Uses of Inorganic Chemistry in Medicine

.

Richmond. Virginia.Uses of Inorganic Chemistry in Medicine Edited by Nicholas P. Farrell Virginia Commonwealth University. U S A RSmC ROYAL SOCIETY OF CHEMISTRY .

UK For further information see our web site at www. in any form or by any means. Apart from any fair dealing for the purposes o research or private study. Bodmin. 0The Royal Society of Chemistry 1999 All rights reserved.Cover illustration by Philip Mattes ISBN 0-85404-444-2 A catalogue record for this book is available from the British Library. Published by The Royal Society of Chemistry.rsc. this publication may not be reproduced. stored or transmitted. or in the case of f reprographic reproduction only in accordance with the terms of the licences issued by the Copyright Licensing Agency in the UK. 1988. Milton Road. without the prior permission in writing o The Royal Society of Chemistry. or criticism or f review as permitted under the terms of the UK Copyright. or in accordance with the terms of the licences issued by the apropriate Reproduction Rights Organization outside the UK. UK . Cornwall. Cambridge CB4 OWF. Enquiries concerning reproduction outside the terms stated here should be sent to The Royal Society of Chemistry ut the address printed on this page. Science Park. Designs and Patents Act.org Typeset in Great Britain by Vision Typesetting. Manchester Printed and bound by M P G Books Ltd. Thomas Graham House.

one of the most interdisciplinary. Further. Understanding of how chemical structure determines the above factors is a critical cornerstone for the future. types of ligands and kinetic reactivity may be tailored to the use of medicinal chemistry. . Studies in the field of inorganic-based pharmaceuticals continue to expand. by definition. To do this. While emphasis tends to be placed on mechanistic studies (target interactions) the pharmacokinetics and tissue distribution are equally important in determining whether a drug will have a sufficient therapeutic index to be of clinical use. for example. Thus. it is important to have as great as possible an understanding and appreciation of the myriad factors which determine the success or otherwise of a clinical agent. Likewise. the trend. Advances in genetic and structural biology underscore the genetic basis of many diseases as residing in metalloproteins. opening new leads for drug design. rather than as isolated examples of inorganic chemistry. especially in cancer. These findings offer a rare opportunity for the inorganic chemist to contribute to the genetic understanding of such diseases. descriptions in this book interweave chemistry with pharmacology and molecular biology. New developments in metal-based drugs must be seen in the light of utilising the body of knowledge on existing drugs. away from an approach of simple testing to drug design based on specific protein and DNA targets also offers an opportunity for the inorganic chemist. The chapters here have been chosen for this purpose as they all strive to combine the chemical understanding of inorganic molecules with their biological activity. The field of drug design and discovery is. More detailed understanding of the current drugs continues to be accumulated. It is the purpose of this monograph to present overviews of the chemistry and biology both of wellknown clinically used agents and of new potential applications. we have selected case studies for the reader to study.Preface Any compilation of clinically used drugs must include important inorganic pharmaceuticals such as lithium carbonate and cisplatin. it is important to know and understand how the many structural features of inorganic compounds capable of subtle fine-tuning such as coordination number and geometry. The introduction in Chapter 1 gives an overview and discusses the outline of the book and will not be repeated here. which would be suitable for current much-studied targets such as topisomerase and telomerase. while yet representing a relatively small percentage of total drugs. What are the design features of a metal-based agent. the inorganic-based drugs fulfil important roles. In the last years advances have been made on several fronts. Rather than attempt an exhaustive listing of all metal-based drugs.

This volume was long in gestation. so that future examples of metal-based drugs will continue to appear. Nicholas P. I thank Dorothy Silvers for her wonderful editing and finally my family for their support. Farrell .vi Preface to name but two? It is our hope that reading these reviews will excite the reader and especially young researchers to think in similar directions. I thank the authors for their continued enthusiasm and patience.

Farrell 1 Introduction 2 Metal Ions in Disease.Contents Chapter 1 Overview Nicholas P. The Use of Chelating Agents Metalloproteins as Drug Targets Zinc and the Human Immunodeficiency Virus Zinc Fingers as Medicinal Target Matrix Metalloproteinases 3 Modulation of Cellular Responses by Metal-containing Drugs 4 Metal-based Chemotherapeutic Drugs 5 Metal Complexes as Diagnostic Agents 6 Summary and Acknowledgements References 1 1 2 2 4 4 4 6 6 7 9 9 11 Chapter 2 Biomedical Uses of Lithium Nicholas J . Birch Introduction Background New Uses of Lithium Chemistry of Lithium Distribution of Lithium in the Body and in Cells Studies Using Lithium Isotopes Tissue Localisation and Transport of Lithium Cellular Localisation of Lithium Biochemistry of Lithium Magnesium and Calcium Lithium and the Phosphoinositide Signalling System Lithium at the Cell Periphery: A Novel Viewpoint Conclusion References 11 11 13 14 15 16 16 18 18 18 19 20 21 21 .

Anti-tumour and Anti-HIV Activity C. Contents Chapter 3 Gold Complexes with Anti-arthritic. Rhodes 1 Introduction 2 Physical Properties of NO 3 Measurement of NO in Biological and Chemical Systems Direct Measurement Indirect Measurement 4 Nitric Oxide and the Respiratory System 5 Nitric Oxide and the Cardiovascular System 6 Nitric Oxide and the Nervous System Peripheral Nervous System 58 59 60 60 61 64 66 69 69 ..The Thiol Shuttle Model Protein Chemistry Serum Albumin Hemoglobin Aurothioneins Glutathione Peroxidase Cyanide Metabolites Oxidation States In Vivo An tigenic Peptides 4 Anti-Tumour Activity Auranofin and Analogues Bis(dipheny1phosphine)ethaneGold Complexes Gold Complexes of Known Anti-Tumour Agents Cisplatin Analogues Prospects for Gold-Based Anti-Cancer Drugs 5 Anti-HIV Activity 6 Conclusions References 26 26 26 26 28 29 29 30 31 31 32 32 33 35 35 37 38 40 42 43 44 47 47 50 51 51 52 52 53 54 58 Chapter 4 Nitric Oxide in Physiology and Medicine Anthony R. Frank Shaw I I I 1 Introduction Chrysotherapy History of Medicinal Uses Rheumatoid Arthritis and Putative Mechanisms of Action 2 Gold Chemistry Oxidation States Gold(1) Complexes Gold(@ Complexes Oxidation-Reduction Potentials 3 Gold Biochemistry and Pharmacology I n Vivo metabolism and Ligand Displacement Cellular Chemistry . Butler and Peter S.Vlll ..

Riley 1 Introduction 2 Results and Discussion Therapeutic Strategy of Superoxide Dismutase Mimics Stopped Flow Kinetic Analysis: A Direct Assay for Superoxide Dismutase Activity Design of Manganese-based Superoxide Dismutase Mimics Advantages of Superoxide Dismutase Mimics over Superoxide Dismutase Enzymes Conformational Control of Superoxide Dismutase Activity and Stability Inhibition of Neutrophil-Mediated Human Aortic Endothelial Cell Injury by Superoxide Dismutase Mimics Lipophilicity Inhibition of Myocardial Ischemia-Reperfusion Injury in the Isolated Heart Protection against Myocardial Ischemia-Reperfusion Injury I n Viuo Potentiation of Nitric Oxide Anti-Inflammatory Activities of Manganese@)-based Superoxide Dismutase Mimics 3 Summary References Chapter 6 Vanadium Compounds as Possible Insulin Modifiers Chris Oruig. Thompson. Katherine H . Cam and John H . Weiss and Dennis P. McNeil 1 Introduction 2 Characterization of Vanadium’s Insulin-mimetic Effects 3 Sites of Action of Vanadium 4 Animal Studies and Human Trials 5 Toxicological Considerations 6 Improved Tissue Uptake with Metal Chelation 7 Summary References 70 71 71 72 74 77 77 78 78 79 79 81 82 85 85 87 87 88 88 90 90 93 93 95 96 98 100 101 104 104 . Margaret C.Contents ix Synaptic Plasticity Cerebral Circulation Neurodegener ation 7 Nitric Oxide and the Immune System References Chapter 5 Therapeutic Aspects of Manganese(r1)-based Superoxide Dismutase Mimics Randy H .

X Contents Chapter 7 Cisplatin-based Anticancer Agents Lloyd R. Kelland 1 Introduction 2 Clinical Properties Cisplatin versus Carboplatin Iproplatin Determination of Platinum Drug Levels and Pharmacokinetics 3 Platinum Chemistry Mechanism of Action Structure-specific Damage-Recognition Proteins 4 Mechanisms of Resistance to Cisplatin/Carboplatin 5 Circumvention of Tumour Resistance to Cisplatin Development of New Platinum Drugs Diaminocyclohexane Platinum Complexes Orally Active Platinum Complexes: JM2 16 0ther Platinum-based Agents Dose Intensification of Cisplatin/Carboplatin Combinations with Other Anticancer Agents: Paclitaxel Modulation of Platinum Resistance Mechanisms 6 Summary References Chapter 8 Dinuclear and Trinuclear Platinum Anticancer Agents Nicholas Farrell and Silvano Spinelli 1 Introduction 2 Dinuclear and Trinuclear Platinum Complexes as Anticancer Agents 3 Biological Activity of Polynuclear Platinum Complexes. Summary and p53 Status of Human Tumors Treatable by BBR3464 Biological Activity and p53 Status Comparison with Other Clinical Cross-linking Agents 4 Structure-Activity Relationships in Polynuclear Platinum Complexes 5 DNA Binding of Polynuclear Platinum Complexes Cooperative Effects in the Solution Structures of Sitespecific (Pt.Pt) Interstrand Cross-links 6 Summary References 109 109 109 110 111 112 113 114 115 116 118 118 118 118 119 119 120 120 120 120 124 124 125 126 126 127 128 129 132 133 134 .

Adriamycin and Other Cytotoxic Agents That Require Iron or Copper David H . Sreedevi Nyayapati. 4 Redox Active Copper Complexes and the Design of New Drugs References 135 135 136 136 137 138 138 138 140 143 144 145 146 148 149 150 150 153 158 Subject Index . Patricia Fulmer and William E. Petering.Contents xi Chapter 9 Oxidation Damage by Bleomycin. Antholine 1 Introduction 2 Bleomycin Clinical Use Animal and Cellular Studies on the Role of Iron in the Cytotoxicity of Bleomycin Cellular Properties of DNA Strand Scission Biochemical Mechanisms of DNA damage Characterization of DNA Damage by FeBlm Solution Properties of Metallobleomycins Related to the DNA Damage Reaction Properties of Metallobleomycins Bound to DNA Structural Properties of CoBlm Species in Solution and Bound to DNA Role of the Metal Domain in Site Selectivity 3 Adriamycin Aspects of Clinical Use Cytotoxic Mechanisms of Reaction of Adriamycin with Iron Biological Model Reactions of Fe(m)Adr. Jun Xiao.

.

incorporation into protein and the nature and function of metalloproteins.especially with respect to the role of metalloproteins in human health and disease. metal-based drugs where the central metal ion is usually the key feature of the mechanism of action. the drug is usually in direct contact with medium throughout the experiment. rather than a simple listing of clinical and potential uses. Any consideration therefore of the uses of inorganic chemistry in medicine must bridge at least two areas . 1001 W. Metal-based drugs are a commercially important sector of the pharmaceutical business. Nevertheless. is a challenge. USA Introduction The field of inorganic chemistry in medicine may usefully be divided into two main categories . it is important to attempt to do so to harness the diversity of inorganic chemistry to systematic developments in medicine. Most mechanistic work is performed in tissue culture or with isolated proteins. Richmond. FARRELL Department of Chemistry. Medicinal chemistry requires intimate knowledge of the metabolism and stability. whether free or protein-bound. In tissue culture assays to measure the efficacyof a potential drug in inhibiting cell growth.drugs which target metal ions in some form. Alternatively. How to approach this field from a didactic and systematic manner. There is not always a direct extrapolation .CHAPTER 1 Overview NICHOLAS P. Main Street. Virginia Commonwealth University. and secondly. VA 23284. Applications continue to grow and approaches to further clinically useful agents are ever more sophisticated.’ Advances in our understanding of how cells process metals and the genetic basis of disease is naturally expanding the traditional directions of bioinorganic chemistry toward an appreciation of its medical importance . as well as target interactions of the drug.bioinorganic chemistry and medicinal chemistry. genetic factors may lead to failure to incorporate and subsequent metabolic disorders may be caused by free metal ions. In a ‘steady-state’ environment all essential metals are incorporated in the right place at the right time and the organism functions normally. DNA and/or RNA. Bioinorganic chemistry is best considered as understanding all aspects of the role of metal ions in biology and has been traditionally heavily involved in understanding their processing.

In this volume we review aspects of the use of inorganic compounds as drugs and chemotherapeutic agents. A major consideration for the improvement of chelating agents is of course that of metal ion selectivity .2 Overview to the clinically relevant in vivo situation when biodistribution and pharmacokinetics play an increasingly important role in determining drug efficacy.2 In the former case. mercury and lead. The Use of Chelating Agents It is well understood that many metals are essential for the human organism and endogenous concentrations are tabulated in most bioinorganic chemistry textb o o k ~ . a corollary of this situation is that uncontrolled mobilization ~. The reader is referred to the many comprehensive reviews in both bioinorganic and medicinal chemistry for further reading. We do not intend to be comprehensive but rather present specific case studies for reading. Finally. The status with respect to some known drugs is reviewed as well as introductions to newer drugs of potential clinical significance. may also inform on approaches to in vivo efficacy.a drug to reverse a stroke must be aware of the severity of that stroke and concentrations adjusted accordingly. Metalloproteins as Drug Targets A more recent and related question to the specificity of chelating agents is that of metalloprotein targets. The classic examples are those of iron and copper overload. It is not surprising that many metalloproteins and metal- . which untreated is inevitably fatal.2 Further. A patient who submits to an anaesthetic does not expect to be deprived of feeling forever. a graded response is required to balance effects .~ may lead to the presence of excess free metal ion. Alloyed to this is the prospect of disease occurring through adventitious exposure to toxic doses of either essential elements and non-essential elements such as cadmium.few chelating agents can be stated to be specific for simply one metal ion. the drug action must be rapid and essentially reversible. In this introduction we give a broad overview of the area from a didactic point of view. (ii)inorganic-based drugs acting by a pharmacodynamic mechanism. Nevertheless the mechanistic information of tissue culture experiments is very useful and.typical examples are shown in Figure 1. 2 Metal Ions in Disease. Many compounds with exciting in vitro results have failed to display the same promise in vivo. with subsequent health problems. Wilson’s disease is an autosomal disorder of copper accumulation. (iii) inorganic-based chemotherapeutic agents and (iv) inorganic-based imaging agents. medicinal chemistry distinguishes between drugs acting by a pharmacodynamic mechanism and chemotherapeutic drugs. The treatments for copper and iron overload are well documented and a list of clinically used chelating agents is found in most textbooks . aside from target interactions. Chemotherapeutic agents on the other hand involve cell killing. In attempting to do so.Their chemistry and toxicology is also very well documented. four main subdivisions logically present themselves: (i) uses of chelating agents to sequester specific metal ions or metalloproteins. an irreversible process.However.

More recently. few noxious effects of excess zinc have been observed and zinc per se is probably one of the least toxic metals.8 These functions include DNA transcription and regulation as well as oxidation and hydrolysis. Zinc is the second most prominent trace metal in the human body after iron. zinc proteins have been recognised as attractive targets for chemotherapy.Nicholas P . there are two principal areas of interest in pharmaceutical laboratories . Because zinc is not redox active its catalytic functions derive from its properties as a Lewis acid. 6i7 A current and very relevant example is zinc.inhibition of the matrix metalloproteinase enzymes such as collagenase as an approach to treatment of metastatic cancer and inhibition of zinc finger activity as a novel chemotherapeutic attack against HIV infectivity. fulfilling both structural and catalytic roles. cleavage of peptide bonds as well as formation of phosphodiester bonds. Especially. . the diiron enzyme essential for de novo synthesis of deoxyribonucleotides for DNA synthesis has long been recognised as a drug target. Drug design and discovery relies more and more on the elucidation of the three-dimensional structure of a target by X-ray crystallography or nuclear magnetic resonance methods. Zinc is involved in a large number of enzymatic functions. Metalloproteins are being increasingly recognized and examined as drug targets. Ribonucleotide reductase. While deficiency of zinc may cause growth effects.Penicillamine CH2-SH I CH2-SH Cysteamine H*N(CH2)5N-C( CH& C0 HN(CH2) 5 N-C( CH2)2C0 HN (CH2)5 N-CCH3 I II I II I II HO 0 HO 0 HO 0 Desferrioxamine B Figure 1 Structures of some clinically used chelating agents for treatment of copper and iron overload loenzymes play vital metabolic roles as well as being critical in genetic information transfer. followed by modelling and synthesis of potential inhibitors of the protein or enzyme active site. The pharmaceutical and chemical properties of chelating thiosemicarbazones and their potential interference with the active site iron moieties has been an extensively studied problem. Farrell y 3 3 HjC-C-SH I CH-NH2 I C02H D.These apparently diverse goals are united by the common feature that the active-site zinc is in both cases the target of attack.

Matrix Me ta1loproteinases (MMPs) Matrix metalloproteinases are involved in extracellular matrix degradation during cell migration. new effective therapies producing long-lasting permanent effects against HIV infectivity are still urgently needed. . which repeated looked like 'fingers'..-Cys-X. bone remodeling and embryo development. 'u4 While these results are highly promising.. The MMPs cleave one or more components of the extracellular matrix for example collagenase cleaves a specific glycine-leucine bond in collagen. Figure 2). '' A principal approach has been to design chelating agents such as dithiobisbenzamides (DIBAs. X-ray crystallographic evidence has now been obtained for such structures. A relatively recent target for drug design has been the zinc fingers of the nucleocapsid protein.' The role of matrix metalloproteinases in metastasis has sparked intense inter- . deriving from the specific conformations proteins adapt upon complexation by the metal.' Model building suggested that zinc binding to His and Cys folded the protein into a conformation. The MMP activity is inhibited in normal tissue by endogenous tissue inhibitors of metalloproteinases (TIMPs).000 natural and synthetic agents during the last fifteen years. ddI and AZT in conjunction with protease inhibitors is a very promising approach to achieve permanent therapeutic effects. new targets within the viral cycle need to be identified and understood on a molecular basis to allow development of drug design strategies. suggested a role for metal-binding. All these functions are highly regulated. However. Part of the rationale behind combination therapy has been to use drugs which act on different parts of the viral cycle. despite the intense efforts and screening of nearly 220.-His-X. Effective therapies for AIDS are still urgently required. Normal processes include wound healing. Currently. which chemically modify the zinc finger cysteine residues resulting in zinc ejection from the fingers with resultant inhibition of HIV replication..4 Overview Zinc and the Human Immunodeficiency Virus Zinc Fingers as Medicinal Target The role of zinc in transfer of genetic information is believed to be structural. As such.-His-X. Many transcription factors (required for RNA transcription) contain zinc. combination therapy using especially purine and pyrimidine analogs such as ddC. thus limiting development of resistance. they are not optimal as yet but serve as a basis for further rational target-based drug design.' The existence of metal-binding domains in regulatory proteins was first postulated because researchers noted that the systematic repeats of cysteine and histidine residues in X.'o The human immunodeficiency virus type 1 (HIV-1) is the etiologic agent of acquired immune deficiency syndrome (AIDS).-Cys-X. .' Abnormal regulation (elevated peptidase activity) occurs in rheumatoid arthritis and invasion and metastasis of neoplastic cells.

Farrell 5 DIBA-I PD022551 Figure 2 Structures of chelating agents designed to inhibit Zn-jngerfunction. The medicinal approach to drug development is to develop as inhibitors substrate analogs which will competitively bind to the zinc active ~ i t e . Indeed a most interesting case is posed by approaches to attacking zinc-finger sites with metal-based compounds such as authiomalate. 'The ~ first drug to enter clinical trials from this approach is batimastat (Figure 3). even in consideration of chelating agents we see that detailed knowledge of metalloprotein structure leads to new approaches to specijc chelating agents. inhibition of the abnormal regulation of MMPs would in principle retard or eliminate metastasis. . This area is a very active one in drug development and almost all pharmaceutical companies have drug development The coming years should see significant advances in drug efficacy and the validation of this approach. See references 12-14 est as a target of chemotherapeutic intervention .very briefly.21In this approach the cysteines of the zinc fingers are the specific targets and metal-ligand replacement is a viable goal.Nicholas P . Thus.

Since the recognition of the messenger role of the small inorganic molecule NO in the early 199Os. (Chapter 2) and gold-based antiarthritic drugs (Chapter 3). chemotherapeutic . In the first two. This definition therefore covers the antibacterial. that of elimination of the unwanted cells. From the perspective of inorganic drugs in medicine. Clinically used examples discussed in this volume are Li. Sodium nitroprusside is used in cases of severe heart failure and is a short-acting hypotensive drug with a duration lasting 1-10 minutes. Interesting developments may be expected as inorganic chemists design newer M-NO molecules with appropriate release rates as well as molecules which may scavenge endogenous NO or its potentially damaging oxidation products such as peroxynitrite. Thus. Chapters 5 and 6 introduce two exciting and potentially significant advances in the stable of metal-based drugs manganese-based superoxide dismutase mimics (Chapter 5 ) and vanadiumbased insulin mimetics (Chapter 6).6 Overview Bastimastat Figure 3 Structure of the matrix metalloproteinase inhibitor batimastat 3 Modulation of Cellular Responses by Metal-containing Drugs Inorganic drugs may be recognised as acting through a pharmacodynamic mechanism . The further potential for gold-based chemotherapeutic agents is summarized in Dr Shaw’s contribution. the organism is strictly not different but the treatment has a common aim. In the case of cancer. a significant body of data has been accumulated (Chapter 4). the understanding of the importance of NO also made clear how the vasodilating properties of nitroprusside are manifested .modulating cellular responses. This consideration is an excellent example of the difference between pharmacodynamic and chemotherapeutic drugs. a family of diseases characterized by uncontrolled cellular proliferation.through release of NO. antiviral and anticancer agents. 4 Metal-based Chemotherapeutic Drugs Chemotherapy is the use of drugs to injure an invading organism without injury to the host. the invading organism is clearly distinct from the host.CO.

few are likely to advance to full clinical use. The current status is outlined in Chapter 7 by Kelland.27 Finally the natural product bleomycin is always classified as an inorganic-based drug through the imputed DNA strand breakage mechanism facilitated by oxygen radical production on iron. must induce an irreversible cytotoxic effect. none has as yet achieved full clinical use. Clearly. This has been successful to the point that. no pharmacodynamic or chemotherapeutic end is desired. Finally. lead to compounds with a spectrum of clinical activity genuinely complementary to the parent drug. let alone the status of cisplatin. .stability and water solubility are paramount. Clinical experience with both gadolinium28 and technetium2’ agents is summarized in appropriate pharamceutical reference books. We considered that future discovery of clinically useful platinum agents was likely to arise with ‘non-classical’ structures. Will clinical success in the treatment of cancer be limited to platinum-containing drugs? The discovery of the anticancer activity of cisplatin sparked intense interest and research to find other metal-containing anticancer agents.22-26Unfortunately. Rather.Nicholas P. at time of writing. in contrast to pharmacodynamic drugs. The two principal sets are technetiumbased imaging agents and paramagnetic MRI contrast agents.such as silver sulfadiazene and mercurochrome. a novel trinuclear platinum agent has just entered clinical trials. imaging of tissue is achieved. In the platinum field. The considerations for stability are important to maintain specific tissue imaging and safety. Their uses and purported mechanisms have been summarized in a previous monograph22and will not be presented in great length in this volume. This latter consideration led us to formulate the hypothesis that development of platinum compounds structurally dissimilar to cisplatin may. the agent must be relatively unreactive and not be rapidly metabolized or degraded. The chemistry and biology are covered in Chapter 9. This effort has been well documented and there are now many distinct classes of metal-based drugs with antitumour activity in experimental models. by virtue of formation of different types of Pt-DNA adducts. Many common antibacterial agents are silver. All direct structural analogs of cisplatin produce a very similar array of adducts on target DNA and it is therefore not surprising that they induce similar biological consequences. Another possibly exciting development is the recognition of certain ruthenium compounds as metastatic poisons rather than cytotoxic agents. the strict reliance on analog development based on the cisplatin structure has produced many promising compounds but.and mercury-based . Chapter 8 summarizes the chemical and biological activity of this important new agent. again. Farrell 7 drugs. The considerations for clinical development of imaging and contrast agents are somewhat similar to those for drugs . clearance of the agent must also be relatively rapid. 5 Metal Complexes as Diagnostic Agents A further subset of inorganic drugs in medicine is comprised of diagnostic agents. By far the greatest success of inorganic chemotherapy is the advent of cisplatin and carboplatin into the clinic. In this application.

8 Overview ““Y N C 111 CNR Tc(HMPAO) Figure 4 Structures of clinically useful technetium imaging agents Tc(MIBI) R=H D03A R = CH2CH(OH)CH3 HP-D03A DOTA R = CH2COOH OH Figure 5 Structures of Gd-based MRI contrast agents .

J. 13 P. Inorganic Biochemistry. 1995.G. p. Brew. 1995. Enzyme Regul. Pabo. Much useful information can also be found in encyclopedic handbook^.C. L. 3 J. 10 N.N.L.77.y Int.-X. Royal Pharmaceutical Society.31*32 Future generations of agents may include larger chelates such as the ‘texaphrins’.33 Finally.^' Finally it is a pleasure for me to acknowledge the assistance and understanding of all the contributors and of The Royal Society of Chemistry.19. A. Williams. Cory.H. Zaharevitz. Clanton.. 1996. Gomis-Ruth. A.V. Science. Otsuka.W. An Introduction. 1979.R. Cory. Basel.. Summers. 1995.F. Lorico. Med. Besides the principal references given here. 1991. M. . Stalling. G. K. References 1 T. Chem. 55. the in v i m use of metal radioisotopes in cancer detection and imaging is an important and well documented use of metal c h e l a t e ~ .Nicholas P .V.O. Jolles and H. Schaeffer. Verlag. f eds. Rappa. J. eds. R. 2 A. H.W.N. Farrell 9 Selected examples of clinically used technetium compounds are shown in Figure 4.S. Many examples are by now well documented but the field remains an active one for research and improvement. Scholten. Bourenkov.C. P. Bartunik and W..F.G. New York. 6th Edn. Cowan. M.C. Buckheit Jr. 7 E. Science. L. Cove11 J. Moore and A. G. Advan.J.597. Klug and J. Oxford. C. 14 M. Maloney. Sugiura. A.. Wiley and Sons. 4 J.35. in Interface between Chemistry and Biology. Sartorelli. Med.M. USA. 1989.239. Sartorelli.. Science. The challenges of further tissue specificity is an active one while the technology to deliver and prepare coordination complexes with the radioactive isotope is ad~anced. de G u m a n . 31st Edn. in Inhibitors ofRibonucleotide Diphosphate Reductase Activity. J .-S. VCH..809. New York.252. J.A. Tummino. Suzuki...384. 9 A. D.P. K.C. 1997. M. an example of which is also shown in Figure 5. h a d . T. 1991. ~ ~ 6 Summary and Acknowledgements This necessarily brief overview shows the wide scope of inorganic chemistry in medicine. R. Domagala and D. Yoshida.C.~’ Likewise. Schwabe. H.P. 12 W. Pergamon Press. Auld.J. D. M-. P. in Martindale. Betz.3606. Bode. Pappalardo.93. Fujita and Y. O’Halloran. p. Nature.39. Gogliotti. 1995. 8 B. Section 28. 1996. Sci. 6 J.-M. Chem. 1994.F. Graham. Proc.D. Albert.J.G. Liu.P. New York. 203. Cory and A. FASEB J. many aspects of this subject are covered in specific volumes such as those of Metal Ions in Biology.4267. C. Nagase. Cell Bio1.715. Huber. The Extra Pharmacopoeia. T. Vallee and D. K. 259.261. P. 1993.37. Pavlevitch and C. 1998.A. London. p.969. Summers. 15 F. 973. The Biological Chemistry o the Elements.279. J. Natl. 11 R. L. Holler. Hupe. Lin and A. 5 Chelating Agents Antidotes and Antagonists. Baramova and J. R. Borer and M. 16 E. Selective Toxicity.G. Wallqvist and D. Bergner. Rice. f Clarendon Press. Jornvall. da Silva and R. N. 1993. Harvey. Foidart. International Encyclopedia o Pharmacology and Therapeutics.. the use of gadolinium agents in MRI work is now well established and the ‘classic’chelate structures first developed are shown in Figure 5.1996.389.9.

Vol.B. Sutherland. deFazio.J. Biol. C. Royal Pharmaceutical Society. Chem. 32 R.. D. London. Gray J . 18 B.Ugo. McElroy. Adu. New York. Sava..M.D. A. 1990. VCH. Gregson. Chambers. p. G. p. Ward. Fricker.J. VOl. eds. D. SOC. A. Acad. Spatola and R. ed. 215. 1994.J.H. The Extra Pharmacopoeia.263. A. Vols. London.. 1994.O.M. 24 Metal Complexes in Cancer chemotherapy.F. London. . 33 J.P. S. Pacor and S. Science.. Jordan. p.4497. ed. Drewry. The Extra Pharmacopoeia. 1989. 1990. P. Chapman and Hall. London.. B. 1009. McGeehan. 1989. Handel. in Catalysis by Metal Complexes. H. Day and R. Am. J . Keppler.253.5199.265. M. 1994. Bickett. Hnatowich. Foley. B.S Stack. 30 M. 31st Edn.10 Overview 17 P.9. Fricker. Hassell. CRC Press. 20 K. Br. 10.D. 1996. G. 19 F. 25 I. Dordrecht. Davies. Lauffer. Marcelf Dekker Inc. McElroy.674. 1991. Myers. Schoenen and P.. Berthon.G. Med. 21 M. 1463.37. Silvestru. Chem. Luther. in Metal Compounds in Cancer Therapy. M.303.N. 1995. Sessler et al. 22 N.F. in Martindale.L.375. R. 31 P. Vol.293.79.. D. Proc. J. A. 35 Handbook o Metal-Ligand Interactions in Biological Fluids.L. Zorzet. Lambert and S. Med. Reidel-Kluwer Academic Press. Natl. D. 1995. Boca Raton.35. A. Norton.87. 1987. ed.B. 1990. 31st Edn. 10368. M.901. 1 and 2.. Vol. Chapman and Hall. McGeehan. K.A. Reu. USA. Podbielski. Longley. Farrell. 27 G.M. Brown. Reu. J. F. Pharmacology (Life Science Aduances). D. Brown. Chem. Armstrong and S.R. Drewry. Keevil. 1. M. 1993. 1996. Chem. Enzyme Regul.M. 1. 1995. in Martindale. Vol. 23 S. 11. Haiduc and C. 1993. 28 Contrast Media. Organometallics in Cancer Chemotherapy.105.L. James and R.92.R. 2.115. S. Weigl.K. Deaton. Coord. 26 Progress in Clinical Biochemistry and Medicine. J. Chem. Transition Metal Complexes as Drugs and Chemotherapeutic Agents.78. 1994.. G. Metal Compounds in Cancer Therapy. Salovich. 34 D. 29 Radiopharmaceuticals. Darlak. 1987.. D. Basel. Lovejoy. M.P. Sci. Royal Pharmaceutical Society.L. Clarke and L.

The first comprehensive pharmacological review of lithium was carried out by Mogens Schou in 1957. Staffordshire WV8 2ER. Lithium occurs naturally in biological tissues and hence is incorporated into foodstuffs. and yet it has major and diverse effects in the body. with supposed medicinal properties. retroviral infections.2 It occurs widely in drinking water. Recent studies of the biological effects of lithium have stretched the scope of its potential uses to include treatments for major disease types such as neoplasms. and electrical batteries. The fascination of lithium. UK 1 Introduction Lithium is a chemical and biological oddity. its biological effects are not easy to relate to specific loci of action. is that it is the lightest and smallest solid element. Natural waters that contain higher concentrations of this and other metals frequently are designated ‘mineral waters’. Codsall. Only a small proportion of lithium production (less than 1% of the total) is used in medicine. with apparently little subtle chemistry to its credit. it is clear that. Since the tiny lithium ion can cause such significant changes in biological systems. and immunological disorders. BIRCH Academic Consultancy Services Limited.’ 2 Background Lithium extraction takes place mainly in North and South America from rocks and brines associated with volcanic activity and aridity. whatever the disease process involved.CHAPTER 2 Biomedical Uses of Lithium NICHOLAS J. and for this reason we must make parallel advances in the chemical and biological understanding of its cellular mode of action. Lithium was first used medically for the treatment of gout. glass. there is strong possibility that studies at the molecular level will shed some light on the disease itself. Its chemistry is perceived as simple and relatively uninteresting except for its organo-metallic compounds. usually at low concentrations. however. Most of the world’s lithium is used in the production of lightweight metal alloys. lubrication greases. Garrod (1859) first described its medical use in detail and particularly mentioned the use in ‘brain .

lithium’s use has escalated until it is now estimated that about 500. The usual delays in laboratory reporting of results are avoided.4-0. who can thereby be challenged about non-compliance or counselled. Serum lithium concentrations should lie in the range 0. The most common side effect seen on initiation of therapy is a fine hand tremor. Nevertheless. where it does seem to have a role to play. a depressive d i ~ o r d e rLithium urate is the most soluble salt of uric acid . and the observation of the effect of lithium must therefore have been startling and exciting.^ Not only are they accurate within the clinically useful range.000 patients receive it. so that the monitoring and dose adjustment or counselling is simultaneous. worldwide. an Australian p~ychiatrist. Side effects can be minimised by operating with a lower therapeutic range. and the dose regulated accordingly. it is used by 60. and hypoparathyroidism.’.8 mmoll. Higher levels may be associated with side-effects.000 patients in the United Kingdom alone. enhancing the patient’s perception of the treatment. Treatment is monitored using regular estimations of blood lithium taken twelve hours after the previous dose.~ the time there were no effectivetreatments for any of At the major psychiatric diseases. Renal and thy9 . Lithium’s clinical value in psychiatry was discovered in 1949 by Cade.’ It is of limited use for other psychiatric states.~ and hence was expected to increase uric acid excretion to relieve gout symptoms. drowsiness.’ Determination has traditionally been by Flame Emission Spectrometry or Atomic Absorption Spectroscopy. Despite many scares. The ability of lithium to reduce or abolish recurrent mood swings has undoubtedly improved immensely the quality of life of many patients and their families and saved the lives of many who would otherwise have been led to Lithium is administered orally. but they also provide the long-sought opportunity for the psychiatrist to measure blood lithium in the presence of the patient.12 Biomedical Uses o Lithium f gout’. usually as lithium carbonate in tablet form at a total dose of up to 30 millimoles (2g) per day. non-toxic side effects such as dry mouth and nausea also may be experienced by patients when serum lithium concentrations are within the therapeutic range. with the possible exception of pathological aggression. weight gain. They generally occur within four hours of the dose. the recurrent affectivedisorders.’. Transient. lithium is a very safe drug in experienced hands. dizziness. a number of other side-effects have been noted. the time lag may be in the order of months rather than weeks. with an annual E25M saving to the health service. which disappears after a few weeks of treatment.’ After prolonged therapy. The lack of potential for major commercial exploitation has restricted the development of lithium as compared with organic psychotherapeutic agents.” Minor side effects may be experienced. when serum lithium concentrations are at their highest. including mild leucocytosis. and Mood stability may not occur until after a significant period on lithium medication. hypothyroidism. but lithium-ion-selective electrodes are now commercially available and are the most effective way of ensuring compliance in lithium-treated patient^. including tremor. Lithium carbonate is used specifically for the prophylaxis or prevention of recurrent mood changes in patients sufferingfrom manic depressive psychoses. Since the mid-l960s.

~ ~ following bone marrow transplantation. notably its stimulation of leucocytosis. Lithium.Nicholas J .^^ A model of human AIDS is found in mice infected with the Murine Acquired Immune Deficiency Syndrome virus (MAIDS). Initially this effect was exploited to treat drug-induced hematopoietic suppression. ~ ~ The metal also affects the immunologic responses to a number of challenge^. interest in lithium has entered a new phase. Birch 13 roid function should be assessed at least once every year. significantly reduces the myelosuppression and marrow toxicity of AZT. and neutropaenia. but its effectiveness is limited by the myelosuppression and bone marrow toxicity that it induce^. it has become clear that lithium can influence a series of cytokines that regulate such cell differentiation not only in blood-forming cells. which occurs due to enhanced myelopoiesis and to alterations in the marginated pool of polymorphonuclear leucocytes. when joined with AZT and combined in vitro with normal bone marrow cells.^' In the process of these investigations. when administered to normal mice following an initial exposure to AZT. l 2 In a study of lithium intoxications in a group of patients whose total exposure time was about 4900 patient years. though it is very occasionally seen as a consequence of an unsuccessful suicide attempt. These findings have now been discounted.37. however. ~ ~ anti-viral drug Zidovudine (AZT) The ? ~ ~ has been used extensively in the treatment of acquired immune deficiency syndrome (AIDS).^^. 15 cases of deliberate self-poisoning were seen with no fatalities. or when administered in vivo to mice receiving dose-escalation AZT. following an extensive range of studies carried out in major laboratories throughout the world. ~ ' particular the in formation of granulocyte^.^^-^' Lithium influences many aspects of blood cell p r o d ~ c t i o n ." Serious intoxication is rare except when therapy is not well controlled. Lithium also appears to reverse the lymphoma associated with MAIDS infection. These studies support the view that lithium might play a part in the treatment of HIV-infected patients receiving anti-viral therapy.22 or in radiation-induced i n j ~ r y . ~ ~ .^^*^^ Lithium is effective whenever granulocyte production is either faulty or i n a d e q ~ a t e .38Animals receiving AZT alone showed anaemia.21-36Further studies have demonstrated lithium's capacity to modulate AZT toxicity by influencing blood cell production. in chemotherapy of c a n c e r ~ ~ ~ .39-42 . which were dose-related and which were prevented by combination lithium/AZT treatment.2'*22for example. but also in other cell type^. Colony-stimulating factors from bone marrow macrophages are increased in the presence of lithium. which is analogous to that seen in human AIDS.'2-20 3 New Uses of Lithium During the past decade. Many of these effects derive from the well-established modification by lithium of hematopoietic processes. thrombocytopaenia. with the discovery of effects unrelated to its psychiatric use.13*14 Disturbing reports in the late 1970s linked long-term lithium therapy with renal damage and polyuria.'5-'9 Cohort studies have confirmed that in practice there is a low incidence of renal disease amongst the lithium groups.

excessively proliferated in this condition.') induced release of lactoferrin from human granulocytes in uitro. may modulate pre-existing autoimmune processes rather than induce them.2 7. There also was decreased duration of pain in patients with recurrent genital herpes simplex infe~tion. Lithium in high concentration (10mmol 1. thyroid. The lithium salt of y-linolenic acid (LiGLA) is being assessed for its effects in reducing the progression of a range of cancers.57 4 Chemistry of Lithium Alkali metals (Group 1A) easily lose an electron to yield a univalent cation. is not taken up into cells. but not of RNA viruses such as influenza encephalomyocarditi~. Lithium has also been used in the measurement of cardiac output. Lithium induces release of enzymes from human granulocyte granules in uitro. having an effect both on the lipid metabolism of the normal skin fungus. pox. but the secretory response to receptor-dependent chemotactic peptide f-MLP was not increased. the very small diameter of lithium in relation to the aqueous solvent results in a large hydration sphere of . and adenovirus (DNA) viruses. In solution. Alkali metal compounds are almost entirely ionic.'~ Further antiviral and topical uses are expected. in any case.46 Tissue-specific auto-antibodies may or may not be related to autoimmune disease: lithium. following the demonstration in uitro of cytotoxicity of y-linolenic acid and its salts towards a range of malignant ~ e l l s .54 Lithium succinate ointment has proved to be useful and is now licensed for use in the treatment of seborrhoeic dermatitis. ~ ' it stimulates the release of Tumour Necrosis Factor (TNF)from stimulated micro phage^. Parietal cell. Lithium is used as a topical application to treat skin diseases.inhibit the replication of herpes. in the time scale of the measurement (30 seconds). Pityrosporurn ovule. as the area is rapidly developing. a finding that ~-~* may have further significance for the treatment of AIDS.55 High concentrations of lithium. and on the general inflammation that is the normal response to such fungal attack." Lithium itself appears to have some anti-tumour e f f e ~ t : ~ ~ . where it can act as a suitable non-toxic marker that is easily detected by an ion-selective electrode and that.3 2. but lithium effects on other aspects of granulocyte function are less clear. there are contradictory reports of its action. and anti-nuclear antibodies and pituitary auto-antibodies have been reported to increase in lithium-treated patients.~~ A double-blind.43-45 The proliferative responses of hamster lymphoid cells to concanavalin A or phytohaemagglutinin were enhanced by lithium in a serum free culture system. ~ LiGLA also is cytotoxic towards AIDS-infected cells. placebo-controlled trial of ointment containing 8% lithium succinate showed that more rapid healing of herpetic ulcers occurred and viral excretion was reduced. Lithium also potentiates enzyme secretion by receptor-independent cytochalasin. although lithium has rather more tendency to form covalent compounds. about 40 mmol 1.^^^^^ It may be that the cytotoxicity of the GLA salt is a combined effect of the essential fatty acid and the lithium ion.14 Biomedical Uses of Lithium Several processes in the immune response are affected by lithium in vivo and in vitro.

75Similarly.9 K My Ca 1.32 0. In other words. Bone also appears to accumulate lithium to a limited degree (Figure 1). uncertain size (Table 1). or intraperitoneal or intravenous injection to experimental anim a l ~ .36 0. the pituitary. the radius increases out of proportion to the radii of the other Group 1A elements. ~ ~ . however. thymus.05 1.21 2. The chemistry of lithium classically is described in relation to that of magnesium by the so-called 'diagonal relationship'. lithium is the least reactive of the alkali metals.99 3. Birch 15 Table 1 Properties o Group I and Group I1 metals f Li Atomic radius (A) Ionic radius (A) Hydrated radius (A) Polarizing power z/r2 Electronegativity Key: A = Angstrom Nu 1. While it is clear that no tissue appears to have a very high lithium content.74 0. high concentrations may occur locally.80 1. The hydrated radius and polarising power of lithium lie between those of magnesium and calcium.33 0.2 unit 10-"m. z = charge on ion (integer). and most particularly the thyroid glands have rather more than average accumulations of lithium. the electronegativity is the same as for calcium. Most cells of the body.65 4. r radius of ion (A). local concentrations in the fluids bathing cells of the gastrointestinal tract during absorption of a single lithium tablet may be high.57 0. provided visual localisation of lithium in the whole body76 and in the different areas of the brain. are exposed to external lithium concentrations of less than 2mmoll-' even at the highest lithium doses. it is possible that the concentration of lithium in the renal papilla during relative water deprivation may reach a concentration of some 6&65 mmol 1.12 0.o = 2.03 1.In an aqueous environment. Later work demonstrated accumulation in bone66*68*73*74 endocrine glands.8 = 1. in a series of and elegant neutron activation experiments using the isotope 6Li.'.~ ~ 5 Distribution of Lithium in the Body and in Cells Early studies showed that lithium is widely distributed in tissues following oral administration.95 2. .77 It is clear that. resulting in poor ionic mobility and low lipid solubility under physiological conditions.o 1.60 3. Moreover.70 1.67 4. and lithium may interact with magnesium and calcium-dependent processes in p h y s i o l ~ g y .75 Thellier. The atomic and ionic radii of lithium and magnesium are similar. ~ ~ Lithium is the lightest solid element: it has the smallest ionic radius of the alkali metals and the largest field density at its surface. At a plasma concentration of 1 mmoll.Nicholas J .76 1.40 2.as may be seen in urine samples from water-deprived subjects. in solution the small diameter of lithium results in non-conformity to ideal solution b e h a ~ i o u r . of the soft tissues.56 0.33 2. These-reports established the basic understanding of the distribution ~ ~ ~ ~ and rate of uptake of lithium in a number of tissues.

....... .. ..... Eichner and Opitz7' demonstrated a sex difference in 'normal' lithium content of both adrenals and thymus glands....... Analytical problems generally stem ........ and partly because of the difficulties of lithium analysis. adding to the evidence for a physiological role of lithium in endocrine function.............. ....78 analysis was by Flame Atomic Absorption Spectrometry using a high temperature......... partly because of its widespread distribution in the body......... and to a lesser extent in the anterior pituitary and adrenal glands.. Figure 1 The distribution of lithium in a range of tissues following chronic administration in rats By contrast.....005-0. animals have been maintained from weaning onwards specifically on either low lithium diets (0.. in other studies... on the one hand.................... The most significant finding of this was the contrast between the pituitary and adrenal glands.... followed the experimental groups through three generations of progeny....... Patt et al. lithium accumulated significantly in bone and teeth...01 5 ppm) or control diets having 'normal' lithium content.... 6 Studies Using Lithium Isotopes Tissue Localisation and Transport of Lithium One of the problems in the study of lithium action is the lack of precision in localisation of the ion and in the measurement of its movements between cells and between tissues.......... This lack of precision arises partly because lithium is a very mobile ion..... The fertility of secondand third-generation female rats in the lithium-deprived groups was reduced though there was no apparent effecton the growth rate of the young... ............................ The lithium concentration in the two endocrine organs was maintained through three generations despite dietary restriction of lithium...........16 [Li] mmol/kg wet wt Biomedical Uses of Lithium /I ................ while in all other organs lithium content continued to decline with prolonged deprivation. The distributions revealed were generally in line with the findings of lithium administration experiments............................ .. nitrous oxide-acetylene flame.... and all other tissues............................. However.....

82Nuclear Magnetic Resonance S p e c t r o ~ c o p yor Neutron Acti. By using a dual-channel atomic absorption spectrometer. some not.Lithium occurs naturally as a mixture of the two stable isotopes 7Li (95. it is probable that these are for the protection of individual cells rather than a mechanism for bulk uptake of metals into the body. Birch 17 not from lack of sensitivity in the analysis. the spectrum of natural lithium is a triplet. We have previously reported the use of ISAAS in the determination of lithium pharmacokinetics. in order to protect the milieu interieur of the intestinal cells themselves. By measuring the light absorbed from hollow cathode lamps of each lithium isotope. ~ ~ makes ~ * ~ ~ This * ~ good biological sense.8.Nicholas J. that is.95It is likely that such processes occur during the import of many metals into the body from the intestinal lumen. The tissue distribution in the rat. but from the interference of related metals and common anions present in large quantities in animal tissues.0. but instead around the intestinal cell in their passage into the body from the g ~ t .42%). because the spectral resolution of the spectrometer is inadequate.2.85 brain lithium distribution in the mouse86-89and the rat. We have carried out extensive studies on the intestinal transport of lithium and magnesium in human and animal which suggest that these metals are absorbed in the intestinal tract via a paracellular route.58%) and 6Li (7. the two isotopes may be determined simultaneously.82-84 The microlocalisation technique with the stable isotope 6Li uses a beam of neutrons in an atomic reactor. because the lack of useful radioisotopes" has limited the metabolic information available. which are regulated by the epithelial tight junctions. 10-21s). a series of calibration curves is constructed. Thellier and Wissocqg0have described in detail a number of useful analytical techniques for the biological study of lithium. and the proportion of each isotope in the sample is determined by solution of the appropriate exponential equation. Lithium has five isotopes. the shift in the spectrum from 6Lito 7Li is 0. It should be recognised that where transport proteins have been identified for particular metals.76kinetics in the mouse brain.92 and distribution in mutant strains of mice with dysmye1inationg3 have been studied. ~ that magnesium deficiency rather than as a way of regulating gross cell content of the element.82 Briefly. via the pericellular spaces. Thus. since otherwise intestinal cells would be required to regulate the intracellular concentration of a wide range of toxic and non-toxic metals. Lithium metabolism and transport cannot be studied directly. which create tracks in a suitable detector placed in contact with 6Li-containing tissue. We have shown that active processes occur to a small extent in the absorption of magn e ~ i u mbut ~ these are likely to be involved in the defence of the cell against .~~ vation a n a l y ~ i s .90*91 distribution in the mouse embryo. three of which have extremely short half lives (0.99 . which may be determined using Atomic Absorption Spectroscopy.015nm which is identical to the separation of the two lines of the spectrum. These metals pass not through. ~ ~ ~ ~ 'normal circumstances it is impossible to identify Under isotopes by using AAS. The 6Li nucleus absorbs a neutron and immediately undergoes fission to produce an a-particle and a 3H atom. some essential.

83We have carried out extensive studies in human and animal erythrocytes (red blood cells) and less extensive experiments in cells obtained from animal sources. is not sufficiently proven to be taken for granted.108.60mV). hepatocytes. A number of ion channels are known to accept lithium relatively easily.100. this would not imply substantial cellular lithium accumulation if there is also an effective lithium extrusion mechanism. It is clear from studies of erythrocytes. However. and that these signals disappeared following addition of the ionophore gramicidin D.dysprosium tripolyphosphate (shift reagent). The explanation may be either low membrane permeability or. The situation may be different in more highly polarized cells and those that are electrically active in the nervous system. fibroblasts. 7 Biochemistry of Lithium Magnesium and Calcium Magnesium is an activator for more than 300 enzymes and has a critical role in the transfer. using a Bruker WP8OSY multinuclear spectrometer. The case for significant cellular uptake of lithium.' ~ ~ including studies in which an alternative method. 7Li NMR signals were detectable from the interior of phosphatidylcholine (PC) liposomes.83i100~101 7Li NMR and ISAAS studies were Both carried out on erythrocytes loaded by incubation in a range of lithium buffers. we were able to confirm in general terms that in the presence of a dysprosium shift reagent. Modified Inversion Recovery. alternatively. and under ideal conditions these cells may show significant lithium currents. These experiments show that in previously untreated subjects the erythrocyte internal lithium concentration is under 8 % of the external after incubation in a range of external lithium concentrations between 2 and 40 mmoll. Before carrying out studies on three different types of living cells. and utilisation of energy. Other workers have reported broadly comparable r e s ~ l t s .l'~ For all of these cell types.18 Biomedical Uses of Lithium Cellular Localisation of Lithium Using 7Li NMR. it is possible to differentiate between atoms or ions that are within the cell and those that are free in the extracellular bathing We have recorded 31 MHz 7Li NMR spectra. once again. we have come to the conclusion that lithium is less readily transported across the cell membrane than has hitherto been believed.40 and . a mechanism of effective ejection of the ion from the cell interior. and may in fact present a variable that is frequently ignored in developing models for lithium's pharmacological action. was used to determine the intracellular component of the lithium ~igna1. and astrocytoma cells83. Its predominant role is as an . The ~ ~ * from NMR and ISAAS methods are in close ~ results ~ ~ ~ agreement. storage.for up to three h o ~ r s . Cells were incubated in phosphate-buffered saline containing 20% deuterium oxide (as heteronuclear lock) and up to 5 mmol 1. between . even in electrically active cells. ' ~ ~ .109 that lithium does not distribute at equilibrium according to the cellular membrane potential (in these examples.

5)IP3 is ultimately converted to myoinositol.4. fat.which converts phosphatidyl-inosito14. and the regulation of the system as a whole soon becomes unstable and chaotic. signalling.60*112 indeed a number of studies and of the effects of lithium on magnesium-dependent systems have been reported. Lithium also inhibits other enzymes in the interconversion and breakdown of polyphosphoinositides. and regulatory. and protein metabolism. That in turn is converted to phosphatidylinositol. lithium reduces the cell concentrations of myo-inositol.Nicholas J . this attenuates the response to external stimuli.' 3-1 The physiology of calcium and magnesium is directed towards a number of different functions: structural. particularly reactions involving ATP.'25These metabolic products are second messengers: 1.5-triphosphate [(1. Subsequently.5-bisphosphate (PIP2) to 1. * Neuronal and hormonal signals may often be transduced via receptor-mediated activation of phophoinositidase C (inositol lipid-directed phospholipase C). by a combination of these. ~ ~ ~ ' ' '' '* Lithium and the Phosphoinositide Signalling System Lithium selectively interferes with the inositol lipid cycle. which would otherwise be converted into phosphatidylinositol.124. which has cata' ~ * is ~ strophic effects on metab~lism. or.4. ~This ~ ~an unusual mode of inhibition. though not by an uncompetitive mechanism. Magnesium has a pivotal role in carbohydrate.2 DG stimulates protein kinase C and (1. including the hydrolysis and transfer of organic phosphate groups.4.'28 This scheme has been suggested as ' ' . which in turn affect the way in which the body utilises these metals.4. The primary substrate at an earlier stage along the metabolic path is then rapidly depleted.'~' a metabolic pathway at steady state. which would otherwise be converted to phosphatidylinositol. It has been suggested that lithium may compete for magnesium and calcium binding sites on biological ligands.5)IP3 releases intracellular calcium from the endoplasmic reticulum. (1.'10*''' If lithium were to compete for sites on these enzymes or on some of the many other magnesium-dependent enzymes.119*120 which is the basis for the proposal of a unifying hypothesis for lithium actions.5)IP3) in the cell membrane.'lg Lithium has been shown to inhibit inositol monophosphate phosphatase u n c ~ m p e t i t i v e l y . Either because of the uncompetitive inhibition of the monophosphatase or because of the inhibition of the other enzymes.121y'22Lithium reduces the cell concentrations of myoinositol. indeed. widespread metabolic effects might be expected. The reduction in cell inositol content attenuates the brain response to external stimuli. although the exact mechanisms of magnesium's regulation have not been fully e s t a b l i ~ h e d .2-diacylglycerol (1. Their cellular activities are controlled by their simple chemistry and osmotic relations. The regulation of these two elements takes place to some extent reciprocally. the In increase in the extent of inhibition is non-linear as the products of earlier enzymes in the cycle begin to accumulate. Birch 19 activator of phosphate transfer reactions. which is used to replenish PIP2 stores and thus complete the cycle.2 DG) and D-inositol 1.

20 Biomedical Uses of Lithium the mechanism of action of lithium in the affective disorders. By contrast. for which lithium is effective in prophylaxis. each with the potential to be regulated by its own individual molecular neighbourhood of macromolecules and ions. but it does not show these effects in psychiatric patients at therapeutic lithium concentrations. If we consider that metal ions at the cell surface may be either free or restricted in their movement. and the brain cell must therefore rely on endogenous supplies. Perhaps we should now reconsider that and rethink our ideas of regulation.' 30 Charged ions may be densely distributed amongst the protruding. 3 0 ' . Metals within cells may themselves be a means of cell regulation.'29 It is proposed that the mood disorder. Hitherto. so it is likely that a radical new approach will be required to interpret the mechanism of lithium's action if our preliminary findings of a low intracellular lithium are ~onfirrned. which in turn is regulated by metal ions at the cell periphery. For example. or even embedded. The whole surface of the cell could be conceived as a patchwork of such microelectronic 'components' distributed across the surface and at variable depth. may be generated by an as-yet-unidentified group of brain cells that are pathologically overactive. It is possible to envisage a model of cellular regulation iiu that depends on the functioning of the cell membrane as a 'bio-microchip'. research has concentrated on the notion that lithium interferes with intracellular processes.'~' Modern molecular biology stresses that the cell is regulated from the nucleus. A pericellular site of action of lithium would provide greater scope for explaining the diversity of the lithium effects that have been reported in the literature over many years. subject to flexible configuration induced by changes in cell morphology and by the docking and undocking of membrane-active biomolecules such as neurotransmitters and second messengers. portions of glycoproteins and glycolipids that form the glycocalyx. silicon-based. rigid. the main theory about lithium action currently in vogue is that it acts via phosphoinositide metabolism. The implications of such an hypothesis are very broad. since dietary sources of inositol cannot cross the blood-brain barrier. this provides us with a new mechanism for signalling and a means for the integration of signals from a variety of sources. The evidence shows that lithium does indeed have effects on phosphoinositide metabolism. and that such overactivity would be preferentially attenuated by this restriction of Lithium at the Cell Periphery: A Novel Viewpoint The major theories of lithium action depend upon there being a relatively high concentration of lithium in the cell interior. which would become rate-limiting were lithium to act in this manner. the intracellular metals may be functionally separate and regulated as part of the m l e interieur of the cell. making a new assumption that the integrator and regulator of the cell is the pericellular apparatus. electronic microchips. Phosphoinositide involvement in the biochemical pathology of manic depressive disorders is far from proven. a flexible and infinitely adaptive analogue of our man-made.

Walton and Maberly.Hetmar. 1991. P. Ahrens. 1993.S. Basel. Schou. Psychiat. Schou. Rev.N. 7 M. N. 18 M. Coley. D. Acta Psychiat. Pharmacopsychiatry. Pharmacopsychiatry. Vestergaard and M. Ladefoged and T. Med. Schou. Thomas. p. p. 1989. 1989. Hullin. J. Birch. 1992. Gallicchio. Dukes. 1. Handbook o Toxicity o Inorganic Compounds. K. Br.437. Psychiat. Scand. eds. Birch. Vestergaard.80. 1989. Hughes and K. 1989. 1979. London.3. p.218. M. Phillips and R. Schou. 23. Grof. Nilsson and R. Lithium. Edwards. f f Seiler. Karger. ed.G.259. 1991. 31 13011 References 1 M.P.. Update. as a probe. J.G.22. The Nature and Treatment of Gout and Rheumatic Gout. J. Scand. Schou. R..B. N. Thau.D. Tse. 16 0. to investigate the molecular interactions between more complex drugs and their receptors. 9 N. Psychol. V. 1989. Psychiatry. Whatever lithium does.E. 383. Muller-Oerlinghausen. J . London. Thomsen and P. 21 V. 1992. Bolwig. Bolwig. 20 R. 11 D. whose chemistry is relatively simple.J. Side Eet ’c s o Drugs Annual: 10.. Simhandl. it achieves because it is a highly charged cation with a large hydrated radius and chemical properties similar to those of magnesium. G.K. which might lead to modification of receptor activation. Waller and J. Br..19. Birch.. f 1986. K. 1992.53. Encephale. 15 D. 10 M. we may gain insights into the most basic features of the cellular response to drugs: lithium does not. N. Schou. . 6 B.99. Schou. Freeman. 4 J. 8 M. Povlsen.86. 27. J.. 1859. Elsevier.P.15. Sigel and H. M. 14 M. 13 P. 1988.146. Res. Garrod. 2 N. 573. M.B. Lenz.G. U. 1. J .221.45363.F.J.G. Axelsson. 133.825. Psychiat.Nicholas J . B..J. Med. 1. C.F.. after all. 17. ed. Internal Med.S. 12 A. et al. Schou. 1949. It must therefore be assumed that whatever lithium does. 17 0. Srinivasan and N. 1457. Pharmacol.285. 1957.J.J. 233. Larsen and T. Schou. Ladefoged.287. J . Am. Res. 22. Davie. Brun. J . Acta Psychiat. 19 M. H. J. Amsterdam. E. C. Cade. Morgan and T. J . Schou. Wolf. For this reason it may be important. Birch. Academic Press. 1989.H. S. 1989. Lithium and the Cell: Pharmacology and Biochemistry.. 5 M.. Grof.G. Birch. Lithium Treatment of Manic Depressive Illness: A Practical Guide. p.9. have a large and convoluted structure that can make multiple contacts with receptors. H. its action is on fundamental processes. 1988.J. New York. If we can discover what it is that lithium does at a molecular level that makes it so effective in psychiatry. Aust.Hetmar. Birch 21 9 Conclusion The use of lithium in medicine is a significant success in the field of inorganic pharmacology and is of particular interest because lithium is the lightest solid element. Med. Hansen. 101.G.349. Volk. 1989.158. 3 A.J. Marcel Dekker. J . p.J.P.22..

K. Gallicchio and N. 93. Plenum Medical Press. 1992.76. 183.S. Biol. Theor. J . J . Abernathy. Ells.A.W.I.25. Fiers. London. Leukemia Res. 1. Gallicchio. N. 181. Tse and N. Lithium. ed. Marius Press.K.S.573.K. 48 F. Prog.R. Todo and S.3. Morrow. Hughes. 25 R. B. Desportes. P. M. Scott. 34 V.403. Oxford. Noblitt. Magnes. J .219.6. New f York.S. Fujiwara. ed. S . N.150 pp. J . Growth Factors. Immunol. 223. Gallicchio.151. 1986. Hulette and N. Lithium: Inorganic Pharmacology and Psychiatric Use. 49 A. 1990. p. Lithium and the Blood. 1993. Chin. 1993. 1988. 35 D. Inst. Lithium and the Cell.K. S. N. Lithium and the Cell Pharmacology and Biochemistry. Tse. Exp. Gallicchio. J.. Townsley. p.K. 1994. Ells. Cibull. Hughes. Cytokine. K. 1986. D.5.K.Y.A. N.50. Hart. K. eds.S. SpringerVerlag.23.G. Begin.J. Med.9. 1993. Tse and H. Tse. 27 D. Gallicchio. Pharm. Wade.J. 32 V. Israel and K. Kazmi. Nut. Sokoloski.L.J. 289. p.22 Biomedical Uses o Lithium f 22 V.E. 38 V. J . Horrobin. Heyninck. G. 52 R.C. Begin.S. 45 D. Basel. Birch.K. 1994. Pathobiology.1. 28 H. Prostaglandins. C. Kinchington. 44 V. Horrobin. 1991. Lazarus. Bach and V.. 1995.216. Gallicchio. Hughes and K. 1980. 1991.S. Lithium. K. Chen.. K. Padgham and M.1994. J.E.J. Cai. M.3. Hughes and K. M. Van Roy and W. Hughes. Hughes.L.61. 58. 41 V. N. Lipid Res. 46 J. Gallicchio.F. Scott. Cibull. Li. 50 M.Z. J. Messino. 1993. Lithium Therapy Monographs: 4.77. 1986.201. 40 V. 33 V. Gallicchio. F. Abstracts of the Tenth International Conference on AIDS.A. K. Gallicchio. Yokahama. Johnke and R. Hughes.F. Ling.7.. K. 42 E. 29 R.S. Basel. V. 1992. Birch and J. 1. 1991. ed. F. Blood.K. Tse.M. K. E. Gallicchio. S.9. N. 36 V. Beyaert. 117.F. N. 39 V. N.54. Karger. 1991. F. 47 M. Das and G. Lithium and the Blood.J. Lithium Therapy Monographs:4. Fiers.151. 1991.S. Hughes. New York. J . 195. Lithium and Cell Physiology. Fiers. Nigam and A. Hulette and L.S.291. . N. 1062.S.K. N.S. FEBS Lett. McGrath.284. 51 D. London.J.L.C.104. Navel and P.M.S. Boeykens.L. Quesenberry. 177. Cell Cloning.268. 1992.O. p. Gallicchio. 1991. Med. 53 R. 1. Kazmi. 24 Y. Birch. August 1994. Townsley.17. Peking. Carnforth. Van Roy and W. N. Academic Press. 1150. J . Lithium in Medicine and Biology. 1992. 1991.K.62 (Abstract). N. Wang and T. Tse.S. 1992. Das and D. Beyaert. 1993.J.K. 1992.23. Gallicchio. Beyaert and W..H. De Valck.W.. N.S. Leukocyte Biol.330. Plaisance. Clin. ed. 37 V. J .C. Wu and D. Clin.E. Lithium. Hughes. SOC. Trace Microprobe Tech. Exp. Ling. B. Hughes. Randall.284.311. Gallicchio. Winther and D. Gallicchio and M. Schulze Osthoff. Cibull.10.55..S. U.78. Endocrine and Metabolic Efects o Lithium.N. R... Leukotrienes und Medicine. Ke. Hughes. UK. I.F. Y. Sartorelli. Press. 1991. S. p. U. Int. 1991. Birch and V. Birch. K. J. Tse. Med. Cellular Physiol. 580. Kister and P.S. Gallicchio.13. Karger. 43 V.S.690. p. Cibull. 30 V..S. Proc. Scott and N. 1993. M.J. Res. Anel. Gallicchio.K. M. K. p. Birch. A.J. Gallicchio. 23 J.L.F.S. Academic Press. 26 X. Leukaemia.S. 31 V. M.N. Jia. Can. V. Gaines.3. K. Birch. Hart. 1986. Imashuku.. eds.

M.38.F.M. Marin... 1970. Physiol. III. 80 M. Patt. A.J. Academic Press. Thellier.J.J.25.J.324.166. 1991. C. 1992. A.46. 1991. Birch. C. N. Eichner and K. Horrobin and J. Pickett and B. Physiol. Neuropsychobiology. 69 M.30. Bioinorg. Exp. Mebarki and M.47. Swan and G.D.Y. Psychopharmacologia. E. 1992. Am.A. K.1255. 60 N. T. Biol. Birch.W. Stewart. 77 M. A.J.. Cell. Thellier. 83 M.251. 1991. 1959. p. C.Nicholas J . Hendrickson and P.C. J. Sci. 429. 55 D. Med. Haire. 56 G.. Barchas. Nature. Neuropatol. H. 1949. Birch and F. Hughes and N..S. Birch. Pharm. A.M.H. J . R.R.M. Akad.F. Birch and R. 70 J.299. Immunol.. N. S.B.216. Dermatol. Academic Press. Zh. Horrobin. J. ed. Baumann. H. Birch. 63 M. Wissocq and C. Kenicer. 58 K.15. Hullin.D. 1950. N.M. p. Signoret. 72 P. Rev. Baer.I. P. Pharmacol..G. 1978. ed.S. Herman. Harper and P. 61 V.S. S. V. Cream. 1952. 78 E. Jenner. J .8.14.59. Robinson. Amis. Fieve. Histochemistry.. 68 N. Thellier and J. Acad. Am. Gallimore.K. C. Oxholm.N.71.S. Gilman. London.J.42. 49. 1972. MacLeod.D. P. P. 64 L. Stelz and J. Sher and J. San Diego. J. R. Leipzig and Jena. M. 1095.. Wilkinson. M. Microbiol. Nature. 175.168.J. Physiol. Hennequin.. T. Ser. E. S. Ebadi.. p.J. London. A. Lindskov. Kersey. 1972. W. Birch. Wissocq.. Wissocq. Opitz. Heurteaux. Morse.J. 86 C.L. Hullin.J. J .115. J . 1976.F. A. Psikhiat. R. B. Braunlich and C. Clarke. Pharmacol. Pathol. Stelz and M. Inorg. Metal Ions in Biological Systems. Chem.409. Birch. Sigel. Samoilov.ed. Wissocq. Birch. 1974. Karl-Marx Universitat and Friedrich-Schiller Universitat. Stern and E.F.262. Lithium and the Cell: Pharmacology and Biochemistry.35. Proby.J. Adv.A. Inie and R. Buchan. J . Am.. 1993. J.201.. Life Sciences. 57 R. Marks.P. A. Physiol.299. Academic Press.. 1980.1. ed.163. 27. Phillips.168. 1973. Microsc. p.9. Ce!lulaire. G.S. Nelson. Nelson and P.633. .A. Radomski. J . Birch 23 54 D.J. Lever.461. 116.264. 1974. Pharmacol. Heurteaux. Aitken. Vol. Hamill. 1982. 74 N. J.71. J.. Pharm.S. p.. 73 N.C.642.. Am. 127. Hullin. 82 N. Saratikov and V. Green.J. 81 N. eds. Simmons.452. 1979. 87 M.M. D. Biol.A. Hartley. Moss. Br. Heurteaux. Bruckner.G.P.177.283.314. Mudge and A. Anaesthesia. Sci. Toxicol. 1978. J .R. D. Br. Saratikov. F.W.586. R.M. Nauk.J.J. Davis.L. ed. 65 H. Vol.A.S. M.J. D. Psychiat. 273. R.17. Band and K. Mol.295..11. R. 85 S. Acta Pharmacol. 1951. 79 D. K.L.P. Wissocq. Acad. 683. Mortimer. Br.D.26.E.J. Chem. M. 1991.S. 1958. New York. 1980. Sykes.202.H. 86. Davis. Thellier. J . Marcel Dekker. Academic Press.J. Talso and R. Gould. 1950. 257. Schou. 681. p. Kassir and R. O’Dell. Davenport. Chem.J. Horsman and R. 1971. J . J . Allen. 66 N. Skinner. Lacy. 76 J. 67 A. Eur. London. 36. 1976. E. Fuyat. C. 71 J. 1970. Dokl. 153. 1982. Lithium and the Cell. 139. Hughes. 84 M. Birch.G. Am. L. 1974. Strong. 75 N.J. Birch. J .100. 1976. 157. Lithium and the Cell: Pharmacology and Biochemistry.27. Clin. Birch. Spurenelement Symposium (4). Bensch. 62 J.157.1709. Clayton. Anke. Foulks. L. Wright. R. Smith. 1983.C. H. Birch and J. 59. Thellier. 59 N. Linton.

G. Dorus. Birch. Phillips. 101 J.A Relevant Ion?.D. Baumann.1462. Biol. 1. 91 B. Biochem. 94 R. New York.J. 10. 1979. Press.D. Yuregar. 111 F. 1994. Maguire. 1976.A. N.L. Birch.J. Bensch and J.ll.S. A.J.J. London. Leaf.212. 0.J. UK. Birch. Lithium. Turkey. Silberberg. Sigel. 325. Adana. Thellier. Sigel. C. Neurochem. Abraha. Nelson.G.R. R. Heurteaux.J. Mota de Freitas. Lithium: Inorganic Pharmacology and Psychiatric Use.D. London.118.. M... Marcel Dekker.J. p. Bramham.J. Samuel. Press. 52.D. 109 F. 187.205 (Abstract).. Wissocq. Psychiat.C. Birch. 96 C. Riddell and J.27. p.3. Marius Press. R.I.J.208 (Abstract).M.G.227. Durlach. 253.2. E. G. Heaton. 135.T. Hughes.H. Lithium in Medicine and Biology. J. Wissocq and A. J.J. Phillips. 1986. Lachapelle. I. R. eds. M. Scotland.M. Wissocq and M. 99 N. Birch. 1993.G. Thomas and N. Biochem.S. ed. 97 R. N. 1991.523. Academic Press.E.J.R. M.S. Cukerova University Publishing Company. Academic Press. N. J .J. 90 S. 108 J. Pharmacol. p. J . C. Yuwiler.Ligand interactions in Biological Fluids. 1988. Magnesium . R. E. G. Birch. Metal Ions in Biological Systems. N. N. Thellier. Metal Ions in Biological Systems.J. 98 N.C. Inorg.C. Mota de Freitas. Therap.263. Maynes. Birch. 1993.J. Inie and F. p.J. D.eds. Barchas. 125. M.P. Partridge. Academic Press. p. 1991. Sigel and A. Davie. Biochem. Lithium and the Cell: Pharmacology and Biochemistry. J . Dorus and D. PhD thesis. ed. N.65 (Abstract). 85. Wassermann and A. 1990. 95 J. Lithium-7 N M R Investigations ofthe Biological Behaviour of the Lithium Ion. 104 F. 1991.16. E x p . Volume 26: Magnesium and Its Role in Biology.237. Am. Turkey.J. Abraha. Coll. Hughes.Phillips. Nutrition and Physiology. Oxford. Volume 26: Magnesium and Its Role in Biology. Thellier.. New York. G. Patel and M. 243. Bisconte and M. S. 93 C. 110 M. 1988. ed. R. Marcel Dekker.G.J. Whang. London. Hughes.. University of St Andrews. Nutrition and Physiology. 107 A. Birch. p. f Berthon. 1989). 263. Birch.415. 119. Wissocq. Padgham and M.39.J. eds. T.46. Birch. 132P (Abstract). H. Chem.H. I. J. p.J. J . Birch. 1990. Davie. Handbook o Metal . F. Monnier. 103 M. 1983. 92 J. Nature. SOC. Herman. 113 N. Riddell. London. R. Nutr.R. eds. Hughes and N. W. Marcel Dekker.J. Phillips. Davie and J. Heurteaux.D. Dorus. Padgham. J . Br. John Libbey. H. 200. 105 D. Trace 89 (Proceedings ofthe Third International Congress on Trace Metals in Health and Disease..p. Olufunwa. 1988... 271. 1974. Phillips. Trans. p. I.L. Magnesium and the Cell. p. Heurteaux and M. A. 1991. Neuropharmacology. Stelz.W. 1990. Bramham. 1993. Oxford. Pharmac. 1977. ed.Trans.. Riddell. Cytochem. 1988. Radioanal. 102 S. ed. Carpenter. 1990. Kayrin. E. Lasserre and J. 1991. 1980.M. and N. Espanol. Histochem.28. 89 M. Williams. B. Hughes. N. Keighley and N.24 Biomedical Uses o Lithium f 88 J. 1990. Donma and L. Partridge. N. 1991. K. 100 G.T. Sigel and A. Res. Carnforth. Birch.S. J .D. Thellier. Birch. J. Birch.S.16. Thomas.P. Birch.22. C.1317. . 112 J. ed. SOC. Elenz and W. Hughes. Lithium and the Cell. 106 F.S.R. New York. Frausto da Silva and R. p. Cukerova. Hullin. M. J. Davie and J.J.J. 37. 285.. Marr and N. Lithium: Inorganic Pharmacology and Psychiatric Use. eds. M.S. J . Davie.

A. J.1. p. Kajda.275. N.H. Biochem. S.K. 127 A. 1988. 131 N.14. M. 1979.M.E. M. 121 M.19. Biochem. Berridge.R. Trends Pharmacol. Wacker.B.297. Hawk and R. Cell. Magnesium and the Cell. Nahorski.. Thomas. Biochem.P.E. 1. Chem. Berridge.299. 1991. Hughes. N. B. Sherman. Geny and G.J. p.12. 201. Znorg.J.. 1988. C. SOC. Birch. Biochem. Biol. 125 S. London. Birch. 1994.Nicholas J . 1993. 1989.587. Michell. G. Brown. Academic Press. Trace Microprobe Tech. M. 128 S. 16499. Glennon.. 1. N. Birch. Kajda and N.A. ed. C. 120 R. 11.G. J A M A . 1986.J. Birch. . 126 S.266.. Downes and M.J. p.3.17. J.J. Znorg.R. 130 N. Press.J. Thomas. Sci. Lithium and the Cell: Pharmacology and Biochemistry. Komoroski.P. Hullin. Berridge.J.1.W. FEBS Lett. p.C. Lithium: Znorganic Pharmacology and Psychiatric Use. Nogimori. Challiss.M. 115 P. Znorg.. Academic Press.J. 159. C. I. 1991. Res. Hodgson. J . 35.12. N. 361. Birch.288. Hanley.1. Cornish-Bowden. Cockcroft and G.H.B.H. Shears. 116 M.J. 129 M. Oxford. ed. Ragan and R. SOC. 1991. ed.203.J.M. 1991... P.C.. 119 W. 1989..l. Biochem.P. Hughes.J. J . Magnes. 124 S. Partridge. 118 W. 1834.59. Birch.. Birch. Downes and M. 121.R. 122 M. 1981.Trans. 1989. London.L. 1991. London.J. Academic Press. Birch 25 114 P.S.49.. Cockcroft. N. Shears.206.I.K. Biochem. 1993. 123 K. 1982. 1992. J . 117 S.Tran. Birch. Birch.262. Marr and N. Lithium and the Cell. O’Brien and R..R. J . R.J. Hanley. J. J . Biochem. Putney and S.J.411.R. ed.J.

USA 1 Introduction Chrysotherapy In current medical practice.6tetra-0-acetyl-P-l-D-thioghcose ligands. thioglucose gold(& and thiopropanolsulfonate gold@)are oligomeric complexes that contain linear gold(1) ions connected by bridging thiolate ligands.3 Modern medicine dismisses such uses for lack of plausible mechanisms. 0 3 2 . It has a medicinal history that can be traced through the written history of every culture and far into pre-history by means of archeological records. contains coordinated triethylphosphine and 2. Five gold(1)complexes (Figure 1)are widely used throughout the world in these treatments.4. Milwaukee. The last two complexes are discrete monomers with well-defined structures. The University of Wisconsin-Milwaukee. Bis(thiosu1fate)gold(I)contains gold bound to the terminal sulfur donor atoms of S . Yet. WI 53201. FRANK SHAW I11 Department of Chemistry. For AuSTm and AuSTg. the treatment of rheumatoid arthritis with gold-based drugs. however.CHAPTER 3 Gold complexes with Anti.3. a golden-haired heroine of Greek Mythology.'e2 History of Medicinal Uses Gold. the noblest of metals. Anti-tumour and Anti-HIV Activity C . It derives its name from Chryses. Medicinal properties were early assigned to gold on the basis of its mystical importance and its association with the sun. auranofin.arthritic. chrysotherapy. was probably the first pure element recognized by the early inhabitants of the earth. form noncrystalline powders and may have variable chain-lengths in solution. PO Box 413. Thiomalatogold(I). is well established. one can still . the oligomers.. The newest drug. the structures given in Figure 1 are the principal components of complex and somewhat variable mixtures that have been reviewed elsewhere.

a German physician and bacteriologist. primarily thiolate compounds (Figure l). The efficacy of chrysotherapy was confirmed by large multinational trials just after World War 11. During the late 1920s. had its inception in the experimental work of Robert Koch. Although chrysotherapy by rheumatologists has since undergone cycles of popularity and .6-tetra-O-acetyl- 6 -1 -D-glucopyranoaato-S) (aumnofin) (trlethylphosphlne) gold(l) Figure 1 Structures offive widely used gold (I)-based anti-arthritic drugs obtain modern descendants of such treatments. Jacques Forestier.C.3. but without killing them. This discovery led in the 1920s and 1930s to the use of gold compounds. the modern. scientific use of gold compounds in medicine. Chrysotherapy. a Parisian physician. ‘Goldwasser’and ‘Goldschlager’ are spiced liqueurs containing particles of elemental gold foil (about a dime’s worth of gold in a litre) and reputed to have medicinal properties. to treat tuberculosis.has potent bacteriostatic properties: that is. . and he undertook trials of gold complexes as a treatment for rheumatoid arthritis. Frank Shaw I I I 27 AuSTrn = gold(l) sodium thiomalate Au(%O&J - sodium bis(thiosu1fato)ga (myochtysin) (sonochrysin) CHiOH OH L Jn AuSg - AuTpS = goldthiopropanolsulfonate (allochryslne) gold(l) R-D-thioglucose (solgonot) OAc EbPAuSATg = (2. noted that patients with arthritis who were taking gold had significant improvements. He discovered that Au(CN).4. it retards bacteria from growing.

This situation is likely to prevail - . or suffer side effects that require cessation of treatment. They undergo rapid metabolism to form protein and low molecular-weight metabolites that are probably the source of gold for the elusive target sites. is an inflammatory condition that leads to progressive erosion of the articular cartilage lining the interfaces of bones in joints. The resulting tissue destruction releases cell and tissue fragments. which they share with many organic medicines such as aspirin and ibuprofen. is not checked. and then moves into the synovial cavity between the bones. Only occasionally are there life-threatening consequences (ca. which are typically held below 300pgdl-' ( 15pM) to minimize the accumulation of gold in tissues and the resulting side effects. As discussed elsewhere in this review.' Gold drugs have well-documented anti-inflammatory activity. but they also have more profound effects on the underlying joint destruction that simple anti-inflammatory agents lack. ~ Rheumatoid Arthritis and Putative Mechanisms of Action Rheumatoid arthritis. Thus some patients on chrysotherapy obtain long-lasting remission of the disease.28 Gold Complexes with Anti-arthritic. Thus. the gold bound to albumin is less bio-available than are the drugs themselves. which surrounds the joints. injectable gold(1) t h i ~ l a t e s . One problem in examining the mechanisms of chrysotherapy has been a general failure to recognize that gold complexes are in fact pro-drugs. but none can be considered as well-established within the medical community.g. the nature of chrysotherapy treatment has changed very little except for the licensing of auranofin in the mid-1980s. Many mechanisms have been proposed for chrysotherapy. AuSTm. and some are only cosmetic (e. to very specific biochemical processes that act through a progressively multiplied effect on a larger system. Damage to the tissues is effected by lysosomal enzymes including collagenase and other proteases that are released because of the inflammatory condition. Chrysotherapy is effective for about 70% of the patients taking the treatment. etc.) have activity superior to that of auranofin (Et. a cycle of degradation and further release of destructive enzymes is set up. which shows many characteristics of an autoimmune disease. but now appears to be less effective than the traditional.PAuSATg). Auranofin has the advantage of being orally absorbed. 1 : 10.' If the attack. and less uptake into cells occurs in its presence. for which chrysotherapy is used world-wide. which stimulate further inflammation and the migration of immune cells including macrophages into the inflamed area. The initial inflammation occurs in the synovial membrane. The proposed mechanisms listed in Table 1 range from broad and sometimes vague systemic effects. Anti-tumour and Anti-HI V Activity disuse. less frequent treatment with the gold drug. such as immunomodulatory activity. skin rashes).000 patients): usually inhibition of white or red blood cell formation in the bone marrow. The injectable gold(1) thiolates (AuSTg.'V5 Others do not benefit from gold. Most side effects are mild. which is sustained with maintenance therapy. the bones will eventually fuse after complete loss of the cartilage. Patients are usually monitored weekly or monthly for blood gold levels.

which has been identified as a common metabolite of all gold drugs. Major review articles by Sadler. and proteins in the body. Only when a mechanism of action is well established will it be possible to design new (third-generation) drugs that combine the effectiveness of the (first generation) injectable gold(1) thiolates with the safety of auranofin (the only second generation drug). 11. IV.I. Gold unfortunately is rather promiscuous and associates with many tissues. only gold(1)and gold(II1) are known to form compounds that are stable in aqueous media.* and Brown and Smithg detail the subtleties of gold chemistry in relationship to biological systems. G. On the other hand.Frank Shaw III Table 1 Possible mechanisms of chrysotherapy ~ ~~~~ ~ ~ ~ ~ ~ 29 Anti-inflammatoryactivity Altered prostaglandin biosynthesis Inhibition of lysosomal enzymes activity and/or release Immunomodulatory activity Altered peptide presentation Inhibited macrophage phagocytosis Inhibited histamine release by mast cells Inhibited complement activity Inhibited monocyte migration Inhibition T-cell stimulation and proliferation Reduced immunoglobin (includingrheomatoid factor) titres Altered copper homeostasis Altered metabolism of reactive oxygen species Catalysis of singlet-to-tripletoxygen conversion Glutathione peroxidase inhibition Modulation of the oxidative burst Perturbation of Zn-fingers by Au(I) thiol competition in uiuo.-.C. Jones" and in 1986 by Melnick and Parish. Extensive surveys of the structural chemistry of gold compounds were published in 1983 by P. - 2 Gold Chemistry Oxidation States Gold can exist in seven oxidation states: .' Shaw. 0. It occurs at concentrations of 1-5 ppb ( 5-25 nM) in blood and at higher concentrations in urine. Relativistic effects arising from the interaction of the large nuclear charge ( + 79) affect the 6s orbital wavefunction and partially determine the properties of gold atoms. ions and complexes. Apart from gold(0)in the colloidal and elemental forms. .I2 Only the broad outlines of gold(1) and gold(II1) chemistry will be reviewed here. and hence in a biological milieu. but only marginally so: 2-3-fold. cells. Au(CN). I. Puddephatt" surveys the basic features of gold chemistry.6 Some drugs bind to their target sites so tightly that radiolabelling leads to quick identification of the molecular mechanism of action. The concentrations are higher in inflamed joints than in the blood. is more readily taken into cells. and v. 111.

however. Thiolates (such as cysteine and thiomalate). NC-AU PPhS"'" Au "'o PPhS I I PPhS 8.has been isolated or characterized. Like Ag(I). and are linear. thioethers (such as methionine and dimethylsulfide). 3 and 4.\ . Two-coordinate complexes may be generally be obtained with any type of ligand able to coordinate to soft metal centers. despite many efforts to form the corresponding tri-and tetra-cyano complexes. with bond angles approaching 180". Typical examples are shown in Figure 2. Gold(ll1) Complexes CN = 4 CN = 5 N C \ p C N NC' 'CN Figure 2 Examples of gold complexes exhibiting the common oxidation states. These complexes may be neutral. I and 111 . positive or negative. selenols. it forms complexes that may be two-. three-. Gold(1) Complexes CN = 2 CN = 3 CN = 4 P Eta NCS ATgS -Au -P( C H j X 0 -CN NCS -AU . Gold(1) has much less tendency than CU(I)and Ag(1) have to adapt 3. cyanide and alkyl groups (especially A. for gold.30 Gold Complexes with Anti-arthritic. phosphines.-"+' and Ag(CN). Thus Cu(CN). and Hg(rr).' \ PPhS \PPh. or four-coordinate. with various combinations of ligands.or 4-coordination.-"+' are known for n = 2. Cu(r). Anti-tumour and Anti-HI V Activity Gold@)Complexes Gold(1) is a d" metal ion. only Au(CN).

).O ' . are generally four-coordinate and square planar (Figure 2). of the ligands about gold.C. as a result. its complexes have a wide range of physical and chemical properties.and four-coordinate d" complexes preferentially adapt ~ trigonal planar and tetrahedral arrangements.~ +]. with a few notable exceptions..g.g.PAu). AuCl. and a host of nitrogen donor ligands (including amines and heterocycles) form strong bonds and stable complexes. with four anions such as Cl-.-. For example. [(Ph.AuBr.0. With four neutral ligands. the charge is + 3.+] or [Au(OH.1. but are generally reactive and much less stable (e. Frank Shaw I11 31 methyl and phenyl). the charge is . Four-coordinate gold(1) complexes are also typically rich in phosphines and usually neutral or ~ a t i o n i c . even water slowly reduces some gold complexes (e. Thus gold(II1)complexes. positive E . E. 'The three.42 (1) (2) The large. Oxidation-Reduction Potentials Neither gold(1) nor gold(II1) forms a stable aquated ion ([Au(OH.)..858 . Oxygen complexes are known. a configuration it shares with Pt(I1) and Rh(1). from appropriate combinations any intermediate charge can be obtained.68 + 1. such as NH. even though these ligands have very high affinities for gold(1). The ability to modulate charge is important for the design of complexes with the appropriate balance of hydrophilicity and lipophilicity for medicinal use or testing. Three-coordinate complexes are trigonal planar and generally contain at least one and often two neutral ligands such as phosphines or arsines. respectively. respectively)analogous to those found for many transition metal and main group cations.A variety of ligands form stable complexes with this oxidation state.14 No trithiolate complexes of gold have been isolated to date. = (3) (4) + 3e- = Auo + 4Br- + 0. used by Schmidbaur and co-workers to prepare hypervalent 1 gold cluster complexes1'). values indicate that mild reducing agents are able to effect the reduction.14 E. with concomitant release of elemental oxygen.- + le- = Auo + 2CyS- E. -). Both are thermodynamically unstable with respect to elemental gold and can be readily reduced: Au'+ + leAu3+ + 3e- = Auo = Auo E. = = + 1. Gold(II1) Complexes Gold(II1) is a d8 metal ion. The situation can easily be changed by appropriate choice of ligands to stabilize the oxidized forms of gold. for example: Au(SCy).+ = .

+ RS(0)R + 2L + 2H' Au"'L. the negative potential demonstrates significant stabilization 1 oxidation state by coordination to thiolate ligands.. 3 Gold Biochemistry and Pharmacology The gold-based anti-arthritic agents are really pro-drugs that undergo rapid metabolism to form new metabolites.still is a powerful oxidant. The reaction can be driven by naturally occurring reductants such as thioethers.. + RSR + H 2 0 + Au'L. thiols or even disulfides:'5-'7 + Au"'L. a number of exciting observations provided new insights into the chemical transformations '~ that transpire in ~ i v o . Following the early triad of in-depth reviews on the bio-inorganic chemistry of general reviews by Parish and Cotrill. XANES and WAXS structural studies by Elder and E i d ~ n e s and~studies on s~ protein chemistry by Shaw2.32 Gold Complexes with Anti-arthritic. For that reason.22 EXAFS. and on aurothioneins (gold-laden metallothioneins) by Savas and S h a ~ . ~have appeared. Anti-tumour and Anti-HI V Activity The E ."='2"-1'/?rnAu1 RS0.. The use of radiolabelled ligands reveals that gold and its carrier ligands have different distributions and excretion times. ~ ' In Vivo Metabolism and Ligand Displacement Biological studies in man and in laboratory animals show that the gold drugs used clinically undergo rapid metabolism in vivo. respectively. Thus.26-28The phenomenological term 'dissociation' that is sometimes applied to this finding is inapt. gold(II1) of the tetrahalide complexes remain powerful oxidizing agents.+ 4mL + 2nH' + (5) (6) (7) The exchange of monodentate ligands bound to gold@)or gold(II1) is generally rapid and proceeds by associative mechanisms that involve. value for AuBr. + 2RSH + Au'L.' Smith and Reglinski2' and Grootveld et ~ 1 .. threecoordinate gold(1) and five-coordinate gold(II1) transition states (and occasionally intermediates).. The presence of bidentate. but gold(1) can be stabilized by cyanide and thiolate ligands. ligand displacement is the . During the early 1990s. because cellular and biochemical studies demonstrate that ligand exchange reactions are the primary mechanism of metabolism and that free gold ions do not result. For the Au(SCy). + RSSR + 2L + 2H+ + +RSSR + nH20. but indicates that AuBr.complex (CyS = cysteinato).is significantly less than that for free Au3+. along with summaries of previous research that provides a basis for understanding the latest findings. Reduction of gold(II1) to gold(r) or gold(0) is often observed in biological milieux. tridentate or macrocyclic ligands reduces the thermodynamic tendency for ligand exchange and often leads to slower reactions even when they are favorable. These are discussed in several of the following sections. in addition to ' specialized reviews on cellular pharmacology by Crooke et a1. rnAu"'L.

The Thiol Shuttle Model The injectable gold drugs such as AuSTm and AuSTg are not readily taken up by most cells. cyanide alters the situation and facilitates gold uptake in red cells.PAu+ across the cell membrane. Et3PAuf is transferred to other cell sulfhydryls (CSH) and can then undergo several fates.. while the half-life for gold excretion is 20 days. in contrast to the polymeric thiolates.C . Frank Shaw III 33 preferred terminology.32Figure 3 shows the principal features of the model. respectively. Within the cell.28 Mirabelli and co-workers developed a sulfhydryl shuttle model for the uptake and efflux of Et3PAu+ from auranofin. '0-200 pgkg-' oral dose. proteins (1:5). while the triethylphosphine enters the cell but does not accumulate to the same extent as the gold itself.32P-'95Au-35SATg Rat blood in vitroaVb % Dog bloodjurine in vivoa. the three components of radiolabelled auranofin (Et 332P'95A~-35SATg) metabolized differently in dogs (Table 2): the half lives for are excretion of Et332P=0 and 35S are 8 and 16 hours. Likewise.. is extensively and quickly taken up by various cell type.28The relevance of the results to humans is clear from clinical results that detected free thiomalate (TmSH) in the blood of patients receiving myochrysine (AuSTm).^^^^^ The data for red cells in Table 2 show that the acetylthioglucose ligand is displaced before gold is transported into cells. Auranofin. .28 Cellular Chemistry . and a lowmolecular-weight species (2: 5 ) known to be phosphine oxide (Et3P0). 8h 2h 1. Rapid displacement of the ligands from 198A~35STg in mice and '98A~35STm rats has been r e p ~ r t e d .26 Studies of auranofin added to whole blood (Table 2) confirmed that the three components of the drug have very different fates and showed that the ligand displacements are dramatically faster than the excretion rates quoted above. ~The ~ ' in ~ ? retention of lg8Auin various organs is much greater than the retention of 35Slabel from the thiolate ligands. and the triethylphosphine is distributed among the red cells (2: 5). d72hour total excretion.5h tl. based on studies in cultured macrophage cells.c % Component Ig8Au Red cells 65+3 0 38+8 Y O Protein 35f3 82+1 19+1 % LM W <<1 19+1 43&7 t. Further reaction with additional cell sulfhydryls Table 2 Ligand displacement from radio-labelled auranojn: Et. b20min incubation (LMW = low molecular weight fractions). ~As* discussed in a ~ ~' subsequent section. Yet it is clear that they can bind to cell surface thiols and by such a mechanism may affect the cells' overall m e t a b o l i ~ m .2 Excretedd 6 81 90 35S-AtgSH 32P-Et3P 20 d 16h 8h "Ref 28.^.28 Within 20min the gold is primarily protein-bound in the serum. in which sulfhydryl-dependent membrane transport proteins (MSH) provide a vehicle for movement of the Et.

Therefore. With or without the Et. was added to the media of cultured macro phage^. . 32). . CSH. active transport. which contains serum albumin.CSH = cellular thiols (protein and non-protein). RSH = extracellular thiols (C'SH) may lead to displacement of the Et. ~ . the gold(1) can be shuttled out of the cell via the membrane sulfhydryl protein^. MSH = membrane transport-protein thiols.P=O.^^^^^ Although the uptake is mediated by the membrane sulfhydryls.P bound. it is not energy-dependent. intracellular gold concentrations should be in equilibrium with the extracellular sources of gold.34 Gold Complexes with Anti-arthritic.P ligand and its oxidation to Et. and red cells that had accumulated gold upon exposure to auranofin were shown to have more extensive gold efflux in the presence of added serum or just albumin than in isotonic salt solution.An equilibrium must .and extracellular auranofin metabolites. The cytotoxicity to tumour cells of both thiotheopyllinato (triphenylphosphine)gold(I) and auranofin is reduced by fetal calf serum.~ be reversible. This led to slower and less extensive uptake of the gold. which has a relatively high affinity for gold. which reduces the ~ p t a k e . Anti-tumour and Anti-HI V Activity SULFHYDRYL SHUTTLE MODEL RSH -TPUR ET~PAuSATGE~AS MS\sH RSAuSR' csHi[ + "SH MSH CSAUPE T ~ R'SH RSH RSAUET~ MSH S U E~ A P TG CSH OP C 8 S H j O 1 ET~ CSH C' SAUSC-s b MSH SAuSC a RSH i = RSAuSC 1 MSH Figure 3 The Sulfhydryl Shuttle Model (adaptedfrom Ref. This was tested in two ways. Serum albumin.33 These findings strongly support the existence of an equilibrium between intra.^ and B16 melanoma cells35 in increasing amounts. consistent with an equilibrium distribution.

C. will be critical in vivo targets. He and D. showing the domains. 11. including enzymes and transport proteins. subdomains and cys-34. each with two Figure 4 The structure of human albumin. and is paired with cysteine or glutathione (AlbSSCy or AlbSSG) in the remainder. Nature. Carter. Cys-34 is in the reduced form in mercaptalbumin. which comprises about 60-70% of the circulating albumin molecules. M. This is consistent with earlier literature establishing that it is protected from the solvent and only accessible due to motions previously described as crevice opening (Adapted from X.~ ~ ligands3' ~ ~ gests that proteins. Serum Albumin Serum albumin (Figure 4) is the major extracellular protein in the blood. & 111). it is clear that extracellular gold in the blood is primarily protein bound. 1992. The sections below summarize the bio-inorganic chemistry of several well-studied protein and enzyme systems that interact with gold complexes. In addition. It is held together by seventeen disulfide bonds and has an additional unpaired cysteine residue.sug. AlbSH. Cystein 34 is below the surface at the bend between helices h2 and h3 of domain IA.358. suggesting protein-mediated transport of gold during therapy. Albumin consists of three structurally related domains (I. C. Frank Shaw III 35 Protein Chemistry The high affinity of gold(1) for s ~ l f ~ r ~ and~ selenium * ~ ~ .209) .

45The rate of gold binding may correspond either to the rate of opening of the cys-34 crevice to solvent molecules or to the rate of this recently found structural change that accommodates gold binding. is not present in serum. + ATgSH -+ AlbSH + AT$-SCy Et. with an average Au-S distance of 228 2 pm:39 AlbS- + l/n[AuSTm].9 f 0.which indicates that auranofin (and probably its deacetylated metabolite) will have a very short lifetime after entering the bloodstream where albumin is present in large excess ( 400 pM AlbSH and 10 pM A u ) . acetylthioglucose or thioglucose replaces the thiomalate have been isolated and ~ h a r a c t e r i z e d .~' Under conditions approximating those in v i m . + GSH ~ + AlbS-Au-SG + PEt3 (12) (134 PEt.38Cys-34 is located in domain I-A and lies buried below the surface of the protein. Anti-tumour and Anti-HI V Activity subdomains (A and B). leading to its oxidation:47 - - AlbS-SCy AlbS-Au-PEt. AuSTm forms a complex in which gold is coordinated by cys-34 and thiomalate. re~pectively. + ATgSH (9) The Au-P and Au-S bond lengths of the phosphine-gold-albumin complex are 229 and 227 pm.48 AlbS-Au-PEt.36 Gold Complexes with Anti-arthritic.47The oxidant in equation 11 can be either molecular oxygen (equation 13a) or the albumin disulfide (equation 13b). yet the reaction proceeds at a similar rate and to about the same extent aerobically or a n a e r ~ b i c a l l y Since free 0.~~ bonds are the in vivo oxidants. The free acetylthioglucose liberated via equation 9 reacts further with the cysteine 34 disulfide bonds to liberate c y ~ t e i n e ~ ~also displaces the Et. -+ Et3P-0 L+[>Alb-X + H.--/ r+ 1/20.O -+(HS).P and ligand. complex formation is first-order in auranofin and has a rate constant of 2. it is likely that disulfide . ~ ~ ~ ~ ' Auranofin reacts at cys-34 via a ligand exchange reaction that displaces the sulfhydryl g r o ~ p : ~ ' .2 s.-Alb-X + Et3P0 .~ ~ AlbS- + Et3PAuSATg + AlbS-Au-PEt. Sadler ~~ and colleagues have recently discovered a conformational change in albumin that accompanies gold binding to cys-34 (but not disulfide formation at cys34). Extensive bio-inorganic research has established that the principle gold(1) binding site is cys-34. Physiological thiol ligands such as glutathione (GSH) also drive the displacement of phosphine (equation 12). -+ AlbS-AuSTm (8) Similar complexes where glutathione.P=O (10) (11) + ATgSH + AlbS-Au-SATg + PEt.

Pr'.5 01.~ ' Me.7 ppm) during reaction with serum albumin under argon.P Structure-function studies of the anionic ligand (Figure 5) revealed.p-1 -D-selenoglucose complexes. that replacement of the AtgSH ligand by CN.P z (NCEt).00 0.05 mM BSA and 2.c') 24 hours after mixing 4.PAuSATg (6.b. = 35.Frank Shaw 111 37 Structure function studies that substituted Me3P. For the cyanide (b. the Et. AlbSAuPEt.4 ppm) and (c.c') Et. = 61.s I I 30.a'). Hemoglobin (Hb) is 0. = 36.6 ppm) (b.P >> iPr.PAuf moiety crosses the membrane and reacts with hemoglobin and gl~tathione. the formation of Et.P > Et.P for Et3P of auranofin in these reactions have established that the basicity of the phosphine controls the rate and extent of R 3 P 0 f ~ r m a t i o n : ~ ~ . For auranojn (a.b') Et3PAuCN (6.PO (equation 11)than is observed with auranofin i t ~ e l f .4 61. (38. ~ ~ ? ~ ~ Hemoglobin In red cells exposed to 7mM Et3PAuC1.5 I r~ 70 00 I I I I I I ' ~ I I I I I I I 50 40 30 20 ID 0 70 60 50 40 30 20 10 0 PPm ( b W ppm (6W Figure 5 Formation of E t P O ( 6 .6 mM gold complex.4 ppm) from (a.PAuSeATg (6. which bind more tightly to gold than the thiol does.5 ppm) and (Et3P)#u+] (43. This demonstrates that increasing the afinity o the f anionic ligand for gold(1)will accelerate rather than retard the rate of metabolism o phosphinegold(1)-baseddrugs f .a') Et.or the selenium analogue. did not inhibit the first reaction (equation 9) and unexpectedly led to more rapid formation of Et. surprisingly.free drug and protein complex are present with only a trace of Et.c) 1.PO.8 ppm) (a.4 I 38. AtSeH. = 37.4 30.~.P or (NCEt).PO is much greater at both time points.b'.b') and tetraacetyl.C.00 61.5 hours and (a'.

including the anti-tumour drugs chlorambucil (a mustard derivative) and cisplatin' 5 * 5 6 and the gold-based anti-arthritic agent^. it loses the thiomalate ligand and binds to two cysteine sulfhydryls with dAuS= 229 pm:'. The kinetics are biphasic. = 9.). + Au3'STm + xsZn. complex described above. there is efficient interprotein transfer of Et.5 x and k.(SCy).60*61 The resulting aurothioneins have been characterized by a variety of chromatographic and spectroscopic techniques. + Hb(SH)2 (14) It is interesting to note that this transfer occurs spontaneously and that lowmolecular-weight thiols are not required to mediate the process... after administration of gold drugs. clusters. the mediation of Zn2+ and Cu' metabolism and.38 Gold Complexes with Anti-arthritic. Biological functions of MT include the detoxification of the environmentally accumulated Cd2+ ions.Cd-MT -+ Au.(SCy). Gold thiomalate displaces Zn2+ and Cd2+ from the protein in a biphasic reaction. When gold is limiting. A variety of animal experiments summarized elsewhere25 demonstrate that in the kidneys and liver..). and cys-p-112). which contains six cysteine residues (cys-a-104.and N-terminal ends of the protein. indicating that the Hb cys-p-93 residues are lower affinity thiols than are AtgSH and AlbSH. These reaction rates are much greater than are the rates of gold accumulation in the kidney or liver.PAuSTg. along with Zn2+ and Cu+. bind to Et.^^ mediates their cytotoxicity. and complete displacement of both metal ions is possible when excess AuSTm is present.PAuf from hemoglobin to albumin:'.Cd-MT + Tm3'SH + Zn2+ (15) .4 0. which are near the surface. and negligibly with Et. Aurothioneins Metallothionein (MT) is a ubiquitous metal-binding protein found in all vertebrate and invertebrate organism^.53 Consistent with this finding. The bio-inorganic chemistry of gold incorporation into M T is described below. Only the cys-p-93 residues.. cys-p-93. Mammalian MTs utilize 20 cysteine residues among about 61 amino acids to group their seven Zn2+ or Cd2+ ions into M.7 x lo-. Figure 6 shows the two clusters. Zn2+ is displaced in preference to Cd2' when gold is limiting. are not as robust as the AlbSAuPEt.. The resulting complexes. putatively. Anti-tumour and Anti-HI V Activity an a2P1tetramer. where n 5 2.~~ isolated a. Interestingly.and P-domains show that the fast step is associated with the a-domain and the slow step with the P-domain.6 1. Hb-(SAuPEt. and M. to a smaller extent with Et. = 2. adventitious reactions with certain medicinal agents.'. respectively.PAuSATg. which are in domains designated a and p and are formed at the C. gold is bound to metallothioneins. respectively.Zn. with rate constants of k.^'. The hemoglobin reacts completely with Et. + XsAlbSH -+ AlbSAuPEt. Hb-(SAuPEt. and therefore demonstrate that the reaction with MT is not rate-limiting for a c c ~ m u l a t i o nRecent studies with .'^ MTs in mammals and crustaceans bind seven and six metal ions.PAuf.PAuCl. the detoxification of reactive oxygen species.

the thiomalate ligand is retained and the initially bound Cd2 and Zn2-t are completely displaced: X S A U ~ ~ SZn.which have not been reported f to date.Cd-MT -+ (Tm35SAu). (A. the stoichiometry of bound gold approaches the number of available sulfhydryls. and the bond-lengths are similar.F) Isomorphous substitution and thiolate-bridged coordination o gold(1) in MT.Au-MT and (TmSAu)20MT.and three-metal clusters of M7-MT where M = Zn2+or Cd2+with tetrahedral MS.Zn. In both reaction products. (C. respectively.B) The four. 20 per MT. the gold remains gold(1). coordination. dAuS= 230pm. Theformulae under structure C-F are the limiting cases for saturation of the protein by each mode of gold coordination When gold is present in excess.D) Bidentate and monodentate chelation o Au(T) metallothionein as observed in f by Cd. Frank Shaw III A B 39 a-DOMAIN HqSll CLUSTER p-DOMAIN M3Sg CLUSTER C n S -Au-S D - S-AU -STm BIDENTATE CHELATION Au S 10 20 E MONODENTATE CHELATION (TmSAu) 2os20 Au ISOMORPHOUS SUBSTITUTION THIOLATE BRIDGED CLUSTER Figure 6 Metallothionein metal coordination.no evidence . (E.20MT+ 7M2+ + T~ + (16) In the latter case.C .

an effect that is attributed to the altered protein structure that results from two-coordination of Au(I) to the sulfhydryl groups. but Auranofin does not react in vitro with MT.O (17a) The active site of GPx-ase contains a selenol group (SeH) which is oxidized to Se(=O)H with formation of one equivalent of water in the first step of the enzymatic cycle (equation 17b). Four possible coordination models are shown in Figure 6. H 2 0 2 is formed by the oxidative burst at inflamed sites. in turn. Ref. glutathione peroxidase (GPx-ase). the twoIn coordination requirement of the gold(1)must distort the cluster structure. necessitating larger doses than would be needed in the absence of this protein. leading to altered ligation of the gold. and it completely unfolds the tertiary structure in (TmSAu).41-62 gold does bind to M T in vivo and in cultured cells after gold drugs have been administered. a thiol which has important functions as a biological reductant) reduce H .~ o r k e r s ~ ~ * ~ have described the interaction of gold thioglucose with the selenium-dependent enzyme. M T induced by auranofin is degraded more rapidly than it is after Zn2+ or Cd2+ induction (Table 3).63 The potential interaction of gold complexes with M T in vivo may reduce their cytotoxicity. Two equivalents of glutathione (GSH. In addition. Zn-MT.PAuCl. Et. Two equivalents of glutathione then react with Table 3 Biological half-lives of MT with various metal loadingsa Cd2+. GSSG (equation 14a). where surrounding cells can also be protected by the peroxidase activity.~' either case. are oxidized to the corresponding disulfide. Zn2 Au(I)~ + 24 10 0. Zn. bMT induction with auranofin.ion is present.75 aMeasured in cultured Chinese hampster ovary cells after MT induction with the indicated metal ion. reacts more readily and displaces bound Zn2+ from Cd. it is clear that Au. 20MT in monodentate c~ordination. 20MT. The apparent discrepancy reflects the rapid in vivo metabolism of auranofin. in which the less tightly bound Cl. 2GSH + H. 0 2 into water and.02-+GSSG + 2H. 63. Glutathione Peroxidase Elegant biochemical and animal studies by Tappel and c o .40 Gold Complexes with Anti-arthritic. From the AuS. coordination environment. Cd-MT contains Au(I) in bidentate coordination and (Tm35SAu). . This enzyme is essential to red cell integrity. Anti-tumour and Anti-HI V Activity for elemental Au or AU(III)is obtained. the stoichiometries of metal exchange and the retention or loss of the thiomalate ligand. because it detoxifies H 2 0 2formed fortuitously.

leading to cell toxicity that would slow the immune response and contribute to clinical improvement for the patients. therefore inhibition in vivo is likely. which can be interpreted as the concentration of gold which will reduce the enzyme activity by a factor of one-half.. the second GSH attacks the sulfur to release the glutathione disulfide and restore the selenol.3 and 10pM for gold(1) thiolates and triethylphosphinegold(1)complexes. These reactions show that care is required in interpreting the chemistry of gold(1) compounds in biological milieux.O. . The first forms an Se-S bond (Gpx-Se-SG) with release of the second water molecule.3pM.3pM.66The Et. In contrast. Frank Shaw III 41 the oxidized selenium. Tappel determined that the inhibition constant (&) for gold(1) thioglucose (AuSTg) has a value of 2. Et. many other enzymes which are inhibited by gold(1) have Ki values in the millimolar range or higher and are unlikely to be inhibited in viva Tappe16’ has suggested that inhibition of GPx-ase could lead to a buildup of H. because they undergo rapid ligand exchange reactions with thiols and other high affinity ligands that may be present. GPx-SeAuS GPx-SeH Px-SeSG 4 GPx-Se(=O)H Hi0 GSH The active site selenol has a high affinity for gold binding and is the site of inhibitory action (equation 17b). and this has now been confirmed by Roberts66 who showed that authentic Au(SG).PAuCl and Et.C.PAuSATg have similar Ki values. 2.. For the same reason.in the assay mixture. The enzyme assay uses 1 mM GSH for the reaction and Tappel et ul. respectively. at inflammatory sites.64*65 suggested that the gold(1) thioglucose was converted in situ to bis(glutathionato)gold(I). Finally. are below the 5-1 5 p M concentrations of gold observed in patients undergoing chrysotherapy and.exhibits the same Ki value. 10 & 1 pM. The Ki values of 2.P ligand has a stronger affinity than GSH for gold@) and is not displaced to form Au(SG). because they are converted into Et.PAuSG by in situ reaction with GSH.

which is absorbed through the lungs. It can react with gold and will facilitate the transfer of gold into red cells and presumably other cells.69 This suggests that formation of Au(CN).67[For obvious reasons this phenomenon was not observed in laboratory animals.is taken up . Elder has measured two apparent equilibrium constants:73 + 2CySH Au(SCy)i + 2 HCN (23) Traces of Au(CN). which follow the general equations be lo^:^'-^^ The equilibria depend on pH. the affinity of the thiol(ate)s for gold(I).. The first evidence of Au(CN).have been identified as a common metabolite of all three gold drugs (AuSTm. in the metabolism of gold drugs... AuSTg. as discussed later in this chapter.also has anti-HIV activity. Anti-tumour und Anti-HI V Activity Cyanide Metabolites Evidence accumulating over the past decade has established the importance of cyanide complexes.-. Research on the effect of gold compounds on the immune system led to the finding that thiocyanate (SCN-) can be converted into cyanide by hypochlorite generated during the oxidative burst of immune cells. and the ratios of thiol and cyanide to gold that are present in the system under study.. The Au(CN).] The involvement of cyanide also explains the higher incidence of side effects from gold drugs in patients who ~ m o k e . principally Au(CN). but the concentration is somewhat higher in smokers. whose life spans are briefer than the minimum age for cigarette purchases.may be a critical metabolite at the inflammatory sites in the arthritic joints of patients.42 Gold Complexes with Anti-arthritic. although animal red cells do not.67968 inhaled smoke of The tobacco products contains up to 1700ppm of HCN. ~ ~ ? ~ ~ Several studies have examined the chemistry of thiolkyanide exchanges. Au(CN). and auranofin) used clinically in the United state^. .^.^^ Its presence is independent of the smoking habits of the patients.as a metabolite was the report that tobacco smoking enhances the ability of patients' red cells to accumulate gold metabolites of AuSTm and AuSTg.

. The PLN assay is important because it discriminates between the effects of a drug and those of its metabolites.’5*17 Thus. as well as other thioethers.in the blood~tream.14 Even disulfide bonds react rapidly to reduce gold(III). but AuSTm does not.. gold exists.C. ~ ~ * ~ * This finding is based on the observation that after mice have been tested with AuSTm (gold sodium thiomalate. 1 9 . Frunk Shuw III 43 into the cells and can limit the extent of the oxidative The mechanism of Au(CN). gold(II1)elicits a response in the popliteal lymph node assay (PLNA).8.transport into red cells appears to be via the sulfhydryl shuttle mechanism (see Figure 3)..76 exemplified by the protein reactions described above.The. Figure 1) for several weeks. in order to determine which is immunogenic. primarily as gold@). 33.~’ These newly found roles of Au(CN). are also capable of reducing gold(@ to gold(1). A considerable body of evidence suggests that in uiuo. Aurosomes (lysosomes that accumulate large amounts of gold and undergo morphological changes) taken from gold-treated rats contain predominantly gold(I). although only the former is physiologically ~ i g n i f i c a n t . Subsequent research in another laboratory . even when gold(II1) has been a d m i n i ~ t e r e d . stronger interaction is surprisingly large. and would be expected to exist.63 Intact Au(CN).'-' and Smith and Brown actually predicted it.’ Gleichmann and co-workers have observed that gold drugs can be activated in uiuo to a gold(rI1) metabolite that is responsible for some of the immunological side effects observed in c h r y ~ o t h e r a p y . but the ~~ ~~ labile binding is easily reversed and thus is consistent with cellular uptake of Au(CN).~ thionine in proteins and peptides. ~Me-~ ’.may open the door to new research and to studies on the mechanisms of action that will end the long uncertainty about where and how the gold drug metabolites exert their effects on the inflamed joints and/or immune systems of patients..ions bind to serum albumin at one strong and three weak binding sites. various bio-inorganic chemists have always been careful to point out the potential for oxidizing gold@ to gold(II1) in viva. Oxidation States In Vivo Chemical reactions of gold drugs exposed to body fluids and proteins are predominantly ligand exchange reactions that preserve the gold(1) oxidation state. it appears that the bulk of the gold present in viuo is likely to be gold(+ Nonetheless.

industrial workers exposed to platinum.> - AuSR f 40C1- AUC1. thioethers. In another context. can oxidize the gold in AuSTm to Au(III). + RSOC AuCl. as they are too small to be recognized.74 For example. Certain of these peptides . in responding patients are much larger than can be accounted for on a stoichiometric basis. the formation of gold(II1) in vivo is not only feasible. AuSTm.~'This finding has been extended to additional gold compounds. nickel or other metals frequently develop allergies and other immune It is unlikely that the immune system is reacting directly to the metal ions. the changes in tissue levels of metals. auranofin and Au(CN). thiols. etc. they do it by hydrolyzing proteins into short peptides and. but demonstrably relevant to human therapy.44 Gold Complexes with Anti-arthritic. is a redox cycle:74 OC1.P=O (27) (28) This clearly establishes the chemical feasibility of gold oxidation by hypochlorite and provides a plausible mechanism for the findings of G l e i ~ h m a n n and .~' An explanation for the apparent dichotomy between the observations that gold is present primarily as gold(1) in vivo and that T cells are sensitized to gold(m). palladium. which is generated by the enzyme myleoperoxidase during the oxidative burst.are oxidized to AU(III). Antigenic Peptides Immunologists have found that the treatment of mice77 and men7' with antiarthritic gold drugs generates T-cells that react to gold(II1)but not to the original gold(1) drugs. ultimately.. not to the gold drugs themselves. proteins. into individual amino acids. ~ ~ ~~ Ver~ilghen. there are a number of ways that metal ions can alter the formation and presentation of peptides to the immune system as it attempts to distinguish foreign matter from When macrophages and other antigen presenting cells digest foreign substances and broken-down self tissues. Instead. A cursory report by Beverly and Couri suggested that hypochlorous acid (HOCl). disulfides The operation of such a cycle is consistent with observations that while relatively low concentrations of gold are present during chrysotherapy (10-50 pM Au). the ligands are oxidized preceding or in and concert with the gold oxidation:74 > EbPAuSATg + 50CI.(oxidative burst) I 1 Drugs -+ Protein-Au'-Ligand Au'"X4 - (29) 2 thiols. + ATgSOi + Et.^^ Thus. Anti-tumour and Anti-HI V Activity confirmed that T cells from human chrysotherapy patients are sensitized against gold(n1) but not gold(^).

~’ a result. via pathway 1.: H. Although peptides derived from self proteins do not normally react with T-cells in this way. As chrysotherapy was described by the Gleichmann group in Dusseldorf. The first model is based on the ability of Au(111)to alter the response of T-cells to ribonuclease-A RNase injected into mice normally generates.designated cryptic . certain auto-immune diseases arise when T-cells react to self proteins. RNase. which is considered to be immunodominant and stimulates a T-cell response.N-Gln-Ser-Tyr-Ser-Thr-Met-Ser-Ile-Thr-Asp-Cys-Arg-G~uThr-G1y-COOH . Frank Shaw ZZZ 45 bind to large macromolecular structures called major histocompatibility complexes (MHC) or human leucocyte antigens (HLA). another cause is presentation of unfamiliar self peptides . Scheme Interestingly. which can occur due to a variety of reasons. These MHCs are transferred to the outer surface of the macrophage cell membrane where T-cell receptors screen the MHC-peptide complexes (pathway 1 of the Scheme) and react with those carrying peptides of foreign origin.. In that case the T-cells stimulate an immune response.which stimulate a response. They have developed two bioinorganic model systems for studying the effects of gold on the immune system. the peptide composed of amino acids 74-88 in the protein sequence. Sometimes aberrant T-cells learn to recognize dominant self-peptides that are usually ignored. Among these autoimmune diseases is rheumatoid arthritis._.. Germany as a double-edged sword since it halts the autoimmune attack on the joints and simultaneously induces unwanted side effects.C. immunological side effects to the gold drugs are common among those patients who respond successfully to ~hrysotherapy.

. Shaw 111...-.N-Gly-Ile-Val-Glu-Gln-Ser-Cys.-Ala-Ser-Val-Cys .46 Gold Complexes with Anti-arthritic. d s c c : I S-Au-S _ H. Thus.CH. : RNase..-.. Mufioz and C./CCC)...N-Gly-Ile-Val-Glu-Gln-Cys.N-Gly-Ile-Val-Glu-Gln-Ser-Cys. Au-Ins.-. The 1: 1 complex is formed with loss of the thiomalate ligands.-Ser-Val-Tyr-COOH When excess A u ( S T ~ ) . this system models the side effects observed in humans and mice where the secondary response to gold is not to the original drug.N-Asn-Cys-Ala-Tyr-Lys-Thr-Thr-Gln-Ala-Asn-Lys-His-IleIle-Val-COOH Treatment with AuC1.-. demonstrates that the effects of bovine insulin on T-cell response can be moderated by simultaneous exposure to gold(1) thiomalate. gold(1) thiomalate. which are considered to be cryptic since they are not normally recognized are generated (pathway 2) and stimulate a response from other T-cells: RNase. H.) of the RNase16 and. also involving T-cells from mice.to form complexes with one or two Au(I) ions (A.- S-Au-STm3- H. S-Au-STm3- . however. where cysteine 6 is replaced by a serine which effectively changes the cysteine SH group to an OH. In particular the steric constraints of the monogold complex will destroy the extended structure necessary for binding of the peptide to the MHC complex (pathway 3). Recent studies of a mutant (Ins. but to a protein altered by AuC1. hemoglobin and metallothionein show a high affinity for Au(I).-. a digold complex which retains two thiomalates is formed. subtly alters the protein structure leading to the formation of the cryptic peptides. the effect of the gold(1) .-Ala-Ser-Val-Cys.-Cys . unpublished results).. 7 and 11 (designated Ins. peptides 7-21 and 94-108... F./SCC). Hence.’.oxidizes the methionine residues (CH. -Ser-Val-Tyr-COOH r As discussed elsewhere in this chapter. (TmSAU)2-InSA1_. The second model system.SCH. the RNase is first pretreated with AuCl.86 The T-cells react to peptide A1-14 (the first 14 amino acids of the insulin A chain)... Anti-tumour and Anti-HI V Activity If. rSHrSH SH H. The gold adducts will have very different electronic and steric properties.: ./SCC: 1 ...-.-Ala-Ser-Val-Cys -Ser-Val-Tyr-COOH These structures are consistent with the tendency of gold to form two-coordinate linear structures with thiolate ligands. show that it can react with Au(STrn). ~ is used. which contains three cysteine residues located at positions 6 .N-Lys-Phe-Glu-Arg-Gln-His-Met-Asp-Ser-Ser-Thr-Ser-AlaAla-Ser-COOH H.. thereby. the cysteine residues of serum albumin.

PAuS-sugar > R. halide ions or other donor atoms replacing the thiosugar are less potent. ~ ~ . in turn. Oligomeric thiolates. auranofin is not a broad spectrum anti-tumour agent. All of the compounds with strong anti-tumour activity in vivo also have potent cytotoxicity when measured against B16 melanoma in vitro.=: R. Analogues with thiolate ligands. but not sufficient criterion for anti-tumour activity. ) . and (3) complexation of gold@) gold(m) with other active anti-tumour agents in order to or enhance the activity and/or alter the biological distribution of the latter. (2) the structural analogy of square-planar gold(II1) to platinum(I1) complexes. The similarity of to ~ the irreversible inhibition of DNA synthesis by auranofin and by the clinically used cisplatin agent was When tested against P388 leukemia in mice.C. 4 Anti-Tumaur Activity Interest over the last several decades in the potential anti-tumour activity of gold complexes has been driven by three common rationales: (1) analogy to the immunomodulatory properties underlying the benefit from gold(1)complexes in treating rheumatoid arthritis. have very little in vivo Thus. This relationship demonstrates the importance of in uivo rnetabolism and animal testing in drug design and development. ~ ' . ~ by Haiduc and S i l ~ e s t r uSadler et ~ 1 and S ~ h a ~ . which are potent anti-tumour agents..90 The uptake and incorporation of ['HI thymidine into DNA is reduced by brief exposure of cultured HeLa cells9' or EBB-transformed l y r n p h ~ c y t e s ~ 50-100 pg dl. as some highly cytotoxic complexes have limited in viuo efficacy.auranofin. ~ .^'^^^ Further testing of related complexes (Table 4) established that the tertiaryphosphine gold(1) complexes with thiosugar ligands (R3PAuS-sugar) had maximal antitumour activity in vivo when tested against P388 leukemia.PAuX > [AuSR]. including the injectable anti-arthritic drugs myochrysine and solganol. Unfortunately. Its activity in uivo is limited to the P388 cell line (Table 5) when administered intraperitoneally ( i . Cytotoxicity is a necessary. [AuSR].~~ Auranofin and Analogues Lorber and co-workers reported in 1979 that auranofin (Figure 1)inhibited the proliferation of HeLa cells in culture.PAuS-alkyl . This. the general trend in activity is ' R. suggests that the mechanism by which gold therapy causes remission of rheumatoid arthritis might be to prevent the presentation of a cysteine-containing immunogenic self-peptide to which the patients' T-cells react. Three reviews of the topic in various contexts have been published over the last decade. auranofin produced significant. Frank Shaw IiI 47 thiomalate in this immune model system is presumably to alter the peptide structure and prevent its binding to the MHC molecule. . The converse relationship was not observed. dose-dependent increases in life pan.

. 101 100.R' Y X C6H5 C6H5 3-F-C6H4 2-C2H5N C6H5>C2H5 (CH2)3 CH.48 Gold Complexes with Anti-urthritic.PAuS-Glutathione Et.5 2 7 4 4 2 1 60 1 0. (CH2)2 (CH2)2 c 1 c 1 c 1 c 1 c 1 c 1 0. using B6D2F1 mice injected intraperitoneally with the maximum tolerated dose (MTD) daily for 5 days. ']Cl DPPE DPPE-02 Bis(DPPE)gold(I) analogues [Au(&PYPR3. D PP E(A ~"'C13)~ DPPE(Au'SGlu)2 [Au(DPPE). 'C1B16 in vitroa MPM) P388 in vivob ILS.PAuSCN Et. 101 "B16 melanoma tested in uitro is the dose that inhibits growth by 50%).101 93 93 Oligomeric gold(1) thiolates [AuSTm]. Anti-tumour and Anti-HI V Activity Table 4 Activity of gold complexes against P388 leukemia and BI 6 melanoma cells Complex Auranofin analogues Et.6 2 - C2H5 cisplatin (CH.PAuS-P-Gal(Ac). [epi-auranofin] Et..). uiuo. 1.( %) Reference . Et.101 loo. [AuSGlu]. .]X R.30 125 f 38 100. Et.PAuPEt.101 100. Et. 5 17 - 83 +_ 25 70. [auranofin] Et SPAuS-P-Glu Et.101 100.PAuCH.101 loo.101 100. PAuS CH Et .PAuS-a-Glu(Ac).). [AuSGlu(Ac).PAuCN Et.]. DPPE complexes D P P E(Au'Cl).4 1 1 60 166 150 8 15 4 2 60 inactive 70 68 58 65 88 32 36 36 36 68 55 36 24 15 14 82 82 82 93 93 93 93 93 93 93 93 93 93 93 93 93 93 96 100.= (CH.83 90 & 17 89 f 28 92 & 26 45.PAuS-P-Glu(Ac).55 75 f 4 54 +_ 16 40.101 100.PAuS-P-Glu(CONHMe).101 100.PAuCl Et.

6 pM when tested against normal FLC cells and 1. is active (T/C = 167) against Ehrlich ascites tumours in mice.c.8 p M against doxorubicin-resistant cells (Dox-RFLC).PAu). values are 0.PAutTP (Figure 7) is active against several cell types. where DTE = dithioerythritol.p.) active P388 leukemia (i-v. a modified purine base. Ph. Ref:98 active inactive active active - [Au(DPPE).) inactive mammary adenocarcinoma 16/c (s.94 Triphenylphosphine(8-thiotheophyllinato)gold(I). s.c.) inactive Lewis lung carcinoma (i. is the possibility of increased DNA-gold interactions..p.c. The IC.” The rationale for using the tTP ligand.p.. s.) inactive mammary adenocarcinoma 13/c (s.) - active inactive active active inactive active - active inactive active active Stocco et al.) inactive L1210 leukemia (i.) inactive Madison 109 lung carcinoma (i.p.) inactive M5076 reticulum cell sarcoma (i.C .1 inactive B16 Melanoma (i..c.p. Frank Shaw iIi 49 Table 5 Activity of selected gold complexes against various tumour lines ~~~ ~~ ~~ ~~~~ ~~ Turnour (implantation) Auranojin Ref:35 DPPE(AuCZ).] C I Ref: 100 P388 leukemia (i. s. found that (Ph.(p-DTE).p..c.) inactive colon carcinoma 26 (i. 0 8-( thiotheophyllinate)(triphenylphoJphine)QO~d(~) streptonigrin Figure 7 The structures of PhPAutTP and streptonigrin . which carry a multi-drug resistance gene and differ significantly from the normal FLC cells in membrane fluidity.P.

(pDPPE)(AuCl).) has no activity. value was 2 pM (Table 3). were investigated for anti-tumour a ~ t i v i t y ."' P. -t and a precipitate containing Cuand DPPE. and three DPPE-gold complexes with antitumour activity .. but the greatest effect is on protein synthesis. which is more active than the precursor^:^^ The reaction product. slightly more potent than the open chain digold compounds of DPPE.+ is stable in serum for 25 hours. and its oxidation product (DPPE-0. auranofin and its analogues exert their primary effect on DNA ~ y n t h e s i s .uM Au(DPPE). ~ ~ ~ ' ~ shows that the ligand itself has Table 4 limited activity. PPh2 sT9 ST9 0 It WPPh2 Figure 8 DPPE. the against various tumours (Table l). It inhibits DNA.. however. established that the gold complex is a better inhibitor of DNA polymerase c both in vitro and in permeabilized t KB cells. the chelate effect (since DPPE coordinates via two phosphorus donor atoms) and the greater stability of phosphine over thiolate ligands.+. The role of copper and gold in the complexes may be to stabilize DPPE against oxidation. re~pectively).50 Gold Complexes with Anti-arthritic.(DPPE). inhibits DNA biosynthesis [measured as (3H)-thymidineincorporation into DNA] more effectively than it inhibits RNA or protein biosynthesis (measured as ['HIuridine and [3H]leucine incorporation into RNA or proteins.97 as CU(II) also enhances the cytotoxicity of DPPE in the B16 melanoma assay and increases its ability to inhibit DNA polymerase a.+ (Figure 7). showed much broader activity than did the auranofin analogues discussed above (Table 5). The complex reacts with CU(II). DPPE-O.97 (p-DPPE)(AuSGlu). which can be shut down by 90 minutes exposure to 15 . hP n PPhi Ph2 PhiP Au PhZP n I Au C I 1 A PPh2 Ph2P PPh2 I I CI Au I I I Au I / Ph2P \A"4 '. In contrast. the digold complexes rearrange to form a new species..98 The activity is not unique to gold complexes. forming Au."' The stability and inertness result from two factors.97 In culture medium (and presumably in vivo). has a useful spectrum of activity Against B16 melanoma in vitro.97When tested against other cell lines. Anti-tumour and Anti-HI V Activity Bis(dipheny1phosphine)ethane Gold Complexes As an outgrowth of the work on auranofin analogues. and in the presence of glutathione for 8 days. Au(DPPE). digold complexes with a bridging bis(dipheny1phosphine)ethane (DPPE) ligand (Figure 8). ~ ~ ' ~ ~ Au(DPPE). ". while the digold complexes are considerably more potent. RNA and protein synthesis.~~ Comparisons of DPPE and (p-DPPE)(AuSGlu).'oo*'o' IC.

For example.H. or that the entire Cu-DPPE molecule is necessary for a~tivity. many inorganic chemists were attracted to the structural similarities of gold(II1) and platinum(@..] developed into a clinically important chemotherapy agent.. anti-tumour agents is to attach gold@)or gold(m) to a compound that has anti-tumour potency and good metal-ligating donor atoms. It forms a highly stable gold(II1) complexlo5that is not altered by exposure to serum albumin. trans-Ph. and 5 after tumour implantation. C1-. pyridyl or fluorophenyl derivatives. 5-fluorouracil.P-C=C-PPh.PtCl.'~ Gold Complexes of Known Anti-Tumour Agents A common rationale for the design of potential gold-based.105 '.Io4 Streptonigrin (Figure 7) is a substituted 7-amino-quinoline-5. Both are d8 metal ions and form square-planar. = 0.30 mg kg--' on days 1. Ph3PAu-5-fluorodeoxyuridine and Ph.'~ not in vitro activity against several cell lines.P also eliminates the Cu-induced enhancement and but the in vivo a~tivity. Snyder et al.8-dione with anti-tumour activity that is complicated by high toxicity." showed that the diphos analogues that have enhanced cytotoxicity when bound to copper are active in vivo against P388.. suggested that CU(I)may function to transport DPPE.C. and para-Ph. but for the tegafur complex there was no increase.P-C. The latter group includes ligands that are structurally unable to chelate to a metal ion because of rigidity in the linker region: '" Ph..-PPh. fourcoordinate structures. However. Generally the phenylphosphine complexes are more active than are ethyl. conversely.As for Ph.5 pgml-'. 3.N).Frank Shaw I I I 51 Structure-function relationships were examined for complexes with a variety of related diphosphine ligands. Substituting Ph. Ph. Snyder et al.PAu-tegafur [tegafur = 5'-fluoro-l-(tetrahydro-2-furanyl)-2. a high affinity binding site for gold(r). the thymidine complex produced a T/C Of (%) of 171 against P388 tumours in mice treated with 43 pmol kg. a known anti-neoplastic agent] have been characterized and tested against L1210 leukemia in mice: the activity of the gold-5fluorouridine complex was greater than that of the ligand itself. Cisplatin Analogues As cisplatin [(H.PAu-nucleotide complexes of 2-thiouracil.'" As would be expected when a cationic complex is the active species. Sadler's review of anti-tumour activity by compounds of . that DPPE may transport CU(I)to the target site.P-C-C-PPh. and a good reducing agent for gold(II1). the nature of the anion (e.3H)pyrimidinedionato-N3. IC. whereas the analogues that are not more cytotoxic in the presence of copper are inactive. or.'02 From these results.4-( 1H."~ these.g. the activity of the complex against P388 tumour cells in vitro is the same as that of the streptonigrin itself.-) does not substantially affect the potency of the drug. thymidine and 6-mercaptopurine were in~estigated. 5-fiuorodeoxyuridine.or NO. Br.

the DPPE or the metal ions were the active component of the gold and copper complexes.]Cl. and [A~(en).and its copper analogue. the striking activity of the complexes shows the untapped potential of inorganic species in general and gold compounds in particular as medicinal agents.Y (Y = pyridine. in fact.88 Two dimethylgold(II1)complexes noteworthy for their similarity to cisplatin. Au) elaborates this rationale. where the greater oxidation potentials and more facile ligand exchange reactions of gold(II1) lead to rapid.Au(p-SCN).][AsPh.. This is demonstrated. pyAuC1.~~ While the rationale based on coordination geometries is attractive to structural inorganic chemists.AuMe. [Me. it fails in uitro. an enzyme that converts viral RNA into DNA in the host cell. irreversible formation of new and inactive metabolites that cannot cross-link DNA with the same efficiency at Pt(I1).-.'^'^ The center of attention quickly passed from auranofin analogues to (p-DPPE)Au. to Au(DPPE). i. Bu. a variety of gold(II1) complexes without the redox-stabilizing methyl groups were inactive: PhAuC1. Whether. AuC1. The lability of gold(1) and gold(II1) to ligand exchange and of gold(II1) to redox reactions must be recognized in efforts to synthesize new drugs. Ag. Anti-tumour and Anti-HI V Activity Group IB (Cu.inhibits the infection of MT-4 cells by HIV strain HL4-3 without inhibiting the RTase activity in the intact virons. The chemistry and biology of the DPPE complexes of copper(1)and gold(1) and the streptonigrin complex of gold(II1) demonstrates that the chelate effect shows promise for developing compounds that will remain intact for longer periods in viuo and permit the delivery of gold to potential target sites. but it is unable to enter the cells where RTase acts.. the finding of significant cardiotoxicity in animal testing has brought this line of research to a sudden halt.lo8Au(STg).AuCl. had activities exceeding 120%.lo6 Despite the absence of any clinical trials of a gold complex to date. The modification of known anti-tumour agents also deserves further exploration.-. Prospects for Gold-Based Anti-Cancer Drugs The intense testing of new gold complexes for anti-cancer activity in the 1980s was stimulated by the finding that auranofin had anti-tumour and cytotoxic proper tie^.. by the broad-spectrum activity of Au(DPPE). A different mechanism of action has been proposed for Au(STg). Unfortunately. However. The critical target site has been tentatively identified . 5 Anti-HIV Activity Investigations of the anti-HIV activity of gold complexes were stimulated by reports that AuSTg inhibits reverse transcriptase (RTase).X2 digold complexes. Robust new ligand structures that can move gold through cell membranes and into the cytoplasm and perhaps into the nucleus are required. the potential for developing new cytotoxic gold complexes that have anti-tumour activity is clear. which can be generated in situ from AuSTg and TgSH.lo7 AuSTg is indeed active in the cell-free extracts where it was studied. for example.52 Gold Complexes with Anti-arthritic.+.S).] and Me.

PAuCN Cy.. Gurney. it retards the proliferation of HIV in these cells.10 0. . as presented here. is AuSTm and AuSTg. cys-532 on gp160.10 97 108 86 88 78 100 101 85 71 0.PAuCl). L.uM concentrations (Table 6). bMTT Survival Assay using MT-4 cells incubated at 37 "Cfor 5 days (the assay is described in Ref. are consistent with this view. The bioinorganic Table 6 Toxicity and HIV-I antiviral activity of gold phosphines" MTT survival results (%)b Concentration (pM) 1. which suggests that Au(CN).PAuCl Me..is taken up into H9 cells. Okada.-. but the oligomeric 1: 1 thiolates. are not active.97Interestingly. A u ( S T ~ ) .^^ - 6 Conclusions Our understanding of gold complexes at the biochemical and cellular levels has evolved rapidly in the last decade. are inactive against the HL4-3 strain (Table 6). ~ -also active.o Toxicity (Au only) 0.Frank Shaw III 53 as a thiol group.PAuCN Ph. . Among the most important conclusions is that the medicinal agents are actually pro-drugs. A. which are able to enter cells. and protein-gold complexes in the metabolism of gold compounds during chrysotherapy. 61 130 92 0 0 0 0 60 0 122 96 96 73 86 103 91 127 59 "Unpublished data of T.PAuSATg Et. Several trialkylphosphinegold(1) cyanides (R. At concentrations as low as 20ppb. Similar conclusions have been reached from consideration of the pharmacology of gold complexes. F.~ Results from numerous laboratories spanning a range of disciplines.may have promise as a complement for existing HIV treatment^. M.PAuSTg Et.values are % surviving T-cells compared with controls.PAuCN) are inactive below the onset of cytotoxicity to the host T cells at 1 .010 1.C. there has been as little real progress in understanding the mechanisms of chrysotherapy as in the preceding half century.010 Et. These findings are consistent with an earlier prediction that the bioinorganic chemistry of gold complexes requires the formation of common metabolites from the parenteral drugs and a~ranofin. 108).PAuCN KAu(CN). Auranofin and two analogues (Et3PAuSTg and Et.' Despite the great gains in understanding cellular and biochemical aspects of gold metabolism in the last decade. which is a glycoprotein of the viral envelope.o + HIV-l(NLA3) 0. Among the most exciting findings are the roles of Au(CN). a continuous T-cell line that is susceptible to HIV infection.PAuCN iPr.PAuCN Et.'09 Tepperman and colleagues73have found that Au(CN). Arendt and C. gold(III). The concentration used is well tolerated in arthritis patients. Shaw 111.

54

Gold Complexes with Anti-arthritic, Anti-tumour and Anti-HI V Activity

chemistry clearly shows that the medicinal agents are pro-drugs with short half-lives in vivo. There is a compelling need for new experimental designs involving the metabolites, because they either (1) are the active species or (2) generate it at the presently unidentified sites of action. The possibility of developing anti-tumour agents based on gold or using gold to modulate organic drugs through ligation is a tantalizing prospect. The recent reports of anti-HIV activity for cyanide and thioglucose derivatives may be even more significant and clearly warrant active pursuit of their mechanism(s) of action.

References
W.F. Kean, C.J.L. Lock and K. Howard-Lock, Inflammopharl?zacoEogy,1991,1,103. M.C. Grootveld, M.T. Razi and P.J. Sadler, Clin. Rheumatol., 1984,3,5. G.J. Higby, Gold Bull., 1982,15, 130. D.T. Felson, J.J.Anderson and R.F. Menan, Arthr. Rheum., 1990,33, 1449. I. Jaffe, in The Columbia University Complete Home Medical Guide, ed. D.F. Tapley, T.Q. Morris, L.P. Rowland and R.J. Wiess, Crown, New York, 1989, pp. 602-625. 6 R.C. Elder, Z . Zhao, Y. Zhang, J.G. Dorsey, E.V. Hess and K. Tepperman, J . Rheumatol., 1993,20,268. 7 P.J. Sadler, Struct. Bonding, 1976,29, 171. 8 C.F. Shaw 111, Inorg. Persp. Biol. Med., 1979,2,278. 9 D.H. Brown and W.E. Smith, Chem. Soc. Rev., 1980,9,217. 10 R.J. Puddephatt, The Chemistry of Gold, Elsevier, Amsterdam, 1978. 11 P.G. Jones, Gold Bull., 1981,14, 102,159; 1983,16, 114. 12 M. Melnick and R.V. Parish, Coord. Chem. Rev., 1986,70,157. 13 H. Schmidbaur, Gold Bull., 1990,23,11. 14 M.C. Gimeno and A. Laguna, Chem. Rev., 1997,97,511. 15 P.L. Witkiewicz and C.F. Shaw 111, J.C.S. Chem. Commun., 1981,1111. 16 A.A. Isab and P.J. Sadler, Biochim. Biophys. Acta, 1977,492,322. 17 C.F. Shaw 111, M.P. Cancro, P.L. Witkiewicz and J. Eldridge, Znorg. Chem,, 1980,19, 3198. 18 C.F. Shaw 111, ed., ‘Proceedings of the 3rd International Conference on Gold and Silver in Medicine’, Metal Based Drugs, 1994,1,350-529. 19 R.V. Parish and S.M. Cottrill, Gold Bull., 1987,20, 3. 20 W.E. Smith and J. Reglinski, Persp. Bioinorg. Chem., 1991,1, 183. 21 M.C. Grootveld, M.T. Razi and P.J. Sadler, Clin. Rheumatol., 1983,3 Suppl. 1, 5. 22 S.T. Crooke, R.M. Snyder, T.R. Butt, D.J. Ecker, H.S. Allaudeen, B. Monia and C.K. Mirabelli, Biochem. Phavmacol., 1986,35,3423. 23 R.C. Elder and M.K. Eidsness, Chem. Rev., 1987,87,1027. 24 C.F. Shaw 111, Comments Znorg. Chem., 1989,8,233. 25 C.F. Shaw 111 and M.M. Savas, in Metallothioneins: Synthesis, Structure and Properties of Metallothioneins, Phytochelatins and Metal-Thiolate Complexes, ed. M.J. Stillman, C.F. Shaw I11 and K.T. Suzuki, VCH Publishers, New York, 1992, pp. 144-1 63. 26 S.R. Rudge et al., J . Rheumatol., 1984, 11, 150. 27 H.A. Schwartz et al., Am. J . Physiol., 1960,199,76. 28 A.P. Intoccia et al., in Bioinorganic Chemistry of Gold Coordination Compounds, ed. B.M. Sutton and R.G. Franz, SK&F, Philadelphia, 1983, pp. 21. 1 2 3 4 5

C. Frank Shaw 111

55

29 S.M. Cottrill, H.L. Sharma, D.B. Dyson, R.V. Parish and C.A. McCauliffe, J . Chew SOC. Perkin Trans. 2,1989, 53. 30 W.E. Smith and J. Reglinski, Persp. Bioinorg. Chem., 1991,1,183. 31 W.E. Smith, J. Reglinski, S. Hoey, D.H. Brown and R.D. Sturrock, Inorg. Chem., 1990, 29,5190. 32 R.M. Snyder, C. Mirabelli and S.J. Crooke, Biochem. Pharmacol., 1986,35,923. 33 C.F. Shaw 111, A.A. Isab, M.T. Coffer and C.K. Mirabelli, Biochem. Pharmacol., 1990, 40, 1227. 34 R.M. Snyder, C.K. Mirabelli and S.T. Crooke, Biochem. Pharmacol., 1986,35,923. 35 C.K. Mirabelli, R.K. Johnson, C.M. Sung, L. Faucette, K. Muirhead and S.T. Crooke, Cancer Rex, 1985,45, 32. 36 M.P. Arizti, A. Garcia-Orad, F. Sommer, L. Silvestro, P. Massiot, P. Chevallier, J.M. Gutierrez-Zorrilla, E. Colacio, M. Martinez de Pancorbo and H. Tapiero, Anticancer Res., 1991, 11, 625. 37 M.N. Akhtar, A.A. Isab and A.R. Al-Arfaj, J . Inory. Biochem., 1997,66,197. 38 D. Carter and J.X. Ho, Adu. Protein Chem., 1994,45, 153. 39 C.F. Shaw 111, N.A. Schaeffer, R.C. Elder, M.K. Eidsness, J.M. Trooster and G.H.M. Calis, J . Am. Chern. SOC.,1984,106, 3511. 40 M.T. Coffer, C.F. Shaw 111, M.K. Eidsness, J.W. Watkins I1 and R.C. Elder, Inorg. Chem., 1986,25,333. 41 D.J. Ecker, J.C. Hempel, B.M. Sutton, R. Kirsch and S.T. Crooke, Inory. Chem., 1987, 26, 3 139. 42 M.T. Razi, G. Otiko and P.J. Sadler, Am. Chem. SOC.Symp. Ser., 1983,209,371. 43 N.A. Malik, G. Otiko and P.J. Sadler, J . Inory. Biochem., 1980,12, 317. 44 J. Roberts, J. Xiao, B. Schliesman, D.J. Parsons and C.F. Shaw 111, submitted for publication. 45 J. Christodoulou, P.J. Sadler and A. Tucker, Eur. J . Biochem., 1994,225, 363. 46 O.M. Ni Dhubhghaill, P.J. Sadler and A. Tucker, J . Am. Chem. SOC.,1992,114,1117. 47 M.T. Coffer, C.F. Shaw 111, A.L. Hormann, C.K. Mirabelli and S.T. Crooke, J . Inorg. Biochem., 1987,30,177. 48 A.A. Isab, C.F. Shaw I11 and J. Locke, Inorg. Chem., 1988,27,3406. 49 C.F. Shaw 111, A.A. Isab, J.D. Hoeschele, M. Starich, J. Locke, P. Schulteis and J. Xiao, J . Am. Chem. Sac., 1994,116,2254. 50 M. Starich and C.F. Shaw 111, unpublished observations. 51 A.A. Isab, A.L. Hormann, M.T. Coffer and C.F. Shaw 111, J . Am. Chem. SOC.,1988,110, 3278. 52 A.A. Isab, D.T. Hill and C.F. Shaw 111, unpublished observations. 53 C.F. Shaw 111, M.T. Coffer, J. Klingbeil and C.K. Mirabelli, J . Am. Chem. SOC.,1988, 110,729. 54 M.J. Stillman, C.F. Shaw I11 and K.T. Suzuki, eds., Metallothioneins: Synthesis, Structure and Properties of Metallothioneins, Phytochelatins and Metal Thiolate Complexes, VCH Publishers, New York, 1992,443 pp. 55 A. Bakka, L. Endresen, A.B.S. Johnsen, P.D. Edminson and H.E. Rugstad, Toxicol. Appl. Pharm., 1981,61,215. 56 L. Endresen, A. Bakka and H.E. Rugstad, Cancer Res., 1983,43,2918. 57 G. Schmitz, D.T. Minkel, D. Gingrich and C.F. Shaw 111, J . Inorg. Biochem., 1980,12, 293. 58 E.M. Mogilnicka and M. Webb, Biochem. Pharmacol., 1983,32,1341. 59 J.E. Laib, C.F. Shaw 111, D.H. Petering, M.K. Eidsness, R.C. Elder and J.S. Garvey, Biochemistry, 1985,24, 1977.

56 60 61 62 63

Gold Complexes with Anti-arthritic, Anti-tumour and Anti-HI V Activity

M.M. Savas, D.H. Petering and C.F. Shaw 111, Inorg. Chem., 1990,29,403. A.A. Muiioz and C.F. Shaw 111, unpublished data. C.F. Shaw I11 and J. Laib, Inorg. Chim. Acta, 1896,123,197. B.P. Monia, T.R. Butt, D.J. Ecker, C.K. Mirabelli and S.T. Crooke, J . Biol. Chem., 1986,261,10957. 64 J. Chaudiere and A.L. Tappel, J . Inorg. Biochem., 1984,20,313. 65 A.L. Tappel, Cur. Topics Cell. Reg., 1984,24,87. 66 J.R. Roberts and C.F. Shaw 111, Biochem. Pharmacol., 1998,55,1. 67 D.W. James, N.W. Ludvigsen, L.G. Cleland and S.C. Milazzo, J . Rheumatol., 1982,9, 532. 68 G.G. Graham, T.M. Haavisto, H.M. Jones and G.D. Champion, Biochem. Pharmacol., 1984,33,1257. 69 G.G. Graham, G.D. Champion and J.B. Ziegler, Metal-Based Drugs, 1994,1,395. 70 R.C. Elder, W.B. Jones, Z. Zhao, J.G. Dorsey and K. Tepperman, Metal-Based Drugs, 1994,1,363. 71 G. Lewis and C.F. Shaw 111, Znorg. Chem., 1986,25,58. 72 G.G. Graham, J.R. Bales, M.C. Grootveld and P.J. Sadler, J . Znorg. Biochem., 1985,25, 163. 73 K. Tepperman, Y. Zhang, P.W. Roy, R. Floyd, Z. Zhao, J.G. Dorsey and R.C. Elder, Metal-Based Drugs, 1994,1,433. 74 C.F. Shaw 111, S. Schraa, E. Gleichmann, Y.P.Grover, L. Dunemann and A. Jagarlamudi, Metal-Based Drugs, 1994,1,351. 75 A. Canumalla, S. Schraa, A.A. Isab, C.F. Shaw 111, E. Gleichmann, L. Dunemann and M. Turfeld. J . Biol.Inorg. Chem., 1998,3, 1. 76 R.C. Elder, K.G. Tepperman, M.K. Eidsness, M.J. Heeg, C.F. Shaw I11 and N.A. Schaeffer, ACS Symposium Ser., 1983,209,385. 77 D. Schuhmann, M. Kubicka-Muranyi, J. Mirtcheva, J. Gunther, P. Kind and E. Gleichmann, J . Immunol., 1990,145,2132. 78 K. Takahashi, P. Griem, C. Goebel, J. Gonzalez and E. Gleichmann, Metal-Based Drugs, 1994,1,483. 79 J. Verwilghen, G.H. Kingsley, L. Gambling and G.S. Panayi, Arthr. Rheum., 1992,35, 1413. 80 B. Beverly and D. Couri, Fed. Proc., 1987,46,854. 81 P. Griem and E. Gleichmann, Curr. Opin. Zmrnunol., 1995,7, 831. 82 P. Griem, E. Gleichmann and C.F. Shaw I11 in Toxicology oJthe Zmrnune System, ed. D.A. Lawrence, Pergamon Press, New York, 1997, pp. 323-338. 83 C. Goebel, M. Kubicka-Muranyi, T. Tonn, J. Gonzalez and E. Gleichmann, Arch. Toxicol., 1995, 64,450. 84 P. Griem, K. Panthel, H. Kalbacher and E. Gleichmann, Eur. J . Irnrnunol., 1996, 26, 279. 85 P. Griem, C. von Vultee, K. Panthel, S.L. Best, P.J. Sadler and C.F. Shaw 111, submitted, 1997. 86 P. Griem, K. Takahashi, H. Kalbacker and E. Gleichmann, J . Immunol., 1995, 1575. 87 I. Haiduc and C. Silvestru, In Vim,1989,3,285. 88 P.J. Sadler, M. Nasr and V.L. Narayanan, in Platinum Coordination Complexes in Cancer Chemotherapy, ed. M.P. Hacker, E.B. Douple and I.H. Krakoff, Martinus Nijhoff Publishing, Boston, 1984, pp. 290-304. 89 C.F. Shaw 111, in Metal Compounds in Cancer Therapy, ed. S.P. Fricker, Chapman and Hall, London, 1994, pp. 46-64. 90 T.M. Simon, D.H. Kunishima, G.J. Vibert and A. Lorber, Cancer, 1979,44,1965.

T.B. L. Okada. 1986. Assoc. Berners-Price. 223.T.M.. G. P. Vibert and A. Sung and S. Matthews.. Collery. D. Mirabelli. Simon. Beery and G. Johnson. M.100. 106 G.J..E. Mirabelli. Chim. G.. 385-389.41.46. S. 1986. Girard and D. Macia. R. Chem. M. T.C. Ziegler. Kunishima. Arizti.69. Champion and J.M. Eitenne. 105 A.33.J.26.T.-M. 1986. Acta.218. Manchester. 1989.M. Faucette. 101 S. C. M. Hill. Proc.S. R. 1993.F. Miyamoto. 97 R. 110. Abstracts 3rd International Conf.W.B. Ni Dhubhghaill. 1979. Anti-Cancer Drug Des. D. Hill.D.5486. J . G. Stocco.32. Jarrett. Jonassen. P. R. D. Johnson. G.L. L. Chem. Marcus and H. 1. 1985.T. Bugelski.L. P. P. G. Gurney. in Metal Ions in Biology and Medicine. F. Vibert and A.J. Lorber.26. R. Faucette. Zimmerman. P.. M. Chem. B. G. UK. D. J .K.C. 2913. Med.F. J.D. 102 O. Rush and W.P. P.A. 95 M. Mirabelli.J. Med. Berners-Price. Chem. Dean.T. J .M.L. Simon. F. S. Whitman.M. Jarrett and P. Garnier-Suillerot. Bull. Johnson and S. McCabe.-Q. D. Sung. C. Amagai.631.C.1. Chem. Cancer Res. Arendt and C. Am. Crooke. A. Mirabelli and P. unpublished data. Colacia. Mattern. Sadler and S. Ichida and Y. 98 R. De Pancorbo. Mirabelli. Crooke. Jensen.. A. Soc. Chem. Kunishima. B. 5. C. R. Okada. Hoke.L. Soc. John Libbey.M. F. Marfait and J.1386. L. 5054.C. Meunier. 110 G. Kuroda. Shaw 111. 14. L.A. R. Eurotext.. 1426. 1989. Blough. Agrawal.H.254. G. Hill. Ye and M.S. Graham. 1990.K.. Mirabelli.R..K. Gutjierrez-Zorrilla and E.F. P. Assoc. M.M.T.-M. pp. 1990. S. C.K.K. Sadler and R. . Mong.K.J. Kuo. J .213. Patterson.J. Cancer Res. Sadler.H. 94 C. J. 1981. Hill. 1990. 46. Bears.T. A. 1990. S. McCabe. 96 C.D. M.E. Faucette. Berners-Price.J. C.K. Pharmacol. Mattern. Inorg. K. G.L. Appl. Sutton. Faucette. Sasaki. 3074..R. Snyder. Crooke. Proc. Cancer Res. Bryan.D. Gold and Silver in Medicine. 104 T.C . C. 108 T.29. Sadler. 1986.M..P.T. Gurney.F. L.62.K. 93 C. Virology.J. Mong. P. ed.K. Porier.M. C. 107 H.192. Schein.Y.F.R. Am. 92 T. Shaw 111.K. Johnson. Cancer Rex.94. D. Moustatih and A. J. Cancer Res.H.M. Toxicol. Dalton Trans.K. 1987..-M. Suppl. p. Garcia-Orad.99. J .R. Johnson.R. C. 1991.F. Frank Shaw I I I 57 91 T. C. Inorg.G.M. Mirabelli. Paris. B. 109 T.K.F.123.B. Faucette. 100 S.S. Girard.91. 1078. McCabe. 103 K.J. B. Girard. Rheumatol.293. Sung. 1978. L. Johnson. Jpn.C.K. Lorber.K. 1986. Med. H.J. D. Crooke.. Sutton. Sung. Hempel and S.K. Infiammopharmacoloyy. 1989. 99 S.

CHAPTER 4

Nitric Oxide in Physiology and Medicine
ANTHONY R. BUTLER' AND PETER RHODES2
'School of Chemistry, St. Andrews University, St. Andrews KY16 9ST, UK 2Department of Pharmacology, Ninewells Hospital, Dundee D D 1 9SY, UK

1 Introduction
That nitric oxide (NO) has a number of important roles in animal physiology is now established beyond doubt. The first to be delineated was that in vasodilation. A key enzyme in arterial muscle relaxation (and hence artery dilation) is guanylate cyclase, and it had been known for some time that, in vitro, N O will activate the enzyme.' In 1987 it was demonstrated, as well, that NO is the endogenous molecule responsible for activation of guanylate ~ y c l a s e .N O~is ~* also generated in activated macrophages (cells that are an important part of the non-specific immune system) and leads to the formation of nitrite and nitrite The presence of the same process in the brain' and peripheral nerves8 was soon established. The natural substrate for N O production was discovered to be arginine, and the process was shown to be effected by the enzyme nitric oxide synthase (NOS).The other product of the reaction is citrulline.' Details of the discovery and confirmation of the arginine-NO pathway in animal

+
H3N,

C
I

,/NH

Hfl,
C
I

p
4NO

+

0,

- (""
NOS
H3N-CH-CO2-

+ I

physiology have been chronicled in a number of reviews" and will not be retold here. Rather we will concentrate on the consequences of these discoveries for the

Anthony R. Butler and Peter Rhodes

59

physiology, pathology, and therapy. More recently has absolute proof been obtained for production of NO by humans rather than other species. This work employed direct gas chromatography-mass spectrometry (GC-MS) detection of NO in the exhalate of normal subjects." Almost all previous attempts had used indirect signatures of NO production. Direct detection of NO is now possible because of the development in recent years of a number of reliable analytical procedures. These will be reviewed before we proceed to more biological topics, but it is first necessary to describe some of the physical characteristics of NO that have made its quantitation, particularly in biological system, so difficult.

2 Physical Properties of NO
NO is a gas at room temperature, with a low solubility in water M). From aqueous solution it is rapidly lost to the headspace. In oxygenated water it is oxidized slowly, at a rate which it is critically concentration-dependent, to nitrite (equations 1-3). At this stage no nitrate is formed.12 However, in oxygenated water and particularly in certain biological media such as blood, nitrite is rapidly oxidized to nitrate.I3

+ 0, + 2N0, NO, + NO + N 2 0 , N 2 0 3 + H,O + 2HN0,
2NO

(1)

(2)
(3)

NO is a radical species, but unlike most other radicals, it is not particularly reactive. However, it does react very rapidly with iron to give well authenticated iron-nitr~syls.'~ Reaction with iron is responsible for the activation of guanylate cyclase; it may also be responsible for the cytotoxicity of NO and it has been incorporated into a number of analytical procedures for the detection and quantitation of NO. NO has high diffusion rates in both aqueous and lipid media, and travels rapidly between cells and across phospholipid membranes.' It has a biological half-life of less than a minute2 and on oxidation forms relatively stable aniThis instability in uiuo reflects reactions with oxygen, superoxide and haemoglobin, and these reactions limit the volume through which N O can travel to exert influence. The low molecular weight of NO may allow rapid movement in aqueous environments. Molecules larger than NO have much lower diffusion coefficients and rates of spread. Models to describe the kinetic and concentration profiles for NO have been developed, based on diffusion models used to describe heat conduction in s01ids.I~ These models show N O concentrations to be a function of the rate of formation, diffusion coefficient, distance from the source and time. Close to a production source, perhaps within a radius of lOpm, concentrations reach a steady state within a few hundred ms of release. This is an order of magnitude faster than the measured biological half-life of NO, so that metabolism of NO in viuo should not significantly affect these concentration gradients
' ' 9

'

60

Nitric Oxide in Physiology and Medicine

established by diffusion. The metabolism of NO is likely to become more significant the greater the distance from the source of production. Within a close range, far more NO will be lost to a given site by diffusion than by metabolism. NO is able to influence cellular function over spherical volumes with an influence that is both predictable and constant. Close to a production source, this influence will depend on the diffusion properties of NO itself. The inverse relationship that exists between the half-life and the concentration of NO is likely to become increasingly important to biological functions the farther NO travels from the source of production. The reaction between oxygen and NO is a third-order process (equations 1-3) and will be very slow at low concentrations of NO. Thus the low levels of NO involved in bioregulation permit survival of the molecule in the presence of oxygen. Unfortunately, some of the same physiochemical characteristics make quantitation of NO difficult. Most studies in vivo and in vitro have used indirect indices of NO production such as breakdown products of N O metabolism, second messengers, or biological events in effector systems. Recent developments in probe technology and mass spectrometry have allowed the direct detection of NO in some situations. Constitutive N O synthases (cNOS) in vascular endothelium and other tissues produce small quantities of NO for continuous maintenance of vascular tone and cellular communication." Inducible N O synthases (iNOS) are induced by immunological stimuli such as interleukins' and lipopolysaccharides, and synthesize relatively large amounts of NO for long periods.lg In general, therefore, it is much more difficult to study constitutive than inducible NO production. The complex fate of NO biological systems helps confound investigation and analysis. The amount of NO required for biological action never exceeds the solubility of N O in water, and so the aqueous chemistry of NO is important. More specifically, it is the chemistry of NO in complex biological media, particularly whole blood, that is more important to understand.

3 Measurement of NO in Biological and Chemical Systems
Direct Measurement
Direct measurements of NO are attractive because they offer the highest specificity. The most direct approaches are detection of electric current produced when NO is oxidized, detection of light produced when NO reacts with ozone, and direct GC-MS detection of NO. By use of the first effect, two probe systems have been developed for detection of NO, and they have been valuable for in vitro work. Probes can be used to determine the exact locality of NO and to follow in real time the kinetics of a system. Judicial positioning of more than one probe in a system can be used to assess direction and speed of travel.I5 The specificity of NO probes depends on their ability to exclude other molecules that may give rise to a signal. Shibuki has developed an electrochemical sensor based on a modified oxygen electrode

Unfortunately. Most versions of this probe are several millimetres in diameter and so will not allow particularly accurate positioning in a cellular system. which is not necessarily representative of systematic constitutive production.Anthony R. these are commonly grown in culture for experimental purposes. It is a very delicate system and requires considerable technical skill to build and operate. Nafion. They would probably be too delicate to place in viuo routinely. the detection limit 5 x 10-7moll-1 and range of linearity 1 x to 3 x 10-4moll-'. The earliest measurement of NO involved bioassays. the second messenger cyclic guanosine monophosphate (cGMP). such as the degree . they are invasive. These are discussed in a later section in relation to investigations of the respiratory system. and they are predominantly influenced by NO in their immediate vicinity and thus represent local NO production.2-1. which eliminates anionic interference from nitrite. and fast response time (10ms) are useful features for detection of NO in microsystems such as single cells. and the current concentration relationship is linear between 1 x and 3 x 10-3moll-'. detection limits of some commercial forms of this probe may be no lower than 5 x moll-'. The Malinski microsensor has been used to demonstrate travel of NO from endothelium to smooth muscle and the release of NO by platelets during aggregation.20 The response time of this sensor is 3-6 seconds. They do not have the specificity of direct measurements. Butler and Peter Rhodes 61 (Clark electrode) consisting of platinum wire as the anode and silver wire as the cathode. but they are more convenient. small diameter (0. These are most commonly used for the investigation of NO in biological systems. They include NO metabolites. interference and delays in signal equilibration are further problems. The high sensitivity. The paucity of applications to have emerged since the enthusiastic introduction of this probe in 1992 testifies to that. Indirect Measurement There is a wide range of indirect indices for NO production. probes do not offer the potential for satisfactory measurement of systemic NO production in uiuo in humans. They may be of greater value to those investigating the chemistry of N O in aqueous media. these would be unsuitable for work in many physiological systems. This electrode has been applied to measurement of N O in the central nervous system in uitro. and effector organ responses such as vessel dilation and platelet function. The detection limits may be as low as 1 x lo-* moll-'. including measurement of N O in rat brain during ischaemia. In complex media. The other direct forms of N O measurement are chemiluminescence and mass spectrometry. This porphyrinic semiconductor is covered with a cation exchanger. These are biological systems able to provide units of response in an effector system. There have also been successful in uiuo applications of the Malinski probe. M a l i n ~ k i has developed a detection system '~ in which NO is oxidized on a polymeric metalloporphyrin.0pm). such as plasma. In general. a wider range than for the Clark electrode.

Bioassay is a qualitative measure and does not provide quantitative evaluations of NO. including superoxide. by experiments using a cascade of strips of endothelium-denuded artery. in biological fluids it is normal to obtain a mixture of nitrite and nitrate from NO. because it is not possible to inhibit selectively most of the molecules that activate guanylate cyclase. These molecules include all naturetic peptides so far identified (atrial. Steps should be taken to assess the specificity of each system. where conversion of NO into nitrite is quantitative and other sources of nitrite apart from NO can easily be controlled. Bioassay studies have recently reduced any controversy over the chemical identity of the EDRF. with NO to give N20. but recent interest in NO as a mediator cell function has greatly increased interest in measurement of these anions. there are many different forms of guanylate cyclase. Analysis of nitrite and nitrate in biological samples has been performed for many reasons. ONOO-. B type.) Peroxynitrite readily isomerises to nitrate (equation 4).22 Since many of the biological parameters regulated by NO. Concentrations of cGMP have therefore been used as indices of NO p r o d ~ c t i o nUnfor. An additional aspect to .~~ tunately. may affect in vivo NO metabolism. the low specificity of cGMP is difficult to improve. NO also will react with ONOO- 4 NO3- (4) oxyhaemoglobin to produce methaemoglobin and nitrate. The specificity of changes in cGMP as an index of NO production may therefore be poor. in an aqueous solution. such as vessel tone. which is protonated at physiological (This important process is discussed in detail later. The endothelium-derived relaxing factor (EDRF) was first observed using a tension device and a ring of rabbit aorta. and hydroxyl radicals.62 Nitric Oxide in Physiology and Medicine of vessel dilation or platelet aggregation. and many molecules activate these enzymes. Much of the early interest was related to nitrosamine formation and the potential of these compounds in cancer causation. carbon monoxide. an intracellular messenger that interacts with a range of receptor proteins to produce its effects. Cyclic G M P concentrations can also be influenced indirectly by substances that inhibit phosphodiesterase. NO stimulates soluble guanylate cyclase to produce.. is faster than is the hydrolysis of NO. gives a mixture of nitrite and nitrate. C type). including inhibition of NO production and inhibition of other factors with similar biological actions. However. It seems that. from guanosine triphosphate.. nitrosothiols.. NO. the reaction of NO. Thus.. the anhydride of nitrous acid. are also controlled by other molecules.21 It was identified as NO in a bioassay system of vascular strips. there are several potential routes for formation of nitrate from NO. as hydrolysis of NO. In contrast with bioassay systems. specificity is a potential problem with bioassay. Many factors. this measurement works well in aqueous solution in the presence of oxygen. increasing cGMP production. Recent studies suggest that NO reacts rapidly with superoxide anion to form peroxynitrite.12 The sole formation of nitrite is surprising. cyclic guanosine monophosphate (cGMP). the oxidation product of NO. As a measure of NO. haemoglobin and hydroxyl radicals.

and because it is present there in much higher concentrations than is nitrite. and defined methods of venesection. Any assay for the contribution of nitrite and nitrate that is to be applied to biological work must be assessed by its ability to detect nanomolar. not simply from nitrate. Nitrate is stable in whole blood and plasma. Moreover. There are specific problems with using nitrite and nitrate concentrations in whole blood to assess NO formation in vivo. particularly reductions in the NO production rate. or at least micromolar. ideally. storage.Anthony R. it unfortunately leads to aromatic production from a wide range of nitrosothiols present in plasma. by far the most valuable for biological work is capillary electrophoresis.I3 Thus the time delay before centrifugation of the whole blood sample influences the plasma nitrite levels subsequently measured.28 A variety of S-nitrosothiols is believed to be present in human plasma. Although this approach has been extensively used. they should be measured simultaneously.27 Part of the controversy over the identity of the endothelium-derived relaxant factor relates to S-nitrosothiols (such as S-nitrosocysteine). However. and may be oxidized to nitrate. which decompose in solution to give N O that is converted into nitrite in the presence of oxygen. acute fluctuations in N O production. nitrite can probably be formed in plasma after sampling. Therefore. and sample handling. Nitrite is not particularly stable in whole blood. A commonly used method26for the assay of biological nitrate is gas chromatography-mass spectrometry (GC-MS). levels in the presence of millimolar chloride and sulfate. Of the wide range of alternative methods. Butler and Peter Rhodes 63 the use of nitrite and nitrate as indices of NO production is that sources of nitrite and nitrate other than N O are difficult to control in vivo. small nitrate contributions from breakdown of less stable NO metabolites are probably not as important as are those to nitrite. such .~~ standardized to achieve maximum reproducibility. The mechanisms by which these compounds are formed and their biological roles have not yet been fully described.29 and endogenous S-nitrosothiols have been proposed for a variety of biological functions.n i t r o s ~ t h i o l s Duration of sample storage prior to analysis should be . It is possible that nitrosothiols are generated in viuo from the active species produced from NO. by reaction of less stable NO metabolites such as S . the half-life of nitrate in uiuo is believed to be several hours. it is important to measure both anions in most circumstances. The ubiquitous nature of these anions calls for specific dietary preparation of subjects before study. a method based on production of nitroaromatics from benzene or related compounds in strong acid. and it may therefore be chronologically insensitive to small. which permits rapid and virtually contamination-free simultaneous analysis of nitrite and nitrate in plasma at physiological and subphysiological levels.

The relevant chemistry of these species has been reviewed.* + 0. They may constitute an elimination mechanism. this remains one of the fundamental questions about the chemistry and biology of NO. greater than the affinities of oxygen or carbon monoxide. Yet. Whether these nitrosothiols function as a means to remove or donate N O is unclear. because its metabolic fate is not known. Clearly. It has been suggested that redox forms of NO. may have physiological roles that are conflated with those of NO. The complex formed between NO and haem iron (HbNO) has well-defined electron paramagnetic resonance (EPR)spectroscopic proper tie^. because if it did the oxygen-carrying capacity of the blood would be rapidly compromised. despite constant production of N O in the body that does not happen with the NO-haemoglobin complex." and analytical techniques devised for the detection of N O as an atmospheric pollutant have been applied. with sufficient exposure. However. and this problem has also contributed to the slow pace of their assessment in biological systems. to death. there could be important ramifications for the variety of indices of NO production currently in use among biologists.31 4 Nitric Oxide and the Respiratory System Work with NO and the respiratory system provides the only direct evidence of N O production in vivo in humans. Most indices are measured in moles per unit volume. When human subjects are exposed to carbon monoxide.^' The specificity of the HbNO assay is good because the characteristic three-line hyperfine pattern produced is not seen when haemoglobin binds nitrite and other nitrogencontaining compounds. Lowmolecular-weight thiols and S-nitrosated derivatives are difficult to analyse. Increases in HbNO have been shown using EPR after inhalation of N O and in mouse and rat models of sepsis. there are mechanisms to prevent this potentially fatal complex from accumulating. N O has very high affinity for haemoglobin. the formation of a complex between carbon monoxide and haemoglobin can give rise to respiratory failure and. they may form a reservoir that can be used to provide NO. and changes reflect not only production but also metabolism.-+No. sequestering and inactivating NO.64 Nitric Oxide in Physiology and Medicine as peroxynitrite. NO + o. it is not clear in which circumstances HbNO may provide a good index of NO production. Indeed. For these reasons it is not possible to predict whether any of the nitrosothiols are likely to provide reliable indicators of N O production. A number of animal models have applied HbNO as an index of N O production. It is certainly important that NO haemoglobin does not accumulate in the circulation. the nitrosonium ion N O + and the nitroxide ion NO-. The matter will be considered in more detail later in this review. and one would therefore expect HbNO to accumulate in the body. NO2* -+ NO. Reaction between N O and ozone leads to the generation of light (equations 5 and 6). If this were so. an important consideration when assessing the meaning of changes in an index like HbNO. + light (51 (6) .

Therefore. Both lower and upper respiratory tracts release NO into e ~ p i r a t e .33 This method has been used to detect changes in NO production that occur during exercise.' 1-32 The technique is highly sensitive (detection limit 1ppb) and linear over a wide range. It is often treated with vasodilators to reduce pulmonary artery press- ' . Direct GC-MS analysis of NO in vivo is complex. ~ ~ Mass spectrometry has been used to confirm the chemiluminescence analyses of NO in breath. There is good reason to believe that systematic NO production rises during exercise. (m/z 30. Breathing with normal to low tidal volumes (> 1 litre) and normal to high frequency (> 8 rpm) does not give rise to a stable end expiratory concentration of NO. present in parts per thousand) and NO (m/z 29. assessment of NO in the expirate of subjects with chronic dyspnoea or neutromuscular disorders would be difficult by these means. 1 to lOOOppb can easily be covered without recalibration of equipment. as well as reduction in PO. are known to elevate NO release from the endothelium. ~ ~ . so exhaled NO levels can be determined in a single breath.Anthony R. temperamental. it seems likely that the measurement of NO in breath will continue to be a useful tool for investigation of systematic NO generation. present in parts per billion). Whether or not useful representation of systematic NO production can be obtained from breath analyses has still to be determined. The implications of these findings are not clear.99799. using the nitrosation of thioproline and analysis of the nitrosothioproline derivative by GC-MSO3' GC-MS confirmation of the presence of NO in the breath in humans" demands differentiation between "N. Further work is still required to establish the exact origins of NO contributing to exhalation profiles recorded with this technology. This can be demonstrated indirectly after activation by oxidation. The response time of chemiluminescence equipment is rapid (< 0. Asthmatics exhale greater than normal amounts of NO and express inducible NO synthase in lung tissue.5 s).. The principal mode of action is stimulation of soluble guanylate cyclase with consequent rises in intracellular cyclic GMP. At present. '* The measurement of NO in breath by chemiluminescence has been used to access exhaled NO in asthma and primary pulmonary hypertension (PPH). regular breathing pattern with prolonged exhalation time (> 5 seconds) is necessary to create the NO plateau required for standardized measurements. since a slow (4 rpm). and increases in blood flow and sheer stress.40 ARDS is a severe and often fatal condition associated with pulmonary oedema and reduced right ventricular function. Butler and Peter Rhodes 65 A variety of chemiluminescence analysers is now available for quantitation of NO. Inhalation of NO gives selective reduction in pulmonary vascular resistance in pulmonary hypertension3' and in adult respiratory distress syndrome (ARDS).38 NO is able to cause vascular and bronchial smooth muscle relaxation. It is likely that raised values reflect the activation of inducible NOS in the airways as part of a generalized inflammatory response. and such equipment can be adapted readily for routine measurement of endogenous NO excreted in expired air. The physiology cardiovascular changes that occur in exercise are profound.The technique is not suitable for assessment of patients with poor control over their respiratory cycle. Analyses are rapid and highly reproducible.00022. and not practical for routine use.

It therefore appears that production of NO by the vascular endothelium is of great functional importance. There are a number of physiological stimuli for N O release by the endothelium that help to regulate blood flow under normal physiological conditions." When such an analogue is infused into the forearms of normal subjects. The magnitude of the response seen after inhibition of NO synthesis is particularly striking when compared with the minor changes of blood flow that can be achieved by manipulation of many other vasoactive hormones in humans. This work4' demonstrates the continuous release of N O in the forearm arterial bed and its major role in determining basal blood flow. from one of a rigid system of conducting pipes predominantly acted upon by vasoconstrictors. The risks involved relating to NO. The full potential of inhaled NO for treatment of pulmonary hypertension and ARDS will be realized only after full clinical trials. so ventilation-perfusion mismatching does not occur.66 N t i Oxide in Physiology and Medicine irc ure. formation and methaemoglobinaemia appear real. Not only does inhaled NO give superior pulmonary vasodilation as compared with systematically administered vasodilators. profound vasoconstriction occurs. confirming enantiomeric specificity. resulting in as much as a 50% fall in basal blood flow. An additional disadvantage of this approach is that systematically infused vasodilators dilate the entire pulmonary vascular bed. Theoretical calculations suggest that without reflex constrictor changes. reducing systemic arterial pressure. These include the flow and amplitude of wave forms created by the passage of blood through the vessel and the partial pressure of oxygen. We now know that in the testing state the vascular system is under a profound. inhibition of N O production in the vasculature would result in a rise in total peripheral resistance sufficient to double the mean arterial blood pressure. is avoided.. but NO.42 Profound changes in blood flow occur during exercise. and leads to recovery of damaged tissue. reaches only parts of the lung still functioning. with blood directed towards active muscle groups and away from large viscera such as the gastrointestinal tract. However. prolonged administration of NO followed by withdrawal gives sustained improved lung function. to one of a highly flexible tree under constant regulation by vasoconstrictor and vasodilator forces. increasing pulmonary venous mixing and reducing arterial PO. NO . constant vasodilator influence. as well as modifications in blood flow patterns with appropriate stimuli. The vasoconstricting effect in the human forearm achieved with LNMMA cannot be reproduced with its enantiomer but can be reversed with L-arginine. as a gas. augmenting blood flow to non-ventilated or poorly ventilated lung regions. and the risk of reduced PO. but unfortunately general systemic vasodilation also occurs. This is an exciting development in the treatment of lung disease. The enzymes responsible for NO production can be inhibited competitively by analogues of L-arginine such as N-monomethyl-L-arginine (LNMMA). suggesting improved blood flow and oxygen exchange. 5 Nitric Oxide and the Cardiovascular System Our perception of the cardiovascular system has changed over recent decades.

Abnormal N O production may occur in hypercholesterolaemia and may be related to subsequent development of atherosclerosis. This protection disappears. It has been the mainstay of anti-anginal therapy ever since. Such a therapeutic approach would be exciting. There are important differences. or at least an amplifying factor. may be of particular benefit in the presence of coronary atherosclerosis. hypercholesterolaemia. particularly as women have greater longevity than men and suffer less ischaemic heart disease. Septic shock is characterized by progressive reduction in systemic blood pressure and resistance to exogenously administered vasoconstrictors. and with catecholamines that cause peripheral vasoconstriction and increased cardiac contractility. particularly in the pharmaceutical industry.45 There have been suggestions that nitroglycerin can substitute for defective NO production in atherosclerosis. allowing constitutive physiological NO production to continue to fulfil its normal function. and other organic nitrates such as isosorbide dintrate and isosorbide mononitrate. A great deal of research effort is currently under way. The potential therapeutic benefit of inhibitors of NOS has recently been demonstrated in a small number of patients with septic Larger trials will be required to establish the benefits and the correct administration protocol for such patients. metabolized to N O in the vasculature. It. and other conditions associated with endothelial dysfunction. endothelium-derived NO.Anthony R. after the menopause. Pre-menopausal women suffer less cardiovascular disease than men do. yet only in recent years has the chemical basis of this therapy become clear. nitroglycerin has only minor effects on coronary microvessels less . while preventing excessive release of NO from pathological expressed iNOS. poor tissue perfusion. and vascular leakage. There is also some recent involving chronic exercise of animals that suggests that increased expression of NOS occurs in vascular endothelium as a response to the repeated stimulus of exercise. During the last century a role for nitroglycerin in the treatment of acute attacks of cardiac angina was suggested. Butler and Peter Rhodes 67 is likely to take part in such profound redirection of blood. The mortality of this condition remains high. but as yet there is no direct evidence to show that NO generation is different between men and women. Nitroglycerin is an N O donor. between nitroglycerin and endogenous.however. Animal work suggests that oestrogens dilate blood vessels by an endothelium-dependent mechanism. Whereas endogenous NO and a number of other NO-donor drugs dilate all classes of coronary microvessels. There is now a body of experimental data to suggest that an abnormality in NO production or function may be causal. to pursue specific inhibitors of inducible NOS. There is considerable evidence to suggest that the site of metabolic conversion into NO is the vascular smooth muscle. This is an area of considerable interest. The basis of current therapy is the maintenance of tissue perfusion pressure with plasma expanders that raise colloidal osmotic pressure in the intravascular space. It is now believed that excessive NO production occurs in septic shock and contributes to pathological changes. in both hypercholesterolaemia and atherosclerosis. and hormone replacement therapy in post-menopausal women reduces cardiovascular mortality. however.

They increase precolateral perfusion pressure. and the effectiveness of nitrate therapy is markedly reduced during treatment regimes maintained throughout a 24 hour period. . Nitrate tolerance is an important clinical problem. NO-donor drugs that produce uniform dilation of all sizes of coronary microvessels will not be effective and may well cause coronary steal. against which any new NO-donor drug must compete. In contrast. which spontaneously releases NO and does not show significant cross-tolerance with organic nitrates. dilation of large coronary arteries improving conductivity and blood supply. One of the major disadvantages of the organic nitrates as therapeutic agents is the development of t o l e r a n ~ e . Strategies for avoiding tolerance are usually based on a nitrate-free interval.68 Nitric Oxide in Physiology and Medicine than 100pm in diameter. In addition. However. even SIN-1 induces some tolerance. These differential effects are the most challenging feature of organic nitrates. Nitrovasodilators reduce cardiac preload and wall tension and thus myocardial oxygen consumption. It is quite possible that chemists will be able to develop compounds that mimic the pattern of action nitroglycerin in the coronary microcirculation. These reflex neurohormonal changes reduce the dynamic effect of nitrate-induced vasodilation. worsening regional ischaemia. The generalized vasodilation caused by many NO-donor drugs is not wanted in most clinical situations. Impaired nitrate metabolism within the vasculature is an important aspect of organic nitrate tolerance.~~ such as sodium nitroprusside and nitrosothiols. and the absence of effects on vessels that would unfavourably reduce perfusion pressure. metabolic step leading to nitrate ~~ ~~ tolerance is believed to precede the final formation of N O in uivo. The . it is likely that another feature of impaired metabolism is neurohormonal activation. . dilation of the venous bed giving preload reduction. Although preliminary studies4* suggest that diuretic therapy or enzyme inhibitors may reduce tolerance. The favourable anti-ischaemic effect of nitrates is based on this unique pattern of vascular relaxation. The differential vasomotor responses result in a unique pattern of therapeutic uses and relevant drug action. New NO-donor drugs for anti-anginal therapy must therefore be evaluated for their effects on the coronary microcirculation. Studies of the coronary microcirculation both in vitro and in viuo suggest that the smaller coronary vessels are incapable of converting nitroglycerin into its vasodilator m e t a b ~ l i t e Other NO-donor compounds. augmenting oxygen delivery to ischaemic sections. Tolerance to nitroglycerin and isosorbide dinitrate can be compensated for by administration of a different NO-donor such as SIN-1. further clinical studies are required to consolidate the information. The combination of these two aspects would reduce vasodilation and induce sodium ion and water retention. and these compounds may be useful in the treatment of regional myocardial ischaemia. Delivery of N O spontaneously from NO-donor drugs and continuous stimulation of endogenous N O production by increased blood flow both lead to the desensitization that is witnessed during continuous nitrate administration. when applied at high concentrations for several days. are not dependent on the biotransformation needed for release of NO from nitroglycerin and are potent vasodilators of all microvessels.

These agents are also fundamental to the treatment of acute left ventricular failure. They also have potential roles in the treatment of hypercholesterolaemia and atherosclerosis. regulation of blood flow.Anthony R. the pain caused by myocardial ischaemia.50 The clearest role for the N O has been established in the peripheral autonomic nervous system. and a major trigger for its release in the nervous system is the activation of glutamate receptors.^' For the CNS. where N O is released from non-adrenergic. There are alternatives to organic nitrates as NO-donor drugs. most interest has revolved around the three main functions of synaptic plasticity.16 The availability of agents to supply exogenous NO and to inhibit NOS has proved essential for investigation of these effects. including adaptive dilation of the stomach following ingestion of Histochemical studies of . These diffusion properties are highly characteristic of N O and could not be achieved by nitrosothiols or other contenders for the NANC nerve transmitter. the organic nitrates present formidable therapeutic competition because of their specific regional metabolism and ideal ‘design’ for the treatment of cardiac ischaemia. In most patients it can be adequately managed with a rest period from the therapy at night. The tolerance developed to organic nitrate therapy may be an area amenable to new product development. The chemical identity of the transmitter has been much debated. and those of NO itself. and neurodegeneration. N O seems to mediate many forms of relaxation. but convincing demonstration that N O fills this function has now emerged. but the problem should not be overestimated. However. Nitric Oxide and the Nervous System It has been known for over two decades that synaptic excitation in the central nervous system (CNS) is accompanied by an increase in cGMP. and much interest in developing new NO-donor drugs exists. non-cholinergic (NANC) nerves and mediates relaxation of smooth muscle in many tissue^. The most direct demonstrations of NO production in the nervous system to date have come from use of the Shibuki probe. In the gastrointestinal tract. It has been claimed that they reduce uterine contractions in premature labour49 but confirmation is lacking. where they reduce preload on the heart and dilate the venous system. allowing improved cardiac output. upon stimulation of NANC nerves in canine ileocolonic function. there are other potential benefits to be had from NO-donor drugs in treatment of myocardial ischaemia and infarction. most recently by comparison between diffusion properties of the NANC nerve transmitter. l7 Superfusion bioassay has shown release of a vasorelaxant factor with properties similar to those of NO. Butler and Peter Rhodes 69 Although the main role of NO-donor drugs has been in treatment of cardiac angina. This phenomenon is now known to be caused by NO. Peripheral Nervous System Inhibitory NANC nerves are recognized as important components of autonomic innervation to many organs.

in the gastrointestinal tract. Part of this phenomenon involves retrograde messengers that return from a stimulated neurone to relay information to the preceding neurone. high concentrations being found in major pelvic ganglia. causing outflow obstruction. In the intestine. Involvement of the NO-arginine pathway in penile erection may explain the aphrodisiac affects of amyl nitrite. In impotent diabetic men. NO is an attractive candidate for a retrograde messenger mediating use-dependent . ~The high functional NOS activity found in the membranous urethra ~. N O also contributes to NANC-induced vasodilation and relaxation of guinea pig and human tracheal muscle. cholinergic and peptidergic nerves. the membranous urethra. as in the cardiovascular system.53Therefore. Flaccidity is enhanced by contraction of corporal tissue after activation of adrenergic receptors by noradreline released from sympathetic nerve terminals. and bladder neck. there is an impairment of both neurogenic and endothelium-mediated relaxation in penile corporal smooth muscle. It is involved in cerebral functions such as memory and in complex automated motor functions. but this process is not mediated simply by passive engorgement with blood. such as playing a piano concerto fluently.5 4 Other work has indicated dysfunction of the N O system in disorders associated with male i mp~t e nce. there seems to be a constant NO-dependent dilator tone crucial for function of these organs.56 Immunohistochemical staining of both rat and canine tissue has shown that NOS is widely distributed throughout the urogenital tract. active filling of the sinusoids compresses the peripheral venules of the corpora against the rigid tunica albuginea. Synaptic Plasticity Synaptic plasticity is a process by which neuronal connections may be reenforced or altered. as well as in the penis i t ~ e l f . and it is proving to be as important as those in adrenergic.58 Clearly there is a widespread system of nerves throughout the body using N O as a transmitter.’~ During erection. Immunohistochemical evidence of NO-containing nerves has been found in penile tissue. from a clinical perspective. muscle relaxation involved in peristalsis is also mediated by NO. The penile flaccidity found in diabetes may be a direct consequence of impaired N O synthesis.~’ and its involvement in relaxation of the bladder neck further suggests that NO may be important in the regulation of micturition and urine continence. this provides a rationale for the treatment of certain types of impotence by intracavernosal administration of vasodilators.70 Nitric Oxide in Physiology and Medicine biopsies from infants with hypertrophic plyoric stenosis and adults with achalasia suggest that these conditions are caused by a lack of NOS in pyloric and gastro-oesophageal tissue. Conversely. it is possible that inhibition of NOS will be beneficial in the treatment of priapism. Functional studies with isolated human and animal corpus cavernosum tissue have shown that relaxation induced by electrical stimulation of autonomic nerves is abolished by blockade of NOS. whereas responses to exogenous donors of NO are preserved. NO is responsible for relaxation of the blood vessels and smooth muscle of the male corpus cavernosum and thus development of penile erection.

Neurodegeneration The possibility that N O can contribute to neurodegeneration in the CNS has been raised. because development of tolerance can be inhibited by NMDA antagonists. activated . its dependence on afferent fibre activity.16y60j61 Tolerance to morphine is a further form of synaptic plasticity in which NO may be involved. That would cause the vasospasm and the consequent cerebral ischaemia responsible for much of the damage that can occur in the recovery phase following intracranial haemorrhage. Another example of synaptic plasticity in hyperalgesia. and arterial p r e ~ s u r e .O within the brain parenchyma may also be a factor that links the neuronal activity to local increases in cerebral blood flow.NOS-containing fibres originating in one of the ~~. and its blockade by (NMDA) antogonists. as well as dilator responses to stimuli appear to be fundamental mechanisms. A range of evidence suggests a role for N O in the plastic changes giving rise to the hyperalgesia. but glutamate acting at NMDA receptors is probably involved. releases red blood cells from their intravascular compartment. or indeed any form of intracranial haemorrhage. It is believed to act within cardiovascular centres in the CNS to regulate sympathetic outflow.24 In cultures. a region of the brain involved in memory and learning. Cerebral Circulation The possibility that neuronally-derived N O may be important in the regulation of cerebral circulation has received much attention. a state of enhanced sensitivity to painful stimuli. These mechanisms may all be significant. particularly its endurance. It has recently been shown62that NOS inhibitors given with morphine completely prevent tolerance without affecting the actions of morphine itself. The mechanism of hyperalgesia shares several features with hypocampal long-term potentiation. vascular tone. The mechanisms of morphine tolerance are poorly understood.Anthony R. and vasoconstriction in the region of the bleed. Subarachnoid haemorrhage.66 One of the most puzzling and damaging consequences of subarachnoid haemorrhage is vapospasm. but the role of N O in this phenomenon remains controversial and uncertain. but further experimental evidence is required to understand the basic mechanisms involved. because it is a potentially toxic molecule and is able to inactivate key enzymes either directly or by peroxynitrite formation.~~ term potentiation in the hypocampus.~~ cranial autonomic ganglia innovate major cerebral muscles and provide neurogenic dilator control over the cerebral c i r ~ u l a t i o nN~ ~generated by neurones . and participation of N O in the maintenance of resting cerebrovascular tone. suggesting a selective effect on the tolerance mechanism. It is possible that this could lead to sequestering of NO by free haemoglobin. Butler and Peter Rhodes 71 changes in synaptic transmission by acting at pre-synaptic sites following postsynaptic p r o d ~ c t i o nThe most studied example of synaptic plasticity is long.

careful examination of the chemistry involved reveals difficulties. it has been proposed7' that the redox forms of N O (i. As has already been mentioned in ~' the Introduction.69This is an exciting area where the chemical synthesis of selected inhibitors of neuronal NOS is greatly needed. NO' cannot exist in aqueous solution. can produce enough NO to kill cocultured cerebral neurone~. such as tumor cells. + H' HNOeH' +NO- (7) (8) + HNO + N. Therefore. to avoid the potentially detrimental action of blocking vital NO-dependent mechanisms. So far. In in vivo models of focal cerebral ischaemia. In view of the apparently conflicting roles of N O in the brain (it can be both neurotoxic and neuroprotective). as there is very rapid hydrolysis to nitrous acid (equation 7). it is possible to see how nitrosothiols are formed in vivo.is formed by ionization of HNO (equation 8).72 Nitric Oxide in Physiology and Medicine microglias. within a cell this probably means to another thiol. The system produces a number of reactive oxygen intermediates (ROIs) in response to inflammatory ~ t i m u l i . The nitroxide ion NO. NO' and NO-) play parts in its activity. was then shown to be dependent upon a r g i ~ ~ i n eNO must . Demonstrations then followed that showed that cultured macrophages. Although that presents an attractive solution to the paradox of N O behaviour. The ability of macrophages to kill abnormal cells. such as cerebral blood vessel dilation and platelet de-aggregation. these inhibitors can also increase damage. one of the main cell groups involved in the immune protective system also produce This process is dependent upon the presence of arginix~e. The first indication of this came when it was noted72 that high levels of nitrate were found in subjects suffering from gastroenteritis. but there is reaction in the presence of an oxidizing agent.~~ These observations led to the conclusion that the arginine-NO pathway occurs in activated macrophages. N O ' could be said to interact with physiologically significant thiols. nitrosothiols can transfer NO' to other nucleophilic sites. N O itself does not react with thiols to give nitrosothiols. coli lipopolysaccharide also leads to high nitrate levels. NO' HNO + H 2 0 + HNO.O + H 2 0 (9) 7 Nitric Oxide and the Immune System The immune system provides a range of defence mechanisms that protect against foreign proteins and invasion of the intracellular environment by microorganisms. Thus. Treatment of rats with E.68However. However. NOS inhibitors have been found to give marked protection. macrophage-like cells in the CNS. rescuing up to 70% of cortical neurones in some studies.~' it is therefore possible that N O production and in vivo could lead to neuronal damage. it now appears that NO is also involved in the physiology of the immune system. there is not direct evidence for the involvement of NOd in physiological processes. but the physiological importance of transnitrosation to another thiol has not been delineated. but there is competing dimerization of HNO to give nitrous oxide (equation 9).'~ .e.

The question now engaging the attention of scientists in a number of disciplines is the nature of the cytotoxic action of NO. When tested against Leishmania major. from SNP or RBS. Of course. it has been proposed that NO produced by macrophages reacts with superoxide to give the peroxynitrite ion (equation This ion is a relatively strong base77 and is protonated at physiological pH (equation 11).82 The exact significance of this finding is at present unclear. Peroxynitrous acid then decomposes to give nitrogen dioxide and hydroxyl radicals. O as NO+. + HO* (1 1) (12) Two studies of the antibacterial action of NO-donor drugs suggest that. inorganic molecule. which are certainly highly reactive (equation 12). S-nitro-N-acetylpencillamine (SNAP) was found to be an effective toxic agent.78 However. In this role they must undergo chemical85 or enzymatics6 reactions to release NO. not all the experimental evidence supports this view. it was initially suggested that this might be sufficient. and the former has been used in hypotensive surgery. diatomic. the effect of which was enhanced by addition of superoxide dismutase (SOD). Butler and Peter Rhodes 73 therefore be a natural cytotoxic agent for the defence of the body against foreign agents and against abnormal cells within the body. rather than the formation of hydroxyl radicals.80A similar conclusion was reached in a study of the effect of peroxynitrite on the aggregation of washed platelets. an aqueous solution of NO is not a particularly powerful antibacterial agent. under the circumstances of these studies. and RBS is the most effective of a range of NO-donor drugs against E . NO is not particularly reactive. In contrast.81 ONOO- + H + + ONOOH ONOOH -+NO. Sodium nitroprusside (SNP) contains NO as a ligand where it is formally present as N O f . coli.75 As macrophage activity also produces ROIs. dioxygen is also a simple. and this would depend upon the environment in which it finds itself. . peroxynitrite was not particularly It may be that the reaction of peroxynitrite with cellular targets. The key may be the formation. it is neither NO nor hydroxyl radicals that is responsible for their toxic action. In view of the complexity of most physiological processes it is refreshing to note that a simple diatomic inorganic molecule has such a significant role in so much physiological chemistry. of nitrosothiols that can both transfer NO' in a heterolytic process and give NO in a homolytic one.Anthony R. it is not without precedent. but it may be that the crucial element in the toxic effect of SNP and RBS is transfer of NO+ to a physiologically significant thiol. SNP is an effective toxic agent against Clostridium s p i r o g e n e ~N~ ~ is also present in Roussin's black salt (RBS) . Both SNPs3 and RBS84 are hypotensive agents. As N O is a radical. When tested against Clostridium.which removes any superoxide present. but unlike most radicals. is responsible for its biological activity.

1987.M. Byrns and G. Proc. Rex. Sci.43. Francis. 1993. Spagnuolo. D. Michel.10043. Sci. I. 62. 1991. 17 J. 3 L.. Exp. Natl. Iyengar. Osborne. Beckman.368. H. Pharmacol. Res. M. A.S. Acad. L. Nature. 87. 1991. Rinelli. Neurosciences. Knowles. J.J.J.J. 87. 1980. Shibuki. 1988.L. Biophys. Busse and A.358. Kubrina. Neurosci. Taintor.169. 7 R. Chaudhuri. R. D. R.327. M. Tends Neurosci.S. Nature.G.A. R.W. K. Radomski. 23. Med. T. 1’71. 6 D.1099. Jpn. E. 87. Biochim.33.A.S.. References 1 S. Rachlin.. Nathan. Zawadzki.S.883. Ferrige and S. England. Proc. 5 J.M. D. Natl. Natl. te Poel. Neuropharrnacoloyy. Zamora. FEBS Lett.A. T.F. Natl. R.S. A. 10 S.S. Chiacone and F. W. T. R. Res.Jaraki.201.8706. Moncada. Chem. Mullins. T. Saruta and R. 25 J.L. Palmer. Esumi. Stuehr. Marshall and B.S. Gross. Caldwell.F. Taha. ROE.S.87.B.23790. Commun. 4 M. 15 T. 1992.W.V. Buga. Kato. 11 A. Palacios. Moncada.41. Leaf and J. Wood. Biochern.J. Acad. 16 J. USA. 1994. Higgs.N.I.J. U S A . Moncada. Nakaki.M.163. Vanin.9265.69. 9 A. Palmer. Moncada.109.G. Moncada.M. Chen.. Wood and J. Wishnok. Z. Ignarro. Coletta.P. Freeman.. P.84. J. Suzuki. N. 18 R. R. Biochim. 20 K.M. 1994.266. Simon.M. 1992. H.275. S. 1011. M. 24 J. Malinski and Z.V. but principally we thank Professor Salvador Moncada and his collaborators at The Wellcome Research Laboratories.. Deussen and S. C.373. Proc.74 N t i Oxide in Physiology and Medicine irc Acknowledgements Our grateful thanks go to many people who have assisted us in exploring this fascinating area on the borders of chemistry and physiology. Vavrin and E.1235. Rev.A. Moncada. Malenkova and A. Ascoli. 8 J. 21 R.157. R. J . Biophys.I. Heart J . P A . M.M.33.J. Ohshima.G. 1987. R. Reu.86.911. Palmer and S. 0. I. Acad. H. Commun. Garthwaite.39. Persson. Gustafsson. ..676. Lev1 and C.. Murad. Biol. 23 K. Pharmacol. Nature. Leone. P.C.D. 1987. Darbyshire. Palmer and S . Hibbs.E.M. 1990. Sci. Proc. J.F. 1990. Acta. 14?60. P. 2 R. who have been responsible for so much of what has been described. Biochern. Leone. U S A . G. Moncada.5. Wiklund and S. 22 M.. Furchgott and J. 12 D.5159. Biochemistry. Res.27. R. J. R. Sakuma.524. 19 M. 1989. 1992. 14 L. 13 C. 1990. Feelisch. Singe1 .233. Acta.1620. Ashton and S. Garthwaite. Stamler. Nims. Miilsch.9.A. Biophys. Beckman. Palmer and E. Sci. M. USA. P. Francis. 59. 1994. 1991. Mordvintcev. Wink. M. 1989. J .W. 1993. Nature.E. Acad.G. P. Waldman and F.. Garthwaite.6.288.J. Ford. Marletta. Yoon.M. 1987.F. Chem. Biophys.E. Saavedra and P. Tsuda. 1988. 1990.R. Knowles. Hishikawa. Toxicol.A.J.

Acta Physiol. J.1210. Frostell.S.H. Scand. Clin.37. Chem.89. Severs. T. Newby and A. Godard.126. Circulation.338.M. Am.Anthony R. Hibbs and Higgs. J. 37 I.308. Abrams. I.709. Baert and J. Moncada. Shibuki. Zanzinger. 2038.8. Wain. Vol. McAninly and D. Seyedi. H. T. Persson.H. S. Gustaffson. Hintze.E. p.E. Acad. Nature.69.444.R. H..H. Chem. 1991. Boeckxstaens. Lowenstein. Bennett and P.28. Vanhoutte. 38 Q. Chem. Williams. 1976. Sci. Perkin Trans.743. K. Campbell. J .L. Collier and S. Leone.. Seki and P. Barbier.M. 1991. 55 N. 27 A. 1986.M. Suburo. Chaud.M.64B.E. Redington.C. Proc.M. J .200.779.W. Pelckmans. 88. V. Dalzeil. Brown. 49 C. J . 1991. Franchi.441. Westfall and K. Fratacci. Vallance.H. Lancet. P. 1990. Riveros-Moreno. 1973.R. R. H. Acad.141. Respir.S. J . 2. Lees. 293. D. Gustafsson. G. Circulation Res.H.951. Lewis. Cardiol. Wallwork. 41 P. Commun. 1510. Barnett. A. P.C. 106. 1989. Science. Burnett. Busse and E.23B. Lancet. 1992. 42 V. 54 A. McCann and M. 57 A.W. Gustaffson. 29 J. J.1173. 1991. . Soc. 539.M. Ward. in The Biology ofNitric Oxide. U S A . Henderson. Romero and P. J. Horowitz.. D. D. 1990. J . Askew. ed. Hutcheson and T. Cardiol.C. 112. Chromatoyr. Persson.345. F. 1993. 18. E.A. Herman. Wiklund and L.91. 47 P. D. Snyder. Rhodes and S.148. Wang and T.9568. 1994. Lancet. H257. Holgate and J.E.346. 52 R. Physiol. 1992. V.S. D.C. Rubanyi. ed. Exptl... Res. A. Trends in Pharmacol.H.W. 32 T. C. K. Moncada and J. Sanders.D. Dis. 1991. 1. Ther..184. 1994.261. Bredt. 35 L.G. Natl. 1991. M.. F. S. Biochem. 34 M..91. 1991.M. Res. Ramsay. Andersson.F. Nature. 28 S. 50 K. 44 A. 33 M. Moncada.A.349. Kim. Lefebvre. ComInun. Hibbs and Higgs. in The Biology ofNitric Oxide. Chang and S.G.C. Biophys. de Tejada. U S A .401. A. K. D. H1145. Ozawa.. Am. Am. J . P. D. Invest.M. 1992. A.G. Bouquet. 1992.741. Zapol. Wiklund and S.. S. S. Soc. Am. Physiol. E. P. 31 A. Martin. Jordaens.250. Natl. Griffith. Higenbottam. Polak. A. Moncada.D. Hypertension. Hoshino.1325. Loscalzo. Stone and J. Vallance. 43 W.9.S. Springall. Azadzoi. Petros.P. Am. Neurosci.E. Agrenius. Loscalzo. 1992. Moncada.Zetterstrom.A.-E. J . N. Bassenge and J. W. Sci. 16. Ihre and L.L. R.343.D. 51 H.C.J. E.. Seea.M. Sci. Chanez. 45 J. Griffith. Goldstein and I. D.338. 26 J. Am. 146. Butler and Peter Rhodes 75 and J. Lancet. M. N.M.M. Van Maercke and A. 1994. V. 39 J.257.M. Dinh-Xuan. Jones and W... Williams.T. Res. Persson. 30 M. Tesch. J . Bult. 1993. Flitney and D. Rettori. K. 48 J. M. Butler. P.181. 1992. Johnson. 74. P. Edwards.J. 997.P. M.. Rev. Hedlund and K. Leone.645. 1986. 1995.70. B. Lancet. Gibb. J. Pepke-Zaba.H.852. 343.. Jauniax. p. Vol.K. Hamid. Y. Med. Gimeno. Needleman and E. Biophys. Holmquist. Am. J. 1993. Stamler and J. P. Marletta. 1992. 46 E.1557.1995. Howarth.64.83. 295. Marletta. Moncada. 0. Pritchard. 1. Biochem.H.G.R. 40 C. 53 S.M. J . Phol.M.. Bassenge.R. Lancet. 1994.J. Thronbury.L.M. Pharmacol. J . Francis. Anal.. R. 1984. 1991. 36 G. Ford. Rehg and R.L. 56 F.342.. Proc.115. 1994.

R.R. Iadecola. Skipper. Butler and L.J. Butler.R.399. Brit. Sucher. Maves. 1992. Marshall and B.47. K. C. W. Beckman. J. Freeman. Reivich. Exptl. Kitto. N. S. Duval. V.343. Flitney. Sakuma. Krane and R.. O’Donnell..G. 63 L.J. Acad. Epperlein. P. K.50. Taintor.S. Toxicol.262. 71 B. 65 A. 1991. Reis. Scatton. Koppenol. Cohen.B. Poignet and B. P. Moncada. Moncada. 834.24. 1991. I. 80 W. J . Sci.S.S. Wilcox.576.Y. Stamler. Gebhart and T. Azadzoi.250.C. USA. 1992. Zamora-Pino. 85 A. Chen.. M. 1992. Glogowski. Blough and O. 62 Y. Ther. Pharmacol.916. Hughes and R.A.L. A.. J .S.3502. Res. Brain Res.W. F.3913. 67 K.T. Inorg. 66 C. F E M S Microbiol.91.89. J . Kowaluk. 74 J. Feelisch. C. Megson. Lett. Cui.16.M. P. Cereb. F. D. Postgrad.A.. Anal Biochem. 60 K..H. de Tejada. Assruey.Z. Haley and G. Read. Seth and H. Zafiriou. Ischiropoulos and J.12.. Lett.262.. 1992.A..J. H. Meller. Lei. 1992. Sci. Greenberg and M. Moro. 183. 76 N. 84 F. Yamamoto.W.. C. Beckman.76 Nitric Oxide in Physiology and Medicine 58 I.672. J . Hibbs. D. 86 E. 69 S . Gross and R. Loscalzo.132. 1991.6369. Rev.S. Bush and B. Eur.G. 82 A. I.F. Pobegailo. Pharmacol. Eur.126.R. V.-L. 64 H. Med. 98. Med. Charles and R.G. Y. Marletta. Freeman. Z.364. 75 X.-B.4244. Ysuda.A.V. Z. 1992. Halliwell.204. 78 J. Kolensnikov. Wishnok and S. 1992. 1989. J. 1987. S.A. S. Togashi.717.M. Noronha-Dutra. Renton. 77 R.V.. 59 J. 1994. Singe1 and J. 842. Green.. N .M. Chen. Nature. J .A. Z. Proc.S.L.L. Liew and S . Radomski and S .F.C.B. Proc.. Chess-Williams. Kitabatake. 67. Pasternak. G...A. J . Nature. Neurosci. 70 S.1620. J . Kobayashi. Glidewell.J. 1988. Pryor.626. H. Levi. Nowicki. J . Pechman. .Q. Shapoval.N. Natl. J.84. Goldstein.K.N. 1994. P. Lett. Beckman. Wagner. Berger and D.-H. Csaki.339.R. Proc. Choi. M.148. 1993. Beckman. D. 1. J. 81 M. U S A .7.320. Vavrin and R.V.A.L. Chem.P. Natl.. C. 68 J. J . Golanov.131. C. 1987. Natl. J .S. Soc. Biol. Neurosci. Moreno.. Proc. 5. A. J.-S. Natl. 1994.W. Cunha. Lancet. Benyo. Pick and G.S.. H. Eur.58. Neuroscience.A. Butler and C. Arora. Acad. Sci. Acad. J. J . U S A . J. Kovach.221. Stuehr and M. 1992. Lipton. 1982. E. 1974.M. Cammack. S.361. Verner. 61 S.266. U S A .385.24. Acad.A. Sci. 1990. Saito. Iyengar. Radi. J.A. J. Pharmacol. M. 1987. 1992. R. Goodwin..J.J.107. Chem. M. Boje and P.A. Yoshioka.721. Exptl. Chem. Pharmacol.H. 79 J.A. 1992.R. Flitney and A.B. M. Physiol. Tannenbaum.W.J.E. unpublished observations.336.550. Chem. I.F. D. Fung. 1992. H.E.449. Pharmacol. Pan. H. Garthwaite. 1992.6702. Immunol. D.50. N. Blood Flow Metab.. Ther.1025.344. Immunol.R. 72 L. T. 83 J.138. 73 R. Joannou.87.S. Szabo.L. Darley-Usmar. D. Sagach and L. Eny. 1985. T.

. -+Mn(m) + H. The role of the enzyme is to regulate the lifetime of superoxide and protect against superoxide-mediated cytotoxicity.~ have passed since its discovery. amyotrophic lateral s c l e r o ~ i s ) . and radiation-induced i n j ~ r y . USA 1 Introduction Since the discovery of the functional activity of the enzyme superoxide dismutase (SOD) by McCord and Fridovich. encouraging results in clinical trials have been obtained with polyethylene glycol (PEG)-conjugated Cu. Although almost 30 years ~. (1) . 800 N.Zn SOD for the treatment of severe head injuries4 and with recombinant human SOD for the prevention of multiple organ failure after multiple t r a ~ m a .CHAPTER 5 Therapeutic Aspects of Manganese(ZZ)-based Superoxide Dismutase Mimics RANDY H. inflammatory bowel disease. organ transplantation.' intensive efforts have been made to develop the enzyme as a therapeutic agent for the treatment of a wide range of diseases and disorders. WEISS AND DENNIS P. superoxide undergoes a slow self-dismutation process (equation 3). Lindbergh Blvd. the results of clinical trials to date with the SOD enzyme for the treatment of myocardial ischemia-reperfusion injury have been disappointing. ~ However. the only clinically approved use for SOD has been in Europe for the treatment of osteoarthritis.6-8 Another surge of interest in SOD developed with the discovery that there is a defective gene that codes for SOD in one inherited form of Lou Gehrig's disease (ALS. ~ ~ ' ~ The enzyme superoxide dismutase is an oxidoreductase that catalyses the dismutation of superoxide to hydrogen peroxide and molecular oxygen (equations l and 2). osteoarthritis. St. rheumatoid arthritis. bronchopulmonary dysplasia." In the absence of the enzyme.O.. post-traumatic and post-ischemic neuropathies. Mn(r1) + HO. Recently. Louis. including reperfusion injury to the ischemic myocardium. RILEY Monsanto Company. MO 63167.

78 Therapeutic Aspects of Manganese(r1)-basedSuperoxide Dimutase Mimics (3) The lack of success in the development of the SOD enzyme as a therapeutic agent can be attributed to several factors: (1)a narrow efficaciousdose characterized by a bell-shaped dose-response curve.'. We felt that many of the limiting factors could be overcome by the development of synthetic.). low-molecularweight mimetics of the SOD enzyme and by use of a direct method to measure the SOD activity of these complexes. the formation of these deleterious oxidants can be inhibited without consumption of the SOD mimic. By catalyzing the dismutation of superoxide. This mechanism is in contrast to the mechanism of action of a general .. before cytotoxic superoxide-derived oxidants. O. * G H. (2) lack of an accurate and routine method to monitor and quantitate SOD activity.7.13-pentaazacyclopentadecane 2 Results and Discussion Therapeutic Strategy of Superoxide Dismutase Mimics The basis of the therapeutic strategy utilizing SOD mimics (see Figure 2) is to more efficiently catalyze the dismutation of superoxide to generate the less innocuous non-radical compounds. Superoxide reacts in a near diffusioncontrolled manner with nitric oxide to produce the potent oxidant peroxynitrite. pharmacology and cost. Figure 1 Structural class of SOD mimics based upon the Mn(rr) dichloro complex of 1.-* + H O . + 0. which can abstract hydrogen atoms from unsaturated fatty acids to initiate lipid per~xidation. ( 5 )lack of oral bioavailability. Under ischemic conditions where the pH of tissue can decrease. stability. the formation of the perhydroxyl radical (HO. hydrogen peroxide and molecular oxygen. We will discuss how the Mn(r1)-based SOD mimics were designed and illustrate the advantages of these SOD mimics over the SOD enzymes with regard to their SOD activity. (6) immunogenicity.'~ occur to a will greater extent.1O.O. (4)instability of the enzyme. (3) inability of the enzyme to gain access to the intracellular space.4. and (7) cost. particularly for the treatment of myocardial ischemia-reperfusion injury. such as peroxynitrite and the perhydroxyl radical can be produced. Our goal has been to develop the Mn(r1)-based SOD mimics shown in Figure 1 as therapeutic agents for diseases mediated by superoxide.

we developed methodology to measure the catalytic dismutation of superoxide by stopped-flow kinetic By this technique. we directly monitor the decay of superoxide spectrophotometrically in the presence or absence of a putative SOD mimic at a given pH. At least a tenfold excess of superoxide over the putative SOD mimic is used in the stopped-flow assay. Moreover. see equations 1 and 2). these assays. which typically rely on a spectrophotometric change of a redox indicator to measure superoxide levels. the indirect assays are prone to false positives or false negatives respectively. see equation 3). when the putative SOD mimic oxidizes or reduces the spectrophotometric indicator. cannot kinetically distinguish between a catalytic dismutation of superoxide and a stoichiometric interaction of superoxide with the putative SOD mimic. to eliminate contributions due to a stoichiometric reaction of the complex with superoxide. Stopped Flow Kinetic Analysis: A Direct Assay for Superoxide Dismutase Activity Numerous indirect assays. ' ~ Design of Manganese-based Superoxide Dismutase Mimics Manganese-based complexes are particularly attractive as potential SOD mimics because they have a reduced capacity to react with the dismutation . have been used in attempts to measure the SOD activity of putative SOD However. The SOD mimic catalyzes the dismutation of superoxide. ' ~ . such as the cytochrome c assay. Kinetic analysis of this decay can determine whether the complex is a SOD mimic (decay of superoxide becomes first-order in superoxide and firstorder in complex. which would be consumed by reacting stoichiornetrically with the oxidant or with a radical derived from the oxidant. We recognized the need for methodology to measure SOD activity directly that would be more accessible to the bench-top scientist than is the method of pulse radiolysis. thereby preventing the formation of the cytotoxic oxidants. Consequently. A catalytic rate constant (kcat)for the dismutation of superoxide by the complex can be determined from the observed rate constants of superoxide decay as a function of catalyst c ~ n c e n t r a t i o n . another direct measure.Randy H . Riley 79 H+ HO+/H+ H w Perhydroxyl Radical Cytotoxicity Figure 2 Therapeutic strategy for the SOD mimics. Weiss and Dennis P. peroxynitrite and the perhydroxyl radical antioxidant. or is inactive (decay of superoxide remains second-order for its self-dismutation.

I8 cyclam. as shown by conductivity measurements.13-pentaazacyclopentadecane (Figure 1. only the Mn(I1) complex of 1.80 Therapeutic Aspects of Manganese(@-based Superoxide Dimutase Mimics product hydrogen peroxide to generate cytotoxic hydroxyl radicals (Fenton chemistry). The high oxidative stability of these complexes is significant in that there are likely to be very few biological species with sufficient oxidizing power to convert the Mn(1r) complex into the Mn(II1) complex. d5 complexes. although oxidants such as the perhydroxyl radical (HO. ~ ~ As will be discussed later.288-quinolin01. type of heteroatoms (nitrogen replaced by oxygen). ~ ' .at physiological pH. Mn(I1)-basedcomplexes with stereochemically-defined substituents can be prepared so that such substituents enhance both the SOD activity and stability of these complexes.13 x lo7 M-'s.10.28The crystal structure of this complex (Figure 3) indicates that the five nitrogen atoms are coordinated to the Mn(I1) center. and also have the desirable characteristic of high stability due to the enhanced kinetic stabilizing effect of the m a c r o ~ y c l e . R = H) had significant SOD activity.7V us.4. NHE). with a catalytic rate constant (kcat) of 4. reaction of the complex with superoxide.and iron-based complexes and their aquated ions are known to carry out such undesirable ~ h e m i s t r y . and in aqueous solution are univalent cations.. ' ~Early work by many -~~ researchers identified a number of manganese-based complexes that had apparent SOD activity by indirect However. The Mn(I1)-based SOD mimics are high-spin. we observed no SOD activity with aquo Mn(I1) and manganese complexes of desferal. The Mn(I1) complexes of this pentaaza crown family are white crystalline solids synthesized by the addition of the macrocyclic ligand to a methanolic solution of anhydrous MnC1." It is absolutely critical that the assay used for measuring SOD activity of putative SOD mimics be able to distinguish between a catalytic dismutation of superoxide and a stoichiometricreaction with superoxide. Additionally.~~ ~ a l e n . We proposed that a Mn(I1) complex of a macrocyclic polyamine could have SOD activity. not catalytic.. and largely in the plane of the Mn with the chloride ions coordinated in the trans-axial positions. .7.28 We have developed a number of versatile. and degree of unsaturation within the ring. we ~~ synthesized over twenty different Mn(I1)-based macrocyclic amines with macrocycles differing in size (9-21 membered rings). ~ ' apparent 'SOD activity' of or The the above manganese complexes is due to a stoichiometric. asymmetric syntheses of highly functionalized polyazamacrocycles via the reduction of cyclic peptide precursors and via a bis(ch1oroacetamide) a p p r ~ a c h .) and the hydroxyl radical (HO-)are indeed competent for this oxidation. it is apparent that such an assay must also be capable of rapidly and reliably providing quantitative rate data so that useful structure-activity relationships can be established for the optimization/design of such synthetic catalysts.To this end. The SOD activity of each complex was assessed by stopped-flow kinetic analysis. Of these complexes. number of donor nitrogen atoms (3-7 nitrogens). when we assessed the SOD activity of these complexes by the direct method of stopped-flow kinetic analysis.28 The Mn(n) complexes have high oxidation potentials ( > 0. whereas copper.

Zn SOD. immunogenicity effects. selective reactivity for superoxide.M11-CI2 Mn-NI Mn-N2 Mn-Na Mn-N. permits the complexes to gain access to . and cost.2(4) CllMnN2 97. lower than SOD enzyme by a factor of almost 100.0(4) CIlMllNd 92. Molecular Weight. stability.7(4) CIlMnNs 86.40( 1) 2.6(6) 67.2(4) CIIMIIN~ 78.57 l(4) 2. including membrane permeability. The low molecular weight of the SOD mimics..Randy H.5(5) Figure 3 Crystal structure of the SOD mimic SC-52608 Advantages of Superoxide Dismutase Mimics over Superoxide Dismutase Enzymes The Mn(I1)-based SOD mimics have numerous potential advantages over the SOD enzymes as potential therapeutic agents." ~ ~ 6 % ~ @ .27(2) CifiCr-----i7g3'(Z)--NlMnN5 NlMnN2 N2MnN3 79. Table 1 compares characteristics of the Mn(I1)-based SOD mimics with R = H versus those of bovine erythrocyte Cu.6(6) N3MnN4 7 1..26( 1) 2. Weiss and Dennis P . Riley 81 B--" --"' Selected Bond Lenglhs ------A ond Length (A) T Pe Y ------------------A n g le ( 0 ) Selected Bond Angles(0) --M n-~ ~ .1(6) 71.2(6) CllMnNl 84.41(1 2A5( 11 2.4 Mn-Ns 2.9(6) N4MnN5 74.

. therefore. The SOD mimics catalyze the decomposition of superoxide only and do not react with hydrogen peroxide or with peroxynitrite. the apoenzyme form of the Cu. whereas the enzymes are isolated from natural sources or prepared by recombinant DNA techniques. it is not orally bioavailable. OONO-. non-immunogenicity Unstable Stability Isolation cost the intracellular space of certain cell types.ZnSOD enzyme Advantage of SOD mimic 32 500 In tracellular O. The SOD enzyme structure is peptide-based and is thus readily degraded by proteases. The selective reactivity of the SOD mimics for superoxide makes these complexes useful probes to establish the role of superoxide in biological mechanisms. whereas the SOD mimics.~ inactivating the enzyme. Reactivity. do not have that problem. Structure. The Mn(I1)-based SOD mimics are prepared by total synthesis..-. Consequently. such as EDTA. The Cu.82 Therapeutic Aspects of Manganese(zr)-based Superoxide Dimutase Mimics ~~ ~~ Table 1 Advantages of Mn(ll)-based SOD mimics over SOD enzymes Characteristic Molecular weight Reactivity Structure Kinetic stability with EDTA Source 341 O. Low toxicity Peptide Oral bioavailability. Kinetic Stability.Zn SOD enzyme can readily be prepared by dialyzing the enzyme with EDTA. the SOD mimics have high kinetic stability (discussed below) and.O.36i37 Source.Zn SOD enzyme reacts with both hydrogen peroxide34 and with per~xynitrite. which are not peptide-based. we have observed that the conformation of substituents on the macrocyclic backbone of the . for example. the cost of manufacturing the SOD mimics would be potentially much less than that for manufacturing the enzymes. It should also be possible to develop an orally effective Mn(I1)-based SOD mimic. H. Conformational Control of Superoxide Dismutase Activity and Stability From extensive studies of the structure-activity relationships. However.only Non-peptidic Stable Synthesis Bovine erythrocyte Cu. are stable in the presence of the transition metal ion chelators.33 whereas the cells are generally impermeable to the enzyme. Immunogenicity problems are also a concern with the non-human-derived SOD enzymes. Due to the stabilizing effect of the macrocycle.

Randy H . two rate-determining steps are implied: a step that is dependent on the concentration of protons. shows in each case that the superoxide decay rates are first-order in both [Mn(L)] and superoxide concentrations.R)-trans-cyclohexano substituted SC-54417.R)-S. We have identified the trans-2. the rate of the pH-independent path in D 2 0is affected indicating that water exchange on the Mn(I1) center is involved. The near diffusion controlled rate for subsequent electron transfer by this path and the proton coupled electron transfer of the outer-sphere path dictate that the structure of the oxidized and reduced forms of the active catalytic species be similar in structure.substituents arranged in a favorable stereochemistry so that the ligand is forced by steric repulsive forces to fold about the spherically symmetrical Mn(I1) ion. Figure 4 shows the SOD activities and kinetic stabilities (defined by the second-order rate constant kdiss (first-order in [H'] and first-order in [Mn(L)] complex) of the unsubstituted SOD mimic SC-52608. then the . SC-52608.3-(R.4. nearly diffusion controlled for the pH-dependent path and near 1 x 1 0 + 7 s. Results of mechanistic studies and kinetic studies indicate that the rate-determining step in the catalytic cycle involves oxidation of the Mn(1r) to Mn(II1). are significant.'.R)-bis(tran~-cyclohexano) substituted SC-55858.37 and the 2. although the kinetics are less than first-order in proton concentration. Wis and Dennis P. and one that is independent of pH. This mechanistic interpretation is also consistent with deuteration results in which an isotope effect of nearly 6 is observed for the proton-dependent step.. and SC-55858.39 By analysis of the kinetics of superoxide decay. Consequently. Additionally. The proton-dependent pathway is consistent with the oxidation of the Mn(I1)-based complex by the perhydroxyl radical via an outer-sphere mechanism. the rate of superoxide decay was studied at different pH values. the mechanism by which these Mn(I1) complexes catalyze the dismutation of superoxide has been The kinetics of superoxide decay for all three complexes.g-(R. SC-54417. Further. The proton-independent pathway is consistent with the oxidation of the Mn(I1) complex by an inner-sphere pathway involving binding of the superoxide anion to a vacant coordination site on Mn(I1). Thus.e. i. with SC-55858 (kcat = 1.. and we observed that the kcat value increases in a linear fashion as the proton concentration increases. We have proposed that this can be accomplished by folding of the ligand to generate a pseudo-octahedral six-coordinate geometry about the high-spin d 5 Mn(I1)ion. similar to the rate for the inner-sphere process measured for these complexes.39 The k. Riley es 83 Mn(I1)-based SOD mimics is critical in determining the complex's SOD activity and stability. The magnitude of the relative contribution of each pathway is different with each Mn(I1)complex.3-cyclohexano group as a substitu~~ ~~ ent that increases both the SOD a ~ t i v i t y and -kinetic stability of the Mn complex. = 4.' for the pH-independent path. The exchange rate of water on Mn(I1) generating a vacant coordination site is on the order of 10' s .20 x lo8 M-' s-') having three times more SOD activity that SC-52608 (k.') at pH 7. if the ligand has the proper substitution pattern.3-(R. the 2..13 x lo7 M-' s... The magnitude of the rate constants measured for these two pathways. values for the dismutation of superoxide by these complexes increase as additional trans-cyclohexano groups are added.

and the kdissvalues are measured by spectrophotometrically monitoring the exchange of Cu(l1) the Mn(11)in the for complex . possess catalytic activity.. values are measured by and The stopped-flow kinetic analysis. the corresponding Mn(I1) complex is over 200 times more thermodynamically stable than the unsubstituted SC-52608. first-order in [H'] and first-order in [Mn(L)] SC-52608 SC-54417 sc-55858 km 7. k. with log K values greater than 10. Consistent with this is that complexes derived from ligands in this series which modeling shows are rigidly planar do not. SC-55858. by a factor of almost 10.1 Increasing SOD Activity and Stability Figure 4 Catalytic rate constants (kcat) the dismutation of superoxide and the kinetic for stabilities (kdissthe dissociative rate constant) of the Mn(ri)-based SOD mimics SC-52608. The thermodynamic stabilities (logK) of the Mn(I1)complexes were assessed by potentiometric titration.84 Therapeutic Aspects of Manganese(@-based Superoxide Dimutase Mimics ligand will force a geometry about the Mn(I1) complex which can have high catalytic SOD activity. The thermodynamic stability of the complexes can be substantially increased by the addition of the trans-cyclohexano substituent. As can be seen from Figure 4. in fact.8. the Mn(I1) complexes are very thermodynamically stable. with two trans-cyclohexano substituents.09x lo7M" sec" 1. The mechanism of ligand loss for these Mn(1r) complexes has been studied3' and is a second-order process.4 4. Substituents on the macrocyclic ligand exert major effects on the stability as well as on the catalytic activity. SC-54417.For SC-55858.~'For example.20 x lo9M. 08 1. 16 13. the substituents on the ligand governs the electron transfer process by affecting the degree of folding of the macrocyclic ligand on Mn(I1). the mono-substituted SC-54417 is more thermodynamically stable than the unsubstituted SC-52608.' sec-I 281 M' sec" 4 1375 M-' sec-' 33 M-' sec" log K 1.13x lo7M ' sec-' pH &is¶ 9. Thus.. which is believed to increase the preorganizational rigidity of the Mn(I1)-based complexes similarly to that observed with gem-dimethyl substituents on macrocyclic tetrathioether complexes of Ni(11).

In general we observe that increasing the number of substituents increases the kinetic and thermodynamic stability of the resultant Mn(I1) complexes. These results are consistent with human neutrophil-mediated injury to aortic endothelial cells being mediated by superoxide. such as EDTA. a potential advantage of the Mn(I1)-based SOD mimics over the SOD enzymes is the ability of the SOD mimics to permeate the cell membrane. Bovine erythrocyte Cu. designated as kdiss. but zero-order in added metal ions or added chelating ligands.Zn SOD protected against the superoxide-mediated injury to the endothelial cells. did not protect the cells against the superoxide-mediated injury.37941 second-order rate The constant.Randy H . However. The rate of exchange is pH-dependent and independent of the CU(II) concentration. where L = ligand). SC-54385. Mn(L)2+ Cu2+ + Mn2+ + C U ( L ) ~ + (4) The kinetic stability of the mono-trans-cyclohexano substituent SC-54417 is twice that of the unsubstituted SC-52608. but the results were quite variable. even at concentrations as high as 300 pM. Inhibition of Neutrophil-Mediated Human Aortic Endothelial Cell Injury by Superoxide Dismutase Mimics We have reported the observation that the Mn(I1)-based SOD mimics inhibit superoxide-mediated injury to human aortic endothelial cells3* The amount of injury to human aortic endothelial cells was assessed by pre-labeling the cells with [51Cr]chromate. Weiss and Dennis P. The amount of injury to cells was directly correlated to the amount of superoxide the cells were exposed to. also decreases with increasing numbers of transcyclohexano groups reflecting the increased kinetic stability. This permits a facile spectrophotometric kinetic assay in which the kinetic stability of the complexes is measured by the exchange rate of Cu(rr) for Mn(I1)in the complex (equation 4. Lipophilicity Clearly. Riley 85 complex. the SOD mimics SC-52608 and SC-54417 showed reproducible dose-dependent protection against the injury. the lipophilicities of mono-substituted Mn(I1)-based complexes can be varied . which is released into the medium when the cells are injured. and two trans-cyclohexano groups increase the kinetic stability of the Mn complex SC-55858 to over 85 times that of the unsubstituted complex SC-52608. Superoxide was produced either by activated human neutrophils or by xanthine/xanthine oxidase. and a bellshaped dose-response curve was observed. As shown in Table 2.33The lipophilicity of the complexes can be controlled by placing the appropriate substituents onto the carbon backbone of the macrocycle. with nearly complete inhibition being obtained with 150pM complex. The increased kinetic stability of the Mn complexes by the addition of trans-cyclohexano groups is believed to be due to the greater preorganizational rigidity of the complexes. a Mn(1r) complex that has no detectable SOD activity.

with one cyclohexano group.6 .4 .5 .18 - + P is the ratio of the partitioning of the complex between n-octanol and aqueous Hepes buffer.4.1.).30 0.87 .The unsubstituted complex (R = H in Table 2) has a logP o . in Table 2) has a positive log P of 0.2. For example.1.02 + 0. which also increases the SOD activity and the kinetic and thermodynamic stability of the complexes.CH. thereby catalyzing the dismutation of intracellular superoxide. has a logP of . a complex can be prepared that is either more or less lipophilic than the unsubstituted complex. the lipophilicities of the SOD mimics are expected to be critical in obtaining optimal protection of the intracellular space against super oxide-mediated cyt ot oxicit y.2. has a logP of .0.0.18.2. which has two trans-cyclohexano substituents. Of particular interest is the ability to increase the lipophilicity of the complexes with the trans-cyclohexano substituent.9.9 .38 These results are consistent with the SOD mimics gaining access to the intracellular space of the cells.9. which the SOD enzyme cannot gain access to. . over a factor of 1000.0. endothelial cell injury assay. In the neutrophil-mediated. pH 7. the 4aminobutyl derivative is less lipophilic (more negative log P) than is the unsubstituted complex.(CH.1.63.86 Therapeutic Aspects of Manganese@/)-basedSuperoxide Dimutase Mimics Table 2 Lipophilicities of Mn(Ii)-based SOD mimics R log P - 1-Aminobutyl Hydro xy met hyl H Methyl 2-Propynyl I sobu t yl Benzyl Alkenyl Phenyl Cyclohexyl Cyclohexylmethy1 1-Naphthyl n-Octadecyl 4.2. The n-octadecyl derivative [R = CH.3 .1 3.5 .as assessed by log P (P = the ratio of the partitioning of the complex between n-octanol and aqueous Hepes buffer.1. indicating that the complex is slightly more soluble in the n-octanol phase than in the aqueous phase.1. whereas SC-54417. respectively). indicating that the f complex is about 1000 times more soluble in aqueous bufer than in the n-octanol phase.. Therefore. the SOD mimics protected against the injury to a greater extent than the SOD enzyme did under conditions where the same amounts of SOD activity of the SOD mimics and of the enzyme were used.3 . SC-55858. Depending on the substituent.2 .

46have examined the protective effect of the SOD mimic SC-52608 in an in vivo dog model of myocardial ischemia-reperfusion injury. Intravenous administration of the SOD mimic SC-52608 at a total dose of 16mgkg-' significantly reduced the myocardial infarct size observed in untreated dogs.44 however. Protection against Myocardial Ischemia-Reperfusion Injury In Vivo The SOD enzyme has been reported to protect against myocardial reperfusion to ischemic tissue. SC-54385. which was attenuated when the hearts were perfused with 20pM SC-52608. they illustrate the potential usefulness of the Mn(I1)-based SOD mimics in protecting against such injuries in viuo.Rundy H . The results indicate that the SOD mimics provide significant protection against myocardial ischemia/reperfusion injury. . there are also many reports in the literature that indicate the enzyme does not offer p r ~ t e c t i o nIn those cases where protection . and an 18 hour period of reperfusion followed. different doses of the enzyme. Riley 87 Inhibition of Myocardial Ischemia-Reperfusion Injury in the Isolated Heart The protective effects of the Mn(I1)-based SOD mimics against myocardial ischemia/reperfusion injury in the isolated rabbit heart and monkey heart have been investigated. Perfusion of 48 pM SC-52608 attenuated the increase in end-diastolic pressure and the increase in creatine kinase release following reperfusion. Perfusion of SC-52608 also inhibited the release of creatine kinase and intracellular potassium. an increase in left ventricular enddiastolic pressure occurred. at the same dose failed to protect against the myocardial injury.~~ has been observed and dose responses have been examined.^^'^^ Black et al. Upon reperfusion. variable periods of ischemia and reperfusion and short half-lives of the SOD enzyme^. These studies indicated that SC-52608 is cardioprotective to the reperfused. These results are consistent with superoxide mediating the myocardial ischemia-reperfusion injury. We have investigated the protective effects of the Mn(II)-based SOD mimics in a feline model of myocardial ischemia-reperfusion injury. isolated rabbit hearts to a 30 minute period of global ischemia followed by a 45 minute period of reperfusion. and reduced the extent of antibody binding to the intracellular protein myosin which occurs upon reperfusion of the ischemic heart. a Mn(I1) complex with no SOD activity. Weiss and Dennis P. ischemic isolated rabbit heart. the SOD enzyme shows a bell-shaped dose-response curve. The discrepancy among reports of the effectiveness of the SOD enzyme may be due to a narrow bell-shaped dose-response curve. The isolated hearts were subjected to 35 minutes of global ischemia followed by 45 minutes of reperfusion. Kilgore et subjected Langendorff perfused. The left circumflex coronary artery of the anesthetized dog was occluded for 90 minutes. However. have also demonstrated a protective effect of SC-52608 from Friedrichs et ischemia-reperfusion injury in the isolated primate heart.47 In this in vivo model.

Intravenous administration of SC-52608 into conscious rats resulted in a transient. as the mimic does not exhibit a bell-shaped dose-response curve.48 The aortic ring relaxation was endothelium-dependent and inhibitable by a nitric oxide synthase inhibitor. with an effective dose being observed at 5pmolkg-' intravenously. In the feline model. The Mn(I1)-based SOD mimic SC-55858 also protected against the reperfusion injury in the feline model at an intravenous dose of 5pmolkg-'.However. and has been proposed to be a mediator of i n f l a m m a t i ~ n . which indicates that the protection observed with the Mn(I1)-based SOD mimics is due to the intact complex. Anti-Inflammatory Activities of Manganese@)-based Superoxide Mimics Superoxide is a product of activated polymorphonuclear leukocytes. At that time the animals were sacrificed and the degree of necrosis and the area at risk in the myocardium were determined.'2 Superoxide reacts with nitric oxide in a diffusion-controlled manner to produce peroxynitrite. significant protection by SC-54417 was observed against the reperfusion injury. SOD mimics would be expected to increase nitric oxide levels. By catalyzing the dismutation of superoxide. which causes the observed relaxation of the aortic rings and the decrease in blood pressure.'. A Mn(I1)-based complex with no SOD activity did not protect at an intravenous dose of 5 pmol kg. Potentiation of Nitric Oxide We evaluated whether the Mn(I1)-based SOD mimics could potentiate the levels of nitric oxide.88 Therapeutic Aspects of Manganese(rr)-based Superoxide Dimutase Mimics the hearts of the animals were made ischemic for a period of 75 minutes.suggesting that superoxide is a mediator of the reperfusion injury to the ischemic myocardial Furthermore.48 SC-52608 induced the relaxation of preconstricted rat aortic rings. which was followed by a reperfusion period of 4.We . direct administration of MnCl. the bovine erythrocyte Cu.5 hours. Further structure-activity relationship studies have been carried out in uivo to assess the role of superoxide in myocardial ischemia-reperfusion injury. did not inhibit the formation of the necrotic tissue in this model. '. such as the neutrophils.Zn SOD enzyme showed a bell-shaped dose-response curve with significant protection against necrosis only at an intravenous dose of 16 mg kg. indicating that there is a difference between the protective effects of the SOD mimic and those of the enzyme. a potent vasorelaxant. the SOD mimic SC-54417 inhibited the necrosis in a dose-dependent manner. Even at a dose as high as 50pmol kg-'. dose-dependent decrease in blood pressure.48 The results are consistent with the SOD mimic SC-52608 potentiating levels of nitric oxide. The SOD mimic SC-52608 enhanced nitric oxide levels (as assessed by cyclic GMP activity) in rat lung fibroblasts in a dose-dependent manner. neutrophil-depend- . ~ ~ ~~ have evaluated the role of superoxide in acetic acid-induced.

for SC-52608 is 10mg kg. Mn(u) complexes were given intra-colonically 30 minutes prior to acetic acid. all of the SOD mimics are anti-inflammatory. Consistent with this hypothesis is the observation that SC-52608 inhibits neutrophil-dependent inflammation induced by the intradermal administration of leukotriene B. and also histologically..14-pentaazacycloheptadecane. SC-52608 and the other SOD mimics attenuate the acetic acid-induced colitis in mice by inhibiting the influx of neutrophils. were tested for antiinflammatory activity in the mouse acetic acid-induced colitis model. After 24 hours. specifically the Mn(I1) dichloro complexes of 1. when administered intracolonically. Colonic myeloperoxidase activity correlates with the degree of tissue inflammation as assessed by visual and histological analysis. ~ ~ As shown by histological analysis. Of particular importance is the observation that Mn(I1) complexes that have no SOD activity. as assessed by tissue myeloperoxidase activity. Tissue samples were assessed biochemically for the neutrophil marker enzyme myeloperoxidase (MPO). which catalyze the dismutation of superoxide as shown by stopped-flow kinetic analysis..7. a neutrophil ~ h e m o a t t r a c t a n t .1 1.10.4. Monosubstituted Mn(1r)-based complexes.'.4.7. A possible mechanism for inhibition of the neutrophil infiltration Table 3 Anti-inflammatory activity of Mn(lI)-based SOD mimics R X YOInhibition MPO activity" (n = 1 69 47 70 48 77 32 H Methyl Isobutyl Phen ylmethyl Cyclohexylmethyl H c 1 c 1 c 1 c 1 c 1 OAc "A solution of 3% acetic acid was instilled intracolonically into mice. not protect against the colonic inflamdo mation induced by acetic acid when the compounds are administered intracolonically at a dose of 30 mg kg. Weiss and Dennis P.13-pentaazacyclohexadecaneand 1. Histological analysis of the colonic tissue confirmed these results.Randy H . . the mice were killed and colonic tissue samples were collected. When administered intravenously at a dose of lomgkg-'. SC-52608 inhibited the acetic acidinduced inflammation by 50%. inhibited the acetic acid-induced inflammation of the colon in a dose-dependent manner (ED. These results are consistent with a role for superoxide as a mediator of neutrophil-dependent inflammation. As can be seen from Table 3.'). Riley 89 ent inflammation of the colon in mice. The Mn(I1)-based SOD mimic SC-52608.'l The inflammation of the colonic tissue was evaluated biochemically for the neutrophil marker enzyme. myeloperoxidase.

Soriano. 9 H. Herzfeldt.. E. Y.A. Halperin. J . A. L. D.R. Hu. H.R. K.L. Mark Currie and Daniela Salvemini of the G. Ichikawa. C. Cardiol. Acknowledgments We wish to thank everyone who has contributed positively to this project. and our external collaborators. Schneider and H. A. J . Van den Bergh. A.~ ~ 3 Summary Mn(I1)-basedcomplexes of the macrocyclic pentaamine 1.A.P. Rep. Trauma. McKennaYasek. Cayabyab. Smyth. D. thereby potentiating nitric oxide levels. Figlewicz.35. P.13-pentaazacyclopentadecane catalyze the dismutation of superoxide and protect against superoxide-mediated tissue injury both in uitro and in uiuo. E.1047. B. Cayabyab. G. N. 8 Y. J. 1994. 17. 10 D. Siddique. V. References J. B. We want to acknowlegde with special thanks the expert molecular pharmacology leadership of Drs. Parge. 6 J. D. Tainer. O’Regan. D. 1987.. Y.A. 1993. T.A. R. D. D. Biihren. B. Roos. R. Goto. T. Searle Co. J. Fridovich. Getzoff. Kereiakes. J.S. Roses.W. Gaston. Warner. R. A. Patterson.M. J . Hentati. Herzfeldt. A. Brinker. D. A. Yamamoto. Donaldson. Tanzi. H. A.W. 1986. Mancini. Life Chem. F.J.D. 1982.67. Nippon Kyobu Geka Gakkai Zasshi. Searle Co.-Y.10.P.. N. A. Hattori and C. 7 T. Petkau.244. Iqbal. Muizelaar. ~ ~ . R. Becker. 1. Mulder.M. Topol and S. Marmarou. J.89. 1993.Trentz. D.G. Z . Sapp. Rothbaum. Michelson. Pericak-Vance and T. Bertram. J. W. our consultants. J . Circulation. including our colleagues at Monsanto Corporate Research and G. R. Murohara.F. Young. Kawazoe. S. Bird. Deitchman. R. Deng. 375. J.90 Therupeu t ic Aspects of M anyanese(rz)bused Superoxide Dimutase Mimics is that the SOD mimics effectively catalyze the dismutation of superoxide.765. 1969. The complexes are also antiinflammatory and potentiate nitric oxide levels by inhibiting production of peroxynitrite. Ohashi. C. Flaherty.4. Biol.C. Y. Smyth. Siddique. R.-X.261. R. P. Gustafson. Hung. Deng. H. M. which prevents the migration of the inflammatory ~ e l l s . A. The Mn(I1)-based SOD mimics are useful as probes to elucidate biological mechanisms that may be mediated by superoxide. Werns.A.R. Heuser.P.. 1991.J. 1993. Gruber. Neurosury. Nakajima and Y. Burwell.E.-X. Kawashima. Shimada. McCord and I. Ahmed.41. Berger.A.A. Hallewell. Marzi. Komai. L.. H. N..13.C.7. Cancer Treat. G. Nitric oxide inhibits the adhesion of neutrophils to endothelial cells. H.78. Hentati. Rev. Wolf. S. Shibata. Kawai.110. D. Rahman. T. A. Hung. Science. A. 5 I.D.M.-Y. Choi.E. J. Deng.6. 1993. Z. P. Am. Kontos. Deng. The SOD mimics are currently undergoing pre-clinical studies as potential therapeutic agents for treating myocardial ischemia-reperfusion injury.W. George. Yui. Schiittler and 0. W. Chem. C. Kosakai. J. Krizus. D..6049. A.427. E. 1 2 3 4 . Rosen.T.

Koppenol. 1994. Schasteen. Weiss.. Am.K. 35 H. Rivers. Neumann. Schasteen.105. 1989. Am. Baker.. Chem.L. Am. A. W. D.L.12-13. 1991.H. Biol. Am. 1989. Free Radicals Biol. J . Malfroy. 26 J.995.P.J. 37 D. J .192. 1991. Beyer Jr. Aston. 31 K. Sparks.H.P. D’Aurora and A. Am. H. 40 J.E. M. 33 M.269. W. Pantoliano. 19 C. S. Bid. Weiss and Dennis P. U.W. Tsai.. C. G. Michel.93.330. Biochern. 1994.H.3687. 21 D. 1991.. Riley. K..L. 20 Y. D. Chem. Sample..J. E. 39 D. Zhu. W. U. R.S. 30 D. J . Znorg. 1984. 24 W.W.268. Ischiropoulos. Znorg. J. Busch and W. J . 1717. 27 M. Rahman. Chem. Goldstein. Czapski.26. Biophys. SOC.W.23049... Tetrahedron Lett. Luo. 569. J . Szulbinski. M.R..1976. Scholler.2987.L. A.L. Neumann. 1993.L. Smith and C. Biochem..35.M. C. Cell. Chem. 1993.G. J .. 1993. 34 S.27. Res. J . J . Free Radicals Res. Hardy.18535. Riley. Rush. Howie and D. Hardy. 14 S. Riley. Bruce. 853.H. Beckman and J.W.P. Chem.431.. Cushing. Weiss and U.289. SOC.M.W.. L. Meyerstein. D. J. D.D.L.112. Chem. Horvitz and R. 104.A.264. Commun. D. Haines.. 11 I. M. Dix. P. 1973.35.D.. Lennon. Rivers. Lucchesi.H. Hickman.98. Gusella. J . 2269. SOC. Methods Enzymol.-J.H.110. 1991. Chem. R. J . SOC.R..S. Biophys. Riley. Weiss.95. Walling.W.P. Graham. Kustin and I. Biochem.H. Weiss. Goosen. Biochem. 17 D.P. Masarwa. 1982. Chem. Brown Jr. Riley.6732. D. Riley and R. S. G. 1992. Eur. Flohe and F. S.A.H. C. 12 J. Biophys. 18 R.2200. 1994. K. Weiss. 23 M. K. Arch. Acc. J. 344. Gellman. Am. Chem.R.6698. A. Weiss. R. Chem. 25 J. 13 J. Riley 91 Laing.. 254. Riley. W.. Ross. Chem. 16 L. Anal. J. 1993. Saran and G..97. 22 C . E. Tetrahedron Lett. K.P. 1996.. Maskos and W.H. R.12. Commun.H.P.. Kilgore. S. Aston. 42 K. H. Chem.268. and I. Pericak-Vance. 1973. Rahman..D.8. Shieh.125. Soriano. Baudry.93. Beckman. Biol. 119. C.M.-S. J . J.Randy H . Stern. Bors.116. Biochem. 1993.. Rivers. Sigman. 36 M. and D. D.J. 1988. Science. Fridovich.. Aiken and T..S. Biochem.M..271.R. Chem.J. 1975. M. A.4293.P. Etienne..387. 1990. Jewett. 6522. J . Chen. Smith and S.S. Ryan.. 1994. H. Modak.P.R. SOC. J. Busch. Weiss. Henke and D.298. Goldstein and G . Flickinger.196. S.R. Desper and S. Hambright. Walling and A. Smith and J. V. Biol. U.7761. Martin. Lennon. .. Mammone and D.2489.S. K. Neumann. 1989. SOC. Ling. W.266.C. Am.149.M. Mol. P.P. Gillespie.295. C. Czapski. Crow.21. Aston. Z. Riley. 28 D. Trans. Chem. Ryan and B.4.23049. R. 1997. J .35.T.S.M..S.362. A. 15 S. 29 R. Aston. 59. Fridovich.S. W. Lennon. H. J. Riley. 180. unpublished results.847. Chem. Otting. 38 M. 41 H. Johnson. Riley.A. 1989. Sawyer. Cardiol. Wegner and L.G. Ryan. Chem. 1988. Espenson.. Biophys. A. L. 5. Hardy. Jacobsen and B.J.H. J . Epstein.H.S. 1993. Cohen.J. Arch. M.P.L. Ryan and D. Sample.H.. Weiss.S.S. Rivers and R. Chem. Znorg. M.R. R. Henke. Biol. Hardy.1994. 15 091. Freidrichs. Flickinger.S. Aston. SOC.5213. D. Arch. SOC. Henke. Biochem. J . M. F. Biol. Weiss and W.F. 32 P. Chem.964. Flickinger. W. Med. 1979.H. Rouleau. P. K.J. K. Rev. Valentine. Palucki. Res. Bakac and J.

J. C. J . 50 M. E x p . U. P. Ooiwa and J. Granger.. Sawyer. Gaboury. D.1208. B. Hoff. Barron. Black..1133.265. Kilgore. Kubes.. ed.P.M. Kubes.W.7. 52 J. 1994.R. A425. C.A. Fretland..B.74. B.A. Moncada and E. Omar. Higgs. 45 J. Free Radicals Res. R. Reinhardt and P. 1995. Niu. 1993.. 1994.270.M.A..703. Tamura. Settle. B i d . S. FASEB J. Grisham. Currie and G.J. M..-F.C. 49 B. J . Lucchesi.2867. . 271. Downey.R. Weiss. R. Palomo..-F. D.208. D. Kasten.S. Lucchesi.C.P. Smith.G. Riley. Med.R. 1993.. Gaboury. in The Biology of Nitric Oxide. Riley. Lucchesi. Chi. McMahon. 44 Y. Blake.P.344.P..A.859. E. Freidrichs. FASEB J.S. Driscoll and B. Freeman. Chem. U. 1994.92 Therapeutic Aspects of Manganese(r.26 149. D. Niu and J. Ryan.P. Kubes. Driscoll Jr. Vol. Schasteen.S. Woodman. Misko. London. FASEB J. Halliwell.1293. K.G. Portland Press.7. Nickols. 53 P. 1993.S.R.H. Weiss.P. R. 12-13. J . 46 S.. Circ.S. R. X. Riley and C. Chi. McCord. Weiss. 3. Riley.S. H. E. 1996. P.A. D. S. D. S. Physiol.R.M.E. 170. Res. Circulation Res. L.P. Therapeut. Biol. 1988. W. SOC. Venturini.H. Proc. 51 R.M. Gallagher and B.B. Exp. Hoult and D. Lancet. Kanwar. H862. 1991. 1994.63. K.P.H. L. M.r)-basedSuperoxide Dimutase Mimics 43 G. 47 C. 48 T.T. Pharmacol. Am.944. Comrnun. 54 X. Schasteen. Smith and P.N. 1988.2.H. Weiss. E. T. Ryan and B. D. M.

' Insulin receptors (IRs) are membrane-spanning tyrosine-specific protein kinases. V6T 123. Insulin is a signaling hormone with numerous regulatory roles. In diabetes.l MARGARET C. University of British Columbia. 2036 Main Mall. glucose uptake into peripheral tissues such as skeletal muscle and fat is impaired. University of British Columbia. Canada 2Faculty of Pharmaceutical Sciences.CHAPTER 6 Vanadium Compounds as Possible Insulin Modifiers CHRIS ORVIG. Canada 1 Introduction Vanadium has been used therapeutically on rare occasions since the turn of the century. a disease state characterized by an absolute or relative lack of insulin. whose function is the mobilization of these molecular forms of stored energy. by binding on the .~-' important aspect of this renewed interest is the potential for An improved clinical efficacy of vanadium at very low concentrations by complexation with appropriate ligands. including uptake of glucose. BC.' KATHERINE H.' Insulin also serves to counteract catabolic hormones. and glucose utilization in the energy-dependent processes within cells is abn0rma1. insulin. V6T 1Z1. amino acids. proteins in muscle. CAM2AND JOHN H. respectively.8-'2 Insulin-mimetics are of interest principally as alternative or adjunct therapeutic agents in diabetes. McNEILL2 'Department of Chemistry. and triglycerides in adipose tissue. glycogen in muscle and liver. there has been a surge of interest in vanadium as an orally effective therapeutic agent due to its insulin-mimetic properties.16 early in the insulin signaling cascade. Vancouver.l~ The normal uptake and metabolism of glucose in nondiabetic individuals is initiated by a series of intracellular reactions known as the insulin signaling cascade. 2146 East Mall. which have been demonstrated in v ~ v o . Vancouver BC.' More recently. Insulin is not orally active. Several recent volumes provide an overview ~-~ of current research interests and findings in this rapidly expanding field of in~estigation. and/or by insulin resistance. THOMPSON. and fatty acids for storage as.

body weight.20the particu1. In vitro studies of the insulin-mimetic actions of vanadium compounds are usually undertaken with phosphatases and kinases related to the insulin signaling cascade. and/or several protein kinases and phosphatases downstream. and reliable.day7-20 The dose of vanadium required to achieve good diabetic control varies with the initial diabetic state of the animal. inexpensive. This step.1 and 0. a reproducible model of diabetes is obtained by administering streptozotocin (STZ) to rats at doses of 45-75 mg kg . For in vivo studies. activates the intracellular protein tyrosine phosphorylation of IRs. not shown) f stimulation by phosphatase inhibition is independent of the insulin cascade. or with enzymes relevant to glucose and lipid metabolism. STZ-treated rats are insulinopenic. The model does not completely parallel Type I diabetes in humans. and is particularly susceptible to vanadyl stimulation . at oral doses of between 0. in adipocytes or other tissues. with several vanadium compounds. IRS-2. in that STZ-diabetic rats can survive without administration of exogenous insulin. it is relatively simple. Moderate to good diabetic control has been obtained in the STZ-diabetic rat.7 mmol kg. shc. Intracellular kinases afected by tyrosine phosphorylation activatioddeactivation under phosphotyrosine phosphatase (PTPase) regulation (vanadium-inhibitable)include (especially) IRS-I.' phosphatidyl inositoId-kinasF\ (P13K) growth factor receptor bound protein (GRB2) glucose transport + 1 "?) dephosphorylatio> phosphoryiation/ cascade radraf complex v mitogen-activated protein kinase (MAPK) DNA synthesis lipogenesis amino acid metabolism other insulin-mediated signal transduction Figure 1 Insulin signal transduction cascade (simplijied). hyperphagic. Cytosolic protein tyrosine kinase (CytPTK. may be potential sites of action by vanadium (Figure 1). and catabolic. and MAPK. however.3*691 insulin receptor (tyrosine residues) INSULIN CASCADE insulin receptor substrate@) (IRS-1/1RS-2khc) -. V? indicates possible sites of vanadium's mechanism o action. but is also mukti-step.94 Vanadium Compounds as Possible Insulin Modifiers extracellular side of cell membranes.

Chris Oruig. whether + 4 as vanadyl. to hemoglobin in erythrocytes. Vanadium (IV) is found as the vanadyl ion.~~-~~ orthovanadate.Lipid. a mixture of [HV0412. bound to small molecules or proteins. in most cases. stimulated glycogen synthesis in rat diaphragm and liver.30Vanadium is multivalent. Margaret C. (Orally administered vanadium has effectively normalized blood glucose in this animal model of diabetes only when used in conjunction with exogenous in~ulin. aerobic solution. ~ ~ ~~ and glycolytic pathways were affected in specific tissues. or + 5 as vanadate. ~ Vanadyl is known to be a less effective inhibitor of cellular phosphatases than an ad ate. mimicking insulin’s effects. depends on ligands. The . For instance. the + 5 and +4 oxidation states predominate in physiological environments (pH 3-7. there may be enough redox activity intracellularly to allow reversible formation of vanadate (and/or pervanadate) such that inhibition of tyrosine phosphatases may still be a crucial feature of vanadium’s mechanism of insulin-like action. pH. balanced by oxygen tension and the presence of endogenous reducing agents such as ascorbate and catecholamines. and c h i ~ k s .39 2 Characterization of Vanadium’s Insulin-mimetic Effects Characterization of the effects of inorganic vanadium on cell homogenates led to the first recognition of its insulin-mimetic effects. vanadate inhibited lipolysis and stimulated lipogenesis in a d i p ~ c y t e s . although no biological role has been ascribed to it. ambient t e m p e r a t ~ r e ) . and therefore the predominant route of elimination is via the feces. and inhibited hepatic gluconeogenesis. and solute concentration. The majority of intracellular vanadium appears to be present as vanadyl. however.Vanadium in blood plasma exists in both oxidation states. ~ ’ Vanadium(v) is generally present as . the total body pool of vanadium is estimated to be 100pg. McNeill 95 lar vanadium compound given. Cum and John H . however.30931 Most orally ingested vanadium is not absorbed. V 0 2 + . vanadium has been shown to be essential for growth and development of goats. and to glutathione or other low-molecular-weight compounds intracellularly.Katherine H . especially those containing thiol g r o ~ p s .* M) in virtually all cells in plants and animals. Thompson. ~ ~ .and [H2V04] -.21v22 addition of other trace and as yet undetermined individual f a ~ t o r s .~’) Vanadium is a putative ultratrace element.^' however. mice.35 Physiological vanadium is largely protein bound (as are many other trace elements): to transferrin in plasma.BioBreeding (BB) Wistar rat is a ~~ ~~ spontaneous model of diabetes that closely resembles Type I diabetes. ~ ~is normally present at It * ~ * very low concentrations ( < 10. The redox state of vanadium. ~ Glucose uptake and transport were enhanced by vanadate in ’*~~ . Sodium vanadate stimulated glucose uptake and glucose oxidation in rat adipocytes.26Reproducible and consistent deficiency symptoms have not been described. rats. It is characterized by rapid onset of hyperglycemia at between 60 and 120 days of age and by loss of pancreatic insulin production.29 In man. usually at millimolar concentrations of added vanadate or ~ a n a d y l .

was increased by concentrations greater that 25 pM. via .0 pM. in vitro and in vivo effects were found to be In contradictory. ~ ~ and increased glucose transporter expression in the presence of vanadate was shown in vitro in NIH 3T3 mouse fibroblasts4’ and in vivo in rat skeletal muscle. vanadate inhibited glucose transport in rat intestine when added to the perfusate of an everted gut sac or when added as a supplement to the drinking water of diabetic and non-diabetic rats at concentrations up to 0. vanadium effects on liver enzymes have tended to be more consistently insulin-like.70 hypothesis of insulin receptor (IR) A .5-1.and glycogen ~~’~~ ~ynthase. when using a concentration range of 0.43 rat skeletal muscle.08 mM for 2-4 weeks.53*54 other experiments. ~ ~ phosphoenolpyruvate carboxykinase. in vanadate-treated diabetic rats having plasma concentrations of 17.1-1. and enzymes involved in glycogen metabolism were affected in an insulin-like manner.47 Stimulated translocation of the GLUT-4 glucose transporter (which is regulated by insulin) to the plasma membrane was demonstrated in rat a d i p ~ c y t e s .' 8 i 2 4 * 6 9 resulting in an apparently enhanced peripheral insulin ~ e n s i t i v i t yand a lowered insulin demand. glycogenolytic (glycogen breakdown) effects5 and insulin-like.44 and isolated villus cells. on the effects of vanadium in vitro on pancreatic cells: enhancement of glucose-stimulated insulin release in rat and mouse islets was observed at concentrations of 0.^. there was an increase in liver glycogen content.68Vanadium treatment in vivo has been consistently and reproducibly associated with a decrease in plasma insulin in control animal^.1--5. these apparently opposing effects were also demonstrable in hepatocytes isolated from diabetic There have been conflicting reports.M. Differences between the reported effects in vivo and in vitro may be dependent on the actual concentrations of vanadium to which tissues are exposed. Furthermore. as well.0mM vanadate in rat hepatocytes.’ 8 9 6 4 pyruvate k i n a ~ e . glycolytic effects56 were demonstrated. al~~ though direct inhibitory effects cannot be completely ruled out.59’60 other investigators have shown inhibition at yet lower concentrations (1-50 P M ) .96 Vanadium Compounds as Possible Insulin Modifiers other tissue types: mouse brain. such as g l u ~ o k i n a s e .50 Vanadate increased sensitivity to insulin in activating glucose transport in normal and insulin-resistant rat adipocytes51 (but not human a d i p o ~ y t e s ~ ~ )enhanced insulin binding. ~ ~ In vivo. Some insulin-like effects of vanadium are concentration-dependent. for instance. glucose output was inhibited by the addition of but vanadate in the range of 0. Thus.45 Conversely. 3 Sites of Action of Vanadium The sites of action of vanadium’s insulin-mimetic effects would appear to be along the insulin cascade (Figure l). For example. in perfused rat liver. at least on restored liver and muscle glycogen.0 mM.~’ that these effects. and are not secondary to a restoration of normal plasma glucose levels.0p.62 A series of in vivo studies has also consistently demonstrated that vanadate treatment of diabetic animals partially or completely restores liver and muscle enzyme activities involved in glycolysis and glycogenesis. both non-insulin-like.5-22.

had any effect on the kinase activity of partially purified adipocyte IR preparation^.O.86 This effect was secondary to the inhibition of phosphotyrosine phosphatase. or by a stimulatory effect on protein tyrosine kinases downstream from the IR.^^*'^ Vanadate may act at a post-receptor level through its inhibitory effects on several protein tyrosine phosphatases. oral tion at concentrations that produced a marked anti-lipolytic vanadate treatment failed to change IR tyrosine kinase activity in STZ-diabetic rats. a cellular protein tyrosine kinase. unlike vanadate. markedly enhanced the IR kinase activity and increased the phosphorylation of at least four related proteins in rat hepatoma (Fao) cells.O. unlike insulin. was strongly activated by vanadate in rat adipocytes.’ Peroxovanadate inhibited lipolysis.~~~~ ing hexose uptake or inhibition of l i p o l y s i ~Activation~of CytPTK may result from the oxidation of vanadyl to vanadate by H.) presumably represents a mechanism of insulin mimesis different from that of vanadate or vanadyl. Thompson. enhanced insulin sensitivity. a substrate of IR tyrosine k i n a ~ eThe~ activity induced by vanadate (10-200 pM) augmented insulin binding. but did not phosphorylate serine or threonine residues of the r e ~ e p t o r .9 Other studies have tended to rule out any involvement of the IR in vanadium’s insulin-mimetic effects. and may mediate vanadate-stimulated lipogenesis and glucose oxidation without affect. the activation of glucose transport by vanadate in rat adipocytes was shown to be unaffected by a loss of 60% of insulin receptors. as a possible modulator of effects of the intracellular vanadyl insulin m~dification. and prolonged insulin-stimulated lipolysis in rat a d i p ~ c y t e s 7~ ~ .. McNeill 97 tyrosine kinase stimulation and/or IR /?-subunit phosphotyrosine phosphatase (PTPase) inhibition has received some experimental s ~ p p o r t .. ~ Vanadate 1-75 activated autophosphorylation of solubilized IRs.80*81 Moreover.^' Vanadate.O. Katherine H . increased tyrosine kinase activity in intact cells.Chris Orvig.~’ Stimulation of insulin receptor kinase by peroxovanadates (complexes of vanadate and H. either isolated during the basal state or in response to a submaximal insulin lamp. .’~ Recent evidence suggests the possibility of in situ formation of per- . similarly to insulin. thus implicating cellular H.~’ Vanadate also activated tyrosine kinase in the IR /?-subunit76and stimulated turnover of a phosphoryl . stimulated protein synthesis and lipogenesis. ’ ~confirming its inhibitory *~~l~~ -73 action exclusively on tyrosine pho~phatases. potentiated insulin-stimulated glucose uptake in rat adipocytes. Cam and John H .92 Although peroxovanadate. CytPTK.’ 1*90*91Low concentrations (5-20 pM) of peroxovanadate.O. neither compound. but not vanadate. Examination of tyrosine phosphorylation of the IR and other intracellular proteins in rat adipocytes revealed that vanadate had minimal effects on tyrosine phosphorylaAlso. Margaret C. and promoted autophosphorylation and activation of the IR tyrosine kinase. an effect that correlated with an increase in protein phosphotyrosine content. effects that were accompanied by inhibition of tyrosine phosphatase a~tivity. ~ activated tyrosine kinase group of pp15. with added H.82 suggesting a post-receptor mechanism of action. I n vivo studies showed that oral administration of vanadate produced profound insulin-like effects on glycogen metabolism without activating IR kinase.

relatively non-toxic.~~ Thus. but also partly because of vanadium's anorexogenic effects. probably because of the normalization of metabolism. ~ ~ ever portion of intracellular vanadium is present as vanadyl is probably acting by some alternative mechanism. thus changing the kinetics of tyrosine kinase activation. the addition of sodium orthovanadate (0."' or vanadate with added hydrogen peroxide. Subsequent studies have shown that the glucose-lowering effect with orally administered vanadate or vanadyl solutions is persistent. perhaps also involving a cytosolic (i.54 mmol kgday-') to the drinking water of STZ-diabetic rats over a period of six weeks dramatically lowered plasma glucose levels without improving depressed plasma insulin levels. and accompanied by other desirable effects in plasma parameters. which may result in irreversible inhibition of some protein tyrosine phosphatases. 4 Animal Studies and Human Trials The first definitive study to demonstrate in uiuo glucose-lowering and antidiabetic effects of vanadium was performed by McNeill and co-workers in 1985. non-receptor) protein tyrosine kinase that is stimulated preferentially by vanadium and may be insulin i n d e ~ e n d e n t . vanadate's substitution for phosphate within the enzyme structure may result in formation of a transition state analog of protein tyrosine phosphatase."' Decreases in plasma glucose levels and improved tolerance to an oral glucose tolerance test were recently reported to be similar regardless of the type of vanadium salt (vanadate or vanadyl) administered. ~ ~ .^^ In addition.39-0. particularly for tyrosine kinase assays.97 The preponderance of evidence favors a post-receptor mechanism in stimulating glucose utilization.102 the mechanism appears to be increased autophosphorylation and activation of the insulin receptor. is testimony of the ability of vanadate to preserve the phosphotyrosine content of cells. as opposed to the readily reversible phosphatase inhibition seen with van ad ate^.2 In this study.35i96 (which may be formed in sit^).8 5 1 0 7 Although overall mean body weight in diabetic rats is not improved.98 Vanadium Compounds as Possible Insulin Modijiers oxovanadates. However. ' Vanadyl is not a potent PTPase i n h i b i t ~ r .101. effects of intracellular vanadium on calcium i n f l ~ x ~ ' * 'as well as intracellular '~ have not been ruled out as important and intravesicular pH factors in the mechanism of action of vanadium as an insulin-mimetic agent.Routine ~ ~ ~ i ~ addition of vanadate to cell lysates. ~ The current preferred postulated mechanism of vanadium's in uitro effects is that vanadate stimulates specific protein-tyrosine phosphorylation by virtue of its inhibitory actions on appropriate P T P a ~ e s . and restoration of thyroxine levels to normal.^^ vanadate In the case of peroxovanadate11*12 plus superoxide radicals. such as triglyceride and cholesterol lowering. 3 2 5962. ~ ~ . both characteristic symptoms of STZdiabetes. l o 9 The apparent biphasic 9 7 . the treatment consistently normalizes the elevated intake of food (hyperphagia) and fluids (polydipsia). with consequent stimulation of glucose uptake. ~ ' -what~. and effectively acting as a means of fine tuning the intracellular balance between phosphatase inhibition and kinase a ~ t i v a t i o n .e.

cardiac. which suggests that insulin and vanadyl work in a complementary manner in viuo. " * ~ This' may be attributed partly to an accumulation of vanadium in ~~ '~ various tissue stores"' or to the preservation or improvement of pancreatic P-cell insulin content in vanadium treated rats. whereas pair feeding only partially reversed these parameters.') nonetheless had significantly lower plasma triglyceride and cholesterol levels and improved glucose tolerance as compared to untreated diabetic animals.day. along with plasma. wherein an increased sensitivity to insulin was ~ e e n . its exogenous l ~. Katherine H . hepatic glucose production and glucose utilization in response to submaximal or maximal insulin levels have been shown to be improved with vanadium treatment in STZ-diabetic rats84*85 in partially and pancreatect omized rats.5 mmol kg . ~ ~ . ~ 3 That peripheral insulin resistance is also reversed by vanadium treatment has been shown by glucose clamp studies in which indwelling venous and arterial catheters permit infusion of tritiated glucose to maintain plasma glucose levels in the presence of high insulin levels (hyperinsulinemia). reduced weight gain. suggested by in vitro studies." * In oblob mice. and adipose tissue abnormalities were maintained for a long period (up to 30 weeks) after vanadyl treatment was withd r a ~ n .Chris Oruig.' Vanadate treatment improved insulin-mediated glucose utilization. has also been confirmed by administration of insulin to vanadium-treated control and diabetic rats. improvement in pancreatic function as measured in situ. or non-insulin dependent diabetes mellitus (NIDDM)."6 In animal models of Type I1 diabetes such as thefalfa Zucker rat.20~"' The interrelationship of vanadium and insulin in the whole animal. although it does not completely eliminate. Diabetic rats that were not rendered normoglycemic by low-dose treatment with vanadium ( < 0. which in this animal model is normally exhausted as a result of hypersecretion. " ~ In addition.' insulin r e q ~ i r e m e n t . McNeill 99 nature of the glucose-lowering response to chronic vanadium treatment20*22i24 has led to the suggestion of a 'dual sensitivity' to oral vanadium treatment.lIg Vanadate treatment preserved pancreatic insulin content. The insulin-lowering effect of vanadium has particular relevance to Type 11. and improved glucose tolerance. An intriguing discovery was that euglycemic and improved glucose tolerance.12' In contrast to vanadium's effects on STZ- ' '' '' . leading to an indefinite period of near-normal glucose homeostasis in the fed state in these animals (unpublished observations). Glucose tolerance of genetically obese (dbldb) diabetic mice was also improved by vanadate treatment. treatment of the genetically diabetic BB rat reduces. Using this method. without increasing glucose transporter levels in muscle. but significant. vanadate treatment reduced food and fluid intake. Thompson. plasma glucose and insulin were reduced by 50% within one week of beginning treatment with vanadate.22*'10It was subsequently found that the more severely diabetic rats (as determined by relatively higher blood glucose levels and lower plasma insulin levels before treatment) either could not be rendered normoglycemic or required higher concentrations of vanadyl to become so.in which hyperinsulinemia is a frequent concomitant and may be a contributing factor in the development of secondary complications of the disorder. as well as a slight. Cum and John H . Margaret C. lowered hyperinsulinemia towards normal levels.

Vanadyl therapy for 6 weeks at 50 mg V day. At 25 mg V day . was corrected by ~ a n a d y l .') for 2 weeks resulted in significant increases in mean rates of glucose metabolism during a euglycemic clamp in two out of five subjects with IDDM and in five out of five subjects with NIDDM.resulted in improved insulin sensitivity in three of five human NIDDM subject^. which were sustained for 2 weeks after treatment was withdrawn.100 Vanadium Compounds as Possible Insulin Modijiers diabetic rats. common in STZ-diabetic rats by 8 weeks after induction of diabetes. which often precedes cataract development.107v'2'~122 Sorbitol accumulation. ~ and " ~ alterations in kidney morphology ~ ? the (swelling in the distal tubules) observed in untreated STZ diabetic rats were absent when they were treated with ~ a n a d y 1 . and the adrenal glands (as compared to the significantly elevated ratios observed in untreated diabetic rats). The development of cataracts.1 24.0 ng ml. Treatment of NIDDM subjects with vanadyl sulphate (100mg day.there was no change in glucose and lipid metabolic parameters and the peak serum vanadium was 15. indicated by diarrhea and subsequent dehydration.7 ng ml. creatinine) and liver function (AST). clinical trials of vanadium compounds have been initiated on human Treatment with Type I (IDDM) and Type I1 (NIDDM) diabetic sodium orthovanadate (125 mg day . liver.2v'08 It has been suggested that vanadate is perhaps not as well tolerated as ~ a n a d y l . Reversal by vanadium treatment of a number of secondary complications relevant to diabetes has been reported.3~69~107~122*'24~'25 treatment also resulted in near-normal organ/body weight ratios of lung.'^' while basal hepatic glucose production was unchanged. measures that are aberrant in the diabetic state. heart.Several plasma parameters that l~~ measure kidney function (blood urea nitrogen.'. characterized by increased lipolytic rates. and an increased rate of glucose disposal. kidney. was also reduced by oral vanadate treatment of diabetic rats.7 3. ' ~ ' * ' ~ ' however. was eliminated with vanadyl treatment. were reported to be Vanadyl normalized by vanadate or vanady1. Peak serum vanadium at this dose was 79.'. and lowered serum cholesterol levels in all subjects. as shown by working heart studies.') for 3 weeks caused an improved insulin sensitivity.' 2 7 * 1 2 8 In both of the studies.126 There were decreased insulin requirements in the IDDM subjects.122*125 Recently. there were reported incidences of mild gastrointestinal intolerance. no effect on plasma triglycerides was seen in models of Type I1 diabetes.'23 Improved cardiac function in STZ-diabetic rats. differences may be slight. ' ' 5 Toxicological Considerations The type of toxicity most often associated with oral vanadium treatment is gastrointestinal. There was no increase in thiobarbituric acid reactive substances (an indicator of in vivo lipid peroxidation) at these doses."' The group of Domingo and colleagues has conducted a number of studies which showed an unfavorable toxicological . ~ ' Abnormal adipose tissue function in diabetic rats. a reduction in hepatic glucose production. resulted from orally administered vanadate2 or ~ a n a d y l .

71 mmol kg.ugg-' in kidney. and 0.4 pg g. sodium orthovanadate or vanadyl sulfate. but also no acceleration in the development of morphological abnormalities in a variety of organs (by histopathological tests) and no outstanding changes in hematological p a r a r n e t e r ~ . tartrate. whether sodium metavanadate. There was no obvious toxicity at this dose.and diperoxo-vanadates). all the test animals died of diarrhea within four days. and cysteine methyl ester.') body weight by gavage was slightly more effective than other ligands in lowering plasma glucose levels within 24 hours of administration. Tissue vanadium levels (at the end of one year) ranged from 6.16-0. Selected coordination complexes of vanadium for which data (either in vivo or in vitro) have been published are shown in Figure 2.2mmol kg. Margaret C. these ligands have generally been chosen to impart specific features to the resulting vanadium complexes: improved lipophilicity (vanadyl cysteine methyl ester. malate.'^^ However. naglivan). A series of bis(ligand)oxovanadium(Iv) complexes were tested in STZ-diabetic rats. or facilitation of transmembranal ion uptake (RL-252 and analogs).1pgg-' in bone. vanadium intake was usually.136 ' 6 Improved Tissue Uptake with Metal Chelation An important advance in the use of vanadium compounds as insulin mimics has been the development of various ligands in order to improve substantially the absorption. The vanadyl cysteine methyl ester complex.1-0. ligands have been chosen to chelate vanadyl. compared with 60% in the untreated diabetic rats. ' ~ ~ Overall mortality was 19% in the vanadyl-treated diabetic rats. Katherine H .'^'-'^^ In these studies. however. improved oral absorption by passive diffusion (BMOV). but not always associated with lowering of plasma glucose levels. oxalate. ' ~ ~ .' day.6 to 7.5 to 15. Naglivan doses of 0.day-' showed not only normalized plasma glucose and lipid levels in treated STZ-diabetic rats.Chris Orvig. Cam and John H . Intraperitoneal administration of the iron chelator tiron with oral vanadate in STZ-diabetic rats lowered the accumulation of vanadium in several organs but did not diminish the anti-diabetic efficacy of ana ad ate. at ten times the glucoselowering dose. suggesting the necessity for careful ligand design if the concept of chelated vanadyl is to prove worthwhile.12' Further reduction in such toxicological effects as decreased weight gain and reduced food intake. thereby reducing the dose required for optimal insulin mimesis. tissue uptake and intracellular mobility of vanadium compounds. The water-insoluble vanadyl complex naglivan [bis(N-octylcysteineamide) 0x0-vanadium(~v)]has been given to STZ-diabetic rats in a suspension of 3 % acacia gum by oral gavage. 3. at a dose of lOmg V kg-' (0.effective- . a one-year toxicology study involving vanadyl sulfate at doses of 0.2 to 0. McNeill 101 profile of vanadium regardless of the salt administered. With the exception of the peroxovanadium(v) complexes.3.* ligands included salicylate. may be possible by using low-dose vanadium in mildly diabetic animals. These are discussed below. potentiation of in vitro insulin-mimetic effect (the monoperoxo. with an increased incidence of mortality and the accumulation of vanadium in tissue^. Thompson.in plasma.3 mmol kg .

21The increased absorption of vanadium from BMOV was reflected in higher tissue vanadium concentrations. however. electrically neutral. neither was there diarrhea associated with naglivan treatment. although the onset of action was significantlyslower than with uncomplexed vanadate or vanadyl. BMOV [bis(maltolato)oxovanadium(~v)] designed as a vanadyl complex was that would be water soluble.4 mmol kg . maltol. is an approved food additive. it was 37-1 expected that the compound would also exhibit a positive safety profile. At a dose of 0.'O-' Naglivan treatment of both control and experimentally diabetic animals was not accompanied by weight loss or a reduction in food or fluid intake over an eight week period. .BMOV reduced blood glucose and lipid levels to near normal with no diarrhea and no mortality during a six-month test period.(im)lImidazoleoxobisperoxovanadate(V) Figure 2 Representative insulin-mimeticcoordination complexes of vanadium(Iv)and (v) ly lowered blood glucose levels to near normal.102 Vanadium Compounds as Possible Insulin Modifiers Vanadyl cysteine methyl ester Naglivan RL-252 v-P [v0(o~2(~L')InLigandoxobis(peroxo)vanadate(V) [V0(03(H2O)2(L-L')lnLigandoxoperoxovanadium(V) [vo(od.~*' 39 Since the ligand. and readily available for gastrointestinal absorption from drinking water (by passive diffu~ion). of low molecular weight.day. as compared with those from a similar treatment '' ' '.

l ~ an RL-252 was maximally effectiveat molar ratios of 10: 1 vanadyl sulfate: chelator.1mmol kg.Intraperitoneal administration of this compound proved to be more effective that oral treatment.'49 The coordination of vanadium(v) to imidazole presents structural analogies to the coordination of vanadium to histidine residues in vanadium-containing haloperoxidases' 5 0 and some phosph~rylases. BMOV was found to be 2-3 times as potent (in terms of glucose-lowering) as vanadyl sulphate. followed by a maintenance dose of 0. A series of dihydroxamic acid chelators have been designed as hydrophobic carriers of ~ a n a d y 1 .structureI2inhibited phosphotyrosine phosphatase (PTPase). and lowered plasma glucose when administered intraperitoneally.day. a six-coordinate bisperoxovanadium imidazole compound (see Figure 2) has been synthesized and shown to enhance insulin receptor autophosphorylation in human liver cell culture. when injected in the portal vein of rat 1 i ~ e r s . I ~ ~ Peroxovanadate lowered blood glucose in diabetic rats when administered intraperitoneally.'^^^'^* Because of the potent insulin-mimetic properties of these uncharacterised peroxovanadate mixtures.and diperoxo-vanadate(v) complexes have recently been prepared and tested in v i m .I2 These potent insulin mimics activated insulin receptor tyrosine kinase and inhibited PTPase in rat liver endosomes at 5-80 pM vanadium concentrations. at concentrations ranging from 1 pM to 1mM. and optically chiral. Pancreatectomized rats treated with a low concentration of oral sodium vanadate (0.L-L']" . They also had a synergistic effect when co-administered with insulin.I2 More recently. with very interesting result^. In ~ assay of lipogenic stimulation in rat adipocytes. a variety of well-characterised monoperoxo.'2*'44Complexation of vanadium(v) with hydrogen peroxide increased phosphorylation of the pstimulated lipogenesis.5g 1. vanadium has also been complexed with peroxide in order to potentiate its insulin-mimetic effect. In addition to organic ligands. lipid-soluble. but was ineffective when administered rally.'^^'^^ Twelve different peroxovanadium compounds of the [VO(O.Chris Orvig. It was administered orally to STZ diabetic rats at an initial dose '~~ (for 2 days) of 0.''T92 Peroxovanadate also increased the phosphorylation of several proteins.2mmolkg-' day-' to achieve normoglycemia. when administered by oral gavage or intraperitoneal injection.). including those of the insulin receptor. . inhibited lipolysis. McNeill 103 course with vanadyl In a number of acute experiments. suggesting a shuttle mechanism of action. Thompson. as well as increase glucose transport in rat adipocytes. they released the bound metal ion when treated with aqueous glutathione solutions. Margaret C. and promoted protein synthesis in rat adipocytes at micromolar concentrations of vanadium. and lowered plasma glucose at doses in the mmol kg-' body weight range. but both achieved significant glucose-lowering. These compounds were electrically neutral. 141 VP [bis(pyrrolidine-N-carbodithioato)oxovanadium(~v)] was initially tested as an in vitro insulin mimetic by inhibition of free fatty acid release from rat a d i p ~ c y t e s . stimulated insulin receptor tyrosine kinase.'~' Combinations of vanadium and other trace elements have also been studied as a way to potentiate the action of vanadium.Katherine H .') and lithium carbonate '. Cam and John H .

C. Fantus. Sigel. 1995. D. Marcel Dekker.359. Science. J. Serrano. Biol. Res. Shechter..20. Z. Clin. and had partially restored liver and kidney superoxide dismutase activity after 16 days of treatment. 6 N. 1899. J . Ann.M. Vanadium and Its Role in L$e. Biochern. 193. J . Ng.G.H. Brichard. 1993. 1994. Rev.G.46. K. Sack and Y. 1987.104 Vanadium Compounds as Possible Insulin Modifiers showed improved glucose uptake and skeletal muscle glycogenic rate by glucose ~ 1 a m p . l ’Addition of magnesium and zinc to some of the vanadium. Diabetes is characterized by insulin deficiency or insulin resistance. R. Vanadium Compounds: Biochemical and Therapeutic Applications.8. G. 7 A. Biochenz. 1992. Lederer and J. Cros. Chew.35.P. So0 Lum and A.H.191. Both organically-chelated and inorganic vanadium compounds have been shown to lower blood glucose and ameliorate other diabetic symptoms in a variety of animal models of both insulin-dependent and non-insulin-dependent diabetes... Bevan. Tahiliani and J. 7 Summary The potential of vanadium compounds to mimic insulin when administered orally has stimulated renewed interest in the pharmacological effects of these unique and enigmatic compounds. Czech. H. Farfel. Sakurai. I. V. Posner. Srivastava and J.A.4596.-J. McNeill.and ~ lithium-treated rats produced a further slight. Chiasson. 12 B. Kluwer Academic Publishers. G . Mol. Nukatsuka. Because insulin cannot be absorbed intact following oral administration and must be administered parenterally. 4 S. 3 J. 435. Shaver. 17. by a mechanism unrelated to insulin.. J. H. Yuen. Nutrit. Shisheva and Y. Dordrecht. 2 C. Biol. A. Faure. A. . Kawada.. 11 A.W.K. Komatsu. 1985. Kluwer Academic Publishers. 13 M. Heyliger.G. Burgess. McNeill. J . Hall. 5 H.227.. J. Sigel and A.-L. were rendered normoglycemic within 4 days.H. Determining a therapeutic dosing regimen without significant toxicity for orally administered vanadium compounds would present a significant advance over currently available treatments. Chasteen. Diab. 111.-C. Hoveyda and C. Metal Ions in Biological Systems.B. S. but significant. Cell Biochem. (Zn2+ has also been found to mimic several actions of insulin.’ In separate experiments. Dordrecht. Ishikawa. Prac.269. Vanadium in Biological Systems: Physiology and Biology. McNeill. D.. J . Lachance. References 1 B. 8 H. Orvig. Med.133. 1990. Chem. Cam. La Presse Medicale. Tsuchiya. J. 10 M. Endocrinology. 31. Lyonnet and E. 1977. M. J. in uitro and in uiuo.P.E. Clin. Vol.6658.Chem. Zhang-Sun.262.R. Meyerovitch. B. Martz Martin. available therapies are cumbersome at best.1562. 153. Henquin. Inc.M. 1991.I.1489. Lazaro and J. Yoshida and M. Diabete Metabolisme (Paris).1474.D.7. improvement in tissue glucose uptake.. 1990. 1993. diabetic rats co-administered sodium vanadate and lithium in their drinking water.H. New York. 9 J. Shechter. J. R.

20.J.283. Sci. 1997. Mol. 48 M.J.36. P. Williams. Koch. H.M. 41 G.S. J. Saxena.H. R. p.107.2754.G. 20 K. McNeill. Saperstein. 22 M.. Ed. Carrano. Struct.V. Chasteen. 38 L.N.149. Shechter and S.E.K.J. Engl. 19 N. Bruech. Uthus. Biol.Blondel.263. Kanthasamy. Chem. Hand and D. 1978. ed.. McNeill. 49 K..K.R. 1991. Brownsey and J. Kahn..25. 1981. Biswis.H. Holloway.H.1151.31.. J . J..86. Tsaprailis. Physiol. Diabetologia. Adv..G. Mueckler. 16 M.E. Biophys. 46 G. 43 S.R.J.311. Invest.80.126. Nielsen and E. J. Pharmacol. Versieck. Kelly. E459.32. Pederson. 35 M. Ramanadham.E. E. J. 1984. Butler and C. 1990. Biophys. Cantley.258.. 47 K. J. Legrum. McNeill. Amir. J. Cornelis. J . Biochem.F. Gingras. 24 M. Grady and C. Degani. Chevalier and B. S. 39 G.R. Coord. Dordrecht. SOC. 2969.5306. Mees. 1990. 1993. 38. Diabetes. J.W.10. Endocrinology. A. 1980. Liepnieks. Kahn and M. 197. Orvig and J. 1988. Barker.R.71. Inorg. Solomon. Med. Biochim..O. Sci. Diabetes.. 1988. 1991. Biochemistry.J. Brady. Cam. S. 1990.G. Slot.H. McNeill 105 14 G. 1984. McNeill. B. Biol. Pederson.G. Kellett and E. Path. Hajjar. Duckworth. 1993.F. R. Subramanian and S. Endocrinology.D.P. J . 30 A. Quintanilla. Meyerovitch and Y. Cam and John H . Serrano and J. 1990. 50 H. James and M. 1390.. Baquer.H. Kleinzeller.A.207. Biol. Res.D. Aisen. Wever and K. Nechay.92. G.L. Berger.K.R. Chem.109. 1990. Shimazu.284.843.. Res. 1980... Int.C. 42 W. 1989. Okumura and T. Kosta. Brain Res.J.R.D. S. Biochem. Chasteen.H. Liu. Shechter. Paquet. Simon. Toxicol. Bendayan and D..103. P. Dobish and T.W. Ramanadham.122. 1989. 218. Yuen.A.P. J . Biol. Angew.257.M. Byrne and L. M.H. Katherine H . . Ann. Ramachandran. J. Can. Chasteen. Pharmacol. 40 Y.2285. 1993. C. McNeill.255.. Karlish and Y. Fedorak. Pharmacol.R. L. 32 R. 1987. Barbier. Inorg.283.979. Dubyak and A. Gresser and C. Res. 18 0. Proc. S. J.D..61. 27 N. and P. 31 R. Romanek and R. V. 1989. J . 17 R. Trends Biochem. 1981. S.1992.L. Portha. J. Tomicic.. 1990. Physiol. 1993. S. (Tokyo). 1989. G.5. Porter and R.419. Moffat. Sci. Strout. 1988. Diabetologia. Kale and N. Pharmacol. R. M. J. 21 V. Biol.J. 1979. Gochin.35.E. 34 A. 1993. Shechter. K.5795. Fuhrmann. Am. Toxicology.. Nature (London).D. White. Cros.148. R.254. Kennedy.. N. Kustin. Srivastava.J.68. S.F. Thompson. 105.109. Physiol.53. J.1549.. Leichter and J. Total Envir.D. Flier. S. Biol.2025. Sekar. Trace Element Res. White. Bonding (Berlin).. A. Keutmann and C.F. Endocrinology. Biochem. Huyer. P.1126. J . D. Jr.556. Hoste and F. Can.22. Chem.127. 1983.272. 1980. Clin. Mountjoy and J. 23 P.Chris Orvig. 263.486.J. Lienhard.3. A. Payette. M. Chem. 33 F. 37 H.J. F. Rev.24. Peavy.K. 1986.17.E. Acta.C. Commun. Kluwer.Exp..2728.81.561. 1992. Commun. Chem. Govinsasamy.E.M.J.501. W.. 36 N. Hamel. 35.190. Rehder. Thompson.H. 25 S.C. Madsen. B. Sargeant. C.S. 29 D. 51. 44 N. J. Pharrnacol. Buchan and J.267. Chem. Mongold. Margaret C. Am. Karlish. 1992. Macara. 45 J. 26 I.M. J . Shaleson.112. M. Rev. 1781.392.42. 28 B. R. J . Chem. Chem. Chem. 15 C. Pilchand J.82. Cell. J. 1990. Netter and G.Z. in Vanadium in Biological Systems. Vicario.

FEBS Lett. Strout. Fillat. Gil.M. F.M. Ahmad and G. Biophys. Diabetes. J.J.. 56 A.J. Brown. Nenquin. Bartrons. I. T. 86 A. 64 A.216. Waldhausl. H. J .. 11.43.M. Y. M. Diabetologia. Fujio. Shechter and S.. Saxena. Hepatology.. Clin. Gilon. Diabetologia.218.Y.A. K. 59 J. 113.V.G.Invest. Henquin.663. 124. Sci. Biochem. 74 A. Rotenberg.L. Clin.124..M. 1988. Sesti and R. Cordera. 267. Zhang. 1989. Acad. Gomez-Foix.84. Natl.35. 1987.J.E. R. Stalmans. Bernier.P.W. Ventura. K. 159.903. Drews and J. Steffan. M.. R. Rodriguez-Gil.. Chem.. Eriksson. P.152.. Res. 1993. 66 S.R. Gomez-Foix and J. Sparks. Toth. S. Arino. Prskavec.. 1992. 1988. Carreras and R. Z.8. Mongold.. J . Vierhapper and W. 77 M. A.. Mooney.W. J . Laughlin. 61 C.I. 1989.106 Vanadium Compounds as Possible Znsulin MolliJiers 51 J. Biol. 83 R.L. 1448. A.A. 1991. G. Biochem. A.238. Smith and G. Jackson. S1.C. Fillat. Liener.256. Vicario.I.263. Biol.E. G. 113. Res.R. J. Krepel. Salhanick. Khandelwal. Shisheva and Y. Res. Furnsinn. Diabetes.4. N.. K. 1992. Biochem. J . 79 I. Montemurro.83.13 626. Larner.M. 1987. 1990. 1986. Cohen and D.V. 1234. P. 1988. Bosch. Cros and J. H. Deragon. Deragon. 1983. J.127. 1990. Bosch and J.G. Gao. Commun. J.E. Y. J.38. Exp. Gherzi.422. Eur. Tracey and M. Ramanadham. 1994.Q. Lonnroth and U.M. J . F. P. Amatruda. Bertolini. Ahmad and G. Srivastava and N. Mol. B. 60 A. Biophys.. 71 G. Sparks. A. 54 M.H. USA.. Chem. Portha. Speeg.W. Caratti. Diabetes. 1989. Proc. Smith. Guinovart. Eriksson.1474. 58 J. 57 J. J.507. Garbers. 1988. Sale. Pharmacol. Rodriguez-Gil. J. Roden.K.266. 1993.14. Hartmann and H. 1993. 1992. Fantus. Bruck.Z.793. G. Blondel. P. 63 J. Metab. Endocrinology.821. 84 0.21649.91. Voss.B.J. Biochem.. Bosch. Arino. P. Arakaki. Giunovart.32. Diabetologia.609. McNeill.E. Davidson. Lane. 73 A. J.A. 144. Chew.93. Z. Biol. Bolognino and J. 70 E. Miralpeix. J. P. Luhowskyj and J. 1993.257. Bollen. J. Commun.300. Gresser. 1991. Physiol. Tsiani and I. Biochem. Bailbe and B. C. J. Diabetes. 62 S.91. 67 M. 75 R.D.7298. Rodriguez-Gil. Brichard. Diabetologia.39.E. 81 H. 1868.M.99.92. Smith. Chem. 55 F. F.. Fantus. Cell.. Posner and U.J. Guinovart and F. R. Biol. G. J. Lonroth. M. Clin. B. Bordwell. J . Nowotny. 80 T.269.37. Endocrinol. Herrmann. Casnelle. H904. K. Prigozin. Fagin.262. Levin. Trends Endocrinol. A. I.M. 540. Shechter. C. Biochem. Invest.40.602. J .J. 185.J. 78 I. Ikejiri and S. 69 S. 892. Biochem. A. Venkatesan. F. Girard. Am. J.80. Endocrinol. Diabetes.E.G. Biol.K. 82 A. Miralpeix. C. Bosch. Baquer. Bartrons and W.123. 72 S. Avidan and M. K. Pugazhenthi and R.. D. Bosch. Biophys. Endocrinology.263. R. Green. Ueno. 53 R.E. 1989. Laird and M.E. Guinovart and F. 1989. C.492.255. 1990. 1987. Commun. Bar-Meir.D. S.36.H. Andraghetti.1355. 1988. C.. Rodriguez-Gil and F. Gomez-Foix.I. J . 1989. B. 1986. J. 65 A. Desbuquois and J. Tamura.257. Dubler and J. Rossetti and M.36.J.510. Zuhlke. Diabetes. Saperstein and E. Valera. Fantus. J. Endocrinology. 1991. M. D. M. Slater. Gomez-Foix. 1988.1918. Takeda and H. Chem. 1992.36.40. 85 N. Brownsey. Swarup. 76 D. 1991. . 1997. 68 L.375. S.S. 1982.2716. 52 P. Diabetes.

Li and Y. Chem.I.265. Henquin. J. Pharmacol. 114 R.268.72. Biol. 1990. Endocrinology. Biochem.35.-C. 110 V. Biol. Palacin and A. Cam and John H . 113 M. J...288. G.665. S . Angel and R. 91 D. 99 A. 1994. Diabetes.-X..8314. Biol. L. Zick. Thompson. Trudel. Brichard. Res. Grinstein. Becker. Commun. 1987. 522. Pharmacol. Exp.276. Margaret C.477. 93 D. Kadota. U S A . Shechter. G.941. Lohse. Jr. Heffetz. Li. 102 D.15. C.D. McDonald. R.267. F. J . Shechter. 115 R.. McNeill. G. Chem.40. Diabetes.169. Pathol. Z. Sekar and Y. Macara. Biol. F. Testar.-C.74. McNeill. Bennett. 1996. Glaser. Biochemistry. J .9521. Fantus.46. R. Hersh and B. Henquin.. 116 G. Pottier and J. Posner. 94 J.2177. Henquin. Brichard. 1991. S. 1990. 1989. 118 S. Res.. Khandelwal. Chasteen. Metabolism. 119 S.J. 122 S. Elberg.H. G. Physiol.. Biol. M.. N. J . 1990. Zick. Behavior. McReynolds. Y..39.A. Ongemba and J. Meyerovitch. 109 D. Thompson and J. McNeill 107 A. 2.-N. Sack.201. K. M. Acta. Methods Enzmol. H.36. Zhou. Pugazhenthi.188.1. Rossetti. Li. Pathol. Dordrecht. Chem.H.R.235. Degani and S. 1989. M. Henquin.R. J. M.611. 87 88 89 90 . Commun. Environ. G. Metabolism.E.L. Pharmacol. Deragon.1684.259.. Furriols. Toxicol.. 28. Eur. Brichard.J. Katherine H . Shechter and S.8864. Guyda. Path. J. Biochem.G.L.125. Lozeman. Chem. J. 92 I.W. Diabetoloyia. Cantley. V.-C.M. 97 J. B. 104 D. Shisheva and Y.C.J. McNeill. Res.191. Guma.291. 111 M. Zhuang and L.p. B.121. M. Yuen and J.G.. 123 J.M. 1993. M.M. Camps.M. B. Reaven. Barbera. Newsholme. K. Pharmacol. 1978. Elberg. G. 1993. Zhang. Zorzano. Gordon. M. Pochal. M. L. Acara and G. R. Hecht and Y. J. He. Metab.C.. Elberg. 1995. Deragon.W. Pharmacol.C. 1993. Rutter and Y.H. Dai. Acad. Thompson and J.E. W.D. J. 1022.A. X. Van Etten and C. 1991. Chem. Res. in Vanadium in Biological Systems. 96 G. Fantus.P. J. Shechter.38. 108 J. 126.269. I..36. Commun. 1996.A. Biophys. J . 1992. Denu.G. Med. A. Zick and R. Cordera. Stauffacher. Biochim.M. Commun. Chem.10240.H. 1994. Sharma. 675.A.L. Biophys. 187.G.D. M. Arch. Sci. 101. Biophys.M. K. Foster and B. Biophys.J.44. Biochem. 101 Y. D. Vijayalakshmi.J. Rev.. 1984. Faun and J. 98 I.N. 1994.10381. 107 S.H.J. 106 P.44.118..241. Dixon. Biochemistry.95.147. Bruck. Andraghetti. Bailey and J.R. A.F. S..2.Chris Orvig. Commun. 100 S.2896.6463.269.M. Can. Posner.773. Clin. Kluwer. Challiss. J. J. 120 S.270. Ramanadham. Chem.H. Meyerovitch. Saper and J. J . Physiol..H. Farfel. 1997. J.260.A. 1991. J . Dror and Y. Orvig. Amir. 1993. Sagi-Eisenberg. Proc. 1992. Biochem.H. Heffetz. Munoz. J. J. 1990.357. Biochemistry.L.H. L.I. Battell. 1985.. S. 1989. ed. Rodriguez-Gil and J. Mongold and J. 332. Bar-Meir. Endocrinology.J. Toxicol. C. McNeill. 1992. 1113. Kadota.2493. 121 K. 117 S.. 112 R. Lohr. 1987.71. Pharmacol. Yuen.. D. Pharmacol.629. Commun. Ongemba and J. Guinovart. Biochemistry. Annu. J. Biochem.80..J. DeFronzo and L. J. Res. McNeill. Shechter.2510.29. 129. 105 Y. Parker and R. 1980. Leighton. Cam.45.C. 95 M. Biochem. Zick.H. Amir. 1989. Willsky.M. 1990. Cassel. Kustin and L.I. Shechter. J. H.631.V. Crans and Y. Biochem.93. Delfert and J. 103 D. Bushkin.20047.A. Y.35. N. Natl. 1992. Paquet and S..1326. Thompson and J. Brownsey. Biophys. Budohoski and E. Karlish. 1992. 1994. Cros. 1997. McNeill. R. R. 1991.

Shlimovich. 1994. Proc. J. Ng.J. M. 146 Y. Shisheva.41. A. McNeill. D. 135 J. Schneider and P. C. Cohen.J. Chem. Endocrinology. B.J. Cohen. D. 2063.L.N.J.. H. Pharrnacol. 1990. Endocrinol. Corbella.L. Klein-Robbenhaar and L. Hanson.N. L. Y. Gefel and Y.J.9.C. 128 M.J. D. Komazawa and H. Orvig. Battell.-E. M. Orvig..J. Toxicol. Pharmacol. J.M. Inorg. J. Keramidas. 133 J.2501.37. Diaz. D. 1996.L. 1993.153. M. H. Shisheva. Rossomando and L. D. Cell. Toxicology. 1995. 137 P.142. 134 J.M. 150 A. 127 N. 1994. Rossetti. 149 D. M. 152 L. Orvig. McNeill and C. Domingo. S. Patti and C. 39.139. A. 140 LA. 138 Y. Path. 1987. Sanchez. Dai and J. Domingo. Vandenberg. Li. Mol. Llobet and J. Ortega. M.R. Rosetti. J . I. K.. L. Llobet and C. Domingo. B. Chem. A. M. Can. A. Physiol. J. Cell.L. Wever. Toxicol. Inorg. R. C. Cros.227. Zick. J. Libman and A. Endocrin.M. Miller. Mongold.L. M. Watanabe. J . 1998.H. 1990. P. Keen. Sweet. Llobet. L. 181. J.95. Sanchez. Rettig. 153 A.Ilonomov and Y.75. Shisheva.C. Ruth. F. Eur.P. C. G. C. Biochem. Am. G. Physiol. 148 A. S. 143 Y. 147 A.M. Tep.A.8252. J . McNeill and J. Crans. I.J. Hall. Sun.G. 3109. Nakai. Hoover-Litty. Lazar. Goldfine. Nadiv. 1996. Lemoine. Acad. Folli. Goldfine. Chem.117. J . B.-F.495.262. Shechter. Inorg.H. 130 J..G. Appl. J . 1996. Thompson. Gomez. USA. J. Biochem.J. Herring. 1991. F. Corbella and C. Halberstam. Biol. 136 J.G. 131 J.119.73. Vo. G.M. Shechter.35. Yuen. Orvig.279.. Roberts. Yuen. Guyda and B. J . A. S. Diabetes. Sabanay. Deragon.55. James.H. Vogel.L. Endocrinology.369.C. Halberstam.35. Vihko.134. M. 144 A. 1992. J . Glover. Chem. Lett.M.45. 1996. 1995. Metab. Pharmacol. Shlimovich. Keen. S. So0 Lum and B.D. Bevan. Chem.. Posner..B. Domingo. M. Chem. J.S. Natl. Leroith and Y.I.J. Posner. Shechter. Gomez.B.. M.T.G.A. I. Yale.G. Pleasic-Williams. Serrano. Caravan. Messerschmidt and R. 1994. 129 A... 0. 1991. Toxicol. Sakurai. Vet. Gelmini. J. 151 Y.1667.J. SOC. Clin Invest. Orvig and J. 145 S. 1994. J . E. Mol. 1992. 192. D. L. 1995. Clin.555. 1993. Am. J. Shamoon. H. McNeill. Kadota. Res.659.49. Ramanadham. 110..H. 1995. personal communication.982.R. V. Anderson. 1991. Carton. E. Sun. Yuen and C. P. Llobet and J.L. Posner.35. Setyawati. Geiger.84. Chang. Toxicol.R.12 759. 1243.23. Gomez. S. Corbella. 569.876.M. Shuter.M. J . Diabetes. K.H. 1992. Biochemistry. 126 A. H. 0ngembaandD.108 Vanadium Compounds as Possible Insulin Modifiers 124 J. Llobet. Tracey. Siou. Rettig and C.8. Shamoon and L.5447. 1994. Chem.97.. Commun. Sun and C. 1995. . Becker. Llobet and J. P. A. L. J. Lyster. Corbella and C. 66..I. Shaver and B.. 1993. J.74. M.I.249. J. Y. Human Toxicol.G. McNeill. Fantus. Pharmacol.L.67. D.. 132 J. O. A.R. 1984. J. Sun. Gomez. Shanzer.68. Med. Lindqvist.6507. V. Siminson. Sci. Rossetti. Shaver. Pharmacol.32. Y.507.H. G.B. 3311.221. L.1997. N. Chem.93.. Keen. J .. 142 H. Zeisler. 0. Drake. and H. F. Domingo.. Diabetes. T. Giaccari. Kahn. 141 V.H.. A. SOC. 125 S.P.31. J. 139 G. Domingo. Setyawati.392. Hadari. Vian.R. Trace Elements Med.. Henquin.80..

Comparative clinical properties of cisplatin and carboplatin are summarised in Table 1. carboplatin is the only platinum analogue to have received worldwide registration.CHAPTER 7 Cisplatin-based Anticancer Agents LLOYD R. 2 Clinical Properties Cisplatin was introduced into clinical practice in 1971 (only some five years after the initial discovery of its cell-killing properties). in 1981. Interestingly. 15 Cotswold Road. Before 1975.’ This chapter describes the clinical properties of cisplatin. Surrey SM2 5NG. cis-diamminedichloroplatinum(~~)]. its analogue carboplatin [CBDCA. Following the introduction of cisplatin into a regimen also containing vinblastine and bleomycin (PVB regime). and of its analogues that have undergone extensive clinical trials (particularly carboplatin). and. To date. the chemical identity of cisplatin was first established in the mid-19th century as Peyrone’s chloride. carboplatin. the mechanisms of tumour resistance to cisplatin and their circumvention. platinum chemistry relating to mechanism of action. was rare. Sutton. who are predominately young adults. Initial studies at the Royal Marsden Hospital (London) by Wiltshaw and . Undoubtedly the most dramatic impact of cisplatin has been observed in men presenting with testicular cancer. The Institute of Cancer Research.2 around 85% of these patients now are essentially cured of the disease. KELLAND CRC Centre for Cancer Therapeutics. cis-diammine(1’1and cyclobutanedicarboxylato)platinum(~~)] among those most frequently used are today. its antitumour properties might have remained unknown. Belmont. the cure of such patients. but for the fortuitous and welldocumented observations of Barnett Rosenberg while performing experiments to investigate the effects of electric fields on bacteria. and the less toxic analogue. UK 1 Introduction The anticancer drugs based on platinum: cisplatin [CDDP.

4 Cisplatin also confers a major palliative effect in patients with small-cell lung carcinoma.' Cisplatin versus Carboplatin Numerous clinical trials have demonstrated that carboplatin is substantially less toxic (especially in terms of nephrotoxicity and gastrointestinal effects.v.daily for 5 days.110 Cisplatin-based Anticancer Agents Table 1 Comparative clinical properties of cisplatin and carboplatin C isplat in T y pica 1 doselsc hed ule 100mgm-2. ~Notably. head and neck carcinoma.~' . q 3-4 weeks Carboplatin Major toxicities Nephrotoxicity Severe nausea and vomiting Neurotoxicity (peripheral neuropathy) 0tot oxicity (tinnitus/hearing loss) M yelosuppression (mainly thrombocytopaenia) Pharmacokinetics22 (A) Total platinum t112 * hi2 20-40 min 44-190 h (B) Free (ultrafilterable) platinum tll2 a 22-78 rnin tll2 P not seen B 1-3 h 6. q 3-4 weeks or 20 mg m. i. secondary responses to .2 . Combination regimens including cisplatin (typically with cyclophosphamide) produce clinical complete remissions in approximately 50% of patients with advanced disease.&' However. i.7. q 3-4 weeks 400 mg m . Accrued long-term survival data from the Netherlands for women presenting with advanced ovarian cancer have shown that combination chemotherapy with cisplatin can improve survival rates at 5 and 10 years by more than 10% as compared with the best available treatment of the pre-cisplatin era. bladder carcinoma.v. it also appears that the two drugs are effective against the same population of tumours and thus share cross-resistance with one a n ~ t h e r . Overview analyses of randomised studies comparing the activity of cisplatin versus that of carboplatin (mainly in ovarian and testicular cancers) have concluded that the two are broadly comparable in terms of response rates and disease-free intervals. see Table 1) than is cisplatin.> 24 h 87 min 354 min 65 24 h Urinary platinum excretion (% of dose)22 16-35 colleagues established that cisplatin also conferred promising activity against ovarian cancer. however. or cervical carcinoma.

rather than the square-planar configuration of cisplatin and Cisplatin Carboplatin . less nephrotoxicity but antitumour activity comparable to that of cisplatin.cPt H C H lproplatin CI OCOCb DadFplatinum e.e. like carboplatin. cis-dichloro-trans-dihydroxo-bis(isopropy1amine)platinum(1v).12this issue remains contentious.' Iproplatin was the first quadrivalent platinum(1v) complex possessing an octahedral configuration.Lloyd R. Iproplatin Iproplatin [CHIP. carboplatin and iproplatin) or are new drugs currently in early clinical trials . response rates increase with increasing progression-free intervals. both with respect to cost and because many of the randomised studies of cisplatin versus carboplatin are still. see Figure 1 for structure]. was selected for clinical evaluation because of its favourable efficacy profile in preclinical studies. in relatively early stages without long-term follow up. Omplatin JM218 Figure 1 Structures of platinum drugs that have undergone extensive clinical trials (cisplatin. Kelland 111 either cisplatin or carboplatin may occur in patients with ovarian cancer who have previously responded to the drugs.g. i. '' While some have advocated replacing cisplatin with the less toxic carboplatin as first-line treatment for all patients with ovarian tumours. in the context of clinical trials.

Ultrafiltrates are usually prepared by centrifugation (1000g. Its future clinical role is limited.I n vitro studies with plasma showed that. median survival values were. iproplatin was more toxic (especially to the gastrointestinal tract) and significantly less For example.2 every 4 weeks)? In conclusion. that entered clinical trials. cut-off of approximately 50 000). but shares cross-resistance with cisplatin and carboplatin.22pharmacokinetic studies have focused on the fate of free (ultrafilterable) platinum. iproplatin appears to be less active and more toxic than carboplatin. The three major conclusions from these clinical trials are: (a) Iproplatin has a toxicity profile similar to that of carboplatin. 30 min. respectively. Studies have determined either 'total' platinum levels. in a randomised trial of 120 patients with advanced ovarian cancer. with myelosuppression (mainly thrombocytopaenia).l4 (b) Iproplatin does confer responses in 'platinum-sensitive' tumour types such as ovarian cancer. transcribing versus nontranscribing genes). and ELISA techniques utilising p~lyclonal'~ monor oclona12' antibodies to detect various DNA intrastrand crosslinks. The measurement of platinum following therapeutic doses of drug (levels down to about 5 ng ml.008) for carboplatin (400mgm-2 every 4 weeks) and iproplatin (300mg m . Determination of Platinum Drug Levels and Pharmacokinetics Platinum levels in cells are generally measured by flameless atomic absorption spectrometry (FAAS). in randomised trials with carboplatin. as platinum bound to plasma proteins is not cytotoxic. it is the only platinum drug besides cisplatin and carboplatin to have undergone extensive (including phase 111)clinical trials. while cisplatin decomposes rapidly (half- .e. 114 weeks and 68 weeks (p = 0. which include drug bound to plasma proteins. Methods to measure cisplatin-induced DNA lesions include alkaline filter elution. and that specifically measures the absorbance of platinum at a wavelength of 265. The pharmacokinetics of cisplatin and carboplatin are clearly different and reflect largely the differences in chemical stability of the two drugs (Table 1). DNA-protein crosslinks (and DNA strand breaks)." which detects DNA interstrand crosslinks. 4 "C) using Amicon ultrafiltration cones ( M .'4-' (c) Most importantly. In addition.21 Identification of individual metabolites (and intact drug) may be determined using high-performance liquid chromatography (HPLC) separation and subsequent analysis by mass spectrometry.can be detected) is now achievable. techniques have recently been described that measure binding and removal of DNA crosslinks in specific regions of the genome (i. being dose-limiting. a technique that uses high temperatures to atomise molecules. To date. However. or 'free' (ultrafilterable) levels.9 nm.112 Cisplatin-based Anticancer Agents carboplatin.

and two located axially.g. The presence of high chloride ion concentrations in extracellular fluid (approximately 100mM) is considered to suppress the aquation reactions and allow the uncharged complex to penetrate .Lloyd R. carboplatin binds much less avidly to plasma proteins (around 24% protein-bound 4 hours after infusion) and is excreted primarily via the kidneys by glomerular filtration. In contrast. tetraplatin. 3 Platinum Chemistry Platinum exists in two main oxidation states. carboplatin is more stable (half-life of 30 hours). the platinum atom has four bonds directed to the corners of a square plane at which the four ligand atoms are located.those to nitrogen or sulfur) are essentially irreversible under physiological conditions. Pt2+ and Pt4+. RNA.22 Two independent studies have demonstrated an excellent relationship between renal function (as assessed by glomerular filtration rate. In Pt(I1) complexes such as cisplatin and carboplatin. ~ ~ .' for previously treated patients and 7 mgml-' min.for previously untreated patients. respectively. Kelland 113 life at 37 "C for binding to proteins of around 2 hours).' min. the majority of the plasma platinum is accounted for as the intact drug over the initial 4-6 hours after administration. cisplatin and transplatin (which is inactive). The stereospecificity of the bonds is also of importance. directly above and below the platinum. The final positively charged. Thus. Consequently. in Pt(1v) complexes (such as iproplatin. observations then led to Calvert and colleagues developing a simple dosing equation for carboplatin based on pretreatment kidney function:23 Dose(mg) = AUC x (GFR + 25) Use of this formula allows the adjustment of the dose of carboplatin according to renal function in order to produce optimal AUC values of 5 mgml. there are six bonds and ligands: four in a square-planar configuration. as exemplified by the contrasting biological properties of the two isomers.22 Cisplatin is inactivated primarily by avid binding to plasma proteins ( > 90% is protein-bound 4 hours after infusion). GFR) and the area under the plasma platinum concentration versus the time curve (AUC) for carboplatin (and hence the therapeutic efficacy and the severity of thromb ~ c y t o p a e n i a ) These~ ~ .25 The chemical properties of platinum coordination complexes are largely dependent on the relative displacement reactions of the various ligands. Compared with cisplatin. the stability of bonds to halogens and especially to aquo (H20) is much lower.usually designated Pt(I1) and Pt(Iv). thus producing an octahedral configuration. see Figure l). cisplatin reacts primarily by stepwise exchange of its two labile chlorides (leaving ligands) for water or hydroxyl ions (Figure 2). and JM216. or proteins. highly reactive diaquo species (which is the same for cisplatin and carboplatin) is then capable of reacting with nucleophilic sites on DNA. While some platinum bonds (e.

114

Cisplatin-bused Anticancer Agents
DNA cmss link btmatii and m w l

- c-c-

-T--C-

membrane

Figure 2 Mechanisms of action and tumour resistance to cisplatin

cell membranes. However, upon entering the cell, where the cytoplasmic chloride concentration is much lower (as low as 4mM), the chloride ligands begin to exchange. Replacement of the two chloride ligands of cisplatin by the bidentate cyclobutane dicarboxylic acid (CBDCA) ligand in carboplatin results in a complex more than 100-fold as resistant to the above aquation reactions; in chloridefree phosphate buffer, pH 7, cisplatin had a half-life of 2.4 hours, compared with a half-life of 268 hours for carboplatin.26

Mechanism of Action
The cell-killing effects of cisplatin (and carboplatin) appear to be due to the formation of various stable bifunctional adducts on DNA, which then block replication or inhibit transcription. Supportive evidence for DNA as the critical target for the antitumour activity of cisplatin is provided by observations that cells from patients with diseases where DNA repair processes are deficient (e.g. Xeroderma pigmentosum) are hypersensitive to ~ i s p l a t i nIn~addition, correla. ~ tions have been shown between levels of platinum-DNA adducts in peripheral blood cells (leukocytes) and disease response in patients receiving cisplatin or carboplatin.28 The nature and proportions of some of the various cisplatin-induced adducts on DNA are shown in Figure 2. These have been determined either by in uitro incubation of DNA and cisplatin or by extraction of DNA from cells exposed to cisplatin followed by separation and analysis of digested platinum-DNA adducts by HPLC.29i30The most common adduct (60-65%) involves binding of platinum to the nitrogen in position 7 of the imidazole ring of adjacent deoxyguanosines along the same strand of DNA (the GpG 1,2-intrastrand adduct). In addition, ApG (20-25%), GpXpG and ApXpG (5-6%; where X is any base) 1,3-intrastrand adducts, monofunctional adducts (2-3%), DNA-protein crosslinks, and G-G interstrand crosslinks (ISCs, 2%) are also found. Once bound to DNA, the adducts induced by cisplatin and carboplatin are similar in

Lloyd R. Kelland

115

nature. The large difference in the concentrations required to produce equal numbers of adducts and similar cell-killing effects reflects the very much faster aquation rate for cisplatin.26 Perhaps surprisingly, the relative role of each of the various DNA adducts induced by cisplatin (especially intra- versus inter-strand adducts) in mediating cell-killing remains unclear. On one hand, there is supportive evidence emphasising the importance of 1,2-intrastrand adducts. The inactive transplatin is sterically unable to form the major GpG and ApG 1,2-intrastrand adducts formed by cisplatin. Instead a high proportion of DNA monoadducts are formed (up to 85% of adducts following a 1-2 hour drug incubation with DNA).3' These are mainly detoxified through rapid reaction with glutathione. A minority slowly rearrange to form bifunctional 1,3 or 1,4 G-G intrastrand crosslinks or DNA interstrand crosslinks. Moreover, experiments have shown that the GpG intrastrand adduct is poorly repaired compared with GpXpG and the monofunctional

ad duct^.^ 2i3
Although they represent only approximately 2% of the total adducts formed by cisplatin, other findings suggest that ISCs may be important determinants of cytotoxicity. Early studies in L1210 leukaemia cells, again using cis- and transplatinum, showed a relationship between cell killing and levels of I S C S . Al~~ though both isomers produce ISCs in naked DNA, only cisplatin produces substantial ISCs in whole cells. Furthermore, studies determining repair at the level of individual genes in paired cisplatin-sensitive and -resistant human ovarian carcinoma cell lines have shown a marked increase in the gene-specific removal of ISCs in resistant lines, but no difference between sensitive and resistant lines in the removal of intrastrand a d d ~ c t s .Recent ~ ~ have also ~ ~ , data indicated that the nature of the ISCs formed by transplatin and by cisplatin in purified DNA differs, cisplatin favouring crosslink formation between guanines, and transplatin between guanine and complementary cyt~sines.~' Studies using both bacterial enzymes (Escherichia coli UVrABC n ~ c l e a s e ~ ~ and mammalian cell extracts3') indicate that platinum-DNA adducts are removed from DNA by nucleotide excision repair. The E. coli UVrABC nuclease has been shown to incise the eighth phosphodiester bond 5' and the fourth bond 3' to G-G intrastrand crosslink, thereby excising an oligomer containing the a d d ~ c tThere is also considerable understanding of the molecular aspects of .~~ platinum-DNA interaction^.^' Interestingly, both the G-G and A-G intrastrand adducts unwind DNA by 13", whereas the GXG 1,3-intrastrand adduct unwinds DNA by 23"; bending of the DNA double helix is similar (32-35") for all three ad duct^.^'

Structure-specific Damage-RecognitionProteins
Two classes of proteins ( 28 kilodaltons and 80-100 kilodaltons) have been identified in mammalian cell extracts that bind specifically to the two major types of cisplatin-induced DNA 1,2-intrastrand crosslink^.^^^^^ The proteins do not bind to adducts produced by the inactive t r a n ~ p l a t i nA~ ~ that encodes . gene one protein of molecular weight of approximately 81 kilodaltons (termed struc-

-

116

Cisplatin-based Anticancer Agents

ture-specific recognition protein, SSRP1) has been cloned and sequenced; it shares a region of homology with the high mobility group proteins, HMGl and HMG2, the HMG It has also been shown by two independent groups that HMGl and HMG2 (proteins of 28.5 and 26.5 kilodaltons, respectively)can themselves recognise and bind to DNA platinated with ~ i s p l a t i n . ~ ~ * ~ ~ The exact function of these damage recognition proteins, however, is currently unknown. Models have been proposed for their involvement in repairing cisplatin-damaged DNA or, somewhat to the contrary, in blocking access of repair enzymes to damaged DNA. Some support for the repair supposition is proved by observations that certain cisplatin-resistant cell tumour lines (some of which have an increase in DNA repair capacity) also exhibit an increase in damage recognition In addition, a damage recognition protein complex, B1, has been shown to contain human single-stranded binding protein (HSSB),49 a protein involved in an early stage of mammalian excision repair.” However, studies using other cell lines with resistance to cisplatin have found no obvious differences in levels of damage recognition proteins or in HSSB protein levels.42*49i5 Furthermore, support for the model involving shielding of intrastrand crosslinks from DNA repair enzymes has arisen from studies in yeast, where a gene, lxrl, encoding a structure-specific recognition protein, when inactivated, resulted in a twofold decrease in sensitivity to ~ i s p l a t i n . ~ ~

4 Mechanisms of Resistance to Cisplatin/Carboplatin
Over recent years an intensive area of study has sought to determine mechanisms of tumour resistance (both intrinsic and acquired) to cisplatin/carboplatin. It is well established that resistance limits the clinical efficacy of the currently available platinum drugs, and thus an understanding of resistance mechanisms should aid the discovery of either improved platinum drugs or new strategies of modulating the efficacy of cisplatin/carboplatin. Studies have focused largely on the use of in vitro murine and human tumour cell lines and have revealed that resistance is often multifactorial in origin. It may arise through one or more of three major mechanisms: namely, decreased intracellular transport of drug, cytoplasmic detoxification through increased levels of thiol-rich species such as glutathione and/or metallothioneins, and enhanced removal of platinum-induced adducts from DNA and/or increased tolerance to platinum-DNA adducts (Figure 2: A, B, C).54 Many acquired cisplatin-resistant cell lines exhibit a decrease in platinum accumulation (typically 2- to 4-fold) as compared with their respective parent lines. However, the mechanisms by which platinum drugs enter cells are still unclear, with laboratory evidence supporting roles for both passive diffusion (e.g. the uptake of cisplatin in not saturable) and active transport (e.g. uptake can be modulated by a variety of pharmacological agents, including the N a + K + ATPase inhibitor, ouabain). Notably, in contrast to the increased efflux of drug observed in tumour cells with multiple resistance to other commonly used anticancer drugs (e.g. doxorubicin, paclitaxel, etoposide and the Vinca alkaloids) mediated through 170 and/or 190 kilodalton membrane proteins, cisplatin resis-

60*61 There is some evidence to suggest that this hypersensitivity. Kelland 117 tance mediated at the level of the plasma membrane occurs mainly through reduced drug influx. involving 2 mol of GSH complexed with 1 mol of platinum to form bi~(g1utathionato)platinum. and a 60 kilodalton heat-shock protein (HSP60).Lloyd R.to 8-fold increase in cisplatin r e ~ i s t a n c eCisplatin has been shown to induce .” As with reduced drug accumulation. Increased levels of metallothioneins have been reported in at least some acquired-cisplatin-resistant cell lines.. and BCL2/BAX). modulation of various intracellular signal transduction pathways (e.68 . studies using tumour cells derived from testicular cancers indicate that they are hypersensitive to cisplatin. the most resistant line possesses GSH levels 4-fold higher than the most ~ensitive.(or carboplatin-) based therapy had expression levels of ERCCl in their tumour biopsies 2. at least some pathways of cell death appear to be activated through the wild-type p53 gene. p21RAS.~~ cell death through a programmed death pathway (apoptosis).6-fold higher than those for patients who responded to that therapy. including those in our own studies.~ Elimination of this complex from tumour cells has been proposed to occur via an ATP-dependent glutathione S-conjugate export pump. many in vitro studies have produced evidence of a role for the cytoplasmic thiol-containing tripeptide glutathione (GSH) in mediating platinum drug resistance. appears to be due to enhanced repair/removal of platinum adducts from DNA. M YC.63Transformation of mouse fibroblasts with ras oncogenes produced a 4.58 The acquired resistance to cisplatin in many cell lines.’~ Evidence has also been reported for a direct interaction between cisplatin and GSH inside tumour cells. those mediated by protein kinase C or via the epidermal growth factor) may also influence sensitivity and resistance to cisplatin.67Finally. ~ ~ ~ ~ ~ Other genes/proteins that may be involved in determining sensitivity to platinum drugs include various oncoproteins (e.54 Moreover. the possible relevance of repair in determining responses to platinum-based therapy in ovarian cancer patients has been highlighted in studies measuring RNA levels encoding for the ERCCl human DNA repair gene. In our own studies using eight human ovarian carcinoma cell lines exhibiting a 100-fold range in intrinsic sensitivity to cisplatin. which correlates with the clinical sensitivity of such tumours. a class of cysteine-rich. as compared with its parent line. whereas tumours less responsive to chemotherapy commonly acquire p53 mutations. the p53 tumour suppressor gene. low molecular weight isoproteins. In addition.66 Elevated constitutive levels of HSP60 have been observed in a human ovarian carcinoma cell line with acquired cisplatin resistance. testicular tumours have been reported to exhibit p53 mutations only rarely.g.g. GSH levels showed a significant correlation with cisplatin sensitivity.65 Interestingly.’ Cytoplasmic detoxification of platinum-based drugs may also occur through binding to metallothioneins.59Patients who were clinically resistant to cisplatin. might also be related to defective removal of platinum-DNA a d d u ~ t s .

2-diarninocyclohexaneplatinum(1v). who showed that platinum complexes of this class retained activity in a cisplatinresistant murine leukaemia (L1210) model?’ Following chemical refinement (primarily to improve aqueous solubility).118 Cisplatin-based Anticancer Agents 5 Circumvention of Tumour Resistance to Cisplatin Although the discovery of carboplatin has unquestionably resulted in reduced morbidity for patients receiving platinum-based chemotherapy. tetraplatin (ormaplatin) and oxaliplatin [oxalato-1. In addition. and the Section of Drug Development at the Institute of Cancer Research (Sutton) has resulted in the discovery of a novel class of . Diaminocyclohexane (DACH) Platinum Complexes Interest in platinum complexes based on the 1. such as the L1210. administration of higher doses of cisplatin/carboplatin.'^*'^ Unfortunately. particularly with regard to circumvention of tumour resistance.2-diaminocyclohexane (DACH) carrier ligand [e.g. there remains an urgent need for further improvements in clinical efficacy. see Figure 13 arose from observations by Burchenal and colleagues. there have been relatively few leads to the discovery of novel platinum drugs capable of circumventing tumour resistance to cisplatin/carboplatin. and combination with other pharmacological agents capable of modulating one or more of the known mechanisms of resistance to cisplatin. Four broad strategies may be envisaged for attempting to circumvent resistance: development of more broad-spectrum platinum drugs. an important caveat for adopting a single acquired-resistance murine tumour model.~ Orally Active Platinum Complexes: JM216 The success of carboplatin calls attention to the importance of pursuing possibilities for improved patient quality of life in the treatment of cancer. Development of New Platinum Drugs Despite the synthesis of many hundreds of cisplatin analogues over the past 20 years. severe neurotoxicity appears to represent a major limitation with ormaplatin and to a lesser extent with oxaliplatin which has recently exhibited some activity in colorectal cancer when used in combination with 5-fluorouracil. A collaborative programme among the Johnson Matthey Technology Centre.2-diaminocyclohexaneplatinum(11). Bristol Myers Squibb.1. tetraplatin. two DACH-based complexes. Ormaplatin. As both cisplatin and carboplatin require intravenous administration. patient comfort could be further facilitated by use of an orally available platinum complex. tetrachloro. 1-OHP] entered clinical trial^. combination chemotherapy with other active anticancer drugs. for platinum drug development has been highlighted by studies where ormaplatin conferred no activity against another cisplatin-resistant murine model (the ADJ/PC6)72 and exhibited poor activity against a variety of human ovarian carcinoma xenograft~.

to date. in the hope of achieving improved response rates.1. ~ ~ ~ compound. shown convincing evidence of activity against cisplatin-resistant tumours. Dose Intensification of Cisplatin/Carboplatin In recent years.JM216. Other Platinum-basedAgents Other platinum-based drugs currently undergoing early clinical evaluation are predominantly carboplatin-like. Our own data. and 5-HT3 inhibitor antiemetics. 54 Examples include CI-973 [cis. using a panel . myelosuppression has been the dose limiting toxicity. or n e u r o t ~ x i c i t y . loboplatin [1. and DWA2114R [2-aminomethylpyrrolidine-( 1.Kelland 119 platinum(1v) complexes. Worldwide phase II/III clinical trials are now ongoing. Figure 11. Toxicology studies in rodents showed that JM216 possesses carboplatin-like.75Antitumour effects appeared to be schedule-dependent. ~ ~ .~ ~ given JM216 orally showed comparable or superior activity to that observed for intravenous cisplatin or carboplatin against a variety of murine and human tumour models. hepatotoxicity. disabling neurotoxicity has proven to be dose-limiting.79~80 JM216 has now undergone phase I clinical evaluation according to both single dose and daily x 5 schedules. versus the same total dose given as a single bolus every 21 days.or neurotoxicity.Lloyd R. with greater activity being observed against a human ovarian carcinoma xenograft when a split daily dose (for five days every three weeks) was given.83 As the main toxicities of carboplatin are haematological. properties. with no observations of nephro. as well as promising in vitro cytotoxic effects against human ovarian carcinoma cells exhibiting intrinsic resistance to c i ~ p l a t i n One~such ~ . the ammine/amine (or ‘mixed amine’) platinum(1v) dicarboxylates. NK 1211. Although the renal and gastrointestinal toxicities of cisplatin may be ameliorated through intravenous hydration and forced diuresis. entered clinical trials at the Royal Marsden Hospital (Sutton) in 1992 as the first orally administerable platinum drug.~’ carboplatin may be better suited to such studies (especially in combination with haematological support with growth factors). D-194663. JM216 has exhibited preclinical circumvention of acquired cisplatin resistance.l-cyclobutanedicarboxylato)platinum(II) monohydrate].4-butanediamineplatinum(I1).*lls2In agreement with the rodent data. and dosage calculation based on kidney function are appli~able. it remains controversial whether increased doses will be reflected in improved patient survival. [bis(acetato)(ammine)(dich1oro)-cyclohexylamineplatinum(Iv). attempts have been made to increase the doses of cisplatin administered to patients.84 However. rather than cisplatin-like.1.78 Moreover. and have not.2-diarninomet hylcyclobutaneplatinum(~~)lactate. and especially of resistance mediated through reduced drug accumulation. respectively.l-cyclobutanedicarboxylato( 2R)-2-methyl. with myelosuppression dose-limiting and with no nephrotoxicity. which demonstrate promising oral activity against a variety of murine and human tumour preclinical models.

178. however. 6 Summary Over the past 25 years. 2 L. Coordination Complexes in Cancer Chemotherapy.. resistant to cisplatin.. 1977. Donohue. However.120 Cisplatin-based Anticancer Agents of human ovarian carcinoma cell lines. Roberts. 1974. Cancer (Phila). 3 E. Carr. p.293. Heidelberg. have shown a difference in intrinsic cellular sensitivity to carboplatin and cisplatin of 30.D. either through the discovery of additional platinum-based complexes possessing a wider therapeutic spectrum of activity (see Chapter 8) or through pharmacological modulation of platinum-resistance mechanisms. has shown promising levels of activity. Modulation of Platinum Resistance Mechanisms To date. Annals Intern.J.87. attempts at the modulation of platinum-resistance mechanisms have generally been conducted in preclinical models and have not reached the stage of clinical trials. the platinum-based drugs (cisplatin and. but in recent years.and 100-fold. Einhorn and J. Connors and J. Rosenberg.88 and at least some may reach clinical trial in the near future. paclitaxel (taxol). there remains scope for substantial improvement in the clinical utility of platinum-based metal coordination complexes.H. ed. the less toxic analogue carboplatin) have brought significant therapeutic benefit to a large number of cancer sufferers.86 Response rates in the region of 20-30% have been reported. References 1 B. have shown promise. reduction of GSH levels using buthionine sulfoximine. . respectively. A diverse range of agents. 1985. Wiltshaw and B. or have become. Examples include modulation at the level of the plasma membrane using the antifungal agent. Combinations with Other Anticancer Agents: Paclitaxel The vast majority of currently available anticancer drugs do not exhibit significant activity against ovarian tumours that are. Med. The likelihood of achieving these important goals has undoubtedly been enhanced through our vastly increased understanding of the chemistry and the biochemical properties of cisplatin. Clinical studies combining cisplatin and paclitaxel have been completed including a pivotal study in advanced ovarian cancer where the paclitaxel/cisplatin combination showed a marked improvement in survival compared to the conventional cyclophosphamide/cisplatin arm. in Platinum.A. latterly.85Thus dose escalations of a similar magnitude may be necessary in the clinic. a natural tubulin-binding compound extracted from the bark of the Pacific Yew. T. Springer-Verlag. and inhibition of DNA repair using aphidicolin.552303. amphotericin B.

J.. H. 1991. Thill. USA..P.A.T.123. J. Sancar and V. Yuspa. D. Kelland 121 4 J.7. 1994. J .1367. Oncol. Erickson and L. 1991. van Lent. R. H.J. Rozencweig. A. Hines. Natl. Dawson. Cancer. Clagett-Carr and B. J. 1979. Acad. Kohn. D. 12. L.D.Y. Hakes. Canetta. Kaye. McGuire 111.E.J.. R. Almadrones and J. M. J.T. Eastman. C. Sessa. J.H. J. Harding. ed.A.C. E. Reed. K. Tobias and R. Chadwick. 1990. Arseneau. Weiss. Clin. Hoskins.R. Siddik. J. Given. 1991. P. New York. Willemse. 1988. D. McBrien and T. Taylor.Lloyd R . S. Cancer Res. 1992.DiSaia.A. J .395. Philadelphia.303.C. Annals Intern. E.5024. J .L.E. Burnell. van der Burg. Irwin. 1984. 6 Advanced Ovarian Cancer Trialists Group.P. Van Echo. Newell. Fisher. Kerrigan. M.767.63.. Ushay.I. Alberts.J. B. Clin. D. Pierce. W. Ash.S.. A. 379. R. Yuspa and M. 13 B. Boxall. I. E.C. 16 C. Pinedo.44. M.44.J.. E. 19 M.E. Roberts.Stewart.J. 18 K. M. Sci. R. 1986.C. 1748.M. Heintz.Y. 28 E. B.A. L.C. S. Calvert and K. Robinson and A. Oncol.J. D. D.P. Markman. Bohr. J . L. Oncol. Friedlos and D. Whitacre. Swerton. Acad. Pharmacol.W. Z. Reed. Cancer. Poirier. Zhen.A. W. Tipping. Grunberger and S.J.. R.R. Foster. Bunn. W. Z. 21 J. Parker. 17 W. S. 1989.1499. Oncol. 1982. J. S. D. Rubinstein. Clin. 1989. Siddik.J. Smith. A.H. J. Clin. Alberts. 29 A.H. p.135. Cancer Res.T.5432. Tilby. R.L. Hanawalt. Harrap. Green. Knox. W. Mason-Liddil. Roberts.. A. Tarone.M. Einhorn. Neijt.. Gore and E. Creasman. M. A. Judson.51. 2066. F. E. Oncol.M. in Handbook of DNA repair. Lydall and J.S. 120. Dean.B. Med. B. Eisenhauer.6.B. D.E. Wiltshaw. Fryatt and C. J. R. Rubin.. Aisner. 26 R. Cancer Treat. Bakker. C. Oncol. Newell. Cordell. Calvert. L. 1991. R. Marcel Dekker Inc. I. S. Med.W. Oncol.T.J.289. de Meijer. 27 J.J.B. Hanson. E. J .N. Ewig. Cavalli and H. R. USA.R. 24 M. Chem. 1990.J. S. F. in Carboplatin. Hatch. H. R. J.J.A.A. Goodwin.C. 8 A. S. S. 29. W. Ozols. J . Silverstone.3912. p..E.H. Jr. Cancer Res.1462.H. O’Reilly and R..100. Friedlos.7. Wolpert-DeFilippes. ten Bokkel Huinink.L.46. Gore. Trimbos and A. Cancer Chemother. Olman. Slater. 1986. T. Zwelling.J..7101. Proc. 1991. 1981. 1972. M. H.R. O’Reilly. J. Lippard. C. J. K.25. Saunders Company.W.M. Teng and W. 20 M. Sci. C. Lydall. . R.S. Semin.54.9.J. Oxford. Rep. Gumbrell. Cancer Res. Ozols and M. D. Biol.M van Lindert. 14 C. Proc. Blessing. Clin. Poirier. K. p. Biochemistry.G.F. L.W. K.266. Connors. 12 D. 23 A. Current Perspectives and Future Directions.A. Williams.98. Wiltshaw. Canetta. 10 M. van Oostersom. Thompson and J. J .25.A. Oncol. Natl.B. Earl. IRL Press. ten Bokkel Huinink. M. 1991. 1986. Clin. Parmar and D.R. A. Semin. Aartsen. Cleare and K. Vermorken. W.H. Jones. Fryatt.K. Calvert.11. 15 G.A. P. Egorin.. 1984.. 1991.17. ed.6443. Knox. 9 E. 133.H. F. Reichman. 1131. Santella. Leyland-Jones. Roberts and C.J.27. Canetta and N. Wiltshaw.. F.19(Suppl2). Eur. Baker and J.C. I. Zwelling. Jones. F. 5 P. L. 1987.. Sturgeon..F. R. T.A. Renard. Souhami. Gore. A. Guthrie. N.J. Johnson.C.9. P. Br.J. Harland. Cancer.60. 1984.J. 11 M. 25 T. Br. Rothman.84. Knox.H. Lewis Jr. 1989. J . Eur. Harrap. Friedberg and P. J.C. Thigpen. J. L.884. 1693. 1990. B. Loehrer and L. Fine. 22 S. Trask. M. M.79.H.704. ed.H.J. H. 7 C. S. in Biochemical Mechanisms of Platinum Antitumour Drugs. Vermorken.

1994. G. Rev. Pil. Chao. R.3324.52. Cancer. O’Connor. Friedlos.. Proc.R. K. Chang. vol. Brown. Coren. 39 J. Leng. Fichtinger-Schepman.J.T. Mills and J. J. 31 A.A. Mistry. L.M. 1993. Science. Bellon.F. D. Acid Rex.D. 42 B. Br. Lazo.R. Page.D.J. Anderson and K. D. S. B.D. 1992. D. Essigmann.67.24. USA. Augot. M. 1992.M. S.64. Yeung. S . J. Sancar and S. Harstrick.8073.A. F. A. 62 B. Chem. K. Sci.J. Johnson. Hansson and R. Lippard. 48 K. J .. 1994. L. Thiebaut. Biol. S.D. Biol.26. Sci.89. 61 L. 41 G.S. Popoff.J. 36 S. 1171.1512.1. Chaney. Housman and S. 832.47.. 1993..K. Eur.. 1. Neidle and M. USA. R. Brabec and M. Kelley. Chu and E. Hamilton..29. 51 P.538.. Pera.. Cancer. 1993. Hacker.603. Science. 1993. 54 L.B. J . Nucl.. 1710. K. 21 5 . Lee and S . C.R..J. 55 D.-L. Natl.87. 1993. Toney. Cancer. Sidhar and K. B.47. Kelland.F.53. Bruhn. Ishikawa and F. Reed. 1991.W. M. P. 10772. den Hartog. Egwuagu and E.M. Waring. Pi1 and S.-K. Szymkowski.K.K. J. 44 E. S. Proc. Wood. Mann.742. Pharmacol. Hamer and J.. Scanlon. 57 T. J.M. Kelland. Biochem. Biochemistry. Acad. Kohn.15. 1992. Zhen. Pera. 1992.J.R.C. in Anticancer Drug-DNA Interactions.Y.J. L.89.52. J. 1987. Comess and S. J .N. Andrews and J.13.L. Isonishi. Parker. Reedijk. 45 P. Lohman and J. 1991.29. J . S.J.T. 47 C. Clugston. F. Acid Res. Masters and R. V.J.K. Chang. J . Lin-Chao.5345.A. 33 D.A. 1990.267.M. Howell.B.20116. S. 1993. 134.90.875. Abel.J.J. McLaughlin. 6375.A. Cancer Res. 32 J. Nucl. 38 D. 63 L. S. Acad. Natl. Hill. M. 59 M.H.M. Kenny. Friedlos.707. Rupp. Biochern. Lippard.P. Lippard. D. 52 D. J.W.3303. Zwelling. M. Lippard and J.241. Treiber. Oncol/Hematol. Harrap. Jones. Gately and S.E. Acad. S.S. Donahue. Ali-Osman.J. Cancer Res. P. Brown.K.A.17.C. Hughes. Hom.. 725. 1992. Essigmann. MacMillan Press Ltd. 50 D. J . Barry.B. L. 58 S.F. J. 49 C. 1993. Natl. Sancar and W. Kenny and R. Chem.122 Cisplatin-based Anticancer Agents 30 A. J . Teicher. A. R. Mistry.E. 1992.1813. 1991.M.30A. Eur. Nature. Harrap. Cancer Commun. Science.P. P. Basu. 37 V.277.J. Roberts. Cancer. Huang. Crit. 46 G. Cancer. Kelland and R. A. Howell and V. Brown. Abel. Mol.564. P.B. Lippard. Billings. Beck. G. Acad. Loh. T. Engelsberg and P. 1989.L. 365.349. 1990. 40 K. Wood. Kelland.A.L.R. E. M.A. Bohr.. 1985. McLaughlin. Lippard and R.A. J .191.242.662. G . K.M.P. W.84. F. Godwin. Shellard. Proc. Basu.67. Brown. Cell Biol.G. Eastman and M. 3689. Fichtinger-Schepman and S. Wood. Cancer. p. 64 S .H. Cancer Inst. Perez. London.5903.J. Coverley.H. 34 L.P. 1992. 1988. 1991.C. Br. 1993. USA.39.M. 35 W. Dabholkar. Husain. Howell.30A. Weber... 1990. Science.H. Link. 1992. 60 M. 1988. Bostick-Bruton. Br. C.689. 1991. Reed. Roberts and K. Cancer Res. Essigmann.12.5872.R. F. Yarema.261.J. J . Lane and R..234.M. A. Natl. Rupp. van der Veer. Cancer Res. Biochemistry. USA. S. Biochemistry. 1987.2307. Munn. Lazo. .. Proc. A.A. Hansson. 53 S. A.. M. Sci. I.J. Biochemistry.R. 6810.M. J.K. Chu and E. P. A. J. Kelland. Eastman and S. ed. Natl.D. Andrews. McLaughlin.A. A.268. J. 56 P. 1985.D. Bohr.256. Handel. Cancer Res.51.3. 43 S.13 520. Sci. 1994.1016. 7395.C. P.. Kellett and S. Ozols and T. J. Bissett. C. Int.-Y.N. D. 1979.

L. A. F. G. F. S. M.A. M. L. 71 D. 74 L. Schilder. Br. G. Valenti. P. Roby. 86 E. Judson.R. Goss.629. Kelland.J. Enns. S. K. M. Siracky. Seitz and M.R. B. 1993. 1997. Newel1 and M. S. E. Clarke-Pearson and M.l.J. D. L. Francois.R.R.E. Cancer.53. Ozols. ed. M. Harrap. Valenti. Wilson and K. McKeage. S.C.279.A. 69 J. A.B. P. p.J.E. 77 M.E. Goddard. Brennan. T.B.R. M. L. Look. M. N. Hogg. O’Dwyer.R.R. B. Clin. D.-J. Goddard. McGuire.Y.R.McKeage.. Harrap and I. J. E. 1995.R. N’Daw.G. Loh.H. McKeage. R. Ward. F.H.1493. M. Santabarbara. New York. Mistry.Lloyd R.451. R. Ychou.P. J. Hom.4118... Jones. P. J. Hamilton. Harrap and I. J.2581. Ellis. New Engl. 1992. E. Isonishi. Judson. S. Clin. Jewett. 84 A.32.P. J . Cancer Res.W. Murrer.L.66. Boxall. Friedman. McKeage..A. B. Annals Oncol.H. Rogatko.. Kucera. 1991. Davidson. Myers and R.2. J. Cancer. G. Harrap. Mistry and K. D.3. Kelland and K.R. J. 88 H. 1993.J. 1246. Harrap. Hawkins. Harrap.J. McKeage.36. M.E. D. C. M. D. Cancer Chemother.. F. 1993. Cell. P. Brady. 68 S.. Loh.M. Boxall.-F. Morgan. 1993. de Gramont. Ruley.M. Jacks and D. Fisherman. Lowe. Housman.R. Andrews. Young. Harrap. Brienza. Oncol. Marty..A. Murrer. K. S. Schilf. J .E. L. M. J.. P. D. Ozols and P. Metzger.R. Santabarbara. Calvert. Gamelin. Cancer. A. T.68. 70 R. 1991.996.J. 1996. Adams.52. F. J. Sastre.E.527.2691. B.957. H. Gallie. Melland.R.822. Peng. Bulbul. 1994. Kelland. Boxall. Giandomenico. 1992. S. Pharmacol.G. Cancer Res. R. Howell.E.M. . Mistry. Jones. Bailey. Br. Ostchega. 87 W. Johnson. C. R. Kelland. J.L. LaCreta.66. in Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy.C. 76 M. S. Mann and D. Ward. M. Murrer and K. Murrer and K. 19. Cancer. 1993.J.E. K. O’Neill. Thiebaut and S.R. 82 M. Dew and L.R. Itzhaki. Cancer Res. J . Med.A. de Vries.R. Semin. Harrap.. Hayn and S. Timmer-Bosscha.3574. Abel. Diaz-Rubio.R.67. J. J . 1992. F.Y. 66 H. Lokys. Christian. Oncol.334.R.54. Vena..C. Kimura. Harrap. R. J . G. Oncol. 72 P. B. R.15. Onetto. G. M. 75 L. Siracky. Abel. Cancer Treat. E.W.M..67. Cancer. Murrer. 1994. Arbuck. Christen. Clin.63. 1979. 85 C. Morgan.E.R. F. R. Berry.277. P. K. Valenti and K. Kelland. 73 M. J .. Kelland.C. J .24. Young.R. M. Rep. McKeage. Br. Goddard. 173. M. Mulder and E. Cancer Res.1109. L. Machover. B. B. Gwynne and K. Br. 155.C. R.74.54.L. L. Cancer Chemother.R. Cancer Rex. M. P.R. Jones and K. Abad. Abel. Howell.E. Kelland. Abel. Pharmacol.P. Raynaud.. N.J. Y. J.535.R.R. J. Odell.E. 81 M. Hard and K. Kelland and K.D.54. McAleer.95. 78 M. Oncol. C. C. P.-Q.C. Partridge. Howell. K. Canetta. W. 1989. Malkin. J.R. Boxall.F. 1994.F.E. P. Burchenal. S. C. D. 1992.E..S. 1985.7. 1996. J. 11. Mellish. Hoskins. Trimble.P.709. D. A. Br. Harrap.A. Annals Oncol. A. 83 R. G. Cancer. Buchanan and P. 79 K.A. Kelland 123 65 S.F. Harrap.M. Cancer Res. 67 E. 1993. Jones. Nash. Plenum Press.. Murrer. Br. 1993. Gastiaburu.C. J. Hills.59.53.J. Vignoud. J .E.R. P. Kalaher.J.240. 1993. Gore. D. 80 S.2405.B. Harrap. Perez.

Via Monza.). entered Phase 1 clinical trials. [cis-[PtCl.{ H.l-cyclobutanedicarboxylate)do not indicate. cisplatin. the first genuinely new platinum agent not based on the ‘classical’ cisplatin structure to do so. curable and is significant in treatment of ovarian and bladder cancers. a broader spectrum of activity in comparison with the parent drug. VA 23284.N(CH. by virtue of formation of different types of Pt-DNA adducts. epithelial ovarian cancer and small cell lung cancer as well as for palliation of bladder. testicular cancer. Monza.).).CHAPTER 8 Dinuclear and Tinuclear Platinum Anticancer Agents NICHOLAS FARRELL~ AND SILVANO SPINELLI~ ‘Department of Chemistry. BBR3464. 1001 W. Italy 1 Introduction In early 1998 a novel trinuclear platinum compound. The use of cisplatin (usually as a principal component of combination regimens) has rendered at least one cancer. lead to compounds with a spectrum of clinical activity genuinely complementary to the parent We considered that future discovery of clinically useful platinum agents was likely to . This latter consideration led us to challenge the empirical structure-activity relationships and formulate the hypothesis that development of platinum compounds structurally dissimilar to cisplatin may. A 1994 World Health Organization consultation document recognized 24 drugs as essential for rational management of advanced malignant neoplasms. gestational trophoblastic tumors.Boehringer Mannheim Italia. disappointingly.] units linked by a non-covalent tetra-amine [Pt(NH. Main Street.] (Figure 1). cis-DDP] is cited for treatment of germ-cell cancers.)]. CBDCA = 1. Its structure is based on the dinuclear concept developed in our laboratories and is best described as two [PtCl(NH.). cervical. Richmond. Despite considerable effort the consensus is that ‘second-generation’ structural analogues such as carboplatin ([Pt(CBDCA)(NH3).’ All direct structural analogs of cisplatin produce a very similar array of adducts on target DNA and it is therefore not surprising that they induce similar biological consequences.). nasopharyngeal. oesophageal and head and neck cancers. Virginia Commonwealth University. Of these.(NH. USA . Research Division.NH.].

Thus [(tran~-PtCl(NH.Nicholas Farrell and Silvano Spinelli 125 l. the geometry of the overall complex is automatically Jixed.l/t.thus BBR3464 contains one Cl on the end platinum units while the central unit is a tetra-amine which contains no (zero) Cl and is incapable o forming covalent bonds. are amongst the best studied. For a convenient abbreviation o polynuclear platinum complexes we have f adopted a system where the numbers refer to the number o chlorides (or anionic f leaving groups) on each platinum atom. variable chain length).t series (see Figure 1 for abbreviations) as having the most promising pattern of .t arise with ‘non-classical’ structures of which the subject of this chapter.lA. or a planar group such as pyridine or quinoline) and linker (flexible.t etc. Once these two parameters are specijied. geometry (leaving chloride groups cis or trans to the diamine bridge). Where there are two chlorides on the same Pt. We and our collaborators have extensively studied these new classes of platinum compounds in order to understand the patterns of DNA modification induced by various structural motifs and further relate these patterns to cytotoxicity and antitumour a ~ t i v i t yThis has been successful to the point where one compound has now .l/t.).lA. dinuclear and trinuclear bifunctional DNA-binding agents. The dinuclear structure is extremely flexible and capable of producing a wide series of compounds differing in functionality (bifunctional to tetrafunctional DNA-binding).t.~~~ advanced to full clinical testing.l/t. 2 Dinuclear and Trinuclear Platinum Complexes as Anticancer Agents During the last ten years.t. the lettering refers to the geometry with respect to the nitrogen o the bridging f diamine.O.t 1. as potential antitumor drugs. Since the f Cl atoms are trans to the diamine bridge the abbreviation is l. we have prepared a number of novel dinuclear Pt compounds. where two platinum-amine coordination units are linked by a variable length diamine linker. is I. Our initial synthetic efforts identified the l.)~)$p( CH.c BBR 3464 (1 .l/c.O.t) Figure 1 Structure for the trinuclear clinical drug BBR3464 and its dinuclear analogues. non-leaving groups in the Pt coordination spheres (NH. the lettering specijies their mutual geometries (cis or trans). Trinuclear agents are abbreviated in the same manner .NH2]C1. For those possibilities where there is only one chloride in a coordination sphere.

non-smallcell lung. In this section we compare briefly the antitumor activity of the lead clinical candidate with cisplatin in a range of human tumor xenografts. Summary and p53 Status of Human Tumors Treatable by BBR3464 The profile of preclinical activity of BBR3464 mirrors its unique structure and is characterized by activity in human tumor (e. where there is considerable historical precedent for cisplatin in both cases. The p53 protein is recognized as an important cell regulatory element necessary for cell cycle arrest and apoptosis induction. The 4+ charge.allowing for molecular weight differences. Biological Activity and p53 Status A highly relevant area of research interest is the p53 gene status of human tumors. The biological testing has been performed in collaboration with Boehringer Mannheim Italia. The general approach has been (i) to obtain cytotoxicity and in uiuo data in murine L1210 and human ovarian A2780 models both sensitive and resistant to cisplatin. a high activity in a broad spectrum of human tumors commonly insensitive to chemotherapeutic intervention (e. the presence of at least two P t coordination units capable of binding to DNA and the consequences of such DNA binding is a remarkable departure from the cisplatin structural paradigm and argue that the drug should be considered the first representative of an entirely new structural class of anticancer agents. Noteworthy also is the optimal dose of 0.g. we envisaged complexes with improved DNA affinity for long-range cross-linking. 3 Biological Activity of Polynuclear Platinum Complexes.~ more effectivethan cisplatin.3 mg/kgfor BBR3464 . Table 1 summarizes the overall profile from 18 human tumors t e ~ t e dIn almost all cases BBR3464 is . lung. As synthetic efforts progressed. ovarian) xenografts resistant to cisplatin and alkylating agents. Wild type p53 function .g. gastric) and characterized as p53-mutant. (ii) to extend the most promising compounds to testing in other solid tumor murine (M5076) and human solid tumor (IGROV-1 (ovarian). LX-1 (small cell lung) models and (iii) finally extended testing on a range of ovarian. With this advance the paradigm of cisplatin-based antitumor agents is altered. The structure. now Roche-Boehringer Mannheim. colon and gastric xenografts.6 is notable for the presence of the central Pt which contributes to DNA affinity only through electrostatic and H-bonding interactions. This was accomplished first through the incorporation of a third platinum center within the alkane diamine chain (Figure 1).126 Dinuclear and Trinuclear Platinum Anticancer Agents antitumor activity and DNA-binding. the new agent is approximately 100-fold more effective than cisplatin for the maximum therapeutic effect. This approach has the advantage of obtaining appropriate pharmacological data on effective and toxic doses on the murine L1210 animals and on the A2780 human ovarian animals. which arose from our initial reports on trinuclear systems.

' Loss of normal p53 function could lead to intrinsic drug resistance as a result of reduced cell ability to engage an apoptotic response. The plausible explanation of the hypersensitivity of human tumors with mutant p53 to BBR3464 is that apoptosis induced by the drug is not mediated by p53. relative resistance and sensitivity are most likely to be seen at these TWI levels. F. In addition to a role in modulating apoptosis. The protein is further implicated in playing a role in modulating nucleotide excision repair pathways. 2 gastric.Nicholas Farrell and Silvano Spinelli 127 Table 1 Comparision at maximum tolerated dose of BBR3464 (0. Table 3 compares activity with the clinically used alkylating agents.') and cisplatin (3-6 mg kg. 2 ovarian. 5 NSCLC. p53-mediated cellular responses include cell growth arrest required to allow DNA repair prior to genome replication. A subset of the tumors was evaluated by Dr. In 5 out of 6 cases the TWI is indicative of clinical sensitivity. SCLC: small cell lung cancer. although there is no simple relationship between DNA damage and p53 functional status. Zunino for p53 status. the reponse is still significantly better than cis-DDP. Comparison with Other Clinical Cross-linking Agents BBR3464 clearly represents a significantly promising antitumor agent and a distinct departure from all previous mononuclear cisplatin agents. plays a critical role in cellular response to drug-induced DNA damage. In these three cases. is that the novel DNA-binding modes of BBR3464 are not recognised by p53 and so repair and/or apoptotic pathways are not induced. 1 gastric.good tumor sensitivity was observed in 15/18 cases for BBR3464. not only can this new clinical candidate be presented as active in tumors with both acquired and inherent resistance to cis-DDP but it can be presented as maintaining activity in tumors with mutant p53.' The new agent displays remarkably high activity in mutant p53 tumors (Table 2).4 mg kg . 2 NSCLC. 1 prostatic) 15 (3 SCLC. . A plausible extension of this thinking. NSCLC: non-small cell lung cancer. chosen because of strict ability to compare. 1 gastric. This is also true in comparison with other DNA-DNA cross-linking agents. 1 bladder) 2 (1 ovarian. T W I < 50% Relative resistance. a prostatic tumor. Thus. TWI% is Tumor Weight Inhibition compared with controls. 2 ovarian. BBR3464 is significantly more potent than cisplatin . 1 SCLC) 'See reference 7. T W I > 70% BBR 3464 0 Cisplatin 3 (1 NSCLC. So. TWI 5O-7O% Sensitivity. BBR3464 is equivalent in efficacy to cisplatin in one and significantly better in the other and significantly more potent than the alkylating agents with the one exception of Mitomycin C in LXFE839. then.l bladder) 9 (4NSCLC. In the sixth. 1 prostatic) 7 (2 SCLC.') after i.u. repeated treatment on staged tumors Clinical parameter" Resistance. The clinical parameter refers to the fact that clinical resistance. 5 ovarian.2-0.

H.v.C.128 Dinuclear and Trinuclear Platinum Anticancer Agents Table 2 Activity and p53 status of human tumours treated with platinum compounds. TWI%: tumor weight inhibition (on day 22): 100 . Relative tumour where Tois the tumor weight at the start of treatment (day 1) and T.whether it is necessary to contain a third Pt unit (trinuclear class of compounds) in the molecule to achieve this remarkable antitumor potency.C.) 1.8. 15(i.3 Cisplatin 6 3464 0.3 Cisplatin 4 3464 0. Tumor fragments S.4 Cisplatin 6 3464 0.) 1.0 Mitomycin 2. Route 1. LXFE: epidermoid lung cancer.) WIFE839 31 - LXFA526 43 46 23 51 23 OVXF899 Cyclophosphamide 200 Ifosfamide 130.) 1.H.v.) 1.v.v. 8. Treatment i. All tumors transplanted S. LXFA lung cancer. This was achieved through synthesis of dinuclear platinum complexes with hydrogenbonding ligands spermine (total charge 4 + ) and spermidine (total charge 3 +) . 15(i. 15(i.a Tumor modelltype POVD/DDP(SCLC) Calu-3(NSCLC) LX-l(NSCLC) POCS (SCLC) IGROV/DDP (Ovarian) Drugb mg kg- ' T W YO I 70 93 60 92 38 73 56 92 68 80 Resistance ~~~ ~ p53 Status Cisplatin 6 3464 0.p. Klinikum der Albert-Ludwigs-Universitat Freiburg through Boehringer Mannheim Italia.8.0 85 62 55 21 34 53 75 30 Data and comparison supplied by Dr. 4 Structure-Activity Relationships in Polynuclear Platinum Complexes An immediate question relating to the structure-activity relationships within this class of compounds is immediately raised . 15(i.0 BBR3464 0.15 when tumor weight reached an average of 100mg. weight is determined as Tx/To is the tumor weight at day x. All compounds were dissolved in saline before use.tumor area T/C) x 100. OVXF: ovarian cancer. 15(i.4 Cisplatin 6 3464 0.p. Fiebig. on days 1. into CD1 nu/nu female mice (day 0). Table 3 Comparison o eficacy in three solid tumors ( T W I % ) o platinum agents f f with clinically used DNA-DNA crosslinking agents Dose mg kFday- ' Agent Schedule.3 Cisplati n 3.(mean relative tumor weight of treated mice/mean relative tumor weight of controls x 100). TWI is (1 . The most closely related compounds to BBR3464 would contain a diamine backbone with some hydrogen-bonding capacity.3 Acquired Intrinsic Intrinsic Intrinsic Acquired Mutant Mutant Mutant Mutant Mutant "Adapted from reference 9.

Just as a molecule of structure [cis-PtX.l o Selective platination to the terminal NH.although chain length is not absolutely critical but the extra charge is. The results show that a family of dinuclear compounds similar to BBR3464 can be achieved through use of polyamine linkers. the absence of the charged central NH. di. The DNA binding profile of BBR3464 is similar to that described for the dinuclear agents (Table 5).] is now expected to be antitumor active with predictable DNA binding characteristics [high proportion of d(GpG) intrastrand cross-links. by definition. This program developed from the hypothesis that DNA damage different to cisplatin will lead to a profile of antitumor activity that is also different and. Their biological activity in human tumors such as LX-1 reproduces that of BBR3464 and is nothing short of exceptional and significantly improved over the 'original' straight-chain diamines such as BBR3005 (Table 4). 5 DNA Binding of Polynuclear Platinum Complexes The interactions of bifunctional polynuclear platinum complexes with target DNA are unlike those of any DNA-damaging agent in clinical use. group is achieved by selective blocking and deblocking of the secondary nitrogens. strand breakage (bleomycin). anthracyclines). The question remains .Nicholas Farrell and Silvano Spinelli 129 linkers (Figure 2). the Boc group may be displaced readily by dilute acid.topoisomerase inhibition (VP-16.2 * 3 The formally bifunctional DNA binding is distinctly different to that of cisplatin and is highlighted by a high percentage of interstrand . few interstrand cross-links. groups reduces cytotoxicity considerably (compare the spermidine pair BBR3571 and BBR3537 where the Boc is still present on the central nitrogens) and the Boc loses its activity in a resistant cell line. complementary in the clinic.and tri-nuclear complexes incorporating the principal features of trans-Pt monofunctional coordination spheres. The ability to modify biological response by suitable coordination chemistry on DNA could be a significant contribution of inorganic chemistry to medicine. In the case of the polynuclear class of compounds we consider that cross-link formation is also the major lesion but these are sufficiently different in structure and consequence to previously studied covalent adducts that the cells' response follows different pathways.how are they connected? The mechanistic implications must be that the range of adducts produced by the new agents are sufficiently unique that DNA processing and repair pathways are distinctly different from those of any other agents.(amine). separated by a longchain.cross-link formation (cisplatin/alkylating agents). activity in p53 mutant tumours and a predictable DNA binding profile. Interestingly. The shorter spermidine compound is also very potent. bending of DNA into major groove]. The mechanistic importance of the polynuclear platinum drugs is that we tend to consider classes of DNAdamaging agents as acting by different global mechanisms . flexible linker capable of hydrogen-bonding and with an overall charge of 3 + or 4 + are now expected to reproduce high antitumor activity. We have summarized this profile of DNA binding in recent reviews.' Upon complexation. Designed synthesis of polyamine backbones in this manner mimics the essential features of BBR3464 such as charge and chain length .

1 2+ kN\ Y+ w i w N-BC I !w .l/t.1. DMF MeOH. A g N 4 . HzO. DMF 2+ BBR 3005 (l. TDDP. + 3. Structures and abbreviations f o r active dinuclear f platinum compounds with polyumine compounds which are discussed in text .t) BBR 3535 Figure 2 Synthetic scheme for fortnation o dinuclear compounds with spermidine as bridging ligand. HCI Pt I 3+ 2.

1. The kinetics of DNA binding of charged platinum compounds is significantly greater than for cisplatin. Data adapted from references 10 and 23.. on days 1.3: 1. '10' cells i. is Inhibiting Concentration 50% of cellular growth after 48 h of drug exposure.25 3 03 . Table 5 Effects of bifunctional DNA binding of mono-. IC. 480 and 1996. 4 ~ 78 72 73 38 aSee Figure 2 for structures.Nicholas Farrell and Silvano Spinelli 131 Table 4 Cytotoxicity and antitumor activity of platinum polyamine complexes in murine leukemia sensitive (L1210) and resistant to CDDP (L121O/CDDP) and in the L X .NH2]2+.081 2.9. into CD2Fl male mice.. 1994. bMean of 198 experiments.5 0.and tri-nuclear platinum complexesa Complex Cisplatin Transplatin l.All compounds were dissolved in NaCl 0.26 4.054 3.63 0.389 llO(l(xT122) ~~ ~~ ~ 1 - 88 - 0.the high charge .. All compounds were dissolved in saline just before use. 1995..9% and further diluted in complete culture medium.25 6 161.03 0.35.389 references therein.p.t" BBR3464d Unwinding Bending Interstrand Intrastrand 11" 9" B +2 HMGb 30-35" < 5% 10" 14" Hinge < 5% joint Not 52% directed - d(GG).4 0.p.34.25 4. see also Anri-Cancer Drug Design. BBR3005 (2 + ) and cisplatin respectively (even after allowing for the 3: 2: 1 Pt ratio in the compounds). Chain length and geometry affects the kinetics of DNA binding.8: 1 for BBR3464 (4 + charge). T/C: median survival time of treated mice/median survival time of controls x 100.d(AG) d(GNG. di. The rapid binding of BBR3464 could affect sequence specificity .0094 24 .I human tumor xenograft" In vitro IC.88 0. 15 16705. and '[{ PtCl(NH3)2}2H2N(CH2). cross-links. a lack of rigid 'directed' bending and the ability to irreversibly form Z-DNA in appropriate purine-pyrimidine sequences. Treatment i.3 1 2 0.178 100 26 1 144 239. For a 2 h incubation (20 pM compound) on Calf Thymus DNA the relative amounts of Pt bound are 7.64 0.9 after tumor transplantation (day 0).0 No 0. minor) d(GNG) major d(GG) observed - No No Yes Yes 1.5. (PM) Compound In vivo (Ll21O/CDDP onlyjc In vivo LX-1 Dose (mgkg-' day-') L1210 L121 O/ CDDP Dose (mgkg-' day-') TIC% TWZ% BBR3535 BBR3537 BBR 3571 BBR3005 BBR 3464 Cisplatinb 0.0075 22.bHMG recognition for global array of adducts. the cross-link preference (intra versus inter) and the structures of the cross-links formed. dData adapted from reference 24.3 Not detached 40-50% "Adapted from Biochemistry.l/t.

Interstrand cross-linking of poly(dGdC). Treatment of supercoiled DNA of various superhelical density.poly(dG-dC) by cis-DDP does not cause a B -+2 tran~ition. Using denaturing agarose gels.In a 1. This apparent dichotomy can be resolved because the short-chain compounds are very weak interstrand cross-linking agents. DNA crosslinking can be quantitated as the amount of the DNA species migrating more slowly than the single-stranded DNA.NH2]2 or [{ tran~-PtCl(NH.NHJ2+ stabilizes poly(dG-dC). to begin to examine for the structural origin of differential effects between mononuclear and polynuclear platinum complexes is in the structures of interstrand cross-links. From the DNA binding view (assuming reasonably similar uptake) it is therefore not induction of a left-handed form per se which is cytotoxic but both the efficiency of induction and irreversibility through cross-linking contribute to cytotoxicity.' Short-chain ( n = 2." + Cooperative Effects in the Solution Structures of Site-specific (Pt. leading to enhanced sequence specificity. Modelling and distance considerations indicate that from 1.).'~ interThe .). propanediamine) '~ linkers induce the B + 2 transition but are very poorly c y t o t ~ x i c .Pt) interstrand cross-link is implicated in cytotoxicity and the ability to irreversibly induce the B + 2 conformational change is also a feature of the most potent agents.the adduct bends DNA and is recognized by HMG proteins. ethylenediamine. DNA sequences which have the potential to form 2-DNA have been found within the mammalian cell genome including transcriptional regulatory regionsI6 and DNA replication origins. ' The (Pt.6 A. At this stage.from the crystal structure the distance between the two Pt-C1 centers is 23.6 cross-links are possible for BBR3464 . relaxed and linearized DNAs with the platinum complexes allows a quantitative assessment of the number of DNA interstrand cross-links formed per DNA adduct. its analogues and even the alkylating agents.Pt) Interstrand Cross-links A logical place.6 cross-link the guanines are separated by four base pairs whereas a 1. [trans-{PtCl(pyridine)2}2H2N(CH.2 cross-link is formed from guanines on neighboring base pairs. a very relevant finding is that the presence of planar ligands in the coordination sphere (i. 3.e. Likewise.uM in both cases).5i15 Both compounds are efficient interstrand cross-linking agents but not potent cytotoxic agents (ID50 in L1210 leukemia > 15.2 to at least 1." The induction of Z-DNA within the cell would have serious consequences with regard to transcription and DNA replication. therefore.)(quinoline)}~H~N(CH.poly(dG-dC) in the B form. long-range interstrand cross-linking is the dominant DNAbinding feature which distinguishes our compounds from cis-DDP.132 Dinuclear and Trinuclear Platinum Anticancer Agents could lead to initial electrostatic interactions very different to small molecules such as cisplatin and the alkylating agents. An interstrand cross-link is a relatively minor event for cis-DDP but there is still controversy as to the biological implications of the lesion -.

As stated. The potential for systematic alteration of DNA structure by platination is demonstrated by the novel structure of a site-specific 1.2-cross-link between adjacent guanines. ) ~ } .20The structure is totally different from that of the cisplatin cross-link and is a dumbbell shape caused by looping of each single complementary strand back on itself with the two ‘looped’ single strands stacked together by the nature of the dinuclear Pt-G binding. a very interesting NMR structure of a 172-interstrandcross-link in 5’d(CATAG*CTATG)-d(GTATCG*ATAC) a left-handed but highly localshows ized 2-DNA formation at the binding site due in part to an induced syn conformation of the platinated bases. 6 Summary This contribution succinctly reviews the chemical and biological data of a new structural class of anticancer agents based on a poly(di.P ~ C ~ ( N H .. with the ~ formed by reaction of [ { ~ ~ u ~ ~ . Acknowledgements It is a great pleasure for the authors to acknowledge the dedication and enthusiasm of all the team in Boehringer Mannheim for this project.2-interstrand cross-link . tri)nuclear structure. if altered to a left-handed form. This combination of interactions could have the ability to induce the conformational change in a cooperative fashion beyond the actual site of platination. This is especially attractive for formation in 2-like sequences. The salt induced B -+ 2 transition is highly cooperative . For dinuclear and trinuclear complexes separated by long-chain diamines. N ( C Hi-) .I2 This conformation results in a large bending angle and the interstrand cross-link is thus recognized by HMG proteins. Brabec and F. Thus. unlike the relatively localized distortions induced by cisplatin-DNA adducts.21 A four to six base-pair sequence is inherently achievable by long-range interstrand cross-links and. Zunino for their valuable cooperation and insight.a four base-pair section is required to ‘nucleate’ adjacent sequences toward the left-handed form. ‘nearest neighbour’ 4 2 cross-linking will not be favoured over longer-range cross-links.22 In design of DNA-binding drugs there is much discussion of the need to target 10-15 base-pair sequences to begin to approach a gene-selective strategy. The differences and similarities we will observe between these agents and the ‘classical’ cisplatin compounds as they hopefully proceed through the clinic will add to our understanding of drug discovery and design for further rational approaches to the drug treatment of cancer. The final structures and conformational change will also be affected by interactions of the linking backbone with the non-platinated base pairs. N H ~ ]central d(GC) pair of the symmetrical { 5‘-CATG*CATG-3’}. the possibility exists that the Pt-DNA adduct is disseminated through a longer piece of the helix. . H .Nicholas Farrell and Siluano Spinelli 133 strand cross-link of cisplatin is a 1. is capable of being propagated through appropriate joining sequences. We wish to also thank V. The ability to cooperatively alter conformation over large distances is an exciting and important direction to pursue and a further unique advantage of the polynuclear approach.

E. 1995..N.l995. E.63. Proc.373. Hoeschele and N.2077. Comments Inorg. S.1697. Biochemistry. A. Pratesi. Inorg. Huang. Proc. 1996.D.375. 20 D.D.C.15 480. Kasparkova.88.26. 1995.16.-J. Wang.32. Farrell. L. C. H. Spinelli. Farrell. Skov.S. Johnson.H. Biochemistry. Polizzi.Kohwi.377. Menta. R.M. L.C.M. 3 N. 21 F. 1997. 16 Y. Supino. D. Randisi.ed.. 1997. 22 P. Di Domenico. Nucleic Acids Res. 1972. De Cillis.M.Houston.Science.H. Frederick and S.19. M.20. Farrell.134 Dinuclear and Trinuclear Platinum Anticancer Agents References 1 M. Kharatishvili. R. Zhu.270. Inorg.Qu. T. F. 12 H. J. Perego.p. Chem. Nucleic Acids Res. 23 H. 1996. Pezzoni. 17 A.3451. Qu. Chetn. Y. G. Manzotti. Proc. f 13.7229. R. 10 H. 1989. Rauter. J . DiDomenico. Farrell.1. Chaires.C. Y. JAI Press. Mathieson and N. .577.34. Y. Spinelli. 8 C. 1993. J. Dalrymple.11.B. 7 C. N.2080. Farrell.2. Rich in Proceedings o the Robert A. 1993. Farrell... Righetti. 1997. 24 N.B. J . Chem. Farrell. Rahmouni and M. Reid. B. 5 N . Van Houten and N. Farrell and A. Tognella.P. De Giorgi.720. G. 1997. 1995. Cancer Investigation. Heintz and M. K. O’Sullivan and D.. J . L. 6 T. Nature Struct. Leng. A. 2 N . S.C. Piazzoni and N. Zou. 18 A.39. Farrell and F. S. 1996. Farrell. Lippard. 13 P. Biol. 1992.H. Acta. Christian. Pezzoni. Lotto. 37. 4 l.C. Manzotti. P. Inorg. Bianchi.. V. Drobny and P. E. 1995. Jovin.38. G. Appleton. Menta. 1442. Q u and N. Lotto and S.67. Chirn. Seminars Oncol..187.2591.D. Vol.P. 2. Fiebig. Torriani. Welch Foundation.21789. Pohl and T.. Wells. P. Farrell. Farrell and F. Wu. A. D. van Boom. Cancer Inst. 1990. N.255.G. Farrell. Y. AACR. M. Tetrahedron Lett. Inorg. M..A.4493. AACR. Rosenzweig. Conti. Y. Roberts. 14 A.578. Zunino. Chem. 1993. Menta. 1995. in press. Takahara. Di Domenico E. R. Mol. J. Rauter. Harris. Br. van Boom. Biol. Y. Da Re. Q u and N.36. A. M. 15 M.3919. R. G.36. Menta. C.H. B.C.. Nature.J. 649. Cancer. J.17.265.M. Hopkins. E. Manzotti. 9 G.C. Caddle. Spinelli. Giuliani. Giuliani. AACR. Yang. J . Soares Fontes.Qu.38. Biochem. N. F.R. J. S. Kharatishvili. Vol. 11 M. 1096. C. Hurley and J. G. Natl. S. Biol. M. Oliva. Valsecchi. New Haven. 19 A. Qu. G. M. Pavesi. Da Re. N. Reedijk.l987. Farrell. in Advances in DNA Sequence Specific Agents.1842. Appleton. Brabec.9.G. Wu and Y. S.A. 1998. Spinelli. A. 1992.. J. Conti..

Other quinone antitumor drugs used in the clinic. USA 1 Introduction There are now more than forty drugs approved for the treatment of human cancer in the United States. ( )-1. Tr. may also function as iron complexes or utilize cellular iron in an indirect way in their mechanisms of action. + . Phen.9-dimethyl-l. this is a highly select group.CHAPTER 9 Oxidation Damage by Bleomycin.7-phenylsulfonyl-l. topoisomerase 11. BPS.' In addition. 2. GSH. RSH. PATRICIA FULMER AND WILLIAM E. NC. is a natural product that must form an iron complex to display cytotoxicity. doxorubicin or adriamycin. Top0 11.5-dioxopiperazinyl-l-yl)propane compound is now called (this ADR-925 by its manufacturer). 4.2-bis(3. glutathione.lO-phenanthroline. Bleomycin causes cellular DNA damage by an oxidative mechanism. 3-ethoxy-2-oxobutyraldehyde bis(thiosemicarbazonato)copper(n). Blm.l0phenanthroline.lO-phenanthroline. PETERING7JUN XIAO. bleomycin. the hydrolyzed form of ICRF-187 (called ADR-529 by the manufacturer). ANTHOLINE Department of Chemistry. including mitomycin C. two anthracycline natural products. adriamycin and daunomycin also produce oxidative damage as part of their repertoire of reactions in cells. CuKTS. Among them are two drugs that require a metal ion as part of their structures. and daunomycin. thiol compound. bleomycin. ICRF-187. mitroxanthone. Thus. adriamycin or doxorubicin. Medical College of Wisconsin. One is the simple metal complex cis-diamminedichloroplatinum(I1). Adriamycin and Other Cytotoxic Agents That Require Iron or Copper DAVID H." Considering the thousands of compounds that have been tested as candidate antitumor agents. transferrin. The other. quite apart from *Abbreviations: Adr. may also require cellular iron to catalyze formation of reactive oxidants. ICRF-198. University of Wisconsin-Milwaukee and Department of Radiation Biophysics. 1. SREEDEVI NYAYAPATI. and mithramycin.

metal-dependent drugs in cancer chemotherapy.* It is administered as the metal-free natural product. The relationship between the efficacy of Blm chemotherapy and its access to circulating or cellular iron has not been examined. probably composed of five labeled nitrogens (Figure 1). it is a highly efficient reagent for the generation of DNA double strand breaks. Nevertheless. Cancer patients frequently display a generalized stress response or cachexia which inhibits appetite. Experimental antitumor agents such as streptonigrin. delivery of FeBlm would circumvent problems that the drug may have in finding sources of iron.136 Oxidation Damage by Bleomycin.l' Therefore. At the least. according to experiments described below. variable R group. it appears that Blm retains full activity in iron-deficient animals. The sections below provide a coordinated review and perspective on the subject of redox-active. it is surprising that trials have not been conducted to determine whether FeBlm offers therapeutic advantages over metal-free Blm. 2 Bleomycin Bleomycin (Blm) is a natural product that is in regular use in combination The left side of its structure contains a versatile metal-binding domain.^'^ In sum. bisthiosemicarbazones. it has remained underdeveloped as a theme for study and application. In this state.' Clinical Use Bleomycin is commonly employed in the treatment of some human cancers. Blm must be converted into FeBlm to inhibit cell proliferation and degrade cellular DNA. and perhaps monothiosemicarbazones must form iron or copper complexes to become biologically active as catalysts of oxidant damage to cell^. FeBlm seems to cause less general host toxicity than Blm d0es.' Whether this influences the availability of iron for the drug is unknown. since FeBlm is the active form of the drug. Adriamycin and Other Cytotoxic Agents deliberate interest in metallodrugs or metal-based redox chemistry. yet. metal-based catalysis of redox reactions in cells describes a major topic in cancer therapeutics. plasma iron can be markedly depressed. A well-defined role for the disaccharide unit has yet to be established. As such. as many as 15% of the anticancer drugs approved for human use may need a metal ion for activity. . Furthermore. It is an effective antitumor agent and a model for an effective metallodrug. indeed. which can bind to DNA. As a subject for mechanistic study. Blm is an enormously important molecule. as discussed below. it has also become the prototype for the design of synthetic chemical nucleases involving redox-active metal sites. The right side includes a bithiazole moiety and a positively charged.'' However.

13 Although it was originally assumed that CuBlm was the active form of the drug. Thus. S . However. E. Fulmer and W. Antholine METAL BiNDING DOMAIN I 1 DNA BINDING DOMAIN - 137 Figure 1 Structure o bleomycin f Animal and Cellular Studies on the Role of Iron in the Cytotoxicity of Bleomycin There are few reports that have dealt with the interaction of Blm with biologically essential transition metal ions that may be involved in its mechanism of action. Blm fully saturated with Cu2+ displayed antitumour properties similar to the original preparations of the drug. Jun Xiao.12 Indeed. P. metal-free Blm was as effective and caused fewer side-effects. Initial reports of the isolation of Blm mentioned that the product isolated from Streptomyces verticillus contained significant amounts of Cu. Soon. which showed that cellular iron was required for both single and double '' . it can be garnered from the host. Petering.'^'^^ Animal experiments were then undertaken to determine whether Fe or some other transition metal was required to elicit the activity of Blm. if a metal is required for activity. two sufficiently iron-deficient cells were finally produced. it was observed that the drug caused the strand cleavage of DNA and that the reaction was enhanced by the presence of Fe2+ and re duct ant^. Nyayapati. to the extent that the powdered drug was bluish.D. Early cellular experiments were unable to define FeBlm as the active form of the drug. It was anticipated that host Fe or Cu deficiency might depress the activity of Blm.and copperdeficient mice. because less of the activating metal ion would be available in the organism for complexation by the drug. l o Nevertheless. Blm retained full activity in iron. H .

26-28If no 0. Since iron is closely regulated and exclusively bound to transferrin in plasma. as in cells. and is predominantly associated with ferritin and other proteins in cells. ~ ~ . and also base Evidently. One such type of cell was the microorganism Euglena gracilis. A second question relates to mechanisms of biological interconversions of metallobleomycins.24Both findings support the central role of double strand cleavage in the cytotoxic process. Cu-. The other was human HL-60 cells adapted to grow in defined medium on iron-pyridoxal isonicotinoylhydrazone and then cultured in the absence of this source of iron.~ ~ The extent of formation of double. capacity to damage DNA in this system. These DNA lesions can also be detected in isolated nuclei exposed to FeBlm. A number of properties of these reactions have been established. most of it may not be readily available to Blm. double strand breaks persist.25 Biochemical Mechanisms of DNA Damage Characterization of DNA Damage by FeBlm Reactions of species of Blm with DNA have been examined in much detail in model reactions. few experiments have examined whether Blm might compete for Fe bound to metalloproteins. is present. However. but not single. and Zn-complexes are similarly cytotoxic to cells exposed under conditions that minimize extracellular cross reaction with other metal ions. by using alkaline and neutral filter elution methods r e ~ p e c t i v e l y . First. and cellular Zn2+may well be available for reaction with B l m . whereas Blm and CuBlm display little. which could be made rigorously iron-deficient upon growth in defined medium containing less than 10 nM Fe. Adriainycin and Other Cytotoxic Agents DNA strand scission and for inhibition of cell proliferation and reduction in viability. implying that each can be converted into FeBlm. Opposite conclusions have been reached on whether copper can be removed from Blm by reductive ligand substitution using thiol Cellular Properties of DNA Strand Scission Both single and double strand scission can be detected in cells exposed to Blm and its Fe(II1) and CU(II)complexes. strand lesions parallels the drug concentration dependence for inhibition of cell proliferation. FeBlm-dependent reduction of dioxygen leads to a combination of single and double strand cleavage. ' ~ ~ ~ has also CuBlm been detected in urine after administration of Blm. each process begins with the abstraction of a C4' hydrogen atom on a deoxyribose site along the backbone of DNA (Figure 2). Blm and its Fe-. A fundamental question about the mechanism of action of Blm is how it acquires iron in the host organism. base release . Cu2+ binds to Blm with higher affinity than does Fe3+. if any.138 Oxidation Damage by Bleomycin.16 exposure of cells to CuBlm did not overcome the effect of iron deficiency." Yet. These experiments indicated that cells must be highly Fe-deficient before any reduction in Blm activity can be detected.21 While single strand scission is readily repaired.

Nyayapati. Antholine 139 + I 0 o< 0 o=a-o 0 1 (L I I ' u 0-q. O 0 1 a x I 0-4-0 0Za-o 0 0 a 1 .D. E . H .- 0-a-0 0 O-Q-0 0 0 . S . Petering.-0 0 I + O0pI : 0 oa0 . Jun Xiao. Fulmer and W. P .

double strand cleavage probably results from the action of a single molecule of FeBlm acting on both strands of DNA at a particular site. . a peptide linker between them and a disaccharide. base release at the ~~’~~ preferred site occurs without further damage to the other strand. ~ ~ In ~ * ~ . ’i3 Solution Properties of Metallobleomycins Related to the DNA Damage Reaction The structure of Blm shown in Figure 1 may be divided into a metal-binding domain. Although some results have suggested an intercalative mode of binding. 5-3 Furthermore. ~ described below. single strand breakage at a preferred site can be accompanied by single strand cleavage on the other strand at the same base pair or offset by one base pair. Both the bithiazole and the variable. With the finding that Blm and ZnBlm adopt extended conformations in solution. in the cell. Thus. culminating in apparent double strand cleavage. a DNA-binding domain.it was originally thought that the site selectivity of the drug for DNA cleavage and base release resided in structural characteristics of the b i t h i a ~ o l e As ~ ~ ~ . and so continues to serve as a model for the metal-ligand structure of various metallobleomycins. producing double strand scission. as well as the kinetics of cleavage. its copper ion is bound to the five nitrogen atoms denoted by dots in Figure 1.39 Then. The early publication of the crystal structure of the metal domain of CuBlm. ~ ~either~case.Alternatively. positively charged R group in the domain contribute to binding. demonstrate that both single and double strand cleavage occur at all concentrations of d r ~ g . The drug strongly prefers cleavage sites between 5’-GpC-3’or 5’-GpT-3’.~ nature of the interaction between the DNA domain of The Blm and DNA itself has been the subject of many studies. remains the only X-ray structure of a metallobleomycin species.. and release of C or T from these sites. this view has evolved to include the metal domain in the mechanism of site selectivity.38. the redox chemistry at the metal center occurs in proximity to DNA and efficiently causes damage to its structure. This description resembles that for synthetic nucleases that have been constructed using this m0de1. association along the minor groove has also been p r o p o ~ e d . this must be the case. Adriamycin and Other Cytotoxic Agents occurs. in the presence of 0. or can be accompanied by base release on the other ~ t r a n d . it was assumed for a number of years that the DNA domain and the linker act to tether the reactive metal domain in the vicinity of DNA. where the concentration of drug in the nucleus is no more than one molecule per lo5 bases. but instead represents a subset of the DNA damage events occurring at a given site. Second.’ There is site selectivity in the reactions of FeBlm and DNA. which lacks the disaccharide.140 Oxidation Damage by Bleomycin. which has not been the subject of much study. the other pathway competes for the C4’ radical.31-34 Certainly. the observed cleavage patterns of supercoiled plasmids and synthetic hairpin DNA duplexes. and frank cleavage of the backbone results.44 As seen in Figure 3. ~ ’ -This ~‘ indicates that double strand scission is not the outcome of the introduction of many single strand breaks.

at least. to Fe(Ir)Blm. . --Fe(m)Blm with DNA as overall redox reactions (Figure 2). + Fe(I1r)Blm -+ HO. Horwitz and colleagues has demonstrated that in uitro DNA strand scission is dependent upon the presence of iron in the reaction mixture and involves the following pathway of activation of the drug to carry out the strand scission reaction: + Blm -+ Fe(1II)Blm Fe(I1)Blm Fe(m) + Blm + eFe(I1)Blm + 0. this species begins to attack its own structure in a reaction that inactivates it for subsequent reactions with DNA. Antholine NH-CO 141 H i H2N Figure 3 Structures of C U P . Mossbauer.14 Upon addition of 0.~"' It can also be formed by the direct reaction of H 2 0 2with F e (~ ~ )B ln i:~ ' H20. + - Fe3 (1) (2) (3) -+ (4) The reducing agents in Reaction 2 can be reagents such as thiols or ascorbate. a fleeting intermediate is observed.D. Nyayapati.45*46 That HO.~ A .F ~ ( I I I ) B ~C4'-H ~ + + C4'-H+ + Fe(m)Blm + H 2 0 + OH- (6) .~~ DNA is not present. probably the dioxygenated species.-Fe(11)Blm 20. it has been shown that this species is competent to initiate strand scission or. modijied metal domain of CuBlm the An elegant series of papers by Peisach. and EXAFS analysis of its struct~re. E.. is the only observable precursor of a form that starts the process of DNA d e g r a d a t i ~ nIndeed.--Fe(m)Blm is formed in this reaction is supported by ESR. base release involves a simple stoichiometry of reaction: H O . which rapidly reacts to generate the products of Reaction 4. S . --+ 0. Petering. P.--Fe(111)Blm + H- (5) Using Reaction 5 to generate H02--Fe(111)Blm.-Fe(11)Blm + H + -+ HO. H . mass spectroscopic. Jun Xiao.--Fe(m)Blm + Fe(I1I)Blm + 0.46*50*51 Considering the various outcomes of the reaction of HO. if . Fulmer and W.

142

Oxidation Damage by Bleomycin, Adriamycin and Other Cytotoxic Agents

Reaction of H02--Fe(111)Blm with DNA to produce a single strand break had been conceived of in two ways: in the first, abstraction of the C 4 hydrogen formally involves a species equivalent to a hydroxyl As such, there is homolytic cleavage of the peroxide, leaving a ferryl species, equivalent to a hydroxyl radical and Fe(III)Blm, as the other product. H0,--Fe(111)Blm + C4’-H -+ C4’.

+ HO-Fe(1v)Blm+ H,O

(7)

As this C 4 radical is converted into a hydroperoxide, an electron is required from some source in the reaction mixture. According to the other proposal, HO, --Fe(III)Blm undergoes heterolytic cleavage to produce hydroxide and a perferryl species that does the hydrogen a b ~ t r a c t i o nThe net reaction is the .~~ same as shown in Reaction 7 :

H02--Fe(111)Blm+ O=Fe(v)Blm O=Fe(v)Blm

+ OH+ C4-H -+HO-Fe(1v)Blm+ C4’-

(8) (9)

It is further assumed that the other electron needed to reach the H0,-C4 intermediate is supplied by the ferryl product in Reaction 10: HO-Fe(1v)Blm + C4‘*+ 0, + H + -+ HO,-C4’

+ O=Fe(v)Blm

(10)

In this model, the perferryl species is one of the products. Thus, in neither hypothetical pathway of single strand scission is Fe(1n)Blmgenerated. Instead, a reactive species remains. How either ferryl or perferryl iron centers are reduced to Fe(II1) is unresolved. The problem of the extra oxidant species in isolated single strand cleavage is an asset for the construction of mechanisms of double strand cleavage by single molecules of FeBlm. Although it is unknown whether Fe(rI1)Blm must be activated twice to obtain double strand cleavage, in principle, either of the mechanisms above (Reaction 7 or Reactions 8-10) can provide a reactive FeBlm species to abstract a hydrogen atom from the C4’-H on the second strand: HO-Fe(1v)Blm + C4’-H -+ C 4 - + Fe(rI1)Blm + H,O or O=Fe(v)Blm (11)(ref. 31)

+ C4‘-H

-+

C4*

+ HO-Fe(1v)Blm

(12) (ref. 34)

In the second pathway, O=Fe(v)Blm will again be generated as H0,-C4‘ is formed. This raises the interesting question of whether HO, - -Fe(m)Blm might serve as a catalytic reagent for double strand cleavage.34 Evidence to date indicates that strand cleavage is stoichiometrically dependent upon added reducing equivalents or added HO, - - F ~ ( I I I ) B ~ ~ . ~ ~ * ’ ~

D. H . Petering, Jun Xiao, S. Nyayapati, P. Fulmer and W. E. Antholine

143

When single strand scission on the site-specific strand is followed by base release on the second strand, the presence of HO-Fe(1v)Blm at the site will provide only one of the two oxidizing equivalents needed to start this pathway (Reaction 6). In contrast, O=Fe(v)Blm can supply both equivalents. Thus, neither of these formulations for the reaction of a single molecule of HO,-Fe(1II)Blmwith a DNA site is completely satisfactory.

Properties of Metallobleomycins Bound to DNA
Early studies of the interaction of ON-Fe(I1)Blm and OC-Fe(I1)Blm with DNA showed that DNA perturbs the metal site. The ESR signal of the former was dramatically changed by DNA, and the chemical shift of the NMR signal of the H2 proton of metal-bound imidazole in the latter was altered upon binding to DNA.53,54 Furthermore, the low- to high-spin transition of Fe(m)Blm at pH 4, thought to result from protonation of the axial amine in the metal domain, was reversed upon addition of DNA, suggesting substantial interaction between DNA and the iron coordination site.55Thus, it is evident that the metal domain chemistry of FeBlm is perturbed by the presence of DNA. What has become clear recently is that the alteration of metal domain properties upon binding to DNA occurs only when the structure interacts with specific 5’-GpPyrimidine-3’ sites.56 Such evidence along with NMR structural results described below strongly link the metal domain structure with site specification that precedes the DNA damage reactions. In order to systematically investigate the roles that DNA may play in the chemistry of FeBlm bound to DNA, the properties of CoBlmDNA were first examined. The cobalt substitution was made because the dioxygen chemistries of Fe(r1) and CO(II)are similar, but the redox reactions of CoBlm are slower than those of FeBlm and are thus amenable to more detailed study. The pathway of reaction of Co(I1)Blmwith 0, is analogous to that proposed for Fe(11)Blm:~’ Co(r1)Blm+ 0, -+ 0,-Co(11)Blm 20,-Co(11)Blm -+ BlmCo(I1)-0,-Co(I1)Blm + 0, BlmCo(II)-O,-Co(Ir)Blm + H + -+ HO,--Co(111)Blm (Form I)

(13)
(14) (15)

+ Co(1II)Blm
(Form 11)

In this case, however, the presence of each species and the kinetics of reaction of each step can be observed, so that the detailed reaction mechanism above can be set forth. The products are analogous to those formed starting with Fe(r1)Blm.In contrast, they are stable and can be conveniently studied. The rate of product formation is markedly changed in the presence of DNA.58 As the ratio of base pairs to CoBlm increases and the average distance between drug molecules distributed along the DNA structure lengthens, the rate of Reaction 14plunges, because metal domains cannot dimerize or make contact to accomplish electron transfer. As a result, 0,-Co(I1)BlmDNA becomes highly stable at base pair to drug ratios of 10 or more.

144

Oxidation Damage by Bleomycin, Adriamycin and Other Cytotoxic Agents

When the reaction of Fe(I1)Blmwith 0, was examined in the presence of DNA, a similar mechanistic pathway was defined.58With ratios of base pairs to drug of 20 or greater, 02-Fe(11)BlmDNA is stabilized, indicating that Reaction 4 is no longer a feasible reaction to generate the H02--Fe(111)Blmneeded to initiate the DNA damage pathways. Interestingly, O,-CO(II)B~~DNA and 0,Fe(I1)BlmDNA are unusually resistant to dissociation of O,, implying that the dioxygen complex is stabilized by DNA.59960 Given the fact that in cells DNA is in overwhelming excess relative to nuclear Blm ( lo5 to l), it is evident that Reaction 4 cannot contribute to the activation process.21 Thus another source of electrons must be involved in this step as well as in Reaction 2.

-

The stabilization of dioxygenated CO(II)- Fe(II)-Blmon DNA suggests that and the complex is not able to undergo the rapid cycles of dissociation and rebinding necessary to find a partner to carry out Reactions 4 or 14. Nor is a rapid sliding process available to facilitate these reactions.

Structural Properties of CoBlm Species in Solution and Bound to DNA
The conformational relationship between the metal domain of several metallobleomycins and DNA has been probed by ESR studies of Fe(m)Blm, NOFe(II)Blm,and 0,-Co(11)Blrn bound to oriented fibers of Remarkably, the dioxygen ligand of 0,-Co(11)Blm and the nitrosyl group of NO-Fe(II)Blm, which largely bear the unpaired spins in these complexes, are rigorously constrained to a plane approximately perpendicular to the helix axis. These results provide clear evidence that the metal domain in these adducts, like the DNA domain, must be closely associated with DNA, and not loosely tethered to the polymer. They also indicate that Fe- and Co-Blm species bind to DNA with similar steric constraints. The stability of H02--Co(111)Blm(Form I) as well as its diamagnetic character offer the opportunity to determine its structure and compare it with that of Co(m)Blm (Form 11).A two-dimensional NMR study has revealed that both of these molecules share similar, highly folded metal domain and linker regions that form close packed They differ in that the bithiazole group of Form I is also folded back upon the metal domain, generating a globular conformation for the structure (Figure 4). Consequently, the sixth coordination site holding the peroxyl group in Form I is a well-defined pocket, and the site in Form I1 is more open. Considering the interaction of the Form I structure with the DNA minor groove, where binding has been thought to occur, and assuming that H 0 , exists in the same conformation as 0, in O,-CO(II)B~~DNA, postulates that one both the DNA and metal domains will be in close contact with DNA. Furthermore, the peroxide ligand will be directed into the interior of the minor groove in

+-

02-Fe(~r)BlmDNA e+

+H

-+

HO, --Fe(III)BlmDNA

(16)

E.. the structure of H02--Co(111)Blm(Form I) bound to DNA oligomers have been under s t ~ d y .The~ . 5’-GpC-3’ or 5’-GpT-3’. Nyayapati. P.~ c a l eIndeed. Stubbe and coworkers showed that Form I binds to DNA in part by intercalation of the bithiazole between the base pairs C6-C7 of the first 10-mer and T5-A6 of the second 0 1 i g o m e r . Jun Xiao. it is the metal domain that . d(CCAGGCCTGG).-Fe(rI1)Blm. Antholine 145 Figure 4 Solution conformation of H 0 2--Fe(ttl)Blm proximity to the C4’ hydrogens that are the initial target of reaction for HO. S .65-67 Using the first two of these oligomers. ~ ~ ~ ~ ~ bonds are made between the ring nitrogen and Hydrogen adjacent amine group of the pyrimidinyl ligand of Blm and N-3 and 2-amino nitrogens on the minor groove edge of G of the first two DNA structures a b ~ v e . The key to this work was finding that ~ ~ . To test these ideas. Fulmer and W. Petering.. In the process.. suggesting that they interact in slow exchange on the NMR t i m e . which incorporate the selective sites of strand cleavage or base release.~~ Form I binds in slow exchange to several oligomers. ~ ~ the formation of an unusual base-triple structure in the ~ result is minor groove comprised of the G-pyrimidine base pair and the pyrimidinyl ring of the drug. and d(AAACGTTT). For purposes of specification of the site of reaction.~ ~ both Form I and Form I1 bind tightly to DNA.namely. the peroxide ligand becomes oriented toward one of the two C4’ hydrogens in the minor groove as if in preparation for hydrogen atom abstraction.. d(CCAGTACTGG). it has been demonstrated that . H .D. d(GGAAGCTTCC).

Phleomycin. would appear necessary to specify both the guanine and pyrimidine components of the cleavage sites. Role of the Metal Domain in Site Selectivity Papers by Hecht and co-workers have examined the role of the metal and DNA domains in the site selectivity of DNA damage caused by FeBlm. the drug can undergo complex oxidation-reduction chemistry. the metal domain tethered to distamycin causes DNA breakage based on the binding preference of distamycin. to generate oxy-radicals like the hydroxyl radical. These mixed results point to a joint role of the whole folded structure in binding and site selection. when the metal and DNA domains are bound together by oligoglycine linkers of 1 to 4 monomers.68Although this has been construed to mean that the metal domain determines the site of strand scission.7476 Furthermore.69970 However. It can bind to DNA and may inhibit DNA replication after The drug may undergo redox cycling in the presence of cellular reductants and 0.6*68 Intercalation appears to be nonspecific in that the bithiazole-tail can intercalate between at least four pairs of stacked bases. redox chemical options are added to the molecule. perhaps because they do not bind DNA as well as FeBlm does. which contains a partially reduced thiazole ring that is no longer planar. anthracycline chromophore connected to an amino sugar residue (Figure 5).72Thus. each of which contains one of the elements of site selectivity.76v82 Adr + e- + Adr- (17) . combined structural determinants in both metal and DNA domains. Interestingly.~' Both are less efficient agents than FeBlm.77If Fe2+ or Fe3+ is bound. cytosine or thymine. Nevertheless. cleaves DNA with the same sequence specificity as does Blm.73 The adriamycin structure consists primarily of the fused-ring. Finally. 3 Adriamycin Adriamycin has been heavily used for many years as a chemotherapeutic agent for the treatment of a variety of cancers. particularly solid tumors. cleavage site specificity does not change. not Blm.146 Oxidation Damage by Bleomycin. That the metal domain participates in site specification had been hypothesized. which directly degrade DNA. modifications of the DNA domain may have either no effect or a dominant effect on sequence selectivity. Adriamycin and Other Cytotoxic Agents interacts invariantly with guanine. a monothiazole derivative shows no sequence ~electivity. and perhaps in the linker region. the planarity and extensive double bond character of the anthracycline portion of the molecule gives it both the lipid solubility to interact with membranes and the structure to intercalate between DNA base Because of the multiple facets to the chemistry of Adr. Through its set of quinone-hydroquinone functional groups. the DNA domain of Blm tethered to Fe-EDTA indiscriminately cleaves DNA. the positioning of these groups offers reasonable binding sites for metal ions. different sites and modes of action of the drug have been hypothesized. Finally.

the lipid solubility of Adr may direct it to the cellular membrane. .O. Petering. where its capacity to undergo redox reactions and bind Fe may lead to lipid peroxidation and destruction of membrane analogous to its proposed effects on DNA.’ Nevertheless. Copper adriamycin is not stable in plasma and thus is unlikely to display significant stability in biological systems.85*86 A number of proposed mechanisms of action of Adr postulate direct or indirect roles for redox-active metal ions.+ OH- (22) which could participate in the reductive activation of 0. but its ill-defined nature weakened interest in biological Fe-Adr interactions. H . In addition. copper and iron. Generally. P. since Adr is administered as a metal-free preparation. + Fe(I1) + Fe(II1) + OH. this mechanism might include direct binding of Fe to catalyze hydroxyl radical formation. to the hydroxyl radical by Adr (Reactions 17-22). Jun Xiao. H.D . S .” A complex of Fe(I1) and Adr called quelamycin received some attention years ago. a question of wide concern has been the role of biological sources of iron as catalysts of the Fenton reaction. and those are considered below. Antholine 147 0 OH COCHEOH H . chemically rigorous studies of iron-adriamycin chemistry and biochemistry have been undertaken. for those exploring the elaborate redox chemistry of Adr and its participation in the drug’s cytotoxic activities.CO 0 HO HO ADR IA MYC I N (DOXORUBICI N) Figure 5 Structure of adriamycin Indeed. Fulmer and W. Nyayapati. any interactions between the drug and metal ions must occur after it enters the organism. attention has centered in the essential redox-active metals. E.83i84Alternatively. However.

9 7 8 9 9 HOOC-CH2 H~NOC-CHZ Figure 6 Structures of (a) ICRF-187and (b) ICRF-198 .198 (ADR-529). Attention has focused on the proligand. The nature of the interaction between Adr and Top0 I1 has not been determined. which occurs in humans and animals after a certain cumulative dose of the d r ~ g .148 Oxidation Damage by Bleomycin. should be a favorable target for oxidant damage by Adr. ~Building on the ideas l-~~ developed above. the other hypothesized pathway of cytotoxicity with extensive support involves the inhibition of topoisomerase I1 (Topo 11) by a d r i a m y ~ i n .1-yl)propane.93*94 Efforts have been ongoing to find means of preventing the cardiotoxicity of Adr. Adriamycin and Other Cytotoxic Agents Besides proposed mechanisms that focus on oxidant-based damage to cells. called ICRF.5-dioxopiperazin.it has been argued that heart muscle. that Adr exerts its effect through the generation of reactive 0. containing two carboxyl and two amino groups that can bind a number of transition metal ions (Figure 6). Administration of ICRF-187 to animals and humans effectively ameliorates the cardiotoxicity of Adr. ICRF-187 (ADR925). nor has a link between cellular iron and inhibition of this enzyme been suggested. one approach has been to try to deny Adr access to reactive pools of cellular Fe that catalyze Reactions 20-22. which is easily hydrolyzed to a tetradentate ligand. ( + )-1. Aspects of Clinical Use The application of Adr in cancer treatments is limited by severe. life-threatening cardiomyopathy (heart muscle degeneration). species. * ~It ~is hy* ' pothesized that the inhibition of Top0 I1 as it participates in DNA replication causes DNA double strand breakage. with its intense oxygen metabolism and relatively low concentration of antioxidants such as superoxide dismutase and catalase.2-bis(3.95p96ICRF-198 can both compete successfully with Adr for Fe(r1) and remove Fe from transferrin. Given the centrality of iron in the mechanisms of redox-based toxicity of the drug.

'O'*''O In the presence of dioxygen. and at the same time move Fe necessary for the production of hydroxyl radicals into close proximity to DNA.lO-phenanthroline may be required as a thermodynamic sink for Fe(I1) to drive these r e a c t i ~ n s .D. E.'02~'0L'08In the first two cases. Antholine 149 Cytotoxic Mechanisms and the Role of Iron Despite the many studies that have been conducted on the mechanism of cytotoxic action of Adr in tumor and heart. Fe(w)Adr. Fulmer and W. 1. In an effort to determine whether iron is required for the cytotoxic actions of Adr. Model studies have suggested that Adr may be able to mobilize iron from physiological iron binding sites such as transferrin and ferritin through either reductive dissociation or direct binding of F ~ ( I I I ) . where a plethora of other iron-binding ligands also exist. Nyayapati.which is more labile and available for chelation. is formed.99*'00 Adriamycin is completely without effect on cell proliferation in severely irondeficient Euglena gracilis and has substantially diminished activity in the Fedeficient mammalian HL-60 tumor and H9c2 (2-1) heart myoblast cell Thus. a variety of one electron donors have been employed. several types of cells have been made iron deficient and tested for sensitivity to Adr. it slowly undergoes internal oxidation-reduction.lO-phenanthroline. the . H . '09 Once Fe(m)Adr. S .84 Alternatively. has yet to be d e ~ i d e d . ( n = 2 or 3) can form and exist long enough to target sensitive sites in the cell. ~ ~ ~ ' ~ ' ~ ' ~ ~ Whether or not Fe(III)Adr. The nonphysiological acceptor ligand 1. Jun Xiao. the availability of cellular iron is necessary for much. Adr serves as an oxidation-reduction site to deliver electrons to reduce Fe(II1) associated with ferritin or microsomal membranes to Fe(II). ' ~If it is~unstable. thereby excluding reaction with transferrin circulating in the blood. Petering. ' ~ ~ * ' ~ ~ ~ ' For the reduction of Adr to Adr--.reactions of Adr and reducing agents with ferritin or microsomes to liberate iron were conducted in the presence of the Fe(I1) binding ligand. P. has the stability to exist in the cell or in its surrounding medium. of the cytotoxic effects of Adr in tumor and cardiac cell models. Biochemical Mechanisms of Reaction of Adriamycin with Iron One class of mechanisms for the biological activity of Adr assumes that FeAdr. then the indirect mechanisms of interaction ~*' ~ between Adr and pools of cellular iron gain in interest. the iron complex may associate with membrane and initiate lipid p e r o x i d a t i ~ n . those with transferrin occur only at pH 6 and below. a role for iron in these processes has yet to be defined. It may bind to DNA through its anthraquinone ring system and positively charged amino sugar.(R=O) .(ROH) + Fe(II)Adr. ' ~ The' ~ ~ ~ . if not all. including NADPH cytochrome P-450 reductase and NADH dehydrogenase.0 (23) in which Fe(II1) is reduced. and those that have documented interactions between ICRF-187 and Adr. a quinone function is converted into its semiquinone and its side chain alcohol is oxidized.

the ligand must have the capacity to compete for the limited intracellular pools of iron or copper. Thus. and that are dependent upon the presence of copper in the host. several basic considerations have to be addres~ed.' l6 In contrast. the redox chemistry of the drug has to be accessible to cellular reductants such as ascorbate or glutathione. Bis(thiosemicarbazones) have been made that display powerful activity against established solid tumours in rodents.150 Oxidation Damage by Bleomycin. causes DNA strand scission that is largely eliminated by the addition of catalase. The parent ligand or metal complex must be able to diffuse or be transported across cell membranes to reach target sites of reaction. causes lipid peroxidation and induces DNA damage in uitro in the presence of reductants. Fe(m)Adr. Mono. ligand substitution reactions are unfavorable. which can be interrupted by the dismutation of H 2 0 2to O2 and H 2 0 by catalase.117*'18 They have formed the basis for continuing inquiry into the mechanisms of action of cytotoxic copper complexes. or can generate a hydroxyl radical with hydrogen peroxide. Adriamycin and Other Cytotoxic Agents Fe(II1)-anthracyclineis regenerated. which.11g The N2S2metal coordination site in these ligands reacts with Cu2+ to form highly stable complexes. the cleavage of DNA by Adr is inhibited under iron-deficient conditions. these studies reveal the breadth of reactivity that can be selectively modulated to design new chemotherapeutic agents. The second shows that.~ proposed complex must have a relatively large The stability constant so that it can exist in the midst of an organism full of compounds with the potential to chelate metal ions. Examples of these properties are described below. presumably as an end product of the initial formation of superoxide anion."' Fe(m)Adr. and the hydroxyl radical is formed. until it is out of range for . DNA strand scission results from a cascade of reactions leading to oxyradical formation. features of the structure of the drug will direct it toward a specific site of localization and reaction. If the metal-free ligand is administered and metal complexation is to occur in vivo. in turn.' l2i1l 3 Ascorbate is also an effective reductant of Fe(rIr)Adr.' In cells. In turn. "-' 4 Redox Active Copper Complexes and the Design of New Drugs To design new metallodrugs that cause oxidant damage.. at least when Fe(m)Adr. can transfer an electron to dioxygen. The first result extends the iron-dependence of cytotoxicity described above to a particular site of reaction. is used. the redox potential declines as much as 100-200mV. can also react with glutathione to yield Fe(u)Adr.' l4 Biological Model Reactions of Fe(III)Adr.and bis(thiosemicarbazones) were among the first compounds that were deliberately constructed as metal-binding ligands for the purpose of inhibiting and killing tumor cells (Figure 7). As the number of alkyl groups increases. such as DNA. but is insensitive to the presence of added catalase.. Ideally. Once in cells. Fe(m)Adr.

+ CU(II)SR+ 0 2 . S . Alternatively. + H 2 0 . the oxidation of sensitive protein thiols may also elicit cytotoxicity..(right)bis(thiosemicarbazones)CuKTS and CuKTSM.H . such that CuKTS is reductively dissociated 250 times faster than is CUKTSM. Jun Xiao. (28) CU(I)+ H 2 0 .-* (26) (27) + 2H' -+ 0. The production of the hydroxyl radical may be responsible for the cytotoxic properties of CuKTS. and is at the lower end of the range of facile reduction by glutathione (E" = .' Synthesizing this information. Fulmer and W. reaction of the N4-dimethyl species with ascorbate. In particular.+ OH- (29) Reaction 28 may occur spontaneously or be catalyzed by superoxide dismutase or a copper complex.' CU(II)KTS RSH + -+ CU(I)KTS+ RS*(iRSSR) + H' (24) (25) CU(I)KTS RSH + + H + + CU(I)SR+ H2KTS Once CU(I)SRis formed. + CU(II)+ OH. R group substitution is directly correlated with antitumor and cytotoxic effectiveness..CU(II)SR CU(I)+ 3RSSR --+ 20. l 2 For ' . Cu has the capacity to couple the oxidation-reduction reaction of thiols or other reductants with 0.12' This trend is directly reflected in the rates of reductive dissociation of these copper complexes in Ehrlich tumor cells. the initial reaction between CU(II)KTS cells is thought to be the following: and . Petering.D. Nyayapati. Antholine 151 NH* R. E. Copper bis(thiosemicarbazones) are neutral complexes that also display a range of partition coefficients between nonpolar solvents and water. one can envision the following sequence of reactions: CU(I)SR+ 0.'~' In turn.240mV). 3 R4" Rl CuKTS R2 R 3 ' R4 CH3CH20CHCH3 H H CH3 C U K T S M ~ CH3CH20CHCH3 H Figure 7 Structures of thiosemicarbazonatoCU(II) complexes: (left) monothiosemicarbazones. P.

1. Copper monothiosemicarbazones. 26 The following sequence of reactions ' '-' CU(II)L RSH + + CU(II)-SR H + + (30) (31) CU(II)L-SR CU(I)L+ RS*(iRSSR) + CU(I)L+ 0.].152 Oxidation Damage by Bleomycin.'. Cu(Phen). [Cu(NC). copper monothiosemicarbazones differ from bis(thiosemicarbazonato)Cu(II) complexes in that they can form adducts with Lewis bases.. The cytotoxic activity of monothiosemicarbazones such as a-N-heterocyclic aldehyde thiosemicarbazones is enhanced upon formation of iron or copper complexes.-* CU(I)L+ RSH CU(I)SR+ HL + (32) (33) shows that a key distinction between the pathways of reaction of mono. solubility and partition coefficients are readily adjustable parameters. CuKTS exhibits substantial solubility in both water and 1-octanol. -+ CU(II)L O. alone or tethered to various DNA-binding domains.lO-phenanthroline). ' ~ That~alters their biological chemis~*' ~ try.. CU(II)( 1. which should modulate both initial uptake by cells and distribution between their aqueous and membrane fractions. then site-specific. causes DNA strand scission in vitro in the presence of reductants. acting as a site-directed catalyst for the generation of oxyradicals. if the ligand structure can be designed to favor association with particular sites within the cell. is an unstable oxidation state in the presence of 0. Thus. it is likely that the complex binds with directly to DNA. toxic species. tumor cells. whereas reoxidation is apparently not an important pathway for bis(thiosemicarbazonato) CU(I). to reactive.lO-phenanthroline). Indeed..catalytic generation of hydroxyl radicals can be maintained with this type of copper complex. 23 Because they involve tridentate ligands. which involves the formation of the hydroxyl radical or its equivalent.'24 In various model systems including cells.In principle. 28 An intriguing variation of this model involves the cellular chemistry of CU(II)( 2. because drug distribution becomes dependent in part upon the possibilities for adduct formation. its Fe(I1) complex is readily detected in urine. An important difference is that they do so while maintaining a modicum of structural stability. In contrast. like bis(thiosemicarbazonato) Cu complexes.'*127 Upon reaction of Cu(~~)(Phen).g-dimethyl. prefers nonpolar media.and bis(thiosemicarbazone) copper complexes is that direct reoxidation of CU(I)L competes with ligand substitution to form CU(I)SR (Reactions 32 and 33). Recent studies with copper phenanthroline complexes expand on the possibilities for site-directed oxidation-reduction chemistry of copper complexes. whereas CuKTSM. thiol and histidine-like adducts of 2-formylpyridine thiosemicarbazonato CU(II) (CuL) have been detected by ESR s p e c t r o ~ c o p y . can support reduction of 0. does not catalyze these reactions when a noncoordinating reductant such as ascorbate is ' . Cu(NC). Adriamycin and Other Cytotoxic Agents example. readily catalyzes the redox reaction of reductants with O. because Cu(r)(Phen). upon administration of 5-hydroxy-2-formylpyridine thiosemicarbazone to humans.

. a coordination site becomes available for rapid inner-sphere reaction with 0. Hacker.A. there appears to be a wide variety of reaction pathways involving copper complexes that can cause oxidant damage in cells.W. p.H. 53. Second.-* The adduct hypothetically participates in reaction rate enhancement in two ways. in Metal Ions in Biological Systems. Hajdu. p. 1. Basel.H. despite the large stability constant of Cu(NC). Nevertheless. + GSH + CU(II)(NC)SG NC + H + + (34) -.E. Moldave. Byrnes and W. J. Nyayapati. because CU(I)(NC). H . Petering and W. upon internal oxidation-reduction. New York. 133. 9. New York. Elsevier. G . Marcel Dekker. W. Powis. H. by thiols.R. in unpublished model experiments. 1990. is which is rapidly reduced upon uptake into tumor cells. E. Interact. Kozarich. 1994. 1985. Rev. In summary. Jun Xiao. in The Toxicity o Anticancer Drugs. Hodgson.. S. p. Stubbe and J. 3 D.129”30 However.1107. Sigel. competes effectively for CU(I). 4 J.. and that it then serves as the site for catalysis of reduction of 0. Hecht.P. p. Volume 19. the prevalent tripeptide thiol in cells. stable in aerobic solution. + CU(II)(NC) 0. ed. at pH 7. as in Reactions 26-29. in Progress in Nucleic Acid Research and Molecular Biology.4 of approximately lo”. Powis and M. f Pergamon Press. Academic Press. 1991.’ An initial hypothesis to explain these observations was that glutathione (GSH). Chem. Vol. Petering.-Biol. New York. P. ed. Antholine.W.(4RSSR) + (35) (36) (37) + + CU(II)(NC) GSH + CU(II)(NC)SG H + + + CU(I)(NC) 0. ed. Antholine.I~’This has led to a modified model for reaction.. References 1 G. Bend and R.E. 5 D. it has been shown that catalysis by the reaction mixture of Cu(NC). in which a glutathione adduct is a key intermediate: Cu(n)(NC).’. Philpot. 1988. is a highly effective redox catalyst for strand breakage of DNA. 2 J. ed.M. 6 S. Antholine 153 used. Acknowledgements The authors appreciate the support of NIH grant CA-22108 and American Cancer Society grant DHP 31. R.E. Chem. Cohn and K. These take advantage of different chemical characteristics of various complexes and may be able to target particular sites of reaction in cells. . CU(II)(NC)SG CU(I)(NC) RS. 73.D. 313. Kane and S.M. 1987. E. 225. in Reviews in Biochemical Toxicology. Petering.87. First. CU(II)(NC). and GSH is much more effective than that by CuSG a10ne. it undergoes oxidation-reduction faster than CU(II)SG does because the presence of NC in the coordination sphere raises the redox potential of Cu. Fulmer and W.

185-200.L. Yoshioka. Meister and C. p.W.C. G.210. Povirk. W. Silen. Stone. T. Sausville. 1967. Inoi-y. Biochem. 28 L.H.. Petering. Frank. Akkerman. A. 10 S. 1978. H. 1988. Takeuchi and H. 25 R. Proc..521. 1980. Rao.N. A. J. Chem. Natl. Pharmacol.T. . 1978.R.28. Jr. W. Antholine and D. Hilbers. 17.. 207.4273. Antholine and D. Biomol. 1990. Pandit. Acad.2076. Solaiman. Takita and K.. Adriamycin and Other Cytotoxic Agents 7 D. Natl.F.5983.H.A. Haasnoot. Peisach. Stubbe. 9 L. Antibiot.A. L. 1980.C. Biochem.W.J.W. Antholine and D.F. Cohen. Petering.H. Abelson. D.W. Antibiot.H.F. Med.A. Petering.154 Oxidation Damage by Bleomycin.245. Suhara. Sarkar. Han. J .489. Biochim. Povirk. Pharmacol. 1991. 32 R. 33 L. Grollman.302. J . Cerami and S. Science.A. S. Biochemistry. Haseltine.2.41. Sawa. P. J. in The Toxicity o Anticancer Drugs. Absalon.W. J . Steighner and L.. 15 E. 106. Stubbe. Petering. Richardson. Rao.B. Li and Q. 1984. Lyman and D. Umezawa. Mao.P. W.. B. Byrnes and D.2746. Stubbe and J.H. T.G. C. Pharmacol. L.21. Proc. Radiat. W. 17 D. Horwitz and J. 1991. H.. 14 E.A. Hacker. Inorg. D. Dynam. D’andrea and W. Sausville. J .134. Worth. Acad. C. Lornitzo. Sigman and C-h. M. M. in Genetic Responses to Metals. 16 K.H.W. Fulmer. 23 R. 17. Christner. Fong.E. Palo Alto. Petering. Freedman.D. F. 19 J. Cancer Res.B. Takeshita. Hori. J .H. T.137. 36 M. Stubbe. ed. 31 D.2601.575.27. 23. New York. Umezawa. Kozarich. Marino. Y. 22 R.32. Maeda. Hilbers. 1989. Antibiot. 34 M.. Sci. Byrnes and D.K. Biochem. 20 W. T. Horwitz. A.B. 35 A. T. D.A. Biochemistry. L.F. Y. Biochem.34. 8 G. B.-H...A.3608. Petering.F. 18 S.A. Y . Petering. Biochemistry.. 21 R. Walsh.S. Hutchinson. D.J. Petering. J.. J.24.4.1241. Moldawer. U.A. Petering. W. New York.J. Biophys.H. Petering. Saryan and D. Biochem. 1966. J. Ohtsubo. 1985.343. K. T. Wu. Biochemistry. 13 M. W. Frank.A. 29 L. Kiihnlein and F. J . 1310. E. Byrnes and D. Manogue. Kozarich and J. Inc. 1978. Petering. Res. 1982. 12 H.P. Res.126.. Radtke. Kruk and C.19. 17. Biochern.F. Powis and M. S.J. 7562. Sci. 15. Y.B. F. Absalon. D. 5275.E. Biochemistry. Absalon. J . 1989. 1994. Povirk. in Annual Review ofBiochemistry. Nucleic Acids Res.E.E.A. ed.H.20. B. Haasnoot and C. Ishizuka. Lyman.H.655. Stubbe. Vlassara.W.W. 1974.. Wu.L. 1989. Sem. Takeuchi and H.173.A. W. ed.J.W. 27 J.3. 37 L. 1994.2203. G.H. FASEB J.. 1994. J. R. U S A . Worth. Kiihnlein and F. 1995.38. Powis. 1990.F.H.201.. M. Lowry. Wei. H. J. J . E. 1978. Peisach and S. Antholine and D. Acad. Takita. Horwitz. Templin. 1977.L. Eur. 1990. 26 J.1396.8350. H. U S A . Kuo. Hutchinson. Biochemistry. P. Antholine.W. 3573. 1637. 1995. Taylor. 5808. Kozarich. Chen. Lornitzo. C. pp.E.75. 30 M.162. Marcel Dekker. Biochemistry.87.2740. Christner. Weir.50. 38 C. A. Jr. 1995.. D.449. W. Kozarich and J. 1978. Proc.W. Byrnes. Byrnes and D. Biochem. Steighner. Sci. Kozarich and J.211.E. J.75. Umezawa. 24 R. Acta.. 1989. Povirk. 12.L. Biochemistry. Han and R. 1993. 39 M.75. Takayama.W. p.W.2065. Vanderwall and J.-H. Peisach and S. 1993.34. f Pergamon Press. Struct. Byrnes.. Ohtsubo and H. 1982.48. U S A . Natl. J. Radiat. Annual Reviews. 11 E. Solaiman.G.. Saryan.

91. A. Turner.B.H. Soc. Petering. D. Westre. Biol. Reddy. Biochemistry. M. J .16. J . Proc. Loeb.J.. Dattagupta. T. Solomon. Turner. Stubbe. W.A. 1980. 1979. Antholine and D. SOC. A. J. Antholine 40 41 42 43 44 155 M.1559. Li. Burger. Horwitz. 57 R. 1982.950. Li.B.H.E. Burger. 52 D. Petering.11636. W. Peisach and J. L. Muraoka. Povirk. Sci.33. 66 W.21478. Arcamone.117. Beery.. E.18. Otvos and D.H.H.J.615. Peisach and S.43.A.F. 43 10.665.M.B. 1979. Templin. J . Sugiura. Nakatani. 258. Fulmer. 1984. 203.E.A. J .A.W. Biophys.J. 21. Drugs and the Pharmaceutical Sciences. Chem. 50 R. Wang and S. Horwitz. C.19. Antholine and D. Nakamura.H. Chem. D. 56 P.A..5319. 1994. Chem. USA.E. Natl.F. 1982. Acad. J . Zhao. W.23.P.-H.264.E.B.. USA. Takita and H. p. 47 R. Lyman.118. W. 1985. Fuji. Petering.M. 1990. X. Saryan.S. 48 J. Biol.114. 1997. Henner. Biochemistry. S. Natrajan and S. Petering.45. Butler. 55 J. Acad. Tang. Hodgson and E. J . 49 T. ed. T. Fulmer.4367. Biol.33. 1984. Y. 12 299. 68 B.Wu.257. Antholine and D. C.1309. Xu. A. W. Xu. 1978. M. Jun Xiao.W. Biol. Petering. 61 M. Am.M. J. Am.E.-H.. D.I. Proc.. K. 6185. Horwitz. H. 373. S. 63 R. Turner. 70 L. Hogan. Petering. W. Biochemistry. S.E. K. Saryan and C. Antholine and D. Petering.M. Biol. Miinck and J. M. Burger.267. R. Petering. 73 F.E.944. C. Sam. Ottenbrite and G. 1992.X. 1070.E. Cancer Res.E. Nettesheim.B.D.118. Povirk. S.H. J. Volume 24. Chem.3372. Kross.M. Brown. L. Antholine and L. H .H. E.B. Natl. 1983. Vanderwall.. Tang and J. 1994. Chem. Hedman. V.256. 1988.1268. 1994.M. Kane. DeRose.116. Natrajan and S. Soc. Biol.36. W.4193. W. J . W. J . L. Petering. Petering.. 1977. Meares.. Kuwahara and Y. Am. 1994. Biochemistry. Chem. 1984. Biol. Iitaka.47.J. Crooke. Chem.78.A. Chem. Wu. Biol. Zaleski. Chien.H.M.lO 899.X.H. X.85. Mao. Huang.M. J .D. 1761.5250. C. KO. 254. 1992.E.-N. 1992. Biochemistry. L. Wittenberg. Hogan and N. Lui. J. J. Sci. S. 3641. Biol. in Anticancer and Znterferon Agents. Res. S. C. A. Chem.. Kozarich and J.E. Garnier-Suillerot. Umezawa.F. Galvan and S.D. 1979. Y.96.H. Vanderwall. DeRose and D. 1995. 1981.-J.M. Murty.269.X. Chem.H. Chem. J . Am.528. Stubbe. Haseltine. .6278. Byrnes.23. J.S. 72 S. Horwitz. Burger. Antholine and D.1992. Nyayapati. Chem. Soc. Biochemistry. Biochemistry. Marcel Dekker.5995. Lui. 58 C. 1981. P.20.. Kent. E. Commun. 1997. J..2459. J.H. J . R. Peisach. F. Hamamichi. 46 R. Buechner and N. P. Antholine and D.. Chem. Biochemistry. E. W.31. Wu.M.P. Peisach. Biochemistry.267. W. Xu. 53 W. Antholine and D.. Antibiot.7517. Soc. Hecht and W. J . Kozarich and J.E. Biochem. 1981. New York. unpublished information. 59 R.W. Antholine. 45 R.. 1996.4. S . J . Hecht.. Chem.. Petering. Biochem. Carter. Horwitz. Biochemistry. 64 W.M. 67 D.907. 1996. Am.. 69 J.M. T. 1989. 60 P. Kozarich and J. Fulmer and D. K. Fulmer and W.. Hecht. Hecht.T. B. Bid. Chem.1281.E. 265. Grollman and S. 54 W. 51 J. 71 N. Albertini and A. J. Pharmacol.E. Petering. 1996. Chem. 65 Q.W. Vanderwall.271. Peisach and S. J . Sakamoto. Stubbe. Dattagupta. J .B. Biol. D. Chang and C. Chikira. S.E.2268. 62 M. Petering. Chikira.. J .

Gelvan.319. 1990..8. Biochem. 1984. Niggeler and V. 319.5645.277. Petrescu. Byrnes.P.A. 81 D. Saltman and A. Biochemistry.W. 104 S. C. ed. Olson and P. Res. Tritton.4551. Etii-. Petering. 97 B.. Gianni and C.D. Kennedy and E. Vigano. Biophys. Tosi and F.V.. Biochem.283. 1990. Mimnaugh. Liu. Pathol.I. Biochem.S. Doroshow.... 1991. Malatesta. 93 R. Stacy. 84 H.26.B.E.B. Arch. 108 E. Pedersen and N.684. L.20.317. Lornitzo.. 77 H. 101 L. Kala.H. 1989.. C. Gianni. Biochim.L. New York. Biochem. Gelvan and A. 78 S. FEBS Lett. 89 K.. Nyayapati. 99 S.P. 1987. M. R. 1152. Powis. 87 K. Minotti. Pharmacol. Molec. Biochem. Ferrans. R. Abstracts of the Meeting of the American Society for Biochemistry and Molecular Biology. Acta.585.268.284.D. P. 1988. Tallent and J.C. 75 G. 1988. 1984.W. 1990. 106 C. Znorg.4. Garnier-Suillerot.378. Vigano. Green. 266. Green.F.248.97. 103 D. C. Arch. J .G.23. G.176. 103. Mol. 1994. 44. Bernaldo.. 1984. Myers.H.19. Biochem. Lanzi. Samuni. Biophys. 76 B.. Jordan.E. 1993. Biochem. 85 L. Niggeler and V. Gianni. 1981. J. 100 S. J . Mudd.25. 17.L.30. 1988. Feit.39. 83 E.. Gross and M. Minotti. Biochemistry. F. Fr. Powis and M. Minotti. 1996. Koch. V. Mailer and D. C..50. Pharmacol. Prog. Rowe. Speyer. Arch. Bachur. Hacker. P. Murphree. 98 B. 1991. Sanger. 95 P. Tewey. R.745..F. 1985.277. Rex.26. Massoud and R. Biochemistry.48. Arch.. Porumb and I.2085. in The Toxicity of Anticancer Drugs.W. J.398. Abstract 1212.1986. . 1988. P..48. 142. M. Chem. T.H..1104. Sinha. Alderton.6. Thomas and S. Chem.S. Am. Natl. R. B. Agents Actions. 1984. N . 288. Zeleniuch-Jacquotte. Arch. Science. Agents Actions. Winterbourn and H. 616. FASEB J. Enyl. A...1289. Quigley and A. Slater. Demant.928.. Biochem. Ughetto. 102 G. Berg.268. Corbett and N. 1987. Med. 2892. 1991.571. Petering.259. E. Malatesta. 90 A. Trush. Biophys. M. Demant. Lown.3076..-S. Res. Hasinoff. Pergamon Press. 1990. Rich. Sartorelli. Quesada and D.. Xiao. Cancer Inst. 1990. F. Biol.A. Muggia. 94 H.3972. 106. Sutton. J. Smith and A. A1467. F. Pharmacol. J. Eliot.2.1104. Cederbaum.D. L.268. Biophys. 1993.24.C. L. 1981. Mol. 1983.55. Biophys. 92 J.P. 105 G. J . Berger and K. E. 1976..J. M. Afshan.R. Cancer Inst.H. Nyayapati. Cancer Res. A. 88 D. C. 80 H. Yang. C..F.H.-J. J . Arch. Lavelle. 5064. 1984. Hasinoff and S. Toxicol. Biochemistry. Blum and F. Mushlin. Adriamycin and Other Cytotoxic Agents 74 J. Med. Schechter. Wang. Pan. 1986.M. K.M. Ward. G. 1985. Aust. FASEB J. G. Rey. N. 79 E. 82 S. R. Biochem. 1990. L. Natl. 96 J. W.R.B. 86 G. 1993. Kramer. Biophys.J. Chem. Breslauer. Dubin. L. Remeta.J.326. Lornitzo. Biophys. Toxicol.A. 107 G. Biol. Byrnes and D.J. 184. Cancer Res. Rad. M. Samuni.4. Halligan and L.D. Kukielka and A. Vile. 32.F. 118. Lanzi.H. L. 91 G.80. A.C. Wernz.156 Oxidation Damage by Bleomycin. Gaudiano and T. G. Petering. 466. Osheroff. p.D. 1989. Cancer.649.L. Biochem. Biophys.80. J . Cell..5136.30.C.

3.H.W. Byrnes. Biochem.H. .207. Biochemistry. 1990. Kukielka and A. Johns.6820. R. Myers. R. Cancer Res. Sartorelli and D.12.. Biochemistry. 1977.27. Winkelmann.E.B. A. 717. Gokey.I.T. Saryan and D.Greene. American Chemical Society. Petering. p. 1981.457. C. Chem. 2. Antholine. Myers.H. Toxicol. 125 L. 128 R. 112 J.H. (Pt l). Myers. Petering.H. in Antineoplastic and Immunosuppressive Agents. Antholine. Fulmer and W. in Metal Ions in Biological Systems. 113 H. 115 C. Miura.H. K. Biol. Antholine and D. 1994. 1985.H. Stability Constants.A. Jun Xiao. Martell. C. ed.. Supplement No. to be submitted for publication. Sillen and A. Petering. Ogiso. Sigman and C-h.283. 129 R.124. Tullius. 1980. Saryan. M. Byrnes. Fr. Pharmacol.A. Springer-Verlag. Gianni. Cell Biol.89. J . H. 1707. ed. Krishamurti. Minkel. p. 2. Hasinoff. S .G. P.261.74.16. Knight and D. E. Petering. Rep.. A. Petering. 122 L. Mayer. Petering. Petering and H.C.569. 1992. DC. 118 K. Byrnes.H. 126 R.H. Cancer Chemother. 130 L.22.. Metal-DNA Chemistry. 793. Xu and D. 116 J. Vol. Petering. to be submitted for publication. Eliot. Handbook of Experimental Pharmacology 38. 1971.38. L. 131 M. Rad.E.260. Antholine and D. J. T. Petering. C. Petering.X. Y. Chem.M.W.1595. Blumenfield. Biophys. Rad. Biol. Mol.7046. Zweier.L. W. S.H. Berlin.H.58.29. 121 D. 111 B.H.B. Krishnamurti and D... K. Handbook of Experimental Pharmacology 38. Biochem. 1975.68. 127 D. Hong. Vol. Gianni and C.G. B. Biol. Simone. p. Pharmacol. Vol. Med.326. Petering. Petering. Sinha. 1990.1331. The Chemical Society. Bethune and J. Arch. 1990. Johns. Krakoff. Etcubanas.. I . 119 D. W. Sartorelli.K. Ankel.E. E. T. Cederbaum. Xiao and D..G. Levy and C. p. 114 T.E. J. 1978. Inorg.. C. ed. Agrawal and A. Bermke and D. Med. Mailer. Biochemistry. Biochem.B.356. 1989. L. Berlin. J . Med. ed. 120 D. Mohan. 1992.G. Tan.H. 841. 1984.13. Antholine 157 109 E. Burchenal. Bioinory. 124 W.W. Muraoka and T. Gianni.A. S.. Nyayapati. Muindi. 1979. Fr. 117 D.928. Petering. 110 L. Logan. E.21.C. 24. L. Marcel Dekker. Pharmacol.S. 197. W. 123 I. Q.469.Klecker and R.23. L. 1975.. V. Mao and D. 1985. W. Myers. H .. Springer-Verlag.E. p. Chem. 2. 1218. London. in Antineuplastic and Immunosuppressive Agents. Sartorelli and D. Basel. 1974.C. Washington.D. 1982. Chem. Gianni and C. Sigel. Antholine and D.30.A. Saryan.E.C.H. A. Petering. 1974.

in disease. 116 versus cisplatin. 149 Anti-HIV activity. 109 combination with paclitaxel. 144 polynuclear platinum.126 clinical resistance. 116 structure specific damage. proteins. 7 DNA damage. gold compounds. 138 metallobleomycins. 37 . 129 Glutathione. anti-tumour activity. 114 mechanism of resistance. 47 Aurothioneins. SOD mimics. 137 Carboplatin dose intensification. 117 Gold compounds and adriamycin. cooperative effects. 114 Co bleomycin. 146 bleomycin. 150 dinuclear and trinuclear platinum compounds. 138 role of iron. 79 Anti-tumour activity adriamycin. and cisplatin resistance. 138 DNA binding bleomycin. 136 cisplatin and analogues. 150 Diagnostic agents.Subject Zndex Adriam ycin anti-tumour activity. 118 resistance. 119 mechanism of action. 40 and hemoglobin. 150 Anti-tumour drugs BBR3464. 113 modulation of resistance. modulation by metal drugs. 116 metabolites. 47 Auranofin. 126 comparison with cross-linking agents. 115 versus carboplatin. 6 Chelating agents. 140 properties related to DNA strand scission. 143 metallobleomycins. 126 strand scission. 127 DNA binding. 47 metal-based drugs. metal complexes. 116 gold complexes. 146 aspects of clinical use. 119 mechanism of resistance. 138 cisplatin. 149 and glutathione peroxidase. 51 Anti-inflammatory activity. 38 BBR3464 anti-tumour activity. 129 DNA binding. 110 Cellular response. 125 gold compounds. 149 reaction with iron. 132 Bleomycin anti-tumour activity. 110 copper compounds. 148 cytotoxic mechanisms and the role of iron. 120 dose intensification. 110 Copper complexes. design of new drugs. 136 clinical use. 2 Cisplatin clinical properties. 136 DNA strand scission.

matrix metalloproteinases. 12 cellular distribution. 47 biochemistry and pharmacology. 16 lithium carbonate as drug.83. 26 cyanide metabolites. 81. 16 value in psychiatry. 88 Platinum compounds anti-tumour activity. and gold. and gold. direct. 118 oral activity. 4 as medicinal targets. 79. 33 Vanadium compounds as insulin mimetics. 118 metabolism and ligand displacement. 37 anti-HIV activity. 113 cisplatin analogues.Subject Index 159 Nitric oxide and the cardiovascular system. 28 attachment to anti-tumour agents. 12 Manganese compounds as SOD mimics. 6 Metal ions in disease. as diagnostic agents. 19 isotopes. 30 Au(II1) oxidation state. 35 rheumatoid arthritis. 33 chemistry. use of chelating agents. 2 . 4 Metal complexes.95-97 chelation and tissue uptake. 80 Matrix metalloproteinases. 20 biochemistry. 7 Metal drugs chemotherapeutic uses modulation of cellular responses. 72 and the nervous system. 124 Polyamines. 18 cellular localization. 69 and the resporatory system. 28 Lithium and AIDS. 32 oxidation states in viuo. 29 chrysotherapy. 15 inositol lipid cycle. 146 JM-216. 2 Metalloproteins. anti-tumor activity. 101 Zinc fingers and human immunodeficiency virus. 77 assay for activity. 29 protein chemistry. 110 diaminocyclohezane (DACH). 66 and the immune system. 111 new drug development. indirect. 79 of macrocyclic polyamines. 112 polynuclear. 61 nitrosothiol production. 51 Au(1) oxidation state. 118 pharmacokinetics. platinum compounds. (SOD) activity. 110. 124 iproplatin. reaction with adriamycin. 60 measurement in chemical and biological systems.93. 92. 14 distribution in body. 31 auranofin. 124 trinuclear. 51 anti-tumour activity. 127 Rheumatoid arthritis. 42 history of medicinal use. as drug targets. 73 physical properties. 13 tissues distribution. as drug targets. 12 new uses. 28 Superoxide dismutase. 93 tissue uptake. 126 chemistry. 84 Thiol shuttle. 59 potentiation by SOD mimics. 4 Zinc proteins. 13 at cell periphery. 32 cullullar chemistry. 15 measurement in chemical and biological systems. 78 mimics. 101 historic uses. 18 biomedical uses. 47 as anti-arthritis drugs. 63 NO release from metal compounds. 11 chemistry. 64 cellular localization. 26 iron. 4 and serum albumin. 79 functional activity. 15 chemical properties. 118 dinuclear.

Sign up to vote on this title
UsefulNot useful

Master Your Semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master Your Semester with a Special Offer from Scribd & The New York Times

Cancel anytime.