Clinical Consequences 3rd Edition
A John Wiley and Sons, Ltd., Publication
ayyappan
9780470010297.jpg
Clinical Consequences 3rd Edition
A John Wiley and Sons, Ltd., Publication
This edition first published 2009 2009 John Wiley & Sons Ltd
Registered office John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom
For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com.
The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
The publisher and the author make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of fitness for a particular purpose. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for every situation. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. The fact that an organization or Website is referred to in this work as a citation and/or a potential source of further information does not mean that the author or the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work was written and when it is read. No warranty may be created or extended by any promotional statements for this work. Neither the publisher nor the author shall be liable for any damages arising herefrom.
Library of Congress Cataloging-in-Publication Data
Crichton, Robert R. Iron Metabolism – From Molecular Mechanisms to Clinical Consequences / Robert Crichton. – 3rd ed.
p. ; cm. Includes bibliographical references and index. ISBN 978-0-470-01028-0
1. Iron–Metabolism. 2. Iron proteins. 3. Iron–Metabolism–Disorders. I. Title. [DNLM: 1. Iron–metabolism. 2. Cells–metabolism. 3. Iron–physiology. 4. Metabolic Diseases–
physiopathology. 5. Metalloproteins–metabolism. QV 183 C928i 2009] QP535.F4C75 2009 572′.5174–dc22
2008042470
A catalogue record for this book is available from the British Library
ISBN 978-0470-010280
Typeset in 10/12pt Times-Roman by Laserwords Private Limited, Chennai, India
Printed and bound in Singapore by Fabulous Printers Private Ltd
Cover image: The front cover shows the 24-mer ferritin heavy chain, exhibiting a spherical shape with a central nearly spherical cavity for iron binding. Figure reproduced with permission from Human Blood Plasma Proteins by Johann Schaller, Simon Gerber, Urs Kampfer, Sofia Lejon and Christian Trachsel (Wiley, 2008, ISBN 9780470016749).
Preface xi
1. Solution Chemistry of Iron in Biological Media 1 1.1 Aqueous Solution Chemistry of Iron 1
1.1.1 Oxygen Free Radicals 2 1.1.2 Iron Hydrolysis – a Ubiquitous Phenomenon 5 1.1.3 Hydrolysis of Iron(III) in Acid Media – Formation of Polynuclear
Species 7 1.1.4 Ageing of Amorphous Ferrihydrite to more Crystalline Products 9
1.2 Biomineralisation 10 1.2.1 Magnetite Biomineralisation by Magnetotactic Bacteria 12 References 15
2. The Importance of Iron for Biological Systems 17 2.1 Introduction 17 2.2 Physical Techniques for the Study of Iron in Biological Systems 19 2.3 Haemoproteins 24
2.3.1 Oxygen Carriers 24 2.3.2 Activators of Molecular Oxygen 28 2.3.3 Electron Transport Proteins 35
2.4 Iron–Sulfur Proteins 39 2.5 Other Iron Containing Proteins 45
2.5.1 Mononuclear Non-Haem Iron Enzymes 45 2.5.2 Dinuclear Non-Haem Iron Enzymes 50 References 56
3. Microbial Iron Transport and Metabolism 59 3.1 Introduction 59 3.2 Siderophores 63
3.2.1 Iron Transport Across the Outer Membrane in Gram-Negative Bacteria 66
3.2.2 Transport Across the Periplasm and Cytoplasmic Membrane 75 3.2.3 Iron Release from Ferric Siderophores and Ferric Reduction 79 3.2.4 Fe2+ Transport Systems in E. coli 80 3.2.5 Fe3+ Iron Acquisition by Pathogens 80
3.3 Intracellular Iron Metabolism 86
vi Contents
3.4 Control of Gene Expression by Iron 88 References 94
4. Iron Uptake by Plants and Fungi 103 4.1 Iron Acquisition by Plants 103
4.1.1 Introduction 103 4.1.2 Iron Acquisition by the Roots of Plants 105 4.1.3 Long Distance Iron Transport 110 4.1.4 Intracellular Iron Transport 112
4.2 Iron Acquisition by Yeast 118 4.2.1 Introduction – Pathways for Iron Uptake 118 4.2.2 Cell Surface Reductases 119 4.2.3 High Affinity Iron Transport System 120 4.2.4 Low Affinity Ferrous Iron Transport 125 4.2.5 Siderophore-Mediated Iron Uptake 126 4.2.6 Intracellular Iron Metabolism 129 4.2.7 Iron Transport in Other Fungi 130 4.2.8 Regulation of Iron Uptake/Homeostasis in Yeast 131 References 131
5. Cellular Iron Uptake and Export in Mammals 141 5.1 The Transferrins 141 5.2 Structure of Transferrins 143 5.3 Transferrin Iron Binding and Release 146 5.4 Iron Uptake by Mammalian Cells – Uptake of Transferrin Bound Iron 151
5.4.1 The Transferrin Receptor 151 5.4.2 The Transferrin-to-Cell Cycle 155 5.4.3 Transferrin Binding to its Receptor 157
5.5 Cellular Iron Uptake and Export 161 5.5.1 Red Blood Cell Precursors 162 5.5.2 Tissue Macrophages 164 5.5.3 Hepatocytes 167
5.6 Uptake of Iron from Other Sources than Transferrin 168 5.7 Nontransferrin Bound Iron 169 5.8 Ferritin Bound Iron 170 5.9 Haptoglobin and Haemopexin as Iron Transporters 170
References 173
6. Intracellular Iron Storage and Biomineralisation 183 6.1 Intracellular Iron Storage 183
6.1.1 Ferritin: Distribution and Primary Structure 184 6.1.2 Three-Dimensional Structure 186 6.1.3 The Mineral Core 198 6.1.4 Iron Deposition in Ferritin 200 6.1.5 Iron Mobilisation from Ferritin 207 6.1.6 Haemosiderin 209
Contents vii
7. Intracellular Iron Metabolism and Cellular Iron Homeostasis 223 7.1 Intracellular Iron Metabolism 223
7.1.1 The Labile Iron Pool 224 7.1.2 Mitochondrial Iron Uptake and Metabolism 226 7.1.3 Haem Biosynthesis 228 7.1.4 Iron–Sulfur Protein Biogenesis in Eukaryotes 233 7.1.5 Intracellular Haem Degradation – Haem Oxygenase 240
7.2 Cellular Iron Homeostasis 247 7.2.1 Structural Features of IREs 251 7.2.2 Hereditary Hyperferritinaemia Cataract Syndrome 253 7.2.3 Iron Regulatory Protein 1 254 7.2.4 Iron Regulatory Protein 2 259 References 261
8. Iron Absorption in Mammals, with Particular Reference to Man, and Regulation of Systemic Iron Balance 271 8.1 Iron Metabolism in Man: An Overview 271 8.2 Sources of Dietary Iron in Man and the Importance of Luminal Factors 273 8.3 Iron Losses and Requirements for Absorbed Iron 275 8.4 Molecular Mechanisms of Mucosal Iron Absorption 276
8.4.1 Iron Uptake at the Apical Pole 279 8.4.2 Iron Transfer Across the Mucosal Cell 282 8.4.3 Release of Iron at the Basolateral Membrane and Uptake
by Apotransferrin 283 8.5 Regulation of Iron Uptake by the Enterocyte 284 8.6 Regulation of Systemic Iron Balance 285
References 294
9. Pathophysiology of Iron Deficiency and Iron Overload in Man 299 9.1 Introduction: Acquired and Genetic Disorders of Iron Metabolism 299 9.2 Homeostatic Control of the Internal Milieu and Consequences of Its
Disruption 300 9.3 Iron Overload Syndromes 303 9.4 Primary Iron Overload, Hereditary Haemochromatosis (HH) 303
9.4.1 HFE Haemochromatosis (Type 1) 304 9.4.2 Other Types of Haemochromatosis 308 9.4.3 Nonhaemochromatotic Primary Iron Overload 311 9.4.4 Treatment of Primary Iron Overload 312
9.5 Secondary Iron Overload 313 9.5.1 Treatment of Secondary Iron Overload 315
9.6 Iron Deficiency and IDA 318 9.6.1 Epidemiology 318 9.6.2 Causes 319
viii Contents
9.6.3 Clinical Stages of Iron Deficiency and Laboratory Diagnosis 320 9.6.4 Treatment of Iron Deficiency 322
9.7 Anaemia of Chronic Disease 324 9.8 Conclusions 324
References 325
10. Iron and Oxidative Stress 335 10.1 Introduction to Free Radicals 335 10.2 Reactive Oxygen Species (ROS) 336 10.3 Cytoprotective Enzymes and Antioxidants 338
10.3.1 Superoxide Dismutase 338 10.3.2 Catalase and Glutathione Peroxidase 340 10.3.3 Glutathione 340 10.3.4 Thioredoxin System 342 10.3.5 Haem Oxygenase 342 10.3.6 Ferritin 343 10.3.7 Low Molecular Weight Antioxidants 343
10.4 Ageing and Cytoprotection 345 10.5 Oxidative Stress 348
10.5.1 NFκB 348 10.5.2 Caspases 350 10.5.3 JNK 353 10.5.4 MAPK Signalling Pathway-amplification Cascade 353
10.6 Cyclin Dependent Kinases 356 10.7 Deregulation of Calcium Homeostasis and Oxidative Stress 357 10.8 Nitric Oxide and Cyclic Guaylate Cyclase 358 10.9 Activation of cAMP Dependent PKA 360 10.10 Importance of Iron, ROS and RNS in Phagocytic Cells 360
10.10.1 Macrophages 360 10.10.2 Glial Cells 364 References 365
11. Brain Iron Homeostasis and Its Perturbation in Various Neurodegenerative Diseases 371 11.1 Introduction 371 11.2 Mechanisms for Iron Transport into Brain 372 11.3 Importance of Iron in the Developing Foetus 376 11.4 Iron Uptake and Turnover Within the Brain 377 11.5 Importance of IRPs in Brain Iron Homeostasis 378 11.6 Brain Iron Speciation 379 11.7 Neurodegenerative Diseases 379
11.7.1 Parkinson’s Disease 381 11.7.2 Alzheimer’s Disease 389 11.7.3 Frederich’s Ataxia 393 11.7.4 Aceruloplasminaemia 395 References 398
Contents ix
12. Interactions Between Iron and Other Metals 403 12.1 Introduction 403 12.2 Interactions Between Iron and Essential Metals 404
12.2.1 Mars and Venus – Iron and Copper 404 12.2.2 Iron and Zinc 414 12.2.3 Iron and Manganese 420 12.2.4 Iron and Cobalt 424
12.3 Iron and Toxic Metals 429 12.3.1 Iron and Aluminium 429 12.3.2 Iron and Lead 432 12.3.3 Iron and Cadmium 434 References 435
Concluding Remarks 445
Preface
Two roads diverged in a wood, and I – I took the one less traveled by, And that has made all the difference.
Robert Frost ‘The road not taken’ from Mountain Interval (1916).
I still feel that ‘the road not taken’ reflects many of the choices that I made early in my career – deciding to read biochemistry instead of continuing with the rest of my schoolmates to read chemistry; carrying out my final year project in a protein chemistry laboratory when the rest of J.N. Davidson’s Department in Glasgow was working on nucleic acids; going off on a post-doctoral fellowship to Germany, without having any more competence in the language of Goethe and Schiller than that required to translate a passage of scientific German (still a requirement for science graduates in the 1960s) with the aid of the faithful Patterson’s Ger- man/English scientific dictionary. Yet, on more considered reflection of the really significant roads that one has taken over the years, it is striking how many of them seem to involve an important element of predestination1.
My ‘Blood and Iron’ connection began during my final year undergraduate project in Glasgow, when my supervisor Dr George Leaf suggested that I should work on the haemopeptide from horse heart cytochrome c. Although my subsequent doctoral thesis involved gas chromatography of amino acid derivatives2 and mass spectrometry of peptides, I was working in a laboratory just next door to that of Hamish Munro, who, with his doctoral student Jim Drysdale, was studying the regulation of ferritin synthesis in rat liver. In May 1966 my head of Department, J. Norman Davidson, informed me that there would be a lectureship vacant in the autumn, so I had better write up my thesis3, and – by the way – if I wanted to have a post-doctoral year abroad I should get going fast to organise it! An interview for the post of lecturer and, two days later, a doctoral viva (these were, after all, the golden sixties) in early September, led to me turning up on the doorstep of the Max-Planck Institut fur Biochimie in Munich in January 1967 with a PhD and a lectureship in Biochemistry in my pocket. There I began isolating and characterising insect haemoglobins
1The notion that events somehow conspire to lead to decisions which are not really taken, but rather impose themselves. I already referred in previous editions to the impact of the writings of Soren Kierkegaard, in many ways the philosophical father of predestination, on my thinking, particularly when I was envisaging entering theology.
2The method failed to be of any practical use for the analysis of hydrolysates of peptides purified by 2-dimensional paper chro- matography and electrophoresis, no doubt because of the lack of specificity of the detection system.
3My doctoral students will understand my reluctance to believe that it needs 6–8 months to write a thesis – I wrote mine between the end of May and the middle of August, while still carrying out final experiments and travelling to Edinburgh four nights a week for rehearsals with the Edinburgh Festival Chorus. We performed the Mahler symphony number 8 and the Britten War Requiem (with Peter Pears, Dietrich Fischer-Dieskau and Galina Vishneskya as soloists, the Melos Ensemble conducted by Benjamin Britten himself, with the Scottish National Orchestra and the Chorus conducted by Sir Alexander Gibson).
xii Preface
in Gerhard Braunitzer’s laboratory, working together with Volkmar Braun, which resulted in my first publication (Braun et al., 1968).
Back in Glasgow, no doubt influenced by Hamish Munro, I began my long involvement with ferritin. But, at the International Congress of Biochemistry in Switzerland, an invitation from Heinz-Gunther Wittmann to give a seminar in Berlin led to my leaving the permanent position in Glasgow for a nonpermanent position. So, from 1970 till 1973 I was a Senior Fellow of the European Molecular Biology Organisation (EMBO) at the Max-Planck Institut fur Genetik in Berlin. There, I continued my work on ferritin, but also worked on riboso- mal protein–RNA interactions. During my stay in Berlin (still then within the confines of the Wall), I travelled extensively, attending symposia and workshops on protein structure and on protein synthesis. Among the many colleagues and friends whom I met were Goetz Domagk, then Professor of Biochemistry at the Catholic University of Louvain, and Hubert Chantrenne, of the Free University of Brussels, whom I had entertained during his visit to Glasgow to give a lecture on protein biosynthesis in my capacity as Secretary of the student Biochemical Society4. Without my realising this, they played key roles in my being invited to succeed Domagk in autumn 1973 as chair of biochemistry in the Chemistry Department of the Catholic University of Louvain in Louvain-la-Neuve. In none of these decisions was I remotely involved – things simply happened. So it was that I came in a preordained fashion to protein chemistry, to iron and to Belgium.
My interest in the broader field of biological inorganic chemistry developed during the organisation, with Cees Veeger5, of seventeen Advanced Courses on ‘Chemistry of Metals in Biological Systems’ (with the financial support of the Federation of European Biochemi- cal Societies (FEBS), the European Science Foundation and the European Union). With the enthusiastic help of a devoted faculty, we have trained more than 750 young scientists in the techniques required to investigate the role of metals in biological systems. It is with pride and a great deal of pleasure that I continue to see on the platforms of both European and International Congresses of Biological Inorganic Chemistry many of the students who passed through the Louvain-la-Neuve courses. In many ways, the frustration of not being able to secure the funding to continue these courses over the last few years has resulted in my first venture into writing a textbook for students (Crichton, 2008).
A relatively limited number of inorganic elements play important roles in biology, the environment and medicine. Their relative abundance in the earth’s crust, in seawater, and examples of their specific functions, are presented in Table P.1 (as are those of a few selected nonmetals). The basic principles involved in the bioselection of elements conform to four fundamental rules: (i) abundance, (ii) efficacity, (iii) basic fitness for a given task, and (iv) evolutionary pressure (Frausto da Silva and Williams, 2001; Crichton, 2008). A rapid exam- ination of Table P.1 shows that abundance, for example, is not an adequate requirement for biological fitness (aluminium is perhaps the best example, and owes its inclusion to the fact that it has more or less been brought into our present day biological environment by man himself). Individual elements are particularly fitted for specific functions, often as a direct consequence of their chemical properties. Na+ and K+, which form complexes of very low stability and are therefore very mobile in biological media, are ideally suited for use in ionic
4In the same capacity I also entertained Feodor Lynen, whom I had got to know during my stay in Munich – the cost being, as I recall, a bottle of Glenfiddich after a well lubricated dinner!
5Cees and I were also involved as members of the Steering Committee of the European Science Foundation ‘Chemistry of Metals In Biological Systems’ Programme, which was largely responsible for the creation of both the Journal of Biological Inorganic Chemistry and the Society of Biological Inorganic Chemistry.
Preface xiii
Table P.1 Relative abundance and examples of functions of inorganic elements (and a few selected nonmetals) which play an important role in biology
Metal Crustal Seawater Examples of specific functions Average mg/l (ppm)
Sodium 2.8 × 104 1.1 × 104 osmotic control, electrolytic equilibria, currents Magnesium 2.1 × 104 1.4 × 103 phosphate metabolism, chlorophyll Aluminium 8.1 × 104 1 × 10−3 neurotoxic, solubilised by acid rain Silicon 2.8 × 105 3 prevents aluminium toxicity Potassium 2.6 × 104 3.9 × 10−2 osmotic control, electrolytic equilibria, currents Calcium 3.6 × 104 4.1 × 10−2 second messenger, muscle activation, biominerals Vanadium 135 2 × 10−3 nitrogenase, peroxidases Chromium 100 5 × 10−4 glucose metabolism ? Manganese 950 2 × 10−3 oxygen production and metabolism, structure Cobalt 25 4 × 10−4 B12 coezymes, alkyl transfer Nickel 75 7 × 10−3 hydrogenases, urease Copper 55 3 × 10−3 electron transfer, oxidases, oxygen transport Zinc 70 1 × 10−2 Lewis acid catalysis, regulation (DNA binding) Selenium 5 × 10−2 9 × 10−9 glutathione peroxidase Molybdenum 1.5 1 × 10−2 nitrogenase, oxidases, oxo-transfer Tungsten 1.5 1 × 10−4 dehydrogenases Iron 5 × 104 3 × 10−3 Oxygen transport, storage, activation and detoxification,.
electron transfer, nitrogen fixation, ribose reduction, etc.
Source: from Mason and Moore (1982).
equilibria and electrolytic circuits. Mg2+ is generally involved with phosphate compounds and phosphate metabolism, while Ca2+, in addition to its structural role in biological minerals like bone, plays a key role as an intracellular signalling messenger. Zn2+, redox inactive, is found in more than 300 enzymes, functions in many situations as a Lewis acid, but is also involved as a component of gene regulatory proteins. Copper, like iron is involved in many electron transfer reactions as well as in enabling cells to cope with dioxygen. Manganese, also redox active, constitutes the catalytic centre of the photosynthetic water splitting complex, oxidising two molecules of water to dioxygen (and in the process transforming our atmosphere from reducing to oxidising). And so one could continue; yet, if each of these elements have their own particular specificities with regard to biological function, we will consider one metal only (with the exception of a brief excursion into its interactions with other metals in Chapter 12) in what follows. This metal, which I consider to be of capital importance, is iron, which, as a glance at the table will show, has a multiplicity of functions. The reader will, I trust, forgive me this selectivity, for it is with iron that I have passed the last four decades, and it is the metal with which I am the most familiar.
As I set out on this next step along a road already well travelled, I would like to address myself first of all to you, my dear readers and colleagues, and to thank you for the enormous encouragement that you have given me over the last two decades to continue, and develop, the project which I began in 1990 and continued in the 2001 edition. It has been an important stimulus to undertake the preparation of this third edition, to seeing well-thumbed copies of both editions on your desks and bookshelves, and of hearing from so many how useful you,
xiv Preface
and particularly your students, have found them to be in giving a critical as well as panoramic view of iron metabolism. I continue to adopt the position that when writing a review (or even more important, an overview), it is important not simply to provide a seed catalogue of the current literature, but to take a reasoned position concerning the probability that one particular viewpoint is correct. As Aneurin Bevan6 pointed out, ‘We know what happens to people who stay in the middle of the road. They get run down’. But, enough of this; rather than the critique emitted by one of the characters in John Osborne’s, ‘Look back in Anger’ , ‘They spend their time mostly looking forward to the past’, let us address what has been happening since the last edition.
The importance of well-defined amounts of iron for the survival, replication and differ- entiation of the cells of animals, plants and almost all microorganisms is well established (Crichton, 2001). One notable exception is constituted by the so-called ‘lactic acid bacte- ria’, which include members of the Lactobacillus family together with some Lactococcus and Streptococcus . They are of particular importance in the manufacture of dairy products, including cheeses7 and yoghurt, as well as in the spoiling of milk. Their adaptation to growth in the presence of the strongly iron binding protein of milk, lactoferrin (Chapter 5), has led to the conclusion that they can grow in the absence of iron, although unequivocal proof of this is not easy to obtain.
However, for most living organisms, iron in excess is toxic, and iron deficiency is also a general problem in biology, which means that iron homeostasis is extremely important, both at the cellular and the systemic level. And since humans have little capacity to excrete iron, it follows, as was originally suggested by McCance and Widdowson (1937), that iron balance in man is primarily determined by iron absorption. Some examples of the multiple roles of iron in biology have been selected here, to give the reader a clear impression of the importance of this element. This panorama is by no means comprehensive, nor is it intended to be. It should, like the ‘ameuse geule’ served before the meal in many French restaurants, simply whet the appetite of the reader for what is to follow.
While iron is the fourth most abundant element in the earth’s crust, it is only present in trace concentrations in seawater (Table P.1). In particular, the surface waters of the Southern Ocean contain extremely low concentrations of iron (20–50 pM) thereby limiting primary production of phytoplankton (Martin and Fitzwater, 1988). Since the Southern Ocean exerts a major control on the partial pressure of carbon dioxide (pCO2) in the atmosphere, low rates of photosynthesis and biological carbon export in Antarctic waters result in macronutrients being largely unused. The resulting up-welled CO2 enters the atmosphere, sustaining the relatively high interglacial atmospheric pCO2 of the present day. Martin subsequently proposed (Martin, 1990) that natural variations in the atmospheric iron flux could ultimately regulate primary production in the Southern Ocean and influence the pCO2 of the atmosphere, potentially contributing to global warming of the planet.
Mesoscale iron addition experiments (FeAXs) have unequivocally shown that iron supply limits production in one third of the world’s oceans, where surface macronutrient concen- trations are perennially high (Boyd et al., 2007). The findings of these 12 FeAXs show that iron supply exerts control over the dynamics of plankton blooms, which in turn affects the
6The left-wing miner and trades union official from South Wales became Minister of Health in the 1945 post-war Atlee Government, and was responsible for establishing the UK National Health Service. The NHS was created sixty years ago, on the 5 July 1948, enabling the government to take over responsibility for all medical services and supplying free diagnosis and treatment for all.
7One is reminded of the classic reply of Charles de Gaulle, in response to the question ‘Why is France such a difficult country to govern?’ – ‘Do you know of any other country that produces more than 360 different sorts of cheese’.
Preface xv
biogeochemical cycles of carbon, nitrogen, silicon and sulfur, and ultimately influences the world’s climate system. In the Southern Ocean Iron Experiment in 2002, in which 1.7 tonnes of iron sulfate was dropped in the sea (Coale et al., 2004), the data accumulated imply that each atom of iron added to the sea could pull down between 10 000 and 100 000 atoms of car- bon out of the atmosphere by encouraging plankton growth, which captures carbon and sinks it deep towards the ocean floor (Bishop et al., 2004). Scaling up of such iron fertilization of the sea could have a real impact on the high level of carbon dioxide in the atmosphere, which is causing global warming, with some researchers estimating that, in the Southern Ocean alone, the technique could absorb 15% of the carbon dioxide build-up. However, ecologists are quick to warn that the technique could damage marine ecosystems in ways that remain to be established, and the future of massive fertilization of the oceans is still in a state of turmoil (Buesseler et al., 2008).
Iron added to so-called high nitrate low chlorophyll (HNLC) regions of the world’s oceans has clearly been shown to promote growth of autotrophic as well as heterotrophic microorganisms. What, then, are the molecular mechanisms by which marine microorganisms compete for the added iron (reviewed in Butler, 2005)? In common with other bacterial species many marine bacteria have been shown to produce siderophores (see next para- graph). The marinobactins, aquachelins, amphibactins, ochrobactins and synechobactins are siderophores isolated from distinct genera of marine bacteria. They all contain a unique head group that coordinates iron(III) and one of a series of fatty acid tails (C8−C18). The mari- nobactins, aquachelins and most likely the amphibactin lipopeptides are characterized by low critical micelle concentrations (CMCs). At concentrations exceeding their CMC, the apo mari- nobactins form spherical micelles which shrink in diameter upon coordination of Fe(III). Upon addition of excess Fe(III), the Fe(III)-marinobactin micelles undergo a transition to form vesi- cles (Martinez et al., 2000; Owen et al., 2005). These properties seem to have evolved as a common iron acquisition strategy for marine bacteria, the properties of which remain to be elucidated (Martinez et al., 2003).
Ironically, the propensity of some pathogenic bacteria, and particularly highly virulent strains, to infect their hosts, whether other microbes, plants or mammals, depends on their capacity to produce highly effective iron uptake systems. The first to be clearly identified was the aerobactin system for iron uptake in virulent strains of E. coli , which is encoded by the ColV plasmid (Williams, 1979). To combat the effectiveness of the siderophores produced by highly virulent bacterial strains, animals may synthesize specific proteins, like lipocalin, to bind and nullify their action. However, these countermeasures can be thwarted by at least one bacterium, Salmonella , glycosylating its siderophore (enterobactin/enterochelin) so that binding of the modified siderophore (now termed salmochelin) with lipocalin no longer occurs (Ratledge, 2007). This early conceptual observation has subsequently been underlined by the identification, not only of pathogenic solutions to gaining access to host iron supplies, but also to the increasing realisation that horizontally transmissable mobile genetic elements, designated as ‘pathogenic islands’ constitute a source of enhanced virulence, frequently associated with uptake of essential nutrients, like iron (Hacker and Kaper, 2000).
In previous editions of this book I have evoked the way in which many plants in the Season of mists and mellow fruitfulness (John Keats To Autumn) shed their leaves, and with it much of their mineral content, including the iron, so essential for photosynthetic electron transfer pathways. Come the spring, they must recuperate this iron from the soil and send it, sometimes tens of metres high in the air, to regenerate their photosynthetic capacity in the newly formed leaves. While we continue to enlarge our understanding of the uptake of soil iron by the
xvi Preface
roots, we are now beginning to have a much better idea of how iron is transported around plants. Since the publication of the last edition of this book, the second fully sequenced and annotated plant genome, that of rice (International Rice Genome Sequencing Project, 2005), has joined that of Arabidopsis (Arabidopsis Genome Initiative, 2000) providing scientists with the possibility to compare a monocot against a dicot plant in terms of their gene content and genome structure. However, it could also allow the identification of possible candidate genes which play a role in iron uptake, transport and homeostasis by analogy with fungal and mammalian systems. These two genomes are both relatively small in size (125 Mbp and 389 Mbp for Arabidopsis and rice, respectively), and an even more abundant harvest is awaited from the ongoing analysis of the much larger cereal genomes, such as barley, maize and wheat. The abundance of highly conserved repetitive elements has contributed to the considerable increase in physical size that is observed in these cereal genomes when compared with rice and Arabidopsis . Approximately 80% of the maize genome8 is composed of repetitive DNA mainly long terminal repeat (LTR)-retrotransposons.
Proteins with iron–sulfur ([Fe−S]) cofactors play important roles in many cellular pro- cesses, such as redox reactions, metabolic catalysis and the regulation of gene expression (Chapter 2). Mitochondria play a central role in [Fe−S] protein biogenesis and our under- standing of the way in which these clusters are assembled has evolved rapidly in the last few years. [Fe−S] clusters are constructed from cysteine as a source of sulfur, combining it with iron to synthesize the [Fe−S] cluster on scaffold proteins, and the cluster is finally incorporated into recipient apoproteins. In eukaryotes, such as yeast and human cells, more than 20 components are known that facilitate the maturation of [Fe−S] proteins in mitochon- dria, cytosol, and nucleus (Lill and Muhlenhoff, 2006). These components are also involved in other cellular pathways, such as the regulation of iron homeostasis or the modification of tRNA. The final step of haem biosynthesis, the insertion of iron into protoporphyrin IX by ferrochelatase, also takes place in the mitochondria.
In both [Fe−S] cluster formation and haem biosynthesis, the protein frataxin is thought to the play the role of an Fe2+ chaperone (Lill and Muhlenhoff, 2006; Zhang et al., 2005, 2006). The expansion of a trinucleotide repeat in the first intron of the frataxin gene is the origin of Friedreich’s ataxia9, the most common hereditary ataxia and the most prevalent cerebrellar ataxia among children and adults in Europe. This neurodegenerative disease is characterized by the accumulation of large amounts of iron within the mitochondria, as is also observed in a number of haematological disorders associated with defects in Fe/S protein biosynthesis. One such example is cross-linked sideroblastic anaemia with associated ataxia (Bekri et al., 2000), in which a mutation in a membrane transporter results in impaired transport of [Fe−S] clusters from the mitochondria to the cytoplasm.
I referred earlier to the studies of Hamish Munro on ferritin biosynthesis, which resulted in the identification in the 5’-Untranslated Region (UTR) of both H and L chain ferritin mRNAs of putative stem loops consisting of a highly conserved sequence of 28 nucleotides termed Iron Regulatory Elements (IREs). These IREs, are found in both the 3’- and 5’-UTRs of a number of other proteins involved in iron metabolism. With the discovery of iron regulated cytosolic RNA binding proteins, Iron Regulatory Proteins (IRPs), which bind to the IREs (Liebold and Munro, 1988; Rouault et al., 1988), it became clear how in many mammalian cells there is a reciprocal regulation, at the level of mRNA translation, of iron uptake and intracellular iron
8The 2.4–2.7 gigabase pairs (Gbp) maize genome is almost as large as the human genome (2.8 Gbp) and the determination of its complete DNA sequence is imminent.
9Ataxia – total or partial inability to coordinate voluntary bodily movements, particularly muscular movements.
Preface xvii
utilisation. IRP1 exists in an iron free form, which binds to the IRE stem loop, and an iron bound form, with an [Fe−S] cluster, which has aconitase activity. Our understanding at the molecular level of the conformational changes involved in this transition have been greatly advanced by the determination of the structure of the protein in both its forms (Dupuy et al., 2006; Walden et al., 2006).
In the previous edition we highlighted the impressive developments in our understanding of iron metabolism through the application of the revolutionary techniques of molecular biology. This resulted in the identification of new genes and their gene products involved in iron uptake and cellular utilisation, with the prospect of progressively approaching an understanding of their function. The role of many of these proteins has become increasingly clear in the last few years. But we can also hail the appearance of an impressive number of new genes and gene products in the increasingly complex network of reactions that constitute iron metabolism. Not the least impressive of these is the antimicrobialpeptide hormone, hepcidin, which has swept over the mammalian iron field like some kind of virus10 (which it is certainly not!).
The initial discovery in Rennes that hepcidin was significantly up-regulated (Pigeon et al., 2001) in the liver of iron overloaded mice was obtained by performing a suppressive subtrac- tive hybridisation between cDNAs from the liver of iron overloaded and control mice – the same strategy that had been employed to identify IREG1, the iron transporter up-regulated in conditions of increased iron absorption in the duodenum (McKie et al., 2000). This was followed by the serendipitous discovery in Paris that inactivation of the mouse hepcidin gene resulted in iron overload (Nicolas et al., 2001) – the authors were seeking to knock-out an adjacent gene, but the genetic ‘scissors’ used to excise it removed hepcidin expression as well. In summary (for more detail see Chapter 8), the way in which hepcidin exerts its key role in the regulation of systemic iron homeostasis is as follows (Nemeth et al., 2004; de Dominico et al., 2007). Hepatocytes secrete hepcidin, which binds to IREG1 (ferroportin), the iron efflux transporter expressed in all iron exporting cells. This results in the phosphory- lation, internalization, ubiquitination and ultimate degradation of ferroportin. When hepcidin concentrations are high, ferroportin degradation exceeds its rate of synthesis, resulting in decreased iron uptake from the intestine, and retention of iron within macrophages and hep- atocytes. In contrast, when hepcidin concentrations are low, ferroportin synthesis exceeds its rate of degradation, increased amounts of iron are released at the basolateral membrane of intestinal enterocytes, from macrophages (iron derived from senescent erythrocytes) and from hepatocytes into the extracellular fluid.
On account of our incapacity to excrete significant amounts of iron, pathological disorders of iron metabolism associated with excessive iron accumulation, principally in parenchymal tissues, are often observed in man. These diseases, designated as Hereditary Haemochromatosis (HH), are now known to be a family of conditions, characterised by excessive dietary iron absorption. The classical form of HH (type 1) is an autosomal recessive, HLA linked disease. It is the most widely prevalent form of HH, and indeed one of the most frequent genetic disorders in man, more common than cystic fibrosis, muscular dystrophy and phenylketonuria combined, with an estimated carrier frequency of 1 in 200 in Caucasian populations. The identification of the candidate gene, Hfe, the determination of its X-ray structure, the demonstration of its interaction with transferrin receptor 1 (TfR1), and the determination of the structure of its complex with the transferrin receptor, were all described in the previous edition. But it is only with the recognition that the site of action of HFE is not in the gastrointestinal
10Since the six publications in 2000 and eight in 2001, the total to date (August 2008) is in excess of 1100 (PubMed).
xviii Preface
tract, but rather in the liver, that we have recognised that, as with all of forms of HH, the primary defect in type I HH is in the regulation of hepatic hepcidin synthesis. Mice with hepatocyte specific Hfe ablation develop systemic iron overload, whereas enterocyte specific or macrophage specific disruption of Hfe results in a normal iron phenotype (Vujic Spasic et al., 2007, 2008). Taken together with the lack of progression of hereditary haemochromatosis in human liver transplant recipients receiving normal livers, it seems clear that hepatocytes are the principal cells controlling systemic iron homeostasis, through the secretion of hepcidin. Indeed, we might consider that hepcidin is to iron homeostasis what insulin is to the homeostasis of blood glucose (Pietrangelo, 2007).
Iron is particularly indicated for catalysing reactions which necessitate a free radical mech- anism, like the transformation of ribose to deoxyribose. This reaction, which is of primordial importance for DNA replication and cell division, is catalysed by ribonucleotide reductases, the best class of which has a dinuclear iron centre and a stable tyrosyl free radical as cofactor. However, as we will see in Chapters 10 and 11 (which have been prepared in collaboration with my long-standing (and long suffering) collaborator Roberta Ward), iron can also catalyse reactions with molecular oxygen to produce highly reactive oxygen species (ROS). This is the so-called oxygen paradox – oxygen is an absolute necessity for our energy-economical anaerobic life style, yet it is a potential toxin. On the plus side, the arrival of oxygen enabled organisms which developed respiratory chains to extract almost 20 times more energy from metabolism than was available when using redox balanced fermentations. The down-side was that molecular oxygen proved to be toxic, generating the potentially dangerous hydroxyl ion, OH•, a short lived but highly reactive free radical, which causes enormous damage to all biological molecules. As we develop in Chapter 11, this has enormous consequences for age related disorders, particularly neurodegenerative disorders. As we have argued (Crichton and Ward, 2006; Crichton et al., 2008), dysregulation of metal ion homeostasis, particularly of redox active metals like iron and copper in specific brain regions, leads to the generation of ROS, which can either directly damage proteins, or lead to the formation of highly reac- tive aldehydes. These, in turn, generate protein carbonyls, resulting in protein denaturation, aggregation and a subsequent failure of the ubiquitin/proteasome system to eliminate these defective proteins. The end result, as in Alzheimer’s, Parkinson’s, Huntington’s disease and countless others, is neurodegeneration accompanied by the appearance of neuropathological lesions characterised by intranuclear, cytoplasmic or extracellular protein aggregates (Crichton and Ward, 2006; Crichton et al., 2008).
Finally, we can cite the powerful influence on human civilization of the materials out of which weapons could be manufactured, reflected in the so-called three ages of Man (Stone, Bronze and Iron). While the Stone Age, which began around two million years ago, left many vestiges, it used stone, in the apparent absence of metals, as the sole means of making martial instruments. It was superseded by the Bronze Age, which was marked by important inventions, like the wheel and the ox-drawn plough, and was characterised by the use of metals, notably copper. True bronze, an alloy of copper and tin, in Europe, was heavily dependent on the tin deposits of Cornwall. But by around 1200 BC, the ability to heat and to forge iron saw the end of the Bronze Age, and ushered in the Iron Age, of which few artefacts remain on account of the poor stability of iron when confronted by oxygen and water (rust is no way to preserve archaeological artefacts!). As we discuss in Chapter 12, iron does interact with copper, as well as with a number of other metals. While little remains from the Iron Age, compared to the Stone, to the Bronze, or even to the Gold and Silver, iron always has the last word.
This policy cannot succeed through speeches and shooting matches and songs; it can only be carried out through blood and iron
Preface xix
Otto von Bismarck, speech to the Prussian House of Deputies, 28 January 1886 These few examples give only the superficial mariner’s view of the iceberg that remains
unseen concerning the fundamental importance of iron in biological systems. There remains a great deal more that we have not seen in this titillation of the reader’s palate, and we shall surely make amends in what follows. For the vast majority of living organisms iron is absolutely necessary for their maintenance, defence, differentiation and last, but by no means least, for their growth and cellular division.
That is why I have devoted this book to the inorganic biochemistry of iron. I accept that any errors or omissions are my responsibility, and I would be most grateful to my readers to let me know what needs to be amended in the next edition.
Robert Crichton
August 2008
References
Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana , Nature, 408, 796–815.
Bekri, S., Kispal, G., Lange, H. et al. (2000) Human ABC7 transporter: gene structure and mutation causing X-linked sideroblastic anemia with ataxia with disruption of cytosolic iron-sulfur protein maturation, Blood , 96, 3256–64.
Bishop, J.K.B., Wood, T.J., Davis, R.E. and Sherman, J.T. (2004) Robotic observations of enhanced carbon biomass and export at 55 degrees during SOFeX, Science, 304, 417–20.
Boyd, P.W., Jickells, T., Law, C.S. et al. (2007) Mesoscale iron enrichment experiments 1993–2005: synthesis and future directions, Science, 315, 612–7.
Braun, V., Crichton, R.R. and Braunitzer, G. (1968) Ueber monomere und dimere Insekten-hamoglobine (Chironomus thummi), Hoppe Zeyler’s Z. Physiol. Chem., 18, 369–74.
Buesseler, K.O., Doney, S.C., Karl, D.M. et al., (2008) Ocean Iron Fertilization – Moving Forward in a Sea of Uncertainty, Science, 319, 162.
Butler, A. (2005) Marine siderophores and Microbial Iron Mobilization, Biometals , 281, 207–10. Coale, K.H., Johnson, K.S., Chavez, F.P. et al. (2004) Southern Ocean iron enrichment experiment: carbon
cycling in high- and low-Si waters, Science, 304, 408–14. Crichton, R.R. (2001) Inorganic Biochemistry of Iron Metabolism From Molecular Mechanisms to Clinical
Consequences , John Wiley & Sons Ltd, Chichester, pp. 326. Crichton, R.R. (2008) Biological Inorganic Chemistry An Introduction , Elsevier, Amsterdam and Oxford,
pp. 369. Crichton, R.R. and Ward, R.J. (2006) Metal-based Neurodegeneration From Molecular Mechanisms to
Therapeutic Perspectives , John Wiley & Sons Ltd, Chichester, pp. 227. Crichton R.R., Dexter D.T. and Ward R.J. (2008) Metal based neurodegenerative diseases – From molecular
mechanisms to therapeutic strategies, Coord. Chem. Rev., 252, 1189–99. de Domenico, I., McVey Ward, D., Langelier, C. et al. (2007) The molecular mechanism of
hepcidin-mediated ferroportin down-regulation, Mol. Biol. Cell., 18, 2569–78. Dupuy, J., Volbeda, A., Carpentier, P. et al. (2006) Crystal structure of human iron regulatory protein 1 as
cytosolic aconitase, Structure, 14, 129–39. Frausto da Silva, J.J.R. and Williams, R.J.P. (2001) The biological chemistry of the elements 2nd edn,
Oxford University Press, Oxford, pp. 575. Hacker, J. and Kaper, J.B. (2000) Pathogenicity islands and the evolution of microbes, Ann. Rev. Microbiol.,
54, 641–79. International Rice Genome Sequencing Project (2005) The map-based sequence of the rice genome, Nature,
436, 793–800.
xx Preface
Leibold, E.A. and Munro, H.N. (1988) Cytoplasmic protein binds in vitro to a highly conserved sequence in the 5′ untranslated region of ferritin heavy- and light-subunit mRNAs, Proc. Nat. Acad. Sci. USA, 85, 2171–5.
Lill, R. and Muhlenhoff, U. (2006) Iron-Sulfur Protein Biogenesis in Eukaryotes: Components and Mech- anisms, Ann. Rev. Cell Dev. Biol., 22, 457–86.
Martin, J.H. (1990), Glacial–Interglacial CO2 Change: The Iron Hypothesis, Paleoceanography , 5, 1–13. Martin, J.H. and Fitzwater, S.E (1988) Iron deficiency limits phytoplankton growth in the north-east Pacific
subarctic, Nature, 331, 341–3. Martinez, J.S., Zhang, G.P., Holt, P.D. et al. (2000) Self assembling amphiphilic siderophores from marine
bacteria, Science, 287, 1245–7. Martinez, J.S., Carter-Franklin, J.N., Mann, E.L., (2003) Structure and membrane affinity of a suite of
amphiphilic siderophores produced by a marine bacterium, Proc. Nat. Acad. Sci. USA, 100, 3754–9. Mason, B. and Moore, C.B. (1982) Principles of Geochemistry , 4th edn, John Wiley & Sons, Inc., New
York. McCance, R.A. and Widdowson, E.M. (1937) Absorption and excretion of iron, Lancet ii, 680–4. McKie, A.T., Marciano, P., Rolfs, A. et al. (2000) A novel duodenal iron-regulated transporter, IREG1,
implicated in the basolateral transfer of iron to the circulation, Mol. Cell , 5, 299–309. Nemeth, E., Tuttle, M.S., Powelson, J. et al. (2004) Hepcidin regulates cellular iron efflux by binding to
ferroportin and inducing its internalization, Science, 306, 2090–3. Nicolas, G., Bennoun, M., Devaux, I. et al. (2001) Lack of hepcidin gene expression and severe tissue
iron overload in upstream stimulatory factor 2 (USF2) knockout mice, Proc. Nat. Acad. Sci. USA, 98, 8780–5.
Pietrangelo, A. (2007) Hemochromatosis: An Endocrine Liver Disease, Hepatol., 46, 1291–301. Pigeon, C., Ilyin, G., Courselaud, B. et al. (2001) A new mouse liver-specific gene, encoding a protein
homologous to human antimicrobial peptide hepcidin, is overexpressed during iron overload, J. Biol. Chem., 276, 7811–9.
Owen, T., Pynn, R. Martinez, J.S. and Butler, A. (2005) Micelle-to-vesicle transition of an iron-chelating microbial surfactant, marinobactin E, Langmuir , 21, 12109–14.
Ratledge, C. (2007) Iron metabolism and infection, Food Nutr. Bull., 28, S515–23. Rouault, T.A., Hentze, M.W., Dancis, A. et al. (1988) Binding of a cytosolic protein to the iron-responsive
element of human ferritin messenger RNA, Science, 241, 1207–10. Vujic Spasic, M., Kiss, J., Herrmann, T. et al. (2007) Physiologic systemic iron metabolism in mice deficient
for duodenal Hfe, Blood , 109, 4511–7. Vujic Spasic, M., Kiss, J., Herrmann, T. et al. (2008) Hfe acts in hepatocytes to prevent hemochromatosis,
Cell Metab., 7, 173–8. Walden W.E., Selezneva, A.I., Dupuy, J. et al. (2006) Structure of dual function iron regulatory protein 1
complexed with ferritin IRE-RNA, Science, 314, 1903–8. Williams, P.H. (1979) Novel iron uptake system specified by ColV plasmids: an important component in
the virulence of invasive strains of Escherichia coli, Infect. Immunol., 26, 925–32. Zhang, Y., Lyver, E.R., Knight, S.A. et al. (2005) Frataxin and mitochondrial carrier proteins Mrs3p and
Mrs4p, cooperate in providing iron for heme synthesis, J. Biol. Chem., 280, 19794–807. Zhang, Y., Lyver, E.R., Knight, S.A. et al. (2006) Mrs3p, Mrs4p, and frataxin provide iron for Fe-S cluster
synthesis in mitochondria, J. Biol. Chem., 281, 22493–502.
1 Solution Chemistry of Iron
in Biological Media
1.1 Aqueous Solution Chemistry of Iron
Iron, element 26 in the periodic table, is the second most abundant metal (after aluminium) and the fourth most abundant element of the earth’s crust. Its position in the middle of the elements of the first transition series (so designated because their ions have incompletely filled d orbitals) implies that iron has the possibility of various oxidation states (from −II to +VI), the princi- pal being II (d6) and III (d5), although a number of iron-dependent monooxygenases generate high valent Fe(IV) or Fe(V) reactive intermediates during their catalytic cycle. Whereas Fe2+ is extremely water soluble, Fe3+ is quite insoluble in water (Ksp = 10−39 M and at pH 7.0, [Fe3+] = 10−18 M) and significant concentrations of water-soluble Fe3+ species can be attained only by strong complex formation. Iron (III) is a hard acid which prefers hard oxygen ligands while iron (II) is on the borderline between hard and soft, favouring nitro- gen and sulfur ligands. The interaction between Fe2+ and Fe3+ and ligand donor atoms will depend on the strength of the chemical bond formed between them.
An idea of the strength of such bonds can be got from the concept of ‘hard’ and ‘soft’ acids and bases (HSAB). ‘Soft’ bases have donor atoms of high polarisability with empty, low energy orbitals; they usually have low electronegativity and are easily oxidised. In contrast ‘hard’ bases have donor atoms of low polarisability, and only have vacant orbitals of high energy; they have high electronegativity and are hard to oxidise. Metal ions are ‘soft’ acids if they are of low charge density, have a large ionic radius and have easily excited outer electrons. ‘Hard’ acid metal ions have high charge density, a small ionic radius and no easily excited outer electrons. In general ‘hard’ acids prefer ‘hard’ bases and ‘soft’ acids form more stable complexes with ‘soft’ bases (Pearson, 1963). Fe(III), with an ionic radius of 0.067 nm and a charge of 3+, is a ‘hard’ acid and will prefer ‘hard’ oxygen ligands like phenolate and
Iron Metabolism – From Molecular Mechanisms to Clinical Consequences, Third Edition Robert Crichton 2009 John Wiley & Sons, Ltd
2 Iron Metabolism
carboxylate to imidazole or thiolate. Fe(II), with an ionic radius of 0.083 nm and a charge of only 2+, is on the borderline between ‘hard’ and ‘soft’, favouring nitrogen (imidazole and pyrrole) and sulfur ligands (thiolate and methionine) over oxygen ligands.
The coordination number of six is the most frequently found for both Fe(II) and Fe(III), giving octahedral stereochemistry, although four (tetrahedral) and particularly five coordinate complexes (trigonal bipyramidal or square pyrimidal) are also found. For octahedral com- plexes, two different spin states can be observed. Strong field ligands (e.g. F−, OH−), where the crystal field splitting is high and hence electrons are paired, give low spin complexes, while weak field ligands (e.g. CO, CN−), where crystal field splitting is low, favour a maxi- mum number of unpaired electrons and give high spin complexes Changes of spin state affect ion size of both Fe(II) and Fe(III), the high spin ion being significantly larger than the low spin ion. As will be seen in Chapter 2, this is put to good use as a trigger for cooperative binding of dioxygen (O2) to haemoglobin. High spin complexes are kinetically labile, while low spin complexes are exchange inert. For both oxidation states only high spin tetrahedral complexes are formed. Both oxidation states are Lewis acids, particularly the ferric Fe(III) state.
The unique suitability of iron comes from the extreme variability of the Fe2+/Fe3+ redox potential, which can be fine tuned by well chosen ligands, so that iron sites can encompass almost the entire biologically significant range of redox potentials, from about −0.5 V to about +0.6 V.
1.1.1 Oxygen Free Radicals
Molecular oxygen was not present when life began on earth, with its essentially reducing atmosphere, and both the natural abundance of iron and its redox properties predisposed it to play a crucial role in the first stages of life on earth. About one billion (109) years ago, photosynthetic prokaryotes (Cyanobacteria) appeared and dioxygen was evolved into the earth’s atmosphere. It probably required 200–300 million years, a relatively short time on a geological time scale, for oxygen to attain a significant concentration in the atmosphere, since at the outset the oxygen produced by photosynthesis would have been consumed by the oxidation of ferrous Fe(II) ions in the oceans. Once dioxygen had become a dominant chemical entity, iron hydroxides precipitated, as the Precambrian deposits of red ferric oxides laid down in the geological strata at that time bear witness. Concomitant with the loss of iron bioavailability, the oxidation of Cu(I) led to soluble Cu(II). While enzymes active in anaerobic metabolism were designed to be active in the lower portion of the redox potential spectrum, the presence of dioxygen created the need for a new redox active metal with EoMn+1/Mn
from 0 to 0.8 V. Copper, now bioavailable (Crichton and Pierre, 2001), was ideally suited for this role, and began to be used in enzymes with higher redox potentials (as a di-copper centre in laccase and a mixed iron–copper centre in cytochrome oxidase) to take advantage of the oxidising power of dioxygen. Some typical redox potentials for iron and copper proteins and chelates are given in Figure 1.1.
Although oxygen must ultimately completely oxidise all biological matter, its propensity for biological oxidation is considerably slowed by the fact that in its ground state (lowest energy state) it exists as a triplet spin state (Figure 1.2), whereas most biological molecules are in the singlet state as their lowest energy level. Spin inversion is relatively slow, so that oxygen reacts much more easily with other triplet state molecules or with free radicals than with singlet state molecules.
Solution Chemistry of Iron in Biological Media 3
Phen3Fe + 1.1
+ 0.37 (plastocyanin) + 0.33 (azurin) + 0.32 (Cu-ZnSOD)
+ 0.16 (Cu aquo)
Ferrous or cuprous state does not give Fenton reaction
The ferrous or the cuprous state give Fenton reaction
The ferric or the cupric form can be reduced by superoxide
(iron SOD) + 0.27
The ferric or the cupric form is not reduced by superoxide
(haemoglobin) + 0.14
(ferritransferrin) − 0.52
(ferrienterobactin) − 0.75
O2/O2 − − 0.16
Figure 1.1 Some redox potentials of iron and copper enzymes and chelates at pH 7 in Volts relative to the standard hydrogen electrode. (Copyright 2001, John Wiley & Sons Ltd.)
Orbital
Peroxide ion
(O2 2−)
Singlet O2
Figure 1.2 Bonding in the diatomic oxygen molecule. (Copyright 2001, John Wiley & Sons Ltd.)
4 Iron Metabolism
The arrangement of electrons in most atoms and molecules is such that they occur in pairs, each of which have opposite intrinsic spin angular momentum. Molecules which have one or more unpaired electrons are termed free radicals: they are generally very reactive, and will act as chain carriers in chemical reactions. Thus, the hydrogen atom, with one unpaired electron, is a free radical, as are most transition metals and the oxygen molecule itself. The dioxygen molecule has two unpaired electrons, each located in a different π∗ anti-bonding orbital. Since these two electrons have the same spin quantum number, if the oxygen molecule attempts to oxidise another atom or molecule by accepting a pair of electrons from it, both new electrons must have parallel spins in order to fit into the vacant spaces in the π∗ orbitals. A pair of electrons in an atomic or molecular orbital would have anti-parallel spins (of +1/2 and −1/2) in accordance with Pauli’s principle. This imposes a restriction on oxidation by dioxygen, which means that dioxygen tends to accept its electrons one at a time (Figure 1.2), and slows its reaction with nonradical species (Halliwell and Gutteridge, 1984). Transition metals can overcome this spin restriction on account of their ability to accept and donate single electrons. The interaction of iron centres and oxygen is of paramount importance in biological inorganic chemistry; some of the main features have been summarised in Figure 1.3.
The reactivity of dioxygen can be increased in another way, by moving one of the unpaired electrons in a way that alleviates the spin restriction to give the two singlet states of dioxygen (Figure 1.2). The most important of the two forms of singlet O1
2δg in biological systems has no unpaired electrons, is not a radical, and can be obtained when a number of biological pigments such as chlorophylls, retinal, flavins or porphyrins are illuminated in the presence of dioxygen. When a single electron is accepted by the ground-state dioxygen molecule, it must enter one of the π∗ anti-bonding orbitals, to form the superoxide radical, O2
−. Addition of a second electron to O2
− gives the peroxide ion O2 2− with no unpaired electrons. At
physiological pH O2 2− will immediately protonate to give hydrogen peroxide, H2O2. The
third reactive oxygen species found in biological systems is the hydroxyl free radical. Two hydroxyl radicals, •OH can be formed by homolytic fission of the O−O bond in hydrogen peroxide either by heating or by irradiation. However, a simple mixture of hydrogen peroxide and an Fe(II) salt also produces the •OH radical (Reaction 1.1) in the reaction first attributed
FeII
FeIII
FeIII
η2 - peroxo (or side-on)
Figure 1.3 Iron–oxygen chemistry. Multi-bridged species have been omitted. (Copyright 2001, John Wiley & Sons Ltd.)
Solution Chemistry of Iron in Biological Media 5
to Fenton (Fenton, 1894). In fact, what Fenton observed was the oxidation of tartaric acid; we now know that the reactive oxygen species involved is the hydroxyl rediacal.
Fe2+ + H2O2 → Fe3+ + •OH + OH− (1.1)
In the presence of trace amounts of iron, the superoxide radical can then reduce Fe3+ to molecular oxygen and Fe2+. The sum of this reaction (1.2) plus the Fenton reaction (1.1) pro- duces molecular oxygen, hydroxyl radical and hydroxyl anion from the superoxide radical and hydrogen peroxide, in the presence of catalytic amounts of iron – the so-called Haber–Weiss1
reaction (1.3) (Haber and Weiss, 1934).
Fe3+ + O2 − → Fe2+ + O2 (1.2)
O2 − + H2O2 → O2 + •OH + OH− (1.3)
Iron or copper complexes will catalyse Fenton chemistry only if two conditions are met simultaneously, namely that the ferric complex can be reduced and that the ferrous complex has an oxidation potential such that it can transfer an electron to hydrogen peroxide. How- ever, it must also be added that this reasoning supposes that it is under standard conditions, and at equilibrium, which is rarely the case for biological systems. A simple example will illustrate the problem – whereas under standard conditions reaction (1.2) has a redox potential of −330 mV (at an O2 concentration of one atmosphere), in vivo with [O2] = 3.5 × 10−5 M and [O2
−] = 10−11 M the redox potential is +230 mV (Pierre and Fontecave, 1999). In aqueous solution in the absence of oxygen, iron is present as the hydrated hexa-aqua
ferrous ion, Fe(H2O)6 2+. In the early stages of evolution the atmosphere was thought to be
essentially reducing with a very low oxygen pressure, and thus a high concentration of reduced iron would have been present. The appearance of molecular oxygen, which accompanied the arrival of photosynthetic organisms capable of fixing atmospheric carbon dioxide with concomitant water splitting to yield electrons, protons and oxygen, changed dramatically the situation, since the following reaction (1.4) (here simplified by neglecting the hydration of the ferrous ion) would result:
Fe(II)aq + O2 → Fe(III)aq + O2 − (1.4)
Except at very low pH values, the hexa-aqua ferric ion, Fe(H2O)6 3+, would then undergo
a series of hydrolysis and polymerisation reactions leading progressively to more and more insoluble ferric polynuclear species, which would precipitate to give the geologic evidence of the oxygenation of the atmosphere by the presence around the mid Precambrian of intense red deposits of ferric oxides. The inorganic chemistry involved in these processes is becoming better understood (Jolivet et al., 2004); the rest of this chapter concerns the pathways of iron hydrolysis and polymerisation, and concludes with some thoughts on biomineralisation.
1.1.2 Iron Hydrolysis – a Ubiquitous Phenomenon
Metal salts, when they are dissolved in water, undergo hydrolysis – iron forms hexacoordinate aquo complexes, [Fe(H2O)6]z+, in which polarisation of the coordinated water molecules depends on the oxidation state and the size of the cation. Ferric aquo complexes are more
1The reaction was originally described by Haber and Wilstatter (1931), but the original paper was published in German! The more frequently cited Haber and Weiss paper does cite the original, but in neither is a reference to Fenton given.
6 Iron Metabolism
pH
Figure 1.4 Speciation of [Fe(OH)h(H2O)6–h](z–h)+ complexes of: (a) Fe(II); (b) Fe(III). (From Jolivet, J.-P., Chaneac, C. and Trone E. (2004) Iron oxide chemistry. From molecular clusters to extended solid networks, Chem. Commun., 481–487. Reproduced by permission of The Royal Society of Chemistry.)
acidic that ferrous, and hydroxylation of the cations occurs in very distinct ranges of pH, as can be seen from the speciation diagram (Figure 1.4). Hydrolysis originates from the loss of protons from the aqua metal ion – going from [Fe(OH)hFe(H2O)6−h](z−h)+, where h = 0, with progressively increasing values of h, each step accompanied by release of H+. Between pH 5 and pH 9, which is clearly of relevance to living organisms as well as aquatic systems, ferric salts hydrolyse immediately, whereas ferrous salts, in the absence of oxygen or other oxidising agents, give solutions of ferrous aqua ions, Fe(H2O)6
2+. Thus, in biological media, the hydrated ferrous ion is a real species, as can be seen from the speciation diagram (Figure 1.4), whereas the hydrated ferric ion is relatively rare (Jolivet et al., 2004), although significant concentrations of Fe(H2O)6
3+ are present at very low pH values. In most lakes, estuaries, streams and rivers, iron levels are high, and Fe2+ is produced by photolysis of inner-sphere complexes of particulate and colloidal iron (III) hydroxides with biogenic organic ligands. Since the photic zones in which this takes place are aerobic, there is continuous reoxidation of iron, producing secondary colloidal iron (III) hydroxides. In deeper waters, settling organic matter can supply reducing equivalents to convert FeO.OH to Fe2+. In contrast, iron levels in surface seawater are extremely low, 0.02–1 nM (Wu and Luther, 1996).
Hydroxylated complexes can condense by two different mechanisms, depending on the nature of the coordination sphere of the cation (Jolivet, 2000). Aquohydroxo complexes can condense by elimination of water and formation of hydroxo bridges (olation), whereas oxo- hydroxy complexes, where there is no water molecule, condense in a two step mechanism leading to the formation of oxo bridges (oxolation):
H2O M OH d−d+
+ M OH2
d−d+
M OH d−
M O M + H2O
For ferric complexes, condensation occurs from strongly acidic media (pH ∼ 1), whereas ferrous complexes condense only above pH 6, and the formation of polycationic ferrous
Solution Chemistry of Iron in Biological Media 7
(b)(a)
Figure 1.5 Examples of polycationic structures formed by ferric ions in the presence of strongly complexing ligands: (a) [Fe19O6(OH)14(L)10(H2O)12]+ (L = N(CH2COOH)2(CH2CH2OH)); (b) Fe8(PhCOO)12(thme)4.2Et2O (thme: trishydroxymethylethane). (From Jolivet, J.-P., Chaneac, C. and Trone E. (2004) Iron oxide chemistry. From molecular clusters to extended solid networks, Chem. Commun., 481–487. Reproduced by permission of The Royal Society of Chemistry.)
species is poorly documented. On account of their high reactivity, ferric complexes condense very rapidly and the process is difficult to stop without the use of very strongly complexing polydentate ligands. However, a range of species containing polynuclear Fe(III) cores have been characterised using a number of polycarboxylate or amino ligands (Lippard, 1988; Taft and Lippard, 1990; Taft et al., 1993; Schmitt et al., 2001; Jones et al., 2002); two of them are illustrated in Figure 1.5.
1.1.3 Hydrolysis of Iron(III) in Acid Media – Formation of Polynuclear Species
Hydrolysis of ferric solutions is readily induced by the addition of a base. Upon addition of a base at rather acid pH, the purple ferric aqua-ion Fe(H2O)6
3+ initially undergoes a first deprotonation step (Reaction 1.5), which is followed by reversible dimerisation (Reaction 1.6), giving a yellow solution of mono and dinuclear species:
2Fe3+ + 2H2O → 2FeOH2+ + 2H+ (1.5)
FeOH2+ + FeOH2+ → Fe2(OH)2 4+ (1.6)
The equilibria leading to mono and dinuclear hydrolysis products such as FeOH2+, Fe(OH)2
+ and Fe2(OH)2 4+ are established rapidly and are well understood (Cornell et al.,
1989). The low molecular species interact to produce species with a higher nuclearity (Reaction 1.7):
Fe2(OH)2 4+ + FeOH2+ + H2O → Fe3(OH)4
5+ + H+ (1.7)
Addition of a base to solutions of ferric ions at pH values >3 immediately leads to precipitation of a poorly ordered, amorphous, red–brown ferric hydroxide precipitate. This synthetic precipitate resembles the mineral ferrihydrite, and also shows some similarity to the iron oxyhydroxide core of ferritin (Chapter 6). Ferrihydrite can be considered as the
8 Iron Metabolism
least stable but most reactive form of iron(III), the group name for amorphous phases with large specific surface areas (>340 m2/g). The transformation of ferrihydrite into other more crystalline products such as goethite and haematite is discussed shortly, after beginning with some remarks concerning the biological distribution and structure of ferrihydrite (Jambor and Dutrizac, 1998).
Although ferrihydrite is of great importance in metallurgical processing and in the natu- ral environment, its presence is often underestimated because of difficulties in its definitive identification and also because of its common designation (covering a range of poorly ordered compounds), as amorphous iron hydroxide, colloidal ferric hydroxide, Fe(OH)3 and so on. Ferrihydrite has been identified as a preterrestrial component of meteorites and may be a constituent of the soils of Mars. On Earth, ferrihydrite is ubiquitous in natural waters, in the sediments derived from these waters and is a constituent of a wide variety of soils, particularly those formed under cool and moist conditions as the precursor of haematite. It is abundantly present in the precipitates resulting from acid mine drainage. Ferrihydrite is routinely used in industrial applications such as coal liquefaction and metallurgical processing, and because of its extremely high surface area and reactivity it is manufactured for use as a very effective heavy metal scavenger in wastewater treatments.
As pointed out above, rapid hydrolysis of Fe(III) solutions (for example neutralisation of ferric solutions with an excess of alkali) gives a red–brown precipitate of ferrihydrite. The conventional classification of ferrihydrite is based on the number of X-ray diffraction (XRD) peaks. Normally, a distinction is drawn between two types of ferrihydrite, referred to as “2-line ferrihydrite”, which describes material that exhibits little crystallinity and ‘6-line ferrihydrite’, which has the best crystallinity. In a typical XRD pattern of these materials, the 2-line form displays two broad peaks at 1.5 and 2.5 A, while the more crystalline 6-line form displays six peaks at 1.5 (a doublet), 1.7, 2.0, 2.2 and 2.5 A (Jambor and Dutrizac, 1998). The degree of order found in ferrihydrite depends on the method of preparation and the time of its ageing. Brief heating of Fe(III) solutions to about 80 C typically produces “6-line ferrihy- drite”, whereas the 2-line variety is typically produced at ambient temperatures by addition of alkali to raise the pH to about 7. It seems to be agreed that ferrihydrite is not amorphous and has at least some degree of crystallinity. Despite the ease of its synthesis in the laboratory, no single formula is widely accepted, and compositions ranging from Fe5HO8.4H2O (Towe and Bradley, 1967), through 5Fe2O3.9H2O (Towe, 1981) to the recent Fe10O14(OH)2 (Michel et al., 2007a) have been proposed. It has been demonstrated that almost all of the water can be replaced by adsorbed species in quantities that cannot be accommodated within the crystal structure, and it was proposed that the bulk structural unit for ferrihydrite is an Fe(O,OH)6
octahedron, where the surface structure is a mixture of octahedrally and tetrahedrally coor- dinated iron (Jambor and Dutrizac, 1998). These ‘coordination-unsaturated’ surface sites are readily accessible to the adsorption of foreign species and, together with the large surface area referred to above, most likely account for the high adsorptive capacity of ferrihydrite.