Metal-Carbon Bonds in Enzymes and Cofactors 9783110436587, 9783110442793

Citation preview

METAL IONS IN LIFE SCIENCES

VOLUME 6

Metal-Carbon Bonds in Enzymes and Cofactors

METAL IONS IN LIFE SCIENCES edited by Astrid Sigel,(1) Helmut Sigel,(1) and Roland K. O. Sigel(2) ^ Department of Chemistry Inorganic Chemistry University of Basel Spitalstrasse 51 CH-4056 Basel, Switzerland ® Institute of Inorganic Chemistry University of Zürich Winterthurerstrasse 190 CH-8057 Zürich, Switzerland

VOLUME 6

Metal-Carbon Bonds in Enzymes and Cofactors

DE GRUYTER

First published by the Royal Society of Chemistry in 2009. Publication Details: ISBN: 978-1-84755-915-9 ISSN: 1559-0836 DOI: 10.1039/9781847559159 A cataloque record for this book is available from the British Library

ISBN 978-3-11-044279-3 e-ISBN (PDF) 978-3-11-043658-7 Set-ISBN (Print + Ebook) 978-3-11-043659-4 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. ©2015 Walter de Gruyter GmbH, Berlin/Munich/Boston Cover image: The figure on the dust cover shows coenzyme B12 (= 5'-deoxy5'-adenosylcobalamin) containing a cobalt-carbon bond; prepared by Roland K. O. Sigel using the CSD coordinations DADCBL. www.degruyter.com

Historical Development and Perspectives of the Series Metal Ions in Life Sciences*

It is an old wisdom that metals are indispensable for life. Indeed, several of them, like sodium, potassium, and calcium, are easily discovered in living matter. However, the role of metals and their impact on life remained largely hidden until inorganic chemistry and coordination chemistry experienced a pronounced revival in the 1950s. The experimental and theoretical tools created in this period and their application to biochemical problems led to the development of the field or discipline now known as Bioinorganic Chemistry, Inorganic Biochemistry, or more recently also often addressed as Biological Inorganic Chemistry. By 1970 Bioinorganic Chemistry was established and further promoted by the book series Metal Ions in Biological Systems founded in 1973 (edited by H.S., who was soon joined by A.S.) and published by Marcel Dekker, Inc., New York, for more than 30 years. After this company ceased to be a family endeavor and its acquisition by another company, we decided, after having edited 44 volumes of the MIBS series (the last two together with R.K.O.S.) to launch a new and broader minded series to cover today's needs in the Life Sciences. Therefore, the Sigels new series is entitled Metal Ions in Life Sciences. After publication of the first four volumes (2006-2008) with John Wiley & Sons, Ltd., Chichester, U K , we are happy to join forces now in this still new endeavor with the Royal Society of Chemistry, Cambridge, UK; a most experienced Publisher in the Sciences.

Reproduced with some alterations by permission of John Wiley & Sons, Ltd., Chichester, UK (copyright 2006) from pages ν and vi of Volume 1 of the series Metal Ions in Life Sciences (MILS-1).

vi

PERSPECTIVES OF THE SERIES

The development of Biological Inorganic Chemistry during the past 40 years was and still is driven by several factors; among these are (i) the attempts to reveal the interplay between metal ions and peptides, nucleotides, hormones or vitamins, etc., (ii) the efforts regarding the understanding of accumulation, transport, metabolism and toxicity of metal ions, (iii) the development and application of metal-based drugs, (iv) biomimetic syntheses with the aim to understand biological processes as well as to create efficient catalysts, (v) the determination of high-resolution structures of proteins, nucleic acids, and other biomolecules, (vi) the utilization of powerful spectroscopic tools allowing studies of structures and dynamics, and (vii), more recently, the widespread use of macromolecular engineering to create new biologically relevant structures at will. All this and more is and will be reflected in the volumes of the series Metal Ions in Life Sciences. The importance of metal ions to the vital functions of living organisms, hence, to their health and well-being, is nowadays well accepted. However, in spite of all the progress made, we are still only at the brink of understanding these processes. Therefore, the series Metal Ions in Life Sciences will endeavor to link coordination chemistry and biochemistry in their widest sense. Despite the evident expectation that a great deal of future outstanding discoveries will be made in the interdisciplinary areas of science, there are still "language" barriers between the historically separate spheres of chemistry, biology, medicine, and physics. Thus, it is one of the aims of this series to catalyze mutual "understanding". It is our hope that Metal Ions in Life Sciences proves a stimulus for new activities in the fascinating "field" of Biological Inorganic Chemistry. If so, it will well serve its purpose and be a rewarding result for the efforts spent by the authors. Astrid Sigel, Helmut Sigel Department of Chemistry Inorganic Chemistry University of Basel CH-4056 Basel Switzerland

Roland K. O. Sigel Institute of Inorganic Chemistry University of Zürich CH-8057 Zürich Switzerland October 2005 and October 2008

Preface to Volume 6 Metal-Carbon Bonds in Enzymes and Cofactovs

This is the 6th volume within the MILS series; together with the 44 volumes published in our former series Metal Ions in Biological Systems this sums up to in total 50 books. This event is celebrated with a comprehensive Author Index, given at the end of this book. It encompasses the names of all colleagues who contributed to these 44 MIBS and 6 MILS volumes. All these authors deserve our special thanks; without their excellent contributions the two series could not have been successful. The present Volume 6 is devoted to naturally occurring metal-carbon bonds, a topic recently obtaining (again) significant momentum, largely - but not only - due to new insights gained with hydrogenases. The field started out about 50 years ago when coenzyme B 12 was identified as organometallic derivative of vitamin B 12 . This moved the cobalt-carbon bond into the center of interest and consequently, the first two chapters of this book are devoted to the organometallic chemistry of B 12 coenzymes and to the biochemistry of cobalamin- and corrinoid-dependent enzymes. B 12 coenzymes are required in the metabolism of a broad range of organisms including humans; however, only microorganisms have the ability to biosynthesize BI 2 and other natural corrinoids. This fact alone, together with new metabolic insights (e.g., riboswitches), guarantees a continued fascination - not only for the B 12 community. Related to Co-corrin, the Ni-porphinoid unit (F 430 ) is the prosthetic group of methyl-coenzyme Μ reductase. This enzyme, the topic of Chapter 3, catalyzes the methane-forming step in methanogenic archaea and most

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-FP007

viii

PREFACE TO VOLUME 6

probably also the methane-oxidizing step in methanotrophic archaea. Chapter 4 deals with acetyl-coenzyme A synthases/carbon monoxide dehydrogenases, i.e., bifunctional nickel-containing enzymes, which catalyze the synthesis of acetyl-CoA and the reversible reduction of C 0 2 to CO in anaerobic, mostly thermophilic, organisms, able to grow chemiautotrophically on simple inorganic compounds like C 0 2 . Ni-C bonds with methyl, acetyl, carbonyl, and carboxylate groups are evidenced. [NiFe]-, [FeFe]-, and [Fe]-hydrogenases are detailed in the next three chapters. These enzymes, present in many microorganisms, catalyze the oxidation of molecular hydrogen or the reduction of protons. All of them have a Fe(CO) x unit in their active site. Iron-cyanide units occur in [NiFe]and [FeFe]-hydrogenases. However, despite the indicated similarities they clearly have independent evolutionary origins. The participation of the commonly considered toxic ligands CO and CN~ in the active sites of hydrogenases is still a surprise to many; yet, exactly their occurrence incites a great interest in physical chemists as well as evolutionary biologists. The dual role of heme as cofactor and substrate in the biosynthesis of carbon monoxide is the topic of Chapter 8. Carbon monoxide is a ubiquitous molecule in the atmosphere but it is also produced in mammalian, plastidic, and bacterial cells as a byproduct in the catalytic cycle of heme degradation as catalyzed by the enzyme heme oxygenase. Most fascinating is the fact that the biological role of CO spans the range from toxic to cytoprotective, depending on its concentration. CO generated by heme oxygenase is now known to function in several important physiological processes, including vasodilation, apoptosis, inflammation, and possibly neurotransmission. The relevance of the copper-carbon bond in biological inorganic chemistry will probably not easily come to the mind of most biochemical and inorganic researchers. However, there is a vast amount of literature, cunningly presented in Chapter 9. CO as well as CN~ have proven very useful in obtaining insights into the active site structures and mechanisms of copper proteins. Naturally, in these instances both ligands are inhibitors and used as probes. However, there is also the recently described copper-carbon unit present in a carbon monoxide dehydrogenase, which contains a novel molybdenum-copper catalytic site, or the copper(I)-arene unit, which was evidenced in a bacterial copper chaperone. Apparently also a plant receptor site (ETR1) utilizes Cu(I) to sense the growth hormone ethylene. Chapter 10 focuses on the interaction of CN~ with enzymes containing vanadium, manganese, non-heme iron, and zinc, and the inhibiting properties of this ligand, allowing its use as a probe. The reaction mechanism of the molybdenum hydroxylase xanthine oxidoreductase is revisited in Chapter 11; previously a molybdenum-carbon bond was postulated but now proof is presented against its formation. The terminating Chapter 12 reviews

PREFACE TO VOLUME 6

ix

briefly the most popular computational approaches employed in theoretical studies of bioorganometallic species by providing detailed examples. Taken together, MILS-6 summarizes our knowledge on Metal-Carbon Bonds in Enzymes and Cofactors', i.e., it emphasizes the role of metal-carbon bonds for life as well as research. However, there are many metal-carbon bonds which occur in the environment in compounds like alkyl-arsenicals or -mercurials and in lead- or tinorganyls, etc., most of them known as being toxic. Consequently, the next volume ( M I L S - 1 ) will be devoted to Organometallics in Environment and Toxicology. Astrid Sigel Helmut Sigel Roland K. O. Sigel

Contents

HISTORICAL DEVELOPMENT A N D PERSPECTIVES OF THE SERIES

ν

PREFACE TO VOLUME 6

vii

CONTRIBUTORS TO VOLUME 6

xvii

TITLES OF VOLUMES 1 ^ 4 IN THE METAL IONS IN BIOLOGICAL SYSTEMS

SERIES

CONTENTS OF VOLUMES IN THE METAL IONS IN LIFE SCIENCES SERIES

1

xix xxi

O R G A N O M E T A L L I C CHEMISTRY OF B 12 COENZYMES Bernhard Kräutler

1

Abstract 1. Introduction 2. Structure of B 12 Derivatives in the Crystal and in Solution 3. Redox Chemistry of B 12 Derivatives 4. Reactivity of B 12 Derivatives in Organometallic Reactions 5. Organometallic B 12 Derivatives as Cofactors and Intermediates in Enzymes 6. Concluding Remarks and Future Directions Acknowledgments Abbreviations References

2 2 5 18 24

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-FPO11

34 40 41 41 42

CONTENTS

xii 2

COBALAMIN- AND CORRINOID-DEPENDENT ENZYMES Rowena G. Matthews Abstract 1. Introduction. What Is a Corrinoid? 2. Corrinoid-Dependent Methyltransferases 3. Adenosylcobalamin-Dependent Rearrangements and Eliminations 4. Concluding Remarks Acknowledgments Abbreviations and Definitions References

3

NICKEL-ALKYL BOND FORMATION IN THE ACTIVE SITE OF METHYL-COENZYME Μ REDUCTASE Bernhard Jaun and Rudolf K. Thauer Abstract 1. Introduction 2. Nickel-Carbon Bond Formation in Free Coenzyme F 4 3 0 3. Nickel-Alkyl Bond Formation in MCR Upon Inactivation with Alkyl Halides 4. Methyl-Nickel Bond Formation in Methyl-Coenzyme Μ Reductase During Catalysis? 5. Observations to Be Followed Up Acknowledgments Abbreviations References

4

NICKEL-CARBON BONDS IN ACETYL-COENZYME A SYNTHASES/CARBON MONOXIDE DEHYDROGENASES Paul A. Lindahl Abstract 1. Introduction 2. Redox and Catalytic Properties of the A- and C-Clusters 3. Evidence for a Ni-CO Bond in the A red -CO State of the A-cluster

53 54 54 56 84 106 107 107 107

115 116 116 119 120 123 129 129 129 130

133 133 134 137 139

CONTENTS

Evidence for a Ni-CH 3 Bond in the Methylated Intermediate of the A-Cluster 5. Evidence for a N i - C ( 0 ) C H 3 Bond in the Acetyl Intermediate of the A-Cluster 6. Evidence for a Ni-CO Bond in the C-Cluster 7. Evidence for a N i - C ( 0 ) 0 - F e Bond in the C-Cluster 8. Conclusions and Future Studies Acknowledgment Abbreviations and Definitions References

xiii

4.

5

6

139 140 141 143 144 147 147 147

S T R U C T U R E A N D F U N C T I O N O F [NiFe]-HYDROGENASES Juan C. Fontecilla-Camps

151

Abstract 1. Introduction 2. Hydrogenase Structure 3. Hydrogenase Maturation and Active Site Assembly 4. Electron Transfer 5. Proton Transfer 6. Oxidized Inactive States of the [NiFe]-Hydrogenase Active Site 7. Substrate Binding and Catalysis 8. Concluding Remarks Acknowledgments Abbreviations References

152 152 153 160 165 166 168 172 173 173 173 174

CARBON MONOXIDE AND CYANIDE LIGANDS IN THE ACTIVE SITE O F [FeFe]-HYDROGENASES John W. Peters

179

Abstract 1. Introduction 2. [FeFe]-Hydrogenase Structure 3. [FeFe]-Hydrogenase Spectroscopic Studies 4. Η-Cluster Model Complexes 5. Η-Cluster Biosynthesis 6. Future Directions Acknowledgments

180 180 181 192 195 199 206 207

CONTENTS

xiv

7

Abbreviations and Definitions References

208 208

C A R B O N M O N O X I D E AS I N T R I N S I C L I G A N D TO I R O N I N T H E ACTIVE SITE O F [Fe]-HYDROGENASE Seigo Shima, Rudolf K. Thauer, and Ulrich Ermler

219

Abstract 1. Introduction 2. Physiology 3. The Iron Guanylylpyridinol Cofactor in the Enzyme-Free State 4. Structure of [Fe]-Hydrogenase with and without the Iron Guanylylpyridinol Cofactor Bound 5. Ligands to Iron in the Active Site of [Fe]-Hydrogenase 6. Proposed Catalytic Mechanisms 7. Concluding Remarks Acknowledgments Abbreviations References

8

T H E D U A L R O L E O F H E M E AS C O F A C T O R A N D SUBSTRATE I N T H E BIOSYNTHESIS O F C A R B O N MONOXIDE Mario Rivera and Juan C. Rodriguez Abstract 1. Introduction 2. The Biosynthesis of Carbon Monoxide 3. Heme Oxygenase Favors Heme Hydroxylation over Ferryl Formation. The Nature of the Ferric Hydroperoxide Complex in Heme Oxygenase 4. Heme Oxygenase Dynamics and Heme Breakdown. The Distal Ligand Has a Profound Effect in the Dynamic Behavior of Heme Oxygenase 5. The Regioselectivity of Heme Hydroxylation 6. Conclusion and Outlook Acknowledgments Abbreviations References

220 220 222 223 227 231 235 237 237 237 238

241

242 243 247

259

268 276 284 285 286 286

CONTENTS 9

xv

COPPER-CARBON BONDS IN MECHANISTIC AND STRUCTURAL PROBING OF PROTEINS AS WELL AS IN SITUATIONS WHERE COPPER IS A CATALYTIC OR RECEPTOR SITE

295

Heather R. Lucas and Kenneth D. Karlin

10

Abstract 1. Introduction 2. Binuclear Copper Proteins 3. Heterobimetallic Copper-Containing Enzymes 4. Non-Blue Copper Oxidases 5. Blue, Green, and Purple Copper Proteins 6. Copper(I) Recognition Sites or Receptors 7. Miscellaneous 8. General Conclusions Acknowledgments Abbreviations References

296 297 298 317 334 337 344 346 349 350 351 352

INTERACTION OF CYANIDE WITH ENZYMES CONTAINING VANADIUM, MANGANESE, NON-HEME IRON, AND ZINC

363

Martha E. Sosa-Torres

11

and Peter Μ. H. Kroneck

Abstract 1. Introduction 2. Vanadium Enzymes 3. Manganese Enzymes 4. Non-Heme Iron Enzymes 5. Zinc Enzymes 6. Conclusions Acknowledgments Abbreviations References

364 364 368 373 377 383 388 388 388 389

THE REACTION MECHANISM OF THE MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE: EVIDENCE AGAINST THE FORMATION OF INTERMEDIATES HAVING METAL-CARBON BONDS

395

Russ Hille

Abstract 1. Introduction

396 396

xvi

CONTENTS

2.

Electron-Nuclear Double Resonance Studies of the "Very Rapid" Species 3. X-Ray Crystal Structures Relevant to The Reaction Mechanism 4. General Conclusions Acknowledgments Abbreviations and Definitions References

12

400 404 413 414 414 414

COMPUTATIONAL STUDIES OF BIOORGANOMETALLIC ENZYMES A N D COFACTORS 417 Matthew D. Liptak, Katherine M. Van Heuvelen, and Thomas C. Brunold Abstract 1. Introduction 2. Computational Approaches to Bioorganometallic Chemistry 3. Formation and Cleavage of the Co-C Bond of Cobalamin in Enzyme Active Sites 4. Organometallic Chemistry and Catalytic Cycle of Methyl-Coenzyme Μ Reductase 5. Geometric and Electronic Structures of the Carbon Monoxide Dehydrogenase/Acetyl-Coenzyme A Synthase Active Site Clusters 6. Magnetic Properties of the Active Site Cluster of Iron-Only Hydrogenases 7. Concluding Remarks and Future Directions Acknowledgments Abbreviations References

418 419 420 426 435

442 447 450 452 452 454

SUBJECT INDEX

461

A U T H O R INDEX OF CONTRIBUTORS TO MIBS-1 TO Μ IBS- 44 A N D MILS-1 TO MILS-6

497

Contributors to Volume 6

Numbers in parentheses indicate the pages on which the contributions begin.

authors'

Thomas C. Brunold Department of Chemistry, University of WisconsinMadison, 1101 University Avenue, Madison, WI 53706, USA (Fax: + 1-608-262-6143) (417) Ulrich Ermler Max Planck Institute for Biophysics, Max-von-Laue-Strasse 3, D-60438 Frankfurt/Main, Germany (219) Juan C. Fontecilla-Camps Laboratoire de Cristallographie et de Cristallogenese des Proteines, Institut de Biologie Structurale "Jean-Pierre Ebel" (CEA-CNRS-UJF), 41 rue Jules Horowitz, F-38027 Grenoble Cedex 1, France ( F a x : + 33-4-3878-5122) (151) Russ Hille Department of Biochemistry, University of California, Riverside, CA 92521, USA (Fax: +1-951-827-3719) (395) Bernhard Jaun Organic Chemistry, ΕΤΗΖ, ΕΤΗ Hönggerberg, HCl E317, CH-8093 Zürich, Switzerland (115) Kenneth D. Karlin Department of Chemistry, The Johns Hopkins University, New Chemistry 213, 3400 N. Charles Street, Baltimore, M D 21218, USA (Fax: + 1-410-516-7044) (295) Bernhard Kräutler Institute of Organic Chemistry and Centre of Molecular Biosciences, University of Innsbruck, A-6020 Innsbruck, Austria (Fax: + 43-512-507-2892) (1) Peter Μ. H. Kroneck Fachbereich Biologie, Universität Konstanz, Postfach M665, D-78457 Konstanz, Germany (363)

xviii

CONTRIBUTORS TO VOLUME 6

Paul A. Lindahl Department of Chemistry and of Biochemistry and Biophysics, Texas A&M University, College Station, TX 77843-3255, USA (133) Matthew D. Liptak Department of Chemistry, University of WisconsinMadison, Madison, WI 53706, USA (417) Heather R. Lucas Department of Chemistry, The Johns Hopkins University, New Chemistry 213, 3400 N. Charles Street, Baltimore, M D 21218, USA (Fax: + 1-410-516-7044) (295) Rowena G. Matthews Department of Biological Chemistry and Life Sciences Institute, University of Michigan, Room 4002, 210 Washtenaw Avenue, Ann Arbor, MI 48109-2216, USA (Fax: + 1-734-763-6492) (53) John W. Peters Montana State University, Department of Chemistry and Biochemistry, Bozeman, M T 59717, USA (Fax: +1-406-994-7212) (179) Mario Rivera Ralph N. Adams Institute for Bioanalytical Chemistry, Department of Chemistry, The University of Kansas, Multidisciplinary Research Building, 2030 Becker Dr., Lawrence, KS 66047, USA (Fax: + 1-785-864-5396) (241) Juan C. Rodriguez Ralph N. Adams Institute for Bioanalytical Chemistry, Department of Chemistry, The University of Kansas, Multidisciplinary Research Building, 2030 Becker Dr., Lawrence, KS 66047, USA (241) Seigo Shima Max Planck Institute for Terrestrial Microbiology, Karl-von-Frisch-Strasse, D-35043 Marburg, Germany (219) Martha E. Sosa-Torres Facultad de Quimica, Universidad Nacional Autönoma de Mexico, Ciudad Universitaria, Coyoacän, 04510, D.F. Mexico, Mexico (Fax: +52-55-5616-2010) (363) Rudolf K. Thauer Max Planck Institute for Terrestrial Microbiology, Karlvon-Frisch-Strasse, D-35043 Marburg, Germany (Fax: +49-6421-178-109) (115, 219) Katherine M. Van Heuvelen Department of Chemistry, University of Wisconsin-Madison, Madison, WI 53706, USA (417)

Titles of Volumes 1-44 in the Metal Ions in Biological Systems Series edited by the and published by DekkerjTaylor

Volume 1: Volume 2: Volume 3: Volume 4: Volume 5: Volume 6: Volume 7: Volume 8: Volume 9: Volume 10: Volume 11: Volume 12: Volume 13: Volume 14: Volume 15: Volume 16: Volume Volume Volume Volume Volume

17: 18: 19: 20: 21:

Volume 22: Volume 23:

SIGELs & Francis (1973—2005)

Simple Complexes Mixed-Ligand Complexes High Molecular Complexes Metal Ions as Probes Reactivity of Coordination Compounds Biological Action of Metal Ions Iron in Model and Natural Compounds Nucleotides and Derivatives: Their Ligating Ambivalency Amino Acids and Derivatives as Ambivalent Ligands Carcinogenicity and Metal Ions Metal Complexes as Anticancer Agents Properties of Copper Copper Proteins Inorganic Drugs in Deficiency and Disease Zinc and Its Role in Biology and Nutrition Methods Involving Metal Ions and Complexes in Clinical Chemistry Calcium and Its Role in Biology Circulation of Metals in the Environment Antibiotics and Their Complexes Concepts on Metal Ion Toxicity Applications of Nuclear Magnetic Resonance to Paramagnetic Species ENDOR, EPR, and Electron Spin Echo for Probing Coordination Spheres Nickel and Its Role in Biology

XX

Volume 24: Volume 25:

VOLUMES IN THE MIBS SERIES

Aluminum and Its Role in Biology Interrelations among Metal Ions, Enzymes, and Gene Expression Volume 26: Compendium on Magnesium and Its Role in Biology, Nutrition, and Physiology Volume 27: Electron Transfer Reactions in Metalloproteins Volume 28: Degradation of Environmental Pollutants by Microorganisms and Their Metalloenzymes Volume 29: Biological Properties of Metal Alkyl Derivatives Volume 30: Metalloenzymes Involving Amino Acid-Residue and Related Radicals Volume 31: Vanadium and Its Role for Life Volume 32: Interactions of Metal Ions with Nucleotides, Nucleic Acids, and Their Constituents Volume 33: Probing Nucleic Acids by Metal Ion Complexes of Small Molecules Volume 34: Mercury and Its Effects on Environment and Biology Volume 35: Iron Transport and Storage in Microorganisms, Plants, and Animals Volume 36: Interrelations between Free Radicals and Metal Ions in Life Processes Volume 37: Manganese and Its Role in Biological Processes Volume 38: Probing of Proteins by Metal Ions and Their Low-Molecular-Weight Complexes Volume 39: Molybdenum and Tungsten. Their Roles in Biological Processes Volume 40: The Lanthanides and Their Interrelations with Biosystems Volume 41: Metal Ions and Their Complexes in Medication Volume 42: Metal Complexes in Tumor Diagnosis and as Anticancer Agents Volume 43: Biogeochemical Cycles of Elements Volume 44: Biogeochemistry, Availability, and Transport of Metals in the Environment

Met. Ions Life Sei. 2009, 6, 1-51

1 Organometallic Chemistry of B 1 2 Coenzymes Bernhard Kräutler Institute of Organic Chemistry and Centre of Molecular Biosciences, University of Innsbruck, A-6020 Innsbruck, Austria < bernhard. kraeutler @uibk .ac. at >

ABSTRACT 2 1. INTRODUCTION 2 2. STRUCTURE OF B12 DERIVATIVES IN THE CRYSTAL AND IN SOLUTION 5 2.1. "Incomplete" Corrinoids 5 2.2. "Complete" Corrinoids 7 2.2.1. "Complete" Corrinoids with an "Inorganic" ß-Ligand 7 2.2.2. "Complete" Corrinoids with an "Organic" ß-Ligand 10 2.2.3. Spectroscopic Studies of the Solution Structure of B 12 Derivatives 13 2.3. The "Base-On/Base-Off' Constitutional Switch of "Complete" Corrinoids 16 3. REDOX CHEMISTRY OF B12 DERIVATIVES 18 3.1. Thermodynamics of Redox Processes 20 3.2. Kinetics of the Redox Processes 22 4. REACTIVITY OF B12 DERIVATIVES IN ORGANOMETALLIC REACTIONS 24 4.1. Formation of the (Co-C) Bond in Organocorrinoids 25 4.2. Cleavage of the (Co-C) Bond in Organocorrinoids 31 5. ORGANOMETALLIC B12 DERIVATIVES AS COFACTORS AND INTERMEDIATES IN ENZYMES 34 5.1. Methylcorrinoids in B12-Dependent Methyltransferases 35 5.2. Adenosylcorrinoids in Enzymes Dependent on Coenzyme B 12 38 Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00001

2

KRÄUTLER

6. C O N C L U D I N G R E M A R K S A N D F U T U R E DIRECTIONS ACKNOWLEDGMENTS ABBREVIATIONS REFERENCES

40 41 41 42

ABSTRACT: When coenzyme B 12 was identified as organometallic derivative of vitamin B12, metal-carbon bonds were revealed to be relevant in life processes. Vitamin B12, the "antipernicious anaemia factor" required for human health, was isolated earlier as a crystallizable cyano-Co(III)-complex. B 12 cofactors and other cobalt corrinoids play important roles not only in humans, but in the metabolism of archaea and other microorganisms, in particular. Indeed, the microorganisms are the only natural sources of the B12 derivatives. For other B 12 -requiring organisms the corrinoids are thus "vitamins". However, vitamin B 12 also needs to be converted into organometallic B 12 -forms, which are the typical coenzymes in metabolically important enzymes. One of these, methionine synthase, catalyzes the transfer of a methyl group and its corrinoid cofactor is methylcobalamin. Another one, methylmalonyl-CoA mutase uses a reversible radical process, and coenzyme B 12 (adenosylcobalamin) as its cofactor, to transform methylmalonyl-CoA into succinyl-CoA. In such enzymes, the bound B 12 derivatives engage (or are formed) in exceptional organometallic enzymatic reactions, which depend upon the organometallic reactivity of the B 12 cofactors. Clearly, organometallic B12 derivatives hold an important position in life and have thus attracted particular interest from the medical sciences, biology, and chemistry. This chapter outlines the unique structures of B12 derivatives and recapitulates their redox properties and their organometallic chemistry, relevant in the context of the metabolic transformation of B 12 derivatives into the relevant coenzyme forms and for their use in Bi 2 -dependent enzymes. KEYWORDS: cobalt-carbon bond · cobalt complex · coenzyme B12 · electrochemistry · homolysis · methyl group transfer · organometallic bond · radical reaction · vitamin B12

1.

INTRODUCTION

The importance of a metal-carbon bond in enzymatic processes was first revealed in the early 1960s, when coenzyme B 12 was identified as organometallic derivative of vitamin B 12 [1]. Vitamin B 12 was discovered and isolated 60 years ago as a crystallizable, red complex [2,3], and was revealed to be a cobalt complex of the remarkable corrin ligand, a unique member of the natural tetrapyrroles [4]. Organometallic B 12 forms are the coenzymes in a variety of metabolically important enzymes. In humans, methionine synthase and methylmalonyl-CoA mutase use methylcobalamin and coenzyme B 12 , respectively, as their B 12 cofactors [5-12]. Bi 2 coenzymes are now known to be required in the metabolism of a broad range of organisms. However, only microorganisms have the capacity

Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 12 COENZYMES

3

to biosynthesize B 12 and other natural corrinoids [13,14]. For other I n dependent organisms, such as humans, B 12 derivatives are thus vitamins [15]. Their functioning metabolism depends on the uptake and binding of B 12 derivatives [16], on their metabolic transformation into the relevant B 12 cofactors [17], the controlled transport of these [16] and the catalysis by B 12 -dependent enzymes [18-23]. Perhaps B 12 coenzymes are Nature's physiologically most relevant organometallic cofactors [9-12]. Organometallic B 12 derivatives engage in protein-activated reactions in unique Bi 2 -dependent enzymes. In these, Bi 2 cofactors (co)catalyze exceptional enzymatic reactions [18-23] that directly depend upon the reactivity of the cobalt-carbon bond [11,12,22]. Since the surprising identification of coenzyme B 12 as an organometallic compound, corrinoid cofactors have thus played central roles in the discovery and better understanding of biological organometallic processes [24-28]. Vitamin B 12 was identified as the "antipernicious anemia factor" over 60 years ago. It was isolated as the red cyanide-containing cobalt complex cyanocob(III)alamin (1, CNCbl), which crystallized readily and was revealed to be a relatively inert Co(III) complex [2,3]. It is the most important commercially available form of the naturally occurring B 12 derivatives, but it appears to have no physiological function itself [15]. The physiologically relevant vitamin B 12 derivatives are the highly light-sensitive and chemically more labile organometallic coenzymes, coenzyme B 12 (2, 5'-deoxy-5'-adenosylcobalamin, AdoCbl) and methylcobalamin (3, MeCbl), as well as the "inorganic" and easily reducible B 12 derivatives aquacob(III)alamin (4 + , H 2 O C b l + ) and hydroxocob(III)alamin (5, HOCbl) [6,29,30] (Figure 1). During the last six decades, remarkable scientific advances towards the solution of some of the major "B 12 -mysteries" have been achieved. Five European Symposia on "Vitamin B 12 and B 12 -Proteins" were dedicated to this subject, the first two of which took place in Hamburg (1956 and 1961, Germany), followed by one in Zürich (1979, Switzerland) [7], in Innsbruck (1996, Austria) [9], and in Marburg (2000, Germany). Among the top achievements in the B 12 field the elucidation of the structure of vitamin B 12 [4,31] and of coenzyme B 12 [1,32] are to be highlighted, the synthetic conquest of the vitamin B 12 structure [33-35], the elucidation of the biosynthetic pathways to B 12 [13,14], as well as crystal structures [36-44] and a solution structure [45] of a variety of B 12 -binding proteins and B 12 -dependent enzymes. Several concise books on B 12 have been written, with earlier ones by Pratt [5] and by Friedrich [6]. More recent ones on "B 1 2 " [7], "Vitamin B 12 and B 12 -Proteins" [9] and on "Chemistry and Biochemistry of B 1 2 " [10] and extensive reviews [11,12] describe the more recent findings on the chemistry of B 12 and on the biological roles of the B 12 derivatives.

Met. Ions Life Sei. 2009, 6, 1-51

4

KRÄUTLER

Figure 1. General structural formula. Left: cobalamins (Cbls = DMB-cobamides, Ado = adenosyl). Vitamin BI 2 (1, CNCbl, R = CN), coenzyme BI 2 (2, R = 5'-deoxy5'-ado), methylcobalamin (3, MeCbl, R = CH 3 ), aquacobalamin (4 + , R = H 2 0 + ) , hydroxocobalamin (5, HOCbl, R = HO), cob(II)alamin (6, B 1 2 r , R = e~), chlorocobalamin (18 , R = C1), nitroxylcobalamin (19, R = NO), 2,3-dihydroxypropyl-Cbl (21, R = 2,3-dihydroxy-propyl), a-adenosyl-Cbl (22, R = 5'-deoxy-5'-a-Ado), adeninylpropyl-Cbl (23, R = 3-adeninyl-propyl), homocoenzyme B 12 (24, R = 5'-deoxy-5'Ado-methyl), trifluoromethyl-Cbl (25, R = CF 3 ), difluoromethyl-Cbl (26, R = CHF 2 ), vinylcobalamin (28, R = C H = C H 2 ) , cw-chlorovinyl-Cbl (29, R = CH=CHC1), bishomocoenzyme B 12 (33, R = 2-[5'-deoxy-5'-Ado]-ethyl), 2'-deoxycoenzyme B 12 (48, R = 2',5'-dideoxy-5'-Ado), 2',3'-dideoxycoenzyme B 12 (49, R = 2',3',5'-trideoxy-5'-Ado). Right: Structural formulae of other naturally occurring "complete" corrinoids (cobamides with other nucleotide functions " N u " [6,11]: Co p cyano-imidazolylcobamide (14, R = CN, N u = imidazole), Coß-methyl-imidazolylcobamide (27, R = CH 3 , N u = imidazole); pseudovitamin B 12 (Co p -cyano-7"-adeninylcobamide, 16, R = C N , N u = adenine), factor A (Co p -cyano-7"-[2'-methyl]adeninyl-cobamide, 17, R = CN, N u = 2-methyl-adenine). Bottom: symbols used.

Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 12 COENZYMES

2. 2.1.

5

STRUCTURE OF B 1 2 DERIVATIVES IN THE CRYSTAL AND IN SOLUTION "Incomplete" Corrinoids

The structures of vitamin B 12 (1) and of coenzyme B 12 (2) were established largely through the pioneering X-ray crystallographic studies of Hodgkin et al. [4,31,32], who discovered the composition of the corrin core of 1 and the organometallic nature of 2 (see recent reviews [46^18]). These two cobalamins belong to the "complete" corrinoids, in which a pseudonucleotide function is attached via an amide linkage to the corrin moiety. The resulting combination is unique, as the B 12 -nucleotide function may switch between the cobalt-coordinated "base-on" form and the de-coordinated (so called "base-off form) - coenzyme Bi 2 thus can be considered to be a "molecular switch" [49]. Earlier important X-ray investigations also specifically relied on some crystallizable "incomplete" Co(III)-corrinoids, such as a-cyano-ß-aqua cobyric acid (7) (see review [46]). Cobyric acid is the natural corrinoid moiety of the vitamin B 12 and it was the initial target for Eschenmoser and Woodward for the total synthesis of vitamin B 12 [33-35] (as it had already been shown how 7 could be converted to vitamin B 12 ). Since these times, crystallographic work with "incomplete" Co(III)-corrinoids has focused on obtaining detailed structural information. More recently analyzed structures include that of dicyano-heptamethyl-cobyrinate (8, "cobester" [50]) [51,52] and of its analogues [53], including 15-norcobester (9) [54,55] (as reviewed in [47]). Very recently, a new type of B 12 dimer structure was found in the crystal of Cop-cyano-neocobyric acid (10) [56] (Figure 2). A similar dimer structure has not been observed in the "normal" corrinoids. However, in the "neo" corrinoid 10, the configuration at the corrin ring C was inverted, apparently reducing the steric hindrance due to a propionate side chain, and making the dimer formation possible [56]. The crystal structure of heptamethyl-cob(II)yrinate (cob(II)ester (as the Perchlorate complex 11) gave the first detailed insights into the structure of a paramagnetic Co(II)-corrin [57]. It revealed a five-coordinated Co(II) center in the "incomplete" Co(II)-corrin 11, to which a Perchlorate ligand was coordinated via a long axial cobalt-oxygen bond (2.31 A) [57]. The coordination of the axial ligand at the sterically less hindered "upper" ß-face is in contrast with the preference seen in the "complete" cob(II)alamin (6, B 12r ) [58,59] (see below).

Met. Ions Life Sei. 2009, 6, 1-51

6

KRÄUTLER H 3 CO 2 C

CONHn

,CO2CH3 H 3 CO 2 C

CONHn

HnNOC

CO,CH,

H 3 CO 2 C'

HnCOnC 034 073 N34 Q33 w 023 \ C72—N73 / W /C32 C51 C71 C3 / \ ,J N23 1 ,^ C7A C5 C 2 1 ^ C3. •C4 w H , C8'HC81 N84 ' \ / 'C10 C82-C83 / N4 L' N3-C11 »C12B 084 0183 / ® // / \ C12 ,,„ \ C16 ,C14 C182 C17 # C13 "/C12A ^ C15 I N183 C17B C171 C131 C151 0^72 ^0132 I C173 \ C133 N134 0174 N174 0134 I C175 I C176 / \ C177 0177

CO 2 CH 3

OH

Figure 2. Top left: Structural formulae of "incomplete" corrinoids: Co a -cyano-Co p aqua cobyric acid (7, R = H 2 0 , Ri = propionamide, R'i = Η, V = CN, X = OH); COa-aqua-Cop-cyano-13-epicobyric acid (10, R = CN, R'j = propionamide, R j ^ H , V = H 2 0 , X = OH), Co p -5'-deoxy-5'-adenosylcobinamide (32, R = 5'-deoxy-5'-Ado, R i = propionamide, R, - H, L' - Η,Ο, X - N H ( Ή , ( HOH ( H ;). Top right: Structural formulae of heptamethyl-cobyrinates: cobester (8, L = L' = CN, X = CH 3 ), 15-nor-cobester (9, L = L' = CN, X = H); perchlorato-heptamethylcob(II)yrinate (11, L = C10 4 , L' = vacant, X = CH 3 ). Bottom left: Atom numbering for cobinamide [60]. Bottom right: general symbol used for a cob(III)inamide.

Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES

2.2.

7

" C o m p l e t e " Corrinoids

The " c o m p l e t e " corrinoid vitamin B 1 2 (1, C N C b l ) is called a " c o b a l a m i n " (or a 5',6'-dimethylbenzimidazolyl-cobamide = D M B - c o b a m i d e ) , in which a cyanide ligand is b o u n d at the " u p p e r " axial c o o r d i n a t i o n site (on the ß-face, see e.g., [60]). In other " c o b a l a m i n s " , the cyanide g r o u p at the ß-face of C N C b l is replaced by a different ligand, e.g., an organometallic g r o u p , as in coenzyme B 1 2 (AdoCbl, 2). Purinyl-cobamides are a n o t h e r i m p o r t a n t class of naturally occurring " c o m p l e t e " corrinoids, in which a purine base is p a r t of the nucleotide f u n c t i o n (see Figure 1) [6,11]. T h e systematic a t o m n u m bering used in this chapter for vitamin B 1 2 derivatives [60] builds on the convention that a t o m n u m b e r s of the heavy a t o m s of a substituent reflect the n u m b e r of the points of a t t a c h m e n t to the corrin ligand and are indexed consecutively [61] (see Figure 2); however, it deviates f r o m that introduced by H o d g k i n et al. [31,32,46] and still used by some a u t h o r s [12].

2.2.1.

"Complete"

Corrinoids with an "Inorganic"

ß-Ligand

Only " b a s e - o n " cobalamins, where the nucleotide functionality coordinates in an intramolecular m o d e , have been analyzed by X-ray crystallography [46-48]. The " o l d " structure of vitamin B 1 2 [7] was re-analyzed using m o d e r n cryo-crystallography techniques [47,62], which showed the molecular geometry of the B I 2 moiety to agree within experimental error of H o d g k i n ' s original result. N e o v i t a m i n B 1 2 (12) was also studied by crystallography and was revealed to be the cyano-13-epicobalamin, a derivative of vitamin B 1 2 where the p r o p i o n a m i d e chain at the C I 3 position is in the ß-configuration [63,64]. A notable difference between the two structures is an increased "nonplanarity" in the corrin ring of the neo-derivative (expressed as a 6° larger fold angle, 23.7° versus 17.9°). T h e C8 epimer of vitamin B 1 2 , cyano-8-epicobalamin (13), has an even larger fold angle of the corrin core (23.8°) [65] (Figure 3). T h e discovery of the replacement of the cobalt-coordinated 5,6-dimethylimidazole base by a protein-derived imidazole in several B 1 2 -dependent enzymes [36-38]), gave the analysis of Cop-cyano-imidazolylcobamide (14) [62] particular interest. The less bulky and m o r e nucleophilic imidazole base of 14 caused a n u m b e r of structural differences. The corrin ring fold angle of 14 decreased to 11.3° and the axial ( C o - N ) b o n d shortened (from 2.011 Ä in 1 to 1.968 Ä in 14). In addition, the " b a s e tilt" of 12 (i.e., half the difference between the two C o - N - C angles to the coordinating base) decreased to near zero, within experimental error. In all crystal structures of 5',6'-dimethylbenzimidazoyl-cobamides a " t i l t " of a b o u t 5° is Met. Ions Life Sei. 2009, 6, 1-51

8

KRÄUTLER •S ό ο" g -Ö Ν s 3 κ 11 > >> Ο ft .. 1—1 -ö .a Ο ^ qj ~ s ω "5. ο ω 5 'S μ ί-Η ft π) 6 η ο Ο Η ϋ£ ÖJ π Ο -Ö —I —1 -Ö Ö ω ω χ; 3 £ ιι r> β

! U

£ υ >> II Ν * S α 8

>η

< β

's § π) Ο .3 S

β 13 " S .Sf -Ο ω 3 Pi Ο -Ö β (]J 03 3 m Ö Ο £ & A -ö 2 ο > ι Μ ο 2 öS CN £ II pi a o(Ν |f11

ffi Pi

pi

I3. a a -Ö I

3 c. ο ρ N^ I—I OS £ β .„ ^ ö μ ιä §JH 1 Η \^CH3 OH

41+

42

Figure 6. Outline of the redox transitions between cob(III)inamide cob(II)inamide (41 + ) and cob(I)inamide (42).

(40 2+ ),

Met. Ions Life Sei. 2009, 6, 1-51

22

KRÄUTLER

potential (at - 3 0 °C) for the methylcob(III)alamin (3)/methylcob(II)alamin redox couple was estimated as —1.60 V versus SCE [139,149] similar to the value obtained for the coenzyme B 12 /5'-deoxy-5'-adenosyl-cob(II)-alamin pair [144]. The standard potential of the typical Co(III)/Co(II) redox pair of organometallic B 12 derivatives is significantly more negative than that of Bi2r (6)/B 12s (39~) and out of the reach of biological reductants. However, upon one-electron reduction of 3 fast decoordination of the nucleotide base occurs, followed by rapid decomposition to a methyl radical and cob(I)alamin (39~). The thermodynamic features of B 12 -redox systems can be summarized as: (i) Intramolecular coordination of the nucleotide base or strong coordinating or nucleophilic ligands (such as cyanide ions) stabilize the corrin-bound cobalt center against one-electron reduction and shift the Co(III)/Co(II) redox couples to more negative potentials. (ii) The one-electron reduction of organometallic Co(III)-corrins typically occurs at more negative potentials than the Co(II)/Co(I) redox couple B 12r /B 12s [139]. Exceptions to this are provided by organometallic B 12 derivatives with electron withdrawing substituents on the organometallic group, such as methoxycarbonylmethyl-cob(III)alamin [150].

3.2.

Kinetics of the Redox Processes

One-electron transfer reactions of cobalt corrinoids are accompanied by either cleavage or formation of a bond to an axial ligand. Typically, a reduction is accompanied by an expulsion, and an oxidation by the coordination, of the ligand [139]. The electron transfer step accordingly takes place either in a concerted fashion or in a rapid associated step with coordination or dissociation of the axial ligand. Electron transfer in the protonated Co(II)/Co(I) couple B 1 2 r -H + (6-H + )/ Bi 2 s -H (39-H) is fast in aqueous solution (kl p p >0.1 c m s - 1 ) as the presumed axial water ligand is only kinetically weakly bound in the base-off Co(II)corrin 6 - H + [139,149]. However, when the aqua ligand in 6 - H + is substituted by a stronger axial ligand, e.g., by the nucleotide base as in base-on B 12r , the electron transfer is slowed down sufficiently so that its kinetics can be conveniently measured by cyclic voltammetry [139,151,152]. For example, in the Co(II)/Co(I) redox couple 6/39" k f p = 0.0002cms" 1 [139], the electron transfer is at least a thousand times slower than in the base-off forms 6 - H + / 39-H. Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 12 COENZYMES

23

The trend in kinetics for Co(III)/Co(II) couples follows the same trend as those for the corresponding Co(II)-/Co(I)-couples, albeit much slower. The Co(III)/Co(II) couple aquacob(III)alamin (4 + )/B 1 2 r (6) has a rate constant for heterogeneous electron transfer of about K T 5 c m s - 1 [139]. The electron transfer steps for the cyano-cob(III)- and cyano-cob(II)alamins 1 - C N " and 23-CN" are slower still [139,147], There is an approximate linear correlation between the equilibrium constant for the coordination of the axial ligand and the standard apparent rate constant for electron transfer [139]. This correlation has been rationalized by a model, in which stretching of the bond between the cobalt ion and the axial ligand represents the main factor of the kinetics of the electron transfer. As a consequence, kinetic and thermodynamic dependence of the electron transfer on the strength of the complexation of the axial ligands both add up, resulting in more negative reduction potentials as the strength of the ligand increases. Organocobalamins, such as coenzyme B 12 (2) and MeCbl (3), have a different kinetic behavior from CNCbl (1) and other Co(III)-corrins with strong axial ligands [139,153,154], Whereas the Co(III)/Co(II) reduction potentials are quite negative, the kinetics of electron transfer are fast. The one-electron reduction of 3 to the unstable methylcob(II)alamin anion (43~) was estimated to have a rate constant of 1200 s - 1 at 30 °C. However, the product of the one-electron reduction of methylcobinamide (44 + ), methylcob(II)inamide (45), has a half life of only about 0.1s at — 20 °C and decomposes into a methyl radical and cob(I)inamide (42) (see Figure 7). An Arrhenius plot of the kinetics of the decomposition of 45 gave the activation

Figure 7. One-electron reduction of methylcob(III)inamide (44 + ) gives methylcob(II)inamide (45), which rapidly decomposes into cob(I)inamide (42) and a methyl radical. Met. Ions Life Sei. 2009, 6, 1-51

KRÄUTLER

24

energy to be 19kcal/mol and a pre-exponential factor A = 1 0 1 7 ' 6 s _ 1 [153]. F r o m the values of the ( C o - C ) bond dissociation energy (37kcal/mol) of M e C b l (3) [155] and the kinetics of the decomposition of the intermediate 43", the one-electron reduction is suggested to reduce the strength of the ( C o - C ) b o n d of 3 (by a b o u t 12kcal/mol) to " h a l f of its value [139,155],

4.

REACTIVITY OF B 12 DERIVATIVES IN ORGANOMETALLIC REACTIONS

F o r m a t i o n and cleavage of the ( C o - C ) b o n d in organometallic B 1 2 cofactors are crucial steps n o t only in the reactions catalyzed by B 1 2 -dependent enzymes, but also for the chemistry of B 1 2 derivatives in solution [912,22,25,156-159]. T h e reactivity of B 1 2 derivatives in organometallic reactions thus holds the key to the understanding of the biological activity of the B 1 2 -dependent enzymes. T h e most c o m m o n methods for the p r e p a r a t i o n of such organometallic B 1 2 derivatives typically rely on the efficient alkylation of Co(I)-corrins. One practical m e t h o d is an electrochemical a p p r o a c h , as also described below. In solution cleavage and f o r m a t i o n of the ( C o - C ) b o n d have been observed to occur in all three basic oxidation levels of the corrin-bound cobalt center [22,156-159]. So far, two m a i n paths for these organometallic reactions have also been f o u n d to be relevant in enzymes: (i) the homolytic m o d e is typical of the reactivity of coenzyme B 1 2 (AdoCbl, 2): 5'-adenosyl-Co(III)-corrin ^ Co(II)-corrin + 5'-adenosyl radical F o r m a l l y it is a one-electron reduction/oxidation of the corrin-bound cobalt center and it results in the cleavage or f o r m a t i o n of a single axial b o n d [11,22,158-160], (ii) the nucleophile-induced, heterolytic m o d e is typical of the reactivity of M e C b l (3): methyl-Co (III)-corrin + nucleophile ^ Co(I)-corrin + methylating agent Formally, it involves a two-electron reduction/oxidation of the corrinb o u n d cobalt center and the cleavage or f o r m a t i o n of two (trans) axial b o n d s [11,22], Met. Ions Life Sei. 2009, 6, 1-51

25

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES

4.1.

Formation of the (Co-C) Bond in Organocorrinoids

One i m p o r t a n t type of reactivity of B i 2 derivatives is represented by the highly nucleophilic Co(I)-corrins [140]. These provide the basis of the standard (heterolytic) m o d e of f o r m a t i o n of the ( C o - C ) bond, also important in methyl-corrinoids in enzyme-catalyzed methyl transfer reactions [23,161,162]. This m o d e is represented by the reaction of Co(I)-corrins with alkylating agents in the f o r m a t i o n of the ( C o - C ) b o n d (and the nucleophileinduced demethylation of methyl Co(III)-corrins for the cleavage of the ( C o - C ) bond). Overall an oxidative trans addition occurs at the corrinb o u n d cobalt center [125,163] (Figure 8). Alkylation at the c o r r i n - b o u n d Co(I) center normally proceeds via the "classical" bimolecular nucleophilic substitution (SN 2 ) mechanism, where the Co(I)-corrin acts as a "supernucleophile" [140,164]. However, in certain cases alkylation occurs via a two-step one-electron transfer p a t h , where Co(I)-corrins act as strong one-electron reducing agents and the process goes via Co(II)-corrin intermediates [112]. W i t h " c o m p l e t e " corrins, such as B 1 2 s (39~), either p a t h w a y results in alkylation at the ß-face, which allows the nucleotide to coordinate at the α-face of alkylcobalamins, such as M e C b l [22,125]. W h e n the nucleotide base has been changed f r o m a D M B base to an imidazole, little effect on the t h e r m o d y n a m i c s of the methyl transfer reaction occurs [94]. T h e studies of Co(I)-corrins, like B 1 2 s (39~), have shown the following reactivity patterns relevant for the S N 2 alkylation pathway: (i) the nucleophilicity of Co(I)-corrins is virtually independent of the presence of the D M B nucleotide, b o t h " c o m p l e t e " and " i n c o m p l e t e " Co(I)-corrins react preferentially at their ß-face, which is essentially m o r e nucleophilic [125]. The immediate p r o d u c t of the ß-alkylation

?H 3

39

?H 3

3

Figure 8. Methylation of cob(I)alamin B 12s (39 ) by an SN 2 mode is directed to the "upper" ß-face (by both, kinetic and thermodynamic reasons) and yields MeCbl (3). Met. Ions Life Sei. 2009, 6, 1-51

26

KRÄUTLER

may be a penta-coordinate (or already solvated and effectively hexacoordinate) Cop-alkyl-Co(III)-corrin; (ii) in aqueous solution and at room temperature the "base-on" (hexacoordinate) methylcob(III)alamin is more stable by about 4 kcal/mol than the " b a s e - o f f Co a -aqua-Cop-methylcob(III)alamin [126]. From N M R studies, the latter has been estimated to still be more stable in water, by around 7 kcal/mol, than the corresponding " b a s e - o f f and dehydrated form of Cop-methylcob(III)alamin, which has a pentacoordinate Cop-methyl-Co(III) center [165].

With "incomplete" cobalt-corrins the situation is again more complex, with two diastereoisomeric alkylation products often formed [11,12,112]. In specific cases, under suitable kinetic control, one of the alkyl-Co(III)-corrin diastereoisomers can form with high selectivity. For example with the Co(I) form of the lipophilic heptamethylcob(II)yrinate (11), the SN 2 -pathway can provide ß-methylation with high diastereoselectivity ( > 9 6 % ) , whilst the one-electron transfer mechanism permits the formation of the COo,-methylated product with high diastereoselectivity ( > 9 8 % ) [22,112]. However, configurational equilibration via rapid methyl group transfer reactions (involving Co(I)-, Co(II)-, and unalkylated Co(III)-corrins as methyl group acceptors) may give another overall outcome [22,126]. Electrochemistry is an excellent method for the selective and controlled production of reduced B 12 forms under potentiostatic control. As alkyl halides or alkyl tosylates react quickly and efficiently with Co(I)-corrins [138], which are cleanly generated at controlled electrode potentials near that of Co(II)/Co(I) couples, electrochemistry provides a suitable method for the synthesis of organometallic B 12 derivatives [139]. Indeed, the one-electron reduction of organometallic Co(III)-corrins typically occurs at more negative potentials than the Co(II)/Co(I) redox couple B 1 2 r / B 1 2 s [139]! Using electrolysis at a controlled potential o f - 1 . 1 V versus SCE, coenzyme B 1 2 (2) was prepared in 95% yield from vitamin B 1 2 (1) or from aquacobalamin (4 + ) by alkylating cob(I)alamin (39~) with 5'-chloro-5'deoxyadenosine [166]. Other organometallic B 1 2 derivatives produced in an analogous method were, e.g., pseudocoenzyme B 1 2 (37) (78% yield from pseudovitamin B 12 ) [110], neocoenzyme B 12 (39) (89% from neovitamin B 12 ) [111] and homocoenzyme B 12 (24) (99% from 4 + ) [85], Cop-methyl-imidazolylcobamide (31) (90% yield from Cop-cyano-imidazolylcobamide) [94] and methyl-13-epicob(III)alamin (46) (88% yield from neovitamin B 12 ) [111] were synthesized by alkylation with methyl iodide. Also, dimeric B 1 2 derivatives, such as the Cop-alkyl bridged and sterically crowded tetramethylene-Cop-l,4-biscobalamin (30) [98], and a strained organometallic B 1 2 -rotaxane [167], were synthesized by similar methods (see Figure 9). Met. Ions Life Sei. 2009, 6, 1-51

O R G A N O M E T A L L I C C H E M I S T R Y OF B 1 2 C O E N Z Y M E S

27

Figure 9. Electrochemistry as means for the preparation of alkyl-bridged biscorrinoids. Structural formulae of tetramethylene-bridged biscobalamin (30, n = l ) [97] and of a dodecamethylene-bridged biscobalamin (n = 5); symbolic representations of alkyl-bridged biscobalamins and of a cyclodextrin-based B 1 2 -rotaxane [166]. Met. Ions Life Sei. 2009, 6, 1-51

28

KRAUTLER

T h e high nucleophilicity of cob(I)alamin (39~) towards alkylating agents makes it a versatile tool for the detection of toxicologically relevant electrophilic reagents. Such analytical m e t h o d s are facilitating in vitro and in vivo studies of genotoxic c o m p o u n d s in cancer risk assessment. M a n y genotoxic c o m p o u n d s are directly (or indirectly) electrophilically reactive. The use of cob(I)alamin (39~) as an analytical tool has been investigated in the t r a p p i n g of oxiranes, metabolites of alkenes, to f o r m alkyl-Cbls (Figure 10) [168,169]. It is presumed that the reaction proceeds according to an S N 2 reaction following attack at the least hindered c a r b o n [170]. In the w o r k of Fred et al. [168], the 1,2-epoxide metabolites (oxiranes) of 1,3-butadiene were studied. F o r each metabolite a specific alkyl-Cbl was f o r m e d and it was possible to discriminate between the products by H P L C U V and by L C - M S . T h e cob(I)alamin (39"), used in this study, h a d the advantage of reacting a b o u t 400,000 times faster t h a n , e.g., nicotinamide, and therefore gave a better on-the-spot account [168]. Similar processes m a y also be relevant in vivo, as m a d e likely by recent studies, which aimed to mimic the chemical reactions t h a t could deplete vitamin B 1 2 as a result of h u m a n exposure to electrophilic xenobiotics (styrene, chloroprene, and 1,3butadiene) [171], T h e electrochemical m e t h o d o l o g y (see previous section) has been f u r t h e r developed as a m e t h o d for the clean p r e p a r a t i o n of easily reduced

Figure 10. Illustration of the alkylation of a cob(I)alamin (39 ) by reaction with an oxirane [167]. Met. Ions Life Sei. 2009, 6, 1-51

29

ORGANOMETALLIC CHEMISTRY OF B 12 COENZYMES

organocorrinoids, such as Cop-[(methoxycarbonyl)methyl]-cob(III)alamin (47) via the alkylation of cob(II)alamin (6) [172]. Easily reducible organocob(III)alamins are known to be cleaved by direct electrochemical reduction or by reduction with cob(I)alamin (39~) [150]. An acceptor-substituted C atom is directly bound to the Co center in 47 inducing it to be reduced at a peak potential of 0.90 V versus SCE (in D M F , room temperature) [173]. This value is close to that of the redox couple (base-on) B 12r (6)/B 12s (39~) (—0.85 V) and explains the difficulties encountered when preparing 47 via alkylation of 39" [173]. However, when aquacob(III)alamin chloride (4 + Cl") was submitted to a controlled potential of —0.45 V (versus 0.1 ν CE) under potentiostatic conditions, it gave 6 cleanly. After the addition of an excess of methylbromoacetate and continued electrochemical reduction at —0.45 V, the crystalline alkyl cob(III)alamin was isolated in 75% yield [172]. The reaction was proposed to take place directly via 6 and radical intermediates [172]. The alkylation of "complete" Co(II)corrinoids with sufficiently reactive alkylating agents (methyl iodide, methylbromoacetate, etc.) thus is an efficient and alternative method to the more established synthetic procedures via Co(I)corrinoids for the synthesis of reduction-labile Co(III) organocorrinoids [172,174]. Organocobalamins are also accessible by the reaction between Co(II)corrins and radicals. In particular, the radicaloid cob(II)alamin (B 12r , 6) has a penta-coordinated Co(II) center and can be considered to fulfill all the structural criteria of a highly efficient "radical trap", as revealed by the crystal structure of 6 [58]: the reactions of B 12r (6) with alkyl radicals are indicated to occur with negligible restructuring of the (DMB nucleotide coordinated) cobalt corrin moiety and to furnish coenzyme B 12 (2) and other organo-Cbls directly by the "homolytic" mode of formation of the (Co-C) bond [58]. From this it is understandable that the remarkably high reaction rate of 6 with alkyl radicals (such as the 5'-deoxy-5'-adenosyl radical), and the diastereospecificity for the reaction at the ß-face, are both consistent and explainable due to the structure of cob(II)alamin (see Figure 11). The

R

6 Figure 11. The "radical trap" cob(II)alamin (6) rapidly combines with radicals on the "upper" ß-face. Met. Ions Life Sei. 2009, 6, 1-51

30

KRÄUTLER

coordination of the DMB-nucleotide in 6 controls the (a/ß)-diastereoface selectivity (in both a kinetic and thermodynamic sense) in alkylation reactions at the Co(II) center, which give ß-alkyl-Cbls directly [141,159]. The stereochemical situation is appreciably more complex in "incomplete" corrins, such as cob(II)ester (11) and "base-off'-forms of "complete" corrins [22]. The axial ligand at the corrin-bound Co(II) center is expected to direct the formation of the (Co-C) bond. In this way kinetic control can lead with high efficiency to the "rare" a-alkyl-Co(III)-corrins [112,175]. In such radical recombination reactions the axial ligand at the a- or ß-side of the metal center will not only steer the diasteroselectivity of the alkylation but also may contribute to significant altering of the cage effects [160,176]. The two most relevant modes of formation (and cleavage) of the (Co-C) bond of the cobalt center differ significantly in their structural requirements (see Figure 12): • The heterolytic mode of formation (and cleavage) of the (Co-C) bond, in which significant reorganizations at both faces of the corrin-bound cobalt center occur; • The homolytic mode of formation (and cleavage), in which the cobaltcorrin portion of complete cob(III)amides (such as 2 and 3) hardly changes structure. Photolysis of methylcobalamin (MeCbl, 3) in deoxygenated aqueous solution saturated with (pressurized) carbon monoxide, gave acetyl-cobalamin in good yield and in a radical reaction, which was considered to finally involve the (re)combination of cob(II)alamin (11) with an acetyl radical [177]. This experiment turned out not to be relevant for the biological

Co Ό

7

•CH,

-FT

-Nu

supernucleophile

Q tz

-Ft*

Ο methylating agent

ISiu source of alkyl radical

-o

^

N'u

radical trap

Figure 12. Elementary reaction steps in organometallic and redox transformations of "complete" corrinoids, and their patterns of reactivity relevant for their cofactor function in B 1 2 -dependent enzymes. Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES

31

assembly of acetyl-CoA by the n o w well k n o w n enzyme acetyl-CoA synthase [178]. However, this organometallic t r a n s f o r m a t i o n with a B 1 2 -derivative turned out to find considerable interest as a model for the "slaving-in" mechanism in radical reactions [179]. A n o t h e r special m o d e of the f o r m a t i o n of ( C o - C ) b o n d s in alkyl Co(III)corrins involves nucleophilic alkylating agents and the electrophilic p r o p erties of aqua-Co(III)-corrins [156,157,180]. A f u r t h e r means of preparing methyl-corrinoids is opened by methyl g r o u p transfer reactions between corrinoids and methyl-corrinoids and some other methyl-organic c o m p o u n d s [22,126,181].

4.2.

Cleavage of the (Co-C) Bond in Organocorrinoids

As coenzyme B 1 2 (AdoCbl, 2) is considered to be a "reversible carrier of an alkyl radical" (or a reversibly functioning "radical source" [159]), the homolytic m o d e of the cleavage of the ( C o - C ) b o n d of 2 is of particular i m p o r t a n c e in its role as a cofactor. T h e strength of the ( C o - C ) b o n d of A d o C b l has been calculated to be a b o u t 30 kcal/mol by using detailed kinetic analyses of the thermal decomposition of 2 [159,160,182]. Considerable cage effects, and the presence of b o t h " b a s e - o n " and " b a s e - o f f f o r m s of 2, caused complications in the quantitative t r e a t m e n t of the homolytic ( C o - C ) bond dissociation energy (BDE) [160]. In several organocobalamins, the nucleotide coordinated " b a s e - o n " f o r m s decomposed faster t h a n their corresponding nucleotide-deficient organocobinamides or their p r o t o n a t e d ( " b a s e - o f f ) f o r m s of the o r g a n o c o b a l a m i n s [183,184]. The intramolecular c o o r d i n a t i o n of the nucleotide was therefore considered to cause a " m e c h a n o c h e m i c a l " means of labilizing the ( C o - C ) b o n d of organometallic B 1 2 derivatives [159,183,184]. The extension of this idea to the enzymatic reactions with 2 as cofactor was disputed [58]. In the time since, crystallographic studies of coenzyme B 1 2 -dependent enzymes also helped to dismiss m u c h of the original idea concerning the direct " m e c h a n o c h e m i c a l " mechanism. They rather suggested the specific stabilization of the homolysis fragments to be an i m p o r t a n t means of producing destabilization in the p r o t e i n - b o u n d A d o C b l (2) - and thus activating 2 towards homolysis [58]. F r o m the m o r e recent crystal structures of A d o C b l - d e p e n d e n t enzymes, a distant aden(os)ine-binding pocket is n o w recognized to provide the required structural means for this [58,86,88,185]. Indeed, the c o n t r i b u t i o n of the nucleotide c o o r d i n a t i o n to the ease of homolytic cleavage of A d o C b l (2) was f o u n d to be relatively small: O n the basis of available t h e r m o d y n a m i c d a t a concerning the c o o r d i n a t i o n of the nucleotide in 2 and of the homolysis p r o d u c t cob(II)alamin (6), the Met. Ions Life Sei. 2009, 6, 1-51

32

KRÄUTLER

coordination of the nucleotide was estimated to weaken the (Co-C) bond by only 0.7kcal/mol [22,126]. In contrast, in MeCbl (3) the intramolecular coordination of the nucleotide was determined to increase the homolytic (Co-C)-BDE of 3, by a bout 4kcal/mol according to studies of the methylgroup transfer equilibrium between MeCbl (3)/cob(II)inamide (41) and methylcobinamide (44)/cob(II)alamin (6) [22], see Figure 13. The nucleophile-induced dealkylations of alkyl-Co(III)-corrins is another well known means for cleavage of the (Co-C) bond, in particular for methyl Co(III)-corrins. It has been less studied with MeCbl (3), due to the impediment of the nucleophile-induced demethylation by the intramolecular coordination of the nucleotide base [125,186]. Indeed, thiolates demethylate methylcobinamide (44) to cob(I)inamide (42) approximately 1000 times faster than MeCbl (3) to B 12s (39~) [186], reflecting the strong stabilizing effect of the coordinated nucleotide in 3 [22,126]. This effect is of relevance also for enzymatic methyl group transfer reactions involving protein-bound Co(I)- and methyl-Co(III)-corrins, where considerable axial base effects are seen [187], The electrophile-induced dealkylation of the cobalt-bound methyl group by polarizable metal ions, such as H g 2 + ions, is a crucial path to methylmetal complexes, such as the poisonous Hg-CH 3 ion [26]. Aqua-Co(III)corrins can also demethylate methyl-Co(III)-corrins slowly at room temperature [181].The coordination of the D M B nucleotide modifies the reactivity of the cobalt center by enhancing the ease of abstraction by electrophiles, in both a kinetic and thermodynamic sense [125]. The (Co-C) bond of alkyl-Co(III)-corrins is rather inert against proteolytic cleavage under physiological conditions. The acid-induced heterolytic cleavage of the (Co-C) bond of MeCbl (3) is not documented, the cleavage of coenzyme B 12

M e C b l (3)

Cbi(ll) (41+)

MeCbi (44+)

B 1 2 r (6)

Figure 13. Methyl transfer reaction involving M e C b l (3) and methyl-cob(III)inamide (44 + ) as methyl group donors and B 12r (6) and cob(II)inamide (41 + ) as methyl group acceptors. Met. Ions Life Sei. 2009, 6, 1-51

33

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES

R

R

,n\UH

CS "Co"'

4+7

S

I

II

I I

Nu

H3CH2CO2C R=

,CO 2 CH 2 CH ;'3

CHG

Figure 14. Methylcobalamin (3) as a methylating agent for organic radicals, a hypothetical new mechanism of biological methylation [189].

(2) occurs in acidic aqueous solution at low p H [188], but less readily when compared to 2'-deoxycoenzyme B 12 (48) and 2',3'-di-deoxycoenzyme B 12 (49) [163]. The reactivity difference can be traced back largely to the effect of the ease of protonation of the cobalt-bound organic group [163]. Interestingly, the replacement of the D M B base by an imidazole in Cop-adenosylimidazolyl-cobamide also results in a more readily dealkylated analogue of AdoCbl [188], A recently recognized further mode of cleavage of the (Co-C) bond of organometallic B 1 2 derivatives, is represented by the radical-induced substitution at the cobalt-bound carbon center [22,98,189] (Figure 14). This type of thermodynamically favorable reaction holds strong interest due to the observation of unusual biological (C-C) bond forming reactions and methylations at seemingly inactivated carbon centers [190,191]. In a formally related radical abstraction reaction, the cobalt-bound methyl group of methylcobalamin (3) and other methylcorrinoids is rapidly abstracted by Co(II)-corrinoids, such as cob(II)inamide (41 + ), (giving methylcob(III)inamide, 44 + ) and cob(II)alamin (B 12r , 6) (Figure 14) [22], This type of reaction does not involve free methyl radicals and, under appropriate conditions (aprotic solvents), it is not (even) sensitive to the presence of molecular oxygen [192]. The (Co-C) bond of various organocorrinoids is cleaved homolytically by absorption of visible light and organometallic B 12 derivatives have long been know to be sensitive to visible light [193], which induces cleavage with a Met. Ions Life Sei. 2009, 6, 1-51

34

KRÄUTLER

q u a n t u m yield of a b o u t 0.3 [141,194,195]. Organo-cob(III)amides are also labile to strong one-electron reducing agents, as it has been f o u n d that after one electron reduction of organyl-Co(III)-corrins the ( C o - C ) b o n d is considerably weakened [139,155]. As noted above, this aspect m a y render it difficult to prepare organo-cob(III)amides with electron-withdrawing substituents, via alkylation of the strongly reducing cob(I)amides [173].

5.

ORGANOMETALLIC B 12 DERIVATIVES AS COFACTORS AND INTERMEDIATES IN ENZYMES

" C o m p l e t e " methyl- and adenosyl-corrinoids, such as M e C b l (3) and A d o C b l (2), typically are considered to be the relevant cofactor f o r m s of corrinoids in enzymatic reactions. These organometallic corrinoids are frequently observable in the resting states (and m a y be f o u n d in isolation forms) of functioning enzymes, and m a y then be b o u n d characteristically " b a s e - o n " , " b a s e - o f f / H i s - o n " or " b a s e - o f f to the protein p a r t of the enzyme [18-23]. T h e catalytically equally i m p o r t a n t dealkylated cofactor forms, such as the Co(II)- and Co(I)-corrinoids cob(II)alamin (6) and cob(I)alamin (39~), are less well observable species, and transient in the enzyme reactions, for reasons of their t h e r m o d y n a m i c instability under typical physiological conditions. In most organisms, physiologically inactive f o r m s of the ("complete") corrinoids are taken u p and converted into active organometallic cofactor f o r m s enzymatically. In the h u m a n metabolism, (inactive) vitamin B 1 2 (1) is converted into the adenosyl-corrinoid coenzyme B 1 2 (2) by an ATP-using adenosyl-transferase, or into the methyl-corrinoid M e C b l (3) by methylation with S-adenosyl-methionine ( S A M ) in the methionine synthase complex [17]. However, in the course of the biosynthesis of " c o m p l e t e " corrinoids in microorganisms, such as of the cobalamins, " i n c o m p l e t e " organometallic B 1 2 derivatives already play an i m p o r t a n t role at an early stage as obligate biosynthesis intermediates [196]: Thus, the biosynthetic build-up of the " c o m p l e t e " corrinoids firsts leads to the " i n c o m p l e t e " (organometallic) Cop5'-deoxy-5'-adenosyl-cobinamide (32), to proceed f u r t h e r to the " c o m p l e t e " B 1 2 derivatives by assembly of the nucleotide moiety [196]. Organometallic B 1 2 derivatives are also considered as intermediates in B 1 2 -dependent reductive dehalogenases [97], which play an i m p o r t a n t role in the detoxification of chlorinated c o m p o u n d s [197,198]. Several B 1 2 dependent dehalogenases have been purified with nearly all containing one or m o r e iron-sulfur clusters, in addition to the corrinoid cofactor [97,197]. In the anaerobic bacterium Sulfurospirillum multivorans, which catalyzes the reductive dehalogenation of tetrachloroethene and Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 12 COENZYMES

35

trichloroethene to c«-l,2-dichloroethene [97], a novel corrinoid cofactor was found that had slightly different catalytic properties to other cobamides [66]. This cofactor was isolated as norpseudovitamin B 12 (15, Cop-cyano-7"adeninyl-176-norcobinamide or 176-norpseudovitamin B 12 ), a homologue of pseudovitamin B 12 (16) lacking the methyl group attached to carbon 176 [66], In the B 12 -catalyzed reductive dechlorination of tetrachloroethylene the first step is likely to involve an electron transfer from the fully reduced Co(I)corrin (such as 39") to tetrachloroethylene, leading to a Co(II)corrin (such as 6) and the formation of a trichlorovinyl radical by loss of chloride [199]. Chlorovinylcobalamin (29) and vinylcobalamin (28) were thus synthesized as model compounds [200]. It was shown that chlorinated organometallic derivatives could be possible intermediates in reductive dehalogenation reactions, as 39" promoted reactions can reduce such compounds back to the active form of the catalyst [95,96].

5.1.

Methylcorrinoids in B 12 -Dependent Methyltransferases

The B 12 -dependent methyltransferases play an important role in amino acid metabolism in many organisms (including humans) as well as in one-carbon metabolism and C 0 2 fixation in anaerobic microbes [21]. The reactivity of the "supernucleophilic" Co(I)-corrins and of methyl-Co(III)-corrins make Bi 2 derivatives ideal as cofactors in such enzymatic methyl group transfer reactions [11,12]. B 12 -dependent methionine synthase has been particularly well studied (see e.g. [21,201]) as have methyltransferases in aerobic acetogenesis (see, e.g., [202]), methanogenesis (see, e.g., [203]), and in the anaerobic catabolism of acetic acid to methane and C 0 2 (see, e.g., [204]). Various substrates act as sources of methyl groups, such as methanol, methyl amines, aromatic methyl esters, methylated heavy metals or N 5 methyltetrahydro-pterins (such as N 5 -methyltetrahydromethanopterin or N 5 -methyltetrahydrofolate). For N 5 -methyltetrahydrofolate as a source of the methyl group it has been suggested that the methyl group donor is more likely to be the protonated form of N 5 -methyltetrahydrofolate [21]. Thiols are the methyl group acceptors in methionine synthesis (homocysteine) [21] and methanogenesis (coenzyme M) [23]. In the anaerobic biosynthesis of acetyl-coenzyme A from one-carbon precursors the methyl group acceptor is suggested to be the nickel center attached to the Fe/S cluster [205]. The methyl group transfers catalyzed by methionine synthase (MetH) from E. coli [21] and other Β ^-dependent methyltransferases are all indicated to proceed with an overall retention of configuration (i.e., consistent with two nucleophilic displacement steps, each with inversion of Met. Ions Life Sei. 2009, 6, 1-51

36

KRÄUTLER

configuration) [206,207]. These stereochemical findings exclude free methyl cations or radicals as intermediates, even t h o u g h , in a f o r m a l sense, the methyl transfer reactions catalyzed by B 1 2 enzymes involve (nucleophilicb o u n d ) methyl " c a t i o n s " and heterolytic cleavage/formation of the ( C o C H 3 ) bond. The methyl g r o u p transfer thus, relies on the catalytic properties of enzyme-bound Co(I)corrins and methyl-Co(III)-corrins [22] and is amenable to considerable control f r o m the protein environment [21], due to the great structural changes expected to a c c o m p a n y the transitions f r o m (tetracoordinate) Co(I)corrins to (hexacoordinate) methyl-Co(III)-corrins [11,12]. T h e X-ray crystal analysis of the B 1 2 -binding d o m a i n of M e t H provided the first insight into the three-dimensional structure of a B 1 2 -binding protein [36,116,208,209]. T h e astounding revelation of this work was the finding that the cobalt-coordinating D M B nucleotide tail of the p r o t e i n - b o u n d cofactor M e C b l (3) was displaced by a histidine imidazole and extended into the core of the " R o s s m a n n f o l d " [36,209]. Consequently, in methionine synthase the corrinoid cofactor is b o u n d by histidine ligation to the metal center and in a " b a s e - o f f constitution, i.e., b o u n d in a " b a s e - o f f / H i s - o n " mode. Various other B 1 2 -dependent methyltransferases are indicated to have either a "baseo f f / H i s - o n " b o u n d methyl-Co(III)-corrinoid, or even a corrinoid cofactor in " b a s e - o f f f o r m (where His-coordination is absent) [128] (Figure 15). In a catalytic cycle of B 1 2 -dependent methyltransferases the corrinoid is indicated to cycle between a methyl-Co(III)-corrin and a Co(I)-corrin [21,23]. T h e changing between the hexacoordinate methyl-Co(III) f o r m and (presumably) tetracoordinate Co(I) f o r m is accompanied by constitutional/ c o n f o r m a t i o n a l changes which are highly likely to provide a means for controlling the organometallic reactivity of the b o u n d cofactor [22], subject to H + u p t a k e or H + release (see Figure 15). In response, a H + - m e d i a t e d switch mechanism m a y result, mediated via the " r e g u l a t o r y " His-Asp-Ser triad, which provides the crucial c o n f o r m a t i o n a l alterations associated with the enzyme [21,201,210]. T h e nucleophile-induced methyl g r o u p transfers, involving heterolytic cleavage and f o r m a t i o n of the organometallic ( C o C H 3 ) b o n d at the corrin-bound cobalt center, are expected to be in-line attacks (incoming nucleophile/CH 3 -group/leaving group) and to be subject to strict geometric control: a m a i n role of the His-Asp-Ser-triad appears to be participating in maintaining c o n f o r m a t i o n a l control of the m u t u a l placement of the corrinoid cofactors and the enzyme-bound substrates [210,211], A significant second role of the His-Asp-Ser triad in organometallic reactions is associated with the t h e r m o d y n a m i c effect of the α-axial base c o o r d i n a t i o n on the strength of the ( C o p - C H 3 ) bond. Solution studies showed a significant t h e r m o d y n a m i c trans effect of the D M B c o o r d i n a t i o n in methylcobalamin (3) [11,22,125] and of the imidazole c o o r d i n a t i o n in Cop-CH 3 -imidazolyl-cobamide (27) [94] on heterolytic methyl g r o u p Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 12 COENZYMES

tetrahydrofolate

37

N5-methyl-tetrahydrofolate

Figure 15. Illustration of methionine formation catalyzed by MetH (Enz signifies the MetH-apoenzyme) [21], where the bound corrinoid shuttles between MeCbl (3), in a "base-off/His-on" form, and cob(I)alamin (B12s, 3 9 ) .

transfer reactions. The result showed that the stronger coordinating (nitrogen) ligand stabilizes the methyl-Co(III) f o r m against nucleophilic abstraction of the methyl g r o u p by a b o u t 4 kcal/mol in 3 [126]. This m a y be seen mainly as an "electronic" effect [11,22,125], consistent with the observation of a n o m a l o u s structural trans effects in other methyl-Co(III) complexes [100]. M o r e recent studies with 27 suggested the imidazole base exerts similar "electronic effects" as the D M B base in 3 but 27 is m o r e basic and, therefore, imidazolyl-cobamides (or the " b a s e - o f f / H i s - o n " f o r m of 3) are m o r e readily p r o t o n a t e d near neutral p H [94]. T h e His-Asp-Ser triad m a y then represent a " r e l a y " for H + uptake/release, assumed to f u n c t i o n in the enzymatic methylation/demethylation cycles [212,213]. of

In conclusion, the axial ( C o - N a ) b o n d in the methyl-Co(III) f o r m the p r o t e i n - b o u n d cofactor of M e t H (and other B 1 2 -dependent Met. Ions Life Sei. 2009, 6, 1-51

KRÄUTLER

38

methyltransferases) appears to have three i m p o r t a n t consequences. The weakening of this b o n d activates b o t h (i) the methyl g r o u p for heterolytic abstraction by a nucleophile and (ii) the Co(II) f o r m for reduction to the Co(I) f o r m and (iii) helps to position the methyl-cob(III)amide cofactor for methyl g r o u p transfer [22,201,211].

5.2.

Adenosylcorrinoids in Enzymes Dependent on Coenzyme B 1 2

A b o u t ten coenzyme B 1 2 -dependent enzymes are n o w k n o w n . These enzymes are f o u r c a r b o n skeleton mutases, two diol dehydratases, ethanolamine a m m o n i a lyase, two a m i n o mutases and B 1 2 -dependent ribonucleotide reductase (see [18-20,214—216]). The coenzyme B 1 2 -dependent enzymes are disproportionately distributed in the living world. Only methylmalonylC o A m u t a s e is indispensable in h u m a n metabolism. T h e coenzyme B 1 2 -dependent enzymes p e r f o r m chemical t r a n s f o r m a t i o n s that are difficult to achieve by typical " o r g a n i c reactions" [18]. W i t h the exception of the enzymatic ribonucleotide reduction [215], the results of coenzyme B 1 2 -catalyzed enzymatic reactions correspond to isomerizations with vicinal exchange of a hydrogen a t o m and of a g r o u p with heavy a t o m centers. H o m o l y t i c cleavage of the ( C o - C ) bond of the protein-bound A d o C b l (2) to a 5'-deoxy-5'-adenosyl radical and cob(II)alamin (6) was indicated early to be the entry to Η abstraction reactions induced by the 5'deoxy-5'-adenosyl radical [217]. Therefore, homolysis of the ( C o - C ) bond of 2, which is the thermally most easily achieved reaction of 2 in solution (homolytic C o - C B D E of a b o u t 30 kcal/mol [159,176]) appears to be its biologically most significant reactivity: coenzyme B 1 2 (2) is characterized as a "reversible free radical carrier" [159]) (see Figures 11, 12, and 16). However, the homolysis of the ( C o - C ) b o n d of the p r o t e i n - b o u n d coenzyme needs to be accelerated by a factor of a b o u t 10 12 to agree with the observed rates of reaction of catalysis by the coenzyme B 1 2 -dependent enzymes [159,160]. T h e deduced d r a m a t i c labilization of the b o u n d organometallic cofactor towards homolysis of the ( C o - C ) b o n d is an intriguing feature of the coenzyme B 1 2 -dependent enzymes [159,160,215]. The mechanism of the enzyme- (and substrate-) induced labilization of the ( C o C) b o n d still is a m u c h discussed problem. Covalent restructuring of the b o u n d cofactor (except for the f o r m a t i o n of the " b a s e - o f f / H i s - o n " - f o r m in the c a r b o n skeleton mutases) is n o t indicated [86,216]. In addition, protein and solvent molecules can only weakly stabilize a radical center [218]. Steric distortions of the p r o t e i n - b o u n d cofactor are thus considered as a likely means for the enhanced rate of ( C o - C ) b o n d homolysis [85,86,88,216]. In Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES HO

Η coenzyme B-\2 (2)

39

OH

! Λ - ? cob(ll)alamin (6)

5'-deoxy-5'-adenosyl radical

Figure 16. Coenzyme B 1 2 (AdoCbl, 2), a reversible source of the 5'-deoxy-5'-adenosyl radical and of cob(II)alamin (B 12r , 6).

view of the available crystal structures of cob(II)alamin (6) [58] and of coenzyme B 1 2 -dependent enzymes [36-44], Halpern's earlier suggestion of an "upwards conformational distortion" of the cobalt-corrin part of 2 [159] is not likely to be of relevance. However, labilization may come about largely from a protein- and substrate-induced strain on the organometallic group, separation of the largely nonstrained homolysis fragments and strong binding by the protein of the separated pair, 5'-deoxy-5'-adenosyl radical and 6 (in either a "base-off/His-on" or "base-on" form) [85,86,88,216]. One explanation is the existence in some of these enzymes of a binding interface (e.g., of an "adenosine-binding pocket") which does not allow for unstrained binding of the organometallic moiety [85,86,88,216]. Fixed placement of the corrin moiety at the interfaces of the B 1 2 -binding and substrate-binding/ activating domains appears to be of high significance and movements of the corrin moiety are not required. The "regulatory triads" logically appears not to be involved in proton-transfer steps and may conserve its structure largely during enzymatic turnover. "Electronic effects" of the axial trans ligand on the (Co-C) bond homolysis in 2 and MeCbl (3) are now seen to be of less importance [22]. To conclude, all coenzyme B 1 2 -dependent enzymes appear to rely on the reactivity of bound organic radicals, which are formed (directly or indirectly) by a Η atom abstraction by the 5'-deoxy-5'-adenosyl radical, that originates from the homolysis of the (Co-C) bond of AdoCbl (2). In these enzymatic reactions, the 5'-deoxy-5'-adenosyl radical is the established reactive partner in the actual enzymatic reaction, so that 2 should be looked at as a "precatalyst" (or catalyst precursor) [22]. Coenzyme B 1 2 (2) might then be considered to be a structurally highly sophisticated, reversible source for an Met. Ions Life Sei. 2009, 6, 1-51

KRÄUTLER

40

alkyl radical, whose ( C o - C ) b o n d is labilized in the protein b o u n d state [159], and the m a i n role of the b o u n d cofactor A d o C b l (2) is indeed, the p r o d u c t i o n and controlled presentation of the 5'-deoxy-5'-adenosyl radical f r o m homolysis of its ( C o - C ) bond [159]. The f u n c t i o n of the remaining Co(II)-corrin f r a g m e n t 6 of the coenzyme (as a " s p e c t a t o r " or a "cond u c t o r " ) has recently again become a m a t t e r of discussion [129] and has been re-addressed by calculations [219]. T h e rearrangement steps of B 1 2 -dependent enzymatic rearrangements are n o w assumed to be accomplished by tightly protein-bound radicals that are controlled in their reaction space [18]. Consequently, the m a j o r functions of the enzyme concern n o t only the catalysis of its p r o p e r reactions but also the reversible generation of the radical intermediates and the protection of its proteinic environment f r o m non-specific radical chemistry, called "negative catalysis" [217].

6.

CONCLUDING REMARKS AND FUTURE DIRECTIONS

The discovery of B 1 2 coenzymes by Barker et al. [220] and of heir o r g a n o metallic n a t u r e by Lenhert and H o d g k i n [1], as well as subsequent studies of the organometallic chemistry and biological f u n c t i o n of A d o C b l and M e C b l have helped to open the field of " b i o o r g a n o - m e t a l l i c " chemistry. Clearly, N a t u r e makes use of the capacities of organometallic catalysis in a r e m a r k a b l e way, as is particularly a p p a r e n t , e.g., in alternative pathways of c a r b o n fixation in anaerobes [205,221]. In the Bi 2 -dependent metabolism, the B 1 2 cofactors are b o u n d to proteins and are subjected to the m u t u a l interaction with the proteins. Recently, n a t u r a l B 1 2 -binding nucleotides have also been discovered and suggested to function as "riboswitches", relevant in a new f o r m of controlling gene expression [222]. This finding has begun to open a new area of research in the B 1 2 field [223,224] and to induce f u r t h e r studies and c o m p l e m e n t a r y work with B 1 2 nucleotide conjugates [49,225]. Indeed, the capacity for m u t u a l interaction between the evolved B 1 2 cofactors and functional nucleotides is hardly explored. This subject m a y intensify the search for evidence for a role of corrinoids in an early f o r m of life, such as is represented by the (hypothesis of an) R N A world [226]. Corrinoids clearly are unique c o m p o u n d s extending the capacity of (biological) organometallic catalysis. In another direction, recent studies on the proteins involved in the uptake of B 1 2 derivatives in microorganisms [227], h u m a n s [133], and other Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES

41

mammals are giving a good foundation for investigations on other complex questions, such as: (i) how corrinoids are selectively acquired from the environment, and what forms of mutual dependencies and symbiosis may result from the metabolic need of B 12 -catalyzed organometallic processes (see, e.g., [228]). (ii) how vitamin B 12 derivatives can be used as carriers to shuttle small ligands or larger molecules into cells, simple and higher organisms [74], to diagnose and to influence their metabolism [73]. The growing understanding of the unique reactivity of corrinoids in organometallic processes may also lead to an increasing use of these natural cobalt complexes in the (in vitro and in vivo) analysis of normal and aberrant life processes, such as chemical modifications and damage to D N A [229]. Along the same lines, organometallic processes with Β ] ί derivatives also provide a remarkable potential as the general basis of novel synthetic and analytical developments [137]. Clearly, the "most beautiful" cofactor [230] and its unique organometallic reactivity [22] will continue to fascinate not only the "B 1 2 fraternity", but it will keep a special place in a range of current and future scientific developments.

ACKNOWLEDGMENTS Over the years our work was generously supported by the Austrian National Science Foundation (FWF P13595) and by the European Commission (HPRN-CT-2002-00195).

ABBREVIATIONS Ado AdoCbl BDE Cbi Cbl CE CNCbl CoA DMB

adenosyl 5'-deoxy-5'-adenosylcobalamin, adenosylcobalamin, coenzyme B 12 bond dissociation energy cobinamide cobalamin (DMB-cobamide) calomel electrode cyanocobalamin coenzyme A 5,6-dimethylbenzimidazole Met. Ions Life Sei. 2009, 6, 1-51

42 DMF ESR H2OCbl HOCbl HPLC LC-MS MeCbl MetH NHE NMR NOE Nu SAM SCE UV-Vis

KRÄUTLER dimethylformamide electron spin r e s o n a n c e aquacobalamin, B12a hydroxocobalamin high p e r f o r m a n c e liquid c h r o m a t o g r a p h y liquid c h r o m a t o g r a p h y - m a s s s p e c t r o m e t r y methylcobalamin m e t h i o n i n e synthase n o r m a l h y d r o g e n electrode nuclear m a g n e t i c r e s o n a n c e nuclear O v e r h a u s e r effect nucleotide f u n c t i o n S-adenosylmethionine s a t u r a t e d calomel electrode ultraviolet visible a b s o r b a n c e s p e c t r u m

REFERENCES 1. P. G. Lenhert and D. C. Hodgkin, Nature, 1961, 192, 937. 2. E. L. Rickes, N. G. Brink, F. R. Koniuszy, T. R. Wood and K. Folkers, Science, 1948, 107, 396-397. 3. E. L. Smith and L. F. J. Parker, Biochem. J., 1948, 43, R8-R9. 4. D. C. Hodgkin, J. Pickworth, J. H. Robertson, Κ. N. Trueblood, R. J. Prosen and J. G. White, Nature, 1955, 176, 325-328. 5. J. M. Pratt, Inorganic Chemistry of Vitamin B12, Academic Press, New York, 1972. 6. W. Friedrich, in Fermente, Hormone und Vitamine, Ed. R. Ammon and W. Dirscherl, Georg Thieme Verlag, Stuttgart, 1975, Vol. III/2. 7. B. Zagalak and B. Friedrich, in 3rd European Symposium on Vitamin B12 and Intrinsic Factor, Walter de Gruyter, Berlin, New York, 1979. 8. Bl2, Ed. D. Dolphin, Vol. I, John Wiley & Sons, New York, Chichester, 1982. 9. Vitamin B12 and Β12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-VCH, Weinheim, 1998. 10. Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999. 11. B. Kräutler and S. Ostermann, in The Porphyrin Handbook, Ed. K. M. Kadish, K. M. Smith and R. Guilard, Vol. 11, Elsevier Science, Oxford, 2003, p. 229-276. 12. K. L. Brown, Chem. Rev., 2005, 105, 2075-2149. 13. A. R. Battersby, in Vitamin B12 and B12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-VCH, Weinheim, 1998, p. 47-61. 14. A. I. Scott, C. A. Roessner and P. J. Santander, in The Porphyrin Handbook, Ed. Κ. M. Kadish, Κ. M. Smith and R. Guilard, Vol. 12, Elsevier Science, Oxford, 2003, p. 211-228.

Met. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES

43

15. L. Ellenbogen and B. A. Cooper, in Handbook of Vitamins, Nutritional and Clinical Aspects, Food Science and Technology, Ed. L. J. Machlin, Marcel Dekker, New York, 1984, , p. 491-536. 16. E. Nexo, in Vitamin B12 and B12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-YCH, Weinheim, 1998, p. 461-475. 17. D. Padovani, T. Labunska, B. A. Palfey, D. P. Ballou and R. Banerjee, Nature Chem. Biol., 2008, 4, 194-196. 18. W. Buckel and Β. T. Golding, Ann. Rev. Microbiol., 2006, 60, 27-49. 19. Ε. N. G. Marsh and C. L. Drennan, Curr. Opin. Chem. Biol., 2001, 5, 499-505. 20. P. A. Frey, A. D. Hegeman and G. H. Reed, Chem. Rev., 2006, 106, 3302-3316. 21. R. G. Matthews, Acc. Chem. Res., 2001, 34, 681-689. 22. B. Kräutler, in Vitamin B12 andB12 Proteins, Ed. B. Kräutler, D. Arigoni and B. T. Golding, Wiley-VCH, Weinheim, 1998, p. 3-43. 23. K. Sauer and R. K. Thauer, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999, p. 655-679. 24. B. Kräutler and B. Jaun, in Concepts and Models in Bioinorganic Chemistry, Ed. H.-B. Kraatz and N. Metzler-Nolte, Wiley V C H , Weinheim, 2006. 25. P. A. Butler and B. Kräutler, in Bioorganometallic Chemistry, Ed. G. Simonneaux, Topics of Organometallic Chemistry, Vol. 17, Springer Verlag, Heidelberg, 2006, ρ 1-55. 26. The Biological Alkylation of Heavy Elements, Ed. P. J. Craig and F. Glockling, Royal Soc. Chem., London, 1988. 27. Bioorganometallic Chemistry, Ed. G. Simonneaux, Topics of Organometallic Chemistry, Vol. 17, Springer Verlag, Berlin, 2006. 28. P. A. Frey and A. D. Hegeman, Enzymatic Reaction Mechanisms, Oxford University Press, New York, 2007. 29. W. Friedrich, Vitamins, Walter de Gruyter, Berlin, 1988. 30. Z. Schneider and A. Stroinski, Comprehensive B12, Chemistry, Biochemistry, Nutrition, Ecology, Medicine, Walter de Gruyter, Berlin, New York, 1987. 31. D. C. Hodgkin, J. Kamper, M. Mackay, J. Pickworth, Κ. N. Trueblood and J. G. White, Nature, 1956, 178, 64-66. 32. D. Hodgkin-Crowfoot, Angew. Chem. Int. Ed., 1965, 77, 954-962. 33. Α. Eschenmoser and C. E. Wintner, Science, 1977, 196, 1410-1426. 34. A. Eschenmoser, Nova Acta Leopoldina, 1982, 55, 47. 35. R. B. Woodward, in Vitamin B12, Proceedings of the Third European Symposium on Vitamin B12 and Intrinsic Factor, Ed. B. Zagalak and W. Friedrich, Walter de Gruyter, Berlin, 1979, p. 37. 36. C. L. Drennan, S. Huang, J. T. D r u m m o n d , R. G. Matthews and M. L. Ludwig, Science, 1994, 266, 1669-1674. 37. F. Mancia, Ν. H. Keep, A. Nakagawa, P. F. Leadlay, S. McSweeney, B. Rasmussen, P. Bösecke, Ο. Diat and P. R. Evans, Structure, 1996, 4, 339-350. 38. R. Reitzer, K. Gruber, G. Jogi, U. G. Wagner, H. Bothe, W. Buckel and C. Kratky, Structure, 1999, 7, 891-902. 39. N. Shibata, J. Masuda, T. Tobimatsu, T. Toraya, K. Suto, Y. Morimoto and N. Yasuoka, Structure, 1999, 7, 997-1008.

Met. Ions Life Sei. 2009, 6, 1-51

44

KRÄUTLER

40. Μ. D. Sintchak, G. Arjara, Β. A. Kellogg, J. Stubbe and C. L. Drennan, Nature Struct. Biol., 2002, 9, 293-300. 41. J. Würges, G. Garau, S. Geremia, S. N. Fedosov, Τ. E. Petersen and L. Randaccio, Proc. Natl. Acad. Sei. USA, 2006, 103, 4386-4391. 42. K. P. Locher, A. T. Lee and D. C. Rees, Science, 2002, 296, 1091-1098. F. S. Mathews, Μ. M. Gordon, Z. Chen, K. R. Rajashankar, S. E. Ealick, D. H. 43. Alpers and N. Sukumar, Proc. Natl. Acad. Sei. USA, 2007, 104, 17311-17316. C. B. Bauer, Μ. Y. Fonseca, Η. M. Holden, J. B. Thoden, Τ. B. Thompson, J. C. 44. Escalante-Semerena and I. Rayment, Biochemistry, 2001, 40, 361-374. M. Tollinger, R. Konrat, Β. H. Hilbert, Ε. Ν. G. Marsh and B. Kräutler, 45. Structure, 1998, 6, 1021-1033. J. P. Glusker, in B12, Ed. D. Dolphin, John Wiley & Sons, New York, 1982, 46. Vol. I, p. 23-106. C. Kratky and B. Kräutler, in Chemistry and Biochemistry of B12, Ed. R. 47. Banerjee, 1999, p. 6-41. L. Randaccio, S. Geremia, G. Nardin and J. Würges, Coord. Chem. Rev., 2006, 48. 250, 1332-1350. S. Gschösser, Κ. Gruber, C. Kratky, C. Eichmüller and B. Kräutler, Angew. 49. Chem. Int. Ed., 2005, 44, 2284-2288. R. Keese, L. Werthemann and A. Eschenmoser, see S. Müller, A. Wolleb, 50. L. Walder, R. Keese, Helv. Chim. Acta, 1990, 73, 1659-1668, unpublished. A. Fischli and J. J. Daly, Helv. Chim. Acta, 1980, 63, 1628-1643. 51. K. Kamiya and O. Kennard, J. Chem. Soc. Perkin Trans 1, 1982, 2279-2288. 52. B. Kräutler, C. Caderas, R. Konrat, M. Puchberger and C. Kratky, Helv. Chim. 53. Acta, 1995, 78, 581-599. N. J. Lewis, R. Nussberger, B. Kräutler and A. Eschenmoser, Angew. Chem. Int. 54. Ed., 1983, 22, 736-737. C. Nussbaumer and D. Arigoni, Angew. Chem. Int. Ed., 1983, 22, 737-738. 55. S. Murtaza, P. A. Butler, C. Krakty, K. Gruber and B. Kräutler, Chem. Eur. J., 2008, 14, 7521-7524. 56. B. Kräutler, W. Keller, M. Hughes, C. Caderas and C. Kratky, J. Chem. Soc. Chem. Commun., 1987, 1678-1680. 57. B. Kräutler, W. Keller and C. Kratky, J. Am. Chem. Soc., 1989, 111, 8936-8938. P. Langan, M. Lehmann, C. Wilkinson, G. Jogl and C. Kratky, Acta Crystall. 58. Section D - Biol. Cry St., 1999, 55, 51-59. B. Kräutler, in Vitamin B12 and Β12-Proteins, Ed. B. Kräutler, D. Arigoni and 59. B.T. Golding, Wiley-VCH, Weinheim, 1998, p. 517-521. G. P. Moss, Pure Appl. Chem, 1987, 59, 779-832. 6 0 . B. Kräutler, R. Konrat, E. Stupperich, G. Färber, Κ. Gruber and C. Kratky, Inorg. Chem., 1994, 33, 4128-4139. 6 1 . H. Stoeckli-Evans, E. Edmond and D. C. Hodgkin, J. Chem Soc. Perkin Trans II, 1972, 605-614. 6 2 . K. L. Brown, D. R. Evans, J. D. Zubkowski and E. J. Yalente, Inorg. Chem., 1996, 35, 4 1 5 ^ 2 3 . 63. K. L. Brown, X. Zou, G. Z. Wu, J. D. Zubkowski and E. J. Valente, Polyhedron, 1995, 14, 1621-1639. 64. 65. Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B 1 2 COENZYMES

45

66. B. Kräutler, W. Fieber, S. Ostermann, Μ. Fasching, Κ. Η. Ongania, Κ. Gruber, C. Kratky, C. Mikl, A. Siebert and G. Diekert, Helv. Chim. Acta, 2003, 86, 3698-3716. 67. P. J. Anderson, J. Lango, C. Carkeet, A. Britten, B. Kräutler, B. H a m m o c k and J. R. Roth, J. Bacteriol., 2008, 190, 1160-1171. 68. C. Kratky, G. Färber, K. Gruber, K. Wilson, Z. Dauter, H. F. Nolting, R. K o n r a t and B. Kräutler, J. Am. Chem. Soc., 1995, 117, 4654^1670. 69. L. Randaccio, M. Furlan, S. Geremia and M. Slouf, Inorg. Chem., 1998, 37, 5390-5393. 70. L. Hannibal, C. A. Smith, D. W. Jacobsen and Ν. E. Brasch, Angew. Chem. Int. Ed, 2007, 46, 5140-5143. 71. S. Kunze, T. Zobi, P. Kurz, B. Spingier and R. Alberto, Angew. Chem. Int. Ed., 2004, 43, 5025-5029. 72. C. C. Smeltzer, M. J. Cannon, P. R. Pinson, J. D. J. Munger, F. G. West and C. B. Grissom, Org. Lett., 2001, 3, 799-801. 73. Η. P. C. Hogenkamp, D. A. Collins, C. B. Grissom and F. G. West, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999, p. 385-410. 74. A. K. Petrus, A. R. Vortherms, Τ. J. Fairchild and R. P. Doyle, ChemMedChem., 2007, 2, 1717-1721. 75. E. Jörin, Α. Schweiger and Η. H. Günthard, J. Am. Chem. Soc., 1983, 105, 4277-4286. 76. E. Hohenester, C. Kratky and B. Kräutler, J. Am. Chem. Soc., 1991, 113, 4523-4530. 77. P. G. Lenhert, Proc. Roy. Soc. Series A, 1968, 303, 45-84. 78. H. F. J. Savage, P. F. Lindley, J. L. Finney and P. A. Timmins, Acta Cryst. Sect. Β - Struct. Sei., 1987, 43, 280-295. 79. J. P. Bouquiere, J. L. Finney, M. S. Lehmann, P. F. Lindley and H. F. J. Savage, Acta Cryst. Sect. Β - Struct. Sei., 1993, 49, 79-89. 80. Ν. W. Alcock, R. M. Dixon and Β. T. Golding, J. Chem. Soc., Chem. Commun., 1985, 603-605. 81. K. L. Brown, S. Cheng, X. Zou, J. Li, G. D. Chen, E. J. Yalente, J. D. Zubkowski and Η. M. Marques, Biochemistry, 1998, 37, 9704-9715. 82. G. N. Sando, R. L. Blakley, H. P. C. Hogenkamp and P. J. H o f f m a n n , J. Biol. Chem., 1975, 250, 8774-8779. 83. J. S. Krouwer, B. Holmquist, R. S. Kipnes and Β. M. Babior, Biochim. Biophys. Acta, 1980, 612, 153-159. 84. T. G. Pagano, G. L. Marzilli, Μ. M. Flocco, L. Tsai, H. L. Carrell and J. P. Glusker, J. Am. Chem. Soc., 1991, 113, 531-542. 85. S. Gschösser, R. B. Hannak, R. Konrat, K. Gruber, C. Mikl, C. Kratky and B. Kräutler, Chem. Eur. J., 2005, 11, 81-93. 86. K. Gruber, R. Reitzer and C. Kratky, Angew. Chem. Int. Ed., 2001, 40, 3377-3380. 87. S. Gschösser, R. Hannak, R. Konrat, K. Gruber, C. Mikl, C. Kratky and B. Kräutler, unpublished.

Met. Ions Life Sei. 2009, 6, 1-51

KRÄUTLER

46 88.

89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106.

107. 108.

109. 110. 111.

Μ. Fukuoka, Υ. Nakanishi, R. Β. Hannak, Β. Kräutler and Τ. Toraya, FEBS J., 2005, 272, 4787^796. Μ. Rossi, J. P. Glusker, L. Randaccio, M. F. Summers, P. J. Toscano and L. G. Marzilli, J. Am. Chem. Soc., 1985, 107, 1729-1738. M. Tollinger, R. Konrat and B. Kräutler, Helv. Chim. Acta, 1999, 82, 1596-1609. L. Randaccio, M. Furlan, S. Geremia, M. Slouf, I. Srnova and D. Toffoli, Inorg. Chem., 2000, 39, 3403-3413. K. L. Brown, S. F. Cheng, X. Zou, J. D. Zubkowski, E. J. Yalente, L. Knapton and Η. M. Marques, Inorg. Chem., 1997, 36, 3666-3675. T. Wagner, C. E. Afshar, H. L. Carrell, J. P. Glusker, U. Englert and H. P. C. Hogenkamp, Inorg. Chem., 1999, 38, 1785-1794. M. Fasching, W. Schmidt, B. Kräutler, Ε. Stupperich, A. Schmidt and C. Kratky, Helv. Chim. Acta, 2000, 83, 2295-2316. Κ. M. McCauley, D. A. Pratt, S. R. Wilson, J. Shey, T. J. Burkey and W. A. van der Donk, J. Am. Chem. Soc., 2005, 127, 1126-1136. Κ. M. McCauley, S. R. Wilson and W. A. van der Donk, J. Am. Chem. Soc., 2003, 125, 4410^411. G. Wohlfahrt and G. Diekert, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999, p. 871-893. B. Kräutler, Τ. Derer, P. L. Liu, W. Mühlecker, Μ. Puchberger, C. Kratky and K. Gruber, Angew. Chem. Int. Ed., 1995, 34, 84-86. Κ. F. Purcell and J. C. Kotz, Inorganic Chemistry, Holt-Saunders Intl. Eds., Philadelphia, 1977, p. 694ff. D. J. A. De Ridder, E. Zangrando and H.-B. Bürgi, J. Mol. Struct., 1996, 374, 63-83. V. B. Pett, Μ. N. Liebman, P. Murrayrust, K. Prasad and J. P. Glusker, J. Am. Chem. Soc., 1987, 109, 3207-3215. B. Kräutler, Chimia, 1988, 42, 91-94. K. L. Brown, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999, p. 197-237. B. Kräutler, FEMS Microbiol. Rev., 1990, 87, 349-354. B. Hoffmann, M. Oberhuber, E. Stupperich, H. Bothe, W. Buckel, R. Konrat and B. Kräutler, J. Bacteriol., 2000, 182, 4773^782. R. Konrat, M. Tollinger and B. Kräutler, in Vitamin B12 and B12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-VCH, Weinheim, 1998, p. 349-368. M. F. Summers, L. G. Marzilli and A. Bax, J. Am. Chem. Soc., 1986, 108, 4285-4294. A. Bax, L. G. Marzilli and M. F. Summers, J. Am. Chem. Soc., 1987, 109, 566-574. T. G. Pagano, P. G. Yohannes, B. P. Hay, J. R. Scott, R. G. Finke and L. G. Marzilli, J. Am. Chem. Soc., 1989, 111, 1484-1491. W. Fieber, Β. Hoffmann, W. Schmidt, E. Stupperich, R. Konrat and B. Kräutler, Helv. Chim. Acta, 2002, 85, 927-944. G. Kontaxis, D. Riether, R. Hannak, M. Tollinger and B. Kräutler, Helv. Chim. Acta, 1999, 82, 848-869.

Ions Life Sei. 2009, 6, 1-51

ORGANOMETALLIC CHEMISTRY OF B12 C O E N Z Y M E S

47

112. B. Kräutler and C. Caderas, Helv. Chim. Acta, 1984, 67, 1891-1896. 113. W. Fieber, R. Konrat and B. Kräutler, unpublished results. 114. G. J. Gerfen, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, Chichester, 1999, p. 165-195. 115. E. Stupperich, H. J. Eisinger and S. P. J. Albracht, Eur. J. Biochem., 1990, 193, 105-109. 116. C. L. Drennan, R. G. Matthews and M. L. Ludwig, Curr. Opin. Struct. Biol., 1994, 4, 919-929. 117. S. Van Doorslaer, A. Schweiger and B. Kräutler, J. Phys. Chem. B, 2001, 105, 7554-7563. 118. S. Van Doorslaer, G. Jeschke, B. Epel, D. Goldfarb, R. A. Eichel, Β. Kräutler and Α. Schweiger, J. Am. Chem. Soc., 2003, 125, 5915-5927. 119. Τ. A. Stich, N. R. Buan, J. C. Escalante-Semerena and T. C. Brunold, J. Am. Chem. Soc., 2005, 127, 8710-8719. 120. Τ. A. Stich, M. Yamanishi, R. Banerjee and T. C. Brunold, J. Am. Chem. Soc., 2005, 127, 7660-7661. 121. P. Renz, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999, p. 557-576. 122. P. A. Limbach, P. F. Crain and J. A. Mccloskey, Nucleic Acids Res., 1994, 22, 2183-2196. 123. Μ. Ε. Taga, Ν. A. Larsen, Α. R. Howard-Jones, C. T. Walsh and G. C. Walker, Nature, 2007, 446, 4 4 9 ^ 5 3 . 124. Α. Eschenmoser, Angew. Chem. Int. Ed., 1988, 27, 5-39. 125. B. Kräutler, in The Biological Alkylation of Heavy Elements, Ed. P. J. Craig and F. Glockling, Royal Soc. Chem., London, 1988, p. 31-45. 126. B. Kräutler, Helv. Chim. Acta, 1987, 70, 1268-1278. 127. M. Fasching, H. Perschinka, C. Eichmüller, S. Gschösser and B. Kräutler, Chem. Biodiv., 2005, 2, 178-197. 128. T. Svetlitchnaia, V. Svetlitchnyi, O. Meyer and H. Dobbek, Proc. Nat. Acad. Sei. USA, 2006, 103, 14331-14336. 129. W. Buckel, C. Kratky and Β. T. Golding, Chem. Eur. ./., 2006, 12, 352-362. 130. E. Stupperich, H. J. Eisinger and B. Kräutler, Eur. J. Biochem., 1989,186, 657-661. 131. E. Stupperich, R. Konle and M. Lehle, in Vitamin B12 and B12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley VCH, Weinheim, 1998, p. 179-187. 132. E. Stupperich and E. Nexo, Eur. J. Biochem., 1991, 199, 299-303. 133. S. Fedosov, N. Fedosova, B. Kräutler, E. Nexo and T. Petersen, Biochemistry, 2007, 46, 6446-6458. 134. R. B. Hannak, R. Konrat, W. Schüler, B. Kräutler, Μ. Τ. M. Auditor and D. Hilvert, Angew. Chem. Int. Ed., 2002, 41, 3613-3616. 135. A. Nahvi, N. Sudarsan, M. S. Ebert, X. Zou, K. L. Brown and R. R. Breaker, Chem. Biol., 2002, 9, 1043-1049. 136. S. W. Ragsdale, Crit. Rev. Biochem. Mol. Biol., 1991, 26, 261-300. 137. B. Steiger, A. Ruhe and L. Walder, Anal. Chem., 1990, 62, 759-766. 138. B. Kräutler, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley, New York, 1999, p. 315-339.

Met. Ions Life Sei. 2009, 6, 1-51

48

KRÄUTLER

139. D. Lexa and J. M. Saveant, Acc. Chem. Res., 1983, 16, 235-243. 140. G. N. Schrauzer, E. Deutsch and R. J. Windgassen, J. Am. Chem. Soc., 1968, 90, 2441-2442. 141. J. F. Endicott and T. L. Netzel, J. Am. Chem. Soc., 1979, 101, 4 0 0 0 ^ 0 0 2 . 142. D. Lexa, J. M. Saveant and J. Zickler, J. Am. Chem. Soc., 1977, 99, 2786-2790. 143. N. R. de Tacconi, D. Lexa and J. M. Saveant, J. Am. Chem. Soc., 1979, 101, 467-473. 144. Κ. A. Rubinson, Η. Y. Parekh, E. Itabashi and Η. B. Mark, Inorg. Chem., 1983, 22, 4 5 8 ^ 6 3 . 145. S. M. Chemaly and J. M. Pratt, ./. Chem. Soc., Dalton Trans., 1984, 595-599. 146. D. Lexa and J. M. Saveant, J. Chem. Soc. Chem. Commun., 1975, 872-874. 147. D. Lexa, J. M. Saveant and J. Zickler, J. Am. Chem. Soc., 1980,102, 2654-2663. 148. D. Lexa, J. M. Saveant and J. Zickler, J. Am. Chem. Soc., 1980,102, 4851-4852. 149. D. Lexa and J. M. Saveant, J. Am. Chem. Soc., 1976, 98, 2652-2658. 150. D. L. Zhou, Ο. Tinembart, R. Scheffold and L. Walder, Helv. Chim. Acta, 1990, 73, 2225-2241. 151. D. Faure, D. Lexa and J. M. Saveant, J. Electroanal. Chem., 1982,140, 269-284. 152. D. Faure, D. Lexa and J. M. Saveant, J. Electroanal. Chem., 1982,140, 285-295. 153. D. Lexa and J. M. Saveant, J. Am. Chem. Soc., 1978, 100, 3220-3222. 154. R. L. Birke, Q. D. Huang, Τ. Spataru and D. K. Gosser, J. Am. Chem. Soc., 2006, 128, 1922-1936. 155. B. D. Martin and R. G. Finke, J. Am. Chem. Soc., 1992, 114, 585-592. H. P. C. Hogenkamp, in B12, Ed. D. Dolphin, Vol. I, John Wiley & Sons, 156. New York, 1982, p. 295. J. Halpern, in B12, Ed. D. Dolphin, Vol. I, John Wiley & Sons, New York, 1982, 157. p. 501-541. Β. T. Golding, in B12, Ed. D. Dolphin, Vol. I, John Wiley & Sons, New York, 158. 1982, p. 543-582. J. Halpern, Science, 1985, 227, 869-875. 159. R. G. Finke, in Vitamin B12 and B12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-VCH, Weinheim, 1998, p. 383-402. 160. R. G. Matthews, R. V. Banerjee and S. W. Ragsdale, BioFactors, 1990, 2, 147-152. 161. R. G. Matthews, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999, p. 681-706. 162. B. Kräutler, in Organic Reactivity, Physical and Biological Aspects, Ed. Β. T. Golding, R. J. Griffin and H. Maskill, Royal Soc. Chem., London, 1995, p. 209. 163. G. N. Schrauzer and E. Deutsch, J. Am. Chem. Soc., 1969, 91, 3341-3350. P. A. Milton and T. L. Brown, J. Am. Chem. Soc., 1977, 99, 1390-1396. 164. 165. R. B. H a n n a k and B. Kräutler, unpublished, see R. B. Hannak, Ph.D. thesis, University of Innsbruck, 1996. 166. R. B. Hannak, G. Färber, R. K o n r a t and B. Kräutler, J. Am. Chem. Soc., 1997, 119, 2313-2314. 167. C. Fred, J. Haglund, T. Alsberg, P. Rydberg, J. Minten and M. Tornqvist, J. Separation Sei., 2004, 27, 607-612. 168.

Ions Life Sei. 2009, 6, 1-51

O R G A N O M E T A L L I C C H E M I S T R Y OF B 1 2 C O E N Z Y M E S

49

169. J. Haglund, A.-L. Magnusson, L. Ehrenberg and M. Törnqvist, Toxicol. Environ. Chem., 2003, 85, 81-94. 170. D. L. Zhou, P. Walder, R. Scheffold and L. Walder, Helv. Chim. Acta, 1992, 75, 995-1011. 171. W. P. Watson, T. Munter and Β. T. Golding, Chem. Res. Toxicol., 2004, 17, 1562-1567. 172. M. Puchberger, R. Konrat, B. Kräutler, U. Wagner and C. Kratky, Helv. Chim. Acta, 2003, 86, 1453-1466. 173. O. Tinembart, L. Walder and R. Scheffold, Ber. Bunsen-Ges. Phys. Chem., 1988, 92, 1225-1231. 174. M. Tollinger, Τ. Derer, R. Konrat and Β. Kräutler, J. Mol. Catal., 1997, 116, 147-155. 175. Κ. L. Brown, L. Zhou, D. Zhao, S. Cheng and Z. Xiang, in Vitamin B12 andB12Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-YCH, Weinheim, 1998, p. 417^32. 176. B. P. Hay and R. G. Finke, J. Am. Chem. Soc., 1986, 108, 4820^829. 177. B. Kräutler, Helv. Chim. Acta, 1984, 67, 1053-1059. 178. S. W. Ragsdale and M. Kumar, Chem. Rev., 1996, 96, 2515-2539. 179. Η. Fischer, J. Am. Chem. Soc., 1986, 108, 3925-3927. 180. K . L. Brown, in B12, Ed. D. Dolphin, Vol. I, John Wiley & Sons, New York, 1982, p. 245-294. 181. Y-T. Fanchiang, G. T. Bratt and H. P. C. Hogenkamp, Proc. Natl. Acad. Sei. USA, 1984, 81, 2698-2702. 182. R. G. Finke and B. P. Hay, Inorg. Chem., 1984, 23, 3041-3043. 183. S. M. Chemaly and J. M. Pratt, J. Chem. Soc., Dalton Trans., 1980, 2259-2266. 184. J. H. Grate and G. N. Schrauzer, J. Am. Chem. Soc., 1979, 101, 4601-4611. 185. T. Toraya, Cell. Mol. Life Sei., 2000, 57, 106-127. 186. H. P. C. Hogenkamp, G. T. Bratt and S. Sun, Biochemistry, 1985, 24, 6428-6432. 187. J. S. Dorweiler, R. G. Finke and R. G. Matthews, Biochemistry, 2003, 42, 14653-14662. 188. Κ. L. Brown, X. Zou, R. R. Banka, C. B. Perry and Η. M. Marques, Inorg. Chem., 2004, 43, 8130-8142. 189. H. Mosimann and B. Kräutler, Angew. Chem. Int. Ed., 2000, 39, 393-200. 190. P. K. Galliker, O. Gräther, M. Rümmler, W. Fitz and D. Arigoni, in Vitamin B12 and B12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, WileyVCH, Weinheim, 1998, p. 447-458. 191. R. D. Woodyer, G. Li, H. Zhao and W. A. van der Donk, Chem. Commun., 2007, 359-361. 192. B. Kräutler, M. Hughes and C. Caderas, Helv. Chim. Acta, 1986, 69,1571-1575. 193. W. H. Pailes and H. P. C. Hogenkamp, Biochemistry, 1968, 7, 4160-4166. 194. A. G. Cole, L. Μ. Yoder, J. J. Shiang, N. A. Anderson, L. A. I. Walker, Μ. M. B. Holl and R. J. Sension, J. Am. Chem. Soc., 2002, 124, 434-441. 195. L. A. I. Walker, J. J. Shiang, N. A. Anderson, S. H. Pullen and R. J. Sension, J. Am. Chem. Soc., 1998, 120, 7286-7292.

Met. Ions Life Sei. 2009, 6, 1-51

50

KRÄUTLER

196. Μ. J. Warren, Ε. Raux, Η. L. Schubert and J. C. Escalante-Semerena, Nat. Prod. Rep., 2002, 19, 390-412. 197. C. Holliger, G. Wohlfarth and G. Diekert, FEMS Microbiol. Rev., 1999, 22, 383-398. 198. B. L. Sun, Β. M. Griffin, H. L. Ayala-del-Rio, S. A. Hashsham and J. M. Tiedje, Science, 2002, 298, 1023-1025. 199. J. Shey and W. A. van der Donk, J. Am. Chem. Soc., 2000, 122, 12403-12404. Κ. M. McCauley, S. R. Wilson and W. A. van der Donk, Inorg. Chem., 2002, 200. 41, 3 9 3 ^ 0 4 . M. L. Ludwig and R. G. Matthews, in ACS Symposium Series, Structures and 201. Mechanisms. From Ashes to Enzymes, Vol. 827, 2002, p. 186-201. S. W. Ragsdale, M. Kumar, S. Zhao, S. Menon, J. Seravalli and T. Doukov, in 202. Vitamin B12 and Β12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-YCH, Weinheim, 1998, p. 167-178. 203. R. K. Thauer, Microbiology, 1998, 144, 2377-2406. 204. J. G. Ferry, Annu. Rev. Microbiol., 1995, 49, 305-333. 205. S. W. Ragsdale, Chem. Rev., 2006, 106, 3317-3337. 2 0 6 . Τ. M. Zydowsky, L. F. Courtney, V. Frasca, K. Kobayashi, H. Shimizu, L. D. Yuen, R. G. Matthews, S. J. Benkovic and H. G. Floss, J. Am. Chem. Soc., 1986, 108, 3152-3153. 207. L. D. Zydowsky, Τ. M. Zydowsky, E. S. Haas, J. W. Brown, J. N. Reeve and H. G. Floss, J. Am. Chem. Soc., 1987, 109, 7922-7923. 208. C. L. Drennan, Μ. M. Dixon, D. M. Hoover, J. T. Jarrett, C. W. Goulding, R. G. Matthews and M. L. Ludwig, in Vitamin B12 and B12-Proteins, Ed. B. Kräutler, D. Arigoni and Β. T. Golding, Wiley-VCH, Weinheim, 1998, p. 133-155. 209. M. L. Ludwig, C. L. Drennan and R. G. Matthews, Structure, 1996, 4, 506-512. V. Bandarian, M. L. Ludwig and R. G. Matthews, Proc. Nat. Acad. Sei. USA, 210. 2003, 100, 8156-8163. V. Bandarian, K. A. Pattridge, B. W. Lennon, D. P. Huddler, R. G. Matthews 211. and M. L. Ludwig, Nat. Struct. Biol., 2002, 9, 53-56. B. Kräutler and C. Kratky, Angew. Chem. Int. Ed., 1996, 35, 167-170. 212. Y. Bandarian and R. G. Matthews, Biochemistry, 2001, 40, 5056-5064. 213. R. Banerjee, Chem. Rev., 2003, 103, 2083-2094. 214. J. Stubbe, D. G. Nocera, C. S. Yee and M. C. Y. Chang, Chem. Rev., 2003,103, 215. 2167-2201. 2 1 6 . T. Toraya, Chem. Rev., 2003, 103, 2095-2127. 217. J. Retey, Angew. Chem. Int. Ed., 1990, 29, 355-361. 218. D. Griller and D. D. M. Wayner, Pure Appl. Chem., 1989, 61, 717-724. 219. T. Kamachi, T. Toraya and K. Yoshizawa, Chem. Eur. J., 2007, 13, 7864-7873. 220. H. A. Barker, H. Weissbach and R. D. Smyth, Proc. Nat. Acad. Sei. USA, 1958, 44, 1093-1097. 2 2 1 . R. K. Thauer, Science, 2007, 318, 1732-1733. 2 2 2 . W. C. Winkler and R. R. Breaker, Ann. Rev. Microbiol., 2005, 59, 487-517. 223. S. Gallo, R. K. O. Sigel, M. Oberhuber and B. Kräutler, Chimia, 2007, 61, 457. 224. S. Gallo, Μ. Oberhuber, R. K. O. Sigel and B. Kräutler, ChemBioChem., 2008, 9, 1408-1414.

Ions Life Sei. 2009, 6, 1-51

O R G A N O M E T A L L I C C H E M I S T R Y OF B 1 2 C O E N Z Y M E S

51

225. S. Gschösser and B. Kräutler, Chem. Eur. J., 2008, 14, 3605-3619. 226. R. R. Breaker, in The RNA World, Ed. R. F. Gesteland, Τ. R. Cech and J. F. Atkins, , Laboratory Press, Cold Spring Harbor, 2006, p. 89-107. 227. C. Bradbeer, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999, p. 489-506. 228. Μ. T. Croft, A. D. Lawrence, E. Raux-Deery, M. J. Warren and A. G. Smith, Nature, 2005, 438, 90-93. 229. J. Haglund, A. Rafiq, L. Ehrenberg, Β. Τ. Golding and Μ. Tornqvist, Chem. Res. Toxicol., 2000, 13, 253-256. 230. J. Stubbe, Science, 1994, 266, 1663-1664.

Met. Ions Life Sei. 2009, 6, 1-51

Met. Ions Life Sei. 2009, 6, 53-114

2 Cobalamin- and Corrinoid-Dependent Enzymes Rowena G.

Matthews

Department of Biological Chemistry and Life Sciences Institute, University of Michigan, Ann Arbor MI 48109-2216, USA

ABSTRACT 1. INTRODUCTION. WHAT IS A CORRINOID? 2. CORRINOID-DEPENDENT METHYLTRANSFERASES 2.1. Overview of the Metabolic Roles of Corrinoid-Dependent Methyltransferases 2.2. Cobalamin-Dependent Methionine Synthase 2.3. Corrinoid-Dependent Methyltransferases in Methanosarcina spp. 2.4. Membrane-Associated Energy-Conserving Corrinoid Methyltransferase 2.5. The Corrinoid Iron-Sulfur Protein 2.6. Aromatic O-Demethylases 2.7. Reductive Dehalogenases 2.8. Modes of Activation of Corrinoid-Dependent Methyltransferases 2.9. Methyl Transfer in Fosfomycin Biosynthesis 3. ADENOSYLCOBALAMIN-DEPENDENT REARRANGEMENTS AND ELIMINATIONS 3.1. Enzymes Catalyzing Carbon Skeleton Rearrangements 3.1.1. Glutamate Mutase 3.1.2. Methyleneglutarate Mutase 3.1.3. Methylmalonyl-Coenzyme A Mutase

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00053

54 54 56 56 60 67 72 72 76 77 79 82 84 86 86 89 90

54

MATTHEWS

3.1.4. Isobutyryl-Coenzyme A Mutase 3.1.5. MeaA - A Mutase of Unknown Function 3.2. Aminomutases and Diol Dehydrases. Isomerization and Elimination 3.2.1. Diol Dehydrase 3.2.2. Ethanolamine Ammonia Lyase 3.2.3. Lysine 5,6-Aminomutase 3.2.4. Ornithine 4,5-Aminomutase 3.3. Adenosylcobalamin-Dependent Ribonucleotide Triphosphate Reductase 4. C O N C L U D I N G R E M A R K S ACKNOWLEDGMENTS ABBREVIATIONS A N D DEFINITIONS REFERENCES

94 94 95 95 100 101 103 103 106 107 107 107

ABSTRACT: This chapter reviews the literature on cobalamin- and corrinoid-containing enzymes. These enzymes fall into two broad classes, those using methylcobalamin or related methylcorrinoids as prosthetic groups and catalyzing methyl transfer reactions, and those using adenosylcobalamin as the prosthetic group and catalyzing the generation of substrate radicals that in turn undergo rearrangements and/or eliminations. KEYWORDS: adenosylcobalamin · methylcobalamin · methyltransferase

1.

INTRODUCTION. WHAT IS A CORRINOID?

The structure of cobalamin, or dimethylbenzimidazolylcobamide, is shown in Figure 1. In cob(III)alamin derivatives like methyl- or adenosylcobalamin the cobalt is in the + 3 oxidation state and is typically six-coordinate. Four nitrogens from the corrin macrocycle serve as the equatorial ligands, and a substituent of the corrin ring known as the nucleotide loop, which terminates in a dimethylbenzimidazole base, provides the lower axial or α ligand to the cobalt. Cobamides in which the benzimidazole moiety is coordinated to the cobalt are referred to as "base-on" cobamides. The upper axial or β ligand, shown as R in Figure 1, is a methyl group in methylcobalamin, an adenosyl group in adenosylcobalamin (AdoCbl), or may be occupied by an exchangeable ligand such as water in aquacobalamin. In the domains Archaea and Prokarya, cobalamin is only one of many variants grouped under the name corrinoids. Some of these variants involve changes in the structure of the dimethybenzimidazole (DMB) nucleotide base, such as 5'-methoxybenzimidazole cobamide, while other variants involve replacement of the D M B base by compounds such as adenine (pseudovitamin Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

55

Figure 1. The structure of cobalamin. R is a methyl group in methylcobalamin or an adenosyl group in adenosylcobalamin.

B 12 ) or ^-cresol (^-cresolylcobamide). The latter two bases cannot be coordinated to the cobalt of the free cobamide, which instead contains a water in the lower axial position and is referred to as a base-off corrinoid. Corrinoid-dependent methyltransferases are found in all three kingdoms of life, and in all cases, the cofactor is bound to its enzyme in a base-off manner. It is probably for this reason that so much variation in the nucleotide loop is tolerated. In a subset of corrinoid-dependent methyltransferases, the corrinoid is bound with a histidine (His) from the protein as Met. Ions Life Sei. 2009, 6, 53-114

MATTHEWS

56

the lower axial ligand, and this m o d e of binding is referred to as base-off, His-on. T h e first observation of base-off binding was m a d e by Ragsdale and his colleagues [1], w h o characterized the corrinoid in the corrinoid ironsulfur protein f r o m Moorella thermoaceticum by electron paragmagnetic resonance ( E P R ) and M ö s s b a u e r spectroscopy. Extracts of the bacterium Sporomusa ovata were subsequently shown to contain the base-off corrinoid /?-cresolyl-cob(II)amide in the Stupperich l a b o r a t o r y [2]. W h e n a protein containing the b o u n d corrinoid was isolated, the corrinoid was f o u n d to exhibit the E P R properties of a base-on corrinoid in the + 2 oxidation state. W h e n the bacterial cells were grown on [ 1 5 N]-His, the E P R spectrum was altered, indicating that the axial nitrogen ligand of the corrinoid was derived f r o m the imidazole g r o u p of His. However, after release f r o m the protein, the corrinoid remained in the base-off f o r m . A d o C b l - d e p e n d e n t enzymes are only f o u n d in the d o m a i n s E u k a r y o t a and P r o k a r y o t a , and cobalamin is the only corrinoid to be f o u n d in these enzymes. As will be discussed f u r t h e r in the second m a j o r section of this chapter, A d o C b l - d e p e n d e n t enzymes m a y contain the cobalamin b o u n d either in the base-on f o r m with the D M B ligand coordinated to the cobalt, or in the base-off,His-on f o r m . As the cobalt in corrinoids is reduced, the preferred c o o r d i n a t i o n n u m b e r decreases. Corrinoids in the + 2 oxidation state are preferentially fivecoordinate, with only one axial ligand, while corrinoids in the + 1 oxidation state are preferentially four-coordinate, with n o axial ligands.

2. 2.1.

CORRINOID-DEPENDENT METHYLTRANSFERASES Overview of the Metabolic Roles of CorrinoidDependent Methyltransferases

In h u m a n s , only one corrinoid-dependent methyltransferase, methionine synthase, is f o u n d , and this appears to be the only such corrinoid-dependent methyltransferase to be f o u n d in the d o m a i n E u k a r y a . However, in the d o m a i n s P r o k a r y a and Archaea, a wide variety of corrinoid-dependent methyltransferases play central roles in metabolism, particularly in organisms t h a t grow u n d e r anaerobic conditions. W e will begin by briefly enumerating these methyltransferases and their roles in c a r b o n assimilation and energy generation. M a n y organisms belonging to these two d o m a i n s m a k e use of the reactions in the W o o d - L j u n g d a h l p a t h w a y (Figure 2), first elucidated by the studies of H a r l a n d W o o d and Lars L j u n g d a h l and their groups (recently reviewed in [3]). The enzymes in the Eastern b r a n c h of the W o o d - L j u n g d a h l p a t h w a y catalyze the reduction of C 0 2 to f o r m methyl groups t h a t are Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

57

"Western branch" N*

CO?

Η,Λ-tt-JL 00 M i-iong·»« dpf*»*— CO CHfC-Ni Q aq/tcaisyr^amj

cöwiiyme Μ

Figure 2. The role of corrinoid methyltransferases in the central metabolic pathways of anaerobic eubacteria and methanogens. The corrinoid methyltransferases are labeled in red. Red arrows indicate pathways, not shown in detail, that also involve corrinoid proteins. The boxed reactions are not part of the Wood-Ljungdahl pathway but are unique to methanogens. R ' is tetrahydrofolate or a tetrahydropterin analogue of tetrahydrofolate. The charge on the nickel that serves as the methyl acceptor on acyl-CoA synthase/decarbonylase is shown as + 1, but whether it is Ni 1 + or Ni° remains a matter for debate [175,176].

intially bound to tetrahydrofolate or tetrahydropterin analogues of tetrahydrofolate. The reducing power that is needed is provided by three hydride ion transfers. In organisms that can grow on C 0 2 and hydrogen, hydrogenases catalyze the reversible conversion of hydrogen gas to hydride and a proton. The Western branch of the Wood-Ljungdahl pathway involves the reduction of C 0 2 to CO, catalyzed by CO dehydrogenase, and the incorporation of CO into a methyl-nickel bond to form an acetyl-Ni at the Ni-Ni metal center of acyl-CoA synthase. The acetyl group generated by carbonylation can then be transferred to coenzyme A (CoA) to form acetyl-CoA. While none of the reactions of the Eastern or Western branches of the Wood-Ljungdahl pathway involve corrinoids, the corrinoid iron-sulfur protein (highlighted in red in Figure 2) plays a central role in transferring the methyl group generated in the Eastern branch to the nickel in acyl-CoA synthase. All of the reactions in the Wood-Ljungdahl pathway are Met. Ions Life Sei. 2009, 6, 53-114

MATTHEWS

58

reversible, and some organisms will r u n portions of the W o o d - L j u n g d a h l p a t h w a y in reverse, as we shall see. A s u b g r o u p of organisms in the k i n g d o m A r c h a e a are obligate anaerobes that derive their c a r b o n f r o m C 0 2 when grown in the presence of hydrogen and p r o d u c e m e t h a n e as the final p r o d u c t . These organisms are k n o w n as methanogens. M e t h a n o g e n s convert a p o r t i o n of the methyl groups generated in the Eastern b r a n c h of the W o o d - L j u n g d a h l p a t h w a y into methane. This reaction occurs in two irreversible steps, in which the methyl g r o u p is first transferred to coenzyme Μ (ethanethiolsulfonate) by coenzyme Μ m e t h y t r a n s f e r a s e s , and then reduced to m e t h a n e by coenzyme Μ reductase. These two steps are energy generating, and are coupled to the creation of an ion gradient across the cell m e m b r a n e that is used to generate energy for cellular growth. The energy-conserving coenzyme Μ methyltransferase in the m e t h a n o g e n Methanobacterium thermoautotrophicum is a complex containing a corrinoid protein. The complex catalyzes the transfer of a methyl g r o u p f r o m m e t h y l t e t r a h y d r o m e t h a n o p t e r i n , a methyltetrahydrofolate analogue f o u n d in this organism, to the cobalt of the corrinoid protein, and thence to the sulfur of coenzyme Μ to f o r m methylcoenzyme M . T h e basic p a t t e r n for corrinoid-dependent methyltransferases is shown in Figure 3. These enzymes comprise a m i n i m u m of three modules, a central m o d u l e that binds the corrinoid and modules that present the methyl d o n o r to the corrinoid in the cobalt(I) oxidation state, and activate the d o n o r if necessary, and t h a t present the methyl acceptor to the methylcorrinoid and activate it if necessary. These three modules may reside on three separate proteins, or they m a y be present as modules on a single polypeptide or on several polypeptides. A c o m m o n feature of these methyltransferases is that they must stabilize b o t h the methylcorrinoid and the corrinoid in the cobalt(I) oxidation state. F u r t h e r m o r e , they must be capable of undergoing

X-CH3

Co+1

Y-CH3

corrinoid binding module

X

CHg-CO + 3

X - C H 3 = methanol methylamines methylthiols methyltetrahydrofolate aromatic O-methyl ethers

Figure 3.

Y Y = coenzyme Μ Ni +1 in a c y l C o A s y n t h a s e tetrahydrofolate homocysteine

Basic pattern for corrinoid methyltransferases.

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

59

conformational changes that allow the corrinoid prosthetic group access to both donor and acceptor modules. Some methanogens can also grow on acetate by converting it to acetylCoA and then reversing the acyl-CoA synthase reaction to decarbonylate the acetyl group and produce CO and the methylated form of the corrinoid ironsulfur protein. The CO is oxidized to C 0 2 with the generation of a hydride ion. The reversal of the methyl transfers catalyzed by the corrinoid ironsulfur protein complex will then produce a methyltetrahydropterin, which will be converted to methane. The hydride needed for the conversion of methylcoenzyme Μ to methane is generated by the oxidation of CO. Members of the genus Methanosarcina can also use other simple onecarbon compounds as sources of carbon and energy, including methylamines, methylthiols, and methanol. The methyl groups of these compounds are transferred to coenzyme Μ by specific non-energy conserving corrinoid methyl transferases that follow the basic pattern shown in Figure 3. Methanogens growing on substrates other than acetate synthesize acetylCoA from methyltetrahydropterins and C 0 2 by reversal of the corrinoid iron-sulfur complex methyl transfers to generate methyl-Ni on acyl-CoA synthase, followed by the acyl-CoA synthase reaction and oxidation of the resultant CO. From the acetyl-CoA thus formed, all other carbon-containing cellular components are generated. Acetogenic bacteria do not generate energy by methanogenesis, but rather generate energy by the anaerobic fermentation of glucose or by anaerobic growth on hydrogen and C 0 2 . Glucose is converted to two molecules of pyruvate, which in turn is decarboxylated to form two molecules of acetylCoA in a reaction coupled to the generation of ATP from A D P and inorganic phosphate. They use the Eastern branch of the Wood-Ljungdahl pathway to reduce C 0 2 to a methyl group and the Western branch of the pathway to produce CO. These two reagents are then coupled to form an additional molecule of acetyl-CoA, from which all other carbon-containing compounds are generated. A variant use of the Wood-Ljungdahl pathway is made by hydrogenogenic bacteria such as Carboxydothermus hydrogenoformans, which can grow on CO as the sole source of energy and carbon under anaerobic conditions. Some of the CO is oxidized to C 0 2 by the action of CO dehydrogenase, with protons serving as the terminal electron acceptors. The Eastern branch of the Wood-Ljungdahl pathway is used for production of methyl groups, using hydride equivalents generated by oxidation of CO to C 0 2 . The remainder of the CO is converted to acetyl-CoA by the action of acyl-CoA synthase and the corrinoid iron-sulfur protein. Another variant use of the Wood-Ljungdahl pathway is provided by bacteria that use aromatic O-methyl ethers as the source of both carbon and energy such as Sporomusa ovata. Corrinoid aromatic O-demethylases Met. Ions Life Sei. 2009, 6, 53-114

MATTHEWS

60

catalyze transfer of the methyl g r o u p to tetrahydrofolate. The methyl g r o u p can then be oxidized to f o r m a t e a n d / o r C 0 2 by reversal of the Eastern branch of the p a t h w a y with the accompanying generation of reducing equivalents, and then converted to acetate using corrinoid iron-sulfur protein and acyl-CoA synthase. Finally corrinoid-dependent reductive dehalogenases f o u n d in P r o k a r y o t a use the corrinoid protein to catalyze the anaerobic dehalogenation of a variety of halogenated alkyl and aryl c o m p o u n d s . Since the p r o d u c t s of such dehalogenations will vary with the halogenated substrate, the reductive dehalogenases are n o t shown on the central metabolic scheme in Figure 2.

2.2.

Cobalamin-Dependent Methionine Synthase

C o b a l a m i n - d e p e n d e n t methionine synthase ( M e t H ) catalyzes the reaction shown in e q u a t i o n (1).

Methyltetrahydrofolate

L-Homocysteine

Tetrahydrofolate

L-Methionine

(1)

The enzyme is f o u n d in m a n y members of the k i n g d o m P r o k a r y o t a , including Escherichia coli, but has not been f o u n d in the Archaea. It is one of only two B 1 2 -dependent enzymes f o u n d in h u m a n s and other m a m m a l s , and is widely distributed a m o n g the animal E u k a r y o t a . D u e to its overexpression in r e c o m b i n a n t f o r m [4,5] and the resultant ease of purification under aerobic conditions, large a m o u n t s of purified E. coli protein have been available for biochemical and structural characterization. It was one of the first corrinoid proteins to be characterized, and has subsequently been extensively studied in the laboratories of H e r b e r t Weissbach, F r a n k H u e n nekens, and m o r e recently in m y own laboratory. The enzyme consists of four modules, that are arranged linearly with single interdomain linkers to form a single 136 k D a polypeptide. The N-terminal module, the methyl donor module in the parlance used in Figure 3, binds and activates methyltetrahydrofolate and presents it to the cobalamin prosthetic group for methyl transfer. The next module in the sequence is the methyl acceptor module, this module binds and activates homocysteine and presents it to methylcobalamin to allow methyl transfer to form methionine. The third module is the cobalamin-binding module and also contains a four helix bundle Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT E N Z Y M E S

61

at the N-terminus of the domain. The final module binds adenosylmethionine (AdoMet) and is required for reductive activation of the protein. Its raison d'etre requires discussion of the reactions catalyzed by the enzyme, which are shown in Figure 4. M e t H is active during aerobic growth in E. coli, and also under in vitro turnover in microaerophilic conditions. In vitro, the cob(I)alamin form of the enzyme is oxidized to the inactive species cob(II)alamin about once in every 2000 turnovers [6]. Return of the prosthetic group to the active methylcobalamin form requires a reductive remethylation. In E. coli, the electron is provided by the electron transfer protein flavodoxin, the fldA gene product [7,8]. The reduction potential of the flavodoxin semiquinone/hydroquinone is —440 mV vs. the standard hydrogen electrode [9], and the quinone/ semiquinone potential, which is probably the more relevant one for cells grown under microaerophilic conditions, is —260 mV. In contrast, the cob(II)alamin/

Hey

Met

Ν Deactivation/activation Figure 4. Reactions catalyzed by methionine synthase. During primary turnover, the enzyme-bound cobalamin cycles between methylcobalamin and cob(I)alamin forms as the prosthetic group is alternately methylated by methyltetrahydrofolate (CH 3 -H 4 folate) and demethylated by transfer of the methyl group to homocysteine (Hey). During turnover under microaerophilic conditions, the cob(I)alamin form of the enzyme is oxidized to cob(II)alamin about once in every 2000 turnovers. This form of the enzyme is inactive, and reactivation requires a reductive methylation in which the reduction of cob(II)alamin to cob(I)alamin is coupled to a highly exergonic methylation using adenosylmethionine (AdoMet) as the methyl donor. Met. Ions Life Sei. 2009, 6, 53-114

62

MATTHEWS

cob(I)alamin reduction potential is —490 mV at p H 7 [10]. Thus the reduction of cob(II)alamin is a highly endergonic reaction and must be driven to completion by coupling to a highly exergonic methyl transfer. For this purpose, AdoMet is used as the methyl donor for reductive activation. The transfer of the methyl group of AdoMet is associated with a driving force of about 17kcal/mol, helping to assure that the reductive remethylation proceeds quantitatively. The C-terminal domain of MetH binds AdoMet and catalyzes this alternate methyl donor reaction, and is designated as the reactivation module. Indeed, if this module is removed from the protein by limited proteolysis of the native enzyme, the methylated enzyme turns over until all the cobalamin accumulates as cob(II)alamin, at which point the enzyme can no longer be reactivated [6]. Further insights into the complicated catalytic and reactivation cycles of methionine synthase came as X-ray structures were determined for fragments of methionine synthase in the laboratory of Martha Ludwig. The entire enzyme has never been crystallized, presumably because of the conformational lability of the enzyme. The first fragment to be crystallized was the cobalamin-binding module [11]. This was the first structure of cobalamin bound to a protein, and it revealed the remarkable displacement of the D M B axial ligand by a histidine residue from the protein—what we now call the base-off,His-on state of cobalamin. As shown in Figure 5, His759 is coordinated to the cobalt of methylcobalamin and also linked by a network of hydrogen bonds to the carboxyl group of Asp757, which in turn is hydrogenbonded to the hydroxyl of Ser810. The binding of the prosthetic group in the base-off,His-on form was shown to be associated with a signature His-xAsp-xxGly—•41/42—Ser-x-Leu-25/26—Gly-Gly sequence that had originally been identified in glutamate mutase [12]. The ß-face of the cobalamin prosthetic group in the structure is shielded by a four helix bundle that forms the N-terminal portion of the module sequence and is referred to as the "cap". Thus far, only indirect evidence suggests that this conformation of the intact MetH protein exists in solution [13]. The structure of the C-terminal reactivation module of MetH was determined next [14], and then a structure was obtained of the entire C-terminal half of the protein comprising the cobalamin-binding and reactivation modules [15]. This structure was determined with a fragment of His759Gly MetH, and revealed that the fragment had crystallized in a conformation in which the ß-face of the cobalamin prosthetic group was now in contact with the reactivation module (Figure 6, right). Although, His759 is absent in this structure, the distance between Ca of Gly759 and the cobalt of the cob(II)alamin prosthetic group is 2.3 A greater than in the Cap:Cob conformation assumed by the isolated cobalamin-binding module. This increased distance would be predicted to cause the cobalamin of the wild-type enzyme to assume a base-off,His-off conformation, enforced in part by the juxtaposition of residues from the AdoMet-binding module between the cobalamin-binding domain and Met. Ions Life Sei. 2009, 6, 53-114

C O B A L A M I N - A N D C O R R I NO I D - D E P E N D E N T E N Z Y M E S

63

CH,

7

Co

-N His759

-N Η ο

,ο

Η I ο,

Asp757, Η

«

Η

Λ

C SerSlO

Figure 5. Coordination of methylcobalamin in the cobalamin-binding module of MetH. A hydrogen bonded network comprising His759, Asp757 and Ser810 links His759 to the external solvent; these three residues that are absolutely conserved in all M e t H enzymes, are referred to as the ligand triad. Reprinted f r o m [177] with permission of Annual Reviews of Biochemistry, copyright 1997.

the corrin ring. Indeed, the methylated form of the wild-type enzyme has been shown to undergo interconversions between His-on and His-off forms that are induced by binding of ligands, shifts in temperature, or the binding of flavodoxin [15,16]. These interconversions are thought to reflect rearrangements of the four modules of methionine synthase, as shown in Figure 7. The left hand side of Figure 6 shows the structure of the N-terminal substrate-binding modules of methionine synthase from Thermotoga maritima [17]. The homocysteine- and folate-binding modules of methionine synthase are both α 8 β 8 barrels, with their openings positioned orthogonally with respect to each other. There is a large buried surface area between the two barrels, strongly suggesting that they move as a unit rather than individually. Thus, as cartooned in Figure 7, large modular rearrangements are required to allow the cobalamin to access the Hcy-binding and Fol-binding modules alternately during catalysis. Originally, it was proposed that the base-off,His-on state of cobalamin in methionine synthase might accelerate the methyl transfer reactions [11]. However, mounting evidence suggests that the primary role of the ligand Met. Ions Life Sei. 2009, 6, 53-114

MATTHEWS

Figure 6. Structures of the N-terminal and C-terminal halves of methionine synthase. The structure of the homocysteine-binding (green) and folate-binding (gold) modules of MetH was determined with protein from Thermotoga maritima [17], and the structure of the cobalamin-binding (red) and reactivation (blue) modules of MetH was determined with the His759Gly mutant of the enzyme from Escherichia coli [178]. The cobalamin-binding module also contains a four helix bundle, referred to as the "cap" and shown in grey.

replacement is to facilitate the conformational changes necessary for catalysis. In the initial studies, the wild-type His759 enzyme was compared with mutations of each residue of the ligand triad: His759Gly, Asp757Glu and Asp757Asn, and Ser810Ala. While the His759Gly mutant was inactive in steady-state assays, the Asp757 mutants showed /veat values that were 4—6% of the wild-type enzyme and that for the Ser810Ala mutant was 56% [5]. When the approach to steady-state was examined by enzyme-monitored turnover, the Asp757 mutants were 33 to 54% as fast as the wild-type enzyme and the Ser810Ala mutant was 61% as fast, indicating that these mutants were barely compromised in establishing the initial steady-state distribution of methylcobalamin and cob(I)alamin enzyme forms. The rate constant for the AdoMet- and reduced flavodoxin-dependent reactivation of enzyme in the cob(II)alamin form, which occurs in state 4 of Figure 7, was also measured. The His759Gly mutant was methylated 14 times faster than wild-type enzyme, as was the Asp757Glu mutant. The Asp757Asn and Ser810Ala mutants were methylated about twice as fast as the wild-type enzyme. Based on these data, Jarrett [18] proposed that mutations of residues in the ligand triad might alter the distribution of states Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- A N D C O R R I NO I D - D E P E N D E N T E N Z Y M E S Hey: Cob

Fol : Cob

Cap : Cob

65 AdoMet: Cob

Figure 7. Conformational states of methionine synthase. The four modules are shown in gold (Hcy-binding), green (folate-binding), red (cobalamin-binding domain) and gray (cap domain), and blue (AdoMet-binding). The corrin ring of methylcobalamin is indicated by the rectangle on top of the cobalamin-binding domain, and His759 is indicated by the vertical line. In the AdoMet:Cob conformation, the histidine is displaced as indicated in the cartoon, and the corrin ring tilts away from the cobalamin-binding domain and from Ca of His759. Reprinted from [20] with permission of the National Academy of Sciences of the USA.

shown in Figure 7, and that, as the ligation of the histidine was weakened and then finally abolished, the distribution would increasingly favor the AdoMet:Cob conformation. In agreement with this proposal, it was found that the EPR spectra of wild-type and mutant enzymes in the cob(II)alamin form increasingly favored the His-off conformation in the order: His759Gly (100%) > Asp757Glu (65%) > Asp757Asn (25%) > wild-type (15%) > Ser810Ala (5%). The next contribution to our understanding was the demonstration that addition of oxidized flavodoxin to methionine synthase in the cob(H)alamin form resulted in the conversion of the His-on state of the prosthetic group to the His-off state, as shown by the loss of superhyperfine coupling in the EPR spectrum of the cobalamin [9]. Further studies [19] established that enzyme in the presumably His-off four-coordinate cob(I)alamin form could not interconvert between catalytic conformations (states 1, 2, and 3 in Figure 7) and the AdoMet:Cob conformation (state 4 in Figure 7). If MetH in the cob(I)alamin state is produced by reduction of cob(H)alamin, the enzyme thus formed reacts with AdoMet but not with methyltetrahydrofolate, suggesting that the enzyme can assume the AdoMet:Cob conformation but not the Fol:Cob conformation. Conversely, if cob(I)alamin is generated by demethylation of enzyme in the methylcobalamin state in the presence of homocysteine, the cob(I)alamin reacts with methyltetrahydrofolate 30,000fold more rapidly than it reacts with AdoMet, suggesting that it can readily assume the Fol:Cob conformation but has very limited access to the AdoMet:Cob conformation. Further evidence that the two forms of cob(I)alamin are in different protein conformations came from the observation that limited proteolysis of the native enzyme resulted in different patterns. The Met. Ions Life Sei. 2009, 6, 53-114

66

MATTHEWS

pattern seen with cob(I)alamin enzyme generated by reduction was also seen with wild-type cob(II)alamin enzyme with flavodoxin b o u n d (previously shown to be His-off [9]) and also with cob(II)alamin b o u n d to the His759Gly m u t a n t . In contrast, the cob(I)alamin enzyme generated by demethylation with homocysteine showed the same cleavage p a t t e r n as wild-type enzyme in the methylcobalamin and cob(II)alamin (in the absence of flavodoxin) forms. T h u s the results suggested that the first cleavage pattern was characteristic of enzyme in the A d o M e t : C o b c o n f o r m a t i o n , while the second cleavage p a t t e r n was characteristic of enzyme in one of the catalytic conf o r m a t i o n s (states 1, 2, and 3 in Figure 7). T h e results described thus far indicated that enzyme in the cob(I)alamin f o r m can n o t freely interconvert between catalytic and reactivation conf o r m a t i o n s , while enzyme in the cob(II)alamin f o r m interconverts readily on addition of flavodoxin. B a n d a r i a n et al. [15] discovered t h a t enzyme in the methylcobalamin f o r m can also be induced to interconvert between catalytic and reactivation c o n f o r m a t i o n s , and t h a t the reactivation c o n f o r m a t i o n is associated with an a b s o r b a n c e spectrum typical of base-off methylcobalamin. H e showed that the H i s - o f f / H i s - o n equilibrium was < 5 : 9 5 for the full length wild-type enzyme, while it was 12:88 for the Asp757Glu m u t a n t , confirming that this ligand triad m u t a t i o n weakens the ligation of His759 to the cobalt. Addition of A d o H c y , which is b o u n d at the interface between the A d o M e t m o d u l e and the C o b m o d u l e in the A d o M e t : C o b c o n f o r m a t i o n of the protein, shifts the equilibrium to favor the His-off state, consistent with the a r g u m e n t t h a t the His-off state is associated with the A d o M e t : C o b c o n f o r m a t i o n . A d d i t i o n of methyltetrahydrofolate also shifts the equilibrium to favor the His-off state, presumably because the methyl g r o u p of methyltetrahydrofolate is in steric conflict with the methyl g r o u p of methylcobalamin in the F o l : C o b state, which is therefore disfavored. The two ligands exert their effects on the equilbrium independently, favoring the His-off state with free energy changes of 0.9 and 0.6 kcal/mol, respectively. The picture that emerges is of a delicately balanced equilibrium between alternate c o n f o r m a t i o n s of methionine synthase, with small free energy changes induced by ligand binding able to shift the distribution of conformers because these ligands have different affinities for the different states. The next advance in our understanding of the dynamic equilibrium of conformers in M e t H came f r o m the studies of Fleischhacker [16]. She examined the effect of substitutions in the upper axial (β) ligand of cob(III)alamin on the conformational equilibrium. She showed that methionine synthase in the propylcobalamin f o r m had a His-off/His-on equilibrium of 31:69 in the absence of ligands, while the His-off f o r m was undetectable when the enzyme prosthetic group was aquacobalamin. The His-off/His-on equilibrium was predicted by the ligand trans influence, with the more electron-donating propyl substituent favoring the A d o M e t : C o b conformation. Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

67

These studies provided a rationale for the base-off,His-on substitution. The histidine ligand serves as a protein sensor of the oxidation and ligation state of the cobalamin, biasing the equilibrium in accord with the net formal charge on the cobalt and its resultant effect on the histidine ligation. The cob(II)alamin state of methionine synthase showed properties intermediate between methylcobalamin and propylcobalamin in its ability to enter the AdoMet:Cob conformation. Oxidation of the prosthetic group to the cob(II)alamin state would lead to an enhanced propensity to enter the AdoMet:Cob conformation, which is favored by addition of flavodoxin and/ or by addition of AdoMet (there is no steric conflict between cob(II)alamin and AdoMet). Once the prosthetic group is reduced and methylated, the resultant His-off methylcobalamin is converted to the His-on state and thus returns to the catalytic cycle. A further role for the His759 ligand was discovered when the structure of a C-terminal fragment of "wild-type" methylated enzyme was determined [20]. To stabilize the AdoMet: Cob conformation of the enzyme, a disulfide crosslink was introduced between the " c a p " and the cobalamin-binding domain by mutation of Ile690 and Gly743 to cysteine residues. The crystal structure of this fragment revealed that it was indeed in the AdoMet:Cob conformation, but that the histidine had now moved about 7 A away from the cobalt and was now involved in intermodular hydrogen bonding with the AdoMet-binding domain (Figure 8). These unanticipated intermodular contacts would be expected to stabilize the His-off forms of the wild-type enzyme in the AdoMet:Cob conformation by as much as 3-5kcal/mol.

2.3.

Corrinoid-Dependent Methyltransferases in Methanosarcina spp.

Methanogens in the genus Methanosarcina use protein complexes containing corrinoid-binding proteins to catabolize simple one-carbon compounds such as methylamines and methylthiols as well as methanol. These complexes typically consist of a substrate-specific methyltransferase, a cognate corrinoid protein that receives the methyl group, and a second methyltransferase that catalyzes the transfer of the methyl group from the corrinoid protein to coenzyme Μ (ethane thiol sulfonate). More recently, tetramethylammonium-coenzyme Μ methyltransferase activity has been identified in Methanococcoides sp. [21]. Figure 9 diagrams the individual complexes that have been studied: methanol-coenzyme Μ methyltransferase [22], dimethylsulfide-coenzyme Μ methyltransferase [23], monomethylamine-coenzyme Μ methyltransferase [24,25], dimethylaminecoenzyme Μ methyltransferase [26], trimethylamine-coenzyme Μ methyltransferase [27], and tetramethylammonium-coenzyme Μ methyltransferase [28]. However, the genome sequence of Methanosarcina acetivorans contains 10 Met. Ions Life Sei. 2009, 6, 53-114

68

MATTHEWS

El 069

Figure 8. Intermodular contacts between His759 and residues in the AdoMetbinding reactivation module of MetH. Νε2 of His759 interacts with the AdoMet module directly through a hydrogen bond to Aspl093 and via a water-mediated hydrogen bond to Glul069. Ν δ ΐ of His759 forms a hydrogen bond with the amide of the propionamide side chain of ring Β of the cobalamin (not shown). Adapted f r o m [20] with permission of the National Academy of Sciences, USA.

sequences with homologies to substrate-specific methyltransferases, 15 putative corrinoid protein sequences, and 14 sequences with homologies to coenzyme Μ methyltransferases. The substrates for many of these proteins remain unidentified [28,29], The reactions catalyzed by these cytoplasmic enzymes bear striking similarity to the overall reaction catalyzed by cobalamin-dependent methionine synthase, and indeed the corrinoid-binding proteins in those complexes that have been sequenced show homology with the cobalamin-binding domain of methionine synthase, including the characteristic Asp-X-His-X-X-Gly motif indicative of a corrinoid cofactor with a histidine axial ligand. However, the methylcorrinoid-coenzyme Μ methyltransferase proteins do not show significant sequence homology with the homocysteine-binding domain of methionine synthase, but instead show sequence similarity with uroporphyrinogen decarboxylase (TJroD) [30]. The substrate-methylcorrinoid methyl Met. Ions Life Sei. 2009, 6, 53-114

C O B A L A M I N - A N D C O R R I NO I D - D E P E N D E N T E N Z Y M E S MtaB

MtaC

MtaA

MtsA

MtsB

MtsA

69

(mercaptopropionic acid) MtmB

MtmC

MtbA

MtbBl

MthC

MtbA

MttBl

MttC

MtqB

MtqC

MtbA

MtqA

Figure 9. Complexes catalyzing methyl transfer from methyl donors to coenzyme M. The gene product designations are shown above each protein. The first two letters, mt, indicate involvement of the gene product in methyl transfer, the third letter indicates the substrate: a for methanol, s for methylthiols, m for monomethylamine, b for dimethylamine, t for trimethylamine and q for tetramethylammonium. The final letter designates the polypeptide function: where Β is the substrate-specific methyltransferase that methylates the corrinoid protein with a methyl group derived f r o m substrate, C is the corrinoid binding polypeptide and A is the CoM-methylating protein, as originally suggested by Krzycki and coworkers [27].

Met. Ions Life Sei. 2009, 6, 53-114

70

MATTHEWS

transferases neither resemble methionine synthase domains nor each other. The genes specifying MtmC, MtbC, and MttC all contain an in-frame U A G stop codon in the middle of the open reading frame [27], and this U A G has been shown to specify pyrrolysine [31]. This unique residue is located at the active site of MtmB and is thought to be involved in activation of the amine substrate for methyl transfer. Despite the lack of sequence similarity between MtmB and the methyltetrahydrofolate-binding domain of MetH, their overall structures are similar. Of this group of enzymes, the most mechanistically characterized system is the Mta complex catalyzing methanol-coenzyme Μ methyl transfer. As mentioned above, M t a C shows sequence homology with other corrinoid proteins involved in cytoplasmic methyl transfers to coenzyme M, and with the cobalamin-binding module of MetH. The active-site histidine responsible for the base-off,His-on ligation of the 5-hydroxybenzimidazolylcobamide corrinoid of M t a C was shown to be Hisl36, the histidine in the signature AspX-His-X-X-Gly sequence [22]. M t a C is isolated in a complex with MtaB, and the complex was shown to catalyze methylation of the corrinoid prosthetic group using methanol as the methyl donor [32]. Recently, an X-ray structure of the MtaBC complex has been determined [33]. Thus far, this is the only structure of a methyl transferase corrinoidbinding protein in complex with one of its substrate binding domains. M t a C is indeed structurally related to the cobalamin-binding module of MetH, and as in that structure it contains both a four helix bundle (the cap) and the Rossmann domain responsible for cobalamin binding. As in the structure of the C-terminal fragment of His759Gly MetH, the cap is displaced to allow juxtaposition of MtaB with the Rossmann domain. MtaB is composed of an α 8 β 8 barrel with similarities to the substrate-binding domains of MetH, and a unique helical layer that is not seen in MetH. A zinc atom is located at the C-terminus of the barrel in a deep funnelshaped pocket (Figure 10); this zinc was previously shown to be required for activity [34]. In the M t m B C structure the barrel is positioned over the Rossmann domain so as to position the zinc above the cobalt of the corrinoid prosthetic group and to define the methanol binding site at the interface. The catalytic zinc ion is ligated by Glul64, Cys220, and Cys269, and although it exhibits approximately tetrahedral geometry, is apparently lacking a fourth ligand. The authors assume that methanol binds to the fourth site on the zinc through its hydroxyl group. Additional electron density, which has been modeled as a potassium ion, is located 3.1 A away from the zinc, and this putative potassium ion is also ligated by Glul64 as well as by other oxygen ligands. The authors suggest that the methanol will actually bridge between the potassium ion and the zinc. Hisl36, the cobalt of the corrinoid, the methyl group and the oxygen of methanol and the catalytic zinc all form a line, favoring an S N 2 mechanism in Met. Ions Life Sei. 2009, 6, 53-114

C O B A L A M I N - A N D C O R R I NO I D - D E P E N D E N T E N Z Y M E S

S-V

s 7,n

\

\

ΜI Gin 1614 I

Co

\

..-x"^,

H1129; h^··. /

71

v

.

|Glu2äf I

'''OVA ] I

Zn1*

Ώ—C—

Η'

- - *-"-NZ pLys26T]

0hS4

ν

C>Co{l) s δ λ fci y. Ο K U

δ ί^ £ .3

Κ ^ sα^ α -Ci 9 α κ κ ^ ε; α α S § "5 "5 ω -S

κ α -g

ft ο Ö , >> ο α ί -Ο -ο -α ' - -Ö ο ο 60 ο Ν 3 u oj 5 •Β -ö π) oj >> > Ο > >> I a ö £ ο υ 13 ω Μ Ο ö α I Ο ΟΜ* .9 1 S3SH ο ω S ® ο ä Ο ϋ Ö >> >> .S π! -α 'ν >> Λ Ν ω) -3 S S 2 λ > ο ο .a ο ωΝ 2 3 rt π) λ g "θ U V? 'ν

Κ rl Ö ai fc ^ ,3 ^ % ο ο j .0 Λ I }Η U

*

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

81

cob(II)alamin [9]. This shift required stoichiometric concentrations of methionine synthase and flavodoxin. W h e n the structure of the C-terminal half of His759Gly methionine synthase in the A d o M e t : C o b conformation was determined, it was apparent that assumption of this conformation led the cobalamin to assume a base-off conformation. T h u s oxidized flavodoxin, which is incapable of electron transfer, is assuming a role as a chaperone to facilitate the formation of base-off cobalamin in the A d o M e t : C o b conformation when the enzyme is in the cob(II)alamin form. This is of course the conformation needed for reductive remethylation of the cofactor. If the enzyme is in the aquacob(III)alamin form, flavodoxin binds tightly, but the shift to the base-off conformation does not occur. M a m m a l s lack flavodoxin and flavodoxin (ferredoxin) reductase. F o r the activation of methionine synthase they instead employ a fusion protein with an N-terminal d o m a i n h o m o l o g o u s to flavodoxin and a C-terminal d o m a i n h o m o l o g o u s to flavodoxin reductase [90]. This protein, methionine synthase reductase, is required for the in vivo activation of methionine synthase, and patients with severe deficiencies of this enzyme present with symptoms resembling those of patients with severe deficiencies of methionine synthase itself [91]. M e t h i o n i n e synthase reductase is n o t only required for the reductive reactivation of h u m a n methionine synthase. The oxidized reductase substantially stabilizes the apoenzyme, which rapidly undergoes irreversible d e n a t u r a t i o n on incubation at 37 °C [82]. F u r t h e r m o r e , methionine synthase reductase greatly increases the yield of holoenzyme f o r m e d on incubation of apo-methionine synthase with a q u a c o b a l a m i n and dithiothreitol [82]. These findings suggest a chaperone-like f u n c t i o n for methionine synthase reductase. T h u s far, insufficient a m o u n t s of h u m a n methionine synthase have been available to permit experiments to determine whether methionine synthase reductase also stabilizes holoenzyme in a base-off AdoMet:Cob conformation. T h u s far, A d o M e t - d e p e n d e n t reductive activation appears to be unique to methionine synthase. W h e r e activation of other methyltransferases has been shown to occur, reductive activation appears to be coupled to A T P rather t h a n A d o M e t . F o r a listing of methyltransferase activating proteins see Table 1. The best-studied reactivation protein is the methyltransferase-activating protein ( M A P ) which is involved in the activation of the cytoplasmic methylaminexoenzyme Μ methyltransferases and m e t h a n o l x o e n z y m e Μ methyltransferase in Methanosarcina barkeri. Sustained activity of these coenzyme Μ methyltransferases requires A T P , hydrogen, hydrogenase and M A P [83]. Incubation of A T P with M A P at 1:1 concentration ratios led to the phosphorylation of M A P . Phosphorylated M A P substituted for A T P in the stoichiometric activation of M T i , the corrinoid-containing methanol:5hydroxybenzimidazolyl cobamide c o m p o n e n t of the cytoplasmic methan o l x o e n z y m e Μ methyltransferase complex. If A T P were present in excess, Met. Ions Life Sei. 2009, 6, 53-114

82

MATTHEWS

M A P could catalyze multiple rounds of activation, indicating that A T P hydrolysis occurred during the activation process. If M T j in the aquacob(III)inamide form was incubated with hydrogen and hydrogenase, base-on cob(II)amide was formed. Addition of M A P and A T P resulted in formation of a mixture of base-on (60%) and base-off (40%) cob(H)inamide, and if methanol was then added, the prosthetic group was quantitatively converted to the methylcobinamide form [93]. Thus, phosphorylated M A P acts as a chaperone, inducing a conformation change in M T i that favors the formation of base-off cob(II)inamide, similar to the effect of oxidized flavodoxin on M e t H in the cob(H)alamin form. The difference is that methanol, the substrate, also serves as the methyl donor in reductive reactivation. A protein thought to be similar or identical to M A P can also serve to activate the corrinoid-proteins in the methylamine methyltransferase complexes [97]. However, annotation of the Methanosarcina acetivorans genome sequence [28,98] indicates that a gene specifying an iron-sulfur protein designated R a m M is responsible for activation of methylamine methyltransferases. A number of homologues of ramM have been identified in the M. acetivorans genome, although their functions have not yet been determined.

2.9.

Methyl Transfer in Fosfomycin Biosynthesis

The biosynthesis of fosfomycin was initially proposed to involve a unique function for methylcobalamin, namely the transfer of the methyl group as a methyl anion (Figure 12). However, all characterized methylcobalamin-dependent methyltransferases transfer the methyl group as a methyl cation. The

phosphonacelaldehyde

H+

2

V

Ο *

fosfomycin

Ο

Ο"

CH, / c 7 ~ 7

/

co*/

Figure 12. Originally proposed mechanism for generation of the methyl group of fosfomycin. Adapted f r o m [101] with permission f r o m the Royal Society of Chemistry, copyright 2007. Met. Ions Life Sei. 2009, 6, 53-114

C O B A L A M I N - A N D C O R R I NO I D - D E P E N D E N T E N Z Y M E S

83

evidence for the involvement of methylcobalamin as the methyl donor comes from the studies of Seto and Kuzuyama and their colleagues [99,100]. They showed that a mutant strain of Streptomyces with a block in the B 12 biosynthetic pathway could not produce fosfomycin, and that feeding of 14C-labeled methylcobalamin to this blocked mutant resulted in 14C-labeled fosfomycin. Recently, van der Donk and his colleagues [101] identified the entire biosynthetic cluster for fosfomycin and showed that it induced fosfomycin production in a Streptomyces strain lacking this capability. Analysis of the gene sequence of one of the components of the cluster, fom3, showed that it contained a region with similarities to a B 12 -binding domain and another region with homology to the family of proteins that utilize AdoMet as a radical generator. The fom3 gene was shown to be essential for the biosynthesis of fosfomycin. Furthermore the authors provided a strong inference that the actual substrate for methylation was hydroxyethylphosphonate rather than phosphonoacetaldehyde, leading to the proposal of a mechanism much more in keeping with known B 12 chemistry (Figure 13). We must await the purification and characterization of Fom3. But if the proposed mechanism is indeed correct, we may have a fascinating clue to the origin of AdoCbl-dependent enzymes. As pointed out by Sauer and Thauer in their review [53], thus far all corrinoid protein characterized from methanogenic Archaea have been methyltransferases containing methylcobalamin as a prosthetic group and no AdoCbl-dependent enzymes have been found. Furthermore, the cobO gene required for synthesis of AdoCbl appears to be lacking. Fom3, and its analogues in methylation reactions required for the biosynthesis of other antibiotics produced in Streptomyces, may represent a

hydroxyethylphosphonate

Ο

Ο

dAdo + Met

Η

AdoMet

Figure 13. Mechanism for generation of the methyl group of fosfomycin proposed by van der D o n k and colleagues. Adapted from [101] with permission from the Royal Society of Chemistry, copyright 2007. Met. Ions Life Sei. 2009, 6, 53-114

84

MATTHEWS

step in the direction of development of the AdoCbl family of enzymes. In the mechanism proposed for Fom3, AdoMet is cleaved to generate an adenosyl radical which abstracts a hydrogen from hydroxyethylphosphonate. The substrate radical is then methylated by methylcobalamin leaving cob(II)alamin as the product. Regeneration of methylcobalamin might then occur by way of a process similar to the reactivation of methionine synthase, using an external reductant and AdoMet as the methyl donor. It should be noted that this proposed mechanism would require two molecules of AdoMet per methylation reaction: one to generate the initial radical and a second to serve as a methyl donor.

3.

ADENOSYLCOBALAMIN-DEPENDENT REARRANGEMENTS AND ELIMINATIONS

Although the AdoCbl-dependent enzymes initially attracted the greatest attention from organic and inorganic chemists because of their fascinating chemistry, we now know that they represent but a small branch of the corrinoid-dependent enzymes. They are found most frequently in eubacteria (Domain Prokaryota) and just one AdoCbl-dependent enzyme, methylmalonyl-CoA mutase, is found in mammals. N o AdoCbl-dependent enzymes have yet been identified in Archaea. The basic mechanism of AdoCbl-dependent rearrangements is shown in Figure 14. This mechanism was simultaneously elucidated in Abeles' laboratory at Brandeis University and Arigoni's laboratory in Zürich. In classic papers, Frey and Abeles [102] showed that AdoCbl bound to propanediol dehydrase is tritiated as the enzyme reacts with [l- 3 H]l,2-propanediol and that tritium could subsequently be transferred from the isolated tritiated coenzyme to unlabeled propanediol, and Retey and Arigoni [103] then showed that AdoCbl that had been labeled with tritium when catalyzing the propanediol dehydrase reaction could also transfer tritium to methylmalonylCoA. Frey, Essenberg, and Abeles [104] showed that tritium is transferred from [l- 3 H]propanediol to the C-5' position of the AdoCbl of diol dehydrase, and from [5'- 3 H]AdoCbl to C2 of the product propionaldehyde. They also showed that the hydrogen abstracted from CI of the substrate becomes equivalent with one or both of the hydrogens of the C5' position of the cofactor following the initial hydrogen transfer. Cleavage of the Co-C5' bond of AdoCbl to form cob(II)alamin and 5'-deoxyadenosine was first observed in the suicide inactivation of diol dehydrase by glycolaldehyde, and in the same paper the substrate propanediol was also shown to induce transient formation of cob(II)alamin [105]. At the same time, Retey, Umani-Ronchi, Seibl, and Arigoni [106], used l s O-enriched

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- A N D C O R R I NO I D - D E P E N D E N T E N Z Y M E S

I I

—C-C—

=

Η X

Η X

Ado-Cbl

Ado· + Cbl(ll)·

I I

-C-C—

Χ Η

Ado-Cbl Figure 14.

-C-C—

I I

-9-9X Η

Ado· + Cbl(ll)·

85

I I

-c-cAdoH + Cbl(ll)·

I I Λ1

-C-CAdoH +Cbl(ll)·

Basic mechanism of AdoCbl-dependent rearrangements.

propanediols to demonstrate the migration of l s O from C2 of propanediol to CI of propionaldehyde, thus demonstrating the migration of the substituent at CI (X in Figure 14) in another classic paper. In the sections that follow, I will review the literature on each of the AdoCbl-dependent enzymes in turn. First, however, I wish to emphasize several challenges to understanding the mechanisms of each of these enzymes: • Activation of the C-Co bond of AdoCbl for cleavage: The carboncobalt bond of AdoCbl is estimated to have a bond dissociation energy of 30kcal/mol [107]. Thus Finke and Hay have estimated that diol dehydrase must lower the barrier for Co-C bond homolysis by at least 14.7 kcal/mol for a rate acceleration of 1010! How this is achieved in any AdoCbl-dependent enzyme remains controversial. • Transfer of Η from substrate to deoxyAdo and from deoxyAdo to product: For some of the AdoCbl-dependent enzymes, the Η must traverse long distances (6-10 A) during the reaction. We are just now beginning to understand how Η is transferred by the various enzymes. • Catalysis of the migration of X: The issue of whether cob(II)alamin is a participant in catalysis or a bystander is an argument that has persisted over decades. However, recent studies have greatly illuminated the mechanisms of migration, especially in the enzymes that catalyze carbon skeleton rearrangements. It has become clear that the distance between cob(II)alamin and the substrate radical following homolytic cleavage of AdoClb in diol dehydrase and ethanolamine ammonia lyase is too great for the cobalamin to participate in the subsequent rearrangment. However, in the mutases, the distance between cob(II)alamin and

Met. Ions Life Sei. 2009, 6, 53-114

86

MATTHEWS

substrate radical would be consistent with participation of the cobalamin in the rearrangement [108], and in fact density functional theory calculations support this role for cobalamin [109]. In reviewing the voluminous literature on these enzymes, I have tried to emphasize recent developments in the field, particularly emphasizing what we have learned as X-ray structures of these proteins have been determined. But I have not done justice to the earlier stereochemical experiments that elucidated the details of the overall reactions, nor the extensive characterization of the substrate and product radicals. Rather, I have attempted to focus on the role of the B12 cofactor.

3.1.

Enzymes Catalyzing Carbon Skeleton Rearrangements

This class of enzymes catalyzes rearrangements that require cleavage of a carbon-carbon bond to allow migration of X. These enzymes are glutamate mutase, methylmalonyl-CoA mutase, and isobutyryl CoA mutase. In all three of these enzymes, the AdoCbl cofactor is ligated by a histidine residue from the protein. Indeed, following the cloning of glutamate mutase, Marsh and Holloway first recognized the Asp-X-His-X-X-Gly motif that characterizes His-on ligation in methionine synthase and the enzymes that catalyze carbon skeleton rearrangements [12]. While all three enzymes have similar cobalamin-binding domains or subunits, they differ considerably in the structures/sequences of the substrate-binding regions of the proteins.

3.1.1.

Glutamate

Mutase

Glutamate mutase catalyzes the reaction shown in equation (4), the conversion of (S) -glutamate to ('25',i5'J-i-methylaspartate. In this reaction,

a hydrogen on Cß is exchanged with the glycyl group on Ca. The enzyme is an a2p2 oligomer; the small subunit GlmS (σ) is 14.7 kDa while the large

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

87

subunit GlmE (ε) is 50kDa [110]. As noted above, the small subunit contains the Asp-X-His-X-X-Gly motif associated with binding of B12 in a DMB-off,His-on state. However, neither the large nor the small subunit binds AdoCbl tightly, and the binding site is created at the interface between the two subunits. For some studies, a fusion protein comprising both large and small subunits has been used to avoid the complications of subunit dissociation and loss of activity and AdoCbl [111]. The structure of glutamate synthase with AdoCbl substituted by methyl- or cyanocobalamin reveals that the architecture of GlmS is highly similar to that of the B12-binding domain of methionine synthase from E. coli [112]. The DMB nucleotide is deeply buried in a hydrophobic pocket in this subunit, and Hisl6, which is carried on the loop between the first strand and the first helix of the barrel, coordinates the α position of the cobalt in B12. As suggested by the conserved motif, Hisl6 is also hydrogen bonded to Aspl4; however, the third amino acid in the "ligand triad" is missing and instead Asp 14 also forms hydrogen bonds to main-chain amide groups and to a water molecule. A structure of GlmS apoenzyme has been determined by NMR and provides insights into how the holoenzyme may be formed [113]. In this structure, which otherwise resembles the architecture of the holoenzyme small subunit, residues 13-27 form a disordered and highly mobile loop. The region corresponding to the first α-helix in the holoenzyme, residues 18-27, rapidly interconverts between unstructured forms and an α-helical conformation. The unfolding of the first helix exposes to solvent the cavity where the DMB will reside in the holoenzyme, so that the apoenzyme is preorganized for incorporation of the Bi 2 cofactor. I have long argued that only cobalamins with relatively high propensities to form base-off cobalamin (e.g., methyl- and adenosyl-cobalamin, cob(II)alamin, and cob(I)alamin) will lend themselves to incorporation into proteins that bind the cofactor in the base-off,His-on state, but this NMR structure adds another element to our understanding of the process by which holoenzyme formation might occur. The GlmE subunit is an α 8 β 8 barrel, with the open end packed against the β face of the B 12 in the holoenzyme structure [112]. The crystallization medium contained tartrate, an analogue of methylaspartate, which bound in the barrel in close proximity to the B12 cofactor. In a subsequent paper [114], the structure of active glutamate mutase with AdoCbl and glutamate bound was determined. In this structure, the electron density of the adenine is clearly modeled, but fitting the electron density with the ribose moiety of adenosine requires modeling in a mixture of C2'-endo and Ci'-endo conformations (Figure 15). The Ci'-endo conformation places C5' of the ribose within bonding distance of the cobalt of cobalamin, while the CT-endo conformation leads to a 4.2 A distance between C5' and Co, but places C5'

Met. Ions Life Sei. 2009, 6, 53-114

88

MATTHEWS

C3'-endo ribose

C2'-endo ribose

Figure 15. Alternate conformations of the ribose of AdoCbl at the active site of glutamate mutase. On the left the ribose assumes a CV-endo conformation, placing the 5'-carbon within bonding distance of the cobalt of cobalamin and distant f r o m Cy of the glutamate substrate. On the right the ribose assumes a C2'-endo conformation, leading to breakage of the carbon-cobalt bond and placing C5' within van der Waal's distance of Cy of the substrate.

within 3.3 A of Cy of the glutamate substrate. Thus the authors propose that a simple pseudorotation of the ribose between two low-energy conformers leads to cleavage of the carbon-cobalt bond and abstraction of hydrogen from Cy of glutamate. Figure 16 shows the mechanism proposed for glutamate mutase. The initial research supporting this mechanism was performed in Horace Barker's laboratory in Berkeley and has been recently reviewed [110]. These studies established the stereochemistry of the reaction, and showed that there was no exchange of the hydrogens of substrate or product with solvent, and no exchange of potential intermediates such as glycine or acrylate. The reaction is unique among the enzymes catalyzing rearrangement of carbon skeletons in that the migrating carbon is sp3 hybridized, as shown in equation (4). Initial

,

3

NH3+

NH3+

Ado-CH3

Ado-CH3

•ooc-^coo-^oocA/Soo ^.0QCJAdo-CH 2 ·

NhV

NH 3 "

o o c ' Y 0 0 " —-ooc-^Y000 CHj CH3 Ado-CH,

Ado-CH2-

Figure 16. Proposed mechanism for glutamate mutase. The adenosyl radical formed by cleavage of the C-Co bond of AdoCbl abstracts a hydrogen from glutamate to yield the glutamyl radical. This leads to the elimination of acrylate and formation of glycyl radical, which in turn condenses with acrylate to form the ß-methylaspartate radical. In the final step, the product radical abstracts hydrogen from 5'-deoxyadenosine to regenerate the adenosyl radical, which can now recombine with cob(II)alamin to reform AdoCbl. Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

89

evidence for the fragmentation to form a glycyl radical and acrylate came from the observation that glycine and acrylate inhibit the enzyme synergistically, and induce the formation of an EPR spectrum similar to those observed for the substrate or product radical (reviewed in [110]). More recently, direct evidence for formation of glycine and acrylate was obtained by rapid-quench analysis of the reaction, and the rate of formation of these intermediates was shown to be faster than the overall rate of reaction [115].

3.1.2.

Methyleneglntcircite

Mutase

Methyleneglutarate mutase catalyzes a reaction that is very similar to that catalyzed by glutamate mutase, as shown in equation (5). One key difference however, in the reaction is that the migrating carbon is sp2 hybridized, a

(5)

)H methyleneglutarate

(/?) -3- meth y I itacon ate

property that is shared with all the other enzymes that catalyze carbon skeletal rearrangement except glutamate mutase. This property permits the rearrangement to proceed through a cyclopropylcarbinyl intermediate rather than requiring a fragmentation-recombination as shown in Figure 17 [110]. Rearrangements of cyclopropinyl radicals are well precedented in model chemistry and proceed extremely rapidly [116].

Η. Η OOC

ΌΟ C Η COOCOO"

COO*

Figure 17. Mechanism proposed for radical rearrangment in methyleneglutarate mutase. Because the migrating carbon is sp 2 , a cyclopropylcarbinyl radical can form to mediate the transfer of the radical between Ca and Cß [116]. Met. Ions Life Sei. 2009, 6, 53-114

90

MATTHEWS

The protein is isolated from Clostridium barkeri as a homotetramer of 60kDa subunits. The deduced amino acid sequence of the protein shows significant sequence homology in its C-terminal region with the cobalaminbinding regions of methylmalonyl-CoA mutase, glutamate mutase and MetH, including the conserved Asp-X-His-X-X-Gly sequence that is the hallmark of DMB-off,His-on binding of the cobalamin cofactor [117]. Mutation of the corresponding residues, His485 and Asp483, decreases the rate of substrate turnover by > 4000-fold and by 2000-fold respectively [118],

3.1.3.

Methylmalonyl-Coenzyme

A Mutase

Methylmalonyl-CoA mutase catalyzes the reaction shown in equation (6). The migrating carbon is sp 2 , allowing radical rearrangement to proceed

by way of a cyclopropylcarbinyl radical, as in methyleneglutarate mutase. The bacterial enzymes are aß heterodimers, with considerable homology between the α and β subunits, while the mammalian enzyme is an a 2 homodimer. Only one molecule of AdoCbl is bound per bacterial heterodimer, however. The X-ray structure of the enzyme from Propionibacterium shermanii was the first structure to be determined of a complete cobalaminbinding protein [119]. The a- and ß-chains exhibit similar folds, but only the α-chain contains bound cofactor. Each chain consists of an N-terminal α 8 β 8 barrel and a C-terminal domain that exhibits a fold similar to the cobalaminbinding domain of methionine synthase. Νε of HisA610 coordinates the lower axial position of the B 12 cofactor, which appears to be cob(H)alamin in this structure. Ν δ ΐ of HisA610 is hydrogen bonded to the carboxyl oxygen of AspA608, and the other carboxyl oxygen of the Asp608 side chain is hydrogen bonded to LysA604. As in glutamate mutase and MetH, the D M B substituent of the corrin ring is deeply buried in a hydrophobic pocket in the cobalamin-binding domain. The N-terminal barrel of the α-chain of methylmalonyl-CoA mutase is juxtaposed against the ß-face of the B 12 , similar to its position in glutamate mutase [119]. The protein was crystallized in the presence of desulfo-CoA, a

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

91

substrate analogue that lacks the terminal thiol and the succinyl group of succinyl-CoA. One equivalent of this analogue was bound in a narrow tunnel along the axis of the barrel of the α-chain, completely buried in the interior of the barrel [119]. Structures of methylmalonyl-CoA mutase were subsequently obtained for the substrate-free enzyme and for enzyme in a non-productive complex with CoA [120]. These structures, which were similar, revealed that in the absence of productively bound CoA, the α 8 β 8 barrel is split apart and the CoA binding site is accessible to solvent. When CoA binds, the barrel closes up and encapsulates the substrate. The adenosyl group of AdoCbl could be seen in the substrate-free complex, but when the active site closes it is no longer visible, and the TyrA89 side chain now occupies a position that overlaps with the adenosyl binding region in the substrate-free enzyme. The authors propose that the closing of the active site cavity forces the carbon-cobalt bond cleavage. Support for the role of TyrA89 as a "molecular wedge" comes from studies of TyrA89Phe and TyrA89Ala mutant enzymes that demonstrate 1000-fold reductions in catalytic activity, and the disappearance of cob(II)alamin from the spectrum of the enzyme taken under steady-state conditions [121]. In the wild-type enzyme the ratio of AdoCbl:cob(II)alamin is about 4:1 when the enzyme is catalyzing the conversion of methylmalonyl- to succinyl-CoA. These studies enforce the view that the enzyme uses conformational changes driven by the binding of the CoA substrate to break the carbon-cobalt bond of the cofactor, consistent with the earlier observation that substrate binding accelerates carbon-cobalt bond cleavage by a factor of 10 12 [122], Padmakumar and Banerjee have measured the Co-C homolysis rate of AdoCbl bound to methylmalonyl-CoA mutase [122]. When the rates of homolysis are compared in the presence of [CH 3 ]methylmalonyl-CoA and [CD 3 ]methylmalonyl-CoA, the rate in the presence of the deuterated substrate is at least 20-fold slower. One would not expect the deuteration of the substrate to affect the rate of cleavage of AdoCbl unless the situation shown in Figure 18 were to prevail. The large isotope effect on formation of the substrate radical slows the overall rate of cleavage of AdoCbl. In a subsequent study [123] Chowdhury and Banerjee measured the temperature dependence of the activation parameters for reaction with [CH 3 ]methylmalonyl-CoA, providing a better estimate of magnitude of the kinetic isotope effect at 49.9. Subsequently computational analysis of the reaction confirmed that the cleavage of AdoCbl was indeed a stepwise process, rather than being concerted with hydrogen atom transfer from substrate to the deoxyadenosyl radical formed on cleavage of AdoCbl [124]. Surprisingly, these computations predicted that the rate constants k\ and k_i in Figure 18 are actually much faster than the rate constant for transfer of the hydrogen atom from the substrate to deoxyadenosine. The equilibrium governing the first step is unfavorable as shown in Figure 18.

Met. Ions Life Sei. 2009, 6, 53-114

92

MATTHEWS .SCoA

Ο: "OOC

AdoCbl

Η

dAdo· +

cob(ll)alamin

("Η H(D)

.ςτηΔ

Η dAdoH(D) +

cob(ll)alamin

Figure 18. Coupling between cleavage of AdoCbl and formation of the substrate radical in methylmalonyl-CoA mutase. A rapid but unfavorable equilibrium between AdoCbl and the homolytically cleaved dAdo radical and cob(II)alamin precedes a rate-limiting hydrogen atom abstraction f r o m the substrate [124].

Electron paramagnetic resonance has been used to estimate the distance between cob(II)alamin and a succinyl-CoA radical at the active site and their relative orientations [125]. Line broadening induced by heavy atom substitutions in succinyl-CoA indicated that the radical was centered on the carbon α to the free carboxyl. The interspin distance was about 6 A between the two radical centers, and the radical could be modeled in a position very similar to that occupied by succinyl-CoA in a product complex determined by X-ray crystallography. The interspin distance is large for a system that exhibits such large deuterium kinetic isotope effects, which are well above the classical limit and suggest a significant contribution due to hydrogen atom tunneling. In a recent paper, the contribution of hydrogen tunneling to the radical transfer catalyzed by methylmalonyl-CoA mutase has been rigorously analyzed [126]. The authors conclude that the large kinetic isotope effect can only be explained if corner-cutting tunneling decreases the distance over which the system tunnels. Human methylmalonyl-CoA mutase is a mitochondrial enzyme and the only AdoCbl-dependent enzyme in humans, and mitochondrial Bi 2 processing involves reduction of cob(II)alamin to cob(I)alamin, conversion of cob(I)alamin to AdoCbl and then transfer to the methylmalonyl-CoA mutase apoenzyme. Human adenosyltransferase catalyzes the conversion of cob(I)alamin to AdoCbl using ATP as the source of the adenosyl group [127]. However, cob(II)alamin can not be used as the substrate, indicating that the adenosyltransferase does not itself catalyze the reduction of cob(II)alamin to cob(I)alamin. If adenosyltransferase is incubated with cob(II)alamin in the presence of methionine synthase reductase, ATP and NADPH, AdoCbl is formed [128]. Addition of a cob(I)alamin scavenging agent, iodoacetamide, has no effect on this conversion, indicating that the cob(I)alamin is sequestered. The reasonable assumption is therefore that

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

93

reduction occurs while cob(II)alamin is bound to adenosyltransferase. However, we note that methionine synthase reductase is probably not the physiological reducing agent for adenosyltransferase. This reaction occurs in mitochondria, while methionine synthase reductase is cytoplasmic, and patients lacking methionine synthase reductase do not show abnormalities in AdoCbl synthesis. The physiological reducing agent for adenosyltransferase thus remains to be identified. Banerjee and Brunold and their colleagues have shown that adenosyltransferase binds cob(II)alamin in the base-off,His-off form [129], which leads to a more favorable potential for reduction to cob(I)alamin. Furthermore, the cob(II)alamin becomes four-coordinate, lacking water as a ligand, when ATP is present. It is proposed that adenosyltransferase also functions as a chaperone, transferring the base-off AdoCbl to methylmalonyl-CoA mutase [130], which has a higher affinity for the cofactor than adenosyltransferase. While it remains to identify the protein responsible for the reduction of cob(II)alamin to cob(I)alamin in mitochondria, and it remains to determine whether cob(II)alamin is indeed bound to human adenosyltransferase during reduction, the idea of chaperoning this rare and reactive cofactor is highly compelling. The situation should be compared to that in cytoplasmic MetH, where reduction of cob(II)alamin to cob(I)alamin takes place when bound to M e t H itself, using electrons from a partner electron transfer protein, and methylation requires AdoMet bound to its own module in methionine synthase. In addition to the reducing agent and adenosyltransferase required for the activity of methylmalonyl-CoA in human mitochondria, a third component, M M A A , is also strongly stimulatory ( M M A A is the gene designation for this protein). This protein is a homologue of MeaB, a bacterial protein that is frequently found in operons also containing methylmalonyl-CoA mutase. Mutant strains lacking MeaB are unable to convert methylmalonyl-CoA to succinyl-CoA, although they retain the ability to synthesize AdoCbl [131]. Human patients lacking M M A A belong to the cblA complementation group of patients who present with methylmalonic aciduria [132]. Several studies have been carried out on MeaB from Methylobacterium extorquens, the organism in which MeaB was first studied. MeaB shows homology with GTPases, a family that includes many enzymes involved in assembly of metal cofactors [131]. It forms complexes with holomethylmalonyl-CoA mutase that are enhanced when G T P is bound, and methylmalonyl-CoA mutase stimulates the GTPase activity of MeaB [133]. Furthermore the GTP-bound form of MeaB slows the rate of oxidative inactivation of methylmalonyl-CoA mutase (to form aquacob(III)alamin) by about 15-fold [134]. However, the physiological role of MeaB and its human homologue M M A A in maintaining methylmalonyl-CoA mutase activity remains to be elucidated.

Met. Ions Life Sei. 2009, 6, 53-114

94

3.1.4.

MATTHEWS

Isobutyryl-Coenzyme

A Mutase

Isobutyryl CoA mutase catalyzes the reaction shown in equation (7). This reaction is very similar to the reaction catalyzed by methylmalonyl-CoA

isobutyryl-CoA

n-butyryl-CoA

mutase, with the carboxyl group in methylmalonyl-CoA being replaced by a methyl group in isobutyryl-CoA. Inactivation of the icmA gene in Streptomyces cinnamonensis leads to a strain that is unable to use valine or isobutyryl-CoA as carbon sources [135]. The genes specifying the large and small subunits of isobutyryl-CoA mutase in Streptomyces cinnamonensis have been cloned and sequenced and expressed in E. coli. The icmA gene specifies a 62 kDa large subunit with ~ 4 0 % sequence identity to the large subunits of bacterial methylmalonyl-CoA mutases [136]. However, homologies to the C-terminal cobalamin-binding regions of the latter proteins are lacking. Instead, homologies to the cobalamin-binding regions of methylmalonyl-CoA mutases are found in the icmB gene specifying the 14 kDa small subunit [137]. These homologies include the Asp-X-His-X-X-Gly motif associated with DMB-off,His-on binding of the cofactor. The purified protein is an α2β2 heterodimer. Given the extensive homologies with methylmalonyl-CoA mutase, it is likely that the catalytic mechanisms of these two proteins will be highly similar. Early studies showed that the enzyme catalyzes an intramolecular rearrangement in which the carbonyl thioester of isobutyryl-CoA undergoes a 1,2-migration to the pro-fSJ methyl and is replaced by a hydrogen atom at C(3) of «-butyryl-CoA with overall retention [138].

3.1.5.

MeaA - A Mutase of Unknown Function

A second protein with extensive homology to methylmalonyl-CoA mutase, MeaA, has been identified in Streptomyces collimis and Methylobacterium extorquens. The meaA gene from S. collimis has been cloned and sequenced [139]. It specifies a putative 74 kDa protein with 40% homology to methylmalonyl-CoA mutase and significant homology to the large and small subunits of isobutyryl-CoA mutase. Growth studies suggest that MeaA is Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

95

crucial for the production of methylmalonyl-CoA in Streptomyces cinnamonensis, although the actual reaction catalyzed remains to be elucidated.

3.2.

Aminomutases and Diol Dehydrases. Isomerization and Elimination

Aminomutases catalyze the 1,2 exchange of an amino group with a hydrogen atom, while diol dehydrases catalyze a 1,2 exchange between a hydroxyl group with a hydrogen atom. In a subset of the aminomutase reactions and all the diol dehydrase reactions, the 1,2 exchange is followed by elimination of water. Two enzymes of this type that have been particularly well characterized are diol dehydrase and ethanolamine ammonia lyase, and these will be discussed first. We will then turn to the aminomutases that do not catalyze a subsequent elimination, lysine 2,3-aminomutase and ornithine aminomutase.

3.2.1.

Diol

Dehydrase

Diol dehydrase catalyzes the conversion of (SJ-l^-propanediol (equation 8)

HSC J2Ü (S/l-l ,2-propanediol

Η © V c ^ D 2,2-propanediol

ρ ropio η aldehyde

(8)

and of (R)-1,2 propanediol (equation 9) to propionaldehyde and water. HaO JL^

H3C

ft OH

(Rj-Λ ,2-propanediol

2,2-propanediol

propionaldehyde

(9)

While the lack of stereoselection between (S)- and fi^J-l^-propanediols is unusual, the reader will note that the stereochemistry of the two reactions is Met. Ions Life Sei. 2009, 6, 53-114

96

MATTHEWS

different [106,140]. As discussed at the beginning of the section on AdoCbldependent mutases, the role of the AdoCbl cofactor as a radical generator was first elucidated for diol dehydrase by observing the exchange of tritium between the 5'-position of AdoCbl and the substrate. In contrast to the AdoCbl-dependent enzymes that catalyze carbon skeleton rearrangements, diol dehydrase and ethanolamine ammonia lyase bind AdoCbl in the DMB-on form. This was initially demonstrated by labeling the D M B moiety of AdoCbl with 1 5 N and then determining the EPR spectrum of the enzyme after homolytic cleavage of the carbon-cobalt bond was induced with the suicide substrate 2-methyl-l,2-propanediol [141]. The X-ray structure of diol dehydrase confirmed this conclusion [142]. Diol dehydrase is an α 2 β2δ 2 dimer, and the cobalamin is bound at the interface of the a- and ß-subunits of each monomer. The α-subunit is an α 8 β 8 barrel, with the cofactor bound at the C-terminal ends of the central ß-strands and the substrate 1,2-propanediol bound more deeply in the barrel. The lower face of the cofactor, with its D M B nucleotide coordinated to the cobalt, interacts primarily with the ß-subunit. Monovalent cations were reported to be essential cofactors for diol dehydrase [143], and the structure revealed a potassium ion coordinated to the two hydroxyls of the substrate. The distance between the cobalt of the cofactor and CI and C2 of propanediol were 8.4 A and 9.0 A, respectively, in good agreement with data estimating the distance between the radical intermediate observed during steady-state turnover and cob(II)alamin as about 9 A [144]. However, this places the 5'-deoxyadenosyl radical that would be generated by homolytic cleavage of AdoCbl too far from the substrate to permit substrate radical formation. Toraya and his colleagues proposed that a simple rotation of the 5'-deoxyadenosyl radical around the glycosidic bond would bring C5' within 2 A of CI of the substrate [145], as shown in Figure 19. Subsequent studies clarified the nature of the radical intermediate observed during steady-state turnover of diol dehydrase with 1,2-propanediol. EPR spectroscopy of radicals derived from 13 C- and deuterium-labeled substrates established that the radical center resided on CI [146]. Thus, the intermediate is a substrate-derived radical generated by hydrogen atom abstraction from CI. Further insight into the structure of the radical came from EPR studies of the effects of incorporation of solvent deuterium on the radical signal. These studies indicated that the unpaired electron on the radical center couples with the solvent exchangeable proton on the hydroxyl group at CI [147]. The latter studies have important implications for the mechanism of hydroxyl migration from CI to C2, as will be discussed below. The deuterium kinetic isotope effect on kCSit observed with [l- 2 H]-l,2-propanediol is 12, indicating that hydrogen abstraction from CI of the substrate is rate-limiting in catalysis. However, accumulation of a 5'-deoxyadenosyl radical is not seen in either steady-state or stopped-flow analyses. This radical Met. Ions Life Sei. 2009, 6, 53-114

C O B A L A M I N - A N D C O R R I NO I D - D E P E N D E N T E N Z Y M E S

97

r

Figure 19. Mechanism for hydroxyl migration in diol dehydrase. Electron paramagnetic analysis of diol dehydrase reconstituted with 3',4'-anhydro AdoCbl and subjected to homolytic cleavage to form the anhydroadenosyl radical and cob(II)alamin in the presence (A) and absence (B) of (^,S)-l,2-propanediol was used to model the structures. In the absence of substrate, the anhydroribosyl moiety rotates about 60° relative to its position in the presence of substrate, bringing the radical into a position that would be appropiate for hydrogen atom abstraction from substrate and that would not permit formation of a carbon-cobalt bond. Reprinted f r o m [149] with permission from the American Chemical Society, copyright 2006.

would be expected to accumulate if hydrogen abstraction from the substrate were rate limiting. A satisfactory resolution to this dilemma is cartooned in equation (10). The unfavorable equilibrium between AdoCbl and its

AdoCbl...SH

Ado·—SH cob(II)alamin

dAdoH-S· cob(II)alamin

(10)

homolytic cleavage products prevents the accumulation of observable 5'deoxyadenosyl radical, and leads to the observed kinetic isotope effect on the accumulation of the substrate-derived radical intermediate in the rate-limiting step of catalysis. Also consistent with this proposal is the observation that detectable formation of cob(H)alamin requires the presence of substrate [148]. Support for this postulate also comes from studies with diol dehydrase reconstituted with 3',4'-anhydroAdoCbl (Figure 20). The introduction of a double bond adjacent to the radical center generated by homolytic cleavage of Met. Ions Life Sei. 2009, 6, 53-114

98

Figure 20.

MATTHEWS

Structure of 3',4'-anhydroadenosylcobalamin.

this cofactor analogue greatly stabilizes the resulting radical, and results in its accumulation during reaction with 1,2-propanediol [149]. In fact, homolytic cleavage of 3'-4'-anhydroAdoCbl bound to diol dehydrase is also observed in the absence of substrate. Still to be resolved is the mechanism of hydroxyl migration from CI to C2 following hydrogen atom abstraction from CI of 1,2-propanediol. A variety of mechanisms have been proposed (Figure 21) involving (a) general base catalysis, (b) general acid catalysis, (c) partial protonation (hydrogen bonding), (d) electrophilic catalysis by the activating potassium ion, or (e) combined general acid/base (push-pull) catalysis (discussed in [147]). The demonstration that the proton remains on the CI hydroxyl group of the substrate-derived radical argues against the base-catalyzed mechanism for rearrangement [147]. Activation of the enzyme by thallous ion instead of K + , which introduces a spin 1/2 metal in close proximity to the substrate radical, fails to lead to spin coupling with the substrate-derived radical, also disfavoring a role for K + in catalyzing the hydroxyl migration [147]. Hisl43 is in hydrogen bonding distance to the hydroxyl on C2 of the substrate and Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRINOID-DEPENDENT ENZYMES

99

(A)

OH Η I I -c—C· I 1 Η OH B-J

PH u s ν VC-C. H C Λ H;jC

iv OH Η l l " H 3 C~C~C* •« * 1 I ^ Η oJ -BH

Hu

—BH

Η OH I I —H3C-C-C-I ι OH -B:

(B)

OH Η

Λ Η H-O —χ *f ¥

H,C—C-C· 3 I I Η OH

-A-H

Η OH I I -C-CH • I OH

)

OH

(C)

-a-h.

OH Η 3

I I

Η OH

-A--

H3C

Η

~0H +

Η OH I I HjC-C-C—Η 3 · I OH

^'H

OH

(D)

,K + HO Η ι ι ; H3C-C-C'/ = I 1/ Η Ο Η

HO

Η OH I I H3C—C—C — Η

h3CVl

ÜIH OH

OH

(E)

Hs 3

OH Η 1 I I Η Ov

OH H3C Η

Η OH I I HjC —C—C—Η Ο. Η-

+

0H,

Β

Figure 21. Proposed mechanisms for hydroxyl migration in diol dehydrase. The mechanisms proposed involve (a) general base catalysis, (b) general acid catalysis, (c) partial protonation (hydrogen bonding), (d) electrophilic catalysis by the activating potassium ion, or (e) combined general acid/base (push-pull) catalysis. Reprinted from Schwartz et al. [147] with permission f r o m Protein Science, copyright 2007. Met. Ions Life Sei. 2009, 6, 53-114

100

MATTHEWS

Glu 170 is in hydrogen bonding distance to the hydroxyl on C I . Mutation of these residues to alanine reduces k c a t by 77-fold and 38,000-fold respectively [150], making the push-pull mechanism shown in (e) of Figure 21 highly attractive. What then is the role of the essential monocation activator? Recent studies have shown that K + activates the spontaneous cleavage of AdoCbl in the absence of substrate [151]. The reaction is observed as the formation of aquacob(III)alamin and requires both 0 2 and K + . These studies suggest that the binding energy of K + is used to enforce a conformational change in the enzyme that strains the carbon-cobalt bond of AdoCbl, accelerating cleavage. This conclusion is in agreement with conclusions drawn from crystallographic studies of diol dehydrase with the coenzyme analogue adeninylpentylcobalamin bound [152]. This analogue has an increased distance between the adenosine and the cobalamin. These studies suggested that in diol dehydrase with K + bound, the AdoCbl would be bound in a strained conformation even in the absence of substrate. Spontaneous cleavage of AdoCbl bound to diol dehydrase would result in inactivation in the cellular milieu, which contains both oxygen and potassium ions, because the aquacobalamin is extremely tightly bound and this form of the enzyme is inactive. Mori and Toraya [153] have identified a reactivating factor, a chaperone-like protein with ATPase activity. In the presence of ADP, the chaperone mediates the release of aquacobalamin from the inactive enzyme and binds tightly to the apoenzyme, while in the presence of ATP, the apoenzyme is released and can be reconstituted with AdoCbl. This chaperone comprises two subunits, gene products of the ddrA and cklrB genes in Klebsiella oxytoca.

3.2.2.

Ethanolamine Ammonia Lyase

Ethanolamine ammonia lyase catalyzes the conversion of ethanolamine to acetaldehyde and ammonia. The reaction proceeds as shown in equation (11). The enzyme is an α 6 β 6 oligomer, and recent studies suggest that a mol NH4+ J *

V-CH, Η

(11)

of AdoCbl is bound per aß protomer [154]. The predicted masses of the aand ß-subunits from E. coli are 50 and 32kDa, respectively. Although the subunits show little homology with those of other B 12 -dependent enzymes, Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

101

the larger subunit does show limited homology with the subunits of methylmalonyl-CoA mutase from P. shermanii [154], and recent modeling of the large subunit of the enzyme from Salmonella typhimurium suggests that this subunit is indeed a β 8 α 8 barrel [155]. By analogy with diol dehydrase, we might expect the AdoCbl to be bound at the interface between the small and large subunits. As in diol dehydrase, the AdoCbl is bound in the " D M B - o n " form [156], The X-ray structure of ethanolamine ammonia lyase has not been determined, but sophisticated spectroscopic studies have revealed its mechanistic similarity with diol dehydrase and also provided some unique insights into the catalytic mechanism. One of the first issues was whether hydrogen transfer actually occurred directly between the AdoCbl radical and the substrate, or whether an intermediary such as a protein radical might be involved. Following cleavage of AdoCbl in the presence of substrate, the substrate-derived radical was found to be more than 10 A from the unpaired electron of cob(II)alamin [157]. In a landmark study, electron nuclear double resonance and specific 2 H- and 13 C-labeling of the substrate was used to show that the methyl group of 5'-dAdo was positioned 3.4 A from C I ' of the substrate radical, in a perfect position to mediate direct hydrogen transfer between cofactor and substrate [158]. Similar conclusions were reached using pulsed EPR [159]. Thus, the 5'-carbon of AdoCbl migrates ~ 7 A from its position when the cofactor is intact, to its position when the substrate radical is formed. One major difference between diol dehydrase and ethanolamine ammonia lyase is that, while the stereochemistry of the diol dehydrase reaction requires that the hydroxyl group migrates from CI to C2 prior to elimination of water, there is no such compelling evidence that the amino group of ethanolamine migrates rather than simply undergoing elimination from C2 after hydrogen atom abstraction at CI. However, arguing by analogy with all the other AdoCbl-dependent isomerases, we might expect that such migration does actually occur. A second difference is the apparent absence of a requirement for a cation (potassium ion or ammonia) for catalysis. In the modeling of the active site of the Salmonella enzyme, Sun and Warncke suggest that the guanidinium side chain of Arg 160 occupies the same position as the potassium ion in the active site of diol dehydrase [155].

3.2.3.

Lysine

5,6-Aminomutase

Lysine 5,6-aminomutase was initially characterized in the laboratory of Theresa Stadtman, and was shown to be an AdoCbl-dependent enzyme that catalyzed a 1,2-migration of the ε-amino group of D-lysine with concomitant Met. Ions Life Sei. 2009, 6, 53-114

102

MATTHEWS

reverse migration of a hydrogen atom, as shown in equation (12) [160]. The stereochemistry of the reaction has not yet been determined, so we do nh3+ (12) 2,5-diaminohexanoate

not yet know which of the two hydrogens on the δ-carbon of D-lysine migrates. More recently, the enzyme was cloned, sequenced, expressed and purified from Clostridium sticklandii [161]. In agreement with earlier studies, in which the enzyme was isolated as a 170 kDa complex of 55 and 30kDa subunits, the enzyme is an a 2 p2 heterodimer composed of 57 kDa a-subunits and 29 kDa ß-subunits. The small subunit contains the Asp-X-His-X-X-Gly sequence characteristic of His-on binding of the AdoCbl cofactor. The substrate D-lysine forms an external aldimine linkage (Figure 22) with the pyridoxal phosphate cofactor that is required for enzyme activity [162]. The pyridoxal phosphate is proposed to stabilize radical intermediates formed following hydrogen atom abstraction from the substrate as shown in Figure 22 [163], The X-ray structure of the enzyme [164] revealed that the large subunit is an α 8 β 8 barrel and positions the pyridoxal phosphate cofactor at the Cterminal end of the barrel strands. The crystal structure was obtained in the absence of D-lysine, and revealed that the pyridoxal phosphate was linked as

COO

N, «•OaPO" V ^ V

U M

2

O3po^V^v-um

2

O 3 PO

Ν substrate radical derived from external aldimine

azacyclopropylcarbinyl radical

product radical derived f r ° m external aidimine

Figure 22. Proposed role for pyridoxal phosphate in the reaction catalyzed by lysine 5,6-aminomutase [163]. Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

103

an internal aldimine with Lysl44 of the small subunit. The AdoCbl is sandwiched between the large and small subunits, in the predicted DMB-off,His-on conformation. However, the AdoCbl is far from the pyridoxal phosphate binding site in this "resting" conformation of the enzyme. The covalent bond between pyridoxal phosphate and Lysl44ß locks the two subunits together in a conformation that leaves the active site accessible to solvent and substrate and prevents cleavage of AdoCbl. The authors propose that substrate binding will lead to a rearrangement that will sequester the active site and position AdoCbl appropriately for catalysis.

3.2.4.

Ornithine

4,5-Aminomntcise

This enzyme catalyzes the reaction shown in equation (13). Until very recently, the enzymatic reaction had only been studied in crude extracts,

(13) (2R,4S)-diaminopentanoate

ornithine

where the product was shown to be (^i^SO-diaminopentanoate. However the stereochemistry of the migrating hydrogen atom in ornithine has not been determined. In 2001, the oraE and oraS genes specifying the large and small subunits of D-ornithine amino mutase were cloned and sequenced from Clostridium stick!andii [165], and in 2004, the enzyme was successfully expressed in E. coli and purified to homogeneity [166]. Despite the lack of sequence homology with other AdoCbl-dependent enzymes, the properties of the purified ornithine aminomutase are very similar to those of lysine 5,6aminomutase. The enzyme is an a 2 p2 heterodimer, and requires pyridoxal phosphate for activity in addition to AdoCbl.

3.3.

Adenosylcobalamin-Dependent Ribonucleotide Triphosphate Reductase

Ribonucleotide triphosphate reductase catalyzes the reaction shown in equation (14), which results in oxidation of two active site thiols to form a Met. Ions Life Sei. 2009, 6, 53-114

MATTHEWS

104

e

/

s h

+

p?p0-r(°

-f"

_ _

E

/f

disulfide. Regeneration of the enzyme by reduction of the disulfide bond is required for sustained turnover, and is accomplished by a series of electron transfers from N A D P H to thioredoxin reductase, to thioredoxin, and then to ribonucleotide triphosphate reductase as shown in equation (15). S ^S

+ NADPH

thioredoxin thioredoxin reductase

E

/

SH SH

+NADP+

(15)

Enzyme catalysis requires AdoCbl, which serves as a radical generator as it does in other AdoCbl-dependent enzymes. However, a unique feature of ribonucleotide triphosphate reductase is that hydrogen atom abstraction from the substrate is not catalyzed by the 5'-deoxyadenosyl radical, but rather by a thiyl radical generated by hydrogen atom abstraction from a cysteine residue on the protein. The enzyme from Lactobacillus leichmanii has been most extensively studied. In contrast to the intramolecular tritium transfer seen in other AdoCbldependent enzymes, if ribonucleotide triphosphate reductase labeled with tritium in the 5' position of AdoCbl is incubated with the allosteric activator dGTP and reductant, the label is quantitatively transferred to solvent [167]. Stubbe and her colleagues then showed using single turnover experiments that ribonucleotide triphosphate reductase catalyzes the cleavage of the 3'-carbon-hydrogen bond of UTP with quantitative release of label to solvent when the 3'-position is tritiated [168]. The authors proposed a working hypothesis for the enzyme that involved the generation of a protein radical that in turn generated the substrate radical by direct hydrogen atom abstraction of the 3'-hydrogen of UTP. Cysteine 408 was subsequently shown to be the site of formation of the protein radical [169], and rapid freeze quench electron paramagnetic resonance of an intermediate in the exchange of the 5'-hydrogens of AdoCbl with solvent catalyzed by ribonucleotide triphosphate reductase containing specifically deuterated cysteines demonstrated a cysteine-centered radical that was spin-coupled to cob(II)alamin. These studies also demonstrated that the formation and decay of the cysteine-based radical signal proceeded at rates consistent with its involvement in catalysis. The detailed mechanism proposed for the catalytic cycle [170] is shown in Figure 23. Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

105

AdoCbl Cys408SH

Ado* cob(ll)alamin Cys408SH

dAdoH cob(ll)alamin Cys408S·

Cys408S·

Cys408S·

•Ο. Β

PPPO Ho

pppo"

ht? HO

ΐ"Η

HO

OH

Β

ν ° γ Η s-s

HS SH

It Cys408SH Β

Ο R P P P O ^

OH

HO

ρρρο^γ°\/ HC)

Cys408SH

HS SH

HH S-S

Cys408SH

ρρρθ'/\/0γ·Β HO HS SH

Figure 23.

Proposed mechanism for AdoCbl-dependent ribonucleotide reductase [170].

Following the cloning and sequencing of ribonucleotide reductase f r o m L. leichmanii [171], a crystal structure of the obtained with the A d o C b l analogue adeninylpentylcobalamin The m o n o m e r i c enzyme folds as a ten stranded α/β barrel

triphosphate enzyme was b o u n d [172]. with Cys408

Met. Ions Life Sei. 2009, 6, 53-114

106

MATTHEWS

positioned at the tip of a hairpin loop at the bottom of the barrel. The AdoCbl analogue is bound with D M B in the "base-on" conformation, as previously predicted [173]. One issue that has recently been clarified concerns the mechanism of formation of the thiyl radical, which could be generated in a stepwise fashion by cleavage of AdoCbl to form 5'-deoxyadenosyl radical and cob(II)alamin, or in a concerted fashion in which a 5'-deoxyadenosyl radical is not an intermediate. When ribonucleotide reductase was incubated with stereoselectively deuterated (5'i?)-[5'- 2 H]AdoCbl and the allosteric activator dGTP, the 5'deuterium was stereochemical^ scrambled [174]. This scrambling occurred in both the wild-type enzyme and the Cys408Ala mutant, even though exchange of deuterium with solvent did not occur in the mutant enzyme. The implication is that transient cleavage of the Co-C5' bond of AdoCbl occurs in the presence of the allosteric activator even in the absence of Cys408. The second mechanistic issue concerns the sequence by which the activesite disulfide is reduced at the end of each turnover. As previously mentioned, the electrons ultimately come from N A D P H , and are transferred to the redox active disulfide of thioredoxin by thioredoxin reductase. Reduction of ribonucleotide triphosphate reductase by thioredoxin requires two cysteines at the C-terminus of the protein, Cys731 and 736 [169], and these two cysteines are thought to shuttle reducing equivalents between reduced thioredoxin and the active site disulfide. This C-terminal extension is disordered in the crystal structure of the enzyme [172].

4.

CONCLUDING REMARKS

The author hopes that this review highlights the explosion of information recently obtained about the growing family of characterized corrinoiddependent enzymes, and particularly the mounting insights available from structural analyses of these proteins. We now appreciate the vital role that corrinoid-dependent methyltransferases play in central metabolic pathways in prokaryotes and Archaea. A challenge for the future will be to reach a detailed understanding of how the membrane-associated energy conserving corrinoid methyl transferase couples sodium ion transport with methyl transfer from methylcobalamin to the thiol of coenzyme M. The elegant spectroscopies that have been applied to the AdoCbldependent enzymes have greatly clarified the nature of the radical intermediates and structural analyses have revealed the type of molecular motions required to allow hydrogen atom transfer from substrate to the deoxyadenosyl radical and back to product. These complicated enzymes have posed a challenge to enzymologists and chemists for the past 70 years, Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRI NO ID-DEPENDENT ENZYMES

107

and it has been thrilling to review the elegant experiments that step-by-step revealed the details of catalysis.

ACKNOWLEDGMENTS Research in the a u t h o r ' s l a b o r a t o r y is f u n d e d by N a t i o n a l Institutes of Health Grant GM24908.

ABBREVIATIONS AND DEFINITIONS AdoCbl AdoMet ADP ATP CoA dGTP DMB EPR EXAFS GTP Hey MAP MetH NADP(H) NMR UroD UTP

adenosylcobalamin S-adenosylmethionine adenosine 5'-diphosphate adenosine 5'-triphosphate coenzyme A deoxyguanosine 5'-triphosphate dimethylbenzimidazole electron paramagnetic resonance extended X-ray a b s o r p t i o n fine structure guanosine 5'-triphosphate homocysteine methyltransferase-activating protein cobalamin-dependent methionine synthase f r o m E. coli nicotinamide adenine dinucleotide p h o s p h a t e (reduced) nuclear magnetic resonance u r o p o r p h y r i n o g e n decarboxylase uridine 5'-triphosphate

REFERENCES 1. S. W. Ragsdale, P. A. Lindahl and E. Munck, J. Biol. Chem., 1987, 262, 14289-14297. 2. E. Stupperich, H. J. Eisenger and S. P. J. Albracht, Eur. J. Biochem., 1990, 193, 105-109. 3. S. W. Ragsdale, The Acetogenic Corrinoid Proteins, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999. 4. R. Y. Banerjee, N. L. Johnston, J. K. Sobeski, P. Datta and R. G. Matthews, J. Biol. Chem., 1989, 264, 13888-13895.

Met. Ions Life Sei. 2009, 6, 53-114

108

MATTHEWS

5. Μ. Amaratunga, Κ. Fluhr, J. Τ. Jarrett, C. L. Drennan, M. L. Ludwig, R. G. Matthews and J. D. Schölten, Biochemistry, 1996, 35, 2453-2463. 6. J. T. Drummond, S. Huang, R. M. Blumenthal and R. G. Matthews, Biochemistry, 1993, 32, 9290-9295. 7. K. Fujii, J. H. Galivan and F. M. Huennekens, Arch. Biochem. Biophys., 1977, 178, 662-670. 8. C. Osborne, L.-M. Chen and R. G. Matthews, J. BacterioL, 1991, 173, 1729-1737. 9. D. M. Hoover, J. T. Jarrett, R. H. Sands, W. R. Dunham, M. L. Ludwig and R. G. Matthews, Biochemistry, 1997, 36, 127-138. 10. J. T. Jarrett, C. Y. Choi and R. G. Matthews, Biochemistry, 1997, 36, 15739-15748. 11. C. L. Drennan, S. Huang, J. T. Drummond, R. G. Matthews and M. L. Ludwig, Science, 1994, 266, 1669-1674. 12. Ε. N. G. Marsh and D. E. Holloway, FEBS Lett., 1992, 310, 167-170. 13. J. T. Jarrett, C. L. Drennan, M. Amaratunga, J. D. Schölten, Μ. L. Ludwig and R. G. Matthews, J. Bioorgan. Med. Chem., 1996, 4, 1237-1246. 14. Μ. M. Dixon, S. Huang, R. G. Matthews and M. Ludwig, Structure, 1996, 4, 1263-1275. 15. Y. Bandarian, M. L. Ludwig and R. G. Matthews, Proc. Natl. Acad. Sei. USA, 2003, 100, 8156-8163. 16. A. S. Fleischhacker and R. G. Matthews, Biochemistry, 2007, 46, 12382-12392. 17. J. C. Evans, D. P. Huddler, Μ. T. Hilgers, G. Romanchuk, R. G. Matthews and M. L. Ludwig, Proc. Natl. Acad. Sei. USA, 2004, 101, 3729-3736. 18. J. T. Jarrett, M. Amaratunga, C. L. Drennan, J. D. Schölten, R. H. Sands, M. L. Ludwig and R. G. Matthews, Biochemistry, 1996, 35, 2464-2475. 19. J. T. Jarrett, S. Huang and R. G. Matthews, Biochemistry, 1998, 37, 5372-5382. 20. S. Datta, M. Koutmos, K. A. Pattridge, M. L. Ludwig and R. G. Matthews, Proc. Natl. Acad. Sei. USA, 2008, 105, 4115-4120. 21. K. Tanaka, J. Ferment. Bioeng., 1994, 78, 386-388. 22. K. Sauer and Τ. K. Thauer, Eur. J. Biochem., 1998, 253, 698-705. 23. T. C. Tallant, L. Paul and J. A. Krzycki, J. Biol. Chem., 2001, 276, 4485-4493. 24. S. A. Burke and J. A. Krzycki, J. Biol. Chem., 1997, 272, 16570-16577. 25. S. A. Burke, S. L. Lo and J. A. Krzycki, J. Bacteriol., 1998, 180, 3432-3440. 26. D. J. Ferguson Jr., N. Gorlatova, D. A. Grahame and J. A. Krzycki, J. Biol. Chem., 2000, 275, 9053-29060. 27. L. Paul, D. J. Ferguson Jr. and J. A. Kryzycki, J. Bacteriol., 2000, 182, 2520-2529. 28. J. E. Galagan, C. Nusbaum, A. Roy, M. G. Endrizzi, P. Macdonald, W. FitzHugh, S. Calvo, R. Engels, S. Smirnov, D. Atnoor, A. Brown, N. Allen, J. Naylor, N. Stange-Thomann, K. DeArellano, R. Johnson, L. Linton, P. McEwan, K. McKernan, J. Talamas, A. Tirrell, W. Ye, A. Zimmer, R. D. Barber, I. Cann, D. E. Graham, D. A. Grahame, A. M. Guss, R. Hedderich, C. Ingram-Smith, H. C. Kuettner, J. A. Krzycki, J. A. Leigh, W. Li, J. Liu, B. Mukhopadhyay, J. N. Reeve, K. Smith, T. A. Springer, L. A. Umayam, O. White, R. H. White, E. Conway de Macario, J. G. Ferry, K. F. Jarrell, H. Jing,

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRINOID-DEPENDENT ENZYMES

29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53.

54. 55.

109

A. J. Macario, I. Paulsen, M. Pritchett, K. R. Sowers, R. Y. Swanson, S. H. Zinder, E. Lander, W. W. Metcalf and B. Birren, Genome Res., 2002, 12, 532-542. J. E. Galagan, et αι., Genome Res., 2002, 12, 532-542. U. Harms and R. K. Thauer, Eur. J. Biochem., 1996, 325, 653-659. B. Hao, W. Gong, Τ. K. Ferguson, C. M. James, J. A. Krzycki and Μ. K. Chan, Science, 2002, 296, 1462-1466. K. Sauer, U. Harms and R. K. Thauer, Eur. J. Biochem., 1997, 243, 670-677. C. H. Hagemeier, M. Krer, R. K. Thauer, B. Warkentin and U. Ermler, Proc. Natl. Acad. Sei., USA, 2006, 103, 18917-18922. K. Sauer and R. K. Thauer, Eur. J. Biochem., 1997, 249, 280-285. L. D. Zydowsky, T. M. Zydowsky, E. S. Haas, J. W. Brown, J. N. Reeve and H. G. Floss, J. Am. Chem. Soc., 1987, 109, 7922-7923. D. A. Grahame, J. Biol. Chem., 1989, 264, 12890-12894. C. W. Goulding and R. G. Matthews, Current Opin. Chem. Biol., 1997, 1, 332-339. K. Sauer and R. K. Thauer, Eur. J. Biochem., 2000, 267, 2498-2504. M. Krüer, M. Haumann, W. Meyer-Klaucke, R. K. Thauer and H. Dau, Eur. J. Biochem., 2002, 269, 2117-2123. S. Gencic, G. M. LeClerc, N. Gorlatova, K. Peariso, J. E. Penner-Hahn and D. A. Grahame, Biochemistry, 2001, 50, 13068-13078. M. Blaut, Y. Müller and G. Gottschalk, J. Bioenerg. Biomembr., 1992, 24, 529-546. P. Gärtner, Α. Ecker, R. Fischer, D. Linder, G. Fuchs and R. Thauer, Eur. J. Biochem., 1993, 213, 537-545. U. Harms, D. S. Weiss, P. Gartner, D. Linder and R. K. Thauer, Eur. J. Biochem., 1995, 228, 640-648. E. Stupperich, A. Juza, M. Hoppert and F. Mayer, Eur. J. Biochem., 1993, 217, 115-121. B. Kräutler, J. Moll and R. K. Thauer, Eur. J. Biochem., 1987, 162, 275-278. U. Harms and R. K. Thauer, Eur. J. Biochem., 1996, 241, 149-154. U. Harms and R. K. Thauer, Eur. J. Biochem., 1997, 250, 783-788. P. Gärtner, D. S. Weiss, U. Harms and R. K. Thauer, Eur. J. Biochem., 1994, 226, 465-472. D. S. Weiss, P. Gärtner and R. K. Thauer, Eur. J. Biochem., 1994, 226, 799-809. V. Svetlitchnyi, H. Dobbek, W. Meyer-Klaucke, T. Meins, B. Thiele, P. Romer, R. Huber and O. Meyer, Proc. Natl. Acad. Sei. USA, 2004, 101, 446-451. U. Holder, D.-E. Schmidt, E. Stupperich and G. Fuchs, Arch. Microbiol., 1985, 141, 229-238. B. Eikmanns, G. Fuchs and R. K. Thauer, Eur. J. Biochem., 1985,146, 149-154. K. Sauer and R. K. Thauer, The Role of Corrinoids in Methanogenesis, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999. W.-P. Lu, I. Schiau, J. R. Cunningham and S. W. Ragsdale, J. Biol. Chem., 1993, 268, 5605-5614. J. Maupin-Furlow and J. G. Ferry, J. Bacteriol., 1996, 178, 340-346.

Met. Ions Life Sei. 2009, 6, 53-114

110

MATTHEWS

56. D. A. Grahame, J. Biol. Chem., 1991, 266, 22227-22233. 57. D. A. Grahame, Biochemistry, 1993, 32, 10786-10793. 58. K. C. Terlesky, M. J. K. Nelson and J. G. Ferry, J. Bacterial., 1986, 168, 1053-1058. 59. S. R. Harder, W.-P. Lu, B. A. Feinberg and S. W. Ragsdale, Biochemistry, 1989, 28, 9080-9087. 60. T. Svetlitchnaia, V. Svetlitchnyi, O. Meyer and H. Dobbek, Proc. Natl. Acad. Sei. USA, 2006, 103, 14331-14336. 61. P. E. Jablonski, W.-P. Lu, S. W. Ragsdale and J. G. Ferry, J. Biol. Chem., 1993, 268, 325-329. 62. T. A. Stich, J. Seravalli, S. Yenkateshrao, T. G. Spiro, S. W. Ragsdale and T. C. Brunold, J. Am. Chem. Soc., 2006, 128, 5010-5020. 63. S. Menon and S. W. Ragsdale, J. Biol. Chem., 1999, 274, 11513-11518. 64. S. Menon and S. W. Ragsdale, Biochemistry, 1998, 37, 5689-5698. 65. T. Doukov, J. Seravelli, J. J. Stezowski and S. W. Ragsdale, Structure, 2000, 8, 817-830. 66. D. Lexa and J.-M. Saveant, Acc. Chem. Res., 1983, 16, 235-243. 67. J. Seravalli, K. L. Brown and S. W. Ragsdale, J. Am. Chem. Soc., 2001, 128, 1786-1787. 68. F. Kaufmann, G. Wohlfarth and G. Diekert, Eur. J. Biochem., 1998, 257, 5125-5521. 69. F. Kaufmann, G. Wohlfarth and G. Diekert, Eur. J. Biochem., 2008, 253, 706-711. 70. D. Naidu and S. W. Ragsdale, J. Bacterial., 2001, 183, 3272-3281. 71. U. E. Krone, R. K. Thauer and H. P. C. Hogenkamp, Biochemistry, 1989, 28, 4908-4914. 72. Κ. M. McCauley, D. A. Pratt, S. R. Wilson, J. Shey, T. J. Burkey and W. A. van der Donk, J. Am. Chem. Soc., 2005, 127, 1126-1136. 73. G. Wohlfarth and G. Diekert, Reductive Dehalogenases, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999. 74. W. Schumacher, C. Holliger, A. J. B. Zehnder and W. R. Hagen, FEBS Lett., 1997, 409, 421-425. 75. A. Neumann, G. Wohlfarth and G. Diekert, J.Bacteriol., 1998,180, 4140-4145. 76. B. A. van de Pas, H. Smidt, W. R. Hagen, J. van der Oost, G. Schraa, A. J. M. Stams and W. M. de Vos, J. Biol. Chem., 1999, 274, 20287-20292. 77. J. Thibodeau, A. Gauthier, M. Duguay, R. Villemur, F. Lepine, P. Juteau and R. Beaudet, Appl. Environ. Microbiol., 2004, 70, 4532-4537. 78. Z. Studer, E. Stupperich, S. Vuilleumier and T. Leisinger, Eur. J. Biochem., 2001, 268, 2931-2938. 79. T. Vannelli, M. Messmer, A. Studer, S. Vuilleumier and T. Leisinger, Proc. Natl. Acad. Sei. USA, 1999, 96, 4615-4620. 80. A. Studer, S. Vuilleumier and T. Leisinger, J. Biochem., 1999, 264, 242-249. 81. M. Yamanishi, T. Labunska and R. Banerjee, J. Am. Chem. Soc., 2005, 127, 526-527. 82. K. Yamada, R. A. Gravel, T. Toraya and R. G. Matthews, Proc. Natl. Acad. Sei. USA, 2006, 103, 9476-9481.

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRINOID-DEPENDENT ENZYMES

111

83. P. J. H. Daas, R. W. Wassenaar, P. Willemsen, R. J. Theunissen, J. T. Keltjens, C. van der Drift and G. D. Vogels, J. Biol.Chem., 1996, 271, 22339-22345. 84. K. L. Brown and S. Peck-Siler, Inorg. Chem., 1988, 27, 3548-3555. 85. K. L. Brown and X. Zou, Inorg. Chem., 1991, 30, 4185-4191. 86. J. H. Mangum and K. G. Scrimgeour, Fed. Proc., 1962, 21, 242. 87. K. Fujii and F. M. Huennekens, J. Biol. Chem., 1974, 249, 6745-6753. 88. K. Fujii and F. M. Huennekens, Methionine Synthetase: Characterization of Protein Components and Mechanisms for Activation and Catalysis, in Biochemical Aspects of Nutrition, Ed. K. Yagi, Japan Scientific Societies Press, Tokyo, 1979. 89. D. E. Hall, T. C. Jordan-Starck, R. O. Loo, M. L. Ludwig and R. G. Matthews, Biochemistry, 2000, 39, 10711-10719. 90. D. LeClerc, A. Wilson, R. Dumas, C. Gafuik, D. Song, D. Watkins, Η. H. Q. Heng, J. M. Rommens, S. W. Scherer, D. S. Rosenblatt and R. A. Gravel, Proc. Natl. Acad. Sei. USA, 1998, 95, 3059-3064. 91. D. S. Rosenblatt, Inherited Disorders of Folate Transport and Metabolism, in The Metabolic and Molecular Bases of Inherited Disease, Ed. C. R. Scriver, A. L. Beaudet, W. S. Sly and D. Yalle, McGraw Hill, New York, 1995. 92. H. Olteanu and R. Banerjee, J. Biol. Chem., 2001, 276, 35558-35563. 93. P. J. H. Daas, W. R. Hagen, J. T. Keltjens, C. Van der Drift and G. D. Vogels, J. Biol. Chem., 1996, 271, 22346-22351. 94. R. Fischer, P. Gärtner, A. Yeliseev and R. K. Thauer, Arch. Microbiol., 1992, 158, 208-217. 95. A. Siebert, T. Schubert, T. Engelmann, S. Studenik and G. Diekert, Arch. Microbiol., 2005, 183, 378-384. 96. F. Kaufmann, G. Wohlfarth and G. Diekert, Eur. J. Biochem., 1998, 253, 706-711. 97. R. W. Wassenaar, P. J. H. Daas, W. J. Geerts, J. T. Keltjens and C. van der Drift, J. Bacterial., 1996, 178, 6937-6944. 98. G. Srinivasan, C. M. James and J. A. Krzycki, Science, 2002, 296, 1459-1462. 99. T. Kuzuyama, T. Hidaka, K. Kamigiri, S. Imaig and H. Seto, J. Antibiot., 1992, 45, 1812-1814. 100. H. Seto, T. Hidaka, T. Kuzuyama, S. Shibahara, T. Usui, O. Sakanaka and S. Imai, J. Antibiot., 1991, 44, 1286-1288. 101. R. D. Woodyer, G. Li, H. Zhao and W. A. van der Donk, Chem. Commun., 2007, 359-361. 102. P. A. Frey and R. H. Abeles, J. Biol. Chem., 1966, 241, 2732-2733. 103. J. Retey and D. Arigoni, Experientia, 1966, 22, 783-784. 104. P. A. Frey, Μ. Κ. Essenberg and R. Η. Abeles, J. Biol. Chem., 1967, 242, 5369-5377. 105. Ο. W. Wagner, H. A. Lee Jr., P. A. Frey and R. H. Abeles, J. Biol. Chem., 1966, 241, 1751-1762. 106. J. Retey, A. Umani-Rouchi, J. Seibl and D. Arigoni, Experientia, 1966, 22, 502-503. 107. R. G. Finke and B. P. Hay, Inorg. Chem., 1984, 23, 3041-3043. 108. W. Buckel, Β. Τ. Golding and C. Kratky, Chem. Eur. J., 2006, 12, 352-362.

Met. Ions Life Sei. 2009, 6, 53-114

112

MATTHEWS

109. P. Μ. Kozlowski, Τ. Kamachi, Τ. Toraya and Κ. Yoshizawa, Angew. Chem. Int. Ed., 2007, 46, 980-983. 110. W. Buckel, G. Bröker, Η. Bothe, A. Piere and Β. T. Golding, Glutamate Mutase and 2-Methyleneglutarate Mutase, in Chemistry and Biochemistry of BI2, Ed. R. Banerjee, John Wiley & Sons, New York, 1999. 111. H.-P. Chen and Ε. N. Marsh, Biochemistry, 1997, 36, 14939-14945. 112. R. Reitzer, K. Gruber, G. Jogi, U. G. Wagner, H. Bothe, W. Buckel and C. Kratky, Structure, 1999, 7, 891-902. 113. M. Tollinger, R. Konrat, Β. H. Hilbert, Ε. Ν. Marsh and Β. Kräutler, Structure, 1998, 6, 1021-1033. 114. Κ. Gruber, R. Reitzer and C. Kratky, Angew. Chem. Int. Ed. Engl., 2001, 40, 3377-3380. 115. H. W. Chih and Ε. N. Marsh, Biochemistry, 1999, 38, 13684-13691. 116. W. Buckel and Β. T. Golding, Chem. Soc. Rev., 1996, 25, 329-338. 117. Β. Beatrix, Ο. Zelder, D. Linder and W. Buckel, Eur. J. Biochem., 1994, 221, 101-109. 118. A. J. Pierik, D. Ciceri, R. F. Lopez, F. Kroll, G. Broker, B. Beatrix, W. Buckel and Β. T. Golding, Biochemistry, 2005, 44, 10541-10551. 119. F. Mancia, Ν. H. Keep, A. Nakagawa, P. F. Leadlay, S. McSweeney, B. Rasmussen, P. Boseck, O. Diat and P. R. Evans, Structure, 1996, 4, 339-350. 120. F. Mancia and P. R. Evans, Structure, 1998, 6, 711-720. 121. Μ. D. Ylasie and R. Banerjee, J. Am. Chem. Soc., 2003, 125, 5431-5435. 122. R. Padmakumar and R. Banerjee, Biochemistry, 1997, 36, 3713-3718. 123. S. Chowdhury and R. Banerjee, Biochemistry, 2000, 39, 7998-8006. 124. R. A. Kwiecien, Ι. V. Khavrutskii, D. G. Musaev, K. Morokuma, R. Banerjee and P. Paneth, J. Am. Chem. Soc., 2006, 128, 1287-1292. 125. S. O. Mansoorabadi, R. Padmakumar, N. Fazliddinova, M. Vlasie, R. Banerjee and G. H. Reed, Biochemistry, 2005, 44, 3153-3158. 126. A. Dybala-Defratyka, P. Paneth, R. Banerjee and D. G. Truhlar, Proc. Natl. Acad. Sei. USA, 2007, 104, 10774-10779. 127. N. A. Leal, S. D. Park, P. E. Kima and T. A. Bobik, J. Biol. Chem., 2003, 278, 9227-9234. 128. N. A. Leal, Η. Olteanu, R. Banerjee and Τ. A. Bobik, J. Biol. Chem., 2004, 279, 47536^7542. 129. Τ. A. Stich, M. Yamanishi, R. Banerjee and T. C. Brunold, J. Am. Chem. Soc., 2005, 127, 7660-7661. 130. M. Yamanishi, M. Ylasie and R. Banerjee, Trends Biochem. Sei., 2005, 30, 304-308. 131. N. Korotkova and Μ. E. Lidstrom, J. Biol. Chem., 2004, 279, 13652-13658. 132. C. M. Dobson, T. Wai, D. Leclerc, A. Wilson, X. Wu, C. Dore, T. Hudson, D. S. Rosenblatt and R. A. Gravel, Proc. Natl. Acad. Sei. USA, 2002, 99, 15554-15559. 133. D. Padovani, T. Labunska and R. Banerjee, J. Biol. Chem., 2006, 281,17838-17844. 134. D. Padovani and R. Banerjee, Biochemistry, 2006, 45, 9300-9306. 135. J. W. Vrijbloed, K. Zerbe-Burkhardt, A. Ratnatilleke, A. Grubelnik-Leiser and J. A. Robinson, J. Bacteriol., 1999, 181, 5600-5605.

Met. Ions Life Sei. 2009, 6, 53-114

COBALAMIN- AND CORRINOID-DEPENDENT ENZYMES

113

136. K. Zerbe-Burkhardt, A. Ratnatilleke, N. Philippon, A. Birch, A. Leiser, J. W. Yrijbloed, D. Hess, P. Hunziker and J. A. Robinson, J. Biol. Chem., 1998, 273, 6508-6517. 137. A. Ratnatilleke, J. W. Yrijbloed and J. A. Robinson, J. Biol. Chem., 1999, 274, 31679-31685. 138. D. Gani, D. O'Hagan, K. Reynolds and J. A. Robinson, J. Chem. Soc. Chem. Commun., 1985, 1002-1004. 139. W. Zhang and K. A. Reynolds, J. Bacteriol., 2001, 183, 2071-2080. 140. B. Zagalak, P. A. Frey, G. L. Karabatsos and R. H. Abeles, J. Biol. Chem., 1966, 241, 3028-3035. 141. M. Yamanishi, S. Yamada, H. Muguruma, Y. Murakami, T. Tobimatsu, A. Ishida, J. Yamauchi and T. Toraya, Biochemistry, 1998, 37, 4799^803. 142. N. Shibata, J. Masuda, T. Tobimatsu, T. Toraya, K. Suto, Y. Morimoto and N. Yasuoka, Structure, 1999, 7, 997-1008. 143. H. A. Lee Jr. and R. H. Abeles, J. Biol. Chem., 1963, 238, 2367-2373. 144. K. L. Schepler, W. R. Dunham, R. H. Sands, J. A. Gee and R. H. Abeles, Biochim. Biophys. Acta, 1975, 397, 510-518. 145. J. Masuda, N. Shibata, Y. Morimoto, T. Toraya and N. Yasuoka, Structure, 2000, 8, 775-788. 146. M. Yamanishi, H. Ide, Y. Murakami and T. Toraya, Biochemistry, 2005, 44, 2113-2118. 147. P. A. Schwartz, R. Lobrutto, G. H. Reed and P. A. Frey, Protein Sei., 2007, 16, 1157-1164. 148. Ο. W. Wagner, H. A. Lee Jr., P. A. Frey and R. H. Abeles, J. Biol. Chem., 1966, 249, 1751-1762. 149. S. O. Mansoorabadi, Ο. T. Magnusson, R. R. Poyner, P. A. Frey and G. H. Reed, Biochemistry, 2006, 45, 14362-14370. 150. T. Toraya, Chem. Rev., 2003, 103, 2095-2127. 151. P. A. Schwartz and P. A. Frey, Biochemistry, 2007, 46, 7293-7301. 152. N. Shibata, J. Masuda, Y. Morimoto, N. Yasuoka and T. Toraya, Biochemistry, 2002, 41, 12607-12617. 153. K. Mori and T. Toraya, Biochemistry, 1999, 38, 13170-13178. 154. Y. Bandarian and G. H. Reed, Ethanolamine Ammonia-Lyase, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, New York, 1999. 155. L. Sun and K. Warncke, Proteins: Structure, Function, andBioinformatics, 2006, 64, 308-319. 156. A. Abend, Y. Bandarian, R. Nitsche, E. Stupperich, J. Retey and G. H. Reed, Arch. Biochem. Biophys., 1999, 370, 138-141. 157. J. F. Boas, P. R. Hicks, J. R. Pilbrow and T. D. Smith, J. Chem. Soc, Faraday Trans. 2, 1978, 74, 417-430. 158. R. Lobrutto, V. Bandarian, Ο. T. Magnusson, X. Chen, V. L. Schramm and G. H. Reed, Biochemistry, 2001, 40, 9-14. 159. K. Warncke and A. S. Utada, J. Am. Chem. Soc., 2001, 123, 8564-8572. 160. C. G. Morley and T. C. Stadtman, Biochemistry, 1971, 10, 2325-2329. 161. C. H. Chang and P. A. Frey, J. Biol. Chem., 2000, 275, 106-114. 162. C. D. Morley and T. C. Stadtman, Biochemistry, 1972, 11, 600-605.

Met. Ions Life Sei. 2009, 6, 53-114

114

MATTHEWS

163. S. D. Wetmore, D. M . Smith and L. R a d o m , J. Am. Chem. Soc., 2001, 123, 8678-8689. 164. F. Berkovitch, E. Behshad, Κ . H. Tang, E. A. Enns, P. A. Frey and C. L. D r e n n a n , Proc. Natl. Acad. Sei. USA, 2004, 101, 15870-15875. 165. H . P. Chen, S. H . W u , Y. L. Lin, C. M . Chen and S. S. Tsay, J. Biol. Chem., 2001, 276, 44744-44750. 166. H . P. Chen, F. C. Hsui, L. Y. Lin, C. T. Ren and S. H . W u , Eur. J. Biochem., 2004, 271, 4 2 9 3 ^ 2 9 7 . 167. W. S. Beck, R. H . Abeles and W. G. Robinson, Biochem. Biophys. Res. Commun., 1966, 25, 421-425. 168. G. W. Ashley, G. Harris and J. Stubbe, J. Biol. Chem., 1986, 261, 3958-3964. 169. S. Booker, S. Licht, J. Broderick and J. Stubbe, Biochemistry, 1994, 33, 1267612685. 170. S. Licht and J. Stubbe, Mechanistic Investigations of Ribonucleotide Reductases, in Comprehensive Natural Products Chemistry, Ed. D . Poulter, Elsevier, Amsterdam, 1999. 171. S. Booker and J. Stubbe, Proc. Natl. Acad. Sei. USA, 1993, 90, 8352-8356. 172. M . D . Sintchak, G. Arjara, B. A. Kellogg, J. Stubbe and C. L. D r e n n a n , Nat. Struct. Biol., 2002, 9, 293-300. 173. C. C. Lawrence, G. J. Gerfen, V. Samano, R. Nitsche, M . Robins, J. Retey and J. Stubbe, J. Biol. Chem., 1999, 274, 7039-7042. 174. D. Chen, A. Abend, J. Stubbe and P. A. Frey, Biochemistry, 2003, 42, 45784584. 175. S. J. George, J. Seravalli and S. W. Ragsdale, J. Am. Chem. Soc., 2005, 127, 13500-13501. 176. P. A. Lindahl, J. Biol. Inorg. Chem., 2004, 9, 516-524. 177. M . L. Ludwig and R. G. Matthews, Ann.Rev. Biochem., 1997, 66, 269-313. 178. V. Bandarian, K . A. Pattridge, B. W. Lennon, D . P. Huddler, R. G. Matthews and M . L. Ludwig, Nat. Struct. Biol., 2002, 9, 53-56.

Met. Ions Life Sei. 2009, 6, 53-114

Met. Ions Life Sei. 2009, 6, 115-132

3 Nickel-Alkyl Bond Formation in the Active Site of Methyl-Coenzyme Μ Reductase Bernhard Jauna and Rudolf K. a

Thauerh

Organic Chemistry ΕΤΗΖ, ΕΤΗ Hönggerberg H C l E317, CH-8093 Zürich, Switzerland

b M a x Planck Institute for Terrestrial Microbiology, Karl-von-Frisch-Strasse, D-35043 Marburg, Germany

ABSTRACT 116 1. INTRODUCTION 116 2. NICKEL-CARBON BOND FORMATION IN FREE COENZYME F 4 3 0 119 3. NICKEL-ALKYL BOND FORMATION IN MCR UPON INACTIVATION WITH ALKYL HALIDES 120 3.1. 3-Sulfonatopropyl-Ni(III)F 4 3o Formation by Reaction of MCR re di with 3-Bromopropane Sulfonate 120 3.2. Methyl-Ni(III)F43o Formation in MCR r e d i by Reaction with Methyl Bromide 122 4. METHYL-NICKEL BOND FORMATION IN METHYLCOENZYME Μ REDUCTASE D U R I N G CATALYSIS? 123 4.1. Methylation of Ni(I)F 430 in the Active Site via Nucleophilic Substitution 125 4.2. Methylation of Ni(I)F 430 in the Active Site via Oxidative Addition 127 4.3. Catalytic Mechanism of Methyl-Coenzyme Μ Reductase Not Involving Metal-Carbon Bond Formation 128

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00115

116

JAUN and THAUER

5. OBSERVATIONS TO BE F O L L O W E D U P ACKNOWLEDGMENTS ABBREVIATIONS REFERENCES

129 129 129 130

ABSTRACT: Methyl-coenzyme Μ reductase (MCR) catalyzes the methane-forming step in methanogenic archaea and most probably also the methane-oxidizing step in methanotrophic archaea. The enzyme contains coenzyme F 4 3 0 as prosthetic group. F 4 3 0 is a nickel porphinoid that has to be in the reduced Ni(I) state for the enzyme to be active. The presently discussed catalytic mechanisms of MCR can in principle be divided into two basic models. In one model the key intermediate features a methyl-Ni(III) species being either formed in a nucleophilic substitution reaction or in an oxidative addition reaction. In the other model first the thioether sulfur of methyl-coenzyme Μ binds to the Ni(I), which subsequently results in the release of the methyl group as methyl radical leaving behind a Ni(II)-sulfur bond. The experimental evidence for and against a methyl-nickel intermediate is reviewed. KEYWORDS: anaerobic oxidation of methane · catalytic mechanism of methane formation · coenzyme F 4 3 0 · methane formation · methyl-coenzyme Μ reductase · methylnickel intermediate · nickel enzymes · nickel porphinoid

1.

INTRODUCTION

Methyl-coenzyme Μ reductase (MCR) is a nickel enzyme found in relatively high concentrations in all methanogenic and methanotrophic archaea [1^1]. The enzyme catalyzes the reversible reaction of methyl-coenzyme Μ (CH 3 -S-COM; 2-(methylthio)ethanesulfonate) with coenzyme Β (CoBSH; N-7-thioheptanoyl-O-phospho-L-threonine) to methane and the heterodisulfide (CoM-S-S-CoB) of coenzyme Μ (CoM-SH) and coenzyme Β (reaction 1) at a specific rate of approximately 100 μηιοί methane formed per min and per mg protein which is equivalent to a turnover number of 250 s" 1 [5], CH3-S-COM+COB-SH^ CH4 + CoM-S-S-CoB

AG 0 ' = —30kJ/mol

(1)

M C R has a molecular mass of 300 k D a and is composed of three different subunits in an α2β2Υ2 arrangement, which form two structurally and probably also functionally coupled active sites, each containing one coenzyme F 4 3 0 as prosthetic group. Coenzyme F 4 3 0 is a nickel hydroporphyrin of Met. Ions Life Sei. 2009, 6, 115-132

METHYL-COENZYME Μ REDUCTASE: Ni-C BOND FORMATION Coenzyme F430

H3C'

Ss

Ο

^^S03~

117

ß-face

Methyl-coenzyme Μ COO" Coenzyme Β

Figure 1. Structure of coenzyme F 430 with its nickel axially coordinated by the carboxamide group of glutamine 147 in the active site of methyl-coenzyme Μ reductase from Methanothermobacter marburgensis. View onto the ß-face of the corphin ring. Also shown are the structures of the substrates methyl-coenzyme Μ and of coenzyme B.

unique structure (Figure 1), corphin being the name proposed by Eschenmoser for this class of tetrapyrroles [6]. F 4 3 0 has to be in the reduced Ni(I) state for the enzyme to be active and it has been proposed that also the Ni(III) state is involved in the catalytic cycle [7-9]. The pentamethyl ester of coenzyme F 4 3 0 (F 4 3 0 M) shows a redox potential E°' of nearly - 0 . 6 V for the Ni(II)F 4 3 0 M/Ni(I)F 4 3 0 M couple (in acetonitrile) [10] and one of nearly + 1.6 V for the Ni(III)F 4 3 0 M/Ni(II)F 4 3 0 M couple (in acetonitrile) [11]. Compared to the normal hydrogen electrode (NHE) the first redox potential is lower than that of the hydrogen electrode in water at p H 7 (—0.42 V) and the second redox potential is higher than that of the oxygen electrode in water at p H 7 (0.81 V) explaining why only Ni(II)F 4 3 0 is stable in water. Ni(II)F 4 3 0 can be reduced to Ni(I)F 4 3 0 in water at a mid potential of approximately —600 mV but only if the proton concentration is kept below p H 7. Consistently, in M C R the nickel corphin is buried deeply Met. Ions Life Sei. 2009, 6, 115-132

118

JAUN and THAUER

within the protein in a water free hydrophobic pocket accessible from the outside only via a narrow 50 A long channel as revealed by the crystal structures of inactive Ni(II) forms [12-15]. The redox potentials of F 4 3 0 in its MCR-bound form are not known. However, that of the Ni(II)F43o/Ni(I)F 4 3o couple is probably very close to that of the free cofactor as deduced from the similar UV-visible spectra of free and MCR-bound Ni(II)F 43 o and of the similar UV-visible and EPR spectra of free- and MCR-bound Ni(I)F 4 3 0 . In contrast, the redox potential of the Ni(III)F43o/Ni(II)F 4 3o couple of free- and MCR-bound F430 are predicted to be significantly different. Neither the UV-visible nor the EPR spectrum of free Ni(III)F 4 3oM show similarity with the spectra of the different MCR-bound forms of coenzyme F 4 3 0 that are formally in the Ni(III) valence state [1]. Active M C R is referred to as MCR r e di and its Ni(I)-derived EPR signal as the MCR r e di signal. In Ni(I)F 4 3 0 and in MCR r e ( n the unpaired electron resides mainly in the d X 2 _ y 2 orbital, the nickel dZ2 orbital being filled with two electrons. MCR r e ( n can be converted to other EPR active but enzymatically inactive forms which are distinguished via differences in their Ni(I)-[d 9 ] or Ni(III)-[d 7 ] EPR spectra: M C R r e d l / r e d 2 (Ni(I)) by reaction of M C R r e d l with coenzyme Μ and coenzyme Β (reaction 2) [16,17]; MCR o x i(Ni(III)) by reaction of MCR r e d i/2 with polysulfide (reaction 3) [18]; M C R o x i back to MCR r e [ |i by reduction with titanium(III)citrate [Ti(III)] (reaction 4) [5]; MCRps (Ni(III)) by reaction of M C R r e d i with 3-bromopropane sulfonate (BPS) (reaction 5) [19]; M C R P S back to M C R r e d l by reaction with sulfide (reaction 6) [20]; MCR M e (Ni(III)) by reaction of M C R r e d i with methyl bromide (BrMe) or methyl iodide (reaction 7) [21,22]; and M C R M e back to MCR re< ii by reaction with coenzyme Μ (reaction 8) [22]. (In the literature M C R P S is also referred to as MCR B ps and M C R M e as MCRßrMe)· MCRredl + CoM-SH + CoB-SH ^ M C R r e d l / 2 MCRredi/2 + polysulfide -s· MCR o x i MCRoxi + Ti(III)

MCRredi

(at p H 7)

(2)

(at p H 7)

(3)

(at p H 10)

(4)

MCRredi + 3-bromopropane sulfonate —>· MCRps + Br

(at p H 7)

MCRps +

(at p H 10)

MCRredi +3-mercaptopropane sulfonate

MCRredi + BrCH 3

M C R M e + Br"

Met. Ions Life Sei. 2009, 6, 115-132

(at p H 7)

(5) (6) (7)

METHYL-COENZYME Μ REDUCTASE: Ni-C BOND FORMATION

MCRjvie + CoM-SH —> MCR r edi + CH3-S-C0M

(at p H 7)

119 (8)

In the presence of coenzyme Μ (which is an inhibitor of MCR) and coenzyme Β maximally 50% of the MCR r e di signal is converted into the MCR re d2 signal indicating half-of-the-sites reactivity [17]. The MCR re d2 signal is composed of an axial component (MCR re d2a) and a strongly orthorhombic component (MCR re d2 r )· The two forms are in a temperature dependent equilibrium with MCR r e di [23]. In MCR r e di, MCRoxi, M C R P S and M C R M e the nickel of F 4 3 0 is axially ligated from the α-face (Figure 1) by the oxygen of the carboxamide group of glutaminel47 a ' c " [1,9], This ligand is displaced in MCR re d2 [24], The upper axial ligand in MCR o x i is the sulfur of the thiolate group of coenzyme Μ [25], in MCRps it is a 3-sulfonatopropyl group [26] and in M C R M e a methyl group [21,22]. In MCR re d2a it is a hydride and in MCR re d2r a hydrogen and a sulfur, presumably as the result on an oxidative addition of the S-H bond of coenzyme Μ to Ni(I)F 4 3 0 giving a (S)(H)Ni(III)F 4 3 0 species [24]. In MCR r e di the ß-axial coordination site appears to be unoccupied [27]. The properties on M C R and of its nickel corphin cofactor coenzyme F 4 3 0 have been extensively reviewed by the authors two years ago in Volume 2 of "Metal Ions in Life Sciences" [1]. This allows the present chapter to completely focus on what is known about metal-carbon bonds in M C R and its cofactor.

2.

NICKEL-CARBON BOND FORMATION IN FREE COENZYME F43O

Ni(II)F 4 3 0 M (M for pentamethyl ester) in dry dichloromethane, chloroform or acetonitrile is in a diamagnetic d 8 low spin state. In this state the nickel in F43q is square planar tetracoordinated, but has a pronounced tendency to bind additional ligands in the axial positions such as chloride (log K j = 5.4; log K2 = 3.7), imidazole (log^T; = 2.7; log K2 = 2.2) or water. Penta- and hexacoordination is associated with a change from low spin to high spin, a shift of the absorbance band at 422 nm (ε = 21 m M _ 1 c m _ 1 ) by a few nm and a gain in band intensity [1]. In water Ni(II)F 4 3 0 has an absorbance maximum at 430 nm and an extinction coefficient of 23 m M _ 1 c m _ 1 . When methyl iodide is allowed to react with Ni(I)F 4 3 0 M in a 1:2 stoichiometric ratio at low temperature, the color of the solution changes from green Ni(I)F 4 3 0 M to brown-orange within seconds without methane being formed. When acid is added, the color changes to that of Ni(II)F 4 3 0 M and more than 80% of the theoretical amount of methane is generated. Addition of deuterated acid gives over 85% CH 3 D. This experiment proves that an Met. Ions Life Sei. 2009, 6, 115-132

120

JAUN and THAUER

intermediate is formed, which can be dissociated to Ni(II)F 4 3 0 M and methane by protonation [28-30]. Its properties are consistent with those expected for a methyl-Ni(II) derivative with an axial nickel-carbon bond. CD 3 -Ni(II)F 4 3 0 M and CH 3 -Ni(II)F 4 3 0 M were independently generated at low temperature by stoichiometric reaction with (CD 3 ) 2 Mg and (CH 3 ) 2 Mg, respectively, and their structure was verified by 2 H and N M R spectroscopy [30]. The dissociation energy for the Ni(II)-C bond was estimated to be of the order of 80kJ/mol [31], Formation of methyl-Ni(III)F 43 oM as an intermediate is discussed for the reaction of Ni(I)F 430 M with methyl iodide to give methyl-Ni(II)F 430 M. However, all attempts to demonstrate the existence of this compound outside the enzyme have failed until now although theoretical modeling (DFT) predicts that methyl-Ni(III) should be stable [32]. One reason probably is that any methyl-Ni(III)F 430 M formed will immediately react with excess Ni(I)F 430 M to methyl-Ni(II)F 430 M and Ni(II)F 430 M. The redox potential of the methyl-Ni(III)F 43 oM/methyl-Ni(II)F 430 M couple has been estimated to be near + 0.45 V [1] which is one Volt more positive than the redox potential E°'=—0.6 V of the Ni(II)F 430 M/Ni(I)F 430 M couple [10], Coenzyme F 4 3 0 has been shown to catalyze the reductive dehalogenation of CC14, CHC13, CH2C12, and CH3C1 with Ti(III) in aqueous solution. Trichloromethyl-Ni(III)F 430 , dichloromethyl-Ni(III)F 430 , monochloromethyl-Ni(III)F 430 , and methyl-Ni(III)F 430 , respectively, have been proposed to be likely intermediates [33]. The reaction of Ni(I)-octaethylisobacteriochlorin (a structural cousin of F 430 ) with alkyl halides has been investigated in detail and has been interpreted as proceeding via alkyl-Ni(III) species that undergo reduction to the alkyl-Ni(II), followed by protonolysis yielding the alkane [34].

3. 3.1.

NICKEL-ALKYL BOND FORMATION IN MCR UPON INACTIVATION WITH ALKYL HALIDES 3-Sulfonatopropyl-Ni(lll)F43o Formation by Reaction of MCRredi with 3-Bromopropane Sulfonate

3-Bromopropane sulfonate is the most potent inhibitor of MCR known to date. The methyl-coenzyme Μ analogue binds in the active site of MCR re( n with a nanomolar inhibition constant ( Κ w h e r e it reacts with the Ni(I) of F 4 3 0 to give MCRps which has a nickel based, axial X-band continuouswave (CW) EPR spectrum [19]. In view of the known reactivity of free Ni(I)F 430 towards alkyl halides [33] and the results of D F T calculations [32] the possibility was considered that the MCR P S species might Met. Ions Life Sei. 2009, 6, 115-132

M E T H Y L - C O E N Z Y M E Μ R E D U C T A S E : Ni-C B O N D F O R M A T I O N

121

correspond to an alkyl-Ni(III)F 430 derivative resulting from reaction of Ni(T)F430 with BPS to give a bromide ion and ~0 3 S(CH 2 ) 3 -Ni(III)F 43 o in the active site of MCR through what is formally an oxidative addition of BPS to Ni(I): -0 3 S(CH 2 ) 3 Br + Ni(I)F 430 ^ - O 3 S ( C H 2 ) 3 - N i ( m ) F 4 3 0 + Br"

(9)

To test this hypothesis, Hinderberger et al. [26] synthesized [3-13C]-3bromopropane sulfonate, then treated the active enzyme with it, and investigated the resulting 13 C-MCRps samples by using high frequency CWEPR (W-band, microwave frequency ca. 94 GHz), high-resolution pulse electron nuclear double resonance (ENDOR), and hyperfine sublevel correlation (HYSCORE) spectroscopy. As a control, MCR P S samples induced by reaction of MCR re( n with unlabeled 3-bromopropane sulfonate were also studied. To further clarify the coordination geometry, the proton signals derived from the two γ protons and the two β protons of the propane sulfonate moiety were investigated. The interpretation of the spectroscopic results in terms of the binding situation in MCR P S is shown in Figure 2A. In MCRps there is a bond between the nickel center of F 4 3 0 and the Cj atom of the propane sulfonate residue. Approximately 7% of the spin

OGIn«'147 Figure 2. Proposed coordination of the nickel in the active site of M C R after reaction of M C R r e d l with 3-bromopropane sulfonate to form M C R P S . (A) The p 2 orbital of the proposed near sp 2 -hybrid carbon atom has been drawn to indicate the direction of the Ni-Cy bond. The asterix marks the position of 13 C labeling. Regarding the detailed ligation of Ni see Figure 1. (B) Newman projection along the Cy-Cß bonding. For clarity the Ni-Cy has been elongated, the p 2 orbital of Cy has been drawn in gray and the γ protons have been omitted. R = -CH 2 -SO^\ Reproduced from Hinderberger et al. [26] with permission of Wiley-VCH, copyright (2006). Met. Ions Life Sei. 2009, 6, 115-132

122

JAUN and THAUER

population resides on the Cy atom and 75% in the Ni d X 2 _ y 2 orbital in close agreement with the results of D F T calculations on a CH 3 -Ni(III)F 4 3o model structure [32]. The Ni-Cy bond is slightly tilted (ca. 20°) away from the g z principle axis direction. Hybridization of the Cj atom is closer to sp 2 than sp3 resulting in a flattened bonding geometry at Cj: the two protons Ηγΐ and Ηγ2 are situated in (or close to) a nodal plane of the nonhybridized p z orbital of the Cj atom that hence has very little overlap with the hydrogen Is orbitals [26]. The Newman projection of the conformation at the Cß atom as deduced from proton coupling is shown in Figure 2B. It was noted by the authors that the EPR data do not allow the determination of the total number of electrons in the system, and that therefore the assignment of a formal alkyl-Ni(III) state rests on the assumption that after the reaction of MCR re di with 3-bromopropane sulfonate a bromide ion is generated as a reaction product [26]. In the meantime it was found that upon acid denaturation of MCR P S the C-Ni bond is protonolized yielding propane sulfonate [35]. This product was also identified when MCR P S was re-reduced with Ti(III) to MCR re di· MCR P S can also be converted to active MCR re di by treatment with sodium borohydride [20]. Upon addition of thiols to MCR P S at pH 10 the active MCR re di state is regenerated with the concomitant formation of the respective thioether [20,35,36]. All these findings point to the presence of an alkyl-Ni(III) bond in MCR P S . MCR re di was found to react with 4-bromobutyrate and other ω-brominated carboxylic acids with carbon chain length up to 16. All of these compounds give rise to an alkyl-Ni intermediate with an EPR signal similar to that of the MCRps species. Reaction of the alkyl-Ni adduct, formed from brominated acids with eight or fewer total carbons, with CoM-SH as nucleophile at pH 10 results in the formation of a thioether coupled to the regeneration of the active MCR re di state [22,36]. Reactivation is highest with the smallest free thiol HS . Interestingly, MCR P S can also be reductively activated with analogues of HS-CoB such as HS-CoB 8 (N-8-thiooctanoyl-O-phospho-L-threonine) and HS-CoBg (N-9-thiononanoyl-O-phospho-L-threonine) but not with coenzyme Β (N-7-thioheptanoyl-O-phospho-L-threonine) [20]. MCR re di alkylated with 4-bromobutyrate was surprisingly found to undergo "self-reactivation" at pH 10 to generate the MCR re di state and the ester between 4-bromobutyrate and 4-hydroxybutyrate [36].

3.2.

Methyl-Ni(lll)F 4 3o Formation in M C R r e d i by Reaction with Methyl Bromide

The active enzyme MCR re di reacts rapidly and irreversibly with simple alkyl halides such as chloroform. The latter has been routinely used to quench the Met. Ions Life Sei. 2009, 6, 115-132

METHYL-COENZYME Μ REDUCTASE: Ni-C BOND FORMATION

123

Ni(I) state yielding an EPR silent M C R . Also the inhibitor 2-bromoethanesulfonate (BES) quenches the Ni(I) state with the concomitant formation of a protein based radical [19]. To the contrary, reaction of BPS with M C R r e d i yields M C R P S with an alkyl-Ni(III) bond. Apparently it needs a methyl-coenzyme Μ related structure for the alkyl-Ni(III) derivative to be stable. It was therefore surprising that upon reaction of methyl bromide with MCR r e di the enzyme was converted into a form M C R M e with a n UV-visible and EPR spectrum very similar to that of M C R P S [21,22]. Complete conversion of the MCR r e di signal into the MCRMe signal was achieved by incubation of MCR r e di in 50 m M Tris/HCl p H 9 . 0 with an approximately 50 fold excess of methyl bromide or methyl iodide (added as saturated aqueous solution). The signal decays to an EPR silent form with a tij 2 of 20 min at room temperature. Continuous wave and pulse electron nuclear double resonance and hyperfine sublevel correlation spectroscopy of the enzyme alkylated with CH 3 Br, 1 3 CH 3 Br or CD 3 Br (or the respective iodides) revealed that the methyl group from the methyl halide becomes directly bound to the nickel ion after reaction. From the EPR data it was calculated that the nickel has a spin population of 81 % which resides in the Ni(III) d X 2 _ y 2 orbital. The most reasonable picture of the H 3 C-Ni(III) coordination is via an interaction of the filled nickel dZ2 orbital with the empty orbital from the cation C H ^ , indicating that the halogen atom is lost as halogenide ion [21,22]. When the CH 3 -Ni(III) species was reacted with CoM-SH, active MCR r e di was regenerated with a rate constant of 0.044 s - 1 forming methyl-coenzyme Μ as demonstrated by mass spectrometry [22]. The puzzling observation that the Ni(I) ion of MCR r e di and the Ni(III) ion of M C R M e a n d M C R P S both have d X 2 _ y 2 ground configurations, even though the formal electron count on Ni differs by two, is made understandable through the cartoon bonding scheme for methyl-Ni(III) in Figure 3 [22]. The scheme explains why the odd-electron orbital is largely unperturbed by the oxidative addition reaction and why the transfer of spin to CH 3 is minimal: the Ni-C two electron bond does not involve the odd electron, which in effect is a spectator to the reaction! The change in g-values can be assigned primarily to changes in the d-orbital splitting [22].

4.

METHYL-NICKEL BOND FORMATION IN METHYLCOENZYME Μ REDUCTASE DURING CATALYSIS?

Different catalytic mechanisms are presently discussed for the reaction of methyl-coenzyme Μ and coenzyme Β to methane and the heterodisulfide (reaction 1). They are based on density function theory calculations, on the Met. Ions Life Sei. 2009, 6, 115-132

124

JAUN and THAUER Qxidctive Addition .\Ϊ(ΐ)

iyni)-Vfc

CH, +

Η

N

«Ii \


+ 1 V and of the Ni(II)F43o/Ni(I)F43o couple o f - 0 . 6 V, respectively. Met. Ions Life Sei. 2009, 6, 115-132

METHYL-COENZYME Μ REDUCTASE: Ni-C BOND FORMATION

4.1.

125

Methylation of Ni(l)F 4 3 0 in the Active Site via Nucleophilic Substitution

It is assumed in this mechanism that the first step in the catalytic cycle is the reaction of the methyl group of methyl-coenzyme Μ with Ni(I) in a nucleophilic substitution reaction (Figure 4A) yielding methyl-Ni(III) and coenzyme Μ (reaction 10). Methyl-Ni(III) is then thought to be protonolyzed to methane and Ni(III) in an electrophilic substitution reaction, the proton donor most probably being coenzyme Β (reaction 11). Subsequently Ni(III) oxidizes the thiol group of coenzyme Μ to the thiyl radical (-S') (reaction 12) which in turn reacts with coenzyme Β to the disulfide radical anion (-S'-S-)~ with a two center-three electron (2c-3e) bond (reaction 13). The latter is a strong reductant capable of re-reducing Ni(II)F 43 o to Ni(I)F 43 o (reaction 14) thus closing the catalytic cycle [12]. CH3-S-C0M + Ni(I)F 43 o — CH 3 -Ni(III)F 4 3o + CoB-S~

(10)

CH 3 -Ni(III)F 4 3 o + C o B - S H ^ CH 4 + Ni(III)F 4 3 0 + CoB-S~

(11)

Ni(III)F 4 3o + CoM-S~ ^ Ni(II)F 4 3 0 + CoM-S'

(12)

CoM-S· + CoB-S~ ^ CoM-S· -S-CoB"

(13)

CoM-S· -S-CoB~ + Ni(II)F 4 3 0 ^ C0M-S-S-C0B + Ni(I)F 4 3 0

(14)

This mechanism is consistent with the finding that MCR-catalyzed ethylcoenzyme Μ reduction proceeds with inversion of configuration at CI of its ethyl group [40], since reaction (10) is predicted to proceed with inversion and reaction (11) with retention of configuration. Also the finding that M C R catalyzes the reduction of ethyl-coenzyme Μ with less than 1 % of the catalytic efficiency of methyl-coenzyme Μ reduction is in agreement with a nucleophilic substitution as first step, which predicts that the electrophilic attack of Ni(I) on CI of the ethyl group should be sterically hindered [19] (Figure 4A). But there are also problems. The formation of CH 3 -Ni(III)F 4 3 o from CH 3 S-CoM and Ni(I)F 4 3 0 has no yet been shown, neither with free Ni(I)F 4 3 0 nor with Ni(I)F 4 3 0 bound in MCR r e di· Based on D F T calculations, Siegbahn et al. concluded that reaction (10) would be energetically too unfavorable [41,42]. In the case of free Ni(I)F 4 3 0 this is intuitively understandable since the CH 3 -S bond (approximately 300kJ/mol) is much stronger than the CH 3 -Ni bond (80kJ/mol) [31]. In agreement with this prediction methyl-coenzyme Μ is readily formed from M C R M e (generated from Met. Ions Life Sei. 2009, 6, 115-132

126

JAUN and THAUER

Figure 4. Optimal position of methyl-coenzyme Μ in the active site of methylcoenzyme Μ reductase assuming (A) methylation of Ni(I) via nucleophilic substitution [12,15], (B) methylation of the Ni(I) center via oxidative addition [47], and (C) reaction of the Ni(I) center with the sulfur of the thioether bond of methylcoenzyme Μ [41,42]. The long aliphatic arm of coenzyme Β (Figure 1) can reach into the channel only to the extent where its terminal thiol group is in a distance of 8 A from Ni(I).

M C R r e d i and methyl bromide; reaction 7) u p o n addition of coenzyme Μ [22], F o r methyl-coenzyme Μ to react with free Ni(I)F 4 3 o the C H 3 - S b o n d in methyl-coenzyme Μ has to be activated by p r o t o n a t i o n (or methylation) of the sulfur [1]. But, there is n o indication f r o m the crystal structure of M C R that an activation of the C H 3 - S b o n d by p r o t o n a t i o n occurs. The two conserved tyrosines in the active site, which are potential p r o t o n donors, are not positioned appropriately [12-15]. It has to be kept in mind, however, that the crystal structure is that of the inactive M C R in the Ni(II) oxidation state and that there is evidence for substantial c o n f o r m a t i o n a l changes in the active site u p o n binding of coenzyme Β to the active enzyme [17,24], Methyl-coenzyme Μ reacts with Ni(I)F 4 3o in M C R r e d i only in the presence of coenzyme B, which u p o n binding to the active enzyme induces a c o n f o r m a t i o n a l change forcing methyl-coenzyme Μ and Ni(I) of the prosthetic g r o u p to interact. This enforced interaction is deduced f r o m the finding that inactivation of M C R r e ( n by b r o m o e t h a n e s u l f o n a t e and by other suicide inhibitors is dependent on the presence of coenzyme Β [19] and that the substrate analogue coenzyme Μ binds with its thiol g r o u p to Ni(I) in M C R r e d i o n l y in the presence of coenzyme Β as revealed by E P R spectroscopy [43,44]. In single turnover experiments m e t h a n e f o r m a t i o n f r o m methyl-coenzyme Μ was f o u n d to be dependent on coenzyme Β [45]. Considering the involvement of methyl-coenzyme Μ reductase in anaerobic m e t h a n e oxidation with sulfate, the m e t h a n e - f o r m i n g step in the Met. Ions Life Sei. 2009, 6, 115-132

METHYL-COENZYME Μ REDUCTASE: Ni-C BOND FORMATION

127

catalytic cycle must be reversible (reaction 11). Indeed, the reaction of methane, either end-on or side-on, with Ni(III)F 4 3o as described for the activation of C-H bonds by other high-valent metal complexes can be envisaged [46].

4.2.

Methylation of Ni(l)F 4 3 0 in the Active Site via Oxidative Addition

This mechanism, which is mainly based on D F T calculations, also involves methyl-nickel bond formation but is otherwise quite different [47]. The cycle begins with the protonation of Ni(I)F 4 3 0 either on Ni(I) or on the C-ring nitrogen of the corphin ring to yield N i ( I ) H + F 4 3 0 (reaction 15). N i ( I ) H + F 4 3 0 is predicted to oxidatively add CH3-S-C0M (reaction 16) to give a 4-coordinate Ni center (Figure 4B) with the nickel above the plane of the corphin ring, to which the nickel is now coordinated only to two of the four nitrogen atoms of the corphin ring. The two other ligands are the methyl group and S C H 2 C H 2 S O f " . By binding of CoB-SH, the Ni (and the attached C H 3 and S C H 2 C H 2 S O f " ligands) moves towards the "S-CoB (deprotonated HSCoB) cofactor allowing a 2c-3e interaction to form between the two sulfur atoms and allowing the transfer of two electrons from ~S-CoB to the Ni center. A Ni(II)-coordinated heterodisulfide radical anion is an intermediate. The release of the heterodisulfide yields C H 3 - N i ( I ) H + F 4 3 0 = CH 3 -Ni(III)H~F 4 3 0 (reaction 17) from which methane is reductively eliminated (reaction 18) Ni(I)F 43 o + CoB-SH ^ N i ( I ) H + F 4 3 o + CoB-S~ N i ( I ) H + F 4 3 0 + CH3-S-C0M ^

(CH3-)(CoM-S-)Ni(III)H+F 4 3o

(CH3-)(CoM-S-)Ni(III)H+F 4 3o + CoB-S~ ^ CH3-Ni(I)H + F 4 3 0 + C0M-S-S-C0B CH 3 -Ni(I)H + F 4 3o ^ CH3-Ni(III)irF 4 3o ^ C H 4 + Ni(I)F 4 3 0

(15) (16)

(17) (18)

The oxidative addition (reaction 16) and the reductive elimination (reaction 18) both occur via retention of stereochemistry, which appears to exclude this mechanism since, as indicated above, it has been shown that the reduction of ethyl-coenzyme Μ with coenzyme Β proceeds with inversion of stereochemistry at CI of the ethyl group [40]. However, the catalytic efficiency of M C R with ethyl-coenzyme Μ is less than 1 % of that with methyl coenzyme Μ [17]. Therefore, it is argued by Duin and McKee [47] that with Met. Ions Life Sei. 2009, 6, 115-132

JAUN and THAUER

128

such a big difference it is possible that the mechanism followed for the reaction of ethyl-coenzyme Μ is very different from the one followed for methyl-coenzyme M. In line with this mechanism is the recent finding that MCR re d2 r is formed from MCR r e di by the coenzyme Β induced oxidative addition of the H-S bond of coenzyme Μ to Ni(I)F 4 3 0 to give a Ni(III) center, to which a hydride and the sulfur of "SCH 2 CH 2 SOf" are ligated [24],

4.3.

Catalytic Mechanism of Methyl-Coenzyme Μ Reductase Not Involving Metal-Carbon Bond Formation

This mechanism, which is based on predictions of D F T calculations, is generally referred to as the Siegbahn mechanism [41,42]. It starts with the reaction of the thioether sulfur of methyl-coenzyme Μ with the Ni(I) center in MCRredi (Figure 4C) yielding CoM-S-Ni(II)F 4 3 0 and a methyl radical (CH 5) (reaction 19) which subsequently reacts with coenzyme Β forming methane and the thiyl radical (-S') of coenzyme Β (reaction 20). The thiyl radical subsequently reacts with the sulfur attached to Ni(II) resulting in a Ni(II)-coordinated heterodisulfide radical anion (-S-'S-) with a two center three electron (2c-3e) bond, from which the third electron is transferred to the Ni(II) center yielding Ni(I)F 4 3 0 and CoM-S-S-CoB (reaction 21). C H 3 - S - C 0 M + Ni(I)F 43 o — CoM-S-Ni(II)F 4 3o + C H j

(19)

CH · + CoB-SH ^ CH 4 + CoB-S ·

(20)

•S-C0B + CoM-S-Ni(II)F 4 3o^Ni(I)F 4 3 o + CoM-S-S-CoB

(21)

Reaction (19) is calculated to have a much lower energy barrier than reaction (10), which is the main reason for this mechanistic proposal. In their second paper Siegbahn et al. [41,42] propose that the release of the methyl radical (reaction 19) and its reaction with coenzyme Β (reaction 20) proceed in a concerted manner such that the transfer of the methyl group is associated with an inversion of stereo configuration. The mechanism does not explain why ethyl-coenzyme Μ is so much more slowly reduced to ethane than methyl-coenzyme Μ to methane and why allyl-coenzyme Μ is not at all reduced although allyl-coenzyme Μ is bound in the active site [19]. The argument is that the ethyl radical and the allyl radicals are much better leaving groups than the methyl radical. Steric reasons for why ethyl-coenzyme Μ and allyl-coenzyme Μ are reduced much Met. Ions Life Sei. 2009, 6, 115-132

METHYL-COENZYME Μ REDUCTASE: Ni-C BOND FORMATION

129

more slowly or not at all are not evident from the MCR crystal structure [12]. In the active site of MCR there is enough space for the ethyl and the allyl rest to be positioned like the methyl group of methyl-coenzyme Μ when being reduced by the Ni(I) center (Figure 4C).

5.

OBSERVATIONS TO BE FOLLOWED UP

Coenzyme Μ has been shown to bind with its thiol(ate) sulfur to the MCR Ni center in the Ni(III) (MCR o x l ) state [25,48,49] and Ni(II) (MCR oxl _ si i ent ) state [12], as well as in the Ni(I) (MCR r e d 2 r ) state [24,43,44], which might have to be formally described as (H)(CoMS)Ni(III) [23,24], However, the thioether sulfur of methyl-coenzyme Μ is expected to show much weaker coordination than the thiol(ate). Recent EPR results on the interaction of methyl-coenzyme Μ with the Ni(I) center in the MCR r e ( n m state (methyl-coenzyme Μ bound in the active MCR re( n form) point to an extremely weak coordination with a SNi distance of >3.5 A (unpublished results). Apparently, only the binding of coenzyme B, which seems to induce a conformational change and rearrangement of the coordination sphere, triggers the crucial step, namely the breaking of the CH 3 -S bond in the substrate. When methyl-coenzyme Μ is added to MCR r e ( n, the EPR signal changes only slightly. There is also an only minor change when additionally coenzyme Β is added despite the fact that now methane is actively formed [16]. The MCR re di signal is quenched, however, when MCR re( n is supplemented with both methyl-coenzyme Μ and HS-CoB 6 (N-6-thiohexanoyl-O-phospho-Lthreonine) (unpublished results). The coenzyme Β homologue has been shown to induce, as coenzyme B, the conformational changes but without being a substrate [17]. The quenching of the signal indicates that methyl-coenzyme Μ reacted with the Ni(I). A kinetic analysis of the inactivation reaction could reveal which of the considered mechanisms are to be excluded.

ACKNOWLEDGMENTS This work was supported by the Max Planck Gesellschaft, by the Fonds der Chemischen Industrie, and the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung.

ABBREVIATIONS 2c-3e bond BES

two-center three-electron bond 2-bromoethanesulfonate Met. Ions Life Sei. 2009, 6, 115-132

JAUN and T H A U E R

130 BPS BrMe CH3-S-C0M C0M-S-S-C0B CoB-SH CoM-SH

cw DFT ENDOR F430

F430M HYSCORE MCR NHE

3-bromopropane sulfonate methyl b r o m i d e methyl-coenzyme M; 2-(methylthio)ethane sulfonate heterodisulfide of coenzyme Μ a n d coenzyme Β c o e n z y m e B; N - 7 - t h i o h e p t a n o y l - O - p h o s p h o L-threonine coenzyme M, 2-thioethanesulfonate c o n t i n u o u s wave density f u n c t i o n t h e o r y electron nuclear d o u b l e r e s o n a n c e nickel p o r p h i n coenzyme F 4 3 0 p e n t a m e t h y l e s t e r of F 4 3 0 h y p e r f i n e sublevel correlation methyl-coenzyme Μ reductase n o r m a l h y d r o g e n electrode

REFERENCES 1. B. Jaun and R. K. Thauer, in Nickel and Its Surprising Impact in Nature, Vol. 2 of Metal Ions in Life Sciences, Ed. A. Sigel, H. Sigel and R. K. O. Sigel, John Wiley & Sons, Ltd, Chichester, U K , 2007, pp. 323-356. 2. S. Shima and R. K. Thauer, Curr. Opin. Microbiol., 2005, 8, 643-648. 3. R. K. Thauer and S. Shima, in Archaea, Evolution, Physiology and Molecuar Biology, Ed. R. Garrett and H.-P. Klenk, Blackwell Publishing, Inc., Maiden, USA, 2007, pp. 275-283. 4. R. Thauer and S. Shima, Ann. NY Acad. Sei., 2008, 1125, 158-170. 5. Μ. Goubeaud, G. Schreiner and R. K. Thauer, Eur. J. Biochem., 1997, 243, 110-114. 6. A. Eschenmoser, Ann. NY Acad. Sei., 1986, 471, 108-129. 7. R. K. Thauer, Microbiology, 1998, 144, 2377-2406. 8. U. Ermler, Dalton Transactions, 2005, 3451-3458. 9. E. C. Duin, in Tetrapyrroles, Ed M. J. Warren and A. G. Smith, Landes Bioscience and Springer Science + Business Media, 2007. 10. R. Piskorski and B. Jaun, J. Am. Chem. Soc., 2003, 125, 13120-13125. 11. B. Jaun, in Properties of Metal Alkyl Derivatives, Vol. 29 of Metal Ions in Biological Systems, Ed. H. Sigel and A. Sigel, Marcel Dekker, New York, 1993, pp. 287-337. 12. U. Ermler, W. Grabarse, S. Shima, M. Goubeaud and R. K. Thauer, Science, 1997, 278, 1457-1462. 13. W. Grabarse, F. Mahlert, S. Shima, R. K. Thauer and U. Ermler, J. Mol. Biol., 2000, 303, 329-344. 14. W. Grabarse, F. Mahlert, Ε. C. Duin, M. Goubeaud, S. Shima, R. K. Thauer, V. Lamzin and U. Ermler, J. Mol. Biol., 2001, 309, 315-330.

Met. Ions Life Sei. 2009, 6, 115-132

METHYL-COENZYME Μ R E D U C T A S E : Ni-C B O N D FORMATION

131

15. W. Grabarse, S. Shima, F. Mahlert, Ε. C. Duin, R. K. Thauer and U. Ermler, in Handbook of Metalloproteins, Ed. A. Messerschmidt, R. Huber, T. Poulos and K. Wieghardt, John Wiley & Sons, Chichester, 2001, pp. 897-914. 16. F. Mahlert, W. Grabarse, J. Kahnt, R. K. Thauer and E. C. Duin, J. Biol. Inorg. Chem., 2002, 7, 101-112. 17. M. Goenrich, E. C. Duin, F. Mahlert and R. K. Thauer, J. Biol. Inorg. Chem., 2005, 10, 333-342. 18. F. Mahlert, C. Bauer, B. Jaun, R. K. Thauer and E. C. Duin, J. Biol. Inorg. Chem., 2002, 7, 500-513. 19. M. Goenrich, F. Mahlert, Ε. C. Duin, C. Bauer, B. Jaun and R. K. Thauer, J. Biol. Inorg. Chem., 2004, 9, 691-705. 20. R. C. Kunz, Μ. Dey and S. W. Ragsdale, Biochemistry, 2008, 47, 2661-2667. 21. N. Yang, M. Reiher, Μ. Wang, J. Harmer and Ε. C. Duin, J. Am. Chem. Soc., 2007, 129, 11028-11029. 22. M. Dey, J. Telser, R. C. Kunz, Ν. S. Lees, S. W. Ragsdale and Β. M. Hoffman, J. Am. Chem. Soc., 2007, 129, 11030-11032. 23. D. I. Kern, M. Goenrich, B. Jaun, R. K. Thauer, J. Harmer and D. Hinderberger, J. Biol. Inorg. Chem., 2007, 12, 1097-1105. 24. J. Harmer, C. Finazzo, R. Piskorski, S. Ebner, Ε. C. Duin, M. Goenrich, R. Thauer, M. Reiher, Α. Schweiger, D. Hinderberger and B. Jaun, J. Am. Chem. Soc., 2008, 130, 10907-10920. 25. J. Harmer, C. Finazzo, R. Piskorski, C. Bauer, B. Jaun, E. C. Duin, M. Goenrich, R. K. Thauer, S. Van Doorslaer and A. Schweiger, J. Am. Chem. Soc., 2005, 127, 17744-17755. 26. D. Hinderberger, R. P. Piskorski, M. Goenrich, R. K. Thauer, A. Schweiger, J. Harmer and B. Jaun, Angew. Chem. Int. Ed. Engl., 2006, 45, 3602-3607. 27. E. C. Duin, N. J. Cosper, F. Mahlert, R. K. Thauer and R. A. Scott, J. Biol. Inorg. Chem., 2003, 8, 141-148. 28. B. Jaun and A. Pfaltz, J. Chem. Soc., Chem. Commun., 1988, 293-294. 29. S.-K. Lin and B. Jaun, Helv. Chim. Acta, 1992, 75, 1478-1490. 30. S.-K. Lin and B. Jaun, Helv. Chim. Acta, 1991, 74, 1725-1738. 31. Μ. H. Schofield and J. Halpern, Inorg. Chim. Acta, 2003, 345, 353-358. 32. T. Wondimagegn and A. Gosh, J. Am. Chem. Soc., 2000, 122, 6375-6381. 33. U. E. Krone, F. Laufer, R. K. Thauer and H. P. C. Hogenkamp, Biochemistry, 1989, 28, 10061-10065. 34. G. K. Lahiri and A. M. Stolzenberg, Inorg. Chem., 1993, 32, 4409^413. 35. R. C. Kunz, Y. C. Horng and S. W. Ragsdale, J. Biol. Chem., 2006, 281, 34663-34676. 36. M. Dey, R. C. Kunz, D. M. Lyons and S. W. Ragsdale, Biochemistry, 2007, 46, 11969-11978. 37. S. Licht, G. J. Gerfen and J. A. Stubbe, Science, 1996, 271, 477-481. 38. J. A. Stubbe and W. A. van der Donk, Chem. Rev., 1998, 98, 705-762. 39. S. P. Mezyk and D. A. Armstrong, J. Chem. Soc. Perkin Trans. 2, 1999, 1411-1419. 40. Y. Ahn, J. A. Krzycki and H. G. Floss, J. Am. Chem. Soc., 1991,113, 4700-4701. 41. V. Pelmenschikov and P. E. Siegbahn, J. Biol. Inorg. Chem., 2003, 8, 653-662.

Met. Ions Life Sei. 2009, 6, 115-132

132

JAUN and T H A U E R

42. Y. Pelmenschikov, M. R. Blomberg, P. E. Siegbahn and R. H. Crabtree, J. Am. Chem. Soc., 2002, 124, 4039^049. 43. C. Finazzo, J. Harmer, B. Jaun, E. C. Duin, F. Mahlert, R. K. Thauer, S. Van Doorslaer and A. Schweiger, J. Biol. Inorg. Chem., 2003, 8, 586-593. 44. C. Finazzo, J. Harmer, C. Bauer, B. Jaun, E. C. Duin, F. Mahlert, Μ. Goenrich, R. K. Thauer, S. Van Doorslaer and A. Schweiger, J. Am. Chem. Soc., 2003,125, 4988-4989. 45. Y. C. Horng, D. F. Becker and S. W. Ragsdale, Biochemistry, 2001, 40, 12875-12885. 46. A. E. Shilov and G. B. Shul'pin, Chem. Rev., 1997, 97, 2879-2932. 47. E. C. Duin and M. L. McKee, J. Phys. Chem. B, 2008, 112, 2466-2482. 48. E. C. Duin, L. Signor, R. Piskorski, F. Mahlert, Μ. D. Clay, M. Goenrich, R. K. Thauer, B. Jaun and Μ. K. Johnson, J. Biol. Inorg. Chem., 2004, 9, 563-574. 49. J. L. Craft, Y. C. Horng, S. W. Ragsdale and T. C. Brunold, J. Am. Chem. Soc., 2004, 126, 4068-4069.

Met. Ions Life Sei. 2009, 6, 115-132

Met. Ions Life Sei. 2009, 6, 133-150

4 Nickel-Carbon Bonds in Acetyl-Coenzyme A Synthases/Carbon Monoxide Dehydrogenases Paul A. Lindahl Departments of Chemistry and of Biochemistry and Biophysics, Texas A&M University, College Station, TX 77843, USA < lindahl @mail. chem.tamu.edu >

ABSTRACT 1. INTRODUCTION 2. REDOX AND CATALYTIC PROPERTIES OF THE A- AND C-CLUSTERS 3. EVIDENCE FOR A Ni-CO BOND IN THE A R E D -CO STATE OF THE A-CLUSTER 4. EVIDENCE FOR A Ni-CH 3 BOND IN THE METHYLATED INTERMEDIATE OF THE A-CLUSTER 5. EVIDENCE FOR A Ni-C(0)CH 3 BOND IN THE ACETYL INTERMEDIATE OF THE A-CLUSTER 6. EVIDENCE FOR A Ni-CO BOND IN THE C-CLUSTER 7. EVIDENCE FOR A Ni-C(0)0-Fe BOND IN THE C-CLUSTER 8. CONCLUSIONS AND FUTURE STUDIES ACKNOWLEDGMENT ABBREVIATIONS AND DEFINITIONS REFERENCES

133 134 137 139 139 140 141 143 144 147 147 147

ABSTRACT: Acetyl-coenzyme A synthases/carbon monoxide dehydrogenases are bifunctional enzymes that catalyze the synthesis of acetyl-CoA and the reversible reduction of C 0 2 to CO. The active site for the first reaction, called the Α-cluster, consists of a [Fe 4 S 4 ] cubane bridged to a dinuclear nickel subcomponent. The active site for the second reaction, the C-cluster, consists of a [Fe 3 S 4 ] subsite linked to a Ni-Fe dinuclear Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00133

134

LINDAHL

site. There is evidence for the formation of five Ni-C bonds, involving methyl, acetyl, carbonyl, and carboxylate groups. In this review, the current evidence for each of these bonds is described. The mechanism of catalysis is discussed, highlighting the role of these species. The unique coordination environments of nickel that may facilitate the formation of organometallic species is discussed. Current puzzles in the field and future research directions that might resolve them are outlined. KEYWORDS: A-cluster · C-cluster · Fe/S clusters · organometallic · EPR spectroscopy· Mössbauer spectroscopy · infrared spectroscopy

1.

INTRODUCTION

Acetyl-coenzyme A synthases/carbon monoxide dehydrogenases (ACS/ CODH) are bifunctional Ni-containing enzymes found in certain bacteria and archaea. The organisms that contain such enzymes are anaerobic, mostly thermophilic, and able to grow chemiautotrophically on simple inorganic compounds (e.g., C 0 2 as a source of carbon). The metabolic and phylogenetic characteristics of the organisms containing this family of enzymes have been reviewed recently [1]. Other recent reviews have focused on the structure and function of the metal centers in these enzymes and on synthetic model complexes [2-4]. The major objective of this review is to describe the evidence for Ni-C bonds contained within these enzymes (when prepared in particular states). There are five possible catalytic intermediates which have been proposed to contain such bonds, involving methyl, acetyl, carbonyl, and carboxylate groups. Evidence for these bonds ranges from unambiguous and definite, to indirect and somewhat speculative. Before describing each candidate organometallic species, and the evidence for it, the structure and physiochemical properties of the active-site metal clusters will be outlined, as will the most popular mechanisms of catalysis. Readers should consult other reviews for a more complete description of the structure of the enzymes and associated metal centers, and for a more detailed comparison of competing proposed catalytic mechanisms. The best-studied ACS/CODH, from the mesophile Moorella thermoacetica, is an 310,000 Da α 2 β 2 tetramer. The two β subunits, which form the central core (Figure la), catalyze the reversible oxidation of CO to C 0 2 (CODH reaction 1). The active site for this activity, called the C-cluster, is CO + H 2 0 ^ C 0 2 + 2e" + 2H+

(1)

a NiFe 4 S 4 cluster (Figure lb) (for complete structural details, see [5-10]). The C-cluster consists of an [Fe3S4] subsite linked to a [NiFeJ subsite. Two of the three μ 2 -bridging sulfides of the [Fe3S4] subsite additionally coordinate to the Ni ion while the third μ 2 -bridging sulfide additionally coordinates Met. Ions Life Sei. 2009, 6, 133-150

N I C K E L - C A R B O N B O N D S IN A C S / C O D H

135

a

CH2x s

\ Fe— S

/

\t~rf

/CHa

Fe

CH2-S

C H 2

N

HC CH \ / ΗΝ—C^ "CH'2

c o^yS Fe i\

& |\

S i f - fe-L

CHo j

/ CHx ο i^ii L-Mj Ο

c

\

/ '2

Nip

Nid

Ο

s, >cvs

ι

CH2—CH

Figure 1. Structures of A C S / C O D H and associated metal centers, (a) symbolic protein structure of A C S / C O D H . Each α subunit contains three domains which are in slightly different conformations. Each α subunit contains a single metal center called the Α-cluster (green). The two β subunits form the central core of the enzyme. These subunits contain three types of metal centers, including the active-site Cclusters (green) and electron-transfer clusters Β and D (brown); (b) structure of the Ccluster; (c) structure of the A-cluster.

t o a u n i q u e F e called F e a . T h e N i a n d F e a c a n be b r i d g e d b y a h y d r o x o g r o u p ( a l m o s t c e r t a i n l y d e r i v e d f r o m t h e s u b s t r a t e H 2 0 ) [10], o r b y s u l f i d o [5,9] a n d p e r h a p s o t h e r a n i o n s s u c h as C N " [11], S C N " , O C N " a n d ΝίΓ g r o u p s [12,13]. T h e N i F e a b r i d g i n g site c a n also be u n o c c u p i e d [7,8]. T h e C cluster N i is also c o o r d i n a t e d t o a cysteine t h i o l a t e , r e s u l t i n g in a p l a n a r t h r e e - c o o r d i n a t e T - s h a p e d g e o m e t r y w h e n t h e b r i d g i n g p o s i t i o n is e m p t y Met. Ions Life Sei. 2009, 6, 133-150

LINDAHL

136

and a square-planar geometry when it is occupied. Completing the coordination sphere of Fe a is a cysteine thiolate and a histidine imidazole. Three cysteine thiolates affix the [Fe3S4] subsite to the protein. The C-cluster is buried ca. 18 A below the protein surface [7]. Two [Fe4S4] clusters in the β subunits form a "wire" that electronically connects the C-cluster to redox agents in the solvent. The B-cluster is proximal to the C-cluster and located I I A away from the C-cluster (Figure la). The distal D-cluster, about 10 A away from the B-cluster, bridges the two β subunits and is essentially at the protein surface. The Α-cluster active site within the α subunits (Figure la) catalyzes the acetyl-CoA synthase (ACS) activity (reaction 2). CH 3 -Co 3 + FeSP + CO + CoASH ^ CH 3 -C(0)-CoA + Co 1 + FeSP + H+

(2)

CoFeSP, the corrinoid-iron-sulfur protein, is the specific methyl group donor to ACS/CODH. In a separate reaction, CoFeSP accepts a methyl group from methyl-tetrahydrofolate (H 4 F), in accordance with reaction (3), which is catalyzed by a methyltransferase. CH3-H4F + Co 1 + FeSP + H+ ^ CH 3 -Co 3 + FeSP + H-H 4 F

(3)

The Α-cluster consists of a [Fe4S4] cubane bridged through a cysteine residue to a Ni ion (called proximal Ni p ) which is additionally bridged, through 2 cysteine residues, to a second Ni ion called distal Ni d (Figure lc) (for complete structural details, see [7-8,14]). Ni p is square-planar, including the three μ 2 -bridging cysteine thiolates and an endogenous unidentified ligand which is presumably replaced by substrates during catalysis. The N 2 S 2 square-planar geometry of Ni d is completed by the coordination to two amide nitrogens originating from the protein backbone. The acetyl-CoA decarbonylase/synthases (ACDS) represent another class of enzymes within the Ni-containing CODH family [15]. These trifunctional αβγδε enzymes have ACS, CODH and methyltransferase activities, and they function in acetoclastic methanogenesis. The isolated α subunit of ACS/ CODH and the homologous subunit from other enzymes can be prepared by recombinant genetic methods, and are active catalysts after incubation in NiCl 2 [14-18]. The monomeric ACS from Carboxydothermus hydrogenoformans has been structurally characterized [14]. There are also monofunctional Ni-dependent enzymes that catalyze only the CODH reaction, with the ß 2 dimeric enzyme from Rhodospirillum rubrum (CODH R r ) being the best-studied [19], Met. Ions Life Sei. 2009, 6, 133-150

NICKEL-CARBON BONDS IN ACS/CODH

2.

137

REDOX AND CATALYTIC PROPERTIES OF THE A- AND C-CLUSTERS

The C-cluster can be stabilized in 4 redox states, including C o x (S = 0), C r e d i ( S = 1/2, with EPR principal g-values of 2.01, 1.81, 1.65 for ACS/CODH), Qnt (S = 0 or integer), and C r e d 2 ( S = l / 2 , with EPR principal g-values of 1.97, 1.87, 1.75, again for A C S / C O D H ) [11,20,21], The C o x state is most oxidized and the others are progressively more reduced in one-electron increments. Reasonable electronic configurations for the metal ions in the C o x and C r e d l states include {[Fe3 + , F e 2 + , F e 2 + ] [Ni 2 + Fe a 3 + ]} and {[Fe 3 + , F e 2 + , F e 2 + ] [Ni 2 + Fe a 2 + ]}, respectively [11,22], while those for C i n t and C r e d 2 are less certain. The best-supported or "standard" C O D H catalytic mechanism is shown in Figure 2a. The C-cluster becomes activated for catalysis when it is reduced from the C o x to the C r e d i state [23]. A hydroxide ion bridges the Ni and Fe a in the C r e d i state [13,24]; this ion is derived from the substrate H 2 0 and it serves as nucleophile to attack a Ni-bound carbonyl carbon (see below) [25]. Deprotonation of the resulting carboxylic acid leads to a [Ni-C02-Fe a ] intermediate [10]. Subsequent decarboxylation leads to the C r e d 2 state. Two one-electron oxidations, the binding of water, and the deprotonation o f t h a t water return the C-cluster to the C r e d i state, completing the catalytic cycle. The Α-cluster can be stabilized in three well-established states (A o x , CH 3 A o x , and A r e d -CO) and perhaps in two less-established states (A red _ act and CH 3 C(0)-A o x ). The S = 0 AOX state is obtained when A C S / C O D H (or the isolated Ni-activated α subunit) is prepared in an inert-atmosphere glove box without added reductants or oxidants. The electronic configuration of A o x appears to be {[Fe 4 S 4 ] 2+ Ni^ + Ni^ + } [26,27], The S = 1/2 A r e d -CO state is obtained by treating A C S / C O D H with CO or by treating the α subunit with CO and a low-potential reductant like dithionite or Ti(III) citrate. The Ni becomes reduced and bound with CO, affording the electronic configuration {[Fe 4 S 4 ] 2+ Nip + -CO Ni^+}. The A r e d -CO state exhibits the well-studied NiFeC EPR signal [28]. In the so-called diamagnetic mechanism (Figure 2b), A r e d -CO represents an inhibited form of catalysis. However, in the paramagnetic mechanism, A r e d -CO is considered to be a catalytic intermediate [2,3,29,30], According to the diamagnetic mechanism, the Α-cluster in the A o x state is reduced by two low-potential electrons to the reductively-activated A red _ act state before it is active for catalysis [27], The electronic configuration for this state is controversial, but it is probably either {[Fe 4 S 4 ] 2+ Ni® Ni d + } or {[Fe 4 S 4 ] 1+ Nip+ Ni d +} [31-34], The diamagnetic CH 3 -A o x state is the methylated intermediate (described below) with electronic configuration {[Fe 4 S 4 ] 2+ Ni^+-CH 3 Ni^ + } [27], This state is obtained by reducing ACS/ C O D H (or the isolated α subunit) with a low-potential reductant, followed Met. Ions Life Sei. 2009, 6, 133-150

138

LINDAHL

H

x

[Ni2+

Fe a 2+ ]

CO

y

[Ni 2+

0 1{[Fe]2Nil+} or {[Fep- Ni2+}

! Mn+

M"+ Credl

Mn+ I n Ni2+ j \ M+ / \ I XS2c" -O

Cred2

01.

C02-Cred1

Figure 3. Comparison of the Ni p and C-cluster Ni coordination environment and reactivity. See text for details. Met. Ions Life Sei. 2009, 6, 133-150

146

LINDAHL

Where should we go from here? Clearly, obtaining high-quality X-ray diffraction structures of the remaining four Ni-C intermediates will be critical in providing conclusive evidence for their existence. With this structural information in hand, additional computational studies will be critical in defining the ACS and CODH catalytic mechanisms at the atomic level of detail. With regard to the ACS mechanism, the most critical issue (in my mind) is whether the A red _ act state represents a {[Fe 4 S 4 ] 2+ Ni®} or a {[Fe 4 S 4 ] 1+ Nip + } configuration. The order of substrate addition must also be settled; recent studies suggest a random mechanism of substrate addition, with either CO or the methyl group adding first and second, and CoA adding last [29]. The enzyme contains an exotic tunnel that connects the Cand A-clusters [7,8]. This tunnel serves not simply to channel CO from the site where it is made (the C-cluster) to the site where it is consumed (the Acluster), but probably to control delivery of CO to the Α-cluster such that it arrives at the appropriate step of catalysis [68,69]. How this view of the tunnel's function can be reconciled with the evidence for the random addition of CO and methyl groups will require further mechanistic studies. With regard to the CODH mechanism, perhaps the most pressing problem is the electronic configuration of Cred2 and a better understanding of the effects of C 0 / C 0 2 on the EPR spectra of the C-cluster. Mössbauer spectroscopy could be used to evaluate iron oxidation states and spin-coupling arrangements, but the problem is that the C-cluster represents at best ~ 3 0 % of the iron in ACS/CODH. Given the heterogeneity of the metal centers in the enzyme, it would appear that only ~ 8 % of the Fe in a sample of ACS/ CODH actually contributes to the C r e d2 state. Moreover, the B-cluster is reduced and magnetic whenever the C-cluster is in the Cred2 state, and the magnetic spectral features of the two centers overlap. Mutant strains should be developed in which the Α-, B-, and D-clusters are either abolished or locked into diamagnetic states, as this would make the study of the electronic properties of the C re d2 state feasible. Another interesting research path would be to probe the C0 2 -bound state of the C-cluster using 13 C and magnetic resonance techniques. If the C-cluster when bound with C 0 2 ( C 0 ) is in the C re di(C re d2) state, perhaps even weak hyperfine interactions could be observed using ESEEM spectroscopy. Ultimately, the study of these enzymes might facilitate the more practical goal of designing new metalloenzymes that catalyze other desirable organometallic reactions. Given the plethora of Ni-C bonds in ACS/CODH, it seems that this enzyme would be an ideal platform upon which such studies could ensue. We have recently attempted (perhaps successfully) to incorporate Pd and Pt into the proximal site of the A-cluster [70]. Future combinations of transition metal coordination chemistry with recombinant genetic engineering might prove successful in developing new bio-organometallic catalysts. Met. Ions Life Sei. 2009, 6, 133-150

NICKEL-CARBON BONDS IN ACS/CODH

147

ACKNOWLEDGMENT I would like to thank the National Institutes of Health (GM46441) graciously funding my laboratory to work on this enzyme for the past 15 years.

ABBREVIATIONS AND DEFINITIONS ACDS ACS ACS/CODH Aox -^red-act

A r e d -CO CH 3 -A OX CH 3 C(0)-A, CoA CODH CODH R R CoFeSP C CREDL? ^ I II I c

red2

ENDOR EPR ESEEM H4F IR Ni d Ni p phen UV

acetyl-CoA decarbonylase/synthase acetyl-coenzyme A synthase the bifunctional enzyme from Moorella thermoacetica fully oxidized state of the A-cluster the reductively activated state of the A-cluster form of the A-cluster one electron more reduced than A o x and bound with CO the methylated state of the A-cluster the acetylated state of the A-cluster coenzyme A carbon monoxide dehydrogenase C O D H from Rhodospirillum rubrum corrinoid-iron-sulfur protein fully oxidized state of the C-cluster redox states of the C-cluster that are 1, 2, and 3 electrons more reduced, respectively, relative to C o x electron nuclear double resonance electron paramagnetic resonance electron spin echo envelop modulation tetrahydrofolate infrared the distal Ni of the A-cluster the proximal Ni of the A-cluster 1,10-phenanthroline ultraviolet

REFERENCES 1. D. E. Graham, P. A. Lindahl, in Metal Ions in Life Sciences, A. Sigel, H. Sigel, R. K. O. Sigel, (Ed.), John Wiley & Sons, Ltd.Chichester, UK2007, Vol. 2, pp. 357^416. 2. S. W. Ragsdale, J. Inorg. Biochem., 2007, 101, 1657-1666. 3. S. W. Ragsdale, Chem. Rev., 2006, 106, 3317-3337. Met. Ions Life Sei. 2009, 6, 133-150

148

LINDAHL

4. Τ. C. H a r r o p and P. K . M a s c h a r a k , Coord. Chem. Rev., 2005, 249, 3007-3024. 5. Η. D o b b e k , V. Svetlitchnyi, L. Gremer, R. H u b e r and O. Meyer, Science, 2001, 293, 1281-1285. 6. C. L. D r e n n a n , J. Heo, M . D . Sintchak, E. Schreiter and P. W. Ludden, Proc. Natl. Acad. Sei. USA, 2001, 98, 11973-11978. 7. Τ. I. D o u k o v , Τ. M . Iverson, J. Seravalli, S. W. Ragsdale and C. L. D r e n n a n , Science, 2002, 298, 567-572. 8. C. Darnault, A. Yolbeda, E. J. Kim, P. Legrand, X. Vernede, P. A. Lindahl and J. C. Fontecilla-Camps, Nat. Struct. Biol., 2003, 10, 271-279. 9. Η. D o b b e k , Y. Svetlitchnyi, J. Liss and Ο. Meyer, J. Am. Chem. Soc., 2004, 126, 5382-5387. 10. J. H . Jeoung and H. D o b b e k , Science, 2008, 318, 1461-1464. 1 1 . Z . H u , N . J. Spangler, Μ . Ε. Anderson, J. Xia, P. W. Ludden, P. A. Lindahl and E. M ü n c k , J. Am. Chem. Soc., 1996, 118, 830-845. 12. J. Seravalli, M . K u m a r , W. P. Lu and S. W. Ragsdale, Biochemistry, 1995, 34, 7879-7888. 13. J. Feng and P. A. Lindahl, J. Am. Chem. Soc., 2004, 126, 9094-9100. 14. V. Svetlitchnyi, H. D o b b e k , W. Meyer-Klaucke, T. Meins, B. Thiele, P. Römer, R. H u b e r and O. Meyer, Proc. Natl. Acad. Sei. USA, 2004, 101, 446-451. 15. D . A. G r a h a m e , Trends Biochem. Sei., 2003, 28, 221-224. 16. Η . Κ. Loke, X. T a n and P. A. Lindahl, J. Am. Chem. Soc., 2002, 124, 8667-8672. 17. T. F u n k , W. W. Gu, S. Friedrich, Η . X. Wang, S. Gencic, D . A. G r a h a m e and S. P. Cramer, J. Am. Chem. Soc., 2004, 126, 88-95. 18. S. J. George, J. Seravalli and S. W. Ragsdale, J. Am. Chem. Soc., 2005, 127, 13500-13501. 19. W. B. Jeon, S. W. Singer, P. W. Ludden and L. M . Rubio, J. Biol. Inorg. Chem., 2005, 10, 903-912. 20. P. A. Lindahl, E. M ü n c k and S. W. Ragsdale, J. Biol. Chem., 1990, 265, 3873-3879. 21. D . M . Fräser and P. A. Lindahl, Biochemistry, 1999, 38, 15706-15711. 22. Ρ. Α. Lindahl, Biochemistry, 2002, 41, 2097-2105. 23. J. Feng and P. A. Lindahl, Biochemistry, 2004, 43, 1552-1559. 24. Y. J. DeRose, J. Telser, Μ . E. Anderson, P. A. Lindahl and Β. M . H o f f m a n , J. Am. Chem. Soc., 1998, 120, 8767-8776. 25. J. Seravalli, M . K u m a r , W. P. Lu and S. W. Ragsdale, Biochemistry, 1997, 36, 11241-11251. 26. J. Xia, Z. H u , C. V. Popescu, P. A. Lindahl and E. M ü n c k , J. Am. Chem. Soc., 1997, 119, 8301-8312. 27. M . R. Bramlett, A. Stubna, X. Tan, I. Surovtsev, E. M ü n c k and P. A. Lindahl, Biochemistry, 2006, 45, 8674-8685. 28. S. W. Ragsdale, H . G. W o o d and W. E. Antholine, Proc. Natl. Acad. Sei. USA, 1985, 82, 6811-6814. 29. J. Seravalli and S. W. Ragsdale, J. Biol. Chem., 2008, 283, 8384-8394. 30. J. Seravalli, M . K u m a r and S. W. Ragsdale, Biochemistry, 2002, 41, 1807-1819. 31. P. A. Lindahl, J. Biol. Inorg. Chem., 2004, 9, 516-524.

Met. Ions Life Sei. 2009, 6, 133-150

NICKEL-CARBON BONDS IN ACS/CODH

149

32. X. Tan, M. Martinho, A. Stubna, P. A. Lindahl and E. Münck, J. Am. Chem. Soc., 2008, 130, 6712-6713. 33. R. P. Schenker and T. C. Brunold, J. Am. Chem. Soc., 2003, 125, 13962-13963. 34. P. Amara, A. Yolbeda, J. C. Fontecilla-Camps and M. J. Field, J. Am. Chem. Soc., 2005, 127, 2776-2784. 35. E. Pezacka and H. G. Wood, J. Biol. Chem., 1988, 263, 16000-16006. 36. W. P. Lu, S. R. Harder and S. W. Ragsdale, J. Biol. Chem., 1990, 265, 3124-3133. 37. M. Kumar, D. Qiu, T. G. Spiro and S. W. Ragsdale, Science, 1994, 270, 628-630. 38. D. P. Barondeau and P. A. Lindahl, J. Am. Chem. Soc., 1997, 119, 3959-3970. 39. X. S. Tan, C. Sewell and P. A. Lindahl, J. Am. Chem. Soc., 2002,124, 6277-7284. 40. X. Tan, C. Sewell, Q. Yang and P. A. Lindahl, J. Am. Chem. Soc., 2003, 125, 318-319. 41. X. Tan, Ι. V. Surovtsev and P. A. Lindahl, J. Am. Chem. Soc., 2006, 128, 12331-12338. 42. J. Y. Chen, S. Huang, J. Seravalli, H. Gutzman, D. J. Swartz, S. W. Ragsdale and K. A. Bagley, Biochemistry, 2003, 42, 14822-14830. 43. C. L. Fan, C. M. Gorst, S. W. Ragsdale and Β. M. Hoffman, Biochemistry, 1991, 30, 431-435. 44. P. Stavropoulos, M. C. Muetterties, M. Carrie and R. H. Holm, J. Am. Chem. Soc., 1991, 113, 8485-8492. 45. T. C. Harrop, Μ. M. Olmstead and P. K. Mascharak, Inorg. Chem., 2006, 45, 3424-3436. 46. W. Shin and P. A. Lindahl, J. Am. Chem. Soc., 1992, 114, 9718-9719. 47. T. C. Harrop, Μ. M. Olmstead and P. K. Mascharak, Chem. Comm., 2004, 15, 1744-1745. 48. S. W. Ragsdale and M. Kumar, Chem. Rev., 1996, 96, 2515-2539. 49. W. Shin, Μ. E. Anderson and P. A. Lindahl, J. Am. Chem. Soc., 1993, 115, 5522-5526. 50. E. Pezacka and H. G. Wood, J. Biol. Chem., 1986, 266, 3554-3564. 51. S. E. Ramer, S. A. Raybuck, W. H. Orme-Johnson and C. T. Walsh, Biochemistry, 1989, 28, 4675-4680. 52. W. P. Lu and S. W. Ragsdale, J. Biol. Chem., 1991, 266, 3554-3564. 53. B. Bhaskar, E. DeMoll and D. A. Grahame, Biochemistry, 1998, 37, 14491-14499. 54. D. M. Fraser and P. A. Lindahl, Biochemistry, 1999, 38, 15697-15705. 55. V. J. DeRose, J. Telser, Μ. E. Anderson, P. A. Lindahl and Β. M. Hoffman, J. Am. Chem. Soc., 1998, 120, 8767-8776. 56. J. Heo, C. R. Staples, J. Telser and P. W. Ludden, J. Am. Chem. Soc., 1999, 121, 11045-11057. 57. S. A. Macgregor, Z. Lu, O. Eisenstein and R. H. Crabtree, Inorg. Chem., 1994, 33, 3616-3618. 58. C. Saint-Joly, A. Mari, A. Gleizes, M. Dartiguenave, Y. Dartiguenave and J. Galy, Inorg. Chem., 1980, 19, 2403-2410. 59. D. H. Nguyen, H. -F. Hsu, M. Millar and S. A. Koch, J. Am. Chem. Soc., 1996, 118, 8963-8964. 60. Μ. E. Anderson and P. A. Lindahl, Biochemistry, 1994, 33, 8702-8711.

Met. Ions Life Sei. 2009, 6, 133-150

150 61. 62. 63. 64. 65. 66. 67. 68. 69. 70.

LINDAHL Μ. Ε. Anderson and P. A. Lindahl, Biochemistry, 1996, 35, 8371-8380. W. K. Russell and P. A. Lindahl, Biochemistry, 1998, 37, 10016-10026. S. A. Ensign, Biochemistry, 1995, 34, 5372-5381. Z. Lu and R. H. Crabtree, J. Am. Chem. Soc., 1995, 117, 3994-3998. S. Sakaki, J. Am. Chem. Soc., 1992, 114, 2055-2062. M. Isaacs, J. C. Canales, M. J. Aguirre, G. Estiu, F. Caruso, G. Ferraudi and J. Costamagna, Inorg. Chim. Acta, 2002, 339, 224-232. C. A. Grapperhaus and Μ. Y. Darensbourg, Acc. Chem. Res., 1998, 31, 4 5 1 ^ 5 9 . X. Tan, H. -K. Loke, S. Fitch and P. A. Lindahl, J. Am. Chem. Soc., 2005, 127, 5833-5839. X. Tan, A. Yolbeda, J. C. Fontecilla-Camps and P. A. Lindahl, J. Biol. Inorg. Chem., 2006, 11, 371-378. X. Tan, I. Kagiampakis, I. V. Surovtsev, B. Demeler and P. A. Lindahl, Biochemistry, 2007, 46, 11606-11613.

Met. Ions Life Sei. 2009, 6, 133-150

Met. Ions Life Sei. 2009, 6, 151-178

5 Structure and Function of [NiFe]-Hydrogenases Juan C.

Fontecilla-Camps

Laboratoire de Cristallographie et de Cristallogenese des Proteines, Institut de Biologie S t r u c t u r a l J. P. Ebel (CEA-CNRS-UJF), 41 rue Jules Horowitz, F-38027 Grenoble Cedex 1, France

ABSTRACT 1. INTRODUCTION 2. HYDROGENASE STRUCTURE 2.1. The Three-Dimensional Folding 2.2. The Active Site at Medium Resolution 2.3. The Active Site at Higher Resolution 3. HYDROGENASE MATURATION AND ACTIVE SITE ASSEMBLY 3.1. Maturation of the Large Subunit 3.2. Cyanide, Carbon Monoxide, and Iron Insertion 3.3. Nickel Insertion 3.4. Proteolytic Cleavage of the Large Subunit C-Terminal Extension 4. ELECTRON TRANSFER 5. PROTON TRANSFER 6. OXIDIZED INACTIVE STATES OF THE [NiFe]HYDROGENASE ACTIVE SITE 7. SUBSTRATE BINDING AND CATALYSIS 8. CONCLUDING REMARKS ACKNOWLEDGMENTS ABBREVIATIONS REFERENCES Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00151

152 152 153 153 156 158 160 161 161 164 164 165 166 168 172 173 173 173 174

152

FONTECILLA-CAMPS

ABSTRACT: [NiFe(Se)]-hydrogenases are hetero-dimeric enzymes present in many microorganisms where they catalyze the oxidation of molecular hydrogen or the reduction of protons. Like the other two types of hydrogen-metabolizing enzymes, the [FeFe]- and [Fe]-hydrogenases, [NiFe]-hydrogenases have a Fe(CO) x unit in their active sites that is most likely involved in hydride binding. Because of their complexity, hydrogenases require a maturation machinery that involves several gene products. They include nickel and iron transport, synthesis of CN~ (and maybe CO), formation and insertion of a FeCO(QSr) 2 unit in the apo form, insertion of nickel and proteolytic cleavage of a C-terminal stretch, a step that ends the maturation process. Because the active site is buried in the structure, electron and proton transfer are required between this site and the molecular surface. The former is mediated by either three or one Fe/S cluster(s) depending on the enzyme. When exposed to oxidizing conditions, such as the presence of 0 2 , [NiFe]-hydrogenases are inactivated. Depending on the redox state of the enzyme, exposure to oxygen results in either a partially reduced oxo species probably a (hydro)peroxo ligand between nickel and iron or a more reduced OFT ligand instead. Under some conditions the thiolates that coordinate the NiFe center can be modified to sulfenates. Understanding this process is of biotechnological interest for H 2 production by photosynthetic organisms. KEYWORDS: active site assembly · electron transfer · enzyme maturation · hydrogen oxidation · iron with CO and CN~ coordination · [NiFe]-hydrogenases · proton transfer

1.

INTRODUCTION

H y d r o g e n is utilized by microorganisms as a source of reducing power or produced when p r o t o n s are used as final electron acceptors. The reaction is mediated by metallo-enzymes called hydrogenases. The need of catalytic transition metal centers is explained by the significant increase in the acidity of molecular hydrogen when b o u n d to metal [1]. Three phylogenetically unrelated classes of these enzymes exist: [NiFe]and [FeFe]-hydrogenases with hydrogen being the only substrate or p r o d u c t , according to the following reaction: H

2

^2H++2e-

(1)

In the third class, hydrogen u p t a k e is coupled to methenyltetrahyd r o m e t h a n o p t e r i n reduction [2]. Recently, the latter has been called [Fe]hydrogenase because, as opposed to the other two enzymes, its active site contains a single metal ion. Previously, the structures of the apo-enzyme and a photolyzed derivative of its Fe-binding cofactor were reported [2]. There is n o w a crystal structure available for the holoenzyme (S. Shima, personal c o m m . [97]). There are five structures of [NiFe]-hydrogenases available, all f r o m sulfate-reducing bacteria [3-7]. F o r [FeFe]-hydrogenase, there are structures f r o m the sulfate-reducing bacterium Desulfovibrio desulfuricans A T C C 7757 [8] and f r o m Clostridium pasteurianum [9]. The Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

153

former is a hydrogen-oxidizing periplasmic enzyme whereas the latter is cytoplasmic and produces hydrogen. A low-spin Fe(CO) x motif is common to the three enzyme classes indicating that it represents Nature's solution to hydrogen catalysis. The active sites of [NiFe]- and [FeFe]-hydrogenases are buried within the structure, requiring electron and proton transport between the catalytic center and the molecular surface. Also, hydrogen has to access the active site or diffuse from it depending on the direction of reaction (1). Because of the unusual structure of the active sites of these enzymes, a considerable effort has been invested in understanding their assembly and significant progress has resulted from this. Last but not least, a recent and quite popular subject of study concerning these enzymes is their degree of sensibility to molecular oxygen. Besides being of basic scientific interest, this study is important when hydrogenases are considered as catalysts in bio-fuel cells or in the industrial production of bio-hydrogen. Prior to the structural analyses of hydrogenases, a series of electron paramagnetic resonance (EPR) spectroscopic studies had shown that the oxidized enzyme is characterized by two paramagnetic states, called Ni-A and Ni-B that disappeared upon activation with hydrogen or other reducing agents. Because enzyme in the Ni-A state was difficult to activate, it was also called the unready form. On the other hand, the Ni-B species was rapidly activated and was consequently considered to be in a ready state. The remaining EPR-active species, obtained under reducing conditions, was called Ni-C and was considered to be part of the catalytic cycle. Two diamagnetic species were observed upon reduction of the Ni-B and Ni-C states and were called Ni-SI and Ni-R, respectively (Figure 1). Theoretical fitting of D. gigas [NiFe]-hydrogenase redox titrations favored the Two Electron Difference (TED) model with the active Ni-B form being two electrons more oxidized than the reduced ready Ni-C state. Thus, the sequence Ni-A/B-»NiSI -*• Ni-C -*• Ni-R was postulated to represent one-electron reducing steps.

2. 2.1.

HYDROGENASE STRUCTURE The Three-Dimensional Folding

Crystals of Desulfovibrio (D.) vulgaris Miyazaki [NiFe]-hydrogenase [4] and the D. gigas enzyme [10] were first reported in 1987. However, the first structure was not published until 1995, from a better quality crystal form of the latter enzyme [3]. The crystals were grown by the vapor diffusion method and diffraction data were collected in house with a rotating anode Cu-Ko, Xray source at room temperature. X-ray data had to be collected immediately Met. Ions Life Sei. 2009, 6, 151-178

154

FONTECILLA-CAMPS

1 2800

1 1 Γ 3000 3200 3400 Magnetic Field (Gauss)

1 3600

Figure 1. Ni EPR spectra of the unready Ni-A, ready Ni-B, and active Ni-C paramagnetic species. The spectra of the diamagnetic Ni-SI and Ni-R are also depicted. The successive Ni-B/SI/C/R are thought to be separated by one-electron reductions [84]. Adapted from [92].

after crystal growth because the crystals became colorless and rapidly lost their diffracting power upon exposure to air. A complete 2.85 A resolution X-ray data set was collected from several native crystals and many others were tested for heavy atom derivatives in soaking experiments. The crystal structure was subsequently solved by a combination of multiple isomorphous replacement (MIR) and non-crystallographic symmetry density averaging over the two enzyme heterodimers in the asymmetric unit. The two subunits establish an extensive contact interface of about 3,500 A 2 (Figure 2). The small subunit has a N-terminal flavodoxin-like Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

155

Figure 2. Stereo pair of the crystal structure of [NiFe]-hydrogenase from Desulfovibrio gigas. The large and small subunits are depicted in red and blue, respectively. The three Fe/S clusters and the active site are represented by spheres. Color codes: yellow for S, red for Fe, and green for Ni. Figures 2, 3, 5, 7 and 8 were prepared with the Y M D program [96].

domain that represents 65% of the whole polypeptide chain. Superposition of this domain to 89 Ca atoms of a clostridian flavodoxin [11] results in a root mean square deviation of about 2.7 A [3]. The domain coordinates a [Fe4S4] cluster, topologically equivalent to the site occupied by the phosphate group of F M N in flavodoxin. This (proximal) cluster is the closest one to the active site and is located near the enzyme center. The mesial [Fe3S4] and distal [Fe4S4] clusters are bound to the remaining domain that represents 35% of the small subunit and has little secondary structure. This region is missing in some [NiFe]-hydrogenases, such as NAD(P) + -reducing and energy converting enzymes. With the exception of one of the Fe ligands of the distal cluster, that is the Νδ atom of a surface exposed histidine side chain, the cluster Fe-atoms are coordinated by cysteine thiolates. The Ni-Fe active site of the D. gigas enzyme is located in the large subunit (Figure 2). Towards the end of the large subunit chain tracing process, we found that there was no electron density corresponding to the last 15 C-terminal residues of the gene sequence [3]. This was surprising because the last visible residue, His536, was deeply buried in the structure. However, shortly thereafter, we noticed a paper by Menon et al. that reported the proteolytic cleavage of a 15-residue C-terminal peptide as part of the maturation process of the large subunit of D. gigas [NiFe]-hydrogenase [12]. In the homologous [NiFeSe]-hydrogenase from Methanococcus voltae, the equivalent of His536 is the C-terminal residue of a 25-residue long third Met. Ions Life Sei. 2009, 6, 151-178

156

FONTECILLA-CAMPS

subunit [13]. Depending on selenium availability, this third subunit can contain either Cys or SeCys as nickel ligand. High resolution structures of the D. vulgaris Miyazaki [4] and D. gigas hydrogenases showed that the C-terminal histidine binds a Mg ion. U p o n the cleavage of the C-terminal peptide, the large subunit undergoes a substantial conformational rearrangement burying two C-terminal α-helices. Additional structures of [NiFe]-hydrogenases from D. fructosovorans [5], D. desulfuricans [7], and the [NiFeSe] enzyme from Desulfomicrobium baculatum [6] indicated that all these enzymes have essentially the same domain structure and polypeptide fold. Only the [NiFeSe] enzyme from Dm. baculatum [6], is somewhat different from the Desulfovibrio [NiFe]-hydrogenases: (a) the mesial [Fe 3 S 4 ] cluster is replaced by a third [Fe 4 S 4 ] cluster (Figure 2); (b) the Mg ion that binds to the C-terminal histidine of the large subunit in the other enzymes is replaced by an iron ion; (c) near the active site there is a tightly bound H 2 S molecule, and (d) the Cys530 Ni-ligand of D. gigas is replaced by a seleno-cysteine (SeCys). Additional differences are found at the enzyme surface.

2.2.

The Active Site at Medium Resolution

In 1993, after having defined the molecular envelope, we calculated a 5 A resolution anomalous scattering difference map of D. gigas hydrogenase. This map showed four peaks more or less linearly arranged and separated by a center-to-center distance of about 12 A [14]. The two [Fe4S4] clusters were assigned to the strongest peaks, whereas the intermediate [Fe 3 S 4 ] cluster and the active site were considered to generate the third and fourth weaker peaks, respectively. The latter was thought to correspond to a single Ni ion. Higuchi et al. found the same distribution of metal sites in a 4 A resolution analysis of D. vulgaris Miyazaki [NiFe]-hydrogenase [15]. However, when the resolution of the X-ray data from D. gigas enzyme crystals was extended to 2.85 A, we realized that the weak anomalous scattering difference peak at the active site corresponded to a different metal ion close to Ni [3]. With the benefit of hindsight we can now conclude that this was to be expected; the anomalous differences we used were obtained using Cu K a X-ray radiation, the wavelength of which corresponds to the low energy side of the Ni absorption edge. [NiFe]-hydrogenases have two CxxC conserved motifs at the N- and C-terminal ends of the large subunit, respectively. Before the crystal structure was determined, site-directed mutagenesis and E X A F S studies had indicated that the four cysteines of the two conserved motifs coordinated the Ni ion [16]; this was later confirmed by the X-ray analysis [3]. Thus, it was easy to assign the Ni site in the electron density because it was known, from Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

157

additional EXAFS analyses of homologous [NiFeSe]-hydrogenases, that it had to be terminally bound to the thiolate of Cys530 [17-19]. The second metal ion was then tentatively assigned to iron because of the reported content of 12 ± 1 iron atoms per hydrogenase heterodimer [3,20]. The active site model at 2.85 A resolution was completed with three ligands, modelled as water molecules, bound to the putative iron ion [3] (X in Figure 3). When in 1996 we started using cryogenic methods and synchrotron radiation with tunable X-ray wavelengths, it became much easier to obtain high resolution, good quality X-ray diffraction data. Crystals could be stored indefmitively in liquid nitrogen and were much more resistant to radiation damage when kept at 100 Κ during X-ray data collection. Also, complete data sets could be measured much faster from single crystals thanks to the intense synchrotron X-ray beam. Thus, crystallographic evidence for the presence of an active site iron ion was obtained using anomalous scattering X-ray data collected at 1.733 A and 1.750 A (at either side of the iron absorption edge) at the European Synchrotron Radiation Facility

Cys530

Figure 3. Model of the active site of [NiFe]-hydrogenase from Desulfovibrio gigas at 2.85 A resolution [3]. Three water molecules (Wat) were modeled as ligands of X that, in turn, was refined as an Fe ion. I represents a bridging ligand, it was modeled as an oxygen atom. Met. Ions Life Sei. 2009, 6, 151-178

FONTECILLA-CAMPS

158

(ESRF). A significant peak for the second active site metal ion was only observed in the anomalous difference m a p calculated f r o m data collected at λ = 1.733 A [21]. A similar analysis of D. vulgaris Miyazaki [NiFe]-hydrogenase confirmed this result [4]. In the latter study the assignment of the nickel next to the active site iron was also confirmed crystallographically.

2.3.

The Active Site at Higher Resolution

A 2.54 A resolution analysis of D. gigas [NiFe]-hydrogenase using a new crystal form and data collected at the ESRF, confirmed the presence of three non-protein ligands to the active site iron ion. However, the shape of the electron density peaks was elongated and could not correspond to monoatomic ligands [21]. Earlier, a F T I R spectroscopic analysis of Chromatium vinosum [NiFe]-hydrogenase (now Allochromatium vinosum) had revealed the existence of three intrinsic high frequency bands that shifted when the redox state of the enzyme was modified [22,23]. When bacteria were grown in a medium containing 13 C and 1 5 N, the isotopic effect made these bands shift in a way that indicated they corresponded to two cyanides and one carbon monoxide [24]. The two cyanide bands were found to arise from vibrationally coupled species [25]. Species that were EPR-silent could now also be characterized by F T I R spectroscopy. For instance, an EPR-silent diamagnetic unready oxidized species was identified and designated as Ni-SU. Because equivalent F T I R bands were subsequently obtained with the D. gigas enzyme (Figure 4), we proposed that they originated from the iron ligands, named L I , L2 and L3 that were modeled as diatomic species in our electron density maps (Figure 5) [21]. L3 sat in a hydrophobic pocket and, consequently, it was assigned to CO. On the other hand, LI and L2 were hydrogen-bonded to the protein (Figure 5), and were modelled as cyanides [26]. Initially, the nature of the diatomic iron ligands was controversial because in the 1.8 A resolution structure of the D. vulgaris Miyazaki hydrogenase a strong electron density peak corresponding to LI was assigned to SO [4). In order to explain the F T I R spectra, which were essentially identical to those of the other enzymes (S. Albracht, personal communication), L2 and L3 were assigned to a mixture of CO and CN~. However, there was no evidence for a SO ligand in the other structures [6,7,21,27], that had very similar electron density for the three diatomic ligands. The strong LI electron density observed in the first D. vulgaris Miyazaki hydrogenase structure remains unexplained. In fact, recent high-resolution crystal structures of this enzyme have very similar electron densities for L I , L2, and L3 [28,29]. A detailed F T I R study has shown that, indeed, D. vulgaris Miyazaki hydrogenase also contains a Fe(CN) 2 CO unit [30]. Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

159

wavenumber (cm 1 ) Figure 4. FTIR spectra of Desulfovibrio gigas [NiFe]-hydrogenase poised at different redox potentials A = -210mV , B=-300mV, C = -400mV, D = -500mV, and E = -600mV. For details see [26].

There are two coordination sites available for substrate binding at the active site: one that bridges the Ni and Fe ions, called El and one that is terminal to the Ni, called E2 (Figure 5). In the initial as-prepared (oxidized, unready) [NiFe]-hydrogenase structures El was modeled as occupied by either an oxo ligand [21] or by sulfide [4,7]. These models have been recently corrected, as we and others have found crystallographic evidence showing that in the unready oxidized hydrogenases there is a peroxo, rather than an Met. Ions Life Sei. 2009, 6, 151-178

160

FONTECILLA-CAMPS

\J

El 8

/

C65

Figure 5. The active site of Desulfovibrio gigas [NiFe]-hydrogenase and its environment at 2.56 A resolution [21]. The three diatomic ligands are labeled LI, L2, and L3. Based on their environments LI was modeled as CO and L2 and L3 as CN~ (see text).

oxo, ligand bound to El [21,31] (see below). On the other hand, the crystals of two reductively activated enzymes did not contain any detectable electron density corresponding to the El or E2 sites [6,32]. However, this does not mean that the sites are empty; hydride binding to one or both sites in the reduced active center would not be detected at the resolution of these analyses.

3.

H Y D R O G E N A S E MATURATION AND ACTIVE SITE ASSEMBLY

As stated in the Introduction, hydrogenases are metallo-enzymes that represent one of the most striking examples of convergent evolution; their active site iron ions share the unique feature of being coordinated, in the case Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

161

of the [NiFe] and [FeFe] enzymes, with CO and CN~ and, in [Fe]-hydrogenase, with two CO ligands [2,33]. The synthesis and insertion of these metal centers follow a complex pathway that requires the involvement of accessory proteins with novel biochemical properties.

3.1.

Maturation of the Large Subunit

The maturation of the large subunit of [NiFe]-hydrogenases requires at least seven proteins. The process can be divided into: (a) synthesis of the apoenzyme, (b) transport and storage of nickel and iron, (c) C O / C N " ligand synthesis/binding to iron and partial active site assembly, (d) insertion of nickel, (e) proteolysis of a C-terminal amino acid stretch, and (f) folding of the nascent C-terminal region into the rest of the large subunit. On the other hand, the Fe/S clusters are assembled by the general-purpose Iron sulfur cluster (Isc) scaffold proteins [34]. Hydrogenase synthesis depends on the environment. Under anoxic conditions, Escherichia coli undergoes a major re-organization of its metabolism through the expression of over 300 genes [35]. When hydrogen is present, some microorganisms can detect it thanks to a [NiFe]-hydrogenase-like sensor protein, also called regulatory hydrogenase. Subsequent activation of a kinase and a DNA-binding protein, results in the expression of [NiFe]-hydrogenase structural genes [36]. Both microbial iron and nickel transport are highly regulated in order to avoid Fenton chemistryinduced damage. E. coli and related bacteria have an A T P Binding Cassette (ABC) nickel transporter the corresponding proteins of which are coded by the nik operon. Nickel transport is regulated by NikR, another D N A binding protein of known three-dimensional structure [37]. Besides relying on specific proteins, iron transport depends on the presence of small molecules called siderophores. We have found that this may also be the case for nickel [38]. In subsequent steps, the active site ligands have to be synthesized and/or ligated to iron and the active site has to be completely assembled and inserted into the large subunit. This process is controlled by the six gene products from the hyp operon and a nickeldependent protease.

3.2.

Cyanide, Carbon Monoxide, and Iron Insertion

Several years ago, Barrett and coworkers [39] found that inactivation of the enzyme carbamoylphosphate (CP) synthase (PyrA) completely abolished hydrogenase activity in Salmonella typhimurium. However, the basis for this Met. Ions Life Sei. 2009, 6, 151-178

162

FONTECILLA-CAMPS

phenomenon was not further investigated. More recently, Paschos, Glass, and Böck showed that CN~ synthesis in E. coli requires two hyp operon gene products, namely, HypE and HypF [40]. HypF is an 82 kDa protein homologous to eukaryotic acylphosphatases at its N-terminal region and to O-carbamoyltransferases at its C-terminal end [41]. The homology of the N-terminal stretch of HypF to acylphosphatases has been confirmed by X-ray crystallography [42]. HypF hydrolyzes CP, transferring the carbamoyl moiety to the C-terminal Cys thiolate of HypE. The formation of a HypF-HypE complex has been shown experimentally [43-45]. HypE is a 35 kDa monomeric protein with amino acid sequence similarities to aminoimidazole ribonucleotide synthase (PurM) and selenophosphate synthase (SelD) [46]. The former catalyzes the dehydration of aminoimidazole ribonucleotide and requires ATP. Its ATP-binding motif is conserved in HypE. The process consists of the following steps: carbamoyl gets transferred by HypF to the HypE C-terminal cysteine, where it is dehydrated generating a HypE-CysSCN complex; next, and in sequential steps, two modified HypEs donate two CN~ ligands to the nascent Fe unit of hydrogenase active site. It was originally speculated that CP was also the source of CO [40]. However, this notion has been discredited by recent experiments using 1 3 C 0 2 because only CN~ was labeled ( C 0 2 is a CP precursor). Since CO was not labeled, its synthesis must follow a different pathway. Roseboom et al. have postulated that CO is synthesized from acetate or one of its precursors [47]. Another possibility is that CO is incorporated directly from the medium without having any specific synthetic pathway. HypC interacts very strongly with both the hydrogenase large subunit and with HypD. HypC is a 9.6 kDa protein with a CxxxP motif in its N-terminal region [48], HypD is a 41.4 kDa protein that has a [Fe4S4] cluster [47,49], The HypC-HypD complex is not detected in the presence of citrulline, a CP source. In the absence of nickel, HypC forms a complex with the hydrogenase large subunit HycE. It is thought that the HypC-HypD complex gets the CN~ ligands from HypE and then transfers them to the hydrogenase, a notion that is comforted by the fact that a HypC-HypD-HypE complex has been characterized [49]. A very recent crystallographic study of HypC, HypD, and HypE has provided a structural interpretation concerning the mechanism by which CN~ ligands get transfered to the prospective active site Fe (Figure 6). The [Fe4S4] cluster environment in HypD is similar to that of ferredoxin. Although under reducing conditions the complex between HypC and the hydrogenase large subunit precursor preHycE is stable, it dissociates when exposed to alkylating agents. This is consistent with the idea that HypC forms a covalent complex with preHycE by means of its C-terminal cysteine [48]. Nickel also plays a role in the dissociation of the HypC-preHycE complex. Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

163

Met. Ions Life Sei. 2009, 6, 151-178

164

3.3.

FONTECILLA-CAMPS

Nickel Insertion

A l t h o u g h , as shown by a H y p B - m u t a n t t h a t lacked hydrogenase activity, H y p B is involved in nickel transport/insertion, the enzymatic activity could be restored with added nickel in the growth medium. This suggests t h a t H y p B is n o t essential for nickel insertion [50]. W h e n nickel is replaced by zinc there is n o proteolytic m a t u r a t i o n showing that the protease is nickelspecific [48,51]. T h e structure of H y p B f r o m Μ. jannaschii has been reported recently [52]. Nickel insertion in C O dehydrogenase and urease is mediated by similar G T P a s e s [53,54]. Some H y p B s m a y store nickel ions using a histidine-rich stretch at their Ν termini. In the absence of this region, H y p B only binds one N i ion per protein molecule [55]. Indeed, after the histidinerich sequence, the N-terminal region of E. coli H y p B has a high-affinity Ni-binding site with a CxxCxxxxxC sequence [56]. H y p A interacts with H y p B and it m a y be the protein that provides nickel to H y c E . T h e H y p B G T P a s e activity m a y p r o v o k e the c o n f o r m a t i o n a l change needed for nickel insertion by H y p A . SlyD is a proline cis/trans isomerase that interacts with H y p B . As with the latter, the deletion of the corresponding gene lowers hydrogenase activity but nickel addition abolishes this effect.

3.4.

Proteolytic Cleavage of the Large Subunit C-Terminal Extension

This process represents the last step in the m a t u r a t i o n process [57]. D e p e n d i n g on the hydrogenase, the C-terminal extension can have between five and 32 a m i n o acid residues [58]. Thus, a b o u t two-thirds of the C-terminal 25-residue extension of the E. coli hydrogenase 3 can be genetically removed without consequences for either cleavage or subunit m a t u r a t i o n . O n the oher h a n d , f u r t h e r t r u n c a t i o n strongly reduces precursor stability [58]. T h e proteolytic enzyme is very specific and there is only one isoform for each hydrogenase in a given organism. T h e crystal structure of the 17.5 k D a H y b D protease is available [59]. Instead of nickel, the structure contains a c a d m i u m ion coordinated by the carboxylate oxygens of glutamic and aspartic acids, the imidazole nitrogen of a histidine and a water molecule. The as-purified H y b D does n o t contain metal [59,60], and in vitro it only cleaves its substrate u p o n nickel addition. F u r t h e r m o r e , the enzyme activity is insensitive to standard protease inhibitors [61,62]. These observations suggest that H y b D recognizes nickel b o u n d to the large subunit precursor and that c a d m i u m replaced physiological nickel in the crystal structure [62]. The m a t u r a t i o n endopeptidase Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

165

proteolytic site of different hydrogenases can vary being located between either His or A r g and a n o n p o l a r residue such as Met, lie, Val or Ala. Even this relative conservation is n o t a strict requirement, as shown by the fact that most m u t a t i o n s at these sites d o n o t prevent proteolysis. It is thus tempting to conclude t h a t the nickel ion, when b o u n d to the large subunit precursor, determines the regiospecificity of the cleavage site [58,62,63].

4.

ELECTRON TRANSFER

The identity of the hydrogenase physiological redox p a r t n e r depends on its intracellular location and on its eventual association with a n o t h e r catalytic entity. Examples of such associations are the cytoplasmic bidirectional h y d r o g e n : N A D ( P ) + oxidoreductases f o u n d in b o t h bacteria and archaea. M o n o f u n c t i o n a l periplasmic hydrogenases, which are mostly hydrogenoxidizing enzymes, transfer electrons t h r o u g h o t y p e cytochromes to the cytoplasmic side of the m e m b r a n e . The resulting p r o t o n s remain in the periplasm and contribute to the f o r m a t i o n of an ATP-generating transm e m b r a n e p r o t o n gradient. All the available [NiFe(Se)]-hydrogenase structures correspond to periplasmic u p t a k e enzymes of sulfate-reducing bacteria. In enzymes f r o m Desulfovibrio species, electrons move f r o m the active site to the redox partner t h r o u g h a proximal [Fe 4 S 4 ], a mesial [Fe 3 S 4 ] and a distal [Fe 4 S 4 ] cluster (see Figure 2 in Section 2.1). Consecutive redox centers in this p a t h w a y are separated by center-to-center distances of a b o u t 12 A , which are typical for electron transfer [64]. The mesial position of the [Fe 3 S 4 ] cluster is surprising given the fact t h a t its redox potential is m u c h m o r e positive t h a n that of hydrogen oxidation [65]. Indeed, this cluster should remain reduced during catalysis. It m a y be t h a t reduction of either the proximal or the distal [Fe 4 S 4 ] cluster transiently decreases the mesial cluster redox potential m a k i n g it m o r e compatible with electron transfer in hydrogenase. In order to elucidate the role of the [Fe 3 S 4 ] mesial cluster we converted it to a [Fe 4 S 4 ] center in the D. fructosovorans [NiFe]-hydrogenase [66]. The m u t a t e d enzyme h a d wild-type activity indicating that internal electron transfer is n o t rate-limiting. T h e reaction rate must then depend on the other processes such as substrate access to the active site, hydrogen heterolytic cleavage, p r o t o n transfer f r o m the active site to the solution or electron transfer to a redox partner. Studies with different electron acceptors suggest that the catalytic rate is defined by their interaction with hydrogenase [67]. [NiFe]-hydrogenases are good examples of enzymes having specific redox partner recognition sites. Thus, a crown of acidic residues surrounds the His ligand of the distal [Fe 4 S 4 ] cluster in different Desulfovibrio species. This Met. Ions Life Sei. 2009, 6, 151-178

166

FONTECILLA-CAMPS

negatively-charged crown is likely to interact with positively-charged patches found near one of the heme groups on both soluble and membranebound otype cytochromes [3]. The redox complex is probably short-lived, given the binding K m values which are typically micromolar [68]. Electrons are ultimately discarded through the reduction of a varity of cytoplasmic terminal acceptors, such as sulfate in sulfate-reducing bacteria.

5.

PROTON TRANSFER

Several years ago, [NiFe]-hydrogenase three-dimensional structures were used to propose plausible proton tranfer (PT) pathways. In the 2.8 A resolution structure of an oxidized form of the D. gigas enzyme (pdb entry 1FRV), the side chains of four histidines, two water molecules and a carboxylate group formed a plausible pathway for proton transfer from the Ni-Fe bridging Cys533 ligand to the protein surface [3]. However, this proposition had two problems: a subsequent study at 2.54 A resolution [21] showed that one of these residues, modelled as histidine, was leucine instead (M. Rousset, personal communication). In addition, electron density that bridged the side chains of the C-terminal His residue and a Glu carboxylate was modelled as water but, in reality, it corresponded to the Mg(II) ion discussed above (pdb entry 2FRV). The C-terminal His seems to help stabilizing the C-terminal region, which may be its main role, rather than PT. A very recent study on PT pathways obtained using a quantum mechanical/ molecular mechanical (QM/MM) approach is shown in Figure 7. As mentioned above, in Dm. baculatum [NiFeSe]-hydrogenase the terminal Ni ligand of the D. gigas enzyme, Cys530, is replaced by a SeCys. The 2.1 A resolution crystal structure of the active, reduced form of this enzyme indicated that the Se atom forms a hydrogen bond with the carboxylate of a glutamate side chain, corresponding to Glul8 in D. gigas. This residue is conserved in all [NiFe]-hydrogenases [6]. The SeCys and Glu residues, along with the Ni ion, have temperature (B) factors that are higher than those of atoms in their surroundings. This may be due to a mixture of protonation states of these species in the crystal structure. Similar high B-factors for these two residues have been found in other crystal structures. Consequently, the (Se)Cys and Glu residues are probably constituents of the proton transfer pathway in all [NiFe]-hydrogenases. In fact, in oxidized forms of D. desulfuricans ATCC 2111A [7] and D.fructosovorans [NiFe]-hydrogenases [27] the thiolate of the cysteine residue corresponding to Cys530 in D. gigas displays more than one conformation. There is a precedent for proton transfer from a Ni site to a thiolate ligand in a model compound [69]. Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

167

Figure 7. Two possible proton transfer pathways (A and B) in D. fructosovorans [NiFe]-hydrogenase based on a semi-empirical quantum mechanical/molecular mechanical study [94]. Similar results have been recently obtained by Texeira et al. [95].

Met. Ions Life Sei. 2009, 6, 151-178

168

FONTECILLA-CAMPS

T h e Ni-based E P R spectroscopic properties of the D. fructosovorans enzyme were n o t modified by the m u t a t i o n of the glutamate, corresponding to G l u l 8 in D. gigas, to Gin. However, the m u t a t e d enzyme was catalytically inactive and did n o t p e r f o r m the H / D isotope exchange. Only the para-H2/ ortho-H2 conversion was still detectable [70]. These results imply t h a t although the m u t a t e d enzyme still catalyzed the heterolytic cleavage of H 2 , the p r o t o n / d e u t e r i o n exchange between solvent and the active site did n o t take place. Thus, as potulated f r o m the crystallographic results, G l u l 8 is essential for p r o t o n transfer. Matias et al. have p r o p o s e d a p r o t o n transfer p a t h w a y based on the crystal structure of oxidized D. desulfuricans A T C C 27774 [NiFe]-hydrogenase [7],

6.

OXIDIZED INACTIVE STATES OF THE [NiFe]HYDROGENASE ACTIVE SITE

As mentioned above, hydrogenases can be reversibly inactivated by oxygen. Recent revisions and higher resolution oxidized hydrogenase structures have revealed a series of modifications of the thiolates of the cysteine ligands of their active sites. T h e 1.8 A resolution o r t h o r h o m b i c crystal structure of the as-prepared D. fructosovorans enzyme we reported in 1998 [71] has been recently re-analyzed using q u a n t u m refinement and our X-ray diffraction d a t a by Söderhjelm and R y d e [72]. Their results indicate that Cys68 and Cys530 (D. gigas numbering), were 5 % and 2 0 % modified to sulfenic acid, respectively. In addition to this o r t h o r h o m b i c crystal structure, a 1.8 A resolution monoclinic crystal structure of as-prepared hydrogenase was originally published by us in 2002 [27]. M o r e recently, we have revised the structure of this monoclinic f o r m [31] and f o u n d t h a t , as in the revised o r t h o r h o m b i c structure described above [72], its active site has b o t h an alternative c o n f o r m a t i o n of Cys530 and the 3 0 % modification of Cys68 to sulfenic acid. Our revision also included the proposition t h a t the bridging E2 site is a b o u t 7 0 % occupied by a peroxide species (replacing the oxo ligand we had reported earlier for this site). Similarly, in the recent crystal structure of an u n r e a d y state of D. vulgaris Miyazaki [NiFe]-hydrogenase, the thiolates of the side chains corresponding to D. gigas Cys68 and Cys530 are 53% and 39% modified to sulfenates, respectively; there is also a 59% occupied N i F e bridging putative peroxo ligand [29]. Thus, the active site seems to be structurally heterogenous in all the crystals containing oxidized u n r e a d y enzyme. There is a consensus that the oxidized ready Ni-B f o r m has Ni(III) and a bridging h y d r o x o ligand [31,73,74] (Figure 8A). In our recently revised crystal structure of as-prepared enzyme (Figure 8B), the two a t o m s of the peroxo ligand d o n o t refine Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

169

Figure 8. Different states of the active site of D. fructosovorans [NiFe]-hydrogenase. (A) the putative Ni-B form; (B) putative Ni-A species with a bridging (hydro)peroxo ligand; (C) a sulfenate-containing structure. Results taken from [29] and [31].

t o t h e s a m e o c c u p a n c y : t h e o x y g e n a t o m closer ( p r o x i m a l ) t o t h e F e i o n a p p e a r s t o be m o r e o c c u p i e d t h a n t h e distal o n e . T h i s , in t u r n , i n d i c a t e s t h a t t h e p r o x i m a l site is likely t o be o c c u p i e d by a ( h y d r ) o x o l i g a n d in a sign i f i c a n t p e r c e n t a g e of h y d r o g e n a s e m o l e c u l e s . So, w h e n c o m p a r e d t o t h e N i B f o r m , t h e s t r u c t u r e s of t h e u n r e a d y N i - A a n d N i - S U states a p p e a r n o t t o be as well d e f i n e d . B o t h t h e p e r o x o l i g a n d a n d cysteine s u l f e n a t e s ( F i g u r e 8C) c o u l d in p r i n c i p l e be s i g n a t u r e s of t h e u n r e a d y state. T h e f o l l o w i n g r e a c t i o n s m a y explain t h e o b s e r v e d s t r u c t u r e s . A t r o o m t e m p e r a t u r e , t h e r e a c t i o n of m o l e c u l a r o x y g e n w i t h partially r e d u c e d h y d r o g e n a s e is likely t o g e n e r a t e a ( h y d r o ) p e r o x o b r i d g i n g species a c c o r d i n g to:

0 2 + N i ( I I ) + e " + H + -s· N i ( I I I ) - O O H "

(2)

T a k i n g t h e r e d o x p o t e n t i a l s i n t o a c c o u n t , t h e e l e c t r o n in r e a c t i o n (2) is likely t o c o m e f r o m t h e o x i d a t i o n of t h e [Fe 3 S 4 ]° cluster t o [Fe 3 S 4 ] 1 + . T h e Met. Ions Life Sei. 2009, 6, 151-178

FONTECILLA-CAMPS

170

resulting (hydro)peroxo ligand can subsequently react with a cysteine thiolate to yield a bridging (hydr)oxo and a sulfenate according to: Ni(III)-OOH" + Cys-S~

Ni(III)-OH" + Cys-SCT

(3)

We believe that reaction (3) leads to dead enzyme as suggested by the fact that anaerobically purified hydrogenases are much more active than their reductively re-activated, aerobically purified counterparts [31]. In addition, sulfenic acid formation according to reaction (3) might explain the gradual irreversible enzyme inactivation that is observed in electrochemical studies [F. Armstrong, personal communication]. Indeed, according to the recent D F T analysis of [NiFe]-hydrogenase discussed above, the AG 0 of reaction (3) is about -180kJ/mol [72], which is more than twice as high as the activation energy of + 88 kJ/mol reported for the enzyme in the unready state [26,75]. An -SO ligand will stabilize the lower accessible redox state of Ni, in this case Ni(II), because sulfenate is not as good a σ donor as thiolate [76]. We have previously argued that the (hydro)peroxo bridging species is the signature responsible for the unready Ni-A EPR signal [31]. However, the situation is complicated by the fact that (i) E N D O R results from Carepo et al. indicated that after the oxidation of the Ni-C state by air, the labeled oxygen atom of H 2 1 7 0 from the solvent medium is bound to Ni in the Ni-A form [77], and (ii) 1 7 0 2 also modifies the EPR spectra of the oxidized states of hydrogenases [78]. This leads to the need of an exchange reaction between the putatively 0 2 -generated (hydro)peroxo ligand and water. A possibility is provided by reaction (4) [79] were it is assumed that the exchanging species is Ni-SU, a diamagnetic form, one electron more reduced than Ni-A (Figure 9). Indeed, recent electrochemical results postulate that in the activation process the unready Ni-A state is first reduced to the diamagnetic Ni-SU and subsequently undergoes a slow, reversible step to a transient state [80]. Reaction (4) is analogous to the exchange reaction of peroxide with water in the model heme catalyst, microperoxidase-8 (the Fe(II) ion has been omitted for clarity) [79]: Ni(II)-OOH" + e" + H+ -s· Ni(III)-0 2 ~ + H 2 0 Ni(II)-OOH" + e" + H+

(4)

The (hydro)peroxo-containing species comes from equation (2) and the bridging peroxo ligand will be labeled if this reaction is carried out in the presence of 1 7 0 2 . Conversely, if activation takes place in the presence of H 2 1 7 0 , the labeled water oxygen atom will exchange as in equation (4). The electron would come from the oxidation of [Fe3S4]°. Reaction (4) explains the observed absence of H 2 1 7 0 exchange in the Ni-A state [77] Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

/ Fe(ll) / Fe(II)

ο η ο> υ

/ Fe(II)

ΟΟΗ \ Νί(ΠΙ) "ΟΟΗ \ Ν1(ΙΙ) ΟΗ

\

Fe(II) / Fe(II) / Fe(II)

HH"

\ \

171

Unready Ni-A* Unready Ni-SU

Νϊ(ΙΠ)

Ready Ni-B*

Ni(II)

Silent Ni-SI

Ni(Iil)

Active Ni-C*

Ni(II)

Active Ni-R

Figure 9. Postulated redox states of the different stable active site intermediates of [NiFe]-hydrogenase. Adapted f r o m [91]. "?" indicates that the bridging site E2 may be occupied in Ni-SI.

because in this form the enzyme has no electrons available to form Ni(HI)-0 2 -. In the active site structures of unready hydrogenase, both oxygen atoms of the peroxo ligand are at a binding distance from Ni, which can explain the previous 1 7 0 2 EPR and H 2 1 7 0 E N D O R spectroscopic results. The possibility of Ο exchange between H 2 0 and a peroxo ligand at the hydrogenase active site was evoked by Lamle, Albracht, and Armstrong [80]. As a function of their environment, microorganisms have hydrogenases with various degrees of oxygen tolerance. Understanding the structural basis of this resistance is central to ongoing, technologically-oriented, hydrogenase research. To this date, only hydrogenase structures from sulfatereducing bacteria are known. They all have very limited oxygen tolerance. On the other hand, the most remarkable and best characterized oxygenresistant enzymes come from the Ralstonia Knallgas bacteria for which no structure is available. Ralstonia eutropha (now R. cuprivorans) has a membrane-bound [NiFe]-hydrogenase coupled to the respiratory chain and a cytoplasmic soluble enzyme, which oxidizes H 2 and reduces N A D + . [81]. The latter is postulated to have two extra CN~ ligands, one on each of the metal ions of the active site. In addition, the Ni ion would have O/N coordination [82]. The additional CN~ ligands would be involved in 0 2 resistance but may dissociate during catalysis. Met. Ions Life Sei. 2009, 6, 151-178

172

7.

FONTECILLA-CAMPS

SUBSTRATE BINDING AND CATALYSIS

As shown in e q u a t i o n (1) the reaction catalyzed by hydrogenases involves three components: molecular hydrogen, p r o t o n s , and electrons. In addition, hydrogen cleavage is heterolytic, implying t h a t the active site must also bind hydride. These species are undetectable by X-ray crystallography at the resolutions of the reported studies. Consequently, X-ray crystallography can only provide indirect evidence of substrate binding. Several stable N i - F e active site intermediates have been characterized by b o t h E P R and F T I R spectroscopy (see Figures 1 and 4 in Sections 1 and 2.2, respectively). Above, we have addressed the crystallographic d a t a concerning the n a t u r e of oxygenic ligands in the active site of u n r e a d y (Ni-A and N i - S U ) and ready enzyme f o r m s (Ni-B). N o exogenous ligands are evident in the crystal structures of reduced, active hydrogenases [6,32]. However, a hydride is likely to occupy the E2 site (see Figure 5 in Section 2.2) making the N i c o o r d i n a t i o n square pyramidal. Also, the N i - F e distance in the reduced enzyme is 2.6 A, consistent with the binding of a bridging hydride [83]. D u e to electron and p r o t o n transfer reactions, different N i - F e active site states are at t h e r m o d y n a m i c equilibrium m a k i n g it difficult to obtain hydrogenase crystals in h o m o g e n e o u s states. Theoretical fitting of D. gigas [NiFe]-hydrogenase redox titrations favored the two electron difference model with the active N i - C f o r m being two electrons m o r e reduced t h a n the oxidized ready Ni-B state [84]. These two E P R active states are considered to be Ni(III) species [85] implying that the two-electron difference arises f r o m the presence of a hydride in Ni-C. Indeed, using H Y S C O R E and E N D O R spectroscopies, Brecht et al. f o u n d direct evidence for b o u n d hydride in the N i - C state of the regulatory [NiFe]-hydrogenase f r o m R. eutropha [86]. In addition, the g-tensor orientation obtained in a single-crystal E P R study of the N i - C state of D. vulgaris Miyazaki hydrogenase indicated that the hydride was b o u n d to the N i - F e bridging E2 position [87]. This result has been confirmed by a m o r e recent E N D O R / H Y S C O R E study [88], Intriguingly, only the N i - C / N i - R and one [Fe 4 S4] 2 + /[Fe 4 S4]~ couples were f o u n d to be at redox equilibrium with molecular hydrogen [84,89]. This result implies that the enzyme in the hydride-bound N i - C state can react with hydrogen, and that, consequently, hydride, molecular hydrogens, and protons could occupy b o t h E l and E2 in a transient f o r m , two electrons m o r e reduced t h a n Ni-C. This transient f o r m would rapidly oxidize to the N i - R f o r m . There is another observation favoring a putative hydride b o u n d at the bridging E2 site in the N i - C species and an additional hydrogen binding site at E l : in the crystal structure of D. vulgaris Miyazaki F hydrogenase the added C O inhibitor binds terminally to the N i at the E l site [28]. This result agrees with I R spectroscopic d a t a indicating that the vibrational frequency

Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

173

of extrinsic bound CO did not correspond to a bridging binding mode in two [NiFe]-hydrogenases [22,90].

8.

CONCLUDING REMARKS

Hydrogen metabolism is mediated by enzymes that coordinate at least one Fe(CO) x unit. This convergent evolutionary event is remarkable because CO, like CN~, is normally toxic to the cells. The diatomic ligands make the iron ion in [NiFe]-hydrogenases both low-spin and redox inactive. It can stay in the 2 + valence state through catalysis thanks to the π-accepting properties of CO and hydrogen-bonded CN~. In addition, the CN~ ligands serve to anchor the iron centers to the protein in both [NiFe]- and [FeFe]hydrogenases. An electron-rich, low spin iron ion has properties that are similar to those of 2 n d and 3 r d row transition metals, which are known to be good hydrogen catalysts and hydride binders. Thus, Nature has found a cheap solution to hydrogen use and production. Similar approaches may prove useful for the future biotechnological production of this gas as well as its utilization in fuel cells. A more general review on hydrogenases in general can be found in [91].

ACKNOWLEDGMENTS I thank the Commissariat ä l'Energie Atomique and the Centre National de la Recherche Scientifique for institutional support. I also thank Dr. Patricia Amara for critical reading of the manuscript and valuable help with the figures.

ABBREVIATIONS ABC CP DFT ENDOR EPR ESRF EXAFS FMN

ATP-binding cassette carbamoylphosphate density functional theory electron nuclear double resonance electron paramagnetic resonance European Synchrotron Radiation Facility extended X-ray absorption fine structure flavin mononucleotide

Met. Ions Life Sei. 2009, 6, 151-178

174 FTIR HYSCORE Isc MIR NAD(P) PT PurM PyrA QM/MM SelD TED

FONTECILLA-CAMPS Fourier transform infrared h y p e r f i n e sublevel c o r r e l a t i o n i r o n s u l f u r cluster multiple isomorphous replacement nicotinamide adenine dinucleotide (phosphate) proton transfer aminoimidazole ribonucleotide synthase carbamoylphosphate synthase q u a n t u m mechanical/molecular mechanical selenophosphate synthase two electron difference

REFERENCES 1. J. P. Collman, Nat. Struct. Biol., 1996, 3, 213-217. 2. S. Shima, E. J. Lyon, R. K. Thauer, B. Mienert and E. Bill, J. Am. Chem. Soc., 2005, 43, 10430-10435. 3. A. Yolbeda, Μ. H. Charon, C. Piras, E. C. Hatchikian, M. Frey and J. C. Fontecilla-Camps, Nature, 1995, 373, 580-587. 4. Y. Higuchi, T. Yagi and N. Yasuoka, Structure, 1997, 5, 1671-1680. 5. Y. Montet, P. Amara, A. Yolbeda, X. Vernede, E. C. Hatchikian, M. J Field, M. Frey and J.C. Fontecilla-Camps, Nat. Struct. Biol., 1997, 4, 523-526. 6. E. Garcin, X. Vernede, E. C. Hatchikian, A. Volbeda, M. Frey and J. C. Fontecilla-Camps, Structure, 1999, 7, 557-566. 7. P. M. Matias, C. M. Soares, L. M. Saraiva, R. Coelho, J. Morais, J. Le Gall and M. A. Carrondo, J. Biol. Inorg. Chem., 2001, 6, 63-81. 8. Y. Nicolet, C. Piras, P. Legrand, C. E. Hatchikian and J. C. Fontecilla-Camps, Structure, 1999, 7, 13-23. 9. J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt, Science, 1998, 282, 1853-1858. 10. V. Niviere, E. C. Hatchikian, C. Cambillaud and M. Frey, J. Mol. Biol., 1987, 195, 969-971. 11. W. W. Smith, R. M. Burnett, G. D. Darling and M. L. Ludwig, J. Mol. Biol., 1977, 117, 195-225. 12. Ν. K. Menon, J. Robbins, M. DerVartanian, D. Patil, H. D. Peck Jr., A. L. Menon, R. L. Robson and A. E. Przybyla, FEBS Lett., 1993, 331, 91-95. 13. O. Sorgenfrei, D. Linder, Μ. Karas and A. Klein, Eur. J. Biochem., 1993, 213, 1355-1358. 14. A. Volbeda, C. Piras, M.-H. Charon, E. C. Hatchikian, M. Frey and J. C. Fontecilla-Camps, ESF/CCP4 Newslett., 1993, 28, 30-33. 15. Y. Higuchi, T. Okamoto, K. Fujimoto and S. Misaki, Acta Cryst. D, 1994, 50, 781-785. 16. S. P. J. Albracht, Biochim. Biophys. Acta, 1993, 1144, 221-224.

Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

175

17. S. H. He, M. Teixeira, J. Le Gall, D. S. Patil, I. Moura, J. J. Moura, D. Y. DerVartanian, Β. H. Huynh and H. D. Peck Jr., J. Biol. Chem., 1989, 264, 2678-2682. 18. Μ. K. Eidsness, R. A. Scott, B. C. Prickril, D. Y. DerVartanian, J. LeGall, I. Moura, J. J. Moura and H. D. Peck Jr., Proc. Natl. Acad. Sei. USA, 1989, 86, 147-151. 19. O. Sorgenfrei, A. Klein and S. P. J. Albracht, FEBS Lett., 1993, 332, 291-297. 20. E. C. Hatchikian, M. Bruschi and J. LeGall, Biochem. Biophys. Res. Commun., 1978, 82, 451-461. 21. A. Yolbeda, E. Garcin, C. Piras, A. L. De Lacey, Υ. M. Fernandez, E. C. Hatchikian, M. Frey and J. C. Fontecilla-Camps, J. Am. Chem. Soc., 1996, 118, 12989-12996. 22. K. A. Bagley, C. J. Van Garderen, Μ. Chen, Ε. C. Duin, S. P. Albracht and W. H. Woodruff, Biochemistry, 1994, 33, 9229-9236. 23. K. A. Bagley, E. C. Duin, W. Roseboom, S. P. J. Albracht and W. H. Woodruff, Biochemistry, 1995, 34, 5527-5535. 24. R. P. Happe, W. Roseboom, A. J. Pierik, S. P. J. Albracht and K. A. Bagley, Nature, 1997, 385, 126. 25. A. J. Pierik, W. Roseboom, R. P. Happe, K. A. Bagley and S. P. J. Albracht, J. Biol. Chem., 1999, 274, 3331-3337. 26. A. L. De Lacey, E. C. Hatchikian, A. Yolbeda, M. Frey, J. C. Fontecilla-Camps and V. M. Fernandez, J. Am. Chem. Soc., 1997, 119, 7181-7189. 27. A. Volbeda, Y. Montet, X. Vernede, E. C. Hatchikian and J. C. FontecillaCamps, Int. J. Hydr. Energy, 2002, 27, 1449-1461. 28. H. Ogata, Y. Mizoguchi, N. Mizuno, K. Miki, S. Adachi, N. Yasuoka, T. Yagi, O. Yamauchi, S. Hirota and Y. Higuchi, J. Am. Chem. Soc., 2002, 124, 11628-11635. 29. H. Ogata, S. Hirota, A. Nakahara, H. Komori, N. Shibata, T. Kato, K. Kano and Y. Higuchi, Structure, 2005, 13, 1635-1642. 30. C. Fichtner, C. Laurich, Ε. Bothe and W. Lubitz, Biochemistry, 2006, 45, 9706-9716. 31. A. Volbeda, L. Martin, C. Cavazza, M. Matho, B. W. Faber, W. Roseboom, S. P. Albracht, E. Garcin, M. Rousset and J. C. Fontecilla-Camps, J. Biol. Inorg. Chem., 2005, 10, 239-249. 32. Y. Higuchi, H. Ogata, K. Miki, N. Yasuoka and T. Yagi, Structure, 1999, 7, 549-556. 33. E. J. Lyon, S. Shima, R. Boecher, R. K. Thauer, F. W. Grevels, Ε. Bill, W. Roseboom and S. P. J. Albracht, J. Am. Chem. Soc., 2004,126, 14239-14248. 34. D. C. Johnson, D. R. Dean, A. D. Smith and Μ. K. Johnson, Ann. Rev. Biochem., 2005, 74, 247-281. 35. P. J. Kiley and H. Beinert, FEMS Microbiol. Rev., 1999, 22, 341-352. 36. P. M. Vignais and A. Colbeau, Curr. Issues Mol. Biol., 2004, 6, 159-188. 37. E. R. Schreiter, S. C. Wang, D. B. Zamble and C. L. Drennan, Proc. Natl. Acad. Sei. USA, 2006, 103, 13676-13681. 38. Μ. V. Cherrier, L. Martin, C. Cavazza, L. Jacquamet, D. Lemaire, J. Gaillard and J. C. Fontecilla-Camps, J. Am. Chem. Soc., 2005, 127, 10075-10082.

Met. Ions Life Sei. 2009, 6, 151-178

176

FONTECILLA-CAMPS

39. Ε. L. Barrett, Η. S. Kwan and J. Macy, J. Bacteriol., 1984, 158, 972-977. 40. A. Paschos, R. S. Glass and A. Böck, FEBS Lett., 2001, 488, 9-12. 41. I. Wolf, T. Buhrke, J. Dernedde, A. Pohlmann and B. Friedrich, Arch. Microbiol., 1998, 170, 451-459. 42. C. Rosano, S. Zuccotti, M. Bucciantini, M. Stefani, G. Ramponi and M. Bolognesi, J. Mol. Biol., 2002, 321, 785-796. 43. J. C. Rain, L. Selig, Η. De Reuse, Y. Battaglia, C. Reverdy, S. Simon, G. Lenzen, F. Petel, J. Wojcik, V. Schachter, Y. Chemama, A. Labigne and P. Legrain, Nature, 2001, 409, 211-215. 44. M. Blokesch, A. Paschos, A. Bauer, S. Reissmann, Ν. Drapal and A. Böck, Eur. J. Biochem., 2004, 271, 3428-3436. 45. A. K. Jones, O. Lenz, Α. Strack, Τ. Buhrke and B. Friedrich, Biochemistry, 2004, 43, 13467-13477. 46. C. Li, T. J. Kappock, J. Stubbe, Τ. M. Weaver and S. E. Ealick, Structure, 1999, 7, 1155-1166. 47. W. Roseboom, Μ. Blokesch, Α. Böck and S. P. J. Albracht, FEBS Lett., 2005, 579, 4 6 9 ^ 7 2 . 48. A. Magalon and A. Böck, J. Biol. Chem., 2000, 275, 21114-21120. 49. M. Blokesch, S. P. J. Albracht, B. F. Matzanke, Ν. M. Drapal, A. Jacobi and A. Böck, J. Mol. Biol., 2004, 344, 155-167. 50. R. Waugh and D. H. Boxer, Biochimie, 1986, 68, 157-166. 51. T. Maier, A. Jacobi, M. Sauter and A. Böck, J. Bacteriol., 1993, 175, 630-635. 52. R. Gasper, A. Scrima and A. Wittinghofer, J. Biol. Chem., 2006, 281, 27492-27502. 53. Μ. H. Lee, S. B. Mulrooney, M. J. Renner, Y. Markowicz and R. P. Hausinger, J. Bacteriol., 1992, 174, 4324-4330. 54. R. L. Kerby, P. W. Ludden and G. P. Roberts, J. Bacteriol., 1997, 179, 2259-2266. 55. J. W. Olson and R. J. Maier, J. Bacteriol., 2000, 182, 1702-1705. 56. M. R. Leach, S. Sandal, H. Sun and D. B. Zamble, Biochemistry, 2005, 44, 12229-12238. 57. R. Böhm, Μ. Sauter and A. Böck, Mol. Microbiol., 1990, 4, 231-243. 58. Ε. Theodoratou, R. Huber and Α. Böck, Biochem. Soc. Trans., 2005, 33, 108-111.

59. E. Fritsche, A. Paschos, H. G. Beisel, A. Böck and R. Huber, J. Mol. Biol., 1999, 288, 989-998. 60. R. Rossmann, T. Maier, F. Lottspeich and A. Böck, Eur. J. Biochem., 1995, 227, 545-550. 61. A. L. M e n o n and R. L. Robson, J. Bacteriol., 1994, 176, 291-295. 62. E. Theodoratou, A. Paschos, A. Magalon, E. Fritsche, R. Huber and A. Böck, Eur. J. Biochem., 2000, 267, 1995-1999. 63. E. Theodoratou, A. Paschos, S. Mintz-Weber and A. Böck, Arch. Microbiol., 2000, 173, 110-116. 64. D. Leys and Ν. S. Scrutton, Curr. Opin. Struct. Biol., 2004, 14, 642-647. 65. M. Teixeira, I. Moura, Α. V. Xavier, J. J. G. Moura, J. LeGall, D. V. DerVartanian and H. D. Peck Jr., J. Biol. Chem., 1989, 264, 16435-16450.

Met. Ions Life Sei. 2009, 6, 151-178

STRUCTURE AND FUNCTION OF [NiFe]-HYDROGENASES

177

66. M. Rousset, Y. Montet, B. Guigliarelli, N. Forget, M. Asso, P. Bertrand, J. C. Fontecilla-Camps and E. C. Hatchikian, Proc. Natl. Acad. Sei. USA, 1998, 95, 11625-11630. 67. P. Bertrand, F. Dole, Μ. Asso and B. Guigliarelli, J. Biol. Inorg. Chem., 2000, 5, 682-691. 68. V. Niviere, E. C. Hatchikian, P. Bianco and J. Haladjian, Biochim. Biophys. Acta, 1988, 935, 3 4 ^ 0 . 69. W. Clegg and R. A. Henderson, Inorg. Chem., 2002, 41, 1128-1135. 70. S. Dementin, Β. Burlat, A. L. De Lacey, A. Pardo, G. Adryanczyk-Perrier, B. Guigliarelli, V. M. Fernandez and M. Rousset, J. Biol. Chem., 2004, 279, 10508-10513. 71. Y. Montet, Ph.D. Thesis, Universite Joseph Fourier, Grenoble, 1998. 72. P. Söderhjelm and U. Ryde, J. Mol. Struct. Theochem., 2006, 770, 199-219. 73. M. Stein, E. Van Lenthe, Ε. J. Baerends and W. Lubitz, J. Am. Chem. Soc., 2001, 123, 5839-5840. 74. C. Stadler, A. L. De Lacey, Y. Montet, A. Volbeda, J. C. Fontecilla-Camps, J. C. Conesa and V. M. Fernandez, Inorg. Chem., 2002, 41, 4424-4434. 75. V. M. Fernandez, E. C. Hatchikianand and R. Cammack, Biochim. Biophys. Acta, 1985, 832, 69-79. 76. P. J. Farmer, J. H. Reibenspies, P. A. Lindahl and Μ. Y. Darensbourg, J. Am. Chem. Soc., 1993, 115, 4 6 6 5 ^ 6 7 4 . 77. M. Carepo, D. L. Tierney, C. D. Brondino, T. C. Yang, A. Pamplona, J. Telser, I. Moura, J. J. M o u r a and Β. M. H o f f m a n , J. Am. Chem. Soc., 2002, 124, 281-286.

78. J. W. Van der Zwaan, J. M. Coremans, E. C. Bouwens and S. P. J. Albracht, Biochim. Biophys. Acta, 1990, 1041, 101-110. 79. J. L. Primus, K. Teunis, D. M a n d o n , C. Yeeger and I. M. C. M. Rietjens, Biochem. Biophys. Res. Commun., 2000, 272, 551-556. 80. S. E. Lamle, S. P. J. Albracht and F. A. Armstrong, J. Am. Chem. Soc., 2004, 126, 14899-14909. 81. T. Buhrke, M. Brecht, W. Lubitz and B. Friedrich, J. Biol. Inorg. Chem., 2002, 7, 897-908. 82. B. Blejlevens, T. Buhrke, E. van der Linden, B. Friedrich and S. P. J. Albracht, J. Biol. Chem., 2004, 279, 46686-46691. 83. A. Ceriotti, P. Chini, A. Fumagalli, T. F. Koetzle, G. Longoni and F. Tagusagawa, Inorg. Chem., 1984, 23, 1363-1368. 84. L. M. Roberts and P. A. Lindahl, J. Am. Chem. Soc., 1995, 117, 2565-2572. 85. J. C. Salerno, in The Bioinorganic Chemistry of Nickel, Ed. J. R. Lancaster, VCH, Weinheim, F R G , 1988, pp. 53-71. 86. M. Brecht, Μ. Van Gastel, T. Buhrke, B. Friedrich and W. Lubitz, J. Am. Chem. Soc., 2003, 125, 13075-13083. 87. S. Foerster, M. Stein, M. Brecht, Η. Ogata, Y. Higuchi and W. Lubitz, J. Am. Chem. Soc., 2003, 125, 83-93. 88. S. Foerster, M. Van Gastel, M. Brecht and W. Lubitz, J. Biol. Inorg. Chem., 2005, 10, 51-62.

Met. Ions Life Sei. 2009, 6, 151-178

178

FONTECILLA-CAMPS

89. J. Μ. C. C. Coremans, C. J. Van Garderen and S. P. J. Albracht, Biochim. Biophys. Acta, 1992, 1119, 148-156. 90. A. L. De Lacey, C. Stadler, Υ. M. Fernandez, E. C. Hatchikian, H.-J. Fan, S. Li and Μ. B. Hall, J. Biol. Inorg. Chem., 2002, 7, 318-326. 91. J. C. Fontecilla-Camps, A. Yolbeda, C. Cavazza and Y. Nicolet, Chem Rev., 2007, 107, 4273-4303. 92. C. Bagyinka, J. P. Whitehead and M. J. Maroney, J. Am. Chem. Soc., 1993, 115, 3576-3585. 93. S. Watanabe, R. Matsumi, T. Arai, H. Atomi, T. Imanaka and K. Miki, Mol. Cell, 2007, 27, 2 9 ^ 0 . 94. I. Fdez. Galvan, A. Volbeda, J. C. Fontecilla-Camps and M. J. Field, Proteins: Struct. Funct. and Bioinformatics, DOI: 10.1002/prot.2204 (published online: April 15, 2008). 95. V. H. Texeira, C. M. Soares and A. M. Batista, Proteins, 2008, 70, 1010-1022. 96. W. Humphrey, A. Dalke and K. Schulten, J. Molec. Graphics, 1996, 14, 33-38. 97. S. Shima, O. Pilak, S. Vogt, Μ. Schick, Μ. S. Stagni, W. Meyer-Klaucke, E. Warkentin, R. K. Thauer and U. Ermler, Science, 2008, 321, 572-575.

Met. Ions Life Sei. 2009, 6, 151-178

Met. Ions Life Sei. 2009, 6, 179-218

6 Carbon Monoxide and Cyanide Ligands in the Active Site of [FeFe]-Hydrogenases John W. Peters Montana State University, Department of Chemistry and Biochemistry and the Astrobiology Biogeocatalysis Research Center, Bozeman, M T 59717, USA

ABSTRACT 1. INTRODUCTION 2. [FeFe]-HYDROGENASE STRUCTURE 2.1. [FeFe]-Hydrogenase Overall Structure 2.2. [FeFe]-Hydrogenase Active Site Η-Cluster Structure 2.2.1. Biologically Unique Ligands 2.2.2. Ligand Exchangeable Site 2.2.3. Carbon Monoxide Inhibition 2.2.4. Non-Protein Dithiolate Composition 3. [FeFe]-HYDROGENASE SPECTROSCOPIC STUDIES 3.1. Electron Paramagnetic Resonance Spectroscopy and Mössbauer Spectroscopy 3.2. Infrared Spectroscopy 4. Η-CLUSTER MODEL COMPLEXES 4.1. 2Fe Subsite Model Synthesis 4.2. Advanced 2Fe Subsite and Η-Cluster Model Synthesis 5. Η-CLUSTER BIOSYNTHESIS 5.1. Relevance to Prebiotic Chemistry 5.2. Identification of Genes 5.3. Gene Clusters and Operons Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00179

180 180 181 181 185 187 188 190 191 192 192 193 195 195 198 199 199 200 201

180

PETERS

5.4.

Radical-S-Adenosylmethionine Chemistry and Potential Biological/Biochemical Sources 6. F U T U R E DIRECTIONS ACKNOWLEDGMENTS ABBREVIATIONS A N D DEFINITIONS REFERENCES

202 206 207 208 208

ABSTRACT: The [FeFe]-hydrogenases, although share common features when compared to other metal containing hydrogenases, clearly have independent evolutionary origins. Examples of [FeFe]-hydrogenases have been characterized in detail by biochemical and spectroscopic approaches and the high resolution structures of two examples have been determined. The active site Η-cluster is a complex bridged metal assembly in which a [4Fe-4S] cubane is bridged to a 2Fe subcluster with unique non-protein ligands including carbon monoxide, cyanide, and a five carbon dithiolate. Carbon monoxide and cyanide ligands as a component of a native active metal center is a property unique to the metal containing hydrogenases and there has been considerable attention to the characterization of the Η-cluster at the level of electronic structure and mechanism as well as to defining the biological means to synthesize such a unique metal cluster. The chapter describes the structural architecture of [FeFe]-hydrogenases and key spectroscopic observations that have afforded the field with a fundamental basis for understanding the relationship between structure and reactivity of the Η-cluster. In addition, the results and ideas concerning the topic of Η-cluster biosynthesis as an emerging and fascinating area of research, effectively reinforcing the potential linkage between ironsulfur biochemistry to the role of iron-sulfur minerals in prebiotic chemistry and the origin of life. KEYWORDS: hydrogenase · hydrogen evolution · hydrogen oxidation · iron-sulfur enzymes · iron-sulfur cluster biosynthesis · nitrogenase · nitrogen fixation · non-protein ligands · prebiotic chemistry · proton reduction

1.

INTRODUCTION

The presence of both carbon monoxide and cyanide as ligands to Fe atoms at the active site of the [NiFe]- and [FeFe]-hydrogenases is still a surprise to many even after more than ten years since the first definite evidence was presented on their existence [1]. To date, no enzymes other than the aforementioned hydrogenases have been shown to possess both ligands to Fe in the active form. The three classes of hydrogenases, described in this chapter [FeFe] (previously termed Fe-only hydrogenases), the preceding chapter [NiFe], and the next chapter [Fe], share in common carbon monoxide ligation to Fe atoms and also have common structural features and reactivity [2-28]. Despite this, however, it seems clear that these three classes of enzymes have separate evolutionary origins, and common structural and functional features have arisen by divergent evolution [29-33]. These Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

181

inferences can be attributed to the lack of primary sequence similarity between the three classes and distinct differences in the occurrence of these enzymes in various microorganisms. The aforementioned hydrogenases occur only in microorganisms where they function in microbial metabolism to either couple hydrogen oxidation to energy-yielding processes or to dispose of excess electrons that accumulate during anaerobic metabolism [2,3]. Amongst microorganisms, these hydrogenases have been found to occur in bacteria, archaea, and lower eukaryotes including protists and algae. Analysis of available genomes indicates that the [NiFe]-hydrogenases are widely spread in bacteria and archaea and are prevalent in cyanobacteria but have yet to be observed to occur in any eukaryotes [29-31,33]. In contrast, [FeFe]-hydrogenases are widely distributed among anaerobic and facultative anaerobic bacteria, some protists and algae, but do not occur in the genomes of cyanobacteria or archaea [30-33]. A third class of hydrogenase exclusive to methanogens, termed [Fe]-hydrogenases, occurs with an iron carbonyl in association with an organic cofactor [5,28,34] (Chapter 7). Regardless of evolutionary origin, it is rational to think that one or more of the [Fe]-, [FeFe]-, and/or [NiFe]hydrogenases have an ancient earth origin given the potential roles for iron-sulfur minerals and derivatives in prebiotic chemistry [35-43] and the presumed importance of reversible hydrogen oxidation chemistry in early energetics [44]. Further examination of the phylogeny of organisms in which these enzymes occur will provide insights into the origins of these enzymes. The unique structural character of the carbon monoxide in all three classes [1,12,14,21,45-59] and cyanide in the [FeFe]- [12,21,48,51,52,57] and [NiFe]-hydrogenases [1,12,14„45-51,53,54,56,58,59] distinguishes these enzyme active sites from other organometallic cofactors found in nature [60] and makes these enzymes of significant interest to basic scientists from evolutionary biologists to physical chemists. In addition, the relevance to prebiotic chemistry [60-63] coupled with the potential for the enzymes as Noble metal-free solutions to alternative and renewable energy [64-74] may suggest that the studies on the structure, function, biosynthesis, and origin of hydrogenases may hold both the keys to the origin of life as well as the future of life on earth.

2. 2.1.

[FeFe]-HYDROGENASE STRUCTURE [FeFe]-Hydrogenase Overall Structure

In the late 1990s, the structure of [FeFe]-hydrogenases from two different microbial sources (Clostridium pasteurianum and Desulfovibrio desulfuricans) Met. Ions Life Sei. 2009, 6, 179-218

182

PETERS

were revealed [11,18]. The [FeFe]-hydrogenase from Clostridium pasteurianum (Cpl) determined by our group is a ~ 60 kD monomeric enzyme that is localized in the cytoplasm and functions in the coupling of proton reduction to the oxidation of reduced electron carriers that accumulate during carbohydrate fermentation [11]. The Desulfovibrio desulfuricans [FeFe]-hydrogenase (DdH) determined by the Fontecilla-Camps group is a dimeric enzyme, with a large and a small subunit, is localized in the periplasm and consequentially is presumably involved physiologically in hydrogen oxidation [18]. In addition to the differences in the quaternary structure and cellular localization of the C. pasteurianum and D. desufuricans [FeFe]-hydrogenases, the enzymes also differ in overall size and in their respective complement of accessory FeS clusters. The C. pasteurianum [FeFe]-hydrogenase exists with an active site Η-cluster and four additional FeS clusters including three [4Fe-4S] clusters and a [2Fe-2S] cluster (Figure 1A) while the D. desufuricans enzyme exists with an accessory cluster composition of just two [4Fe-4S] clusters in addition to the active site Η-cluster (Figure 1C). In general, all [FeFe]-hydrogenases that have either been characterized biochemically [2,75-91] or simply implicated by searches of deduced amino acid sequences in the genomes of various organisms [30,32,33] share a common architecture with most apparent differences on the complement of various FeS clusters that link the active site Η-cluster for reversible hydrogen oxidation to various external electron donors and acceptors. These differences presumably affect the differences in the physiology of organisms that harbor these enzymes by accommodating specific and distinct external electron donors and/or acceptors. The C. pasteurianum monomeric [FeFe]-hydrogenase exists with an overall mushroom shape with a large domain represented by the mushroom cap in which the active site Η-cluster resides (Figure 1A and IB). In addition to the larger Η-cluster domain, three other smaller domains can be assigned according to the structural similarity to known ferredoxin proteins [92,93]. The domain proximal to the Η-cluster domain contains two [4Fe-4S] clusters and is similar in overall structure to the numerous two [4Fe-4S] cluster-containing ferredoxins for which the structures of several are known (Figure 1A, green domain) [93-98]. This domain is common to many [FeFe]-hydrogenases [18,99] and is also found as a domain in additional redox proteins and enzymes involved for example in respiration and photosynthesis [100,101]. The ferredoxins and the structurally analogous domains are easily distinguished in searches of deduced amino acid sequences in genomes because they possess a well-defined and conserved arrangement of cysteine ligands (eight conserved cysteine residues) that function as covalent thiolate ligands to the [4Fe-4S] clusters. The far N-terminal domain contains a [2Fe-2S] Met. Ions Life Sei. 2009, 6, 179-218

C A R B O N M O N O X I D E A N D C Y A N I D E IN [ F e F e ] - H Y D R O G E N A S E S

Λ

183

c

Figure 1. A. Cartoon depiction of the X-ray crystal structure of Cpl [11] showing the overall fold and division into 4 separate structural domains based on FeS cluster content and structural similarity to known ferredoxin proteins. The active site Hcluster domain is shown in blue, the 2 [4Fe-4S] cluster ferredoxin-like domain is shown in green, the N-terminal [2Fe-2S] cluster ferredoxin-like domain is shown in light magenta, the single [4Fe-4S] domain is shown in purple, and the C-terminal region is shown in red. The FeS clusters and CO and CN~ ligands are shown as space filling models (Fe: dark red, S: orange, O: red, N: blue). B. Zoom view of the active site domain showing the Η-cluster cleft formed by the pseudosymmetric symmetry of the two twisted ß-sheets on either side. The C-terminus is perpendicular to the ßsheets and wraps around the cleft as occupied by the Η-cluster. C. Cartoon depiction of the X-ray crystal structure of D d H [18] where equivalent structural regions to Cpl are shown in the same colors. D. Cartoon depiction of the homology structure of C. reinhardtii to Cpl as determined by the homology server Phyre [225,226]. The green algae C. remhardtü lacks additional accessory cluster domains and is represented by the homologous active site domain only.

cluster and is also structurally analogous to a class of what is termed planttype ferredoxins for which structures have been described (Figure 1A, magenta domain) [92,102-109]. Again, a characteristic motif of cysteine residues facilitates the identification of structurally analogous domains or ferredoxins of this type in deduced amino acid sequences. An additional small domain containing a single [4Fe-4S] cluster separates the two ferredoxin-like domains (Figure 1A, purple domain). This [4Fe-4S] cluster possesses a unique set of coordinating ligands with the combination of three cysteine thiolates and secondary amine of a histidine imidazole side chain. This combination of ligands in an accessory cluster is a feature only observed in this [FeFe]-hydrogenase and in the characterized structures of Met. Ions Life Sei. 2009, 6, 179-218

184

PETERS

the [NiFe]-hydrogenases from sulfate-reducing bacteria [4,7,8,16,110] but to my knowledge not observed elsewhere in biology. The specific biochemical and physiological role of this unique cluster is not known, but it is possible that the ligand arrangement would contribute to defining the oxidation-reduction potential of the cluster and the presence of the histidine coordination may contribute to proton-coupled electron transfer. In comparison, the D. desulfuricans structure lacks the histidine-coordinating cluster, as well as the [2Fe-2S] clusters and their associated domains (Figure 1C) [18]. The complement of only the additional two [4Fe-4S] cluster containing ferredoxin-like domains (Figure 1, green domains) is observed in many [FeFe]-hydrogenases [18,99] and phylogenetic analysis [30,32,33] reveals this feature in the most deeply rooted organisms suggesting that this arrangement of domains may be ancestral. The aforementioned [FeFe]-hydrogenase architectures (those with either four or two accessory clusters) represent the majority of known [FeFe]hydrogenases. However, as mentioned above, the specific complement of accessory clusters and their associated domains is a distinguishing property and a notable additional variation on this theme observed in [FeFe]-hydrogenases of thermophiles [111], hyperthermophiles [112], and eukaryotes including a group that include the [FeFe]-hydrogenases observed in algae [113-119], which lack accessory clusters entirely (Figure ID) and other more complex versions occur in the hydrogenosomes of the ciliate Nyctotherus ovalis in which the [FeFe]-hydrogenase functionality is fused to domains that have primary sequence similarities to electron transport chain components [120]. This arrangement implies that the latter [FeFe]hydrogenase can directly reoxidize N A D H and has been suggested to support the idea that the hydrogenosome evolved from ciliate mitochondrion [121], Given the relative position of the accessory clusters of the C. pasteurianum [FeFe]-hydrogenase, a single pathway for the transfer of electrons to and from the active site Η-cluster and external donors and acceptors cannot be rationalized [11]. The histidine-ligated [4Fe-4S] cluster and the [2Fe-2S] cluster are located close to the protein surface at the termini of what could be a branched pathway providing the means to interact with electron transfer partners with different chemical characters and may serve as some sort of metabolic switch for the enzymes between hydrogen production and hydrogen oxidation. An assessment of the relative distribution of charge on the surface of the C. pasteurianum [FeFe]-hydrogenase indicates that the protein surface surrounding is distinctly positively charged and in contrast the region surrounding the [2Fe-2S] cluster is distinctly negatively charged which may support interactions with multiple external electron transfer partners. Alternatively, the histidine-coordinated cluster may have a different role that has yet to be determined. Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

185

The active site domain is the largest domain in [FeFe]-hydrogenases ( ~ 3 0 0 amino acid residues) and exists as a pseudosymmetric structure in which two twisted β-sheet structures come together to form a cleft in which the Η-cluster is located (Figure 1A and IB). Covalent coordination of the 6Fe Η-cluster to this domain is provided by four cysteine thiolates, two each provided from each pseudosymmetric half. The residues involved in covalent coordination and the associated residues within the environment of the Η-cluster are highly conserved [30] providing the basis for a primary sequence signature for the identification of [FeFe]-hydrogenases from genome sequences (Figure 2). These motifs also provide the basis of differentiating between bona fide [FeFe]-hydrogenases and related sequences including eukaryotic narF gene sequences [122-124]. For the C. pasteurianum, the cleft where the H-cluster is located is covered by the C-terminal region wrapping around the cleft perpendicular to the approximate axis of the pseudosymmetric structure (Figure 1A and IB). In the dimeric D. desulfuricans structure, the region analogous to the C-terminus of C. pasteurianum [FeFe]-hydrogenase is provided by the small subunit (Figure 1C). The small subunit (ß-subunit) also encodes the determinants that direct the enzyme to the twin arginine transporter for localization of the enzyme in the periplasm.

2.2.

[FeFe]-Hydrogenase Active Site Η-Cluster Structure

Much of the attention in the structures of [FeFe]-hydrogenases can be correlated to the interest in the structure of the active site H-cluster. The surprises realized in the early structural characterization of the [NiFe]-hydrogenase active sites [4,7,8,16] coupled with the inability to come to a structural consensus based on spectroscopic characterization placed an imperative on defining the structural details of the H-cluster in the late 1990s. In 1996, in a seminal review on iron-sulfur enzymes [60], authored by Richard Holm of Harvard and Edward Solomon together with his postdoctoral associate at the time, Pierre Kennepohl, of Stanford University, highlighted this imperative and described the H-cluster as the most conspicuously undescribed complex iron-sulfur cluster of the time. With the presence of the carbon monoxide and cyanide ligands observed in the [NiFe]-hydrogenase active site [1,12,14,45-50], it was envisioned that this complement of ligands could also be present in the [FeFe]-hydrogenase, but this had not been definitely determined until after the first structures were revealed. The Cpl structure, which was described first [11], was originally refined to 1.8 A resolution and afforded sufficient information to make tentative assignments to most of the features of the cluster. The H-cluster in the Cpl Met. Ions Life Sei. 2009, 6, 179-218

186

PETERS

CpiMKT I I [NSVQfNTDEDTT I LKFARDNN lOISALCFLNNCNNBI NKCE Ε CT V EV E GT GL V Τ SO & M8ALVLK • *- -. - . . . . . . .. . * . . . , - - - PCAAVS1 R···· * — -»..·. » , *3CRARQ 23 DdH - - - . · - . -> * - - - -- -- KClQCO Tma C Τ L KP Y EG MKVKTNTPE t YEMRRN I LEI. 1 LATHNRDC1 Γ CORN GSCKLQK 109 - w w v Cpf SKPFE PKOKTEYVDER. · • SKS Ε. Τ V DR Τ Κ C L L C SR CV NACOKN Τ Ε Τ YAMK FL Ν ΚΝ G 173 Crl VAPRAPLAASTVRVAL AT L E APARR L GNVAC ·νΑΑΑΡAAEAP L Ξ HVQQAL - - - 72 DdH TCSQYCPTAAI FGEWG- ----- - EPKSI PHIEACINCGQCLΤHCPENA IYEAQSWV- - - - 66 Troa YAEDFGl BK I R F U A L KK £HVR 0 Ε SAPVVRDT S KC I LCGBCVRVCΕΕ IOGVGVI CFAKRGF 169 "λ/1 ^ -v/VWvCplKTl IQAEBEKCFODTMCl LCGQC I AC PVAAL S t=KSHMORV KNA L NAP Ε KHV I VAMAPS V 233 CrlA Ε: AK Ρ KODP χ, . . TRKHVCVQVAPAV 95 DdH • PEVEKKLKD G K V K C 1 A M F A P A V 110 TmaESVVTTAIOTPLIETECVLCGQCVAYCPIGALSIRNDIDKL r Ε A L Ε•SÜKIVIGM-APAV 228 289 CpE RAS (GEI FNM :F GVI1VTGKI 1 I ALROLGFDKI FBI Η F GADMT 1MEEATELVQRI EN Crl RVA I AETLGLAPGATTPKaLAEGLRRLGFOEVFDTLFGADLT 1 MEEGSEL LHRLTEH . L- / 155 EJdll RYALGD-- FGMP VGSV TT GKM L AALQK L GFAH CWDT : FTAQVT 1WEEG5Ε FVERLTK - - - - 166 TnwRAAIQE FGIO DVAMAEKLV5FLΚ Γ i GFDKVFDV5FGADLVAY£EARΕFYERLKK 26+ ^A/VW-s/VVi» * \ΑΑΑΛ —V Cpl • NGPFPMFTSCCP CWV ROAE rs'YYP Ε LLNNLSSAKSPÜQE FGTASKTYYPS I SGL DP Κ 345 Crl HPHSDEPLPMFTSCCPGW1 AtfLEKSYPDL I Ρ YV S S CKSPQMM LAAMV KS Y LA Ε Κ KG I ΑΡΚ 215 E3JH - K5DMP LPQ F TSCCPOTlfQK YA Ε TYYPELLPHF5TCK5P I GMN GA LAKT YGAER'.'K Y DP Κ 224 Tma .GERLPOFTSCCPAWVKHAE! Τ YPQYLQN L SSVKSPQQA LGT V I KK I Y ARK L βν Ρ Ε Ε 340 Cpl .MVFTVTVMPCTSKKFEADRPOMEKDG I. R D I DAV I TTRELÄKM I KL1AK I PFAHL EDSE ι rif-Fe^ ^P Ά2 Ξ—

"Fe.-^ppt.^ 0

"12-

S-Fe ,Fe OC / OC

F^ \ CO CO

\> S "« TΛs' \ Ξ—

Figure 5. Overview of the general development and progression of synthetic model compounds to the 2Fe subcluster and Η-cluster of [FeFe]-hydrogenases. A. Diironhexacarbonyldisulfide. B. Insertion of the P D T ligand and single cyanide substitution at each Fe atom of the hexacarbonyl compound. C. Further variation on the dithiolate ligand including insertion of nitrogen and oxygen atoms at the bridge head position. D. Development of advanced models to improve similarity and functionality of the enzyme's 2Fe subcluster including (from left to right): mixed valance F e : F e : l species with a bridging CO ligand, bridging hydride, and terminal hydride species. E. Synthesis of the basic 6Fe Η-cluster framework.

( P D T ) ( F i g u r e 5B) [189], T h e resulting c o m p o u n d [ F e 2 ( S C H 2 C H 2 C H 2 S ) ( C O ) 6 ] is o f t e n c o n s i d e r e d t o be t h e p a r e n t m o d e l f o r t h e e n s u i n g 2 F e s u b cluster m o d e l s . I n t e r e s t i n g l y , this line of synthetic c h e m i s t r y evolved a d e c a d e earlier f r o m r e p o r t of t h e first [ F e F e ] - h y d r o g e n a s e crystal s t r u c t u r e s a n d i d e n t i f i c a t i o n of t h e C O a n d C N ~ l i g a n d s at t h e 2 F e s u b c l u s t e r site b y I R Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

197

studies as discussed above. Shortly after, stoichiometric cyanide substitution at each Fe center on the [Fe 2 (SCH 2 CH 2 CH 2 S)(CO) 6 ] compound was achieved and synthesis of the [Fe 2 (SCH 2 CH 2 CH 2 S)(CO) 4 (CN) 2 ] 2 " compound was reported by three different groups (Figure 5B) [151,153,158]. This complex already exhibited key features of the 2Fe subcluster site such has its edge shared bipyramidal geomeotry, Fe-Fe distance (2.6 A), and placement of an open coordination site above the distal Fe atom. Most similar to the Hred state of the D d H enzyme, the model compound exists with nonparamagnetic Fe 1 centers. The next feature addressed by development of model complexes to the 2Fe subcluster was introduction of variations on the dithiolate ligand. Because the reported X-ray crystal structures could not distinguish between isoelectronic groups at the middle atom of the bridging ligand, it is possible that either -CH 2 -, -NH-, or -O- could exist at that position and synthesis of two other possible bridging ligands, (SCH 2 NRCH 2 S) R = H, Me) [159] and (SCH 2 OCH 2 S) [174], was reported (Figure 5C). Looking at the case of the amine dithiolate bridge, from a mechanistic perspective the nitrogen was described as being an attractive possibility at that position as it could act as base to extract and transfer protons [131,133,190-192] at the ligandexchangeable site above the distal Fe atom during hydrogen catalysis. There is however, no crystallographic evidence to date that supports this hypothesis and it is important that model complex studies keep in mind all possibilities for the bridging ligand (PDT, D T N , DTO). Given the rapid developments of model compounds to the 2Fe subcluster site, these early studies still failed to address a stable mixed valence state of the Hox state and the bridging CO ligand present in this state due to low stability and reactivity. These two characteristics are thought to be important to H 2 catalysis, especially as in the H 2 catalytic cycle the Η-cluster cycles between just two states ( H o x and Hred). Picket and coworkers were the first able to spectrochemically characterize a mixed valence Fe'Fe 1 1 species with a bridging CO ligand, however due to the low stability, isolation of the compound remained elusive [169]. This work prompted the development of diferrous compounds that resemble the Hox state where an isocyanide ligand is bridging between the two Fe atoms. The diferrous compound [Fe 2 SCH 2 CH 2 CH 2 S(CNMe) 7 ][PF 6 ] 2 closely mimics the geometry of the Hox active state and includes the presence of a bridging ligand, however, the stability of the compound can be greatly attributed to the higher oxidation state of the diiron dithiolate complex [170]. In 2007, Liu and Darensbourg reported the first isolation of a stable mixed valence Fe'Fe 1 1 species with a bridging CO ligand [183] and shortly after, Justice, Rauchfuss and Wilson [182] reported another similar species (Figure 5D). EPR and F T I R on the species revealed a similar rhombic EPR spectra and F T I R spectra [182,183] to the Hox D d H state and the reported Met. Ions Life Sei. 2009, 6, 179-218

198

PETERS

models were the first to exhibit a Fe'Fe 11 paramagnetic couple with a bridging CO ligand. Significantly, these models were a breakthrough in modeling the active form of the 2Fe subcluster, however, both included bulky strong donor ligands, IMES and dppv, in place of the CN~ ligand at one of the Fe atoms. These ligands were presumably used to lock the unique geometry in place with the bridging CO ligand and mixed valence Fe'/Fe 11 state and may not be effective mimics of the actual biological Η-cluster ligands. In addition, at the proximal Fe, in both cases, a PMe 3 ligand was used in place of the CN~ ligand such that these models capture the key features of the 2Fe subcluster in the active Hox state but they have not been able to include simple electron donating CN~ ligands and ongoing work is directed toward capture of simple Fe'/Fe 11 models containing CN~ ligands, a bridging CO ligand, an unsaturated Fe atom.

4.2.

Advanced 2Fe Subsite and Η-Cluster Model Synthesis

With respect to the most recent work in this area, much emphasis is placed on modeling the hydrogenase functionality emphasizing hydride complexes and synthesis of the complete 6Fe Η-cluster framework. Darensbourg et al. have reported the synthesis of 2Fe compounds containing a bridging hydride ([Fe 2 (SCH 2 CH 2 CH 2 S)^-H)(CO) 4 (PMe 3 ) 2 ] + ) (Figure 5D) [165,166], Significantly, this compound begins to model the functionality of the 2Fe subcluster as it has the ability to bind and activate H 2 at an Fe 11 site. Although the biochemical and structural characterization have not implicated a role for a bridging hydride in the enzymes, this is the first compound to show H 2 binding at an unsaturated Fe 11 center. Terminal hydride compounds have also been synthesized and the synthesis of a 2FeT species having a bridging carbon monoxide ligand and a terminal hydride ligand has been reported (Figure 5D) [184]. In the Hox state, the bridging carbon monoxide ligand may prevent protonation of the FeFe bond such that proton binding at the distal Fe would be more favorable [193]. Development of hydride binding species is important in unraveling the mysteries still present at the 2Fe subcluster and mechanistic features of H 2 catalysis. In 2005 Picket et al. reported the synthesis of an entire 6Fe H-cluster framework with a 2Fe subcluster model bridged to a [4Fe-4S] subcluster using a tridentate thiolate ligand (Figure 5E) [194]. On a similar time frame it was realized through D F T calculations that there is a signficant impact of the presence of the [4Fe-4S] cluster on the overall electronic structure of the 2Fe subcluster and likely catalysis [150,195]. This highlights the fact that the groundwork for D F T studies has been well set in place from the development of synthetic models to the 2Fe subcluster and synthetic model studies have opened opportunities for D F T studies to explore H 2 oxidation and H + Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

199

reduction at the 2Fe subcluster and the complete 6Fe cluster. The application of D F T to modeling the Η-cluster during H 2 catalysis will be discussed in much greater detail in Chapter 12.

5. 5.1.

Η-CLUSTER BIOSYNTHESIS Relevance to Prebiotic Chemistry

Η-cluster biosynthesis is an emerging area of research that is key to generating effective metabolic engineering of hydrogen-producing microorganisms and heterologous expression and mass production of hydrogenase enzymes for use in biomimetic and biohybrid materials in biotechnology. But as in the case of the synthesis of small molecule mimetics described in Section 4, there are links to these enzyme active sites to prebiotic chemistry [61-63] and understanding the relationship between biosynthesis, structure, and reactivity and relating these findings to the ongoing work exploring the structure and reactivity of iron-sulfur minerals and their derivatives may hold keys to life's origins. Of course, the suggestion that iron-sulfur minerals may have had a role in prebiotic chemistry and the origin of life are not new and the potential for this has been expounded upon in recent works of Wächtershäuser and others [41]. The complex iron-sulfur cluster at the active sites of nitrogenases, hydrogenases, carbon monoxide dehydrogenases, and acetyl-CoA synthase reactions have been suggested to be among the first cofactors since these enzymes catalyze interconversions of small molecules that could have been important in increasing the pool of reactants to participate in basic condensation reactions important for forming the building blocks of life. These enzymes, all having deeply rooted lineages in a "metabolism first" model for the origin of life, may represent highly evolved derivatives of iron-sulfur minerals which supported the reactivity to make the transition from the non-living to the living world. These ideas are dependent on a mechanism to generate a variety or combinatorial allotment of modified iron-sulfur mineral catalysts tuned to specific reactions equivalent to enzymes in a metabolic pathway. We anticipate that reduced iron-sulfur minerals would have been plentiful on the early earth and the means to generate a variety of structures of modified iron-sulfur minerals exists today in environments that many feel are closest to mimics of early earth's environments. Examples of these environments are the hydrothermal vents in the deep oceans or the effluent streams leading from thermal springs like those found in Yellowstone National Park. These environments represent highly mineral-rich gradients and those rich in iron and sulfur are not difficult to find today. With large Met. Ions Life Sei. 2009, 6, 179-218

200

PETERS

gradients in temperature, pH, gas solubility, mineral solubility, etc. it is not difficult to imagine the introduction of carbon monoxide or heteroatoms such as Mo or Ni to tune various types of reactivity. A complete understanding of the structure and reactivity of these complex iron-sulfur enzymes together with a full understanding of the chemical reactions that bring about the biosynthesis of the prosthetic groups will provide the basis for a better understanding of the potential for generating modified iron-sulfur minerals that resemble these clusters on the early earth.

5.2.

Identification of Genes

Until only very recently essentially nothing was known about [FeFe]hydrogenase Η-cluster biosynthesis. One of the reasons for this might be that there has been limited success in terms of genetically manipulating organisms that harbor [FeFe]-hydrogenases. For the nitrogenase system, genetic studies were key in the identification of genes other than the nitrogenase structural genes nifH, nifD, and nifK required for the process of nitrogen fixation (see reviews [196,197]). These studies provided significant insights into understanding the regulation of nitrogen fixation, the linkages between nitrogen fixation and energy metabolism, and iron-sulfur cluster biosynthesis. Gene products were initially implicated in being involved in iron-sulfur cluster biosynthesis through the analysis of nitrogenases produced in various genetic backgrounds in which nitrogen fixation specific gene products had either insertion or deletion mutations. In these genetic backgrounds it was found that a number of mutants produced nitrogenases that had a complement of iron-sulfur clusters that differed from the native enzyme. From these studies a number of gene products were specifically implicated in being involved in nitrogenase iron-molybdenum cofactor biosynthesis. This approach and the analysis of nitrogenase mutants later proved to be the driving force in some of the first breakthrough studies that provided insights into general mechanisms of iron-sulfur cluster biosynthesis. Some interesting insights have been made into the active site cluster biosynthesis of [NiFe]-hydrogenases including the source of cyanide ligands (discussed in detail in Chapter 5) [198-202]. For [FeFe]-hydrogenase Η-cluster biosynthesis, it was not until 2004 when an approach similar to that described for the identification of nitrogen fixation gene products involved in cluster biosynthesis was applied [203]. In this work, several strains of Chlamydomonas reinhardtii produced by transposon mutagenesis that were unable to produce hydrogen were identified. These strains were analyzed with regard to what specific genes were interrupted to eliminate hydrogen production. Several gene disruptions were identified that were not Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

201

localized to either of the two [FeFe]-hydrogenase structural genes present in C. reinhardtii including within a gene cluster encoding two novel members of the S-adenosylmethionine dependent (radical-SAM) enzymes. These gene products termed hydEF and hydG were not found to be clustered in the C. reinhardtii genome with either of the [FeFe]-hydrogenase structural genes hydA, however, it was found that the expression of the [FeFe]-hydrogenase in a background of co-expressed HydEF and HydG heterologously in E. coli resulted in the synthesis of active [FeFe]-hydrogenase [203]. Although only a fraction of the anticipated activity ( ~ 2 0 % ) was obtained, expression of the [FeFe]-hydrogenases in E. coli without co-expression of the identified accessory enzymes does not yield any active enzyme. The results are highly significant and the ability to synthesize an active enzyme having such a unique active site cluster will be tremendously valuable in defining the biochemistry of Η-cluster biosynthesis. For nitrogenase, the analogous feat of expression of active enzyme in an E. coli host has not been achieved.

5.3.

Gene Clusters and Operons

These studies provide a framework for thinking about how the Η-cluster is synthesized. A closer look at the deduced amino acid sequences of hydEF and hydG and their homologs in other organisms provides additional insights. All other organisms that are known to harbor active [FeFe]hydrogenases have homologs of the C. reinhardtii genes, however, in other organisms hydE and hydF occur as separate genes. In addition, although in many cases the genes hydE, hydF, and hydG and the structural gene(s), do not occur in gene clusters and are distributed throughout organisms essentially indiscriminately, there are several examples in which the genes do occur in clusters (Figure 6). In a number of cases these clusters contain genes, in addition to hydE, hydF, and hydG, and the structural gene(s) however, these are the only genes yet to be determined to be common in all organisms. Interestingly the aal gene product ammonia aspartate lyase is present in a number of gene clusters but is not universal in its occurrence with hydrogen genes. The occurrence of ammonia aspartate lyase may implicate a link between biosynthesis and central carbohydrate or amino acid metabolism. Thus it is interesting to contemplate about [FeFe]-hydrogenase in the context of only these products of the three genes hydE, hydF, and hydG. One thing that is clear in comparing the sequences and properties of genes involved in the [FeFe]- and [NiFe]-hydrogenases (Chapter 5) is that the biochemistry of cluster synthesis and hydrogenase maturation in the two systems is very different. Although there is good evidence that the source of cyanide ligands in the [NiFe]-hydrogenases is carbamoyl phosphate Met. Ions Life Sei. 2009, 6, 179-218

202

PETERS

A metalliredigens Β. thetaiotaomicron D. vulgaris F nodosum

fransposas^>

Ξ, oneidensis

|

S.

metanesiensis

HydA HydA

HyäA

/ .)

HydG

HydE )•

>• >j

Η yd Ε

HydE

|

HydG

• ^ranapo3ase>

HydG

>

Figure 6. Gene clusters of various [FeFe]-hydrogenases showing the observed genomic arrangement of [FeFe]-hydrogenase maturation gene products HydE, HydF, and HydG along with the [FeFe]-hydrogenase structure encoding gene product HydA. Also present in some operons is AspA ammonia aspartate lyase encoded by the aal gene. [198,199], this similarity is not found to occur with [FeFe]-hydrogenase genes and thus, cyanide is most likely to be derived from another source. Some may consider the enzyme carbon monoxide dehydrogenase which catalyzes reversible carbon monoxide oxidation to be a potential mechanism for deriving carbon monoxide ligands in both [FeFe]- and [NiFe]-hydrogenases, however, this possibility can be dismissed. The processes of carbon monoxide oxidation and anaerobic carbon dioxide fixation occur much less frequently in biology than hydrogen metabolism and thus, many if not most hydrogenases occur in organisms that do not possess a carbon monoxide dehydrogenase. In addition, the only cases in which the genes encoding carbon monoxide dehydrogenase occur with hydrogenases are in cases where carbon monoxide oxidation is metabolically coupled to hydrogen metabolism as in the case of the purple non-sulfur bacteria [204], making the hypothesis of carbon monoxide derivation for active site formation necessitated by an active hydrogenase untenable.

5.4.

Radical-S-Adenosylmethionine Chemistry and Potential Biological/Biochemical Sources

We have thought long and hard about the potential of only three gene products in Η-cluster biosynthesis. This would be in stark contrast to the nitrogenase F e M o cofactor biosynthesis in which approximately ten gene products can be genetically and biochemically linked to the process [196,197]. However, making several assumptions based on the observations associated with F e M o cofactor biosynthesis and taking into account the clear implications concerning the potential function of the H-cluster Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

203

accessory gene a n n o t a t i o n a n d preliminary biochemical c h a r a c t e r i z a t i o n has led t o the ability to suggest possible r a t i o n a l s f o r Η-cluster f o r m a t i o n . T h e o b s e r v a t i o n t h a t t h e F e S cluster c o n t a i n i n g proteins H y d E a n d H y d G are clearly r a d i c a l - S A M enzymes as by t h e observance t h a t they b o t h have the C - X , - C - X 2 - C r a d i c a l - S A M signature m o t i f (Figure 7) [203,205] a n d t h a t

HjilZ HydC BioB

1ψ.\

PEL A E LAM Hill

130 t • ·» .ι >J .M KVML R L 1 Ε R I V L PL I - ·-, . q V Q V 5 1 LL IK : Jr. ι r N H ι* Τ Λ Τ ; . · • - R LL 1. a R I' D ft V L R I V M Ρ L

•

KR ν Ν Ν Γ ED Τ R R R Μ v ·•

1*0 1 Βκ.1 a8κ aL 3 1V

A NH

Β

ΗΝ Η

IS I. , I

U C t Κ [ HC Kl ( J 5 Η II : V ;JI(1RPV Ι Έ D E T ι HRft SGt^J Κ Η Κ MC H l

no .!

I

I .... I ·

Ei I. Ε Μ Κ D Κ

l·

K i t Υ Η Κ Η 1 . I. Κ C

1. Κ Γ> S V I I'l^VBIIVL I I. Q D J T Η Κ Η I.

ANH

ΗΝ Biotin Synthase (BioB)

Η

int

..I

ER Κ t i l> D D I I Ε L KU R H ^ . - L - i f l D E L H E E V K L I) Χ C Β L Μ V Ε Κ V L () Ε . \ Q L Ρ V C T P - - l l t V e ® L t Κ * 1 Μ V r ΚΚ } L Q O I I HE RKK L s g E O V K H ET 1

— f

S—S

Η

r

Lipayl Sjmllia«(UfA) Η Η E-Gly· β E-Giy ~~"·-5> Pyruvate Ferriöteξ Lyase Activating EnEjme

I w Η Η OO' H3N Η

Ν OH

TMH +• Adoraet ^ ^ OH

Figure 7. A. Portion of sequence alignment of radical-SAM [FeFe]-hydrogenase proteins HydE and HydG relative to the known functional radical-SAM proteins biotin synthase (BioB), lipoate synthase (LipA), pyruvate formate lyase activating enzyme (PFL-AE), lysine 2,3-aminomutase (LAM), and thiazole synthase (ThiH). The signature radical-SAM motif C-X 3 -C-X 2 -C is highlighted red and conserved and similar residues are highlighted light blue and green, respectively. B. Reactions catalyzed by radical-SAM proteins. BioB and and LipA catalyze hydrogen atom abstraction/sulfur insertion reactions and PFL-AE, LAM, and ThiH catalyze various reactions via the formation of an amino acid radical. Met. Ions Life Sei. 2009, 6, 179-218

204

PETERS

HydF protein is a GTPase [203,206,207] gives a solid foundation to think about Η-cluster biosynthesis. In fact, radical-SAM chemistry alone clues for potential mechanism of Η-cluster biosynthesis. In 2007, we published an article and detailed a potential hypothetical mechanism for Η-cluster biosynthesis based on what is known concerning FeMo cofactor biosynthesis together with observations and considerations of the Η-cluster structure and the involvement of radical-SAM based biochemistry [208]. We had made several assumptions to frame our hypothetical mechanism. One of the most important assumptions was that the maturation machinery was directed only at the 2Fe subcluster of the Η-cluster. We made this assumption for several reasons, including (1) the limited number of gene products implicated in Η-cluster biosynthesis [203], (2) the ability of organisms processing [FeFe]-hydrogenase to synthesize [4Fe-4S] clusters [209], and (3) the observation that the only covalent linkage between the two subclusters is provided by the protein [11,18,21]. The other important assumption we made was that the synthesis of the 2Fe subcluster does not occur on the hydrogenase and occurs on a protein scaffold in analogy to nitrogenase FeMo cofactor biosynthesis [210] and thus, we anticipate that the [FeFe]-hydrogenase produced in a background devoid of the maturation genes should be produced in a form that is capable of being activated upon the insertion of the 2Fe subcluster. With this foundation we then began to consider the involvement of radical-SAM chemistry and specifically tandem radical-SAM enzymes catalyzing sequential steps in the process which would include the synthesis and placement of the non-protein ligands and production of the 2Fe subcluster. In the context of precedented radical-SAM chemistry, one connection that appears to be intuitive is that sulfur insertion chemistry similar to that observed in BioB [211] and LipA [212] in the synthesis of biotin and lipoic acid is likely invoked in the synthesis of the dithiolate ligand (Figure 7). In these reactions it has been shown that the source of sulfur for these cofactors originates from a second conserved FeS cluster unique from the SAM-binding [4Fe-4S] cluster [213-217], For this reaction, it is not intuitive what might be the substrate for the sulfur insertion. It has been suggested that perhaps propane or dimethylamine could serve as substrates resulting in a PDT or D T N ligand, respectively, but these substrates are hard to rationalize in the context of microbial metabolism. Since only there are only three gene products and there are no clear conserved gene products that would be involved in providing the substrates for either of the radical-SAM enzymes, it may be rational to suggest that the precursors come from the central metabolism and in the case of [FeFe]-hydrogenase this would be specific to anaerobic metabolism. Although this perhaps narrows the possibilities, there are still a large number of compounds that could serve as precursors and since the composition of Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

205

the dithiolate has yet to be determined, this makes rationalizing potential substrates even more difficult. We have therefore kept our thoughts on this step of cluster biosynthesis fairly generalized with the notion that the reaction, whatever the precursor, resembles the reactivity observed for LipA in that a sulfur insertion occurs by the hydrogen atom abstraction at two positions of an organic substrate [215] and the source of the thiolate sulfurs are sulfides from an FeS cluster. In contrast to the LipA reaction in which the FeS cluster is destroyed in the process of lipoic acid synthesis, we envision a reaction in which a [2Fe-2S] cluster is converted to a dithiolatecoordinated 2Fe cluster that remains intact. If we think about this reaction as the first step of Η-cluster biosynthesis and we examine this from the perspective of the reactivity of a hypothetical [2Fe-2S] precursor, the reactive sulfides are converted to less reactive thiolates allowing us to think about a second step of the biosynthetic pathway in which the Fe ions are the focus of the reactivity. Additional ideas can be gleaned from precedented radical-SAM chemistry. One set of reactions or themes that we found particularly attractive are reactions involving amino acid radicals as intermediates. The reactions catalyzed by pyruvate formate lyase activating enzyme [218], lysine amino mutase [219], and the thiH gene product [220,221] involved in synthesis of the thiazole ring of thiamin all occur via the formation of amino acid radicals (Figure 7). The observation of multiple examples of radical-SAM enzymes that utilize amino acids as substrates and catalyze bioconversions of amino acids involving the formation of amino acid radicals makes it attractive to think of an amino acid substrate and an amino acid radical intermediate as a source of the carbon monoxide and cyanide ligands in the Η-cluster. There are some attractive features with regard to the potential utilization of an amino acid as a substrate for carbon monoxide and cyanide formation, probably the most significant being that an amino acid substrate and the decomposition of an amino acid at a iron-sulfur cluster as a means for forming these ligands represents a mechanism in which both carbon monoxide and cyanide ligands can be formed and deposited on the metal center in a concerted manner from a fixed source such that free carbon monoxide and cyanide do not freely diffuse through cells. Considering the availability of substrates that could be involved in a reaction forming carbon monoxide and cyanide in a concerted manner it is attractive to think about the decomposition of an amino acid. In our hypothesis we suggested glycine as a potential substrate for a radical-SAM enzyme yielding upon decomposition carbon monoxide, cyanide, and water, essentially the ligand set observed at the distal Fe of the 2Fe subcluster [208]. Glycine is an interesting potential substrate from the perspective of availability since glycine is required for nucleotide metabolism and thus, should be available in all organisms. A computational analysis of the Met. Ions Life Sei. 2009, 6, 179-218

PETERS

206

viability of such a reaction was analyzed using a hypothetical in-silico generated simple thiolate c o m p o u n d and yielded an overall t h e r m o dynamically favorable reaction with an activation barrier for the f o r m a t i o n of the glycine radical that binds and decomposes at the site. T h e barrier can be overcome by the activity of a radical-SAM enzyme perhaps lending credence to this overall concept. These ideas are interesting but not f o u n d e d in experimental d a t a and only provide the basis for experimental design to prove or disprove them. A n o t h e r interesting perspective of the idea of an a m i n o acid decomposition as a c o m p o n e n t of Η-cluster biosynthesis returns us to the idea of a link between iron-sulfur enzymes and prebiotic chemistry. If we consider the reverse of the reaction described above we have the condensation of c a r b o n monoxide, and water to f o r m an a m i n o acid. In essence, the form a t i o n of a building block f r o m small molecule precursors is conceptually similar to the early experiments of Miller and U r e y [222] but a d a p t e d in the spirit of the current ideas of W ä c h t e r s h ä u s e r in clearly identifying a central role for iron-sulfur c o m p o u n d s in early H a e d i a n and prebiotic reactions [41,43,223],

6.

FUTURE DIRECTIONS

The ability to examine the above ideas concerning the f o r m a t i o n of the nonprotein ligands (dithiolate, c a r b o n monoxide, and cyanide) and screen for the actual substrates for radical-SAM enzymes ( H y d E and H y d G ) is dependent on a functional assay for their activities. A recent advancement in this regard is the development of an in vitro system in which it has been shown that a hydrogenase structural gene p r o d u c t (HydA) f r o m Clostridium or Chlamydomonas expressed in an E. coli host in the absence of any accessory enzymes (without H y d E , H y d F , or H y d G ) is capable of being converted into an active hydrogenase by the addition of E. coli cell extracts in which all three accessory enzymes are expressed in concert [224]. T h e results of this w o r k suggest that the bulk of cluster biosynthesis occurs at a site other t h a n the structural enzyme implicating a potential role of some scaffold in the process. Our recent results implicate a role of H y d F in this process and one could envision t h a t a precursor of the intact Η-cluster could be synthesized on H y d F in a m a n n e r involving sequential radical-SAM catalyzed reactions steps as described above (Figure 8). T h e in vitro system is just a small step in unraveling the complex process of Η-cluster biosynthesis but an i m p o r t a n t step in defining a viable biochemical in vitro assay to design f u t u r e experiments p r o b i n g the f u n c t i o n of the radical-SAM enzymes specifically. Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

207

[Sulfur insertion

filycinc radical Formation and decomposition

C l u s t e r insertion

Figure 8. Schematic representation depicting the proposed chemical steps of Η-cluster biosynthesis involving radical-SAM enzyme activities. The 2Fe subcluster assemblies and intermediates are represented as ball-and-stick models with carbon atoms in gray, oxygen in red, nitrogen in blue, sulfur in orange, and iron in rust brown. The central atom of the dithiolate linkage and the unknown iron ligands are represented in magenta. The ribbon representation of the overall structure of the Fe-only hydrogenase is from Clostridium pasteurianum (Cpl).

ACKNOWLEDGMENTS Hydrogenase research in this lab is supported by grants from the AirForce Office of Scientific Research (FA9550-05-01-0365), the Department of Energy (DE-FC36-06-G086060 and DE-FG02-07ER46477), and the NASA Astrobiology Biogeocatalysis Center of Montana State University Funded by NASA (NNA08CN85A). Portions of this research were carried out at the Standford Synchrotron Radiation Laboratory, a national user facility operated by Stanford University on behalf of the U.S. Department of Energy, Office of Basic Energy Sciences. The SSRL Met. Ions Life Sei. 2009, 6, 179-218

208

PETERS

Structural Molecular Biology Program is supported by the Department of Energy, Office of Biological and Environmental Research and by the National Institutes of Health, National Center for Research Resources, Biomedical Technology Program, and the National Institute of General Medical Sciences. Special thanks to David Mulder for assistance in preparation of this chapter.

ABBREVIATIONS AND DEFINITIONS BioB Cpl Crl DdH DFT dppv DTN DvH ENDOR EPR ESEEM FTIR HYSCORE IMES IR LAM LipA NADH PDT PFL-AE SAM Tma ThiH

biotin synthase Clostridium pasteurianum [FeFe]-hydrogenase Chlamydomonas reinhardtii [FeFe]-hydrogenase Desulfovibrio desulfuricans [FeFe]-hydrogenase density functional theory cis- l,2-C 2 H 2 (PPh 2 ) 2 dithiolate Desulfovibrio vulgaris Hildenborough [FeFe]-hydrogenase electron nuclear double resonance spectroscopy electron paramagnetic resonance spectroscopy electron spin echo envelope modulation spectroscopy Fourier transform infrared spectroscopy hyperfine sublevel correlation spectroscopy l,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene infrared lysine 2,3-aminomutase lipoate synthase nicotinamide adenine dinucleotide, reduced propanedithiolate pyruvate formate lyase activating enzyme S-adenosylmethionine Thermotoga maritima [FeFe]-hydrogenase thiazole synthase

REFERENCES 1. R. P. Happe, W. Roseboom, A. J. Pierik, S. P. Albracht and K. A. Bagley, Nature, 1997, 385, 126. 2. M. W. Adams, Biochim. Biophys. Acta, 1990, 1020, 115-145.

3. A. E. Przybyla, J. Robbins, N. Menon and H. D. Peck Jr., FEMS Microbiol. Rev., 1992, 8, 109-135.

Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

209

4. A. Yolbeda, Μ. H. Charon, C. Piras, E. C. Hatchikian, M. Frey and J. C. Fontecilla-Camps, Nature, 1995, 373, 580-587. 5. R. K. Thauer, A. R. Klein and G. C. Hartmann, Chem. Rev., 1996, 96, 3031-3042. 6. A. Yolbeda, J. C. Fontecilla-Camps and M. Frey, Curr. Opin. Struct. Biol., 1996, 6, 804-812. 7. Y. Montet, P. Amara, A. Volbeda, X. Vernede, E. C. Hatchikian, M . J. Field, M . Frey and J. C. Fontecilla-Camps, Nat. Struct. Biol., 1997, 4, 523-526. 8. Y. Higuchi, T. Yagi and N. Yasuoka, Structure, 1997, 5, 1671-1680. 9. J. C. Fontecilla-Camps, M. Frey, Ε. Garcin, C. Hatchikian, Y. Montet, C. Piras, X. Vernede and A. Volbeda, Biochimie, 1997, 79, 661-666. 10. E. Garcin, Y. Montet, A. Volbeda, C. Hatchikian, M. Frey and J. C. FontecillaCamps, Biochem. Soc. Trans., 1998, 26, 396-401. 11. J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt, Science, 1998, 282, 1853-1858. 12. A. J. Pierik, M. Hulstein, W. R. Hägen and S. P. Albracht, Eur. J. Biochem., 1998, 258, 572-578. 13. M. Rousset, Y. Montet, B. Guigliarelli, N. Forget, M. Asso, P. Bertrand, J. C. Fontecilla-Camps and E. C. Hatchikian, Proc. Natl. Acad. Sei. USA, 1998, 95, 11625-11630. 14. A. J. Pierik, W. Roseboom, R. P. Happe, K. A. Bagley and S. P. J. Albracht, J. Biol. Chem., 1999, 274, 3331-3337. 15. C. Popescu and E. Munck, J. Am. Chem. Soc., 1999, 121, 7877-7884. 16. E. Garcin, X. Vernede, E. C. Hatchikian, A. Volbeda, M. Frey and J. C. Fontecilla-Camps, Structure, 1999, 7, 557-566. 17. Y. Higuchi, H. Ogata, K. Miki, N. Yasuoka and T. Yagi, Structure, 1999, 7, 549-556. 18. Y. Nicolet, C. Piras, P. Legrand, C. E. Hatchikian and J. C. Fontecilla-Camps, Structure Folding & Design, 1999, 7, 13-23. 19. B. J. Lemon and J. W. Peters, Biochemistry, 1999, 38, 12969-12973. 20. Y. Nicolet, B. J. Lemon, J. C. Fontecilla-Camps and J. W. Peters, Trends Biochem. Sei., 2000, 25, 138-143. 21. Y. Nicolet, A. L. de Lacey, X. Vernede, V. M. Fernandez, E. C. Hatchikian and J. C. Fontecilla-Camps, J. Am. Chem. Soc., 2001, 123, 1596-1601. 22. A. Volbeda, Y. Montet, X. Vernede, E. Hatchikian and J. Fontecilla-Camps, Int. J. Hydrogen Energy, 2002, 27, 1449-1461. 23. H. Ogata, Y. Mizoguchi, N. Mizuno, K. Miki, S. Adachi, N. Yasuoka, T. Yagi, O. Yamauchi, S. Hirota and Y. Higuchi, J. Am. Chem. Soc., 2002, 124, 11628-11635. 24. A. L. DeLacey, V. M. Fernandez, M. Rousset, C. Cavazza and E. C. Hatchikian, J. Biol. Inorg. Chem., 2003, 8, 129-134. 25. F. A. Armstrong, Curr. Opin. Chem. Biol., 2004, 8, 133-140. 26. A. Volbeda, L. Martin, C. Cavazza, M. Matho, B. W. Faber, W. Roseboom, S. P. Albracht, E. Garcin, M. Rousset and J. C. Fontecilla-Camps, J. Biol. Inorg. Chem., 2005, 10, 239-249.

Met. Ions Life Sei. 2009, 6, 179-218

210

PETERS

27. A. L. de Lacey, Υ. M. Fernandez and M. Rousset, Coord. Chem. Rev., 2005, 249, 1596-1608. 28. S. Shima and R. K. Thauer, Chem. Record, 7, 37-46. 29. L. F. Wu and M. A. Mandrand, FEMS Microbiol. Rev., 1993, 104, 243-270. 30. P. Μ. Yignais, Β. Billoud and J. Meyer, FEMS Microbiol. Rev., 2001, 25, 455-501. 31. P. M. Yignais and A. Colbeau, Curr. Issues Mol. Biol., 2004, 6, 159-188. 32. J. Meyer, Cell. Mol. Life Sei., 2007, 64, 1063-1084. 33. P. M. Vignais and B. Billoud, Chem. Rev., 2007, 107, 4206-4272. 34. C. Zirngibl, W. Van Dongen, B. Schworer, R. Yon Bunau, M. Richter, A. Klein and R. K. Thauer, Eur. J. Biochem., 1992, 208, 511-520. 35. D. O. Hall, R. Cammack and Κ. K. Rao, Orig. Life, 1974, 5, 363-386. 36. G. Wächtershäuser, Proc. Natl. Acad. Sei USA, 1988, 85, 1134-1135. 37. G. Wächtershäuser, Origins of Life and Evolution of the Biosphere, 1990, 20, 173-176. 38. G. Wächtershäuser, Proc. Natl. Acad. Sei. USA, 1990, 87, 200-204. 39. G. Wächtershäuser, Pure Appl. Chem., 1993, 65, 1343-1348. 40. C. Huber and G. Wächtershäuser, Science, 1997, 276, 245-247. 41. G. Wächtershäuser, in: The Molecular Origins of Life: Assembling Pieces of the Puzzle, A. Brack, (Ed.), Cambridge University Press, New York, 1998, pp. 206-218. 42. W. Martin and M. J. Russell, Phil. Trans. Biol. Sei., 2003, 358, 59-85. 43. Μ. Schoonen, A. Smirnov and C. Cohn, Ambio, 2004, 33, 539-551. 44. F. Tian, Ο. B. Toon, A. A. Pavlov and H. De Sterck, Science, 2005, 308, 1014-1017. 45. K. A. Bagley, C. J. Van Garderen, Μ. Chen, Ε. C. Duin, S. P. Albracht and W. H. Woodruff, Biochemistry, 1994, 33, 9229-9236. 46. K. A. Bagley, E. C. Duin, W. Roseboom, S. P. Albracht and W. H. Woodruff, Biochemistry, 1995, 34, 5527-5535. 47. A. Volbeda, E. Garcin, C. Piras, A. L. de Lacey, Υ. M. Fernandez, E. C. Hatchikian, M. Frey and J. C. Fontecilla-Camps, J. Am. Chem. Soc., 1996,118, 12989-12996. 48. Τ. M. van der Spek, A. F. Arendsen, R. P. Happe, S. Yun, K. A. Bagley, D. J. Stufkens, W. R. Hägen and S. P. Albracht, Eur. J. Biochem., 1996, 237, 629-634. 49. A. L. de Lacey, E. C. Hatchikian, A. Volbeda, M. Frey, J. C. Fontecilla-Camps and V. M. Fernandez, J. Am. Chem. Soc., 1997, 119, 7181-7189. 50. R. P. Happe, W. Roseboom and S. P. Albracht, Eur. J. Biochem., 1999, 259, 602-608.

51. A. L. De Lacey, C. Stadler, C. Cavazza, E. C. Hatchikian and V. M. Fernandez, J. Am. Chem. Soc., 2000, 122, 11232-11233. 52. Z. Chen, B. J. Lemon, S. Huang, D. J. Swartz, J. W. Peters and K. A. Bagley, Biochemistry, 2002, 41, 2036-2043. 53. A. L. De Lacey, C. Stadler, V. Μ. Fernandez, Ε. C. Hatchikian, H. J. Fan, S. Li and Μ. B. Hall, J. Biol. Inorg. Chem., 2002, 7, 318-326.

Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

211

54. B. Bleijlevens, F. A. van Broekhuizen, A. L. De Lacey, W. Roseboom, Υ. M. Fernandez and S. P. Albracht, J. Biol. Inorg. Chem., 2004, 9, 743-752. 55. E. J. Lyon, S. Shima, R. Boecher, R. K. Thauer, F. W. Grevels, Ε. Bill, W. Roseboom and S. P. Albracht, J. Am. Chem. Soc., 2004, 126, 14239-14248. 56. T. Burgdorf, S. Löscher, P. Liebisch, E. Van der Linden, M. Galander, F. Lendzian, W. Meyer-Klaucke, S. P. Albracht, B. Friedrich, Η. Dau and M. Haumann, J. Am. Chem. Soc., 2005, 127, 576-592. 57. W. Roseboom, A. L. De Lacey, V. M. Fernandez, E. C. Hatchikian and S. P. Albracht, J. Biol. Inorg. Chem., 2006, 11, 102-118. 58. C. Fichtner, C. Laurich, Ε. Bothe and W. Lubitz, Biochemistry, 2006, 45, 9706-9716. 59. O. Schroder, B. Bleijlevens, Τ. E. de Jongh, Z. Chen, T. Li, J. Fischer, J. Forster, C. G. Friedrich, Κ. A. Bagley, S. P. Albracht and W. Lubitz, J. Biol. Inorg. Chem., 2007, 12, 212-233. 60. R. H. Holm, P. Kennepohl and Ε. I. Solomon, Chem. Rev., 1996, 96, 2239-2314. 61. G. Wächtershäuser, Prog. Biophys. Mol. Biol., 1992, 58, 85-201. 62. G. D. Cody, Ν. Z. Boctor, T. R. Filley, R. M. Hazen, J. H. Scott, A. Sharma and H. S. Yoder Jr., Science, 2000, 289, 1337-1340. 63. M. Dorr, J. Kassbohrer, R. Grunert, G. Kreisel, W. A. Brand, R. A. Werner, H. Geilmann, C. Apfel, C. Robl and W. Weigand, Angew. Chem. Int. Ed. Engl., 2003, 42, 1540-1543. 64. M. L. Ghirardi, L. Zhang, J. W. Lee, T. Flynn, M. Seibert, Ε. Greenbaum and A. Melis, Trends Biotechnol., 2000, 18, 506-511. 65. A. Melis, L. Zhang, M. Forestier, M. L. Ghirardi and M. Seibert, Plant Physiol., 2000, 122, 127-136. 66. P. Tamagnini, J. L. Costa, L. Almeida, M. J. Oliveira, R. Salema and P. Lindblad, Curr. Microbiol., 2000, 40, 356-361. 67. A. Melis and T. Happe, Plant Physiol., 2001, 127, 740-748. 68. T. Happe, A. Hemschemeier, M. Winkler and A. Kaminski, Trends Plant Sei., 2002, 7, 246-250. 69. F. Hawkes, R. Dinsdale, D. Hawkes and I. Hussy, Int. J. Hydrogen Energy, 2002, 27, 1339-1347. 70. A. Melis, Int. J. Hydrogen Energy, 2002, 27, 1217-1228. 71. K. Schutz, Τ. Happe, Ο. Troshina, P. Lindblad, E. Leitao, P. Oliveira and P. Tamagnini, Planta, 2004, 218, 350-359. 72. Μ. L. Ghirardi, P. W. King, M. C. Posewitz, P. C. Maness, A. Fedorov, K. Kim, J. Cohen, K. Schulten and M. Seibert, Biochem. Soc. Trans., 2005, 33, 70-72. 73. O. Kruse, J. Rupprecht, K. P. Bader, S. Thomas-Hall, P. M. Schenk, G. Finazzi and B. Hankamer, J. Biol. Chem., 2005, 280, 34170-34177. 74. P. C. Hallenbeck, Water Sei. Technol., 2005, 52, 21-29. 75. J. S. Chen and L. E. Mortenson, Biochim. Biophys. Acta, 1974, 371, 283-298. 76. Η. M. van der Westen, S. G. Mayhew and C. Veeger, FEBS Lett., 1978, 86, 122-126.

Met. Ions Life Sei. 2009, 6, 179-218

PETERS

212

77. J. S. Chen and D. Κ. Blanchard, Biochem. Biophys. Res. Commun., 1978, 84, 1144-1150. 78. C. Van Dijk, S. G. Mayhew, H. J. Grande and C. Yeeger, Eur. J. Biochem., 1979, 102, 317-330. 79. B. R. Glick, W. G. Martin and S. M. Martin, Can. J. Microbiol., 1980, 26, 1214-1223. 8 0 . C. van Dijk, H. J. Grande, S. G. Mayhew and C. Yeeger, Eur. J. Biochem., 1980, 107, 251-261. 81 C. van Dijk and C. Yeeger, Eur. J. Biochem., 1981, 114, 209-219. 82. H. J. Grande, W. R. Dunham, B. Averiii, C. Van Dijk and R. H. Sands, Eur. J. Biochem., 1983, 136, 201-207. 83. M. W. Adams and L. E. Mortenson, J. Biol. Chem., 1984, 259, 7045-7055. 84. Β. H. Huynh, Μ. H. Czechowski, H. J. Kruger, D. Y. DerVartanian, H. D. Peck Jr. and J. LeGall, Proc. Natl. Acad. Sei. USA, 1984, 81, 3728-3732. 85. M. W. W. Adams, Μ. K. Johnson, I. C. Zambrano and L. E. Mortenson, Biochimie, 1986, 68, 35-41. 8 6 . W. R. Hägen, A. van Berkel-Arts, Κ. M. Kruse-Wolters, G. Voordouw and C. Yeeger, FEBS Lett., 1986, 203, 59-63. 87. G. Fauque, H. D. Peck Jr., J. J. Moura, Β. H. Huynh, Y. Berlier, D. V. DerVartanian, M. Teixeira, A. E. Przybyla, P. A. Lespinat, I. Moura and J. LeGall, FEMS Microbiol. Rev., 1988, 4, 299-344. 8 8 . D. S. Patil, J. J. Moura, S. H. He, M. Teixeira, B. C. Prickril, D. V. DerVartanian, H. D. Peck Jr., J. LeGall and Β. H. Huynh, J. Biol. Chem., 1988, 263, 18732-18738. 89. D. S. Patil, H. Huynh Boi, S. H. He, H. D. Peck, D. V. DerVartanian and J. LeGall, J. Am. Chem. Soc., 1988, 110, 8533-8534. 90. D. S. Patil, S. H. He, D. V. DerVartanian, J. Le Gall, Β. H. Huynh and H. D. Peck Jr., FEBS Lett., 1988, 228, 85-88. 91. E. C. Hatchikian, N. Forget, V. M. Fernandez, R. Williams and R. Cammack, Eur. J. Biochem., 1992, 209, 357-365. 92. Μ. T. Bes, E. Parisini, L. A. Inda, L. M. Saraiva, M. L. Peleato and G. M. Sheldrick, Structure, 1999, 7, 1201-1211. 93. J. M. Moulis, L. C. Sieker, Κ. S. Wilson and Z. Dauter, Protein Sei., 1996, 5, 1765-1775. 94. Ε. Τ. Adman, L. C. Siefker and L. H. Jensen, J. Biol. Chem., 1976, 251, 3801-3806. 95. I. Bertini, Α. Donaire, B. A. Feinberg, C. Luchinat, M. Piccioli and H. Yuan, Eur. J. Biochem., 1995, 232, 192-205. 96. Z. Dauter, K. S. Wilson, L. C. Sieker, J. Meyer and J. M. Moulis, Biochemistry, 1997, 36, 16065-16073. 97. M. Unciuleac, M. Boll, E. Warkentin and U. Ermler, Acta Cryst. D Biol. Cryst., 2004, 60, 388-391. 98. P. Giastas, N. Pinotsis, G. Efthymiou, M. Wilmanns, P. Kyritsis, J. M. Moulis and I. M. Mavridis, J. Biol. Inorg. Chem., 2006, 11, 445-458. 99. G. Voordouw and S. Brenner, Eur. J. Biochem., 1985, 148, 515-520.

Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

213

100. I. Naud, C. Meyer, L. David, J. Breton, J. Gaillard and Y. Jouanneau, Eur. J. Biochem., 1996, 237, 399-405. 101. K. Saeki, K. Tokuda, K. Fukuyama, H. Matsubara, K. Nadanami, M. Go and S. Itoh, J. Biol. Chem., 1996, 271, 31399-31406. 102. T. Tsukihira, K. Fukuyama, M. Nakamura, Y. Katsube, N. Tanaka, M. Kakudo, K. Wada, T. Hase and H. Matsubara, J. Biochem., 1981, 90, 1763-1773. 103. T. Tsukihara, K. Fukuyama, M. Mizushima, T. Harioka, M. Kusunoki, Y. Katsube, T. Hase and H. Matsubara, J. Mol. Biol., 1990, 216, 399^10. 104. W. R. Rypniewski, D. R. Breiter, Μ. Μ. Benning, G. Wesenberg, Β. Η. Oh, J. L. Markley, I. Rayment and Η. M. Holden, Biochemistry, 1991, 30, 4126-4131. 105. B. L. Jacobson, Υ. K. Chae, J. L. Markley, I. Rayment and Η. M. Holden, Biochemistry, 1993, 32, 6788-6793. 106. S. Ikemizu, M. Bando, T. Sato, Y. Morimoto, T. Tsukihara and K. Fukuyama, Acta Cryst. D Biol. Cry St., 1994, 50, 167-174. 107. K. Fukuyama, N. Ueki, H. Nakamura, T. Tsukihara and H. Matsubara, J. Biochem., 1995, 117, 1017-1023. 108. F. Frolow, M. Harel, J. L. Sussman, M. Mevarech and M. Shoham, Nat. Struct. Bio.l, 1996, 3, 452-458. 109. C. Binda, A. Coda, A. Aliverti, G. Zanetti and A. Mattevi, Acta Cryst. D Biol Cryst., 1998, 54, 1353-1358. 110. P. M. Matias, C. M. Soares, L. M. Saraiva, R. Coelho, J. Morais, J. Le Gall and M. A. Carrondo, J. Biol. Inorg. Chem., 2001, 6, 63-81. 111. M. F. Yerhagen, T. O'Rourke and M. W. Adams, Biochim. Biophys. Acta, 1999, 1412, 212-229. 112. B. Soboh, D. Linder and R. Hedderich, Microbiology, 2004, 150, 2451-2463. 113. T. Happe and J. D. Naber, Eur. J. Biochem., 1993, 214, 4 7 5 ^ 8 1 . 114. L. Florin, A. Tsokoglou and T. Happe, J. Biol. Chem., 2001, 276, 6125-6132. 115. R. Wunschiers, K. Stangier, H. Senger and R. Schulz, Curr. Microbiol., 2001, 42, 353-360. 116. T. Happe and A. Kaminski, Eur. J. Biochem., 2002, 269, 1022-1032. 117. M. Winkler, B. Heil, Β. Heil and Τ. Happe, Biochim. Biophys. Acta, 2002,1576, 330-334. 118. M. Forestier, P. King, L. Zhang, M. Posewitz, S. Schwarzer, Τ. Happe, Μ. L. Ghirardi and M. Seibert, Eur. J. Biochem., 2003, 270, 2750-2758. 119. C. Kamp, A. Silakov, M. Winkler, E. J. Reijerse, W. Lubitz and T. Happe, Biochim. Biophys. Acta, 2008. 120. B. Boxma, G. Ricard, A. H. van Hoek, E. Severing, S. Y. Moon-van der Staay, G. W. van der Staay, T. A. van Alen, R. M. de Graaf, G. Cremers, M. Kwantes, N. R. McEwan, C. J. Newbold, J. P. Jouany, T. Michalowski, P. Pristas, M. A. Huynen and J. H. Hackstein, BMC Evol. Biol., 2007, 7, 230. 121. B. Boxma, R. M. de Graaf, G. W. van der Staay, T. A. van Alen, G. Ricard, Τ. Gabaldon, A. H. van Hoek, S. Y. Moon-van der Staay, W. J. Koopman, J. J. van Hellemond, A. G. Tielens, T. Friedrich, Μ. Yeenhuis, Μ. A. Huynen and J. H. Hackstein, Nature, 2005, 434, 74-79.

Met. Ions Life Sei. 2009, 6, 179-218

PETERS

214

122. J. Balk, A. J. Pierik, D. J. Netz, U. Muhlenhoff and R. Lill, Embo J, 2004, 23, 2105-2115. 123. J. Balk, A. J. Pierik, D. J. Aguilar Netz, U. Muhlenhoff and R. Lill, Biochem. Soc. Trans., 2005, 33, 86-89. 124. J. Huang, D. Song, A. Flores, Q. Zhao, S. M. Mooney, L. M. Shaw and F. S. Lee, Biochem. J., 2006. 125. B. Bennett, B. J. Lemon and J. W. Peters, Biochemistry, 2000, 39, 7455-7460. 126. J. W. van der Zwaan, J. M. Coremans, E. C. Bouwens and S. P. Albracht, Biochim. Biophys. Acta, 1990, 1041, 101-110. 127. R. K. Thauer, B. Kaufer, Μ. Zahringer and Κ. Jungermann, Eur. J. Biochem., 1974, 42, 447-452. 128. D. L. Erbes, R. H. Burris and W. H. Orme-Johnson, Proc. Natl. Acad. Sei. USA, 1975, 72, 4795-4799. 129. M. W. Adams, J. Biol. Chem., 1987, 262, 15054-15061. 130. A. T. Kowal, M. W. Adams and Μ. K. Johnson, J. Biol. Chem., 1989, 264, 4342-4348. 131. H. J. Fan and Μ. B. Hall, J. Am. Chem. Soc., 2001, 123, 3828-3829. 132. Α. I. Krasna and D. Rittenberg, J. Am. Chem. Soc., 1954, 76, 3015-3020. 133. Z. Liu and P. Hu, J. Chem. Phys., 2002, 117, 8177-8180. 134. A. S. Pandey, Τ. Y. Harris, L. J. Giles, J. W. Peters and R. K. Szilagyi, J. Am. Chem. Soc., 2008, 130, 4533-4540. 135. J. Telser, M. J. Benecky, M. W. Adams, L. E. Mortenson and Β. M. Hoffman, J. Biol. Chem., 1986, 261, 13536-13541. 136. I. C. Zambrano, A. T. Kowal, L. E. Mortenson, M. W. Adams and Μ. K. Johnson, J. Biol. Chem., 1989, 264, 20974-20983. 137. A. J. Pierik, W. R. Hagen, J. S. Redeker, R. B. Wolbert, M. Boersma, M. F. Yerhagen, H. J. Grande, C. Veeger, P. H. Mutsaers, R. H. Sands and R. W. Dunham, Eur. J. Biochem., 1992, 209, 63-72. 138. S. P. Albracht, W. Roseboom and E. C. Hatchikian, J. Biol. Inorg. Chem., 2006, 11, 8 8 - 1 0 1 .

139. G. Wang, M. J. Benecky, Β. H. Huynh, J. F. Cline, M. W. Adams, L. E. Mortenson, Β. M. Hoffman and E. Munck, J. Biol. Chem., 1984, 259, 14328-14331. 140. H. Thomann, M. Bernardo and M. W. W. Adams, J. Am. Chem. Soc., 1991, 113, 7044-7046. 141. A. Silakov, E. J. Reijerse, S. P. Albracht, E. C. Hatchikian and W. Lubitz, J. Am. Chem. Soc., 2007, 129, 11447-11458. 142. P. J. van Dam, E. J. Reijerse and W. R. Hägen, Eur. J. Biochem., 1997, 248, 355-361. 143. R. Williams, R. Cammack and E. C. Hatchikian, J. Chem. Soc. Faraday Trans., 1993, 89, 2869-2872. 144. F. M. Rusnak, M. W. Adams, L. E. Mortenson and E. Munck, J. Biol. Chem., 1987, 262, 38-41. 145. A. S. Pereira, P. Tavares, I. Moura, J. J. Moura and Β. H. Huynh, J. Am. Chem. Soc., 2001, 123, 2771-2782.

Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

215

146. W. R. Hägen, A. van Berkel-Arts, Κ. M. Kruse-Wolters, W. R. Dunham and C. Yeeger, FEBS Lett., 1986, 201, 158-162. 147. E. C. Hatchikian, V. Magro, N. Forget, Y. Nicolet and J. C. Fontecilla-Camps, J. Bacteriol., 1999, 181, 2947-2952. 148. Z. Cao and Μ. B. Hall, J. Am. Chem. Soc., 2001, 123, 3734-3742. 149. Z. P. Liu and P. Hu, J. Am. Chem. Soc., 2002, 124, 5175-5182. 150. A. T. Fiedler and T. C. Brunold, Inorg. Chem., 2005, 44, 9322-9334. 151. E. J. Lyon, I. P. Georgakaki, J. H. Reibenspies and Μ. Y. Darensbourg, Angew. Chem. Int. Ed. Engl., 1999, 38, 3178-3180. 152. A. Le Cloirec, S. P. Best, S. Borg, S. C. Davies, D. J. Evans, D. L. Hughes and C. J. Pickett, Chem. Commun., 1999, 2285-2286. 153. M. Schmidt, S. M. Contakes and Τ. B. Rauchfuss, J. Am. Chem. Soc., 1999,121, 9736-9737. 154. S. J. George, Z. Cui, M. Razavet and C. J. Pickett, Chemistry, 2002, 8, 4037-4046. 155. Μ. Y. Darensbourg, E. J. Lyon, X. Zhao and I. P. Georgakaki, Proc. Natl. Acad. Sei. USA, 2003, 100, 3683-3688. 156. J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt, Science, 1998, 282, 1853-1858. 157. W. Lubitz, E. Reijerse and M. van Gastel, Chem. Rev., 2007, 107, 4331^365. 158. A. Le Cloirec, S. P. Best, S. Borg, S. C. Davies, D. J. Evans, D. L. Hughes and C. J. Pickett, Chem. Commun., 1999, 2285-2286. 159. J. D. Lawrence, H. Li, Τ. B. Rauchfuss, Μ. Benard and Μ. Μ. Rohmer, Angew. Chem. Int. Ed. Engl., 2001, 40, 1768-1771. 160. J. D. Lawrence, Η. X. Li and Τ. B. Rauchfuss, Chem. Commun., 2001, 1482-1483. 161. E. J. Lyon, I. P. Georgakaki, J. H. Reibenspies and Μ. Y. Darensbourg, J. Am. Chem. Soc., 2001, 123, 3268-3278. 162. F. Gloaguen, J. D. Lawrence, M. Schmidt, S. R. Wilson and Τ. B. Rauchfuss, J. Am. Chem. Soc., 2001, 123, 12518-12527. 163. F. Gloaguen, J. D. Lawrence and Τ. B. Rauchfuss, Am. Chem. Sox, 2001, 123, 9476-9477. 164. M. Razavet, S. C. Davies, D. L. Hughes and C. J. Pickett, Chem. Commun., 2001, 847-848. 165. X. Zhao, I. P. Georgakaki, M. L. Miller, J. C. Yarbrough and Μ. Y. Darensbourg, J. Am. Chem. Soc., 2001, 123, 9710-9711. 166. X. Zhao, I. P. Georgakaki, M. L. Miller, R. Mejia-Rodriguez, C. Y. Chiang and Μ. Y. Darensbourg, Inorg. Chem., 2002, 41, 3917-3928. 167. H. Li and Τ. B. Rauchfuss, J. Am. Chem. Soc., 2002, 124, 726-727. 168. S. George, Z. Cui, M. Razavet and C. J. Pickett, Chemistry, 2002, 8, 4037^046. 169. M. Razavet, S. J. Borg, S. J. George, S. P. Best, S. A. Fairhurst and C. J. Pickett, Chem. Commun., 2002, 700-701. 170. J. D. Lawrence, Τ. B. Rauchfuss and S. R. Wilson, Inorg. Chem., 2002, 41, 6193-6195. 171. A. Kayal and Τ. B. Rauchfuss, Inorg. Chem., 2003, 42, 5046-5048.

Met. Ions Life Sei. 2009, 6, 179-218

216

PETERS

172. Μ. Razavet, S. C. Davies, D. L. Hughes, J. E. Barclay, D. J. Evans, S. A. Fairhurst, X. M. Liu and C. J. Pickett, Dalton Trans., 2003, 586-595. 173. J. L. Nehring and D. M. Heinekey, Inorg. Chem., 2003, 42, 4288-4292. 174. L. C. Song, Ζ. Y. Yang, Η. Z. Bian and Q. M. Hu, Organometallies, 2004, 23, 3082-3084. 175. F. Wang, M. Wang, X. Liu, K. Jin, W. Dong, G. Li, B. Akermark and L. Sun, Chem. Commun., 2005, 3221-3223. 176. L. C. Song, J. Cheng, J. Yan, Η. T. Wang, X. F. Liu and Q. M. Hu, Organometallics, 2006, 25, 1544-1547. 177. L. C. Song, J. H. Ge, X. F. Liu, L. Q. Zhao and Q. M. Hu, J. Organomet. Chem., 2006, 691, 5701-5709. 178. S. Ezzaher, J. F. Capon, F. Gloaguen, F. Y. Petillon, P. Schollhammer, J. Talarmin, R. Pichon and N. Kervarec, Inorg. Chem., 2007, 46, 3426-3428. 179. D. Morvan, J. F. Capon, F. Gloaguen, P. Schollhammer and J. Talarmin, Eur. J. Inorg. Chem., 2007, 5062-5068. 180. P. Y. Orain, J. F. Capon, N. Kervarec, F. Gloaguen, F. Petillon, R. Pichon, P. Schollhammer and J. Talarmin, Dalton Trans., 2007, 3754-3756. 181. G. M. Jacobsen, R. K. Shoemaker, M. Rakowski DuBois and D. L. DuBois, Organometallics, 2007, 26, 4964-4971. 182. A. K. Justice, Τ. B. Rauchfuss and S. R. Wilson, Angew. Chem. Int. Ed. Engl., 2007, 46, 6152-6154. 183. T. Liu and Μ. Y. Darensbourg, J. Am. Chem. Soc., 2007, 129, 7008-7009. 184. Β. E. Barton and Τ. B. Rauchfuss, Inorg. Chem., 2008. 185. A. Reihlen, A. Grul and G. Hessling, Liebigs Ann. Chem., 1929, 472, 268. 186. D. Seyferth, R. S. Henderson and L. C. Song, Organometallics, 1982,1, 125-133. 187. D. Seyferth, G. B. Womack, Μ. K. Gallagher, M. Cowie, B. W. Hames, J. P. Fackler and A. M. Mazany, Organometallics, 1987, 6, 283-294. 188. D. Seyferth, G. B. Womack, C. M. Archer and J. C. Dewan, Organometallics, 1989, 8, 430^42. 189. A. Winter, L. Zsolnai and G. Huttner, Z. Naturforsch., Β: Anorg. Chem., Org. Chem., 1982, 1430-1436. 190. J. A. Ayllon, S. F. Sayers, S. Sabo-Etienne, B. Donnadieu, B. Chaudret and E. Clot, Organometallics, 1999, 18, 3981-3990. 191. D. H. Lee, B. P. Patel, E. Clot, O. Eisenstein and R. H. Crabtree, Chem. Commun., 1999, 297-298. 192. W. Xu, A. J. Lough and R. H. Morris, Inorg. Chem., 1996, 35, 1549-1555. 193. X. M. Liu, S. K. Ibrahim, C. Tard and C. J. Pickett, Coord. Chem. Rev., 2005, 249, 1641-1652. 194. C. Tard, X. Liu, S. K. Ibrahim, M. Bruschi, L. De Gioia, S. C. Davies, X. Yang, L. S. Wang, G. Sawers and C. J. Pickett, Nature, 2005, 433, 610-613. 195. D. E. Schwab, C. Tard, E. Brecht, J. W. Peters, C. J. Pickett and R. K. Szilagyi, Chem. Commun., 2006, 3696-3698. 196. P. C. Dos Santos, D. R. Dean, Y. L. Hu and M. W. Ribbe, Chem. Rev., 2004, 104, 1159-1173. 197. L. M. Rubio and P. W. Ludden, J. Bacteriol., 2005, 187, 405-414.

Met. Ions Life Sei. 2009, 6, 179-218

CARBON MONOXIDE AND CYANIDE IN [FeFe]-HYDROGENASES

217

198. A. Paschos, R. S. Glass and A. Bock, FEBS Lett, 2001, 488, 9-12. 199. S. Reissmann, Ε. Hochleitner, Η. Wang, A. Paschos, F. Lottspeich, R. S. Glass and A. Bock, Science, 2003, 299, 1067-1070. 200. W. Roseboom, M . Blokesch, A. Bock and S. P. Albracht, FEBS Lett., 2005, 579, 469-472. 201. Ο. Lenz, I. Zebger, J. H a m a n n , P. Hildebrandt and B. Friedrich, FEBS Lett., 2007, 581, 3322-3326. 202. L. Forzi, P. Hellwig, R. K . Thauer and R. G. Sawers, FEBS Lett., 2007, 581, 3317-3321. 203. Μ . C. Posewitz, P. W. King, S. L. Smolinski, L. Zhang, Μ . Seibert and M . L. Ghirardi, J. Biol. Chem., 2004, 279, 25711-25720. 204. R. L. Kerby, S. S. Hong, S. A. Ensign, L. J. Coppoc, P. W. Ludden and G. P. Roberts, J. BacterioL, 1992, 174, 5284-5294. 205. A. Bock, P. W. King, M . Blokesch and M . C. Posewitz, Adv. Microb. Physiol., 2006, 51, 1-71. 206. X. Brazzolotto, J. K. R u b a c h , J. Gaillard, S. Gambarelli, M . A t t a and M . Fontecave, J. Biol. Chem., 2006, 281, 769-774. 207. J. K . R u b a c h , X. Brazzolotto, J. Gaillard and M . Fontecave, FEBS Lett., 2005, 579, 5055-5060. 208. J. W. Peters, R. K. Szilagyi, A. N a u m o v and T. Douglas, FEBS Lett., 2006, 580, 363-367. 209. D . C. Johnson, D. R. Dean, A. D. Smith and Μ . K . Johnson, Annu. Rev. Biochem., 2005, 74, 247-281. 210. R. M . Allen, R. Chatterjee, M . S. M a d d e n , P. W. L u d d e n and Υ. K . Shah, Crit. Rev. Biotechnol., 1994, 14, 225-249. 211. E. C. Duin, Μ . E. Lafferty, B. R. Crouse, R. M . Allen, I. Sanyal, D . H . Flint and Μ . K . Johnson, Biochemistry, 1997, 36, 11811-11820. 212. J. R. Miller, R. W. Busby, S. W. Jordan, J. Cheek, Τ. F. Henshaw, G. W. Ashley, J. B. Broderick, J. E. C r o n a n and M . A. Marietta, Biochemistry, 2000, 39, 15166-15178. 213. Ν . Β. Ugulava, Β. R. Gibney and J. Τ. Jarrett, Biochemistry, 2001, 40, 8343-8351. 214. Β. Τ. S. Bui, Μ . Lotierzo, F. Escalettes, D. Florentin and A. M a r q u e t , Biochemistry, 2004, 43, 16432-16441. 215. R. M . Cicchillo and S. J. Booker, J. Am. Chem. Soc., 2005, 127, 2860-2861. 216. Μ . M . Cosper, G. N. L. Jameson, H . L. Hernandez, C. Krebs, Β. Η . H u y n h and Μ . Κ . Johnson, Biochemistry, 2004, 43, 2007-2021. 217. J. Τ. Jarrett, Chem. Biol., 2005, 12, 4 0 9 ^ 1 0 . 218. R. Kulzer, T. Pils, R. Kappl, J. H u t t e r m a n n and J. K n a p p e , J. Biol. Chem., 1998, 273, 4 8 9 7 ^ 9 0 3 . 219. P. A. Frey, Annu. Rev. Biochem., 2001, 70, 121-148. 220. T. P. Begley, J. Xi, C. Kinsland, S. Taylor and F. McLafferty, Curr. Opin. Chem. Biol., 1999, 3, 623-629. 221. R. Leonardi, S. A. Fairhurst, M . Kriek, D. J. Lowe and P. L. Roach, FEBS Lett., 2003, 539, 95-99. 222. S. L. Miller and H. C. Urey, Science, 1953, 117, 528-529.

Met. Ions Life Sei. 2009, 6, 179-218

218

PETERS

223. G. Wächtershäuser, Science, 2000, 289, 1307-1308. 224. S. E. McGlynn, S. S. Ruebush, A. Naumov, L. E. Nagy, A. Dubini, P. W. King, J. B. Broderick, M. C. Posewitz and J. W. Peters, J. Biol. Inorg. Chem., 2007,12, 443-447. 225. L. A. Kelley, R. M. MacCallum and M. J. Sternberg, J. Mol. Biol., 2000, 299, 499-520. 226. R. Bennett-Lovsey, A. Herbert, M. Sternberg and L. Kelley, Proteins, 2008, 70, 611-625.

Met. Ions Life Sei. 2009, 6, 179-218

Met. Ions Life Sei. 2009, 6, 219-240

7 Carbon Monoxide as Intrinsic Ligand to Iron in the Active Site of [Fe]-Hydrogenase Seigo Shima,a Rudolf K. Thciuev, and Ulrich Ermlerh a

Max Planck Institute for Terrestrial Microbiology, Karl-von-Frisch-Strasse, D-35043 Marburg, Germany

b Max Planck Institute for Biophysics, Max-von-Laue-Strasse 3, D-60438 Frankfurt/Main, Germany

ABSTRACT 1. INTRODUCTION 2. PHYSIOLOGY 3. THE IRON GUANYLYLPYRIDINOL COFACTOR IN THE ENZYME-FREE STATE 3.1. Isolation of the Iron Guanylylpyridinol Cofactor and Activity Assay 3.2. Proposed Structure of the Iron Guanylylpyridinol Cofactor 3.3. Electronic and Magnetic State of the Iron 3.4. Stability of the Iron Guanylylpyridinol Cofactor 4. STRUCTURE OF [Fe]-HYDROGENASE WITH AND WITHOUT THE IRON GUANYLYLPYRIDINOL COFACTOR BOUND 4.1. Apo- and Holoenzyme Production 4.2. Crystal Structure Determination 4.3. Protein Fold 4.4. The Binding Sites for the Iron Guanylylpyridinol Cofactor and for Methenyl-H 4 MPT +

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00219

220 220 222 223 223 224 225 225 227 227 228 228 229

220

SHIMA, THAUER, and ERMLER

5. LIGANDS TO I R O N IN THE ACTIVE SITE OF [Fe]HYDROGENASE 5.1. Intrinsic Ligands 5.2. Carbon Monoxide and Cyanide as Extrinsic Ligands 5.3. Inactivation of [Fe]-Hydrogenase by 0 2 and Cu(I) 6. PROPOSED CATALYTIC MECHANISMS 6.1. H 2 Binds First to the Carbocationic C 1 4 a of Methenyl-H 4 MPT + 6.2. H 2 Binds First to the Active Site Iron Carbonyl Complex 7. C O N C L U D I N G R E M A R K S ACKNOWLEDGMENTS ABBREVIATIONS REFERENCES

231 231 232 234 235 235 236 237 237 237 238

ABSTRACT: Structural and spectroscopic studies on [Fe]-hydrogenase revealed an active site mononuclear low spin iron coordinated by the Cysl76 sulfur, two CO, and the sp2 hybridized nitrogen of a 2-pyridinol compound with back bonding properties similar to those of cyanide. Thus, [Fe]-hydrogenases are endowed with an iron-ligation pattern related to that found in the active site of [NiFe]- and [FeFe]-hydrogenases although the three hydrogenases and the enzymes involved in their posttranslational maturation have evolved independently and although CO and cyanide ligands are not found in any other metallo-enzymes. Obviously, low-spin iron complexed with thiolate(s), CO, and cyanide or a cyanide functional analogue plays an essential role in H 2 activation. KEYWORDS: carbon monoxide · cyanide · hydrogen activation · hydrogenase · iron carbonyl · iron guanylylpyridinol cofactor · methanogenic archaea

1.

INTRODUCTION

[Fe]-hydrogenase was discovered in 1990 and according to its systematic name was referred to as H 2 -forming methylenetetrahydromethanopterin dehydrogenase, abbreviated Hmd, which is also the abbreviation for the encoding gene (hmd) [1,2]. The enzyme was found as a homodimer with a molecular mass of ca. 80kDa that contains 2 iron per mol homodimer but no iron-sulfur clusters [1,2]. The iron was first thought to be non-functional, mainly because of its redox inactivity, and therefore the enzyme was addressed as metal-free hydrogenase [3,4] in contrast to the [NiFe]- and [FeFe]-hydrogenases. It had been overlooked that the enzyme is reversibly inhibited by CO and by cyanide, albeit only at relatively high concentrations. Especially CO inhibition indicates an involvement of the iron in catalysis. It took until 2004 to demonstrate unambiguously that the iron is essentially required for the catalytic activity of the enzyme [5]. After another two years the iron center was definitively identified as mononuclear [6] and not Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

221

dinuclear as in [NiFe]-hydrogenases (see Chapter 5 of this volume) and [FeFe]-hydrogenases (Chapter 6) and the enzyme was finally named as [Fe]-hydrogenase. Whereas [NiFe]-hydrogenases are found in many archaea and bacteria and [FeFe]-hydrogenases in bacteria and lower eucarya, the occurrence of [Fe]-hydrogenase appears to be restricted to methanogenic archaea growing on H 2 and C 0 2 as sole energy source. The reason for this is that [Fe]hydrogenase catalyzes one specific reaction in their energy metabolism [7] namely the reduction of methenyltetrahydromethanopterin (methenylH 4 M P T + ) with H 2 to methylenetetrahydromethanopterin (methyleneH 4 MPT) (reaction 1) (Figure 1) [8,9], H 2 + methenyl-H 4 MPT + - methylene-H 4 MPT + H+ AG°'= -5.5kJ/mol

^

[Fe]-hydrogenase contains tightly bound an iron guanylylpyridinol cofactor (FeGP cofactor). The iron is ligated to the guanylylpyridinol molecule, 2 CO, and the sulfur of Cysl76 which covalently links the FeGP cofactor to the protein [6]. This ligation pattern is reminescent to that of [FeFe]- and [NiFe]-hydrogenases which also contain an iron complexed by CO and by cyanide. The pyridinol moiety of FeGP might be a functional analogue to cyanide (see below). Interestingly, neither [Fe]-hydrogenases, [FeFe]hydrogenases and [NiFe]-hydrogenases nor the enzymes involved in their posttranslational maturation, as far as known, are phylogenetically related and intrinsic CO and cyanide ligands are not found in any other metalloenzyme [10]. This finding indicates that the H 2 activation sites in the three enzymes have evolved convergently [11] and attributes to them a unique function in H 2 activation.

H,N Figure 1. Reaction catalyzed by [Fe]-hydrogenase (Hmd). A hydride is transferred from H? to C 1 4 a of methenyl-H 4 MPT + from the pro-R side yielding methyleneH 4 M P T and a proton [8,9]. The methenyl C 1 4 a has carbocation character and is therefore an excellent hydride acceptor [32-34]. H 4 M P T , tetrahydromethanopterin. Met. Ions Life Sei. 2009, 6, 219-240

222

SHIMA, THAUER, and ERMLER

The definition of [Fe]-hydrogenase as "hydrogenase" can be questioned. The pros are that the enzyme catalyzes a reversible reaction with H 2 as substrate and that in H 2 activation it involves an active site iron carbonyl as do [NiFe]- and [FeFe]-hydrogenases (see Chapters 5 and 6). The con is that [Fe]hydrogenase exhibits a ternary complex catalytic mechanism [12] whereas [NiFe]- and [FeFe]-hydrogenases show ping-pong catalytic mechanisms. Thus, [Fe]-hydrogenase does not react with H 2 in the absence of its substrate methenyl-H 4 MPT + [13] whereas [NiFe]- and [FeFe]-hydrogenases are reduced by H 2 in the absence of an external hydride or electron acceptor. This difference is reflected in the finding that [Fe]-hydrogenase catalyzes a single and double exchange of H 2 with protons of water only in the presence of methenylH 4 M P T + [12-16] whereas [NiFe]- and [FeFe]-hydrogenases do this per se. The literature on [Fe]-hydrogenase has recently been reviewed by the authors [17]. This chapter will therefore concentrate on the F e G P cofactor, in particular, on the ligation pattern of the iron.

2.

PHYSIOLOGY

In methanogens reaction (1) is also catalyzed by two other enzymes working together, namely by F 4 2 0 -reducing [NiFe]-hydrogenase (Frh) and F 4 2 0 -dependent methylenetetrahydromethanopterin dehydrogenase (Mtd), which catalyze reactions (2) and (3), respectively [18]. n 2 + F42o - F 4 2 0 H 2

AG°' = - 1 1 k J / m o l

(2)

F 4 2 0 H 2 + methenyl-H 4 MPT+ ^ F 4 2 0 + methylene-H 4 MPT + H+ A G 0 ' = +5.5 k J / m o l

^

This explains why not all methanogens have to contain [Fe]-hydrogenase. The enzyme is absent in all investigated members of the Methanosarcinales, in most members of the Methanomicrobiales, and some members of the Methanobacteriales. The methanogens, that contain [Fe]-hydrogenase, are adapted to niches, where the H 2 partial pressure is relatively high and the nickel concentration can be low. This is deduced from the finding that the apparent K m for H 2 of [Fe]-hydrogenase is ten times higher than that of the F 4 2 0 -reducing [NiFe]-hydrogenases and that the [Fe]-hydrogenase is overproduced in the methanogens under nickel limiting growth conditions, under which the F 4 2 0 -reducing [NiFe]-hydrogenases are not synthesized [18,19]. [Fe]-hydrogenase has been found in methanogens thriving in hot vents (Methanopyrus Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

223

kandleri [20] and Methanocaldococcus jannaschii [21]), in salt marshes (.Methanococcus maripaludis [22]), in anaerobic sewage digestion plants (.Methanothermobacter thermoautotrophicus [23] and M. marburgensis [1]) and the human intestinal tract (Methanobrevibacter smithii [24]). After growth of M. marburgensis under nickel limiting conditions the cell contains [Fe]-hydrogenase at a concentration of nearly 5% of its soluble proteins, from which the enzyme can be purified to homogeneity in high yields [18], All methanogens with [Fe]-hydrogenase were found to harbour a gene nikR, a homolog of which encodes for a protein NikR that senses the intracellular nickel concentration and binds to the promoter region of the nickel importer genes nikABCDE and represses their transcription when containing nickel bound [25]. In E. coli NikR binds specifically to a palindrome sequence (CTATGA-NI 6 -TCATAG) [26]. A similar palindrome sequence is also found in the promoter region of the hmd gene in M. marburgensis ( G T A C T A C - N 1 4 - G T A T T A C ) , M. maripaludis (GTATTA-N15A T A T T A C ) , a n d M. jannaschii ( A T A T T A C - N 1 4 - A T A T T A C [25,27],

3.

THE IRON GUANYLYLPYRIDINOL COFACTOR IN THE ENZYME-FREE STATE

Most of what is known about the cofactor has been worked out with the cofactor of [Fe]-hydrogenase from M. marburgensis. This thermophilic methanogen is easy to grow under nickel limiting conditions in 100 g (wet mass) amounts, from which approximately 100 mg of enzyme are purified to homogeneity in a yield of 50%.

3.1.

Isolation of the Iron Guanylylpyridinol Cofactor and Activity Assay

From 100 mg of [Fe]-hydrogenase (1.25 μηιοί homodimer) approximately 2 μηιοί cofactor are obtained in a procedure, in which the enzyme solution is supplemented with methanol (60% final concentration), ammonia (600 mM) and mercaptoethanol ( I m M ) . Mercaptoethanol or other thiol compounds thereby displace the cysteine sulfur ligand to iron through which the FeGP cofactor is covalently bound to the protein (reaction 4). After incubation for 16 hours at 4°C, 1 Μ NaCl in 50 m M Tris/HCl p H 8.0 is added to a final concentration of 0.2 Μ and the precipitated material removed by centrifugation and the cofactor purified by H P L C [28]. Active enzyme (Hmd holoenzyme) is regenerated upon addition of the extracted cofactor to refolded or Met. Ions Life Sei. 2009, 6, 219-240

224

SHIMA, THAUER, and ERMLER

heterologously produced Hmd apoenzyme (reaction 5) providing an activity assay for the FeGP cofactor [4]. [FeJ-hydrogenase + mercaptoethanol

(in 60% mehanol)

>

^

unfolded Hmd apoenzyme + FeGP cofactor folded Hmd apoenzyme + FeGP cofactor —>

_

[Fe]-hydrogenase + mercaptoethanol

3.2.

Proposed Structure of the Iron Guanylylpyridinol Cofactor

The structure of the protein-free FeGP cofactor is not yet know but can be deduced from the decomposition products after irradiation of the cofactor with white light [5], from the FTIR spectrum [29], and from the XAS spectra, which indicates that the iron in the cofactor is complexed by 2 CO, 1 sulfur and two O/N ligands [6], and from the crystal structure of the [Fe]-hydrogenase holoenzyme [11]. The S ligand is most probably provided by mercaptoethanol required to detach the cofactor from the protein (reaction 4), to which the cofactor is covalently bound via a cysteine sulfur-iron bond. One of the two O/N ligands is provided by the pyridone nitrogen. The other O/N ligand is most likely an oxygen of the 2-car boxy methyl group of the pyridinol as deduced from the finding that during MALDI TOF MS of the FeGP cofactor the carboxyl group is lost whereas during MS of the guanylylpyridone it is not [28]. The proposed structure is shown in Figure 2.

Figure 2. Proposed structure of the protein-free F e G P cofactor. The structure of the iron-free guanylylpyridone was elucidated by N M R and mass spectrometry [28]. The structure of the iron site is deduced f r o m IR spectra [29], Mössbauer spectra [30], and X A spectra [6]. The results are also consistent with a structure in which the carboxyl of the carboxymethyl group of the pyridinol forms a Fe-CO-CH 2 - rather than a F e - 0 - C 0 - C H 2 - bond and in which one of the two CO are bound to the Fe opposite to the pyridinol nitrogen. Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

225

Does the [Fe]-hydrogenase cofactor have the same structure in all methanogens? Probably yes. The cofactor can be extracted from cell extracts and its mass subsequently determined by MALDI TOF MS. All methanogens investigated that contain [Fe]-hydrogenase also possess a cofactor with a guanylylpyridone ligand. The extracted cofactor could also always reconstitute active enzyme from apoenzyme. What is known about FeGP cofactor biosynthesis? Only that one of the two methyl groups attached to the pyridine ring is derived from methionine as revealed by mass spectrometry of the cofactor isolated from cells grown in the presence of [methyl- 13 C]-methionine (unpublished results).

3.3.

Electronic and Magnetic State of the Iron

Mössbauer spectroscopy of the 57 Fe-labelled FeGP cofactor in the absence and presence of an external magnetic field revealed that in the protein-free and in the protein-bound state the iron in the cofactor is in a low oxidation and spin state, either low spin Fe(0) or low spin Fe(II) [30]. Therefore, the oxidation state is not yet clear. In favor of Fe(0) is that in the FeGP cofactor it appears to be pentacoordinated which is characteristic for iron in Fe(0) complexes rather than for iron in low spin Fe(II) complexes, in which the iron is generally hexacoordinated. Note however, that with the methods employed (X-ray diffraction and XAS) a proton or H 2 as sixth ligand cannot be seen. In agreement with Fe(0) is also the finding that the FeGP cofactor is not very stable in water and that the stability decreases with increasing proton concentrations. Fe(0) complexes generally slowly react with protons yielding Fe(II) and H 2 . However, many other properties of the FeGP cofactor are in favor of Fe(II): (i) the cofactor is very light-sensitive which Fe(0) carbonyl complexes are generally not, although also Fe(0) carbonyls decay upon irradiation with light, (ii) The cofactor does not autoxidize in the presence of 0 2 , which Fe(0) complexes generally do. (iii) The two intrinsic CO bound to the iron in the cofactor do not exchange with extrinsic 13 CO as revealed by FTIR spectroscopy [5]. Rapid exchange of extrinsic with intrinsic CO is a property characteristic for Fe(0) CO complexes. And finally (iv) the cofactor does not give rise to the formation of H 2 upon acidification, which Fe(0) complexes generally do (unpublished results).

3.4.

Stability of the Iron Guanylylpyridinol Cofactor

The protein-free FeGP cofactor is very labile and is therefore very difficult to purify to homogeneity. Solutions have to be kept in the dark, at low Met. Ions Life Sei. 2009, 6, 219-240

226

S H I M A , T H A U E R , and E R M L E R

temperatures and at > p H 9, and have to be supplemented with mercaptoethanol or other thiol compounds for the cofactor to remain intact. But even under these conditions the FeGP cofactor has a half life of only a few days. The solutions have to be frozen a t < - 8 0 ° C to fully retain cofactor activity. The FeGP cofactor absorbs light in the UV-A/blue region (350-462 nm tested) (Figure 3A). Upon irradiation with UV-A/blue light at 4°C, the color is irreversibly bleached indicating that the cofactor has decomposed [5]. Decomposition products are an iron, a guanylylpyridone (GP) [28] (Figure 3), two CO, and most probably one mercaptoethanol (reaction 6): FeGP cofactor

UV

^ Fe + GP + 2 CO + mercaptoethanol

(6)

The light sensitivity and the decomposition products substantiate that the cofactor is an iron carbonyl which can also be deduced from the FTIR

400 Wavelenght (nm)

2000

1900 1

Wavenumber {cm" )

Figure 3. Spectra of the protein-free F e G P cofactor. (A) UY-visible spectrum of the F e G P cofactor: black line ( ε 3 6 0 « 2 η ι Μ _ 1 c m - 1 ; ε 3 0 ο ~ 5 . 5 η ι Μ _ 1 c m - 1 ) before irradiation with white light; blue line, after 30 min irradiation; red line ( ε 3 0 0 « 9 η ι Μ _ 1 c m - 1 ) after 60 min irradiation [5]. (B) Infrared absorption in the region between 2100 and 1800 c m - 1 . Reproduced f r o m Lyon et al. [29] with permission of the American Chemical Society, copyright (2004). Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

227

spectrum (Figure 3B) showing two almost equally intense bands at wave numbers 2032 and 1973, respectively, characteristic for two CO bound to iron in an angle of 90° [5]. In the absence of mercaptoethanol the cofactor decomposes within minutes already at room temperature in the dark. The stability increases with the mercaptoethanol concentration up to 1 m M indicating that the thiol compound probably competes with water in binding to the iron and that the complex with a water ligand is less stable than the one with a sulfur ligand which would not be too unusual for iron carbonyl complexes. Interesting in this respect is the facile exchange of the sulfur ligand. U p o n addition of the F e G P cofactor to apoenzyme (reaction 5) it takes only a few seconds until the reconstitution to the active enzyme is completed. At mercaptoethanol concentrations above 10 m M the rate of F e G P cofactor decomposition increases again. An explanation for the decrease in stability might be that at higher mercaptoethanol concentrations two thiol groups bind to the iron in the FeGP cofactor and thereby destabilize the complex.

4.

STRUCTURE OF [Fe]-HYDROGENASE WITH AND WITHOUT THE IRON GUANYLYLPYRIDINOL COFACTOR BOUND

Although [Fe]-hydrogenase from M. marburgensis is easy to purify in relatively large amount it was not yet possible to obtain a crystal structure of the enzyme from this organism. Only perfectly merohedrally twinned crystals were obtained that did not allow phase determination. Purification of the enzyme from other methanogens proved almost impossible because their cells do not grow well under the nickel-limiting growth conditions required to induce [Fe]-hydrogenase biosynthesis. But is was possible to heterologously produce the [Fe]-hydrogenase apoenzyme from two methanogens in E. coli and reconstitute the holoenzyme with F e G P cofactor isolated from the enzyme from M. marburgensis.

4.1.

Apo- and Holoenzyme Production

Unfortunately, the heterologously produced apoprotein from most methanogens is recovered in the inclusion body fraction, from which the native apoprotein is difficult to refold. Until now only the heterologous expression of the hmd gene from Methanocaldococcus jannaschii and Methanopyrus kandleri yielded fully soluble apoprotein, which was correctly folded as Met. Ions Life Sei. 2009, 6, 219-240

228

SHIMA, THAUER, and ERMLER

deduced from the successful reconstitution of the fully active holoenzyme upon addition of the F e G P cofactor [4]. Purification of the recombinant apoprotein from E. coli was essentially performed by heating the soluble cell fraction to a temperature of 70 °C (M. jannaschii) or 80 °C ( Μ . kandleri) and by hydrophobic interaction chromatography with a phenyl-sepharose column. The yields were ca. 40mg/10g wet cells, which - after reconstitution to the holoenzyme - was sufficient for spectroscopic and structural studies. The apoenzyme of Μ. jannaschii requires the presence of dithiothreitol to remain in solution and to retain its ability to reconstitute active enzyme with the F e G P cofactor. In the structure of the apoenzyme from M. jannaschii there are two cysteine residues, those of Cys66 and C y s l l 8 , positioned such that they can form a disulfide bridge. The two cysteines are not conserved in the M. kandleri apoenzyme, which remains stable in the absence of dithiothreitol.

4.2.

Crystal Structure Determination

Crystals suitable for X-ray structure analysis were obtained from the apoenzyme from both hyperthermophiles. The structure of the M. kandleri apoenzyme was determined by the multiple anomalous dispersion (MAD) method based on selenomethionine-labelled protein at medium resolution [31]. Subsequently, the structure of the Μ. jannaschii apoenzyme was solved at 1.75 A using the molecular replacement method for phase determination [31]. A year later the structure of the holoenzyme reconstituted from the heterologuously produced apoenzyme from M. jannaschii and F e G P cofactor was solved to 1.75 A resolution [11].

4.3.

Protein Fold

[Fe]-hydrogenase is a homodimer with dimensions of 90 A χ 50 A χ 40 A that can be subdivided into a central globular unit and two peripheral globular units which are linearly aligned. The central unit is composed of the intertwined C-terminal segments of both subunits, forming a novel intersubunit fold. The two peripheral units consist of the N-terminal domain of each subunit and are composed of an α/β structure that belongs to the Rossmann fold family. A deep cleft is formed between both units that is built up from segments of the peripheral unit of one subunit and segments of the central unit of the other subunit (Figures 4 and 5).

Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

229

Figure 4. [Fe]-hydrogenase apoenzyme (A) from Μ. jannaschii and (B) from Μ. kandleri. The active site cleft is in a closed conformation in A and in an open conformation in Β [31].

4.4.

The Binding Sites for the Iron Guanylylpyridinol Cofactor and for Methenyl-H 4 MPT +

The Rossmann fold-like structure of the N-terminal domain contains a mononucleotide binding site for the G M P moiety of the F e G P cofactor with the binding motif G C G (Gly7, Ala8, and Gly9), which is only the first part of the characteristic dinucleotide binding motif G ( X ) X G X X G reflecting that the F e G P cofactor is a phosphodiester rather than a diphosphate. The other binding site for the cofactor is Cysl76 which is located at the bottom of the intersubunit cleft. Cysl76 is absolutely essential for enzyme activity [6] and therefore considered as iron ligand. Structural studies on the holoenzyme confirmed the previously predicted binding site. The binding site for the substrate methenyl-H 4 MPT + could, so far, not be structurally characterized, but it is highly likely that it is embedded into the intersubunit cleft where sufficient space is available for the bulky molecule. In this position methenyl-H 4 MPT + can be modelled in a manner that the methenyl carbocation (C 14a ) is near the iron site such that H 2 can interact with the iron and attack C 1 4 a of methenyl-H 4 MPT + from the i?e-face [31] (Figure 1). Methenyl-H 4 MPT + binding is presumably accompanied by a large induced-fit movement which optimizes substrate binding Met. Ions Life Sei. 2009, 6, 219-240

230

S H I M A , T H A U E R , and E R M L E R

Figure 5. Electron density (in blue) of the F e G P cofactor bound to [Fe]-hydrogenase of M. jannaschii. (A) Ribbon diagram of the [Fe]-hydrogenase holoenzyme; (B) Magnification of the electron density; (C) Fit of the electron density to the F e G P cofactor [11].

Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

231

and completely shields the active site from bulk solvent. Different orientations between the central and peripheral units dependent on substrate binding is reflected in the crystal structures determined [11,31]. The cleft is in a more closed state in the M. jannaschii apoenzyme structure but in an open state in the M. jannaschii holoenzyme and M. handled apoenzyme structures [31]. The two states can be interconverted by rotation of the peripheral units relative to the central subunit by 35° (Figure 4) [31]. Furthermore, a relatively large conformational change is indicated by the finding that [Fe]-hydrogenase crystals immediately crack upon soaking with methenyl-H 4 MPT + . Also it was found that in the presence of methenylH 4 M P T + , [Fe]-hydrogenase does not crystallize under the same conditions under which the enzyme crystallizes in the absence of methenyl-H 4 MPT + .

5.

LIGANDS TO IRON IN THE ACTIVE SITE OF [Fe]HYDROGENASE

The nature of the iron ligands and their spatial arrangement were identified by IR [29], Mössbauer [30], XAS [6], and N M R [28] spectrocopic methods and by the X-ray structure of the holoenzyme [11].

5.1.

Intrinsic Ligands

In the crystal structure (Figure 5) the electron density of F e G P is well shaped except for the carboxyl group of the carboxymethyl group of the pyridine which is partially disordered [11]. Accordingly, the iron is coordinated by five or six ligands forming a slightly distorted square pyramid or an octahedral arrangement. The iron is ligated by two CO ligands originally derived from IR spectroscopic data and later also seen in the crystal structure. The third iron ligand is the nitrogen and not the hydroxyl group or the carboxylate group of the pyridone which is in its pyridinol tautomeric form as deduced from the [Fe]-hydrogenase holoenzyme structure. The guanylylpyridinol ligand, especially when its hydroxyl group is negatively charged, is predicted to have back-bonding properties similar to a cyanide ligand, which is of interest since in [FeFe]-hydrogenases and [NiFe]-hydrogenases the active site iron is decorated with CO and cyanide ligands. As previously found in the latter hydrogenases the pyridinol nitrogen and the two CO molecules of [Fe]-hydrogenase are oriented perpendicular to each other. The fourth ligation site of iron is occupied by the thiolate group of Cysl76 which was assumed from results of site-specific mutagenesis experiments and clearly verified in the crystal structure [11]. Met. Ions Life Sei. 2009, 6, 219-240

SHIMA, THAUER, and ERMLER

232

The structure of the cofactor bound in the [Fe]-hydrogenase holoenzyme differs from that proposed for the protein-free cofactor (Figure 2) in that Cysl76 from the apoenzyme rather than mercaptoethanol provides the sulfur ligand and in that the carboxymethyl group of the pyridinol is not involved in iron ligation. At the position of the carboxymethyl group in the protein-free cofactor an unknown ligand binds in the holoenzyme whose electron density is clearly connected with iron. The iron coordination is in good agreement to that calculated from XAS spectra which indicate that the iron in the reconstituted holoenzyme is tetra-/penta-coordinated (one sulfur, two CO, and one/two N/O) and in the protein-free cofactor it is pentacoordinated (one sulfur-, two CO, and two N / O ligands) [6]. In the crystal structure of the reconstituted holoenzyme the vacant sixth coordination site contains a spherical electron density interpreted as monoatomic solvent molecule that is, however, at a distance of 2.7 A too far away to be considered as a ligand [11]. This could be the binding site for a proton or H 2 which is not detectable with the methods employed (X-ray diffraction and XAS). The sixth coordination site is definitely the binding site of the competitive inhibitor CO (see below). Differences in coordination do not result in differences in the spin or oxidation state which always remains low spin Fe(II) or Fe(0) as revealed by Mössbauer spectroscopy of 57 Fe-labelled samples [30].

5.2.

Carbon Monoxide and Cyanide as Extrinsic Ligands

[Fe]-hydrogenase is reversibly inhibited by CO. The XAS spectrum of the CO inhibited enzyme reveals three CO, one sulfur and one/two N / O ligated to the iron [6]. The three CO bands show an almost equal intensity in the F T I R spectra (Figure 6) indicating that the extrinsic CO is bound perpendicular to the two intrinsic CO [5]. This finding attributes to the external CO the sixth ligation site of the iron in the holoenzyme. As CO is a competitive inhibitor for [Fe]-hydrogenase this site is presumably also the position of H 2 binding. Unfortunately, the crystal structure of the CO inhibited enzyme was almost identical to that of the uninhibited enzyme. Apparently, the extrinsic CO, which binds only very weakly and is the first to come off upon light irradiation [29], was lost during crystal structure determination. Attempts to show the binding of extrinsic CO to the iron of the proteinfree FeGP cofactor gave ambiguous results. Whereas the Mössbauer spectrum of the cofactor changed significantly when CO was added to the cofactor solution [30], the F T I R spectrum did not [29], which can presently only be explained considering that the experimental conditions were quite different. The measurements of the F T I R spectra of the protein-free cofactor in the absence and presence of extrinsic CO clearly have to be revisited. Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

233

Wavenumber (cm"1) Figure 6. Infrared spectra of the [Fe]-hydrogenase holoenzyme f r o m M. marburgensis in the absence (upper spectrum) and presence of 12 CO (lower spectrum). Reproduced from Lyon et al. [29] with permission of the American Chemical Society, copyright (2004).

Cyanide reversibly inhibits [Fe]-hydrogenase by binding to the cofactor iron as shown by Mössbauer [30], FTIR [29], and XAS [6] data, the latter exhibiting the iron as penta-coordinated (one sulfur, two CO, one cyanide, and one N/O). Cyanide is a non-competitive inhibitor for [Fe]-hydrogenase which implicates that it does not bind to the sixth ligation site. The most attractive binding site is that of the unknown ligand which is corroborated by a significantly higher electron density in the crystal structure of the cyanide-inhibited enzyme [11]. To the contrary, cyanide appears to irreversibly react with the protein-free cofactor yielding various decomposition products. However, the irreversible reaction is relatively slow and it might be Met. Ions Life Sei. 2009, 6, 219-240

SHIMA, THAUER, and ERMLER

234

speculated that the cyanide replaces the carboxylate g r o u p at the iron and thereby weakens the Fe-pyridinol linkage.

5.3.

Inactivation of [Fe]-Hydrogenase by 0 2 and Cu(l)

In cell extracts of methanogenic archaea [Fe]-hydrogenase is extremely sensitive towards 0 2 , in the presence of which the activity is rapidly lost. T o the contrary, purified [Fe]-hydrogenase is stable and catalytically active in the presence of air. This p r o p e r t y indicates t h a t in cell extracts [Fe]-hydrogenase is inactivated by a reactive 0 2 species rather t h a n 0 2 itself, most likely by the superoxide anion radical 0 2 . A n d indeed it was f o u n d that purified [Fe]-hydrogenase is rapidly inactivated in the presence of 0 2 and xanthine when also xanthine oxidase is present which catalyzes the f o r m a t i o n of 0 2 under these conditions (Figure 7). [NiFe]- and [FeFe]hydrogenases are also rapidly inactivated by 0 2 . A n o t h e r p r o p e r t y in c o m m o n is the sensitivity of [Fe]-hydrogenase towards copper ions. T h e contaminating concentrations in buffers are generally sufficient to inactivate the enzyme at low concentrations, which is the

Incubation time (min)

Figure 7. Inactivation of [Fe]-hydrogenase from M. marburgensis by superoxide anion radical. Open squares: Purified enzyme in 100mM Tris/HCl pH 8.0 at 30 °C under aerobic conditions (liquid phase equilibrated with air). Filled circles: The enzyme solution was supplemented with 0.5 mM xanthine and 15mU xanthine oxidase to generate O2. Reproduced from Buurman [51]. Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

235

reason why the buffers in contact with the enzyme have to be supplemented either with E D T A or citrate in order to complex the copper ions. The group or groups in the active site of Fe-hydrogenase, with which 0 2 and copper ions react, are not yet known. The most attractive candidate is the sulfur ligand. If so, it indicates that the sulfur is highly reactive, which may be a property exploited for catalysis.

6.

PROPOSED CATALYTIC MECHANISMS

Any mechanism has to explain why [Fe]-hydrogenase catalyzed an exchange of H 2 with protons of bulk water only in the presence of methenyl-H 4 MPT + and why for the methenyl-H 4 MPT + dependent exchange the iron in the cofactor is required. Two principally different mechanisms are presently discussed both relying on the finding that C 1 4 a of methenyl-H 4 MPT + , when bound to [Fe]-hydrogenase, has carbocation character [32-34]. In the first mechanism the iron in [Fe]-hydrogenase is assumed to have the function of a Lewis base and in the second mechanism the function of a Lewis acid [13],

6.1.

H 2 Binds First to the Carbocationic C 1 4 a of Methenyl-H 4 MPT +

Carbocations can bind H 2 either side-on or end-on. It is therefore assumed that the first step in the catalytic mechanism of [Fe]-hydrogenase is the binding of H 2 to + C 1 4 a of methenyl-H 4 MPT + [35,36] (reaction 7). The bound H 2 is positioned such that it can interact with the cofactors, iron functioning as a Lewis base. Via the interaction of H 2 with both the carbocation and the Lewis base the H - H bond is polarized such that it is heterolytically cleavaged, the hydride reacting with the carbocation and the proton with the iron (reaction 8), where the proton is in exchange with protons of bulk water (reaction 9). The activation barrier for the reaction has been calculated to be reasonable only when in the transition state H 2 binds end-on to the carbocation and the base is positioned relative to H 2 such that it can directly accept the proton. (Actually, in the calculation the base was assumed to be the amine group of lysine or the carboxyl group of an aspartate or glutamate rather than a transition metal) [37-39]. The proposed mechanism predicts that any H 2 / H + exchange catalyzed by [Fe]hydrogenase should be absolutely dependent on the presence of methenylH 4 M P T + , which is what was found. Met. Ions Life Sei. 2009, 6, 219-240

SHIMA, THAUER, and ERMLER

236

+C 14a + H 2 ^ [ H 2 - C 1 4 a ]

+

(7)

[H 2 -C 14a ] + + [Fe] ^ H - C 1 4 a + H-[Fe] +

(8)

H-[Fe] + + H 2 0 ^ [Fe] + H 3 0 +

(9)

In Eqs. (7)-(9) + C 1 4 a is used as abbreviation for methenyl-H 4 MPT + , H-C 1 4 a for methylene-H 4 MPT and [Fe] for the iron in the F e G P cofactor. Chemical precedent for such a pathway has recently been provided [40]. The proposed mechanism is consistent with the inhibition of [Fe]-hydrogenase by CO and by cyanide, whose binding to the iron site is expected to significantly affect its Lewis basicity. Competitive CO inhibition relative to H 2 has been interpreted in a manner that H 2 binds to an open coordination site rather than to a free electron pair although competitive inhibition only indicates that the binding of the competitors is mutually exclusive. The function of Fe in [Fe]-hydrogenase as Lewis base implies that in the catalytic cycle an Fe(II) hydride or an Fe(IV) hydride is formed ([H-Fe] + in reactions 8 and 9), dependent on whether the iron before protonation is in a Fe(0) or Fe(II) oxidation state [41-43]. Since mononuclear Fe(IV) hydrides are, for thermodynamic reasons, not stable, this mechanism would favor an Fe(0) complex as proton acceptor. The formation of metal hydrides in the active site of enzymes is not without precedents. A Ni(III) hydride has recently been characterized as intermediate in the catalytic cycle of [NiFe]-hydrogenase [44]. A Ni(III) hydride was also shown to be formed when the Ni(I) in methyl-coenzyme Μ reductase interacts with coenzyme Μ in the presence of coenzyme Β [45]. These findings indicate that transition metals in enzymes can have Lewis base character. Despite all these pros, the mechanism fails until now to explain why the base required has to be an iron and cannot simply be a proton accepting group of the protein.

6.2.

H 2 Binds First to the Active Site Iron Carbonyl Complex

Iron carbonyl complexes with an open coordination site form side-on (r| 2 -H 2 )Fe complexes, in which the ρΚΆ of H 2 is lowered from 35 to below 15 [46-48] and in which H 2 can be in proton exchange with bulk water [49,50]. Such a Lewis acid function is proposed for the iron in [Fe]-hydrogenase in the second mechanism [13]. In the first step H 2 binds side-on to the iron (reaction 10) but binding is only weak with the consequence that the p^Tis Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

237

lowered only somewhat and that therefore the proton exchange rates are too low to be detected by the methods employed. This has to be proposed since in the absence of methenyl-H 4 MPT + [Fe]-hydrogenase does not catalyze such an exchange. After H 2 , methenyl-H 4 MPT + binds such to the enzyme that its carbocationic C 1 4 a is juxtapositioned to the iron allowing the interaction of H 2 with both the iron and + c 1 4 a (reaction 11). The interaction would lead to a polarization of H 2 to an extent in which a bridging hydride and a bridging proton are formed between the two nucleophiles. The cationic (μ-Η2) complex thus formed is predicted to be a relatively strong acid and therefore to be in exchange with bulk water (reaction 12).

7.

[Fe] + H 2 ^ ( r i 2 - H 2 ) [ F e ]

(10)

(r| 2 -H 2 )[Fe] + + C 1 4 a ^ (μ-Η 2 ) complex +

(11)

(μ-Η 2 ) complex + + H 2 0 ^ (μ-Η) complex + H 3 0 +

(12)

CONCLUDING REMARKS

A crystal structure of [Fe]-hydrogenase with its substrate methenylH 4 M P T + bound is not yet available. It is therefore not known, whether or not the Fe of the FeGP cofactor and the C 1 4 a of methenyl-H 4 MPT + are positioned in the active site of the enzyme such that H 2 can simultaneously interact with both the iron and the carbocationic C 14a . For density function theory calculations all coordinates will have to be known. Model complexes to be constructed on the basis of the iron center of [Fe]-hydrogenase should give further insight into the essential but not yet understood function of the low spin iron with its unique ligands in H 2 activation.

ACKNOWLEDGMENTS This work was supported by the Max Planck Society and by the Fonds der Chemischen Industrie.

ABBREVIATIONS +

C14a EDTA F420

methenyl group of methenyl-H 4 MPT + ethylenediamine-N,N,N',N'-tetracetate coenzyme F 4 2 0 Met. Ions Life Sei. 2009, 6, 219-240

238 F e G P cofactor Frh FTIR GMP GP H4MPT H-C14a Hmd HPLC MAD MALDI TOF MS Mtd NMR XAS

SHIMA, THAUER, and ERMLER i r o n g u a n y l y l p y r i d o l c o f a c t o r of [Fe]-hydrogenase F 4 2 o-reducing [ N i F e ] - h y d r o g e n a s e Fourier transform infrared guanosine 5'-monophosphate guanylylpyridol tetrahydromethanopterin m e t h y l e n e g r o u p of m e t h y l e n e - H 4 M P T Hydrogen forming methylenetetrahydromethanopterin d e h y d r o g e n a s e , [Fe]-hydrogenase high p e r f o r m a n c e liquid c h r o m a t o g r a p h y m u l t i p l e a n o m a l o u s dispersion m a t r i x assisted laser d e s o r p t i o n / i o n i s a t i o n time-offlight mass spectrometry F42o-dependent methylenetetrahydromethanopterin dehydrogenase nuclear magnetic resonance X-ray absorption spectroscopy

REFERENCES 1. C. Zirngibl, R. Hedderich and R. K. Thauer, FEBS Lett., 1990, 261, 112-116. 2. C. Zirngibl, W. van Dongen, B. Schwörer, R. von Bünau, M. Richter, A. Klein and R. K. Thauer, Eur. J. Biochem., 1992, 208, 511-520. 3. R. K. Thauer, A. R. Klein and G. C. Hartmann, Chem. Rev., 1996, 96, 3031-3042. 4. G. Buurman, S. Shima and R. K. Thauer, FEBS Lett., 2000, 485, 200-204. 5. E. J. Lyon, S. Shima, G. Buurman, S. Chowdhuri, A. Batschauer, Κ. Steinbach and R. Κ. Thauer, Eur. J. Biochem., 2004, 271, 195-204. 6. M. Korbas, S. Vogt, W. Meyer-Klaucke, E. Bill, E. J. Lyon, R. K. Thauer and S. Shima, J. Biol. Chem., 2006, 281, 30804-30813. 7. R. K. Thauer, Microbiology, 1998, 144, 2377-2406. 8. J. Schleucher, C. Griesinger, Β. Schwörer and R. Κ. Thauer, Biochemistry, 1994, 33, 3986-3993. 9. J. Schleucher, B. Schwörer, R. K. Thauer and C. Griesinger, J. Am. Chem. Soc., 1995, 117, 2941-2942. 10. A. Böck, P. W. King, M. Blokesch and M. C. Posewitz, Adv. Microb. Physiol., 2006, 51, 1-71. 11. S. Shima, O. Pilak, S. Vogt, Μ. Schick, Μ. S. Stagni, W. Meyer-Klaucke, E. Warkentin, R. K. Thauer and U. Ermler, Science, 2008, 321, 572-575. 12. B. Schwörer, V. Μ. Fernandez, C. Zirngibl and R. K. Thauer, Eur. J. Biochem., 1993, 212, 255-261. 13. S. Vogt, Ε. J. Lyon, S. Shima and R. K. Thauer, J. Biol. Inorg. Chem., 2008, 13, 97-106.

Met. Ions Life Sei. 2009, 6, 219-240

CARBON MONOXIDE IN THE ACTIVE SITE OF [Fe]-HYDROGENASE

239

14. A. R. Klein and R. K. Thauer, Eur. J. Biochem., 1995, 227, 169-174. 15. A. R. Klein, G. C. H a r t m a n n and R. K. Thauer, Eur. J. Biochem., 1995, 233, 372-376. 16. G. C. Hartmann, E. Santamaria, V. M. Fernandez and R. K. Thauer, J. Biol. Inorg. Chem., 1996, 1, 446-450. 17. S. Shima and R. K. Thauer, Chem. Ree. (Jp.), 2007, 7, 37-46. 18. C. Afting, A. Hochheimer and R. K. Thauer, Arch. Microbiol., 1998, 169, 206-210.

19. C. Afting, E. Kremmer, C. Brucker, A. Hochheimer and R. K. Thauer, Arch. Microbiol., 2000, 174, 225-232. 20. Κ. Ma, C. Zirngibl, D. Linder, Κ. Ο. Stetter and R. K. Thauer, Arch. Microbiol., 1991, 156, 4 3 ^ 8 . 21. H. P. Klenk, R. A. Clayton, J. F. Tomb, O. White, Κ. E. Nelson, K. A. Ketchum, R. J. Dodson, M. Gwinn, Ε. K. Hickey, J. D. Peterson, D. L. Richardson, A. R. Kerlavage, D. E. Graham, N. C. Kyrpides, R. D. Fleischmann, J. Quackenbush, Ν. H. Lee, G. G. Sutton, S. Gill, E. F. Kirkness, B. A. Dougherty, K. McKenney, M. D. Adams, B. Loftus and J. C. Venter, Nature, 1997, 390, 364-370. 22. E. L. Hendrickson, R. Kaul, Y. Zhou, D. Bovee, P. Chapman, J. Chung, E. Conway de Macario, J. A. Dodsworth, W. Gillett, D. E. Graham, M. Hackett, A. K. Haydock, A. Kang, M. L. Land, R. Levy, T. J. Lie, T. A. Major, B. C. Moore, I. Porat, A. Palmeiri, G. Rouse, C. Saenphimmachak, D. Soil, S. Van Dien, Τ. Wang, W. Β. Whitman, Q. Xia, Y. Zhang, F. W. Larimer, Μ. V. Olson and J. A. Leigh, J. Bacterial., 2004, 186, 6956-6969. 23. D. R. Smith, L. A. Doucette-Stamm, C. Deloughery, H. Lee, J. Dubois, T. Aldredge, R. Bashirzadeh, D. Blakely, R. Cook, K. Gilbert, D. Harrison, L. Hoang, P. Keagle, W. Lumm, B. Pothier, D. Qiu, R. Spadafora, R. Vicaire, Y. Wang, J. Wierzbowski, R. Gibson, N. Jiwani, A. Caruso, D. Bush and J. N. Reeve, J. Bacterial., 1997, 179, 7135-7155. 24. B. S. Samuel, Ε. E. Hansen, J. K. Manchester, P. M. Coutinho, B. Henrissat, R. Fulton, P. Latreille, K. Kim, R. K. Wilson and J. I. Gordon, Proc. Natl. Acad. Sei. USA, 2007, 104, 10643-10648. 25. D. A. Rodionov, P. Hebbeln, Μ. S. Gelfand and Τ. Eitinger, J. Bacteriol., 2006, 188, 317-327. 26. E. R. Schreiter, S. C. Wang, D. B. Zamble and C. L. Drennan, Proc. Natl. Acad. Sei. USA, 2006, 103, 13676-13681. 27. A.-K. Kaster, Hochregulation der [Fe]-Hydrogenase-Synthese in Methanococcus maripaludis und Methanocaldococcus jannaschii bei Wachstum unter NickelMangelbedingungen., Diploma-Thesis, Philipps-Universität Marburg, 2007. 28. S. Shima, E. J. Lyon, M. Sordel-Klippert, M. Kauß, J. Kahnt, R. K. Thauer, K. Steinbach, X. Xie, L. Verdier and C. Griesinger, Angew. Chem. Int. Ed. Engl., 2004, 43, 2547-2551. 29. E. J. Lyon, S. Shima, R. Boecher, R. K. Thauer, F. W. Grevels, E. Bill, W. Roseboom and S. P. Albracht, J. Am. Chem. Soc., 2004, 126, 14239-14248. 30. S. Shima, E. J. Lyon, R. K. Thauer, B. Mienert and E. Bill, J. Am. Chem. Soc., 2005, 127, 10430-10435.

Met. Ions Life Sei. 2009, 6, 219-240

240

SHIMA, THAUER, and E R M L E R

31. O. Pilak, B. Mamat, S. Vogt, C. H. Hagemeier, R. K. Thauer, S. Shima, C. Vonrhein, Ε. Warkentin and U. Ermler, J. Mol. Biol., 2006, 358, 798-809. 32. S. Bartoschek, G. Buurman, Β. H. Geierstanger, J. Lapham and C. Griesinger, J. Am. Chem. Soc., 2003, 125, 13308-13309. 33. S. Bartoschek, G. Buurman, R. K. Thauer, Β. H. Geierstanger, J. P. Weyrauch, C. Griesinger, Μ. Nilges, Μ. Hutter and V. Helms, ChemBioChem, 2001, 2, 530-541. 34. Β. H. Geierstanger, T. Prasch, C. Griesinger, G. Hartmann, G. Buurman and R. K. Thauer, Angew. Chem. Int. Ed. Engl., 1998, 37, 3300-3303. 35. A. Berkessel and R. K. Thauer, Angew. Chem. Int. Ed. Engl., 1995, 34, 2247-2250. 36. A. Berkessel, Curr. Opin. Chem. Biol., 2001, 5, 486-490. 37. J. Cioslowski and G. Boche, Angew. Chem. Int. Ed. Engl., 1997, 36, 107-109. 38. A. P. Scott, Β. T. Golding and L. Radom, New J. Chem., 1998, 22, 1171-1173. 39. J. H. Teles, S. Brode and A. Berkessel, J. Am. Chem. Soc., 1998,120, 1345-1346. 40. G. C. Welch, R. R. San Juan, J. D. Masuda and D. W. Stephan, Science, 2006, 314, 1124-1126. 41. X. Zhao, C. Y. Chiang, M. L. Miller, Μ. V. Rampersad and Μ. Y. Darensbourg, J. Am. Chem. Soc., 2003, 125, 518-524. 42. A. Kayal and Τ. B. Rauchfuss, Inorg. Chem., 2003, 42, 5046-5048. 43. E. J. Daida and J. C. Peters, Inorg. Chem., 2004, 43, 7474-7485. 44. M. Brecht, Μ. van Gastel, T. Buhrke, B. Friedrich and W. Lubitz, J. Am. Chem. Soc., 2003, 125, 13075-13083. 45. J. Harmer, C. Finazzo, R. Piskorski, S. Ebner, Ε. C. Duin, M. Goenrich, R. Thauer, M. Reiher, Α. Schweiger, D. Hinderberger and B. Jaun, J. Am. Chem. Soc., 2008, 130, 10907-10920. 46. D. M. Heinekey and W. J. Oldham, Chem. Rev., 1993, 93, 913-926. 47. G. J. Kubas, Science, 2006, 314, 1096-1097. 48. G. J. Kubas, Catalysis Lett., 2005, 104, 79-101. 49. J. W. Tye, M. B. Hall, I. P. Georgakaki and Μ. Y. Darensbourg, Synergy between Theory and Experiment as Applied to HjD Exchange Activity Assays in Fe H2ase Active Site Models, in Advances in Inorganic Chemistry-Including Bioinorganic Studies, Ed. R. van Eldik, Elsevier, Academic Press, San Diego, Vol. 56, 2004, 1-26. 50. J. W. Tye, Μ. Y. Darensbourg and Μ. B. Hall, Inorg. Chem., 2006, 45, 1552-1559. 51. G. Buurman, Zum Katalysemechanismus der H 2 -bildenden N 5 ,N 1 0 -Methylentetrahydromethanopterin-Dehydrogenase (Hmd) aus methanogenen Archaea: Untersuchung zur Stereospezifität und Nachweis einer prosthetischen Gruppe., PhD Thesis, Philipps-Universität Marburg, 2001.

Met. Ions Life Sei. 2009, 6, 219-240

Met. Ions Life Sei. 2009, 6, 241-293

8 The Dual Role of Heme as Cofactor and Substrate in the Biosynthesis of Carbon Monoxide Mario Rivera and Juan C. Rodriguez Ralph N. Adams Institute for Bioanalytical Chemistry, Department of Chemistry, University of Kansas, Multidisciplinary Research Building, 2030 Becker Dr., Lawrence, KS 66047, USA

ABSTRACT 1. I N T R O D U C T I O N 1.1. C a r b o n Monoxide: Properties and E n v i r o n m e n t a l Sources 1.2. The T w o Faces of C a r b o n Monoxide: Toxic and Cytoprotective Effects 1.3. H e m e Oxygenase Is a U b i q u i t o u s Enzyme 2. T H E B I O S Y N T H E S I S O F C A R B O N M O N O X I D E 2.1. H e m e B r e a k d o w n and C a r b o n M o n o x i d e Release. Overview of the H e m e Oxygenase Catalytic Cycle 2.2. The Structure of H e m e Oxygenase 2.3. F o r m a t i o n of H y d r o p e r o x i d e at the Catalytic Center of H e m e Oxygenase 2.4. A Conserved N e t w o r k of H y d r o g e n Bonded W a t e r s Facilitates F o r m a t i o n of the Ferric H y d r o p e r o x i d e Intermediate 2.5. Oxidation of M e s o - H y d r o x y h e m e to V e r d o h e m e and the Release of C a r b o n M o n o x i d e

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00241

242 243 243 243 245 247 247 248 251

253 256

242

R I V E R A and R O D R I G U E Z

3. HEME OXYGENASE FAVORS HEME HYDROXYLATION OVER FERRYL FORMATION. THE NATURE OF THE FERRIC HYDROPEROXIDE COMPLEX IN HEME OXYGENASE 3.1. Studies with Model Heme Complexes 3.2. Studies with the Hydroxide Complex of pa-Heme Oxygenase (/M-HO-OH) 3.3. Studies with the Azide Complex of pa-HO ( / M - H O - N 3 ) 3.4. A Hydrogen Bond from Glyl25 N-H to Coordinated Azide-N May Promote the Unusual Electronic Structure of the />a-HO-N3 3.5. Implications to the Mechanism of Heme Oxidation by Heme Oxygenase 4. HEME OXYGENASE DYNAMICS AND HEME BREAKDOWN. THE DISTAL LIGAND HAS A PROFOUND EFFECT IN THE DYNAMIC BEHAVIOR OF HEME OXYGENASE 4.1. H/D Exchange 4.2. Microsecond-Millisecond Dynamics 4.3. A Unifying View of Protein Dynamics and Heme Oxygenase Reactivity 5. THE REGIOSELECTIVITY OF HEME HYDROXYLATION 5.1. Hydroxylation of the α-Meso Carbon Leads to Its Release as CO with Subsequent Formation of a-Biliverdin 5.2. pa-Heme Oxygenase Exhibits Unique δ-Regioselectivity 5.3. Polypeptide-Heme Interactions Control the Regioselectivity of Heme Oxidation 5.4. The 'II NMR Spectra of Cyanide-Inhibited Heme Oxygenase as a Diagnostic Tool of Heme Oxidation Regioselectivity 6. CONCLUSION AND OUTLOOK ACKNOWLEDGMENTS ABBREVIATIONS REFERENCES

259 259 260 263

265 267

268 269 271 274 276 276 277 279

282 284 285 286 286

ABSTRACT: Carbon monoxide (CO) is a ubiquitous molecule in the atmosphere. The metabolism of mammalian, plastidic, and bacterial cells also produces CO as a byproduct of the catalytic cycle of heme degradation carried out by the enzyme heme oxygenase (HO). The biological role of CO spans the range from toxic to cytoprotective, depending on concentration. CO generated by the catalytic activity of HO is now known to function in several important physiological processes, including vasodilation, apoptosis, inflammation, and possibly neurotransmission. Consequently, understanding the details of the reaction that leads to the formation of this important gaseous molecule from heme has become an important aspect in the study of the chemistry and Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

243

biochemistry of HO, which utilizes heme in the dual capacity of substrate and cofactor. In this chapter, a summary, and when appropriate, discussion of the current understanding of the structural, dynamical, and reactive properties that allow HO to breakdown heme into iron, biliverdin, and CO is presented. KEYWORDS: carbon monoxide · enzyme dynamics · ferric hydroperoxide · heme degradation · heme oxygenase · N M R spectroscopy of heme · oxygen activation · protein dynamics

1. 1.1.

INTRODUCTION Carbon Monoxide: Properties and Environmental Sources

Carbon monoxide is a colorless and odorless gas under atmospheric temperature and pressure, with a melting point of —205 °C, a boiling point of - 1 9 1 . 5 °C and a density of 1.250 g/L at 0 ° C [1], CO is a ubiquitous atmospheric component because it is generated from the incomplete burning of biomass (wild fires), it is released from the earth's mantle, where it is dissolved in molten rock, by volcanic eruptions, and is produced by oxidation of methane emitted from wetlands and ruminants, and from the oxidation of higher hydrocarbons emitted by vegetation [2]. CO in the atmosphere is also a byproduct of human activity because the molecule is also produced from the incomplete combustion of fuels in households and in the now omnipresent internal combustion engine. As of 2001, the combined emission of CO is estimated to be 2780 teragrams per year; approximately 50% of the total is thought to be contributed by human endeavors and the remainder by the natural sources mentioned above [3,4]. In addition, cellular metabolism in mammals, plants and bacteria produce CO, mainly from the degradation of heme by the enzyme heme oxygenase [5-8].

1.2.

The Two Faces of Carbon Monoxide: Toxic and Cytoprotective Effects

Similar to several other molecules, the effect of CO on living organisms can range from toxic to essential, depending on dose. The toxicity and lethality of CO at elevated concentrations has been recognized as early as the mid 1800s, when it was identified as a poisoning agent in coal gas. Since then a very large number of confirmed cases involving CO poisoning have been documented [9]. The largest number of unintentional fatalities is associated with automobile exhaust, followed by the burning of wood and coal, and the use of kerosene stoves in improperly ventilated spaces [2]. Clinical signs of Met. Ions Life Sei. 2009, 6, 241-293

244

R I V E R A and R O D R I G U E Z

acute CO toxicity are not specific but include dizziness, headache, weakness, confusion, abdominal pain, and muscle cramping. Several of these symptoms are consistent with hypoxia, which is a consequence of the fact that carbon monoxide competes effectively with molecular oxygen ( 0 2 ) for binding to hemoglobin (Hb), thus forming carboxyhemoglobin (Hb-CO). The binding affinity of hemoglobin for CO is approximately 200 times larger than the affinity for 0 2 [10], which allows a gradual outcompeting of 0 2 for one or more of the four heme molecules in the tetrameric hemoglobin. Partial saturation of hemoglobin with CO has important implications on its affinity for 0 2 , a fact that is manifested in a left shift of the oxygen-hemoglobin dissociation curve, which is left shifted and is accompanied by a change in shape from sigmoidal to hyperbolic. These changes in the dissociation curve indicate that partial saturation of H b with CO impair the release of 0 2 at the tissue level, thus ensuing hypoxia [11]. It is important to note, however, that CO binding to hemoglobin does not account for all the pathophysiologic consequences of CO-induced toxicity, and that carbomonoxy-hemoglobin levels can be valuable in confirming CO exposure, although these levels alone cannot be used to assess the severity of the exposure or to guide a treatment plan [12]. Interestingly, animals transfused with blood containing highly saturated Hb-CO but minimal free CO did not show clinical symptoms, an observation that has led to the suggestion that a small fraction of CO dissolved in plasma may play an important role in mediating CO toxicity [9,13]. In this context, it is interesting that approximately 15% of the total CO absorbed is bound to extravascular heme proteins and enzymes such as myoglobin, cytochrome P450, cytochrome c oxidase, soluble guanylate cyclase, inducible nitric oxide synthase, and the heme-heme oxygenase complex. Taken together, these observations led to the idea that CO toxicity is a combination of tissue hypoxia and direct CO-mediated cellular damage [9]. In fact, although a complete understanding is still lacking, there are some clues indicating that cellular damage may be mediated by interactions between these heme proteins and CO. For instance, it has been shown that the activity of cytochrome c oxidase (complex IV of the mitochondrial respiratory chain) is partially inhibited upon exposure to CO, likely due to the binding of CO to the heme a 3 [14] and binding of CO to the distal site of cytochromes P450 inhibits the activity of these widespread and important enzymes [15]. In comparison to its toxic and lethal side, the beneficial properties of CO have been recognized very recently. In fact, this has become an area of intense investigation, where new observations and plausible mechanisms of action span the range from fundamental studies to possible clinical applications. The interested reader is referred to exhaustive reviews addressing the role of CO in biology and medicine [16-19]. Carbon Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

245

monoxide in m a m m a l s is generated by the enzymatic activity of at least two isoforms of the enzyme heme oxygenase (HO), the inducible H O - 1 and the constitutively expressed HO-2; the existence of a third isoform, H O - 3 remains controversial [19]. Although, the endogenous synthesis of c a r b o n monoxide h a d been suspected for a long time on the basis that patients with hemolytic anemia exhaled air with relatively high concentrations of C O , the biochemical basis for C O biosynthesis was n o t set in place until 1968, when T e n h u n e n , M a r v e r , and Schmid established the existence of a previously undescribed microsomal enzyme capable of degrading heme to biliverdin [20]. In the subsequent year these investigators established that this enzyme, n o w k n o w n as heme oxygenase, uses N A D P H and 0 2 to oxidize heme to biliverdin, with the c o n c o m i t a n t release of C O and iron [5]. In vitro and in vivo studies have shown t h a t C O generated endogenously by the activity of H O plays a role in a n u m b e r of physiological processes, including vasodilation, inhibition of s m o o t h muscle cell proliferation, inhibition of apoptosis, and possibly in neurotransmission. These cytoprotective properties of C O have been reviewed exhaustively [16-19]. Endogenously generated C O also appears to have an a n t i i n f l a m m a t o r y effect, although its role independent of the participation of other byproducts of heme degradation (biliverdin and up-regulated ferritin), is n o t yet clear. The possibility that C O generated by induction of H O - 1 m a y exert a n t i i n f l a m m a t o r y effect has been mirrored in the a n t i i n f l a m m a t o r y response induced by exogenous CO, which was released f r o m administered methylene chloride u p o n enzymatic d e g r a d a t i o n [21]. These findings, which include protection of grafts in animal studies of organ t r a n s p l a n t a t i o n [21,22], stimulated some investigators to explore the idea of synthesizing c o m p o u n d s capable of delivering C O to tissues, with the expectation of using t h e m as potential a n t i i n f l a m m a t o r y drugs [23-25],

1.3.

Heme Oxygenase Is a Ubiquitous Enzyme

The degradation of heme is catalyzed by the enzyme heme oxygenase [5], which carries out a complicated set of reactions that result in opening of the heme macrocycle, release of the heme iron, and p r o d u c t i o n of biliverdin and c a r b o n m o n o x i d e (Figure 1). H O - 1 has been the most actively investigated isoform because it is induced by n u m e r o u s stimuli such as heme, metals, h o r m o n e s , and oxidizing agents [18]. H O - 2 is a constitutively synthesized enzyme present in highest concentration in the testes and brain [18]. N o t long ago, the catalytic activity of H O was regarded only in the context of the maintenance of cellular heme homeostasis as a catabolic enzyme, and the products of H O activity were t h o u g h t of as toxic waste material. This view changed drastically once it became a p p a r e n t t h a t the products of heme Met. Ions Life Sei. 2009, 6, 241-293

246

R I V E R A and R O D R I G U E Z

Figure 1. Schematic representation of the heme oxidation path, leading to the formation of CO, iron, and biliverdin.

degradation play important biological functions. In addition to the now recognized importance of CO in biology and medicine (see above), the other two products of H O activity are known to have important physiological roles. Thus, heme breakdown by H O is crucial for the recycling of iron because only 1-3% of the daily iron requirement is obtained from dietary intakes [26] and biliverdin and bilirubin are powerful antioxidants [27-29]. Heme oxygenase has also been identified in some pathogenic bacteria. In this context, it is interesting that a primary obstacle for successful bacterial colonization of a mammalian host (infection) is the lack of available iron because the concentration of free iron in mammals is maintained at a very low level, ~ 10~9 Μ [30]. Some bacterial pathogens are capable of utilizing heme as a sole source of iron, suggesting that bacterial HOs are integral part of a pathway to mine iron from host heme. Nevertheless, a relatively small number of bacterial heme-degrading enzymes have been identified so far [6,31-35]. The biochemical and biophysical characterization of these enzymes has provided an additional avenue to explore the mechanism of heme catabolism at the molecular level. Heme oxygenase also plays an important role in plants where it facilitates the synthesis of phycobilins and open-chain tetrapyrroles, which are produced from the final product of heme degradation, biliverdin [7,8]. Heme oxygenase enzymes have been found to be essential in the cyanobacterium Synechocystis sp. PCC6803, the red algae Cyanidium caldarium [36], and in the higher plant Arabidopsis thaliana [37], thus H O seems to be present ubiquitously in the plant kingdom. A recent study reported the bacterial expression and spectroscopic characterization of heme oxygenase from Synechocystis sp. PCC6803 [38] (Syn HO-1) and concluded that the heme binding site of this enzyme is likely to be more similar to that of the bacterial Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

247

cd-HO than to the mammalian enzymes. Future detailed characterization of plant HOs is also likely to prove fruitful in the quest to understand the mechanism of heme degradation. There are several excellent reviews summarizing the rich chemistry and biochemistry [39-43] and the biology [18,28,44,45] of heme oxygenase enzymes, thus this chapter is in no way a comprehensive recapitulation of the immense field of heme oxygenase research. Rather, it will focus on the structural, dynamic and reactive properties inherent in heme oxygenase that allow this enzyme to channel dioxygen activation toward heme oxidation, which leads to the biosynthesis of CO, with concomitant production of iron and the antioxidant biliverdin.

2. 2.1.

THE BIOSYNTHESIS OF CARBON MONOXIDE Heme Breakdown and Carbon Monoxide Release. Overview of the Heme Oxygenase Catalytic Cycle

Heme oxygenase is unusual in that it utilizes heme in a dual role of substrate and cofactor. It is now clear that the catalytic cycle of H O (Figure 1) parallels that of cytochrome P450 in that the ferric enzyme is reduced to its ferrous state, with subsequent formation of an oxyferrous complex ( F e n - 0 2 ) that accepts a second electron and is thereby transformed into an activated ferric hydroperoxy ( F e m - O O H ) oxidizing species [46]. Thereafter the mechanisms of heme hydroxylation and monooxygenation diverge, as shown schematically in Figure 2. Whereas the F e m - O O H oxidizing species in monooxygenases decays to an oxoferryl species ( F e v = 0 ) , the F e m - O O H intermediate in H O reacts with the heme macrocycle to give a-meso hydroxyheme. The latter undergoes a subsequent 0 2 -dependent elimination of the hydroxylated α-meso carbon as CO with the concomitant formation of verdoheme (see Figure 1). Verdoheme is subsequently oxidized to Fe m -biliverdin in a reaction that requires both 0 2 and reducing equivalents [46,47]. In mammals, Fe m -biliverdin must be reduced to Fe n -biliverdin previous to the sequential release of Fe 11 and biliverdin [48]. In total, 7 electrons and 3 molecules of 0 2 are needed to oxidize heme to biliverdin. Cytochrome P450 reductase (CPR) supports the catalytic activity of HO-1 by donating all 7 electrons to mammalian HO-1 enzymes [42,48]. Electron transfer from CPR to H O is preceded by the formation of a ternary complex involving CPR, N A D P H and HO; the binding interface between HO-1 and CPR has been partially delineated using results obtained from fluorescence resonance energy transfer (FRET) [49] and surface plasmon resonance (SPR) studies [50]. In the case of bacterial HOs, it has recently been Met. Ions Life Sei. 2009, 6, 241-293

248

RIVERA and RODRIGUEZ

Figure 2. The obligatory Fe m -OOH intermediate is common to monooxygenase (cyt P450) and peroxidase enzymes, in which it decays to a formally Fe v oxo oxidizing species. The Fe m -OOH intermediate is also obligatory in heme oxygenase where it reacts with the heme to form meso-hydroxyheme.

established that the electron donor to H O from P. aeruginosa (pa-HO) is an NADPH-dependent ferredoxin reductase, /w-FPR [51]. The genes coding for pa-HO and /?A-FPR are upregulated when P. aeruginosa is challenged with low iron concentrations [52], an observation that underscores the synergistic function of these enzymes in the release of iron from heme [51].

2.2.

The Structure of Heme Oxygenase

The crystal structure of human HO-1 (h-HO-1), first reported in 1999 [53], catalyzed an intense spark of activity that allowed relatively fast progress toward the fundamental understanding of H O chemistry and biophysics. This seminal report was rapidly followed by the elucidation of the X-ray crystal structure of rat heme oxygenase-1 (r-HO-1), which is 84% identical to h-HO-1 [54]. More recently, the crystal structures of heme oxygenase enzymes from pathogenic bacteria such as Neisseria meningitidis (nm-HO) [55], Corynebacterium diphtheriae (CÖ?-HO) [56] and Pseudomonas aeruginosa (pa-HO) [57] have been solved. Despite the low sequence identity of nm-HO (22%) [32,58], cd-HO (33%) [31], and pa-HO (19%) [33] relative to HO-1 the bacterial H O enzymes share the fold characteristic of the mammalian enzymes. The H O fold is highly α-helical (8 helices) and harbors the heme between helices A and F, as shown in Figure 3. This fold is also shared Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

249

by the constitutively expressed mammalian enzyme HO-2 [59,60]. H u m a n HO-2 is one of several H O enzymes whose structure has been solved in its substrate-bound and apo forms. In all cases, the overall fold of the holo-enzyme is maintained in the absence of heme [59,61]. A common feature among all known holo-HO structures is the presence of several water molecules encased between helix C, E, and F at the core of the protein. These structural water molecules, which are detected in X-ray diffraction maps, can also be inferred from their N M R spectroscopic characteristics in solution [62], and constitute an integral part of a hydrogen bonding network that is known to play essential catalytic roles in HO. Site-directed mutagenesis and spectroscopic studies conducted before the structure of HO-1 was available suggested that the heme environment in the heme-HO complex is similar to that of myoglobin in that its heme is coordinated by a proximal, non-ionized His residue (yellow in Figure 3) and by a distal H 2 0 or OH~ ligand [63-65]. The advent of the crystal structure of HO-1 [53] revealed the absence of polar residues in the distal binding site, and thus the absence of a distal residue. In fact, as can be seen in the view of Figure 4, the backbone of a section of helix F is in close contact with the heme, a structural motif that is unusual in heme proteins. Consequently, the

Figure 3. X-ray crystal structure of h-HO-1 (PDB ID 1N45) showing the α-helical fold characteristic of HO enzymes. The heme (red) is sandwiched between helices A and F and is coordinated by a proximal histidine (yellow). Met. Ions Life Sei. 2009, 6, 241-293

250

R I V E R A and R O D R I G U E Z

Figure 4. Blow-out of helices A and F in ρα-ΆΟ to illustrate that the distal helix (F) above the heme does not contain polar side chains. It is also apparent that the backbone of helix F makes contact with the face of the heme. This is unusual in heme proteins but characteristic of H O enzymes of known structure.

only polar groups near the heme-iron that can stabilize the distal H 2 0 or 0 2 ligands are the carbonyl oxygen of Glyl39 and the amine nitrogen of Glyl43 [53]. The crystal structure of rat HO-1 [54] also showed that Glyl39 and Glyl43 are near the heme iron and likely contribute to stabilizing the distal H 2 0 or 0 2 ligands. The authors of this study also suggested that the N - H group of Glyl43 may form a hydrogen bond to the coordinated oxygen atom of the hydroperoxide molecule in the F e m - O O H oxidizing species, thus assisting in the catalysis of heme hydroxylation [54]. We will return to this issue latter in this chapter. The fold of bacterial HOs is almost identical to that of their mammalian counterparts. The proximal side of the heme pocket in the bacterial enzymes also harbors a non-ionized His ligand [55]. Interestingly, but perhaps not surprisingly given the low sequence similarity, the amino acid identity among residues comprising the distal helices of mammalian and bacterial H O enzymes is low. This is manifested in several structural differences in the distal helix, where there is no conservation of acidic and basic side chains, which would appear to eliminate the possibility of a conserved role in sustaining catalysis for these types of residues [55]. Nevertheless, it is noteworthy that G l y l l 6 and Glyl20 in nm-HO, Glyl35 and Glyl39 in ctf-HO, and Glyl21 and Glyl25 in ^ - H O take the place of Glyl39 and Glyl43 in the distal pocket of HO-1. It has therefore been suggested that the distal helix kink and flexibility imparted by these two Gly residues is an important Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

251

c o m p o n e n t of H O f u n c t i o n [55]. Hence, a l t h o u g h the distal helix of H O enzymes can tolerate significant differences in a m i n o acid composition and structure, the presence of G l y l 3 9 and G l y l 4 3 (h-HO-1), or their equivalent in distinct heme oxygenase enzymes, is necessary to s u p p o r t heme hydroxylation. Consistent with this idea, when G l y l 3 9 or G l y l 4 3 in H O - 1 is replaced, the F e m - O O H intermediate in the m u t a n t enzymes does n o t attack the heme to f o r m meso-hydroxyheme but rather decays into a ferryl species [66]. Recent reports suggest t h a t the p r o d u c t of the ChuS gene in pathogenic E. coli 0157:H7 is a heme oxygenase possessing a fold distinct f r o m t h a t displayed by all H O s of k n o w n structure [67,68]. In contrast to the classical α-helical fold of H O , the structure of C h u S consists of a central core m a d e u p of two ß-sheets, each consisting of nine antiparallel ß-strands flanked by a pair of parallel ß-helices. It is interesting that a l t h o u g h the structure of C h u S consists of two identical t a n d e m repeats, the structure of the heme complex shows t h a t heme binds only at the C-terminal motif. Clearly, additional work is needed to establish whether the C h u S fold is mirrored in H O enzymes f r o m other bacterial strains.

2.3.

Formation of Hydroperoxide at the Catalytic Center of Heme Oxygenase

The activation of 0 2 by H O starts with the one-electron reduction of the h e m e - H O complex, followed by the binding of dioxygen to f o r m an oxyferrous complex ( F e n - 0 2 ) [46], as shown schematically in Figure 1. The efficient utilization of 0 2 depends in p a r t on the f o r m a t i o n of a stable F e n - 0 2 complex because unstable complexes are n o t only expected to favor the dissociation of 0 2 but would also be susceptible to autoxidation, thus lowering the efficiency of 0 2 utilization. Indeed, it has been shown that the affinity of ferrous H O - 1 and H O - 2 for 0 2 is 30 μ Μ " 1 and 80 μ Μ " 1 , respectively, which is 30-90-fold greater t h a n t h a t exhibited by m a m m a l i a n myoglobins [69]. T h e enhanced stability of the oxyferrous complex in H O is manifested in relatively high 0 2 association rate constants (similar to those exhibited by myoglobin) and in significantly slower 0 2 dissociation rates, which are notably slower t h a n those measured for myoglobin. T h e relatively high affinity for 0 2 also imparts H O with a relative immunity t o w a r d inhibition by the C O f o r m e d during heme oxidation. This point will be addressed in m o r e detail in Section 2.5. Early w o r k with heme oxygenase established t h a t the oxidation of heme to biliverdin requires t h a t the F e n - 0 2 complex accepts a second reducing equivalent [46]. This i n f o r m a t i o n was interpreted to suggest t h a t the Met. Ions Life Sei. 2009, 6, 241-293

252

R I V E R A and R O D R I G U E Z

reduction of the coordinated 0 2 molecule in the oxyferrous complex of H O leads to the formation of a coordinated peroxide, which acts as the heme hydroxylating species [70]. The same investigators demonstrated that the addition of limited amounts of H 2 0 2 to a solution of rat HO-1 results in the oxidation of the heme to verdoheme. Interestingly, substituting H 2 0 2 by alkyl hydroperoxides did not result in heme hydroxylation or the formation of verdoheme. Instead, it was observed that acyl- and alkylhydroperoxides promote the formation of a ferryl intermediate [70]. The results obtained upon reacting H 2 0 2 with H O stand in stark contrast with those obtained upon reacting H 2 0 2 or alkylhydroperoxides with ferric porphyrinates [71] or with heme proteins such as peroxidases [72] and globins [73], which produce a ferryl complex, regardless of the nature of the oxidant i.e., H 2 0 2 or R O O H . It was therefore concluded that the nature of the species that oxidizes the HO-bound heme to α-meso-hydroxyheme is a ferric hydroperoxide ( F e m - O O H ) [70]. The oxidation of heme to verdoheme upon addition of limited amounts of H 2 0 2 to a solution of HO-1 was subsequently observed with HO-2 [74] and with the bacterial heme oxygenases cd-HO [75] and pa-HO [76,77], thus strongly supporting the notion that hydroxylation of the heme-HO complex is fully supported by a ferric hydroperoxide ( F e m - O O H ) intermediate. Spectroscopic evidence supporting the need for a one-electron reduction of the oxyferrous complex to form the F e m - O O H , which in turn hydroxylates the heme in HO, was obtained from cryogenic E N D O R (electron nuclear double resonance) spectroscopic studies. These experiments showed that upon radiolytic reduction of a frozen solution (77 K) of the F e n - 0 2 complex of HO, a low-spin ferriheme complex ( F e m - 0 0 ~ ) is formed, which rapidly accepts a proton to form the F e m - O O H species [78]. The latter exhibits a rhombic EPR (electron paramagnetic resonance) spectrum characteristic of low-spin ferrihemes and is stable at 77 K. U p o n increasing the temperature to 218 Κ for a short period of time and cooling back to 77 Κ the EPR spectrum of the F e m - O O H intermediate is replaced by the spectrum of the α-meso-hydroxyheme complex of H O [78]. These findings confirmed the idea that a ferric hydroperoxide intermediate is a precursor of α-hydroxyheme and reinforce the notion that the oxidation of heme by a ferric hydroperoxide species is a novel reaction performed by heme proteins [39,70]. It is interesting to appreciate that theoretical studies have led to the suggestion that compound I, [Fe I V =0], is formed by O-O bond heterolysis of the F e m - O O H intermediate, followed by attack of the released H 2 0 to form meso-hydroxyheme [79]. Experimental studies aimed at testing this hypothesis, however, demonstrate that compound I in HO-1 does not hydroxylate its heme meso carbon, thus apparently ruling out the involvement of compound I in heme catabolism [80]. Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

2.4.

253

A Conserved Network of Hydrogen Bonded Waters Facilitates Formation of the Ferric Hydroperoxide Intermediate

Heme oxygenase and monooxygenase enzymes share a similar mechanism of 0 2 activation which leads to the formation of a ferric hydroperoxide intermediate. However, once formed, the F e m - O O H intermediate can follow two divergent paths (schematically illustrated in Figure 2): (a) the formation of an oxoferryl species, thus avoiding the hydroxylation of heme, or (b) the efficient reaction with the porphyrin macrocycle to produce mesohydroxyheme. The existence of these two competing pathways implies that heme proteins whose primary function is that of heme catabolism favor (accelerate) the reaction between the ferric hydroperoxide intermediate and the heme macrocycle, whereas monooxygenase enzymes accelerate the decay of the F e m - O O H complex into a ferryl species. The remainder of this section will be devoted to discuss the current state of understanding regarding the structure-function relationships that in H O result in heme hydroxylation rather than the formation of an oxoferryl. As has been pointed out earlier, the distal pocket structure in H O is significantly different from the equivalent structures in monooxygenase and peroxidase enzymes. An important signature of the distal pocket in H O is the absence of a polar side chain capable of stabilizing a coordinated H 2 0 or 0 2 ligand. Stabilization of these ligands is likely accomplished by the presence of Glyl43 in HO-1, or equivalent Gly in other H O enzymes, which is thought to donate a hydrogen bond to the coordinated sixth ligand [53,81]. The crystal structure of HO-1 showed that the distal helix also exhibits several side chains that although not positioned to interact directly with the distal ligand, can interact with the latter via bridging hydrogen bonding water molecules [53-54,82]. When one of these residues (Aspl40) was mutated to Ala, His or Phe, the mutants exhibited less than 3% of the biliverdin forming activity characteristic of wild-type HO-1 [82]. Moreover, these mutants were found to exhibit peroxidase activity when incubated with H 2 0 2 and guaiacol as substrates. Similar observations where made with Asp 140 mutants of rat HO-1 [83], The tantalizing observations made with Asp 140 mutants of h-HO-1 and r-HO-1 led to additional investigations by E N D O R spectroscopy [78,84,85]: Radiolytic cryoreduction (77 K) of the F e n - 0 2 results in the formation of a F e m - O O H species, as is the case when the F e n - 0 2 complex of cytochrome P450 c a m is radiolytically reduced at cryogenic temperatures [86]. In contrast, radiolytic cryoreduction of the F e n - 0 2 complex of myoglobin (not an oxygen activating heme protein) leads to the formation of a ferric peroxo ( F e m - 0 0 ~ ) complex, which is converted to a F e m - O O H species only when the sample is annealed to temperatures above 180 Κ [86,87]. These Met. Ions Life Sei. 2009, 6, 241-293

254

R I V E R A and R O D R I G U E Z

Figure 5. Schematic representation of hydrogen bonding networks in (a) cytochrome P450 c a m and (b) h-HO-1.

differences in the readiness with which the F e m - 0 0 ~ complex accepts a proton have been taken [84] to support the idea that 0 2 -activating enzymes harbor a network of hydrogen bonds that is utilized to deliver a proton to the nascent F e m - 0 0 ~ complex [88,89]. Along this vein, it is interesting to consider that the F e m - O O H complex of HO does not hydroxylate the heme at 77 K. Instead, exposing this intermediate to progressively longer periods of time at 200 K, followed by cooling back to 77 K, results in slight changes in the EPR spectrum that have been attributed to rearrangements of the hydrogen bonding network in the distal pocket that lead to a reactive F e m - O O H intermediate, denoted as R [84], The 1 f I E N D O R spectrum of R revealed two signals, one originating from the proton in the F e m - O O H moiety, denoted as H I (see Figure 5b) and a second signal, also attributed to an exchangeable hydrogen, which has been denoted as H2. At 200 Κ the EPR spectrum of R and the signals of H I and H2 in the E N D O R spectrum rapidly loose intensity, with the concomitant increase in intensity of a signal that corresponds to α-meso-hydroxyheme. In contrast, the EPR spectrum obtained upon radiolytic cryoreduction of the F e n - 0 2 complex of the Aspl40 mutants of r-HO-1 is consistent with the formation of a F e m - 0 0 ~ complex. Protonation of the latter to form the F e m - O O H intermediate occurs only after the sample is annealed to temperatures above 180 K. In addition, and distinct from the observations made with the wild-type Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON

MONOXIDE

255

enzyme, the F e m - O O H intermediate in the Asp 140 mutants is not converted to meso-hydroxyheme when annealed above 200 K; rather, it forms an EPR-silent species proposed to be the oxoferryl species [84], whose formation was inferred from electronic absorption spectroscopic studies of similar Asp 140 mutants [82]. These observations clearly implicate Asp 140 as an important component of the network of hydrogen bonds in the distal pocket of human and rat HO-1 that relays a hydrogen ion to the nascent F e m - 0 0 " moiety. It is noteworthy that X-ray crystallographic studies conducted with the N O and CO complexes of H M - H O revealed that A s n l l 8 takes the place of the catalytically essential Asp 140 in mammalian HO-1 [90]. It was also found that the side chain of Arg77 moves upon ligand binding and is likely to participate in the network of hydrogen bonds that deliver a proton to the nascent F e m - 0 0 ~ . Hence, it was concluded that the precise identity of the side chains involved in the hydrogen-bonding network of the distal pocket appears not to be important. On the other hand, conservation of a strong network of hydrogen bonds in the distal pocket of H O appears to be an important structural motif shared by all known HOs [90]. In this hydrogen bonding network Asp 140 forms a hydrogen bond to a water molecule, which in turn forms a hydrogen bond with the terminal oxygen of the nascent F e m - 0 0 ~ complex [83], as illustrated schematically in Figure 5b. The acidity of the water molecule donating the hydrogen bond (HI in Figure 5b) to the nascent ferric hydroperoxo complex is thought to be diminished by virtue of being simultaneously hydrogen-bound to Asp 140. Accordingly, the coordinated peroxide would not be fully protonated and would exhibit a decreased tendency to cleave its O-O bond, thereby allowing sufficient time for the heme hydroxylating reaction to take place [83]. In comparison, the terminal oxygen in the F e m - 0 0 ~ intermediate of cytochrome P450 c a m is hydrogen-bonded with the side chain of Thr252 and a water molecule that interacts with the amido group of Asp251 [91], as illustrated schematically in Figure 5a. In this set of hydrogen bonds the interaction between the N - H group of Asp251 and the water molecule hydrogen-bonded to the terminal oxygen in the F e m - 0 0 ~ moiety render this water molecule relatively acidic. Since theoretical investigations suggest that protonation of the F e m - 0 0 ~ species results in significant weakening of the O-O bond [92,93], the hydrogen bonds to the terminal oxygen of F c " ' - 0 0 are expected to facilitate cleavage of the O-O bond. The crystal structures of HO-1 coordinated by azide [81] and those obtained from the D140A mutant of human HO-1 in its ferric aqua, ferrous, and ferrous-NO forms [94] have demonstrated that the Η-bonding network of water molecules is similar in all structures. In the structure of the D140A mutant of human HO-1, however, a new water molecule (water-3) takes the place formerly occupied by the carboxylate side chain of Aspl40, thus Met. Ions Life Sei. 2009, 6, 241-293

256

RIVERA and RODRIGUEZ

lifting the structural restriction imposed on water-1. This structural freedom of water-1 in the D140A mutant allows it to donate a hydrogen bond to the terminal oxygen atom of the F e m - O O H group. Theoretical investigations suggest that protonation of the F e m - 0 0 ~ species results in significant weakening of the O-O bond [92,93]. In fact, it has been observed that the O-O stretching frequency (vO-O) in F e m - O O H is at least 50 cm lower than that in similar F e m - 0 0 ~ , an observation that has been interpreted to indicate that a weakened O-O bond in F e m - O O H is primed for dissociation to convert it into an oxoferryl species [95]. Consequently, donation of a hydrogen bond from water-1 to the terminal oxygen atom in the F e m - O O H complex of the D140A mutant is expected to accelerate its decay to a ferryl moiety, thus explaining the peroxidase activity exhibited by the D140A mutant of HO-1 [94], As is apparent from the discussion above, the crystal structures of human HO-1 [53,96] and rat HO-1 [54] suggested the presence of a number of well ordered water molecules in the distal pocket of these enzymes. Subsequent structures of bacterial heme oxygenase enzymes obtained from N. meningitidis [55], Corynebacterium diphtheriae [56], and Pseudomonas aeruginosa [57] confirmed the presence of several structural water molecules in the distal pocket of these enzymes. In addition, N M R spectroscopic studies conducted with cyanide-inhibited human HO-1 [62,97,98] and cd-HO [99] revealed the presence of exchangeable ( N H and OH) resonances exhibiting strong downfield shifts 10-17ppm) [97], Large downfield N H or O H 1 f I chemical shifts are diagnostic of the presence of very strong hydrogen bonds [100], where the backbone/side chain N H or O H protons participate as donors. The identity of these residues was obtained from resonance assignments and that of the corresponding hydrogen bond acceptors was attained from analysis of the crystal structure. These observations led to the conclusion that there is a relatively rigid network of hydrogen bonds in the distal pocket of H O that is more extensive than what was possible to infer from the X-ray crystal structures [62,97,98]. Hence, solution state investigations reinforced the idea that a hydrogen bond network, which involves Asp 140 in h-HO-1 and r-HO-1, is critical to stabilizing the water molecule that donates a hydrogen bond to the nascent F e m - 0 0 ~ complex.

2.5.

Oxidation of Meso-Hydroxyheme to Verdoheme and the Release of Carbon Monoxide

From the above discussion it is evident that the ferric hydroperoxide species in H O hydroxylates the heme to produce meso-hydroxyheme. The latter reacts with 0 2 to produce verdoheme (see Figure 1), with the concomitant Met. Ions Life Sei. 2009, 6, 241-293

DUAL R O L E OF H E M E R E G A R D I N G C A R B O N M O N O X I D E

257

elimination of the hydroxylated carbon atom as CO. It has been suggested that ferric meso-hydroxyheme is converted to ferric verdoheme upon contact with 0 2 , without the need of an electron [101]. An alternative mechanism in which ferric meso-hydroxyheme is reduced by one electron to ferrousmeso-hydroxyheme, which subsequently reacts with 0 2 to form ferrous verdoheme, has also been proposed [102,103]. This apparent discrepancy may be explained by the suggestion that under physiological conditions both reaction paths are likely, with one dominating over the other depending on the relative concentrations of N A D P H and cytochrome P450 reductase in the cell [41]. It is noteworthy however, that there is complete agreement on the fact that in vitro the CO complex of ferrous meso-hydroxyheme is converted to the CO complex of ferrous verdoheme when the former is reacted with 0 2 , a chemical property that is unique to verdoheme [101-103]. This unique property of the CO complex of meso-hydroxyheme was used in studies aimed at dissecting mechanistic differences between the heme oxygenation reaction, as observed with HO, and the coupled oxidation reaction [104]. The latter is a process [105] whereby heme proteins not involved in heme catabolism degrade heme to verdoheme and biliverdin, with the accompanying release of CO [106-111]. Meso-hydroxyheme exists in at least three different resonance forms: a ferric phenolate anion, a ferric keto, and a ferrous keto π neutral radical. It is interesting that the relative contribution of these different resonance structures changes upon exposure to CO because coordination of the latter to the iron stabilizes the ferrous keto π neutral radical species [112]. In general, CO binds to ferrous heme with higher affinity than 0 2 as is evident from the fact that the binding affinity of free heme for CO is significantly larger than its affinity for 0 2 , as indicated by the ratio of the corresponding equilibrium association constants K c o j K o 2 , which is ~ 25,000 [113]. The structural properties in the binding pocket of globins reduce this ratio to approximately 40 (see Table 1); the relatively high KC0jK0 ratio exhibited by globins explains the fact that exposure to CO

Table 1. Equilibrium association constants for the binding of 0 2 and CO to protoheme-X in different environments.

Protoheme Myoglobin HO-1 pa-HO «m-HO cd- H O

Ko2 ( μ Μ - 1 )

K c o (μΜ" 1 )

K co /Ko 2

Reference

0.015 0.86 28 2.6 3.3 21

440 35 34-150 7.5 6.9 150

- 25,000 41 1.2-5.3 2.9 2.1 7.1

[113] [119] [69] [115] [115] [174]

Met. Ions Life Sei. 2009, 6, 241-293

258

R I V E R A and R O D R I G U E Z

results in the formation of carbomonoxy-hemoglobin and myoglobin. It is therefore interesting that CO generated from heme degradation by H O does not inhibit the enzyme, which in principle could be inhibited by binding CO at several stages of the catalytic cycle, including ferrous heme, meso hydroxyheme, and verdoheme. In fact, the relative immunity of H O to inhibition by CO was made evident in a study demonstrating that 20% CO in air is necessary to inhibit catalytic activity, which is arrested at the verdoheme stage due to the formation of a verdoheme-CO complex. Clearly this concentration of CO is significantly larger than those likely to be encountered under normal physiologic conditions [114]. A more quantitative description of the relative immunity of H O to inhibition by CO was obtained from measurements of the equilibrium association constants for the binding of CO and 0 2 to ferrous H O enzymes (Table 1). These investigations revealed that on average, the structure of H O enzymes lowers the magnitude of the K c o j K o 2 ratio by approximately one order of magnitude relative to the K c o j K o 2 value exhibited by myoglobin. The data in Table 1 also indicate that the affinity of ferrous HO-1 for 0 2 is approximately 30-fold larger than that of myoglobin, whereas the affinity of bacterial HOs for 0 2 is, on average, ~ 5-fold larger than that of Mb. The lower affinity of the bacterial HOs (pa-HO and nm-HO) for 0 2 relative to HO-1 appears to be compensated by a lower affinity for CO relative to HO-1, such that the magnitude of the K c o j K o 2 ratio is similar for all the enzymes; the magnitudes of Kco and KQ2 for cd-HO, on the other hand, are very similar to those exhibited by HO-1. Although some important clues regarding the structural properties that allow H O enzymes to strongly discriminate against CO have emerged, a thorough understanding has not yet been obtained. The existing clues come from details in the measurement of binding affinities, which showed that although the 0 2 association rate constants are similar to those exhibited by the globins, the 0 2 dissociation rate constants are significantly slower [69,113,115]. These observations have been placed in context using information obtained from the crystal structures of H O and from insights derived from extensive work aimed at understanding the discrimination of CO by globins [116-119]. The emerging picture suggests that the slower 0 2 dissociation from H O enzymes stems from the ability of these enzymes to donate several hydrogen bonds to the coordinated 0 2 . This general view is consistent with the polar nature of the distal binding site, which also harbors a network of structural waters that contribute to the stabilization of the coordinated 0 2 . The number of hydrogen bonds and the details of the interactions, however, appear to be different in distinct H O enzymes [69,113,115]. Recent studies demonstrated that the dynamic behavior of the H O polypeptide is largely dependent on the oxidation state and on the nature of the distal ligand coordinating the heme iron [120,121] (see Section 4). It is therefore possible Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

259

that an additional source of discrimination against CO comes from the dynamic plasticity of the heme pocket in HOs, which may be different depending on whether the distal ligand is CO or 0 2 .

3.

3.1.

HEME OXYGENASE FAVORS HEME HYDROXYLATION OVER FERRYL FORMATION. THE NATURE OF THE FERRIC HYDROPEROXIDE COMPLEX IN HEME OXYGENASE Studies with Model Heme Complexes

In the EPR spectrum of the ferric hydroperoxide intermediate of HO-1 [84] the sum of the squares of the g values (Σ^·2) is 14.14. This compressed g anisotropy (Σ^·2 < 1 6 ) suggested [122] the possibility of an electronic configuration, (dxz,dyz)4(dxy)1, ( d x y h e r e a f t e r [123-126]. Ferric porphyrinates exhibiting the ( d ^ ) 1 electronic configuration are known to be highly nonplanar (ruffled) and to place a large amount of unpaired electron density at the meso carbons [123-128]. In comparison, ferric porphyrinates with the more common (dxy)2(dxz,dy^)i electron configuration, d It hereafter, exhibit planar macrocycles and place significant spin density at the α-pyrrole carbons and negligible spin density at the meso carbons [127-129]. To understand the role exerted by a hydroperoxide axial ligand on the electronic structure of ferrihemes, meso- 1 3 C labeled F e m - t e t r a phenylporphyrinates coordinated by a methoxide and an alkyl hydroperoxide ligand, [meso- 1 3 C-TPPFe(OCH 3 )(OO t Bu)]", was investigated by E N D O R and 13 C N M R spectroscopy. The E N D O R spectrum unequivocally indicated that the electron configuration of [meso- 1 3 C-TPPFe (OCH 3 )(OO t Bu)]~ at 8 Κ is low-spin d It , despite the fact that the EPR spectrum displays compressed g anisotropy (Σ^ 2 ~ 14) [122], In striking contrast, results obtained with 13 C N M R spectroscopy at more elevated temperature suggested a different picture: The meso 13 C shift of [meso- 13 CTPPFe(OCH 3 )(OO t Bu)]~ at 218 Κ occurs at 422 ppm. The relevance of this observation is that ferrihemes with the ( d ^ ) 1 electron configuration exhibit large spin density at the meso carbons [127], which induce diagnostically large 1000 ppm) meso carbon shifts [130,131]. In comparison, ferrihemes with the d It electron configuration exhibit spin density into ß-pyrrole carbons [125,127] and negligible spin density at the meso carbons, which results in meso carbon resonances near the diamagnetic position, 70 ppm [132,133]. Hence, the 422 ppm chemical shift observed for [meso- 13 CTPPFe(OCH 3 )(OO t Bu)]~ indicates significant population of the (d M )' electron configuration at 218 Κ [122]. Temperature-dependent investigations Met. Ions Life Sei. 2009, 6, 241-293

260

RIVERA and RODRIGUEZ

of these resonances led the authors to suggest that a population of molecules with a ά π electron configuration and planar heme is in dynamic equilibrium with a population of molecules with a ( d ^ ) 1 electron configuration and a ruffled macrocycle [122,134], as indicated by equation (1). The equilibrium shifts to the left as the temperature decreases and at 8 Κ the fraction with d It planar macrocycles approaches unity, which is in agreement with the results obtained from the E N D O R experiments. In contrast, at the temperature of the 13 C N M R experiments the population with planar and ruffled hemes is nearly equal and at ambient temperature the population with ruffled heme and ( d ^ ) 1 electronic structure is expected to predominate [122]. The authors thus suggested that if a similar equilibrium between planar and ruffled hemes takes place in the heme pocket of H O , the relatively large electron density at the meso-carbons would facilitate attack of the terminal O H group in the F e m - O O H complex [122], planar(dn)

3.2.

?± ruffled{dxy)

(1)

Studies with the Hydroxide Complex of pa-Heme Oxygenase (pa-HO-OH)

The equilibrium between planar and ruffled hemes observed with the model complex described above implied that the electronic structure of the F e m - O O H intermediate in heme oxygenase should be studied at ambient temperature [122]. In an attempt to circumvent the extremely high reactivity of this intermediate the electronic structure of the hydroxide complex of pa-HO was studied as a model of the highly reactive F e m - O O H [135]. This approach capitalizes from methodology developed for the biosynthesis of 13 C-labeled heme [136,137] and from relatively straightforward correlations between core carbon shifts and the coordination state and electronic structure of hemes [138-140,130,132], The 13 C N M R spectra of the hydroxide complex of pa-HO (Figure 6) revealed the presence of at least three populations of molecules in dynamic exchange, each exhibiting a distinct electronic configuration [135]. The most abundant population has an S = 3/2, (dxz, dyz)3(dxy)l(dz2)1 (S = 3/2, (d^) 1 , hereafter) electronic structure [141], which is characterized by pyrrole carbon-α (C a ) and pyrrole-β (Cp) resonances near 400 ppm and meso carbon (C m ) resonances near zero ppm [139] (blue boxes in Figure 6). The spectra also demonstrate the presence of two other populations with relatively low concentrations. One of them has an S = 3/2 (dxy)2(dxz,dyz)2(dz2)1 (S = 3/2, d It , hereafter) electron configuration, with Co, resonances ca. 650 ppm, Cp resonances ca. 1000 ppm and C m resonances ca. -200 ppm (red boxes). The other low-concentration species exhibits an S = l / 2 , ( d ^ ) 1 electron configuration, with C m resonances ca. Met. Ions Life Sei. 2009, 6, 241-293

261

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

Cm, S = 3/2, (d„)'

1 000

800

600 13

-too

200

0

-200

-100

-600

C Chemical Shift (ppm)

Figure 6. 13 C N M R spectra (37 °C) of the hydroxide complex of ρα-ΆΟ reconstituted with heme labeled at the C a and C m carbons (A) and C a and Cß carbons (B). Peaks corresponding to the populations exhibiting the S = 3/2, (d.TV)' spin state are in blue boxes, peaks corresponding to the population with the S = 3/2, d„. spin state are in red boxes, and peaks corresponding to the population with the S = 1/2, (d.TV)' spin state are in black boxes. Adapted from [135].

1300 ppm and C a resonances ca. -400 ppm (black boxes). These electronic configurations are highly unusual among heme active sites. In fact, this is the first example of a heme active site with S = 3/2, (dx-v)1 and S = 1/2, (d vi )1 electronic configurations [135]. More important, ferric porphyrinates with these unusual electron configurations are always associated with non-planar distortions of the macrocycle [142-144,123,125,139], and also place significant unpaired electron density at the meso positions [135]. In contrast, it is noteworthy that the hydroxide complex of globins has the S = 1/2, ά π electron configuration typical of planar hemes [145]. Met. Ions Life Sei. 2009, 6, 241-293

262

R I V E R A and R O D R I G U E Z

A recent 111 N M R spectroscopic study of the hydroxide complex of heme oxygenase from Neisseria meningitidis (nm-HO-OH) concluded [146] that the electronic ground state of this complex is the conventional d It orbital ground state and not the unusual S = 1/2 ( d ^ ) 1 reported for the hydroxide complex of /)α-ΗΟ-ΟΗ. This statement, unfortunately, only adds confusion to current understanding: The authors of the study involving /ja-HO-OH did not conclude that S = 1/2 ( d ^ ) 1 is the ground electronic state of this complex [135]. In fact, as can be seen in Figure 6, the 13 C N M R spectra clearly indicate that the /?a-HO-OH complex exists in at least 3 interconverting electronic states; the S = 1/2 ( d ^ ) 1 electronic state (corresponding resonances in black boxes) is the least populated and therefore it is highly unlikely to be the electronic ground state in the /?a-HO-OH complex. Thus, if the F e m - O O H complex in H O were to behave similarly, heme hydroxylation may occur via the population with the reactive ( d ^ ) 1 electronic structure; the dynamic interconversion of populations would constantly repopulate the low abundance but reactive ( d ^ ) 1 state, therefore facilitating efficient heme oxidation. It is also important to emphasize that discovery of the unusual electronic structures in/?a-HO-OH [135] was possible because this complex was studied with the aid of 13 C N M R spectroscopy, in addition to the traditional approach of observing only N M R resonances. Additional discussion regarding this issue is illustrative: As pointed out above, there are three 13 C N M R observable populations of /ja-HO-OH, each exhibiting a characteristic set of core porphyrin carbon chemical shifts [135]. The relative intensity of the peaks corresponding to the most abundant population facilitates the study of their temperature dependence (see Figure 7). Thus, it is apparent that at 37 °C (1/T ~ 3 . 2 χ 1 0 " 3 K ~ l ) the magnitude of Co, and Cp chemical shifts from the most abundant population is near 500 ppm and 400 ppm, respectively (Δ and Ο in Figure 7). These values are significantly larger than those corresponding to Co, and Cp in /?a-HO-CN (A and · symbols), which is known to exhibit a pure low-spin d It electron configuration [141,147]. In addition, the temperature dependence of the Co, and Cp shifts from /?a-HO-OH is very steep, which also stands in sharp contrast to the modest temperature dependence of the Co, and Cp shifts from /)fl-HO-CN. In contrast, the C m resonances in both, pa-lIO-OII and /?a-HO-CN are very similar and near their corresponding diamagnetic shifts. There is no question that the information content of the 13 C shifts is rich and points to the fact that the electronic structure of this most abundant population of /ja-HO-OH is not the typical low-spin ά π . In stark contrast, the paramagnetically resolved resonances in the ' H N M R spectrum of /?a-HO-OH do not alert the experimentalist of an unusual situation. In fact, if one were to study only the 111 N M R spectrum of pa-l IO-OI I (Figure 8B), which is somewhat similar that of /»a-HO-CN (Figure 8A), it would not be Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

263

500 Δ

c

400 - Ο

Λ C

Δ Ο

ß

Έ Q. £ 300 •C cn "5 ο Ε ®

υ

δ 6

200 -

υ

C O τ—

•Cß

100 -

DA

• ο 3.2

Ca I 3.3

•

•

Q •

A

• °cm

• CIm

I 3.4

3.5 1

• •

Γ • 3.6

3.7

3

1/Τ (Κ" Χ 10 ) Figure 7. Temperature dependence of core carbon chemical shifts corresponding to /kz-HO-OH (open symbols) and />a-HO-CN (filled symbols). Adapted from [135]. difficult to conclude that nothing is unusual in the electronic structure of the /m-HO-OH complex. However, as discussed above, the remarkable 13 C chemical shifts and their temperature dependence prompted detailed N M R spectroscopic probing of the /?a-HO-OH complex [135]. These investigations revealed that the most abundant population of /?a-HO-OH exists as an equilibrium mixture of two species, one exhibiting an S = 1/2 d It electron configuration and planar heme and a second with a novel S = 3/2, ( d ^ ) 1 spin state and nonplanar heme. As has been already mentioned, the same investigations revealed two other populations of lower concentration, one with the S = 3/2, d It electronic state and the lowest concentration population with the S = 1/2 ( d ^ ) 1 electron configuration.

3.3.

Studies with the Azide Complex of pa-HO (pa-HO-N 3 )

Observations made with the hydroxide complex of pa-HO suggest that the nature of the axial ligand is important in inducing the unusual heme Met. Ions Life Sei. 2009, 6, 241-293

264

RIVERA and RODRIGUEZ

(A) pa-HO-CN

30

10

20 1

Figure 8.

0

-10

H Chemical Shift (ppm)

' H N M R spectra of ρα-ΗΟ-CN (A) and pa-HO-OH (B).

electronic structures and deformations described above. For example, when C N " is the distal ligand in ρα-ΆΟ [147], human [148], rat [149], or bacterial [99,150] HOs, the corresponding Ή and 13 C N M R spectra are always indicative of planar low-spin d It electronic hemes. In contrast, when the distal ligand is capable of accepting a hydrogen bond at the coordinating atom (i.e., OH~) the resulting complex no longer exhibits the common low-spin d It electron configuration characteristic of planar ferrihemes. Thus, the azide (Ni~) complex of pa-HO (/?a-HO-N 3 ) was studied to further probe this idea [141] because the N J ligand is also capable of accepting a hydrogen bond at the coordinating atom. Once again, 13 C N M R spectroscopic investigations demonstrated that the /M-HO-N3 complex does not harbor a planar ferriheme with the S = 1/2, d It electronic structure. Instead, the binding of azide leads to nonplanar hemes and electronic structures similar to those described above for /?a-HO-OH. The magnitude and temperature dependence of the chemical shifts from the /M-HO-N3 complex [141] are strikingly similar to those exhibited by the most abundant population of /?a-HO-OH (see above). Detailed analysis of these observations in the context of molecular orbitals and concomitant measurements of magnetic moment led to the conclusion that the /?(2-HO-N3 complex also exists at ambient temperature as a mixture of two interconverting populations, one with the d It electron configuration Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

265

and planar heme and the other with the S = 3/2, ( d ^ ) 1 electron configuration and nonplanar heme [141].

3.4.

A Hydrogen Bond from Gly125 N-H to Coordinated Azide-N May Promote the Unusual Electronic Structure of the pa-HO-N3

The above-described studies of the heme active site in H O led to the conclusion that the enzyme channels the activation of 0 2 toward heme hydroxylation aided by unusual heme electronic structures of the F e m - O O H oxidizing species. An advantage of /?a-HO-N 3 as a model of the F e m - O O H intermediate in H O is that it can be studied at neutral pH, which permits concomitant probing of the polypeptide with multinuclear and multidimensional N M R experiments. Several N M R studies of the polypeptide in H O have been carried out with the aid of homonuclear methods that rely on the larger than normal dispersion of : H resonances affected by heme paramagnetism. These studies provided several assignments of residues in relative close proximity to the paramagnetic heme active site [151-153,62]. Nevertheless, understanding polypeptide dynamics and polypeptide-heme interactions in H O enzymes requires the availability of complete heteronuclear and sequential assignments that permit global probing of the enzyme. A study reporting the sequential assignment of backbone ( N - ' H , 15 N-H, C a , Cp, and C') resonances of the 198-residue-long, paramagnetic pa-HO in complex with azide and with cyanide paved the way to study the dynamic behavior of HOs using N M R methods [120]. The strategy used for obtaining resonance assignments from residues strongly affected by the heme iron paramagnetism made use of selective amino acid labeling and N M R experiments tailored for the observation of fast relaxing resonances [120]. Selective amino acid labeling was also used to probe the notion that the lower than typical field strength of N J coordinated to pa-HO likely stems from accepting a hydrogen bond at the coordinating atom. The hydrogen bond donor would be either the N - H from Glyl25, either directly or mediated by a water molecule [134—135,141]. Thus, /?a-HO-N 3 selectively labeled with 1 5 N Gly ( 15 N-Gly-/?a-HO) was used in these investigations [120]. Despite considerable effort, the cross peak corresponding to Glyl25 was not found in the H S Q C spectrum of /?a-HO-N 3 , presumably because the proximity to the paramagnetic iron greatly increases the relaxation rate of the amide hydrogen in Glyl25. On the other hand, since the gyromagnetic ratio of 1 5 N is 10-fold smaller than that of : H , the effect of the paramagnetic center on 1 5 N relaxation is expected to be approximately 100-fold smaller. Met. Ions Life Sei. 2009, 6, 241-293

266

R I V E R A and R O D R I G U E Z

Thus, at 37 °C the 1 5 N resonance of Glyl25 in /?a-HO-N 3 was observed at 128 ppm and 133 ppm in a one-dimensional 15 N spectrum (Figure 9a) [120]. Two peaks (128 and 133 ppm) with relative intensity 70:30 are observed because /?a-HO-N 3 exists in solution as a mixture of two heme orientational isomers with equilibrium concentrations ~ 70:30 [147]. Clearly, these resonances are at least 3-fold broader than other 15 N resonances in /?a-HO-N 3 , are significantly downfield shifted relative to other Gly 15 N resonances, and exhibit pronounced temperature dependence (Figure 9). These unusual properties strongly suggest through-bond (Fermi contact) spin derealization from the heme iron into the Glyl25 1 5 N atom. Through-bond spin derealization would only be possible if the Glyl25 N-H forms a hydrogen bond with the iron-bound nitrogen of azide. Additional support for this idea

5 1 5 N (ppm) Figure 9. One-dimensional 1 5 N N M R spectrum o f / w - H O - N 3 labeled with 15 N-Gly obtained at (a) 37 °C, (b) 32 °C, (c) 20 °C and (d) 8 °C. Adapted f r o m [120]. Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

267

was found in the uncommonly large upfield deuterium isotope effect observed for the 1 5 N resonance of Glyl25 which also suggests unpaired spin derealization onto the 1 5 N atom of Glyl25 via a hydrogen bond [120]. In comparison, it is important to underscore that the 1 5 N and resonances of Glyl25 in /?a-HO-CN, where the coordinated carbon atom cannot accept a H-bond, display "standard" chemical shifts, line widths, and deuterium isotope effects.

3.5.

Implications to the Mechanism of Heme Oxidation by Heme Oxygenase

The above-discussed observations led to the postulation [134,135,141] that coordinated OH~ or N J , by virtue of accepting a hydrogen bond to the coordinating atom, lower their σ-donating ability, thereby lowering their field strength. Decreasing axial ligand field strength is typically accompanied by: (i) strengthening of the equatorial field, i.e., shorter bonds between the iron and the pyrrole nitrogens, which induce nonplanar porphyrin deformations and (ii) stabilization of the dz2 orbital which facilitates attainment of the S = 3/2 electron configuration [154-156]; and/or stabilization of the dxz and d y z orbitals relative to the in-plane dxy orbital, which is conducive to the ( d x y e l e c t r o n i c structure [144,157-159]. Consequently, if the field strength of the HOCT ligand is modulated similarly via a hydrogen bond to the coordinating Ο atom, the nonplanar deformations and unpaired spin density at the meso carbons observed in the populations with the ( d ^ ) 1 and S = 3/2 spin states would facilitate homolytic cleavage of the O-O bond by efficiently trapping the O H radical at a sterically unprotected meso carbon. Thus, if a limiting resonance structure in which the electron is fully transferred from the porphyrinate ring to the metal is considered (Figure 10b), attack of O H at a meso carbon would produce c, which would rapidly rearrange its Fe-O electron configuration, lose a proton from the attacked meso carbon to re-aromatize the porphyrin ring, and protonate the F e m - O i ~ complex to yield the ferric meso-hydroxyheme complex d [122,141]. In this context, it is interesting that a theoretical study has suggested that heme hydroxylation by H O is more likely to occur via a stepwise mechanism in which homolytic cleavage of the O-O bond is followed by trapping of the O H radical at the porphyrin [160]. Importantly, more recent calculations [161] also suggest that homolytic cleavage of the O-O bond proceeds via a species with an electronic structure resembling a porphyrin radical, P o r + ' - F e m O H — H O ' . Hence, the notion derived from experimental studies with /?a-HO-OH and /?a-HO-N 3 , which suggests that the porphyrin in the F e m - O O H intermediate has significant unpaired spin density at the meso carbons, is in good agreement with theoretical developments. Met. Ions Life Sei. 2009, 6, 241-293

268

RIVERA and RODRIGUEZ

ρ

Me < Οf °

H

y

ι'

d

c

Figure 10. Schematic illustration of the proposed activation of meso heme carbons in the F e m - O O H complex of HO. (a) The population with the ( d ^ ) 1 electronic structure can be viewed as the limiting resonance structure (b), which facilitates homolytic cleavage of the O-O bond to produce (c). The latter is expected to rapidly rearrange its Fe-O electron configuration, rearomatize the porphyrin ring and protonate the F e m - 0 2 ~ complex to form the meso hydroxyheme (d).

4.

HEME OXYGENASE DYNAMICS AND HEME BREAKDOWN. THE DISTAL LIGAND HAS A PROFOUND EFFECT IN THE DYNAMIC BEHAVIOR OF HEME OXYGENASE

Protein dynamics has been largely recognized as an essential determinant of protein function [162-164]. Because the H O catalytic cycle involves multiple changes in coordination and redox states, as well as in the structure of the substrate (heme), while undergoing only minor structural alterations, the polypeptide matrix of H O can be regarded as being inherently dynamic. Furthermore, since H O conserves the fold of the holo-protein when devoid of heme [59,61,165], it seems intuitive that for efficient substrate binding and product (biliverdin) release to occur, the polypeptide must be flexible. Met. Ions Life Sei. 2009, 6, 241-293

DUAL R O L E OF H E M E R E G A R D I N G C A R B O N M O N O X I D E

269

Recent work on bacterial HOs has addressed the issue of protein dynamics and its potential role in enzyme catalysis [120,121]. In what follows, a general overview of the most recent advances related to the dynamic nature of the backbone in H O is presented, followed by a general view of the dynamic behavior that H O exhibits at specific stages of the catalytic cycle, as determined from hydrogen-deuterium (H/D) exchange studies and N M R relaxation measurements conducted with CN~, Ni~, and CO-inhibited HO.

4.1.

H/D Exchange

The propensity of hydrogen atoms of backbone amides to undergo exchange reactions with the solvent in solution is a sensitive indicator of protein flexibility [166,167]. Global probing of protein flexibility by H / D exchange is commonly performed by monitoring the loss of signal intensity in an N M R spectrum as a function of the amount of time that the protein has been in contact with D 2 0 . Values of the exchange rate constants (k ex ) measured in this fashion are then compared to values of the intrinsic rate constant (k ch ). The latter, which is a measure of the rate of exchange exhibited by an amide proton in an unstructured polypeptide, is sequence-dependent and can be calculated [168,169]. A commonly used term to express the propensity of amide hydrogens to exchange is the protection factor. Protection factors expressed in logarithmic form (log P = log kexjkcll) provide a quantitative estimation of the propensity of a given amide proton to exchange with deuterons in the solvent [168,169]. H / D exchange studies were recently performed with azide- and cyanideinhibited pa-HO [120]. Rates of H / D exchange (k e x ) were measured for /?(2-HO-N3 and /?a-HO-CN and converted into protection factors. Analysis of the data obtained with each complex are summarized in Figure 11 A. To appreciate the information content in this figure it is important to keep in mind that the difference in protection factors, calculated as Alog Ρ = log P N 3 - log Pcn, has been plotted per residue. In addition, the horizontal line at Alog Ρ ± 0 . 5 represents the average obtained from summing the absolute value of Alog Ρ for all residues exhibiting different protection toward exchange. Values of Alog Ρ = 1 . 5 correspond to cases where a residue in /?(2-HO-N3 exchanges with an experimentally measurable kex but the corresponding residue in /?a-HO-CN exchanges too fast to allow measurement. This means that although the magnitude of Alog Ρ cannot be evaluated for these residues, they are clearly located in sections of the protein with the most pronounced differences in propensity to exchange. The data revealed that the majority of residues where there is a difference in propensity to exchange exhibit Alog Ρ > 0, which indicates that globally, the backbone amide groups of /?a-HO-N 3 are distinctively less prone to Met. Ions Life Sei. 2009, 6, 241-293

270

R I V E R A and R O D R I G U E Z

A 1 : : : : τ J: ?:

1 1 •: 1: • . . . . 1. ·.·:;!(! ;· Π •| 10

1 1 1 1

•

1 • J 1 \

1 r 20

1 1 1

I1 1 1 i l l 1 1 1 .1. 1 ; |·; 1 30

ι

1! l l

1

50 Residue 1

II-

III i 1

!i III 1 i • i 1 a' IK · | ! τ·;··;: 1 1

1

!1

ll 1 1 . . . :1 1 1

Ii 1

1

1

1

60

70

80

90 • 1 • 1 • • t i t

·

llllll II; IIIIII 1 1. .. 1··(':·:• 1 1 1 1 1

1

, ll III .1 • •

ll L .

1

i!

j

1

i 1 1

1 :

1

I i

jl

1 1 1 1• 40

Ii

mi, 1 R

'

1 ;i 1

• 1 1 1 1

:

1 it ll 1 1 )

ί I .. t 1

Iii

15 10 0 5

a.

• -0 5

«

:

- -1.0

κ )0

Γ

1 0

-

0.5

--0.5

£L f l

r -1 0 - -1 5

100

110

120

130

140

150

160

170

180

190

200

Residue

Β pa-HO-N3 > » pa-HO-CN

pa-HO-N3 » pa-HO-CN

pa-HO-N3 > pa-HO-CN

Figure 11. (A) Per residue differences in protection factors (Alog Ρ = log P a 2 i d e - log Pcyanide) between />ö-HO-N 3 has enhanced protection to exchange relative to /w-HO-CN (AlogP is positive in A). Portions exhibiting 0 < A l o g P < 0 . 5 are green, 0.5 average (0.5) in yellow and residues with Alog Ρ < average in green. Remarkable about these findings is the fact that the nature of the distal ligand, the diatomic CN~ or the triatomic N J , appear to exert a large influence on the flexibility of the enzyme, not only in regions of close proximity to the heme, but also in portions of the protein far removed from it. Among residues that are most protected in /?a-HO-N 3 relative to pa-HOC N (pink and yellow in Figure IIB), those located in helices V, VI, and X and the loop preceding helix X stand out because they are remote from the heme iron and distal ligand. Interestingly, inclusion of the crystallographic structural waters in the model (blue in Figure IIB) strongly suggests the possibility that long range communication between the distal ligand and helices V, VI, and X occurs via these networked structural waters. As has been discussed above, this H-bond network is thought to function in H O as an integral part of a proton delivery machinery used to protonate the nascent F e m - 0 0 ~ moiety in the process of 0 2 activation leading to meso-carbon hydroxylation [84,94]. The observations made with H / D exchange experiments, therefore, suggest that proton delivery is not the only function of the network of Η-bonded water molecules. Instead, the data implies the tantalizing possibility that an additional function of the structural waters in H O is to provide adaptable interactions between otherwise remote structural elements, thus permitting rapid propagation of conformational changes in the active site with minimum perturbation of secondary structure.

4.2.

Microsecond-Millisecond Dynamics

N M R spectroscopy has emerged as the technique of choice to gain insights into conformational motions of enzymes in solution. N M R relaxation experiments designed to quantitatively determine the contribution of conformational exchange to the overall transverse relaxation rates of 1 5 N nuclei have been proven particularly powerful in this regard [170,171]. Such an approach was recently applied to study /?a-HO-N 3 , /»a-HO-CN, and pa-HO-CO, in order to directly establish whether the axial ligand exerts measurable effects on the dynamic characteristics of the protein backbone [121]. The information derived from these studies, which is complementary to that obtained from H / D exchange measurements, allowed for direct identification of amino acids undergoing μβ-ηιβ conformational motions. To achieve this end, the relaxation-compensated Carr-Purcell-Meiboom-Gill Met. Ions Life Sei. 2009, 6, 241-293

272

R I V E R A and R O D R I G U E Z

(rc-CPMG) pulse sequence was used to measure the effective transverse relaxation rates (ARF F ) of 1 5 N nuclei in / ) Ö - H O - N 3 and / W - H O - C N . A comparative analysis of the AR| F F values obtained from these experiments revealed that, similar to the findings obtained from the H/D exchange studies, CN~ and Ν J" exert a distinct influence on the dynamic behavior of the protein backbone. Again, the most significant differences between the CN~ and NJ" inhibited enzymes were located in regions that are far removed from the heme iron. It is noteworthy that residues whose relaxation rates were differentially affected by the nature of the axial ligand are involved in hydrogen bonding interactions, via backbone or corresponding side chain, with water molecules in the hydrogen bonding network [121] (Figure 12). This indicates that the differential effects exerted by the ligand on backbone dynamics can be accounted for if direct participation of the hydrogen bonding network is invoked. Consequently, the hydrogen bonding network may be envisioned as a plastic structural motif that allows HO to adapt readily to changes in the reactive (coordination) state of the heme iron. To directly test the involvement of the H-bond network in modulating backbone dynamics in HO, the Arg80Leu mutant of pa-l IΟ (R80L-/>«-[ IO)

Figure 12. Blown-out view of ρα-ΆΟ (PDB:lsk7) highlighting the hydrogen bonding interactions between water molecules and amino acid side chains undergoing μβ-ηιβ backbone conformational motions. R o m a n numerals are used to identify helices in the structure according to the nomenclature used in the P D B file. Arabic numerals identifying crystallographic waters (blue spheres) are given as assigned in the structure coordinates. Reproduced from [121] with permission of the American Chemical Society, copyright (2007). Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF H E M E R E G A R D I N G C A R B O N M O N O X I D E

273

was prepared and used to measure per-residue values of AR| f f . R80 in pa-HO is intimately involved in the hydrogen bonding network of this enzyme (see Figure 12) and is equivalent to D140 in HO-1 (Section 2.4). Replacing this residue for an aliphatic side chain disrupts the hydrogen bonding network and significantly decrease the catalytic activity of pa-HO [121]. Remarkably, a large number of residues in the R80L-/?a-HO-N 3 mutant undergo μβ-ηιβ motions, as is indicated in Figure 13B by the large number of residues with AR| f f values larger than the average, which is indicated by the segmented horizontal line. This is in stark contrast to the nearly complete μβ-ηιβ stillness of the wild-type /?a-HO-N 3 complex, which is apparent from the few residues with AR| f f values above the average in Figure 13A. A comparison of data in Figures 13A and 13B, therefore, reveals that residues unaffected by conformational exchange in wild-type /?a-HO-N 3 become clearly affected by μβ-ηιβ motions in R80L-/?a-HO-N 3 . These findings, which were interpreted to be indicative of chaotic motions provoked by the severe disruption of the Η-bonding network, highlight the importance

Figure 13. Conformational motions in wild-type />a-HO-N3 (A) and R80L-/>a-HON 3 (B). Amino acid residues exhibiting AR| ff values greater than three standard deviations above the average (dashed line) are affected by μβ-πιβ timescale conformational exchange. The widespread occurrence of residues affected by conformational exchange in R80L-/>a-HO-N3 relative to wild-type />a-HO-N3 is indicative of the role of the hydrogen bonding network in modulating the dynamic behavior of the protein. Partial reproduction of Figure 4 in [121] by permission of the American Chemical Society, copyright (2007). Met. Ions Life Sei. 2009, 6, 241-293

274

RIVERA and RODRIGUEZ

of this structural motif in controlling the dynamic characteristics of the protein [121]. Relaxation measurements were also conducted with CO-inhibited pa-HO. Unlike the N J - and CN~-inhibited enzymes, pa-HO-CO possesses iron in the + 2 oxidation state, which renders it diamagnetic and more amenable for N M R spectroscopic investigations, which allows dynamic characterization of regions near the heme. These investigations showed that residues located in the distal helix, in the region known as the helix kink, are relatively rigid with respect to μβ-ηιβ dynamics. It is possible that the dynamic characteristics exhibited by ferrous pa-HO-CO approximate the dynamic state of the pa-W0-02 complex in the catalytic cycle, prior to formation of the ferric peroxo intermediate. In such scenario, the rigidity of the helix kink can be envisioned to conformationally "organize" the enzyme prior to accepting the second electron and therefore facilitate the formation of the F e m - 0 - 0 ~ intermediate [121]. The effects of CO-inhibition on the μβ-ηιβ motions of the /?a-HO-R8() mutant have also been studied. Surprisingly, per-residue AR| f f values measured for R80L-/)«-! IO-CO are similar to those of equivalent residues in wild-type pa-HO-CO, suggesting conformational stillness of the CO-inhibited mutant in this time scale [121]. A remarkable feature of spectra obtained with R80L-/?a-HO-CO, however, is the presence of a noticeably greater number of cross peaks in well-resolved sections of the H S Q C spectrum, relative to equivalent sections of spectra acquired with wild-type pa-HO-CO (Figure 14). This suggested that a single cross-peak in the spectrum of wild-type pa-HO-CO is split into two or more cross-peaks in the spectrum of R80L-/)«-! IO-CO. Sequential assignment of the "doubled" resonances confirmed that single amino acids in R80L-/>«-[ IO-CO produce two well-resolved cross-peaks, indicating the existence of at least two distinct conformations of the mutant in solution. The conformational heterogeneity of the CO-inhibited mutant exchanges slowly ( > 300 ms) compared to the μβ-ηιβ motions of the ferric R80L-/?a-HO-N 3 complex. These observations demonstrated that the time scale of the conformational disorder brought about by disruption of the H-bond network in the R80L-/?a-HO mutant are clearly dependent on both, the nature of the distal axial ligand and the oxidation state of the heme iron [121].

4.3.

A Unifying View of Protein Dynamics and Heme Oxygenase Reactivity

:

H N M R spectroscopic investigations conducted with human HO-1 and nm-HO axially coordinated by water, cyanide, and hydroxide in the distal side have established that the hydrogen bond network in these enzymes is Met. Ions Life Sei. 2009, 6, 241-293

DUAL R O L E OF H E M E R E G A R D I N G C A R B O N M O N O X I D E

Ο

£-

0 5

1

S41

*

109

β

110

T199

X

0

G153 1 θ F39

r

ε

f

G181 F3S

ΐ 108

Ο T196 8

8.0

Ή (wm)

β

] Λ

Ο

G98 I 8.5

S41 / \

r

°

G181 11 CL

Β

A

a

104

275

ess

η S>-in. Λ»

8.5

0.0 1

7.5

H (ppm)

Figure 14. Representative expansions of 'H- 1 5 N HSQC spectra of ferrous [U- 15 N]^ / - H O - C O (A) and ferrous [U- 1 5 N]-R80L-^/-HO-CO (B). Labels identify crosspeaks according to their sequential assignment. A single backbone amide gives rise to two cross-peaks in spectrum B; indicating the presence of two distinct conformations of R80L-/>fl-HO-CO in solution. In contrast, equivalent residues in /w-HO-CO give rise to a single cross-peak as shown in spectrum A. Reproduced partially from Figure 10 in [121] by permission of the American Chemical Society, copyright (2007).

structurally robust [62,146,153,172]. This notion was inferred from the low level of deuterium incorporation into amide groups of residues located near the active site after the protein was exposed to D 2 0 for extended periods of time. In addition, results from recent theoretical Q M / M M calculations suggest that while robust, the Η-bonding network it is prone to undergo reorganization caused by changes in the pattern of Η bond to and from water. The robustness of the Η bond network was rationalized to exist in order to restrict the orientation and location of the OH radical in the active site prior to hydroxylation of heme [173]. In the context of the experimental dynamic behavior discussed above, it is likely that the ability of the water network to alter its Η-bonding pattern is also related to the need that HO has to exert dynamic control on remote regions of the enzyme, as suggested by the N M R relaxation and H / D exchange data described above. Hence, the emerging picture seems to suggest that while maintaining a rigid structure near the active site is important to stabilize reactive intermediates for oxygen activation and heme cleavage, exerting dynamic control over residues located in the protein periphery may be equally essential to facilitate substrate binding, docking with the reductase that provides the electrons needed in the catalytic cycle and product release. Although more experimentation is needed to corroborate this assertion, it seems undeniable that the fold and sequence of H O is exquisitely equipped to sense the changes in redox and coordination state that occur Met. Ions Life Sei. 2009, 6, 241-293

276

RIVERA and RODRIGUEZ

in the active site at each stage of the catalytic cycle. These changes, in turn, elicit signals that instill in the protein the need to tune the dynamic behavior of the polypeptide, according to the demands imposed by the distinct stages of the catalytic cycle.

5. 5.1.

THE REGIOSELECTIVITY OF HEME HYDRO XYLATION Hydroxylation of the α-Meso Carbon Leads to Its Release as CO with Subsequent Formation of a-Biliverdin

The catalytic activity of H O produces iron, CO, and exclusively one isomer of biliverdin; typically α-biliverdin. This high level of regioselectivity is dictated at the stage of meso carbon hydroxylation because the hydroxylated meso carbon is subsequently released from the macrocycle as CO with the concomitant formation of the corresponding isomer of verdoheme and subsequently, biliverdin (see Figure 1). This section summarizes findings from work aimed at understanding the properties of H O that allow it to convert heme to biliverdin with a high degree of regioselectivity. The crystal structure of human HO-1 revealed that its distal pocket is considerably more polar than that of globins and cytochromes and exhibits several polypeptide backbone-heme contacts (see Figure 4) [53]. The distal helix approaches the heme within 3 - 4 A along the entire width of the macrocycle and therefore provides steric protection to the β-, γ-, and δ-meso carbons. This observation strongly suggested that exclusive attack of the unprotected α-meso carbon is controlled by steric steering of the HOO~ moiety in the F e m - O O H complex [53]. Corroborating evidence for significant steric control of regioselectivity in heme hydroxylation came from the crystal structure of the azide ( F e m - N 3 ) complex of rat HO-1 [81]. As is clearly illustrated in Figure 15, the coordinated azide molecule is bent ( F e m - N ( 3 ) - N ( l ) angle = 116°) and directed toward the ameso carbon [81]. Interactions between the coordinated NJ" molecule and Glyl39, Serl42, Glyl43, and Glyl44 not only orient the terminal N ( l ) atom to lie almost on top of the α-meso carbon, but also provide steric protection to the β-, γ-, and δ-meso carbons [81]. A very similar picture has emerged from the crystallization of the F e I I - 0 2 complex of cd-HO, which demonstrates that the 0 2 molecule is bent toward the α-meso axis and prevents access to the β-, γ-, and δ-meso carbons by steric interactions with Glyl35, 139, and 140 [174]. These observations also underscore the importance of the Gly residues in the distal helix of HO, Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

277

Figure 15. Space filling representation illustrating the interactions between coordinated azide ( F e m - N 3 ) and Glyl39, Serl42, Glyl43, and Glyl44 in rat HO-1, which are responsible for steering the terminal nitrogen ( N l ) toward the α-meso carbon. Similar steric interactions are expected to orient the H O O ~ ligand to facilitate a-meso hydroxylation. Adapted from reference [81]. Color key: heme (red), coordinated azide (yellow), Glyl39 (green), Serl42 (magenta), Glyl43 (cyan) and Glyl44 (orange).

which serve at least three critical roles: (z) impart flexibility for binding and release of substrate and product, (z'z) make close contacts with the heme and coordinated HOO~ ligand with the purpose of not only steering the latter toward the α-meso carbon but also protecting the remaining meso carbons, and (in) donate a hydrogen bond to the coordinated Ο in F e m - O O H to lower its ligand field strength, thereby promoting the electronic structure that activates the substrate (heme).

5.2.

pa-Heme Oxygenase Exhibits Unique δ-Regioselectivity

Although most of the discussion so far has been in the context of understanding the exquisite α-regioselectivity exhibited by heme oxygenase, it is interesting at this point to discuss the only known H O that oxidizes heme at a carbon other than the α-meso carbon. This discussion is illustrative not only because these investigations contributed significantly to the notion that polypeptide-heme propionate interactions control the regioselectivity of heme oxidation, but also because it illustrates the usefulness of N M R spectroscopy applied to these systems. Heme oxygenase from Pseudomonas aeruginosa, coded by the Pseudomonas iron inducible gene Pig A, was initially reported to oxidize heme to produce mostly ß-biliverdin [33]. Met. Ions Life Sei. 2009, 6, 241-293

278

R I V E R A and R O D R I G U E Z

CO,H

Figure 16. (A) The heme seating in ρα-ΆΟ locates the δ-meso carbon where it is susceptible to hydroxylation by the sterically constrained F e m - O O H moiety. Rotation of the heme by 180° about the α-γ-meso axis places the ß-meso carbon where it is susceptible to hydroxylation. (B) The heme in all α-hydroxylating H O s of known structure is rotated ~ 100° relative to the heme seating in ρα-ΆΟ, thus placing the α-meso carbon where it is susceptible to hydroxylation by the sterically constrained F e m - O O H group. Color key: heme (red), proximal histidine (yellow); Phel 17 i n ρ α - Ά Ο , equivalent to Tyrl 12 in nm-HO (blue); A s n l 9 i n p a - H O , equivalent to Lysl6 in nm-HO (green), Lys34 in pa-HO (magenta). P D B numbers are 1SK7 and iJ77, respectively.

Subsequent studies established that ^a-HO oxidizes heme to a mixture of β- (30%) and δ-biliverdin (70%) [147]. N M R spectroscopic investigations conducted with the cyanide-inhibited enzyme revealed that the heme in ^ö-HO is rotated in-plane approximately 100° relative to the heme in all other α-hydroxylating heme oxygenase enzymes (see Figure 16A). This inplane rotation locates the δ-meso carbon within the ^a-HO fold in the same place where all α-hydroxylating heme oxygenase enzymes place the a-meso carbon [147] (see Figure 16B ). It is also clear that a 180° rotation of the heme in Figure 16A about the α-γ-meso axis places the ß-meso carbon in ^a-HO where it can be hydroxylated, thus explaining the formation of 30% ß-biliverdin [147], Amino acid sequence alignments in the context of the known structures of heme oxygenase suggested that in-plane rotation of the heme in ^a-HO Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

279

stems from the absence of hydrogen bonding and electrostatic interactions between the heme propionates and the side chains of A s n l 9 and P h e l l 7 (see Figure 16A). When these residues in pa-HO were mutated for Lys and Tyr, respectively, in an attempt to introduce the hydrogen bonding and electrostatic interactions typical of heme propionates in α-hydroxylating heme oxygenase enzymes (see Figure 16B), the heme was found to undergo a dynamic in-plane rotation 100°) [147]. N M R spectroscopic experiments also revealed that the in-plane rotation of the heme exchanges the position of the α-meso carbon in one of the in-plane conformers (heme seating) with the δ-meso carbon of the second heme seating. As a result of this dynamic equilibrium the N19K/F117Y double mutant of pa-HO oxidizes heme to a mixture of 55% α-, 35% δ-, and 10% ß-biliverdin [147]. α-Biliverdin is produced from the heme in-plane conformation (seating) similar to that of α-hydroxylating heme oxygenase (Figure 16B), δ-biliverdin is produced from the heme seating characteristic of pa-HO (Figure 16A) and ß-biliverdin is produced from a heme orientational isomer obtained by rotating the heme in Figure 16A 180° about the α-γ-meso axis [147]. The advent of the X-ray crystal structure of/?a-HO [57] corroborated the unusual seating of the heme in ^ a - H O and demonstrated that a key interaction between Lys34 and heme propionate-7 (see Figure 16A) stabilizes the heme seating that in pa-HO is conducive to β- and δ-hydroxylation [57]. A similar conclusion was reached almost simultaneously from site-directed mutagenesis investigations [175]. The high amplitude in-plane rotation of the heme 100°), which occurs in a submillisecond time scale, is highly unusual for heme proteins. Typically, steric interactions between heme substituents and side chains lining the heme pocket provide a strong steric friction that maintains the heme in place. Although these interactions are not as pronounced in heme oxygenase, the unusual large-amplitude in-plane rotation of the heme is likely accompanied by polypeptide motions that facilitate the long in-plane excursions of the heme in HO. Although this phenomenon was initially observed only with the N19K/F117Y mutant of pa-HO [147], subsequent work clearly demonstrated that removing key heme propionate-polypeptide interactions leads to large-amplitude in-plane heme disorder and consequently to changes in regioselectivity [147,150,176,177]. The following section summarizes the findings leading to this conclusion.

5.3.

Polypeptide-Heme Interactions Control the Regioselectivity of Heme Oxidation

The first documented change of regioselectivity brought about by site directed mutagenesis of a heme oxygenase enzyme was observed with the R183D and R183E mutants of rat HO-1, which under coupled oxidation Met. Ions Life Sei. 2009, 6, 241-293

280

RIVERA and R O D R I G U E Z

conditions render a 65:35 mixture of a - and ß-biliverdin, respectively [178]. These investigators proposed two plausible mechanisms to explain the change in regioselectivity: (1) Electrostatic repulsion between D 1 8 3 or E183 and one of the heme propionates, which force the heme to rotate and (2) f o r m a t i o n of a new hydrogen bonding n e t w o r k caused by rearrangement of SI42 and Κ179. The mechanism invoking heme r o t a t i o n was disfavored on the basis that the p a t t e r n of p a r a m a g n e t i c heme methyl resonances of the R183 m u t a n t s is very similar to that exhibited by wild-type rat H O - 1 [178]. This issue was reexamined by studying the corresponding R 1 7 7 D and R 1 7 7 E m u t a n t s in heme oxygenase f r o m C. diphtheriae [150]. Replacement of R177 for Ε or D in cd-HO results in the oxidation of heme to a mixture consisting of ~ 50% of each, a - and δ-biliverdin. Detailed N M R spectroscopic analysis of the m u t a n t s revealed t h a t f o r m a t i o n of δ-biliverdin in the m u t a n t stems f r o m electrostatic repulsion of the heme propionates by the side chain of D or Ε at position 177. Hence, in addition to d e m o n s t r a t i n g that electrostatic repulsion between the negatively charged E-177 and heme propionates indeed causes heme r o t a t i o n , the findings f r o m this study contributed to strengthen the n o t i o n that the H O O ~ ligand is sterically restrained and oriented t o w a r d the α-meso c a r b o n , or whichever meso c a r b o n takes its place u p o n in-plane heme r o t a t i o n [147,150]. Subsequent N M R spectroscopic investigations carried out with m u t a n t s of H O - 1 also d e m o n s t r a t e d that placement of negatively charged side chains in close proximity to one of the heme propionates causes ~ 90° in-plane r o t a t i o n of the heme, which was also accompanied by f o r m a t i o n of a mixture of a - and δ-biliverdin [177]. T o f u r t h e r investigate the idea that key heme propionate-polypeptide interactions p r o m o t e the a p p r o p r i a t e in-plane heme c o n f o r m a t i o n t h a t positions the correct meso c a r b o n where it is unhindered to react with the F e m - O O H intermediate, a chimeric protein was constructed and investigated spectroscopically [176]. T h e chimeric protein was prepared by replacing the distal helix of the α-regioselective nm-HO with the distal helix of the δ-regioselective pa-HO, as shown schematically in Figure 17. The resultant nm-HO chimera, « m - H O c h hereafter, oxidizes heme to p r o d u c e primarily (95%) α-biliverdin and a very small a m o u n t of δ-biliverdin (5%). Since the chimeric protein h a r b o r s the distal helix (and environment) of the δ-biliverdin producing pa-HO, the f o r m a t i o n of mostly α-biliverdin clearly eliminates the possibility that the regioselectivity of heme hydroxylation is determined by hydrogen b o n d i n g interactions in the distal pocket t h a t steer the hydroperoxide ligand t o w a r d α-meso c a r b o n . N M R spectroscopic studies conducted with « m - H O c h revealed in-plane disorder of the heme. However, the angle of r o t a t i o n , a l t h o u g h large, is n o t sufficient to place the δ-meso c a r b o n where it can be efficiently attacked by the F e m - O O H oxidizing species. Thus, the 5 % δ-biliverdin produced is likely a representation Met. Ions Life Sei. 2009, 6, 241-293

DUAL R O L E OF H E M E R E G A R D I N G C A R B O N M O N O X I D E nm-HO

281

ρ a-Η Ο

nm-HO chimera

Figure 17. The chimeric protein, »w-HOch (bottom) was constructed by replacing distal helix residues 107-142 (green) in nm-HO with the corresponding residues (112-147) in ρα-ΆΟ (purple). The proximal helix is shown in cyan. Adapted from [176].

of the population of molecules where in-plane heme rotation places the δ-meso carbon where it is susceptible to attack [176]. To understand the origin of in-plane disorder in the chimeric enzyme it is important to consider that replacement of the distal helix in «/«-HOch substitutes T y r l l 2 in nm-HO (see Figure 18B) with the equivalent residue in ^a-HO, P h e l l 7 (see Figure 18A). This eliminates a stabilizing H-bonding interaction between T y r l l 2 and heme propionate-7 in the «/«-HOch protein and "loosens" the heme, which is now able to undergo relatively large inplane rotations. To test this notion the authors eliminated the second hemepropionate-polypeptide interaction in nm-HO by replacing Lysl6 with Ala in the «/«-HOch to create the «;;?-HOch-K16A mutant [176]. This protein oxidizes heme to 60% a- and 40% δ-biliverdin, a mixture that suggested an increased population of the heme in-plane conformation that places the δ-meso carbon where it can be attacked by the F e m - O O H moiety. This conclusion was supported by N M R spectroscopic analysis of the «/«-HOch K16A mutant, where the heme was also found to be in a dynamic equilibrium between two in-plane conformations. One of these conformations makes the α-meso carbon susceptible to hydroxylation, whereas the other in-plane conformation presents the α-meso carbon to the Fe m -hydroxylating group. The information thus far indicated that absence of heme propionatepolypeptide interactions triggers the large in-plane heme disorder seen in the Met. Ions Life Sei. 2009, 6, 241-293

282

RIVERA and RODRIGUEZ

His127

Phem

Cyste

(A) pa-HO

(B) nm-HO

Figure 18. Crystal structures of ρα-ΆΟ (A) and nm-HO (B) highlighting the positions of key amino acids in the control of in-plane heme conformation, which in turn controls the regioselectivity of heme oxidation. Note that the in-plane heme orientation of the heme in the two enzymes is related to one another by ~ 90° in-plane rotation. The crystal structures of ρα-ΆΟ and nm-HO are Protein Data Bank entries 1SK7 and 1J77, respectively. Reproduced f r o m [176] by permission of the American Chemical Society, copyright (2005).

«/«-HOch protein. Consequently, if heme propionate-polypeptide interactions are paramount to direct the regioselectivity of heme oxidation by anchoring the heme in an appropriate in-plane conformation within the enzyme, it follows that introducing the key interactions seen in the crystal structure of ^a-HO into the chimeric protein should render a δ-hydroxylating chimera. Results obtained from probing this idea demonstrated that indeed, replacement of Met31 in nm-HO with the equivalent residue in /w-HO, Lys34 (see Figure 18), to make the «ni-HOch K16A/M31K double mutant, results in a recombinant enzyme that oxidizes heme to 95% δ-biliverdin. N M R analysis of this enzyme corroborated that the majority of molecules ( > 9 5 % ) harbor a heme with an in-plane heme conformation that presents the δ-meso carbon to the F e m - O O H moiety.

5.4.

The 1 H NMR Spectra of Cyanide-Inhibited Heme Oxygenase as a Diagnostic Tool of Heme Oxidation Regioselectivity

As has been pointed out above, bacterial HOs participate in the mining of heme-iron. By comparison, HOs in photosynthetic bacteria are involved in the production of biliverdin, which serves as a precursor in the biosynthesis Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

283

of c y a n o b a c t e r i a l light-harvesting phycobiliproteins a n d t h e p h o t o r e c e p t o r p h y t o c h r o m e s [179]. A l t h o u g h p h y t o c h r o m e s are traditionally k n o w n as biliprotein p h o t o r e c e p t o r s in p l a n t s a n d p h o t o s y n t h e t i c bacteria, these have been recently discovered in n o n - p h o t o s y n t h e t i c bacteria [180]. U n l i k e p l a n t

1

H Chemical shift

Figure 19. Downfield portions of the ' H N M R spectra of nm-HO (A), trf-HO (B), ρα-ΆΟ (C) and BphO (D). The presence of only one heme methyl resonance (3 methyl, or 3Me) is diagnostic of α-hydroxylating HOs, where the proximal histidineimidazole plane lies approximately parallel to the β-δ-meso axis. The same region of the ' H N M R spectrum of/w-HO, where the heme is rotated in-plane ~90° relative to the α-hydroxylating HOs shows three methyl resonances (5Me, lMe, and 8Me). The magnitude of the chemical shifts and the order of these resonances (5Me > l M e > 8Me) in the context of well developed theory [183] indicate that the proximal histidine-imidazole plane in ρα-ΆΟ lies parallel to the α-γ-meso axis. Reproduced from [182] by permission of the American Society for Biochemistry and Molecular Biology, copyright (2004). Met. Ions Life Sei. 2009, 6, 241-293

284

RIVERA and RODRIGUEZ

and cyanobacterial phytochromes, bacteriophytochromes (BphP) from nonphotosynthetic prokaryotes have been shown to utilize biliverdin as a chromophore [181]. Pseudomonas aeruginosa was among the first bacteria identified to harbor a gene coding for a BphP, which functions as a typical light-regulated histidine kinase. The BphP gene is located in an operon downstream from a BphO gene, which encodes for a putative heme oxygenase enzyme [182]. The product of the BphO gene, BphO, was shown to catalyze the degradation of heme to α-biliverdin, iron, and CO, demonstrating that it is indeed a heme oxygenase [182]. BphO was reconstituted with 13 C-labeled heme to assign and 13 C resonances from the heme binding site of cyanide inhibited BphO (BphO-CN). Observations made during the study of other cyanide-inhibited bacterial HOs established that the pattern of paramagnetically affected heme methyl resonances in the ' H N M R spectra of cyanide-inhibited H O enzymes (HO-CN) is diagnostic of the heme in-plane orientation, and hence regioselectivity of meso carbon oxidation [147,150]. For instance, Figures 19A and 19B show that only the heme methyl resonance from the major (3Me) and minor (3me) heme orientational isomers in «m-HO-CN [99] and cd-\IO-CN [150] is resolved from the diamagnetic region. The N M R spectra of the α-biliverdin producing human and rat H O - l - C N are essentially identical [148,149]. In comparison, the spectrum of/?a-HO-CN (Fig. 19C), where the heme is rotated in-plane ~ 100° relative to the heme in α-biliverdin producing HOs, exhibits three heme methyl resonances (5Me, l M e and 8Me) resolved from the diamagnetic region [147], Consequently, the fact that the : H N M R spectrum of BphO-CN exhibits only the 3Me resonance above 12ppm (Figure 19D) corroborates that BphO is an α-biliverdin producing H O [182]. Further, the order of the heme methyl resonances in BphO (3Me > 8Me > 5Me > lMe) indicates that the proximal His-imidazole plane forms an angle φ ~ 133° with respect to the molecular χ axis. This means that the proximal His-imidazole plane lies almost parallel to the β-δ-meso axis, as is the case for all α-biliverdin producing HOs of known structure (human [53] and rat [54] HO-1, cd AIΟ [56], and nm-HO [55]). Consequently, it is possible to conclude that the pattern of heme methyl resonances in the N M R spectrum of CN~-inhibited HOs is an effective diagnostic tool of heme oxidation regiochemistry [182].

6. CONCLUSION AND OUTLOOK The catalytic cycle of heme degradation by H O follows a complicated set of reactions that are not yet fully understood. Key to the process is the Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

285

generation of a ferric hydroperoxide intermediate in the active site of HO, which is at a crossroad of reactivity because hydroxylation of heme commits the process to heme degradation, whereas formation of compound I is a trademark of monooxygenase and peroxidase enzymes. Hence, the path followed by the ferric hydroperoxide intermediate has enormous repercussions on the biochemistry and biology of 0 2 activation at heme centers. Despite its importance, we have yet to attain detailed understanding of the determinants that allow H O to bind 0 2 , reduce it to F e m - 0 0 ~ , deliver a proton to form F e m - O O H and finally hydroxylate the heme, while avoiding being inhibited by the CO released from the process. A very large effort encompassing chemical, biochemical and theoretical approaches, as well as biophysical techniques such as X-ray crystallography, EPR, N M R , resonance R a m a n and other spectroscopies, have shed significant light on the process. Nevertheless a significant portion of the path remains uncharted. Among the challenges to overcome in the future is the study of the electronic structure of the highly reactive ferric peroxide and ferric hydroperoxide intermediates at ambient temperatures, because investigations with model heme complexes suggest that their electronic structures at physiologically relevant temperatures are distinct from those observed at the cryogenic temperatures used to trap and study them. In addition, recent investigations suggest that dynamic motions involving the hydrogen bonding network in the distal site of H O are significantly affected by the nature of the distal ligand and by the oxidation state of the heme-iron. Moreover, disruption of the hydrogen bonding network creates global conformational disorder of the enzyme in the μβ-ηιβ time scale, which is accompanied by a decrease in reactivity. Thus, in order to fully understand the control that enzyme dynamics exerts on reactivity of heme degradation it will be important to conduct studies aimed at comparing the polypeptide motions in distinct coordination and oxidation states with specific rates of ligand binding and 0 2 activation. Although challenging, investigations of this nature would not only generate important information to further understand the mechanism of 0 2 activation by H O including its inhibition by CO, but would also represent an unprecedented description of dynamic-reactivity relationships in heme containing proteins and enzymes.

ACKNOWLEDGMENTS The author's research reported in this manuscript was carried out with support from grants from the National Institutes of Health (GM 50503) and the National Science Foundation (MCB-0446326). Met. Ions Life Sei. 2009, 6, 241-293

286

RIVERA and RODRIGUEZ

ABBREVIATIONS BphP cd CPR cyt P450 ENDOR EPR FAD FMN FPR FRET H/D Hb Hb-CO h-HO-1 HO HSQC Mb MM NADPH nm «m-HOch NMR pa Por QM rc-CPMG r-HO-1 SPR

bacteriophytochrome Corynebacterium diphteriae cytrochrome P450 reductase cytochrome P450 electron nuclear double resonance electron paramagnetic resonance flavin adenine dinucleotide flavin mononucleotide NADPH-dependent ferredoxin reductase fluorescence resonance energy transfer hydrogen-deuterium hemoglobin carboxyhemoglobin human heme oxygenase-1 heme oxygenase heteronuclear single quantum coherence spectroscopy myoglobin molecular mechanics nicotinamide dinucleotide phosphate reduced Neisseria meningitidis chimeric nm-HO nuclear magnetic resonance Pseudomonas aeruginosa porphyrin quantum mechanics relaxation-compensated Carr-Purcell-Meiboom-Gill rat heme oxygenase-1 surface plasmon resonance

REFERENCES 1. The Merck Index, Ed. S. Budavari, M. J. O'Neil, A. Smith, P. E. Heckelman and J. F. Kinneary, Merck Research Laboratories, Rahway, 1996. 2. J. Raub, Carbon Monoxide, http://www.inchem.org/documents/ehc/ehc/ ehc213.htm. 3. ICPP, Climate Change 2001, The Scientific Basis. Contribution of Working Group 1 to the Third Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, New York, 2001. 4. M. G. Sanderson, Emission of Carbon Monoxide by Vegetation and Soils. A Literature Review and Recommendations for STOCHEM.

Met. Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF HEME REGARDING CARBON MONOXIDE

287

5. R. Tenhunen, H. S. Marver and R. Schmid, J. Biol. Chem., 1969, 244, 6388-6394. 6. M. P. Schmitt, J. Bacterial., 1997, 179, 838-845. 7. S. I. Beale and J. Cornejo, Arch. Biochem. Biophys., 1983, 227, 279-286. 8. S. I. Beale, Chem. Rev., 1991, 93, 785-802. 9. A. Ernst and J. D. Zibrak, New Eng. J. Med., 1998, 339, 1603-1608. 10. F. L. Rodkey, J. D. O'Neil, H. A. Collison and D. E. Uddin, Clin. Chem., 1974, 20, 83-84. 11. F. J. W. Roughton and R. C. Darling, Am. J. Physiol., 1944, 141, 17-31. 12. K. R. Hardy and S. R. Thorn, J. Toxicol. Clin. Toxicol., 1994, 32, 613-629. 13. L. R. Goldbaum, R. G. Ramirez and Κ. B. Absalon, Aviat. Space Envrion. Med., 1975, 46, 1289-1291. 14. J. R. Alonso, F. Cardellach, S. Lopez, J. Casademont and O. Miro, Pharm. Toxicol., 2003, 93, 142-146. 15. R. W. Estabrook, M. R. Franklin and A. G. Hildebrandt, Ann. NY Acad. Sei., 1970, 174, 218-232. 16. S. W. Ryter and L. E. Otterbein, Bioessays, 2004, 26, 270-280. 17. P. F. Mannaioni and Υ. E. Masini, Inflamm. Res., 2006, 55, 261-273. 18. L. Ε. Otterbein and Α. Μ. Κ. Choi, Am. J. Physiol. Lung Cell Mol. Physiol., 2000, 279, L1029-L1037. 19. S. W. Ryter, J. Alam and Α. Μ. K. Choi, Physiol. Rev., 2006, 86, 583-650. 20. R. Tenhunen, H. S. Marver and R. Schmid, Proc. Natl. Acad. Sei. USA, 1968, 61, 748-755. 21. C. Chauveau, D. Bouchet, J.-C. Roussel, P. Mathieu, C. Braudeau, J.-P. Soulillou, S. Iyer, R. Buelow and I. Anegon, Am. J. Transplant., 2002, 2, 581-592. 22. R. Buelow, S. G. Tullius and H. D. Yolk, Am. J. Transplant., 2001, 1, 313-315. 23. R. Motterlini, J. E. Clark, R. Foresti, P. Sarathchandra, Β. E. Mann and C. J. Green, Circ. Res., 2002, 90, E17-E23. 24. A. Vanacci, A. Di Felice, L. Giannini, C. Marzocca, S. Pierpaoli and G. Zagli, Inflamm. Res., 2004, 53, S9-S10. 25. Y. Guo, A. B. Stein, W. J. Wu, W. Tan, X. Zhu and Q. H. Li, Am. J. Physiol. Heart Circ. Physiol., 2004, 286, H1649-H1653. 26. C. Uzel and Μ. E. Conrad, Semin. Hematol., 1998, 35, 27-34. 27. R. Stocker, Y. Yamamoto, A. F. McDonagh, A. N. Glazer and Β. N. Ames, Science, 1987, 235, 1043-1046. 28. M. D. Maines, Annu. Rev. Pharmacol. Toxicol., 1997, 37, 517-554. 29. G. Marilena, Biochem. Mol. Med., 1997, 61, 136-142. 30. J. R. Chipperfield and C. Ratledge, BioMetals, 2000, 13, 165-168. 31. A. Wilks and M. P. Schmitt, J. Biol. Chem., 1998, 273, 837-841. 32. W. Zhu, D. J. Hunt, A. R. Richardson and I. Stojiljkovic, J. Bacteriol., 2000, 182, 439-447. 33. M. Ratliff, W. Zhu, R. Deshmukh, A. Wilks and I. Stojiljkovic, J. Bacteriol., 2001, 183, 6394-6403. 34. E. P. Skaar, A. H. Caspar and O. Schneewind, J. Biol. Chem., 2004, 279, 436^43. 35. M. L. Pendrak, M. P. Chao, S. S. Yan and D. D. Roberts, J. Biol. Chem., 2004, 279, 3426-3433.

Met. Ions Life Sei. 2009, 6, 241-293

288

R I V E R A and R O D R I G U E Z

36. J. Cornejo and S. I. Beale, Photosynth. Res., 1995, 51, 223-230. 37. T. Muramoto, T. Kohchi, A. Yokota, I. Hwang and Η. M. Goodman, Plant Cell, 1999, 11, 335-347. 38. C. T. Migita, X. Zhang and T. Yoshida, Eur. J. Biochem., 2003, 270, 687-698. 39. P. R. Ortiz de Montellano, Acc. Chem. Res., 1998, 31, 543-549. 40. P. R. Ortiz de Montellano, Curr. Opin. Chem. Biol., 2000, 4, 221-227. 41. P. R. Ortiz de Montellano and A. Wilks, Adv. Inorg. Chem., 2000, 51, 359^07. 42. P. R. Ortiz de Montellano and K. Auclair, in The Porphyrin Handbook, Ed. Κ. M. Kadish, Κ. M. Smith and R. Guilard, Academic Press, Elsevier Science, Amsterdam, 2003, pp. 183-210. 43. T. Yoshida and C. Taiko Migita, J. Inorg. Biochem., 2000, 82, 33-41. 44. S. Shibahara, T. Kitamuro and M. Takahashi, Antioxid Redox Signal, 2002, 4, 593-602. 45. S. Shibahara, Tohoku, J. Exp. Med., 2003, 200, 167-188. 46. Τ. Yoshida, Μ. Noguchi and G. Kikuchi, J. Biol. Chem., 1980, 255, 4418-4420. 47. T. Yoshida and G. Kikuchi, J. Biol. Chem., 1978, 253, 4230-4236. 48. Y. Liu and P. R. Ortiz de Montellano, J. Biol. Chem., 2000, 275, 5297-5307. 49. J. Wang and P. R. Ortiz de Montellano, J. Biol. Chem., 2003, 278, 20069-20076. 50. Y. Higashimoto, H. Sakamato, S. Hayashi, M. Sugishima, K. Fukuyama, A. G. Palmer and M. Noguchi, J. Biol. Chem., 2005, 280, 729-737. 51. A. Wang, Y. Zeng, H. Han, S. Weeratunga, Β. N. Morgan, P. Moenne-Loccoz, E. Schönbrunn and M. Rivera, Biochemistry, 2007, 46, 12198-12211. 52. U. A. Ochsner, P. J. Wilderman, A. I. Yasil and M. L. Yasil, Mol. Microbiol., 2002, 45, 1277-1287. 53. D. J. Schuller, A. Wilks, P. R. Ortiz de Montellano and T. L. Poulos, Nature Struct. Biol., 1999, 6, 860-867. 54. M. Sugishima, Y. Omata, Y. Kakuta, H. Sakamoto, M. Noguchi and K. Fukuyama, FEBS Lett., 2000, 471, 61-66. 55. D. J. Schuller, W. Zhu, I. Stojiljkovic, A. Wilks and T. L. Poulos, Biochemistry, 2001, 40, 11552-11558. 56. S. Hirotsu, G. C. Chu, M. Unno, D.-S. Lee, T. Yoshida, S.-Y. Park, Y. Shiro and M. Ikeda-Saito, J. Biol. Chem., 2004, 279, 11937-11947. 57. J. Friedman, L. Lad, H. Li, A. Wilks and T. L. Poulos, Biochemistry, 2004, 43, 5239-5245. 58. W. Zhu, A. Wilks and I. Stojiljkovic, J. Bacterial., 2000, 182, 6783-6790. 59. C. M. Bianchetti, L. Yi, S. W. Ragsdale and G. N. J. Phillips, J. Biol. Chem., 2007, 282, 37624-37631. 60. L. Yi and S. W. Ragsdale, J. Biol. Chem., 2007, 282, 21056-21067. 61. L. Lad, D. J. Schuller, Η. Shimizu, J. Friedman, H. Li, P. R. Ortiz de Montellano and T. L. Poulos, J. Biol. Chem., 2003, 278, 7834-7843. 62. Y. Li, R. T. Syvitski, K. Auclair, P. R. Ortiz de Montellano and G. N. La Mar, J. Am. Chem. Soc., 2003, 125, 13392-13403. 63. J. Sun, A. Wilks, P. R. Ortiz de Montellano and Τ. M. Loehr, Biochemistry, 1993, 32, 14151-14157. 64. J. Sun, T. Loehr, A. Wilks and P. R. Ortiz de Montellano, Biochemistry, 1994, 33, 13734-13740.

Met. Ions Life Sei. 2009, 6, 241-293

DUAL R O L E OF H E M E R E G A R D I N G C A R B O N M O N O X I D E

289

65. M. Ito-Maki, I. Kazunobu, Κ. M. Matera, M. Sato, M. Ikeda-Saito and T. Yoshida, Arch. Biochem. Biophys., 1995, 317, 253-258. 66. Y. Liu, L. K. Lightning, H. Huang, P. Moenne-Loccoz, D. J. Schuller, Τ. L. Poulos, Τ. M. Loehr and P. R. Ortiz de Montellano, J. Biol. Chem., 2000, 275, 34501-34507. 67. M. D. Suits, G. P. Pal, K. Nakatsu, A. Matte, M. Cygler and Z. Jia, Proc. Natl. Acad. Sei. USA, 2005, 102, 16955-16960. 68. M. D. Suits, N. Jaffer and Z. Jia, J. Biol. Chem., 2006, 281, 36776-36782. 69. C. T. Migita, Κ. M. Matera, M. Ikeda-Saito, J. S. Olson, H. Fujii, T. Yoshimura, H. Zhou and T. Yoshida, J. Biol. Chem., 1998, 273, 945-949. 70. A. Wilks and P. R. Ortiz de Montellano, J. Biol. Chem., 1993, 268, 22357-22362. 71. B. Meunier, Chem. Rev., 1992, 92, 1411-1456. 72. P. R. Ortiz de Montellano, Annu. Rev. Pharmacol. Toxicol., 1992, 32, 89-107. 73. Ν. K. King and A. M. Winfield, J. Biol. Chem., 1963, 238, 1520-1528. 74. K. Ishikawa, N. Takeuchi, S. Takahashi, K. Mansfield Matera, M. Sato, S. Shibahara, D. L. Rousseau, M. Ikeda-Saito and T. Yoshida, J. Biol. Chem., 1995, 270, 6345-6350. 75. G. C. Chu, K. Katakura, X. Zhang, T. Yoshida and M. Ikeda-Saito, J. Biol. Chem., 1999, 274, 21319-21325. 76. C. O. Damaso, R. A. Bunce, Μ. Y. Barybin, A. Wilks and M. Rivera, J. Am. Chem. Soc., 2005, 127, 17852-17853. 77. C. O. Damaso, N. D. Rubie, P. Moenne-Loccoz and M. Rivera, Inorg. Chem., 2004, 43, 8470-8478. 78. R. M. Davydov, T. Yoshida, M. Ikeda-Saito and Β. M. H o f f m a n , J. Am. Chem. Soc., 1999, 121, 10656-10657. 79. T. Kamachi and K. Yoshizawa, J. Am. Chem. Soc., 2005, 127, 10686-10692. 80. T. Matsui, S. H. Kim, H. Jin, Β. M. H o f f m a n and M. Ikeda-Saito, J. Am. Chem. Soc., 2006, 128, 1090-1091. 81. M. Sugishima, H. Sakamoto, Y. Higashimoto, Y. Omata, S. Hayashi, M. Noguchi and K. Fukuyama, J. Biol. Chem., 2002, 277, 45086^5090. 82. L. Koenigs Lightning, H. Huang, P. Moenne-Loccoz, Τ. M. Loehr, D. J. Schuller, Τ. L. Poulos and P. R. Ortiz de Montellano, J. Biol. Chem., 2001, 276, 10612-10619. 83. H. Fujii, X. Zhang, T. Tomita, M. Ikeda-Saito and T. Yoshida, J. Am. Chem. Soc., 2001, 123, 6475-6484. 84. R. Davydov, Y. K o f m a n , H. Fujii, T. Yoshida, M. Ikeda-Saito and Β. M. H o f f m a n , J. Am. Chem. Soc., 2002, 124, 1798-1808. 85. R. Davydov, T. Matsui, H. Fujii, M. Ikeda-Saito and Β. M. H o f f m a n , J. Am. Chem. Soc., 2003, 125, 16208-16209. 86. R. Davydov, I. D. G. Macdonald, Τ. M. Makris, S. G. Sligar and Β. M. H o f f m a n , J. Am. Chem. Soc., 1999, 121, 10654-10655. 87. R. Kappl, M. H ö h n Berlage, J. Hütterman, Ν. Bartlett and Μ. C. R. Symons, Biochim. Biophys. Acta, 1985, 827, 327-343. 88. R. Davydov, Τ. M. Makris, Y. K o f m a n , D. E. Werst, S. G. Sligar and Β. M. H o f f m a n , J. Am. Chem. Soc., 2001, 123, 1403-1415.

Met. Ions Life Sei. 2009, 6, 241-293

29C

RIVERA and RODRIGUEZ

89. R. Davydov, J. D. Satterlee, Η. Fujii, A. Sauer-Masarwa, D. H. Busch and Β. M. H o f f m a n , J. Am. Chem. Soc., 2003, 125, 16340-16346. 90. J. Friedman, L. Lad, R. Deshmukh, H. Li, A. Wilks and T. L. Poulos, J. Biol. Chem., 2003, 278, 34654-34659. 91. I. Schilichting, J. Berendzen, K. Chu, A. M. Stock, S. A. Maves, D. E. Benson, R. M. Sweet, D. Ringe, G. A. Petsko and S. G. Sligar, Science, 2000, 287, 1615-1622. 92. D. L. Harris and G. H. Loew, J. Am. Chem. Soc., 1998, 120, 8941-8948. J. Zheng, d. Wang, W. Thiel and S. Shaik, J. Am. Chem. Soc., 2006, 128, 93. 13204-13215. L. Lad, J. Wang, H. Li, J. Friedman, B. Bhaskar, P. R. Ortiz de Montellano and 94. T. L. Poulos, J. Mol. Biol., 2003, 330, 527-538. R. Υ. N. H o , G. Roelfes, B. L. Feringa and L. Que Jr., J. Am. Chem. Soc., 1999, 95. 121, 264-265. L. Lad, D. J. Schuller, Η. Shimizu, J. Friedman, H. Li, P. R. Ortiz de 96. Montellano and T. L. Poulos, J. Biol. Chem., 2003, 278, 7834-7843. R. T. Syvitski, Y. Li, K. Auclair, P. R. Ortiz de Montellano and G. N. La Mar, 97. J. Am. Chem. Soc., 2002, 124, 14296-14297. Y. Li, R. T. Syvitski, K. Auclair, A. Wilks, P. R. Ortiz de Montellano and G. N. 98. La Mar, J. Biol. Chem., 2002, 277, 33018-33031. Y. Li, R. T. Syvitski, G. C. Chu, M. Ikeda-Saito and G. N. La Mar, J. Biol. 99. Chem., 2003, 278, 6651-6663. Τ. K. Harris and A. S. Mildvan, Proteins: Struct. Funct. Genet., 1999, 35, 275-282. 100. 101. Y. Liu, P. Moenne-Loccoz, Τ. M. Loehr and P. R. Ortiz de Montellano, J. Biol. Chem., 1997, 272, 6909-6917. 102. Κ. M. Matera, S. Takahashi, H. Fujii, H. Zhou, K. Ishikawa, T. Yoshimura, D. L. Rousseau, T. Yoshida and M. Ikeda-Saito, J. Biol. Chem., 1996, 271, 6618-6624. 103. C. T. Migita, H. Fujii, Κ. M. Matera, S. Takahashi, H. Z h o u and T. Yoshida, Biochim. Biophys. Acta, 1999, 1432, 203-213. 104. L. Avila, Η. Huang, C. O. Damaso, S. Lu, P. Moenne-Loccoz and M. Rivera, J. Am. Chem. Soc., 2003, 125, 4103-4110. 105. R. Lemberg, Rev. Pure Appl. Chem., 1956, 6, 1-23. 106. S. Sano, T. Sano, I. Morishima, Y. Shiro and Y. Maeda, Proc. Natl. Acad. Sei. USA, 1986, 83, 531-535. 107. P. O'Carra and E. Colleran, FEBS Lett., 1969, 5, 295-298. H. Sakamoto, Y. Omata, G. Palmer and M. Noguchi, J. Biol. Chem., 1999, 274, 108. 18196-18200. T. Murakami, I. Morishima, M. Toshitaka, S.-i. Ozaki, I. Hara, H.-J. Yang and 109. Y. Watanabe, J. Am. Chem. Soc., 1999, 121, 2007-2011. J. C. Rodriguez and M. Rivera, Biochemistry, 1998, 37, 13082-13090. 110. J. K. Rice, I. M. Fearnley and P. D. Barker, Biochemistry, 1999, 16847-16856. 111. I. Morishima, H. Fujii and Y. Shiro, Inorg. Chem., 1995, 34, 1528-1535. 112. F. Tani, M. Matsu-ura, K. Aryanna, T. Setoyama, T. Shimada, S. Kobayashi, 113. T. Hayashi, T. Matsuo, Y. Hisaeda and Y. Naruta, Chem. Eur. J., 2003, 9, 862-870.

Ions Life Sei. 2009, 6, 241-293

DUAL R O L E OF H E M E R E G A R D I N G C A R B O N M O N O X I D E

291

114. T. Yoshida, M. Noguchi and G. Kikuchi, J. Biochem., 1980, 88, 557-563. 115. J. Friedman, Υ. T. Meharenna, A. Wilks and T. L. Poulos, J. Biol. Chem., 2007, 282, 1066-1071. 116. F. Draghi, A. E. Miele, C. Travaglini-Allocatelli, B. Yallone, M. Brunori, Q. H. Gibson and J. S. Olson, J. Biol. Chem., 2002, 277, 7509-7519. 117. S. J. Smerdon, G. G. Dodson and A. J. Wilkinson, Biochemistry, 1991, 30, 6252-6260. 118. B. A. Springer, K. D. Egeberg, S. G. Sligar, R. J. Rohlfs, A. J. Mathews and J. S. Olson, J. Biol. Chem., 1989, 264, 3057-3060. 119. B. A. Springer, S. G. Sligar, J. S. Olson and G. N. Phillips Jr., Chem. Rev., 1994, 94, 699-714. 120. J. C. Rodriguez, A. Wilks and M. Rivera, Biochemistry, 2006, 45, 4578-4592. 121. J. C. Rodriguez, Y. Zeng, A. Wilks and M. Rivera, J. Am. Chem. Soc., 2007, 129, 11730-11742. 122. M. Rivera, G. A. Caignan, Α. V. Astashkin, A. M. Raitsimring, Τ. K. Shokhireva and F. A. Walker, J. Am. Chem. Soc., 2002, 124, 6077-6089. 123. Μ. K. Safo, F. A. Walker, A. M. Raitsimring, W. P. Walters, D. P. Dolata, P. G. Debrunner and W. R. Scheidt, J. Am. Chem. Soc., 1994, 116, 7760-7770. 124. F. A. Walker, H. Nasri, I. Torowska-Tyrk, K. Mohanrao, C. T. Watson, Ν. V. Shkhirev, P. G. Debrunner and W. R. Scheidt, J. Am. Chem. Soc., 1996, 118, 12109-12118. 125. F. A. Walker, Coord. Chem. Rev., 1999, 186, 471-534. 126. G. Simonneaux, V. Schünemann, C. Morice, L. Carel, L. Toupet, H. Winkler, Α. X. Trautwein and F. A. Walker, J. Am. Chem. Soc., 2000, 122, 4366-4377. 127. F. A. Walker, in The Porphyrin Handbook, Ed. Κ. M. Kadish, Κ. M. Smith and R. Guilard, Academic Press, New York, 2000, pp. 81-183. 128. A. Ghosh, E. Gonzalez and T. Yangberg, J. Phys. Chem. B, 1999, 103, 1363-1367. 129. F. A. Walker and U. Simonis, in Biological Magnetic Resonance, Ed. L. J. Berliner and J. Reuben, Plenum Press, New York, 1993, pp. 133-274. 130. T. Ikeue, Y. Ohgo, S. Takashi, M. Nakamura, H. Fujii and M. Yokoyama, J. Am. Chem. Soc., 2000, 122, 4068^076. 131. T. Ikeue, Y. Ohgo, T. Saitoh, T. Yamaguchi and M. Nakamura, Inorg. Chem., 2001, 40, 3423-3434. 132. J. Mispelter, M. Momenteau and J. M. Lhoste, in Biological Magnetic Resonance, Ed. L. J. Berliner and J. Reuben, Plenum Press, New York, 1993, pp. 299-355. 133. M. Rivera, F. Qiu, R. A. Bunce and R. E. Stark, J. Biol. Inorg. Chem., 1999, 4, 87-98. 134. M. Rivera and Y. Zeng, J. Inorg. Biochem., 2005, 99, 337-354. 135. G. A. Caignan, R. Deshmukh, Y. Zeng, A. Wilks, R. A. Bunce and M. Rivera, J. Am. Chem. Soc., 2003, 125, 11842-11852. 136. M. Rivera and F. A. Walker, Anal. Biochem., 1995, 230, 295-302. 137. M. J. Rodriguez-Maranon, Q. Feng, R. E. Stark, S. P. White, X. Zhang, S. I. Foundling, Y. Rodriguez, C. L. Schilling III, R. A. Bunce and M. Rivera, Biochemistry, 1996, 35, 16378-16390.

Met. Ions Life Sei. 2009, 6, 241-293

R I V E R A and R O D R I G U E Z

292

138. M. Rivera and G. A. Caignan, Anal. Bioanal. Chem., 2004, 378, 1464-1483. 139. T. Ikeue, Y. Ohgo, T. Yamaguchi, M. Takahashi, M. Takeda and M. Nakamura, Angew. Chem. Int. Ed. Engl., 2001, 40, 2617-2620. 140. A. Ikezaki and M. Nakamura, Inorg. Chem., 2002, 41, 6225-6236. Y. Zeng, G. A. Caignan, R. A. Bunce, J. C. Rodriguez, A. Wilks and M. Rivera, 141. J. Am. Chem. Soc., 2005, 127, 9794-9807. T. Ikeue, T. Saitoh, T. Yamaguchi, Y. Ohgo, M. Nakamura, M. Takahashi and 142. M. Takeda, Chem. Commun., 2000, 1989-1990. T. Sakai, Y. Ohgo, T. Ikeue, M. Takahashi, M. Takeda and M. Nakamura, 143. J. Am. Chem. Soc., 2003, 125, 13028-13029. L. A. Yatsunyk and F. A. Walker, Inorg. Chem., 2004, 43, 757-777. 144. 145. G. N. La Mar, J. D. Satterlee and J. S. De Ropp, in The Porphyrin Handbook, Ed. Κ. M. Kadish, Κ. M. Smith and R. Guilard, Academic Press, 2000, pp. 185-297. 146. L.-H. Ma, Y. Liu, X. Zhang, T. Yoshida and G. N. La Mar, J. Am. Chem. Soc., 2006, 128, 6657-6668. 147. G. A. Caignan, R. Deshmukh, A. Wilks, Y. Zeng, H. Huang, P. MoenneLoccoz, R. A. Bunce, M. A. Eastman and M. Rivera, J. Am. Chem. Soc., 2002, 124, 14879-14892. 148. C. M. Gorst, A. Wilks, D. C. Yeh, P. R. Ortiz de Montellano and G. N. La Mar, J. Am. Chem. Soc., 1998, 120, 8875-8884. 149. G. Hernandez, A. Wilks, R. Paolesse, Κ. M. Smith, P. R. Ortiz de Montellano and G. N. La Mar, Biochemistry, 1994, 33, 6631-6641. 150. Y. Zeng, R. Deshmukh, G. A. Caignan, R. A. Bunce, M. Rivera and A. Wilks, Biochemistry, 2004, 43, 5222-5238. 151. Y. Li, R. T. Syvitski, K. Auclair, P. R. Ortiz de Montellano and G. N. La Mar, J. Am. Chem. Soc., 2003, 125, 13392-13403. 152. Y. Liu, X. Zhang, T. Yoshida and G. N. La Mar, Biochemistry, 2004, 43, 10112-10126. 153. Y. Liu, X. Zhang, T. Yoshida and G. N. La Mar, J. Am. Chem. Soc., 2005,127, 6409-6422. 154. J.-P. Simonato, J. Pecaut, L. Le Pape, J.-L. Oddou, C. Jeandey, M. Shang, W. R. Scheldt, J. Wojaczynski, S. Wolowiec, L. Latos-Grazynski and J.-C. Marchon, Inorg. Chem., 2000, 39, 3978-3987. 155. C. A. Reed, T. Mashiko, S. P. Bentley, Μ. E. Kastner, W. R. Scheidt, Κ. Spartalian and G. Lang, J. Am. Chem. Soc., 1979, 101, 2948-2958. 156. R.-J. Cheng, P.-Y. Chen, P.-R. Gau, C.-C. Chen and S.-M. Peng, J. Am. Chem. Soc., 1997, 119, 2563-2569. 157. Μ. K. Safo, G. P. Gupta, C. T. Watson, U. Simonis, F. A. Walker and W. R. Scheldt, J. Am. Chem. Soc., 1992, 114, 7066-7075. 158. G. N. La Mar, T. J. Bold and J. D. Satterlee, Biochim. Biophys. Acta, 1977, 498, 189-207. 159. G. Simmoneaux, F. Hindre and M. Le Plouzennec, Inorg. Chem., 1989, 28, 823-825. 160. P. K. Sharma, R. Kevorkiants, S. P. de Yisser, D. Kumar and S. Shaik, Angew. Chem. Int. Ed., 2004, 43, 1129-1132.

Ions Life Sei. 2009, 6, 241-293

DUAL ROLE OF H E M E R E G A R D I N G C A R B O N M O N O X I D E

293

161. D. K u m a r , S. P. de Yisser and S. Shaik, J. Am. Chem. Soc., 2005, 127, 8204-8213. 162. R. M. Daniel, R. V. Dunn, J. L. Finney and J. C. Smith, Annu. Rev. Biophys. Biomol. Struct., 2003, 32, 69-92. 163. P. K. Agarwal, Microbial Cell Factories, 2006, 5, 2-13. 164. Ε. Z. Eisenmesser, Ο. Millet, Y. Labeikovsky, D. M. Korzhnev, M. Wolf-Watz, D. A. Bosco, J. J. Skalicky, L. E. Kay and D. Kern, Nature, 2005, 438, 117-121. 165. M. Sugishima, H. Sakamato, Y. Kakuta, Y. Omata, S. Hayashi, M. Noguchi and K. Fukuyama, Biochemistry, 2002, 41, 7293-7300. 166. S. W. Englander, Annu. Rev. Biophys. Biomol. Struct., 2000, 29, 213-238. 167. H. Maity, W. K. Lim, J. N. Rumbley and S. W. Englander, Prot. Sei., 2003, 12, 153-160. 168. A. Hvidt and S. Ο. Nielsen, in Advances in Protein Chemistry, Ed. C. B. Anfinsen, M. L. Lanson, J. T. Edsall and F. M. Richards, Academic Press, New York, 1966, pp. 288-380. 169. Y. Bai, J. S. Milne, L. Mayne and S. W. Englander, Proteins: Struct. Funct. Genet., 1994, 20, 4-14. 170. A. G. Palmer, Curr. Opin. Struct. Biol., 1997, 7, 732-737. 171. L. E. Kay, Nat. Struct. Biol., 1998, 5, 513-517. 172. L. H. M a , Y. Liu, X. Zhang, T. Yoshida, K. C. Langry, Κ. M. Smith and G. N. La Mar, J. Am. Chem. Soc., 2006, 128, 6391-6399. 173. H. Chen, Y. Moreau, E. Derat and S. Shaik, J. Am. Chem. Soc., 2008, 130, 1953-1965. 174. M. Unno, T. Matsui, G. C. Chu, M. Couture, T. Yoshida, D. L. Rousseau, J. S. Olson and M. Ikeda-Saito, J. Biol. Chem., 2004, 279, 21055-21061. 175. H. Fujii, F. Zhang and T. Yoshida, J. Am. Chem. Soc., 2004, 126, 4 4 6 6 ^ 4 6 7 . 176. R. Deshmukh, Y. Zeng, L. M. Furci, H.-w. Huang, Β. N. Morgan, S. Sander, A. Alontaga, R. A. Bunce, P. Moenne-Loccoz, M. Rivera and A. Wilks, Biochemistry, 2005, 44, 13713-13723. 177. J. Wang, J. P. Evans, H. Ogura, G. N. La M a r and P. R. Ortiz de Montellano, Biochemistry, 2006, 45, 61-73. 178. H. Zhou, C. T. Migita, M. Sato, D. Sun, X. Zhang, M. Ikeda-Saito, H. Fujii and T. Yoshida, J. Am. Chem. Soc., 2000, 122, 8311-8312. 179. B. L. Montgomery and J. C. Lagarias, Trends in Plant Sei., 2002, 1360-1385. 180. S. J. Davis, A. Y. Vener and R. D. Vierstra, Science, 1999, 286, 2517-2520. 181. S.-H. Bhoo, S. J. Davis, J. Walker, B. Karniol and R. D. Viestra, Nature, 2001, 414, 776-779. 182. R. Wegele, R. Tasler, Y. Zeng, M. Rivera and N. Frankenberg-Dinkel, J. Biol. Chem., 2004, 279, 45791-45802. 183. Ν. V. Shokhirev and F. A. Walker, J. Biol. Inorg. Chem., 1998, 3, 581-594.

Met. Ions Life Sei. 2009, 6, 241-293

Met. Ions Life Sei. 2009, 6, 295-361

9 Copper-Carbon Bonds in Mechanistic and Structural Probing of Proteins as well as in Situations where Copper is a Catalytic or Receptor Site Heather R. Lucas and Kenneth D. Karlin Department of Chemistry, The Johns Hopkins University, 3400 N. Charles Street, Baltimore, MD 21218, USA

ABSTRACT 1. INTRODUCTION 2. BINUCLEAR COPPER PROTEINS 2.1. Coupled Binuclear Copper Proteins 2.1.1. Carbon Monoxide versus Dioxygen Binding to Copper(I) 2.1.2. Relationship of Overall Structure to Protein Function 2.1.3. Trends in Spectroscopic Properties of CarbonmonoxyHemocyanin and Tyrosinase 2.2. Noncoupled Binuclear Copper Proteins 2.2.1. Static Structure of the Catalytic Core of Peptidylglycine α-Hydroxylating Monooxygenase as Determined by X-Ray Crystallography 2.2.2. Active Site Probing of the Catalytic Core of Peptidylglycine α-Hydroxylating Monooxygenase and Dopamine ß-Monooxygenase through Carbon Monoxide Coordination

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00295

296 297 298 298 300 302 304 308

309

311

LUCAS and KARLIN

296

2.2.3.

Coordination of Isocyanide to C u M and C u H in the Catalytic Core of Peptidylglycine a-Hydroxylating Monooxygenase and Dopamine ß-Monooxygenase 3. H E T E R O B I M E T A L L I C C O P P E R - C O N T A I N I N G E N Z Y M E S 3.1. Heme-Copper Oxidases 3.1.1. Structural and Mechanistic Probing through Photochemical Methods 3.1.2. Proton Translocation Pathways 3.1.3. Mixed-Valent Heme-Copper Oxidases and Electron Transfer 3.1.4. Interaction of Cyanide 3.1.5. Nitric Oxide Reduction Capabilities of Cytochrome c Oxidase 3.2. Cu-Zn Superoxide Dismutase 3.3. Molybdenum-Copper Carbon Monoxide Dehydrogenase 4. N O N - B L U E C O P P E R OXIDASES 4.1. Copper Amine Oxidase 4.2. Galactose Oxidase 5. BLUE, G R E E N , A N D P U R P L E COPPER P R O T E I N S 5.1. Blue Electron Transfer Proteins 5.2. Multi-Copper Oxidases 5.3. Copper Enzymes in Denitrification 5.3.1. Nitrite Reductase 5.3.2. Nitrous Oxide Reductase 6. COPPER(I) R E C O G N I T I O N SITES OR R E C E P T O R S 6.1. Bacterial Copper Chaperone CusF 6.2. Copper-Ethylene Receptor 6.3. Copper Ion in an Olfactory Receptor Site? 7. M I S C E L L A N E O U S 7.1. Bleomycin 7.2. Copper-Alkyl Complexes from Biologically Derived Carbon Radicals 8. G E N E R A L C O N C L U S I O N S ACKNOWLEDGMENTS ABBREVIATIONS REFERENCES

315 317 317 318 323 324 327 329 330 332 334 334 337 337 337 338 340 340 342 344 344 345 345 346 346 349 349 350 351 352

ABSTRACT: While copper-carbon bonds are well appreciated in organometallic synthetic chemistry, such occurrences are less known in biological settings. By far, the greatest incidence of copper-carbon moieties is in bioinorganic research aimed at probing copper protein active site structure and mechanism; for example, carbon monoxide (CO) binding as a surrogate for 0 2 . Using infrared (IR) spectroscopy, CO coordination to cuprous sites has proven to be an extremely useful tool for determining active site copper ligation (e.g., donor atom number and type). The coupled (hemocyanin,

Met. Ions Life Sei. 2009, 6, 295-361

COPPER-CARBON BONDS

297

tyrosinase, catechol oxidase) and non-coupled (peptidylglycine α-hydroxylating monooxygenase, dopamine ß-monooxygenase) binuclear copper proteins as well as the hemecopper oxidases (HCOs) have been studied extensively via this method. In addition, environmental changes within the vicinity of the active site have been determined based on shifts in the CO stretching frequencies, such as for copper amine oxidases, nitrite reductases and again in the binuclear proteins and HCOs. In many situations, spectroscopic monitoring has provided kinetic and thermodynamic data on Cu : -CO formation and CO dissociation from copper(I); recently, processes occurring on a femtosecond timescale have been reported. Copper-cyano moieties have also been useful for obtaining insights into the active site structure and mechanisms of copper-zinc superoxide dismutase, azurin, nitrous oxide reductase, and multi-copper oxidases. Cyanide is a good ligand for both copper(I) and copper(II), therefore multiple physical-spectroscopic techniques can be applied. A more obvious occurrence of a "Cu-C" moiety was recently described for a CO dehydrogenase which contains a novel molybdenum-copper catalytic site. A bacterial copper chaperone (CusF) was recently established to have a novel d-π interaction comprised of copper(I) with the arene containing side-chain of a tryptophan amino acid residue. Meanwhile, good evidence exists that a plant receptor site (ETR1) utilizes copper(I) to sense ethylene, a growth hormone. A copper olfactory receptor has also been suggested. All of the above mentioned occurrences or uses of carbon-containing substrates and/or probes are reviewed and discussed within the framework of copper proteins and other relevant systems. KEYWORDS: carbon monoxide · copper proteins · copper receptors · cyanide · ethylene and cation-n-interactions · heterobimetallic (Fe, Zn, Mo) units · isocyanide

1.

INTRODUCTION

Even for most biochemical or inorganic researchers focusing on metals in biology, the bioinorganic chemistry of the copper-carbon bond will probably not easily come to mind. For example, the vast majority of proteins or enzymes with copper ion containing active sites deal either with electron transfer (e.g., azurin or plastocyanin), nitrogen oxide processing (for nitrite reduction to nitrogen monoxide ('NO) or nitrous oxide (N 2 0) reduction to N 2 ), or dioxygen processing. The latter include the 0 2 -carrier hemocyanins, coupled or uncoupled dicopper monoxygenases, copper oxidases which couple substrate oxidation/dehydrogenation to 0 2 -reduction to H 2 0 2 or water and others. In none of these very many cases are there carbon based active-site ligands; by contrast, carbon monoxide (CO) and cyanide (CN~) are natural metal-ligands in iron or iron-nickel hydrogenases (see Chapters 5, 6, and 7 in this volume). And for NO x and (^-processing proteins, the copper binding substrates are nitrogen oxide- or dioxygen-derived small molecules. It is the purpose of this review to survey the incidence of carbon-based metal ligands in copper bioinorganic chemistry. In fact, there is a substantial literature such that we will only highlight the field, while trying to provide details in many of the cases. The major area of occurrence of Cu-C bonds is in the use of chemical and/or spectroscopic probing of copper protein active Met. Ions Life Sei. 2009, 6, 295-361

298

LUCAS and KARLIN

sites. Such applications go back as far as 1919, when Craifaleanu observed that bubbling of CO through solutions of a highly colored hemocyanin (oxyform) led to a change to colorless [1]. This was again observed in 1922 by Dhere and Schneider, but they showed that hemocyanin (He) previously exposed to CO would recolor when exposed to air. More detailed insights obtained by Root in 1934 [1] revealed that both CO and 0 2 bind to hemocyanin in a two-copper to one small molecule stoichiometry, and that in sharp contrast to the behavior of hemoglobin, CO binds less strongly (X eq about 20 times smaller) to copper in He than does 0 2 . Thus, just as has been extensively employed for the interrogation of heme protein iron centers, CO, CN~ and even isocyanides (RNC) have been and are used for copper proteins. This is particularly true for the investigation of copper(I), the reduced redox partner of copper(II), in proteins processing nitrogen oxides and dioxygen. As is well known from inorganicorganometallic chemistry, these three carbon-based ligands are good π-acceptors, thus excellent ligands for low-valent copper(I). Cuprous ion is also well known to ligate to olefins and even arenes, and as such we will draw attention to copper proteins where Cu(I)-R' (R' = olefin or arene) interactions are intimately involved. As cuprous ion is a d 1 0 metal ion, there are few spectroscopic handles; for example, UV-vis and EPR spectroscopies do not apply. However, IR signatures for CO, CN~, and R N C tell a great deal about local environment. As an anionic ligand, cyanide is also a very good ligand for cupric ion, and as will be seen, various spectroscopic methods can then be utilized to interrogate the ligand field about Cu(II), thus obtaining insights into ligation at copper protein active sites, these being new entities in coordination chemistry. As for all of biological chemistry, structure and function are integrally intertwined. To fully comprehend enzyme reactivity and mechanisms, active site structural insights for both Cu(I) and Cu(II) are key. In fact, in one biological example, CO is produced as a product of the copper dioxygenase protein quercetinase, although it does not appear that a Cu-CO interaction ever occurs. However, very recent and exciting findings reveal that copper(I) binds carbon monoxide as the enzyme substrate in molybdenumcopper CO dehydrogenase ( M o C u C O D H ) , involving the oxidation of CO to C 0 2 . This system will be discussed in some detail (see Section 3.3).

2. 2.1.

BINUCLEAR COPPER PROTEINS Coupled Binuclear Copper Proteins

Hemocyanin, tyrosinase (Tyr), and catechol oxidase make up a class of proteins that have undergone study for over a century [1-3]. Each protein Met. Ions Life Sei. 2009, 6, 295-361

COPPER-CARBON BONDS

β β

j

299

«W ^ ' 1 Η Cu c u ' . n b

* Lt Jt ·® ?

·

9 »

-

„ 02

« « —* t J ü

C u L Cu» * «

ι Ρ-· β/

••L* V *

Figure 1. X-ray crystal structures of deoxy-hemocyanin (left) and oxy-hemocyanin (right) from Limulus II (horseshoe crab). Adapted from [4].

has a binuclear active site consisting of two copper centers (labeled C u A and Cu B ), independently coordinated by three imidazole histidines as shown in Figure 1. T h e close proximity (2.9-4.6 A) of C u A and C u B enables fast electron transfer (ET) for the two electron reduction of dioxygen by deoxy-Hc to f o r m oxy-Hc, best described as a side-on b o u n d μ - η 2 : η 2 peroxo dicopper(H) species [4,5]; there has never been any evidence for an initially f o r m e d single-copper ion adduct, C u - ( 0 2 ) . F o r oxy-Hc and other met (i.e., oxidized) enzyme forms, a strong magnetic interaction (—2/ > 1200 c m - 1 , H = — 2 / S i · S 2 ) exists between the C u A and C u B ions, which led to their designation by S o l o m o n and coworkers as Type III coupled binuclear proteins [6]. D u e to their contradictory functions, the n a t u r e of the related coupled binuclear proteins have baffled researchers since their discovery. H e is the dioxygen t r a n s p o r t protein in a r t h r o p o d s and mollusks, analogous to hemoglobin in vertebrates. Tyr is f o u n d in all aerobic organisms and is responsible for catalyzing the hydroxylation of m o n o p h e n o l s to orthodiphenols and the subsequent two-electron oxidation to or/Ao-quinones. As illustrated in Scheme 1, the first monoxygenase-type process is referred to as phenolase or cresolase activity and the second oxidase-type process is referred to as catecholase activity. Catechol oxidase is also involved in the p r o d u c t i o n of or/Ao-quinones but lacks the cresolase ability like that of Tyr. The p r o d u c t i o n of quinone p r o d u c t initiates the synthesis of melanin, which

phenolase activity tyrosinase

catecholase activity tyrosinase, catechol oxidase

Scheme 1. Met. Ions Life Sei. 2009, 6, 295-361

300

LUCAS and KARLIN

is responsible for the browning processes found in fruits, vegetables, and the skin of mammals [7]. Catechol oxidase is found only in plants and aids in their protection from pathogens or insects.

2.1.1.

Carbon Monoxide

versus Dioxygen

Binding to Copper ( I )

As mentioned, early work by Root established that only one small molecule, CO or 0 2 , binds per dicopper active site in deoxy-Hc [1], and it binds CO with a much lower affinity than 0 2 . Through isotopic 13 C and l s O studies, Fager and Alben confirmed that carbon monoxide coordinates in a terminal fashion to a single copper site rather than in a bridged coordination mode between both copper sites, like for 0 2 [8]. Extensions by van der Deen and Hoving later confirmed that CO linkage to copper was through the carbon atom [9]. Differences in the observed coordination abilities of the Cu A and Cu B sites has led to a plethora of research focusing on the chemical nature of the dicopper active site. Bourne and coworkers measured the kinetics and equilibrium binding constants of CO and 0 2 establishing the cooperative nature in which the multi-subunit He's bind 0 2 , but not carbon monoxide [10]. In establishing fundamental aspects of CO and 0 2 binding to copper(I) centers, recent studies by Karlin and coworkers have led to the determination of CO binding kinetics and equilibrium parameters for a series of synthetic mononuclear copper(I) compounds possessing tripodal tetradentate ligands [11]. The study of mononuclear copper(I) carbonyl species is very applicable for comparison to He since the CO coordination in Hc-CO involves only one copper ion (see Figure 2). As shown in Table 1, the CO binding constant CK co ) for [Cu I (tmpa)(Solv)] + is similar to that for deoxy-Hc [10,11]. The model compound data was further supplemented by transient absorbance laser flash photolysis experiments. By monitoring CO re-binding following photodissociation from [Cu I (tmpa)(CO)] + , the CO association rates (kCo)

Figure 2. Schematic of carbonmonoxy-hemocyanin (left) and a mononuclear synthetic carbonyl complex, [Cu : (tmpa)(CO)] + (right). Met. Ions Life Sei. 2009, 6, 295-361

COPPER-CARBON BONDS

301

ο λ;

00 00 Ο Ο — ' ι— ' ι Χ Χ , , m ίη \l

\ V V V"l II II-

Antibonding mostly metal

ν(*ν '

*

/ '

/ /

Bonding mostly ligand

CN"

Figure 1. Schematic molecular orbital energy level diagram for a complex [M(CN) 6 ] n ~ 6 . Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

368

Cyanide was also developed into a useful ligand-directed probe of the coordination site of Cu,Zn-superoxide dismutase employing both infrared and R a m a n spectroscopy [26]. Finally, cyanide has been successfully employed to inactivate molybdoenzymes, such as xanthine oxidase or aldehyde oxidase, producing the desulfo form of the active site. The so-called cyanolyzable sulfur bound to the molybdenum center reacts with cyanide to form thiocyanate, SCN~, which can be quantitatively determined [27].

2.

VANADIUM ENZYMES

2.1.

Vanadium in Biology. Structures and Functions

The biological function of vanadium is well established and the chemistry of the element is gaining considerable interest. The metal is an inherent part of enzymatic active sites, prominent examples are the vanadium-containing haloperoxidases and vanadium nitrogenase [28]. For both types of Vdependent enzymes there exist functional analogues in nature which are either more widely spread or more efficient, for example, the heme-containing haloperoxidases and the conventional Mo-dependent nitrogenases, respectively. One might ask how these enzyme systems evolved, and in particular, whether the vanadium-containing enzymes known today are retained functional analogues, which withstood evolutionary forces. The widespread physiological effects of vanadium are mainly attributed to the similarity of the vanadate(V) ions and phosphate ions ( P O ^ ) . Note that, depending on pH, also important differences exist between theses two anions with regard to charge, redox and coordination properties [28,29].

2.1.1.

Vanadium Haloperoxidases

Vanadium-containing haloperoxidases are enzymes isolated primarily from marine algae, although they also have been purified from other organisms [30]. These enzymes catalyze the two-electron oxidation of halide ions (X~) to the corresponding hypohalous acids, H O X (equation 6). H O X can further react with a broad range of acceptors to form a diversity of halogenated compounds [31]. Generally, the haloperoxidases are named after the most X " + H+ + H 2 0 2 -s· 0 2 + H 2 0 + H O X

(6)

electronegative halide they can oxidize, thus vanadium chloroperoxidase, in addition to chloride, will also transform bromide. Recently, two different Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

369

haloperoxidases, one specific for the oxidation of iodide, and the second that can oxidize both iodide and bromide, were obtained from the brown alga Laminaria digitata. The iodoperoxidase activity was significantly more efficient than the bromoperoxidase fraction in the oxidation of iodide. The two enzymes differed markedly in their molecular masses, trypsin digestion profiles, and immunological characteristics [32]. One of the best characterized V-dependent chloroperoxidases is the fungal enzyme isolated and purified from Curvularia inaequalis [33]. Its X-ray structure (2.1 A), in complex with azide, reveals that the ligand coordinates directly to the vanadium center. Furthermore, three non-protein oxygen ligands and one histidine nitrogen are bound. In the native state, vanadium is bound as hydrogen vanadate(V) in a trigonal bipyramidal geometry, with the metal coordinated to three oxygens in the equatorial plane, and HO~ (from water) and a nitrogen of histidine at the apical positions. The vanadium-containing bromoperoxidase from the seaweed Ascophyllum nodosum shows high similarities in the regions of the metal binding site, with all hydrogen vanadate(V) interacting residues conserved except for Lys353 which has been replaced by an asparagine. An interesting observation results from a trapped phosphate intermediate in the crystal structure (1.5 A) of vanadium-free apochloroperoxidase from Curvularia inaequalis which catalyzes a dephosphorylation reaction [34]. Since the chloroperoxidase is functionally and evolutionary related to several acid phosphatases including human glucose-6-phosphatase and a group of membrane-bound lipid phosphatases, the structure may help to understand the mechanism of action of these enzymes as well. The trapped intermediate is bound to the active site as metaphosphate, PO^, with its phosphorus atom covalently attached to the nitrogen atom of His496. An apical water molecule is within hydrogen-bonding distance to the phosphorus atom, and it is in a perfect position for a nucleophilic attack on the metaphosphate-histidine intermediate to form the inorganic phosphate. Recently, the 3D structure of a vanadium-dependent bromoperoxidase dodecamer from the red algae Corallina officinalis has been determined at 2.3 A resolution, with each subunit made up of 595 amino acid residues [35]. A cavity is formed by the Ν terminus of each subunit in the center of the dodecamer. The subunit fold and dimer organization of the bromoperoxidase are similar to those of the dimeric enzyme from the brown algae Ascophyllum nodosum, with which it shares 33% sequence identity. The different oligomeric states of the two algal enzymes seems to reflect separate mechanisms of adaptation to harsh environmental conditions and/or to chemically active substrates and products. The residues involved in the vanadate binding are conserved between the two bromoperoxidases and the vanadium chloroperoxidase from Curvularia inaequalis. However, most of the other residues forming the active-site cavity are different in the three Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

370

enzymes, which reflects differences in the substrate specificity and stereoselectivity of the reaction. The v a n a d a t e is directly linked to the protein only t h r o u g h the axially b o u n d histidine residue, and is embedded in the protein t h r o u g h an extensive h y d r o g e n - b o n d i n g network.

2.1.2.

Vanadium Nitrogenase

Nitrogenase, the microbial enzyme catalyzing biological nitrogen fixation accounts for the cycling of at least 9 χ 10 10 kg of nitrogen fixed annually [36]. All N 2 -fixing organisms (diazotrophs) which have been investigated have a nitrogenase system based on M o and Fe. In addition, it is n o w k n o w n that some organisms have alternative nitrogenases based on V and Fe, or apparently Fe alone [37]. U n d e r optimal conditions Mo-nitrogenase catalyzes the A T P - d e p e n d e n t reduction of dinitrogen to a m m o n i a and p r o t o n s to dihydrogen employing a t w o - c o m p o n e n t enzyme machinery (reaction 7). N 2 + 8H+ + 8e~ + 1 6 M g A T P -s· 2 N H 3 + H 2 + 1 6 M g A D P + 16P ;

(7)

V a n a d i u m was shown unequivocally to have a role in biological nitrogen fixation, when V-containing nitrogenases were isolated f r o m m u t a n t strains unable to synthesize Mo-nitrogenase [38,39]. T h e reality of a M o - i n d e p e n d e n t r o u t e for nitrogen fixation was established beyond question when the structural genes encoding Mo-nitrogenase in Azotobacter vinelandii were specifically removed and the resulting deletion m u t a n t strains shown to be able to fix N 2 in Mo-deficient m e d i u m [40]. Subsequently, V-nitrogenases were purified f r o m strains of Azotobacter chroococcum and Azotobacter vinelandii [37]. The V F e protein contains Ρ cluster redox centers and a catalytic F e V cofactor, in which v a n a d i u m is in a polynuclear cluster with iron, sulfur, and h o m o c i t r a t e with a chemical environment similar to m o l y b d e n u m in M o F e proteins [41]. T h e crystal structures of b o t h individual proteins of Mo-nitrogenase and the putative A D P - A 1 F 4 transition state complex of the two proteins have been determined [42,43]. The F e protein is a dimer which has a single [4Fe-4S] center ligated at the subunit interface and two nucleotide-binding sites, one on each subunit. T h e structures of the M o F e proteins have revealed an a 2 ß 2 - s u b u n i t structure in which each dimeric a ß - s u b u n i t pair binds the [8Fe-7S] P-cluster positioned at the subunit interface and the [7Fe-9S-homocitrate] F e M o cofactor center within the α-subunit. This i n f o r m a t i o n , together with comparative spectroscopic d a t a enabled a m o r e meaningful interpretation of d a t a obtained for V-nitrogenase, for which the structure has yet to come (Figure 2) [44]. Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

371

\

Figure 2. Stick-and-ball model o f the nitrogenase F e M o - c o f a c t o r including the μ 6 interstitial atom (blue) in the center o f the Fe-S moiety (Fe, S, and M o are shown as grey, yellow and orange spheres) [45]. In the current model o f the FeV-cofactor, vanadium would replace molybdenum [46].

The redox centers of V-nitrogenase have been investigated by E P R , M C D , Mössbauer, and X-ray absorption spectroscopies [47-50]. The spectroscopic data for V F e proteins are fully consistent with the presence of redox centers very similar to those of Mo-nitrogenase. An indication of the similarity of these systems is the ability of the components of V-nitrogenase to form fully functional hybrid nitrogenases with components of the Mo-nitrogenase [47]. V-nitrogenase of Azotobacter chroococcum has been reported to consume more A T P and to produce more dihydrogen by comparison with M o nitrogenase, the reasons and possible significance of these differences remain currently unclear [28]. The V-nitrogenase from Azotobacter chroococcum produces ammonia and traces of hydrazine [51], and V-nitrogenase also Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

372

exhibits an alkyne reductase activity as well as an isocyanide reductase activity [52-54],

2.2. 2.2.1.

Vanadium Enzymes and Their Interaction with Cyanide Vanadium Haloperoxidases

Vanadium haloperoxidases catalyze the oxidation of halide anions, C P , Br~ and P , by hydrogen peroxide (reaction 6). Recently, Pecoraro and coworkers suggested that the V-haloperoxidases may actually be peroxidases and that the well-known halogenation reactions may simply reflect the abundance of chloride in the marine environment [55]. The potentials for the oxidation of P , Br~ and C P span a range of 0.53-1.36 V (versus NHE), the potentials for the oxidation of the pseudohalide anions, such as CN~, OCN~, or SCN~, fall within or below this range, suggesting the vanadium haloperoxidases may also catalyze the transformation of these compounds. Indeed, cyanide ( E ° = +0.375 V) acts as an inhibitor of V-bromoperoxidase isolated from the marine alga Ascophyllum nodosum through preferential oxidation of CN~. Similarly, thiocyanate ( E ° = + 0.77 V) was reported to inhibit bromide peroxidation through preferential oxidation of thiocyanate over bromide. 13 C N M R studies of the oxidation of K 1 3 C N by H 2 0 2 catalyzed by V-bromoperoxidase showed the formation of several oxidized thiocyanate species, including the putative (unstable) dithiocyanate ether, hypothiocyanate, thiooxime, and bicarbonate [46,56]. Bromine K-edge EXAFS experiments on samples containing bromide and V-peroxidase, in buffer pH 8, were carried out, with biomimetic vanadium compounds carrying Br-V, Br-C(aliphatic), and Br-C(aromatic) bonds, as reference. From these studies it appears that bromide does not coordinate to the vanadium center of the peroxidase, but binds covalently to carbon, with an active site serine as possible candidate. In this series of experiments, two oxovanadium complexes, [V0(H 2 0) 2 (sal-L-Leu)] and [V0(H 2 0) 2 (5-Br-sal-Gly)] have been structurally characterized [46]. These complexes contain the water ligands in cis and trans positions to the oxo group, at V-OH 2 distances ranging from 2.008 to 2.228 A, and were used to model the apical electron density feature observed in the structures of fungal and algal V-haloperoxidases.

2.2.2.

Vanadium Nitrogenase

Mo-nitrogenase is able to reduce both CN~ and N^ to NH 3 . Because V-nitrogenase has been shown to release hydrazine, N 2 H 4 , in the course of Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

373

N 2 reduction it was chosen to investigate the conversion of cyanide and azide. Sensitive assay procedures were developed to monitor the production of either H C H O or C H 3 O H from H C N . Like Mo-nitrogenase, the vanadium enzyme suffered electron flux inhibition when interacting with CN~, but in contrast to Mo-nitrogenase, M g A T P hydrolysis was also inhibited by CN~ in the case of V-nitrogenase. At high concentrations of cyanide, the vanadium enzyme directed a significant percentage of electrons into the production of excess N H 3 . Under these experimental conditions, a substantial amount of formaldehyde (HCHO) but not methanol (CH 3 OH) was detected for the first time [57]. Azide inhibited both the total electron flux as well as MgATP hydrolysis in the case of V-nitrogenase but not Monitrogenase. V-nitrogenase, unlike Mo-nitrogenase, revealed no preference between the two electron reduction to N 2 -plus-NH 3 and the six-electron reduction to N 2 H 4 -plus-NH 3 . V-nitrogenase formed more excess N H 3 , but reduction of the N 2 produced by the two electron reduction of N ^ . Unlike Mo-nitrogenase, CO could not completely eliminate either cyanide or azide reduction by V-nitrogenase. CO did, however, eliminate the inhibition of both electron flux and MgATP hydrolysis by CN~, but not that caused by azide. These different responses to CO suggest different sites or modes of interaction for these two substrates with V-nitrogenase.

3. 3.1.

MANGANESE ENZYMES Manganese in Biology. Structures and Functions

Manganese can play many roles in biology ranging from acting as a simple Lewis acid catalyst to being an element that can transverse several oxidation states to carry out water oxidation [58,59]. Manganese is an essential constituent of the tetranuclear M n cluster that is involved in dioxygen production in photosynthetic plants, algae, and cyanobacteria. Water splitting is driven by the membrane pigment-complex known as photosystem ΙΙ-water oxidizing complex (PSII-WOC), or oxygen-evolving complex (PSII-OEC). The electrons and protons are ultimately used to store energy in the form of ATP and to reduce C 0 2 to carbohydrate. The complex is directly involved in the oxidation of water, it carries four M n ions embedded in the protein matrix [58]. Crystallographic investigations of cyanobacterial PSII-OEC provided several medium-resolution structures (3.8-3.2 A) that explained important features of the protein matrix and cofactors, but did not produce a highly resolved picture of the complex [60-63]. The most complete cyanobacterial photosystem II structure reported so far (3 A) showed locations of and interactions between 20 protein subunits and 77 cofactors per monomer. Assignment of 11 ß-carotenes yielded insights into electron and Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

374

energy transfer and photo-protection mechanisms in the reaction center and antenna subunits, and the structure provided further information about the Mn 4 Ca cluster, where oxidation of water takes place [64]. Nevertheless, many questions remain to be solved about the structure and mode of action of PSII-OEC as discussed most recently [65]. Manganese is the cofactor for superoxide dismutases, catalases and some peroxidases, which are all used for the detoxification of reactive oxygen species (ROS). An important property of manganese in its 2 + oxidation state, which has important biochemical consequences, is that it is a close but not exact surrogate of M g 2 + . M n 2 + with its relatively similar ionic radius can readily exchange with M g 2 + in most structural environments, and exhibits much of the labile, octahedral coordination chemistry. However, it can more easily accommodate the distortions in coordination geometry in progressing from the substrate-bound to the transition state and to the bound product. Consequently, M n 2 + in the active site of a Mg 2 + -enzyme often results in improved enzyme efficacy [10].

3.1.1.

Manganese Superoxide

Dismutase

There are relatively few characterized proteins having mononuclear M n sites. By far the best characterized mononuclear site is that in Mn-superoxide dismutase which catalyzes the disproportionation of superoxide ( O j ) to hydrogen peroxide and dioxygen (reaction 8). The overall reaction is a redox 2C>2 + 2H+

H202 + 02

(8)

process that involves the alternate oxidation and reduction of the catalytic active site metal. In the first step, superoxide binds to the resting Mn(III) enyzme, and is oxidized to dioxygen, with Mn(III) being reduced to Mn(II). In the second step, a second molecule of superoxide binds to the Mn(II)enzyme, and is reduced to hydrogen peroxide, with Mn(II) being reoxidized to the Mn(III) state. There are four different types of superoxide dismutases known, carrying Cu and Zn, Fe, Μη, or Ni in the active site. In prokaryotes, Mn-superoxide dismutase is most commonly found as a dimer in which two monomers come together at a highly conserved interface. The eukaryotic enzyme is usually tetrameric, formed when two of the prokaryotic-like dimers come together to form a dimer of dimers. Each subunit in the oligomer has one metalbinding active site, found at the junction of the two domains of which the monomer is composed. The crystal structures of several Mn-superoxide dismutases including the human enzyme have been determined [66,67]. Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

375

The metal site has an approximately trigonal bypyramidal structure with an equatorial plane of two histidine imidazoles and one aspartate carboxylate, and axially coordinated solvent and imidazole.

3.1.2.

Manganese

Catcilases

The Mn-catalases possess a dinuclear Mn active site and catalyse the disproportionation of hydrogen peroxide (reaction 9). Catalases play an important protective role by converting toxic hydrogen peroxide into dioxygen and water. In contrast to their heme-containing counterparts, which are ubiquitous in aerobic organisms, a broad range of microorganisms, living in anoxic, or close to anoxic environments, have Mn-catalases. A most notable property of this type of catalase appears to be its insensitivity to the classical heme poisons, cyanide and azide [68]. 2H202

2H20 + 02

(9)

Most of the work on Mn-catalases has focused on the enzymes isolated from Lactobacillus plantarum and Thermits thermophilic. The X-ray structure of Mn-catalase from Lactobacillus plantarum [69] revealed the dimanganese active site (Figure 3). The oxidized [Mn(III)Mn(III)] cluster is

Figure 3.

The active site of Lactobacillus

plantarum

Mn-catalase [69].

Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

376

bridged by two solvent molecules (oxo and hydroxo, respectively) together with a μ-l,3-bridging glutamate carboxylate and is embedded into a network of hydrogen bonds involving a tyrosine residue (Tyr42). Spectroscopic and structural studies indicate that disruption of the hydrogen-bonded network significantly perturbs the active site in the Y42F variant. This variant has less than 5% of the catalase activity and much higher K m for H 2 0 2 ( « 1.4 mM) at neutral pH than the wild-type enzyme, although the activity is slightly restored at high pH. The occurrence of μ-l,3-bridging carboxylates appears to be a universal feature of bimetalloproteins and complexes which perform two-electron or multi-electron chemistry. Bridging carboxylates probably serve a functional role beyond merely that of a passive structural bridge to bring the metals together. The dinuclear Mn center carries out a two-electron catalytic cycle, interconverting between reduced [Mn(II)Mn(II)] and oxidized [Mn(III)Mn(III)] states during turnover [70].

3.2. 3.2.1.

Manganese Enzymes and Their Interaction with Cyanide Manganese Superoxide Dismutase

Mn-superoxide dismutases are usually insensitive towards inhibition by cyanide, in fact that is the way of recognizing them. About three deacades ago, in studies on micronutrient interactions in plants, leaf extracts of Pisum sativum showed the presence of three electrophoretically distinct superoxide dismutases, two of which were inhibited by cyanide while the third one was insensitive towards cyanide [71]. The CN-resistant activity, as judged by its dependence on manganese, appeared to be rather a Mn-superoxide dismutase than a Fe-superoxide dismutase. Azide (N^), on the other hand, proved to be a competitive inhibitor of Mn-SOD, and it is believed to bind analogously to superoxide. As expected, the crystal structure of the azide derivative of Mn-SOD from Thermus thermophilus shows that azide coordinates directly to the metal without causing release of the metalcoordinated solvent molecule. Azide binding takes place without loss of protein ligands. The increased coordination number of the metal ion opens the His-Mn-His angle and the bent azide molecule across from the aspartate ligand. A similar arrangement was also observed for the Fe-superoxide dismutase from E. coli [72]. Note that in the case of the Cu,Zn-superoxide dismutase, cyanide, which acts as a competitive inhibitor of the enzyme, binds directly to the active site copper atom and replaces the bound water molecule. The coordination geometry of Cu(II) is partly altered with respect Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

377

to the uninhibited enzyme but its coordination number has not been increased [73,74]).

3.2.2.

Manganese

Catalase

The Mn-catalases were originally identified on the basis of their insensitivity towards azide and cyanide. However, kinetic studies have subsequently shown that these enzymes are inhibited by azide and other anions, albeit at much higher concentrations than required for inhibition of heme catalases [59]. In the case of azide, the inhibition is competitive, suggesting that azide and peroxide bind at the same metal site. EPR spectroscopy first revealed that Mn catalase contained a binuclear Mn center, and this technique has been applied extensively to characterize the electronic structure of three of the four known oxidation states (see Table III.Β in [70]). The distance between the Mn(II) varied between 3.31 A (F~) and ~ 3 . 7 A (unliganded, pH 7) for a variety of anions in a systematic way with size of the anion, consistent with binding to both Mn ions at a μ-bridging position. Anions, with two or more lone pairs of electrons per atom, exhibited evidence of binding to the bridging site in the Mn(II),Mn(II) oxidation state and of simultaneously inhibiting the catalase activity. Note that cyanide showed a special behavior in that it did not inhibit catalase activity of the unliganded enzyme and actually reversed inhibition caused by the inhibiting anions. HCN, most likely, was the active form involved in reactivation of catalase activity. Cyanide appears to bind to a terminal (non-bridging) site on one or both Mn(II) ions, as seen by the increase in the intermanganese distance. This terminal ligation preference has been ascribed to the linear sp-hybridized electronic structure of cyanide. Apparently, bridging anions exchanged more slowly than terminally ligated anions and thus inhibited effectively. The stimulation of activity by cyanide suggests that bridging anions can be displaced to a labile terminal coordination site on one Mn ion.

4. 4.1.

NON-HEME IRON ENZYMES Non-Heme Iron Enzymes. Structures and Functions

Non-heme iron enzymes perform a wide range of important biological functions involving dioxygen in parallel to those of heme enzymes [75]. Oxygen-activating enzymes with mononuclear non-heme iron active sites participate in many metabolically important reactions that have environmental, pharmaceutical, and medical significance. For example, catechol Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

378

dioxygenases and Rieske dioxygenases are involved in the degradation of aromatic molecules in the environment. Lipoxygenases oxidize unsaturated fatty acids into precursors of leukotrienes and lipoxins and are potential targets for antiinflammatory drugs. Enzymes like isopenicillin Ν synthase and deacetoxycephalosporin C synthase, an α-keto acid-dependent enzyme, are important in the biosynthesis of antibiotics such as penicillin and cephalosporin [76]. Most mononuclear non-heme iron enzymes contain iron ligated by oxygen and/or nitrogen ligands. Usually, these enzymes promote dioxygen activation, resulting in the formation of highly reactive ironperoxo (Fe m -OOH, F e m - 0 2 ) or iron-oxo ( F e I V = 0 or F e v = 0 ) oxidation catalysts. The flexible coordination environment of enzymes containing 2-His-l-carboxylate (N 2 0)-ligated iron leaves room for substrate as well as dioxygen activation to occur at the metal site [77]. There also exists a class of iron-containing metalloenzymes utilizing an oxygen-bridged binuclear non-heme iron cluster, and some members of this class have been extensively characterized. The first member of the oxygenactivating binuclear iron class to be recognized, Fe-containing ribonucleotide reductase, is not an oxygenase during its primary catalyzed reaction, which is the reduction of ribonucleotides at the active site on its large subunit (Rl). However, it does activate oxygen at a secondary active site on its small subunit (R2) in conjunction with the generation of a stable tyrosyl radical that is essential to the overall mechanism. The first true oxygenase of this class to be described was methane monooxygenase. The studies of this enzyme made it possible for the first time to compare the proposed oxygenactivating mechanism of P450 with that of a structurally dissimilar metallooxygenase. Recently, several other enzymes which are apparently similar to MMO, have been described in the literature [78].

4.1.1.

Enzymes Carrying the 2-His-l-Carboxylate Triad Structural Motif and Variants

Facial

Great progress has been made in recent years toward our understanding of mononuclear non-heme Fe(II) enzymes thanks to a rapidly increasing number of crystal structures available for this class of enzymes. They activate dioxygen with a common structural motif, the iron(II) center in these enzymes is invariably coordinated by three protein residues, two histidines and one aspartate or glutamate, constituting one face of an octahedron, a recurring motif referred to as the 2-His-l-carboxylate facial triad (Figure 4) [79]. The 2-His-l-carboxylate facial triad serves as an excellent monoanionic three-pronged platform for binding divalent metal ions. The three remaining sites on the opposite face of the octahedron are consequently available for exogenous ligands. In the as-isolated enzymes, these sites are usually Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

Figure 4.

379

Structure of the 2-His-l-carboxylate facial triad motif [75].

occupied by solvent molecules but can a c c o m m o d a t e b o t h substrate (or co-substrate) and 0 2 in later steps of the catalytic cycle. Despite the m a n y different t r a n s f o r m a t i o n s catalyzed, a general mechanistic pattern at the iron(II) center has emerged f r o m spectroscopic and crystallographic studies of the various enzymes in this family. T h e Fe(II) center is six-coordinate at the start of the catalytic cycle and relatively unreactive t o w a r d 0 2 . Subsequent substrate a n d / o r cofactor binding to the active site makes the metal center five-coordinate and increases its affinity for 0 2 . 0 2 binding then initiates the oxidative mechanism specific for each subclass. Thus, the metal center binds 0 2 only when substrate and cofactor(s) are present at the active site, thereby p r o m o t i n g strong coupling between the reduction of 0 2 and the oxidation of substrate. This structural motif thus allows the metal center to activate b o t h substrate and 0 2 and bring t h e m into close proximity for subsequent reaction, thereby accounting in large part for its versatility [80-82]. T h e application of the 2-His + Asp/ Glu Fe(II) binding motif d o c u m e n t s h o w n a t u r e can catalyze a r e m a r k a b l e variety of oxygen activation chemistries with one functional structural building block. The organization of the second sphere environment can also be tuned by the strategic placement of acid/base catalysts, hydrogen b o n d i n g partners, n o n - b o n d i n g substrates, electron-supplying metal clusters, and cofactors. However, by controlling the structural elements of the Fe(II) center and its local environment, the point at which the O - O bond breaks during the reaction cycle can be regulated [83]. Relatively recently, a new class of cysteinate sulfur-ligated n o n - h e m e iron enzymes emerged which includes nitrile hydratases, superoxide reductases, and peptide deformylases. Superoxide reductase shares the N 2 X triad (X = 0 , S) seen with the majority of n o n - h e m e iron enzymes, with a cysteinate (Figure 4) replacing the m o r e c o m m o n carboxylate residue, and two histidines replacing two of the waters. Nitrile hydratase and superoxide reductase have in c o m m o n an active site containing a cysteinate sulfur trans to the substrate binding site. T w o additional cysteinates ligate the iron of nitrile hydratase, completing one of the faces of an octahedron, with the Met. Ions Life Sei. 2009, 6, 363-393

380

SOSA-TORRES and KRONECK

remaining coordination sites occupied by two peptide amide nitrogens and either a hydroxide or N O [77].

4.1.2.

Protocatechuate

3,4-Dioxygenase

Many soil bacteria express dioxygenases that are involved in the oxidative aromatic ring cleavage of catechol and its derivatives such as 3,4dihydroxybenzoate. These are important enzymes in the aerobic biodegradation pathways that allow these organisms to derive their carbon and energy from aromatic hydrocarbons. Depending on the position of the cleaved double bond relative to the hydroxyl groups, catechol dioxygenases can be split into two families: the intradiol-cleaving catechol dioxygenases, which cleave the carbon-carbon bond of the enediol moiety, and the extradiol-cleaving catechol dioxygenases, which cleave adjacent to the enediol. Although these enzymes share similar substrates, the intradiol- and extradiol-cleaving enzymes exhibit near exclusivity in their oxidative cleavage products, suggesting that there are two different mechanisms for cleavage. Furthermore, intradiol-cleaving catechol dioxygenases use an [Fe III (His) 2 (Tyr) 2 ] active site, while extradiol-cleaving catechol dioxygenases contain a [M II (His) 2 (Asp/Glu)] active site, typically iron(II) but manganese(II) in a few cases [79]. Within this class of non-heme iron enzymes, perhaps the most extensively investigated is protocatechuate (3,4-dihydroxybenzoate) 3,4-dioxygenase. Crystallographic information is available for several complexes of the enzyme, including the as-isolated state, several enzyme-substrate complexes, and many complexes with inhibitors. The crystal structure of as-isolated protocatechuate (3,4-dihydroxybenzoate) 3,4-dioxygenase reveals a trigonal bipyramidal iron center, with four endogenous protein ligands (His460, His462, Tyr408, and Tyr447). The 5th coordination position, located in the trigonal plane, is occupied by a solvent-derived ligand [75,79].

4.1.3.

The Non-Heme Iron Center of the Oxygen-Evolving of Photosystem II

Complex

A common property of the photosynthetic bacterial center and that of photosystem II of green plants is the presence of a non-heme iron center located between the quinones Q A and Q B . A number of spectroscopic similarities and sequence homologies suggest that in PSII-OEC, as in bacteria, the linear quinone-non heme iron-quinone arrangement spanning approximately 18 A is conserved. Where the analogy deviates significantly is in the coordination and the properties of the iron. While in both systems Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

381

the metal is coordinated by four histidine residues, there are important differences in the 5th and 6th coordination positions. In bacteria, these positions are occupied by a bidentate glutamate residue [84], whereas in plants at least one of these positions is occupied by bicarbonate [85]. A number of carboxylate anions, or NO, can bind to the PSII-OEC non-heme iron center in competition with bicarbonate. In addition, the non-heme iron has been shown, unlike in its the bacterial reaction center homologue, to undergo redox changes between Fe(II) and Fe(III) [86]. Usually, in purple bacteria, as well as in algae and higher plants, the non-heme iron site appears to be in the reduced high-spin Fe(II) state, but can be oxidized to the high-spin Fe(III) state [87,88],

4.2. 4.2.1.

Non-Heme Iron Enzymes and Their Interaction with Cyanide Enzymes Carrying the 2-His-l-Carboxylate Triad Structural Motif and Variants

Facial

Superoxide reductases remove potentially toxic superoxide anions from anaerobic organisms without forming 0 2 as a side product (reaction 10). In contrast, the better known superoxide dismutases produce one equivalent 2H+ + e" + 0 2

H202

(10)

of 0 2 for every two molecules of 0 2 (reaction 8). In the mechanism by which superoxide reductase presumably reduces 0 2 , superoxide is proposed to bind to the reduced Fe(II) state of the enzyme at diffusion controlled rates ( > 10 9 M _ 1 S"1). The transfer of an electron from the Fe(II) center metal ion to the bound substrate via an innersphere pathway is then proposed to afford an Fe(III)-peroxide intermediate. Consistent with an innersphere pathway, exogenous ligands including azide, nitric oxide, and cyanide have been shown to bind to the iron site of superoxide reductase [89]. EPR and MCD studies provided evidence of azide, ferrocyanide, hydroxide, and cyanide binding via displacement of the glutamate ligand. For the first three ligands, ligand binding occurred with retention of the high-spin (S = 5/2) ground state, whereas cyanide binding resulted in a low-spin (S = 1/2) species, with g-values at 2.29, 2.25, 1.94. The ability to bind exogenous ligands to both the Fe(III) and the Fe(II) active sites is consistent with an innersphere mechanism for superoxide reduction. Most likely, superoxide binds at the vacant coordination site of the reduced enzyme coupled with electron transfer from iron to superoxide. Met. Ions Life Sei. 2009, 6,

363-393

SOSA-TORRES and KRONECK

382

The cysteine ligand, which is in trans position to the superoxide binding site, would play a crucial role in pushing electron density on to the iron in order to p r o m o t e Fe(II)-to-superoxide electron transfer, p r o d u c t dissociation f r o m the Fe(III)-(hydro)peroxo intermediate, or both. Superoxide reductase is inhibited by cyanide but n o t by azide. A possible mechanism for cyanide inhibition would involve cyanide coordinating to the open c o o r d i n a t i o n site of the catalytically active Fe(II) site. This would m a k e the F e site inaccessible to C>2· However, this does n o t explain why superoxide reductase is not inhibited by azide. A n o t h e r possibility is t h a t C N ~ inhibition involves the oxidized Fe(III) state. Experiments with model complexes showed that the redox properties of these complexes were dramatically altered by cyanide in c o m p a r i s o n to azide. Presumably, cyanide prevents the enzyme f r o m turning over by preventing the reduced, catalytically active Fe(II) state f r o m being regenerated. N o t e t h a t cyanide inhibits Fe-superoxide dismutase in a similar m a n n e r , by preventing the Fe(II) state f r o m being regenerated [77,90].

4.2.2.

Protocatechuate

3,4-Dioxygenase

P r o t o c a t e c h u a t e 3,4-dioxygenase utilizes a Fe(III) center to catalyze the aromatic ring cleavage of 3,4-dihydroxybenzoate by i n c o r p o r a t i o n of b o t h atoms of dioxygen to yield ß - c a r b o x y - c « , c « - m u c o n a t e . T h e crystal structures of several complexes of this enzyme with its substrate and with heterocyclic substrate analogs have been recently determined [91]. Complexation of the active site Fe(III) led to a dissociation of the endogenous axial tyrosinate ligand (Tyr447). A f t e r its release, Tyr447 becomes stabilized by hydrogen b o n d i n g and f o r m s the t o p of a small cavity adjacent to the C3-C4 b o n d of the substrate. The equatorial Fe(III) c o o r d i n a t i o n site within this cavity is unoccupied in the anaerobic enzyme-substrate complex but coordinates a solvent molecule in the complexes with the substrate analogs. In addition, ternary complexes with cyanide b o u n d could be obtained. This shows t h a t an 0 2 analogue can occupy the cavity and suggests t h a t electrophilic 0 2 attack on the substrate 3,4-dihydroxy-benzoate is initiated f r o m this site. Both the dissociation of the endogenous Tyr447 and the expansion of the iron c o o r d i n a t i o n sphere are novel features of the 3,4p r o t o c a t e c h u a t e 3,4-dioxygenase-substrate complex which a p p e a r to play essential roles in the activation of substrate for 0 2 attack.

4.2.3.

The Non-Heme Iron Center of the Oxygen-Evolving of Photosystem II

Complex

Recently, it was shown that cyanide binds at the n o n - h e m e iron site of the P S I I - O E C complex [92]. At lower p H , cyanide competed with N O for Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

383

binding to the iron (approximately 10 m M CN~). At higher concentration of cyanide (approximately 50-80 mM), cyanide modified the Q^ F e 2 + EPR signal at g = 1.82-1.9 to a new form ( g = 1.98). It was assumed that at least two cyanide anions could bind to this iron site. More recently, in the presence of N a C N (30-300mM), at p H 6.5, the reduced state Q A F e 2 + , produced by illumination at < 2 0 0 K , or by reduction in the dark with sodium dithionite, was characterized by a g = 1.98 EPR signal. This signal decayed with increasing p H above 6.5 and was almost absent at p H 8.1 and N a C N concentrations above 300 m M . Complementary to the disappearance of the g = 1.98 EPR signal with increasing p H or incubation time, a new EPR signal developed which could be assigned to the uncoupled semiquinone Q^. These high pH, high cyanide concentration effects are accompanied by the conversion of the characteristic F e 2 + (S = 2) Mössbauer doublet to a new one with parameters characteristic of an F e 2 + (S = 0) state which explains the loss of the magnetic coupling of Q^ with the iron center. A progressive binding of two or even three cyanide anions to the non-heme F e 2 + site is proposed to explain the spectroscopic data [88]. The conversion of the iron to a diamagnetic state by cyanide treatment should help to obtain further valuable structural and functional informations on the Q^ F e 2 + site as important component of the PSII-OEC.

5. 5.1.

ZINC ENZYMES Zinc in Biology. Structures and Functions

Zinc is essential for growth and development in all forms of life. It is found in more than 300 enzymes, where it plays both a catalytic and a structural role. It is the only metal to have representatives in each of the six fundamental classes of enzymes: (i) oxidoreductases like alcohol dehydrogenase and superoxide dismutase, (ii) transferases like R N A polymerase and aspartate transcarbamoylase, (iii) hydrolases like carboxypeptidase A and thermolysin, (iv) lyases like carbonic anhydrase and fructose-1,6-bisphosphate aldolase, (v) isomerases like phosphomannose isomerase, and (vi) ligases like pyruvate carboxylase and aminoacyl-tRNA synthases [10,93]. There are growing numbers of nucleic acid binding proteins depending on Z n 2 + , indicating that Z n 2 + is also widely involved in the regulation of the transcription and translation of the genetic message. Z n 2 + ion is redox inactive, as a consequence of its filled 3d 10 configuration it does not have d-d transitions, and therefore no absorption in the visible region. Zinc is able to adopt a highly flexible coordination geometry, however, in most zinc proteins there is a strong preference for tetrahedral Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

384

Χ, Υ, Ζ Y

\

y

/

Zn—o

\

OoC

HoN+

ϋ

Ν, Ο, S donors of His, Asp, Glu & Cys residues O?C

Η N-

\C(H)—CH2 / HoN+

His

Ο C^ OH

Asp

O?C

\C(H)—CH -SH

\

HoN+

Glu

Csy Figure 5.

^

C(H)—CH2—CH2—CQ / OH

2

Common structural features of zinc enzymes [94].

coordination, frequently slightly distorted, which enhances both the Lewis acidity of the metal center and the acidity of a coordinated water molecule. Since Zn is of borderline hardness, it will bind oxygen (Asp, Glu, H 2 0 ) , nitrogen (His) and sulfur (Cys) ligands (Figure 5).

5.1.1.

Zinc Carbonic

Anhydrase

Carbonic anhydrase has played the most pivotal role in the development of enzymology [95-97]. It was the first enzyme recognized to contain a mononuclear zinc site, and is one of the most efficient enzymes known. It also has widespread occurrence in prokaryotes and has therefore been classified as an "ancient" enzyme [98]. The essential physiological function of the enzyme is to catalyze the hydration of carbon dioxide, and it thus plays an important role in respiration and intracellular CC^/HCO^ equilibration (equation 11). C 0 2 + H 2 0 ^ H C 0 3 + H+

(11)

These enzymes are very efficient catalysts for the reversible hydration of carbon dioxide to bicarbonate, but at least the α-class enzyme possesses a Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

385

high versatility, being able to catalyze different other hydrolytic processes, such as (i) the hydration of cyanate to carbamic acid, (ii) the hydration of cyanamide to urea, (iii) the aldehyde hydration to gem-diols, (iv) the hydrolysis of carboxylic or sulfonic acid esters, and (v) other less investigated hydrolytic processes, such as hydrolysis of halogeno derivatives, arylsulfonyl halides, and other hydrolyzable substrates. It is not known whether reactions catalyzed by carbonic anhydrases other than the hydration of C 0 2 or dehydration of H C O ^ may have physiological relevance in organisms where these enzymes are present. The zinc ion is located at the bottom of a conical cavity (ca. 15 A deep) and is coordinated to the protein by three histidine residues and a water molecule (or hydroxide ion, depending upon pH) forming a tetrahedral site (Figure 6). Recent crystallographic studies on the carbonic anhydrase from Methanosarcina thermophila (prototype of the γ-class) revealed that the active site of this enzyme contained additional metal-bound water ligands, so the overall coordination geometry was trigonal bipyramidal [99]. The mechanism of action of carbonic anhydrase comprises three steps: (i) deprotonation of the coordinated water with a ρΚΆ κ 7 to give [(His) 3 ZnO H ] + , (ii) nucleophilic attack of the zinc-bound hydroxide at C 0 2 to give the I ICO, intermediate [ ( H i s ) 3 Z n - 0 C 0 2 H ] + , and (iii) displacement of the H C O ^ anion by H 2 0 to complete the catalytic cycle. Details concerned with the nature of the H C O ^ intermediate and how it is displaced from the metal center are still not fully understood [100,101]. To date, it is well established that carbonic anhydrase is widely distributed among phylogenetically and physiologically diverse prokaryotes, indicating a far greater role for this enzyme in nature than previously recognized. The comparison of sequences and crystal structures of the mammalian and plant enzymes demonstrates that they evolved independently and have

Figure 6.

Schematic drawing of the active site of carbonic anhydrase (α-class) [94].

Met. Ions Life Sei. 2009, 6, 363-393

386

S O S A - T O R R E S and K R O N E C K

been designated the α-class and ß-class, respectively. A third class, the γ-class carbonic anhydrase, was reported in 1994, with a typical representative isolated from the methanogenic archaeon Methanosarcina thermophila [102].

5.1.2.

Zinc Hydrolases

Much of the importance of zinc enzymes derives from their peptidase and amidase activity involving the cleavage of R C ( 0 ) - N H ( R ' ) amide bonds. For example, with respect to peptidase activity, zinc enzymes include both endopeptidases (cleaving peptides or proteins at positions within the chain) and exopeptidases (cleaving a terminal amino acid from the chain). Of the exopeptidases, zinc enzymes function as both carboxypeptidases (which remove a C-terminal amino acid) and aminopeptidases (which remove a N-terminal amino acid). Other examples of zinc enzymes that function by cleaving amide bonds are (i) ß-lactamases that destroy ß-lactams (such as penicillin) by hydrolyzing and cleaving the four-membered lactam ring and (ii) matrix metalloproteinases that degrade extracellular matrix components such as collagen. In addition to the cleavage of amide bonds, zinc enzymes play an important role in the cleavage of the P-OR bond in phosphates, [(R0)P0 3 ] 2 ~ and [(R0) 2 P0 2 ]~, as exemplified by their nuclease activity pertaining to the hydrolysis of D N A and R N A . Examples of enzymes that incorporate two zinc centers include metalloß-lactamases, aminopeptidases such as bovine lens leucine aminopeptidase, and alkaline phosphatases (Figure 7). There are also enzymes known that incorporate three zinc centers, such as phospholipase C and nuclease P I . Note that the third zinc centers of phospholipase C and nuclease PI are not directly associated with the binuclear zinc site [94].

Glu

Asp

Ηίδχ His

H2

r

His

^

Asp

OH,

/X His

ri2-Asp,„ / His

X = His, Cys

^His .Zn

\

Zn.1 0

H,

ti 2 -GIu

Lys,„_ I ^ O ^ l ,xAsp 'Zn Zn Asp''

N

/

N

Asp

O^Asp

Figure 7. Schematic structures of the active sites of dinuclear zinc enzymes: metalloß-lactamases and aminopeptidases [94]. Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

5.2. 5.2.1.

387

Zinc Enzymes and Their Interaction with Cyanide Zinc Carbonic

Anhydrase

Inhibitors and activators of the zinc enzyme carbonic anhydrase have a large n u m b e r of applications in therapy. M a n y types of such new derivatives have been reported recently, together with their potential applications as antiglaucoma, anticancer and antiosteoporosis agents or for the management of a variety of neurological disorders, a m o n g others. C a r b o n i c anhydrases are inhibited primarily by two m a i n classes of c o m p o u n d s : the metal complexing inorganic anions, such as cyanide, cyanate, thiocyanate, azide, hydrogen sulfide, and the unsubstituted sulfonamides. Sulfonamide inhibitors are useful as diuretics, or in the t r e a t m e n t and prevention of a variety of diseases. Inhibitors of b o t h types directly bind to the metal ion, either by substituting the Z n - b o u n d H 2 0 / 0 H ~ , or adding to the c o o r d i n a t i o n sphere, leading thus to p e n t a c o o r d i n a t e d zinc. In the structure of the iodide complex of h u m a n carbonic anhydrase, the inhibitor I~ replaced the f o u r t h coordinated H 2 0 / 0 H ~ ligand (Zn-I~ b o n d 2.7 A), whereas A u ( C N ) 2 was b o u n d in a different p a r t of the active site cavity and not directly to the Z n 2 + ion. The nitrogen a t o m of A u ( C N ) 2 is within h y d r o g e n - b o n d i n g distance of the zinc-bound H 2 0 / 0 H ~ g r o u p which shifts by a b o u t 0.4 A away f r o m the zinc ion in relation to its position in the native enzyme. It is proposed t h a t binding of the inhibitor A u ( C N ) 2 leads to a c o n f o r m a t i o n a l reorientation of the activity-linked g r o u p , due to hydrogenbond f o r m a t i o n with the inhibitor, which in t u r n sterically hinders the binding of the substrate C 0 2 molecule in the active site [103]. In a related study on the crystal structure of h u m a n carbonic anhydrase inhibited by cyanide and cyanate, it could be shown that the inhibitors replaced a water which f o r m e d a hydrogen b o n d to a peptide nitrogen ( T h r l 9 9 ) in the native structure. T h e c o o r d i n a t i o n of the Z n ion was hereby left unaltered, i.e., Z n 2 + coordinated three histidines and the H 2 0 / 0 H ~ g r o u p in a tetrahedral fashion. The binding site of the two inhibitors proved to be identical to w h a t earlier has been suggested to be the position of the substrate, C 0 2 , when attacked by the Z n - b o u n d hydroxyl ion. The peptide chain underwent n o significant alterations u p o n binding of either inhibitor [104], Finally, in a recent study, the enzyme f r o m the a r c h a e o n Methanosarcina thermophila was exposed to a large n u m b e r of anions including halides, C N " , H C 0 3 , CO^ , N O 2 , H S " , and H S 0 3 . The best Zn-carbonic anhydrase inhibitors were H S " and O C N " , whereas S C N " , N3~, COij , and HSO3" were weaker inhibitors. F o r the Co-substituted enzyme, the metal poisons were less effective, with cyanide possessing an inhibition constant of approximately 50 m M [105]. Met. Ions Life Sei. 2009, 6, 363-393

SOSA-TORRES and KRONECK

388

5.2.2.

Zinc Hydrolases

In experiments performed close to six decades ago, the effect of both zinc and magnesium ions on the activity of cyanide-inhibited (2 m M cyanide) kidney alkaline phosphatase was investigated. In low concentrations, Z n 2 + reactivated the inhibited enzyme whereas M g 2 + had little effect [106]. A soluble form of alkaline phosphatase purified from Walterinnesia aegyptia snake venom was recently purified and characterized. Zinc and cyanide ions, at concentrations of 1 5 m M and 10 mM, respectively, completely inhibited the activity of the phosphatase [107].

6.

CONCLUSIONS

The cyanide anion, CN~, and its transition metal complexes, are important compounds with interesting properties which turn them into valuable tools for both chemists and biologists. Based on the early chemistry of Prussian Blue, new cyano metal complexes have been designed which exhibit fascinating magnetic and electronic properties. The potential role of H C N as a precursor in prebiotic chemistry has been supported by the discovery that the hydrolytic products of its polymers including amino acids, purines, and orotic acid, a biosynthetic precursor of uracil, with the rapid formation of adenine by aqueous polymerization of H C N one of the key events in these studies. The cyanide anion is usually toxic for most aerobic organisms because of its inhibitory effects on respiratory enzymes on the one hand, but as a substrate it also represents an important form of carbon and nitrogen for many microorganisms, fungi, and plants. Finally, the cyanide anion is an important constituent of important metaldependent biomolecules, such as the hydrogenases and the cobalt site in vitamin B 12 .

ACKNOWLEDGMENTS This work was supported by Deutscher Akademischer Austauschdienst and Universidad Nacional Autönoma de Mexico (M.E.S.T.) and Deutsche Forschungsgemeinschaft (P.Μ.H.K).

ABBREVIATIONS ADP ATP

adenosine 5'-diphosphate adenosine 5'-triphosphate

Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn CD EPR EXAFS MCD MMO NHE NMR PSII QA,

ROS Sal SOD

QB

woe

389

circular d i c h r o i s m electron paramagnetic resonance extended X-ray absorption fine structure magnetic circular dichroism methane monooxygenase n o r m a l hydrogen electrode nuclear magnetic resoance photosystem II quinone centers A and Β r e a c t i v e o x y g e n species m o n o a n i o n of s a l i c y l a l d e h y d e superoxide dismutase water-oxidizing complex

REFERENCES 1. C. J. Knowles, Bacteriol. Rev., 1976, 40, 652-680. 2. B. Yennesland, Cyanide in Biology, Academic Press, London & New York, 1982. 3. Cyanide Compounds in Biology, Ciba Foundation Symposium 140, Ed. D. Evered and S. Harnett, John Wiley & Sons, Chichester, New York, Brisbane, Toronto, Singapore, 1988. 4. J. Woodward, Philos. Trans., 1724, 33, 15-17. 5. A. G. Sharpe, The Chemistry of Cyano Complexes of the Transition Metals, Academic Press, London, New York, San Francisco, 1976. 6. M. Shatruk, A. Dragulescu-Andrasi, Κ. E. Chambers, S. A. Stoian, E. L. Bominaar, C. Achim and K. R. Dunbar, J. Am. Chem. Soc., 2007, 129, 6104-6116. 7. Handbook on Metalloproteins, Ed. I. Bertini, A. Sigel and H. Sigel, Marcel Dekker, Inc., New York, Basel, 2001. 8. Handbook of Metalloproteins, Ed. A. Messerschmidt, R. Huber, T. Poulos and K. Wieghardt, Vols 1 and 2, John Wiley & Sons, Ltd., Chichester, New York, Weinheim, Brisbane, Singapore, Toronto, 2001. 9. Handbook of Metalloproteins, Ed. W. Bode, Y. Cygler and A. Messerschmidt, Vol. 3, John Wiley & Sons, Ltd., Chichester, New York, Weinheim, Brisbane, Singapore, Toronto, 2004. 10. R. C. Crichton, Biological Inorganic Chemistry. An Introduction, Elsevier, Amsterdam, 2008. 1 1 . A . Volbeda and J. C. Fontecilla-Camps, Coord. Chem. Rev., 2005, 249,1609-1619. 12. J. Alper, Science, 2003, 299, 1686-1687. 13. S. Reissmann, Ε. Hochleitner, Η. Wang, A. Paschos, F. Lottspeich, R. S. Glass and A. Böck, Science, 2003, 299, 1067-1070. 14. J. Orö and A. Lazcano-Araujo, in Cyanide in Biology, Β. Vennesland, Ε. Ε. Conn, C.J. Knowles, J. Westley and F. Wissing, (Ed.), Academic Press, London, U K , 1981, pp. 517-541. Met. Ions Life Sei. 2009, 6, 363-393

390

SOSA-TORRES and KRONECK

15. 16. 17. 18. 19.

R. A. Sanchez, J. P. Ferris and L. E. Orgel, Science, 1966, 154, 784-785. R. Shapiro, Proc. Natl. Acad. Sei. USA, 1999, 96, 4396-4401. A. Eschenmoser, Chemistry & Diversity, 2007, 4, 554-573. S. Ebbs, Curr. Opin. Biotechnol., 2004, 15, 231-236. M. Barclay, Y. A. Trett and C. J. Knowles, Enzyme Microb. Techno!., 1998, 23, 321-330. S. Yoshikawa and W. S. Caughey, J. Biol. Chem., 1990, 265, 7945-7958. W. P. Fehlhammer and M. Fritz, Chem. Rev., 1993, 93, 1243-1280. T. P. Hanusa, in: Encylopedia of Inorganic Chemistry, R. B. King, (Ed.), John Wiley & Sons, Ltd, Chichester, U K , 2005, pp. 1231-1241. A. S. Vinogradov, A. B. Preobrajenski, A. Knop-Gericke, S. L. Molodtsov, S. A. Krasnikov, S. V. Nekipelov, R. Szargan, M. Hävecker and R. Schlögl, J. Electron. Spectroscopy and Related Phenomena, 2001, 114-116, 813-818. Η. M. Goff, in: Iron Porphyrins, Part ΙΑ. B. P. Lever and H. B.Gray, (Ed.), Addison-Wesley Publishing Company, London, Amsterdam, Ontario, Sidney, Tokyo, 1983, pp. 237-281. G. Palmer, in: Iron Porphyrins, Part IIA. Β. P. Lever and H. B.Gray, (Ed.), Addison-Wesley Publishing Company, London, Amsterdam, Ontario, Sidney, Tokyo, 1983, pp. 43-88. J. Han, N. J. Blackburn and Τ. M. Loehr, Inorg. Chem., 1992, 31, 3223-3229. R. C. Wahl and Κ. V. Rajagopalan, J. Biol. Chem., 1982, 257, 1354-1359. D. C. Crans, J. J. Smee, E. Gaidamauskas and L. Yang, Chem. Rev., 2004, 104, 849-902. W. Plass, Angew. Chem. Int. Ed., 1999, 38, 909-912. A. Butler and J. V. Walker, Chem Rev., 1993, 93, 1937-1944. J. S. Martinez, G. L. Carroll, R. A. Tschirret-Guth, G. Altenhoff, R. D. Little and A. Butler, J. Am. Chem. Soc., 2001, 123, 3289-3294. C. Colin, C. Leblanc, E. Wagner, L. Delage, E. Leize-Wagner, A. Van Dorsselaer, B. Kloareg and P. Potin, J. Biol. Chem., 2003, 278, 23545-23552. A. Messerschmidt and R. Wever, Proc. Natl. Acad. Sei. USA, 1996, 93, 392-396. S. de Macedo-Ribeiro, R. Renirie, R. Wever and A. Messerschmidt, Biochemistry, 2008, 47, 929-934. Μ. N. Isupov, A. R. Dalby, Amanda A. Brindley, Y. Izumi, T. Tanabe, G. N. Murshudov and J. A. Littlechild, J. Mol. Biol., 2000, 299, 1035-1049. W. H. Schlesinger, Biogeochemistry: An Analysis of Global Change, Academic Press, San Diego, 1991. R. Eady, Coord. Chem. Rev., 2003, 237, 23-30. R. L. Robson, R. R. Eady, Τ. H. Richardson, R. W. Miller, M. Hawkins and J. R. Postgate, Nature, 1986, 322, 388-390. B. J. Hales, Ε. E. Case, J. E. Morningstar, M. F. Dzeda and L. A. Mauterer, Biochemistry, 1986, 25, 7251-7255. Β. E. Bishop, R. Premakumar, D. Dean, M. R. Jacobson, J. R. Chisnell, Τ. M. Rizzo and J. Jopczynski, Science, 1986, 232, 92-94. Β. K. Burgess and D. J. Lowe, Chem. Rev., 1996, 96, 2983-3012. J. B. Howard and D. C. Rees, Chem. Rev., 1996, 96, 2965-2982.

20. 21. 22. 23.

24.

25.

26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

391

43. R. L. Robson, P. R. Woodley, R. N. Pau and R. R. Eady, EMBO J., 1989, 8, 1217-1224. 44. J. M. Arber, Β. Ε. Dobson, R. R. Eady, P. Stevens, S. S. Hasnain, C. D. Garner and Β. E. Smith, Nature, 1987, 325, 372-374. 45. J. Kim and D. C. Rees, Science, 1992, 257, 1677-1682. 46. D. Rehder, C. Schulzke, H. Dau, C. Meinke, J. Hanss and M. Epple, J. Inorg. Biochem., 2000, 80, 115-121. 47. R. R. Eady, R. L. Robson, Τ. H. Richardson, R. W. Miller and M. Hawkins, Biochem. J., 1987, 244, 197-207. 48. J. E. Morningstar, Μ. K. Johnson, Ε. E. Case and Β. H. Hales, Biochemistry, 1987, 26, 1795-1800. 49. N. Ravi, Y. Moore, S. G. Lloyd, B. J. Hales and Β. H. Huynh, J. Biol. Chem., 1994, 269, 20920-20924. 50. G. N. George, C. L. Coyle, B. J. Hales and S. P. Cramer, J. Am. Chem. Soc., 1988, 110, 4057^059. 51. M. J. Dilworth and R. R. Eady, Biochem. J., 1991, 277, 465^68. 52. M. J. Dilworth, R. R. Eady, R. L. Robson and R. W. Miller, Nature, 1987, 327, 167-168. 53. K. Schneider, A. Müller, Ε. Krahn, W. R. Hägen, Η. Wassink and Κ. -H. Kuettel, Eur. J. Biochem., 1995, 230, 666-675. 54. M. Kelly, J. R. Postgate and R. L. Richards, Biochem. J., 1967, 102, 1-3C. 55. C. Slebodnick, B. J. Hamstra and Y. L. Pecoraro, Struct. Bonding, 1991, 89, 51-108. 56. D. Rehder, H. Dau, J. Dittmer, M. Epple, J. Hanss, C. Schulzke and H. Vilter, FEBS Lett., 1999, 457, 237-240. 57. K. Fisher, M. J. Dilworth and E. Newton, Biochemistry, 2006, 45, 4190^198. 58. C. F. Yocum and V. L. Pecararo, Curr. Opin. Biol. Chem., 1999, 3, 182-187. 59. A. J. Wu, J. E. Penner-Hahn and Y. L. Pecoraro, Chem. Rev., 2004, 104, 903-938. 60. A. Zouni, H. -T. Witt, J. Kern, P. Fromme, Ν. Krauss, W. Saenger and P. Orth, Nature, 2001, 409, 739-743. 61. N. Kamiya and J. R. Shen, Proc. Natl Acad. Sei. USA, 2003, 100, 98-103. 62. Κ. N. Ferreira, Τ. Μ. Iverson, Κ. Maghlaoui, J. Barber and S. Iwata, Science, 2004, 303, 1831-1838. 63. J. Biesiadka, B. Loll, J. Kern, K.-D. Irrgang and A. Zouni, Phys. Chem. Chem. Phys., 2004, 6, 4733-4736. 64. B. Loll, J. Kern, W. Saenger, A. Zouni and J. Biesiadka, Nature, 2005, 438, 1040-1044. 65. S. Zein, L. V. Kulik, J. Yano, J. Kern, A. Zouni, V. K. Yachandra, W. Lubitz, F. Neese and J. Messinger, Phil. Trans. Roy. Soc. London B, 2008, 363, 1167-1177. 66. G. E. Borgstahl, Η. Ε. Parge, Μ. J. Hickey, W. F. Beyer Jr., R. A. Halewell and J. A. Tainer, Cell, 1992, 71, 107-118. 67. A. F. Miller, Curr. Opin. Chem. Biol., 2004, 8, 162-168. 68. E. A. Delwhiche, J. Bacteriol., 1961, 81, 416-418.

Met. Ions Life Sei. 2009, 6, 363-393

392

SOSA-TORRES and KRONECK

69. V. V. Barynin, Μ. Μ. Whittaker, S. V. Antonyuk, Y. S. Lamzin, P. M. Harrison, P. J. Artymiuk and J. W. Whittaker, Structure, 2001, 9, 725-738. 70. G. C. Dismukes, Chem. Rev., 1996, 96, 2909-2926. 71. L. A. del Rio, F. Sevilla, Μ. Gomez, Y. Yanez and J. Lopez, Planta, 1978, 140, 221-225. 72. Μ. Ε. Stroupe, Μ. D i D o n a t o and J. A. Tainer, in: Handbook of Metalloproteins, A. Messerschmidt, R. Huber, T. Poulos and K. Wieghardt, (Ed.), Vol. 1, John Wiley & Sons, Ltd., Chichester, New York, Weinheim, Brisbane, Singapore, Toronto, 2001, pp. 941-951. 73. K. Djinovic Carugo, A. Battistoni, Μ. T. Carri, F. Polticelli, A. Desideri, G. Rotilio, A. Coda and M. Bolognesi, FEBS Lett., 1994, 349, 93-98. 74. D. Bordo, A. Pesce, M. Bolognesi, M. E. Stroppolo, M. Falconi and A. Desideri, in: Handbook of Metalloproteins, A. Messerschmidt, R. Huber, T. Poulos and K. Wieghardt, (Ed.), Vol. 2, John Wiley & Sons, Ltd., Chichester, New York, Weinheim, Brisbane, Singapore, Toronto, 2001, pp. 1284-1300. 75. Μ. Υ. M. Pau, J. D. Lipscomb and Ε. I. Solomon, Proc. Natl. Acad. Sei. USA, 2007, 104, 18355-18362. 76. L. Que Jr. and R. Υ. N. Ho, Chem. Rev., 1996, 96, 2607-2624. 77. J. A. Kovacs, Chem. Rev., 2004, 104, 825-848. 78. B. J. Wallar and J. D. Lipscomb, Chem. Rev., 1996, 96, 2625-2658. 79. M. Costas, Μ. M. Mehn, M. P. Jensen and L. Que Jr., Chem. Rev., 2004, 104, 939-986. 80. E. L. Hegg and L. Que Jr., Eur. J. Biochem., 1997, 250, 625-629. 81. L. Que Jr., Nature Struct. Biol., 2000, 7, 182-184. 82. K. D. Koehntop, J. P. Emerson and L. Que Jr., J. Biol. Inorg. Chem., 2005, 10, 87-93. 83. E. G. Kovaleva and J. D. Lipscomb, Nature Chem. Biol., 2008, 4, 186-193. 84. B. A. Diner and V. Petrouleas, Biochim. Biophys. Acta, 1987, 893, 138-148. 85. V. Petrouleas and B. A. Diner, FEBS Lett., 1982, 147, 111-114. 86. V. Petrouleas and B. A. Diner, Biochim. Biophys. Acta, 1987, 893, 126-137. 87. J. L. Zimmermann and A. W. Rutherford, Biochim. Biophys. Acta, 1986, 851, 416-423. 88. Y. Sanakis, V. Petrouleas and B. A. Diner, Biochemistry, 1994, 33, 9922-9928. 89. M. D. Clay, F. E. Jenney Jr., P. L. Hagedoorn, G. N. George, M. W. W. Adams and Μ. K. Johnson, J. Am. Chem. Soc., 2002, 124, 788-805. 90. S. Ozaki, J. Hirose and Y. Kidani, Inorg. Chem., 1988, 27, 3746-3751. 91. A. M. Orville, J. D. Lipscomb and D. H. Ohlendorf, Biochemistry, 1997, 36, 10052-10066. 92. D. Koulougliotis, T. Kostoupolos, V. Petrouleas and B. Diner, Biochim. Biophys. Acta, 1993, 1141, 275-282. 93. W. N. Lipscomb and N. Sträter, Chem. Rev., 1996, 96, 2375-2434. 94. G. Parkin, Chem. Rev., 2004, 104, 699-768. 95. D. W. Christianson and C. A. Fierke, Acc. Chem. Res., 1996, 29, 331-339. 96. D. W. Christianson and J. D. Cox, Ann. Rev. Biochem., 1999, 68, 33-57.

Met. Ions Life Sei. 2009, 6, 363-393

ENZYMES CONTAINING V, Μη, NON-HEME Fe, AND Zn

393

97. D. M. Duda and R. McKenna, in: Handbook of Metalloproteins, W. Bode, Y. Cygler, A. Messerschmidt, (Eds.), Vol. 3, John Wiley & Sons, Ltd., Chichester, New York, Weinheim, Brisbane, Singapore, Toronto, 2004, pp. 249-263. 98. K. S. Smith, C. Jakubzick, T. S. Whittam and J. G. Ferry, Proc. Natl. Acad. Sei. USA, 1999, 96, 15184-15189. 99. Τ. M. Iverson, Β. E. Alber, C. Kisker, J. G. Ferry and D. C. Rees, Biochemistry, 2000, 39, 9222-9231. 100. Κ. M. Merz Jr. and L. Banci, J. Am. Chem. Soc., 1997, 119, 863-871. 101. M. Bräuer, J. L. Perez-Lustres, J. Weston and E. Anders, Inorg. Chem., 2002, 41, 1454-1463. 102. B. C. Tripp, K. Smith and J. G. Ferry, J. Biol. Chem., 2001, 276, 48615^18618. 103. V. Kumar, Κ. K. Kannan and P. Sathyamurthi, Acta Cryst., 1994, D50, 731-738. 104. M. Lindahl, S. L. Anders and A. Liljas, Proteins: Structure, Function, and Genetics, 1993, 15, 177-182. 105. A. Innocenti, S. Zimmerman, J. G. Ferry, A. Scozzafava and C. T. Supuran, Bioorg. Med. Chem. Lett., 2004, 14, 3327-3331. 106. R. Hoare and G. E. Delory, Arch. Biochem. Biophys., 1955, 59, 465-472. 107. S. M. Al-Saleh and S. M. Saad, J. Natural Toxins, 2002, 11, 357-365.

Met. Ions Life Sei. 2009, 6, 363-393

Met. Ions Life Sei. 2009, 6, 395-416

11 The Reaction Mechanism of the Molybdenum Hydroxylase Xanthine Oxidoreductase: Evidence Against the Formation of Intermediates Having Metal-Carbon Bonds Russ

Hille

Department of Biochemistry, The University of California, Riverside CA 92521, USA

ABSTRACT 1. INTRODUCTION 2. ELECTRON-NUCLEAR DOUBLE RESONANCE STUDIES OF THE "VERY RAPID" SPECIES 2.1. ENDOR of the Intermediate Seen with Xanthine as Substrate 2.2. ENDOR of the Intermediate Seen with 2-Hydroxy-6methylpurine as Substrate 3. X-RAY CRYSTAL STRUCTURES RELEVANT TO THE REACTION MECHANISM 3.1. Alloxanthine-Complexed Xanthine Oxidoreductase 3.2. Xanthine Oxidoreductase in Complex with FYX-051 3.3. Xanthine Oxidoreductase Reacting with 2-Hydroxy-6methylpurine and Lumazine 3.4. Desulfo Xanthine Oxidoreductase in Complex with Xanthine 3.5. Substrate Orientation and the Basis of Enzyme Catalysis 3.5.1. Glu730 3.5.2. Glu232 3.5.3. Arg310 Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00395

396 396 400 400 401 404 404 405 406 409 410 410 412 413

396

HILLE

4. G E N E R A L C O N C L U S I O N S ACKNOWLEDGMENTS ABBREVIATIONS A N D D E F I N I T I O N S REFERENCES

413 414 414 414

ABSTRACT: ENDOR spectra of the catalytically relevant "very rapid" Mo(V) species generated in the course of the reaction of xanthine oxidoreductase with substrate have been examined by two different groups. While the data themselves are virtually identical, the analysis has been variously interpreted as supporting or refuting the existence of a molybdenum-carbon bond in the signal-giving species. While the basis for this difference in interpretation has now been generally agreed upon - the Mo-C distance in the signal-giving species is now understood to be too long to represent a direct Mo-C bond - independent information concerning the structure of the signal-giving species is highly desirable. Recently, several X-ray crystal structures of catalytically relevant complexes of the enzyme with several substrates and inhibitors have been reported. Taken together, these structures strongly and unambiguously support the interpretation that the intermediate giving rise to the "very rapid" EPR signal, as well as the Mo(IV) intermediate that precedes it in the reaction mechanism, has product coordinated to the active site molybdenum via the catalytically introduced hydroxyl group in a simple "end-on" fashion, with no metal-carbon bond character to the complex. The manner in which product is bound and its orientation within the active site provide important clues as to the specific catalytic roles of active sites in accelerating the reaction rate. KEYWORDS: catalysis · molybdenum hydroxylase · reaction mechanism · xanthine oxidoreductase

1.

INTRODUCTION

The enzyme xanthine oxidoreductase catalyzes the final two steps of purine metabolism in most organisms (including humans), oxidatively hydroxylating hypoxanthine to xanthine and xanthine on to uric acid. The enzyme is most frequently encountered as a dehydrogenase that utilizes N A D + to remove the reducing equivalents taken from substrate in the course of hydroxylation, but is also found as an oxidase that utilizes 0 2 rather than N A D + [1]. The enzyme from cow's milk (typically isolated as the oxidase) was first purified by Dixon and Thurlow in 1924 [2] and since then has been the subject of intensive investigation using a variety of sophisticated physicochemical methods, as well as both steady-state and rapid reaction kinetic methods. It was the first system to be examined using freeze-quench methodologies, following the formation and decay of paramagnetic Mo(V) intermediates formed in the course of the reaction with xanthine by electron paramagnetic resonance spectroscopy (EPR) [3,4]. This work identified several discrete Mo(V) species, ultimately designated "very rapid", "rapid", and "slow" on the basis of the kinetics of their formation and decay in the Met. Ions Life Sei. 2009, 6, 395-416

M O L Y B D E N U M H Y D R O X Y L A S E XANTHINE O X I D O R E D U C T A S E

397

course of the reaction of enzyme with xanthine [5]. The "slow" EPR signal was soon shown to arise from an inactive form of the enzyme lacking a labile sulfur at the molybdenum center (see below) [6]. The "rapid" species was subsequently shown to arise from a complex of partially reduced enzyme (formed by prior turnover) with substrate [7] and as such represented a paramagnetic analog of the E*S Michaelis complex rather than a bona fide catalytic intermediate. It was immediately recognized that the "very rapid" species likely represented an authentic catalytic intermediate, and beginning in the early 1980's Bray and coworkers embarked on an elegant series of isotopic substitution experiments aimed at elucidating the structure of the signal-giving species, and by implication the reaction mechanism. Using 8-[ 2 H]-xanthine, it was shown that the C8 proton of substrate was initially transferred to a strongly magnetically coupled site of the molybdenum center, but was rapidly lost from this site by exchange with solvent [8]. Using 8-[ 13 C]xanthine, observation of a small superhyperfine coupling demonstrated that the purine nucleus was an integral component of the signal-giving species [9]. It had been shown that treatment of enzyme with cyanide removed a catalytically essential sulfur from the molybdenum center (as thiocyanate) in a manner that could be reversed by incubation of the enzyme with sulfide under appropriate conditions [10], and 33 S incorporated into the enzyme using Na 2 3 3 S. The "very rapid"species generated with enzyme thus labeled exhibited strong and anisotropic coupling to the 33 S nucleus and it was concluded that the sulfur was likely present as a terminal M o = S group in the active site molybdenum center [11,12]. This conclusion was subsequently confirmed by X-ray absorption spectroscopy (XAS) [13,14], which established the sulfur to be present at a distance of 2.15 A, consistent with a M o = S formulation. The XAS analysis also identified a M o = 0 at 1.68 A, consistent with a terminal M o = 0 group. Using EPR, a detailed analysis of the hyperfine coupling of 9 5 Mo- and 9 7 Mo-labeled enzyme (which constituted the initial evidence that molybdenum was an integral component of the enzyme active site [15]) was also interpreted in the context of a MoOS core [16]. In other isotope-labeling work, enzyme turnover in 1 7 0-labeled water yielded a "very rapid" EPR signal exhibiting strong and isotropic coupling [17]. Presciently, it was concluded that this strongly coupled oxygen was not likely to be the M o = 0 group identified by XAS, but rather the bridging oxygen of a M o - O R moiety that represented hydroxylated product coordinated to the molybdenum via the catalytically introduced oxygen atom (itself ultimately derived from solvent, as had been established previously [18]). Subsequent electron-nuclear double resonance ( E N D O R ) work established that apart from the M o - O R oxygen, no other oxygen in the metal center, including the M o = 0 group, exchanged rapidly with solvent under turnover conditions [19]. Met. Ions Life Sei. 2009, 6, 395-416

398

HILLE

The basic elements of the active site molybdenum center established by EPR and XAS studies have been confirmed as crystal structures have become available, first of the closely related aldehyde oxidoreductase from Desulfovibrio gigas [20,21] and subsequently by the xanthine oxidoreductases from Bos taurus [22] and Rhodobacter capsulatus [23]. These structures have established the molybdenum coordination sphere of oxidized enzyme as LMoOS(OH) in a distorted square-pyramidal geometry (Figure 1). L here represents a pyranopterin cofactor coordinated to the molybdenum via an enedithiolate side chain, as shown. (Although this organic component of the metal center is frequently referred to as molybdopterin, this terminology is confusing on at least two counts: first, the term refers to the organic component alone, excluding the metal connoted by the prefix; and second, the same cofactor is found in tungsten-containing enzymes.) The M o = S group was originally assigned in the apical position of the molybdenum coordination sphere, which was surprising in light of the known structures of simple MoO-containing inorganic compounds. Subsequent work with the bovine enzyme [24] and related enzymes such as carbon monoxide dehydrogenase [25] and quinoline-2-oxidoreductase [26] has clearly established that it is the M o = 0 rather than M o = S that occupies the apical position. This assignment is consistent with the considerable literature on oxomolybdenum complexes [27,28]. In the protein, the M o - O H ligand points into the solvent access channel toward the substrate binding site, a point of mechanistic significance as discussed below. With the active site of the oxidized enzyme established, other mechanistic information could next be incorporated to give a comprehensive picture of

Figure 1. Coordination geometry of the active site molybdenum center of XOR {top) and of the pyranopterin cofactor common to both molybdenum- and tungstencontaining enzymes (bottom). Met. Ions Life Sei. 2009, 6, 395-416

M O L Y B D E N U M H Y D R O X Y L A S E XANTHINE O X I D O R E D U C T A S E

399

the catalytic sequence. Most importantly, while it had long been known that the oxygen atom incorporated into product is ultimately derived from solvent [19], single-turnover experiments demonstrated that it is a catalytically labile site on the enzyme that is the proximal oxygen atom donor in the course of reaction, regenerated by oxygen from solvent at the completion of reaction [29]. Although originally thought to be the M o = 0 identified by XAS, it was subsequently established that a single turnover in 1 7 0-labeled water incorporates 1 7 0 into a strongly and anisotropically coupled site in the molybdenum coordination sphere [30]. By analogy to EPR studies of isotopically labeled model compounds [31,32], it could be concluded that it is the M o - O H rather than M o = 0 that represents the catalytically labile oxygen. This interpretation was subsequently substantiated when the reported crystal structures showed the M o - O H ligand pointing directly at the substrate binding site in the protein, as discussed above. Other mechanistic information comes from the p H dependence of the reaction of enzyme with xanthine, where it has been shown that both the steady-state parameter kQ!ii\Km and rapid reaction parameter kred/Kd (each reflecting the reaction of free enzyme with free substrate in the low-[S] regime), exhibit a bell-shaped p H dependence with ρΚΆ values of 6.6 and 7.4 from the acid and alkaline limbs of the curve, respectively [33]. The latter value agrees well with that for the ionization of neutral xanthine to the monoanion; the former ρΚΆ has been assigned to a universally conserved glutamate residue in the active site that is thought to act as an active site base [21,34]. Finally, use has been made of the slow substrate 2-hydroxy-6methylpurine, which in contrast to xanthine forms copious amounts of the "very rapid" EPR signal (and on a tens of seconds rather than milliseconds time scale [35]), to understand the kinetics of the reaction. Single-turnover experiments with this substrate have demonstrated that the reaction proceeds in three kinetic phases: the formation of a species having a difference maximum relative to oxidized enzyme at 470 nm, decay of this first species to one having a difference maximum at 540 nm, and finally the decay of this species back to oxidized enzyme (the experiment having been performed under aerobic conditions) [36]. The formation and decay of the 540 nmabsorbing species correlates well with formation and decay of the "very rapid" EPR signal, demonstrating that formation of the 540 nm-absorbing species is an oxidative event: an electron is lost from the enzyme in forming this species from the 470 nm-absorbing species that precedes it [36]. It can thus be concluded that the 470 nm-absorbing species represents a Mo(IV)product complex, which subsequently decays to the Mo(V)-product species giving rise to the "very rapid" EPR signal. Under the conditions used (pH 10 with 2-hydroxy-6-methylpurine as substrate) the "very rapid" species is essentially quantitatively generated. With other substrates and at moderate pH, however, much less of the signal-giving species is formed, presumably Met. Ions Life Sei. 2009, 6, 395-416

400

HILLE

due to dissociation of product from the Mo(IV)-product species prior to oxidation to the Mo(V) state.

2. 2.1.

ELECTRON-NUCLEAR DOUBLE RESONANCE STUDIES OF THE "VERY RAPID" SPECIES ENDOR of the Intermediate Seen with Xanthine as Substrate

Even though the "very rapid" species is not always generated to a significant degree under many (even most) experimental conditions, it remains extremely valuable as a paramagnetic reporter on the nature of the Mo-product interaction in the obligatory Mo(IV) species that precedes it in the catalytic sequence. To further investigate the manner in which product was bound to the active site molybdenum in the signal-giving species, Howes et al. examined the 13 C coupling in the "very rapid" species generated with 8-[ 13 C]-xanthine by E N D O R spectroscopy [19]. High quality data yielded a hyperfine tensor A = [11.1, 7.6, 7.6] MHz, with the A tensor rotated from the z-axis of the g-tensor by 25°. From the relationship A| | + 2 A ± = 3 A i s o , the experimental A tensor yielded an A i s o of 8.8 M H z due to the small amount of spin density resident on C8 and an anisotropic term of the form Τ = [2 Τ, -Τ, -Τ] due to the through-space interaction of the magnetic moments of the unpaired electron and the 1 = 1 / 2 nucleus of the 13 C nucleus, with T = 1.17 M H z (reported as A a n i s o =3.5 MHz). This in turn was used to estimate a Mo-C distance of 2.2 A in the signal-giving species. This distance estimate was recognized to be dependent on the hybridization of the C8 carbon and was taken as an upper limit; a more realistic distance of ~ 2.1A was subsequently estimated assuming that sp hybridization yielded the longest possible distance (this in turn was based on the assumption that an sp-hybridized orbital was less spherically symmetric than an sp 3 -hybridized one). On the basis of the conclusion that the C8 of bound product was likely to be within bonding distance of the molybdenum in the "very rapid" species, a reaction mechanism was proposed in which the C8-H bond of substrate inserts across the M o = S bond of oxidized enzyme to yield an intermediate in which the C8 carbon directly bonds to the active site molybdenum. A "buried water" (the authors specifically considered the possibility that it was coordinated to the molybdenum [17]) was then proposed to attack the Mo-C bond to give an intermediate in which the C8 carbonyl of the newly formed uric acid was coordinated to the now reduced molybdenum in a side-on η 2 fashion, in which the Met. Ions Life Sei. 2009, 6, 395-416

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

401

r ι

Η ^

e, H'

Figure 2. The reaction mechanism proposed by Bray and coworkers [17], on the basis of a short M o - C distance in the species giving rise to the "very rapid" E P R signal. The reaction is proposed to begin with the insertion of the C8-H bond across the M o = S group of the molybdenum center to give the species shown with no reduction of the molybdenum itself.

π electrons of the carbonyl bond donate into the molybdenum d x y orbital (Figure 2). This bonding interaction was taken as a vestige of the direct Mo-C bond of the previous intermediate. The reaction was completed by sequential oxidation of the molybdenum and displacement of product by hydroxide from solvent. Precedent for insertion chemistry of the type proposed exists in the reactivity M o = S units within complexes such as the binuclear [MoOS^-S) 2 MoO(Cp)] 2 ~, which are susceptible to insertion reactions with compounds such as carbon disulfide or dicarbomethoxyacetylene [37].

2.2.

ENDOR of the Intermediate Seen with 2-Hydroxy-6methylpurine as Substrate

The interpretation that the Mo-C8 distance in the "very rapid" species was sufficiently short to be interpreted as a direct Mo-C bond was controversial and prompted a second E N D O R study examining the "very rapid" signal generated using 8-[ 13 C]-2-hydroxy-6-methylpurine as substrate [38]. This study was facilitated by the high levels of signal-giving species afforded by this substrate (and with the signal-giving species forming on a more convenient time scale than with xanthine), and yielded A = [10.2, 7.0, 6.5] MHz. Met. Ions Life Sei. 2009, 6, 395-416

402

HILLE

While modestly different from the values obtained with xanthine as substrate, the value for Τ of 1.15 MHz derived with 2-hydroxy-6-methylpurine is essentially indistinguishable from the value of 1.17 MHz seen with xanthine, and yielded an identical apparent Mo-C distance of 2.2 A. These workers noted, as did Howes et al. [18], that Τ has an anisotropic local contribution arising from the finite spin density on C8 and that only the non-local (through-space or dipolar) contribution to Τ should be used in the distance calculation. T]oc is dependent on the degree of hybridization at C8, introducing some uncertainty into its estimation, but a table was constructed demonstrating that the Mo-C distance increased from 2.4 A with sp hybridization to 2.7 A with sp 2 , and 3.0 A with sp 3 hybridization. That the correction increased with the degree of ρ hybridization arises from the fact that a given sp orbital is more spherically symmetric than a given sp 3 orbital, a point not appreciated in the earlier work. Assuming nominal sp 2 hybridization and taking into account the finite spin density on the catalytically introduced oxygen, a value of 2.8 A was estimated for the Mo-C distance [38]. This was deemed too long for a direct Mo-C bond, and was in fact in good agreement with the distance expected for product coordinated to the molybdenum via the catalytically introduced oxygen with a Mo-O-C bond angle of 109°. This geometry is depicted in Figure 3, which also illustrates that this geometry is also consistent with Mo-Η distances of 3.2 and 6.1 A for the N 7 -H and 6-methyl protons (averaged) of 2-hydroxy-6-methylpurine, as previously determined using pulsed EPR methods [39]. It is now generally accepted that the E N D O R data support a Mo-C distance too great to reflect formation of a direct Mo-C bond in the course of the hydroxylation reaction. As has been reviewed elsewhere [40-42], the catalytic sequence is instead envisaged as being initiated by abstraction of the Mo-OH proton by Glul261 in the bovine structure

2.8 A

Η

Figure 3. A summary of the metric information for the "very rapid" species obtained with 2-hydroxy-6-methylpurine from ESEEM [38] and E N D O R [39] analysis. The M 0 - O - C 8 bond angle agrees well with that subsequently seen in the crystal structure of the intermediate [47]. Met. Ions Life Sei. 2009, 6, 395-416

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

403

Η V HO", H+ P, e"

Figure 4. A reaction mechanism for xanthine oxidoreductase predicated on transfer of a M o - O H moiety on C8 of substrate, as originally proposed by Wedd and coworkers [32]. The position of the M o - O H in the active site, as subsequently seen in the crystal structure of the D. gigas aldehyde oxidoreductase [20], was appropriate for such group transfer, and it was suggested that an active site glutamate (Glul261 in the structure of the bovine xanthine oxidoreductase) functioned as an active site base to deprotonate the M o - O H and initiate catalysis.

(Glu730 in the R. capsulatus enzyme), followed by nucleophilic attack on the substrate site to be hydroxylated and concomitant hydride transfer from C8 of substrate to the M o = S group, as depicted in Figure 4. The initial intermediate can be formulated as LMo(IV)0(SH)OR, and in this first step of the reaction the C-H bond of substrate is broken, O-C bond of product is formed, and the molybdenum center formally reduced from the (VI) to (IV) oxidation state. This key intermediate may decay either by transfer of an electron to other redox-active centers in the enzyme (it possesses two [2Fe-2S] iron-sulfur clusters and F A D in addition to the molybdenum center) to give the "very rapid" species, or by displacement of bound product by hydroxide from solvent prior to electron transfer out of the molybdenum center, thereby circumventing formation of the EPR-active intermediate. Decay of the "very rapid" species, to the extent that it is formed, occurs by displacement of product by hydroxide and concomitant transfer of the second electron out of the molybdenum center. Such a mechanism, involving transfer of a Mo-OH ligand to substrate in forming product, was first suggested by Wedd and coworkers [32] on the basis of a comparison of 1 7 0 hyperfine coupling seen in the EPR of a series of structurally defined model compounds with that observed in various enzyme species. The disposition of the Mo-OH facing into the solvent access channel to the active site in the enzyme, as established by subsequent crystallographic work, is consistent with such a mechanism. The available data are thus fully consistent with a reaction mechanism that proceeds as indicated above and shown in Figure 4, with product coordinated to molybdenum in the initially formed E red *P intermediate in a simple end-on fashion via the catalytically introduced oxygen. Met. Ions Life Sei. 2009, 6, 395-416

404

3.

HILLE

X-RAY CRYSTAL STRUCTURES RELEVANT TO THE REACTION MECHANISM

Over the past several years a number of X-ray crystal structures in addition to that of oxidized enzyme have appeared, including several that are of direct mechanistic significance specifically regarding the structure of the E red *P intermediate. What follows is a discussion of these structures, then a consideration of the catalytic roles of specific active site residues in the context of our present understanding of substrate orientation within the active site.

3.1.

Alloxanthine-Complexed Xanthine Oxidoreductase

Allopurinol (4-hydroxypyrazolo[3,4-i/]-pyrimidine) was developed in the late 1960's as an inhibitor of xanthine oxidoreductase to treat hyperuricemia and as a tandem drug to potentiate the effectiveness of 6-mercaptopurine and related first-generation chemotherapeutic agents [43]. George Hitchings and Gertrude Elion shared the 1988 Nobel Prize in Physiology or Medicine (with Sir James Black, for unrelated work) for the successful development of this tandem drug therapy in the treatment of cancer. Allopurinol was found to be oxidatively hydroxylated to alloxanthine (4,6-dihydroxypyrazolo[3,4i/]-pyrimidine) in a manner analogous to the conversion of hypoxanthine to xanthine, but with the product alloxanthine bound tightly to the reduced form of the molybdenum center to inhibit the enzyme [44]. Interest in the structure of the inhibitory complex was greatly stimulated by the discovery that partial oxidation of the E red *alloxanthine complex resulted in an EPR signal that was strongly reminiscent o f t h a t of the "very rapid" species, with unresolved superhyperfine structure due to coupling to a nitrogen nucleus [45,46]. It was immediately recognized that the likely structure of the signalgiving species involved direct coordination of the inhibitor to the molybdenum via the nitrogen at the position analogous to C8 in xanthine. (It was also inferred at the time that the "very rapid" species involved the product uric acid, coordinated via the catalytically introduced oxygen, an interpretation later abandoned in favor of a direct Mo-C bond, as discussed above). The crystal structure of alloxanthine-complexed enzyme was first reported for the xanthine dehydrogenase from Rhodobacter capsulatus [23], which has a very high degree of sequence and structural homology to the vertebrate enzymes. As illustrated in Figure 5, the structure shows the inhibitor bound in the active site between two phenylalanine residues (one interacting face-on, the other side-on with the aromatic inhibitor), and coordinated to the molybdenum via N2, just as predicted on the basis of the EPR work. Inhibitor is oriented in the active site such that its C6-carbonyl is directed Met. Ions Life Sei. 2009, 6, 395-416

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

405

Figure 5. The structure of the reduced R. capsulatus xanthine dehydrogenase in complex with the tight-binding inhibitor alloxanthine [22]. The orientation of the pyrimidine ring of substrate is such that its C6 carbonyl (equivalent to the C2 carbonyl of xanthine) is oriented toward the conserved Arg310 in the active site.

toward Arg310. The structure of the alloxanthine complex defines the substrate binding site, with complexed product occupying the position otherwise occupied by the Mo-OH group in the coordination sphere of oxidized enzyme.

3.2.

Xanthine Oxidoreductase in Complex with FYX-051

The next structure of mechanistic relevance is that of the bovine enzyme in complex with the inhibitor FYX-051, another pharmaceutical^ important inhibitor of the enzyme (marketed as Febuxostat). Like allopurinol, FYX051 is hydroxylated by enzyme [24], but unlike alloxanthine product remains tightly coordinated to the molybdenum center via the catalytically introduced hydroxyl group. In this sense, FYX-051 is a true mechanismbased inhibitor, as opposed to allopurinol which is simply a target-activated drug. As shown in Figure 6, the crystal structure of the reduced enzymeinhibitor complex shows the hydroxylated product coordinated to the molybdenum center with a Mo-O-C geometry closely resembling that predicted on the basis of the ENDOR/ESEEM data discussed above (Figure 3), with a Mo-O-C bond angle of ~110°. Like alloxanthine, FYX-051 has a hydrophobic portion that inserts between the two active site phenylalanine residues. FYX-051 is a much larger molecule than is alloxanthine, however, and its therapeutic effectiveness appears to derive from a close structural complementarity to most (if not all) of the solvent access channel of the mammalian enzyme. Interestingly, FYX-051 does not inhibit the R. capsulatus enzyme, presumably because the solvent access channel of the bacterial enzyme is shaped differently than in the mammalian enzyme (a point suggesting that species-specificity may in at least some cases be engineered into mechanism-based inhibitors). Met. Ions Life Sei. 2009, 6, 395-416

406

HILLE

Figure 6. The crystal structure of bovine xanthine oxidoreductase in complex with the inhibitor FYX-051 [23]. The complex is of a particularly stable intermediate formed in the course of hydroxylation of the inhibitor, corresponding to E red »P, in which the hydroxylated product is coordinated to the metal center via the catalytically introduced hydroxylate.

3.3.

Xanthine Oxidoreductase Reacting with 2-Hydroxy-6methylpurine and Lumazine

M o r e recently, t h e crystal s t r u c t u r e of t h e bovine enzyme d u r i n g t u r n o v e r with t h e slow substrate 2 - h y d r o x y - 6 - m e t h y l p u r i n e h a s been d e t e r m i n e d [47]. This substrate, which generates large a m o u n t s of t h e "very r a p i d " species as discussed above, reacts o n a sufficiently slow time scale (tens of seconds) t h a t it is possible t o arrest the enzyme d u r i n g catalysis simply by briefly soaking p r e - g r o w n crystals of the enzyme with substrate p r i o r t o freezing the crystal f o r d a t a acquisition. As seen in F i g u r e 7, p r o d u c t is f o u n d b o u n d in t h e active site, inserted again between t h e t w o p h e n y l a l a n i n e

Figure 7. The crystal structure of bovine xanthine oxidoreductase in complex with 2-hydroxy-6-methylpurine and lumazine, and of desulfo enzyme with xanthine. As with FYX-051, the first two complexes represent particularly stable catalytic intermediates in which the hydroxylated substrate coordinates the metal center via the catalytically introduced hydroxylate. The orientation of the pyrimidine subnucleus of 2-hydroxy-6-methylpurine (top) is comparable to that of alloxanthine in the reduced R. capsulatus protein, with the C2 carbonyl oriented toward Arg880. The orientation with lumazine, however, is inverted with the C2 carbonyl (equivalent ot C6 of xanthine) oriented toward the arginine (middle). For xanthine bound to the (in)active site of the desulfo enzyme the pyrimidine subnucleus of substrate oriented such that its C6 carbonyl is oriented toward Arg880 (bottom), analogous to the orientation seen with lumazine but opposite that seen with 2-hydroxy-6-methylpurine. Met. Ions Life Sei. 2009, 6, 395-416

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

407

Glu 802 2-hydroxy-6-metbylpurine

Arg 880 Lumazine

Arg 880 xanthine

Arg 880

Met. Ions Life Sei. 2009, 6, 395-416

408

HILLE

residues, and coordinated to the molybdenum via the catalytically introduced hydroxyl group in a manner directly analogous to the structure seen with the inhibitor FYX-051. The sole carbonyl of the pyrimidine subnucleus, at C2, is oriented toward Arg880 (Arg310 in the R. capsulatus enzyme), with the purine in an orientation analogous to that of alloxanthine seen in its complex with reduced enzyme. There is clear electron density bridging between the C8 carbon and molybdenum [47], and the structure is clearly that of the E red *product complex for this slow substrate. The structure provides compelling support for the interpretation above of the E N D O R data regarding the nature of product binding in the E red *P complex. On the basis of a best estimate for the Mo-S distance in the equatorial plane (2.0 ± 0.2 Ä), it would appear that the complex is that of the Mo(V)=S species (rather than the Mo(IV)-SH species) that gives rise to the "very rapid" EPR signal, although this could not be definitively established given the well-known difficulties associated with Fourier transform truncation artifacts introduced into the electron density in the vicinity of atoms as heavy as molybdenum [48]. Interestingly, the structure described above is seen in only one of the two active sites of the homodimeric enzyme (of which one dimer is present in the asymmetric unit of the crystal). In the second site, the purine nucleus sits somewhat further back from the molybdenum center with no clear evidence of intervening electron density [47]. The structure is apparently that of the Michaelis complex of the enzyme, with substrate situated in the active site in an appropriate orientation for catalysis, but with no evidence of the catalytic sequence having been initiated. The implication is that in the crystallographically defined dimer, only one of the two subunits has progressed into the catalytic sequence. This suggests that, in the crystal if not in free solution, the enzyme may function in a reciprocating fashion, with first one then the other subunit functioning catalytically. Such half-sites reactivity for the enzyme has been suggested previously [49], but given the absence of any obvious structural basis for the implied subunit-subunit interaction (the two subunits in the crystal structure appear to have exactly the same protein folds) it remains to be determined whether the enzyme indeed functions in such a way. The structure of the same intermediate seen in the course of the reaction of the bovine enzyme with lumazine (2,4-dihydroxypteridine) has also been determined very recently (J. M. Pauff, H. Cao, and R. Hille, unpublished). Lumazine is the pterin analog to the purine xanthine, and has been shown to be hydroxylated by xanthine oxidoreductase to give violapterin (2,4,7-trihydroxypteridine) [50]. In the course of the reaction a long-lived intermediate absorbing at long wavelength is formed that has been shown to represent a catalytically competent Ered*P species [51]. As compared with 2-hydroxy-6-methylpurine, lumazine binds with its pyrimidine subnucleus Met. Ions Life Sei. 2009, 6, 395-416

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

409

oriented such that the C2 carbonyl (equivalent to C6 in xanthine) points toward Arg880, i.e., inverted relative to the orientation seen with 2-hydroxy6-methylpurine, as shown in Figure 7. As discussed further below, this is also the orientation seen with xanthine bound to the protein and appears to represent the catalytically preferred orientation for substrate binding.

3.4.

Desulfo Xanthine Oxidoreductase in Complex with Xanthine

The final crystal structure of catalytic relevance is (ironically) that of the inactive desulfo form of the bovine enzyme in complex with xanthine (J. M. Pauff, H. Cao, and R. Hille, unpublished). As shown in Figure 7, the structure shows substrate bound at the active site with the C6 carbonyl oriented toward Arg880 (Figure 8); it is evident from the shape of the electron density that an inverted orientation similar to that seen with 2-hydroxy-6-methylpurine does not fit well into the observed electron

Vi

N ^ N ^ O Η

g )

H2N^nh2 ^880

Γ /'S,,, II ^ S

ν ^ - Ν'

η

Ο Ί

οH2N^nh2

_

^880

_

Figure 8. Proposed stabilization of negative charge accumulation on the C6 carbonyl of xanthine in the course of nucleophilic attack on substrate. Stabilization of negative charge on C2 in a substrate orientation inverted relative to that shown is expected to be less effective given the expected lower degree of negative charge distribution onto C2. Met. Ions Life Sei. 2009, 6, 395-416

410

HILLE

density. This orientation is inverted from that seen with 2-hydroxy-6methylpurine (where the C2 carbonyl is oriented toward the active site arginine) but analogous to the orientation seen with lumazine above. In an earlier kinetic study of a homologous series of purine substrates, these could be segregated into two distinct groups: those that were effective substrates of wild-type enzyme and which were profoundly affected by mutation of Arg310 (in the R. capsulatus enzyme), with kred in the reductive half-reaction reduced by approximately 104; and those that were poor substrates but which were not profoundly affected by the mutation [52]. It was suggested that the dichotomy in substrate reactivity was due to differences in the orientation of substrates in the active site. Effective substrates had a C6 carbonyl (or thio) group and were proposed to bind in the active site as seen with xanthine, with this group oriented toward Arg310 (Arg880 in the bovine sequence). This arginine appeared positioned to stabilize negative charge accumulation on the heterocycle in the course of nucleophilic attack by charge complementation at the C6 carbonyl oxygen, as shown in Figure 8. Substrates that were less effective (e.g., 2-hydroxy-6methylpurine) bound in the opposite orientation, with C2 oriented toward Arg310. Although not reacting as rapidly with wild-type enzyme, these poorer substrates were also much less sensitive to mutation of Arg310 to Met - in the case of 2,6-diaminopurine, the substrate is actually somewhat more reactive with the R310M mutant than wild-type enzyme [52].

3.5.

Substrate Orientation and the Basis of Enzyme Catalysis

The above crystallographic information provides crucial information not only about the structures of the critical Ered*product and "very rapid" intermediates in the catalytic sequence (and by inference the specific mechanism by which substrate is chemically converted to product) but also the roles of specific amino acid residues in accelerating reaction rate by establishing the orientation of substrate in the active site. We consider here three specific residues, Glu767, Glul261, and Arg880 (Glu232, Glu730, and Arg310 in the numbering system of the R. capsulatus xanthine dehydrogenase, with which the site-directed mutagenesis studies to be discussed were performed). The role of each of these residues, depicted in Figure 9, in catalysis will now be considered.

3.5.1.

Glu730

Glu730 is universally conserved in the purine- and aldehyde-hydroxylating molybdenum enzymes, and as indicated above on the basis of the crystal Met. Ions Life Sei. 2009, 6, 395-416

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

Gin 767

411

Gin 802

Figure 9. Residues of the active site of xanthine oxidoreductase that interact with substrate or the molybdenum center of the active site.

structure of the first of these enzymes to be determined (that of the aldehyde oxidoreductase from D. gigas) is thought to act as a general base, deprotonating the Mo-OH of the molybdenum center to facilitate nucleophilic attack on substrate (be it a purine as illustrated in Figure 4, or an aldehyde) [21]. The protonation state of the ligand to the metal is important, and although the initial X-ray crystallographic work concluded that the ligand was a doubly protonated water molecule, it has subsequently been established using the more precise method of X-ray absorption spectroscopy (with the bovine enzyme) that it is a singly protonated Mo-OH [53], whose deprotonation leads to the much more nucleophilic Mo-O". Mutation of Glu730 profoundly affects catalysis: while the wild-type enzyme is fully reduced by 100 μΜ xanthine within 100 ms under anaerobic conditions at pH 7.8, the E730A mutant is not perceptibly reduced in the course of an overnight incubation with substrate at the same concentration [34]. The effect on the limiting rate of reduction is conservatively estimated to be a factor of 107, amounting to ~ 10 kcal/mol of transition state stabilization contributed by Glu730. Interestingly, XAS analysis of the (bovine) enzyme at pH 10 suggests a shortening of the Mo-O bond consistent with deprotonation [53] and indeed much of the disparity in reactivity between wild-type and E730A mutant is lost in going from pH 7.4 to 10 (Ibdah and Hille, unpublished). The effect is approximately equally due to Met. Ions Life Sei. 2009, 6, 395-416

412

HILLE

loss of activity of wild-type enzyme as gain of activity of the mutant at high pH.

3.5.2.

Glu232

Glu232 sits opposite Glu730, atop the active site as shown in Figure 9, and is strictly conserved only among enzymes that hydroxylate xanthine - it is typically absent in aldehyde-oxidizing enzymes. Mutation of this residue to an alanine results in a ~ 10-fold decrease in the limiting rate of enzyme, /cred, reduction at high [xanthine] and a similar ~ 10-fold increase in K d for xanthine - this residue thus contributes approximately 1.5 kcal/mol to transition state stabilization and a similar amount to substrate affinity [34]. In a computational analysis of tautomeric forms of substrate likely to be encountered in the course of catalysis [54], it was found that the tautomer with protons on nitrogens 1, 7, and 9 was significantly stabilized relative to the predominant tautomer in solution (with protons on nitrogens 1,3, and 7) once nucleophilic attack has occurred (Figure 10). Glu232 was proposed to

substrate

s p 3 intermediate

Figure 10. Possible role of tautomerism in the course of the reaction of xanthine oxidoreductase. Left: the f o u r tautomeric forms of neutral xanthine. Right: the relative stabilities of the three most stable tautomeric forms of xanthine in solution and as part of an E r e d »P complex.

Met. Ions Life Sei. 2009, 6, 395-416

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

413

facilitate this tautomerization, accounting for the rate acceleration attributable to this residue. The orientation of substrate seen in the crystal structure of desulfo (bovine) xanthine oxidase in complex with xanthine is consistent with such a role. It has alternatively been proposed that this tautomerization is facilitated by Glu730 [55], and it would indeed be elegant were Glu730 to simultaneously deprotonate the Mo-OH and transfer the proton thus obtained to N9 of substrate once nucleophilic attack has been initiated. Such a role, however, is predicated on a substrate orientation in the active site opposite to that seen crystallographically and now seems unlikely.

3.5.3.

Arg310

As indicated above, it is the orientation of substrate relative to Arg310 that is proposed to determine whether a given purine substrate will be an effective or ineffective substrate for enzyme, due to the ability of the arginine to stabilize negative charge accumulation on the C6 carbonyl oxygen in the course of nucleophilic attack. Although stabilization via the C2 carbonyl may occur to some degree when substrate is in the unproductive orientation, it is expected to be less effective than stabilization via the C6 carbonyl as evidenced in the Kekule structures shown in Figure 8, where it can be seen that it is more difficult to delocalize negative charge on the C2 carbonyl oxygen. Thus as with Glu232, substrate orientation dictates the role of Arg310 in catalysis.

4.

GENERAL CONCLUSIONS

Although the formation of a direct Mo-C bond in the course of the reaction of xanthine oxidoreductase was advocated on the basis of the analysis of the original ENDOR data obtained with the "very rapid" intermediate, subsequent analysis strongly suggests that the Mo-C distance is no shorter than 2.8 A and that it is unlikely that the signal-giving species possesses a direct bond between metal and substrate carbon. Several subsequently determined X-ray crystal structures, including those of catalytic intermediates seen with FYX-051, 2-hydroxy-6-methylpurine and lumazine, support the latter ENDOR interpretation of substrate binding in a simple end-on fashion in the critical intermediate(s). The structures imply a reaction mechanism that is initiated by base-assisted nucleophilic attack on substrate after deprotonation of a Mo-OH group of the molybdenum center. Such a mechanism provides a ready explanation for the observed effects seen on mutagenesis in the active site of the enzyme. Met. Ions Life Sei. 2009, 6, 395-416

414

HILLE

ACKNOWLEDGMENTS Work in the author's laboratory has been supported by the National Institutes of Health (GM 075036). The author wishes to acknowledge Mr. J. M. Pauff for preparation of the structural Figures 5, 6, 7, and 9.

ABBREVIATIONS AND DEFINITIONS allopurinol alloxanthine ENDOR EPR ESEEM FAD lumazine NAD+ violapterin XAS XOR

4-hydroxypyrazolo[3,4-i/]-pyrimidine 4,6-dihydroxypyrazolo[3,4-i/]-pyrimidine electron-nuclear double resonance electron paramagnetic resonance electron spin echo envelope modulation spectroscopy flavin adenine dinucleotide 2,4-dihydroxypteridine nicotinamide adenine dinuclotide 2,4,7-trihydroxypteridine X-ray absorption spectroscopy xanthine oxidoreductase

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

R. Hille and T. Nishino, FAS ΕΒ J., 1995, 9, 995-1003. Μ. Dixon and S. Thurlow, Biochem. J., 1924, 18, 971-988. G. Palmer, R. C. Bray and H. Beinert, J. Biol. Chem., 1964, 239, 2657-2666. R. C. Bray, G. Palmer and H. Beinert, J. Biol. Chem., 1964, 239, 2667-2676. R. C. Bray and T. Vänngärd, Biochem. J., 1969, 114, 725-734. P. F. Knowles, F. M. Pick and R. C. Bray, Biochem. J., 1968, 107, 601-602. R. Hille, J. H. Kim and C. Hemann, Biochemistry, 1993, 32, 3973-3980. S. Gutteridge, S. J. Tanner and R. C. Bray, Biochem. J., 1978, 175, 869-878. S. J. Tanner, R. C. Bray and F. Bergmann, Biochem. Soc. Trans., 1978, 6, 227-229. V. Massey and D. E. Edmondson, J. Biol. Chem., 1970, 245, 6595-6598. J. P. G. Malthouse and R. C. Bray, Biochem. J., 1980, 191, 265-267. J. P. G. Malthouse, G. N. George, D. J. Lowe and R. C. Bray, Biochem. J., 1981, 199, 629-637. T. D. Tullius, D. M. Kurtz Jr., S. D. Conradson and K. O. Hodgson, J. Am. Chem. Soc., 1979, 101, 2776-2777. J. Bordas, R. C. Bray, C. D. Garner, S. Gutteridge and S. Hasnain, Biochem. J., 1980, 191, 499-508. R. C. Bray and L. S. Meriwether, Nature, 1966, 212, 467-469.

Met. Ions Life Sei. 2009, 6, 395^116

MOLYBDENUM HYDROXYLASE XANTHINE OXIDOREDUCTASE

415

16. G. N. George and R. C. Bray, Biochemistry, 1988, 27, 3603-3609. 17. S. Gutteridge and R. C. Bray, Biochem. J., 1980, 189, 615-623. 18. B. D. Howes, R. C. Bray, R. L. Richards, N. A. Turner, B. Bennett and D. J. Lowe, Biochemistry, 1996, 35, 1432-1443. 19. Κ. N. Murray, J. G. Watson and S. Chaykin, J. Biol. Chem., 1966, 241, 4798-4801. 20. M. J. Romäo, M. Archer, I. M o u r a , J. J. G. Moura, J. LeGall, R. Engh, M. Schneider, P. Hof and R. Huber, Science, 1995, 270, 1170-1176. 21. R. Huber, P. Hof, R. O. Duarte, J. J. G. Moura, I. M o u r a , J. LeGall, R. Hille, Μ. Archer and M. J. Romäo, Proc. Natl. Acad. Sei. USA, 1996, 93, 8846-8851. 22. C. Enroth, Β. T. Ebger, K. Okamoto, T. Nishino and E. F. Pai, Proc. Natl. Acad. Sei. USA, 2000, 97, 10723-10728. 23. J. J. Truglio, K. Theis, S. Leimkühler, R. Rappa, Κ. Y. Rajagopalan and C. Kisker, Structure, 2002, 10, 115-125. 24. K. Okamoto, K. Matsumoto, R. Hille, Β. Τ. Eger, Ε. F. Pai and T. Nishino, Proc. Natl. Acad. Sei. USA, 2004, 101, 7931-7936. 25. H. Dobbek, L. Gremer, R. Kiefersauer, R. Huber and O. Meyer, Proc. Natl. Acad. Sei USA, 2002, 99, 15971-15976. 26. I. Bonin, Β. M. Martins, Y. Purvanov, S. Fetzner, R. Huber and H. Dobbek, Structure, 2004, 12, 1425-1435. 27. R. H. Holm, Coord. Chem. Rev., 1990, 100, 183-221. 28. J. H. Enemark, J. J. A. Cooney, J.-J. Wang and R. H. Holm, Chem. Rev., 2004, 104, 1175-1200. 29. R. Hille and H. Sprecher, J. Biol. Chem., 1987, 262, 10914-10917. 30. M. Xia, R. Dempski and R. Hille, J. Biol. Chem., 1999, 274, 3323-3330. 31. G. L. Wilson, M. Kony, E. R. T. Tiekink, J. R. Pilbrow, J. T. Spence and A. G. Wedd, J. Am. Chem. Soc., 1988, 110, 6923-6925. 32. R. J. Greenwood, G. L. Wilson, J. R. Pilbrow and A. G. Wedd, J. Am. Chem. Soc., 1993, 115, 5385-5392. 33. J. H. Kim, M. G. Ryan, H. K n a u t and R. Hille, J. Biol. Chem., 1996, 271, 6771-6780. 34. S. Leimkühler, A. L. Stockert, K. Igarashi, T. Nishino and R. Hille, J. Biol. Chem., 2004, 279, 40437-40444. 35. R. C. Bray and G. N. George, Biochem. Soc. Trans., 1985, 13, 560-567. 36. R. B. McWhirter and R. Hille, J. Biol. Chem., 1991, 266, 23724-23731. 37. D. Coucouvanis, A. Toupadakis, J. D. Lane, S. M. Koo, C. G. Kim and A. Hadjikyriacou, J. Am. Chem. Soc., 1991, 113, 5271-5282. 38. P. Manikandan, E.-Y. Choi, R. Hille and Β. M. H o f f m a n , J. Am. Chem. Soc., 2001, 123, 2658-2663. 39. G. A. Lorigan, R. D. Britt, J. H. Kim and R. Hille, Biochim. Biophys. Acta, 1994, 1185, 284-294. 40. R. Hille, Chem. Rev., 1996, 96, 2757-2816. 41. R. Hille, Trends Biochem. Sei., 2002, 27, 360-367. 42. R. Hille, Eur. J. Inorg. Chem., 2006, 10, 1913-1926. 43. G. B. Elion, A. Kovensky, G. H. Hitchings, E. Metz and R. W. Rundles, Biochem. Pharmacol., 1966, 15, 863-880.

Met. Ions Life Sei. 2009, 6, 395-416

416

HILLE

44. Y. Massey, Η. Komai, G. Palmer and G. B. Elion, J. Biol. Chem., 1970, 245, 2837-2844. 45. J. W. Williams and R. C. Bray, Biochem. J., 1981, 195, 753-760. 46. T. R. Hawkes, G. N. George and R. C. Bray, Biochem. J., 1984, 218, 961-968. 47. J. M. Pauff, J. Zhang, C. E. Bell and R. Hille, J. Biol. Chem., 2008, 283, 4818-4824. 48. C. Kisker, H. Schindelin and D. C. Rees, Annu. Rev. Biochem., 1997, 66, 233-268. 49. L. A. Tai and K. C. Hwang, Biochemistry, 2004, 43, 4869-4876. 50. Μ. D. Davis, J. S. Olson and G. Palmer, J. Biol. Chem., 1982, 257, 3526-3583. 51. M. D. Davis, J. S. Olson and G. Palmer, J. Biol. Chem., 1984, 259, 14730-14737. 52. J. M. Pauff, C. F. Hemann, S. Leimkühler and R. Hille, J. Biol. Chem., 2007, 282, 12785-12790. 53. C. J. Doonan, A. L. Stockert, R. Hille and G. N. George, J. Am. Chem. Soc., 2005, 127, 4518-4522. 54. P. Ilich and R. Hille, Inorg. Chim. Acta, 1997, 263, 87-94. 55. Y. Yamaguchi, T. Matsumura, K. Ichida, K. Okamoto and T. Nishino, J. Biochem. (Tokyo), 2007, 141, 513-524.

Met. Ions Life Sei. 2009, 6, 395-416

Met. Ions Life Sei. 2009, 6, 417-460

12 Computational Studies of Bioorganometallic Enzymes and Cofactors Matthew D. Liptak, Katherine M. Van Heuvelen, and Thomas C. Brunold* Department of Chemistry, University of Wisconsin-Madison, Madison WI 53706, USA

ABSTRACT 1. INTRODUCTION 2. COMPUTATIONAL APPROACHES TO BIOORGANOMETALLIC CHEMISTRY 2.1. Overview 2.2. Popular Methods for Computing Geometric and Electronic Structures 2.3. Computation of Spectroscopic Observables 3. FORMATION AND CLEAVAGE OF THE CO-C BOND OF COBALAMIN IN ENZYME ACTIVE SITES 3.1. Overview 3.2. Co-C Bond Formation During Adenosylcobalamin Biosynthesis and Methylcobalamin Activation by Methionine Synthase 3.3. Co-C Bond Homolysis in Adenosylcobalamin-Dependent Enzymes 3.4. Co-C Bond Heterolysis in Methylcobalamin-Dependent Enzymes 4. ORGANOMETALLIC CHEMISTRY AND CATALYTIC CYCLE OF METHYL-COENZYME Μ REDUCTASE Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00417

418 419 420 420 421 424 426 426 427 430 432 435

418

LIPTAK, V A N H E U V E L E N , and B R U N O L D

4.1. Overview 4.2. Properties of the Free F 430 Cofactor 4.3. Ni-Alkyl Species of Methyl-Coenzyme Μ Reductase: Computational Evaluation of Viable Reaction Mechanisms 4.3.1. Ni-Methyl Pathway 4.3.2. Methyl Radical Pathway 4.3.3. Protonation of Ni(I)MCR Redl 5. GEOMETRIC AND ELECTRONIC STRUCTURES OF THE CARBON MONOXIDE DEHYDROGENASE/ACETYLCOENZYME A SYNTHASE ACTIVE SITE CLUSTERS 5.1. Overview 5.2. Mechanism of Carbon Monoxide Oxidation/Carbon Dioxide Reduction by Carbon Monoxide Dehydrogenase 5.3. Computational Insights into the Catalytic Cycle of AcetylCoenzyme A Synthase 6. MAGNETIC PROPERTIES OF THE ACTIVE SITE CLUSTER OF IRON-ONLY HYDROGENASES 6.1. Overview 6.2. Structure and Oxidation States of the Active Site Cluster of [FeFe] Hydrogenases 6.3. Calculation of the Magnetic Properties of the H-Cluster 7. CONCLUDING REMARKS AND FUTURE DIRECTIONS ACKNOWLEDGMENTS ABBREVIATIONS REFERENCES

435 437 439 439 441 442 442 442 444 445 447 447 447 449 450 452 452 454

ABSTRACT: Because of their complex geometric and electronic structures, the active sites and cofactors of bioorganometallic enzymes, which are characterized by their metal-carbon bonds, pose a major challenge for computational chemists. However, recent progress in computer technology and theoretical chemistry, along with insights gained from mechanistic, spectroscopic, and X-ray crystallographic studies, have established an excellent foundation for the successful completion of computational studies aimed at elucidating the electronic structures and catalytic cycles of these species. This chapter briefly reviews the most popular computational approaches employed in theoretical studies of bioorganometallic species and summarizes important information obtained from computational studies of (i) the enzymatic formation and cleavage of the Co-C bond of coenzyme B 12 ; (ii) the catalytic cycle of methyl-coenzyme Μ reductase and its nickel-containing cofactor F 430 ; (iii) the polynuclear active-site clusters of the bifunctional enzyme carbon monoxide dehydrogenase/acetyl-coenzyme A synthase; and (iv) the magnetic properties of the active-site cluster of Fe-only hydrogenases. KEYWORDS: bioorganometallic enzymes and cofactors · carbon monoxide dehydrogenase/acetyl-coenzyme A synthase · coenzyme B 12 -density functional theory · electronic structure description · Fe-only hydrogenase · methyl-coenzyme Μ reductase

Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

1.

419

INTRODUCTION

Because of their complex geometric and electronic structures, the active sites and cofactors of bioorganometallic enzymes, which are characterized by the rare occurrence of metal-carbon bonds in Nature, represent major challenges for computational chemists. Nevertheless, owing to the remarkable progress in computer technology, the development of advanced computational methodologies, and the dedicated efforts of X-ray crystallographers that have afforded a wealth of high-resolution structures of the bioorganometallic enzymes and cofactors, the necessary foundation has been established for the successful completion of computational studies aimed at elucidating the electronic structures and catalytic cycles of these species. Density functional theory (DFT) has proven to be a particularly popular method for the quantum chemical treatment of these large metal-containing enzyme active sites and cofactors. The application of D F T in computational bioinorganic chemistry is the subject of an excellent review that has recently been published by Deeth [1]. As delineated in this chapter, computational studies have played a vital role in advancing our understanding of the electronic structures and catalytic cycles of the bioorganometallic enzyme active sites and cofactors. While the success in obtaining high-resolution X-ray structures has permitted detailed insight into the coordination environments of the metal centers in these species, in many cases delivering surprising information regarding the composition of polynuclear metal clusters and revealing the identities of unusual active-site ligands, these structures often raised more questions concerning the corresponding catalytic mechanisms than they answered. Therefore, computational studies - when properly evaluated on the basis of the results obtained in X-ray crystallographic, kinetic and/or spectroscopic investigations - will undoubtedly continue to play a key role in future research into the reaction cycles of the bioorganometallic systems. This chapter is organized as follows. In Section 2, the most popular computational approaches employed in theoretical studies of bioorganometallic species and methods for validating computational results on the basis of spectroscopic data are briefly reviewed. Section 3 summarizes relevant information obtained from computational studies of enzymes that are involved in the biosynthesis of coenzyme B 12 or utilize this unusually complex cofactor for carrying out methyltransfer and radical-rearrangement reactions. The geometric and electronic properties of the nickel-containing cofactor F 4 3 0 and the catalytic cycle of the enzyme that requires this species for its activity, methyl-coenzyme Μ reductase, are the subjects of Section 4. Section 5 provides an overview of important insights that have recently been gained from computational studies of the polynuclear active-site clusters of the bifunctional enzyme CO dehydrogenase/ acetyl-coenzyme A synthase. Lastly, in Section 6, a DFT-based quantitative

Met. Ions Life Sei. 2009, 6, 417-460

420

LIPTAK, VAN HEUVELEN, and BRUNOLD

analysis is presented of the magnetic properties of the active-site cluster of Feonly hydrogenases.

2. 2.1.

COMPUTATIONAL APPROACHES TO BIOORGANOMETALLIC CHEMISTRY Overview

The number of computational studies applied to bioorganometallic systems has grown so dramatically in the last decade that it is now almost commonplace to complement experimental studies with computations. Undoubtedly, a major contributor to this development has been the remarkable progress in computer technology, which is nicely demonstrated by the fact that the speed of microprocessor chips has increased by more than a factor o f t e n from 1997 to 2007 [2]. However, it would be a gross oversimplification to consider technological progress as the only driving force for the explosive growth of computational studies on bioorganometallic systems in the last decade. Computational approaches have been utilized with increasing frequency because they have proven to be an extremely useful complement to experimental investigations. For example, computations have been successfully used to investigate the formation and cleavage of the Co-C bond in cobalamin (Cbl, Section 3), evaluate putative catalytic intermediates in methyl-coenzyme Μ reductase (MCR, Section 4), explore the electron distributions and exchange interactions among the active site metal ions in carbon monoxide dehydrogenase/acetyl-coenzyme A synthase (CODH/ACS, Section 5), and elucidate the magnetic properties of the [FeFe] hydrogenase active-site cluster (Section 6). The foundation of computational approaches to chemistry is the prediction of energies and molecular geometries. There is a diverse selection of computational methods available to compute the electronic structure of a system at a fixed geometry. Below, these methods are categorized into distinct classes and the typical applications and limitations for each class are discussed using methylcobalamin (MeCbl) as a representative example (Section 2.2). All of these methods employ fundamentally similar strategies to find the energyminimized (or optimized) geometry of a bioorganometallic species. In particular, they all rely on the Born-Oppenheimer approximation [3], according to which the total wavefunction of a molecule is separated into electronic and nuclear components - a reasonable assumption considering that the electrons move much faster than the nuclei due to the large difference in mass. A typical geometry-optimization strategy can be illustrated by considering the Co-C bond of MeCbl [4]. First, a reasonable initial guess is made for the Co-C bond length and the total energy of MeCbl is computed for this particular nuclear configuration. Next, the Co-C bond length is changed slightly Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

421

and the total energy associated with this new geometry is computed. The geometry-optimization algorithm will continue to vary the C o - C bond length until it reaches a minimum for the total energy of the molecule. At this point, the geometry-optimization algorithm has optimized the C o - C bond length within the limitations of the method chosen to compute the total energy. In cases where the positions of all atoms of the molecule are simultaneously optimized, more complex algorithms are required [5], but the general conceptual approach remains the same. In general, the accuracy of the energyminimized (i.e., optimized) structure of the molecule is limited by the quality of the computational method used to calculate the total energy for a given nuclear configuration. It is of utmost importance to validate the molecular structure obtained in the energy minimization process on the basis of experimental data. In cases where detailed structural information about the molecule of interest is available, e.g., from X-ray crystallography or N M R spectroscopy, the accuracy of the optimized structure can be assessed from a direct comparison between key experimental and computed metric parameters. However, for structurally ill-defined species, such as short-lived intermediates, and to validate the computed electronic structure of a given molecule, it is useful to compare computationally predicted and experimentally determined spectroscopic parameters. A wide variety of methods have been developed to calculate spectroscopic observables from the computed electronic structure. Several methods that have been utilized to validate the computed electronic structure description for various Cbl species are described in Section 2.3. Since satisfactory agreement was achieved between theory and experiment for these species, the computational results provide a sensible framework for answering structural and mechanistic questions of interest.

2.2.

Popular Methods for Computing Geometric and Electronic Structures

Any cursory examination of the literature reveals that an enormous number of different methods is available for predicting the geometric structures of bioorganometallic species or chemical compounds in general. We do not attempt here to provide an exhaustive listing of all available methods or to present the quantitative details for any single method, as these topics have been thoroughly discussed in several text books [6-10]. Instead, this section highlights four methods that can be, and in fact have been, used to compute the geometric structures of bioorganometallic species (Table 1). Molecular mechanics (MM) is typically the method of choice to predict the structure of a large bioorganometallic species, such as a protein. M M is particularly appealing for studies of proteins because the large number of Met. Ions Life Sei. 2009, 6, 417-460

LIPTAK, V A N HEUVELEN, and B R U N O L D

422 Table 1. Method

Summary of the computational methods discussed in Section 2.2. Typical target

Molecular mechanics Protein (MM) Density functional Active site model theory (DFT) Multiconfigurational Active site model self-consistent field valence MOs (MC-SCF) Quantum mechanics/ Active site (QM) molecular mechanics + protein (MM) (QM/MM)

C o - C bond description Harmonic spring Bonding M O Linear combination of MOs Depends on choice of QM method

Limitations Parameterization No electrons Single reference configuration Extreme computational cost Development of realistic Q M / M M partitioning and coupling

atoms in these systems makes other currently available methods prohibitively expensive in terms of computational cost. The application of M M to protein systems has been recently reviewed [11,12]. A particularly popular implementation for bioorganometallic species is the Amber 95 force field [13,14]. Returning to the MeCbl example, this force field models the Co-C bond as a harmonic oscillator in which the two nuclei are connected by a spring. The equilibrium bond length and force constant of this spring are determined by parameters derived from a fit to reliable experimental or theoretical data. For example, the Amber 95 description of the C o - C bond of MeCbl required the careful development of the bond-length and forceconstant parameters for Co 3 + corrinoids by Marques and coworkers [15]. This extensive parameterization is a major limitation of MM, especially for bioorganometallic species where changes in the metal ion oxidation state and/or ligand environment due to substrate binding, product release, etc., will significantly affect the M - C bond length(s). A second major limitation of M M is that this method does not explicitly treat electrons and, thus, cannot be used for investigating the electronic structures of bioorganometallic species [16]. Density functional theory is often the method of choice for the explicit treatment of the electronic structure of a small bioorganometallic species, such as an active site model. Because D F T accounts for electron correlation in an approximate way, it is a more popular single reference configuration method than Hartree-Fock (HF) theory, which ignores this term altogether [10], for studying transition metal complexes where electron correlation is an important determinant of electronic structure [17]. The application of D F T Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

423

to bioinorganic species has been recently reviewed [1,18,19]. Note that in a D F T approach, the composition of every MO is simultaneously determined by minimizing the total energy of the molecule with the Kohn-Sham (KS) self-consistent field (SCF) method [20,21]. This description of the electronic structure as a single set of MOs is a significant limitation of DFT, especially for transition metal complexes where extensive mixing can occur between energetically proximate states that arise from different electronic configurations (so-called configuration interaction, CI) [22]. Multiconfigurational (MC)-SCF is usually the preferred method for properly treating CI mixing in transition metal complexes. If different electronic configurations give rise to multiple states that have similar total energies, then the electronic structure of the molecule cannot be accurately described by a single electronic configuration; rather, a mixture of these configurations must be considered. The application of the MC-SCF approach to transition metal complexes has been recently reviewed [22-24]. A relatively straightforward implementation of this approach is provided by the complete active space SCF (CASSCF) method [25], CASSCF introduces multiconfigurational character into the ground-state wavefunction by allowing for CI mixing among the complete set of states arising from all possible electron configurations associated with a user-defined set of socalled "active" MOs. Because the CASSCF approach is intensely demanding of computational resources, only a relatively small number of active MOs can be considered. Consequently, the application of the CASSCF approach to bioorganometallic species requires that the user be extremely careful in selecting the proper set of active MOs. Combined quantum mechanics/molecular mechanics (QM/MM) is a computational strategy that employs a D F T or MC-SCF method for the treatment of a small portion of the molecule and an M M method for the remaining atoms. In principle, this is an ideal strategy for bioorganometallic protein systems, because it allows the electronic structure of the active site to be treated explicitly with D F T or MC-SCF, while the remainder of the protein can be modeled with M M to reduce the overall computational cost [16,26,27]. For example, when applied to a MeCbl-dependent enzyme, it would be sensible to describe the relevant portion of the cofactor and certain key amino-acid residues in the enzyme active site by a QM method and include all other atoms in the M M region to suitably account for steric and electrostatic interactions within the entire protein. The most challenging aspects of any Q M / M M calculation are the partitioning of the full system into a QM region and a M M region and the treatment of the coupling between them. A proper treatment of the coupling terms is essential to assure that the M M region realistically perturbs the geometric and electronic structures of the QM region. The ideal partitioning and coupling of the QM and M M regions for a particular bioorganometallic species can require Met. Ions Life Sei. 2009, 6, 417-460

424

LIPTAK, V A N H E U V E L E N , and B R U N O L D

significant user insight, but Q M / M M ultimately offers an excellent strategy for accurately modeling the active sites of large enzymatic systems.

2.3.

Computation of Spectroscopic Observables

As stated above, it is of paramount importance that the results obtained with any theoretical method be validated on the basis of experimental data [18]. This is particularly true for computational studies of bioorganometallic species, whose electronic structures are too complex to be described by any currently available method without making certain approximations. The spectroscopic validation of computed electronic-structure descriptions entails the calculation of spectroscopic observables that can be compared directly to experimental data. Once satisfactory agreement between computed and experimental data is achieved, it is reasonable to trust the underlying electronic structure. The remainder of this section outlines some recent examples of how spectroscopic data were used as the basis for validating computational results obtained for the Cbl cofactor (Table 2). Electron paramagnetic resonance (EPR) spectroscopy offers one of the most sensitive probes of the ground state electronic structures of paramagnetic bioorganometallic species [28]. Therefore, several researchers have used EPR spectroscopy to investigate the paramagnetic cob(II)alamin (Co 2+ Cbl) and cob(II)inamide (Co 2 + Cbi + ; a derivative of Co 2 + Cbl in which the nucleotide loop including the DMB base is absent and a water molecule is axially coordinated to the cobalt ion) forms of the Cbl cofactor [29-32], and accurate values for both the g and 59 Co A tensors could be extracted from quantitative analyses of the corresponding EPR spectra. The g and 59 Co A tensors are sensitive to both the relative energies of the Co 3d-based MOs and the composition of the

Table 2. Summary of the spectroscopically validated calculations on the cobalamin cofactor discussed in Section 2.3.

Spectroscopy

Electronic structure method

Spectroscopic calculation method

F o r m of cobalamin

References

EPR Raman rR Abs Abs Abs Abs

B3LYP B3LYP PBE B3LYP BP86 PBE CASSCF

CPSCF SQM Harmonie TD-DFT TD-DFT TD-DFT CASPT2

Co2+Cbl Co3+Cbl Co1+Cbl Co3+Cbl Co2+Cbl Co1+Cbl Co1+Cbl

[37] [52-54] [42] [41] [37] [42] [60]

Met. Ions Life Sei. 2009, 6, 417-460

425

C O M P U T A T I O N A L S T U D I E S OF B I O O R G A N O M E T A L L I C S

Co 3dZ2-based singly occupied MO (SOMO), thus providing an excellent basis for assessing the feasibility of computational results. Hence, we have used the experimental EPR parameters for validating our D F T (B3LYP) computed [33-36] electronic-structure descriptions for Co 2 + Cbl and C o 2 + C b i + [37] by calculating the g and 59 Co A tensors for each species with the coupled-perturbed SCF (CPSCF) method [38,39], For both species the g and 59 Co A tensors calculated by this method were found to agree reasonably well with the experimental data and, more importantly, accurately reproduced the nearly CQ

"I"

uniform increase of the g and ^ C o A values from C o ^ C b l to C c T X b r . This satisfactory agreement between the computationally predicted and experimentally observed trends in the g and 59 Co A values indicated to us that the differences between the D F T computed relative energies and compositions of the Co 3d-based MOs of Co 2 + Cbl and C o 2 + C b i + are meaningful. Raman spectroscopy, which is excellently suited for probing the vibrational modes of bioorganometallic species [40], has been extensively utilized to study several biologically relevant forms of the Cbl cofactor [37,41-51]. Importantly, the vibrational frequencies obtained by Raman spectroscopy can be directly compared to the normal mode frequencies extracted from a calculated potential energy surface. Kozlowski and coworkers have taken advantage of this relationship and developed a D F T (B3LYP) based scaled quantum mechanical (SQM) force field for several forms of cob(III)alamin (Co 3 + Cbl) [52-54] that was validated on the basis of experimental data. Although relatively poor agreement was achieved in the corrin stretching mode region (1450-1600 c m - 1 ) , this force field satisfactorily reproduced the C o - C stretching mode region (350-650 c m - 1 ) of the Raman spectra of MeCbl and several other alkyl C o 3 + C b l species and, therefore, provided a suitable basis for exploring the origin of the different C o - C bond strengths in these species. More recently, we have been able to reproduce the corrin stretching mode region (1450-1600 c m - 1 ) of the C o 1 + C b l R a m a n spectrum almost quantitatively, without the need of introducing empirical scaling factors, by using the PBE density functional [34,42,55,56]. In particular, excellent agreement between theory and experiment was achieved for the vibrational frequencies, ^H —• 2 D isotopic shifts, and off-resonance R a m a n intensities. This study also attempted to predict the resonance R a m a n (rR) intensities of the corrinbased stretching modes. In contrast to the off-resonance R a m a n intensities, the rR intensities exhibit a strong dependence on the excitation wavelength used because only those normal modes that couple to the electronic transition at this particular wavelength are resonance enhanced [40]. We found that the PBE method successfully identified the resonance-enhanced modes in the corrin stretching mode region across the visible/near-UV regions of the C o 1 + C b l absorption spectrum. Owing to this good agreement between theory and experiment, a more detailed analysis of the D F T (PBE) computed bonding description for C o 1 + C b l was warranted. Met. Ions Life Sei. 2009, 6, 417-460

426

LIPTAK, VAN HEUVELEN, and BRUNOLD

Electronic absorption (Abs) spectroscopy has been the most frequently used spectroscopic tool in studies of the Cbl cofactor [57], in part because it can be utilized to probe electronic excited states regardless of the cobalt ion spin and oxidation states [58]. Although it is not typically possible to resolve the individual electronic transitions contributing to the Abs trace of a Cbl species directly, this task can usually be accomplished by collecting complementary data using circular dichroism (CD) and magnetic C D (MCD) spectroscopic techniques [37,41,42]. Collectively, the information that can be gained from a combined analysis of Abs, CD, and M C D spectroscopic data provides an almost ideal experimental framework for validating computational results. Thus, to assess the feasibility of the D F T computed electronic structure descriptions for our C o 3 + C b l , Co 2 + Cbl, and C o 1 + C b l models, we have used the time-dependent D F T (TD-DFT) method [39,59] to predict the corresponding Abs spectra, which could then be compared directly to the results obtained from our Abs, CD, and M C D spectral analysis. Collectively, these studies revealed that D F T is well suited for obtaining reasonable electronic structure descriptions for all biologically relevant forms of Cbl. Jensen also used the Abs spectrum of Co 1 + Cbl to validate his computed electronic structure description for this species [60], which he obtained by employing the CASSCF method [25]. In this study, the complete active space second-order perturbation theory (CASPT2) method was employed to compute the electronic transition energies and Abs intensities for C o 1 + C b l . Although the CASPT2 method successfully reproduced the energies of several electronic transitions observed experimentally, it failed to predict the correct number of electronic transitions contributing to the visible region of the Abs spectrum [42]. This example illustrates that while MC-SCF methods are generally more accurate than D F T methods, it may not be possible to include all of the relevant frontier MOs into the CASSCF active space for large bioorganometallic species due to current technological limitations.

3. 3.1.

FORMATION AND CLEAVAGE OF THE CO-C BOND OF COBALAMIN IN ENZYME ACTIVE SITES Overview

Vitamin B 1 2 deficiency, which may be due to a number of factors, is a serious medical condition that causes megaloblastic anemia, hyperhomocystinuria, and methylmalonic aciduria [61,62]. In humans, these clinical symptoms can be triggered by defects in three distinct enzymes; namely, an ATP:corrinoid adenosyltransferase (ATR) [63,64], methylmalonyl-CoA mutase ( M M C M ) [65-67], and methionine synthase (MetH) [68,69]. A T R catalyzes the final step in the conversion of vitamin B 12 to its physiologically active derivative Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

427

adenosylcobalamin (AdoCbl or coenzyme B12) by transferring the adenosyl (Ado) group from a molecule of adenosine-5'-triphosphate (ATP) to Co 1 + Cbl. The AdoCbl product generated in this process is utilized by MMCM to catalyze the radical rearrangement of methylmalonyl-CoA (MMCoA) to succinyl-CoA (SCoA), which is a key step in the catabolism of branched-chain amino acids, odd-chain fatty acids, and cholesterol for entry into the Krebs cycle. Alternatively, MetH utilizes the MeCbl derivative of vitamin B 12 for synthesizing methionine (Met) by transferring a methyl cation (CH3") from methyltetrahydrofolate to homocysteine (Hey). This reaction is essential for maintaining the proper distribution of cellular folate derivatives and preventing the accumulation of Hey. Due, in part, to the fact that the reactions catalyzed by these three enzymes provide rare examples of organometallic transformations occurring in Nature, they have been the subjects of numerous computational investigations aimed at elucidating key steps in the corresponding catalytic cycles. The physiologically active forms of the Co 3 + Cbl cofactor, AdoCbl and MeCbl, contain a low-spin Co 3 + ion that is equatorially ligated by four nitrogens from the corrin macrocycle and axially coordinated by a nitrogen from 5,6-dimethylbenzimidazole (DMB) on the α-face and a carbon atom from the Ado moiety or the CH 3 group on the ß-face (Figure 1) [57]. The formation of the axial Co-C bonds in AdoCbl and MeCbl involves the nucleophilic attack of a transient Co 1 + Cbl species on the Ado moiety of ATP and the C H ^ group of a methylated substrate, respectively. Enzymes utilizing AdoCbl induce Co-C bond homolysis of the cofactor to produce Co 2 + Cbl and an organic radical centered on the 5'-carbon of the Ado moiety (Ado - ) that is capable of abstracting a hydrogen atom from substrate as the first step in a protein-mediated substrate rearrangement reaction. Alternatively, the MeCbl-dependent enzymes catalyze methyl transfer reactions in which the cofactor's Co-C bond is cleaved heterolytically, leaving behind a protein-bound Co 1 + Cbl species. The remainder of this section is devoted to a discussion of how computational studies have enhanced our understanding of Co-C bond formation and cleavage by enzymatic systems.

3.2.

Co-C Bond Formation During Adenosylcobalamin Biosynthesis and Methylcobalamin Activation by Methionine Synthase

Since aquacobalamin is thought to be the circulating form of Cbl in the body, axial Co-C bonds of AdoCbl and MeCbl must be formed in vivo in reactions that are catalyzed by ATR and MetH, respectively. In each case a transient Co 1 + Cbl species is generated in the enzyme active site that is a sufficiently strong nucleophile to abstract the Ado moiety of ATP or a methyl group Met. Ions Life Sei. 2009, 6, 417-460

428

LIPTAK, V A N H E U V E L E N , and B R U N O L D X = A d o or C H 3

Figure 1. Chemical structure and numbering scheme for the physiologically active forms of Co 3 + Cbl, AdoCbl and MeCbl. In aqueous solution, the coordination number of the Co ion is directly related to the oxidation state [106]. Upon reduction of Co 3 + Cbl to Co 2 + Cbl, the C o - X bond is cleaved. Further reduction of Co 2 + Cbl to Co 1 + Cbl also eliminates the axial C o - N bond. Note that in cobinamides the nucleotide loop and D M B base are absent and a water molecule is axially coordinated to the Co ion in the +3 and + 2 oxidation states (corresponding to C o 3 + C b i + and C o 2 + C b i + , respectively).

f r o m adenosylmethionine (AdoMet). While these C o - C b o n d f o r m a t i o n processes are relatively s t r a i g h t f o r w a r d b e c a u s e C o 1 + C b l is o n e of t h e m o s t p o t e n t n u c l e o p h i l e s k n o w n [71,72], t h e g e n e r a t i o n of a n e n z y m e - b o u n d C o 1 + C b l species via t h e r e d u c t i o n of C o 2 + C b l b y a physiologically available Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF

BIOORGANOMETALLICS

429

reductant poses a significant thermodynamic challenge. Even removal of the D M B moiety and its replacement by a solvent-derived water molecule (a weaker σ-donor ligand) to produce base-off C o 2 + C b l is expected to raise the reduction potential by a mere 120 mV, to -490 mV versus SHE in aqueous solution (corresponding to the value reported for cob(II)inamide ( C o 2 + C b i + ) [73]). This value is still significantly more negative than the reduction potential of nicotinamide adenine dinucleotide phosphate ( N A D P H , -320 mV versus SHE [74]), the ultimate source of the reducing equivalents under physiological conditions. Therefore, A T R and MetH must have devised a different strategy for raising the C o 2 + C b l reduction potential to a sufficient degree that allows these enzymes to generate a C o 1 + C b l intermediate and, thus, to catalyze C o - C bond formation. Because the redox-active Co 3dZ2-based molecular orbital of C o 2 + C b l is oriented along the axial coordination sites, which are singly occupied in C o 2 + C b l and vacant in C o 1 + C b l (Figure 1), it is to be expected that enzymatic manipulation of the C o 2 + / 1 + reduction potential is accomplished by a destabilization of the C o 2 + C b l state rather than a stabilization of the C o 1 + C b l state. Indeed, Abs, M C D , and EPR spectroscopic data obtained for free and A T R ( + A T P ) - b o u n d C o 2 + C b l exhibit significant differences, most notably the appearance of a prominent feature in the near-IR region of the M C D spectrum and widely spread resonances in the low-field region of the EPR spectrum when the cofactor binds to the enzyme active site. To interpret these changes in terms of geometric and electronic structural perturbations of C o 2 + C b l , we have used our spectroscopically-validated D F T computations as the basis for developing a spectro/structural correlation. Specifically, we have carried out a series of D F T and T D - D F T calculations to predict how distortions of the axial ligand-Co 2 + bonding interaction affect the Abs, M C D , and EPR spectra of C o 2 + C b l [37,75], This correlation has enabled us to interpret the observed spectral changes accompanying the binding of C o 2 + C b l to A T R complexed with A T P in terms of the formation of an essentially four-coordinate C o 2 + C b l species that lacks any significant axial bonding interactions [75,76]. In this "activated" C o 2 + C b l species, the C o 2 + / 1 + reduction midpoint potential is raised by an estimated 250 mV, which is sufficient to ensure that Co 2 + ->-Co 1 + reduction can be accomplished in vivo. Using the same spectro/structural correlation for interpreting the Abs, M C D , and EPR spectroscopic data obtained for C o 2 + C b l bound to MetH locked into the cofactor activation conformation, we found that the C o - N bond on the α-face of the cofactor is broken and a water molecule is axially coordinated to the C o 2 + center on the ß-face [77]. Furthermore, the axial 2+ C O - 0 ( H 2 ) bond of the MetH-bound C o C b l species is elongated relative to 2+ + that of C o C b i (a model of base-off C o 2 + C b l , see Figure 1), presumably due to a hydrogen-bonding interaction with the phenolic group of the Y1139 Met. Ions Life Sei.

2009,

6, 417-460

430

LIPTAK, VAN HEUVELEN, and BRUNOLD

Figure 2. Enzyme-induced destabilization of the C o 2 + C b l state should raise the C o 2 + C b l reduction potential into a physiologically-accessible range.

residue that is located above the ß-face [78]. Collectively, these perturbations to the axial coordination environment of Co 2 + Cbl in MetH should be sufficient to raise the redox potential above that of C o 2 + C b i + and thus into a physiologically accessible range (Figure 2).

3.3.

Co-C Bond Homolysis in AdenosylcobalaminDependent Enzymes

Of the different Co-C bond formation and cleavage processes discussed in this section, the enzyme-catalyzed C o - C bond homolysis of AdoCbl has by far received the most attention from the computational chemistry community. Since the last comprehensive review in 2001 [79], several research groups have engaged in theoretical studies that were aimed at elucidating the molecular mechanism of C o - C bond activation by AdoCbl-dependent enzymes [70,80-95]. The degree of Co-C bond activation accomplished by these enzymes is astonishing; in the case of M M C M , the rate of homolytic Co-C bond cleavage is elevated by as much as 10 12 -fold over that of the free cofactor [96]. This trillion-fold acceleration of Co-C bond homolysis by M M C M corresponds to a ~17kcal/mol reduction in the Co-C bond Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

431

dissociation enthalpy of AdoCbl in the enzyme active site, which could be achieved by a destabilization of the AdoCbl "ground state", a stabilization of the transition state, or both. Because Co-C bond homolysis is an endergonic process, the energy of the transition state could also be lowered indirectly, via a stabilization of the post-homolysis products Co 2 + Cbl and the A d o ' radical, according to Hammond's postulate. Therefore, enzymatic activation of the Co-C bond for homolysis can potentially be effected by any combination of AdoCbl destabilization, transition-state stabilization, and post-homolysis product stabilization. To explore how M M C M and a related enzyme, glutamate mutase (GM), activate the Co-C bond of AdoCbl for homolysis, we have carried out detailed spectroscopic studies of these enzymes and interpreted the corresponding data within the framework of spectro/structural correlations developed for AdoCbl and Co 2 + Cbl [81-83], While D F T computations predict that alterations in the axial bonding interactions or corrin fold angle should lead to fairly large shifts of the dominant electronic transition of AdoCbl, no such shifts were actually observed in our Abs, CD, and M C D spectra upon AdoCbl binding to M M C M or GM, even in the presence of substrate (analogues). Alternatively, rather significant differences in the M C D spectra of protein-bound versus free Co 2 + Cbl were noted, especially in the region that is dominated by Co 3d corrin π* charge transfer transitions. Analysis of these changes within the framework of D F T computations led us to propose that M M C M and G M induce a fairly uniform stabilization of the Co 3d-based MOs of Co 2 + Cbl and, because these Orbitals are filled, a stabilization of the enzyme-bound cofactor as a whole. As such, the results obtained in these combined spectroscopic and computational studies strongly suggest that stabilization of the post-homolysis product Co 2 + Cbl, rather than destabilization of the AdoCbl "ground state", is a significant source of enzymatic Co-C bond activation. D F T (B3LYP) computations have also been used by Dölker and coworkers to investigate the effect of the M M C M active site on the Co-C bond homolysis rate of AdoCbl [85]. The computational model used in this study was derived from the crystal structure of M M C M [97] and included the full Ado moiety, the MMCoA substrate, and three active-site residues. MMCoA and the protein residues were frozen in their crystal structure orientations, while the AdoCbl geometry was optimized at three points along the Co-C bond dissociation coordinate. These geometry optimizations revealed that the post-homolysis products could be stabilized by a distortion of the Ado moiety that creates a favorable dipole-dipole interaction between the Co 2 + ion and two oxygen atoms of the Ado" radical. This proposal is consistent with our hypothesis that product stabilization is a primary reason for the 1012-fold enhancement of the homolytic Co-C bond cleavage rate of AdoCbl by MMCM. Met. Ions Life Sei. 2009, 6, 417-460

432

LIPTAK, VAN HEUVELEN, and BRUNOLD

Most recently, two laboratories have utilized the Q M / M M methodology to investigate the effects of the G M and M M C M active sites on the AdoCbl Co-C bond homolysis rate. Jensen and Ryde have used Q M / M M computations to scan the potential energy surface of the C o - C bond of GM-bound AdoCbl from the ground state to the biradical product state [87]. These calculations confirmed that the Co 2 + Cbl/Ado" biradical product state is stabilized by an electrostatic interaction between the Co 2 + ion and two oxygen atoms from the Ado" radical. They revealed further that an even larger contribution to the activation of the C o - C bond might involve the differential stabilization of the Ado moiety in the AdoCbl "ground state" and product states. Subsequent Q M / M M calculations performed by Paneth and coworkers indicated that essentially the same strategy for C o - C bond activation of AdoCbl is employed by M M C M [92]. This study also successfully located the transition state along the C o - C bond dissociation coordinate. While the corrin ring conformation is minimally perturbed in the transition state, the Ado moiety is significantly distorted from its ground state structure (Figure 3). This result is consistent with the emerging view that the differential stabilization of the Ado moiety and stabilization of the Co 2 + Cbl post-homolysis product are the primary contributions to the remarkable acceleration of C o - C bond homolysis by AdoCbl-dependent enzymes.

3.4.

Co-C Bond Heterolysis in MethylcobalaminDependent Enzymes

Compared to the Co-C bond homolysis step that is common to the catalytic cycles of all AdoCbl-dependent enzymes, the mechanism of enzyme-catalyzed Co-C bond heterolysis of MeCbl has received considerably less attention from the computational chemistry community [78,86,89,98-101]. While heterolytic Co-C bond cleavage in MeCbl formally produces a C H ^ cation (along with Co 1+ Cbl), enzymes couple this process with substrate methylation; hence, the C H ^ cation does not actually represent an isolable intermediate [102]. This kinetic coupling complicates a theoretical analysis of the reaction profile of Co-C bond heterolysis by MeCbl-dependent methyl transferases, because both reactants, MeCbl and substrate, must evidently be considered when modeling this reaction. The intimate role that substrate plays in the methyl transfer reaction also suggests that there may not be a consensus mechanism employed by MeCbl-dependent methyl transferases. One of the best characterized members of this class of enzymes is MetH, which catalyzes a methyl group transfer from MeCbl to Hey to produce Co 1 + Cbl and Met [68,69]. MetH has been estimated to increase the rate of methyl transfer by 106-fold [103,104], with a 102-fold rate enhancement Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL S T U D I E S OF B I O O R G A N O M E T A L L I C S

433

Figure 3. The AdoCbl ground state (gray) and transition state (blue) of M M C M as computed by Q M / M M . Reproduced from [92] with permission form the American Chemical Society, copyright (2006).

stemming from MeCbl binding and the remaining 10 -fold acceleration arising from Hey binding [105]. These results indicate that the mechanism employed by MetH to achieve the million-fold rate enhancement for Co-C bond heterolysis involves activation of both the cofactor and the substrate. Activation of the Met. Ions Life Sei. 2009, 6, 417-460

434

LIPTAK, VAN HEUVELEN, and BRUNOLD

Co-C bond of MeCbl could potentially entail an enzyme-induced elongation of the axial ligand-Co bond on the α-face (i.e., trans to the Co-CH 3 bond, see Figure 1) of the cofactor, given that the preferred coordination numbers of the Co ion in MeCbl and Co 1 + Cbl are six and four, respectively [106], Computational support for this proposal was provided by a D F T (B3LYP) study performed by Dölker et al. in which the energy of Co-C bond heterolysis was found to decrease as a function of the trans C o - N bond length [86]. Moreover, evidence for the existence of a long axial ligand-Co bond opposite to the Co-CH 3 bond has been obtained for at least one member of the family of methyltransferases. Specifically, in collaboration with other groups, we have performed a detailed characterization of the corrinoid/iron-sulfur protein (CFeSP), a MeCbl-dependent enzyme that binds the cofactor in the "base-off form. A quantitative analysis of the spectroscopic data obtained for CFeSP within the framework of D F T and TD-DFT computations revealed that the axial C o - O H 2 bond of CFeSPbound MeCbl is elongated by ~ 0.2 Ä relative to that of MeCbi + [41,83,100]. This axial bond elongation should greatly facilitate the methyl group transfer from CFeSP-bound MeCbl to the "proximal" Ni center of the Α-cluster of acetyl-coenzyme A synthase (see Section 5). Two additional strategies that could potentially be employed by MetH to enhance the rate of methyl transfer have been explored computationally by Jensen and Ryde [101]. This study utilized D F T (B3LYP) and a suitably truncated model system to compute the transition state for the methyl transfer from MeCbl to Hey under several different reaction conditions (Figure 4). To justify the use of D F T in this study, the authors demonstrated that they achieved excellent agreement between the computed and experimental rates for the reaction in the absence of MetH by using the same computational methodology [103]. The first alternative strategy explored in this study involved Hey deprotonation by enzyme-bound Zn [107,108]. Importantly, the D F T predicted activation energy for the methyl transfer reaction was found to be 22kcal/mol larger for neutral Hey than for its deprotonated (i.e., thiolate) form, suggesting that for the reaction to occur, substrate must first be deprotonated. The second strategy considered was the creation of a hydrophobic environment in the enzyme active site. The D F T computational results obtained by Jensen and Ryde [101] indicate that the lower dielectric constant expected for a hydrophobic active site can increase the rate of methyl transfer by up to 10 14 -fold. Therefore, the 106-fold rate enhancement by MetH could readily be achieved by deprotonation of Hey and the creation of a hydrophobic active site. To this end, it is interesting to note that our spectroscopic studies of MetH locked into a catalytic conformation revealed that the active site is indeed inaccessible to solvent [75,77], Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

1

435

H 3 C.

H3C

k

H,C

CH 3

ύ

.0

y

ύ

Figure 4. The truncated model used in a D F T computational study of the methyl transfer reaction catalyzed by MetH. Reproduced from [101] with permission from the American Chemical Society, copyright (2003).

4. 4.1.

ORGANOMETALLIC CHEMISTRY AND CATALYTIC CYCLE OF METHYL-COENZYME Μ REDUCTASE Overview

Methyl-coenzyme Μ reductase (MCR) catalyzes the final step of methanogenesis in anaerobic archaea, a process that generates ~ 109 tons of methane each year [109]. Crystallographic studies revealed that MCR is a 300 kDa hexamer of the form a2p2Y2 that contains two equivalents of the F 4 3 0 cofactor, which are separated by 50 A [110]. This cofactor consists of a Ni center that is equatorially ligated by four nitrogens from the highly reduced hydrocorphin macrocycle and axially coordinated by 0(Glnl47) in the lower position and a variable ligand on the upper side (Figure 5) [111]. Access to the active site is controlled by a 30 A channel that both excludes solvent and exerts steric control over the substrates, methyl-coenzyme Μ (Me-SCoM) and coenzyme Β (HSCoB). In the first step of the catalytic cycle, Me-SCoM proceeds through the access channel and binds to the active site parallel to the F 4 3 0 cofactor. Next, the thiol moiety of HSCoB is positioned approximately 8 A above the Ni center by the amino acid residues along the channel [110]. The steric constraints imposed by the tunnel play a significant role in governing reactivity, as evidenced by the fact that enzymatic activity declines sharply when substrate analogues are used in which the length of the carbon chain differs from that of HSCoB [112]. Met. Ions Life Sei. 2009, 6, 417-460

LIPTAK, VAN HEUVELEN, and BRUNOLD

436

"O2CH2CH2Q h

2

no

2

ch

2

c """"CH2CH2C02-

CH2CH2C02 CH2CH2C02

CH2CH2-Gln147 Cofactor F430

methyl-coenzyme Μ (Me-SCoM)

Figure 5. Top: Chemical structure of the cofactor F 4 3 0 in the active Ni(I)MCR R e d l form of MCR, in which the lower axial position is occupied by 0(Glnl47) and the upper axial coordination site is vacant. Bottom: Chemical reaction catalyzed by MCR. Numerous inactive forms of the enzyme, generated by the reaction of M C R with H S C o M and H S C o B , have been characterized and revealed that F 4 3 0 can support + 1 , + 2 , and + 3 oxidation states. Ni(II)MCRsiient is coordinated to the heterodisulfide C 0 M - S - S - C 0 B via the sulfonate oxygen of C o M , whereas Ni(II)MCR 0 x i-silent binds the thiol of C o M in the upper axial position of F 4 3 0 [110,111]. Cryoirradiation of Ni(II)MCR 0 x i-siient yields N i M C R 0 x i , which can be described as Ni(II) ion coupled to a thiyl radical in resonance with a Ni(III) thiolate [113]. N i ( I ) M C R R e d i , in which the upper axial position is vacant, serves as the active form of the enzyme. Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

437

Each form of M C R has unique spectroscopic signatures that provide a means of identifying the metal oxidation state and assessing the feasibility of computational models. The Ni(II)-bound forms of M C R exhibit a dominant Abs feature at 430 nm (F 4 3 0 derives its name from this characteristic electronic transition) that undergoes a blue-shift to 378 nm upon reduction of the metal ion to the Ni(I)MCR R e di state. M C D spectroscopic studies revealed that the Ni(I) form of the protein displays multiple low-energy transitions that are absent in the Ni(II)-bound states [113]. The vibrations of the tetrapyrrole ring, which can be probed by rR spectroscopy, also provide sensitive oxidation state markers, with key features downshifting upon nickel ion reduction [114]. Computational studies aimed at exploring the origin of these trends are discussed in Sections 4.2 and 4.3.

4.2.

Properties of the Free F 4 3 0 Cofactor

Early computational studies of M C R focused on the isolated F 4 3 0 cofactor, with the initial goals of identifying the structural and electronic characteristics that allow F 4 3 0 to access multiple Ni oxidation states and evaluating the degree of ruffling in the cofactor. Nonplanar distortions are known to occur in a wide variety of tetrapyrrole-type species, both in natural systems such as heme and in synthetic complexes, and such distortions can significantly influence the corresponding electronic properties [115,116]. In 1991, Farber et al. crystallized a pentamethylester derivative of the native cofactor [117]. When F 4 3 0 is isolated from M C R , it converts to its diepimer form, eventually reaching an equilibrium of 88% conversion to the catalytically-inactive and highly ruffled 12,13-diepimer ( F 4 3 0 D i e p i ) [118]. Unlike the diepimer, native F 4 3 0 is relatively planar [119] and has N i - N bonds that are longer than those of its ruffled counterpart (1.99-2.14 A in F 4 3 0 as compared to 1.96-1.99 A in F 4 3 0 D i e p i ) [118]; these longer bonds in turn stabilize the Ni(I) oxidation state. Similarly, the longer N i - N bonds in the native cofactor are more conducive to high spin Ni(II) than the smaller low spin Ni(II) ion [118]. D F T geometry optimizations of the native cofactor and F 4 3 0 D i e p i accurately reproduced this trend in N i - N bond lengths [110]. The planar geometry necessary for the catalytic function is preserved by the steric influence of amino acid residues within the M C R active site, a finding supported by both D F T and molecular mechanics studies [118-121]. Computational data also support the somewhat counterintuitive claim that F 4 3 0 can stabilize the Ni(III) oxidation state, even though it favors high-spin Ni(II) over its smaller low-spin counterpart. D F T studies by Wondimagegn and Ghosh [119] of [Ni(III)(F 430 )(ClO 4 )(Am)] + and [Ni(III)(F 430 )(CH 3 )(Am)] + (Am = O-bound acetamide) revealed that the upper axial ligand determines the nature of the singly occupied Ni 3d orbital. The methyl moiety, which is a Met. Ions Life Sei. 2009, 6, 417-460

438

LIPTAK, V A N HEUVELEN, and B R U N O L D

strong σ-donor, results in a preferential ( 3 d X 2 _ y 2 ) 1 electron configuration, consistent with EPR studies of M C R 0 x i [122], whereas the Perchlorate ligand favors a (3d Z 2electron configuration. Further, a geometry optimization of the [Ni(III)(F 430 )(CH 3 )(Am)] + model resulted in N i - N bond lengths similar to those found in F 4 3 0 , as expected by considering that in each case the Ni 3 d 2 _ based orbital is singly occupied. Collectively, these results indicate that the long N i - N distances favored by the planar cofactor are crucial for supporting oxidation states ranging from Ni + to N i 3 + [123], Because of the large size of F 4 3 0 , truncated models have typically been used in computational studies of this species. To assess the minimal model size needed for accurately reproducing the structural and electronic properties of the complete F 4 3 0 cofactor, Pelmenschikov et al. [124] performed D F T computations on two differently truncated models. This study revealed that a relatively small model with a simplified 12-carbon macrocycle is sufficient to preserve the key electronic properties of F 4 3 0 . However, a larger model in which the pyrrole and lactam rings were preserved and the propionate and acetate groups replaced with hydrogen atoms was needed to properly describe the steric influences imposed by neighboring amino acid residues and subtle conformational changes in the cofactor during catalysis [124,125]. Consistent with these findings, our D F T and T D - D F T computations for models of F 4 3 0 that contained the pyrrole rings accurately reproduced the experimental Abs spectra and EPR g and A tensors, even when the lactam ring was replaced with hydrogen atoms [126]. Collectively, the results obtained in these studies demonstrate that the pyrrole rings must be preserved in computational models of F 4 3 0 in order to accurately reproduce experimental observations, whereas truncations of peripheral side chains and even the lactam ring are relatively unproblematic. Additionally, these studies established D F T as a reliable method for computing structures, energies, and electronic properties of the F 4 3 0 cofactor. X

y 2

The results obtained in D F T studies of the free F 4 3 0 cofactor also permitted us to develop an MO-based explanation for an intriguing spectroscopic property of this species; namely, that the dominant Abs feature in the visible spectral region undergoes a rather dramatic blue-shift from 430 to 380 nm upon reduction of nickel ion from the +2 to the +1 oxidation state [126]. On the basis of our spectroscopically-validated bonding descriptions obtained from D F T computations on Ni(II)F 4 3 0 to Ni(I)F 4 3 0 models, this feature can be assigned to a hydrocorphin-centered π - > π * transition. In the oxidized Ni(II)F 4 3 0 species, the Ni-based 3d orbitals are considerably lower in energy than the π-based frontier orbitals of the macrocycle and thus do not participate in electronic transitions contributing to the visible region of the Abs spectrum. However, upon one-electron reduction of the Ni(II) ion, the occupied Ni 3d orbitals are raised in energy, shifting between, and strongly mixing with, the hydrocorphin π and 7i*-based MOs in Ni(I)F43Q. Met. Ions Life ScL 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

439

These ground-state changes have a dramatic effect on the excited-state structure, causing a blue-shift of the dominant π->π* transition and the appearance of numerous Ni 3d hydrocorphin π* charge-transfer features in the visible/near-IR region that give rise to a series of prominent features in the corresponding MCD spectrum [126]. Hence, Abs and MCD spectroscopic studies are well suited for assessing the Ni oxidation state in each of the plentiful known states of MCR-bound F 430 , especially when combined with EPR experiments.

4.3.

Ni-Alkyl Species of Methyl-Coenzyme Μ Reductase: Computational Evaluation of Viable Reaction Mechanisms

The rapid kinetics associated with the reaction catalyzed by MCR has thus far precluded the observation of any intermediates. Therefore, models of viable reaction intermediates have been prepared using substrate analogues and characterized by spectroscopic methods, and computations have been carried out to evaluate possible reaction mechanisms. These studies led to the proposal of two fundamentally different mechanisms for the catalytic cycle of MCR. Mechanism 1 involves a Ni(III)-Me intermediate, whereas mechanism 2 instead invokes the intermediacy of a methyl radical and a Ni(II)-thiol species [127], 4.3.1.

Ni-Methyl

Pathway

Analogous to known cobalamin chemistry (Section 3.4), mechanism 1 invokes an organometallic intermediate. Here, Ni(I) performs a nucleophilic attack on the methyl moiety of Me-SCoM, resulting in a putative Ni(III)-CH3 intermediate [128,129]. Two variants of this mechanism have been suggested in the literature. In the scheme put forward by Ragsdale and coworkers on the basis of steady-state kinetic data, the first step in the reaction involves the generation of a S'CoB radical species [129]. The second variation, favored by Thauer and coworkers (Figure 6, top), instead invokes the intermediacy of an S~CoB anion [128]. According to this latter mechanism, residues in the substrate access channel, namely Tyr333 and Tyr367, serve as proton donors to the reactants at key steps along the reaction pathway. Both of these mechanisms are consistent with structural data. Recent studies of MCR using substrate analogues revealed that the formation of alkyl-Ni(III)F43o species is chemically viable. Specifically, the reaction of Ni(I)MCR Re di with bromopropanesulfonate (BPS) was found to yield a Ni(III)-propyl sulfonate species, termed MCR PS , that has EPR properties Met. Ions Life Sei. 2009, 6, 417-460

440

LIPTAK, V A N HEUVELEN, and B R U N O L D

[125,128], similar to M C R 0 x i [130]. Similarly, the reaction of Ni(I)MCR R e d l with CH 3 X (X = Br, I) was shown to generate a Ni(III)-CH3 species known as M C R M e [131,132]. EPR and electron-nuclear double resonance (ENDOR) studies established that, like [Ni(III)(F43 0 )(CH 3 )(Am)] + (discussed in Section 4.2), MCRMe has a (Sd^yj) 1 ground state [131], Both alkyl-Ni(III)MCR species mimic the reactivity of native reaction intermediates of MCR; Ragsdale and coworkers demonstrated that M C R M e c a n be reacted with HSCoB to produce methane [131], and M C R P S can be treated with a thiolate to regenerate Ni(I)MCR R e d l [133], However, theoretical studies have questioned the viability of mechanism 1. Crabtree and co-workers used D F T computations to quantify the energy Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

441

barrier associated with the initial reaction step; namely, the breaking of the C S bond in Me-SCoM to form an alkyl-Ni(III) species. For the two forms of the putative organometallic intermediate considered, CH 3 -Ni(II)F 4 3o and CH 3 -Ni(III)F 4 3o + , D F T yielded N i - C binding energies of 24.9 and 18.0 kcal/ mol, respectively. In comparison, the C - S bond strength in Me-SCoM was calculated to be 70kcal/mol [124]. On the basis of these results, it would be expected that the methyl transfer from Me-SCoM to Ni(I)MCR R e d i is highly endothermic, by at least ~ 45 kcal/mol. This finding led to the proposal of mechanism 2 described next [124].

4.3.2.

Methyl

Radical

Pathway

The obvious shortcomings associated with mechanism 1 led to the suggestion that the nucleophilic attack of Ni(I)MCR R e di on Me-SCoM causes the C - S bond to cleave homolytically (as opposed to heterolytically, as implicated in mechanism 1) to generate a methyl radical and a CoM-S-Ni(II)F 4 3 o species (Figure 6, bottom). Methane is then produced through hydrogen atom abstraction from HSCoB, a process that generates a S'CoB radical species that subsequently reacts with CoM-S-Ni(II)F 4 3 o to form the mixed disulfide C o M - S - S - C o B and restore N i ( I ) M C R R e d l [118,124,125,134], Mechanism 2 was evaluated computationally by Siegbahn and coworkers [124,125]. Importantly, the computed N i - S bond strength of 39kcal/mol for the putative C o M - S - N i ( I I ) F 4 3 0 species that is formed in this process is much closer to the 70 kcal/mol needed to cleave the S - C bond in Me-SCoM (vide supra). In further support of mechanism 2, the D F T computed energy barrier associated with this mechanism is only 20 kcal/mol and the overall reaction was found to be exothermic [124,125]. A somewhat troubling aspect of mechanism 2 is that it relies on the direct transfer of the methyl radical from Me-SCoM to HSCoB over a relatively large distance. Although the F 4 3 0 cofactor and the active-site tunnel may provide the necessary steric restrictions for the controlled transfer of a reactive radical to occur, both substrates are held in place through a series of hydrogen bonds, such that their thiol groups are separated by ~ 6 . 5 A [133]. Another potential problem with mechanism 2 is that the driving force associated with the disulfide bond formation between two distant thiol moieties may not be sufficient to drive the N i - S bond cleavage in the putative CoM-S-Ni(II)F 4 3 0 species [134]. Lastly, it should be noted that experimental studies of M C R using the substrate analogue ethyl-SCoM revealed that an inversion of configuration at the reactive carbon occurs on reduction; thus, if a methyl radical is indeed formed in the M C R catalytic cycle, then it must be quenched by HSCoB on a timescale that is sufficiently short to prevent it from rotating [127,135]. Met. Ions Life Sei. 2009, 6, 417-460

442

LIPTAK, VAN HEUVELEN, and BRUNOLD

4.3.3.

Protonation of

Ni(I)MCRRedl

Recently Duin and McKee [136] proposed a third type of mechanism for M C R based on a D F T computational study. This mechanism relies on the protonation of Ni(I)MCR R e di, at either the Ni(I) site or the tetrapyrrole ring, followed by oxidative addition of Me-SCoM. The disulfide bond between MeSCoM and HSCoB is formed through a two-center three-electron interaction between the two thiol moieties, leaving the liberated methyl group on the protonated NiF 4 3 0 cofactor. Protonation of the methyl group then leads to release of methane and regeneration of Ni(I)MCR R e di· Some novel aspects of this mechanism, such as the displacement of Ni from the tetrapyrrole ring by 1.5 A and the coordination of Ni to both the thiol and the methyl of MeSCoM, bear further experimental and computational investigation [136].

5.

5.1.

GEOMETRIC AND ELECTRONIC STRUCTURES OF THE CARBON MONOXIDE DEHYDROGENASE/ ACETYL-COENZYME A SYNTHASE ACTIVE SITE CLUSTERS Overview

The bifunctional enzyme CO dehydrogenase/acetyl-coenzyme A synthase (CODH/ACS) is crucial for the autotrophic growth of some archaea and eubacteria. The C O D H / A C S from the anaerobic mesophile Moorella thermoacetica (formerly called Clostridium thermoaceticum) [137], which is probably the best studied member of this class of enzymes, is a 310 k D a tetramer of configuration α 2 β 2 [138,139]. Each aß dimer contains four polynuclear metal clusters that are labeled A- through D-clusters [140-142]. The B-, C-, and D-clusters reside in the ß-subunit and are involved in the reversible reduction of C 0 2 to CO (equation 1) for subsequent generation of acetyl-Co A at the Α-cluster. The B- and D-clusters each consist of relatively typical [Fe 4 S 4 ] cubanes and act as a "redox wire", shuttling the electrons necessary for the reduction of carbon dioxide between the C-cluster and external reducing agents. The C-cluster itself contains an [Fe 3 S 4 ] core that is connected to a N i - S - F e moiety (Figure 7, bottom). Remarkably, the CO generated at the C-cluster via C 0 2 reduction passes to the Α-cluster through a 138 Ä intramolecular tunnel within C O D H / A C S [143], 2e" + 2H+ + C 0 2 ^ CO + H 2 0

(1)

The C-cluster can be stabilized in multiple oxidation states, including COXj Credi, C red2 , and C int . The electron distribution among the multiple redox-active sites of the C-cluster in these states remains incompletely understood. Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

443

Nevertheless, spectroscopic studies revealed that C ox is an EPR-silent (presumably S = 0) species, as is Cj,,,. In contrast, both C redl and C red2 are S= 1/2 systems that can be readily distinguished on the basis of their characteristic g tensors. Redox titrations demonstrated that the one-electron reduction of C ox generates C redl , which can undergo sequential one-electron reductions to yield first C tot and then C red2 [144-146], The Α-cluster, found in the α subunit, catalyzes the synthesis of acetylCoA according to the following reaction: CH 3 -CFeSP + CO + Co A ^ acetyl-CoA + CFeSP

(2)

where CH 3 -CFeSP and CFeSP are the methylated and de-methylated forms, respectively, of the corrinoid/iron-sulfur protein (see Section 3.4). Two

C cluster

Figure 7. (bottom).

Chemical structures of the CODH/ACS Α-cluster (top) and C-cluster

Met. Ions Life Sei. 2009, 6, 417-460

444

LIPTAK, VAN HEUVELEN, and BRUNOLD

relatively recent X-ray crystallographic studies [147,148] revealed that the Acluster consists of an [Fe 4 S 4 ] cubane that is linked through a cysteine bridge to a tetrahedral, proximal Ni site (Ni p ), which in turn is connected to a square-planar, distal Ni site (Ni d ) via two cysteine bridges (Figure 7, top). While initially the presence of two nickel centers at the Α-cluster came as a surprise, it was soon recognized that these structures provided the necessary framework for understanding previously un-interpretable spectroscopic data. Additionally, they provided definitive clues as to the origin of the inherent heterogeneity of the A cluster, alternately showing Zn and Cu (in addition to Ni) at the proximal metal site, which is therefore referred to as the labile site. Subsequent studies afforded compelling evidence that Ni is present in the active form of ACS [145,147,148], Collectively, spectroscopic studies on C O D H / A C S revealed that the Acluster is stable in at least three states: the as-isolated oxidized state (A ox ) in which both metal centers are low-spin Ni(II), a one-electron reduced state (A red ), and a CO-bound, one-electron reduced state (A red -CO) [137]. Conversion of A o x to the A red state leads to the development of an S = 3/2 EPR signal that is characteristic of a reduced [Fe 4 S 4 ] 1+ cubane, indicating that the reducing equivalent localizes on the FeS cluster. The A r e d -CO state exhibits the well-characterized S = 1/2 NiFeC EPR signal, so named for the hyperfine broadening associated with labeled 61 Ni, 5 7 Fe, and 1 3 CO [149]. The relevance of this A r e d -CO species to catalysis is debated (see Section 5.3).

5.2.

Mechanism of Carbon Monoxide Oxidation/Carbon Dioxide Reduction by Carbon Monoxide Dehydrogenase

The mechanism of C O D H is most commonly discussed as the oxidation of CO, as many enzymes, such as the Mo/Cu-dependent C O D H and the monofunctional Ni-dependent C O D H (which lacks the α-subunit and thus the ability to synthesize acetyl-CoA formation), generate C 0 2 . As carbon monoxide metabolism was recently reviewed [145], we provide only a brief overview here. Although the complex geometric and electronic structures of the C-cluster greatly complicate experimental studies of reaction intermediates, it is generally believed that CO and H 2 0 bind to the Ni(II) and Fe(II), respectively, of the N i S-Fe moiety in the C r e d l state (note that this Fe center is referred to as ferrous component II (FCII, Figure 7, bottom) because of an early Mössbauer study [140]). The Fe OH species then attacks the N i 2 + - C O moiety to form a N i 2 + COOH intermediate, which rapidly decays by the release of C 0 2 and the elimination of a proton to yield Cred2. The transfer of two electrons through the "redox wire" regenerates C redl while reducing the B- and D-clusters, which in Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

445

turn are oxidized by ferredoxin to return the β subunit to the "ready" state [111,145,146]. Until recently, the structure and composition of the C-cluster were subjects of considerable debate, which precluded computational studies of the CODH reaction mechanism. However, the high-resolution X-ray crystal structures of the C-cluster that were recently determined [147,150] have afforded the necessary foundation for the successful completion of future computational studies aimed at elucidating the catalytic cycle of CODH at the molecular level.

5.3.

Computational Insights into the Catalytic Cycle of Acetyl-Coenzyme A Synthase

Studies of the catalytic cycle of ACS are complicated by the intricate, multimetallic nature of the active site Α-cluster. Computational studies have been instrumental in identifying the metal in the proximal site of the catalytically active Α-cluster and in discriminating between putative catalytic cycles. As mentioned in Section 5.1, early X-ray crystal structures of CODH/ACS showed Zn, Cu, and Ni in the proximal metal site of the A-cluster [145,147,148]. To determine which metal is catalytically active, we conducted a D F T study that considered two putative models of A red -CO (note that even though this species may not actually be catalytically relevant, the intensity of the NiFeC EPR signal correlates directly with enzyme activity [151]), one containing Cu and the other Ni in the proximal site [149]. In each case, a CO moiety was bound to the proximal metal site and D F T was used to carry out a constrained geometry optimization in which the positions of certain C atoms of ligating amino acid residues were kept frozen to account for the constraints imposed by the protein backbone. As both models yielded structures consistent with experimental data for the Α-cluster, the D F T / ZORA CP-SCF method was used to compute the corresponding 57 Fe, 61 Ni, and 13 CO nuclear hyperfine parameters. It was found that only the model containing Ni in the proximal site successfully reproduced all of the experimental Aiso, indicating that the NiFeC species, and thus the catalytically active Α-cluster form possesses a proximal nickel ion [149]. The NiFeC species (i.e., A red -CO state of the Α-cluster) can therefore be described as possessing a [Fe 4 S 4 ] 2 + -Nij^CO-Nid + core. It is widely accepted that the C-cluster generates CO that subsequently migrates to the Α-cluster and that CFeSP delivers a methyl cation to the Α-cluster. Additionally, because treatment of the Α-cluster with 1,10-phenanthroline (phen) abolishes catalytic activity by removing the labile Ni p center, it is generally assumed that Ni p serves as the primary site of substrate CO binding [152]; however, this is the limit of consensus regarding the mechanism of ACS. There are two general classes of proposed mechanisms to date; namely, the paramagnetic and diamagnetic pathways. In the paramagnetic mechanism the Met. Ions Life Sei. 2009, 6, 417-460

446

LIPTAK, VAN HEUVELEN, and BRUNOLD

EPR-active A red -CO state serves as an early intermediate that binds the methyl group from CH 3 -CFeSP to generate a Ni p (CO)(CH 3 ) species. Migratory insertion of CO into the Ni-CH 3 bond then yields a Ni-acetyl intermediate, and attack of this species by CoAS~ leads to the formation of the product acetyl-CoA [145,153]. In an alternate proposal that also employs a proximal Nip + / 1 + cycle, methylation precedes carbonylation [154]. Some variants of the paramagnetic mechanism involve the cycling of the iron-sulfur cluster between + 1 and + 2 states [153]; however, experimental evidence has cast doubt on this proposal, as kinetic studies revealed that the [Fe 4 S 4 ] 2+/1 + redox conversion is orders of magnitude slower than the observed reaction rate of the A-cluster [155]. An alternative to the paramagnetic mechanism is a diamagnetic scheme, in which C H ^ binds first to the proximal Ni center and all intermediates are EPRsilent, assuming a Nip + / 0 + redox cycling of the proximal nickel ion and thus invoking an unprecedented zero-valent Ni center. This description arises from the apparent two-electron reduction of A ox , a [Fe 4 S 4 ] 2+ -Nip + -Nid + species, to generate the catalytically active form of the A-cluster [156]. Additionally, Riordan and coworkers recently demonstrated that a methylated cobalt species, MeCo(dmgBF2)2py (dmgBF 2 =(difluoroboryl)dimethylglyoximato and py = pyridine), can transfer a - C H 3 moiety to a Ni° model compound, a reactivity that mirrors the Nip + / 0 + diamagnetic proposal [157]. As with the F 430 cofactor, theoretical studies of the A-cluster must carefully consider the degree of active site truncation that can be used without compromising the accuracy of computational results. Webster et al. [158] employed a model of the A-cluster in which the [Fe4S4] cubane was modeled by a mononuclear high-spin Fe(II) tetrathiolate. Interestingly, D F T computations performed on this model seemed to lend support to the diamagnetic Nip + / 0 + mechanism over the alternative Nip + / 1 + pathway [158]. However, two independent studies by us and Field et al. demonstrated that the redox-active [Fe 4 S 4 ] 2+/1 + cluster must be included in its entirety in order to obtain accurate computational results [149,159]. In particular, inclusion of the complete iron-sulfur cluster in D F T computations caused an initial [Fe 4 S 4 ] 2 + -Nip + Nid + model to optimize to a [Fe 4 S 4 ] 1+ -Nip + -Nid + species in which the ironsulfur cluster and proximal Ni are antiferromagnetically coupled, producing an overall diamagnetic species in accordance with experimental data [149]. The fact that the computed electron distributions within the A-cluster vary considerably depending on the size of the active site model used clearly warrants additional computational studies of this highly elaborate metal cluster. Field and coworkers [159] performed a series of DFT studies to evaluate proposed reaction pathways; namely, they considered the possibility of the distal Ni center acting as a binding site (the so-called binuclear pathway [153]), the order in which CH^ and CO bind to Ni, and the stability of the putative zerovalent nickel center. While their findings did not definitively discriminate between initial methylation versus carbonylation, they found the structures Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

447

associated with a mononuclear mechanism, in which both CO and C H ^ bind to Nip, to be energetically more favorable than a binuclear mechanism involving both the proximal and distal Ni centers [159]. Their findings suggest that Ni d remains in the +2 oxidation state throughout the reaction, in good agreement with experimental data. Additionally, their results revealed that while the reduced [Fe 4 S 4 ] 1+ cubane may aid in the stabilization of the zero-valent state of the proximal Ni center, in its oxidized state (which is invoked in the diamagnetic mechanism as required by experimental data) the [Fe 4 S 4 ] 2+ cluster oxidizes Nip + to Nip + , consistent with our findings described above [149,159].

6. 6.1.

MAGNETIC PROPERTIES OF THE ACTIVE SITE CLUSTER OF IRON-ONLY HYDROGENASES Overview

Central to hydrogen metabolism in aerobic and anaerobic microorganisms is a family of enzymes known as hydrogenases that catalyze the consumption and production of H 2 via the following reversible reaction [160,161]: H 2 ^ 2H+ + 2e~

(3)

There are three known classes of hydrogenases, which differ in their activesite compositions: the nickel-iron ([NiFe]), iron-only ([FeFe]), and iron-sulfur cluster-free hydrogenases [162]. The [NiFe] and [FeFe] hydrogenases are primarily used for dihydrogen oxidation and proton reduction, respectively, while the iron-sulfur cluster-free enzymes activate H 2 for use in catabolic processes. The catalytic mechanisms of the [NiFe] and [FeFe] hydrogenases have been the subjects of numerous D F T computational studies. For a summary of the results obtained in these studies, we refer the reader to an excellent review by Siegbahn, Tye, and Hall that has very recently been published [162]. Instead, this section focuses on a topic not covered elsewhere in this chapter; namely, the use of computations for elucidating the magnetic properties of a polynuclear active-site cluster. Specifically, we summarize here our computational efforts to model the exchange interactions that operate within the active-site cluster of the [FeFe] hydrogenases.

6.2.

Structure and Oxidation States of the Active Site Cluster of [FeFe] Hydrogenases

While not as widespread as [NiFe] enzymes, the [FeFe] hydrogenases have attracted a great deal of interest due, largely, to their unusual 6-Fe active-site cluster known as the Η-cluster. X-ray crystallographic studies revealed that Met. Ions Life Sei. 2009, 6, 417-460

448

LIPTAK, V A N H E U V E L E N , and B R U N O L D

the Η-cluster consists of a [Fe 4 S 4 ] H cubane linked to a [2Fe] H subcluster (Figure 8) [161,163]. Each iron atom of the [2Fe] H component is terminally coordinated by one CN~ and one CO ligand, and the two Fe centers are bridged by a dithiolate ligand and an additional CO ligand (CO b ). The proximal Fe (Fe p ) is connected to the [Fe 4 S 4 ] H cubane via a cysteine bridge, while the distal Fe (Fe d ) has a labile coordination site (L) that is the putative site of Η-binding during catalysis [162]. During catalysis, the [2Fe] H component likely cycles between Fe(II)-Fe(II) and Fe(I)-Fe(I) states, making [FeFe] hydrogenases the only known biological system to employ the Fe(I) oxidation state [164-166]. The active form of the Η-cluster is the paramagnetic ( S = 1/2) H o x state in which the [2Fe] H component possesses a mixed-valence {Fe p -Fe d } 3 + configuration. The H o x state readily binds CO, a potent inhibitor of [FeFe] hydrogenases, at the Fe d center to generate the H o x - C O form, while treatment with dithionite yields the EPR-silent H r e d state that features an {Fe p -Fe d } 2 + unit. The [Fe 4 S 4 ] H cubane is formally in its diamagnetic +2 state in all well-characterized forms of the Η-cluster; however, spectroscopic studies have shown that in the H o x and H o x - C O forms the four Fe sites of the cubane display 5 7 Fe hyperfine coupling parameters (A values) that are diagnostic of extensive spin d e r ealization from the [2Fe] H site onto the cubane [167-170]. Nearly all theoretical investigations of the Η-cluster performed to date have omitted the [Fe 4 S 4 ] H cubane and instead modeled this moiety by adding a proton to the Cys residue bound to Fep [171-174]. While this approach may be justified by the fact that the [2Fe] H subunit is presumably the site of H 2

Figure 8.

Structure of the active site Η-cluster of [FeFe]-hydrogenases.

Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

449

uptake/formation, spectroscopic evidence suggests that a high degree of electronic communication exists between the two units, and it is thus likely that the [Fe4S4]H cubane influences both the geometric and electronic structures of the [2Fe]H component. To address these issues, we employed DFT to perform the first computational investigation of the Η-cluster in its entirety and subsequently performed a detailed analysis of the magnetic properties of the H ox and Hox-CO states using DFT methods [175]. Importantly, our geometry optimizations of complete and truncated Η-cluster models revealed that the positions of the Scys and COb ligands are strongly correlated, as the shorter Fep-Scys bonds in models lacking the [Fe4S4]H cubane caused the CO b ligand to shift away from Fe p by ~0.10 A.'Thus, it appears that an important structural role of the [Fe4S4]H cluster is to lengthen the Fep-SCys bond, which in turn modulates the relative position of the CO b ligand. This finding raises some concerns as to the credibility of previous computational studies that omitted the [Fe4S4]H cubane from Η-cluster models when investigating the catalytic mechanism of [FeFe] hydrogenases.

6.3.

Calculation of the Magnetic Properties of the H-Cluster

The primary difficulty in treating the complete Η-cluster with DFT arises from the exchange interactions that operate between the two subclusters as well as within the [Fe4S4]H cubane. Even though the cubane as a whole is diamagnetic, it actually consists of two pairs of high-spin Fe 2 + (S = 2) and Fe 3 + (S = 5/2) centers [176]. Double-exchange coupling yields two ferromagnetically coupled (S= 9/2) mixed-valence pairs of Fe centers, which in turn couple antiferromagnetically to produce the S = 0 ground state observed experimentally. The Η-cluster can thus be modeled as a three-spin system with Si = Sn = 9/2 and = 1/2, where Si and Sn represent the spins of the two mixed-valence Fe dimers that comprise the cubane, and represents the spin associated with the [2Fe]H subunit in the H ox and H ox CO states (Figure 9) [175]. Consequently, this model requires two exchange parameters, / cu b e and / H , to describe the Heisenberg exchange coupling between the two Fe dimers within the cubane and between the [2Fe]H component and the adjacent Fe dimer, respectively. Using this spin-coupling model along with the broken-symmetry (BS) approach developed by Noodleman and coworkers [177,178], / cu b e and / H can be computed from the relative energies of the high-spin and BS states, as indicated in Figure 9 (note that because in DFT wave functions are represented by single determinants, only the high-spin (S = 19/2) state can be properly described by this method; however, the energies of the other pure spin states, and thus the / cu b e and / H values, can be calculated from the energies of the BS states [175]). Met. Ions Life Sei. 2009, 6, 417-460

LIPTAK, VAN HEUVELEN, and BRUNOLD

450

High-Spin States

s = 19/2

c

Three-spin model used in DFT analysis

Ms = 17/2

ΔΕ

= 9Λ

ΛΕ = (81/2)Jcube /W S =1/2(F)

BS States

M S =1/2(AF) [Fe 4 S 4 ] H 2 -

J ΛΕ = (9/2) JH

[Fe 4 S 4 ] H 2+ +- [2Fe],lH

Figure 9. Left: Schematic diagram of the "three-spin model" used to treat the H-cluster. Right: Spin state diagram showing how the relative energies of the high-spin and brokensymmetry (BS) states of the [Fe4S4]n+ cluster are modulated by ferromagnetic (F) and antiferromagnetic (AF) interactions with the S — 1 /2 spin of the [2Fe]H component.

The computed / c u b e values of ~ 400 c m - 1 for both the H o x and H ox -CO states are consistent with previous computational [177,178] and experimental [179] studies of [Fe 4 S 4 ] 2+ clusters. More importantly, the calculated Ju parameter of + 1 5 ± 1 0 c m _ 1 for the H o x state agrees remarkably well with the experimental value of ~ 2 0 c m _ 1 determined from Mössbauer studies [167,168]. Our computations also successfully reproduced the large increase in / H upon binding of CO to Fed, predicting a / H value of 150±50cm _ 1 for the H o x -CO state that closely matches the experimental value of ~ 100 cm" 1 [167,168], Given this good agreement between the experimental and computed exchange-coupling parameters, it was reasonable to use the electronic-structure descriptions provided by D F T as the basis for exploring the origin of the large increase in Ju upon CO-binding to the H-cluster [175]. This analysis revealed that in the H o x state, the unpaired spin density is almost entirely localized on the Fe d center, which lacks an effective pathway for exchange coupling to the [Fe4S4]H cubane. Alternatively, the large amount of unpaired spin density on Fe p in the H ox -CO state facilitates exchange interactions with the cubane via the bridging Cys residue. Thus, the extent of derealization of unpaired spin density over the Fe dimer determines the magnitude of the exchange coupling to the [Fe 4 S 4 ] H cluster, and the Ju value therefore serves as a useful indicator of the spin distribution within the [2Fe]H component.

7.

CONCLUDING REMARKS AND FUTURE DIRECTIONS

Collectively, the remarkable progress in computer technology, the development of advanced computational methodologies, and the dedicated efforts of enzymologists, spectroscopists, and X-ray crystallographers have established Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

451

the necessary foundation for the successful completion of computational studies aimed at elucidating the electronic structures and catalytic cycles of the bioorganometallic enzymes and cofactors. Although the majority of the computational studies completed to date relied on D F T , a stable and relatively reliable method for the quantum chemical treatment of large metalcontaining species, it can be anticipated that more sophisticated theoretical tools will be employed with increasing frequency in future studies of bioorganometallic species. Among these, the combined Q M / M M and MC-SCF approaches appear particularly well suited for evaluating viable catalytic cycles by utilizing complete protein models and for elucidating the magnetic properties of exchange-coupled polynuclear active-site clusters, respectively. Additionally, it is to be expected that theoretical chemists will continue to make improvements in the computational prediction of spectroscopic observables, thus establishing a more rigorous framework for evaluating calculated bonding descriptions, structures of hypothetical reaction intermediates, and viable catalytic mechanisms on the basis of experimental data. Computational studies have played a vital role in elucidating key steps in the catalytic cycles of enzymes involved in the biosynthesis of the B 12 cofactors as well as of AdoCbl-dependent enzymes (Section 3). In the case of the C o - C bond-forming ATP:corrinoid adenosyltransferases, combined spectroscopic/computational studies revealed that these enzymes effect the thermodynamically difficult Co 2 + ->-Co 1 + reduction through the formation of an essentially square-planar Co 2 + corrinoid intermediate, so as to stabilize the redox-active, Co 3dZ2-based molecular orbital that is oriented along the axial coordination sites of the C o 2 + ion. A consensus has also been largely reached regarding the mechanism of C o - C bond activation by AdoCbldependent enzymes; namely, that the differential stabilization of the Ado moiety and stabilization of the C o 2 + C b l post-homolysis product are the primary contributors to the remarkable acceleration of C o - C bond homolysis by these enzymes. In contrast, because the MeCbl-dependent enzymes have received much less attention from the computational chemistry community, relatively little is known to date about the mechanism of enzymecatalyzed C o - C bond heterolysis. Likewise, despite numerous kinetic, spectroscopic, and computational studies, the reaction mechanism of M C R is an enduring subject of intense debate (Section 4). Substrate-analogue studies have revealed that the formation of alkyl-Ni(III)F 4 3o species is chemically viable and that such species can react with HSCoB to generate methane and a disulfide, even as theoretical studies have cast doubt on the possibility of forming such an organometallic intermediate in the reaction with the native substrate Me-SCoM. Computational methods have also been used with great success in studies of the ACS Α-cluster (Section 5) and the [FeFe] hydrogenase Η-cluster (Section 6). In both cases, D F T has been utilized to assign the oxidation states of Met. Ions Life Sei. 2009, 6, 417-460

LIPTAK, VAN HEUVELEN, and BRUNOLD

452

individual metal ions within the active-site clusters and to evaluate key steps in the corresponding catalytic cycles. However, nearly all theoretical investigations of the Α-cluster and Η-cluster performed to date have omitted the [Fe4S4] cubane. While this approach may seem reasonable because the substrates bind to other metal centers in these clusters, spectroscopic and computational evidence suggests that the [Fe4S4] cubanes significantly influence both the geometric and electronic structures of the remaining portions of the active sites. Consequently, future computational studies aimed at elucidating the catalytic mechanisms of ACS and [FeFe] hydrogenases should employ Acluster and Η-cluster models that include the complete [Fe4S4] cubane. Lastly, because the structure and composition of the CODH C-cluster have recently been determined by X-ray crystallography, the necessary foundation has also been laid for future computational studies to evaluate the proposed catalytic cycle of CODH at the molecular level.

ACKNOWLEDGMENTS T.C.B, thanks his current and former graduate students and postdoctoral fellows for their hard work and valuable discussions on this project, his superb collaborators for generous protein supply, Professor Frank Neese for providing us with a free copy of his ORCA computational software package and for his advice on electronic structure calculations, and acknowledges the National Science Foundation (CAREER award MCB-0238530) for financial support. K.M.V.H. was supported by the National Science Foundation Graduate Research Fellowship Program.

ABBREVIATIONS Abs ACS Ado AdoCbl AdoMet Am ATP ATR B3LYP BP86 BPS

electronic absorption acetyl-coenzyme A synthase adenosyl adenosylcobalamin S-adenosylmethionine acetamide adenosine-5'-triphosphate adeno syltransferase Becke's three-parameter hybrid functional for exchange coupled with the Lee-Yang-Parr correlation functional nonlocal gradient corrections of Becke for exchange and Perdew for correlation bromopropanesulfonate

Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

BS CAS CASPT2 Cbi Cbl CD CFeSP CI CODH CP DFT DMB dmgBF 2 ENDOR EPR FCII Fe d Fe p Gin GM Hey HF HSCoB HSCoM IR KS MC MCD MCR Me MeCbl Me-SCoM Met MetH MM MMCM MMCoA MO NADPH Ni d Ni p PBE

453

broken symmetry complete active space complete active space second-order perturbation theory cobinamide cobalamin circular dichroism corrinoid iron-sulfur protein configuration interaction carbon monoxide dehydrogenase coupled-perturbed density functional theory 5,6-dimethylbenzimidazole (difluoroboryl)dimethylglyoximato electron-nuclear double resonance electron paramagnetic resonance ferrous component II of the C-cluster in CODH/ACS distal metal site in the Η-cluster of [FeFe]-hydrogenase proximal metal site in the Η-cluster of [FeFe]-hydrogenase glutamine glutamate mutase homocysteine Hartree-Fock coenzyme Β coenzyme Μ infrared Kohn-Sham multiconfigurational magnetic circular dichroism methyl-coenzyme Μ reductase methyl methylcobalamin methyl-coenzyme Μ methionine methionine synthase molecular mechanics methylmalonyl-CoA mutase methylmalonyl-CoA molecular orbital nicotinamide adenine dinucleotide phosphate distal metal site of the Α-cluster in CODH/ACS proximal metal site of the Α-cluster in CODH/ACS Perdew-Burke-Ernzerhof generalized gradient approximation Met. Ions Life Sei. 2009, 6, 417-460

LIPTAK, VAN HEUVELEN, and BRUNOLD

454 phen py QM rR SCF SCoA SHE SOMO SQM TD ZORA

1,10-phenanthroline pyridine q u a n t u m mechanics resonance R a m a n self-consistent field succinyl-CoA standard hydrogen electrode singly-occupied m o l e c u l a r o r b i t a l scaled q u a n t u m m e c h a n i c a l time-dependent zeroth order regular approximation

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

14. 15. 16. 17.

R. J. Deeth, Struct. Bond. (Berlin), 2004, 113, 37-69. Intel Corporation 2008, Santa Clara, CA. Μ. Born and R. Oppenheimer, Ann. Phys. (Leipzig), 1927, 389, 457^184. L. Randaccio, M. Furlan, S. Geremia, M. Slouf, I. Srnova and D. Toffoli, Inorg. Chem., 2000, 39, 3403-3413. Η. B. Schlegel, J. Comp. Chem., 2003, 24, 1514-1527. C. J. Cramer, Essentials of Computational Chemistry: Theories and Models, John Wiley & Sons, Chichester, 2004. F. Jensen, Introduction to Computational Chemistry, John Wiley & Sons, Chichester, 1999. E. G. Lewars, Computational Chemistry: Introduction to the Theory and Applications of Molecular and Quantum Mechanics, Kluwer Academic, Boston, 2003. W. Koch and M. C. Holthausen, A Chemist's Guide to Density Functional Theory, Wiley-YCH, Weinheim, 2001. I. N. Levine, Quantum Chemistry, Prentice Hall, Upper Saddle River, NJ, 1999. P. Comba and R. Remenyi, Coord. Chem. Rev., 2003, 238-239, 9-20. R. J. Deeth, in Fundamentals: Physical Methods Theoretical Analysis and Case Studies, Ed. A. B. P. Lever, Elsevier Pergamon, Amsterdam, 2004, pp. 457^-66. W. D. Cornell, P. Cieplak, C. I. Bayly, I. R. Gould, J. Κ. M. Merz, D. M. Ferguson, D. C. Spellmeyer, T. Fox, J. W. Caldwell and P. A. Kollman, J. Am. Chem. Soc., 1995, 117, 5179-5197. Η. M. Marques, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, Inc., New York, 1999, pp. 289-313. Η. M. Marques, B. Ngoma, T. J. Egan and K. L. Brown, J. Mol. Struc., 2001, 561, 71-91. B. Kirchner, F. Wennmohs, S. Ye and F. Neese, Curr. Opin. Chem. Biol., 2007, 11, 134-141. C. H. Martin and M. C. Zerner, in Inorganic Electronic Structure and Spectroscopy, Ed. Ε. I. Solomon and A. B. P. Lever, John Wiley & Sons, Inc., New York, 1999, pp. 555-659.

Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL S T U D I E S OF BIOORGANOMETALLICS

455

18. F. Neese, J. Biol. Inorg. Chem., 2006, 11, 702-711. 19. T. Lovell, F. Himo, W.-G. Han and L. Noodleman, Coord. Chem. Rev., 2003, 238-239, 211-232. 20. P. Hohenberg and W. Kohn, Phys. Rev., 1964, 136, B864-B871. 21. W. Kohn and L. J. Sham, Phys. Rev., 1965, 140, A1133-A1138. 22. F. Neese, T. Petrenko, D. Ganyushin and G. Olbrich, Coord. Chem. Rev., 2007, 251, 288-327. 23. A. Ghosh and P. R. Taylor, Curr. Opin. Chem. Biol., 2003, 7, 113-124. 24. B. O. Roos and U. Ryde, in Fundamentals: Physical Methods, Theoretical Analysis, and Case Studies, Ed. A. B. P. Lever, Elsevier Pergamon, Amsterdam, 2004, pp. 457-466. 25. B. O. Roos, R. T. Taylor and P. Ε. M. Siegbahn, Chem. Rev., 1980, 48, 157-173. 26. R. A. Friesner and Y. Guallar, Annu. Rev. Phys. Chem., 2005, 56, 389-427. 27. Η. M. Senn and W. Thiel, Curr. Opin. Chem. Biol., 2007, 11, 182-187. 28. G. Palmer, in Physical Methods in Bioinorganic Chemistry, Ed. L. Que Jr, University Science Books, Sausalito, CA, 2000, pp. 121-185. 29. S. Van Doorslaer, G. Jeschke, B. Epel, D. Goldfarb, R.-A. Eichel, Β. Kräutler and Α. Schweiger, J. Am. Chem. Soc., 2003, 125, 5915-5927. 30. J. Harmer, S. Van Doorslaer, I. Gromov and A. Schweiger, Chem. Phys. Lett., 2002, 358, 8-16. 31. G. N. Schrauzer and L.-P. Lee, J. Am. Chem. Soc., 1968, 90, 6541-6543. 32. J. H. Bayston, F. D. Looney, J. R. Pilbrow and Μ. E. Winfield, Biochemistry, 1970, 9, 2164-2172. 33. A. D. Becke, Phys. Rev. A, 1988, 38, 3098-3100. 34. J. P. Perdew and Y. Wang, Phys. Rev. B: Condens. Matter, 1992, 45, 13244-13249. 35. C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 1988, 37, 785-789. 36. A. D. Becke, J. Chem. Phys., 1993, 98, 5648-5652. 37. T. A. Stich, Ν. R. Buan and T. C. Brunold, J. Am. Chem. Soc., 2004, 126, 9735-9749. 38. F. Neese, J. Chem. Phys., 2001, 115, 11080-11096. 39. F. Neese, Max-Planck-Institut fur Bioanorganische Chemie, Mülheim, Germany, 2004. 40. Τ. G. Spiro and R. S. Czernuszewicz, in Physical Methods in Bioinorganic Chemistry, Ed. L. Que Jr, University Science Books, Sausalito, CA, 2000, pp. 59-119. 41. T. A. Stich, A. J. Brooks, N. R. Buan and T. C. Brunold, J. Am. Chem. Soc., 2003, 125, 5897-5914. 42. M. D. Liptak and T. C. Brunold, J. Am. Chem. Soc., 2006, 128, 9144-9156. 43. E. Mayer, D. J. Gardiner and R. E. Hester, Mol. Phys., 1973, 26, 783-787. 44. E. Mayer, D. J. Gardiner and S. R. Harder, J. Chem. Soc., Faraday Trans. II, 1973, 69, 1350-1358. 45. W. T. Wozniak and T. G. Spiro, J. Am. Chem. Soc., 1973, 95, 3402-3404. 46. S. Salama and T. G. Spiro, J. Raman Spec., 1977, 6, 57-60.

Met. Ions Life Sei. 2009, 6, 4 1 7 ^ 6 0

456

LIPTAK, VAN HEUVELEN, and BRUNOLD

47. S. Nie, P. A. Marzilli, L. G. Marzilli and N. -T. Yu, J. Chem. Soc., Chem. Commun., 1990, 770-771. 48. L. Quaroni, J. Reglinski and W. E. Smith,./. Raman Spec., 1995, 26, 1075-1076. 49. J. Μ. Puckett, Μ. Β. Mitchell, S. Hirota and L. G. Marzilli, Inorg. Chem., 1996, 35, 4656^662. 50. S. Dong, R. Padmakumar, R. Y. Banerjee and T. G. Spiro, J. Am. Chem. Soc., 1996, 118, 9182-9183. 51. S. Dong, R. Padmakumar, R. V. Banerjee and T. G. Spiro, Inorg. Chim. Acta, 1998, 270, 392-398. 52. T. Andruniow, Μ. Z. Zgierski and P. M. Kozlowski, J. Phys. Chem. A, 2002, 106, 1365-1373. 53. P. M. Kozlowski, T. Andruniow, A. A. Jarzecki, Μ. Z. Zgierski and T. G. Spiro, Inorg. Chem., 2006, 45, 5585-5590. 54. T. Andruniow, Μ. Z. Zgierski and P. M. Kozlowski, Chem. Phys. Lett., 2000, 331, 502-508. 55. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865-3868. 56. J. P. Perdew, M. Ernzerhof, A. Zupan and K. Burke, J. Chem. Phys., 1998,108, 1522-1531. 57. J. M. Pratt, Inorganic Chemistry of Vitamin B12, Academic Press Inc., New York, 1972. 58. D. R. McMillin, in Physical Methods in Bioinorganic Chemistry, Ed. L. Que Jr., University Science Books, Sausalito, CA, 2000, pp. 1-58. 59. E. Runge and Ε. K. U. Gross, Phys. Rev. Lett., 1984, 52, 997-1000. 60. K. P. Jensen, J. Phys. Chem. B, 2005, 109, 10505-10512. 61. S. P. Stabler and R. H. Allen, Ann. Rev. Nutrition, 2004, 24, 299-326. 62. R. Banerjee, John Wiley & Sons, Inc., New York, 1999. 63. C. M. Dobson, T. Wai, D. Leclerc, H. Kadir, M. Narang, J. P. Lerner-Ellis, T. J. Hudson, D. S. Rosenblatt and R. A. Gravel, Human Mol. Gen., 2002, 11, 3361-3369. 64. Ν. A. Leal, S. D. Park, P. E. Kima and T. A. Bobik, J. Biol. Chem., 2003, 278, 9227-9234. 65. R. Banerjee and S. Chowdhury, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, Inc, New York, 1999, pp. 707-729. 66. M. Flavin, P. J. Ortiz and S. Ochoa, Nature (London), 1955, 176, 823-826. 67. J. Katz and I. L. Chaikoff, J. Am. Chem. Soc., 1955, 77. 68. R. G. Matthews, in Chemistry and Biochemistry of B12, Ed. R. Banerjee, John Wiley & Sons, Inc, New York, 1999, pp. 681-706. 69. R. T. Taylor and H. Weissbach, J. Biol. Chem., 1967, 242, 1502-1508. 70. R. Banerjee, A. Dybala-Defratyka and P. Paneth, Phil. Trans. Royal Society of London, Series B: Biological Sciences, 2006, 361, 1333-1339. 71. G. N. Schrauzer, E. Deutsch and R. J. Windgassen, J. Am. Chem. Soc., 1968, 90, 2441-2442. 72. G. N. Schrauzer and E. Deutsch, J. Am. Chem. Soc., 1969, 91, 3341-3350. 73. D. Lexa, J. M. Saveant and J. Zickler, J. Am. Chem. Soc., 1980,102, 4851-4852. 74. K. R. Wolthers, J. Basran, A. W. Munro and N. S. Scrutton, Biochemistry, 2003, 42, 3911-3920.

Met. Ions Life Sei. 2009, 6, 417-460

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

457

75. T. A. Stich, Ν. R. Buan, J. C. Escalante-Semerena and T. C. Brunold, J. Am. Chem. Soc., 2005, 127, 8710-8719. 76. T. A. Stich, Μ. Yamanishi, R. Banerjee and T. C. Brunold, J. Am. Chem. Soc., 2005, 127, 7660-7661. 77. M. D. Liptak, A. S. Fleishchhacker, R. G. Matthews and T. C. Brunold, Biochemistry, 2007, 46, 8024-8035. 78. M. D. Liptak, S. Datta, R. G. Matthews and T. C. Brunold, J. Am. Chem. Soc., 2008, 130, in press. 79. P. M. Kozlowski, Curr. Opin. Chem. Biol., 2001, 5, 736-743. 80. K. L. Brown, Dalton Trans., 2006, 1123-1133. 81. A. J. Brooks, M. Ylasie, R. Banerjee and T. C. Brunold, J. Am. Chem. Soc., 2005, 127, 16522-16528. 82. A. J. Brooks, C. C. Fox, Ε. N. G. Marsh, M. Ylasie, R. Banerjee and T. C. Brunold, Biochemistry, 2005, 44, 15167-15181. 83. A. J. Brooks, M. Vlasie, R. V. Banerjee and T. C. Brunold, J. Am. Chem. Soc., 2004, 126, 8167-8180. 84. N. Dölker, A. Morreale and F. Maseras, J. Biol. Inorg. Chem., 2005, 10, 509-517. 85. N. Dölker, F. Maseras and P. Ε. M. Siegbahn, Chem. Phys. Lett., 2004, 386, 174-178. 86. N. Dölker, F. Maseras and A. Lledos, J. Phys. Chem. B, 2003, 107, 306-315. 87. K. P. Jensen and U. Ryde, J. Am. Chem. Soc., 2005, 127, 9117-9128. 88. K. P. Jensen and U. Ryde, J. Phys. Chem. A, 2003, 107, 7539-7545. 89. K. P. Jensen, S. P. A. Sauer, Τ. Liljefors and P.-O. Norrby, Organometallics, 2001, 20, 550-556. 90. M. Jaworska, P. Lodowski, T. Andruniow and P. M. Kozlowski, J. Phys. Chem. B, 2007, 111, 2419-2422. 91. J. Kuta, S. Patchkovskii, Μ. Z. Zgierski and P. M. Kozlowski, J. Comp. Chem., 2006, 27, 1429-1437. 92. R. A. Kwiecien, I. Y. Khavrutskii, D. G. Musaev, K. M o r o k u m a , R. Banerjee and P. Paneth, J. Am. Chem. Soc., 2006, 128, 1287-1292. 93. A. Dybala-Defratyka, P. Paneth, R. Banerjee and D. G. Truhlar, Proc. Nat. Acad. Sei. USA, 2007, 104, 10774-10779. 94. P. M. Kozlowski, T. Kamachi, T. Toraya and K. Yoshizawa, Angew. Chem. Int. Ed., 2007, 46, 980-983. 95. P. K. Sharma, Ζ. T. Chu, Μ. Η. M. Olsson and A. Warshel, Proc. Nat. Acad. Sei. USA, 2007, 104, 9661-9666. 96. S. Chowdhury and R. Banerjee, Biochemistry, 2000, 39, 7998-8006. 97. F. Mancia and P. R. Evans, Structure, 1998, 6, 711-720. 98. R. L. Birke, Q. Huang, T. Spataru and D. K. Gosser Jr., J. Am. Chem. Soc., 2006, 128, 1922-1936. 99. T. Spataru and R. L. Birke, J. Phys. Chem. A, 2006, 110, 8599-8604. 100. T. A. Stich, J. Seravalli, S. Venkateshrao, T. G. Spiro, S. W. Ragsdale and T. C. Brunold, J. Am. Chem. Soc., 2006, 128, 5010-5020. 101. K. P. Jensen and U. Ryde, J. Am. Chem. Soc., 2003, 125, 13970-13971. 102. R. G. Matthews, Acc. Chem. Res., 2001, 34, 681-689.

Met. Ions Life Sei. 2009, 6, 417-460

458

LIPTAK, V A N HEUVELEN, and B R U N O L D

103. Η. P. C. Hogenkamp, G. T. Bratt and S.-Z. Sun, Biochemistry, 1985, 24, 6428-6432. 104. Y. Bandarian and R. G. Matthews, Biochemistry, 2001, 40, 5056-5064. 105. C. W. Goulding, D. Postigo and R. G. Matthews, Biochemistry, 1997, 36, 8082-8091. 106. D. Lexa and J. M. Saveant, Acc. Chem. Res., 1983, 16, 235-243. 107. J. T. Jarrett, C. Y. Choi and R. G. Matthews, Biochemistry, 1997, 36, 15739-15748. 108. C. W. Goulding and R. G. Matthews, Biochemistry, 1997, 36, 15749-15757. 109. M. Goenrich, E. C. Duin, F. Mahlert and R. K. Thauer, J. Biol. Inorg. Chem., 2005, 10, 333-342. 110. U. Ermler, W. Grabarse, S. Shima, M. Goubeaud and R. K. Thauer, Science, 1997, 278, 1457-1462. 111. S. W. Ragsdale, Chem. Rev., 2006, 106, 3317-3337. 112. J. Eilermann, R. Heddrich, R. Bocher and R. K. Thauer, Eur. J. Biochem., 1988, 172, 669-677. 113. J. L. Craft, Y. C. Horng, S. W. Ragsdale and T. C. Brunold, J. Am. Chem. Soc., 2004, 126, 4068^069. 114. Q. Tang, P. E. Carrington, Y. C. Horng, M. J. Maroney, S. W. Ragsdale and D. F. Bocian, J. Am. Chem. Soc., 2002, 124, 13242-13256. 115. A. B. Parusel, T. Wondimagegn and A. Ghosh, J. Am. Chem. Soc., 2000, 122, 6371-6374. 116. J. A. Shelnutt, X.-Z. Song, J.-G. Ma, S.-L. Jia, W. Jentzen and C. J. Medforth, Chem. Soc. Rev., 1998, 27, 31-41. 117. G. Farber, W. Keller, C. Kratky, B. Jaun, A. Pfaltz, C. Spinner, A. Kobelt and A. Eschenmoser, Helv. Chim. Acta, 1991, 74, 697-716. 118. A. Ghosh, T. Wondimagegn and H. Ryeng, Curr. Opin. Chem. Biol., 2001, 5, 744-750. 119. T. Wondimagegn and A. Ghosh, J. Am. Chem. Soc., 2000, 122, 6375-6381. 120. T. Wondimagegn and A. Ghosh, J. Phys. Chem. B, 2000, 104, 10858-10862. 121. L. N. Todd and M. Zimmer, Inorg. Chem., 2002, 41, 6831-6837. 122. J. Telser, Y. C. Horng, D. F. Becker, Β. M. Hoffman and S. W. Ragsdale, J. Am. Chem. Soc., 2000, 122, 182-183. 123. T. Wondimagegn and A. Ghosh, J. Am. Chem. Soc., 2001, 123, 1543-1544. 124. Y. Pelmenschikov, M. R. A. Blomberg, P. Ε. M. Siegbahn and R. H. Crabtree, J. Am. Chem. Soc., 2002, 124, 4039-4049. 125. V. Pelmenschikov and P. Ε. M. Siegbahn, J. Biol. Inorg. Chem., 2003, 8, 653-662. 126. J. L. Craft, Y. C. Horng, S. W. Ragsdale and T. C. Brunold, J. Biol. Inorg. Chem., 2004, 9, 77-89. 127. U. Ermler, Dalton Trans., 2005, 3451-3458. 128. W. Grabarse, F. Mahlert, Ε. C. Duin, M. Goubeaud, S. Shima, R. K. Thauer, V. Lamzin and U. Ermler, J. Mol. Biol., 2001, 309, 315-330. 129. Y. C. Horng, D. F. Becker and S. W. Ragsdale, Biochemistry, 2001, 40, 12875-12885. 130. D. Hinderberger, R. P. Piskorski, M. Goenrich, R. K. Thauer, A. Schweiger, J. Harmer and B. Jaun, Angew. Chem. Int. Ed., 2006, 45, 3602-3607.

Met. Ions Life Sei. 2009, 6, 4 1 7 ^ 6 0

COMPUTATIONAL STUDIES OF BIOORGANOMETALLICS

459

131. M. Dey, J. Telser, R. C. Kunz, Ν. S. Lees, S. W. Ragsdale and Β. M. Hoffman, J. Am. Chem. Soc., 2007, 129, 11030-11031. 132. N. Yang, M. Reiher, Μ. Wang, J. Harmer and Ε. C. Duin,./. Am. Chem. Soc., 2007, 129, 11028-11029. 133. R. C. Kunz, Μ. Dey and S. W. Ragsdale, Biochemistry, 2008, 47, 2661-2667. 134. S. Shima and R. K. Thauer, Curr. Opin. Microbiol., 2005, 8, 643-648. 135. Y. Ahn, J. A. Krzycki and H. G. Floss, J. Am. Chem. Soc., 1991, 113, 4700-4701. 136. E. C. Duin and M. L. McKee, J. Phys. Chem. B, 2008, 112, 2466-2482. 137. T. C. Brunold, J. Biol. Inorg. Chem., 2004, 9, 533-541. 138. J. Xia, J. F. Sinclair, T. O. Baldwin and P. A. Lindahl, Biochemistry, 1996, 35, 1965-1971. 139. S. W. Ragsdale and M. Kumar, Chem. Rev., 1996, 96, 2515-2539. 140. Z. G. Hu, N. J. Spangler, Μ. Ε. Anderson, J. Xia, P. W. Ludden, P. A. Lindahl and E. Münck, J. Am. Chem. Soc., 1996, 118, 830-845. 141. P. A. Lindahl, E. Munck and S. W. Ragsdale, J. Biol. Chem., 1990, 265, 3873-3879. 142. J. Q. Xia, Z. G. Hu, C. Y. Popescu, P. A. Lindahl and E. Münck, J. Am. Chem. Soc., 1997, 119, 8301-8312. 143. Τ. I. Doukov, L. C. Blasiask, J. Seravalli, S. W. Ragsdale and C. L. Drennan, Biochemistry, 2008, 47, 3474-3483. 144. V. J. DeRose, J. Telser, Μ. E. Anderson, P. A. Lindahl and Β. M. Hoffman, J. Am. Chem. Soc., 1998, 120. 145. S. W. Ragsdale, Crit. Rev. Biochem. Mol. Biol., 2004, 39, 165-195. 146. P. A. Lindahl, Biochemistry, 2002, 41, 2097-2105. 147. C. Darnault, A. Volbeda, E. J. Kim, P. Legrand, X. Vernede, P. A. Lindahl and J. C. Fontecilla-Camps, Nat. Struct. Biol., 2003, 10, 271-279. 148. Τ. I. Doukov, T. M. Iverson, J. Seravalli, S. W. Ragsdale and C. L. Drennan, Science, 2002, 298, 567-572. 149. R. P. Schenker and T. C. Brunold, J. Am. Chem. Soc., 2003, 125, 13962-13963. 150. H. Dobbek, V. Svetlitchnyi, L. Gremer, R. Huber and O. Meyer, Science, 2001, 293, 1281-1285. 151. S. J. George, J. Seravalli and S. W. Ragsdale, J. Am. Chem. Soc., 2005, 127, 13500-13501. 152. W. K. Russell, C. Μ. V. Stalhandske, J. Q. Xia, R. A. Scott and P. A. Lindahl, J. Am. Chem. Soc., 1998, 120, 7502-7510. 153. E. L. Hegg, Acc. Chem. Res., 2004, 37, 775-783. 154. S. Gencic and D. A. Grahame, J. Biol. Chem., 2003, 278, 6101-6110. 155. X. Tan, C. Sewell, Q. Yang and P. A. Lindahl, J. Am. Chem. Soc., 2003, 125, 318-319. 156. X. Tan, Ι. V. Surovtsev and P. A. Lindahl, J. Am. Chem. Soc., 2006, 128, 12331-12338. 157. N. A. Eckert, W. G. Dougherty, G. P. A. Yap and C. G. Riordan, J. Am. Chem. Soc., 2007, 129, 9286-9287. 158. C. E. Webster, Μ. Y. Darensbourg, P. A. Lindahl and Μ. B. Hall, J. Am. Chem. Soc., 2004, 126, 3410-3411.

Met. Ions Life Sei. 2009, 6, 4 1 7 ^ 6 0

LIPTAK, VAN HEUVELEN, and BRUNOLD

46C

159. P. Amara, A. Yolbeda, J. C. Fontecilla-Camps and M. J. Field, J. Am. Chem. Soc., 2005, 127, 2776-2784. 160. F. A. Armstrong, Curr. Opin. Chem. Bio!., 2004, 8, 133-140. 161. Y. Nicolet, C. Piras, P. Legrand, C. E. Hatchikian and J. C. Fontecilla-Camps, Structure with Folding & Design, 1999, 7, 13-23. 162. Ρ. Ε. Μ. Siegbahn, J. W. Tye and Μ. B. Hall, Chem. Rev., 2007,107, 4414-4435. 163. J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt, Science, 1998, 282, 1853-1858. 164. M. W. W. Adams, J. Biol. Chem., 1987, 262, 15054-15061. 165. D. S. Patil, J. J. G. Moura, S. H. He, M. Teixeira, B. C. Prickril, D. Y. Dervartanian, H. D. Peck, J. Legall and Β. H. Huynh, J. Biol. Chem., 1988, 263, 18732-18738. 166. D. S. Patil, Β. H. Huynh, S. H. He, H. D. Peck, D. V. Dervartanian and J. Legall, J. Am. Chem. Soc., 1988, 110, 8533-8534. 167. A. S. Pereira, P. Tavares, I. Moura, J. J. G. Moura and Β. H. Huynh, J. Am. Chem. Soc., 2001, 123, 2771-2782. 168. C. V. Popescu and E. Munck, J. Am. Chem. Soc., 1999, 121, 7877-7884. J. Telser, M. J. Benecky, M. W. W. Adams, L. E. Mortenson and Β. M. 169. Hoffman, J. Biol. Chem., 1987, 262, 6589-6594. J. Telser, M. J. Benecky, M. W. W. Adams, L. E. Mortenson and Β. M. 170. Hoffman, J. Biol. Chem., 1986, 261, 3536-3541. T. J. Zhou, Y. R. Mo, A. M. Liu, Ζ. H. Zhou and K. R. Tsai, Inorg. Chem., 171. 2004, 43, 923-930. Z. P. Liu and P. Hu, J. Am. Chem. Soc., 2002, 124, 5175-5182. 172. Ζ. X. Cao and Μ. B. Hall, J. Am. Chem. Soc., 2001, 123, 3734-3742. 173. H. J. Fan and Μ. B. Hall, J. Am. Chem. Soc., 2001, 123, 3828-3829. 174. A. T. Fiedler and T. C. Brunold, Inorg. Chem., 2005, 44, 9322-9334. 175. L. Noodleman, T. Lovell, T. Q. Liu, F. Himo and R. A. Torres, Curr. Opin. 176. Chem. Biol., 2002, 6, 259-273. 177. L. Noodleman, C. Y. Peng, D. A. Case and J. M. Mouesca, Coord. Chem. Rev., 1995, 144, 199-244. 178. L. Noodleman and D. A. Case, Adv. Inorg. Chem., 1992, 38, 423^70. 179. G. C. Papaefthymiou, E. J. Laskowski, S. Frotapessoa, R. B. Frankel and R. H. Holm, Inorg. Chem., 1982, 21, 1723-1728.

Ions Life Sei. 2009, 6, 417^60

Met. Ions Life Sei. 2009, 6, 461-496

Subject Index

A Absorption spectroscopy cobalamins, 426, 429, 431 F 4 3 0 model, 438, 439 methyl-coenzyme Μ reductase, 437 UV, see UV absorption spectroscopy Acetaldehyde, 100 phosphono-, 83 Acetate (or acetic acid), 59, 72, 76, 162 Acetamide, 437, 438, 440 Acetobacteriwn dehalogenans, 77, 80 Acetogenesis, 35 Acetonitrile, 117, 119 Acetyl-coenzyme A, 31, 59, 73, 77, 136, 139, 451 biosynthesis, see Biosynthesis decarbonylase/synthase, 136, 140 Acetyl-coenzyme A synthase(s), 31, 434, 443 active site, see Active sites catalytic cycle, 445-447, 452 density functional theory calculations, see Density functional theory calculations EPR studies, see EPR mechanism, 146, 445, 446 models, 446 nickel-carbon bonds, see Nickel-carbon bonds

Metal Ions in Life Sciences, Volume 6 © Royal Society of Chemistry 2009

Acetyl-coenzyme A synthase/carbon monoxide dehydrogenase, 133-146, 419, 420 A-cluster, 135-141, 144, 146, 442-444 active site, see Active sites azide in, see Azide B-cluster, 135, 136, 146, 442 catalytic properties, 137, 138 C-cluster, 134-138, 141-146, 442 cyanide in, 135, 142-144 D-cluster, 135, 136, 442 diamagnetic mechanism, 137, 138, 140 electron density, see Electron density electron transfer in, see Electron transfer E N D O R studies, see E N D O R spectroscopy EPR studies, see EPR from Moorella thermoacetica, 133-146, 442 hydrogen bonds, see Hydrogen bonds hydrolysis, 141 infrared studies, see Infrared spectroscopy inhibitors, 142, 143 methyl transfer, see Methyl transfer Mössbauer spectroscopy studies, see Mössbauer spectroscopy nickel-carbon bonds, see Nickel-carbon bonds paramagnetic mechanism, 140, 141 ping-pong mechanism, 140, 141 redox properties, 137, 138 structure, 135, 443, 444

Edited by Astrid Sigel, Helmut Sigel, and Roland K. O. Sigel

Published by the Royal Society of Chemistry, www.rsc.org

DOI: 10.1039/9781847559159-00461

462 Acidity constants (see also Equilibrium constants), 385, 399 B 1 2 derivatives, 20 H 2 , 236 A C S / C O D H , see Acetyl-coenzyme A synthase/carbon m o n o x i d e dehydrogenase Active site (of) acetyl-coenzyme A synthase, 445, 446 A C S / C O D H , 135, 136, 442-447 blue copper proteins, 338 bromoperoxidase, 369 carbonic anhydrase, 385 catechol dioxygenase, 380 catechol oxidase, 303 cobalamins, 426^135 copper amine oxidase, 334 copper-zinc superoxide dismutase, 331 cytochrome c oxidase, 318, 324 deoxyhemocyanin, 300 d o p a m i n e ß-monooxygenase, 308, 311, 315-317 F 4 3 0 , 125-127 [FeFe]-hydrogenases, 153, 161, 179-208, 222, 231, 447-449 galactose oxidase, 337 glutamate mutase, 88, 432 heme, 261, 266 heme-copper oxidases, 317-319, 326 hemocyanin, 303, 308 hydrogenases, 152 manganese catalase, 375 methyl-coenzyme Μ reductase, 115-129, 435, 437 methylmalonyl-coenzyme A mutase, 431, 432 MoQ C O D H , 333 multi-copper oxidase, 339 [NiFe]-hydrogenases, 153, 155-165, 168-172, 200, 222, 231 nitrite reductases, 341, 342 nitrous oxide reductase, 343 non-heme iron enzymes, 379 peptidylglycine a-hydroxylating monooxygenase, 311, 315-317 protocatechuate 3,4-dioxygenases, 382 superoxide reductase, 381 tyrosinase, 303 xanthine oxidoreductases, 398, 400, 403, 406, 409-413 zinc enzymes, 384, 386

Met. Ions Life Sei. 2009, 6, 461-496

SUBJECT INDEX Acyl-coenzyme A synthase, 57-60, 73, 76 decarbonylase, 57, 73, 74 Adenosine 5'-deoxy-, 84, 85, 91 5'-diphosphate, see 5 ' - A D P 5'-triphosphate, see 5 ' - A T P Adenosylcobalamin(s) (see also Coenzyme B i 2 and Vitamin B 1 2 ), 18, 38, 54, 83, 84, 86, 87, 90, 91, 104-106, 427, 428, 430-432 3',4'-anhydro-, 97 as cofactor, 34, 40, 96, 102 base-off, 93 biosynthesis, see Biosynthesis D B M - o f f / H i s - o n , 103 D B M - o n , 96, 101 5'-deoxy-5'-, 3 -dependent rearrangements, 84-106 His-on, 102 reactivity, 24 structure, 4, 11, 12, 14, 55, 88, 98 S'-Adenosylmethionine, 34, 61, 62, 65, 67, 68, 75, 79, 80, 81, 83, 84, 93, 428 radical, see Radicals Adenosyltransferase, 34, 92, 93 chaperone, see C h a p e r o n e s 5 ' - A D P , 59 Affinity constants, see Stability constants Agaricus, 304 bisporus, 307 Alcaligenes sp., 341 Alcohols, see individual names Alcohol dehydrogenase zinc in, 383 Aldehyde(s) (see also individual names), 334, 3335, 337 acet-, see Acetaldehyde glycol, 84 hydration, 385 oxidase, 368 propion-, 84 Aldehyde oxidoreductase, 398, 403, 411 crystal structure, see Crystal structures Aldolase fructose-1,6-diphosphate, 383 Algae (see also individual names), 181, 184, 373 brown, 369 green, 183 marine, 368, 372

SUBJECT INDEX [Algae (see also individual names)] oxygen evolving complex, see Oxygen evolving complex red, 246, 369 Alkaline phosphatases, 386, 388 magnesium, 388 zinc, 288 Alkylation, 26 Co(I), 25 Co(III)-corrins, 31 cob(I)amides, 34 cobalamins, 25, 28, 30, 77 Allochromatium vinosum, 158 Allopurinol as inhibitor, 404 Alloxanthine xanthine oxidoreductase complex, 404^108 Amide bond cleavage, 386 nickel coordination, 136, 139 proton exchange, 269 Amine(s) (see also individual names) aromatic, 339 oxidation, 334, 335 tris(pyrid-2-ylmethyl)-, see Tris(pyrid-2ylmethyl)-amine Amino acids (see also individual names) radicals, see Radicals Amino acid sequences [FeFe]-hydrogenases, 186, 203 [Fe]-hydrogenases, 223 heme oxygenase, 278, 279 olfactory receptors, 346 Aminomutases, 38, 95-103 lysine, see Lysine 2,3-aminomutase ornithine, see Ornithine aminomutase Aminopeptidase (see also Peptidases and individual names), 386 leucine, see Leucine aminopeptidase Ammonia, 100, 101, 370-373 aspartate lyase, 202 Amyotrophic lateral sclerosis, 330 Anemia hemolytic, 245 megaloplastic, 426 Antibacterial agents (see also individual names), 340 Antibiotics (see also individual names), 83, 346-349 biosynthesis, see Biosynthesis

463 Anticancer agents (see also individual names), 347, 387 Antiglaucoma agents, 387 Antiinflammatory drugs, see Drugs effect of carbon monoxide, 245 Antiosteoporosis agents, 387 Antioxidants (see also individual names), 246, 247 Apoptosis inhibition, 245 Aquacobalamin, 54, 66, 427 Co(III), 3, 19-21, 23, 29, 79, 81, 93, 100 electrochemistry, 20, 26 structure, 4, 9, 10, 13 Arabidopsis sp., 345 thaliana, 246 Archaea (see also individual names), 54, 56, 60, 84, 165, 181, 221, 442 anaerobic, 58, 134, 435 methanogenic, 73, 83, 116, 221, 234, 386 methanotrophic, 116 Arene Cu(I) complex, 298, 306, 307 Arrhenius plot, 23 Arthrobacter globiformis, 334, 336 Arthropods (see also individual names) hemocyanin, 299, 301, 302, 304-306 Ascophyllwn nodosum, 369, 372 Ascorbate, 308, 339 as reductant, 312 Ascorbate oxidase, 339 rate constants, see Rate constants structure, 340 zucchini squash, 339 Aspartate transcarbamoylase, 383 Aspartic acid methyl-, 86, 87 Association constants (see also Stability constants) carbon monoxide binding, 257, 258 dioxygen binding, 257, 258 heme oxygenases, 257, 258 Atmosphere carbon monoxide in, 243, 332 5'-ATP, 59, 77, 79-81, 92, 93, 162, 371, 373, 429 biosynthesis, see Biosynthesis corrinoid adenosyltransferase, see Transferases hydrolysis, 75, 82, 373

Met. Ions Life Sei. 2009, 6, 461-496

464

SUBJECT INDEX

[5'-ATP] production, 327 -using adenosyltransferase, 34 ATPase, 100 Azide (in), 255, 366, 373, 375-377, 387 ACS/CODH, 135, 142 chloroperoxidase, 369 heme oxygenase, see Heme oxygenases as inhibitor, 269, 270, 274, 376, 377 manganese catalase, 377 manganese superoxide dismutase, 376 superoxide reductase, 381, 382 Azotobacter chroococcum, 370, 371 nitrogenase, 370 vinelandii, 370 Azurin, 338

Β Bacteria(l) (see also individual names), 56, 161, 165, 181, 231 acetogenic, 59, 73, 74 anaerobic, 34, 78, 134, 181 cyano-, see Cyanobacteria eu-, see Eubacteria heme oxygenase, see Heme oxygenases hyperthermophilic, 184 Knallgas, 171 oxygen evolving complex, 381 pathogenic, 246, 248, 251 photosynthetic, 282, 283 purple non-sulfur, 202 purple, 381 soil, 380 sulfate-reducing, 152, 165, 166, 171, 184 sulfidogenic, 78 thermophilic, 134, 184 Benzimidazole 5,6-dimethyl-, see 5,6-Dimethylbenzimidazole 5-hydroxy-, 70, 74, 80, 81 5-methoxy-, 54 Benzoate chlorinated, 78 Bilirubin, 246 Biliverdin, 245-247, 251, 253, 256, 268 Fe(II), 247 Fe(III), 247 formation, 276-282, 284

Met. Ions Life Sei. 2009, 6, 461-496

Binding constants, see Association constants, Equilibrium constants, and Stability constants Bioorganometallic species computational studies, 419^152 Biosynthesis 13 C-labeled heme, 260 acetyl-coenzyme A, 35, 443 adenosylcobalamin, 427-430 antibiotics, 378 ATP, 317 biotin, 204 carbon monoxide, 241-285 coenzyme B12, 3, 34, 419 [Fe]-hydrogenases, 227 FeMo cofactor, 202, 204 fosfomycin, 82 Η-cluster of [FeFe]-hydrogenases, 199-207 iron guanylylpyridinol cofactor, 225 lipoic acid, 204, 205 melanin, 299 methionine, 35 Biotechnology, 366 Biotin biosynthesis, see Biosynthesis synthase, 203 Bleomycin, 346-349 Co (III), 347 Cu(I), 346-349 Cu(I)-carbon monoxide, 348 Cu(II), 347-349 cytotoxicity, 349 Fe(II), 346-348 hydroperoxo-Fe(III), 347 mechanism of action, 347 metal-free, 347-349 N M R studies, see N M R structure, 347, 348 Blue copper proteins (see also individual names), 337, 338 active site, see Active site electron density, see Electron density electron transfer in, see Electron transfer EPR studies, see EPR redox potentials, see Redox potentials Bonds amide, 386 C-C, 33 C-O, 403 C-S, 441 CH 3 -S, 125, 127

465

SUBJECT INDEX [Bonds] Co-C, see Cobalt-carbon bonds Cu-C, see Copper-carbon bonds Cys-His, 304 dissociation energies, 31, 32, 38 hydrogen, see Hydrogen bonds Mo-C, 402, 404 M o = 0 , 397-399, 411 Mo=S, 397, 400, 401, 408 Ni-C, see Nickel-carbon bonds Ni(II)-S, 124 O-O, 252, 255, 256, 267, 268, 379 thioether, 304 Born-Oppenheimer approximation, 420 Bos taurus (see also Bovine) xanthine oxidoreductase, 398 Bovine copper-zinc superoxide dismutase, 330, 331 cytochrome c oxidase, 318, 320, 326, 329 heart heme-copper oxidase, 318, 321, 322,327 xanthine oxidoreductase, see Xanthine oxidoreductase Brain, 245 Bromide methyl-, see Methyl bromide oxidation, 372 Bromoperoxidases active site, see Active site inhibition, 372 vanadium, 369, 372 3-Bromopropane sulfonate, 118, 120-122, 439 [3-13C]-, 121 as inhibitor, 120 Buffer Tris, see Tris(hydroxymethyl)methylamine Busycon canaliculatum, 302, 308 carica, 301 Butyrate 4-bromo-, 122 4-hydroxy-, 122

c Cadmium(II) in HybD protease, 164 Calorimetry photoacoustic, 320, 321

Camberus loevis, 305 robustus, 305 Cancer (see also Carcinomas and individual names) ovarian, 347 risk assessment, 28 testicular, 347 Cancer irroratus, 305 Carbamates, 385 thio-, see Thiocarbamates Carbamoyl phosphate, 201 Carbamoyl phosphate synthase, 161, 162 Carbohydrates (see also individual names), 373 fermentation, 182 metabolism, see Metabolism Carbon u C-labeling, 350 12 C-labeling, 233, 331 13 C ENDOR, see ENDOR 13 C NMR, see N M R 13 C-labeling, 96, 101, 123, 139, 146, 158, 162, 225, 259, 260, 300, 331, 397, 400, 444, 445 14 C-labeling, 139 -carbon bond, 23, 86 -molybdenum bond, see Bonds nickel bond, see Nickel-carbon bonds -oxygen bond, see Bonds skeleton rearrangements, 85-95 -sulfur bond, see Bonds Carbon dioxide (in), 58, 60, 72, 73, 76, 77, 221, 298, 332, 385, 387 13 CO z , 162 ACS/CODH, 134, 143-145 fixation, 35 hydration, 384 reduction, 56, 57, 373, 442, 444, 445 Carbonic anhydrase (from) active site, see Active sites catalysis of hydrolytic processes, 385 classes, 386 crystal structure, see Crystal structures human, 387 hydrogen bonds, see Hydrogen bonds inhibitors, 387 iodide complex, 387 mammalian, 385 mechanism of action, 385 Methanosarcina thermophila, 385

Met. Ions Life Sei. 2009, 6, 461-496

466 [Carbonic anhydrase (from)] plant, 385 zinc in, 383-387 Carbon monoxide (in), 57, 73, 75, 77, 137, 142-144, 146, 297, 366, 367, 373, 442, 445, 447 12 CO, 233 " C O , 139, 225, 444, 445 as inhibitor, see Inhibition as probe in copper amine oxidase, 336 atmospheric levels, 243, 332 biosynthesis, see Biosynthesis Cu(I) binding, see Copper(I) cytoprotective effects, 243-245 dehydrogenase, see Carbon monoxide dehydrogenases dopamine ß-monooxygenase, 311-316 environmental sources, 243 Fe(II)-CO, see Iron(II) [FeFe]-hydrogenases, 179-208, 448, 450 [Fe]-hydrogenases, 219-237 heme oxygenases, see Heme oxygenases heme-copper oxidases, 319-325 hemes, see Hemes metabolism, see Metabolism [NiFe]-hydrogenases, 158, 160-164, 172 nitrite reductase, 341, 342 oxidation, 59, 134, 202, 298, 444, 445 peptidylglycine a-hydroxylating monooxygenase, 311-316 photodissociation, 302 properties, 243 release from heme, see Heme role in biology and medicine, 244 stretching frequencies, 305, 315, 349 toxicity, see Toxicity Carbon monoxide dehydrogenases, 57, 59, 73, 75, 80, 202, 398, 444, 445 catalytic mechanism, 146, 445 electron density, see Electron density ferredoxin, 445 from Carboxythermus hydrogenoformans, 142 from Rhodospirillum rubrum, 136, 141 inhibitors, 142, 143 MoCu, see Molybdenum-copper carbon monoxide dehydrogenase nickel insertion, 164 X-ray diffraction spectroscopy studies, see X-ray diffraction spectroscopy

Met. Ions Life Sei. 2009, 6, 461-496

SUBJECT INDEX Carbon monoxyhemocyanin, 304-308 mutation, 307 Carboxydothermus hydrogenoformans, 59, 73-75, 136, 142 Carboxylase pyruvate, see Pyruvate Carboxylate groups, 375 bridging, 376 Carboxypeptidases (see also Peptidases), 386 A, 383 Carcinoma (see also Cancer) head, 347 neck, 347 skin, 347 Catalases, 374 manganese, see Manganese catalase Catecholase activity, 299, 304 Catechol dioxygenases, 377, 378, 380 extradiol-cleaving, 380 Fe(II)-active site, 380 intradiol-cleaving, 380 Μη(ΙΓ)-active site, 380 Catechol oxidase (from), 298-300, 306 active site, see Active sites crystal structure, see X-ray crystal structure Ipomoea, 303 reaction cycles, 303, 304 Cell damage, 244 death, 327 proliferation, 245 smooth muscle, 245 Cephalosporin, 378 biosynthesis, see Biosynthesis Ceruloplasmin, 339, 340 CD, see Circular dichroism Chaperones (for) adenosyltransferase, 93 bacterial copper CusF, 344, 345, 350 diol dehydrase, 100 flavodoxin, 81 methyltransferase-activating protein, 82 Charge transfer, 439 metal-to-ligand, 306, 307 Chemotherapy, 347, 404 Chlamydomonas reinhardtii, 183 Chloride Ni(II) complex, 119 oxidation, 372

SUBJECT INDEX Chlorinated compounds dechlorination, 77, 78 detoxification, see Detoxification Chloroform, 119, 122 Chloroperoxidase azide in, see Azide vanadium, 368, 369 Chromatium vinosum, see Allochromatium vinosum Chromatography high performance liquid, see High performance liquid chromatography hydrophobic interactions, 228 Chromophores, 284 Ciliates (see also individual names), 184 Circular dichroism (studies of) cobalamins, 426, 431 far-UV, 346 Citrate, 235 homo-, see Homocitrate titanium, 72 Clostridium barkeri, 90 pasteurianum, 152, 181-192 stricklandii, 102, 103 thermoaceticum, see Moorella thermoacetica Clusters (in) [7Fe-9S-homocitrate], 370 [8Fe-7S], 370 [Fe2S2], see [Fe2S2] cluster [Fe3S4], see [Fe3S4] cluster [Fe4S4], see [Fe4S4] cluster [Mn(III)Mn(III)], 375, 376 2Fe sub-, see [FeFe]-hydrogenases ACS/CODH, see Acetyl-coenzyme A synthase/carbon monoxide dehydrogenases Cu 4 S, 343 cubane (see also [Fe4S4] cluster), 136, 139-141, 442-444, 448, 449 H-, see [FeFe]-hydrogenases iron-sulfur, 34, 35 Mn 4 Ca, 374 nitrogenases, 370 oxygen-bridged binuclear non-heme iron, 378 P-, 370

467 Cobalamins (see also Coenzyme B 12 , Corrinoids, Vitamin B 12 , and individual names) absorption spectroscopy, see Absorption spectroscopy active site, see Active sites adenosyl-, see Adenosylcobalamin adenylylpentyl-, 100, 105 alkylation, see Alkylation aqua-, see Aquacobalamin base-off, 5, 12, 14, 15, 20, 21, 26, 74-76, 78, 79, 81 base-off/His off, 62, 63 base-off/His on, 39, 62, 63, 67, 74, 77, 79, 90, 93 base-on, 5, 7, 12, 14, 19, 20, 22, 26, 29, 39, 74, 76, 106 chloro-, 4 circular dichroism studies, see Circular dichroism Co(I), 8, 19, 20, 22, 28, 29, 32, 37, 61, 65, 66, 77, 78, 87, 92, 93, 424^130, 432, 434 Co(II), 5, 9, 10, 15, 19, 20, 29, 31-34, 38-40, 61, 62, 65-67, 76-79, 81, 82, 84, 85, 87, 88, 90-93, 96, 97, 106, 424^132, 451 Co(III), 9, 10, 13, 19, 22, 26, 29, 34, 54, 66, 79, 424-426, 428 cyano-, see Cyanocobalamin density functional theory calculations, see Density functional theory calculations -dependent enzymes, 53-107 electron density, see Electron density electron transfer in, see Electron transfer epi-, 7, 8, 13, 26 EPR studies, see EPR EPR studies, see EPR His-off, 65, 66 His-on, 65, 66, 86 hydrogen bonds, see Hydrogen bonds hydroxo-, see Hydroxocobalamin magnetic circular dichroism studies, see Magnetic circular dichroism methyl transfer, see Methyl transfer methyl-, see Methylcobalamin nitroxyl-, 4 propyl-, 66, 67 Raman spectroscopy studies, see Raman spectroscopy

Met. Ions Life Sei. 2009, 6, 461-496

468 [Cobalamins (see also Coenzyme B 12 , Corrinoids, Vitamin B 12 , and individual names)] spectroscopically validated calculations, 424 structure, 4, 9, 55, 428 trichlorovinyl-, 77, 78 vinyl-, 12, 35 Cobalt (oxidation state undefined) 59 Co, 424, 425 -carbon bonds, see Cobalt-carbon bonds Cobalt© (in) alkylation, see Alkylation B 12 derivatives, 18-20, 26 cobalamins, see Cobalamins corrinoids, see Corrinoids corrins, see Corrins redox couples, see Redox potentials Cobalt(II) (in), 72 B 12 derivatives, 18-20 cobalamin, see Cobalamins cobinamides, see Cobinamides cobyrinate, see Cobyrinate corrinoids, see Corrinoids corrins, see Corrins methylcobalamin, see Methylcobalamin redox couples, see Redox potentials reduction, 451 Cobalt(III) (in), 1-41, 138 bleomycin, see Bleomycin cobalamins, see Cobalamins cobamides, see Cobamides cobinamides, see Cobinamides cobyrinate, see Cobyrinate corrinoid iron-sulfur protein, see Corrinoid iron-sulfur protein corrinoids, see Corrinoids corrins, see Corrins methylcobalamin, see Methylcobalamin redox couples, see Redox potentials Cobalt-carbon bonds, 145 activation, 84 cleavage, 24, 30-34, 84, 87, 91, 106, 420, 430^135 formation, 24-31, 426-435, 451 heterolytic cleavage, 31, 36, 432-435, 451 heterolytic formation, 30, 36 homolytic cleavage, 31, 32, 38-40, 96, 97, 100, 430^132, 451 homolytic formation, 30 reactivity, 3

Met. Ions Life Sei. 2009, 6, 461-496

SUBJECT INDEX Cobamides, 4 adenosyl-, 4, 33 alkylation, see Alkylation base-off, 82 base-off/His-on, 70 base-on, 54, 82 Co(II), 74-76, 82 Co (III), 34, 38, 74 5,6-dimethylbenzimidazole-, see Cobalamins 5-hydroxybenzimidazolyl-, 70, 74, 80, 81 imidazolyl-, 7, 12, 13, 26, 33, 36, 37 5'-methoxybenzimidazole, 54 methyl transfer, see Methyl transfer methyl-, 74, 82 /?-cresolyl-, 18, 55, 56 structure, 8, 9, 15 Cobesters (see also Corrinoids), 5 Co(II), 5, 15, 30 structure, 6 Cobinamides (see also Corrinoids and Vitamin B 12 ) adenosyl-, 14, 34 aqua-, 82 base-off, 75 Co(I), 21, 23, 32, 76 Co(II), 21, 32, 33, 75, 82, 424, 428, 429 Co(III), 21, 428 density functional theory calculations, see Density functional theory calculations diaqua, 21 EPR studies, see EPR methyl-, see Methylcobinamides structure, 6, 428 Cobyrinate, 5, 13 Co(II), 26 Co(III), 13, 14 structure, 6 CODH, see Carbon monoxide dehydrogenases Coenzyme A, 57, 73, 91 acetyl-, 57 dephospho-, 140 isobutyryl-, 94 methyl malonyl-, 93, 94, 431 radiolabeled, 140 succinyl-, 91, 93

469

SUBJECT INDEX Coenzyme B, 116-119, 123, 125-128, 236, 435, 436, 440-442, 451 density functional theory calculations, see Density functional theory calculations structure, 117, 436 Coenzyme B 12 (and derivatives) (see also Adenosylcobalamin and Vitamin B 12 ), 1-41 base-off, 31 base-off,His-on, 87 base-on, 31 biosynthesis, see Biosynthesis 2'-deoxy-, 4, 33 -dependent enzymes, 38, 83 derivatives as cofactor, 34-40 derivatives as intermediates in enzymes, 34-40 homo-, 4, 11, 14, 15 "inorganic" derivatives, 9, 13 molecular switch, see Molecular switch neo-, 8, 14, 15, 26 organometallic chemistry, 1—41 pseudo-, 8, 14, 15, 18, 26 reactivity, 24-34 redox chemistry, 18-24, 26 -rotaxane, 26, 27 spectroscopic studies, 13-16 structure, 4-18 Coenzyme (cofactor) F 4 3 0 , see F 4 3 0 Coenzyme M, 57, 59, 67, 71-73, 118, 122, 123, 125, 126, 129, 236, 436 allyl-, 128 ethyl-, 125, 127, 128 list of methyl donors, 69 methyl transfer, see Methyl transfer methyl-, see Methyl-coenzyme Μ methyltransferase, see Methyltransferases reductase, see Reductases Cofactors computational studies, 417^152 F 430 , see F 4 3 0 FeV, 370, 371 heme, see Heme(s) iron guanylylpyridinol, see Iron guanylylpyridinol cofactor iron-molybdenum, see FeMo cofactor trihydroxyphenylalanine quinone, see 2,4,5-Trihydroxyphenylalanine quinone

Combustion of fuels, 243 Complete active space second-order perturbation theory, 426 Compound I, 252, 285 Computational studies (of) ACS/CODH, 139, 144, 146 cofactors, 417^152 enzymes, 417-452 methods, 421^124 spectroscopic observables, 424-426 Conformational changes, 36, 59, 82, 91, 126, 156, 229, 247, 272, 275, 319, 320, 324, 326 dynamics, 15 Copper (different oxidation states) (in) 63 65 ' Cu, 334 alkyl complex, 349 at catalytic sites, 295-350 carbon bonds, see Copper-carbon bonds imidazole coordination, 298-349 M O Q

C O D H , see M o l y b d e n u m - c o p p e r

carbon monoxide dehydrogenase Copper(I) (in), 295-350 alkyl complexes, 349 -arene π interactions, 306, 307, 344, 345 benzo[A]quinoline complex, 306 bleomycin, see Bleomycin carbon monoxide binding, 300-302, 311-315, 319-325, 328, 337, 341-343 copper amine oxidase, see Copper amine oxidase CusF chaperone, 344, 345 cyanide binding, 327-329, 331, 332, 343, 350 dioxygen binding, 300-302, 308, 310 ethylene receptor, see Copper-ethylene receptor inactivation of [Fe]-hydrogenase, 234, 235 nitrite reductase, see Nitrite reductase olefin interaction, 298, 345 receptors, 344-346 recognition sites, 344-346 stability constants, see Stability constants (synthetic) carbonyl complexes, 300-302, 305-308, 312-317, 323 -tryptophan interaction, 344, 345 Copper(II) (in), 295-350 alkyl complexes, 349 bleomycin, see Bleomycin copper amine oxidase, see Copper amine oxidase

Met. Ions Life Sei. 2009, 6, 461-496

470 [Copper(II) (in)] -nitric oxide, 330 nitrite reductase, see Nitrite reductase reduction, 331, 332 -superoxide complex, 310, 311 Copper(III), 349 Copper amine oxidase (from), 334-337 active site, see Active sites Arthrobacter globiformis, 334, 336 catalytic cycle, 335 Cu(I), 335-337 Cu(II), 335 cyanide inhibition, 335, 336 ping-pong mechanism, 335 structure, 334, 335 X-ray absorption spectroscopy studies, see X-ray absorption spectroscopy Copper-carbon bonds, 295-350 Copper chaperones CusF, 344, 345 Copper proteins bacterial copper chaperone CusF, 344, 345, 350 blue, 337-344 coupled binuclear, 298-308 green, 337-344 non-coupled binuclear, 308-317 purple, 337-344 Type III, 299 Copper-zinc superoxide dismutase, 330-332 active site, see Active sites bovine, 330, 331 cyanide binding, 331, 332, 376 cyanide probe, 368 electron transfer in, see Electron transfer hydrogen bonds, see Hydrogen bonds infrared studies, see Infrared spectroscopy mechanism, 331 Raman spectroscopy studies, see Raman spectroscopy structure, 331 Corallina officinalis, 369 Corphin coenzyme F 430 , see F 4 3 0 origin of name, 117 Corrin(s) (see also individual names) alkylation, see Alkylation base-off, 22, 29 Co(I)-, 24-26, 35, 36 Co(II)-, 20-22, 24-26, 35

Met. Ions Life Sei. 2009, 6, 461-496

SUBJECT INDEX [Corrin(s) (see also individual names)] Co(III)-, 23, 26, 29, 32, 36 "complete", 25, 29 cyano-Co(III)-, 18 -dependent enzymes, 53-107 disproportionation, 21 "incomplete", 26, 29 methyl transfer, see Methyl transfer Corrinoid iron-sulfur protein, 56, 57, 59, 60, 73-76, 79, 80, 434, 443, 445, 446 Co (III), 136, 138-140 conformational change, 76 EPR studies, see EPR magnetic circular dichroism studies, see Magnetic circular dichroism Mössbauer spectroscopy studies, see Mössbauer spectroscopy structure, 75 Corrinoids (see also individual names) 1-41 adenosyl-, 34, 38-40 base-off, 34, 36, 55, 78 base-off/His-on, 34, 36, 38, 56 base-on, 34, 56 Co (I), 29, 58 Co(II), 5, 29, 33 Co (III), 5, 422 "complete", 4, 7-20, 29, 34 electron density, see Electron density epi-, 13 EPR studies, see EPR "incomplete", 5, 6, 13, 15, 21 methyl transfer, see Methyl transfer methyl-, 34, 74 Raman spectroscopy studies, see Raman spectroscopy structure, 4. 6 Corynebacterium diphtheriae, 248, 250, 252, 256-258, 276, 280, 283, 284 Crab horseshoe, 299, 306 Crosslinks cysteine-histidine, 303-305 cysteine-tyrosine, 337 histidine-tyrosine, 323, 324 Crystal structures (of) (see also X-ray crystal structures) adenosylcobalamin, 14 aldehyde oxidoreductase, 403 aquacobalamin, 9,10 carbonic anhydrase, 385, 387 coenzyme B12, 10

SUBJECT INDEX [Crystal structures (of) (see also X-ray crystal structures)] [Fe]-hydrogenase, 224, 228-230, 232 HybD protease, 164, 165 Hyp, 163, 164 manganese superoxide dismutase, 374, 376 methylcobalamin, 11, 14 methyl-coenzyme Μ reductase, 129 methylmalonyl-coenzyme A mutase, 90, 92, 431 [NiFe]-hydrogenase, 153-158, 162, 168. 172 [NiFeSe]-hydrogenase, 166 protocatechuate 3,4-dioxygenase, 380 ribonucleotide triphosphate reductase, 105 xanthine oxidoreductase, 398, 404-413 Curvularia inaequalis, 369 Cyanamide hydration, 385 Cyanate (in), 387 ACS/CODH, 135, 142 as inhibitor, 387 hydration, 385 thio-, see Thiocyanate Cyanidase, 366 Cyanide(s) (in), 297, 298, 350, 375 U C N , 350 13 CN, 372 ACS/CODH, 135, 142-144 as inhibitor, 256, 269, 270, 274, 282-284, 337, 366, 372, 373, 381, 387, 388 as prebiotic substrate, 365, 366 as probe for transition metal sites, 366-368 biodegradation, 366 bridging, 328-330 copper amine oxidase inhibition, 335, 336 copper binding, see Copper(I) copper-zinc superoxide diamutase, see Copper-zinc superoxide dismutase cytochrome c oxidase, see Cytochrome c oxidase Fe(II), see Iron(II) Fe(III), see Iron(III) [FeFe]-hydrogenases, 179-208, 448 [Fe]-hydrogenases, 220, 221, 236 gold complex, 387 heme oxygenase, see Heme oxygenases heme-copper oxidases, see Heme-copper oxidases hydratase, see Hydratases iso-, see Isocyanide

471 [Cyanide(s) (in)] isotopically labeled, 331 managnese catalase, 377 manganese superoxide dismutase, 376, 377 monoxygenase, see Monoxygenases multi-copper oxidases, 339, 340 [NiFe]-hydrogenases, 158, 160-164, 171 non-heme iron enzymes, 381-383 oxidation, 372 oxygen evolving complex, 382, 383 protocatechuate 3,4-dioxygenase, 380, 382 reactions, 366 reduction, 372 -resistant respiratory systems, 365 superoxide reductase, 381 toxicity, see Toxicity vanadium enzymes, 372, 373 xanthine oxidoreductase, 397 zinc enzymes, 387, 388 zinc hydrolase, 388 Cyanidium caldarium, 246 Cyanobacteria (see also individual names), 181, 246, 283, 284, 373 Cyanocobalamin (see also Vitamin B 12 ), 87 Co(II), 23 Co (III), 3, 23 Cyanohydrin, 365 Cyclase guanylate, see Guanylate cyclase Cyclic voltammetry studies of B 12 derivatives, 22 Cyclodextrins, 27 Cysteine -histidine crosslink, see Crosslinks homo-, see Homocysteine seleno-, see Selenocysteine Cytochrome aa3, 327-329 Cytochromes ba3, 327-329 Cytochrome bo3, 318-321, 324 EPR studies, see EPR oxidase, 318, 320, 321, 324, 326, 327 Cytochrome c, 165, 166 Cytochrome c oxidase, 244, 317, 319-322, 343, 365, 366 active site, see Active sites bovine (heart), 318, 320, 326, 329 cyanide binding, 327-329 electron transfer in, see Electron transfer Fe(II)-carbon monoxide, 318 mammalian, 325, 330

Met. Ions Life Sei. 2009, 6, 461-496

472

S U B J E C T INDEX

[Cytochrome c oxidase] mutagenesis, 323 nitric oxide inhibition, 329, 330 photolysis, 318, 324, 326 proton transfer in, see Proton transfer structure, 323 Cytochrome P450 cam, 253-255 reductase, see Reductases Cytotoxicity of bleomycin, 349

D Decarboxylase uroporphyrinogen, 68 Deformylase peptide, 379 Degradation of aromatic molecules, 378, 380 Dehalogenation reductive, 34, 35, 60, 77, 78, 120 Dehydrases diol, see Diol dehydrase propanediol, 84 Dehydrogenases alcohol, see Alcohol dehydrogenase carbon monoxide, see Carbon monoxide dehydrogenase F 42 o"dependent methylenetetrahydromethanopterin, 222 H 2 -forming methylenetetrahydromethanopterin, see [Fe]-hydrogenases xanthine, 404, 405, 410 O-Demethylases, 59, 76, 77 vanillate, 77 Denitrification bacterial, 342 copper enzymes in, 340-344 Density functional theory calculations, 419-452 acetyl-coenzyme A synthase, 445, 446 cobalamins, 86, 425, 429, 431, 434, 435 cobinamides, 425 coenzyme B, 123 F 4 3 0 , 125, 127 , 43 7 [FeFe]-hydrogenases, 189, 192, 193, 198, 199, 447, 449

Met. Ions Life Sei. 2009, 6, 461-496

[Density functional theory calculations] Η-cluster of [FeFe]-hydrogenases, 449, 450 heme-copper oxidases, 324 methyl-coenzyme Μ reductase, 120, 440^142 methyl-coenzyme M, 123 MoCll C O D H , 333, 334 [NiFe]-hydrogenases, 170, 447 peptidylglycine a-hydroxylating monooxygenase, 310 time-dependent, 426, 429, 434, 438 truncated models, 434, 435, 438 5'-Deoxyadenosylcobalamin, see Coenzyme Bl2

Deoxyhemocyanin, 299 active site, 300 Deoxyribonucleic acid, see D N A Deprotonation constants, see Acidity constants Desulfitobacterium dehalogenans, 78 frappieri, 78 Desulfomicrobium baculatum, 156, 166 Desulfovibrio desulfuricans, 152, 156, 166, 168, 181-192 fructosovorans, 156, 165-169 gigas, 153, 155, 156, 168, 187, 398, 403, 411 vulgaris, 153, 156, 168, 189, 190 Detoxification of chlorinated compounds, 34 environmental pollutants, 340 reactive oxygen species, 374 Deuterium 2 H N M R , see N M R CD 3 -Ni(II)F 4 3 0 M, 120 hydrogen exchange, 168, 269-272, 275 isotope effects, 92 label, 96, 101, 106 Diazotrophs, 370 Dielectric constant, 434 D F T , see Density functional theory calculations Dihydrogen (see also Hydrogen), 172, 221, 225, 236, 237 acidity constant, see Acidity constants activation, 221, 222 catalytic cycle, 197 formation, 225

SUBJECT INDEX [Dihydrogen (see also Hydrogen)] heterolytic cleavage, 168, 172, 191, 235 p r o d u c t i o n , 370, 371 3,4-Dihydroxybenzoate, see P r o t o c a t e c h u a t e 2,4-Dihydroxypteridine, see L u m a z i n e 2,6-Dihydroxypurine, see X a n t h i n e 4,6-Dihydroxypyrazolo[3,4-