies confirm that cAMP increases dye transfer from the endothelium into the media whereas Gap 27 promotes sequestration of dye within the intima. In perfused rabbit ear preparations, phenylephrine stimulates an endothelium-dependent efflux of cAMP that is abolished by blockade of gap junctions and therefore mediated by a signal transmitted from activated smooth muscle to overlying endothelial cells directly. However, cAMP formed via this pathway does not influence the constrictor response of the smooth muscle to phenylephrine, indicating that endothelium-derived cAMP does not affect tone in the absence of co-existent endothelial hyperpolarization. Taken together, the findings suggest that the contribution of cAMP to the EDHF phenomenon is to enhance passive electrotonic spread of endothelial hyperpolarization. Its action may therefore be regarded as permissive.

Signalling between adjacent cells may involve direct communication via gap junction channels that are constructed from connexin protein subunits surrounding an aqueous central pore (Perkins et al., 1998). These channels allow the transfer of polar molecules ⬍1000Da in size and provide electrical continuity between coupled cells with communication occurring predominantly at focal sites in the plasma membrane where gap junctions cluster in plaques of up to several hundred individual units (Brink, 1998; Bukauskas et al., 2000). In the blood vessel wall, the distinctive punctate appearance of such plaques, abundant in the endothelial monolayer but relatively sparse in the media, is readily demonstrated by immunostaining with specific antibodies (Chaytor et al., 2001; Berman et al., 2002). By contrast, electron microscopy is required to visualize myoendothelial plaques which are much smaller and less numerous than plaques coupling endothelial cells (Sandow and Hill, 2000).

212

Tudor M. Griffith et al.

In arterioles, myoendothelial gap junctions appear to behave as simple ohmic resistors without rectification, so that action potentials originating in smooth muscle cells are readily detectable in the endothelium and in the converse sense, endothelial hyperpolarizations induced by acetylcholine or direct injection of electrical current may be passively conducted to underlying smooth muscle cells (Segal and Beny, 1992; Beny and Pacicca, 1994; Emerson and Segal, 2000b; Coleman et al., 2001a; Yamamoto et al., 2001). This mechanism provides an alternative to the hypothesis that EDHF-type responses evoked by agonists such as acetylcholine involve the release of a freely diffusible factor into the extracellular space (Popp et al., 1996; Campbell and Harder, 1999; Edwards and Weston, 2001). Indeed, bioassay experiments with sandwich preparations of rabbit conduit arteries, constructed from closely apposed strips of endothelium-intact and -denuded vessel fail to demonstrate the release of a freely diffusible EDHF following administration of acetylcholine, whereas agents that interrupt communication via gap junctions inhibit EDHF-type relaxations in arterial and venous ring preparations from a variety of species (Chaytor et al., 1998, 2001, 2002; Taylor et al., 1998; Griffith and Taylor, 1999). Because the large bulk of the media will readily dissipate electrical signals in thick-walled arteries, it has been argued that the endothelium cannot act as a significant source of hyperpolarizing current without the involvement of an additional regenerative mechanism (Beny, 1999). However, since there is evidence that elevations in cAMP levels may enhance gap junctional communication (Burghardt et al., 1995; Chanson et al., 1996; Abudara et al., 2000; Gladwell and Jefferys, 2001; Grazul-Bilska et al., 2001), the hypothesis was tested that this nucleotide underpins EDHF-type relaxations in conduit arteries by enhancing the passive electrotonic spread of hyperpolarizing current from the endothelium into and through the media (Griffith et al., 2002). The role of gap junctional communication was investigated with an inhibitory peptide homologous to the Gap 27 domain of the second extracellular loop of connexins 37 and 43 and with 18␣-glycyrrhetinic acid, a lipophilic aglycone that disrupts gap junction plaques (Chaytor et al., 1998; Taylor et al., 1998). EDHF-type relaxations were correlated with measurements of cAMP levels in smooth muscle and membrane potential, and the effects of cAMP on the permeability of gap junctions was assessed by loading the entire endothelial monolayer of perfused arterial segments with a cell-permeant tracer dye. The mechanisms that led to cAMP synthesis by the endothelium were also investigated by assaying cAMP release into the effluent from isolated bufferperfused ear preparations following activation of smooth muscle cells by phenylephrine and endothelial cells by acetylcholine. Phenylephrine does not stimulate endothelial cells directly but can elevate endothelial Ca2⫹ levels indirectly through a mechanism that involves transmission of signals from activated smooth muscle via myoendothelial gap junctions (Dora et al., 1997). 1. MATERIALS AND METHODS All experiments were performed with tissue harvested from male New Zealand White rabbits (2–2.5 kg, euthanized with sodium pentobarbitone 120 mg/kg; i.v.) employing oxygenated (95% O2, 5% CO2, pH 7.4) Holmans buffer of composition (mM): 120 NaCl, 5 KCl, 2.5 CaCl2, 1.3 NaH2PO4, 25 NaHCO3, 11 glucose and 10 sucrose. Nitric oxide (NO) synthase and cyclooxygenase were inhibited pharmacologically by including NG-nitro-L-arginine methyl ester (L-NAME, 3 ⫻ 10⫺4 M) and indomethacin (1 ⫻ 10⫺5 M) in the buffer.

cAMP and electrotonic hyperpolarization

213

1.1. Isolated ring preparations 1.1.1. Mechanical responses Iliac artery rings 2–3 mm wide were suspended in organ chambers with resting tension set at ~0.25 g. During an initial equilibrium period of one hour the tissues were repeatedly washed with fresh buffer and tension readjusted after stress relaxation. Rings were constricted with phenylephrine (1 ⫻ 10⫺6 M) and cumulative concentration–relaxation curves to acetylcholine obtained. Some preparations were also incubated with Gap 27 (6 ⫻ 10⫺4 M, amino acid sequence SRPTEKTIFII; Sigma Genosys, UK) or with the cAMP phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (IBMX, 2 ⫻ 10⫺5 M) for 40 min prior to addition of acetylcholine and phenylephrine. As IBMX depressed contraction, the concentration of phenylephrine used in experiments with this agent was increased from 1⫻ 10⫺6 M to 3⫻10⫺6 M. In separate experiments, rings constricted by phenylephrine (1⫻10⫺6 M) were partially relaxed by 8-bromo-cAMP (1⫻10⫺3 M) and tension restored to control level with phenylephrine (3 ⫻ 10⫺6 M) before concentration–relaxation curves for acetylcholine were obtained. The endothelium-dependent nature of the response to acetylcholine was confirmed in rings from which the endothelium had been removed by gentle abrasion. 1.1.2. Cyclic nucleotide levels Unmounted rings of iliac artery were incubated at 37 ⬚C in the presence or absence of Gap 27 or IBMX as described before. Subsequently, phenylephrine (1 ⫻ 10⫺6 M) was introduced into the buffer followed by acetylcholine (3 ⫻ 10⫺6 M) after 3 min. The rings were then frozen in liquid N2 at time points up to 180 s and stored at ⫺70 ⬚C before their cAMP and cGMP content was determined by radioimmunoassay (Amersham Biosciences, UK). Nucleotide levels were expressed relative to protein content determined by a dye-binding assay (Bio-Rad, UK) and reflect smooth muscle levels as the mass of the endothelium is negligible compared to the media (Taylor et al., 2001; Chaytor et al., 2002).

1.2. Membrane potential studies Iliac artery strips were held adventitia down in an oxygenated organ chamber by a Harp slice grid (ALA Scientific Instruments, USA) and superfused at 37 ⬚C. The membrane potential of subintimal smooth muscle cells was recorded with glass capillary microelectrodes filled with 3 M KCl (tip resistance 60–110 M⍀ advanced into the media with a PCS-5000 micromanipulator (Burleigh Instruments, UK). Successful impalements were characterized by a sudden negative drop in potential from baseline and a stable signal for at least 2 min.

1.3. Dye transfer studies Femoral arteries were perfused at 35 ⬚C at a pressure of 25 mmHg and a flow rate of 0.1 ml/min in a Living Systems myograph (LSI, USA) in the absence or presence of Gap 27 (6 ⫻ 10⫺4 M), IBMX (2 ⫻ 10⫺5 M), 8-bromo-cAMP (1 ⫻ 10⫺3 M) or 8-bromo-cGMP (1 ⫻ 10⫺3 M) for 30 min. The preparations were then perfused for 30 min with calcein AM (1 ⫻ 10⫺5 M) (Molecular Probes, Europe) at room temperature, before washout with dye free buffer at 35 ⬚C for 30 min. This procedure loads the endothelium so that after intracellular

214

Tudor M. Griffith et al.

endothelial cleavage of the acetoxymethyl moiety responsible for uptake, a reduction in molecular weight (from 1000 to 623 Da) allows diffusion of calcein into the media via gap junctions. In control experiments, arteries were perfused with calcein (1 ⫻ 10⫺5 M), which is membrane impermeable. Following fixation, transverse rings (10 ␮m thick) were prepared and imaged with a TCS 4D confocal laser scanning system (Leica, UK) with the filters set for 490 nm excitation and 525 nm emission. A pixel intensity profile across the vessel wall was fitted to a monoexponential to derive a space constant describing the decay of medial fluorescence as a function of distance from the intima (i.e.) the distance over which fluorescence decremented by 1/e or ~ 63%.

1.4. Isolated perfused vascular bed studies Ear preparations were cannulated via the central artery and perfused for 30 min at 2 ml/min at 35 ⬚C. Acetylcholine was administered 20 min after subsequent administration of phenylephrine (1 ⫻ 10⫺6 M) when perfusion pressure had risen to a stable plateau. The role of direct communication between endothelial and smooth muscle cells was investigated with 18␣glycyrrhetinic acid (18␣-GA, 1⫻10⫺4 M) that was included in the perfusate for 30min before administration of phenylephrine. The contribution of endothelial and smooth muscle cAMP synthesis was assessed by endothelial denudation, achieved by including 0.1% Triton X-100 in the buffer for 30 seconds, and evaluated by loss of functional response to acetylcholine. Cyclic AMP concentrations in the effluent were determined by collecting aliquots over 5s intervals and stored at ⫺70⬚C before radioimmunoassay (Amersham Biosciences, UK).

1.5. Statistical analysis Results are given as mean ⫾ SEM, where n denotes the number of animals studied for each data point, and were compared by the Student’s t-test for paired or unpaired data as appropriate. Concentration-response curves and nucleotide accumulation were assessed by ANOVA followed by the Bonferroni multiple comparisons test. P less than 0.05 was considered as statistically significant. 2. RESULTS

2.1. Ring preparations Maximal EDHF-type reductions in phenylephrine-induced tone were observed with acetylcholine at a concentration of 3 ⫻ 10⫺6 M, and were respectively amplified from 32.2 ⫾ 4.1% (n ⫽ 14) to 65.2 ⫾ 6.5% and 64.0 ⫾ 9.2% by IBMX (2 ⫻ 10⫺5 M) or 8-bromo-cAMP (1 ⫻ 10⫺3 M) (P ⬍ 0.05 in each case, n ⫽ 6 and 4; Figure 28.1). Corresponding EC50 values fell from 561 ⫾ 100 nM to 99.5 ⫾ 30.9 nM and 79 ⫾ 10 nM (P ⬍ 0.05 in each case). Notably, relaxations to acetylcholine were generally transient, but became sustained in the presence of IBMX or 8-bromo-cAMP (Figure 28.1). Incubation with Gap 27 (6 ⫻ 10⫺4 M) or removal of the endothelium abolished the relaxations (n ⫽ 4 and 5; Figure 28.1).

2.2. Nucleotide accumulation Basal cAMP levels were ~4 pmol/mg protein in both endothelium-intact and -denuded rings (n ⫽ 5 and 10, respectively) and not significantly altered by incubation with L-NAME

A

B

Figure 28.1 (A) Representative traces showing EDHF-type responses to acetylcholine in iliac artery rings constricted by phenylephrine. Transient relaxations became sustained in the presence of IBMX (2 ⫻ 10⫺5 M) or 8-bromo-cAMP (1 ⫻ 10⫺3 M) but were abolished by the connexin-mimetic peptide Gap 27 (6 ⫻ 10⫺4 M). (B) Concentration–relaxation curves confirming potentiation by IBMX and 8-bromo-cAMP and loss of response following endothelial denudation or incubation with Gap 27. Initial constrictor tone was matched in all experimental groups by increasing the concentration of phenylephrine from 1 ⫻ 10⫺6 M to 3 ⫻ 10⫺6 M in experiments with IBMX and 8-bromo-cAMP. Initial tension was unaffected by Gap 27 or endothelial denudation.

216

Tudor M. Griffith et al.

14

[cAMP] pmol/mg protein

12

Control IBMX +E IBMX –E Gap27 Denuded

10

*

8

*

6 4 2 0 Control

0

15

30

60

120

180

Time (sec) after addition of ACh

Figure 28.2 Histograms showing that acetylcholine (3 ⫻ 10⫺6 M) induced a transient peak in smooth muscle cAMP levels in rings with intact endothelium that was abolished by Gap 27 (6 ⫻ 10⫺4 M) or removal of the endothelium. IBMX (2 ⫻ 10⫺5 M) caused equivalent sustained increases in cAMP levels in preparations both with and without endothelium that were unaffected by acetylcholine. The asterisk denotes P less than 0.05 compared with time 0.

(3⫻10⫺4 M) and indomethacin (1⫻10⫺5 M) and administration of phenylephrine (1⫻10⫺6 M). In preparations with endothelium, acetylcholine (3 ⫻ 10⫺6 M) induced a transient rise in cAMP levels that peaked at ~7.5 pmol/mg protein after 15–30 s before falling towards baseline (n ⫽ 5, Figure 28.2). This nucleotide response was abolished by endothelial removal and by incubation with Gap 27 (6 ⫻ 10⫺4 M) (n ⫽ 10 and 8, respectively; Figure 28.2). Cyclic AMP levels were significantly increased to ~10 pmol/mg protein in both endothelium-intact and -denuded rings incubated with IBMX (2 ⫻ 10⫺5 M), but were not subsequently altered by the administration of acetylcholine (n ⫽ 7 and 4, respectively; Figure 28.2). By contrast, cGMP levels were not significantly altered by acetylcholine either in the presence or absence of IBMX (data not shown). This confirms that IBMX, which also inhibits the hydrolysis of cGMP (Lugnier and Komas, 1993), did not potentiate the relaxant response to acetylcholine by amplifying the biochemical consequences of residual NO activity (Cohen et al., 1997).

2.3. Smooth muscle membrane potential The resting membrane potential of subintimal smooth muscle cells in arteries with intact endothelium was ⫺45.3 ⫾ 3.8 mV compared to ⫺46.7 ⫾ 3.0 mV in the presence of Gap 27 (6 ⫻ 10⫺4 M, n ⫽ 3). However, this peptide abolished sustained hyperpolarizations evoked by acetylcholine (1⫻10⫺6 M) and reduced maximal membrane potential changes from ⫺19.4 ⫾ 2.3 to ⫺5.0 ⫾ 1.0 mV (P ⬍ 0.01, n ⫽ 3, Figure 28.3). 2⬘,5⬘-DDA (2 ⫻ 10⫺4 M) also abolished sustained responses to acetylcholine (1 ⫻ 10⫺6 M) with maximal hyperpolarizations being reduced from ⫺27.0 ⫾ 1.5 to ⫺5.7 ⫾ 0.4 mV (P ⬍ 0.005, n ⫽ 3, Figure 28.3). Under resting conditions 2⬘,5⬘-DDA also caused a small depolarization from ⫺41.7⫾3.2 to ⫺36.0⫾1.7mV

cAMP and electrotonic hyperpolarization Control Gap27 (600 µm) 2⬘, 5⬘-DDA (200 µm) IBMX (20 µm)

Membrane potential (mV)

–30 –40

[ACh]

217

ACh (1 µm) Gap27 (600 µm) 2⬘, 5⬘-DDA (200 µm) IBMX (20 µm)

1 µm 1 µm

1 µm

–50 –60 –70 2 min –80

Figure 28.3 Subintimal smooth muscle membrane potential changes evoked by acetylcholine (1 ⫻ 10⫺6 M) in strips of iliac artery with intact endothelium before and after incubation with Gap 27 (6 ⫻ 10⫺4 M), 2⬘,5⬘-DDA (2 ⫻ 10⫺4 M) or IBMX (2 ⫻ 10⫺5 M) for 40 min. Representative traces show that rapid initial hyperpolarizations, sustained 10–15 mV below resting level, were effectively abolished by Gap 27 and 2⬘,5⬘-DDA but unaffected by IBMX. 2⬘,5⬘-DDA caused a small depolarization and IBMX a small hyperpolarization, whereas Gap 27 did not affect resting membrane potential. Histograms show results pooled from 3–9 such experiments. The double asterisk denotes P less than 0.01 and the triple asterisk denotes P less than 0.001 compared with control.

(n ⫽ 3). Although IBMX (2 ⫻ 10⫺5 M) slightly hyperpolarized resting membrane potential from ⫺40.6 ⫾ 2.2 to ⫺44.9 ⫾ 2.1 mV (n ⫽ 9), maximal hyperpolarizations to acetylcholine (1 ⫻ 10⫺6 M) were unaffected at ⫺21.6 ⫾ 2.1 and ⫺20.9 ⫾ 2.6 mV in the absence and presence of this phosphodiesterase inhibitor, respectively (n ⫽ 9, Figure 28.3). In preparations from which the endothelium had been partially removed, maximal hyperpolarizations to acetylcholine (1⫻10⫺6 M) were ⫺22.5⫾0.7mV in the absence and ⫺24.8⫾ 1.1 mV in the presence of IBMX (2 ⫻ 10⫺5 M) when subintimal smooth muscle cells were impaled within a region of the vessel still possessing endothelium (Figure 28.4). When impalement was made 1.5 mm within the denuded region remote hyperpolarizations to acetylcholine (1 ⫻ 10⫺6 M) were reduced to a maximum of ⫺5.5 ⫾ 0.6 mV which was increased to ⫺13.4 ⫾ 2.7 mV by IBMX (2 ⫻ 10⫺5 M) (P ⬍ 0.05, n ⫽ 3, Figure 28.4). No hyperpolarizing response to acetylcholine could be detected in strips that had been completely denuded of endothelium, either in the presence or absence of IBMX.

2.4. Dye transfer Dye was detectable both within the intima and the media of arteries perfused with calcein AM with the fall-off in medial fluorescence being described by a space constant of 8.54 ⫾ 0.39 ␮m (Figure 28.5, n ⫽ 3). This parameter was increased respectively to 11.50 ⫾ 0.80 and 12.50 ⫾ 0.95 ␮m by IBMX (2 ⫻ 10⫺5 M) or 8-bromo-cAMP (1 ⫻ 10⫺3 M) (n ⫽ 6 and 5, P ⬍ 0.05), but was unchanged by 8-bromo-cGMP (1 ⫻ 10⫺3 M) (8.39 ⫾ 0.44 ␮m, n ⫽ 4). By contrast, dye was almost exclusively localized within the endothelium in vessels perfused with Gap 27 (6 ⫻ 10⫺4 M). The essential requirement for endothelial uptake was confirmed by perfusion with calcein when no fluorescence could be detected within either endothelial or smooth muscle cells.

1 µm

1 µm

IBMX (20 µm) 1 µm

–40 –50 –60 –70 –80

2 min E+ upstream

E– downstream

E– complete

Membrane potential (mV)

Membrane potential (mV)

[ACh] –30

ACh (1 µm)

Control IBMX (20 µm) –30 –40 –50 –60 –70 –80

E+ E– E– upstream downstream complete

Figure 28.4 Representative traces from iliac artery strips from which the endothelium had been either partially (50%) or completely removed by abrasion. Representative traces show that IBMX did not affect hyperpolarizations evoked by acetylcholine (1 ⫻ 10⫺6 M) just within (~0.2 mm) regions where the anatomically upstream endothelium was still intact (E⫹ upstream), whereas small conducted hyperpolarizations detected 1.5 mm from the edge of the residual endothelium were amplified by incubation with IBMX for 40 min (E- downstream). In strips completely denuded of endothelium there was no electrical response to acetylcholine either in the presence or absence of IBMX (E- complete). Histograms show maximal potential changes pooled from 3 such experiments in each case. The asterisk denotes P less than 0.05 compared with control.

Figure 28.5 Dye transfer in isolated femoral arteries. Following intraluminal perfusion with the cell permeant calcein AM fluorescence was visible in subjacent smooth muscle cells. Dye transfer was prevented by Gap 27 (6 ⫻ 10⫺4 M) and only autofluorescence of the internal elastic lamina was evident following perfusion with the cell impermeant calcein. Medial fluorescence was enhanced by IBMX (2 ⫻ 10⫺5 M) and 8-bromo-cAMP (1 ⫻ 10⫺3 M) but unaffected by 8-bromo-cGMP (1 ⫻ 10⫺3 M). All images are shown at the same magnification (see Color Plate 10).

cAMP and electrotonic hyperpolarization

219

2.5. Perfused ear experiments In isolated perfused rabbit ear preparations phenylephrine (1 ⫻ 10⫺6 M) induced a rise in perfusion pressure of approximately 200 mmHg which reached a plateau after 15–20 min that was reduced by approximately 50% over 3 min following subsequent administration of acetylcholine (3 ⫻ 10⫺6 M, n ⫽ 4, Figure 28.6). The time course and magnitude of the pressor response to phenylephrine were unaffected by endothelial denudation (n ⫽ 5) or inhibition of gap junctional communication with 18␣-GA (n ⫽ 4), whereas the subsequent depressor effects of acetylcholine were abolished by both interventions. A

B

Figure 28.6 Effects of phenylephrine and acetylcholine on perfusion pressure and endothelial cAMP efflux in isolated perfused rabbit ears. (A) Left: The time course and magnitude of the pressor response to phenylephrine (1 ⫻ 10⫺6 M) were unaffected by the gap junction inhibitor 18␣-GA (1⫻10⫺4 M) or endothelial denudation. Right: The subsequent depressor effects of acetylcholine (3⫻10⫺6 M) were abolished by 18␣-GA and endothelial denudation. (B) Left: Under control conditions phenylephrine (1 ⫻ 10⫺6 M) induced a sustained 2-fold increase in cAMP release into the effluent. Following endothelial denudation increments in cAMP efflux declined to baseline before maximal constriction was attained. An equivalent transient cAMP efflux was observed in the presence of 18␣-GA. Right: In endothelium-intact preparations constricted by phenylephrine, acetylcholine evoked comparable increments in cAMP efflux in the presence and absence of 18␣-GA. No cAMP response to acetylcholine was evident in endothelium-denuded preparations.

220

Tudor M. Griffith et al.

Phenylephrine evoked a sustained 2-fold increase in cAMP release into the effluent from preparations with intact endothelium that increased from 2008 ⫾ 100 fmol/min to 4240 ⫾ 520 fmol/min (n ⫽ 3, Figure 28.6). Following endothelial denudation with Triton-X, this response was no longer sustained and a transient increment in cAMP efflux declined to baseline before the maximal phenylephrine-induced rise in perfusion pressure was attained (n ⫽ 4). An equivalent transient nucleotide response was observed in endothelium-intact preparations perfused with 18␣-GA (1 ⫻ 10⫺4 M, n ⫽ 3). In preparations with endothelium constricted by phenylephrine (1 ⫻ 10⫺6 M), acetylcholine (3 ⫻ 10⫺6 M) evoked an additional endothelium-dependent increment in cAMP efflux that was of approximately equal magnitude in the presence and absence of 18␣-GA (n ⫽ 4 in each case), thus indicating that the nucleotide response to this agonist originated from the endothelium and that 18␣-GA does not impair the activation of endothelial adenylyl cyclase (Figure 28.6). 3. DISCUSSION The salient finding of the present study is that the signalling pathways that underlie the EDHF phenomenon in the rabbit vasculature involve electrotonic spread of hyperpolarization from the endothelium into the media via myoendothelial and homocellular smooth muscle gap junctions. The permeability and electrical conductance of these direct communication channels appear to be regulated by co-existent cAMP synthesis, though the mechanisms that lead to production of this nucleotide by endothelial cells and its accumulation in smooth muscle remain to be delineated in detail. Studies with iliac artery rings constricted by phenylephrine demonstrated that EDHF-type relaxations to acetylcholine were often transient, with an initial decrease in tone over seconds decaying within 1–2 min. Relaxation was associated with an endothelium-dependent elevation in smooth muscle cyclic AMP levels ~1.5 fold above baseline, which peaked after approximately 15–30 s before declining to control. The close relationship between relaxation and cAMP accumulation was emphasized by observations that both were abolished by endothelial denudation or incubation with the connexin-mimetic peptide Gap 27 which interrupts gap junctional communication. Evidence that the role of cAMP was permissive was provided by observations that acetylcholine-induced decreases in tone were prolonged and increased in magnitude by 8-bromo-cAMP, a cell permeant analogue of cAMP, and by the phophodiesterase inhibitor IBMX, which caused static endothelium-independent elevations in smooth muscle cAMP content ~2-fold above control. Measurements of membrane potential in the smooth muscle from arterial strips impaled via their intimal surface confirmed the role of direct endothelial-smooth muscle signalling in the EDHF phenomenon as subintimal hyperpolarizations evoked by acetylcholine in strips with intact endothelium were abolished by endothelial denudation and by Gap 27. Experiments with arterial strips partially denuded of endothelium provided direct evidence that the brief elevations in smooth muscle cAMP levels induced by acetylcholine may be unable to support sustained transmission of endothelial hyperpolarization through the media. Remote conducted smooth muscle hyperpolarizations, monitored ~1.5 mm from the edge of the residual endothelium, were attenuated and abbreviated compared to those in regions where the overlying endothelium remained intact. Such responses were prolonged and amplified ~3-fold in magnitude by the presence of IBMX. In contrast, acetylcholineinduced hyperpolarizations of subintimal smooth muscle in strips with intact overlying endothelium were maintained for at least 10 min and not amplified by IBMX, although their dependence on cAMP synthesis was confirmed by inhibition of adenylyl cyclase with 2⬘,5⬘-DDA. A likely explanation for these differences was obtained in experiments with

cAMP and electrotonic hyperpolarization

221

constricted perfused rabbit ear preparations in which an EDHF-type dilator response to acetylcholine was associated with an endothelium-dependent intraluminal release of cAMP unaffected by inhibition of gap junctional communication. This cAMP efflux continued for at least 20 min, contrasting with the transient accumulation of cAMP in smooth muscle observed in iliac artery rings, thus suggesting that sustained endothelial AMP synthesis may selectively maintain myoendothelial gap junctions in a high conductance state. In parallel dye transfer studies cAMP was shown to modulate the molecular permeability of both myoendothelial and smooth muscle gap junctions. In femoral artery segments in which the endothelium was loaded with a fluorescent dye, the subsequent transfer of the tracer to underlying smooth muscle cells was enhanced by both IBMX and 8-bromo-cAMP. These observations are consistent with the amplifying effect of IBMX on remote hyperpolarizations of smooth muscle in strips partially denuded of their endothelium, although the essential role of myoendothelial gap junctions was confirmed by observations that Gap 27 caused sequestration of dye within the endothelium. The ability of cAMP to enhance the permeability of gap junctions appeared to be nucleotide-specific, as it was not mimicked by 8-bromo-cGMP. Although in theory cAMP could hyperpolarize endothelial and smooth muscle cells by promoting the activation of K⫹ channels (Graier et al., 1993), this action does not appear to contribute significantly to the EDHF phenomenon in rabbit arteries as IBMX did not affect smooth muscle membrane potential in preparations without endothelium, despite causing sustained elevations in cAMP levels. Rather, the mechanism of action of this nucleotide may involve phosphorylation of connexin proteins by protein kinase A (PKA) as EDHF-type relaxations of rabbit arteries are attenuated by the PKA inhibitor Rp-8bromoadenosine-3⬘,5⬘-cyclic monophosphorothioate (Taylor et al., 2001). The endothelial hyperpolarization that is critical to the EDHF phenomenon is now thought to be mediated via Ca2⫹-activated K⫹ channels (KCa) located on endothelial cells, whose sustained opening depends on capacitative influx of Ca2⫹ triggered by depletion of the endoplasmic reticulum Ca2⫹ store following agonist stimulation (Fukao et al., 1997a; Davis and Sharma, 1997; Edwards and Weston, 2001). Receptor-independent EDHF-type relaxations can thus be evoked by direct depletion of the endoplasmic reticulum Ca2⫹ store with cyclopiazonic acid (Fukao et al., 1997a). Observations that EDHF-type responses to cyclopiazonic acid are attenuated by inhibition of adenylyl cyclase with 2⬘,5⬘-DDA and blockade of gap junctions with Gap 27 or 18␣-GA (Chaytor et al., 1998; Taylor et al., 1998, 2001) suggest that capacitative Ca2⫹ influx may not simply underpin the activation of endothelial KCa channels, but also regulate transmission of the resulting hyperpolarization into the media by promoting the synthesis of cAMP and modulating the conductance of gap junctions. Indeed, cyclopiazonic acid has been shown to stimulate cAMP release from endothelial cells in the rat vasculature (Kamata et al., 1996). To further investigate the relationship between endothelial cAMP synthesis and intracellular Ca2⫹ levels, efflux of this nucleotide was measured during constriction of rabbit ear preparations by phenylephrine. This ␣1-adrenergic agonist does not stimulate the endothelium directly but elevates endothelial Ca2⫹ following the transmission of a signal (possibly InsP3 or Ca2⫹ ions themselves) from activated smooth muscle into the endothelium via myoendothelial gap junctions (Dora et al., 1997). In rabbit ears, phenylephrine induced a biphasic cAMP response, with an initial endothelium-independent transient efflux (presumed to originate from smooth muscle cells) being followed by sustained nucleotide release that was abolished by endothelial denudation or perfusion with the gap junction inhibitor 18␣-GA. The time course and magnitude of the pressor response to phenylephrine were nevertheless unaffected by such interventions, so that the indirect endothelium-dependent component of the associated cAMP response did not modulate constriction. This observation is consistent with the

222

Tudor M. Griffith et al.

hypothesis that endothelial synthesis of cAMP fails to influence the tone of the smooth muscle in the absence of a co-existent endothelial hyperpolarization that can be conducted into the media. Observations that interventions that increase intercellular Ca2⫹ promote cAMP synthesis by the endothelium whether they are mediated by capacitative or indirect mechanisms, suggest that a common underlying event could be non-specific activation of a Ca2⫹-stimulated adenylyl cyclase isoform. The pathways that contribute to prostanoid-independent elevations in smooth muscle cAMP in response to agonists nevertheless remain speculative. One possibility is that endothelium-derived cAMP diffuses into subjacent smooth muscle cells as this nucleotide readily negotiates gap junction channels (Griffith and Taylor, 1999). However, phenylephrine does not cause sustained elevations in smooth muscle cAMP levels despite promoting endothelial cAMP synthesis (Taylor et al., 2001). An alternative hypothesis, therefore, is that a conducted endothelial hyperpolarization itself contributes to smooth muscle cAMP accumulation, even though mammalian adenylyl cyclase isoforms are not directly regulated by membrane potential (Cooper et al., 1998). Hyperpolarization-associated closure of L-type channels and reductions in smooth muscle [Ca2⫹]i (Bolz et al., 1999) could in theory elevate cAMP levels by inhibiting the Ca2⫹-activated Type I phosphodiesterase (Houslay, 1998) or activating the Ca2⫹-inhibited Type V/VI adenylyl cyclase isoforms that are closely coupled to L-type Ca2⫹ channels in smooth muscle (Murthy and Makhlouf, 1998). Either of these possibilities would be consistent with the observation that Gap 27 prevented the cAMP accumulation promoted by acetylcholine in rabbit iliac artery rings with intact endothelium. A more complex scenario is suggested by sandwich bioassay experiments with the Ca2⫹ ionophore A23187, which elevates endothelial Ca2⫹ levels through a receptor-independent mechanism and also causes the cell membrane to become leaky (Xue et al., 1999). In contrast to acetylcholine, this ionophore promotes luminal release of a relaxant factor from the endothelium of rabbit arteries under conditions of combined NO synthase and cyclooxygenase blockade whose effects appear to be attenuated by 2⬘,5⬘-DDA and are amplified by IBMX (Hutcheson et al., 1999; Chaytor et al., 2002). It is therefore conceivable that endothelial cells are able to synthesize a novel mediator which normally elevates smooth muscle cAMP levels following transfer via gap junctions on stimulation with acetylcholine, whereas an ‘overspill’ of the same factor elevates smooth muscle nucleotide levels via an extracellular route following administration of A23187. Theoretically, the same factor could also contribute to the synthesis of cAMP by endothelial cells demonstrated in the present study. In conclusion, EDHF-type relaxations and hyperpolarizations evoked by acetylcholine in rabbit arteries have a dual requirement for gap junctional communication and co-existent cAMP synthesis. The findings are consistent with the hypothesis that EDHF-type relaxations are mediated by passive electrotonic spread of endothelial hyperpolarization through the vessel wall with elevations in endothelial cAMP levels being necessary, but not sufficient, to evoke the EDHF phenomenon. ACKNOWLEDGEMENT The work was supported by the Medical Research Council.

29 Myoendothelial gap junctions – the critical link for endothelium-derived hyperpolarizing factor Marianne Tare, Shaun L. Sandow, Harold A. Coleman, Susan J. Wigg, Helena C. Parkington and Caryl E. Hill Since its discovery in the late 1980s, controversy has surrounded the identity of endotheliumderived hyperpolarizing factor (EDHF). In some blood vessels its action has been ascribed to the release of a diffusible factor, while in others there is evidence that electrotonic spread of hyperpolarization generated in endothelial cells to the underlying smooth muscle cells via gap junctions accounts for the responses attributed to EDHF. In this study the hypothesis that myoendothelial gap junctions are necessary for transmission of EDHF to the smooth muscle cells was tested. The relationship between endothelial cell hyperpolarization, existence of myoendothelial gap junctions and EDHF-dependent smooth muscle hyperpolarization was compared in two rat arteries: the femoral artery which lacks EDHF-attributed relaxation, and the mesenteric artery, well documented for its robust EDHF-dependent relaxation. In the femoral artery, the resting membrane potential of endothelial cells was significantly less negative than that of smooth muscle cells and acetylcholine evoked large EDHF-attributed hyperpolarization only in the endothelial cells. The lack of hyperpolarization in the smooth muscle cells correlated with an absence of myoendothelial gap junctions in the femoral artery. This contrasted with observations in the mesenteric artery, in which similar resting membrane potentials and EDHF-attributed hyperpolarizations were recorded in both cell types, corresponding with the existence of myoendothelial gap junctions. Inhibition of gap junctions with 37,43 Gap 27 reduced the EDHF-attributed hyperpolarization of the smooth muscle in the mesenteric artery. When hyperpolarization was selectively generated in endothelial or smooth muscle cells it was transmitted to the other cell type only in the mesenteric artery. These results demonstrate that myoendothelial gap junctions are an essential component of the EDHF pathway in these vascular beds.

The characteristics of the nitric oxide (NO) synthase- and cyclooxygenase-resistant hyperpolarization and relaxation attributed to EDHF have yielded regional and species variation. This has fuelled contention about whether the EDHF hyperpolarization and relaxation is mediated by a diffusible factor(s) or by electrical coupling between endothelial and smooth muscle cells (Little et al., 1995; Campbell et al., 1996; Bény, 1997; Chaytor et al., 1998; Edwards et al., 1998; Yamamoto et al., 1998; Fisslthaler et al., 1999; Emerson and Segal, 2000a; Coleman et al., 2001a). In reality, these studies likely demonstrate that the mechanism(s) underlying EDHF varies with the vascular bed being examined. EDHF has been used as an all-encompassing term to describe the NO- and cyclooxygenase-resistant hyperpolarization and relaxation attributed to a diverse range of entities including K⫹ ions, epoxyeicosatrienoic acids (EETs), hydrogen peroxide and electrical coupling (Félétou and Vanhoutte, 1999b; Hill et al., 2001; McGuire et al., 2001). In small arteries and arterioles the hyperpolarization of the smooth muscle and relaxation attributed to EDHF appear to depend to a greater extent on electrical coupling, rather than on the release of a diffusible factor.

224

Marianne Tare et al.

Reflecting this, the importance of EDHF in endothelium-dependent relaxation increases as vessel size decreases (Hwa et al., 1994; Shimokawa et al., 1996). This observation correlated with the relative incidence of myoendothelial gap junctions along the mesenteric arterial bed (Sandow and Hill, 2000). In large blood vessels, even if myoendothelial gap junctions were present, the electrical sink, due to the large volume of smooth muscle cells, would dissipate the endothelial cellderived hyperpolarization to render it ineffectual in bringing about relaxation (Bény, 1999). Under this paradigm, an EDHF response in large arteries could be accomplished by diffusible factors. Further studies have suggested a mechanism for myoendothelial gap junctions playing an integral role in the EDHF response in some larger arteries via a cAMPdependent mechanism (Chaytor et al., 2001; Campbell and Gauthier, 2002; Griffith et al., 2002). Whatever the mechanism, in muscular arteries there is the potential for both EDHF-dependent diffusible and contact-mediated mechanisms to operate in concert. Such a mechanism has been proposed for blood vessels such as the rat mesenteric artery. The functional role of myoendothelial gap junctions in the EDHF response has been difficult to establish, and has been particularly hampered by the non-specific actions of putative gap junction inhibitors (Davidson and Baumgarten, 1988; Chaytor et al., 1997; Imaeda et al., 2000; Santicioli and Maggi, 2000, Coleman et al., 2001a; Tare et al., 2002). To counter this problem peptide mimetics with sequences homologous to the first and second extracellular loops of segments of connexins, the main constituent protein component of gap junctions, have been used to yield more promising results in terms of blocking gap junctions and associated hyperpolarization and relaxation attributed to EDHF (Chaytor et al., 1998; Dora et al., 1999; Edwards et al., 1999a). Given that both contact- and diffusion-mediated mechanisms could potentially underlie the EDHF-mediated response in some muscular arteries, it was intriguing to ask why some muscular arteries such as the femoral artery of the adult rat lack EDHF-mediated relaxation. Stimulation of the endothelium of this artery with acetylcholine evokes submaximal relaxation attributed predominantly to NO (Zygmunt et al., 1995; Wigg et al., 2001). Thus, it is unlikely that a diffusible EDHF is generated by the endothelium of these blood vessels. The existence or otherwise of myoendothelial gap junctions in the femoral artery was unknown, as was the response of its endothelial cells to agonists. These parameters were examined in the present study and compared with findings in the mesenteric artery where there is a very prominent EDHF-mediated relaxation and hyperpolarization. 1. METHODS In all experiments 16-week-old male Wistar rats were used. For the myograph and electrophysiological studies, rats were killed by exsanguination and first order femoral arteries and first and second order mesenteric arteries isolated.

1.1. Myograph experiments Arteries were cleared of surrounding fat and connective tissue and cut into rings of approximately 1 mm in length. Each ring was threaded with two stainless steel wires (40 ␮m in diameter) and secured to two supports on a Mulvany–Halpern style wire myograph. The arteries were stretched in increments to a final tension equivalent to the mean blood pressure of that artery in vivo. The arteries were continuously superfused with physiological saline (mM: NaCl 120, KCl 5, CaCl2 2.5, MgSO4 1.2, KH2PO4 1, NaHCO3 25 and glucose 11) at

Myoendothelial gap junctions

225

35 ⬚C and bubbled with 95% O2 and 5% CO2. Membrane potential of smooth muscle cells was recorded using intracellular glass microelectrodes filled with 1 M KCl and having resistances of 80–100 M⍀. For the myograph studies the endothelium was stimulated for 2 min using discrete application of acetylcholine (10⫺9 M–10⫺5 M).

1.2. Electrophysiological experiments involving dye identified cells A segment of artery was cut longitudinally, opened and carefully pinned flat, with endothelium uppermost, to the base of a recording chamber. Some arteries were left as intact tube preparations. Arteries were continuously superfused with warmed, oxygenated physiological saline. Membrane potentials of cells were recorded using intracellular glass microelectrodes with resistances of 100–150 M⍀. To allow unequivocal identification of every cell recorded from, the tips of the microelectrodes were filled with 2% Lucifer Yellow CH and backfilled with 1 M KCl. Filled cells were visualized during experiments using epifluorescence optics. For this series of experiments the NO synthase inhibitor N␻-nitro-L-arginine methyl ester (L-NAME, 10⫺4 M) and the cyclooxygenase inhibitor indomethacin (10⫺6 M) were present throughout. Endothelial and smooth muscle cells were stimulated for 1 min with acetylcholine. Application of the potassium channel openers 1-ethyl-2-benzimidazolinone (1-EBIO) and levcromakalim was also for 1 min. Gap junctions were disrupted using the connexin mimetic 37,43Gap 27 (sequence SRPTEKTIFII). For these experiments the physiological saline containing the 37,43Gap 27 was recirculated over the preparations for times ranging between 1 and 3 h. Time control experiments were also undertaken over the above mentioned incubation period.

1.3. Serial section electron microscopy Following anaesthesia with an intraperitoneal injection of rompun (8 mg/kg) and ketamine (44 mg/kg), rats were prepared for perfusion with saline containing sodium nitrate (0.1%) via the left ventricle and subsequently fixed at 25 ⬚C with gluteraldehyde (3%) and paraformaldehyde (1%) in 0.1 mM sodium cacodylate with 0.2 mM CaCl2, 0.15 M sucrose and 10 mM betaine (pH 7.35) for 10 min (Sandow et al., 2002). First order femoral and first and second order mesenteric arteries were isolated and further fixed in the same solution for 2 h. Tissues were postfixed in osmium tetroxide (2%) in the same buffer, stained with saturated aqueous uranyl acetate for 2 h and embedded in Araldite 502. Sets of transverse serial sections were cut over a distance of 5 ␮m (50, 100 nm thick sections) from mesenteric and femoral arteries from each of three different Wistar rats. Electron microscopy was undertaken on a Philips 7100 transmission electron microscope, with all photographs being taken on plate film. Serial sections were collected on Formvar and carbon coated slot grids and stained with uranyl acetate and lead citrate. All myoendothelial gap junctions were identified and photographed at ⫻20,000 to ⫻40,000 through the three series of serial sections. The number of myoendothelial gap junctions per endothelial cell was determined from the appropriate vessel circumference in order to calculate lumenal surface area, with the values for the area of endothelial cells. Selected gap junctions between adjacent endothelial cells or between adjacent smooth muscle cells were photographed at ⫻20,000 to ⫻40,000. Blood vessel characteristics were determined from digitized montages (⫻2,500) of individual transverse arterial sections from each of three different rats. Arterial circumference was determined at the level of the internal elastic lamina. Numbers of medial smooth muscle

226

Marianne Tare et al.

cell layers were determined by averaging the number of smooth muscle cell profiles ⱖ5 ␮m in length, from the edge of the internal elastic lamina to the edge of the external elastic lamina along four linear plots 90⬚ apart.

1.4. Drugs Acetylcholine, carbenoxolone, dilithium Lucifer Yellow CH, indomethacin, L-NAME, phenylephrine (all from Sigma Chemical Co., St Louis, MO, USA), apamin and charybdotoxin were synthesized by Auspep (Melbourne, Vic., Australia), 37,43Gap 27 was synthesized by the Biomolecular Resource facility, John Curtin School of Medical Research (Australian National University, Canberra, A.C.T., Australia), 1-EBIO (Tocris, Bristol, UK). Levcromakalim was a kind gift from Drs Grant McPherson and Barbara Kemp, Monash University, Vic., Australia.

1.5. Data analysis Data are presented as mean ⫾SEM and n refers to the number of animals. Comparisons between data were made using Student’s t-test, paired or unpaired as appropriate using the software package InStat3 (GraphPad Software, San Diego, Ca., USA). Values of P less than 0.05 were considered statistically significant. Concentration–response data were fitted to a sigmoid curve using the least-squares method using Prism3 (GraphPad Software). For the electron microscopy data, significance was tested using one-way ANOVA followed by pairwise t tests with Bonferroni correction for multiple group comparisons and with paired or unpaired t tests for groups of two.

2. RESULTS

2.1. Comparison of EDHF-mediated relaxation and hyperpolarization Femoral and mesenteric arteries mounted on the wire myograph were depolarized and constricted with phenylephrine (3 ⫻ 10⫺6 M and 10⫺5 M, respectively). In the absence of any blockers, stimulation of the endothelium with increasing concentrations of acetylcholine evoked concentration-dependent relaxations in both arteries. The relaxation in the femoral artery was submaximal, unlike that observed in the mesenteric artery (Figure 29.1). Incubation of the preparations with L-NAME (10⫺4 M) abolished the relaxation in the femoral artery, whilst in the mesenteric artery the concentration response curve was shifted to the right, but maximal relaxation was still achieved (Figure 29.1). In the mesenteric artery incubation with indomethacin (10⫺6 M) in the continued presence of L-NAME was without further effect. Acetylcholine evoked concentration-dependent hyperpolarizations in the smooth muscle cells of mesenteric arteries, with the response reaching a maximal amplitude of 31 ⫾ 2 mV (n ⫽ 5) (Figure 29.2). The EC50 for the hyperpolarization and maximal hyperpolarization amplitude were not altered in the presence of L-NAME and indomethacin (Figure 29.2). In contrast, acetylcholine evoked hyperpolarizations of only small amplitude in smooth muscle cells of femoral arteries (maximal hyperpolarization 5 ⫾ 1 mV (n ⫽ 5), that were abolished by L-NAME (Figure 29.2).

Myoendothelial gap junctions

227

Figure 29.1 Endothelium-dependent relaxation in mesenteric and femoral arteries mounted on a wire myograph and constricted with phenylephrine (3 ⫻ 10⫺6 M ⫺ 10⫺5 M). Acetylcholine (10⫺9 M ⫺ 10⫺5 M) evoked concentration-dependent relaxations in both arteries. It evoked maximal relaxations in the mesenteric arteries and incubation in L-NAME shifted the curve to the right, but was without effect on the maximal relaxation. Indomethacin was also without further effect. The L-NAME and indomethacin resistant relaxation in the mesenteric artery produced maximal relaxation. Acetylcholine evoked submaximal relaxation in the femoral artery that was abolished by L-NAME.

2.2. Electrophysiological recordings from dye identified endothelial and smooth muscle cells 2.2.1. Endothelium-dependent responses This series of experiments was carried out in the presence of L-NAME plus indomethacin to isolate the response attributed to EDHF and all cells were dye identified using Lucifer Yellow (Figure 29.3). In longitudinal preparations of mesenteric and femoral arteries, the resting membrane potential of the smooth muscle cells was not different in the two arteries (mesenteric ⫺57 ⫾ 1 mV, n ⫽ 5; femoral artery ⫺58 ⫾ 2 mV; n ⫽ 5) (Figure 29.3). While acetylcholine evoked concentration-dependent hyperpolarization of the smooth muscle cells in the mesenteric artery, no hyperpolarization was recorded in smooth muscle cells of the femoral artery. Femoral smooth muscle cells from either the adventitial/medial border or the medial/intimal border did not hyperpolarize in response to acetylcholine (Figure 29.3). The resting membrane potential of mesenteric endothelial cells was not different from that of the smooth muscle cells (Figure 29.3) and similarly, acetylcholine evoked concentration-dependent hyperpolarization of the former. In contrast, the resting membrane potential of endothelial cells of the femoral artery (⫺26 ⫾ 3 mV, n ⫽ 5) was some 32 mV less negative than that of the smooth muscle cells (Figure 29.3). Unlike that seen in the smooth

Figure 29.2 (A) Concentration-dependent hyperpolarization evoked by acetylcholine (ACh) in mesenteric and femoral arteries mounted on a wire myograph and depolarized and constricted with phenylephrine (3 ⫻ 10⫺6 M⫺10⫺5 M). The hyperpolarization was not significantly altered following incubation in L-NAME and indomethacin (indo) in the mesenteric artery. In contrast, the endothelium-dependent hyperpolarization in the femoral artery was abolished in L-NAME. (B) Examples of hyperpolarizations recorded in smooth muscle cells of a mesenteric artery (left) and a femoral artery (right) depolarized with phenylephrine. The response in the mesenteric artery was recorded in the presence of L-NAME and indomethacin and is attributed to EDHF. The small hyperpolarization in the femoral artery was recorded in the absence of inhibitors.

Mesenteric Artery

Femoral Artery

A

Membrane potential (mV)

Endothelial cells –20

–20

–30

–30

–40

–40

–50

–50

–60

–60

–70

–70

B

Membrane potential (mV)

–80

ACh 10–5 M

–80

Smooth muscle cells –40

–40

–50

–50

–60

–60

–70 –80

ACh 10–5 M

–70

ACh

10–5 M

–80

ACh 10–5 M

5 min

Figure 29.3 Endothelium-dependent responses recorded in dye identified endothelial cells and smooth muscle cells of mesenteric and femoral arteries in the presence of L-NAME and indomethacin. (A) an example of dye identified endothelial cells (left panel; bar ⫽ 25 ␮m). In the absence of vasoconstrictors, acetylcholine (ACh) evoked hyperpolarization in endothelial cells of both mesenteric (middle panel) and femoral (right panel) arteries. Note the difference in resting membrane potentials between the endothelial cells of the two arteries. (B) an example of dye filled smooth muscle cells (left panel; bar ⫽ 50 ␮m). EDHF-mediated hyperpolarization of the smooth muscle was recorded in mesenteric arteries (middle panel), but not in femoral arteries (right panel).

Myoendothelial gap junctions

229

muscle cells, acetylcholine evoked hyperpolarization of large amplitude (maximum 45 ⫾ 2 mV, n ⫽ 5) in the endothelial cells of the femoral artery (Figure 29.3). The maximal hyperpolarization in the femoral endothelial cells was approximately two fold larger than that recorded in the endothelial cells of the mesenteric artery. 2.2.2. K⫹ channel openers To investigate the existence of functional electrical cross talk between endothelial cells and smooth muscle cells, K⫹ channel openers were used to selectively open channels in either endothelial cells or smooth muscle cells, with the responses being recorded in both cell types in the two arteries. The ATP-sensitive K⫹ channel (KATP) opener, levcromakalim, was used to generate hyperpolarization in the smooth muscle cells (Murai et al., 1999; White and Hiley, 2000). Levcromakalim evoked hyperpolarization that was recorded in both cell types in the mesenteric artery (5⫻10⫺7 M, endothelial cells: 12⫾1mV; smooth muscle cells: 9.6⫾ 1 mV, n ⫽ 7) but only in the smooth muscle cells of the femoral artery (10⫺6 M, endothelial cells: 0 mV; smooth muscle cells: 10.3 ⫾ 2 mV, n ⫽ 6). When used at concentrations of 10⫺4 M or less, 1-EBIO activates intermediate-conductance Ca2⫹-sensitive K⫹ channels (IKCa) on the endothelial cells (Walker et al., 2001). 1-EBIO evoked hyperpolarizations of small amplitude in endothelial cells of both mesenteric and femoral arteries. Whilst hyperpolarization was also recorded in smooth muscle cells of the mesenteric arteries (10⫺4 M, endothelial cells: 2.4 ⫾ 0.4 mV; smooth muscle cells: 2.0 ⫾ 0.3 mV, n ⫽ 7), it was absent in smooth muscle cells of the femoral artery (10⫺4 M, endothelial cells: 2.7 ⫾ 1 mV; smooth muscle cells: 0 mV, n ⫽ 6). Hyperpolarization evoked by acetylcholine in the endothelial cells in both arteries was sensitive to charybdotoxin (5 ⫻ 10⫺8 M) and apamin (5 ⫻ 10⫺7 M), inhibitors of IKCa- and small-(SKCa) conductance Ca2⫹-activated K⫹ channels on the endothelial cells (Doughty et al., 1999). 2.2.3. Serial section electron microscopy The failure of the hyperpolarizations generated by K⫹ channel openers to conduct from one cell type to the other in the femoral artery, combined with the absence of L-NAME and indomethacin insensitive endothelium-dependent hyperpolarization in the smooth muscle cells is suggestive of a lack of gap junctions between endothelial and smooth muscle cells (myoendothelial gap junctions) in this artery. Gaps in the internal elastic lamina were present in both mesenteric and femoral arteries (Figure 29.4). In mesenteric arteries, projections arising from endothelial cells traversed these gaps in the internal elastic lamina to make contact with smooth muscle cells in the form of characteristic pentalaminar myoendothelial gap junctions (Figure 29.4(B)). Myoendothelial gap junctions occurred with a frequency of 7.3 ⫾ 1.5/5 ␮m (n ⫽ 3) length of artery. Despite the existence of fenestrations in the internal elastic lamina of femoral arteries, no myoendothelial gap junctions were observed (Figure 29.4(A)). Pentalaminar gap junctions occurred between neighbouring endothelial cells and between adjacent smooth muscle cells in both arteries. 2.2.4. Inhibition of gap junctions Earlier studies have found that the putative gap junction inhibitor, glycyrrhetinic acid and its derivates, have non-selective actions. Therefore, in the present study a peptide mimetic of the second extracellular loop of connexin 37 and connexin 43 (37,43 Gap 27; Chaytor et al.,

230

Marianne Tare et al.

Figure 29.4 Ultrastructural relationship of endothelial (ec) and smooth muscle cells (smc) in the femoral (A) and mesenteric artery (B). When examined through serial sections no myoendothelial gap junctions were present in the femoral artery (A); whilst in the mesenteric artery myoendothelial gap junctions were present (B, arrowed and inset). Holes in the internal elastic lamina (iel) were present in both the femoral and mesenteric arteries (e.g. A, asterisk), although myoendothelial gap junctions were only present in the mesenteric artery (arrow). At higher magnification the pentalaminar region is shown in more detail between the arrows in the inset. Bar ⫽ 0.5 ␮m for A and B and 20 nm for the inset.

2001) was used to inhibit gap junctions. The effect of 37,43Gap 27 on endothelial cell and smooth muscle cell hyperpolarization was examined in the mesenteric artery. Following incubation with 37,43Gap 27 (3 ⫻ 10⫺4 M), hyperpolarization attributed to EDHF was reduced to 33 ⫾ 11% (n ⫽ 5) in the smooth muscle cells. The time for maximal inhibition of the EDHF hyperpolarization in smooth muscle cells varied considerably, from 0.5 to 3 h. 37,43 Gap 27 did not inhibit the hyperpolarization recorded in endothelial cells (Figure 29.5). 3. DISCUSSION This study has revealed the obligatory role of myoendothelial gap junctions in conducting the charybdotoxin and apamin-sensitive EDHF to the smooth muscle cells in rat femoral and mesenteric arteries. The absence of myoendothelial gap junctions in the femoral artery explains why EDHF-dependent hyperpolarization and relaxation of the smooth muscle are not observed in this artery. The femoral artery has approximately double the number of smooth muscle cell layers compared with the mesenteric artery and before this study, the lack of EDHF smooth muscle hyperpolarization could have been interpreted in terms of dissipation of the hyperpolarization generated by the single layer of endothelial cells into the electrical sink of a substantial bulk of smooth muscle (Bény, 1999). However, in the present study it was observed that even smooth muscle cells located directly beneath the internal elastic lamina (location verified by dye filling) did not exhibit EDHF-dependent hyperpolarization in the femoral artery. In order to further examine the potential importance of myoendothelial gap junctions as a pathway for electrical cross talk between the two cell types, hyperpolarizations generated in either the smooth muscle cells or the endothelial cells, using K⫹ channel openers (levcromakalim and 1-EBIO, respectively), were not able to spread to the other cell type in the femoral artery, in contrast to the observations in the mesenteric artery. The importance of myoendothelial gap junctions in mediating cross talk between the two cell types is exemplified

Myoendothelial gap junctions

231

Figure 29.5 Actions of 37,43Gap 27 on the EDHF-attributed hyperpolarizations in mesenteric arteries. Membrane potential recordings in a smooth muscle cell in the presence of L-NAME and indomethacin. Acetylcholine (ACh) evoked hyperpolarization of the smooth muscle cell attributed to EDHF. Following incubation in 3 ⫻ 10⫺4 M 37,43Gap 27 the EDHF hyperpolarization was significantly reduced. Both responses were recorded in the same cell. Parallel bars indicate a break in the recording for 3 h. (B) Influence of 37,43 Gap 27 on the magnitude of EDHF-attributed hyperpolarization recorded in dye identified endothelial cells and smooth muscle cells. Hyperpolarization of the endothelial cells was unaltered, however EDHF-mediated hyperpolarization in the smooth muscle cells was reduced significantly (P ⬍ 0.05).

in arterioles where the endothelial cell to smooth muscle cell ratio is close to one and the two cell layers function as a single syncytium. Heterocellular dye coupling (Little et al., 1995) and strong electrical coupling between endothelial cells and smooth muscle cells (Yamamoto et al., 1998; Emerson and Segal, 2000a; Coleman et al., 2001a) exists and is indicative of functional myoendothelial gap junctions. In such blood vessels, changes in membrane potential evoked in one cell type are mirrored in the other cell type. Responses generated exclusively in the smooth muscle cells, such as excitatory junction potentials evoked by perivascular nerve stimulation and action potentials, are also recorded in the endothelial cells of arterioles and are indistinguishable from the recordings in the smooth muscle cells (Coleman et al., 2001a). Thus, each cell type has the ability to influence the membrane potential in the other cell type in a comparable manner. In contrast, in arteries where the disparity between endothelial cell and smooth muscle cell ratio is great, such as in the porcine ciliary artery (Bény et al., 1997), the membrane potential of the smooth muscle cells may drive the membrane potential of the endothelial cells (Bény, 1999). The lack of myoendothelial gap junctions in the femoral artery most likely accounts for the significant difference in the resting membrane potentials of smooth muscle cells and endothelial cells in the femoral artery. The resting membrane potential of smooth muscle cells of femoral arteries is similar to that reported in other arteries, however, that of the

232

Marianne Tare et al.

femoral endothelial cells was some 30 mV less negative than in the smooth muscle cells. In the mesenteric artery, where myoendothelial gap junctions were present, the resting membrane potentials of the endothelial cells were not different from those in the smooth muscle cells, similar to the observation in hamster cheek pouch arterioles (Emerson and Segal, 2000a) and guinea-pig mesenteric arterioles (Coleman et al., 2001b). The electrical connectivity of smooth muscle cells and endothelial cells in the mesenteric artery would thus enable the resting membrane potential of the smooth muscle cells to influence that of the endothelial cells in that blood vessel. However, in the femoral artery the endothelial cells are electrically isolated, so their resting membrane potential will not be influenced by that of the smooth muscle cells. This difference may be reflected in studies of isolated and cultured endothelial cells that are reported to have wide ranging resting membrane potentials, dictated by differences in the predominating ionic conductances. Cells with a dominant inward rectifier K⫹ conductance have resting membrane potentials similar to smooth muscle cells (range ⫺70 to ⫺60 mV) whilst endothelial cells with dominant Cl⫺ conductances have less negative resting membrane potentials (⫺40 to ⫺10 mV) (Nilius and Droogmans, 2001). Definitive evidence for the functional role of myoendothelial gap junctions in EDHFmediated responses has been sought using pharmacological gap junction uncouplers, yet these have often yielded contradictory results. This apparently stems from the finding that many of the agents used such as glycyrrhetinic acid derivatives and heptanol exert non-selective actions, which include inhibition of action potentials, ion transport processes, agonist induced contraction of smooth muscle and even hyperpolarization of endothelial cells (Davidson and Baumgarten, 1988; Chaytor et al., 1997; Imaeda et al., 2000; Santicioli and Maggi, 2000; Coleman et al., 2001a; Tare et al., 2002). Of relevance to the present study was the observation that the water-soluble glycyrrhetinic acid derivative, carbenoxolone, inhibited EDHF-attributed hyperpolarization in identified endothelial cells (Tare et al., 2002). Inhibitory gap junction peptides that possess conserved homology to segments of the first and second extracellular loops of the connexins were developed to examine gap junction function (Dahl et al., 1994; Evans and Bositano, 2001). These peptide inhibitors were utilized and developed further to target vascular gap junction connexins (connexins 37, 40 and 43) and thus specifically examine myoendothelial gap junction-mediated EDHF responses (Chaytor et al., 1998, 2001; Griffith et al., 2002). In the present study 37,43Gap 27 (a peptide homologous to a segment of the sequence of the second extracellular loop common to both connexins 37 and 43), reduced EDHF attributed hyperpolarization in the smooth muscle cells, but not in dye identified endothelial cells of the mesenteric artery. A similar observation has already been described in the same blood vessel (Edwards et al., 1999a). Given that connexons and hence gap junctions, may be composed of more than one connexin subtype, a combination of peptide inhibitors to target connexins 37, 40 and 43 may be more effective. Indeed, in the rat hepatic artery a combination of Gap peptide inhibitors, 43Gap 26, 40Gap 27 and 37,43Gap 27, targeting these connexins was more effective in combination than alone (Chaytor et al., 2001). When combinations of peptides were used in isolated and pressurized rat mesenteric arteries they reduced EDHF-attributed relaxation to an even greater extent than in studies where 37,43 Gap 27 alone was used (Griffith et al., 2001). However, a residual relaxation of around 10% may be accounted for by the actions of K⫹, a putative EDHF, in this artery (Edwards et al., 1998). Although the actions of the Gap peptides seem to spare the EDHF-attributed hyperpolarization in the endothelial cell, the influence of these connexin mimetics are not limited to myoendothelial gap junctions as they also inhibit gap junctions between adjacent smooth muscle cells (Chaytor et al., 1997). However, 37,43Gap 27 appears to have a limited effect on

Myoendothelial gap junctions

233

increasing the input resistance of mesenteric arteriole smooth muscle cells (Crane et al., 2001), a situation that would be expected if they were indeed blocking activity in specific gap junctions. Thus, the limited effectiveness of 37,43Gap 27 in this tissue may reflect the dominance of other connexin types or species differences. In the present study, the femoral artery of 16-week-old rats has provided a suitable model to test the importance of myoendothelial gap junction communication in the EDHFmediated response. Indeed, this blood vessel corresponds to a form of selective ‘knock out’ of EDHF and myoendothelial gap junctions and as such provides an excellent model for pursuing the hypothesis examined in the present study. Furthermore, the present results demonstrate that myoendothelial gap junctions can provide the critical link in the transmission of the charybdotoxin plus apamin-sensitive EDHF to the smooth muscle.

30 Longitudinal spread of agonist-evoked hyperpolarization in the rat mesenteric artery H. Takano, K.A. Dora and C.J. Garland

Functional evidence suggests that myoendothelial gap junctions in the rat mesenteric artery may facilitate the spread of current from hyperpolarized endothelial cells to adjacent smooth muscle cells. Morphological evidence supports this heterocellular coupling, and in addition, the possibility for homocellular coupling. The present study investigated the extent to which hyperpolarization of smooth muscle cells can spread along the longitudinal axis of mesenteric arteries. Membrane potential was measured with glass microelectrodes in smooth muscle. Either acetylcholine or levcromakalim were focally pressure-ejected from glass pipettes to a small segment of artery. Both direct and indirect hyperpolarization of smooth muscle cells evoked similar conducted responses (to over 2 mm upstream), and the conducted response was markedly reduced in arteries without endothelium. The injection of propidium iodide into either endothelial or smooth muscle cells indicated both homo- and heterocellular coupling, but the spread was more effective between the endothelial cells. These data indicate that the hyperpolarization of smooth muscle, stimulated either by endothelium-derived hyperpolarizing factor or directly with agonists, spreads longitudinally along the axis of the artery, and this spread is facilitated by the endothelium.

1. INTRODUCTION Gap junctions play a significant role during endothelium-dependent hyperpolarizations of smooth muscle in rat mesenteric artery (Edwards et al., 1999). Physical evidence for myoendothelial gap junctions strengthens the argument that radial coupling has an important role in this aspect of arterial function (Sandow and Hill, 2000). However, in many small arteries and arterioles, the spread of hyperpolarization occurs longitudinally along the vessel axis (Duling and Berne, 1970; Segal and Duling, 1986). Here both homo- and heterocellular gap junctions could provide a pathway for such a spread (Yamamoto et al., 2001), which could then enable vasodilatation. In arterioles, simultaneous intracellular recording from two cells showed that the injection of either hyperpolarizing or depolarizing current could evoke distant relaxations or contractions of smooth muscle, respectively. This was irrespective of whether the injection was into an endothelial or a smooth muscle cell (Emerson and Segal, 2000a). However, in these arterioles, the spreading changes in membrane potential depended on the integrity of the endothelium (Emerson and Segal, 2000b). The present study established whether or not endothelium-dependent and -independent hyperpolarizations could spread longitudinally along the length of rat small mesenteric arteries, and investigated the role of the endothelium in facilitating any cell to cell coupling.

Longitudinal spread of agonist-evoked hyperpolarization

235

2. METHODS

2.1. Electrical coupling Third order branches of the superior mesenteric artery were obtained from male Wistar rats (200–250 g). A segment (3 mm long) of artery (diameter circa 200 ␮m) was pinned out and superfused with 3-[N-Morpholino]propane-sulfonic acid solution at 37 ⬚C at 2 ml/min. Smooth muscle membrane potential measurements were obtained with glass microelectrodes (tip resistance 50–100 M⍀, 2 M KCl). The impaled cell was close to the upstream end of the artery, and where possible (the majority), the cell remained impaled for an entire series of local and conducted responses. Either acetylcholine (10⫺3 M) or levcromakalim (10⫺3 M) were focally pressure-ejected from glass pipettes (5 ␮m tip) onto a small segment of artery. The stimulation pipette was moved from a position adjacent to the recording electrode (local site), to positions downstream to the direction of flow of the superfusate (1000␮m and 2000 ␮m, conducted sites). Acetylcholine and levcromakalim (White and Hiley, 2000) were used to cause endothelium-dependent and -independent hyperpolarization of smooth muscle, respectively. The duration of stimulation was varied to evoke responses of a similar magnitude at the local site.

2.2. Dye coupling An artery was mounted with a side branch uppermost. The branch was removed, leaving a small hole through which the endothelial cell layer could be viewed. Impalement was then made in either endothelial cells, or by advancing the microelectrode, in smooth muscle cells. Dye coupling was established by injection of the cell impermeant, low molecular weight dye, propidium iodide (Emerson and Segal, 2001). Propidium iodide (1%) was dissolved in 2 M KCl, and filled in the tip of the recording electrode (tip resistance 100–150 M⍀). After recording the membrane potential, the vessel was illuminated with a 100-W mercury lamp through an excitation filter with a transmission range between 510–550 nm. The resulting image was acquired as the dye fluorescence (⬎570 nm) using a cooled CCD camera (Cool snap HQ, Roper scientific, Munich, Germany).

2.3. Materials The MOPS solution contained (in mol⫺3/L): 145 NaCl, 4.7 KCl, 2.0 CaCl2, 1.17 MgSO4, 2.0 MOPS, 1.2 NaH2PO4, 5.0 glucose, 2.0 pyruvate Na, 0.02 EDTA, 2.75 NaOH, and the pH of the solution was 7.39⫺7.41 at 37 ⬚C. All drugs used were purchased from Sigma Chemical Company, Poole, UK. Levcromakalim was a generous gift from Glaxo SmithKline, UK. 3. RESULTS

3.1. Dye coupling Impalement of a single cell was sufficient for diffusion of propidium iodide into the cell, without any requirement for current injection. The dye tended to stain the nuclei of cells, and clearly showed the images of two different types of cells (Figure 30.1). One group of cells were aligned longitudinally (Figure 30.1, left panel, n ⫽ 5), characteristic of endothelial cells; and another group of cells were oriented radially, characteristic of smooth muscle cells

236

H. Takano et al.

* *

Figure 30.1 Assessment of homocellular and heterocellular coupling by dye injection. A single endothelial cell (left panel) or smooth muscle cell (right panel, each indicated by *) was impaled with a glass microelectrode containing propidium iodide (1%). Dye appeared to pass more readily to adjacent endothelial cells than smooth muscle cells. Bar ⫽ 50 ␮m.

Figure 30.2 Experimental set-up for intracellular membrane potential measurements. A single smooth muscle cell was impaled with a sharp glass microelectrode at 0 ␮m. A glass pipette containing either acetylcholine or levcromakalim was positioned at 0 ␮m (local), 1000 ␮m downstream and 2000 ␮m downstream from the sharp glass electrode.

(Figure 30.1, right panel, n ⫽ 3). The dye could spread between endothelial cells much more readily than between smooth muscle cells, and transfer between the two cell types was rarely observed (not shown).

3.2. Electrical coupling The resting membrane potential of smooth muscle cells of the mesenteric artery, with cells impaled from the adventitial side averaged ⫺50.3 ⫾ 0.5 mV (n ⫽ 164). Acetylcholine evoked a rapid hyperpolarization at the site of stimulation (magnitude 14.3 ⫾ 1.2 mV, n ⫽ 8). This response travelled upstream along the vessel (Figures 30.2 and 30.3), decaying with distance (3.6 ⫾ 0.4 mV, n ⫽ 4, at 2000 ␮m). All conducted (but not local) responses could be blocked by cutting the vessel between the site of stimulation and recording (Figure 30.4, left panel). The ability of the endothelium to facilitate a conducted hyperpolarization was assessed by directly hyperpolarizing the smooth muscle cells with the ATP-sensitive K channel opener, levcromakalim. The membrane potential of smooth muscle cells in arteries without endothelium did not differ from those in endothelium-intact arteries (⫺56.7 ⫾ 1.4 mV, n ⫽ 53, P ⬎ 0.05). The magnitude of the levcromakalim-evoked hyperpolarization at the local (0 ␮m) site was not significantly different between arteries with and without endothelium (magnitude 12.1 ⫾ 1.2 mV, n ⫽ 10 and 13.8 ⫾ 1.9 mV, n ⫽ 4, respectively, P ⬎ 0.05). In a manner almost

Figure 30.3 Representative records of local and conducted smooth muscle membrane potential hyperpolarization to agonists. Acetylcholine (10⫺3 M) or levcromakalim (10⫺3 M) were pressure-pulse ejected at the local and distant sites for 5 ms or 20 ms, respectively (indicated by triangle). Each series of traces was obtained from the same cell impalement.

Acetylcholine

Levcromakalim 1.0

Relative AUC

1.0

Control Endothelium-denuded

Control Cut 0.5

0.5

0.0

0.0 0

500

1000

1500

Distance (µm)

2000

0

500

1000

1500

2000

Distance (µm)

Figure 30.4 Summary data for local and conducted hyperpolarization mediated by acetylcholine and levcromakalim. Each response was evaluated as the area under the curve (AUC) of the transient hyperpolarization and then normalized against the local response (agonist pipette at 0 ␮m). A cut between the recording and application pipettes abolished conducted (but not local) hyperpolarization to acetylcholine, ruling out any diffusion artefact. Further, in endothelium-denuded arteries, the local response to levcromakalim was unaffected, whereas the upstream responses were markedly reduced. Values are means ⫾ SEM of n ⫽ 4–10 experiments.

238

H. Takano et al.

identical to that with the acetylcholine-induced responses, the local hyperpolarization travelled upstream along the blood vessel (Figure 30.4, right panel). However, this was only the case in arteries with endothelium. At 1000 ␮m upstream from the stimulated site, the amplitude of hyperpolarization decayed significantly in arteries without endothelium (the amplitude was 2.9 ⫾ 0.7 mV, n ⫽ 4 in arteries with endothelium, compared with 0.5 ⫾ 0.5 mV, n ⫽ 4 in arteries without endothelium). 4. DISCUSSION The present data indicate that hyperpolarization of the smooth muscle, caused in the mesenteric artery by either endothelium-derived hyperpolarizing factor or directly with agonists, can spread longitudinally along the axis of the artery. Furthermore this spread is facilitated by the endothelium. The presence of gap junctions between cells in the arterial wall serves to coordinate changes in blood vessel diameter. Local activation of endothelial cells with acetylcholine evokes dilatation that can spread bi-directionally along the length of arterioles (Duling and Berne, 1970; Segal and Duling, 1986; Segal and Beny, 1992; Kurjiaka and Segal, 1995) and small arteries (Emerson and Segal, 2000b) and in so doing, can increase blood flow to an arteriole (Kurjiaka and Segal, 1995; Dora et al., 2000). The most likely explanation for such spreading dilatation is that changes in membrane potential spread through the vessel as a result of cellular electrical coupling. In arterioles, the endothelium-dependent agonist, acetylcholine stimulates hyperpolarization in both endothelial and smooth muscle cells, both at the point of stimulation (Yamamoto et al., 1999, 2001), and at sites upstream from the stimulation site (Emerson and Segal, 2000a,b). As acetylcholine was able to evoke spreading hyperpolarization in small mesenteric arteries of the rat, it may be that a similar mechanism operates in this slightly larger vessel. This suggestion was supported by the finding that direct hyperpolarization of smooth muscle cells with levcromakalim was not conducted unless the endothelium was intact. The local hyperpolarization was not reduced by removing the endothelium, indicating that the integrity of the smooth muscle cell layer had not been affected. However, the ability of the hyperpolarization to spread was markedly reduced compared with arteries with endothelium. As well as indicating the crucial importance of the endothelium in conducting the hyperpolarization, the lack of spread to conducted sites provides further evidence that the agonist was not diffusing to stimulate the upstream sites directly. These data also show that the smooth muscle cells do not contribute significantly to the spread of hyperpolarization. This may relate to the arrangement of the smooth muscle cells, as the hyperpolarization must pass through a greater number of cells, and hence more gap junctions, to reach distant sites, compared to spread between the longitudinally arranged endothelial cells (Haas and Duling, 1997). This finding is consistent with those obtained in other vascular beds, where the coupling between smooth muscle cells is poor, and the pathway for spread of a hyperpolarizing signal appears crucially dependent on a functional endothelium (Emerson and Segal, 2000b, 2001; Yamamoto et al., 2001). 5. CONCLUSION The present findings show that a locally stimulated endothelium-dependent hyperpolarizing signal can spread to upstream sites and evoke distal hyperpolarization. Further, the pathway for the spread of smooth muscle cell hyperpolarization is crucially linked to the presence of a functional endothelium.

31 Effect of HEPES on EDHF responses in porcine coronary and rat mesenteric arteries M.J. Gardener, G. Edwards, M. Félétou, P.M. Vanhoutte and A.H. Weston

The mechanism underlying the endothelium-derived hyperpolarizing factor (EDHF) response in the porcine coronary artery remains unclear. One possibility is that efflux of K⫹ from the endothelium creates a concentration gradient which drives the movement of K⫹ from the smooth muscle into the endothelial cells through gap junctions thus resulting in the observed hyperpolarization of the smooth muscle. Since HEPES, a frequently-used buffer of physiological salt solutions, inhibits gap junctions (Bevans and Harris, 1999), the aim of the present study was to determine the effect of HEPES on the EDHF response. Overnight incubation of porcine coronary arteries in bicarbonate-buffered Krebs solution had no effect on the EDHF response induced by substance P or bradykinin. However, a similar incubation in HEPESbuffered Tyrode solution or Krebs solution containing HEPES abolished the smooth muscle, but not the endothelial cell hyperpolarization induced by substance P and unmasked an alternative mechanism by which bradykinin hyperpolarizes the smooth muscle. Following overnight exposure of rat mesenteric arteries to HEPES-buffered Tyrode solution, the EDHF response was significantly reduced compared to that seen in both freshly-isolated arteries and blood vessels incubated for a similar period in Krebs solution. Subsequent examination of Cx37, Cx40 and Cx43 protein levels before and after overnight exposure to HEPEScontaining bicarbonate-buffered Krebs solution showed a selective reduction in Cx40. These experiments suggest that HEPES-buffering is unsuitable for the study of EDHF. It selectively reduces Cx40 protein levels and inhibits the component of the EDHF response dependent on myoendothelial gap junctions.

1. INTRODUCTION The mechanism that underlies responses due to release of the endothelium-derived hyperpolarizing factor (EDHF) in the porcine coronary artery remains unclear. There is evidence for the involvement of K⫹ efflux from the endothelium via calcium-sensitive K⫹ channels in rat mesenteric and hepatic arteries and this can be mimicked by a 5 ⫻ 10⫺3 M increase in [K⫹]o (Edwards et al., 1998). However, elevation of [K⫹]o fails to hyperpolarize smooth muscle of the porcine coronary artery. Furthermore, EDHF-induced hyperpolarizations are insensitive to the application of barium and ouabain (blockers of the inwardly rectifying K⫹ channel and the Na⫹/K⫹-ATPase, respectively) unlike those seen in rat arteries. It therefore seems likely that a mechanism additional to that observed in some rat arteries must be involved. One possibility is that K⫹ does efflux from endothelial cells and that the resulting concentration gradient drives the movement of K⫹ from the smooth muscle to the endothelium via gap junctions, thus causing the observed hyperpolarization of the smooth muscle. HEPES is a frequently-used buffer which inhibits gap junctions (Bevans and Harris, 1999). The purpose of the present study was to examine the effect of HEPES buffering on the EDHF response.

240

M.J. Gardener et al.

2. METHODS

2.1. Preparation of blood vessels Porcine hearts were obtained from a local abattoir and transported to the laboratory in ice-cold Krebs solution (the time from heart removal to vessel dissection was approximately 45 min). The left anterior descending coronary artery was dissected free and cleaned of all adherent fat and connective tissue. Mesenteric arteries were obtained from male Sprague–Dawley rats (150–225 g in weight).

2.2. Electrophysiology Intact vessels were continuously superfused (10 ml/min) at 37 ⬚C, either with Krebs solution (gassed with 95% O2/5% CO2) or HEPES-buffered Tyrode solution (gassed with 100% O2) containing 3 ⫻ 10⫺4 M NG-nitro-L-arginine and 1 ⫻ 10⫺5 M indomethacin. Cells were impaled using sharp micro-electrodes of resistance 50–80 M⍀ when filled with 3 M KCl.

2.3. Western blotting 2.3.1. Sample preparation, electrophoresis and blotting Mesenteric vascular beds were dissected and divided into two groups. One half was processed immediately in extraction buffer (t ⫽ 0) and the other incubated for 24 h in Krebs solution or Krebs solution plus 1 ⫻ 10⫺2 M HEPES (pH 7.4) each gassed with 95% O2 /5% CO2 prior to homogenization in extraction buffer. The protein content of samples was quantified with a modified Bradford assay (Bradford, 1976) using a commercially available Coomassie Brilliant Blue G-250-containing protein estimation buffer (BIORAD) in conjunction with a UV spectrophotometer. A595 values were compared with those on a standard curve constructed using bovine serum albumin. Acrylamide gels, 12% ↔ v/v were subsequently loaded with equal weights of total protein from each test group and electrophoresed at 120 V. Protein was then blotted onto PVDF membranes at 80 V (as described by Towbin et al., 1979) and subsequently blocked with 5% w/v dried non-fat milk solution in TwTBS (0.1% v/v tween20 in tris-buffered saline) prior to addition of primary antibodies. 2.3.2. Antibody labelling Primary antibodies (anti- Cx43, Cx40 and Cx37) were each applied overnight at 4 ⬚C at a concentration of 1 ␮g/ml in 1⫻ TwTBS and secondary antibodies (horseradish peroxidaseconjugated goat anti-mouse and goat anti-rabbit) were used at a final concentration of 40 ng/ml in TwTBS. An ECL⫹ chemiluminescence detection system was used to visualize antibody labelling. 2.3.3. Antibodies The primary antibodies used were: mouse monoclonal anti-Cx43 IgG (1 mg/ml (MAB3068): Chemicon, Hampshire UK, affinity-purified rabbit polyclonal anti-Cx37 IgG (CX37A11-A), anti-Cx40 IgG (CX40-A) (1 mg/ml) and corresponding antigenic peptides (CX37A11-P & CX40-P): Autogen-Bioclear, Wiltshire UK. The secondary antibodies used were: goat

Effect of HEPES on EDHF responses

241

anti-rabbit (111-035-003) and goat anti-mouse (115-035-003) HRP conjugated IgG (0.8 mg/ml): Stratech Scientific, Bedfordshire UK. 2.3.4. Solutions and buffers Bicarbonate-buffered Krebs solution contained: NaCl 0.118 M, KCl 3.4 ⫻ 10⫺3 M, CaCl2 2.5 ⫻ 10⫺3 M, KH2PO4 1.2 ⫻ 10⫺3 M, MgSO4 1.2 ⫻ 10⫺3 M, NaHCO3 2.5 ⫻ 10⫺2 M, glucose 1.11 ⫻ 10⫺2 M and was gassed with 95% O2/5% CO2. In the experiments using Krebs solution plus 1 ⫻ 10⫺2 M HEPES, HEPES was added to the above Krebs solution and its pH was adjusted to 7.4 with NaOH while gassing with 95% O2/5% CO2. HEPES-buffered Tyrode solution contained: NaCl 0.140 M, KCl 4.7 ⫻ 10⫺3 M, CaCl2 1.3 ⫻ 10⫺3 M, MgCl2 1.0 ⫻ 10⫺3 M, HEPES 1 ⫻ 10⫺2 M, glucose 1.11 ⫻ 10⫺2 M. The pH was adjusted to 7.4 with NaOH and the solution was gassed with 100% O2. Extraction buffer: the following was made up with distilled water and stored frozen in 5 ml aliquots: Trizma base 2 ⫻ 10⫺2 M, Sucrose 2.5 ⫻ 10⫺4 M, EDTA 5 ⫻ 10⫺3 M, EGTA 1 ⫻ 10⫺2 M, DTT (Dithiothreitol) 1 ⫻ 10⫺2 M, Protease inhibitor cocktail P2714 (containing 4-(2-aminoethyl)benzenesulfonyl fluoride (AEBSF), trans-epoxysuccinyl-L-leucylamido (4-guanidino) butane (E-64), bestatin, leupeptin, aprotinin and sodium EDTA): contents of one vial were added to 100 ml of extraction buffer. When required, the necessary amount was thawed and the protease inhibitor phenylmethylsulphonylfluoride (PMSF) 1 ⫻ 10⫺3 M (0.1 M stock in ethanol made on the day) was added. Tris-buffered saline (TBS): The following was made up to 1l with distilled water and buffered to pH to 8.0 with HCl: Trizma base 2 ⫻ 10⫺3 M, NaCl 1.5 ⫻ 10⫺2 M. TwTBS: 500 ␮l was added to 500 ml TBS. 2.3.5. Reagents 1-EBIO: Aldrich, Dorset UK; acrylamide/Bis solution (40% 37.5:1), BIORAD protein assay dye reagent concentrate (estimation buffer): Biorad, Hertfordshire UK; ethanol, glycine, methanol, NaCl, sucrose: BDH, Leicestershire UK; ECL⫹: Amersham Pharmacia Biotech, Buckinghamshire UK; levcromakalim: SmithKline Beecham, Essex UK; NS1619, substance P: Research Biochemicals International, Nantick USA; PVDF (polyvinylidene difluoride) membrane ‘Immobilon-P’: Millipore, Hertfordshire UK; synthetic iberiotoxin: Latoxan, France; all other compounds: Sigma, Dorset UK. 2.3.6. Statistics Where appropriate data is represented as mean ⫾ SEM. Data was analysed using the paired student’s t-test and was deemed to be significant at P less than 0.05. 3. RESULTS

3.1. Electrophysiology 3.1.1. EDHF responses in porcine coronary arteries incubated in bicarbonate-buffered Krebs solution In porcine coronary arteries incubated for 22–26 h in bicarbonate-buffered Krebs solution 10⫺7 M substance P and 10⫺7 M bradykinin hyperpolarized the smooth muscle by 27.1 ⫾ 1.8 mV (n ⫽ 4) and 26.6 ⫾ 1.0 mV (n ⫽ 4), respectively (Figure 31.1). These changes were

242

M.J. Gardener et al.

Figure 31.1 Typical trace showing the effect of overnight incubation in bicarbonate-buffered Krebs on smooth muscle hyperpolarization in the porcine coronary artery. 1⫻10⫺7 M substance P and 1⫻10⫺7 M bradykinin hyperpolarized the smooth muscle by 27.1 ⫾ 1.8 mV (n ⫽ 4) and 26.6 ⫾ 1.0 mV (n ⫽ 4) respectively, comparable to responses seen in freshly isolated vessels (e.g. substance P: 27.0 ⫾ 0.6 mV, n ⫽ 4). NS1619- and levcromakalim-induced hyperpolarizations also remain like those seen in freshly isolated vessels. Addition of 1⫻10⫺7 M iberiotoxin (IbTX) had no effect on either substance P- or bradykinin-induced hyperpolarizations but abolished responses seen with NS1619. The electrode was not dislodged during the recording; the gaps indicate that parts of the trace have been removed for clarity. Source: Adapted from Edwards et al., 2001.

comparable to those seen in freshly isolated blood vessels (e.g. 1 ⫻10⫺7 M substance P: 27.0 ⫾ 0.6 mV, n ⫽ 4). Hyperpolarizations to 3.3 ⫻10⫺5 M NS1619, an opener of the large-conductance calcium-sensitive K⫹ channel (BKCa) and to 1⫻10⫺5 M levcromakalim, an opener of the ATP-sensitive K⫹ channel (KATP) were also similar to those seen in freshly isolated vessels. The addition of 1⫻10⫺7 M iberiotoxin had no effect on either substance P- or bradykinin-induced hyperpolarizations but abolished responses to 3.3 ⫻10⫺5 M NS1619 (Figure 31.1). 3.1.2. EDHF responses in porcine coronary arteries incubated in HEPES-buffered Tyrode solution Vessels incubated for 24 h in HEPES-buffered Tyrode solution did not respond to 1 ⫻10⫺7 M substance P (Figure 31.2). However, 1 ⫻10⫺5 M levcromakalim, 3.3 ⫻10⫺5 M NS1619 and 1 ⫻10⫺7 M bradykinin all evoked a hyperpolarization of the smooth muscle, although bradykinin-evoked responses (14.8 ⫾ 0.8 mV, n ⫽ 4) were smaller than those seen in vessels incubated in bicarbonate-buffered Krebs solution. Upon addition of 1 ⫻10⫺7 M iberiotoxin, levcromakalim-induced hyperpolarizations remained whereas the effects of both NS1619 and bradykinin were abolished (Figure 31.2). 3.1.3. The effect of addition of HEPES to bicarbonate-buffered Krebs The effect of overnight incubation in HEPES-buffered Tyrode solution could be reproduced by addition of 1 ⫻10⫺2 M HEPES to the bicarbonate-buffered Krebs solution. In these blood vessels, 1 ⫻10⫺7 M substance P failed to induce hyperpolarization. In contrast, the effects of 1 ⫻10⫺7 M bradykinin (16.1 ⫾ 0.4 mV, n ⫽ 4) and 3.3 ⫻10⫺5 M NS1619 (18.0 ⫾ 0.5 mV, n ⫽ 4) were not significantly different from those seen in arteries incubated in HEPES-buffered Tyrode solution (14.8 ⫾ 0.8 mV and 16.3 ⫾ 0.8 mV, respectively, n ⫽ 4) (Figure 31.3).

Effect of HEPES on EDHF responses

243

Figure 31.2 Typical trace showing the effect of overnight incubation in HEPES-buffered Tyrodes on smooth muscle hyperpolarization in the porcine coronary artery. Vessels incubated for 24 h in HEPES-buffered Tyrodes did not respond to 1⫻10⫺7 M substance P but levcromakalim, NS1619 and 1⫻10⫺7 M bradykinin all evoked a smooth muscle hyperpolarization. The bradykinin-induced response (14.8 ⫾ 0.8 mV, n ⫽ 4) was reduced compared to those seen in vessels incubated in bicarbonate-buffered Krebs. Addition of 1⫻10⫺7 M iberiotoxin (IbTX) inhibited both NS1619- and bradykinin-induced hyperpolarizations. The electrode was not dislodged during the recording; the gaps indicate that parts of the trace have been removed for clarity. Source: Adapted from Edwards et al., 2001.

Figure 31.3 Typical trace showing the effect of overnight incubation in bicarbonate-buffered Krebs containing 1⫻10⫺2 M HEPES on hyperpolarization of smooth muscle in the porcine coronary artery. 1⫻10⫺7 M substance P failed to induce a hyperpolarization and the effects of 1⫻10⫺7 M bradykinin (16.1 ⫾ 0.4 mV, n ⫽ 4) were not significantly different to those seen in vessels incubated in HEPES-buffered Tyrodes (14.8 ⫾ 0.8 mV, n ⫽ 4) whilst NS1619-induced hyperpolarizations remained intact (Figure 31.3). The electrode was not dislodged during the recording; the gaps indicate that parts of the trace have been removed for clarity. Source: Adapted from Edwards et al., 2001.

3.1.4. HEPES-buffered Tyrode solution and hyperpolarizations of endothelial cells Microelectrode recordings taken from the endothelial layer of coronary arteries incubated in HEPES-buffered Tyrode solution showed that 1 ⫻10⫺7 M substance P was able to hyperpolarize these cells (26.8 ⫾ 0.6 mV, n ⫽ 4). Responses to 1 ⫻10⫺7 M bradykinin (25.1 ⫾ 0.9 mV, n ⫽ 4) and to 6 ⫻10⫺4 M 1-EBIO (an opener of the intermediate conductance calcium-sensitive K⫹ channel (IKCa) ) (22.6 ⫾ 0.8 mV, n ⫽ 4) were similar to those seen in blood vessels incubated for 24 h in bicarbonate-buffered Krebs solution (e.g. bradykinin: 26.6 ⫾ 1.0 mV, n ⫽ 4) (Figure 31.4).

244

M.J. Gardener et al.

Figure 31.4 Typical trace showing the effect of overnight incubation in HEPES-buffered Tyrodes on endothelial cell hyperpolarizations in the porcine coronary artery. Hyperpolarizations induced by 1⫻10⫺7 M substance P (26.8 ⫾ 0.6 mV, n ⫽ 4), 1⫻10⫺7 M bradykinin (25.1 ⫾ 0.9 mV, n ⫽ 4) and 6⫻10⫺4 M1-EBIO (an opener of the intermediate conductance calcium-sensitive K⫹ channel (IKCa) ) (22.6 ⫾ 0.8 mV, n ⫽ 4) were unaffected compared to freshly isolated vessels (e.g. substance P: 27.8⫾ 0.8mV, n⫽4) or vessels incubated for 24h in bicarbonate-buffered Krebs (e.g. bradykinin: 26.6 ⫾ 1.0 mV, n ⫽ 4). The electrode was not dislodged during the recording; the gaps indicate that parts of the trace have been removed for clarity. Source: Adapted from Edwards et al., 2001.

3.1.5. Buffering and EDHF responses in rat mesenteric arteries Recordings of smooth muscle membrane potential taken from intact rat mesenteric arteries showed that 24 h incubation in bicarbonate-buffered Krebs solution had no effect on hyperpolarizations induced by 3.3 ⫻10⫺5 M NS1619 (7.7 ⫾ 1.0 mV, n ⫽ 4) by 1 ⫻10⫺5 M levcromakdalim (25.4 ⫾ 0.7 mV, n ⫽ 4) or by 1 ⫻10⫺5 M acetylcholine (22.2 ⫾ 0.4 mV, n ⫽ 4) compared to those obtained in freshly isolated vessels (e.g. 1⫻10⫺5 M acetylcholine: 21.4⫾0.7 mV, n⫽4) (Figure 31.5). However acetylcholine-induced hyperpolarizations recorded from arteries incubated in HEPES-buffered Tyrode solution were significantly smaller (14.5 ⫾ 0.5 mV, n ⫽ 4 [P ⬍ 0.05]) (Figure 31.5) than those seen in preparations incubated in bicarbonate-buffered Krebs solution. In contrast, responses to 3.3 ⫻10⫺5 M NS1619 and 1 ⫻10⫺5 M levcromakalim were unaffected.

3.2. Western blotting 3.2.1. Levels of connexins in rat mesenteric arteries after 24 h incubation Levels of connexin protein present in rat mesenteric arteries incubated overnight in either bicarbonate-buffered Krebs solution or bicarbonate-buffered Krebs solution plus 1 ⫻10⫺2 M HEPES were compared to matched time zero controls. In blood vessels incubated in Bicarbonate-buffered Krebs solution there was no difference between the levels of the connexins Cx43, Cx40 and Cx37 compared to time zero controls (Figure 31.6). However, in arteries incubated in Bicarbonate-buffered Krebs solution plus 1 ⫻10⫺2 M HEPES, Cx40 levels were greatly reduced whilst Cx43 and Cx37 levels remained similar to those in the time zero controls (Figure 31.6).

Effect of HEPES on EDHF responses

245

Figure 31.5 Typical trace showing the effect of overnight incubation in bicarbonate-buffered Krebs or HEPES-buffered Tyrodes on smooth muscle hyperpolarization in the rat mesenteric artery. Twenty-four hour incubation in bicarbonate-buffered Krebs had no effect on hyperpolarizations induced by NS1619 (7.7⫾1.0mV, n⫽4), levcromakalim (25.4⫾0.7mV, n ⫽ 4) and 1⫻10⫺5 M acetylcholine (22.2 ⫾ 0.4 mV, n ⫽ 4) compared to freshly isolated vessels (e.g. 1⫻10⫺5 M acetylcholine: (21.4 ⫾ 0.7 mV, n ⫽ 4) (A). Incubation in HEPESbuffered Tyrodes solution significantly reduced the extent of hyperpolarization seen with 1⫻10⫺5 M acetylcholine (14.5 ⫾ 0.5 mV, n ⫽ 4 [P ⬍ 0.05]) (B), whilst responses to NS1619 and levcromakalim remained unaffected.

Bicarbonate-buffered Krebs t =0

t = 24h

Bicarbonate-buffered Krebs + 1 × 10–2M HEPES t =0 t = 24h

C×37

C×40

C×43

Figure 31.6 Western blots of Cx37, Cx40 and Cx43 before and after 24 h incubation of rat mesenteric arteries in either bicarbonate-buffered or bicarbonate plus HEPES-buffered solutions. Incubation in normal bicarbonate-buffered Krebs had no effect on connexin levels compared to controls. However, 24 h incubation in Krebs containing 1⫻10⫺2 M HEPES selectively reduced levels of Cx40.

4. DISCUSSION

4.1. HEPES and EDHF-induced hyperpolarizations The data presented demonstrate that HEPES-buffering is generally unsuitable for the study of EDHF responses, especially if blood vessels are to be studied for long periods. After overnight incubation in bicarbonate-buffered Krebs solution, both porcine coronary and rat

246

M.J. Gardener et al.

mesenteric arteries displayed hyperpolarizations attributed to the release of EDHF that were indistinguishable from those observed in freshly isolated vessels (Figure 31.1). However upon incubation in HEPES-buffered Tyrode solution, porcine coronary arteries had little or no detectable EDHF response, whilst control responses to various K⫹ channel openers remained intact. Hyperpolarizations due to release of EDHF were also inhibited by the addition of 1⫻10⫺2 M HEPES to bicarbonate-buffered Krebs solution, and thus the lack of EDHF responses can be attributed to the presence of HEPES itself. Recordings from the endothelial-cell layer of porcine coronary arteries showed that overnight incubation in HEPES-buffered Tyrode solution did not inhibit hyperpolarizations induced by substance P. Similarly, those induced by bradykinin were unaffected. This demonstrates that HEPES acts ‘downstream’ of the endothelium suggesting that its action could be exerted on gap junctions. Membrane potential recordings from smooth muscle of mesenteric arteries incubated overnight in HEPES-buffered Tyrode solution showed that although EDHF-evoked hyperpolarizations were not abolished, they were significantly attenuated (by approximately 30%) when compared to those seen in preparations incubated in bicarbonate-buffered Krebs solution.

4.2. HEPES and myoendothelial gap junctions A reasonable explanation for the observed actions of HEPES is that this buffer inhibits myoendothelial gap junctions, an action of HEPES and other taurine-based buffers already described in another biological system (Bevans and Harris, 1999). In the porcine coronary artery, gap junctions may play a critical role in the conduction of EDHF-induced hyperpolarization from the endothelium to the underlying smooth muscle (Edwards et al., 2000). To a lesser extent, gap junctions also seem to conduct some of the hyperpolarizing signal in rat mesenteric arteries (Edwards et al., 1999). This would therefore explain the differences between the magnitude of HEPES-induced inhibition of EDHF responses in these vessels. Incubation in a salt solution containing HEPES for 24 h reduced the amount of Cx 40 but had no effect on levels of Cx37 or Cx43. This is consistent with the reduced acetylcholine-induced vasodilatations seen in Cx40 knockout (de Wit et al., 2000). Furthermore, recordings of EDHF-induced hyperpolarizations in the porcine coronary artery (Edwards et al., 2000) showed that these could be inhibited by 43Gap27 (a peptide inhibitor of Cx43 and Cx37 but not Cx40) if electrodes were placed in the outermost layer of smooth muscle. No inhibition was seen in recordings taken from the layer of smooth muscle immediately below the endothelium indicating that the myoendothelial gap junctions in this vessel are likely to comprise primarily Cx40. This conclusion is supported by the finding that 43 Gap27 has no effect on EDHF-mediated renal vasodilatation whereas 40Gap27 (an inhibitor of Cx40) abolished the response (De Vriese et al., 2002). It thus appears that the critical connexin for EDHF conduction from the endothelium to the smooth muscle (at least in the porcine coronary artery) is Cx40 and that HEPES selectively ablates this protein.

4.3. Bradykinin-induced hyperpolarization: two mechanisms In the porcine coronary artery, hyperpolarizations evoked by substance P were abolished by incubation in HEPES-buffered Tyrode solution whereas those induced by bradykinin were attenuated but not completely inhibited. Application of iberiotoxin subsequently abolished the residual bradykinin-induced hyperpolarization. A widely accepted identifying feature of EDHF is that hyperpolarizations attributed to it can be abolished by the addition of apamin,

Effect of HEPES on EDHF responses

247

(a selective blocker of the small conductance calcium-sensitive K⫹ channel (SKCa) ) plus charybdotoxin, (a inhibitor of the intermediate conductance calcium-sensitive K⫹ channel (IKCa) ). In addition to this, iberiotoxin, a selective blocker of BKCa is unable to substitute for charybdotoxin (which inhibits BKCa and IKCa as well). From this, it seems reasonable that bradykinin, as well as liberating EDHF, hyperpolarizes the smooth muscle via a mechanism that is iberiotoxin-sensitive and likely involves smooth muscle BKCa. Substance P, however, only liberates EDHF and hence responses are completely inhibited by incubation in a HEPES-containing salt solution. In the porcine coronary artery, hyperpolarizations evoked by bradykinin and attributed to EDHF are sensitive to blockade by iberiotoxin alone (Fisslthaler et al., 2000). From this present study it would appear that this observation was the result of HEPES-buffering. This inhibited the bradykinin-induced EDHF-mediated component of hyperpolarization and unmasked the residual iberiotoxin-sensitive mechanism which contributes to the total bradykinin-induced hyperpolarization of the smooth muscle.

5. CONCLUSIONS This study suggests that HEPES-buffering is not ideal for the study of EDHF. The HEPES molecule directly inhibits gap junctions comprising Cx26 (Bevans and Harris, 1999) and the present experiments indicate an additional inhibitory action on Cx40. In addition substance P appears to be the agonist of choice when studying EDHF in the porcine coronary artery since bradykinin, as well as activating the EDHF pathway, also hyperpolarizes the smooth muscle of the porcine coronary artery by an, as yet, undefined mechanism that is sensitive to iberiotoxin.

32 Quantification of the amount of potassium released by cultured porcine coronary endothelial cells, stimulated by bradykinin Jean-Louis Bény, Alexander Schuster, Maud Frieden, Monica Sollini and Anne Baron Edwards et al. (1998) hypothesized that the endothelium-derived hyperpolarizing factor (EDHF) is potassium released by the endothelial cells as a consequence of their hyperpolarization when stimulated by endothelium-dependent vasodilators. A recurrent question is to know whether enough potassium is released to cause hyperpolarization of the adjacent smooth muscle cells. Potassium current was measured by the patch-clamp technique during the application of bradykinin to cultured endothelial cells derived from the porcine coronary artery to determine the amount of released potassium. The whole cell configuration showed that 100 ⫻ 10⫺6 M bradykinin evokes a large, transient outward potassium current. The addition of iberiotoxin and charybdotoxin inhibited this current by 97%. The cell-attached mode confirms that bradykinin activates a large conductance and a small conductance calcium-dependent potassium channel. The inside-out configuration showed half-maximal activation by calcium of both channels. Cytosolic free calcium was determined by fluorescence calcium imaging using fura 2 as a probe. These results allowed to model the time course over potassium membrane current during a maximal transient hyperpolarization of the endothelial cells. The integration on time of this current gives the charge. This charge is carried by 7.7 ⫻ 10⫺14 mol of K⫹. The volume under an endothelial cell where K⫹ is diluted is 5280 ⫻ 10⫺15 l. If half the K⫹ is released in this space, this leads to a concentration of 7.5 mM. This is into the range of concentrations compatible with a relaxing effect of potassium on smooth muscle cells.

In porcine coronary arteries, bradykinin relaxes the smooth muscles in an endotheliumdependent manner by releasing nitric oxide from the endothelium and by triggering the phenomenon known as EDHF-mediated response (Bény and Brunet, 1988; Félétou and Vanhoutte, 1988; Pacicca et al., 1992). During the endothelium-dependent relaxation caused by bradykinin, the membrane of both endothelial and the underlying smooth muscle cells hyperpolarizes simultaneously in the same manner (Brunet and Bény, 1989). Edwards et al. hypothesized that the endothelium-derived hyperpolarizing factor (EDHF) could be potassium released by the endothelial cells as a consequence of their hyperpolarization when stimulated by endothelium-dependent vasodilators (Edwards et al., 1998). A recurrent question is to know whether the endothelial cells release enough potassium to cause hyperpolarization of the smooth muscle cells. The aim of the present study was to determine the amount of potassium ions released by endothelial cells during a maximal transient hyperpolarization of the endothelial cells produced by bradykinin. With this aim in view, potassium currents were measured by the patch-clamp technique during the application of bradykinin to cultured endothelial cells derived from the porcine coronary artery. The knowledge of the time course of outward potassium current allowed to calculate the charge, and therefore the amount of potassium released by the endothelial cells during their transient hyperpolarization.

Quantification of potassium released by cultured endothelial cells

249

1. METHODS

1.1. Primary culture of endothelial cells Left anterior descending branches of pig coronary arteries were obtained at the local slaughterhouse. The endothelial cells were collected by gentle rubbing of the intimal face of the artery with a scalpel blade. The cells were suspended in culture medium (M199) and plated on collagen-coated glass coverslips in Petri dishes. Cells were used after 2–5 days in primary culture (Baron et al., 1996).

1.2. Whole cell patch-clamp recording To measure the currents activated by bradykinin, the whole cell patch-clamp technique was used. (Hamill et al., 1981). Recordings were performed on single cells, or small islets never exceeding four cells, to avoid space clamp problems. To determine the current-potential relationship, repetitive 300 ms voltage pulses were applied throughout the recording, usually reaching 30, 50 and 80 mV above the holding potential (varying between ⫺60 and ⫺20 mV). To normalize the results, the current conductance was expressed in density (pS/pF). Thus, membrane capacitance was measured before each experiment by applying a 10 mV voltage step. Experiments were performed at room temperature (20–22 ⬚C) (Frieden et al., 1999).

1.3. Single channel recordings Single channel recordings were obtained in cell-attached and inside-out configurations (Hamill et al., 1981). During the experiments, the patches containing the channel activated by bradykinin were selected. Experiments were performed at room temperature (20 – 22⬚) (Baron et al., 1996). The standard pipette solution contained (mM): KCl 5.6, NaCl 130, and Hepes 10 (pH ⫽ 7.45 adjusted with NaOH 1 N). The standard bathing solution contained (mM): KCl 130, MgCl2 1, CaCl2 2, Hepes 8, glucose 10 (pH ⫽ 7.5 adjusted with KOH 10 N). The Ca2⫹-free bathing solution was obtained by omission of MgCl2 and CaCl2 and addition of 2 mM EGTA. The calcium-dependence of the channel was determined using solutions containing between 100 nM and 100 ␮M of CaCl2 buffered with 5 mM EGTA. Solutions were exchanged using a custom-made solution exchanger. Recording of current amplitude was performed maintaining the holding potential during 10 sweeps of 1 s each in a range of values between ⫺80 and ⫹40 mV. Channel events detection and open time duration histograms were performed using the MacTAC software (Instrutech Corp, and SKALAR Instr Inc, Seattle, WA, USA).

1.4. The modeling The electrophysiological results show that the reversal potential is very close to the K⫹ equilibrium potential. Only the K⫹ currents seem to be important. At rest, the reversal potential has shifted slightly to a less hyperpolarized state, thus indicating the implication of other ions in this “residual” current. The K⫹ current consists of two distinct contributions: the BKCa, a large conductance channel activated by Ca2⫹ and membrane potential, and an apamin-insensitive, charybdotoxin sensitive small conductance channel (SKCa), activated only by Ca2⫹. IK[t] ⫽ Gtot . (Vm[t] ⫺ EK ) . (0.4 . PoBKCa[Ca[t],Vm[t]] ⫹ 0.6 . PoSKCa[Ca[t]])

(1)

where Gtot is the total potassium channel conductivity, EK is the K⫹ equilibrium potential, PoBKCa is the open-state probability of the BKCa and PoSKCa is the open-state probability of

250

Jean-Louis Bény et al.

the SKCa. The total K⫹ current at maximal activation, consists of two contributions: 40% of the current passes through the BKCa, and 60% through the apamin-insensitive, charybdotoxin sensitive SKCa. This conclusion is derived from the experimental current–voltage curves, where a selective blockade of the BKCa with iberiotoxin decreases the channel conductivity by 40% over the entire range of physiological membrane potentials (see Section 2.2.). Ca[t] as a concentration of cytosolic free calcium in function of time during the stimulation by bradykinin as determined by “fura 2 imaging” (Frieden et al., 1999). The cation channel contribution to the current is being neglected here, as there is also a Na⫹ current through the cation channel, of similar amplitude, but in opposite direction. The net effect of these two currents is supposed to be negligible. Furthermore, the contribution of the Ca2⫹ influx has not been included in the total current (IK[t] ⫹ IR[t]), as it is negligible compared to the potassium current. The “residual” current, can be described by the equation: IR[t] ⫽ GR . (Vm[t] ⫺ Vrest)

(2)

where GR is the residual current conductivity, and Vrest is the membrane resting potential. Note that the open-state probability of the BKCa channel depends both upon the intracellular Ca2⫹ concentration, and the membrane potential. The experimental data available give just the PoBKCa either as a function of Ca2⫹ for different membrane potentials, or as a function of Vm, for different Ca2⫹ concentrations. Plots taking all of this into account, give a twodimensional fit of the PoBKCa. The link between the two variables is made for the [Ca2⫹] EC50 values of the PoBKCa, which gives an exponential fit as a function of Vm. The resting membrane potential dynamics are then simply given by: dVm/dt ⫽ 1/C . (IK[t] ⫹ IR[t])

(3)

where C is the membrane capacitance.

1.5. Drugs The following agents were used: apamin (Alomone Labs, Jerusalem Israel), 1,2-bis (2-aminophenoxy)ethane-N,N,N⬘,N⬘-tetraacetic acid (BAPTA Sigma, St Louis, MO, USA), bradykinin (Bachem Feinchemikalien, AG, Budendorf, Switzerland), charybdotoxin (Alomone Labs, Jerusalem Israel), EGTA, (Sigma, St Louis, MO, USA), iberiotoxin (Alomone Labs, Jerusalem Israel), 17-octadecynoic acid (17-ODYA BAPTA Sigma, St Louis, MO, USA, nitro-L-arginine (Aldrich, Steinheim Germany).

1.6. Statistical analysis Data were calculated as the mean ⫾ standard error of the mean (SEM). Student’s t-test for unpaired observations was used to compare results. A P value less than 0.05 was taken as significant; n refers to the number of observations. 2. RESULTS

2.1. Bradykinin activates a K⫹ current The stimulation of the cultured endothelial cells with bradykinin (10⫺7 M) induced an outward current that developed rapidly and then slowly decreased. The conductance of the

Quantification of potassium released by cultured endothelial cells

251

activated currents measured on different cells was expressed as a function of the cell capacitance to normalize the results. The membrane capacitance was 26 pF for one cell, 44 pF for two and 84 pF for three coupled cells (Frieden et al., 1999). The maximal conductance induced by the effect of bradykinin was 268 pS/pF. With 5.6 ⫻ 10⫺3 M extracellular K⫹, the current reversal potential was ⫺74 mV. This value is close to the K⫹ equilibrium potential of ⫺80 mV. With 4 ⫻ 10⫺2 M K⫹, the current reversal potential was ⫺27 mV for a calculated K⫹ equilibrium potential of ⫺30 mV (Frieden et al., 1999). Without intra- and extracellular Ca2⫹ (2 ⫻ 10⫺3 M EGTA in the bath and 5 ⫻ 10⫺3 M BAPTA in the patch pipette), no currents were elicited by bradykinin.

2.2. Contribution of different potassium channels to the total conductance To determine the contribution of the different KCa channels activated by bradykinin, different KCa channel inhibitors were tested on the maximal conductance induced by the peptide. The current slope conductance averaged 280 pS/pF. It was not significantly affected by apamin (10⫺6 M) a blocker of SKCa. Iberiotoxin (5 ⫻ 10⫺8 M), a blocker of BKCa, significantly reduced (by 40 ⫾ 16%, n ⫽ 8) the K⫹ current activated by bradykinin (164 pS/pF). Together, apamin plus iberiotoxin did not produce a greater inhibition than iberiotoxin alone (154 pS/pF). The two toxins did not shift the current reversal potential of the currents activated by bradykinin (Frieden et al., 1999). Charybdotoxin (5 ⫻ 10⫺8 M), an inhibitor of intermediate conductance IKCa reduced the conductance of the bradykinin-stimulated current to 51 pS/pF, corresponding to an inhibition of 75 ⫾ 8% (n ⫽ 9) because in this series of experiments, the control was 202 pS/pF. The addition of charybdotoxin (5 ⫻ 10⫺8 M) plus iberiotoxin (5 ⫻ 10⫺8 M) decreased the conductance of the bradykinin-induced current to 6 ⫾ 6 pS/pF (Sollini et al., 2002).

2.3. Characterization of the channels responsible for the conductance induced by bradykinin 2.3.1. The apamin insensitive, charybdotoxin sensitive SKCa In the cell-attached configuration, 10⫺7 M bradykinin activated a 10 pS K⫹ channel. In the inside-out configuration, the channel was half-maximally activated by 795 nM free Ca2⫹. Therefore this channel is activated by the [Ca2⫹]i reached during the activation by bradykinin. Apamin (10⫺6 M) added to the pipette solution failed to inhibit the channel activity while charybdotoxin (5 ⫻ 10⫺8 M), completely blocked it. Taken together, these results show that bradykinin activates a 10 pS KCa channel, which contributes to 75 ⫾ 8% (n ⫽ 9) of the total K⫹ current activated by these agonists. Despite its small conductance, this channel shares pharmacological characteristics with intermediate conductance KCa channels (Sollini et al., 2002). 2.3.2. The BKCa In the cell-attached configuration, bradykinin activated a high-conductance K⫹ channel (285 pS in high-symmetrical potassium). Its open state probability was increased by depolarization. This channel was half-maximally activated by 4.5 ␮M Ca2⫹ at ⫺30 mV (Baron et al., 1996). This channel would thus never be activated since the [Ca2⫹]i only reaches up to

252

Jean-Louis Bény et al.

1.3 ␮M during stimulation by bradykinin (Frieden et al., 1999). However, in the whole-cell mode, the inhibition of this channel showed that it contributes to 40 ⫾ 16% (n ⫽ 8) of the maximal K⫹ current activated by bradykinin (Frieden et al., 1999). Three observations explain this paradox: First, cytochrome P450 derived compounds are involved in the bradykinin-induced response. Concentration–response curves were obtained on coronary strips with intact endothelium contracted with prostaglandin F2␣, (10⫺5 M) in the presence of indomethacin (10⫺5 M) and nitro-L-arginine (10⫺4 M). A concentration of 3 ⫻ 10⫺6 M, 17ODYA, a suicide substrate of cytochrome P450 enzyme caused a significant rightward shift of the concentration–response curve to bradykinin, with an EC50 of 1.5 ⫻ 10⫺8 M compared to 3 ⫻ 10⫺9 M under control conditions. It decreased the maximal relaxation to bradykinin by about 20 ⫾ 5% (n ⫽ 9). This compound has no effect on the endothelium-dependent relaxations to substance P which is insensitive to indomethacin and nitro-L-arginine (Frieden et al., 1999); Second, the effect of 17-ODYA (3 ⫻ 10⫺6 M) was tested on the bradykinininduced K⫹ current in the whole cell mode. In the presence of the inhibitor, the maximal current slope conductance due to bradykinin was significantly inhibited by 35⫾14% (n⫽11) (179 pS/pF). On the contrary, 17-ODYA did not affect the response to substance P (Frieden et al., 1999). Third, products of cytochrome P450 enzyme, the four epoxyeicosatrienoic acid regioisomers (1.6 ⫻ 10⫺7 M) applied on the cytosolic side of the membrane in insideout patches, transiently increased the open state probability of the BKCa in the presence of 5 ⫻ 10⫺7 M Ca2⫹ (Baron et al., 1997). Altogether, these results show that the epoxyeicosatrienoic acids potentiate endothelial BKCa to Ca2⫹ and that this pathway is activated by bradykinin. However, it is not activated by substance P. In other words, BKCa contribute to 35–40% of the maximal conductance induced by the effect of bradykinin but are not implicated in the substance P stimulation. The reason being that bradykinin but not substance P triggers the epoxyeicosatrienoic acids cascade. 2.3.3. The non-selective cationic channel In the cell-attached mode, bradykinin (10⫺7 M) activated a non-selective cation channel of 44 pS in high symmetrical K⫹. The relative permeability of this channel was PK : PNa : PCa ⫽ 1 : 1 : 0.7. In the inside-out configuration, the channel was half-maximally activated by 700 nM free Ca2⫹ on the intracellular side of the membrane (Baron et al., 1996). Therefore, this channel is activated by the time course of the membrane potential reached during the activation by bradykinin.

2.4. The time course of potassium current The experimental data give the open-state probability Po of the BKCa channel as a function of intracellular Ca2⫹ for two different holding potentials and as function of membrane potential for three different intracellular Ca2⫹ concentrations. These data have been fitted using a hyperbolic tangent fit, x being the variable to be fitted (e.g. Ca2⫹ concentration or membrane potential): Po ⫽ –12 (1 ⫹ tan h[(x ⫺ m3)/m4])

(4)

where m3 and m4 are parameters determined by the fit (m3 corresponds to the half-activation, and m4 to the “spread” of the curve ). Extracting from these graphs the Ca2⫹EC50 value (open-state probability 50%) as a function of membrane potential yields an exponential relationship between the Ca2⫹EC50 value and Vm. Taken together, this allows constructing a

Quantification of potassium released by cultured endothelial cells

253

Figure 32.1 Three-dimensional fit of the BKCa open-state probability (Po) as a function of both cytosolic free calcium (expressed as log [Ca2⫹]i) and membrane potential (Vm). The black dots are experimental measurements.

three-dimensional fit of the BKCa open-state probability as a function of both intracellular Ca2⫹ and Vm (Figure 32.1). The equation of this fit (5) for the PoBKCa is then used in equation (1) in order to calculate the K⫹ current PoBKCa ⫽ 0.5(1 ⫹ tan h[((log[Ca2⫹]⫺c) (Vm⫺b)⫺a)/ (m3 . (Vm⫹a . (log[Ca2⫹]⫺c)⫺b)2⫹m4])

(5)

where a ⫽ 53.3, b ⫽ ⫺80.8, c ⫽ ⫺6.4, m3 ⫽ 1.32 ⫻ 10⫺3 and m4 ⫽ 0.30. The fitted values yield a half-activation Ca2⫹ concentration of 3944 nM at ⫺30 mV, and of 1543 nM at ⫹10 mV. However, these values were obtained in the absence of epoxyeicosatrienoic acids, which strongly decrease these half-activation values (approximately by a factor of 30). Hence, in order to take into account the effect of epoxyeicosatrienoic acids, the PoBKCa was shifted by a factor of 30 as a function of Ca2⫹ concentration in the construction of the current time courses. The half-activation membrane potentials resulting from these fits are ⫺66.6, ⫺23.6, and ⫹31.4 mV for a Ca2⫹ concentration of 2 mM, 3 ␮M, and 1 ␮M respectively. The apamin-insensitive SKCa channel open-state probability depends only upon the intracellular Ca2⫹ concentration, the fit of the experimental data yielding a half-activation concentration for Ca2⫹ of 529 nM (fitted value). There is no dependence of the open-state probability of this channel upon membrane potential. The resulting individual membrane current contributions of the SKCa, BKCa and residual current upon bradykinin stimulation, were predicted by the model (Figure 32.2). The sum of all these currents, the total current, produced the membrane potential hyperpolarization (Figure 32.3).

2.5. Calculation of the K⫹ released The concentration of K⫹ increase in the internal elastic membrane during a 20 mV hyperpolarization of an endothelial cell stimulated by 10⫺7 M bradykinin was calculated on the basis of Figure 32.2. The integral over time of this current is 739 ⫻ 10⫺11 C. This charge is carried by 739 ⫻ 10⫺11 C divided by 96516 C mol⫺1, that is, 7.7 ⫻ 10⫺14 mol of K⫹. To examine whether this amount of K⫹ is compatible with the content of the endothelial cell in K⫹, the proportion of release was calculated. The surface of a patched cultured endothelial cell

254 Jean-Louis Bény et al.

Figure 32.2 Individual membrane current (I) contributions upon bradykinin stimulation as predicted by the model. Solid line: K⫹ current passing through the BKCa channel; dashed line: K⫹ current passing through the SKCa channel; dotted line: residual current.

Figure 32.3 Total current (ITOT) upon bradykinin stimulation as predicted by the model.

averages 880⫾100␮m2. The thickness is about 3␮m. Therefore, the volume is 3␮m⫻ 880␮m2 ⫽2640 ␮m3 ⫽ 2640 ⫻ 10⫺15 l. With an intracellular concentration of K⫹ of 130 mM, the content of the cell is 2640 ⫻ 10⫺15 l ⫻ 130 ⫻ 10⫺3 mol/l ⫽ 34 ⫻ 10⫺14 mol of K⫹. Therefore, an endothelial cell releases about 1/5 to 1/4 of its potassium content during a 20 mV transient hyperpolarization. To approximate the volume in which the released K⫹ is diluted, it was hypothesized that it is diluted in a volume delimited by the surface of the endothelial cells and the thickness of the internal elastic membrane. The surface of an endothelial cell is 880 ␮m2 and the internal elastic membrane in the porcine coronary artery is about 6␮m. Therefore, the volume under an endothelial cell where K⫹ is diluted is 880␮m2 ⫻ 6 ␮m ⫽ 5280 ␮m3 ⫽ 5280 ⫻ 10⫺15 l. If half of the K⫹ (only through the abluminal face) is released into this space, this leads to a concentration increase of 3.85 ⫻ 10⫺14 mol of K⫹ divided by 5280 ⫻ 10⫺15 l ⫽ 7.5 ⫻ 10⫺3 M.

3. DISCUSSION The calculation of the amount of K⫹ released does not arise from speculations, but it is based on [Ca2⫹]i, current and conductance measurements. The model predicts

Quantification of potassium released by cultured endothelial cells

255

current–potential curves and time course of the membrane potential that perfectly match the experimental data. Therefore, this calculated amount is reliable. However, it reflects the amount of K⫹ released by a cultured endothelial cell. Indeed, these cells in primary culture are different from native one. They grow on a thin layer of collagen on glasses, they are larger and they are not stable since they lost the response to substance P after about four days of culture. Two alternative approaches are (a) acutely isolated cells and (b) intact cells in tissue (open arterial wall). The acutely dispersed cells are at least as questionable as the cultured cells. Enzymes that can change the external receptors disperse them. They also undergo mechanical stress. The native cells can only be patched on their luminal face and it could be that the channels are only present on the basal face. Therefore, each approach presents different artifacts. The purpose of the present study was to determine if the order of magnitude of potassium released by endothelial cells would be compatible with the concentration of K⫹ that relaxes the vascular smooth muscle cells. This concentration in hepatic artery from the rat is from five to twenty 10⫺3 M (Edwards et al., 1998). Therefore, the endothelial cells release enough K⫹ during a transient hyperpolarization caused by bradykinin to relax vascular smooth muscle cells. This fulfills a necessary but not sufficient condition to prove that K⫹ can be EDHF. At least, this removes the controversy concerning the amount of released K⫹ by a stimulated endothelial cell.

4. CONCLUSION The order of magnitude of the calculated concentration of potassium reached in the vicinity of smooth muscles during bradykinin stimulation is compatible with the hypothesis that K⫹ could be EDHF in certain blood vessels (Edwards et al., 1998).

33 The intensity of agonist-stimulation influences the mechanism for relaxation in rat mesenteric arteries K.A. Dora, L. McEvoy, M. King, N. Ings and C.J. Garland The rat mesenteric artery was one of the vessels in which it was first proposed that K⫹ might act as an endothelium-derived hyperpolarizing factor (EDHF). The ability of K⫹ to evoke hyperpolarization and relaxation of smooth muscle cells was reduced during increasing contraction to the ␣1-agonist phenylephrine, at a time when acetylcholine still evoked hyperpolarization of the smooth muscle cells. This suggested that myoendothelial gap junctions might also facilitate acetylcholine-mediated hyperpolarization. In the present study, gap junction blockers were used to determine if the functional importance of the gap junctions increases with stimulation intensity, to explain the persistent EDHF-mediated response to acetylcholine. Third-order branches of rat superior mesenteric artery were mounted in a Mulvany–Halpern wire myograph. Variable levels of contraction were stimulated with 10⫺7 to 10⫺5 M phenylephrine during blockade of nitric oxide synthase. With low levels of contraction (close to 10 mN), the gap junction uncoupler carbenoxolone (10⫺4 M) had minimal effects on relaxation to acetylcholine, levcromakalim or exogenous K⫹ (10.8 mM). With higher levels of contraction, relaxation to acetylcholine was not reduced in the presence of an inhibitor of nitric oxide (NO) synthase, but was completely inhibited by the gap junction inhibitor. These data show that as the functional contribution of K⫹ to the EDHF pathway reduces with arterial contraction, the persistent ability of acetylcholine to evoke EDHF-mediated relaxation reflects a contribution of increasing importance from heterocellular gap junctions.

1. INTRODUCTION The rat mesenteric artery was one of the vessels in which it was first proposed that K⫹ may act as an EDHF, although at that time another mechanism was proposed to contribute to the EDHF-mediated response in this artery. It was suggested that myoendothelial gap junctions may underlie the latter, the functional importance of which might increase as the artery contracted (Edwards et al., 1998). The ability of K⫹ to evoke hyperpolarization and relaxation of smooth muscle cells, and thus act as an EDHF, is reduced during increasing contraction to the ␣1-adrenergic agonist phenylephrine (Dora and Garland, 2001). However, when the stimulation is sufficient to block hyperpolarizations and relaxations to exogenous K⫹, acetylcholine can still evoke an EDHF-mediated response. In the present study, the gap junction blocker carbenoxolone was used to determine if the functional importance of gap junctions predominates as the intensity of stimulation increases, and as such explains the persistent EDHF-mediated response evoked with acetylcholine.

Agonist-stimulation influences the mechanism for relaxation

257

2. METHODS

2.1. Wire myography Third-order branches of rat superior mesenteric artery were mounted in a Mulvany–Halpern wire myograph (Danish Myotechnology, 400A) at a tension equivalent to 0.9 times the diameter of the vessel at 100 mmHg. Endothelium viability was taken as ⬎95% control relaxation to 10⫺6 M acetylcholine. Variable levels of contraction were obtained with 10⫺7 to 10⫺5 M phenylephrine. Arteries were incubated with the gap junction uncoupler carbenoxolone (10⫺4 M), and/or the NO synthase inhibitor, N␻-nitro-L-arginine methyl ester, (L-NAME, 10⫺4 M) for at least 30 min before stimulation. Prostacyclin does not play a significant role in this artery (Garland and McPherson, 1992). 3. MATERIALS Arteries were bathed in oxygenated Krebs buffer of the following composition (mM): NaCl 118.0, NaHCO3 25.0, KCl 3.6, MgSO4.7H2O 1.2, KH2PO4 1.2, glucose 11.0 and CaCl2 2.5 continuously aerated with 95% O2 and 5% CO2. All drugs used were from Sigma Chemical Company, Poole UK. 4. RESULTS In the presence of L-NAME (10⫺4 M), acetylcholine (10⫺6 M) evoked a maximal relaxation of 95 ⫾ 1.6% (n ⫽ 8) and 89 ⫾ 2.5% (n ⫽ 8), in arteries contracted by phenylephrine to low (7–10 mN) and high (17–25 mN) levels of tension, respectively (Figures 33.1 and 33.2). In the additional presence of carbenoxolone (10⫺4 M), relaxation to acetylcholine persisted in arteries contracted to low levels of tension, while increasing the intensity of stimulation with phenylephrine was now associated with significant inhibition of relaxation (Figures 33.1 and 33.2). Contraction to phenylephrine per se was unaffected by carbenoxolone (data not shown). At low levels of tension, raising the bath concentration of K⫹ to 10.8 mM relaxed arteries by 92.6 ⫾ 2.9%. In contrast, in arteries almost maximally contracted with phenylephrine, only 14.6 ⫾ 6.8% relaxation was stimulated by 10.8 mM K⫹. At all levels of tension, carbenoxolone did not alter the ability of K⫹ to evoke relaxation (data not shown). 5. DISCUSSION These data show that in rat mesenteric arteries, as the contribution of K⫹ to the EDHF pathway diminishes with arterial contraction, the persistent ability of acetylcholine to evoke an EDHF-mediated relaxation can be explained by a relative increase in the functional importance of heterocellular gap junctions. In arteries contracted with the ␣1-adrenergic agonist phenylephrine, raising the extracellular K⫹ concentration from 4.8 mM evokes hyperpolarizations and relaxations of the smooth muscle. This reflects the activation of Na⫹/K⫹-ATPase and KIR, until concentrations of around 20 mM K⫹, when depolarization and contraction predominate (Edwards et al., 1998; Dora and Garland, 2000). In the rat mesenteric artery, functional KIR activity appears to be restricted to the endothelium, so direct stimulation of the smooth muscle with K⫹ to evoke hyperpolarization and relaxation occurs by stimulation of the Na⫹/K⫹-ATPase (Dora and Garland, 2001).

L-NAME

10 Contraction (mN)

L-NAME + Carbenoxolone

3 × 10–810–7

10–7 3 × 10–7 10–6

3 × 10–7

3 × 10–6

10–6 3 × 10–6

5

5 × 10–7 M PE

3 × 10–7 M PE 0

2 min

2 min

25

3 × 10–7 –6 10

3 × 10–8 10–7

3 × 10–8 10–7

Contraction (mN)

20 15 3 × 10–7 10–6 3 × 10–6

10 5 0

10–5 M PE

2 min

10–5 M PE

2 min

Figure 33.1 Representative traces showing EDHF-mediated relaxation to acetylcholine in the presence of L-NAME at low (top) and high (bottom) levels of tension stimulated by phenylephrine (PE). A blocking effect with carbenoxolone was only apparent in the latter. Dots indicate addition of acetylcholine (M) to the bath.

L-NAME L-NAME + Carbenoxolone

0

0

*

Relaxation (%)

*

50

50

100

100 10–9

10–8

10–7

10–6

[ACh] (M)

10–5

10–9

10–8

10–7

10–6

10–5

[ACh] (M)

Figure 33.2 Relaxation evoked by acetylcholine in the presence of L-NAME (10⫺4 M) at low tension (n ⫽ 8, left panel) and high tension (n ⫽ 8, right panel), showing the inhibitory influence of carbenoxolone (10⫺4 M). Values are means ⫾ SEM. Analysis by Kruskal–Wallis test. The asterisks indicate a statistically significant (P ⬍ 0.05) effect of carbenoxolone.

Agonist-stimulation influences the mechanism for relaxation

259

At the lower end of the concentration range, up to around 12 mM K⫹, the ability to evoke hyperpolarization and relaxation is dramatically influenced by the intensity of ongoing stimulation of the smooth muscle. Once depolarization and contraction exceed around ⫺40mV and 10 mN, respectively, the inhibitory effects of K⫹ are effectively abolished (Dora and Garland, 2001). As the stimulation of smooth muscle cells with constrictor agonists is associated with the activation of Ca2⫹-activated K channels, increasing stimulation intensities will be associated with an increase in the efflux of K⫹ through these channels. The magnitude of the efflux is able to raise the local extracellular concentration of K⫹ sufficiently to block the subsequent action of exogenous K⫹ (Dora et al., 2002). However, it is also clear that in this artery another pathway exists to enable hyperpolarization of the smooth muscle in response to endothelial cell stimulation with acetylcholine (Edwards et al., 1998). As there is morphological evidence for myoendothelial gap junctions in this artery (Sandow and Hill, 2000), these connections could potentially allow the spread of hyperpolarizing current from the activated endothelium to the adjacent smooth muscle cells (Edwards et al., 1998, 1999). This pathway would then be predicted to act in parallel to extracellular K⫹, ensuring that maximal EDHF-mediated relaxation was still possible under conditions where K⫹ was unable to contribute to relaxation (Doughty et al., 2000; Lacy et al., 2000; Dora and Garland, 2001). This possibility was addressed by comparing the effects of uncoupling gap junctions at the different levels of tension. To enable meaningful studies on the functional importance of myoendothelial gap junctions, it is of primary importance to establish that agents such as carbenoxolone can inhibit myoendothelial gap junctions, in addition to homocellular gap junctions, and if it is possible to differentiate between these two effects. It is also important that the hyperpolarization of the endothelial cells evoked by acetylcholine (and other endothelium-dependent vasodilators) is not depressed, as this would indirectly modify the EDHF-mediated response. Carbenoxolone was used in preference to either 18␣- or 18␤glycyrrhetinic acid, because the non-selective effects appear to be less (Terasawa et al., 1992; Tare et al., 2002). In the present study, carbenoxolone did not inhibit contractions to phenylephrine, whereas other uncouplers, including both 18␣- and 18␤-glycyrrhetinic acid and palmitoleic acid did. Further, at low levels of contraction, carbenoxolone did not inhibit the ability of either acetylcholine or exogenous K⫹ to evoke relaxation. This suggests that carbenoxolone did not alter the ability of acetylcholine to stimulate a rise in endothelial cell [Ca2⫹]i, which is important as both the formation of NO and EDHF are Ca2⫹-dependent processes. These data are consistent with one previous report (Edwards et al., 1999), but not another (Tare et al., 2002), in the same artery. Furthermore, they indicate that carbenoxolone did not inhibit KIR channels and/or Na⫹/K⫹-ATPase, both of which are stimulated by small increments in extracellular [K⫹] (Edwards et al., 1998). At high levels of stimulation with phenylephrine, when the ability of K⫹ to contribute to EDHF-mediated relaxation is reduced, carbenoxolone markedly reduced the ability of acetylcholine to cause endothelium-dependent hyperpolarizations. This strongly suggests that gap junctions play an important role in transferring a hyperpolarizing signal from the endothelium to smooth muscle cells, which at high levels of contraction provides the predominant route to relaxation. This route is also presumably acting in parallel to K⫹ diffusion with the functional importance of the latter masked at high levels of contraction (Figure 33.3). The physiological consequences of these data relate to both the activation of the endothelium and the smooth muscle cells. The importance of endothelium-dependent hyperpolarization

260

K.A. Dora et al.

Figure 33.3 Schematic depicting the pathways for endothelium-dependent hyperpolarization of smooth muscle in small mesenteric arteries of the rat. In response to stimulation with acetylcholine, K⫹ efflux from endothelial cells can activate two pathways leading to the net movement of positive change (dashed arrow) from smooth muscle cells via (1) the Na⫹/K⫹-ATPase and (2) gap junctions. Pathway 1 is activated by phenylephrine, thereby reducing further activation by the efflux of K⫹ from endothelial cells, and pathway 2 is uncoupled by carbenoxolone.

in relaxation increases as artery size decreases. At least in part, this reflects the importance of voltage gated Ca2⫹entry in the contraction of smooth muscle cells in resistance arteries. Thus, endothelial and smooth muscle cell hyperpolarization are of direct physiological relevance in controlling the diameter of vessels intimately involved in the control of blood pressure and flow.

34 Small and intermediate conductance Ca2⫹-activated K⫹ channels (SKCa and IKCa) in porcine coronary endothelium: relevance to EDHF R. Bychkov, M.P. Burnham, G.R. Richards, C. Thollon, G. Edwards, A.H. Weston, P.M. Vanhoutte and M. Félétou EDHF-mediated responses are sensitive to the combination of apamin plus charybdotoxin. The purpose of this study was to characterize the targets of these two toxins in porcine coronary artery. Freshly isolated endothelial cells were studied with the patch-clamp technique. Outward potassium currents were inhibited significantly by iberiotoxin only at very positive potentials (⬎ 50 mV). In contrast, charybdotoxin and apamin partially inhibited these currents over the whole range of test potentials. Single-channel recordings revealed K⫹ conductances of 6.8 and 17 pS. The open probability of each underlying channel was increased by Ca2⫹ and reduced by apamin or charybdotoxin, respectively. Iberiotoxin was without effect. Substance P and bradykinin each activated an iberiotoxin-insensitive outward current which was blocked partially by apamin and charybdotoxin and abolished by their combination. In contrast, l-ethyl-2-benzimidazolinone (1-EBIO) activated mainly a charybdotoxin-sensitive component. Messenger RNA encoding the SK2 and SK3, but not the SK1, subunits of SKCa was detected by RT-PCR in samples of endothelium. Western blot and immunofluorescence labeling confirmed that these protein were highly expressed in the plasmalemma of endothelial cells. Membrane potential was recorded from the endothelium and smooth muscle cells of isolated arteries. 1-EBIO, bradykinin and substance P each produced endothelial hyperpolarizations and endothelium-dependent hyperpolarizations of the smooth muscle. These were unaffected by iberiotoxin, but abolished by charybdotoxin alone for 1-EBIO and apamin plus charybdotoxin for substance P and bradykinin. The present study shows that endothelial cells express both SKCa (most likely the SK3 subunit) and IKCa and that EDHF-mediated hyperpolarization of smooth muscle cells requires the activation of these two potassium channels to elicit endothelial cell hyperpolarization. Additionally, BKCa are poorly expressed in freshly isolated endothelial cells from porcine coronary arteries and are unlikely to play a prominent role at physiological membrane potentials.

1. INTRODUCTION Endothelium-dependent hyperpolarizations are sensitive to a combination of the two toxins, apamin and charybdotoxin. The former is a specific blocker of small conductance calciumactivated potassium channels (SKCa) while the latter is a non-specific blocker of intermediate conductance (IKCa) and large conductance calcium-activated potassium channels (BKCa) as well as some populations of voltage-dependent potassium channels. However, the combination of iberiotoxin, a specific blocker of BKCa, and apamin is ineffective, indicating that BKCa channels are not involved in EDHF-mediated responses (Corriu et al., 1996a; Zygmunt and Högestätt, 1996; Petersson et al., 1997; Chataigneau et al., 1998; Edwards et al., 1998, 2000).

262

R. Bychkov et al.

The combination of the two toxins (apamin and charybdotoxin) blocks EDHF-mediated responses if selectively applied to the endothelium (Doughty et al., 1999), and inhibits the hyperpolarization of the endothelial cells produced by acetylcholine or bradykinin (Edwards et al., 1998, 2000; Ohashi et al., 1999) suggesting that the site of action of apamin and charybdodotoxin is the endothelium (inhibition of endothelial hyperpolarization) and not the smooth muscle cells [inhibition of the action of endothelium-derived hyperpolarizing factor (EDHF)]. The purpose of this work was to characterize the charybdotoxin- and apamin-sensitive channels in porcine coronary endothelial cells and to further evaluate the role of these channels in the endothelium-dependent hyperpolarization of the porcine coronary artery. 2. MATERIAL AND METHODS

2.1. Tissue dissection Pig hearts were taken from Large-White pigs (male or female, 25 kg) anaesthetized with a mixture of tiletamine plus zolazepam (i/m., each 10 mg/kg) or obtained from the local abattoir and transported to the laboratory in ice-cold Krebs solution ( (mM), NaCl 118, KCl 3.4, CaCl2 2.5, KH2PO4 1.2, MgSO4 1.2, NaHCO3 25, glucose 11, gassed with 5% CO2 in O2). Left anterior descending coronary arteries were dissected free of surrounding tissue while maintained in ice-cold Krebs solution.

2.2. Patch-clamp experiments Endothelial cells were isolated mechanically and placed on cover-slips pretreated with Matrigel® (basement membrane matrix, Becton Dickinson). K⫹ currents were measured at room temperature (20–24 ⬚C) using whole-cell, outside-out or perforated-patch configurations of the patch-clamp technique. The external solution contained: (mM) NaCl 140, CaCl2 1.8, MgCl2 1, KCl 5.4, and Na-HEPES 10 (pH 7.4). Recording pipettes (3–8 M⍀ resistance) were filled with a solution containing: (mM) K-aspartate 80, KCl 40, NaCl 20, MgCl2 1, Mg-ATP 3, EGTA 10, K-HEPES 5 (pH 7.4). EGTA was titrated with Ca2⫹ to provide solutions containing 100, 200 and 500 nM free Ca2⫹. For perforated-patch experiments, whole-cell access was achieved within 2–5 min by the inclusion of nystatin (50–100 ␮g/ml) in the pipette solution. Patch–clamp data were recorded at 5–10 kHz using an Axopatch 200B amplifier (Axon Instruments) with compensation for series resistance and cell capacitance. Data were filtered at 1 kHz, digitized using an Axon interface and data analysis was performed using pClamp-6 software (Axon Instruments). Variance analysis and histogram distribution were analysed to determine the amplitude of single K⫹ channels. Both methods were compared and provided similar results. Mean current (I) and variance (␴ 2) were calculated for the variance analysis. Least-squared parabolic fits to the resulting values were used to derive estimates of single channels properties according to the equation ␴ 2 ⫽ Ii ⫺ (I 2/N), where N is the number of channels in the patch and i the current amplitude through a single channel.

2.3. Microelectrode studies Segments of intact vessel were opened longitudinally, pinned to the Sylgard base of a heated 10 ml bath and superfused with Krebs solution (10 ml/min, 37 ⬚C, gassed with 5% CO2 in O2)

SKCa and IKCa in porcine coronary endothelium

263

containing 300 ␮M NG-nitro-L-arginine and 10 ␮M indomethacin. Endothelial cells were impaled using micro-electrodes filled with 3 M KCl (resistance 40–80 M⍀). Successful impalements were signalled by a sudden change in membrane potential which remained stable for at least 2 min before the experiment was commenced. Substance P and 1-EBIO were each added as bolus injections directly into the bath in quantities calculated to give (transiently) the final concentrations indicated. Charybdotoxin and apamin were added to the reservoir of Krebs superfusing the bath. Only recordings from impalements that were maintained successfully for the entire protocol were analyzed. Recordings were obtained using a conventional high impedance amplifier (Intra 767, WPI Instruments) and digitized for analysis using a MacLab system (AD Instruments). Interference of 50 Hz at the amplifier output was removed selectively using an active processing circuit (Humbug, Digitimer).

2.4. Gene-specific RT-PCR Samples of endothelium were obtained as luminal scrapings. Pig brain tissue was taken from the hippocampal area. Total RNA was isolated using RNeasy Mini Prep kits (Qiagen) and, following DNase treatment (Gibco Life Technologies), RNA was reversetranscribed using Sensiscript (Qiagen), all according to manufacturers’ instructions. Forward and reverse primers for RT-PCR (35 cycles, 64 ⬚C annealing temperature and 1.5 mM Mg2⫹) were as follows: GAPDH, 5⬘-GACCACTTCGTCAAGCTCATTTCC and 5⬘-GATGGTACATGACGAGGCAGGTC; SK1, 5⬘-CCCACATAGTCATGAACAGCCACAG and 5⬘-GGATCTCCCGGGCATGGTAGAGG; SK2, 5⬘-CCCCGAGATCGTGGTGTCTAAGC and 5⬘-CACACACCAGTATTTCCAAGCAGATG; SK3, 5⬘-TCCATGTTTT CGTTGGCCCTG and 5⬘-AGCATGACTCGGGCGATCAGGTA. Products were resolved by electrophoresis on 1.5% (w/v) agarose gels and the identity of each was confirmed by cloning and sequencing (Big Dye Chemistry). The SK3 cDNA was sequenced using the additional primers 5⬘-GACACTTCTG GGCACTTCCATGA and 5⬘-CAACTGCTTGAACTTGTGTATGG to amplify overlapping segments of the coding region. 5⬘ and 3⬘ RACE (Rapid Amplification of cDNA Ends) was performed (GeneRacer kit, Qiagen) to obtain the complete cDNA sequence. Each product was sequenced in triplicate.

2.5. Western blotting Whole arteries were homogenized on ice using ground-glass homogenizers in: (mM) Tris pH 7.5 20, sucrose 250, EDTA 5, EGTA 10, protease inhibitor (Sigma P2714). Nuclear and post-nuclear fractions were produced by centrifuging for 5 min at 12,000 rpm in a bench-top centrifuge. Samples of endothelium were obtained by scraping the luminal surface with a clean scalpel blade. Protein estimation was performed with the method of Bradford (1976). Sample preparation for sodium dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) on 10% (w/v) acrylamide separating gels and electrophoretic transfer to polyvinylidene difluoride membrane was performed. Membranes were blocked overnight in 50 mg/ml non-fat dried milk in tween–tris-buffered saline (Tween–TBS; 1 ␮l/ml Tween-20, 20 mM Tris pH 8.0, 150 mM NaCl). Anti-SK2 and SK3 primary antibodies were used with and without control antigen pre-incubation (incubated for 1 h with 3-fold excess by weight of control peptide) and diluted to 1 ␮g/ml in Tween-TBS containing 10 mg/ml bovine serum albumin (BSA). Membranes were incubated with primary antibody for 1 h. Detection was

264

R. Bychkov et al.

achieved using horseradish peroxidase-conjugated secondary antibodies and chemiluminescent reagents.

2.6. Immunofluorescence histochemistry Arteries were fixed for 20 min, cryoprotected overnight in 0.3 g/ml sucrose in phosphatebuffered saline (PBS), and embedded in OCT® compound. Cryostat sections (thickness 8 ␮m) were collected on silanated microscope slides. To label intact endothelium viewed en-face, small fragments of artery were first pinned flat, fixed and washed in PBS. Both preparations were treated with 1 mg/ml SDS in PBS for 30 min, washed in PBS and blocked for 1 h or overnight with blocking buffer (50 ␮l ml⫺1 normal goat serum, 10 mg/ml BSA in PBS). Primary antibodies were prepared with and without control peptide pre-incubation, diluted to 6 ␮g/ml in blocking buffer and applied for 1 h. In some cases, negative controls were performed by omission of primary antibody. Secondary antibodies conjugated with Cy3 or fluorescein iosthiocyanate (FITC) were applied for 30 min, with either 4, 6-diamidino2-phenylindole (DAPI; 5 ␮g/ml final) or propidium iodide (1 ␮g/ml final) included as nuclear labels. Sections were viewed on a Zeiss Axioplan 2 microscope equipped with Hamanatsu CCD camera and Zeiss KS300 software. Identical microscope, camera and software settings were used when imaging labelled sections and negative controls.

2.7. Materials Anti-SK2 and anti-SK3 (Alomone Labs), supplied with control peptides, anti-von Willebrand’s factor (Novocastra) and secondary antibody conjugates (Jackson Immunoresearch) were used. Apamin, synthetic charybdotoxin and iberiotoxin were obtained from Latoxan, France and Matrigel® was obtained from Becton Dickinson (MA, USA). OCT® compound was purchased from R.A. Lamb (UK). All other substances were supplied by Sigma, except 1-EBIO (l-ethyl-2-benzimidazolinone; Aldrich), levcromakalim (SmithKline Beecham) and substance P (RBI).

2.8. Data analysis All values are given as mean ⫾ standard error (SEM). The number of tested cells or arteries from individual animals is given by n. Statistical analysis was carried out using the paired and unpaired Student’s t test and a value of P ⬍ 0.05 was considered to be statistically significant. 3. RESULTS

3.1. Membrane potential of the endothelial cells in the isolated porcine coronary artery Under control conditions, the membrane potential of the endothelial cells was ⫺49.1⫾0.4mV. Both 100 nM substance P and 600 ␮M 1-EBIO each induced robust hyperpolarizations (substance P, 27.8 ⫾ 0.8 mV; 1-EBIO, 24.1 ⫾ 1.0 mV; each n ⫽ 4). The addition of 100 nM charybdotoxin abolished the response to 1-EBIO but only slightly reduced the hyperpolarization to substance P (25.8 ⫾ 0.3 mV, n ⫽ 4). Further addition of 100 nM apamin to the superfusing solution abolished the response to substance P. Neither the hyperpolarization

SKCa and IKCa in porcine coronary endothelium

265

Figure 34.1 Membrane potential responses in porcine coronary artery endothelium to 100 nM substance P, 600 ␮M 1-EBIO and 10 ␮M levcromakalim (LK) were recorded using sharp micro-electrodes. The addition of 100 nM charybdotoxin (ChTX) and 100 nM apamin to the bathing solution is indicated (modified from Burnham et al., 2002).

to 10 ␮M levcromakalim nor the resting membrane potential was affected by either of these toxins (Figure 34.1).

3.1. Patch-clamp experiments 3.2.1. Global currents Endothelial cells were studied with the whole-cell configuration of the patch-clamp technique. Iberiotoxin (100 nM) produced no significant inhibition at negative and physiological potentials. However, at positive potential (above ⫹50 mV) iberiotoxin produced a significant inhibition of the outward K⫹-current (data not shown). Charybdotoxin (100 nM) inhibited the outward K⫹-current over the whole range of potentials. Cumulative application of charybdotoxin plus apamin (100 nM) produced a further decrease of the K⫹-current (Figure 34.2). 3.2.2. Single channel recordings With the outside-out patch configuration, single-channel recordings were performed with either 100, 250 and 500 nM free Ca2⫹ pipette solution in asymmetric K⫹ concentrations (internal solution: 120mM K⫹ and external solution: 5.4mM K⫹). Under these experimental conditions, at least three different potassium channels were observed in the excised patches. The unitary conductance of the first channel, determined by plotting unitary current amplitude against membrane potential, was 17.1 ⫾ 0.4 pS (n ⫽ 9; under asymmetric K⫹ concentration, Figure 34.3). The open probability of the 17 pS conductance was dependent on the free Ca2⫹ concentration and increased from 0.17 ⫾ 0.02 (n ⫽ 8) to 0.38 ⫾ 0.04 (n ⫽ 8) and to 0.65 ⫾ 0.03 (n ⫽ 7), in the presence of 100, 250 and 500 nM Ca2⫹, respectively (Figure 34.3). In the presence of 250 nM Ca2⫹, the addition of apamin (100 nM) did not significantly reduce the open probability of the channel (from 0.42 ⫾ 0.03 to 0.44 ⫾ 0.02, n ⫽ 6, in the absence and presence of apamin, respectively). Similarly, in the presence of iberiotoxin (100 nM) the open probability of the channel was not affected. In contrast, the addition of charybdotoxin (100 nM) significantly decreased the open probability of the channel (Figure 34.4).

Figure 34.2 Effect of iberiotoxin, charybdotoxin and apamin (100 nM each) on the outward K⫹ currents in freshly isolated endothelial cells from porcine coronary artery (whole-cell configuration). (Top) Representative K⫹-currents elicited by 10 mV voltage steps in control, after application of iberiotoxin (IbTX), iberiotoxin ⫹ apamin and iberiotoxin ⫹ apamin⫹charybdotoxin (IbTX⫹ChTX⫹apamin). (Bottom) Summary graph: the presence of apamin and charybdotoxin produced a statistically significant inhibition of the amplitude of the K⫹-currents (modified from Bychkov et al., 2002).

Figure 34.3 Intermediate conductance in freshly isolated endothelial cells from porcine coronary artery (outside-out configuration). Single-channel activity recorded with the outside-out configuration in asymmetrical K⫹ gradients with Ca2⫹ fixed at 250 nM in the patchpipette solution. Corresponding voltages are indicated in these panels. The closed-state current is indicated by the broken line (top left panel) The amplitude of single-channel currents was plotted against voltage and fitted with a linear function. The slope conductance obtained from the fit was 17 pS (top right panel). Ca2⫹-dependency of the K⫹-single channels recorded with the outside-out patch configuration (bottom panels). The Ca2⫹ concentration in the patch pipette is indicated above the recording. The amplitudes of the histograms of distribution were fitted with double or triple Gaussian functions (modified from Bychkov et al., 2002).

SKCa and IKCa in porcine coronary endothelium

267

Figure 34.4 Intermediate conductance in freshly isolated endothelial cells from porcine coronary artery (outside-out configuration). Effect of KCa-channel blockers, iberiotoxin and charybdotoxin (100 nM, each) on the single-channel activity recorded in the outside-out patch configuration. The amplitudes of the histograms of distributions were fitted with a double Gaussian function (modified from Bychkov et al., 2002).

In the presence of charybdotoxin (100 nM) and with 250 nM free Ca2⫹ pipette solution and asymmetric K⫹ concentrations, two other channels were observed with unitary conductances of 6.8 ⫾ 0.04 and 2.7 ⫾ 0.03 pS (n ⫽ 9 and 7, respectively). The 6.8 pS conductance showed essentially no voltage-dependency, with only a slight increase in open probability over the tested voltage range. In the absence of charybdotoxin, the open probability of this channel increased from 0.20 ⫾ 0.03 (n ⫽ 8) to 0.52 ⫾ 0.05 (n ⫽ 8) and to 0.77 ⫾ 0.06 (n ⫽ 7) for 100, 250 and 500 nM Ca2⫹, respectively (Figure 34.5). With 250 nM Ca2⫹ in the pipette, the addition of 100 nM apamin significantly reduced the open probability of the channel (Figure 34.5). In the presence of 100 nM charybdotoxin (with 250 nM Ca2⫹ in the pipette), the open probability of the 6.8 pS channel (0.35 ⫾ 0.05, n ⫽ 6) was only marginally lower than that of controls recorded in this set of experiments (0.52 ⫾ 0.05, n ⫽ 8). The 2.7 pS conductance was voltage-insensitive but was observed infrequently (7 out of 38 patches) and therefore was not characterized further. 3.2.3. Effect of substance P, bradykinin and 1-EBIO on outward K⫹-currents Substance P (100nM) did not affect the K⫹-current in the endothelial cell dialysed with 10mM EGTA (n ⫽ 5, data not shown). With the perforated-patch configuration, outward potassium currents were elicited by voltage steps and endothelial cells were held at a holding potential of ⫺ 80 mV. Substance P produced a transient and rapid increase in the net outward current followed by a steady-state activation. Iberiotoxin (100 nM) did not significantly affect the increase in current produced by substance P while charybdotoxin (100 nM) or apamin (100 nM) significantly inhibited the effects of substance P (Figure 34.6).

Figure 34.5 Small conductance calcium-activated potassium channels in freshly isolated endothelial cells from porcine coronary artery (outside-out configuration). Single-channel activity recorded with the outside-out configuration in asymmetrical K⫹ gradients with Ca2⫹ fixed at 250 nM in the patch pipette solution in the presence of charybdotoxin (100 nM). Corresponding voltages are indicated in panel. The closed state of the current is indicated by the broken line (top left panel). The amplitude of the single channel current was plotted against voltage and fitted with a linear function. The slope conductance obtained from the fit was 6.8 pS (top right panel). Apamin inhibited the 6.8 pS channel activity (bottom panels). Data obtained in the outside-out patch configuration under a physiological K⫹ gradient (in the absence of charybdotoxin). Distribution of unitary conductance amplitude recorded at 0 mV together with representative traces: (data were fitted with a double Gaussian function; modified from Burnham et al., 2002).

Figure 34.6 Substance P (SP, 100 nM) and K⫹ currents in freshly isolated endothelial cells from porcine coronary artery (perforated-patch configuration). Effect of the potassium channel inhibitors charybdotoxin (ChTX), iberiotoxin (IbTX) and apamin (100 nM, each) on the outward K⫹-currents elicited by voltage steps with increments of 10 mV from a potential of ⫺120 mV. Endothelial cells were held at a holding potential of ⫺80 mV. Bar graphs represents the increase of the K⫹-current amplitude induced by SP and the inhibition produced by toxins. Data are shown as means ⫾ SEM and are expressed in percentage of the control (modified from Burnham et al., 2002). (A) Charybdotoxin; (B) Iberiotoxin; (C) Apamin.

SKCa and IKCa in porcine coronary endothelium

269

Figure 34.7 Effect of bradykinin and 1-EBIO on K⫹-currents in freshly isolated endothelial cells from porcine coronary artery (perforated-patch configuration). Left panel: K⫹-currents elicited by voltage steps to 50 mV from a holding potential of ⫺80 mV were recorded in control conditions, 10 s after the application of bradykinin 100 nM in control conditions, or in presence of iberiotoxin 100 nM (IbTX), or in presence of iberiotoxin plus apamin, each 100 nM (IbTX ⫹ apamin) or in presence of iberiotoxin plus apamin plus charybdotoxin 100 nM (IbTX ⫹ apamin ⫹ ChTX). Right panel: K⫹-currents, elicited by voltage steps to 50 mV from a holding potential of ⫺80 mV, were recorded in control conditions, after the application of 1-EBIO (300 ␮M) in control conditions, or in presence of iberiotoxin 100 nM (EBIO ⫹ IbTX), or in the presence iberiotoxin plus apamin 100 nM (EBIO ⫹ IbTX ⫹ apamin), or in presence of iberiotoxin plus apamin plus charybdotoxin 100 nM (EBIO ⫹ IbTX ⫹ apamin ⫹ ChTX). The increase in leak current that appeared after the last application of the ‘cocktail’ of K⫹-channels blockers was subtracted from the original traces (modified from Bychkov et al., 2002).

The effects of bradykinin were qualitatively similar to those of substance P. Voltage steps elicited K⫹-currents that were increased by bradykinin. This current was not affected significantly by iberiotoxin and was reduced significantly by the subsequent addition of apamin and nearly completely abolished by the further addition of charybdotoxin (Figure 34.7). The addition of 1-EBIO (300 ␮M) induced an increase in the outward current that was not significantly affected by the presence of iberiotoxin (100 nM). The addition of apamin (100 nM) produced a small but statistically significant inhibition of this current which was abolished by the subsequent addition of charybdotoxin (100 nM) (Figure 34.7).

3.3. SKCa subunits 3.3.1. SKCa subunit mRNA Samples of cDNA prepared from the endothelium (n ⫽ 5) contained SK3 (detected in 5 out of 5 samples, Figure 34.8) and SK2 (3/5 samples) transcripts (data not shown). No SK1 transcripts were detected (data not shown). The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) housekeeping gene, included as a positive control, was detected in all cases.

270

R. Bychkov et al.

Figure 34.8 Small conductance calcium-activated potassium channels subunit in porcine coronary artery endothelium. Left panel: RT-PCR analysis of SK3 expression in samples of endothelium prepared with (2) and without (3) reverse transcription. 1: Control lane without template. Middle panel: Localization of SK3 protein in porcine coronary arteries by Western blotting. Samples were prepared from endothelium (Endothel) only or whole artery. Primary antibodies were used with (⫹) and without (⫺) control peptide preincubation. Molecular weight markers are indicated (kDa). 10 ␮g protein loadings. Right panel: Localization of SK3 protein in porcine coronary arteries by immunofluorescence labeling. (B) cross-section labeled for SK3 (red) and nuclei (blue), with internal elastic laminae (green) visible (inset, primary antibody pre-incubated with control peptide). (D) arteries opened longitudinally allowing the endothelium to be viewed en-face were labeled for SK3 (green) and nuclei (orange) (inset, primary antibody omitted). All scale bars are 25 ␮m (modified from Burnham et al., 2002) (see Color Plate 11).

3.3.2. Identification of SK2 and SK3 protein by Western blotting The SKCa protein content of endothelium only versus whole arteries was examined by Western blotting (n ⫽ 2). Samples of whole artery were prepared as nuclear and post-nuclear fractions. Anti-SK3 immunoreactivity was absent from whole artery fractions but present in endothelial samples (Figure 34.8, 10 ␮g protein each). The ~74 kDa SK3 band migrated close to the predicted molecular weight of hSK3 of 81 kDa, while a related epitope was present at ~134 kDa. Strong anti-SK2 immunoreactivity was observed at ~69 kDa, close to the predicted molecular weight of hSK2 of 64 kDa. This reactivity was present in the nuclear, but absent from the post-nuclear, fractions of whole artery. The reactivity was also observed in samples of the endothelium (data not shown). 3.3.3. Immunofluorescence histochemistry The endothelium was positively identified in cryostat sections by anti-von Willebrandt’s factor reactivity (data not shown). In arterial sections, intense SK3 labeling was observed in

SKCa and IKCa in porcine coronary endothelium

271

the endothelium (Figure 34.8, n ⫽ 4), with reactivity in the smooth muscle limited to the occasional cell. Labeling was completely absent when the primary antibody was pre-incubated with control peptide (Figure 34.8, Inset). The endothelium was also viewed en-face, with nuclei labeled red using propridium iodide and SK3 reactivity detected with green-fluorescent (FITC) secondary antibodies. In agreement with the observations from cryostat sections, intense SK3 labeling was visible throughout the endothelium (Figure 34.8, n ⫽ 2). SK3 appeared localized to the periphery of each endothelial cell, near or within the plasmalemma. No labeling was observed when the primary antibody was omitted (Figure 34.8, Inset). SK2 immunoreactivity in artery sections was present in both endothelial and smooth muscle cells (data not shown). When viewed en-face or in sections, SK2 labeling was restricted to the area immediately surrounding the nuclear staining (data not shown, n ⫽ 6). 4. DISCUSSION The present study shows that endothelial cells express both SKCa and IKCa, the respective targets of apamin and charybdotoxin (Burnham et al., 2002). The localization of the charybdotoxin/apamin-sensitive potassium channels to the endothelial cells strongly indicates that a critical step in generating the EDHF-mediated hyperpolarization of smooth muscle cells is the activation of SKCa and IKCa to elicit hyperpolarization of the endothelial cell. In other words, hyperpolarization of the latter is a prerequisite for the endotheliumdependent hyperpolarization of the underlying vascular smooth muscle cells.

4.1. Endothelial IKCa and SKCa In freshly non-enzymatically isolated endothelial cells from porcine coronary artery, four types of calcium-activated potassium channels (two SKCa, IKCa and possibly BKCa) were observed. BKCa channels (100–250 pS) were first reported in cultured bovine aortic endothelial cells (Fichtner et al., 1987) and have been described extensively in endothelial cells of porcine coronary in primary culture (Baron et al., 1996) and cultures of a derived endothelial cell line from human umbilical vein (Haburcak et al., 1997). They are inhibited by nanomolar concentration of the non-specific inhibitor charybdotoxin (Miller et al., 1985) and by the selective inhibitor iberiotoxin (Galvez et al., 1990; Bychkov et al., 1999). Surprisingly, in the present study, BKCa-channels were not found in outside-out patches or in the cell-attached configuration. However, iberiotoxin blocked a non-linear outward current at very positive potentials (⬎ ⫹80 mV). This observation strongly suggests that this channel is expressed but is not activated at physiological potentials. Additionally, in some endothelial cells such as those freshly isolated from the mesenteric arteries taken from healthy humans, there is virtually no expression of BKCa (Köhler et al., 2000). The expression (or the activity) of BKCa could depend on the duration of the culture of the endothelial cell (Kestler et al., 1998). Altogether, the results obtained in the present and earlier studies do not argue for a prominent role, if any, of BKCa in freshly isolated endothelial cells. In the present study the expression of a channel with all the characteristics of IKCa was observed. The current is weakly voltage-sensitive, calcium dependent with a conductance (under physiological, asymmetrical K⫹ conditions) of 17pS and was blocked by charybdotoxin but unaffected by iberiotoxin (Ishii et al., 1997b; Jensen et al., 1998; Vergara et al., 1998). 1-EBIO is an activator of calcium-activated potassium channels which opens IKCa and possibly SKCa but not BKCa (Devor et al., 1996; Edwards et al., 1999b; Cao et al., 2001).

272

R. Bychkov et al.

1-EBIO produced a marked increase in the outward current that was inhibited by charybdotoxin but not iberiotoxin, a pharmacological confirmation that the current recorded transited via IKCa. Apamin produced also a small but significant inhibition of the current in the presence of 1-EBIO. However, it is not known at present whether this effect of apamin indicates some activation of SKCa by 1-EBIO (Cao et al., 2001) or the inhibition of a SKCa current already activated under control conditions (Burnham et al., 2002). In the presence of charybdotoxin, another Ca2⫹-sensitive K⫹ channel with a unitary conductance 6.8 pS (under physiological, asymmetrical K⫹ conditions) was detected. The current was virtually voltage-insensitive, blocked by apamin and resistant to charybdotoxin, which are the hallmarks of a SKCa (Marchenko and Sage, 1996; Vergara et al., 1998; Castle, 1999; Frieden et al., 1999). Additionally, a 2.7 pS current was observed in a small number of membrane patches but was difficult to study because of its small conductance and contamination by other channels in the same patch. This current was observed only in the presence of charybdotoxin and not in the presence of apamin, and was blocked by d-tubocurarine (Burnham et al., 2002). Therefore, this channel most likely represents an additional SKCa The few studies performed in intact or freshly isolated endothelial cells suggest that low-conductance Ca2⫹-activated K⫹ channels are physiologically active with a channel of 9 pS conductance identified in rabbit aortic cells (Sakai, 1990) and conductances of 6.7 and 2.8 pS in rat aortic cells (Marchenko and Sage, 1996). The blockade of the endothelial SKCa by 100nM apamin could suggest the presence of an SKCa channel comprising SK2 or SK3 ␣-subunits as some studies have shown that the SK1 subtype is not, or very poorly, sensitive to apamin (Grunnet et al., 2001a). However, the human SK1 channel cloned and re-expressed in various cell types is sensitive to apamin in the nanomolar range (Shah and Haylett, 2000; Strøbæk et al., 2000). Furthermore, expression of SK1–SK2 dimers results in channels that display an apamin-sensitivity greater than SK1 homomultimeric channels (Ishii et al., 1997a). Using only electrophysiological techniques, it is thus difficult to conclude exactly which subunit underlies the observed 6.8 pS and 2.7 pS channels. The SK3 subunit was highly expressed in the endothelium of the porcine coronary artery. SK2 was also detected but there was little evidence to indicate the presence of SK1. At the mRNA level, SK2 and SK3 but not SK1 transcripts were detected in samples of the endothelium. The fact that SK3 transcripts were detected in all cases while SK2 was detected in 3 out of 5 samples suggests that SK mRNA may be relatively unstable and that SK3 mRNA is the more abundant or resistant to degradation. The SK3 protein was detected easily in samples prepared from the endothelium as opposed to those from whole arteries. Finally, SK3 staining was abundant only in the endothelium and not in the smooth muscle cells. Furthermore, when viewed en-face, SK3 fluorescence was observed at the perimeter of each endothelial cell, suggesting a plasmalemmal location which would be consistent with a functional role in mediating endothelial hyperpolarizations. In contrast, SK2 reactivity was found in both endothelium and smooth muscle, and in both situations at an intracellular location surrounding the nuclear staining. To confirm this, whole artery samples for Western blot analysis were fractionated into nuclear and post-nuclear fractions. Strong SK2 immunoreactivity was present in the nuclear but not in the post-nuclear fraction, consistent with an intracellular localization and confirming the immunofluorescence results (Burnham et al., 2002). Collectively, these results strongly suggest that the endothelial cells of the porcine coronary artery express a plasmalemmal-located, apamin-sensitive SKCa channel which contains an SK3 subunit. The effects of substance P and bradykinin were qualitatively similar as they both produced an increase in the outward currents. These effects were partially blocked by

SKCa and IKCa in porcine coronary endothelium

273

charybdotoxin, virtually unaffected by iberiotoxin, partially blocked by apamin and abolished by the combination of charybdotoxin plus apamin. Therefore, these results indicate that substance P and bradykinin activates both IKCa and SKCa in porcine coronary endothelial cells. The activation of SKCa by substance P has previously been reported in cultured endothelial cells from the same tissue (Frieden et al., 1999). However, in the present study the effects of bradykinin, in contrast to a previous report do not involve the activation of BKCa (Baron et al., 1996; Frieden et al., 1999). This might be explained by the differences in experimental conditions as the present study was performed with freshly non-enzymatically isolated endothelial cells while the previous investigations involved endothelial cells in primary culture which may affect the expression or the activity of BKCa (Kestler et al., 1998). The activation of IKCa and SKCa produces the hyperpolarization of the endothelial cells. The measurement of membrane potential with intracellular electrodes shows that substance P and 1-EBIO hyperpolarize the endothelial cells. As shown in the present study, 1-EBIO opens charybdotoxin-sensitive IKCa channels in vascular endothelial cells but is without effect on vascular smooth muscle cells (Edwards et al., 1999b). 1-EBIO can also activate apamin-sensitive channels formed by hSK1, rSK2 or rSK3 subunits expressed in Xenopus oocytes and mammalian HEK293 cells (Syme et al., 2000; Cao et al., 2001; Pedarzani et al., 2001; Grunnet et al., 2001b). However, the apamin-sensitive component was small and the charybdotoxin-sensitive component was preponderant. As expected, hyperpolarization induced by substance P was blocked fully by charybdotoxin only in combination with apamin, indicating an activation of both IKCa and SKCa, and confirming observations in freshly isolated rabbit aortic valve endothelial cells (Ohashi et al., 1999). The hyperpolarizing effect of the 1-EBIO was blocked completely by charybdotoxin alone, indicating that the underlying current was carried mainly by IKCa.

4.2 Relevance for EDHF studies Substance P, bradykinin and 1-EBIO each produced a hyperpolarization of endothelial cells but not a direct hyperpolarization of the myocytes (Edwards et al., 2000, 2001). They also produce endothelium-dependent hyperpolarizations of the smooth muscle cells, the effects of the two peptides being inhibited by the combination of charybdotoxin plus apamin (Edwards et al., 2000), while the effect of 1-EBIO was sensitive to charybdotoxin alone (Edwards et al., 1999b; Coleman et al., 2001a). Finally, charybdotoxin and apamin block EDHF-mediated responses (i.e. smooth muscle hyperpolarization) when selectively applied to the endothelium (Doughty et al., 1999). Taken into conjunction, these indicate that the toxin combination targets the endothelial cells (i.e. prevents endothelial cell hyperpolarization) rather than the EDHF-mediated hyperpolarization of the smooth muscle cells and suggest that the hyperpolarization of the endothelial cells is required in order to observe endotheliumdependent hyperpolarization of the underlying smooth muscle cells (Quignard et al., 2000).

35 The role of KCa in endothelial cell hyperpolarization and endothelium-dependent relaxation in the rabbit aorta X. Kuang, L. Chan, W. Liang, I. Laher, C. van Breemen and X. Wang Vasodilatation is modulated by a variety of vasoactive factors released by endothelial cells including nitric oxide (NO), prostaglandins, and an endothelium-dependent hyperpolarizing factor(s) (EDHF). Charybdotoxin and apamin sensitive Ca2⫹ activated K⫹ (KCa) channels play an important role in the activity of EDHF observed after inhibition of the synthesis of NO and prostacyclin. This study characterized acetylcholine-induced, non-NO, non-prostacyclin mediated relaxation in the intact rabbit aorta and investigated the participation of KCa in this response. In addition, using the patch-clamp technique in freshly isolated rabbit aortic endothelial cells, the role of KCa channels in acetylcholine and cyclopiazonic acid-induced membrane hyperpolarization was examined. Inhibition of NO and prostacylin production with N-Nitro-L-Arginine Methyl Ester (L-NAME) and indomethacin respectively reduced acetylcholine-induced vasodilatation. The remaining relaxation was abolished upon addition of both charybdotoxin and apamin, or the Kir blocker BaCl2. The maxi K channel blocker iberiotoxin had no effect on relaxation. Inhibition of gap junctions with heptanol had no effect on the remaining relaxation. In a superfusion bioassay system L-NAME plus indomethacin abolished acetylcholine induced relaxations, suggesting that the non-NO, non-prostacyclin -mediated response was not diffusible or readily diluted. In isolated cells, acetylcholine induced transient membrane hyperpolarization. This transient hyperpolarization was converted to a sustained response following pretreatment with either bradykinin or the sarco/endoplasmic reticulum Ca2⫹ ATPase (SERCA) blocker cyclopiazonic acid. Addition of charybdotoxin or pretreatment with apamin abolished the maintained phase of the hyperpolarization, while the maxi K channel inhibitor iberiotoxin or the opener NS1619 had no effect on membrane potential in these cells. It is concluded from this study that the intermediate conductance (charybdotoxin sensitive) and the small conductance (apamin sensitive) KCa channels mediate acetylcholine-induced endothelial membrane hyperpolarization that is responsible for non-NO, non-prostacyclin mediated vasodilatation (e.g. EDHF-type response) in the rabbit aorta. These data are consistent with the idea that K⫹ functions as an EDHF in this preparation.

1. INTRODUCTION It is now more than 20 years since Furchgott discovered that the endothelium secretes endothelium-derived relaxing factor (EDRF) (Furchgott and Zawadzki, 1980), a key regulator of vascular tone. Soon after this discovery, it was shown that NO (nitric oxide) and prostacyclin are the two major EDRFs. However, in many vessels inhibition of the synthesis of NO and prostacyclin does not completely abolish endothelium-dependent relaxation (De Mey et al., 1982). The putative non-NO/prostacyclin mediator that hyperpolarizes vascular smooth muscle has been termed endothelium- derived hyperpolarizing factor (EDHF).

The role of KCa in rabbit aortic endothelium

275

The intensive search for EDHF has resulted in a vast number of publications (reviewed by Mombouli and Vanhoutte, 1997; Félétou and Vanhoutte, 1999; Triggle et al., 1999). Based on superfusion bioassay data (Rubanyi and Vanhoutte, 1987; Félétou and Vanhoutte, 1988) and the so-called ‘sandwich preparation’ (Dong et al., 1997), the existence of a diffusible EDHF has been suggested, although its chemical nature is still hotly debated. Thus, many substances have been suggested as candidates for EDHF: NO itself (Cohen et al., 1997), L-Citrulline (Ruiz and Tejerina, 1998), anandamide (an endogenous cannabinoid) (Randall et al., 1996), EETs (epoxyeicosatrienoic acids) (Harder et al., 1995), K⫹ (De May et al., 1980; Edwards et al., 1998) and adenosine (Berne, 1980; Malmsjo et al., 1998). In the rat hepatic artery, acetylcholine produces an endothelium-dependent hyperpolarization of smooth muscle, that depends on charybdotoxin/apamin sensitive KCa channels in the endothelial cells, which is blocked by Ba2⫹ and ouabain (Edwards et al., 1998). Acetylcholine also produces an endothelium-dependent increase in extracellular K⫹, suggesting that K⫹ exiting from endothelial KCa channels functions as EDHF (Edwards et al., 1998). The increased extracellular K⫹ concentration is thought to activate Na⫹,K⫹-ATPase and the inward-rectifier K⫹ channels on the smooth muscle cell membrane which will lead to hyperpolarization and relaxation of the vascular smooth muscle. Bi-directional conductance of acetylcholine-mediated vasodilatation occurs in microvessels (Segal and Duling, 1986; Beny and Paccica, 1994). Electron-microscopic imaging revealed the existence of gap junctions between the endothelium and the underlying vascular smooth muscle (Spagnoli et al., 1982). Gap 27, a synthetic peptide presenting the sequence of 11 residues within the extracellular loop of connexin 43, also inhibits EDHF-mediated relaxation (Chaytor et al., 1998). Substantial differences in EDHF responses exist not only between different species, but also between different blood vessels derived from the same species (Triggle et al., 1999). For example, in rabbit vasculature, a small EDHF response occurs in the aorta, while there are prominent EDHF responses in mesenteric and carotid artery (Dong et al., 1997; Triggle et al., 1999). In freshly isolated rabbit aortic endothelial cells, acetylcholine causes membrane hyperpolarization mediated by the activation of Ca2⫹ dependent K⫹ currents (Wang et al., 1996). The focus of the present study was to determine the roles played by small and intermediate conductance Ca2⫹-activated K⫹ channels (KCa) in the process of EDHF-mediated dilatation in the rabbit aorta. 2. METHODS

2.1. Organ bath study Adult New Zealand White rabbits weighing 2.0–2.5 kg were euthanized by CO2 asphyxiation followed by exsanguinations from the carotid arteries. The thoracic aorta was cut into 4-mm wide ring segments and placed in 25 ml organ baths containing normal physiological saline solution, which was maintained at 37 ⬚C and aerated with oxygen. The rings were placed under a resting tension of 2 g and allowed to equilibrate in normal physiological saline solution and then were contracted with phenylephrine (2 ⫻ 10⫺6 M). The vasoactive agents were added directly to the bath to reach the final concentration as indicated. The data are recorded and analyzed using the MacLab system (ADInstruments, Castle Hill, Australia).

2.2. Bioassay studies A perfusion-superfusion bioassay method was used (Rubanyi et al., 1985; Johns et al., 1987). In brief, a segment of the rabbit aorta (donor segment, 3 cm in length) was cleaned

276

X. Kuang et al.

Figure 35.1 Superfusion bioassay set-up. A segment of endothelium-denuded aorta is mounted on a transducer. The effluent from donor, endothelium denuded artery (supply 2) is dripped on the detector ring. Supply 1 with no tissue is used as a control.

of adventitia and then mounted on a polyethylene tube to allow internal perfusion. The tissue was suspended in a glass tube so that the perfusate from inside of the vessel was sufficient to keep the entire tissue moist. The segment was perfused with physiological saline solution. The endothelium was removed from a separate aortic ring (3 mm) by rubbing the luminal surface. This bioassay ring was then suspended below the donor between stationary stainless steel hooks attached to a force displacement transducer and connected to a recorder. A resting tension of 1.5 g was applied to the bioassay ring at the beginning of an experiment. The bioassay ring was superfused with donor effluent (supply 2, Figure 35.1) and/or an additional source of solution which was not in contact with endothelium (supply 1) at a flow rate of 3 mL/min for each solution. The transit time between the distal end of the donor segment and the bioassay ring was less than 5 s. All the solutions were maintained at 37 ⬚C by passing through a spiral tube in a water jacket.

2.3. Isolation of fresh endothelial cells Endothelial cells were dispersed from New Zealand White rabbit (2–2.5 kg) arteries (Wang and van Breemen, 1999). Briefly, rabbits from a local supplier were euthanized with CO2 asphyxiation and exsanguinated. The thoracic aorta was removed and placed in normal physiological saline solution. After careful removal of the surrounding fat and connective tissue the arteries were placed into a test tube containing nominally zero Ca2⫹ solution with 0.1mg/ml Collagenase and 0.1% Elastase. After 35min of enzyme treatment at 37⬚C, arteries were placed in normal physiological saline solution containing 1mg/ml bovine serum albumin. Endothelial cells were dispersed by trituration using a pipette. The final preparation consisted of single cells and small clusters of cells that maintained their typical tile-like morphology. Before experiments, cells were seeded on a glass coverslip coated with poly-D-lysin at room temperature and cells were used within 2–8 h after seeding.

The role of KCa in rabbit aortic endothelium

277

2.4. Electrophysiology on isolated cells The Nystatin perforated whole-cell patch-clamp method was used to study changes in membrane potential (current-clamp mode) (Wang et al., 1996). An EPC-7 patch-clamp amplifier (List-electronic, Darmstadt, Germany) and a compatible computer with PClampsoftware package (Axon-Instrument, Inc., Foster City, CA) were used to generate the command pulse and record data. Continuous data traces were also recorded onto a videotape via a PCM digitizer (Medical system) for later analysis using Axo-Tape software (Axon-Instrument). Patch pipettes were made from borosillicate glass (Clark Electromedical Instruments, Pangbourne Reading, England) with a tip resistance of approximately 4 M⍀. The pipette was first tip-filled with nystatin-free solution and then back-filled with pipette solution containing 24 ⫻ 10⫺8 M nystatin. Electrical contact with the cytosol was established in about 8 min after the giga seal. This was reflected in a decrease in the access resistance below 40 M⍀. To calculate the access resistance, the current trace using a 10 ms voltage pulse of 4 mV (V) was integrated to estimate the total charge (Q). The time constant (T) was estimated by exponential fitting to the declining phase of the current. The access resistance (Ra) was calculated using the equation Cm ⫽ Q/V and Ra ⫽ T/Cm. All solutions were superfused through the experimental chamber and a vacuum suction pump was used to keep a constant fluid level. The time for complete solution exchange (3 ml) was less than 10 s. The membrane potential experiments were carried out at room temperature (23 ⬚C).

2.5. Solutions and chemicals All chemicals were purchased from Sigma Chemicals (St Louis, Mo., USA). Ca2⫹ free solution used for enzyme treatment contained (10⫺3 M) NaCl 126, KCl 5, MgCl2 1.2, HEPES (N-[2Hydroxyethyl]piperazine-N⬘⫺[2-ethanesulfonic acid]) 10, D-glucose 10. Normal physiological saline solution contained (10⫺3 M) NaCl 126, KCl 5, MgCl2 1.2, HEPES 10, D-glucose 10, and CaCl2 1. Zero Ca2⫹ solution contained NaCl 126, KCl 5, MgCl2 1.2, HEPES 10, D-glucose 10, and EGTA 0.1. All solutions were adjusted to pH 7.4.

2.6. Statistics For multiple experiments, mean value and standard error were calculated where applicable. Unpaired Student’s tests were used to determine statistical significance, which was accepted when P was less than 0.05.

3. RESULTS

3.1. Organ bath studies The endothelium intact aortic ring was contracted with 10⫺5 M phenylephrine, an ␣-adrenergic receptor agonist. After tension reached a plateau, 2 ⫻ 10⫺6 M acetylcholine applied into the bath caused partial relaxation (49.3 ⫾ 8.4% of maximal phenylephrine induced response, n ⫽ 8). Treatment of the rings with 2 ⫻ 10⫺5 M indomethacin and 2 ⫻ 10⫺4 M L-NAME to block prostacyclin and nitric oxide production, reduced the acetylcholine induced relaxation to 5.1⫾1.1% of maximal phenylephrine contraction (n⫽8). Although this relaxation was

278

X. Kuang et al.

Figure 35.2 Isometric tension measurement in rabbit aorta. (A) Aortic rings were contracted using 10⫺5 M phenylephrine. 2 ⫻ 10⫺6 M acetylcholine was added at the plateau. After wash-out and recovery same protocol was repeated after treatment with L-NAME (2⫻10⫺5 to 2⫻10⫺4 M) and 2⫻10⫺5 M indomethacin(indo). (B) Summary of isometric tension measurement derived from (A). Bar graph represents maximal value of relaxation obtained. Data shown as means ⫾ SEM. The asterisks indicate that the values after L-NAME and indomethacin are statistically different from the phenylephrine control (P ⬍ 0.05, unpaired t-test).

relatively small, it was significant (P ⬍ 0.05). Therefore, EDHF-type relaxation contributes to about 10% of the total acetylcholine-induced relaxation in the rabbit aorta (Figure 35.2). When arteries were treated with either 2 ⫻ 10⫺6 M bradykinin or 5 ⫻ 10⫺7 M cyclopiazonic acid, the subsequent application of acetylcholine induced partial relaxation (bradykinin plus acetylcholine: 56.7% ⫾ 5; cyclopiazonic acid plus acetylcholine: 34.9% ⫾ 5.2). Bradykinin and cyclopiazonic acid at the concentrations used had no significant effect on arterial tension. Treatment with either bradykinin or cyclopiazonic acid increased the NO/prostacyclin resistant component of acetylcholine-induced relaxation (bradykinin: 11.3 ⫾ 3.6%; cyclopiazonic acid:11.6 ⫾ 3.2% ) (Figure 35.3). The same protocol was repeated in the presence of the endothelial KCa inhibitors charybdotoxin (10⫺7 M) and apamin (3 ⫻ 10⫺5 M). Charybdotoxin and apamin applied together abolished the NO/prostacyclin resistant acetylcholine-induced relaxation (Figure 35.3). The effects of BaCl2, an inhibitor of the inward rectifier K⫹ channel, on acetylcholineinduced relaxation in the rabbit aorta were tested. BaCl2 (10⫺3 M ) abolished the remaining component of the acetylcholine-induced, non-NO/prostacyclin mediated relaxation.

The role of KCa in rabbit aortic endothelium

279

Figure 35.3 Treatment with bradykinin and/or cyclopiazonic acid potentiate acetylcholine induced non-NO/prostacyclin relaxation. The empty bars represent % of relaxation induced by acetylcholine under control condition (left), after blocking NO and prostacyclin production (middle) and in the presence of charybdotoxin (ChTX) and apamin (right). Same protocols were repeated after treatment using bradykinin or cyclopiazonic acid (CPA).

In contrast, iberiotoxin, a large conductance KCa blocker, had no significant effect on the acetylcholine-induced relaxation in the presence of L-NAME and indomethacin (data not shown, n ⫽ 4).

3.2. Bioassay studies The effluent from the endothelium-intact donor vessel caused basal relaxation in the recipient ring. Application of acetylcholine (10⫺5 M) through the donor further relaxed the recipient ring (basal relaxation contributes to 26 ⫾ 12% of maximal acetylcholine-induced relaxation n⫽6), while application of acetylcholine onto the recipient ring had no effect. After treatment of the donor vessel with L-NAME and indomethacin, acetylcholine had no significant effect on tension. L-NAME and indomethacin applied directly to the recipient ring had no effect on tension (Figure 35.4).

3.3. Membrane potential measurements in freshly isolated rabbit aortic endothelial cells The resting membrane potential of freshly isolated endothelial cells had a value of between ⫺30 to ⫺50mV. Application of acetylcholine (10⫺5 M) induced a transient hyperpolarization to ~⫺80 mV, which returned to the resting membrane potential within 100 s (Wang et al., 1996). Pre-treatment with 3 ⫻ 10⫺5 M cyclopiazonic acid potentiated the subsequent acetylcholine-induced hyperpolarization, causing a maintained response to the cholinergic agonist. At the maintained plateau of the membrane potential, application of 10⫺7 M charybdotoxin abolished the hyperpolarization (Figure 35.5). Apamin (3 ⫻ 10⫺5 M) applied after cyclopiazonic acid and acetylcholine had no effect on the membrane potential. However, if the cells were treated with apamin, the subsequent application of cyclopiazonic acid and acetylcholine together only caused a transient hyperpolarization (Figure 35.5). In these cells, cyclopiazonic acid and acetylcholine when applied

Figure 35.4 Bio-assay experiment. Top graph: Representative traces. The recipient ring was contracted using 2 ⫻ 10⫺6 M phenylephrine. Perfusate from the endothelium (EC) intact donor artery was dripped on the recipent ring as indicated by the vertical line under the trace. Acetylcholine (ACh) was then added through the donor artery. After wash out, the same expriment was repeated in the presence of L-NAME and indomethacin (indo). Bottom graph: Summary of the data. The bar graph represents the maximal relaxation caused by each protocol shown above. The basal component after L-NAME and indomethacin was substracted from all data.

Figure 35.5 Membrane hyperpolarization in freshly isolated endothelial cells. Top: Acetylcholine (ACh, 10⫺5 M) and cyclopiazonic acid (CPA, 15 ⫻ 10⫺6 M) were added together to cause maintained hyperpolarization. The Ca2⫹-activated K⫹ channel blockers iberiotoxin (IbTX, 10⫺3 M) and charybdotoxin (ChTX, 10⫺7 M) were added as indicated. Middle: Apamin (3 ⫻ 10⫺5 M) was added after stimulation with acetylcholine (ACh) and cyclopiazonic acid (CPA). Bottom: Apamin (3 ⫻ 10⫺5 M) was added before the stimulation with acetylcholine (ACh) and cyclopiazonic acid (CPA).

The role of KCa in rabbit aortic endothelium

281

together caused hyperpolarization before treatment with apamin. Similar data were obtained using another SK channel blocker, scallytoxin. Neither iberiotoxin nor the BK channel opener NS1619 had any effect on the membrane potential on freshly isolated endothelial cell (data not shown). 4. DISCUSSION The main finding of this study is that the intact rabbit aorta has a component of endotheliumdependent vasodilatation that has the characteristics of an EDHF-mediated response. This response is mediated by Ca2⫹-activated K⫹ channels in endothelial cells, as shown by tension measurements after inhibition of the production of NO and prostacyclin. Treatment of the tissues with the SERCA inhibitor cyclopiazonic acid potentiated both the acetylcholine-induced endothelial hyperpolarization and the EDHF-mediated relaxation of the smooth muscle. Data from the bioassay experiments suggest that EDHF is not diffusible in this preparation, or that it is diffusible, but diluted to levels below threshold for activity. The existence of a non-NO/prostacyclin mediated relaxing factor was reported shortly after the discovery of EDRF by Furchgott (De May et al., 1982; Rubanyi and Vanhoutte, 1987). Subsequently, EDHF-mediated responses have been shown in many different preparations from various species. Intensive efforts in understanding EDHF-mediated vasodilatation indicate that it is more prevalent in small blood vessels, while in larger vessels endothelial-dependent relaxation is mainly mediated by NO (reviewed by Triggle et al., 1999). Experiments with a superfusion bioassay system, where the endothelium and smooth muscle layers are physically separated, suggested that EDHF might be a diffusible substance (Rubanyi et al., 1985; Félétou and Vanhoutte, 1988). Subsequent experiments using the “Sandwich” preparation and also patch-clamp techniques provided evidence that EETs, a cytochrome P450-generated metabolite of arachadonic acid, could function as EDHF (Campbell and Harder, 1999). On the other hand, the endothelium and smooth muscle are electrically coupled via gap junctions (Segal and Duling, 1986; Beny and Paccica, 1994). Thus, endothelial hyperpolarization may also spread to the smooth muscle cells and cause vasodilatation. In keeping with this, EDHF-mediated responses can be inhibited in some preparations by gap junction blockers such as Gap27 (Chaytor et al., 1998). A third possibility is that K⫹ can act as an EDHF (Edwards et al., 1998). Indeed in the rat carotid artery the acetylcholine-induced EDHF-mediated response could be correlated with an increase in K⫹ concentration in the myoendothelial junctional spaces. K⫹, which is released via the charybdotoxin and apamin sensitive KCa, appears to activate the Na⫹-K⫹-ATPase and inward rectifier K⫹ channels to cause hyperpolarization of the smooth muscle. In the rabbit aorta, the acetylcholine-induced relaxation is largely blocked by a combination of L-NAME and indomethacin, which inhibits the production of NO from eNOS and that of prostacyclin from endothelial cyclooxygenase. Nevertheless, there is a small component (~10% total acetylcholine-induced relaxation) that is resistant to L-NAME and indomethacin. The component remaining after L-NAME and indomethacin treatment was abolished by application of charybdotoxin and apamin. The application of 10⫺7 M iberiotoxin, an inhibitor of large conductance KCa channels, had no effect on the non-NO/prostacyclin mediated relaxation, indicating that smooth muscle BK channels are not the targets for EDHF. In contrast, BaCl2 abolished the L-NAME- and indomethacin-resistant relaxation, consistent with the notion that an increased K⫹ concentration may activate inward rectifier K⫹ channels to relax the smooth muscle.

282

X. Kuang et al.

Activation of endothelial KCa mediates acetylcholine-induced membrane hyperpolarization in freshly isolated rabbit aortic endothelial cells (Wang et al., 1996). Treatment with either bradykinin or cyclopiazonic acid greatly potentiated subsequent acetylcholine-induced responses. Under normal conditions, acetylcholine alone caused the endothelial cells to hyperpolarize; however, the acetylcholine-induced hyperpolarization was transient and returned to baseline values within 100 s. After treatment with cyclopiazonic acid, acetylcholine caused a maintained hyperpolarization of the same magnitude. The maintained responses were dependent on Ca2⫹ influx, since removal of extracellular Ca2⫹ abolished them. Consistent with the changes in membrane potential in freshly isolated endothelial cells, in the intact aorta, cyclopiazonic acid and bradykinin also potentiated the non-NO/prostacyclin mediated relaxation induced by acetylcholine. Both treatments doubled the size of the remaining component of relaxation after L-NAME and indomethacin. The addition of charybdotoxin, an inhibitor of the intermediate conductance KCa channels abolished the membrane hyperpolarization caused by acetylcholine. The small conductance KCa blocker, apamin had no effect on the maintained response when applied after stimulation, but did prevent the subsequent maintained membrane hyperpolarization. A similar effect was achieved with scallytoxin, another SK channel blocker. These data suggest an interaction between SK and IK channels. As in the force recordings with the intact aorta, a blocker (iberiotoxin) or opener (NS1619) of large conductance KCa had no effect on membrane hyperpolarization. These data suggest that maxi K channels are not involved in the EDHF-mediated response in the rabbit aorta. An alternative to K⫹ activating the Na⫹-K⫹-ATPase and/or KIR to cause hyperpolarization of smooth muscle is the possibility that endothelial hyperpolarization is directly conducted through myoendothelial gap junctions, as demonstrated in some small arteries (Segal and Duling, 1986). However, in large vessels there were no detectible myoendothelial gap junctions (Beny and Paccica, 1994). Heptanol, an inhibitor of gap junctions in many vessels, did not effect the non-NO/prostacyclin mediated relaxation in the rabbit aorta. In conclusion, this study provides evidence that in a large conduit blood vessel, charybdotoxin and apamin-sensitive KCa in endothelial cells mediate smooth muscle relaxation caused by EDHF. A bioassay experiment failed to detect a response with similar characteristics. These data suggest that the EDHF-mediated vasodilatation occurs by direct electrical coupling of endothelial and smooth muscle cells, the release of endothelial K⫹ into the endothelium-smooth muscle interstitial space, the release of a very labile factor, or a combination of the above.

36 The contribution of D-tubocurarine and apamin-sensitive potassium channels to endothelium-derived hyperpolarizing factor-mediated relaxation of small mesenteric arteries from eNOS⫺/⫺ mice Hong Ding, Yanfen Jiang and Chris R. Triggle Wire myograph and immunofluorescence techniques were used to investigate the nature and location of the potassium channels involved in mediating acetylcholine-mediated relaxation of contracted small mesenteric arteries from endothelial nitric oxide synthase knock-out, ⫺/⫺, mice. Previous data demonstrated that the contribution of endothelium-derived hyperpolarizing factor (EDHF) to acetylcholine-mediated relaxation of contracted small mesenteric arteries from ⫺/⫺ mice was up-regulated compared to tissues from ⫹/⫹ mice. In the current study, relaxation to acetylcholine was unaffected by charybdotoxin or apamin alone, but was significantly inhibited by the combination of these agents, or tubocurarine. The additive effects of scyllatoxin and iberiotoxin did not mimic the effects of the apamin/charybdotoxin combination. Thus, acetylcholine-induced EDHF-mediated relaxation in small mesenteric arteries involved the activation of tubocurarine and apamin/charybdotoxin sensitive K channels. Immunofluorescence labeling with antibodies for small conductance potassium channels subtypes, SK2 and SK3, on mouse whole mesenteric arteries and freshly isolated single vascular smooth muscle and endothelial cells was also conducted with antibodies to ␣-smooth muscle actin and von Willebrand factor to identify vascular smooth muscle and endothelial cells respectively. The data indicate that the SK2 was associated with the plasma membrane of endothelial and the nuclear region of vascular smooth muscle cells. SK3 was associated with the plasmamembrane of both vascular smooth muscle and endothelial cells. Together these data indicate that the apamin-sensitive channel involved in mediating endotheliumdependent hyperpolarization may be associated with both vascular smooth muscle and endothelial cells; however, SK2, which possesses the highest affinity for apamin, appears to be present in the greatest density and is only associated with the endothelial cell plasmamembrane.

1. INTRODUCTION The nature and distribution of the K⫹ channels involved in EDHF-induced relaxation remain controversial (McGuire et al., 2001). A frequent observation in many vascular beds is that a combination of apamin, the small conductance calcium-activated K⫹ channel (SKCa) blocker, and charybdotoxin, a blocker of both intermediate conductance (IKCa) and large conductance (BKCa) channels, as well as Kv1.2 and a Kv1.3 (Chandy and Gutman, 1995; Nelson and Quayle, 1995) is required to inhibit endothelium-dependent hyperpolarization (McGuire et al., 2001). Furthermore, in most vascular beds, the more selective BKCa blocker, iberiotoxin (Cook and Quast, 1990), fails to substitute for charybdotoxin (Waldron and Garland, 1994; Zygmunt and Högestätt, 1996; Edwards et al., 1998). The site of action

284

Hong Ding et al.

of the inhibitory effects of apamin and charybdotoxin is likely to be the endothelial cell (Edwards et al., 1998). The frequent requirement for both apamin and charybdotoxin in the inhibition of endothelium-dependent hyperpolarization is indicative that SKCa and an IKCa channels on the endothelial cell are involved in the synthesis/release of EDHF (Edwards and Weston, 1998). Charybdotoxin and apamin block the EDHF-mediated response when applied to the endothelium (lumen), but not when applied by superfusion of the vessel and these data provide additional support for the hypothesis that the charybdotoxin- and apaminsensitive channels reside on the endothelial cell and not on the vascular smooth muscle cell (Doughty et al., 1999). The cellular mechanisms that regulate the synthesis and release of NO and EDHF may share a similar activation process at the level of the endothelial cell (Busse et al., 1993; Nagao and Vanhoutte, 1993) and K⫹-channels on endothelial cell play a critical role in the regulation of the driving force for Ca2⫹ and, thus, can modulate the synthesis of NO and EDHF. There are three highly homologous subtypes of SK channels, SKI, SK2 and SK3, that show differing sensitivity to apamin and IC50 values of, based on displacement of 125Ilabeled apamin from oocytes expressing the cloned channels, 390, 4 and 11 pM respectively (Grunnet et al., 2001). In a study of the distribution of SK channels in porcine coronary arteries SK1 mRNA was not detected by RT-PCR, whereas SK2 and SK3 were found in the endothelium and SK2 protein was present in whole porcine artery preparations (Burnham et al., 2002). In addition, the use of immunofluorescent labeling demonstrated that SK3 is highly expressed in the plasmamembrane of endothelial cells, but not seen in the vascular smooth muscle cells and immunofluorescence of SK2 is restricted to the peri-nuclear region of both endothelial and vascular smooth muscle cells (Burnham et al., 2002). The nature and location of the apamin- and charybdotoxin-sensitive K⫹-channel(s) is an enigma. A novel channel, possessing the binding sites for both apamin and charybdotoxin, may exist on endothelial cells (Zygmunt et al., 1997). In support of this hypothesis is evidence that small changes in critical amino acids can result in dramatic changes in the pharmacology of K-channels (Ishii et al., 1997a,b), and thus, a novel channel associated with endothelial and/or vascular smooth muscle cells in the microvasculature may exist. Alternatively, both apamin-sensitive SKCa and charybdotoxin-sensitive IKCa channels may play a role and the block of only one of these channels leads to a compensatory increase in the contribution of the other channel (Edwards and Weston, 1998). In addition, it is also possible that at least two EDHFs are released and the release process of one is sensitive to apamin and that of the other to charybdotoxin. In some blood vessels apamin alone inhibits EDHF-mediated responses (Adeagbo and Triggle, 1993; Murphy and Brayden, 1995; Ayajiki et al., 2000) and, it is argued, an apaminsensitive mechanism plays a key role in mediating the effects of EDHF in other arteries such as the rat mesenteric artery (Chen and Cheung, 1997). Other inhibitors of SKCa channels, such as tubocurarine and scyllatoxin, also show variable efficacy as inhibitors of EDHF. Thus, tubocurarine significantly reduced bradykinin-induced hyperpolarization of porcine coronary artery endothelial cells (von der Weid and Beny, 1992). Murphy and Brayden (1995b) reported that apamin and scyllatoxin, administered alone, were equipotent as inhibitors of acetylcholine-induced EDHF-mediated responses in rabbit mesenteric arteries and, in addition, that tubocurarine also inhibited EDHF-mediated hyperpolarization. Similarly, in the canine corpus cavernosum, apamin and scyllatoxin inhibited acetylcholinemediated relaxations with equal efficacy (Ayajiki et al., 2000). These data indicate that the identity of the KCa channels involved in the regulation of EDHF-dependent relaxation can show considerable variability.

EDHF-mediated relaxation of small mesenteric arteries

285

EDHF-mediated relaxation in mouse saphenous and small mesenteric arteries is inhibited by the combination of apamin and charybdotoxin (Ding et al., 2000). The aim of the present study was to determine whether, in mouse small mesenteric arteries, KCa blockers could, in combination with charybdotoxin, substitute for apamin as inhibitors of EDHF-mediated relaxation. In addition, using immunofluorescence techniques, the cellular distribution of the SK channel subtypes was determined. 2. METHODS

2.1. Wire myograph experiments Homozygous eNOS⫺/⫺, ⫹/⫹ and C57BL/6J mice purchased from Jackson laboratories (Bar Harbor, Maine, USA) (~16 weeks of age) were killed by cervical dislocation after anesthesia with halothane and in accordance with a research protocol consistent with the standards of the Canadian Council on Animal Care and approved by the local Animal Care Committee of the University of Calgary. Small mesenteric arteries were removed and kept in Krebs’ solution (composition, mM): NaCl 120; NaHCO3 25; KCl 4.8; NaH2PO4 1.2; MgSO4 1.2; Dextrose 11.0; CaCl2 1.8; bubbled with 95% O2 and 5% CO2. First order small mesenteric arteries were cut into 2-mm rings and mounted on a Mulvany–Halpern-type myograph (Mulvany and Halpern, 1977). Studies were performed at 37 ⬚C (see Ding and Triggle, 2000; Ding et al., 2000; Waldron et al., 1999).

2.2. Cell isolation Previous studies have focused on mouse small mesenteric arteries (Ding and Triggle, 2000). In brief, first order mesenteric arteries from eNOS⫹/⫹ mice were first dissected free of adherent fat and connective tissues. This process was carried out in Ca2⫹-free medium: (mM): NaCl 120; NaHCO3 25; KCl 4.2; KH2PO4 1.2; MgCl2 1.2; glucose 11. Single smooth muscle cells were dissociated enzymatically with 2 mg/ml collagenase IV and 0.5 unit/ml elastase III in 10 ␮M Ca2⫹ medium for 12 min at 37 ⬚C. After digestion, the arteries were rinsed three times in Ca2⫹-free medium and triturated with a fire-polished Pasteur pipette to yield single smooth muscle cells and single, or clumps, of endothelial cells. Cells were collected on poly-L-lysine coated microscope slides and fixed by 1% formalin in phosphate buffer saline (composition in mM: NaCl 123, NaH2PO4 9.6, NaN3 0.01%, K2HPO4 43) for 15 min and skinned by 0.1–0.2% Triton for 5 min.

2.3. Immunofluorescence histochemistry C57BL/6J mice (10 –12 weeks of age) were killed by cervical dislocation under halothane. Mesenteric arteries were removed and kept in the smooth muscle dissection solution. Arteries were embedded in OCT® compound. Cryostat sections (thickness 10 ␮m) were collected on poly-L-lysine coated microscope slides and fixed with cold acetone on dry ice for 15 min. Slides were washed in Phosphate Buffer Saline and blocked for 1h in 3% bovine serum albumin in Phosphate Buffer Saline. Primary antibodies (polyclonal anti-SK2 and SK3) were prepared with and without control peptide pre-incubation (1 : 10), diluted to 1 : 200 in 3% bovine serum albumin/Phosphate Buffer Saline and applied for 1 h at room temperature or overnight at 4 ⬚C. For dual-label experiments, we separately applied primary antibodies, such as polyclonal anti-SK2 or SK3 and monoclonal anti-␣-smooth muscle actin

286

Hong Ding et al.

(1 : 500)/von Willebrandt’s factor (1 : 200), and washed with 3% bovine serum albumin/ Phosphate Buffer Saline between exposures. Negative controls were also performed by omission of primary antibody. Secondary antibodies conjugated to Cy3 (1 : 2000) and/or Alexa fluor™ 488 (1 : 1500) were applied for 30 min to 1 h, with or without 4, 6diamidino-2-phenylindole (DAPI; 5 ␮g/ ml final) included as a nuclear label. Sections or cells were mounted with FluorSave, sealed with nail polish and viewed on an Olympus IX70 epifluorescence microscope equipped with Spot-RT CCD camera (Diagnostic Instruments; Sterling Heights MI, USA).

2.4. Drugs All drugs were obtained from Sigma (St Louis, Missouri, USA), except for scyllatoxin, which was obtained from Alexis, anti-SK2 and anti-SK3 supplied with control peptides (Alomone Laboratories, Jerusalem, Israel), NCL-von Willebrandt’s factor mouse monoclonal (Novocastra, Newcastle, UK). All drugs were dissolved in distilled water except for indomethacin and 1H-(1,2,4)oxadiazolo(4,3-a) quinoxalin-1-one (ODQ) which were dissolved in 95% ethanol and dimethyl sulphoxide, respectively.

2.5. Data analysis Data are expressed as pD2 values and pD2 is defined as the negative logarithm to base 10 of the EC50 values. In a number of studies with the K-channel blockers it was not possible to obtain 100% relaxation to acetylcholine and thus only an estimate of the pD2 could be obtained. In all experiments, n equals the number of animals used in myograph experiments. Relaxation is expressed as percentage of cirazoline-induced tone ⫾ SEM. The statistical differences between mean values was calculated by Student’s t-test. Statistical significance of differences between the means of data groups was performed using ANOVA for curve analysis. Significance was assumed if P was less than 0.05.

3. RESULTS

3.1. Acetylcholine-induced relaxations in small mesenteric arteries from eNOS⫹/⫹ and ⫺/⫺ mice In small mesenteric arteries from eNOS⫺/⫺ mice, the presence of N␻-nitro-L-arginine (10⫺4 M) and indomethacin (10⫺5 M) did not cause a significant inhibition vs control: maximal relaxation and pD2 value (P ⬎ 0.05) (Figure 36.1). The combination of ODQ (10⫺5 M) with N␻-nitro-L-arginine and indomethacin also did not significantly reduce, compared to control, either the maximal relaxation response to acetylcholine or the pD2 (P ⬎ 0.05). In preparations with endothelium, first order mesenteric arteries from eNOS⫹/⫹ mice pre-contracted with cirazoline (10⫺7 M), acetylcholine induced a concentration-dependent relaxation. The presence of the nitric oxide synthase inhibitor, N␻-nitro-L-arginine, (10⫺4 M), and the cyclooxygenase inhibitor, indomethacin (10⫺5 M) resulted in a significant inhibition of both maximal relaxation to acetylcholine and pD2 (P ⬍ 0.05) (Figure 36.1). The combination of the putatively specific soluble guanylyl cyclase inhibitor, ODQ (10⫺5 M),

EDHF-mediated relaxation of small mesenteric arteries eNOS–/–

Relaxation (%)

A

B

eNOS+/+

287

C eNOS+/+

0

0

0

20

20

20

40

40

60

60

*

60

80

80

*

80

40

ODQ Control

100 8.0

7.5

7.0 6.5 6.0 [ACh] (–log M)

5.5

5.0

100

8.0

7.5

7.0

6.5

6.0

[ACh] (–log M)

5.5

5.0

100 –9

–8

–7 –6 [NO] (–log M)

–5

Control L-NNA+indomethacin

ODQ+L-NNA+indomethacin

Figure 36.1 Concentration–response curves illustrating acetylcholine- and NO-induced relaxations during cirazoline (10⫺7 M) contractions of mouse mesenteric arterioles (n ⫽ 6). (A) In blood vessels from eNOS⫺/⫺ mice neither the inhibition of nitric oxide synthase (N␻ M-nitro-L-arginine, L-NNA, 10⫺4 M) and cyclooxygenase (indomethacin, 10⫺5 M), nor NOS, cyclooxygenase and soluble guanylyl cyclase (ODQ, 10⫺5 M) reduced acetylcholine-induced relaxations (P ⬎ 0.05). (B) In arteries from eNOS⫹/⫹ mice both NOS, in combination with cyclooxygenase inhibition and the combination of NOS, cyclooxygenase and soluble guanylyl cyclase inhibition significantly reduced both maximally recorded acetylcholine-mediated relaxations and the pD2 (* P ⬍ 0.05). (C) In arteries from eNOS⫹/+ mice sGC inhibition with 10⫺5 M ODQ resulted in a significant reduction in the pD2 for NO-mediated relaxation (P ⬍ 0.05). Data are expressed as mean percentage relaxation, ⫾ SEM, to acetylcholine or NO (C) in the absence and presence of the inhibitors of NOS, cyclooxygenase and soluble guanylyl cyclase.

with N␻-nitro-L-arginine and indomethacin further significantly reduced the maximal relaxation and pD2 value (P ⬍ 0.05). Continued incubation of the vessels with ODQ for half an hour resulted in no further inhibition of relaxation. In the presence of ODQ the pD2 (P ⬍ 0.05), but not the maximal response, for the NO-induced relaxation was significantly different from the control (Figure 36.1).

3.2. Potassium channel blockers Acetylcholine-mediated relaxations of eNOS⫺/⫺ and ⫹/⫹ mouse first order mesenteric arteries, contracted with cirazoline and in the presence of N␻-nitro-L-arginine and indomethacin, were compared following incubation of the tissues for 30 min with the following potassium channel inhibitors: charybdotoxin (10⫺7 M), apamin (10⫺6 M), tubocurarine (10⫺4 M), scyllatoxin (10⫺7 M) and iberiotoxin (10⫺7 M). 3.2.1. Apamin and charybdotoxin In eNOS⫺/⫺ mice neither charybdotoxin (10⫺7 M) nor apamin (10⫺6 M) alone significantly affected the relaxation induced by acetylcholine (Figure 36.2). The combination of apamin and charybdotoxin significantly inhibited maximal relaxation, at least up to the maximal concentration of acetylcholine that could be tested, and pD2 values of the acetylcholineinduced relaxation (P ⬍ 0.05).

288 A

Hong Ding et al. eNOS–/–

B

0

Relaxation (%)

20

eNOS+/+

C eNOS+/+

0

0

20

20

* 40

40

60

60

60

80

80

80

*

40

Control apamin apamin/ChTX 100

100 8

7 6 [ACh] (–log M)

5

100 8

7 6 [ACh] (–log M)

5

9

8

7

6

[NO] (–log M)

L-NNA/indomethacin L-NNA/indomethacin/apamin L-NNA/indomethacin/ChTX L-NNA/indomethacin/apamin/ChTX

Figure 36.2 Concentration–response curves illustrating acetylcholine- and NO-induced relaxation during cirazoline contractions of mouse mesenteric arteries in the presence of either apamin and charybdotoxin (ChTX) alone, or the combination of apamin and charybdotoxin (n ⫽ 6). (A) In blood vessels from eNOS⫺/⫺ mice neither apamin alone (10⫺6 M) nor charybdotoxin (10⫺7 M) significantly reduced acetylcholine-induced relaxation (P ⬎ 0.05). The combination of apamin and charybdotoxin significantly inhibited the acetylcholine-induced maximal relaxation that was measured as well as reduced the pD2 (* P ⬍ 0.05). (B) In arteries from eNOS⫹/⫹ mice apamin and charybdotoxin produced a reduction in the sensitivity (pD2) of the tissue to acetylcholine (P ⬍ 0.05). The combination of apamin and charybdotoxin significantly reduced the pD2 and inhibited the acetylcholine-induced maximal relaxation (* P ⬍ 0.05). (C) In blood vessels from eNOS⫹/⫹ mice the combination of apamin and charybdotoxin, but not apamin alone, significantly reduced the pD2 for the NO-induced relaxation (P ⬍ 0.05). Data are expressed as mean percentage relaxation, ⫾ SEM, to acetylcholine or NO (C) in the presence of inhibitors of nitric oxide synthase, cyclooxygenase, [N␻-nitro-L-arginine (L-NNA, 10⫺4 M) and indomethacin, 10⫺5 M, respectively], and the potassium channel inhibitors apamin or charybdotoxin (ChTX) alone or the combination of apamin and charybdotoxin.

In mesenteric arteries from eNOS⫹/⫹ mice treatment with charybdotoxin or apamin alone in tissues which had not been treated with N␻-nitro-L-arginine and indomethacin produced no significant inhibition of the maximal acetylcholine-induced relaxation response when compared to control (Figure 36.2), but significantly shifted the concentration– response curve for acetylcholine as reflected by the pD2 values (P ⬍ 0.05). In the presence of N␻-nitro-L-arginine and indomethacin the combination of apamin and charybdotoxin significantly inhibited the pD2 (P ⬍ 0.05). Apamin alone had no effect on NO-mediated relaxation, but charybdotoxin and apamin together reduced the sensitivity (pD2) of the tissue. 3.2.2. Tubocurarine In eNOS⫺/⫺ mice, 10⫺4 M tubocurarine alone almost completely abolished the acetylcholine-induced relaxation (P⬍0.05) (Figure 36.3). However, in eNOS⫹/⫹ mice, treatment with tubocurarine alone produced no inhibition of maximal relaxation and pD2, but the combination of charybdotoxin and tubocurarine nearly abolished the acetylcholine-induced relaxation (P⬍0.05). Tubocurarine alone had no effect on NO-induced relaxation (Figure 36.3).

EDHF-mediated relaxation of small mesenteric arteries A

eNOS–/–

B

0

eNOS+/+

C

* *

eNOS+/+ 0

0

20

289

20

20

Relaxation (%)

* 40

40

40

60

60

60

80

80

80

100

8.0

7.5

7.0 6.5 6.0 [ACh] (–log M)

5.5

5.0

100

100

8.0

7.5

7.0 6.5 6.0 [ACh] (–log M)

5.5

5.0

9.0

8.5

8.0 7.5 7.0 [NO] (–log M)

6.5

6.

L-NNA / indomethacin L-NNA / indomethacin/tubocurarine L-NNA / indomethacin/tubocurarine/ChTX

Figure 36.3 Concentration–response curves showing acetylcholine- and NO-induced relaxation in the presence of tubocurarine and charybdotoxin (ChTX) (n ⫽ 6). (A) In arteries from eNOS⫺/⫺ mice tubocurarine (10⫺4 M) alone almost completely inhibited acetylcholineinduced relaxations (* P ⬍ 0.05). (B) In blood vessels from eNOS⫹/⫹ mice the combination of tubocurarine and charybdotoxin, but not tubocurarine alone, significantly inhibited both the maximal acetylcholine-induced relaxations as well as the pD2 (* P ⬍ 0.05). (C) In arteries from eNOS⫹/⫹ mice the combination of tubocurarine and charybdotoxin, but not tubocurarine alone, resulted in a significant reduction in the pD2 and significantly reduced the maximal NO-induced relaxation that was recorded (* P ⬍ 0.05). Data are expressed as mean percentage relaxation, ⫾ SEM, to acetylcholine or NO. (C) In the presence of the inhibitors of nitric oxide synthase, cyclooxygenase, [N␻-nitro-L- arginine (L-NNA, 10⫺4 M) and indomethacin, 10⫺5 M, respectively], and the potassium channel inhibitors tubocurarine and/or charybdotoxin (ChTX).

3.2.3. Scyllatoxin In eNOS⫺/⫺ mice, treatment with 10⫺7 M scyllatoxin alone resulted in no significant inhibition in either maximal or pD2 values for acetylcholine-induced relaxation when compared to the response to the muscarinic agonist obtained in the presence of N␻-nitro-L-arginine and indomethacin (P ⬎ 0.05) (Figure 36.4). However, scyllatoxin combined with charybdotoxin produced a significant reduction in both maximal relaxation and pD2 (P ⬍ 0.05). Treatment with scyllatoxin alone in the presence of N␻-nitro-L-arginine and indomethacin in eNOS⫹/⫹ mice resulted in no significant inhibition in either maximal or pD2 values (P ⬎ 0.05) (Figure 36.4). The combination of scyllatoxin and charybdotoxin produced a significant shift in the pD2 value (P ⬍ 0.05) but produced no significant inhibition of the maximal relaxation in response to acetylcholine. 3.2.4. Iberiotoxin In eNOS⫺/⫺ mice, treatment of the tissues with 10⫺7 M iberiotoxin alone resulted in no significant inhibition compared to the relaxation mediated by acetylcholine in the presence of N␻-nitro-L-arginine and indomethacin (P ⬎ 0.05). The combination of iberiotoxin and apamin resulted in a significant inhibition of maximal relaxation and pD2 value (P ⬍ 0.05) (Figure 36.5).

Relaxation (%)

A

eNOS–/–

B

eNOS+/+

0

0

20

20

*

40

40

60

60

80

80

*

100

100 8.0

7.5

7.0 6.5 6.0 [ACh] (–log M)

5.5

5.0

8.0

7.5

7.0 6.5 6.0 [ACh] (–log M)

5.5

5.0

L-NNA/indomethacin L-NNA/indomethacin/scyllatoxin L-NNA/indomethacin/scyllatoxin/ChTX

Figure 36.4 Concentration–response curves showing acetylcholine- and NO-induced relaxations in the presence of scyllatoxin and charybdotoxin (ChTX) (n ⫽ 6). (A) In blood vessels from eNOS⫺/⫺ mice scyllatoxin (10⫺7 M) alone produced no significant inhibition of acetylcholine-induced relaxation, but the combination of scyllatoxin and charybdotoxin significantly reduced both the maximal relaxation response as well as the pD2 (* P ⬍ 0.05). (B) In arteries from eNOS ⫹/⫹ mice the combination of scyllatoxin and charybdotoxin, but not scyllatoxin alone, significantly reduced both the maximal acetylcholine-induced relaxation as well as the pD2 (* P ⬍ 0.05). Data are expressed as mean percentage relaxation, ⫾ SEM, to acetylcholine in the presence of the inhibitors of nitric oxide synthase, cyclooxygenase, [N␻-nitro-L-arginine (L-NNA, 10⫺4 M) and indomethacin, 10⫺5 M, respectively], and the potassium channel inhibitors scyllatoxin alone or together with charybdotoxin (ChTX).

Relaxation (%)

A

eNOS–/–

B

eNOS+/+

C

eNOS+/+

0

0

0

20

20

20

* *

40

40

*

40

60

60

60

80

80

80

100

100 8.0

7.5

7.0 6.5 6.0 [ACh] (–log M)

5.5

5.0

100

8.0

7.5

7.0

6.5 6.0 5.5 [ACh] (–log M)

5.0

9.0

8.5

8.0 7.5 7.0 [NO] (–log M)

6.5

6.0

L-NNA/indomethecin L-NNA/indomethecin/IbTX L-NNA/indomethecin/IbTX/apamin

Figure 36.5 Concentration–response curves illustrating acetylcholine- and NO-induced relaxations in the presence of iberiotoxin (IbTX) and apamin (n ⫽ 6). (A) In blood vessels from eNOS⫺/⫺ mice the combination of iberiotoxin (10⫺7 M) and apamin (10⫺6 M), but not iberiotoxin alone, significantly reduced the maximal acetylcholine-induced relaxations as well as the pD2 (* P ⬍ 0.05). (B) In arteries from eNOS⫹/⫹ mice iberiotoxin alone significantly reduced the maximal acetylcholine-induced relaxations and the pD2 (* P ⬍ 0.05). (C) In blood vessels from eNOS⫹/⫹ mice iberiotoxin alone and the combination of iberiotoxin and apamin significantly reduced the pD2 but not the maximal relaxation induced by NO (P ⬍ 0.05). Data are expressed as mean percentage relaxation, ⫾ SEM, to acetylcholine or NO (C) in the presence of the inhibitors of NOS, cyclooxygenase, [N␻-nitro-L-arginine (L-NNA, 10⫺4 M) and indomethacin, 10⫺5 M, respectively], and the potassium channel inhibitors iberiotoxin (IbTX) alone or in combination with apamin.

EDHF-mediated relaxation of small mesenteric arteries

291

Treatment with iberiotoxin alone in mesenteric arteries from eNOS⫹/⫹ mice significantly inhibited both maximal relaxation and the pD2 value (P ⬍ 0.05). Combining apamin with iberiotoxin resulted in no additional inhibition (Figure 36.5). Iberiotoxin alone reduced the sensitivity of the small mesenteric artery to NO (Figure 36.5).

3.3. Immunofluorescence The distribution of SK2 and SK3 channels in cross sections from first order mouse mesenteric artery was examined. The data was derived from tissues obtained from five mice. SK2 immunofluorescence was consistently associated with the nuclear region of vascular smooth muscle cells and the plasmamembrane of endothelial cells. The data for SK3 was, however, more variable and SK3 was associated with the plasmamembrane of both endothelial cells and vascular smooth muscle cells from 4 of 7 animals; however, in tissues and cells from 3 of 7 animals the presence of SK3 was not as obvious. Six 10 ␮m cross-sections were cut from first order mesenteric arteries and stained with anti-SK2 and anti-SK2 pre-incubated with blocking peptide, anti-SK3 and anti-SK3 incubated with blocking peptide and secondary antibody alone or the nuclear stain, 4, 6-diamidino-2-phenylindole (Figure 36.6). All of the above were double stained with NCL-von Willebrandt’s factor which is a mouse monoclonal anti-human von Willebrandt’s factor. Immunofluorescence (green) associated with anti-von Willebrandt’s factor is apparent only in the internal layer of the section. Anti-SK2 (red) and SK3 (red) is associated not only with the cell layer that demonstrates positive antivon Willibrand factor fluorescence, but also with the smooth muscle layers, however, and as illustrated by staining the cell nuclei with 4, 6-diamidino-2-phenylindole (blue) the anti-SK2 flourescence is associated with the smooth muscle cell nuclei. Single smooth muscle cells exposed to anti-SK2 (red), SK3 (red) and blocking peptides for SK2 and SK3 illustrate that SK2 fluorescence is associated with the nuclear membrane of the smooth muscle cell, whereas that for anti-SK3 is confined to the plasmamembrane. The use of blocking peptides for both SK2 and SK3 indicates the specificity of the respective antibodies (Figure 36.7). Sheets of isolated endothelial cells were also exposed to anti-SK2 (red) and anti-SK3 (red) and the results indicate that immunofluorescence for both antibodies is associated only with the plasmamembrane of the endothelial cells (Figure 36.8).

4. DISCUSSION The treatment of mouse small mesenteric arteries from eNOS⫹/⫹ with N␻-nitro-L-arginine and indomethacin resulted in a small, but significant shift in the endothelium-dependent concentration–relaxation curve to acetylcholine. However, the addition of the soluble guanylyl cyclase inhibitor ODQ to the combination of N␻-nitro-L-arginine and indomethacin resulted in an almost complete inhibition of the response to acetylcholine. These data suggest that there is a NOS inhibitor-insensitive/resistant activation of soluble guanylyl cyclase that contributes to the endothelium-dependent relaxation to acetylcholine. Comparable data have also been presented for the rabbit carotid artery where the combination of two nitric oxide synthase inhibitors, N␻-nitro-L-arginine and N␻-nitro-L-arginine methyl ester, failed to inhibit NO synthesis (Cohen et al., 1997). Thus, the NOS inhibitor insensitive relaxation in mouse small mesenteric arteries may reflect the action of NO synthesized from, potentially, a nitric oxide synthase-independent pathway (Ding et al., 2000; Andrews et al.,

292

Hong Ding et al. A

B

C

D

E

F

Figure 36.6 The distribution of SK2 and SK3 channels on 10 ␮m cross-sections cut from first order mesenteric arteries was examined by immunofluorescence labelling. The following sections were double stained with anti-von Willebrandt’s factor (green): (A) anti-SK2 (red); (B) anti-SK2 incubated with blocking peptide (red); (C) anti-SK3 (red); (D) anti-SK3 pre-incubated with blocking peptide (red); (E) secondary antibody; (F) nuclear stained with 4, 6-diamidino-2-phenylindole (blue) (see Color Plate 12).

2002). Alternatively, the nitric oxide synthase inhibitor-insensitive, but ODQ-sensitive, relaxation in mesenteric arteries from eNOS⫹/⫹ may reflect a contribution from hydrogen peroxide (H2O2) which has been described as the EDHF in mesenteric arteries from eNOS⫺/⫺ mice (Matoba et al., 2000) and may mediate relaxation via a direct or cGMPmediated activation of KCa channels (Barlow and White, 1998; Hayabuchi et al., 1998a). However, catalase, that would be expected to inhibit the actions of H2O2, failed to inhibit the NOS inhibitor-insensitive relaxation in mesenteric arteries from both eNOS⫺/⫺ and ⫹/⫹ mice (Ding and Triggle, unpublished data).

EDHF-mediated relaxation of small mesenteric arteries

A

293

B

C

D

E

Figure 36.7 The distribution of SK2 and SK3 channels of single smooth muscle cells freshly dispersed from first order mesenteric arteries was examined by immunofluorescence labelling: (A) anti-SK2 (red); (B) anti-SK2 (red) double stained with anti-␣-smooth muscle actin (green); (C) anti-SK2 pre-incubated with blocking peptide (red); (D) antiSK3 (red); (E) anti-SK3 incubated with blocking peptide (red) (see Color Plate 13).

A two-hour exposure of mesenteric vessels from eNOS⫺/⫺ mice to ODQ also abolished the NOS- and cyclooxygenase-insensitive relaxation to acetylcholine, but this may reflect non-specific actions of ODQ (Feelisch et al., 1999). In the rabbit carotid artery, the concentration of NO released by 3 ⫻ 10⫺6 M acetylcholine peaked at about 2.5 ⫻ 10⫺7 M, which, based on the concentration–response curves for NO in eNOS⫹/⫹ mouse small mesenteric arteries (Figure 36.1), would have been almost completely inhibited by 10⫺5 M ODQ (Cohen et al., 1997). In contrast to the data from eNOS⫹/⫹, in eNOS⫺/⫺ mice the acetylcholine-mediated relaxation is completely resistant to the combination of N␻-nitro-Larginine indomethacin and ODQ, but is nearly abolished by the combined treatment with the KCa inhibitors apamin and charybdotoxin. However, the combination of the SKCa inhibitor, apamin, with the IKCa blocker, charybdotoxin, only inhibited acetylcholine-mediated relaxation in mesenteric arteries from eNOS⫹/⫹ mice that had been treated with N␻-nitroL-arginine and indomethacin. Although the exact cellular location of the apamin- and charybdotoxin-sensitive channels involved in determining EDHF-mediated relaxation in mouse small mesenteric arteries is unknown, in rat hepatic and mesenteric arteries, these channels are located on the endothelial cells (Edwards et al., 1998; Doughty et al., 1999).

294

Hong Ding et al.

A

B

C

D

Figure 36.8 The distribution of SK2 and SK3 channels of endothelial cells freshly dispersed from first order mesenteric arteries was examined by immunofluorescence labelling: (A) antiSK2 (red) double stained with anti-␣-smooth muscle actin (green); (B) secondary antibody (Cy3-red) with von Willebrandt’s factor (green); (C) anti-SK3 (red); (D) anti-SK3 incubated with blocking peptide and double stained with von Willebrandt’s factor (green) and nuclear stained with 4, 6-diamidino-2-phenylindole (blue) (see Color Plate 14).

Nonetheless, apamin has no effect on acetylcholine-induced hyperpolarizations in endothelial cells (Chen and Cheung, 1992; Marchenko and Sage, 1996) and apamin alone was very effective in inhibiting acetylcholine-induced hyperpolarizations in rat superior mesenteric arteries (Chen and Cheung, 1997). In the guinea-pig carotid artery studies with glass microelectrode recording techniques demonstrated that the combination of apamin and charybdotoxin depolarized vascular smooth muscle cells by an endothelium-independent mechanism (Corriu et al., 1996a). These observations suggest that apamin-sensitive Kchannels may also reside on vascular smooth muscle cells. Similarly, in the rat superior mesenteric artery apamin alone produced a large inhibition of acetylcholine-induced hyperpolarization and the addition of charybdotoxin resulted in a reversal of the hyperpolarization to a depolarization (Chen and Cheung, 1997). These data could be interpreted as indicating that two EDHFs are released: one acting on apamin-sensitive channels and the second EDHF acting on a charybdotoxin-sensitive channels. The BKCa inhibitor, iberiotoxin, only inhibited acetylcholine-mediated relaxations in eNOS⫹/⫹ and not in tissues from eNOS⫺/⫺ mice thus suggesting that it inhibits the effect of NO on vascular smooth muscle rather than the vascular action of EDHF.

EDHF-mediated relaxation of small mesenteric arteries

295

Treatment of the mesenteric arteries from eNOS⫺/⫺ mice with tubocurarine resulted in a nearly complete inhibition of the relaxation to acetylcholine. When scyllatoxin was substituted for apamin and combined with charybdotoxin a significant inhibition of the maximal relaxation-response to acetylcholine was detected. These data suggest that SKCa play a crucial role in the regulation of EDHF-mediated events in eNOS⫺/⫺ mice. Tubocurarine has also been shown to block hSK1 and rSK2 with IC50 values of 2.7 ⫻ 10⫺5 M and 1.7 10⫺5 M when measured at ⫹80 mV (Strøbæk et al., 2000). At 0.3–30 ⫻ 10⫺6 M tubocurarine blocked a 270 pS BKCa channel from the cytoplasmic side, but not when applied externally even at concentrations as high as 3.0 ⫻ 10⫺4 M (Egan et al., 1993). Tubocurarine reduced the mean open time and conductance of a 285 pS KCa channel when applied in the bath in the inside-out patch configuration (Baron et al., 1996). Tubocurarine was the most powerful inhibitor of bradykinin-induced endothelium-dependent hyperpolarizations among the K-channel inhibitors tested (Baron et al., 1996). However, in the rat hepatic artery, the SKCa inhibitors scyllatoxin (10⫺6 M) and tubocurarine (10⫺6 M) only partially inhibited EDHF-mediated relaxation when each of them was combined with charybdotoxin (3 ⫻ 10⫺7 M) (Andersson et al., 2000). In contrast, the present study indicates that tubocurarine is a potent blocker of EDHF in eNOS⫺/⫺ mouse mesenteric arteries and this most likely reflects its action as a blocker of both IKCa and the SK2 subtype of the SKCa channel (Cai et al., 1998; Strobaek et al., 2000). Differences in the sensitivity of the rat hepatic artery vs mouse mesenteric arteries to tubocurarine may reflect heterogeneity in the nature of the KCa channels that modulate the release and/or action of EDHF in these two blood vessels. Variability in the sensitivity of different tissues to scyllatoxin may reflect the lower potency of this toxin against the SKCa channel (Andersson et al., 2000). In conclusion, the myograph studies indicate that, in mouse small mesenteric arteries, an apamin- and tubocurarine-sensitive channel(s) plays a key role in the regulation of the action of EDHF. Furthermore, an apamin-sensitive channel is important for the regulation of the NOS inhibitor-insensitive synthesis of an ODQ-sensitive relaxation that is, most likely, mediated by NO. Immunofluorescence studies were carried out in whole artery sections of the small mesenteric arteries as well as freshly dispersed single smooth muscle and endothelial cells from the same blood vessels. Endothelial cells were identified by their positive immunoreactivity to von Willebrandt’s factor and smooth muscle by ␣-smooth muscle actin immunoreactivity and nuclei by their staining with 4,6-diamidino-2-phenylindole. Immunocytochemistry identified that both SK2 and SK3 channels were present in both endothelial and vascular smooth muscle cells; however, the SK2 immunofluorescence was associated with vascular smooth muscle cell nuclei and immunofluorescence for SK3 was not observed in every blood vessel or single vascular or endothelial cell; the reason for the variable distribution of SK3 channels (approximately 50% of the mice tissues/cells studied) is unknown. The specificity of the anti-SK2 antibody has also been questioned (Khanna et al., 2001; Ro et al., 2001).

5. CONCLUSION These studies support the importance of apamin and charybdotoxin-sensitive K-channels in determining EDHF-mediated relaxation and illustrate that other small-conductance KCa channel blockers, tubocurarine and scyllatoxin, can substitute for apamin. Immunofluorescence data indicates that the SK2 and SK3 channels that likely mediate the effects of apamin are

296

Hong Ding et al.

located on both smooth muscle and endothelial cell; however, the SK2 channel is associated with the nuclear membrane of the smooth muscle cell and is thus unlikely to be involved in mediating the rapid effects of EDHF. ACKNOWLEDGMENTS The authors wish to thank Professor Gary Kargacin and Mrs Gail McMartin for their valuable advice with the immunofluorescence microscopy.

37 Ouabain blocks EDNO-mediated relaxation in mesenteric veins and EDHF-mediated relaxation in mesenteric arteries of the guinea-pig Simon Roizes and Pierre-Yves von der Weid Mechanisms underlying endothelium-dependent relaxations were compared in veins and arteries from the mesentery of guinea-pigs by examining changes in vessel diameter and smooth muscle membrane potential in response to acetylcholine. Constricted first and second order veins, relaxed during application of acetylcholine. The relaxation was not affected by the cyclooxygenase inhibitor, indomethacin and by BaCl2, but abolished after removal of the endothelium, in the presence of the nitric oxide-synthase inhibitor, NG-nitro L-arginine, and with the guanylate cyclase inhibitor, ODQ. The relaxation was also inhibited by ouabain and by the protein kinase A inhibitor, H89. Sodium nitroprusside induced a relaxation that was inhibited by ouabain and H89. Intercellular microelectrode recordings revealed that acetylcholine hyperpolarized the venous smooth muscle. The acetylcholine-induced hyperpolarization was largely inhibited by NG-nitro L-arginine and abolished by ouabain. Constricted mesenteric arteries responded to acetylcholine with a marked relaxation. The acetylcholineinduced relaxation was blocked by ouabain, but not significantly decreased by NG-nitro L-arginine and ODQ. It was not affected by indomethacin, H89 and BaCl2. Sodium nitroprusside also induced a relaxation that was inhibited by ouabain and ODQ and reduced by H89. Acetylcholine strongly hyperpolarized the arterial smooth muscle, a response not affected by NG-nitro L-arginine. The present results demonstrate that the mechanism underlying endothelium-dependent relaxations to acetylcholine varies amongst mesenteric veins and arteries. They are mainly mediated by nitric oxide and cGMP in the vein, but an additional pathway independent of nitric oxide-synthase and cyclooxygenase activation is involved in the guinea-pig mesenteric artery. In both these vessels, acetylcholine-induced relaxations were blocked by ouabain, suggesting a common activation of the Na⫹/K⫹-ATPase through different intracellular pathways.

1. INTRODUCTION Acetylcholine mediates endothelium-dependent relaxations of blood vessels via mechanisms involving the endothelial release of vasoactive substances like nitric oxide, prostacyclin and endothelium-dependent hyperpolarizing factor(s) (EDHFs) and/or the electrotonic transmission of the endothelial hyperpolarization to the smooth muscle (Bény and von der Weid, 1993; Triggle et al., 1999; Félétou and Vanhoutte, 1999a). Irrespective of its origin, the endothelium-dependent hyperpolarization plays an important role in the relaxation of the smooth muscle. It is consequent to opening of K⫹ channels such as inward-rectified K⫹ channels, Ca2⫹-dependent K⫹ channels or ATP-sensitive K⫹ channels (Triggle, 2000). The Na⫹/K⫹ pump plays an important role in the regulation of the tone of smooth muscle (Clausen and Nielsen, 1994), and has long since been proposed to be involved in endothelium-dependent relaxations, as the administration of ouabain-inhibited relaxations

298

Simon Roizes and Pierre-Yves von der Weid

to acetylcholine and nitric oxide in arteries (DeMey and Vanhoutte, 1980; Rapoport et al., 1983; Ferrer et al., 1999). The importance of Na⫹/K⫹ pump activation in EDHF is supported by the finding that ouabain, together with barium ions, inhibits the effect of an EDHF proposed to be K⫹ (Edwards et al., 1998). Endothelium-dependent relaxations and hyperpolarizations have been investigated in mesenteric arteries, but not in mesenteric veins. Like in arteries, the venous endothelium plays a role in modulating the contractile activity of the underlying vascular smooth muscle (Miller and Vanhoutte, 1989; Miller, 1991). The aim of the present study was to investigate the mechanisms involved in the endotheliumdependent relaxation to acetylcholine in the guinea-pig mesenteric veins and to compare them with those occurring in arteries in the same preparation. 2. METHODS

2.1. Tissue preparation Guinea pigs (15 to 21 days of age) of either sex were killed by decapitation during deep anesthesia consequent to inhalation of halothane (5–10%). This procedure has been approved by the University of Calgary Animal Care and Ethics Committee and conforms to the guidelines established by the Canadian Council on Animal Care. The small intestine and attached mesentery were rapidly removed and placed in a physiological saline solution of the following composition (10⫺3 M): CaCl2 2.5; KCl 5; MgCl2 2; NaCl 120; NaHCO3 25; NaH2PO4 1; glucose 11. The pH was maintained at 7.4 by constant bubbling with 95% 02/5% CO2.

2.2. Myograph studies First and second order veins and arteries were isolated from the mesentery. Four segments (length 1–2 mm) were cut from four different blood vessels and suspended between a micropositioner and a force transducer with stainless-steel wire (diameter 40 ␮m) in a Mulvany–Halpern style organ chamber (Model 610 multi-myograph system, J.P. Trading, Aarhus, DK). Isometric tension was recorded online on a computer via an analog-to-digital converter (PowerLab/4SP, ADInstruments, Mountain View, CA). Resting tensions of 2 mN and 1 mN was set for arteries and veins, respectively, and a period of 45 min was allowed for equilibration. The preparations were routinely contracted with 114 ⫻ 10⫺5 M KCl to determine their viability. They were then contracted with U46619 (10⫺6 M) or histamine (10⫺5 M) and responses to acetylcholine (5 ⫻ 10⫺6 M–10⫺5 M, 1 min) and/or sodium nitroprusside (10⫺5 M, 1 min) were determined.

2.3. Electrophysiology A portion of mesentery containing first or second order veins and arteries was pinned into a small organ bath (volume 100 ␮l), mounted on the stage of an inverted microscope (Nikon, TMS) and superfused (flow rate 3 ml/min) with physiological saline solution heated to 36 ⬚C. The resting membrane potential was measured using conventional glass intracellular microelectrodes with resistances of 150–250 M⍀ when filled with 0.5 M KCl. The electrodes were connected to an amplifier (Intra 767, World Precision Instruments, Sarasota, FL) through an Ag-AgCl half-cell. The resting membrane potential was monitored on a digital oscilloscope

Ouabain blocks EDNO-mediated and EDHF-mediated relaxation

299

(Hitachi, VC6525) and simultaneously recorded on a computer via a PowerLab/4SP. Impalements of smooth muscle cells were obtained from the adventitial side of the blood vessels and characterized by a sharp drop in potential that settled after 10–30 s to a value typically more negative than ⫺45 mV. Agonists were applied first as a control and, in experiments where the effects of inhibitors were investigated, a second time at least 20 min later in the presence of the inhibitor that had been superfused for at least 10 min. This protocol was usually performed during the same impalement. However, in some instances, successive impalements were obtained from neighboring cells in the same segment. No significant difference in the response induced by the same agonist applied at the same concentration 20 min apart in the absence of an inhibitor was observed.

2.4. Chemicals and drugs Acetylcholine, indomethacin, NG-nitro L-arginine, ouabain and sodium nitroprusside were purchased from Sigma/Aldrich (Oakville, ONT), histamine was obtained from ICN (Costa Mesa, CA), H89 from Alexis (San Diego, CA) and U46619 from Cayman Chemicals (Ann Arbor, MI). Drugs were dissolved in deionized water (except for NG-nitro L-arginine that was dissolved in 0.1 M HCl and H89 in methyl sulfoxide) to give 10⫺2 M stock solutions. U46619 was diluted from the acetate buffer solution provided by the manufacturer.

2.5. Statistical analysis Experimental data have been expressed as the mean ⫾ standard error of the mean (SEM). Statistical significance was assessed using paired Student’s t-test with P less than 0.05 being considered significant.

3. RESULTS

3.1. Mesenteric veins Studies were performed on first and second order mesenteric veins with inner diameter ranging from 95 to 110 ␮m. Superfusion with the physiological saline solution containing 10⫺5 M histamine or 10⫺6 M U46619 caused the vessel to contract (5 ⫾ 1 mN, n ⫽ 118). Acetylcholine (10⫺5 M) applied for 1 min caused a relaxation that reversed the contraction by 26 ⫾ 1% (n ⫽ 118). The relaxation was not decreased but significantly increased by indomethacin (10⫺5 M, n ⫽ 4), not affected by BaCl2 (3 ⫻ 10⫺5 M, n ⫽ 12), but abolished after removal of the endothelium (n ⫽ 7). Acetylcholine-induced relaxation was inhibited in the presence of NG-nitro L-arginine (10⫺4 M, n ⫽ 14) and ODQ (10⫺5 M, n ⫽ 6) and sometimes reversed to a contraction (decrease in vessel diameter). Both blockers were observed to increase the contraction. In the presence of ouabain (10⫺5 M), the contraction level of the vessel was not changed, but the acetylcholine-induced relaxation was abolished (n ⫽ 27). Similarly, the protein kinase A inhibitor, H89 (10⫺5 M) abolished acetylcholine-induced relaxations (n ⫽ 17). Sodium nitroprusside (10⫺5 M–10⫺4 M) induced a relaxation of 24 ⫾ 2% (n ⫽ 24) that was inhibited by ODQ (n ⫽ 6), ouabain (n ⫽ 4) and H89 (n ⫽ 4). In the presence of a high K⫹-physiological saline solution (substitution of 6 ⫻ 10⫺2 M NaCl with 6 ⫻ 10⫺2 M KCl), both acetylcholine and sodium nitroprusside-induced relaxations were inhibited by almost 50% (Figure 37.1).

Figure 37.1 Effect of acetylcholine on the tension of mesenteric veins contracted with U46619 (10⫺7 M). (A & B) Original traces illustrating the response to acetylcholine (ACh, 10⫺5 M) in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M), indomethacin (Indo, 10⫺5 M) and ouabain (10⫺5 M). (C & D) Histograms summarizing the relaxation to acetylcholine and sodium nitroprusside (SNP) under control conditions (closed bars) and in the presence of various blockers (open bars). *: P less than 0.05 (paired Student’s t test).

Figure 37.2 Effect of acetylcholine and sodium nitroprusside on the membrane potential of smooth muscle in mesenteric veins. Original recordings in response to acetylcholine (ACh, 5 ⫻ 10⫺6 M) and sodium nitroprusside (SNP, 5 ⫻ 10⫺5 M) under control conditions (top traces, A & B) and in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M) and ouabain (10⫺5 M) (lower traces, respectively, A & B).

Ouabain blocks EDNO-mediated and EDHF-mediated relaxation

301

Acetylcholine (5⫻10⫺6 –10⫺5 M) hyperpolarized the venous smooth muscle by 6⫾1mV, from a resting value of ⫺62 ⫾ 5 mV (n ⫽ 6). In some occasions, a rapid and transient depolarization occurred, while the cell was hyperpolarized by acetylcholine (see Figure 37.2). This response was never observed with sodium nitroprusside (5 ⫻ 10⫺5 M), which only hyperpolarized the smooth muscle. The acetylcholine-induced hyperpolarization was largely inhibited in the presence of NG-nitro L-arginine (10⫺4 M, n ⫽ 3). In the presence of ouabain (10⫺5 M), the membrane of the smooth muscle depolarized and underwent rhythmical oscillations, with amplitudes ranging from a few mV up to 40 mV, depending on the tissue. Under these conditions, the application of acetylcholine or sodium nitroprusside did not cause any hyperpolarization (n ⫽ 2, Figure 37.2).

3.2. Mesenteric arteries Arteries (inner diameter 80–100 ␮m) responded to superfusion with acetylcholine (5 ⫻ 10⫺6 M–10⫺5 M) by a marked relaxation of 63⫾3% during contractions to 10⫺6 M U46619 (n⫽33). The acetylcholine-induced relaxation was blocked by ouabain (10⫺5 M, n ⫽ 11), but not significantly decreased by NG-nitro L-arginine (10⫺4 M, n ⫽ 12) and ODQ (10⫺5 M, n⫽6) and not affected significantly by indomethacin (10⫺5 M, n⫽4), H89 (10⫺5 M, n ⫽ 4) and BaCl2 (3 ⫻ 10⫺5 M, n ⫽ 4). In high K⫹-physiological saline solution, the acetylcholine-induced relaxation was inhibited to an extent similar to ouabain. Sodium nitroprusside (10⫺5 M) induced a relaxation of 60 ⫾ 4% of the contraction, which was inhibited by ouabain (n ⫽ 8), ODQ (n ⫽ 6) and high K⫹-physiological saline solution (n ⫽ 6). The sodium nitroprusside-induced relaxation was also reduced by H89 (n ⫽ 4), but not affected by BaCl2 (Figure 37.3). Acetylcholine hyperpolarized the arterial smooth muscle by 17 ⫾ 4 mV from a resting value of ⫺58 ⫾ 6 mV (n ⫽ 4). This response was not affected by NG-nitro L-arginine (n ⫽ 2, Figure 37.4).

Figure 37.3 Effect of acetylcholine on the tension of mesenteric arteries contracted with U46619 (10⫺7 M). (A) Original trace illustrating the relaxation induced by acetylcholine (ACh) in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M) and ouabain (10⫺5 M). (B & C) Histograms summarizing the relaxation to acetylcholine and sodium nitroprusside (SNP) under control conditions (closed bars) and in the presence of various blockers (open bars). *: P less than 0.01 (paired Student’s t test).

302

Simon Roizes and Pierre-Yves von der Weid

Figure 37.4 Effect of acetylcholine on the membrane potential of smooth muscle in mesenteric artery. Original recordings in response to acetylcholine (ACh, 5 ⫻ 10⫺6 M) under control conditions (A) and in the presence of NG-nitro L-arginine (L-NA, 10⫺4 M), (B).

4. DISCUSSION The present study demonstrates that the smooth muscle of mesenteric arteries and veins from the guinea-pig responded to acetylcholine with a relaxation and a hyperpolarization. Although both responses are endothelium-dependent, they vary amongst vessels. In veins, the acetylcholine-induced relaxation is mainly mediated by endothelium-derived nitric oxide, via a cGMP-dependent mechanism. Importantly, venous smooth muscle hyperpolarizes in response to acetylcholine, an effect also due to nitric oxide, as it was mimicked by sodium nitroprusside and abolished by NG-nitro L-arginine. In arteries, the findings that a pathway, independent of nitric oxide-synthase and cyclooxygenase, is activated in response to acetylcholine, confirm results obtained in the same preparation (Dong et al., 2000) and are consistent with previous reports from studies in mesenteric arteries of guinea-pigs (Bolton et al., 1984), dogs (Komori et al., 1988) and rats (McPherson and Angus, 1991). In these blood vessels, acetylcholine and cholinomimetics cause both relaxation and hyperpolarization. Although the acetylcholine-induced vasorelaxation was consistently decreased by nitric oxide-synthase and/or guanylate cyclase inhibitors, the endothelium-dependent hyperpolarization to the muscarinic agonist was not inhibited by NG-nitro L-arginine in the rat mesenteric artery and involvement of prostacyclin, cytochrome P450 arachidonic acid metabolites and endocanabinoid agonists has been ruled out (Fukao et al., 1997; Vanheel and van de Voorde, 1997). The endothelium-dependent hyperpolarization is disrupted in rat mesenteric arteries by 18␤-glycyrrhetinic acid, a potent gap junction uncoupler (Yamamoto et al., 1998). This finding suggests that in the mesenteric artery of the rat, the endothelium-dependent hyperpolarization of the smooth muscle and the resulting relaxation is due to an electronic transmission of the hyperpolarization induced in the endothelium by acetylcholine. A similar conclusion was reached with a synthetic connexin peptide (Gap 27 peptide) to alter gap junctional communications between endothelial and smooth muscle cells (Dora et al., 1999). Although the present

Ouabain blocks EDNO-mediated and EDHF-mediated relaxation

303

results did not specifically address the question of the identity of EDHF, earlier work concluded to the involvement of the three classes of calcium-activated K⫹ channels in the EDHF-dependent relaxation, excluding a role for K⫹ ions and cytochrome P450 (Dong et al., 2000). Although nitric oxide seemed to be marginally involved in the acetylcholineinduced relaxation and hyperpolarization in arteries, it relaxes these blood vessels and causes a consistent hyperpolarization at the concentration used. ACKNOWLEDGMENTS This study was supported by grants from the Alberta Heritage Foundation for Medical Research (AHFMR) and the Heart and Stroke Foundation of Canada. PYvdW is an AHFMR Scholar. The authors wish to thank Dr J.J. McGuire for helpful discussion and comment on the manuscript.

38 Inhibitors of EDHF-evoked responses and the calcium signal in endothelial cell of mesenteric artery Philippe Ghisdal and Nicole Morel

The EDHF-mediated relaxation induced by acetylcholine requires an increase in the cytosolic Ca2⫹ concentration in the endothelial cells and is associated with hyperpolarization of the smooth muscle cells. The aim of the present work was to investigate whether or not agents inhibiting EDHF-evoked responses in the rat superior mesenteric artery could act by preventing the Ca2⫹ response of endothelial cells. EDHF-dependent relaxation of arteries contracted with prostaglandin F2␣ and hyperpolarization of smooth muscle cells evoked by acetylcholine were abolished by high KCl or by charybdotoxin plus apamin. The relaxations and hyperpolarizations were inhibited by 4-aminopyridine. Ouabain produced a time-dependent inhibition of both responses. In indo-1 loaded arteries, acetylcholine increased the endothelial Ca2⫹ signal with the same potency as it hyperpolarized the smooth muscle cells. Increases in KCl concentration in the bathing solution depressed endothelium Ca2⫹ response to acetylcholine but the inhibition was markedly lower than that of the hyperpolarization of smooth muscle cells. The endothelial Ca2⫹ signal was insensitive to charybdotoxin plus apamin but was depressed by 4-aminopyridine. 4-Aminopyridine abolished the transient thapsigargin-sensitive increase in cytosolic Ca2⫹ evoked by acetylcholine in Ca2⫹-free solution, suggesting that this blocker could affect the Ca2⫹ release process. Ouabain produced a small transitory drop in the Ca2⫹ response of endothelial cells to acetylcholine. These results show that interaction with the Ca2⫹ release in endothelial cells might be involved in the inhibition of EDHF-mediated responses by 4-aminopyridine but not by blockers of Ca2⫹-activated K⫹ channels.

1. INTRODUCTION In many arteries, acetylcholine stimulates the release of endothelium-derived hyperpolarizing factor (EDHF). An important prerequisite of EDHF release is the increase in Ca2⫹ concentration in endothelial cells (Chen and Suzuki, 1990; Fukao et al., 1995). One potential way to inhibit EDHF-dependent responses is to inhibit the increase in Ca2⫹ in endothelial cells. An increase in cytosolic Ca2⫹ after stimulation of the endothelial cells may originate from intracellular stores or from the extracellular compartment. Endothelial cells are devoid of functional voltage-operated calcium channels (Himmel et al., 1993) so the entry of Ca2⫹ into the cell is governed by the electrochemical gradient for this ion. Thus, Ca2⫹ entry is increased following hyperpolarization and is decreased when cell membrane is depolarized (Lückhöff and Busse, 1990). Any change in membrane potential might then influence the driving force for Ca2⫹ entry, the Ca2⫹ signal in response to endothelial cell stimulation, and by consequence the EDHF-dependent hyperpolarization and relaxation of smooth muscle cell. The EDHF-dependent relaxation and hyperpolarization are inhibited by several blockers of K⫹ channels or Na⫹ pumps. The aim of the present study was to investigate whether

Inhibitors of EDHF-evoked responses and the calcium signal

305

inhibition of EDHF-dependent responses could be related to interaction with Ca2⫹ signaling in endothelial cells.

2. METHODS Superior mesenteric arteries isolated from male Wistar Kyoto rats (14 weeks) were used. Physiological solution contained (in mM): NaCl 122, KCl 5.9, NaHCO3 15, glucose 10, MgCl2 1.25 and CaCl2 1.25, gassed with a mixture of 95% O2–5% CO2 (37 ⬚C). All experiments were performed in the presence of N␻-nitro-L-arginine (L-NOARG, 10⫺4 M) and indomethacin (10⫺5 M) to block the nitric oxide synthase and the cyclo-oxygenase, respectively and phentolamine (10⫺6 M) to inhibit the effects of norepinephrine that could be released from nerve terminals. Contractile responses and cell membrane potential were measured in inverted arteries mounted in a myograph under a basal tension of 7–10 mN (Ghisdal et al., 1999). The Ca2⫹ signal in endothelial cells was measured using front surface fluorimetry in indo-1-loaded arteries. Segments of arteries (about 2 mm) were inverted and mounted between two hooks under a tension of about 7 mN in the cuvette of a fluorimeter filled with physiological solution. Rings were loaded with indo-1/AM (5.10⫺6 M) for 3 h at room temperature. After excitation at 340 nm, the fluorescence signals emitted at 405 (F405) and 500nm (F500) were recorded. To isolate the endothelial signal, all experiments were performed in the presence of nimodipine (10⫺6 M). In endothelium-intact preparations, acetylcholine produced a concentration-dependent increase in Ca2⫹ signal. When the endothelium was removed mechanically, acetylcholine (10⫺6 M) did not modify the Ca2⫹ signal. At the end of each experiment, the autofluorescence of the tissue was measured at 405 and 500 nm by quenching the indo-1 fluorescence with MnCl2 (5 mM) and was subtracted from F405 and F500. The Ca2⫹ concentration was estimated by the ratio of the fluorescence emitted at 405 and 500 nm.

3. RESULTS In the presence of L-NOARG and indomethacin, acetylcholine (10⫺6 M) hyperpolarized the smooth muscle cells of the mesenteric artery by 15.7 ⫾ 1.3 mV (n ⫽ 8). Measurement of the endothelial cell Ca2⫹ signal in indo-1-loaded arteries indicated that the concentrationresponses curves to acetylcholine for the increase in Ca2⫹ signal in endothelial cells and for the hyperpolarization of the smooth muscle cells were perfectly superimposed (Ghisdal and Morel, 2001). The effect of increased KCl concentration (40 mM) and of different K⫹ channel blockers on the Ca2⫹ signal evoked by acetylcholine (10⫺6 M) in endothelial cells was investigated, in comparison with their effect on the hyperpolarization of the smooth muscle cells. Charybdotoxin (10⫺7 M) plus apamin (10⫺7 M) totally inhibited the hyperpolarization of smooth muscle cells but did not affect the increase in Ca2⫹ concentration evoked by acetylcholine in endothelial cells. 4-Aminopyridine (5.10⫺3 M) depressed the hyperpolarization of the smooth muscle cells and the Ca2⫹ signal in the endothelial cells induced by acetylcholine by 58 ⫾ 8 % (n ⫽ 6) and 27 ⫾ 4% (n ⫽ 7), respectively. The effect of 4-aminopyridine on the Ca2⫹ signal in endothelial cells was not affected by an increase in KCl concentration in the bath (Figure 38.1).

306

Philippe Ghisdal and Nicole Morel

120

Acetylcholine-evoked responses (% of max)

100

80

*

*

60

*

40

20

0

K40/4-AP

K40

BaCl2

4-AP

Chtx+Apa

Control

–20

Figure 38.1 Effect of acetylcholine (ACh, 10⫺6 M) on Ca2⫹ signal in endothelial cells (open bars) and membrane potential of smooth muscle cells (filled bars) in untreated artery (control) and in the presence of charybdotoxin and apamin (Chtx ⫹ Apa, 10⫺7 M), 4-aminopyridine (4-AP, 5.10⫺3 M), BaCl2 (3.10⫺5 M), in physiological solution containing 40 mM of KCl without (K40) or with 4-AP (K40 ⫹ 4-AP). The asterisk indicates a statistically significant difference vs control for the Ca2⫹ signal in endothelial cells (P ⬍ 0.05) and the cross indicates a statistically significant difference vs control for the membrane potential of the smooth muscle cells (P ⬍ 0.05).

Intracellular Ca2⫹ release evoked by acetylcholine was measured in Ca2⫹-free/EGTA physiological solution containing 40 mM of KCl. Under this condition, acetylcholine (10⫺6 M) evoked a rapid but transient increase in Ca2⫹ signal, which amounted to 51 ⫾ 6% (n ⫽ 8) of the maximal response to acetylcholine in the presence of Ca2⫹ and KCl 40 mM and was abolished by thapsigargin (10⫺6 M). 4-Aminopyridine inhibited the transient calcium peak elicited by acetylcholine by 75 ⫾ 7% (n ⫽ 5, P ⬍ 0.05 vs control) (Figure 38.2). Fifteen minutes incubation with ouabain (10⫺3 M) depressed the EDHF-dependent relaxation of the mesenteric artery by 64 ⫾ 12% (n ⫽ 7, acetylcholine 10⫺6 M). Similarly, the EDHF-dependent hyperpolarization evoked by acetylcholine (10⫺6 M) was inhibited by 57 ⫾ 3.8% (n ⫽ 8). However, the effect of ouabain appeared to be time-dependent as after 45 min incubation in the presence of the Na⫹ pump inhibitor relaxations and hyperpolarizations to acetylcholine were nearly abolished. When added during the plateau phase of the endothelial Ca2⫹ response to acetylcholine ouabain rapidly produced a 21 ⫾ 7% (n ⫽ 4, P ⬍ 0.05 paired t-test) drop in the Ca2⫹ signal but this effect tended to decrease with time.

Inhibitors of EDHF-evoked responses and the calcium signal

307

F405/F500 (% of basal value)

10 8 6 4 2

**

**

0 Control

Thapsi

4-AP

Figure 38.2 Effect of 4-aminopyridine on intracellular Ca2⫹ release in endothelial cells of rat mesenteric arteries. Indo-1 loaded mesenteric arteries were perfused with Ca2⫹ free/EGTA solution. Ca2⫹ release was evoked by acetylcholine 10⫺6 M in untreated arteries (control), in arteries incubated with thapsigargin (Thapsi, 10⫺6 M) or in arteries incubated with 4-aminopyridine (4-AP, 5.10⫺3 M). Values are means from 4–5 determinations. The asterisks indicate a statistically significant difference vs control (P ⬍ 0.01).

4. DISCUSSION In the superior mesenteric artery of the rat, the relaxation evoked by acetylcholine in the presence of inhibitors of nitric oxide synthase and cyclooxygenase is accompanied by a rise in cytosolic Ca2⫹ concentration in endothelial cells and hyperpolarization of the membrane of smooth muscle cells. The present results show that interaction with the Ca2⫹ signal in endothelial cells does not account for the inhibition of EDHF-dependent hyperpolarization of smooth muscle cells by blockers of Ca2⫹-dependent K⫹ channels but might be partly responsible for the inhibition of EDHF-mediated responses by 4-aminopyridine and by high KCl. Fluorescence studies using indo-1-loaded arteries allowed measuring the Ca2⫹ signal in the endothelial cells of the superior mesenteric artery. There was a close correlation between the effect of acetylcholine on the cytosolic Ca2⫹ concentration in endothelial cells and the change in membrane potential of the smooth muscle cells (Ghisdal and Morel, 2001). This is in agreement with previous observation that the hyperpolarization evoked by acetylcholine is initiated by an increase in Ca2⫹ concentration in the endothelial cells (Chen and Suzuki, 1990; Fukao et al., 1995). As a consequence of these observations one might expect that blunting the Ca2⫹ signal in endothelial cells causes a proportional reduction in the EDHF-evoked hyperpolarization of the smooth muscle cells. The increase in Ca2⫹ concentration in the endothelial cells results from both the release of intracellular Ca2⫹ stores and the influx of Ca2⫹ from the extracellular compartment. In the absence of voltage-operated calcium channels in endothelial cells (Himmel et al., 1993), Ca2⫹ influx is controlled by the membrane potential in such a way that cell hyperpolarization promotes Ca2⫹ entry while depolarization opposed Ca2⫹ entry (Lückhöff and Busse, 1990). Increasing extracellular K⫹ concentration during agonist stimulation of endothelial cells indeed diminished the rise in Ca2⫹ signal (Kamouchi et al., 1999a; Wang and van Breemen, 1999). However, in the presence of 40 mM KCl, which completely blocked the hyperpolarization of smooth muscle cells to EDHF, the Ca2⫹ signal in endothelial cells was only depressed by about 25% indicating that an additional effect has to be implicated.

308

Philippe Ghisdal and Nicole Morel

Charybdotoxin and apamin are known blockers of the EDHF-mediated responses (Corriu et al., 1996a; Zygmunt and Höggestätt, 1996; Prieto et al., 1998; Doughty et al., 1999; Quignard et al., 1999). Charybdotoxin is a blocker of Ca2⫹-activated K⫹ channels of intermediate conductance and apamin is a blocker of Ca2⫹-activated K⫹ channels of small conductance (Castle, 1999). Charybdotoxin and apamin-sensitive K⫹ channels have been postulated to be located on the endothelial cells, where they could open in response to the increase in Ca2⫹ concentration produced by cell activation (Marchenko and Sage, 1996; Burnham et al., 2002). They are responsible for the hyperpolarization evoked by acetylcholine in endothelial cells (Wang et al., 1996; Ohashi et al., 1999). The present study ruled out the possibility that these toxins abolish the hyperpolarization of the smooth muscle cells by inhibiting the Ca2⫹ signal in endothelial cells. The effect of 4-aminopyridine was different in that it produced about 60% inhibition of the EDHF-dependent relaxation and of the hyperpolarization of the smooth muscle cells evoked by acetylcholine, and also depressed the Ca2⫹ signal in endothelial cells by about 30%. However, this effect was not related to a change in the membrane potential of endothelial cells since it was not abolished by incubation of the artery in high KCl solution. The decrease in Ca2⫹ signal observed with 4-aminopyridine can be explained by the inhibition of the intracellular Ca2⫹ release process activated by acetylcholine. This observation is in agreement with the interaction of 4-aminopyridine with intracellular K⫹ channels which results in the inhibition of the IP3 response (Wood and Gillespie, 1998). Inhibition of the Na⫹ pump by ouabain results in a decrease in the EDHF-mediated responses of mesenteric artery as in several other arteries (Edwards et al., 1998; Prieto et al., 1998; Dora and Garland, 2001). Inhibition of relaxation as well as of hyperpolarization of the smooth muscle cells increased with incubation time of the artery in the presence of ouabain. The time-dependency of this effect does not result from the kinetic properties of ouabain binding since a similar result is obtained when the Na⫹ pump is inhibited by decreasing the KCl concentration in the bathing solution to 0.1 mM (data not shown). The time-dependency of the effect of Na⫹ pump inhibition is probably related to the change in the Na⫹ and K⫹ concentration gradients resulting from the inhibition of their active transport. Ouabain produced a small decrease in the Ca2⫹ response of endothelial cells stimulated by acetylcholine, which might be attributed to the change in membrane potential consecutive to inhibition of the Na⫹ pump. This observation indicates that time is an important parameter in the analysis of the effect of the Na⫹ pump inhibitor on the EDHF-mediated responses. As observed with the K⫹ channel blockers, the interaction of ouabain with the Ca2⫹ signal in endothelial cells cannot be the only mechanism underlying the inhibition of EDHF-dependent hyperpolarization of the smooth muscle cells. ACKNOWLEDGMENTS This work was supported by a grant from the Ministère de l’Education et de la Recherche Scientifique (Action Concertée no. 00/05-260) and from the FRSM (grant no. 3.4534.98). The authors thank G. Leonardy for her excellent technical support.

39 Roles of the inward-rectifier K⫹ channel and Na⫹/K⫹-ATPase in the hyperpolarization to K⫹ in rat mesenteric arteries Gillian Edwards, Gillian R. Richards, Matthew J. Gardener, Michel Félétou, Paul M. Vanhoutte and Arthur H. Weston In rat mesenteric and hepatic arteries, K⫹ which effluxes from endothelial cell K-channels is an endothelium-derived hyperpolarizing factor (EDHF) which hyperpolarizes the underlying smooth muscle (Edwards et al., 1998). The aim of the present study was to investigate further the mechanism by which elevation of extracellular K⫹ hyperpolarizes rat mesenteric artery smooth muscle in endothelium-denuded arteries. Under basal conditions, elevation of extracellular K⫹ by 5 ⫻ 10⫺3 M induced a hyperpolarization which was insensitive to ouabain but which was essentially abolished in the presence of ouabain ⫹ barium. In the presence of phenylephrine (which depolarized the myocytes), 5 ⫻ 10⫺3 M K⫹ was without effect but hyperpolarization to K⫹ was restored in the presence of iberiotoxin. Under these conditions the response to 5 ⫻ 10⫺3 M K⫹ was abolished by ouabain alone whereas barium alone was without effect. RT-PCR, Western blotting and immunohistochemical techniques showed the presence of Na⫹/K⫹-ATPase ␣1-, ␣2- and ␣3-subunits and of Kir2.1, Kir2.2 and Kir2.3 proteins in the myocytes although Kir2.2 was not located in the plasmalemma. It is concluded that the relative contribution of inwardly rectifying K-channels (encoded by Kir2.1 and/ or Kir2.3) and Na⫹/K⫹-ATPase (comprising ␣2- and/or ␣3-subunits) to K⫹-induced hyperpolarization in rat mesenteric artery smooth muscle is determined by the membrane potential. At potentials close to the resting membrane potential both are involved in the response whereas when the smooth muscle is depolarized inwardly rectifying K-channels no longer contribute to the observed electrical changes.

1. INTRODUCTION Membrane potential studies using sharp microelectrodes have demonstrated that in intact rat mesenteric and hepatic arteries, elevation of extracellular K⫹ by 5 ⫻ 10⫺3 M produces hyperpolarization of the smooth muscle (Edwards et al., 1998). Inhibition of the hyperpolarization requires the presence of both ouabain and barium and it was thus concluded that K⫹ stimulated both the Na⫹/K⫹-ATPase and K⫹ efflux via inwardly rectifying K⫹ channels (KIR), respectively (Edwards et al., 1998). In both the rat hepatic and mesenteric arteries, the hyperpolarizing effect of K⫹ on vascular smooth muscle is endothelium-independent (Edwards et al., 1998; Richards et al., 2001). However, in tension studies, the relaxant effect of small increases in extracellular K⫹ was difficult to obtain and appeared to be endotheliumdependent (Andersson et al., 2000; Doughty et al., 2000; Lacy et al., 2000). In an attempt to explain these discrepancies, the effect of phenylephrine (the spasmogen employed in the tension studies) on the electrophysiological response to K⫹ was investigated in endothelium-denuded mesenteric arteries (Richards et al., 2001). The results confirmed that

310

Gillian Edwards et al.

K⫹ was effective in the absence of the endothelium, but showed that in the presence of phenylephrine the hyperpolarization to K⫹ became endothelium-dependent. It was concluded that the agonist, by depolarizing the smooth muscle and elevating its intracellular calcium concentration, stimulated K⫹ efflux from the myocytes via voltage- and calciumsensitive K⫹ channels (Richards et al., 2001). Thus, it was hypothesized that the smooth muscle-derived K⫹ would form a ‘K⫹ cloud’ around the myocytes. This would result in maximal stimulation of the Na⫹/K⫹-ATPase and KIR channels, thus preventing further activation and hyperpolarization by exogenous K⫹. Inhibition of K⫹ efflux from largeconductance calcium-sensitive K⫹ channels (BKCa) on the smooth muscle with iberiotoxin is sufficient to restore the hyperpolarizing effect of K⫹ in endothelium-denuded arteries in the presence of phenylephrine (Weston et al., 2002). In the present study, the putative molecular identity and functional roles of the Na⫹/K⫹-ATPase and KIR channel ␣-subunits were determined using electrophysiological, immunohistochemical and molecular biological techniques. 2. METHODS Male Sprague–Dawley rats were killed by stunning and cervical dislocation. Whole mesenteric beds were placed in ice-cold Krebs solution (mM, NaCl 118, KCl 3.4; CaCl2 2.5; KH2PO4 1.2; MgSO4 1.2; NaHCO3 25; glucose 11; gassed with 5% CO2 in O2) and arteries were dissected free of surrounding tissue. Blood vessels with endothelium were used to provide samples for RT-PCR and Western blots.

2.1. Microelectrode studies Small mesenteric arteries (2nd and 3rd order), from which endothelial cells had been removed by perfusing vessels with distilled water, were superfused with Krebs solution containing 10⫺5 M indomethacin and 3 ⫻ 10⫺4 M NG-nitro-L-arginine (10 ml/min, 37 ⬚C). Smooth muscle cells were impaled from the adventitial side using microelectrodes filled with 3 M KCl (tip resistance 50–80 M⍀) and the lack of a functional endothelium was confirmed by an absence of response to acetylcholine. KCl was added as a bolus directly into the bath to produce a calculated transient 5 mM elevation in the extracellular K⫹ concentration. Levcromakalim (SmithKline Beecham) was used to indicate the ability of the smooth muscle cells to hyperpolarize following removal of the endothelium and drug treatments. Barium (3 ⫻ 10⫺5 M), iberiotoxin (Latoxan; 100 nM), ouabain (10⫺7 M–5 ⫻ 10⫺7 M) and phenylephrine (10⫺5 M) were added to the Krebs superfusing the bath.

2.2. Gene-specific RT-PCR Total RNA was isolated from rat mesenteric arteries and brain using Qiagen RNeasy Mini kits according to the manufacturer’s instructions. Following DNase treatment, cDNA synthesis was performed using Superscript II reverse transcriptase (Qiagen). RT-PCR primers, designed to amplify bases 3252–3410 of rat ␣1 (Genbank number NM 012504), 4797–5050 of rat ␣2 (NM 012505) and 3324–3529 of rat ␣3 (NM 012506) Na⫹/K⫹-ATPases, comprised: ␣1-forward: 5⬘-GAAGCTCATCATCAGGCGACG-3⬘, ␣1-reverse: 5⬘-CCAGGGTAGAGTTCCGAGCTC-3⬘, ␣2-forward: 5⬘-GGGCCTGACTAATTTGAGATCACTG-3⬘,

Roles of the inward-rectifier K⫹ channel and Na⫹/K⫹-ATPase 311 ␣2-reverse: 5⬘-GTCTCACAGAAGGTCACCAGTAAGG-3⬘, ␣3-forward: 5⬘-CCACACCTCGGTTACCTCTCAC-3⬘, ␣3-reverse: 5⬘-CAGATTTAGAACCGGAGATGGC-3⬘.

Since the full-length sequence of rat Kir2.1 was unknown, 3⬘ RACE (rapid amplification of mRNA ends by PCR) was performed to determine the sequence of the 3⬘ UTRs. For Kir2.2 and Kir2.3, primers were designed to amplify bases 1711–1937 of Genbank number X78461 and 2408–2630 of Genbank number X87635, respectively. The primers comprised: rKir2.1-forward: 5⬘-TGTTAGGTTTCAGCTCTGGGCTCTG-3⬘, rKir2.1-reverse: 5⬘-ACCCATGGTTTGTGCAGTACAATAGG-3⬘, rKir2.2-forward: 5⬘-CATGGTGTAGGGACAGCTGGAATATTC-3⬘, rKir2.2-reverse: 5⬘-GGATCGTAACATGGTCACAAACTGC-3⬘, rKir2.3-forward: 5⬘-CCACCTCTGCCCATTTCTACCTG-3⬘, rKir2.3-reverse: 5⬘-GTTACCAAACGCCATCATCAATCTG-3⬘.

RT-PCR was performed for 35 cycles with (depending on the melting temperatures of the primers) annealing temperatures ranging from 60 to 65 ⬚C and products were visualized using 1.5% agarose-ethidium bromide gels. To confirm identity, reaction products were sequenced.

2.3. Immunofluorescence histochemistry After fixing and embedding in OCT® compound, 4 ␮m cryostat sections were incubated with 1 mg/ml SDS in PBS for 30 min and then blocked (50 ␮l/ml normal goat serum, 10 mg/ml bovine serum albumin in PBS). The sections were incubated for 1 h at room temperature with primary antibodies (monoclonal: anti-␣1, M8P1A3, Affinity Bioreagents; anti-␣2, McB2, kindly provided by Prof K. Sweadner and anti-␣3, XVIF9-G10, Affinity Bioreagents; polyclonal: anti-Kir2.1, Alomone; anti-Kir2.2 and anti-Kir2.3 were a kind gift from Dr R.I. Norman). Secondary antibodies conjugated to Texas Red (␣1) or Cy3 (␣2, ␣3 and Kir2.x) were applied for 30 min together with 5 ␮g/ml 4,6-diamidino-2-phenylindole (DAPI) as a blue-fluorescent nuclear stain. Negative controls were either sections which were exposed to secondary antibody alone (monoclonal antibodies) or to primary antibody which had been exposed to an excess of antigenic peptide (polyclonal antibodies). Identical microscope, camera and software settings were used when imaging labelled sections and negative controls.

2.4. Materials Unless stated otherwise, all reagents were supplied by Sigma UK.

3. RESULTS Under basal conditions, transient elevation of extracellular K⫹ by 5 ⫻ 10⫺3 M induced a brief hyperpolarization of mesenteric artery smooth muscle, irrespective of the presence of the endothelium. In contrast, hyperpolarization to K⫹ in the presence of 10⫺5 M phenylephrine was only observed when the endothelium was intact (Richards et al., 2001; Figure 39.1). Nevertheless, during exposure to phenylephrine, a hyperpolarization to K⫹ was restored in endothelium-denuded arteries when they were exposed simultaneously to 10⫺7 M iberiotoxin

312

Gillian Edwards et al.

Figure 39.1 Effect of endothelium on hyperpolarizations of smooth muscle to KCl and levcromakalim, in the presence and absence of phenylephrine, in rat mesenteric arteries. In the absence of phenylephrine (A), responses to 5 ⫻ 10⫺3 M KCl or to 10⫺5 M levcromakalim (LK) were similar in the presence or absence of the endothelium. In contrast, although the response to levcromakalim was not modified by removal of the endothelium in the presence of phenylephrine (B), the response of the smooth muscle to KCl was abolished in arteries without endothelium. In all experiments, removal of the endothelium was confirmed by a lack of response to 10⫺5 M acetylcholine (ACh) (from Richards et al., 2001).

(Weston et al., 2002; Figure 39.2). Thus, it was possible to study the relative contribution of Na⫹/K⫹-ATPase and KIR to the K⫹-induced smooth muscle hyperpolarization in arteries without endothelium under basal conditions (absence of spasmogen) or in depolarizing conditions (presence of 10⫺5 M phenylephrine ⫹ 10⫺7 M iberiotoxin). Under basal conditions, the K⫹-induced hyperpolarization of the smooth muscle was insensitive to 5 ⫻ 10⫺7 M ouabain but essentially abolished in the additional presence of 3 ⫻ 10⫺5 M barium (Weston et al., 2002; Figure 39.2). In the presence of 10⫺5 M phenylephrine⫹ 10⫺7 M iberiotoxin the hyperpolarization to 5 ⫻ 10⫺3 M mM K⫹ was fully inhibited by 5⫻10⫺7 M ouabain alone whereas 3⫻10⫺5 M barium alone was without effect. This suggests that both inward-rectifier K⫹ channels (barium-sensitive) and Na⫹/K⫹-ATPase may contribute to the response to K⫹. RT-PCR identified the presence of mRNA for Kir2.1, Kir2.2 and Kir2.3 in endothelial cells but of only Kir2.1 and Kir2.2 in the myocytes (Figure 39.3). However, antibody staining of cryostat sections of artery was consistent with the plasmalemmal presence of Kir2.1 and Kir2.3 gene products in both smooth muscle and endothelial cells. In contrast, although antiKir2.2 bound to both endothelium and smooth muscle, the binding was essentially restricted to regions of DAPI staining, suggesting a perinuclear location. The subunit identity of Na⫹/K⫹-ATPase and the KIR channel was further investigated using RT-PCR, Western blotting and immunohistochemical techniques. Single-cell RT-PCR suggested that mRNA for the ␣1- and ␣3-subunits, but not for the ␣2-subunit, was present in myocytes whereas endothelial cells possessed the messages for all three sub-types (Weston et al., 2002; Figure 39.3). A similar result was obtained in an immunohistochemical study

Roles of the inward-rectifier K⫹ channel and Na⫹/K⫹-ATPase 313

Figure 39.2 Effects of ouabain and barium on hyperpolarizations induced by KCl and levcromakalim in rat mesenteric arteries without endothelium. (A) Typical trace showing the smooth muscle hyperpolarization induced by transient application of 5 ⫻ 10⫺3 M KCl and 10–5 M levcromakalim (LK) in the presence of 5 ⫻ 10⫺7 M ouabain and 3 ⫻ 10⫺5 M barium (BaCl2) as indicated by the horizontal bars. In the presence of phenylephrine the KCl-induced hyperpolarization was lost, but was restored by exposure to 10⫺7 M iberiotoxin (IbTX) (B,C). Under these conditions, the KCl-induced hyperpolarization was unaffected by 3 ⫻ 10⫺5 M barium (B) but was abolished by 5 ⫻ 10⫺7 M ouabain. Although not shown, removal of the endothelium was confirmed in all experiments by a lack of response to 10⫺5 M acetylcholine. (Adapted from Weston et al., 2002.)

of the distribution of Na⫹/K⫹-ATPase ␣-subunits in rat mesenteric artery smooth muscle (Juhaszova and Blaustein, 1997). However, in the present study, binding of antibodies to ␣1-, ␣2- and ␣3-subunits of Na⫹/K⫹-ATPase was detected in smooth muscle cells whereas only anti-␣1 and anti-␣3 antibodies became bound to the endothelium (Weston et al., 2002; Figure 39.4). This suggests that endothelial cells may not express ␣2-subunit proteins (in spite of the presence of the appropriate mRNA). 4. DISCUSSION There have been conflicting reports in the literature about whether or not vascular smooth muscle is hyperpolarized by K⫹ in the absence of the endothelium (Andersson et al., 2000; Doughty et al., 2000; Lacy et al., 2000). This confusion has arisen because of the obligatory use of spasmogens in tension studies. The presence of phenylephrine stimulates K⫹ efflux from smooth muscle BKCa channels which elevates intercellular K⫹ concentrations preventing any further effect by exogenous K⫹ (Richards et al., 2001; Weston et al., 2002).

314

Gillian Edwards et al. e.c. B

1

1

10 10

myocyte 1

1

10 10 +

–

GAPDH Kir2.1 Kir2.2 Kir2.3

α1 α2 α3

Figure 39.3 Detection of cellular distribution of mRNA for KIR and Na⫹/K⫹ ATPase ␣-subunits in rat mesenteric arteries. RT-PCR was performed (in duplicate) with samples of RNA derived from single morphologically-identified cells or groups of ten such cells. Amplicons of the appropriate size indicated the presence of mRNA for Kir2.1 and Kir2.2 in myocytes and endothelial cells (e.c.) and of Kir2.3 in endothelial cells alone. Messenger RNA for Na⫹/K⫹ ATPase ␣1- and ␣3-subunits was detected in myocyte and endothelial cell samples, but ␣2-subunit mRNA was only present in the latter. Although there was some evidence for ␣2-subunit mRNA in one of the samples extracted from a group of ten myocytes, this was almost certainly due to endothelial cell contamination since the same sample (but not that for the other group of myocytes) reacted positively for von Willebrandt’s factor. RNA from kidney, heart or brain was used for positive controls (⫹) and negative controls (⫺) were derived from reactions performed without reverse transcription. Additional negative controls (B) were obtained by performing RT-PCR on a sample of bath solution using each set of gene-specific primers. The identity of all PCR products was confirmed later by sequencing. Detection of mRNA for GAPDH (a housekeeping gene) was included to provide an indication of the relative amounts of RNA in the initial cell samples.

In the rat hepatic artery, the hyperpolarization to K⫹ is partially inhibited by barium, which blocks KIR, or by ouabain, an inhibitor of Na⫹/K⫹-ATPase but is abolished in their joint presence (Edwards et al., 1998). The results presented here clearly demonstrate a role for both KIR and Na⫹/K⫹-ATPase in K⫹-induced smooth muscle hyperpolarization of rat mesenteric arteries without endothelium. A low concentration of 5 ⫻ 10⫺7M ouabain (sufficient to inhibit rat Na⫹/K⫹-ATPase comprising ␣2 or ␣3 but not ␣1 subunits; O’Brien et al., 1994; Blanco and Mercer, 1998) was without effect on the hyperpolarizing action of K⫹ on the smooth muscle of the mesenteric artery, although in the additional presence of barium the response to K⫹ was almost abolished. The concentration of barium which was used in the present study would be sufficient to block KIR and partially to inhibit current flow through the ATP-sensitive K⫹ channel, KATP (Nelson and Quayle, 1995). However, under the present conditions, glibenclamide, a selective inhibitor of KATP, has no effect on the mesenteric artery smooth muscle membrane potential

Roles of the inward-rectifier K⫹ channel and Na⫹/K⫹-ATPase 315 Kir 2.1

Na+/K+-ATPase

Kir 2.2

α1

α2

Kir 2.3

α3

Figure 39.4 Localization of KIR and Na⫹/K⫹ ATPase ␣-subunits by immunofluorescence labelling of cross-sections of rat mesenteric arteries. Kir2.1 and Kir2.3 antibody staining was distributed throughout the smooth muscle layer which would be consistent with one, or both, of these proteins forming the KIR channels. In contrast, Kir2.2 had a perinuclear distribution which suggests it does not form plasmalemmal K⫹ channels. Negative controls (preincubation of primary antibody with antigenic peptide) are shown as insets. Na⫹/K⫹ ATPase ␣1-, ␣2- and ␣3- labelling was present in the smooth muscle cell layers of rat mesenteric arteries. Negative controls (no primary antibody) are shown as insets. In all sections, DAPI-stained nuclei appear blue, anti-␣-subunit immunoreactivity appears red and the autofluorescence of the internal elastic lamina is green (see Color Plate 15).

and does not modify the response to K⫹ (Weston et al., 2002). Thus, the data provide a strong indication of a functional role for smooth muscle KIR in the response to K⫹. In previous studies, RT-PCR experiments have suggested the presence of mRNA for Kir2.1 (but not Kir2.2 or Kir2.3) in rat mesenteric artery myocytes (Bradley et al., 1999). In the present study there was good evidence from both antibody binding to sections and using RT-PCR for the presence of Kir2.1 in the smooth muscle. Furthermore, although K⫹induced vasodilation to K⫹ occurs in arteries from Kir2.2 knockout mice, it is not observed in those from Kir2.1 knockouts (Zaritsky et al., 2000), suggesting that Kir2.1 channels play the major role in this response. Evidence from the present study for any involvement of smooth muscle Kir2.2 or Kir2.3 was less conclusive. Although the myocytes possessed mRNA for Kir2.2 and the immunohistochemistry data were consistent with the expression of this protein, there was no evidence for any distribution of the protein within the plasmalemma. Thus, it is highly unlikely that Kir2.2 channels contribute to the cellular response to elevation of [K⫹]o. However, the possibility that Kir2.3 channels also form functional, homo- or heteromultimeric inwardly rectifying K⫹ channels in vascular smooth muscle cannot be excluded. Binding of the anti-Kir2.3 antibody to the plasmalemma was apparent in both the endothelium and smooth muscle in the rat mesenteric artery. Based on

316

Gillian Edwards et al.

the reported selectivity of the antibody (Stonehouse et al., 1999) it would appear that K⫹ channels composed at least partially of Kir2.3 ␣-subunits may be present in the latter. However, this observation was not substantiated by RT-PCR, which detected Kir2.3 mRNA in endothelial cells but not in myocytes. In order to perform the necessary single-cell RTPCR on microscopically identified cells of the appropriate phenotype, primers were designed to the 3⬘ UTR region of the gene. One explanation for the apparent presence of Kir2.3 protein but the absence of the necessary message is that alternative splicing of the gene in the 3⬘ UTR region gives rise to distinct endothelial cell and smooth muscle cell Kir2.3 mRNA. Re-designing PCR primers to a different region of the sequence may resolve this discrepancy. In the presence of phenylephrine, the hyperpolarization of the vascular smooth muscle to K⫹ in rat mesenteric arteries without endothelium can be attributed exclusively to stimulation of Na⫹/K⫹-ATPase since it is inhibited by a low concentration of ouabain and is unaffected by barium alone (Weston et al., 2002). In the rat, the effective concentration of ouabain (5 ⫻ 10⫺7 M) does not inhibit the house-keeping form of Na⫹/K⫹-ATPase which comprises ␣1-subunits (O’Brien et al., 1994; Blanco and Mercer, 1998; Weston et al., 2002). Thus, on the basis of its pharmacology, the response to elevation of K⫹ appeared to be due to stimulation of an isoform of Na⫹/K⫹-ATPase comprising ␣2- and/or ␣3-subunits. The results from the single-cell RT-PCR experiments suggested that the ␣2-subunit was not expressed in myocytes. Although one of the group samples of myocytes reacted positively for ␣2, this was almost certainly due to the presence of endothelial cell contamination since the same sample also indicated the presence of mRNA for the von Willebrandt factor (data not shown). This observation is consistent with a previous report (Juhaszova and Blaustein, 1997) which only found (immunohistochemical) evidence for ␣3- and no evidence for ␣2-subunits in rat mesenteric artery myocytes. Nevertheless, in the present study anti-␣2 antibodies did bind to the myocytes. Again, this discrepancy needs to be resolved by further characterization of the anti-␣2 antibody and the use of primers designed to a different region of the sequence. Irrespective of the outcome of these experiments, it is clear from the immunohistochemical and RT-PCR experiments that Na⫹/K⫹-ATPase ␣3-subunits are present in the smooth muscle of the rat mesenteric artery. The reported sensitivity of this isoform to ouabain is similar to that of the K⫹-induced hyperpolarization. The data are thus consistent with some role for Na⫹/K⫹-ATPase ␣3-subunits in the vascular response to elevation of extracellular K⫹. The depolarizing effect of barium on myocytes in the absence of phenylephrine indicates that inwardly rectifying K⫹ channels contribute to the resting membrane potential. However, the importance of the inwardly rectifying K⫹ channel in the hyperpolarizing response to K⫹ and, by implication, in the EDHF response, was reduced when the membrane was depolarized. The lack of effect of ouabain (5⫻10⫺7 M) on the resting membrane potential suggests that the ouabain-insensitive isoforms of Na⫹/K⫹-ATPase (comprising ␣2 and/or ␣3-subunits) are not active under resting conditions. Thus, at potentials close to the resting membrane potential both Na⫹/K⫹-ATPase and KIR channels are involved in the response to K⫹, whereas when the smooth muscle is depolarized the inwardly rectifying K⫹ channels (which are inhibited by intracellular blocking particles; see Nichols and Lopatin, 1997) no longer contribute. Stimulation of K⫹ efflux from myocytes via (iberiotoxin-sensitive) large-conductance calcium-sensitive K-channels by phenylephrine elevates [K⫹]o and prevents any further response to exogenous K⫹. The further depolarization of the membrane induced by ouabain in the presence of phenylephrine is consistent with this view and suggests that the contractile response to phenylephrine is normally tempered by the stimulation of ouabain-insensitive

Roles of the inward-rectifier K⫹ channel and Na⫹/K⫹-ATPase 317 forms of Na⫹/K⫹-ATPase. Thus it is concluded that both inward-rectifier K-channels and Na⫹/K⫹-ATPase contribute to K⫹-induced hyperpolarization in rat mesenteric artery smooth muscle and that the relative importance of each ion translocator in the response is determined by the membrane potential. This has critically important implications for the use of tension studies to determine the role of endothelium-derived K⫹ in the EDHF response. ACKNOWLEDGEMENTS This work was supported by the British Heart Foundation.

40 Importance of intracellular concentration of sodium in the relaxation of rat isolated mesenteric arteries by potassium Didier X.P. Brochet and Philip D. Langton The effect of increased extracellular potassium ([K⫹]o) on rat isolated mesenteric arteries, constricted with phenylephrine, was studied. An increase of [K⫹]o from 5.9 mM to 11.2 or 21.2 mM caused contraction of arteries with or without endothelium. Decreasing [K⫹]o from 5.9 mM to 1.2 mM for 3 min followed by increase in [K⫹]o to between 5.9 mM and a maximum of 41.2 mM, induced relaxation. Raising [K⫹]o from a range of concentrations, between 1.2 and 5.9 mM, to 13.8 mM resulted in a relaxation whose amplitude and duration was inversely proportional to the initial [K⫹]o. The relaxation induced by an increase of [K⫹]o from 1.2 to 13.8 mM was abolished by ouabain or by substituting extracellular sodium by choline or TRIZMA hydrochloride. The relaxations evoked by an increase of [K⫹]o from a low initial concentration may reflect a build-up of intracellular sodium and the subsequent upturn of electrogenic Na,K-ATPase pumping when extracellular potassium is increased. Thus, potassium-induced relaxation involves an ouabain-sensitive mechanism that is dependent on external sodium, but independent of the endothelium.

In addition to nitric oxide (NO) (Furchgott and Zawdski, 1980) and prostacyclin (Moncada and Vane, 1978), endothelium-derived hyperpolarizing factor (EDHF) appears to be an important mediator of relaxation in the rat isolated mesenteric artery (Chen et al., 1988). EDHF-mediated relaxation of constricted arteries is characterized as a relaxation to muscarinic agonists such as acetylcholine in the presence of inhibitors of NO and prostacyclin, causing hyperpolarization and relaxation of the overlying smooth muscle cells. Although EDHF-mediated relaxation is an important component of vasodilatation in many vascular beds, the identity of EDHF and its mechanism of action remain controversial. Potassium ions have been proposed as candidates-for EDHF (Edwards et al., 1998). The hyperpolarization of the overlying smooth muscle cells, produced by EDHF, is supposed to involve the opening of calcium-sensitive potassium channels. Potassium ions released from endothelial cells accumulate in the extracellular space between endothelial and smooth muscle cells which provokes hyperpolarization of vascular smooth muscle cells by activating the inwardly rectifying potassium conductance and the Na⫹,K⫹-ATPase. Thus, in mesenteric and hepatic artery of the rat, elevating extracellular potassium can mimic the effects of EDHF. There have been reports that potassium does not mimic the effects of EDHF in rat mesenteric (Doughty et al., 2000; Lacy et al., 2000), rat gastric arteries (Van de Voorde and Vanheel, 2000), guinea-pig arterioles (Coleman et al., 2001), guinea-pig carotid and porcine coronary arteries (Quignard et al., 1999). 1. METHODS Male Wistar rats (200–250 g) were killed by stunning followed by cervical dislocation. Third order superior mesenteric arteries were dissected in physiological saline solution containing

Intracellular sodium on rat isolated mesenteric arteries

319

(in mM): NaCl 89; NaHCO3 25; KCl 4.7; KH2PO4 1.18; MgSO4 1.2; CaCl2 1.8; glucose 11; sucrose 60; N␻-nitro-L-arginine methyl ester (L-NAME) 10⫺4 M; indomethacin 2.8⫻10⫺6 M. The pH was maintained at 7.4 by gassing with 95% O2/5% CO2.

1.1. Wire myography Segments of mesenteric artery were mounted at 37 ⬚C in a Mulvany–Halpern wire myograph for the recording of isometric tension in the same physiological saline solution as for the dissection. The segments were stretched until the wall tension was equivalent to a transmural pressure of 100 mmHg, the diameter was calculated and set to 90% of this value (Mulvany and Halpern, 1977). They were constricted with phenylephrine (10⫺7⫺10⫺5 M) and relaxed with acetylcholine (10⫺5 M) to test the presence of an endothelium. The response to elevating extracellular potassium from an initial concentration of 5.9 mM for 4 min during a phenylephrine-induced contraction (70% max) to a final concentration of 11.2 or 21.2 mM during 5 min was also tested. Where responses were required from vessels without endothelium, the endothelium was removed from the same vessels mounted in the myograph by rubbing a hair through the lumen. Relaxation to acetylcholine (10⫺5 M) was then tested again. The K⫹ concentration of the physiological saline solution was then reduced to 1.2 mM by isotonic replacement of KCl with NaCl. The standard protocol was a 4-min period of phenylephrine-induced contraction at 5.9 mM [K⫹]o. Extracellular potassium was then reduced to 1.2 mM for 3 min, after which time [K⫹]o was raised to 5.9, 11.2, 21.2, 31.2, 41.2, 51.2 or 61.2 mM for a final period of 7 min. Another protocol consisted of a 6-min exposure to a range of [K⫹]o from 1.2 to 5.9 mM with phenylephrine, followed by a 5-min exposure to 13.8 mM. The same protocol was used with ouabain (10⫺4 M). Ouabain was added at the beginning of the 6-min period of 1.2 mM [K⫹]o and was present during the 5-min test exposure to 13.8 mM [K⫹]o. Equimolar choline chloride or TRIZMA hydrochloride were substituted for NaCl only during the 6-min period of 1.2 mM [K⫹]o. Forces are expressed as percentage of the force immediately before the final potassium rise.

1.2. Drugs All drugs were made up as stock solutions in milli-Q water, except indomethacin which was dissolved in ethanol and applied in the buffer. All drugs were supplied by Sigma.

1.3. Statistics All data are expressed as mean values ⫾ SEM for ‘n’ experiments. Statistical significance was tested using an appropriate Student’s t-test. A P value less than 0.05 was taken as significant. 2. RESULTS In arteries with endothelium, 10⫺5 M acetylcholine relaxes phenylephrine-induced tone, whereas arteries without endothelium failed to relax to acetylcholine. In third-order superior rat mesenteric arteries with endothelium, raising [K⫹]o in the superfusing physiological saline solution from 5.9 mM (control) to 11.2 mM, by doubling the concentration of KCl, partially dilated (one third of the maximal dilatation) only 2 out

320

Didier X.P. Brochet and Philip D. Langton

Figure 40.1 (A) A representative example of an artery with an endothelium that did not dilate to potassium on stepping from normal (5.9 mM) to high (21.2 mM) [K⫹]o during a contraction to phenylephrine (PE). (B) Mean data (⫾ SEM; n ⫽ 9) of experiments on the effects of increase of [K⫹]o (from 5.9 mM to 11.2 or 21.2 mM) on segments of third-order mesenteric arteries with or without endothelium.

of 16 arteries. In the remaining 14 arteries, raising [K⫹]o produced a transient weak dilatation (n ⫽ 3) or a constriction (n ⫽ 11) (Figure 40.1). When the endothelium was removed, raising [K⫹]o produced only a constriction (n ⫽ 11). Raising superfusing [K⫹]o to 21.2 mM caused only constriction (n ⫽ 7) of arteries with endothelium when compared with 11.2 mM, whereas it produced no further effect (n ⫽ 6) on arteries without endothelium. There was no significant difference between arteries with and without endothelium after a rise in [K⫹]o (Figure 40.1). After a standard 4-min exposure to phenylephrine in control [K⫹], the [K⫹] in the physiological saline solution was lowered to 1.2 mM for a further 3 min. Subsequent restoration of [K⫹]o to 5.9 mM or higher concentrations (11.2, 21.2, 31.2 or 41.2 mM) resulted in a dilatation of all arteries tested (n ⫽ 22) (Figure 40.2). Restoring potassium to 5.9 mM caused a relaxation that could persist longer than 5 min (Figure 40.2). Only if potassium was raised to more than 41.2 mM did the relaxation disappear. Arteries with or without endothelium behaved similarly (data not shown). Raising [K⫹]o from a range of concentrations between 1.2 and 5.9 mM (1.2, 3, 3.5, 4, 4.6 and 5.9 mM) to 13.8 mM, after 6 min of phenylephrine-induced contraction resulted in a relaxation (Figure 40.3). The amplitude and duration of this relaxation was inversely related to the initial [K⫹]o (Figure 40.3). For instance, the amplitude of the relaxation was always more than 90% with an initial [K⫹]o of 1.2 mM (n ⫽ 16), whereas only 2 out of 16 arteries relaxed more than 90% with an initial [K⫹]o of 4.6 mM.

Figure 40.2 (A) A representative example of an artery that relaxes to a rise of [K⫹]o to 21.2 mM after a 4-min period of contraction to phenylephrine (PE) at 5.9 mM [K⫹]o followed by a 3-min period of 1.2 mM [K⫹]o. (B) Mean data (⫾ SEM; n ⫽ 22) from similar experiments using different concentrations of final [K⫹]o (5.9, 11.2, 21.2, 31.2, 41.2, 51.2, 61.2 mM).

Figure 40.3 (A) An example of an artery that relaxes from a 6-min period of phenylephrineinduced contraction at 1.2 mM [K⫹]o to a rise of [K⫹]o to 13.8 mM. (B) Mean data (⫾ SEM; n ⫽ 16) of the relaxations to a final [K⫹]o of 13.8 mM from experiments using a range of initial [K⫹]o between 1.2 and 5.9 mM. (C) Mean data (⫾ SEM; n ⫽ 16) of the length of relaxation to a final [K⫹]o of 13.8 mM from experiments using a range of initial [K⫹]o between 1.2 and 5.9 mM.

322

Didier X.P. Brochet and Philip D. Langton

Figure 40.4 Effect of ouabain (10⫺4 M) on relaxations of arteries when raising [K⫹]o from an initial 1.2 mM to a final 13.8 mM during a contraction to phenylephrine (0.5–2 ⫻ 10⫺6 M). Data shown as means ⫾ SEM (n ⫽ 18).

Ouabain of 10⫺4 M, added at the beginning of the 6-min period of 1.2mM [K⫹]o and present during the rise of [K⫹]o to 13.8mM, abolished the relaxation of the arteries exposed to phenylephrine (n⫽18) previously caused by raising [K⫹]o from 1.2 to 13.8mM (Figure 40.4). The effect of lowering extracellular sodium chloride (from 114 to 25 mM) substituted for choline chloride or TRIZMA hydrochloride was investigated. Reducing [Na⫹]o consistently reduced phenylephrine-induced contractions. Inclusion of Bay K8644 (10⫺6 M) increased phenylephrine-induced contractions towards control levels. After a 6-min exposure to phenylephrine with a [K⫹]o of 1.2 mM, [K⫹]o was raised to 13.8 mM. Substitution of extracellular sodium chloride abolished the relaxation (Figure 40.5). The inclusion of atropine did not affect the relaxation (data not shown). Bay K8644 on its own did not interfere with the relaxation to a rise of [K⫹]o from 1.2 mM to 13.8 mM (Figure 40.5). 3. DISCUSSION Increasing [K⫹]o from 5.9 mM to 11.2 or 21.2 mM did not relax rat mesenteric arteries. This may be explained by the fact that an increase of [K⫹]o tends to depolarize the cell membrane, increasing the calcium entry in the smooth muscle cells and so increasing the force of contraction (Beny and Schaad, 2000). The present finding differs from an earlier one in which hyperpolarization of the smooth muscle cells was observed when [K⫹]o was increased suggesting that potassium is an EDHF (Edwards et al., 1998). According to the proposed model, an increase of [K⫹]o should activate the inwardly rectifying potassium (Kir) channels and Na,K-ATPase that would together hyperpolarize the cell. Kir channels are present on the endothelium of rat mesenteric arteries but not in mesenteric artery myocytes (Nelson and Quayle, 1995; Hinton et al., 2000). It is unexpected then that there is no significant difference in the responses of arteries with or without an endothelium (even when obtaining a relaxation by using a 3-min period of 1.2 mM [K⫹]o). Thus, it seems that Kir channels do not play a key role in this relaxation (Prior et al., 1998; Jiang and Dusting, 2001). If, after a 4-min exposure to phenylephrine in control [K⫹]o, [K⫹]o is lowered to 1.2 mM for 3 min, and then raised back to between 5.9 and 41.2 mM, relaxation was observed. The only difference with the previous experiment is the 3-min period of low [K⫹]o. Lowering

Intracellular sodium on rat isolated mesenteric arteries

323

Figure 40.5 Effect of the substitution of extracellular sodium chloride by choline chloride and Bay K8644 (10⫺6 M) or TRIZMA hydrochloride and Bay K8644 and effect of Bay K8644 alone on the relaxation induced by a rise of [K⫹]o from a 6-min period of 1.2 mM [K⫹]o to 13.8 mM [K⫹]o during a contraction to phenylephrine (0.5–2 ⫻ 10⫺6 M). Data shown as means ⫾ SEM (n ⫽ 12).

[K⫹]o will reduce the activity of the Na,K-ATPase which has a K0.5 for [K⫹]o of about 1–1.6mM at physiological [Na⫹]i (Nakao and Gadsby, 1989; Nakamura et al., 1999). Over time this will increase [Na⫹]i and decrease [K⫹]i which will drive a subsequent upturn of the Na,K-ATPase when [K⫹]o is restored, leading to hyperpolarization and relaxation of the smooth muscle cells (Quignard et al., 1999; Lacy et al., 2000). At concentrations of [K⫹]o above 25 to 30 mM, the observed membrane potential is not different to the Nernst potential for potassium (Mulvany et al., 1982). Thus, the opening of K⫹ channels should not hyperpolarize vascular tissue. The observation that a transient relaxation occurs when the [K⫹]o is increased to 41.2 mM demonstrates the importance of Na,K-ATPase in hyperpolarizing the membrane potential of the smooth muscle cells. Both amplitude and duration of the relaxation during phenylephrine-induced force elicited by raising [K⫹]o to 13.8 mM from a range of concentrations between 1.2 and 5.9 mM, are inversely proportional to the initial [K⫹]o. This suggests that the relaxation may be proportional to the degree of depression of Na,K-ATPase activity. Direct evidence for the involvement of Na,K-ATPase pumping comes from the observation that relaxation induced by an increase of [K⫹]o from 1.2 to 13.8 mM is abolished by ouabain. The concentration of ouabain required to abolish this relaxation is relatively high (data not shown) and this may be explained by the fact that the ␣1 subunit of the Na,K-ATPase is relatively insensitive to ouabain, with an IC50 of about 10⫺5 M, whereas the IC50 for the ␣2 and ␣3 subunits is only 1–50 ⫻ 10⫺8 M (Sweadner, 1989; Therien et al., 1996). The relaxation can also be abolished by substituting NaCl by choline chloride or TRIZMA hydrochloride. Because choline chloride can interact with muscarinic receptors, atropine was used to prevent this potential effect. Bay K8644 increases the intracellular concentration of calcium by promoting the opening of high-voltage activated calcium channels, potentiating phenylephrine-induced contractions (Kanmura et al., 1988). The substitution of NaCl by choline chloride or TRIZMA hydrochloride will prevent an intracellular

324

Didier X.P. Brochet and Philip D. Langton

rise of Na⫹ when the activity of the Na,K-ATPase is decreased by 1.2 mM [K⫹]o. In smooth muscle cells of the guinea-pig mesenteric artery, the Na,K-ATPase is half maximally stimulated (K0.5) at about 22 mM [Na⫹]i (Nakamura et al., 1999). Under physiological conditions, when [Na⫹]i is between 8 and 12 mM, a rise of [Na⫹]i will be a very powerful activator of Na,K-ATPase. The K0.5 for [K⫹]o is about 1–1.6 mM at physiological [Na⫹]i (Nakao and Gadsby, 1989; Nakamura et al., 1999) which means that increments in [K⫹]o beyond the normal resting concentration of 4 mM only have a small effect on pump rate, whereas small increments in [Na⫹]i will have a profound effect. The effect of Na⫹ substitution suggests that accumulation of intracellular sodium when the activity of the Na,K-ATPase is reduced by exposure to low [K⫹]o may be a key factor in the relaxation induced by a subsequent rise of [K⫹]o. 4. CONCLUSION The relaxations evoked by an increase of [K⫹]o from a low initial concentration may reflect an accumulation of intracellular sodium and the subsequent upturn of electrogenic Na,K-ATPase pumping when extracellular potassium is subsequently increased. The outward current generated by pump upturn would tend to hyperpolarize the smooth muscle cells, lower the open probability of voltage-gated calcium channels, reduce the intracellular concentration of calcium and so lead to relaxation. Thus, potassium-induced relaxation involves a ouabain-sensitive mechanism that is dependent on external sodium, but independent of the endothelium. ACKNOWLEDGEMENTS This work was supported by a studentship funded by the Department of Physiology, University of Bristol.

41 Inhibition of bradykinin-induced relaxations by an epoxyeicosatrienoic acid antagonist: 14,15-epoxyeicosa5Z-monoenoic acid Kathryn M. Gauthier, Phillip F. Pratt, J.R. Falck and William B. Campbell Endothelium-dependent hyperpolarizations and relaxations of vascular smooth muscle induced by bradykinin are mediated by endothelium-derived hyperpolarizing factors (EDHFs). In bovine coronary arteries, epoxyeicosatrienoic acids (EETs) appear to function as EDHFs. A 14,15-EET analog, 14,15-epoxyeicosa-5Z-enoic acid (14,15-EEZE) has been characterized as an EET antagonist. Bovine coronary arterial rings were constricted with U46619 and concentration–relaxation curves to increasing concentrations of 14,15-EET (10⫺9–10⫺5 M) were obtained. 14,15-EET induced maximal relaxations averaging 80%. However, preincubation of the arteries with 14,15-EEZE (10⫺5 M) shifted the relaxation curve to the right and inhibited maximal relaxations by over 50%. Concentration–relaxation responses to bradykinin (10⫺12–10⫺7 M) were also evaluated. The bradykinin-induced relaxations were not altered by preincubation with miconazole (2 ⫻ 10⫺5 M) but were shifted to the right by preincubation with nitro-L-arginine methyl ester (L-NAME, 3 ⫻ 10⫺5 M) plus indomethacin (10⫺5 M). Incubation with miconazole plus L-NAME and indomethacin maximally inhibited relaxations by 50%. Similarly, incubation with 14,15-EEZE inhibited the indomethacin and L-NAME-resistant relaxations to bradykinin by over 50%. 14,15-EEZE did not alter relaxations to sodium nitroprusside (SNP, 10⫺9–10⫺6 M). In small cannulated and perfused bovine coronary arteries, incubation with 14,15-EEZE inhibited indomethacin- and L-nitroarginine- resistant dilatations to bradykinin (10⫺8 M). Thus, 14,15-EEZE appears to act as an EET antagonist by blocking relaxations to 14,15-EET but not the relaxations induced by SNP. Additionally, 14,15-EEZE inhibits the EDHF component of bradykinin-induced relaxations. These results further support the role of EETs as EDHFs.

1. INTRODUCTION Epoxyeicosatrienoic acids (EETs) are cytochrome P450 metabolites of arachidonic acid (Hecker et al., 1994; Campbell et al., 1996; Fisslthaler et al., 1999). The endothelium synthesizes and releases EETs in response to bradykinin, acetylcholine and arachidonic acid (Pinto et al., 1987; Fulton et al., 1992; Hecker et al., 1994; Campbell et al., 1996; Fisslthaler et al., 1999). In the coronary circulation, EETs diffuse to the smooth muscle and activate large-conductance, calcium-activated potassium channels (BKCa) of the cell membrane to induce hyperpolarization and relaxation (Campbell et al., 1996). Therefore, EETs function as endothelium-derived hyperpolarizing factors (EDHFs). To study the role of endogenous EETs in agonist-induced endothelium-dependent relaxations, inhibitors of cytochrome P450 have been used with varying results. In some studies, these agents block the relaxations to bradykinin and acetylcholine, while in other studies they are without effect (Pinto et al., 1987; Fulton et al., 1992; Hecker et al., 1994; Campbell et al., 1996; Corriu et al., 1996;

326

Kathryn M. Gauthier et al.

Fukao et al., 1997; Gauthier-Rein and Rusch, 1998; Vanheel and Van de Voorde, 1997). Non-specific vascular effects of cytochrome P450 inhibitors have been documented. In rat mesenteric arteries, cytochrome P450 inhibitors reduced relaxations and/or hyperpolarizations of the smooth muscle induced by the potassium channel openers, cromakalim or pinacidil (Fukao et al., 1997; Van de Voorde and Vanheel, 1997; Vanheel and Van de Voorde, 1997). In the portal vein of the ferret, the cytochrome P450 inhibitor, clotrimazole, directly inhibited large-conductance, calcium-activated potassium currents, whereas ketoconazole increased these same currents (Rittenhouse et al., 1997a,b). Therefore, non-specific actions of P450 inhibitors may modify vascular responses. The endothelium metabolizes arachidonic acid to the vasodilator EETs via the cytochrome P450 2C or 2J epoxygenases (Lin et al., 1996; Rosolowsky and Campbell, 1996; Fisslthaler et al., 1999; Node et al., 1999). Alternatively, vascular smooth muscle metabolizes arachidonic acid by the cytochrome P450 4A ␻-hydroxylase to 20-hydroxyeicosatetraenoic acid (20-HETE) (Ma et al., 1993; Harder et al., 1994; Gebremedhin et al., 2000). 20-HETE acts as an endogenous vascular constrictor and inhibits BKCa channels of smooth muscle (Zou et al., 1996). In the renal, mesenteric and cerebral arterial circulation, 20-HETE contributes to myogenic tone and regulates blood flow (Kauser et al., 1991; Ma et al., 1993; Harder et al., 1994; Gebremedhin et al., 2000). Inhibitors of cytochrome P450 enzymes will block the endothelial cell epoxygenases that produce EETs as well as the smooth muscle ␻-hydroxylase that produces 20-HETE (Figure 41.1) (Kauser et al., 1991). Therefore the effect of cytochrome P450 inhibitors will depend on which pathway of arachidonic acid metabolism predominates. Variability of arachidonic acid metabolism between species and vascular beds may account for the observed conflicting effects of cytochrome P450 inhibitors. Consequently, to determine the role of endogenous EETs in vascular relaxation, a pharmacological tool is required that selectively inhibits the action or synthesis of EETs, but not the action or synthesis of 20-HETE. Therefore, the 14,15-EET

Figure 41.1 Schematic of EET and 20-HETE regulation of vascular tone and the possible interactions with cytochrome P450 inhibitors and the 14,15-EET analog, 14,15-EEZE.

Inhibition of bradykinin-induced relaxations

327

analog, 14,15-epoxyeicosa-5(Z)-enoic acid (14,15-EEZE) was synthesized and tested for its ability to antagonize vasodilator activity. 2. METHODS

2.1. Vascular reactivity studies Bovine hearts were obtained from a local slaughterhouse. Sections of the left anterior descending coronary artery were dissected, cleaned of adhering fat and connective tissue and cut into 1.5- to 2-mm-diameter rings (3-mm length) with care taken not to damage the endothelium. The arterial rings were suspended in a tissue bath containing Kreb’s-bicarbonate buffer containing the following (in mmol/L): NaCl 118, NaHCO3 24, KCl 4.8, CaCl2 3.2, KH2PO4 1.2, MgSO4 1.2, glucose 11, EDTA 0.03. The Kreb’s buffer was equilibrated with 95% O2–5% CO2 and maintained at 37 ⬚C. Isometric tension was measured with forcedisplacement transducers (Grass), amplifiers (AD Instruments ETH-400) and recorded on a computer (Macintosh) using MacLab 8e software. Arterial rings were slowly stretched at 0.5 grams increments to a tension of 3.5 grams and allowed to equilibrate for 1 h. After equilibration, KCl (60 mmol/L) was added repeatedly and rinsed until reproducible stable contractions were observed. The thromboxane mimetic, U46619 (2 ⫻ 10⫺8 M) was added to increase tension to approximately 50–75% of maximal KCl contraction. Cumulative concentrations of EETs or SNP were added. The vessels were then rinsed, treated with 14,15-EEZE and the concentration–response repeated. Similar studies were performed with bradykinin as the agonist in rings pretreated with indomethacin (10⫺5 M) and nitro-L-arginine methyl ester (L-NAME, 3 ⫻ 10⫺5 M) or miconazole (2 ⫻ 10⫺5 M). Results are expressed as % relaxation of the response to U46619 with 100% relaxation representing basal tension.

2.2. Relaxations of small cannulated bovine coronary arteries Small coronary arteries were microdissected from the left ventricle and immediately cannulated on tapered glass micropipettes in a heated (37 ⬚C) lucite perfusion-superfusion chamber in a solution of the following composition (in mmol/L): NaCl 119, KCl 4.7, CaCl2 1.6, MgSO4 1.17, glucose 5.5, NaHCO3 24, NaH2PO4 1.18 and EDTA 0.026. Arteries were maintained at constant perfusion pressure of 60 mmHg and were equilibrated with 21% O2–5% CO2–74% N2 to maintain a pH of 7.4 and a PO2 of 140 mmHg. Diameters were measured using a Nikon SMZ-800 inverted microscope, Spot RT camera (Diagnostic Instruments, Inc.) with images captured and analyzed using Spot/Metaview acquisition/analysis/graphics software. All measurements were performed with U46619 (2 ⫻ 10⫺8 M) in the superfusion solution. Vehicle or 14,15-EEZE (3 ⫻ 10⫺6 M) and indomethacin (10⫺5 M) and L-nitroarginine (3 ⫻ 10⫺5 M) were present in perfusion and superfusion solutions. Bradykinin (10⫺8 M) was added to the superfusate solution. Diameters were measured either as a control experiment with vehicle (0.03%), vehicle with bradykinin (10⫺8 M), 14,15-EEZE (3 ⫻ 10⫺6 M), or 14,15-EEZE with bradykinin.

2.3. Drugs and chemicals Bradykinin, indomethacin, L-nitroarginine, miconazole and sodium nitroprusside were purchased from Sigma Chemical Co. (St Louis, MO, USA). Bradykinin, sodium nitroprusside, and L-nitroarginine were mixed to their appropriate stock concentrations in water.

328

Kathryn M. Gauthier et al.

Miconazole, indomethacin and 14,15-EET were prepared as 10⫺2 M stocks in 95% ethanol. 14,15-EET was synthesized by the method of Corey et al., 1979 and 14,15-EEZE by the method of Falck et al., 1985. 3. RESULTS

3.1. Effect of 14,15-EEZE on EET-induced relaxations The major cytochrome P450, arachidonic acid metabolite in bovine coronary arteries is 14,15-EET (Campbell et al., 1996; Pratt et al., 1996). 14,15-EET causes concentrationrelated relaxations of arteries contracted with U46619, which average 80–90% at 10⫺5 M (Figure 41.2). Treatment of this artery with 14,15-EEZE (10⫺5 M) inhibited the maximal relaxations by 50%. 14,15-EEZE also inhibited relaxations to 11,12-, 8,9- and 5,6-EET (Gauthier et al., 2002). Concentration-dependent relaxations to the NO donor, sodium nitroprusside were not altered by incubation with 14,15-EEZE (10⫺5 M) (Figure 41.3). 14,15EEZE also failed to alter the relaxations to iloprost, bimakalim and NS1619 (Gauthier et al., 2002). 14,15-EEZE did not significantly alter the constriction to U46619 and 20-HETE (Gauthier et al., 2002). Therefore, the inhibition induced by 14,15-EEZE appears to be specific.

3.2. Effect of 14,15-EEZE on agonist-induced relaxations of coronary arterial rings In U46619-contracted coronary arterial rings, bradykinin induced concentration-dependent relaxations, which were shifted to the right by incubation with nitro-L-arginine methyl ester (L-NAME, 3 ⫻ 10⫺5 M) plus indomethacin (10⫺5 M)(Figure 41.4). Incubation with miconazole (2⫻ 10⫺5 M) alone did not significantly alter the bradykinin-induced relaxations. However, the combination of miconazole with L-NAME and indomethacin shifted the curve to the right and blocked the maximal relaxations by approximately 50%. These results show that bradykinin stimulates the production of a cytochrome P450-dependent relaxing factor

Figure 41.2 Original tracings showing the effect of 14,15-EEZE (10⫺5 M) on 14,15-EET-induced relaxations of a bovine coronary artery. The artery was contracted with U46619. Increasing concentrations of 14,15-EET were added and changes in isometric tension were recorded either under control conditions (top trace) or with the artery treated with 14,15-EEZE (bottom trace). g ⫽ grams of tension.

Inhibition of bradykinin-induced relaxations

329

Figure 41.3 Original tracings showing the effect of 14,15-EEZE (10⫺5 M) on sodium nitroprusside (SNP)-induced relaxations of a bovine coronary artery. The artery was contracted with U46619. Increasing concentrations of SNP were added and changes in isometric tension were recorded either under control conditions (top trace) or with the artery treated with 14,15-EEZE (bottom trace). g ⫽ grams of tension.

Figure 41.4 Effect of L-nitroarginine methyl ester (L-NAME, 3 ⫻ 10⫺5 M), indomethacin (indo, 10⫺5 M) and miconazole (2 ⫻ 10⫺5 M) on bradykinin-induced relaxations of bovine coronary arteries. Arterial segments were contracted with U46619 (2 ⫻ 10⫺8 M). Changes in isometric tension were measured. Data shown as mean ⫾ SEM, n ⫽ 10–14.

in bovine coronary arteries which is unmasked when the synthesis of NO and prostacyclin is inhibited. The ability of 14,15-EEZE to block the non-NO, non-prostacylin component of bradykinininduced relaxations was evaluated next. In coronary arteries incubated with L-NA and indomethacin, 14,15-EEZE inhibited the concentration-dependent relaxations to bradykinin at all concentrations tested (Figure 41.5). Therefore, the EDHF component of bradykinininduced relaxations is inhibited by 14,15-EEZE.

3.3. Effect of 14,15-EEZE on bradykinin-induced dilatations of small bovine coronary arteries Small, bovine coronary arteries were cannulated, perfused and treated with indomethacin and L-NA. The arteries were constricted with U46619 (2 ⫻ 10⫺8 M). The addition of bradykinin (10⫺8 M) to a U46619-constricted artery induced a dilatation (from 65 to 195␮m) (Figure 41.6). Incubation with 14,15-EEZE (3 ⫻ 10⫺6 M) did not alter the U46619-induced

330

Kathryn M. Gauthier et al.

Figure 41.5 Original tracings showing the effect of 14,15-EEZE (10⫺5 M) on bradykinininduced relaxations of a bovine coronary artery. The artery was treated with L-nitroarginine (3 ⫻ 10⫺5 M) and indomethacin (10⫺5 M) and contracted with U46619 (2 ⫻ 10⫺8 M). Increasing concentrations of bradykinin were added and changes in isometric tension were recorded either under control conditions (top trace) or with the artery treated with 14,15-EEZE (bottom trace). g ⫽ grams of tension. Control

14, 15-EEZE

U46619

U46619 + Bradykinin 100 µm

Figure 41.6 Photomicrographs of a cannulated bovine coronary artery showing the effects of 14,15-EEZE (3 ⫻ 10⫺6 M) on dilatations induced by bradykinin (10⫺8 M). The artery was constricted with U46619 (2 ⫻ 10⫺8 M) and bradykinin was added. Diameter changes were measured either under control conditions (left column) or in the presence of 14,15-EEZE (right column).

constriction (70 ␮m), but nearly eliminated the dilatation to bradykinin (20 ␮m). Similarly, 14,15-EEZE inhibited bradykinin-induced hyperpolarizations of the smooth muscle cells of these small coronary arteries (Gauthier et al., 2002). 4. DISCUSSION In bovine coronary arteries, EETs function as EDHFs (Fulton et al., 1995; Campbell et al., 1996; Popp et al., 1996; Gebremedhin et al., 1998; Hayabachi et al., 1998; Fisslthaler et al., 1999). They are synthesized by the endothelium and diffuse to the smooth muscle to activate

Inhibition of bradykinin-induced relaxations

331

BKCa channels which results in hyperpolarization and relaxation (Campbell et al., 1996; Li and Campbell., 1997; Hayabachi et al., 1998). All four EET regioisomers are synthesized by the coronary endothelium; however 14,15-EET is synthesized in the highest concentration (Pratt et al., 1996). 14,15-EEZE inhibits relaxations induced by 14,15-, 11,12-, 8,9- and 5,6EET (Gauthier et al., 2002). It is most effective in inhibiting 14,15-EET-induced relaxations (Gauthier et al., 2002). This inhibition was specific for the EETs since 14,15-EEZE did not alter the relaxations to sodium nitroprusside, iloprost, NS1619, or bimakalim, or the constrictions to U46619 or 20-HETE (Gauthier et al., 2002). In L-NA and indomethacintreated arterial rings from this study, bradykinin induced a concentration-dependent relaxation which is dependent upon cytochrome P450 metabolism because the relaxations were blocked by miconazole. These relaxations were also inhibited to a similar extent by 14,15-EEZE. Therefore, 14,15-EEZE blocks EET-induced relaxations as well as bradykinininduced relaxations that are dependent on cytochrome P450 metabolism. The antagonism by 14,15-EEZE of the relaxation does not occur through the inhibition of EET synthesis and/or stimulation of 20-HETE synthesis (Gauthier et al., 2002). In rat cortical renal microsomes, 14,15-EEZE did not alter 20-HETE production. It increased EET concentrations while simultaneously decreasing the corresponding dihydroxyeicosatrienoic acid (DHET) concentrations possibly by inhibiting epoxide hydrolase, which converts EETs to DHETs (Gauthier et al., 2002). These results suggest that 14,15-EEZE acts at the level of the smooth muscle to inhibit EET-induced actions and not at the endothelium to inhibit its release. 14,15-EEZE differs from 14,15-EET in that the double bonds at carbons, 8,9 and 11,12 are saturated. 14,15-EEZE has little agonist activity. It relaxed constricted bovine coronary rings by a maximum of 21% at 10⫺5 M (Gauthier et al., 2002). These findings demonstrate that chemical manipulation of the EET molecule can alter agonist activity and confer antagonist properties. The requirement for a distinct molecular structure supports the theory of a vascular EET binding site(s) or receptor(s). High-affinity binding of 14,15-EET has been shown in guinea pig mononuclear cells (Wong et al., 2000). It is therefore possible that 14,15-EET induces its effects through a specific binding site on the coronary smooth muscle. Inhibitors of cytochrome P450 enzymes block the endothelial cell cytochrome P450 epoxygenase, which produce EETs as well as the smooth muscle cytochrome P450 ␻-hydroxylase, which produces 20-HETE (Ortiz de Montellano et al., 1981; Capdevila et al., 1988; Harder et al., 1995) To specifically evaluate the vascular contributions of EETs an inhibitor that selectively blocks the epoxygenase pathway is needed. The present study shows that the 14,15-EET analog, 14,15-EEZE, inhibits EET-induced relaxations as well as the EDHF component of bradykinin-induced relaxations in bovine coronary arteries and therefore represents an EET antagonist. This and other EET analogs will provide useful pharmacological tools required for the further examination of the biological effects of EETs. ACKNOWLEDGMENTS This research was supported by grants from the National Institutes of Health (HL-51055 and GM-31278) and the Robert A. Welch Foundation. K. Gauthier is a post-doctoral fellow of the American Heart Association-Northland Affiliate and was supported by the National Institutes of Health, Training Grant (HL-07792). The authors thank Mrs Gretchen Barg for secretarial assistance.

42 Local release of EDHF initiates a conducted dilatation, but is not the upstream mediator in arterioles of the hamster Cor de Wit, Bernd Hoepfl, Steffen-Sebastian Bolz and Ulrich Pohl Vasomotor reactions upon focal stimulation of arterioles are conducted along the vascular wall. This study aimed to identify which endothelial autacoid(s) act as mediators of the local and conducted dilator responses. Hitherto, arterioles in the hamster cremaster microcirculation were locally stimulated with acetylcholine or sodium nitroprusside and the resulting changes in diameter were measured using videomicroscopy at the site of application and up to 1.4 mm upstream at distant sites. Experiments were also performed after blockade of NO-synthase, cyclooxygenase, P450-monooxygenase or K⫹-channels. Dilatations in response to acetylcholine, but not sodium nitroprusside, were conducted rapidly to upstream sites with diminished amplitude. Maximal amplitudes of acetylcholine-induced dilatations were not attenuated by inhibition of NO-synthase and cyclooxygenase. However, P450-monooxygenase blockers (sulfaphenazole or ODYA) attenuated dilatations at local and distant sites. Charybdotoxin or iberiotoxin attenuated dilatations at the local and remote site after focal application at the stimulation site. In contrast, treatment of the upstream site with these blockers was without any effect. It is concluded that upon local stimulation with acetylcholine a P450monooxygenase product is generated, which induces local dilatation via the activation of Ca2⫹-dependent K⫹-channels and initiates conduction of the dilatation. In contrast to the local site, neither activation of these K⫹-channels nor the synthesis of nitric oxide (NO) or prostaglandins are necessary to dilate the arterioles at remote, distant sites. This suggests that endothelium-derived hyperpolarizing factor (EDHF) serves as an important mediator to initiate conducted dilatations and may act as a key player in the coordination of arteriolar behaviour in the microcirculatory network.

1. INTRODUCTION To match tissue needs, blood flow is regulated in a wide dynamic range. For instance, during exercise blood flow to skeletal muscle can increase more than 20-fold of its resting level. Such large increases of flow can only be achieved by an overall dilatation of resistance vessels and their feeding arteries. An upstream dilatation that accompanies that of downstream arterioles can potentially be accomplished by the NO-mediated dilatation following an increase of shear stress exerted by augmented blood flow (de Wit et al., 1997). Other mechanisms beside the ‘flow-induced dilatation’ may contribute to the coordination of vessel behaviour in the arteriolar tree. Of potential relevance in this connection is the conduction of vasomotor signals along the cells of the vascular wall, which may add to the coordination of arteriolar behaviour within and throughout the vascular segments (Christ et al., 1996). To study the conduction of vasomotor signals in the microcirculation, arterioles are usually stimulated by focal, transient application of various vasoactive compounds. Upon such focal stimulation with certain compounds changes of arteriolar diameter cannot only

Local release of EDHF initiates a conducted dilatation

333

be observed at the site of stimulation, but also several millimetres upstream and downstream of the application site (Segal and Duling, 1986; Doyle and Duling, 1997). The vasomotor signal travels at a very high velocity along the vascular wall and therefore it was assumed that the signal transmitted is a change of the membrane potential of vascular cells. In fact, upon focal stimulation locally induced hyperpolarizations and depolarizations have been measured not only at the site of stimulation but also at upstream, remote locations (Xia and Duling, 1995; Welsh and Segal, 1998). Therefore, it is generally accepted that in the microcirculation and small conduit arteries, the conduction of signals leading to vasodilatation or vasoconstriction at remote sites reflects the initiation and spread of electrical signals that travel along endothelial and/or vascular smooth muscle cells. In different tissues and animals focal application of a acetylcholine induces such a remote dilatation (Segal and Duling, 1986; Welsh and Segal, 1998; Bartlett and Segal, 2000; de Wit et al., 2000). Acetylcholine acts by releasing different endothelial autacoids, namely nitric oxide (NO), prostaglandins and a hyperpolarizing factor (EDHF) (Ignarro et al., 1987a; Palmer et al., 1987; Campbell et al., 1996; Fisslthaler et al., 1999; Bolz et al., 2000). However, it is not clear whether autacoids and which one in particular is the key mediator to initiate a propagated response in the microcirculation of the cremaster of hamsters. Since EDHF, by definition, acts by a change of the membrane potential, it is hypothesized that EDHF is the crucial factor to initiate a dilatation in response to acetylcholine that conducts along the arteriole. This hypothesis was tested by studying the effect of sequential inhibition of the synthesis of NO, prostaglandins and EDHF on local and remote dilatations initiated by acetylcholine. If indeed EDHF evokes the local signal that is transmitted along the vascular wall, it does not necessarily imply that EDHF itself is also the mediator of the dilatation at distant sites. This distant dilatation could be induced either by a hyperpolarization per se or by a secondary release of autacoids. To address this issue, adequate blockers of K⫹-channels were applied in a focal manner, either at the site of acetylcholine stimulation or at distant sites. 2. METHODS

2.1. Animal preparation The care of the animals and the conduct of the experiments were in accordance with the rules of the German animal protection law. Male golden syrian hamsters (80–150 g body wt) were anaesthetized by intraperitoneal injection of pentobarbital sodium (75 mg/kg). After cannulation of the right jugular vein with a polyethylene catheter, anaesthesia was maintained by infusion of a combination of droperidol (0.1 mg/kg), fentanyl (0.1 mg/kg) and midazolam (2 mg/kg) at a rate of 0.2–0.3 ml/min/kg. Arterial pressure was measured continuously via a catheter placed in the right carotid artery by means of a pressure transducer (Statham, Costa Mesa, California). Data were sampled at a rate of 2 Hz by an analog-digital board, processed with a data acquisition system (ONLINE), and stored on computer disc for later analysis. The animals were ventilated artificially (7025 Rodent Ventilator, Hugo Sachs Elektronik, Freiburg, Germany) to maintain PO2 and PCO2 at physiological values (~85 and 40 mmHg, respectively). The right cremaster muscle was prepared for intravital microscopy.

2.2. Experimental setup The cremaster muscle was superfused with warm (34 ⬚C) bicarbonate-buffered salt solution at a rate of 8 ml/min. The superfusion fluid had a pH of 7.4, a PO2 of ~30 mmHg and a PCO2 of ~38 mmHg as measured in samples taken from the surface of the muscle. One or

334

Cor de Wit et al.

two arterioles were studied in each animal and monitored by means of a microscope (Metallux, Leitz, Wetzlar, Germany) equipped with a video camera (CCD, Computer Optics, Weilheim, Germany). The microscopic images were acquired using a 32⫻ objective (numerical aperture: 0.40), and displayed on a video monitor at a 1000-fold magnification as well as recorded on videotape (S-VHS, Sony) for offline measurements of luminal diameters (video dimension analyser, IPM, California).

2.3. Experimental protocols After surgery, the preparation was allowed to stabilize from surgery for 30min before starting an experimental protocol. In order to study the conduction of vascular responses, a micropipette was positioned in the vicinity of an arteriole. In some experiments the pipette was repositioned and placed directly into the tissues to check for diffusion and/or convection of the vasoactive substance after pressure ejection. Pipettes were pulled, using a BrownFlaming puller (Model P-97, Sutter, USA), from borosilikate glass (Hilgenberg, Germany, outer diameter 1 mm, inner diameter 0.5 mm). The tip opening was 1–2 ␮m. Acetylcholine (10⫺2 M) or sodium nitroprusside (10⫺2 M) were applied by a pressure pulse. If a response at the site of stimulation (local) was obtained, the same pulse stimulation was used and vasomotor responses were studied at sites located between 0.67 and 1.40 mm upstream. Thereafter, NO-synthase and cyclooxygenase were blocked by adding a combination of NGnitro-L-arginine (L-NA, 3 ⫻ 10⫺5 M) and indomethacin (3 ⫻ 10⫺6 M) to the superfusion fluid 30 min before restudying the response. In additional groups aimed to study the role of EDHF, control dilatations in response to acetylcholine at the local and a upstream site (0.67mm) were obtained in the presence of L-NA and indomethacin. Thereafter, sulfaphenazole (10⫺5 M) or 17-octadecynoic acid (ODYA, 5 ⫻ 10⫺5 M) was added to the superfusion fluid. While the blocker of the P450-monooxygenase sulphaphenzole was continuously present, the suicide inhibitor of P450 monooxygenases, ODYA (Zou et al., 1994), was added to the superfusion fluid for 30 min and washed out before reexamination of the arteriolar responses. In other experimental groups, dilatations were studied before and after application of specific blockers of Ca2⫹-dependent K⫹-channels (charybdotoxin or iberiotoxin) (Gribkoff et al., 1996). These blockers were applied via a micropipette either at the site of acetylcholine stimulation (local) or at the remote site. The topical application of these blockers was done by repeated pressure ejections (130 kPa, 200 ms, 10–20 times). The efficacy of the K⫹-channel blockade was verified by checking the dilatation in response to acetylcholine applied locally at the same location. The experiments lasted typically between 3 and 6 h. The maximal diameter of the investigated arterioles was measured by the superfusion of a combination of different vasodilators (adenosine 10⫺4 M, sodium nitroprusside 10⫺5 M) at the end of the experimental protocol and the animal was killed by an overdose of anaesthesia.

2.4. Solutions and drugs The salt buffer used for superfusion had the following composition (in mmol/L): Na⫹ 143, 2⫺ ⫺ K⫹ 6, Ca2⫹ 2.5, Mg2⫹ 1.2, Cl⫺ 128, HCO⫺ 3 25, SO4 1.2, and H2PO4 1.2. Acetylcholine, sodium nitroprusside, charybdotoxin, iberiotoxin, and sulfaphenazole were purchased from Sigma (Deisenhofen, Germany). Sodium nitroprusside was dissolved in 10⫺3 M Na-acetate on the day of experiment and stored in the dark. Stock solutions of charybdotoxin and iberiotoxin were prepared in water and stored at ⫺20 ⬚C until use. Sulfaphenazole was dissolved in 60% ethanol at a concentration of 10⫺2 M on the day of experiment and further diluted

Local release of EDHF initiates a conducted dilatation

335

using the superfusion buffer. For all other solutions and further dilutions freshly prepared superfusion buffer was used.

2.5. Statistics and calculations Vascular tone is expressed as a quotient of the vessel’s resting diameter divided by its maximal diameter. Changes of the inner diameter of the vessels were normalized to the maximal possible constriction or dilatation according to the relationship: % of maximal response ⫽ (DTr ⫺ DCo)/(DM ⫺ DCo) ⫻ 100 where DTr represents the diameter observed after treatment and DCo the control diameter before treatment. DM represents (for dilator responses) the diameter at maximal dilatation or (for constrictions) the minimal luminal diameter (zero). This normalization allows the comparison of vessels of different tone and size. Comparisons within groups were performed using Student’s paired t-tests and, for multiple comparisons, P values were corrected according to Bonferroni. Data between groups were compared by analysis of variance followed by post hoc analysis of the means. Differences were considered significant at a corrected error probability of P less than 0.05. Means ⫾ SEM of all data are presented. 3. RESULTS

3.1. Basal data A total of 44 arterioles were studied in 39 hamsters. The animals exhibited a mean arterial pressure of 69 ⫾ 2 mmHg. The arterioles studied were of varying size and branching order and their maximal diameter ranged from 25 to 72 ␮m (mean value: 46 ⫾ 2 ␮m). The resting tone of the arterioles amounted to 0.62 ⫾ 0.05 in untreated preparations. Arteriolar tone in vessels treated from the start of the experiment with L-NA and indomethacin amounted to 0.44 ⫾ 0.02 and was significantly higher than in untreated arterioles (P ⬍ 0.05).

3.2. Local stimulation and conduction of vasodilator responses Upon short local stimulation with a bolus of the endothelium-dependent vasodilator acetylcholine applied via the micropipette (130 kPa for 20 –100 ms) the arterioles in untreated preparations dilated to a maximum peak of 71 ⫾ 12% within 16 ⫾ 5 s at the stimulation site (local, n ⫽ 5). The dilatation was transient and the blood vessels gradually returned to their initial resting diameter, which was reattained within 83 ⫾ 4 s. This local dilatation was conducted rapidly to distant upstream sites. However, the maximal amplitude was diminished to 35 ⫾ 7% at a distance of 0.77 mm (P ⬍ 0.05 vs local site) and this smaller peak was reached earlier (7 ⫾ 1 s, P ⬍ 0.05 vs local site). If the stimulation pipette was placed in the tissue at a distance of 0.67 mm from the arteriole, ejection of acetylcholine led only to a small dilatation of 6 ⫾ 2%. Like acetylcholine, the application of the exogenous NO-donor sodium nitroprusside (n ⫽ 5) induced a local dilatation of a similar amplitude (71 ⫾ 7%). Although the local dilatation upon sodium nitroprusside in the same arterioles was of a similar amplitude, only a small dilatation of 5 ⫾ 2% (P ⫽ 0.07) was observed at upstream sites. This small peak was reached only after 25 ⫾ 5 s (P ⬍ 0.05 vs acetylcholine).

336

Cor de Wit et al.

Figure 42.1 Absolute changes in diameter upon focal application of acetylcholine at the local and remote site (0.67 mm) before (control, white bars) and after inhibition of NO-synthase (NG-nitro-L-arginine, 3 ⫻ 10⫺5 M) and cyclooxygenase (indomethacin, 3 ⫻ 10⫺6 M) (L-NA ⫹ indo, black bars) are depicted. The maximal dilator amplitude observed in response to acetylcholine at the local and at the remote site remained unaffected by L-NA and indomethacin (n ⫽ 4).

3.3. Role of endothelial autacoids in the initiation of conducted vasodilatations Compared to the respective controls obtained in the same blood vessels, the maximal amplitude of the dilatation upon application of acetylcholine was not reduced at the local site (58⫾ 6% vs 52 ⫾ 3%, control vs L-NA ⫹ indomethacin, respectively) or at the distant site (0.67 mm: 40 ⫾ 6% vs 36 ⫾ 8%) in the presence of L-NA and indomethacin (Figure 42.1). Only the sustained phase of the local dilatation was diminished after L-NA and indomethacin (data not shown). In further experiments, a blocker of the P450-monooxygenase, sulfaphenazole (10⫺5 M), was applied to the superfusion. In the presence of L-NA and indomethacin, sulfaphenazole attenuated the dilatation in response to acetylcholine at the local site. The peak dilatation was reduced from 62 ⫾ 9% to a small remaining amplitude of 17 ⫾ 5%. The dilatation at the remote site was virtually abrogated after sulfaphenazole (at 0.67 mm: 4 ⫾ 3%, P ⫽ 0.20, Figure 42.2). In other animals (n ⫽ 6 arterioles in 4 animals) a chemically different blocker of P450-monooxygenases was used instead. In this series, the microcirculation was incubated with the suicide inhibitor ODYA (5 ⫻ 10⫺5 M) for 30 min. This incubation reduced the peak dilatation in response to acetylcholine at the local site from 76 ⫾ 4% to 15 ⫾ 4% and at the remote site from 61 ⫾ 4% to 20 ⫾ 5% (P ⬍ 0.05, Figure 42.2). Nevertheless, the superfusion of a combination of sodium nitroprusside and adenosine still induced a nearly maximal dilatation (85 ⫾ 3%).

3.4. Site of action of EDHF during conducted vasodilatations A second micropipette was used to apply specific blockers of Ca2⫹-dependent K⫹-channels in a focal manner. All these experiments were performed in the continuous presence of L-NA and indomethacin. The application of charybdotoxin at the site of acetylcholine

Local release of EDHF initiates a conducted dilatation

337

Figure 42.2 Effect of P450-monooxygenase blockers on dilatations induced by acetylcholine at the local and remote site in the absence of NO and prostaglandins (L-NA ⫹ indo). Sulfaphenazole (10⫺5 M, sulfa) attenuated the dilatation in response to acetylcholine at the local site and abolished it at the upstream site. Likewise after incubation with ODYA (5 ⫻ 10⫺5 M) the dilatations were reduced at both sites. * indicates significant difference as compared to L-NA and indomethacin (P ⬍ 0.05). Absolute maximal diameter changes are depicted.

stimulation attenuated the peak dilatation from 46 ⫾ 9% to 19 ⫾ 5% (P ⬍ 0.05, n ⫽ 4). Although the remote site was left untreated, the dilatation was also attenuated at this upstream site (0.67 mm: from 29 ⫾ 4% to 10 ⫾ 5%, P ⬍ 0.05, Figure 42.3). If charybdotoxin was applied in a similar way at a distant site (0.67 mm) in another series of experiments, the dilatation upon acetylcholine remained unaltered at the local (45 ⫾ 4% vs 47 ⫾ 8%; L-NA ⫹ indomethacin vs charybdotoxin, respectively; P ⫽ 0.83) and also at this remote site (33 ⫾ 4 vs 30 ⫾ 6, P ⫽ 0.71, n ⫽ 6, Figure 42.3). The efficacy of the applied charbydotoxin at the remote site was tested by repositioning of the acetylcholine pipette to this site. In marked contrast to the remote response, the dilatation upon acetylcholine ejected directly at the site treated with charbydotoxin was significantly attenuated (from 51 ⫾ 8% to 24 ⫾ 5%, P ⬍ 0.05). However, all sites treated with charbydotoxin dilated close to maximal (77 ⫾ 6%) upon superfusion with sodium nitroprusside and adenosine. In a further experimental series iberiotoxin was used instead of charybdotoxin. As was found with charybdotoxin, the application of iberiotoxin at the site of acetylcholine stimulation attenuated the dilatation from 45 ⫾ 4% to 12 ⫾ 6% (P ⬍ 0.05, n ⫽ 6). In contrast, treatment of the remote site with iberiotoxin (distance from acetylcholine pipette: 0.67 mm) did not alter the response at this site (30 ⫾ 8% vs 23 ⫾ 5%, P ⫽ 0.53, n ⫽ 6, L-NA ⫹ indomethacin and iberiotoxin, respectively). 4. DISCUSSION This study shows that the release and action of EDHF, which is most likely a product of the P450 monooxygenase pathway, is a prerequisite to initiate a vasodilatation upon stimulation with acetylcholine that conducts along the vascular wall in the microcirculation of the cremaster of hamsters. The activation of Ca2⫹-dependent K⫹-channels, presumably by EDHF, is necessary only at the site of stimulation, but not at remote, distant sites. NO and

338

Cor de Wit et al.

L32

L32

L32

L32

Figure 42.3 Ca2⫹-dependent K⫹-channels were blocked in specific regions along the arteriole by injection of charybdotoxin (ChTx) in the vicinity of the arteriole with a micropipette as indicated by the grey area in insets. They also denote site of observation and site of stimulation with acetylcholine (pipette). Charybdotoxin injected at the local site (upper panels, n ⫽ 4) not only reduced the dilatation at this site but also at the upstream remote site. In marked contrast, charybdotoxin application at the remote site (lower panels, n ⫽ 6) did not alter dilatations upon acetylcholine at either location. Significant differences (P ⬍ 0.05) are indicated by *. All experiments were done in the presence of L-NA (3 ⫻ 10⫺5 M) and indomethacin (3 ⫻ 10⫺6 M). Absolute maximal changes in diameter is depicted in all panels.

prostaglandins are less important to initiate and accomplish a conducting dilatation. Several observations support these conclusions. The conduction of a dilatation after endothelial stimulation remained unaffected by inhibitors of NO-synthase and cyclooxygenase. However, two chemically different inhibitors of the P450 monooxygenase pathway (sulfaphenazole and ODYA) attenuated local and conducted dilatations in these animals. The site of action of EDHF was shown to be the location at which the vessel was stimulated by acetylcholine. Blocking the Ca2⫹-dependent K⫹-channels, the presumed target of EDHF, by means of topical application of charybdotoxin or iberiotoxin at a conducted site did not alter the local or conducted dilatation. In contrast, topical blockade at the acetylcholine stimulation site attenuated the local and also the conducted response. Thus, the signal

Local release of EDHF initiates a conducted dilatation

339

transmitted along the vascular wall after stimulation with acetylcholine is most likely a hyperpolarization induced by the opening of K⫹-channels generated by a P450 monooxygenase product at the stimulation site. The fact that dilatations were observed at remote sites at a distance of 0.67 mm with a time delay of less than one second excludes diffusion of mediators and can virtually only be achieved by a signal that is transmitted electrotonically. Therefore, it is concluded that endothelial stimulation with acetylcholine elicits a local hyperpolarization by the release and action of EDHF at the local site. The data suggest that this hyperpolarization is conducted to upstream sites via the endothelial and/or smooth muscle layer to induce a dilatation along the arteriole. At remote sites, the release and action of all endothelial factors studied is not required. The conduction of signals along arterioles is classically studied by focal, transient application of vasomotor stimuli in the vicinity of arterioles either on isolated vessels (Xia et al., 1995; Xia and Duling, 1995; Dietrich et al., 1996; Doyle and Duling, 1997) or in the intact microcirculation (Welsh and Segal, 1998; Segal et al., 1999; Welsh and Segal, 2000b). The microejection of acetylcholine evoked not only local, but also upstream vasodilatations in this preparation. In contrast to acetylcholine, the NO-donor sodium nitroprusside did not initiate a remote dilatation despite a local dilatation of a similar amplitude. This demonstrates that a local dilatation per se is not sufficient to elicit a conducted response. Moreover, this also rules out that distant responses were due to diffusion of the locally ejected vasodilator. The endothelium releases several mediators upon acetylcholine in this preparation (de Wit et al., 1999). Other endothelium-dependent vasodilators are also capable of initiating a conducted response (e.g. bradykinin, (de Wit et al., 2000) ). Therefore, the mechanism of action is most likely related to the release of an endothelial autacoid. To identify their potential roles, selective blockers of the known endothelial mediators were applied sequentially. Blockade of NO-synthase and cyclooxygenase did not diminish the maximal amplitude of local as well as remote responses upon stimulation with acetylcholine. Furthermore, a NO-donor did not elicit conducted responses supporting the view that NO was only of minor importance. This is in agreement with studies in different vascular beds (Rivers, 1997) or in experiments on isolated arteries (Doyle and Duling, 1997; Welsh and Segal, 2000b). However, the additional blockade of P450-monooxygenase by sulfaphenazole (Fisslthaler et al., 1999) or ODYA attenuated the local dilatation in response to acetylcholine. This supports the concept that EDHF is most likely a product of the P450-monooxygenase pathway in the microcirculation of the hamster (de Wit et al., 1999) and isolated small arteries (Bolz et al., 2000). Moreover, the data demonstrate that the formation of presumably EETs is necessary to initiate a dilatation at remote sites as these responses were completely abrogated by sulfaphenazole and reduced by ODYA. The crucial role of EDHF is supported by the observation that the arteriole dilates nearly synchronously at local and remote sites. Only changes of membrane potential (evoked by EDHF) are conducted sufficiently fast for this synchronicity to be observed. And indeed it was shown previously that in arterioles of hamsters a hyperpolarization was induced by acetylcholine, but not NO (Bolz et al., 1999). Furthermore, it could be shown that EDHF is not the mediator of the dilatation at remote sites. This was achieved by applying in a topical manner blockers of Ca2⫹-dependent K⫹channels, which have been shown to be a target of EDHF in this (de Wit et al., 1999) and other preparations (Hashitani and Suzuki, 1997; Mieyal et al., 1998). The application of charybdotoxin at the stimulation site attenuated not only the local dilatation, but also the dilator response at the conducted site. This suggests that EDHF, in particular EETs, are released at the site of acetylcholine application and induce at this site a hyperpolarization of the endothelial and/or vascular smooth muscle cells. Without active Ca⫹⫹-dependent

340

Cor de Wit et al.

K⫹-channels at the stimulation site a conduction of the response was not found. On the contrary, the blockade of these channels at the remote site did not affect the dilatation at this site and the local dilator response was also preserved. The inability of charybdotoxin to block remote dilatations after topical application at these sites was not due to an insufficient blockade. A sufficient K⫹-channel blockade was clearly demonstrated by the strong attenuation of the dilatation in response to acetylcholine directly applied at this site. Virtually similar results were obtained with iberiotoxin, another blocker of Ca2⫹-dependent K⫹-channels, further emphasizing the critical role of the activation of Ca⫹⫹-dependent K⫹channels at the site of stimulation with acetylcholine. If active Ca2⫹-dependent K⫹-channels as well as generation of NO and prostaglandins are not a prerequisite to induce a remote dilatation, it remains to be resolved, how this response is accomplished. The present data suggest that hyperpolarization is conducted via the smooth muscle and/or endothelial layer to these sites and is sufficient to induce a dilator response. Local electrical activation and hyperpolarization of the endothelium is sufficient to induce a dilatation which conducts along the vascular wall (Emerson and Segal, 2000a, 2001). This initial local hyperpolarization necessary to evoke a signal that travels along the arteriole seems to be achieved by the focal EDHF release and action as shown in the present study. Whether EDHF hyperpolarizes not only the smooth muscle, but also the endothelium remains to be elucidated. Thus, EDHF may serve as a critical endothelial autacoid to coordinate arteriolar behaviour in the microcirculatory network.

43 Interaction of astrocytes and cerebral endothelial cells: function of astrocytic epoxyeicosatrienoic acids in the differentiation of endothelial cells David R. Harder, Chenyang Zhang, Jayashree Narayanan and Meetha Medhora Angiogenesis is initiated by regulation of localized factors that results in matrix dissolution, proliferation, invasion, migration and differentiation of endothelial cells to form blood vessels. The arachidonic acid metabolite, epoxyeicosatrienoic acid (EET), promotes growth of rat cerebral microvascular endothelial cells in vitro. Besides endothelial cells, a second source of EETs in the brain are astrocytes, which express epoxygenases, including the isoform cytochrome P450 2C11 (CYP2C11), that can metabolize arachidonic acid to EETs. The regioisomer 8,9-EET is as potent as vascular-endothelial growth factor (VEGF) in causing proliferation of cerebral endothelial cells. In addition, it significantly increases VEGFinduced growth of these cells as measured by incorporation of 3H-thymidine in presence of these factors. Even though VEGF has no effect on growth of astrocytes, it stimulates proliferation of these cells in the presence of the EET precursor, arachidonic acid. Thus interaction of angiogenic factors with arachidonic acid metabolites, especially EETs, enhances proliferation of endothelial cells and surrounding astrocytes in the brain. One of the mechanisms by which EETs increase growth of endothelial cells is by regulation of gene expression of specific growth promoting genes, as demonstrated by DNA array analysis of vehicle-treated endothelial cells vs cells treated with 8,9-EET. In summary, these results indicate that EETs have additive effect on VEGF-driven growth of rat cerebral endothelial cells and astrocytes. This is the first report of the interaction of EETs and VEGF in proliferation of endothelial cells, an important step in angiogenesis.

Epoxyeicosatrienoic acids (EETs) are derivatives of arachidonic acid that have multiple biological functions. One of the most noted of these is the characterization of EETs as an endothelium-derived hyperpolarizing factor (EDHF) in bovine (Campbell et al., 1996; Gebremedhin et al., 1998) and porcine (Fisslthaler et al., 1999) coronary and rat cerebral circulation. What has made this finding even more interesting is that EETs also promote growth of vascular endothelial cells, so that this EDHF is also responsible for angiogenesis and maintenance of vascular integrity. This function increases in importance in the coronary and cerebral beds since these are often injured by hypoxic insults and would benefit from increased angiogenic activity subsequent to injury. While both vascular beds pose this challenge, the present study focuses its attention on EETs-mediated angiogenesis in the brain. The cerebral circulation is unique because it maintains the blood brain barrier, and vessels are surrounded by astrocytes and neurons, of which astrocytes are more abundant. The astrocytes have been seen to extend foot processes onto the vasculature, but the functions of these structures are not known. Cultured astrocytes are capable of metabolizing arachidonic acid (Amruthesh et al., 1993) and the predominant metabolites produced by intact cells are EETs (Alkayed et al., 1996, 1997). The enzymes that carry out the reaction

342

David R. Harder et al.

are from a family of ~500 known cytochrome P450 isoforms, one of which is the epoxygenase isoform 2C11, also expressed by astrocytes (Alkayed et al., 1996). Differentiation of rat cerebral endothelial cells into tubes in vitro, requires the presence of astrocytes (Laterra et al., 1990). Astrocyte-conditioned medium promotes growth of cerebral endothelial cells (Munzenmaier and Harder, 2000) derived from rat microvessels. In addition, when astrocyteendothelial cell cocultures are incubated with a cytochrome P450 (CYP450) enzyme inhibitor, 17-ODYA, the differentiation of endothelial cells into tubes is attenuated (Munzenmaier and Harder, 2000). Formation of EETs from the substrate arachidonic acid is dependent on CYP450 epoxygenase activity, giving support to the hypothesis that astrocytes contribute to angiogenesis via production of EETs. To investigate this interaction and resolve the mechanism(s), the role of EETs in astrocyte and endothelial cell growth was studied with emphasis on the synergy between angiogenic action of EETs and vascular endothelial growth factor (VEGF). VEGF is a very important factor in initiating angiogenesis and EETs may sustain and complement the actions of VEGF to promote angiogenesis. EETs may putatively mediate this activity in a number of ways, including regulation of gene expression in cerebral endothelial cells. To investigate this, DNA array technology was used and modulation of a number of growth-promoting genes was detected. In summary, the work described here supports the role of the arachidonic acid metabolites, EETs, in growth. Astrocytes as well as cerebral endothelial cells were used to explore the interactions of EETs with VEGF. 1. METHODS

1.1. Cell culture Primary cultures of astrocytes were prepared from hippocampi of postnatal 3-day-old rat pups as described previously (Alkayed et al., 1996). Cerebral capillary endothelial cells were prepared from brains of 4-week-old rats (Munzenmaier and Harder, 2000) with some modifications. Briefly, cerebral cortex were dissected and homogenized in ice-cold HEPESbuffered saline solution with 0.9% glucose (Alkayed et al., 1996). Vascular tissues were separated from the rest of brain by centrifugation in a HEPES-buffered salt solution containing 15% dextran. The vascular tissue was filtered through a 150-micron mesh screen to remove large vessels. The eluates were loaded on a glass bead column. Capillaries adhering to the beads were released by vigorous shaking. The microvessel pellet was digested with collagenase (500 ␮g/ml) in RPMI-1640 (BioWhittaker, Walkersville, MD, USA) containing 10% fetal bovine serum for 15 min at room temperature. The formation of a single-cell suspension was monitored under a phase-contrast microscope. After centrifugation, the cell pellet was resuspended in a L-valine free medium (Life Technologies, Gaithersburg, MD, USA), which inhibits growth of other cell types but not endothelial cells (Picciano et al., 1984), and plated in T-25 flasks pre-coated with fibronectin at 5 ␮g/cm2. Cells were incubated at 37 ⬚C in a 95%–5% mixture of atmospheric air and CO2. After three days, the medium was changed to microvascular endothelial cell growth medium that was formulated to promote growth of endothelial cells (Clonetics, Walkersville, MD, USA). This medium is made from serum-free endothelial cell basal medium-2 (EBM-2) supplemented with vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), fibroblast growth factor (FGF), insulin-like growth factor (IGF), ascorbic acid, hydrocortisone, heparin, fetal bovine serum (FBS) and antibiotics according to manufacturer’s instruction (Clonetics, Walkersville, MD, USA). Confluent first passages of endothelial cells were used in all experiments.

Interaction of astrocytes and cerebral endothelial cells

343

1.2. 3H-thymidine incorporation Confluent cerebral endothelial cells or astrocytes in T-25 flasks were detached by trypsin. After centrifugation, the cells were suspended in the endothelial cell growth medium and plated at 7500 cells/well into 24-well plates, and incubated for one day so that they reached 80% confluency. The medium was then changed to EBM-2 with 0.1% bovine serum albumin (BSA) for two days to make the cells quiescent. EETs at 1 ⫻ 10⫺7 M and/or VEGF at 1 ⫻ 10⫺9 M were added to the medium as required and incubated for 18 h. 3H-thymidine at a concentration of 2 ␮Ci/ml was then added to pulse the cells for an additional 3 h. Cells were washed 3 times with phosphate-buffered saline, and precipitated with ice-cold 15% trichloroacetic acid for 30 min at 4 ⬚C. Wells were washed gently with water and allowed to dry. Cells were lysed with 1N NaOH and incubated at 37 ⬚C for 30 min. After neutralization with 1N HCl, the radioactivity from the sample of each well was determined by liquid scintillation spectrometry after adding Ecoscint (National Diagnostics, Atlanta, GA, USA). Results were expressed as counts per minute per well. Each experimental data point represented at least 4 wells from three or more independent experiments. All values were expressed as mean ⫾ SEM (standard error of mean). Student’s “t” tests were employed to compare results from vehicle treated controls vs reagent treated samples. A significant difference was reported for “P” value of less than 0.05.

1.3. Infection of astrocytes with Ad5-2C11 and astrocytic conditioned medium Subconfluent astrocytes were infected with Ad5-2C11 (at 4.5 ⫻ 107 pfu) or Ad5-GFP (Medhora and Harder, 1998). For making conditioned medium from astrocytes, cells infected with Ad5-2C11 or Ad5-GFP were washed with Dulbecco’s phosphate buffered saline (Sigma Chemicals, St Louis, MO). Dulbecco’s modified Eagle’s medium with 0.1% bovine BSA (fatty acid free) was added to cells for conditioning overnight. The conditioned medium was combined with EBM-2 ⫹ 0.1% BSA at 1:1 ratio, and was added to the quiescent endothelial cells seeded in 24-well plates.

1.4. DNA arrays Array analysis for rat cerebral microvascular endothelial cells were done by growing the cells in 100 mm dishes to 80% confluency, starving for 48 h with EBM-2 ⫹ 0.1% BSA and treating them separately with vehicle (ethanol less than 0.1%v/v) or 8,9-EETs (3 ⫻ 10⫺7 M) for 18 h. The EETs were replenished after 12 h. Total RNA was isolated with Trizol reagent (Gibco, Grand Island, NY, USA) as described (Medhora, 2000). Briefly, both control as well as treated cells were washed with phosphate buffered saline (PBS), homogenized with the reagent (3.0 ml/100 mm dish) and treated with 0.2 volumes chloroform to partition the solution. The aqueous layer from each was recovered after centrifugation and the RNA was precipitated at room temperature by adding 0.5 volumes of isopropanol. The pellet was recovered by centrifugation at 15,000 ⫻ g in an Eppendorf tabletop centrifuge and washed with 70% RNAse-free ethanol. The pellet was dried at room temperature and resuspended in 30–50 ␮l of RNAse-free water. Concentration of RNA was determined by A260/280 nm in a spectrophotometer. The remaining procedure was followed using the Atlas Rat 1.2 Array from Clontech, (Palo Alto, CA) using the buffers, enzymes and a matched set of gene array filters supplied

344

David R. Harder et al.

by the manufacturer. Briefly, probes were synthesized by reverse transcribing 5 ␮g of purified total RNA, using the cDNA synthesis primer mix and [␣-32P]dATP (deoxyadenosine triphosphate, 3000 Ci/mmol, Amersham, Piscataway, NJ, USA). The labeled cDNA probes were purified from unincorporated nucleotides using Atlas NucleoSpin columns, and radioactivity of the probes were determined using a scintillation counter (Wallac, 1450 Micro Beta). Equal counts of cDNA probes were hybridized to Atlas Array membranes in a hybridization oven at 68 ⬚C. After overnight hybridization and high-stringency washes as mentioned by the manufacturer, the membranes were exposed to a phosphorimager screen and the data directly analyzed after normalization using Atlas Image 2.0 software. 2. RESULTS

2.1. 8,9-EETs promote VEGF-induced proliferation of rat cerebral endothelial cells Cerebral endothelial cells were grown to 70% confluence and growth arrested by starving in EBM-2 containing 0.1% BSA for 48 h. The cells were then treated for 18 h with 8,9-EET (1 ⫻ 10⫺7 M), VEGF (1 ⫻ 10⫺9 M) and a combination of 8,9-EET and VEGF. Cell proliferation was assayed by 3H-thymidine incorporation. Cell growth was stimulated to the same extent by 8,9-EET and VEGF and was further increased by adding both factors together (n ⫽ 12; P ⬍ 0.05, Figure 43.1).

2.2. Epoxygenase CYP2C11 overexpression in astrocytes increases growth of rat cerebral endothelial cells To test the effect of CYP2C11 in astrocytes on proliferation of cerebral endothelial cells, astrocytes were infected with adenovirus expressing recombinant epoxygenase CYP2C11

5000 **

**

##

4000

**

3000 2000 1000

3

H-thymidine incorporation per well (CPM)

6000

0 Vehicle

8,9-EET

8,9-EET VEGF

VEGF

Figure 43.1 Graph showing effect of 8,9-EET (1 ⫻ 10⫺7 M) and VEGF (1 ⫻ 10⫺9 M) on 3Hthymidine incorporation in rat cerebral microvascular endothelial cells. ** indicate that P ⬍ 0.001 as compared to vehicle; ## indicates that P ⬍ 0.05 as compared to 8,9-EET or VEGF. CPM ⫽ counts per minute.

3

H-thymidine incorporation per well (CPM)

Interaction of astrocytes and cerebral endothelial cells 7000

345

**

6000 5000 4000 3000 2000 1000 0

Ad5-GFP

Ad5-2C11

Figure 43.2 Graph showing effect of conditioned medium from astrocytes infected with adenovirus carrying cDNA for CYP2C11 or green fluorescent protein (GFP) on 3 H-thymidine incorporation in rat cerebral microvascular endothelial cells. ** indicate that P ⬍ 0.001. CPM ⫽ counts per minute.

cDNA driven by a CMV promoter (Medhora and Harder, 1998). Recombinant CYP2C11 was previously cloned and characterized in an adenoviral vector. The enzyme CYP2C11 metabolizes arachidonic acid to EETs (Yoshioka et al., 1987) and the adenoviral vector is able to infect astrocytes with very high efficiency (Medhora and Harder, 1998). Control cells were infected with Ad5-GFP (green fluorescent protein). Most of the astrocytes in the cultures expressed recombinant CYP2C11 after infection with virus at a multiplicity of infection of 50 plaque forming units of virus/astrocyte. Cultured rat cerebral endothelial cells were serum starved and then treated for 18–24 h with conditioned media from the astrocytes, before estimating DNA replication by 3H-thymidine incorporation (see Methods). Conditioned medium from astrocytes expressing either CYP2C11 or GFP was used for comparison. DNA synthesis was significantly increased over 2.5-fold in astrocytes over-expressing CYP2C11 than cells expressing Ad5-GFP (Figure 43.2).

2.3. VEGF and AA, together, stimulate growth of astrocytes Since cerebral angiogenesis involves interaction and growth of endothelial cells and astrocytes, the effect of VEGF on growth of astrocytes was tested. VEGF did not stimulate growth of astrocytes in culture (Figure 43.3). There was no effect of the epoxygenase substrate, arachidonic acid, on growth of astrocytes (Figure 43.3). However, treatment of the primary cultures of astrocytes with VEGF (2 ⫻ 10⫺9 M) and AA (5 ⫻ 10⫺7 M) significantly increased DNA synthesis in astrocytes over 3-fold (Figure 43.3).

2.4. Effect of 8,9-EET on gene expression in rat cerebral endothelial cells An array of rat genes was tested for expression of endothelial mRNAs after treatment of rat cerebral microvascular endothelial cells in culture with 8,9-EET. Control cells were treated with vehicle. Total RNA was extracted from both types of cells and labeled with ␣32P-dATP. The 32P-labelled RNA from vehicle and control cells were used separately to hybridize to identical arrays of 1176 rat cDNAs (Clontech Rat 1.2 array). Autoradiographs of the arrays were analyzed and the intensity of signal between vehicle and 8,9-EET (3 ⫻ 10⫺7 M) treated cells was compared. This dose of 8,9-EET had previously given maximal growth response

346

David R. Harder et al. 1600 1400 1200

* n = 16

Vehicle AA VEGF VEGF + AA

CPM

1000 800 600 400 200 0

Figure 43.3 Graph showing effect of arachidonic acid (5 ⫻ 10⫺7 M) and VEGF (2 ⫻ 10⫺9 M) on 3 H-thymidine incorporation in rat cerebral microvascular endothelial cells. * indicates that P ⬍ 0.05 as compared to vehicle. CPM ⫽ counts per minute from incorporated 3H-thymidine.

when applied to rat cerebral endothelial cells. The signal from the two arrays was normalized by comparing nine housekeeping cDNAs on the membrane. Plasmid and bacteriophage DNAs on the filters were used as negative controls to confirm hybridization specificity. The autoradiographs were analyzed after normalization and genes with detectable intensity of signal (adjusted intensity above 0.12 using the AtlasImage 2.0 software from Clontech) and expression difference greater than 1.4-fold were used for the results (Table 43.1). 3. DISCUSSION The list of cardiovascular-related functions of the arachidonic acid metabolites, EETs, is rapidly growing. In addition to anti-inflammatory (Node et al., 1999) and fibrinolytic (Node et al., 2001) effects in endothelial cells the EETs are protective following ischemia (Zeldin, 2001). In addition, epoxygenase activity in astrocytes has been hypothesized to mediate functional hyperemia in the brain. Metabolic spillover of neurotransmitters accompanying neuronal stimulation has been conventionally believed to be “mopped up” by astrocytes. Stimulation of EET synthesis in cultured astrocytes results from application of the most abundant neurotransmitter, glutamate (Alkayed et al., 1997; Harder et al., 2002). Since EETs are dilators in the cerebral circulation (Ellis et al., 1990; Gebremedhin et al., 1992) and because of their anatomical juxtaposition with the vasculature (Harder et al., 2002) a link for neuronal function being coupled to local vasodilation via release of EETs from astrocytes, was hypothesized. More recently, a mitogenic function of EETs in endothelial cells was observed (Zhang et al., 2001; Medhora et al., 2002), which has prompted investigation of the effect of EETs on vascular remodeling. Of the four commonly described regioisomers of EETs, 14,15-, 11,12-, 8,9- and 5,6-EETs, the 8,9-EET has been the most potent stimulator of growth of primary cultures of rat cerebral endothelial cells (Zhang et al., 2001). To study the interaction of endothelial cells and astrocytes the effect of 8,9-EET on VEGF-induced growth of cerebral endothelial cells was examined. Endothelial cell proliferation is not common in the adult brain but angiogenesis can occur under pathological conditions (Xu and Severinghaus, 1998). VEGF has been detected in brain tumor tissue, rat

Interaction of astrocytes and cerebral endothelial cells

347

Table 43.1 Genes involved in growth and signaling that are modulated by treatment of cultural rat cerebral microvascular endothalial cells with 8,9-EET Genes upregulated Growth

● ● ● ● ● ● ●

c-fos proto-oncogene c-jun proto-oncogene PDGFa receptor ATP synthase, subunit c, P2 gene b-nerve growth factor precursor FGF 10 precursor Plasminogen activator inhibitor

Genes downregulated ●

● ● ● ● ● ● ●

Receptors and cell signaling

● ● ● ● ●

● ● ● ● ● ● ● ● ●

●

Cadherin 6 precursor Integrin ␣1 STAT3 erbB3 EGF receptor-related proto-oncogene erbB2 receptor protein-tyrosine kinase precursor PKC-d Phospholipase C d Phosphotidylinositol 4 kinase Rab-related GTP binding protein Neuronal acetylcholine receptor protein a 5 serotonin receptor 4 serotonin receptor 5B GABA-A receptor 7 glutamate receptor, ionotropic, kainate 5 metabotropic glutamate receptor 7 precursor

●

NGF-inducible anti-proliferative protein Bone morphogenetic protein 4 Inhibin, b A subunit JAK1 ERK 1 G1/S-specific cyclin E (CCNE) Cyclin (PCNA) Copper-zinc containing superoxide dismutase Endothelin 1 precursor

cerebellum and mouse choroid plexus (Breier et al., 1992) proceeding surgical trauma (Severinghaus, 1995) and after inhalation hypoxia in rat brain (Xu and Severinghaus, 1998). Of the many putative angiogenic factors described including VEGF, epidermal growth factor, tumor angiogenesis factor, acidic and basic fibroblast growth factor, platelet-derived endothelial cell growth factors (PDGFa and b) and interleukin-8, VEGF is thought to be the most potent especially in initiating angiogenesis. This study shows that 8,9-EET is as potent as VEGF in promoting cell growth of rat cerebral microvascular endothelial cells, and enhances the effect of VEGF on growth. This can explain a role for astrocyte-released EETs in supporting angiogenesis of cerebral vessels. VEGF, which is usually synthesized by endothelial cells, does not cause increase in proliferation of cultured astrocytes. However, in the presence of both VEGF and arachidonic acid, astrocyte proliferation is significantly increased. The major metabolites of exogenous arachidonic acid in intact astrocytes are EETs (Alkayed et al., 1996), which imply that EETs may act in concert with VEGF in promoting astrocyte growth during angiogenesis. However, it must be kept in perspective that cultured astrocytes proliferate readily; which is atypical of astrocytes residing in adult brain, though certain astrocytes are capable of dividing in adults (Rakic, 1985). It is therefore intriguing that growth factors (VEGF and EETs) from both types of cells, astrocytes and endothelial cells act in synergy to promote proliferation of each other.

348

David R. Harder et al.

Further evidence for the role of EETs as a mediator of this interaction is the finding that conditioned media from astrocytes expressing large amounts of the epoxygenase CYP2C11, stimulate the growth of cultured endothelial cells. As mentioned CYP2C11, while being able to metabolize a number of substrates, also epoxygenates arachidonic acid to produce EETs. The Ad5-2C11 construct increases EETs in cultured astrocytes (Medhora and Harder, 1998). Thus, it was not surprising to see enhanced thymidine incorporation in endothelial cells treated with conditioned media from CYP 2C11 over-expressing astrocytes, as compared to astrocytes expressing GFP. We did notice a marked decrease in proliferation of endothelial cells treated with conditioned media from cells infected with Ad5, as compared to uninfected astrocytes (not shown). This could be the effect of expression of proteins coded by the adenoviral genome that is part of the adenoviral vector. These proteins, that are involved with viral growth and packaging, do not support growth of the host cells and may be released into the media from lysed cells. However, the conditioned medium from Ad5GFP infected cells showed increased stimulation of growth as compared to unconditioned (fresh) basal medium. EETs are found endogenously in astrocytes as well as endothelial cells and are increased in human coronary endothelial cells by stimulation with physiological agonists such as bradykinin (Medhora et al., 2001a–c). EETs also increase proliferation of human coronary and pulmonary endothelial cells (Fleming et al., 2001a; Medhora et al., 2001a, 2002). Proliferative pathways such as the extracellular signal-regulated kinase (erk), p38 mitogenactivated protein kinase (p38 MAPK) and tyrosine kinases are modulated by application of EETs (Chen et al., 1999; Fleming et al., 2001a; Zhang et al., 2001; Potente et al., 2002). The present experiments tested if expression of growth-regulatory genes were altered by application of 8,9-EETs. 8,9-EETs increased the message levels of some important genes involved in growth, including the immediate early genes c-fos and c-jun in rat cerebral endothelial cells. The receptor for PDGFa is also upregulated. Thus EETs can regulate growth by activating signaling cascades (Chen et al., 1999; Fleming et al., 2001a) as well as by modulating specific growth-promoting genes. A number of growth factors were downregulated by 8,9-EET (Table 43.1). The results of the arrays however should be confirmed at the protein level and with increasing concentrations of EETs to narrow down the functional products that are modulated by 8,9-EETs and to verify that increase in mRNA is followed by increase in protein. In conclusion, EETs are proliferative agents that may act alone or in concert with other powerful angiogenic factors such as VEGF to promote endothelial cell growth. In fact EETs also increase EGF-mediated rise in [Ca2⫹]i in renal proximal tubule cells, thereby enhancing mitogenesis induced by EGF (Chen et al., 1999), another angiogenic factor. The role of EETs in endothelial cell differentiation, another step in angiogenesis, is currently under investigation. ACKNOWLEDGMENTS The authors acknowledge the help of Joan Anders and Rachel Kraemer for preparing primary cultures of endothelial cells and astrocyte. We thank Dr Paul Vanhoutte for organizing this meeting. Financial support from NIH/NHLBI (USA) PO1 HL-59996, NIH/NHLBI (USA) RO1 HL 33833-16, HL 069996 and 3440-06P from the Department of Veterans Affairs, USA.

44 11,12-EETs hyperpolarize human platelets Florian Krötz, Tobias Riexinger, Matthias Keller, Hae-Young Sohn and Ulrich Pohl

Endothelium-derived hyperpolarizing factor (EDHF) is, at least in some blood vessels, assumed to be identical with 11,12-epoxyeicosatrienoic acid (11,12-EET), a product of endothelial CYP450 2C metabolism. It is investigated whether 11,12-EETs affect membrane potential not only of vascular smooth muscle cells but also of platelets. Platelet membrane potential was assessed by flow cytometry using the membrane potential sensitive fluorescent dye DiBac4(3). Calibration was performed using increasing extracellular K⫹ concentrations in the presence of valinomycin. Resting membrane potential of washed human platelets was calculated to ⫺58 ⫾ 9 mV. The Na⫹/K⫹-ionophor gramicidin induced a membrane depolarization of platelets, while the K⫹-ionophor valinomycin decreased fluorescence indicating hyperpolarization. 11,12-EET also hyperpolarized platelets in a dose-dependent manner. This effect was abolished by charybdotoxin indicating involvement of calcium-gated K⫹ (KCa) channels. In contrast, valinomycin-induced hyperpolarization could not be prevented by charybdotoxin. Depolarization of platelets caused a strong release of superoxide, which was scavenged by superoxide dismutase. This amount of superoxide anions released could also be decreased by 11,12-EETs. In conclusion, 11,12-EETs that have been reported to be released from endothelial cells, hyperpolarize platelets, which inhibits the depolarization-induced platelet release of superoxide anions.

1. INTRODUCTION Endothelial dysfunction and activation of platelets are key determinants in the development of cardiovascular disorders. Endothelial dysfunction is characterized by an impairment of nitric oxide (NO)-dependent dilatation. Besides NO and prostacyclin, which potently inhibit platelet activation (Moncada et al., 1977; Bassenge, 1991), the endothelium releases a hyperpolarizing factor (endothelium-derived hyperpolarizing factor, EDHF) (Pfister et al., 1996), which in certain blood vessels may be an arachidonic acid metabolite produced by cytochrome P4502C (CYP2C), namely 11,12-epoxyeicosatrienoic acid (11,12-EET) (Campbell et al., 1996; Fleming, 2001). Although various epoxyeicosatrienoic acids, including 11,12-EETs strongly inhibit platelet activation, the mechanism of their action has not been clarified (Fitzpatrick et al., 1986). Hypothetically, a change in membrane potential may be an integral part in the action of 11,12-EETs on platelets since these compounds may, as in vascular smooth muscle cells, activate calcium-gated K⫹-channels (KCa-channels). The latter are expressed in platelets (Fine et al., 1989; Mahaut-Smith, 1995). Such influences may have functional consequences since, the endothelial NAD(P)H-oxidase mediates a superoxide anion (O⫺ 2 ) production which is modulated by changes in membrane potential (Sohn et al., 2000). Platelets also contain a O⫺ 2 producing NAD(P)H-oxidase (Krötz et al., 2002).

350

Florian Krötz et al.

Therefore, the present study was designed to investigate, whether 11,12-EETs affect the membrane potential of platelets and whether release of O⫺ 2 by platelets depends on the membrane potential and influenced by 11,12-EETs. 2. METHODS

2.1. Platelet isolation Venous blood was drawn from healthy volunteers who had not taken any medication for at least ten days. Informed consent was obtained from all subjects. Supernatants of blood, that was anticoagulated by 3.13% sodium citrate and centrifuged at 150 g for 15 min was used as platelet-rich plasma. Washed platelets were prepared by another centrifugation step at 600 g for 10 min by re-suspending in calcium-free modified tyrode buffer containing 138 mmol/L sodium chloride, 2.7 mmol/L potassium chloride, 12 mmol/L sodium hydrogen carbonate, 400 ␮mol/L disodium phosphate, 1 mmol/L magnesium chloride, 5 mmol/L D-glucose and 5 mmol/L N-[2-hydroxyethyl]piperazine-N’-[2-ethanesulfonic acid] (Buffer A). Washed platelets were used within 2 h. Platelet and leukocyte counts were obtained using a resistance particle counter (Coulter Z2, Beckman Coulter, Germany).

2.2. 11,12-EETs 11,12-EETs were purchased from Biomol (USA) as solutions in 100% ethanol. As ethanol is known to be a potent inhibitor of thrombin-induced platelet activation (Fenn and Littleton, 1982), which is due to an alteration of lipid composition of platelet membranes, 11,12-EETs were lyophilized by speed-vac-centrifugation and re-suspended in PBS before use in experiments.

2.3. Measurement of platelet membrane potential Platelet membrane potential was assessed using the potential-sensitive fluorescent dye DiBac4(3). two lakhs washed platelets/␮1 were incubated with 500 nM of DiBac4(3) for 30 min at darkness and room temperature in the presence of 27 nM prostacyclin (Iloprost, Schering, Germany). After addition of 1 mM CaCl2 they were exposed to the respective substances and mean fluorescence of 10,000 platelets was measured after 10 min. For calibration, platelets were re-suspended in buffer containing 0.1, 10, 20, 30, 60 or 90 mM of KC1 in the presence of valinomycin.

2.4. Superoxide anions Superoxide anion production was measured by a chemiluminescence assay using the dye L-012 (Sohn et al., 1999). Only leukocyte-free solutions were used for experiments (final reaction volume 300 ␮L) containing 150,000–300,000 washed platelets/␮L, CaCl2 (1 mmol/L), L-012 (100 ␮mol/L), and stimulating substances were added. Photon emission was expressed as % increase in relative light units (RLU) versus control conditions.

2.5. Materials Superoxide dismutase was purchased from Roche Molecular Biochemicals (Basel, Switzerland). L-012 was from Aventis (Bad Soden, Germany), Iloprost was from Schering

EDHF and platelets

351

(Berlin, Germany). DiBac4(3) was from Molecular Probes (Leiden, Netherlands). 11,12-EETs were purchased from Biomol (Plymouth Meeting, PA, USA). All other substances were obtained from Sigma Chemicals Co (Munich, Germany).

2.6. Statistical analysis All data are expressed as means ⫾ SEM. Data were analyzed using one-way ANOVA or student’s t-test for paired or unpaired data. Differences were considered significant when the error probability level was P ⬍ 0.05. 3. RESULTS

3.1. Platelet membrane potential Incubation of platelets with DiBac4(3) (5⫻10⫺6 M) for 30min led to a stable fluorescence signal. In platelet buffer containing 2.7 ⫻10⫺3 M K⫹, addition of valinomycin, a K⫹-ionophor, led to a decrease of fluorescence which reached a constant level at 5–7min, indicating that the resting membrane potential of platelets was different from the K⫹ equilibrium potential. The levels of fluorescence decreased upon addition of up to 5 ⫻ 10⫺6 M of valinomycin. Therefore, this concentration was used for calibration. In contrast, addition of the Na⫹/K⫹ ionophor gramicidin (10⫺6–10⫺7 M), increased the fluorescence indicating a membrane depolarization. For calibration of the membrane potential the so-called valinomycin null-point method was used (Freedman and Novak, 1989). First, platelets were re-suspended in modified buffer A containing 0.1% K⫹ (corrected for osmalarity by adding the respective amount of Na⫹). Then valinomycin and increasing concentrations of K⫹ were added and the fluorescence changes monitored after 5 min. Plotting of platelet DiBac4(3)-fluorescence changes against K⫹ concentrations resulted in a correlation, that was linear between concentrations of external potassium (Ke⫹) from 0.1 to 60 ⫻ 10⫺3 M (Figure 44.1). On the

Fluorescence change (%)

300 250 200 150 100 50 0 0

20

40 60 Potassium (mM)

80

100

Figure 44.1 Calibration of platelet membrane potential using the fluorescent dye DiBac4(3). DiBac4(3) fluorescence changes were calibrated to membrane potential by adding increasing concentrations of KCl to the suspensions in the presence of the potassium (K⫹) inonophore valinomycin (5 ⫻ 10⫺6 M).

352

Florian Krötz et al.

assumption that, in the presence of valinomycin, the membrane potential is close to the K⫹ equilibrium potential, membrane potential was calculated according to the Nernst equation EK⫹ ⫽ RT(FzK⫹)⫺1 ⫻ ln([K⫹]e/[K⫹]i), where R is a gas constant (taken as 8.314 J K⫺1 mol⫺1), T is the absolute temperature, F is the constant of Faraday (9.65 ⫻ 104 A s mol⫺1) and z is the charge of the respective ion (1 for K⫹). Thus membrane potential was calculated according to the simplified equation: EK⫹ ⫽ ⫺61 mV ⫻ log([K⫹]i/[K⫹]e). Taken that platelets contain an intracellular potassium concentration (Ki⫹) of 140 mM (Friedhoff and Sonenberg, 1983; Pipili, 1985), the resulting membrane potential was obtained by calculating the respective [K⫹]e that corresponded to the measured fluorescence value according to the equation obtained from the calibration curve (fluorescence blotted against [K⫹]e, Figure 44.1) (0.1–60 mM K⫹ external). In physiological platelet buffer, resting membrane potential calculated by this technique amounted to ⫺57.7 ⫾ 9.4 mV (n ⫽ 10).

3.2. Influence of 11,12-EETs on platelet membrane potential Hyperpolarization of washed platelets caused by 11,12-EETs reached a constant level after 10 min (Figure 44.2). 11,12-EETs hyperpolarized platelets in a dose-dependent manner from 57.7 to ⫺71.4⫾18mV (10⫺6 M, n⫽10) or to ⫺61.1⫾14mV (10⫺7 M, n⫽9, Figure 44.3). The hyperpolarization was reversible after addition of tetrabutylammonium chloride (10⫺3 M, n ⫽ 3 not shown) and was fully inhibited by Cbtx (10⫺6 M EET ⫹ 10⫺6 M Cbtx: ⫺52.2 ⫾ 16 mV, n ⫽ 10). In contrast, the effect of valinomycin (10⫺6 M, n ⫽ 8), which hyperpolarized platelets to ⫺67.5 ⫾ 17 mV (n ⫽ 8) could not be reversed by blockade of Intermediate Conductance (IK(Ca)) using Cbtx (10⫺6 M, n⫽3). Depolarization by gramicidin induced a dose-dependent increase in membrane potential to ⫺13.3 ⫾ 20 mV (10⫺6 M, n ⫽ 8) or to ⫺38.3 ⫾ 17 mV (10⫺7, n ⫽ 9). 11,12-EETs (10⫺7) also reduced the gramicidin-induced depolarization significantly (to ⫺47.4 ⫽ 12 mV, n ⫽ 10).

3.3. Platelet superoxide release Platelet exhibited a low basal O⫺ 2 formation which was increased 3.2-fold by gramicidin (10⫺7 M, n⫽4, P⬍0.01, Figure 44.3). This increase was significantly attenuated by 11,12-EETs (n ⫽ 4, P ⬍ 0.05), while superoxide dismutase (250 U/ml) completely scavenged gramicidinmediated release of O⫺ 2. 4. DISCUSSION Measurement of human platelet membrane potential can be achieved with high sensitivity with the fluorescent dye DiBac4(3) which changes fluorescence with high sensitivity in response to changes in membrane potential. In untreated platelets, the resting membrane potential of washed platelets amounted to ⫺58 mV. This value is consistent with that obtained by other laboratories using the single-cell patch-clamp technique (Maruyama, 1987; Mahaut-Smith et al., 1990) or other potential-sensitive fluorescent dyes (MacIntyre and Rink, 1982; Friedhoff and Sonenberg, 1983; Fine et al., 1989).

Counts

EDHF and platelets

100

Counts

0

Counts

[K+]e = 0.1 mM 100

101

102

100

0

353

103

104

[K+]e = 10 mM 100

101

102

100

103

104

[K+]e = 30 mM

0

Counts

100 100

Counts

0

101

102

103

104

102

[K+]e = 60 mM 100

101

103

104

100

101 102 103 Fluorescence Intensity

104

100

0

Figure 44.2 Flow cytometric measurement of platelet membrane potential. Representative flow cytometric histograms of DiBac4(3)-labeled platelets at different concentrations of external K⫹ in the presence of valinomycin (5 ⫻ 10⫺6 M).

There are conflicting results concerning an involvement of changes of platelet membrane potential during platelet activation. Typical platelet agonists like thrombin, ADP or plateletactivating factor caused slight depolarization of platelets, dependent on influx of sodium (Pipili, 1985), but depolarization or hyperpolarization did not influence aggregation or shape change. Other studies demonstrated an increased aggregability in response to ADP, when extracellular potassium levels were raised (Greil et al., 1972; Friedhoff et al., 1981) or when there was inhibition of platelet activation by 11,12-EETs and other stereoisomers of epoxyeicosatrienoic acid (Fitzpatrick et al., 1986). However, the mechanism of this inhibition remained unrevealed, and it was not investigated whether this was related to a change in platelet membrane potential. The relevance of these results remains uncertain, as concentrations larger than 10⫺7 M are not likely to be released by the endothelium (Fisslthaler, 1999) or to play a role in vivo (Nithipatikom et al., 2001). The membrane potential of resting platelets is determined mainly by the activation state of voltage-gated potassium channels (Kv), which are abundantly expressed in their membranes (Mahaut-Smith et al., 1990). However, platelets not only express Kv-channels, but also calcium-activated potassium channels (KCa), the target of 11,12-EETs (Pipili, 1985; Fine et al., 1989). These channels are not sensitive to apamin, an inhibitor of small conductance (SKCa)-channels, but are sensitive to charybdotoxin (Fine et al., 1989). The characterization of platelet KCa-channels by means of the nystatin whole-cell patch-clamp technique strongly

40

20

0

–20 G ra m ic id in G 10 –6 ra Va M m lin om . 10 – Va 7 l. yc M 10 in –6 1 M 0 –6 + M C bt 11 x 11 10 ,1 –6 ,1 22EE M EE T T 10 10 –6 –6 G M 11 ra m ,1 M + 210 C EE –7 bt T x M 10 + –7 EE M T 10 –7 M

Change in membrane potential (mV)

Resting membrane potential: –57,7 mV

Figure 44.3 Changes in platelet membrane potential by 11,12-EETs. Platelet membrane potential values shown as change in mV from resting membrane potential. Gramicidin (Gram., 10⫺6–10⫺7 M, n ⫽ 9) depolarized platelets in a dose-dependent manner, whereas valinomycin (10⫺6 M, n ⫽ 8) led to a hyperpolarization. Hyperpolarization could also be achieved using different doses of 11,12-EETs (n ⫽ 9), which was reversible by blocking Ca2⫹-gated potassium channels (KCa-channels) with charybdotoxin (Cbtx, 5 ␮M, n ⫽ 9 or tetrabutylammonium chloride, not shown). Hyperpolarization induced by valinomycin could not be reduced by charybdotoxin (n ⫽ 4). 11,12-EETs also reduced the gramicidininduced depolarization (n ⫽ 9). The resting membrane potential of washed human platelets in buffer containing 2.7 10⫺3 M KC1 was 57.7 ⫾9 mV (n ⫽ 10).

Gramicidin

Fluorescence (arbitrary units)

11,12-EETs

-2

0

2

4

6

8

10 12

t (min)

Figure 44.4 Time course of 11,12-EET-induced hyperpolarization. Representative tracing of the time course of a flow cytometric DiBac4(3) fluorescence measurement of platelet hyperpolarization induced by 11,12-EETs (10⫺7 M) and the effect of gramicidin (10⫺7 M).

EDHF and platelets Control Gramicidin 11,12-EET Gram+EET Gram+SOD

1400 Superoxide release (RLU)

355

1200 1000 800 600 400 200 0 –5

0

5 10 minutes

15

20

Figure 44.5 Depolarization-induced release of superoxide anions from platelets. Time course of superoxide anion release from platelets upon depolarization. Platelet membrane depolarization using gramicidin (10⫺7 M) induced a 3.2-fold increase in platelet release of superoxide anions (O⫺ 2 ) compared to control conditions (n ⫽ 4, P ⬍ .01), which could be scavenged by SOD (250 U/ml). 11,12-EETs only gradually prevented platelet O2⫺ release but decreased gramicidin-induced O2⫺ release significantly (n ⫽ 4, P ⬍ 0.05).

suggests the existence of a total number of five to seven intermediate-conductance IKCa-channels in each platelet (Mahaut-Smith, 1995). A functional role of large or intermediate conductance KCa-channels is supported by the present finding that platelethyperpolarization is induced by 11,12-EET and can be prevented by charybdotoxin. Such complete reversal is in contrast to the effects of EDHF in resistance vessels of mice (Brandes et al., 2000), in guinea-pig carotid arteries (Chataigneau et al., 1998), in bovine (Hecker et al., 1994), or porcine coronary artery segments (Hecker et al., 1994; Fisslthaler et al., 1999), where only the combination of both KCa-channel-inhibitors, charybdotoxin plus apamin, abolishes EDHF-mediated effects. Recruitment of further platelets to an existing thrombus is increased by the release of O⫺2 from the platelets during aggregation (Krötz et al., 2002). Furthermore, in cultured endothelial cells, the NAD(P)H-oxidase-dependent production of O2⫺ is increased by depolarization. The present study shows, that the mechanism of membrane potential dependent superoxide formation also occurs in platelets and can be inhibited by hyperpolarization with 11,12-EETs. Therefore, 11,12-EETs might also decrease platelet recruitment. Assuming that EDHF is identical with 11,12-EETs in certain blood vessels, it may act as an inhibitor of platelet recruitment.

45 Epoxyeicosatrienoic acid activates cloned BKCa channel ␣-subunit through ADP-ribosylation of the G-protein G␣s Mitsuhiro Fukao, Helen S. Mason, Satoshi Nawate, Takamitsu Soma, Ichiro Sakuma, Soichi Miwa, James L. Kenyon, Burton Horowitz and Kathleen D. Keef Recent evidence suggests that cytochrome P450-derived arachidonic acid metabolites, epoxyeicosatrienoic acids (EETs) act as endothelium-derived hyperpolarizing factor EDHF in some arteries. EETs are released from vascular endothelial cells and dilate arteries. The dilatation seems to be caused by activation of large-conductance Ca2⫹-activated K⫹ channels (BKCa) leading to membrane hyperpolarization. Previous studies suggest that EETs activate BKCa channels via ADP-ribosylation of the G-protein G␣s with a subsequent membranedelimited action on the channel in native cells. The present study examined whether this pathway is present in HEK cells when the cloned canine BKCa ␣-subunit (cslo-␣) is expressed without the ␤-subunit. 11,12-EET increased outward K⫹ current in whole-cell recordings of human embryonic kidney (HEK) cells. In cell-attached patches, 11,12-EET also increased the activity of cslo-␣ channels without affecting unitary conductance. This action was mimicked by cholera toxin. The ADP-ribosyltransferase inhibitors 3-aminobenzamide and m-iodobenzylguanidine (MTBG) blocked the stimulatory effect of 11,12-EET. In inside-out patches 11,12EET was without effect on channel activity unless GTP was included in the bathing solution. GTP alone also activated cslo-␣ channels. Dialysis of cells with anti-G␣s antibody completely blocked the activation of cslo-␣ channels by 11,12-EET, whereas anti-G␣i/o and anti-G␤␥ antibodies were without effect. The protein kinase A inhibitor KT5720 and the adenylate cyclase inhibitor SQ22536 did not reduce the stimulatory effect of 11,12-EET on cslo-␣ channels in cell-attached patches. These data suggest that EET leads to G␣s-dependent activation of the cslo-␣ subunits expressed in HEK cells and that the cslo-␤ subunit is not required.

The endothelium releases a number of different contracting and relaxing factors. Amongst these, endothelin, nitric oxide and prostacyclin have been particularly well characterized (Furchgott and Vanhoutte, 1989). The additional factor, epoxyeicosatrienoic acid (EET), which is a product of the cytochrome P450 pathway, is also synthesized and released from the endothelium (Hecker et al., 1994; Campbell et al., 1996). Endothelial cells possess cytochrome mono-oxygenase activity (Abraham et al., 1985; Pinto et al., 1987) and several cytochrome P450 isoforms have been described in endothelial cells. In vitro studies have shown that EETs relax coronary, pial, cerebral, caudal and renal arteries and membrane hyperpolarization has been observed (Hecker et al., 1994; Campbell et al., 1996; Fukao et al., 1997; Eckman et al., 1998). These results suggest that EETs contribute to endotheliumdependent relaxation and hyperpolarization in some blood vessels. The vasodilator action of EET is due to activation of large conductance Ca2⫹-activated K⫹ channels (BKCa). Patch clamp studies have shown that EETs increase the open probability of

EETs activate BKCa channel

357

BKCa channels in native cells (Campbell et al., 1996; Li and Campbell, 1997). EET-induced hyperpolarization is blocked by the BKCa channel blocker iberiotoxin (Eckman et al., 1998). Finally, EET-induced relaxation can be reduced or abolished by either iberiotoxin or tetraethylammonium (Campbell et al., 1996; Li and Campbell, 1997; Eckman et al., 1998; Li et al., 1999). Previous studies of native cells suggest that EETs may enhance BKCa activity by activating the G-protein G␣s (Li and Campbell, 1997) via ADP ribosylation (Li et al., 1999). The mechanism by which EETs modulate BKCa channel activity remains unclear. For example, a specific receptor for EETs has yet to be positively identified. Furthermore, the mechanism by which the G-protein G␣s leads to activation of the BKCa channel activity is unknown. It is possible that G-protein subunits interact specifically with the ␣-subunit of the BKCa channel. Alternatively the ␤-subunit may be involved in the process. Expression systems are particularly useful to address these questions since it is possible to express a known isoform of the BKCa ␣-subunit in the absence of the ␤-subunit and ultimately to manipulate the predicted components of the pathway with molecular biology techniques. The goal of the present study was therefore to determine whether the pathway previously described for EETinduced modulation of BKCa channels in native cells (Campbell et al., 1996; Li and Campbell, 1997; Li et al., 1999) is present when a known isoform of the ␣-subunit of the BKCa channel is expressed in a mammalian cell. BKCa ␣-subunits (cslo-␣) expressed in HEK293 cells give rise to voltage-gated, Ca2⫹-sensitive currents with electrophysiological and pharmacological features similar to those of native BKCa (Adelman et al., 1992; Esguerra et al., 1994; Fukao et al., 1999). The experiments were designed to determine: (a) whether EETs enhance cslo-␣ channel activity in this expression system; (b) whether the G-protein ␣ and/or ␤␥ subunit is involved; (c) whether activation involves ADP-ribosylation; and (d) whether activation of cslo-␣ involves the adenylyl cyclase/PKA pathway or a direct membrane-delimited pathway. 1. MATERIALS AND METHODS

1.1. Expression of cslo-␣ channels The cDNA encoding the ␣-subunit of BKCa channel (cslo-␣) was cloned from canine colonic smooth muscle using reverse transcription and a polymerase chain reaction. The cslo-␣ construct was subcloned into the mammalian expression vector pZeoSV (Invitrogen, CA, USA) and expressed in HEK293 cells by electroporation. Current recording was performed one to four days after the electroporation procedure.

1.2. Electrophysiological recording The patch-clamp technique was used to measure membrane currents in whole cell and isolated patch configurations. Data acquisition and analysis were performed with pClamp software (version 6.0.4., Axon Instruments, USA). Channel open probability (NPo) in patches was determined from recordings of more than three minutes by fitting the sum of Gaussian functions to all-points histogram plot at each potential. Single-channel conductance was determined from all-point amplitude histograms using Fechan and P-stat programs (Axon Instruments).

1.3. Solutions and drugs For whole cell recordings of HEK293 cells, the bath solution contained (mM) NaCl 135, KCl 5, CaCl2 1.8, MgCl2 1, HEPES 10, glucose 10 (pH 7.4) and the pipette solution

358

Mitsuhiro Fukao et al.

contained (mM) KCl 50, KAsp 70, NaCl 8, CaCl2 0.826, MgCl2 1, MgATP 2, NaGTP 0.1, HEPES 10, N-(2-hydroxyethyl)-ethylenediamine-triacetic acid (HEDTA) 1 (pH 7.2). For single channel recordings in the inside-out mode, the bath solution contained (mM) KCl 140, MgCl2 1, HEPES 10, HEDTA 1 (pH 7.2). The concentration of free Ca2⫹ in the bath solution was changed ranging from 10⫺8 M to 10⫺4 M to determine the Ca2⫹ sensitivity of BKCa channels. The pipette solution contained (mM) KCl 140, CaCl2 1.8, MgCl2 1, HEPES 10 (pH 7.4). For single channel recordings in the cell-attached mode, the bath solution contained (mM) KCl 140, CaCl2 1.8, MgCl2 1, HEPES 10, glucose 10 (pH 7.4), and the pipette solution contained (mM) KCl 140, CaCl2 1.8, MgCl2 1, HEPES 10 (pH 7.4). All patch-clamp experiments were performed at room temperature (22 ⬚C). 11,12- EET was purchased from Cayman (MI, USA). KT5720, SQ22536, cholera toxin and all antibodies were obtained from Calbiochem (CA, USA). Anti-G␣i/o is a mixture of two antibodies: (a) Anti-Gi␣-1-and Gi␣-2 subunit antibody, (b) Anti-Gi␣-3 and Go␣-subunit antibody. 3-aminobenzamide (3-AM) and m-iodobenzylguanidine (MIBG) were obtained from Sigma (MO, USA). 11,12-EET was dissolved in ethanol and KT5720, forskolin in DMSO. Solvent per se had no effect on channel activity at final concentration (ethanol: 0.03%, DMSO: 0.1%).

1.4. Statistics Data are expressed as mean ⫾ SEM. Statistical significance between groups was determined using ANOVA for repeated measures. A Student’s t-test was used to examine the significance of differences between observations within groups. P less than 0.05 was considered significant. 2. RESULTS

2.1. Characterization of cslo-␣ currents expressed in HEK293 cells The endogenous currents of HEK293 cells and the currents recorded in cells expressing cslo-␣ were characterized. Endogenous currents (0.29 ⫾ 0.04 nA at ⫹ 50 mV, n ⫽ 10) are much smaller in amplitude than those recorded from cells expressing cloned cslo-␣ channels (4.56 ⫾ 0.42 nA at ⫹ 50 mV, n ⫽ 10). Iberiotoxin (10⫺7 M) inhibited the cslo-␣ current. The current remaining in the presence of iberiotoxin was not different from that of native currents.

2.2. RT-PCR examination of cslo-␤ subunit in HEK293 cells Native HEK293 cells were examined for endogenous expression of BKCa ␤-subunit mRNA and compared to freshly isolated human jejunal smooth muscle cells. No detectable product was observed in HEK293 cells whereas abundant message was detected in jejunal cells (data not shown). These results suggest that the actions of EET on cslo-␣ subunits can be examined in HEK293 cells in the absence of the cslo-␤ subunit.

2.3. Whole cell cslo-␣ currents Experiments were undertaken to determine the action of 11,12-EET on cells expressing cslo-␣ using the whole-cell patch-clamp mode. Cytosolic Ca2⫹ concentration was buffered at 10⫺5 M with HEDTA in these experiments. Addition of 11,12-EET (10⫺6 M) to the

EETs activate BKCa channel

359

Figure 45.1 Effect of 11,12-EET on whole-cell cslo-␣ current expressed in HEK293cells. (A) Representative traces of cslo-␣ current expressed in HEK293 cells in the absence and presence of 11,12-EET (10⫺6 M). Membrane potential was held at ⫺70 mV and stepped at 15-s intervals to potentials between ⫺50 mV and ⫹80 mV in 10 mV increments for 200 ms and then held at ⫺30 mV for 20 ms. (B) Time course of the steady state current amplitude before and after treatment of 11,12-EET (10⫺6 M) at a membrane potential of ⫹50 mV. (C) current-voltage relationship of cslo-␣ currents before (䊉) and after (䊊) application of 11,12-EET (10⫺6 M). (D) averaged cslo-␣ currents of control and after application of 11,12-EET (10⫺6 M). Data shows averaged peak currents at voltage clamp steps from ⫺70 to ⫹50mV at 30-s intervals. Values are means⫾SEM (n⫽8). The asterisk indicates a statistically significant difference (*P ⬍ 0.05) compared with control.

bathing solution led to a significant increase in whole-cell outward current. Steady state current was obtained after approximately 5 min. A representative voltage-current relationship of cslo-␣ current before and after treatment of 11,12-EET is shown in Figure 45.1. In the eight cells tested, 11,12-EET significantly (P ⬍ 0.05) increased outward current amplitude two-fold at ⫹ 50 mV (Figure 45.1).

2.4. cslo-␣ channel activity in cell-attached patches Additional experiments were performed to determine whether changes also occur in single channel activity recorded in cell-attached patches. 11,12-EET increased cslo-␣ channel activity in a concentration-dependent manner (Figure 45.2). 11,12-EET at concentrations between 10⫺7 and 10⫺6 M produced a 2–3.5-fold increase in NPo of cslo-␣ channels (Figure 45.2) while having no effect on the unitary conductance (control; 238 ⫾ 9.5, 11,12-EET 10⫺6 M; 246 ⫾ 10, n ⫽ 7, P ⬎ 0.05).

2.5. cslo-␣ channel activity in inside-out patches In contrast to the stimulatory effect of 11,12-EET on cslo-␣ channel in cell-attached patches, 11,12-EET did not significantly affect cslo-␣ channel activity in inside-out patches

360

Mitsuhiro Fukao et al.

Figure 45.2 Effect of 11,12-EET on cslo-␣ channel activity in cell-attached patches. (A) representative recording of cslo-␣ channel under control condition (top trace) and after application of 3 ⫻ 10⫺7 M 11,12-EET (middle trace) and 10⫺6 M 11,12-EET (bottom trace) to the bath solution. Cells were held at ⫹40 mV. (c) the closed state. (B) Summary of the effect of 11,12-EET (10⫺7, 3 ⫻ 10⫺7 and 10⫺6 M) on NPo of the cslo-␣ channels (n ⫽ 6–13). Values are means ⫾ SEM. The asterisk indicates statistically significant differences (*P ⬍ 0.05) compared with control.

when applied to the cytosolic surface of the membrane (n ⫽ 12). Since activation of channels by 11,12-EET may involve phosphorylation, additional experiments were undertaken with ATP. Application of ATP (10⫺3 M) to the bath solution had no effect on cslo-␣ channel activity (n ⫽ 8). Likewise 11,12-EET was without effect in the presence of ATP (n ⫽ 4). These results suggest that ATP alone is insufficient to support activation of cslo-␣ channels by 11,12-EET in isolated patches.

2.6. Effect of GTP There is evidence from native cell experiments that EET can lead to activation of BKCa channels via G␣s (Li et al., 1997). To explore the role of G-proteins in the used expression system, additional experiments were undertaken with GTP and the non-hydrolyzable analogs GDP␤S. GTP (10⫺4 M) significantly increased cslo-␣ channel activity in inside-out patches without affecting the single-channel conductance. In the presence of 10⫺4 M GTP, addition of 11,12-EET (10⫺6 M) led to a further increase in cslo-␣ channel activity. GDP␤S (2⫻10⫺4 M) completely blocked the stimulatory effect of GTP as well as GTP plus 11,12-EET. These data suggest that 11,12-EET activates cslo-␣ via a GTP-dependent mechanism.

2.7. Anti-G␣s antibody To further investigate the nature of the GTP-dependent response, antibodies to various G-proteins were tested in experiments using the whole-cell configuration. Specific antibodies against G␣s, G␣i/o and G␤␥ were included in the pipette solution. After obtaining a stable whole-cell current, 11,12-EET was added to the bathing solution to activate cslo-␣ channels. There was no significant difference in the ability of 11,12-EET to stimulate cslo-␣ current in the presence of anti-G␣i/o or anti-G␤␥ antibody. In contrast, 11,12-EET was without effect on cslo-␣ current in the presence of anti-G␣s antibody (Figure 45.3) suggesting that 11,12-EET activated cslo-␣ via G␣s.

EETs activate BKCa channel

361

Figure 45.3 Effect of various agents, which modify G-protein function on responses to 11,12-EET in HEK293 cells. (A) Effect of G-protein antibodies. Bar graph shows summary of the effect of anti-G␣s, anti-G␣i/o, and anti-G␤␥ antibodies on the stimulatory effect of 11,12-EET of cslo-␣ current in whole-cell patch clamp mode. Antibodies were included in the pipette solution and dialyzed into cells. Peak current attained during a voltage step from ⫺70 to ⫹50 mV in the presence of 11,12EET was normalized to the current obtained in the absence of 11,12-EET. Antibodies were diluted to produce a final concentration, which was two times greater than that necessary for a western blot (1:500) (n ⫽ 610). (B) Effect of the G␣s activator cholera toxin (CTX). Bar graph shows summary of the effect of 11,12EET (10⫺6 M, n ⫽ 13), CTX (100 ng/ml, n ⫽ 8) and CTX plus 11,12-EET (10⫺6 M, n ⫽ 5) on cslo-␣ channel activity in cell-attached patches. Channel activity following 15 min exposure to CTX was compared to the control channel activity (normalized to 1). Membrane potential was clamped at ⫹40 mV. (C) Effect of ADP-ribosyltransferase inhibitors on the stimulatory effect of 11,12-EET on cslo-␣ channel activity in cell-attached patches. Channel activity in the presence of inhibitor plus 11,12-EET was compared to the activity obtained with 11,12-EET in the absence of inhibitor (normalized to 1). The ADP-ribosyltransferase inhibitors, 3-aminobenzamide (3-AB, n⫽6) and MIBG, n⫽7 both blocked the stimulatory effect of 11,12-EET on cslo-␣ channel. Membrane potential was clamped at ⫹40 mV. Values are means ⫾ SEM. The asterisk indicates a statistically significant difference (*P ⬍ 0.05) compared with control.

2.8. Cholera toxin To provide further evidence for coupling of G␣s to cslo-␣ channels, cholera toxin which activates G␣s by ADP-ribosylation was tested. Inclusion of cholera toxin in the bathing solution gave rise to a significant increase in cslo-␣ channel activity in cell-attached patches

362

Mitsuhiro Fukao et al.

(Figure 45.3) without a change in single-channel conductance (control; 248 ⫾ 13 pS, cholera toxin 250 ⫾ 12 pS, n ⫽ 8). In the presence of cholera toxin, 11,12-EET did not produce a further increase cslo-␣ channel activity (Figure 45.3).

2.9. Mono-ADP-ribosyltransferase inhibitors Activation of native BKCa channels by EET involves ADP-ribosylation of G␣s (Li et al., 1999). To determine whether this same pathway is present in HEK293 cells, the effect of two different inhibitors of mono-ADP-ribosyltransferase, 3-aminobenzamide (3-AM) (Purnell and Whish, 1980) and MIBG (Smets et al., 1990) was investigated. 3-AM (10⫺3 M) did not affect basal channel activity in cell-attached patches. However, in the presence of 3-AM the stimulatory effect of 11,12-EET (10⫺6 M) was abolished. Likewise, MIBG (10⫺4 M) was without effect on basal channel activity but blocked the stimulatory effect of 11,12-EET on cslo-␣ channels (Figure 45.3).

2.10. KT 5720 and SQ22536 The results suggest that 11,12-EET activates BKCa via ADP-ribosylation of the G-protein G␣s. G␣s is a well-known activator of the adenylyl cyclase/PKA pathway. To investigate the role of this pathway in the actions of G␣s and 11,12-EET additional experiments were undertaken. The PKA inhibitor KT5720 (2 ⫻ 10⫺7 M) was without effect on basal cslo-␣ channel activity in cell-attached patches (Figure 45.4). In addition, the stimulatory effect of 11,12-EET on cslo-␣ channel was not inhibited by pretreatment with KT5720. In contrast, KT5720 abolished the stimulatory effect of the adenylyl cyclase activator forskolin on cslo-␣ channels. The adenylate cyclase inhibitor SQ22536 (2 ⫻ 10⫺4 M) was also without affect on basal cslo-␣ channel activity in cell-attached patches. Stimulation of BKCa channels by either 11,12-EET or cholera toxin was unchanged in the presence of SQ22536 (Figure 45.4). These results indicate that the adenylyl cyclase/PKA pathway is present in HEK293 cells but that this pathway cannot represent the predominant mechanism by which 11,12-EET and G␣s activate cslo-␣ channels. 3. DISCUSSION 11,12-EET is a cytochrome P450 product of the arachidonic acid cascade that is synthesized and released by the endothelium and may serve as one of several different endotheliumderived factors which relax and hyperpolarize the adjacent smooth muscle (Hecker et al., 1994; Campbell et al., 1996). The present study found that 11,12-EET leads to activation of the cloned ␣-subunit of the BKCa channel (cslo-␣) when expressed in HEK293 cells. This activation involves a novel pathway in which 11,12-EET leads to ADP-ribosylation of the G-protein G␣s. G␣s in turn activates cslo-␣ via a membrane-delimited pathway that is independent of PKA and may involve a direct action of G␣s on the channel. EETs relax blood vessels by activating BKCa channels in the smooth muscle (Hecker et al., 1994; Campbell et al., 1996; Eckman et al., 1998). In the present study, 11,12-EET stimulated cslo-␣ current in whole-cell recordings and enhanced the activity of cslo-␣ channels in cell-attached patches without a change in single-channel conductance. GTP per se gave rise to a significant increase in cslo-␣ channel activity that was blocked by GDP␤S. These data suggest that the predominant action of G-proteins on cslo-␣ channels in HEK293 cells is stimulatory. In the presence of GTP, 11,12-EET caused a significant increase in channel

EETs activate BKCa channel

363

Figure 45.4 Effect of blockers of the adenylyl cyclase/PKA pathway on the stimulatory effect of 11,12-EET on cslo-␣ channel activity in cell-attached patches. (A) Representative recording of cslo-␣ channel recorded at a membrane potential of ⫹40 mV (top trace). The PKA inhibitor KT 5720 alone (2 ⫻ 10⫺4 M) was without effect on channel activity (middle trace) and 11,12-EET (10⫺6 M) still enhanced cslo-␣ channel activity in the presence of KT5720 (bottom trace). Cytosolic Ca2⫹ concentration was maintained at 10⫺6 M. (c) the closed state. (B) Summary of the effect of KT5720 (n ⫽ 16), KT5720 ⫹ 11,12-EET (n ⫽ 9), forskolin (10⫺5 M, n ⫽ 7) and KT5720 plus forskolin (n ⫽ 6) on cslo-␣ channel activity. (C) Effect of the adenylyl cyclase inhibitor SQ22536 on the actions of 11,12-EET and CTX on HEK293 cells. Bar graph shows summary of the effects of SQ22536 (2 ⫻ 10⫺4 M, n ⫽ 17), 11,12-EET (10⫺6 M, n ⫽ 13); SQ22536 plus 11,12-EET (n ⫽ 7) and SQ22536 plus CTX (100 ng/ml; n ⫽ 6) on cslo-␣ channel activity in cell-attached patches. Effects on channel activity were compared to the control channel activity (normalized to 1). Values are means⫾SEM (n⫽6–10). The asterisk indicates a statistically significant difference (*P ⬍ 0.05) compared with control.

activity and this effect was also blocked by GDP␤S. In contrast, in the absence of GTP, 11,12-EET was without effect. Thus, 11,12-EET appears to activate channels via a GTP-dependent mechanism. Since anti-G␣s antibody but not anti-Gi/o or anti-G␤␥ antibodies blocked the actions of 11,12-EET, this suggests that 11,12-EET stimulates cslo-␣ via the GTP-binding protein G␣s. The known G␣s activator cholera toxin also enhanced BKCa channel activity providing additional evidence for coupling between G␣s and cslo-␣ channels. This conclusion is in agreement with previous studies of 11,12-EET in native bovine coronary artery cells (Li and Campbell, 1997). Recently, it has been suggested that 11,12-EET can activate mono-ADP-ribosyltransferase leading to the transfer of ADP-ribose to the 52-kDa G-protein G␣s, resulting in activation

364

Mitsuhiro Fukao et al.

of BKCa channels in small bovine coronary arteries (Li et al., 1999). A similar result was previously reported for EET in the rat liver (Seki et al., 1992). In agreement with these data, the stimulatory effect of 11,12-EET on cslo-␣ was blocked by two different mono-ADPribosyltransferase inhibitors, 3-AM and MIBG. This suggests that ADP-ribosylation of G␣s is also important in the regulation of the cloned ␣-subunit of BKCa by 11,12-EET. The pathway by which 11,12-EET leads to activation of mono-ADP-ribosyltransferase remains unclear. A high-affinity binding site for 14(R),15(S)-EET in guinea-pig mononuclear membranes has been reported, suggesting that a receptor for EET may exist (Wong et al., 1993). Thus, 11,12-EET may stimulate specific receptors that activate mono-ADPribosyltransferase leading to ADP ribosylation of G␣s. This action mimics cholera toxin, which is an exogenous ADP-ribosyltransferase also activating G␣s by ADP ribosylation. In both native cell experiments and in expression systems, PKA activation leads to an increase in BKCa channel activity (Standen and Quayle, 1998) via phosphorylation of serine 869 (Nara et al., 1998). Indeed, BKCa channel activity was increased in the present study by the adenylyl cyclase activator forskolin. Inhibition of this effect by the PKA inhibitor KT 5720 implies the existence of a functional adenylyl cyclase/PKA pathway in HEK293 cells and the possibility was considered that this pathway might contribute to the G␣s-dependent responses to 11,12-EET. However, in cell-attached patches the stimulatory effect of 11,12-EET was not blocked by either the adenylyl cyclase inhibitor SQ22536 nor the PKA inhibitor KT 5720 providing direct evidence that the actions of 11,12-EET are independent of the adenylyl cyclase/PKA pathway. This conclusion is in agreement with the finding that 11,12-EET relaxes the bovine coronary artery without a significant change in tissue levels of either cAMP or cGMP (Campbell et al., 1996). In studies of native cells, EET activates BKCa channels through a PKA-dependent mechanism in renal arteries (Imig et al., 1999) whereas in porcine (Hayabuchi et al., 1998b) and bovine coronary arteries (Campbell et al., 1996) a PKA-independent pathway has been proposed. Multiple isoforms of adenylate cyclase and PKA exist (Houslay and Milligan, 1997). The variable role of PKA in the actions of EET may be related to: (a) the presence of different isoforms of adenylyl cyclase and PKA in different cells; (b) the quantity of isoforms present; and (c) the degree of coupling between G␣s and adenylyl cyclase. In HEK293 cells the membrane-delimited actions of G␣s appear to far outweigh those of the adenylyl cyclase/PKA pathway since SQ22536 was also without effect on cholera toxin, which activates all G␣s within the cell. This suggests very poor coupling between G␣s and adenylyl cyclase in these cells. Thus in HEK293 cells 11,12-EET appears to stimulate cslo-␣ via a direct membrane-delimited action of G␣s. The nature of this interaction between channel and G-protein requires further investigation but appears to involve ADP-ribosylation of G␣s. BKCa channels play a fundamental role in the regulation of membrane potential in smooth muscle, particularly under circumstances where intracellular calcium is elevated (Brayden and Nelson, 1992). The activity of these channels can be importantly modulated by a variety of different physiological stimuli including EET. Native BKCa channels are composed of pore forming ␣-subunits (i.e. ␣-slo) plus a regulatory ␤-subunit (predominantly ␤1 in smooth muscle (Jiang et al., 1999) raising the possibility that the ␤-subunit may play a role in regulation of BKCa channel activity by G␣s. HEK293 cells transfected with specific BKCa subunits provide an excellent system to investigate this issue because endogenous currents in general are small and there are no endogenous BKCa currents (Yu and Kerchner, 1998; Fukao et al., 1999) or message encoding the cslo-␤ subunit. Accordingly, the present results indicate that cslo-␣ channels are expressed in the absence of ␤-subunits. Further evidence of the lack of ␤-subunits is that the voltage for half maximal activation of the expressed

EETs activate BKCa channel

365

BKCa currents with 10⫺5 M free Ca2⫹ is ⫹20 mV (Fukao et al., 1999), similar to values reported by others for activation of the ␣-slo subunits in the absence of the ␤-slo subunits (Toro et al., 1998). Thus, the present study suggests that 11,12-EET activates BKCa channels through a direct action on the ␣-subunit independent of the ␤-subunit. However, it does not rule out a modulatory role for the ␤-subunit in this process. 4. CONCLUSION The cslo-␣ expressed in HEK293 cells is activated by 11,12-EET in both the whole-cell and single-channel configuration. Activation involves ADP ribosylation of G␣s but is independent of the adenylyl cyclase/PKA pathway suggesting a direct, membrane-delimited pathway. These results agree well with previous studies of native cells (Campbell et al., 1996; Li and Campbell, 1997; Li et al., 1999) and suggest that the ␤-subunit of BKCa is not required for this pathway.

46 Different role of epoxyeicosatrienoic acids (EET11,12) in EDHF-mediated relaxation in small porcine coronary and pulmonary arteries Guo-Wei He, Wei Zou, Yu Huang, APC Yim and Qin Yang The cytochrome P450-monooxygenase metabolites of arachidonic acid epoxyeicosatrienoic acids (EETs), such as EET11,12 , may be endothelium-derived hyperpolarizing factor (EDHF) in various blood vessels. The present study was designed to examine the role of EET11,12 in small porcine coronary and pulmonary arteries. Small porcine coronary and pulmonary arteries (diameter 200–450 ␮m) were studied in a myograph. The artery rings were set at 90% of the circumference at 100 mmHg for coronary, and 40 mmHg for pulmonary arteries, respectively. During contraction with U46619, EET11,12 or bradykinin-induced relaxation was obtained in the presence of inhibitors for cyclooxygenase (indomethacin) and nitric oxide (NO) synthase (NG-nitro-L-arginine), and a scavenger of NO, hemoglobin. EET11,12 induced a concentrationdependent relaxation in coronary arteries with a maximal relaxation that was significantly less than that in response to bradykinin. In contrast, in pulmonary arteries, bradykinin induced a marked relaxation whereas EET11,12 did not have any effect. These experiments suggest that in small porcine coronary arteries, EET11,12 may partially mimic the action of EDHF whereas in pulmonary arteries, this substance is unlikely to be involved in the EDHF-mediated response.

1. INTRODUCTION Endothelium-dependent relaxations are mediated by a variety of endothelium-derived relaxing factors (EDRFs) including nitric oxide (NO), prostacyclin, and endothelium-derived hyperpolarizing factor (EDHF) (Furchgott and Zawadzki, 1980; Ignarro et al., 1987; Chen et al., 1988; Félétou and Vanhoutte, 1988; He et al., 1996; Ge et al., 2000). EDHF may back up or enhance the relaxing action of NO, particularly when the NO-mediated relaxation is impaired, as seen in some pathologic states such as hypercholesterolemia, hypertension, and diabetes mellitus (Cohen and Vanhoutte, 1995). The EDHF-mediated relaxation may involve different pathways of intracellular communication, such as release of potassium and myoendothelial gap junctions (Edwards et al., 1998; Dora et al., 1999; Hutcheson et al., 1999). EDHF may play a more important role in the regulation of vascular tone in microcirculation than in large conductance arteries (Garland, 1995; Bolz et al., 1999). Unlike NO and prostacyclin, the chemical nature of EDHF has not been finally identified. The candidates include epoxyeicosatrienoic acids (EETs), K⫹, anandamide, NO, prostacyclin, ATP, ammonium, and citruline, etc. (Vanhoutte, 1998). In particular, cytochrome P450monooxygenase metabolites of arachidonic acid, possibly endogenous cannabinoids EETs, have been suggested to be EDHF (Campbell et al., 1996; Baron et al., 1997; Zou et al., 2001). EETs are vasodilators, particularly in small and more peripheral vessels such as intestinal microvessels, caudal, cerebral, renal, and coronary arteries, as well as resistance

The different role of EET11,12 367 arterioles of the kidney. EETs hyperpolarize vascular smooth muscle cells by increasing the open probability of KCa channels (Hu and Kim, 1993; Hecker et al., 1994; Li and Campbell, 1997). These characteristics support the hypothesis that EETs are EDHF mimics. The present study was designed to examine the role of EET11,12 in small porcine coronary and pulmonary arteries. 2. METHODS Fresh porcine hearts and lungs collected from a local slaughterhouse were placed in a container filled with cold Krebs’ solution and immediately transferred to the laboratory. Upon receipt of the hearts and lungs, intramyocardial small coronary arteries (usually the tertiary branches of the left anterior descending artery) and small intralobular pulmonary arteries (usually small branches of the upper lobe artery, diameter 200–450 ␮m) were dissected carefully and removed under a microscope. The blood vessels were cleaned of fat and connective tissue and cut into rings of 2-mm length under a microscope. The Krebs’ solution was aerated with a gas mixture of 95% O2–5% CO2 at 37 ⬚C and had the following composition (in mM): NaCl 118.4, KCl 4.7, MgSO4.7H2O 1.2, KH2PO4 1.2, NaHCO3 25, (⫹)-Glucose 11.1, and CaCl2.2H2O 2.5. This gives the following final molar concentration (in mM): Na⫹ 143.4, K⫹ 5.9, Ca2⫹ 2.5, Mg2⫹ 1.2, Cl⫺ 128.7, HCO3⫺ 25, SO42⫺ 1.2, H2PO4⫺ 1.2 and glucose 11.1. During the above procedure, the endothelium was intentionally preserved by cautiously dissecting and mounting the rings. After the rings were mounted in a two-channel myograph (Model 500A; J.P. Trading, Aarhus, Denmark), the vascular rings were normalized under a condition simulating the transmural pressure encountered in vivo in the coronary and pulmonary circulation (He et al., 1996; Ge et al., 2000). Briefly, the rings were stretched progressively until the passive transmural pressure reached 100mmHg (in coronary arteries) or 40mmHg (in pulmonary arteries). The internal circumference was then set to a normalized value, estimated to be equivalent to 90% of the circumference at a passive transmural pressure of 100 or 40 mmHg. The transmural pressure at this point was 75 mmHg (in coronary arteries) or 30 mmHg (in pulmonary arteries). The passive tension at these points was maintained throughout the experiments. The blood vessels were contracted with the thromboxane A2 mimetic U46619 (⫺8.2 log M in coronary arteries and ⫺7.5 log M in pulmonary arteries). These concentrations were chosen from previous studies for the coronary arteries (He et al., 1996; Ge et al., 2000) and from pilot experiments for the pulmonary arteries. EDHF-mediated relaxations were evoked by either bradykinin (⫺10 to ⫺6.5 log M) (Group a) or EET11,12 (⫺10 to ⫺6.5 log M) (Group b) in the presence of the cyclooxygenase inhibitor indomethacin (⫺5.2 log M), the NO synthase inhibitor NG-nitro-L-arginine (⫺3.5 log M), and a potent NO scavenger hemoglobin (⫺4.7 log M). Two rings taken from the same coronary (Group I) or pulmonary (Group II) artery were allocated in the different groups, one for bradykinin and the other for EET11,12. In all experiments, only one concentration–relaxation curve was obtained from each arterial ring. A mean concentration–relaxation curve was calculated from a group of experiments (n ⫽ 8). During the experiment, the myograph chambers were continuously bubbled with a gas mixture of 95% O2 and 5% CO2.

2.1. Data analysis Relaxation was expressed as the percentage decrease of the isometric force induced by U46619. The effective concentration of bradykinin or EET11,12 that caused 50% of maximal

368

Guo-Wei He et al.

relaxation was defined as EC50. The EC50 was determined from each concentration–relaxation curve by a logistic, curve-fitting equation: E ⫽ MAP/(AP ⫹ KP), where E is response, M is maximal relaxation, A is concentration, K is EC50 concentration, and P is the slope parameter. From this fitted equation, the mean EC50 ⫾ SEM was calculated for each group.

2.2. Statistical analysis Data were expressed as means ⫾ SEM and were analyzed with unpaired student’s t-test. Values of P less than 0.05 were considered to be statistically significant.

2.3. Drugs Chemicals used and their sources were as follows: bradykinin, EET11,12, NG-nitro-L-arginine, indomethacin, hemoglobin (Sigma Chemical Co, St Louis, Mo), U46619, (Cayman Chemical, Ann Arbor, Mitch). NG-nitro-L-arginine (dissolved in distilled water) and indomethacin (dissolved in ethanol) were stored at 4⬚C. The solution of U46619, hemoglobin, bradykinin, and EET11,12 were held frozen until required. 3. RESULTS

3.1. Resting force The resting force of the coronary was significantly higher than that of pulmonary arteries (P ⬍ 0.05) (Table 46.1).

3.2. U46619-induced contraction In all pulmonary and coronary arterial rings, U46619 induced a stable and rapidly developed tension (Table 46.1).

3.3. Relaxations to bradykinin In the presence of indomethacin, NG-nitro-L-arginine, and hemoglobin, bradykinin induced a maximal relaxation of 72.8⫾7.8% with an EC50 of ⫺7.30⫾0.89 log M in coronary and 69.6⫾ 6.3% with an EC50 of ⫺7.54⫾0.18 log M in pulmonary artery preparations (Figure 46.1). Table 46.1 Resting force (RF) and U46619-induced contraction in small porcine coronary (CA, group I) and pulmonary arteries (PA, group II). One asterisk indicates P ⬍ 0.05, compared with CA (unpaired t-test) Bradykinin Group (a)

EET11,12 Group (b)

Group I (CA)

Group II (PA)

Group I (CA)

Group II (PA)

1.0 ⫾ 0.1(*) 2.3 ⫾ 0.3

3.2 ⫾ 0.6 4.3 ⫾ 0.6

1.1 ⫾ 0.1(*) 2.7 ⫾ 0.5

RF (mN) 2.5 ⫾ 0.2 Contraction (mN) 4.7 ⫾ 0.5 Note * P ⬍ 0.05.

0

0

20

20 Relaxation (%)

Relaxation (%)

The different role of EET11,12 369

40 60 PA CA

80 100 10

9.5

*

8.5

8

7.5

Bradykinin (–log M)

7

6.5

*

*

*

**

40 60 PA 80

9

*

100 10

CA

9.5

9

8.5

8

7.5

7

6.5

EET (–log M)

Figure 46.1 Mean concentration (log M)-relaxation (% of contraction by U46619) curves to bradykinin (BK; left) and EET11,12 (right) in small porcine coronary (CA) and pulmonary (PA) arteries with endothelium, during contractions to U46619 in the presence of indomethacin (⫺5.2 log M), NG-nitro-L-arginine (⫺3.5 log M), and Hb (⫺4.7 log M). Vertical error bars are 1 SEM. One asterisk indicates P ⬍ 0.05 and two means P ⬍ 0.01 (n ⫽ 8, unpaired student’s t-test).

3.4. Relaxations to EET11,12 In the presence of indomethacin, NG-nitro-L-arginine, and hemoglobin, EET11,12 induced a maximal relaxation of 18.4 ⫾ 3.3% with an EC50 of ⫺8.30 ⫾ 0.82 log M in the coronary arteries. In comparison, EET11,12 did not have any relaxing effect in the pulmonary arteries (Figure 46.1). 4. DISCUSSION The present study suggests that in small porcine coronary and pulmonary arteries that exhibit a large EDHF-mediated relaxation, EET11,12 may play a different role. It may partially mimic the action of EDHF in the coronary but is unlikely to be involved in the EDHF-mediated response in the pulmonary arteries. EDHF plays a role in the endothelium-dependent relaxation of large coronary arteries (Chen et al., 1988; Félétou and Vanhoutte, 1988; He et al., 1996; Ge et al., 2000) including that of humans (He, 1997; Liu et al., 2000; He and Liu, 2001). However, it plays an even more important role in the regulation of the vascular tone in microcirculation than in the large conductance arteries (Bolz et al., 1999; Ge et al., 2000). The present study demonstrates that when the production of NO and prostacyclin is inhibited, EDHF-mediated relaxations to bradykinin average 70% both in porcine coronary or pulmonary small arteries. EDHF-mediated responses exist both in large and small coronary arteries (Chen et al., 1988; He et al., 1996; Ge et al., 2000). However, little is known about the role of EDHF in smaller pulmonary blood vessels, although the endothelium plays an important role in the control of pulmonary arterial tone and blood flow (Hasunuma et al., 1991; Liu et al., 1998). In the lung, vasomotor responses have a major impact on pulmonary vascular resistance, and ventilation–perfusion matching occurs mainly in small, distal resistance pulmonary arteries rather than in large, proximal conduit blood vessels. Hyperpolarization contributes to the endothelium-dependent response to acetylcholine in the rat main pulmonary artery

370

Guo-Wei He et al.

(Chen et al., 1988). EDHF-mediated hyperpolarizations and relaxations also exist in the pulmonary arteries of dog, rat, and piglet (Gambone et al., 1997; Horibe et al., 2000; Torok, 2000; Karamsetty et al., 2001; Ogawa et al., 2001). The present study suggests that the EDHF pathway also plays a role in small porcine pulmonary arteries. EETs share many of the known properties of EDHF. For example, both EETs and EDHF can be synthesized by endothelial cells, are released by the endothelium in response to endothelium-dependent vasodilators, relax vascular smooth muscle, and increase the open probabilities of KCa channels in cell-attached patches (Fultou et al., 1995; Campbell et al., 1996). In addition, both EDHF and EET11,12 affect signaling and proliferation of vascular cell (Fleming, 2001). In the present study, in the presence of indomethacin, NG-nitro-L-arginine, and hemoglobin, EET11,12 only induced a small portion of the relaxation induced by bradykinin. This result does not support the hypothesis that in small porcine coronary arteries, EETs are a major EDHF. At the most, they may partially mimic the effect of the latter. Furthermore, in the small pulmonary arteries, in the presence of the above three inhibitors, EET11,12 did not induce any relaxation, compared to the major response to bradykinin. ACKNOWLEDGMENTS This study was fully supported by grants from the Research Grants Council of the Hong Kong Special Administrative Region (project No. CUHK7246/99M & CUHK4127/01M), China and the Providence St Vincent Medical Foundation, Portland, OR USA.

47 EDHF 2002: the take home message Michel Félétou and Paul M. Vanhoutte

These proceedings attempt to provide a scientifically rational explanation for most, if not all, EDHF-mediated responses. Furthermore, thanks to a better understanding of the phenomenon and to new tools that have been developed, the answer to the fundamental question, whether or not endothelium-derived hyperpolarizing factor (EDHF) is an epiphenomenon, whether or not it has physiopathological relevance, begins to emerge with a clear hint that indeed endothelium-dependent hyperpolarizations contribute to the local regulation of the vascular function. The highlights are: 1. MEMBRANE POTENTIAL OF THE ENDOTHELIAL CELLS The amplitude of the endothelium-dependent hyperpolarization of the smooth muscle cells is closely correlated to the changes in intracellular calcium concentration in the endothelial cells. However, the combination of charybdotoxin plus apamin or raising the extracellular concentration in potassium, blocks the former without affecting (or minimally affecting, in the case of potassium) the latter. These results demonstrate the obligatory role of endothelial potassium channels in the endothelium-dependent hyperpolarization of the smooth muscle cells. IKCa channels have been characterized electrophysiologically and pharmacologically in freshly isolated endothelial cells but little evidence supports the expression of this potassium channel in vascular smooth muscle cells presenting the contractile phenotype. Similarly, the endothelial SKCa channels have been fully characterized with the SK3 subunit being expressed in porcine endothelial cells while in murine endothelium the SK2 subunit could predominate. In contrast, in freshly isolated vascular cells, BKCa channels are preferentially expressed in smooth muscle and poorly expressed in endothelial cells. This is of importance since a response that is blocked by iberiotoxin, a specific inhibitor of BKCa, but not IKCa, can only be explained by the diffusion of a mediator acting on the smooth muscle cells (unless some endothelial cells do express BKCa). However, most of the responses attributed to EDHF are not sensitive to iberiotoxin alone. 2. ENDOTHELIUM-DEPENDENT DEPOLARIZATION The endothelium-dependent depolarization of the smooth muscle cells and the resulting contraction are poorly understood. These depolarizations are closely associated with the EDHF-mediated responses, as they can be observed only in blood vessels that exhibit endothelium-dependent hyperpolarizations and only when the endothelial potassium channels are blocked. Furthermore, in analogy with the EDHF-mediated responses, the endotheliumdependent depolarization of the smooth muscle cells are associated with the depolarization

372

Michel Félétou and Paul M. Vanhoutte

of the endothelial cells. The mechanisms underlying this phenomenon, whether or not a factor is involved and its physiopathological significance deserve to be explored further. 3. GAP JUNCTIONS The role of homocellular coupling has been further documented in the last two years. Gap junctions between endothelial cells facilitate the longitudinal spread of both agonist-induced hyperpolarization and relaxation while the gap junctions between smooth muscle cells, possibly through the diffusion of either IP3 or calcium itself, coordinate the basal tone of entire arteriolar segments. The involvement of myoendothelial gap junctions in EDHFmediated responses, in small blood vessels, has also been substantiated further. These junctions, characterized by pentalaminar structures between adjacent endothelial and smooth muscle cells, are observed in blood vessels (e.g. rat mesenteric artery) that exhibit endothelium-dependent relaxations and hyperpolarizations which are resistant to the combination of inhibitors of cyclooxygenase and nitric oxide synthase. In contrast, blood vessels (e.g. rat femoral artery) in which such responses cannot be evoked do not contain these myoendothelial gap junctions. Further support for an obligatory role of myoendothelial gap junctions in EDHF-mediated responses in the rat mesenteric artery comes from the observation than acetylcholine hyperpolarizes both the endothelial cells from the mesenteric and femoral arteries but that hyperpolarization of the smooth muscle cells only occurs in the former. Finally, the regulation of the expression of myoendothelial gap junctions may explain the increased compensatory role of EDHF-mediated responses in the caudal artery of the hypertensive rat. Whether gap junctions are a privileged site for the transfer of electrical charges or do actually transfer a messenger molecule (EDHF) is still an unresolved question. However, the generation of cAMP is involved in the response by facilitating both myoendothelial gap junction communication and homocellular coupling. 4. K⫹ IONS The conclusion, first demonstrated in rat mesenteric and hepatic arteries, that K⫹ ions flow through the endothelial potassium channels and accumulate in the intercellular space to provoke hyperpolarization of the surrounding smooth muscle cells, has been documented and substantiated further. This mechanism is also involved in the EDHF-mediated responses of the renal arteries of the rats and in humans as well as in small porcine coronary arteries. The two targets of potassium ions on the smooth muscle cells, Kir and the Na⫹/K⫹-ATPase, have been characterized in rat arteries. The inwardly rectifying potassium channels, Kir2.1 and/or Kir2.3, and the Na⫹/K⫹-ATPase (comprising ␣2 and/or ␣3 subunits) are involved, and their relative contribution to endothelium-dependent hyperpolarizations is determined by the membrane potential of the smooth muscle cells. Both contribute at potentials close to the resting membrane potential while in depolarized preparations only the latter could participate. This dependency on the membrane potential of the ability of K⫹ ions to cause hyperpolarization (and relaxation) of smooth muscle cells may explain the intense controversy which has followed the proposal that K⫹ ions act as EDHF, as in very depolarized tissues the contribution of K⫹ ions in the EDHF-mediated response disappears. An attempt has been made to quantify and model the amount of potassium flowing out of the endothelial cells and accumulating in the intercellular space This model, in favor of a role of K⫹ in EDHF-mediated response, requires further validation.

EDHF 2002: the take home message

373

5. 11–12 EETs In coronary arteries of various species the role of endothelial cytochrome P450 monooxygenase in modulating endothelium-dependent hyperpolarizations has been substantiated further. One major breakthrough came from the synthesis and characterization of specific antagonists of epoxyeicosatrienoic acids (EETs) the 14–15-epoxyeicosa-5Z monoenoic acid (14,15-EEZE) and its methylsulfonimide analogue. This antagonist inhibits endotheliumdependent relaxations and hyperpolarizations resistant to inhibitors of NO-synthase and cyclooxygenase without inhibiting NO-dependent relaxations or those to openers of K-ATP and BKCa. The exact target (receptor, channel or enzyme) of 14,15-EEZE is still unknown as is its exact location (endothelial or smooth muscle cells). Furthermore, the effect of 14,15-EEZE on SKCa and IKCa has to be assessed. Nevertheless, this compound could be a precious tool to identify the role of EETs in EDHF-mediated responses. In the cremaster of the hamster the endothelium-dependent conducted dilatation to focal application of acetylcholine is insensitive to inhibitors of NO-synthase and cyclooxygenase but is blocked by charybdotoxin, iberiotoxin and inhibitors of cytochrome P450 monooxygenase, suggesting that in this vascular bed a release of EETs occurs. Similarly, in gracilis muscle arterioles taken from female eNOS knock-out mice, shear stress releases a transferable cytochrome P450 epoxygenase product. This finding is corroborated by the observation that cyclic stretch and shear stress augments the expression and the activation of cytochrome P450 2C in porcine coronary endothelial cells. Additional roles for EETs have been suggested such as hyperpolarization of platelets and, in the cerebral circulation, an angiogenic effect of astrocyte-derived 8,9-EET that promotes endothelial cell proliferation. 6. HYDROGEN PEROXIDE A novel aspect is the suggestion that H2O2 is an EDHF not only in murine blood vessels but also in the canine coronary and the human mesenteric artery. Whether or not endotheliumdependent hyperpolarizations (and relaxations) and those caused by H2O2 are equally sensitive to the combination of charybdotoxin plus apamin or to iberiotoxin, alone or in combination with apamin, needs to be determined. 7. PHYSIOLOGICAL ROLE FOR EDHF-MEDIATED RESPONSES Two major new findings indicate a role for EDHF-mediated responses in the regulation of the vascular system. First, in anesthetized rats, blockade of NO-synthase increases the basal conductance in the mesenteric bed and in the hindlimb. Combined local application of charybdotoxin plus apamin had no further effect in either vascular bed. However, the acetylcholine-induced increase in conductance was sensitive to both NO-synthase blockade and the combination of charybdotoxin plus apamin (but not iberiotoxin), indicating that an EDHF-mediated response contributes little to basal conductance while it accounts for a significant component of the increase during endothelial stimulation by acetylcholine. Second, in isolated mesenteric arteries from rats with conditional overexpression (⫻3) of small conductance calcium-activated SK3 potassium channels, the sensitivity to the vasoconstrictor phenylephrine is decreased when compared to wild type animals. This is blocked by the combination of charybdotoxin plus apamin. In these animals, the blood vessel diameter is increased and the density of

P450 - Regulation K+ R DAG PLC

AA

IP3

P450

2+

Ca SKCa +

K lbTX

1

A23187, Thapsigargin, Cyclopiazonic acid

ACh, Bk, SP

Endothelial cells

R

R Ca2+

Ca2+

Hyperpolarization

EETs 3 2 4 IKCa EETs K+

SKCa Apamin K

+

Hyperpolarization

K

+

IKCa ChTX

α GA,

–

IKCa I-EBIO

K

K+

Hyperpolarization BKCa

+

Ba2+ Na2/ K+ pump (inhibited by ouabain)

GAP 27 Heptanol, [Ouabain?]

KIR

Smooth muscle cells

Hyperpolarization

= Gap junction = Inhibition

Figure 47.1 EDHF. A unifying explanation EDHF-mediated responses are initiated by an increase in the endothelial [Ca2⫹]i and the consequent activation of endothelial SKCa and IKCa eliciting the hyperpolarization of the endothelial cells. The endothelial hyperpolarization could then either spread to the adjacent smooth muscle cells through myoendothelial gap junctions and/or the efflux of K⫹ through the endothelial SKCa and IKCa channels could elicit the hyperpolarization of the surrounding myocytes by activating KIR and/or the Na⫹/K⫹-ATPase. In some blood vessels, such as large coronary arteries, the activation of endothelial cytochrome P450 and the resulting generation of EETs could be a required step for the regulation of endothelial [Ca2⫹]i (1), KCa activation (2) or gap junction communication (3). The release of EETs, and the subsequent activation of BKCa in smooth muscle cells (4), is unlikely to account for the majority of EDHF-mediated responses in most blood vessels. A23187: calcium ionophore; ACh: acetylcholine; BK: bradykinin; SP: substance P; R: receptor; AA: arachidonic acid; PLC: phospholipase C; PLA2: phospholipase A2; DAG: diacyl-glycerol; P450: cytochrome P450 monooxygenase; EET: epoxy-eicosatrienoic acid; IP3: inositol trisphosphate; Hyperpol. hyperpolarization. Iberiotoxin (IBX) is a specific inhibitor of large conductance calcium-activated potassium channels (BKCa). Charybdotoxin (CTX) is an inhibitor of BKCa, intermediate conductance calcium-activated potassium channel (IKCa) and some voltage-dependent potassium channels. Apamin is a specific inhibitor of small conductance calciumactivated potassium channels (SKCa). 1-ethyl-2-benzimidazolinone (1-EBIO) is a specific opener of IKCa. Barium (Ba2⫹) in the micromolar range is a specific inhibitor of the inward rectifier potassium channel (Kir). Gap27, an eleven amino acid peptide possessing conserved sequence homology to a portion of the second extracellular loop of connexin, 18␣-glycyrrhetinic acid (␣GA) and heptanol are gap junction uncouplers. Ouabain in the sub-micromolar range is an inhibitor of the Na⫹/K⫹-ATPase (Na⫹/K⫹ pump) and at higher concentrations of gap junctions (modified from Busse et al., 2002).

EDHF 2002: the take home message

375

arteries is larger. Suppression of the overexpression by doxycycline provokes an increase in blood pressure, and a marked decrease in the absolute value of the membrane potential of the smooth muscle cells. 8. PATHOPHYSIOLOGY AND THERAPEUTICS Various vascular diseases have already been associated with a reduction in EDHF-mediated responses. Experiments with isolated human arteries taken from biopsies of the myometrium demonstrate that pre-eclamptic pregnant women also have a specific loss of EDHF-mediated responses. Further evidence supports the improvement of EDHF-mediated responses by estrogen supplementation in ovariectomized rats and by angiotensin converting enzyme inhibition in aging rats. 9. CONCLUSION Not so long ago, there were nearly as many potential hypotheses to explain the EDHF phenomenon as there were scientists working on the subject. However, more recent evidence suggests that a common universal mechanism may explain most of the responses attributed to EDHF whatever the blood vessel and the species studied or the geographical location of the scientists involved. This unifying concept involves first the increase of the endothelial intracellular calcium concentration and then the activation of endothelial SKCa and/or IKCa provoking the hyperpolarization of the endothelial cells. In the endothelial cells, the activation of potassium channels and the increase in gap junction communication could be, depending on the tissue studied, under the control of the endothelial cytochrome P450 monooxygenase. The endothelial hyperpolarization could spread electrotonically to the adjacent smooth muscle cells through myoendothelial gap junction especially in blood vessels where these two cell types form a closed syncytium. Additionally, the efflux of potassium through the opening of endothelial SKCa and IKCa could also produce the hyperpolarization of the underlying smooth muscle cells by activating Kir or/and the Na⫹/K⫹-ATPase. These mechanisms are not necessarily mutually exclusive and, in a given blood vessel, they can occur simultaneously, sequentially or act synergistically. The relative proportion of each mechanism will depend on numerous parameters including the state of activation of the smooth muscle cells, the density of myoendothelial gap junction and the level of expression of the relevant isoforms of cytochrome P450 monooxygenase, Kir and Na⫹/K⫹-ATPase (Figure 47.1).

References

Abraham, N.G., Pinto, A, Mullane, K.M., Levere, R.D., Spokas, E. (1985) Presence of cytochrome P-450-dependent monooxygenase in intimal cells of the hog aorta. Hypertension, 7: 899–904. Abudara, V., Eyzaguirre, C., Saez, J.C. (2000) Short- and long-term regulation of rat carotid body gap junctions by cAMP. Identification of connexin43, a gap junction subunit. Adv. Exp. Med. Biol., 475: 359–369. Adeagbo, A.S.O., Triggle, C.R. (1993) Varying extracellular [K⫹]: a functional approach to separating EDHF- and EDNO-related mechanisms in perfused rat mesenteric arterial bed. J. Cardiovasc. Pharmacol., 21: 423–429. Adeagbo, A.S.O., Oyekan, A.O., Joshua, I.G., Falkner, K.C., Prough, R.A., Pierce, W.M. (2001) Epoxyeicosatrienoic acid (EET) release mediates nitric oxide-independent dilation of rat mesenteric vessels. In EDHF 2000, edited by P.M. Vanhoutte, chapter 22, pp. 195–206. London and New York: Taylor & Francis. Adelman, J.P., Shen, K.Z., Kavanaugh, M.P., Warren, R.A., Wu, Y.N., Lagrutta, A., Bond, C.T., North, R.A. (1992) Calcium-activated potassium channels expressed from cloned complementary DNAs. Neuron, 9: 209–216. Aizman, O., Uhlen, P., Lal, M., Brismar, H., Aperia, A. (2001) Ouabain, a steroid hormone that signals with low calcium oscillations. Proc. Nat. Acad. Sci., 98: 13420–13424. Al-Ani, B., Saifeddine, M., Hollenberg, M.D. (1995) Detection of functional receptors for the proteinase-activated-receptor-2-activating polypeptide, SLIGRL-NH2, in rat vascular and gastric smooth muscle. Can. J. Physiol. Pharmacol., 73: 1203–1207. Alkayed, N.J., Narayanan, J., Gebremedhin, D., Medhora, M., Roman, R.J., Harder, D.R. (1996) Molecular characterization of an arachidonic acid epoxygenase in rat brain astrocytes. Stroke, 27: 971–979. Alkayed, N.J., Birks, E.K., Narayanan, J., Petrei, K.A., Kohler-Cabot, A.E., Harder, D.R. (1997) Role of P450 arachidonic acid epoxygenase in the response of cerebral blood flow to glutamate in rats. Stroke, 28: 1066–1072. Alvarez, J., Montero, M., Garcia-Sancho, J. (1992) High affinity inhibition of Ca2⫹-dependent K⫹ channels by cytochrome P-450 inhibitors. J. Biol. Chem., 267: 11789–11793. Amano, K.-I., Hori, M., Ozaki, H., Matsuda, Y., Karaki, H. (1994) Ca⫹⫹ mobilization mediated by endothelin ETA receptor in endothelium of rabbit aortic valve. J. Pharmacol. Exp. Ther., 271: 1359–1364. Amruthesh, S.C., Boerschel, M.F., McKinney, J.S., Willoughby, K.A., Ellis, E.F. (1993) Metabolism of arachidonic acid to epoxyeicosatrienoic acids, hydroxyeicsatetraenoic acids and prostaglandins in cultured rat hippocampal astrocytes. J. Neurochem., 61: 150–159. Andersson, D.A., Zygmunt, P.M., Movahed, P., Andersson, T.L., Hogestatt, E.D. (2000) Effects of inhibitors of small- and intermediate-conductance calcium-activated potassium channels, inwardly-rectifying potassium channels and Na(⫹)/K(⫹) ATPase on EDHF relaxations in the rat hepatic artery. Br. J. Pharmacol., 129: 1490–1496.

References

377

Andrews, K.L., Triggle, C.R., Ellis, A. (2002) NO and the vasculature: Where does it come from and what does it do? Heart Failure Reviews., 7: 423–445. Andriambeloson, E., Kleschyov, A.L., Muller, B., Beretz, A., Stoclet, J.C., Andriantsitohaina, R. (1997) Nitric oxide production and endothelium-dependent vasorelaxation induced by wine polyphenols in rat aorta. Br. J. Pharmacol., 120: 1053–1058. Arreola, J., Melvin, J.E., Begenisich, T. (1996) Activation of calcium-dependent chloride channels in rat parotid acinar cells. J. Gen. Physiol., 108: 35–47. Arystarkhova, E., Wetzel, R.K., Asinovski, N.K., Sweadner, K.J. (1999) The gamma subunit modulates Na⫹ and K⫹ affinity of the renal Na,K-ATPase. J. Biol. Chem., 274: 33183–33185. Ashworth, J.R., Warren, A.Y., Baker, P.N., Johnson, I.R. (1996) A comparison of endothelium-dependent relaxation in omental and myometrial resistance arteries in pregnant and non-pregnant women. Am. J. Obstet. Gynecol., 175: 1307–1312. Ashworth, J.R., Warren, A.Y., Baker, P.N., Johnson, I.R. (1997) Loss of endothelium-dependent relaxation in myometrial resistance arteries in pre-eclampsia. Br. J. Obstet. Gynaecol., 104: 1152–1158. Atkinson, J., Tatchum-Talom, R., Corman, B. (1994) Effects of chronic ANG I-converting enzyme inhibition on aging processes. III. endothelial function of mesenteric arterial bed of rats. Am. J. Physiol., 267: R136–R143. Ayajiki, K., Fujioka, H., Okamura, T. (2000) Mechanisms underlying endothelium-dependent, nitric oxide/prostacyclin-independent, acetylcholine-induced relaxation in canine corpus cavernosum Naunyn Schmiedebergs Arch. Pharmacol., 362: 448–451. Balzer, M., Lintschinger, B., Groschner, K. (1999) Evidence for a role of Trp proteins in the oxidative stress-induced membrane conductances of porcine aortic endothelial cells. Cardiovasc. Res., 42: 543–549. Banerjee, M., Kang, K.H., Morrow, J.D., Roberts, L.J., Newman, J.H. (1992) Effects of a novel prostaglandin, 8-epiPGF2a, in rabbit lung in situ. Am. J. Physiol., 263: H660–H663. Bang, L., Nielsen-Kudsk, J.E., Gruhn, N., Trautner, S., Theilgaard, S.A., Olesen, S.P., Boesgaard, S., Aldershvile, J. (1998) Hydralazine-induced vasodilation involves opening of high conductance Ca2⫹-activated K⫹ channels. Eur. J. Pharmacol., 361: 43–49. Barker, D.J.P., Bull, A.R., Osmond, C. (1990) Fetal and placental size and risk of hypertension in later life. Br. Med. J., 301: 259–261. Barlow, R.S., White, R.E. (1998) Hydrogen peroxide relaxes porcine coronary arteries by stimulating BKCa channel activity. Am. J. Physiol., 275: H1283–H1289. Barlow, R.S., El-Mowafy, A.M., White, R.E. (2000) H2O2 opens BKCa channels via the PLA2-arachidonic acid signaling cascade in coronary artery smooth muscle. Am. J. Physiol., 279(2): H475–H83. Baron, A., Frieden, M., Chabaud, F., Beny, J.-L. (1996) Ca2⫹-dependent non-selective cation and K⫹ channels, activated by bradykinin in pig coronary artery endothelial cells. J. Physiol., 493: 691–706. Baron, A., Frieden, M., Beny, J.L. (1997) Epoxyeicosatrienoic acids activate a high-conductance, Ca2⫹-dependent K⫹ channel on pig coronary artery endothelial cells. J. Physiol., 504: 537–543. Bartlett, I.S., Segal, S.S. (2000) Resolution of smooth muscle and endothelial pathways for conduction along hamster cheek pouch arterioles. Am. J. Physiol., 278: H604–H612. Bassenge, E. (1991) Antiplatelet effects of endothelium-derived relaxing factor and nitric oxide donors. Eur. Heart J., 12(Suppl E): 12–15. Bauersachs, J., Popp, R., Hecker, M., Sauer, E., Fleming, I. Busse, R. (1996) Nitric oxide attenuates the release of endothelium-derived hyperpolarizing factor. Circulation, 94: 3341–3347. Bauersachs, J., Christ, M., Ertl, G., Fisslthaler, B., Busse, R., Fleming, I. (2001) Cortisol increases EDHFmediated relaxations in porcine coronary arteries and up-regulates the expression of cytochrome P450 2C9. In EDHF 2000, edited by P.M. Vanhoutte, pp. 173–179. London and New York: Taylor & Francis. Bauersachs, J., Christ, M., Ertl, G., Michaelis, U.R., Fisslthaler, B., Busse, R., Fleming, I. (2002) Cytochrome P450 2C expression and EDHF-mediated relaxation in porcine coronary arteries is increased by cortisol. Cardiovasc. Res., 54: 669–675. Beech, D.J. (1997) Actions of neurotransmitters and other messengers on Ca2⫹ channels and K⫹ channels in smooth muscle cells. Pharmacol. Ther., 73: 91–119.

378

References

Beguin, P., Wang, X., Firsov, D., Puoti, A., Claeys, D., Horisberger, J.D., Geering, K. (1997) The ␥ subunit is a specific component of the Na,K-ATPase and modulates its transport function. EMBO J., 16: 4250–4260. Beguin, P., Crambert, G., Guennoun, S., Garty, H., Horisberger, J.-D., Geering, K. (2001) CHIF, a member of the FXYD protein family, is a regulator of Na,K-ATPase distinct from the ␥-subunit. EMBO J., 20: 3993–4002. Bény J.-L. (1997) Electrical coupling between smooth muscle cells and endothelial cells in pig coronary arteries. Pflügers Archiv., 433: 364–367. Beny, J.-L. (1999) Information networks in the arterial wall. News in Physiol. Sci., 14: 68–73. Bény, J.-L., Brunet, P.C. (1988) Neither nitric oxide nor nitroglycerin accounts for all the characteristics of the endothelially mediated vasodilatation of pig coronary arteries. Blood Vessels, 25: 308–311. Beny, J-L., Pacicca, C. (1994) Bidirectional electrical communication between smooth muscle and endothelial cells in the pig coronary artery. Am. J. Physiol., 266: H1465–H1472. Beny, J.L., Schaad, O. (2000) An evaluation of potassium ions as endothelium-derived hyperpolarizing factor in porcine coronaries arteries. Br. J. Pharmacol., 131: 965–973. Beny, J.-L., von der Weid, P.Y. (1991) Hydrogen peroxide: an endogenous smooth muscle cell hyperpolarizing factor. Biochem. Biophys. Res. Commun., 176: 378–384. Bény, J.-L., von der Weid, P.-Y. (1993) Hyperpolarizing factors. In Coronary Artery Disease. Reviews in Depth, edited by (Sobel, B.E.), pp. 50–57 Current Science, Bell and Bain Ltd, Glasgow. Bény, J.-L., Zhu, P., Haefliger, I.O. (1997) Lack of bradykinin-induced smooth muscle cell hyperpolarization despite heterocellular dye coupling and endothelial cell hyperpolarization in porcine ciliary artery. J. Vasc. Res., 34: 344–350. Berg, T., Koteng, O. (1997) Signalling pathways in bradykinin- and nitric oxide-induced hypotension in the normotensive rat; role of K⫹-channels. Br. J. Pharmacol., 121: 1113–1120. Bergamini, S., Rota, C., Canali, R., Staffieri, M., Daneri, F., Bini, A., Giova, F., Tomasi, A., Iannone, A. (2001) N-acetylcysteine inhibits in vivo nitric oxide production by inducible nitric oxide synthase. Nitric Oxide, 5: 349–360. Berger, M.G., Rusch N.J. (1999) Voltage and calcium-gated potassium channels: functional expression and therapeutic potential in the vasculature. Perspectives in Drug Discovery and Design, 15/16: 313–332. Berman, R.S., Martin, P.E.M., Evans W.H., Griffith, T.M. (2002) Relative contributions of NO and gap junctional communication to endothelium-dependent relaxations of rabbit resistance arteries vary with vessel size. Microvasc. Res., 63: 115–128. Berne, R.M. (1980) The role of adenosine in the regulation of coronary blood flow. [Review] [61 refs] Circ. Res., 47(6):807–813. Beswick, R.A., Zhang, H., Marable, D., Catravas, J.D., Hill, W.D., Webb, R.C. (2001) Long-term antioxidant administration attenuates mineralocorticoid hypertension and renal inflammatory response. Hypertension, 37: 781–786. Betts, L.C., Kozlowski, R.Z. (2000) Electrophysiological effects of endothelin-1 and their relationship to contraction in rat renal arterial smooth muscle. Br. J. Pharmacol., 130: 787–796. Bevans, C.G., Harris, A.L. (1999) Regulation of connexin channels by pH. Direct action of the protonated form of taurine and other aminosulfonates. J. Biol. Chem., 274: 3711–3719. Bilsel, A.S., Moini, H., Tetik, E., Aksungar, F., Kaynak, B., Ozer, A. (2000) 17Beta-estradiol modulates endothelin-1 expression and release in human endothelial cells. Cardiovasc. Res., 46: 579–584. Blanco, G., Mercer, R. W. (1998) Isozymes of the Na,K-ATPase: heterogeneity in structure, diversity in function. Am. J. Physiol., 275: F633–F650. Blanco, G., Koster, J.C., Sánchez, G., Mercer, R.W. (1995a). Kinetic properties of the ␣2␤1 and ␣2␤2 isozymes of the Na,K-ATPase. Biochemistry, 34: 319–325. Blanco, G., Sánchez, G. Mercer, R.W. (1995b). Comparison of the enzymatic properties of the Na,KATPase ␣3␤1 and ␣3␤2 isozymes. Biochemistry, 34: 9897–9903.

References

379

Blanco, G., Melton, R.J. , Sánchez, G. Mercer, R.W. (1999) Functional characterization of a testesspecific ␣-subunit isoform of the Na,K-ATPase. Biochemistry, 38: 13661–13669. Blaustein, M.P. (1993) Physiological effects of endogenous ouabain: control of intracellular Ca2⫹ stores and cell responsiveness. Am. J. Physiol., 264: C1367–C1387. Bolego, C., Cignarella, A., Ruzza, R., Zaarour, C., Messi, E., Zanisi, M., Puglisi, L. (1997) Differential effects of low- and high-dose estrogen treatments on vascular responses in female rats. Life Sci., 60: 2291–2302. Bolton, T.B., Lang, R.J., Takewaki, T. (1984) Mechanisms of action of noradrenaline and carbachol on smooth muscle of guinea-pig anterior mesenteric artery. J. Physiol., 351: 549–572. Bolz, S.S., de Wit, C., Pohl, U. (1999) Endothelium-derived hyperpolarizing factor but not NO reduces smooth muscle Ca2⫹ during acetylcholine-induced dilation of microvessels. Br. J. Pharmacol., 128: 124–134. Bolz, S.S., Fisslthaler, B., Pieperhoff, S., de Wit, C., Fleming, I., Busse, R., Pohl, U. (2000) Antisense oligonucleotides against cytochrome P450 2C8 attenuate EDHF-mediated Ca2⫹ changes and dilation in isolated resistance arteries. FASEB J., 14: 255–260. Bond, C.T., Maylie, J., Adelman, J.P. (1999) Small-conductance calcium-activated potassium channels. Ann. N.Y. Acad. Sci., 868: 370–378. Bondy, S.C., Naderi, S. (1994) Contribution of hepatic cytochrome P450 systems to the generation of reactive oxygen species. Biochem. Pharmacol., 48: 155–159. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal. Biochem., 72: 248–54. Bradley, K.K., Jaggar, J.H., Bonev, A.D., Heppner, T.J., Flynn, E.R., Nelson M.T., Horowitz, B. (1999) Kir2.1 encodes the inward rectifier potassium channel in rat arterial smooth muscle cells. J. Physiol., 515: 639–651. Brandes, R.P., Schmitz-Winnenthal, F.H., Félétou, M., Godecke, A., Huang, P.L., Vanhoutte, P.M., Fleming, I., Busse, R. (2000) An endothelium-derived hyperpolarizing factor distinct from NO and prostacyclin is a major endothelium-dependent vasodilator in resistance vessels of wild-type and endothelial NO synthase knockout mice. Proc. Natl. Acad. Sci. USA, 97: 9747–9752. Brayden, J.E., Nelson, M.T. (1992) Regulation of arterial tone by activation of calcium-dependent potassium channels. Science, 256: 532–535. Breier, G., Albrecht, U., Sterrer, S., Risau ,W. (1992) Expression of vascular endothelial growth factor during embryonic angiogenesios and endothelial cell differentiation., Development, 114: 521–532. Brenner, R., Perez, G.J., Bonev, A.D., Eckman, D.M., Kosek, J.C., Wiler, S.W., Patterson, A.J., Nelson, M.T., Aldrich, R.W. (2000) Vasoregulation by the beta1 subunit of the calcium-activated potassium channel. Nature, 407: 870–876. Brink, P.R. (1998) Gap junctions in vascular smooth muscle. Acta Physiol. Scand., 164: 349–356. Britton, F., Ohya, S., Horowitz, B., Greenwood, I.A. (2002) Comparison of the properties of CLCA1 generated currents and ICl(Ca) in murine portal vein smooth muscle cells. J. Physiol., 539: 107–117. Brownlee, M., Vlassara, H., Kooney, A., Ulrich, P., Cerami, A. (1986) Aminoguanidine prevents diabetes-induced arterial wall protein cross-linking. Science, 232: 1629–1632. Brunet, P.C., Bény, J.-L. (1989) Substance P and bradykinin hyperpolarize, pig coronary artery endothelial cells in primary culture. Blood Vessels, 26: 228–234. Bukauskas, F.F., Jordan, K., Bukauskiene, A., Bennett, M.V., Lampe, P.D., Laird, D.W., Verselis, V.K. (2000) Clustering of connexin 43-enhanced green fluorescent protein gap junction channels and functional coupling in living cells. Proc. Natl. Acad. Sci. USA., 97: 2556–2561. Burghardt, R.C., Barhoumi, R., Sewall, T.C., Bowen, J.A. (1995) Cyclic AMP induces rapid increases in gap junction permeability and changes in the cellular distribution of connexin43. J. Membr. Biol., 148: 243–253. Burnham, M.P., Bychkov, R., Félétou, M., Richards, G.R., Vanhoutte, P.M., Weston, A.H., Edwards, G. (2002) Characterization of an apamin-sensitive small-conductance Ca2⫹-activated K⫹ channel in porcine coronary artery endothelium: relevance to EDHF. Brit. J. Pharmacol., 135: 1133–1143.

380

References

Bush, T.L., Barrett-Connor, E., Cowan, L.D., Criqui, M.H., Wallace, R.B., Suchindran, C.M., Tyroler, H.A., Rifkind, B.M. (1987) Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the lipid research clinics programm follow-up study. Circulation, 75: 1102–1109. Busse, R., Fleming, I. (1998) Pulsatile stretch and shear stress: physical stimuli determining the production of endothelium-derived relaxing factors. J. Vasc. Res., 35: 73–84. Busse, R., Fleming, I., Hecker, M. (1993) Signal transduction in endothelium-dependent vasodilatation. Eur. Heart J., 14: Suppl I: 2–9. Busse, R., Edwards, G., Félétou, M., Fleming, I., Vanhoutte, P.M., Weston, A.H. (2002) EDHF: bringing the concepts together. Trends Pharmacol. Sci., 23: 374–380. Buus, N.H., Simonsen, U., Pilegaard, H.K., Mulvany, M.J. (2000) Nitric oxide, prostanoid and non-NO, non-prostanoid involvement in acetylcholine relaxation of isolated human small arteries. Br. J. Pharmacol., 129: 184–192. Bychkov, R., Pieper, K., Ried, C., Milosheva, M., Bychkov, E., Luft, F.C., Haller, H. (1999) Hydrogen peroxide, potassium currents, and membrane potential in human endothelial cells. Circulation, 99: 1719–1725. Bychkov, R., Burnham, M.P., Richards, G.R., Edwards, G., Weston, A.H., Félétou, M., Vanhoutte, P.M. (2002) Characterisation of a charybdotoxin-sensitive intermediate conductance Ca2⫹-activated K⫹ channel in porcine coronary endothelium: relevance to EDHF. Br. J. Pharmacol., 137(8): 1346–1354. Cai, H., Harrison, D.G. (2000) Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ. Res., 87: 840–844. Cai, S., Garneau, L., Sauve, R. (1998) Single-channel characterization of the pharmacological properties of the K(Ca2⫹) channel of intermediate conductance in bovine aortic endothelial cells. J. Memr. Biol., 163:147–158. Campbell, W.B., Gauthier, K.M. (2002) What is new in endothelium-derived hyperpolarizing factors? Curr. Opinion in Nephrology and Hypert. Res., 11: 177–183. Campbell, W.B., Harder, D.R. (1999) Endothelium-derived hyperpolarizing factors and vascular cytochrome P450 metabolites of arachidonic acid in the regulation of tone. Circ. Res., 84: 484–488. Campbell, W.B., Gebremedhin, D., Pratt, P.F., Harder, D.R. (1996) Identification of epoxyeicosatrienoic acids as endothelium-derived hyperpolarizing factors. Circ. Res., 78: 415–423. Campbell, W.B., Falck, J.R., Gauthier, K. (2001) Role of epoxyeicosatrienoic acids as endotheliumderived hyperpolarizing factor in bovine coronary arteries. Med. Sci. Monit., 7: 578–584. Cannell, M.B., Cheng, H., Lederer, W.J. (1995) The control of calcium release in heart muscle. Science, 268: 1045–1049. Cao, Y., Dreixler, J.C., Roizen, J.D., Roberts, M.T., Houamed, K.M. (2001) Modulation of recombinant small conductance Ca(2⫹)-activated K(⫹) channels by the muscle relaxant chlorzoxalone and structurally related compounds. J. Pharmacol. Exp. Ther., 296: 683–689. Capdevila, J., Gil, L., Orellana, M., Marnett, L.J., Mason, J.I., Yadagiri, P., Falck, J.R. (1988) Inhibitors of cytochrome P-450-dependent arachidonic acid metabolism. Arch. Biochem. Biophys., 261: 257–263. Carpenter, G. (2000) The EGF receptor: a nexus for trafficking and signaling. Bioessays, 22: 697–707. Carr, A.C., Zhu, B.Z., Frei, B. (2000) Potential antiatherogenic mechanisms of ascorbate (vitamin C) and alpha-tocopherol (vitamin E). Circ. Res., 87: 349–354. Castilla, M.A., Arroyo, M.V., Aceituno, E., Aragoncillo, P., Gonzales-Pacheco, F.R., Texeiro, E., Bragado, R., Caramelo, C. (1999) Disruption of cadherin-related junctions triggers autocrine expression of vascular endothelial growth factor in bovine aortic endothelial cells: effects on cell proliferation and death resistance. Circ. Res., 85: 1132–1138. Castle, N.A. (1999) Recent advances in the biology of small conductance calcium-activated potassium channels. Perspect. Drug. Discov., 15/16: 131–154. Chan, A.K., von der Weid, P.-Y. (2002) Effect of PAR2 activation on guinea-pig mesenteric lymphatic vessels. FASEB J., 16: A77.

References

381

Chandy, K.G., Gutman, G.A. (1995) Voltage-gated potassium channels, their electrophysiology and function. In Potassium channel modulators, edited by A.H. Weston and T.C. Hamilton pp. 144–180. Oxford: Blackwell Scientific. Chanson, M., White, M.M., Garber, S.S. (1996) cAMP promotes gap junctional coupling in T84 cells. Am. J. Physiol., 271: C533–C539. Chao, J., Stallone, J.N. et al. (1997) Kallistatin is a potent new vasodilator. J. Clin. Invest., 100(1): 11–17. Chataigneau, T., Félétou, M., Duhault, J., Vanhoutte, P.M. (1998) Epoxyeicosatrienoic acids, potassium channel blockers and endothelium-dependent hyperpolarization in the guinea-pig carotid artery. Br. J. Pharmacol., 123: 574–580. Chaytor, A.T., Evans, W.H., Griffith, T.M. (1997) Peptides homologous to extracellular loop motifs of connexin 43 reversibly abolish rhythmic contractile activity in rabbit arteries. J. Physiol., 503: 99–110. Chaytor, A.T., Evans, W.H., Griffith, T.M. (1998) Central role of heterocellular gap junctional communication in endothelium-dependent relaxations of rabbit arteries. J. Physiol., 508(2): 561–573. Chaytor, A.T., Marsh, W.L., Hutcheson, I.R., Griffith, T.M. (2000) Comparison of glycyrrhetinic acid isoforms and carbenoxalone as inhibitors of EDHF-type relaxations mediated via gap junctions. Endothelium, 7: 265–278. Chaytor, A.T., Martin, P.E., Edwards, D.H., Griffith, T.M. (2001) Gap junctional communication underpins EDHF-type relaxations evoked by ACh in the rat hepatic artery, Am. J. Physiol., 280: H2441–H2450. Chaytor, A.T., Taylor, H.J., Griffith, T.M. (2002) Gap junction-dependent and -independent EDHF-type relaxations may involve smooth muscle cAMP accumulation. Am. J. Physiol., 282: H1548–H1555. Chen, C., Wagoner, P.K. (1991) Endothelin induces a nonselective cation current in vascular smooth muscle cells. Circ. Res., 69: 447–454. Chen, G., Cheung, D.W. (1992a) Pharmacological distinction of the hyperpolarization response to caffeine and acetylcholine in guinea-pig coronary endothelial cells. Eur. J. Pharmacol., 223: 33–38. Chen, G., Cheung, D.W. (1992b) Characterization of acetylcholine-induced membrane hyperpolarization in endothelial cells. Circ. Res., 70: 257–263. Chen, G., Cheung, D.W. (1996) Modulation of endothelium-dependent hyperpolarization and relaxation to acetylcholine in rat mesenteric artery by cytochrome P450 enzyme activity. Circ. Res., 79: 827–833. Chen, G., Cheung, D.W. (1997) Effect of K⫹ channel blockers on ACh-induced hyperpolarization and relaxtion in mesenteric arteries. Am. J. Physiol., 272: H2306–H2312. Chen, G., Suzuki, H. (1989) Some electrical properties of the endothelium-dependent hyperpolarization recorded from rat arterial smooth muscle cells. J. Physiol., 410: 91–106. Chen, G., Suzuki, H. (1990) Calcium dependency of the endothelium-dependent hyperpolarization in smooth muscle cells of the rabbit carotid artery. J. Physiol., 421: 521–534. Chen, G., Suzuki, H., Weston, A.H. (1988) Acetylcholine releases endothelium-derived hyperpolarizing factor and EDRF from rat blood vessels. Br. J. Pharmacol., 95: 1165–1174. Chen, G., Yamamoto, Y., Miwa, K., Suzuki, H. (1991) Hyperpolarization of arterial smooth muscle induced by endothelial humoral substances. Am. J. Physiol., 260: H1888–H1892. Chen, J.K., Wang, D.W., Falck, J.R., Capdevila, J., Harris, R.C. (1999) Transfection of an active cytochrome P450 arachidonic acid epoxygenase indicates that 14,15-epoxyeicosatrienoic acid functions as an intracellular second messenger in response to epidermal growth factor. J. Biol. Chem., 274: 4764–4769. Chen, P.Y., Sanders, P.W. (1991) L-arginine abrogates salt-sensitive hypertension in Dahl/Rapp rats. J. Clin. Invest., 88: 1559–1567. Cheng, H., Lederer, W.J., Cannell, M.B. (1993) Calcium sparks: elementary events underlying excitation– contraction coupling in heart muscle. Science, 262: 740–744. Cherry, P.D., Furchgott, R.F., Zawadzki, J.V., Jothianandan, D. (1982) Role of endothelial cells in relaxation of isolated arteries by bradykinin. Proc. Natl. Acad. Sci. USA, 79: 2106–2110. Chien, S., Li, S., Shyy, Y.J. (1998) Effects of mechanical forces on signal transduction and gene expression in endothelial cells. Hypertension, 31: 162–169.

382

References

Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J., Rutter, W.J. (1979) Isolation of biochemically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 18: 5294–5299. Chomczynski, P., Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156–159. Christ, G.J., Spray, D.C., el-Sabban, M., Moore, L.K., Brink, P.R. (1996) Gap junctions in vascular tissues. Evaluating the role of intercellular communication in the modulation of vasomotor tone. Circ. Res., 79: 631–646. Chueh, S.-C., Guh, J.-H., Chen, J., Lai, M.-K., Teng, C.-M. (2001) Dual effects of ouabain on the regulation of proliferation and apoptosis in human prostatic smooth muscle cells. J. Urol., 166: 347–353. Churchill, G.C., Louis, C.F. (1998) Roles of Ca2⫹, inositol trisphosphate and cyclic ADP-ribose in mediating intercellular Ca2⫹ signaling in sheep lens cells. J. Cell Sci., 111: 1217–1225. Cicala, C. (2002) Protease activated receptor 2 and the cardiovascular system. Br. J. Pharmacol., 135: 14–20. Clausen, T., Nielsen, O.B. (1994) The Na⫹,K⫹-pump and muscle contractility. Acta. Physiol. Scand., 152: 365–373. Clément-Chomienne, O., Ishii, K., Walsh, M.P., Cole, W.C. (1999) Identification, cloning and expression of rabbit vascular smooth muscle Kv1.5 and comparison with native delayed rectifier K⫹ current. J. Physiol., 515: 653–667. Coats, P., Hillier, C. (1999) Determination of a reliable normalization procedure for the study of isolated resistance arteries on the pressure myograph. Exp. Physiol., 84: 1085–1094. Coats, P., Johnston, F. (2001) Endothelium-derived hyperpolarizing factor: identification and mechanisms of action in human subcutaneous resistance arteries. Circulation, 103(12): 1702–1708. Coats, P., Johnston, F., MacDonald, J., McMurray, J.J.V., Hillier, C. (2001) Endothelium-derived hyperpolarising factor: Identification and mechansims of action in human subcutaneous resistance arteries. Circulation, 103: 1702–1708. Coetzee, W.A., Amarillo, Y., Chiu, J., Chow, A., Lau, D., McCormack, T., Moreno, H., Nadal, M.S., Ozaita, A., Pountney, D., Saganich, M., Vega-Saenz De. Miera, E., Rudy, B. (1999) Molecular diversity of K⫹ channels. Ann. N.Y. Acad. Sci., 868: 233–285. Cohen, R.A., Vanhoutte, P.M. (1995) Endothelium-dependent hyperpolarization. Beyond nitric oxide and cyclic GMP. Circulation, 92: 3337–3349. Cohen, R.A., Plane, F., Najibi, S., Huk, I., Malinski, T., Garland, C.J. (1997) Nitric oxide is the mediator of both endothelium-dependent relaxation and hyperpolarization of the rabbit carotid artery. Proc. Nat. Acad. Sci. USA, 94(8): 4193–4198. Coleman, H.A., Tare, M., Parkington, H.C. (2001a) K⫹ currents underlying the action of endotheliumderived hyperpolarizing factor in guinea-pig, rat and human blood vessels. J. Physiol., 531: 359–373. Coleman, H.A., Tare, M., Parkington, H.C. (2001b) EDHF is not K⫹ but may be due to spread of current from the endothelium in guinea pig arterioles. Am. J. Physiol., 280: H2478–H2483. Coleman, H.A., Tare, M., Parkington, H.C. (2001c) Components of the potassium currents underlying the actions of endothelium-derived hyperpolarizing factor in arterioles. In EDHF 2000, edited by P.M. Vanhoutte, pp. 259–269. London and New York: Taylor & Francis. Coleman, H.A., Tare, M., Parkington, H.C. (2002) Myoendothelial electrical coupling in arteries and arterioles and its implications for EDHF. Clin. Exp. Pharmacol. Physiol., 29: 630–637. Collier, M.L., Thomas, A.P., Berlin, J.R. (1999) Relationship between L-type Ca2⫹ current and unitary sarcoplasmic reticulum Ca2⫹ release events in rat ventricular myocytes. J. Physiol., 516: 117–128. Collier, M.L., Ji, G., Wang, Y., Kotlikoff, M.I. (2000) Calcium-induced calcium release in smooth muscle: loose coupling between the action potential and calcium release. J. Gen. Physiol., 115: 653–662. Cook, N.S., Quast, U. (1990) Potassium Channel Pharmacology. In Potassium channels: structure classification, function and therapeutic potential, edited by, N.S., Cook, pp. 181–325. New York: Halsted Press. Cooper, D.M., Schell, M.J., Thorn, P., Irvine, R.F. (1998) Regulation of adenylyl cyclase by membrane potential. J. Biol. Chem., 273: 27703–27707.

References

383

Cordellini, S. (1999) Endothelium dysfunction in DOCA-salt hypertension: possible involvement of prostaglandin endoperoxides. Gen. Pharmacol., 32: 315–320. Corey, E.J., Niwa, H., Falck, J.R. (1979) Selective epoxidation of eicosa-cis-5,8,11,14-tetraenoic (arachidonic) acid and eicosa-cis-8,11,14-trienoic acid. J. Am. Chem. Soc., 101: 1586–1587. Corriu, C., Félétou, M., Canet, E., Vanhoutte, P.M. (1996a) Endothelium-derived factors and hyperpolarization of the carotid artery of the guinea-pig. Br. J. Pharmacol., 119: 959–964. Corriu, C., Félétou, M., Canet, E., Vanhoutte, P.M. (1996b) Inhibitors of the cytochrome P450-monooxygenase and endothelium-dependent hyperpolarizations in the guinea-pig isolated carotid artery. Br. J. Pharmacol., 117: 607–610. Corriu, C., Félétou, M., Puybasset, L., Bea, M.L., Berdeaux, A., Vanhoutte, P.M. (1998) Endotheliumdependent hyperpolarization in isolated arteries taken from animals treated with NO-synthase inhibitors. J. Cardiovasc. Pharmacol., 32: 944–950. Cosentino, F., Patton, S. (1998) Tetrahydrobiopterin alters superoxide and nitric oxide release in prehypertensive rats. J. Clin. Invest., 101(7): 1530–1537. Cosentino, F., Barker, J.E., Brand, M.P., Heales, S.J., Werner, E.R., Tippins, J.R., West, N., Channon, K.M., Volpe, M., Luscher, T.F. (2001) Reactive oxygen species mediate endothelium-dependent relaxations in tetrahydrobiopterin-deficient mice. Arterioscler. Thromb. Vasc. Biol., 21: 496–502. Cotgreave, I.A. (1997) N-acetylcysteine: pharmacological considerations and experimental and clinical applications. Adv. Pharmacol., 38: 205–227. Cowan, C.L., Cohen, R.A. (1991) Two mechanisms mediate relaxation by bradykinin of pig coronary artery: NO-dependent and -independent responses. Am. J. Physiol., 261: H830 –H835. Cox, D.H., Aldrich, R.W. (2000) Role of the beta1 subunit in large-conductance Ca2⫹-activated K⫹ channel gating energetics. Mechanisms of enhanced Ca2⫹ sensitivity. J. Gen. Physiol., 116: 411–432. Cox, R.H., Folander, K., Swanson, R. (2001) Differential expression of voltage-gated K⫹ channel genes in arteries from spontaneously hypertensive and Wistar Kyoto rats. Hypertension, 37: 1315–1322. Crane, G.J., Hines, M.L., Neild, T.O. (2001) Simulating the spread of membrane potential changes in arteriolar networks. Microcirculation, 8: 33–43. Creager, M.A., Cooke, J.P., Mendelson, M.E., Gallagher, S.J., Coleman, S.M., Loscalzo, J., Dzau, V.J. (1990) Impaired vasodilation of forearm resistance vessels in hypercholesterolemic humans. J. Clin. Invest., 86: 228–234. Cui, Z.J., Kanno, T. (1997) Photodynamic triggering of calcium oscillation in the isolated rat pancreatic acini. J. Physiol., 504: 47–55. Curcio, F., Ceriello, A. (1992) Decreased cultured endothelial cell proliferation in high glucose medium is reversed by antioxidants: new insights on the pathophysiological mechanisms of diabetic vascular complications. In Vitro Cell Dev. Biol., 28A: 787–790. Curtis, T.M., Scholfield, C.N. (2001) Evidence for two endothelin EtA receptor subtypes in rabbit arteriolar smooth muscle. Br. J. Pharmacol., 134: 1787–1795. Dahl, G., Nonner, W., Werner, R. (1994) Attempts to define functional domains of gap junction proteins with synthetic peptides. Biophys. J., 67: 1816–1822. Daniel, E.E. (2002) The smooth muscle cell membrane and peripheral sr: their functions are organized and may be crucial. J. Hypert., 21: 367–370. Davidson, J.S., Baumgarten, I.M. (1988) Glycyrrhetinic acid derivatives: a novel class of inhibitors of gap-junctional intracellular communication. Structure-activity relationships. J. Pharmacol. Exp. Ther., 246: 1104–1107. Davis, M.J., Sharma, N.R. (1997) Calcium-release-activated calcium influx in endothelium. J. Vasc. Res., 34: 186–195. Davson, H. (1980) The aqueous humour and the intraocular pressure. In Physiology of the Eye, edited by H. Davson, p. 26. London: Churchill Livingstone. Declaration of Helsinki. (1997) Cardiovasc. Res., 35: 2–3. De Mey, J.G., Vanhoutte, P.M. (1980) Interaction between Na⫹,K⫹ exchanges and the direct inhibitory effect of acetylcholine on canine femoral arteries. Circ. Res., 46: 826–836.

384

References

De Mey, J.G., Claeys, M., Vanhoutte, P.M. (1982) Endothelium-dependent inhibitory effects of acetylcholine, adenosine triphosphate, thrombin and arachidonic acid in the canine femoral artery. J. Pharmacol. Exp. Ther., 222: 166–173. De Vriese, A.S., Van de Voorde, J., Blom, H.J., Vanhoutte, P.M., Verbeke, M., Lameire, N.H. (2000a) The impaired renal vasodilator response attributed to endothelium-derived hyperpolarizing factor in streptozotocin – induced diabetic rats is restored by 5-methyltetrahydrofolate. Diabetologia, 43: 1116–1125. De Vriese, A.S., Verbeuren, T.J., Van de Voorde, J., Lameire, N.H., Vanhoutte, P.M. (2000b) Endothelial dysfunction in diabetes. Br. J. Pharmacol., 130: 963–974. De Vriese, A.S., Van de Voorde, J., Lameire, N.H. (2002) Effects of connexin-mimetic peptides on nitric oxide synthase- and cyclooxygenase-independent renal vasodilation. Kidney Int., 61: 177–185. de Wit, C., Schäfer, C., von Bismarck, P., Bolz, S.S., Pohl, U. (1997) Elevation of plasma viscosity induces sustained NO-mediated dilation in the hamster cremaster microcirculation in vivo. Pflügers Arch., 434: 354–361. de Wit, C., Esser, N., Lehr, H., Bolz, S., Pohl, U. (1999) Pentobarbital-sensitive EDHF comediates AChinduced arteriolar dilation in the hamster microcirculation. Am. J. Physiol., 276: H1527–H1534. de Wit, C., Roos, F., Bolz, S.S., Kirchhoff, S., Krüger, O., Willecke, K., Pohl, U. (2000) Impaired conduction of vasodilation along arterioles in connexin 40-deficient mice. Circ. Res., 86: 649–655. Devor, D.C., Singh, A.K., Frizzell, R.A., Bridges, R.J. (1996) Modulation of C1⫺ secretion by benzimidazolones. I. Direct activation of Ca(2⫹)-dependent K⫹ channel. Am. J. Physiol., 271: L775–L784. Dietrich, H.H., Kajita, Y., Dacey, R.G. (1996) Local and conducted vasomotor responses in isolated rat cerebral arterioles. Am. J. Physiol., 271: H1109–H1116. Ding, H., Kubes, P., Triggle, C.R. (2000) Potassium- and acetylcholine-induced vasorelaxation in mice lacking endothelial nitric oxide synthase. Br. J. Pharmacol., 129: 1194–1200. Ding, H., Triggle, C.R. (2000) Novel endothelium-derived relaxing factors: Identification of factors and cellular targets. J. Pharmacol and Toxicol. Methods, 44: 441–452. Dittrich, M., Daut, J. (1999) Voltage-dependent K⫹ current in capillary endothelial cells isolated from guinea pig heart. Am. J. Physiol., 277: H119–H127. Doné, S.C., Leibiger, I.B., Efendiev, R., Katz, A., Leibiger, B., Berggren, P.-O., Pedemonte, C., Bertorello, A. (2002) Tyrosine 537 within the Na⫹,K⫹-ATPase ␣-subunit is essential for AP-2 binding and clathrin-dependent endocytosis. J. Biol. Chem., 277: 17108–17111. Dong, H., Waldron, G.J., Galipeau, D., Cole, W.C., Triggle, C.R. (1997) NO/PGI2-independent vasorelaxation and the cytochrome P450 pathway in rabbit carotid artery. Br. J. Pharmacol., 120(4): 695–701. Dong, H., Waldron, G.J., Cole, W.C., Triggle, C.R. (1998) Roles of calcium-activated and voltagegated delayed rectifier potassium channels in endothelium-dependent vasorelaxation of the rabbit middle cerebral artery. Br. J. Pharmcol., 123: 821–832 Dong, H., Jiang, Y., Cole, W.C., Triggle, C.R. (2000) Comparison of the pharmacological properties of EDHF-mediated vasorelaxation in guinea-pig cerebral and mesenteric resistance vessels. Br. J. Pharmacol., 130: 1983–1991. Dora, K.A., Garland, C.J. (2001) Properties of smooth muscle hyperpolarization and relaxation to K⫹ in the rat isolated mesenteric artery. Am. J. Physiol., 280: H2424–H2429. Dora, K.A., Doyle, M.P., Duling, B.R. (1997) Elevation of intracellular calcium in smooth muscle causes endothelial cell generation of NO in arterioles. Proc. Natl. Acad. Sci. USA., 94: 6529–6534. Dora, K.A., Martin, P.E.M., Chaytor, A.T., Evans, H., Garland, C.J., Griffith, T.M. (1999) Role of heterocellular gap junctional communication in endothelium-dependent smooth muscle hyperpolarization: inhibition by a connexin-mimetic peptide. Biochem. Biophys. Res. Comm., 254: 27–31. Dora, K.A., Damon, D.N., Duling, B.R. (2000) Microvascular dilation in response to occlusion: a coordinating role for conducted vasomotor responses. Am. J. Physiol. Heart Circ. Physiol., 279: H279–H284. Dora, K.A., Ings, N.T., Garland, C.J. (2002) KCa channel blockers reveal hyperpolarization and relaxation to K⫹ in the rat isolated mesenteric artery. Am J Physiol. Heart. Circ. Physiol., 283: H606–H614.

References

385

Doughty, J.M., Plane, F., Langton, P.D. (1999) Charybdotoxin and apamin block EDHF in rat mesenteric artery if selectively applied to the endothelium. Am. J. Physiol., 276: H1107–H1112. Doughty, J.M., Boyle, J.P., Langton, P.D. (2000) Potassium does not mimic EDHF in rat mesenteric arteries. Br. J. Pharmacol., 130: 1174–1182. Doyle, M.P., Duling, B.R. (1997) Acetylcholine induces conducted vasodilation by nitric oxidedependent and -independent mechanisms. Am. J. Physiol., 272: H1364–H1371. Drummond, G.R., Selemidis, S., Cocks, T.M. (2000) Apamin-sensitive, non-nitric oxide (NO) endothelium-dependent relaxations to bradykinin in the bovine isolated coronary artery: no role for cytochrome P450 and K⫹. Br. J. Pharmacol., 129: 811–819. Dube, S., Canty, J.M.J. (2001) Shear stress-induced vasodilation in porcine coronary conduit arteries is independent of nitric oxide release. Am. J. Physiol. Heart Circ. Physiol., 280: H2581–H2590. Dudgeon, S., Benson, D.P., MacKenzie, A., Paisley-Zynzkiewicz, K., Martin, W. (1998) Recovery by ascorbate of impaired NO-dependent relaxation resulting from oxidant stress in rat aorta. Br. J. Pharmacol., 125: 782–786. Duling, B.R., Berne, R.M. (1970) Propagated vasodilation in the microcirculation of the hamster cheek pouch. Circ. Res., 26: 163–170. Dvornik, E., Simard-Duquesne, N., Krami, M., Sestanj, K., Gabbay, K.H., Kinoshita, J.H., Varma, S.D., Merola, L.O. (1973) Polyol accumulation in galactosemic and diabetic rats: control by an aldose reductase inhibitor. Science, 182: 1146–1148. Ebihara, S., Shirato, K., Harata, N., Akaike, N. (1995) Gramicidin-perforated patch recording: GABA response in mammalian neurones with intact intracellular chloride. J. Physiol., 484: 77–86. Eckman, D.M., Hopkins, N., McBride, C., Keef, K.D. (1998) Endothelium-dependent relaxation and hyperpolarization in guinea-pig coronary artery: role of epoxyeicosatrienoic acid. Br. J. Pharmacol., 124: 181–189. Edwards, G., Weston, A.H. (1998) Endothelium-derived hyperpolarizing factor – a critical appraisal. Prog. Drug. Res., 50: 107–133. Edwards, G., Weston, A.H. (2001) EDHF – are there gaps in the pathway? J. Physiol., 531: 299. Edwards, G., Dora, K.A., Gardener, M.J., Garland, C.J., Weston, A.H. (1998) K⫹ is an endotheliumderived hyperpolarizing factor in rat arteries. Nature, 396: 269–272. Edwards, G., Félétou, M., Gardener, M.J., Thollon, C., Vanhoutte, P.M., Weston, A.H. (1999a) Role of gap junctions in the responses to EDHF in rat and guinea-pig small arteries. Br. J. Pharmacol., 128: 1788–1794. Edwards, G., Gardener, M.J., Félétou, M., Brady, G., Vanhoutte, P.M., Weston, A.H. (1999b) Further investigation of endothelium-derived hyperpolarizing factor (EDHF) in rat hepatic artery: studies using 1-EBIO and ouabain. Br. J. Pharmacol., 128: 1064–1070. Edwards, G., Thollon, C., Gardener, M.J., Félétou, M., Vilaine, J.-P., Vanhoutte, P.M., Weston, A.H. (2000) Role of gap junctions and EETs in endothelium-dependent hyperpolarization of porcine coronary artery. Br. J. Pharmacol., 129: 1145–1154. Edwards, G., Félétou, M., Gardener, M.J., Glen, C.D., Richards, G.R., Vanhoutte, P.M., Weston, A.H. (2001) Further investigations into the endothelium-dependent hyperpolarizing effects of bradykinin and substance P in porcine coronary artery. Br. J. Pharmacol., 133: 1145–1153. Egan, T.M., Dagan, D., Levitan, I.B. (1993) Properties and modulation of a calcium-activated potassium channel in rat olfactory bulb neurons. J. Neurophysiol., 69: 1433–1442. Eizawa, H., Yui, Y., Inoue, R., Kosuga, K., Hattori, R., Aoyama, T., Sasayama, S. (1995) Lysophosphatidylcholine inhibits endothelium-dependent hyperpolarization and N omega-nitro-Larginine/indomethacin-resistant endothelium-dependent relaxation in the porcine coronary artery. Circulation, 92: 3520–3526. Elble, R.C., Ji, G., Nehrke, K., DiBiasio, J., Kingsley, P.D., Kotlikoff, M.I., Pauli, B.U. (2002) Molecular and functional characterization of a murine calcium-sensitive chloride channel expressed in smooth muscle. J. Biol. Chem., 277: 18586–18591. Elliott, S.J., Koliwad, S.K. (1997) Redox control of ion channel activity in vascular endothelial cells by glutathione. Microcirc., 4: 341–347.

386

References

Ellis, E.F., Police, R.J., Yancey, L., McKinney, J.S., Amruthesh, S.C. (1990) Dilation of cerebral arterioles by cytochrome P-450 metabolites of arachidonic acid. Am. J. Physiol., 259: H1171–H1177. Ellmann, G.L. (1959) Tissue sulfhydryl groups. Arch. Biochem. Bio-phys., 82: 70. El-Mas, M.M., Abdel-Rahman, A.A. (1999) Role of the sympathetic control of vascular resistance in ethanol-clonidine hemodynamic interaction in SHRs. J. Cardiovasc. Pharmacol., 34: 589–596. Emerson, G.G., Segal, S.S. (2000a) Electrical coupling between endothelial cells and smooth muscle cells in hamster feed arteries: role in vasomotor control. Circ. Res., 87: 474–479. Emerson, G.G., Segal, S.S. (2000b) Endothelial cell pathway for conduction of hyperpolarization and vasodilation along hamster feed artery. Circ. Res., 86: 94–100. Emerson, G.G., Segal, S.S. (2001) Electrical activation of endothelium evokes vasodilation and hyperpolarization along hamster feed arteries. Am. J. Physiol. Heart Circ. Physiol., 280: H160–H167. Emilsson, K., Wahlestedt, C., Sun, M.K., Nystedt, S., Owman, C., Sundelin, J. (1997) Vascular effects of proteinase-activated receptor 2 agonist peptide. J. Vasc. Res., 34: 267–272. Escalante, B., Omata, K., Sessa, W., Lee, S.G., Falck, J.R., Schwartzman, M.L. (1993) 20-Hydroxyeicosatetraenoic acid is an endothelium-dependent vasoconstrictor in rabbit arteries. Eur. J. Pharmacol., 235: 1–7. Esguerra, M., Wang, J., Foster, C.D., Adelman, J.P., North, R.A., Levitan, I.B. (1994) Cloned Ca2⫹dependent K⫹ channel modulated by a functionally associated protein kinase. Nature, 369: 563–565. Evans, W.H., Bositano, S. (2001) Connexin mimetic peptides: specific inhibitors of gap-junctional intercellular communication. Biochem. Soc. Trans., 29: 606–612. Falck, J.R., Manna, S., Viala, J., Siddhanta, A.K. (1985) Arachidonate epoxygenases: inhibitors and metabolite analogs. Tetrahedron Lett., 26: 2287–2290. Feelisch, M., Kotsonis, P., Siebe, J., Clement, B., Schmidt, H.H. (1999) The soluble guanylyl cyclase inhibitor 1H-(1,2,4)oxadiazolo-(4.3-a)quinoxalin-1 one is a nonselective heme protein inhibitor of nitric oxide synthase and cytochrome P-450 enzymes involved in nitric oxide donor bioactivation. Mol. Pharmacol., 56: 243–253. Félétou, M., Vanhoutte, P.M. (1996) Endothelium-derived hyperpolarizing factor. Clin. Exp. Pharmacol. Physiol., 23: 1082–1090. Félétou, M., Vanhoutte, P.M. (1988) Endothelium-dependent hyperpolarization of canine coronary smooth muscle. Br. J. Pharmacol., 93: 515–524. Félétou, M., Vanhoutte, P.M. (1999a) The alternative: EDHF. J. Mol. Cell Cardiol., 31: 15–22. Félétou, M., Vanhoutte, P.M. (1999b) The third pathway: endothelium-dependent hyperpolarization. J. Physiol. Pharmacol., 50: 525–534. Félétou, M., Vanhoutte, P.M. (2000) Endothelium-dependent, hyperpolarization of vascular smooth muscle cells. Acta Pharmacol. Sin., 21: 1–18. Fenn, C.G., Littleton, J.M. (1982) Inhibition of platelet aggregation by ethanol in vitro shows specificity for aggregating agent used and is influenced by platelet lipid composition. Thromb. Haemost., 48: 49–53. Ferguson, M.K. (1992) Modulation of lymphatic smooth muscle contractile responses by the endothelium. J. Surg. Res., 52: 359–363. Ferrer, M., Marin, J., Encabo, A., Alonso, M.J., Balfagon, G. (1999) Role of K⫹ channels and sodium pump in the vasodilation induced by acetylcholine, nitric oxide, and cyclic GMP in the rabbit aorta. Gen. Pharmacol., 33: 35–41. Fichtner, H., Frobe, U., Busse, R., Kohlhardt, M. (1987) Single non-selective cation channels and Ca2⫹-activated K⫹ channels in aortic endothelial cells. J. Membr. Biol., 98: 125–133. Fine, B.P., Hansen, K.A., Salcedo, J.R., Aviv, A. (1989) Calcium-activated potassium channels in human platelets. Proc. Soc. Exp. Biol. Med., 192: 109–113. Fisslthaler, B., Popp, R., Kiss, L., Potente, M., Harder, D.R., Fleming, I., Busse, R. (1999) Cytochrome P450 2C (CYP2C) is an EDHF synthase in coronary arteries. Nature, 401: 493–497. Fisslthaler, B., Hinsch, N., Chataigneau, T., Popp, R., Kiss, L., Busse, R., Fleming, I. (2000) Nifedipine increases cytochrome P4502C expression and endothelium-derived hyperpolarizing factor-mediated responses in coronary arteries. Hypertension, 36: 270–275.

References

387

Fisslthaler, B., Popp, R., Michaelis, U.R., Kiss, L., Fleming, I., Busse, R. (2001) Cyclic stretch enhances the expression and activity of coronary endothelium-derived hyperpolarizing factor synthase. Hypertension, 38: 1427–1432. Fisslthaler, B., Michaelis, U.R., Randriamboavonjy, V., Busse, R., Fleming, I. (2003) Cytochrome P450 epoxygenases and vascular tone: novel role for HMG Co A reductase inhibitors in the regulation of CYP 2C expression. Biochim. Biophys. Acta., 1619(3): 332–339. Fitzpatrick, D.F., Hirschfield, S.L., Coffey, R.G. (1993) Endothelium-dependent vasorelaxing activity of wine and other grape products. Am. J. Physiol., 265: H774–H778. Fitzpatrick, F.A., Ennis, M.D., Baze, M.E., Wynalda, M.A., McGee, J.E., Liggett, W.F. (1986) Inhibition of cyclooxygenase activity and platelet aggregation by epoxyeicosatrienoic acids. Influence of stereochemistry. J. Biol. Chem., 261: 15334–15338. Fleming, I. (2001) Cytochrome p450 and vascular homeostasis. Circ. Res., 89: 753–762. Fleming, I., Bauersachs, J. (1997) Calcium-dependent and calcium-independent activation of the endothelial NO synthase. J. Vasc. Res., 34(3): 165–174. Fleming, I., Fisslthaler, B., Michaelis, U.R., Kiss, L., Popp, R., Busse, R. (2001a) The coronary endothelium-derived hyperpolarizing factor (EDHF) stimulates multiple signaling pathways and proliferation in vascular cells. Pflugers Arch., 442: 511–518. Fleming, I., Michaelis, U.R., Bredenkotter, D., Fisslthaler, B., Dehghani, F., Brande, R.P., Busse, R. (2001b) Endothelium-derived hyperpolarizing factor synthase (Cytochrome P450 2C9) is a functionally significant source of reactive oxygen species in coronary arteries. Circ. Res., 88: 44–51. Florey, H.W. (1927) Observations on the contractility of lacteals. Part I. J. Physiol., 62: 267–272. Fontana, L., McNeill, K.L., Ritter, J.M., Chowienczyk, P.J. (1999) Effects of vitamin C and of a cell permeable superoxide dismutase mimetic on acute lipoprotein induced endothelial dysfunction in rabbit aortic rings. Br. J. Pharmacol., 126: 730–734. Forbush, B., Kaplan, J.H., Hoffman, J.F. (1978) Characterization of a new photoaffinity derivative of ouabain: labeling of the large polypeptide and of a proteolipid component of the Na,K-ATPase. Biochemistry, 17: 3667–3676. Frangos, S.G., Knox, R., Yano, Y., Chen, E., Luozzo, G., Chen, A.H., Sumpio, B.E. (2001) The integrinmediated cyclic strain-induced signaling pathway in vascular endothelial cells. Endothelium, 8: 1–10. Freedman, J.C., Novak, T.S. (1989) Optical measurement of membrane potential in cells, organelles, and vesicles. Methods Enzymol, 172: 102–122. Freise, J., Schmidt, F.W. et al. (1985) ‘Gabexate mesilate and camostate: new inhibitors of phospholipase A2 and their influence on the alpha-amylase activity in serum of patients with acute pancreatitis. Clin. Biochem., 18(4): 224–229. Fribel, M., Klotz, K.F., Ley, K., Gaehtgens, P., Pries, A.R. (1995) Flow-dependent regulation of arteriolar diameter in rat skeletal muscle in situ: role of endothelium-derived relaxing factor and prostanoid. J. Physiol., 483: 715–726. Frieden, M., Sollini, M., Bény, J.-L. (1999) Substance P and bradykinin activate different types of KCa channels to hyperpolarize porcine coronary artery endothelial cells. J. Physiol., 519: 361–371. Friedhoff, L.T., Sonenberg, M. (1983) The membrane potential of human platelets. Blood, 61: 180–185. Friedhoff, L.T., Kim, E., Priddle, M., Sonenberg, M. (1981) The effect of altered transmembrane ion gradients on membrane potential and aggregation of human platelets in blood plasma. Biochem. Biophys. Res. Commun., 102: 832–837. Fromy, B., Abraham, P., Saumet, J.L. (1998) Non-nociceptive capsaicin-sensitive nerve terminal stimulation allows for an original vasodilatory reflex in the human skin. Brain Res., 811: 166–168. Fromy, B., Abraham, P., Saumet, J.L. (2000a) Progressive calibrated pressure device to measure cutaneous blood flow changes to external pressure strain. Brain Res. Protocols, 5: 198–203. Fromy, B., Merzeau, S., Abraham, P., Saumet, J.L. (2000b) Mechanisms of the cutaneous vasodilator response to local external pressure application in rats: Involvement of CGR, neurokinins, prostaglandins and NO. Br. J. Pharmacol., 131: 1161–1171.

388

References

Fry, T., Evans, J.H., Sanderson, M.J. (2001) Propagation of intercellular calcium waves in C6 glioma cells transfected with connexins 43 or 32. Microsc. Res. Tech., 52: 289–300. Fu, X., Kamps, M.P. (1997) E2a-Pbx1 induces aberrant expression of tissue-specific and developmentally regulated genes when expressed in NIH 3T3 fibroblasts. Mol. Cell. Biol., 17: 1503–1512. Fujii, K., Tominaga, M., Ohmori, S., Kobayashi, K., Koga, T., Takata, Y., Fujishama, M. (1992) Decreased endothelium-dependent hyperpolarization to acetylcholine in smooth muscle of the mesenteric artery of spontaneously hypertensive rats. Circ. Res., 70: 660–669. Fujii, K., Ohmori, S., Tominaga, M., Abe, I., Takata, Y., Ohya, Y., Kobayashi, K., Fujishama, M. (1993) Age-related changes in endothelium-dependent hyperpolarization in the rat mesenteric artery. Am. J. Physiol., 265: H509–H516. Fujii, K., Goto, K., Abe, I. (2001) Inhibition of converting enzyme prevents the age-related decline in endothelium-dependent hyperpolarization. In EDHF 2000, edited by P.M. Vanhoutte, pp. 410–416. London and New York: Taylor & Francis. Fukao, M., Hattori, Y., Kanno, M., Sakuma, I., Kitabatake, A. (1995) Thapsigargin- and cyclopiazonic acid-induced endothelium-dependent hyperpolarization in rat mesenteric artery. Br. J. Pharmacol., 115: 987–992. Fukao, M., Hattori, Y., Kanno, M., Sakuma, I., Kitabatake, A. (1997a) Sources of Ca2⫹ in relation to generation of acetylcholine-induced endothelium-dependent hyperpolarization in rat mesenteric artery. Br. J. Pharmacol., 120: 1328–1334. Fukao, M., Hattori, Y., Kanno, M., Sakuma, I., Kitabatake, A. (1997b) Evidence against a role of cytochrome P450-derived arachidonic acid metabolites in endothelium-dependent hyperpolarization by acetylcholine in rat isolated mesenteric artery. Br. J. Pharmacol., 120: 439–446. Fukao, M., Hattori, Y., Kanno, M., Sakuma, I., Kitabatake, A. (1997c) Alterations in endotheliumdependent hyperpolarization and relaxation in mesenteric arteries from streptozotocin-induced diabetic rats. Br. J. Pharmacol., 121: 1383–1391. Fukao, M., Mason, H.S., Britton, F.C., Kenyon, J.L., Horowitz, B., Keef, K.D. (1999) Cyclic GMPdependent protein kinase activates cloned BKCa channels expressed in mammalian cells by direct phosphorylation at serine 1072. J. Biol. Chem., 274: 10927–10935. Fuller, C.M., Benos, D.J. (2000) Electrophysiological characteristics of the Ca2⫹-activated Cl⫺ channel family of anion transport proteins. Clin. Exp. Pharm. Physiol., 27: 906–910. Fuller, C.M., Ji, H.-L., Tousson, A., Elble, R.C., Pauli, B.U., Benos, D.J. (2001) Ca2⫹-activated Cl⫺ channels: a newly emerging anion transport family. Pflügers Archiv, 443: S107–S110. Fulton, D., McGiff, J.C., Quilley, J. (1992) Contribution of NO and cytochrome P450 to the vasodilator effect of bradykinin in the rat kidney. Br. J. Pharmacol., 107: 722–725. Fulton, D., Mahboubi, K., McGiff, J.C., Quilley, J. (1995) Cytochrome P450-dependent effects of bradykinin in the rat heart. Br. J. Pharmacol., 114: 99–102. Fulton, D., McGiff, J.C., Quilley, J. (1998) Pharmacological evaluation of an epoxide as the putative hyperpolarizing factor mediating the nitric oxide-independent vasodilator effect of bradykinin in the rat heart. J. Pharmacol. Exp. Ther., 287: 497–503. Furchgott, R.F. (1988) Studies on relaxation of rabbit aorta by sodium nitrite: the basis for the proposal that the acid-activatable inhibitory factor from bovine retractor penis is inorganic nitrite and the endothelium-derived relaxing factor is nitric oxide. In Vasodilatation; Vascular smooth muscle, peptides, autonomic nerves and endothelium, edited by P.M. Vanhoutte, pp. 401–414, New York: Raven Press. Furchgott, R.F., Zawadzki, J.V. (1980) The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature, 288: 373–376. Furchgott, R.F., Vanhoutte, P.M. (1989) Endothelium-derived relaxing and contracting factors. Faseb. J., 3: 2007–2018. Galvez, A., Gimenez-Gallego, G., Rueben, J.P., Roy-Contancin, L., Feigenbaum P., Kaczorowski, G.J., Garcia, M.L. (1990) Purification and characterization of a unique, potent, peptidyl probe for the high conductance calcium-activated potassium channel from venom of the scorpion Buthus tamulus. J. Biol. Chem., 265: 11083–11090.

References

389

Gambone, L.M., Murray, P.A., Flavahan, N.A. (1997) Synergistic interaction between endotheliumderived NO and prostacyclin in pulmonary artery: potential role for K⫹ATP channels. Br. J. Pharmacol., 121: 271–279. Gandhi, R., Elble, R.C., Gruber, A.D., Schreur, K.D., Ji, H.L., Fuller, C.M., Pauli, B.U. (1998) Molecular and functional characterization of a calcium-sensitive chloride channel from mouse lung. J. Biol. Chem., 273: 32096–32101. Gao, J., Zhao, J., Rayner, S.E., Van Helden, D.F. (1999) Evidence that the ATP-induced increase in vasomotion of guinea-pig mesenteric lymphatics involves an endothelium-dependent release of thromboxane A2. Br. J. Pharmacol., 127: 1597–1602. Garcia-Cohen, E.C., Diez-Picazo, J.M.L.D., Baena, A.B., Salaices, M., Rodriguez-Martinez, M.A. (2000) Oxidative stress induced by tert-butyl hydroperoxide causes vasoconstriction in the aorta from hypertensive and aged rats: role of cyclooxygenase-2 isoform. J. Pharmacol. Exp. Ther., 293: 75–81. Gardiner, S.M., Compton, A.M., Bennett, T., Palmer, R.M.J., Moncada, S. (1990) Regional haemodynamic changes during oral ingestion of NG-monomethyl-L-arginine or NG-nitro-L-arginine methyl ester in conscious Brattleboro rats. Br. J. Pharmacol., 101: 10–12. Gardiner, S.M., Kemp, P.A., Bennett, T. (1991) Effect of NG-nitro-L-arginine methyl ester on vasodilator responses to acetylcholine, 5⬘-N-ethylcarboxamidoadenosine or salbutamol in conscious rats. Br. J. Pharmacol., 103: 1725–1732. Gardner, P.R., Nguyen, D.D., White, C.W. (1996) Superoxide scavenging by Mn(II/III) tetrakis (1-methyl-4-pyridyl) porphyrin in mammalian cells. Arch. Biochem. Biophys., 325: 20–28. Garland, C.J., McPherson, G.A. (1992) Evidence that nitric oxide does not mediate the hyperpolarization and relaxation to acetylcholine in the rat small mesenteric artery. Br. J. Pharmacol., 105: 429–435. Garland, C.J., Plane, F., Kemp, B.K., Cocks, T.M. (1995) Endothelium-dependent hyperpolarization: a role in the control of vascular tone. Trends Pharmacol. Sci., 16: 23–30. Gauthier-Rein, K.M., Rusch, N.J. (1998) Distinct endothelial impairment in coronary microvessels from hypertensive dahl rats. Hypertension, 31 [part2]: 328–334. Gauthier, K.G., Rusch, N.J. (2001) Rat coronary endothelial cell membrane potential responses during hypertension. Hypertension, 37: 66–71. Gauthier, K.G., Liu, C., Popovic, A., Albarwani, S., Rusch, N.J. (2002). Freshly isolated coronary endothelial cells do not express the BKCa channel gene. J. Physiol., 545: 829–836. Gauthier, K.M., Deeter, C., Krishna, U.M., Reddy, Y.K., Bondlela, M., Falck, J.R., Campbell, W.B. (2002) 14,15-Epoxyeicosa-5(Z)-enoic acid: a selective epoxyeicosatrienoic acid antagonist that inhibits endothelium-dependent hyperpolarization and relaxation in coronary arteries. Circ. Res., 90: 1028–1036. Ge, Z.-D., Zhang, X.-H., Peter, C.-W., Fung, He, G.-W. (2000) Endothelium-dependent hyperpolarization and relaxation resistance to NG-nitro–arginine and indomethacin in coronary circulation. Cardiovasc. Res., 46: 547–556. Geary, G.G., Krause, D.N., Duckles, S.P. (1998) Estrogen reduces myogenic tone through a nitric oxide-dependent mechanism in rat cerebral arteries. Am. J. Physiol., 275: H292–H300. Gebremedhin, D., Ma, Y.-H., Falck, J.R., Roman, R.J., VanRollins, M., Harder, D.R. (1992) Mechanism of action of cerebral epoxyeicosatrienoic acids on cerebral arterial smooth muscle. Am. J. Physiol., 263: H519–H525. Gebremedhin, D., Harder, D.R, Pratt, P.F., Campbell, W.B. (1998) Bioassay of an endothelium derived hyperpolarizing factor from bovine coronary arteries: role of a cytochrome P450 metabolite. J. Vasc. Res., 35: 274–284. Gebremedhin, D., Lange, A.R., Lowry, T.F., Taheri, M.R., Birks, E.K., Hudetz, A.G., Narayanan, J., Falck, J.R., Hirotsugu, O., Roman, R.J., Nithipatikom, K., Campbell, W.B., Harder, D.R. (2000) Production of 20-HETE and its role in autoregulation of cerebral blood flow. Circ. Res., 87: 60–65. Ghisdal P., Morel, N. (2001) Cellular target of voltage and calcium-dependent K⫹ channel blockers involved in EDHF-mediated responses in rat superior mesenteric artery. Br. J. Pharmacol., 134: 1021–1028.

390

References

Ghisdal, P., Godfraind, T., Morel, N. (1999) Effect of nitro-L-arginine on electrical and mechanical responses to acetylcholine in the superior mesenteric artery from stroke-prone hypertensive rat. Br. J. Pharmacol., 128: 1513–1523. Gilligan, D.M., Quyyumi, A.A., Cannon, R.O. 3rd (1994) Effects of physiological levels of estrogen on coronary vasomotor function in postmenopausal women. Circulation, 89: 2545–2551. Gladwell, S.J., Jefferys, J.G. (2001) Second messenger modulation of electrotonic coupling between region CA3 pyramidal cell axons in the rat hippocampus. Neurosci. Lett.. 300: 1–4. Golden, P.F., Jane, J.A. (1970) Survival in profound hypovolemia: differentiation of the role of the heart, lung and the brain. Ann. Roy. Coll. Surg., 3: 63–64. Golding, E.M., You, J., Robertson, C.S., Bryan, R.M. Jr. (2001) Potentiated endothelium-derived hyperpolarizing factor-mediated dilations in cerebral arteries following mild head injury. J. Neurotrauma., 18: 691–697. Gollasch, M., Wellman, G.C., Knot, H.J., Jaggar, J.H., Damon, D.H., Bonev, A.D., Nelson, M.T. (1998) Ontogeny of local sarcoplasmic reticulum Ca2⫹ signals in cerebral arteries: Ca2⫹ sparks as elementary physiological events. Circ. Res., 83: 1104–1114. Goto, K., Fujii, K., Onaka, U., Abe, I., Fujishima, M. (2000a) Renin-angiotensin system blockade improves endothelial dysfunction in hypertension. Hypertension, 36: 575–580. Goto, K., Fujii, K., Onaka, U., Abe, I., Fujishima, M. (2000b) Angiotensin-converting enzyme inhibitor prevents age-related endothelial dysfunction. Hypertension, 36: 581–587. Goto, K., Fujii, K., Kansui, Y., Abe, I., Iida, M. (2002) Critical role of gap junctions in endotheliumdependent hyperpolarization in rat mesenteric arteries. Clin. Exp. Pharmacol. Physiol., 29: 595–602. Graier, W.F., Kukovetz, W.R., Groschner, K. (1993) Cyclic AMP enhances agonist-induced Ca2⫹ entry into endothelial cells by activation of potassium channels and membrane hyperpolarization. Biochem. J., 291: 263–267. Graier, W.F., Hoebel, B.G., Paltauf-Doburzynska, J., Kostner, G.M. (1998) Effects of superoxide anions on endothelial Ca2⫹ signaling pathways. Arterioscler. Thromb. Vasc. Biol., 18: 1470–1479. Grazul-Bilska, A.T., Reynolds, L.P., Bilski, J.J., Redmer, D.A. (2001) Effects of second messengers on gap junctional intercellular communication of ovine luteal cells throughout the estrous cycle. Biol. Reprod., 65: 777–783. Green, L.C., Wagner, D.A., Glogowski, J., Skipper, P.L., Wishnok, J.C., Tannenbaum, S.R. (1982) Analysis of nitrate, nitrite and [15N] nitrate in biological fluids. Anal. Biochem., 126: 131–138. Greenwood, I., Ledoux, J., Leblanc, N. (2001) Differential regulation of Ca2⫹-activated Cl⫺ currents in rabbit arterial and portal vein smooth muscle cells by Ca2⫹-calmodulin-dependent kinase. J. Physiol., 534: 395–408. Greenwood, I.A., Miller, L.J., Ohya, S., Horowitz, B. (2002) The large conductance potassium channel ␤ subunit can interact with and modulate the functional properties of a calcium activated chloride channel CLCA1. J. Biol. Chem., 277: 22119–22122. Greil, W., Patscheke, H., Brossmer, R. (1972) Effect of lithium and other monovalent cations on the ADP induced platelet aggregation in human platelet rich plasma. FEBS Lett., 26: 271–273. Gribkoff, V.K., Lum-Ragan, J.T., Boissard, C.G., Post-Munson, D.J., Meanwell, N.A., Starrett, J.E., Kozlowski, E.S., Romine, J.L., Trojnacki, J.T., McKay, M.C., Zhong, J., Dworetzky, S.I. (1996) Effects of channel modulators on cloned large-conductance calcium-activated potassium channels. Mol. Pharmacol., 50: 206–217. Griffith, T.M., Taylor, H.J. (1999) Cyclic AMP mediates EDHF-type relaxations of rabbit jugular vein. Biochem. Biophys. Res. Commun., 263: 52–57. Griffith, T.M., Chaytor, A.T., Berman, R.S., Edwards, D.H. (2001) Heterogeneity of EDHF-type relaxations of rabbit and rat arteries analysed with peptides homologous to the extracellular loops of connexins 37, 40, 43. In EDHF 2000, edited by P.M. Vanhoutte, pp. 62–74. London and New York: Taylor & Francis. Griffith, T.M., Chaytor, A.T., Taylor, H.J., Giddings, B.D., Edwards, D.H. (2002) cAMP facilitates EDHF-type relaxations in conduit arteries by enhancing electrotonic conduction via gap junctions. Proc. Natl. Acad. Sci. USA, 99: 6392–6397.

References

391

Grishin, A.V., Reinhard, J., Dunbar, L.A., Courtois-Coutry, N., Wang, T., Giebisch, G., Caplan, M.J. (1999) Nongastric H⫹,K⫹-ATPase: cell biologic and functional properties. Sem. Nephro., 19: 421–430. Groschner, K., Graier, W.F., Kukovetz, W.R. (1994) Histamine induces K⫹, Ca2⫹, and Cl⫺ currents in human vascular endothelial cells. Circ. Res., 75: 304–314. Gruber, A.D., Schreur, K.D., Ji, H.L., Fuller, C.M., Pauli, B.U. (1999), Molecular cloning and transmembrane structure of hCLCA2 from human lung, trachea, and mammary gland. Am. J. Physiol., 276: C1261–C1270. Gruchow, H.W., Anderson, A.J., Barboriak, J.J., Sobocinski, K.A. (1988) Postmenopausal use of estrogen and occlusion of coronary arteries. Am. Heart J., 115: 954–963. Grunnet, M., Jensen, B.S., Olesen, S-P., Klaerke, D.A. (2001a) Apamin interacts with all subtypes of cloned small conductance Ca2⫹-activated K⫹ channels. Pflügers Atrch. Eur. J. Physiol., 441: 544–550. Grunnet, M., Jespersen, T., Angelo, K., Frøkjær-Jensen, C., Klaerke, D.A., Olesen, S-P., Jensen, B.S. (2001b) Pharmacological modulation of SK3 channels. Neuropharmacology, 40: 879–887. Grunt, M., Glaser, J., Schmidhuber, H., Pauschinger, B., Born, J., (1993) Effects of corticotrop inreleasing factor on isolated rat heart activity. Am. J. Physiol., 264: H1124–H1129. Guibert, C., Beech, D.J. (1999) Positive and negative coupling of the ETA receptor to Ca2⫹-permeable channels in rabbit cerebral cortex arterioles. J. Physiol., 514: 843–856. Haas, M., Wang, H., Tian, J., Xie, Z. (2002) Src-mediated inter-receptor cross talk between the NaK ATPase and the EFGR relays the signal from ouabain to MAP kinases. J. Biol Chem., 277: 18694–18702. Haas, T.L., Duling, B.R. (1997) Morphology favors an endothelial cell pathway for longitudinal conduction within arterioles. Microvasc. Res., 53: 113–120. Haburcak, M., Weil, Viana, F., Prenen, J., Droogmans, G., Nilius, B. (1997) Calcium-activated potassium channels in cultured human endothelial cells are not directly modulated by nitric oxide. Cell Calcium, 21: 291–300. Haddock, R.E., Hirst, G.D.S., Hill, C.E. (2002) Voltage independence of vasomotion in isolated irideal arterioles of the rat. J. Physiol., 540: 219–229. Haefliger, J.-A., Castillo, E., Waeber, G., Bergonzelli, G.E., Aubert, J.F., Sutter, E., Nicod, P., Waeber, B., Meda, P. (1997) Hypertension increases connexin43 in a tissue-specific manner. Circulation, 95: 1007–1014. Halcox, J.P., Narayanan, S. (2001) Characterization of endothelium-derived hyperpolarizing factor in the human forearm microcirculation. Am. J. Physiol., 280(6): H2470–H2477. Halliwell, B., Gutteridge, J.M.C. (1989) In Free radicals in biology and medicine, edited by B. Halliwell, J.M.C. Gutteridge, Oxford: Clarendon Press. Hamill, O.P., Marty, A., Neher, E., Sackmann, B., Sigworth, F.J. (1981) Improved patch-clamp techniques for the high-resolution current recording from cells and cell-free membrane patches. Pflügers Archiv, 391: 85–100. Hamilton, J.R., Nguyen, P.B., Cocks, T.M. (1998) Atypical protease-activated receptor mediates endothelium-dependent relaxation of human coronary arteries. Circ. Res., 82: 1306–1311. Hammarström, A.K., Parkington, H.C., Tare, M., Coleman, H.A. (1999) Endothelium-dependent hyperpolarization in resting and depolarized mammary and coronary arteries of guinea-pigs. Br. J. Pharmacol., 126: 421–428. Harder, D.R., Gebremedhin, D., Narayanan, J., Jefcoat, C., Falck, J.R., Campbell, W.B., Roman, R. (1994) Formation and action of a P-450 4A metabolite of arachidonic acid in cat cerebral microvessels. Am. J. Physiol., 266: H2098–H2107. Harder, D.R., Campbell, W.B., Roman, R.J. (1995) Role of cytochrome P-450 enzymes and metabolites of arachidonic acid in the control of vascular tone. [Review] [114 refs] J. Vascul. Res., 32(2): 79–92. Harder, D.R., Zhang, C., Gebremedhin, D. (2002) Astrocytes function in matching blood flow to metabolic activity. News in Physiol. Sci., 17: 27–31. Harris, D., Martin, P.E.M., Evans, H.W., Kendall, D.A., Griffith, T.M., Randall, M.D. (2000) Role of gap junctions in endothelium-derived hyperpolarising factor responses and mechanisms of K⫹relaxation. Eur. J. Pharmacol., 402: 119–128.

392

References

Harris, D., McCulloch, A.I., Kendall, D.A., Randall, M.D. (2002) Characterization of vasorelaxant responses to anadamide in th erat mesenteric arterial bed. J. Physiol., 539: 893–902. Hashitani, H., Suzuki, H. (1997) K⫹ channels which contribute to the acetylcholine-induced hyperpolarization in smooth muscle of the guinea-pig submucosal arteriole. J. Physiol., 501: 319–329. Hasler, U., Wang, X., Crambert, G., Beguin, P., Jaisser, F., Horisberger, J.-D., Geering, K. (1998) Role of ␤-subunit domains in the assembly, stable expression, intracellular routing, and functional properties of Na,K-ATPase. J. Biol. Chem., 273: 30826–30835. Hasunuma, K., Yamaguchi, T., Rodman, D.M., O’Brien, R.F., McMurtry, I.F. (1991) Effects of inhibitors of EDRF and EDHF on vasoreactivity of perfused rat lungs. Am. J. Physiol., 260: L97–L104. Hayabuchi, Y., Nakaya, Y., Matsuoka, S., Kuroda, Y. (1998a) Hydrogen peroxide-induced vascular relaxation in porcine coronary arteries is mediated by Ca2⫹-activated K⫹ channels. Heart Vessels, 13: 9–17. Hayabuchi, Y., Nakaya, Y., Matsuoka, S., Kuroda, Y. (1998b) Endothelium-derived hyperpolarizing factor activates Ca2⫹-activated K⫹ channels in porcine coronary artery smooth muscle cells. J. Cardiovasc. Pharmacol., 32: 642–649. He, G.-W. (1997) Hyperkalemic exposure impairs EDHF-mediated endothelial function in the human coronary artery. Ann. Thorac. Surg., 63: 84–87. He, G.-W., Liu, Z.-G. (2001) Comparison of nitric oxide release and endothelium-derived hyperpolarizing factor-mediated hyperpolarization between human radial and internal mammary arteries. Circulation, 104 [suppl I]: I344–I349. He, G.-W., Yang, C.-Q., Graier, W.F., Yang, J.-A. (1996) Hyperkalemia alters EDHF-mediated hyperpolarization and relaxation in porcine coronary arteries. Am. J. Physiol., 271 (Heart Circ. Physipol., 40): H760–H767. Hecker, M., Bara, A.T., Bauersachs, J., Busse, R. (1994) Characterization of endothelium-derived hyperpolarizing factor as a cytochrome P450-derived arachidonic acid metabolite in mammals. J. Physiol., 481: 407–414. Heldwein, K.A., Redick, D.L., Rittenberg, M.B., Claycomb, W.C., Stenzel-Poore, M.P. (1996) Corticotropin-releasing hormone receptor expression and functional coupling in neonatal cardiac myocytes and AT-1 cells. Endocrinology, 137: 3631–3639. Helmlinger, G., Berk, B.C., Nerem, R.M. (1995) Calcium responses of endothelial cell monolayers subjected to pulsatile and steady laminar flow differ. Am. J. Physiol., 269: C367–C375. Herrera, G.M., Nelson, M.T. (2000) Differential regulation of SK and BK channels by Ca2⫹ signals from Ca2⫹ channels and ryanodine receptors in guinea pig urinary bladder myocytes. J. Physiol., 541(2): 483–492. Hibbard, B., Milner, D. (1994) Health Trends, 26: 26–28. Hill, C.E., Eade, J., Sandow, S.L. (1999) Mechanisms underlying spontaneous rhythmical contractions in irideal arterioles of the rat, J. Physiol., 521: 507–516. Hill, C.E., Hickey, H., Sandow, S.L. (2000) Role of gap junctions in acetylcholine-induced vasodilatation of proximal and distal arteries of the rat mesentery. J. Autonom. Nerv. Syst., 81: 122–127. Hill, C.E., Phillips, J.K., Sandow, S.L. (2001) Heterogeneous control of blood flow amongst different vascular beds. Med. Res. Rev., 21: 1–60. Himmel, H.M., Whorton, A.R., Strauss, H.C. (1993) Intracellular calcium, currents, and stimulusresponse coupling in endothelial cells. Hypertension, 21: 112–127. Hinton, J.M., Walker, S.D., Garland, C.J. (2000) Identification and distribution of inwardly rectifying potassium channels in rat microvascular endothelial cells. J. Physiol., 523P: 18P. Hirst, G.D.S., Edwards, F.R., Gould, D.J., Sandow, S.L., Hill, C.E. (1997) Electrical properties of iridial arterioles of the rat. Am. J. Physiol., 273: H2465–H2472. Höfer, T., Politi, A., Heinrich R. (2001) Intercellular Ca2⫹ wave propagation through gap-junctional Ca2⫹ diffusion: a theoretical study. Biophys. J., 80: 75–87. Hogg, D.S., Albarwani, S., Davies, A.R.L., Kozlowski, R.Z. (1999) Endothelial cells freshly isolated from resistance-sized pulmonary arteries possess a unique K⫹ current profile. Biochem. Biophys. Res. Comm., 263: 405–409.

References

393

Hollenberg, M.D., Compton, S.J. (2002) International Union of Pharmacology XXVIII. ProteinaseActivated Receptors. Pharmacol Rev., 54(2): 203–217. Hollenberg, M.D., Saifeddine, M., Al Ani, B., Kawabata, A. (1997) Proteinase-activated receptors: structural requirements for activity, receptor cross-reactivity, and receptor selectivity of receptoractivating peptides. Can. J. Physiol. Pharmacol., 75: 832–841. Honing, M.L., Smits, P. (2000) Bradykinin-induced vasodilation of human forearm resistance vessels is primarily mediated by endothelium-dependent hyperpolarization. Hypertension, 35(6): 1314–1318. Honkakoski, P., Negishi, M. (2000) Regulation of cytochrome P450 (CYP) genes by nuclear receptors. Biochem. J., 347: 321–337. Horibe, M., Ogawa, K., Sohn, J.T., Murray, P.A. (2000) Propofol attenuates acetylcholine-induced pulmonary vasorelaxation: role of nitric oxide and endothelium-derived hyperpolarizing factors. Anesthesiology, 93: 447–455. Houghton, J.L., Davidson, C.A., Kuhner, P.A., Torossov, M.T., Strogatz, D.S., Carr, A.A. (1998) Heterogenous vasomotor responses of coronary conduit and resistance vessels in hypertension. J. Am. Coll. Cardiol., 31: 374–382. Houslay, M.D. (1998) Adaptation in cyclic AMP signalling processes: a central role for cyclic AMP phosphodiesterases. Semin. Cell Dev. Biol., 9: 161–167. Houslay, M.D., Milligan, G. (1997) Tailoring cAMP-signalling responses through isoform multiplicity. Trends. Biochem. Sci., 22: 217–224. Hu, S., Kim, H.S. (1993) Activation of K⫹ channel in vascular smooth muscles by cytochrome P450 metabolites of arachidonic acid. Eur. J. Pharmacol., 230: 215–221. Huang, P.L., Huang, Z., Mashimo, H., Bloch, K.D., Moskowitz, M.A., Bevan, J.A., Fishman, M.C. (1995) Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature, 377: 239–242. Huang, A.H., Busse, R., Bassenge, E. (1988) Endothelium-dependent hyperpolarization of smooth muscle cells in rabbit femoral arteries is not mediated by EDRF (nitric oxide). Naunyn Schmiedebergs Arch. Pharmacol., 338: 438–442. Huang, Y., Chan, F.L., Lau, C.W., Tsang, S.Y., He, G.W., Chen, Z.Y., Yao, X. (2002) Urocortin-induced endothelium-dependent relaxation of rat coronary artery: role of nitric oxide and K⫹ channels. Br. J. Pharmacol., 135: 1467–1476. Hubel, C.A. (1999) Oxidative stress in the pathogenesis of preeclampsia. Exp. Biol. Med., 222: 222–235. Hulme, J.T., Martens, J.R., Navarro-Polanco, R.A., Nishiyama, A., Tamkun, M.M. (2001) Voltagegated K⫹ channels in the myocardium. In Potassium Channels in Cardiovascular Biology, edited by S.L. Archer and N.J. Rusch, pp. 337–356. New York, NY: Kluwer Academic/Plenum Publishers. Hungerford, J.E., Sessa, W.C., Segal, S.S. (2000) Vasomotor control in arterioles of the mouse cremaster muscle. FASEB J., 14: 197–207. Hutcheson, I.R., Chaytor, A.T., Evans, W.H., Griffith, T.M. (1999) Nitric oxide-independent relaxations to acetylcholine and A23187 involve different routes of heterocellular communication. Role of gap junctions and phospholipase A2. Circ. Res., 84: 53–63. Hwa, J.J., Ghibaudi, L., Williams, P., Chatterjee, M. (1994) Comparison of acetylcholine-dependent relaxation in large and small arteries of rat mesenteric vascular bed. Am. J. Physiol., 266: H952–H958. Hwa, J.J., Ghibaudi, L., Williams, P., Chintala, M., Zhang, R., Chatterjee, M., Sybertz, E. (1996) Evidence for the presence of a proteinase-activated receptor distinct from the thrombin receptor in vascular endothelial cells. Circ. Res., 78: 581–588. Ignarro, L.J., Buga, G.M., Wood, K.S., Byrns, R.E., Chaudhuri, G. (1987a) Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc. Natl. Acad. Sci. USA, 84: 9265–9269. Ignarro, L.J., Byrns, R.E., Buga, G.M., Wood, R.S., Chaudihuri, G. (1987b) Pharmacological evidence that endothelium-derived relaxing factors is nitric oxide: use of pyrogallol and superoxide dismutase to study the endothelium-dependent and nitric oxide elicited vascular smooth muscle relaxation. J. Pharmacol. Exp. Ther., 244: 181–189. Ignarro, L.J., Byrns, R.E., Wood, K.S. (1988) Biochemical and pharmacological properties of endothelium-derived relaxing factor and its similarity to nitric oxide radical. In Vasodilatation;

394

References

Vascular smooth muscle, peptides, autonomic nerves and endothelium, edited by P.M. Vanhoutte, pp. 427– 435, New York: Raven Press. Illiano, S., Nagao, T., Vanhoutte, P.M. (1992) Calmidazolium, a calmodulin inhibitor, inhibits endothelium-dependent relaxations resistant to nitro-L-arginine in the canine coronary artery. Br. J. Pharmacol., 107: 387–392. Imaeda, K., Yamamoto, Y., Fukuta, H., Koshita, M., Suzuki, H. (2000) Hyperpolarizationinduced dilatation of submucosal arterioles in the guinea-pig ileum. Br. J. Pharmacol., 131: 1121–1128. Imig, J.D., Inscho, E.W., Deichmann, P.C., Reddy, K.M., Falck, J.R. (1999) Afferent arteriolar vasodilation to the sulfonimide analog of 11, 12-epoxyeicosatrienoic acid involves protein kinase A. Hypertension, 33: 408–413. Inoue, R., Kuriyama, H. (1993) Dual regulation of cation-selective channels by muscarinic and ␣1adrenergic receptors in the rabbit portal vein. J. Physiol., 456: 427–448. Intengan, H.D., Schiffrin, E.L. (2000) Structure and mechanical properties of resistance arteries in hypertension. Role of adhesion molecules and extracellular matrix determinants. Hypertension, 36: 312–318. Ishii, T.M., Maylie, J., Adelman, J.P. (1997a) Determinants of apamin and d-tubocurarine block in SK potassium channels. J. Biol. Chem., 272: 23195–23200. Ishii, T.M., Silvia, C., Hirschberg, B., Bond, C.T., Adelman, J.P., Maylie, J. (1997b) A human intermediateconductance calcium-activated potassium channel. Proc. Natl. Acad. Sci. USA, 94: 11651–11656. Iwamuro, Y., Miwa, S., Zhang, X.-F., Minowa, T., Enoki, T., Okamoto, Y., Hasegawa, H., Furutani, H., Okazawa, M., Ishikawa, M., Hashimoto, N., Masaki, T. (1999) Activation of three types of voltageindependent Ca2⫹ channel in A7r5 cells by endothelin-1 as revealed by a novel Ca2⫹ channel blocker LOE 908. Br. J. Pharmacol., 126: 1107–1114. Jackson, W.F., Blair, K.L. (1998) Characterization and function of Ca2⫹-activated K⫹ channels in arteriolar muscle cells. Am. J. Physiol., 274: H27–H34. Jaggar, J.H., Wellman, G.C., Heppner, T.J., Porter, V.A., Perez, G.J., Gollasch, M., Kleppisch, T., Rubart, M., Stevenson, A.S., Lederer, W.J., Knot, H.J., Bonev, A.D., Nelson, M.T. (1998a) Ca2⫹ channels, ryanodine receptors and Ca2⫹-activated K⫹ channels: a functional unit for regulating arterial tone. Acta Physiol. Scand., 164: 577–587. Jaggar, J.H., Stevenson, A.S., Nelson, M.T. (1998b) Voltage dependence of Ca2⫹ sparks in intact cerebral arteries. Am. J. Physiol., 274: C1755–C1761. Jaggar, J.H., Porter, V.A., Lederer, W.J., Nelson, M.T. (2000) Calcium sparks in smooth muscle. Am. J. Physiol., 278: C235–C256. Jain, V., Vedernikov, Y.P., Saade, G.R., Chwalisz, K., Garfield, R.E. (1997) The relaxation responses to corticotropin-releasing factor in rat aorta are endothelium dependent and gestationally regulated. Am. J. Obstet. Gynecol., 176: 234–240. Jensen, B.S., Strobaek, D., Christophersen, P., Jorgensen, T.D., Hansen, C., Silahtaroglu, A., Olesen, S.P., Ahring, P.K. (1998) Characterization of the cloned human intermediate-conductance Ca2⫹activated K⫹ channel. Am. J. Physiol., 275: C848–C856. Jeremy, R.W., McCarron, H. (2000) Effect of hypercholesterolemia on Ca(2⫹)-dependent K(⫹) channel-mediated vasodilatation in vivo. Am. J. Physiol. Heart. Circ. Physiol., 279: H1600–H1608. Jiang, F., Dusting, G.J. (2001) Endothelium-dependent vasorelaxation independent of nitric oxide and K⫹ release in isolated renal arteries of rats. Br. J. Pharmacol., 132: 1558–1564. Jiang, Z., Wallner, M., Meera, P., Toro, L. (1999) Human and rodent maxiK channel beta-subunit genes: cloning and characterization. Genomics, 55: 57–67. Johns, A., Lategan, T.N., Lodge, N.J., Ryan, U.S., van Breemen, C., Adams, D.J. (1987) Calcium entry through receptor-operated channels in bovine pulmonary artery endothelial cells. Tissue & Cell. 19(6): 733–745. Jow, F., Sullivan, K., Sokol, P., Numann, R. (1999) Induction of Ca2⫹-activated K⫹ current and transient outward current in human capillary endothelial cells. J. Membr. Biol., 167: 53–64. Juhaszova, M., Blaustein, M.P. (1997) Na⫹ pump low and high ouabain affinity ␣-subunit isoforms are differently distributed in cells. Proc. Natl. Acad. Sci. USA, 94: 1800–1805.

References

395

Kalin, M.F., Zumoff, B. (1990) Sex hormones and coronary disease: a review of the clinical studies. Steroids, 55: 330–352. Kamata, H., Hirata, H. (1999) Redox regulation of cellular function. Cell Signall., 11: 1–14. Kamata, K., Umeda, F., Kasuya, Y. (1996) Possible existence of novel endothelium-derived relaxing factor in the endothelium of rat mesenteric arterial bed. J. Cardiovasc. Pharmacol., 27: 601–606 Kamouchi, M., Droogmans, G., Nilius, B. (1999a) Membrane potential as a modulator of the free intracellular Ca2⫹ concentration in agonist-activated endothelial cells. Gen. Physiol. Biophys., 18: 199–208. Kamouchi, M., Philipp, S., Flockerzi, V., Wissenbach, U., Mamin, A., Raeymaekers, L., Eggermont, J., Droogmans, G., Nilius, B. (1999b) Properties of heterologously expressed hTRP3 channels in bovine pulmonary artery endothelial cells. J. Physiol., 518: 345–358. Kampf, C., Roomans, G.M. (2001) Effect of hypochlorite on cultured respiratory epithelial cells. Free Radic. Res., 34: 499–511. Kanmura, Y., Itoh, T., Kuriyama, H. (1988) Agonist actions of Bay K 8644, a dihydropyridine derivative, on the voltage-dependent calcium influx in smooth muscle cells of the rabbit mesenteric artery. J. Pharmacol. Exp. Ther., 231(3): 717–723 Kansui, Y., Fujii, K., Goto, K., Abe, I., Iida, M. (2002) Angiotensin II receptor antagonist improves age-related endothelial dysfunction. J. Hypertens., 20: 439–446. Karamsetty, M.R., Nakashima, J.M., Ou, L., Klinger, J.R., Hill, N.S. (2001) EDHF contributes to strain-related differences in pulmonary arterial relaxation in rats. Am. J. Physiol. Lung Cell. Mol. Physiol., 280: L458–L464. Karschin, A. (1999) G protein regulation of inwardly rectifying K⫹ channels. News Physiol. Sci., 14: 215–220. Katusic, Z.S. (1996) Superoxide anion and endothelial regulation of arterial tone. Free Radical Biology & Medicine, 20(3): 443–448. Katusic, Z.S. (2001) Vascular endothelial dysfunction: does tetrahydrobiopterin play a role? Am. J. Physiol. Heart Circ. Physiol., 281(3): H981–H986. Katusic, Z.S., Vanhoutte, P.M. (1989) Superoxide anion is an endothelium-derived contracting factor. Am. J. Physiol., 257: H33–H37. Kauser, K., Rubanyi, G.M. (1997) Potential cellular signaling mechanisms mediating upregulation of endothelial nitric oxide production by estrogen. J. Vasc. Res., 34: 229–236. Kauser, K., Clark, J.E., Masters, B.S., Ortiz de Montellano, P.R., Ma, Y-H., Harder, D.R., Roman, R.J. (1991) Inhibitors of cytochrome P-450 attenuate the myogenic response of dog renal arcuate arteries. Circ. Res., 68: 1154–1163. Kawanabe, Y., Hashimoto, N., Masaki, T. (2002) Ca2⫹ channels involved in endothelin-induced mitogenic response in carotid artery vascular smooth muscle cells. Am. J. Physiol., 282: C330–C337. Keaney, J.F., Jr., Vita, J.A. (1995) Atherosclerosis, oxidative stress, and antioxidant protection in endothelium-derived relaxing factor action. Prog. Cardiovasc. Dis., 38: 129–154. Kenny, L.C., Baker, P.N., Randall, M.D., Kendall, D.A., Dunn, W.R. (2002) The role of gap junctions in mediating endothelium-dependent responses to bradykinin in small myometrial arteries isolated from pregnant women. Br. J. Pharmacol., 136(8): 1085–1088. Kerr, S., Brosnan, M.J., McIntyre, M., Reid, J.L., Doniniczak, A.F., Hamilton, C.A. (1999) Superoxide anion production is increased in a model of genetic hypertension. Role of endothelium. Hypertension, 33: 1353–1358. Kestler, H.A., Janko, S., Haussler, U., Muche, R., Hombach, V., Hoher, M., Wiecha, J. (1998) A remark on the high-conductance calcium-activated potassium channel in human endothelial cells. Res. Exp. Med. (Berlin), 198: 133–143. Khanna, R., Roy, L., Zhu, X., Schlichter, L.C. (2001) K⫹ channels and the microglial respiratory burst. Am. J. Physiol., 280: C796–C806. Kharitonov, S.A., Logan-Sinclair, R.B., Busset, C.M., Shinebourne, E.A. (1994) Peak expiratory nitric oxide differences in men and women: relation to the menstrual cycle. Br. Heart. J., 72: 243–245. Kilpatrick, E.V., Cocks, T.M. (1994) Evidence for differential roles of nitric oxide (NO) and hyperpolarization in endothelium-dependent relaxation of pig isolated coronary artery. Br. J. Pharmacol., 112: 557–565.

396

References

Kimura, K., Nishio, I. (1999) Impaired endothelium-dependent relaxation in mesenteric arteries of reduced renal mass hypertensive rats. Scand. J. Clin. Invest., 59: 199–204. King, G.L., Ishii, H., Koya, D. (1997) Diabetic vascular dysfunctions: a model of excessive activation of protein kinase C. Kidney Int. Suppl., 60: S77–S85. Kiss, L., Bieniek, E., Weissmann, N., Schutte, H., Sibelius, U., Gunther, A., Bier, J., Mayer, K., Henneking, K., Padberg, W., Grimm, H., Seeger, W., Grimminger, F. (1998) Simultaneous analysis of 4- and 5-series lipoxygenase and cytochrome P450 products from different biological sources by reversed-phase high-performance liquid chromatographic technique. Anal. Biochem., 261: 16–28. Klein, M.G., Cheng, H., Santana, L.F., Jiang, Y.H., Lederer, W.J., Schneider, M.F. (1996) Two mechanisms of quantized calcium release in skeletal muscle. Nature, 379: 455–458. Klockner, U., Isenberg, G. (1991) Endothelin depolarizes myocytes from porcine coronary and human mesenteric arteries through a Ca-activated chloride current. Pflügers Arch., 418: 168–175. Knaus, H.G., Eberhart, A., Kaczorowski, G.J., Garcia, M.L. (1994) Covalent attachment of charybdotoxin to the beta-subunit of the high conductance Ca2⫹-activated K⫹ channel. Identification of the site of incorporation and implications for channel topology. J. Biol. Chem., 269: 23336–23341. Knekt, P., Jarvinen, R., Reunanen, A., Maatela, J. (1996) Flavonoid intake and coronary mortality in Finland: a cohort study. BMJ, 312: 478–481. Knock, G.A., Poston, L. (1996) Bradykinin-mediated relaxation of isolated maternal resistance arteries in normal pregnancy and preeclampsia. Am. J. Obstet. Gynecol., 175: 1668–1674. Knot, H.J., Nelson, M.T. (1998) Regulation of arterial diameter and wall [Ca2⫹] in cerebral arteries of rat by membrane potential and intravascular pressure. J. Physiol., 508: 199–209. Knot, H.J., Zimmerman, P.A., Nelson, M.T. (1996) Extracellular K⫹-induced hyperpolarizations and dilatations of rat coronary and cerebral arteries involve inward rectifier K⫹ channels. J. Physiol., 492: 419–430. Knot, H.J., Standen, N.B., Nelson, M.T. (1998) Ryanodine receptors regulate arterial diameter and wall [Ca2⫹] in cerebral arteries of rat via Ca2⫹-dependent K⫹ channels. J. Physiol., 508: 211–221. Knot, H.J., Lounsbury, K.M., Brayden, J.E., Nelson, M.T. (1999) Gender differences in coronary artery diameter reflect changes in both endothelial Ca2⫹ and ecNOS activity. Am. J. Physiol., 276: H961–H969. Kohler, R., Dgenhardt, C., Kuhn, M., Runkel, N., Paul, M., Hoyer, J. (2000) Expression and function of endothelial Ca2⫹-activated K⫹ channels in human mesenteric artery. Circ. Res., 87: 496–503. Kohler, R., Brakemeier, S., Kuhn, M., Behrens, C., Real, R., Degenhardt, C., Orzechowski, H.-D., Pries, A.R., Paul, M., Hoyer, J. (2001) Impaired hyperpolarization in regenerated endothelium after balloon catheter injury. Circ. Res., 89: 174–179. Koksoy, A.A., Allen, J.C. (2001) Regulation of Na pump expression by vascular smooth muscle cells. Amer. Jour. Physiol., 280: H1869–H1874. Koksoy, A.A., Abramowitz, J., Allen, J.C. (2001) Ouabain induced signaling and vascular smooth muscle cell proliferation. J. Biol. Chem., 276: 46605–46611. Koller, A., Kaley, G. (1990) Prostaglandins mediate arteriolar dilatation to increased blood flow velocity in skeletal muscle microcirculation. Circ. Res., 67: 529–534. Koller, A., Sun, D., Kaley, G. (1993) Role of shear stress and endothelial prostaglandins in flow-induced and viscosity-induced dilation of arterioles in vitro. Circ. Res., 72: 1276–1284. Komaru, T., Kanatsuka, H., Shirato, K. (2000) Coronary microcirculation. Physiology and pharmacology. Pharmacol. Ther., 86: 217–261. Komiya, T., Tanigawa, Y., Hirohashi, S. (1999) Cloning and identification of the gene gob-5, which is expressed in intestinal goblet cells in mice. Biochem. Biophys. Res. Commun., 255: 347–351. Komori, K., Lorenz, R.R., Vanhoutte, P.M. (1988) Nitric oxide, ACh, and electrical and mechanical properties of canine arterial smooth muscle. Am. J. Physiol., H207–H212. Krötz, F., Sohn, H.-Y., Gloe, T., Zahler, S., Riexinger, T., Schiele, T.M., Becker, B.F., Theisen, K., Klauss, V., Pohl, U. (2002) NAD(P)H-oxidase-dependent platelet superoxide anion release increases platelet recruitment. Blood, 100(3): 917–924.

References

397

Kumar, K.V., Das, U.N. (1993) Are free radicals involved in the pathobiology of human essential hypertension? Free Rad. Res. Commun., 19: 59–66. Kurjiaka, D.T., Segal, S.S. (1995) Conducted vasodilation elevates flow in arteriole networks of hamster striated muscle. Am. J. Physiol. Heart Circ. Physiol., 269: H1723–H1728. Lückhöff, A., Busse, R. (1990) Activators of potassium channels enhance calcium influx into endothelial cells as a consequence of potassium currents. Naunyn Schmiedebergs Arch. Pharmacol., 342: 94–109. Lüscher, T.F., Vanhoutte, P.M. (1986) Endothelium-dependent contractions to acetylcholine in the aorta of the spontaneously hypertensive rat. Hypertension, 8: 344–348. Lacy, F., Kailasam, M.T., O’Connor, D.T., Schmid-Schnbein, G.W., Parmer, R.J. (2000) Plasma hydrogen peroxide production in human essential hypertension: role of heredity, gender, and ethnicity. Hypertension, 36: 878–884. Lacy, P.S., Pilkington, G., Hanvesakul, R., Fish, H.J., Boyle, J.P., Thurston, H. (2000) Evidence against potassium as an endothelium-derived hyperpolarizing factor in rat mesenteric small arteries. Br. J. Pharmacol., 129: 605–611. Lagaud, G.J., Skarsgard, P.L., Laher, I., van Breemen, C. (1999) Heterogeneity of endotheliumdependent vasodilation in pressurized cerebral and small mesenteric resistance arteries of the rat. J. Pharmacol. Exp. Ther., 290: 832–839. Large, W.A., Wang, Q. (1996) Characteristics and physiological role of the Ca2⫹-activated Cl⫺ conductance in smooth muscle. Am. J. Physiol., 271: C435–C454. Large, W.A. (2002) Receptor-operated Ca2⫹-permeable nonselective cation channels in vascular smooth muscle: a physiologic perspective. J. Cardiovasc. Electrophysiol., 13: 493–501. Laterra, J., Guerin, C., Goldstein, G.W. (1990) Astrocytes induce neural microvascular endothelial cells to form capillary-like structures in vitro. J. Cell. Physiol., 144: 204–215. LeBel, C.P., Bondy, S.C. (1990) Sensitive and rapid quantitation of oxygen reactive species formation in rat synaptosomes. Neurochem. Int., 17: 435–440. Lefroy, D.C., Crake, T., Uren, N.G., Davies, G.J., Maseri, A. (1993) Effect of inhibition of nitric oxide synthesis on epicardial coronary artery calibre and coronary blood flow in humans. Circulation, 88: 43–54. Levine, G.N., Frei, B., Koulouris, S.N., Gerhard, M.D., Keaney, J.F., Jr., Vita, J.A. (1996) Ascorbic acid reverses endothelial vasomotor dysfunction in patients with coronary artery disease. Circulation, 93: 1107–1113. Li, P.-L., Campbell, W.B. (1997) Epoxyeicosatrienoic acids activate K⫹ channels in coronary smooth muscle through a guanine nucleotide binding protein. Circ. Res., 80: 877–884. Li, Z., Duckles, S.P. (1994) Influence of gender on vascular reactivity in the rat. J. Pharmacol. Exp. Ther., 268: 1426–1431. Li, P.L., Zou, A.P., Campbell, W.B. (1997) Regulation of potassium channels in coronary arterial smooth muscle by endothelium-derived vasodilators. Hypertension, 29: 262–267. Li, K.-X., Fouty, B., McMurtry, I.F., Rodman, D.M. (1999a) Enhanced ETA-receptor-mediated inhibition of KV channels in hypoxic hypertensive rat pulmonary artery myocytes. Am. J. Physiol., 277: H363–H370. Li, P.L., Chen, C.L., Bortell, R., Campbell, W.B. (1999b) 11,12-epoxyeicosatrienoic acid stimulates endogenous mono-ADP-ribosylation in bovine coronary arterial smooth muscle. Circ. Res., 85: 349–356. Lin, J.H-C., Kobari, Y., Zhu, Y., Stemerman, M.B., Pritchard, K.A. (1996) Human umbilical vein endothelial cells express P450 2C8 mRNA: cloning of endothelial P450 epoxygenase. Endothelium, 4: 219–229. Lind, L., Granstam, S-O., Millgard, J. (2000) Endothelium-dependent vasodilation in hypertension: A review. Blood Pressure, 9: 4–15. Lipp, P., Niggli, E. (1996) Submicroscopic calcium signals as fundamental events of excitation– contraction coupling in guinea-pig cardiac myocytes. J. Physiol., 492: 31–38. Little, T.L., Xia, J., Duling, B.R. (1995) Dye tracers define differential endothelial and smooth muscle coupling patterns within the arteriolar wall. Circ Res., 76: 498–504.

398

References

Liu, Q., Wiener, C.M., Flavahan, N.A. (1998a) Superoxide and endothelium-dependent constriction to flow in porcine small pulmonary arteries. Br. J. Pharmacol., 124: 331–336. Liu, X., Hymel, L.J., Songu-Mize, E. (1998b) Role of Na and Ca in stretch induced Na K ATPase ␣ subunit regulation in aortic smooth muscle cells. Amer. J. Physiol., 274: H83–H89. Liu, Y., Hudetz, A.G., Knaus, H.-G., Rusch, N.J. (1998c) Increased expression of Ca2⫹-sensitive K⫹ channels in the cerebral microcirculation of genetically hypertensive rats. Circ. Res., 82: 729–737. Liu, Z.-G., Ge, Z.-D., He, G.-W. (2000) Difference in endothelium-dervied hyperpolarizing factorsmediateed hyperpolarization and nitric oxide release between human interal mammary artery and saphenous vein. Circulation, 102(III): 296–301. Liu, M.Y., Hattori, Y., Fukao, M., Sato, A., Sakuma, I., Kanno, M. (2001) Alterations in EDHFmediated hyperpolarization and relaxation in mesenteric arteries of female rats in long-term deficiency of oestrogen and during oestrus cycle. Br. J. Pharmacol., 132: 1035–1046. Lopez-Lopez, J.R., Shacklock, P.S., Balke, C.W., Wier, W.G. (1994) Local, stochastic release of Ca2⫹ in voltage-clamped rat heart cells: visualization with confocal microscopy. J. Physiol., 480: 21–29. Lounsbury, K.M., Hu, Q., Ziegelstein, R.C. (2000) Calcium signaling and oxidant stress in the vasculature. Free Radic. Biol. Med., 28: 1362–1369. Loutzenhiser, R.D., Epstein, M., Fischetti, F., Horton, C. (1989) Effects of amlodipine on renal hemodynamics. Am. J. Cardiol., 64: 122I–127I. Lückhöff, A., Busse, R. (1990) Calcium influx into endothelial cells and formation of endotheliumderived relaxing factor is controlled by the membrane potential. Pflugers Arch., 416: 305–311. Lugnier, C., Komas, N. (1993) Modulation of vascular cyclic nucleotide phosphodiesterases by cyclic GMP: role in vasodilatation. Eur. Heart. J., 14 (Suppl I): 141–148. Luscher, T. (1990) Imbalance of endothelium-derived relaxing and contracting factors: a new concept in hypertension. Am. H. Hypertension, 3: 317–330. Lüscher, T.F., Vanhoutte, P.M. (1986) Endothelium-dependent contractions to acetylcholine in the aorta of the spontaneously hypertensive rat. Hypertension, 8: 344–348. Ma, Y.H., Gebremedhin, D., Schwartzman, M.L., Falck, J.R., Clark, J.E., Masters, B.S., Harder, D.R., Roman, R.J. (1993) 20-Hydroxyeicosatetraenoic acid is an endogenous vasoconstrictor of canine renal arcuate arteries. Circ. Res., 72: 126–136. McCulloch, A.I., Randall, M.D. (1998) Sex differences in the relative contributions of nitric oxide and EDHF to agonist-stimulated endothelium-dependent relaxations in the rat isolated mesenteric arterial bed. Br. J. Pharmacol., 123: 1700–1706. McCarthy, A.L., Woolfson, R.G., Raju, S.K., Poston, L. (1993) Abnormal endothelial cell function of resistance arteries from women with preeclampsia. Am. J. Obstet. Gynecol., 168: 1323–1330. McCulloch, A.I., Bottrill, F.E., Randall, M.D., Hiley, C.R. (1997) Characterization and modulation of EDHF-mediated relaxations in the rat isolated superior mesenteric arterial bed. Br. J. Pharmacol., 120: 1431–1438. McGuire, J.J., Triggle, C.R., Hollenberg, M.D. (2000) Effects of SLIGRL-NH2 on femoral arteries from eNOS knockout mice. FASEB, 14: A1403. McGuire, J.J., Hollenberg, M.D., Andrade-Gordon, P., Triggle, C.R. (2002) Multiple mechanisms of vascular smooth muscle relaxation by the activation of proteinase-activated receptor 2 in mouse mesenteric arterioles. Br. J. Pharmacol., 135: 155–169. McGuire, J.J., Ding, H., Triggle, C.R. (2001) Endothelium-derived relaxing factors: a focus on endothelium-derived hyperpolarizing factor(s). Can. J. Physiol. Pharmacol., 79: 443–470. McHale, N.G., Allen, J.M. (1983) The effect of external Ca2⫹ concentration on the contractility of bovine mesenteric lymphatics. Microvasc. Res., 26: 182–192. McLean, P.G., Aston, D., Sarkar, D., Ahluwalia, A. (2002) Protease-activated receptor-2 activation causes EDHF-like coronary vasodilation: selective preservation in ischemia/reperfusion injury: involvement of lipoxygenase products, VR1 receptors, and C-fibers. Circ. Res., 90: 465–472. McNeill, A.M., Kim, N., Duckles, S.P., Krause, D.N., Kontos, H.A. (1999) Chronic estrogen treatment increases levels of endothelial nitric oxide synthase protein in rat cerebral microvessels. Stroke, 30: 2186–2190.

References

399

McNeish, A.J., Wilson, W.S., Martin, W. (2001) Dominant role of an endothelium-derived hyperpolarizing factor (EDHF)-like vasodilator in the ciliary vascular bed of the bovine isolated perfused eye. Br. J. Pharmacol., 134: 912–920. McNeish, A.J., Wilson, W.S., Martin, W. (2002) Ascorbate blocks endothelium-derived hyperpolarizing factor (EDHF)-mediated vasodilatation in the bovine ciliary vascular bed and rat mesentery. Br. J. Pharmacol., 135: 1801–1809. McPherson, G.A., Angus, J.A. (1991) Evidence that acetylcholine-mediated hyperpolarization of the rat small mesenteric artery does not involve the K⫹ channel opened by cromakalim. Br. J. Pharmacol., 103: 1184–1190. McPherson, G.A., Keily, S.G. (1995) Electrophysiological properties of the rat middle cerebral artery at different levels of passive wall tension. Clin. Exp. Pharmacol. Physiol., 22: 724–731. MacIntyre, D.E., Rink, T.J. (1982) The role of platelet membrane potential in the initiation of platelet aggregation. Thromb. Haemost., 47: 22–26. Mahaut-Smith, M.P. (1995) Calcium-activated potassium channels in human platelets. J. Physiol., 484 (Pt 1): 15–24. Mahaut-Smith, M.P., Rink, T.J., Collins, S.C., Sage, S.O. (1990) Voltage-gated potassium channels and the control of membrane potential in human platelets. J. Physiol., 428: 723–735. Makino, A., Ohuchi, K., Kamata, K. (2000) Mechanisms underlying the attenuation of endotheliumdependent vasodilatation in the mesenteric arterial bed of the streptozotocin-induced diabetic rat. Br. J. Pharmacol., 130: 549–556. Malmsjo, M., Edvinsson, L., Erlinge, D. (1988) P2U-receptor mediated endothelium-dependent but nitric oxide-independent vascular relaxation. Br. J. Pharmacol., 123: 719–729. Mano, T., Masuyama, T., Yamamoto, K., Naito, J., Kondo, H., Nagano, R., Tanouchi, J., Hori, M., Inoue, M., Kamada, T. (1996) Endothelial dysfunction in the early stage of atherosclerosis precedes appearance of intimal lesions assessable with intravascular ultrasound. Am. Heart. J., 131: 231–238. Marchenko, S.M., Sage, S.O. (1993) Electrical properties of resting and acetylcholine-stimulated endothelium in intact rat aorta. J. Physiol., 462: 735–751. Marchenko, S.M., Sage, S.O. (1994a) Mechanism of acetylcholine action on membrane potential of endothelium of intact rat aorta. Am. J. Physiol., 266: H2388–H2395. Marchenko, S.M., Sage, S.O. (1994b) Smooth muscle cells affect endothelial membrane potential in rat aorta. Am. J. Physiol., 267: H804–H811. Marchenko, S.M., Sage, S.O. (1996) Calcium-activated potassium channels in the endothelium of intact rat aorta. J. Physiol., 492: 53–60. Marrelli, S.P. (2000) Selective measurement of endothelial or smooth muscle [Ca(2⫹)](i) in pressurized/ perfused cerebral arteries with fura-2. J. Neurosci. Methods, 97: 145–155. Marrelli, S.P. (2001) Mechanisms of endothelial P2Y(1)- and P2Y(2)-mediated vasodilatation involve differential [Ca2⫹]i responses. Am. J. Physiol. Heart Circ. Physiol., 281: H1759–H1766. Martin, S., Andriambeloson, E., Takeda, K., Andriantsitohaina, R. (2002) Red wine polyphenols increase calcium in bovine aortic endothelial cells: a basis to elucidate signalling pathways leading to nitric oxide production. Br. J. Pharmacol., 135: 1579–1587. Maruyama, Y. (1987) A patch-clamp study of mammalian platelets and their voltage-gated potassium current. J. Physiol., 391: 467–485. Maruyama, T., Kanaji, T., Nakade, S., Kanno, T, Mikoshiba, K. (1997) 2-APB, 2-aminoethoxydiphenyl borate, a membrane-permeable modulator of Ins(1,4,5)P3-induced Ca2⫹ release. J. Biochem., 122: 498–505. Matoba, T., Shimokawa, H., Nakashima, M., Hirakawa, Y., Mukai, Y., Hirano, K., Kanaide, H., Takeshita, A. (2000) Hydrogen peroxide is an endothelium-derived hyperpolarizing factor in mice. J. Clin. Invest., 106: 1521–1530. Matoba, T., Shimokawa, H.C. (2002) Hydrogen peroxide is an endothelium-derived hyperpolarizing factor in human mesenteric arteries. Biochem. Biophys. Res. Commun., 290: 909–913. May, J.M. (2000) How does ascorbic acid prevent endothelial dysfunction? Free Radic. Biol. Med., 28: 1421–1429.

400

References

Medhora, M.M. (2000) Retinoic acid up-regulates the beta-1 integrin subunit in vascular smooth muscle cells and alters adhesion to fibronectin. Am. J. Physiol., 279: H382–H387. Medhora, M., Harder, D. (1998) Functional role of epoxyeicosatrienoic acids and their production in astrocytes: approaches for gene transfer and therapy (review). Int. J. Mol. Med., 2: 661–669. Medhora, M., Narayanan, J. Harder, D.R (2001a) Analysis of the role of the putative EDHFepoxyeicosatrienoic acid on gene expression in human coronary endothelial cells. FASEB J., 15: A477. Medhora, M., Narayanan, J., Harder, D. (2001b) Dual regulation of the cerebral microvasculature by epoxyeicosatrienoic acids. Trends Cardiovasc. Med., 11: 38–42. Medhora, M., Narayanan, J., Harder, D., Maier, K.G. (2001c) Identifying endothelium-derived hyperpolarizing factor: recent approaches to assay the role of epoxyeicosatrienoic acids. Jpn. J. Pharmacol., 86: 369–375. Medhora, M., Daniels, J., MacAndrew, R.P., Zhang, C., Jacobs, E.R., Harder, D.R. (2002) EETsinduced angiogenesis in human pulmonary microvascular endothelial cells. FASEB J., 16: A88. Meerson, F.Z., Alechina, G.M. (1968) Action of intensity of the contractile function of muscle cells of the heart on the DNA synthesis of the intermediate connective tissue of the myocardium. Acta Anat. Bazel., 70: 559–567. Mehrke, G., Daut, J. (1990) The electrical response of cultured guinea-pig coronary endothelial cells to endothelium-dependent vasodilators. J. Physiol., 430: 251–272. Meij, I.C., Koenderink, J.B., Bokhoven, H.V., Assink, K.F.H., Groenestege, W.T., Joep, J., de Pont, H.H.M., Bindels, R.J.M., Monnens, L.A.H., van den Heuvel, L.P.W.J., Knoers, N.V.A.M. (2000) Dominant isolated renal magnesium loss is caused by misrouting of the Na⫹,K⫹-ATPase ␥subunit. Nature Gene., 26: 265–266. Mesaeli, N., Tappia, P.S., Suzuki, S., Dhalla, N.S., Panagia, V. (2000) Oxidants depress the synthesis of phosphatidylinositol 4,5-bisphosphate in heart sarcolemma. Arch. Biochem. Biophys., 382: 48–56. Mieyal, P., Fulton, D., McGiff, J.C., Quilley, J. (1998) NO-independent vasodilation to acetylcholine in the rat isolated kidney utilizes a charybdotoxin-sensitive, intermediate-conductance Ca⫹⫹activated K⫹ channel. J. Pharmacol. Exp. Ther., 285: 659–664. Millar, C., Kaufman, P.L. (1995) Aqueous humor: secretion and dynamics. In Duane’s Foundations of Clinical Ophthalmology, edited by W. Tasmin, pp. 1–52. London and Philadelphia: Harper & Row. Miller, C., Moczydlowski, E., Latorre, R., Phillips, M. (1985) Charybdotoxin, a protein inhibitor of single Ca2⫹-activated potassium channels from mammalian skeletal muscle. Nature, 313: 316–318. Miller, V.M. (1991) Selective production of endothelium-derived nitric oxide in canine femoral veins. Am. J. Physiol., 261: H677–H682. Miller, V.M., Vanhoutte, P.M. (1989) Relaxations to SIN-1, nitric oxide, and sodium nitroprusside in canine arteries and veins. J. Cardiovasc. Pharmacol., 14: suppl 11, S67–S71. Millette, E., de Champlain, J., Lamontagne, D. (2000) Altered coronary dilation in deoxycorticosterone acetate-salt hypertension. J. Hypertension, 18: 1783–1793. Mineo, C., Gill, G., Anderson, R. (1999) Regulated migration of EGFR from caveolae. J. Biol. Chem., 274: 30636–30643. Mitchell, J.B., Samuni, A., Krishna, M.C., DeGraff, W.G., Ahn, M.S., Samuni, U., Russo, A. (1990) Biologically active metal-independent superoxide dismutase mimetics. Biochemistry, 29: 2802–2807. Miura, H., Liu, Y., Gutterman, D.D. (1999) Human coronary arteriolar dilation to bradykinin depends on membrane hyperpolarization: contribution of nitric oxide and Ca2⫹-activated K⫹ channels. Circulation, 99: 3132–3138. Miura, H., Wachtel, R.E., Liu, Y., Loberiza, F.R., Saito, T., Miura, M., Gutterman, D.D. (2001) Flowinduced dilation of human coronary arterioles: important role of Ca2⫹-activated K⫹ channels. Circulation, 103: 1992–1998. Miyoshi, A., Suzuki, M., Masai, M., Iwasaki, T. (1997) Impairment of endothelial function in saltsensitive hypertension in humans. Am. J. Hypertension, 10: 1083–1090. Moffatt, J.D., Cocks, T.M. (1998) Endothelium-dependent and -independent responses to proteaseactivated receptor-2 (PAR-2) activation in mouse isolated renal arteries. Br. J. Pharmacol., 125: 591–594.

References

401

Mohammed, K., Ayanafshar, B., Liu, L., Gable, M., Xie, Z., Askari, A. (2002) Ouabain interaction with cardiac Na K ATPase induces Src translocation to caveolae. Abstract No. 835, Exp. Biol., New Orleans. Mombouli, J.V., Vanhoutte, P.M. (1997) Endothelium-derived hyperpolarizing factor(s): updating the unknown. Trends Pharmacol. Sci., 18(7): 252–256. Mombouli, J-V., Vanhoutte, P.M. (1999) Endothelial dysfunction: from physiology to therapy. J. Mol. Cell. Cardiol., 31: 61–74. Mombouli, J.-V., Illiano, S., Nagao, T., Scott-Burden, T., Vanhoutte, P.M. (1992) Potentiation of endothelium-dependent relaxations to bradykinin by angiotensin I converting enzyme inhibitors in canine coronary artery involves both endothelium-derived relaxing and hyperpolarizing factors. Circ. Res., 71: 137–144. Mombouli, J.V., Bissiriou, I., Vanhoutte, P.M. (1995) Simultaneous release of endothelium-derived hyperpolarizing and depolarizing factors. Circulation, 92: I-751. Mombouli, J.V., Bissiriou, I., Vanhoutte, P.M. (1996) Bioassay of endothelium-derived hyperpolarizing factor: Is endothelium-derived depolarizing factor a confounding element? In Endothelium-derived hyperpolarizing factor, edited by P.M. Vanhoutte, pp. 51–55. Amsterdam: Harwood Academic Publishers. (First Workshop; Endothelium-derived hyperpolarizing factor. Abbaye des Vaux de Cernay, France.) Moncada, S., Vane, J.R. (1978) Pharmacology and endogenous roles of prostaglandin endoperoxides, thromboxane A2, and prostacyclin. Pharmacol.Rev., 30: 293–331. Moncada, S., Vane, J.R. (1979a) The role of prostacyclin in vascular tissue. Fed. Proc., 38: 66–71. Moncada, S., Vane, J.R. (1979b) Pharmacology and endogenous roles of prostaglandin endoperoxides, thromboxane A2 and prostacyclin. Pharmacol. Rev., 30: 293–331. Moncada, S., Gryglewski, R.J., Bunting, S., Vane, J.R. (1976) An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature, 263: 663–665. Moncada, S., Higgs, E.A., Vane, J.R. (1977) Human arterial and venous tissues generate prostacyclin (prostaglandin x), a potent inhibitor of platelet aggregation. Lancet, 1: 18–20. Morita, H., Cousins, H., Onoue, H., Ito, Y., Inoue, R. (1999) Predominant distribution of Nifedipineinsensitive, high voltage-activated Ca2⫹ channels in the terminal mesenteric artery of guinea pig. Circ. Res., 85: 596–605. Morrison B.W., Moorman, J.R., Kowdley, G.C., Kobayashi, Y.M., Jones, L.R., Leder, P. (1995) Mat-8, a novel phospholemman-like protein expressed in human breast tumors, induces a chloride conductance in Xenopus oocytes. J. Biol. Chem., 270: 2176–2182. Mulvany, M.J., Halpern, W. (1977) Contractile properties of small mesenteric arteries resistance vessels in spontaneously hypertensive and normotensive rats. Circ. Res., 41: 19–26. Mulvany, M.J., Nilsson, H., Flatman, J.A. (1982) Role of membrane potential in the response of rat small mesenteric arteries to exogenous noradrenaline stimulation. J. Physiol., 332: 363–373. Mulvany, M.J., Baumbach, G.L., Aalkjaer, C., Heagerty, A.M., Korsgaard, N., Schiffrin, E.L., Heistad, D.D. (1996) Vascular remodeling. Hypertension, 28: 505–506. Munzenmaier, D.H., Harder, D.R. (2000) Cerebral microvascular endothelial cell tube formation: role of astrocytic epoxyeicosatrienoic acid release. Am. J. Physiol., 278: H1163–H1167. Murai, T., Muraki, K., Imaizumi, Y., Watanabe, M. (1999) Levcromakalim causes indirect endothelial hyperpolarization via a myo- endothelial pathway. Br. J. Pharmacol., 128: 1491–1496. Muramatsu, M., Fujii, S. (1968) Inhibitory effects of ␻-guanidino acid esters on tyrpsin, plasmin, thrombin and plasma kallikrein. J. Biochem., 64: 807–814. Murphy, J.G., Khalil, R.A. (2000) Gender-specific reduction in contractility and [Ca2⫹]i in vascular smooth muscle cells of female rat. Am. J. Physiol., 278: C834–C844. Murphy, M.E., Brayden, J.E. (1995a) Nitric oxide hyperpolarizes rabbit mesenteric arteries via ATPsensitive potassium channels. J. Physiol., 486: 47–58. Murphy, M.E., Brayden, J.E. (1995b) Apamin-sensitive K⫹ channels mediate an endothelium-dependent hyperpolarization in rabbit mesenteric arteries. J. Physiol., 489: 723–734.

402

References

Murthy, K.S., Makhlouf, G.M. (1998) Regulation of adenylyl cyclase type V/VI in smooth muscle: interplay of inhibitory G protein and Ca2⫹ influx. Mol. Pharmacol., 54: 122–128. Nagao, T., Vanhoutte, P.M. (1993) Endothelium-derived hyperpolarizing factor and endotheliumdependent relaxations. Am. J. Respir. Cell. Mol. Biol., 8: 1–6. Nagao, T., Illiano, S., Vanhoutte, P.M. (1992) Calmodulin antagonists inhibit endothelium-dependent hyperpolarization in the canine coronary artery. Br. J. Pharmacol., 107: 382–386. Nakajima, T., Iwasawa, K., Hazama, H., Asano, M., Okuda, Y., Omata, M. (1997) Extracellular Mg2⫹ inhibits receptor-mediated Ca2⫹-permeable non-selective cation currents in aortic smooth muscle cells. Eur. J. Pharmacol., 320: 81–86. Nakamura, Y., Ohya, Y., Abe, I., Fujishima, M. (1999) Sodium-potassium pump current in smooth muscle cells from mesenteric resistance arteries of the guinea-pig. J. Physiol., 519.1: 203–212. Nakao, M., Gadsby, D.C. (1989) [Na] and [K] dependence of the Na/K pump current-voltage relationship in guinea pig ventricular myocytes. J. Gen. Physiol., 94: 539–565. Nakashima, M., Vanhoutte, P.M. (1996) Decreased endothelium-dependent hyperpolarization with aging and hypertension in the rat mesenteric artery. In Endothelium-Derived Hyperpolarizing Factor, edited by P.M. Vanhoutte, pp. 227–233. The Netherlands: Harwood Academic Publishers. Nakashima, M., Mombouli, J.V., Taylor, A.A., Vanhoutte, P.M. (1993) Endothelium-dependent hyperpolarization caused by bradykinin in human coronary arteries. J. Clin. Invest., 92: 2867–2871. Nakayama, F. (1997) Panniculitis with eosinophilic infiltration due to gabexate mesilate (FOY): possibility of allergic reaction. J. Dermatol., 24(4): 235–242. Nara, M., Dhulipala, P.D., Wang, Y.X., Kotlikoff, M.I. (1998) Reconstitution of beta-adrenergic modulation of large conductance, calcium-activated potassium (Maxi-K) channels in Xenopus oocytes. Identification of the cAMP-dependent protein kinase phosphorylation site. J. Biol. Chem., 273: 14920–14924. Naruse, K., Sokabe, M. (1993) Involvement of stretch-activated ion channels in Ca2⫹ mobilization to mechanical stretch in endothelial cells. Am. J. Physiol., 264: C1037–C1044. Nekooeian, A.A., Lim, S.L., Man, R.Y.K., Pang, C.C.Y. (1998) Chronic 17␤-estradiol augments relaxant role of basal nitric oxide in blood vessels from rats with heart faillure. Naunyn-Schmiedeberg’s Arch Pharmacol., 358: 671–677. Nelson, M.T., Quayle, J.M. (1995) Physiological roles and properties of potassium channels in arterial smooth muscle. Am. J. Physiol.- Cell Physiol., 268: C799–C822. Nelson, M.T., Cheng, H., Rubart, M., Santana, L.F., Bonev, A.D., Knot, H.J., Lederer, W.J. (1995) Relaxation of arterial smooth muscle by calcium sparks. Science, 270: 633–637. Nichols, C.G., Lopatin, A.N. (1997) Inward rectifier potassium channels. Annu. Rev Physiol., 59: 171–191. Niggli, E., Lederer, W.J. (1990) Voltage-independent calcium release in heart muscle. Science, 250: 565–568. Nilius, B., Droogmans, G. (2001) Ion channels and their functional role in vascular endothelium. Physiol. Reviews, 81(4): 1415–1459. Nilius, B., Viana, F., Droogmans, G. (1997a) Ion channels in vascular endothelium. Annu. Rev. Physiol., 59: 145–170. Nilius, B., Prenen, J., Voets, T., Van den, B.K., Eggermont, J., Droogmans, G. (1997b) Kinetic and pharmacological properties of the calcium-activated chloride-current in macrovascular endothelial cells. Cell Calcium, 22: 53–63. Nishikawa, Y., Stepp, D.W., Chilian, W.M. (1999) In vivo location and mechanism of EDHF-mediated vasodilation in canine coronary microcirculation. Am. J. Physiol., 277: H1252–H1259. Nishikawa, Y., Stepp, D.W., Chilian, W.M. (2000) Nitric oxide exerts feedback inhibition on EDHFinduced coronary arteriolar dilation in vivo. Am. J. Physiol. Heart. Circ. Physiol., 279: H459–H465. Nishimura, Y., Usui, H., Kurahashi, H., Suzuki, A. (1995) Endothelium-dependent contraction induced by acetylcholine in isolated rat renal arteries. Eur. J. Pharmacol., 275: 217–221. Nishiyama, M., Hashitani, H., Fukuta, H., Yamamoto, Y., Suzuki, H. (1998) Potassium channels activated in endothelium-dependent hyperpolarization in guinea-pig coronary artery. J. Physiol., 510: 455–465.

References

403

Nithipatikom, K., DiCamelli, R.F., Kohler, S., Gumina, R.J., Falck, J.R., Campbell, W.B., Gross, G.J. (2001) Determination of cytochrome P450 metabolites of arachidonic acid in coronary venous plasma during ischemia and reperfusion in dogs. Anal. Biochem., 292: 115–124. Node, K., Huo, Y. (1999) Anti-inflammatory properties of cytochrome P450 epoxygenase-derived eicosanoids. Science, 285(5431): 1276–1279. Node, K., Huo, Y., Ruan, X., Yang, B., Spiecker, M., Ley, K., Zeldin, D.C., Liao, J.K. (1999) Antiinflammatory properties of cytochrome P450 epoxygenase-derived eicosanoids. Science, 285: 1276–1279. Node, K., Ruan, X.L., Dai, J., Yang, S.X., Graham, L., Zeldin, D.C., Liao, J.K. (2001) Activation of G{alpha}s mediates induction of tissue-type plasminogen activator gene transcription by epoxyeicosatrienoic acids. J. Biol. Chem., 276: 15983–15989. Novikova, E.B. (1972) Autoregulation in the coronary system. Fiziol. Zh. SSSR, 58: 61–72. Nystedt, S., Emilsson, K., Wahlestedt, C., Sundelin, J. (1994) Molecular cloning of a potential proteinase activated receptor [see comments]. Proc. Natl. Acad. Sci. USA, 91: 9208–9212. O’Brien, W.J., Lingrel, J.B., Wallick, E.T. (1994) Ouabain binding kinetics of the rat alpha two and alpha three isoforms of the sodium-potassium adenosine triphosphatase. Arch. Biochem. Biophys., 310: 32–39. Ogawa, K., Tanaka, S., Murray, P.A. (2001) Inhibitory effects of etomidate and ketamine on endothelium-dependent relaxation in canine pulmonary artery. Anesthesiology, 94: 668–677. Ohashi, M., Satoh, K., Itoh, T. (1999) Acetylcholine-induced membrane potential changes in endothelial cells of rabbit aortic valve. Br. J. Pharmacol., 126: 19–26. Ohhashi, T., Takahashi, N. (1991) Acetylcholine-induced release of endothelium-derived relaxing factor from lymphatic endothelial cells. Am. J. Physiol., 260: H1172–H1178. Oltman, C.L., Weintraub, N.L., VanRollins, M., Dellsperger, K.C. (1998) Epoxyeicosatrienoic acids and dihydroxyeicosatrienoic acids are potent vasodilators in the canine coronary microcirculation. Circ. Res., 83: 932–939. Onaka, U., Fujii, K., Abe, I., Fujishima, M. (1998) Antihypertensive treatment improves endotheliumdependent hyperpolarization in the mesenteric artery of spontaneously hypertensive rats. Circulation, 98: 175–182. O’Reilly, D.R., Miller, L.K., Luckow, V.A. (1992) In, Baculovirus Expression Vectors. A Laboratory Manual., W.H. Freeman and Co., New York. Ortiz de Montellano, P.R., Mico, B.A., Matthews, J.M., Kunze, K.L., Miwa, G.T., Lu, A.Y.H. (1981) Selective inactivation of cytochrome P-450 isozymes by suicide substrates. Arch. Biochem. Biophys., 210: 717–729. Oyekan, A.O., McGiff, J.C. (1998) Functional response of the rat kidney to inhibition of nitric oxide synthesis: role of cytochrome P450-derived arachidonate metabolites. Br. J. Pharmacol., 125: 1065–1073. Pacicca, C., Von der Weid, P.-Y., Bény, J.-L. (1992) Effect of Nitro-L-Arginine on endotheliumdependent hyperpolarizations and relaxations of pig coronary arteries. J. Physiol., 457: 247–256. Palmer, R.M.J., Ferrige, A.G., Moncada, S. (1987) Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature, 327: 524–526. Pannirselvam, M., Verma, S., Anderson, T.J., Triggle, C.R. (2002) Cellular basis of endothelial dysfunction in small mesenteric arteries from spontaneously diabetic (db/db ⫺/⫺) mice: role of decreased tetrahydrobiopterin bioavailability. Br. J. Pharmacol., 136: 255–263. Panza, J.A., Quyyumi, A.A., Brush, Jr. J.E., Epstein, S.E. (1990) Abnormal endothelium-dependent vascular relaxation in patients with essential hypertension. N. Engl. J. Med., 323: 22–27. Papassotiriou, J., Kohler, B., Prenen, J., Krause, H., Akbar, M., Eggermont, J., Paul, M., Distler, A., Nilius, B., Hover, J. (2000) Endothelial K⫹ channel lacks the Ca2⫹ sensitivity-regulating ␤ subunit. FASEB J., 14: 885–894. Park, J.B., Touyz, R.M., Chen, X., Schiffrin, E.L. (2002) Chronic treatment with a superoxide dismutase mimetic prevents vascular remodelling and progression of hypertension in salt-loaded stroke-prone spontaneously hypertensive rats. Am. J. Hypertension, 15: 78–84.

404

References

Parkes, D.G., Vaughan J., Rivier J., Vale, W., May, C.N. (1997) Cardiac inotropic actions of urocortin in conscious sheep. Am. J. Physiol., 272: H2115–H2122. Parkington, H.C., Tare, M., Tonta, M.A., Coleman, H.A. (1993) Stretch reveals three components in the hyperpolarization of guinea-pig coronary artery in response to acetylcholine. J. Physiol., 465: 459–476. Parkington, H.C., Tonta, M.A., Coleman, H.A., Tare, M. (1995) Role of membrane potential in endothelium-dependent relaxation of guinea-pig coronary arterial smooth muscle. J. Physiol., 484: 469–480. Parkington, H.C., Chow, J.M., Evans, R.G., Coleman, H.A., Tare, M. (2002) Role for endotheliumderived hyperpolarizing factor in vascular tone in rat mesenteric and hindlimb circulations in vivo. J. Physiol., 542: 929–937. Pascoal, I.F., Lindheimer, M.D., Nalbantian-Brandt, C., Umans, J.G. (1998) Preeclampsia selectively impairs endothelium-dependent relaxation and leads to oscillatory activity in small omental arteries. J. Clin. Invest., 101: 464–470. Pedarzani, P., Mosbacher, J., Rivard, A., Cingolani, L.A., Oliver, D., Stocker, M., Adelman, J.P., Fakler, B. (2001) Control of electrical activity in central neurons by modulating the gating of small conductance Ca2⫹-activated K⫹ channels. J. Biol. Chem., 276: 9762–9769. Peng, M., Huang, L., Huang, W.-H., Askari, A. (1996) Partial inhibition of Na K ATPase by ouabain induces the Ca dependent expressions of early response genes in cardiac myocytes. J. Biol. Chem., 271: 10372–10378. Perez, G.J., Bonev, A.D., Patlak, J.B., Nelson, M.T. (1999) Functional coupling of ryanodine receptors to KCa channels in smooth muscle cells from rat cerebral arteries. J. Gen. Physiol., 113: 229–238. Perez, G.J., Bonev, A.D., Nelson, M.T. (2001) Micromolar Ca2⫹ from sparks activates Ca2⫹-sensitive K⫹ channels in rat cerebral artery smooth muscle. Am. J. Physiol., 281: C1769–C1775. Perkins G.A., Goodenough, D.A., Sosinsky, G.E. (1998) Formation of the gap junction intercellular channel requires a 30 degree rotation for interdigitating two apposing connexons. J. Mol. Biol., 277: 171–177. Petersson, J., Zygmunt, P.M., Högestätt, E.D. (1997) Characterization of the potassium channels involved in EDHF-mediated relaxation in cerebral arteries. Br. J. Pharmacol., 120: 1344–1350. Petkova-Kirova, P.S., Gagov, H.S., Duridanova, D.B. (2000) Urocortin hyperpolarizes stomach smooth muscle via activation of Ca2⫹-sensitive K⫹ currents. J. Muscle Res. Cell Motil., 21: 639–645. Pfister, S.L., Spitzbarth, N., Edgemond, W., Campbell, W.B. (1996) Vasorelaxation by an endotheliumderived metabolite of arachidonic acid. Am. J. Physiol., 270: H1021–H1030. Picciano, P.T., Johnson, B., Walenga, R.W., Donovan, M., Borman, B.J., Douglas, W.H., Kreutzer, D.L. (1984) Effects of D-valine on pulmonary artery endothelial cell morphology and function in cell culture. Exp. Cell Res., 151: 134–147. Pinto, A., Abraham, N.G., Mullane, K.M. (1987) Arachidonic acid-induced endothelial-dependent relaxation of canine coronary arteries: contribution of a cytochrome P-450-dependent pathway. J. Pharmacol. Exp. Ther., 240: 856–863. Piper, A.S., Greenwood, I.A., Large, W.A. (2002) Dual effect of blocking agents on Ca2⫹-activated Cl⫺ currents in rabbit pulmonary artery smooth muscle cells. J. Physiol., 539: 119–131. Pipili, E. (1985) Platelet membrane potential: simultaneous measurement of diSC3(5) fluorescence and optical density. Thromb. Haemost., 54: 645–649. Placer, Z.A., Cushman, L.L., Johnson, B.C. (1966) Estimation of product of lipid peroxydation (malondialdehyde) in biochemical systems. Anal. Biochem., 16: 359–364. Plane, F., Holland, M., Waldron, G.J., Garland, C.J., Boyle, J.P. (1997) Evidence that anandamide and EDHF act via different mechanisms in rat isolated mesenteric arteries. Br. J. Pharmacol., 121: 1509–1511. Pluger, S., Faulhaber, J., Furstenau, M., Lohn, M., Waldschutz, R., Gollasch, M., Haller, H., Luft, F.C., Ehmke, H., Pongs, O. (2000) Mice with disrupted BK channel beta1 subunit gene feature abnormal Ca2⫹ spark/STOC coupling and elevated blood pressure. Circ. Res., 87: E53–E60.

References

405

Pomposiello, S., Yang, X.-P., Liu, Y.-H., Surakanti, M., Rhaleb, N.-E., Sevilla, M., Carretero, O.A. (1997) Autacoids mediate coronary vasoconstriction induced by nitric oxide synthesis inhibition. J. Cardiovasc. Pharmacol., 30: 599–606. Popp, R., Bauersachs, J., Hecker, M., Fleming, I., Busse, R. (1996) A transferable, ␤-naphthoflavoneinducible, hyperpolarizing factor is synthesized by native and cultured porcine coronary endothelial cells. J. Physiol., 497: 699–709. Popp, R., Fleming, I., Busse, R. (1998) Pulsatile stretch in coronary arteries elicits release of endothelium-derived hyperpolarizing factor: a modulator of arterial compliance. Circ. Res., 82: 696–703. Porter, V.A., Bonev, A.D., Knot, H.J., Heppner, T.J., Stevenson, A.S., Kleppisch, T., Lederer, W.J., Nelson, M.T. (1998) Frequency modulation of Ca2⫹ sparks is involved in regulation of arterial diameter by cyclic nucleotides. Am. J. Physiol., 274: C1346–C1355. Potente, M., Michaelis,U.R., Fisslthaler, B., Busse, R., Fleming, I. (2002) Cytochrome P450 2C9induced endothelial cell proliferation involves induction of mitogen-activated protein (MAP) kinase phosphatase-1, inhibition of the c-Jun N-terminal kinase, and up-regulation of cyclin D1. J. Biol Chem., 277: 15671–15676. Pratt, P.F., Rosolowsky, M., Campbell, W.B. (1996) Mediators of arachidonic acid-induced relaxation of bovine coronary artery. Hypertension, 28: 76–82. Prieto, D., Simonsen, U., Hernandez, M., Garcia-Sacristan, A. (1998) Contribution of K⫹ channels and ouabain-sensitive mechanisms to the endothelium-dependent relaxations of horse penile small arteries. Br. J. Pharmacol., 123: 1609–20. Prior, H.M., Webster, N., Quinn, K., Beech, D.J., Yates, M.S. (1998) K⫹ induced dilation of a small renal artery: no role for inward rectifier K⫹ channels. Cardiovasc. Res., 37: 780–790. Pu, H.X., Cluzeaud, F., Goldshleger, R., Karlish, S.J.D., Farman N., Blostein, R. (2001) Functional role and immunological localization of the ␥a and ␥b forms of the Na,K-ATPase ␥ subunit. J. Biol. Chem., 276: 20370–20378. Puntarulo, S., Cederbaum, A.I. (1998) Production of reactive oxygen species by microsomes enriched in specific human cytochrome P450 enzymes. Free Rad. Biol. Med., 24: 1324–1330. Purnell, M.R., Whish, W.J. (1980) Novel inhibitors of poly(ADP-ribose) synthetase. Biochem, J., 185: 775–777. Quayle, J.M., McCarron, J.G., Brayden, J.E., Nelson, M.T. (1993) Inward rectifier K⫹ currents in smooth muscle cells from rat resistance-sized cerebral arteries. Am. J. Physiol., 265: C1363–C1370. Quayle, J.M., Nelson, M.T., Standen, N.B. (1997) ATP-sensitive and inwardly rectifying potassium channels in smooth muscle. Physiol. Rev., 77: 1165–1232. Quignard, J.F., Félétou, M., Thollon, C., Vilaine, J.P., Duhault, J., Vanhoutte, P.M. (1999) Potassium ions and endothelium-derived hyperpolarizing factor in guinea-pig carotid and porcine coronary arteries. Br. J. Pharmacol., 127: 27–34. Quignard, J-F., Félétou, M., Edwards, G., Duhault, J., Weston, A.H., Vanhoutte, P.M. (2000) Role of endothelial cell hyperpolarization in EDHF-mediated responses in the guinea-pig carotid artery. Br. J. Pharmacol., 29: 1103–1112. Quilley, J., McGiff, J.C. (2000) Is EDHF an epoxyeicosatrienoic acid? Trends Pharmacol. Sci., 21: 121–124. Quilley, J., Fulton, D., McGiff, J.C. (1997) Hyperpolarizing factors. Biochem. Pharmacol., 54: 1059–1070. Rakic, P. (1985) Limits of neurogenesis in primates. Science, 227: 1054–1056. Randall, M.D., Kendall, D.A. (1998) Endocannabinoids: a new class of vasoactive substances. Trends Pharmacol. Sci., 19: 55–58. Randall, M.D., Alexander, S.P.H., Bennett, T., Boyd, E.A., Fry, J.R., Gardiner, S.M., Kemp, P.A., McCulloch, A.I., Kendall, D.A. (1996) An endogenous cannabinoid as an endothelium-derived vasorelaxant. Biochem. & Biophys. Res. Commun., 229(1): 114–120. Rapoport, R.M., Murad, F. (1983) Effect of ouabain and alterations in potassium concentration on relaxation induced by sodium nitroprusside. Blood Vessels, 20: 255–264. Rayner, S.E., Van Helden, D.F. (1997) Evidence that the substance P-induced enhancement of pacemaking in lymphatics of the guinea-pig mesentery occurs through endothelial release of thromboxane A2. Br. J. Pharmacol., 121: 1589–1596.

406

References

Recchia, F.A., Senzaki, H., Saeki, A., Byrne, B.J., Kass, D.A. (1996) Pulse pressure-related changes in coronary flow in vivo are modulated by nitric oxide and adenosine. Circ. Res., 79: 849–856. Reeder, L.B., Yang, L.H., Ferguson, M.K. (1994) Modulation of lymphatic spontaneous contractions by EDRF. J. Surg. Res., 56: 620–625. Rees, D.D., Palmer, R.M.J., Moncada, S. (1989) Role of endothelium-derived nitric oxide in the regulation of blood pressure. Proc. Natl. Acad. Sci. USA, 86: 3375–3378. Regoli, D., D’Orleans-Juste, P., Escher, E., Mizrahi, J. (1984) Receptors for substance P. I. The pharmacological preparations. Eur. J. Pharmacol., 97: 161–170. Renaud, S., de Lorgeril, M. (1992) Wine, alcohol, platelets, and the French paradox for coronary heart disease. Lancet, 339: 1523–1526. Richards, G.R., Weston, A.H., Burnham, M.P., Félétou, M., Vanhoutte, P.M., Edwards, G. (2001) Suppression of K⫹-induced hyperpolarization by phenylephrine in rat mesenteric artery: relevance to studies of endothelium-derived hyperpolarizing factor. Br. J. Pharmacol., 134: 1–5. Ridge, K.M., Dada, L., Lecuona, E., Bertorello, A., Katz, A., Mochly-Rosen, D., Sznajder, J. (2002) Dopamine induced exocytosis of NaKATPase is dependent on activation of protein kinase C-␧ and –␦;. Mol. Biol. Cell., 13: 1381–1389. Rittenhouse, A.R., Parker, C., Brugnara, C., Morgan, K.G., Alper, S.L. (1997a) Inhibition of maxi-K currents in ferret portal vein smooth muscle cells by the antifungal clotrimazole. Am. J. Physiol., 273: C45–C56. Rittenhouse, A.R., Vandorpe, D.H., Brugnara, C., Alper, S.L. (1997b) The antifungal imidazole clotrimazole and its major in vivo metabolite are potent blockers of the calcium-activated potassium channel in murine erythroleukemia cells. J. Membr. Biol., 157: 177–191. Rivers, R. (1997) Conducted arteriolar dilations persist in the presence of nitroarginine. J. Cardiovasc. Pharmacol., 30: 309–312. Ro, S., Hatton, W.J., Koh, S.D., Horowitz, B. (2001) Molecular properties of small-conductance Ca2⫹activated K⫹ channels expressed in murine smooth muscle. Am. J. Physiol., 281: G964–G973. Roberts, J.M., Taylor, R.N., Goldfien, A. (1991) Endothelial-cell activation as a pathogenetic factor in preeclampsia. Semin Perinatol., 15: 86–93. Robertson, B.E., Schubert, R., Hescheler, J., Nelson, M.T. (1993) cGMP-dependent protein kinase activates Ca-activated K channels in cerebral artery smooth muscle cells. Am. J. Physiol., 265: C299–C303. Robertson, B.E., Bonev, A.D., Nelson, M.T. (1996) Inward rectifier K⫹ currents in smooth muscle cells from rat coronary arteries: block by Mg2⫹, Ca2⫹, and Ba2⫹. Am. J. Physiol., 271: H696–H705. Rosolowsky, M., Campbell, W.B. (1993) Role of PGI2 and epoxyeicosatrienoic acids in relaxation of bovine coronary arteries to arachidonic acid. Am. J. Physiol., 264(2 Pt 2): H327–H335. Rosolowsky, M., Campbell, W.B. (1996) Synthesis of hydroxyeicosatetraenoic (HETEs) and epoxyeicosatrienoic acids (EETs) by cultured bovine coronary endothelial cells. Biochem. Biophys. Acta., 1299: 267–277. Roy, S., Saifeddine, M., Triggle, C.R., Loutzenhiser, R., Hollenberg, M.D. (1997) Endotheliumdependent contractile effect of trypsin-activated receptor (PAR2) stimulation in rat vascular tissue. Proc. West. Pharmacol. Soc., 40: 53–55. Rubanyi, G.M., Vanhoutte, P.M. (1986) Oxygen-derived free radicals, endothelium, and responsiveness of vascular smooth muscle. Am. J. Physiol., 250: H815–H821. Rubanyi, G.M., Vanhoutte, P.M. (1987) Nature of endothelium-derived relaxing factor: are there two relaxing mediators?. Circ. Res., 61(5 Pt 2): II61–II67. Rubanyi, G.M., Lorenz, R.R., Vanhoutte, P.M. (1985) Bioassay of endothelium-derived relaxing factor(s): inactivation by catecholamines. Amer. J. Physiol., 249(1 Pt 2): H95–101. Rubanyi, G.M., Romero, J.C., Vanhoutte, P.M. (1986) Flow-induced release of endothelium-derived relaxing factor. Am. J. Physiol., 250: H1145–H1149. Ruiz, E., Tejerina, T. (1998) Relaxant effects of L-citrulline in rabbit vascular smooth muscle. Br. J. Pharmacol., 125(1): 186–192. Ruiz-Marcos, F.M., Ortiz, M.C., Fortepiani, L.A., Nadal, F.J., Atucha, N.M., Garcia-Estan, J. (2001) Mechanisms of the increased pressor response to vasopressors in the mesenteric bed of nitric oxidedeficient hypertensive rats. Eur. J. Pharmacol., 412: 273–279.

References

407

Rummery, N., McKenzie, K., Whitworth, J.A., Hill, C.E. (2002) Decreased endothelial size and connexin expression in rat caudal arteries during hypertension. J. Hypert., 20: 247–253. Ruppersberg, J.P. (2000) Intracellular regulation of inward rectifier K⫹ channels. Pflügers Arch., 441: 1–11. Rusch, N.J., De Lucena, R.G., Wooldridge, T.A., England, S.K., Cowley, A.W.J. (1992) A Ca2⫹-dependent K⫹ current is enhanced in arterial membranes of hypertensive rats. Hypertension, 19: 301–307. Sage, S.O., Marchenko, S.M. (2001) Single-channel properties of Ca2⫹-activated K⫹ channels in the vascular endothelium. In Potassium Channels in Cardiovascular Biology, edited by S.L. Archer and N.J. Rusch, pp. 651–666. New York, NY: Kluwer Academic/Plenum Publishers. Saifeddine, M., Al Ani, B., Cheng, C.H., Wang, L., Hollenberg, M.D. (1996) Rat proteinase-activated receptor-2 (PAR-2): cDNA sequence and activity of receptor-derived peptides in gastric and vascular tissue. Br. J. Pharmacol., 118: 521–530. Sakai, T. (1990) Acetylcholine induces Ca-dependent K currents in rabbit endothelial cells. Jpn. J. Pharmacol., 53: 235–246. Salter, K.J., Kozlowski, R.Z. (1996) Endothelin receptor coupling to potassium and chloride channels in isolated rat pulmonary arterial myocytes. J. Pharmacol. Exp. Ther., 279: 1053–1062. Sandow, S.L., Hill, C.E. (2000) Incidence of myoendothelial gap junctions in the proximal and distal mesenteric arteries of the rat is suggestive of a role in endothelium-derived hyperpolarizing factormediated responses. Circ. Res., 86: 341–346. Sandow, S.L., Tare, M., Coleman, H.A., Hill, C.E., Parkington, H.C. (2002) Involvement of myoendothelial gap junctions in the actions of endothelium-derived hyperpolarizing factor. Circ. Res., 90: 1108–1113. Sandow, S.L., Bramich, N.J., Bandi, H.P., Rummery, N., Hill, C.E., (2003) Structure, function and EDHF in the caudal artery of the SHR and WKY rat. Arterio. Thromb. Vasc. Biol., 23: 822–828. Santicioli, P., Maggi, C.A. (2000) Effect of 18␤-glycyrrhetinic acid on electromechanical coupling in the guinea-pig renal pelvis and ureter. Br. J. Pharmacol., 129: 163–169. Schafer, F.Q., Buettner, G.R. (2001) Redox environment of the cell as viewed through the redox-state of the glutathione disulfide/glutathione couple. Free Radic. Biol. Med., 30: 1191–1212. Schiffrin, E.L., Deng, L.Y. (1995) Comparison of effects of angiotensin I-converting enzyme inhibition and ␤-blockade for 2 years on function of small arteries from hypertensive patients. Hypertension, 25(pt 2): 699–703. Schiffrin, E.L., Park, J.B., Intengan, H.D., Touyz, R.M. (2000) Correction of arterial structure and endothelial dysfunction in human essential hypertension by the angiotensin receptor antagonist losartan. Circulation, 101: 1653–1659. Schildmeyer, L.A., Bryan, R.M., Jr (2002) Effect of NO on EDHF response in rat middle cerebral arteries. Am. J. Physiol. Heart Circ. Physiol., 282: H734–H738. Schilling, W.P., Elliot, S.J. (1992) Ca2⫹ signaling mechanisms of vascular endothelial cells and their role in oxidant-induced endothelial cell dysfunction. Am. J. Physiol., 262: H1617–H1630. Schnackenberg, C.G., Wilcox, C.S. (1999) Two-week administration of tempol attenuates both hypertension and renal excretion of 8-iso prostaglandin F2a. Hypertension, 33: 424–428. Schnackenberg, C.G., Welch, W.J., Wilcox, C.S. (1998) Normalization of blood pressure and renal vascular resistance in SHR with a membrane-permeable superoxide dismutase mimetic: role of nitric oxide. Hypertension, 32: 59–64. Schubert, L.-L. Schubert, W., Spray, D.C., Lisanti, M. (2002) Connexin family members target to lipid raft domains and interaction with Caveolin-1. Biochem., 41, 5754–5764. Scornik, F.S., Toro, L. (1992) U46619, a thromboxane A2 agonist, inhibits KCa channel activity from pig coronary artery. Am. J. Physiol., 262: C708–C713. Segal, S.S., Beny, J.-L. (1992) Intracellular recording and dye transfer in arterioles during blood flow control. Am. J. Physiol., 263: H1–H7. Segal, S.S., Duling, B.R. (1986) Flow control among microvessels coordinated by intercellular conduction. Science, 234: 868–870. Segal, S.S., Welsh, D.G., Kurjiaka, D.T. (1999) Spread of vasodilatation and vasoconstriction along feed arteries and arterioles of hamster skeletal muscle. J. Physiol., 516: 283–291.

408

References

Seki, K., Hirai, A., Noda, M., Tamura, Y., Kato, I., Yoshida, S. (1992) Epoxyeicosatrienoic acid stimulates ADP-ribosylation of a 52KDa protein in rat liver cytosol. Biochem. J., 281: 185–190. Serafini, M., Maiani, G., Ferro-Luzzi, A. (1998) Alcohol-free red wine enhances plasma antioxidant capacity in humans. J. Nutr., 128: 1003–1007. Severinghaus, J.W. (1995) Hypothetical roles of angiogenesis, osmotic swelling, and ischemia in highaltitude cerebral edema. J. Appl. Physiol., 79: 375–379. Severs, N.J. (1999) Cardiovascular disease. Novartis Foundation Symposium, 219: 188–211. Shah, M., Haylett, D.G. (2000) The pharmacology of hSK1 Ca2⫹-activated K⫹ channels expressed in mammalian cell lines. Br. J. Pharmacol., 129: 627–630. Shimoda, L.A., Sylvester, J.T., Booth, G.M., Shimoda, T.H., Meeker, S., Undem, B.J., Sham, J.S.K. (2001) Inhibition of voltage-gated K⫹ currents by endothelin-1 in human pulmonary arterial myocytes. Am. J. Physiol., 281: L1115–L1122. Shimokawa, H. (1999) Primary endothelial dysfunction: atherosclerosis. J. Mol. Cell Cardiol., 31(1): 23–37. Shimokawa, H., Urakami-Harasawa, L. (1999) Importance of endothelium-derived hyperpolarizing factor in human arteries. In Endothelium-derived hyperpolarizing factor, edited by P.M. Vanhoutte, pp. 391–398. Amsterdam: Harwood Academic Publishers. Shimokawa, H., Yasutake, H., Fujii, K., Owada M.K., Nakaike, R., Fukumoto, Y., Takayanagi, T., Nagao, T., Egashira, K., Fujishima, M., Takeshita, A. (1996) The importance of the hyperpolarizing mechanism increases as the vessel size decreases in endothelium-dependent relaxations in rat mesenteric circulation. J. Cardiovasc. Pharmacol., 28: 703–711. Shirahase, H., Kanda, M., Kurahashi, K., Nakamura, S., Usui, H., Shimizu, Y. (1995) Endotheliumdependent contraction in intrapulmonary arteries: mediation by endothelial NK1 receptors and TXA2. Br. J. Pharmacol., 115: 1215–1220. Singer, H.A., Saye, J.A., Peach, M.J. (1984) Effects of cytochrome P-450 inhibitors on endotheliumdependent relaxation in rabbit aorta. Blood Vessels, 21: 223–230. Skarsgard, P., van Breemen, C., Laher, I. (1997) Estrogen regulates myogenic tone in pressurized cerebral arteries by enhanced basal release of nitric oxide. Am. J. Physiol., 273: H2248–H2256. Smets, L.A., Loesberg, C., Janssen, M., Van Rooij, H. (1990) Intracellular inhibition of mono(ADPribosylation) by meta-iodobenzylguanidine: specificity, intracellular concentration and effects on glucocorticoid-mediated cell Lysis. Biochim. Biophys. Acta., 1054: 49–55. Smith, R.O. (1949) Lymphatic contractility: a possible mechanism of lymphatic vessels for the transport of lymph. J. Exp. Med., 90: 497–509. Sneyd, J., Wilkins, M., Strahonja, A., Sanderson, M.J. (1998) Calcium waves and oscillations driven by an intercellular gradient of inositol(1,4,5)-trisphosphate. Biophys. Chem., 72: 101–109. Sobey, C.G., Cocks, T.M. (1998) Activation of protease-activated receptor-2 (PAR-2) elicits nitric oxide-dependent dilatation of the basilar artery in vivo. Stroke, 29: 1439–1444. Sobey, C.G., Moffatt, J.D., Cocks, T.M. (1999) Evidence for selective effects of chronic hypertension on cerebral artery vasodilatation to protease-activated receptor-2 activation. Stroke, 30: 1933–1940. Sohn, H.Y., Gloe, T., Keller, M., Schoenafinger, K., Pohl, U. (1999) Sensitive superoxide detection in vascular cells by the new chemiluminescence dye L-012. J. Vasc. Res., 36: 456–464. Sohn, H.-Y., Keller, M., Gloe, T., Morawietz, H., Rueckschloss, U., Pohl, U. (2000) The small G-protein Rac mediates depolarization-induced superoxide formation in human endothelial cells. J. Biol. Chem., 275: 18745–18750. Sollini, M., Frieden, M., Bény, J-L. (2002) Charybdotoxin-sensitive small conductance Kca channel activated by bradykinin and substance P in endothelial cells. Br. J. Pharmacol., 136: 1201–1209 Solodkov, A.P., Belyaeva, L.Eu., Shebeko, V.I. (2001) Coronary vessels autoregulation following acute hemorrhage and its combination with preexisting immobilization stress. Russian J. Physiol., 87: 1250–1259. Som, S., Raha, C., Chatterjee, I.B. (1983) Ascorbic acid: a scavenger of superoxide radical. Acta Vitaminol. Enzymol., 5: 243–250.

References

409

Somers, M.J., Harrison, D.G. (1999) Reactive oxygen species and the control of vasomotor tone. Curr. Hypertens. Rep., 1(1): 102–108. Somers, M.J., Mavromatis, K., Galis, Z.S., Harrison, D.G. (2000) Vascular superoxide production and vasomotor function in hypertension induced by deoxycorticosterone acetate-salt. Circulation, 101: 1722–1728. Spagnoli, L.G., Villaschi, S., Neri, L., Palmieri, G. (1982) Gap junctions in myo-endothelial bridges of rabbit carotid arteries. Experientia, 38(1): 124–125. Stampfer, M.J., Colditz, G.A., Willett, W.C., Manson, J.E., Rosner B, Speizer, F.E., Hennekens, C.H. (1991) Postmenopausal estrogen therapy and cardiovascular disease. Ten-year follow-up from the nurses health study. N. Engl. J. Med., 325: 756–762. Standen, N.B., Quayle, J.M. (1998) K⫹ channel modulation in arterial smooth muscle. Acta. Physiol. Scand., 164: 549–557. Stoclet, J.C., Kleschyov, A., Andriambeloson, E., Diebolt, M., Andriantsitohaina, R. (1999) Endothelial NO release caused by red wine polyphenols. J. Physiol. Pharmacol., 50: 535–540. Stonehouse, A.H., Pringle, J.H., Norman, R.I., Stanfield, P.R., Conley, E.C., Brammar, W.J. (1999) Characterisation of Kir2.0 proteins in the rat cerebellum and hippocampus by polyclonal antibodies. Histochem. Cell Biol., 112: 457–465. Strøbæk, D., Jø´rgensen, T.D., Christophersen, P., Ahring, P.K., Olesen, S.P. (2000) Pharmacological characterization of small-conductance Ca2⫹-activated K⫹ channels stably expressed in HEK 293 cells. Br. J. Pharmacol., 129: 991–999. Sudhir, K., Jennings, G.L., Funder, J.W., Komesaroff, P.A. (1996) Estrogen enhances basal nitric oxide release in the forearm vasculature in perimenopausal women. Hypertension, 28: 330–334. Sullivan, J.C., Pollock, D.M., Pollock, J.S. (2002) Altered nitric oxide synthase 3 distribution in mesenteric arteries of hypertensive rats. Hypertension, 39: 597–602. Sun, C.W., Alonsogalicia, M., Taheri, M.R., Falck, J.R., Harder, D.R., Roman, R.J. (1998) Nitric oxide20-hydroxyeicosatetraenoic acid interaction in the regulation of K⫹ channel activity and vascular tone in renal arterioles. Circ. Res., 83: 1069–1079. Sunano, S., Watanabe, H., Tanaka, S., Sekiguchi, F., Shimamura, K. (1999) Endothelium-derived relaxing, contracting and hyperpolarizing factors of mesenteric arteries of hypertensive and normotensive rats. Br. J. Pharmacol., 126: 709–716. Sweadner, K.J. (1989) Isozymes of the Na⫹/K⫹-ATPase. Biochim. Biophys. Acta, 988: 185–220. Sweadner, K.J., Rael, K. (2000) The FXYD gene family of small ion transport regulators or channels: cDNA sequence, protein signature sequence, and expression. Genomics, 68: 41–56. Swei, A., Lacy, F., Delano, F.A., Parks, D.A., Schmid-Schonbein, G.W. (1999) A mechanism of oxygen free radical production in the Dahl hypertensive rat. Microcirculation, 6: 179–187. Syme, C.A., Gerlach, A.C., Singh, A.K., Devor, D.C. (2000) Pharmacological activation of cloned intermediate- and small-conductance Ca2⫹-activated K⫹ channels. Am. J. Physiol., 278: C570–C581. Taddei, S., Virdis, A., Mattei, P., Ghiadoni, L., Gennari, A., Fasolo, C.B., Sudano, I., Salvetti, A. (1995) Aging and endothelial function in normotensive subjects and patients with essential hypertension. Circulation, 91: 1981–1987. Taddei, S., Ghiadoni, L., Virdis, A., Buralli, S., Salvetti, A. (1999) Vasodilation to bradykinin is mediated by an ouabain-sensitive pathway as a compensatory mechanism for impaired nitric oxide availability in essential hypertensive patients. Circulation, 100: 1400–1405. Taenaka, N., Shimada, Y. et al. (1982) New approach to regional anticoagulation in hemodialysis using gabexate mesilate (FOY). Crit. Care. Med., 10(11): 773–775. Tagawa, H., Shimokawa, H. (1999) Long-term treatment with eicosapentaenoic acid augments both nitric oxide-mediated and non-nitric oxide-mediated endothelium-dependent forearm vasodilatation in patients with coronary artery disease. J. Cardiovasc. Pharmacol., 33(4): 633–640. Takahashi, K., Nammour, T.M., Fukunaga, M., Ebert, J., Morrow, J.D., Roberts, L.J. II., Hoover, R.L., Badr, K.F. (1992) Glomerular actions of a free radical-generated novel prostaglandin 8-epi-prostaglandin F2a, in the rat: evidence for interaction with thromboxane A2 receptors. J. Clin. Invest., 90: 136–141.

410

References

Takamura, Y., Shimokawa, H. (1999) Important role of endothelium-derived hyperpolarizing factor in shear stress – induced endothelium-dependent relaxations in the rat mesenteric artery. J. Cardiovasc. Pharmacol., 34(3): 381–387. Tamura, K., Manabe, T. et al. (1992) Effect of synthetic protease inhibitors on superoxide (O2⫺), hydrogen peroxide (H2O2) and hydroxyl radical, production by human polymorphonuclear leukocytes. Hepatogastroenterology, 39(1): 59–61. Tanaka, Y., Meera, P., Song, M., Knaus, H.-G., Toro, L. (1997) Molecular constituents of maxi KCa channels in human coronary smooth muscle: predominant ␣ ⫹ ␤ subunit complexes. J. Physiol., 502: 545–557. Taniguchi, K., Kaya, S. (eds) (2000) Na/K ATPase and related ATPases, Elsevier. Tare, M., Parkington, H.C., Coleman, H.A., Neild, T.O., Dusting, G.J. (1990) Hyperpolarization and relaxation of arterial smooth muscle caused by nitric oxide derived from the endothelium. Nature, 346: 69–71. Tare, M., Coleman, H.A., Parkington, H.C. (2002) Glycyrrhetinic derivatives inhibit hyperpolarization in endothelial cells of guinea pig and rat arteries. Am. J. Physiol., 282: H335–H341. Taylor, S.G., Weston, A.H. (1988) Endothelium-derived hyperpolarizing factor: a new endogenous inhibitor from the vascular endothelium. Trends Pharmacol. Sci., 9: 272–274. Taylor, P.D., Graves, J.E., Poston, L. (1995) Selective impairment of acetylcholine-mediated endotheliumdependent relaxation in isolated resistance arteries of the streptozotocin-induced diabetic rat. Clin. Sci. (Lond), 88: 519–524. Taylor, H.J., Chaytor, A.T., Evans, W.H., Griffith, T.M. (1998) Inhibition of the gap junctional component of endothelium-dependent relaxations in rabbit iliac artery by 18-alpha glycyrrhetinic acid. Br. J. Pharmacol., 125: 1–3. Taylor, H.J., Chaytor, A.T., Edwards, D.H., Griffith, T.M. (2001) Gap junction-dependent increases in smooth muscle cAMP underpin the EDHF phenomenon in rabbit arteries. Biochem. Biophys. Res. Commun., 283: 583–589. Terasawa, T., Okada, T., Hara, T., Itoh, K. (1992) Glycyrrhetinic acid derivatives as potent inhibitors of Na⫹,K⫹-ATPase. Synthesis and structure-activity relationships. Eur. J. Med. Chem., 27: 345–351. Terui, K., Higashiyama, A., Horiba, N., Furukawa, K.I., Motomura, S., Suda, T. (2001) Coronary vasodilation and positive inotropism by urocortin in the isolated rat heart. J. Endocrinol., 169: 177–183. Therien, A.G., Nestor, N.B., Ball, W.J., Blostein, R. (1996) Tissue-specific versus isoform-specific differences in cation activation kinetics of the Na,K-ATPase. J. Biol. Chem., 271: 7104–7112. Therien, A.G., Karlish, S.J.D., Blostein, R. (1999) Expression and functional role of the ␥ subunit of the Na,K-ATPase in mammalian cells. J. Biol. Chem., 274: 12252–12256. Thyberg, J. (2000) Difference in caveolae dynamics in vascular smooth muscle cells of different phenotypes. Lab. Invest., 80: 915–929. Tominaga, T., Katagi, H., Ohnishi, S.T. (1998) Is Ca2⫹ -activated potassium efflux involved in the formation of ischemic brain edema? Brain Res., 460: 376–378. Tomioka, H., Hattori, Y., Fukao, M., Watanabe, H., Akaishi, Y., Sato, A., Kim, T.Q., Sakuma, I., Kitabatake, A., Kanno, M. (2001) Role of endothelial Ni2⫹-sensitive Ca2⫹ entry pathway in regulation of EDHF in porcine coronary artery. Am. J. Physiol., 280: H730–H737. Toro, L., Wallner, M., Meera, P., Tanaka, Y. (1998) Maxi-KCa, a unique member of the voltage-gated K channel superfamily. News Physiol. Sci., 13: 112–115. Torok, J. (2000) Histamine-induced relaxation in pulmonary artery of normotensive and hypertensive rats: relative contribution of prostanoids, nitric oxide and hyperpolarization. Physiol. Res., 49: 107–114. Tosun, M., Paul, R.J., Rapoport, R.M. (1998) Role of extracellular Ca2⫹ influx via L-type and non-Ltype Ca⫹⫹ channels in thromboxane A2 receptor-mediated contraction in rat aorta. J. Pharmacol. Exp. Ther., 284: 921–928. Towbin, H. Staehelin, T., Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA, 76, 4350–4354.

References

411

Treasure, C.B., Manoukian, S.V., Klein, J.L., Vita, J.A., Nabel, E.G., Renwick, G.H., Selwyn, A.P., Alexander, R.W., Ganz, P. (1992) Epicardial coronary artery responses to acetylcholine are impaired in hypertensive patients. Circ. Res., 71: 776–781. Triggle, C.R. (2001) Endothelial cell K⫹ channels, membrane potential and the release of vasoactive factors from the vascular endothelium. In Potassium Channels in Cardiovascular Biology, edited by S.L. Archer and N.J. Rusch, pp. 667–689. New York, NY: Kluwer Academic/Plenum Publishers. Triggle, C.R., Dong, H., Waldron, G.J., Cole, W.C. (1999) Endothelium-derived hyperpolarizing factor(s): species and tissue heterogenity. Clin. Exp. Pharmacol. Physiol., 26: 176–179. Trottier, G., Triggle, C.R., O’Neill, S.K., Loutzenhiser, R. (1998) Cyclic GMP-dependent and cyclic GMP-independent actions of nitric oxide on the renal afferent arteriole. Br. J. Pharmacol., 125: 563–569. Trottier, G., Hollenberg, M., Wang, X., Gui, Y., Loutzenhiser, K., Loutzenhiser, R. (2002) PAR-2 elicits afferent arteriolar vasodilation by NO-dependent and NO- independent actions. Am. J. Physiol. Renal Physiol., 282: F891–F897. Tschudi, M.R., Criscione, L., Novosel, D., Pfeiffer, K., Lüscher, T.F. (1994) Antihypertensive therapy augments endothelium-dependent relaxations in coronary arteries of spontaneously hypertensive rats. Circulation, 89: 2212–2218. Tsugorka, A., Rios, E., Blatter, L.A. (1995) Imaging elementary events of calcium release in skeletal muscle cells. Science, 269: 1723–1726. Ungvari, Z., Csiszar, A., Koller, A. (2002) Increases in endothelial Ca2⫹ activate KCa channels and elicit EDHF-type arteriolar dilation via gap juctions. Am. J. Physiol., 282: H1760–H1767. Urakami-Harasawa, L., Shimokawa, H., Nakashima, M., Egashira, K., Takeshita, A. (1997) Importance of endothelium-derived hyperpolarizing factor in human arteries. J. Clin. Invest., 100: 2793–2799. Urata, H., Nishimura, H., Ganten, D. (1996) Chymase-dependent angiotensin II forming systems in humans. Am. J. Hypertens., 9: 277–284. Ushio-Fukai, Milenski, L., Santanam, N., Becker, P., Ma, Y., Griendling, K., Alexander, R. (2001) Cholesterol depletion inhibits EGFR transactivation by angiotensin II in vascular smooth muscle cells. J. Biol. Chem., 276: 48269–48275. Vaca, L., Licea, A., Possani, L.D. (1996) Modulation of cell membrane potential in cultured vascular endothelium. Am. J. Physiol., 270: C819–C824. Vale, W., Spiess, J., Rivier, C., Rivier, J. (1981) Characterization of a 41-residue ovine hypothalamic peptide that stimulates secretion of corticotropin and ␤-endorphin. Science, 213: 1394–1397. Vallance, P., Collier, J., Moncada, S. (1989) Effect of endothelium-derived nitric oxide on peripheral arteriolar tone in man. Lancet, ii: 997–1000. Van de Voorde, J., Vanheel, B. (1997) Influence of cytochrome P-450 inhibitors on endotheliumdependent nitro-L-arginine-resistant relaxation and cromakalim-induced relaxation in rat mesenteric arteries. J. Cardiovasc. Pharmacol., 29: 827–832. Van de Voorde, J., Vanheel, B. (2000) EDHF-mediated relaxation in rat gastric small arteries: Influence of ouabain/Ba2⫹ and relation to potassium ions. J. Cardiovasc. Pharmacol., 35: 543–548. Van de Voorde, J., Vanheel, B., Leusen, I. (1992) Endothelium-dependent relaxation and hyperpolarization in aorta from control and renal hypertensive rats. Circ. Res., 70: 1–8. Van Helden, D.F. (1993) Pacemaker potentials in lymphatic smooth muscle of the guinea-pig mesentery. J. Physiol., 471: 465–479. Van Helden, D.F., von der Weid, P.-Y., Crowe, M.J. (1996) Intracellular Ca2⫹ release: a basis for electrical pacemaking in lymphatic smooth muscle. In Smooth Muscle Excitation, edited by T. Tomita, and T. B. Bolton, Academic Press, London, pp. 355–373. Vanheel, B., Van de Voorde, J. (1997) Evidence against the involvement of cytochrome P450 metabolites in endothelium-dependent hyperpolarization of the rat main mesenteric artery. J. Physiol., 501: 331–341. Vanheel, B., Van de Voorde, J. (1999) Barium decreases endothelium-dependent smooth muscle responses to transient but not to more prolonged acetylcholine applications. Pflügers Arch., 439: 123–129. Vanhoutte, P.M. (1996) Endothelial dysfunction in hypertension. J. Hypert., 14: S83–S93.

412

References

Vanhoutte, P.M. (1997) Endothelial dysfunction and atherosclerosis. Eur. Heart. J., 18(suppl E): E19–E29. Vanhoutte, P.M. (1998) Vascular biology. Old-timer makes a comeback. Nature, 396: 213, 215–16. Vanhoutte, P.M., Boulanger, C.M., Mombouli, J.V. (1995) Endothelium-derived relaxing factors and converting enzyme inhibition. Am. J. Cardiol., 76: E3–12. VanWijk, M.J., Kublickiene, K., Boer, K., Van Bavel, E. (2000) Vascular function in preeclampsia. Cardiovasc. Res., 47: 38–48. Vasquez-Vivar, J., Hogg, N., Martasek, P., Karoui, H., Pritchard, K.A. Jr, Kalyanaraman, B. (1999) Tetrahydrobiopterin-dependent inhibition of superoxide generation from neuronal nitric oxide synthase. J. Biol. Chem., 274: 26736–2642. Vergara, C., Latorre, R., Marrion, N.V., Adelman, J.P. (1998) Calcium-activated potassium channels. Curr. Opin. Neurobiol., 8: 321–329. Vogalis, F., Vincent, T., Qureshi, I., Schmalz, F., Ward, M.W., Sanders, K.M., Horowitz, B. (1996) Cloning and expression of the large-conductance Ca2⫹-activated K⫹ channel from colonic smooth muscle. Am. J. Physiol., 271: G629–G639. Von Beckerath, N., Dittrich, M., Klieber, H.-G., Daut, J. (1996) Inwardly rectifying K⫹ channels in freshly dissociated coronary endothelial cells from guinea-pig heart. J. Physiol., 491: 357–365. von der Weid, P.Y., Beny, J.L. (1992) Effect of Ca2⫹ ionophores on membrane potential of pig coronary artery endothelial cells. Am. J. Physiol., 262: H1823–H1831. von der Weid, P.-Y., Bény, J.-L. (1993) Simultaneous oscillations in the membrane potential of pig coronary artery endothelial and smooth muscle cells. J. Physiol., 471: 13–24. von der Weid, P.Y., Van Helden, D.F. (1997) Functional electrical properties of the endothelium in lymphatic vessels of the guinea-pig mesentery. J. Physiol., 504: 439–451. von der Weid, P.Y., Crowe, M.J., Van Helden, D.F. (1996) Endothelium-dependent modulation of pacemaking in lymphatic vessels of the guinea-pig mesentery. J. Physiol., 493: 563–575. von der Weid, P.-Y., Zhao, J., Van Helden, D.F. (2001) Nitric oxide decreases pacemaker activity in lymphatic vessels of guinea pig mesentery. Am. J. Physiol., 280: H2707–H2716. Vu, T.K., Hung, D.T., Wheaton, V.I., Coughlin, S.R. (1991) Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell, 64: 1057–1068. Waldron, G.J., Cole, W.C. (1999) Activation of vascular smooth muscle K⫹ channels by endotheliumderived relaxing factors. Clin. Exp. Pharmacol. Physiol., 26: 180–184. Waldron, G.J., Garland, C.J. (1994) Effect of potassium channel blockers on L-NAME insensitive relaxations in rat small mesenteric artery. Can. J. Physiol. Pharmacol., 72(S1): 115. Waldron, G.J., Ding, H., Lovren, F., Kubes, P., Triggle, C.R. (1999) Acetylcholine-induced relaxation of peripheral arteries isolated from mice lacking endothelial nitric oxide synthase. Br. J. Pharmacol., 128: 653–658. Walker, S.D., Dora, K.A., Ings, N.T., Crane, G.J., Garland, C.J. (2001) Activation of endothelial cell IK(Ca) with 1-ethyl-2-benzimidazolinone evokes smooth muscle hyperpolarization in rat isolated mesenteric artery. Br. J. Pharmacol., 134: 1548–1554. Wallis, S.J., Firth, J., Dunn, W.R. (1996) Pressure-induced myogenic responses in human isolated cerebral resistance arteries. Stroke, 27: 2287–2291. Wang, J., Juhaszova, M., Rubin, L.J., Yuan, X.-J. (1997) Hypoxia inhibits gene expression of K⫹ channel ␣ subunits in pulmonary artery smooth muscle cells. J. Clin. Invest., 100: 2347–2353. Wang, X., Loutzenhiser, R. (2002a) Determinants of the renal microvascular response to acetylcholine, afferent and efferent arteriolar actions of EDHF. Am. J. Physiol. Renal Physiol., 282: F124–F132. Wang, X., Loutzenhiser, R.D. (2002b) The response of the renal afferent arteriole to bradykinin may involve two distinct EDHFs. FASEB, 16: A826. Wang, X., van Breemen, C. (1999) Depolarization-mediated inhibition of Ca2⫹ entry in endothelial cells. Am. J. Physiol., 277: H1498–H1504. Wang, X., Chu, W., van Breemen, C. (1996) Potentiation of acetylcholine-induced responses in freshly isolated rabbit aortic endothelial cells. J. Vascul. Res., 33(5): 414–424.

References

413

Wang, Z., Huang, Y., Zou, J., Cao, K., Xu, Y., Wu, J.M. (2002) Effects of red wine and wine polyphenol resveratrol on platelet aggregation in vivo and in vitro. Int. J. Mol. Med., 9: 77–79. Wellman, G.C., Bevan, J.A. (1995) Barium inhibits the endothelium-dependent component of flow but not acetylcholine-induced relaxation in isolated rabbit cerebral arteries. J. Pharmacol. Exp. Ther., 274: 47–53. Wellman, G.C., Brayden, J.E., Nelson, M.T. (1996) A proposed mechanism for the cardioprotective effect of oestrogen in women: enhanced endothelial nitric oxide release decreases coronary artery reactivity. Clin. Exp. Pharmacol. Physiol., 23: 260–266. Wellman, G.C., Santana, L.F., Bonev, A.D., Nelson, M.T. (2001) Role of phospholamban in the modulation of arterial Ca2⫹ sparks and Ca2⫹-activated K⫹ channels by cAMP. Am. J. Physiol., 281: C1029–C1037. Welsh, D.G., Segal, S.S. (1998) Endothelial and smooth muscle cell conduction in arterioles controlling blood flow. Am. J. Physiol., 274: H178–H186. Welsh, D.G., Nelson, M.T. (2000a) A case for myoendothelial gap junctions. Circ. Res., 87: 427–428. Welsh, D.G., Segal, S.S. (2000b) Role of EDHF in conduction of vasodilation along hamster cheek pouch arterioles in vivo. Am. J. Physiol., 278: H1832–H1839. Weston, A.H., Richards, G.R., Weston, A.H., Burnham, M.P., Félétou, M., Vanhoutte, P.M., Edwards, G. (2002) K⫹-Induced hyperpolarization in rat mesenteric artery: identification, localization and role of Na⫹/K⫹-ATPases. Br. J. Pharmacol., 136: 918–926. White, R., Hiley, C.R. (2000) Hyperpolarisation of rat mesenteric endothelial cells by ATP-sensitive K(⫹) channel openers. Eur. J. Pharmacol., 397: 279–290. White, R.M., Rivera, C.O., Davison, C.A. (2000) Nitric oxide-dependent and -independent mechanisms account for gender differences in vasodilation to acetylcholine. J. Pharmacol. Exp. Ther., 292: 375–380. Widman, M.D., Weintraub, N.L., Fudge, J.L., Brooks, L.A., Dellsperger, K.C. (1998) Cytochrome P450 pathway in acetylcholine-induced canine coronary microvascular vasodilation in vivo. Am. J. Physiol., 274: H283–H289. Wigg, S.J., Tare, M., Tonta, M.A., O’Brien, R.C., Meredith, I.T., Parkington, H.C. (2001) Comparison of effects of diabetes mellitus on an EDHF-dependent and an EDHF-independent artery. Am. J. Physiol., 281: H232–H240. Wolin, M.S. (2000) Interactions of oxidants with vascular signaling systems. Arterioscler. Thromb. Vasc. Biol., 20: 1430–1442. Wong, P.Y., Lin, K.T., Yan, Y.T., Ahern, D., Iles, J., Shen, S.Y., Bhatt, R.K., Falck, J.R. (1993) 14(R),15(S)-epoxyeicosatrienoic acid (14(R),15(S)-EET) receptor in guinea pig mononuclear cell membranes. J. Lipid Mediat., 6: 199–208. Wong, P.Y-K., Lai, P-S., Falck, JR. (2000) Mechanisms and signal transduction of 14(R), 15(S)epoxyeicosatrienoic acid (14,15-EET) binding in guinea-pig monocytes. Prostaglandins Other Lipid Mediat., 62: 321–333. Wood, P.G., Gillespie, J.I. (1998) In permeabilised endothelial cells IP3-induced Ca2⫹ release is dependent on the cytoplasmic concentration of monovalent cations. Cardiovasc. Res., 37: 263–270. Xia, J., Duling, B.R. (1995) Electromechanical coupling and the conducted vasomotor response. Am. J. Physiol., 269: H2022–H2030. Xia, J., Little, T.L., Duling, B.R. (1995) Cellular pathways of the conducted electrical response in arterioles of hamster cheek pouch in vitro. Am. J. Physiol., 269: H2031–H2038. Xia, Y., Tsai, A.L., Berka, V., Zweier, J.L. (1998) Superoxide generation from endothelial nitric-oxide synthase. A Ca2⫹/calmodulin-dependent and tetrahydrobiopterin regulatory process. J. Biol. Chem., 273: 25804–25808. Xu, C., Lu, Y., Tang, G., Wang, R. (1999) Expression of voltage-dependent K⫹ channel genes in mesenteric artery smooth muscle cells. Am. J. Physiol., 277: G1055–G1063. Xu, F. Severinghaus, J.W., (1998) Rat brain VEGF expression in alveolar hypoxia: possible role in high-altitude cerebral edema. J. Appl. Physiol., 85: 53–57. Xue, D., Xu, J., McGuire, S.O., Devitre, D., Sun, G.Y. (1999) Studies on the cytosolic phospholipase A2 in immortalized astrocytes (DITNC) revealed new properties of the calcium ionophore, A23187. Neurochem. Res., 24: 1285–1291.

414

References

Yada, T., Shimokawa, H., Hiramatsu, O., Kajita, T., Shigeto, F., Goto, M., Ogasawara, Y., Kajiya, F. (2003) Hydrogen peroxide, an endogenous endothelium-derived hyperpolarizing factor, plays an important role in coronary autoregulation in vivo. Circulation, 107: 1040–1045. Yamamoto, Y., Fukuta, H., Nakahira, Y., Suzuki, H. (1998) Blockade by 18␤-glycyrrhetinic acid of intercellular electrical coupling in guinea-pig arterioles. J. Physiol., 511: 501–508. Yamamoto, Y., Imaeda, K., Suzuki, H. (1999) Endothelium-dependent hyperpolarization and intracellular electrical coupling in guinea-pig mesenteric arterioles. J. Physiol., 514: 505–513. Yamamoto, Y., Klemm, M.F., Edwards, F.R., Suzuki, H. (2001) Intercellular electrical communication among smooth muscle and endothelial cells in guinea-pig mesenteric arterioles. J. Physiol., 535: 181–195. Yamazaki, J., Kitamura, K. (2001) Cell-to-cell communication via nitric oxide modulation of oscillatory Cl⫺ currents in rat intact arterioles. J. Physiol., 536: 67–78. Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y., Yazaki, Y., Goto, K., Masaki, T. (1988) A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature, 332: 411–415. Yang, J., Jan, Y.N., Jan, L.Y. (1995) Control of rectification and permeation by residues in two distinct domains in an inward rectifier K⫹ channel. Neuron, 14: 1047–1054. Yang, D., Félétou, M., Boulanger, C.M., Wu, H.-F., Levens, N., Zhang, J.-N., Vanhoutte, P.M. (2002) Oxygen-derived free radicals mediate endothelium-dependent contractions to acetylcholine in aortas from spontaneously hypertensive rats. Br. J. Pharmacol., 136: 104–110. Yokoyama, S., Ohhashi, T. (1993) Effects of acetylcholine on spontaneous contractions in isolated bovine mesenteric lymphatics. Am. J. Physiol., 264: H1460–H1464. Yonehara, N., Takiuchi, S. (1997) Involvement of calcium-activated potassium channels in the inhibitory prejunctional effect of morphine on peripheral sensory nerves. Regul. Pept., 68: 147–153. Yoshioka, H., Morohashi, K., Sogawa, K., Miyata, T., Kawajiri, K., Hirose, T., Inayama, S., FujiKuriyama, Y., Omura, T. (1987) Structural analysis and specific expression of microsomal cytochrome P-450 (M-1) mRNA in male rat livers. J. Biol. Chem., 262: 1706–1711. You, J., Johnson, T.D., Marrelli, S.P., Bryan, R.M., Jr (1999a) Functional heterogeneity of endothelial P2 purinoceptors in the cerebrovascular tree of the rat. Am. J. Physiol., 277: H893–H900. You, J., Johnson, T.D., Marrelli, S.P., Mombouli, J.V., Bryan, R.M., Jr (1999b) P2u receptor-mediated release of endothelium-derived relaxing factor/nitric oxide and endothelium-derived hyperpolarizing factor from cerebrovascular endothelium in rats. Stroke, 30: 1125–1133. Yu, S.P., Kerchner, G.A. (1998) Endogenous voltage-gated potassium channels in human embryonic kidney (HEK293) cells. J. Neurosci. Res., 52: 612–617. Yuan, X.-J., Wang, J., Juhaszova, M., Gaine, S.P., Rubin, L.J. (1998) Attenuated K⫹ channel gene transcription in primary pulmonary hypertension. Lancet, 51: 726–727. Zanzinger, J., Czachurski, J., Seller, H. (1996) Role of calcium-dependent K⫹ channels in the regulation of arterial and venous tone by nitric oxide in pigs. Pflügers Arch., 432: 671–677. Zaritsky, J.J., Eckman, D.M., Wellman, G.C., Nelson, M.T., Schwarz, T.L. (2000) Targeted disruption of Kir2.1 and Kir2.2 genes reveal the essential role of the inwardly rectifying K⫹ current in K⫹mediated vasodilatation. Circ. Res., 87: 160–166. Zeldin, D.C. (2001) Epoxygenase pathways of arachidonic acid metabolism. J. Biol. Chem., 276: 36059–36062. Zhang, H., Inazu, M., Weir, B., Daniel, E. (1994) Endothelin-1 inhibits inward rectifier potassium channels and activates nonspecific cation channels in cultured endothelial cells. Pharmacology, 49: 11–22. Zhang, R., Pinson, A., Samuni, A. (1998) Both hydroxylamine and nitroxide protect cardiomyocytes from oxidative stress. Free Rad. Biol. Med., 24: 66–75. Zhang, X.F., Iwamuro, Y., Enoki, T., Okazawa, M., Lee, K., Komuro, T., Minowa, T., Okamoto, Y., Hasegawa, H., Furutani, H., Miwa, S., Masaki, T. (1999) Pharmacological characterization of Ca2⫹ entry channels in endothelin-1-induced contraction of rat aorta using LOE 908 and SK&F 96365. Br. J. Pharmacol., 127: 1388–1398.

References

415

Zhang, C., Medhora, M., Harder, D. (2001) Astrocyte-endothelial cell interaction: Effect of epoxyeicosatrienoic acids on mitogenesis and gene expression of rat cerebral microvascular endothelial cells in culture. FASEB J., 15: A460. Zhou, M.S., Nishida, Y., Chen, Q.H., Kosaka, H. (1999) Endothelium-derived contracting factor in carotid artery of hypertensive Dahl rats. Hypertension, 34: 39–43. ZhuGe, R., Fogarty, K.E., Tuft, R.A., Lifshitz, L.M., Sayar, K., Walsh, J.V.J. (2000) Dynamics of signaling between Ca2⫹ sparks and Ca2⫹- activated K⫹ channels studied with a novel image-based method for direct intracellular measurement of ryanodine receptor Ca2⫹ current. J. Gen. Physiol., 116: 845–864. Zou, A.P., Ma, Y.H., Sui, Z.H., Ortiz-de Montellano, P.R., Clark, J.E., Masters, B.S., Roman, R.J. (1994) Effects of 17-octadecynoic acid, a suicide-substrate inhibitor of cytochrome P450 fatty acid omega-hydroxylase, on renal function in rats. J. Pharmacol. Exp. Ther., 268: 474–481. Zou, A-P., Fleming, J.T., Falck, J.R., Jacobs, E.R., Gebremedhin, D., Harder, D.R., Roman, R.J. (1996) 20-HETE is an endogenous inhibitor of the large-conductance Ca2⫹-activated K⫹ channel in renal arterioles. Am. J. Physiol., 270: R228–R237. Zou, W., Yang, Q., Yim, A.P.C., He, G.-W. (2001) Epoxyeicosatrienoic acids (EET11,12) may partially restore endothelium-derived hyperpolarizing factor-mediated function in coronary micro-arteries. Ann. Thorac. Surg., 72: 1970–1976. Zygmunt, P.M., Ryman, T., Högestätt, E.D. (1995) Regional differences in endothelium-dependent relaxation in the rat – contribution of nitric oxide and nitric oxide-independent mechanisms. Acta Physiol. Scand., 155: 257–266. Zygmunt, P.M., Edwards, G. Weston, A.H., Davis, S.G., Högestätt, E.D. (1996) Effects of cytochrome P450 inhibitors on EDHF-mediated relaxation in the rat hepatic artery. Br. J. Pharmacol., 118(5): 1147–1152. Zygmunt, P.M., Högestätt, E.D. (1996) Role of potassium channels in endothelium-dependent relaxation resistant to nitroarginine in the rat hepatic artery. Br. J. Pharmacol., 117: 1600–1606. Zygmunt, P.M., Edwards, G., Weston, A.H., Larsson, B.I., Högestätt, E.D., (1997) Involvement of voltage-dependent potassium channels in the EDHF-mediated relaxation of rat hepatic artery. Br. J. Pharmacol., 121: 141–149.

Index

Author’s articles appear where page numbers are bold. A23187 222 Abramowitz, J. 20 acetylcholine, age effects 118–22; agonist-stimulation intensity 257–8, 260; ascorbate 183–4; calcium signal 304–8; cAMP permissive role 213, 214–22; diabetic mice 126–7, 128, 129, 130–1; DOCA-salt induced hypertension 141, 143–4, 145–6, 148; endothelial cells 105; endothelial stimulation 203; endothelium-dependent depolarization 200, 202; eNOS⫺/⫺ mice 284–96; gabexate mesilate 191; HEPES effects 245; human subcutaneous resistance arteries 207, 208, 210; hyperpolarization longitudinal spread 235, 236–8; hypertension 111, 114–15, 116; K⫹ channel remodeling 9; KCa role 275–82; Kv channels 6; local release of EDHF 333–40; mesenteric arterioles 197; mesenteric lymphatic endothelium 96; mesenteric veins/arteries 297–303; myoendothelial gap junctions 226, 227–9; renal blood flow 80, 81–2, 85; sodium intracellular concentration 319; vascular tone 102 N-acetylcysteine, red wine polyphenolic compounds 169, 172; vascular tone 156–64 Ad5–2C11, astrocytic EETs 343, 348 Ad5-GFP, astrocytic EETs 343, 345, 348 Adeagbo, A.S.O. 139 adenosine, local release of EDHF 334 adenylyl cyclase, cAMP permissive role 222 ADP-ribosylation, BKCa ␣-subunit 361–2; G-protein G␣s 356–65 age-related impairment, renin-angiotensin system blockade 117–23 Allen, J.C. 20 allopurinol, red wine polyphenolic compounds 169–70 3-aminobenzamide 362, 364 2-aminoethoxydiphenyl borate (2APB), basal tone 197, 198 4-aminopyridine, calcium signal 305–7, 308; DOCA-salt induced hypertension 141, 145; potassium channels 5, 6

Anderson, T.J. 124 angiotensin II 123 antibodies, anti-G␣s 360, 361, 363; anti-G␤␥ 360, 361, 363; anti-G␣i/o 360, 361, 363; HEPES effects 240–1; hypertension 110; KIR and Na⫹/K⫹-ATPase 311, 313, 315–16 aorta, KCa role 274–82; PAR2 48–9, 50, 53 apamin 371, 373; ascorbate 183, 184, 185, 186, 187; Ca2⫹-activated K⫹ channels 105; calcium signal 305, 306, 308; diabetic mice 127, 128, 129, 130; DOCA-salt induced hypertension 141, 143; endothelial potassium release 251; endothelial stimulation 203; endothelium-dependent depolarization 200; eNOS⫺/⫺ mice 283–96; estrogen effect 177–8, 179; HEPES effects 246–7; human platelets 353, 355; human subcutaneous resistance arteries 207, 208; hypertension 111, 114, 115; KCa role 278, 279–81; myoendothelial gap junctions 229; PAR2 51–2, 54; red wine polyphenolic compounds 168, 171; SKCa channels 7; SKCa and IKCa 261–73; skin pressure-induced vasodilatation 152, 153–4; smooth muscle cells 106; urocortin-induced relaxations 89, 90, 92 2APB see 2-aminoethoxydiphenyl borate arachidonic acid, astrocytic EETs 345, 346; cytochrome P450 2C 56, 61; DOCA-salt induced hypertension 149–50; free radicals 142, 146, 147; microsomal metabolism 141, 146, 147, 148 arachidonic acid epoxygenases, cytochrome P450 2C 58 L-arginine 101–2 ascorbate, EDHF inhibition 181–7 astrocytes, cerebral endothelial cells interaction 341–8 astrocytic epoxyeicosatrienoic acids 341–8 atherosclerosis, hydrogen peroxide as EDHF 66 ATP, dilatations attenuation 72, 73, 74, 77; Na,K-ATPase 27 ATP-sensitive K⫹ channel (KATP) 185, 229, 314

418

Index

autacoids, local release of EDHF 333–40 Awe, S.O. 139 Aydemir-Koksoy, A. 20 ␤-adrenergic agonists, Na pump 21 Baker, P.N. 132 Bandi, H.P. 108 barium, diabetic mice 128, 129, 130; endothelium-dependent depolarization 200; KIR channels 4; KIR and Na⫹/K⫹-ATPase 312, 313, 314; mesenteric veins/arteries 298; potassium release 105; urocortin-induced relaxations 88, 89, 90, 91 barium chloride, KCa role 278; mesenteric veins/arteries 299, 301; PAR2 54; urocortin-induced relaxations 88 Baron, A. 248 basal vascular tone, mesenteric arterioles 193–8 Belyaeva, L.E. 156 Bény, J.-L. 248 BH4 see tetrahydrobiopterin bimakalim, bradykinin-induced relaxations 331 bioassay method, KCa role 275–6, 279, 280 BKCa see large-conductance Ca2⫹-sensitive K⫹ channels Blanco, G. 27 blood pressure, connexin-mimetic peptides 78–86; DOCA-salt induced hypertension 142–3, 144, 149; RAS blockade 118, 122; skin pressure-induced vasodilatation 152, 153; vascular tone dysregulation 158 Bolz, S.-S. 332 bradykinin, ascorbate 183, 184–5; diabetic mice 126–7, 128, 129, 130–1; 11,12-EET role 367–70; endothelial cells 105; endothelial stimulation 203; 14,15-epoxyeicosa-5Zmonoenoic acid 325–31; HEPES effects 241–2, 243–7; hydrogen peroxide as EDHF 64, 65, 66, 67; KCa role 278, 279, 282; PAR2 54; potassium release quantification 248–55; pre-eclampsia 134–6, 137, 138; SKCa and IKCa 269, 272–3 bradykinin receptors, endothelium-dependent depolarization 200 Bramich, N.J. 108 Britton, F. 13 Brochet, D.X.P. 318 8-bromo-cAMP 213, 214, 215, 217, 218, 220, 221 8-bromo-cGMP 217, 218, 221 Bryan, R.M. Jr 70 Burnham, M.P. 261 Busse, R. 56 tert-butyl hydroperoxide (t-BOOH), DOCA-salt induced hypertension 144, 146, 148, 149, 150 Bychkov, R. 261

Ca2⫹-activated Cl⫺ channels 200, 201, 203 Ca2⫹-activated Cl⫺ currents (ICl(Ca)) 13–19 Ca2⫹-activated K⫹ channels (KCa) 6–8, 172; agonist-stimulation intensity 259; cAMP permissive role 221; EETs 104; endothelial stimulation 203; genes 6–7; 20-HETE 202; human platelets 353, 355; local release of EDHF 337–8, 339–40; rabbit aorta 274–82; skin pressure-induced vasodilatation 153; urocortin-induced relaxations 91–2; vascular tone 105 Ca2⫹-calmodulin complex 172 Ca2⫹ channels see also voltage-gated Ca2⫹ channels L-type 222 Ca2⫹-induced Ca2⫹ release 36 Ca2⫹ influx, K⫹ channels 12; potassium channels link 2–3 Ca2⫹-permeable cation channels, U44619 202 Ca2⫹ signal, EDHF-evoked responses inhibition 304–8 Ca2⫹ sparks, membrane potential 35–46 calcein, cAMP permissive role 214, 217 calcium see also Ca2⫹, cAMP permissive role 222; dynamics, dilatations attenuation 70–7; gap junctions 372; mesenteric arterioles 195 calcium phosphate co-precipitation method 15 calmidazolium, red wine polyphenolic compounds 168, 171–2 calmodulin, red wine polyphenolic compounds 171–2; SKCa channels 7 cAMP see cyclic adenosine monophosphate Campbell, W.B. 325 cancer, K⫹ channel remodeling 9 candesartan cilexetil, renin-angiotensin system blockade 118, 119–20, 122 carbenoxolone, agonist-stimulation intensity 257–8, 259, 260; myoendothelial gap junctions 232; pre-eclampsia 133, 135, 137, 138 cardiac contractility, vascular tone dysregulation 161 cardiac muscle, calcium sparks 35–6, 38 cardiomyocytes, Na pump 22, 23, 25 cardiovascular diseases, K⫹ channel remodeling 9 catalase 64; diabetic mice 128, 129, 131; hydrogen peroxide as EDHF 66; red wine polyphenolic compounds 169, 172 caudal artery, hypertension 109–16 caveolae 25, 26 caveolin 25, 26 cell-attached patches, BKCa channel ␣-subunit 359 cerebral artery, diameter measurement 72, 73; dilatations attenuation 70–7; harvesting and mounting 71; myocytes, calcium sparks 40 cerebral endothelial cells, astrocytes interaction 341–8 cGMP see cyclic guanosine monophosphate

Index 419 Chan, A.K. 93 Chan, F.L. 87 Chan, L. 274 charybdotoxin 371, 373; ascorbate 183, 184–5, 186, 187; Ca2⫹-activated K⫹ channels 105; calcium-sensitive potassium channels 7; calcium signal 305, 306, 308; diabetic mice 128, 129, 130; DOCA-salt induced hypertension 141; endothelial potassium release 251; endothelial stimulation 203; endothelium-dependent depolarization 200; eNOS⫺/⫺ mice 283–96; estrogen effect 177–8, 179; HEPES effects 247; human platelets 352, 353, 354, 355; human subcutaneous resistance arteries 207, 208; hypertension 111, 114, 115; KCa role 278, 279; local release of EDHF 334, 336–7, 338; myoendothelial gap junctions 229; PAR2 52, 54; potassium channels 5; red wine polyphenolic compounds 168, 169, 171; SKCa and IKCa 261–73; skin pressureinduced vasodilatation 152–4; smooth muscle cells 106; urocortin-induced relaxations 89 Chataigneau, T. 165, 174 Chaytor, A.T. 211 Chen, Z.-Y. 87 CHIF see corticosteroid hormone-induced factor cholera toxin 361–2, 363 choline chloride, sodium intracellular concentration 319, 322, 323–4 cirazoline, DOCA-salt induced hypertension 141, 145; eNOS⫺/⫺ mice 286, 287 CLCA1 13–19 CLCA3 15, 18 CLCA4 14, 15, 18 clotrimazole, calcium-sensitive potassium channels 7; DOCA-salt induced hypertension 146, 147 clso-␣ 357, 358–65 clso-␤ 358, 364–5 Coats, P. 205 Coleman, H.A. 101, 199, 223 colonic adenocarcinoma, K⫹ channel remodeling 9 connexins see also Gap 27; connexin 37 229–30; connexin 40 107, 109, 110; connexin 43 107, 109, 229–30; HEPES effects 244–5; mimetic peptides 78–86 coronary artery, ascorbate inhibition 181–7; 11,12-EET 366–70; endothelial cells, cytochrome P450 2C 56–62; 14,15epoxyeicosa-5Z-monoenoic acid 327–31; HEPES 239–47; hydrogen peroxide as EDHF 64, 66–7; K⫹ channel remodeling 9; red wine polyphenolic compounds 165–73; urocortininduced relaxations 87–92 coronary epithelium, SKCa and IKCa 261–73

coronary flow, vascular tone dysregulation 157–60, 162–3 coronary perfusion pressure 157, 158–9, 162, 163 coronary vascular tone, dysregulation 156–64 corticosteroid hormone-induced factor (CHIF) 29, 31–4 corticotropin-releasing factor (CRF) 87; urocortin-induced relaxations 90–1 cyclic adenosine monophosphate (cAMP), 8-bromo-cAMP 213, 214, 215, 217, 218, 220, 221; calcium sparks 40; gap junctions 372; permissive role 211–22 cyclic guanosine monophosphate (cGMP), 8-bromo-cGMP 217, 218, 221; electronic hyperpolarization 213, 216; urocortin-induced relaxations 91, 92; vascular tone dysregulation 158, 159, 162, 163 cyclic stretch, cytochrome P450 2C 57, 58–9, 60–1, 62 cyclooxygenase, diabetic mice 127, 128, 129, 130; DOCA-salt induced hypertension 143, 149; endothelium-dependent depolarization 202; local release of EDHF 334, 339; microsomal metabolism 146; PAR2 54; red wine polyphenolic compounds 172 cyclooxygenase-independent renal vasodilatation, connexin-mimetic peptides 78–86 cyclopiazonic acid, cAMP permissive role 221; KCa role 278, 279–82; mesenteric lymphatic endothelium 94, 96–7, 98, 99–100 cytochrome P450 67–8, 373, 375; 2C11 344–5, 348; 2C 56–62, 172; astrocytic epoxyeicosatrienoic acid 342; bradykinin-induced relaxations 326, 328–9, 331; diabetic mice 128, 131; DOCA-salt induced hypertension 139–50; endothelial potassium release 252; endothelium-dependent depolarization 202; epoxygenase 67–8; pre-eclampsia 138; vascular tone 104–5 cytoplasmic translocation, Na pump 21–2, 24 DCF see dichlorofluorescein DCFDA see 2⬘,7⬘-dichlorofluorescein diacetate deoxycorticosterone-salt induced hypertension 139–50 detaNONOate, renal blood flow 80, 82–3, 85 De Vriese, A.S. 78 de Wit, C. 332 DHETs see dihydroxyeicosatrienoic acids diabetic mice, mesenteric arteries 124–31 DiBac4(3), human platelets 351, 352, 354 dichlorofluorescein (DCF), DOCA-salt induced hypertension 149 2⬘,7⬘-dichlorofluorescein diacetate (DCFDA), DOCA-salt induced hypertension 146, 149; free radicals 142

420

Index

dihydroxyeicosatrienoic acids (DHETs), DOCA-salt induced hypertension 146, 148 Ding, H. 283 diphenylene iodonium, red wine polyphenolic compounds 169–70 dithiothreitol (DTT), Ca2⫹-activated Cl⫺ currents 17, 18 DNA arrays, astrocytic EETs 343–4 dopamine, Na pump 22 Dora, K.A. 234, 256 dose-response curves, Na,K-ATPase 31 doxycycline 375 drug–receptor interaction model 24 DTT see dithiothreitol D-tubocurarine 7, 272, 283–96 Dunn, W.R. 132 dye coupling, hyperpolarization longitudinal spread 235–6 dye transfer studies, cAMP permissive role 213–14, 217–18, 221 1-EBIO see 1-ethyl-2-benzimidazolinone Eckman, D.M. 35 EDCFs see endothelium-derived contracting factors Edwards, D.H. 211 Edwards, G. 239, 261, 309 EETs see epoxyeicosatrienoic acids EEZE see 14,15-epoxyeicosa-5Z-monoenoic acid EGFR see epidermal growth factor receptor eicosanoids, cytochrome P450 2C 60; vascular tone 104 electrical coupling 203; hyperpolarization longitudinal spread 235, 236–7 electrophysiology, hypertension 110; mesenteric veins/arteries 298–9 enalapril, renin-angiotensin system blockade 118, 119–20, 122 endocannabinoids 138 endothelin-1 201 endothelin receptors 200, 201, 202 endothelium-derived contracting factors (EDCFs) 202 epidermal growth factor receptor (EGFR), Na⫹ pump 25, 26 14,15-epoxyeicosa-5Z-monoenoic acid (14,15-EEZE) 325–31, 373 8,9-epoxyeicosatrienoic acids (8,9-EET), astrocytic EETs 345–7, 348; cerebral endothelial cells 343, 344 11,12-epoxyeicosatrienoic acids (11,12-EET) 58, 366–70, 373; BKCa channel ␣-subunit 358–60, 362–3, 364, 365; human platelet hyperpolarization 349–55 14,15-epoxyeicosatrienoic acids (14,15-EET) 58 epoxyeicosatrienoic acids (EETs) 67; astrocytic 341–8; BKCa channel ␣-subunit 356–65;

cytochrome P450 2C 56–62; diabetic mice 131; DOCA-salt induced hypertension 148, 150; endothelial potassium release 253; 14,15-epoxyeicosa-5Z-monoenoic acid 325–31; microsomal metabolism 141, 146; vascular tone 104–5 ERK 1/2 60 17␤-estradiol, dilatations attenuation 71, 73–7; vascular tone 176–80 estrogen, depletion/repletion 71; vascular tone 174–80 ESTs see expressed sequence tags 1-ethyl-2-benzimidazolinone (1-EBIO), estrogen effect 179, 180; HEPES effects 243, 244; myoendothelial gap junctions 225, 229, 230; SKCa and IKCa 269, 271–2, 273 expressed sequence tags (ESTs) 31–2 eye, ascorbate inhibition 181–7 Falck, J.R. 325 Falkner, K.C. 139 Félétou, M. 239, 261, 309 femoral artery, cAMP permissive role 213–14, 217–18, 221; hypertension 109–16; myoendothelial gap junctions 224–33; PAR2 48–9, 50, 53 Ferens, D.M. 70 Fisslthaler, B. 56 Fleming, I. 56 fluid shear stress, cytochrome P450 2C 58–9, 60–1, 62 fluorescence studies, calcium signal 305, 307 forskolin, calcium sparks 40 free radicals, hypertension 139–50 Frieden, M. 248 Fromy, B. 151 Fujii, K. 117 Fukao, M. 356 fura 2 72 FXYD see phenylalanine-X-tyrosine-aspartate G12 protein dependent pathway 201 GA see 18␣-glycyrrhetinic acid; 18␤-glycyrrhetinic acid gabexate mesilate 188–92 Gap 27 302, see also connexin; cAMP permissive role 212, 213, 214, 215–16, 217–18, 220; 40Gap 27 81–4, 85, 246; 43 Gap 27 80, 82–4, 85, 246; 37,43Gap 27 225, 229–30, 231, 232–3 GAPDH housekeeping gene 269, 314 gap junctions 79, 107, 108–16, 223–33, 372, 375; agonist-stimulation intensity 256–60; ascorbate 186; basal tone 196, 198; cAMP 211–22; connexin-mimetic peptides 85–6; EDHF-mediated relaxations 68; endotheliumdependent depolarization 200, 203, 204;

Index 421 HEPES effects 246; human subcutaneous resistance arteries 205–10; hyperpolarization longitudinal spread 234–8; hypertension 108–16; K relationship 26; pre-eclampsia 138; RAS inhibitors 123; vascular tone 2 Gardener, M.J. 239, 309 Garland, C.J. 234, 256 Garry, A. 151 Gauthier, K.M. 1, 325 GDP␤S 360, 362–3 gender differences, calcium changes 74–7; vascular reactivity 70 gene expression, astrocytic EETs 345–6 gene-specific RT-PCR, KIR and Na⫹/K⫹-ATPase 310–11; SKCa and IKCa 263 gene transcription, Na⫹ pump 24 Ghisdal, P. 304 glibenclamide 185–6; KIR and Na⫹/K⫹ATPase 314–15; urocortin-induced relaxations 89, 92 glutathione, vascular tone dysregulation 158, 159, 164 glyceryl trinitrate, ascorbate 185 glycosides, Na,K-ATPase 28 18␣-glycyrrhetinic acid (18␣-GA) 64, 68, 69; agonist-stimulation intensity 259; cAMP permissive role 212, 214, 219–20, 221; human subcutaneous resistance arteries 207, 208–9, 210; mesenteric arterioles 198; pre-eclampsia 133, 135, 137, 138 18␤-glycyrrhetinic acid (18␤-GA), agonist-stimulation intensity 259; mesenteric arterioles 196, 198 Golding, E.M. 70 Goto, K. 117 G-protein coupled receptors, PAR2 47–55 G-protein G␣s, ADP-ribosylation 356–65 Gq/phospholipase C pathway 201 Gq protein dependent pathway 201 gramicidin, human platelets 352, 354, 355 Greenwood, I.A. 13 Griffith, T.M. 211 guanosine triphosphate (GTP), BKCa channel ␣-subunit 360 guanylyl cyclase, diabetic mice 127, 130 Gui, Y. 47 H89, calcium sparks 40; mesenteric veins/arteries 299, 301 Hamada, T. 188 Harder, D.R. 341 heart failure, Na,K-ATPase 28 heart mass, DOCA-salt induced hypertension 142–3 heart rate, connexin-mimetic peptides 84; hemorrhage 159; vascular tone dysregulation 158

He, G.-W. 87, 366 HEK293 cells, BKCa channel ␣-subunit 361, 362, 364; Ca2⫹-activated Cl⫺ currents 15–17, 18; clso-␣ 357, 358 hemoglobin, 11,12-EET role 367, 368–9, 370 hemorrhage, endothelium-dependent coronary vascular tone 156–64 HEPES, coronary and mesenteric arteries 239–47 20-HETE see 20-hydroxyeicosatetraenoic acid heterocellular coupling 236 Hill, C.E. 108, 223 Hillier, C. 205 Hirakawa, Y. 63 histamine, mesenteric veins/arteries 299 Hoepfl, B. 332 Hollenberg, M.D. 47 homocellular coupling 236 Horowitz, B. 13, 356 Huang, Y. 87, 366 Hudlett, F. 174 hydralazine 119–20; skin pressure-induced vasodilatation 153–4 hydrochlorothiazide 119–20 hydrogen peroxide 373; diabetic mice 131; eNOS⫺/⫺ mice 292; gabexate mesilate 192; PAR2 52, 53; red wine polyphenolic compounds 169, 172; role of 63–9 20-hydroxyeicosatetraenoic acid (20-HETE) 146, 202, 326; bradykinin-induced relaxations 331; DOCA-salt induced hypertension 148 hyperglycaemia 124 hypertension, age effects 122; deoxycorticosterone-salt induced 139–50; K⫹ channel remodeling 9–11; myoendothelial gap junctions 108–16; pre-eclampsia 132; vascular K⫹ channels 8–9 hypocalciuria 28 hypomagnesemia 28 iberiotoxin 45; BKCa channels 6, 7, 8, 11, 358; diabetic mice 128, 129, 130; endothelial potassium release 251; eNOS⫺/⫺ mice 289–91, 294; HEPES effects 242, 246–7; human subcutaneous resistance arteries 207; KCa role 279, 281; KIR and Na⫹/K⫹-ATPase 311–12, 313; local release of EDHF 334, 337, 338; SKCa and IKCa 265–9, 271–2, 273; urocortin-induced relaxations 88, 90, 92; vascular tone 103, 104, 105 iberiotoxin-sensitive potassium currents 7 IBMX see 3-isobutyl-1-methylxanthine Iida, M. 117 IKCa see intermediate-conductance Ca2⫹-sensitive K⫹ channels iliac artery, cAMP permissive role 213–22 iloprost, bradykinin-induced relaxations 331 imidazole, cytochrome P450 104

422

Index

immobilization stress, vascular tone 156–64 immunofluorescence histochemistry, eNOS⫺/⫺ mice 285–6, 291, 295–6; KIR and Na⫹/K⫹ATPase 311, 315; SKCa and IKCa 264, 270–1 immunohistochemistry, cytochrome P450 2C 58; hypertension 110 indomethacin, age effects 118–19, 121–2; ascorbate 183; bradykinin-induced relaxations 328–9, 331; calcium signal 305; cAMP permissive role 216; diabetic mice 126–7, 128, 129; dilatations attenuation 72, 74–5, 76, 77; DOCA-salt induced hypertension 143; 11,12-EET role 367, 368–9, 370; endothelial potassium release 252; eNOS⫺/⫺ mice 286–8, 289, 290, 291, 293; estrogen effect 177; gabexate mesilate 189, 190–1, 192; human subcutaneous resistance arteries 207, 208; hydrogen peroxide as EDHF 64, 65; hypertension 111, 114–15; KCa role 277, 278, 279–80, 281, 282; local release of EDHF 334, 335, 336, 337; mesenteric veins/arteries 299, 300, 301; microsomal metabolism 141, 146; myoendothelial gap junctions 226–7, 228, 229; PAR2 51–2; red wine polyphenolic compounds 167, 168, 169–70, 172; renal blood flow 80–6; urocortin-induced relaxations 89; vascular tone 102, 103 Ings, N. 256 inositol triphosphate (IP3) 372; basal tone 197, 198; Ca2⫹ release 201; endothelium-dependent depolarization 200 inside-out patches, BKCa channel ␣-subunit 359–60 intermediate-conductance Ca2⫹-sensitive K⫹ channels (IKCa) 5, 6, 7–8, 12, 371, 373; ascorbate 186, 187; coronary epithelium 261–73; disease 9; endothelial stimulation 203; gap junctions 229; human platelets 355; red wine polyphenolic compounds 171, 172 inward-rectifier K⫹ channels (KIR) 3–5, 12, 202, 322, 372, 375; endothelium-dependent depolarization 200, 201; genes 3–4, 5, 311, 312, 315–16; mesenteric arteries 309–17; urocortin-induced relaxations 91 IP3 see inositol triphosphate isethionate, Ca2⫹-activated Cl⫺ currents 18 3-isobutyl-1-methylxanthine (IBMX), cAMP permissive role 213, 214, 215–16, 217–18, 220, 221 isolated perfused vascular bed studies, cAMP permissive role 214 isometric tension 118–19 Jiang, Y. 283 Joshua, I.G. 139

K⫹ channels 200, see also Ca2⫹-activated K⫹ channels; intermediate-conductance Ca2⫹sensitive K⫹ channels; inward-rectifier K⫹ channels; large-conductance Ca2⫹-sensitive K⫹ channels; small-conductance Ca2⫹sensitive K⫹ channels; voltage-gated K⫹ channels; ATP-sensitive 185, 229, 314; blockers 204; diabetic mice 127–8, 130, 131; endothelial cells 105; endothelial and smooth muscle cells 1–12; endothelin-1 201; eNOS⫺/⫺ mice 283–96; openers 229; remodeling in disease 8–11; urocortin-induced relaxations 87–92 kallistatin 192 Kansui, Y. 117 KCNMB1 18–19 Keef, K.D. 356 Keller, M. 349 Kendall, D.A. 132 Kenny, L.C. 132 Kenyon, J.L. 356 kidneys, PAR2 49, 52 King, M. 256 KIR see inward-rectifier K⫹ channels Krötz, F. 349 KT 5720 362, 363, 364 Kuang, X. 274 Kubota, H. 63 Laher, I. 274 Lameire, N.H. 78 Langton, P.D. 318 Lansbery, K. 27 large-conductance Ca2⫹-sensitive K⫹ channels (BKCa) 5, 6–7, 8, 12, 271; ␣ subunit 43, 356–65; ␤1 subunit 43, 44–5; bradykinin-induced relaxations 331; calcium sparks 36, 37, 39–40, 43–5, 46; disease 9, 10–11; endothelial potassium release 251–2, 253; endothelium-dependent depolarization 200; KIR and Na⫹/K⫹-ATPase 313; potassium release quantification 249–50; urocortin-induced relaxations 90; vasodilators 40–3 L-arginine 101–2 Lau, C.-W. 87 levcromakalim, age effects 119, 122; ascorbate 184, 185, 187; HEPES effects 242, 243–5; hydrogen peroxide as EDHF 66, 67; hyperpolarization longitudinal spread 235, 236–7, 238; KIR and Na⫹/K⫹-ATPase 313; myoendothelial gap junctions 225, 229, 230; red wine polyphenolic compounds 169 Liang, W. 274 lipid alkoxyl radicals 149 lipid peroxide anions 149 lipid peroxyl radicals 149

Index 423 lipid profile, estrogen effect 174 lipid radicals 149 lipoxygenases, DOCA-salt induced hypertension 149; PAR2 52, 53, 55 L-NA see NG-nitro-L-arginine L-NAME see N-nitro-L-arginine methyl ester L-NNA see Nw-nitro-L-arginine LOE-908 201 Loutzenhiser, R.D. 47 LRGILS-NH2, mesenteric lymphatic endothelium 98 L-type Ca2⫹ channels 222 LY 294002, Na pump 21 lymphatic endothelium, nitric oxide 93–100 McEvoy, L. 256 McGuire, J.J. 47 McNeish, A.J. 181 malondialdehyde, vascular tone dysregulation 158, 159 mammary tumor protein of 8-kDA (Mat-8) 29, 31–2 Marrelli, S.P. 70 Martin, W. 181 Mason, H.S. 356 Mat-8 see mammary tumor protein of 8-kDA Matoba, T. 63 maxi-K potassium channels, DOCA-salt induced hypertension 141, 144, 145 Medhora, M. 341 membrane potential, BKCa channel ␣-subunit 357–65; calcium signal 305–6; calcium sparks 35–46; cAMP permissive role 213, 216–17; coronary artery 264–5; endothelial cells 371; endothelial and smooth muscle cells 1–12; human platelets 350, 351–4; hydrogen peroxide as EDHF 66; hypertension 110; K⫹ channel remodeling 9; KIR and Na⫹/K⫹ATPase 309–17; Kv channels 5, 6; mesenteric arterioles 194, 195–6, 197–8; mesenteric lymphatic endothelium 95, 96–7; mesenteric veins/arteries 298–9, 300–2; myoendothelial gap junctions 225, 227, 231–2; rabbit aorta 279–81; RAS blockade 118; sodium intracellular concentration 319–24; vascular tone 104 Mendenhall, M.L. 27 Mercer, R.W. 27 Merzeau, S. 151 mesenteric arteries, agonist-stimulation intensity 256–60; calcium signal 304–8; diabetic mice 124–31; DOCA-salt induced hypertension 141–50; EDHF-mediated relaxations 68, 69; eNOS⫺/⫺ mice 283–96; estrogen substitution 174–80; gabexate mesilate 188–92; HEPES 239–47; hydrogen peroxide as EDHF 63–9; hyperpolarization

longitudinal spread 234–8; hypertension 109–16; inward-rectifier K⫹ channels 309–17; myoendothelial gap junctions 224–33; Na⫹/K⫹-ATPase 309–17; ouabain 297–303; PAR2 48–9, 50–2, 53, 54, 55; renin-angiotensin system blockade 118–23; sodium intracellular concentration 318–24; vascular tone 193–8 mesenteric lymphatic endothelium, nitric oxide 93–100 mesenteric veins, ouabain 297–303 MIBG see m-iodobenzylguanidine Michaelis, U.R. 56 miconazole 107; bradykinin-induced relaxations 329; cytochrome P450 enzymes 67 microelectrodes, KIR and Na⫹/K⫹-ATPase 310; red wine polyphenolic compounds 166–7; SKCa and IKCa 262–3 microsomal metabolism, arachidonic acid 141, 146, 147, 148 microsomal production, free radicals 142 m-iodobenzylguanidine (MIBG) 362, 364 mitochondrial respiratory chain 172 mitogen-activated protein kinases 60, 62 Mitsumizo, S. 188 Miwa, S. 356 MnTMPyP, red wine polyphenolic compounds 169–70, 172 mono-ADP-ribosyltransferase inhibitors 362, 363–4 Morel, N. 304 Morikawa, K. 63 muscarinic receptors, endothelium-dependent depolarization 200; PAR2 54 myocytes, Ca2⫹-activated Cl⫺ currents 15–18; calcium sparks 40, 44 myoendothelial gap junctions see gap junctions myography 224–5, 226, see also wire myography; 11,12-EET 367; mesenteric veins/arteries 298 myometrial arteries, pre-eclampsia 133–8 myxothiazol, red wine polyphenolic compounds 169–70 Na⫹/K⫹-ATPase see also Na⫹pump; ␣ subunit 27–9, 32; BKCa channel ␣-subunit 360; ␤ subunit 27–9, 32; isoforms 27–34; mesenteric arteries 309–17; sodium intracellular concentration 322–3, 324; ␥ subunit 27–9, 31–4 Na⫹ pump, calcium signal 308; mesenteric veins/arteries 297–8; trafficking and transduction 20–6 N-acetylcysteine, red wine polyphenolic compounds 169, 172; vascular tone 156–64 NADPH, DOCA-salt induced hypertension 149; microsomal metabolism 146, 147

424

Index

NADPH oxidase, endothelium-dependent depolarization 202; red wine polyphenolic compounds 169, 172 nafamostat mesilate 192 Nakashima, M. 188 ␤-naphthoflavone, DOCA-salt induced hypertension 146, 147, 149 Narayanan, J. 341 Nawate, S. 356 Ndiaye, M. 165 Nelli, S. 181 Nelson, M.T. 35 Nernst equation 352 nicorandil, calcium sparks 40 nifedipine, basal tone 196–7, 198 nitric oxide, age effects 120; cytochrome P450 2C 60; endothelial cells 105; endothelium-dependent depolarization 204; estrogen substitution 174–80; flow 106; hypertension 111; KCa role 277, 281; local release of EDHF 333; lymphatic endothelium 93–100; mesenteric veins/arteries 302–3; PAR2 52–3; pre-eclampsia 137–8; red wine polyphenolic compounds 165–6, 171, 172; skin pressure-induced vasodilatation 151, 154–5; urocortin-induced relaxations 90, 91; vascular tone 101–3, 158, 159; vasodilatation 104, 106 nitric oxide synthase 373; connexin-mimetic peptides 78–86; diabetic mice 127, 128, 129, 130; DOCA-salt induced hypertension 143; endothelium-dependent depolarization 202; estrogen effect 175; local release of EDHF 334, 339; PAR2 52–4, 55; red wine polyphenolic compounds 170–1, 172–3; vascular tone dysregulation 158, 159, 163, 164 nitric oxide synthase knockout (eNOS⫺/⫺ mice), mesenteric arteries 283–96; PAR2 50–1, 53–4 NG-nitro-L-arginine (L-NA), age effects 118, 120, 121; bradykinin-induced relaxations 327, 329, 331; 11,12-EET role 367, 368–9, 370; endothelial potassium release 252; human subcutaneous resistance arteries 207, 208; local release of EDHF 334, 335, 336, 337; mesenteric lymphatic endothelium 96–8, 99–100; mesenteric veins/arteries 299, 300, 301, 302 Nw-nitro-L-arginine (L-NNA), calcium signal 305; diabetic mice 126–7, 129; eNOS⫺/⫺ mice 286–8, 289, 290, 291; estrogen effect 177, 178, 179; hydrogen peroxide as EDHF 64, 65; red wine polyphenolic compounds 167, 168, 169–70, 171; skin pressure-induced vasodilatation 154–5 N-nitro-L-arginine methyl ester (L-NAME), agonist-stimulation intensity 257–8; ascorbate

182, 183; bradykinin-induced relaxations 328–9; cAMP permissive role 214; dilatations attenuation 72, 74–5, 76, 77; DOCA-salt induced hypertension 143; gabexate mesilate 189, 190–1, 192; hypertension 111, 114–15; KCa role 277, 279–80, 281, 282; myoendothelial gap junctions 226–7, 228, 229; PAR2 50–2, 53, 54; pre-eclampsia 133, 135–6, 137; renal blood flow 80–6; urocortin-induced relaxations 89, 91; vascular tone 102, 103, 158, 163 non-selective cation channels (NSCC), endothelial potassium release 252; NSCC1 200, 201; NSCC2 200, 201 norepinephrine, age effects 118–19, 120–1 NS1619, bradykinin-induced relaxations 331; HEPES effects 242, 243, 244, 245; KCa role 281, 282 17-octadecynoic acid (17-ODYA) 64, 68, 107; astrocytic epoxyeicosatrienoic acid 342; diabetic mice 128, 129; endothelial potassium release 252; local release of EDHF 334, 336, 338, 339; PAR2 54 ODQ see 1H-[1,2,4]oxadiazolo [4,2-␣]quinoxalin-1-one Ohya, S. 13 organ bath, KCa role 275, 277–9 organ chambers, cAMP permissive role 213; gabexate mesilate 189; hydrogen peroxide as EDHF 64 ouabain, calcium signal 306, 308; diabetic mice 128, 129, 130; DOCA-salt induced hypertension 145; KIR and Na⫹/K⫹-ATPase 312, 313, 314, 316; mesenteric veins/arteries 297–303; Na⫹ pump 21, 22–4, 26; Na,K-ATPase 27, 28, 31, 32–4; PAR2 54; potassium release 105–6; sodium intracellular concentration 319, 322 ovariectomy, dilatations attenuation 70–7; estrogen substitution 174–80 1H-[1,2,4]oxadiazolo[4,2-␣]quinoxalin-1-one (ODQ), diabetic mice 126, 127; eNOS⫺/⫺ mice 286–7, 291–3; mesenteric veins/arteries 299; PAR2 51–2, 53; urocortin-induced relaxations 88, 89, 91 P13 kinase inhibitor LY 294002 21 p38 MAP kinase 60 P450-monoxygenase, local release of EDHF 334, 336, 337, 338, 339 palmitoleic acid, agonist-stimulation intensity 259; pre-eclampsia 133, 135, 137, 138 pancreatitis, gabexate mesilate 188, 192 Pannirselvam, M. 124 papaverine, renal blood flow 80, 82–3, 85 PAR2 see proteinase-activated receptor-2 Parkington, H.C. 101, 199, 223

Index 425 patch-clamp method, BKCa channel ␣-subunit 357, 358; K⫹ channels 9; KCa role 277, 281; mesenteric arterioles 194; potassium channels 1, 5, 7; potassium release quantification 249; SKCa and IKCa 262, 265–9 penitrem A, DOCA-salt induced hypertension 141, 143, 144, 145 perfused ear experiments, cAMP permissive role 219–20, 221 perfusion studies, DOCA-salt induced hypertension 140–1 Pernot, F. 174 phentolamine 305 phenylalanine-X-tyrosine-aspartate (FXYD) 29–34 phenylalanine-X-tyrosine-aspartate 7 (FXYD7) 29, 30, 31–2 phenylephrine, agonist-stimulation intensity 257–8, 259; cAMP permissive role 212, 213, 214–16, 219–20, 221; diabetic mice 126, 127, 130; gabexate mesilate 189–91; KCa role 277, 278, 280; KIR and Na⫹/K⫹-ATPase 311, 312, 313, 316; myoendothelial gap junctions 226; sodium intracellular concentration 319–24; vascular tone 177, 178 phosphatidylinositol bisphosphate (PIP2) 202 phosphodiesterase Type I 222 phosphohippolin (PHP) 29, 30, 31–2 phospholamban (PLB), calcium sparks 38, 40–1 phospholemman (PLM) 29, 31–2, 33 phospholipase A2 61, 202 PHP see phosphohippolin pinacidil, renal blood flow 80, 82–3, 85 PIP2 see phosphatidylinositol bisphosphate PKA see protein kinase A PKC see protein kinase C PKG see protein kinase G platelet hyperpolarization, 11,12-EETs 349–55 PLB see phospholamban PLM see phospholemman Pohl, U. 332, 349 polyphenolic compounds, red wine 165–73 poorly selective cation channels (PSCC) 200, 201, 203 portal vein myocytes, Ca2⫹-activated Cl⫺ currents 15–17 potassium 105–6, 372, see also K⫹ channels; agonist-stimulation intensity 256–60; human subcutaneous resistance arteries 207, 209, 210; mesenteric artery relaxation 318–24; Na⫹ pump 26; pre-eclampsia 135–6, 138; released quantification 248–55 potassium chloride, calcium signal 305–6, 308; dilatations attenuation 72, 75, 77; DOCA-salt induced hypertension 141, 144, 145; gabexate mesilate 189, 190, 192; KIR and Na⫹/K⫹-ATPase 310, 312, 313; PAR2 52, 54;

red wine polyphenolic compounds 166–7; sodium intracellular concentration 319 Pratt, P.F. 325 pre-eclampsia 132–8 pressure-induced vasodilatation 151–5 pressure myography 133–4 prostacyclin, KCa role 277, 281; vascular tone 104 prostaglandins, F2␣ 252; H2 202; H2 synthase 202; hypertension 111; local release of EDHF 333, 338 proteinase-activated receptor-2 (PAR2) 47–55 protein assays, Na,K-ATPase 30–1 protein kinase A (PKA) 29; BKCa channel ␣-subunit 364, 365; calcium sparks 45; cAMP permissive role 221; Na⫹ pump 21, 24 protein kinase C (PKC) 29; calcium sparks 45; Na⫹ pump 21, 24 protein kinase G (PKG), calcium sparks 45 protein kinases, cytochrome P450 2C 62 proteinuria, pre-eclampsia 132 Prough, R.A. 139 PSCC see poorly selective cation channels pulmonary arteries, 11,12-EET 366–70 pulmonary hypertension, K⫹ channels 9–10 pulsatile stretch 69 Randall, M.D. 132 reactive oxygen species, endothelium-dependent depolarization 200, 202; red wine polyphenolic compounds 169–70, 172 redox-sensitive mechanisms, red wine polyphenolic compounds 165–73 red wine polyphenolic compounds 165–73 related to ion channel (RIC) 29, 31–2 renal blood flow, connexin-mimetic peptides 78–86 renin-angiotensin system blockade, age-related impairment 117–23 resistance arteries see also mesenteric arteries; calcium sparks 38; gap junctions 205–10 reverse transcriptase-polymerase chain reaction (RT-PCR) 15, 17–18; CLCA1 14; clso-␤ 358; cytochrome P450 2C 57–8, 60; KIR and Na⫹/K⫹-ATPase 310–11, 312, 314, 315–16; Na,K-ATPase 30, 31–2; SKCa and IKCa 263 RIC see related to ion channel Richards, G.R. 261, 309 Riexinger, T. 349 RNA isolation, CLCA1 14; cytochrome P450 2C 57–8 Roizes, S. 297 Rp-8-bromoadenosine-3⬘,5⬘-cyclic monophosphorothioate 221 RT-PCR see reverse transcriptase-polymerase chain reaction

426

Index

Rummery, N.M. 108 Rusch, N.J. 1 ryanodine, calcium sparks 42–3 ryanodine receptors, calcium sparks 35–6, 37, 38, 39 Sakuma, I. 356 Sanchez, G. 27 Sandow, S.L. 108, 223 Santana, L.F. 35 sarcoplasmic reticulum, calcium 38, 40 Saumet, J.L. 151 scallytoxin 281 Schini-Kerth, V.B. 165, 174 Schuster, A. 248 scyllatoxin, eNOS⫺/⫺ mice 284, 289, 290, 295 serial section electron microscopy 229 shear stress 69; cytochrome P450 2C 58–9, 60–1, 62; vascular tone 106–7, 158, 163 Shebeko, V.I. 156 Shimokawa, H. 63 single channel recordings, potassium release quantification 249; SKCa and IKCa 265–8 SK1, eNOS⫺/⫺ mice 284 SK2 269–71, 272, 371; eNOS⫺/⫺ mice 284, 291, 292–4, 295–6 SK3 269–71, 272, 371, 373; eNOS⫺/⫺ mice 284, 291, 292–4, 295 SKCa see small-conductance Ca2⫹-sensitive K⫹ channels SKF96365 201 skin-pressure-induced vasodilatation 151–5 SLIGRL-NH2, mesenteric lymphatic endothelium 94, 98, 99, 100; PAR2 47–53 small-conductance Ca2⫹-sensitive K⫹ channels (SKCa) 6–8, 12, 371, 373; ascorbate 186, 187; coronary epithelium 261–73; disease 9; endothelial potassium release 251, 253; endothelial stimulation 203; gap junctions 229; human platelets 353; potassium release quantification 249–50; red wine polyphenolic compounds 171, 172 smooth muscle, basal vascular tone 193–8; Ca2⫹-activated Cl⫺ currents 13–19; calcium changes 74–5, 77; calcium signal 305–8; calcium sparks 35–46; cAMP permissive role 220–2; endothelium-dependent depolarization 202–3; estrogen effect 174–5; HEPES effects 239–47; 20-HETE 202; hydrogen peroxide as EDHF 66; hyperpolarization longitudinal spread 234–8; hyperpolarization spread 106; hypertension 108–16; membrane potential 1–12, 197–8, 216–17; mesenteric arteries 309–17; mesenteric lymphatic endothelium 93–100; mesenteric veins/arteries 300–1, 302; myoendothelial

gap junctions 223–33; Na pump 20–6; potassium channels 1–12; potassium release quantification 248–55; red wine polyphenolic compounds 169; vascular tone 103–4 sodium see also Na⫹intracellular concentration 318–24 sodium nitroprusside, age effects 119, 120; bradykinin-induced relaxations 329, 331; calcium sparks 40; DOCA-salt induced hypertension 144, 146, 149; estrogen effect 179, 180; local release of EDHF 334, 335, 339; mesenteric lymphatic endothelium 96; mesenteric veins/arteries 299, 300, 301, 302; urocortin-induced relaxations 89, 91 Sohn, H.-Y. 349 Sollini, M. 248 Solodkov, A.P. 156 Soma, T. 356 spectrofluorimetry 142 SQ22536, BKCa channel ␣-subunit 362, 363, 364 SR1417116A, pre-eclampsia 133, 135, 137, 138 Src 22, 23, 26 Src kinase 60 Stoclet, J.C. 165 store-operated channel (SOCC) 200, 201 substance P, HEPES effects 244, 246, 247; SKCa and IKCa 267–8, 272–3 sulfaphenazole 64; cytochrome P450 enzymes 68, 69; local release of EDHF 334, 336, 337, 338, 339; red wine polyphenolic compounds 169–70 superoxide anions 149; endothelium-dependent depolarization 202; human platelets 350, 355; red wine polyphenolic compounds 172 superoxide dismutase, human platelets 352; red wine polyphenolic compounds 169–70, 172 Suzuki, H. 193 systemic hypertension, K⫹ channels 10 Takano, H. 234 Takeshita, A. 63 Tare, M. 101, 199, 223 Taylor, J.A. 27 tempol, DOCA-salt induced hypertension 144, 145, 148, 149, 150 tert-butyl hydroperoxide (t-BOOH), DOCA-salt induced hypertension 144, 146, 148, 149, 150 tetrabutylammonium chloride, human platelets 352 tetraethylammonium, diabetic mice 127, 130; potassium channels 5; urocortin-induced relaxations 88, 89, 90, 92 tetrahydrobiopterin (BH4), red wine polyphenolic compounds 170–1, 172–3 thapsigargin, calcium signal 306, 307 thiocyanate, Ca2⫹-activated Cl⫺ currents 18

Index 427 Thollon, C. 261 thromboxane A2 202; analog 167, 168; endothelium-dependent depolarization 200; gabexate mesilate 192 3 H-thymidine, astrocytic EETs 343, 344, 345 tiron, red wine polyphenolic compounds 169, 172 Totoki, T. 188 transcription, Na pump 24 transduction, Na pump 20–6 transduction cascade, Na⫹ pump 25 transient receptor potential (TRP) 200, 202, 203 Triggle, C.R. 47, 124, 283 TRIZMA hydrochloride, sodium intracellular concentration 319, 322, 323–4 TRP see transient receptor potential Tsang, S.-Y. 87 D-tubocurarine 7, 272, 283–96 tyrosine kinase 62 tyrosine kinase c-Src 60 U44619 202 U46619, ascorbate 183, 184; bradykinin-induced relaxations 328, 329–30, 331; 11,12-EET role 367, 368–9; mesenteric veins/arteries 299, 301; red wine polyphenolic compounds 167, 169–70; urocortin-induced relaxations 90, 92 urocortin-induced relaxations, coronary artery 87–92 valinomycin, human platelets 351–2, 354 van Breemen, C. 274 Van de Voorde, J. 78 Vanhoutte, P.M. 239, 261, 309 vanilloid receptors, PAR2 52, 53, 55 vascular endothelial growth factor (VEGF), astrocytic EETs 342, 343, 344, 345, 346–7, 348 vascular flow 106–7 vascular reactivity, gender differences 70 vascular tone 69; diabetes 124; DOCA-salt induced hypertension 141; EDHF role in vivo 101–7; endothelium-dependent coronary 156–64; estrogen substitution 174–80; K⫹ channel regulation 2–3; local release of EDHF 335; mesenteric arterioles 193–8 vasodilatation 106; ascorbate 181–7; cyclooxygenase-independent renal 78–86; cytochrome P450 68; DOCA-salt induced hypertension 144, 148; eicosanoids 104; endothelium-dependent coronary 158; local

release of EDHF 335–7, 339; nitric oxide 102; PAR2 48, 52, 53, 54; skin-pressure-induced 151–5; urocortin-induced relaxations 87, 90–1; vascular tone 69 vasodilators, BKCa channels 40–3; calcium sparks 40–3 vasorelaxation 106–7 VEGF see vascular endothelial growth factor Vehige, L.C. 27 video-edge detection, mesenteric arterioles 195 viral constructions/infections, Na,K-ATPase 30 voltage-gated Ca2⫹ channels 2–3; basal tone 196–7, 198; calcium sparks 36, 38–9; vascular tone 103–4 voltage-gated Ca2⫹ influx 12 voltage-gated K⫹ channels (Kv) 5–6, 12; DOCA-salt induced hypertension 141, 144, 145; endothelin-1 201; endothelium-dependent depolarization 200; human platelets 353; K⫹ channel remodeling 9–10 voltage-insensitive cation channels 2 von der Weid, P.-Y. 93, 297 von Willebrandt’s factor 291, 295, 314, 316 Wang, X. 47, 274 Western blotting, HEPES effects 240–1, 244–5; SKCa and IKCa 263–4, 270 Weston, A.H. 239, 261, 309 whole cell patch-clamp method, KCa role 277; potassium release quantification 249 Wigg, S.J. 223 Wilson, W.S. 181 wire myography, agonist-stimulation intensity 257; eNOS⫺/⫺ mice 285; sodium intracellular concentration 319 xanthine oxidase, endothelium-dependent depolarization 202; red wine polyphenolic compounds 172 Yada, T. 63 Yamamoto, Y. 193 Yang, Q. 366 Yao, X. 87 Yim, A.P.C. 366 Zerr, M. 174 Zhang, C. 341 Zou, W. 366