Nitric Oxide and the Cell: Proliferation, Differentiation, and Death 9781400887538

Table of contents :
Contents
Preface
Contributors
Abbreviations
1. Nitric oxide, mitochondria and cytotoxicity
2. Nitric oxide and its congeners in mitochondria: implications for apoptosis
3. Interactions between nitric oxide and cytochrome-c oxidase in turnover
4. Nitric oxide interactions with mitochondria and oxygen metabolism
5. Resistance factors for nitrosative and oxidative stress
6. S-Nitrosylation of thiol groups in haemoglobins from various species
7. Differential inhibitory action of nitric oxide and peroxynitrite on the R2 subunit of ribonucleotide reductase
8. Hydroxyguanidines as modifiers of cell proliferation and death: a possible role for nitric oxide
9. Control of cell growth by nitric oxide: the role of cGMP-dependent and -independent events occurring at various steps of growth factor-activated signalling processes
10. Nitric oxide controls cell proliferation during Drosophila development
11. Cytoprotective actions of nitric oxide in hepatic inflammation
12. Activation of nuclear factor-κB and activator protein- I, and steady-state level of mRNA encoding tumour necrosis factor-α and inducible nitric oxide synthase in rat liver during different types of cell proliferation
13. The mechanism of nitric oxide-induced apoptosis in the pancreatic β-cell may involve cGMP and protein kinase G
14. Normal and pathological expression of nitric oxide synthase in the central nervous system
15. The role of astroglial cell-derived nitric oxide and prostanoids in death and repair mechanisms in the brain
16. Role of nitric oxide and nerve growth factor in the mechanisms of neurotoxicity induced by the HIV-1 coat protein gp l 20 in rat
17. Mechanisms of neurotoxicity induced by the HIV-1 coat protein, gp l 20, in human neuroblastoma cells in culture
18. Nitric oxide mediates differentiation and survival of neuronal cells
19. Neuromodulatory and neuroprotective actions of nitric oxide in the striatum
20. Microdialysis study in vivo of the nitric oxide/cGMP pathway in the hippocampus and cerebellum of adult and aged rats
21. Mechanisms of neuronal apoptosis elicited by glutamate or nitric oxide donors
22. Evidence that nitric oxide induces apoptotic cell death in the lateral geniculate nucleus of dark-reared rabbits
23. Nitric oxide and prostaglandins in inflammation
24. A further insight into the mechanism of lipoxygenase inhibition by nitric oxide
25. The role of nitric oxide in T-cell functions
26. CD95-induced calcium release and apoptosis are regulated by nitric oxide via a cGMP dependent pathway in human γδ T-lymphocytes
27. Nitric oxide-initiated apoptotic cell death: activating and antagonistic signalling components
28. Nitric oxide-induced apoptosis of RAW264.7 peritoneal macrophages: relationship between apoptosis and integrin expression profiles
29. Nitric oxide, p53 and cancer
30. The role of nitric oxide in Bacillus Calmette-Guérin-mediated anti-tumour effects in human bladder cancer
31. Nitric oxide priming inhibits nitric oxide-mediated apoptosis
Subject Index

Citation preview

Nitric Oxide and the Cell Proliferation, Differentiation and Death

Nitric Oxide and the Cell Proliferation, Differentiation and Death Proceedings of the Symposium held in Calabria, Italy, in September 1996, under the auspices of the British and Italian Pharmacological Societies

Editors

S.Moncada G.Nistico G.Bagetta E.A.Higgs

PRINCETON UNIVERSITY PRESS PRINCETON, NEW JERSEY

Published in North America by Princeton University Press, 41 William Street, Princeton, New Jersey 08540 Published in the United Kingdom by Portland Press Ltd, 59 Portland Place, London WIN 3AJ, U.K. Tel: (+44) 171 580 5530; e-mail: [email protected]

© 1998 Portland Press Ltd, London

ISBN 0-691-00716-0

Princeton Legacy Library edition 2017 Paperback ISBN: 978-0-691-60090-1 Hardcover ISBN: 978-0-691-65423-2 All rights reserved

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, this publication may be reproduced, stored or transmitted, in any forms or by any means, only with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency. Inquiries concerning reproduction outside those terms should be sent to the publishers at the above-mentioned address. Although, at the time of going to press, the information contained in this publication is believed to be correct, neither the authors nor the editors nor the publisher assume any responsibility for any errors or omissions herein contained. Opinions expressed in this book are those of the authors and are not necessarily held by the editors or the publishers.

Typeset by Portland Press Ltd Printed in Great Britain by Cambridge University Press, Cambridge http:/ /pup.princeton.edu

1 3 5 7 9 10 8 6 4 2

ix

Preface

xi

Contributors

xxi

Abbreviations Nitric oxide, mitochondria and cytotoxicity Richard G. Knowles and Salvador Moncada

II

Nitric oxide and its congeners in mitochon for apoptosis Christoph Richter

23

Interactions between nitric oxide and in turnover Jaume Torres and Michael T.Wilson

31

Nitric oxide interactions with metabolism Guy C. Brown

41

Resistance factors for nitrosative and Alfred Hausladen and Jonathan S. Stamler

47

S-Nitrosylation of thiol groups in haem species P. Di Simplicio, L. Lusini, F. Giannerini, D. Giustarini, Boumis, G.Amiconi and R. Rossi

60

Differential inhibitory action of nitric oxide peroxynitrite on the R2 subunit of ri Olivier Guittet, Beatrice Ducastel, Beatrice Roy

71

Hydroxyguanidines as modifiers of cell death: a possible role for nitric oxide S.A. Everett, K.Anne Smith and P.Wardman

81

Control of cell growth by nitric oxide: the dependent and -independent events steps of growth factot'-activated signalling Emilio Clementi, Clara Sciorati, Maria Riccio, I jacopo Meldolesi and Giuseppe Nistico

93

Nitric oxide controls cell proliferation development Grigori Enikolopov and Boris Kuzin

I

I

I

I I

vi

Contents

99

Cytoprotective actions of nitric oxide in inflammation Timothy R. Billiar andYoung-Myeong Kim

107

Activation of nuclear factoi'-KB and steady-state level of mRNA encoding tu factol'-a and inducible nitric oxide synth during different types of cell M. Menegazzi,A. Carcereri de Prati, M. Pibiri, G.M. Ledda-Columbano, E. Cavalieri and H.

113

The mechanism of nitric oxide-induced pancreatic 13-cell may involve cGMP and Anne C. Loweth, Gwyn T. Williams, Roger D. Scarpello and Noel G. Morgan

123

Normal and pathological expression of in the central nervous system M. Bentivoglio, G. Bertini, R. Marioui and Z.-C. ·

135 prostanoids in death and repair Vincenzo Mollace, Domenicantonio Rotiroti, Cristina Caroleo and Giuseppe Nistic6 145

Role of nitric oxide and nerve growth mechanisms of neurotoxicity induced protein gp 120 in rat G. Bagetta, L. Berliocchi, N. Costa, E. Palma, L.

ISS

protein, gp 120, in human M. Tiziana Corasaniti, Michele Navarra, Melino, Giuseppe Nistic6 and Alessandro 169

Nitric oxide mediates differentiation neuronal cells Natalia Peunova and Grigori Enikolopov

177

Neuromodulatory and neuroprotective oxide in the striatum K.M. Kendrick, R. Guevara-Guzman, L. Trabace

191

Microdialysis study in vivo of the nitric in the hippocampus and cerebellum of Ernesto Fedele and Maurizio Raiteri

14

Contents

213

Mechanisms of neuronal apoptosis elicited nitric oxide donors P. Nicotera, M. Leist, E. Bonfoco and S.A. Lipton

219

Evidence that nitric oxide induces lateral geniculate nucleus of dark-reared C. Nucci,A.M. Paoletti, S. Piccirilli, G. Bagetta, G.

227

Nitric oxide and prostaglandins in L. Sautebin,A. lalenti,A. lanaro, and M. Di Rosa

235

A further insight into the mechanism of inhibition by nitric oxide Mauro Maccarrone, M.Tiziana Corasaniti, Pietro Nistico and Alessandro Finazzi Agro

243

The role of nitric oxide in T-cell functions F.Y. Liew

247

CD95-induced calcium release and by nitric oxide via a cGMP·dependent -yo T-lymphocytes Clara Sciorati, Patrizia Revere, Marina Ferrarini Giuseppe Nistico and Emilio Clementi

257

Nitric oxide-initiated apoptotic cell death: antagonistic signalling components Bernhard Brune, Udo K. Messmer, Katrin Sandau Knethen

269

Nitric oxide-induced apoptosis of macrophages: relationship between expression profiles Raymond Judware,Thomas S. McCormick,Jong K. G. Lapetina

275

Nitric oxide, p53 and cancer S. Calmels, P. Hainaut and H. Ohshima

281

The role of nitric oxide in Bacillus Olof Jansson, Edward Morcos, Lou Brundin,Jon Adolfsson and N. Peter Wiklund

289

Nitric oxide priming inhibits nitric Jose F. Ponte and Anne E. Huot

301

Subject index

vii

II

II

-·

Nitric oxide has proven to be a molecule with wide biological significance. It is involved in myriad actions which range from physiology to pathophysiology. One of the fundamental questions in relation to its biological relevance concerns the paradoxical nature of some of its actions. There is, for example, a whole range of effects related to cytoprotection, cell proliferation, differentiation and cell death.The way in which nitric oxide becomes involved in pathophysiology is slowly being elucidated. This volume is a summary of the discussions that took place at the Second International Paraelios Symposium on Nitric Oxide research. The editors hope that the book will make a significant contribution to the clarification of this very exciting subject.

S. Moncada and E.A. Higgs The Cruciform Project, London

ix

J.Adolfsson Department of Surgery-Section of Urology, Karolinska Hospital, S-171 76 Stockholm, Sweden LAioe CNR Institute of Neurobiology, Rome, Italy G.Amiconi Department of Biochemical Sciences and CNR Center of Molecular Biology, University 'La Sapienza', Rome, Italy G.Bagetta Experimental Neurobiology Center, 'Mondino-Tor Vergata', Department of Biology, Un·iversity of Rome 'Tor Vergata', Via della Ricerca Scientifica, 00133 Rome, Italy A. Bellelli Department of Biochemical Sciences and CNR Center of Molecular Biology, University 'La Sapienza', Rome, Italy M. Bentivoglio Institute of Anatomy and Histology, Medical Faculty, Strada Le Grazie 8, 37134 Verona, Italy

L. Berliocchi Experimental Neurobiology Center, 'Mondino-Tor Vergata', University of Rome 'Tor Vergata', Via della Ricerca Scientifica, 00133 Rome, Italy

G. Bertini Institute of Anatomy and Histology, Medical Faculty, Strada Le Grazie 8, 37134 Verona, Italy

T.R. Billiar Department of Surgery, A1010 Presbyterian University Hospital, 200 Lothrop Street, Pittsburgh, PA 15213, U.S.A. E. Bonfoco Institute of Environmental Medicine, Division of Toxicology, Karolinska Institute, Stockholm, Sweden G.Boumis Department of Biochemical Sciences and CNR Center of Molecular Biology, University 'La Sapienza', Rome, Italy G.C.Brown Department of Biochemistry, University of Cambridge, Tennis Court Road, Cambridge CB21QW, U.K. xi

xii

Contributors

L.Brundin

Department of Neurology, Karolinska Hospital, Stockholm, Sweden B. Brune

University of Erlangen Nurnberg, Faculty of Medicine, Department of Medicine IV-Experimental Division, Loschgestrasse 8, 91054 Erlangen, Germany S.Calmels

Unit of Endogenous Cancer Risk Factors, International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon Cedex 08, France A. Carcereri de Prati

Istituto di Chimica Biologica, Universira di Verona, 37134 Verona, Italy M.C. Caroleo

Department of Pharmaco-Biology, Calabria University at Cosenza, Italy M.V.Catani

Department of Experimental Medicine and Biochemical Sciences, University of Rome 'Tor Vergata', Rome, Italy E. Cavalieri

Istituto di Chimica Biologica, Universita di Verona, 37134 Verona, Italy L.Cerulli

Department of Surgery, University of Rome 'Tor Vergata', Rome, Italy E. Clementi

Dip. Farmacologia, DIBIT, Scientific Institute San Raffaele, Via Olgettina 58, 20132 Milano, Italy A. Columbano

Istituto di Patologia Sperimentale, Universita di Cagliari, 09124 Cagliari, Italy M.T. Corasaniti

Department of Biology, University of Rome 'Tor Vergata', Via della Ricerca Scientifica, 00133 Rome, Italy, and Faculty of Pharmacy, University of Reggio Calabria, Catanzaro, Italy N.Costa

Faculty of Pharmacy and IBAF-CNR, Catanzaro, Italy M.Di Rosa

Department of Experimental Pharmacology, University of Naples 'Federico II', Via D.Montesano, 49-80131 Naples, Italy

Contributors

P. Di Simplicio Department of Environmental Biology, University of Siena, Siena, Italy B.Ducastel INSERM U350, Institut Curie, Centre Universitiiire, F-91405 Orsay Cedex, France P.C.Emson Department of Neurobiology, The Babraham Institute, Babraham, Cambridge CB2 4AT, U.K. G. Enikolopov Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 11724, U.S.A. S.A. Everett Gray Laboratory Cancer Research Trust, P.O. Box 100, Mount Vernon Hospital, Northwood, Middlesex HA6 2JR, U.K. E. Fedele lstituto di Farmacologia e Farmacognosia, Universidi di Genova, Viale Cembrano 4, 16148 Genova, Italy M. Ferrarini Receptor Biochemistry Unit-DIBIT and Second Department of Medicine-Laboratory of Immunology, Scientific Institute H.S. Raffaele, School of Medicine, University of Milano, Italy A. Finazzi-Agro Department of Experimental Medicine and Biochemical Sciences, University of Rome 'Tor Vergata', Via di Tor Vergata 135, I-00133 Rome, Italy F. Giannerini Department of Environmental Biology, University of Siena, Siena, Italy D. Giustarini Department of Environmental Biology, University of Siena, Siena, Italy P.Guerrier Department of Experimental Medicine and Biochemical Sciences, University of Rome 'Tor Vergata', Via di Tor Vergata 135, I-00133 Rome, Italy R. Guevara-Guzman Department of Physiology, Faculty of Medicine, UNAM, Mexico O.Guittet CNRS, URA 1116, Bat. 430, Universite Paris-Sud, F-91405 Orsay Cedex, France

xiii

xiv

Contributors

P.Hainaut Unit of Mechanisms of Carcinogenesis, International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon Cedex 08, France A. Hausladen Department of Medicine, Divisions of Respiratory and Cardiovascular Medicine, Duke University Medical Center, Durham, N C 27710, U.S.A. A.E.Huot Department of Biomedical Technology, Rowell Building-Room 302, University of Vermont, Burlington, VT 05405, U.S.A. R.D. Hurst Department of Neurology, University College London, Gower Street, London, U.K. A.lalenti Department of Experimental Pharmacology, University of Naples 'Federico II', Via D.Montesano, 49-80131 Naples, Italy A.lanaro Department of Experimental Pharmacology, University of Naples 'Federico II', Via D.Montesano, 49-80131 Naples, Italy O.Jansson Department of Surgery-Section of Urology, Karolinska Hospital, S-171 76 Stockholm, Sweden R.Judware Case Western Reserve University School of Medicine, Molecular Cardiovascular Research Center, 10900 Euclid Ave., Cleveland, OH 44106-4958, U.S.A. K.M. Kendrick Department of Neurobiology, The Babraham Institute, Babraham, Cambridge CB2 4AT, U.K. Y.-M.Kim

Department of Surgery A1010 Presbyterian University Hospital, 200 Lothrop Street, Pittsburgh, PA 15213, U.S.A., and Department of Pharmacology, University of Pittsburgh, Pittsburgh, PA 15261, U.S.A. R.G. Knowles Enzyme Pharmacology, Glaxo Well come Research, Gunnels Wood Road, Stevenage, Hertfordshire SG 1 2NY, U.K. B.Kuzin

Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 11724, U.S.A.

Contributors

E.G. Lapetina

Case Western Reserve University School of Medicine, Molecular Cardiovascular Researcl1 Center, 10900 Euclid Ave., Cleveland, OH 44106-4958, U.S.A. G.M. Ledda-Columbano

lstituto di Patologia Sperimentale, Universira di Caglian, 09124 Cagliari, Italy M.Leist

Faculty of Biology, University of Konstanz, POB 5560 X911, D-78434 Konstanz, Germany. M.Lepoivre

CNRS, URA 1116, Bat. 430, Universite Paris-Sud, F-91405 Orsay Cedex, France F.Y.Liew

Department of Immunology, University of Glasgow, Glasgow G 11 6NT, U.K. S.A.Lipton

Laboratory of Cellular and Molecular Neuroscience, Children's Hospital and Program in Neuroscience, Harvard Medical School, Boston, MA, U.S.A. A.C.Loweth

Department of Biological Sciences, Keele University, Staffordshire ST5 5BG, U.K. jon Lundberg

Department of Pharmacology, Karolinska Institute, Stockholm, Sweden L. Lusini Department of Environmental Biology, University of Siena, Siena, Italy M. Maccarrone

Department of Experimental Medicine and Biochemical Sciences, University of Rome 'Tor Vergata', Via di Tor Vergata 135, I-00133 Rome, Italy, and IDI-IRCCS Biochemistry Laboratory, University of Rome 'Tor Vergara', Rome, Italy A.A. Manfredi

Receptor Biochemistry Unit-DIBIT and Second Department of Medicine-Laboratory of Immunology, Scientific Institute H.S. Raffaele, School of Medicine, University of Milano, Italy R. Mariotti

Institute of Anatomy and Histology, Medical Faculty, Strada Le Grazie 8, 37134 Verona, Italy

xv

xvi

Contributors

T.S. McCormick

Case Western Reserve University School of Medicine, Molecular Cardiovascular Research Center, 10900 Euclid Ave., Cleveland, OH 44106-4958, U.S.A. j. Meldolesi

Consiglio Nazionale delle Ricerche Cellular and Molecular Pharmacology Center, and B. Ceccarelli Center, DIBIT-H San Raffaele Scientific Institute, University of Milano, Milano, Italy G.Melino

Department of Experimental Medicine and Biochemical Sciences, University of Rome 'Tor Vergata', Rome, Italy, and IDI-IRCCS, Rome, Italy M. Menegazzi

Istituto di Chimica Biologica, Universita di Verona, 37134 Verona, Italy U.K. Messmer

University of Erlangen Niirnberg, Faculty of Medicine, Department of Medicine IV-Experimental Division, Loschgestrasse 8, 91054 Erlangen, Germany V.Mollace

Department of Biology, University of Rome 'Tor Vergara', Via della Ricerca Scientifica, 00173 Rome, Italy S.Moncada

The Cruciform Project, University College London, 140 Tottenham Court Road, London W1P 9LN, U.K. E.Morcos

Department of Surgery-Section of Urology, Karolinska Hospital, S-171 76 Stockholm, Sweden N.G.Morgan

Department of Biological Sciences, Keele University, Staffordshire ST5 5BG, U.K. C.Muscoli

Department of Biology, University of Rome 'Tor Vergata', Via della Ricerca Scientifica, 00173 Rome, Italy M.Navarra

Faculty of Pharmacy, University of Catanzaro, Catanzaro, Italy P. Nicotera

Faculty of Biology, University of Konstanz, POB 5560 X911, D-78434 Konstanz, Germany.

Contributors

G. Nistico Department of Biology, University of Rome 'Tor Vergata', Via della Ricerca Scientifica, 00133 Rome, Italy C. Nucci Department of Surgery, University of Rome 'Tor Vergata', Rome, Italy H.Ohshima

Unit of Endogenous Cancer Risk Factors, International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon Cedex 08, France E. Palma

Faculty of Pharmacy and IBAF-CNR, Catanzaro, Italy A.M. Paoletti

Department of Biology, University of Rome 'Tor Vergata' Rome, Italy Z.-C.Peng

Institute of Anatomy and Histology, Medical Faculty, Strada Le Grazie 8, 37134 Verona, Italy N.Peunova

Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 11724, U.S.A. M.Pibiri

Istituto di Patologia Sperimentale, Universita di Cagliari, 09124 Cagliari, Italy S. Piccirilli Department of Biology, University of Rome 'Tor Vergata' Rome, Italy R.Piga

Istituto di Patologia Sperimentale, Universita di Cagliari, 09124 Cagliari, Italy

J.F. Ponte Department of Cell, University of Vermont, Burlington, VT 05405, U.S.A. M.Raiteri

Istituto di Farmacologia e Tossicologia, Universira di Genova, Viale Cembrano 4, 16148 Genova, Italy M.Riccio

Department of Pharmacology, Faculty of Pharmacy, University of Reggio Calabria, Catanzaro, Italy C. Richter

Laboratory of Biochemistry I, Swiss Federal Institute of Technology (ETH), Universitatstr. 16, CH-8092 Zurich, Switzerland

xvii

xviii

Contributors

R. Rossi Department of Environmental Biology, University of Siena, Siena, Italy D. Rotiroti Faculty of Pharmacy, Catanzaro, and IBAF Catanzaro, Italy P. Rovere Receptor Biochemistry Unit-DIBIT and Second Department of MedicineLaboratory of Immunology, Scientific Institute H.S. Raffaele, School of Medicine, University of Milano, Italy B. Roy CNRS, URA 1116, Bat. 430, Universite Paris-Sud, F-91405 Orsay Cedex, France

K.Sandau University of Erlangen Niirnberg, Faculty of Medicine, Department of Medicine IV-Experimental Division, Loschgestrasse 8, 91054 Erlangen, Germany L.Sautebin Department of Experimental Pharmacology, University of Naples 'Federico II', Via D.Montesano, 49-80131 Naples, Italy J.H.B. Scarpello Department of Medicine, Keele University, Staffordshire ST5 5BG, U.K. C.Sciorati Receptor Biochemistry Unit-DIBIT and Second Department of MedicineLaboratory of Immunology, Scientific Institute H.S. Raffaele, School of Medicine, University of Milano, Italy, Consiglio Nazionale delle Ricerche Cellular and Molecular Pharmacology Center, and B. Ceccarelli Center, DIBIT-H San Raffaele Scientific Institute, University of Milano, Milano, Italy K.Anne Smith Gray Laboratory Cancer Research Trust, P.O. Box 100, Mount Vernon Hospital, Northwood, Middlesex HA6 2JR, U.K. j.S. Stamler Department of Cell Biology, Duke University Medical Center, MSRB Rm 321, Box 2612, Durham, NC 27710, U.S.A. H. Suzuki Istituto di Chimica Biologica, Universita di Verona, 37134 Verona, Italy L.Trabace Institute of Pharmacology, Medical School, University of Bari, Italy

Contributors

J.Torres

Department of Biological Sciences (Central Campus), University of Essex, Colchester, C04 3SQ, U.K. I. Vecchio

Department of Pharmacology, Faculty of Pharmacy, University of Reggio Calabria, Catanzaro, Italy P.Wardman Gray Laboratory Cancer Research Trust, P.O. Box 100, Mount Vernon Hospital, Northwood, Middlesex HA6 2JR, U.K. N.P.Wiklund Department of Surgery-Section of Urology, Karolinska Hospital, S-171 76 Stockholm, Sweden G.T.Williams Department of Biological Sciences, Keele University, Staffordshire, STS SBG, U.K. M.T.Wilson

Department of Biological Sciences (Central Campus), University of Essex, Colchester, C04 3SQ, U.K. A. von Knethen University of Erlangen Niirnberg, Faculty of Medicine, Department of Medicine IV-Experimental Division, Loschgestrasse 8, 91054 Erlangen, Germany j.K.Yun

Case Western Reserve University School of Medicine, Molecular Cardiovascular Research Center, 10900 Euclid Ave., Cleveland, OH 44106-4958, U.S.A.

xlx

AP BCG CGC

cox DA DNQX

OTT EM ECM EGF GABA GSNO GST HOG ICE lEG IFN--y IGF IL IP IP 3 LOX

LPS MAPK L-NAME L-NMMA NFKB NGF

NMDA NO

eNOS iN OS nNOS PARP PG PH

RA sGC

SNAP SNP SOD TcR Tg TN F-a

TUNEL

activator protein Bacillus Calmette-Guerin cerebellar granule cell cycle-oxygenase dopamine 6, 7-dinitroquinoxaline-2,3-dione dithiothreitol excitatory amino acid extracellular matrix epidermal growth factor -y-aminobutyric acid S-nitrosoglutathione glutathione $-transferase hydroxyguanidine interleukin-113 converting enzyme immediate early gene interferon--y insulin-like growth factor interleukin inositol phosphate inositol 1,4,5-trisphosphate lipoxygenase lipopolysaccharide mitogen-activated protein kinase Nw-nitro-L-arginine methyl ester Nw-monomethyi-L-arginine nuclear factor KB nerve growth factor N-methyi-D-aspartate nitric oxide endothelial nitric oxide synthase inducible nitric oxide syr.thase neuronal nitric oxide synthase poly(ADP-ribose) polymerase prostaglandin partial hepatectomy rheumatoid arthritis soluble guanylate cyclase S-nitroso-N-acetylpenicillamine sodium nitroprusside superoxide dismutase T-cell receptor thapsigargin tumour necrosis factor-a deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labelling

xxi

Nitric oxide, mitochondria and cytotoxicity Richard G. Knowles* and Salvador Moncadat *Enzyme Pharmacology, GlaxoWellcome Research, Gunnels Wood Road, Stevenage, Herts. SG I 2NY, U.K., and t The Cruciform Project, University College London, 140 Tottenham Court Road, London WI P 9LN, U.K.

Introduction Nitric oxide (NO) is generated in many mammalian tissues and is an important mediator of both physiological and pathological responses [1,2]. It is synthesized by a family of three isoforms of the enzyme nitric oxide synthase (NOS, EC 1.14.13.39) [3]. These are: nNOS (also known as Type I), first identified as being expressed in neuronal cells of the central and peripheral nervous systems, and now known to be expressed also in other cell types, e.g. skeletal muscle and some epithelial cells; eNOS (also known as Type III), first identified in vascular endothelial cells; and iN OS (also known as Type II), which unlike nNOS and eNOS is not (usually) constitutively expressed but can be induced in a wide range of cell types. Again unlike nNOS and eNOS, the activity of iN OS is independent of the extracellular calcium concentration in the physiological range. The iN OS isoform, which is often referred to as 'the inducible form', has been shown to be expressed in response to a wide variety of bacterial products (e.g. lipopolysaccharide, lipotechoic acid, muramyl peptides) as well as by cytokines (e.g. interferon-)', tumour necrosis factor, interleukin (IL)-1 and IL-6, often acting in synergy) [3,4]. Activation of the transcription factor nuclear factor KB (NFKB) has been shown to play an important role in iN OS induction by these agents ([5] and references therein). Recent evidence suggests that hypoxia may also contribute to induction of iNOS [6]. A further range of agents (e.g. glucocorticoids and cytokines such as IL-13) suppress iNOS induction [1-4]. There is apparently little constitutive expression of iN OS in most normal cells and tissues although this has been reported to occur (see [2]). Although eNOS and nNOS are widely expressed in normal cells and tissues and are often collectively referred to as 'constitutive NOS', relatively recently it has become clear that they can also be induced, albeit by quite different stimuli from those that induce iN OS. Our studies in pregnant and oestrogen- and tamoxifen-treated guinea-pigs have shown that oestrogen induces both isoforms, resulting in substantial (up to 4-fold) increases in calcium-dependent NOS activity in a range of tissues [7]. The consequences of this in the brain are at present unclear, but induction of eNOS and nNOS in other tissues may be responsible for some of the changes which occur in pregnancy, e.g. redistribution of blood flow and increased gastrointestinal transit time. In addition, others have *To whom correspondence should be addressed.

2

R.G. Knowles and S. Moncada

shown that eNOS can be induced within vascular endothelial cells by shear stress [8], and that nNOS can be induced in neuronal cells and tissue, e.g. by certain proconvulsive agents [9]. It is therefore clear that the expression of all three isoforms is highly regulated. The NO generated by these enzymes has a wide variety of roles in the body, both physiological and pathophysiological [1-4,1 0] (Table 1).

Overproduction of NO in pathological conditions The isoforms of NOS are being considered as novel therapeutic targets for a range of disease states associated with an excessive production of NO [2]. Non-selective inhibition of NOS by N"'-monomethyl-L-arginine hydrochloride (LNMMA ·HCl) is currently undergoing clinical trial as a treatment for septic shock, a condition associated with a high mortality (30-50% ). Many aspects of this syndrome, particularly the severe refractory hypotension, vascular leakage and tissue damage, are believed to be a result of overproduction of NO by iN OS (induced by circulating bacterial products and cytokines), so that moderation of this overproduction by L-NMMA should help patients to survive the hypotensive crisis. If, as has been demonstrated in animals, induction of iN OS is responsible for some of these dramatic signs and symptoms of shock in humans, then it is likely that selective inhibition of the iN OS isoform will constitute an even more attractive strategy in the future, because it would leave eNOS functional and able to fulfil its protective roles in vascular homoeostasis. Induction of iNOS by cytokines has also been implicated in the pathogenesis of a range of inflammatory diseases, including arthritis, inflammatory bowel diseases, and lung diseases such as asthma [1,2]. For such conditions, it would be essential to use selective iN OS inhibitors to inhibit the overproduction of NO, in order to avoid the hypertension and other side effects of inhibition of eNOS. In the brain, activation of glutamate receptors causes an influx of calcium ions, leading to stimulation of the calcium-dependent nNOS, resulting in increases in cyclic GMP [10]. NO synthesis in the brain has been implicated in many different physiological responses (Table 1), as well as in pathological conditions such as cerebral ischaemia. The brain contains a high NOS activity compared with other tissues [11]; during cerebral ischaemia this can result in exposure of the brain to high concentrations of NO (>1 fLM [12]). Inhibition of NO synthase and genetic manipulation to 'knock out' expression of nNOS have both been shown to result in significant protection of brain tissue from ischaemic injury [13,14]. These experiments identify NO as an important agent in the tissue damage associated with excitotoxicity in this context. Interestingly, it has recently been shown that even a brief (5 min) exposure of striatal neurons to the glutamate receptor agonistN-methyl-D-aspartate (NMDA) results in prolonged (up to 16 h) generation of NO, associated with cell death which can be blocked by treatment with L-NMMA [15]. Whether such long-lasting NO production is a consequence of prolonged activation of nNOS or of induction of nNOS or iN OS remains to be established.

Table 1 nNOS Central nervous system neurotransmitter/ neuromodulator Responses to glutamate Nociception Synaptic plasticity? Peripheral nitrergic nerve neurotransmitter Gl tract Penile erection Sphincter relaxation (e.g. bladder) Blood flow

eNOS Cardiovascular system Relaxation of vascular smooth muscle: regulation of tissue conductance blood flow blood pressure Inhibition of platelet aggregation and reactivity Protection of vascular endothelium from toxic mediators (e.g. PAF) Protection of the vasculature from atherosclerosis

Pathological roles lschaemic brain damage Hyperalgesia Epilepsy? Parkinson's disease?

Modified from [42]. Abbreviation used: PAF, platelet-activating factor.? Indicates areas of particular controversy. Postulated roles for NO synthesized by the three NOS isoforms

iN OS Non-specific immunity Resistance to infection by Protozoa Fungi Bacteria Viruses? Pathological roles Shock states Septic, endotoxic Cytokine-induced Inflammatory/autoimmune diseases Acute inflammation Ulcerative colitis Asthma Transplant rejection Arthritis, prosthetic joint failure Multiple sclerosis? Dementias (Alzheimer's, Lewy body, viral)? Tumours Promotion of vascularization, growth and metastasis?

z

q· r;·

~

...a: 3

[ 0

"' .,0.. ;;:;· ~

0..

1 ;;·

~-

w

4

R.G. Knowles and S. Moncada

Mechanisms of cytotoxicity caused by NO and peroxynitrite The mechanisms of cell damage by high concentrations of NO include inhibition of a number of cellular processes, such as DNA synthesis and mitochondrial respiration [16-19]. Some of these effects may be direct and others may arise from the reaction of NO with superoxide (0 2 -)to form peroxynitrite [20]. In the brain, ischaemia results in release from compromised cells of glutamate, which then activates nNOS in susceptible partially oxygenated cells adjacent to the infarct [2,10]. Through this mechanism, damage to neuronal cells may extend from the site of the original insult over a period of several hours. The activation of the NMDA receptor by high glutamate concentrations in these oxygenated cells is also associated with formation of oxygen free radicals such as 02- [21-23]. At sites of inflammation iN OS is frequently found to be expressed in the affected tissue in cell types, such as macrophages, which are known also to produce 0 2 -. As one recent example, we have found that human macrophages express iN OS in the foreign-body inflammatory tissue around failing prosthetic joints [24]. It is therefore possible that, in addition to NO, peroxynitritc may also be produced in quantities sufficient to mediate damage to mitochondria and other targets in the cells. It has been shown that both NO and peroxymtrite can disrupt mitochondrial function [17,21,25,26] and therefore either NO or peroxynitrite could potentially be responsible for mitochondrial damage occurring in the ischaemic brain and at inflammatory sites. However, the sites at which NO and peroxynitrite interact with the respiratory chain and the mechanism of inhibition appear to be different. It has been known for some time that NO binds to cytochrome-c oxidase, the terminal member of the mitochondrial respiratory chain [27,28], but only recently was it shown that NO may act as an inhibitor of this enzyme at physiological concentrations [17]. This reaction is reversible and competitive with oxygen [29]. In contrast, peroxynitrite has little or no effect on cytochrome-c oxidase but inhibits respiratory complexes I-III [25]. Despite the potential significance of NO-dependent disruption of the mitochondrial electron-transport chain to neurodegenerative diseases and to both acute and chronic inflammatory conditions, a direct comparison of the NO and peroxynitrite-dependent inhibition of the respiratory chain had not been reported until very recently. Furthermore, it is known that peroxynitrite may react rapidly with a broad range of molecules such as thiols, low-molecular-mass antioxidants and sugars [20,30-32], any or all of which could conceivably scavenge this oxidant and prevent reaction with proteins such as the respiratory complexes. However, since the relative rates of reaction of peroxynitrite with these potential scavengers and mitochondrial electron transfer proteins were not clear, the potential of peroxynitrite as an inhibitor of mitochondrial respiration remained uncertain.

Nitric oxide, mitochondria and cytotoxicity

The effect of NO on mitochondrial respiration We have shown [19] that the addition of NO-donating mixtures [3-morpholinosydnonimine and superoxide dismutase (SIN-1 +SOD) or S-nitrosoglutathione and dithiothreitol (SNOG + DTT)] to rat brain submitochondrial particles results in the inhibition of oxygen consumption when using NADH, succinate or N,N,N' ,N' -tetramerhyl-p-phenylenediamine (TMPD)/ascorbate as substrate, consistent with inhibition of complex IV (cytochrome-c oxidase). Three observations implicate NO in these effects: (1) SIN-1 alone and SNOG alone do not inhibit respiration and do not produce significant concentrations of NO, although they do both when in the presence of SOD and DTT respectively; (2) SIN -1 + SOD and SN OG + DTT inhibit respiration at similar concentrations of NO; and (3) haemoglobin reverses the inhibition of respiration by both NOdonating mixtures. These effects of NO on respiration are consistent with those previously reported for NO from donors, aqueous solutions of NO gas or NOS [16,17,21,26]. The inhibition of NADH respiration by NO occurred with an IC 50 of 2 f.LM. Nitric oxide concentrations higher than this have been reported in the brain during cerebral ischaemia [12], suggesting that inhibition of respiration by NO is hkely to occur during such episodes, and may play a role in the neurodegeneration which results, either because of progression to irreversible inhib1tion of respiration (see below) or because of the irreversible consequences of prolonged ATP depletion. Moreover, the inhibition of cytochrome-c oxidase is competitive with oxygen and occurs through the concerted binding of NO to the oxygen binding site when the enzyme is in a redox state which is only populated during turnover [29]). One consequence of this mechanism of inhibition is that, for example, the IC50 of NO will be decreased from 2 f.LM at normal arterial oxygen concentrations to approximately 0.6 f.LM at a concentration of 50 f.LM 0 2 and to 0.1 f.LM under near-anoxic conditions. Thus, in vivo, greater inhibitory effects may occur in regions of tissue with compromised blood flow [29].

The effect of peroxynitrite on mitochondrial respiration In contrast to NO, we have found [19] that peroxynitrite inhibits respiration at complexes I-III and II-III without affecting TMPD/ascorbate respiration (dependent only on complex IV), confirming previous reports [16,25). Although SIN-1 in the absence of SOD generates peroxynitrite [31], the amounts formed from 500 f.LM SIN-1 are not sufficient to inhibit respiration during short experiments ( 1 mM) may modulate the effect of NO on respiration to some extent, and this suggests that GSH depletion might make cells more susceptible to the cytotoxicity of NO.

Nitric oxide, mitochondria and cytotoxicity

It has been proposed (e.g. [16]) that peroxynitrite causes toxicity to neuronal cells via mitochondrial dysfunction, since its addition to neurones in primary culture resulted in the inhibition of succinate-cytochrome-c reductase and release of lactate dehydrogenase into the culture medium. However, our observation that direct effects of peroxynitrite are suppressed by cellular constituents such as GSH suggests that this effect of peroxynitrite on cellular respiration is most likely to be indirect, either by generation of NO or by other effects, perhaps initially at the plasma membrane.

NO may cause irreversible damage to mitochondria via peroxynitrite The accumulated data indicate that the direct effect of NO itself on respiration is at complex IV, with complexes I-III functioning normally. Thus, the initial effect of NO on the respiratory chain would be the reversible inhibition of mitochondrial respiration at cytochrome-c oxidase. However, 0 2 - is a by-product of mitochondrial respiratory electron transport and its production increases in the presence of mitochondrial inhibitors such rotenone, antimycin A and cyanide [40]; the site of this additional 0 2 - formation is believed to be within complexes I and III. It is therefore not surprising that NO itself has recently been demonstrated to stimulate 0 2 - production by the respiratory chain [41]. Thus a sustained production of NO, together with this resulting formation of 0 2 - (and perhaps also the depletion of antioxidant defences) may, because of formation of peroxynitrite within the mitochondria, cause irreversible inhibition of complexes I-III, leading to cell death. This mechanism could contribute significantly to tissue damage during ischaemia in the brain and following iNOS induction in a range of cells and tissues. However, if peroxynitrite is not formed in the immediate vicinity of a redox site withm the respiratory chain, it is unlikely to inhibit respiration directly unless produced in such large quantities that it overwhelms the thiol! carbohydrate defences. This mechanism could also reconcile the apparent contradiction between the data [17,19,21] showing that NO does not directly affect complexes I and II, and the observations which show irreversible inhibition of these complexes in cells exposed to NO, e.g. from cytotoxic-activated macrophages [18].

Conclusions NO is a biologically active molecule which may, through an effect on cytochrome-c oxidase, exert physiological control on cellular respiration. More intense or longer-term inhibition of this enzyme may lead to generation of 0 2 - in the mitochondria and production of peroxynitrite. This could result in irreversible inhibition of complexes I-III of respiration and therefore cytotoxicity. The details of this sequence of events have not been clarified, but it does provide a working hypothesis for the mechanism of cell death and tissue damage caused by exposure

7

8

R.G. Knowles and S. Moncada

to high concentrations of NO during acute or chronic inflammation or during ischaemic brain damage.

We are indebted to our friends and colleagues with whom this work was carried out: Victor Darley-Usmar, Ignacio Lizasoain, Maria-Angeles Moro and Lu Tao. References 1. 2. 3. 4. 5. 6.

7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.

Moncada, S., Palmer, R.M.J. and Higgs, E.A. (1991) Pharmacal. Rev. 43,109-142 Moncada, S. and H1ggs, E.A. (1995) FASEBJ. 9,1319-1330 Knowles, R.G. and Moncada, S. (1994) Biochem.J 298,249-258 Nathan, C. (1992) FASEBJ. 6, 3051-3064 Xie, Q.-W., Kashiwabara, Y. and Nathan, C. (1994) J. Bioi. Chern. 269, 4705-4708 Melillo, G., Musso, T., Sica, A., Taylor, L.S., Cox, G.W. and Varesio, L. (1995)]. Exp. Med. 182,1683-1693 Weiner, C., Lizasoain, I., Baylis, S.A., Knowles, R.G., Charles, I. G. and Moncada, S. (1994) Proc. Nad. Acad. Sci. U.S.A. 91,5212-5216 Nishida, K., Harrison, D.G., Navas, J.P., Fischer, A.A., Dockery, S.P., Uematsu, M., Nerem, R.M., Alexander, R.W. and Murphy, T.J. (1992) J. Clin. Invest. 90, 2092-2096 Bagetta, G., Massoud, R., Rodino, P., Federici, G. and Nistico, G. (1993) Eur. J. Pharmacal. 237,61-64 Garthwaite,]. and Boulton, C.L. (1995) Annu. Rev. Physiol. 57, 683-706 Salter, M., Knowles, R.G. and Moncada, S. (1991) FEBS Lett. 291, 145-149 Malinski, T., Bailey, F., Zhang, Z.G. and Chopp, M. (1993) J. Cereb. Blood Flow Meta b. 13, 355-358 Huang, Z., Huang, P.L., Panahian, N., Dalkara, T., Fishman, M.C. and Moskowitz, M A. (1994) Sctence265, 1883-1885 Nowicki, J.P., Duval, D., Poignet, H. and Scatton, B. (1991) Eur.]. Pharmacal. 204, 339-340 Strijbos, P.J.L.M., Leach, M.J. and Garthwaite,]. (1996)]. Neurosci. 16,5004-5013 Bolanos, J.P., Heales, S.J.R., Land, J.M. and Clark, J.B. (1995) ]. Neurochem. 64, 1965-1972 Cleeter, M.W.J., Cooper, J M., Darley-Us mar, V.M., Moncada, 5. and Schapira, A.H.V. (1994) FEBS Lett. 345, 50-54 Hibbs Jr., J.B., Tamtor, R.R., Vavnn, Z., Granger, D.L., Drapier, J.C., Amber, I.J and Lancaster Jr., J.R. (1990) in Nitric Oxide from L-Arginine: A Bioregulatory System (Moncada, S. and Higgs, E.A., eds.), pp. 189-223, Elsevier, Amsterdam L1zasoain, I., Moro, M.A., Knowles, R.G., Darley-Usmar, V. and Moncada, S. (1996) Biochem.J. 314,877-880 Beckman, J.S., Beckman, T. W., Chen, J ., Marshall, P.A. and Freeman, B.A. ( 1990) Proc. Nat!. Acad. Sci. U.S.A. 87,1620--1624 Brown, G.C. and Cooper, C.E. (1994) FEBS Lett. 356,295-298 Lafon-Cazal, M., P1etri, 5., Culcasi, M. and Bockaert, J. (1993) Nature (London) 364,535-537 McCord,J.M. (1985) N. Engi.J. Med. 312, 159-163 Moilanen, E., Moilanen, T., Knowles, R., Charles, I., Kadoya, Y., Al-Saffar, N., Revell, P.A. and Moncada, S. (1997) Am. J. Pathol. 150, 881-887 Radi, R., Rodriguez, M., Castro, L. and Tellen, R. (1994) Arch. B10chem. Biophys. 308, 89-95 Schweizer, M. and Richter, C. (1994) Biochem. Biophys. Res. Commun. 204, 169-175 Tiesjema, R.H. and Van Gelder, B.F. (1974) Biochim. Biophys. Acta 347,202-214 Brudvig, G.V., Stevens, T.H. and Chan, S I. (1980) BIOchemistry 18,5275-5285 Torres,}., Darley-Usmar, V. and Wilson, M.T. (1995) Biochem.J. 312,169-173 Radi, R., Beckman,J.S. and Freeman, B.F. (1991)]. Bioi. Chern. 266,4244-4250 Hogg, N, Darley-Usmar, V.M., Wilson, M.T. and Moncada, S. (1992) Biochem. J. 281, 419-424 Vander Vliet, A., Smith, D., O'Neil, C.A., Kaur, !.I., Darley-Usmar, V., Cross, C.E. and Halliwell, B. (1994) Biochem.]. 303,295-301 Knowles, R.G., Lu, T. and Moncada, S. (1996) in The Biology of Nitric Oxide, Part 5 (Moncada, S., Stamler, J., Gross, S. and Higgs, E.A., eds.), p. 96, Portland Press, London Borutaite, V and Brown, G.C. (1996) Biochem.]. 315,295-299 Brudvig, O.W., Stevens, O.H. and Chan, 0.1. (1980) Biochemistry 19,5275-5285 Clarkson, R.B., Norby, S.W., Smirnov, A., Boyer, 5., Vahidi, N., Nims, R.W. and Wink, D.A. (1995) Biochim. Biophys. Acta 1243, 496-502 Moro, M.A., Darley-Usmar, V.M., Goodwin, D.A., Read, N.G., Zamora-Pino, R., Feelisch, M., Radomski, M.W. and Moncada, S. (1994) Proc. Nat!. Acad. Sci. U.S.A. 91, 6702-6706

Nitric oxide, mitochondria and cytotoxicity

38.

39. 40. 41.

42

Moro, M.A., Darley-Usmar, V.M., Lizasoain, I., Su, Y., Knowles, R.G., Radomski, M.W. and Moncada, S. ( 1995) Br. J. Pharmacal. 116, 1999-2004 Whiteman, M., Tritschler, H. and Halliwell, H. (1996) FEBS Lett. 379,74-76 Turrens,J.E and Boveris, A. (1980) Biochem. J. 191,421-427 Poderoso, J.J., Carreras, M.C., Lisdero, C., Riobo, N., Schopfer, F. and Boveris, A. (1996) Arch. Biochem. Biophys. 328, 85-92 Knowles, R.G. (1996) Biochem. Soc. Trans. 24, 875-878

9

Nitric oxide and its congeners in mitochondria: implications for apoptosis Christoph Richter Laboratory of Biochemistry I, Swiss Federal Institute ofTechnology (ETH), Universitatstr. 16, CH-8092 Zurich, Switzerland

Cell Ca2+ and mitochondria Cellular Ca2+ homoeostasis Intracellular Ca 2 + regulates many processes. Its concentration is adjusted by binding to non-membranous proteins, by mitochondria, and by membranebound Ca2+-ATPases located primarily in the plasma, nuclear and endoplasmic reticular membrane [1]. Mitochondria contain Ca 2+-sensitive targets regulated by moderate Ca2 + transients. These organelles are also able to take up large amounts of Ca 2 + and buffer the cytosolic Ca 2+. They thereby act as safety devices against potentially toxic increases of cytosolic Ca1+ [2]. Mitochondria take up and release Ca2+ by separate routes. As a consequence, Ca2+ is 'cycled' across their inner membrane [1]. The importance of mitochondria as short-term modulators of cytosolic Ca2+ under physiological conditions was until recently considered minor. However, there is now compelling evidence [3,4] that, during physiological cell stimulation, mitochondrial CaH transport directly participates in the modulation and maintenance of cellular CaZ+ homoeostasis. Thus, mitochondria are of central importance for physiological Ca2+ handling. They act as a reservoir for Ca2 +, provide much of the ATP used by Ca2 +-ATPases, and regulate with Ca2+ the activity of intramitochondrial dehydrogenases as well as nucleic acid and protein synthesis [5]. Reactive oxygen and other pro-oxidants in mitochondria Reactive oxygen species (ROS) such as superoxide radical (0 2 -), hydrogen peroxide, hydroxyl radical and singlet oxygen are physiological metabolites [6]. Mitochondria consume about 90% of the body's oxygen, and are a particularly rich source of ROS since about 1-2% of oxygen metabolized by mitochondria are normally converted to 0 2 -. A rat liver mitochondrion produces physiologically about 3 X 10 7 0 2 - molecules per day [7]. Additional ROS are formed in mitochondria exposed to certain compounds such as tumour necrosis factor-a (TNF-a) or azidothymidine, used in treatment of AIDS, and in some pathological states, e.g. ischaemia/reperfusion [8].

II

I2

C. Richter

Mitochondrial 0 2 - and hydrogen peroxide (or organic peroxides) are metabolized by the Mn-containing superoxide dismutase (SOD) and theSecontaining glutathione peroxidase, respectively. They, together with glutathione reductase and the energy-linked transhydrogenase, connect ROS and organic hydroperoxides with the redox state of mitochondrial pyridine nucleotides [9].

Ca2+ release from mitochondria In principle, Ca 2 + can leave mitochondria in three ways: by non-specific leakage through the inner membrane or by Na +-dependent or -independent Ca2 +-specific release pathways [5]. The latter two operate when the mitochondrial membrane potential (ll Y) is high. Collapse of Ll Y due to, for example, inhibition of respiration leads to non-specific release of Ca2 + from mitochondria. The Na+ -independent specific pathway is linked to the redox state of mitochondrial pyridine nucleotides. Compounds causing their oxidation ('prooxidants') and the hydrolysis of oxidized pyridine nucleotides promote Ca2+ release from intact mitochondria. Valid criteria for the intactness of mitochondria during Ca2+ release are the maintenance of inner membrane integrity as determined by ll Y measurements and the demonstration of the specificity of Ca2 + movement across the inner membrane. The pyridine nucleotide-linked Ca2+ release from mitochondria meets these criteria, and is accompanied by protein mono-ADP-ribosylation (see below). Mitochondrial Ca2 + in cell death Attention had been drawn to Ca2+ -induced cell death many years ago [1 OJ. Excessive intracellular Ca2 + is thought to contribute to a final common pathway of cytotoxic events leading to ROS formation, necrosis or apoptosis. These events include over-activation of protein kinase C, Ca 2+f calmodulin-dependent protein kinase II, phospholipases, proteases, protein phosphatases, xanthine oxidase, endonucleases and nitric oxide synthase (NOS). Although the exact role of Ca2 + in cell killing is unclear, a disturbance of mitochondrial Ca 2 + handling can be fatal. The ll!lrmal Ca 2 + 'cycling' across the inner mitochondrial membrane requires little energy [1]. However, when the Ca2+ release pathway is stimulated by pro-oxidants, 'cycling' may become excessive and lead to loss of ll Y, general leakiness of the inner mitochondrial membrane, inhibition of ATP synthesis, mitochondrial damage and cell death [2]. Compatible with this concept, cyclosporin A (CSA), an inhibitor of prooxidant-induced Ca2+ release from mitochondria, protects against loss of cell viability induced by pro-oxidants [11] or by nitric oxide (NO) [12], and favourably alters liver mitochondrial functions in the post-ischaemic phase at the organ level [13].

Apoptosis General considerations: activation and execution Apoptosis is an evolutionarily conserved form of physiological cell death important for tissue development and homoeostasis. Its hallmarks are distinct

Nitric oxide and its congeners in mitochondria

morphological alterations such as nuclear condensation, cell shrinkage and bleb formation, and the absence of inflammatory responses of the affected tissue. Deranged apoptosis plays a major role in diseases such as cancer, AIDS, autoimmune diseases and neurodegeneration [14]. The program for apoptotic cell death appears to be present constitutively in virtually all mammalian cells and can be activated by a variety of extraand intra-cellular signals. Although the various steps and biochemical mechanisms participating in apoptosis are not completely understood, it is clear from genetic studies in lower organisms and comparative investigations in mammalian systems that apoptosis generally comprises four distinct stages, namely the decision to die, the execution of death, the engulfment of dead cells or fragments thereof and their degradation [15].

Mediators of apoptosis Several conditions, molecules or organelles, such as oxidative stress, ROS, Ca 2 +, proteases or mitochondria, are considered mediators of apoptosis, but at present it is not always clear whether they are required for or are the consequence of apoptosis. This is due to our incomplete knowledge of the cellular machinery and the timing of the various stages of apoptosis. Also, studies with reconstituted systems have shown that characteristic apoptotic responses occur in isolated nuclei as well as in nucleus-free cell fractions. Thus, it is conceivable that some of the conditions or molecules are required to direct the cell to undergo apoptosis, whereas others may be required for its execution. There may also exist cell typespecific requirements. Our understanding of apoptosis IS further limited by the fact that until recently most studies were done on cell batches, i.e. they relied on bulk measurements. Unless cells are in perfect synchrony, such measurements conceivably hide cause-effect relationships. There is ample evidence that apoptosis is accompanied by oxidative stress, and antioxidants can delay or prevent apoptosis [16). A valuable tool used to elucidate the importance of oxidative stress is the proto-oncogene, bcl-2 (see also below), which stimulates an antioxidative response in cells and prevents apoptosis [17,18]. The importance of oxidative stress as a cause for apoptosis was questioned because cells can undergo bcl-2-inhibitable apoptosis under anaerobic conditions as well [19,20], which preclude ROS formation. The requirement of Ca 2 + for apoptosis is also controversial [21]. Early reports suggested that a rise of the intracellular CaZ+ leads to apoptosis via endonuclease activation, and more recent work indicated that apoptosis is accompanied by shifts of CaZ+ between various intracellular pools. It is worth noting that cellular Ca2+ handling and ROS production are related. Thus, increased 'cycling' of Ca2+ by mitochondria stimulates ROS production (see below). At present, the role of proteases in apoptosis receives much attention [22]. Gene analysis in Caenorhabditis elegans identified a key pro-apoptotic gene, ced-3, which encodes a putative cysteine protease that is related to the mammalian interleukin (IL)-113-converting enzyme (ICE). ICE-related proteases cleave poly(ADP-ribose) polymerase, which results in the activation of the Ca2 +/Mg2+dependent endonuclease implicated in internucleosomal DNA cleavage character-

13

14

C. Richter

istic of apoptosis. Protease inhibitors also inhibit apoptosis. This suggests that pro teases are important for the initiation of apoptosis.

Bel-l links oxidative stress, Ca2+ and D..Y to apoptosis Given that expression of bcl-2 elicits an antioxidative response in cells, what are the biochemical mechanism(s) by which bcl-2 prevents apoptosis? We showed (see later) that one mechanism is the prevention of ROS-induced mitochondrial Ca2 + cycling, a process which results in a collapse of 8 Y and in cellular ATP depletion. Thus, bcl-2 prevents disturbances of the cellular Ca2+ homoeostasis and ROS production at the mitochondrial level. Another mechanism could relate to a shift of the NADH/NAD+ ratio in favour of the reduced form: bcl-2 prevents Ca2+ release from the endoplasmic reticulum via the inositol trisphosphate-insensitive pathway [23], known to be stimulated by NAD+-derived cyclic ADP-ribose [24). It is conceivable that bcl-2 prevents this release by shifting NAD+ to NADH, thereby preventing the formation of cyclic ADP-ribose.

Are mitochondria (or parameters which they control) essential for apoptosis? As mentioned above, ROS and Ca2+, two entities presumably important for apoptosis, are linked to mitochondria. These two entities are, under certain conditions, even directly related: ROS stimulate a specific Ca2+ release from mitochondria, and when mitochondria 'cycle' Ca2+ excessively their ROS production increases [25-27]. Accordingly, TNF-a stimulates ROS production by mitochondria, and the TNF-a-induced apoptotic killing of at least some cell types is due to ROS-induced mitochondrial Ca2+ cycling since apoptosis is prevented by mitochondrial Ca2+ uptake inhibitors. Mitochondria change their structural and functional properties, particularly those of the respiratory chain, during apoptosis. Whether these changes are required for, or are secondary to, apoptosis is presently unclear. The suspected requirement of mitochondria for apoptosis was recently investigated more directly. In one study cells harbouring respiratory chain-deficient (p 0) mitochondria were capable of apoptosis and it was concluded that mitochondria are not required [28). However, this conclusion is flawed. It is very likely that such cells up-regulate glycolysis and maintain high ATP levels and it has been shown [29-31] that their mitochondria maintain a high 8 Y, presumably at the expense of intramitochondrial ATP hydrolysis [31). The other study used isolated nuclei in combination with various cell fractions. In this reconstituted system, which is believed to be a valid model for apoptosis, mitochondria were indeed required for apoptosis [32]. The exact function of mitochondria was not established, however, and it was postulated that they may produce a hitherto unidentified factor necessary for apoptosis (see below). A clue as to which mitochondrial parameter(s) may be important for the control of apoptosis comes from the observation that the stabilization of 8 Y, either by overexpression of bcl-2 or by the ionophore nigericin, prevents apoptosis induced by TNF-a.

Nitric oxide and its congeners in mitochondria

ATP levels as a gauge, or mitochondria as the central switchboard, for apoptosis The biochemical and molecular mechanisms that mediate apoptosis are poorly understood. However, there is a general agreement that apoptosis is an active, energy-requiring process. We proposed [33) that the cellular ATP level is an important determinant for cell death, either by apoptosis or necrosis. This hypothesis is supported by circumstantial evidence, is consistent with most available data, has a corollary in ageing and is amenable to direct experimental testing. We argue that a cell stays alive as long as a certain ATP level is maintained. When ATP falls below this level apoptosis ensues provided enough ATP is still available for energy-requiring apoptotic processes such as enzymic hydrolysis of macromolecules, nuclear condensation and bleb formation. Only when there is a severe drop in cellular ATP does controlled cell death cease and necrosis ensue. A decreased cellular ATP level is characteristic of cell death, but there has been no systematic investigation into whether the decrease IS the cause or the consequence of cell death. However, it has been shown with six human and murine leukaemia/lymphoma cell lines [34] that the expression of the bcl-2 gene and the cellular energy status are highly correlated and that both parameters are inversely related to the sensitivity for glucocorticoid-induced apoptosis. An ADP/ ATP ratio of about 0.2 was the critical discriminator between survival and apoptosis in all cell types. Also, acute inhibition of the mitochondrial respiratory chain with rotenone or antimycin A induces apoptosis in cells carrying normal mitochondria, but not in p0 cells [35). This was taken as support for the argument that active mitochondria are required for apoptosis and it was suggested that cells with impaired mitochondrial energy metabolism reach an energy threshold which triggers apoptosis. It was similarly argued that ageing and age-related diseases are caused by a fall of the energy charge below a threshold [36]. A critical evaluation of whether mitochondria are important for apoptosis and possibly a hint as to which mitochondrial parameter(s) are significant in it, could come from the analysis of apoptosis, 11 Y, mitochondrial Ca2+ handling and ROS production and the cellular ATP level at the single-cell level and in selected cell populations. It is possible to determine apoptosis, necrosis and 11 Y quantitatively at the single-cell level by fluorescence-activated cell sorting [37,38]. Using this novel technology it was shown with dexamethasone-treated rat thymocytes, a classical model of apoptosis, that mitochondria are functionally intact during apoptosis, when DNA fragmentation already occurs, whereas alterations in their potential and mass take place after DNA damage. Loss of plasma membrane integrity, which indicates necrosis, eventually follows [38). Similar studies should now be performed with other cell types and apoptosis inducers and should be paralleled by ATP determinations. ATP cannot be measured in single cells, but sorting of cells into normal, apoptotic and necrotic subpopulations followed by analysis of their ATP content would satisfactorily circumvent this problem.

Damaged mitochondria release an apoptosis-inducing factor Recently it was shown with a reconstituted, cell-free system that damaged mitochondria release a (proteinaceous?) factor which causes apoptosis [39].

IS

16

C. Richter

Damage was provoked by a host of agents such as Ca2 +, ROS, uncoupler and thiol reagent, which had prevwusly been found to cause apoptosis in cells also. Mitochondrial damage, and therefore apoptosis also, were inh1b1ted by expression of bcl-2. These important observations strengthen the suggestion that destabilization and stabilization of mitochondria result in cell death and cell survival, respectively.

Nitric oxide (NO) Overview NO is a biologically active molecule that is currently the subject of much research. Note that here NO indicates nitric oxide independent of its redox state, whereas NO" and NO- (see later) refer to the nitrogen monoxide radical and nitroxyl anion, respectively. NO is synthesized by NO synthase (NOS). The family of NOSs [40,41] consists of three isoenzymes, the neuronal constitutive (nNOS), the endothelial constitutive (eNOS) and the inducible (iN OS) isoforms. The constitutive isoforms are strictly Ca 2 + -dependent, whereas iN OS is not, although calmodulin is necessary for its activity. Activation of both constitutive enzymes occurs via small Ca2+ -transients, whereas iN OS, once expressed, stays active as long as substrate is available. The constitutive isoforms show a typical interaction with calmodulin controlled by Ca2+. In contrast, iNOS forms a tight complex with calmodulin at very low Ca 2 + concentrations and thus appears to be Ca2 +independent. The location of NOS in cells is not clear. The eNOS is Nmyristoylated, which targets it to the Golgi [42], while nNOS and iN OS seem to be cytosolic proteins. nNOS in skeletal muscle was recently shown to be membrane- and possibly mitochondria-associated [43), and recent immunochemical studies showed iN OS associated with the inner membrane of rat liver and brain mitochondria [44]. NO mediates beneficial responses such as maintenance of blood pressure or destruction of foreign invaders in the immune response and is probably of major importance in long-term memory. However, NO can also inhibit T-cell proliferation, be tumoricidal, inhibit viral replication in macrophages, suppress protein synthesis and mediate cell lysis by multiple mechanisms [40,41]. The range of actions of NO is particularly evident from studies of the brain, where NO either acts as a second messenger in the central nervous system, or protects or kills N-methyl-o-aspartate (NMDA)-respons1ve neurons (45,46]. Ca 2+ is an important factor in the killing of neurons (47] although the mechanisms underlying neuronal cell death are not clear. However, many findings point to a central role of mitochondrial Ca 2 + 'cycling'. First, death is favoured by a pro-oxidant state, as is evident from its stimulation by peroxides, free radicals, NO, defective SOD and deprivation of nerve growth factor (NGF) (NGF induces catalase). Second, death requires Ca2 + but an increase in cytosolic Ca 2 + is not always evident, which points to an event distal to Ca 2 + entry. Third, the effective cellular antioxidant bcl-2, localized in mitochondria and other membranes, is cytoprotective [17,48}. As mentioned, bcl-2 stabilizes 6. Y and the

Nitric oxide and its congeners in mitochondria

cellular ATP level, and rescues cells from apoptotic death triggered by prooxidant-induced Ca2+ 'cycling'. The many actwns of NO are in part due to the broad array of derived redox species with distinctive properties and reactivities: NO+ (nitrosonium), NO" (itself a free radical) and NO- (nitroxyl anion) [49,50], and the ability of NO to combine with 0 2 - to yield peroxynitrite (ONOO-) [51]. ONOO- is an efficient oxidant of thiols [52]. Its formation in vivo was recently shown in macrophages and other cells. ONOO- production is associated with the activation and expression of iN OS and implicated in the pathophysiology of diseases such as acute endotoxaemia, inflammatory bowel disease, neurological disorders and atherosclerosis. Inhibition by SOD, the 0 2 - scavenger, of NO-mediated cytotoxicity suggests that ONOO- may contribute to the NOmediated biological effects [53]. NO- is formed from NO by reduced SOD [54]. It is another NO congener which oxidizes thiols. NO- is, like ONoo-, neuroprotective at NMDA-receptors because these compounds lead to disulphide formation at critical thiols of the redox modulator site of the receptor [55].

NO and ONOO- in apoptosis It was recently shown in several systems that NO can cause apoptosis [56-59]. Thus, NO mediates apoptosis in murine macrophages and activated murine macrophages can induce apoptosis in tumour cells through NO-dependent orindependent mechanisms. NO-induced apoptosis can be the consequence of DNA damage and subsequent expression of the tumour suppressor gene,p53 [59]. However, according to Mannik eta!. [60], NO inhibits apoptosis in lymphocytes. Recently it was found that ONOO- also induces apoptosis in a time- and concentration-dependent manner [53] and that, depending on the concentration of ONOo-, cells die by either apoptosis or necrosis [61].

NO and the regulation of cytochrome oxidase The most-cited and best-understood physiological target of NO is guanylate cyclase. NO binds to it and stimulates It, and thus controls cell functions via cGMP, cGMP-gated channels, cGMP-dependent protein kinases and phosphodiesterases. However, NO also binds to cytochrome oxidase and reversibly inhibits mitochondrial respiration by competing with oxygen for this enzyme [62,63]. Concentrations of NO measured in a range of biological systems are Similar to those shown to inhibit cytochrome oxidase and mitochondrial respiration, and inhibition of NO synthesis results in a stimulation of respiration in many systems. Therefore, it was recently proposed that NO exerts a good part of its physiological and pathological effects on cells by inhibiting cytochrome oxidase [64].

Presence of NOS in mitochondria As mentioned above, there is evidence that NOS is associated with mitochondria [43,44]. If the enzyme were stimulated by Ca 2 + and located in the matrix or on the inner side of the inner mitochondrial membrane, this would provide a selfregulating system for mitochondrial Ca2+ homoeostasis in which CaZ+ uptake by mitochondria would lead to NO formation. NO could then promote Ca2 + release

17

18

C. Richter

by collapsing Ll Y via inhibition of cytochrome oxidase. Alternatively, NO could combine with 0 2 - and form ONOO-, which could then stimulate CaZ+ release from mitochondria with maintenance of Ll Y.

Regulation of Ca2+ release from mitochondria by NO and its congeners Pyridine nucleotide-linked Ca2 + release CaZ+ release from mitochondria was first associated with the oxidation of the mitochondrial pyridine nucleotides by Lehninger's group [65], who showed that enzymic oxidation of NAD(P)H by acetoacetate or oxaloacetate promotes release whereas reduction of NAD(P)+ by [3-hydroxybutyrate prevents it. Shortly thereafter we reported [66,67] that hydroperoxides such as t-butyl hydro peroxide or hydrogen peroxide promote mitochondrial pyridine nucleotide oxidation followed by hydrolysis ofNAD+ to ADP-ribose and nicotinamide and by CaZ+ release. Since then, many pro-oxidants have been identified stimulating Ca2 + release from intact liver, heart, brain and kidney mitochondria secondarily to pyridine nucleotide oxidation and hydrolysis followed by protein mono-ADPribosylation [2,5,8]. CSA and its derivatives are useful for the study of mitochondrial Ca 2 + handling. We have tested a number of them as inhibitors of pyridine nucleotide hydrolysis, Ca2+ release and the matrix-located pept1dyl-prolyl cis-trans isomerase [68]. There is an impressive positive correlation between the extent of inhibition of these three parameters by the different cyclosporin derivatives. This strongly suggests that the ADP-ribose-dependent Ca2 + release engages peptidylprolyl cis-trans isomerase, further documenting the specificity of this pathway. Hypoxia/reperfusion injury is characterized by an increased cytoplasmic CaZ+ content and an increased ROS production in mitochondria and the cytosol [69]. We have shown [70] that CSA protects mitochondria in a model in vitro of hypoxia/reperfusion injury against loss of complex I activity by preventing pyridine nucleotide hydrolysis and CaZ+ 'cycling'.

NAD+ hydrolysis is under the control of vicinal thiols NAD+ hydrolysis, and therefore ADP-ribosylation and Ca2+ release, are under the control of vicinal thiols. Phenylarsine oxide, which reversibly forms a fivemembered ring with vicinal thiols, promotes the CaZ+ -dependent intramitochondrial NAD+ hydrolysis and thereby the specific Ca2 + release [71]. Gliotoxin, a fungal metabolite carrying a disulphide moiety, also promotes the Ca2+dependent intramitochondrial NAD+ hydrolysis and thereby the specific Ca2+ release, but is inactive when its disulphide is reduced or methylated [72]. Thus, intramitochondrial, Ca2 +-dependent NAD+ hydrolysis is prevented when some vicinal thiols are in the SH form, and occurs when they are connected, either by a cross-linking reagent or by oxidation to the disulphide form.

Nitric oxide and its congeners in mitochondria

NO causes Cal+ release from mitochondria NO at sub-micromolar, physiologically relevant, concentrations potently 'deenergizes' isolated mitochondria [62]. De-energization is observed when mitochondna utilize respiratory substrates such as pyruvate plus malate, succinate or ascorbate plus tetramethylphenylenediamine, but not when mitochondria are energized with ATP, and is due to a transient inhibition of cytochrome oxidase. The extent and duration of de-energization is determined by the concentration of NO and oxygen, and the type of respiratory substrate. The NO-induced changes of the mitochondrial energy state are transient and are paralleled by release and reuptake of mitochondrial Ca 2 +. Importantly, cytochrome oxidase is particularly sensitive to NO at oxygen concentrations below 30 fLM [63], i.e. at intracellular oxygen tensions. These findings reveal a direct action of NO on the mitochondrial respiratory chain and suggest that NO exerts some of its physiological and pathological effects by de-energizing mitochondria. NO a:lso de-energizes mitochondria in freshly prepared hepatocytes [12). De-energization is reversible at low NO concentrations, but longer-lasting at higher ones. The drop and the recovery of d Yare accompanied by a rise and fall of cytosolic Ca 2 + levels. NO at higher concentrations, provided by nitrosoglutathione in combination with dithiothreitol (GSNO/DTT), kills hepatocytes. Killing is reduced when the cytosolic Ca 2 + is chelated, or when Ca 2 + 'cycling' by mitochondria is prevented by CSA. We conclude that NO can kill cells by releasing Ca2+ from mitochondria and thereby flooding the cytosol with Ca2+.

Mitochondrial Ca2 +, pro-oxidants and apoptosis We showed [26] that TNF-a kills L929 cells apoptotically by stimulating ROS production and Ca2+ 'cycling' in mitochondria. A decrease in mitochondrial enzyme activities and ATP production preceded cell death. TNF-a-induced mitochondrial ROS production in these cells was recently confirmed [73). L929 cells overexpressing bcl-2 have an increased .1 Y [74]. This stabilizes mitochondria and protects L 929 cells from apoptosis caused by the ROS-induced mitochondrial Ca2+ 'cycling' and increased cytosolic Ca 2 + levels. Also the ionophore nigericin, which increases d Y, protects the cells [74]. Based on these and other findings we suggested [75] that a pro-oxidant-induced Ca2 + release from mitochondria, followed by CaZ+ 'cycling' and ATP depletion, is a common cause of apoptosis. Accordingly, maintenance of d Y stabilizes mitochondria and thereby prevents apoptosis. Bcl-2 thus provides the link between the antioxidant defence system, Ca2 + and d Y [76]. In this context it is interesting to recall that many carcinoma cells have an increased .1Y l77].

ONOO- stimulates the specific mitochondrial Ca2 + release pathway Since ONOO- oxidizes thiols, and since vicinal thiols control the specific mitochondrial Ca2+ release pathway, we tested whether ONOO- is able to activate it. We found [78] that ONOO- indeed induces Ca2+ release from rat liver mitochondria. This release occurs (i) with preservation of d Y, (ii) when mitochondrial pyridine nucleotides are oxidized but not when they are reduced, (iii) parallel to NAD+ hydrolysis, (iv) in a CSA-inhibitable manner, (v) without

19

20

C. Richter

inhibition of respiration and (vi) without entry of extramitochondrial solutes such as sucrose into mitochondria. We conclude from this that ONoo- can mobilize mitochondrial Ca2+ by stimulating the specific, ADP-ribose-dependent release pathway.

Determination of NO-induced apoptosis at the single-cell level In pilot experiments (A. Cossarizza, C. Franceschi and C. R1chter, unpublished work) we measured, with a fluorescence-assisted cell sorter, NO-induced changes in Ll Y, apoptosis and necrosis at the single-cell level. We confirmed that a bolus of NO transiently decreases Ll Y and found that, in the myelomonocytic human cell line U937, GSNO/DTT induces apoptosis in a time- and concentrationdependent manner. With 2 mM GSNO/DTT, 30% of the cells underwent apoptosis and about the same proportion of cells were necrotic within 17 h. By that time about 50% of the mitochondria were de-energized. Throughout the experiment Ll Y changes preceded apoptosis. The work done in the laboratory of C.R. was generously supported over the years by the Schweizerischer Nationalfonds, by the Schweizerische Krebsliga, by the European Science Foundatzon and recently by an anonymous sponsor.

References 1. 2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

Carafoli, E. (1987) Annu. Rev. Biochem. 56, 395-433 Richter, C. and Kass, G.E.N. (1991) Chern. Bioi. Interactions 77, 1-23 Rizzuto, R., Bastianutto, C., Brini, M., Murgia, M. and Pozzan, T. (1994) J. Cell Bioi. 126, 1183-1194 Hajnoczky, G., Robb-Gaspers, L.D., Seitz, M. and Thomas, A.P. (1995) Cell82, 415-424 Richter, C. (1992) m Molecular Mechanisms in Bioenergetics (L. Ernster, ed.), New Comprehensive Biochemistry Series (Neuberger, A. and Van Deenen, L.L.M., eds.), pp. 349-358, Elsevier, Amsterdam Cadenas, E. (1989) Annu. Rev. Biochem. 58, 79-110 Richter, C. (1988) FEBS Lett. 241, 1-5 Richter, C. and Schlegel,]. (1993) Toxicol. Lett. 67, 119-127 Richter, C. and Frei, B. (1988) Free Radical Bioi. Med. 4, 365-375 Schanne, F.A., Kane, A., Young, E. and Farber,]. (1979) Science 206, 699-700 Kass, G.E.N., Juedes, M. and Orrenius, S. (1992) B10chem. Pharmacal. 44, 1995-2003 Richter, C., Gogvadze, V., Schlapbach, R., Schweizer, M. and Schlegel, J. (1994) Biochem. Biophys. Res. Commun. 205,1143-1150 Kurokawa, T., Kobayashi, H., Harada, A., Nakao, A., Sugiyama, S., Ozawa, T. and Takagi, H. (1992) Transplantation 53, 308-311 Thompson, C.B. (1995) Science267, 1456-1462 Steller,H.(1995)Sctence267, 1445-1449 Buttke, T.M. and Sandstrom, P.A. (1994) Immunol. Today 15, 7-10 Garcia, I., Martinou, I., Tsujimoto, Y. and Martinou, J.-C. (1992) Science 258, 302-304 Steinman, H.M. (1995) J. Btol. Chern. 270,3487-3490 Jacobson, M.D. and Raff, M.C. {1995) Nature (London) 374, 814-816 Shimizu, S., Eguchi, Y., Kosaka, H., Kamiike, W., Matsuda, H. and Tsujimoto, Y. (1995) Nature (London) 374, 811-813 Beaver, J.P. and Waring, P. (1994) Immunol. Cell Bioi. 72,489-499 Martin, S.J. and Green, D.R. (1995) Cell82, 349-352 Lam, M., Dubyak, G., Chen, L., Nunez, G., Miesfeld, R.L. and Distelhorst, C.W (1994) Proc. Nat!. Acad. Sci. U.S.A. 91, 6569-6573 Lee, H. C. (1994) Cell. Signallmg 6, 591-600 Chacon, E. and Acosta, D. (1991) Toxicol. Appl. Pharmacal. 107, 117-128 Hennet, T., Richter, C. and Peterhans, E. (1993) Biochem.J. 289, 587-592 Lafon-Cazal, M., Pietri, S., Culcasi, M. and Bockaert,J. (1993) Nature (London) 364,535-537

Nitnc oxide and its congeners in mitochondria

28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67.

Jacobson, M.D., Burne, J.F., King, M.P., Miyashita, T., Reed, J.C. and Raff, M.C. (1993) Nature (London) 361, 365-369 Skowronek, P., Haferkamp, 0. and Rode!, G. (1992) Biochem. Biophys. Res. Commun. 187, 991-998 Vayssierre,J.-L., Peut, PX., Risler, Y. and Mignotte, B (1994) Proc. Nat!. Acad. Sci U.S.A. 91, 11752-11756 Di Lisa, F., Blank, P.S., Colonna, R., Gambassi, G., Silverman, H.S., Stern, M.D. and Hansford, R.G. (1995)J.Physiol. 486,1-13 Newmeyer, D.O., Farschon, D.M. and Reed,J.C. (1994) Cell79, 353-364. Richter, C., Schweizer, M., Cossarizza, A. and Franceschi, C. (1996) FEBS Lett. 378, 107-110 Smets, L.A., Van den Berg, J., Acton, D., Top, B., Van Rooij, H. and Verwijs-Janssen, M. (1994) Blood 84, 1613-1619 Wolvetang, E.J., Johnson, K.L., Krauer, K., Ralph, S.J. and Linnane, A.W. (1994) FEBS Lett. 339,40-44 Wallace, D.C. (1992) Annu. Rev. Btochem. 61, 1175-1212 Cossarizza, A., Baccarani Contri, M., Kalashnikova, G. and Franceschi, C. (1993) Biochem. Biophys. Res. Commun. 197,40-45 Cossarizza, A., Kalashnikova, G., Grasstlh, E., Chiappelli, F., Salvwh, S., Capri, M., Barbieri, D., Troiano, L., Monti, D. and Franceschi, C. (1994) Exp. Cell Res. 214, 323-330 Zanzami, N., Susin, S.A., Marchetti, T., Hirsch, M., Castedo, M. and Kraemer, G. (1996)J. Exp.Med. 183,1533-1544 Kroncke, K.-D., Fehsel, K. and Kolb-Bachofen, V. (1995) Bioi. Chern. Hoppe-Seyler 376, 327-343 Snyder, S.H. (1995) Nature (London) 377, 196-197 Sessa, W.C., Garcia-Cardenas, G., Liu,J., Pollock,J.S., Bradley,]., Thiru, S., Braverman, l.M. and Desai, K.M. (1995)]. Bioi. Chern. 270, 17641-17644 Kobzik, L., Stringer, B, Balligand,J.L., Reid, M.B. and Stamler,J.S. (1995) Biochem. Biophys. Res. Commun. 211, 375-381 Bates, T.E., Loesch, A., Burnstock, G. and Clark, J.B. (1995) Biochem. Biophys. Res. Commun. 213, 896-900 Lowenstein, C.J. and Snyder, S.H. (1992) Cell70, 705-707 Lipton, S.A., Choi, Y.B., Pan, Z.H., Lei, S.Z., Chen, H.S.V., Sucher, N.J., Loscalzo, J., Singcl, D.J. and Stamler, J.S. (1993) Nature (London) 364, 626-632 Stesjo, B.K. (1988) Ann. N.Y. Acad. Sci. 522,638-661 Hockenbery, D.M., Oltvai, Z.N., Yin, X.M., Milliman, G.L. and Korsmeyer, S.J. (1993) Cell 75,241-251 Stamler, J.S., Singe!, D.J. and Loscalzo, J. (1992) Science 258, 1898-1902 Stamler,J.S. (1994) Cell78, 931-936 Beckman,J.S., Beckman, T.W., Chen,J., Marshall, P.A. and Freeman, B.A. (1990) Proc. Nat!. Acad. Sci. U.S.A. 87, 1620-1624 Radi, R., Beckman,J.S., Bush, K.M. and Freeman, B.A. (1991) J. Bioi. Chern. 266, 4244-4250 Lin, K.-T., Xue, J.-Y., Nomen, M., Spur, B. and Wong, P.Y.-K. (1995) J. Bioi. Chern 270, 16487-16490 Murphy, M.E. and Sie~, H. (1991) Proc. Nat!. Acad. Sci. U.S.A. 88, 10680-10684 Lipton, S.A., Kim, W.-K., Rayudu, P.V., Asaad, W., Arnelle, D.R. and Stamler,J.S. (1995) Endothelium 3, S44, abstract 174 Albina,J.E., Cui, S., Mateo, R.B. and Reichner,J.S. (1993) J. Immunol. 150, 5080-5085 Cui, S., Retchner, J.S., Mateo, R.B. and Albina,J.E. (1993) Cancer Res. 54, 2462-2467 Ankacrona, M., Dypbukt,J.M., Briine, B. and Nicotera, P. (1994) Exp. Cell Res. 213, 172-177 Messmer, U.K., Ankacrona, M., Nicotera, P. and Briine, B. (1994) FEBS Lett. 355,23-26 Mannik,J.B., Asano, K., Izumi, K., Kieff, E. and Stamler,J.S. (1994) Cell79, 1137-1146 Bonfoco, E., Krainc, D., Ankacrona, M., Nicotera, P. and Ltpton, S.A. (1995) Proc. Nat!. Acad. Sci. U.S.A. 92,7162-7166 Schweizer, M. and Richter, C. (1994) Biochem. Biophys. Res. Commun. 204,169-175 Richter, C, Gogvadze, V., Laffranchi, R., Schlapbach, R., Schweizer, M., Suter, M., Walter, P. and Yaffee, M. (1995) Biochim. Biophys. Acta 1271, 67-74 Brown, G.C. (1995) FEBS Lett. 369,136-139 Lehninger, A.L., Vercesi, A and Bababunmi, E.A. (1978) Proc. Nat!. Acad. Set. U.S.A. 75, 1690-1694 Lotscher, H.R., Winterhalter, K.H., Carafoli, E. and Richter, C. (1979) Proc. Nat!. Acad. Sci. U.S.A. 76, 4340-4344 Lotscher, H.R., Winterhalter, K.H., Carafoli, E. and Richter, C. (1980) J. Bioi. Chern. 255, 9325-9330

21

22

C. Richter

68. 69. 70. 71. 72.

73. 74. 75. 76. 77. 78.

Schweizer, M., Schlegel,]., Baumgartner, D. and Richter, C. (1993) Biochem. Pharmacol. 45, 641--646 Bolli, R. (1991) Cardiovasc. Drugs Ther. 5, 249-268 Gogvadze, V. and Richter, C. (1993) FEBS Lett. 333,334-336 Schweizer, M., Durrer, P. and Richter, C. (1994) Biochem. Pharmacol. 48,967-973 Schweizer, M. and Richter, C. (1994) Biochemistry 33, 13401-13405 Shoji, Y., Uedono, Y., Ishikura, H., Takeyama, N. and Tanaka, T. (1995) Immunology 84, 543-548 Hennet, T., Peterhans, E., Richter, C. and Bertoni, G. {1993) Cancer Res. 53, 1456-1460 Richter, C. (1993) FEBS Lett. 325,104-107 Bornkamm, G.W. and Richter, C. (1994) Curr. Topics Microbiol. Immunol. 194,323-330 Chen, L.-B. (1988) Annu. Rev. Cel!Biol. 4, 155-181 Schweizer, M. and Richter, C. (1996) Biochemistry 35, 4524--4528

Interactions between nitric oxide and cytochrome-c oxidase in turnover JaumeTorres and MichaeiT.Wilson* Department of Biological Sciences (Central Campus), University of Essex, Colchester, C04 3SQ, U.K.

Introduction Nitric oxide (NO) is a free radical generated in vivo by NO synthases. The biological roles of this molecule are diverse, including activation of guanylate cyclase, neurotransmission and killing of tumour cells or parasites. In addition to these roles, the ability of NO to inhibit cytochrome-c oxidase in mitochondria has attracted considerable attention, as NO concentrations in the range of those found in vivo can rapidly and reversibly inhibit the enzyme [1-3]. Furthermore, the ability of NO to inhibit cytochrome-c oxidase is greatly influenced by the ambient oxygen concentration, being much more effective at the low oxygen concentrations found in vivo [2,3]. There is also growing evidence that cytochrome-c oxidase is able to metabolize NO. Under anaerobic conditions, purified cytochrome-c oxidase has been found to catalyse the reduction of NO to N 2 0 [4], and a cyanide-sensitive depletion of NO by mitochondria bas been measured directly using an intracellular EPR technique [5]. Mitochondria have also been found to catalyse the breakdown of NO in the presence of oxygen. This breakdown is partially inhibited by azide and cyanide, classical inhibitors of cytochrome-c oxidase [6]. In addition, we have data Q. Torres and M.T. Wilson, unpublished work) suggesting that electron donation to NO or N0/0 2 can occur in the binuclear centre. It was shown previously that under anaerobic conditions the fully reduced enzyme and, to some extent, the fully oxidized form can bind NO [7]. However, there are few studies at present on the interactions of the enzyme in turnover with NO. We have undertaken such a study with the aim of investigating the mechanisms through which cytochrome-c oxidase becomes inhibited and through which NO may be metabolized by the enzyme. Cytochrome-c oxidase (EC 1. 9.3 .1) is the final electron acceptor of the mitochondrial respiratory chain, transferring electrons from cytochrome c to molecular oxygen to form water. Coupled to this redox reaction, the enzyme pumps protons across the membrane, contributing to the generation of the electrochemical potential gradient (~fLH +)which drives ATP synthesis and, in turn, controls the activity of the enzyme. The enzyme contains two haem a groups which are placed in protein environments which confer on them distinctive *To whom correspondence should be addressed.

23

24

J.Torres and M.T.Wilson

properties giving rise to moieties termed cytochrome a and cytochrome ar The former acts as an electron transfer protein and is in rapid redox equilibrium with CuA, the electron entry site of the enzyme. Cytochrome a 3 , together with CuB, forms a binuclear metal centre where oxygen is bound and reduced [8]. NO reacts with fully oxidized or reduced cytochrome-c oxidase exclusively at the binuclear oxygen binding site with no apparent involvement of either cytochrome a or CuA (the electron entry sites). In the fully reduced enzyme, NO binds tightly to ferrocytochrome a 3 and CuB+ acts as a second binding site with lower affinity. On the basis of low-temperature EPR spectroscopy, Stevens et al. showed that in the fully oxidized resting enzyme NO binds solely to CuB 2 + [9]. Recently, we have studied the interactions of NO with isolated bovine cytochrome-c oxidase under conditions in which the enzyme is turning over, e.g. in the simultaneous presence of a reductant and oxygen [1 0]. Under these conditions, we have confirmed that low concentratiOnS of NO, somewhat surprisingly, compete effectively with oxygen and inhibit the enzyme (K,- 10- 7 M). The final product of this inhibition is an enzyme species in which cytochrome a 3 is reduced and bound to NO. However, it appears that inhibition is mediated by prior binding of NO to a partially reduced form of the enzyme for which oxygen has low affinity. We have suggested that this form may be one in which CuB is reduced while cytochrome a 3 remains oxidized [1 0]. An alternative view, that cytochrome a 3 is reduced and CuB oxidized, has been proposed by Giuffre and co-workers [11]. To attempt a resolution of this problem and in order to investigate the early events in the inhibition we have undertaken a series of rapid-scan stoppedflow experiments in which the enzyme in turnover was mixed with NO. The reductant system used (ascorbate/hexaminoruthemum) is spectrally silent; thus this methodology allows us to collect spectra of the enzyme in turnover and in the presence of NO as a function of time, and hence offers the opportunity of identifying spectral intermediates formed during the onset of inhibition. This system is inherently complex, as the enzyme in turnover comprises a number of species varying in redox state and in the nature of the ligands bound to the cytochrome-a/CuB centre. Figure 1 illustrates the experimental protocol and the different species which are proposed to constitute the catalytic cycle of the enzyme in turnover.

Materials and methods Cytochrome-c oxidase was prepared according to a method based on that of Yonetani [12]. Figure 1 shows the design for the stopped-flow experiments in which the enzyme in turnover was mixed with NO. The fully reduced enzyme was formed by mixing 60 f!.i of 136 iJ.M (expressed in functional units) cytochrome-c oxidase, 4 IJ.l of 0.4 mM hexaminoruthenium, 100 ILl of 50 mM sodium ascorbate and 335 IJ.I of buffer (0.1 M Hepes, pH 7.4, 1% Tween 80), giving final concentrations of 8.16 fJ.M cytochrome-c oxidase, 1.6 ILM ruthenium

Interactions between nitric oxide and cytochrome-c oxidase in turnover

25

Figure 1

(a) Hexam1noruthenium + Ascorbate 'Restmg' cytochrome-c OXIdase

+ 02

Reduced enzyme

02 SYRINGE A 'Pulsed' enzyme~

(b)

Schematic representations of the stopped-flow experiments and the catalytic cycle (a) Protocol for stopped-flow experiments to study the reaction of NO with cytochrome-c oxidase in turnover. (b) The catalytic cycle of cytochrome< oxidase. F, ferry/ species.

hexamine and 5 mM ascorbate. Under these conditions the enzyme consumes the oxygen present in the buffer in a few minutes and becomes fully reduced. The reduced enzyme was mixed with 0.5 ml of air-equilibrated buffer (- 250 1-LM Oz) in a syringe {syringe A). The addition of oxygen very rapidly (1-2 ms) reoxidizcs the enzyme and generates the so-called 'pulsed' enzyme which is generally considered more homogeneous in its properties than the enzyme as prepared,

26

J.Torres and M.T.Wilson

termed the 'resting' enzyme. The enzyme now enters a steady state in which its average redox state and its spectrum remain essentially unchanged until oxygen is again consumed. This period was typically -2.5 min, after which the enzyme again became fully reduced. Before re-reduction occurred (e.g. -1 min into the 'turnover period') the solution in syringe A was mixed with an anaerobic solution of NO from syringe B in the stopped-flow apparatus (Applied Photophysics DX17, Leatherhead, Surrey, U.K.).

Results On mixing the enzyme in the steady state with NO, a complex series of spectroscopic changes occurred. Some of these took place in the first milliseconds following mixing and the whole reaction was complete after some SO s. Figure 2 shows the three-dimensional representation of these spectral changes and Figure 3 the time courses at some selected wavelengths. Spectra collected at various times during the progress of the reaction are shown in Figure 4. The initial spectrum in the So ret region, centred at 426 nm, is very similar to the spectrum of the enzyme in turnover before mixing with NO and is characteristic of the enzyme being largely in an oxidized state. This is to be expected as the rate of electron entry is very low and the reactions with oxygen are rapid. The spectrum collected at 20 s clearly shows that the enzyme is inhibited, as indicated by the peak at 445 nm, which indicates that electron efflux from cytochrome a has greatly decreased and

Figure 2 0.7 0.6 Q) (.)

c:

0.5

-e0

0.4

Cll Cl)

.c .c

400 --ISDN

±

~GSNO

9A!I1'119l::!

Cell-cycle effects of NO at the G I, S and G2/M phases EGFR-T 17 mono/ayers were treated with EGF (I 0 nM) for 20 h in the presence or absence of either staurasporine (I nM) or hydroxyurea (I mM), or for 6 h in the presence of nocodazo/e ( 1.5 /-(M).After

being washed they were cultured with EGF alone (control) or together with SNAP (200 /-(M) for the times indicated to the right of each panel.

Control of cell growth by nitric oxide

mentioned inhibitory effect exerted by NO at these levels. Taken together, these results indicate that the inhibitory, cGMP-independent effect of NO on cell growth is a complex process occurring at multiple steps during G 1 and S phases of the cell cycle. The only phase apparently not affected by NO appears to be the G2/M transition.

Discussion The exact role of NO in the control of cell growth, although analysed in several cell systems, has never been unambiguously clarified, since both stimulatory and inhibitory effects have been reported [1-13]. A striking example of this double effect occurs in the vascular system. On the one hand, NO produced by endothelial cells is known to induce quiescence of smooth-muscle cells, thus playing a major role in the overall control of peripheral vascular resistance [30]. On the other hand, its spectrum of actions includes stimulation of endothelial cell proliferation, which has led to NO being proposed as an autocrine mediator for neovascularization [11]. The variability of the NO effects is not so unexpected if one considers the multiplicity of the intracellular pathways affected by the gaseous messenger. The mechanisms by which NO operates to control cell growth have just begun to be elucidated. Except for the cGMP-independent inhibition of ribonucleotide reductase, documented in a variety of reports [18], no direct link with growth control has been established for the other molecular functions attributed to NO, including stimulation of immediate early gene expression [14, 15,24,31], MAPK dephosphorylation and increased EGF receptor phosphorylation [16, 17]. The main task of the present investigation was the correlation of the effects of NO on cell growth with specific, growth-related events, dissected by appropriate biochemical and pharmacological means. The results we have obtained reveal a number of important new aspects. NO does indeed exert dual, conflicting effects in the same cell system: a slight stimulation of cell growth, mediated by cGMP generation and ensuing activation of cGMP-dependent protein kinases; and a strong inhibition, mediated by cGMP-independent pathways. The stimulation of growth mediated by cGMP could be attributed, at least m part, to increased Ca2+ influx, an effect that has been demonstrated to be mitogenic in these cells [20], and to the activation of AP-1-specific transcription. The specific roles of fos, jun and myc in the response to NO are now under investigation. As far as the NO inhibitory, cGMP-independent effect is concerned, it appears to be exerted at the level of the cell cycle, during the G 1 and the S phases, as revealed after pharmacological synchronization of the cells at these steps. In contrast, the G2/M step was unaffected by NO, a negative observation that confirms the specificity of the inhibition found at the other steps of the cell cycle. The signal transduction pathways other than cGMP production that are responsible for the cell-cycle effects of NO remain at present largely undefined. Since the most effective among the NO donors employed were SNAP and GSNO, both S-nitroso compounds, a possible mechanism could be S-nitrosyl-

91

92

E. Clementi et al.

ation of proteins. Additional pathways, such as ADP-ribosylation, could also be involved [18]. From the results reported here, the functional role of NO in cell growth begins to be elucidated. The cGMP-dependent actions might function as modulatory events, because of both the short times required for their onset/ offset and the fine control of the cytosolic cGMP levels. The cGMP-independent inhibitory events, on the other hand, might function collectively as a multi-switch device to be activated whenever growth arrest is required at crucial steps of cell life, such as during differentiation towards a defined phenotype. The unique property of NO of operating differentially and at multiple levels, taken together with its diffusibility and its potential action as an intercellular messenger, suggests a particularly important role in the co-ordinate and diversified growth of multicellular structures and organs. At least in the case of the nervous system such a role for NO has indeed begun to be recognized. NO-induced growth arrest appears critical for neuronal differentiation [32] and for the co-ordinate development of important brain areas, such as the visual cortex and the olfactory bulb [33,34].

We thank L. Beguinot and L. Vallar for critical discussion and support, M. Miloso and A. Grebe for their help with the immunoprecipitations and the CAT assay. We also thank E.K. Rooney for careful reading of the manuscript. This work was supported in part by grants from the AIRC (Italzan Associatzon of Cancer Research) and from the ACRO target project of the Italian Consiglio Nazionale delle Ricerche. References 1.

2 3. 4. 5. 6 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Newby, A. C., Southgate, K.M. and Assender,J.W. (1992) Herz 17,291-299 Lmcoln, T.M., Komalavilas, P. and Cornwell, T.L. (1994) Hypertens10n 23, 1141-1147 Garg, U.C. and Hassid, A. (1990) Biochem. Biophys. Res. Commun. 171, 474-479 Miyazaki, M., Wahid, S., Bai, L. and Namba, M. (1992) Exp. Cell Res. 200, 404-409 Punjabi, C.J., Laskin, D.L., Heck, D.E. and Laskin, J.D. (1992)]. Immunol. 149,2179-2184 Goureau, 0., Lepoivre, M., Becquet, F. and Curtois, Y. (1993) Proc. Nat!. Acad. Sci. U.S.A. 90,4276-4280 Garg, U.C. and Hassid, A. (1989) Am.]. Physiol. 257, F60-F66 Yang, W., Ando,J., Korenaga, R., Toyo-oka, T. and Kamiya, A. (1994) Biochem. Biophys. Res. Commun. 203, 1160-1167 Garg, U.C., Devi, L., Turndorf, H., Goldfrank, L.R. and Bansinath, M. (1992) Brain Res. 592, 208-212 Konturek, S.J., Brzozowski, T., Majka,]., Pytko-Polonczyk,]. and Stachura, J. (1993) Eur. J. Pharmacal. 239,215-217 Ziche, M., Morbidelli, L., Masini, E., Amerini, S., Granger, H.J ., Maggi, C. A., Geppetti, P. and Ledda, F. (1994)]. Clin. Invest. 94,2036-2044 Leibovich, S.J., Polverini, P.J., Pong, T.W., Harlow, L.A. and Koch, A.E. (1994) Proc. Nat!. Acad. Sci. U.S.A. 91,4190-4194 Munoz-Fernandez, M.A. and Fresno, M. (1993) Biochem. Biophys. Res. Commun. 194, 319-325 Peunova, N. and Enikolopov, G. (1993) Nature (London) 364,450-453 Pilz, R.B.,Suhasini, M., Idriss, S., Meinkoth,J.L. and Boss, G.R. (1995) FASEB J. 9, 552-558 Baker, T.L. and Buss,J.E. (1996) in The Biology of Nitric Oxide, part 5 (Moncada, S., Stamler, ]., Gross, S. and Higgs, E.A., eds.), p 104, Portland Press, London Peranovich, T.M., da Silva, A.M., Fries, D.M., Stern, A. and Monteiro, H.P. (1995) Biochem. J. 305,613-619 Gross, S.S. and Wolin, M.S. (1995) Annu. Rev. Physiol. 57, 737-769 Clementi, E., Sciorati, C. and Nistico, G. (1995) Mol. Pharmacal. 48, 1068-1077 Magni, M., Meldolesi,]. and Pandiella, A.(1991)J. Bioi. Chern. 266, 6329-6335

Control of cell growth by nitric oxide

21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.

Miloso, M., Mazzotti, M., Vass, W.C. and Beguinot, L. (1995) J. Biol. Chern. 270, 19557-19562 Nerlov, C., De Cesare, D., Pergola, F., Caraccwlo, A., Blasi, F., Johnsen, M. and Verde, P. (1992) EMBO J. 11,4573-4582 Remh, P., Nerlov, C., Blasi, F. and Johnsen, M. (1990) Nucleic Acid Res. 18,5009-5017 Gudi, T., Huvar, I., Meinecke, M., Lohmann, S.M., Boss, G.R. and Pilz, R.B. (1996) J. Biol. Chern. 271, 4597-4600 Clementi, E. and Meldolesi,J. (1996) Cell Calcium 19,269-279 Gukovskaya, A. and Pando!, S. (1994) Am.J. Physiol.266, G350-G356 Gadbois, D.M., Crissman, H.A., Tobey, R.A. and Bradbury, E.M (1992) Proc. Nat!. Acad. Sci. U.S.A. 89, 8626-8630 Moore, E. C. and Hurlbert, R.B. (1985) Pharmacal. Therapeut. 27, 167-196 Jordan, M.A., Thrower, D. and Wilson, L. (1992)J. Cell Sci. 102,401-416 De Mey, J.G.R., Dijkstra, E.H. and Vrijdag, M.J.J.F. (1991) Am. J. Physiol. 260, H1128-H1134 Morris, B.J. (1995) J. Biol. Chern. 270, 24740-24744 Peunova, N. and Emkolopov, G. (1995) Nature (London) 375,68-73 Roskams, A.J., Bredt, D.S., Dawson, T.M. and Ronnett, G.V. (1994) Neuron 13,289-299 Wu, H.H., Williams, C.V. and McLoon, S.C. (1994) Science 265, 1593-1596

93

Nitric oxide controls cell proliferation during Drosophila development Grigori Enikolopov* and Boris Kuzin Cold Spring Harbor Laboratory, Cold Spring Harbor, NY I 1724, U.S.A.

Introduction Organ development requires a tightly controlled programme of cell proliferation followed by growth arrest and differentiation and, often, programmed cell death. The balance between the number of cell divisions and the extent of subsequent cell death determines the final size of the organ [1-3]. The signals that cause discrete groups of cells and organs to terminate growth at the appropriate cell number and size are not known but they probably involve as-yet undetermined inter- and intra-cellular anti-proliferative second messengers. We have recently shown that the cytostatic activity of nitric oxide (NO) is crucial for the transition from proliferation to differentiation during neuronal differentiation of cultured cells [4]. To investigate whether NO has a similar role during normal development of intact organisms, we have used Drosophila as a model organism. During Drosophila development, the structure, size and shape of most of the organs of the adult fly are determined in the imaginal structures of the larvae [5,6]. Disc cells divide rapidly throughout larval development and cease proliferating at the end of the third-instar period (Figure 1). In leg, wing and haltere discs, progression through the cell cycle stops in G2 phase 3-4 h before puparium formation. It resumes 15-18 h later, and then stops again in a defined spatial pattern after 12-14 h [7-9]. Although most of the dividing cells in the late larvae and in the early pupae are already committed to their adult fate, they do not develop a fully differentiated phenotype until growth arrest is firmly established. Thus, cell proliferation is temporally separated from cell differentiation, which takes place later during metamorphosis. We tested whether NO is involved in the signalling pathways that control co-ordinated temporary growth arrest in larvae and pupae and subsequent terminal growth arrest in pupae and adults. We have found that NO acts as an anti-proliferative agent during Drosophila development and controls the transition from cell proliferation to growth arrest and differentiation. Our study indicates a novel role for NO as a regulator of the balance between cell proliferation and cell differentiation during development, and suggests that NO controls cell number and thus the size of anatomical structures in an intact animal.

*To whom correspondence should be addressed.

95

96

G. Enikolopov and B. Kuzin

Figure 1 Pupa

Larva

Adult

~

..0

E

:l

c

Qi

NO mediates growth arrest

()

@ Cell division

(f-"'TSP ~/

--

Temporary Cell growth arrest division

~~--===

~

Terminal growth arrest and differentiation

Schematic summary of cell proliferation in imaginal discs and organ development in Drosophila Anti-proliferative properties of NO mediate part of the temporary and, perhaps, terminal growth arrest during development of imaginal discs, and thus control the balance between cell proliferation and cell differentiation.

NO synthase is expressed in imaginal discs during larval development When imaginal discs of the third instar and early pupae were examined for the NADPH-diaphorase activity of NO synthase (NOS), staining was observed in all imaginal discs, imaginal rings, histoblasts and the brain of the larvae, beginning in the third instar. Staining became more intense as development proceeded, indicating accumulation of the NOS activity. In late third-instar larvae and early pupae, a highly specific and reproducible pattern of very intense staining was evident. Later in development, when the discs began to evert in the prepupae, diaphorase staining of the forming structures became less intense and a distinct characteristic staining pattern of individual segments became evident. Diaphorase staining further decreased in a specific spatial pattern during early pupal development. Thus, there is a gradual and specific accumulation of NOS in those developing imaginal structures, which undergo radical changes during metamorphosis before giving rise to adult organs. This staining reaches its highest levels at the time when progression through the cell cycle slows down.

Inhibition of NOS results in hypertrophy of leg segments The strong anti-proliferative properties of NO and the specific pattern of diaphorase staining seen in mature imaginal discs imply that NO may act as a growth-arrest agent in these structures, and be capable of inhibiting DNA

Nitric oxide controls cell proliferation during Drosophila development

synthesis and supporting temporary cytostasis during the switch to metamorphosis. If NO indeed acts as an anti-proliferative agent during the late stages of larval development, then inhibition of NOS might result in excessive growth of organs and tissues, whereas ectopic overexpression of the NOS gene might have the opposite effect. To test this hypothesis, we inhibited NOS activity by injecting various NOS inhibitors in the developing larvae at the end of the third instar, several hours before metamorphosis. The resulting adults exhibited dramatic enlargements of many structures - legs, wings, genital structures, tergites, sternites etc. The changes most often effected and most profound were in the legs of the adults, where the diameter of certain segments increased 3-4-fold. The number of bristles and the number of rows of bristles were also increased, confirming that hyperproliferation of the cells had occurred. The leg segments most strongly affected were those (first and second tarsal segments, tibia and femur) whose primordi?- had the highest levels of NOS at the larval and prepupal stages. Identical changes were observed when structurally unrelated inhibitors of NOS were used, indicating that the observed effect resulted specifically from blocking NOS activity. To test whether changing the NO levels directly affects DNA synthesis in the imaginal discs, we labelled the nuclei with 5-bromodeoxyuridine (BrdU) after inhibition of NOS activity. There were significantly more BrdU -labelled cells in the imaginal discs of flies in which NOS activity was blocked than in those of control flies. Thus, inhibition of NOS at the late stages of larval development results in excessive cell proliferation and increased size of the structures of the body of the adult fly.

Ectopic expression of a mouse NOS transgene results in reduced size of leg segments The ability of NO to inhibit DNA synthesis and cell proliferation suggests that overexpression of NOS in developing larvae may lead to diminished cell proliferation in the imaginal discs and to a reduction in the size of organs of the adult fly. To increase NO production, we induced expression of NOS transgene in transformed larvae carrying the mouse inducible NOS eDNA gene under the control of the heat-shock promoter. Inducible NOS is a calcium-independent form of NOS that is capable of efficient constitutive NO production. Transgenic larvae were heat-shocked within 1 h after pupariation to induce ectopic expression of NOS before the final cell divisions took place. This resulted, among other changes, in a reduction in the size of the limbs of the fly. The distal segments of the legs were affected most frequently and to the greatest degree. In extreme cases, the whole tarsus was shortened 1.5-2-fold and several segments were fused together with poorly defined boundaries. The most terminal structures of the appendage remained intact in these defective legs, suggesting that the observed reduction in size was due to incomplete growth of the developing appendage, rather than to complete loss of its distal structures. The number of bristles in a row on the affected segments also decreased, although the number of rows did not change. The segments of the adult leg most often affected by the overexpression of NOS

97

98

G. Enikolopov and B. Kuzin

(third, fourth and fifth tarsal segments) were those that were not affected by the NOS inhibitors and whose precursors exhibited particularly low levels of diaphorase staining in the early prepupal stages. In contrast to experiments with inhibition of NOS, when imaginal discs were labelled with BrdU after induction of ectopic NOS expression, there were markedly fewer labelled cells in imaginal discs from induced NOS-transformed flies, than in uninduced controls. Thus, ectopic expression of NOS at the late stages of larval development results in a decrease in cell proliferation and a reduction in the size of the structures of the body of the adult fly.

Inhibition of apoptosis unmasks excessive proliferation For some Drosophila structures, for instance the eye, we consistently detected an increase in the number of cells inS phase in the imaginal disc after inhibition of NOS, but the resulting adult organ usually appeared normal. We tested the possibility that the apparently normal eye phenotype occurred as a result of programmed cell death, which counteracts excessive cell proliferation induced by NOS inhibition and restores the normal number of cells in the eye during metamorphosis. To suppress programmed cell death, we used GMR-P35 flies in which apoptosis in the developing eye is largely prevented by expression of recombinant baculovirus p35 protein [10]. When NOS was inhibited in GMR-P35larvae, the eyes of the adult flies showed numerous changes, reflecting the surplus proliferation of various cell types in the developing eye. The most dramatic of these changes was in the number of ommatidia in the adult eye, which increased from the nearly invariant complement of 750 in wild-type flies and untreated GMR-P35 flies, to nearly 820 after NOS inhibition in GMR-P35 flies. In addition, different types of accessory cells in the ommatidia proliferated after NOS inhibition to levels higher than those achieved by blocking apoptosis by p35 [10]. For instance, the number of secondary and tertiary pigment cells was increased from 12 per sample area in normal eyes to more than 35 in GMR-P35 flies after inhibition of NOS. This number exceeds the maximal number of pigment cells saved from programmed cell death in untreated GMR-P35 flies ( -25 per sample area) and suggests that extra pigment cells arise as a result of excessive cell proliferation caused by inhibition of NOS combined with suppression of cell death caused by p35. Furthermore, the number of bristles was increased in some areas of the eye in GMR-P35 flies after NOS inhibition, up to 4-5 per ommatidium, instead of the 3 seen in normal flies and untreated GMR-P35 flies. Similarly, the number of cone cells was increased from 4 in normal and in untreated GMR-P35 ommatidia, to 5 and 6 in many ommatidia of GMR-P35 flies after NOS inhibition. Thus, prevention of apoptosis in the developing Drosophila eyes revealed excessive proliferation of various cell types after NOS inhibition, which was otherwise masked by programmed cell death in the larvae and pupae. Together, our experiments with Drosophila have demonstrated that inhibition of NOS in larvae leads to enlargement of organs in adults and, conversely, that ectopic expression of NOS in larvae leads to a reduction in the

Nitric oxide controls cell proliferation during DrosophJ/a development

size of organs in adults. Also, the distribution of affected segments in the adult leg corresponds to the distribution of NOS in the larvae and the changes in segment size can be directly correlated with changes in DNA synthesis in imaginal discs after manipulations of NOS activity. The increased cell proliferation that occurs in response to NOS inhibition is masked in some structures by apoptosis and it can be revealed by suppressing programmed cell death. Taken together, these results provide support for the hypothesis that NO acts as an anti-proliferative agent and controls the cell number in an intact developing organism (Figure 1). We propose that induction of NOS is a crucial step in Drosophtla development and that NO may be a general regulator of cell proliferation and differentiation during organism development and morphogenesis. References 1. 2. 3. 4. 5.

6.

7. 8 9 10.

Bryant, P.J. and Simpson, P. (1984) Q. Rev. Bioi. 59,387-415 Raff, M.C. (1992) Nature (London) 356,397-400 Raff, M.C. (1996) Cel\86, 327-329 Peunova, N. and Enikolopov, G. (1995) Nature (London) 375,68-73 Cohen, S.M. ( 1993) in The Development of Drosophila melanogaster (Bate, M. and MartinezArias, A., eds.), pp. 747-841, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Fristrom, D. and Fristrom,J.W. (1993) in The Development of Drosophila melanogaster (Bate, M. and Marttnez-Arias, A., eds.), pp. 843-897, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Fain, M J. and Stevens, B. (1982) Dev. B10L 92, 247-258 Graves, B.J. and Schubtger, G. (1981) Dev. Bioi. 93, !04-110 Schubiger, M. and Palka, J. (1987) Dev. Bioi. 123, 145-153 Hay, B.A., Wolff, T. and Rubin, G.M. (1994) Development 120,2121-2129

99

Cytoprotective actions of nitric oxide in hepatic inflammation Timothy R. Billiar* andYoung-Myeong Kimt:j: *tDepartment of Surgery, A I0 I0 Presbyterian University Hospital, 200 Lothrop Street, Pittsburgh, PA 15213, U.S.A., and :j:Department of Pharmacology, University of Pittsburgh, Pittsburgh, PA I5261, U.S.A.

Intr-oduction From the time of the dis~overy of the L-arginine:nitric oxide pathway, it has been clear that nitric oxide (NO) contributes to the life-and-death struggle of cells and micro-organisms. The seminal observations of John Hibbs, Jr. eta!. [1,2] in 1987 described the induction of the inducible NO synthase (iNOS) pathway in activated macrophages and its key role in macrophage tumoricidal activity in vitro. The cytotoxic capacity of NO and its reaction products has since been confirmed in numerous systems by using diverse target cells and organisms, and is thought to contribute to tissue damage in many acute and chronic disease processes such as stroke [3] and chronic inflammatory diseases [4,5]. These toxic roles of NO are in contrast to evidence that NO has cellular protective actions [6,7] and even promotes cellular growth [8]. This brief review examines the cytopro tective actions of NO by discussing the protective roles of NO in hepatocellular injury during inflammation in the liver.

Regulation of NO synthesis in the liverOf the three known NOS isoforms-neuronal NOS (nNOS), inducible NOS (iN OS), and endothelial NOS (eNOS)-iNOS and eNOS have been identified in the liver. eNOS is expressed constitutively in the liver [9], and iNOS can be induced in heparocytes [10,11], Kupffer cells [12], Ito cells [13] and endothelial cells [14,15] in rodents. Hepatocytes were the first non-macrophage cell type [10] and the first human cell type shown to express iN OS [16], and the human iN OS was first cloned from cytokine-treated human hepatocytes [17]. The expression of iN OS is transcriptionally up-regulated in hepatocytes in vitro by exposure to combinations of tumour necrosis factor (TNF), interleukin-1 (IL-l), imerferon--y, and lipopolysaccharide (LPS) [18] or high concentrations of IL-1 [19]. Hepatocytes are stimulated to express iN OS in vivo under many circumstances, including endotoxaemia [12, 14, 15,20], chronic inflammation [11 ], hepatitis [20a], regeneration [21,22], parasitic infection [23] and ozone exposure [24]. In a human liver cell line, the transcriptional activation of iNOS by cyrokine exposure involves the interaction of transcriptional factors with binding in promoter and/or *To whom correspondence should be addressed.

101

I 02

T.R. Billiar andY.-M. Kim

enhancer regions between 3.8 and 7.0 kb upstream of the transcriptional start site [25 J. Despite the initial difficulty in demonstrating iN OS expression in human macrophages, iN OS expression by human hepatocytes was as robust as that seen in activated rodent macrophages. We have recently detected iN OS mRNA in liver biopsies, within the first few hours following trauma, of essentially every patient undergoing emergency abdominal operation, suggesting that the expression of iN OS in human livers is a facile process.

Initial observations using NOS inhibitors in vivo With the knowledge that iN OS was expressed in the liver in endotoxaemia and inflammation, and that the iN OS pathway was associated with cytotoxicity in macrophages, we initiated studies 8 years ago to test the hypothesis that induced NO production contributes to liver damage. We first chose a rodent model of LPS hypersensitivity in which there was massive liver damage and high-level iN OS expression in the liver. In this model, mice are first treated with killed Corynebacterium parvum which causes a hepatitis. This is followed 5-7 days later with a small dose of LPS (1 f.Lg/mouse), which causes massive hepatocellular damage within hours. When a bolus of N"'-monomethyl-L-arginine (L-NMMA), a non-selective NOS inhibitor, was given with the LPS, a 3- to 5-fold increase in liver damage was seen [26]. The dose of L-NMMA needed both to block plasma elevations in N0 2 - + N0 3 - and to increase injury was high, and the effect was seen only with the LPS, not when the L-NMMA was given without LPS. Histological examination of the livers from L-NMMA-treated animals revealed infarcts throughout the liver with microthrombi in the vessels. Further studies using this model demonstrated that the L- NMMA -associated damage could be attenuated by the co-injection of superoxide dismutase [27] and dramatically worsened by simultaneous cyclo-oxygenase inhibition [28]. The damage could also be reduced by co-injection of prostacyclin or heparin (27]. Whereas NOS inhibition alone increased the LPS-induced liver damage, cyclo-oxygenase inhibition alone did not, suggesting a more dominant role for NO than eicosanoids in protecting the liver from damage. Thus, this initial series of studies provided evidence that NO from some source protected the liver from damage in this model of LPS hypersensitivity and that the mechanism of protection involved interaction with oxygen radicals and the prevention of thrombosis. Administration of high doses of LPS alone to animals also causes acute liver damage, although not nearly as severe as seen in models where animals have been rendered hypersensitive. We (29] and others [30] have shown that nonselective NOS inhibition with boluses of L-NMMA or N"'-nitro-L-arginine methyl ester (L-NAME) also increases liver injury in these models. Again, doses of the inhibitors adequate to block plasma N0 2 - + N0 3 - elevations were required for the increased damage. It is likely that polymorphonuclear leucocytes (PMN) contribute to the increased injury. Non-selective NOS inhibition with even low doses of LPS has been shown to increase PMN adhesion in the hepatic sinusoids [31 ].

Cytoprotective actions of nitric oxide in hepatic inflammation

Only recently has the issue of isoform specificity been addressed. Administration of boluses of isothioureas to rats given high doses of LPS was shown to inhibit the LPS-induced elevations in hepatic enzyme release and bilirubin levels [32). Isothioureas exhibit partial selectivity against iN OS and the authors postulated that excess NO generated by iN OS led to the hepatic damage. Another explanation could be that iN OS inhibition improved organ perfusion through effects on systemic vascular tone. Furthermore, it has been discovered recently that isothioureas also block the expression of iN OS protein, suggesting that these compounds may have salutary actions independent of iN OS enzyme inhibition [33]. Additional insight into the importance of distinguishing between the NOS isoforms has come from studies using iNOS knockout mice. Two studies have shown that iNOS-deficient mice were protected from LPS-induced death [34,35], while a third study, which utilized F 2 offspring, reported no improvement in 96 h survival using high doses of LPS [36). In one of the studies a more detailed examination of liver injury found no differences in plasma hepatocellular levels 5.5 h after LPS (10 mg/kg) between iNOS +I+ and iNOS -/mice [33]. Although the precise roles of iN OS in the lethality of septic shock remain controversial after these studies, the work did begin to indicate that NO from iNOS was not playing a clear role in promoting or preventing hepatic necrosis m endotoxaemia.

Studies using infusion of selective inhibitors We have had three major concerns about the previous studies (including our own) which examined the cytotoxic and cytoprotective actions of NO in the liver in inflammation. First, most have utilized non-selective inhibitors of the NOS isoforms. Secondly, all have given the inhibitors systemically and as a bolus. Therefore, specific inhibition in the liver is not assured and constant inhibition is also an uncertainty. Finally, relying on knockout studies alone raises the issue of altered responses in these animals, i.e. genetic mutations which alter the normal response. To address these concerns, we initiated a series of experiments in which we devised a method to infuse NOS inhibitors directly and constantly into the liver. Catheters from Alzet osmotic pumps were inserted into branches of the portal vein, and the pumps were loaded with various NOS inhibitors and placed in the abdominal cavity. Using this approach, we effectively blocked NOS activity for up to 7 days [37]. In short-term experiments (less than 24 h), we have examined the consequences of constant hepatic infusion of NOS inhibitors in LPS-treated rats. Immediately after pump placement, animals received LPS or saline and the level of liver damage was determined at 16 h. The infusion of non-selective inhibitors (L-NMMA and L-NAME) resulted in a significant increase in hepatocellular necrosis indicated by increases in plasma levels of transaminases [38]. Infusion of N-iminoethyl-L-lysine (~IL) or aminoguanidine effectively blocked elevations in plasma NO, levels but had no effect on hepatocellular necrosis. Examination of PMN infiltration revealed an increase in LPS-induced PMN accumulation in the L-NAME group but no difference with NIL. L-NAME infusion also increased

I 03

I 04

T.R. Billiar andY.-M. Kim

LPS-induced expression of the intercellular adhesion molecule, ICAM-1. In contrast to the observations on LPS-induced hepatocellular necrosis in which NIL had no effect, infusion of NIL resulted in the appearance of apoptosis within the liver. All the effects of NOS inhibitors were attenuated by the simultaneous infusion via the jugular vein of an NO donor, 0 2 -vinyll-(pyrrolidin-1-yl)diazen1-ium-1,2-diolate (V-PYRRO/NO), which concentrates NO release in the liver, indicating that the changes induced by the inhibitors were due to the inhibition of NO synthesis. Thus, eNOS inhibition promoted necrosis whereas iNOS inhibition led to detectable apoptosis without enhanced necrosis. It has also been suggested that increased NO synthesis leads to metabolic inhibition in the liver in endotoxaemia [39,40]. We have examined the role of NO in the inhibition of hepatic gluconeogenesis, which is known to occur in sepsis [41]. Alzet osmotic pumps for portal infusion of NOS inhibitors were again placed in rats prior to LPS injection. Hepatocytes isolated from LPS-treated rats were shown to exhibit the expected 60% decrease in gluconeogenic capacity. This decrease was not altered by NOS inhibition with L-NMMA or aminoguanidine. Furthermore, no difference in the LPS-induced decrease in gluconeogenesis was seen between hepatocytes isolated from iN OS -I- deficient and iN OS +I+ mice. Thus, even though NO exposure has been shown to block gluconeogenesis in vitro [39,40], there was no evidence for this in vivo. These results point to the importance of confirmation in vivo of results obtained in vitro.

Cytoprotective actions of NO in hepatocytes Hepatocytes are very resistant to toxicity by NO. In fact, millimolar concentrations of donors are required to kill hepatocytes in vitro. In vivo, our data indicated that NO generated by iN OS blocked apoptosis. It has been shown that pre-exposure of hepatocytes to NO donors renders the cells less susceptible to H 2 0 2 or high concentrations of NO donors [42]. This is due to the induction of haem oxygenase, which prevents oxidative damage. Although this study provided evidence for NO-induced cytoprotective mechanisms in hepatocytes, it did not indicate how NO might protect from apoptosis. Therefore, we examined the capacity of cultured hepatocytes to resist TNF-induced apoptosis (43]. Preexposure to an NO donor, S-nitroso-N-acetylpenicillamine (SNAP), rendered the cells resistant to TNF + actinomycin D (TNFI ActD )-induced apoptosis. The protection correlated with the SNAP-induced expression of heat-shock protein 70 (HSP70) and was blocked by inhibiting HSP70 expression using a specific antisense oligonucleotide to HSP70. The protection and HSP70 expression following SNAP exposure exhibited a lag period of 12-16 h. The induction of HSP70 may have been due to the oxidation of intracellular glutathione. Exposure to short-lived NO donors probably does not mimic endogenous iN OS expression. Whereas iN OS expression is gradual in onset and then sustained for hours to days, most NO donors release NO in high levels for short periods of time. To study the capacity of cells which expressed sustained iNOS expression to resist injury, we prepared a human liver cell line, AKN-1, which stably expressed human iN OS. A retroviral vector was used to transfer

Cytoprotective actions of nitric oxide in hepatic inflammation

I OS

human iNOS driven by a cytomegalovirus promoter into the cells. Sustained expression of iNOS for up to 8 days resulted in no toxicity. The cells were resistant to TNF I ActD-induced apoptosis but HSP70 was not induced. Furthermore, the cells had to be producing NO at the time of TNF/ ActD exposure to exhibit the protection. Primary cultures of rat hepatocytes were also protected from TNF/ ActD-induced apoptosis by simultaneous exposure to V-PYRRO/NO, an NO donor which releases NO specifically in hepatocytes. The release is sustained for 12-24 h in culture. V-PYRRO/NO was found to protect hepatocytes from apoptosis through a cGMP-dependent mechanism m vitro. Rats were protected from massive liver damage and fulminant hepatic failure in response to an injection of D-galactosamine and TNF by a constant infusion of V-PYRRO/NO. Thus, quantities of NO adequate to stimulate soluble guanylate cyclase also protect hepatocytes from TNF damage.

A perspective The evolving story of the role of NO in mflammatory diseases has become increasingly more complex. At the same time, we are refining our understanding of the important issues. It would appear that the site and quantity of NO produced are key factors. These, in turn, must be dictated by the expression patterns of eNOS and iN OS. Our observations indicate that eNOS activity is a key to limiting inflammatory responses that would lead to hepatocellular necrosis in the liver (Figure 1). It is likely that the damaging effects of non-selective NOS inhibitors in early studies were due predominantly to eNOS inhibition. eNOS probably protects by inhibiting, either directly or indirectly, the adherence of PMN. Constitutive NO at the endothelial surface probably also prevents platelet adhesion and may neutralize oxygen radicals. iNOS expression in the parenchymal cells also has a protective role to play. High-level iN OS expression may trigger the expression of protective proteins such as haem oxygenase and other heat-shock proteins. Even low-level sustained iN OS expression could block the toxic effects of TNF via a cGMP-dependent process.

Reduces Inflammation-Induced Necrosis • Blocks PMN accumulation • Inhibits platelet accumulation • Neutralizes oxygen radicals

Constitutive NOS (eNOS) \

?\ 44 Inducible NOS (iN OS)

--...

Prevents Cell Death/Apoptosis • cGMP-dependent mechanism • Induction of stress proteins

Postulated protective actions of NO produced by eNOS and iN OS in the liver in inflammation

Figure 1

106

Figure 2

T.R. Billiar andY.-M. Kim

A

Cytotoxicity

ir c

Oxidative or nitrosative stress Result: induction of HSP70

Q)

0

c 0 0

0

z

Result: induction of haem oxygenase

Release of prosthetic iron from proteins

Cell signalling functions Result: activation of guanylate cyclase

Time

B

ic

Q)

0

c 0 0

0

l

z

Time

...

Responses to NO depend on the rate of NO formation (A) The response to NO by liver cells depends largely on its rate of formation. A low rate of formatiOn results in signalling functions, such as the activation of soluble guanylate cyclase. As the rate of NO formation increases, NO may interact with other cellular targets, resulting in either the release of iron or oxidativelnitrosative stress. (B) The rate response will be dependent on other factors such as redox status of the cell. For example, a reduction in oxtdative defences may result in the shift of the curve to the left, whereas an increase in antioxidants could result in a shift to the right

We believe that another critical aspect of enhanced NO formation is not the total amount of NO produced but the rate of NO formation. Based on the accumulated data, we suggest that there is a hierarchy of responses by liver cells to NO such that low rates are adequate to stimulate specific cellular signalling pathways (e.g. cGMP) and various redox reactions resulting from NO and oxygen radical interaction (Figure 2A). As rates increase, NO or its reaction products interact with heavy metals in some proteins, perhaps disrupting the metal prosthetic group. This is demonstrated in hepatocytes by the induction of haem oxygenase, resulting from the liberation of haem iron. Higher rates yet will induce

Cytoprotective actions of nitric oxide in hepatic inflammation

oxidative or nitrosative stress, resulting in stress protein expression (HSP70) and yet another cytoprotective system. At the highest rates, we postulate that protective mechanisms are overwhelmed and cytotoxicity is manifested. It is conceivable that this rate-response curve can be shifted (Figure 2B) by factors such as changes in antioxidant levels with the cells (e.g. glutathione levels) and the presence of other free radicals. Understanding these relationships will be essential to an understanding of the functions of NO in inflammation.

References 1.

Hibbs, Jr., J.B., Taintor, R.R. and Vavrin, Z. (1997) Science 235,473-476 Hibbs,Jr.,J.B. Vavrin, Z. and Taimor, R.R. (1987)J. Immunol.138, 550-565 3. Huang, Z., Huang, P.L., Panahian, N., Dalkara, T., Fishman, M.C. and Moskowitz, M.A. (1994) Science 265, 1883-1885 4. Nussler, A.K. and Billiar, T.R. (1993) J. Leukocyte Bioi. 54, 171-178 5. Kroncke, K.D., Fehsel, K., Sommer, A., Rodriguez, M.L. and Kolb-Bachofen, V. (1995) Bioi. Chern. Hoppe Seyler 376,179-185 6. Wink, D.A., Han bauer, I., Krishna, M.C., DeGraff, W., Gamson,J. and Mitchell, J.B. (1993) Proc. Nat!. Acad. Set. U.S:A. 90, 9813-9817 7. Mannick,J.B., Asano, K., Izumi, K., Kieff, E. and Stamler,J.S. (1994) Cell79, 1137-1146 8. Morbidelli, L., Chang, C.-H., Douglas,J.G., Granger, H.J., Ledda, O.F. and Zichc, M. (1996) Am.J. Physiol. 270, H411-H415 9. Mittal, M.K., Gupta, T.K., Lee, F.- Y., Sieber, C. C. and Groszmann, R.J. (1994) Am.J. Physiol. 267, G416-G422 10. Curran, R.D., Billiar, T.R., Stuehr, D.J., Hofmann, K. and Simmons, R.L. (1989) J. Exp. Med. 170,1769-1774 11. Bilhar, T.R., Curran, R.D., Stuehr, D.J., Stadler, J., Simmons, R.L. and Murray, S.A. (1989) Biochem. Biophys. Res. Commun. 168, 1034-1040 12. Billiar, T.R., Curran, R.D., Stuehr, D.J., West, M.A., Bentz, B. G. and Simmons, R.L. (1989)J. Exp.Med. 169,1467-1472 13. Rockey, D.C. and Chung,].]. (1995) J. Clin. Invest. 95, 1199-1206 14. Spitzer,J.A. (1994) Hepatology 19,217-228 15. Laskin, D.L., Heck, D.E., Gardner, C.R., Feder, L.S. and Laskin, J.D. (1994) J. Leukocyte Bioi. 56, 751-758 16. Nussler, A., Di Silvio, M., Billiar, T.R., Hoffman, R.A., Geller, D.A., Selby, R., Madariaga,J. and Simmons, R.L. (1992) J. Exp. Med. 176,261-264 17. Geller, D.A., Lowenstein, C.J., Shapiro, R.A., Nussler, A.K., Di Silvio, M., Wang, S.C., Nakayama, D.K., Simmons, R.L., Snyder, S.H. and Billiar, T.R. (1993) Proc. Nat!. Acad. Sci. U.S.A. 90, 3491-3495 18. Geller, D.A., Nussler, A.K., Di Silvio, M., Lowenstein, C.J., Shapiro, R.A., Wang, S.C., Simmons, R.L. and Billiar, T.R. (1993) Proc. Nat!. Acad. Sci. U.S.A. 90, 522-526 19. Geller, D.A., de Vera, M.E., Russel, D.A., Shapiro, R.A., Nussler, A.K., Simmons, R.L. and Billiar, T.R. (1995)]. Immunol. 155,4890-4898 20. Geller, D.A., Di Silvio, M., Nussler, A.K., Wang, S.C., Shapiro, R.A., Simmons, R.L. and Billiar, T.R. (1993) J. Surg. Res. 55, 427-432 20a. Kane, J.M., Shears, L.L., Hierholzer, C., Ambs, S., Billiar, T.R. and Posner, M.C. (1997) J. Surg. Res. 69,321-324 21. Obolenskaya, M., Schulze-Specking, A., Plaumann, B., Frenzer, K., Freudenberg, N. and Decker, K. (1994) Biochem. Biophys. Res. Commun. 204, 1305-1311 22. Hortelano, S., Dewez, B., Genaro, A.M., Diaz-Guerra, M.J.M. and Bosca, L. (1995) Hepatology 21, 776-786 23. Klotz, F.W., Scheller, L.F., Seguin, M.C., Kumar, N., Marietta, M.A., Green, S.J. and Azad, A.F. (1995) J. lmmunol. 154, 3391-3395 24. Laskin, D.L., Pendino, K.J., Punjabi, C.J., Rodriguez del Valle, M. and Laskin, J.D. (1994) Environ. Health Perspect. 102 (Suppl. 10), 61-64 25. de Vera, M.E., Shapiro, R.A., Nussler, A.K., Mudgett, J.S., Simmons, R.L., Morris, Jr., S.M., Billiar, T.R. and Geller, D.A. (1996) Proc. Nat!. Acad. Sci. U.S.A. 93, 1054--1059 26. Billiar, T.R., Curran, R.D., Harbrecht, B.G., Stuehr, D.J., Demetris, A.J. and Simmons, R.L. (1990) J. Leukocyte Bioi. 48,565-569 27. Harbrecht, B.G., Billiar, T.R., Stadler, J., Demetris, A.J., Ochoa, J., Curran, R.D. and Simmons, R.L. (1992) J. Leukocyte Bioi. 52, 390-394 2.

I 07

I 08

T.R. Billiar andY.-M. Kim

28. 29. 30. 31. 32. 33. 34.

35. 36. 37. 38. 39. 40. 41. 42. 43.

Harbrccht, B. G., Stadler, J., Demetris, A.J., Simmons, R.L. and Billiar, T.R. (1994) Am. J. Physiol. 266, G1004-G1010 Harbrecht, B.G., Billiar, T.R., Stadler, J., Demetris, A.J., Ochoa, J.B., Curran, R.D. and Simmons, R.L. (1992) Crit. Care Med. 20, 1568-1574 Wang, G.-S. and Liu, G.-T. (1995) Biochem. Pharmacal. 49, 1277-1281 Nishida, J., McCuskey, R.S., McDonnell, D. and Fox, E.S. (1994) Am. J. Physiol. 267, G1135-G1141 Szabo, C., Southan, G.J. and Thiemermann, C. (1994) Proc. Nat!. Acad. Sci. U.S.A. 91, 12472-12476 Ruetten, H. and Thiemermann, C. (1996) Biochem. Biophys. Res. Commun. 225, 525-530 MacMicking, J.D., Nathan, C., Hom, G., Chartrain, N., Fletcher, D.S., Trumbauer, M., Stevens, K., Xie, Q.-W., Sokol, K., Hutchinson, N., Chen, H. and Mudgett,J.S. (1995) Cell81, 641-650 Wei, X.-Q., Charles, I. G., Smith, A., Ure,J., Feng, G.-J., Huang, F.-P., Xu, D., Muller, W., Moncada, S. and Liew, F.Y. (1995) Nature (London) 375, 408-411 Laubach, V.E., Shesely, E.G., Smithies, 0. and Sherman, P.A. (1995) Proc. Nat!. Acad. Sci. U.S.A. 92, 10688-10692 Luss, H., Di Silvio, M., Litton, A.L., Molina y Vedia, L., Nussler, A.K. and B1lliar, T.R. (1994) Biochem. Biophys. Res. Commun. 204, 635-MO Ou, J., Carlos, T.M., Saavedra,J.E., Keefer, L.K., Kim, Y.M., Watkins, S.C., Harbrecht, B.G. and Billiar, T.R. (1997) Nitric Oxide Bioi. Chern., in the press Stadler,]., Trockfeld,J., Schmahx, W.A., Brill, T., Siewert,J.R., Greim, H. and Doehmer,J. (1994) Proc. Nat!. Acad. Sci. U.S.A. 91, 3559-3563 Horton, R.A., Ceppi, E.D., Knowles, R.G. and Titheradge, M.A. (1994) Biochem. J. 299, 735-739 Ou,J., Molina, L., Kim, Y.-M. and Billiar, T.R. (1996) Am.J. Physiol. 271, G621-G628 Kim, Y.M., Bergonia, H. and Lancaster,Jr.,J.R. (1995) FEBS Lett. 374,228-232 Kim, Y.M., de Vera, M.E., Watkins, S.C. and Billiar, T.R. (1997)J. Bioi. Chern. 272,1402-1411

Activation of nuclear factor-KB and activator protein- I, and steadystate level of mRNA encoding tumour necrosis factor-a and inducible nitric oxide synthase in rat liver during different types of cell proliferation M. Menegazzi*t,A. Carcereri de Prati*, M. Pibiri:j:, R. Piga:j:,A. Columbano:j:, G.M. Ledda-Columbano:j:, E. Cavalieri* and H. Suzuki* *lstituto di Chi mica Biologica, Universita' diVerona, 37134 Verona, Italy and :j:lstituto di Patologia Sperimentale, Universita' di Cagliari, 09124 Cagliari,ltaly.

Introduction

The ability of mammalian hepatocytes to proliferate either after loss or death of liver cells or in direct response to mitogens is a useful model for the study of cell proliferation of highly differentiated tissues. Moreover, it provides a good system for the characterization in vivo of the factors involved in the control of cell growth in response to physiological stimuli. A precise pattern of events, such as release and modulation of growth factors and cyrokines, is followed in the recruitment of cells to entry into the cell cycle. Tumour necrosis factor-a (TNF-a) has been suggested to play a major role in liver cell proliferation on the basis of the ability of recombinant TNF-a to induce liver cell proliferation [1] and of the inhibition by pretreatment with anti-TNF-a antibodies of liver regeneration after partial hepatectomy (PH) [2]. TNF-a is also a powerful inducer of the activation of nuclear factor (NF)-KB and of inducible nitric oxide synthase (iNOS) gene expression [3,4]. NF-KB consists of a heterodimeric complex of the Rei family of proteins, which is retained in an inactive form in the cytoplasm through association with the I -KB inhibitor protein. The release of I -KB following cellular stimulation allows the translocation of NF-KB into the nucleus where it can interact with KB binding sites and regulate transcription of the target genes [3 ]. Recently, activation of NF-KB has been suggested to play an important role in making the hepatocytes competent to proliferate and, together with other transcription factors such as activator protein (AP)-1, to be responsible for the increased hepatic expression of several immediate early genes [5,6]. tTo whom correspondence should be addressed.

109

I I0

M. Menegazzi et al.

It is well known that the production of nitric oxide (NO) has an important role in the maintenance of the integrity of the liver tissue [7]. An enhanced production of NO during liver restoration after PH has been observed [8,9]. Moreover, the cloned promoter of the murine gene coding for the iNOS protein contains two NF-KB and AP-1 putative binding sites [1 0]. Xie et al. [11] have recently shown, using additional constructs, that the activation ofNF-KB was necessary for the iN OS promoter inducibility by lipopolysaccharides. We have previously shown that compensatory liver regeneration occurring after cell loss and direct hyperplasia induced by primary mitogens displays a different pattern of immediate early gene and growth factor gene expression [12]. In order to establish whether changes in the transcription factors and gene expression are essential events for liver cell proliferation induced by different stimuli, in the present work we show the effect of direct hyperplasia induced by cyproterone acetate (CPA) and nafenopin (NAF), and compensatory regeneration induced by PH and carbon tetrachloride (CC1 4 ) on changes in: (1) the hepatic expression of TNF-a, (2) the induction of the iN OS gene and (3) the activation of NF-KB and AP-1.

Materials and methods Eight-week-old male Wistar rats were purchased from Charles River (Milano, Italy). The animals were fed a laboratory chow diet provided by Piccioni (Brescia, Italy) and had free access to food and water. CC1 4 (2 ml/kg; Sigma, St. Louis, MO, U.S.A.); NAF (200 mg/kg; a gift from Dr. V. Preat) and CPA (60 mg/kg; Pentagone Pharmaceuticals Inc., Montreal, Canada) were dissolved in corn oil and administered intragastrically. In other rats, PH (2/3 original mass) was performed according to the method of Higgins and Anderson [13]. Rats were killed at various times thereafter. Immediately after death, liver sections were fixed in 10% buffered formalin and processed for H&E staining. The remaining liver was frozen in liquid nitrogen and stored at -80 °C.

Northern blot analysis Total RNA was isolated from frozen liver by the procedure described by Chirgwin et al. [14]. Total RNA (40 !Lg) was separated on 1% agaroseformaldehyde gels and blotted on Hybond N membrane (Amersham, Bucks, U.K.). The RNA concentration was determined spectrophotometrically at 260 nm. To control the amount of total RNA in each lane, before blotting the gels were stained with ethidium bromide. Poly(A)+ RNA, when needed, was isolated by two cycles of affinity chromatography on oligo(dT) cellulose (Boehringer, Mannheim, Germany). UV-irradiated filters were then hybridized with randomprimed 32 P-cDNAs for rat iN OS [15] and TNF-a. Membranes were exposed to Kodak XAR film and the intensity of the bands was quantified by Phosphorimager (Molecular Dynamics, Sunnydale, CA, U.S.A.).

Nuclear factor-KB and inducible NO synthase expression in liver cells

Ill

Electrophoretic mobility shift assay (EMSA) Nuclear extracts were prepared from 200 mg of liver tissue according to Schreiber eta!. [16] in the presence of 10 f!.-g/mlleupeptin, 5 f!.-g/ml antipain and pepstatin, and 1 mM PMSF (Sigma). Protein concentration in the nuclear extracts was determined using the method of Bradford [17]. Nuclear extracts (8 ~-Lg) was incubated at room temperature for 30 min with (2-5 X 104 ) cpm of the 32 P-labelled double stranded oligonucleotide containing the consensus NF-KB DNA binding site (5' GATCCAGAGGGGACTTTCCGAGTAC 3') or the AP-1 DNA binding site (5' CTAGTGATGAGTCAGCCGGATC 3') in a 15 I-Ll reaction mixture containing 20 mM Hepes, pH 7.9, 50 mM KCI, 10% glycerol, 0.5 mM dithiothreitol, 0.1 mM EDTA, 2 f!.-g poly(di-dC), 1 f!.-g salmon sperm DNA. To check that nuclear extracts were not degraded, the EMSA was performed with an oligonucleotide containing the consensus YY1 DNA binding site (5' CTGCAG TAACGCCATTTTGCAAGGCATGAA 3'). Products were fractionated on a non-denaturing 5% polyacrylamide gel. In competition assays, 100-fold oligonucleotide competitor was·added 15 min before addition of the labelled probe. The intensity of the retarded bands was measured by Phosphorimager.

Results Treatment with CC1 4 caused a severe centrilobular hepatic necrosis followed by regeneration. On the other hand, single treatments with CPA and NAF induced an increase in liver weight and mitotic activity with no signs of cell necrosis, as reported previously [18]. The compensatory regeneration processes, induced by CCI4 administration, elicited a rapid and transient induction of the TNF-a gene expression, with a peak at 3-5 h; when we analysed the steady-state mRNA level for iN OS, we did not detect the bands in any of the control samples, whereas they were found in the treated samples 12 and 24 h after CC14 addition [19}. An induction of hepatic levels of iN OS mRNA starting at 4 h was observed also in another model of compensatory regeneration, namely the one occurring after 2/3 PH (Figure 1). In order to detect the pattern of transcriptional activation during the hepatic proliferative response, mobility shift assays were performed with oligonucleotide probes representing the binding sites for NF-KB and APl transcriptional factors. The NF-KB activity was present in the control samples as faint gelControl Time(h)

8 24 72 120

Figure 1

Partial hepatectomy

1

4 8

24 48 72 96 120 360 iN OS

Northern blot analysis of iN OS mRNA The steady-state level of mRNA encoding iN OS was measured in the liver tissue of control and partially hepatectomized rats over 1-360 h.

I 12

M. Menegazzi et al.

Figure 2

NAF Time (h) Competitor

free

2

6

C

12

12

6

12

12

+

DNA ~

EMSA of NF-KB transcription factor DNA binding activity af NF-KB transcription factor was estimated in the liver of control rats (C) and of rats after NAF or CC/ 4 treatment. Nuclear extracts from livers taken at 2. 6 and 12 h after drug administration were incubated with 32 P-Iabe/led double-stranded oligonucleotide containing the consensus sequence for the NF-KB binding site. The specificity of the bands was demonstrated by competing with a

I 00-fo/d excess of the NF-KB unlabelled oligonucleotide.

retarded bands (Figure 2). Following CCI4 administration there is an approximate 4-fold, and more than 10-fold, activation of KB-site binding activity after 6 and 12 h, respectively (Figure 2). This activity rapidly drops after 24 and 48 h. The specificity of all these bands was demonstrated by competition with a 100-fold excess of the NF-KB unlabelled oligonucleotide (Figure 2). In our previous experiments carried out with the same system, we demonstrated a transient induction of both c-fos and c-jun mRNA levels between 1 and 4 h [12]. We also demonstrated that the binding activity of AP-1 was markedly stimulated. The AP-1 gel-retarded band increases more than 20-fold, 6 and 12 h after CCl4 treatment [19]. By contrast, no increase in TNF-a mRNA levels or induction of iN OS gene could be observed during liver cell proliferation induced by the mitogens NAF and CPA. Moreover, no activation ofNF-KB (Figure 2) or AP-1 [19] was observed following treatment with these mitogenic agents. To rule out the possibility that the nuclear extracts are degraded, the EMSA was also performed using an oligonucleotide probe carrying the YYl site. With this probe the pattern of DNA-protein interaction is similar in control and CCI 4 -, NAF- and CPAtreated rat liver (data not shown). These results are in agreement with our

Nuclear factor-KB and inducible NO synthase expression in liver cells

previous report showing that NAF and CPA induced liver cell proliferation in the absence of increased expression of c-fos and c-jun [12].

Discussion Liver cell proliferation provides a unique system to study in vivo the transition of quiescent cells into the G 1 stage of the cell cycle. The liver cells start a series of timed responses of cytokine release and production of specific transcription factors [20]. In the present work we have demonstrated that an increase in steadystate levels of TNF-a occurs during compensatory liver regeneration induced by a necrogenic dose of CC1 4 • This was associated with activation of NF-KB and a transient induction of iNOS expression, similar to that seen after PH [5,9,21] (Figure 1). Although a prominent feature of iNOS expression is the marked synergy that occurs between various signals including cytokines [22], TNF-cx on its own can induce the expression of iN OS in rat hepatocytes in tissue cultures [4]. Thus, the early induction of TNF-a. mRNA levels might be sufficient to induce hepatic iN OS expression in vivo. It is known that TNF-cx is able to produce an oxidative stress leading to the activation of NF-KB [23]. Accordingly, a CC14 -elicited activation of the NFKB binding was demonstrated at times corresponding to 6 and 12 h after CC1 4 treatment, when TNF-a mRNA levels decreased. This is the first evidence that NF-KB can be activated in the compensatory liver cell regeneration triggered by CC1 4 • Moreover, the results show that c-fos and c-jun expression [12] precedes an increase of AP-1 DNA binding activity [19], in agreement with Zawaski et al. [24]. Since both NF-KB and AP-1 consensus sequences are present in the promoter region of the mouse iNOS gene [10], the activation of DNA binding of these transcription factors might represent a crucial step in the induction of iN OS gene expression by CC14 treatment. In order to understand whether liver cell proliferation was always associated with activation of NF-KB and/or AP-1 and TNF-cx and iN OS gene expression, we have also studied changes in these variables using a model in which liver cell proliferation is not preceded by cell loss (direct hyperplasia). Our previous data [12] showed that liver cell proliferation elicited by the primary mitogens NAF and CPA occurs without expression of the immediate early genes c-fos, c-jun and c-myc. In this work, we have demonstrated that liver cell proliferation elicited by those primary mitogens takes place in the complete absence of increased expression of TNF-cx or iN OS, and NF-KB or AP-1 activation. NAF belongs to a class of chemicals that bind to nuclear receptors, the peroxisome proliferator activated receptors [25]. The latter bind to the peroxisome proliferator responsive elements which in turns may act as a transcription factor [26]. Thus, it is possible that an alternative pathway to that requiring cytokines and NF-KB might be involved in liver cell proliferation induced by certain mitogens (hormone-receptor-mediated cell proliferation), probably through a direct action on downstream molecules.

113

I 14

M. Menegazzi et al.

In conclusion, these results indicate the existence of different signalling pathways in the triggering of liver cell proliferation, depending upon the nature of the proliferative stimulus. Furthermore, the present data show that the induction of iNOS expression is associated with the production of TNF-a and the successive activation of NF-KB, as seen in compensatory liver regeneration.

This work was supported by funds from C.N.R. (P.F. Oncologia) and Target Project on Aging, Associazione Italiana Ricerca sul Cancro, MURST 40% and 60%, and C.l.S.M.l., Italy. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

Akerman, P., Cote, P., Yang, S.Q., McClain, C., Nelson, S., Bagby, G.]. and Diehl, A.M. (1992) Am. J. Physiol. 263, G579-G585 Feingold, K.K., Barker, M.R.,Jones, A.L. and Grunfeld, C. (1991) Hepatology 13, 773-779 Gnlli, M., Chiou,J.S. and Lenardo, M.J. (1993) Int. Rev. Cytol. 143, 1--{,2 Geller, D.A., Nussler, A.K., Di Silvio, M., Lowenstein, C.J., Shapiro, R.A., Wang, S.C., Simmons, R.L. and Billiar, T.R. (1993) Proc. Nat!. Acad. Sci. U.S.A. 90, 552-526 FitzGerald, M.J., Webber, E.M., Donovan,J.R. and Fausto, N. (1995) Cell Growth Diff. 6, 417-427 Scearce, L.M., Lee, J., Naji, L., Greenbaum, L., Cressman, D.E. and Taub, R. (1996) Cell Death Differ. 3, 51-59 Suzuki, H., Menegazzi, M., Carcereri de Prati, A., Mariotto, S. and Armato, U. (1995) Adv. Neuroimmunol. 5, 379-410 Obolenskaya, M.Y., Vamn, A.F., Mordvintcev, P.I., Mulsch, A. and Decker, K. (1994) Biochem. Biophys. Res. Commun. 202,571-576 Hortelano, S., Dewez, B., Genaro, A.M., Diaz-Guerra, M.J.M. and Bosca, L. (1995) Hepatology 21, 776--786 Xie, Q., Whisnant, R. and Nathan, C. (1993)}. Exp. Med. 177,1779-1784 Xie, Q., Kashiwabara, Y. and Nathan C. (1994) J. Bioi. Chern. 269, 4705-4708 Coni, P., Simbula, G., Carcereri de Prati, A., Menegazzi, M., Suzuki, H., Sarma, D.S.R., Ledda-Columbano, G.M. and Columbano, A. (1993). Hepatology 17, 1109-1116 Higgins, G.M. and Anderson, B.R. (1931) Arch. Pathol. 12, 186--202 Chirgwin,J.M., Pryzbyla, R.J., MacDonald, R.J. and Rutter, W.J. (1978) Biochemistry 18, 5294-5299 Adachi, H., Iida, S., Oguchi, S., Oshima, H., Suzuki, H., Nagasaki, K., Kawasaki, H., Sugimura, T. and Esumi, H. (1993) Eur.J. Biochem. 217,37-43 Schreiber, E., Mattias, P., Muller M.M. and Schaffiner, W. (1989) Nucleic Acids Res. 17,6419 Bradford, M.M (1976) AnaL Biochem 72,248-254 Columbano, A., Ledda-Columbano, G.M., Lee, G., Rajalakshmi, S. and Sarma, D.S.R. (1987) Cancer Res. 47,5557-5559 Menegazzi, M., Carcereri de Prati, A., Suzuki, H., Shinozuka, H., Pibiri, M., Piga, R., Columbano, A. and Ledda-Columbano, G.M. (1997) Hepatology 25,585-592 Nadal, C. (1992) in Liver Regeneration (Bernuau, D. and Feldmann, G.,eds.), pp. 123-127, John Libbey Eurotext, Montrouge, France Cressman, D.E., Greenbaum, L.E., Haber, B.A. and Taub, R. (1994) J. Bioi. Chern. 269, 30429-30435 Curran, R.D., Billiar, T.R., Stuehr, D.J., Ochoa, B.G., Harbrecht, B.G., Flint, S.G. and Simmons, R.L. (1990) Ann. Surg. 212, 462-471 Collart, M.A., Baeuerle, P. and Vassalli, P. (1990) Mol. Cell Bioi. 10, 1498-1506 Zawaski, K., Gruebele, A., Kaplan D., Reddy, S., Mortensen, A. and Novak, R F. (1993) Biochem. Biophys. Res. Commun. 197,585-590 Tugwood,J.D., Issemann, I., Anderson, R.G., Bundell, K.R., McPheat, W.L. and Green, S. (1992) EMBO J. 11,432-439 Bardot, 0., Aldridge, T.C., Latruffe, N.D. and Green, S. (1993) Bwchem. Biophys. Res. Commun. 192,37-45

The mechanism of nitric oxideinduced apoptosis in the pancreatic f3-cell may involve cGMP and protein kinase G Anne C. Loweth*, Gwyn T. Williams*, Roger D. Hurst:j:, John H.B. Scarpellot and Noel G. Morgan*§ Cellular Pharmacology Group, Departments of *Biological Sciences and tMedicine, Keele University, Staffordshire STS SBG, U.K. and :j:Department of Neurology, University College London, Gower Street, London, U.K.

Pancreatic islet 13-cells Mammalian fuel homoeostasis is maintained by the reciprocal effects of insulin and counter-regulatory hormones, including glucagon, on metabolic processes. Insulin and glucagon are secreted in a co-ordinated manner from pancreatic islets in response to nutritional, neuronal and hormonal signals. The central importance of insulin in glycaemic control has long been recognized; during the 7 decades since its identification by Banting and Best, much research has been focused on the biochemistry of insulin synthesis and secretion, with particular emphasis on the lesions underlying the pathology of diabetes mellitus. This condition results from an absolute or relative deficiency of circulating insulin and is often associated with impaired insulin production by islet [3-cells. Type I, or insulin-dependent diabetes mellitus (IDDM), is aT-cell-mediated autoimmune disease resulting from the specific destruction of islet [3-cells, with concomitant loss of 13-cell mass and insulin secretion. In IDDM, [3-cellloss progresses over a number of years before ultimately leading to overt diabetes, leaving the affected individual insulindependent for life. Type II, or non-insulin-dependent diabetes mellitus (NIDDM), is a complex, multifactorial disease characterized by an abnormal insulin secretory response to glucose and, in many cases, peripheral insulin resistance. 13-cellloss is also a feature of NIDDM since a decreased 13-cell mass is frequently observed in NIDDM patients at post-mortem examination. The mechanisms underlying 13-cell loss in IDDM and NIDDM are not well understood and the factors regulating 13-cell mass in normal individuals are also poorly characterized. One surprising feature that has emerged recently is that, far from being static, as was widely believed, the islet 13-cell population is dynamically responsive to changing circumstances throughout life.

§To whom correspondence should be addressed.

liS

I 16

A. C. Loweth et al.

Physiological !3-cell mass homoeostasis In humans, insulin-secreting fj-cells comprise some 60-80% of islet mass, and the islet mass itself changes in response to the overall demand for insulin. This process begins early in development since the relative islet mass occupies 20-30% of the pancreatic volume in the foetus but decreases to 1-2% in the adult, as the proportion of exocrine tissue increases. Despite this, the absolute islet mass actually increases 4-5-fold as adulthood is reached, reflecting the greater requirements of a larger body mass for insulin [1]. Another stimulus to increasing !3-cell mass is obesity. A post-mortem comparison of obese control subjects and NIDDM patients suggested that diabetic individuals may fail to respond adequately to the demand for greater fj-cell mass since obese NIDDM individuals were found to have only 50% of the number of fj-cells when compared with obese non-diabetics [2]. Pregnancy also creates greater demands for insulin. In pregnant rats, fj-cell mass increases by some 53% during gestation, and this extra tissue is then lost during the post-partum period [3]. The increase in mass is thought to reflect cell proliferation rather than hypertrophy, as the protein/DNA and insulin/DNA ratios remain unaltered during pregnancy [4,5]. In this context, gestational diabetes can be considered as a dysfunctional adaptive response to pregnancy, which results in failure to maintain adequate fj-cell numbers and leads to a shortfall in insulin release [6]. f3-cell replication has been measured at 3% per day in adult mice and rats, which would be sufficient to allow a doubling of the total fj-cell mass within 1 month were it not compensated by a similar rate of cell death [7]. Fluctuations in fj-cell numbers can, in theory, reflect a combination of cell proliferation, hypertrophy, atrophy and cell death, and these mechanisms must be triggered by particular stimuli. Glucose is known to stimulate fj-cell replication [8], whereas there is some evidence from implantation of insulinomas in rats that hypoglycaemia may induce fj-cell atrophy [9]. Certain growth factors and hormones are also thought to regulate fj-cell mass. Prolactin stimulates fj-cell proliferation in vitro [10] and may induce functional changes in insulin synthesis and secretion during pregnancy, in concert with placental lactogen [11,12]. Insulin-like growth factors I and II (IGF-1, IGF-11) may also regulate pancreatic cell growth, in that they stimulate thymidine incorporation in cultured islets [13]. Furthermore, IGF-1 expression increases after 90% pancreatectomy in rats [14], and hybridization in situ revealed that the areas in which enhanced IGF-1 expression occurred correlated with regions of greatest regeneration. Genetic control of islet cell replication has been suggested by the selective expression of a gene (termed reg for 'regenerating') in rat islets regenerating following surgical removal of 90% of pancreatic mass [15]. Increased expression of this gene has also been correlated with thymidine incorporation following treatment of cultured rat islets with various nutrients and growth factors [16] and it is thought that reg expression may be involved in islet repair mechanisms [17]. Based on this information, a dynamic model of fj-cell population control has been postulated [1], in which fj-cell turnover occurs at a constant basal rate but is rapidly responsive to changing needs. The initial response may involve hypertrophy of senescent cells but further demand for increased fj-cell mass leads

Mechanism of nitric oxide-induced apoptosis in the pancreatic ~-cell

to mitotic division of pre-senescent cells and perhaps differentiation of new [3cells from pancreatic ductal precursors. Conversely, decreased requirements for insulin, for example, following pregnancy or weight loss, probably result in [3-cell atrophy. If sustained, this leads to a decrease in cell population, effected by a reduced replication rate or an increase in cell death.

[3-Cell death in IDDM The possibility that an increased rate of cell death might be involved in the control of [3-cell mass may also have relevance to the alterations in homoeostasis that contribute to the pathology of diabetes. In particular, recent research has been directed towards understanding the physiological control of apoptosis and, specifically, to determining whether this apparently ubiquitous process may contribute to [3-celllos~ in IDDM.

Cytokine-mediated cell death Although the factors which initiate the autoimmune attack leading to [3-cell destruction are not understood, there is considerable evidence to suggest that cytokines, secreted by macrophages and T-cells infiltrating the pancreas, provoke [3-cell cytotoxicity. lnterferon--y (IFN--y), tumour necrosis factor-a (TNF-a) and, in particular, interleukin-1[3 (IL-1[3) are among the cytokines implicated in [3-cell damage. IL-1[3 is reported to have a bimodal effect in rat islets and clonal[3-cells, in that acute doses stimulate insulin secretion whereas multiple inhibitory effects result from longer-term treatment. These include inhibition of glucose-induced insulin secretion and glucose oxidation, depletion of NAD+, cAMP and ATP levels and reduction in DNA synthesis [18]. Human islets are also susceptible to cytokine damage, although intact human islets generally appear to be more resistant to the effects of cytokines than their rodent counterparts [19]. IL-1[3, IFN--y and TNF-a, alone and in combination, have been shown to destroy cultured human islet [3-cells [20,21] and exposure of human islets to this combinatiOn of cytokines is inhibitory to insulin secretion [22].

NO-mediated cell death The intracellular pathways that mediate the effects of cytokines are still under investigation although a strong candidate as a transducer of cytokine signalling is the free radical, nitric oxide (NO). NO acts as a signalling molecule in the nervous system; it also mediates endothelium-derived relaxing factor-induced vasodilation and is known to inhibit platelet aggregation [23]. NO is produced from the oxidation of L-arginine to yield citrulline and NO, a reaction catalysed by NO synthase (NOS). Rat islets and clonal[3-cells have been shown to contain a constitutive form of the enzyme as well as a calcium-independent inducible NOS (iNOS) [24,25]. There is substantial evidence in support of NO mediating

I 17

I 18

AC. Loweth et al.

cytokine damage in a range of other cell types, including macrophages, smoothmuscle cells and hepatocytes [23]. NOS inhibitors, such as the arginine analogues Nw-nitro-L-arginine methyl ester (L-NAME) and N .. -monomethyl-L-arginine (L-NMMA), have been shown to influence cytokine effects. L-NAME is reported to inhibit IL-113-induced inhibition of insulin secretion as well as inhibiting cytokine-induced NO production [26,27]. L-NMMA also blocks cytokine effects, including the inhibition of mitochondrial enzyme activity [28,29]. The cytotoxic effects of macrophages on islet cells are also, at least partially, dependent on NO production [30], although NO-independent routes of cytokine damage have also been reported. For example, nicotinamide, a free-radical scavenger, was reported to inhibit cytokine-induced cell death without affecting NO production [31]. In addition, cytokine-induced damage to human islets may be independent of NO production under certain conditions [32] although NO donors have recently been shown to impair human islet function and to promote cell death in this tissue [33]. There are several potential mechanisms by which NO may be cytotoxic. One of these involves the formation of iron-nitrosyl complexes with FeScomaining enzymes, such as aconitase. Inhibition of this enzyme leads to a reduction in glucose and amino acid metabolism and a fall in ATP generation [34]. As a free radical and a progenitOr of other reactive oxygen species, NO may induce cell damage as a result of oxidative stress. The cytotoxic effects of cytokines on rat islets have been ascribed to free-radical formation leading to lipid peroxidation [35]. NO also promotes direct DNA damage in islet cells by deaminating purines and pyrimidines [36), which is preventable by NOS inhibitors (37,38]. Although the mechanisms by which NO evokes cytokine-mediated islet toxicity are not fully understood, an important component undoubtedly involves DNA damage. Since DNA damage is thought to act as a trigger for apoptosis, this ability of NO to induce DNA damage and cell death raises the possibility that cell Figure 1

5.0

~

1.0 0.0

Lt-t--t--t-+-t-+-t--t--+-+--1-=:;:::;::~=+=+::::f=:;:::,.._~-+

0 1 2 3

4 5 6 7 8

9 10 11 12 1314 15 161718 19 20 21 22 2324

Time (h)

NO production from 200 j.i.M GSNO in culture medium NO production from 200 JLM GSNO in RPM I 1640 culture medium was measured using on NO

electrode, over a penod of 24 h. Low levels of NO ( 1-2 JLM) were sustained for up to 5 h after addition of the GSNO and NO production wos detectable for up to 20 h.

Mechanism of nitric oxide-induced apoptosis in the pancreatic ~-cell

I 19

A

0

c::

KA

J!

200

'

*

* fl

-

-,.I.

100

c

Ol

•

BrGMP

10

3

M

10

1

M

#.

,:X:

.: