Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life

Ideas in Chemistry and Molecular Sciences Where Chemistry Meets Life Edited by Bruno Pignataro Ideas in Chemistry and...

Author: Bruno Pignataro

94 downloads 1155 Views 10MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Ideas in Chemistry and Molecular Sciences Where Chemistry Meets Life

Edited by Bruno Pignataro

Ideas in Chemistry and Molecular Sciences Edited by Bruno Pignataro

Related Titles Pignataro, Bruno (ed.)

Waldmann, H., Janning, P. (eds.)

Ideas in Chemistry and Molecular Sciences

Chemical Biology

Advances in Synthetic Chemistry 2010

Learning through Case Studies 2009 ISBN: 978-3-527-32330-2

ISBN: 978-3-527-32539-9

Jacob, Claus/Winyard, Paul G. (eds.) Pignataro, Bruno (ed.)

Ideas in Chemistry and Molecular Sciences Advances in Nanotechnology, Materials and Devices

Redox Signaling and Regulation in Biology and Medicine 2009 ISBN: 978-3-527-31925-1

2010 ISBN: 978-3-527-32543-6

Carta, Giorgio/Jungbauer, Alois

Gjerde, Douglas T./Hoang, Lee/ Hornby, David

RNA Puriﬁcation and Analysis

Protein Chromatography

Sample Preparation, Extraction, Chromatography

Process Development and Scale-Up

2009

2010

ISBN: 978-3-527-32116-2

ISBN: 978-3-527-31819-3

Pignataro, Bruno (ed.)

Hartmann, Roland K./Bindereif, Albrecht/ ¨ Astrid/Westhof, Eric (eds.) Schon,

Tomorrow’s Chemistry Today

Handbook of RNA Biochemistry

Concepts in Nanoscience, Organic Materials and Environmental Chemistry 2nd edition

Student Edition

2009 ISBN: 978-3-527-32623-5

2008 ISBN: 978-3-527-32534-4

Ideas in Chemistry and Molecular Sciences Where Chemistry Meets Life

Edited by Bruno Pignataro

The Editor Prof. Bruno Pignataro University of Palermo Department of Physical Chemistry Viale delle Scienze 90128 Palermo Italy

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Cover

Library of Congress Card No.: applied for

We would like to thank Dr. Adriana Pietropaolo (ETH Z¨urich) for providing us with the graphic material used in the cover illustration.

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. 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 .  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microﬁlm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not speciﬁcally marked as such, are not to be considered unprotected by law. Cover Design Adam Design, Weinheim Typesetting Laserwords Private Limited, Chennai, India Printing and Binding betz-druck GmbH, Darmstadt Printed in the Federal Republic of Germany Printed on acid-free paper ISBN: 978-3-527-32541-2 Set ISBN: 978-3-527-32875-8

V

Contents

Preface XIII List of Contributors Part I 1

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.7.1 1.7.2 1.7.3 1.7.4 1.7.5 1.7.6 1.8 1.8.1 1.8.2 1.9

2 2.1 2.2

XIX

Biochemical Studies

1

The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders 3 Fabio Arnesano Introduction 3 Metal Ions in the Brain 4 Brain Copper Homeostasis 5 Brain Copper and Neurodegenerative Disorders 8 The Role of Ubiquitin in Protein Degradation 9 Failure of the Ubiquitin System in Neurodegenerative Disorders 13 Interaction of Ubiquitin with Metal Ions 15 Thermal Stability of Ubiquitin 15 Spectroscopic Characterization of CuII Binding 15 Possible Implications for the Polyubiquitination Process 17 CuII -Induced Self-Oligomerization of Ub 17 Cooperativity between CuII -Binding and Solvent Polarity 18 Comparison with Other Metal Ions 19 Biological Implications 21 The Redox State of Cellular Copper 21 Ubiquitin and Phospholipids 22 Conclusions and Perspectives 23 Acknowledgments 24 References 24 The Bioinorganic and Organometallic Chemistry of Copper(III) 31 Xavi Ribas and Alicia Casitas Introduction 31 Bioinorganic Implications of Copper(III) 33

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

VI

Contents

2.2.1 2.2.2 2.2.3 2.2.4 2.3 2.3.1 2.3.1.1 2.3.1.2 2.3.1.3 2.3.2 2.3.3 2.3.3.1 2.3.3.2 2.4 2.5

3 3.1 3.2 3.3 3.4 3.4.1 3.4.1.1 3.4.1.2 3.4.1.3 3.4.1.4 3.4.2 3.4.3 3.4.3.1 3.4.3.2 3.4.3.3 3.5 3.5.1 3.5.2 3.5.2.1 3.5.2.2 3.5.2.3 3.5.2.4 3.5.3 3.5.3.1

Dinuclear Type-3 Copper Enzymes 33 Particulate Methano Monooxygenase (pMMO) 36 Mononuclear Monooxygenating Copper-based Enzymes 38 Trinuclear Copper Models for Laccase 40 Organometallic CuIII Species in Organic Transformations 41 C–C Bond Formation in Organocuprate(I) Catalysis 42 Conjugate Addition to α-Enones 42 Acetylene Carbocupration 43 SN 2 and SN 2 Alkylations 43 Aryl–Heteroatom Bond Formation in Cu-mediated Cross-coupling Processes 44 Aromatic and Aliphatic C–H Bond Organometallic Functionalizations 45 Catalytic Systems 45 Stoichiometric Systems 47 Miscellany: Cuprate Superconducting Materials 51 Overview and Future Targets 51 References 52 Chemical Protein Modiﬁcation 59 Gon¸calo J. L. Bernardes, Justin M. Chalker, and Benjamin G. Davis Introducing Diversity by Posttranslational Modiﬁcation 59 Chemistry: A Route to Modiﬁed Proteins 60 Challenges in Chemical Protein Modiﬁcation 61 Traditional Methods for Protein Modiﬁcation 61 Lysine Modiﬁcation 62 Activated Esters 62 Isocyanates and Isothiocyanates 62 Reductive Alkylation 62 IME Reagents 63 Glutamic and Aspartic Acid Modiﬁcation 64 Cysteine 64 Alkylation 65 Disulﬁdes 66 Desulfurization at Cysteine 67 Recent Innovations in Site-Selective Protein Modiﬁcation 70 Dehydroalanine: A Useful Chemical Handle for Protein Conjugation 71 Metal-Mediated Protein Modiﬁcation 71 Modiﬁcation at Natural Residues 72 Iridium-Catalyzed Reductive Alkylation of Lysine 74 Modiﬁcation of Unnatural Residues 74 Oleﬁn Metathesis at S-Allyl Cysteine 77 Metal-Free Methods for Modifying Unnatural Amino Acids 78 Oxime Ligation at Aldehydes and Ketones 78

Contents

3.5.3.2 3.5.3.3 3.5.3.4 3.5.4 3.6

Azide and Alkyne Modiﬁcation 79 Selective Modiﬁcation of Tetrazole-Containing Proteins 80 Tetrazine Ligation 81 Dual Modiﬁcation 81 Conclusion and Outlook 81 References 82 Part II

4

4.1 4.2 4.3 4.3.1 4.3.2 4.4 4.4.1 4.4.2 4.4.3 4.4.4 4.4.4.1 4.4.4.2 4.5

5

5.1 5.1.1 5.1.2 5.1.2.1 5.1.2.2 5.1.2.3 5.1.2.4 5.2 5.2.1 5.2.1.1 5.2.2 5.2.2.1 5.2.2.2 5.2.3

Drug Delivery 93

Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery 95 Pilar Ruiz-S´anchez Introduction 95 Transport Mechanism 96 Metabolism of B12 97 Adenosylcobalamin-Dependent Reactions 97 Methylcobalamin-Dependent Reactions 99 Vitamin B12 Derivatives 99 Structure 99 b-, d-, e-Cobalamin Derivatives 99 Modiﬁcations on the Ribose Moiety 101 β-Axial Position 102 Cobalamin Alkylation 102 Heterodinuclear Concept 103 Outlook 108 Acknowledgments 108 References 109 Strategies for Microsphere-Mediated Cellular Delivery 117 Rosario M. Sanchez-Martin, Lois M. Alexander, Juan Manuel Cardenas-Maestre, and Mark Bradley Introduction 117 Cellular Delivery 117 Delivery Devices 117 Liposomes 117 Cell-penetrating Peptides 118 Dendrimers 118 Nanomaterials 118 Microspheres 119 Biodegradable Microspheres 119 Preparation 119 Biostable Microspheres 120 Applications of Biostable Microspheres 120 Preparation 120 Microspheres and Solid-phase Chemistry 121

VII

VIII

Contents

5.2.3.1 5.2.3.2 5.2.3.3 5.2.3.4 5.2.4 5.2.4.1 5.2.4.2 5.2.5 5.2.5.1 5.2.5.2 5.2.6 5.2.6.1 5.3

Preparation of Microspheres 121 Fmoc Chemistry on Microspheres 121 Dual Functionality of Microspheres 122 Coupling Agents 124 Noncleavable Link to Microspheres 126 Microsphere-based Intracellular Sensing 126 siRNA Delivery 127 Cleavable Linkers 130 Ester Bonds 130 Disulﬁde Bonds 132 Bioconjugation 133 Streptavidin–Biotin 134 Future Perspectives 135 Acknowledgments 135 References 135 Part III

6

6.1 6.1.1 6.1.2 6.2 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 6.2.6 6.3

7 7.1 7.2 7.2.1 7.2.2

Research in Therapeutics 141

Fundamental Processes in Radiation Damage to DNA: How Low-Energy Electrons Damage Biomolecules 143 Ilko Bald Radiation Damage and the Role of Low-Energy Electrons 143 How Chemical Bonds are Broken by Low-energy Electrons 145 DEA Studies of Gas-Phase DNA Building Blocks: The Nucleobases 147 DEA Studies on Model Compounds for the DNA Backbone 148 Electron Attachment to d-Ribose 148 Cross-Ring Cleavage of d-Ribose Proceeds with Selective Charge Retention 150 The Nature of the Transient Negative d-Ribose Anions 154 One Step Further: Tetraacetyl-d-Ribose 155 The Use of Laser-Induced Acoustic Desorption (LIAD) to Study DEA to Larger Biomolecules 159 Sugar–Phosphate Cleavage Induced by 0 eV Electrons: DEA to d-Ribose-5 -Phosphate 159 Outlook and Future Prospects 161 Acknowledgments 162 References 163 Structure-Based Design on the Way to New Anti-infectives 167 Anna Katharina Herta Hirsch Introduction 167 Isoprenoids and the Nonmevalonate Pathway 169 4-Diphosphocytidyl-2C-methyl-d-erythritol Kinase (IspE) 170 Structure of IspE 170

Contents

7.2.3 7.3 7.3.1 7.3.1.1 7.3.1.2 7.3.2 7.3.3 7.3.4 7.3.4.1 7.3.4.2 7.3.5 7.4 7.4.1 7.4.2 7.4.3 7.4.4 7.5 7.5.1 7.5.2

8

8.1 8.1.1 8.1.2 8.1.3 8.2 8.2.1 8.2.2 8.2.2.1 8.2.2.2 8.2.2.3 8.2.2.4 8.3 8.3.1 8.3.2 8.4 8.4.1 8.4.2

Active Site of IspE 170 Targeting the CDP-Binding Pocket of IspE 174 Design 174 Possible Ribose Analogues 175 Design of the Vector 176 Optimization of the Ribose Analogue 176 Importance of the Vector 178 Optimization of the Filling of the Small, Hydrophobic Pocket 179 The ‘‘55% Rule’’ 179 Evaluation of Inhibitors Featuring Different Sulfone Substituents 180 Summary of the First-Generation Inhibitors 182 X-ray Cocrystal Structure Analysis 182 Design of Water-Soluble Inhibitors 182 Enzyme Assays of Inhibitors Designed to be Water Soluble 183 Structural Analysis 184 Lessons Learnt from the Cocrystal Structure 185 Conclusions and Outlook 185 Conclusions 185 Outlook 186 Acknowledgments 186 List of Abbreviations 186 References 187 Drug–Membrane Interactions: Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs 191 Marlene Lucio, ´ Jos´e L. F. C. Lima, and Salette Reis Biological Membranes 191 Role of Membranes in Life Maintenance 191 Structure and Composition of Membranes 191 Dynamic Molecular Organization of Membranes 193 Drug–Membrane Interactions 195 Possible Effects of Drugs on Membranes 195 Clinical Relevance of the Drug–Membrane Interaction Studies 195 Contribution for Drug Development 195 Understanding Therapeutic and Toxic Effect of Drugs 197 Understanding Mechanisms of Multidrug Resistance 198 Controlling Enzymatic Inhibition 198 Analysis and Quantiﬁcation of Drug–Membrane Interactions 199 Membrane Model Systems 199 Experimental Techniques 200 Drug–Membrane Interactions Applied to the Study of Nonsteroidal Anti-inﬂammatory Drugs (NSAIDs) 201 Drug Fundamental Physical–Chemical Studies 202 Membrane Structural and Dynamic Studies 203

IX

X

Contents

8.4.3 8.5

Results 205 Conclusions and Future Research Directions Acknowledgments 206 References 207

9

Targeting Disease with Small Molecule Inhibitors of Protein–Protein Interactions 215 Fedor Forafonov, Elena Miranda, Ida Karin Nordgren, and Ali Tavassoli Introduction 215 High-Throughput Screening of Chemical Libraries 216 High-Throughput Screening of Biosynthesized Libraries 220 Cyclic Peptide Inhibitors of AICAR Transformylase Activity 222 Cyclic Peptide Inhibitors of HIV Budding 224 Future Direction 226 Small Molecule Inhibitors of Tumor Hypoxia Response Network 226 Targeting Protein–Protein Interactions in Asthma 228 Targeting the Protein Interaction Networks of Inﬂuenza Virus 230 Acknowledgments 232 References 232

9.1 9.2 9.3 9.3.1 9.3.2 9.4 9.4.1 9.4.2 9.4.3

10 10.1 10.1.1 10.1.2 10.1.3 10.1.3.1 10.1.3.2 10.1.3.3 10.1.4 10.1.4.1 10.1.4.2 10.1.4.3 10.1.4.4 10.2 10.2.1 10.2.2 10.2.3 10.3

206

Cracking the Glycocode: Recent Developments in Glycomics 239 Lars Hillringhaus and J¨urgen Seibel State of the Art 239 Introduction 239 Carbohydrate-Based Drugs 239 Carbohydrate Synthesis 240 Chemical Synthesis 241 Enzymatic Synthesis 243 Glycoprotein Synthesis 244 Glycomics 245 Mass Spectrometry 245 Microarrays 246 Cell, Tissue, and Metabolic Labeling 246 Bioinformatics 247 Some New Insights in Glycomics 247 Microwave-Assisted Glycosylation for the Synthesis of Glycopeptides 247 Highly Efﬁcient Chemoenzymatic Synthesis of Novel Branched Thiooligosaccharides by Substrate Direction with Glucansucrases 251 Identiﬁcation of New Acceptor Speciﬁcities of Glycosyltransferase R with the Aid of Substrate Microarrays 257 Future Perspectives 259 Acknowledgments 259 References 260

Contents

Part IV 11

11.1 11.2 11.2.1 11.2.2 11.3 11.3.1 11.3.2 11.3.3 11.3.3.1 11.3.3.2 11.4

O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity 267 Anna Company General Introduction: O2 Activation and Model Systems 267 Copper Proteins Involved in O2 Activation 268 Hemocyanin 269 Tyrosinase 270 O2 Binding and Activation at Biomimetic Cu Complexes 272 Copper–Dioxygen Adducts 272 Ligand Architecture: Inﬂuence on Reactivity toward O2 275 Hydroxylation of Aromatic Rings: Mimicking Tyrosinase Activity 278 Intramolecular Aromatic Hydroxylation 278 Intermolecular ortho-Hydroxylation of Phenolic Compounds 280 Concluding Remarks 285 References 286 Part V

12

12.1 12.1.1 12.1.2 12.1.3 12.1.4 12.2 12.3 12.3.1 12.3.2 12.3.3 12.3.4 12.4

13

13.1 13.2 13.3

Enzyme Chemistry 265

Structure–Property Relationship and Biosensing 291

Chirality in Biochemistry: A Computational Approach for Investigating Biomolecule Conformations 293 Adriana Pietropaolo Introduction 293 Molecular Chirality in Living Systems 293 Protein Secondary Structures 295 Protein Secondary Structure Assignment 296 Intrinsic Chirality and Protein Secondary Structures 297 Computational Techniques for Studying Protein Dynamics 297 Employing Chirality to Analyze Protein Motions 298 The Chirality Index 298 Using Chirality to Understand Protein Structure 300 Chirality Index as a Tool for Monitoring Protein Dynamics 302 Chirality and Circular Dichroism 305 Perspectives 308 Acknowledgments 309 References 310 Collisional Mechanism–Based E-DNA Sensors: A General Platform for Label-Free Electrochemical Detection of Hybridization and DNA Binding Proteins 313 Francesco Ricci Introduction 313 E-DNA Signaling Mechanism 315 E-DNA Sensor for DNA Binding Proteins Detection 319

XI

XII

Contents

13.4 13.5

Conclusions and Future Perspectives 321 Acknowledgments 324 References 324 Index 327

XIII

Preface The idea of publishing books based on contributions given by emerging young chemists arose during the preparation of the ﬁrst EuCheMs (European Association for Chemical and Molecular Sciences) Conference in Budapest. In this conference I cochaired the competition for the ﬁrst European Young Chemist Award aimed at showcasing and recognizing the excellent research being carried out by young scientists working in the ﬁeld of chemical sciences. I then proposed to collect in a book the best contributions from researchers competing for the Award. This was further encouraged by EuCheMs, SCI (Italian Chemical Society), RSC (Royal Society of Chemistry), GDCh (Gesellschaft Deutscher Chemiker), and Wiley-VCH and brought out in the book ‘‘Tomorrow’s Chemistry Today’’ edited by myself and published by Wiley-VCH. The motivation gained by the organization from the above initiatives was, to me, the trampoline for co-organizing the second edition of the award during the second EuCheMs Conference in Torino. Under the patronage of EuCheMs, SCI, RSC, GDCh, the Consiglio Nazionale dei Chimici (CNC), and the European Young Chemists Network (EYCN), the European Young Chemist Award 2008 was again funded by the Italian Chemical Society. In Torino, once again, I personally learned a lot and received important inputs from the participants about how this event can serve as a source of new ideas and innovations for the research work of many scientists. This is also related to the fact that the areas of interest for the applicants cover many of the frontier issues of chemistry and molecular sciences (see also Chem. Eur. J. 2008, 14, 11252–11256). But, more importantly, I was left with the increasing feeling that our future needs of new concepts and new technologies should be largely in the hands of the new scientiﬁc generation of chemists. In Torino, we received about 90 applications from scientists (22 to 35 years old) from 30 different countries all around the world (Chem. Eur. J. 2008, 14, 11252–11256). Most of the applicants were from Spain, Italy, and Germany (about 15 from each of these countries). United Kingdom, Japan, Australia, United States, Brazil, Morocco, Vietnam, as well as Macedonia, Rumania, Slovenia, Russia, Ukraine, and most of the other European countries were also represented. In terms of applicants, 63% were male and about 35% were PhD students; the number of Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

XIV

Preface

postdoctoral researchers was only a small percentage, and only a couple of them came from industry. Among the oldest participants, mainly born between 1974 and 1975, several were associate professors or researchers at Universities or Research Institutes and others are lecturers, assistant professors, or research assistants. The scientiﬁc standing of the applicants was undoubtedly very high and many of them made important contributions to the various symposia of the 2nd EuCheMs Congress. A few ﬁgures help to substantiate this point. The, let me say, ‘‘h index’’ of the competitors was 20, in the sense that more than 20 applicants coauthored more than 20 publications. Some patents were also presented. Five participants had more than 35 publications, and, h indexes, average number of citations per publication, and number of citations, were as high as 16, 35.6, and 549, respectively. Several of the papers achieved further recognition as they were quoted in the reference lists of the young chemists who where featured on the covers of top journals. The publication lists of most applicants proudly noted the appearance of their work in the leading general chemistry journals such as Science, Nature, Angewandte Chemie, Journal of the American Chemical Society, or the best niche journals of organic, inorganic, organometallic, physical, analytical, environmental, and medicinal chemistry. All of this supported the idea of publishing a second book with the contributions of these talented chemists. However, in order to have more homogeneous publications and in connection to the great number of interesting papers presented during the competition, we decided to publish three volumes. This volume represents indeed one of the three edited by inviting a selection of young researchers who participated in the European Young Chemist Award 2008. The other two volumes concern the two different areas of synthetic chemistry and nanotechnology/materials-science and are, respectively, entitled ‘‘Ideas in Chemistry and Molecular Science: Advances in Synthetic Chemistry’’ and ‘‘Ideas in Chemistry and Molecular Sciences: Advances in Nanotechnology, Materials, and Devices’’. It is important to mention that the contents of the books are a result of the work carried out in several topmost laboratories around the world both by researchers who already lead their own group and by researchers who worked under a supervisor. I would like to take this occasion to acknowledge all the supervisors of the invited young researchers for their implicit or explicit support to this initiative that I hope could also serve to highlight the important results of their research groups. The prospect of excellence of the authors was evident from the very effusive recommendation letters sent by top scientists supporting the applicants for the Award. A ﬂavor of these letters is given by the extracts from some of the sentences below: ‘‘The candidate creativity in polymer design for therapeutic impact is extremely impressive. I believe that he is an excellent example of an outstanding young European chemist’’; ‘‘Is an energetic, enthusiastic, articulate, bright researcher’’; ‘‘Talented and enthusiastic scientist who has a lot of determination and persistence. I was impressed by the strong preparation, enthusiasm and motivation. First rate

Preface

researcher’’; ‘‘. . .outstanding, very broad and far reaching research achievements’’; ‘‘The personality and the scientiﬁc research achievements will certainly pave to the candidate the way as a future leader of an independent and creative research group. Due to his original and uncommon approach together with the high quality of results, he will become well known in the ﬁeld of medicinal chemistry. All this exciting work relied on his ability as a chemist and on his bright intellect, which allows him to move across disciplines seamlessly’’; ‘‘The candidate is a very articulate, knowledgeable, driven, clear minded, and extremely personable young scientist’’; ‘‘Application in this work has been outstanding, strong, powerful and very dedicated. The candidate will be in a very good position to make tremendous impact in the area of therapeutics and to become a leader in the area of organic chemistry in the next generation’’; ‘‘As one of the most talented graduate students to have emerged for our group he has the vision, skill and drive to make a real difference to the national research community’’; ‘‘. . .outstanding young scientist and a charming person’’; ‘‘. . .has an astonishing level of productivity’’; ‘‘. . .work of very high caliber’’; ‘‘Such deep and insightful investigations are refreshing and invigorating in my view especially in times that tend to favor simple, phenomenological results’’; ‘‘The candidate has been the intellectual driver and forged the necessary collaborations to address this complex problem. He deft integration of the various experimental and theoretic disciplines needed for the work is impressive’’; ‘‘The candidate performed outstanding research projects’’; ‘‘He should be a compelling and deserving candidate of the European Young Chemist award’’; ‘‘The candidate is for sure one of the most outstanding and promising young researchers that I know in the ﬁeld of molecular science. Despite being young, he has a lot of experience and high motivation. He is excellent in the lab and extremely smart person’’; ‘‘The candidate has performed work of top quality and originality and of great breadth’’; ‘‘. . .a special individual who ranks among the very best’’; ‘‘Exceptional in many respect. The candidate commands superior experimental and intellectual skills and has shown great chemical structure intuition in the project’’. ‘‘. . .outstanding scientiﬁc debater. The personality is impeccable’’. ‘‘Extremely gifted young researcher who has already developed an outstanding track record of achievement while working in one of the world’s premier laboratories in the USA’’; ‘‘Has the skills, expertise and leadership potential to discover new compounds that will revolutionize the therapy of some difﬁcult diseases. The work is well-aligned to the University’s strategy of supporting cross disciplinary research of the highest quality’’; ‘‘. . .unusually broad dimension of experimental skills’’; ‘‘This is a rare combination of expertise’’; ‘‘enables to undertake a diverse scope of key problems. In my opinion these are the type of individuals who will drive the ﬁeld forward’’; ‘‘. . .has an engaging, outgoing, attractive personality to which others instinctively respond’’; ‘‘Brilliant young chemist with a great potential’’. The chapters written by the various contributors cover several areas of life science and range from more or less typical biochemical studies, to research relevant in the therapeutic ﬁeld, to the enzyme or drug activity, to drug delivery, to computational structure–activity relation or sensors.

XV

XVI

Preface

In the area of biochemical studies, the book collects a contribution in the ﬁeld of inorganic chemistry of the brain with the chapter on the role of metal ions (Cu(II) in particular) and the ubiquitin–proteasome system on neurodegenerative disorder (Arnesano). In another chapter, the strategies for chemical modiﬁcation of proteins as well as selective installation of biochemical probes and tethering of therapeutic cargo to proteins are highlighted (Bernardes et al.). Another chapter (Ribas et al.) is centered on Cu(III) ion and its bioinorganic and organometallic chemistry. In particular, this contribution turns out to be relevant for understanding copper-dependent metalloenzymes. In the drug delivery domain, a chapter is dedicated to vitamin B12 as potential targeting molecule for therapeutic drug delivery (Ruiz S´anchez). This potentiality is vividly illustrated by the author with the following deﬁnition: vitamin B12 is an attractive ‘‘Trojan horse’’ for therapeutic drug delivery. A second contribution starts from the consideration that the cellular membrane poses a formidable barrier to the intracellular ﬂux of materials circulating extracellularly due to its selective permeability based predominantly on ionic charge, hydrophobicity, and size. As such, many drugs, nucleic acids, proteins, and other investigative constructs are not able to translocate the cellular membrane alone and often require a delivery vehicle for function and/or activity. After considering various vehicles this contribution focuses on a highly competitive cellular delivery technology based on polymeric microspheres-mediated cellular delivery (Sanchez-Martin). In particular, the author presents different strategies that can be applied for the attachment of biomolecules to the microspheres and their applications as a delivery system. Monodispersed populations of robust cross-linked microspheres of deﬁned sizes (from 200 nm to 2 µm) have been synthesized and standard solid phase multistep protocols have been applied to them. Several contributions are primarily dedicated to research that has relevance to the therapeutic ﬁeld. The chapter (Lucio et al.) on molecular mechanisms, underlying therapeutic and toxic effects of drugs, highlights the importance of membrane composition and the dynamic molecular organization of membranes in drug–membrane interaction. Also, the work aims to exemplify an application of drug–membrane interaction studies in the evaluation of the nonsteroidal anti-inﬂammatory drug effect on membranes. These studies may prove valuable in the design of novel drug formulations with increasing efﬁcacy and reduced side effects. A second contribution in the area deals with targeting diseases with small molecule inhibitors of protein–protein interactions (Tavassoli et al.). The chapter starts from the consideration that there are several thousand protein–protein interactions that control the majority of biological processes. There is therefore great potential for small molecule therapeutics that affect disease through modulation of protein–protein interactions. In this chapter, there is then a discussion on recent approaches taken toward controlling protein interactions with small molecules, from logical design using structural data to high-throughput screening of chemical and biological libraries. The authors go on to outline their own efforts in this

Preface

ﬁeld, and end with a detailed discussion of some of the important protein–protein interactions currently being targeted. Another chapter (Hirsh) is dedicated to the design and synthesis of inhibitors of the kinase IspE as potential antimalarians opening a number of research avenues summarized in the chapter itself, while another contribution (Hillringhaus) deals with recent developments in glicomics – a technology that can be anticipated that it will strongly inﬂuence tomorrow’s therapy in the areas of various diseases. A last contribution (Bald) in this area deals with the important studies on the electron-induced modiﬁcation of biomolecules, which may have potential implications for the design of new chemotherapeutic and radiosensitizing drugs and for the development of more efﬁcient protocols in cancer therapy. Understanding how nature works in the area of enzyme chemistry is the goal of a chapter (Company) dedicated to the tyrosinase reactivity. In this chapter, in particular, it is shown how the model chemistry approach has allowed to reproduce and to better understand the mechanisms by which O2 is activated in the dinuclear copper protein just called tyrosinase. The studies summarized in another chapter (Ricci) refer to electrochemical biosensors, termed E-DNA sensors, that are based on the target binding–induced folding of electrode-bound DNA probes and gives contribution on E-DNA signaling mechanisms. In addition, studies on E-DNA sensors for DNA binding protein detection are also reported. The E-DNA sensing platform has then demonstrated to be not only a promising and appealing approach for the sequence-speciﬁc detection of DNA and RNA but also to be ﬂexible enough to be adaptable for the detection of DNA–protein interaction. Taking into account that chirality selection is a key issue in many important biochemical phenomena such as protein folding and enzyme recognition, another chapter discusses the role of chirality in the chemistry of the life sciences. In particular, a novel methodology for the study of local chirality is presented, which provides one with a deeper understanding of the connection between secondary structure and protein ﬂexibility. This computational study for investigating biomolecule conformation reported in the chapter (Pietropaolo) suggests that chirality is a central organizing principle in life that confers order at every length scale up to the full cell. Probably, everyone who reads this book will have their own opinion on what is relevant for the future of life chemistry, and in this respect I would like to notify that, due to its peculiar genesis, the book reﬂects the opinions of a select group of young chemists and therefore does not pretend to cover the whole area. The main aim is just to offer a variety of individual views that will provoke thought possibly giving an attractive insight into the minds and research areas for the next generation of chemical and molecular sciences, especially those related with life sciences and technologies. Starting from this point, I hope that the many ideas that can be grasped sailing through the various contributions by the young authors of the book should be very useful for helping to make several step forward in the ﬁeld of chemistry and molecular sciences to solve problems related to our health or more generally to our

XVII

XVIII

Preface

life as well as to discover new tools in medicinal chemistry needed for the present and the future society. I cannot end this preface without acknowledging all the authors and the persons who helped me in the book project together with all the societies (see the book cover) that motivated and sponsored the book. I’m personally grateful to Professors Giovanni Natile, Francesco De Angelis and Luigi Campanella for their motivation and support in this activity. Palermo, October 2009

Bruno Pignataro

XIX

List of Contributors Lois M. Alexander University of Edinburgh Chemical Biology Section School of Chemistry Joseph Black Building West Mains Road Edinburgh EH9 3JJ Scotland Fabio Arnesano University of Bari ‘‘A. Moro’’ Department of Farmaco-Chimico Via E. Orabona 4 70125 Bari Italy Ilko Bald Freie Universit¨at Berlin Institut f¨ur Chemie und Biochemie–Physikalische und Theoretische Chemie Takustraße 3 D-14195 Berlin Germany

Gonc¸alo J. L. Bernardes University of Oxford Department of Chemistry Chemistry Research Laboratory 12 Mansﬁeld Road Oxford OX1 3TA United Kingdom Mark Bradley University of Edinburgh Chemical Biology Section School of Chemistry Joseph Black Building West Mains Road Edinburgh EH9 3JJ Scotland Alicia Casitas Universitat de Girona Department de Qu´ımica Grup de Qu´ımica Inorg`anica i Supramolecular (QBIS) Campus Montilivi 17071 Girona – Catalonia Spain

and Interdisciplinary Nanoscience Center (iNANO) Aarhus University Ny Munkegade 8000 Aarhus C Denmark

Justin M. Chalker University of Oxford Department of Chemistry Chemistry Research Laboratory 12 Mansﬁeld Road Oxford OX1 3TA United Kingdom

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

XX

List of Contributors

Anna Company Technische Universit¨at Berlin Institut f¨ur Chemie. Sekretariat C2 Straße des 17.Juni 135 D-10623 Berlin (Germany) Benjamin G. Davis University of Oxford Department of Chemistry Chemistry Research Laboratory 12 Mansﬁeld Road Oxford OX1 3TA United Kingdom Fedor Forafonov University of Southampton School of Chemistry University Road, Highﬁeld Southampton SO17 1BJ United Kingdom Lars Hillringhaus University of Oxford Department of Chemistry Chemistry Research Laboratory 12 Mansﬁeld Road Oxford OX1 3TA United Kingdom Ida Karin Nordgren University of Southampton School of Chemistry University Road, Highﬁeld Southampton SO17 1BJ United Kingdom Anna Katharina Herta Hirsch Universit´e de Strasbourg Institut de Science et d’Ing´enierie Supramol´eculaires 8 All´ee Gaspard Monge Strasbourg 67000 France

Jos´e L. F. C. Lima Universidade do Porto Faculdade de Farm´acia Requimte, Servic¸o de Qu´ımica-F´ısica Rua An´ıbal Cunha, 164 Porto 4099-030 Portugal Marlene L´ucio Universidade do Porto Faculdade de Farm´acia Requimte, Servic¸o de Qu´ımica-F´ısica Rua An´ıbal Cunha, 164 Porto 4099-030 Portugal Juan Manuel Cardenas-Maestre University of Edinburgh Chemical Biology Section School of Chemistry Joseph Black Building West Mains Road Edinburgh EH9 3JJ Scotland Elena Miranda University of Southampton School of Chemistry University Road, Highﬁeld Southampton SO17 1BJ United Kingdom

List of Contributors

Adriana Pietropaolo Universit´a di Catania Dipartimento di Scienze Chimiche Viale A. Doria 6 Catania I-95125 Italy

Francesco Ricci University of Rome Tor Vergata Dipartimento di Scienze e Tecnologie Chimiche Via della Ricerca Scientiﬁca 00133 Rome Italy

and

Pilar Ruiz-S´anchez Universit¨at Z¨urich Anorganisch-chemisches Institut Winterthurerstr. 190 8057 Z¨urich Switzerland

´ ETH Zurich Department of Chemistry and Applied Biosciences Institute of Computational Science USI Campus Via Giuseppe Bufﬁ 13 CH-6900 Lugano Switzerland Salette Reis Universidade do Porto Faculdade de Farm´acia, Requimte, Servic¸o de Qu´ımica-F´ısica Rua An´ıbal Cunha, 164 Porto 4099-030 Portugal Xavi Ribas Universitat de Girona Department de Qu´ımica Grup de Qu´ımica Inorg`anica i Supramolecular (QBIS) Campus Montilivi 17071 Girona – Catalonia Spain

Rosario M. Sanchez-Martin University of Edinburgh Chemical Biology Section School of Chemistry Joseph Black Building West Mains Road Edinburgh EH9 3JJ Scotland J¨ urgen Seibel University of W¨urzburg Institute of Organic Chemistry Am Hubland D-97074 W¨urzburg Germany Ali Tavassoli University of Southampton School of Chemistry University Road, Highﬁeld Southampton SO17 1BJ United Kingdom

XXI

1

Part I Biochemical Studies

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

3

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders Fabio Arnesano

1.1 Introduction

Ubiquitin (Ub) plays a crucial role in intracellular protein degradation via the proteasome and the autophagy–lysosome pathways [1]. Failure to eliminate misfolded proteins can lead to the formation of toxic aggregates and cell death [2]. Insoluble protein aggregates enriched with Ub are a hallmark of most neurodegenerative disorders including Parkinson’s, Alzheimer’s, amyotrophic lateral sclerosis and prion diseases [3, 4]. All of these disorders have been linked to metal accumulation and disturbance of redox and metal homeostasis in the brain [5–7], and metal ions have been implicated in the aggregation of disease-related, amyloidogenic proteins [8–12]. The potential role of metal ions in the aggregation of Ub has recently been examined [13, 14]. CuII is different from ZnII , NiII , AlIII , or CdII in that it binds to the N-terminal end of Ub, destabilizes the protein, and promotes its oligomerization into spherical particles. By mimicking the condition of low dielectric constant experienced near a membrane surface, the assembly of spherical oligomers of Ub yields a series of intermediate species leading to an extended nonﬁbrillar ﬁlament network. Aggregate disassembly is triggered by CuII chelation or reduction [14]. Intermediate annular and porelike structures, stabilized by the interaction of CuII -induced Ub oligomers with lipid bilayers, resemble toxic protoﬁbrillar species produced by amyloidogenic proteins, which cause membrane permeabilization and disruption of metal homeostasis [15–19]. Susceptibility to aggregation of Ub represents a potential risk factor for disease onset or progression while cells attempt to tag and process toxic substrates. CuII binding and proximity to biological membranes appear to dramatically increase the aggregation propensity of Ub and other disease-related proteins, thus emphasizing the importance of preserving cellular compartmentalization and metal homeostasis for the correct functioning of protein degradation systems. Recent ﬁndings reinforce the vision of metal ions as key factors and promising therapeutic targets in protein conformational disorders [20, 21]. New strategies are being developed that will help to investigate their functional and pathogenic interactions in vivo. Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

4

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

1.2 Metal Ions in the Brain

The brain is a specialized organ that controls cognitive and motor functions. To carry out its functions the brain requires the highest concentrations of metal ions in the body and the highest per-weight consumption of body oxygen [22]. Metal ions in the brain fulﬁll catalytic and structural roles, which include the stabilization of biomolecules (e.g., MgII in nucleic acids, ZnII in Zn-ﬁnger transcription factors) or dynamic processes (e.g., NaI and KI in ion channels, CaII in neuronal cell signaling) [23]. The dynamic partitioning of these metal ions is controlled by ion-speciﬁc channels that selectively allow passage of ions in and out of cells. In the brain, the uneven distribution of NaI and KI ions across a cell membrane creates a potential that enables transmission of nervous pulses. CaII is also a key modulator of molecular information transfer within and between cells during neurotransmission; most eukaryotic cells either export or store CaII within membrane-enclosed vesicles to maintain cytosolic- free CaII levels at 100–200 nM, roughly 10 000-fold less than in the extracellular space [23]. More recently, considerable attention has been directed to the role of transition metal ions in the brain [22, 24]. Zn, Fe, Cu, and related d-block metals are emerging as signiﬁcant players in both neurophysiology and neuropathology, particularly with regard to aging and neurodegenerative diseases [25]. Relatively high concentrations of these d-block metals are present within the different cellular compartments, the values ranging from 100 to 1000 µM [22]. The metal concentrations in brain tissue are up to 10 000-fold higher than those in common neurotransmitters and neuropeptides. Not only do these metals serve as components of various proteins and enzymes essential for normal brain function, but, in the labile form, are also involved in specialized brain activities; therefore, if misregulation of their homeostasis occurs, toxicity, mediated also by oxidative stress in the case of Cu and Fe [26], could ensue. Oxidative stress has been identiﬁed in many neurodegenerative diseases, and is commonly associated with increased levels of at least one of these transition metal ions in speciﬁc brain regions [27]. Transporters for Cu, Zn, Fe, and Mn play an important part in the intracellular distribution of these metals [28], such that defects in their regulation, which could possibly occur with aging, may create an environment that could result in protein misfolding and aggregation, thereby accelerating degenerative conditions [2, 29, 30]. Notably, brain homeostasis of metals is intertwined with changes in one metal leading to changes in the levels of other metals [24]. This is well established for Cu and Fe, where decreased Cu bioavailability may result in altered Fe levels, and for Fe and Mn, where Fe deﬁciency leads to a signiﬁcant increase in brain Mn. The following discussion will be focused on basic aspects of brain Cu homeostasis. The widespread distribution and mobility of Cu required for normal brain function, along with the numerous correlations between Cu misregulation and a variety of neurodegenerative diseases, have prompted interest in studying its roles in neurophysiology and neuropathology [26, 31].

1.3 Brain Copper Homeostasis

1.3 Brain Copper Homeostasis

Copper is the third-most abundant transition metal in the brain, after Fe and Zn, with average neuronal Cu concentrations of ∼0.1 mM. This redox-active nutrient is distributed unevenly within brain tissue, as Cu levels in the gray matter are twoto threefold higher than those in the white matter. Cu is particularly abundant in the locus coeruleus (1.3 mM), the neural region responsible for physiological responses to stress and panic, as well as the substantia nigra (0.4 mM), the center for dopamine production in the brain [32]. The major oxidation states for Cu ions in biological systems are cuprous CuI and cupric CuII ; the former is more common in the reducing intracellular environment, and the latter is dominant in the more oxidizing extracellular environment [33]. Levels of extracellular CuII vary from 10−25 µM in blood serum, 0.5−2.5 µM in cerebrospinal ﬂuid (CSF), and 30 µM in the synaptic cleft. Intracellular Cu levels within neurons can reach concentrations higher than 2–3 orders of magnitude [32]. Like Zn and Fe, brain Cu is partitioned into tightly bound and labile pools. Owing to its redox activity, Cu is an essential cofactor in numerous enzymes that handle the chemistry of oxygen or its metabolites, including cytochrome c oxidase (CcO), Cu, Zn superoxide dismutase (Cu,Zn-SOD1), ceruloplasmin (CP), dopamine β monooxygenase (DβM), peptidylglycine α-hydroxylating monooxygenase (PHM), and tyrosinase [31]. Because of its propensity to trigger aberrant redox chemistry and oxidative stress when unregulated, the brain maintains strict control over its Cu levels and distributions. An overview of homeostatic Cu pathways in the brain is given in Figure 1.1. Many of the fundamental concepts for neuronal Cu homeostasis are derived from studies in yeast, but the brain provides a more complex system with its own unique and largely unexplored inorganic physiology. There is little ‘‘free’’ Cu in the yeast cytoplasm, which is due to the tight regulation of metallochaperones [35, 36]; however, many open questions remain concerning the homeostasis of organelle Cu stores, particularly in higher organisms with specialized tissues. Some data suggest that yeast and mammals possess pools of labile Cu in the mitochondrial matrix [37]. Uptake of Cu by the blood–brain barrier (BBB) is considered to occur through the P-type ATPase ATP7A, which can pump Cu into the brain [38]. Mutations in the related gene lead to Menkes disease, an inherited neurodegenerative disorder that is globally characterized by brain Cu deﬁciency. This phenotype is mirrored by Wilson disease, which involves mutations in the ATP7B gene responsible for excretion of excess Cu from the liver into the bile. Loss of ATP7B function leads to abnormal increase of Cu in the liver [39]. The extracellular trafﬁcking of brain Cu differs from that in the rest of the body. CSF, the extracellular medium of the brain and central nervous system, possesses a distinct Cu homeostasis from blood plasma, which carries Cu to organs in the rest of the body. Cp, a multicopper oxidase that is essential for Fe metabolism, is

5

6

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

Cu+

1

tr hC

Cu2+

Synapse

Reductase MT3 c

CcO

E Sco1/ Cox11

Atox1

S CC

PrP

ome ndos

Cox17

Metalloprotein

SOD1 AT

P7

?

?

A

Presynaptic cell

AP

P

Postsynaptic cell

Figure 1.1 A schematic model of neuronal copper homeostasis. Reprinted with permission from [34]. Copyright 2008 American Chemical Society.

the major carrier of CuII in the plasma, but houses less than 1% of Cu in CSF [32]. The primary protein or small-molecule ligands for Cu in CSF remain unidentiﬁed. Uptake of Cu into brain cells requires reduction of CuII to CuI . Steap proteins may fulﬁll this role like the yeast ferric and cupric reductases Fre1 and Fre2 [40]. Following reduction, CuI ions can be transported into cells through a variety of trafﬁcking pathways [41, 42]. A major class of proteins involved in cellular Cu uptake is the copper transport (Ctr) protein family. Ctr1 is a representative member that is ubiquitously expressed. It resides predominantly in the plasma membrane and is essential for the survival of mammalian embryos and for Cu import into neurons and astrocytes [43]. Elevated Cu stimulates rapid endocytosis and degradation of Ctr [44]. Ctr1 contains three transmembrane helices, an N-terminal extracellular domain, and a C-terminal cytosolic domain. Electron crystallography revealed that Ctr1 is trimeric and possesses the type of radial symmetry associated with the structure of certain ion channels [45]. A region of low protein density at the center of the trimer is consistent with the existence of a Cu permeable pore. Mutagenesis studies have established that a methionine(Met)-rich motif in the N-terminal domain and a Met-rich motif at the extracellular end of the second transmembrane helix of Ctr1 play a pivotal

1.3 Brain Copper Homeostasis

role in the mechanism of Cu uptake [46, 47]. The mechanisms of Cu translocation across cellular membranes, however, remain largely unknown. In addition to Ctr1, prion protein (PrP) and amyloid precursor protein (APP) are two other abundant Cu-binding proteins, found speciﬁcally at brain cell surfaces, implicated in Cu uptake/efﬂux [48, 49]. In particular, PrP is localized in synaptic membranes of presynaptic neurons. Mammalian PrP contains at least four octapeptide repeats in the N-terminal region that can bind CuII . Millimolar concentrations of CuII induce endocytosis of PrP, suggesting that PrP may act as a buffer for Cu in the synaptic cleft, maintaining presynaptic Cu concentrations while preventing CuII -related toxicity in the extracellular space [49, 50]. Upon its entry into brain cells, CuI can be funneled to its ultimate intracellular destinations through the use of Cu chaperone proteins or buffering by metallothioneins (MTs), such as MT1 (ubiquitously expressed) and MT3 (expressed in the brain) [51]. The metallochaperones function not only as intracellular Cu delivery proteins but also as protective agents against toxicity resulting from unbound and unregulated Cu ions [35, 36]. Three human Cu chaperones have been characterized so far: Atox1, CCS, and Cox17. Atox1 loads CuI into the Menkes and Wilson P-type ATPases, ATP7A and ATP7B, which mediate Cu delivery to the secretory pathway from the trans-Golgi network (TGN) to the plasma membrane [41]. Both Atox1 and ATP7A/B contain CXXC sequence motifs that are essential for CuI binding and exchange of CuI between the two partner proteins [52]. The combination of available structural and biochemical data suggests a docking model that involves CuI transfer through twoand three-coordinate intermediates [53–56]. ATP7A and ATP7B play multiple roles in neurons from the delivery of Cu to cuproenzymes involved in neurotransmitter synthesis and metabolism, such as DβM, to the removal of excess Cu via secretion or vesicular sequestration [39]. To carry out this function, ATP7A undergoes Cu-stimulated translocation from the Golgi to the plasma membrane [57]. Metabolic studies also revealed that translocation of ATP7A after N-methyl d-aspartate (NMDA) receptor activation is associated with rapid release of Cu from hippocampal neurons [58, 59], a ﬁnding that suggests a role for Cu in the modulation of synaptic activity [60]. The copper chaperone for superoxide dismutase (CCS) inserts Cu into SOD [61]. Cu,Zn-SOD1 is a ubiquitous component of the cellular antioxidant system, which catalyzes disproportionation of the superoxide anion to oxygen and hydrogen peroxide [62]. Active Cu,Zn-SOD1 is a dimer; each subunit binds one Cu and one Zn ion, and contains an intramolecular disulﬁde bond [63]. On the contrary, in the immature, apo form of SOD1, cysteines are in the reduced state and the protein is a monomer [64, 65]. CCS docks with and transfers the Cu ion to the latter form of SOD [66], and also catalyzes disulﬁde bond formation [67, 68]. CCS is made of three domains: the N-terminal domain I has a fold similar to Atox1 and contains a conserved CXXC motif, domain II has a fold similar to SOD1 and participates in target recognition, domain III is constituted by ∼30 residues and contains a CXC Cu-binding motif [69]. While domain III is essential for CCS activity in vivo, the requirement for domain I is only apparent under Cu-limiting conditions [69].

7

8

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

A third Cu chaperone, Cox17, is involved in Cu delivery to CcO in mitochondria [70, 71]. CcO is a 13-subunit complex embedded in the mitochondrial inner membrane and a key component of the respiratory chain that reduces oxygen to water [72]. Two Cu ions form a dicopper cluster, designated CuA , in a CcO subunit, while a third Cu ion, designated CuB , forms a dinuclear site with heme a in another CcO subunit [72]. Cox17 acts as CuI donor for Sco1 and Cox11 [73]. Sco1, in turn, transfers Cu to CuA [74], while Cox11 is involved in CuB assembly [75]. Cox17 contains two CX9 C motifs implicated in two intramolecular disulﬁde bonds [76] and a conserved CC motif essential for CuI binding [77]. Fully reduced Cox17 binds up to four CuI ions in a polycopper cluster and undergoes oligomerization [76, 78].

1.4 Brain Copper and Neurodegenerative Disorders

Disruption of Cu homeostasis is implicated in a number of neurodegenerative diseases, including Alzheimer’s disease (AD), prion diseases, Parkinson’s disease (PD), familial amyotrophic lateral sclerosis (ALS), Menkes disease, and Wilson disease [5]. In all these disorders, the deleterious effects of Cu stem from its dual abilities to bind ligands and trigger uncontrolled redox chemistry. A dominant risk factor associated with most neurodegenerative diseases is increasing age. A positive correlation with chronic occupational exposure to Cu and other metal ions in industrialized countries has also been recognized [79, 80]. Several studies have reported a rise in the levels of brain Cu from youth to adulthood [32]. However, biologically available Cu levels drop markedly with advanced age and in AD brain [20, 81]. The connection between Cu and AD pathology is due mainly to its reactions with APP and its β-amyloid cleavage product (Aβ), that result in imbalance of extracellular and intracellular brain Cu pools [20]. Aberrant binding of CuII to APP triggers its reduction to CuI with concomitant disulﬁde bond formation; this intermediate can then participate in reactive oxygen species (ROS) production [82]. Extracellular Aβ deposits from AD brains (amyloid plaques) are rich in Cu, in addition to Zn and Fe [83]. The MT3, released in the synaptic cleft by neighboring astrocytes, has the potential to ameliorate this adverse interaction, but is downregulated in AD [84]. Moreover, the β-secretase β-site of amyloid precursor protein cleaving enzyme (BACE1), involved in APP cleavage, possesses a CuI -binding site in its C-terminal cytosolic domain through which it interacts with domain I of CCS, indicating that intracellular Cu levels can have an impact on Aβ generation [85]. Altered brain Cu distribution in AD, with abnormal accumulation of Cu in amyloid plaques and Cu deﬁciency in neighboring cells, is accompanied by a loss of Cu-dependent enzymes (e.g., CcO, Cu,Zn-SOD1, CP). Therefore, administration of Cu chelators such as clioquinol, that can reverse Aβ aggregation and redistribute brain Cu pools acting as ionophores, can have dual beneﬁcial effects [20]. It is also found that Cu-bound clioquinol and other Cu complexes can exhibit proteasome-inhibitory abilities [86, 87].

1.5 The Role of Ubiquitin in Protein Degradation

Prion diseases are also linked to brain Cu misregulation, where opposing CuII and MnII levels may inﬂuence the conversion of PrP into the toxic, protease-resistant form, PrPSc [88, 89]. PrP may act as a Cu-chelating agent, when extracellular Cu reaches high concentration peaks (15–300 µM) such as during synaptic transmission and depolarization [8]. Another hypothesis is that the binding of Cu to PrP could act directly to detoxify ROS, performing SOD-like activity [90]. In one proposal for prion toxicity, excess free Cu further exacerbates the disease by promoting oxidative stress [91]. Onset of PD is accompanied by death of dopaminergic neurons and intracellular accumulation of Lewy bodies [92], which are protein aggregates containing α-synuclein (α-syn), an abundant protein in the brain whose function is still unclear [93]. Monomeric α-syn, which has no persistent structure in aqueous solution, is known to bind anionic lipids [94] with a resulting increase in α-helix structure [95, 96]. Factors including oxidative stress and presence of various metal ions promote its ﬁbrillation [97, 98]. CuII also promotes the self-oligomerization of α-syn [10] and its oxidation and aggregation in the presence of H2 O2 [99]. Although α-syn is widely expressed in the brain, inclusions of α-syn are commonly localized in the substantia nigra, locus coeruleus, and cerebral cortex, which are the regions where Cu is abundant [32]. In PD brain, increased levels of Cu are found in the CSF [100]. Familial ALS is an inherited neurodegenerative disorder stemming from mutations in Cu,Zn-SOD [101, 102]. Three main hypotheses exist regarding the molecular mechanisms of this disease: (i) the loss-of-function mechanism, which results in toxic accumulation of superoxide by lack of SOD1 protection, (ii) the gain-of-function mechanism, in which SOD1 exhibits enhanced peroxidase activity by aberrant redox chemistry, and (iii) the aggregation mechanism, where SOD1 aggregates are induced by decreased availability of Cu and Zn ions and are stabilized by intermolecular disulﬁde bonds [103, 104]. The role of Cu homeostasis in this disease remains uncertain, however molecular machineries controlling redox homeostasis in mitochondria [68] appear to be essential for intramolecular disulﬁde bond formation and correct metal incorporation into SOD1, two processes involving the CCS metallochaperone [105].

1.5 The Role of Ubiquitin in Protein Degradation

The unique morphology of neurons (with specialized zones for presynaptic neurotransmitter release and postsynaptic receptor activation) and the plasticity of synapses (which is tightly coupled to changes in the synaptic proteome) impose special challenges on the cellular machinery for both protein synthesis and degradation [106]. Protein degradation has important roles in both neuronal development and long-term synaptic plasticity. Moreover, many neurodegenerative diseases are associated with abnormal protein aggregates, implicating degradative dysfunction [4, 107].

9

10

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

Major proteases in eukaryotic cells are conﬁned to specialized protein complexes (proteasomes) and organelles (lysosomes) to prevent nonspeciﬁc proteolysis. The ubiquitin-proteasome system (UPS) is responsible for degrading most intracellular, soluble proteins, but it can also degrade transmembrane proteins if they are extracted from the membrane into the cytosol (Figure 1.2). The lysosome degrades most membrane and endocytosed proteins, but it can also digest cytosolic proteins through autophagy (Figure 1.3) [1]. In both cases ubiquitin (Ub) plays a crucial role. Ub is a small protein of 76 aminoacids folded in a compact globular structure in which a mixed parallel/antiparallel β-sheet packs against an α-helix generating a hydrophobic core [108]. Not found in bacteria, this protein is ubiquitous in eukaryotes and has highly conserved sequences, the human and the yeast proteins differing by only three residues. The remarkable degree of sequence conservation underscores its important physiological role [109]. Ubiquitination is a posttranslational modiﬁcation that forms an isopeptide bond between a lysine residue on the protein and the C-terminus of Ub. The ubiquitination requires four different classes of enzymes: E1–E4 [110, 111]. First, Ub is covalently conjugated to E1 (Ub-activating enzyme) in an ATP-dependent reaction, and then it is transferred to E2 (Ub-conjugating enzyme). E3 (Ub-protein ligase) then transfers Ub from E2 to the substrate protein and is largely responsible for target recognition through physical interactions with the substrate. After the ﬁrst Ub has been attached (monoubiquitination), E3 can also elongate the Ub chain by creating Ub–Ub isopeptide bonds. Finally, E4 enzymes (chain elongation factors) are a subclass of E3-like enzymes that only catalyze chain extension (Figure 1.2) [110, 111]. Ub has seven lysine residues (K6, K11, K27, K29, K33, K48, and K63), all of which are available and indeed used in vivo for chain extension. The signiﬁcance of complex ubiquitination patterns is only partially understood: K48 chains lead to degradation of the substrate by the 26S proteasome, whereas monoubiquitination and K63 chains are known to activate cell signals in several pathways including tolerance of DNA damage, inﬂammatory response, ribosomal protein synthesis, endocytosis, and protein trafﬁcking [111]. The 26S proteasome is a large multisubunit complex of ∼2 MDa, localized in the cytosol and nucleus, and composed of a 20S proteolytic core and one or two 19S regulatory caps. After substrate–proteasome association, deubiquitinating enzymes (DUBs) and ATP-dependent unfoldase activities help the substrate to enter the proteolytic lumen of the 20S core by regenerating monomeric Ub [110, 111]. Notably, the cleavage of the isopeptide bond between the substrate and the most proximal Ub of the polyUb chain requires the metalloprotease activity of a 19S proteasome subunit, which contains a JAMM (JAB1 (Jun activation domain-binding protein-1)/MPN (Mpr1 Pad1 N-terminal domain)/Mov34 metalloenzyme) domain with a coordinated ZnII ion [112, 113]. Ub is also involved in the lysosomal degradation pathway. Lysosomes are organelles that contain acid hydrolases that break down biomolecules. The hydrolases in the lumen of lysosomes (pH 4–5) and late endosomes (pH 5–6) are highly active in acidic environments but loose their activities in the cytosol (pH ∼ 7.2) [114]. Many types of signals can regulate endocytosis and sorting of lysosomes, including

1.5 The Role of Ubiquitin in Protein Degradation

11

ATP ADP E2

E2

E1 Ubiquitin

E3 E3/E4

(K63)

DUB

DUB Target protein

DUB

DUB E3/E4 (K48) Proteasome translocation

Diffusion, chaperone, shuttling factor

26S proteasome

ATP ADP

DUB, unfoldase

Figure 1.2 The ubiquitin-proteasome system. Reprinted by permission from Macmillan Publishers Ltd: Nature Reviews Neuroscience [1], copyright 2008.

12

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

Micro autophagy

Exocytosis Endocytosis Recycling endosome

Macro autophagy Early endosome

Autophagosome

Lysosome Multivesicular body

Surface protein

Chaperone-mediated autophagy

Late endosome

Ubiquitin

Acid hydrolase

LAMP2 protein

Cytosolic protein

Chaperone

Figure 1.3 The autophagy-lysosome pathway. Reprinted by permission from Macmillan Publishers Ltd: Nature Reviews Neuroscience [1], copyright 2008.

monoubiquitination and K63-polyubiquitination [115, 116]. Intracellular proteins can enter lysosomes through several autophagic mechanisms. In macroautophagy, large amounts of cytosolic materials or even organelles are surrounded by a double-membrane structure (autophagosome) that fuses with lysosomes. In microautophagy, a small amount of the cytoplasm is internalized through lysosomal invagination (Figure 1.3) [117, 118]. Different classes of misfolded proteins partition between two separate intracellular compartments, one next to the nucleus (juxtanuclear) and the other near vacuoles (perivacuolar). Soluble ubiquitinated proteins go to the juxtanuclear compartment, whereas insoluble proteins accumulate in the perivacuolar compartment [119]. In addition to the spatial segregation, the fate of the proteins in these two compartments is divergent. Proteins in the juxtanuclear compartment are in close proximity to the 26S proteasome, whereas the perivacuolar compartment is targeted by proteins implicated in autophagy. Misfolded proteins may interact inefﬁciently with the normal quality control machinery and as a consequence are shunted to the perivacuolar pathways, as observed for PrP [119]. Enhancing the ubiquitination of a PrP enhances its targeting to the juxtanuclear compartment. Likewise, blocking the ubiquitination of proteins that normally go to the juxtanuclear compartment leads them to the perivacuolar compartment.

1.6 Failure of the Ubiquitin System in Neurodegenerative Disorders

1.6 Failure of the Ubiquitin System in Neurodegenerative Disorders

Protein misfolding and the subsequent assembly of protein molecules into aggregates of various morphologies represent common mechanisms that link a number of important human disorders, known as conformational diseases. These disorders include numerous neurodegenerative diseases and many systemic, localized, and familial amyloidoses [29]. A hallmark of these diseases is the accumulation of insoluble protein aggregates, termed amyloid, within affected cells. Amyloid ﬁbrils have a characteristic cross-β structure with a ribbon-like β-sheet that extends over the length of the ﬁbril and is comprised of β-strands that run approximately perpendicular to the long axis of the ﬁbril. Backbone H-bonds that link the β-strands are nearly parallel to the ﬁbril axis [30]. Numerous recent reports suggest that the toxicity of amyloidogenic proteins lies not in the insoluble ﬁbrils that accumulate but rather in the soluble oligomeric intermediates [16]. Early oligomers have been identiﬁed as the primary toxic species for a number of amyloid diseases and shown to be cytotoxic to cells in culture and in tissue [15, 120]. These soluble oligomers include spherical particles and curvilinear structures, called protoﬁbrils, that appear to represent strings of the spherical particles [30]. An antibody speciﬁcally recognizes soluble oligomers formed by several types of amyloidogenic proteins and peptides, which indicates that they have a common structure and may share a common pathogenic mechanism [17]. Such a mechanism appears to be related to the ability of these oligomers to interact with cell membranes and to form annular protoﬁbrils that cause membrane permeabilization [19, 121, 122]. Disruption of intracellular CaII homeostasis and ROS production are among the earliest biochemical modiﬁcations following the interaction of protoﬁbrillar assemblies with membranes [123]. In the case of extracellular aggregates, cell damage may be accomplished via a nonspeciﬁc pathway (pores) or through oligomer interaction with speciﬁc cell surface receptors, such as glutamatergic receptors involved in CaII inﬂux [124]. Most, if not all, protein inclusions associated with neurodegenerative diseases are enriched with Ub [4]. Lewy bodies in PD, amyloid plaques and neuroﬁbrillary tangles in AD, skeinlike inclusions and Lewy body-like inclusions in ALS, polyglutamine inclusions in Huntington’s disease (HD): all these inclusions show Ub immunoreactivity (Figure 1.4). Ub immunoreactivity is also present in vesicles in some areas of degeneration in AD [3, 125]. The frequent detection of proteasomes and lysosomes around Ub-enriched aggregates in postmortem brains implies that proteins within these inclusions are marked for degradation but not efﬁciently removed. Hence, it has been proposed that neurodegeneration might be linked to degradative dysfunction by several mechanisms, and protein aggregates might arise, at least in part, from impaired proteasomal and/or autophagic removal of the damaged proteins [4, 107]. One prevalent idea is that misfolded proteins resist degradation by, but not engagement with, the proteasome, resulting in proteasome inhibition. In this model, soluble misfolded proteins would be tagged with Ub and partly enter the

13

14

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

Ab

Ubiquitin

Merged

Figure 1.4 Ubiquitin-positive amyloid plaques from AD brain revealed by immunoﬂuorescence. Reprinted by permission from Wiley-Blackwell: Journal of Cellular and Molecular Medicine, copyright 2008 [126].

proteasome, but because of their abnormal structures, would resist full entry and cause steric occlusion [127]. In particular, β-sheet structures, which have been implicated in aberrant conformations in several diseases, seem to be difﬁcult for the proteasome to process when they occur at the site at which misfolding initiates. Moreover, overall proteasome activity in brain tissue decreases with age, and further loss is observed in degenerative conditions such as AD and PD [128]. Importantly, impairment of the UPS is an early response to soluble protein aggregates and not a consequence of inclusion formation [129]. Alternatively, the impairment of degradation might be responsible for the etiology of disease, and aggregate formation might be a secondary phenomenon [4]. The link between the Ub system and neurodegeneration has been strengthened by the identiﬁcation of disease-causing mutations in genes coding for Ub [130] and several ubiquitination enzymes related to AD and PD [131]. The most compelling evidence for the involvement of the UPS in AD pathogenesis comes from a transcriptional misreading, which causes the deletion of two nucleotides in the mRNA coding for Ub [130]. The resulting frameshift mutant form of Ub, called UBB+1 , has a 19-aminoacid extension at the C-terminus and cannot bind to target proteins. However, it can be ubiquitinated by wild-type Ub. The polyubiquitinated UBB+1 cannot be degraded by the proteasome and probably inhibits its activity. Furthermore, it cannot be deubiquitinated [132]. One of the more common types of familial PD is caused by mutations in the gene that encodes the E3 Ub-ligase parkin [133]. Parkin is reported to possess monoubiquitination and K63-polyubiquitination activities [134]. In addition to parkin, an increasing number of E3 genes are now linked to neurodegenerative disorders [1]. Mutations in parkin cause early disease onset, with the loss of dopaminergic neurons in the substantia nigra in the general absence of Lewy bodies. On the other hand, the genetic association of UCHL1 (ubiquitin C-terminal hydrolase L1), a DUB, with familial and sporadic PD is controversial, but UCHL1 is found in Lewy bodies [131].

1.7 Interaction of Ubiquitin with Metal Ions

UCHL1 is a neuron-speciﬁc DUB and one of the most abundant proteins in the brain, which helps to maintain monomeric Ub levels. It possesses DUB activity in its monomeric form and ligase activity in its dimeric form. A mutation linked to familial PD promotes its dimerization and the K63-polyubiquitination of α-syn [135]. The cellular concentrations of the two forms of Ub, free Ub and polyUb chains, are closely interconnected and may change because of various cellular events; for instance heat stress induces an increase in polyUb chains at the expenses of free Ub [136]. The ﬁnding that modest reduction in the level of free Ub in brain is linked to synaptic dysfunction and neuronal degeneration associated with loss-of-function mutations in DUBs, suggests that adequate neuronal Ub supply should be maintained. Decreased Ub availability in neurons of mice is sufﬁcient to cause neuronal dysfunction and death [137]. On the other hand, increased Ub concentrations in CSF of patients affected by neurodegenerative diseases and amyloidosis may have a neuroprotective effect [138]. The presence of Ub within inclusion bodies was noted in a large variety of neurodegenerative diseases [3, 4, 107, 125], suggesting that disruption of Ub homeostasis may be a common factor in the pathogenesis of these disorders [137].

1.7 Interaction of Ubiquitin with Metal Ions 1.7.1 Thermal Stability of Ubiquitin

Ub has been widely used as model for stability, folding, and structural studies, the protein remaining soluble and folded over a wide pH range and at high temperatures [139]. The effect of various cations on the thermal stability of Ub was assessed by differential scanning calorimetry (DSC) experiments carried out on Ub solutions of different metals (CuII , ZnII , NiII , AlIII , and CdII ). CuII , in contrast with other cations, has a speciﬁc negative effect on the thermal stability of Ub, both in terms of unfolding temperature (Tm ) and enthalpy (H) [13]. Far–UV circular dichroism (CD) spectra indicated very little overall change in Ub secondary structure upon CuII binding, however thermal denaturation curves revealed a decrease of Tm from 100 ◦ C for native Ub to 90 ◦ C for the CuII –Ub system. 1.7.2 Spectroscopic Characterization of CuII Binding

Electrospray ionization mass spectrometry (ESI-MS) indicated binding of two CuII ions to the protein with ∼fourfold different afﬁnities. The ﬁrst CuII site has an afﬁnity constant of ∼107 M−1 , as determined from spectrophotometric measurements, and the EPR (electron paramagnetic resonance) parameters (g|| = 2.30 and A|| = 159 × 10−4 cm−1 ) are consistent with a tetragonal N1 O3 (type II)

15

16

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

CuII site [13]. The unpaired electron of a type II CuII site is characterized by relatively long electronic relaxation times (10−8 –10−9 seconds); therefore, its coupling with the nuclear spins has a dramatic effect on the nuclear relaxation and consequently severely affects the NMR linewidths [140]. For most residues in the proximity of the metal center that completely escaped signal identiﬁcation in 1 H-detected heteronuclear single quantum coherence (HSQC) spectra, 13 C resonances could be recovered using tailored NMR experiments based on 13 C direct detection [141, 142]. These latter experiments are intrinsically less affected by paramagnetism-enhanced relaxation than conventional experiments. As the paramagnetic dipolar contributions to nuclear relaxation depend on the square of the gyromagnetic ratio of the observed nucleus, going from 1 H(γH = 2.67 × 108 ) to 13 C(γC = 6.73 × 107 ) detection produces a decrease in relaxation rate by a factor of ∼16[140, 143]. The ‘‘protonless’’ approach was originally applied to a CuII -binding protein involved in Cu trafﬁcking and homeostasis in bacteria [144, 145], and subsequently extended to human Cu,Zn-SOD1 [146, 147]. In the case of Ub, conventional and 13 C-detected experiments allowed the identiﬁcation of resonances for most aminoacid residues with the only exception of those directly coordinated to the Cu ion: the ﬁrst CuII site is located at the N-terminus of the protein and involves the Met1 nitrogen and three oxygen donor ligands (residues 16–18) in a tetragonal arrangement; the second CuII -anchoring site involves the imidazole nitrogen of His68 [13] (Figure 1.5). CuII has excited electronic states of relatively high energy, which produce small orbital contributions to the electronic spin moment [140, 143]. As a consequence, the magnetic susceptibility anisotropy of CuII is generally small. Nonetheless some nuclei of Ub experienced small but measurable pseudocontact shifts (PCSs) arising from dipolar coupling with the unpaired electron [13]. The size and the sign of PCS allowed to deﬁne the orientation of the magnetic susceptibility tensor χ, whose main axis is nearly orthogonal to the tetragonal plane formed by the CuII donor atoms. Such a CuII coordination environment was also supported by the presence

Met1

His68 Lys63 Figure 1.5

The two CuII -binding sites of ubiquitin determined by paramagnetic NMR [13].

1.7 Interaction of Ubiquitin with Metal Ions

of a weak absorption band at 680 nm in the visible spectrum, assignable to the d–d electronic transition [13]. All NMR spectral changes induced by CuII on Ub are abolished upon addition of stoichiometric amounts of ascorbic acid and reduction of CuII to CuI [13, 148]. Therefore, under ‘‘normal’’ conditions, in which intracellular Cu is mainly CuI , Ub does not bind Cu. On the other hand, addition of the tridentate ligand iminodiacetic acid (IDA) mobilizes CuII from the ﬁrst to the second afﬁnity site, in which three coordination positions of CuII are ﬁrmly taken by the IDA ligand, and the protein only provides the fourth donor atom coincident with an easily accessible nitrogen of the His68 imidazole ring [13, 149] (Figure 1.5). Cu reduction or chelation can suppress the CuII -induced protein destabilization. 1.7.3 Possible Implications for the Polyubiquitination Process

Since residues close to the N-terminus are involved in the formation of a β-strand, CuII binding may destabilize the protein starting from its N-terminal region. Potential oxygen donor ligands for the ﬁrst site are Met1 and Val17 carbonyl groups, and Glu16 and Glu18 carboxylate groups [13]. In native Ub, the ﬁrst residue, Met1, is involved in two key H-bonds [108]: one occurring between the amino-terminal group and the CO of Val17 and the other between the side chain sulfur atom of Met1 and the amide NH of Lys63 (Figure 1.5). Therefore, CuII binding to Ub might hamper the proteins’ turnover and other in vivo signaling events regulated by Lys63-polyubiquitination [111] such as the autophagic clearance of protein inclusions [150]. In the polyubiquitination process, Lys63 of the acceptor Ub attacks Gly76 (the C-terminal residue) of the donor Ub forming a Lys63–Gly76 isopeptide bond. The reaction is catalyzed by the Ub-conjugating enzyme Ubc13 and a Ub-ligase. The structural elucidation of Ub covalently bound to Ubc13 has revealed the molecular determinants of Lys63-polyubiquitination [151]. In this structure, several H-bonds help positioning the Lys63 of the acceptor Ub in the active site of Ubc13 near the C-terminus of the donor Ub [151]. On this basis, binding of CuII to the acceptor protein could compromise the correct positioning of Lys63 by affecting the electrostatic interactions at the interface between the conjugating enzyme and the acceptor Ub. 1.7.4 CuII -Induced Self-Oligomerization of Ub

Since conditions that destabilize the native state of a protein render the macromolecular system more prone to aggregation, the interaction of Ub with CuII was hypothesized to be a factor promoting Ub aggregation. Furthermore, CuII ions were found to target the most aggregation-prone regions of Ub (the N-terminus and His68), as predicted by an algorithm based on the propensity of two residues to be found paired in neighboring strands within a β-sheet [152].

17

18

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

CuII

TFE

CuII + TFE

1 µm

1 µm

1 µm

Figure 1.6 AFM images of ubiquitin aggregate morphologies upon long-term incubation with CuII and/or TFE [14].

These predictions were corroborated by an in vitro study of Ub aggregation [14]. It was observed that micromolar concentrations of CuII ions were sufﬁcient to induce self-oligomerization of Ub. Ub oligomers appear as spherical particles ranging from 5 to 25 nm (Figure 1.6), that can progress from dimers to large Ub conglomerates resistant to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). CuII reduction by ascorbic acid or CuII chelation by ethylenediaminetetraacetic acid (EDTA) or IDA, can trigger disruption of Ub oligomers. On the other hand, when the CuII -stabilized oligomers were added to a low-polarity medium, Ub aggregation increased dramatically. 1.7.5 Cooperativity between CuII -Binding and Solvent Polarity

An aqueous solution with a moderate amount (20%, v/v) of 2,2,2-triﬂuoroethanol (TFE) was used to mimic the local decrease of dielectric constant in the proximity of a membrane surface [153]. In the absence of CuII , 20% TFE does not affect Ub secondary structure; the protein forms short beaded chains composed of SDS-sensitive spherical subunits of 6–8 nm (Figure 1.6). In contrast, in the mixture containing both CuII and TFE, the Ub structure shows an increase of β-sheet content, as determined by far-UV CD and attenuated total reﬂectance-Fourier transform infrared (ATR-FTIR) spectroscopy [14]. Atomic force microscopy (AFM) indicates that the aggregation process proceeds through distinct steps characterized by clustering of spherical particles in annular species, formation of trigonal branched structures growing radially from the annular species, and interconnection of these branched structures in ﬁlament networks (Figure 1.6). Large aggregates can be disrupted by CuII reduction yielding a homogeneous population of spherical particles similar to those formed in TFE alone. The CuII -stabilized spherical oligomers of Ub form annular and porelike structures, in liposomes and planar phospholipid bilayers respectively [14] (Figure 1.7). The membrane-penetrating structures are not observed after pretreatment with EDTA, which was shown to destroy the Ub oligomers formed by incubation with CuII . The negativity to tests for amyloid conﬁrmed that the aggregation process of Ub does not lead to ﬁbril formation [14].

1.7 Interaction of Ubiquitin with Metal Ions

19

nm 5 4 3 2 1 0 250

nm 6 4 2 0 600

200 400

600

nm

nm

150 100

250 200

200 400

200 0

150

50 100

nm

nm

50

0

0

0

1 µm

1 µm

Figure 1.7 AFM images of annular and porelike structures formed by CuII -stabilized ubiquitin oligomers in phospholipid bilayers [14].

Kinetic stabilization of α-syn protoﬁbrils, under conditions inhibiting their conversion to mature ﬁbrils, was shown to enhance the harmful effects of aggregation and to accelerate disease progression [15, 154–156]. It was also shown that natural lipids destabilize and rapidly revert inert Aβ amyloid ﬁbrils to soluble neurotoxic protoﬁbrils [157]. Changes in the Aβ peptide structure promoted by metal ions appear to modify the effect of Aβ on lipid membranes. In particular, addition of CuII induces a greater peptide association with lipids and membrane insertion, and also results in an increased β-sheet content for the peptide [158]. 1.7.6 Comparison with Other Metal Ions

CuII appears to play a unique role in Ub destabilization and aggregation. Other metal ions (ZnII , NiII , AlIII , and CdII ) do not affect Ub stability, even at very high metal : protein ratios [13]. The different behavior of various metal ions could be ascribed to differences in their coordination properties. The N-terminus of Ub is the preferred CuII -binding site in pure water as well as in water with 20%

20

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

Zn

Figure 1.8

Packing of ubiquitin molecules in orthorhombic ZnII -ubiquitin crystals [159].

TFE, as determined by NMR spectroscopy [14]. Despite the afﬁnity of CuII for the imidazole nitrogen of histidine, the lack of a preorganized set of donor atoms near His68 renders the binding of CuII to this site less effective than its binding to the N-terminus, where the presence of carboxylate groups (Glu16 and Glu18) helps to accommodate the metal ion in a preorganized anchoring site. In contrast, His68 is the favorite binding site for ZnII , as deduced from the available X-ray structure of the ZnII –Ub adduct [159] (Figure 1.8). The ZnII ion completes its tetrahedral coordination by binding two water molecules and His68 from an adjacent Ub molecule. In the solid state, the adjacent Ub molecule plays a role similar to the IDA ligand in solution experiments, that is, to provide additional donor atoms to the metal ion. All trials to crystallize Ub in the presence of CuII were unsuccessful, even when using very small amounts of metal [159]. This could be a consequence of CuII destabilizing the protein structure and hampering crystallization. The analysis of molecular packing shows that in orthorombic crystals of the ZnII –Ub adduct, the N-terminus of one Ub molecule packs against helix α1 of another molecule. Each of the two Ub molecules, in turn, establishes complementary contacts with a third molecule, thus giving rise to a symmetric trimer stabilized by a series of intermolecular electrostatic interactions (H-bonds and salt bridges) (Figure 1.8). Helix α1 of native Ub contains a large number of unshielded backbone H-bonds, called dehydrons [160], which are generally correlated with membrane association [161] and aggregation propensity of a protein [162]. Mapping on the Zn–Ub structure reveals that dehydrons are clustered in the core of the trimer [159] (shown in black in Figure 1.8). Therefore, missing H-bond protection in the isolated Ub molecules is partially offset by crystal packing. Similarly, protein desolvation associated with a decrease in dielectric constant near a membrane surface is expected to foster intermolecular electrostatic interactions at the trimer interface. From this analysis it is also inferred that CuII binding at the N-terminus may enhance the aggregation propensity of these surface regions of Ub, thus shifting the equilibrium toward the formation of oligomeric species with bridging metal ions.

1.8 Biological Implications

1.8 Biological Implications 1.8.1 The Redox State of Cellular Copper

CuII reduction to CuI offsets Ub destabilization and aggregation [14]. As described above, Cu is generally believed to be transported in the cells by the plasma membrane permease Ctr1 in the +1 oxidation state; it is yet unproven, however, that CuI is the only permeant species [163]. There is unequivocal evidence that Ctr1 is not the only protein capable of mediating Cu entry into mammalian cells, and it is quite possible that CuII rather than CuI is transported by Ctr1-independent mechanisms, based on divalent metal ion inhibition [164, 165]. Consistent with the reducing environment of the cytosol, X-ray absorption spectroscopy indicates the presence of low-coordinate monovalent CuI in this compartment [33, 166, 167]; however, CuII appears to be abundant inside both normal and PD neurons of substantia nigra [168]. That CuII could gain access to special intracellular districts of normal or diseased cells is also supported by the recent identiﬁcation of the Fre6 vacuolar cupric reductase [169]. In neurodegenerative diseases the intravesicular material undergoes ubiquitination [3, 170]. It has been also shown that a portion of α-syn is present in the lumen of intracellular vesicles and can be secreted to the extracellular space, thus possibly contributing to disease propagation [171]. Intravesicular α-syn is more prone to aggregation than its cytosolic counterpart [171] and this could be explained by the microenvironment of the vesicular lumen and increased metal ion concentrations. Preliminary ﬁndings indicate that α-syn and Ub, the two main components of Lewy bodies, form hybrid oligomers in the presence of CuII (Arnesano et al., unpublished observations). Intravesicular aggregates and Lewy bodies can then be propagated from diseased tissues to the cytosol of healthy neurons through prion-like transmission [172]. Importantly, it has been reported that extracellular aggregates including AD amyloids, can be internalized by mammalian cells, gain access to the cytosol, and colocalize with Ub and UPS components [173]. The mechanism of internalization is not clear; however, the recent discovery that PrP mediates Aβ-induced synaptic dysfunction [174], has led to the hypothesis that internalization of PrP may allow Aβ oligomers to reach intracellular compartments and interfere with proteasomal degradation [175] (Figure 1.9). Altogether, these ﬁndings strongly indicate that Ub may come in contact, in the cytosol, with extracellular aggregates enriched with oxidized CuII . Elemental mapping of amyloid deposits of AD brain revealed ‘‘hot spots’’ of accumulated metal ions, particularly Cu and Zn [83, 176]. Moreover, an increment of Cu density was observed in brain tissues that were positively stained for Ub [177], and AD amyloid plaques can be dissolved by CuII chelators [178]. Therefore, the interaction of Ub with CuII could be a pathological event taking place inside intracellular organelles or in the cytosolic compartment of cells which

21

22

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

Ab oligomer

Synapse

PrP coreceptor

PrP Lipid raft Neuron ?

?

Signal transduction

?

Transcription Proteasome

Nucleus

Figure 1.9 Prion and Aβ-induced synaptic toxicity. Reprinted by permission from Macmillan Publishers Ltd: Nature [175], copyright 2009.

attempt to tag and process toxic aggregates. Oxidative stress, membrane breaching, abnormal metal ion homeostasis, and metal miscompartmentalization can foster this process [6, 179]. 1.8.2 Ubiquitin and Phospholipids

Factors that favor protein and metal desolvation (e.g., lower dielectric constant near a membrane surface or in the proximity of inclusion bodies) [180] may signiﬁcantly increase the CuII binding afﬁnity and aggregation propensity of Ub. As an example, the CuII binding afﬁnity of α-syn was shown to increase by 1 order of magnitude in the presence of lipids [181]. Lipids and vesicle membranes were found in Lewy bodies on autopsy [182], and lipid-bound soluble cytosolic oligomers were increased in brain extracts from patients with PD or dementia, with Lewy bodies [183]. The lipid composition of different subcellular compartment membranes can modulate protein folding and aggregation rates. Intracellular accumulation of Aβ was seen predominantly in multivesicular bodies and lysosomes and Aβ ﬁbrillogenesis was found to be accelerated in the presence of endosomal and

1.9 Conclusions and Perspectives

lysosomal membranes. Furthermore, Aβ oligomerization was accelerated through the interaction with ganglioside clusters in lipid rafts [184]. Many physiologically relevant functions of Ub are carried out in the proximity of membrane surfaces. The ubiquitination process regulates endocytosis of membrane proteins, multivesicular body formation, and Golgi and endoplasmic reticulum functions [115]. A membrane-bound form of Ub, phosphatidyl–Ub, has been found in baculovirus particles [185] and in the virions of several other enveloped viruses [186]. The release of Ub from phosphatidyl–Ub, rather than the free cellular pool of Ub, facilitates protein ubiquitination events at membranes [187]. In-cell NMR spectroscopy allows one to determine how protein structures are inﬂuenced by their intracellular environment [188]. The application of this technique to the study of Ub in human cells has revealed an increase of protein dynamics and a decrease of folding stability in vivo as compared to those in vitro, as assessed by hydrogen exchange measurements [189]. This somewhat unexpected behavior has been attributed to nonspeciﬁc interactions of Ub with intracellular macromolecules, the cytoskeleton, and inner membranes [189], which may decrease the folding stability of Ub in cells. Ub exhibits large structural heterogeneity both in the free form and after binding to different partners [190]. Thus, binding events in cells might cause interconversion between different conformations of Ub, which may also destabilize folding and increase hydrogen exchanges rates. Alternatively, some of the substrates may preferentially bind to less folded states of Ub [190].

1.9 Conclusions and Perspectives

We found that Ub, the protein responsible for the cellular clearance of toxic aggregates linked to neurodegenerative disorders, is able to coordinate CuII and form in vitro assemblies similar to those it is supposed to break up [14]. The new ﬁndings underscore the importance of preserving cellular compartmentalization and Cu homeostasis for the correct functioning of protein degradation systems. Ub stability and function might be otherwise compromised [13]. The inorganic chemistry of the brain is inherently rich and remains an open frontier. The results linking Cu trafﬁcking to neurodegenerative disorders are engaging, but much work remains to be done to fully elucidate the molecular roles of bound and labile metal ions, their physiological redox state and speciation, and their contributions to basic aspects of signaling within and between brain cells. Considerable attention is presently focused on the early stages of protein misfolding and aggregation (i.e., protein destabilization and oligomerization), with the aim of elucidating the molecular and biochemical basis of conformational and neurodegenerative disorders. Biophysical studies have been increasingly inﬂuential owing to their ability to provide a mechanistic rationale to better explain the effects of disease-causing mutations, oxidative stress, and the exacerbating role of Cu and other metal ions.

23

24

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders

Of the features discussed above, NMR characterization of paramagnetic CuII proteins represents a challenge, however NMR experiments based on 13 C direct detection open new possibilities for solution and solid-state studies of disease-related CuII -binding proteins and their oligomeric assemblies [140, 143]. The obvious shortcoming of such studies is that they require experimental conditions that have little resemblance to the human brain. Moreover, the pathological processes are difﬁcult to study in vitro, because it is difﬁcult to reconstitute protein aggregation in an artiﬁcial environment. However, the advancements of in-cell NMR techniques [188], coupled to a variety of molecular and elemental imaging techniques including ﬂuorescence and X-ray detection methods [34, 191], will allow detailed investigations at the atomic and microscopic level of globular, amyloid-forming, or intrinsically unstructured proteins in neuronal cells, and will help to characterize functional and pathogenic interactions of metal ions in vivo.

Acknowledgments

I am grateful to Prof. Giovanni Natile for precious suggestions and stimulating discussions. I also thank young collaborators and all the people involved in this research and, in particular, Prof. Ivano Bertini, Prof. Lucia Banci, and Prof. Enrico Rizzarelli. References 1. Tai, H.C. and Schuman, E.M. (2008) 2.

3.

4. 5.

6. 7.

8.

Nat. Rev. Neurosci., 9, 826–838. McClellan, A.J., Tam, S., Kaganovich, D., and Frydman, J. (2005) Nat. Cell Biol., 7, 736–741. Lowe, J., Blanchard, A., Morrell, K., Lennox, G., Reynolds, L., Billett, M., Landon, M., and Mayer, R.J. (1988) J. Pathol., 155, 9–15. Ciechanover, A. and Brundin, P. (2003) Neuron, 40, 427–446. Gaggelli, E., Kozlowski, H., Valensin, D., and Valensin, G. (2006) Chem. Rev., 106, 1995–2044. Barnham, K.J. and Bush, A.I. (2008) Curr. Opin. Chem. Biol., 12, 222–228. Davies, P., Fontaine, S.N., Moualla, D., Wang, X., Wright, J.A., and Brown, D.R. (2008) Biochem. Soc. Trans., 36, 1299–1303. Brown, D.R., Qin, K., Herms, J.W., Madlung, A., Manson, J., Strome, R., Fraser, P.E., Kruck, T.A., Von Bohlen, A., Schulz-Schaeffer, W., Giese, A., Westaway, D., and

9.

10.

11.

12.

13.

14.

Kretzschmar, H.A. (1997) Nature, 390, 684–687. Atwood, C.S., Moir, R.D., Huang, X., Scarpa, R.C., Bacarra, N.M., Romano, D.M., Hartshorn, M.A., Tanzi, R.E., and Bush, A.I. (1998) J. Biol. Chem., 273, 12817–12826. Paik, S.R., Shin, H.J., Lee, J.H., Chang, C.S., and Kim, J. (1999) Biochem. J., 340, 821–828. Morgan, C.J., Gelfand, M., Atreya, C., and Miranker, A.D. (2001) J. Mol. Biol., 309, 339–345. Drago, D., Bolognin, S., and Zatta, P. (2008) Curr. Alzheimer Res., 5, 500–507. Milardi, D., Arnesano, F., Grasso, G., Magri, A., Tabbi, G., Scintilla, S., Natile, G., and Rizzarelli, E. (2007) Angew. Chem. Int. Ed. Engl., 46, 7993–7995. ` V., Arnesano, F., Scintilla, S., Calo, Bonfrate, E., Ingrosso, C., Losacco, M., Pellegrino, T., Rizzarelli, E., and Natile, G. (2009) PLoS ONE, 4, e7052.

References 15. Lashuel, H.A., Hartley, D., Petre, B.M.,

16.

17.

18.

19. 20. 21.

22. 23.

24. 25.

26. 27.

28. 29. 30. 31. 32.

33.

Walz, T., and Lansbury, P.T. (2002) Nature, 418, 291. Bucciantini, M., Giannoni, E., Chiti, F., Baroni, F., Formigli, L., Zurdo, J., Taddei, N., Ramponi, G., Dobson, C.M., and Stefani, M. (2002) Nature, 416, 507–511. Kayed, R., Head, E., Thompson, J.L., McIntire, T.M., Milton, S.C., Cotman, C.W., and Glabe, C.G. (2003) Science, 300, 486–489. Kayed, R., Sokolov, Y., Edmonds, B., McIntire, T.M., Milton, S.C., Hall, J.E., and Glabe, C.G. (2004) J. Biol. Chem., 279, 46363–46366. Lashuel, H.A. (2005) Sci. Aging Knowledge Environ., 2005, e28. Bush, A.I. and Tanzi, R.E. (2008) Neurotherapeutics, 5, 421–432. Bolognin, S., Drago, D., Messori, L., and Zatta, P. (2009) Med. Res. Rev, 29, 547–570. Bush, A.I. (2000) Curr. Opin. Chem. Biol., 4, 184–191. Burdette, S.C. and Lippard, S.J. (2003) Proc. Natl. Acad. Sci. U.S.A, 100, 3605–3610. Thiele, D.J. and Gitlin, J.D. (2008) Nat. Chem. Biol., 4, 145–147. Sigel, A., Sigel, H., and Sigel, R. K. O. (eds) (2006) Neurodegenerative Diseases and Metal Ions: Metal Ions in Life Sciences, John Wiley & Sons, Ltd, Chichester, pp. 1–488. Madsen, E. and Gitlin, J.D. (2007) Annu. Rev. Neurosci., 30, 317–337. Barnham, K.J., Masters, C.L., and Bush, A.I. (2004) Nat. Rev. Drug Discov., 3, 205–214. Finney, L.A. and O’Halloran, T.V. (2003) Science, 300, 931–936. Soto, C. (2003) Nat. Rev. Neurosci., 4, 49–60. Chiti, F. and Dobson, C.M. (2006) Annu. Rev. Biochem., 75, 333–366. Kim, B.E., Nevitt, T., and Thiele, D.J. (2008) Nat. Chem. Biol., 4, 176–185. Smith, R.M. (1983) in Neurobiology of the Trace Elements, Vol. 1 (eds I.E.Dreosti and R.M. Smith), Humana Press, Clifton, pp. 1–40. Yang, L., McRae, R., Henary, M.M., Patel, R., Lai, B., Vogt, S., and Fahrni,

34.

35.

36.

37.

38. 39. 40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

C.J. (2005) Proc. Natl. Acad. Sci. U.S.A, 102, 11179–11184. Que, E.L., Domaille, D.W., and Chang, C.J. (2008) Chem. Rev., 108, 1517–1549. Rae, T., Schmidt, P.J., Pufahl, R.A., Culotta, V.C., and O’Halloran, T.V. (1999) Science, 284, 805–808. O’Halloran, T.V. and Culotta, V.C. (2000) J. Biol. Chem., 275, 25057–25060. Cobine, P.A., Pierrel, F., Bestwick, M.L., and Winge, D.R. (2006) J. Biol. Chem., 281, 36552–36559. Mercer, J.F. and Llanos, R.M. (2003) J. Nutr., 133, 1481S–1484S. Lutsenko, S. and Petris, M.J. (2003) J. Membr. Biol., 191, 1–12. Ohgami, R.S., Campagna, D.R., McDonald, A., and Fleming, M.D. (2006) Blood, 108, 1388–1394. Arnesano, F., Banci, L., Bertini, I., and Cioﬁ-Baffoni, S. (2004) Eur. J. Inorg. Chem., 2004, 1583–1593. Arnesano, F. and Banci, L. (2007) in Handbook of Metalloproteins (ed. A.Messerschmidt), John Wiley & Sons, Ltd. Lee, J., Prohaska, J.R., and Thiele, D.J. (2001) Proc. Natl. Acad. Sci. U.S.A., 98, 6842–6847. Petris, M.J., Smith, K., Lee, J., and Thiele, D.J. (2003) J. Biol. Chem., 278, 9639–9646. De Feo, C.J., Aller, S.G., Siluvai, G.S., Blackburn, N.J., and Unger, V.M. (2009) Proc. Natl. Acad. Sci. U.S.A, 106, 4237–4242. Puig, S., Lee, J., Lau, M., and Thiele, D.J. (2002) J. Biol. Chem., 277, 26021–26030. Guo, Y., Smith, K., Lee, J., Thiele, D.J., and Petris, M.J. (2004) J. Biol. Chem., 279, 17428–17433. Barnham, K.J., McKinstry, W.J., Multhaup, G., Galatis, D., Morton, C.J., Curtain, C.C., Williamson, N.A., White, A.R., Hinds, M.G., Norton, R.S., Beyreuther, K., Masters, C.L., Parker, M.W., and Cappai, R. (2003) J. Biol. Chem., 278, 17401–17407. Brown, D.R. and Kozlowski, H. (2004) Dalton Trans., 1907–1917.

25

26

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders 50. Mendola, D.L., Pietropaolo, A.,

51.

52.

53.

54.

55.

56.

57.

58. 59.

60. 61.

62. 63.

64.

65.

Pappalardo, G., Zannoni, C., and Rizzarelli, E. (2008) Curr. Alzheimer Res., 5, 579–590. Hidalgo, J., Aschner, M., Zatta, P., and Vasak, M. (2001) Brain Res. Bull., 55, 133–145. Arnesano, F., Banci, L., Bertini, I., Cioﬁ-Baffoni, S., Molteni, E., Huffman, D.L., and O’Halloran, T.V. (2002) Genome Res., 12, 255–271. Wernimont, A.K., Huffman, D.L., Lamb, A.L., O’Halloran, T.V., and Rosenzweig, A.C. (2000) Nat. Struct. Biol., 7, 766–771. Arnesano, F., Banci, L., Bertini, I., Cantini, F., Cioﬁ-Baffoni, S., Huffman, D.L., and O’Halloran, T.V. (2001) J. Biol. Chem., 276, 41365–41376. Arnesano, F., Banci, L., Bertini, I., and Bonvin, A.M.J.J. (2004) Structure, 12, 669–676. Banci, L., Bertini, I., Cantini, F., Felli, I.C., Gonnelli, L., Hadjiliadis, N., Pierattelli, R., Rosato, A., and Voulgaris, P. (2006) Nat. Chem. Biol., 2, 367–368. Petris, M.J., Mercer, J.F., Culvenor, J.G., Lockhart, P., and Camakaris, J. (1996) EMBO J., 15, 6084–6095. Schlief, M.L., Craig, A.M., and Gitlin, J.D. (2005) J. Neurosci., 25, 239–246. Schlief, M.L., West, T., Craig, A.M., Holtzman, D.M., and Gitlin, J.D. (2006) Proc. Natl. Acad. Sci. U.S.A, 103, 14919–14924. Schlief, M.L. and Gitlin, J.D. (2006) Mol. Neurobiol., 33, 81–90. Sturtz, L.A., Diekert, K., Jensen, L.T., Lill, R., and Culotta, V.C. (2001) J. Biol. Chem., 276, 38084–38089. Fridovich, I. (1995) Annu. Rev. Biochem., 64, 97–112. Bordo, D., Djinovic, K., and Bolognesi, M. (1994) J. Mol. Biol., 238, 366–386. Arnesano, F., Banci, L., Bertini, I., Martinelli, M., Furukawa, Y., and O’Halloran, T.V. (2004) J. Biol. Chem., 279, 47998–48003. Doucette, P.A., Whitson, L.J., Cao, X., Schirf, V., Demeler, B., Valentine, J.S., Hansen, J.C., and Hart, P.J. (2004) J. Biol. Chem., 279, 54558–54566.

66. Rae, T.D., Torres, A.S., Pufahl, R.A.,

67.

68. 69.

70.

71.

72. 73.

74.

75.

76.

77.

78.

79. 80.

81.

and O’Halloran, T.V. (2001) J. Biol. Chem., 276, 5166–5176. Furukawa, Y., Torres, A.S., and O’Halloran, T.V. (2004) EMBO J., 23, 2872–2881. Kawamata, H. and Manfredi, G. (2008) Hum. Mol. Genet., 17, 3303–3317. Schmidt, P.J., Rae, T.D., Pufahl, R.A., Hamma, T., Strain, J., O’Halloran, T.V., and Culotta, V.C. (1999) J. Biol. Chem., 274, 23719–23725. Beers, J., Glerum, D.M., and Tzagoloff, A. (1997) J. Biol. Chem., 272, 33191–33196. Cobine, P.A., Pierrel, F., and Winge, D.R. (2006) Biochim. Biophys. Acta, 1763, 759–772. Yoshikawa, S. (2002) Adv. Protein Chem., 60, 341–395. Horng, Y.C., Cobine, P.A., Maxﬁeld, A.B., Carr, H.S., and Winge, D.R. (2004) J. Biol. Chem., 279, 35334–35340. Lode, A., Kuschel, M., Paret, C., and Rodel, G. (2000) FEBS Lett, 485, 19–24. Hiser, L., Di Valentin, M., Hamer, A.G., and Hosler, J.P. (2000) J. Biol. Chem., 275, 619–623. Arnesano, F., Balatri, E., Banci, L., Bertini, I., and Winge, D.R. (2005) Structure, 13, 713–722. Banci, L., Bertini, I., Cioﬁ-Baffoni, S., Janicka, A., Martinelli, M., Kozlowski, H., and Palumaa, P. (2008) J. Biol. Chem., 283, 7912–7920. Heaton, D.N., George, G.N., Garrison, G., and Winge, D.R. (2001) Biochemistry, 40, 743–751. Tanner, C.M. (1989) Trends Neurosci., 12, 49–54. Gorell, J.M., Johnson, C.C., Rybicki, B.A., Peterson, E.L., Kortsha, G.X., Brown, G.G., and Richardson, R.J. (1999) Neurotoxicology, 20, 239–247. Zecca, L., Stroppolo, A., Gatti, A., Tampellini, D., Toscani, M., Gallorini, M., Giaveri, G., Arosio, P., Santambrogio, P., Fariello, R.G., Karatekin, E., Kleinman, M.H., Turro, N., Hornykiewicz, O., and Zucca, F.A. (2004) Proc. Natl. Acad. Sci. U.S.A, 101, 9843–9848.

References 82. Multhaup, G., Schlicksupp, A., Hesse,

83.

84.

85. 86. 87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

L., Beher, D., Ruppert, T., Masters, C.L., and Beyreuther, K. (1996) Science, 271, 1406–1409. Lovell, M.A., Robertson, J.D., Teesdale, W.J., Campbell, J.L., and Markesbery, W.R. (1998) J. Neurol. Sci., 158, 47–52. Meloni, G., Sonois, V., Delaine, T., Guilloreau, L., Gillet, A., Teissie, J., Faller, P., and Vasak, M. (2008) Nat. Chem. Biol., 4, 366–372. Dingwall, C. (2007) Biochem. Soc. Trans., 35, 571–573. Chen, D. and Dou, Q.P. (2008) Expert Opin. Ther. Targets, 12, 739–748. Cvek, B., Milacic, V., Taraba, J., and Dou, Q.P. (2008) J. Med. Chem., 51, 6256–6258. Brown, D.R., Haﬁz, F., Glasssmith, L.L., Wong, B.S., Jones, I.M., Clive, C., and Haswell, S.J. (2000) EMBO J., 19, 1180–1186. Kralovicova, S., Fontaine, S.N., Alderton, A., Alderman, J., Ragnarsdottir, K.V., Collins, S.J., and Brown, D.R. (2009) Mol. Cell Neurosci, 41, 135–147. Brown, D.R., Wong, B.S., Haﬁz, F., Clive, C., Haswell, S.J., and Jones, I.M. (1999) Biochem. J., 344, 1–5. Requena, J.R., Groth, D., Legname, G., Stadtman, E.R., Prusiner, S.B., and Levine, R.L. (2001) Proc. Natl. Acad. Sci. U.S.A, 98, 7170–7175. Wakabayashi, K., Tanji, K., Mori, F., and Takahashi, H. (2007) Neuropathology, 27, 494–506. Maroteaux, L., Campanelli, J.T., and Scheller, R.H. (1988) J. Neurosci., 8, 2804–2815. Sharon, R., Goldberg, M.S., Bar-Josef, I., Betensky, R.A., Shen, J., and Selkoe, D.J. (2001) Proc. Natl. Acad. Sci. U.S.A, 98, 9110–9115. Eliezer, D., Kutluay, E., Bussell, R. Jr., and Browne, G. (2001) J. Mol. Biol., 307, 1061–1073. Ulmer, T.S., Bax, A., Cole, N.B., and Nussbaum, R.L. (2004) J. Biol. Chem., 11, 9595–9603. Uversky, V.N., Li, J., and Fink, A.L. (2001) J. Biol. Chem., 276, 44284–44296.

98. Rasia, R.M., Bertoncini, C.W.,

99.

100.

101.

102.

103.

104.

105.

106. 107. 108. 109.

Marsh, D., Hoyer, W., Cherny, D., Zweckstetter, M., Griesinger, C., Jovin, T.M., and Fern´andez, C.O. (2005) Proc. Natl. Acad. Sci. U.S.A., 102, 4294–4299. Paik, S.R., Shin, H.J., and Lee, J.H. (2000) Arch. Biochem. Biophys., 378, 269–277. Pall, H.S., Williams, A.C., Blake, D.R., Lunec, J., Gutteridge, J.M., Hall, M., and Taylor, A. (1987) Lancet, 2, 238–241. Rosen, D.R., Siddique, T., Patterson, D., Figlewicz, D.A., Sapp, P., Hentati, A., Donaldson, D., Goto, J., O’Regan, J., Deng, H.X., Rahmani, Z., Krizus, A., McKenna-Yasek, D., Cayabyab, A., Gatson, S.M., Berger, R., Tanzi, R.E., Halperin, J.J., Herzfeldt, B., Van der Bergh, R., Hung, W.-Y., Bird, Deng, G., Mulder, D.W., Smyth, C., Laing, N.G., Soriano, E., Pericak-Vance, M.A., Haines, J., Rouleau, G.A., Gusella, J.S., Horvitz, H.R., Brown, R.H. Jr. (1993) Nature, 362, 59–62. Valentine, J.S., Doucette, P.A., and Potter, S.Z. (2005) Annu. Rev. Biochem., 74, 563–593. Deng, H.X., Shi, Y., Furukawa, Y., Zhai, H., Fu, R., Liu, E., Gorrie, G.H., Khan, M.S., Hung, W.Y., Bigio, E.H., Lukas, T., Dal Canto, M.C., O’Halloran, T.V., and Siddique, T. (2006) Proc. Natl. Acad. Sci. U.S.A, 103, 7142–7147. Banci, L., Bertini, I., Durazo, A., Girotto, S., Gralla, E.B., Martinelli, M., Valentine, J.S., Vieru, M., and Whitelegge, J.P. (2007) Proc. Natl. Acad. Sci. U.S.A, 104, 11263–11267. Furukawa, Y., Kaneko, K., Yamanaka, K., O’Halloran, T.V., and Nukina, N. (2008) J. Biol. Chem., 283, 24167–24176. DiAntonio, A. and Hicke, L. (2004) Annu. Rev. Neurosci., 27, 223–246. Ross, C.A. and Pickart, C.M. (2004) Trends Cell Biol., 14, 703–711. Vijay-Kumar, S., Bugg, C.E., and Cook, W.J. (1987) J. Mol. Biol., 194, 531–544. Pickart, C.M. and Eddins, M.J. (2004) Biochim. Biophys. Acta, 1695, 55–72.

27

28

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders 110. Hershko, A. and Ciechanover, A.

111. 112.

113. 114.

115. 116. 117. 118.

119.

120. 121.

122. 123.

124.

125. 126. 127. 128.

(1998) Annu. Rev. Biochem., 67, 425–479. Pickart, C.M. and Fushman, D. (2004) Curr. Opin. Chem. Biol., 8, 610–616. Verma, R., Aravind, L., Oania, R., McDonald, W.H., Yates, J.R. III, Koonin, E.V., and Deshaies, R.J. (2002) Science, 298, 611–615. Ambroggio, X.I., Rees, D.C., and Deshaies, R.J. (2004) PLoS Biol., 2, E2. Luzio, J.P., Pryor, P.R., and Bright, N.A. (2007) Nat. Rev. Mol. Cell Biol., 8, 622–632. Hicke, L. and Dunn, R. (2003) Annu. Rev. Cell Dev. Biol., 19, 141–172. Piper, R.C. and Luzio, J.P. (2007) Curr. Opin. Cell Biol., 19, 459–465. Klionsky, D.J. and Emr, S.D. (2000) Science, 290, 1717–1721. Todde, V., Veenhuis, M., and van der Klei, I. (2009) Biochim. Biophys. Acta, 1792, 3–13. Kaganovich, D., Kopito, R., and Frydman, J. (2008) Nature, 454, 1088–1095. Haass, C. and Selkoe, D.J. (2007) Nat. Rev. Mol. Cell Biol., 8, 101–112. Quist, A., Doudevski, I., Lin, H., Azimova, R., Ng, D., Frangione, B., Kagan, B., Ghiso, J., and Lal, R. (2005) Proc. Natl. Acad. Sci. U.S.A., 102, 10427–10432. Lashuel, H.A. and Lansbury, P.T. Jr. (2006) Q. Rev. Biophys., 39, 167–201. Demuro, A., Mina, E., Kayed, R., Milton, S.C., Parker, I., and Glabe, C.G. (2005) J. Biol. Chem., 280, 17294–17300. Pellistri, F., Bucciantini, M., Relini, A., Nosi, D., Gliozzi, A., Robello, M., and Stefani, M. (2008) J. Biol. Chem., 283, 29950–29960. Lowe, J., Hand, N., and Mayer, R.J. (2005) Methods Enzymol., 399, 86–119. Oddo, S. (2008) J. Cell Mol. Med., 12, 363–373. Bence, N.F., Sampat, R.M., and Kopito, R.R. (2001) Science, 292, 1552–1555. Gray, D.A., Tsirigotis, M., and Woulfe, J. (2003) Sci. Aging Knowledge Environ., 2003, RE6.

129. Bennett, E.J., Bence, N.F.,

130.

131.

132.

133.

134. 135.

136. 137.

138. 139. 140. 141.

142.

Jayakumar, R., and Kopito, R.R. (2005) Mol. Cell, 17, 351–365. van Leeuwen, F.W., de Kleijn, D.P., van den Hurk, H.H., Neubauer, A., Sonnemans, M.A., Sluijs, J.A., Koycu, S., Ramdjielal, R.D., Salehi, A., Martens, G.J., Grosveld, F.G., Peter, J., Burbach, H., and Hol, E.M. (1998) Science, 279, 242–247. Leroy, E., Boyer, R., Auburger, G., Leube, B., Ulm, G., Mezey, E., Harta, G., Brownstein, M.J., Jonnalagada, S., Chernova, T., Dehejia, A., Lavedan, C., Gasser, T., Steinbach, P.J., Wilkinson, K.D., and Polymeropoulos, M.H. (1998) Nature, 395, 451–452. Lindsten, K., de Vrij, F.M., Verhoef, L.G., Fischer, D.F., van Leeuwen, F.W., Hol, E.M., Masucci, M.G., and Dantuma, N.P. (2002) J. Cell Biol., 157, 417–427. Shimura, H., Hattori, N., Kubo, S., Mizuno, Y., Asakawa, S., Minoshima, S., Shimizu, N., Iwai, K., Chiba, T., Tanaka, K., and Suzuki, T. (2000) Nat. Genet., 25, 302–305. Moore, D.J. (2006) Biochem. Soc. Trans., 34, 749–753. Liu, Y., Fallon, L., Lashuel, H.A., Liu, Z., and Lansbury, P.T. Jr. (2002) Cell, 111, 209–218. Parag, H.A., Raboy, B., and Kulka, R.G. (1987) EMBO J., 6, 55–61. Ryu, K.Y., Garza, J.C., Lu, X.Y., Barsh, G.S., and Kopito, R.R. (2008) Proc. Natl. Acad. Sci. U.S.A, 105, 4016–4021. Sixt, S.U. and Dahlmann, B. (2008) Biochim. Biophys. Acta, 1782, 817–823. Jackson, S.E. (2006) Org. Biomol. Chem., 4, 1845–1853. Arnesano, F., Banci, L., and Piccioli, M. (2005) Q. Rev. Biophys., 38, 167–219. Bermel, W., Bertini, I., Duma, L., Emsley, L., Felli, I.C., Pierattelli, R., and Vasos, P.R. (2005) Angew. Chem. Int. Ed., 44, 3089–3092. Bermel, W., Bertini, I., Felli, I.C., Piccioli, M., and Pierattelli, R. (2006) Prog. NMR Spectrosc., 48, 25–45.

References 143. Bertini, I., Luchinat, C., Parigi, G.,

144.

145.

146.

147.

148. 149.

150.

151.

152.

153.

154.

155.

156.

157.

and Pierattelli, R. (2008) Dalton Trans., 3782–3790. Arnesano, F., Banci, L., Bertini, I., Felli, I.C., Luchinat, C., and Thompsett, A.R. (2003) J. Am. Chem. Soc., 125, 7200–7208. Arnesano, F., Banci, L., Bertini, I., Mangani, S., and Thompsett, A.R. (2003) Proc. Natl. Acad. Sci. U.S.A., 100, 3814–3819. Bermel, W., Bertini, I., Felli, I.C., K¨ummerle, R., and Pierattelli, R. (2003) J. Am. Chem. Soc., 125, 16423–16429. Bertini, I., Felli, I.C., Luchinat, C., Parigi, G., and Pierattelli, R. (2007) Chembiochem, 8, 1422–1429. Rice, M.E. (2000) Trends Neurosci., 23, 209–216. Nomura, M., Kobayashi, T., Kohno, T., Fujiwara, K., Tenno, T., Shirakawa, M., Ishizaki, I., Yamamoto, K., Matsuyama, T., Mishima, M., and Kojima, C. (2004) FEBS Lett., 566, 157–161. Tan, J.M., Wong, E.S., Kirkpatrick, D.S., Pletnikova, O., Ko, H.S., Tay, S.P., Ho, M.W., Troncoso, J., Gygi, S.P., Lee, M.K., Dawson, V.L., Dawson, T.M., and Lim, K.L. (2008) Hum. Mol. Genet., 17, 431–439. Eddins, M.J., Carlile, C.M., Gomez, K.M., Pickart, C.M., and Wolberger, C. (2006) Nat. Struct. Mol. Biol., 13, 915–920. Trovato, A., Chiti, F., Maritan, A., and Seno, F. (2006) PLoS. Comput. Biol., 2, e170 . Munishkina, L.A., Phelan, C., Uversky, V.N., and Fink, A.L. (2003) Biochemistry, 42, 2720–2730. Conway, K.A., Lee, S.J., Rochet, J.C., Ding, T.T., Williamson, R.E., and Lansbury, P.T.Jr. (2000) Proc. Natl. Acad. Sci. U.S.A., 97, 571–576. Conway, K.A., Rochet, J.C., Bieganski, R.M., and Lansbury, P.T.Jr. (2001) Science, 294, 1346–1349. Herrera, F.E., Chesi, A., Paleologou, K.E., Schmid, A., Munoz, A., Vendruscolo, M., Gustincich, S., Lashuel, H.A., and Carloni, P. (2008) PLoS ONE, 3, e3394 . Martins, I.C., Kuperstein, I., Wilkinson, H., Maes, E.,

158.

159.

160. 161.

162.

163.

164.

165.

166.

167.

168.

169. 170.

171.

Vanbrabant, M., Jonckheere, W., Van, G.P., Hartmann, D., D’Hooge, R., De, S.B., Schymkowitz, J., and Rousseau, F. (2008) EMBO J., 27, 224–233. Lau, T.L., Ambroggio, E.E., Tew, D.J., Cappai, R., Masters, C.L., Fidelio, G.D., Barnham, K.J., and Separovic, F. (2006) J. Mol. Biol., 356, 759–770. Falini, G., Fermani, S., Tosi, G., Arnesano, F., and Natile, G. (2008) Chem.Commun.(Camb.), 5960–5962. Fernandez, A., Chen, J., and Crespo, A. (2007) J. Chem. Phys., 126, 245103. Fernandez, A. and Berry, R.S. (2003) Proc. Natl. Acad. Sci. U.S.A, 100, 2391–2396. Fernandez, A., Kardos, J., Scott, L.R., Goto, Y., and Berry, R.S. (2003) Proc. Natl. Acad. Sci. U.S.A, 100, 6446–6451. Maryon, E.B., Molloy, S.A., Zimnicka, A.M., and Kaplan, J.H. (2007) Biometals, 20, 355–364. Lee, J., Petris, M.J., and Thiele, D.J. (2002) J. Biol. Chem., 277, 40253–40259. Moriya, M., Ho, Y.H., Grana, A., Nguyen, L., Alvarez, A., Jamil, R., Ackland, M.L., Michalczyk, A., Hamer, P., Ramos, D., Kim, S., Mercer, J.F., and Linder, M.C. (2008) Am. J. Physiol. Cell Physiol., 295, C708–C721. Pufahl, R.A., Singer, C.P., Peariso, K.L., Lin, S.-J., Schmidt, P.J., Fahrni, C.J., Cizewski Culotta, V., Penner-Hahn, J.E., and O’Halloran, T.V. (1997) Science, 278, 853–856. Xiao, Z., Loughlin, F., George, G.N., Howlett, G.J., and Wedd, A.G. (2004) J. Am. Chem. Soc., 126, 3081–3090. Chwiej, J., Adamek, D., Szczerbowska-Boruchowska, M., Krygowska-Wajs, A., Bohic, S., and Lankosz, M. (2008) J. Trace Elem. Med. Biol., 22, 183–188. Rees, E.M. and Thiele, D.J. (2007) J. Biol. Chem., 282, 21629–21638. Doherty, F.J., Osborn, N.U., Wassell, J.A., Heggie, P.E., Laszlo, L., and Mayer, R.J. (1989) Biochem. J., 263, 47–55. Lee, H.J., Patel, S., and Lee, S.J. (2005) J. Neurosci., 25, 6016–6024.

29

30

1 The Role of Copper Ion and the Ubiquitin System in Neurodegenerative Disorders 172. Li, J.Y., Englund, E., Holton, J.L.,

173.

174.

175. 176.

177. 178.

179.

180.

181.

Soulet, D., Hagell, P., Lees, A.J., Lashley, T., Quinn, N.P., Rehncrona, S., Bjorklund, A., Widner, H., Revesz, T., Lindvall, O., and Brundin, P. (2008) Nat. Med., 14, 501–503. Ren, P.H., Lauckner, J.E., Kachirskaia, I., Heuser, J.E., Melki, R., and Kopito, R.R. (2009) Nat. Cell Biol., 11, 219–225. Lauren, J., Gimbel, D.A., Nygaard, H.B., Gilbert, J.W., and Strittmatter, S.M. (2009) Nature, 457, 1128–1132. Cisse, M. and Mucke, L. (2009) Nature, 457, 1090–1091. Miller, L.M., Wang, Q., Telivala, T.P., Smith, R.J., Lanzirotti, A., and Miklossy, J. (2006) J. Struct. Biol., 155, 30–37. Ide-Ektessabi, A. and Rabionet, M. (2005) Anal. Sci., 21, 885–892. Cherny, R.A., Legg, J.T., McLean, C.A., Fairlie, D.P., Huang, X., Atwood, C.S., Beyreuther, K., Tanzi, R.E., Masters, C.L., and Bush, A.I. (1999) J. Biol. Chem., 274, 23223–23228. Donnelly, P.S., Xiao, Z., and Wedd, A.G. (2007) Curr. Opin. Chem. Biol., 11, 128–133. Liang, X., Campopiano, D.J., and Sadler, P.J. (2007) Chem. Soc. Rev., 36, 968–992. Lee, E.N., Lee, S.Y., Lee, D., Kim, J., and Paik, S.R. (2003) J. Neurochem., 84, 1128–1142.

182. Gai, W.P., Yuan, H.X., Li, X.Q.,

183.

184.

185. 186.

187.

188. 189.

190.

191.

Power, J.T., Blumbergs, P.C., and Jensen, P.H. (2000) Exp. Neurol., 166, 324–333. Sharon, R., Bar-Joseph, I., Frosch, M.P., Walsh, D.M., Hamilton, J.A., and Selkoe, D.J. (2003) Neuron, 37, 583–595. LaFerla, F.M., Green, K.N., and Oddo, S. (2007) Nat. Rev. Neurosci., 8, 499–509. Guarino, L.A., Smith, G., and Dong, W. (1995) Cell, 80, 301–309. Webb, J.H., Mayer, R.J., and Dixon, L.K. (1999) FEBS Lett., 444, 136–139. Welchman, R.L., Gordon, C., and Mayer, R.J. (2005) Nat. Rev. Mol. Cell Biol., 6, 599–609. Burz, D.S. and Shekhtman, A. (2009) Nature, 458, 37–38. Inomata, K., Ohno, A., Tochio, H., Isogai, S., Tenno, T., Nakase, I., Takeuchi, T., Futaki, S., Ito, Y., Hiroaki, H., and Shirakawa, M. (2009) Nature, 458, 106–109. Lange, O.F., Lakomek, N.A., Fares, C., Schroder, G.F., Walter, K.F., Becker, S., Meiler, J., Grubmuller, H., Griesinger, C., and de Groot, B.L. (2008) Science, 320, 1471–1475. Fahrni, C.J. (2007) Curr. Opin. Chem. Biol., 11, 121–127.

31

2 The Bioinorganic and Organometallic Chemistry of Copper(III) Xavi Ribas and Alicia Casitas

2.1 Introduction

Copper in oxidation state +3 is known in a limited number of compounds. In view of the electronic conﬁguration of CuIII ion, 3d8 , the ligand-ﬁeld theory predicts an enhanced stabilization in a square-planar type of geometry [1], and, indeed, this is the common geometry found in the vast majority of CuIII structures reported. Also, pairing of all d8 electrons affords diamagnetic species. Some of them are stable only in the solid state, such as NaCuIII O2 [2, 3], CuIII -bis-biuret [4], and CuIII -bis-oxamide [5], but decompose rapidly in solution; the generation of CuIII in water or acetonitrile has been achieved by means of pulse radiolytic [6, 7] or electrochemical experiments in the presence of macrocyclic amines [8, 9], but their reduction to CuII is usually very rapid. Diperiodato and ditellurato cuprates(III) have been generated in solution and studied as oxidizing agents of organic substrates [10]. However, it was not until the discovery by Margerum et al. in 1975 of CuIII –peptides complexes, stable in aqueous solutions and room temperature, that it boosted the interest in CuIII and its implications in biological systems. Examples of these are the tetrapeptide CuIII –glycine [11] and tri-α-aminoisobutyric acid tripeptide CuIII complex, [CuIII (H-3 Aib3 )] [12]. Since then, several other CuIII species have been stabilized at room temperature and in solution by using anionic ligands [13, 14]. For example, ligands containing two deprotonated oxime groups in combination with two imine N atoms stabilize CuIII complexes [15], as well as tetraanionic bis-oximato chelating ligands [16], allowing the characterization of their CuII /CuIII redox properties. Highly stable CuIII complexes with N-amido and alkoxo polyanionic chelating ligands were reported by Anson et al. [17] and studied their stability alteration by modifying the donor properties of ligands. More recently, Hanss et al. reported CuIII complexes bearing deprotonated hydroxyiminoamide ligands [18]. The comparison of CuIII /CuII redox potentials among several CuIII species in water and CH3 CN revealed the trend of enhanced stabilization of CuIII oxidation state by increasing the number of anionic donor atoms [19], and by introducing thiolate donor groups (Figure 2.1).

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

32

2 The Bioinorganic and Organometallic Chemistry of Copper(III) − O

O

N

N CuIII R1

R1

O O

R1=

O

O−

Me

E1/2 (CuIII/CuII) vs. SCE (mV)

N−

O

0.41

0.02

H3C H3C −0.51

O−

H3C H3C

S−

−0.78

N N H2

N CuIII O

0.10

O

O N O

N

CuIII N O OH 0.26

N

Figure 2.1 Redox potentials for well-established CuIII species bearing disubstituted o-phenylene-bis(amide) ligands (four on the left) [16], one tri-α-aminoisobutyric acid tripeptide ligand [12], and one bis(hydroxyiminoamide) ligand (on the right) [18].

Besides square-planar CuIII compounds, there are several examples reporting weak axial coordination yielding square-pyramidal geometry, and only two compounds have been reported exhibiting octahedral (Oh) geometries: the well-known paramagnetic K3 [CuIII F6 ] described in text books [1] and a more recent heterometallic compound, [LCoIII CuIII CoIII L](ClO4 )3 , with a CuIII S6 coordination environment [20]. In the latter case, the Oh CuIII species is no longer diamagnetic and the high spin paramagnetic (S = 1) species is characterized by EPR (electron paramagnetic resonance) spectroscopy. Additionally, Santo et al. have also shown a diamagnetic-to-paramagnetic conversion of a tris(2-pyridylthio)methylcopper(III) through a geometry change from trigonal bypiramidal to Oh upon addition of chloride [21]. The direct assignment of the oxidation state of the copper ion is not straightforward in many complexes. Transition metal complexes containing dithiolene ligands or bidentate ortho-disubstituted aromatic ligands bearing O, N, S, or Se as coordinating atoms [22] show a rich electrochemical behavior that opens the question whether the redox processes are ligand- or metal centered. Experimental and theoretical evidence has been reported on dearomatization of the benzene rings to give an open-shell o-semiquinonate type ligand, and the singlet diradical ground states of some neutral complexes have been characterized [23, 24]. Thus, the ligand role upon oxidation of the complex, usually referred as innocent or noninnocent character of the ligands, is still under discussion. A number of dithiolene- and diselenolene-type copper complexes in the formal oxidation state +3 are known [25–28], and several works discussing the real oxidation state of the metal center have been reported [26, 29]. In any case, the existence of an experimental technique to prove the oxidation state of copper was mandatory. DuBois et al. used synchrotron light source to set up the Cu K-edge X-ray absorption spectroscopy (XAS) technique in 2000, which provided a clear means of identifying CuIII [30]. They studied a varied number of CuIII species, from Margerum’s CuIII –peptide compounds to bis(µ-oxo)CuIII 2 species with peralkylated amine ligands; the distinct feature of CuIII is a 1 s → 3d transition centered on 8981 eV, approximately 2 eV

2.2 Bioinorganic Implications of Copper(III)

higher than the characteristic energy of this transition for CuII oxidation state. In mixtures with CuII , CuIII is identiﬁable if present in at least a 25%.

2.2 Bioinorganic Implications of Copper(III)

Copper versatility is reﬂected in the high diversity of biological functions that copper-dependent enzymes are capable to catalyze, and in the diverse number of Cu ions contained in each metalloenzyme active site [31]. Copper can be present as mononuclear, dinuclear, or trinuclear complexes, sharing the active site with other metals such as Fe and Zn or forming polymetallic aggregates. Among the different catalyzation processes, O2 activation and electron transfer processes are the most widely found. Enzymes eliciting monooxygenation (tyrosinase, Tyr; dopamine-β-monooxygenase, DβM; peptidylglycine α-hydroxylating monooxygenase, PHM; particulated metano monooxygenase, pMMO), dioxygenation (quercitinase and indole dioxygenase, IDO), and oxidation processes (galactose oxidase, GO; catechol oxidase, CO; laccase, Lc; cytochrome c oxidase, CcO) are widely spread in a large number of microorganisms, arthropods, and superior organisms including mammals. Electron transfer processes are catalyzed by mononuclear blue copper proteins (azurin and plastocyanin, Pc) found in plants. The participation of copper(III) species in processes catalyzed by mono-, di,and polynuclear copper metalloproteins has been often invoked [32]; despite this it has not been directly detected in natural systems so far. Research in bioinorganic chemistry and the development of low-molecular weight chemical models have, however, allowed the characterization of CuIII species that may be relevant to natural systems. It was actually after Margerum’s discovery of stable CuIII –peptide compounds in aqueous solutions that the scientiﬁc community began to think seriously in the possible biological implication of CuIII . Hereafter, we will brieﬂy describe the most relevant biological processes where CuIII might have an important role. 2.2.1 Dinuclear Type-3 Copper Enzymes

Type-3 copper enzymes, namely hemocyanin (Hc), Tyr, and CO, are capable of binding molecular O2 to form a µ-η2 : η2 -(O2 )-CuII 2 side-on peroxo species as the enzymatic oxidized form. Hc function is exclusively limited to O2 transport in arthropods and mollusks [33], whereas Tyr is usually found in bacteria, fungi, plants, and mammals, and is involved in the synthesis of melanins. Speciﬁcally, it catalyzes the hydroxylation of monophenols to obtain ortho-diphenols (catechols) in a ﬁrst step, and reoxidizes the latter to o-quinones in a second step [34]. In dinuclear copper(I)-reduced deoxy forms, each metal center is coordinated to three histidinic-N residues, and reaction with O2 in a cooperative manner allows for the

33

34

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

NHis NHis

CuII NHis

NHis O CuII N His O NHis

O2 transport (Hemocyanin)

NHis NHis

OH

CuI

CuI

NHis

NHis

NHis NHis

+ O2

R Phenolase reactivity (Tyrosinase)

R NHis NHis

O CuII NHis

O CuII O

NHis

OH OH

R

NHis

NHis

Figure 2.2 Distinct functionality of µ-η2 : η2 -(O2 )-CuII 2 side-on peroxo species in hemocyanin and tyrosinase. An electrophilic attack of the peroxo core to the phenolic substrate is proposed to account for tyrosinase reactivity [35].

stabilization of the peroxo bridging group in a side-on geometry fashion for the oxidized form of both enzymes (Figure 2.2). Spectroscopic characterization of bioinorganic synthetic models of Hc and Tyr active sites anticipated the identical oxidized forms for both enzymes [34, 36–40], before X-ray structural data were recently available [41, 42]. The difference in reactivity is likely related to the accessibility of phenolic substrates to the active site, which is modulated by the tertiary structure of the polypeptidic chain. The latter has been also suggested by partial denaturalization of hemocyanin with urea, resulting in an active enzyme for phenol hydroxylation [43, 44]. The active species responsible for the hydroxylation of phenol in Tyr is thought to be the well-characterized µ-η2 : η2 -(O2 )-CuII 2 side-on peroxo species, performing an electrophilic attack on the π-system of the phenol aromatic ring (Figure 2.2) [35, 45, 46]. Several synthetic compounds modeling the side-on peroxo-CuII 2 species support its capability to perform an electrophilic σ ∗ attack to a properly oriented aromatic ring [47], either in intramolecular ligand hydroxylations [48–53] or in intermolecular hydroxylation of exogenous phenolic substrates[54–58]. However, the real active species in Tyr was initially challenged in 1995 by Tolman and coworkers, who ﬁrst described the existence of a bis(µ-oxo)CuIII 2 core in a synthetic model [59] and shortly after reported the coexistence of µ-η2 : η2 -(O2 )-CuII 2 side-on peroxo species and bis(µ-oxo)CuIII 2 isomers in rapid equilibrium in solution at very low temperatures (Figure 2.3a) [60, 61]. The bis(µ-oxo)CuIII 2 core represented a step forward in the comprehension of the Cu2 –O2 interaction mechanism because the O–O bond is already broken. Since then, and considering the facile interconversion between the two isomeric species by changing solvent or counteranions [61–65], several groups have pushed the idea of the bis(µ-oxo)CuIII 2 core capability of performing aromatic hydroxylation, and thus proposing its implication in the reactivity of Tyr [66–68]. The most representative work in this line was published by Mirica et al. in 2005 [67], when

2.2 Bioinorganic Implications of Copper(III)

CuII

O

35

O CuII

CuIII

O

CuIII O

3.6 Å

2.8 Å

(a) +

2+ X N

N O

N

CuIII

CuIII N O

N

N

ONa N

X = Cl, F, CO2Me, CN Acetone −90 °C

OH

N

Warming up

O N

CuIII O N

(b)

N

CuIII O

N

X

Figure 2.3 (a) Equilibrium between isomeric µ-η2 : η2 -(O2 )-CuII 2 side-on peroxo and bis(µ-oxo)CuIII 2 cores and (b) an example of exogenous phenolate hydroxylation by a bis(µ-oxo)CuIII 2 core [68].

they reported a dinuclear synthetic copper(I) complex bearing peralkylated diamine ligands that bind O2 to form a side-on peroxo species at −120 ◦ C, which evolutes to a bis(µ-oxo)CuIII 2 core upon coordination of an exogenous 2,4-di-tert-butylphenolate. Finally, the phenolic substrate is hydroxylated by the bis(µ-oxo)CuIII 2 core to afford quantitatively a mixture of the corresponding catechol and quinone. Theoretical analysis suggested an electrophilic σ ∗ attack of bis(µ-oxo)CuIII 2 core to the phenol ring. Company et al. have also made an important contribution substantiating the capability of a perfectly established bis(µ-oxo)dicopper(III) species to perform Tyr-like reactivity [68, 69]. In this synthetic model, the bis(µ-oxo)CuIII 2 core rapidly binds 4-chlorophenolate affording a new metastable intermediate species at −90 ◦ C, [CuIII 2 (µ-O)2 (p-Cl-C6 H4 O)(m-XYLMeAN )]+ , which decomposes toward the hydroxylation of the phenolate moiety to afford the corresponding 4-chlorocatechol (Figure 2.3b). Stopped-ﬂow UV–Vis kinetic monitoring allowed to determine an electrophilic type of reactivity for a series of para-substituted phenolates, thus in line with Stack’s observations [67]. Our work contribution is centered in the fact that O–O bond cleavage can occur not only prior to C–O bond formation but also prior to phenol binding, further substantiating the idea that the bis(µ-oxo) core is competent for performing enzyme-like activity. Very recently, Herres-Pawlis et al. have reported another bis(µ-oxo)dicopper(III) species based on a permethylated-amine-guanidine ligand performing 2,4-di-tert-butylphenolate hydroxylation to 3,5-di-tert-butylcatecholate [70]. The fact that both isomeric Cu2 O2 cores are capable of performing the electrophilic aromatic hydroxylation of arene rings (either of phenolate-type or of intramolecular xylyl linkers), along with the known facile interconversion between both isomers depending on experimental conditions, suggests that either one may

X

OH

36

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

be the ﬁnal core responsible for the aromatic hydroxylation. But, most importantly, what is really crucial is that the aromatic substrate can approach in intimate proximity and correct orientation to the Cu2 O2 core, and here seems to rely on the difference in reactivity between Tyr and hemocyanin. It has to be indicated, however, that a very recent report from Casella and coworkers rules out the existence of the bis(µ-oxo)CuIII 2 core in the natural Tyr [71], as previously suggested theoretically by Siegbahn et al. [72]. A complete discussion on reactivity of isomeric Cu2 O2 cores is found in Dr. Company’s chapter of this book is Chapter 11. 2.2.2 Particulate Methano Monooxygenase (pMMO)

Methane oxidation to methanol is performed by methanotrophic bacteria using molecular oxygen [73]. These bacteria reside at the boundary of aerobic and anaerobic environments, where CH4 and O2 are available. Two different enzymes with the same function are known to be expressed in methanotrophic bacteria such as Methylococcus capsulatus: a cytoplasmatic soluble methane monooxygenase (sMMO) and a membrane-bound known as particulate-methane monooxygenase (pMMO). The crystal structure of sMMO is known since 1993 [74], and establishes the existence of a dinuclear iron center that is responsible for the activation of O2 and reaction with CH4 to produce CH3 OH. On the other hand, it was known that pMMO contained Cu ions in its active site by means of spectroscopic studies [75], but unraveling its structure has been one of the most intriguing and exciting topics in bioinorganic chemistry. Very recently, Lieberman et al. overcame the difﬁculties of purifying and crystallizing a membrane-bound enzyme, and reported the crystal structure of pMMO from M. capsulatus at a resolution of 2.8 A˚ [76]. This structure showed the existence of three different subunits, with three metal centers: one mononuclear copper site, a dinuclear copper site 21 A˚ apart, and a mononuclear Zn site 19 A˚ apart from the dicopper site and 32 A˚ from the monocopper site (Figure 2.4a). The zinc site is considered to be occupied by Cu or Fe since its presence was not spectroscopically anticipated and pMMO was crystallized using a Zn buffer. Therefore, in sharp contrast to the iron mediated methane conversion to methanol in sMMO, the pMMO appears to be a copper-dependent enzyme that also hydroxylates CH4 . It has been intensively discussed in the literature whether monocopper or dicopper sites (or both) can perform O2 activation and subsequent methane oxidation to methanol [77–79]. Several highly oxidized copper-oxo species have been invoked to play a role in pMMO activity, and among them, mononuclear [CuIII =O]+ species [77] and dinuclear bis(µ-oxo)CuII CuIII core [77, 80] appear as the most reliable active species that can elicit the hydroxylation of methane (Figure 2.4b). Yoshisawa et al. evaluated computationally the formation of [CuIII =O]+ and bis (µ-oxo)CuII CuIII species by reproducing the biologic coordination environment certiﬁed by the X-ray structure [77]. According to their DFT (density functional

2.2 Bioinorganic Implications of Copper(III)

His 48

HN N

HN

N

Cu

Cu

N

NH

His 33

NH

C NH2 Gln 404

HN

N

HN His 137

NH

CuI

His 139

NH

O

HN

His 137 N

N

His 173

O N H

HN

His 33

His 33

HN His 137

NH N

CuIII O His 139 CuIII N NH

O

O His 139

CuII N NH

O C

Mononuclear metal site

CuII

O2

CuI His 139 NH

His 33

Glu 195

N

N

HN (b)

N H

O

Zn

N O

O

N

HN

Cu

Dinuclear copper site

N

N

O C

Glu 35

Monocopper site

HN His 137

Asp 156

NH N

O

(a)

His 160

His 137

His 72

37

N e−

CuIII O His 139 CuII N NH

O HN

NH N

His 33

Figure 2.4 (a) Three different copper sites identiﬁed in the pMMO crystal structure of M. capsulatus. (b) Proposal for O2 activation by the dinuclear copper site to obtain the most likely active bis(µ-oxo)CuII CuIII core.

theory) calculations, the mononuclear CuI center can activate O2 to form a CuII -superoxo species, followed by two consecutive H atom transfer from near tyrosine residues to generate a [CuIII =O]+ species and H2 O. This process was computed to be 17.8 kcal mol−1 endothermic, although was still considered possible under physiological conditions. On the other hand, the activation of O2 by the dinuclear CuI center affords a µ-η2 : η2 -(O2 )-CuII 2 side-on peroxo species that interconverts to a bis(µ-oxo)CuIII 2 isomer. One electron reduction of the latter leads to the formation of the dinuclear bis(µ-oxo)CuII CuIII species, an overall process computed to be 59.3 kcal mol−1 exothermic, thus much more favorable than the mononuclear [CuIII =O]+ . The latter result suggests that the dinuclear copper site might be the real executor of the pMMO catalytic activity. Indeed, the computed reaction path of methane hydroxylation by bis(µ-oxo)CuII CuIII (doublet ground state) shows a methane molecule weakly bound to the CuIII center (Figure 2.5), followed by H atom abstraction by a bridging oxo group (barrier 16.1 kcal mol−1 ) to form a nonradical intermediate, and ﬁnal recombination of OH and CH3 ligands coordinated to the two distinct copper centers (rate determining barrier of 24.3 kcal mol−1 ). It is interesting to note that the computed mechanistic proposal of these authors shows the participation of organometallic CuIII –CH3 moieties in the reaction path of both [CuIII =O]+ and bis(µ-oxo)CuII CuIII species with methane. This mechanistic proposal is consistent with experimental hydroxylation of chiral ethane [81] and butane [82] by pMMO from M. capsulatus, since the alcohol products are obtained with 100% retention of conﬁguration. The rate determining recombination step is, however, not consistent with the experimental H/D kinetic isotope effect (KIE = 5.2–5.5, 30 ◦ C for ethane hydroxylation) [81], although the authors

38

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

OH CH3 CH3

CuII

H O CuII

CuIII

O 16.1

TS2

TS1

1.9

−0.7

CH4 CuII

O O

CuIII O 24.3

OH

CuIII CuII

−43.6 CH3 CuIII

O

HO CuI

CH3 CuII

O

Figure 2.5 DFT energy diagram for the hydroxylation of methane by bis(µ-oxo)CuII CuIII species (energies in kilocalories per mole) [79].

believe that the protein environment can signiﬁcantly stabilize the transition state for OH migrations. Full consideration of protein environment is computationally demanding, but necessary to address the viability of the mechanism proposed via bis(µ-oxo)CuII CuIII species from a computational point of view. Balasubramanian et al. also indicate that the coordination of the carboxylate moiety from Glu35 may not be correctly considered in these computational studies, since it is found 4 A˚ apart from the closer Cu center [78]. The calculated activation energies for the reaction of bis(µ-oxo)CuIII CuIII species with CH4 were found high enough to clearly discard this species as playing a role in pMMO activity [79]. In a recent report, Hakemian et al. have crystallized the pMMO enzyme from Methylosinus trichosporium OB3b [83], and found no copper in the mononuclear site, thus suggesting its nonparticipation in the reactivity, in line with theoretical proposals of bis(µ-oxo)CuII CuIII species as active species. 2.2.3 Mononuclear Monooxygenating Copper-based Enzymes

Peptidylglycine α-hydroxylating monooxygenase (PHM) and dopamine β-monooxygenase (DβM) are copper-dependent enzymes found in higher eukaryotes that catalyze the regio and stereospeciﬁc α-hydroxylation of a C-terminal glycine and the hydroxylation of dopamine to norepinephrine, respectively (Figure 2.6) [84]. The active sites of PHM and DβM are almost identical, and both consist of two inequivalent mononuclear Cu centers (CuA and CuB ) far apart in space (11 A˚ in PHM) [85] with no direct bridging ligands and no observable magnetic interactions between metals (Figure 2.7a). The crystal structure of PHM indicates that the copper atom is coordinated by two histidine and one methionine ligands at the CuB site (where O2 coordinates and substrate hydroxylation occurs), whereas copper is coordinated to three histidine ligands at

2.2 Bioinorganic Implications of Copper(III) O

O HO H

2e−, 2H+ COO−

N H

Peptide

H 2O

H OH

2e−, 2H+

NH2

NH2

DbM O2

(b)

Figure 2.6

COO−

N H

Peptide

PHM O2

(a)

39

H2O

Global aliphatic hydroxylation reactions catalyzed by (a) PHM and (b) DβM. NHis NHis

CuAI R H

SMet

O O CuBII NHis

(a)

(b)

S

CuBIII N

NH N

NH

R

NHis

CuAI

R NHis

H

NHis

OH SMet

NHis

NHis

H 3C H3C

H 2N

OH OH

H2N ∆E = 3.8 kcal mol−1

O CuBII NHis

H abstraction

O H3C

NHis

OH OH

H2N H3C

11 Å

R H

OH II S CuB N

OH OH

OH O rebinding

NH N NH

∆E = 6.5 kcal mol−1

II H3C S CuB H3C N

Figure 2.7 (a) Mononuclear CuII B -superoxo proposal as active species responsible for aliphatic H atom abstraction in both natural DβM and PHM enzymes (mononuclear CuA lies 11 A˚ apart from CuB site). (b) Theoretical QM/MM proposal of a CuIII B -oxo as the most active species in DβM [88b].

the CuA site (which provides an additional electron through long-range electron transfer (ET) to the CuB site) [85, 86]. The structural similarities and parallel H/D KIEs strongly suggest that both enzymes perform the C–H abstraction reaction via very similar molecular mechanisms. Based on the available experimental data, Klinman tested four mechanistic hypotheses involving different active species [84]: a one-electron reduced intermediate (copper-superoxo, CuB II (O−· 2 )), two-electron reduced species II − (copper-peroxo, CuB II (O2− 2 ) or a copper-hydroperoxo, CuB (OOH )), and, ﬁnally, III 2− a highly reduced copper-oxo CuB (O ) formed via the reductive cleavage of CuB II (OOH− ) by a proximal tyrosine. Based on the large amount of experimental data for PHM and DβM reactivity, the author concluded that a copper-superoxo intermediate is the most consistent alternative as active species toward H atom abstraction from the substrate (Figure 2.7a). A recent synthetic model supports the copper(II)-superoxo alternative [87].

NH N NH

40

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

There are, however, two recent computational works that challenge the CuB II -superoxo proposal as the active species [88, 89]. Yoshisawa et al. carried out calculations for a full enzyme Quantum Mechanics-Molecular Mechanics (QM–MM model) obtained by using the crystal structure information from DβM and PHM. They evaluated the reactivity of copper-superoxo, -hydroperoxo, and -oxo species in DβM. The activation energies for the H abstraction by the CuII -superoxo moiety were found to be 23.1 kcal mol, thus suggesting that the CuB II -superoxo species can promote the H abstraction in DβM. The CuB II -hydroperoxo species was ruled out as an active oxidant, because the activation barrier of that H abstraction was found to be more than 40 kcal mol−1 . Finally, they concluded that the CuB III -oxo species reasonably promotes the H abstraction because the activation energy was found to be only 5.4 kcal mol−1 , appearing as the most active species within the protein active site environment (Figure 2.7b) [88]. The hydrogen-bonding network between dopamine and three amino acid residues (Glu268, Glu369, and Tyr494) was found to play an essential role in substrate binding and the stereospeciﬁc hydroxylation of dopamine to norepinephrine. A second computational report on PHM reactivity also supports a CuIII -oxo as the active species for H atom abstraction [89]. Potential energy proﬁles computed for the H abstraction reaction performed by copper-superoxo species, CuB II (O−· 2 ), suggest that it is not the active species due to the high activation energy found. Inspection of the potential energy surfaces shows that the CuB II -hydroperoxo species is spontaneously formed by abstracting a proton from the surrounding solvent, but cannot abstract the hydrogen atom from the substrate. However, this species can spontaneously abstract another proton from the surrounding III +2 solvent to form [L·+ ([CuB III O]+ moiety with two unpaired electrons 3 CuB O] ferromagnetically coupled with L·+ 3 , in the quartet spin ground state), along with the release of a water molecule. Despite enormous efforts devoted to investigate DβM and PHM reactivity, there is no deﬁnitive consensus on the mechanism and/or the identity of the active species responsible for H abstraction. Bioinorganic models usually help in the understanding of mechanistic issues of natural systems. In the context of mononuclear monooxygenase enzymes, Hong et al. reported a mononuclear CuI -α-ketocarboxylate complex bearing a bulky bidentate ligand that upon activation of O2 undergoes an intramolecular hydroxylation of an arene moiety [90]. Although not experimentally directly detected, theoretical data pointed toward a CuIII -oxo species as the active species for this aromatic hydroxylation. Experimental isolation of these active intermediates to prove their oxidizing power toward exogenous substrates is a desired target for bioinorganic chemists, which may help in the interpretation of intimate mechanistic details in enzymes. 2.2.4 Trinuclear Copper Models for Laccase

Lc, ceruloplasmin (Cp), and ascorbate oxidase (AO) are the representative enzymes of type (2+3) trimer, and are involved in the lignin formation (plants)

2.3 Organometallic CuIII Species in Organic Transformations

3 [LCuI(CH3CN)]+

L=

NMe2

−80 °C, CH2Cl2 O2 3 CH CN 3

[L3Cu3O2]3+ ON

N II

III

Cu Cu Cu NMe2

N

O N

N II

N

Figure 2.8 Synthetic functional model for the O2 activation by a trinuclear copper center forming a (bis-µ3 -oxo)[CuII CuII CuIII ] species [93].

and degradation (fungi), among others [31c, 34]. They are responsible for the 4e− reduction of O2 to two H2 O molecules by using the electrons coming from one-electron oxidation of four substrate molecules. The active sites of these enzymes contain a trinuclear CuI active site (capable of O2 reduction) [91], with an additional mononuclear ‘‘blue’’ CuI site at 12 A˚ far apart (as a proton transfer site to inject one electron to the trinuclear site) [92]. A genuine functional model of these trinuclear centers is a monocopper(I) complex with N-permethylated (1R,2R)-cyclohexanediamine ligand, which reacts with O2 and generates a (bis-µ3 -oxo)[CuII CuII CuIII ] (Figure 2.8) [93]. The spectroscopic properties of this species, and especially, the bands at 290 and 355 nm assigned to oxo-to-CuIII ligand-to-metal charge transfer (LMCT), do not match the spectroscopic properties of the oxidized forms of the enzymes. The latter was interpreted in the sense that the fourth electron to reduce O2 to two O2− comes from the ‘‘blue’’ CuI site oxidation to CuII in the natural system, whereas from one CuII center to generate one CuIII in the synthetic model. Therefore, although CuIII cations probably do not play a role in the enzymatic reactivity, full characterization of the synthetic model has been crucial for the understanding of the natural systems.

2.3 Organometallic CuIII Species in Organic Transformations

The implication of organometallic CuIII –C species in different organic transformations catalyzed by copper reagents has been proposed for a long time since 1941, when Kharash discovered [94] the reactivity of organocuprates (R2 CuLi, for instance) [95] toward nucleophilic alkyl-, vinyl, and aryl anion delivery in regio and stereoselective C–C bond forming processes as conjugate additions [96], carbocuprations [97], SN 2 alkylations [98], and SN 2 allylations [99]. General consensus exists on the formation of R–CuIII –R’ species followed by subsequent reductive elimination to account for organocuprate reactivity [95, 100]. Indeed, the more than one century-old Ullmann chemistry [101–103], referred in general terms to all aromatic nucleophilic substitution reactions in

41

42

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

the presence of copper(I) salts to form new Caryl –Caryl , Caryl –O, Caryl –N, or Caryl –S bonds, is thought to proceed by participation of high-valent organometallic aryl-CuIII –Nucleophile, which undergoes reductive elimination to form the cross-coupling aryl–Nucleophile product [104, 105]. From a general perspective, the accessibility of four oxidation states for copper (from 0 to +3) is a distinctive characteristic that governs its chemistry and differentiates it from Pd reactivity, which basically functions with the Pd0 /PdII redox pair (although several systems working with PdII /PdIV redox pair have recently appeared) [106, 107]. In this section, the most intriguing implications of reported CuIII organometallic species in catalytic or stoichiometric organic transformations involving CuI and CuII reagents are described. 2.3.1 C–C Bond Formation in Organocuprate(I) Catalysis

Four basic types of organocopper are known: RCuI , R2 CuI M (M = Li, MgX), R3 CuI Li2 , and R3 CuIII . The polymeric nature of RCuI renders unreactive species, whether R2 CuI M and R3 CuI Li2 species generated in situ are the most useful carbon nucleophile reagents. The exact structural nature of the latter in solution remains a matter of debate [95]; however, it is clear that lithium cations play a role in the stabilization and reactivity of the intermediate species. On the other hand, R3 CuIII species are proposed in numerous mechanistic discussions, although there are only three structurally characterized examples of di- and tetraalkyl CuIII species: [CuIII (CF3 )2 (SC(S)NEt2 )] [108], [CuIII (CF3 )4 ]- [109], and [CuIII (CF2 H)4 ]- [110]. Their square-planar geometry can be regarded as a planar T-shape geometry with a fourth ligand to enhance stabilization. 2.3.1.1 Conjugate Addition to α-Enones Conjugated additions to α, β-unsaturated ketones and esters are among the most important reactions of organocuprates. Based on NMR spectroscopic studies and theoretical calculations, the reaction of R2 CuI Li with enones is directed by Li-carbonyl and copper-oleﬁn interactions [111]. The key intermediate is a trialkyl-CuIII species that forms upon two-electron inner-sphere electron transfer (Figure 2.9). The organocopper R3 CuIII species then undergoes reductive elimination to afford the conjugate 1,4-addition product. Dorigo et al. already reported back in 1995 the theoretical stability of (CH3 )3 CuIII moieties further stabilized with one H2 O or NH3 molecule to complete a square-planar geometry [112]. Indeed, Snyder also predicted the possibility to isolate and even structurally characterize these CuIII species [113]. Very recently, Bertz et al. have used rapid injection nuclear magnetic resonance (RI-NMR) techniques at low temperatures (−100 ◦ C) to unequivocally characterize the largely sought and elusive trialkyl-CuIII intermediate upon reaction of 2-cyclohexenone and Me2 CuI Li

2.3 Organometallic CuIII Species in Organic Transformations

R1 CuI R1 Li Li R1 CuI R1

O R2

OLi R1 CuI R1 O Li Li R1 CuI R1

Reductive elimination

2

R

R1

OLi R2

CuIII

R12CuILi

43

R1

R1

+ R1CuI

Proposed mechanism for 1,4-conjugate addition of R2 CuI Li to α-enones.

Figure 2.9

Me Me Si Me

1. Inject TMSCI

Me2CuLi LiCN (−100 °C) 2. Inject O

O

H

Warming Li+ to −80 °C N− C Reductive CuIII Me elimination Me

OTMS + MeCuICNLi Me

NMR characterized Figure 2.10 Low-temperature RI-NMR characterization of an R3 CuIII intermediate species in an organocuprate conjugate addition reaction [100].

Li

X

Li I

R Cu R + H

H

Figure 2.11

“Trap” (oxidative addition) X = RCuR, halogen

Li

X

R Cu H

Li R “Bite” (reductive X Li CuI R R elimination)

Li III

H

H

Transmetalation

H

R Li

CuI

X Li H

Proposed trap-and-bite mechanism of acetylene carbocupration [95].

(Figure 2.10) [100], also supported by DFT calculations [114]. Warming of the CuIII intermediate up to −80 ◦ C afforded the conjugate 1,4-addition enolate. 2.3.1.2 Acetylene Carbocupration Synthesis of alkenyl derivatives can be achieved in a stereoselective manner by reaction of organocuprate(I) reagents with acetylene, a process that involves the intermediacy of a R2 CuIII (acetylide) species in a trap-and-bite mechanistic proposal (Figure 2.11) [95, 97]. These reactions are usually performed at room temperature or below. 2.3.1.3 SN 2 and SN 2 Alkylations The alkylation of alkyl halides (and epoxides) with organocuprate(I) reagents (Me2 CuI Li) occurs through a SN 2 mechanism in which the C–X (X = halide) bond is substituted by an incoming R–Cu σ bond to generate unstable Me2 CuIII R intermediates. The latter undergoes reductive elimination to afford the cross-coupled Me–R product. This mechanistic proposal is fully supported by the recent RI-NMR characterization of several CuIII tetracoordinate alkyl species upon reaction of ethyl iodide (EtI) with a variety of methyl Gilman reagents, Me2 CuLi·LiX (X = I, CN,

R H

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

44

CH3CH2I

Me2CuILi LiX (−100 °C)

X = I, CN, SCN, SPh

Me Et CuIII X Me

Li

Me2CuILi LiCN MeCuIXLi LiCN

Lil

Reductive Me elimination III CH3CH2CH3 Et Cu Me Li Me I Me2Cu Li

(a)

3 Me2CuLi (−100 °C) (b)

2

Cl

Me CuIII Me

MeCuIClLi + LiCI

+

Reductive Me elimination 2 CuIII Me Li Me Me2CuILi + MeCuI

CH3

Figure 2.12 Low-temperature RI-NMR characterization of (a) the ﬁrst nonﬂuorinated R4 CuIII intermediate species in an SN 2 alkylation [115] and (b) η1 σ -allyl and η3 π -allyl-CuIII species [117].

SCN, SPh) [115, 116]. Among them, [CuIII Me3 Et]Li is the most stable and also the ﬁrst example of a tetraalkylcopper(III) without ﬂuorinated alkyl substituents (Figure 2.12a) [109, 110]. Similar reactivity is observed if allylic halides are used, but usually a mixture of SN 2 and SN 2 products is obtained. This observation suggested the existence of a regioisomeric σ -allyl-CuIII species, which has been certiﬁed very recently by means of RI-NMR with the identiﬁcation of both η1 σ -allyl and η3 π-allyl-CuIII complexes upon reaction of allyl chloride and Me2 CuI Li·LiI (Figure 2.12b) [117, 118]. The σ complex is formed initially but transforms rapidly to the more stable π-allyl-CuIII , which undergoes reductive elimination to form 1-butene upon warming (the SN 2 product). Depending on the allylic halide nature and reaction conditions, different ratios of SN 2 and SN 2 products are obtained. 2.3.2 Aryl–Heteroatom Bond Formation in Cu-mediated Cross-coupling Processes

Impressive development of palladium-catalyzed methods for C–heteroatom and C–C bond formation [106] has obscured the older and classical Ullmann chemistry of copper-catalyzed C–N, C–O, and C–S bond formation reactions [101, 103, 104]. However, some of the deﬁciencies related to the use of Pd reagents such as their cost, high toxicity, use of sophisticated ligands, and low tolerance to reactive functional groups have prompted the revitalization of research on copper-based cross-coupling reactions between haloarenes and a wide range of N-nucleophiles (i.e., aromatic [119] and aliphatic [120] amines, imidazoles [121], and amides [122]), O-nucleophiles (phenols [123] and aliphatic alcohols [124]), and S-nucleophiles (aryl and alkyl thiols [125]). Equivalent methods for the synthesis of aryl-selenides [126] and aryl-phosphines [127] are also known. Examples of all these cross-coupling reactions exist with the mediation of copper(I) salts (CuI X, X = OAc, Br, I) (the use of a soft base and a ligand auxiliary additive is often required). The proposed mechanistic pathway consists in the oxidative addition of CuI to the haloarene to form an aryl-CuIII species, followed by coordination of the nucleophile source

2.3 Organometallic CuIII Species in Organic Transformations

45

Y X

CuIX

X

X = OAc, Br, I, Cl oxidative addition

L

Figure 2.13

Y = N, O, S CuIII X

X L

X−

CuIII Y

Reductive elimination

Cross-coupling of haloarenes and nucleophiles mediated by CuI salts.

and subsequent reductive elimination, to ﬁnally afford the arylation product (Figure 2.13) [104]. An impressive number of parallel methods have been reported by using organometallic-aryl compounds instead of haloarenes [104, 128], such as arylboronic acids [129], triaryl-bismuth(V)-diacetate (i.e., Ph3 Bi(OAc)2 ) [130], aryl-lead(IV)-triacetates [131], arylsiloxanes (i.e., PhSi(OMe)3 ) [132], and arylstannanes (i.e., PhSnMe3 ) [133], among others. These methods permit the obtention of the same O-, N-, and S-arylation products under signiﬁcantly milder experimental conditions and shorter reaction times. Mechanistic proposals include initial transmetallation with the copper salt and suggest also the intermediacy of aryl-CuIII species prior to cross-coupling product formation for those methodologies using CuII salts (CuII (OAc)2 , CuII (OTf)2 , CuII Cl2 , CuII SO4 ·5H2 O) [104, 105]. In the latter, reactions require O2 atmosphere, what suggests the oxidation of aryl-CuII to aryl-CuIII in order to facilitate the subsequent reductive elimination step. 2.3.3 Aromatic and Aliphatic C–H Bond Organometallic Functionalizations

The transition metal activation of an inert alkane C–H bond is a challenging task [134], thus coupling a C–H bond activation with a subsequent functionalization step in a catalytic manner is even more complicated [107]. Enzymes can efﬁciently perform oxygenase processes through metal-oxo, -peroxo, or -superoxo species as active species [38–40, 84, 135], and it is in general believed that these catalyses do not occur through the intermediacy of M–C bonds. Hereafter, systems where a transition metal complex regioselectively cleaves a C–H bond to form organometallic M–C species that undergo functionalization via reductive elimination, provided a nucleophile is present, are discussed. 2.3.3.1 Catalytic Systems The well-known Shilov system is based on PtII /PtIV chloride complexes in aqueous solution capable of catalytic methane functionalization to methanol at 120 ◦ C [136]. Mechanistically, a PtII –CH3 species is initially formed, followed by oxidation to PtIV and ﬁnal reductive elimination in the presence of H2 O to obtain CH3 OH and regenerate the PtII salt to restart the catalytic cycle. Some palladium-based systems that catalytically functionalize aromatic and aliphatic C–H bonds achieve their activation by means of the coordination-directed metallation strategy [134], where a functional group of the substrate coordinates

Y

+ CuIX

46

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

R1 (a)

But

( ) R2 1-2 FG H MLn

−H+

R1

( ) R2 1-2 MLn FG

( ) R2 1-2 FG X

X = C, O, N, Halide

O

Pd(OAc)2 (10 mol%)

But

Phl(OAc)2 (1 equiv.) I2 (1 equiv.) CH2Cl2 24 °C, 13 h

N

(b) R-X

L2PdII

Reductive elimination X L2PdIV X

X X

R-H

O N I

C-H bond cleavage HX

X L2PdII

R

R = alkyl, aryl X = OAc, OR, X, OH (c)

R1

X R

Oxidation X2

Figure 2.14 (a) Coordination-directed metallation strategy for C–H functionalizations; (b) example of arene halogenation mediated by PdII species; and (c) catalytic cycle proposed for PdII systems.

the metal in such a way that it is well positioned to regioselectively interact with a proximal C–H bond (Figure 2.14a). By using PdII reagents, PdII –C species are generated, followed by the subsequent functionalization step, which is facilitated by previously resorting to the oxidation of metallated PdII to PdIV species with the concomitant coordination of selected nucleophiles. This is an effective strategy to lower the barrier for the ﬁnal reductive elimination step (Figure 2.14c) [107]. Following this methodology, broad scope halogenation of unactivated aromatic C–H bonds [137] (Figure 2.14b) and oxygenation of aliphatic sp3 C–H bonds [138] have been reported. A growing number of examples of C–H bond functionalization using CuII salts and involving a coordination-directed metallation have been reported recently [139–141], although the intimate mechanistic details remain quite obscure. For example, Chen et al. reported the ortho-acetoxylation and chlorination of aryl C–H bonds in phenyl-pyridine substrates mediated by CuII (OAc)2 and CuII Cl2 , respectively [139]. The coordination of CuII center to the pyridinic N accounts for the ortho-regioselectivity of the C–H activation step (Figure 2.15). A single electron transfer (SET) from the arene to reduce the metal to CuI , followed by the nucleophile insertion [140], proton loss, and a second SET mechanistic pathway,

2.3 Organometallic CuIII Species in Organic Transformations

N

X H

X = H, Me, F, Cl, CF3, OMe, COOMe, CHO, CH=CH2

Cu(OAc)2, O2 Nucleophiles (Nu−)

N

X

Solvent, 130 °C (40–90% yield)

Nu Nu = OH, OAc, p-CN-PhO, Cl, Br, I, PhS, MeS, TsNH, CN

Figure 2.15 Coordination-directed metallation strategy for the functionalization of arenes with a wide range of nucleophiles [139].

is proposed based on a nonobservable KIE; interestingly, the formation of a cyclometallated aryl-CuII species followed by reductive elimination to product and Cu0 is not discarded. These reactions are conducted under O2 , which is believed to reoxidate low-valent copper species to regenerate the CuII catalyst. Other recent CuII -mediated C–H bond catalytic functionalizations include the synthesis of benzimidazoles from amidines (C–N bond formation) [142] and of 2-arylbenzoxazoles from benzanilides via an intramolecular C–O bond formation [143]. In neither of the previous reports, a clear mechanistic picture is presented. 2.3.3.2 Stoichiometric Systems Although stoichiometric C–H bond functionalizations have minor synthetic impact compared to catalytic ones, they may allow for a more detailed mechanistic investigation of the overall process, and, in selected cases, important intermediates may be trapped. Ribas et al. have developed discrete systems based on triazamacrocyclic ligands (Ltriazamac ) that exhibit aromatic C–H bond activation upon complexation with CuII salts at exceedingly mild conditions [144]. The outcome of the reaction is the formation of equimolar amounts of stable, diamagnetic aryl-CuIII and CuI species (Figure 2.16). Full characterization of the metallated aryl-CuIII product by means of X-ray diffraction, NMR spectroscopy, cyclic voltammetry, and Cu K-edge XAS conﬁrmed the high oxidation state of the metal center. Crystal structures showed ˚ and XAS spectra short CuIII –Caryl bond distances ranging from 1.85 to 1.90 A, exhibited a very intense transition at 8981 ± 0.5 eV, which has been demonstrated to be unequivocally indicative of the 1 s → 3d transition of CuIII compounds [30]. The singlet multiplicity for a d8 metal in a square-planar geometry agrees with the possibility of full NMR assignment of the diamagnetic compounds. The scope of this reaction was expanded to a whole family of triazamacrocyclic ligands [145], allowing us to understand the inﬂuence of steric and electronic factors in the properties of the aryl-CuIII complexes. It was found that the redox potential for the CuIII /CuII couple depends, as expected, on the electronic nature of the para-phenyl substituents as well as on the nature of the central macrocyclic amine (E1/2 (CuIII /CuII ) in the range between −201 and +6 mV versus Saturated Calomel Electrode(SCE)). Electron-withdrawing substituents in the phenyl ring destabilize the oxidation state +3, whereas electron donors produce the opposite

47

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

48

R1

HN

Ha NH CuII N

R1

2+

CH3CN 25 °C, < 1 min

0.5 HN CuIII

R

R2 [CuII(Ltriazamac)]2+ R1 = H, CH3, NO2 n = 0,1

2

[aryl-CuIII]2+ + 0.5 [CuI(CH3CN)4]+

R2 = H, CH3

C1

NH n

N

n

2+

1.901 N1

CuIII

N3

1.952

1.960

2.002 N2 (R1 = H; R2 = H, n = 1)

+ 0.5

(LtriazamacH)+

Figure 2.16 Aromatic C–H bond cleavage involving CuII disproportionation to afford aryl-CuIII and CuI species in a family of triazamacrocyclic ligands (crystal structure of one aryl-CuIII species included; H atoms omitted for clarity) [144, 145].

effect. Tertiary amines have a lower σ -donating capacity than secondary amines owing to a lower overlap capacity and also to solvation effects (N-methylation effect) [146], and this was manifested in a cathodic shift of the E1/2 redox potential. The isolation of stable aryl-CuIII species is important from three points of view. The ﬁrst one is that no other example of structurally characterized aryl-CuIII species has been reported in the literature, and that these species can be considered as intermediate models of aryl–heteroatom cross-coupling reactions. The second one is that C–H activation is achieved very rapidly (seconds) and at room temperature, thus the mechanistic pathway by which this overall CuII disproportionation reaction takes place may provide some important clues to explain its fast reactivity [147], Finally, the third remarkable issue is the fact that we can make use of these aryl-CuIII species to evaluate their reactivity with nucleophiles as models of the aryl–heteroatom cross-coupling reactivity via a reductive elimination ﬁnal stage. Huffman et al. have used three of these aryl-CuIII complexes to prove their ability for C–N coupling reactions with amide-type nucleophiles (Figure 2.17) [148]. The reactions occurred in a facile manner at room temperature to produce quantitatively the C–N coupled organic compound and [CuI (CH3 CN)4 ]+ , the expected products of a reductive elimination. The decay of aryl-CuIII features monitored by UV–Vis and 1 H-NMR indicate pseudo-ﬁrst-order kinetics under excess of amide. They found that electron-withdrawing groups (−NO2 ) in the para position of the aromatic ring with respect to the ipso-carbon accelerate the reaction, whereas electron-donating groups disfavor it, thus clearly suggesting that the more electrophilic aryl-CuIII species react faster with a given N-nucleophilic substrate. On the other hand, the secondary amides exhibiting a higher pKa reacted faster, and a linear correlation between the logarithm of kobs and pKa (acidity) was found. Although the precise

2.3 Organometallic CuIII Species in Organic Transformations

R1

49

2+ H N O

HN CuIII NH

+

N

HN

CH3CN

O N R1

N

+ [CuI(CH3CN)4]+

25 °C, < 0.5 hours (And other amides)

N H

R1 = H, CH3, NO2 Figure 2.17 Facile C–N bond formation in aryl-amide coupling via reductive elimination of well-deﬁned, isolated aryl-CuIII species [148].

mechanism of the reductive elimination step is not clear since no intermediate species are detected upon addition of the amide nucleophile, it seems clear that amide deprotonation is key for the reaction to occur. We are currently exploring the reactivity of aryl-CuIII species with O-nucleophiles to generate the aryl-O coupled products. Preliminary experiments conﬁrm the possibility to obtain the products of reductive elimination upon reaction with aromatic and aliphatic alcohols [149]. Furthermore, the reactivity with hydroxide affords dinuclear bis-phenoxo-CuII 2 compounds in moderate yields [144, 150, 151]. We reported a few years ago the intramolecular aromatic hydroxylation of a 12-membered Ltriazamac type of ligand (named as H22m) upon reaction of its CuI complex with molecular O2 [152]. We have now evidence regarding the intermediacy of aryl-CuIII species in that process, although not isolable in this particular system due to the reduced size of the macrocyclic cavity. When H22m is reacted with CuII salts in the presence of water, the same bis-phenoxo-CuII 2 compound is obtained (Figure 2.18). Our reasoning to justify the isolation of ﬁnal CuII species is that the thermodynamic stability of the ﬁnal dinuclear bis-phenoxo species may be the driving force to oxidize the CuI species and recombine with the aryl-OH organic compounds, the original products of reductive elimination. The study of the reactivity of the corresponding CuI complexes, bearing the same Ltriazamac ligands, allowed us to report for the ﬁrst time a family of CuI complexes that undergo regiospeciﬁc isotopic H/D exchange in D6-acetone, which represents an example of aromatic C–H activation by a CuI center (Figure 2.19) [153]. The reaction is reversible, quantitative (100% deuteration was found for the aromatic C–Ha bond, from minutes to hours depending on the ligand), and solvent dependent (occurs in D6-acetone, not in CD3 CN or CD2 Cl2 ). Our mechanistic proposal to explain the H/D exchange highlights the formation of a transient oxidative addition aryl-CuIII –hydride intermediate, followed by the H/D exchange assisted by the close proximity of a D6-acetone in its enolic form (Figure 2.19b). The observed reductive elimination CuI and Ltriazamac products by the reaction of aryl-CuIII species with hydride as nucleophile supports the H/D exchange mechanism. Noteworthy, the isolation of monodeuterated ligands at

50

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

+ CuI ; O2, CH3CN, 298 K

N1 II N3 Cu

2+

H2O NH

O1′

+ CuII ; CH3CN, H2O, 298 K HN CuIII NH N H

HN N H

N2

O1

N2′ N1′

Proposed aryl-CuIII intermediate

CuII N3′

Figure 2.18 Aromatic hydroxylation involving CuII disproportionation to afford a transient aryl-CuIII , which reacts with a nucleophilic water molecule to ﬁnally isolate bis-phenoxo-Cu2 II species (crystal structure ellipsoid diagram included; H atoms omitted for clarity) [152].

R1

R1 CD3CN

R1-H33m + CuI(CH3CN)4PF6

HN

Ha CuI NH

(a)

D

(b)

CD2

D3C

O

O H

CuI

N H

H, CH3, NO2

CD2

D3C

D CuI NH

HN CH3COCH3

N H R1:

CD3COCD3

R1

D

Oxidative addition

O

H

CuIII

CD3

D2C

1

H

R

H/D exchange

CD2

D 3C H

D R

CuIII

1

O D CuI

R1

Reductive elimination

Figure 2.19 (a) C–H activation in triazamacrocyclic ligands by CuI : isotopic H/D exchange at room temperature in D6-acetone. (b) Mechanistic proposal for H/D exchange reaction [153].

the C–Ha position will be an important tool to unravel important aspects of the reactivity of C–H bond activations by CuII within these systems. Up to this point, the work completes an overview on the different reactivity of these copper-containing systems depending on its oxidation state (CuII or CuI reactant species), but in either case the role of aryl-CuIII species has been demonstrated. The fundamental investigation on the mechanistic details of these intramolecular functionalization reactions within triazamacrocyclic systems will help in the development of catalytic versions of C–H bond activation processes, and may give important clues in the understanding of relevant CuII -catalyzed acetoxylation, halogenation, and arylation of aryl C–H bonds, among others [139].

2.5 Overview and Future Targets

Other reported stoichiometric C–H bond activation processes mediated by CuII species are the reaction of doubly-N-confused porphyrins (NCPs) with CuII (OAc)2 at room temperature to yield the NCP–CuIII species quantitatively in few hours [154], as well as a recent work from Ban et al. [155] that report the multiple C–H arylation of indoles and pyrroles mediated by stoichiometric amounts of CuII (O2 CCF3 )2 upon heating to 80 ◦ C for 24 hours. The latter reactivity is intriguing since it does not follow the coordination-directed functionalization strategy, thus further mechanistic investigation is required.

2.4 Miscellany: Cuprate Superconducting Materials

Owing to their relevance in potential technological applications, the inclusion of the CuIII role in high-temperature superconducting (HTSC) cuprates warrants their mention in this chapter, despite the still unclear participation of this oxidation state in the superconducting free electron ﬂow. Cuprates as the well-known perovskite-type YBa2 Cu3 O7 (YBCO) or La0.85 Sr0.15 CuO4 present copper-oxide layers where the metal oxidation state is formally a mixture of CuII and CuIII [156]. Although it has been claimed that the holes may reside on the 2p-oxygen atom orbitals [157], many studies support the partial oxidation of CuII to CuIII , thus involving the unoccupied 3d Cu orbitals [158, 159]. Cu K-edge XAS and spectroscopic titration with p-anisidine have been reported to quantify the CuIII content in superconducting cuprate materials. In any case, the CuIII role in the origin of cuprate superconductivity remains unclear since their discovery 20 years ago [160]. The recently reported high-temperature iron-arsenic superconductors have rebursted the ﬁeld [161] and may help in the understanding of the superconducting phenomena for HTSC, which falls beyond the Bardeen–Cooper–Schrieffer theory that can explain superconductivity only at low temperatures (<20 K).

2.5 Overview and Future Targets

The chemistry of CuIII is living a golden period since the mostly proposed or invoked organometallic CuIII species for many years are now being experimentally characterized. The full characterization of R3 CuIII species in organocuprate chemistry and of aryl-CuIII species in the context of cross-coupling reactions to form new aryl–C or aryl–heteroatom bonds are key to understand the reactivity of these processes, and allow for the solid establishment of a CuI /CuIII redox chemistry involving oxidative addition to form CuIII species and subsequent reductive elimination step to afford the coupled products and CuI species. The organometallic C–H bond functionalizations mediated by CuII salts are less understood, specially the initial formation of the organometallic CuIII –C species; subsequent reductive elimination step is now better understood.

51

52

2 The Bioinorganic and Organometallic Chemistry of Copper(III)

On the other hand, the implication of CuIII -oxo species in the chemistry of several copper-dependent metalloenzymes has been proposed but observed only in synthetic models. Nevertheless, their characterization will help for a deeper understanding of enzymatic processes. Theoretical studies, walking usually a step forward compared to experimental works, have already supported the implication of bis(µ-oxo)CuII CuIII core as the active species in pMMO for achieving the conversion of methane to methanol. Intriguingly, the implication of an organometallic CuIII –methyl intermediate in a theoretically predicted reaction path brings out the question of whether a link between the organometallic and bioinorganic worlds can be made. The general unstability of organometallic species with ﬁrst-row transition metals is a requirement for a catalytic turnover in any process, and it is not discardable that future exploration of Cu-based metalloenzymes may function through organometallic pathways.

References 1. Cotton, F.A. and Wilkinson, G. (1988)

2. 3. 4. 5.

6.

7. 8. 9.

10. 11.

12.

Advanced Inorganic Chemistry, 5th edn, John Wiley & Sons, Inc., New York, p. 774. Scholder, R. and Voelskow, U. (1951) Z. Anorg. Allg. Chem., 266, 256. Magee, J.S. and Wood, R.H. (1965) Can. J. Chem., 43, 1234. Bour, J.J. and Steggerda, J.J. (1967) Chem. Commun., 85. Bour, J.J., Birker, P.J.M.W.L., and Steggerda, J.J. (1971) Inorg. Chem., 10, 1202–1205. (a) Meyerstein, D. (1971) Inorg. Chem., 10, 638–641; (b) Meyerstein, D. (1971) Inorg. Chem., 10, 2244–2249; (c) Cabelli, D.E., Wishart, J.F., Holcman, J., Meier, M., and van Eldik, R. (1997) J. Phys. Chem. A, 101, 5131–5136. Ulanski, P. and von Sonntag, C. (2000) Eur. J. Inorg. Chem., 1211–1217. Olson, D.C. and Vasilekskis, J. (1971) Inorg. Chem., 10, 463–470. Kumar, K., Rotzinger, F.P., and Endicott, J.F. (1983) J. Am. Chem. Soc., 105, 7064–7074. Keyworth, D.A. and Stone, K.G. (1955) Anal. Chem., 27, 833–835. Margerum, D.W., Chellappa, K.L., Bossu, F.P., and Burce, G.L. (1975) J. Am. Chem. Soc., 97, 6894–6896. (a) Margerum, D.W. (1983) Pure Appl. Chem., 55, 23–34; (b) Diaddario, L.L.,

13.

14.

15. 16.

17.

18.

19.

20.

21.

Robinson, W.R., and Margerum, D.W. (1983) Inorg. Chem., 22, 1021–1025. Keyes, W.E., Dunn, J.B.R., and Loehr, T.M. (1977) J. Am. Chem. Soc., 99, 4527–4529. Stevens, P., Waldeck, J.M., Strohl, J., and Nakon, R. (1978) J. Am. Chem. Soc., 100, 3632–3634. Sulfab, Y. and Al-Shatti, N.I. (1984) Inorg. Chim. Acta, 87, L23–L24. Ruiz, R., Surville-Barland, C., Aukauloo, A., Anxolabehere-Mallart, E., Journaux, Y., Cano, J., and Mu˜ noz, M.C. (1997) J. Chem. Soc., Dalton Trans., 745–751. Anson, F.C., Collins, T.J., Richmond, T.G., Santarsiero, B.D., Toth, J.E., and Treco, B.G.R.T. (1987) J. Am. Chem. Soc., 109, 2974–2979. Hanss, J., Beckmann, A., and Kr¨uger, H.-J. (1999) Eur. J. Inorg. Chem., 163–172. Youngblood, M.P. and Margerum, D.W. (1980) Inorg. Chem., 19, 3068–3072. Krebs, C., Glaser, T., Bill, E., Weyherm¨uller, T., Meyer-Klaucke, W., and Wieghardt, K. (1999) Angew. Chem. Int. Ed., 38, 359–361. Santo, R., Miyamoto, R., Tanaka, R., Nishioka, T., Sato, K., Toyota, K., Obata, M., Yano, S., Kinoshita, I., Ichimura, A., and Takui, T. (2006) Angew. Chem. Int. Ed., 45, 7611–7614.

References 22. Chaudhuri, P., Verani, C.N., Bill, E.,

23.

24.

25. 26.

27.

28.

29.

30.

31.

32. 33.

Bothe, E., Weyherm¨uller, T., and Wieghardt, K. (2001) J. Am. Chem. Soc., 123, 2213–2223. Verani, C.N., Gallert, S., Bill, E., Weyherm¨uller, T., Wieghardt, K., and Chaudhuri, P. (1999) Chem. Commun., 1474. Hsieh, C.-H., Hsu, I.-J., Lee, C.-M., Ke, S.-C., Wang, T.-Y., Lee, G.-H., Wang, Y., Chen, J.-M., Lee, J.-F., and Liaw, W.-F. (2003) Inorg. Chem., 42, 3925–3933. Meln´ık, M. and Kabeˇsov´a, M. (2000) J. Coord. Chem., 50, 323–338. Ribas, X., Dias, J.C., Morgado, J., Wurst, K., Molins, E., Ruiz, E., Almeida, M., Veciana, J., and Rovira, C. (2004) Chem. Eur. J., 10, 1691–1704. Ribas, X., Dias, J.C., Morgado, J., Wurst, K., Almeida, M., Veciana, J., and Rovira, C. (2002) Cryst. Eng. Commun., 4, 564–567. Ribas, X., Maspoch, D., Dias, J.C., Morgado, J., Almeida, M., Wurst, K., Vaughan, G., Veciana, J., and Rovira, C. (2004) Cryst. Eng. Commun., 6, 589–592. Sarangi, R., DeBeer George, S., Rudd, D.J., Szilagyi, R.K., Ribas, X., Rovira, C., Almeida, M., Hodgson, K.O., Hedman, B., and Solomon, E.I. (2007) J. Am. Chem. Soc., 129, 2316–2326. DuBois, J.L., Mukherjee, P., Stack, T.D.P., Hedman, B., Solomon, E.I., and Hodgson, K.O. (2000) J. Am. Chem. Soc., 122, 5775–5787. (a) Lippard, S.J. and Berg, J.M. (1994) Principles of Bioinorganic Chemistry, University Science Books, Mill Valley; (b) Holm, R.H., Kennepohl, P., and Solomon, E.I. (1996) Chem. Rev., 96, 2239–2314; (c) Solomon, E.I., Chen, P., Metz, M., Lee, S.-K., and Palmer, A.E. (2001) Angew. Chem. Int. Ed., 40, 4570–4590. Klinman, J.P. (1996) Chem. Rev., 96, 2541–2561. (a) Volbeda, A. and Hol, W.G.J. (1989) J. Mol. Biol., 209, 249–279; (b) Magnus, K.A., Ton-That, H., and

34.

35.

36.

37. 38.

39.

40. 41.

42.

43.

44.

45.

46.

47.

48.

49.

Carpenter, J.E. (1994) Chem. Rev., 94, 727–735. Solomon, E.I., Sundaram, U.M., and Machonkin, T.E. (1996) Chem. Rev., 96, 2563–2605. Decker, H., Schweikardt, T., and Tuczek, F. (2006) Angew. Chem. Int. Ed., 45, 4546–4550. Kitajima, N., Fujisawa, K., Fujimoto, C., Moro-oka, Y., Hashimoto, S., Kitagawa, T., Toriumi, K., Tatsumi, K., and Nakamura, A. (1992) J. Am. Chem. Soc., 114, 1277–1291. Solomon, E.I. and Lowery, M.D. (1993) Science, 259, 1575–1581. Karlin, K.D., Kaderli, S., and Zuberb¨uhler, A.D. (1997) Acc. Chem. Res., 30, 139–147. Mirica, L.M., Ottenwaelder, X., and Stack, T.D.P. (2004) Chem. Rev., 104, 1013–1045. Lewis, E.A. and Tolman, W.B. (2004) Chem. Rev., 114, 1047–1076. Cuff, M.E., Miller, K.I., van Holde, K.E., and Hendrickson, W.N. (1998) J. Mol. Biol., 278, 855–870. Matoba, Y., Kumagai, T., Yamamoto, A., and Sugiyama, M.J. (2006) J. Biol. Chem., 281, 8981–8990. Morioka, C., Tachi, Y., Suzuki, S., and Itoh, S. (2006) J. Am. Chem. Soc., 128, 6788–6789. Suzuki, K., Shimokawa, C., Morioka, C., and Itoh, S. (2008) Biochemistry, 47, 7108–7115. Decker, H., Dillinger, R., and Tuczek, F. (2000) Angew. Chem. Int. Ed., 39, 1591–1595. Inoue, T., Shiota, Y., and Yoshizawa, K. (2008) J. Am. Chem. Soc., 130, 16890–16897. Pidcock, E., Obias, H.V., Zhang, C.X., Karlin, K.D., and Solomon, E.I. (1998) J. Am. Chem. Soc., 120, 7841–7847. Karlin, K.D., Nasir, M.S., Cohen, B.I., Cruse, R.W., Kaderli, S., and Zuberb¨uhler, A.D. (1994) J. Am. Chem. Soc., 116, 1324–1336. Casella, L., Monzani, E., Gullotti, M., Caravagnoni, D., Cerina, G., Santagostini, L., and Ugo, R. (1996) Inorg. Chem., 35, 7516–7525.

53

54

2 The Bioinorganic and Organometallic Chemistry of Copper(III) 50. Menif, R., Martell, A.E., Squattrito, P.J.,

51.

52.

53.

54.

55.

56.

57.

58. 59.

60.

61.

62.

63.

and Clearﬁeld, A. (1990) Inorg. Chem., 29, 4723–4729. Costas, M., Ribas, X., Poater, A., Balvuena, J.M., Xifra, R., Company, A., Duran, M., Sol`a, M., Llobet, A., Corbella, M., Us´on, M.A., Mah´ıa, J., Solans, X., Shan, X., and Benet-Buchholz, J. (2006) Inorg. Chem., 45, 3569–3581. Poater, A., Ribas, X., Llobet, A., Cavallo, L., and Sola, M. (2008) J. Am. Chem. Soc., 130, 17710–17717. Sander, O., Henß, A., N¨ather, C., W¨urtele, C., Holthausen, M.C., Schindler, S., and Tuczek, F. (2008) Chem. Eur. J., 14, 9714–9729. Santagostini, L., Gullotti, M., Monzani, E., Casella, L., Dillinger, R., and Tuczek, F. (2000) Chem. Eur. J., 6, 519–522. Battaini, G., Granata, A., Monzani, E., Gullotti, M., and Casella, L. (2006) Adv. Inorg. Chem., 58, 185–233. Palavicini, S., Granata, A., Monzani, E., and Casella, L. (2005) J. Am. Chem. Soc., 127, 18031–18036. Itoh, S., Kumei, H., Taki, M., Nagatomo, S., Kitagawa, T., and Fukuzumi, S. (2001) J. Am. Chem. Soc., 123, 6708–6709. Itoh, S. and Fukuzumi, S. (2007) Acc. Chem. Res., 40, 592–600. Mahapatra, S., Halfen, J.A., Wilkinson, E.C., Pan, G., Cramer, C.J., Que, L. Jr., and Tolman, W.B. (1995) J. Am. Chem. Soc., 117, 8865–8866. Halfen, J.A., Mahapatra, S., Wilkinson, E.C., Kaderli, S., Young, V.G. Jr., Que, L.Jr., Zuberb¨uhler, A.D., and Tolman, W.B. (1996) Science, 271, 1397–1400. Mahapatra, S., Halfen, J.A., Wilkinson, E.C., Pan, G., Wang, X., Young, V.G. Jr., Cramer, C.J., Que, L. Jr., and Tolman, W.B. (1996) J. Am. Chem. Soc., 118, 11555–11574. Mahadevan, V., Henson, M.J., Solomon, E.I., and Stack, T.D.P. (2000) J. Am. Chem. Soc., 122, 10249–10250. Mahadevan, V., Hou, Z., Cole, A.P., Root, D.E., Lal, T.K., Solomon, E.I., and Stack, T.D.P. (1997) J. Am. Chem. Soc., 119, 11996–11997.

64. Maiti, D., Woertink, J.S.,

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

Narducci Sarjeant, A.A., Solomon, E.I., and Karlin, K.D. (2008) Inorg. Chem., 47, 3787–3800. Henson, M.J., Vance, M.A., Zhang, C.X., Liang, H.-C., Karlin, K.D., and Solomon, E.I. (2003) J. Am. Chem. Soc., 125, 5186–5192. Holland, P.L., Rodgers, K.R., and Tolman, W.B. (1999) Angew. Chem. Int. Ed., 38, 1139–1142. Mirica, L.M., Vance, M., Rudd, D.J., Hedman, B., Hodgson, K.O., Solomon, E.I., and Stack, T.D.P. (2005) Science, 308, 1890–1892. Company, A., Palavicini, S., Garcia-Bosch, I., Mas-Ballest´e, R., Que, L. Jr., Rybak-Akimova, E.V., Casella, L., Ribas, X., and Costas, M. (2008) Chem. Eur. J., 14, 3535–3538. Company, A., Lamata, D., Poater, A., SolS, M., Llobet, A., Parella, T., Fontrodona, X., Que, L. Jr., and Costas, M. (2006) Inorg. Chem., 45, 5239–5241. Herres-Pawlis, S., Verma, P., Haase, R., Kang, P., Lyons, C.T., Wasinger, E.C., Fl¨orke, U., Henkel, G., and Stack, T.D.P. (2009) J. Am. Chem. Soc., 131, 1154–1169. Spada, A., Palavicini, S., Monzani, E., Bubacco, L., and Casella, L., (2009) Dalton Trans., 33, 6468–6471. Siegbahn, P.E.M. and Wirstam, M. (2001) J. Am. Chem. Soc., 123, 11819–11820. Merkx, M., Kopp, D.A., Sazinsky, M.H., Blazyk, J.L., M¨uller, J., and Lippard, S.J. (2001) Angew. Chem. Int. Ed., 40, 2782–2807. (a) Rosenzweig, A.C., Frederick, C.A., Lippard, S.J., and Nordlund, P. (1993) Nature, 366, 537–543; (b) Woodland, M.P., and Dalton, H. (1984) J. Biol. Chem., 259, 53–59; (c) Fox, B.G., Froland, W.A., Dege, J.E., and Lipscomb, J.D. (1989) J. Biol. Chem., 264, 10023–10033. Nguyen, H.-H.T., Nakagawa, K.H., Hedman, B., Elliott, S.J., Lidstrom, M.E., Hodgson, K.O., and Chan, S.I. (1996) J. Am. Chem. Soc., 118, 12766–12776.

References 76. Lieberman, R.L. and Rosenzweig, A.C. 77. 78.

79. 80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

(2005) Nature, 434, 177–182. Yoshizawa, K. and Shiota, Y.J. (2006) J. Am. Chem. Soc., 128, 9873–9881. Balasubramanian, R. and Rosenzweig, A.C. (2007) Acc. Chem. Res., 40, 573–580. Shiota, Y. and Yoshizawa, K. (2009) Inorg. Chem., 48, 838–845. (a) Chan, S.I., Chen, K.H.-C., Yu, S.S.-F., Chen, C.-L., and Kuo, S.S.-J. (2004) Biochemistry, 43, 4421–4430; (b) Elliott, S.J., Zhu, M., Tso, L., Nguyen, H.-H., Yip, J.H.-K., and Chan, S.I. (1997) J. Am. Chem. Soc., 119, 9949–9955. Wilkinson, B., Zhu, M., Priestley, N.D., Nguyen, H.-H.T., Morimoto, H., Williams, P.G., Chan, S.I., and Floss, H.G. (1996) J. Am. Chem. Soc., 118, 921–922. Yu, S.S.-F., Wu, L.-Y., Chen, K.H.-C., Luo, W.-I., Huang, D.-S., and Chan, S.I. (2003) J. Biol. Chem., 278, 40658–40669. Hakemian, A.S., Kondapalli, K.C., Telser, J., Hoffman, B.M., Stemmler, T.L., and Rosenzweig, A.C. (2008) Biochemistry, 47, 6793–6801. (a) Klinman, J.P. (2006) J. Biol. Chem, 281, 3013–3016; (b) Klinman, J.P. (1996) Chem. Rev., 96, 2541–2562. Prigge, S.T., Kolhekar, A.S., Eipper, B.A., Mains, R.E., and Amzel, L.M. (1997) Science, 278, 1300–1306. Prigge, S.T., Kolhekar, A.S., Eipper, B.A., Mains, R.E., and Amzel, L.M. (1999) Nat. Struct. Biol., 6, 976–983. Kunishita, A., Kubo, M., Sugimoto, H., Ogura, T., Sato, K., Takui, T., and Itoh, S. (2009) J. Am. Chem. Soc., 131, 2788–2789. (a) Yoshizawa, K., Kihara, N., Kamachi, T., and Shiota, Y. (2006) Inorg. Chem., 45, 3034–3041; (b) Kamachi, T., Kihara, N., Shiota, Y., and Yoshizawa, K. (2005) Inorg. Chem., 44, 4226–4263. Crespo, A., Mart, M.A., Roitberg, A.E., Amzel, L.M., and Estrin, D.A. (2006) J. Am. Chem. Soc., 128, 12817–12828. Hong, S., Huber, S.M., Gagliardi, L., Cramer, C.C., and Tolman, W.B. (2007) J. Am. Chem. Soc., 129, 14190–14192.

91. (a) Allendorf, M.D., Spira, D.J., and

92.

93.

94.

95. 96.

97. 98.

99. 100.

101. 102. 103.

104. 105.

106.

Solomon, E.I. (1985) Proc. Natl. Acad. Sci. U.S.A., 82, 3063–3067; (b) Ryden, L. and Bjork, L. (1976) Biochemistry, 15, 3411–3417; (c) Zaitseva, I., Zaitsev, V., Card, G., Moshkov, K., Bax, B., Ralph, A., and Lindley, P. (1996) J. Biol. Inorg. Chem., 1, 15–23; (d) Messerschmidt, A., Luecke, H., and Huber, R. (1993) J. Mol. Biol., 230, 997–1014. Messerschmidt, A., Ladenstein, R., Huber, R., Bolognesi, M., Avigliano, L., Petruzzelli, R., Rossi, A., and Finazzi-Agr´o, A. (1992) J. Mol. Biol., 224, 179–205. Cole, A.P., Root, D.E., Mukherjee, P., Solomon, E.I., and Stack, T.D.P. (1996) Science, 273, 1848–1850. Kharacsh, M.S. and Tawney, P.O. (1941) J. Am. Chem. Soc., 63, 2308–2315. Nakamura, E. and Mori, S. (2000) Angew. Chem. Int. Ed., 39, 3750–3771. Perlmutter, P. (1992) Conjugate Addition Reactions in Organic Synthesis, Pergamon, Oxford. Normant, J.F. and Alexakis, A. (1981) Synthesis, 841–870. Klunder, J.M. and Posner, G.H. (1991) in Comprehensive Organic Synthesis, Vol. 3 (eds B.M.Trost and I. Fleming), Pergamon, Oxford, pp. 207–239. Marshall, J.A. (1989) Chem. Rev., 89, 1503–1511. Bertz, S.H., Cope, S., Murphy, M., Ogle, C.A., and Taylor, B.J. (2007) J. Am. Chem. Soc., 129, 7208–7209. Ullmann, F. and Bielecki, J. (1901) Chem. Ber., 34, 2174. Ullmann, F. (1903) Chem. Ber., 36, 2389. Beletskaya, I.P. and Cheprakov, A.V. (2004) Coord. Chem. Rev., 248, 2337–2364. Ley, S.V. and Thomas, A.W. (2003) Angew. Chem. Int. Ed., 42, 5400–5449. Hassan, J., Svignon, M., Gozzi, C., Schulz, E., and Lemaire, M. (2002) Chem. Rev., 102, 1359–1470. Canty, A.J. (2002) in Handbook of Organopalladium Chemistry, Vol. 1 (ed. E.I.Negishi), John Wiley & Sons, Inc., New York, p. 189.

55

56

2 The Bioinorganic and Organometallic Chemistry of Copper(III) 107. Hartwig, J.F. (2008) Nature, 455, 108.

109.

110. 111.

112.

113. 114. 115.

116.

117.

118. 119.

120.

121.

122.

123.

124.

125.

126.

314–322. Willert-Porada, M.A., Burton, D.J., and Baenziger, N.C. (1989) J. Chem. Soc., Chem. Commun, 1633–1634. Naumann, D., Roy, T., Tebbe, K.F., and Crump, W. (1993) Angew. Chem. Int. Ed. Engl., 32, 1482–1483. Eujen, R., Hoge, B., and Brauer, D.J. (1996) J. Organomet. Chem., 519, 7–20. Nakamura, E., Mori, S., and Morokuma, K. (1997) J. Am. Chem. Soc., 119, 4900–4910. Dorigo, A.E., Wanner, J., and Schleyer, P.R. (1995) Angew. Chem. Int. Ed., 34, 476–478. Snyder, J.P. (1995) J. Am. Chem. Soc., 117, 11025–11026. Hu, H. and Snyder, J.P. (2007) J. Am. Chem. Soc., 129, 7210–7211. Bertz, S.H., Cope, S., Dorton, D., Murphy, M., and Ogle, C.A. (2007) Angew. Chem. Int. Ed., 46, 7082–7085. Grtner, T., Henze, W., and Gschwind, R.M. (2007) J. Am. Chem. Soc., 129, 11362–11363. Bartholomew, E.R., Bertz, S.H., Cope, S., Murphy, M., and Ogle, C.A. (2008) J. Am. Chem. Soc., 130, 11244–11245. Karlstr¨om, A.S.E. and B¨ackvall, J.-E. (2001) Chem. Eur. J., 7, 1981–1989. Kelkar, A.A., Patil, N.M., and Chaudhari, R.V. (2002) Tetrahedron Lett., 43, 7143–7146. Okano, K., Tokuyama, H., and Fukuyama, T. (2003) Org. Lett., 5, 4987–4990. Wu, Y.J., He, H., and L’Heureux, A. (2003) Tetrahedron Lett., 44, 4217–4218. Klapars, A., Huang, X.H., and Buchwald, S.L. (2002) J. Am. Chem. Soc., 124, 7421–7428. Marcoux, J.F., Doye, S., and Buchwald, S.L. (1997) J. Am. Chem. Soc., 119, 10539–10540. Wolter, M., Nordmann, G., Job, G.E., and Buchwald, S.L. (2002) Org. Lett., 4, 973–976. Bates, C.G., Gujadhur, R.K., and Venkataraman, D. (2002) Org. Lett., 4, 2803–2806. Beletskaya, I.P., Sigeev, A.S., Peregudov, A.S., and Petrovskii,

127. 128.

129.

130.

131.

132.

133.

134. 135.

136. 137.

P.V. (2003) Tetrahedron Lett., 44, 7039–7041. Van Allen, D. and Venkataraman, D. (2003) J. Org. Chem., 68, 4590–4593. Finet, J.-P., Fedorov, A.Y., Combes, S., and Boyer, G. (2002) Curr. Org. Chem., 6, 597–626. (a) Chan, D.M.T., Monaco, K.L., Wang, R.-P., and Winters, M.P. (1998) Tetrahedron Lett., 39, 2933–2936; (b) Evans, D.A., Katz, J.L., and West, T.R. (1998) Tetrahedron Lett., 39, 2937–2940; (c) Lam, P.Y.S., Clark, C.G., Saubern, S., Adams, J., Winters, M.P., Chan, D.M.T., and Combs, A. (1998) Tetrahedron Lett., 39, 2941–2944; (d) Liu, Y., Yan, W., Chen, Y., Petersen, J.L., and Shi, X. (2008) Org. Lett., 10, 5389–5392. (a) Dodonov, V.A., Gushchin, A.V., and Brilkina, T.G. (1984) Zh. Obshch. Khim., 54, 2157–2158; (b) Dodonov, V.A., Gushchin, A.V., and Brilkina, T.G. (1985) Zh. Obshch. Khim., 55, 2514–2519; (c) Dodonov, V.A., Starostina, T.I., Kuznetsova, Y.L., and Gushchin, A.V. (1995) Russ. Chem. Bull., 44, 151–153. (a) Abramovitch, R.A., Barton, D.H.R., and Finet, J.-P. (1988) Tetrahedron, 44, 3039–3071; (b) Barton, D.H.R., Donnelly, D.M.X., Finet, J.-P., and Guiry, P.J. (1991) J. Chem. Soc., Perkin Trans. 1, 2095–2102; (c) Elliott, G.I. and Konopelski, J.P. (2001) Tetrahedron, 57, 5683–5705. Lam, P.Y.S., Deudon, S., Averill, K.M., Li, R., He, M.Y., DeShong, P., and Clark, C.G. (2000) J. Am. Chem. Soc., 122, 7600–7601. Lam, P.Y.S., Vincent, G., Bonne, D., and Clark, C.G. (2002) Tetrahedron Lett., 43, 3091–3094. Godula, K. and Sames, D. (2006) Science, 312, 67–72. Costas, M., Mehn, M.P., Jensen, M.P., and Que, L. (2004) Chem. Rev., 104, 939–986. Shilov, A.E. and Shul’pin, G.B. (1997) Chem. Rev., 97, 2879–2932. Giri, R., Chen, X., and Yu, J.-Q. (2005) Angew. Chem. Int. Ed., 44, 2112–2115.

References 138. Desai, L.V., Hull, K.L., and Sanford,

139.

140. 141. 142. 143. 144.

145.

146.

147.

148.

M.S. (2004) J. Am. Chem. Soc., 126, 9542–9543. Chen, X., Hao, X.-S., Goodhue, C.E., and Yu, J.-Q. (2006) J. Am. Chem. Soc., 128, 6790–6791. Yu, J.-Q., Giri, R., and Chen, X. (2006) Org. Biomol. Chem., 4, 4041–4047. Uemura, T., Imoto, S., and Chatani, N. (2006) Chem. Lett., 35, 842–843. Brasche, G. and Buchwald, S.L. (2008) Angew. Chem. Int. Ed., 47, 1932–1934. Ueda, S. and Nagasawa, H. (2008) Angew. Chem. Int. Ed., 47, 6411–6413. Ribas, X., Jackson, D.A., Donnadieu, B., Mah´ıa, J., Parella, T., Xifra, R., Hedman, B., Hodgson, K.O., Llobet, A., and Stack, T.D.P. (2002) Angew. Chem. Int. Ed., 41, 2991–2994. Xifra, R., Ribas, X., Llobet, A., Poater, A., Duran, M., Sol`a, M., Stack, T.D.P., Benet-Buchholz, J., Donnadieu, B., Mah´ıa, J., and Parella, T. (2005) Chem. Eur. J., 11, 5146–5156. (a) Golub, G., Cohen, H., Paoletti, P., Bencini, A., Messori, L., Bertini, I., and Meyerstein, D. (1995) J. Am. Chem. Soc., 117, 8353–8361; (b) Meyerstein, D. (1999) Coord. Chem. Rev., 185-186, 141–147; (c) Bernhardt, P.V. (1997) J. Am. Chem. Soc., 119, 771–774; (d) Bernhardt, P.V. (2001) Inorg. Chem., 40, 1086–1092. Ribas, X., Calle, C., Poater, A., Xifra, R., Casitas, A., G´omez, L., Parella, T., Benet-Buchholz, J., Schweiger, A., Mitrikas, G., Sol`a, M., Llobet, A., and Stack, T.D.P., in preparation. Huffman, L.M. and Stahl, S.S. (2008) J. Am. Chem. Soc., 130, 9196–9197.

149. Casitas, A., Huffman, L.M., Costas, M.,

150.

151.

152.

153.

154.

155.

156.

157. 158.

159. 160. 161.

Stahl, S.S., and Ribas, X., in preparation. Ribas, X., Donnadieu, B., Jackson, D., Hodgson, K.O., Hedman, B., Stack, T.D.P., and Llobet, A. (2001) J. Inorg. Biochem., 86, 399. Casitas, A., Donnadieu, B., Jackson, D., Hodgson, B.K.O., Corbella, M., Stack, T.D.P., Llobet, A., and Ribas, X., in preparation. Fusi, V., Llobet, A., Mah´ıa, J., Micheloni, M., Paoli, P., Ribas, X., and Rossi, P. (2002) Eur. J. Inorg. Chem., 987–990. Ribas, X., Xifra, R., Parella, T., Poater, A., Sol`a, M., and Llobet, A. (2006) Angew. Chem. Int. Ed., 45, 2941–2944. Furuta, H., Maeda, H., and Osuka, A. (2000) J. Am. Chem. Soc., 122, 803–807. Ban, I., Sudo, T., Taniguchi, T., and Itami, K. (2008) Org. Lett., 10, 3607–3609. Smart, L. and Moore, E. (1992) Solid State Chemistry: An Introduction, Chapman & Hall, London. Guo, Y., Langlois, J.-M., and Goddard, W.A.III (1988) Science, 239, 896–899. Choy, J.-H., Kim, D.-K., Hwang, S.-H., and Park, J.-C. (1995) J. Am. Chem. Soc., 117, 7556–7557. Murthy, N.K. and Murthy, B.S. (1995) Talanta, 42, 101–103. Cho, A. (2008) Science, 320, 870–871. Takahashi, H., Igawa, K., Arii, K., Kamihara, Y., Hirano, M., and Hosono, H. (2008) Nature, 453, 376–378.

57

59

3 Chemical Protein Modiﬁcation Gonc¸ alo J. L. Bernardes, Justin M. Chalker, and Benjamin G. Davis

3.1 Introducing Diversity by Posttranslational Modiﬁcation

The proteome of a speciﬁc organism is encoded by its genome, giving rise to a number of proteins that are responsible for the majority of biological events. With the ﬁrst draft of the human genome, only 30 000 genes were found [1] – just twice the number of genes as nematodes [2]. The apparent disparity between our paucity of genes and functional complexity is a central problem in biology. At least one factor reconciling gene economy and phenotypical complexity is the posttranslational processing of proteins [3, 4]. Indeed, the structural and functional variety of such modiﬁcations is staggering. A selection of common posttranslational modiﬁcations (PTMs) is shown in Figure 3.1 [3, 4]. The covalent attachment of complex carbohydrates to proteins is a widespread form of PTM [5]. Indeed, approximately 1% of all genes are used to access such modiﬁcations [6]. Glycosylation is involved in several biological processes such as cellular differentiation, fertilization, and immune response [7]. Phosphorylation is a ubiquitous and essential modiﬁcation in the control of cellular function [8]. Acylation of lysine [9] and methylation of lysine [10] and arginine [11] side chains are key markers in the control of gene expression. Polyprenylation of cysteine is critical in membrane localization and protein–protein interactions [12]. Oxidation of cysteine is a common way of cross-linking peptide chains to increase the stability of protein architecture and is critical in cellular redox regulation [13]. Hydroxylation is also an important form of PTM and is key for correct maturation of collagen ﬁbers [14]. These key modiﬁcations of proteins highlight nature’s repertoire of protein processing enzymes and the resulting diversity of macromolecules that arise from a comparatively restricted genome.

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

60

3 Chemical Protein Modiﬁcation

A

Genetic

Processing

Control

Enzymes

F D B

DNA

C

Protein

PTM Proteins E

A. Phosphorylation

B. N-Acetylation and methylation

O O P O O

O O P O O N H

N H

O

NHAc

N H

O

N H

O

N -AcLys

N H

O

N -MenLys n = 1–3

OH O

O O

O HO

AcHN O N H

H N O

AcHN

R

N H

O

R = CH3 GalNAc-O -Thr R = H GalNAc-O -Ser

S N H

N H

GlcNAc-N -Asn

O

Man-C -Trp

CysS-SCys

N H

NH3 R2

HO

NH2 N O

O

3-OH-Asn

Figure 3.1

n = 2 Farnesyl-Cys n = 3 Geranylgeranyl-Cys

R1

O

O

n

F. Hydroxylation

S N H

O

O

HO N H

O

N -Me2Arg

H N

HO O HO HO OH

O

N H

O

D. Polyprenylation

E. Cysteine oxidation S

NH

N -Men Arg n = 1–2

C. O -, N -, and C -Glycosylation O

NHMe

NH

O -phospho-Ser O -phospho-Tyr

HO

NMen H(2-n) MeHN

H2N

NMenH(3-n)

R1, R2 = OH or H 3- or 4-OH-Pro

N H

O

5-OH-Lys

Examples of posttranslational modiﬁcations.

3.2 Chemistry: A Route to Modiﬁed Proteins

Given the ubiquity and variety of natural covalent modiﬁcations of proteins, it is important to understand their precise role. Since modiﬁed proteins are often

3.4 Traditional Methods for Protein Modiﬁcation

difﬁcult to obtain from pure sources or from recombinant translation systems, chemical access can resolve supply issues. Moreover, in certain cases it may not even be necessary to synthesize a natural PTM. Often a chemical mimic will sufﬁce [15]. The capacity to access these modiﬁed proteins facilitates the understanding of their natural function and gives insight useful in the manipulation of these systems. Modiﬁed proteins have also found increasing use in therapeutic strategies, including vaccine constructs in campaigns against HIV [16, 17], cancer [18], and malaria [19]. Precisely modiﬁed proteins for these purposes will allow a more clear assessment of therapeutic efﬁcacy than the mixtures that are typically employed. Well-deﬁned protein therapeutics will also simplify regulatory requirements and batch-to-batch reproducibility. In order to access modiﬁed proteins for the purposes described above, the reactions employed must be selective for the site of interest. The inherent challenges for such modiﬁcations are discussed next.

3.3 Challenges in Chemical Protein Modiﬁcation

Covalent protein modiﬁcation is an attractive strategy for access, study, and functional modulation of proteins. Central to precise chemical modiﬁcation of proteins is the need for reactions that are compatible with protein systems. This is a daunting challenge in organic chemistry. The reaction must proceed efﬁciently in buffered aqueous media at or near room temperature and select a single reactive residue among several competing functional groups. The reaction is most useful if it is fast and quantitative, eliminating the need for complicated and tedious puriﬁcation procedures. The fragile tertiary structure of proteins adds an additional level of complexity to chemical modiﬁcation. Despite these challenges, many methods for chemical modiﬁcation of proteins are available. The development of these techniques was a challenge to the chemist and biochemist alike and much was learned in both chemistry and biology. What follows is a survey of both traditional and emerging techniques for chemical modiﬁcation of proteins. Highlighted are key advances in reaction and protein engineering that continues to raise the bar for biochemical endeavors.

3.4 Traditional Methods for Protein Modiﬁcation

Most traditional methods for protein modiﬁcation rely on electrophilic reagents that react with nucleophilic side chains. Amino groups on lysine side chains have been frequently chosen as a target [20]. Glutamic and aspartic acid are also common targets for amidation of their carboxylate side chains [21]. Cysteine has also received considerable attention due to its low abundance in naturally occurring

61

62

3 Chemical Protein Modiﬁcation

proteins [22, 23], and the unique reactivity of the sulfhydryl side chain [24, 25]. Chemical modiﬁcation reactions at these amino acid side chains are discussed below. 3.4.1 Lysine Modiﬁcation

Many bioconjugation methods rely on lysine [20]. The side chain is abundant, nucleophilic, and often exposed to the solvent. Since there are many lysine residues in most proteins, reactions are generally incomplete and result in mixtures of modiﬁed proteins. Still, these methods ﬁnd wide use in applications where regioselectivity is not an immediate concern. A number of these strategies are summarized below. 3.4.1.1 Activated Esters N-hydroxy succinimide (NHS) esters and homobifunctional NHS derivatives such as disuccinimidyl glutarate (DSG) and disuccinimidyl suberate (DSS), along with the related bifunctional squarate derivatives, have been extensively used to convert the side chain amino group of lysine residues to an amide bond (Scheme 3.1) [20, 26]. While the synthetic simplicity and availability of NHS esters makes this a popular strategy, the efﬁciency of the coupling reaction is normally low. Poor regioselectivity in the conjugation and potential immunogenicity of the linker of the modiﬁed proteins are major drawbacks [27, 28]. Human serum albumin (HSA) can be modiﬁed with LewisB hexasaccharide using DSG and DSS as cross-linkers, derivatizing 25 of the 59 potential sites of reaction [29]. For the reported modiﬁcation, an average of 40 equiv. of sugar per lysine residue was necessary to achieve such levels of incorporation. When diethyl squarate is used instead, the average incorporation level of glycan structures usually does not exceed 25% [30, 31]. 3.4.1.2 Isocyanates and Isothiocyanates Isocyanates and isothiocyanates have also been successfully applied for linking conjugates to proteins [20]. Both reagents react with the amino side chain of lysine residues to form substituted urea and thiourea derivatives, respectively. Typically, phosgene or thiophosgene is reacted with an amine and excess base to form the corresponding isocyanate or isothiocyanate, which in turn reacts with nucleophilic lysine side chains. This method has been used to conjugate various glycans to bovine serum albumin in the investigation of HIV vaccine candidates [16]. 3.4.1.3 Reductive Alkylation The strategies described above neutralize the charged amine side chains of lysine. This modiﬁcation changes the overall charge of the protein with potential alteration of solubility and other protein properties. If a reductive alkylation strategy is used instead, such effects can be avoided. An aldehyde and the amino group of a lysine residue condense to an imine that can be reduced by sodium cyanoborohydride

3.4 Traditional Methods for Protein Modiﬁcation

O

O HN

R

R

O

O

RO X

NH2

N H X = Spacer

O

N H

O

OEt

Squarate derivatives

NHS esters

N H

O Lys O

N H

O

O O

O HN

O N

R-NH2

Scheme 3.1

H N

O

HN

O

RO X

O N

O

N H

63

n

N H

O O

R

n

O

N

O

n = 1 DSG n = 3 DSS

O

Strategies for lysine modiﬁcation using NHS esters and squarate derivatives.

to afford the corresponding secondary amine. These conditions are harsh for proteins and suffer from low conversion levels. When applied to carbohydrates, the amine group of a lysine residue reacts with a reducing sugar in its open-chain aldehyde form. This creates an additional problem, since the carbohydrate at the reducing terminus will be in its open-chain form, which alters the conformation compared to naturally occurring glycosylated proteins (Scheme 3.2a). To overcome this problem, aldehyde-terminated spacer arms have been developed. In this strategy, an oligosaccharide carrying a spacer at the anomeric center with a terminal double bond is subjected to ozonolysis to yield the corresponding aldehyde (Scheme 3.2b) [32]. This method has been used, for example, in the attachment of Globo-H, a tumor-associated antigen, to an immunogenic protein carrier [33]. 3.4.1.4 IME Reagents Use of 2-imino-2-methoxyethyl reagents (IME) have also been successfully applied in lysine modiﬁcation through the formation of an amidine linkage. In particular, 2-imino-2-methoxy-1-thioglycosides have proven to be useful for the selective modiﬁcation of lysine side chains on polypeptides (Scheme 3.3) [34, 35]. This method was used in the galactosylation of a rhamnosidase [36]. The resulting neoglycoprotein was selectively targeted in vivo as a result of glycosylation.

O

64

3 Chemical Protein Modiﬁcation

OH O

OH OH

OH

HO

HO

OH OH HO

O

H N

NaBH3CN

H2N

Protein Protein

(a) OH O HO

OH O

O3

H N

HO

H2N

H N O

O

Protein

O NaBH3CN OH O HO

H N

(b)

N H

O

Protein

Scheme 3.2 Reductive alkylation of lysine with carbohydrates; (a) direct coupling and (b) spacer route.

OH O

OH O

NH Buffer pH 8–9

S

HO

O

HO

+

H2N

NH S

N H

Protein

Protein

Scheme 3.3

IME strategy for modiﬁcation of lysine side chains with glycans.

3.4.2 Glutamic and Aspartic Acid Modiﬁcation

Several reactions have been described for the chemical modiﬁcation of carboxylic side chains of glutamic and aspartic acids. Amidation of carboxylate groups using an amine containing conjugate and a water-soluble coupling carbodiimide reagent such as 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC), is the most common method in protein modiﬁcation (Scheme 3.4) [37]. 3.4.3 Cysteine

Cysteine has been one of the most extensively studied amino acid side chains for protein modiﬁcation and many conjugation methods are available [24, 25]. The low natural abundance <1.6% [22, 23] and accessible pKa (∼8–9) makes cysteine a convenient target for chemoselective conjugation. No charge balance problems

3.4 Traditional Methods for Protein Modiﬁcation

R

HN CO2H

N

n

N H

C

N

N

65

O

n

EDC R

N H

NH2

O

O

n = 1 Asp n = 2 Glu Scheme 3.4 Aspartic/glutamic acid modiﬁcation with 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide.

HN

R

N O

O I

N H

O SH

Iodoacetamides N H

S N H

Vinyl sulfones

O

N H

SO2R

S

Aminoethylation

O

NMen H(3- n)

N H

+

Cl

O

Maleimides

Cys

Br

Scheme 3.5

S

N R

R

O

SO2R

O

O

S N H

R

O

O NMen H(2-n )Cl − Lys and MeLys analogs or

+

NMe3 Br − Base

Direct alkylation of cysteine.

are generated when modifying cysteine, since the thiol side chain as well as the resulting thioether or disulﬁde linkages are typically noncharged. 3.4.3.1 Alkylation The direct alkylation of cysteine with suitable electrophiles is a common strategy for cysteine modiﬁcation. It is possible, by careful control of pH, to selectively modify the sulfhydryl group in the presence of other nucleophilic side chains such as lysine or histidine [24, 25]. Recent studies have delineated considerations regarding electrophile selection [38]. The most common methods used for the direct alkylation of cysteine are summarized in Scheme 3.5 Reaction of the sulfhydryl group of cysteine residues with α-halocarbonyls is one of the most common methodologies for alkylating cysteine residues on protein surfaces [24, 25]. In particular, iodoacetamides have been very useful for the covalent addition of complex oligosaccharides to proteins [39]. This approach has

66

3 Chemical Protein Modiﬁcation

S K9C

NMenH(3-n)

Histone H3 (K9C, C110A)

Figure 3.2

Incorporation of methyl lysine analogs into histone H3 by aminoethylation.

allowed the construction of homogeneous glycosylated erythropoietin [40] and the modulation of protein thermal stability by glycosylation [41]. Competing alkylation of lysine and histidine residues is often observed, but can be resolved by the use of chloroacetamides [42]. Conjugate addition of cysteine to Michael acceptors such as maleimides or vinyl sulfones is also a reliable way to selectively modify cysteine [24, 25]. Recent examples of the use of maleimides include the glycosylation of myoglobin for oxygen afﬁnity enhancement [43] and the conjugation of carbohydrate epitopes to protein carriers in synthetic cancer vaccine strategies [44]. Although less common than maleimides, vinyl sulfones have also been used in cysteine alkylation, for example, in the conjugation of enzymes to a ubiquitin-like protein [45]. Site-selective PEGylation at reduced disulﬁdes with bis-(vinyl sulfones) has also been reported [46]. Aminoethylation of cysteine has long been used as a reliable way to convert cysteine to lysine analogs [47]. An enzyme containing a deactivating lysine to cysteine mutation can be chemically mutated to a lysine analog by aminoethylation, regenerating catalytic activity [48]. Recent access to methyl lysine analogs was also accomplished using aminoethylation of cysteine [49]. This methodology gives access to mimics of an important class of PTMs in histone proteins, which are involved in encoding the epigenetic status of the cell (Figure 3.2) [10]. 3.4.3.2 Disulﬁdes Conjugation at cysteine by formation of disulﬁde linkages is a useful and easy method for the selective modiﬁcation of cysteine residues [24, 25]. Disulﬁde-linked protein conjugates can be achieved either by simple air oxidation [24], disulﬁde exchange [50], or by activated thiol-selective electrophiles such as glyco-methanethiosulfonates (glyco-MTSs) [51], glyco-phenylthiosulfonates (glyco-PTSs) [52], glyco-5-nitropyridine-2-sulfenyl (glyco-pNPys) [53], and glyco-selenenylsulﬁde (glyco-SeS) (Scheme 3.6) [54]. The addition of activated thiol electrophiles to a single cysteine protein mutant has resulted in a number of synthetic, homogeneous glycoproteins. Importantly, the high rate of disulﬁde formation allows for the use of small amounts of modifying reagents. The control offered by this approach has even allowed the construction of a trisulﬁde-linked glycoprotein [55]. Recently, Lawesson’s reagent has been described as an efﬁcient reagent for the conversion of reducing sugars and allylic alcohols to the corresponding thiol derivatives [56, 57]. The product thiols were used to site-speciﬁcally glycosylate and prenylate a single cysteine protein mutant by disulﬁde formation (Scheme 3.7).

3.4 Traditional Methods for Protein Modiﬁcation

SH N H

Cysteine– specific sulfenylation N H

O

R–SH Air oxidation

Scheme 3.6

R

OH

S S

R

S

S

R

Disulfide exchange

R

O

R S X Activated reagent X = SO2Me (MTS), SO2Ph (PTS), SePh (SeS), I, Br, Cl pNPys = NO2 S N

Methods for disulﬁde formation at cysteine. a

Lawesson’s reagent (LR)

R

S

b

SH PhSe

S

R

Protein

Protein

OH O HO HO

S

R

OH O

HOO HO

HO O HO

SH

2

OH O HO O HO

OH O HO O HO

OH O SH OH

SH

Scheme 3.7 Disulﬁde protein conjugation of oligosaccharide and prenyl thiols: (a) direct thionation using Lawessson’s reagent and (b) chemoselective disulﬁde formation using a phenylselenyl activated cysteine mutant.

3.4.3.3 Desulfurization at Cysteine Disulﬁde Contraction to Thioether The instability of disulﬁde conjugates in reducing environments has driven the search for methods to stabilize this reducible linkage. On the basis of seminal chemistry developed by Harpp and Gleason [58–60], the Davis group has used hexamethylphosphorus triamide (HMPTL) to convert a variety of disulﬁde glycoaminoacids and glycopeptides to the corresponding thioether derivatives [61]. This methodology was also used for the ﬁrst conversion of cysteine to a thioether-linked glycoprotein (Scheme 3.8). The products of this reaction are typically epimeric mixtures at the α-C of cysteine. Epimerization arises, at least in part, by formation of dehydroalanine, followed by conjugate addition of the liberated thiol. This mechanistic course was veriﬁed by 1 H-NMR as at least one productive pathway to the thioether product (Scheme 3.8). Native Chemical Ligation Native chemical ligation (NCL) is the most effective chemical route to large polypeptides [62]. On the basis of seminal ligations

67

68

3 Chemical Protein Modiﬁcation

AcO AcO

OAc O S OAc

(a)

BnO BnO

OBn O OBn

S

S

P(NMe2)3

Protein

AcO AcO

pH 9.5

P(NMe2)3 MeOD

S

O

BnO BnO

H

S = P(NMe2)3 AcHN H+

Protein

OAc

S−

OBn H

OMe

AcHN

OBn O

BnO BnO

OAc O S

OBn O

S

OBn OMe

AcHN

OMe

O O

H

H OMe

AcHN O

78

6 (b)

6.3

6.2

6.1

6.0

5.9

Time (min)

ppm

Scheme 3.8 Desulfurization of disulﬁde conjugates: (a) conversion of disulﬁde glycoprotein to the corresponding thioether derivative and (b) model desulfurization and observation of dehydroalanine formation (oleﬁn peaks) and consumption during the desulfurization reaction.

described by Wieland and coworkers in 1953 [63], NCL comprises the chemo- and regioselective reaction of a C-terminal peptide thioester and an N-terminal cysteine. Both peptides involved in NCL can be in the side chain unprotected form and the resulting product contains a native amide bond at the ligation site (Scheme 3.9a). NCL not only allows the synthesis of large proteins but also provides access to homogeneous posttranslationally modiﬁed proteins [64]. Incorporation of peptide fragments that contain preinstalled glycosylated sites into a protein backbone using NCL has been often used to produce homogeneous glycoproteins [64–66]. A semisynthetic homogeneous glycosyl phosphatidylinositol (GPI)-anchored prion protein was also successfully accessed using NCL to conjugate the GPI to a thioester containing protein (Scheme 3.9b) [67]. NCL has also provided access to the modiﬁed tails of histone proteins [68–70]. The utility of NCL in chemical protein synthesis has been reviewed in detail elsewhere and the reader is directed to several excellent reviews [64, 71–74]. Rather than reiterate these accomplishments, we instead turn focus to recent developments that take advantage of cysteine in NCL, but then convert the cysteine of the ligated product to another residue. Formally, this transformation constitutes a ligation at the ﬁnal residue.

3.4 Traditional Methods for Protein Modiﬁcation

69

O O Peptide 1

HS SR

O

Thioester

Peptide 2

Exchange

H 2N

H2N

Peptide 1

Peptide 2

S

O Cysteine

S O

N acyl transfer

HS Peptide 2

Peptide 1

N H

(a)

O

SH O

H N N H

O

Native chemical ligation

O O

OH P O OH O HO HO HO O O HO HO O HO HO

OH O OH O

O HO

+

H3N −

(b)

O

O

HO OH

O O P OC18H37

Scheme 3.9 (a) Native chemical ligation and (b) example of posttranslationally modiﬁed protein obtained by native chemical ligation; in this case, a homogeneous semisynthetic GPI prion protein.

As mentioned above, a major limitation of NCL is the inherent reliance on cysteine. Cysteine is a low-incidence residue in natural proteins and ligations at other residues are highly desirable. While auxiliaries have been used at other residues [74, 75] and direct ligation to N-terminal histidine has been reported [76], an alternative approach is to use cysteine in a ligation as usual and then convert it to another residue. A key reaction to accomplish such a transformation is the reduction of cysteine to alanine. The overall result of this transformation is a formal ligation at alanine. Moreover, alanine unlike cysteine, is an abundant amino acid side chain. Yan and Dawson introduced the hydrogenolytic desulfurization of cysteine using different palladium sources and Raney nickel

OH OH

70

3 Chemical Protein Modiﬁcation

NCL at alanine O

HS Peptide 2 N H

Peptide 1 (a)

Selective desulfurization

O

Peptide11 Raney Ni, H2 or Pd, H2 peptide or TCEP, radical initiator

O

Cysteine

Peptide 2

N H Ala

O

NCL at phenylalanine O Peptide 1

HS SR

NCL at valine O

(c)

Peptide 2 H2N

1. NCL

Peptide 2

Peptide11 2. NiCl2, NaBH4 peptide

HS SR

Peptide 2 H 2N O Penicillamine

Scheme 3.10

1. NCL 2. Desulfurization Peptide 1 TCEP radical initiator, thiol

Ph

O N H

O b-Mercapto phenylalanine

(b)

peptide Peptide11

Ph

O Phe

O Peptide 2

N H

O Val

Formal native chemical ligation at alanine, phenylalanine, and valine.

to synthesize cysteine-free peptides by NCL (Scheme 3.10a) [77]. Crich used a related nickel boride reduction of β-mercapto phenylalanine for a formal ligation at phenylalanine (Scheme 3.10b) [78]. Recently, Wan and Danishefsky reported a mild radical-based desulfurization reaction that resolves chemoselectivity issues and low yields common in metal-mediated reductions [79]. This transformation uses the water soluble-phosphine tris(2-carboxyethyl)phosphine (TCEP) and yields alanine at the ligation site (Scheme 3.10a). When NCL is performed with penicillamine followed by TCEP-mediated desulfurization, valine is obtained (Scheme 3.10c) [80, 81]. While the transformations described expand, signiﬁcantly, the use of NCL, new methods to access other amino acid residues directly from cysteine are required and yet to be developed.

3.5 Recent Innovations in Site-Selective Protein Modiﬁcation

Most reagents for protein modiﬁcation developed to date react with lysine and cysteine residues. Despite the general utility of these reagents for the construction of protein conjugates, development of novel reactions that target other amino acid side chains in a selective fashion is desirable. Below, we discuss recent progress in chemical methods for protein modiﬁcation that provide efﬁcient alternatives to the traditional modiﬁcation of nucleophilic side chains such as lysine and cysteine.

3.5 Recent Innovations in Site-Selective Protein Modiﬁcation

3.5.1 Dehydroalanine: A Useful Chemical Handle for Protein Conjugation

Dehydroalanine is a convenient handle for peptide and protein modiﬁcation. Conjugate addition of thiol nucleophiles to this α, β-unsaturated amino acid has been used to access S-linked peptide conjugates [82, 83]. Early access to dehydroalanine on proteins was accomplished by a two-step sulfonylation and elimination of the active site serine in serine proteases [84, 85]. Dehydroalanine can also be accessed by elimination of dialkylated cysteine [86]. These methods require high pH or elevated temperature and may not be compatible with any given protein. Alternative access to dehydroalanine is realized by oxidative elimination of alkylcysteine [87] or phenylselenocysteine [88, 89] containing peptides. Ribosomal synthesis of dehydroalanine containing peptides is also known [90]. Schultz and coworkers reported the genetic incorporation of phenylselenocysteine into a protein using amber codon suppression [91]. Oxidative elimination followed by thiol conjugate addition methodology was used to generate S-hexadecyl- and S-mannosylcysteine conjugates of green ﬂuorescent protein [91]. It is worth noting that the diastereoselective outcome of thiol conjugate additions to dehydroalanine is usually quite low [82, 83], and dependent on sequence and the local stereochemical environment [92]. Davis and coworkers discovered that when cysteine is treated with excess O-mesitylenesulfonylhydroxylamine (MSH) under basic conditions, dehydroalanine is rapidly formed (Scheme 3.11) [93]. This is a novel and direct method for the conversion of cysteine to dehydroalanine. Methionine also reacts with MSH, but the resulting sulﬁlimine is unstable under basic conditions or reducing conditions dithiothreitol (DTT), and decomposes to regenerate methionine. The chemical recovery of methionine may resolve issues of chemoselectivity that arise when using H2 O2 for the oxidative elimination of phenylselenocysteine [90, 91]. Conjugate addition of a suitable thiol derivative allowed site-speciﬁc glycosylation, phosphorylation, prenylation, peptide conjugation, and incorporation of methyl lysine analogs into a model protein (Scheme 3.11) [93]. The methyl lysine analogs and corresponding acyl lysine analog were later incorporated in a similar manner into the H3 histone by the Schultz laboratory [94]. The acyl mimic was further validated as a substrate for histone deacetylase 3. Alkylcysteines also react with MSH, allowing regeneration of dehydroalanine from a previously synthesized thioether modiﬁed protein [93]. This functional switch is a consequence of the unique reactivity of both cysteine and alkylcysteines to MSH, and allows controlled conversion between different protein modiﬁcations. 3.5.2 Metal-Mediated Protein Modiﬁcation

Reactions mediated by transition metals have the potential to dramatically expand the toolkit for site-selective protein modiﬁcation [95]. Organometallics have had an enormous impact on synthetic organic chemistry over the past decades, allowing

71

72

3 Chemical Protein Modiﬁcation

O S O NH2 O (MSH) HS

R-SH Conjugate RS addition

pH 8.0, 20 min Cys

Dha O HO P

SH

OH

R-SH

OH O

HO HO

Homogeneous modified proteins

SH

HO HO HO

AcHN

OH O SH

SH

H(3-n )Men N +

2

O

O

HO NH2

N H

SH H N

SH

O OH

O

Scheme 3.11 Oxidative elimination of cysteine to dehydroalanine and conversion to posttranslationally modiﬁed proteins.

new transformations with high efﬁciency and chemoselectivity [96]. Many of these transition metal–mediated reactions could potentially be adapted to protein modiﬁcation, where high efﬁciency and selectivity are major challenges. While some promising advances have been reported [95], the full potential of transition metal–mediated reactions for protein modiﬁcation is yet to be realized due to the inevitable challenges of using proteins as substrates. Among these challenges, efﬁcient metal-mediated reactions in water are required [97–99]. Moreover, these reactions must proceed readily in the presence of hundreds of side chains that may compete for a coordination site at the metal center. Finally, the reaction must be efﬁcient at or near room temperature and over a narrow pH range. These are signiﬁcant challenges to the organometallic and protein chemist alike and make successful conjugations all the more worthy of comment. It is to these modiﬁcations that we now turn. 3.5.2.1 Modiﬁcation at Natural Residues Reductive Desulfurization of Cysteine The reductive desulfurization of cysteine was discussed above in the context of NCL (Section 3.4.3.3.2). The conversion of cysteine to alanine by palladium or nickel mediated hydrogenolysis is one of the earliest deployments of transition metals in protein modiﬁcation [77]. While this reaction has expanded the scope of NCL, other modiﬁcations that form new bonds present many new and exciting opportunities. These modiﬁcations are discussed next.

3.5 Recent Innovations in Site-Selective Protein Modiﬁcation

CO2R

NH N

R′

Rh N H

OR R′

O

O

N H

O +

Rh2(OAc)4 t BuNHOH (pH 6.0) NH CO2R

N2 OR R′ O Scheme 3.12

R′ R = -(CH2CH2O)3CH3 R′ = -CH=CHPh

N H

O

Tryptophan modiﬁcation with Rh carbenoids.

Rhodium Carbenoid Alkylation of Tryptophan One of the earliest reports of metal-mediated protein modiﬁcation that results in productive bond formation is tryptophan alkylation with rhodium carbenoids (Scheme 3.12) [100]. The reactive carbenoid species was generated in situ via catalytic degradation of an α-diazo ester and Rh2 (OAc)4 . Despite competing insertion of the carbenoid into water, the reaction was successfully applied to the site-selective modiﬁcation of tryptophan in several model proteins [100, 101]. Low pH (<3.5) was required for early examples, but milder conditions (pH 6.0) have since been enabled by the addition of N-(tert-butyl)hydroxylamine [101]. The ligation results in a mixture of both N- and C-2 alkylation of in the indole ring, but it is nonetheless one of the ﬁrst examples of successful metal-mediated carbon–carbon bond formation on protein surfaces and one of the few methods to selectively modify tryptophan. This technique has since been applied to the analysis of the tryptophan domain of β-lactoglobulin using ﬂuorescent probes [102]. Palladium-Catalyzed Allylation of Tyrosine Palladium-catalyzed reactions in water are well known [98]. Since aqueous media is one of the prerequisites for protein modiﬁcation, it is perhaps not surprising that the use of Pd was also explored for in protein ligation strategies. Francis and coworkers reported the use of π-allylpalladium complexes for the selective O-alkylation of surface exposed tyrosine residues with a rhodamine dye and lipid anchors [103]. Allyl acetates and carbamates are converted to electrophilic π-allylpalladium species using Pd(OAc)2 and triphenylphosphine tris(sulfonate). At pH 9, these complexes selectively O-alkylate tyrosine (Scheme 3.13). This strategy is useful for lipidation since the carbonate leaving group (X in Scheme 3.13) can impart water solubility to the otherwise insoluble hydrophobic chain. This ‘‘solubility switch’’ allows tyrosine

73

74

3 Chemical Protein Modiﬁcation

R

X

Pd(OAc)2 TPPTS

R + X−

Phosphate buffer pH 9.0

LnPd+

R = Rhodamine, farnesyl, C17

OH

X = OCONH(CH2)2SO3Na

N H

O

O

N H Scheme 3.13

R

O

Pd-catalyzed allylation of tyrosine.

selective lipidation – a well-deﬁned modiﬁcation for anchoring proteins to membranes [12]. In other applications, this tyrosine selective alkylation has been used to study the local polarity of protein tyrosine domains [104]. 3.5.2.2 Iridium-Catalyzed Reductive Alkylation of Lysine Reductive alkylation of lysine residues on a protein surface was performed using a Cp*-Ir bipyridyl complex in combination with formate as the stoichiometric reducing agent (Scheme 3.14) [105]. This system is a mild alternative to sodium cyanoborohydride, the typical reagent in reductive alkylations of amines. The active Ir-hydride species is generated in situ from sodium formate at physiological pH and room temperature. This reactive species efﬁciently reduces the imine formed by condensation of the amine side chain with aliphatic and aromatic aldehydes. 3.5.2.3 Modiﬁcation of Unnatural Residues Cu-Catalyzed [3 + 2] Cycloaddition of Azides and Alkynes The Cu(I)-promoted [3 + 2] cyclization of azides and alkynes [106, 107], a catalyzed version of the Huisgen cycloaddition reaction [108], is the staple reaction in the ‘‘click’’ chemistry [109] arsenal. The reaction results in the regioselective formation of stable 1,4-disubstituted 1,2,3-triazoles (Scheme 3.15) and several studies have shed light on the mechanism of this transformation [110–112]. This reaction is now ubiquitous in bioconjugation [113, 114] and much attention has been given to the installation of azide and alkyne functionalities into proteins.

3.5 Recent Innovations in Site-Selective Protein Modiﬁcation

OMe

2+ R

N NH2

HN

Ir H2 O

N OMe HCO2Na

N H

Phosphate buffer pH 7.4 O

O Lys

R Scheme 3.14

N

N N

N H

O

R = Alkyl, aryl, PEG

H

Reductive alkylation of proteins with aldehydes and Ir-hydrides.

R1

+ R2 Scheme 3.15

Cu(I)

N

N

1 N R

R2 Regioselective Cu(I)-catalyzed Huisgen cycloaddition.

These reactive chemical handles can be chemically installed through a linker coupled to lysine or cysteine [115, 116]. Direct conversion of lysines to azides is also possible by diazo transfer [117]. The cycloaddition is mild and tolerant of a number of different functionalities. Sharpless and Finn reported the ﬁrst use of the Cu(I)-catalyzed [3 + 2] cycloaddition between azides and alkynes on a protein by labeling a viral capsid [115]. Since, the reaction has found broad application including activity-based protein proﬁling [118], and cell surface labeling [119]. Other conjugates that were attached to proteins using this two-step process consisting of conjugation of a coupling partner bearing an azide or alkyne functionality followed by subsequent Cu(I)-catalyzed [3 + 2] cycloaddition include carbohydrates [116, 120, 121], peptides [116], polymers [116, 122], and even other proteins [116]. The application of this chemistry for more precise protein modiﬁcation was enabled by efﬁcient methods for the genetic incorporation of azides and alkynes. Azidohomoalanine (Aha) [123] and homopropargylglycine [124], for instance, are methionine surrogates and are efﬁciently incorporated into proteins by methionine autotrophic strains of Escherichia coli. This strategy has allowed labeling of expressed proteins on the surface of E. coli cells by biotinylation or ﬂuorescent tag attachment (Scheme 3.16) [125, 126]. Alternatively, Schultz and coworkers have reported a tyrosyl-tRNA/tRNA-synthetase pair capable of recognizing a nonsense amber codon and incorporating azide-containing amino acids into proteins. This system was used to efﬁciently introduce p-azidophenylalanine into speciﬁc sites of superoxide dismutase. These residues were then efﬁciently labeled with ﬂuorophores and PEG polymer chains (Scheme 3.16) [127, 128]. A recent report by Chin describes a

75

76

3 Chemical Protein Modiﬁcation N N N

N3 Cu(I)

R

R Biosynthetic incorporation

N N N Cu(I)

R

R N3 R = fluorophore, PEG, or biotin

Scheme 3.16

Cu(I)-catalyzed triazole formation on proteins.

related strategy for the incorporation of aliphatic azides and alkynes [129]. In this report, a pyrrolysyl tRNA/tRNA-synthetase pair was used to incorporate unnatural azide- and alkyne-containing amino acids into proteins in response in response to the amber codon. Palladium-Catalyzed Carbon–Carbon Bond Formation Palladium-catalyzed crosscoupling of aryl halides with organometallic cross-coupling partners is among the most useful methods for carbon–carbon bond formation in organic synthesis [130]. Progress in aqueous cross-coupling [131], particularly in Suzuki– Miyaura [132, 133] and Sonogashira [134, 135] cross-coupling on peptides, has prompted many chemical biologists to adapt these transformations to protein modiﬁcation. The ﬁrst criteria for cross-coupling on protein surfaces is an efﬁcient method for incorporation of a cross-coupling partner. Accordingly, several strategies for the incorporation of aryl halides have been described. Tirrell and coworkers have reported the incorporation of p-bromophenylalanine [136] and p-iodophenylalanine [137] into proteins as phenylalanine analogs in phenylalanine auxotrophic E. coli. Incorporation of 3-iodotyrosine into proteins is also possible using a mutant tyrosyl-tRNA synthetase from E. coli. in a cell-free translation system [138]. In this report, the amber stop codon was reassigned as a unique codon for 3-iodotyrosine. This method was later adapted to mammalian expression systems [139]. Translational capacity for p-iodophenylalanine [140] and p-bromophenylalanine [141] in response to the amber codon in E. coli hosts has also been described. Related procedures for genetic incorporation aryl halides into proteins and peptides have since been reported [142, 143]. With a range of methods available for the incorporation of aryl halides into proteins, it is possible to study carbon–carbon bond formation by cross-coupling. The reported palladium mediated reaction at p-iodophenylalanine in proteins was the Mizoroki–Heck conjugation of biotin (Scheme 3.17) [144]. This transformation is notable in that it was the ﬁrst palladium mediated carbon–carbon bond forming reaction on a protein. The conversion, however, was extremely low (2%). A modest increase in efﬁciency (28% conversion) was observed when a Sonogashira coupling of biotin was investigated in the same system [145].

3.5 Recent Innovations in Site-Selective Protein Modiﬁcation R

R Pd (0), Cu (I)

N H

Sonogashira O R

I R

Mizoroki-Heck

Pd (0) N H

N H

O

O Ar

OH B

OH

Ar-B(OH)2 Pd (0) ArI

Suzuki-Miyaura N H

Pd (0) N H

O

O

Scheme 3.17 Sonogashira, Mizoroki–Heck, and Suzuki–Miyaura coupling strategies applied to polypeptides carrying p-iodophenylalanine and p-boronophenylalanine.

The opportunity for Suzuki–Miyaura cross-coupling on proteins has been facilitated by the genetic incorporation of p-boronophenylalanine at the amber stop codon using an orthogonal tRNA/aminoacyl-t-RNA synthetase pair [146]. Efﬁcient Suzuki–Miyaura coupling at this residue on proteins remains an outstanding problem since only modest conversion was observed even at prolonged reaction times and elevated temperatures (12 hours at 70 ◦ C). New catalyst systems will likely render this transformation feasible, rendering one of the most widely used cross-couplings in organic synthesis as a useful strategy for site-selective protein modiﬁcation. 3.5.2.4 Oleﬁn Metathesis at S-Allyl Cysteine Oleﬁn metathesis is an efﬁcient, chemoselective way to form C–C bonds [147–150]. Despite its wide application in organic synthesis, its use in modifying biomolecules has been hampered by the limited scope of aqueous metathesis. The emergence of novel, water-soluble catalysts [151] has encouraged the use of aqueous metathesis in chemical biology [152], especially peptide and protein modiﬁcation [153]. It was recently observed that allyl sulﬁdes undergo rapid cross-metathesis in aqueous media with the Hoveyda–Grubbs second-generation metathesis catalyst [154]. This result was unexpected not only because the reaction was run in water, but also because ruthenium-based metathesis catalysts are often incompatible with substrates containing sulfur(II) donor sites [155]. Allyl sulﬁdes are not just compatible

77

78

3 Chemical Protein Modiﬁcation 1.

2.

O S O NH2 O (MSH) R

SH

t

HS

S

R

MesN NMes Cl Ru Cl i PrO

BuOH / buffer, MgCl2

HO HO

OH O

S

HO HO HO

O

OH O

OH HO

Scheme 3.18

R

Cross-metathesis

O O

O n

Aqueous cross-metathesis for site-selective protein modiﬁcation.

with metathesis catalysts, they enhance the rate of metathesis relative to their carbon and heteroatom counterparts. Taking advantage of the allyl sulﬁde reactivity and its easy installation at dehydroalanine [93], glycosylation and PEGylation were accomplished on a protein surface. These modiﬁcations are the ﬁrst examples of cross-metathesis on a protein (Scheme 3.18) [154]. Several chemical routes to S-allyl cysteine on protein surfaces that enable the study of the scope and limitations of oleﬁn metathesis in protein modiﬁcation, have since been described [156]. 3.5.3 Metal-Free Methods for Modifying Unnatural Amino Acids

It is apparent from many of the transition metal–mediated reactions described above that unnatural amino acid incorporation [157, 158] has dramatically expanded the capacity for selective protein modiﬁcation. There are many other transformations that take advantage of the bio-orthogonality of unnatural chemical tags that do not rely on metals. We now turn to the reactions developed for speciﬁc modiﬁcation of these nonnative chemical handles in proteins and even living cells. 3.5.3.1 Oxime Ligation at Aldehydes and Ketones The nucleophilic addition of alkoxyamines and hydrazides to ketones and aldehydes gives relatively stable oximes and hydrazones, respectively. Schultz and coworkers reported the efﬁcient incorporation of a ketone containing amino acid, p-acetyl-l-phenylalanine, through the use of an orthogonal tRNA-synthetase pair in the Z-domain of Staphylococcal protein A [159, 160]. The ketone handle was used for the site-selective installation of hydrazide containing ﬂuorophore and biotin tags [159, 160]. A GlcNAc oxime derivative was also readily accessed, which was further elaborated through the use of glycosyltransferases (Scheme 3.19a) [161]. Oxime formation can also be used as a ligation method at formyl glycine [162, 163]. This residue can be accessed by the oxidation of cysteine by formylglycine-generating enzyme when the cysteine is contained in the sequence of a small sulfatase motif.

3.5 Recent Innovations in Site-Selective Protein Modiﬁcation Ketone handle modification O

HO HO

OH O O HO HO AcHN NH2

OH O O N AcHN

1. UDP-galactose b-1,4-GalT HO HO OH CO2H OHOH OH 2. CMP-sialic acid O O O O N O O AcHN a-2,3-SialylT HO OH HO AcHN

Genetically encoded ketone (a) Aldehyde tag modification O

R O NH 2

H

N

O R H

Genetically encoded aldehyde (b)

Scheme 3.19

Site-speciﬁc modiﬁcation of ketones and aldehydes.

The aldehyde-tagged protein was chemically modiﬁed by selective reaction with hydrazide- and hydroxylamine-functionalized reagents (Scheme 3.19b). This technique was recently used for the site-speciﬁc modiﬁcation of therapeutically relevant monoclonal antibodies [163].

Scheme 3.20

Cell

N N

ob

ob Pr

+

Strain promoted cycloaddition

Pr

e

e

3.5.3.2 Azide and Alkyne Modiﬁcation The copper-catalyzed azide-alkyne cycloaddition was discussed above, and many examples of protein modiﬁcation were described. Still, it is useful to have copper-free variants of this reaction since copper can lead to protein precipitation and is toxic to living cells. With an eye to overcome these limitations and ultimately adapt this reaction for use in living cells, Bertozzi and coworkers have introduced a strain-promoted [3 + 2] cycloaddition between azides and cyclooctynes [164]. (Scheme 3.20). This methodology has enabled in vivo imaging of membrane-associated glycans in developing zebraﬁsh [165, 166]. The azide handle also reacts with arylphosphine reagents in the Staudinger ligation, allowing site-selective modiﬁcation of proteins, even in living systems [167]. Genetically incorporated azides such as Aha and p-azidophenylalanine have allowed chemoselective conjugation by Staudinger and traceless Staudinger ligations in a number of contexts (Scheme 3.21) [123, 168]. Azide stability under physiological conditions allows in vivo incorporation of azides into biomolecules. N-azidoacetylmannosamine (ManNAz), for instance, is

N3

79

N Cell

Strain-promoted [3 + 2] cycloaddition between azides and cyclooctynes.

80

3 Chemical Protein Modiﬁcation H N

O

O N H

R O

Ph2P O

NH

H N

OCH3

O Ph

O P Ph O

Staudinger

HN R

R = Peptide, dye, fluorophores

N3 O O Biosynthetic incorporation of Aha or p-AzPhe

R

Ph2P

H N O

Traceless staudinger

O

R +

OH

Ph2P

R = Fluorophore

Scheme 3.21

Site-selective modiﬁcation by Staudinger and traceless Staudinger ligation.

readily processed by Jurkat cells and ultimately displayed on the cell surface in the sialic acid derivatives that decorate the glycocalix. Staudinger ligation allows modiﬁcation of the azide labeled cell surface [169, 170]. This strategy was later adapted to living animals where Staudinger modiﬁcation veriﬁed the incorporation of ManNAz into murine splenocytes [171]. 3.5.3.3 Selective Modiﬁcation of Tetrazole-Containing Proteins It is possible to genetically incorporate of O-allyl-tyrosine (Oat) into proteins using E. coli translation systems [93, 124, 158]. This unnatural amino acid contains an alkene handle useful for chemical modiﬁcation. Lin and coworkers recently reported the selective labeling of oat-containing proteins by using a photoactivated 1,3-dipolar cycloaddition. (Scheme 3.22) [172]. A nitrile imine is generated in situ by photolysis of a tetrazole. Remarkably, this ‘‘photoclick chemistry’’ can even be used in living cells when using cell-permeable reagents [173]. R

MeO2C

N N N

Protein

R

N

O

O

N CO2Me

N

320 nm, -N2

Protein

Scheme 3.22 Speciﬁc modiﬁcation of genetically encoded O-allyl-tyrosine via a photoinduced 1,3-dipolar cycloaddition reaction.

3.6 Conclusion and Outlook

R

O

Protein

Protein

N N

Protein

O

O

R

Isomerize

N N 5 min, -N2

R

R

HN

N N

Scheme 3.23

R

N

H R

Inverse-electron-demand Diels–Alder bioconjugation reaction.

3.5.3.4 Tetrazine Ligation The cycloaddition of tetrazine and trans-cyclooctene has also been shown to be an efﬁcient method for the site-selective modiﬁcation of proteins (Scheme 3.23) [174]. This bioorthogonal reaction is extremely efﬁcient and proceeds to completion in a matter of minutes, without the need for catalysis. This feature enables efﬁcient reactions at low concentrations. 3.5.4 Dual Modiﬁcation

Many proteins possess two or more forms of PTMs at different sites [3, 4]. Davis and coworkers have reported a method for the incorporation of two different PTMs at speciﬁc sites into proteins [175]. Their strategy consists of the introduction of two orthogonal functionalities followed by sequential chemical modiﬁcation. A single methionine mutant of the β-galactosidase from Sulfolobus solfataricus (SsβG) was expressed with Aha as a methionine surrogate, allowing site-speciﬁc incorporation of an azide tag [123]. Site-directed mutagenesis was also used to incorporate a solvent exposed cysteine at a deﬁned distance from Aha. Use of Cu(I)-catalyzed [3 + 2] cycloaddition at the azide side chain and disulﬁde formation at cysteine allowed sequential introduction of two different modiﬁcations (Scheme 3.24) [175]. This strategy gave access to a reconstituted mimic of a naturally occurring protein, P-selectin glycoprotein ligand-1 (PSGL-1), and was validated by showing strong binding to human P-selectin. Since the PSGL-1 mimic was constructed on a galactosidase scaffold, this innate enzymatic activity could be exploited as a reporter for inﬂammation. This example highlights the utility of multiple, orthogonal ligation methods and is an encouraging validation of the power of chemical modiﬁcation of proteins. Next we consider what the future may hold in advances in both the chemistry and biology associated with selective protein modiﬁcation.

3.6 Conclusion and Outlook

The impact of chemical modiﬁcation of proteins on understanding biology cannot be understated. This review has highlighted multiple methods for such

81

82

3 Chemical Protein Modiﬁcation 43

SSO2Me 43

O O S O

N3

N3

Aqueous buffer

439

O

S

O S

439 HS

S

O

Aqueous buffer Cu(I) ligand

HO HO

OH O

AcHN HO

CO2H OH OH OH O O O O O O OH AcHN O OH OH HO

HO HO

OH O

AcHN HO

CO2H OH OH OH O O O O O OH O AcHN O OH OH HO

N

O O S

N

43

N 439 S

S

O

Scheme 3.24 Double differential modiﬁcation of a protein by triazole and disulﬁde conjugation.

modiﬁcations as well as their application in the dissection of biochemical phenomena. Although there have been many advances in site-selective protein modiﬁcation, much remains to be explored and the discovery of novel, highly speciﬁc reactions is still desirable. Many therapeutic proteins bear some kind of PTM, and vaccination strategies often rely on conjugation of antigens to protein carriers. The chemistry used for these modiﬁcations must be tailored to the compounds of interest and the bigger the chemical toolkit at one’s disposal, the better. It is our intention that the methods reviewed herein will facilitate an understanding of protein modiﬁcation and its potential to provide precisely modiﬁed macromolecules. These materials will facilitate the development of protein-based therapies, and the study of naturally modiﬁed proteins and complex biochemical processes.

References 1. Venter, J.C. et al. (2001) The sequence

of the human genome. Science, 291 (5507), 1304–1351. 2. The C. elegans Sequencing Consortium (1998) Genome sequence of the nematode C elegans: a platform for investigating biology. Science, 282 (5396), 2012–2018. 3. Walsh, C.T. (2005) Posttranslational Modiﬁcation of Proteins: Expanding

Nature’s Inventory, Roberts and Company Press, Englewood. 4. Walsh, C.T., Garneau-Tsodikova, S., Gregory, J., and Gatto, J. (2005) Protein posttranslational modiﬁcations: the chemistry of proteome diversiﬁcations. Angew. Chem. Int. Ed., 44 (45), 7342–7372. 5. Dwek, R.A. (1996) Glycobiology: toward understanding the function of sugars. Chem. Rev., 96 (2), 683–720.

References 6. Lowe, J.B. and Marth, J.D. (2003) A

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

genetic approach to mammalian glycan function. Annu. Rev. Biochem., 72, 643–691. Ohtsubo, K. and Marth, J.D. (2006) Glycosylation in cellular mechanisms of health and disease. Cell, 126 (5), 855–867. Cohen, P. (2002) The origins of protein phosphorylation. Nat. Cell. Biol., 4 (5), E127–E130. Yang, X.-J. and Seto, E. (2008) Lysine acetylation: codiﬁed crosstalk with other posttranslational modiﬁcations. Mol. Cell, 31 (4), 449–461. Trojer, P. and Reinberg, D. (2006) Histone lysine demethylases and their impact on epigenetics. Cell, 125 (2), 213–217. Bedford, M.T. and Richard, S. (2005) Arginine methylation: an emerging regulator of protein function. Mol. Cell, 18 (3), 263–272. Zhang, F.L. and Casey, P.J. (1996) Protein prenylation: molecular mechanisms and functional consequences. Annu. Rev. Biochem., 65 (1), 241–269. Jacob, C., Giles, G.I., Giles, N.M., and Sies, H. (2003) Sulfur and selenium: the role of oxidation state in protein structure and function. Angew. Chem. Int. Ed., 42 (39), 4742–4758. Vranka, J.A., Sakai, L.Y., and Bachinger, H.P. (2004) Prolyl 3-hydroxylase 1, enzyme characterization and identiﬁcation of a novel family of enzymes. J. Biol. Chem., 279 (22), 23615–23621. Davis, B.G. (2004) Mimicking posttranslational modiﬁcations of proteins. Science, 303 (5657), 480–482. Astronomo, R.D., Lee, H.-K., Scanlan, C.N., Pantophlet, R., Huang, C.-Y., Wilson, I.A., Blixt, O., Dwek, R.A., Wong, C.-H., and Burton, D.R. (2008) A glycoconjugate antigen based on the recognition motif of a broadly neutralizing human immunodeﬁciency virus antibody, 2G12, is immunogenic but elicits antibodies unable to bind to the self glycans of gp120. J. Virol., 82 (13), 6359–6368. Joyce, J.G., Krauss, I.J., Song, H.C., Opalka, D.W., Grimm, K.M., Nahas,

18.

19.

20. 21.

22.

23.

24.

25.

26.

D.D., Esser, M.T., Hrin, R., Feng, M., Dudkin, V.Y., Chastain, M., Shiver, J.W., and Danishefsky, S.J. (2008) An oligosaccharide-based HIV-1 2G12 mimotope vaccine induces carbohydrate-speciﬁc antibodies that fail to neutralize HIV-1 virions. Proc. Natl. Acad. Sci. U.S.A., 105 (41), 15684–15689. Krug, L.M., Ragupathi, G., Ng, K.K., Hood, C., Jennings, H.J., Guo, Z., Kris, M.G., Miller, V., Pizzo, B., Tyson, L., Baez, V., and Livingston, P.O. (2004) Vaccination of small cell lung cancer patients with polysialic acid or N-propionylated polysialic acid conjugated to keyhole limpet hemocyanin. Clin. Cancer Res., 10 (3), 916–923. Schoﬁeld, L., Hewitt, M.C., Evans, K., Siomos, M.A., and Seeberger, P.H. (2002) Synthetic GPI as a candidate anti-toxic vaccine in a model of malaria. Nature, 418 (6899), 785–789. Hermanson, G.T. (2008) Bioconjugation Techniques, 2nd edn, Academic Press. Tawﬁk, D.S. (2007) Amidation of carboxyl groups, The Protein Protocols Handbook, 2nd edn, Humana Press 477–478. Barrai, I., Volinia, S., and Scapoli, C. (1995) The usage of oligopeptides in proteins correlates negatively with molecular weight. Int. J. Pept. Protein Res., 45 (4), 326–331. Fodge, M.N. and Al-Karadaghi, S. (2002) Occurrence, conformational features and amino acid propensities for the helix. Protein Eng., 15 (5), 353–358. Crankshaw, M.W. and Grant, G.A. (1996) Modiﬁcation at cysteine, Current Protocols in Protein Science, Unit 15.1, John Wiley & Sons, Inc, 1–18. Chalker, J.M., Bernardes, G.J.L., Lin, Y.A., and Davis, B.G. (2009) Chemical modiﬁcation of proteins at cysteine: opportunities in chemistry and biology. Chem. Asian J., 4 (5), 630–640. Izumi, M., Okumura, S., Yuasa, H., and Hashimoto, H. (2003) Mannose-BSA conjugates: comparison between commercially available linkers in reactivity and bioactivity. J. Carbohydr. Chem., 22 (5), 317–329.

83

84

3 Chemical Protein Modiﬁcation 27. Buskas, T., Li, Y., and Boons, G.-J.

28.

29.

30.

31.

32.

33.

34.

(2004) The immunogenicity of the tumor-associated antigen Lewisy may be suppressed by a bifunctional cross-linker required for coupling to a carrier protein. Chem. Eur. J., 10 (14), 3517–3524. Ingale, S., Wolfert, M.A., Gaekwad, J., Buskas, T., and Boons, G.-J. (2007) Robust immune responses elicited by a fully synthetic three-component vaccine. Nat. Chem. Biol., 3 (10), 663–667. Lahmann, M., B¨ulow, L., Teodorovic, P., Gyb¨ack, H., and Oscarson, S. (2004) Synthesis of the Lewis b hexasaccharide and HSA-conjugates thereof. Glycoconj. J., 21, 251–256. Tietze, L.F., Schroeter, C., Gabius, S., Brinck, U., Goerlach-Graw, A., and Gabius, H.J. (1991) Conjugation of p-aminophenyl glycosides with squaric acid diester to a carrier protein and the use of the neoglycoprotein in the histochemical detection of lectins. Bioconjug. Chem., 2 (3), 148–153. Kamath, V.P., Diedrich, P., and Hindsgaul, O. (1996) Use of diethyl squarate for the coupling of oligosaccharide amines to carrier proteins and characterization of the resulting neoglycoproteins by MALDI-TOF mass spectrometry. Glycoconj. J., 13, 315–319. Bernstein, M.A. and Hall, L.D. (1980) A general synthesis of model glycoproteins: coupling of alkenyl glycosides to proteins, using reductive ozonolysis followed by reductive amination with sodium cyanoborohydride. Carbohydr. Res., 78 (1), C1–C3. Ragupathi, G., Slovin, S.F., Adluri, S., Sames, D., Kim, I.J., Kim, H.M., Spassova, M., Bornmann, W.G., Lloyd, K.O., Scher, H.I., Livingston, P.O., and Danishefsky, S.J. (1999) A fully synthetic Globo H carbohydrate vaccine induces a focused humoral response in prostate cancer patients: a proof of principle. Angew. Chem. Int. Ed., 38 (4), 563–566. Lee, Y.C., Stowell, C.P., and Krantz, M.J. (1976) 2-Imino-2-methoxyethyl

35.

36.

37.

38.

39.

40.

41.

42.

43.

1-thioglycosides: new reagents for attaching sugars to proteins. Biochemistry, 15 (18), 3956–3963. Stowell, C.P. and Lee, Y.C. (1982) Preparation of neoglycoproteins using 2-imino-2-methoxyethyl 1-thioglycosides. Methods Enzymol., 83, 278–288. Robinson, M.A., Charlton, S.T., Garnier, P., Wang, X.-T., Davis, S.S., Perkins, A.C., Frier, M., Duncan, R., Savage, T.J., Wyatt, D.A., Watson, S.A., and Davis, B.G. (2004) LEAPT: Lectin-directed enzyme-activated prodrug therapy. Proc. Natl. Acad. Sci. U.S.A., 101 (40), 14527–14532. Rao, M.J. and Acharya, A.S. (1994) Amidation of basic carboxyl groups of hemoglobin. Methods Enzymol., 231, 246–267. Weerapana, E., Simon, G.M., and Cravatt, B.F. (2008) Disparate proteome reactivity proﬁles of carbon electrophiles. Nat. Chem. Biol., 4 (7), 405–407. Davis, N.J. and Flitsch, S.L. (1991) A novel method for the speciﬁc glycosylation of proteins. Tetrahedron Lett., 32 (46), 6793–6796. Macmillan, D., Bill, R.M., Sage, K.A., Fern, D., and Flitsch, S.L. (2001) Selective in vitro glycosylation of recombinant proteins: semi-synthesis of novel homogeneous glycoforms of human erythropoietin. Chem. Biol., 8 (2), 133–145. Swanwick, R.S., Daines, A.M., Tey, L.-H., Flitsch, S.L., and Allemann, R.K. (2005) Increased thermal stability of site-selectively glycosylated dihydrofolate reductase. ChemBioChem, 6 (8), 1338–1340. Nielsen, M.L., Vermeulen, M., Bonaldi, T., Cox, J., Moroder, L., and Mann, M. (2008) Iodoacetamide-induced artifact mimics ubiquitination in mass spectrometry. Nat. Methods, 5 (6), 459–460. Zhang, Y., Bhatt, V.S., Sun, G., Wang, P.G., and Palmer, A.F. (2008) Site-selective glycosylation of hemoglobin on Cys beta93. Bioconjug. Chem., 19 (11), 2221–2230.

References 44. Betting, D.J., Kaﬁ, K., Abdollahi-Fard,

45.

46.

47.

48.

49.

50.

51.

52.

A., Hurvitz, S.A., and Timmerman, J.M. (2008) Sulfhydryl-based tumor antigen-carrier protein conjugates stimulate superior antitumor immunity against B cell lymphomas. J. Immunol., 181 (6), 4131–4140. Hemelaar, J., Borodovsky, A., Kessler, B.M., Reverter, D., Cook, J., Kolli, N., Gan-Erdene, T., Wilkinson, K.D., Gill, G., Lima, C.D., Ploegh, H.L., and Ovaa, H. (2004) Speciﬁc and covalent targeting of conjugating and deconjugating enzymes of ubiquitin-like proteins. Mol. Cell Biol., 24 (1), 84–95. Shaunak, S., Godwin, A., Choi, J.-W., Balan, S., Pedone, E., Vijayarangam, D., Heidelberger, S., Teo, I., Zloh, M., and Brocchini, S. (2006) Site-speciﬁc PEGylation of native disulﬁde bonds in therapeutic proteins. Nat. Chem. Biol., 2 (6), 312–313. Lindley, H. (1956) A new synthetic substrate for trypsin and its application to the determination of the amino-acid sequence of proteins. Nature, 178 (4534), 647–648. Smith, H.B. and Hartman, F.C. (1988) Restoration of activity to catalytically deﬁcient mutants of ribulose bisphosphate carboxylase/oxygenase by aminoethylation. J. Biol. Chem., 263 (10), 4921–4925. Simon, M.D., Chu, F., Racki, L.R., de la Cruz, C.C., Burlingame, A.L., Panning, B., Narlikar, G.J., and Shokat, K.M. (2007) The site-speciﬁc installation of methyl-lysine analogs into recombinant histones. Cell, 128 (5), 1003–1012. Ryle, A.P. and Sanger, F. (1955) Disulphide interchange reactions. Biochem. J., 60 (4), 535–540. Davis, B.G., Maughan, M.A.T., Green, M.P., Ullman, A., and Jones, J.B. (2000) Glycomethanethiosulfonates: powerful reagents for protein glycosylation. Tetrahedron: Asymmetry, 11 (1), 245–262. Gamblin, D.P., Garnier, P., Ward, S.J., Oldham, N.J., Fairbanks, A.J., and Davis, B.G. (2003) Glycosyl phenylthiosulfonates (glyco-PTS): novel reagents

53.

54.

55.

56.

57.

58.

59.

60.

61.

for glycoprotein synthesis. Org. Biomol. Chem., 1 (21), 3642–3644. Macindoe, W.M., van Oijen, A.H., and Boons, G.-J. (1998) A unique and highly facile method for synthesising disulﬁde linked neoglycoconjugates: a new approach for remodelling of peptides and proteins. Chem. Commun., (7), 847–848. Gamblin, D.P., Garnier, P., van Kasteren, S., Oldham, N.J., Fairbanks, A.J., and Davis, B.G. (2004) Glyco-SeS: selenenylsulﬁde-mediated protein glycoconjugation-a new strategy in post-translational modiﬁcation. Angew. Chem. Int. Ed., 43 (7), 828–833. Bernardes, G.J.L., Marston, J.P., Batsanov, A.S., Howard, J.A.K., and Davis, B.G. (2007) A trisulﬁde-linked glycoprotein. Chem. Commun., (30), 3145–3147. Bernardes, G.J.L., Gamblin, D.P., and Davis, B.G. (2006) The direct formation of glycosyl thiols from reducing sugars allow as one-pot protein glycoconjugation. Angew. Chem. Int. Ed., 45 (24), 4007–4011. Gamblin, D.P., van Kasteren, S., Bernardes, G.J.L., Chalker, J.M., Oldham, N.J., Fairbanks, A.J., and Davis, B.G. (2008) Chemical site-selective prenylation of proteins. Mol. BioSyst., 4 (6), 558–561. Harpp, D.N. and Gleason, J.G. (1971) Preparation and mass spectral properties of cystine and lanthionine derivatives. Novel synthesis of L-lanthionine by selective desulfurization. J. Org. Chem., 36 (1), 73–80. Harpp, D.N. and Gleason, J.G. (1971) Organic sulfur chemistry. X. Selective desulfurization of disulﬁdes. Scope and mechanism. J. Am. Chem. Soc., 93 (10), 2437–2445. Harpp, D.N., Gleason, J.G. and Snyder, J.P. (1968) Organic sulfur chemistry. I. The disulﬁde-phosphine reaction. Desulfurization with tris(diethylamino)phosphine. J. Am. Chem. Soc., 90 (15), 4181–4182. Bernardes, G.J.L., Grayson, E.J., Thompson, S., Chalker, J.M., Errey, J.C., Oualid, F.E., Claridge, T.D.W., and Davis, B.G. (2008) From disulﬁde-

85

86

3 Chemical Protein Modiﬁcation

62.

63.

64.

65.

66.

67.

68.

69.

70.

to thioether-linked glycoproteins. Angew. Chem. Int. Ed., 47 (12), 2244–2247. Dawson, P.E., Muir, T.W., Clark-Lewis, I., and Kent, S.B. (1994) Synthesis of proteins by native chemical ligation. Science, 266 (5186), 776–779. Wieland, V.T., Bokelmann, E., Bauer, L., Lang, H.U., and Lau, H. (1953) ¨ Uber peptidsynthesen. 8. Mitteilung bildung von S-haltigen peptiden durch intramolekulare wanderung von aminoacylresten. Liebigs Ann. Chem., 583 (1), 129–149. Muir, T.W. (2003) Semisynthesis of proteins by expressed protein ligation. Annu. Rev. Biochem., 72 (1), 249–289. Piontek, C., Ring, P., Harjes, O., Heinlein, C., Mezzato, S., Lombana, N., P¨ohner, C., P¨uttner, M., Silva, D.V., Martin, A., Schmid, F.X., and Unverzagt, C. (2009) Semisynthesis of a homogeneous glycoprotein enzyme: ribonuclease C: part 1, Angew. Chem. Int. Ed., 48 (11), 1936–1940. Piontek, C., Silva, D.V., Heinlein, C., P¨ohner, C., Mezzato, S., Ring, P., Martin, A., Schmid, F.X., and Unverzagt, C. (2009) Semisynthesis of a homogeneous glycoprotein enzyme: ribonuclease C: part 2. Angew. Chem. Int. Ed., 48 (11), 1941–1945. Becker, C.F.W., Liu, X., Olschewski, D., Castelli, R., Seidel, R., and Seeberger, P.H. (2008) Semisynthesis of a glycosylphosphatidylinositol-anchored prion protein. Angew. Chem. Int. Ed., 47 (43), 8215–8219. Shogren-Knaak, M.A., Fry, C.J., and Peterson, C.L. (2003) A native peptide ligation strategy for deciphering nucleosomal histone modiﬁcations. J. Biol. Chem., 278 (18), 15744–15748. He, S., Bauman, D., Davis, J.S., Loyola, A., Nishioka, K., Gronlund, J.L., Reinberg, D., Meng, F., Kelleher, N., and McCafferty, D.G. (2003) Facile synthesis of site-speciﬁcally acetylated and methylated histone proteins: reagents for evaluation of the histone code hypothesis. Proc. Natl. Acad. Sci. U.S.A., 100 (21), 12033–12038. McGinty, R.K., Kim, J., Chatterjee, C., Roeder, R.G., and Muir, T.W. (2008)

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

Chemically ubiquitylated histone H2B stimulates hDot1L-mediated intranucleosomal methylation. Nature, 453 (7196), 812–816. Flavell, R.R. and Muir, T.W. (2009) Expressed Protein Ligation (EPL) in the study of signal transduction, ion conduction, and chromatin biology. Acc. Chem. Res., 42 (1), 107–116. Schwarzer, D. and Cole, P.A. (2005) Protein semisynthesis and expressed protein ligation: chasing a protein’s tail. Curr. Opin. Chem. Biol., 9 (6), 561–569. Macmillan, D. (2006) Evolving strategies for protein synthesis converge on native chemical ligation. Angew. Chem. Int. Ed., 45 (46), 7668–7672. Dawson, P.E. and Kent, S.B.H. (2000) Synthesis of native proteins by chemical ligation. Annu. Rev. Biochem., 69 (1), 923–960. Hofmann, R.M. and Muir, T.W. (2002) Recent advances in the application of expressed protein ligation to protein engineering. Curr. Opin. Biotechnol., 13 (4), 297–303. Lianshan, Z. and Tam, J.P. (1997) Orthogonal coupling of unprotected peptide segments through histidyl amino terminus. Tetrahedron Lett., 38 (1), 3–6. Yan, L.Z. and Dawson, P.E. (2001) Synthesis of peptides and proteins without cysteine residues by native chemical ligation combined with desulfurization. J. Am. Chem. Soc., 123 (4), 526–533. Crich, D. and Banerjee, A. (2007) Native chemical ligation at phenylalanine. J. Am. Chem. Soc., 129 (33), 10064–10065. Wan, Q. and Danishefsky, S.J. (2007) Free-radical-based, speciﬁc desulfurization of cysteine: a powerful advance in the synthesis of polypeptides and glycopolypeptides. Angew. Chem. Int. Ed., 46 (48), 9248–9252. Haase, C., Rohde, H., and Seitz, O. (2008) Native chemical ligation at valine, Angew. Chem. Int. Ed., 47 (36), 6807–6810. Chen, J., Wan, Q., Yuan, Y., Zhu, J., and Danishefsky, S.J. (2008) Native

References

82.

83.

84.

85.

86.

87.

88.

89.

90.

chemical ligation at valine: a contribution to peptide and glycopeptide synthesis. Angew. Chem. Int. Ed., 47 (44), 8521–8524. Zhu, Y. and van der Donk, W.A. (2001) Convergent synthesis of peptide conjugates using dehydroalanines for chemoselective ligations. Org. Lett., 3 (8), 1189–1192. Galoni´a, D.P., van der Donk, W.A., and Gin, D.Y. (2003) Oligosaccharide-peptide ligation of glycosyl thiolates with dehydropeptides: synthesis of S-linked mucin-related glycopeptide conjugates. Chem. Eur. J., 9 (24), 5997–6006. Strumeyer, D.H., White, W.N., and Koshland, D.E. (1963) Role of serine in chymotrypsin action. Conversion of the active serine to dehydroalanine. Proc. Natl. Acad. Sci. U.S.A., 50 (5), 931–935. Weiner, H., White, W.N., Hoare, D.G., and Koshland, D.E. (1966) The formation of anhydrochymotrypsin by removing the elements of water from the serine at the active site. J. Am. Chem. Soc., 88 (16), 3851–3859. Holmes, T.J. and Lawton, R.G. (1977) Cysteine modiﬁcation and cleavage of proteins with 2-methyl-N1-benzenesulfonyl-N4bromoacetylquinonediimide. J. Am. Chem. Soc., 99 (6), 1984–1986. Rich, D.H., Tam, J., Mathiaparanam, P., Grant, J.A., and Mabuni, C. (1974) General synthesis of didehydroamino acids and peptides. J. Chem. Soc., Chem. Commun., (21), 897–898. Hashimoto, K., Sakai, M., Okuno, T., and Shirahama, H. (1996) Phenylselenoalanine as a dehydroalanine precursor – efﬁcient synthesis of alternariolide (AM-toxin I). Chem. Commun., (10), 1139–1140. Okeley, N.M., Zhu, Y. and van der Donk, W.A. (2000) Facile chemoselective synthesis of dehydroalanine-containing peptides. Org. Lett., 2 (23), 3603–3606. Seebeck, F.P. and Szostak, J.W. (2006) Ribosomal synthesis of dehydroalanine-containing peptides. J. Am. Chem. Soc., 128 (22), 7150–7151.

91. Wang, J., Schiller, S.M., and Schultz,

92.

93.

94.

95.

96.

97.

98.

99.

100.

101.

P.G. (2007) A biosynthetic route to dehydroalanine-containing proteins. Angew. Chem. Int. Ed., 46 (36), 6849–6851. ¨ Schmidt, U. and Ohler, E. (1976) Optical induction during biomimetic formation of cysteine. Angew. Chem. Int. Ed., 15 (1), 42. Bernardes, G.J.L., Chalker, J.M., Errey, J.C., and Davis, B.G. (2008) Facile conversion of cysteine and alkyl cysteines to dehydroalanine on protein surfaces: versatile and switchable access to functionalized proteins. J. Am. Chem. Soc., 130 (15), 5052–5053. Guo, J., Wang, J., Lee, J.S., and Schultz, P.G. (2008) Site-speciﬁc incorporation of methyl- and acetyl-lysine analogues into recombinant proteins. Angew. Chem. Int. Ed., 47 (34), 6399–6401. Antos, J.M. and Francis, M.B. (2006) Transition metal catalyzed methods for site-selective protein modiﬁcation. Curr. Opin. Chem. Biol., 10 (3), 253–262. Hegedus, L.S. (1999) Transition Metals in the Synthesis of Complex Organic Molecules, 2nd edn, University Science Books. Li, C.-J. (2005) Organic reactions in aqueous media with a focus on carbon-carbon bond formations: a decade update. Chem. Rev., 105 (8), 3095–3166. Shaughnessy, K.H. and DeVasher, R.B. (2005) Palladium-catalyzed cross-coupling in aqueous media: recent progress and current applications. Curr. Org. Chem., 9 (7), 585–604. Herrerias, C.I., Yao, X., Li, Z., and Li, C.-J. (2007) Reactions of C-H bonds in water. Chem. Rev., 107 (6), 2546–2562. Antos, J.M. and Francis, M.B. (2004) Selective tryptophan modiﬁcation with rhodium carbenoids in aqueous solution. J. Am. Chem. Soc., 126 (33), 10256–10257. Antos, J.M., McFarland, J.M., Iavarone, A.T., and Francis, M.B. (2009) Chemoselective tryptophan labeling with rhodium carbenoids at mild pH. J. Am. Chem. Soc., 131 (17), 6301–6308.

87

88

3 Chemical Protein Modiﬁcation 102. Bao, Z., Wang, S., Shi, W., Dong, S.,

103.

104.

105.

106.

107.

108.

109.

110.

111.

and Ma, H. (2007) Selective modiﬁcation of Trp19 in β-lactoglobulin by a new diazo ﬂuorescence probe. J. Proteome Res., 6 (9), 3835–3841. Tilley, S.D. and Francis, M.B. (2006) Tyrosine-selective protein alkylation using π -allylpalladium complexes. J. Am. Chem. Soc., 128 (4), 1080–1081. Chen, S., Li, X., and Ma, H. (2009) New approach for local structure analysis of the tyrosine domain in proteins by using a site-speciﬁc and polarity-sensitive ﬂuorescent probe. ChemBioChem, 10 (7), 1200–1207. McFarland, J.M. and Francis, M.B. (2005) Reductive alkylation of proteins using iridium catalyzed transfer hydrogenation. J. Am. Chem. Soc., 127 (39), 13490–13491. Tornoe, C.W., Christensen, C., and Meldal, M. (2002) Peptidotriazoles on solid phase: [1,2,3]-Triazoles by regiospeciﬁc copper(I)-catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem., 67 (9), 3057–3064. Rostovtsev, V.V., Green, L.G., Fokin, V.V., and Sharpless, K.B. (2002) A stepwise Huisgen cycloaddition process: Copper(I)-catalyzed regioselective ‘‘ligation’’ of azides and terminal alkynes. Angew. Chem. Int. Ed., 41 (14), 2596–2599. H¨uisgen, R. (1984) 1,3-Dipolar Cycloaddition Chemistry, John Wiley & Sons, Inc., New York. Kolb, H.C., Finn, M.G., and Sharpless, K.B. (2001) Click chemistry: diverse chemical function from a few good reactions. Angew. Chem. Int. Ed., 40 (11), 2004–2021. Himo, F., Lovell, T., Hilgraf, R., Rostovtsev, V.V., Noodleman, L., Sharpless, K.B., and Fokin, V.V. (2005) Copper(I)-catalyzed synthesis of azoles. DFT study predicts unprecedented reactivity and intermediates. J. Am. Chem. Soc., 127 (1), 210–216. Ahlquist, M. and Fokin, V.V. (2007) Enhanced reactivity of dinuclear copper(I) acetylides in dipolar cycloadditions. Organometallics, 26 (18), 4389–4391.

112. Rodionov, V.O., Fokin, V.V., and

113.

114.

115.

116.

117.

118.

119.

120.

121.

Finn, M.G. (2005) Mechanism of the ligand-free Cu(I)-catalyzed azide-alkyne cycloaddition reaction. Angew. Chem. Int. Ed., 44 (15), 2210–2215. Kolb, H.C. and Sharpless, K.B. (2003) The growing impact of click chemistry on drug discovery. Drug Discov. Today, 8 (24), 1128–1137. Hein, C., Liu, X.-M., and Wang, D. (2008) Click chemistry, a powerful tool for pharmaceutical sciences. Pharm. Res., 25 (10), 2216–2230. Wang, Q., Chan, T.R., Hilgraf, R., Fokin, V.V., Sharpless, K.B., and Finn, M.G. (2003) Bioconjugation by copper(I)-catalyzed azide-alkyne [3 + 2] cycloaddition. J. Am. Chem. Soc., 125 (11), 3192–3193. Gupta, S.S., Kuzelka, J., Singh, P., Lewis, W.G., Manchester, M., and Finn, M.G. (2005) Accelerated bioorthogonal conjugation: a practical method for the ligation of diverse functional molecules to a polyvalent virus scaffold. Bioconjug. Chem., 16 (6), 1572–1579. van Dongen, S.F.M., Teeuwen, R.L.M., Nallani, M., van Berkel, S.S., Cornelissen, J.J.L.M., Nolte, R.J.M., and van Hest, J.C.M. (2009) Single-step azide introduction in proteins via an aqueous diazo transfer. Bioconjug. Chem., 20 (1), 20–23. Speers, A.E., Adam, G.C., and Cravatt, B.F. (2003) Activity-based protein proﬁling in vivo using a copper(I)-catalyzed azide-alkyne [3 + 2] cycloaddition. J. Am. Chem. Soc., 125 (16), 4686–4687. Link, A.J., Vink, M.K.S., and Tirrell, D.A. (2004) Presentation and detection of azide functionality in bacterial cell surface proteins. J. Am. Chem. Soc., 126 (34), 10598–10602. Gupta, S.S., Raja, K.S., Kaltgrad, E., Strable, E., and Finn, M.G. (2005) Virus-glycopolymer conjugates by copper(i) catalysis of atom transfer radical polymerization and azide-alkyne cycloaddition. Chem. Commun., (34), 4315–4317. Lin, P.-C., Ueng, S.-H., Tseng, M.-C., Ko, J.-L., Huang, K.-T., Yu, S.-C., Adak, A.K., Chen, Y.-J., and Lin, C.-C.

References

122.

123.

124.

125.

126.

127.

128.

129.

(2006) Site-speciﬁc protein modiﬁcation through Cu(I)-catalyzed 1,2,3-triazole formation and Its implementation in protein microarray fabrication. Angew. Chem. Int. Ed., 45 (26), 4286–4290. Dirks, A.J., van Berkel, S.S., Hatzakis, N.S., Opsteen, J.A., van Delft, F.L., Cornelissen, J.J.L.M., Rowan, A.E., van Hest, J.C.M., Rutjes, F.P.J.T., and Nolte, R.J.M. (2005) Preparation of biohybrid amphiphiles via the copper catalysed Huisgen [3 + 2] dipolar cycloaddition reaction. Chem. Commun., (33), 4172–4174. Kiick, K.L., Saxon, E., Tirrell, D.A., and Bertozzi, C.R. (2002) Incorporation of azides into recombinant proteins for chemoselective modiﬁcation by the Staudinger ligation. Proc. Natl. Acad. Sci. U.S.A., 99 (1), 19–24. van Hest, J.C., Kiick, K.L., and Tirrell, D.A. (2000) Efﬁcient incorporation of unsaturated methionine analogues into proteins in vivo. J. Am. Chem. Soc., 122 (7), 1282–1288. Link, A.J. and Tirrell, D.A. (2003) Cell surface labeling of Escherichia coli via copper(I)-catalyzed [3 + 2] cycloaddition. J. Am. Chem. Soc., 125 (37), 11164–11165. Beatty, K.E., Xie, F., Wang, Q., and Tirrell, D.A. (2005) Selective dye-labeling of newly synthesized proteins in bacterial cells. J. Am. Chem. Soc., 127 (41), 14150–14151. Deiters, A., Cropp, T.A., Mukherji, M., Chin, J.W., Anderson, J.C., and Schultz, P.G. (2003) Adding amino acids with novel reactivity to the genetic code of Saccharomyces Cerevisiae. J. Am. Chem. Soc., 125 (39), 11782–11783. Deiters, A., Cropp, T.A., Summerer, D., Mukherji, M., and Schultz, P.G. (2004) Site-speciﬁc PEGylation of proteins containing unnatural amino acids. Bioorg. Med. Chem. Lett., 14 (23), 5743–5745. Nguyen, D.P., Lusic, H., Neumann, H., Kapadnis, P.B., Deiters, A., and Chin, J.W. (2009) Genetic encoding and labeling of aliphatic azides and alkynes in recombinant proteins via a pyrrolysyl-tRNA synthetase/tRNACUA

130.

131.

132.

133.

134.

135.

136.

137.

138.

pair and click chemistry. J. Am. Chem. Soc. doi: 10.1021/ja900553w. Negishi, E.-I. and Meijere, A.D. (2002) Handbook of Organopalladium Chemistry for Organic Synthesis, 1st edn, Wiley-Interscience. Shaughnessy, K.H. (2006) Beyond TPPTS: new approaches to the development of efﬁcient palladium-catalyzed aqueous-phase cross-coupling reactions. Eur. J. Org. Chem., 2006 (8), 1827–1835. Ojida, A., Tsutsumi, H., Kasagi, N., and Hamachi, I. (2005) Suzuki coupling for protein modiﬁcation. Tetrahedron Lett., 46 (19), 3301–3305. Vilaro, M., Arsequell, G., Valencia, G., Ballesteros, A., and Barluenga, J. (2008) Arylation of Phe and Tyr side chains of unprotected peptides by a Suzuki-Miyaura reaction in water. Org. Lett., 10 (15), 3243–3245. Dibowski, H. and Schmidtchen, F.P. (1998) Bioconjugation of peptides by palladium-catalyzed C-C cross-coupling in water. Angew. Chem. Int. Ed., 37 (4), 476–478. Bong, D.T. and Ghadiri, M.R. (2001) Chemoselective Pd(0)-catalyzed peptide coupling in water. Org. Lett., 3 (16), 2509–2511. Sharma, N., Furter, R., Kast, P., and Tirrell, D.A. (2000) Efﬁcient introduction of aryl bromide functionality into proteins in vivo. FEBS Lett., 467 (1), 37–40. Kirshenbaum, K., Carrico, I.S., and Tirrell, D.A. (2002) Biosynthesis of proteins incorporating a versatile set of phenylalanine analogues. ChemBioChem, 3 (2-3), 235–237. Kiga, D., Sakamoto, K., Kodama, K., Kigawa, T., Matsuda, T., Yabuki, T., Shirouzu, M., Harada, Y., Nakayama, H., Takio, K., Hasegawa, Y., Endo, Y., Hirao, I., and Yokoyama, S. (2002) An engineered Escherichia coli tyrosyl-tRNA synthetase for site-speciﬁc incorporation of an unnatural amino acid into proteins in eukaryotic translation and its application in a wheat germ cell-free system. Proc. Natl. Acad. Sci. U.S.A., 99 (15), 9715–9720.

89

90

3 Chemical Protein Modiﬁcation 139. Sakamoto, K., Hayashi, A.,

140.

141.

142.

143.

144.

145.

Sakamoto, A., Kiga, D., Nakayama, H., Soma, A., Kobayashi, T., Kitabatake, M., Takio, K., Saito, K., Shirouzu, M., Hirao, I., and Yokoyama, S. (2002) Site-speciﬁc incorporation of an unnatural amino acid into proteins in mammalian cells. Nucl. Acids Res., 30 (21), 4692–4699. Xie, J., Wang, L., Wu, N., Brock, A., Spraggon, G., and Schultz, P.G. (2004) The site-speciﬁc incorporation of p-iodo-L-phenylalanine into proteins for structure determination. Nat. Biotechnol., 22 (10), 1297–1301. Kwon, I., Wang, P., and Tirrell, D.A. (2006) Design of a bacterial host for site-speciﬁc incorporation of p-bromophenylalanine into recombinant proteins. J. Am. Chem. Soc., 128 (36), 11778–11783. Kodama, K., Fukuzawa, S., Sakamoto, K., Nakayama, H., Kigawa, T., Yabuki, T., Matsuda, N., Shirouzu, M., Takio, K., Tachibana, K., and Yokoyama, S. (2006) A new protein engineering approach combining chemistry and biology, Part I; Site-speciﬁc incorporation of 4-iodo-L-phenylalanine in vitro by using misacylated suppressor tRNA-Phe. ChemBioChem, 7 (10), 1577–1581. Hartman, M.C.T., Josephson, K., Lin, C.-W., and Szostak, J.W. (2007) An expanded set of amino acid analogs for the ribosomal translation of unnatural peptides. PLoS ONE, 2 (10), e972. Kodama, K., Fukuzawa, S., Nakayama, H., Kigawa, T., Sakamoto, K., Yabuki, T., Matsuda, N., Shirouzu, M., Takio, K., Tachibana, K., and Yokoyama, S. (2006) Regioselective carbon-carbon bond formation in proteins with palladium catalysis; new protein chemistry by organometallic chemistry. ChemBioChem, 7 (1), 134–139. Kodama, K., Fukuzawa, S., Nakayama, H., Sakamoto, K., Kigawa, T., Yabuki, T., Matsuda, N., Shirouzu, M., Takio, K., Yokoyama, S., and Tachibana, K. (2007) Site-speciﬁc functionalization of proteins by organopalladium reactions. ChemBioChem, 8 (2), 232–238.

146. Brustad, E., Bushey, M.L., Lee, J.W.,

147.

148.

149.

150.

151.

152.

153.

154.

155.

156.

157.

Groff, D., Liu, W., and Schultz, P.G. (2008) A genetically encoded boronate-containing amino acid. Angew. Chem. Int. Ed., 47 (43), 8220–8223. Connon, S.J. and Blechert, S. (2003) Recent developments in oleﬁn cross-metathesis. Angew. Chem. Int. Ed., 42 (17), 1900–1923. F¨urstner, A. (2000) Oleﬁn metathesis and beyond. Angew. Chem. Int. Ed., 39 (17), 3012–3043. Grubbs, R.H. (2003) Handbook of Metathesis, Wiley-VCH Verlag GmbH, Weinheim. Hoveyda, A.H. and Zhugralin, A.R. (2007) The remarkable metal-catalysed oleﬁn metathesis reaction. Nature, 450 (7167), 243–251. Burtscher, D. and Grela, K. (2009) Aqueous oleﬁn metathesis. Angew. Chem. Int. Ed., 48 (3), 442–454. Binder, J.B. and Raines, R.T. (2008) Oleﬁn metathesis for chemical biology. Curr. Opin. Chem. Biol., 12 (6), 767–773. Lin, Y.A., Chalker, J.M., and Davis, B.G. (2009) Oleﬁn metathesis for site-selective protein modiﬁcation. ChemBioChem, 10 (6), 959–969. Lin, Y.A., Chalker, J.M., Floyd, N., Bernardes, G.J.L., and Davis, B.G. (2008) Allyl sulﬁdes are privileged substrates in aqueous cross-metathesis: application to site-selective protein modiﬁcation. J. Am. Chem. Soc., 130 (30), 9642–9643. F¨urstner, A., Seidel, G., and Kindler, N. (1999) Macrocycles by ring-closing-metathesis, XI: syntheses of (R)-( + )-lasiodiplodin, zeranol and truncated salicylihalamides. Tetrahedron, 55 (27), 8215–8230. Chalker, J.M., Lin, Y.A., Boutureira, O., and Davis, B.G. (2009) Enabling oleﬁn metathesis on proteins: chemical methods for installation of S-allyl cysteine. Chem. Commun., (24), 3714–3716. Kiick, K.L. and Tirrell, D.A. (2000) Protein engineering by in vivo incorporation of non-natural amino acids: control of incorporation of methionine analogues by methionyl-tRNA

References

158.

159.

160.

161.

162.

163.

164.

165.

166.

synthetase. Tetrahedron, 56 (48), 9487–9493. Xie, J. and Schultz, P.G. (2006) A chemical toolkit for proteins – an expanded genetic code. Nat. Rev. Mol. Cell Biol., 7 (10), 775–782. Cornish, V.W., Hahn, K.M., and Schultz, P.G. (1996) Site-speciﬁc protein modiﬁcation using a ketone handle. J. Am. Chem. Soc., 118 (34), 8150–8151. Wang, L., Zhang, Z., Brock, A., and Schultz, P.G. (2003) Addition of the keto functional group to the genetic code of Escherichia coli. Proc. Natl. Acad. Sci. U.S.A., 100 (1), 56–61. Liu, H., Wang, L., Brock, A., Wong, C.H., and Schultz, P.G. (2003) A method for the generation of glycoprotein mimetics. J. Am. Chem. Soc., 125 (7), 1702–1703. Carrico, I.S., Carlson, B.L., and Bertozzi, C.R. (2007) Introducing genetically encoded aldehydes into proteins. Nat. Chem. Biol., 3 (6), 321–322. Wu, P., Shui, W., Carlson, B.L., Hu, N., Rabuka, D., Lee, J., and Bertozzi, C.R. (2009) Site-speciﬁc chemical modiﬁcation of recombinant proteins produced in mammalian cells by using the genetically encoded aldehyde tag. Proc. Natl. Acad. Sci. U.S.A., 106 (9), 3000–3005. Agard, N.J., Prescher, J.A., and Bertozzi, C.R. (2004) A strain-promoted [3 + 2] azide-alkyne cycloaddition for covalent modiﬁcation of biomolecules in living systems. J. Am. Chem. Soc., 126 (46), 15046–15047. Laughlin, S.T., Baskin, J.M., Amacher, S.L., and Bertozzi, C.R. (2008) In vivo imaging of membrane-associated glycans in developing zebraﬁsh. Science, 320 (5876), 664–667. Codelli, J.A., Baskin, J.M., Agard, N.J., and Bertozzi, C.R. (2008) Second-generation diﬂuorinated cyclooctynes for copper-free click chemistry. J. Am. Chem. Soc., 130 (34), 11486–11493.

167. K¨ ohn, M. and Breinbauer, R. (2004)

168.

169.

170.

171.

172.

173.

174.

175.

The Staudinger ligation – a gift to chemical biology. Angew. Chem. Int. Ed., 43 (24), 3106–3116. Tsao, M.L., Tian, F., and Schultz, P.G. (2005) Selective staudinger modiﬁcation of proteins containing p-azidophenyl-alanine. ChemBioChem, 6 (12), 2147–2149. Saxon, E. and Bertozzi, C.R. (2000) Cell surface engineering by a modiﬁed staudinger reaction. Science, 287 (5460), 2007–2010. Saxon, E., Luchansky, S.J., Hang, H.C., Yu, C., Lee, S.C., and Bertozzi, C.R. (2002) Investigating cellular metabolism of synthetic azidosugars with the staudinger ligation. J. Am. Chem. Soc., 124 (50), 14893–14902. Prescher, J.A., Dube, D.H., and Bertozzi, C.R. (2004) Chemical remodelling of cell surfaces in living animals. Nature, 430 (7002), 873–877. Song, W., Wang, Y., Qu, J., Madden, M.M., and Lin, Q. (2008) A photoinducible 1,3-dipolar cycloaddition reaction for rapid, selective modiﬁcation of tetrazole-containing proteins. Angew. Chem. Int. Ed., 47 (15), 2832–2835. Song, W., Wang, Y., Qu, J., and Lin, Q. (2008) Selective functionalization of a genetically encoded alkene-containing protein via ‘‘photoclick chemistry’’ in bacterial cells. J. Am. Chem. Soc., 130 (30), 9654–9655. Blackman, M.L., Royzen, M., and Fox, J.M. (2008) Tetrazine ligation: fast bioconjugation based on inverse-electron-demand Diels-Alder reactivity. J. Am. Chem. Soc., 130 (41), 13518–13519. van Kasteren, S.I., Kramer, H.B., Jensen, H.H., Campbell, S.J., Kirkpatrick, J., Oldham, N.J., Anthony, D.C., and Davis, B.G. (2007) Expanding the diversity of chemical protein modiﬁcation allows post-translational mimicry. Nature, 446 (7139), 1105–1109.

91

93

Part II Drug Delivery

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

95

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery Pilar Ruiz-S´anchez

4.1 Introduction

Cyanocobalamin (vitamin B12 ) is an essential, water-soluble vitamin for humans. This vitamin plays an important role in enzymatic processes in the mitochondria, cell nucleus, and cytoplasm [1]. It is necessary for the synthesis of red blood cells [2], maintenance of the nervous system [3], and growth and development in children [4]. Historically, B12 was discovered from the disease pernicious anemia [5, 6]. This disease was fatal until Minot and Murphy found that ingesting liver seemed to cure the disease. Castle observed a protein in the stomach which enhanced the curative effect of liver. He called this protein intrinsic factor (IF) [7]. After these ﬁndings, and for the next 20 years, liver was the only source of treatment. Several groups started to look for the curative substance, and ﬁnally, two independent researchers, Folkers and Smith, isolated the red crystalline B12 from liver [8–11]. Since 1952, vitamin B12 analogs had been isolated and characterized from natural sources and by biosynthesis [12–15]. In 1964, Dorothy Crowfoot Hodgkin was awarded the Nobel prize in chemistry after determining the crystal structure of vitamin B12 , as well as the structure of cholesterol, penicillin, and insulin [16]. In 1971, Woodward and Eschenmoser were able, in a cooperative effort, to chemically synthesize B12 [17, 18]. The biosynthetic pathway of the corrin ring in the aerobic Pseudomonas denitriﬁcans and in the anaerobic Propionibacterium shermanii was accomplished [19, 20]. These bacteria are very efﬁcient producers of vitamin B12 , making the vitamin accessible for medical application and scientiﬁc research. The discovery of the coenzymatic function of B12 by Barker et al. created a new ﬁeld of research based on the molecular function of coenzyme B12 and isolation of related enzymes [21, 22]. The biological function of B12 transport proteins was mainly studied by medical laboratories, in an effort to understand the crucial role that these proteins play in binding B12 released from food and transporting it across the cell membrane [23–25]. Deﬁciency in absorption, transportation and/or metabolism of B12 results in intracellular deﬁciency leading to anemia, degradation of nerves, and irreversible neurological damage. Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

HC

HC-B12

Dietary B12

96

H+ IF

IF-B12

Blood

TCII-B12

Figure 4.1 Gastrointestinal pathway for cobalamin absorption from food (HC, haptocorrin; HC-B12 , haptocorrin bound to B12 ; IF, intrinsic factor; IF-B12 , intrinsic factor bound to B12 ; TCII-B12 , transcobalamin II bound to B12 ).

4.2 Transport Mechanism

Humans require vitamin B12 as a cofactor for two essential enzymatic processes. The bioavailability of vitamin B12 is very low. Therefore, mammals have developed a very complex biochemical uptake involving three different transport proteins: transcobalamin I (TC) (also known as haptocorrin (HC)), IF, and TCII (Figure 4.1). HC is present in saliva and carries the released B12 to the stomach. The name HC was introduced to include cobalamin-binding proteins with identical protein backbones that had different names including TCI and TCIII, cobalophilins, R-binder, and saliva binder [26]. Low pH and hydrolytic enzymes in the stomach release B12 from the transport protein HC. In the lumen of the stomach, gastric parietal cells secrete a speciﬁc binding protein for B12 , known as intrinsic factor

4.3 Metabolism of B12 Transport protein (TCII) B12

B12 TCII-B12

Receptor

Receptor

B12 B12

Figure 4.2

Cellular uptake of vitamin B12 (receptor–mediator endocytosis).

[27]. IF binds to B12 only under neutral conditions in the duodenum. After B12 is bound, holoIF becomes resistant to the proteases present in duodenum, which degrade most proteins [28]. The mucosal cells of the small intestine have transport sites for IF-B12 . The receptor protein of the mucosal cells, cubilin, recognizes the IF-B12 complex. The transcytotic mechanism by which cobalamin is transported across ileal epithel cells to the blood is still unknown [29–31]. Once B12 is in the blood, it is captured and transported by TCII to the cells of organs and to the fetus [32]. Although, 80% of the total cobalamins found in the circulation are bound to HC [33], the remaining part is bound to TCII, which mediates the transport of cobalamins into the cell. The transport through the cell membrane is mediated by a speciﬁc protein receptor for TCII-B12 , and B12 is thereby internalized by receptor-mediated endocytosis [34] (Figure 4.2).

4.3 Metabolism of B12

Once cobalamin has entered the cell, TCII is degraded inside the lysosomes to release cobalamin into the cytoplasm. Conversion of B12 to methylcobalamin (MeCbl) and adenosylcobalamin (AdoCbl) provides the cell with the cofactors needed for the cobalamin-dependent reactions (Figures 4.3 and 4.4). 4.3.1 Adenosylcobalamin-Dependent Reactions

To convert vitamin B12 to coenzyme B12 (AdoCbl), the cell has to reduce the cobalt center from Co(III) to Co(I). The ﬁrst step is the reduction of cob(III)alamin to cob(II)alamin, which is achieved nonenzymatically by dihydroﬂavins (FMNH2 , FADH2 ) in the cytosol [35]. The second step is the reduction of cob(II)alamin

97

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

98

Homocysteine

Degraded TCII

MeCbI

MeFH4 Methyl transferase

B12

Methionine B12

FH4

OHCbI (CblIII)

B12

CblIII Reductase CblII Reductase

Mitochondria

CblI Adenosyl transferase AdoCbl

Figure 4.3 Metabolism of cobalamin (MeFH4 , methyltetrahydrofolate; FH4 , tetrahydrofolate; HOCbl, hydroxycobalamin; MeCbl, methylcobalamin; TCII, transcobalamin II).

Ado H2C Co

OH Co

Activation Adenosylmethionine + flavodoxin O CH3 N R HN H2N

N

II

Co

His CH3 Co

COO−

O H2N (a)

N

H N N H

Co

Ado CH2 + H COSCoA CH3 HOOC

Ado CH2 + H COSCoA CH2 H COOH

Ado CH3 + H COSCoA CH2 HOOC

Ado CH3 + COSCoA H CH H COOH

HS COO−

HN

+

NH3+

His

N H

Ado CH2

+

I

R

Co

H3CS

NH3

(b) Figure 4.4 Cobalamin-dependent enzymatic reactions in humans (a: methylcobalamin; b: adenosylcobalamin).

to cob(I)alamin by reduced ﬂavodoxin (FldA), a nonspeciﬁc electron-transfer protein [36, 37]. The ﬁnal process is catalyzed by an adenosine triphosphate (ATP):co(I)rrinoid adenosyltransferase enzyme [38], in humans known as PduO [39]. In the active site of the enzyme, Cob(I)alamin attacks the 5 C of ATP forming a Co–C with 5 -deoxyadenosine yielding AdoCbl and releasing triphosphate as a by-product. AdoCbl is a cofactor for the methylmalonyl-CoA mutase in the mitochondria. This enzyme (EC 5.4.99.2) converts l-methylmalonyl-CoA to succinyl-CoA [40]. This is

4.4 Vitamin B12 Derivatives

an intermediate step in the catabolism of odd-chain fatty acids and branched-chain amino acids. 4.3.2 Methylcobalamin-Dependent Reactions

MeCbl serves as cofactor for the methionine synthase. This enzyme is located in the cytoplasm and methylates homocysteine to form methionine. The methyl group of MeCbl comes from 5-methyltetrahydrofolate. In this reaction, cob(I)alamin acts as an intermediate to transfer the methyl group from 5-methyltetrahydrofolate to homocysteine [41, 42]. In the methyl transfer from MeCbl to homocysteine, MeCbl is in the base-off form (benzimidazole not coordinated to the cobalt center), with a histidine ligand from the protein coordinated to the cobalt in the α-axial position [43]. Corresponding heterolytic cleavage of the Co–C bond yields cob(I)alamin. MeCbl is formed again by methylation of cob(I)alamin with 5-methyltetrahydrofolate closing the catalytic cycle.

4.4 Vitamin B12 Derivatives 4.4.1 Structure

The central part of vitamin B12 consists of a tetrapyrrole corrin ring, with the four nitrogens bound to the cobalt center. Extending down from the corrin ring, attached at the C17 is a phosphoric-5,6-dimethylbenzimidazole chain (α-ligand). The N3 of the benzimidazole is bound to the cobalt center forming a Co–N bond. The upper β-ligand can be occupied by several different ligands, determining the name and properties of vitamin B12 [44]. In mammals, the major forms are hydroxycobalamin, glutathionylcobalamin, MeCbl, and AdoCbl. The industrially produced form of cobalamin is cyanocobalamin. Since the 1990s, syntheses of cobalamin analogs has been a main research target for possible therapeutical applications. The structure of vitamin B12 is depicted in Figure 4.5. 4.4.2 b-, d-, e-Cobalamin Derivatives

Analogs of vitamin B12 with functional groups attached to the propionamide of the corrin ring at b, d, or e positions, have been synthesized by different laboratories. Evaluation of cobalamin analogs in vitro and in vivo have shown interferences with cobalamin metabolism, as evidenced by increased levels of the substrates of the two mammalian enzymes dependent on cobalamin, homocysteine and methylmalonic acid [45]. As mentioned in Section 4.2, vitamin B12 is internalized into the cell by the transport protein TCII. Modiﬁcations of cobalamin must not affect its

99

100

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

H2NOC

b

c

H3C

H2NOC

CONH2 CH3

a

A

H3C H3C H

N

d

CONH2

N Co

N H2NOC

B

R

N

D

g

CH3

C

CH3

H3C

CH3

e

f

O

HN H3C

HO

H O P O Figure 4.5

O O

N

CONH2 CH3

R Short form =

N

Co

CH3 O

OH

Molecular structure of vitamin B12 .

interactions with this protein to ensure transport. Competitive binding studies of b-, d-, e-cobalamin derivatives with TCII showed lower-afﬁnity binding with TCII than native vitamin B12 , for instance, b-acid conjugation on B12 shows a 2.6% binding inhibition to TCII as compared to pure B12 in a competitive binding assay. The corresponding relative inhibitions constants for d-acid and e-acid conjugation are 2.3 and 33%, respectively [46]. These results explained the in vivo effects on B12 metabolism due to low TCII afﬁnity, and therefore low B12 uptake by the cells. However, several derivatives at these positions had been synthesized for therapeutic uses showing high tumor accumulation in mice [47, 48]. In the 1950s several researchers demonstrated that vitamin B12 can be labeled with cobalt radionuclides (57 Co, 58 Co, 60 Co) [49–52]. These derivatives are attractive for biochemical studies; but the long biosynthesis process, high cost, and long half-life times of the radioisotopes, limited the dose administrated to humans, and excluded external imaging of biodistribution. Developing a new method for radiolabeling B12 is still of great interest. While using arylstannane derivatives conjugated to the cobalamin moiety attached by a linker to the b-, d-, e-propionamide chain of cobalamin, Morgan et al. demonstrated that radioiodination is facile and allowed preliminary biodistribution studies in mice [53]. Collins and Hogenkamp developed diethylenetriaminepentaacetate (DTPA) cobalamin analog for the purpose of imaging TCII receptors in malignant and nonmalignant tissue [54]. Labeling of DTPA–cobalamin conjugates with 111 In allowed biodistribution studies in mice and pigs, and later in patients diagnosed with different tumors [47, 55]. Thereafter, Collins et al. patented a series of novel cobalamin conjugates linked via one or more peptide residues, wherein at least one of the peptides is linked to chelating

4.4 Vitamin B12 Derivatives

groups, which are attached to radionuclides. As an alternative, peptide residues comprising one or more nonmetallic radionuclides (11 C, 18 F, 76 Br, 123 I, 124 I) had been proposed. These derivatives are useful for tumor imaging [56]. Cobalamin linked to a residue of a chemotherapeutic agent (doxorubicin, paclitaxel) [57] or molecule comprising 10 B or 157 Gd [58], useful for treatment of abnormal cellular proliferation, had been also patented by Collins et al. Russell-Jones et al. had shown potential cellular uptake of ﬂuorescent polymers and nanoparticles by their covalent binding to B12 , with the idea of oral delivery of drugs by the B12 transport system [59–61]. Innovative studies carried out by Alberto et al. have demonstrated, by abolishing interaction with the transport protein TCII, preferential accumulation of the radiolabeled B12 derivative, 99m Tc-PAMA4 -B12 , in mice is observed in cancer tissue [48, 62]. 4.4.3 Modiﬁcations on the Ribose Moiety

Owing to the low synthetic yields for the derivatization of the b, d, e side chain acids of B12 , the 5 -hydroxyl moiety of the ribofuranoside has been the most productive position for additional conjugation. For using vitamin B12 derivatives as drug delivery agents, high afﬁnity for TCII is essential. Upon derivatization of the 5 -hydroxyl at the ribose group, B12 conjugates have been shown to essentially retain the binding afﬁnity to TCII [46]. Crystallographic data obtained from human TCII assessed that the 5 -hydroxyl group on the ribose of B12 can accommodate the attachment of a larger ligand without disturbing the interactions between TCII and B12 [63]. The ﬁrst ribose derivative was synthesized in 1975, by reaction of vitamin B12 with succinic anhydride. The ﬁrst succinylation takes place on the 5 -OH of the ribose, and, more slowly, on 2 -OH. They can be easily backconverted into B12 by mild acid or base hydrolysis [64]. Russell-Jones et al. proposed a new procedure to attach spacers to the ribose group of B12 . This method followed preactivation of the 5 -OH on the ribose by the use of 1,1 -carbonyldiimidazole, 1,1 -carbonyldi-(1,2,4-triazole), or di(1-benzotriazolyl) carbonate with subsequent nucleophilic attack to introduce the desired spacer. The 5 -carbamate derivatives are very stable under physiological conditions. Carbamates are poorly cleaved by esterases and not hydrolyzed by the acidic pH in the stomach. These new B12 conjugates had higher afﬁnity to IF and HC than the corresponding e-acid conjugates. Therefore, they were suitable for oral delivery [65]. Following this derivatization method, Russell-Jones et al. developed an oral delivery system in which the B12 -based drug was encapsulated within B12 -coated nanospheres that could be targeted to the circulation by the B12 -IF pathway. The nanoparticle surface was modiﬁed with succinic anhydride and conjugated with amino group of B12 -carbamate derivatives. These nanoparticle derivatives were loaded with insulin, and were found to avoid decomposition of insulin against gut proteases. Oral administration to diabetic rats showed reduction of plasma glucose [66, 67].

101

102

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

A novel ﬂuorescent cobalamin analog was proposed by Fedosov et al. as a tool for analysis of the binding properties of cobalamin derivatives to TCII and IF. This ﬂuorescent cobalamin was prepared by coupling 5-carboxyrhodamine succinimidil ester to an amino derivative of B12 modiﬁed at 5 -OH of the ribose [68]. This new strategy became an alternative to radioactive 57 Co–B12 for measuring and calculating protein binding constants. 4.4.4 β-Axial Position 4.4.4.1 Cobalamin Alkylation According to the natural model, the most attractive modiﬁcation of B12 is alkylation of the cobalt center, producing a stable but light-sensitive cobalt–carbon bond. The ﬁrst cobalamin derivatives following this strategy were reported by Jacobsen et al. in 1979. They synthesized a set of coenzyme analogs in which the β-axial ligand was replaced by ﬂuorescent moieties. Analogs of AdoCbl were prepared by using ﬂuorescent nucleosides and a MeCbl congener was synthesized by introduction of a dansylamidopropyl group. In all cases, ﬂuorescence was switched on after cleavage of the cobalt–carbon bond by photolysis or cyanolysis, making these derivatives interesting for studying the enzymatic mechanism involving the cobalt–carbon bond [69]. The attachment of ligands at the β-axial position had shown little effect on binding with TCII [46]. Grissom et al. followed this strategy to synthesize a series of cobalamin analogs containing cytotoxic therapeutic agents (chlorambucil, doxorubicin, or colchicines) and ﬂuorescent markers (Oregon green, ﬂuorescein, or naphthoﬂuorescein) [70–74]. To trigger ligand release from cobalamin, Grissom proposed four different methods: photolysis, which produced homolytic cleavage of the Co–C bond; reducing conditions to produce Co(II) and the corresponding radical ligand; sonolysis yielding the cleavage of the Co–C bond by an indirect process mediated by the radical formation of H· and OH· ; or by hydrazone-containing cobalamin conjugate, which hydrolyzed within lysosomes inside the cell due to slightly acidic pH [72, 74]. New oligomethylene-bridged vitamin B12 dimers [75] and a nonnatural B12 rotaxane were reported by Kr¨autler et al. and applied as ‘‘molecular springs’’ [76]. Conjugation of vanadium for oral treatment of diabetes has been reported by Brasch et al. Since vanadium(V) complexes are known to be reduced to vanadium(IV) inside cells [77] thereby provoking insulin-enhancing properties [78]. They linked 3-hydroxy-2-methyl-1-propyl-1H-pyridin-4-one to cobalamin and vanadium(V) via the β-axial site of cobalamin [79]. The anesthetic gas nitrous oxide (N2 O) has been noted to cause cobalamin disturbance in vivo. Prolonged administration of N2 O can result in megaloblastic anemia and neural abnormalities [80, 81]. It is not surprising that nitric oxide (NO) can also interact with cobalamins. Nitrosylcobalamin has been widely studied to determine if NO alters the enzymatic activity of cobalamins [82–84]. However, NO is an ideal chemotherapeutic agent [85]. Therefore, Bauer et al. proposed the use of nitrosylcobalamin as a tumor targeting drug [86, 87].

4.4 Vitamin B12 Derivatives

The only functionality that has not received much attention is the cyanide group coordinated to Co(III) in cyanocobalamin (CNCbl). It is well known that M–CN moieties tend to bridge two metal centers to form the M–CN–M motif [88–92]. Examples with tetrapyrrols such as porphyrines are, however, rare [93–95]. The inverse situation with respect to cyanide has been studied in great detail and [Fe(CN)6 ]4− or [Fe(CN)5 (NO)]2− were coordinated to Co(III) of aquocobalamin (Figure 4.6) [96, 97]. 4.4.4.2 Heterodinuclear Concept To the best of our knowledge, the ﬁrst examples of cobalamin-CN-{M} derivatives were reported by Kunze and Alberto et al. in 2004. Their interest lies in radiopharmaceuticals containing the fac-[99m Tc(H2 O)(L2 )(CO)3 ] unit, where L2 is a bidentate ligand. The remaining water molecule can be replaced by the nitrogen atom from cyanide bound to Co(III) in vitamin B12 [98]. This strategy allowed the direct labeling of B12 for radiodiagnosis, without disrupting the corrin ring and, therefore, allowing recognition by the transport proteins. While using the core fac-[Re(CO)3 ]+ variety of drugs or biomolecules can be coordinated, with rhenium acting as a mediator between B12 and the additional molecule. The Tc/Re examples are attractive since they imply the application of other metal complexes along the same line, binding one vacant site of a metal complex to the Co(III)-bound cyanide. Cisplatin, for example, is well known in cancer therapy; however, its low selectivity causes severe side effects [99]. To enhance speciﬁcity by targeting tumor cells is one of the main objectives, nowadays, to improve the therapeutic index of cytotoxic drugs [100]. Coordination of B12 to platinum(II) complexes was described by Wilson et al. via linkers bound to the e-acid in cobalamins [101]. Interactions of the activated form of cisplatin, [Pt(H2 O)2 (NH3 )2 ]2+ , with adenosylcobalamin resulted in platinum coordination to the N7 or N1 of the adenosyl unit. In MeCbl, cobalamin base-off is formed with cisplatin coordinated to the N3 of dimethylbenzimidazole in the back loop [102]. Binding to cyanide was not considered or not observed. We could not ﬁnd binuclear complexes in literature containing the structural {Co–CN–Pt} arrangement. A few examples were reported for the {Co–NC–Pt} complexes [103, 104], and only one for the case where cobalamin is attached via the cyanide bridge to tetracyanoplatinate(II) [105]. Mundwiler et al. reported the ﬁrst cyanide-bridged vitamin B12 –cisplatin conjugate by reacting the monoactivated form of cisplatin, cis-[PtCl(H2 O)(NH3 )2 ]+ , with cyanocobalamin [106]. Further studies followed this discovery by reacting trans-[PtCl(H2 O)(NH3 )2 ]+ , K[PtCl3 (NH3 )], and K2 [PtCl4 ] with vitamin B12 (Scheme 4.1) [107]. Applicability of these heterodinuclear conjugates depends primarily on serum stability. It is well known that cisplatin and transplatin bind very rapidly to human serum albumin (HSA) [108]. After intravenous infusion of cisplatin, 65–98% of platinum in blood plasma is protein-bound [108, 109]. Many studies have demonstrated that the most favorable site for coordination in HSA is cysteine [110, 111]. This process is essentially irreversible. Cisplatin bound to albumin may have

103

104

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

H2NOC

In111-DTPA H3C

H2NOC H3C H 3C H

N

H 3C O

HO P

O

N

CONH2 CH3

N

CH3

N N

O

N

O

N

HO

N

P

O

Collins and Hogenkamp et al.

CH3 CH3 CONH2 CH3 CH3

O

O

O

CONH2

N

CH3

HN H3C H

CN Co

H 3C

O OH

O

O

CH3 CH3

CH3

HN O

H3C H3C H H2NOC

CONH2

N

Co N

CONH2 CH3

H3C

H2NOC

CH3

CN

N

H2NOC

H3C H

CONH2

O

O

NH(CH2)xNHCO

Russell-Jones et al.

nanoparticle

O N N H H2NOC

H3C H3C H

N N

O

P

N

OCH3

CH3 N

HO

N

CH3

O

O OH

O

O

Grissom et al.

O CH2OH

CO CONH2 CH3

OH

OHOH

OCH3 OCH3

CH3 CH3

O

HN O

O

CONH2

N

OCH3

NH

Co

H2NOC H3C

H3C H

CONH2 CH3

H3C

H2NOC

Co

O

O

O Tc99m N

OC OC H2NOC

N N CONH2 CH3 CH3 C

H2NOC H 3C H3C H

N

Kräutler et al.

Co

H3 C

N CH3

CH3 CH3

O

N

CONH2 CH3

HO

N

CH3

HN O

Co

CONH2

N

N

H2NOC

H3C H

O

O

O OH Kunze and Alberto et al.

O

Figure 4.6

P

O

Structures overview of vitamin B12 derivatives reported by different authors.

4.4 Vitamin B12 Derivatives

+

NH3 H3N

Pt

Cl

N

+

+

H2O Cl Pt NH3

H 2O H3N Pt NH3

NH3

Cl

+

Cl H3N

Pt

NH3

N C

C Co

Co

N C Co

Cl H3N

Pt

NH3 Cl

Cl

N

N C Co

Pt

+

Cl

Cl Cl Pt Cl

−

NH3

Cl Cl Pt Cl Cl

2− Cl

−

Cl Pt

Cl

N

C

C

Co

Co

Scheme 4.1 Syntheses of vitamin B12 –Pt(II) conjugates with cis-[PtCl(H2 O)(NH3 )2 ]+ , trans[PtCl(H2 O)(NH3 )2 ]+ , K[PtCl3 (NH3 )], and K2 [PtCl4 ].

lower antitumor activity [112, 113], although the corresponding data reported are sometimes controversial [111]. Two kinds of interactions with serum proteins for B12 –Pt(II) conjugates can be expected. Noncovalent lipophilic binding is usually a fast and reversible process, detectable shortly after incubation. Covalent, and therefore, irreversible binding through coordination to platinum is slow and evolves with time as observed for cisplatin. It should be noted here that free B12 has almost no interaction with HSA (only 1–2% binding) Conjugates [{Co}-CN-{cis-PtCl(NH3 )2 }]+ , [{Co}-CN-{trans-PtCl(NH3 )2 }]+ , and [{Co}-CN-{cis-PtCl2 (NH3 )}] were incubated with 1 equiv. of bovine serum albumin (BSA) in phosphate buffer (10 mM, pH 7.4, 0.8% NaCl). Stability under these conditions was investigated with two different HPLC (high performance liquid chromatography) systems. Reversed-phase HPLC determined the authenticity of B12 conjugates and size-exclusion HPLC quantiﬁes the ratio of BSA bound/unbound B12 derivatives. Protein association of B12 –Pt(II) derivatives was about 9–20% after 48 hours. The appearance of free B12 would be indicative of the decomposition product of Pt(II) cleavaged from B12 . Around 20% of CNCbl could be detected after 48 hours, while the remaining 80% were still the respective B12 conjugates. The low protein-binding values observed for B12 –Pt(II) conjugates indicated that vitamin B12 protects cisplatin from serum interactions. Axial coordination to the cobalt center depends on pH and the formal oxidation state of the cobalt ion. In the thermodynamically predominant forms of cobalt

105

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

106

e−

R III

Co

e− II

FIdA-Fpr

Co

I

FIdA/FMNH

Co

CobA + ATP

Ado III

Co

Scheme 4.2 Corrinoid adenosylation pathway: electrons for reduction of cob(III)alamin to cob(I)alamin are from nicotinamide adenine dinucleotide (NAD(P)H) through ﬂavodoxin proteins. The adenosylation process with ATP takes place in the active site of CobA enzyme.

corrins, both axial ligands are bound to the Co(III) center yielding a hexacoordinated complex. The reduction to Co(II) induces the release of one of the axial ligands yielding a pentacoordinated complex. Further reduction to Co(I) yields a tetracoordinated complex (Co(I) is d8 ) where the axial ligands are absent [114, 115]. Therefore, release of the platinum complexes from B12 should be driven by intracellular reduction of Co(III) → Co(II) → Co(I) and subsequent adenosylation by the adenosyltransferase enzyme (Figure 4.3, Scheme 4.2). A crucial result of further studies was the electrochemical behavior of the cobalt center, and therefore, the possible release of platinum complexes. The cyclic voltametry spectrum of vitamin B12 exhibited a reduction wave at E1/2 = −670 mV (vs. Ag/AgCl) for the reduction from Co(III) to Co(II). However, the corresponding spectra of derivatives [{Co}-CN-{cis-PtCl(NH3 )2 }]+ , [{Co}-CN-{trans-PtCl(NH3 )2 }]+ , and [{Co}-CN-{cis-PtCl2 (NH3 )}] showed more positive potentials (E1/2 = −515,−486, and −584 mV, respectively). A shift of the electrochemical potentials toward more positive values supports the theoretical ability to perform enzymatic reduction in adenosylation assays. For the purpose of targeted metal complex delivery by using the heterodinuclear approach, after reduction of the cobalt center from Co(III) to Co(I), the axial ligands should be released. If this was the case, enzymatic corrinoid adenosylation assays would yield AdoCbl, as a further conﬁrmation of platinum complexes cleavage. Since corrinoid adenosyltransferase CobA from Salmonella enterica is a well-established system [36, 116, 117], the behavior of B12 –Pt(II) conjugates was investigated with these conditions (Scheme 4.2). The corrinoid adenosylation system requires the ﬂavin mononucleotide–dependent reductase, the electron-transfer protein FldA, and the ferrodoxin (Fpr) as reductases for the steps Co(III) → Co(II) and Co(II) → Co(I), respectively. In the ﬁnal process, Co(I) is adenosylated by the S. enterica CobA enzymes using ATP as a substrate, via oxidative addition. Enzymatic corrinoid adenosylation assays of B12 –Pt(II) adducts with CobA showed recognition and conversion to AdoCbl and release of Pt(II) species (see Figure 4.7) [107]. Although, reduction of Co(III) to Co(II), under physiological pH, released the β-axial [115], in this case of the platinum derivative, the formation of AdoCbl is an experimental probe of platinum cleavage. The conﬁrmation of AdoCbl formation during the process was obtained by in vitro experiments. The S. enterica strain JE7179 grows only in the presence of AdoCbl. The absence of CobA enzyme in this bacterial strain avoids transformation of

4.4 Vitamin B12 Derivatives

107

OH OH N N

O

N

H2N

N H2NOC CH3

H2NOC H3C H3C H

1. NADPH + FldA + Fpr

N

CONH2

N

N

H3C

N

CH3 CH3

CH3

Pt release HN H3C H O O

0.25

Co →Co III

II

0.2

0.2

O

N

HO

N

O O

CONH2 CH3 CH3

O OH

Co →AdoCbl II

0.15 A

A

0.15 0.1

0.1 0.05

0.05 0 400

P

CH3

Co

H2NOC

2. CobA + ATP

CONH2

450

500 nm

550

600

0 400

450

500

550

600

nm

Figure 4.7 Spectral changes associated with the reduction of [{Co}-CN-{cis-PtCl2 (NH3 )}] from cob(III)alamin to cob(II)alamin (monitored every 10 minutes), and subsequent conversion of cob(II)alamin to adenosylcobalamin in the presence of ATP and CobA enzyme (monitored every 5 minutes).

naturally occurring vitamin B12 into the corresponding coenzyme, stopping cellular metabolism. Strong growth was observed with AdoCbl generated in adenosylation assays using B12 –Pt(II) conjugates as a substrates. These data further conﬁrmed the release of the β-axial ligand from vitamin B12 with the formation of AdoCbl. This Pt conjugation to B12 can be used for speciﬁc targeting, since fast proliferating cells, such as tumor cells or bacterial infections, overexpress B12 receptors to accumulate high amounts of this vitamin and to satisfy their high B12 demand. In the heterodinuclear concept, B12 acts as a simple ligand providing water solubility and targeting capacity of cisplatin. This approach could help to reduce side effects and to increase quality of life of the patients. A novel radiolabeled B12 pathway, without modiﬁcations of the B12 molecule was proposed by Ruiz-S´anchez et al. As mentioned in Section 4.4.2, the long-lived 57 Co radionuclide limited the use of 57 Co–vitamin B12 in humans, and for imaging the B12 metabolism in tumor cells. The cisplatin adduct of vitamin B12

650

108

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

[{Co}-CN-{cis-PtCl(NH3 )2 }]+ , reacted with iodide in aqueous solution in good yield. Since iodide exerts a strong trans effect, the amine trans to the ﬁrst substituted iodide was replaced by a second iodide to produce the bis-substituted product [{Co}-CN-{cis-PtI2 (NH3 )}]. Labeling could be achieved by exchange of Cl− for 131 I in [{Co}-CN-{cis-PtCl(NH3 )2 }]+ . Since labeling with 131 I is carried out at nanomolar concentrations, bis-substitution was not observed, and the main product of the reaction was [{Co}-CN-{cis-Pt131 I(NH3 )2 }]+ [118]. Furthermore, the chloride of [{Co}-CN-{cis-PtCl(NH3 )2 }]+ can be exchanged by donors that enable introduction of ﬂuorescent markers [119], antibiotics, or other biologically active molecules. Drug resistance in bacteria has become an increasing problem. The main reasons known for antibiotic resistance are based on drug inactivation, mutations, permeability, and the so-called efﬂux pump effect, based on the expulsion of antibiotics outside the cell. Focusing on efﬂux pump and permeability, and the high demand of B12 in bacteria [120], it could be expected that hiding antibiotics by coupling them with vitamin B12 , will incorporate the new derivatives into bacteria, via cobalamin recognition, independent of resistance. These experiments are currently under investigation.

4.5 Outlook

The development of novel derivatives of vitamin B12 for targeted drug delivery is a major focus in B12 chemistry and biology. The high uptake by fast proliferating cells makes this vitamin attractive for therapeutical application. Modiﬁcations on the corrin ring, ribose, and β-axial positions have been widely explored by several research groups. Some of these derivatives are potential targeting molecules for drug delivery and actually are in clinical trials. The cyanide of vitamin B12 as a conjugation group merits more attention and will be further explored. It is attractive as a linking group for derivatization with biologically or active molecules or, if ﬂuorescent, for analytical purposes to follow the behavior of B12 in vitro and in vivo. Platinum as a cytotoxic metal is not limiting our purposes. From coordination chemistry considerations, one might predict that other metal complexes will react similarly, and ruthenium for example, might be an attractive candidate to extend the concept presented here. As a conclusion, vitamin B12 is an attractive ‘‘Trojan horse’’ for therapeutic drug delivery.

Acknowledgments

The author thanks Prof. Dr. Roger Alberto for his support and enthusiasm during the PhD study. Prof. Dr. Jorge C. Escalante-Semerena, Dr. Nicole Buan, and Paola

References

Mera from Department of Bacteriology, University of Wisconsin at Madison, are acknowledged for their support with adenosylation experiments, and Dr. Stefano Ferrari and Christiane K¨onig from the Institute of Molecular Cancer Research, University of Z¨urich, are acknowledged for their great help in biological in vitro cell experiments. Financial support from the Swiss National Science Foundation is also acknowledged.

References 1. Battersby, A.R. (1994) How nature

2.

3.

4.

5.

6.

7.

8.

9.

10.

builds the pigments of life: the conquest of vitamin B12 . Science, 264, 1551–1557. Vijayaraghavan, P.K. and Dunn, M.S. (1951) Effect of vitamin B12 , vitamin B12a and vitamin B12b on red blood cell counts in experimental anemia in mice. Arch. Biochem. Biophys., 31, 248–250. Robertson, K.R., Stern, R.A., Hall, C.D., Perkins, D.O., Wilkins, J.W., Gortner, D.T., Donovan, M.K., Messenheimer, J.A., Whaley, R., and Evans, D.L. (1993) Vitamin B12 deﬁciency and nervous system disease in HIV infection. Arch. Neurol., 50, 807–811. Black, M.M. (2008) Effects of B12 and folate deﬁciency on brain development in children. Food Nutr. Bull., 29, S126–S131. Okuda, K. (1999) Discovery of vitamin B12 in the liver and its absorption factor in the stomach: a historical review. J. Gastroenterol. Hepatol., 14, 301–308. Chanarin, I. (2000) A history of pernicious anaemia. Br. J. Haematol., 111, 407–415. Strauss, M.B. and Castle, W.B. (1932) The extrinsic (deﬁciency) factor in pernicious and related anaemias. Lancet, 2, 111–115. Murphy, W.P. and Minot, G.R. (1926) A special diet for patients with pernicious anemia. Boston Med. Surg. J., 195, 410–411. West, R. (1948) Activity of Vitamin B12 in addisonian pernicious anemia. Science, 107, 398–398. Rickes, E.L., Brink, N.G., Koniuszy, F.R., Wood, T.R., and Folkers, K.

11.

12.

13.

14.

15.

16.

17.

18.

(1948) Crystalline Vitamin B12 . Science, 107, 396–397. Smith, E.L. (1948) Puriﬁcation of antipernicious anaemia factors from liver. Nature, 161, 638–639. Bernhauer, K., Becher, E., and Wilharm, G. (1959) Biosynthesen ¨ in der ‘‘Cobalamin’’-Reihe. V. Uber die Umwandlung verschiedener Cobalamin-Analoga in Vitamin B12 . Arch. Biochem. Biophys., 83, 248–258. Koppenhagen, Vb., Elsenhans, B., Wagner, F., and Pﬁffner, J.J. (1974) Methylrhodibalamin and 5 -deoxyadenosylrhodibalamin, rhodium analogs of methylcobalamin and cobalamin coenzyme. J. Biol. Chem., 249, 6532–6540. Kolhouse, J.F. and Allen, R.H. (1977) Absorption, plasma transport, and cellular retention of cobalamin analogs in rabbit. Evidence for existence of multiple mechanisms that prevent absorption and tissue dissemination of naturally occurring cobalamin analogs. J. Clin. Invest., 60, 1381–1392. Kolhouse, J.F. and Allen, R.H. (1978) Isolation of cobalamin and cobalamin analogs by reverse afﬁnity chromatography. Anal. Biochem., 84, 486–490. Hodgkin, D.C. (1965) The X-ray analysis of complicated molecules. Science, 150, 979–988. Woodward, R.B. (1979) Synthetic vitamin B12 , in Vitamin B12 (eds B. Zagalak and W. Friedrich), Walter de Gruyter, Berlin, pp. 37–87. Eschenmoser, A. (1979) Chemical synthesis of corrinoids: current problems and recent advances, in Vitamin B12 (eds B. Zagalak and W. Friedrich), Walter de Gruyter, Berlin, pp. 89–117.

109

110

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery 19. Battersby, A.R. and Leeper, F.J. (1999)

20.

21.

22.

23.

24.

25.

26.

27.

28.

Biosynthesis of B12 in the aerobic organism Pseudomonas denitriﬁcans, in Chemistry and Biochemistry of B12 (ed. R. Banerjee), John Wiley & Sons, Inc., New York, pp. 507–535. Scott, A.I. (1993) How nature synthesizes vitamin B12 . A survey of the last four billion years. Angew. Chem. Int. Ed. Engl., 32, 1223–1243. Barker, H.A., Smyth, R.D., Weissbach, H., Munchpetersen, A., Toohey, J.I., Ladd, J.N., Volcani, B.E., and Wilson, R.M. (1960) Assay, puriﬁcation, and properties of the adenylcobamide coenzyme. J. Biol. Chem., 235, 181–190. Barker, H.A., Smyth, R.D., Weissbach, H., Toohey, J.I., Ladd, J.N., and Volcani, B.E. (1960) Isolation and properties of crystalline cobamide coenzymes containing benzimidazole or 5,6-dimethylbenzimidazole. J. Biol. Chem., 235, 480–488. Ellenbogen, L., Chow, B.F., and Williams, W.L. (1958) Effect of intrinsic factor on absorption of vitamin B12 in healthy individuals. Am. J. Clin. Nutr., 6, 26–29. Highley, D.R. and Ellenbogen, L. (1964) A rapid and simpliﬁed method for preparing puriﬁed intrinsic factor. Arch. Biochem. Biophys., 107, 298–301. Nexo, E. and Andersen, J. (1977) Unsaturated and cobalamin saturated transcobalamin I and transcobalamin II in normal human plasma. Scand. J. Clin. Lab. Invest., 37, 723–728. Morkbak, A.L., Poulsen, S.S., and Nexo, E. (2007) Haptocorrin in humans. Clin. Chem. Lab. Med., 45, 1751–1759. Levine, J.S., Nakane, P.K., and Allen, R.H. (1980) Immunocytochemical localization of human intrinsic factor: the nonstimulated stomach. Gastroenterology, 79, 493–502. Allen, R.H., Seetharam, B., Podell, E., and Alpers, D.H. (1978) Effect of proteolytic enzymes on the binding of cobalamin to R protein and intrinsic factor. In vitro evidence that a failure to partially degrade R protein is responsible for cobalamin malabsorption in

29.

30.

31.

32.

33.

34.

35.

36.

37.

pancreatic insufﬁciency. J. Clin. Invest., 61, 47–54. Moestrup, S.K., Kozyraki, R., Kristiansen, M., Kaysen, J.H., Rasmussen, H.H., Brault, D., Pontillon, F., Goda, F.O., Christensen, E.I., Hammond, T.G., and Verroust, P.J. (1998) The intrinsic factor-vitamin B12 receptor and target of teratogenic antibodies is a megalin-binding peripheral membrane protein with homology to developmental proteins. J. Biol. Chem., 273, 5235–5242. Birn, H., Verroust, P.J., Nexo, E., Hager, H., Jacobsen, C., Christensen, E.I., and Moestrup, S.K. (1997) Characterization of an epithelial similar to 460-kDa protein that facilitates endocytosis of intrinsic factor-vitamin B12 and binds receptor-associated protein. J. Biol. Chem., 272, 26497–26504. Kapadia, C.R., Serﬁlippi, D., Voloshin, K., and Donaldson, R.M. (1983) Intrinsic factor-mediated absorption of cobalamin by guinea pig ileal cells. J. Clin. Invest., 71, 440–448. Dencker, L. (1983) Placental accumulation of 57 Co-vitamin B12 in mice studied by light- and electron-microscopic autoradiography. Placenta, 4, 207–215. Hall, C.A. (1977) The carriers of native vitamin B12 in normal human serum. Clin. Sci. Mol. Med., 53, 453–457. Amagasaki, T., Green, R., and Jacobsen, D.W. (1990) Expression of transcobalamin II receptors by human leukemia K562 and HL-60 cells. Blood, 76, 1380–1386. Fonseca, M.V. and Escalante-Semerena, J.C. (2000) Reduction of cob(III)alamin to cob(II)alamin in Salmonella enterica serovar typhimurium LT2. J. Bacteriol., 182, 4304–4309. Fonseca, M.V. and Escalante-Semerena, J.C. (2001) An in vitro reducing system for the enzymic conversion of cobalamin to adenosylcobalamin. J. Biol. Chem., 276, 32101–32108. Hall, D.A., Vander Kooi, C.W., Stasik, C.N., Stevens, S.Y., Zuiderweg, E.R.P., and Matthews, R.G. (2001) Mapping the interactions between ﬂavodoxin and its physiological partners ﬂavodoxin reductase and cobalamin-dependent

References

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

methionine synthase. Proc. Natl. Acad. Sci., 98, 9521–9526. Walker, G.A., Murphy, S., and Huennekens, F.M. (1969) Enzymatic conversion of vitamin B12a to adenosyl-B12 : Evidence for existence of two separate reducing systems. Arch. Biochem. Biophys., 134, 95–105. Leal, N.A., Park, S.D., Kima, P.E., and Bobik, T.A. (2003) Identiﬁcation of the human and bovine ATP : Cob(I)alamin adenosyltransferase cDNAs based on complementation of a bacterial mutant. J. Biol. Chem., 278, 9227–9234. Banerjee, R. (ed) (1999) Chemistry and Biochemistry of B12 , John Wiley & Sons, Inc., New York. Gulati, S., Chen, Z.Q., Brody, L.C., Rosenblatt, D.S., and Banerjee, R. (1997) Defects in auxiliary redox proteins lead to functional methionine synthase deﬁciency. J. Biol. Chem., 272, 19171–19175. Banerjee, R. (1997) The Yin-Yang of cobalamin biochemistry. Chem. Biol., 4, 175–186. Drennan, C.L., Huang, S., Drummond, J.T., Matthews, R.G., and Ludwig, M.L. (1994) How a protein binds B12 : a 3.0 Angstrom X-Ray structure of B12 -binding domains of methionine synthase. Science, 266, 1669–1674. Kr¨autler, B. (2005) Vitamin B12 : chemistry and biochemistry. Biochem. Soc. Trans., 33, 806–810. Stabler, S.P., Brass, E.P., Marcell, P.D., and Allen, R.H. (1991) Inhibition of cobalamin-dependent enzymes by cobalamin analogs in rats. J. Clin. Invest., 87, 1422–1430. Pathare, P.M., Wilbur, D.S., Heusser, S., Quadros, E.V., McLoughlin, P., and Morgan, A.C. (1996) Synthesis of cobalamin-biotin conjugates that vary in the position of cobalamin coupling. Evaluation of cobalamin derivative binding to transcobalamin II. Bioconjugate Chem., 7, 217–232. Collins, D.A., Hogenkamp, H.P.C., and Gebhard, M.W. (1999) Tumor imaging via indium 111-labeled DTPA-adenosylcobalamin. Mayo Clin. Proc., 74, 687–691.

48. Waibel, R., Treichler, H., Schaefer,

49.

50.

51.

52.

53.

54.

55.

56.

N.G., van Staveren, D.R., Mundwiler, S., Kunze, S., Kuenzi, M., Alberto, R., Nuesch, J., Knuth, A., Moch, H., Schibli, R., and Schubiger, P.A. (2008) New derivatives of vitamin B12 show preferential targeting of tumors. Cancer Res., 68, 2904–2911. Anderson, R.C. and Delabarre, Y. (1951) The preparation of radioactive vitamin B12 by direct neutron irradiation. J. Am. Chem. Soc., 73, 4051–4052. Smith, E.L., Hockenhull, D.J.D., and Quilter, A.R.J. (1952) Tracer studies with the vitamins B12 . II. Biosynthesis of vitamin B12 labelled with 60Co and 32P. Biochem. J., 52, 387–388. Smith, E.L. (1952) Tracer studies with the vitamins B12 . I. Neutron irradiation of vitamin B12 . Biochem. J., 52, 384–387. Chaiet, L., Rosenblum, C., and Woodbury, D.T. (1950) Biosynthesis of radioactive vitamin B12 Containing Cobalt-60. Science, 111, 601–602. Wilbur, D.S., Hamlin, D.K., Pathare, P.M., Heusser, S., Vessella, R.L., Buhler, K.R., Stray, J.E., Daniel, J., Quadros, E.V., McLoughlin, P., and Morgan, A.C. (1996) Synthesis and nca-radioiodination of arylstannyl-cobalamin conjugates. Evaluation of aryliodo-cobalamin conjugate binding to transcobalamin II and biodistribution in mice. Bioconjug. Chem., 7, 461–474. Collins, D.A. and Hogenkamp, H.P.C. (1997) Transcobalamin II receptor imaging via radiolabeled diethylene-triaminepentaacetate cobalamin analogs. J. Nucl. Med., 38, 717–723. Collins, D.A., Hogenkamp, H.P.C., O’Connor, M.K., Naylor, S., Benson, L.M., Hardyman, T.J., and Thorson, L.M. (2000) Biodistribution of radiolabeled adenosylcobalamin in patients diagnosed with various malignancies. Mayo Clin. Proc., 75, 568–580. Hogenkamp, H.P.C. and Collins, D.A. (2001) Cobalamin conjugates useful as imaging and therapeutic agents, US Patent No. WO0128595.

111

112

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery 57. Hogenkamp, H.P.C. and Collins, D.A.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

(2001) Cobalamin conjugates useful as imaging agents and as antitumor agents, US Patent No. WO0128592. Collins, D.A. and Hogenkamp, H.P.C. (2000) Cobalamin conjugates useful as antitumor agents, US Patent No. WO0062808. Russell-Jones, G.J., Arthur, L., and Walker, H. (1999) Vitamin B12 mediated transport of nanoparticles across Caco-2 cells. Int. J. Pharm., 179, 247–255. Russell-Jones, G., McTavish, K., and McEwan, J. (2003) Enhanced cellular uptake of polymers targeted with vitamin B12 and folate. J. Control. Release, 91, 259–260. Russell-Jones, G., McTavish, K., and McEwan, J. (2003) Enhanced tumour uptake of polymers targeted with vitamin B12 and folate. J. Control. Release, 91, 260–262. Spingler, B., Mundwiler, S., Ruiz-S´anchez, P., van Staveren, D.R., and Alberto, R. (2007) Structures of the b- and d-acid derivatives of vitamin B12 and their complexes with [M(CO)3 ]+ (M = 99mTc, Re). Eur. J. Inorg. Chem., 2007, 2641–2647. Wuerges, J., Garau, G., Geremia, S., Fedosov, S.N., Petersen, T.E., and Randaccio, L. (2006) Structural basis for mammalian vitamin B12 transport by transcobalamin. Proc. Natl. Acad. Sci., 103, 4386–4391. Toraya, T., Ohashi, K., Ueno, H., and Fukui, S. (1975) Preparation, properties and biological activities of succinyl derivatives of vitamin B12 . Bioinorg. Chem., 4, 245–255. McEwan, J.F., Veitch, H.S., and Russell-Jones, G.J. (1999) Synthesis and biological activity of ribose-5 -carbamate derivatives of vitamin B12 . Bioconjug. Chem., 10, 1131–1136. Chalasani, K.B., Russell-Jones, G.J., Yandrapu, S.K., Diwan, P.V., and Jain, S.K. (2007) A novel vitamin B12 -nanosphere conjugate carrier system for peroral delivery of insulin. J. Control. Release, 117, 421–429. Chalasani, K.B., Russell-Jones, G.J., Jain, A.K., Diwan, P.V., and Jain, S.K.

68.

69.

70.

71.

72.

73.

74.

75.

76.

(2007) Effective oral delivery of insulin in animal models using vitamin B12 -coated dextran nanoparticles. J. Control. Release, 122, 141–150. Fedosov, S.N., Grissom, C.B., Fedosova, N.U., Moestrup, S.K., Nexo, E., and Petersen, T.E. (2006) Application of a ﬂuorescent cobalamin analogue for analysis of the binding kinetics. FEBS J., 273, 4742–4753. Jacobsen, D.W., Holland, R.J., Montejano, Y., and Huennekens, F.M. (1979) Cryptoﬂuorescent analogs of cobalamin coenzymes. Synthesis and characterization. J. Inorg. Biochem., 10, 53–65. Hogenkamp, H.P.C., Collins, D.A., Grissom, C.B., and West, F.G. (1999) Diagnostic and therapeutic analogues of cobalamin, in Chemistry and Biochemistry of B12 (ed. R. Banerjee), John Wiley & Sons, Inc., New York, pp. 385–410. Grissom, C.B., West, F.G., McGreevy, J., and Bentz, J.S. (2001) Fluorescent cobalamins and uses thereof, US Patent No. WO0130967. Bagnato, J.D., Eilers, A.L., Horton, R.A., and Grissom, C.B. (2004) Synthesis and characterization of a cobalamin-colchicine conjugate as a novel tumor-targeted cytotoxin. J. Org. Chem., 69, 8987–8996. Smeltzer, C.C., Cannon, M.J., Pinson, P.R., Munger, J.D.J., West, F.G., and Grissom, C.B. (2001) Synthesis and characterization of ﬂuorescent cobalamin (CobalaFluor) derivatives for imaging. Org. Lett., 3, 799–801. Howard, W.A., Bayomi, A., Natarajan, E., Aziza, M.A., ElAhmady, O., Grissom, C.B., and West, F.G. (1997) Sonolysis promotes indirect Co-C pond cleavage of alkylcob(III)alamin bioconjugates. Bioconjug. Chem., 8, 498–502. Krautler, B., Derer, T., Liu, P.L., Muhlecker, W., Puchberger, M., Kratky, C., and Gruber, K. (1995) Oligomethylene-bridged vitamin B12 dimers. Angew. Chem., 34, 84–86. Hannak, R.B., Farber, G., Konrat, R., and Krautler, B. (1997) An organometallic B12 -rotaxane and a

References

77.

78.

79.

80.

81.

82.

83.

B12 -dimer, relaxed and loaded forms of a molecular spring. J. Am. Chem. Soc., 119, 2313–2314. Thompson, K.H. and Orvig, C. (2006) Vanadium in diabetes: 100 years from Phase 0 to Phase I. J. Inorg. Biochem., 100, 1925–1935. Rehder, D., Pessoa, J.C., Geraldes, C., Castro, M., Kabanos, T., Kiss, T., Meier, B., Micera, G., Pettersson, L., Rangel, M., Salifoglou, A., Turel, I., and Wang, D.R. (2002) In vitro study of the insulin-mimetic behaviour of vanadium (IV, V) coordination compounds. J. Biol. Inorg. Chem., 7, 384–396. Mukherjee, R., Donnay, E.G., Radomski, M.A., Miller, C., Redfern, D.A., Gericke, A., Damron, D.S., and Brasch, N.E. (2008) Vanadium-vitamin B12 bioconjugates as potential therapeutics for treating diabetes. Chem. Commun., 3783–3785. Drummond, J.T. and Matthews, R.G. (1994) Nitrous oxide inactivation of cobalamin-dependent methionine synthase from Escherichia Coli: Characterization of the damage to the enzyme and prosthetic group. Biochemistry, 33, 3742–3750. Kondo, H., Osborne, M.L., Kolhouse, J.F., Binder, M.J., Podell, E.R., Utley, C.S., Abrams, R.S., and Allen, R.H. (1981) Nitrous oxide has multiple deleterious effects on cobalamin metabolism and causes decreases in activities of both mammalian cobalamin-dependent enzymes in rats. J. Clin. Invest., 67, 1270–1283. Brouwer, M., Chamulitrat, W., Ferruzzi, G., Sauls, D.L., and Weinberg, J.B. (1996) Nitric oxide interactions with cobalamins: biochemical and functional consequences. Blood, 88, 1857–1864. Rochelle, L.G., Morana, S.J., Kruszyna, H., Russell, M.A., Wilcox, D.E., and Smith, R.P. (1995) Interactions between hydroxocobalamin and nitric oxide (NO): evidence for a redox reaction between NO and reduced cobalamin and reversible NO binding to oxidized cobalamin. J. Pharmacol. Exp. Ther., 275, 48–52.

84. Greenberg, S.S., Xie, J.M., Zatarain,

85.

86.

87.

88.

89.

90.

91.

J.M., Kapusta, D.R., and Miller, M.J.S. (1995) Hydroxocobalamin (vitamin B12a ) prevents and reverses endotoxin-induced hypotension and mortality in rodents: Role of nitric oxide. J. Pharmacol. Exp. Ther., 273, 257–265. Sagar, S.M., Singh, G., Hodson, D.I., and Whitton, A.C. (1995) Nitric oxide and anticancer therapy. Cancer Treat. Rev., 21, 159–181. Bauer, J.A. (1998) Synthesis, characterization and nitric oxide release proﬁle of nitrosylcobalamin: a potential chemotherapeutic agent. Anti-Cancer Drugs, 9, 239–244. Bauer, J.A., Morrison, B.H., Grane, R.W., Jacobs, B.S., Dabney, S., Gamero, A.M., Carnevale, K.A., Smith, D.J., Drazba, J., Seetharam, B., and Lindner, D.J. (2002) Effects of interferon beta on transcobalamin II-receptor expression and antitumor activity of nitrosylcobalamin. J. Natl. Cancer Inst., 94, 1010–1019. Karadag, A., Onal, I., Senocak, A., Ucar, I., Bulut, A., and Buyukgungor, O. (2008) Syntheses, IR spectra, thermal properties and crystal structures of novel cyano-bridged polymeric complexes of zinc(II) and cadmium(II) with tetracyanoplatinate(II). Polyhedron, 27, 223–231. Zhan, S.Z., Li, W., Wang, J.G., Liang, A.Q., and Deng, Y.F. (2007) Assembly of cyano-bridged Cu(II)/Cu(II) and Cu(I)/Cu(II) compounds obtained by controlled ration of cyanide. J. Organomet. Chem., 692, 3568–3573. Ni, W.W., Ni, Z.H., Cui, A.L., Liang, X., and Kou, H.Z. (2007) Cyanide-bridged Mn(III)-Fe(III) bimetallic complexes based on the pentacyano(1-methylimidazole) ferrate(III) building block: Structure and magnetic characterizations. Inorg. Chem., 46, 22–33. Toma, L., Lescouezec, R., Vaissermann, J., Delgado, F.S., Ruiz-Perez, C., Carrasco, R., Cano, J., Lloret, F., and Julve, M. (2004) Nuclearity controlled cyanide-bridged bimetallic

113

114

4 Vitamin B12 : A Potential Targeting Molecule for Therapeutic Drug Delivery

92.

93.

94.

95.

96.

97.

98.

99.

100.

CrIII-MnII compounds: synthesis, crystal structures, magnetic properties and theoretical calculations. Chem. Eur. J., 10, 6130–6145. Albores, P., Slep, L.D., Weyhermuller, T., and Baraldo, L.M. (2004) Fine tuning of the electronic coupling between metal centers in cyano-bridged mixed-valent trinuclear complexes. Inorg. Chem., 43, 6762–6773. Lee, S.C., Scott, M.J., Kauffmann, K., Munck, E., and Holm, R.H. (1994) Cyanide poisoning: an analog to the binuclear site of oxidized cyanide-inhibited cytochrome C oxidase. J. Am. Chem. Soc., 116, 401–402. Corsi, D.M., Murthy, N.N., Young, V.G., and Karlin, K.D. (1999) Synthesis, structure, and solution NMR studies of cyanide-copper(II) and cyanide-bridged iron(III)-copper(II) complexes. Inorg. Chem., 38, 848–858. Takano, S., Naito, T., and Inabe, T. (1998) Cyano-bridged cobalt-phthalocyanine dimer: a trapped species by crystallization with a pi radical cation. J. Mater. Chem., 8, 511–513. Wolak, M., Stochel, G., and van Eldik, R. (2003) Mechanistic studies on the interaction of reduced cobalamin (Vitamin B12r ) with nitroprusside. J. Am. Chem. Soc., 125, 1334–1351. Stochel, G., Vaneldik, R., Kunkely, H., and Vogler, A. (1989) Kinetics and mechanism of the anation of aquocobalamin (vitamin B12a ) by cyanoferrates. Isolation and identiﬁcation of a cyano-bridged product and mechanistic information from pressure effects. Inorg. Chem., 28, 4314–4322. Kunze, S., Zobi, F., Kurz, P., Spingler, B., and Alberto, R. (2004) Vitamin B12 as a ligand for technetium and rhenium complexes. Angew. Chem. Int. Ed. Engl., 43, 5025–5029. Lippert, B. (ed.) (1999) Cisplatin: Chemistry and Biochemistry of a Leading Anticancer Drug, Wiley-VCH Verlag GmbH, Weinheim. Dubowchik, G.M. and Walker, M.A. (1999) Receptor-mediated and

101.

102.

103.

104.

105.

106.

107.

108.

enzyme-dependent targeting of cytotoxic anticancer drugs. Pharmacol. Ther., 83, 67–123. Wilson, S., Reinhard, K.S., and Gao, X. (2004) in Drug delivery and targeting with vitamin B12 conjugates. US Patent No. 2004047917. Fanchiang, Y.T., Bratt, G.T., and Hogenkamp, H.P.C. (1983) Kinetics and mechanism of the interaction of cis-diaminediaquaplatinum(II) with organocobalamins. J. Chem. Soc. Dalton Trans., 1929–1934. Ribas, J., Serra, M., and Escuer, A. (1984) Solid-state kinetic parameters for the deaquation-anation of the tetracyanonickelate(II), tetracyanopalladate(II), and tetracyanoplatinate(Ii) of aquopentaamminecobalt(III). Inﬂuence of the metal-metal interactions on the platinum complex. Inorg. Chem., 23, 2236–2242. Escuer, A. and Ribas, J. (1985) Dinuclear complexes with a u-cyano ligand. Part X. Synthesis and characterization of several derivatives of cis-diaquaamminecobalt(III) complexes. Inﬂuence of pH. Inorg. Chim. Acta, 104, 143–147. Kunkely, H., Pawlowski, V., and Vogler, A. (1993) Photochemistry of Vitamin B12 Derivatives. New Observations and Conclusions, 14th Conference on Coordination Chemistry, Butterworths, London, pp. 363–366. Mundwiler, S., Spingler, B., Kurz, P., Kunze, S., and Alberto, R. (2005) Cyanide-bridged vitamin B12 -cisplatin conjugates. Chem. Eur. J., 11, 4089–4095. Ruiz-Sanchez, P., Mundwiler, S., Spingler, B., Buan, N.R., Escalante-Semerena, J.C., and Alberto, R. (2008) Syntheses and characterization of vitamin B12 -Pt(II) conjugates and their adenosylation in an enzymatic assay. J. Biol. Inorg. Chem., 13, 335–347. DeConti, R.C., Toftness, B.R., Lange, R.C., and Creasey, W.A. (1973) Clinical and pharmacological studies with cis-diamminedichloroplatinum(II). Cancer Res., 33, 1310–1315.

References 109. Gullo, J.J., Litterst, C.L., Maguire, P.J.,

110.

111.

112.

113.

114.

115.

Sikic, B.I., Hoth, D.F., and Woolley, P.V. (1980) Pharmacokinetics and protein binding of cis-dichlorodiammine platinum (II) administered as a one hour or as a twenty hour infusion. Cancer Chemother. Pharmacol., 5, 21–26. Trynda-Lemiesz, L., Keppler, B.K., and Kozlowski, H. (1999) Studies on the interactions between human serum albumin and imidazolium [trans-tetrachlorobis(imidazol) ruthenate(III)]. J. Inorg. Biochem., 73, 123–128. Ivanov, A.I., Christodoulou, J., Parkinson, J.A., Barnham, K.J., Tucker, A., Woodrow, J., and Sadler, P.J. (1998) Cisplatin binding sites on human albumin. J. Biol. Chem., 273, 14721–14730. Kratz, F. (1993) New tumor-inhibiting ruthenium complexes, in Metal Complexes in Cancer Chemotherapy (ed. B.K. Keppler), Wiley-VCH Verlag GmbH, Weinheim, pp. 391–429. Canal, P. and Chatelut, E. (1996) Pharmacokinetic-pharmacodynamic relationships, in Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy 2 (eds H.M. Pinedo and J.H. Schornagel), Plenum Press, New York, pp. 53–66. Lexa, D. and Saveant, J.M. (1976) Electrochemistry of vitamin B12 . 1. Role of base-on/base-off reaction in oxidoreduction mechanism of B12r -112s system. J. Am. Chem. Soc., 98, 2652–2658. Lexa, D., Saveant, J.M., and Zickler, J. (1977) Electrochemistry of vitamin B12 .

116.

117.

118.

119.

120.

2. Redox and acid-base equilibria in B12a /B12r system. J. Am. Chem. Soc., 99, 2786–2790. Fonseca, M.V., Buan, N.R., Horswill, A.R., Rayment, I., and Escalante-Semerena, J.C. (2002) The ATP:co(I)rrinoid adenosyltransferase (CobA) enzyme of Salmonella enterica requires the 2 -OH Group of ATP for function and yields inorganic triphosphate as its reaction byproduct. J. Biol. Chem., 277, 33127–33131. Bauer, C.B., Fonseca, M.V., Holden, H.M., Thoden, J.B., Thompson, T.B., Escalante-Semerena, J.C., and Rayment, I. (2001) Three-dimensional structure of ATP : corrinoid adenosyltransferase from Salmonella typhimurium in its free state, complexed with MgATP, or complexed with hydroxycobalamin and MgATP. Biochemistry, 40, 361–374. Ruiz-Sanchez, P., Mundwiler, S., Medina-Molner, A., Spingler, B., and Alberto, R. (2007) Iodination of cisplatin adduct of vitamin B12 [{B12 }-CN-{cis-PtCl(NH3 )2 }]+ . J. Organomet. Chem., 692, 1358–1362. Ruiz-Sanchez, P., Mundwiler, S., and Alberto, R. (2007) Syntheses of ﬂuorescent vitamin B12 -Pt(II) conjugates and their Pt(II) release in a spectroelectrochemical assay. Chimia, 61, 190–193. Giannella, R.A., Broitman, S.A., and Zamcheck, N. (1971) Vitamin B12 uptake by intestinal microorganisms: mechanism and relevance to syndromes of intestinal bacterial overgrowth. J. Clin. Invest., 50, 1100–1107.

115

117

5 Strategies for Microsphere-Mediated Cellular Delivery Rosario M. Sanchez-Martin, Lois M. Alexander, Juan Manuel Cardenas-Maestre, and Mark Bradley

5.1 Introduction 5.1.1 Cellular Delivery

The cellular membrane poses a formidable barrier to the intracellular ﬂux of materials circulating extracellularly due to its selective permeability based predominantly on ionic charge, hydrophobicity, and size [1]. As such, many drugs, nucleic acids, proteins, and other investigative constructs are not able to translocate the cellular membrane alone and often require a delivery vehicle for function and/or activity. 5.1.2 Delivery Devices

There are a diverse range of delivery vehicles available, which allow the intracellular introduction of foreign material, although all have associated advantages and disadvantages. Some of these devices are discussed in this section. 5.1.2.1 Liposomes Liposomes are colloidal, vesicular structures (20 nm to 10 µm in diameter) based on phospholipid bilayers. These can range from unilamellar (meaning only one bilayer surrounding an aqueous core) to multilamellar (several bilayers oriented concentrically around an aqueous core), with the structure depending on the processing protocol, concentrations, and choice of bilayer components [2]. The liposome drug-carrier concept was ﬁrst proposed and demonstrated in the early [3, 4], when it was shown that soluble molecules could be entrapped within the core of the liposomes which could be endocytosed. Uptake by cells can be highly efﬁcient (depending on the cell line); however, poor release of the material from the endosomal compartments (especially for high-molecular-weight species) can be a serious limitation. Endosomal escape can be enhanced by Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

118

5 Strategies for Microsphere-Mediated Cellular Delivery

modiﬁcation of the liposomes, by the use of pH-sensitive monomers (e.g., phosphatidylethanolamine) [5, 6], the addition of a lipid helper with ‘‘fusogenic’’ properties (e.g., 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE)) [7], or complexation of the liposome with a membrane disruptive agent [8]. Liposome-based delivery is an extensively applied method, which has been used to deliver many low-molecular-weight drugs, peptides, enzymes [9], and oligonucleotides [10, 11] as well as ionic sensor-ﬁlled microparticles (also known as PEBBLES) [12]. 5.1.2.2 Cell-penetrating Peptides Cell-penetrating peptides (or CPPs), of which HIV TAT is perhaps the most well-known example, have been used to deliver an assorted range of cargos into cells [13]. Often, these peptides are rich in lysine or arginine residues, for example, CPPs based on TAT typically have a cationic cluster of arginine and lysine residues, a notable example being GRKKRRQRRR [14]. Over the past decade, several CPPs that enable the intracellular delivery of active compounds into the cells have been described, typically made up of 10–30 amino acids [15, 16], but they all share some common features, including hydrophilicity, helical moment, and the ability to interact with the cellular membrane. A range of peptidomimetics structurally inspired by the CPPs have been reported, which potentially represent a highly efﬁcient means of translocating a variety of materials across the cell membrane, enhancing the pharmacokinetic properties of any drug, sensor, or protein [17–21]. 5.1.2.3 Dendrimers Dendrimers and hyper-branched polymers are a relatively new class of materials with unique molecular architectures and dimensions in comparison to traditional linear polymers. Highly branched functionalized dendrimers have the potential to act as efﬁcient drug-carrier systems [22] in two different ways: (i) the encapsulation of soluble drugs within the dendritic core (which acts as a reservoir for drug molecules) [23] and (ii) the conjugation of drug molecules to surface functional groups [24, 25]. These functional groups can also be coupled to molecules that address the polymer to particular tissues or cells (targeted delivery) or enhance solubility. Dendrimer research so far has predominantly been focused on the delivery of ‘‘DNA drugs’’ – where its use has enhanced the transfection of DNA into the cellular nucleus [26], but they have also been used to deliver drugs [27]. 5.1.2.4 Nanomaterials Nanomaterials, such as carbon nanotubes, have been exploited as delivery systems [28]. These devices were popularized in 29 and since then much work has been focused on their intracellular use, for example, in the delivery of cell impermeable dyes and short interfering RNA (siRNA) [30–32]. They may be prepared by a number of methods, including chemical vapor deposition and laser ablation, and have additionally been manipulated to form magnetic nanotubes to address site-speciﬁc targeting issues in vivo [33]. However, some controversy surrounds their intracellular uptake with both endocytic-based mechanisms [34] and passive diffusion [35] being considered and additional concerns over their in vivo toxicity.

5.2 Microspheres

Additionally, attention has focused on the development of a diverse range of polymeric particles as delivery systems in order to enhance transport and uptake at the cellular level. One type of polymeric materials that has been shown to be engulfed and actively transported throughout cells are microspheres. The properties, chemistry, and applications of microspheres as delivery method are discussed in this chapter.

5.2 Microspheres

The term microsphere typically refers to uniform spherical polymer particles with diameters ranging from nanometers to several microns composed of various materials such as polystyrene, silica, and methacrylates. A suspension of polymeric microspheres is often described as a ‘‘polymer colloid’’ but microspheres are also known as latex beads. Nowadays, the terms latex and polymer colloid are used interchangeably and have essentially the same meaning. Polymeric microspheres have been used for many years in applications as diverse as in inks, paints, catalysts, and more recently in medical applications [36]. The appeal of microspheres lies largely in the diverse range of applications these particles have. They are not only used as vehicles for intracellular delivery but also for applications ex vitro, for example, in the investigation of DNA hybridization (using DNA coupled to microspheres) [37, 38] or as ﬂuorescent probes in ﬂow cytometry to investigate phagocytosis [39]. Furthermore, they are cheap, easy to produce and functionalize, have a long shelf-life, and they can be produced from a diverse range of materials and in a multitude of sizes. In terms of intracellular investigations, biodegradable microspheres have been studied more extensively than their biostable counterparts [40], largely due to the applications of biodegradable microspheres in drug delivery, where the delivery device can degrade resulting in release of an encapsulated drug [41]. 5.2.1 Biodegradable Microspheres

Microspheres can be tailored to biodegrade by the introduction of monomers that will break down under physiological conditions, with classic examples being poly(lactic acid) and poly(glycolic acid) microspheres. 5.2.1.1 Preparation Microspheres can be prepared via the polymerization of monomers using suspension, dispersion, or emulsion polymerization to achieve particles of the desired size [42–44]. However, due to the availability of linear biodegradable polymers, it is some what more common for biodegradable microspheres to be prepared directly from these and there are several methods by which they may be made. Amongst these are the so-called solvent evaporation method, sometimes called

119

120

5 Strategies for Microsphere-Mediated Cellular Delivery

the double emulsion technique and spray-drying [45–48]. Regardless of the method, it is important to choose the appropriate conditions for preparing biodegradable microspheres since the polymer molecular weight, diameter, and morphology are all important factors for biodegradability. 5.2.2 Biostable Microspheres

The sensitivity of biodegradable polymers to aggressive solvents can limit the chemistry that can be performed on the beads and, in addition, they are not particularly suitable as investigative tools where it would be undesirable for the microsphere to diminish, for example, as biological sensors or reporters. Thus, microspheres prepared with stable monomers and with cross-linking may be more desirable for many applications. 5.2.2.1 Applications of Biostable Microspheres A major established application of biostable microspheres is the investigation of phagocytotic cells by ﬂow cytometry. Biostable microspheres can be easily labeled with a ﬂuorophore, yielding a stable ﬂuorescent construct that can allow analysis of phagocytosis by quantitative ﬂow cytometric methods. As such, microspheres have been used to investigate the phagocytic activity of pulmonary alveolar macrophages (PAMs) [39], pinocytosis and phagocytosis in rat peritoneal macrophages [49], and activation of alveolar macrophages [50]. 5.2.2.2 Preparation Several approaches may be adopted to produce biostable microspheres and a wide range of monomers are available. Typically, dispersion and emulsion polymerization are employed as described previously [51, 52], due to their reasonably simple setup and the controllability of the microsphere size and quality. Dispersion Polymerization Dispersion polymerization gives high conversions and the microspheres are typically narrowly dispersed in size and easily manipulated to give a range of diameters between 0.5 and 10 µm, ideal for biological analysis [53, 54]. However, although the mechanical setup of a dispersion polymerization reaction is easy, the mechanism is complex and not well understood. The required monomers and initiators are dispersed in an organic phase containing a stabilizer to produce a homogenous dispersion [55]. Emulsion Polymerization Emulsion polymerization typically results in the preparation of particles less than 0.5 µm in diameter [53] and is a well-studied and widely applied procedure for the production of a range of polymers, including synthetic rubbers and latex paints [56–58]. The mechanism by which emulsion polymerization results in the production of a polymer precipitate was ﬁrst established in the [59, 60], and has recently been reevaluated by the help of computer simulations [61]. Simplistically, the process of emulsion polymerization can be

5.2 Microspheres

divided into three main stages. First, an emulsion is formed with the monomer in an aqueous solution containing a surfactant, resulting in the production of large micelles into which the monomer can diffuse. Secondly, a water-soluble initiator is added and is able to react with monomer in the micelles. Thus, the monomer polymerizes producing a polymer particle, resulting in a system that contains both monomer droplets and early polymer particles. In the last stage of the emulsion polymerization, monomer from the excess monomer droplets diffuses into the polymer particles, where it may react with initiator resulting in further oligomeric chains. 5.2.3 Microspheres and Solid-phase Chemistry 5.2.3.1 Preparation of Microspheres In our approach, polystyrene microspheres (0.5 and 2 µm) are typically prepared using a dispersion polymerization method or an emulsiﬁer-free emulsion polymerization (0.2 µm), with styrene as the main monomer. To afford stable particles, a cross-linking agent (divinylbenzene, DVB) can be used at 2 mol% with regard to styrene, as at this concentration efﬁcient cross-linking may be achieved without severe disruption to the mechanism by which the polymerizations occur [62]. Finally, vinylbenzylamine [53] can be included to yield aminomethylated microspheres (Scheme 5.1) [63]. An SEM image of microspheres typical of our procedure is shown in Figure 5.1, where it is seen that the microspheres are evenly sized with smooth, spherical morphologies. Depending on the particular use of the microspheres, different chemical approaches were applied to the coupling of cargos to the polymeric particle. In this section we describe some strategies that have been applied. 5.2.3.2 Fmoc Chemistry on Microspheres These polymer particles could be readily employed in multistep solid-phase synthesis and the resulting constructs are rapidly taken up by cells and exploited for a variety of applications. From a synthetic chemistry point of view, the beads

+

(a) or (b)

+

HCl·H2N

NH2·HCl Scheme 5.1 Preparation of aminomethyl microspheres. (a) 0.5- and 2-µm microspheres prepared by dispersion polymerization with 2,2 -Azobis(2-methylpropionitrile) (AIBN) and Polyvinylpyrrolidone (PVP)

40 000, respectively, in EtOH/H2 O (93 : 7) or EtOH at 70 ◦ C for 12 hours and (b) 0.2-µm microspheres prepared by emulsiﬁer-free emulsion polymerization with V-50 and MgSO4 in H2 O at 80 ◦ C for 2 hours.

121

122

5 Strategies for Microsphere-Mediated Cellular Delivery

Acc.V

Spot Magn Det WD

30.0 kV 2.0 9456x SE 9.0

2 µm

Figure 5.1 Polystyrene/Divinylbenzene (PS/DVB) microspheres. SEM image of 500-nm microspheres typical of a dispersion polymerization.

are comparable with the use of protecting group chemistry while also acting as a mode for traversing the membrane of living cells and are able to deliver cargos ranging from sensors to proteins and siRNA. Our strategy consists of an Fmoc solid-phase synthesis protocol (Scheme 5.2), which is based on simple coupling and deprotection steps, where coupling is normally carried out in NMP or DMF (dimethylformamide) with standard coupling agents. Fmoc chemistry allows spacers of a variety of different lengths and hydrophilicities, such as Fmoc-aminohexanoic acid (Fmoc-Ahx-OH) or Fmoc-polyethylene glycolic acid (Fmoc-PEG-OH), to be coupled to the microspheres in order to alter the physical properties of the microspheres (e.g., hydrophilicity) (Figure 5.2), and to distance the cargo from the bead itself [64]. The efﬁciency and reliability of the resulting cellular uptake was found to be dramatically improved by the incorporation of small PEGylated spacers between the cargo and the bead. Using B16F10 cells as a representative example, Table 5.1 shows the results of cellular uptake of 200 nm ﬂuorescein-labeled microspheres using Fmoc-Ahx-OH (Ahx) or Fmoc-PEG-OH (PEG) as spacer. Presumably, this is directly related to the improved hydrophilicity of the overall microsphere resulting in an improved cell–microsphere interaction [65]. 5.2.3.3 Dual Functionality of Microspheres Fmoc-Dde Full Orthogonality Orthogonality between protecting groups is essential for the design of a solid-phase strategy. For the controlled synthesis of our scaffold, both the N-α-amino group and the side chain amino functionality of the lysine derivative should be blocked with orthogonal protecting groups. Where the classic strategy is based on a base-labile N-Fmoc group for protection of the α-amino

5.2 Microspheres

Fmoc HN

OH Fmoc C + HN O

Activation

Deprotection Act C O

Fmoc HN

HN Repeat of deprotection and coupling cycles

Coupling

HN C O

Fmoc HN Final deprotection

HN C O

H2N

Scheme 5.2

Fmoc

H N

Basic steps in Fmoc chemistry.

O OH

Fmoc-Ahx-OH spacer Figure 5.2

Fmoc

H N

H N

O 3

O OH

O Fmoc-PEG-OH spacer

Microsphere spacers. Structure of spacers used.

and an acid-labile side chain protecting group (e.g., Boc), our modiﬁed protocol is based on the use of a nucleophilically labile protecting group on the side chain (1-(4,4-dimethyl-2,6-dioxacyclohexylidene) ethyl, Dde) [66]. Dde has been used previously as a protecting group for primary amines in solid-phase chemistry, but its standard conditions for deprotection (3% v/v hydrazine in DMF) [42] also cleave Fmoc. However, recently complete orthogonality between Fmoc and Dde has been demonstrated using hydroxylamine hydrochloride and imidazole in 1-methyl-2-pyrrolidinone (NMP) to afford the removal of Dde without deprotecting Fmoc [67]. Applying this methodology yields a powerful approach to facilitate dual functionalization of microspheres. Multifunctionalization Strategy Dual functionalized microspheres allow the production of microspheres capable of carrying two cargos simultaneously (e.g., a biologically relevant cargo and a tracking ﬂuorophore) (Scheme 5.3 and Figure 5.3). Subsequently, selective cleavage of Fmoc or Dde permits the directed coupling

123

124

5 Strategies for Microsphere-Mediated Cellular Delivery Results of cellular uptake (% of labeled cells from total population) of 0.2-µm Ahx- and PEG microspheres at varying concentrations on B16F10 cells after 12 hours (at 37 ◦ C and 5% CO2 atmosphere). Table 5.1

% Cellular incorporation Spacer (mg ml –1 )

0.1

0.2

0.3

Ahx PEG

49 95

64 97

90 98

N- a protecting group

H N

O

O

Side-chain protecting group OH

O O HN

O

Figure 5.3

Dual functionality. Structure of Fmoc-Lysine(Dde)-OH.

of the cargo so that its position can be known, and facilitates the introduction of a scaffold that could be easily attached to the microspheres and elongated using Fmoc protocols in a directed manner. 5.2.3.4 Coupling Agents Dicyclohexylcarbodiimide (DCC) and diisopropylcarbodiimide (DIC) are commonly used coupling agents in the preparation of amides, esters, and acid anhydrides from carboxylic acids. Dicyclohexylurea, the by-product formed from DCC, is virtually insoluble in most organic solvents and precipitates from the reaction mixture as the reaction progresses. Hence, although DCC is very useful in solution-phase reactions, it is not appropriate for solid-phase couplings. DIC may be used since the urea by-product is soluble in DMF and can be easily removed from the microspheres via sequential washing and centrifugation. Carbodiimide couplings may be enhanced by the presence of 1-hydroxybenzotriazole (HOBt) as a catalyst, which forms an active ester. In certain applications a water-soluble coupling agent maybe applicable (for example, siRNA couplings – see Section 2.4.2),

5.2 Microspheres Fmoc HN Spacer

Fmoc HN Spacer

Spacer attachment

CH

H2N

Fmoc Dde O NH

H N

N Spacer H

Dde

N 4 H Deprotect Naprotecting group

N H

Dde

N H

O Activating agent 4 Couple O

Spacer attachment

Fmoc HN Spacer

O

H N

H N

N Spacer H

O HN 4 Dde

O

4

H N

H2N Spacer

O

H N

N Spacer H

Deprotect Naprotecting group

O NH

O H2N

H N O

O

Fmoc

125

O

Deprotect side chain protecting group Fmoc HN

CARGO O

H N

Spacer

O HN

H N

N Spacer H 4

Fmoc Couple HN Spacer

O

O

H N O H 2N

CARGO

H N

N Spacer H 4

O

Deprotect Naprotecting group CARGO B HN

Spacer

H N O HN

CARGO B

O H N

N Spacer H 4

HN Couple

O

CARGO

Spacer

O

H N O HN

H N

N Spacer H 4

CARGO

Scheme 5.3 Multifunctionalization. General solid-phase synthesis strategy applied to microspheres.

whereby both the reagent and the urea by-product are water soluble. In these cases, ethyl-(N , N -dimethylamino)propylcarbodiimide hydrochloride (EDC) is frequently employed. Depending on the nature of the functional group available on the cargo, there are two possibilities to facilitate a linkage: amino-functionalized microspheres reacting with a carboxylic group on the cargo or vice versa. These established and well-studied techniques have led to the development of microspheres functionalized in a variety of ways with a diversity of cargos, some of which are discussed in this chapter.

O

126

5 Strategies for Microsphere-Mediated Cellular Delivery

5.2.4 Noncleavable Link to Microspheres

For many applications of microspheres, the robustness of the connection between the cargo and the microsphere is essential for the success of the device, and so a noncleavable amide bond is constructed between the cargo and the polymeric particle. Classic solid-phase protocols have been applied using HOBt/DIC or N-(3-dimethylaminopropyl)-N -ethylcarbodiimide hydrochloride (EDAC) for coupling under organic or aqueous conditions respectively. In the following sections, we describe some of the approaches that have been applied in the attachment of cargos to microspheres. 5.2.4.1 Microsphere-based Intracellular Sensing In the biochemical and medicinal ﬁelds, it is becoming increasingly important to be able to monitor the intracellular levels of certain ions. Fluorescent sensors for a variety of ions, such as Ca2+ and Na+ , are among the most useful tools available for studying ion ﬂuxes within the intracellular environment following the modulation or stimulation of a range of complex biological processes. There are several advantageous reasons for binding sensors covalently to microspheres. First, it facilitates the cellular entrance of these non–membrane permeable molecules (both the acid and salt forms of these types of sensors are unable to cross cellular membranes, an essential requirement in allowing such molecules to be used within the cellular environment). Secondly, covalent linkage to the microsphere eliminates leakage of the dye from cells and ensures a highly localized sensor that allows ‘‘on-bead’’ analysis. Thirdly, due to their micron size the labeled microspheres are easily visualized by microscopy. We have demonstrated that microspheres can be used successfully for calcium sensing, avoiding cellular dilution of the signal [68]. This was achieved by attaching the ﬂuorescent indicator for Ca2+ ions, Indo-1, to the microspheres. In order to achieve this, the original synthesis of Indo-1 was modiﬁed to allow the incorporation of a benzyl ester at the indole 6-position, while maintaining four ethyl ester functionalities on the coordinating carboxy groups [69]. This generated orthogonality between the acid groups that chelate the Ca2+ ions and the additional carboxylic acid group required for resin attachment. The synthesis of the novel derivative was carried out in nine steps (Scheme 5.4), with hydrogenation of the benzyl ester to give the tetraethylester indolecarboxylic acid ready for coupling to the aminomethylated microspheres. Subsequent hydrolysis of the ethyl ester gave the microsphere-immobilized sensor. Other sensors have been used for quantitative and qualitative intracellular pH monitoring [70] with the sensor-loaded microsphere prepared by coupling 5(6)-carboxyﬂuorescein to 2-µm aminomethyl microspheres derivatized with an aminohexanoic acid spacer (Scheme 5.5). Covalent binding of ﬂuorescein to the microspheres dramatically improves the stability of the indicator over time and eliminates leakage from the cell.

5.2 Microspheres

N(CH2COOEt)2 N(CH2COOEt)2 O

O 2

CH2PPh3Br NO2

+

CHO

(a)

127

N(CH2COOEt)2 N(CH2COOEt)2 O O 2

COOBn NO2

COOBn (b) N(CH2CO2K)2 O

N(CH2CO2K)2

N(CH2COOEt)2 N(CH2COOEt)2 O O 2

O 2 (c– e)

NH

NH H N

O

O N H

Scheme 5.4 Calcium sensing. Synthesis and covalent attachment of modiﬁed Indo-1 to microspheres to give the Indo-1-microsphere sensor. Reagents and conditions: (a) K2 CO3 , DMF, 100 ◦ C, 3 hours (46%); (b) (EtO)3 P, 160 ◦ C, N2 , overnight (60%); (c) H2 , 10% Pd/C, ethanol,

COOBn 4 hours (91%); (d) microspheres derivatized with an aminohexanoic acid spacer, HOBt, (benzotriazol-1-yloxy)tripyrrolidinophosphonium hexaﬂuorophosphate (PyBOP), 4-ethylmorpholine, DMF, room temperature, 16 hours (quant.); and (e) 1 M KOH, 6 hours (quant.).

5.2.4.2 siRNA Delivery siRNA (short interfering RNA or small interfering RNA) is a short (typically 18–21 nucleotides long) double-stranded piece of RNA that may be produced naturally and is involved in a pathway known as the RNA interference (RNAi) pathway [71, 72]. Since its discovery in the 73, interest has grown exponentially due to the applications it may have in the biomedical ﬁeld, for example, silencing of genes that are known to result in a diseased state. The RNAi pathway starts with cytoplasmic double-stranded siRNA forming a complex with a multiprotein unit, forming a complex known as the RNA-induced silencing complex, or, more commonly, RISC [74]. This complex then facilitates the degradation of endogenous mRNA with a complementary sequence to the siRNA guide strand [75, 76]. As a consequence, the protein that is coded for by the mRNA will not be produced and there will be a decrease in the concentration of

128

5 Strategies for Microsphere-Mediated Cellular Delivery

O H 2N

(a – b)

H 2N

N H (c)

HO

O

O

COOH H N O Scheme 5.5 Preparation of ﬂuorescein-labeled amino-functionalized polystyrene microspheres. (a) N-Fmoc-aminohexanoic acid (5 equiv.), HOBt (5 equiv.), DIC (5 equiv.), DMF, 12 hours; (b) 20% piperidine in DMF;

O N H and (c) 5(6)-carboxyﬂuorescein (5 equiv.), HOBt (5 equiv.), DIC (5 equiv.), DMF, 12 hours. HOBt, N-hydroxybenzotriazole; DIC, N,N -diisopropylcarbodiimide; DMF, dimethylformamide.

the protein (described as knockdown or gene silencing). It was of interest to investigate if microspheres could act as a delivery vector for nonnatural siRNA. An easily measurable and well-studied knockdown target is enhanced green ﬂuorescent protein (EGFP). Transformed cell lines, for example, human cervical cancer (HeLa) cells, can be easily genetically modiﬁed to encode for the permanent production of green ﬂuorescent protein (GFP), which renders the cells ﬂuorescent. Silencing of the gene that codes for the production of EGFP can be easily quantiﬁed by ﬂow cytometry and visualized by microscopy, making it an easily measurable target. In order to study microsphere-mediated EGFP silencing, siRNA with a sequence tailored to target EGFP expressed in HeLa cells was coupled covalently to 0.5-µm polystyrene microspheres with an independent ‘‘tracking’’ ﬂuorophore (Cy5) (Scheme 5.6) [77]. This allowed the microsphere location to be determined and allowed cells containing the delivery vehicle (and therefore the siRNA) to be distinguished from non-‘‘beadfected’’ cells, giving an accurate analysis of microsphere-mediated gene silencing. Thus, HeLa-EGFP cells were treated with Cy5-co-siRNA 0.5-µm microspheres and EGFP silencing was analyzed after 24, 48, and 72 hours by ﬂow cytometry (Figure 5.4) selecting those cells positive for Cy5 ﬂuorescence and, hence, beadfected cells containing siRNA (Figure 5.5). Gene silencing was observed after 24 hours, where EGFP was reduced by 40% and was even more prevalent at 48 and 72 hours where reductions of up to approximately 90% were noted. No signiﬁcant toxicity was noted in treated cells (cellular viability was found to be over 90% after 72 hours by methylthiozodiphenyl-tetrazolium bromide (MTT) assay), demonstrating that siRNA microspheres have no detrimental effect on cellular metabolism.

5.2 Microspheres

129

Modified PEG H2N

O

3

FmocHN–PEG

O

H N

(a – c)

N H

O

DdeHN

NH H N 4 O

O

O

H N 3

N H

O

(d – f) O HN PEG NH O H H N Cy5 N 4 O

OH

O

O

H N 3

(g) N H

O

HN PEG NH O H H N Cy5 N 4 O

GFP fluorescence intensity (a.u.)

O

O

O

H N 3

O

N H

piperidine/DMF, 25 ◦ C; (g) succinic anhydride, N,N-diisopropylethylamine (DiPEA), 18 hours, 25 ◦ C; (h) N-(3-dimethylaminopropyl)-N -ethylcarbodiimide hydrochloride (EDAC), 2-(N-morpholino)ethanesulfonic acid (MES) (pH 5.5), 2 hours, 25 ◦ C; and (i) amino-siRNA, sterile RNase-free water, 24 hours, 25 ◦ C.

14 000

Negative GFP Positive GFP

12 000

siRNA Microspheres

~40%

8 000 6 000 4 000

0

3

6

16 000

2 000

O

H N

NH

Scheme 5.6 Dual functionalization 0.5 µm Cy5-co-siRNA microspheres. (a) Fmoc-lysine-(Dde)-OH, HOBt, DIC, DMF, 18 hours, 25 ◦ C; (b) 20% piperidine/DMF, 25 ◦ C; (c) Fmoc-PEG-OH, HOBt, DIC, DMF, 18 hours, 25 ◦ C; (d) hydroxylamine hydrochloride, imidazole, (NMP), 1 hour, 25 ◦ C; (e) carboxylated-Cy5, HOBt, DIC, DMF, 18 hours, 25 ◦ C; (f) 20%

10 000

O

N

(h – i) si RNA O

H2N–PEG NH H O H N N 4 N 5 O

>90%

24 h 48 h 72 h

Figure 5.4 Microsphere-mediated gene silencing. siRNA covalently linked (via the antiguide sense strand) to 0.5-µm Cy5 microspheres silenced EGFP expressed in HeLa cells over 24–72 hours.

N H

130

5 Strategies for Microsphere-Mediated Cellular Delivery HeLa GFP knockdown uber 190308-: 100 75 Count

Only Cy5 positive cells are considered for EGFP expression

50 25

105 0

Cy 5 +ve 104

(a)

102

Q1 Q2

103

105

Serum free - couplings in water-24 h:

Q3 Q4

Cy 5 −ve 102

103 104 FITC-A

70 60 104 105 GFP +ve

50 Count

102 103 GFP −ve

40 30 20 10 0

(b)

Figure 5.5 Microsphere-mediated cell selection. On the basis of the independent Cy5 tracking label expressed on microspheres, cells containing the delivery device (and therefore siRNA) can be selected against cells not containing microspheres and

102

103 104 FITC-A

105

analyzed independently for their EGFP expression. (a) Positive GFP untreated control cells and (b) cells treated with siRNA microspheres after 72 hours. ‘‘FITC-A’’ refers to EGFP.

5.2.5 Cleavable Linkers

In some circumstances, it may be desirable for a cargo to be disengaged from the delivery device, for example, in the use of constructs that require nuclear localization [21]. Cleavability may be achieved in a number of manners, for example, a light source may be applied to facilitate a change that results in the release of a cargo (photosensitive linkers) [78]. Alternatively, naturally occurring enzymes or cellular factors could be harnessed to cleave a chemical bond, for example, an ester or a disulﬁde. This could be a more attractive avenue as it rarely results in undesired cytotoxicity. 5.2.5.1 Ester Bonds One of these delivery approaches is the cleavage of an ester bond between the molecule and the delivery system by an intracellular esterase [79]. We decided to use this method of cleavage to investigate if microspheres could transport molecules linked via an ester bond, which following cleavage in the intracellular space would become active. This would prove that microspheres could be used to deliver

5.2 Microspheres

131

molecules into cells, but additionally the enzymatic reactions on microspheres could take place intracellularly. For this assay, resoruﬁn (7-hydroxy-3H-phenoxazin-3-one, λex = 570 nm, λem = 590 nm) was chosen as a ﬂuorogenic reporter due to its nonﬂuorescent nature as an ester and its ﬂuorescent characteristics when seen as the free phenolate [80]. When covalently conjugated through its oxy anion, resoruﬁn shows a drop in its absorbance extinction coefﬁcient relative to the anion and a 40-fold difference in the ﬂuorescence quantum yield is observed [81]. However, free ﬂuorescent resoruﬁn is cell impermeable and thus requires a carrier system to transport it intracellularly. Cells incubated with resoruﬁn–microspheres should thus emit red ﬂuorescence only if the resoruﬁn ester is hydrolyzed from the beads in the intracellular space. This assay tests both the property of the microspheres to cross the cell membrane and the enzymatic cleavage within the cell. The inability of deprotonated resoruﬁn to cross the cell membrane of its own accord provides a suitable control. Microspheres supporting resoruﬁn were therefore synthesized and the intracellular release of the dye was studied [40]. Microspheres (0.5 and 2 µm) were dual labeled with ﬂuorescein and resoruﬁn via an amide and an ester linkage, respectively, as shown in Scheme 5.7. The intracellular cleavage of the ester bond will release resoruﬁn that will become ﬂuorescent. In addition the noncleavable linkage of ﬂuorescein to the microspheres (via an amide linkage) allows the trafﬁcking of the loaded beads to be continuously monitored within the cells. After incubation with B16F10 cells for 6 hours, the resoruﬁn ﬂuorescence increased 11-fold compared to free resoruﬁn as a control (Figure 5.6) with the cells becoming ﬂuorescent after treatment with Amide bond (non-cleavable linker)

Ester bond (cleavable linker)

OH

O

O

HOOC O

PEG

O

O

O

O

H N

N H

O

N Nonfluorescent

O

O

HOOC

O

O +

N

O

O

Intracellular cleavage

PEG

HO O

N H

Fluorescent Scheme 5.7 Resoruﬁn. Intracellular cleavage of ester bound resoruﬁn leading to release of ﬂuorogenic resoruﬁn from ﬂuorescein-labeled PEG microspheres (0.5 and 2.0 µm).

H N O

OH

5 Strategies for Microsphere-Mediated Cellular Delivery

Fluorescence intensity

132

25000 20000 15000 10000 5000 0

6h

12 h Time of incubation

Untreated cells

Free resorufin

0.5 um microspheres

2.0 um microspheres

Figure 5.6 Resoruﬁn microspheres. Intracellular resoruﬁn release from 0.5- and 2-µm microspheres in B16F10 cells after 6 and 12 hours.

resoruﬁn–microspheres. This establishes these materials as effective delivery devices of exogenous cellular cargo via esterase cleavage. 5.2.5.2 Disulﬁde Bonds Glutathione is a tripeptide that is naturally present in cells as a protective agent. It reduces disulﬁde bonds by acting as an electron donor and is itself oxidized to glutathione disulﬁde. To demonstrate that this disulﬁde bridge could be used to facilitate the intracellular cleavage of a cargo from microspheres, 0.2- and 0.5-µm microspheres were functionalized with dithiodipropionic acid, followed by ﬂuoresceinamine to obtain a ﬂuorescent construct, whereby the ﬂuorophore was coupled to the microspheres via a disulﬁde bond (Scheme 5.8). (a – b)

H2N

H2N

O

H N

O

3

N H

O

(c)

O

(d – e)

S

HO

H N

S O

O

O

O

H N 3

O

N H

O O HO

HOOC

N H

S

H N

S

O

O

Scheme 5.8 Cleavable microspheres. Coupling of ﬂuorescein to 0.2- and 0.5-µm microspheres via a cleavable disulﬁde linkage. (a) Fmoc-PEG-OH, HOBt, DIC, DMF, 18 hours, 25 ◦ C; (b) 20% piperidine/DMF, 25

O

H N 3 ◦

O

N H

C; (c) Dithiodipropionic acid, HOBt, DIC, DMF, 18 hours, 25 ◦ C; (d) HOBt, DIC, DMF, 2 hours, 25 ◦ C; and (e) ﬂuoresceinamine, DMF, 18 hours, 25 ◦ C.

5.2 Microspheres

100000 Fluorescence intensity (a.u.)

90000 80000 70000 60000 50000 40000 30000 20000 10000 0 0h

1h

6h

24 h

Time (h) 0.2 µm DTT

0.2 µm glutathione

0.5 µm DTT

0.5 µm glutathione

Figure 5.7 Disulﬁde cleavage. Dissociation of ﬂuorescein from 0.2- and 0.5-µm disulﬁde microspheres by treatment with 10 mM DTT or 10 mM glutathione in phosphate buffered saline (PBS), after 1, 6, and 24 hours.

Initially, the ex vivo cleavability of the disulﬁde bond was investigated, with microspheres treated with dithiothreitol (DTT) and glutathione (Figure 5.7). After 1 hour treatment with DTT, the ﬂuorescence associated with the FAM beads decreased by 55% on 0.5-µm microspheres and by 75% on 0.2-µm microspheres. Use of glutathione facilitated a decrease in ﬂuorescence intensity of 50% on 0.5-µm microspheres and by 60% on 0.2-µm microspheres after 1 hour. After 24 hours, the ﬂuorescence intensity had decreased to a base value (90% reduction in ﬂuorescence). In order to study the cleavage of the disulﬁde linkage in vitro, HeLa cells were beadfected with 0.2- and 0.5-µm disulﬁde-FAM microspheres and intracellular cleavage was assessed. The dye was initially associated with microspheres but was noted to be dispersed within the cytoplasmic region of the cell over time, indicating the successful dissociation of ﬂuorescein from the microspheres indicative of disulﬁde cleavage. 5.2.6 Bioconjugation

Although chemical approaches, linking cargo covalently via a synthetic linkage, can be desirable, an increase in diversity may be achieved by facilitating the coupling of material by alternative approaches. An example of this is bioconjugation, where a cargo is linked to the microspheres via interaction between biologically compatible components (e.g., a receptor and a ligand).

133

134

5 Strategies for Microsphere-Mediated Cellular Delivery

5.2.6.1 Streptavidin–Biotin The high afﬁnity of the biotin–streptavidin interaction has allowed many applications to be developed, which make use of this general bridge system in various branches of biotechnology [82–85]. There are some remarkable and notable properties of this system. The small size of biotin makes it a perfect tag for biologically active macromolecules because biotinylation does not usually alter biological activity, while the biotin–streptavidin interaction is not disrupted by serum proteins. The exceptionally tight interaction between biotin and streptavidin [86] suggested that this might be a practical way to add a range of molecules to microspheres and subsequently aid their delivery into cells. On the basis of these properties of the biotin–streptavidin interaction, a strategy for microsphere-mediated delivery has been developed, where the microspheres are loaded with streptavidin following an efﬁcient procedure (Scheme 5.9) and different biotinylated cargoes are attached to them [87]. The ﬁrst stage was the preparation of streptavidin-loaded microspheres. For this purpose, the streptavidin was coupled to 0.5- and 2-µm aminomethyl microspheres by overnight reaction in the presence of glutaraldehyde [88]. Subsequently, a solution of streptavidin in PBS buffer (pH 7.4) was added and allowed to react overnight. Finally, the Schiff base was reduced with sodium cyanoborohydride to give the covalently loaded streptavidin microspheres (Scheme 5.9). In order to be able to evaluate the cellular uptake, the microspheres were labeled with biotin-4-ﬂuorescein, which has been reported to be a superior ﬂuorescent biotin derivative for accurate measurement of streptavidin in crude bioﬂuids [89]. This ﬂuorescent labeling allowed us to follow the microspheres inside the cells and to study their cellular uptake (found to be up to 85% for 500-nm microspheres after 24 hours). Additionally, the fact that we were able to label the microspheres with this biotinylated compound also demonstrated the integrity of the streptavidin when bound to the microspheres. It also allowed determination of the loading of streptavidin on the microspheres and the coupling efﬁciency was found to be >90%. A further study was carried out applying this biotin–streptavidin approach, in which the cellular uptake of oligonucleotides loaded onto the microspheres was analyzed. A biotinylated ﬂuorescently labeled dsDNA (biotin-dsDNA-Cy3) was H2N

(a)

OHC

N

(b)

Streptavidin

N

N (c)

Streptavidin

Scheme 5.9 Bioconjugation. Preparation of streptavidin-loaded microspheres. (a) glutaraldehyde, PBS, 15 hours; (b) streptavidin, PBS, 15 hours; and (c) NaCNBH3 , PBS, ethanol, 2 hours.

N H

N H

References

successfully attached to the streptavidin microspheres and was taken up by the cells with high efﬁciency. The application of the biotin–streptavidin technology to our microspheres opens up a large number of applications such as the binding and delivery of targeting ligands (for example, monoclonal antibodies) in an efﬁcient manner.

5.3 Future Perspectives

The successful and efﬁcient delivery of materials into cells is of fundamental importance throughout many areas of biology. In this chapter we have presented different strategies that can be applied for the attachment of biomolecules to the microspheres and their applications as a delivery system. Mono-dispersed populations of robust cross-linked microspheres of deﬁned sizes (from 200 nm to 2 µm) have been synthesized and standard solid-phase multistep protocols have been applied to them. Thanks to their stability in organic solvents provided by the cross-linking structure; numerous chemical reactions have been carried out easily on these materials. Their robust nature has allowed the successful development of several approaches for the attachment of biologically active cargo to the microspheres, such as siRNAs or biosensors. These microspheres provide a reliable way to perform cellular delivery and long-term cell monitoring and encourage the possibility of the use of this approach for alternative applications, such as monitoring of enzymatic activity or for the delivery of therapeutic macromolecules (protein, plasmid, etc.) or small molecules such as drugs. We believe that the microspheres, with their ﬂexible and diversiﬁed chemistry together with their biological properties (lack of toxicity, efﬁcient cellular uptake, and the opportunity to sort at a single cell level), mean that they are a highly competitive cellular delivery technology, rivaling current nanoparticle techniques, as they have controlled cellular loading while exhibiting no toxicity.

Acknowledgments

The authors would like to thank the BBSRC and EPSRC for funding and RSM would like to thank the Royal Society for a Dorothy Hodgkin Fellowship. The authors are grateful to all the Bradley’s group for help and support and in particular to M. Lopalco for the Cy5 dye.

References 1. Adrian, C. and Hogben, M.

(1960) Movement of material across cell membranes. Physiologist, 3, 56–62.

2. Gregoriadis, G. (1995) Engineering lipo-

somes for drug delivery: progress and problems. Trends Biotechnol., 13 (12), 527–537.

135

136

5 Strategies for Microsphere-Mediated Cellular Delivery 3. Gregoriadis, G. (1976) Carrier potential

4.

5.

6.

7.

8.

9.

10.

11.

12.

of liposomes in biology and medicine. 1. New England. J. Med., 295 (13), 704–710. Poste, G. and Papahadjopoulos, D. (1976) Drug-containing lipid vesicles render drug-resistant tumor-cells sensitive to Actinomycin-D. Nature, 261 (5562), 699–701. Torchilin, V.P.Z.F. and Huang, L.J. (1993) pH-sensitive liposomes. J. Liposome Res., 3, 201–255. Simoes, S., Moreira, J.N., Fonseca, C., Duzgunes, N., and de Lima, M.C.P. (2004) On the formulation of pH-sensitive long circulation times. Adv. Drug Deliv. Rev., 56 (7), 947–965. Sakurai, F., Nishioka, T., Yamashita, F., Takakura, Y., and Hashida, M. (2001) Effects of erythrocytes and serum proteins on lung accumulation of lipoplexes containing cholesterol or DOPE as a helper lipid in the single-pass rat lung perfusion system. Eur. J. Pharm. Biopharm., 52 (2), 165–172. Yessine, M.A. and Leroux, J.C. (2004) Membrane-destabilizing polyanions: interaction with lipid bilayers and endosomal escape of biomacromolecules. Adv. Drug Deliv. Rev., 56 (7), 999–1021. Chang, J.S., Choi, M.J., Cheong, H.S., and Kim, K. (2001) Development of Th1-mediated CD8+effector T cells by vaccination with epitope peptides encapsulated in pH-sensitive liposomes. Vaccine, 19 (27), 3608–3614. Meyer, O., Kirpotin, D., Hong, K.L., Sternberg, B., Park, J.W., Woodle, M.C., and Papahadjopoulos, D. (1998) Cationic liposomes coated with polyethylene glycol as carriers for oligonucleotides. J. Biol. Chem., 273 (25), 15621–15627. Zelphati, O. and Szoka, F.C. (1996) Liposomes as a carrier for intracellular delivery of antisense oligonucleotides: a real or magic bullet? J. Control. Release, 41 (1-2), 99–119. Clark, H.A., Hoyer, M., Philbert, M.A., and Kopelman, R. (1999) Optical nanosensors for chemical analysis inside single living cells. 1. Fabrication,

13.

14.

15.

16.

17.

18.

19.

20.

21.

characterization, and methods for intracellular delivery of PEBBLE sensors. Anal. Chem., 71 (21), 4831–4836. Zhao, M. and Weissleder, R. (2004) Intracellular cargo delivery using tat peptide and derivatives. Med. Res. Rev., 24 (1), 1–12. Ruben, S., Perkins, A., Purcell, R., Joung, K., Sia, R., Burghoff, R., Haseltine, W.A., and Rosen, C.A. (1989) Structural and functional-characterization of human immunodeﬁciency virus tat protein. J. Virol., 63 (1), 1–8. Jarver, P. and Langel, U. (2004) The use of cell-penetrating peptides as a tool for gene regulation. Drug Discov. Today, 9 (9). doi: PII S1359-6446(04)03042-9. Temsamani, J. and Vidal, P. (2004) The use of cell-penetrating peptides for drug delivery. Drug Discov. Today, 9 (23). doi: PII S1359-6446(04)03279-9. Wender, P.A., Mitchell, D.J., Pattabiraman, K., Pelkey, E.T., Steinman, L., and Rothbard, J.B. (2000) The design, synthesis, and evaluation of molecules that enable or enhance cellular uptake: peptoid molecular transporters. Proc. Natl. Acad. Sci. U.S.A., 97 (24), 13003–13008. Wender, P.A., Rothbard, J.B., Jessop, T.C., Kreider, E.L., and Wylie, B.L. (2002) Oligocarbamate molecular transporters: design, synthesis, and biological evaluation of a new class of transporters for drug delivery. J. Am. Chem. Soc., 124 (45), 13382–13383. Peretto, I., Sanchez-Martin, R.M., Wang, X.H., Ellard, J., Mittoo, S., and Bradley, M. (2003) Cell penetrable peptoid carrier vehicles: synthesis and evaluation. Chem. Commun., 18, 2312–2313. Fara, M.A., Diaz-Mochon, J.J., and Bradley, M. (2006) Microwave-assisted coupling with DIC/HOBt for the synthesis of difﬁcult peptoids and ﬂuorescently labelled peptides – a gentle heat goes a long way. Tetrahedron Lett., 47 (6), 1011–1014. Unciti-Broceta, A., Diezmann, F., Ou-Yang, C.Y., Fara, M.A., and Bradley, M. (2009) Synthesis, penetrability and intracellular targeting

References

22.

23.

24.

25.

26.

27.

28.

29.

of ﬂuorescein-tagged peptoids and peptide-peptoid hybrids. Bioorg. Med. Chem., 17 (3), 959–966. Aulenta, F., Hayes, W., and Rannard, S. (2003) Dendrimers: a new class of nanoscopic containers and delivery devices. Eur. Polym. J., 39 (9), 1741–1771. Capitosti, G.J., Guerrero, C.D., Binkley, D.E., Rajesh, C.S., and Modarelli, D.A. (2003) Efﬁcient synthesis of porphyrin-containing, benzoquinone-terminated, rigid polyphenylene dendrimers. J. Org. Chem., 68 (2), 247–261. Milhem, O.M., Myles, C., McKeown, N.B., Attwood, D., and D’Emanuele, A. (2000) Polyamidoamine Starburst (R) dendrimers as solubility enhancers. Int. J. Pharm., 197 (1-2), 239–241. How, S.E., Yingyongnarongkul, B., Fara, M.A., Diaz-Mochon, J.J., Mittoo, S., and Bradley, M. (2004) Polyplexes and lipoplexes for mammalian gene delivery: from traditional to microarray screening. Comb. Chem. High Throughput Screen., 7 (5), 423–430. (a) How, S.E., Unciti-Broceta, A., Sanchez-Martin, R.M., and Bradley, M. (2008) Solid-phase synthesis of a lysine-capped bis-dendron with remarkable DNA delivery abilities. Org. Biomol. Chem., 6 (13), 2266–2269; (b) Diaz-Mochon, J.J., Fara, M.A., Sanchez-Martin, R.M., and Bradley, M. (2008) Peptoid dendrimers-microwave-assisted solid-phase synthesis and transfection agent evaluation. Tetrahedron Lett., 49 (5), 923–926. Gillies, E.R. and Frechet, J.M.J. (2005) Dendrimers and dendritic polymers in drug delivery. Drug Discov. Today, 10 (1). doi: PII S1359-6446(04)03276-3. Sahoo, S.K. and Labhasetwar, V. (2003) Nanotech approaches to delivery and imaging drug. Drug Discov. Today, 8 (24). doi: PII S1359-6446(03)02903-9. Kam, N.W.S., Jessop, T.C., Wender, P.A., and Dai, H.J. (2004) Nanotube molecular transporters: internalization of carbon nanotube-protein conjugates into mammalian cells. J. Am. Chem. Soc., 126 (22), 6850–6851.

30. Kam, N.W.S. and Dai, H.J. (2005) Car-

31.

32.

33.

34.

35.

36.

37.

38.

bon nanotubes as intracellular protein transporters: generality and biological functionality. J. Am. Chem. Soc., 127 (16), 6021–6026. Pantarotto, D., Briand, J.P., Prato, M., and Bianco, A. (2004) Translocation of bioactive peptides across cell membranes by carbon nanotubes. Chem. Commun., 1, 16–17. Kam, N.W.S., Liu, Z., and Dai, H.J. (2005) Functionalization of carbon nanotubes via cleavable disulﬁde bonds for efﬁcient intracellular delivery of siRNA and potent gene silencing. J. Am. Chem. Soc., 127 (36), 12492–12493. Son, S.J., Reichel, J., He, B., Schuchman, M., and Lee, S.B. (2005) Magnetic nanotubes for magnetic-ﬁeld-assisted bioseparation, biointeraction, and drug delivery. J. Am. Chem. Soc., 127 (20), 7316–7317. Cherukuri, P., Bachilo, S.M., Litovsky, S.H., and Weisman, R.B. (2004) Near-infrared ﬂuorescence microscopy of single-walled carbon nanotubes in phagocytic cells. J. Am. Chem. Soc., 126 (48), 15638–15639. Bianco, A., Hoebeke, J., Godefroy, S., Chaloin, O., Pantarotto, D., Briand, J.P., Muller, S., Prato, M., and Partidos, C.D. (2005) Cationic carbon nanotubes bind to CpG oligodeoxynucleotides and enhance their immunostimulatory properties. J. Am. Chem. Soc., 127 (1), 58–59. Kawaguchi, H. (2000) Functional polymer microspheres. Prog. Polym. Sci., 25 (8), 1171–1210. Zhang, Y., Eniola, A.O., Graves, D.J., and Hammer, D.A. (2003) Speciﬁc adhesion of micron-sized colloids to surfaces mediated by hybridizing DNA chains. Langmuir, 19 (17), 6905–6911. Rogers, P.H., Michel, E., Bauer, C.A., Vanderet, S., Hansen, D., Roberts, B.K., Calvez, A., Crews, J.B., Lau, K.O., Wood, A., Pine, D.J., and Schwartz, P.V. (2005) Selective, controllable, and reversible aggregation of polystyrene latex microspheres via DNA hybridization. Langmuir, 21 (12), 5562–5569.

137

138

5 Strategies for Microsphere-Mediated Cellular Delivery 39. Steinkamp, J.A., Wilson, J.S., Saunders,

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

G.C., and Stewart, C.C. (1982) Phagocytosis – ﬂow cytometric quantitation with ﬂuorescent microspheres. Science, 215 (4528), 64–66. Freiberg, S. and Zhu, X. (2004) Polymer microspheres for controlled drug release. Int. J. Pharm., 282 (1-2), 1–18. Christie, R.J. and Grainger, D.W. (2003) Design strategies to improve soluble macromolecular delivery constructs. Adv. Drug Deliv. Rev., 55 (3), 421–437. Piirma, I. (1985) Colloids, in Encyclopedia of Polymer Science and Engineering, 2nd edn (eds H.F. Mark, N.M. Bikales, C.G. Overberger, G. Menges, and J.I. Kroschwitz), John Wiley & Sons, Inc., New York, pp. 125–130. (a) Barrett, K.E.J. (1975) Dispersion Polymerisation in Organic Media, John Wiley & Sons, Ltd, London; (b) Strover, H.D.H. and Li, K. (1996) 1900–1905: dispersion polymerisation, Polymer Materials Encyclopedia, CRC Press, Boca Raton. Weissman, J.M., Sunkara, H.B., Tse, A.S., and Asher, S.A. (1996) Thermally switchable periodicities and diffraction from mesoscopically ordered materials. Science, 274 (5289), 959–960. Ogawa, Y., Yamamoto, M., Okada, H., Yashiki, T., and Shimamoto, T. (1988) A new technique to efﬁciently entrap leuprolide acetate into microcapsules of polylactic acid or copoly(Lactic Glycolic) acid. Chem. Pharm. Bull., 36 (3), 1095–1103. Vasir, J.K., Tambwekar, K., and Garg, S. (2003) Bioadhesive microspheres as a controlled drug delivery system. Int. J. Pharm., 255 (1-2), 13–32. Edlund, U. and Albertsson, A.C. (2002) Degradable polymer microspheres for controlled drug delivery. Degradable Aliphatic Polyesters, 157, 67–112. Bodmeier, R. and Chen, H.G. (1988) Preparation of biodegradable poly (+/−) lactide microparticles using a spray-drying technique. J. Pharm. Pharmacol., 40 (11), 754–757. Pratten, M.K. and Lloyd, J.B. (1986) Pinocytosis and phagocytosis – the effect of size of a particulate substrate on its mode of capture by rat

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

peritoneal-macrophages cultured in vitro. Biochim. Biophys. Acta, 881 (3), 307–313. Ng, K., Stringer, K.A., Cohen, Z., Serravo, R., Tian, B., Meyer, J.D., Falk, R., Randolph, T., Manning, M.C., and Thompson, D.C. (1998) Alveolar macrophage cell line is not activated by exposure to polymeric microspheres. Int. J. Pharm., 170 (1), 41–49. (a) Tseng, C.M., Lu, Y.Y., Elaasser, M.S., and Vanderhoff, J.W. (1986) Uniform polymer particles by dispersion polymerization in alcohol. J. Polym. Sci. A, Polym. Chem., 24 (11), 2995–3007; (b) Paine, A.J. and McNulty, J. (1990) Uniform polymer particles by dispersion polymerization in alcohol [Comment]. J. Polym. Sci. A, Polym. Chem., 28 (9), 2569–2574. Delair, T., Marguet, V., Pichot, C., and Mandrand, B. (1994) Synthesis and characterization of cationic amino functionalized polystyrene latexes. Colloid Polym. Sci., 272 (8), 962–970. Ober, C.K. and Lok, K.P. (1987) Formation of large monodisperse copolymer particles by dispersion polymerization. Macromolecules, 20 (2), 268–273. Ye, Q., Zhang, Z.C., Jia, H.T., He, W.D., and Ge, X.W. (2002) Formation of monodisperse polyacrylamide particles by radiation-induced dispersion polymerization: particle size and size distribution. J. Colloid Interface Sci., 253 (2), 279–284. Paine, A.J. (1990) Dispersion polymerization of styrene in polar-solvents .1. Grafting mechanism of stabilization by hydroxypropyl cellulose. J. Colloid Interface Sci., 138 (1), 157–169. Whitby, G.S. and Katz, M. (1933) Synthetic rubber (concluded). Ind. Eng. Chem., 25, 1338–1348. Farbenfabrik, A.G. (1909) First Process for Manufacturing artiﬁcial rubber Germany. Luther, M. and Heuck C. (1927) Polymerizing Butadiene Hydrocarbons, Germany. Smith, W.V. and Ewart, R.H. (1948) Kinetics of emulsion polymerization. J. Chem. Phys., 16 (6), 592–599.

References 60. Harkins, W.D. (1947) A general theory

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

of the mechanism of emulsion polymerization. J. Am. Chem. Soc., 69 (6), 1428–1444. Gilbert, R.G. (1996) Emulsion Polymerisation: A Mechanistic Approach, Academic Press, London. Zhang, H.T., Huang, J.X., and Jiang, B.B. (2002) Kinetics of polymerization and particle stabilization mechanism on dispersion copolymerization of styrene and divinylbenzene. J. Appl. Polymer Sci., 85 (10), 2230–2238. Sanchez-Martin, R.M., Muzerelle, M., Chitkul, N., How, S.E., Mittoo, S., and Bradley, M. (2005) Bead-based cellular analysis, sorting and multiplexing. Chembiochem, 6 (8), 1341–1345. Page, P., Bradley, M., Walters, I., and Teague, S. (1999) Solid-phase synthesis of tyrosine peptide aldehydes. Analogues of (S)-MAPI. J. Org. Chem., 64 (3), 794–799. Harris, J.M. (1992) Poly(ethylene glycol) chemistry: biotechnical and biomedical applications, in Topics in Applied Chemistry (ed. J.M. Harris), Plenum Press, New York. Bycroft, B.W., Chan, W.C., Chhabra, S.R., and Hone, N.D. (1993) A novel lysine-protecting procedure for continuous-ﬂow solid-phase synthesis of branched peptides. J. Chem. Soc. Chem. Commun., (9), 778–779. Diaz-Mochon, J.J., Bialy, L., and Bradley, M. (2004) Full orthogonality between Dde and Fmoc: the direct synthesis of PNA-peptide conjugates. Org. Lett., 6 (7), 1127–1129. Sanchez-Martin, R.M., Cuttle, M., Mittoo, S., and Bradley, M. (2006) Microsphere-based real-time calcium sensing. Angew. Chem. Int. Ed., 45 (33), 5472–5474. Grynkiewicz, G., Poenie, M., and Tsien, R.Y. (1985) A new generation of Ca-2+ indicators with greatly improved ﬂuorescence properties. J. Biol. Chem., 260 (6), 3440–3450. Bradley, M., Alexander, L., Duncan, K., Chennaoui, M., Jones, A.C., and Sanchez-Martin, R.M. (2008) pH sensing

71.

72.

73.

74.

75.

76.

77.

78.

79.

in living cells using ﬂuorescent microspheres. Bioorg. Med. Chem. Lett., 18 (1), 313–317. Fire, A., Xu, S.Q., Montgomery, M.K., Kostas, S.A., Driver, S.E., and Mello, C.C. (1988) Potent and speciﬁc genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature, 391 (6669), 806–811. Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K. and Tuschl, T. (2001) Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature, 411 (6836), 494–498. Hamilton, A.J. and Baulcombe, D.C. (1999) A species of small antisense RNA in posttranscriptional gene silencing in plants. Science, 286 (5441), 950–952. Hong, J., Wei, N., Chalk, A., Wang, J., Song, Y.T., Yi, F., Qiao, R.P., Sonnharnmer, E.L.L., Wahlestedt, C., Liang, Z.C., and Du, Q. (2008) Focusing on RISC assembly in mammalian cells. Biochem. Biophys. Res. Commun., 368 (3), 703–708. Schwarz, D.S., Hutvagner, G., Du, T., Xu, Z.S., Aronin, N., and Zamore, P.D. (2003) Asymmetry in the assembly of the RNAi enzyme complex. Cell, 115 (2), 199–208. Schubert, S., Grunweller, A., Erdmann, V.A., and Kurreck, J. (2005) Local RNA target structure inﬂuences siRNA efﬁcacy: systematic analysis of intentionally designed binding regions. J. Mol. Biol., 348 (4), 883–893. Wang, K.P., Wang, C.H., Xiao, F., Wang, H.X., and Wu, Z.G. (2008) JAK2/STAT2/STAT3 are required for myogenic differentiation. J. Biol. Chem., 283 (49), 34029–34036. Berg, K., Folini, M., Prasmickaite, L., Selbo, P.K., Bonsted, A., Engesaeter, B.O., Zaffaroni, N., Weyergang, A., Dietze, A., Maelandsmo, G.M., Wagner, E., Norum, O.J., and Hogset, A. (2007) Photochemical internalization: a new tool for drug delivery. Curr. Pharm. Biotechnol., 8 (6), 362–372. Simpson, D.J., Unkefer, C.J., Whaley, T.W., and Marrone, B.L. (1991) A mechanism-based ﬂuorogenic probe for the cytochrome-P-450 cholesterol

139

140

5 Strategies for Microsphere-Mediated Cellular Delivery

80.

81.

82.

83.

84.

side-chain cleavage enzyme. J. Org. Chem., 56 (18), 5391–5396. Gao, W.Z., Xing, B.G., Tsien, R.Y., and Rao, J.H. (2003) Novel ﬂuorogenic substrates for imaging 6-lactamase gene expression. J. Am. Chem. Soc., 125 (37), 11146–11147. Ishiyama, M., Furusawa, H., Shiga, M., Ohseto, F., and Sasamoto, K. (1999) A resoruﬁn derivative as a ﬂuorogenic indicator for cell viability. Anal. Sci., 15 (10), 1025–1028. Pardridge, W.M. (2007) shRNA and siRNA delivery to the brain. Adv. Drug Deliv. Rev., 59 (2-3), 141–152. Susumu, K., Uyeda, H.T., Medintz, I.L., Pons, T., Delehanty, J.B., and Mattoussi, H. (2007) Enhancing the stability and biological functionalities of quantum dots via compact multifunctional ligands. J. Am. Chem. Soc., 129, 13987–13996. Niemeyer, C.M., Adler, M., and Wacker, R. (2007) Detecting antigens by quantitative immuno-PCR. Nat. Protoc., 2 (8), 1918–1930.

85. Diamandis, E.P. and Christopoulos,

86. 87.

88.

89.

T.K. (1991) The Biotin (Strept)Avidin System – principles and applications in biotechnology. Clin. Chem., 37 (5), 625–636. Green, N.M. (1990) Avidin and Streptavidin. Methods. Enzymol., 184, 51–67. Bradley, M., Alexander, L., and Sanchez-Martin, R.M. (2008) Cellular uptake of ﬂuorescent labelled biotin-streptavidin microspheres. J. Fluoresc., 18 (3-4), 733–739. Wu, C.W., Lee, J.G., and Lee, W.C. (1998) Protein and enzyme immobilization on non-porous microspheres of polystyrene. Biotechnol. Appl. Biochem., 27, 225–230. Kada, G., Falk, H., and Gruber, H.J. (1999) Accurate measurement of avidin and streptavidin in crude bioﬂuids with a new, optimized biotin-ﬂuorescein conjugate. Biochimi. Biophys. Acta, 1427 (1), 33–43.

141

Part III Research in Therapeutics

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

143

6 Fundamental Processes in Radiation Damage to DNA: How Low-Energy Electrons Damage Biomolecules Ilko Bald

6.1 Radiation Damage and the Role of Low-Energy Electrons

The severe effects of high-energy radiation such as X rays (0.1–100 keV) and γ rays (>100 keV) are well known and range from a simple sunburn to alterations of the genome, which might lead to cancer or long-term genetic diseases. Experimental evidence accumulates indicating that the damage caused by high-energy radiation can to a large extent be ascribed to a series of low-energy events [1–3]. In a ﬁrst approximation, this can be rationalized when considering the typical dimensions of DNA that is assumed to be the most sensitive part of a cell, since an alteration of the genome might lead to the loss of genetic information. DNA is packed into the cell nucleus and surrounded by proteins to form a highly viscous polymer gel with high density [4]. The DNA double helix is wrapped around an octamer of histone proteins making up a nucleosome containing 146 bp [5]. The nucleosomes are further packed into a 30-nm ﬁber constituting the chromatin [6], which condenses into the chromosome during cell division. The mass of chromatine comprises twice as much protein than DNA, and in total the DNA covers only about 1% of the volume of a cell. Hence, the probability that high-energy radiation directly deposits energy into the DNA molecule is very low, and most of the damage is rather due to secondary particles generated along the radiation track, which are able to travel close enough to the DNA to induce damage. On a molecular level, radiation damage comprises a series of complex physical, chemical, and biological events that are only partially understood. Some of the processes eventually leading to DNA damage are depicted in Figure 6.1 ordered by the timescale of their appearance. After deposition of the initial radiation quantum, molecules are ionized and excited within femtoseconds or shorter thereby producing secondary electrons and ions. The thus generated electrons might possess sufﬁcient energy to induce further ionization processes resulting in a distribution of the high energy of the primary quantum to a large number of secondary particles. Electrons are the most abundant secondary particles and are highly reactive before they are thermalized and solvated within 10−12 seconds. Molecules such as DNA can effectively be damaged by dissociative electron attachment (DEA) as Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

144

6 Role of Low-Energy Electrons in DNA Radiation Damage

hw

Chromosome

Primary radiation 100 eV to MeV

DNA

T C C A C A G G T A A T A T

Physical stage +

H2O+ → H+ + ·OH −

e + DNA → DNA-# → Fragments 10−12 s

Ionisation, excitation

e−aq

Final biological effects: mutations, cell death

A G C

Chemical stage Stand breaks Base loss Cross links

−

H2O + hw → H2O + e

10−15 s

T A T

10−9 s

10−3 s Diffusion

Figure 6.1 Outline of relevant processes in DNA radiation damage according to their timescale, starting from the initial events of ionization and excitation up to the biological effects of DNA damage. Shown is the type of damage at different time domains,

Biological stage Enzymatic response: repair/ misrepair

1s

weeks Enzymatic reactions

the respective effects on the DNA, and the involved pivotal physical, chemical, and biological processes. Dissociative electron attachment is a crucial process since it takes place very early in timescale.

is shown later in this chapter. Besides ions and electrons another highly reactive secondary particle is the hydroxyl (OH) radical that is produced by ionization and subsequent dissociation of water. In diffusion controlled reactions within 10−9 and 10−3 seconds the OH radical is able to induce severe chemical modiﬁcations in DNA such as strand breaks and base loss. The effects of the chemical stage are traditionally referred to as indirect damage; however, also damage caused by secondary electrons and ions taking place in the physical stage can be understood as indirect damage. Single strand breaks (SSBs) can be repaired by enzymes using the complementary strand as a template, but double strand breaks (DSBs) are barely repairable and usually result in cell death. Enzymatic reactions are triggered after some seconds and represent the biological stage of radiation damage. A misrepair or a lack of damage repair results in mutations or cell death. Apparently the ﬁnal biological effects are determined by the initial ionization processes taking place very early after impact of the high-energy quantum and the majority of the DNA lesions is due to the indirect damage caused by secondary particles. According to very recent Monte–Carlo simulations [7], the most probable energy of secondary electrons that are produced in liquid water by high-energy proton or electron radiation (1–100 MeV) is about 9 eV. When all further ionization events that are caused by energetic secondary electrons are taken into account, almost one third of the electrons possess energy below 1 eV [7]. The importance of secondary

6.1 Radiation Damage and the Role of Low-Energy Electrons

electrons is obvious when considering that more than 80% of the primary energy of 1 MeV proton radiation is carried by low-energy electrons after the primary ionizations [7]. In the meantime it is well known that these low-energy electrons efﬁciently induce DSBs in plasmid DNA [2, 8, 9] at energies above 5 eV and SSBs even down to 0 eV with basically no threshold energy [10, 11]. A striking feature observed in the yield curves of SSB and DSB caused in plasmid DNA by low-energy electrons is a resonant behavior, that is, the presence of distinct maxima at energies around 10 (DSB and SSB) [2, 9], 2, and 0.8 eV (only SSB) [10]. A DNA strand break requires the cleavage of one or more chemical bonds within the DNA backbone, presumably a C–O or P–O bond between the sugar- and phosphate moieties. The characteristic resonant features of the DNA strand break yields indicate that bond cleavage is due to DEA, and a similar behavior was observed in the analysis of degradation products using high performance liquid chromatography (HPLC) [12] and very recently also by single photon ionization of neutral products [8] desorbed from the surface during electron irradiation of DNA. A direct observation of DNA damage caused by low-energy electrons by means of scanning tunneling microscopy (STM) was reported recently [13]. The observed structural changes of the physisorbed DNA upon irradiation with 8 eV electrons (extension of DNA islands) were ascribed to strand breaks or an unraveling of the double helix. A direct comparison of the damage induced by X-ray photons and secondary electrons reveals that the yield of strand breaks caused by secondary electrons is higher than that is caused directly by X rays [14]. 6.1.1 How Chemical Bonds are Broken by Low-energy Electrons

The dissociation energy of a typical chemical bond in an organic molecule is about 4–5 eV and superﬁcially it is remarkable that electrons with considerably lower energy are able to induce dissociation of these bonds. In such a case the dissociation is driven by the electron afﬁnity (EA) of the formed fragment anion, which might compensate the bond dissociation energy of a broken chemical bond. The initial step in DEA is the formation of a temporary negative ion [15–18] (TNI) by electron capture thereby occupying a formerly empty molecular orbital (MO) – usually an antibonding MO, which then leads to dissociation along a certain bond: e− + ABC −→ ABC#− −→ AB + C

(6.1)

The TNI (ABC#− ) may dissociate along different bonds or even involve complex rearrangements in polyatomic molecules: e− + ABC −→ ABC#− −→ A− + BC

(6.2)

e− + ABC −→ ABC#− −→ AC− + B

(6.3)

Depending on the speciﬁc energy of the incoming electron, different parent anionic states can be generated, which are commonly referred to as resonances. A TNI

145

146

6 Role of Low-Energy Electrons in DNA Radiation Damage

ABC−

E

ABC

Γ

E

VAE = D(AB − C) − EA (AB) + E* AB + C EA (AB)

VAE

E* AB− + C D(AB − C)

0

0 EA (ABC)

QC Q −eq

Q(AB − C)

Figure 6.2 Schematic two-dimensional potential energy diagram illustrating the vertical transition from the neutral ground state of the molecule ABC to the anionic state ABC#− requiring the vertical attachment energy (VAE). As the anionic potential energy surface is repulsive in the Franck–Condon region, the generated molecular anion is referred to as transient negative ion (TNI) and dissociates into a stable

Ion yield

anionic and one or more neutral fragments. The energy dependence of the fragment ion formation reﬂects the initial Franck–Condon transition. The maximum of the ion yield corresponds to a Franck–Condon transition at longer bond distances where the dissociation probability is high due to the longer lifetime of the TNI with respect to AD and the shorter time it takes for dissociation resulting in a redshift of the DEA signal.

is formed in the geometry of the neutral molecule by a vertical Franck–Condon transition as is indicated in a simpliﬁed two-dimensional potential energy diagram in Figure 6.2. In a polyatomic molecule this represents a cut through the multidimensional potential energy surface. On the right-hand side of Figure 6.2, a typical DEA spectrum is displayed, which is obtained by scanning the energy of an electron beam within typically 0–15 eV while the yield of an anionic fragment is recorded by mass spectrometry as a function of the electron energy [17, 19]. The TNI in Eqs. (6.1–6.3) is labeled by ‘‘#’’ to indicate that it is intrinsically unstable toward loss of the extra electron (autodetachment (AD)). The AD lifetime τAD of the TNI is determined by the energy width of the resonance due to the Heisenberg uncertainty relation (τAD ≥ h/2π, with h being Planck’s constant (h = 6.6 · ×10−16 eV s) and the energy width of the resonance), and the probability for dissociation is determined by the ratio of the AD lifetime and the time it takes for dissociation: P = exp(−τDEA /τAD ). If the dissociation takes place along one single bond, the dissociation time can be as short as one vibrational period (≈10−14 seconds). In summary, the energetic position of a DEA signal in the ion yield curve is determined by the vertical attachment energy (VAE), convoluted with the (energy-dependent) probability for dissociation into a particular fragment ion.

6.1 Radiation Damage and the Role of Low-Energy Electrons

The intensity of the signal is governed by the AD rate and the Franck–Condon factor of the initial transition; that is, the square of the overlap integral of the neutral ground state and the anionic wave function. The intensity is expressed as (energy-dependent) cross section and is for biomolecules usually in the order of the geometrical cross section of the molecule (10−21 –10−22 m2 ; see [20] for a critical discussion of the absolute DEA cross section of the nucleobase thymine), but can be orders of magnitudes larger in halogen containing molecules. Dissociative electron attachment processes can be investigated experimentally by crossing an electron beam, having deﬁned but variable energy, with a molecular beam. In the experiments described here the electron beam is generated by a trochoidal electron monochromator having an energy resolution of about 100 meV. Negative ions are detected by a quadrupole mass spectrometer as a function of the primary electron energy resulting in ion yield curves shown in the next section and schematically on the right-hand side of Figure 6.2. 6.1.2 DEA Studies of Gas-Phase DNA Building Blocks: The Nucleobases

Electron-induced reactions in biomolecules are discussed in detail in recent review articles [3, 17, 21, 22]. DEA to the DNA and RNA nucleobases thymine [20, 23–28], cytosine [23, 24], adenine [29, 30], guanine [29], and uracil [31, 32] has been studied extensively in the gas phase. All nucleobases (NBs) capture electrons at various energies below 10 eV and dissociate mainly by abstraction of a neutral hydrogen atom to form the closed shell anion [NB–H]− ,1) but also a decomposition of the pyrimidine or purine ring system, respectively, is observed. The abstraction of neutral H proceeds with remarkable site and energy selectivity [28, 33]; it was demonstrated that in thymine and uracil the hydrogen abstraction occurs within a sharp signal peaking at 1.0 eV exclusively by cleavage of the N1–H bond (which is connected to the sugar in nucleosides), whereas at an electron energy of 1.8 eV predominantly the N3–H bond dissociates. In the purine nucleobase adenine it was shown that at electron energies below 1.5 eV only the N9–H bond is broken [30]. These studies lead to the assumption that in DNA the NBs act as antennas for low-energy electrons, in that an excess electron gets initially localized on the nucleobase moiety and the extra charge then induces a bond cleavage within the DNA backbone resulting in a strand break. Different scenarios of DEA to single nucleotides have been calculated by different research groups using density functional theory (DFT) [34–40]. In general, it is conﬁrmed that electrons of energy below 2 eV can attach to the NBs, which might be followed by sugar–phosphate bond cleavage due to coupling of the π ∗ orbital localized at the nucleobase with a σ ∗ orbital at the sugar–phosphate moiety. On the other hand, electron attachment to the phosphate group was predicted to proceed only at higher energies around 2 eV. An interesting alternative mechanism for low-energy electron–induced strand 1) The notation [M–H]− is commonly used

in mass spectrometry and the dash reads

‘‘minus’’and does not refer to a chemical bond.

147

148

6 Role of Low-Energy Electrons in DNA Radiation Damage

breaks was suggested by Dabkowska et al. [41], who considered initial formation of a stable hydrogenated closed shell anion of cytidine followed by intramolecular proton transfer from the sugar moiety to the nucleobase eventually leading to C–O bond cleavage in the backbone. Owing to the low barriers of the reactions and the fact that only bound anions are involved, the rates of strand breakage are much higher than that in other proposed mechanisms. Experimental data on direct low-energy electron attachment to the DNA backbone is scarce, which is also due to the limited accessibility of suitable gas-phase model compounds for the DNA backbone. Nevertheless, in a recent study of low-energy electron–induced DNA strand breaks, it was demonstrated that the yield of neutral fragmentation products associated with a decomposition of the sugar moiety in the DNA backbone can be correlated very well with the resonant features in the strand break yield curves, whereas fragmentation products associated with the decomposition of the NBs give rise to different resonances [8]. High resolution electron energy loss (HREELS) and XPS (X-ray photoelectron spectroscopy) studies of self-assembled monolayers of DNA also suggest that electrons mainly interact with the DNA backbone [42]. In the following section, the experimental challenges that are encountered when investigating DNA backbone subunits are described, along with results demonstrating their ability to dissociatively capture low-energy electrons.

6.2 DEA Studies on Model Compounds for the DNA Backbone

The DNA backbone is composed of alternating 2-deoxy-d-ribose and phosphate moieties (Figure 6.3), but model compounds for the sugar unit and the phosphate group are not accessible in a straightforward way. In this section, DEA studies on the sugar molecule d-ribose are presented followed by an investigation of acetylated d-ribose, which is regarded as a considerably improved surrogate for the sugar unit in DNA or RNA due to its more similar chemical structure. Finally, an experimental approach that allows the study of DEA to more complex biomolecules in the gas phase is presented, and the discussion of DEA to the DNA backbone is completed by the presentation of results on d-ribose-5 -phosphate. The chemical structures of the investigated molecules are shown in Figure 6.3. 6.2.1 Electron Attachment to D-Ribose

The sugar moiety in DNA occupies a central position since it connects two phosphate groups of the backbone with a nucleobase, and a damage of the sugar would lead to either base release or a strand break. And indeed, the isolated (gas phase) RNA sugar d-ribose captures electrons at very low energies followed by numerous fragmentation reactions [43]. Examples of typical ion yields obtained from d-ribose are shown in Figures 6.4–6.6. All fragment ions are observed

6.2 DEA Studies on Model Compounds for the DNA Backbone

O O P O O−

CH3 O C O

H O

NB

4

H

O

5 H O

CH3 C O O

O HO

OH

O

P

O

HO O O P O O− DNA backbone

H

H

3

2 OH 1

OH D-Ribose

R C5H10O5 150 amu

OH

O O O C C O CH3 CH3

H

Tetraacetyl-D-ribose TAR (CH3COO)4C5H6O 350 amu

149

OH H

H

H

OH

H OH

D-Ribose-5′-phosphate

RP H2PO3C5H9O5 230 amu

Figure 6.3 Part of the DNA backbone compared to the molecular structures of D-ribose in the pyranose form, tetraacetyl-D-ribose in the furanose form, which is more comparable to the sugar unit in DNA, and D-ribose-5 -phosphoric acid representing a model compound for the entire DNA backbone.

within a sharp signal close to 0 eV, and some fragment ions also appear from a resonant feature located at 6–8 eV (as shown in Figure 6.6). A typical fragmentation reaction of monosaccharide anions is the loss of neutral water molecules as was demonstrated with the pentoses 2-deoxy-d-ribose [44] and d-ribose [43], and the hexose d-fructose [45]: e− + M −→ M−# −→ [M–nH2 O]− + nH2 O

(6.4)

with n = 1–2 for d-ribose [43] and 2-deoxy-d-ribose [44], and 1–3 for d-fructose [45]. The thermodynamic threshold of a DEA reaction is given by the sum of the bond dissociation energies D that need to be broken subtracted by the sum of bond dissociation energies of newly formed bonds and the EA of the formed anionic fragment. For the abstraction of one H2 O molecule this is R H = D(C–OH) + D(C–H) − D(H–OH) − D(C=C) − EA(C5 H8 O4 ) (6.5) The EA of C5 H8 O4 is not known, but must be positive, and the sum of bond dissociation energies is zero when using typical values for the involved bonds [47] (D(C–OH) = 4.0 eV, D(C–H) = 4.2 eV, D(H–OH) = 5.2 eV, D(C = C) = 3 eV). Thus, the reaction is in total exothermic, but it is required that the activation barriers are very low so that the bond breaking and bond formation must proceed simultaneously. This reaction remarkably demonstrates how a complex reaction involving multiple bond breaking is solely triggered by the excess charge as it occurs close to 0 eV electron energy. The loss of different numbers of water molecules from the TNI most likely leaves the sugar ring intact. Nevertheless, several reactions that are associated with a fragmentation of the carbon–oxygen ring have been observed, that is, a cross-ring cleavage. The most dominant signals that arise from cross-ring cleavage are shown in Figures 6.4–6.6. The unambiguous assignment of the fragment masses to

150

6 Role of Low-Energy Electrons in DNA Radiation Damage

5-13C-D-Ribose

D-Ribose

OH

C

OH

OH OH

Ion count rate (s−1)

101 amu: C4H5O3−

H C O OH H OH

1.0

1.5

SF6− OH

OH OH

OH OH

102 amu: 13CC H O − 3 5 3

CH2O, H2O

C4H6O3− (102 amu)

103 amu: 13CC H O − 3 6 3

102 amu: C4H6O3− 2.0

C

OH OH

102 amu: C4H6O3− 0.5

O 13OH

101 amu: C4H5O3−

e− (0 eV) +

0.0

O OH

13

O OH

1-13C-D-Ribose

0.0

0.5

1.0

1.5

2.0

0.0

0.5

1.0

1.5

2.0

Electron energy (eV) Figure 6.4 Ion yield curves for fragment anions at 101 and 102 amu, which are assigned to C4 H5 O3 − and C4 H6 O3 − , respectively. The DEA spectra of D-ribose labeled at C5 with 13 C (middle) are identical to the spectra of nonlabeled D-ribose (left) indicating that the loss of the carbon containing

neutral fragment occurs exclusively from C5 (as indicated in the reaction scheme). Conversely, the signals obtained from D-ribose labeled at C1 with 13 C (right) are shifted to 102 and 103 amu, respectively, conﬁrming that the negative charge remains on the C1 site.The DEA spectra are adapted from [43].

speciﬁc formulas was only possible by the use of different isotope-labeled d-ribose molecules [43] containing either 13 C or deuterium (1-13 C-d-ribose, 5-13 C-d-ribose, and 1-D-d-ribose). Furthermore, detailed fragmentation studies of deprotonated d-ribose and d-fructose using matrix-assisted laser desorption and ionization (MALDI) [48] revealed a systematic fragmentation pattern, in that the negatively charged sugars loose multiples of neutral H2 O molecules and CH2 O units. Three examples of cross-ring cleavage reactions are described in detail in the next section. 6.2.2 Cross-Ring Cleavage of D-Ribose Proceeds with Selective Charge Retention

The dominant fragment anions associated with cross-ring cleavage of d-ribose subsequent to electron attachment were observed at 101, 102, 71, and 72 amu. The assignment of the product ions to a speciﬁc stoichiometry was supported by measurements with isotope-labeled molecules resulting in the following fragmentation scheme:

6.2 DEA Studies on Model Compounds for the DNA Backbone

5-13C-D-Ribose 13

400

OH 200

C

O

151

1-13C-D-Ribose

OH

600

O OH

OH OH

300

OH C

13

OH OH 71 amu

71 amu C3H3O2−

Ion count rate (s−1)

0 600

400

300

200 72 amu C3H4O2−

72 amu − 2H3O2

13CC

0 600

400

200

73 amu

300 13

73 amu CC2H4O2−

0 0.0

0.5

1.0

1.5

2.0

0.0

0.5

1.0

1.5

Electron energy (eV) Figure 6.5 DEA spectra obtained at 71, 72, and 73 amu from 5-13 C-D-ribose (left) and 1-13 C-D-ribose (right), which are assigned to a cross-ring cleavage resulting in C3 H3 O2 − and C3 H4 O2 − , respectively. Two carbon atoms are excised as neutral fragments

exclusively from C5 and C4, since the spectra of 5-13 C-D-ribose are identical to nonlabeled D-ribose (not shown here) and in 1-13 C-D-ribose the signals are shifted to 72 and 73 amu, respectively.The DEA spectra are adapted from [43].

e− (ε ≈ 0 eV) −→ R−# −→ C4 H6 O− 3 (102 amu) + CH2 O + H2 O

(6.6)

−→ C4 H5 O− 3 (101 amu) + CH2 O + H2 O + H

(6.7)

−→ C3 H4 O− 2 (72 amu) + C2 H4 O2 + H2 O

(6.8)

−→ C3 H3 O− 2 (71 amu) + C2 H4 O2 + H2 O + H

(6.9)

Furthermore, the use of isotope-labeled molecules revealed another fascinating aspect of the DEA reactions to d-ribose and other monosaccharides, that is, the selective excision of neutral fragments containing C5 (see atom labeling in Figure 6.3). In Figure 6.4, the ion yield curves of the fragment anions at 102 and 101 amu from nonlabeled d-ribose are illustrated on the left-hand side. The spectra

2.0

6 Role of Low-Energy Electrons in DNA Radiation Damage

1-13C-D-Ribose

D-Ribose – nonlabeled

59 amu O OH OH

Ion yield (a.u.)

152

O OH 13 C

OH OH

OH OH OH

59 amu C2H3O2−

0

1

2

3

4

5

6

7

8

9 10 11

60 amu 0

1

2

3

13

C

OH

5

6

7

8

O OH

O OH OH

OH OH

59 amu

OH OH

59 amu

D

60 amu 0

1

2

3

4

5

9 10 11

1-D-D-Ribose

5- C -D-Ribose 13

4

6

7

8

60 amu

0 1 2 3 9 10 11 Electron energy (eV)

Figure 6.6 Energy dependence of fragment ions at 59 and 60 amu obtained from DEA to D-ribose and its isotope-labeled analogs 5-13 C-D-ribose, 1-13 C-D-ribose, and 1-D-D-ribose. These anions are formed from two different resonances located close to 0 eV and at

4

5

6

7

8

9 10 11

6–8 eV. The selectivity of charge retention is less pronounced than for the heavier fragment anions; however, a clear preference for the C2 H3 O2 − anion to be formed from C1 is apparent.The DEA spectra are adapted from [46].

in the middle are taken from 5-13 C-d-ribose and show the same curves at the same masses as the nonlabeled d-ribose demonstrating that the neutral fragment in this reaction is exclusively generated from the C5 position. In contrast to this, both spectra are completely shifted by one mass unit to higher masses in 1-13 C-d-ribose (right-hand side) indicating that in this case the fragment anions contain 13 C, that is, the anomeric carbon atom. The situation is similar for the fragment anions at 72 and 71 amu, which corresponds to the loss of a neutral fragment containing two carbon atoms. In Figure 6.5 the spectra recorded at 71, 72, and 73 amu from 5-13 C-d-ribose (left) and 1-13 C-d-ribose (right) are compared. Apparently the signals obtained from 1-13 C-d-ribose are almost completely shifted to 72 and 73 amu, whereas the spectra of 5-13 C-d-ribose exhibit signals at 71 and 72 amu corresponding to the spectra of nonlabeled d-ribose (not shown here). Hence, these reactions proceed with selective charge retention on C1, similar to the prompt dissociation of deprotonated d-ribose and d-fructose recently observed in MALDI mass spectra [48]. Selective bond breaking involving the excision of C5 from d-ribose was also recently demonstrated in experiments using low-energy ion beams [49]. It should be noted that in DNA

6.2 DEA Studies on Model Compounds for the DNA Backbone

the C5 atom of the sugar moiety is connected to the adjacent phosphate group and an excision of this atom directly results in a strand break. In addition to the cross-ring cleavage associated with loss of one or two carbon atoms also the abstraction of three neutral CH2 O units (or one C3 H6 O3 fragment) was observed resulting in the fragment anion at 59 amu [46]: e− (ε ≈ 0 eV) −→ R−# −→ C2 H3 O− 2 (59 amu) + C3 H6 O3 + H

(6.10)

The DEA spectra are shown in Figure 6.6 and the C2 H3 O2 − fragment ion is different compared to the ones discussed above in the sense that it is also generated in considerable amounts from a resonance located at higher energies between 6 and 8 eV. Similar to the selectivity of Reactions 6.6–6.9, the C2 H3 O2 − ion is mainly formed from the C1 site of the d-ribose molecule and the neutral fragment most likely consists of C5, C4, and C3 as can be seen on the very weak signal at 60 amu in 5-13 C-d-ribose. However, the selectivity is not complete and even slightly less pronounced on the higher energy resonance indicating that two different reaction pathways exist and a small fraction of the C2 H3 O2 − ions contains C5. The origin of the two resonances near 0 eV and at 6–8 eV is explained in the next subsection. The electron-induced selective bond breaking is a fascinating aspect of the DEA reactions, since selective bond breaking and bond formation has always been a major vision and challenge in chemistry and was pursued with enormous experimental effort especially in laser chemistry [50, 51]. Moreover, low-energy electron beams can be used as a comparatively convenient tool [52] to generate speciﬁc reaction products by site- and energy-selective bond breaking [53] at speciﬁc electron energies via the formation of deﬁned negative ion states. According to the preceding discussion, a large number of fragment anions arise when a transient negative ion of d-ribose is formed near 0 eV. The DEA experiments reveal the mass-to-charge ratio of the anions and the energy at which they are formed. By means of isotope-labeled molecules, the stoichiometry of ionic and neutral fragments could be derived and the neutrals were found to be exclusively produced from the C5 site of the d-ribose molecule. The chemical structure of the fragments can be derived from the fact that the reactions leading to these fragments have to be exothermic as they occur close to 0 eV. DFT calculations at the B3LYP/6-31++G∗∗ level of theory using Gaussian 03 revealed the thermodynamic thresholds of Reactions 6.7 and 6.8 for different structures of the anionic and neutral fragments. In Figure 6.7 schematic energy diagrams are depicted starting from the lowest energy conformation of d-ribose, which is stabilized by three intramolecular hydrogen bonds between OH2, OH3, and OH4 [54]. According to the DFT calculations only one possible structure of the C4 H5 O3 − ion (at 101 amu) out of nine different linear and cyclic structures (not shown here) [55] is stable enough to render the reaction exothermic. The anion contains a carboxyl group and is stabilized by an intramolecular hydrogen bond as shown in Figure 6.7. The EA of the corresponding neutral molecule is with 3.9 eV remarkably high, and all other possible geometries for C4 H5 O3 − (not shown here) are at least 1.9 eV higher in energy [55]. Figure 6.7 shows two alternative compositions for the

153

154

6 Role of Low-Energy Electrons in DNA Radiation Damage D-Ribose:

72 amu: C3H4O2−

C5H10O5 101 amu: C4H5O3− CH2O H2O H

+ H2O CH(OH)CH(OH)

+0.003 eV

0

+ −0.098 eV −0.506 eV

+

+ CH3OH OH

H 2O

CH3COOH

−1.545 eV

Figure 6.7 Thermodynamic thresholds for the formation of C4 H5 O3 − (101 amu) and C3 H4 O2 − (72 amu), respectively, as calculated at the B3LYP/6-31++G** level of theory. For both fragment anions two neutral decay channels are indicated. The C4 H5 O3 − ion can either be associated with formation of neutral CH2 O + H2 O + H resulting in a threshold of −98 meV, or with CH3 OH

+ OH resulting in a threshold of −0.5 eV. For the C3 H4 O2 − (72 amu) anion thermodynamic thresholds of +3 meV for the neutral channel CH(OH)CH(OH) + H2 O or, alternatively, −1.5 eV for the the formation of neutral CH3 COOH + H2 O were calculated. For the anionic fragments, the most stable isomers or conformers out of many are displayed.

neutral fragments (CH2 O + H2 O + H, and CH3 OH + OH), which both result in exothermic reactions. However, the formation of water and formaldehyde might require less activation energy than the formation of methanol and the hydroxyl radical due to considerable rearrangements in the latter case. The most stable structure for C3 H4 O2 − (72 amu) found by DFT calculations is the propenoic acid anion as shown in Figure 6.7. Two alternative neutral channels accompanying the propenoic acid anion generation are shown on the left-hand side of Figure 6.7. The reaction is exothermic by 1.5 eV if acetic acid and water are formed, and slightly endothermic by 3 meV for the case that Z-dihydroxyethene and water are produced in addition to the propenoic acid anion. It is interesting to note that the spectra shown in Figure 6.5 exhibit rather broad signals, and the slightly better electron energy resolution employed in the measurements of 5-13 C-d-ribose even reveals two distinct features contributing to the formation of C3 H4 O2 − . Thus, two different reaction pathways having different thermodynamic thresholds might be operative from the same resonance, and the slightly endothermic reaction might appear only at the high-energy side of the resonance as a shoulder. 6.2.3 The Nature of the Transient Negative D-Ribose Anions

As is described in Section 6.1.1 electron attachment at energies below 4 eV proceeds without changing the electronic conﬁguration of the neutral molecule,

6.2 DEA Studies on Model Compounds for the DNA Backbone

what is referred to as a shape resonance [15, 17, 56] due to the explicit shape of the interaction potential between the incoming electron and the neutral molecule that is responsible for the trapping of the excess electron. As the d-ribose molecule is a saturated molecule the excess electron must occupy a σ ∗ orbital. However, a σ ∗ shape resonance is usually located at appreciably higher energies than 0 eV due to the high energy of the σ ∗ orbital in the neutral molecule that is not expected to be reduced in the negative sugar ion such that it appears close to 0 eV.2) For instance, in a recent systematic DEA study of different aliphatic and cyclic alcohols ∗ shape resonance is located around 2 eV. [57] it was demonstrated that a σO−H ∗ Furthermore, a σ shape resonance is expected to be comparatively localized on certain bonds that are cleaved to yield, for instance, [M–H]− in the case of alcohols ∗ arising from an O–H bond dissociation due to occupation of the σO−H antibonding MO. However, the manifold of different fragment anions observed in d-ribose arising from the 0 eV resonance through multiple bond breaking indicates that the initial negative ion state is rather delocalized. Such a ‘‘diffuse’’ negative ion state can be formed due to the large dipole moment of the sugar molecule. The extra electron is attracted by the positive site of the dipole moment and gets trapped into a very diffuse rydberg-like orbital to form a dipole-bound state (DBS) [58, 59]. It was shown previously by ab initio calculations [60] that d-fructose indeed can form a DBS, which may serve as a doorway to dissociation when the DBS couples to a valence state. The formation of DBS by electron attachment followed by dissociation was previously observed in the NBs uracil [61] and thymine [62] and also in other molecules, for example, ethylene carbonate [63]. The negative ion signals close to 0 eV are thus tentatively assigned to a DBS, but recent electron scattering calculations on d-ribose [46] predicted a series of shape resonances located between 7.98 and 9.49 eV. In this energy range, between 5.5 and 9.0 eV, the experiments indeed show weaker signals due to the formation of C3 H4 O2 − , C2 H3 O2 − (Figure 6.6), and OH− . This is consistent with the analysis of the spatial distribution of the calculations, which predict that the shape resonances lead to C–OH and C–C bond breaking [46]. 6.2.4 One Step Further: Tetraacetyl-D-Ribose

The DEA studies on d-ribose demonstrated that electron attachment to a sugar molecule is possible and, moreover, leads even at 0 eV to manifold fragmentation accompanied by a decomposition of the ring structure. Consequently, these reactions are highly relevant for the discussion of molecular mechanisms of DNA strand breaks since the fragmentation of the sugar moiety in DNA corresponds to both strand breaks and loss of a nucleobase and thus loss of genetic information. 2) According to Koopmans’ theorem, the

attachment energy (or electron afﬁnity) corresponds to the negative of the virtual orbital energy (VOE) as obtained from a self-consistent ﬁeld calculation. However,

orbital relaxation and electron correlation energies are neglected in this approximation, so that the VOE must be scaled by at least 0.5 eV. See [18] (J. Simons, J. Phys. Chem. A (2008), 112, 6401) for details.

155

156

6 Role of Low-Energy Electrons in DNA Radiation Damage

However, apparently the structure of isolated gas-phase d-ribose is considerably different from the structure of sugar in DNA, since the latter adopts the furanose form in contrast to the pyranose structure of gas-phase d-ribose [54], which is shown in Figure 6.3. Additionally, the 2-deoxy-d-ribose unit in DNA does not possess any free hydroxyl groups, but it may be these hydroxyl groups that enable the electron attachment reactions described above in d-ribose. In order to elucidate whether the reactions described above for bare d-ribose also take place in a sugar that is bound to a more complex molecular environment, we investigated DEA to peracetylated d-ribose [64], which additionally can be obtained in pure furanosic form (Figure 6.3). Of course, the acetyl groups in tetraacetyl-d-ribose (TAR) are only a rough model for the nucleobase and the phosphate groups of the DNA molecule; nevertheless, the TAR molecule is a very impressing example to demonstrate how low-energy electrons react with a complex molecule containing several polar functional groups. Figure 6.8 shows a selection of DEA spectra obtained from TAR. The most dominant fragment anion was observed at 59 amu (top panel of Figure 6.8) and is due to the acetate ion generated through a single C–O bond dissociation: (CH3 COO)4 C5 H6 O + e− (0 eV; 1–3 eV; 7–11 eV) −→ CH3 COO− (59 amu) + (CH3 COO)3 C5 H6 O

(6.11)

It arises from three different resonant features located close to 0 eV, between 1 and 3 eV, and with smaller intensity at 7–11 eV. These distinct energy regimes are characteristic of the behavior of different parts of the molecule. The second spectrum was recorded at 70 amu and represents a group of anions [64] (100, 84, and 70 amu, but only the latter is shown here) that are only formed close to 0 eV and additionally at higher electron energies, but no intensity was observed at medium energies. The shape of the spectra is reminiscent to the DEA spectra of bare d-ribose, and so are the corresponding masses indicating that these fragment anions are associated with processes in the central sugar ring resulting in its decomposition. The 70 amu fragment is ascribed to the following reaction: (CH3 COO)4 C5 H6 O + e− (0 eV; 7–11 eV) −→ C3 H2 O− 2 (70 amu) + fragments

(6.12)

That is, the 0 eV signals are due to the initial formation of a DBS followed by a dissociation of the sugar ring and eventually leading to similar fragment anions as observed from bare d-ribose. In contrast to this the third spectrum of Figure 6.8 represents a group of heavier fragment anions (215, 161, 154, 119, and 113 amu; see [64] for details), that are exclusively generated from the resonance located at 1–3 eV and is due to the loss of the exocyclic acetate groups. The ion at 154 amu is ascribed to the following reaction: (CH3 COO)4 C5 H6 O + e− (1–3 eV) −→ C7 H6 O− 4 (154 amu) + 2CH3 COOH + CH3 CHO

(6.13)

6.2 DEA Studies on Model Compounds for the DNA Backbone

40

59 amu CH3COO−

Ion yield (102 s−1)

20

0 6

0

1

2

3

4

5

6

7

8

4

9 10 11

70 amu C3H2O2−

2 0 12

0

1

2

8

3

4

5

6

7

8

9 10 11

154 amu C7H6O4−

4 0 0

1

2

3 4 5 6 7 8 Electron energy (eV)

Figure 6.8 Representative ion yield curves obtained from DEA to peracetylated D-ribose. The upper panel shows the DEA spectrum of the acetate anion, which is formed within three distinct resonant features. The signal close to 0 eV is tentatively assigned to the initial formation of a dipole-bound state, the peaks at 1–3 eV are ascribed to π ∗ shape resonances located on the acetate groups, and the signals at 7–11 eV are most likely due to core excited resonances

9 10 11

within the sugar ring. The middle spectrum is reminiscent to the DEA spectra of bare D-ribose and is associated with a decomposition of the sugar ring, thus no signal appears at 1–3 eV. The lower spectrum is representative for a group of heavier fragments associated with the abstraction of exocyclic groups exclusively through the π ∗ shape resonance at 1–3 eV.The DEA spectra are adapted from [64].

These spectra clearly indicate that the dissociation through DBS as observed in bare d-ribose is preserved in the appreciably more complex molecule TAR. More general, the experiments demonstrate that in a molecule composed of different moieties characteristic resonant features of the individual subunits are preserved. In TAR these are the polar exocyclic acetate groups leading to π ∗ shape resonances located at 1–3 eV and the central sugar ring resulting in ion formation close to 0 eV and at higher energies between 7 and 11 eV. In Figure 6.9 it is illustrated in the picture of a simpliﬁed two-dimensional potential energy diagram how different resonances contribute to the ion yield of, for example, the CH3 COO− ion at 59 amu. A TNI (indicated as TAR#− ) is formed by a

157

158

6 Role of Low-Energy Electrons in DNA Radiation Damage

TAR−

TAR# −

TAR

E

E CH3COO + fragments

TAR− (DBS) 0

CH3COO− + fragments

0 Ion yield

Q Figure 6.9 Schematic potential energy diagram illustrating the formation of the fragment anion CH3 COO− through the initial formation of two different resonant states of tetraacetyl-D-ribose (TAR) resulting in two distinct signals in the ion yield curve. At 0 eV a dipole-bound state can be formed serving as a doorway to dissociation since the dissociative valence anionic state is occupied

via intramolecular charge transfer. The valence anionic state can also be accessed by a direct Franck–Condon transition resulting in a π ∗ shape resonance. An example for an MO representation of the excess electron in the π ∗ shape resonance is displayed, which was calculated at the B3LYP/6-31++G** level of theory using the stabilization method [64].

Franck–Condon transition requiring energy of about 2 eV. The MO representation of the TNI is a representative example resulting from ab initio calculations using the stabilization method [64]. The TNI formed at 2 eV is unstable with respect to dissociation and different fragment anions can result from this particular TNI, among them the acetate anion. The right-hand side of Figure 6.9 shows the resulting ion yield curve. The same fragment anion can also be formed at 0 eV when a DBS is initially formed. In the DBS the excess electron is still located relatively far away from the molecular framework, thus the geometry of the DBS corresponds to the geometry of the neutral and the binding energy of the excess electron is slightly positive. Consequently, the DBS can be formed at 0 eV, and an intramolecular charge transfer to the valence-bound state then leads to dissociation. The experiments with TAR yield important indications about the electron attachment reactions in DNA. The sugar unit is very active toward low-energy electrons also when it is bound within a more complex molecular environment; the only prerequisite seems to be the presence of polar side groups. In a DNA nucleotide, these are one nucleobase and two phosphate groups that in turn possess considerable electron afﬁnities resulting in shape resonances located at the respective moiety.

6.2 DEA Studies on Model Compounds for the DNA Backbone

In the next two sections, the investigation of a combined sugar–phosphate compound with respect to DEA is described along with the experimental challenges that are connected with the study of such molecules. 6.2.5 The Use of Laser-Induced Acoustic Desorption (LIAD) to Study DEA to Larger Biomolecules

The study of biomolecules in the gas phase, which are larger than single NBs or monosaccharides such as nucleosides or nucleotides, raises problems due to their thermal fragility. It was demonstrated previously that conventional sublimation of nucleosides such as thymidine in high vacuum by appropriate heating results in thermal decomposition [65]. As gas-phase experiments yield a wealth of information and due to their function as a reference for both theoretical calculations and experiments in more biologically relevant environments, more sophisticated methods are required to transfer biomolecules into the gas phase to make them accessible for DEA measurements. A very gentle method to desorb only neutral molecules without ionization is laser-induced acoustic desorption (LIAD) [66–68]. The sample molecules are deposited on a thin titanium foil, which is then irradiated from the backside with a pulsed Nd:YAG laser operated at 532 nm (15 Hz, 3 mJ per pulse, 2–6 ns). The laser pulse induces a shock wave that propagates through the metal foil and leads to a gentle desorption of sample molecules from the opposite surface, without resulting in fragmentation or ionization of the sample molecules. The desorbed neutral molecules then interact with an electron beam of deﬁned but variable energy, and the thus formed anions are analyzed by a mass spectrometer. The proof-of-principle experiments were performed by using the nucleoside thymidine [69] and the sugar–phosphate compound d-ribose-5 -phosphate [69] (RP), which is shortly discussed in the next section. 6.2.6 Sugar–Phosphate Cleavage Induced by 0 eV Electrons: DEA to D-Ribose-5 -Phosphate

d-Ribose-5 -phosphate can be understood as a model compound for the entire DNA or RNA backbone. Figure 6.10a shows DEA spectra exhibiting formation of the H2 PO4 − anion and the ribose anion C5 H9 O5 − that are due to a C–O or a P–O bond cleavage, respectively: e− (≈ 0 eV) + (HO)2 OPO–C5 H9 O4 −→ H2 PO− 4 + C5 H9 O4

(6.14)

e− (≈ 0 eV) + (HO)2 OP–OC5 H9 O4 −→ H2 PO3 + C5 H9 O− 5

(6.15)

Both reactions occur close to 0 eV and represent a sugar–phosphate cleavage corresponding to a strand break in DNA or RNA. A more detailed analysis of the reactions by means of DFT calculations at the B3LYP/6-31G++∗∗ level of theory [69] reveals the energetics of the reaction pathways as shown in Figure 6.10b. The

159

0

2

4

6

8

0

2

4

6

8

10

12

14

−1 0

1

2

3 E (eV)

4

5

6 7

149 amu C5H9O5−

97 amu H2PO4−

8

9

(b)

HO

A

OH

P

O O

OH

O

OH

H OH

H

−0.19 eV

H

H

B

+0.43 eV

+0.78 eV

O

−

C

HO

H

OH

P

O

OH

H

O

+ H2PO3

−2.10 eV

O− + C H O 5 9 4

−0.20 eV

H OH

H

OH

Figure 6.10 (a) Ion yields of the phosphate anion H2 PO4 − and the D-ribose anion C5 H9 O5 − generated by DEA to D-ribose-5 -phosphate (RP) through a sugar–phosphate cleavage. Compared to the situation in DNA, both reactions correspond to a strand break. (b) Activation barriers for both reaction pathways. All the molecules were fully optimized at the B3LYP/6-31++G** level of theory and the given energies are in respect to the neutral ground state RP molecule, which is not shown here. A represents the substrate anion, B the transition states and C the sum of the isolated products. The DEA spectra and illustration are adapted from [69].

(a)

Ion count rate (s−1)

160

6 Role of Low-Energy Electrons in DNA Radiation Damage

6.3 Outlook and Future Prospects

calculations indicate that ribose-5 -phosphate can form a stable molecular anion with an electron binding energy of 0.2 eV. The dissociation into the phosphate anion is driven by its thermodynamic stability reﬂected by the EA of the corresponding neutral (EA(H2 PO4 ) = 4.9 eV [70]). In contrast, the generation of the ribose fragment anion leads to products that are barely more stable than the RP− anion. Nevertheless, the activation energy for P–O bond breaking is appreciably lower than for the H2 PO4 − formation, making it a kinetically controlled reaction. The fact that two different bonds are broken and that the excess charge is subsequently localized on either the phosphate group or the sugar unit suggests that two different resonances are involved. The electron attachment mechanisms to d-ribose are discussed in detail above, but also the phosphate group is known to capture electrons at various energies down to 0 eV [71]. The low energy of the signals indicates that also in d-ribose-5-phosphate DBSs may be involved, which is supported by a recent theoretical and experimental study excluding low-lying π ∗ shape resonances in phosphate compounds [72].

6.3 Outlook and Future Prospects

The investigation of molecular mechanisms of DNA radiation damage and the role of low-energy electrons is beneﬁcial for an improved understanding of the mode of action of radiosensitizers in radiation and chemoradiation therapy for tumor treatment [73, 74]. For instance, the chemotherapeutic agent cisplatin was shown to enhance the yield of electron-induced SSBs and DSBs in DNA by factors of 1.3–4.4 depending on the electron energy [74]. The enhancement is largest for 10 eV electrons, which is ascribed to an increase in the magnitude of DEA processes in the presence of cisplatin. A detailed study of the underlying DEA processes may have implications for the design of new chemotherapeutic and radiosensitizing drugs and for the development of more efﬁcient protocols in cancer therapy. The understanding of molecular mechanisms of electron-induced DNA strand breaks has improved considerably over the past few years, but a number of controversial results still demonstrate the complexity of the system. For example, it still remains to be evaluated whether DNA strand breaks are mainly induced by initial localization of low-energy electrons at the NBs or in the backbone. The NBs are usually regarded as antennas for low-energy electrons, and the role of NBs is reﬂected, for instance, in the sequence dependence of electron capture by DNA [75]. However, at different electron energies different electron attachment mechanisms prevail and various experiments point out the active role of the sugar–phosphate backbone in electron attachment [8, 42, 69, 71, 76]. The experiments described in the present chapter indicate that electron attachment to the backbone subunits proceeds very effectively presumably via the initial formation of DBSs. It must be investigated if such DBSs are also relevant in a more biologically realistic environment [59].

161

162

6 Role of Low-Energy Electrons in DNA Radiation Damage

In general, it is crucial to determine the effect of solvation by water molecules on electron-induced reactions in biomolecular systems. This question was tracked recently by experiments performed with ﬁlms of short oligonucleotides covered by water [77] showing that fragment anion formation is increased by a factor of 1.6 in the presence of water. The recently initiated experiments combining DEA and LIAD offer a new scope of experiments that can be performed, for example, with oligonucleotides or peptides. The role of DNA binding proteins in electron-induced DNA damage is an important issue that was recently addressed by experiments using self-assembled monolayers [78] and must be further investigated. Apart from their relevance to radiation damage, the experiments on electroninduced reactions in isotope-labeled sugars (and previously in NBs) revealed that in these cases bond cleavage is highly energy- and site selective. As is indicated in Section 6.1.2 it has been shown in the case of pyrimidine nucleobases [33] and the purine nucleobase adenine [30] how the electron-induced hydrogen abstraction from the nucleobase can be controlled by simply changing the energy of the electron beam. More precisely, the energy of the incoming electron determines, on the one hand, the speciﬁc atom from which hydrogen is abstracted (either N1 or N3) and, on the other hand, if the negative charge remains on the nucleobase resulting in the dehydrogenated nucleobase anion or on the hydrogen atom leading to the hydride anion [26, 28]. In another DEA study about organic molecules containing different functional groups [79, 80], it was demonstrated that different TNIs are formed at electron energies speciﬁc for a certain functional group leading to fragmentation products at characteristic electron energies. In other words, a particular bond cleavage can be triggered just by tuning the energy of the electron beam. This can technically be applied to a controlled manipulation of materials at the nanoscale (nanolithography) [52]. Low-energy electron–induced polymerization of CH3 SSCH3 involving electron attachment, and subsequent dissociation was demonstrated very recently by means of STM [81]. The initial step is attachment of an electron emitted by the STM tip followed by dissociation along the S–S bond and subsequent reaction with an adjacent CH3 SSCH3 molecule. This example demonstrates the potential of DEA reactions as a route to synthesize novel materials using low-energy electron sources such as the STM, which has the advantage of possessing a remarkable spatial resolution.

Acknowledgments

The work presented here was conducted within my PhD project that was supervised by Prof. Eugen Illenberger who was an excellent mentor whom I hold in great respect and gratitude. Furthermore, I acknowledge very fruitful collaborations and discussions with Dr. Janina Kopyra and Dr. Iwona Dabkowska. Finally, a PhD fellowship by the Studienstiftung des deutschen Volkes is gratefully acknowledged.

References

References 1. Michael, B.D. and O’Neill, P. (2000) 2.

3. 4. 5.

6. 7. 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

Science, 287, 1603. Boudaiffa, B., Cloutier, P., Hunting, D., Huels, M.A., and Sanche, L. (2000) Science, 287, 1658. Sanche, L. (2005) Eur. Phys. J. D, 35, 367. Ball, P. (2003) Nature, 421, 421. Luger, K., Mader, A.W., Richmond, R.K., Sargent, D.F., and Richmond, T.J. (1997) Nature, 389, 251. Felsenfeld, G.and Groudine, M. (2003) Nature, 421, 448. Pimblott, S.M.and LaVerne, J.A. (2007) Radiat. Phys. Chem., 76, 1244. Chen, Y.F., Aleksandrov, A., and Orlando, T.M. (2008) Int. J. Mass Spectrom., 277, 314. Huels, M.A., Boudaiffa, B., Cloutier, P., Hunting, D., and Sanche, L. (2003) J. Am. Chem. Soc., 125, 4467. Martin, F., Burrow, P.D., Cai, Z., Cloutier, P., Hunting, D., and Sanche, L. (2004) Phys. Rev. Lett., 93, 068101. Panajotovic, R., Martin, F., Cloutier, P., Hunting, D., and Sanche, L. (2006) Radiat. Res., 165, 452. Zheng, Y., Wagner, J.R., and Sanche, L. (2006) Phys. Rev. Lett., 96, 208101. Mochiji, K., Hashimoto, H., Tanaka, Y., Ninomiya, N., and Takeo, M. (2007) Phys. Rev. B, 75, 094302. Cai, Z., Cloutier, P., Hunting, D., and Sanche, L. (2005) J. Phys. Chem. B, 109, 4796. Illenberger, E., and Momigny, J. (1992) Gaseous Molecular Ions. An Introduction to Elementary Processes Induced by Ionization, Springer, Steinkopff . Illenberger, E. (2000) Electron attachment processes to free and bound molecules, in Photoinonization and Photodetachment, Advanced Series in Physical-Chemistry, Vol. 10B (ed. C.-Y. Ng), World Scientiﬁc, Singapore Part II. Bald, I., Langer, J., Tegeder, P., and Ingolfsson, O. (2008) Int. J. Mass Spectrom., 277, 4.

18. Simons, J. (2008) J. Phys. Chem. A, 112,

6401. 19. Balog, R., Langer, J., Gohlke, S.,

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

Stano, M., Abdoul-Carime, H., and Illenberger, E. (2004) Int. J. Mass Spectrom., 233, 267. Kopyra, J., Koenig-Lehmann, C., and Illenberger, E. (2009) Int. J. Mass Spectrom., 281, 89. McConkey, J.W., Malone, C.Y., Johnson, P.V., Winstead, C., McKoy, V., and Kanik, I. (2008) Phys. Rep.: Rev. Sec. Phys. Lett., 466, 1. Baccarelli, I., Bald, I., Illenberger, E., and Gianturco, F.A. (2009) Phys. Rep. Rev. Sec. Phys. Lett., in preparation. Huels, M.A., Hahndorf, I., Illenberger, E., and Sanche, L. (1998) J. Chem. Phys., 108, 1309. Abdoul-Carime, H., Gohlke, S., and Illenberger, E. (2004) Phys. Rev. Lett., 92, 168103. Abouaf, R., Pommier, J., and Dunet, H. (2003) Int. J. Mass Spectrom., 226, 397. Ptasinska, S., Deniﬂ, S., Grill, V., Maerk, T.D., Scheier, P., Gohlke, S., Huels, M.A., and Illenberger, E. (2005) Angew. Chem. Int. Ed., 44, 1647. Ptasinska, S., Deniﬂ, S., Mroz, B., Probst, M., Grill, V., Illenberger, E., Scheier, P., and Mark, T.D. (2005) J. Chem. Phys., 123, 124302. Ptasinska, S., Deniﬂ, S., Scheier, P., Illenberger, E., and Maerk, T.D. (2005) Angew. Chem. Int. Ed., 44, 6941. Abdoul-Carime, H., Langer, J., Huels, M.A., and Illenberger, E. (2005) Eur. Phys. J. D, 35, 399. Deniﬂ, S., Sulzer, P., Huber, D., Zappa, F., Probst, M., Mark, T.D., Scheier, P., Injan, N., Limtrakul, J., Abouaf, R., and Dunet, H. (2007) Angew. Chem. Int. Ed., 46, 5238. Hanel, G., Gstir, B., Deniﬂ, S., Scheier, P., Probst, M., Farizon, B., Farizon, M., Illenberger, E., and Mark, T.D. (2003) Phys. Rev. Lett., 90, 4188104. Aﬂatooni, K., Scheer, A.M., and Burrow, P.D. (2005) Chem. Phys. Lett., 408, 426.

163

164

6 Role of Low-Energy Electrons in DNA Radiation Damage 33. Swiderek, P. (2006) Angew. Chem. Int. 34.

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

42.

43. 44.

45.

46.

47. 48.

49.

50.

51.

Ed., 45, 4056. Berdys, J., Anusiewicz, I., Skurski, P., and Simons, J. (2004) J. Am. Chem. Soc., 126, 6441. Li, X., Sevilla, M.D., and Sanche, L. (2003) J. Am. Chem. Soc., 125, 13668. Simons, J. (2006) Acc. Chem. Res., 39, 772. Gu, J.D., Wang, J., and Leszczynski, J. (2006) J. Am. Chem. Soc., 128, 9322. Gu, J.D., Xie, Y.M., and Schaefer, H.F. (2006) J. Am. Chem. Soc., 128, 1250. Kumar, A.and Sevilla, M.D. (2007) J. Phys. Chem. B, 111, 5464. Kumar, A. and Sevilla, M.D. (2008) J. Am. Chem. Soc., 130, 2130. Dabkowska, I., Rak, J., and Gutowski, M. (2005) Eur. Phys. J. D, 35, 429. Vilar, M.R., do Rego, A.M.B., Ferraria, A.M., Jugnet, Y., Nogues, C., Peled, D., and Naaman, R. (2008) J. Phys. Chem. B, 112, 6957. Bald, I., Kopyra, J., and Illenberger, E. (2006) Angew. Chem. Int. Ed., 45, 4851. Ptasinska, S., Deniﬂ, S., Scheier, P., and Mark, T.D. (2004) J. Chem. Phys., 120, 8505. Sulzer, P., Ptasinska, S., Zappa, F., Mielewska, B., Milosavljevic, A.R., Scheier, P., Mark, T.D., Bald, I., Gohlke, S., Huels, M.A., and Illenberger, E. (2006) J. Chem. Phys., 125, 044304. Baccarelli, I., Gianturco, F.A., Grandi, A., Sanna, N., Lucchese, R.R., Bald, I., Kopyra, J., and Illenberger, E. (2007) J. Am. Chem. Soc., 129, 6269. Lide D. (ed.) (2002) CRC Handbook of Chemistry and Physics. Bald, I., Flosadottir, H.D., Kopyra, J., Illenberger, E., and Ingolfsson, O. (2009) Int. J. Mass Spectrom., 280, 190. Deng, Z.W., Bald, I., Illenberger, E., and Huels, M.A. (2008) Angew. Chem. Int. Ed., 47, 9509. Levis, R.J., Menkir, G.M., and Rabitz, H. (2001) Science, 292, 709. Brixner, T. and Gerber, G. (2003) Chemphyschem, 4, 418.

52. Mason, N.J. (2008) Int. J. Mass Spec-

trom., 277, 31. 53. Bald, I., Dabkowska, I., Illenberger, E.,

54.

55.

56. 57.

58.

59. 60. 61.

62.

63.

64.

65.

66.

67.

68.

and Ingolfsson, O. (2007) Phys. Chem. Chem. Phys., 9, 2983. Guler, L.P., Yu Y.-Q., and Kenttaemaa, H.I. (2002) J. Phys. Chem. A, 106, 6754. Bald, I. (2007) Low Energy Electron Induced Reactions in Gas Phase Biomolecules – The Role of the Sugar and Phosphate Moiety in DNA Damage PhD thesis, Freie Univerit¨at Berlin, Berlin. Schulz, G.J. (1973) Rev. Mod. Phys., 45, 423. Ibanescu B.C., May O., Monney, A., and Allan, M. (2007) Phys. Chem. Chem. Phys., 9, 3163. Hotop, H., Ruf, M.W., Allan, M., and Fabrikant, I.I. (2003) Adv. Atom. Mol. Opt. Phys., 49, 85. Jordan, K.D. and Wang, F. (2003) Ann. Rev. Phys. Chem., 54, 367. Sommerfeld, T. (2007) J. Chem. Phys. 126, 124301/1. Scheer, A.M., Aﬂatooni, K., Gallup, G.A., and Burrow, P.D. (2004) Phys. Rev. Lett., 92 068102. Burrow, P.D., Gallup, G.A., Scheer, A.M., Deniﬂ, S., Ptasinska, S., Mark, T., and Scheier, P. (2006), J. Chem. Phys., 124, 124310. Stepanovic, M., Pariat, Y., and Allan, M. (1999) J. Chem. Phys., 110, 11376. Bald, I., Kopyra, J., Dabkowska, I., Antonsson, E., and Illenberger, E. (2007) J. Chem. Phys., 126, 074308. Ptasinska, S. Deniﬂ, S., Gohlke, S., Scheier, P., Illenberger, E., and Maerk, T.D. (2006) Angew. Chem. Int. Ed., 45, 1893. Golovlev, V.V., Allman, S.L., Garrett, W.R., Taranenko, N.I., and Chen, C.H. (1997) Int. J. Mass Spectrom., 169/170, 69. Perez, J., Petzold, C.J., Watkins, M.A., Vaughn, W.E., and Kenttamaa, H.I. (1999) J. Am. Soc. Mass Spectrosc., 10, 1105. Shea, R.C., Petzold, C.J., Campbell, J.L., Li, S., Aaserud, D.J., and

References

69.

70.

71.

72.

73. 74.

Kenttaemaa, H.I. (2006) Anal. Chem., 78, 6133. Bald, I., Dabkowska, I., and Illenberger, E. (2008) Angew. Chem. Int. Ed., 47, 8518. Wang, X.-B., Vorpagel, E.R., Yang, X., and Wang, L.-S. (2001) J. Phys. Chem. A, 105, 10468. Konig, C., Kopyra, J., Bald, I., and Illenberger, E. (2006) Phys. Rev. Lett., 97, 018105. Burrow, P.D., Gallup, G.A., and Modelli, A. (2008) J. Phys. Chem. A, 112, 4106. Gohlke, S. and Illenberger, E. (2002) Europhys. News, 33, 207. Zheng, Y., Hunting, D.J., Ayotte, P., and Sanche, L. (2008) Phys. Rev. Lett., 100, 4198101.

75. Ray, S.G., Daube, S.S., and Naaman, R.

76. 77. 78. 79.

80.

81.

(2005) Proc. Natl. Acad. Sci. U.S.A., 102, 15. Pan, X. and Sanche, L. (2006) Chem. Phys. Lett., 421, 404. Ptasinska, S. and Sanche, L. (2007) Phys. Rev. E, 75, 031915. Solomun, T. and Skalicky, T. (2008) Chem. Phys. Lett., 453, 101. Prabhudesai, V.S., Kelkar, A.H., Nandi D., and Krishnakumar, E. (2005) Phys. Rev. Lett., 95, 143202. Prabhudesai, V.S., Nandi, D., Kelkar, A.H., and Krishnakumar, E. (2008) J. Chem. Phys., 128, 154309. Maksymovych, P., Sorescu, D.C., Jordan, K.D., and Yates, J.T. (2008) Science, 322, 1664.

165

167

7 Structure-Based Design on the Way to New Anti-Infectives Anna Katharina Herta Hirsch

7.1 Introduction

With an estimated 300–500 million new infections and three million deaths annually, malaria and tuberculosis undoubtedly still pose a major health concern [1]. The need for the development of novel therapeutic approaches is ever-growing in light of the emergence of multi-drug-resistant parasites [2]. How can a new drug be identiﬁed? First, a lead compound, that is, a molecule with a promising biological activity that does not yet fulﬁll all the requirements but represents the starting point on the way to a new drug, has to be identiﬁed. A number of different strategies exist to achieve this goal: • High-throughput screening (HTS) of a large library of small molecules is of particular interest in cases in which no structural information or characterization of the biological target is available. The majority of lead compounds still comes from hits identiﬁed by HTS [3]. • Virtual screening has established itself as an alternative or complementary approach to classical HTS. Potentially active and/or drug-like compounds are selected from a library of compounds, using elaborate docking and scoring functions [4]. • Combinatorial chemistry is useful for the formation of large, small-molecule libraries. However, this approach is less effective for generating a great deal of structural diversity. • Nature can be of help by providing a rich and diverse source of structural inspiration [5]. The scaffold of a natural product, displaying interesting biological properties, could be developed into a new drug. • Structure-based design, a comparatively new ﬁeld, has established itself in pharmaceutical research as a valuable alternative to traditional screening; the X-ray crystal structure of a target enzyme is used as a basis for lead compound identiﬁcation and optimization. The increasing number of leads, identiﬁed and/or optimized using this rational approach, used for the development of new drugs illustrates Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

168

7 Structure-Based Design on the Way to New Anti-Infectives

this fact [6]. Until now, this strategy has been mainly used in the later stages of lead optimization. The last mentioned strategy will be employed for the purposes of this project. The aim of this approach is to identify synthetically accessible target molecules, with optimal stereoelectronic properties that are complementary to the binding site of the target enzyme and show minimal or no repulsive interactions when complexed to the enzyme. The identiﬁcation of promising leads is aided by the medicinal chemist’s understanding of molecular recognition. While hydrophobic interactions between a lead and an enzyme are the main driving force for complexation, H-bonding interactions account for selectivity (Figure 7.1). Later stages of drug development are not the subject of this chapter. To determine the biological activity of a potential lead compound, a new biological target [7], that is, an enzyme or receptor that upon interference by the ligand/drug has an impact on the disease causing pathogen in the desired way, has to be identiﬁed. In an ideal case, the target is essential for the pathogen and not present in humans, thereby precluding any selectivity issues. Isoprenoids are an essential class of natural products, requiring the essential precursors isopentenyl diphosphate (IPP, 1) and dimethylallyl diphosphate (DMAPP, 2) for their biosynthesis (Scheme 7.1). Until recently, only one route to the universal isoprenoid precursors was known, the so-called mevalonate pathway, using acetyl-coenzyme A as the only building block [8]. A completely distinct alternative to this well-established biosynthetic route, now known as the nonmevalonate pathway, was discovered in the early 1990s, starting from pyruvate (3) and glyceraldehyde 3-phosphate (4) [9]. Interestingly, this biosynthetic pathway is exclusively used by a number of pathogens such as the malarial parasite Plasmodium falciparum and the tuberculosis-causing Mycobacterium tuberculosis and not by higher eukaryotes (e.g., humans), which means that inhibition of the constituent enzymes of the nonmevalonate pathway affects and kills only the parasites, leaving the patient untouched (Scheme 7.1). Thus, this pathway has provided a rich source of new, highly attractive drug targets. To illustrate the use of a structure-based design cycle, the development of the ﬁrst inhibitors of the kinase IspE of the nonmevalonate pathway is described below, constituting a novel approach toward anti-infectives. First, the biosynthetic pathway

Target enzyme

Inhibitor

Key requirements: Steric complementarity Electronic complementarity Noncovalent interactions

Figure 7.1

Schematic representation of structure-based inhibitor design.

7.2 Isoprenoids and the Nonmevalonate Pathway

O

O S

O CO2−

CoA

Acetyl CoA

+

H

3

Pathogens: nonmevalonate pathway

Humans: Mevalonate pathway

OPP

OP OH 4

+

OPP 2

1

Target: block the kinase IspE

Isoprenoids Scheme 7.1 Schematic representation of the biosynthesis of the isoprenoid precursors IPP (1) and DMAPP (2). P = phosphate group.

is brieﬂy presented with an overview of the known inhibitors of the constituent enzymes. Subsequently, the chosen target, the kinase IspE, is introduced and its structure is discussed in some detail. This will provide the basis for the rationale of the design of the ﬁrst-generation inhibitors. Second, the active site is explored by showing the possibility of applying concepts from supramolecular chemistry to an enzymatic context. Finally, the development of water-soluble inhibitors is described with the aim of obtaining an X-ray cocrystal structure to verify the proposed binding mode.

7.2 Isoprenoids and the Nonmevalonate Pathway

There are more than 35 000 known isoprenoids, which fulﬁll a myriad of important biological functions. Despite their striking structural diversity, all isoprenoids are biosynthesized from the two simple ﬁve-carbon building blocks 1 and 2 (Scheme 7.1). This concept is also known as the isoprene rule [10]. The nonmevalonate pathway starts with the head-to-tail condensation of the two- and three-carbon precursors 3 and 4. A total of seven enzymes catalyze the conversion of these starting materials into the essential isoprenoid precursors 1 and 2. Fosmidomycin, an inhibitor of the second enzyme of the pathway (IspC), has been shown to cure malaria in rodents, thereby validating the constituent enzymes as drug targets [11]. This ﬁnding triggered research efforts aimed at the elucidation of the structures

169

170

7 Structure-Based Design on the Way to New Anti-Infectives

and mechanisms of the participating enzymes. As a result, detailed structural and mechanistic data exist for most enzymes [12], setting the stage for lead generation by structure-based design. A number of enzymes of the nonmevalonate pathway have been chosen as targets to achieve this goal. The rather hydrophilic nature of the active sites renders the development of low-molecular-weight inhibitors challenging. Thus, it comes as no surprise that the few reported inhibitors either bear phosphate or phosphonate groups or display rather modest inhibition (Table 7.1) [13]. 7.2.1 4-Diphosphocytidyl-2C-methyl-D-erythritol Kinase (IspE)

The absence of known inhibitors, the fact that the kinase IspE belongs to the nonmevalonate pathway, and the availability of an X-ray crystal structure make the fourth enzyme of the pathway an ideal target for rational design of potent, drug-like inhibitors without the use of the problematic phosphate or phosphonate moieties. IspE (EC 2.7.1.148) employs adenosine 5 -triphosphate (ATP) and Mg2+ cations for the phosphorylation of the C(2)-hydroxyl group of 4-diphosphocytidyl2C-methyl-d-erythritol (5) to afford 4-diphosphocytidyl-2C-methyl-d-erythritol 2-phosphate (6, Scheme 7.2) [18]. This central reaction is the only ATP-dependent step of the whole biosynthetic pathway. Sequence comparisons have shown that IspE belongs to the galactose/ homoserine/mevalonate/phosphomevalonate (GHMP) kinase superfamily [19]. 7.2.2 Structure of IspE

In 2003, the ﬁrst crystal structures of IspE from Thermus thermophilus and Escherichia coli were solved, of the apoenzyme and of a ternary complex, respectively [20, 21]. Recently, the crystal structure of Aquifex aeolicus IspE was solved as a complex with a number of natural ligands [22], a synthetic substrate mimic [22], and synthetic cytidine [23] as well as cytosine (Section 7.4) [24] derivatives. To date, no crystal structure is available for IspE from a pathogen, for example, M. tuberculosis or P. falciparum. IspE generally crystallizes as a homodimer with each monomer displaying the characteristic two-domain fold of the GHMP kinase superfamily that consists of an ATP- and a substrate-binding domain. The dimer clasps around a solvent-ﬁlled channel, featuring two active sites at either end (Figure 7.2). 7.2.3 Active Site of IspE

The active site of E. coli IspE was used for modeling. It can be divided into three main pockets: the ATP-, the cytidine 5 -diphosphate (CDP)-, and the methylerythritol (ME)-binding pockets. Molecular modeling, using the program MOLOC

R2

N

O

N

R4 R3

Cl

O

OH

N

F

H

O

PO32−

CO2−

n

CDP 2 X+ O2 S N H

+ 2 CDP 2 X

Substrate and transition-state analogues

O

R

1

O

O

H N

2

R

OH

O

N

N

R3

N

N

R4

H N

NH

Mechanism-based inhibitors

NH

4

HN

HO

N NH2

IC50 = 0.45 mM [16] Ki = 0.011 µM [17]

Ki = 0.9 nM [15]

IC50 = 0.08 mM [14]

Inhibition

The inhibition of the best inhibitor is given. CDP = cytidine 5 -diphosphate; DXS and IspF: ﬁrst and ﬁfth enzyme of the nonmevalonate pathway, respectively; IC50 = concentration of inhibitor at which 50% maximum initial velocity is observed; Ki = inhibition constant.

NH2 O

No known inhibitors at the onset of the present design cycle NMe2

R1

O

O

O

Inhibitors

Known inhibitors of enzymes of the nonmevalonate pathway.

IspF IPP isomerase

IspC IspE

DXS

Enzyme

Table 7.1

7.2 Isoprenoids and the Nonmevalonate Pathway 171

172

7 Structure-Based Design on the Way to New Anti-Infectives

NH2

DXS O

O

IspC

+

OP

H

CO2−

IspD

OH

3

1 OH

4

N

HO 2

3

O

4

O

O P O P O O−

OH

O

N

O

O− 5

HO

OH

ATP IspE ADP

OPP 1

NH2

IspF IPP isomerase

IspG IspH

N

PO

OH

O OH

O

O P O P O O− O−

OPP 6

2 Scheme 7.2

Simpliﬁed version of the nonmevalonate pathway.

AMP-PNP

5 Figure 7.2 Schematic representation of the ternary complex of E. coli IspE, 5 -adenyl-β, γ -amidotriphosphate (AMP-PNP) and the substrate 5, cocrystallized as a homodimer (Protein Data Bank (PDB) code: 1OJ4) [20].

O

HO

N

OH

O

7.2 Isoprenoids and the Nonmevalonate Pathway Small, hydrophobic pocket

ME-binding pocket

Leu15

Leu28

Lys10 H 2N

Asn12

Lys96 Asn10

O H2N

Phe185

+

His26 O NH

Val157 O

H3N+

N

−O

Pro182

NH

Thr181 OH N

O

O

S

NH3 O

Val156 Asp141 Met100

H2N H N

HN Tyr25

173

Leu66

OH CDP-binding pocket

ATP-binding pocket

Figure 7.3 Schematic representation of the four pockets of the active site of E. coli IspE. The residues of the glycine-rich loop were omitted for clarity.

[25], revealed that an additional small, hydrophobic pocket lies adjacent to the CDP-binding site (Figure 7.3). The ATP-binding pocket features a glycine-rich phosphate-binding loop, typically displaying an adjacent positively charged N terminus of an α helix [20]. The adenine moiety is accommodated in a hydrophobic cleft lined by Val57, Val60, Leu66, Ile67, Lys96, and Met100. Numerous hydrophobic contacts offered by this pocket certainly make a large contribution to the binding enthalpy, as predicted by MOLOC. Additional stabilization is derived from a network of H bonds to the nucleobase moiety. The ribose moiety of 5 -adenyl-β, γ -amidotriphosphate (AMP-PNP) is solvent-exposed and does not show any contacts with the protein. The CDP-binding pocket accommodates the cytosine moiety in a π sandwich, consisting of the side chains of Tyr25 and Phe18 held in place by – stacking interactions. The ribose moiety beneﬁts from stabilization by a pseudo-π sandwich that is composed of the aromatic side chain of Tyr25 and the aliphatic Pro182. A pseudo-π sandwich can be deﬁned as a π sandwich, in which one of the two aromatic rings is replaced by an aliphatic ring. His26 is a key residue for the recognition of cytosine, which involves a total of three H-bonding interactions. The ribose and phosphate groups are stabilized by a number of solvent-mediated interactions and a H bond from the side chain of Tyr25 (Figure 7.4). IspE was shown to have high substrate speciﬁcity [21]. Hence, the cytosine moiety must play a key role in substrate recognition and thus selectivity, providing the starting point for the structure-based design of the ﬁrst-generation inhibitors (Section 7.3).

lle67 Asn65 Val160

174

7 Structure-Based Design on the Way to New Anti-Infectives

Asn12

Leu15

Lys10

Leu28

Asp141

Phe185 His26

Val156 Thr181 Pro182 (a)

Tyr25

(b)

Figure 7.4 (a) CDP- and ME-binding and the small, hydrophobic pockets of E. coli IspE. (b) Active site of E. coli IspE (PDB code: 1OJ4) [20]. Color code: protein skeleton: C: gray; skeleton of 5: C: green; O: red; N: blue; P: orange.

7.3 Targeting the CDP-Binding Pocket of IspE 7.3.1 Design

Careful examination of the active site revealed the CDP-binding pocket to be more attractive as a target for inhibitor design than the other binding pockets for a number of reasons. A perfect setup for a double π sandwich and other recognition features should endow potential inhibitors with higher selectivity. Furthermore, given the myriad of proteins that use ATP as a cofactor, inhibitors designed to target the ATP-binding pocket bear a high risk of selectivity issues. Given that neither substrate nor cofactor bind to the adjacent, hydrophobic pocket, presumably, both afﬁnity and selectivity could be gained by occupying it. As this small cavity lies adjacent to the CDP-binding pocket, potential inhibitors should be designed to occupy both the CDP-binding pocket and the newly discovered subpocket, leaving the hydrophilic ME- as well as the ATP-binding pockets unoccupied. In a ﬁrst round of design, cytosine was chosen as a central scaffold to position potential inhibitors in the CDP-binding pocket. According to modeling, the nucleobase moiety was predicted to be sandwiched between Tyr25 and Phe185. By analogy to the natural substrate 5, the cytosine moiety was postulated to be able to form H bonds to His26. It was concluded that the central platform should be decorated with a suitable ribose analogue at the N(1) position, which should be held in place by the pseudo-π sandwich and a vector designed such that its

7.3 Targeting the CDP-Binding Pocket of IspE

175

substituent would be placed in the hydrophobic subpocket. If this were to be achieved, numerous hydrophobic interactions would result. By connecting the linker to the C(5) position, it should serve to address the catalytically essential residues Lys10 and Asp141. Because of the modular design, the different components, that is the ribose analogue, the vector, and the central scaffold, should be easy to vary and optimize (Figure 7.5a). 7.3.1.1 Possible Ribose Analogues As ribose analogues, both heteroalicyclic and aromatic rings can be envisaged to ﬁll the space provided by the pseudo-π sandwich (Figure 7.5b, compounds of type 7 and 8). Introduction of a saturated ring featuring a sulfur atom, for example, a tetrahydrothiophenyl ring, presumably would enable an additional sulfur–aromatic interaction with the phenolic ring of Tyr25 [27]. Modeling predicted both enantiomers of a tetrahydrothiophenyl derivative to bind with similar strength due to the conformational ﬂexibility of ﬁve-membered rings. Four questions need to be answered regarding the choice of the ribose analogue: (i) Which is the ideal ring size? (ii) Is an aromatic or an aliphatic system favored? (iii) Are heteroatoms beneﬁcial? (iv) Is a connecting methylene linker beneﬁcial or detrimental to afﬁnity? NH2 NH2 4 N 5 2 N 6 O

CDP-binding pocket

R2

N N R1

O

Small, hydrophobic pocket

R1 = ribose analogue

Inhibitor R2 Ribose analogue

(a)

O NH2 N O

N H

(c)

clogP

(±)-9 Alicyclic amine 10 OH

O S

R

O NH2

2

O

(b)

Saturated cyclic Methylene bridged

S

Inhibitor R

clogPa 0.3 to 1.0 −0.4 to 1.5

Figure 7.5 (a) Modular design of ﬁrst-generation inhibitors. (b) Ribose analogues, (c) different vectors, and (d) sulfone substituents envisaged for ﬁrst-generation inhibitors. a clogP values (calculated

O S

N

R2 = Et, c -Pr, CH2CF3

7 8

N H

N

N R1

Inhibitor R1

1.1 to 2.3 1.2

(d)

(±)-11 Acyclic (±)-12 Cyclic

clogP 0.3 to 2.0 −0.03 to 2.6

partitioning coefﬁcient) were calculated with the program ACD/LogP [26]. The tetrahydrothiophenyl and the tetrahydrofuranyl derivatives of type 7 and 8, respectively, are chiral.

R

176

7 Structure-Based Design on the Way to New Anti-Infectives

7.3.1.2 Design of the Vector For the vector, a propargylic sulfonamide substituent at the C(5) position of cytosine was envisaged, which displays three attractive features. First, the alkyne ensures a certain rigidity and linearity. Second, N-substituted sulfonamides are known to prefer a conformation in which the lone pair of the nitrogen atom bisects the O–S–O angle, resulting in a staggered arrangement [28]. In its preferred conformation, the sulfonamide is expected to form ionic H bonds to the side chains of Lys10 and Asp141. Finally, small complementary sulfone substituents (R2 = Et, Figure 7.5a), should orient directly into the subpocket. However, before this part of the vector is optimized (Section 7.3.4), the importance of the sulfonamide’s contribution to afﬁnity should be evaluated (Section 7.3.3). An exemplary inhibitor of type 7 features a tetrahydrothiophenyl ring as a ribose mimic (R1 ) and an ethyl group as a sulfone substituent (R2 ) (Figure 7.5b). According to modeling, the ligand should beneﬁt from H bonds to His26, Lys10, and Asp141 as well as from the postulated sulfur–aromatic interaction (Figure 7.6a). An overlay of the natural substrate 5 and the potential inhibitor showed the sulfone moiety to be almost perfectly superimposed with the C(2)-hydroxyl group that is to be phosphorylated (Figure 7.6b). The synthesis of the ﬁrst representative of the ﬁrst-generation inhibitors was achieved using a convergent strategy based on the Sonogashira cross-coupling reaction [29, 30]. With the synthetic route in place, the optimization of both modules of the inhibitors could be initiated in parallel. 7.3.2 Optimization of the Ribose Analogue

It was envisaged to introduce a number of different ribose analogues (Figure 7.5b). Different ring types had to be tested: saturated, aromatic, or heteroaromatic rings, all of which should be directly attached to the N(1) of the nucleobase (ligands of type 7). Furthermore, the inﬂuence of a connecting methylene group between N(1) Asn12

Leu15 Leu28 Phe185

His26 2.9

2.9

Phe32 3.1 2.9

3.0

Lys10 Asp141

Pro182 3.4 (a)

Val156 Tyr25

(b)

Figure 7.6 (a) MOLOC-generated molecular model of the exemplary inhibitor in the active site of E. coli IspE (PDB code: 1OJ4) [20]. (b) Superposition of the potential inhibitor and the natural substrate 5. Color code: S: yellow; C skeleton of the inhibitor: cyan.

7.3 Targeting the CDP-Binding Pocket of IspE

of cytosine and the ribose analogue needed to be evaluated both for aliphatic and (hetero-)aromatic substituents (compounds of type 8). In addition, it was deemed important to test an acyclic ribose mimic, featuring an ester group. Replacement of the sulfur atom of the tetrahydrothiophenyl ring of the exemplary inhibitor by a methylene moiety to afford a cyclopentyl ring as a ribose analogue was expected to yield information on the postulated sulfur–aromatic interaction. Synthesis of this set of potential inhibitors was achieved using similar synthetic routes [24, 29]. Using the established photometric assay to determine the activity of potential inhibitors against IspE, the IC50 and Ki values for all new ligands were determined (Table 7.2) [24, 29, 31]. A competitive mechanism with respect to substrate binding was assigned to most ligands. In a few cases, however, a mixed competitive (Kic )–uncompetitive (Kiu ) mode of inhibition was found. In agreement with the observation that small cytosine derivatives possess remarkably low water solubility [24], none of the inhibitors showed high water solubility, despite the low calculated partitioning coefﬁcient (clogP) values (Figure 7.5b). The results obtained are highly satisfactory: First, as no inhibitors of IspE had been described at the onset of this design cycle, obtaining the ﬁrst active compounds with Ki values in the upper-nanomolar range constitutes an important achievement. Second, inhibition was possible in the absence of a phosphate or phosphonate group. The ligands described constitute the ﬁrst example of potent, drug-like inhibitors of an enzyme of the nonmevalonate pathway. Furthermore, variation of the ribose mimic has a clear effect on afﬁnity, that is, structure–activity relationships (SARs) could be observed. A methylene-bridged tetrahydrofuranyl ring is the poorest ribose substitute with a double-digit micromolar Ki value. Introduction of a methylene-bridged aromatic ring (benzyl) or an open alkyl chain bearing an ester moiety as a ribose mimic clearly also has a negative effect on afﬁnity. However, it seems that a methylene-bridged ring is tolerated as long as it is not too bulky: the methylene-bridged cyclobutyl and pyrazolyl derivatives feature similar afﬁnities to that of the inhibitor bearing a cyclopentyl ring directly attached to N(1) of cytosine. On the basis of the afﬁnities, there does not seem to be any Inhibitory activities (E. coli IspE) of compounds of type 7 and 8.

Table 7.2

Ribose analogue

Kic (µM)a

Tetrahydrothiophenyl Cyclopentyl Cyclobutylmethyl CH2 –3-pyrazolyl Benzyl CH2 CO2 Et CH2 –2-tetrahydrofuranyl

0.29 ± 0.1 1.5 ± 0.2 1.5 ± 0.2 1.6 ± 0.1 3.7 ± 0.5b 4.2 ± 0.6c 32.3 ± 2.8

a The

IC50 values can be found in 24, 29, 30. inhibition: Kiu = 23.5 ± 7.1 µM. c Mixed inhibition: K = 21.6 ± 6.2 µM. iu b Mixed

177

178

7 Structure-Based Design on the Way to New Anti-Infectives

signiﬁcant difference between a heteroaromatic ring such as pyrazolyl and the alicyclic cyclopentyl ring. However, if clogP values are taken into account, the latter inhibitor is signiﬁcantly more lipophilic; hence, the resulting more favorable partitioning from the aqueous buffer into the less polar protein environment could partially account for its afﬁnity. This implies that heteroatoms afford some afﬁnity. Finally, the tetrahydrothiophenyl ring is the best ribose mimic, affording the lowest Ki value. Presumably, its improved afﬁnity could be ascribed to the favorable sulfur–aromatic interaction. This interaction was quantiﬁed when comparing the afﬁnities of the tetrahydrothiophenyl derivative with an inhibitor bearing the same sulfone substituent (cyclopropyl) and a cyclopentyl ring as a ribose analogue. The afﬁnity is lowered by a factor of ﬁve, corresponding to a free-enthalpy increase of G300 K ≈ 1 kcal mol−1 , in agreement with published data [27]. Both rings beneﬁt from stabilization by the pseudo-π sandwich. Thus, the higher afﬁnity could be explained by favorable interactions of the phenolic side chain and the sulfur atom of the sandwiched ring. 7.3.3 Importance of the Vector

Before optimizing the propargylic sulfonamide vector, its signiﬁcance has to be conﬁrmed. According to modeling, the sulfonamide group forms two H bonds to Lys10 and Asp141. To verify this prediction experimentally, four derivatives featuring modiﬁed vectors and established ribose analogues were designed and synthesized (Figure 7.5c): (i) The N-methylated derivative of the most potent inhibitor, featuring a cyclopropyl ring as sulfone substituent and a tetrahydrothiophenyl ring as ribose analogue, served to quantify the importance of the ionic H bond between the sulfonamide NH and the side chain of Asp141. (ii) Different propargyl amine derivatives [compounds of type (±)-9] as well as propargyl alcohol derivative 10 were envisaged to evaluate the effect of substituting the sulfonamide altogether. The set of target molecules were synthesized following similar routes [24]. Using the established enzymatic assay, the IC50 and Ki values of the potential inhibitors were determined [24, 29, 31]. N-methylation of the sulfonamide nitrogen atom clearly affected afﬁnity, illustrating the importance of the postulated ionic H bond. Modeling showed the inhibitor to be still accommodated in the active site without any repulsive interactions but with one less H bond. Comparison with the most potent inhibitor enabled the quantiﬁcation of the postulated ionic H bond, given that both ribose analogue and sulfone substituent remained unchanged. The inhibitory potency was reduced by a factor of nearly 10 upon N-methylation (Ki = 2.5 µM); thus, a contribution of up to G300 K = 1.3 kcal mol−1 to the overall binding free enthalpy could be ascribed to this H bond alone (Figure 7.7a). As expected, substitution of the sulfonamide moiety by heterocyclic amines led to a decrease in inhibitory potency. A piperidyl-substituted derivative maintains the highest afﬁnity (Ki = 4.7 µM). A possible explanation could be an ionic H bond between the side chain of Asp141 and the piperidinium residue (Figure 7.7b). According to modeling, the piperidinyl ring should be located at the entrance of the

7.3 Targeting the CDP-Binding Pocket of IspE Leu15 Asn12

Leu28 His26

Phe185

Leu28 3.2

Lys10

2.9

Phe32

His26 Phe185

Phe32 2.9

Asn12

Leu15

2.9

2.8

2.9

3.2

3.1 2.9

Lys10

Pro182 Pro182 (a)

Val156 Tyr25

Val156 Asp141

Asp141 (b)

Tyr25

Figure 7.7 MOLOC-generated molecular model of (a) the N-methylated and (b) the piperidinyl-substituted inhibitors in the active site of E. coli IspE (PDB code: 1OJ4) [20]. Reproduced with permission from the Royal Society of Chemistry.

small, hydrophobic pocket. In this way, a number of hydrophobic interactions with the residues lining this cavity are still possible. A pyrrolidinyl-substituted derivative was proposed to bind in a similar manner, displaying fewer hydrophobic contacts because of its smaller ring. As a result, a decrease in binding afﬁnity was observed (Ki = 11.8 µM). The alcohol 10 shows very weak afﬁnity for IspE; consequently, no Ki value could be determined. This observation could be explained by the lack of both a sulfonamide moiety and an alkyl substituent to beneﬁt from potential interactions in the small, hydrophobic cavity. In summary, important contributions of the sulfonamide moiety to the observed binding afﬁnity were clearly conﬁrmed through this set of derivatives. With the sulfonamide moiety validated as a good vector, the next logical step was therefore the optimization of the sulfone substituent R (Figure 7.5d). 7.3.4 Optimization of the Filling of the Small, Hydrophobic Pocket

Inspired by the ﬁnding that optimal volume occupancy can be an important contributor to afﬁnity [32], the small, hydrophobic pocket of IspE was carefully examined. For this purpose, a series of derivatives of type (±)-11 and (±)-12 were designed, differing only in the sulfone substituent as shown in Figure 7.5d. By keeping the scaffold, the vector and the ribose analogue constant, the effects of different substituents on afﬁnity should be directly comparable. The target molecules were readily synthesized [24]. The choice of substituents was guided by the predictions made by molecular modeling and a concept from conventional supramolecular chemistry – the ‘‘55% rule’’ – that was recently applied to enzymes for the ﬁrst time [33]. 7.3.4.1 The ‘‘55% Rule’’ Investigation of the optimal volume occupancy of the cavity space conﬁned by capsular synthetic receptors by Mecozzi and Rebek led to the ‘‘55% rule,’’ stating

179

180

7 Structure-Based Design on the Way to New Anti-Infectives

that the most stable inclusion complex forms if 55 ± 9% of the apolar space is occupied by the guest [34]. This concept holds true, from synthetic supramolecular chemistry [35]. Recently, Z¨urcher et al. applied it to the ﬁlling of a hydrophobic cavity in the active site of the antimalarial target plasmepsin II [33]. The van-der-Waals interactions in the cavity are not ideal at smaller packing coefﬁcients (PCs). At higher PCs, however, large entropic losses, resulting from a decrease in the mobility of the binding partners, counteract enthalpic gains. The small, hydrophobic pocket of E. coli IspE was estimated to be around 100 A˚ 3 by ﬁlling the pocket with a hydrocarbon network [34, 36]. Thus, the PCs were calculated for different target molecules, which guided their design. 7.3.4.2 Evaluation of Inhibitors Featuring Different Sulfone Substituents A series of potential inhibitors were subjected to the enzymatic assay. The inhibitory activities as well as the calculated PCs for selected compounds are summarized in Table 7.3. A number of derivatives prepared feature Ki values in the nanomolar range. In general, small alkyl chains with a maximum chain length of two carbon atoms or cyclic alkyl substituents up to a ring size of ﬁve are well suited to ﬁll the cavity; hence, it is presumably more ﬂexible on the sides than at its bottom. The three-membered ring seems to be ideal to ﬁll this lipophilic pocket. A 2,2,2-triﬂuoroethyl group almost affords the same afﬁnity as the most potent inhibitor (featuring a cyclopropyl ring)

Inhibitory activities (E. coli IspE) and PCs of compounds of type (±)-11 and (±)-12.

Table 7.3

Sulfone substituent

µM) Kic (µ

PC (%)

Cyclopropyl 2,2,2-Triﬂuoroethyl Isopropyl Cyclobutyl Ethyl Cyclopentyl 1,1,1-Triﬂuoromethyl sec-Butyl n-Hexyl Cyclohexyl Methyl n-Butyl n-Propyl Phenyl

0.29 ± 0.1 0.36 ± 0.1 0.52 ± 0.1 0.56 ± 0 0.64 ± 0.1 0.89 ± 0.1a 1.2 ± 0.3 1.8 ± 0.3 2.0 ± 0.3 2.5 ± 0.4b 2.6 ± 0.1 8.0 ± 0.1 8.2 ± 1.7c 16.3 ± 1.0

56 n.d. 62 69 45 83 n.d. n.d. 107 97 28 76 61 n.d.

inhibition: Kiu = 19.0 ± 9.2 µM. inhibition: Kiu = 67.7 ± 35 µM. c Mixed inhibition: K = 27.3 ± 11 µM. iu n.d. = not determined. a Mixed

b Mixed

7.3 Targeting the CDP-Binding Pocket of IspE

and is nearly twice as strong as the corresponding ethyl derivative (Figure 7.6a). The afﬁnity of isopropyl- and cyclobutyl-substituted ligands is very similar to that of the ethyl-substituted derivative; however, the alkyl residues seem to be slightly too large for optimal volume occupancy, leading to a decrease in afﬁnity. As the pocket is not properly ﬁlled by ligands with a smaller alkyl substituent (methyl), the binding afﬁnity is reduced. Evaluation of the set of derivatives in terms of their PCs showed binding afﬁnity to correlate with volume occupancy: a cyclopropyl ring has a PC of 56% and the lowest Ki value (Ki = 0.29 µM, Figure 7.8a). Lower (28% for methyl and 45% for ethyl substituents) or higher PCs (62% for isopropyl and 69% for cyclobutyl) are mirrored by weaker inhibition. A slight increase in size of the sulfone substituent – that is, extension of the ethyl substituent by one or two carbon atoms – led to a strong decrease in afﬁnity. According to modeling, the n-propyl substituent might still ﬁt into the cavity at the cost of adopting the energetically less favorable gauche conformation (PC 61%). The n-butyl substituent, however, cannot be accommodated by the pocket, even when contorted. Thus, the propargylic sulfonamide linker could equally well undergo a conformational change to direct this substituent out of the pocket into solvent-exposed space. Larger substituents (PC > 80%), such as cyclopentyl, cyclohexyl, or n-hexyl, were predicted to direct their alkyl substituents toward the opposite direction, that is, toward the solvent (Figure 7.8b). The increased lipophilicity (cf. clogP values) and the resulting more favorable partitioning to the less polar protein environment from the aqueous buffer could explain the increased binding afﬁnity. In summary, by exploring the small, hydrophobic pocket, the afﬁnity of the inhibitors was improved and another example for the application of the ‘‘55% rule’’ to an enzymatic context was provided.

Lys10 Phe185 Leu15 Asp141 (a)

Leu28

(b)

Figure 7.8 (a) van-der-Waals surfaces of the cyclopropyl ring of the most potent inhibitor of type (±)-11 and the protein in the small, hydrophobic pocket. (b) MOLOC-generated molecular model of the alkyl-substituted inhibitors in the active site of E. coli IspE

(PDB code: 1OJ4) [20]. Color code: C skeleton of the inhibitors: methyl, cyan; ethyl, magenta; n-propyl, green; n-hexyl, light pink.Reproduced with permission from the Royal Society of Chemistry.

181

182

7 Structure-Based Design on the Way to New Anti-Infectives

7.3.5 Summary of the First-Generation Inhibitors

In conclusion, addressing the CDP-binding pocket was rewarded with the discovery of the ﬁrst, potent, small-molecule inhibitors of IspE. The afﬁnity could be improved through rational modiﬁcations of the lead compound, giving access to an extensive set of SARs; for the time being, they provide the only indication that the proposed binding mode is correct. X-ray-crystallographic studies had to be performed to validate this hypothesis (Section 7.4).

7.4 X-ray Cocrystal Structure Analysis

Efforts were undertaken to obtain a cocrystal structure of the ﬁrst-generation inhibitors and IspE. The kinase is very sensitive to crystallization conditions; even traces of organic cosolvents are sufﬁcient to preclude crystal growth. Presumably, because of the lack of water solubility of the ﬁrst-generation inhibitors, no cocrystal structures could be obtained. Thus, a water-soluble derivative had to be designed (Section 7.4.1). Provided it resembled the structure of the inhibitors in hand, it could prove the binding mode of this class of compounds. 7.4.1 Design of Water-Soluble Inhibitors

Modeling as well as clogP values were used to guide the selection of promising target molecules (Figure 7.9). To improve the water solubility of the ﬁrst-generation inhibitors without a concomitant decrease in afﬁnity, the sulfone substituent or the sulfonamide vector (target molecules of type (±)-13) or the ribose analogue (ligands of type 14) could be modiﬁed. Given that modestly potent inhibitors could be cocrystallized (Section 7.2.2), the proposed series of target molecules looked promising as long as water solubility could be achieved. Two derivatives were obtained by varying NH2

O

Inhibitor (±)-13 14

R2

N R1

R1

R2

clogP

2-Tetrahydrothiophenyl Carboxylic-acid-based

Morpholinyl based Cyclopropyl

–0.49 to 0.9 –0.6 to 0.4

Figure 7.9 Target molecules to improve the water solubility of the ﬁrst-generation inhibitors.

7.4 X-ray Cocrystal Structure Analysis

the vector, whereas the remaining three resulted from a modiﬁcation of the ribose analogue. Introduction of a carboxylic acid or the corresponding ester functionality, a morpholinyl, or an oxetanyl substituent should afford the desired property. Oxetanes, in particular, were recently described to endow compounds with improved physicochemical properties, namely, affording enhanced solubility and decreased lipophilicity [37]. 7.4.2 Enzyme Assays of Inhibitors Designed to be Water Soluble

Among the six compounds speciﬁcally prepared to obtain water-soluble inhibitors, two (the morpholinyl-substituted sulfonamide (Ki = 13.1 µM) and the oxetanyl derivative (compound 15, Figure 7.10b, Ki = 28.7 µM)) do not require addition of dimethyl sulfoxide as a cosolvent to perform the enzyme assay [24]. While the former still requires ethanol as a cosolvent, the latter is water soluble, setting the stage for X-ray crystallographic studies (Section 7.4.3). Water solubility was achieved at the expense of potency: the afﬁnity was decreased by nearly two orders of magnitude, when replacing the tetrahydrothiophenyl ring of the most potent inhibitor by the oxetanyl substituent of inhibitor 15. Because of their poor afﬁnities, only the IC50 values of the carboxylic acid derivatives were determined (upper micromolar range). The weak inhibitory potency of the morpholinyl derivative lacking the sulfonamide moiety could be explained by the bad match between the electronics of the morpholinyl moiety and the small, hydrophobic subpocket. Thus, introduction of an oxygen atom to the piperidinyl derivative of type (±)-9 (Section 7.3.3), affording the corresponding morpholinyl-substituted ligand, resulted in a decrease in afﬁnity by almost a factor of nine. A similar decrease in afﬁnity was observed when changing the sulfone substituent from cyclohexyl (compound of type (±)-12) to morpholinyl. The poor afﬁnity of the carboxylic acid derivatives could be attributed to the less than optimal ribose mimics.

Lys145

Tyr31 O

2.9 3.2

His25 3.1 3.2

3.0

3.0 2.6

NH2

Asn11

2.3 Asp130

O

N O

(a)

N H

N

(b)

Figure 7.10 (a) Binding mode of inhibitor 15 in the CDP-binding pocket of active site B (PDB code: 2VF3) [24]. (b) Structure of compound 15.

CO2Et 15

O S

183

184

7 Structure-Based Design on the Way to New Anti-Infectives

7.4.3 Structural Analysis

The structure of the oxetanyl derivative 15 in complex with A. aeolicus IspE was determined to 2.2 A˚ resolution (PDB code: 2VF3) [24]. Two molecules are present in the asymmetric unit, featuring ‘‘active sites A and B.’’ Evidence for the proposed binding mode of inhibitor 15 was provided by the cocrystal structure. The water-soluble ligand indeed binds in the CDP-binding pocket. As a result of the differences in amino-acid sequences of A. aeolicus and E. coli IspE [30], the sulfonamide and nucleobase moieties form somewhat different H-bonding patterns compared to those predicted by modeling. In particular, two additional H bonds stabilize the cytosine moiety of ligand 15 (Figure 7.10a). The cytosine moiety is engaged in four H bonds to the side chain and backbone amide of His25 in active site B. The amino group forms an additional H bond to the backbone amide C=O of Lys145. The cytosine ring is sandwiched by the side chains of Tyr24 and Tyr175 (= Phe185 in E. coli IspE). A network of H bonds to the side chains of Asn11, Tyr31, and Asp130 stabilizes the sulfonamide moiety (Figure 7.10a). Although AMP was present in the crystallization conditions, the electron density observed in the glycine-rich loop of the ATP-binding pocket was incompatible with this compound. Modeling and reﬁnement of the density as diphosphate proved successful. The latter unit forms numerous H bonds to the backbone-amide NH groups of the glycine-rich loop, in analogy to the diphosphate moiety of AMP-PNP in the crystal structure of the E. coli enzyme complex [20]. With one exception (Gly95), all the backbone-amide NH moieties of glycine residues of the loop are involved in H-bonding. The residues of the glycine-rich loop, which are not glycine, however, do not participate in H-bonding interactions with diphosphate. This underlines the importance of the presence of glycine residues in phosphate-binding pockets. Presumably, glycine is ideally suited for the binding of phosphate groups in such loops as it can adopt the required conformation to wrap the loop around the bound phosphate group [38]. For future efforts aimed at the design of inhibitors for IspE from pathogens such as P. falciparum or M. tuberculosis, the ﬁnding that the cyclopropyl substituent of ligand 15 is accommodated in the hydrophobic pocket is of particular interest (Figure 7.10b). Given that the change from Phe185 (in E. coli IspE) to Tyr175 (in A. Aeolicus IspE) increases the hydrophilicity of the small pocket, the cyclopropyl ring was not expected to be located in this subpocket as it precludes solvation of the hydroxyl group of Tyr175. This is an important ﬁnding, given that the enzymes from P. falciparum and M. tuberculosis possess a conserved tyrosine residue at this position, and that – according to homology modeling – the phenolic hydroxyl group is directed into the small pocket. The propargylic sulfonamide vector of this class of inhibitors is therefore highly suitable for precisely addressing the phenolic hydroxyl group in future design cycles.

7.5 Conclusions and Outlook

7.4.4 Lessons Learnt from the Cocrystal Structure

By designing water-soluble inhibitors and solving the cocrystal structure of one ligand with A. aeolicus IspE, the suggested binding mode was validated. Thus, the ﬁrst design cycle was successfully completed. The inhibitor occupies the CDP-binding pocket and exhibits a Ki value in the lower micromolar activity range. This proof of concept opens the way for further modiﬁcation and optimization of the inhibitors aimed at the development of ligands with activity against IspE from medically important organisms. Thus, the cocrystal structure solved represents an important step on the way to anti-infectives with a novel mode of action.

7.5 Conclusions and Outlook 7.5.1 Conclusions

The active site of the kinase IspE features both highly polar and hydrophobic subpockets. Because the highly polar subpockets do not lend themselves well to structure-based drug design, the inhibitors were targeted to the lipophilic regions of the active site. Using structure-based design, the ﬁrst inhibitors of the enzyme were developed, which display drug-like properties and highly satisfactory potency. A number of conclusions can be drawn from the series of inhibitors: • Inhibition of IspE is possible whilst bypassing the highly polar phosphate-binding pocket in the absence of a phosphate moiety. Therefore, the ligands synthesized constitute the ﬁrst potent and drug-like inhibitors of an enzyme of the nonmevalonate pathway. • The inhibitors target only the CDP-binding pocket, potentially affording better selectivity than inhibitors designed to bind to the ATP-binding pocket. • The majority of the binding free enthalpy is derived from the cytosine scaffold and the propargylic sulfonamide vector. • A systematic variation of the vector’s substituent provided access to SARs, conﬁrming the proposed binding mode. In addition, these modiﬁcations gave rise to the second example of the ‘‘55% rule’’ for the optimal ﬁlling of cavities applied to an enzymatic context. • Preparation of a water-soluble derivative enabled crystallographic studies of an enzyme–inhibitor complex, validating the proposed binding mode. This cocrystal structure provided invaluable structural information, especially in view of ﬁne-tuning the ligands to obtain activity against the pathogenic enzymes. In conclusion, the new target IspE revealed itself to be ideally suited to structure-based inhibitor design. The inhibitors synthesized in the context of this project represent a successful application of this design strategy.

185

186

7 Structure-Based Design on the Way to New Anti-Infectives

7.5.2 Outlook

The design and synthesis of the ﬁrst-generation inhibitors of IspE opened up a number of future research avenues. First, as a result of the observed difference in amino acid sequences between the model system E. coli and the pathogenic enzymes – marked by the replacement of the phenylalanine residue lining the hydrophobic pocket by a tyrosine – the structure of the inhibitors will have to be ﬁne-tuned to address this structural difference. Second, the bisubstrate approach could be tried to obtain even more potent and selective inhibitors of IspE. In principle, it should be possible to decorate the established scaffold to occupy the different pockets of the active site. Third, it would be very rewarding to identify an attractive mimic in the quest for new cytosine analogues. As opposed to the other nucleobases, very few cytosine substitutes have been described to date. Finally, identiﬁcation of a generally applicable way of rendering promising inhibitors water soluble without having to undertake a time-consuming design cycle and synthesis of the new derivatives would be of great interest. In this way, attractive structures could be designed, synthesized, and assayed without worrying about their physicochemical properties such as solubility.

Acknowledgments

This work was carried out under the supervision of Prof. F. Diederich and in collaboration with the groups of Prof. A. Bacher and Prof. W. N. Hunter. Financial support by the Roche Research Foundation and the ETH Research Council is gratefully acknowledged.

List of Abbreviations

AMP-PNP ATP clogP CDP DMAPP GHMP HTS IC50 IPP IspE Ki ME

5 -adenyl-β, γ -amidotriphosphate adenosine 5 -triphosphate calculated partitioning coefﬁcient cytidine 5 -diphosphate dimethylallyl diphosphate galactose/homoserine/mevalonate/phosphomevalonate high-throughput screening concentration of inhibitor at which 50% maximum initial velocity is observed isopentenyl diphosphate 4-diphosphocytidyl-2C-methyl-d-erythritol kinase inhibition constant methyl-erythritol

References

PC PDB SAR

packing coefﬁcient protein data bank structure–activity relationship

References 1. http://www.cdc.gov/malaria/impact/

2.

3.

4.

5.

6.

7.

8.

index.htm (accessed 23 February 2009); http://www.cdc.gov/tb/WorldTBDay/ resources global.htm (accessed 23 February 2009). (a) Baird, J.K. (2005) Effectiveness of antimalarial drugs. N. Engl. J. Med., 352, 1565–1577; (b) Kritski, A.L., de Jesus, L.S.R., Andrade, M.K., Werneck-Barroso, E., Vieira, M.A., Haffner, A., and Riley, L.W. (1997) Retreatment tuberculosis cases. Factors associated with drug resistance and adverse outcomes. Chest, 111, 1162–1167; (c) Oll´e-Goig, J.E. (2006) The treatment of multi-drug resistant tuberculosis – a return to the pre-antibiotic area? Trop. Med. Int. Health, 11, 1625–1628. Haney, S.A., LaPan, P., Pan, J., and Zhang, J. (2006) High-content screening moves to the front of the line. Drug Discov. Today, 11, 889–894. Schneider, G. and B¨ohm, H.-J. (2002) Virtual screening and fast automated docking methods. Drug Discov. Today, 7, 64–70. Butler, M.S. (2004) The role of natural product chemistry in drug discovery. J. Nat. Prod., 67, 2141–2153. (a) Kubinyi, H. (1998) Structure-based drug design. Chim. Oggi, 16, 17–22; (b) Klebe, G. (2000) Recent developments in structure-based drug design. J. Mol. Med., 78, 269–281; (c) Van Dongen, M., Weigelt, J., Uppenberg, J., Schultz, J., and Wikstr¨om, M. (2002) Structure-based screening and design in drug discovery. Drug Discov. Today, 7, 471–478. Knowles, J. and Gromo, G. (2003) Target selection in drug discovery. Nat. Rev. Drug Discov., 2, 63–69. (a) Bloch, K. (1992) Sterol molecule: structure, biosynthesis, and function. Steroids, 57, 378–383; (b) Bochar, D.A., Friesen, J.A., Stauffacher, C.V.,

9.

10.

11.

12.

13.

and Rodwell V.W. (1999)Biosynthesis of mevalonic acid from Acetyl-CoA, in Comprehensive Natural Products Chemistry: Isoprenoids, Including Carotenoids and Steroids (ed. D. E.Cane), Elsevier Science Publishers, Amsterdam, pp. 15–44. (a) Rohmer, M., Knani, M., Simonin, P., Sutter, B., and Sahm, H. (1993) Isoprenoid biosynthesis in bacteria: a novel pathway for the early steps leading to isopentenyl diphosphate. Biochem. J., 295, 517–524; (b) Schwarz, M.K. (1994) Terpen-Biosynthese in Gingko biloba: ¨ Eine Uberraschende Geschichte. Dissertation, ETH Z¨urich; (c) Broers, ¨ S.T.J. (1994) Uber die Fr¨uhen Stufen der Biosynthese von Isoprenoiden in E. coli. Dissertation, ETH Z¨urich; (d) Eisenreich, W., Schwarz, M., Cartayrade, A., Arigoni, D., Zenk, M.H., and Bacher, A. (1998) The deoxyxylulose phosphate pathway of terpenoid biosynthesis in plants and microorganisms. Chem. Biol., 5, R221–R233. Ruzicka, L. (1953) The isoprene rule and the biogenesis of terpenic compounds. Experientia, 9, 357–367. Jomaa, H., Wiesner, J., Sanderbrand, S., Altincicek, B., Weidemeyer, C., Hintz, M., T¨urbachova, I., Eberl, M., Zeidler, J., Lichtenthaler, H.K., Soldati, D., and Beck, E. (1999) Inhibitors of the nonmevalonate pathway of isoprenoid biosynthesis as antimalarial drugs. Science, 285, 1573–1576. Hunter, W.N. (2007) The nonmevalonate pathway of isoprenoid precursor biosynthesis. J. Biol. Chem., 282, 21573–21577. Hirsch, A.K.H. and Diederich, F. (2008) The nonmevalonate pathway to isoprenoid biosynthesis: a potential

187

188

7 Structure-Based Design on the Way to New Anti-Infectives source of new drug targets. Chimia, 62, 226–230. 14. (a) Altincicek, B., Hintz, M., Sanderbrand, S., Wiesner, J., Beck, E., and Jomaa, H. (2000) Tools for discovery of inhibitors of the 1-Deoxy-d-xylulose 5-Phosphate (DXP) synthase and DXP reductoisomerase: an approach with enzymes from the pathogenic bacterium Pseudomonas aeruginosa. FEMS Microbiol. Lett., 190, 329–333; (b) Mueller, C., Schwender, J., Zeidler, J., and Lichtenthaler, H.K. (2000) Properties and inhibition of the ﬁrst two enzymes of the nonmevalonate pathway of isoprenoid biosynthesis. Biochem. Soc. Trans., 28, 792–793. 15. (a) Kuzuyama, T., Shimizu, T., Takahashi, S., and Seto, H. (1998) Fosmidomycin, a speciﬁc inhibitor of 1-Deoxy-d-xylulose 5-phosphate reductoisomerase in the nonmevalonate pathway for terpenoid biosynthesis. Tetrahedron Lett., 39, 7913–7916; (b) Herforth, C., Wiesner, J., Heidler, P., Sanderbrand, S., Van Calenbergh, S., Jomaa, H., and Link, A. (2004) Antimalarial activity of N6-substituted adenosine derivatives. Part 3. Bioorg. Med. Chem., 12, 755–762; (c) Walker, J.R. and Poulter, C.D. (2005) Synthesis and evaluation of 1-Deoxy-d-xylulose 5-phosphate analogues as chelation-based inhibitors of methylerythritol phosphate synthase. J. Org. Chem., 70, 9955–9959; (d) Woo, Y.H., Fernandes, R.P.M., and Proteau, P.J. (2006) Evaluation of fosmidomycin analogs as inhibitors of the Synechocystis sp PCC6803 1-Deoxy-d-xylulose 5-phosphate reductoisomerase. Bioorg. Med. Chem., 14, 2375–2385; (e) Haemers, T., Wiesner, J., Van Poecke, S., Goeman, J., Henschker, D., Beck, E., Jomaa, H., and Van Calenbergh, S. (2006) Synthesis of α-substituted fosmidomycin analogues as highly potent Plasmodium falciparum growth inhibitors. Bioorg. Med. Chem. Lett., 16, 1888–1891; (f) Devreux, V., Wiesner, J., Goeman, J.L., Van der Eycken, J., Jomaa, H., and Van Calenbergh, S. (2006) Synthesis

and biological evaluation of cyclopropyl analogues of fosmidomycin as potent Plasmodium falciparum growth inhibitors. J. Med. Chem., 49, 2656–2660. 16. (a) Crane, C.M., Kaiser, J., Ramsden, N., Rohdich, F., Eisenreich, W., Hunter, W.N., Bacher, A., and Diederich, F. (2006) Fluorescent inhibitors for IspF, an enzyme in the nonmevalonate pathway for isoprenoid biosynthesis and a potential target for antimalarial therapy. Angew. Chem., 118, 1083–1087; Angew. Chem. Int. Ed., 45, 1069–1074; (b) Baumgartner, C., Eberle, C., Diederich, F., Lauw, S., Rohdich, F., Eisenreich, W., and Bacher, A. (2007) Structure-based design and synthesis of the ﬁrst weak non-phosphate inhibitors for IspF, an enzyme in the nonmevalonate pathway of isoprenoid biosynthesis. Helv. Chim. Acta, 90, 1043–1068. 17. (a) Reardon, J.E. and Abeles, R.H. (1986) Mechanism of action of isopentenyl pyrophosphate isomerase: evidence for a carbonium-ion intermediate. Biochemistry, 25, 5609–5616; (b) Muehlbacher, M. and Poulter, C.D. (1988) Isopentenyl-diphosphate isomerase: inactivation of the enzyme with active-site-directed irreversible inhibitors and transition-state analogues. Biochemistry, 27, 7315–7328; (c) Poulter, C.D., Muehlbacher, M., and Davis, D.R. (1989) Isopentenyldiphosphate isomerase. Mechanism of active-site-directed irreversible inhibition by 3-(ﬂuoromethyl)-3-butenyl diphosphate. J. Am. Chem. Soc., 111, 3740–3742; (d) Lu, X.J., Christensen, D.J., and Poulter, C.D. (1992) Isopentenyl-diphosphate isomerase: irreversible inhibition by 3-methyl-3,4-epoxybutyl diphosphate. Biochemistry, 31, 9955–9960. 18. (a) Kuzuyama, T., Takagi, M., Kaneda, K., Watanabe, H., Dairi, T., and Seto, H. (2000) Studies on the nonmevalonate pathway: conversion of 4-(cytidine 5 -diphospho)-2C-methyl-d-erythritol to its 2-phospho derivative by 4-(cytidine 5 -diphospho)-2C-methyl-d-erythritol Kinase. Tetrahedron Lett., 41, 2925–2928;

References

19.

20.

21.

22.

23.

(b) L¨uttgen, H., Rohdich, F., Herz, S., Wungsintaweekul, J., Hecht, S., Schuhr, C.A., Fellermeier, M., Sagner, S., Zenk, M.H., Bacher, A., and Eisenreich, W. (2000) Biosynthesis of terpenoids: YchB protein of Escherichia coli phosphorylates the 2-hydroxy group of 4-diphosphocytidyl-2C-methyl-derythritol. Proc. Natl. Acad. Sci. U.S.A., 97, 1062–1067. Bork, P., Sander, C., and Valencia, A. (1993) Convergent evolution of similar enzymatic function on different protein folds: the hexokinase, ribokinase, and galactokinase families of sugar kinases. Protein Sci., 2, 31–40. Miallau, L., Alphey, M.S., Kemp, L.E., Leonard, G.A., McSweeney, S.M., Hecht, S., Bacher, A., Eisenreich, W., Rohdich, F., and Hunter, W.N. (2003) Biosynthesis of isoprenoids: crystal structure of 4-diphosphocytidyl-2C-methyl-derythritol kinase. Proc. Natl. Acad. Sci. U.S.A., 100, 9173–9178. Wada, T., Kuzuyama, T., Satoh, S., Kuramitsu, S., Yokoyama, S., Unzai, S., Tame, J.R.H., and Park, S.Y. (2003) Crystal structure of 4-(cytidine 5’-diphospho)-2C-methyl-d-erythritol kinase, an enzyme in the nonmevalonate pathway of isoprenoid synthesis. J. Biol. Chem., 278, 30022–30027. Sgraja, T., Alphey, M.S., Ghilagaber, S., Marquez, R., Robertson, M.N., Hemmings, J.L., Lauw, S., Rohdich, F., Bacher, A., Eisenreich, W., Illarionova, V., and Hunter, W.N. (2008) Characterization of Aquifex aeolicus 4-diphosphocytidyl-2C-methyl-derythritol kinase – ligand recognition in a template for antimicrobial drug discovery. FEBS J., 275, 2779–2794. Crane, C.M., Hirsch, A.K.H., Alphey, M.S., Sgraja, T., Lauw, S., Illarionova, V., Rohdich, F., Eisenreich, W., Hunter, W.N., Bacher, A., and Diederich, F. (2008) Synthesis and characterization of cytidine derivatives that inhibit the kinase IspE of the nonmevalonate pathway for isoprenoid biosynthesis. ChemMedChem, 3, 91–101.

24. Hirsch, A.K.H., Alphey, M.S., Lauw,

25.

26.

27.

28.

29.

30.

31.

32.

S., Seet, M., Barandun, L., Rohdich, F., Hunter, W.N., Bacher, A., and Diederich, F. (2008) Inhibitors of the kinase IspE: structure–activity relationships and co-crystal structure analysis. Org. Biomol. Chem., 6, 2719–2730. Gerber, P.R. and M¨uller, K. (1995) MAB, a generally applicable molecular-force ﬁeld for structure modeling in medicinal chemistry. J. Comput.-Aided Mol. Des., 9, 251–268. ACD/LogP & ACD/LogD, Advanced Chemistry Development, Inc. (2005)Toronto. www.acdlabs.com. (a) Meyer, E.A., Castellano, R., and Diederich, F. (2003) Interactions with aromatic rings in chemical and biological recognition. Angew. Chem., 115, 1244–1287; Angew. Chem. Int. Ed., 42, 1210–1250. Brameld, K.A., Kuhn, B., Reuter, D.C., and Stahl, M. (2008) Small molecule conformational preferences derived from crystal structure data. A medicinal chemistry focused analysis. J. Chem. Inf. Model., 48, 1–24. Hirsch, A.K.H., Lauw, S., Gersbach, P., Schweizer, W.B., Rohdich, F., Eisenreich, W., Bacher, A., and Diederich, F. (2007) Nonphosphate inhibitors of IspE protein, a kinase in the nonmevalonate pathway for isoprenoid biosynthesis and a potential target for antimalarial therapy. ChemMedChem, 2, 806–810. Hirsch, A.K.H. (2008) A novel approach towards antimalarials: design and synthesis of inhibitors of the kinase IspE . Dissertation, ETH Z¨urich. Illarionova, V., Kaiser, J., Ostrozhenkova, E., Bacher, A., Fischer, M., Eisenreich, W., and Rohdich, F. (2006) Nonmevalonate terpene biosynthesis enzymes as antiinfective drug targets: substrate synthesis and high-throughput screening methods. J. Org. Chem., 71, 8824–8834. Z¨urcher, M. and Diederich, F. (2008) Structure-based drug design: exploring the proper ﬁlling of apolar pockets at enzyme active sites. J. Org. Chem., 73, 4345–4361.

189

190

7 Structure-Based Design on the Way to New Anti-Infectives 33. Z¨ urcher, M., Gottschalk, T., Meyer,

S., Bur, D., and Diederich, F. (2008) Exploring the ﬂap pocket of the antimalarial target plasmepsin II: the ‘‘55% Rule’’ applied to enzymes. ChemMedChem, 3, 237–240. 34. Mecozzi, S. and Rebek, J.Jr. (1998) The 55% solution: a formula for molecular recognition in the liquid state. Chem. Eur. J., 4, 1016–1022. 35. (a) Purse, B.W. and Rebek, J.Jr. (2006) Self-fulﬁlling cavitands: packing alkyl chains into small spaces. Proc. Natl. Acad. Sci. U.S.A., 103, 2530–2534; (b) Hooley, R.J. and Rebek, J.Jr. (2007) Self-complexed deep cavitands: alkyl chains coil into a nearby cavity. Org. Lett., 9, 1179–1182; (c) Gottschalk, T., Jaun, B., and Diederich, F. (2007) Container molecules with portals:

reversibly switchable cycloalkane complexation. Angew. Chem., 119, 264–268; Angew. Chem. Int. Ed., 46, 260–264. 36. (2006) Spartan’06, Wavefunction Inc., Irvine . 37. (a) Wuitschik, G., Rogers-Evans, M., M¨uller, K., Fischer, H., Wagner, B., Schuler, F., Polonchuk, L., and Carreira, E.M. (2006) Oxetanes as promising modules in drug discovery. Angew. Chem., 118, 7900–7903; (b) Angew. Chem. Int. Ed., 45, 7736–7739. 38. (a) Hirsch, A.K.H., Fischer, F.R., and Diederich, F. (2007) Phosphate recognition in structural biology. Angew. Chem., 119, 342–357; Angew. Chem. Int. Ed., 46, 338–352.

191

8 Drug–Membrane Interactions: Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs Marlene L´ucio, Jos´e L. F. C. Lima, and Salette Reis

8.1 Biological Membranes 8.1.1 Role of Membranes in Life Maintenance

Membranes are of fundamental importance for biological systems [1]. They provide for cellular and subcellular organelles’ compartmentalization and maintain the essential differences between the cellular and extracellular environments. The external membrane (called the plasma membrane) protects the cell from the loss of important molecules and the penetration of undesired substances from outside. This control of the internal cell environment provides for a regulation of the concentration of electrolytes and the maintenance of the osmotic pressure, which are part of the cellular homeostasis. Internal membranes also perform important functions separating incompatible biochemical reactions, maintaining intracellular transport, and playing a major role in energy transfer and storage [2]. Continuous and reliable communication between compartments (intracellular) and between cells (extracellular) is therefore required, involving processes that occur at the surface of membranes or are mediated by membranes. Membranes also serve as a matrix for the spatial organization of proteins among which are the enzymes and their cofactors, holding otherwise scattered molecules in functional contact [3]. 8.1.2 Structure and Composition of Membranes

In the past two decades there has been considerable advance in the knowledge of membrane structure. According to the Fluid Mosaic model [4], biological membranes are based on a ﬂuid lipid bilayer, in which the lipids (and many proteins) show a rapid lateral diffusion. Such rapid diffusion implies that lipids and proteins are more or less randomly distributed in the plane of the membrane with a nonspeciﬁc order. However, this rather simple picture of ‘‘proteins ﬂoating in a sea of lipids’’ Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

192

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

can hardly match either the heterogeneous composition of the membrane or the membrane molecular organization. Indeed, more recent analysis of the bilayer structure suggests that the composition of biological membranes is not uniform, but shows a heterogeneous distribution of components, both laterally and across the bilayer [5, 6]. Cell membranes, nowadays, are considered to consist of solid domains coexisting with ﬂuid membrane lipids that may also contain proteins [7–10]. The most important difference encountered in the actual membrane model from the ﬂuid mosaic model is that a high degree of spatiotemporal order also prevails in the ﬂuid membrane and in membrane domains [11], and this order seems to be essential for the functioning of lipid-embedded and integrated proteins [12, 13]. Therefore, the earlier concept of a lipid matrix, which was merely a support for embedded proteins, has evolved, and evidence is accumulating that lipid bilayer structure and dynamics play a crucial role in membrane functionality. According to this, it is worthwhile to describe in more detail, the lipid composition of biomembranes. Most of the membrane common lipids belong to three classes: phospholipids, glycolipids (lipid–sugar conjugates), and sterols (e.g., cholesterol). On average, a biological membrane contains about 40–70% of phospholipids, 0–50% of cholesterol, and 0–26% glycolipids [2]. The major components of the membranes are thus the phospholipids, which are amphiphilic molecules constituted of four components: one or two fatty acids designated as the hydrophobic tails, a glycerol backbone (or sphingosine in the case of the sphingolipids), a phosphate, and an alcohol attached to the phosphate. The phosphate and alcohol together form a polar head group. The hydrophobic tails of the phospholipids can differ in their length or in the number of unsaturated bonds. Because of the fatty acids synthesis cycles in the metabolism, practically all of the naturally occurring saturated fatty acids in natural membranes are palmitic (C16), stearic (C18), and myristic (C14). From the unsaturated fatty acids, the oleic (C18:1), which is a stearic acid with one double bond in the middle of the chain, is the most important. Most natural lipids contain one or more double bonds in cis conformation (e.g., higher animals, plants) while a trans conformation is found in bacteria. The level of saturation is important in controlling the ﬂuidity of the membranes by lowering the phase transition temperature, as well as the hydration and mobility of the bilayer. Unsaturated fatty acids are therefore important for the proper functioning of biomembranes, and actual nutritional recommendations mirror this knowledge. The heterogeneity of lipids is also reﬂected in their polar heads. The most frequently occurring alcohol moieties are the amino acids serine, ethanolamine, choline, glycerol, and inositol [2]. The polar headgroups can substantially alter the function of a membrane since this is the part of the lipid that is present on the membrane surface, and is often responsible for the interaction with the surrounding. The polar groups of phospholipids are usually neutral or negatively charged. A positive charge is rare (cationic lipids are, however, important in for example, gene delivery, in which they form complexes with nucleic acids; the so-called lipoplexes). The electric charges of membrane lipids usually fall into a range of +1 to −[2]. Most of the neutral lipids are zwitterionic, characterized by strong dipole moments.

8.1 Biological Membranes

The electrostatics of lipid membranes is a complex issue, but strongly affects membrane–protein interactions, domain formation, as well as other membrane functions. In conclusion, membranes are complex supramolecular liquid-crystalline assemblies mainly composed of phospholipids. This complexity also arises from the fact that in most membranes different regions and domains with deﬁned lipid and protein compositions usually coexist and similar domains are not always entirely equal [14]. Finally, membrane proteins and lipids may both be subjected to regulatory processes in response to pathophysiological situations or nutritional pharmacological interventions, which in turn may alter the activity and functions of the membrane [14]. Hence, membranes gather lipids and proteins, both fulﬁlling prominent roles in certain cellular processes, equally relevant in many cases. Therefore, drug–membrane interactions represent a wide and complex ﬁeld in medicinal chemistry and can involve studies of drugs that are capable of inﬂuencing the protein function both directly and indirectly via lipid modulation. The protein regulation achieved by drug–membrane interactions can ﬁnally induce changes in cell signaling and gene expression, which might serve to reverse the pathological state. The literature on drug–membrane interactions is vast and multifaceted, rendering it a difﬁcult task to elaborate a comprehensive review on the subject. In view of this, the present chapter will be restricted to examples of applications of drug–membrane interaction studies focusing on the inﬂuence of drugs on the lipid components. 8.1.3 Dynamic Molecular Organization of Membranes

A characteristic feature of the amphiphilic molecules in general, and lipids in particular, are their ability to spontaneously self-aggregate into organized and functional structures [15, 16]. In an aqueous environment, the lipid molecules conceal their hydrophobic tails constituting an interior shielded by their polar headgroups, whose surface interacts favorably with water. The driving force for lipid self-assembly is the hydrophobic effect [17, 18]. The molecular aggregation process, is itself a complex phenomenon, which still lacks experimental studies in atomic detail. The entire process has been however demonstrated using molecular dynamic simulations [19, 20]. As a starting point, from a random solution of phospholipids in water, a perfect bilayer forms spontaneously. The bilayer, a lamellar phase, is one of the several different possible structures into which lipids can self-organize. The diversity of the most common aggregation forms is shown in Figure 8.1. The extraordinarily rich polymorphism of lipid aggregates results from a subtle balance of attractive and repulsive interactions between these molecules at the water/lipid interface [21]. This balance is strongly affected by external conditions and of the lipid themselves. The major external factors determining the aggregation state of lipids are the temperature and the amount and the nature of the solvent. But other factors, such as the pressure, pH, and ion concentration also

193

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs Increasing temperature

Increasing temperature

Lc

Increasing hydration

194

Lb

Pb ′ QI

HI

HII

La

M

Single chain Short double-chain lipids

Figure 8.1 Schematic dependence of phase transitions of lipid–water mixtures on hydration, temperature, and also on length of the alkyl tail of a phospholipid. Lc corresponds to the lamellar crystalline phase; Lβ

Long double-chain lipids

to the lamellar gel phase; Lα to the lamellar liquid-crystalline phase; HI to the hexagonal phase; HII to the inverted hexagonal phase; M to the micellar phase; Pβ to the ripple phase; and QII to the inverted cubic phase.

have a signiﬁcant inﬂuence. Geometrical properties that affect the phase behavior are the size of the lipid headgroups, the length and saturation of lipid tails. In Figure 8.1, a schematic representation of the phase transitions induced by temperature, hydration, and lipid type is presented. At physiological conditions, the most common and the most biologically relevant lamellar phase is the ﬂuid phase also named as liquid-crystalline phase (Lα ). In this phase, the rotational freedom of a single C–C bond results in the vigorous thermal movement of fatty acid chains of phospholipids ensuring the ﬂuidity of the entire bilayer. Upon lowering the temperature, the bilayer undergoes a number of phase transitions ending in the lamellar crystalline phase (Lc ). The main phase transition from the ﬂuid to the ordered phase occurs at a melting temperature (Tm ). Here, the tails become almost fully ordered and pack into a hexagonal lattice, forming a gel phase (Lβ ). Some disorder is still present, distinguishing the gel phase from the crystalline phase. For some lipids (dimyristoylphosphatidylcholine, DMPC; dipalmitoylphosphatidylcholine, DPPC), upon cooling the liquid-crystalline phase (Lα ) state to a temperature below the Tm , the formation of an intermediate gel phase, called a ripple phase (Pβ ) is observed. Decreasing the temperature further, a tilted gel phase (Lβ ) is formed.

8.2 Drug–Membrane Interactions

As mentioned above, the transition from a lamellar to a nonlamellar phase can be triggered both by temperature changes (thermotropic phase transition) and by (de)hydration (lyotropic phase transition). Upon increased hydration, the lamellar structure ruptures and long wormlike aggregates such as in the hexagonal HI phase, or small micelle aggregates, M form. On the other hand, dehydration (or temperature rise) can trigger the formation of inverted aggregates such as the HII or cubic phases QII . The described phase transitions occurring within a membrane can be considered as integrative phenomena of the dynamic molecular organization of membranes, which can be divided in conformational movements (related with the intramolecular movements and the long range order of lipid packing) and the translational movements (which indicate the lateral position of the molecule in the plane of the bilayer and are related with the short-range order of lipid packing). The common used term membrane ﬂuidity can thus be considered to contain both conformational (microviscosity) and translational (lateral diffusion) dynamics [22].

8.2 Drug–Membrane Interactions 8.2.1 Possible Effects of Drugs on Membranes

It was previously mentioned that cell membranes, composed of lipids and proteins, function as a permeability barrier, maintain ion gradients across the membrane and steady state of ﬂuxes, and possess recognition sites for communication and interaction with other cells. Disturbance of such a complex system by the uptake of exogenous compounds, like drugs and lipids, can cause several effects in membrane structure and dynamics, which in turn, are responsible for severe changes in the performance of the cell, including the function of transmembrane receptor proteins and proteins responsible for signal transduction. The degree of membrane perturbation will depend on both the structure and physicochemical properties of the exogenous molecules and membrane lipids involved, as well as the way of supplying the molecules to the membrane system. The effects on membranes that can arise from drug–membrane interactions are summarized in Figure 8.2. 8.2.2 Clinical Relevance of the Drug–Membrane Interaction Studies 8.2.2.1 Contribution for Drug Development Understanding the effects of drugs on membranes is of great importance for current and future pharmaceutical research. Drug discovery and development has typically considered interactions between the ligand and the molecular target (e.g.,

195

196

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

Conformation Location

CD

Fluorescence XD

Membrane / water partition

Drug

Derivative spectroscopy Drug–membrane Mode of action toxicity Membrane

Phase transitions

Structure AFM, BAM, SAXS WAXS, FFEM, XD

Surface charge Zeta-potential

Figure 8.2 Schematic representation of the analytical tools available for the evaluation of possible effects of drugs on membranes (effects on membrane structure, surface charge, dynamic properties, and

Dynamic properties

DSC

NMR, Fluorescence

phase transition) and effects of the membrane on drug molecules (effects on drug partition between aqueous and lipid phase, drug location and orientation, and drug conformation).

protein receptor), with the assumption that the membrane bilayer acts as a passive surrounding environment. In recent years, however, it has become apparent that membrane constituents, particularly phospholipids, interact with drugs and other hydrophobic chemicals [1, 23–25]. Partitioning of drugs into lipid bilayers of biological membranes plays a signiﬁcant role in their uptake, transport, bioavailability, and distribution [24–26]. As most drugs are administered orally, their ability to transport across the intestinal epithelium, a monolayer of cells that line the interior of the intestine, is an important issue. For efﬁcient absorption from the gastrointestinal tract, a compound has to cross the plasma membrane of enterocytes lining the gut lumen. The majority of the compounds cross the plasma membrane by passive diffusion through the lipid bilayer [27]. The main determinants for absorption are therefore drug–lipid interactions. After the drug has reached the circulation it will redistribute in the body. The distribution can be limited due to some blood–tissue barriers, such as the blood–brain barrier [28]. The blood–brain barrier is formed by endothelial cells sealed by tight junctions in brain capillaries, and several proteins are involved in the formation of tight junctions [28]. The blood–brain barrier is also metabolically very active and contains large amount of transport proteins [28]. However, most of the compounds going through these barriers must

8.2 Drug–Membrane Interactions

cross the lipid bilayer of the endothelial cells [28]. Thus, drug–lipid interactions also play a crucial role in the distribution of compounds across blood–tissue barriers. Furthermore, if the drug target is intracellular, the permeation through the plasma membrane of the target cell is of outmost importance. As different cells can have different plasma membrane compositions and physicochemical properties, a thorough knowledge of drug–lipid interactions would help the development of optimal properties for drug candidates [29–32]. 8.2.2.2 Understanding Therapeutic and Toxic Effect of Drugs In order to understand the therapeutic action of any drug it is pivotal to know the dynamics of interaction with the lipid membrane and its crossing rate. This information is particularly useful when the lipid membrane represents the actual target for the drug and, in this case, the studies of the drug–membrane interaction are obviously fundamental to understand the drugs mode of action and consequently their therapeutic effect. In this context, earlier studies on neuroactive compounds such as anesthetics, tranquillizers, antidepressants, and so on, have suggested a strong involvement of a nonspeciﬁc impairing action of these compounds on the lipid bilayer in their neurological effects. The major cause of the sedative side effect of the antidepressant imipramine has been attributed to its ability to penetrate membrane structures [33, 34]. Studies on anesthetics like halothane and enﬂurane have proved that the preferred position of interaction of inhalation anesthetics with the lipid bilayer is the membrane interface [35]. Besides drugs, other neuroactive compounds include pollutants that should also be investigated for both their environmental and health toxicity. Popular insecticides like parathion, malathion, lindane, dichlorodiphenyl trichloroethane (DDT), and allethrin induce perturbations of membrane permeability and ﬂuidize the membrane and the changes could be partially related to their primary insecticidal activity [36–40]. Also, a number of other drugs have been described to alter lipid order and domain formation. Examples include cyclosporine [41], steroids [42], triﬂuoroperazine [43], and antibiotics like aminoglycosides [44, 45] or macrolides (azithromycin) [46–53]. This latter compound has also shown to markedly inhibit endocytosis [52, 53], probably by interacting with lipids, modifying biophysical properties of membrane and affecting membrane’s dynamics in living cells [46, 47]. The membrane interactions, as a part of the mechanism of action, is further shown for amphiphilic drugs like cisplatin [54–59], amphotericin B [60, 61], and antimicrobial peptides such as magainin, cecropin, and defensin [62–68], for which no protein receptors have been identiﬁed. Knowledge about drug–lipid interactions might contribute to a better understanding of the drug activity but might also shed light on the severe side effects of the drug, of which, the effects on membrane biophysics are pointed out as one of the most important causes. Examples of the correlation between the effect of drugs on membrane biophysics and their toxicity are widely described in the literature [69–83].

197

198

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

8.2.2.3 Understanding Mechanisms of Multidrug Resistance Multidrug resistance (MDR) is a critical problem in cancer chemotherapy. Overexpression of P-glycoprotein (PgP) on the plasma membrane is frequently observed in drug-resistant cancer cells leading to the failure of chemotherapy [84–86]. PgP is an integral membrane protein that also modulates drug absorption, bioavailability, tissue deposition, and excretion. Therefore, not only anticancer drugs, but also several other drugs are known to be efﬁciently pumped out from the cell by PgP, which thus plays an important role in compromising AIDS chemotherapy and other drug-treatable diseases [87–89]. A common conclusion emerging from the studies performed on MDR modulation was that the interaction of drugs with the lipid bilayer was a primary determinant for substrate recognition of PgP. Consequently, substrate binding to PgP should be considered in two steps, the initial partitioning of drugs to the lipid bilayer followed by binding to transmembrane region of the enzyme [86, 90–93]. Despite this knowledge, conventionally, drug–lipid interactions are neglected, either as a possible explanation for increased intracellular concentration of the substrate in studies of PgP function, or as a way of ﬁnding inhibitors of PgP. Indeed, besides the interaction of MDR modulators with transporter proteins, their interaction with plasma membrane lipids may also contribute to reverse the molecular mechanisms of MDR. According to this, a drug that is able to induce alterations of bilayer properties, especially altering ﬂuidity and permeability, should play an essential role in the processes of MDR modulation. Hence, most of the MDR reversing agents are preferentially soluble in lipids and they may also exert an inﬂuence on the physical properties of lipid bilayers [94–96]. For example, phenothiazines, considered as MDR reversing agents affect the properties of the lipid bilayer causing phase separation in membranes containing lipid microdomains of sphingolipids and cholesterol [95]. These induced changes in the lipid environment of membrane transporters like PgP can affect either drug export or import [95]. In conclusion, perturbation of lipid phase of the cell membrane by modulators seems to be important for potentiating the anticancer drug therapy and other therapies, whose efﬁcacy is affected by MDR mechanisms [95]. 8.2.2.4 Controlling Enzymatic Inhibition A growing body of evidence has shown that lipid–bilayer structure and dynamics play a key role in membrane functionality [97]. Hence, the lipid disturbance by the uptake of drugs, can lead to changes in many membrane properties, which then can induce severe alterations in the performance of the enzymes. In fact, binding of drugs to membrane lipids can lead to alterations in the function of proteins, as shown for phospholipase A2 [98], 5-lipoxygenase [99], cytochrome c oxidase [100], lysosomal phospholipases, phosphatidylinositol-speciﬁc phospholipase C, sphingomyelinase [70], and protein kinase C [101]. For instance, it has been reported that the inhibitory effect of the mycotoxin Fumonisin B1 on ceramide synthase activity might not be a typical enzyme–inhibitor type direct interaction, but a consequence of mycotoxin-induced changes in the dynamic organization of the lipid phase [83].

8.3 Analysis and Quantiﬁcation of Drug–Membrane Interactions

Consequently, in addition to speciﬁc interactions between drugs and enzymes, drugs that act as enzymatic inhibitors are also likely to modify the bulk physical properties of the membrane. The mechanism of these effects on membrane biophysics is an area of current interest, and therefore justiﬁes the increasing interest to study drug–membrane interactions [102]. 8.3 Analysis and Quantiﬁcation of Drug–Membrane Interactions 8.3.1 Membrane Model Systems

Membranes of cell surfaces have distinct lateral structures [10, 103] whose molecular organization and dynamics are essential for their interaction with speciﬁc ligands and proteins [104, 105]. Since plasma membranes are too complex to be characterized in simple physical–chemical terms, the quantitative study of such questions proceeds most conveniently by means of model systems [103]. Therefore, to understand the basic functions of membranes, or mechanisms by which they undergo speciﬁc transitions, simple in vitro models typically consisting of only one or two components are used. A lipid bilayer membrane can be artiﬁcially reconstituted in various forms from puriﬁed or synthesized lipid molecules. Indeed, as previously mentioned in Section 8.1.3, in an aqueous environment, phospholipids spontaneously self-assemble into bilayer membranes through hydrophobic interactions between their acyl chains [106]. On the basis of the aggregation properties of lipids in aqueous media, several membrane models were developed for studying drug–membrane interactions namely liposomes, planar bilayers, and monolayers [104]. Liposomes are a good model for biomembranes and have been widely used for studying drug delivery and interaction of drugs with the phospholipid bilayer [107]. Liposomes are spherical vesicles consisting either of a single lipid bilayer or multiple lipid bilayers that enclose an internal aqueous volume [108]. They can be obtained easily by hydrating dry thin ﬁlms of phospholipids. These vesicles form spontaneously as a result of the tendency to minimize the energetically unfavorable interactions of bilayer edges with water [109]. Liposomes can vary widely in size, ranging from few nanometers referred to as small unilamellar vesicles (SUVs) to the micrometer size of real cells also referred to as giant unilamellar vesicles (GUVs). Spontaneously formed liposomes are usually multilamellar and inhomogeneous in shape. However, various techniques can be used to reduce the size of vesicles and the number of bilayers (e.g., by ultrasound, so-called sonication and extrusion through polycarbonate membranes [108]), in order to obtain large unilamellar vesicles (LUVs). Other important models used for studying the structural properties of membranes are the monolayers. In fact, the use of monolayers of phospholipids as a membrane model has regained interest recently, since it is half of a lipid bilayer and thus constitutes a very convenient ﬁrst step to approach the two-dimensional

199

200

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

structure of biomembranes [110–117]. Phospholipidic monolayers dispersed on the top of an aqueous subphase provide a highly informative approach for studying drug–lipid interactions at the air–water interface, as there is no curvature effect (unlike vesicles) and the lateral packing can be precisely controlled [111, 113, 118]. Moreover, the control of the state of the hydrocarbon chains and the relative concentration of the drug (both on the monolayer and in the subphase) allows a molecular control of the parameters associated with the effects of drugs [113]. Supported phospholipid bilayers (SPBs) can also act as simple in vitro models for physiological membranes gathering the advantages of providing for a lipid bilayer with no curvature effect. These bilayers are placed in a solid substrate, and a hydration layer, typically with a thickness of around 1–2 nm, exists between the lower leaﬂet of the SPB and the solid substrate. The presence of this layer provides the bilayer with ﬂexibility and ﬂuidity, creating a dynamic system where lipid molecules can, laterally, diffuse freely [23]. More recently, alternative models made of lipid bilayer-modiﬁed microbeads and nanomaterials [119] have been extensively studied. The microbeads and nanomaterials-supported lipid bilayers (nanoSLBs) somewhat mimic the native environment of membrane-associated biomolecules. Thus, from a bionic point of view, they can be considered as curved lipid membranes with high structural integrity [119]. Compared with other lipid membranes, an attractive feature of nanoSLBs is that some states of mimetic biomembranes could be monitored by tracking the changes of some intrinsic properties of the nanomaterials. Thus, the nanoSLBs could be used as a biomembrane model studying some membrane properties and biological functions, such as the ﬂuidity of membranes, and the interaction between proteins and membranes [119]. 8.3.2 Experimental Techniques

The molecular description of the interactions between drugs and membrane constituents is a powerful tool to understand their mechanism of activity or toxicity. In view of this, a large number of methods have been developed for analyzing and quantifying various aspects of drug–membrane interactions (Figure 8.2). Experimental techniques used to study model membranes can be grouped into the following categories [1]: microscopic (optical and electron microscopy (EM)); spectroscopic (nuclear magnetic resonance (NMR), infrared (IR), electron spin resonance (ESR), ﬂuorescence, X-ray diffraction (XD), neutron diffraction); thermodynamic (calorimetry, electrochemical methods); hydrodynamic (high performance liquid chromatography (HPLC), dynamic light scattering (DLS), zeta potential, viscosity); chemical; and mechanical (pipette aspiration). Before analyzing the properties of lipids of interest, they have to be isolated from the source and then separated. HPLC is a very accurate technique that allows the separation of different fractions of lipids [1]. The size of the vesicles formed can be determined by a simple turbidity method or more accurately by DLS. A macroscopic view of the cell is provided by EM.

8.4 Drug−Membrane Interactions Applied to the Study of NSAIDs

Especially, the application of the freeze-fracture EM (FFEM) technique allows the direct visualization of the architecture of the membrane and the embedded proteins [120]. The surface charge of the membrane can be also measured by determining the zeta potential. To obtain structural properties (such as area per lipid molecule, volume, thickness, or order parameter), microscopic (atomic force microscopy (AFM) and Brewster angle microscopy (BAM)), spectroscopic, diffraction (X-ray; neutron), or scattering methods (small angle X-Ray scattering and wide angle X-Ray scattering (SAXS and WAXS)) are used [103, 117, 121, 122]. Dynamic properties such as diffusion, permeability, or ﬂuidity of membranes are measured by NMR, X-ray, or ﬂuorescence [80, 123, 124]. Phase transitions can be detected and characterized using calorimetric methods such as differential scanning calorimetry (DSC) [102, 107, 124–126]. Besides analyzing the effect on the physicochemical properties of membranes, it is also necessary to evaluate the inﬂuence of the membrane itself in the drug behavior and thus determine drug partition, localization, orientation, and conformation within the membrane. This information can be obtained from a vast number of techniques which include derivative spectroscopy, ﬂuorescence quenching, XD, circular dichroism (CD), and so on [1, 127, 128].

8.4 Drug–Membrane Interactions Applied to the Study of Nonsteroidal Anti-inﬂammatory Drugs (NSAIDs)

In the previous sections of this chapter, the clinical relevance of studying the interactions of drugs with membranes was brieﬂy presented. This section will cover the particular case of nonsteroidal anti-inﬂammatory drugs (NSAIDs) since these drugs can be presented as an iconic example of the clinical importance of studying drug–membrane interactions. NSAIDs are one of the world’s most prescribed groups of drugs for acute and chronic inﬂammatory diseases, and their continuous use is associated with severe gastrointestinal toxicity [127]. Principal targets for the NSAIDs in controlling pain and inﬂammation are the cyclooxygenases (COXs), which are membrane-associated enzymes. To bind with the targets these drugs have to pass through the biomembranes, and thus their interactions with lipid bilayers are expected to play a major role in guiding their COX inhibition [127]. However, the sequence of events resulting from COX inhibition does not totally explain the gastric toxicity of NSAIDs. Such mechanisms are complex and mucosal damage can also be related to a local acid effect of the dissolved drug [127]. Most NSAIDs are weakly acidic, lipid-soluble compounds. Since the cell membranes on the stomach wall contain lipids for protection against strong acids, they offer little resistance to the lipid-soluble NSAIDs.

201

202

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

The observation of the strong inﬂuence of NSAIDs on membranes either by their inhibition effect on the activity of the membrane-located enzyme COX, or by their topical effect on gastric mucosa, kept our investigation group interested in exploring the membrane changes upon the interaction with NSAIDs molecules. Therefore, the focus of our studies has been placed on the effects of NSAIDs on membranes, and speciﬁcally the ability of these drugs to bind to surface-active phospholipids, as well as their effect on the lipid dynamic properties of membrane models, once the results elucidate about the effects of these drugs compromising the integrity of gastric mucosal barrier. In order to identify and isolate the effect of different membrane parameters on the interaction of NSAIDs, the biomimicry of the bilayer lipid membrane has been achieved by the use of membrane models [127, 128] (liposomes and monolayers) once they show structural similarities to the lipid matrix of cell membranes. Occasionally, more complex systems were used like mouse splenocytes, mouse macrophages cell line – J774, human leukemia monocyte cell line – THP-1, human granulocytes, and mononuclear cells [129, 130]. The experimental techniques used in the interaction studies of NSAIDs with membrane model systems can constitute an example for those willing to apply the same approach to other drugs or compounds and will be grouped as techniques more focused on drugs (e.g., drugs physical–chemical studies) and techniques more focused on the effects of drugs on membrane (e.g., membrane structural and dynamic studies). 8.4.1 Drug Fundamental Physical–Chemical Studies

When studying the interaction of drugs with biomembranes, it is highly recommended to start with physical–chemical studies that include the evaluation of drug acid–base properties in aqueous or organized media; drug solubility in several media; drug liposome/water partition coefﬁcient (Kp ); and drug location within membranes. Indeed, one major question is to determine the likely mechanism by which the drug interacts with the membrane, which is dependent of the drugs hydrophilic/lipophilic balance. Therefore, in the study of the interaction of any compound with model membrane systems, the determination of the partition coefﬁcient should be the ﬁrst step. After this information is obtained, structural and dynamic studies can then be carried out. The extent of interaction of a solute with a membrane model system is evaluated in a quantitative way from its partition coefﬁcient. The Kp determinations can be successfully obtained using several techniques (for a review see reference [131]). Among all the techniques, we have chosen the spectroscopic (derivative spectroscopy and ﬂuorescence quenching techniques), which permitted, in the case of NSAIDs, to evaluate and compare the extent of penetration and/or interaction of the drugs with membrane phospholipids [128, 132–134]. When spectroscopic techniques are used, the measured parameter is a combination of signals from the ‘‘aqueous’’ and ‘‘lipidic’’ solute subpopulations, their relative weight depending on the partition

8.4 Drug−Membrane Interactions Applied to the Study of NSAIDs

coefﬁcient. Spectroscopic techniques are thus usually based on the analysis of signals originating from both phases. Therefore, there is no demand for their physical separation, which is a signiﬁcant advantage [131], since physical separation of phases may be laborious and may result in equilibria perturbation. Fluorescence techniques, on the other hand, are amongst the most sensitive; low solute concentrations can be used, leading to a small perturbation of the lipidic membranes. Understanding the role of the lipid or polar moiety of drugs in altering the partitioning between membrane and aqueous environments will also assist in predicting the subcellular localization of these therapeutic substances. Moreover, ﬂuorescence quenching studies and zeta potential measurements can be used to preview the location of drugs within the membrane. Fluorescence quenching can provide information on the position and orientation of drugs within the bilayer, as the precise location in the membrane of the ﬂuorescent probes used is known. Furthermore, for the same probe, the extent of quenching is inversely proportional to its distance to the drug that acts as ﬂuorescence quencher. In this way it was possible to predict the membrane location of NSAIDs, by measuring the decrease in ﬂuorescence of the probe inserted in the lipid membrane when increasing concentrations of drugs were added [128, 132–135]. Additional information on NSAIDs location has been obtained by assessing their inﬂuence on the membrane surface potential, as it provides an indication of the type of interactions between drug and lipid bilayer surfaces [132, 133, 135]. Measurement of acid–base properties of drugs (e.g., pKa ) is also essential since drugs can face different physiologic pH conditions, and consequently their chemical behavior may change. Brieﬂy, this method consists of a titration of a strong basic (or acidic) solution of the drug with a strong acid (or base) added with an automatic piston burette, controlled by a personal computer, under a nitrogen stream. The absorption spectrum of the drug is recorded in a system where a peristaltic pump enables circulation of the solution from the reaction pot, through a ﬂow-cell of the spectrophotometer and back to the reaction pot. The acid–base properties of NSAIDs have been assessed by this method [132, 133]. 8.4.2 Membrane Structural and Dynamic Studies

After drug physical–chemical studies have been conducted, which enabled a deeper knowledge about drugs lipophilicity and about the charged forms of drugs, structural and dynamic studies have been carried out to evaluate the interaction and penetration of NSAIDs in lipid bilayers and the resultant variations in the thermodynamic parameters and lipid phase transitions. In this case, DSC is the appropriate technique and can be used to determine the onset temperature of the lipid chain–melting and other phase transitions, and the area under the curve (i.e., the enthalpy), which is representative of the cooperativity. The ﬁrst effect observed by DSC when NSAIDs were interacting with membranes of DPPC and DMPC was the disappearance of the pretransition peak, which represents the phase transition

203

204

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

between bilayers in an Lβ organization, where the lipid molecules are parallel and tilted with respect to the bilayer normal, to a Pβ phase where the bilayer displays periodic ripples. The second effect was a shift on the main transition temperature, Tm , to a lower temperature, which represents the melting of the acyl chains from the Pβ to the Lα phase. Finally, the third effect was a broadening of the peak, indicating a lower cooperativity during the melting process. The calorimetric enthalpy, H did not change or changed slightly. This behavior is typical of an intercalation of the drugs into the lipid bilayer and indicates that the drug strongly interacts with liposomes. More speciﬁcally, it indicates that the molecule acts as a spacer causing ﬂuidization and disordering of the lipid molecules in the membrane [127, 135]. Modiﬁcations of membrane biophysical properties elicited by the drugs under study have also been evaluated by ﬂuorescence measurements of anisotropy [128]. Measurements of ﬂuorescence anisotropy of probes inserted in lipid bilayers are a powerful tool to report changes in membrane microviscosity. Upon excitation with polarized light, the emission from lipid membrane labeled with probes is also polarized. The extent of polarization of the emission is described in terms of the anisotropy (r). The emission can become depolarized by a number of processes. One cause of depolarization is changing the microviscosity of the lipid media where the probe is inserted, which provokes rotational diffusion changes of the probe. Therefore, a change in ﬂuorescence anisotropy of the probe will report a change in the microviscosity of the membrane where the probe rotates. Thus, in the case of NSAIDs there was a decrease in the ﬂuorescence anisotropy in a concentration-dependent manner and we could conclude that, in agreement with DSC studies, NSAIDs increased membrane ﬂuidity. Besides the effects of drugs on membrane microviscosity, it is also important to evaluate the effects on the order and packing of the lipids. As a result, the effects in symmetry and in the long- and short-range organization of bilayers; in the molecular packing of the bilayers as well as in the chain conformation of the molecules upon the interaction of the bilayers with NSAIDs have been investigated by SAXS and WAXS [127]. Scattering methods, such as SAXS and WAXS, have been useful to measure structural features on length scales between 1 nm up to several hundreds of nanometers by analyzing the scattering pattern at very low or very wide angles from the direct X-ray beam. These techniques provide complementary information: with SAXS it is possible to measure the long distances that correspond to the bilayer thickness; with WAXS it is possible to measure the short distances that correspond to chain order. Another highly sophisticated X-ray technique that was used to study the inﬂuence of NSAIDs in lipid membranes was grazing-incidence X-ray diffraction (GIXD), which provides the most precise structural parameters of condensed monolayers [127]. In GIXD, a monochromatic X-ray beam is directed on the surface at an angle of incidence that is smaller than the critical angle for total external reﬂection. Thus, the beam penetrates the subphase by only about 5 nm, and GIXD becomes a surface-sensitive technique. Consequently, the beam is diffracted by the structural pattern of the monolayer. The scattered intensities are detected as functions of the horizontal and the vertical scattering angle.

8.4 Drug−Membrane Interactions Applied to the Study of NSAIDs

The corresponding in-plane and out-of-plane components of the scattering vector allow the calculation of distinct lattice parameters, the tilt angle, and the direction of the aliphatic chains, as well as the correlation length, that is, the size of a homogeneously structured lipid domain. Therefore, along with the studies in bilayers systems, models consisting of phospholipids spread at the air–water interface as a Langmuir monolayer and NSAIDs used as a second component dissolved in the subphase allowed the variation of physical–chemical parameters of the interface as lateral pressure and surface providing the measurement of pressure-area isotherms, which are simple models for studying the drug’s penetration [127]. Simultaneously, BAM measurements have permitted to follow the drug adsorption to the lipid monolayer and detect possible drug–lipid ﬂuidity changes. 8.4.3 Results

Results obtained with the aforementioned techniques provided evidence that the NSAIDs evaluated have destabilizing effects, especially in the lipid gel phase, with a possible increase of the bilayer’s wettability, which could be transposed to a similar effect in the stomach’s mucus phospholipid gel layer with the development Digestive acids

Prostaglandines

Figure 8.3 Brewster angle microscopy image of the effect of NSAIDs on lipid membranes. Lipid membranes in the absence of NSAIDs (upper image) shows ﬂuid phase (dark area) and solid domains (clear spots). NSAIDs show a ﬂuidizing effect

on membrane (lower image) reducing the solid–lipid domains and increasing the ﬂuid content of the membrane. On the right side is an schematic interpretation of the ﬂuidizing effect in vivo.

205

206

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

of gastric erosions (Figure 8.3). The NSAIDs ﬂuidizing effects have proved to be parallel with their membrane preferential location and are also related with their physical–chemical properties. For instance, in the lipid gel phase (Lβ ), indomethacin and acemetacin showed to be located inside the hydrophobic core of the membrane with the ionized carboxyl group anchored at the headgroup region. This membrane location parallels the highest destabilizing effects presented by these drugs, which in turn can be related with their gastric toxicity. Indeed, the destabilizing effects of these drugs on membranes may be part of the mechanism by which these NSAIDs attenuate the hydrophobic barrier properties of the stomach’s mucus phospholipid gel layer with the consequent increase in the back diffusion of luminal acid into the mucosa and the development of erosions (Figure 8.3). In comparison, nimesulide, one NSAID studied, which has a less damaging gastrointestinal proﬁle, showed minimal effects of the lipid gel phase [127]. Further studies will be made with several other NSAIDs to extend the conclusions achieved. 8.5 Conclusions and Future Research Directions

The aims of this chapter were to highlight the importance of membrane composition and the dynamic molecular organization of membranes in drug membrane interaction studies, to point out the effects of such interactions on membrane properties, and ﬁnally to exemplify an application of drug–membrane interaction studies in the evaluation of the NSAIDs effect on membranes in their normal ﬂuid state (allowing the evaluation of the therapeutic effect of drugs) and in the gel state typical of the gastric protective barrier (allowing the evaluation of the toxic effects of drugs). The studies presented offer starting points for further exploration of the interaction of drugs with membranes. Drugs affect not only the phospholipids, but also other components of the cell membranes such as membrane proteins. Studying the interactions of drugs with different membrane proteins, membrane compositions, and levels of hydration and pH, as well as studying interactions of drugs with the lipid microdomains are fertile ﬁelds for future investigation. Results gathered with these studies can provide useful information to understand the therapeutic effects of drugs, as well as their different selectivity and toxicity. Ultimately, these studies may prove valuable in the design of novel drug formulations with increased efﬁcacy and reduced side effects. Acknowledgments

Partial ﬁnancial support for this work was provided by Fundac¸a˜o para a Ciˆencia e Tecnologia (FCT – Lisbon), through the contract PTDC/SAU-FCF/67718/2006. We are grateful to Dr Gerald Brezesinski and his research group at the Max Planck Institute (Golm, Germany) for his support and cooperation in some of our studies. We thank HASYLAB at DESY, Hamburg, Germany, for beam time and support through the contract I-20080033 EC.

References

References 1. Seydel, J.K., Wiese, M., Mannhold,

2. 3.

4.

5.

6.

7.

8.

9.

10.

11.

R., Kubinyi, H., and Folkers, G. (eds) (2002) Drug-Membrane Interactions: Analysis, Drug Distribution, Modeling, Wiley-VCH Verlag GmbH, Weinheim. Small, D.M. (ed.) (1998) Handbook of Lipid Research, Plenum Press, London. Berg, J.M., Tymoczko, J.L., and Stryer, L. (eds) (2006) Biochemistry, W. H. Freeman and Company. Singer, S.J. and Nicolson, G.L. (1972) The ﬂuid mosaic model of the structure of cell membranes. Science, 175, 720–731. Welti, R. and Glaser, M. (1994) Lipid domains in model and biological membranes. Chem. Phys. Lipids, 73, 121–137. Balasubramanian, K. and Schroit, A.J. (2003) Aminophospholipid asymmetry: a matter of life and death. Annu. Rev. Physiol., 65, 701–734. Lommerse, P.H., Spaink, H.P., and Schmidt, T. (2004) In vivo plasma membrane organization: results of biophysical approaches. Biochim. Biophys. Acta, 1664, 119–131. Brown, D.A. and London, E. (2000) Structure and function of sphingolipidand cholesterol-rich membrane rafts. J. Biol. Chem., 275, 17221–17224. Frankel, D.J., Pfeiffer, J.R., Surviladze, Z., Johnson, A.E., Oliver, J.M., Wilson, B.S., and Burns, A.R. (2006) Revealing the topography of cellular membrane domains by combined atomic force microscopy/ﬂuorescence imaging. Biophys. J., 90, 2404–2413. Jacobson, K., Mouritsen, O.G., and Anderson, R.G. (2007) Lipid rafts: at a crossroad between cell biology and physics. Nat. Cell Biol., 9, 7–14. Vereb, G., Szollosi, J., Matko, J., Nagy, P., Farkas, T., Vigh, L., Matyus, L., Waldmann, T.A., and Damjanovich, S. (2003) Dynamic, yet structured: the cell membrane three decades after the Singer-Nicolson model. Proc. Natl. Acad. Sci. U.S.A., 100, 8053–8058.

12. Simons, K. and Ikonen, E. (1997) Func-

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

tional rafts in cell membranes. Nature, 387, 569–572. Mouritsen, O.G. and Jorgensen, K. (1992) Dynamic lipid-bilayer heterogeneity: a mesoscopic vehicle for membrane function? Bioessays, 14, 129–136. Escriba, P.V., Gonzalez-Ros, J.M., Goni, F.M., Kinnunen, P.K.J., Vigh, L., Sanchez-Magraner, L., Fernandez, A.M., Busquets, X., Horvath, I., and Barcelo-Coblijn, G. (2008) Membranes: a meeting point for lipids, proteins and therapies. J. Cell. Mol. Med., 12, 829–875. Israelachvili, J.N., Mitchell, D.J., and Ninham, B.W. (1977) Theory of self-assembly of lipid bilayers and vesicles. Biochim. Biophys. Acta, 470, 185–201. Whitesides, G.M. and Grzybowski, B. (2002) Self-assembly at all scales. Science, 295, 2418–2421. Tanford, C. (1979) Interfacial free energy and the hydrophobic effect. Proc. Natl. Acad. Sci. U.S.A., 76, 4175–4176. Tanford, C. (1978) The hydrophobic effect and the organization of living matter. Science, 200, 1012–1018. Marrink, S.J. and Tieleman, D.P. (2001) Molecular dynamics simulation of a lipid diamond cubic phase. J. Am. Chem. Soc., 123, 12383–12391. Marrink, S.J., Lindahl, E., Edholm, O., and Mark, A.E. (2001) Simulation of the spontaneous aggregation of phospholipids into bilayers. J. Am. Chem. Soc., 123, 8638–8639. Kocherbitov, V. (2005) Driving forces of phase transitions in surfactant and lipid systems. J. Phys. Chem. B, 109, 6430–6435. Mouritsen, O.G., Kinnunen, P.K.J., Merz, K.M.J., and Roux, B. (eds) (1996) Role of Lipid Organization and Dynamics for Membrane Functionality, Birkh¨auser, Berlin. Nussio, M.R., Sykes, M.J., Miners, J.O., and Shapter, J.G. (2009) Kinetics membrane disruption due to drug

207

208

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

24.

25.

26.

27.

28.

29.

30.

interactions of chlorpromazine hydrochloride. Langmuir, 25, 1086–1090. Seydel, J.K., Coats, E.A., Cordes, H.P., and Wiese, M. (1994) Drug membrane interaction and the importance for drug transport, distribution, accumulation, efﬁcacy and resistance. Arch. Pharmazie, 327, 601–610. Seydel, J.K., Velasco, M.A., Coats, E.A., Cordes, H.P., Kunz, B., and Wiese, M. (1992) The importance of drug-membrane interaction in drug research-and-development. Quant. Struct.-Act. Relat., 11, 205–210. Mouristen, O.G., Anderson, H.K., Anderson, J.S., Davidsen, J., Nielsen, L.K., Jørgensen, K., Testa, B., van De Waterbeemd, H. Folkers, G., and Guy, R. (eds) (2001) Structure of Liposomal Membranes in Relation to Permeation, Wiley-VCH & Verlag Helvetica Chimica Acta, Weinheim, Z¨urich. Kansy, M., Fischer, K., Kratzat, F., Senner, B., Wagner, B., Parilla, I., Testa, B., van De Waterbeemd, H., Folkers, G., and Guy, R. (eds) (2001) High-throughput Artiﬁcial Membrane Permeability Studies in Early Lead Discovery and Development, Wiley-VCH & Verlag Helvetica Chimica Acta, Weinheim, Z¨urich. Kr¨amer, S.D., Abbott, N.J., Begley, D.J., Testa, B., van De Waterbeemd, H., Folkers, G., and Guy, R. (eds) (2001) Biological Models to Study Blood-brain Barrier Permeation, Wiley-VCH Verlag GmbH Helvetica Chimica Acta, Weinheim, Z¨urich. Neves, P., Berkane, E., Gameiro, P., Winterhalter, M., and de Castro, B. (2005) Interaction between quinolones antibiotics and bacterial outer membrane porin OmpF. Biophys. Chem., 113, 123–128. Grancelli, A., Morros, A., Cabanas, M.E., Domenech, O., Merino, S., Vazquez, J.L., Montero, M.T., Vinas, M., and Hernandez-Borrell, J. (2002) Interaction of 6-ﬂuoroquinolones with dipalmitoylphosphatidylcholine monolayers and liposomes. Langmuir, 18, 9177–9182.

31. Giacometti, A., Cirioni, O., Ghiselli, R.,

32.

33.

34.

35.

36.

37.

38.

Mocchegiani, F., Orlando, F., Silvestri, C., Bozzi, A., Di Giulio, A., Luzi, C., Mangoni, M.L., Barra, D., Saba, V., Scalise, G., and Rinaldi, A.C. (2006) Interaction of antimicrobial peptide temporin L with lipopolysaccharide in vitro and in experimental rat models of septic shock caused by gram-negative bacteria. Antimicrob. Agents Chemother., 50, 2478–2486. Chaudhuri, S., Banerjee, A., Basu, K., Sengupta, B., and Sengupta, P.K. (2007) Interaction of ﬂavonoids with red blood cell membrane lipids and proteins: antioxidant and antihemolytic effects. Int. J. Biol. Macromol., 41, 42–48. Bauer, M., Megret, C., Lamure, A., LaCabanne, C., and Fauran-Clavel, M.J. (1990) Differential scanning calorimetry study of the interaction of antidepressant drugs, noradrenaline, and 5-hydroxytryptamine with a membrane model. J. Pharm. Sci., 79, 897–901. Fisar, Z., Fuksova, K., and Velenovska, M. (2004) Binding of imipramine to phospholipid bilayers using radioligand binding assay. Gen. Physiol. Biophys., 23, 77–99. Yoshida, T., Taga, K., Okabayashi, H., Kamaya, H., and Ueda, I. (1989) Changes in surface capacitance and conductance parallel to phospholipid membranes associated with phase transition: effects of halothane. Biochim. Biophys. Acta, 984, 253–256. Moya-Quiles, M.R., Munoz-Delgado, E., and Vidal, C.J. (1994) Interactions of the pyrethroid insecticide allethrin with liposomes. Arch. Biochem. Biophys., 312, 95–100. Antunes-Madeira, M.C., Videira, R.A., and Madeira, V.M. (1994) Effects of parathion on membrane organization and its implications for the mechanisms of toxicity. Biochim. Biophys. Acta, 1190, 149–154. Antunes-Madeira, M.C., Almeida, L.M., and Madeira, V.M. (1993) Depth-dependent effects of DDT and lindane on the ﬂuidity of native membranes and extracted lipids. Implications for mechanisms of toxicity.

References

39.

40.

41.

42.

43.

44.

45.

46.

47.

Bull. Environ. Contam. Toxicol., 51, 787–794. Jorgensen, K., Ipsen, J.H., Mouritsen, O.G., Bennett, D., and Zuckermann, M.J. (1991) The effects of density ﬂuctuations on the partitioning of foreign molecules into lipid bilayers: application to anaesthetics and insecticides. Biochim. Biophys. Acta, 1067, 241–253. Sreeramulu, K., Liu, R., and Sharom, F.J. (2007) Interaction of insecticides with mammalian P-glycoprotein and their effect on its transport function. Biochim. Biophys. Acta-Biomembr., 1768, 1750–1757. Soderlund, T., Lehtonen, J.Y., and Kinnunen, P.K. (1999) Interactions of cyclosporin A with phospholipid membranes: effect of cholesterol. Mol. Pharmacol., 55, 32–38. Wenz, J.J. and Barrantes, F.J. (2003) Steroid structural requirements for stabilizing or disrupting lipid domains. Biochemistry, 42, 14267–14276. Hendrich, A.B., Wesolowska, O., and Michalak, K. (2001) Triﬂuoperazine induces domain formation in zwitterionic phosphatidylcholine but not in charged phosphatidylglycerol bilayers. Biochim. Biophys. Acta-Biomembr., 1510, 414–425. Schanck, A., Mingeot-Leclercq, M.P., Tulkens, P.M., Carrier, D., Smith, I.C., and Jarrell, H.C. (1992) Interactions of aminoglycoside antibiotics with phospholipids. A deuterium nuclear magnetic resonance study. Chem. Phys. Lipids, 62, 153–163. Colome, C., Alsina, M.A., Busquets, M.A., Haro, I., and Reig, F. (1993) Interaction of aminoglycosides and colistin with model membranes – liposomes and monolayers. Int. J. Pharm., 90, 59–71. Berquand, A., Fa, N., Dufrene, Y.F., and Mingeot-Leclercq, M.P. (2005) Interaction of the macrolide antibiotic azithromycin with lipid bilayers: effect on membrane organization, ﬂuidity, and permeability. Pharm. Res., 22, 465–475. Tyteca, D., Schanck, A., Dufrene, Y.F., Deleu, M., Courtoy, P.J., Tulkens, P.M., and Mingeot-Leclercq, M.P. (2003)

48.

49.

50.

51.

52.

53.

The macrolide antibiotic azithromycin interacts with lipids and affects membrane organization and ﬂuidity: studies on Langmuir-Blodgett monolayers, liposomes and J774 macrophages. J. Membr. Biol., 192, 203–215. Berquand, A., Mingeot-Leclercq, M.P., and Dufrene, Y.F. (2004) Real-time imaging of drug-membrane interactions by atomic force microscopy. Biochim. Biophys. Acta-Biomembr., 1664, 198–205. Fa, N., Lins, L., Courtoy, P.J., Dufrene, Y., Van Der Smissen, P., Brasseur, R., Tyteca, D., and Mingeot-Leclercq, M.P. (2007) Decrease of elastic moduli of DOPC bilayers induced by a macrolide antibiotic, azithromycin. Biochim. Biophys. Acta-Biomembr., 1768, 1830–1838. Fa, N., Ronkart, S., Schanck, A., Deleu, M., Gaigneaux, A., Goonnaghtigh, E., and Mingeot-Leclercq, M.P. (2006) Effect of the antibiotic azithromycin on thermotropic behavior of DOPC or DPPC bilayers. Chem. Phys. Lipids, 144, 108–116. Montenez, J.P., VanBambeke, F., Piret, J., Schanck, A., Brasseur, R., Tulkens, P.M., and MingeotLeclercq, M.P. (1996) Interaction of the macrolide azithromycin with phospholipids. 2. Biophysical and computer-aided conformational studies. Eur. J. Pharmacol., 314, 215–227. Tyteca, D., Van Der Smissen, P., Mettlen, M., Van Bambeke, F., Tulkens, P.M., Mingeot-Leclercq, M.P., and Courtoy, P.J. (2002) Azithromycin, a lysosomotropic antibiotic, has distinct effects on ﬂuid-phase and receptor-mediated endocytosis, but does not impair phagocytosis in J774 macrophages. Exp. Cell Res., 281, 86–100. Tyteca, D., Van Der Smissen, P., Van Bambeke, F., Leys, K., Tulkens, P.M., Courtoy, P.J., and Mingeot-Leclercq, M.P. (2001) Azithromycin, a lysosomotropic antibiotic, impairs ﬂuid-phase pinocytosis in cultured ﬁbroblasts. Eur. J. Cell Biol., 80, 466–478.

209

210

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs 54. Maheswari, K.U., Ramachandran, T.,

55.

56.

57.

58.

59.

60.

61.

62.

and Rajaji, D. (2000) Interaction of cisplatin with planar model membranes – dose dependent change in electrical characteristics. Biochim. Biophys. Acta-Biomembr., 1463, 230–240. Speelmans, G., Staffhorst, R.W.H.M., Versluis, K., Reedijk, J., and de Kruijff, B. (1997) Cisplatin complexes with phosphatidylserine in membranes. Biochemistry, 36, 10545–10550. Burger, K.N.J., Staffhorst, R.W.H.M. and De Kruijff, B. (1999) Interaction of the anti-cancer drug cisplatin with phosphatidylserine in intact and semi-intact cells. Biochim. Biophys. Acta-Biomembr., 1419, 43–54. Speelmans, G., Sips, W.H., Grisel, R.J., Staffhorst, R.W., Fichtinger-Schepman, A.M., Reedijk, J., and de Kruijff, B. (1996) The interaction of the anti-cancer drug cisplatin with phospholipids is speciﬁc for negatively charged phospholipids and takes place at low chloride ion concentration. Biochim. Biophys. Acta, 1283, 60–66. Suwalsky, M., Hernandez, P., Villena, F., and Sotomayor, C.P. (2000) The anticancer drug cisplatin interacts with the human erythrocyte membrane. Z. Naturforsch., C: J. Biosci., 55, 461–466. Wang, K., Lu, J.F., and Li, R.C. (1996) The events that occur when cisplatin encounters cells. Coord. Chem. Rev., 151, 53–88. Baginski, M., Sternal, K., Czub, J., and Borowski, E. (2005) Molecular modelling of membrane activity of amphotericin B, a polyene macrolide antifungal antibiotic. Acta Biochim. Pol., 52, 655–658. Matsuoka, S., Ikeuchi, H., Umegawa, Y., Matsumori, N., and Murata, M. (2006) Membrane interaction of amphotericin B as single-length assembly examined by solid state NMR for uniformly C-13-enriched agent. Bioorg. Med. Chem., 14, 6608–6614. Bechinger, B. (1997) Structure and functions of channel-forming peptides: magainins, cecropins, melittin

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

and alamethicin. J. Membr. Biol., 156, 197–211. Matsuzaki, K. (1998) Magainins as paradigm for the mode of action of pore forming polypeptides. Biochim. Biophys. Acta, 1376, 391–400. Oren, Z. and Shai, Y. (1998) Mode of action of linear amphipathic alpha-helical antimicrobial peptides. Biopolymers, 47, 451–463. Sitaram, N. and Nagaraj, R. (1999) Interaction of antimicrobial peptides with biological and model membranes: structural and charge requirements for activity. Biochim. Biophys. Acta, 1462, 29–54. Shai, Y. (1999) Mechanism of the binding, insertion and destabilization of phospholipid bilayer membranes by alpha-helical antimicrobial and cell non-selective membrane-lytic peptides. Biochim. Biophys. Acta, 1462, 55–70. Foged, C., Franzyk, H., Bahrami, S., Frokjaer, S., Jaroszewski, J.W., Nielsen, H.M., and Olsen, C.A. (2008) Cellular uptake and membrane-destabilising properties of alpha-peptide/beta-peptoid chimeras: lessons for the design of new cell-penetrating peptides. Biochim. Biophys. Acta-Biomembr., 1778, 2487–2495. Hitz, T., Iten, R., Gardiner, J., Namoto, K., Walde, P., and Seebach, D. (2006) Interaction of alpha- and beta-oligoarginine-acids and amides with anionic lipid vesicles: a mechanistic and thermodynamic study. Biochemistry, 45, 5817–5829. Molitoris, B.A., Meyer, C., Dahl, R., and Geerdes, A. (1993) Mechanism of ischemia-enhanced aminoglycoside binding and uptake by proximal tubule cells. Am. J. Physiol., 264, F907–F916. Mingeot-Leclercq, M.P. and Tulkens, P.M. (1999) Aminoglycosides: nephrotoxicity. Antimicrob. Agents Chemother., 43, 1003–1012. Gallis, H.A., Drew, R.H., and Pickard, W.W. (1990) Amphotericin B: 30 years of clinical experience. Rev. Infect. Dis., 12, 308–329. Barwicz, J. and Tancrede, P. (1997) The effect of aggregation state of amphotericin-B on its interactions with

References

73.

74.

75.

76.

77.

78.

79.

80.

cholesterol- or ergosterol-containing phosphatidylcholine monolayers. Chem. Phys. Lipids, 85, 145–155. Andreoli, T.E. (1974) The structure and function of amphotericin B-cholesterol pores in lipid bilayer membranes. Ann. N. Y. Acad. Sci., 235, 448–468. Cohen, B.E. (1992) A sequential mechanism for the formation of aqueous channels by amphotericin B in liposomes. The effect of sterols and phospholipid composition. Biochim. Biophys. Acta, 1108, 49–58. de Kruijff, B. and Demel, R.A. (1974) Polyene antibiotic-sterol interactions in membranes of Acholeplasma laidlawii cells and lecithin liposomes. 3. Molecular structure of the polyene antibiotic-cholesterol complexes. Biochim. Biophys. Acta, 339, 57–70. de Paula, E. and Schreier, S. (1996) Molecular and physicochemical aspects of local anesthetic-membrane interaction. Braz. J. Med. Biol. Res., 29, 877–894. Malheiros, S.V.P., Pinto, L.M.A., Gottardo, L., Yokaichiya, D.K., Fraceto, L.F., Meirelles, N.C., and de Paula, E. (2004) A new look at the hemolytic effect of local anesthetics, considering their real membrane/water partitioning at pH 7.4. Biophys. Chem., 110, 213–221. Yun, I., Cho, E.S., Jang, H.O., Kim, U.K., Choi, C.H., Chung, I.K., Kim, I.S., and Wood, W.G. (2002) Amphiphilic effects of local anesthetics on rotational mobility in neuronal and model membranes. Biochim. Biophys. Acta, 1564, 123–132. Colell, A., Garcia-Ruiz, C., Lluis, J.M., Coll, O., Mari, M., and Fernandez-Checa, J.C. (2003) Cholesterol impairs the adenine nucleotide translocator-mediated mitochondrial permeability transition through altered membrane ﬂuidity. J. Biol. Chem., 278, 33928–33935. Cronier, F., Patenaude, A., Gaudreault, R.C., and Auger, M. (2007) Membrane composition modulates the interaction between a new class of antineoplastic agents deriving from aromatic

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

2-chloroethylureas and lipid bilayers: a solid-state NMR study. Chem. Phys. Lipids, 146, 125–135. Grunicke, H., Grunewald, K., Helliger, W., Scheidl, F., Wolff-Schreiner, E., and Puschendorf, B. (1983) Inhibition of tumor growth by an alkylation of the plasma membrane. Adv. Enzyme Regul., 21, 21–30. Ihlenfeldt, M., Gantner, G., Harrer, M., Puschendorf, B., Putzer, H., and Grunicke, H. (1981) Interaction of the alkylating antitumor agent 2,3,5-tris(ethyleneimino)benzoquinone with the plasma membrane of Ehrlich ascites tumor cells. Cancer Res., 41, 289–293. Theumer, M.G., Clop, E.M., Rubinstein, H.R., and Perillo, M.A. (2008) The lipid-mediated hypothesis of fumonisin B1 toxicodynamics tested in model membranes. Colloids Surf. B: Biointerfaces, 64, 22–33. Bosch, I. and Croop, J. (1996) P-glycoprotein multidrug resistance and cancer. Biochim. Biophys. Acta, 1288, F37–F54. Gottesman, M.M. and Pastan, I. (1993) Biochemistry of multidrug resistance mediated by the multidrug transporter. Annu. Rev. Biochem., 62, 385–427. Omote, H. and Al-Shawi, M.K. (2006) Interaction of transported drugs with the lipid bilayer and beta-glycoprotein through a solvation exchange mechanism. Biophys. J., 90, 4046–4059. Ambudkar, S.V., Kimchi-Sarfaty, C., Sauna, Z.E., and Gottesman, M.M. (2003) P-glycoprotein: from genomics to mechanism. Oncogene, 22, 7468–7485. Borst, P. and Elferink, R.O. (2002) Mammalian ABC transporters in health and disease. Annu. Rev. Biochem., 71, 537–592. Leslie, E.M., Deeley, R.G., and Cole, S.P. (2005) Multidrug resistance proteins: role of P-glycoprotein, MRP1, MRP2, and BCRP (ABCG2) in tissue defense. Toxicol. Appl. Pharmacol., 204, 216–237. Lu, P., Liu, R., and Sharom, F.J. (2001) Drug transport by reconstituted P-glycoprotein in proteoliposomes.

211

212

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

91.

92.

93.

94.

95.

96.

97.

98.

99.

Effect of substrates and modulators, and dependence on bilayer phase state. Eur. J. Biochem., 268, 1687–1697. Omote, H. and Al-Shawi, M.K. (2002) A novel electron paramagnetic resonance approach to determine the mechanism of drug transport by P-glycoprotein. J. Biol. Chem., 277, 45688–45694. Romsicki, Y. and Sharom, F.J. (1997) Interaction of P-glycoprotein with deﬁned phospholipid bilayers: a differential scanning calorimetric study. Biochemistry, 36, 9807–9815. Romsicki, Y. and Sharom, F.J. (1999) The membrane lipid environment modulates drug interactions with the P-glycoprotein multidrug transporter. Biochemistry, 38, 6887–6896. Drori, S., Eytan, G.D., and Assaraf, Y.G. (1995) Potentiation of anticancer-drug cytotoxicity by multidrug-resistance chemosensitizers involves alterations in membrane ﬂuidity leading to increased membrane-permeability. Eur. J. Biochem., 228, 1020–1029. Michalak, K., Wesolowska, O., Motohashi, N., Molnar, J., and Hendrich, A.B. (2006) Interactions of phenothiazines with lipid bilayer and their role in multidrug resistance reversal. Curr. Drug Targets, 7, 1095–1105. Pajeva, I.K., Wiese, M., Cordes, H.P., and Seydel, J.K. (1996) Membrane interactions of some catamphiphilic drugs and relation to their multidrug-resistance-reversing ability. J. Cancer Res. Clin. Oncol., 122, 27–40. Bergethon, P.R. and Simons, E.R. (eds) (1990) Biophysical Chemistry – Molecules to Membranes, Springer-Verlag, New York. Mustonen, P. and Kinnunen, P.K. (1991) Activation of phospholipase A2 by adriamycin in vitro. Role of drug-lipid interactions. J. Biol. Chem., 266, 6302–6307. Pande, A.H., Qin, S., and Tatulian, S.A. (2005) Membrane ﬂuidity is a key modulator of membrane binding,

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

insertion, and activity of 5-lipoxygenase. Biophys. J., 88, 4084–4094. Nicolay, K. and de Kruijff, B. (1987) Effects of adriamycin on respiratory chain activities in mitochondria from rat liver, rat heart and bovine heart. Evidence for a preferential inhibition of complex III and IV. Biochim. Biophys. Acta, 892, 320–330. Kumar, R., Holian, O., Cook, B., and Roshani, P. (1997) Inhibition of rat brain protein kinase C by lipid soluble psychotropics. Neurochem. Res., 22, 1–10. Epand, R.M. (1997) Biophysical studies of lipopeptide-membrane interactions. Biopolymers, 43, 15–24. Thormann, E., Simonsen, A.C., Nielsen, L.K., and Mouritsen, O.G. (2007) Ligand-receptor interactions and membrane structure investigated by AFM and time-resolved ﬂuorescence microscopy. J. Mol. Recognit., 20, 554–560. Banerjee, S., Bennouna, M., Ferreira-Marques, J., Ruysschaert, J.M., and Caspers, J. (1999) Lipid-drug interaction and colligative properties in phospholipid vesicles. J. Colloid Interface Sci., 219, 168–177. Ohki, S. (1981) Membrane-potential, surface-potential, and ionic permeabilities. Physiol. Chem. Phys., 13, 195–210. Suzuki, H. and Takeuchi, S. (2008) Microtechnologies for membrane protein studies. Anal. Bioanal. Chem., 391, 2695–2702. Liu, Q., Qu, Y., Van Antwerpen, R., and Farrell, N. (2006) Mechanism of the membrane interaction of polynuclear platinum anticancer agents. Implications for cellular uptake. Biochemistry, 45, 4248–4256. Lasic, D.D. (ed.) (1993) Liposomes – from Physics to Applications, Elsevier, New York. Lasic, D.D. (1988) The mechanism of vesicle formation. Biochem. J., 256, 1–11. Brezesinski, G. and Mohwald, H. (2003) Langmuir monolayers to study interactions at model membrane

References

111.

112.

113.

114.

115.

116.

117.

118.

119.

surfaces. Adv. Colloid Interface Sci., 100-102, 563–584. Brockman, H. (1999) Lipid monolayers: why use half a membrane to characterize protein-membrane interactions? Curr. Opin. Struct. Biol., 9, 438–443. Caetano, W., Ferreira, M., Tabak, M., Sanchez, M.I.M., Oliveira, O.N., Kruger, P., Schalke, M., and Losche, M. (2001) Cooperativity of phospholipid reorganization upon interaction of dipyridamole with surface monolayers on water. Biophys. Chem., 91, 21–35. Hidalgo, A.A., Caetano, W., Tabak, M., and Oliveira, O.N. (2004) Interaction of two phenothiazine derivatives with phospholipid monolayers. Biophys. Chem., 109, 85–104. Jutila, A., Soderlund, T., Pakkanen, A.L., Huttunen, M., and Kinnunen, P.K. (2001) Comparison of the effects of clozapine, chlorpromazine, and haloperidol on membrane lateral heterogeneity. Chem. Phys. Lipids, 112, 151–163. Agasosler, A.V., Tungodden, L.M., Cejka, D., Bakstad, E., Sydnes, L.K., and Holmsen, H. (2001) Chlorpromazine-induced increase in dipalmitoylphosphatidylserine surface area in monolayers at room temperature. Biochem. Pharmacol., 61, 817–825. Varnier Agasoster, A. and Holmsen, H. (2001) Chlorpromazine associates with phosphatidylserines to cause an increase in the lipid’s own interfacial molecular area–role of the fatty acyl composition. Biophys. Chem., 91, 37–47. da Silva, A.M.G. and Romao, R.I.S. (2005) Mixed monolayers involving DPPC, DODAB and oleic acid and their interaction with nicotinic acid at the air-water interface. Chem. Phys. Lipids, 137, 62–76. Nagy, I.B., Haro, I., Alsina, M.A., Reig, F., and Hudecz, F. (1998) Interaction of branched chain polymeric polypeptides with phospholipid model membranes. Biopolymers, 46, 169–179. Zhang, L.X., Liu, J.Y., and Wang, E.K. (2008) A new method for studying

120.

121.

122.

123.

124.

125.

126.

127.

the interaction between chlorpromazine and phospholipid bilayer. Biochem. Biophys. Res. Commun., 373, 202–205. Ladaviere, C., Toustou, M., Gulik-Krzywicki, T., and Tribet, C. (2001) Slow reorganization of small phosphatidylcholine vesicles upon adsorption of amphiphilic polymers. J. Colloid Interface Sci., 241, 178–187. Lewis, R.N., Tristram-Nagle, S., Nagle, J.F., and McElhaney, R.N. (2001) The thermotropic phase behavior of cationic lipids: calorimetric, infrared spectroscopic and X-ray diffraction studies of lipid bilayer membranes composed of 1,2-di-O-myristoyl-3-N,N,N-trimethylaminopropane (DM-TAP). Biochim. Biophys. Acta, 1510, 70–82. Tristram-Nagle, S. and Nagle, J.F. (2004) Lipid bilayers: thermodynamics, structure, ﬂuctuations, and interactions. Chem. Phys. Lipids, 127, 3–14. Panicker, L. and Mishra, K.P. (2006) Nuclear magnetic resonance and thermal studies on the interaction between salicylic acid and model membranes. Biophys. Chem., 120, 15–23. Tsogas, I., Tsiourvas, D., Nounesis, G., and Paleos, C.M. (2006) Modeling cell membrane transport: interaction of guanidinylated poly(propylene imine) dendrimers with a liposomal membrane consisting of phosphate-based lipids. Langmuir, 22, 11322–11328. Castelli, F., Raudino, A., and Fresta, M. (2005) A mechanistic study of the permeation kinetics through biomembrane models: Gemcitabine-phospholipid bilayer interaction. J. Colloid Interface Sci., 285, 110–117. Gicquaud, C., Auger, M., Wong, P.T.T., Poyet, P., Boudreau, N., and CGaudreault, R. (1996) Interaction of 4-tert-butyl-[3-(2-chloroethyl)ureido] benzene with phosphatidylcholine bilayers: a differential scanning calorimetry and infrared spectroscopy study. Arch. Biochem. Biophys., 334, 193–199. Lucio, M., Bringezu, F., Reis, S., Lima, J.L.F.C., and Brezesinski, G.

213

214

8 Molecular Mechanisms Underlying Therapeutic and Toxic Effects of Drugs

128.

129.

130.

131.

(2008) Binding of nonsteroidal anti-inﬂammatory drugs to DPPC: structure and thermodynamic aspects. Langmuir, 24, 4132–4139. Lucio, M., Ferreira, H., Lima, J.L.F.C., Matos, C., de Castro, B., and Reis, S. (2004) Inﬂuence of some anti-inﬂammatory drugs in membrane ﬂuidity studied by ﬂuorescence anisotropy measurements. Phys. Chem. Chem. Phys., 6, 1493–1498. Sousa, C., Nunes, C., Lucio, M., Ferreira, H., Lima, J.L.F.C., Tavares, J., Cordeiro-Da-Silva, A., and Reis, S. (2008) Effect of nonsteroidal anti-inﬂammatory drugs on the cellular membrane ﬂuidity. J. Pharm. Sci., 97, 3195–3206. Ferreira, H., Lucio, M., Lima, J.L.F.C., Cordeiro-da-Silva, A., Tavares, J., and Reis, S. (2005) Effect of anti-inﬂammatory drugs on splenocyte membrane ﬂuidity. Anal. Biochem., 339, 144–149. Santos, N.C., Prieto, M., and Castanho, M.A.R.B. (2003) Quantifying molecular partition into model systems of biomembranes: an emphasis on optical

132.

133.

134.

135.

spectroscopic methods. Biochim. Biophys. Acta-Biomembr., 1612, 123–135. Ferreira, H., Lucio, M., Lima, J.L.F.C., Matos, C., and Reis, S. (2005) Effects of diclofenac on EPC liposome membrane properties. Anal. Bioanal. Chem., 382, 1256–1264. Ferreira, H., Lucio, M., Lima, J.L.F.C., Matos, C., and Reis, S. (2005) Interaction of clonixin with EPC liposomes used as membrane models. J. Pharm. Sci., 94, 1277–1287. Ferreira, H., Lucio, M., de Castro, B., Gameiro, P., Lima, J.L.F.C., and Reis, S. (2003) Partition and location of nimesulide in EPC liposomes: a spectrophotometric and ﬂuorescence study. Anal. Bioanal. Chem., 377, 293–298. Lucio, M., Nunes, C., Gaspar, D., Golebska, K., Wisniewski, M., Lima, J.L.F.C., Brezesinski, G., and Reis, S. (2009) Effect of anti-inﬂammatory drugs in phosphatidylcholine membranes: a ﬂuorescence and calorimetric study. Chem. Phys. Lett., 471, 300–309.

215

9 Targeting Disease with Small Molecule Inhibitors of Protein–Protein Interactions Fedor Forafonov, Elena Miranda, Ida Karin Nordgren, and Ali Tavassoli

9.1 Introduction

There exists in every cell an intricate network of thousands of protein interactions, which control and regulate the majority of cellular processes and functions, from intercellular communication, proliferation, growth, and differentiation to programmed cell death. Alterations in protein interaction networks are the key regulatory components of almost all diseases, from viral infection to heart disease or cancer. There is therefore tremendous potential for small molecule modulators of protein–protein interactions to play a key role in the next generation of healthcare. Such compounds would treat the cause, not the symptoms of disease, at an earlier stage and on a more fundamental level; they would also serve as tools to enhance our current understanding of biological processes. As a result of this potential for novel therapeutic agents, there has been a great deal of recent effort to develop methodologies that allow the facile and rapid identiﬁcation of small molecules capable of selectively disrupting speciﬁc protein–protein interactions [1, 2]. There are however, several factors that make identifying such compounds extremely challenging. Unlike enzymes, which have substrates that serve as an ideal starting point for the design of inhibitors, no such small molecule starting points exist for protein–protein interactions. The often large and uncharted protein surface combined with a lack of cavities and small molecule binding sites, results in the discovery of small molecule inhibitors being quite challenging. The size of the interacting protein interfaces can vary from 750 to 5000 A˚ 2 , and are typically stabilized via a series of buried hydrophobic patches [3]. Such interactions and their subsequent stabilization is thought to occur through van der Waals contacts between nonpolar amino acid residues, driven by the gain in free energy resulting from elimination of the polar aqueous environment. Electrostatic forces and hydrogen bonds are also believed to play an important part in partner recognition and subsequent stabilization of the protein dimer [4]. Such stabilizing regions, and more speciﬁcally the residues responsible for stabilizing the interaction are termed hot spots [4]. These hot spot residues have traditionally been identiﬁed using a combination of X-ray crystallography and site-directed Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

216

9 Protein Interactions with Small Molecules

mutagenesis, replacing individual amino acids with alanine (alanine scanning), and measuring the effect on the total binding energy of the complex. This approach tends to be quite laborious, and consequently available data on hot spot residues is limited, but the recent advances in computational power have resulted in several in silico approaches identifying hot spots on protein surfaces. Given the structural diversity and large number of protein interactions, it seems reasonable to assume that each interacting interface is also speciﬁc, unique, and diverse. It is also thought however, that proteins with multiple interactions tend to use the same hot spots, with structural changes varying the hot spot, adapting it to present the same residues in different structural contexts. Small molecules that target hot spots may also inadvertently affect other protein–protein interactions, with subsequent physiological side effects. The presence of structural and biophysical data for two interacting proteins, allows the possibility of rationally designed small molecule inhibitors, or in silico screening of virtual libraries. Although the availability of such structural data for interacting proteins is relatively rare, there have been several examples of protein–protein interaction inhibitors identiﬁed by these methods [5–7].

9.2 High-Throughput Screening of Chemical Libraries

The recent advances in chemical biology and functional genomics have given rise to new methods and opportunities for drug discovery, and in turn new methods for the generation of small molecule libraries [8–12]. Combinatorial chemistry, the method of choice for assembling chemically synthesized libraries has also undergone rapid development over the last few years [13–15]. Yet, despite providing almost unlimited functional group diversity, chemically synthesized libraries of small molecules still lack a suitable, straightforward method for their decoding. Split and pool protocols allow for the chemical synthesis of libraries with up to ∼105 members [16, 17]; however, the paths to identifying the active members of such libraries are often laborious and complicated [18]. Another approach to identifying inhibitors of protein–protein interactions is to use high-throughput screens. The composition of the libraries used in such screens can vary from small molecules to peptides, with library size varying from a few hundred to a hundred million. The challenge with this method of drug discovery lies in identifying the binding site and mode of action of the active compounds. Also the compound/protein ratio in such screens may be quite high and not reﬂect realistic physiological levels. The advantage of high-throughput approaches is that unlike rational design, they tend to not bias the compounds to one binding site, instead allowing the most effective binding site (even previously unknown allosteric sites) to be targeted. An early example is inhibition of Myc, a basic helix-loop-helix leucine zipper (bHLHZip) transcription factor that has been identiﬁed as an oncogenic effector, involved in inducing lymphoid tumors [19]. Through dimerization with Max, another

9.2 High-Throughput Screening of Chemical Libraries

bHLHZip protein, its transcriptional activity drives cell proliferation, stimulates angiogenesis, and represses differentiation. The Myc/Max protein–protein interaction has therefore become a potent target for the development of small molecule inhibitors. Using a ﬂuorescence-based high-throughput screen, a library of 7000 peptidomimetic compounds was assayed for inhibitors of Myc/Max dimerization [19]. Four compounds were identiﬁed, the most potent of which, produced 38% dissociation between the protein partners at 25 µM. However, it was found that another leucine zipper, Jun, was also inhibited by the identiﬁed compounds and with an IC50 of only 75 µM. Although these compounds did not show potency high enough to warrant further investigation, the methodology illustrated the potential use of high-throughput screens for the identiﬁcation of small molecule inhibitors of protein–protein interactions. Another well-explored protein–protein interaction with relevance to cancer is the one between the human protein double minute 2 (HDM2) and the tumor suppressor protein p53. The tumor suppressor p53 is a potent transcription factor that controls and regulates cellular response to damage and stress [20]. Intracellular p53 levels are elevated in response to a variety of stress signals, leading to cell cycle arrest or apoptosis, and as such, it is frequently inactivated in a variety of cancers through its dimerization with HDM2. The binding of HDM2 to p53, upregulates p53 degradation, resulting in the suppression of its transcriptional activity (and inhibition of cell cycle arrest and apoptosis). This interaction has been identiﬁed as a potential target for new, broad-based anticancer therapies. There have been two recent reports of HDM2/p53 inhibitors from high-throughput screens. Hoffmann-La Roche, in New Jersey, United States, recently reported a series of tetra-substituted imidazoles (termed Nutlins) that bind HDM2 and inhibit its interaction with p53 (Figure 9.1) [21]. The most potent compound, Nutlin-3, had an IC50 of 90 nM. The mode of inhibition and the binding location of these compounds were identiﬁed by determining the crystal structure of the HDM2/Nutlin-2 complex. It had been postulated that an inhibitor of HDM2/p53 interaction should lead to stabilization and accumulation of p53 protein, resulting in activation of p53-regulated pro-apoptotic genes. Small molecule inhibitors of this interaction were therefore expected to cause cell cycle O

OH Br

O

N

N

Cl

N

O

N

N O

O

N

N O

Br

O Cl

Nutlin-2 Figure 9.1

NH

The structure of Nutlin-2 and Nutlin-3.

Nutlin-3

217

218

9 Protein Interactions with Small Molecules

arrest and apoptosis. This was shown to be the case in a series of in vitro and in vivo experiments and in multiple cellular models [21]. The HDM2/p53 interaction was also targeted by Johnson & Johnson, where a library of 338 000 compounds designed by Directed Diversity software were screened in parallel for binding to HDM2, by monitoring changes in the thermostability of the protein complex using a specially designed instrument [22]. This methodology utilizes ﬂuorescent dyes to monitor protein unfolding as a function of temperature, allowing the detection of compounds binding to target proteins by measuring the resultant increase in thermal stability. The natural agonist from p53 ﬁlls a relatively deep hydrophobic pocket of HMD2 through three side chains, Phe19, Trp23, and Leu26 (Figure 9.2) [23]. A novel series of benzodiazepinediones were identiﬁed as potent inhibitors of the HDM2/p53 interaction, exhibiting a Kd of 67 nM and an IC50 of 420 nM [22]. As with the nutlin series, crystal structures were obtained and used to observe the nature of the inhibitor–protein interaction and utilized to improve agonists. The compounds identiﬁed as inhibitors bind to the same region as this α-helical portion, and insert aromatic or aliphatic moieties into the same hotspot pockets of HDM2. In the case of Nutlin-2, the bromophenyl moiety is observed sitting in

Phe19

Leu26

Trp23

Figure 9.2 Structure of the MDM2 oncoprotein bound to the p53 tumor suppressor transactivation domain. Three key residues on p53 are highlighted.

9.2 High-Throughput Screening of Chemical Libraries

the Trp pocket, while the remaining bromophenyl is occupying the Leu pocket, and the ether side chain is directed toward the Phe pocket. As with the bound peptide, the interaction is largely through nonspeciﬁc Van der Waals contacts with no hydrogen bonds involved. The interaction of HDM2 with p53 produces a more open conformation of HDM2, whereas it appears to close over the small molecule inhibitors resulting in a more concave contact. These structures also clearly demonstrate the remarkable efﬁciency of the inhibitors as compared to the p53 peptide. The inhibitors bind 25% less of the exposed HMD2 surface (804 vs. 1073 A˚ 2 ) than the minimal peptide binding epitope with less than a third of the molecular weight, but manage to achieve a 10-fold higher binding afﬁnity. Protein–protein interaction targets need not be limited to anticancer therapeutics. In fact, host proteins are regularly usurped by viruses and other pathogens as part of the infection process and hence constitute an enticing target for combating infections. This is demonstrated by HIV, with protein–protein interactions heavily involved in all steps of the viral life cycle (Figure 9.3). Maraviroc (UK-427 857) is a recent anti-HIV compound, marketed by Pﬁzer, and discovered by screening their in-house compound library against the protein–protein interaction between the human chemokine receptor CCR5 and

Entry

Replication

Budding Figure 9.3 The lifecycle of HIV: viral entry involves the interaction of the HIV gp120 protein with the human CD4 and CCR5 or CXCR4 proteins. HIV uses the host cell’s machinery to replicate its genetic material

and core proteins, which have been integrated into the host’s DNA. Immature virus particles bud from the host cell, by utilizing host proteins, and go on to infect other cells.

219

220

9 Protein Interactions with Small Molecules

F F

O

NH

N

Figure 9.4

N

N N

The structure of the HIV entry inhibitor, maraviroc.

the HIV envelope protein gp 120 [24]. This interaction is a key mediator of cell–cell fusion and HIV invasion. In order to identify a lead compound, Pﬁzer screened their compound ﬁle against CCR5, using a chemokine radioligand-binding assay. This screen monitors the level of dimerization through the measurement of an 125 I-labeled protein partner, in this case gp120, and its subsequent dissociation on protein–protein interaction disruption. One of the most potent and ligand-efﬁcient (deﬁned as binding afﬁnity per heavy atom) lead compound was optimized for binding potency, antiviral activity, absorption, and pharmacokinetics as well as selectivity against key human targets such as the hERG channel by screening a further 1000 analogs. This resulted in maraviroc (Figure 9.4) with a low nanomolar IC90 potency and broad-spectrum antiviral activity. Through this development process, Pﬁzer developed and optimized two bespoke assays, one measuring envelope binding to cell surface receptors and a second modeling subsequent membrane fusion events. Importantly there was no unrelated activity observed, especially with CCR2, which has closest sequence identity to CCR5.

9.3 High-Throughput Screening of Biosynthesized Libraries

In contrast to chemical libraries, biologically synthesized libraries of small molecules are often several-fold larger in size and allow for very straightforward identiﬁcation of the active members [25–27]. When combined with an in vivo screen, biological libraries become part of a powerful, rapid, and facile method for the screening of a large number of compounds against a chosen target [11, 27–29]. Biological libraries are principally polypeptides, typically embedded within, or fused to larger molecules in order to overcome the host cell’s catabolic machinery. Alternatively, intracellular stability is achieved by constraining the ends of the peptide with noncovalent and covalent interactions, with varying results [30, 31]. We have pursued intracellular backbone cyclization to generate biostable peptide libraries [32]. This procedure, termed SICLOPPS [33, 34] utilizes the Synechocystis sp. PCC6803 DnaE split intein [35] By rearranging the order of the elements of the

9.3 High-Throughput Screening of Biosynthesized Libraries

O Active intein

O

O HZ N H

O Thioester

Figure 9.5 SICLOPPS mechanism: the expressed fusion protein folds to form an active intein. An N- to S-acyl shift at the target N-terminal intein junction produces a thioester, which undergoes transesteriﬁcation with a side chain nucleophile (serine or

IN IC

IC

IN

H2N

H N

T

H2N

H N

Z O N H

O Lariat

O

T

O

HN HZ

T

et peptide arg

N H

S NH2

NH2 O

et peptide arg

O

O HZ

T

et peptide arg

H2N

H N

N H

SH

O

O

O et peptide arg

O

O

IC

IN

HS

O Cyclic peptide

cysteine; X = O or S) at the C-terminal intein junction to form a lariat intermediate. An asparagine side chain liberates the cyclic product as a lactone, and an X- to N-acyl shift generates the thermodynamically favored lactam product in vivo.

intein, an active cis-intein (IC : target peptide : IN ) is yielded, which upon splicing results in cyclization of the target protein/peptide sequence (Figure 9.5). In order to utilize SICLOPPS for the biosynthesis of cyclic peptide libraries [33], degenerate oligonucleotides encoding those peptides are introduced between the IC and IN genes, while making sure that the correct reading frame is maintained throughout, using a PCR-based method [32]. The variable segment is encoded in the form NNS where N represents any of the four DNA bases (A, C, G, or T) and S represents C or G. The NNS sequence generates 32 codons and encodes all 20 amino acids while eliminating the ocher (UAA) and opal (UGA) stop codons from the library. The intein chemistry requires the ﬁrst amino acid to be a nucleophilic cysteine or serine. There are no limits on the number of amino acids in the target peptide, allowing cyclic peptides of various sizes. We typically use ﬁve variable amino acids because the theoretical number of library members at DNA levels (34 million) is within the number of transformants that can be readily achieved. We have combined SICLOPPS with homodimeric and heterodimeric bacterial reverse two-hybrid systems (RTHSs) [28, 29]. Our RTHS is based on the bacteriophage regulatory system [36], using chimeric repressor fusions and promoter sequences to link the disruption of targeted fusion protein heterodimers to the expression of three reporter genes (Figure 9.6). The homodimeric RTHS uses the bacteriophage 434 DNA-binding protein, and the heterodimeric RTHS uses a mutant P22.434 and a wild-type 434 DNA-binding protein. HIS3 (imidazole glycerol phosphate dehydratase) [37] and KanR (aminoglycoside 3 -phosphotransferase for kanamycin resistance) are two chemically tunable, conditionally selective reporter genes. The third reporter gene, LacZ (β-galactosidase) is used to quantify the protein–protein interaction through β-galactosidase assays. Combining the aforementioned RTHS with SICLOPPS yields a powerful method in which each bacterial cell is utilized for the intracellular synthesis, and screening of a small molecule against the targeted protein–protein interaction; reporting the afﬁnity and selectivity of potential inhibitors. This approach differs from conventional approaches to drug discovery, as traditionally, compound libraries are screened for in vitro activity with the aim of demonstrating in vivo selectivity

221

222

9 Protein Interactions with Small Molecules

Y

X

Y

X

No transcription 434

P22

434

P22

434

P22

434

P22

HIS3

Kan

LacZ

Kan

LacZ

(a)

434

X

Y

434

P22

P22

434

Transcription

P22

HIS3

(b)

Figure 9.6 Heterodimeric RTHS. (a) The target proteins are expressed as fusions with the 434 or P22.434 bacteriophage DNA-binding proteins, which associate to form a functional repressor that prevents expression of the reporter genes, inhibiting growth on minimal media. (b) A small

molecule inhibits the protein–protein interaction of the target proteins (p6 and UEV), allowing transcription and translation of the reporter genes that rescue growth by induction of HIS3 and KanR . LacZ is used to quantify repression by β-galactosidase assays.

at a later stage. In our genetic selection system, inhibitors were selected on the basis of in vivo activity and screened for in vivo target selectivity intracellularly. The mode of action and site of binding are determined for the most active compounds afterward. We have used this approach to uncover inhibitors of several protein–protein interactions, two of which are outlined below. 9.3.1 Cyclic Peptide Inhibitors of AICAR Transformylase Activity

The de novo purine biosynthetic pathway [29], used by virtually all organisms to produce purine nucleotides consists of 10 enzymatic reactions that convert 5-phosphoribosyl-1-pyrophosphate to inosine monophosphate (IMP) [38]. The ﬁnal two steps of this pathway (Figure 9.7) are catalyzed by the product of the purH gene 5-amino-4-imidazolecarboxamide ribonucleotide transformylase/inosine 5 -monophosphate cyclohydrolase (ATIC), a 64-kDa bifunctional protein possessing two distinct domains [39]. The C-terminal aminoimidazole-4-carboxamide ribonucleotide (AICAR) transformylase domain (residues 200–593) catalyzes the transfer of a formyl group from N10 -formyl-tetrahydrofolate (10-f-THF) to AICAR. The ﬁnal step of the pathway (intramolecular cyclization of 5-formyl-AICAR) is catalyzed by the N-terminal IMP cyclohydrolase domain (residues 1–199). The individual AICAR transformylase and IMP cyclohydrolase domains have been

9.3 High-Throughput Screening of Biosynthesized Libraries O N 2-

O4P O

N

HO OH AICAR

Figure 9.7

223

O N

NH2 10-f-THF

THF

NH2

2-

O4P O

AICAR transformylase

N

HO OH

O NH2

N H

N H2O

O

2-

O4P O

IMP cyclohydrolase

N

HO OH

FAICAR

IMP

Final two steps of the de novo purine biosynthesis pathway catalyzed by ATIC.

shown to be active when expressed separately [40–42]. ATIC is highly conserved from Escherichia coli to humans. Of the two pathways available for purine synthesis, normal cells favor the salvage pathway whereas cancer cells rely on the de novo pathway [43]. Inhibiting enzymes of the de novo pathway therefore represents an attractive approach in the search for anticancer agents. Current approaches rely on folate analogs such as methotrexate, an inhibitor of dihydrofolate reductase that has been in clinical use for over 50 years [44]. Owing to its low sequence homology with other folate-dependent enzymes, speciﬁc antifolate inhibitors of ATIC have been relatively scarce [45]. As well as potential uses in the treatment of malignant diseases, ATIC inhibitors may have uses in the treatment of inﬂammatory diseases such as rheumatoid arthritis. Inhibition of ATIC and the resultant increase in intracellular AICAR levels also leads to the inhibition of adenosine deaminase and adenosine kinase. The subsequent build up of adenosine is in turn, thought to be responsible for the anti-inﬂammatory response of several nonsteroid ATIC inhibitors [46–49]. The crystal structure of ATIC [41] shows the dimer to have an interacting interface of ∼5000 A˚ 2 , demonstrating the magnitude of the challenge involved in disrupting this interaction with a small molecule. We used our genetic selection methodology outlined above, to screen two SICLOPPS libraries of around a hundred million members for inhibitors of ATIC homodimerization. The screen identiﬁed around a hundred sequences, which were ranked for their activity. The two most active cyclic peptides were chemically synthesized for testing in vitro (AICAR transformylase assay). The most potent cyclic peptide was found to have a Ki of 17 µM and we determined its mode of action by progress curve analysis [50–52]. The cyclic peptide showed competitive inhibition of enzyme dimerization with respect to 10-f-THF and noncompetitive inhibition with respect to AICAR, which was consistent with both the ordered binding observed for the enzyme and stabilization of the catalytic dimer by 10-f-THF [53, 54]. Furthermore, the Ki of c-1a obtained by this method (18 µM) closely matches that obtained assuming competitive inhibition using a linear plot. This work demonstrated the potential of our genetic selection methodology, by uncovering a six-membered cyclic peptide, capable of disrupting a protein–protein interaction with an interface of 5000 A˚ 2 . Whether the selected inhibitors prevent dimerization by binding to a hot spot [55] in the ATIC protein interface or elsewhere remains unclear. Studies are currently underway in our laboratory to develop the

NH N

224

9 Protein Interactions with Small Molecules

next generation of inhibitors, and to determine the exact binding site of the inhibitor. 9.3.2 Cyclic Peptide Inhibitors of HIV Budding

Another area of interest for our group is studying, and disrupting the protein–protein interactions involved in viral life cycles [56]. In the case of HIV, the poor ﬁdelity of genomic replication is advantageous to the virus, allowing the evolution of a number of drug-resistant mutants. Around 40% of infected individuals are estimated to harbor drug-resistant strains of HIV, with a further 5% exhibiting complete resistance to reverse transcriptase and protease inhibitor drugs. Recent advances in molecular research have allowed better understanding of the intricate interplay between viral and host proteins during the HIV life cycle, uncovering new points of possible intervention less susceptible to acquired resistance. A number of retroviruses including HIV, Ebola, and human T-cell leukemia virus type 1 contain a highly conserved P(T/S)AP tetrapeptide motif in their late budding domain, thought to be essential for viral budding. In the case of HIV the PTAP motif is found in the p6 region of Gag, the protein that drives virus assembly and release. P6 has been shown to speciﬁcally bind the UEV (Ubiquitin E2 variant) domain of TSG101 [57], a human protein residing on the late endosome, involved in the formation of multivesicular bodies. TSG101 activity is normally mediated by binding to the endosomal protein HRS (hepatocyte growth factor-regulated tyrosine kinase substrate), through a PSAP motif [58]. The HIV p6–TSG101 interaction occurs through a competing PTAP motif. Upon binding to p6, TSG101 is diverted from its normal site of action to the plasma membrane, a process that is vital to the initiation of viral budding [59, 60]. Current HIV therapies target the entry and replication steps (Figure 9.3) in the viral life cycle, with no drugs currently on the market that target HIV budding. We saw great potential in using our genetic selection methodology to uncover a cyclic peptide that disrupts the interaction of the HIV p6, with the human TSG101 protein. We therefore constructed a heterodimeric RTHS (Figure 9.6) to assay SICLOPPS libraries for inhibitors of the p6–UEV protein–protein interaction. We used a plasmid library coding for SGWXXXXX (X = any amino acid) cyclic peptides. The invariable motif of the peptide was designed to contain serine (required nucleophile for intein processing), glycine (avoids racemization during chemical synthesis), and tryptophan, which functions as a chromophore for HPLC puriﬁcation. The library size recovered by electroporation into the selection strain was 1 × 108 transformants. After several rounds of secondary screening, three cyclic peptides remained, with SGWIYWNV being the most active in the p6–UEV RTHS (Figure 9.8). We also constructed and screened a SGWXXPXXPXX library, to target the PTAP binding motif on TSG101. Selectants from this library were not carried forward as they were less active than those from the fully randomized library; this highlights the power and advantage of genetic selection over conventional methodologies.

9.3 High-Throughput Screening of Biosynthesized Libraries

NH O

O N H O HO

HN

NH

HN

NH O

NH

O

OH

H N O

O H2N O

Figure 9.8

O HN

N H

Cyclic peptide SGWIYWNV.

The conventional approach would be to target the p6–UEV interaction with PTAP mimics, artiﬁcially limiting the available inhibitor binding sites. Genetic selection however, allows the whole surface area of the two proteins to be assayed, uncovering the most effective inhibitors from the library, regardless of binding site. The key step for us was to demonstrate that the inhibitor of the p6–UEV interaction uncovered using our genetic selection methodology, also inhibited HIV budding from host cells. The assay used monitors viral budding from human 293T cells transfected with a GFP-tagged Gag construct (Fugene transfection), which causes viruslike particle release [61]. This is based on the fact that the HIV Gag protein is itself sufﬁcient to bud from the cell surface and give rise to particles that are morphologically similar to genuine HIV viral particles [60, 61]. The intracellular delivery of the cyclic peptides was enhanced by utilizing a cell penetrating peptide sequence derived from the HIV Tat internalization domain [62, 63]. The arginine-rich Tat sequence was synthetically bound to the cyclic peptides via a disulﬁde bond, formed by modifying the nonrandom region of the cyclic peptide from SGW to CGW and by adding a cysteine to the beginning of the Tat sequence. The Tat-tagged cyclic peptide (Figure 9.8) was found to inhibit Gag VLP release by more than 60% when administered to the 293T cells at 10 µM. In comparison, the Tat peptide alone (negative control) had no effect at the same concentration. The inhibitory effect of the cyclic peptide on Gag budding was found to be dose-dependent, and speciﬁc to VLP release at all concentrations, with an IC50 of 7 µM. We also assayed the cyclic peptide for its effect on the physiological function of TSG101. Epidermal growth factor receptor (EGFR) degradation is a process that is regulated by, and dependant on the TSG101–HRS interaction. We measured the effect of Tat-tagged cyclic peptide 11 on EFGR degradation in HeLa cells [58], and found that it does not affect this process. The collective data suggests that the cyclic peptide inhibitor acts by inhibiting dimerization of TSG101 and HIV p6, thus inhibiting HIV budding from 293T cells. This compound promises the possibility of targeting the viral budding of HIV,

225

226

9 Protein Interactions with Small Molecules

via the p6–TSG101 interaction. As this process relies on host proteins, it is less likely to be susceptible to acquired resistance than current drugs that only target the viral proteome. We are currently attempting to evolve more potent inhibitors using second-generation SICLOPPS libraries (based on the selected sequences) and peptidomimetics [64]. The isolation of the active cyclic peptide from a library of ∼108 members demonstrates the power of genetic selection and the premise of controlling cellular processes through modulation of speciﬁc protein–protein interactions. The methodology used allows rapid identiﬁcation of small molecule inhibitors of other protein–protein interactions, yielding a powerful and novel approach to drug discovery.

9.4 Future Direction

The majority of protein interaction networks contain weak spots: protein–protein interactions that are critical to a cellular process and yet are not ‘‘backed up’’ by other protein–protein interactions that would take over in case of malfunction (genetic mutations, etc.). Thus, small molecules that target such weak points have great potential to form the next generation of therapeutic agents. In the following sections we outline three areas in which inhibitors of the given protein–protein interactions would represent a major step forward from current therapies. 9.4.1 Small Molecule Inhibitors of Tumor Hypoxia Response Network

As tumors grow, they are constantly outstripping the supply of oxygen and nutrients that are delivered by the vascular network. Growth of primary and metastatic tumors is therefore dependant on angiogenesis, the development of new blood vessels from an existing vascular network, delivering oxygen and nutrients necessary for growth. The cellular hypoxia (deprivation of oxygen) response network is therefore a major driver of tumor progression. Hypoxic areas form when the growth of a tumor outstrips local neovascularization, thereby creating areas of inadequate perfusion. HIF-1 is the most potent inducer of the expression of genes coding for glycolytic enzymes, vascular endothelial growth factor (VGEF) and erythropoietin [65]. HIF-1 is a heterodimer that consists of HIF-1α whose cellular concentration is tightly regulated by oxygen, and HIF-1β, which is constitutively expressed in the nucleus. In the absence of oxygen, HIF-1α forms an active complex with HIF-1β that binds to hypoxia response element, thereby activating the expression of numerous hypoxia response genes (Figure 9.9). The potency of HIF-1 activity in tumors and its correlation to angiogenesis, tumor growth, and metastasis is well established [66, 67]. Immunohistochemical analysis of human tumor biopsies has revealed dramatic overexpression of HIF-1α in common cancers [68], which is associated with the regulation of genes involved

9.4 Future Direction

Normoxia

227

Hypoxia

O2

PHD2

OH OH

PHD2

OH

HIF-1a

HIF-1a

Ub VHL

Ub

VHL

HIF-1a Proteasome degradation

HIF-1a

HIF-1a Stabilization

HIF-1a

HIF-1a

HIF-1b

HIF-1a mRNA Expression of target genes Figure 9.9 The hypoxia response network. In normoxia, HIF-1α is hydroxylated by prolyl hydroxylases (PHDs), allowing the von Hipple–Lindau (VHL) protein to bind, leading to its ubiquitination and degradation. Under hypoxic conditions, the oxygen substrate for PHDs is no longer present,

resulting in stabilization and build up of HIF-1α, which translocates to the nucleus and heterodimerizes with HIF-1β, to form a transcription activator of pro-angiogenic genes, which stimulate new blood vessel formation.

in angiogenesis such as VEGF. Clinically, HIF-1α overexpression has been shown to be a marker of highly aggressive disease and is associated with poor prognosis and treatment failure in a number of cancers [69, 70]. Hypoxic cells convert to a glycolytic metabolism, become resistant to apoptosis, and are more likely to migrate to less hypoxic areas of the body (metastasis). Furthermore, the loss or gain of HIF-1 activity is negatively or positively correlated with tumor growth and angiogenesis [71, 72].

228

9 Protein Interactions with Small Molecules

A small molecule inhibitor of HIF-1 heterodimerization, could potentially act as a very potent anticancer agent, inhibiting angiogenesis in tumors and preventing their growth. We are currently in the process of uncovering such agents using our genetic selection methodology, outlined previously. 9.4.2 Targeting Protein–Protein Interactions in Asthma

Asthma is a chronic disease of the lung where there is an inappropriate response to an innocuous substance. These allergens can be harmless, everyday molecules that when inhaled by an asthmatic subject, trigger an immediate and acute response by their immune system, characterized by inﬂammation and contraction of the muscles surrounding the airways. This intense inﬂammation damages the lining of the lung and over time leads to profound structural changes known as remodeling. At the root of this process (as with most other cellular processes) lie a series of protein–protein interactions that play a key role in disease development and progress. Such interactions are upstream of the majority of current drug targets, and are potentially ideal points for combating disease. It is becoming increasingly evident that many of these interactions cannot be targeted with conventional druglike molecules [73]. The promise of this approach is demonstrated by the current excitement surrounding therapeutic antibodies that target extracellular protein interactions (e.g., inﬂiximab) [74]. Although highly speciﬁc for their molecular targets and stable in human serum, antibodies are difﬁcult and expensive to manufacture and not cell permeable. Small molecule inhibitors that bind protein surfaces with antibody-like afﬁnity and selectivity offer the potential to overcome such issues [75]. An important aspect of allergic inﬂammation in asthma is the migration of Th2 cells and eosinophils into the airway. Th2 cells support the inﬂux of eosinophils via cytokine secretion, and are sources of interleukin-4 (IL-4) and IL-13 that have been shown to induce chemokine production in various airway cells [76–78]. IL-4 and IL-13 are two key cytokines associated with airway inﬂammation and airﬂow obstruction in asthma [79, 80]. The predominant biological effects of both IL-4 and IL-13 are driven through a heterodimeric receptor complex composed of IL-4 receptor α (IL-4Rα) and IL-13Rα1 subunits, or the IL-4Rα and γ C receptor complex (Figure 9.10). IL-4/IL-13 binding to either receptor complex instigates an intracellular signaling cascade resulting in the homodimerization of signal transducer and activator of rranscription factor 6 (STAT-6) (Figure 9.10) [81]. This is thought to be critical for Th2 differentiation of T helper cells and essential for isotype class switching to IgE. STAT-6 expression has furthermore shown to be required for the development of experimental allergic asthma in mice [81, 82]. IL-4 and IL-13 are proposed to play key roles in the pathogenesis of severe asthma, through modulating airway hyperresponsiveness, subepithelial ﬁbrosis, and mucus production [83]. There has been considerable effort directed toward therapies based on the inhibition of either IL-4 or IL-13 by soluble receptor proteins or antibodies

9.4 Future Direction

TYK2

JAK1

P

P

P Homodimerization

P

STAT-6

P STAT-6

STAT-6

STAT-6

P

STAT-6

STAT-6

P

IL-13

IL-4Ra

P

IL-13Ra1

JAK1 IL-4Ra

JAK3

YC

P

or

IL-4

IL-4

P Inflammation and fibrosis

Figure 9.10 The IL-4 and IL-13 signaling pathways in asthma. Binding of the interleukins to the receptor complex activates phosphorylation of janus kinase (Jak) 1 and 3, and tyrosine kinase (Tyk) 2. These phosphorylated sites recruit STAT-6, which

leads to its phosphorylation and subsequent dimerization. STAT-6 dimers translocate to the nucleus, where they activate transcription of a number of target genes that lead to inﬂammation and ﬁbrosis.

[84]. The shared receptor complex of IL-4Rα and IL-13Rα1, however, leads to many overlapping functions and thus renders single-target therapies largely ineffective. Previous work has shown that the expression of STAT-6 in the bronchial epithelium is higher in subjects with severe asthma than in subjects with mild asthma or normal controls [85]. Thus, this presents an ideal therapeutic target for the treatment of asthma. The ability to develop small molecule antagonists has signiﬁcant advantages over the current therapies, which are expensive to manufacture and require either repeated hospital visits or daily injections. As both cytokines are thought to predominantly instigate their intracellular signaling cascade via activation of STAT-6, this seems an ideal point for intervention. Our laboratory is currently studying the role of these protein–protein interactions in asthma, and developing small molecule inhibitors.

229

230

9 Protein Interactions with Small Molecules

9.4.3 Targeting the Protein Interaction Networks of Inﬂuenza Virus

In humans, inﬂuenza viruses are common and highly contagious pathogens of the upper respiratory tract, and seasonal epidemics affect 10–20% of the general population. However, the virus can also be deadly, with pandemic epidemics that killed up to a million people worldwide. All human inﬂuenza viruses belong to the family Orthomyxoviridae. There are three genera, corresponding to inﬂuenza types A, B, and C. Inﬂuenza A viruses, uniquely, are divided into subtypes based on major antigenic speciﬁcities of their hemagglutinin (HA) and neuraminidase (NA) proteins. Type A is the most important of the inﬂuenza viruses, causing alternate annual outbreaks and epidemics during the winter seasons. Pandemics of inﬂuenza A have occurred about three times per century since 1700 and were manifested by the worldwide spread of the disease, typically with high morbidity and mortality. The most extreme pandemic in the twentieth century was that of Spanish inﬂuenza of 1918–1919, which killed an estimated 20–40 million people throughout the world [86]. Emergence of new inﬂuenza strains in the human population occurs via transmission from other animal species, especially poultry. Typically, human and avian inﬂuenza viruses are different with little cross-species infection. Occasionally, direct avian–human transmission can occur, often with enhanced pathogenicity, as demonstrated by the emergence of the recent H5N1 avian inﬂuenza in many countries throughout Southeast Asia; human-to-human spread of H5N1 avian inﬂuenza has not been conclusively documented [87]. The genome of inﬂuenza A viruses consists of eight separate segments of single-stranded, negative-sense RNA encoding for viral RNA polymerase subunits and structural and functional proteins. Mutations resulting from the error-prone replication of the single-stranded RNA, and an ability to produce genetic hybrid viruses by the mixing of gene segments from two viruses (when dual infection occurs), contribute to the evolutionary success of these viruses and the difﬁculty associated with the development of anti-inﬂuenza therapies. Inﬂuenza viruses, like other enveloped viruses, use a membrane fusion strategy to deliver their genomes to the cytosol of target cells. Receptor-binding and fusion are mediated by the viral HA. HA binds to sialic acid (SA) residues on glycoproteins or glycolipids on the host cell surface. The speciﬁcity of the HA determines the tropism of the virus with respect to species and target cells [88]. One potential approach to antiviral therapy involves the inhibition of virus/host cell binding via blocking of the viral receptor. Ideal preventative agents must be nontoxic to mammalian cells, broadly speciﬁc across viral subtypes, and cost effective. Previous reports have demonstrated that millimolar concentrations of solubilized monomeric SA prevent inﬂuenza A agglutination of chicken erythrocytes (cRBC), presumably by binding to the HA molecule on the virus. However, each inﬂuenza virion uses about 300–600 envelope HA spikes to develop strong polyvalent binding to host cell receptors, while the binding of soluble monomeric SA derivative to HA is very weak. Thus, nonphysiological and potentially toxic concentrations of soluble monomeric SA are

9.4 Future Direction

required to competitively disrupt this polyvalent interaction and prevent viral internalization into the cells. To circumvent these problems, multiple sialyl moieties were conjugated to a synthetic polymer backbone to provide a polyvalent inhibitor. Even if ‘‘polymeric inhibitors’’ based on polyacrylamide backbones are more effective than monomeric SA at inhibiting inﬂuenza-induced agglutination of red blood cells, they were toxic to mammalian cells [89]. Other works have focused on the acidiﬁcation of the endosome compartment: it is generally accepted that the M2 proton channel is inhibited by amantadine and its derivate rimantadine [90, 91]. Both compounds appear to be of prophylactic activity against inﬂuenza A, and may be important in the treatment of immunocompromised and elderly patients. They show, however, the following disadvantages: induction of viral resistance and exhibition of adverse effects on the central nervous and gastrointestinal system [92]. Fusion of inﬂuenza virus from within endosomes is inhibited by agents that raise the intraendosomal pH, such as NH4 Cl, monensin, or chloroquine or those that could bind selectively to a pocket of the HA2 chain, thereby stabilizing its nonfusogenic conformation and inhibiting the exposure of the fusion peptide [92, 93]. Inﬂuenza viruses replicate and transcribe their segmented negative-sense single-stranded RNA genome in the nucleus of the infected host cells, processes that are performed by the virally encoded RNA-dependent RNA polymerase, which is a trimeric complex composed by PB1, PB2, and PA [94]. PB1 is the central protein, containing to different domains interacting with PB2 and PA. In a recent work, Ghanem et al. [95] have shown that peptides that speciﬁcally bind to the protein–protein interaction domains in these subunits interfere with polymerase complex assembly and viral replication. In the host nucleus, the viral primary transcription produces proteins necessary for replication. Viral transcription is however, critically dependent on on-going cellular transcription, in particular, on the activities associated with the cellular DNA-dependent RNA polymerase II (Pol II) [96]. Thus if cellular Pol II is blocked by speciﬁc inhibitors, no viral mRNA can be synthesized [97]. The viral polymerase complex is not the only viral component shown to interact with the host Pol II transcription machinery. Many viruses encode proteins that modify the host transcription or translation apparatus to favor the synthesis of viral proteins over those of the host cell. As a result, the synthetic capability of the host cell is directed principally to the production of new virus particles [96]. The viral NS1, a multifunctional 26-kDa protein, was reported to inhibit host immune responses [98]. NS1 binds and sequesters dsRNA, interfering with the host mRNA processing and facilitating the preferential viral mRNA translation. The combination of all these functions makes the NS1 protein a very potent inhibitor of immunity and allows the inﬂuenza virus to efﬁciently escape the immune surveillance and to establish infection in the host. It is however clear that NS1 also acts directly to modulate other important aspects of the virus replication cycle, including viral RNA replication, viral protein synthesis, and general host cell physiology. Two important domains have been described in NS1 accomplishing its multiple functions: the N-terminal RNA-binding domain (residues 1–73) and the C-terminal effector domain (residues 74–230), which predominantly mediates interactions

231

232

9 Protein Interactions with Small Molecules

with host cell proteins, and also stabilizes the RNA-binding domain [99]. The presence of different domains, the resolution of its crystal structure, and the discovery of its interaction with host components provide several potential targets for the development of antiviral compounds. Given the numerous roles of NS1 during virus replication, one potential target for anti-inﬂuenza drug design may be to disrupt conserved interactions of NS1 with RNA or its protein–protein interaction with cellular and viral factors. In this regard, peptide-mediated inhibition of the NS1–CPSF30 interaction has recently been described as a ‘‘proof-of-principle’’ approach to limit virus replication in tissue culture. Unfortunately, such a virus-speciﬁc strategy allows for virus mutation and the development of drug resistance [100]. Other works have been focused on the inhibition of NS1 binding to vRNA [101]. A recent report outlined the use of a yeast-based assay to identify compounds that phenotypically suppress NS1 function [102]. Interestingly, in this study, cells lacking an interferon response were drug resistant, suggesting that the compounds block interactions between NS1 and the interferon system. The effects of the compounds were speciﬁc to NS1, because they had no effect on the ability of the severe acute respiratory syndrome coronavirus papain-like protease protein to block β-interferon promoter activation. These recent ﬁndings suggest that function of NS1 can be modulated by chemical inhibitors and that such inhibitors will be useful probes of biological function, and starting points for development of anti-inﬂuenza therapies [102]. A number of other host factors have also been shown to interact with inﬂuenza viral components, and for some of these interactions, there is evidence that they have functional signiﬁcance during the viral life cycle. For others, however, much needs to be learned before we can fully appreciate their role in the infected cell for either the viral RNA synthesis machinery or for the virus’ attempts at thwarting normal cellular functions. The development of proteomic tools in recent years will certainly aid our attempts to identify some of these physical interactions. The greatest challenge, however, remains in establishing what, if any, role a newly discovered interacting partner plays in the inﬂuenza life cycle and to selective and speciﬁc inhibitors of their function. Acknowledgments

A.T. would like to thank Prof. Stephen Benkovic for his guidance and support. We would also like to thank Prof. Stanley Cohen, Prof. Alex Horswill, Prof. Sergey Savinov, Dr Todd Naumann, and Prof. Quan Lu. References 1. Arkin, M.R. and Wells, J.A.

(2004) Small-molecule inhibitors of protein-protein interactions: progressing towards the dream. Nat. Rev. Drug Discov., 3, 301–317.

2. Wells, J.A. and McClendon, C.L.

(2007) Reaching for high-hanging fruit in drug discovery at proteinprotein interfaces. Nature, 450, 1001–1009.

References 3. Young, L., Jernigan, R.L., and Covell,

4.

5.

6.

7.

8.

9.

10.

11.

12.

D.G. (1994) A role for surface hydrophobicity in protein-protein recognition. Protein Sci., 3, 717–729. Moreira, I.S., Fernandes, P.A., and Ramos, M.J. (2007) Hot spots–a review of the protein-protein interface determinant amino-acid residues. Proteins, 68, 803–812. Gao, Y., Dickerson, J.B., Guo, F., Zheng, J., and Zheng, Y. (2004) Rational design and characterization of a Rac GTPase-speciﬁc small molecule inhibitor. Proc. Natl. Acad. Sci. U.S.A., 101, 7618–7623. Trosset, J.Y., Dalvit, C., Knapp, S., Fasolini, M., Veronesi, M., Mantegani, S., Gianellini, L.M., Catana, C., Sundstrom, M., Stouten, P.F., and Moll, J.K. (2006) Inhibition of protein-protein interactions: the discovery of druglike beta-catenin inhibitors by combining virtual and biophysical screening. Proteins, 64, 60–67. Enyedy, I.J., Ling, Y., Nacro, K., Tomita, Y., Wu, X., Cao, Y., Guo, R., Li, B., Zhu, X., Huang, Y., Long, Y.Q., Roller, P.P., Yang, D., and Wang, S. (2001) Discovery of small-molecule inhibitors of Bcl-2 through structure-based computer screening. J. Med. Chem., 44, 4313–4324. Boldt, G.E., Dickerson, T.J., and Janda, K.D. (2006) Emerging chemical and biological approaches for the preparation of discovery libraries. Drug Discov. Today, 11, 143–148. Borghouts, C., Kunz, C., and Groner, B. (2005) Peptide aptamers: recent developments for cancer therapy. Expert Opin. Biol. Ther., 5, 783–797. Chang, K., Elledge, S.J., and Hannon, G.J. (2006) Lessons from nature: microRNA-based shRNA libraries. Nat. Methods, 3, 707–714. Matsumoto, S., Akashi, H., and Taira, K. (2006) Screening and determination of gene function using randomized ribozyme and siRNA. libraries Handb. Exp. Pharmacol., 173, 197–221. Rothe, A., Hosse, R.J., and Power, B.E. (2006) Ribosome display for improved

13.

14.

15. 16.

17.

18.

19.

20.

21.

22.

biotherapeutic molecules. Expert Opin. Biol. Ther., 6, 177–187. Baldwin, J.J. (1996) Design, synthesis and use of binary encoded synthetic chemical libraries. Mol. Divers., 2, 81–88. Edwards, P.J. and Morrell, A.I. (2002) Solid-phase compound library synthesis in drug design and development. Curr. Opin. Drug Discov. Dev., 5, 594–605. Huwe, C.M. (2006) Synthetic library design. Drug Discov. Today, 11, 763–767. Tan, D.S. (2005) Diversity-oriented synthesis: exploring the intersections between chemistry and biology. Nat. Chem. Biol., 1, 74–84. Webb, T.R. (2005) Current directions in the evolution of compound libraries. Curr. Opin. Drug Discov. Dev., 8, 303–308. Irwin, J.J. (2006) How good is your screening library? Curr. Opin. Chem. Biol., 10, 352–356. Berg, T., Cohen, S.B., Desharnais, J., Sonderegger, C., Maslyar, D.J., Goldberg, J., Boger, D.L., and Vogt, P.K. (2002) Small-molecule antagonists of Myc/Max dimerization inhibit Myc-induced transformation of chicken embryo ﬁbroblasts. Proc. Natl. Acad. Sci. U.S.A., 99, 3830–3835. Vogelstein, B., Lane, D., and Levine, A.J. (2000) Surﬁng the p53 network. Nature, 408, 307–310. Vassilev, L.T., Vu, B.T., Graves, B., Carvajal, D., Podlaski, F., Filipovic, Z., Kong, N., Kammlott, U., Lukacs, C., Klein, C., Fotouhi, N., and Liu, E.A. (2004) In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science, 303, 844–848. Grasberger, B.L., Lu, T., Schubert, C., Parks, D.J., Carver, T.E., Koblish, H.K., Cummings, M.D., LaFrance, L.V., Milkiewicz, K.L., Calvo, R.R., Maguire, D., Lattanze, J., Franks, C.F., Zhao, S., Ramachandren, K., Bylebyl, G.R., Zhang, M., Manthey, C.L., Petrella, E.C., Pantoliano, M.W., Deckman, I.C., Spurlino, J.C., Maroney, A.C., Tomczuk, B.E., Molloy, C.J., and Bone, R.F. (2005) Discovery and cocrystal structure of benzodiazepinedione

233

234

9 Protein Interactions with Small Molecules

23.

24.

25.

26.

27.

28.

29.

30.

31.

HDM2 antagonists that activate p53 in cells. J. Med. Chem., 48, 909–912. Kussie, P.H., Gorina, S., Marechal, V., Elenbaas, B., Moreau, J., Levine, A.J., and Pavletich, N.P. (1996) Structure of the MDM2 oncoprotein bound to the p53 tumour suppressor transactivation domain. Science, 274, 948–953. Dorr, P., Westby, M., Dobbs, S., Grifﬁn, P., Irvine, B., Macartney, M., Mori, J., Rickett, G., Smith-Burchnell, C., Napier, C., Webster, R., Armour, D., Price, D., Stammen, B., Wood, A., and Perros, M. (2005) Maraviroc (UK-427,857), a potent, orally bioavailable, and selective small-molecule inhibitor of chemokine receptor CCR5 with broad-spectrum anti-human immunodeﬁciency virus type 1 activity. Antimicrob. Agents Chemother., 49, 4721–4732. Falciani, C., Lozzi, L., Pini, A., and Bracci, L. (2005) Bioactive peptides from libraries. Chem. Biol., 12, 417–426. Ulrich, H. (2005) DNA and RNA aptamers as modulators of protein function. Med. Chem., 1, 199–208. Uttamchandani, M., Wang, J., and Yao, S.Q. (2006) Protein and small molecule microarrays: powerful tools for high-throughput proteomics. Mol. Biosyst., 2, 58–68. Horswill, A.R., Savinov, S.N., and Benkovic, S.J. (2004) A systematic method for identifying small-molecule modulators of protein-protein interactions. Proc. Natl. Acad. Sci. U.S.A., 101, 15591–15596. Tavassoli, A. and Benkovic, S.J. (2005) Genetically selected cyclic-peptide inhibitors of AICAR transformylase homodimerization. Angew. Chem. Int. Ed. Engl., 44, 2760–2763. Gururaja, T.L., Narasimhamurthy, S., Payan, D.G., and Anderson, D.C. (2000) A novel artiﬁcial loop scaffold for the noncovalent constraint of peptides. Chem. Biol., 7, 515–527. Jermutus, L., Honegger, A., Schwesinger, F., Hanes, J., and Pluckthun, A. (2001) Tailoring in vitro evolution for protein afﬁnity or

32.

33.

34.

35.

36.

37.

38.

39.

40.

stability. Proc. Natl. Acad. Sci. U.S.A., 98, 75–80. Tavassoli, A. and Benkovic, S.J. (2007) Split-intein mediated circular ligation used in the synthesis of cyclic peptide libraries in E. coli. Nat. Protoc., 2, 1126–1133. Scott, C.P., Abel-Santos, E., Jones, A.D., and Benkovic, S.J. (2001) Structural requirements for the biosynthesis of backbone cyclic peptide libraries. Chem. Biol., 8, 801–815. Scott, C.P., Abel-Santos, E., Wall, M., Wahnon, D.C., and Benkovic, S.J. (1999) Production of cyclic peptides and proteins in vivo. Proc. Natl. Acad. Sci. U.S.A., 96, 13638–13643. Wu, H., Hu, Z., and Liu, X.Q. (1998) Protein trans-splicing by a split intein encoded in a split DnaE gene of Synechocystis sp. PCC6803. Proc. Natl. Acad. Sci. U.S.A., 95, 9226–9231. Hu, J.C., O’Shea, E.K., Kim, P.S., and Sauer, R.T. (1990) Sequence requirements for coiled-coils: analysis with lambda repressor-GCN4 leucine zipper fusions. Science, 250, 1400–1403. Joung, J.K., Ramm, E.I., and Pabo, C.O. (2000) A bacterial two-hybrid selection system for studying protein-DNA and protein-protein interactions. Proc. Natl. Acad. Sci. U.S.A., 97, 7382–7387. Zalkin, H. and Dixon, J.E. (1992) De novo purine nucleotide biosynthesis. Prog. Nucleic Acid Res. Mol. Biol., 42, 259–287. Ni, L., Guan, K., Zalkin, H., and Dixon, J.E. (1991) De novo purine nucleotide biosynthesis: cloning, sequencing and expression of a chicken PurH cDNA encoding 5-aminoimidazole-4-carboxamideribonucleotide transformylase-IMP cyclohydrolase. Gene, 106, 197–205. Rayl, E.A., Moroson, B.A., and Beardsley, G.P. (1996) The human purH gene product, 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase. Cloning, sequencing, expression, puriﬁcation, kinetic analysis, and domain mapping. J. Biol. Chem., 271, 2225–2233.

References 41. Greasley, S.E., Horton, P., Ramcharan,

42.

43.

44.

45.

46.

47.

48.

49.

J., Beardsley, G.P., Benkovic, S.J., and Wilson, I.A. (2001) Crystal structure of a bifunctional transformylase and cyclohydrolase enzyme in purine biosynthesis. Nat. Struct. Biol., 8, 402–406. Wolan, D.W., Greasley, S.E., Beardsley, G.P., and Wilson, I.A. (2002) Structural insights into the avian AICAR transformylase mechanism. Biochemistry, 41, 15505–15513. Jackson, R.C. and Harkrader, R.J. (1981) in Nucleosides and Cancer Treatment (eds M.H.N. Tattersall and R.M. Fox), Academic Press, Sydney, pp. 18–31. McGuire, J.J. (2003) Anticancer antifolates: current status and future directions. Curr. Pharm. Des., 9, 2593–2613. Cheong, C.G., Wolan, D.W., Greasley, S.E., Horton, P.A., Beardsley, G.P., and Wilson, I.A. (2004) Crystal structures of human bifunctional enzyme aminoimidazole-4-carboxamide ribonucleotide transformylase/IMP cyclohydrolase in complex with potent sulfonyl-containing antifolates. J. Biol. Chem., 279, 18034–18045. Baggott, J.E., Morgan, S.L., Ha, T., Vaughn, W.H., and Hine, R.J. (1992) Inhibition of folate-dependent enzymes by non-steroidal anti-inﬂammatory drugs. Biochem. J., 282 (Pt 1), 197–202. Allegra, C.J., Drake, J.C., Jolivet, J., and Chabner, B.A. (1985) Inhibition of phosphoribosylaminoimidazolecarboxamide transformylase by methotrexate and dihydrofolic acid polyglutamates. Proc. Natl. Acad. Sci. U.S.A., 82, 4881–4885. Cronstein, B.N., Naime, D., and Ostad, E. (1993) The antiinﬂammatory mechanism of methotrexate. Increased adenosine release at inﬂamed sites diminishes leukocyte accumulation in an in vivo model of inﬂammation. J. Clin. Invest., 92, 2675–2682. Gadangi, P., Longaker, M., Naime, D., Levin, R.I., Recht, P.A., Montesinos, M.C., Buckley, M.T., Carlin, G., and Cronstein, B.N. (1996) The

50.

51.

52.

53.

54.

55.

56.

57.

anti-inﬂammatory mechanism of sulfasalazine is related to adenosine release at inﬂamed sites. J. Immunol., 156, 1937–1941. Kuzmic, P. (1996) Program DYNAFIT for the analysis of enzyme kinetic data: application to HIV proteinase. Anal. Biochem., 237, 260–273. Stone, S.R. and Fromm, H.J. (1980) Studies on the mechanism of adenosine 5 -monophosphate inhibition of bovine liver fructose 1,6-bisphosphatase. Biochemistry, 19, 620–625. Bauer, M., Griengl, H., and Steiner, W. (1999) Kinetic studies on the enzyme (S)-hydroxynitrile lyase from hevea brasiliensis using initial rate methods and progress curve analysis. Biotechnol. Bioeng., 62, 20–29. Vergis, J.M., Bulock, K.G., Fleming, K.G., and Beardsley, G.P. (2001) Human 5-aminoimidazole-4-carboxamide ribonucleotide transformylase/inosine 5 -monophosphate cyclohydrolase. A bifunctional protein requiring dimerization for transformylase activity but not for cyclohydrolase activity. J. Biol. Chem., 276, 7727–7733. Bulock, K.G., Beardsley, G.P., and Anderson, K.S. (2002) The kinetic mechanism of the human bifunctional enzyme ATIC (5-amino-4-imidazolecarboxamide ribonucleotide transformylase/inosine 5 -monophosphate cyclohydrolase). A surprising lack of substrate channeling. J. Biol. Chem., 277, 22168–22174. DeLano, W.L. (2002) Unraveling hot spots in binding interfaces: progress and challenges. Curr. Opin. Struct. Biol., 12, 14–20. Tavassoli, A., Lu, Q., Gam, J., Pan, H., Benkovic, S.J., and Cohen, S.N. (2008) Inhibition of HIV budding by a genetically selected cyclic peptide targeting the Gag-TSG101 interaction. ACS Chem. Biol., 3, 757–764. Pornillos, O., Alam, S.L., Davis, D.R., and Sundquist, W.I. (2002) Structure of the Tsg101 UEV domain in complex with the PTAP motif of the HIV-1 p6 protein. Nat. Struct. Biol., 9, 812–817.

235

236

9 Protein Interactions with Small Molecules 58. Lu, Q., Hope, L.W., Brasch, M.,

59.

60.

61.

62.

63.

64.

65.

66.

67.

Reinhard, C., and Cohen, S.N. (2003) TSG101 interaction with HRS mediates endosomal trafﬁcking and receptor down-regulation. Proc. Natl. Acad. Sci. U.S.A., 100, 7626–7631. Demirov, D.G., Ono, A., Orenstein, J.M., and Freed, E.O. (2002) Overexpression of the N-terminal domain of TSG101 inhibits HIV-1 budding by blocking late domain function. Proc. Natl. Acad. Sci. U.S.A., 99, 955–960. Garrus, J.E., von Schwedler, U.K., Pornillos, O.W., Morham, S.G., Zavitz, K.H., Wang, H.E., Wettstein, D.A., Stray, K.M., Cote, M., Rich, R.L., Myszka, D.G., and Sundquist, W.I. (2001) Tsg101 and the vacuolar protein sorting pathway are essential for HIV-1 budding. Cell, 107, 55–65. Hermida-Matsumoto, L. and Resh, M.D. (2000) Localization of human immunodeﬁciency virus type 1 Gag and Env at the plasma membrane by confocal imaging. J. Virol., 74, 8670–8679. Mae, M. and Langel, U. (2006) Cell-penetrating peptides as vectors for peptide, protein and oligonucleotide delivery. Curr. Opin. Pharmacol., 6, 509–514. Mie, M., Takahashi, F., Funabashi, H., Yanagida, Y., Aizawa, M., and Kobatake, E. (2003) Intracellular delivery of antibodies using TAT fusion protein A. Biochem. Biophys. Res. Commun., 310, 730–734. Andronati, S.A., Karaseva, T.L., and Krysko, A.A. (2004) Peptidomimetics – antagonists of the ﬁbrinogen receptors: molecular design, structures, properties and therapeutic applications. Curr. Med. Chem., 11, 1183–1211. Dery, M.A., Michaud, M.D., and Richard, D.E. (2005) Hypoxia-inducible factor 1: regulation by hypoxic and non-hypoxic activators. Int. J. Biochem. Cell Biol., 37, 535–540. Hewitson, K.S. and Schoﬁeld, C.J. (2004) The HIF pathway as a therapeutic target. Drug Discov. Today, 9, 704–711. Shi, Y.H. and Fang, W.G. (2004) Hypoxia-inducible factor-1 in tumour

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

angiogenesis. World J. Gastroenterol., 10, 1082–1087. Melillo, G. (2006) Inhibiting hypoxia-inducible factor 1 for cancer therapy. Mol. Cancer Res., 4, 601–605. Jensen, R.L. (2006) Hypoxia in the tumourigenesis of gliomas and as a potential target for therapeutic measures. Neurosurg. Focus, 20, E24 . Lauzier, M.C., Michaud, M.D., Dery, M.A., and Richard, D.E. (2006) [HIF-1 activation during tumour progression: implications and consequences]. Bull. Cancer, 93, 349–356. Bardos, J.I. and Ashcroft, M. (2005) Negative and positive regulation of HIF-1: a complex network. Biochim. Biophys. Acta, 1755, 107–120. Escuin, D., Simons, J.W., and Giannakakou, P. (2004) Exploitation of the HIF axis for cancer therapy. Cancer Biol. Ther., 3, 608–611. Hopkins, A.L. and Groom, C.R. (2002) The druggable genome. Nat. Rev. Drug Discov., 1, 727–730. Shealy, D.J. and Visvanathan, S. (2008) Anti-TNF antibodies: lessons from the past, roadmap for the future. Handb. Exp. Pharmacol., 181, 101–129. Millward, S.W., Fiacco, S., Austin, R.J., and Roberts, R.W. (2007) Design of cyclic peptides that bind protein surfaces with antibody-like afﬁnity. ACS Chem. Biol., 2, 625–634. Komiya, A., Nagase, H., Yamada, H., Sekiya, T., Yamaguchi, M., Sano, Y., Hanai, N., Furuya, A., Ohta, K., Matsushima, K., Yoshie, O., Yamamoto, K., and Hirai, K. (2003) Concerted expression of eotaxin-1, eotaxin-2, and eotaxin-3 in human bronchial epithelial cells. Cell. Immunol., 225, 91–100. Yuan, Q., Campanella, G.S., Colvin, R.A., Hamilos, D.L., Jones, K.J., Mathew, A., Means, T.K., and Luster, A.D. (2006) Membrane-bound eotaxin-3 mediates eosinophil transepithelial migration in IL-4-stimulated epithelial cells. Eur. J. Immunol., 36, 2700–2714. Ip, W.K., Wong, C.K., and Lam, C.W. (2006) Interleukin (IL)-4 and IL-13 up-regulate monocyte chemoattractant protein-1 expression in human

References

79.

80.

81.

82.

83.

84.

85.

86.

bronchial epithelial cells: involvement of p38 mitogen-activated protein kinase, extracellular signal-regulated kinase 1/2 and Janus kinase-2 but not c-Jun NH2-terminal kinase 1/2 signalling pathways. Clin. Exp. Immunol., 145, 162–172. Dabbagh, K., Takeyama, K., Lee, H.M., Ueki, I.F., Lausier, J.A., and Nadel, J.A. (1999) IL-4 induces mucin gene expression and goblet cell metaplasia in vitro and in vivo. J. Immunol., 162, 6233–6237. Doucet, C., Brouty-Boye, D., Pottin-Clemenceau, C., Canonica, G.W., Jasmin, C., and Azzarone, B. (1998) Interleukin (IL) 4 and IL-13 act on human lung ﬁbroblasts. Implication in asthma. J. Clin. Invest., 101, 2129–2139. Kuperman, D.A. and Schleimer, R.P. (2008) Interleukin-4, interleukin-13, signal transducer and activator of transcription factor 6, and allergic asthma. Curr. Mol. Med., 8, 384–392. Kuperman, D., Schoﬁeld, B., Wills-Karp, M., and Grusby, M.J. (1998) Signal transducer and activator of transcription factor 6 (Stat6)-deﬁcient mice are protected from antigen-induced airway hyperresponsiveness and mucus production. J. Exp. Med., 187, 939–948. Wills-Karp, M. (2004) Interleukin-13 in asthma pathogenesis. Immunol. Rev., 202, 175–190. Kasaian, M.T. and Miller, D.K. (2008) IL-13 as a therapeutic target for respiratory disease. Biochem. Pharmacol., 76, 147–155. Mullings, R.E., Wilson, S.J., Puddicombe, S.M., Lordan, J.L., Bucchieri, F., Djukanovic, R., Howarth, P.H., Harper, S., Holgate, S.T., and Davies, D.E. (2001) Signal transducer and activator of transcription 6 (STAT-6) expression and function in asthmatic bronchial epithelium. J. Allergy Clin. Immunol., 108, 832–838. Hilleman, M.R. (2002) Realities and enigmas of human viral inﬂuenza: pathogenesis, epidemiology and control. Vaccine, 20, 3068–3087.

87. Hampson, A.W. and Mackenzie, J.S.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

(2006) The inﬂuenza viruses. Med. J. Aust., 185, S39–S43. Sharon, N. and Lis, H. (2004) History of lectins: from hemagglutinins to biological recognition molecules. Glycobiology, 14, 53R–562. Reuter, J.D., Myc, A., Hayes, M.M., Gan, Z., Roy, R., Qin, D., Yin, R., Piehler, L.T., Esfand, R., Tomalia, D.A., and Baker, J.R. (1999) Inhibition of viral adhesion and infection by sialic-acid-conjugated dendritic polymers. Bioconjugate Chem., 10, 271–278. Pinto, L.H. and Lamb, R.A. (2007) Controlling inﬂuenza virus replication by inhibiting its proton channel. Mol. Biosyst., 3, 18–23. Samuel, C.E. (2006) Virus-host interaction minireview series: human immunodeﬁciency virus, hepatitis C virus, and inﬂuenza virus. J. Biol. Chem., 281, 8305–8307. Luscher-Mattli, M. (2000) Inﬂuenza chemotherapy: a review of the present state of art and of new drugs in development. Arch. Virol., 145, 2233–2248. Huckriede, A., Bungener, L., Stegmann, T., Daemen, T., Medema, J., Palache, A.M., and Wilschut, J. (2005) The virosome concept for inﬂuenza vaccines. Vaccine, 23 (Suppl 1), S26–S38. Engelhardt, O.G. and Fodor, E. (2006) Functional association between viral and cellular transcription during inﬂuenza virus infection. Rev. Med. Virol., 16, 329–345. Ghanem, A., Mayer, D., Chase, G., Tegge, W., Frank, R., Kochs, G., Garcia-Sastre, A., and Schwemmle, M. (2007) Peptide-mediated interference with inﬂuenza A virus polymerase. J. Virol., 81, 7801–7804. Amorim, M.J. and Digard, P. (2006) Inﬂuenza A virus and the cell nucleus. Vaccine, 24, 6651–6655. Chan, A.Y., Vreede, F.T., Smith, M., Engelhardt, O.G., and Fodor, E. (2006) Inﬂuenza virus inhibits RNA polymerase II elongation. Virology, 351, 210–217.

237

238

9 Protein Interactions with Small Molecules 98. Kochs, G., Garcia-Sastre, A. and

Martinez-Sobrido, L. (2007) Multiple anti-interferon actions of the inﬂuenza A virus NS1 protein. J. Virol., 81, 7011–7021. 99. Lin, D., Lan, J., and Zhang, Z. (2007) Structure and function of the NS1 protein of inﬂuenza A virus. Acta Biochim. Biophys. Sin., 39, 155–162. 100. De Clercq, E. (2006) Antiviral agents active against inﬂuenza A viruses. Nat. Rev. Drug Discov., 5, 1015–1025.

101. Maroto, M., Fernandez, Y., Ortin, J.,

Pelaez, F., and Cabello, M.A. (2008) Development of an HTS assay for the search of anti-inﬂuenza agents targeting the interaction of viral RNA with the NS1 protein. J. Biomol. Screen., 13, 581–590. 102. Basu, D., Walkiewicz, M.P., Frieman, M., Baric, R.S., Auble, D.T., and Engel, D.A. (2009) Novel inﬂuenza virus NS1 antagonists block replication and restore innate immune function. J. Virol., 83, 1881–1891.

239

10 Cracking the Glycocode: Recent Developments in Glycomics Lars Hillringhaus and J¨urgen Seibel

10.1 State of the Art 10.1.1 Introduction

Cell surfaces are covered by various glycans, which characterize the status of the cell and participate in molecular recognition processes involved in embryonic development, cancer cell metastasis, bacterial and viral infections, and the initiation of immune responses [1]. Changes in the presented glycans are associated with certain disease states like cancer and inﬂammation. In vertebrates there exist different types of glycoconjugates like glycoproteins, glycolipids, proteoglycans, and glycosylphophatidylinositol anchors. Unlike nucleic acids and proteins, which are linear polymers, glycans exist as complex branched structures. Additionally, they are not directly encoded by the genome but synthesized by the combined interaction of carbohydrate synthesizing and modifying enzymes. Owing to their enormous structural diversity they are predestinated to encode biological information. To decipher this so-called glycocode and determine the correlation between speciﬁc glycans and biological events is the goal of glycomics. Several tools have been developed to analyze the glycome ranging from high-throughput analysis on microarrays to in vivo imaging of dynamic changes in the glycome. Cracking the glycocode should enhance the development of carbohydrate-based drugs against infectious diseases and cancer as well as tools for early-stage diagnosis. 10.1.2 Carbohydrate-Based Drugs

As glycans are involved in numerous biological processes they have an enormous potential as therapeutic agents. Nevertheless, the number of carbohydrate-based therapeutics in the market is relatively low, which is mainly due to the difﬁculties in the analysis and synthesis of glycans. Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

240

10 Cracking the Glycocode: Recent Developments in Glycomics

Several small molecules containing carbohydrate moieties are used as therapeutics against infectious diseases and cancer. The most famous ones are zanamivir (Relenza) and oseltamivir (Tamiﬂu), which are used in the treatment of inﬂuenza. Both are structural analogs of neuraminic acid and inhibit inﬂuenza neuraminidase, which is essential for the release of replicated viruses from the host cell. Aminoglycosides belong to a diverse class of aminosugar-containing glycans. They are used to treat bacterial infections as they inhibit the biosynthesis of bacterial proteins by binding bacterial ribosomes. Acarbose is used for the treatment of type II diabetes. It functions as an α-glucosidase and an α-amylase inhibitor regulating intestinal carbohydrate digestion, thereby reducing glucose absorption. It is a pseudotetrasaccharide from microbial origin and is produced by fermentation. The glycosaminoglycan heparin has been used as an antithrombic agent since the 1940s. It activates the serine protease inhibitor antithrombin III, which blocks thrombin and factor Xa in the coagulation cascade [2]. Heparin, which is isolated from animal organs is a heterogeneous mixture of polysaccharides and severe side effects have been observed [3]. By chemical and enzymatic fragmentation low-molecular-weight heparins (LMWHs) can be obtained, which have a longer half-life and fewer side effects, but are still heterogeneous and less active than heparin. In the early 1980s, a pentasaccharide was discovered as the critical sequence that is responsible for the anticoagulant activity of heparin [4]. A synthetic analog of the pentasaccharide is in the market since 2002, and it was shown that its use in major orthopedic surgery led to a decrease in the risk of thrombosis by more than 50% compared to LMWH [5]. Erythropoietin (EPO) is the most widely produced recombinant protein drug in the world and is used in the treatment of anemia. Its in vivo activity and half-life could be increased by the introduction of new glycosylation sites [6]. The surfaces of viruses, parasites, and bacteria are covered by glycans, which usually differ from those of the host cells. Several vaccines containing these glycan motifs have been developed against bacterial and infectious diseases [7]. As tumor cells have altered glycans on their surfaces, vaccines containing these structures were shown to elicit immune responses and are actually used in clinical trials [8, 9]. 10.1.3 Carbohydrate Synthesis

Glycans are difﬁcult to isolate as they are only displayed in small quantities on cell surfaces. Additionally there exist numerous glycoforms of each glycoprotein which differ in their glycosylation pattern and sites. Deﬁned glycan structures are essential to study the roles of glycans in biological processes in detail. Whereas nucleic acids can be synthesized by polymerase chain reaction and proteins by recombinant expression, adequate methods for the synthesis of carbohydrates are still not available. The existing chemical and enzymatic glycosylation techniques are often combined for the synthesis of pure and deﬁned glycans.

10.1 State of the Art

10.1.3.1 Chemical Synthesis Carbohydrates are polyfunctional molecules that carry hydroxyl groups of similar reactivity. Protecting group strategies have to be applied in carbohydrate synthesis to avoid side reactions. Selective protection of hydroxyl groups can be achieved by using orthogonal protecting groups, which can be cleaved independently under different reaction conditions [10]. Another important factor is the control of the stereoselectivity of the glycosylation. In most glycosylation methods anomeric mixtures are obtained. The stereochemistry can be controlled by an accurate choice of neighboring groups, reaction temperature, and solvent [11]. A substituent at C-2, which can participate in the forming of the transition state (e.g., an ester, phenylsulfanyl, or phenylselanyl group) will conduct the synthesis toward the 1,2-trans conﬁgured product, whereas with alkyl or benzyl groups at C-2 or 2-deoxy sugars, the 1,2-cis conﬁgured product is predominantly formed due to the anomeric effect [12]. Nevertheless, the synthesis of the 1,2-cis conﬁgured glycans is a difﬁcult task and anomeric mixtures are often obtained, which have to be separated subsequently by column chromatography. For the glycosylation reaction, a glycosyl donor carrying a leaving group at the anomeric center is ﬁrst activated to an oxocarbenium ion, which is then attacked by a nucleophile that can be a sugar-bearing a free hydroxyl group or an amino acid–like serine or threonine. There exist several glycosylation methods, which differ in the kind of leaving group and the conditions that are necessary for its activation. Many complex glycans including natural products have been synthesized using different glycosylation methods like the Koenings–Knorr, the trichloracetimidate, or the glycal method among many others. Trichloracetimidate glycosyl donors were used in the syntheses of several natural products such as calicheamicin [13, 14], macroviracin D [15], or complex gangliosides [16, 17]. Biantennary, tribranched, and tetrabranched PSA glycopeptide fragments were synthesized using the glycal assembly technique [18]. Chemical synthesis is often the only way to get pure glycans, which are necessary for use in medical applications. Nevertheless, most syntheses require multiple steps and puriﬁcation procedures, reaction times up to several months, and the products are obtained only in low yields. Automated Synthesis The investigation of the glycome requires a large library of structurally deﬁned glycans. Compared to genomics and proteomics where nucleic acids and peptides are synthesized with automated synthesizers, adequate tools for the synthesis of glycans are still in their infancy. However, promising techniques for the automated synthesis of glycans have been developed and applied in various complex glycan syntheses. One-Pot Synthesis The one-pot synthesis is based on the different reactivities of carbohydrate building blocks [19]. The reactivity is determined by the substituents that consist of ﬁnely tuned electron-pushing and -pulling groups. The glycans are synthesized by successive addition of the building blocks to a single reaction vessel, normally starting from the most reactive to the less reactive one. The

241

242

10 Cracking the Glycocode: Recent Developments in Glycomics BzO

BnO BnO

OBn O

(BzN)O

OBz O

HO 2

O(NBz) O STol O(ClBn)

HO 3

STol NHTroc

BnO BnO

STol OLev

NIS, TfOH, CH2Cl2, −20 °C 67%

1

OBn BzO O O OLev

OBz (BzN)O O O NHTroc

O(NBz) O STol O(ClBn) 4: R = Lev

H2NNH2, HOAc, THF, 0 °C 95% 5: R = H HO

OBn O

O OBn BnO

BzO O

STol OBn

BnO OBn

5

7

NIS, TfOH, CH2Cl2, −40 °C 62%

6

OBn O

OBz OBn BzO (BzN)O O(NBz) O O O BnO O O (BnCl)O TrocHN O OBn O OBn O BnO O 8 BnO BnO OBn OBn BnO

OPMP

OBn rt

OBn O

OPMP

OBn

(a) Zn, AcOH (b) Ac2O, pyridine

8

(c) NaOMe, MeOH (d) H2, Pd /C 45%

HO

OH O

HO

HO

OH O

O

O O HO OH

HO

NHAc

OH O HO O

OH

9 HO

OH O OH

O HO

OH O

OPMP

OH

Figure 10.1 One-pot synthesis of Globo H (Tol, toluyl; Lev, levulinyl; Troc, 2,2,2-trichloroethoxycarbonyl; Bn, benzyl; Bz, benzoyl; PMP, 4-methoxyphenyl; NIS, N-iodosuccinimide).

glycan is thereby elongated from the nonreducing to the reducing end. The one-pot synthesis was automated by implementing a computer program named Optimer, which reverts to a database where the reactivities of more than 400 building blocks are saved. The synthesis of the tumor-associated antigen Globo H, a complex hexasaccharide, by automated one-pot synthesis demonstrates the potential of this method (Figure 10.1) [20]. Similarly, Takashi and coworkers synthesized a library of 54 linear and 18 branched trisaccharides by a one-pot, two-step glycosylation using a commercial synthesizer in 63–99% yields [21]. However, the one-pot synthesis is limited to ﬁnding suitable building blocks of adapted reactivity, which are not yet commercially available and often difﬁcult to synthesize. Since the reaction is done in a single reaction vessel no puriﬁcation of the intermediates is necessary. After addition of all building blocks the ﬁnal product has to be puriﬁed properly because by-products and unreacted educts are not removed during the synthesis. Solid-Phase Synthesis An automator for carbohydrate synthesis was developed by Seeberger and coworkers [22]. It is based on a modiﬁed peptide synthesizer. As building blocks, glycosyl phosphates and trichloracetimidates with temporary protecting groups are used. The reducing end of the ﬁrst building block is coupled to a polystyrene resin by an easily cleavable linker. For the glycosylation, an excess

10.1 State of the Art

of the next building block is added followed by washing and ﬁltration to remove by-products and remaining reagents. After selective removal of the temporary protecting group the next building block is added. After several cycles of coupling and deprotection steps the fully protected product is cleaved from the resin, deprotected, and thoroughly puriﬁed. Many biologically important oligosaccharides have been prepared by automated solid-phase synthesis. The dimeric combination of the Lewis blood group antigen Ley -Lex , which is found on tumor cells was prepared using only ﬁve building blocks [23]. The nonasaccharide was synthesized in 6% yield within 23 hours, which is much faster than the solution-phase synthesis, which requires several months. Although only about 36 building blocks are needed to assemble 75% of known mammalian oligosaccharides, the building blocks are laborious and time-consuming to synthesize and not yet commercially available. The advantage is that complex oligosaccharides can be assembled within one day or less and there is only one ﬁnal puriﬁcation step necessary. 10.1.3.2 Enzymatic Synthesis With chemical methods, almost any complex oligosaccharide can be synthesized but the syntheses require an enormous expense. In nature, glycans are synthesized with the help of speciﬁc enzymes termed glycosyltransferases. Isolated enzymes or even whole cells can be used for the synthesis of glycans in vitro [24]. Enzymes are highly regio- and stereoselective so that no protecting groups are necessary. For the enzymatic synthesis of glycans, mainly, glycosidases, glycosynthases, and glycosyltransferases have been used. Glycosidases and Glycosynthases In nature there exist two classes of glycosidases termed endoglycosidases and exoglycosidases, which are responsible for the trimming of oligosaccharides. In vitro they can be used for oligosaccharide synthesis if the acceptor is a better nucleophile than water. Although they are seldom regioselective and give lower yields than glycosyltransferases, they are generally more stable and can be used in some organic solvents [25–28]. Furthermore, glycosidases and the required glycosyl donors are well accessible from bacteria and fungi. Glycosynthases are engineered glycosidases, where a catalytic nucleophile is substituted through mutation by a neutral amino acid like glycine, alanine, or serine. The enzymes thereby lose their hydrolytic activity and are able to form glycosidic bonds using mainly glycosyl ﬂuorides as substrates. Glycosyltransferases Glycosyltransferases synthesize glycans in cells, especially in mammalian cells employing nucleotide-activated sugars as glycosyl donors [29]. Glycosyltransferases catalyze the transfer of a glycosyl donor to a hydroxyl group of another sugar or to a serine, threonine, or asparagine residue of a protein with excellent regio- and anomeric stereoselectivity [30]. Usually they are highly substrate speciﬁc, but tolerate small modiﬁcations of the donor and acceptor [31, 32]. Glycosyltransferases can be further classiﬁed into retaining and inverting ones. Retaining glycosyltransferases do not change the stereochemistry at the anomeric position, whereas inverting glycosyltransferases change the anomeric

243

244

10 Cracking the Glycocode: Recent Developments in Glycomics

stereochemistry. Their in vitro application for glycan synthesis is hampered by limited availability as they exist only in low concentrations in tissue. In addition they are often membrane-bound or -associated, difﬁcult to isolate and often unstable. Recombinantly, they can only be expressed in low yields, which results in a limited large-scale producibility. Though the required nucleotide-activated glycosyl donors are commercially available they are unstable and expensive. Since many glycosyltransferases are strongly inhibited by the released nucleotides, during glycosyltransfer, the nucleotide-activated sugar has to be used in excess. Alternatively, the nucleotide can be degraded or recycled to the nucleotide-activated sugar. A system for the in situ regeneration of nucleotide-activated sugars by which the concentration of the activated sugar and thus, its inhibitory effect on the glycosyltransferase is kept low, was developed by Wong and Whitesides [33]. 10.1.3.3 Glycoprotein Synthesis The structural and functional diversity of proteins is remarkably increased by posttranslational modiﬁcations such as methylation, phosphorylation, sulfation, and glycosylation. The glycosylation of proteins is the most complex posttranslational modiﬁcation and essential for several biological processes like protein folding, cell targeting and adhesion, signal transduction, and many others [34]. Glycoproteins are an emerging class of therapeutics and it has been shown that the glycan moieties are necessary to improve their stability and pharmacokinetic properties [35]. However, glycoproteins exist in nature in numerous glycoforms, which are not appropriate for systematic structure–function studies and therapeutic applications. Thus, different methods have been developed for the preparation of homogeneous glycoproteins. Peptide Ligation Glycopeptides can be synthesized by solid-phase peptide synthesis (SPPS) incorporating glycosylamino acids, by chemoenzymatic manipulation of the glycan moiety of a glycopeptide or by direct glycosylation of a peptide. For SPPS approaches extensive carbohydrate protection is necessary, which is accompanied by the acid and base lability of the glycosylamino acids [36]. SPPS approaches are limited to a maximum peptide size of 50 amino acids. For the synthesis of larger peptides, ligation techniques have to be used where a synthetic glycopeptide is ligated with a peptide or protein based on chemical or biochemical approaches. In the native chemical ligation (NCL) a glycopeptide bearing a C-terminal thioester is reacted with an N-terminal cysteine residue of another peptide whereby a natural peptide bond is formed [37–40]. Several glycopeptides like a HIV viral protein gp 120 fragment [41, 42], complex-type N-linked glycopeptides [43], and a MUC2 motif carrying several O-GalNAc moieties [44] have been synthesized using NCL approaches. The peptide thioester or the cysteine-containing peptide, which are required for the NCL reaction can be generated by recombinant protein expression techniques termed expressed protein ligation (EPL) [38, 45, 46]. Using EPL, Bertozzi et al. synthesized the muzinelike glycoprotein GlyCAM-1, which serves as a ligand for the leukocyte adhesion molecule l-selectin [47]. Peptide fragments can also be ligated chemically by traceless Staudinger ligation where an azide is reacted with

10.1 State of the Art

a phosphinothioester resulting in a native peptide bond [48]. Enzymatic ligation is possible by the use of proteases, which are highly stereo- and chemoselective and do not require a cysteine residue at the ligation site [49]. Enzyme Remodeling Homogenous glycoproteins can be obtained by enzymatic remodeling of their glycan moieties. In a ﬁrst step the glycans are trimmed by glycosidases and in a second step extended by glycosidase- or glycosyltransferase-catalyzed transglycosylation. Hydrolysis as a competing side reaction can be reduced by using organic solvents [50, 51] or oxazoline derivates [49, 52, 53]. Oxazoline donors were used by Wang et al. for the endoglycosidase-catalyzed synthesis of homogenous glycoforms of the protein RNase [54]. Molecular and Cell Biological Techniques Several techniques have been developed to modify the natural glycosylation machinery of a cell. The biosynthetic pathways of cells were augmented by the introduction of new glycosyltransferase genes or regulated by using small molecules that heterodimerize proteins. The natural product rapomyzin, for example, was used for chemically induced dimerization to regulate the production of sLex in living cells [55]. By in vivo suppressor tRNA technology it is possible to introduce unnatural amino acids to the genetic code resulting in a kind of ‘‘pretranslational glycosylation’’ [35]. These unnatural amino acids are either glycosylamino acids [56–58] or carry tags [59] for subsequent site-selective glycosylation. Schultz et al. used this approach to incorporate the glycosylamino acids β-GlcNAc serine and α-GalNAc threonine into myoglobin in vivo [60, 61]. 10.1.4 Glycomics

The glycome is deﬁned as the entity of glycans in a particular cell, tissue, or organism. Glycomics is the structural and functional study of the glycome. The glycome depends on several factors like the genome, which encodes carbohydrate synthesizing, modeling, and transporting enzymes; the transcriptome; and the proteome. In addition, the glycome is also inﬂuenced by nutrient levels, which is not the case for the genome and proteome [62]. Thus the glycome is highly complex and dynamic and its analysis requires a combination of several tools as it cannot be predicted by sole analysis of the genome and proteome. 10.1.4.1 Mass Spectrometry Mass spectrometry (MS) is the primary technique for the analysis of small glycan quantities. Information about the glycan structures as well as the glycan attachment sites of proteins is obtained by various experiments. With modern MS methods, glycomes of whole organisms have been proﬁled and serum samples have been screened to discover disease-related biomarkers [63]. To use MS for high-throughput analysis of, for example, serum samples, the inevitable procedures for glycan sample puriﬁcation had to be accelerated. Methods have been developed by which carbohydrates are captured onto a solid support by an oxime linkage. The solid

245

246

10 Cracking the Glycocode: Recent Developments in Glycomics

support is subsequently washed and the carbohydrates eluted by cleavage of the linker [64] or oxime exchange [65]. However, progress in MS glycan analysis is hampered by extensive sample preparation before analysis. Glycoconjugates have to be separated depending on their attached aglycones (N-linked, O-linked, and glycolipids) and modiﬁcations like O-acetyl, methyl, and sulfate groups are often removed during sample preparation [63]. 10.1.4.2 Microarrays Microarrays are ideally suited for the high-throughput analysis of glycans. Lectins and antibodies are highly speciﬁc for certain glycans and can be used for glycan proﬁling. For this purpose, lectins and antibodies are immobilized on the array surface and afterward, ﬂuorescence labeled glycoproteins from cell lysate or ﬂuid samples are added. This technique is excellently suited for a fast comparison of glycomes, but it reveals only limited structural details. Since 2002, carbohydrate microarrays are used where isolated or synthesized glycans are noncovalently or covalently immobilized on the array surface. The glycans have a deﬁned orientation on the surface, which is similar to their presentation on cell surfaces. The immobilized glycans can be probed with MS [66, 67], ﬂuorescence labeled antibodies or lectins [68], or cell imaging [69]. Glycan arrays have been used to proﬁle carbohydrate speciﬁcities of lectins, growth factors, cytokines, antibodies, microbial toxins, viruses, and whole cells, to determine the speciﬁcity of human, avian, and mutated inﬂuenza hemagglutinin, to develop vaccines against malaria, and to ﬁnd cellular markers of infectious diseases [70]. Furthermore, the activity and selectivity of various glycosyltransferases [70, 71] as well as new substrate speciﬁcities of bacterial glycosyltransferases [72] and inhibitors [70] have been identiﬁed. Thus, glycan microarrays are emerging tools for biomarker and drug discovery. 10.1.4.3 Cell, Tissue, and Metabolic Labeling Molecular imaging techniques have been developed to analyze changes in the glycome during pathogenic or normal biological processes. The glycans in cells in tissues can be imaged by using ﬂuorescent-labeled lectins and antibodies. Lectins have been used in cultured cell lines [73, 74] and tissue sections [75–80], but their in vivo applicability is limited because they possess only low afﬁnity to carbohydrates and are furthermore tissue-impermeable and often toxic. Antibodies have been used for ﬁxed cells and one in vivo application has also been reported [81]. Nevertheless, their use for in vivo applications is limited by their tissue impermeability and low afﬁnity. As well as lectins, they can only be used to study a present state of the glycome, whereas dynamic studies are difﬁcult to conduct. A novel approach for a dynamic study of the glycome in vivo encompasses feeding cells with modiﬁed glycans, which the cell subsequently incorporates in the biosynthesis of cell surface glycans. The modiﬁed glycans carry a chemical reporter group, which can be labeled by click-chemistry with imaging probes and then visualized by ﬂuorescence microscopy. Most glycan subtypes have been

10.2 Some New Insights in Glycomics

imaged by metabolic labeling using azido or alkynyl derivates of sialic acid, N-acetylgalactosamine, N-acetylglucosamine, mannose, and fucose [82]. Glycans could be imaged in many cultured cell lines, and also in living organisms such as zebraﬁsh [83], where changes in the glycome during embryonical development were mapped by metabolic labeling. 10.1.4.4 Bioinformatics A major challenge for bioinformatics is to correlate the data obtained from the various experiments since they reveal information of different kinds and hierarchical levels. Some techniques, for instance, provide information about the glycan structures, whereas others reveal which glycans are presented on speciﬁc cells or tissues. Bioinformatic tools from genomics and proteomics research are seldom adaptable for glycomics since the glycome is not directly encoded by the genome [84]. The correlation of data from glycomics with genomic, transcriptomic, and proteomic data is essential to see, for example, how the expression of enzymes involved in the synthesis or transport of glycans determines the glycome of a certain cell or tissue. In recent years, comprehensive glycan databases have been created and international consortia have been formed to collaborate in a systematic study of the glycome.

10.2 Some New Insights in Glycomics 10.2.1 Microwave-Assisted Glycosylation for the Synthesis of Glycopeptides

Malignant transformations of cells are often associated with the presentation of altered glycoproteins on their surfaces [85]. Synthetic vaccines containing the displayed carbohydrate antigens like Ley or Globo H conjugated to an immunogenic carrier protein like keyhole limpet hemocyanin have been shown to elicit antitumor cell antibody responses in mice and patients [86]. The carbohydrate antigens appear only in small amounts on the surface of tumor cells, and thus cannot be isolated, but rather have to be synthesized. For the solid-phase synthesis of glycopeptides Fmoc-protected glycosylamino acids are generally required. So, there is an urgent need for easy and efﬁcient methods for the synthesis of glycosylamino acids. Carvalho et al. reported about the glycosylation of Fmoc-Ser-OBn with 2-aceta mido-2-deoxy-3,4,6-tri-O-acetyl-α-d-glucopyranosylchloride, which is activated by mercuric bromide [87]. The glycosylation of Fmoc-Ser-OH with β-glucose was reported to proceed under BF3 ·OEt2 promotion in yields from 30 to 37% and reaction times from 2 to 18 hours [88–91]. The described glycosylations require multistep reactions for the synthesis of the glycosyl donors and heavy metals like Hg(CN)2 , HgBr2 , or AgOTf as promoters. Long reaction times are necessary and the products are obtained in low yields.

247

248

10 Cracking the Glycocode: Recent Developments in Glycomics

OH FmocHN 10 O OAc RO AcO

O AcO

FeCl3, 4 Å MS toluene OAc

OBn OAc RO AcO

Microwave

O

O

AcO

OBn

FmocHN O Figure 10.2 FeCl3 -promoted synthesis of glycosylamino acids. Reprinted with permission from [93].

It has been reported that FeCl3 promotes the β-glycosylation of aliphatic alcohols with β-peracetylated glycosides [92]. So, we carried out the reaction with β-peracetylated carbohydrates and Fmoc-Ser-OBn 10 under reﬂux conditions in toluene using FeCl3 as a promoter. Although the β-linked glycosylamino acids were formed, we observed many by-products. The optimal temperature for a minimum of site products and a maximum yield was 45 ◦ C. TLC studies revealed that not the activation but rather the addition of the acceptor was problematic. Neither other promoters nor an excess of the glycosyl donor or an increase of the temperature could improve the yield. This showed that steric reasons obviously hindered the addition. It was reported that steric hindrance could be overcome through microwave radiation. So, we carried out the same reactions under microwave radiation by which the reaction time could be reduced from several hours to just 4 minutes (Figure 10.2). It was not possible to activate the α-anomers of the glycosyl donors. Using a mixture of peracetylated α- and β-glucose 13, only the β-anomer reacted and the α-anomer could be reisolated. We were able to glycosylate Fmoc-Ser-OBn with various peracetylated mono- and disaccharides (Table 10.1). For the synthesis of the β-glycoside 18 from peracetylated lactose 17, a literature method requires three steps with an overall yield of 9%, whereas our method requires only one step with a reaction time of 8 minutes and a yield of 54% [94]. The glycosylation is assumed to proceed under participation of the neighboring group at C-2 to yield selectively the β-glycosides, as reported by Lemieux [95]. The oxocarbenium ion reacts with the more stable acyloxonium ion, which is formed by the attack of the acyl group at C-2. Solvents with low polarity like toluene probably promote this mechanism by inefﬁcient solvation of the oxocarbenium ion [96]. The formation of orthoesters, which is often described in literature was not observed. We also performed the reaction with peracetylated N-acetylglucosamine 19. We were able to isolate the respective oxazoline 20, which can be reacted with various glycosyl acceptors [97]. Nevertheless, it was not possible to react the oxazoline directly with Fmoc-Ser-OBn under FeCl3 promotion and microwave conditions. In conclusion, we could demonstrate that through microwave radiation the reaction time for the synthesis of various glycosylamino acids could be reduced

13

AcO AcO

11

AcO

AcO

Donor

O

OAc

AcO

O

OAc OAc

OAc

14

AcO AcO

12

AcO

AcO

Product

O

O

FmocHN

AcO

O

OAc

FmocHN

AcO

O

OAc

O

O

OBn

OBn

FeCl3 promoted glycosylation of Fmoc-Ser-OBn 10.

AcO

Table 10.1

4

2×4

4

4

Time (min)

Microwave

0 : 0 (0)

0 : 1 (61)

0 : 0 (0)

0 : 1 (52)

α/ß (yield %)

300

720

300

720

Time (min)

0 : 0 (0)

0 : 1 (22)

0 : 0 (0)

0 : 1 (31)

α/ß (yield %)

(continued overleaf )

Conventional

10.2 Some New Insights in Glycomics 249

AcO

AcO AcO

17

AcO

O

OAc

O AcO

19

OAc NHAc

O

OAc

AcO

O

OAc

AcO

O

OAc

AcO

O

OAc

(continued)

AcO O AcO 15

AcO AcO

Donor

Table 10.1

OAc

OAc

AcO AcO

18

AcO

AcO

AcO O AcO

O

OAc

20

N

O O

OAc

O AcO AcO

O

OAc

16

AcO AcO

Product

O

O

AcO FmocHN

O

OAc

FmocHN

AcO

O

OAc

O

O

OBn

OBn

4

2×4

4

2×4

4

4

Microwave

0 : 0 (0)

0 : 1 (85)

0 : 0 (0)

0 : 1 (54)

0 : 0 (0)

0 : 1 (52)

300

–

300

720

300

720

0 : 0 (0)

–

0 : 0 (0)

0 : 1 (16)

0 : 0 (0)

0 : 1 (10)

Conventional

250

10 Cracking the Glycocode: Recent Developments in Glycomics

10.2 Some New Insights in Glycomics

from 5–10 hours to 4–8 minutes. This method is thereby providing easy access for building blocks needed for solid-phase glycopeptide synthesis. Metals such as silver and mercury, which are necessary in most glycosylations for the activation of the glycosyl donor, could be substituted by environmentally friendly FeCl3 . The ß-anomeric selectivity of the glycosyl donor allows separation and discrimination of donor substrates and could be used for multicomponent and multistep reactions. 10.2.2 Highly Efﬁcient Chemoenzymatic Synthesis of Novel Branched Thiooligosaccharides by Substrate Direction with Glucansucrases

Carbohydrates exist as glycolipids or proteins on the cell surfaces. As they inﬂuence signaling processes like viral or bacterial infections, cancer metastasis, and inﬂammatory responses [98], large amounts of deﬁned structure are needed to explore the details of these processes. Glycans exist in nature in extremely wide diversity, whereby the isolation of homogenous forms from natural sources is hardly possible. Even for a single protein, there generally exist several glycoforms, which differ in their glycosylation pattern and sites. Chemical synthesis is often the only way to obtain pure and deﬁned structures, but it also requires complex protecting group strategies resulting in low yields and thorough puriﬁcation steps. In nature, carbohydrates are synthesized by more than 200 glycosyltransferases, which exhibit a high regio- and stereoselectivity, but are often quite substrate speciﬁc. Using glycosyltransferases for the in vitro synthesis of carbohydrates is restricted by expensive substrates and limited availability of the membrane-bound glycosyltransferases. One class of carbohydrate synthesizing enzymes are bacterial glucansucrases also termed non-Leloir glycosyltransferases, which are produced by lactic acid bacteria such as Streptococcus, Lactobacillus, and Leuconstoc spp [99, 100]. Although they have high structural homology with glycosidases, they catalyze the synthesis of glycans, whereas glycosidases catalyze their hydrolysis. It is not clear until now, which structural details are responsible for these contrary functions. Bacterial glucansucrases synthesize glucose polymers of high molecular weight using sucrose as a substrate [24, 101]. They have a highly conserved catalytic domain, which is responsible for the binding of sucrose and a C-terminal domain that is responsible for glucan binding [99]. Depending on the respective glucansucrase, differently linked polymers like dextran (α-1,6), mutan (α-1,3), reuteran (α-1,4), and alternan (α-1,3 and α-1,6) are formed [102]. Besides polymers, oligosaccharides can also be synthesized if small molecules are used as acceptors [103–105]. Nevertheless, the product spectrum is limited, as the products are mostly α-1,6-linked and only glucose can be transferred by using sucrose as a substrate. It was demonstrated that the regioselectivity of enzymatic condensations could be altered by modifying the acceptor substrate as well as the enzyme [106, 107]. The regioselectivity of a glycosidase [108] as well as a Leloir glycosyltransferase [109] was changed by blocking the hydroxyl groups of the acceptor. We were

251

10 Cracking the Glycocode: Recent Developments in Glycomics

252

able to direct the selectivity of glucansucrases by substrate engineering. We used the glucansucrase R (GTFR) from Streptococcus oralis, which forms an α-1,6 glucose polymer with minor α-1,3 branches from sucrose as a substrate. Previous experiments showed that the GTFR can glucosylate various alcohols and amino acids indicating the diversity of this enzyme. The acceptor reaction with maltose as a substrate yielded panose as a product, in which glucose was transferred from α,1-6 to maltose. Using lactose as substrate, glucose was transferred α-1,2. The 6 -hydroxyl group was obviously not sufﬁciently accessible for the enzyme due to steric hindrance caused by the axial 4 -hydroxyl group of lactose. Thus the chemoselectivity of the enzyme could be controlled by the choice of the acceptor. In the acceptor reaction of GTFR with glucose 21 as a substrate, isomaltose 22, a disaccharide with α-1,6 linkage is formed as a product (Figure 10.3). We were able to change the selectivity of GTFR by modifying the substrate glucose at C-6. Glucose was blocked by a tosyl group at C-6 yielding the new acceptor

OH

OTs

O

HO HO

OR

OTs HO HO

OR

25: R = H 27: R = Me (a) 29: R = allyl (a)

21: R = H 23: R = Me (a)

GTFR or GTFA, sucrose

27: R = Me (a) 29: R = allyl (a)

GTFA, sucrose

GTFR, sucrose

OH

OH HO HO

O

O

HO HO

OH

O

HO HO

OTs

HO HO O

HO O

O

OR HO 6-Ts-glucose

HO 6-Ts-glucose

HO Glucose

HO HO

O

HO HO

O

OR

HO

O HO O

OTs O HO

OR

OR

HO

HO 22: R = H 24: R = Me (a) a-1,6

26: R = H 28: R = Me (a) 30: R = allyl (a) a -1,3

Figure 10.3 Acceptor-substrate-directed synthesis by GTFR and GTFA glycosyltransferases. Reprinted with permission from [112].

31: R = Me (a) 32: R = allyl (a) a-1,4

10.2 Some New Insights in Glycomics

substrate 6-tosylglucose 25, which was reacted with GTFR and sucrose. The acceptor product 26 was glucosylated at C-3 in a yield of 48%, which is in correlation with the branching activity of GTFR. The yield could be increased up to 95% by using tosylated methyl-27 and allylglucose 29. The structural diversity was further expanded by using the glucansucrase A (GTFA) from Lactobacillus reuteri, which forms the α-1,4-linked glucose polymer reuteran [110]. Nevertheless, the acceptor products of GTFA are mainly α-1,6-linked [111]. We used the 6-tosylglucose derivates 27 and 29 as acceptor substrates for the GTFA and the products 31 and 32 were α-1,4-linked. So, it was possible to direct the main activity of the glucansucrases GTFR and GTFA to a side activity by substrate engineering. The tosyl as well as the allyl group provide access to further elaboration to build up branched glycoconjugates. Driguez et al. reported the synthesis of β-1,6-thio-linked oligosaccharides by anomeric S-alkylation through substitution of 6-halogenides of S-acetyl protected thiosugars [113]. We applied commercially available 1-thiosugars to minimize the necessary synthetic steps. The α-1,3-linked acceptor product 26 from the reaction with GTFR was ﬁrst acetylated and then substituted with peracetylated thiosugars to yield trisaccharides in yields of around 30% (Table 10.2). The synthesis was further optimized so that no protecting groups were needed. Investigations showed that no substitution took place using monosaccharides, which is due to their lower nucleophilicity. So, the unprotected tosyl sugars could be substituted in DMF (dimethylformamide) at 80 ◦ C by thiosugars, which were previously deprotected and deprotonated with sodium methanolate. The yields were in the range of 90%. With this chemoenzymatic strategy, various branched thiooligosaccharides could be synthesized in a two-step synthesis (Table 10.2). Thiosugars, which carry a sulfur instead of an oxygen atom in their glycosidic bond, possess unique physiochemical properties differing from those of conventional O-glycosides. They have an enhanced water solubility and are tolerated by most biological systems. Besides, the S-glycosidic bond causes an enhanced stability against enzymatic or acid/base-catalyzed hydrolysis [114]. These characteristics qualify them as promising compounds for therapeutic applications. It has been reported that certain thio-linked disaccharides serve as fucosidase inhibitors [114, 115]. Furthermore it was shown, that thio-linked disaccharides decreased tumor cell line viability and hence, are potential antitumor agents [116]. Among the synthesized glycans, is the trisaccharide 45 with a terminal α-2,6-linked sialic acid. Sialic acids are involved in important biological processes like viral infection, immune cell activation, cancer metastasis, and neuronal development among many others [117]. An enhanced expression of α-2,6-linked sialic acids on tumor cells is often associated with poor prognosis for therapy [118]. The synthesized thiooligosaccharides, which are allyl-functionalized at the reducing end provide the opportunity for conjugation to proteins to yield neoglycoproteins or for immobilization to solid supports like microarrays, to investigate their potential as vaccines and their afﬁnity to proteins of biological relevance.

253

AcO AcO

AcO AcO

36

O

OAc

O AcO

O AcO

OAc

AcO

36

O

OAc

OAc

O

OAc

O

OTs

OAc

O

OTs

OAc

OAc 37

34

AcO

AcO

AcO AcO

34

AcO

AcO

SH

SH OAc

O

OAc

OAc

O

OAc

SH OAc

O

OAc

Acceptor

Chemical synthesis of branched thiooligosaccharides.

OTs

AcO

33

AcO AcO

Donor

Table 10.2

AcO

O

OAc

AcO AcO

AcO AcO

39

38

O

AcO

O

OAc

O

AcO

O

OAc

35 AcO AcO

AcO

AcO

Product

O

AcO

AcO

AcO AcO

AcO

S

AcO

AcO

O

OAc

AcO

AcO

O

OAc

OAc

O

O AcO

S

AcO

S

OAc

OAc

34

31

75

Yield (%)

254

10 Cracking the Glycocode: Recent Developments in Glycomics

O

29

HO HO

29

HO HO O HO

O HO

HO

O

OH

HO

O

OH

HO 33 OCH3

HO HO

OTs

OCH3

HO

O OCH3

OTs

HO

O

OTs

OH

O

OH

SH

SH

SH

OH

O

OH

40

O

OH

40

33

HO HO

HO

HO

HO

HO

OH

HO HO

HO HO

41

HO

HO

43

S

O

O

HO

O

OH

42

HO

O

OH

HO HO

HO

O

OH

HO HO

HO

HO

HO

HO

O

S

HO

HO

O

OH

OH

OCH3

O HO

HO

O HO

87

91

(continued overleaf )

OCH3

OCH3

O

S

90

10.2 Some New Insights in Glycomics 255

HO HO

29

HO HO

Donor

(continued)

31

O HO

HO

O

OH

O HO

HO

O

OH

Table 10.2

OCH3

HO

O O

OTs

HO

O

OTs OH

33

OH

O

OH

O

SH

44

AcHN HO HO

HO HO

HO

Acceptor

SH

CO2H

HO HO

HO HO

O

O

HO

O

OH

HO O

46

45

OH O

HO

HO

HO

AcHN HO HO OH

HO

Product

S

O

O HO

OCH3

HO

HO

O

OH

HO

O

S

CO2H

90

47

Yield (%)

256

10 Cracking the Glycocode: Recent Developments in Glycomics

10.2 Some New Insights in Glycomics

257

10.2.3 Identiﬁcation of New Acceptor Speciﬁcities of Glycosyltransferase R with the Aid of Substrate Microarrays

Leloir glycosyltransferases are excellent tools for the regio- and stereoselective synthesis of glycans. Nevertheless, the transferases are not always accessible and as well as the required nucleotide-activated glycosyl donors, expensive and not ﬁtted for large-scale synthesis. Non-Leloir glycosyltransferases, on the other hand, use sucrose as a substrate and it was shown that their regioselectivity can be controlled by substrate engineering [112]. Carbohydrate microarrays are excellent tools to analyze the structure and functions of glycans [70, 119, 120]. In most cases, isolated or readily synthesized glycans are ﬁrst immobilized on the array and proteins added in a second step. There are some reports where glycoarrays are used to study the activity of glycosyltransferases [70, 71]. Our goal was to OH OH O HO HO

OH O

HO HO

HO O O HO

HO 47

HO HO

OH OH

OH

HO

O

O

HO OH HO

OH O HO

O

GTFR

O HO

OH

HO HO

+

OH

48

HO HO

O

HO

O

O OH HO

O

O HO

O

HO

HO O

OH N N N

N N N O

O

H3C N

H3C N O

O H3C N

H3C N

A

B

Figure 10.4 Enzymatic glycosylation of different acceptors on arrays. The immobilized alcohol on slide A and immobilized maltose were successfully glycosylated by the bacterial glycosyltransferase R and could afterward be detected by the binding of ﬂuorescence labeled Concanavalin A. Reprinted with permission from [72].

A

B

10 Cracking the Glycocode: Recent Developments in Glycomics

identify new glycosylation pathways of bacterial glucansucrases on an array by variation of enzyme, acceptor, and substrate. As acceptors, azido-functionalized maltose and lactose were ﬁrst coupled to a linker by 1,3-dipolar cycloaddition, and afterward, a primary aliphatic alcohol was also immobilized on a microtiter plate by NHS/EDC (N-hydroxy succinimide/1-ethyl-3-(3-dimethylaminopropyl)carbodiimide) coupling in order to ﬁnd out which of these substrates is glycosylated by the bacterial glycosyltransferase GTFR (Figure 10.4). Hence, a solution of glucosyltransferase GTFR and sucrose 47 was added to each microtiter well. After blocking the array with bovine serum albumin, ﬂuorescence labeled Concanavalin A (ConA), which is a lectin speciﬁc for mannose and glucose but not for galactose, was added to detect the glycosylation of the immobilized acceptors. In literature, maltose was described to be a good acceptor for glycosyltransferases of the glucansucrase type such as dextransucrase (DSRS) [121], whereas for alcohols, no glycosylation was observed. After scanning the microtiter plate with a ﬂuorescence scanner, we observed a stronger signal for maltose with added enzyme than for maltose without added enzyme (Figure 10.5). Goldstein et al. reported about a twofold increase of the ConA binding afﬁnity for α-(1,6)-glucosylated maltose relative to mannose [122]. So, we interpreted the increased signal for enzymatically treated maltose as its successful glycosylation. The signal for enzymatically treated lactose was lower than for the microtiter wells without enzyme indicating that no glycosylation took place. The decreased signal is presumably due to less nonspeciﬁc binding of ConA. With the DSRS B-512F from L. mesenteroides, lactose is glucosylated at C-2[123]; so the immobilized lactose was probably not glucosylated because of steric hindrance resulting from the attached linker at the reducing end (Figure 10.5). The strongest binding was observed for 30 000 Aliphatic alcohol, sigmoidal Maltose, sigmoidal Lactose, sigmoidal Empty, sigmoidal Maltose Lactose Aliphatic alcohol Empty

25 000 Relative fluorescence

258

20 000 15 000 10 000 5 000 0 0

2

4

6

8

Time [h]

Figure 10.5 The arrays were monitored with the glucose-speciﬁc lectin ConA, with novel acceptor products derived from sucrose being identiﬁed with the glycosyltransferase, GTFR. Reprinted with permission from [72].

10

12

10.3 Future Perspectives

the immobilized alcohol, which shows its successful glycosylation. In summary, we could use the array platform as an excellent tool for the identiﬁcation of new enzyme speciﬁcities as well as new acceptor products for the GTFR. After approval of the concept, the chip format should in future be reduced to small carbohydrate microarrays that offer the possibility to screen several hundreds acceptors in parallel fashion, which is not possible with conventional methods. This makes them important tools to enhance the scope of enzymes, which can be used to synthesize various complex glycans in solution and on-chip.

10.3 Future Perspectives

In 2003, the MIT’s magazine Technology Review had identiﬁed glycomics as one of the top 10 technologies that will change the future. During the past years, many efforts have been devoted to establish tools for the synthesis and investigation of glycans. Promising methods are now available for the synthesis of complex glycans. Nevertheless, each method has its strengths and weaknesses, which are often complementary to each other. Therefore, combined chemical and enzymatic approaches have to be applied for the synthesis of diverse glycan libraries. Although techniques to analyze the glycome are lagging behind those in genomics and proteomics, important advancements have been achieved. Metabolically labeled glycans have already been visualized in vivo and should in future be extended to mammalian disease models and human clinical settings [82]. Presenting the whole glycome of a certain cell on a single microarray would be a great advance in functional glycomics, which is until now only limited by the difﬁculties in the isolation and synthesis of glycans. Modiﬁcation or even synthesis of the glycans on the chip by using engineered enzymes and substrates will enhance the diversity of existing glycan libraries. Microarrays have already proven to be excellent tools to study interactions between carbohydrates and proteins, viruses, and whole cells; sera of patients have also been screened [70]. High-throughput array analysis will in future lead to novel therapeutics and the discovery of highly speciﬁc disease biomarkers. Synthetic carbohydrate-based vaccines are already in clinical trials and are hopeful candidates against infectious diseases and cancer. New tools, algorithms, and data collections for glycobiology have to be developed and cross-linked with existing genomic and proteomic data collections [84]. So, most likely, many future insights in life sciences will be by achieved with the aid of glycomics.

Acknowledgments

J.S. and L.H. wish to express their gratitude to all past and present group members of J.S. who contributed to the described research. L.H. would also like to thank the German Environmental Foundation (DBU) for a PhD scholarship.

259

260

10 Cracking the Glycocode: Recent Developments in Glycomics

References 1. Varki, A. (2009) Essentials of Glycobio-

2. 3. 4. 5.

6. 7. 8. 9. 10. 11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

logy, 2nd edn, Cold Spring Harbor Laboratory Press, Cold Spring Harbor. Capila, I. and Linhardt, R.J. (2002) Angew. Chem. Int. Ed., 41, 390. Tabeur, C. et al. (1999) Bioorg. Med. Chem., 7, 2003. Petitou, M., Casu, B., and Lindahl, U. (2003) Biochimie, 85, 83. Bauer, K.A., Hawkins, D.W., Peters, P.C., Petitou, M., Herbert, J.-M., Boeckel, C.A.A., and Meuleman, D.G. (2002) Cardiovasc. Drug Rev., 20, 37. Macdougall, L.C. (2000) Semin. Nephrol., 20, 375. Seeberger, P.H. and Werz, D.B. (2007) Nature, 446, 1046. Livingston, P.O. and Ragupathi, G. (2006) Hum. Vaccines, 2, 137. Galoni´c, D.P. and Gin, D.Y. (2007) Nature, 446, 1000. Wong, C.H., Ye, X.S., and Zhang, Z. (1998) J. Am. Chem. Soc., 120, 7137. Levy, D.E., F¨ugedi, P., and Levy, F. (2006) The Organic Chemistry of Sugars, Taylor & Francis, Boca Raton. Nicolaou, K.C. and Mitchell, H.J. (2001) Angew. Chem., 113, 1624. Halcomb, R.L., Boyer, S.H., and Danishefsky, S.J. (1992) Angew. Chem. Int. Ed. Engl., 31, 338. Nicolaou, K.C., Schreiner, E.P., Iwabuchi, Y., and Suzuki, T. (1992) Angew. Chem. Int. Ed. Engl., 31, 340. Mlynarski, J., Ruiz-Caro, J., and F¨urstner, A. (2004) Chemistry, 10, 2214. Sugimoto, M., Fujikura, K., Nunomura, S., Horisaki, T., Ito, Y., and Ogawa, T. (1990) Tetrahedron Lett., 31, 385. Sugimoto, M., Fujikura, K., Nunomura, S., Ito, Y., and Ogawa, T. (1990) Tetrahedron Lett., 31, 1435. Dudkin, V.Y., Miller, J.S., and Danishefsky, S.J. (2004) J. Am. Chem. Soc., 126, 736. Zhang, Z., Ollmann, I.R., Ye, X.S., Wischnat, R., Baasov, T., and Wong, C.H. (1999) J. Am. Chem. Soc., 121, 734. Burkhart, F., Zhang, Z., Wacowich-Sgarbi, S., and Wong, C.H.

21.

22.

23.

24.

25. 26. 27.

28.

29.

30. 31.

32.

33. 34. 35.

36. 37. 38.

(2001) Angew. Chem. Int. Ed. Engl., 40, 1274. Takahashi, T., Adachi, M., Matsuda, A., and Doi, T. (2000) Tetrahedron Lett., 41, 2599. Plante, O.J., Palmacci, E.R., and Seeberger, P.H. (2001) Science, 291, 1523. Routenberg Love, K. and Seeberger, P.H. (2004) Angew. Chem. Int. Ed. Engl., 43, 602. Seibel, J., J¨ordening, H.-J., and Buchholz, K. (2006) Biocatal. Biotransform., 24, 311. Chinnathambi, S. and Lachke, A. (1997) Biotechnol. Tech., 11, 589. Laroute, V. and Willemot, R.M. (1992) Enzyme Microb. Technol., 14, 528. Vulfson, E.N., Patel, R., Beecher, J.E., Andrews, A.T., and Law, B.A. (1990) Enzyme Microb. Technol., 12, 950. Beecher, J.E., Andrews, A.T., and Vulfson, E.N. (1990) Enzyme Microb. Technol., 12, 955. Murrell, M.P., Yarema, K.J., and Levchenko, A. (2004) Chembiochem, 5, 1334. Roseman, S. (2001) J. Biol. Chem., 276, 41527. Qian, X., Sujino, K., and Palcic, M.M. (2000) Enzymatic Glycosylations with Non-natural Donors and Acceptors Wiley-VCH Verlag GmbH, vol. 2. Sujino, K., Uchiyama, T., Hindsgaul, O., Seto, N.O.L., Wakarchuk, W.W., and Palcic, M.M. (2000) J. Am. Chem. Soc., 122, 1261. Wong, C.H. and Whitesides, G.M. (1982) J. Org. Chem., 47, 2816. Helenius, A. and Aebi, M. (2001) Science, 291, 2364. Gamblin, D.P., Scanlan, E.M., and Davis, B.G. (2009) Chem. Rev., 109, 131. Kunz, H. (1987) Angew. Chem. Int. Ed. Engl., 26, 294. Dawson, P.E. and Kent, S.B.H. (2000) Annu. Rev. Biochem., 69, 923. Muir, T.W. (2003) Annu. Rev. Biochem., 72, 249.

References 39. Nilsson, B.L., Soellner, M.B., and

40. 41.

42.

43.

44.

45.

46.

47. 48.

49.

50.

51.

52.

53.

54. 55. 56.

Raines, R.T. (2005) Annu. Rev. Biophys. Biomol. Struct., 34, 91. Derek, M. (2006) Angew. Chem. Int. Ed., 45, 7668. Mandal, M., Dudkin, V.Y., Geng, X., and Danishefsky, S.J. (2004) Angew. Chem. Int. Ed., 43, 2557. Geng, X., Dudkin, V.Y., Mandal, M., and Danishefsky, S.J. (2004) Angew. Chem. Int. Ed., 43, 2562. Mezzato, S., Schaffrath, M., and Unverzagt, C. (2005) Angew. Chem. Int. Ed., 44, 1650. Hojo, H., Matsumoto, Y., Nakahara, Y., Ito, E., Suzuki, Y., Suzuki, M., Suzuki, A., and Nakahara, Y. (2005) J. Am. Chem. Soc., 127, 13720. David, R., Richter, M.P.O., and Beck-Sickinger, A.G. (2004) Eur. J. Biochem., 271, 663. Muir, T.W., Sondhi, D., and Cole, P.A. (1998) Proc. Natl. Acad. Sci. U.S.A., 95, 6705. Macmillan, D. and Bertozzi, C.R. (2004) Angew. Chem. Int. Ed., 43, 1355. Saxon, E., Armstrong, J.I., and Bertozzi, C.R. (2000) Org. Lett., 2, 2141. Li, B., Zeng, Y., Hauser, S., Song, H., and Wang, L.X. (2005) J. Am. Chem. Soc., 127, 9692. Fan, J.Q., Takegawa, K., Iwahara, S., Kondo, A., Kato, I., Abeygunawardana, C., and Lee, Y.C. (1995) J. Biol. Chem., 270, 17723. Akaike, E., Tsutsumida, M., Osumi, K., Fujita, M., Yamanoi, T., Yamamoto, K., and Fujita, K. (2004) Carbohydr. Res., 339, 719. Li, H., Li, B., Song, H., Breydo, L., Baskakov, I.V., and Wang, L.X. (2005) J. Org. Chem., 70, 9990. Rising, T.W., Claridge, T.D., Moir, J.W., and Fairbanks, A.J. (2006) Chembiochem, 7, 1177. Li, B., Song, H., Hauser, S., and Wang, L.-X. (2006) Org. Lett., 8, 3081. Kohler, J.J. and Bertozzi, C.R. (2003) Chem. Biol., 10, 1303. Fahmi, N.E., Dedkova, L., Wang, B., Golovine, S., and Hecht, S.M. (2007) J. Am. Chem. Soc., 129, 3586.

57. R¨ ohrig, C.H., Retz, O.A., Hareng, L.,

58.

59.

60.

61.

62. 63. 64.

65.

66.

67.

68.

69.

70.

Hartung, T., and Schmidt, R.R. (2005) Chembiochem, 6, 1805. Zhang, Z., Gildersleeve, J., Yang, Y.Y., Xu, R., Loo, J.A., Uryu, S., Wong, C.H., and Schultz, P.G. (2004) Science, 303, 371. Liu, H., Wang, L., Brock, A., Wong, C.-H., and Schultz, P.G. (2003) J. Am. Chem. Soc., 125, 1702. Xu, R., Hanson, S.R., Zhang, Z., Yang, Y.-Y., Schultz, P.G., and Wong, C.-H. (2004) J. Am. Chem. Soc., 126, 15654. Zhang, Z., Gildersleeve, J., Yang, Y.-Y., Xu, R., Loo, J.A., Uryu, S., Wong, C.-H., and Schultz, P.G. (2004) Science, 303, 371. Freeze, H.H. (2006) Nat. Rev. Genet., 7, 537. Pilobello, K.T. and Mahal, L.K. (2007) Curr. Opin. Chem. Biol., 11, 300. Lohse, A., Martins, R., Jorgensen, M., and Hindsgaul, O. (2006) Angew. Chem. Int. Ed. Engl., 45, 4167. Shimaoka, H., Kuramoto, H., Furukawa, J., Miura, Y., Kurogochi, M., Kita, Y., Hinou, H., Shinohara, Y., and Nishimura, S. (2007) Chem. Eur. J., 13, 1664. Fazio, F., Bryan, M.C., Blixt, O., Paulson, J.C., and Wong, C.H. (2002) J. Am. Chem. Soc., 124, 14397. Laurent, N., Voglmeir, J., Wright, A., Blackburn, J., Pham, N.T., Wong, S.C., Gaskell, S.J., and Flitsch, S.L. (2008) Chembiochem, 9, 883. Blixt, O., Head, S., Mondala, T., Scanlan, C., Huﬂejt, M.E., Alvarez, R., Bryan, M.C., Fazio, F., Calarese, D., Stevens, J., Razi, N., Stevens, D.J., Skehel, J.J., van Die, I., Burton, D.R., Wilson, I.A., Cummings, R., Bovin, N., Wong, C.H., and Paulson, J.C. (2004) Proc. Natl. Acad. Sci. U.S.A., 101, 17033. Nimrichter, L., Gargir, A., Gortler, M., Altstock, R.T., Shtevi, A., Weisshaus, O., Fire, E., Dotan, N., and Schnaar, R.L. (2004) Glycobiology, 14, 197. Liang, P.H., Wu, C.Y., Greenberg, W.A., and Wong, C.H. (2008) Curr. Opin. Chem. Biol., 12, 86.

261

262

10 Cracking the Glycocode: Recent Developments in Glycomics 71. Laurent, N., Voglmeir, J., and Flitsch,

72.

73. 74.

75. 76. 77.

78.

79. 80.

81.

82. 83.

84.

85. 86.

87.

88.

S.L. (2008) Chem. Commun., (37), 4400. Seibel, J., Hellmuth, H., Hofer, B., Kicinska, A.M., and Schmalbruch, B. (2006) Chembiochem, 7, 310. Carlsson, S., Carlsson, M.C., and Lefﬂer, H. (2007) Glycobiology, 17, 906. Lizzi, A.R., D’Alessandro, A.M., Bozzi, A., Cinque, B., Oratore, A., and D’Andrea, G. (2007) Mol. Cell. Biochem., 300, 29. Grifﬁth, C.M. and Sanders, E.J. (1991) Anat. Rec., 231, 238. Grifﬁth, C.M. and Wiley, M.J. (1989) Anatomy Embryol., 180, 567. Paessens, L.C., Garc´ıa-Vallejo, J.J., Fernandes, R.J., and van Kooyk, Y. (2007) Immunol. Lett., 110, 65. Thurston, G., Baluk, P., Hirata, A., and McDonald, D.M. (1996) Am. J. Physiol.: Heart Circ. Physiol., 271, Tian, E. and Ten Hagen, K.G. (2007) Glycobiology, 17, 820. Truong, L.D., Phung, V.T., Yoshikawa, Y., and Mattioli, C.A. (1988) Histochemistry, 90, 51. Licha, K., Debus, N., Emig-Vollmer, S., Hofmann, B., Hasbach, M., Stibenz, D., Sydow, S., Schirner, M., Ebert, B., Petzelt, D., B¨uhrer, C., Semmler, W., and Tauber, R. (2005) J. Biomed. Opt., 10, 041205. Laughlin, S.T. and Bertozzi, C.R. (2009) Proc. Natl. Acad. Sci. U.S.A., 106, 12. Laughlin, S.T., Baskin, J.M., Amacher, S.L., and Bertozzi, C.R. (2008) Science, 320, 664. von der Lieth, C.W., Bohne-Lang, A., Lohmann, K.K., and Frank, M. (2004) Brief. Bioinform., 5, 164. Hakomori, S.I. (1985) Cancer Res., 45, 2405. Ouerfelli, O., Warren, J.D., Wilson, R.M., and Danishefsky, S.J. (2005) Expert Rev. Vaccines, 4, 677. Carvalho, I., Scheuerl, S.L., Kartha, K.P.R., and Field, R.A. (2003) Carbohydr. Res., 338, 1039. Fahmi, N.E., Golovine, S., Wang, B., and Hecht, S.M. (2001) Carbohydr. Res., 330, 149.

89. Salvador, L.A., Elofsson, M., and

90.

91.

92.

93. 94. 95. 96. 97. 98. 99.

100.

101.

102.

103. 104. 105.

106.

Kihlberg, J. (1995) Tetrahedron, 51, 5643. Sol, V., Blais, J.C., Carr´e, V., Granet, R., Guilloton, M., Spiro, M., and Krausz, P. (1999) J. Org. Chem., 64, 4431. Steffan, W., Schutkowski, M., and Fischer, G. (1996) Chem. Commun., 313. Katsuraya, K., Ikushima, N., Takahashi, N., Shoji, T., Nakashima, H., Yamamoto, N., Yoshida, T., and Uryu, T. (1994) Carbohydr. Res., 260, 51. Seibel, J., Hillringhaus, L., and Moraru, R. (2005) Carbohydr. Res., 340, 507. Rajca, A. and Wiessler, M. (1995) Carbohydr. Res., 274, 123. Lemieux, R.U. (1954) Adv. Carbohydr. Chem., 9, 1. Wulff, G. and Rohle, G. (1974) Angew. Chem., 86, 173. Wittmann, V. and Lennartz, D. (2002) Eur. J. Org. Chem., (8), 1363. Varki, A. (1993) Glycobiology, 3, 97. Monchois, V., Willemot, R.M., and Monsan, P. (1999) FEMS Microbiol. Rev., 23, 131. Monsan, P., Bozonnet, S., Albenne, C., Joucla, G., Willemot, R.M., and Remaud-Simeon, M. (2001) Int. Dairy J., 11, 675. van Hijum, S.A., Kralj, S., Ozimek, L.K., Dijkhuizen, L., and van Geel-Schutten, I.G. (2006) Microbiol. Mol. Biol. Rev., 70, 157. Magali Remaud-Simeon, C.A., Joucia, G., Fabre, E., Bozonnet, S., Pizzut, S., Escalier, P., Potocki-Veronese, G., and Monsan, P. (2003)in Glucansucrases: Structural Basis, Mechanistic Aspects, and New Perspectives for Engineering, vol. 849 (eds G.Eggleston. and G.L. Cote), American Chemical Society, p. 90. Robyt, J.F. and Eklund, S.H. (1982) Bioorg. Chem., 11, 115. Robyt, J.F. and Walseth, T.F. (1978) Carbohydr. Res., 61, 433. Dols, M., Simeon, M.R., Willemot, R.M., Vignon, M.R., and Monsan, P.F. (1997) Carbohydr. Res., 305, 549. Montero, E., Alonso, J., Canada, F.J., Fernandez-Mayoralas, A., and

References

107.

108. 109.

110.

111.

112.

113.

Martin-Lomas, M. (1997) Carbohydr. Res., 305, 383. Nilsson, K.G.I. and Fern´andezMayoralas, A. (1991) Biotechnol. Lett., 13, 715. MacManus, D.A. and Vulfson, E.N. (2000) Biotechnol. Bioeng., 69, 585. Lairson, L.L., Watts, A.G., Wakarchuk, W.W., and Withers, S.G. (2006) Nat. Chem. Biol., 2, 724. Kralj, S., van Geel-Schutten, G.H., van der Maarel, M.J., and Dijkhuizen, L. (2004) Microbiology, 150, 2099. Kralj, S., Stripling, E., Sanders, P., van Geel-Schutten, G.H., and Dijkhuizen, L. (2005) Appl. Environ. Microbiol., 71, 3942. Hellmuth, H., Hillringhaus, L., H¨obbel, S., Kralj, S., Dijkhuizen, L., and Seibel, J. (2007) Chembiochem, 8, 273. Greffe, L., Jensen, M.T., Chang-Pi-Hin, F., Fruchard, S., O’Donohue, M.J., Svensson, B., and Driguez, H. (2002) Chemistry, 8, 5447.

114. Witczak, Z.J., Chhabra, R., Chen, H.,

115.

116. 117. 118.

119. 120. 121.

122. 123.

and Xie, X.-Q. (1997) Carbohydr. Res., 301, 167. Witczak, Z.J., Sun, J., and Mielguj, R. (1995) Bioorg. Med. Chem. Lett., 5, 2169. Witczak, Z.J., Kaplon, P., and Dey, P.M. (2003) Carbohydr. Res., 338, 11. Varki, N.M. and Varki, A. (2007) Lab. Invest., 87, 851. Miles, D.W., Happerﬁeld, L.C., Smith, P., Gillibrand, R., Bobrow, L.G., Gregory, W.M., and Rubens, R.D. (1994) Br. J. Cancer, 70, 1272. Paulson, J.C., Blixt, O., and Collins, B.E. (2006) Nat. Chem. Biol., 2, 238. Park, S., Lee, M.R., and Shin, I. (2008) Chem. Commun., 4389. Buchholz, K. and Monsan, P. (2003) Handbook of Food Enzymology, p. Marcel Dekker, 589. Goldstein, I.J., Hollerman, C.E., and Smith, E.E. (1965) Biochemistry, 4, 876. Robyt, J.F. and Eklund, S.H. (1983) Carbohydr. Res., 121, 279.

263

265

Part IV Enzyme Chemistry

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

267

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity Anna Company

11.1 General Introduction: O2 Activation and Model Systems

The use of O2 as the oxidant in chemical transformations has become of great industrial interest, not only because it constitutes an environmentally friendly alternative to the highly toxic oxidants used nowadays in several chemical transformations but also due to its relatively low cost [1]. However, directly using molecular oxygen as an oxidant is not straightforward. The dioxygen molecule has a high potential reactivity that is kept in control by its particular electronic structure: molecular oxygen is a diradical having two unpaired electrons [2]. The two oxygen atoms share six electrons in the σ2pz , π2px , and π2py molecular orbitals and two unpaired electrons reside in the two degenerate antibonding π2px ∗ and π2py ∗ orbitals, leaving O2 with a formal bond order of two. The triplet ground state of O2 (S = 1) direct reaction with singlet molecules (S = 0) – the spin-paired state of most organic molecules representative of biological substrates – a spin-forbidden process [3, 4]. In order to overcome the high kinetic barrier inherent to the reactions of triplet O2 , in nature dioxygen is reduced from its abundant triplet ground state to a reactive singlet or doublet state (O2 − , O2 2− ) by oxidase and oxygenase enzymes. In many cases, these enzymes generate species of even greater reactivity by cleaving the O–O bond. Generally, nature’s strategy involves a transition metal (mainly Cu and Fe but also Mn, Ni, or Co), which in the appropriate oxidation state is able to react directly with triplet O2 to form dioxygen adducts that can participate in reaction pathways leading either to the selective incorporation of oxygen atoms into organic substrates or to their oxidation. Chemists have long sought to unravel the mechanisms of biological oxygen activation [5–9], with the ultimate objectives of inventing new reagents for organic synthesis and industrial catalysis, exerting control over reactivity and selectivity, and getting fundamental information about enzymatic reactions. A key issue concerns the nature of the oxidizing species that is directly responsible for effecting the reaction. In particular, the characterization of metal-based intermediates that act as the active species is critical to obtain detailed mechanistic insight. However, due to their nature as reactive oxidizing molecules, isolation or even direct observation Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

268

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

is, in general, difﬁcult. Model chemistry is a useful tool to achieve this goal [10, 11]. It gains inspiration from biological systems to develop low-molecular-weight complexes that reproduce structural characteristics of an enzyme and hopefully its activity. These complexes are useful to understand how nature works, and in selected cases they constitute excellent catalysts that can elicit unprecedented and highly speciﬁc chemical transformations analogous to enzymes. In model chemistry, the accomplishment of these objectives often begins with the preparation of a suitable ligand (organic synthesis) and continues through several steps toward the ultimate goal of reproducing catalytic reactivity. In this review chapter, we present how the model chemistry approach has allowed to reproduce and to better understand the mechanisms by which O2 is activated in a dinuclear copper protein called tyrosinase is discussed.

11.2 Copper Proteins Involved in O2 Activation

Copper is found in most living organisms, and despite being present only in trace amounts, it is essential for life [12]. Generally, it is considered that biological copper serves exclusively as 1e− shuttle, alternating between CuI and CuII . The CuIII state is generally considered to be inaccessible because of the highly positive CuIII /CuII redox potentials [13]. However, several reaction mechanisms postulate the existence of intermediates in which the copper center reaches such an oxidation state that has so far not been directly detected in biological systems. The versatility of copper is clearly reﬂected by the wide range of biological reactions that it catalyzes. Nevertheless, in most cases, the role of this metal ion is directly related to two basic functions: electron transfer and O2 activation and transport [3, 14]. While the electron transfer activity is performed by the so-called ‘‘blue copper proteins,’’ transport and activation of O2 is carried out by a wide range of enzymes. Indeed, in biological systems, O2 -activating copper proteins emerge as a structurally diverse family of enzymes. Their active center may be constituted by one copper center (mononuclear copper proteins: amine oxidase and galactose oxidase) [15], by two metal sites (dinuclear copper proteins: dopamine β-hydroxylase, hemocyanin, tyrosinase, and catechol oxidase) [15, 16], or even by multinuclear conﬁgurations in which three copper centers are involved (trinuclear sites: laccase, ascorbate oxidase, ceruloplasmin, and FET 3) [16]. This variety of structural motifs allows the performance of a wide range of oxidative transformations including O2 transport, substrate hydroxylation, amine oxidation, or hydrogen atom abstraction. All these reactions take place through the activation of O2 by copper sites leading either to incorporation of oxygen atoms into an organic substrate (oxygenase activity) or to its oxidation accompanied by O2 reduction usually to water (oxidase activity) [9, 17, 18]. In the context of this chapter, those synthetic models capable of reproducing the activity exhibited by tyrosinase (the ortho-hydroxylation of phenols) are reviewed. From a structural point of view, this enzyme possesses a dinuclear coupled center also referred as type 3 copper site. However, before attempting to

11.2 Copper Proteins Involved in O2 Activation

HN

HN N

H N

N

HN N

4.6 Å

CuI

CuI

N

N NH

N

NH

NH

+

2

−

2

1.4 Å

H N

N N N

N H

HN

CuII

CuII

N 3.6 Å NH

269

N N N H

Oxy form

Deoxy form

rRaman: n(O – O) = 750 cm−1 UV– vis: l = 345 nm (20 000 M−1 cm−1), 570 nm (1000 M−1 cm−1) EPR silent Scheme 11.1 Structural changes of the dicopper site of hemocyanin upon oxygenation (• denotes oxygen atom from molecular O2 ).

reproduce the structure or reactivity of such an enzyme, the basic properties of type 3 copper proteins must be understood. In Sections 11.2.1 and 11.2.2, the main features of the two prototypical proteins (hemocyanin and tyrosinase) within this group are described. 11.2.1 Hemocyanin

Hemocyanins are large, multisubunit proteins that act as oxygen carriers (reversible O2 binding) in many molluscs and arthropods, analogous to hemoglobin in vertebrates [19]. Their active site is constituted by two copper atoms, each of them coordinated to three histidine residues. In the reduced (deoxy) form, the copper(I) ions show a trigonally distorted coordination geometry and they are separated by 4.6 A˚ (Scheme 11.1) [20]. However, in the oxygenated (oxy) form, the copper centers are oxidized by one electron to the +2 oxidation state and an oxygen molecule is reduced by two electrons to form a peroxide unit (O2− 2 ), which ends up ligated between the two metal sites in a µ − η2 : η2 coordination mode. Each of the ﬁve-coordinated CuII centers possesses a distorted square-pyramidal geometry; the Cu · · · Cu separation of 3.6 A˚ is signiﬁcantly contracted relative to that in the reduced form (Scheme 11.1). The oxy form of hemocyanin has very distinctive spectroscopical signatures [21]. Its resonance Raman (rRaman) spectrum displays the characteristic O–O stretching frequency at ν = 750 cm−1 and UV–vis spectroscopy shows bands at λmax = 345 nm (ε ∼ 20 000 M−1 cm−1 ) and 570 nm (ε ∼ 1000 M−1 cm−1 ) responsible for the intense blue color of the oxidized form of the protein. Finally, no EPR (electron paramagnetic resonance) signal is detected due to strong antiferromagnetic coupling between the two CuII centers (S = 1/2). Several structural studies [22, 23] indicate that in hemocyanin there is a protein domain that shields the access of substrates to the dicopper active center. Because

270

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

OH

OH + O2

R Phenol Scheme 11.2

O OH

Tyrosinase (Phenolase)

O

Tyrosinase

Melanins

(Catecholase) R Catechol

R

ortho -Quinone

Biological function of tyrosinase toward the ﬁnal synthesis of melanins.

this domain interferes with substrate binding, hemocyanin can play only one role, namely, that of an oxygen transporter by the reversible coordination of dioxygen to the active site. 11.2.2 Tyrosinase

Tyrosinase is an essential copper-containing enzyme that is found in all organisms. It is involved in browning processes of skin, hair, and fruit, and in wound healing or the immune response [16]. Tyrosinase initiates the synthesis of brown melanin pigments by catalyzing the ortho-hydroxylation of phenols to catechols (ortho-diphenols) and subsequent two-electron oxidation to ortho-quinones using molecular O2 as the oxidant (Scheme 11.2). The latter reaction is also catalyzed by the related enzyme catechol oxidase, which, however, is unable to mediate the phenol hydroxylation step. The crystal structure of catechol oxidase is available since 1998 [24], and despite the fact that the overall folding of this enzyme is completely different from the one found in hemocyanin, both proteins present an almost identical type 3 copper site. Remarkably, the X-ray structure of tyrosinase was not solved until very recently [25], but the conﬁguration of its active site had been inferred by the similar spectroscopic features with hemocyanin [26]. Thus, for ages it has been well established that tyrosinase possesses an active site almost identical to the one found in hemocyanin and catechol oxidase both in the reduced and in the oxidized form (Scheme 11.1). This structural assignment was further corroborated by the X-ray studies by Matoba et al. on bacterial tyrosinase, a work that also allowed the characterization of the most relevant intermediates of the catalytic cycle of this enzyme [25]. It was found that the active site of tyrosinase and hemocyanin can be superposed showing a high degree of structural similarity [20, 25]. The different reactivity of both enzymes arises from the particular folding of the polypeptidic chain: while the accessibility to the active site in hemocyanin is blocked, the active center of tyrosinase is situated close to the molecular surface, a useful location to ensure the access of the substrate. Moreover, analysis of the crystal structure of tyrosinase indicates that the substrate-binding pocket has a large vacant space above the dicopper center, which allows direct interaction between the substrate (phenol) and the active site [25]. Remarkably, Itoh and coworkers demonstrated that hemocyanin can exhibit efﬁcient monooxygenase activity (phenolase activity analogous to tyrosinase) if the

11.2 Copper Proteins Involved in O2 Activation

HN

(His) N

= (His)

N

(His) N CuI

CuI

N (His)

(His) N

2

N (His)

N (His)

Deoxy form (A)

N (His) CuII CuII N (His) (His) N N (His) Oxy form (B)

(His) N

O H2

OH

+ H+

R

H+

R

R

R

O (His) N

O II

Cu (His) N N H (His) (D)

CuII N (His)

N (His) N (His)

N (His) CuII N (His) N (His)

(His) N

CuII (His) N N (His) (C)

Scheme 11.3 Catalytic cycle proposed for the hydroxylation of a monophenolic substrate by tyrosinase (• denotes oxygen atom from molecular O2 ).

polypeptidic chain is altered by treatment with urea. This observation corroborates the identical active center in both proteins and the critical role that plays the folding of the protein in determining the ultimate function of a speciﬁc enzyme [27]. The proposed mechanism for phenol hydroxylation performed by tyrosinase is depicted in Scheme 11.3 [28]. In the ﬁrst step of the catalytic cycle, the reduced form of tyrosinase (deoxy form A) interacts with molecular O2 to give the oxygenated intermediate (oxy form B) in which a peroxo ligand is coordinated in a µ − η2 : η2 mode to the two CuII centers. At this point, the substrate approaches the active center and it is preoriented by hydrophobic and hydrogen bonding interactions with nearby protein residues. The substrate binds to one of the metal centers (C) and hydroxylation of the aromatic ring occurs. To this end, the O–O axis of the peroxo ligand rotates and it points toward the phenolic ring. The proximity of the ortho position of the phenolic ring to the side-on coordinated peroxo group enables an electrophilic attack of the Cu2 O2 moiety on the aromatic ring, by which concomitant cleavage of the O–O bond occurs, so that the resulting diphenolic unit binds in a bidentate fashion (D). Finally, release of the ortho-quinone product regenerates the deoxy form (A) and a new catalytic cycle occurs. Overall, the active center of tyrosinase accommodates important changes during the catalytic cycle, which are possible only with a highly ﬂexible dicopper site [25].

271

272

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

11.3 O2 Binding and Activation at Biomimetic Cu Complexes

The biological importance of the reactions involved in copper-regulated O2 metabolism and the potential technological utility of the oxidation reactions catalyzed by copper-dependent enzymes have fueled great effort in understanding the reaction mechanisms regulating O2 activation by these metalloproteins. A particular strategy to achieve these goals is the study of the reactivity exhibited by bioinspired synthetic models with O2 [9, 29, 30]. The study of mono- and dinuclear copper complexes with simple ligands reproducing the coordination environment of the metal center in natural proteins has allowed the characterization of a wide range of novel chemical species resulting from the interaction of O2 with copper(I). This Cun O2 type of compounds (generally formed by the oxygenation of CuI complexes) tends to be highly unstable and, in fact, the advance in stabilizing and characterizing them can be attributed partly to a greater accessibility of appropriate spectroscopic tools and to a better appreciation of the appropriate reaction conditions. Low temperatures (∼200 K), aprotic solvents (CH2 Cl2 , THF, acetone), and weak coordination anions are now standard conditions. Owing to their high instability, Cun O2 compounds are called the intermediate species formed upon CuI oxidation to CuII or CuIII accompanied by total or partial reduction of O2 . The identiﬁcation and characterization of these synthetic intermediates are crucial to postulate possible reaction mechanisms in natural enzymes such as tyrosinase [31]. 11.3.1 Copper–Dioxygen Adducts

In the last couple of decades, several CuI complexes have been synthesized and their reactivity toward O2 has been thoroughly studied, which has allowed the determination of different coordination modes in Cun O2 type of systems. The nuclearity (n, number of copper centers activating a single O2 molecule) can range from 1 to 3 depending on the stoichiometry of the reaction [9, 29]. In tyrosinase there is a cooperative interaction between two copper sites to activate O2 (Scheme 11.3), which makes Cu2 O2 specially relevant for the purpose of mimicking such an enzyme. The most outstanding conﬁgurations exhibiting a Cu2 O2 core are trans-µ-1,2-peroxodicopper(II) (T P), µ-η2 : η2 -peroxodicopper(II) (s P), and bis(µ-oxo)dicopper(III) (O) (Figure 11.1). The biological relevance of trans-µ-1,2-peroxodicopper(II) (T P) intermediates has not been proved yet, but they are synthetically accessible through the use of mononuclear copper(I) complexes with tetradentate ligands that self-assemble upon O2 reaction [32, 37–41]. Their structural assignment is based primarily on distinctive intense UV–vis features that are responsible for their purple colors: λmax ∼ 530 nm (ε ∼ 10 000M−1 cm−1 ), and 600 nm (ε ∼ 7000M−1 cm−1 ). rRaman spectroscopy is also an adequate tool to identify this type of species that are characterized by an O–O stretching vibration at ν ∼ 830cm−1 ([18 O2 ] ∼ –45 cm−1 ) and a Cu–O stretching vibration at ν ∼ 555 cm−1 ([18 O2 ] ∼ –24 cm−1 ) [9].

11.3 O2 Binding and Activation at Biomimetic Cu Complexes

H B N

N

N N

N

−

D5

N tpa

N B

Cu

O

O

CuIIL

trans -µ-1,2-Peroxo (TP) (a)

LCuII

Cu

O

CuIIL

Figure 11.1 (a) X-ray structure of the T P species [CuII 2 (O2 )(tpa)2 ]2+ formed upon reaction of [CuI (tpa)(EtCN)]+ with O2 in EtCN at −90 ◦ C [32]. (b) X-ray structure of the S P species [CuII 2 (O2 )(TpiPr2 )2 ] formed upon reaction of [CuI (TpiPr2 )]

O

CuIIIL O Bis(µ-oxo) (O)

LCuIII

O µ-h2:h2-Peroxo (sP) (b)

N D D

N O

O

Cu

N

LCuII

N

D D D21-Bz3tacn

TpiP r2

O

D5

D D N

N N

N N

273

(c) with O2 in acetone at −80 ◦ C [33, 34]. (c) X-ray structure of the O species [CuIII 2 (O)2 (D21 -Bz3 tacn)2 ]2+ formed upon reaction of [CuI (D21 -Bz3 tacn)(CH3 CN)]+ with O2 in CH2 Cl2 at −80 ◦ C [35, 36].

[CuII 2 (O2 )(tpa)2 ]2+ (Figure 11.1a) constitutes the prototypical example of these compounds and it is synthesized by reversible interaction of two [CuI (tpa) (EtCN)]+ units with a single O2 molecule at cryogenic temperatures [32, 37]. µ-η2 : η2 -Peroxodicopper(II) complexes (S P) were ﬁrst discovered by Kitajima and coworkers with the resolution of the X-ray crystal structure of [CuII 2 (O2 )(TpiPr2 )2 ] [33, 34] (Figure 11.1b). Interestingly, the similarity between the spectroscopic features of this compound and those of the oxy form of hemocyanin (Scheme 11.3) strongly supported the notion that O2 binding in the oxidized form of this protein occurred with the same mode, as was later demonstrated by the resolution of the crystal structure of the natural protein [20]. This was a great success of synthetic model chemistry because through the characterization of a simple coordination complex, the correct and unprecedented coordination mode of O2 in a natural system could be established. S P species have very characteristic spectroscopic features. On the one hand, their UV–vis spectra exhibit a high energy charge transfer band at λmax = 340–380 nm (ε ∼ 18 000–25 000 M−1 cm−1 ) and a weaker lower energy band at 510–550 nm (ε ∼ 1000M−1 cm−1 ) conferring their characteristic violet color. On the other hand, their rRaman spectra display a characteristic low-energy O–O stretching vibration at ν = 730–760 cm−1 ([18 O2 ] ∼ –40 cm−1 ) [9].

D5

274

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

The ﬁrst CuIII (µ-O)2 -CuIII core (O) was reported by Tolman in [35] and it presents completely different spectroscopic features compared to the rest of Cu2 O2 intermediates. The ﬁrst structurally characterized O species was generated with the macrocyclic fac triamine ligand Bz3 tacn [35, 36], and despite its instability above −80 ◦ C, it could be fully characterized by several techniques including X-ray analysis of its perdeutero analog (Figure 11.1c). Since Tolman’s discovery, several O type of intermediates have been spectroscopically and, in selected cases, structurally characterized [42–45]. The structural differences between O and S P isomers portend different UV–vis and rRaman spectroscopic characteristics. O complexes generally exhibit two intense ligand-to-metal charge-transfer absorptions at λmax ∼ 300 nm (ε ∼ 20 000 M−1 cm−1 ) and 400 nm (ε ∼ 24 000 M−1 cm−1 ), whilst rRaman experiments reveal a characteristic and intense vibration at ν ∼ 600 cm−1 ([18 O2 ] ∼ –25 cm−1 ) [9]. Despite the fact that O species have never been directly detected in biological systems, it cannot be excluded that they may be involved in biological processes through the O–O bond breakage of the S P core. Indeed, one of the major contributions from Tolman and coworkers was the demonstration of the existence of a rapid equilibrium between S P and O species, thus indicating reversible scission and formation of the dioxygen O–O bond (Scheme 11.4) [46]. This equilibrium was clearly evidenced by the observation of reversible dramatic spectroscopic changes (UV–vis and rRaman) in a particular S P species upon changing the solvent. The new spectroscopic features could be unambiguously assigned to the new O compound; thanks to the resolution of its X-ray structure. Tolman’s work constitutes the ﬁrst precedent of a transition metal complex where the O–O bond is formed and broken reversibly, and because of its signiﬁcance, this equilibrium has been studied in depth [47–49]. It is inﬂuenced primarily by the steric demands of the ligand [47, 50, 51], although solvent [47, 51–53] or electronic features of the ligand [48, 54] can also affect its position.

LCuIII

O

CuIIIL

LCuII

O

CuIIL

O

O

Bis(µ-oxo) (O)

µ-h2:h2-peroxo (sP)

rRaman: n ~ 600 cm−1 (∆[18O2] ~ −25 cm−1)

rRaman: n ~ 750 cm−1 (∆[18O2] ~ −40 cm−1)

UV– vis: lmax ~ 300 nm (e ~ 20 000 M−1 cm−1) 400 nm (e ~ 24 000 M−1 cm−1)

UV– vis: lmax ~ 360 nm (e ~ 22 000 M−1 cm−1) 530 nm (e ~ 1000 M−1 cm−1)

Scheme 11.4 Equilibrium between O and S P complexes along with their spectroscopic features.

11.3 O2 Binding and Activation at Biomimetic Cu Complexes

Mononuclear models

Dinuclear models

Cu+

Cu+

+

Cu+

Cu+

2

2

Cun +1

−2n 2

Cun +1

Cun +1

−2n 2

Cun +1

Scheme 11.5 Schematic representation of the O2 activation by mononuclear and dinuclear synthetic models to form Cu2 O2 species (• denotes oxygen atom from molecular O2 ).

11.3.2 Ligand Architecture: Inﬂuence on Reactivity toward O2

Our understanding on how oxygen can be activated by CuI centers has been advanced enormously thanks to the development of suitable ligand scaffolds. In general, biological models of dinuclear copper proteins that activate oxygen consist in mononuclear CuI complexes that self-assemble upon reaction with molecular O2 (Scheme 11.5, top). However, as described in Section 11.2.2, the active center of tyrosinase is constituted by a dinuclear copper site in which the two metallic ions are disposed in a particular conﬁguration favoring a cooperative interaction with molecular O2 (Scheme 11.3) [25]. For this reason, an important step toward the mimicking of the natural active site of tyrosinase, and in general of dicopper proteins, is the synthesis of dinucleating ligands capable of establishing some control over the spatial arrangement of the two metal centers in the synthetic system [55]. These ligands are constituted by two coordination sites that hold two copper centers close enough to give rise to intramolecular interaction with O2 (Scheme 11.5, bottom). Traditionally, the connection between the two binding units of a dinucleating ligand is done with an alkyl group [56, 57] or an aromatic ring. A particular case of the latter is constituted by meta-xylyl linked dinuclear CuI complexes [58, 59] (dinucleating ligands, Scheme 11.6). Apart from the ligand nuclearity, another important point to design a good mimic of dinuclear copper proteins is the coordination environment: in order to reproduce the histidine-rich environment present in this type of proteins, the copper center should be surrounded by nitrogen atoms. In this sense, the use of ligands containing secondary or tertiary amines, pyridine groups, or even benzimidazole units has succeeded in the development of synthetic models [9, 29] suitable to reproduce the activity of dinuclear copper proteins in general, and tyrosinase in particular. The most relevant works in this ﬁeld are discussed in Section 11.3.3. Another decisive factor that determines the ﬁnal outcome of the interaction between copper(I) centers and molecular oxygen is the ligand ﬂexibility because change in the oxidation state from +1 to +2 or even to +3 entails important modiﬁcations in the disposition of the coordination atoms around the copper center. This phenomenon is even more important in those cases where the copper–dioxygen adduct oxidizes a speciﬁc substrate. In these situations, direct

275

276

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

Mononucleating ligands

N

N N

NH N N

N N

N

N

D D

N

N

NH

N

N

MeAN L6Ph

N

N

N

N

N

N

N

N

N N

N

N

LAA

R R-PhPyNEt2 R = H, NO2, OMe

LPy2

DBED

LGG

LAG

R

Dinucleating ligands R N N

N

N N N

N

N

N

N

N N

N N

N

N

N

N N

N

N

N

N

N

N

N N

N

m -XYLMeAN

Me3m

R-XYL R = H,t Bu, OMe, F, CN, NO2

L66 (R = H) MeL66 (R = Me)

Scheme 11.6 Ligands used for the preparation of biomimetic copper complexes of tyrosinase.

coordination of exogenous substrates to the metal site is generally required causing even greater modiﬁcations in the environment around the Cun O2 core. The effect of ligand topology on oxygen activation by low-molecular-weight copper complexes has been studied in depth by different research groups [55]. Indeed, it has been evidenced that the structure of the ligand is important not only to achieve the desired O2 activation but also to perform tyrosinase-like reactivity (hydroxylation of aromatic rings). Among these studies, a recent work reported by our group clearly demonstrates the decisive role played by the ﬂexibility of the ligand as far as oxygen activation by dicopper centers is concerned [60]. Two different dinucleating nitrogen-based ligands were used: Me3m and m-XYLMeAN (Scheme 11.6). In Me3m, the two binding triamine sites are connected by two meta-xylyl linkers giving a hexaaza macrocycle, while m-XYLMeAN is constituted by two triamine units connected by a single meta-xylyl group. The corresponding dicopper(I) complexes present important similarities (Scheme 11.7): in both cases, the copper sites are coordinated to three aliphatic nitrogen atoms and they possess a distorted trigonal planar geometry. The coordination sphere of the metal centers is structurally related to the one found in tyrosinase, in which each copper ion is coordinated to three histidine residues in the reduced form [25]. Finally, [CuI (MeAN)]+ (Schemes 11.6 and 11.7) constitutes a mononuclear analog of the above-mentioned dinuclear complexes.

11.3 O2 Binding and Activation at Biomimetic Cu Complexes

N

N I

N Cu N

Cu

I

N N

I

N

N

N I

N

Cu Cu N

N N CuI N

2

N [CuI(MeAN)]+

I

MeAN 2+

[Cu 2(m -XYL [CuI2(Me3m)]2+ 2

)]

Acetone −80 °C

2

N

Slow decomposition to CuII species

N N

CuIII N

N CuIII

N

N

N CuII N

CH2Cl2 −80 °C

N CuII N N

[CuII2(µ-h2:h2-O2)(MeAN)2]2+

[CuIII2(µ-O)2(m -XYLMeAN)]2+ Scheme 11.7 Reactivity toward O2 of three structurally related dinuclear and mononuclear CuI complexes (• denotes oxygen atom from molecular O2 ) [60].

Its X-ray structure [61] shows a copper(I) center with a distorted trigonal planar geometry structurally equivalent to the one found in its dinuclear counterparts. Despite the almost identical structure of the copper(I) centers in the three complexes, their reactivity toward O2 was found to be completely dependent on the speciﬁc ligand. [CuI 2 (Me3m)]2+ reacted slowly with O2 even at room temperature and no metal–oxygen adduct was accumulated or detected in the course of the reaction. In contrast, reaction of [CuI 2 (m-XYLMeAN )]2+ in acetone at−80 ◦ C was much faster and a bis(µ-oxo) core (O) was trapped and spectroscopically characterized by UV–vis (λmax = 308 nm, ε = 20 000 M−1 cm−1 ; λmax = 413 nm, ε = 28 000 M−1 cm−1 ) and rRaman spectroscopy (ν = 600 cm−1 , [18 O2 ] = −23 cm−1 ). Finally, for the mononuclear analog [CuI (MeAN)]+ , a S P species, formed upon reaction of two complex units with a single O2 molecule, was exclusively detected [61]. Given the comparable coordination sphere and electronic properties of the CuI ions in the three complexes, it is rather unlikely that the initial O2 binding to a single CuI depends on the particular complex, and therefore the different O2 reactivity highlights the important role played by the second metal ion. In [CuI 2 (m-XYLMeAN )]2+ , the ligand is ﬂexible, allowing copper sites to approach close enough to promote their synergistic actuation in O2 binding reduction, which is also true for its mononuclear analog [CuI (MeAN)]2+ . Instead, the rather rigid nature of the macrocyclic ligand (Me3m) imposes a higher barrier to this process,

277

278

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

shutting down the reaction. Thus, this rather simple model highlights the importance of cooperative effects between two copper centers to activate O2 . Further support for this concept comes from the recent observation that the active site of the natural enzyme tyrosinase accommodates remarkable changes occurring along its catalytic cycle, which require the presence of a rather ﬂexible active center [25]. 11.3.3 Hydroxylation of Aromatic Rings: Mimicking Tyrosinase Activity

The reactivity of several Cu2 O2 species has been studied and analyzed in detail during the last two decades. The biological relevance of these species and the potential technological interest of the chemical transformations they perform have fueled great effort toward the understanding of basic aspects of their chemistry, such as their electrophilic/nucleophilic character or their acid/base behavior [29]. However, the main objective on synthetic models is reproducing the reactivity performed by enzymes. In particular, in this section, the focus is on those synthetic systems that reproduce the reactivity exhibited by tyrosinase: the hydroxylation of aromatic rings. These model systems contain a Cu2 O2 type of species (analogous to the dinuclear copper site present in the natural system) and they may be divided in two categories: (i) compounds that perform an intramolecular ligand aromatic hydroxylation and (ii) complexes that carry out the ortho-hydroxylation of exogenous phenolic compounds (natural substrates). In the following sections, the most outstanding examples of both classes have been reviewed. 11.3.3.1 Intramolecular Aromatic Hydroxylation Generally, copper–dioxygen species are stable only at low temperatures (−80 ◦ C) and warming up usually causes decomposition, typically to CuII complexes, via processes involving intramolecular oxidation of a supporting ligand. These processes can be considered as destructive and thus they may be undesired from the viewpoint of catalyst development, but they provide fundamental mechanistic information relevant to the reactivity of Cun O2 . As far as Cu2 O2 species is concerned, N-dealkylation [42, 62] or aliphatic hydroxylation [63] has been observed to occur especially upon formation of O and S P isomers. Another intramolecular oxidation occurring upon decomposition of selected Cu2 O2 adducts is the hydroxylation of an aromatic moiety included in the ligand structure. This reaction has attracted much attention of the scientiﬁc community, not only for the inherent interest of this transformation itself but also due to the reminiscence of this process to the activity exhibited by tyrosinase. Probably, the most deeply studied systems exhibiting this type of reactivity were described by Karlin and coworkers using dinucleating ligands of the R-XYL family (Scheme 11.6) [58, 59, 64]. Reaction of complexes [CuI 2 (R-XYL)]2+ with oxygen at room temperature resulted in the formation of [CuII 2 (R-XYL-O)(OH)]2+ , with the two oxygen atoms incorporated into the oxidized product coming from O2 (Scheme 11.8). Monitoring the reaction by UV–vis spectroscopy at low temperatures afforded the detection of a transient copper–dioxygen adduct when electron-withdrawing R groups (such

11.3 O2 Binding and Activation at Biomimetic Cu Complexes R

R

279 R

R +

N N

CuI N

N CuI N

N CuII

2

N

[CuI2(R-XYL)]2+

py

N CuII py

py

N CuII py

[CuII2(O2)(R-XYL)]2+

py

py

H

py

N CuII

N CuII

N CuII py

py

py

H

py

py

[CuII2(R-XYL-O)(OH)]2+

(sP)

Scheme 11.8 Mechanism of intramolecular aromatic ligand hydroxylation performed by [CuI 2 (R-XYL)]2+ system (• denotes oxygen atom from molecular O2 ).

as NO2 ) were used, which was unambiguously assigned as a S P species on the basis of its spectroscopic features (UV–vis: λmax = 358 nm, ε = 20 000 M−1 cm−1 ; λmax = 435 nm, ε = 5000 M−1 cm−1 ; λmax = 530 nm, ε = 1200 M−1 cm−1 . rRaman: ν = 747 cm−1 , [18 O2 ] = −40 cm−1 ); further support for this assignation comes from density functional theory (DFT) calculations. An electrophilic attack of the S P species over the m-xylyl ring followed by an intramolecular proton transfer was the proposed mechanism to explain the intramolecular ligand aromatic hydroxylation reaction, and it was supported by several experimental observations: (i) the lack of hydrogen/deuterium kinetic isotope effect indicated that the rate-determining step was the initial attack of the S P unit on the aromatic ring rather than a hydrogen atom abstraction event and (ii) the decrease in the hydroxylation rate upon increasing the electron-withdrawing character of the R group (measured Hammett parameter ρ = –2.1). The reactivity of several other dinuclear copper(I) systems with O2 also causes the intramolecular hydroxylation of the aromatic moiety of the ligand. However, in most cases, little is known about the exact mechanism of this process as no Cun O2 adducts are experimentally detected [65–70]. By analogy to Karlin’s system and on the basis of theoretical analyses [71], it is generally proposed that a S P type of species is responsible for the observed reactivity. Interestingly, a recent theoretical study shows that the mechanism of intramolecular aromatic hydroxylation by a dinuclear copper(I) complex upon reaction with O2 favors the participation of a S P species. However, the small energetic differences with the corresponding O isomer do not allow to deﬁnitively rule out the involvement of the bis(µ-oxo) core in this hydroxylation process [72]. It is interesting to notice here that there is a reported example in which a Cu2 O2 species different from the S P structure can perform the intramolecular aromatic hydroxylation [73]. In this work, it was observed that decomposition of the O compound supported by R-PhPyNEt2 ligands (R = NO2 ; UV–vis: λmax = 404 nm, ε = 13 000 M – 1 cm−1 ; rRaman: ν = 606 cm−1 , [18 O2 ] = –27 cm−1 ) resulted in partial ligand phenyl ring hydroxylation (Scheme 11.9). The decomposition of the Cu2 O2 species clearly involved the reaction with the aryl group because the nature of the aromatic substituents signiﬁcantly inﬂuenced the decay rate. Similar to

280

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

R

Et2 N CuIII N

R

N CuIII N Et2

1. Warm 2. Aq. NH4OH (- Cu ions)

R

N H

NEt2

+

R

N NEt2

[CuIII2(µ-O)2(R-PhPyNEt2)2]2+ (O)

Scheme 11.9 Thermal decomposition of [CuIII 2 (µ-O)2 (R-PhPyNEt2 )2 ]2+ leading to partial hydroxylation of the aromatic phenyl ring (• denotes oxygen atom from molecular O2 ) [73].

what was observed in Karlin’s system, there was almost no kinetic isotope effect when the phenyl ring was fully deuterated and the rate of decay increased upon increasing the electron-donating properties of the R group. Thus, from this work it seems clear that an electrophilic aromatic substitution pathway is also viable with O type of species. Nevertheless, even though no indication of the presence of a S P isomer was found, hydroxylation by a small, undetectable amount of such an isomer in rapid equilibrium with the experimentally detected O core could not be ruled out in this case. 11.3.3.2 Intermolecular ortho-Hydroxylation of Phenolic Compounds The reactivity of Cu2 O2 species toward exogenous substrates is specially relevant to catalysis and it has been thoroughly studied. These reactions include oxygen atom transfer, hydrogen atom abstraction, oxidation of alcohols to aldehydes or ketones [74–76], and conversion of amines to nitriles [77, 78], among others. However, a specially relevant reaction performed by some Cu2 O2 adducts is the ortho-hydroxylation of phenolic compounds to catechols or quinones [79], due to its direct relevance to the catalytic transformation performed by tyrosinase. Casella and coworkers reported the ﬁrst synthetically prepared Cu2 O2 adduct capable of performing the ortho-hydroxylation of exogenous phenolates to give the corresponding ortho-diphenols (catechols) [80]. It consisted in a dinuclear copper complex bearing the dinucleating ligand L66 that combines an aliphatic amine and two benziimdazole units as the coordinating groups (Scheme 11.6). Reaction of [CuI 2 (L66)]2+ with O2 at low temperature (−78 ◦ C) afforded a species with a UV–vis spectrum (λmax = 362 nm, ε = 15 000 M−1 cm−1 ; λmax = 455 nm, ε = 2000 M−1 cm−1 ) and a rRaman pattern (ν = 760 cm−1 , [18 O2 ] = −41 cm−1 ) fully consistent with a S P compound (Scheme 11.10). Reaction of this transient Cu2 O2 species with p-carbomethoxyphenolate at −60 ◦ C gave the corresponding catechol product in about 40% yield (with respect to the dioxygen adduct formed). The yield of catechol product increased up to 90% when the related S P complex [CuII 2 (O2 )2 (MeL66)]2+ reacted with the phenolic substrate [81]. Interestingly,

11.3 O2 Binding and Activation at Biomimetic Cu Complexes

281

R CuII N

N

II

D

N

N Cu

Cu N

D

D N

N N

II

Cu

D N

N

N

N II

Cu

II

N

N

CuII O−

X

N

X

N O CuII

[CuII2(O2)(LPy2)2 ] 2+ (sP)

II

[Cu 2(O2)(L66)]

2+ s

II

( P) or [Cu 2(O2)(MeL66)]

CuII

2+ s

( P) X

H X

O−

X

OH O

p - Substituted phenolate X = Me, F, Cl, COOMe, CN

Catechol product

CuII

CuII H

Scheme 11.10 Reported synthetic S P species that perform the ortho-hydroxylation of exogenous phenolates to form catechols (left) along with their proposed mechanism of action (right) [81, 83] (• denotes oxygen atom from molecular O2 ).

reaction of [CuI (L6Ph)]+ , the mononuclear analog of [CuI 2 (L66)]2+ , with dioxygen at low temperatures also allowed the detection of a dinuclear S P species capable of oxidizing exogenous phenolates to catechols (oxidation of sodium p-carbomethoxyphenolate was obtained in 88% yield with respect to the theoretical amount of Cu2 O2 species) [82]. One year after the publication of the work by Casella and coworkers, another synthetic system capable of performing a similar reaction was reported by Itoh et al. [83]. In this case, the tridentate mononucleating LPy2 ligand (Scheme 11.6) was used and a S P compound was unambiguously formed upon self-assembling of two mononuclear [CuI (LPy2 )]+ complexes by reaction with O2 at low temperatures. Reaction of a series of lithium salts of para-substituted phenolates with the newly formed S P species resulted in the formation of the corresponding catechol derivates in good yields (ranging from 60 to 90% depending on the speciﬁc substrate) (Scheme 11.10). Both in Itoh (LPy2 ) and Casella’s (L66, MeL66, or L6Ph) systems, it was well established that the oxygen atom incorporated into the catechol product derived from O2 (as ascertained by experiments using 18 O2 ), and that the rate of hydroxylation increased upon increasing the electron-donating abilities of the para-substituents in the phenolic ring (Hammett analysis afforded ρ values of −1.8 for LPy2 and −1.84 for MeL66). Finally, rate constants of the reaction between S P and phenolates were dependent on substrate concentration following Michaelis–Menten type saturation curves. Such a kinetic behavior is indicative of the formation of a complex constituted by an association between the substrate and the peroxo intermediate in the course of reaction. These results are consistent with an electrophilic attack of the S P intermediate to the phenolate ring posterior to the formation of a binary complex between the peroxo core and the phenolic substrate

282

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

(Scheme 11.10, right). The similarities between this mechanism of action and the one proposed for tyrosinase suggest that these synthetic systems can be thought as true models for the natural enzyme. Remarkably, the reaction of [CuII 2 (O2 )(LPy2 )2 ]2+ with neutral phenols (rather than phenolate substrates) resulted in the exclusive formation of C–C coupling dimer products (no catechol was formed) [84]. Presumably, this reaction occurs through a proton-coupled electron transfer mechanism (PCET) by which phenoxyl radical intermediates are generated and they spontaneously collapse. In the biological system (Section 11.2.2), it is proposed that the neutral phenol acting as the natural substrate becomes readily deprotonated upon interaction with the active center, and thus it is the corresponding phenolate that directly coordinates to the copper site. Instead, phenol deprotonation in synthetic systems is not possible, thus precluding direct use of neutral phenols to model tyrosinase reactivity and enforcing reaction with previously deprotonated phenolate salts. Overall, the results arising from the works by Casella and Itoh on synthetic models of tyrosinase indicate that the presence of a S P type of species is necessary to perform the ortho-hydroxylation of exogenous phenolic substrates. This idea is further supported by the fact that this type of peroxo unit is also directly detected in the natural system where the proposed mechanistic pathway entails an O–O bond cleavage occurring concomitantly with hydroxylation (Scheme 11.3). Nevertheless, the aromatic hydroxylation mechanism continues being subject of intense discussion because the existence of a rapid equilibrium between S P and O species [46] (Scheme 11.4) hinders the assignment of the true hydroxylation species in natural systems. Moreover, some recent examples reported in the last three years clearly indicate that the O unit is also competent to perform tyrosinase-like reactivity. The ﬁrst example was reported by Stack and coworkers (Scheme 11.11) [77, 85, 86]. In this case, reaction of the mononuclear [CuI (DBED)]+ with O2 at −80 ◦ C resulted in the formation of [Cu2 (O2 )(DBED)2 ]2+ , which was unequivocally assigned as a S P species on the basis of its characteristic spectroscopic features (UV–vis: λmax = 350 nm, ε = 36 000 M−1 cm−1 ; 485 nm, ε = 1200M−1 cm−1 ; 605 nm, ε = 900 M−1 cm−1 . rRaman: ν = 721 cm−1 , [18 O2 ] = −40 cm−1 ) [85]. Reaction of this Cu2 O2 species with sodium 2,4-di-tert-butylphenolate at −80 ◦ C afforded a 1 : 1 mixture of 3,5-di-tert-butylcatechol and 3,5-di-tert-butyl-1,2-benzoquinone products (∼80% yield based on the initial amount of copper complex). More interestingly, when the reaction between S P species and the phenolic substrate was monitored by UV–vis spectroscopy at extremely low temperatures (−120 ◦ C), the formation of a transient intermediate assigned as a bis(µ-oxo) core (O) with a coordinated phenolic substrate was observed [86]. The transformation of the initially detected S P species into a O compound upon substrate coordination was clearly evidenced by rRaman spectroscopy, in which a characteristic bis(µ-oxo) stretching frequency at 590 cm−1 ([18 O2 ] = −20 cm−1 ) appeared once the substrate was added. Detailed mechanistic studies on the hydroxylation reactions indicated that electron-deﬁcient phenolic substrates caused a decrease in the decay rate (Hammett parameter ρ = –2.2), which supports a reaction path involving an electrophilic aromatic substitution.

11.3 O2 Binding and Activation at Biomimetic Cu Complexes

+2

N H

N H O2 CuI NCMe MeTHF N −120 °C H [CuI(DBED)]+

t

O−

Bu

H N

N H N O CuIII CuIII O O H N

N

t

Bu

O CuII Cu O N N H H II

283

MeTHF −120 °C

t

Bu

OH t

O OH tBu +

Bu t

t

Bu

t

Bu 30% [CuIII2(µ-O)2(phenolate)(DBED)2]+ (O)

[CuII2(O2)(DBED)2]2+ (sP)

O Bu

30%

Scheme 11.11 Schematic representation of the O–O bond breakage prior to substrate hydroxylation in [CuI (DBED)]+ system [86] (• denotes oxygen atom from molecular O2 ).

This mechanism was further favored by the observed inverse secondary C–H/C–D isotope effect and by DFT calculations modeling the hydroxylation process. Therefore, in this work, it was spectroscopically observed not only that the O–O bond cleavage occurs prior to the hydroxylation but also that the O isomer is the real executor of the chemistry (Scheme 11.11). This observation contrasts with the above mentioned idea that formulates the true hydroxylation species as a side-on peroxo (S P) type of intermediate. More recently, our group reported that O species [CuIII 2 (O)2 (m-XYLMeAN )]2+ formed upon reaction of the dinuclear copper complex [CuI 2 (m-XYLMeAN )]2+ with O2 is also capable of performing the ortho-hydroxylation of phenolates to the corresponding catechols (Scheme 11.12) [87]. In fact, sodium para-chlorophenolate was oxidized into 4-chlorocatechol in 67% yield with respect to the initial dicopper complex. Moreover, UV–vis monitoring of the hydroxylation reaction indicated the formation of a new brown intermediate (UV–vis: λmax = 390 nm, ε = 8000 M−1 cm−1 ; 485 nm, ε = 4500 M−1 cm−1 ) showing a rRaman spectrum with the characteristic bis(µ-oxo) core vibration frequency at ν = 597cm−1 ([18 O2 ] = −26 cm−1 ) together with vibrations corresponding to the phenolic

O−Na+

N N N

CuIII

CuIII

N

X

N

X = Cl, F, CN, CO2Me

N

N

Acetone, −80 °C

N

N

[CuIII2(µ-O)2(m -XYLMeAN)]2+ (O)

CuIII

N CuIII N N O

X [CuIII2(µ-O)2(phenolate)(m -XYLMeAN)]+ (O)

Scheme 11.12 ortho-Hydroxylation of phenolates by the O species [CuIII 2 (µ-O)2 (m-XYLMeAN )]2+ [87] (• denotes oxygen atom from molecular O2 ).

OH X

H

284

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

substrate. No isotope-sensitive features that could be assigned to a S P species were observed. This rRaman data together with the similarity with the UV–vis spectrum of [CuIII 2 (µ-O)2 (phenolate)(DBED)2 ]+ [86] provided direct evidence for phenolate binding to the Cu2 O2 in this newly formed transient intermediate that could be formulated as [CuIII 2 (µ-O)2 (phenolate)(m-XYLMeAN )]+ . In analogy to the previously reported systems, the rate of hydroxylation increased upon increasing the electron-donating properties of the phenolate, which is indicative of an electrophilic oxidizing species that attacks the aromatic ring. Thus, the system reported by our group is similar to the previous one described by Stack and coworkers in the sense that the O isomer is the real executor of the hydroxylation. However, a signiﬁcant difference can be found between these two systems: while in [CuIII 2 (µ-O)2 (m-XYLMeAN )]2+ the O–O bond breakage occurs prior to phenolate coordination, in [CuII 2 (O2 )(DBED)]2+ the O–O breakage is concomitant with phenolate binding. Substrate accessibility to Cu2 O2 cores is a very important factor in determining the nature of the oxidative process. Indeed, the active sites of hemocyanin and tyrosinase are almost identical, but the different substrate accessibility to the copper center determines their different ﬁnal activity. Taking this concept into account, it is interesting to refer to a recent work reported by Stack and coworkers [88] in which a series of bidentate ligands were prepared: they contain either two nonbulky peralkylated diamine units (LAA ), or two sterically crowded guanidine coordinating groups (LGG ), or a hybrid of both constituted by one guanidine and one amine group (LAG ) (Scheme 11.6). The corresponding mononuclear copper(I) complexes self-assemble upon reaction with O2 to form [CuIII 2 (µ-O)2 (L)2 ]2+ assigned as O species on the basis of UV–vis (λmax ∼ 297 nm, ε ∼ 19 000 M−1 cm−1 ; λmax ∼ 390 nm, ε ∼ 23 000 M−1 cm−1 ) and EXAFS spectroscopy. Despite the similar bis(µ-oxo) core, their reactivity toward exogenous phenolates was found to be completely different (Scheme 11.13). The O species coordinated to LGG ligands did not react with phenolates presumably due to the hindered access of the phenolic substrates caused by the steric demands of the four guanidine units surrounding the copper sites. Instead, the O species formed with the peralkylated diamine ligand (LAA ) (with minimal steric demands) reacted with phenolates to form radical C–C coupling dimers by an electron transfer mechanism involving initial electron transfer from the phenolate to the CuIII center. Finally, when hybrid ligands LAG were present, the resulting O core exhibited tyrosinase-like reactivity and it performed the ortho-hydroxylation of phenolates to give the corresponding catechols. Presumably, the substitution of one of the amine groups by a guanidine results in a less oxidant copper center (at parity of oxidation state, a guanidine is much stronger nitrogen σ -donor than a peralkylated amine) and thus outer-sphere electron transfer mechanisms are diminished. Overall, this work indicates that ligand LAG forms an O species that balances both Cu2 O2 accessibility and moderate oxidizing power, which allows efﬁcient hydroxylation of phenols.

11.4 Concluding Remarks

N

N CuIII

N

N

CuIII

N

CuIII

N

N

N

N

N N

CuIII N

N

N

N

N

N CuIII

CuIII

N

N

N

285

N N

N

N

[CuIII2(µ-O)2(LAA)2]2+ (O) O

−

[CuIII2(µ-O)2(LAG)2]2+ (O)

[CuIII2(µ-O)2(LGG)2]2+ (O)

O−

OH

O−

OH

No reaction H

HO (a)

C – C coupling

(b) ortho -Hydroxylation

(c)

Scheme 11.13 Reactivity toward exogenous phenolates of three related O compounds with LAA (a), LAG (b), and LGG (c) ligands [88] (• denotes oxygen atom from molecular O2 ).

11.4 Concluding Remarks

In the past decades enormous effort has been devoted to the development of suitable biomimetic systems that can structurally and even functionally reproduce the active site of a speciﬁc protein. These studies have decisively contributed to a better understanding on how nature works and they are considered as a powerful tool to get fundamental insight into the mechanistic and structural details of biological systems. In this review chapter, we have focused on the description of those synthetic model systems capable of reproducing the activity of tyrosinase, a dinuclear copper protein that catalyzes the ortho-hydroxylation of phenols to catechols. Despite the fact that the fundamental structure of this enzyme is nowadays ﬁrmly established by X-ray diffraction studies [25], some key aspects of its mechanism of action remain unclear, such as the characterization of the true hydroxylation species. On the one hand, the spectroscopic detection of a µ-η2 : η2 -peroxodicopper(II) species (S P) in the natural system together with the ability of some S P model systems to perform the ortho-hydroxylation of phenols [81, 83] suggests that this type of conﬁguration is responsible for the hydroxylation event. On the other hand, the recent development of selected functional models of tyrosinase containing a bis(µ-oxo)-dicopper(III) core (O) together with theoretical studies postulating the electrophilic character of such a structure [89] indicates that O is also competent to perform the hydroxylation of phenols. Nevertheless, the

286

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

existence of a well-known equilibrium between S P and O isomers (reversible O–O bond breakage) [46] precludes the clear identiﬁcation of the real oxidant. Indeed, the two isomers may be competent for such a transformation and it might be just a matter of ﬁnding dinuclear copper centers that balance substrate accessibility and suitable oxidizing power [88]. Future work in mimicking tyrosinase activity will be directed toward the transformation of the stoichiometric processes achieved by the reported functional models into catalytic and synthetically useful reactions to perform the selective oxidation of phenols. Most likely, these studies will require the use of adequate exogenous reducing agents in order to complete the catalytic cycle. In spite of the need for further work in this topic, all the successful accomplishments in mimicking tyrosinase’s activity and its active site must be considered as a landmark for the performance of equivalent studies with other O2 -activating copper proteins. Indeed, biomimetic studies on mononuclear and multinuclear copper oxidases and oxygenases are still in their initial stages of development, especially compared to dinuclear copper proteins. All knowledge acquired during the past decades to model such proteins will be undoubtedly essential for an efﬁcient and successful mimicking of other proteins involved in O2 activation.

References 1. B¨ackvall, J.-E. (2004) Modern Oxidation

2. 3.

4. 5.

6.

7. 8. 9.

10.

Methods, Wiley-VCH Verlag GmbH, Weinheim. Sawyer, D.T. (1991) Oxygen Chemistry, Oxford University Press, New York. Bertini, I., Gray, H.B., Stiefel, E.I., and Valentine, S.J. (2007) Biological Inorganic Chemistry: Structure and Reactivity, University Science Books, Sausalito. Kovaleva, E.G. and Lipscomb, J.D. (2008) Nat. Chem. Biol., 4, 186. Holm, R.H., Kennepohl, P., and Solomon, E.I. (1996) Chem. Rev., 96, 2239. Costas, M., Mehn, M.P., Jensen, M.P., and Que, L. Jr. (2004) Chem. Rev., 104, 939. Seigbahn, P.E.M. and Borowski, T. (2006) Acc. Chem. Res., 39, 729. Nam, W. (2007) Acc. Chem. Res., 40, 465. Mirica, L.M., Ottenwaelder, X., and Stack, T.D.P. (2004) Chem. Rev., 104, 1013. Kraatz, H.-B. and Metzler-Nolte, N. (2006) Concepts and Models in Bioinorganic Chemistry, Wiley-VCH Verlag GmbH, Weinheim.

11. Que, L. Jr. and Tolman, W.B. (2008) Na-

ture, 455, 333. 12. Kaim, W. and Schwederski, B. (1998)

13.

14.

15. 16.

17.

18.

19.

Bioinorganic Chemistry: Inorganic Elements in the Chemistry of Life, John Wiley & Sons, Ltd, Chichester. Cotton, F.A., Wilkinson, G., Murillo, C.A., and Bochmann, M. (1999) Advanced Inorganic Chemistry, 6th edn, John Wiley & Sons, Inc., New York. Lippard, S.J. and Berg, J.M. (1994) Principles of Bioinorganic Chemistry, University Science Books, Mill Valley. Klinman, J.P. (1996) Chem. Rev., 96, 2541. Solomon, E.I., Sundaram, U.M., and Machonkin, T.E. (1996) Chem. Rev., 96, 2563. Bento, I., Carrondo, M.A., and Lindley, P.F. (2006) J. Biol. Inorg. Chem., 11, 539. Solomon, E.I., Chen, P., Metz, M., Lee, S.-K., and Palmer, A.E. (2001) Angew. Chem. Int. Ed., 40, 4570. van Holde, K.E., Miller, K.I., and Decker, H. (2001) J. Biol. Chem., 276, 15563.

References 20. Cuff, M.E., Miller, K.I., van Holde, K.E.,

21.

22. 23.

24.

25.

26.

27.

28.

29. 30. 31. 32.

33.

34.

35.

36.

37.

and Hendrickson, W.N. (1998) J. Mol. Biol., 278, 855. Eickman, N.C., Himmelwright, R.S., and Solomon, E.I. (1979) Proc. Natl. Acad. Sci. U.S.A., 76, 2094. Decker, H. and Rimke, T. (1998) J. Biol. Chem., 273, 25889. Himmelwright, R.S., Eickman, N.C., LuBien, C.D., and Solomon, E.I. (1980) J. Am. Chem. Soc., 102, 5378. Klabunde, T., Eicken, C., Sacchettini, J.C., and Krebs, B. (1998) Nat. Struct. Biol., 5, 1084. Matoba, Y., Kumagai, T., Yamamoto, A., and Sugiyama, M.J. (2006) J. Biol. Chem., 281, 8981. Himmelwright, R.S., Eickman, N.C., LuBien, C.D., Lerch, K., and Solomon, E.I. (1980) J. Am. Chem. Soc., 102, 7339. Morioka, C., Tachi, Y., Suzuki, S., and Itoh, S. (2006) J. Am. Chem. Soc., 128, 6788. Decker, H., Schweikardt, T., and Tuczek, F. (2006) Angew. Chem. Int. Ed., 45, 4546. Lewis, E.A. and Tolman, W.B. (2004) Chem. Rev., 104, 1047. Schindler, S. (2000) Eur. J. Inorg. Chem., 2311. Decker, H., Dillinger, R., and Tuczek, F. (2000) Angew. Chem. Int. Ed., 39, 1591. Jacobson, R.R., Tyekl´ar, Z., Farooq, A., Karlin, K.D., Liu, S., and Zubieta, J. (1988) J. Am. Chem. Soc., 110, 3690. Kitajima, N., Fujisawa, K., and Moro-oka, Y. (1989) J. Am. Chem. Soc., 111, 8975. Kitajima, N., Fujisawa, L., Fujimoto, C., Moro-oka, Y., Hashimoto, S., Kitagawa, T., Toriumi, K., Tatsumi,K., and Nakamura, A. (1992) J. Am. Chem. Soc., 114, 1277. Mahapatra, S., Halfen, J.A., Wilkinson, E.C., Pan, G., Wang, X., Young, V.G. Jr., Cramer, C.J., Que, L. Jr., and Tolman, W.B. (1996) J. Am. Chem. Soc., 118, 11555. Halfen, J.A., Mahapatra, S., Wilkinson, E.C., Kaderli, S., Young, V.G. Jr., Que, L.Jr., Zuberb¨uhler, A.D., and Tolman, W.B. (1996) Science, 271, 1397. Tyekl´ar, Z., Jacobson, R.R., Wei, N., Murthy, N.N., Zubieta, J., and Karlin,

38.

39.

40.

41.

42.

43.

44.

45.

46. 47.

48.

49.

50.

51. 52.

K.D. (1993) J. Am. Chem. Soc., 115, 2677. Becker, M., Heinemann, F.W., and Schindler, S. (1999) Chem. Eur. J., 5, 3124. Weitzer, M., Schatz, M., Hampel, F., Heinemann, F.W., and Schindler, S. (2002) J. Chem. Soc., Dalton Trans., 5, 686. Weitzer, M., Schindler, S., Brehm, G., Schneider, S., H¨ormann, E., Jung, B., Kaderli, S., and Zuberb¨uhler, A.D. (2003) Inorg. Chem., 42, 1800. B¨orzel, H., Comba, P., Katsichtis, C., Kiefer, W., Lienke, A., Nagel, V., and Pritzkow, H. (1999) Chem. Eur. J., 5, 1716. Mahadevan, V., Hou, Z., Cole, A.P., Root, D.E., Lal, T.K., Solomon, E.I., and Stack, T.D.P. (1997) J. Am. Chem. Soc., 119, 11996. Cole, A.P., Mahadevan, V., Mirica, L.M., Ottenwaelder, X., and Stack, T.D.P. (2005) Inorg. Chem., 44, 7345. Spencer, D.J.E., Aboelella, N.W., Reynolds, A.M., Holland, P.L., and Tolman, W.B. (2002) J. Am. Chem. Soc., 124, 2108. Aboelella, N.W., Lewis, E.A., Reynolds, A.M., Brennessel, W.W., Cramer, C.J., and Tolman, W.B. (2002) J. Am. Chem. Soc., 124, 10660. Tolman, W.B. (1997) Acc. Chem. Res., 30, 227. Mahadevan, V., Henson, M.J., Solomon, E.I., and Stack, T.D.P. (2000) J. Am. Chem. Soc., 122, 10249. Hatcher, L.Q., Vance, M.A., Sarjeant, A.A.N., Solomon, E.I., and Karlin, K.D. (2006) Inorg. Chem., 45, 3004. Maiti, D., Woertink, J.S., Sarjeant, A.A.N., Solomon, E.I., and Karlin, K.D. (2008) Inorg. Chem., 47, 3787. Lam, B.M.T., Halfen, J.A., Young, V.G. Jr., Hagadorn, J.R., Holland, P.L., Lled´os, A., Cucurull-S´anchez, L., Novoa, J.J., Alvarez, S., and Tolman, W.B. (2000) Inorg. Chem., 39, 4059. Stack, T.D.P. (2003) Dalton Trans., 1881. Liang, H.-C., Henson, M.J., Hatcher, L.Q., Vance, M.A., Zhang, C.X., Lahti, D., Kaderli, S., Sommer, R.D., Rheingold, A.L., Zuberb¨uhler, A.D.,

287

288

11 O2 Reactivity at Model Copper Systems: Mimicking Tyrosinase Activity

53. 54.

55. 56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

Solomon, E.I., and Karlin, K.D. (2004) Inorg. Chem., 43, 4115. Cahoy, J., Holland, P.L., and Tolman, W.B. (1999) Inorg. Chem., 38, 2161. Henson, M.J., Vance, M.A., Zhang, C.X., Liang, H.-C., Karlin, K.D., and Solomon, E.I. (2003) J. Am. Chem. Soc., 125, 5186. De, A., Mandal, S., and Mukherjee, R. (2008) J. Inorg. Biochem., 102, 1170. Mahapatra, S., Young, V.G.Jr., Kaderli, S., Zuberb¨uhler, A.D., and Tolman, W.B. (1997) Angew. Chem. Int. Ed. Engl., 36, 130. Kodera, M., Katayama, K., Tachi, Y., Kano, K., Hirota, S., Fujinami, S., and Suzuki, M. (1999) J. Am. Chem. Soc., 121, 11006. Karlin, K.D., Nasir, M.S., Cohen, B.I., Cruse, R.W., Kaderli, S., and Zuberb¨uhler, A.D. (1994) J. Am. Chem. Soc., 116, 1324. Pidcock, E., Obias, H.V., Zhang, C.X., Karlin, K.D., and Solomon, E.I. (1998) J. Am Chem. Soc., 120, 7841. Company, A., Lamata, D., Poater, A., Sol`a, M., Rybak-Akimova, E.V., Que, L. Jr., Fontrodona, X., Parella, T., Llobet, A., and Costas, M. (2006) Inorg. Chem., 45, 5239. Liang, H.-C., Zhang, C.X., Henson, M.J., Sommer, R.D., Hatwell, K.R., Kaderli, S., Zuberb¨uhler, A.D., Rheingold, A.L., Solomon, E.I., and Karlin, K.D. (2002) J. Am. Chem. Soc., 124, 4170. Mahapatra, S., Halfen, J.A., and Tolman, W.B. (1996) J. Am. Chem. Soc., 118, 11575. Itoh, S., Nakao, H., Berreau, L.M., Kondo, T., Komatsu, M., and Fukuzumi, S. (1998) J. Am. Chem. Soc., 120, 2890. Becker, M., Schindler, S., Karlin, K.D., Kaden, T.A., Kaderli, S., Palanch´e, T., and Zuberb¨uhler, A.D. (1999) Inorg. Chem., 38, 1989. Costas, M., Ribas, X., Poater, A., L´opez Valbuena, J.M., Xifra, R., Company, A., Duran, M., Sol`a, M., Llobet, A., Corbella, M., Us´on, M.A., Mah´ıa, J., Solans, X., Shan, X., and Benet-Buchholz, J. (2006) Inorg. Chem., 45, 3569. Fusi, V., Llobet, A., Mah´ıa, J., Micheloni, M., Paoli, P., Ribas, X., and Rossi, P. (2002) Eur. J. Inorg. Chem., 2002, 987.

67. Ghosh, D. and Mukherjee, R. (1998)

Inorg. Chem., 37, 6597. 68. Menif, R., Martell, A.E., Squattrito, P.J.,

69. 70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

and Clearﬁeld, A. (1990) Inorg. Chem., 29, 4723. Menif, R. and Martell, A.E. (1989) J. Chem. Soc., Chem. Commun., 20, 1521. Casella, L., Gullotti, M., Pallanza, G., and Rigoni, L. (1988) J. Am. Chem. Soc., 110, 4221. Sander, O., Henss, A., N¨ather, C., W¨urtele, C., Holthausen, M.C., Schindler, S., and Tuczek, F. (2008) Chem. Eur. J., 14, 9714. Poater, A., Ribas, X., Llobet, A., Cavallo, L., and Sol`a, M. (2008) J. Am. Chem. Soc., 130, 17710. Holland, P.L., Rodgers, K.R., and Tolman, W.B. (1999) Angew. Chem. Int. Ed. Engl., 38, 1139. Obias, H.V., Lin, Y., Murthy, N.N., Pidcock, E., Solomon, E.I., Ralle, M., Blackburn, N.J., Neuhold, Y.-M., Zuberb¨uhler, A.D. and Karlin, K.D. (1998) J. Am. Chem. Soc., 120, 12960. Zhang, C.X., Liang, H.-C., Kim, E.-I., Shearer, J., Helton, M.E., Kim, E., Kaderli, S., Incarvito, C.D., Zuberb¨uhler, A.D., Rheingold, A.L., and Karlin, K.D. (2003) J. Am. Chem. Soc., 125, 634. Shearer, J., Zhang, C.X., Zakharov, L.N., Rheingold, A.L., and Karlin, K.D. (2005) J. Am. Chem. Soc., 127, 5469. Mirica, L.M., Rudd, D.J., Vance, M.A., Solomon, E.I., Hodgson, K.O., Hedman, B., and Stack, T.D.P. (2006) J. Am. Chem. Soc., 128, 2654. Mahadevan, V., DuBois, J.L., Hedman, B., Hodgson, K.O., and Stack, T.D.P. (1999) J. Am. Chem. Soc., 121, 5583. Battaini, G., Granata, A., Monzani, E., Gullotti, M., and Casella, L. (2006) Adv. Inorg. Chem., 58, 185. Santagostini, L., Gullotti, M., Monzani, E., Casella, L., Dillinger, R., and Tuczek, F. (2000) Chem. Eur. J., 6, 519. Palavicini, S., Granata, A., Monzani, E., and Casella, L. (2005) J. Am. Chem. Soc., 127, 18031. Battaini, G., De Carolis, E., Monzani, E., Tuczek, F., and Casella, L. (2003) Chem. Commun., 726.

References 83. Itoh, S., Kumei, H., Taki, M., Nagatomo, 87. Company, A., Palavicini, S.,

S., Kitagawa, T., and Fukuzumi, S. (2001) J. Am. Chem. Soc., 123, 6708. 84. Osako, T., Ohkubo, K., Taki, M., Tachi, Y., Fukuzumi, S., and Itoh, S. (2003) J. Am. Chem. Soc., 125, 11027. 85. Mirica, L.M., Vance, M., Rudd, D.J., Hedman, B., Hodgson, K.O., Solomon, E.I., and Stack, T.D.P. (2002) J. Am. Chem. Soc., 124, 9332. 86. Mirica, L.M., Vance, M., Rudd, D.J., Hedman, B., Hodgson, K.O., Solomon, E.I., and Stack, T.D.P. (2005) Science, 308, 1890.

Garcia-Bosch, I., Mas-Ballest´e, R., Que, J.L., Rybak-Akimova, E.V., Casella, L., Ribas, X., and Costas, M. (2008) Chem. Eur. J., 14, 5335. 88. Herres-Pawlis, S., Verma, P., Haase, R., Kang, P., Lyons, C.T., Wasinger, E.C., Fl¨orke, U., and Stack, T.D.P. (2009) J. Am. Chem. Soc., 131, 1154. 89. Spuhler, P. and Holthausen, M.C. (2003) Angew. Chem. Int. Ed., 42, 5961.

289

291

Part V Structure–Property Relationship and Biosensing

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

293

12 Chirality in Biochemistry: A Computational Approach for Investigating Biomolecule Conformations Adriana Pietropaolo

12.1 Introduction 12.1.1 Molecular Chirality in Living Systems

Since its discovery by Louis Pasteur in 1848, molecular chirality has captured the collective imagination of the chemical community. Pasteur’s realization that tartaric acid could be observed in number of chemically inequivalent forms (R, S, racemic and meso forms) led to a rich new vein of chemical discovery based on the study of what he christened chirality from the greek, cheir, meaning hand. Any molecule that has no improper rotational symmetry is by deﬁnition chiral because it cannot be superimposed on its mirror image. In other words molecules with no improper rotational symmetry can exist in two nonequivalent, enantiomeric forms, named R, S, following the Cahn–Ingold–Prelog priority rule or l, d in the Fisher notation for glyceraldehyde stereocenters (Figure 12.1), being these single notations not correlated each other. In nature many biomolecules do not have improper rotational symmetry and are thus chiral. All the natural amino acids, with the exception of glycine, are chiral and therefore have two enantiomeric forms commonly termed l and d. Amino acids are almost exclusively observed in the l form in biomolecules, while in natural carbohydrates the individual monomers are found in the d form. This selectivity in biomolecules, which is known as homochirality or using the Pasteur notation dissimilarity, has aroused the curiosity of many scientists; many of whom have expended a great deal of effort in attempts to understand why nature is so strict about chirality. Proteins, the building blocks of the cell, are typically composed of a chain of homochiral l-amino acids. Only in the cell wall of some microorganisms [1] and various aged human tissues (tooth, bone, brain, etc.) [2] d-amino acids are observed. Nanda et al. [3] recently investigated the role of homochirality in the folding free energy landscape of proteins. They showed that syndiotactic chains, in which there is an alternation in the chirality of adjacent monomers, have a large ensemble of accessible conformations (Figure 12.2). Hence, syndiotactic chains have a much Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

294

12 A Computational Approach for Investigating Biomolecule Conformations

Energy

Figure 12.1 Graphical representation of enantiomers, due to stereocenter chirality generated by the reﬂection through a vertical plane.

Achiral Homochiral Heterochiral

Unfolded Homo + heterochiral

Conformational states Figure 12.2 Energy landscape representation as described in [3]. Width of each cone represents the number of accessible conformational states, while their depth underlines the energy linked to structural stability. Unfolded conformations correspond to the highest energy level, here represented by the cone bases, whose perspective is given by the square.

greater degree of ﬂexibility in their backbone, than isotactic polypeptides, in which the chiralities of all stereocenters are identical. The fact that glycine residues, the only amino acid that lacks the Cα stereocenter, can provide ﬂexibility to proteins and can destabilize folded states [4, 5] strengthens the suggestion that chirality plays an important role in protein ﬂexibility. This dependence of protein folding on chirality occurs because the chirality of individual amino acids can affect local secondary structures which in turn can affect the tertiary structure of the protein. For example, exchanging a number of l-amino acids for d-amino acids in Sso7d ribonuclease changes the conformation of the active site and prevents the protein function [6]. Chirality has therefore a deep role in complex systems like

12.1 Introduction

proteins in promoting the formation of inter-residue contacts that stabilize folded conﬁgurations. 12.1.2 Protein Secondary Structures

Protein structures are typically described using a system that separates the various aspects of their structures. The so-called primary structure describes the amino acid sequence of which the protein is composed, while tertiary structure refers to the global structure adopted by a single protein. This tertiary structure can be thought of in terms of a number of highly, regular substructural motifs, or secondary structure elements, the structures of which depend on the underlying primary structure. Proteins can adopt a number of different secondary structures (Figure 12.3), the most common being the right-handed helices (which include the α, the 310 and the π helices), the β sheet, the β turn and the left-handed helices (an example of which is the poly-l-proline II (PPII)). The following are the characteristics of each of these structures:

Figure 12.3 The most common protein secondary structures: α helix is shown by tubes, β sheets by thick ribbons, turn by narrow ribbons. H-bonds are shown by dotted lines.

295

296

12 A Computational Approach for Investigating Biomolecule Conformations

• α helix: This structure, which was discovered by Linus Pauling and which typically involves more than four residues, has a pitch of 5.4 A˚ and is stabilized by hydrogen bonds between the carbonyl oxygen of residue i and the amide hydrogen of residue i + 4. • β sheets: This structure is ﬂat and the hydrogen bonds it contains are either parallel or antiparallel depending on the orientation of the chain. Often an isolated β sheet is referred to as a bulge. • β turns: In these structures hydrogen bonds are formed between the ith and i + 3th residues. Right-handed 310 helices are β turns in which more than three residues are involved. ˚ • Poly-L-proline II: It has an elongated structure that involves a pitch of 9.3 A. • Coil: A Coil is any segment of an amino acid chain that has no secondary structure. Any given protein segment is said to have no secondary structure if every pair of (φ, ψ) angles is independent of all the other pairs of (φ, ψ) angles within the segment [7]. The secondary structure of a protein is very important as it is presumed that the global folding event is driven by local folding. Furthermore, as already mentioned, the tertiary structure, which controls the protein function, can be considered to be a collection of secondary elements. The incorrectly folded (misfolded) secondary structural elements can give rise to a different tertiary structure or even a protein with altered properties and can prevent the protein function. What is more, introduction of induced ﬁt concepts surpassed rigid lock and key ideas in the theories of protein docking. This means that, now more than ever, it is necessary to understand the dynamical structures of proteins as well their static structures, since there is a direct link between protein conformational change and protein function. 12.1.3 Protein Secondary Structure Assignment

A number of methods for secondary structure assignment have been proposed (see, e.g., [8–11] and there is about 80% agreement between structural assignments made with different methods. The ﬁrst key contribution was probably by Ramachandran et al. [12], who correlated the native distribution of the –N–Cα – and –Cα –C– protein dihedral angles (φ, ψ) with the secondary structure [13]. However, this approach fails when there is a great deal of conformational ﬂexibility, which is common for peptides, because the ﬂexibility gives rise to nonstandard backbone angles [14–16]. One of the most commonly used programs for secondary structure determination is the ‘‘dictionary of protein secondary structures’’ (DSSP) [17]. The DSSP is reliant on an algorithm that recognizes hydrogen bond patterns involving the C=O and N–H backbone atoms and neglects the φ and ψ dihedral angles. Structural classiﬁcation is thus based on eight qualitatively, different classes, and small deviations in the backbone dihedral angles from the ideality, which are often important when it comes to proteins’ biological function, can be neglected.

12.2 Computational Techniques for Studying Protein Dynamics

Moreover, DSSP analysis is known to be error-prone when it comes to the exact detection of the edges of a given motif [18]. There are various other methods that improve on the original DSSP method. The ﬁrst of these is the so-called STRIDE algorithm [19], in which a consideration of the contributions from the dihedral angles is included in addition to the descriptions of the hydrogen bond patterns in the identiﬁers for various classes. A recently reported algorithm for classifying turn structures [20] works by screening and clustering a large data set taken from a protein data bank. From this analysis the three normal, four open, and ﬁve reverse turn families emerged along with a number of new turn types. Despite all the algorithmic sophistication, the assignment of secondary structure to conformations that, like PPII structures [21, 22], depart strongly from any ideal backbone structure still remains a challenge. What makes the structural assignment of protein structure in this ‘‘twilight zone’’ [23] so difﬁcult is the high degree of ﬂexibility, which makes the coil state become a kind of catch-all structural assignment unless high resolution data are obtained. 12.1.4 Intrinsic Chirality and Protein Secondary Structures

In Section 12.1, only chirality related to stereocenters has been discussed. Such stereocenters have a chirality due to the asymmetry arising from a noncoplanar arrangement of atoms in which one atom is connected to ‘‘at least’’ three further nonequivalent atoms. The most ubiquitous example of this is a carbon bonded to four atoms that lie on the vertices of a tetrahedron centered on the carbon (Figure 12.1). This, however, is not the only form of chirality that can exist as there are countless other three-dimensional structures that have no stereocenter and are also chiral. An example of an intrinsically chiral structure that does not necessarily need to have any stereocenters is a helicoidal structure. Proteins tend to adopt more chiral (α helix, 310 helix, β turn, PPII) than achiral (β sheets) local, secondary structures.

12.2 Computational Techniques for Studying Protein Dynamics

Several computational techniques are available to study the motions of biological macromolecules, some commonly used examples of which are molecular dynamics (MD), Monte Carlo, and normal mode analysis (NMA) approaches. All these methods generally use the equations of classical mechanics together with a forceﬁeld that is used to calculate the potential energies or forces. These forceﬁelds are based on either terms that describe the interactions between atoms (atomistic models) or terms that describe the interactions between groups of atoms (mesoscale models). If the goal is to analysze the ﬁne details of local arrangements, atomistic scale simulations are preferred. On the other hand, if the purpose is to obtain an

297

298

12 A Computational Approach for Investigating Biomolecule Conformations

understanding of the spatial arrangements of the macromolecule, coarse-grained (mesoscale) approaches are preferred as the longer timescales accessible make it possible to examine huge rearrangements of the system. The simplest approach to study protein folding within a coarse-grained model is provided by the Go− potential [24]. The basic idea of which is to devise a forceﬁeld consistent with the experimentally established structure of the native state, in which only the interactions between Cα –Cα are taken into account. These pairs are within some small cutoff distance (usually 7.5 A˚ [25]) in the native state structure. As only Cα atoms are included, the model provides only a very coarse determination of the protein motions. Another commonly used approach to describe large motions in proteins is NMA, which is based on the harmonic approximation of the potential energy. Here the dynamics of the molecule is then described in terms of a collection of independent harmonic oscillators on a basis of normal modes eigenvectors (an orthonormal set of directional vectors that represent the uncoupled motions of the system). Low-frequency oscillations represent large-amplitude, collective motions that often correlate well with the experimentally observed conformational changes associated with protein function. To perform an NMA, ﬁrst, the potential energy of the system must be minimized, which is calculated using either an atomistic or coarse-grained approach, and then a diagonalization of the matrix of second derivatives of the potential energy [26] (the Hessian) should be performed. The method that can, in theory, provide the greatest amount of detail about protein motion is MD, which is made possible by the fact that forces are expressed as the gradient of a potential energy function. MD is a computational technique in which the time evolution of a set of interacting particles (generally atoms or molecules) is followed step by step by integrating their equations of motion. In contrast to Monte Carlo simulations, which employ stochastic dynamics to explore phase space, MD employs deterministic dynamics, which means that given an initial set of positions and velocities, the subsequent time evolution is completely determined and in principle reversible. The major disadvantage of this technique is that it is currently only possible to run at most a few microseconds of simulation, while most processes of interest in nature take place on the order of seconds. However, more advanced variants of the MD techniques have been developed [27–30] – in which longer timescales can be investigated – and used to obtain intriguing results [31].

12.3 Employing Chirality to Analyze Protein Motions 12.3.1 The Chirality Index

So far, we have discussed the main concepts used in the study of protein structure. However, studying the structural data that can be gleaned from molecular

12.3 Employing Chirality to Analyze Protein Motions

299

simulation is usually far from simple, because proteins are not static entities and instead exist as dynamical entities. Typically, the dynamical motions of proteins involve local rearrangements around the average structure; namely, that which can be probed experimentally using techniques such as NMR. A natural way to describe such local ﬂuctuations in protein structures would involve an examination of changes in the local secondary structure of the protein. However, a description of these changes is beyond the scope of the standard tools such as DSSP [32] or STRIDE [33] that have been described. The exception to this is local chirality which is highly sensitive to local conformational motions and can thus provide insight into local conformational changes and the ﬂexibility of particular protein segments [34]. The chirality method works as it is possible to calculate individual indices for fragments of the protein backbone. For each of these fragments, a chirality index can be computed using a method that was proposed by Marti et al. and Chou [35, 36] for low mass molecules and which succeeded in analyzing facial diastereoselectivity [37]. Here, however, instead of calculating the chirality index for the entire molecule, it is calculated for each individual fragment of the molecule separately (as shown in Figure 12.4). The value of the index for each protein fragment is expressed as follows: Ga,Na =

all p 4 gijkl 3N4a p

(12.1)

i,j,k,l

gijkl =

    

[(rij ×rkl )·ril ](rij ·rij )(rjk ·rkl )

0

if rij , rkl , ril , rjk ≤ re , a ≤ i, j, k, l ≤ Na + a − 1 otherwise

(12.2)

where a is the ﬁrst (closest to the N terminal) atom of a given sequence of Na = 15 consecutive backbone atoms with coordinates r and rc = 12 A˚ is a cutoff radius, added to avoid the computation of negligible long-range terms. The efﬁcacy of this method is that the instantaneous value of the chirality indices can be calculated only from a knowledge of the atomic coordinates. What NR

Ca N a

N C

Ca

C Ca

N

Ca

C Na

Figure 12.4 A backbone composed of 15 atoms. The G index is calculated for every permutation of four atoms inside this 15 atoms fragment. Then, second G index is calculated ahead of one atom, here underlined by a line starting from the Cα .

Ca

C

N

N

Na+a−1 C

N

C Ca

12 A Computational Approach for Investigating Biomolecule Conformations Table 12.1 Average G values and relative standard deviations of G for ideal secondary structures, involving at least NR residues. Each structure was built by sampling φ and ψ angles from a Gaussian distribution, centered at the ideal φ and ψ values with σ = 15◦ (see [30]).

Structure

G

σG

NR

α helix 310 helix β turn I β sheets PPII π helix

−0.04 −0.07 −0.07 +0.00 +0.10 −0.01

0.02 0.01 0.01 0.01 0.03 0.02

>3 >3 2,3 ≥2 >3 >3

0.10 PPII Sheets a helix Type I b turn 310 helix p helix

0.05

Ga

300

0.00

−0.05

−0.10 3

4

5

6

7

8

9

Residue number

Figure 12.5 Chirality index G along the backbone for different ideal secondary structures. The cutoff used is rc = 12 A˚ and Na = 15 atoms is considered.

is more, because it is possible to calculate the chirality index in idealized secondary structures (here reported in Table 12.1, Figure 12.5), we can determine which secondary structure a local region of the protein has and how much the structure deviates from that structure during its dynamical motions. 12.3.2 Using Chirality to Understand Protein Structure

Any function that purports to be able to determine secondary structure in proteins must provide a range of values for each of the possible secondary structures. To

12.3 Employing Chirality to Analyze Protein Motions

assign these ranges, ideal secondary structures were analyzed. To do this, the periodicity in the backbone angles (φ, ψ) was used to generate ideal segments of protein structures for which chirality indices were computed. The periodicities employed were (−67◦ , −41◦ ) for α helix [34], (−49◦ , −26◦ ) for 310 helix [35], (−67◦ , −59◦ ) for π helix [34], (−60◦ , −30◦ , −90◦ , 0◦ ) for type I β turns [36], (−75◦ , 147◦ ) for PPII helix [37], and (−130◦ , 130◦ ) for sheet regions [38]. In all cases, the ω angles were kept ﬁxed to the trans value of 180◦ . The type I β turn conformation is not periodic in proteins, as it generally involves only four consecutive residues; however, for the sake of simplicity, it was considered periodic in [30] so that comparison with the other structural motifs is possible. Figure 12.5 shows the behavior of the G index for the ideal structures along the ˚ The opposite backbone. Here, the index was calculated using Na = 15 and rc = 12 A. chirality indices of the right-handed α helix (negative) and the left-handed PPII helix (positive) is worth noting together with the observation that β sheets have no chirality because of their ﬂat shape. Figure 12.6 shows the chirality index calculated along the backbone of a typical protein taken from the protein data bank of structures. It is possible to note from this ﬁgure that in most of this protein the chirality index is in one of the ranges corresponding to the idealized structures. In addition to its ability to recognize secondary structure types, the chirality index is very sensitive to poly-l-proline dihedrals. A positive peak in the G value indicates that at least one amino acid in a given protein region has the PPII structural motif. Figure 12.7 shows the G value of a model poly-l-proline peptide fragment (1JMQ, residues 51–60), which

Hemoglobin G0s a 310 Turn Bend Coil

0.05 0.00

−0.05

DSSP

Ga

0.10

6 5 4 3 2 1 0

−0.10 10

30

50

70 90 110 Residue number

Figure 12.6 Chirality index G along the backbone for hemoglobin protein, from X-ray structure. Typical secondary structures, with negative periodicity concerning the α helices and with typical G values for the

130

other secondary structures (cf. Figure 12.5), are easily identiﬁed. The DSSP assignment is also plotted as the numeric code: 310 = 0, turn = 1, bend = 2, bridge = 2.5, α = 3, sheets = 4, and coil = 5.

301

12 A Computational Approach for Investigating Biomolecule Conformations

6 5 4 3 2 1 0

0.12 0.10 0.08 0.06 0.04

G G ideal ppII DSSP–coil

0.02 0.00

DSSP

0.14

Ga

302

3

4

5 6 Residue number

7

Figure 12.7 Chirality index G along the backbone for poly-L-proline II model peptide. The DSSP assignment is also plotted as the numeric code: 310 = 0, turn = 1, bend = 2, bridge = 2.5, α = 3, sheets = 4, and coil = 5.

is known to have poly-l-proline dihedral angles. A good overlap between the PPII ideal structure and the PPII model peptide exists in the residues labeled 3–5 in the ﬁgure; however, after residue 5, the G value of 1JMQ drops as it must take residues 7 and 8 into account that are not in the PPII conformation. Detection of the PPII structure using DSSP-like algorithms is hampered because prolines do not form hydrogen bonds. Therefore, despite the fact that this structure is observed in other polypeptides, its extended conformation (9.3 A˚ pitch) is not conducive to the assignment of a well-deﬁned hydrogen bond pattern, which means they are commonly misclassiﬁed as loops or coils [22]. G’s sensitivity to these PPII structures makes the G function an important tool for the better identiﬁcation of this particular structure class. 12.3.3 Chirality Index as a Tool for Monitoring Protein Dynamics

Because the chirality index can be calculated using the atomic coordinates, it can prove useful in examining protein dynamics, for example, the dynamics of a transition from an α helix to a 310 helix or from a coil to a helixes (Figure 12.8a and b, respectively). It is particularly useful for the examination of folding transitions, as it contains information on both the φ and ψ dihedral angles. Moreover, G index is stable during dynamics and in response to random thermal ﬂuctuations [30], which proves that it can be used to follow dynamical processes. Furthermore, the average value of the chirality index can be used to detect protein motions, and

12.3 Employing Chirality to Analyze Protein Motions

Ga

0.01 −0.01 −0.03

DSSP

Residue 5 G7 a 310 Turn Bend

6 5 4 3 2 1 0

−0.05 −0.07 −0.09 −0.11

0

20

40

60

80

100

120

Time (ns)

0.10 Residue 6

Ga

0.05

G10 DSSP a Turn 310 Bend Coil Sheets b bridge

0.00 −0.05 −0.10 −0.15

(b)

0

10

20

30 40 Time (ns)

50

6 5 4 3 2 1 0 DSSP

(a)

60

Figure 12.8 (a) α to 310 helix transition, followed by chirality. (b) Coil to helix transition, underlined from the positive-to-negative switch of Ga .

therefore useful for studying protein ﬂexibility. This is an important issue in the study of intrinsically disordered proteins. Figure 12.9 reports the averages and standard deviations of the chirality indices calculated along the backbone of rigid (a) and ﬂexible structures (b), respectively, during the course of an MD trajectory. As previously discussed, the average chirality provides information on the average local structure adopted, which, together with the standard deviations of G, provides information on the ﬂexibility of local regions in the protein. The typical method used to measure this quantity is the root mean

303

−0.10

−0.05

0.00

3

4

5

6

7

8 9 10 11 12 13 14 15 Residue number

Hemoglobin 1–18 110 ns Hemoglobin 1–18 PDB

(b)

−0.10

−0.05

0.00

0.05

0.10

5

10

15 20 Residue number

25

1 REI 1–30 50 ns 1 REI 1–30 PDB

30

Figure 12.9 Standard deviations of G among (a) hemoglobin 1–18 and (b) 1REI 1–30 conﬁgurations. It is worth to note the persistence of the chirality index for a rigid peptide, as the ﬁrst helix of hemoglobin, and the high values of the standard deviations for a ﬂexible peptide, as the trajectories of the ﬁrst 30 residues of 1REI immunoglobulin antigen. As comparison the G from PDB and from the trajectories is shown.

(a)

Ga

0.05

Ga

304

12 A Computational Approach for Investigating Biomolecule Conformations

12.3 Employing Chirality to Analyze Protein Motions

square deviations (RMSDs), which, in fact, measures only the difference in the structure from a given reference conﬁguration and is not particularly sensitive to the changes in dihedral angles that local chirality measures. In particular high RMSD indicates that the structure highly differs from the reference structure. However, the reference structure may not be the true average structure and, even if it is, the RMSD may not represent signiﬁcant motions particularly if the single conﬁgurations examined are far from the average structure. The standard deviation of the chirality index makes sense, however, as it is a deviation about an average in a quantity that measures secondary structures. This average quantity is zero if no secondary structure is present, whereas if a secondary structure is present then a constant value for the average chirality index is returned and this depends on the average secondary structure. Meanwhile, the standard deviation measures directly the ﬂexibility of the secondary structure in a local region of the protein. 12.3.4 Chirality and Circular Dichroism

Circular dichroism (CD) experiments are connected in depth with chirality because the l and d enantiomers absorb polarized light differently. These experiments thus provide a useful probe for chiral molecules but not for the achiral ones. The CD spectra of proteins are strongly related to the backbone dihedral angles φ and ψ, and can be used to obtain insight into the average structure the proteins adopt in solution. What is more, combining information derived from CD and MD may help in unraveling the conformational ensemble adopted in solution. As an example, we consider the CD spectra of the avian prion hexarepeat region as a function of pH [16]. Prion proteins (PrPC ) are glycoproteins that in mammals, but not in avians, can cause prion diseases, which involve the incorrect folding (misfolding) of proteins. It is commonly believed that this misfolding involves a conversion from a structure rich in α helices to a βsheet–enriched pathogenic isoform (PrPSc ). Until now, NMR structure determination of the prion N-terminal portion has been hampered because of its ﬂexibility. Mammal and avian proteins show different N-terminal tandem repeats, PHGGGWGQ and PHNPGY, respectively. Both these contain histidine, however; only the avian proteins involve tyrosine in their primary sequence. Both these residues are of particular interest because they are highly sensitive to pH variations, having average pKa values of 6.1 and 9.9, respectively [16]. The far UV–CD spectra of the avian tetra-hexarepeat (PHNPGY)4 , reported in Figure 12.10, show a signal varying with pH. At pH 4, a pH at which one would expect all histidyl residues to be protonated (average pKa 6.1), the spectrum is broad and has a minimum at 203 nm, a weak shoulder around 216 nm, and a maximum at 230 nm. In general, this shape indicates an equilibrium between different conformations, suggesting the presence of other secondary structure elements besides the random coil. This spectrum shape is similar to that found for shorter peptide fragments as reported in [15, 39]. For these smaller fragments, the presence of both random coil and β-turn structures was suggested for those primary

305

G

−0.09 −0.12

−0.12

0.00

0.03

0.06

0.09

0.12

3

250

−0.06

18

240

−0.09 21

220 230 λ (nm)

LH84+

9 12 15 Residue number

210

−0.06

6

200

pH 4 pH 7 pH 8 pH 10

−0.03

3

−50 190

−40

−30

−20

−10

−0.03

0.00

0.03

0.06

0.09

0.12

∆e (M−1 cm−1)

0

G

10

6

9 12 15 Residue number

18

LH4

21

LH41

G

−0.12

−0.09

−0.06

−0.03

0.00

0.03

0.06

0.09

3

6

4H+

−4H+

0.12

4H+

−4H+

L4−

LH84+

9 12 15 Residue number

LH42

18

L4−

21

306

12 A Computational Approach for Investigating Biomolecule Conformations

12.3 Employing Chirality to Analyze Protein Motions

sequences that encompass a PXXP motif, a sequence that normally favors β-turn and/or PPII structures [40]. When the pH is increased the spectrum of the avian tetra-hexarepeat (PHNPGY)4 changes in a different way as shown in Figure 12.10. Up to pH 7, there is no change to the band centered at 230 nm, but there is a shift in the position of the minimum, which takes it toward 200 nm, and a decrease in the bands’ intensity. Further increases in the pH cause a general broadening of the spectra and a signiﬁcant decrease of the signal at 230 nm. Finally, at pH 10, a new maximum at 250 nm is observed. It is well known that aromatic side chains give rise to a contribution to the far UV–CD spectra of peptides and that in phenols this far UV signal is redshifted in the deprotonated phenolate ion [41, 42]. Hence, the new maximum, observed at 250 nm, is attributable to the deprotonation of tyrosine residues. Furthermore, the decrease and subsequent disappearance of the positive maximum in the signal at 230 nm is also related to the protonation state of the tyrosine residue as it is caused by the phenolic group on it. Alongside the evidence the UV–CD spectrum provides about the protonation state of the tyrosine residue, changes in the peptide secondary structure, which occur as a function of pH, can also be detected. For instance, the strong positive band at 190 nm (Figure 12.10) and the shoulder found at 216 nm are features typical of β-turn-like conformations [43–45], which appear to be predominant at neutral and basic pH. The weakening of these signals at lower pH suggests that the equilibrium conﬁguration is shifted, thus it would seem that the conformation adopted is strongly dependent on the protonation state of the histidine and tyrosine residues. A chirality analysis of the conformational ensemble obtained from MD simulations reveals that there is an enhancement in the number of turn conformations as pH increases (this is indicated by the negative sharp peaks shown in the bottom side of Figure 12.10). In particular, at acidic pH (LH4+ 8 ), the broad negative peak centered at residue 7 (Pro), which includes residues 6–8 (Tyr, Pro, and His, respectively), conﬁrms the presence of a 310 helix structure [30], while the negative sharp peak centered at residue 11 (Gly) suggests the presence of a turn that includes residues 9–12 (Asn, Pro, Gly, and Tyr, respectively). At neutral pH (LH4 ), chirality analysis indicates the presence of a turn region that includes residues 4–6 (Pro, Gly, and Tyr, respectively). This turn is highly ﬂexible, however, as indicated by the wide standard deviation and the average value for this negative peak is close to the upper bound of values observed for this particular structural motif [–0.1 : −0.06]. Two further turn regions that are signaled by much stronger negative peaks are observed at neutral pH and are centered at residues 11 and 17 (Gly). At basic

Figure 12.10 Up on the left, CD spectra of avian tetra-hexarepeat fragment as a function of pH. At basic pH, the shoulder approximately at 216 nm reveals the formation of type I β turn. Up, on the right, typical conformations of avian tetra-hexarepeat fragment at acidic (LH8 4+ ), neutral (LH4 ), and basic (L4− ) pH. Bottom, chirality index along the backbone of the avian tetra-hexarepeat trajectories in the three protonation states. The shift toward type I β turn is underlined from the negative sharp peaks, adopted mainly in the basic state, L4− , as shown in the CD spectra. A circle underlines the side chains hydrogen bonds. N and C termini are shown, respectively, from left to right.

307

308

12 A Computational Approach for Investigating Biomolecule Conformations

pH (L4− ), those negative peaks that were observed at neutral pH show smaller values for their standard deviations. This suggests that the stability of turn regions, inside the peptide, is enhanced by the increasing pH. Moreover, by examining the chirality pattern along the backbone, we can observe that on changing the pH the chirality index varies most strongly for residues 4–8 (Pro, Gly, Tyr, Pro, and His, respectively) and for the C-terminal region. These are the regions in which His8 and Tyr24 are involved and this indicates the pivotal role that these residues play in any conformational change. The turn structures observed in the MD simulations are of course stabilized by the large number of proline residues. However, in this system it would appear that histidine, particularly histidine 8, also plays a key role in stabilizing turn regions. This is evidenced by the fact that at physiological pH, when this residue is expected to be deprotonated, turn regions become more stable. The above example has demonstrated how a combination of experimental CD and chirality analysis of simulation data can be used to explain how the conformational states accessible to this molecular system depend on pH. Furthermore, Figure 12.10 shows that at physiological pH the chirality pattern of the tetra-hexarepeat region reﬂects the periodicity in the primary structure. This ﬁnding reinforces the suggestion that local folding events mainly drive the global protein folding toward the tertiary structure.

12.4 Perspectives

The connections between chirality and biosystems, particularly proteins, have been discussed here at length. It has been shown that protein structures are particularly amenable to descriptions based on chirality concept since proteins have intrinsically asymmetric structures. In particular, a chirality index has been introduced that can be used to analyze the extent to which proteins are folded and the local ﬂexibility of the protein. This index has well-deﬁned values for the typical secondary structure elements and is particularly effective in detecting PPII motifs, often misclassiﬁed as coils when other techniques are employed. One word of caution regarding coil states must be mentioned here: Coils are often considered to be unstructured regions in the protein and thus on the face of it would be expected to be achiral. However, because proteins are isotactic, regions of proteins that adopt coiled states should present some degree of left-handed chirality and thus should not have a chirality index of zero [30]. A particularly interesting class of protein structures is the so-called coiled coils, examples of which include keratine and the muscle protein tropomyosin. These structures consist of α helices wound together in a manner that is stabilized by hydrophobic interactions. What makes them intriguing from a chirality point of view is that while the α helices themselves are right handed the structure is often overall left handed. Recently, a correlation between the chirality of the coiled coils and their mechanical properties has been found [46]. This correlation links the

12.4 Perspectives

plasticity of coiled coil structures with the partner intrinsic chirality in the selection of biochemical partners. A particularly intriguing question related to these discussions is the chirality of amyloid suprastructures. Amyloids, like A β 1−40 and hen lysozyme, are ﬁbrillar aggregates of proteins with a characteristics cross-β conformation that until now were generally accepted to have a structure composed of left-handed helices. Recently, however, a surprising discovery was made concerning the peptide of serum amyloid A protein (SAA1−12 ). It was found that structures composed of amino acids with S stereocenters form right-handed helices, while those composed of R stereocenters form the left-handed suprastructures [47]. Hence, it is clear that supramolecular chirality depends on the structure organization and it need not be necessarily the same for all ﬁbrils. Fascinating proteins with high symmetries are found in viruses. A virus consists of a fragment of deoxyribonucleic acid (DNA) or ribonucleic acid (RNA) enclosed in a highly regular arrangement of capsomer proteins. In spherical viruses, this external, capsid shell has its individual capsomer molecules located on the vertex positions of an equilateral triangulation of a sphere [48] with an icosahedral symmetry [49, 50]. Viruses can also adopt nonspherical geometries in which there is a skew in the capsid, which can cause it to become chiral. These chiral-shaped capsids are found to require a higher rupture force than the perfect icosahedral capsids [51]. Proteins play a central role in the life sciences as both the molecular workhorses and building blocks of all living things. Simulation and experiment are beginning to provide genuine insight into how these complicated molecules function in isolation, but how they interact or organize themselves to form a cell is an area where a lot of work is still required. Recently, however, in a discovery that surprised the community, cells were found to have a chiral organizing principle, which works in terms of asymmetrical compartmentalization and probably originates in the centrosome [52]. Xu et al. [53] argue that intrinsic functional chirality is a property of eukaryotic cells that probably confers on them a selective advantage during the course of evolution. This ﬁnding suggests that chirality is a central organizing principle in life that confers order at every length scale up to the full cell. This organization on the basis of chirality must have emerged at some point in the past as an evolutionary response that protected the growth of life.

Acknowledgments

A number of people have contributed for the realization of the chirality index methodology; Dr. Luca Muccioli, Dr. Roberto Berardi, and Prof. Claudio Zannoni in thedevelopment of chirality method applied to protein structures. Dr. Diego La Mendola and Prof. Enrico Rizzarelli have made possible the combined experimental–computational approaches for studying the pH dependence

309

310

12 A Computational Approach for Investigating Biomolecule Conformations

of the N-terminal region of avian prion protein, by using circular dichroism and chirality from MD simulations. I gratefully acknowledge Prof. Claudio Zannoni, Prof. Enrico Rizzarelli, and Prof. Michele Parrinello for valuable suggestions and discussions. A special acknowledgement goes to Dr. Gareth Tribello, Dr. Meher Prakash Ayalasomayajula, and Dr. Francesco Mauriello for several useful suggestions in writing this manuscript.

References 1. Corrigan, J. (1969) D-amino acids in 2.

3.

4.

5.

6.

7.

8.

9.

10.

animals. Science, 164, 142–149. Fujii, N. (2002) D-amino acids in living higher organisms. Orig. Life Evol. Biosph., 32, 103–127. Nanda, V., Andrianarijaona, A., and Narayanan, C. (2007) The role of protein homochirality in shaping the energy landscape of folding. Protein Sci., 16, 1667–1675. O’Neil, K.T. and Grado, W.F.D. (1990) Thermodynamic scale for the helix-forming tendencies of the commonly occurring amino-acids. Science, 249, 774–778. Maity, H., Lim, W.K., Rumbley, J.N., and Englader, S.W. (2003) Protein hydrogen exchange mechanism: local ﬂuctuations. Protein Sci., 12, 153–160. Ladik, J.J. and Szekeres, Z.Z. (2006) A study on chirality in biomolecules: the effect of the exchange of L amino acids to D ones in Sso7d ribonuclease. J. Mol. Model., 12, 462–467. Smith, L.J., Fiebig, K.M., Schwalbe, H., and Dobson, C.M. (1996) The concept of a random coil – residual structure in peptides and denatured proteins. Fold Des., 1, R95–R106. Dupuis, F., Sadoc, J.F., and Mornon, J.P. (2004) Protein secondary structure assignment through Voronoi tesselation. Proteins, 55, 519–528. Richards, F.M. and Kundrot, C.M. (1988) Identiﬁcation of structural motifs from protein coordinate data: secondary structure and ﬁrst level supersecondary structure. Proteins, 3, 71–84. Majumdar, I., Krishna, S.S., and Grishin, N.V. (2005) PALSSE: a program to delineate linear secondary

11.

12. 13.

14.

15.

16.

17.

18.

structural elements from protein structures. BMC Bioinformatics, 6, 202. Ramachandran, G.N., Ramakrishnan, C., and Sasisekharan, V. (1963) Stereochemistry of polypeptide chain conﬁgurations. J. Mol. Biol., 7, 95–99. Schlick, T. (2002) Molecular Modeling and Simulation, Springer, New York. Zagrovic, B., Lipfert, J., Sorin, E.J., Millett, I.S., van Gunsteren, W., Doniach, S., and Pande, V.S. (2005) Unusual compactness of a polyproline type II structure. Proc. Natl. Acad. Sci. U.S.A., 102, 11698–11703. Pietropaolo, A., Raiola, L., Muccioli, L., Tiberio, G., Zannoni, C., Fattorusso, R., Isernia, C., La Mendola, D., Pappalardo, G., and Rizzarelli, E. (2007) An NMR and molecular dynamics investigation of the avian prion hexarepeat conformational features in solution. Chem. Phys. Lett., 442, 110–118. Pietropaolo, A., Muccioli, L., Zannoni, C., La Mendola, D., Maccarrone, G., Pappalardo, G., and Rizzarelli, E. (2008) Unveiling the role of histidine and tyrosine residues on the avian prion hexarepeat domain. J. Phys. Chem. B, 112, 5182–5188. Kabsch, W. and Sander, C. (1983) Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features. Biopolymers, 22, 2577–2637. Mizuguchi, K. and Go, N. (1995) Comparison of spatial arrangements of secondary structural elements in proteins. Protein Eng., 8, 353–362. Frishman, D. and Argos, P. (1995) Knowledge-based protein secondary structure assignment. Proteins, 23, 566–579.

References 19. Meissner, M., Koch, O., Klebe, G.,

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

and Schneider, G. (2008) Prediction of turn types in protein structure by machine-learning classiﬁers. Proteins, 74, 344–352. Stapley, B.J. and Creamer, T.R. (1999) A survey of left-handed polyproline II helices. Protein Sci., 8, 587–595. Cubellis, M.V., Caillez, F., Blundell, T.L., and Lovell, S.C. (2005) Properties of polyproline II, a secondary structure element implicated in protein-protein interactions. Proteins, 58, 880–892. Martin, J., Letellier, G., Marin, A., Taly, J.F., de Brevern, A.G., and Gibrat, J.F. (2005) Protein secondary structure assignment revisited: a detailed analysis of different assignment methods. BMC Struct. Biol., 5, 17. Hayward, S. and Go, N. (1995) Collective variable description of native protein dynamics. Annu. Rev. Biophys. Biomol. Struct., 46, 223–250. Clementi, C., Vendruscolo, M., Maritan, A., and Domany, E. (1999) Folding Lennard-Jones proteins by a contact potential. Proteins Struct. Funct. Gen., 290, 253–266. Tama, F. and Brooks, C.L.III (2006) Symmetry, form and shape: guiding principles for robustness in macromolecular machines. Annu. Rev. Biophys. Biomol. Struct., 35, 115–133. Laio, A. and Parrinello, M. (2002) Escaping free-energy minima. Proc. Natl. Acad. Sci. U.S.A., 99, 12562–12566. Barducci, A., Bussi, G., and Parrinello, M. (2008) Well-tempered metadynamics: a smoothly converging and tunable free-energy method. Phys. Rev. Lett., 100, 020603-1–4. Branduardi, D., Gervasio, F.L., and Parrinello, M. (2007) From A to B in free energy space. J. Chem. Phys., 126, 10, Article No. 015403. Carter, E.A. (2008) Challenges in modeling materials properties without experimental input. Science, 321, 800–803. Laio, A. and Gervasio, F.L. (2008) Metadynamics: a method to simulate rare events and reconstruct the free energy in biophysics, chemistry and material science. Rep. Progr. Phys., 71, 126601.

31. Pietropaolo, A., Muccioli, L., Berardi,

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

R., and Zannoni, C. (2007) A chirality index for investigating secondary structure and their time evolution.Proteins, 70, 667–677. Solymosi, M., Low, R.J., Grayson, M., and Neal, M.P. (2002) A generalized scaling of a chiral index for molecules. J. Chem. Phys., 116, 9875–9881. Osipov, M.A., Pickup, B.T., and Dunmur, D.A. (1995) A new twist to molecular chirality: intrinsic chirality indices. Mol. Phys., 84, 1193–1206. Berardi, R., Cainelli, G., Galletti, P., Giacomini, D., Gualandi, A., Muccioli, L., and Zannoni, C. (2005) Can the pi-facial selectivity of solvation be predicted by atomistic simulation? J. Am. Chem. Soc., 127, 10699–10706. Pauling, L., Corey, R.B., and Branson, H. (1951) The structure of proteins; two hydrogen-bonded helical conﬁgurations of the polypeptide chain. Proc. Natl. Acad. Sci. U.S.A., 37, 205–211. Marti, D.N., Schaller, J., and Llin´as, M. (1999) Solution structure and dynamics of the plasminogen kringle 2-AMCHA complex: 3(1)-helix in homologous domains. Biochemistry, 38, 15741–15755. Chou, K.C. (2000) Prediction of tight turns and their types in proteins. Anal. Biochem., 286, 1–16. Adzhubei, A.A. and Stemnberg, M.J.E. (1993) Left-handed polyproline-II helices commonly occur in globularproteins. J. Mol. Biol., 229, 472–493. Branden, C. and Tooze, J. (1999) Introduction to Protein Structure, Garland Publishing, New York. La Mendola, D., Bonomo, R.P., Impellizzeri, G., Maccarrone, G., Rizzarelli, E., Pietropaolo, A., and Zito, V. (2005) Copper(II) complexes with chicken prion repeats: inﬂuence of proline and tyrosine residues on the coordination features. J. Biol. Inorg. Chem., 10, 463–475. Feller, S., Ren, R., Hanafusa, H., and Baltimore, D. (1994) SH2 and SH3 domains as molecular adhesives: the interactions of Crk and Abl. Trends Biochem. Sci., 19, 453–458.

311

312

12 A Computational Approach for Investigating Biomolecule Conformations 42. Krittanai, C. and Johnson, W.C. (1997)

43.

44.

45.

46.

47.

Correcting the circular dichroism spectra of peptides for contributions of absorbing side chains. Anal. Biochem., 253, 57–64. Chakrabartty, A., Kortemme, T., Padmanabhan, S., and Baldwin, R.L. (1993) Helix propensities of the amino acids measured in alanine-based peptides without helix-stabilizing side-chain interactions. Biochemistry, 32, 5560–5565. Che, Y. and Marshall, G.R. (2006) Impact of cis-proline analogs on peptide conformation. Biopolymers, 81, 392–406. Rose, G.D., Gierasch, L.M., and Smith, J.A. (1985) Turns in peptides and proteins. Adv. Protein Chem., 37, 1–109. Bazan, J.F., Fletterick, R.J., McKinley, M.P., and Prusiner, S.B. (1987) Predicted secondary structure and membrane topology of the scrapie prion protein.Protein Eng., 1, 125–135. Neukirch, S., Goriely, A., and Hausrath, A.C. (2008) Chirality of

48.

49.

50.

51.

52.

53.

coiled coils: elasticity matters. Phys. Rev. Lett., 100, 038105-1–4. Rubin, N., Perugia, E., Goldschmidt, M., Fridkin, M., and Addadi, L. (2008) Chirality of amyloid suprastructures. J. Am. Chem. Soc., 130, 4602–4607. Caspar, D. and Klug, A. (1962) Physical principles in the construction of regular viruses. Cold Spring Harb. Symp. Quant. Biol., 27, 1–24. Caspar, D. and Klug, A. (1960) The structure of small viruses. Adv. Virus Res., 7, 225–325. Horne, R.W. and Wildy, P. (1961) Symmetry in virus architecture. Virology, 15, 348–373. Buenemann, M. and Lenz, P. (2008) Elastic properties and mechanical stability of chiral and ﬁlled viral capsids. Phys. Rev. E, 78, 051924-1–21. Xu, J., Keymeulen, A.V., Wakida, N.M., Carlton, P., and Berns, M.W. (2007) Polarity reveals intrinsic cell chirality. Proc. Natl. Acad. Sci. U.S.A., 104, 9296–9300.

313

13 Collisional Mechanism–Based E-DNA Sensors: A General Platform for Label-Free Electrochemical Detection of Hybridization and DNA Binding Proteins Francesco Ricci

13.1 Introduction

Pathogen detection or diagnosis of genetic diseases based on the sequence-speciﬁc identiﬁcation of DNA has attracted signiﬁcant attention [1, 2] due to the fact that these methods are usually more accurate and much faster than culture-based methods. Optical [1–4], electronic [5], acoustic [6], and gravimetric [7, 8] techniques have been applied to this goal. Although each of these approaches has its own advantages, the electrochemical detection of DNA hybridization appears promising due to its rapid response time, low cost, and suitability for mass production [9–12]. For this reason, many electrochemical DNA detection schemes have been described to date [11–13], the best of which achieve limits of detection ranging from picomolar to femtomolar [14–18]. Recently, Plaxco, Heeger, and coworkers, concomitantly with other groups, have introduced a number of single-step, label-free electrochemical biosensors, termed E-DNA sensors, that are based on the target binding–induced folding of electrode-bound DNA probes [19–22]. These sensors, in their original format, were designed as the electrochemical equivalent of optical molecular beacons [19–22]. The original E-DNA sensor is in fact comprised of a redox-modiﬁed ‘‘stem-loop’’ probe that is immobilized on the surface of a gold electrode via self-assembled monolayer chemistry (Figure 13.1). In the absence of target the stem-loop holds the redox moiety in proximity to the electrode, producing a large Faradaic current. Upon target hybridization, the stem is broken and the redox moiety moves away from the electrode surface. This, in turn, reduces the efﬁciency with which electrons are transferred to the electrode, thus suppressing the Faradaic current in proportion to the fraction of probe DNAs that have hybridized (Figure 13.1). This approach seemed to be particularly promising for oligonucleotide detection because of its rapidity, reagentless nature, and operational convenience [20, 22] and demonstrated to be a potentially promising method for the detection of both Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

13 Collisional Mechanism–Based E-DNA Sensors MB

Target

MB

AC current (A)

314

−0.5

−0.4

−0.3

−0.2

−0.1

Potential (mV)

Figure 13.1 E-DNA sensors [13–16], in their original format, were designed as the electrochemical equivalent of optical molecular beacons comprised of a redox-modiﬁed ‘‘stem-loop’’ probe immobilized on the surface of a gold electrode. In the absence of a target, the stem-loop structure holds the electrochemical tag (here methylene

blue) into close proximity with the electrode surface, thus ensuring efﬁcient redox of the electrochemical label (lower panel). Upon hybridization with the target sequence collisions between the redox tag and electrode are inhibited, decreasing the signaling current. Reproduced with permission from [19].

DNA and RNA sequences (see, e.g., commentary by Palececk et al. [23] and Thorp et al. [24]). E-DNA platform, as its optical counterpart, proved to be highly speciﬁc. E-DNA signaling is in fact driven by hybridization, and therefore, the sensor response is sequence-speciﬁc, thereby allowing to distinguish between a fully complementary 17-base target and a 50 000-fold excess of genomic DNA [20]. Unlike other similar optical approaches, the E-DNA platform is very selective and was often demonstrated several times to be effective in measuring DNA sequences in real complex matrices such as blood serum, soil extracts, and foodstuff (e.g., [25–27]). Also, an additional feature that differentiates this approach from the optical one, is that the E-DNA sensor is reusable. The redox-modiﬁed probe DNA is in fact strongly bonded to the interrogating electrode via a thiol-gold linkage and after hybridization, a brief, low ionic strength wash (30 seconds in room temperature distilled water) is usually sufﬁcient to recover >99% of the original sensor signal even for sensors that have been challenged directly in complex matrices [25, 28, 29].

13.2 E-DNA Signaling Mechanism

Fabrication and interrogation procedures for E-DNA sensors are also straightforward [30] and suitable for in situ analysis with portable instrumentation and disposable sensors.

13.2 E-DNA Signaling Mechanism

Even if E-DNA sensors [20–22] have been extensively studied in recent years, key issues in their fabrication were only recently disclosed, shedding light on the possible signal mechanism of the E-DNA platform. We have recently studied the effects of probe surface density, target length, and other aspects of molecular crowding on the signaling properties, selectivity, and response time of the E-DNA sensor trying to elucidate the possible mechanistic principles of this platform [29]. As a test bed for this study, we have employed a signal-off E-DNA sensor directed against 17-bases of the gyrB gene of Salmonella typhimurium [20, 31–33]. As with all E-DNA signal-off sensors, in the absence of the target DNA, the sensor gives a sharp, well-deﬁned ACV peak at ∼–0.25 V (vs Ag/AgCl) consistent with the formal potential of the methylene blue redox moiety used (Figure 13.1). We found that the signal suppression observed at a given target concentration is a biphasic function of probe density (the number of probe DNA strands per square centimeter of electrode). At low probe densities, signal suppression decreases with increasing probe density, presumably because the increased charge density on the electrode repels the target. In contrast, at higher probe densities, the signal suppression increases dramatically with increasing probe density (Figure 13.2). The transition between these two behaviors occurs at the probe density at which the mean probe-to-probe separation approximates the length of the probe-target duplex, suggesting that molecular crowding plays a role in the effect. This result suggested that probably, in contrast with what was previously speculated, in the E-DNA signaling mechanism, even the probe-target duplex can, at certain conditions, efﬁciently transfer electrons to the electrode. This is particularly apparent at low probe densities, where the signal suppression observed upon hybridization is limited because even hybridized probe DNA collides with the electrode surface and transfers electrons relatively efﬁciently. At high probe densities, in contrast, the steric bulk of the closely packed DNA probes inhibits these collisions, leading to higher signal suppression despite presumably lower hybridization efﬁciency. These results provided insights into the mechanism of E-DNA signaling. As said before, the original stem-loop structure of E-DNA sensors was chosen based on the speculation that, as with the optical counterpart, the signal suppression observed upon hybridization was solely due to the increased tunneling distance between the redox moiety and the electrode [20–22]. The results obtained in this work, however, suggest an alternative mechanism for the observed signaling: hybridization induces changes in the rate with which the redox moiety collides with the electrode surface. The formation of the stem induces efﬁcient electron transfer at both low and high probe densities. At low

315

13 Collisional Mechanism–Based E-DNA Sensors

Mean probe-to-probe separation (nm) 50

10

80 70 % Signal suppression

316

60 50 40 30 20 10 1e+ 11

1e+ 12

Surface density (molecules / cm2) Figure 13.2 We have recently demonstrated that signaling of E-DNA sensors is dependent on probe density. We have controlled probe surface density by varying the concentration of probe DNA employed during sensor fabrication. At high probe densities,

the observed signal suppression decreases monotonically with decreasing density. Shown is the dependence observed at 200 nM of a 17-base target DNA is shown. Reprinted with permission from [29]. Copyright (2007) American Chemical Society.

probe densities, however, the signal suppression observed upon hybridization is limited because even hybridized probe DNA can, on occasion, collide with the electrode surface and transfer electrons. At high probe densities, in contrast, the steric bulk of the closely packed probe DNAs precludes these collisions, leading to higher signal suppression despite presumably lower hybridization efﬁciency [29]. This collisional model of E-DNA signaling is further supported by studies of the sensor performance as a function of AC frequency probe and of target length and bulk [29]. For example, longer and bulkier targets produce greater signal suppression, presumably because they reduce the collision rate of the probe-target duplex (Figure 13.3). Likewise, very little target-induced signaling is observed when E-DNA sensors are probed with low frequency AC potentials (AC voltammetry), suggesting that, on a sufﬁciently slow time-scale, electron transfer is efﬁcient from both the free and bound probe. As the ACV frequency increases above a threshold (presumably deﬁned by the slower collision rate of the probe-target duplex), transfer efﬁciency from the bound state is reduced, producing a large, hybridization-linked reduction in the sensing current. Consistent with this argument, the frequencies at which this transition occurs are higher for lower-density sensors, presumably because reduced probe densities leads to higher collision rates for the probe-target duplex [29].

13.2 E-DNA Signaling Mechanism

100

% Signal suppression

80 High density 60 Low density 40 Medium density 20

0 0.1

1

10 100 ACV frequency (Hz)

Figure 13.3 E-DNA signal suppression is eliminated at low ACV frequencies. This presumably occurs when the ACV frequency falls below the rate at which the redox moiety of the probe-target duplex collides with the electrode; under these conditions, electron transfer is efﬁcient from both free and bound probes. As the ACV frequency increases, it

1000

10 000

presumably surpasses the collision rate of the bound probe, reducing transfer efﬁciency and producing a large, hybridization-linked reduction in the sensing current. Shown are the signal suppressions achieved at 200 nM of a 17-base target. Reprinted with permission from [29]. Copyright (2007) American Chemical Society.

We believe our results support the claim made by Anne and Demaille that the intrinsic bending elasticity of DNA controls the dynamics of electron transport in molecular layers comprised of surface-attached, redox-modiﬁed DNA [34]. We provided further evidence for this collisional signaling mechanism also demonstrating that careful optimization of probe density and measurement techniques is necessary in order to achieve maximum performance across this broad and increasingly important class of sensors. Of more importance, we have started to elucidate the possible mechanism of E-DNA sensors proposing a ‘‘collisional’’ signaling mechanism that opens the future to a new class of E-DNA sensors, which is not anymore forced to the stem-loop structure of original E-DNA platforms. All of the groups responsible for the initial development of E-DNA sensors, in fact, employed stem-loop DNA probes [20–22, 25], presumably due to the misconception that, by analogy to molecular beacons, a speciﬁc conformational (i.e., geometric) change is required in order to support robust signaling. As a conﬁrmation of the hypothesis that a stem-loop conformation is not necessarily needed for E-DNA sensors, we have recently proposed the use of a ssDNA linear probe for E-DNA-based sensing and demonstrated that binding-induced changes in DNA dynamics are sufﬁcient to support E-DNA signaling, thus giving further evidence to the ‘‘collisional’’ theory [35]. Linear probe

317

13 Collisional Mechanism–Based E-DNA Sensors

E-DNA sensors maintain all the positive features of their stem-loop counterparts being label-free, reusable, sequence-speciﬁc, and selective enough to employ directly in complex sample matrices such as blood serum, thus rendering them well suited for clinical applications [35]. We have fabricated E-DNA sensors using a 27-base linear probe sequence that, in order to facilitate direct comparison with earlier studies, is directly analogous to the previously characterized stem-loop E-DNA sensor used for density effect experiments [35]. In the absence of a target, the sensor gives rise to a sharp, well-deﬁned AC voltammetry peak consistent with the ∼–0.25 V (vs Ag/AgCl) formal potential of the methylene blue redox moiety employed (Figure 13.4). Upon hybridization to a fully complementary, 17-base target this current is signiﬁcantly reduced. Like the original stem-loop E-DNA architecture, the linear probe E-DNA sensor is label-free and reusable: a 30 seconds wash in room temperature distilled water or MB

MB

MB

MB

MB

MB MB

MB

eT

MB

eT

eT

Hybridization

eT MB

(a) Regeneration AC current (1e-7 A)

318

Buffer

6 Regeneration 4

Serum/buffer 50/50

Target

Target 2 −0.4

(b)

−0.3 −0.2 Potential (V)

−0.1

Figure 13.4 E-DNA signaling arises due to hybridization-induced changes in probe ‘‘collision’’ dynamics (rather than to a conformational change per se) and thus redox-modiﬁed linear probe DNAs serve as effective E-DNA sensors (a) [35]. The Faradaic current arising from such a linear probe DNA is signiﬁcantly reduced in the presence of a complementary target

−0.4 (c)

−0.3 −0.2 Potential (V)

−0.1

sequence because, as demonstrated here, hybridization reduces the rate with which the terminal redox tag collides with the electrode surface and transfers electrons (b). Linear probe E-DNA sensors respond well in either (b) buffer or (c) 50% blood serum. Reproduced with permission of the Royal Society of Chemistry from [35].

13.3 E-DNA Sensor for DNA Binding Proteins Detection

(after deployment in blood serum) room temperature detergent solution is enough to regenerate >97% original sensor current (Figure 13.4). Of note, such linear probe sensors exhibit somewhat improved signal gain over their stem-loop counterparts presumably because target binding no longer competes with breakage of the stem structure. Despite this, however, while the signal suppression observed in serum is quite similar to that obtained in buffer the currents observed in serum are lower. This may be due to the greater viscosity and/or the reduced ionic strength of this medium, which would reduce collision rates and thus reduce electron transfer efﬁciency; no similar drop-off is observed for stem-loop sensors, presumably because the stem structure ﬁxes the (MB) methylene blue near the surface, rendering its electron transfer rate relatively independent of these effects [35]. The disclosure that E-DNA signaling requires only that a target binds to an oligonucleotide probe and, in doing so, changes the efﬁciency with which the attached redox tag strikes the electrode could account for the approach’s generalizability, thus suggesting that the same sensing principle would also support the detection of DNA binding events that lead to the formation of bulky and/or rigid complexes. In order to test this suggestion we have recently proposed the use of E-DNA-like sensors for the detection of DNA binding proteins. These proteins are abundant and essential in cells, interacting with DNA in order to organize their packing, regulate transcription, and perform replication and repair [36, 37].

13.3 E-DNA Sensor for DNA Binding Proteins Detection

For this study we have selected two double-strand binding proteins, the eukaryotic TATA-box binding protein (TBP) (a core component of the eukaryotic transcriptional machinery), and the prokaryotic M.HhaI methyltransferase (M.HhaI) (involved in the restriction–modiﬁcation system of bacteria); and two single-strand binding proteins (SSBPs) involved in the replication machinery, the prokaryotic SSBP and the eukaryotic replication protein A (RPA) [38]. We have fabricated sensors against the double-strand binding proteins using short, stem-loop probe DNAs in which the relevant recognition sequences are contained within the double-stranded stem. These probes were modiﬁed with a 3 thiol group, supporting strong chemisorption to an interrogating electrode, and a methylene blue redox tag pendant on a thymine base along the double-stranded stem (for TBP, Figure 13.5a and c) or within the single-stranded loop (for M.HhaI, Figure 13.5b and d). In the absence of a target, both probes produce a large faradaic peak at the potential expected for the methylene blue redox tag (Figure 13.5c and d). This current is due to the collisions of the relatively dynamic dsDNA probes. The relative dynamicity of the probe is ensured by the fact that this was designed to have a short (ﬁve bases) single-stranded element at the proximity of the gold electrode, thus ensuring enough freedom for the collision of the probe.

319

320

13 Collisional Mechanism–Based E-DNA Sensors

TBP

M. Hhal

MB

MB

MB MB

MB MB

MB

MB

MB

MB MB

MB

MB MB

eT

eT

M. Hhal TBP

(a)

(b) Buffer

−0.5 (c)

−0.3 V vs. Ag/AgCl, 3M NaCl

−0.1

10 nA

+ TBP

Regeneration 10 nA

Regeneration

−0.5 (d)

Figure 13.5 Schematics of E-DNA-like sensors for the detection of DNA binding proteins. (a, b) The sensor is comprised of a DNA hairpin covalently attached to a classic gold rod electrode using thiol-gold self-assembled monolayer chemistry and containing an internal methylene blue redox tag. (c, d) In the absence of target relatively

Buffer

+ M. Hhal

−0.3 V vs. Ag /AgCl, 3 M NaCl

−0.1

efﬁcient collision between the label and the electrode produces a large faradic current. Upon target binding this faradic current is signiﬁcantly reduced, presumably because the bulky structure of the protein reduces the collision rate. (Reprinted with permission from [38]. Copyright (2009) American Chemical Society.

In the presence of saturating TBP and M.HhaI, the binding of these proteins leads to a bulk and rigid DNA–protein complex, which in turns reduces the currents by 45 and 55% respectively. The two sensors support the ready detection of their target proteins at concentrations as low as 2 and 25 nM for TBP and M.HhaI respectively. Again, exactly as in the case of E-DNA sensors for hybridization detection, because all of the sensing components are strongly adsorbed to the electrode surface, this sensing architecture is readily regenerable; a short wash (30 seconds) in 8 M guanidine chloride is sufﬁcient to regenerate 98% of the original signaling current of both sensors (Figure 13.5c and d), allowing multiple cycles of detection and regeneration (data not shown). Of note is that, it is important to stress that not only dsDNA but also ssDNA probes support this analytical approach, enabling the sensitive, convenient detection of proteins that bind to such targets. Using single-stranded, poly-thymine probes (neither of our target proteins exhibits any signiﬁcant sequence speciﬁcity) we have fabricated sensors for the detection of the SSBP (Figure 13.6) and RPA. Of note, because the single-stranded DNA is thought to wrap entirely around SSBP, the gain of these sensors is dependent on the length of the single-stranded probe employed: when targeting this protein the observed signal suppression (at saturating target concentration) increases from 20 to 70% as the probe is lengthened from 20 to 70

13.4 Conclusions and Future Perspectives

321

MB

MB MB

MB MB

eT

eT

MB

SSBP

eT

(a) 40-base probe Buffer 0.2 µA

Regeneration

1 µA

+ SSBP

−0.5

−0.3

(b)

V vs. Ag/AgCl, 3 M NaCl

−0.1

70-base probe Buffer

Regeneration 0.05 µA

20-base probe Buffer

+ SSBP

−0.5

−0.3 V vs. Ag/AgCl, 3M NaCl

Figure 13.6 The sensing mechanism also holds for single-stranded probes, thus enabling the detection of Escherichia. coli single-strand binding protein (SSBP). (a) The faradaic current arising from such a single-stranded probe is signiﬁcantly reduced upon binding with this target. (b) Of note,

−0.1

−0.5

Regeneration

+ SSBP

−0.3 V vs. Ag/AgCl, 3M NaCl

because the single-stranded probe wraps around the protein target, different length probes (20, 40, and 70 bases) giving rise to different responses upon protein binding. Shown are responses to 80 nM of SSBP. Reprinted with permission from [38]. Copyright (2009) American Chemical Society.

bases (Figure 13.6b). The same change in probe length is associated with a 5–70% increase in gain in the presence of RPA, an effect that could arise when multiple RPAs bind to the longer probe [39]. All three sensors are as selective as the DNA probes from which they are fabricated. For example, we do not detect any signiﬁcant cross-reactivity between the two sensors directed against double-strand binding proteins (Figure 13.7a). (Neither of the two SSBPs investigated here exhibits any signiﬁcant sequence speciﬁcity and thus, while the signal gain produced by the two proteins is not identical, the ability of our sensor to discriminate between these two targets is limited.) Likewise, because their signaling is linked to a binding-speciﬁc change in the probe DNA (and not simply to adsorption of target to the sensor surface), our sensors are effective in rejecting false positives arising due to the nonspeciﬁc adsorption of interferents and can be employed directly in complex samples. For example, our single-strand sensor supports the selective detection of exogenous levels of RPA directly in crude Raji cell nuclear extracts (Figure 13.7b). 13.4 Conclusions and Future Perspectives

We believe the studies we have summarized in this chapter represent a strong basis for the hypothesis that signaling in our sensors is dominated by binding-induced

−0.1

0 (a)

TBP

M. Hhal

Figure 13.7 DNA binding protein sensors are as selective as the DNA probes from which they are fabricated. (a) For example, no signiﬁcant cross-reactivity is observed between the two sensors directed against double-strand binding proteins in the presence of 1.5 µM BSA and saturating concentrations of the nontargeted double-strand binding proteins (80 nM M.HhaI or 10 nM TBP). (b) Likewise our sensors are also

Raji 225 µg/ml

Raji 125 µg/ml

Comp. seq. 10 µM

10

Raji 75 µg/ml

20

Raji 25 µg/ml

% Signal suppression

SAH

TBP

BSA

Buffer

TBP M. Hhal

25

BSA

50

Buffer

% Signal suppression

75

Raji 175 µg/ml

13 Collisional Mechanism–Based E-DNA Sensors

M. Hhal

322

0

(b) effective in rejecting false positives arising due to interferents and perform well when challenged with realistically complex sample matrices. For example, we can detect the exogenous level of RPA in crude Raji nuclear extracts and partially restore initial signal by adding high concentration of a competitor sequence (polyT-70). Reprinted with permission from [38]. Copyright (2009) American Chemical Society.

changes in the efﬁciency with which the redox tag collides, and thus transfers electrons to the interrogating electrode. This is of primary importance and we have demonstrated that this opens the possibility to the use of DNA probe with no stringency of stem-loop conformation. As a further advancement, the collisional mechanism makes possible the monitoring of theoretically any DNA–protein or DNA–antibody interactions, which leads to the formation of bulky and/or rigid complexes. We have conﬁrmed this suggestive hypothesis with a three-successes-out-of-three-attempts success rate in fabricating sensors targeting both sequence-speciﬁc, double-strand binding proteins and nonspeciﬁc, SSBPs. The E-DNA sensing platform has then demonstrated to be not only a promising and appealing approach for the sequence-speciﬁc detection of DNA and RNA, but also to be ﬂexible enough to be adaptable for the detection of DNA–protein interactions. For future practical applications of this new sensing platform, however, some drawbacks have to be considered and overcome. For many applications, detection of speciﬁc DNA sequences is still not sensitive enough and will require the coupling of PCR ampliﬁcation. Moreover, miniaturization of the sensing platform, although possible, has not yet been demonstrated. Recently, we have however shown the use of screen printed electrodes with E-DNA sensors, which represent a ﬁrst step toward the automation and mass production of these sensors [40]. Long-term and operational stability of the sensors is another issue, which also deserves a deeper study before this platform could be applied in the real world.

13.4 Conclusions and Future Perspectives

Generalizability and ﬂexibility of E-DNA are among the most appealing advantages. For example, the same sensing platform could be expanded to a wide range of other targets not only limited to those that bind to unmodiﬁed DNA or RNA. The collisional mechanism suggests, in fact, also a ready means of detecting proteins/enzymes that bind small molecules by simply appending the small molecule to the DNA probe, thus expanding widely, the possible E-DNA applications. We have recently demonstrated this interesting suggestion by appending a small molecule recognition element onto a relatively rigid, partially double-stranded DNA scaffold that is chemi-adsorbed to an interrogating gold electrode [41]. One of the two scaffold strands, the anchoring strand, is linked to the electrode using self-assembled monolayer chemistry via a 3 thiol group and is modiﬁed with a redox tag (methylene blue) at its 5 terminus. The second strand, the recognition strand, is complementary to a region of the anchoring strand and is covalently modiﬁed with the relevant small molecule recognition element at one of its termini. As a preliminary test for the development of this technology we have employed the small molecules biotin and digoxigenin as recognition elements to fabricate sensors directed against streptavidin and antidigoxigenein antibodies, respectively. In the absence of a target, the modiﬁed, double-stranded scaffold is free to collide with the electrode surface, and thus produces a large faradaic current at the potential expected for methylene blue. In preliminary experiments, we observed that upon target binding, this current is reduced, signaling the presence of the target. This reduction presumably occurs because target binding reduces the efﬁciency with which the redox tag collides with the electrode, either due to the increase in steric bulk or due to the hydrodynamic radius associated with the target. In the case of multivalent targets (such as antibodies), another cause may be that the two antigen binding sites on the antibody cross-link two scaffolds. This new sensing approach offers several signiﬁcant advantages over other methods for monitoring protein–small molecule interactions or detecting proteins that bind to speciﬁc small molecule targets. For example, standard methods for protein detection, such as ELISAs and western blots, are time and capital intensive, and typically require the addition of reagents and multiple washing steps [42, 43]. Less cumbersome methods for monitoring protein–small molecule interactions, such as ﬂuorescence polarization, and plasmon resonance, surface plasmon resonance (SPR) and quartz crystal microbalance (QCM) approaches, often fail in complex samples because of background ﬂuorescence [44] or nonspeciﬁc adsorption [45, 46]. In contrast, the approach we have demonstrated here is rapid and convenient, and functions even when challenged directly in clinically and environmentally relevant samples. Moreover, it appears that this new approach is general; in principle, given a target macromolecule large enough to alter the collision dynamics of the scaffold, the only limiting factor is the ability to effectively conjugate the relevant small recognition molecule to a relatively rigid and dynamic molecular scaffold [41].

323

324

13 Collisional Mechanism–Based E-DNA Sensors

Acknowledgments

The work described here has been performed at the Prof. Plaxco’s lab of the University of California, Santa Barbara. Grateful acknowledgment for his scientiﬁc and economic support is given to Prof. Kevin W. Plaxco for hosting me in two different periods and for making possible this work. All the members of his research group, for helpful discussions and comments on the manuscript, are also thanked. I am also grateful to my supervisor, Prof. Giuseppe Palleschi of the University of Rome, Tor Vergata for giving me the possibility of this work.

References 1. Cao, Y.W.C., Jin, R.C., and Mirkin, C.A.

2.

3.

4.

5.

6.

7.

8.

(2002) Nanoparticles with Raman spectroscopic ﬁngerprints for DNA and RNA detection. Science, 297, 1536–1540. Gaylord, B.S., Heeger, A.J., and Bazan, G.C. (2002) DNA detection using water-soluble conjugated polymers and peptide nucleic acid probes. Proc. Natl. Acad. Sci. U.S.A., 99, 10954–10957. Peterson, A.W., Wolf, L.K., and Georgiadis, R.M. (2002) Hybridization of mismatched or partially matched DNA at surfaces. J. Am. Chem. Soc., 124, 14601–14607. Ahn, S. and Walt, D.R. (2005) Detection of Salmonella spp. using microsphere-based, ﬁber-optic DNA array. Anal. Chem., 77 (15), 5041–5047. Kelley, S.O., Barton, J.K., Jackson, N.M., and Hill, M.G. (1997) Preparation of electrode-immobilized, redox-modiﬁed oligonucleotides for electrochemical DNA and aptamer-based sensing. Bioconjug. Chem., 8, 31–37. Cooper, M.A., Dultsev, F.N., Minson, T., Ostanin, V.P., Abell, C., and Klenerman, D. (2001) Direct and sensitive detection of a human virus by rupture event scanning. Nat. Biotechnol., 19, 833–837. Endo, T., Kerman, K., Nagatani, N., Takamura, Y., and Tamiya, E. (2005) Label-free detection of peptide nucleic acid-DNA hybridization using localized surface plasmon resonance based optical biosensor. Anal. Chem., 77 (21), 6976–6984. Patolsky, F., Lichtenstein, A., and Willner, I. (2001) Preparation of

9.

10.

11.

12.

13.

14.

15.

electrode-immobilized, redox-modiﬁed oligonucleotides for electrochemical DNA and aptamer-based sensing. J. Am. Chem. Soc., 123, 5194–5205. Park, S.J., Taton, T.A., and Mirkin, C.A. (2002) Array-based electrical detection of DNA with nanoparticle probes. Science, 295, 1503–1506. Boon, E.M., Ceres, D.M., Drummond, T.G., Hill, M.G., and Barton, J.K. (2000) Mutation detection by electrocatalysis at DNA-modiﬁed electrode. Nat. Biotechnol., 18, 1096–1100. Drummond, T.G., Hill, M.G., and Barton, J.K. (2003) Electrochemical DNA sensors. Nat. Biotechnol., 21, 1192–1199. Thorp, H.H. (1998) Cutting out the middleman: DNA biosensors based on electrochemical oxidation. Trends Biotechnol., 16, 117–121. Katz, E. and Willner, I. (2003) Probing biomolecular interactions at conductive and semiconductive surfaces by impedance spectroscopy: routes to impedimetric immunosensors, DNA-sensors, and enzyme biosensors. Electroanalysis, 15, 913–947. Gore, M.R., Szalai, V.A., Ropp, P.A., Yang, I.V., Silverman, J.S., and Thorp, H.H. (2003) Detection of attomole quantities of DNA targets on gold microelectrodes by electrocatalytic nucleobase oxidation. Anal. Chem., 75, 6586–6592. Floch, L.F., Ho, H.A., Harding-Lepage, P., Bedard, M., Neagu-Plesu, R., and Leclerc, M. (2005)

References

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

Ferrocene-functionalized cationic polythiophene for the label-free electrochemical detection of DNA. Adv. Mater., 17, 1251–1254. Gibbs, J.M., Park, S.J., Anderson, D.R., Watson, K.J., Mirkin, C.A., and Nguyen, S.T. (2005) Polymer-DNA hybrids as electrochemical probes for the detection of DNA. J. Am. Chem. Soc., 127, 1170–1178. Kim, E., Kim, K., Yang, H., Kim, Y.T., and Kwak, J. (2003) Enzyme-ampliﬁed electrochemical detection of DNA using electrocatalysis of ferrocenyl-tethered dendrimer. Anal. Chem., 75, 5665–5672. Kerman, K., Matsubara, Y., Morita, Y., Takamura, Y., and Tamiya, E. (2004) Recent trends in electrochemical DNA biosensor technology. Sci. Technol. Adv. Mater., 5, 351–357. Ricci, F. and Plaxco, K.W. (2008) E-DNA sensors for convenient, label-free electrochemical detection of hybridization. Microchim. Acta, 163, 149–155. Fan, C., Plaxco, K.W., and Heeger, A.J. (2003) Electrochemical interrogation of conformational changes as a reagentless method for the sequence-speciﬁc detection of DNA. Proc. Natl. Acad. Sci. U.S.A., 100, 9134–9137. Immoos, C.E., Lee, S.J., and Grinstaff, M.W. (2004) Conformationally gated electrochemical gene detection. Chembiochem, 5, 1100 . Mao, T., Luo, C., and Ouyang, Q. (2003) Studies of temperature-dependent electronic transduction on DNA hairpin loop sensor. Nucleic Acids Res., 31, 108 . Palecek, E. (2004) Surface-attached molecular beacons light the way for nucleotide sequencing. Trends Biotechnol., 22, 55–58. Thorp, H.H. (2003) Reagentless detection of DNA sequences on chemically modiﬁed electrodes. Trends Biotechnol., 21, 522–524. Lubin, A.A., Lai, R.Y., Baker, B.R., Heeger, A.J., and Plaxco, K.W. (2006) Sequence-speciﬁc, electronic detection of oligonucleotides in blood, soil, and foodstuffs with the reagentless, reusable E-DNA sensor. Anal. Chem., 78 (16), 5671–5677.

26. Lubin, A.A., Fan, C., Schafer, M.,

27.

28.

29.

30.

31.

32.

33.

34.

Clelland, C.T., Bancroft, C., Heeger, A.J., and Plaxco, K.W. (2008) Rapid, electronic detection of DNA and non-natural DNA analogs for molecular marking applications. Forensic Sci. Commun., 10 (1), 1–6. Lai, R.Y., Seferos, D.S., Heeger, A.J., Bazan, G.C., and Plaxco, K.W. (2006) Comparison of the signaling and stability of electrochemical DNA sensors fabricated from 6-or 11-carbon self-assembled monolayers. Langmuir, 22, 10796–10800. Lai, R.Y., Lagally, E.T., Lee, S.-H., Soh, H.T., Plaxco, K.W., and Heeger, A.J. (2006) Rapid, sequence-speciﬁc detection of unpuriﬁed PCR amplicons via a reusable, electrochemical sensor. Proc. Natl. Acad. Sci. U.S.A., 103, 4017–4021. Ricci, F., Lai, R.Y., Heeger, A.J., Plaxco, K.W., and Sumner, J.J. (2007) Effect of molecular crowding on the response of an electrochemical DNA sensor. Langmuir, 23, 6827–6834. Xiao, Y., Lai, R., and Plaxco, K.W. (2007) Preparation of electrode-immobilized, redox-modiﬁed oligonucleotides for electrochemical DNA and aptamer-based sensing. Nat. Protoc., 2, 2875–2880. Lubin, A.A., Vander Stoep Hunt, B., and Plaxco, K.W. (2008) The effects of probe length, probe geometry and redox-tag placement on the performance of the electrochemical E-DNA sensor. Anal. Chem., 81 (6), 2150–2158. Lai, R.Y., Lee, S.-H., Soh, H.T., Plaxco, K.W., and Heeger, A.J. (2006) Differential labeling of closely-spaced biosensor electrodes via oxidative desorption. Langmuir, 22, 1932–1936. Pavlovic, E., Lai, R.Y., Wu, T.T., Ferguson, B.S., Sun, R., Plaxco, K.W., and Soh, H.T. (2007) Speciﬁc, electromechanical detection of multiple DNA sequences in an integrated microﬂuidic system. Langmuir, 24 (3), 1102–1107. Anne, A. and Demaille, C. (2006) Dynamics of electron transport by elastic bending of short DNA duplexes. Experimental study and quantitative modeling of the cyclic voltammetric behavior of

325

326

13 Collisional Mechanism–Based E-DNA Sensors

35.

36.

37.

38.

39.

40.

30-ferrocenyl DNA endgrafted on gold. J. Am. Chem. Soc., 128, 542–557. Ricci, F., Lai, R.Y., and Plaxco, K.W. (2007) Linear, redox modiﬁed DNA probes as electrochemical DNA sensors. Chem. Commun., 3768–3770. Kadonaga, J.T. (2004) Regulation of RNA polymerase II transcription by sequence-speciﬁc DNA-binding factors. Cell, 116 (2), 247–257. Bulyk, M.L. (2006) DNA microarray technologies for measuring protein-DNA interactions. Curr. Opin. Biotechnol., 17 (4), 422–430. Ricci, F., Bonham, A.J., Mason, A.C., Reich, N.O., and Plaxco, K.W. (2009) A reagentless, electrochemical approach for the speciﬁc detection of double- and single-stranded DNA binding proteins. Anal. Chem., 81 (4), 1608–1614. Kim, C., Paulus, B.F., and Wold, M.S. (1994) Interactions of human replication protein A with oligonucleotides. Biochemistry, 33, 14197–14206. Ricci, F., Zari, N., Caprio, F., Recine, S., Amine, A., Moscone, D., Palleschi, G.,

41.

42.

43. 44.

45.

46.

and Plaxco, K.W. (2009) Surface chemistry effects on the performance of an electrochemical DNA sensor. Bioelectrochemistry, 76 (1-2), 208–213. Cash, K., Ricci, F., and Plaxco, K.W. (2009) An electrochemical sensor for the detection of protein-small molecule interactions directly in serum and other complex matrices. J. Am. Chem. Soc., 131 (20), 6955–6957. Crowther, J.R. (2001)in The ELISA Guidebook, vol. 149 (ed. J.M.Walker), Humana, Totawa . Wild, D.(ed.) (2001) The Immunoassay Handbook, 2nd edn, Nature, London . Owicki, J.C. (2000) Fluorescence polarization and anisot- ropy in high throughput screening: perspectives and primer. J. Biomol. Screen., 5, 297–306. Homola, J. (2003) Present and future of surface plasmon resonance biosensors. Anal. Bioanal. Chem., 377, 528–539. Cooper, M.A. (2003) Label-free screening of bio-molecular interactions. Anal. Bioanal. Chem., 377, 834–842.

327

Index

a acarbose 240 activated esters 62 adenosylcobalamin 97 – dependent reactions 97–99 5 -adenyl-β, γ -amidotriphosphate (AMP-PNP) 172 aggregation mechanism 9 alanine 69 aldehydes – oxime ligation 78–79 – site-speciﬁc modiﬁcation 79 alkylation 65–66 alkyne – Cu-catalyzed [3 + 2] cycloaddition of 74 – modiﬁcation 79–80 allethrin 197 Alzheimer’s disease, ubiquitin-positive amyloid plaques 14 aminoethylation 66 aminoglycosides 240 aminoimidazole-4-carboxamide ribonucleotide (AICAR) 222–224 – cyclic peptide inhibitors of 222–224 5-amino-4-imidazolecarboxamide ribonucleotide transformylase/inosine 5 -monophosphate cyclohydrolase (ATIC) 222–223 amphotericin B 197 amyloid 13 amyloidogenic proteins, toxicity of 13 anisotropy 204 anti-infectives, structure-based design to 167–187 – 4-diphosphocytidyl-2C-methyl-d-erythritol kinase (IspE) 170–182 – isoprenoids 169–174

– nonmevalonate pathway 169–174 – X-ray cocrystal structure analysis 182–184 aquocobalamin 103 aromatic hydroxylation 50, 278–280 aromatic rings, hydroxylation of 278–285 – intramolecular aromatic hydroxylation 278–280 – phenolic compounds, intermolecular ortho-hydroxylation of 280–285 ascorbate oxidase 40 asthma 228–229 atomic force microscopy (AFM) 18 atox1 7 autophagy-lysosome pathway 12 avian proteins 305 β-axial position 102–108 – cobalamin alkylation 102–103 – heterodinuclear concept 103–108 azide – Cu-catalyzed [3 + 2] cycloaddition of 74 – modiﬁcation 79–80 azithromycin 197

b basic helix-loop-helix leucine zipper (bHLHZip) 216 bioconjugation 133–135 – streptavidin–biotin 134–135 biodegradable microspheres 119–120 – preparation 119–120 biological membranes 191–195 – composition of 191–193 – dynamic molecular organization 193–195 – life maintenance, role in 191 – structure of 191–193 biomimetic Cu complexes 272–285

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life. Edited by Bruno Pignataro Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32541-2

328

Index biostable microspheres 120–121 – applications 120 – preparation 120–121 – – dispersion polymerization 120 – – emulsion polymerization 120–121 biosynthesized libraries, high-throughput screening of 220–226 biotin 134–135 blue copper proteins 33, 268 brain copper 8–9 – homeostasis 5–8 brain, metal ions in 4 Brewster angle microscopy (BAM) 201, 205

c calculated partitioning coefﬁcient (clogP) 175, 177–178, 181 carbohydrate microarrays 257 carbohydrate synthesis 240–245 – chemical synthesis 241–243 – – automated synthesis 241 – – one-pot synthesis 241–242 – – solid-phase synthesis 242–243 – enzymatic synthesis 243–244 – – glycosidases 243 – – glycosyltransferases 243–244 – – glycosynthases 243 – glycoprotein synthesis 244–245 – – cell biological technique 245 – – enzyme remodeling 245 – – molecular biological technique 245 – – peptide ligation 244–245 carbohydrate-based drugs 240 carbon–carbon bond formation, palladium-catalyzed 76 catalytic systems 45–47 catechol 270 cecropin 197 cell-penetrating peptides 118 cell selection 130 cellular copper, redox state of 21–22 ceruloplasmin 40 chemical libraries 216–220 chemical protein modiﬁcation 59–82 – challenges in 61 – posttranslational modiﬁcation 59–60 – route to 60–61 – site-selective 70–81 – – dehydroalanine 71 – – dual modiﬁcation 81 – – metal-mediated protein modiﬁcation 71–78 – – unnatural amino acids, metal-free methods for modifying 78–81

– traditional methods for 61–70 – – aspartic acid modiﬁcation 64 – – cysteine 64–70 – – glutamic acid modiﬁcation 64 – – lysine modiﬁcation 62–64 chemoenzymatic synthesis 251–256 chirality 293–310 – intrinsic 297 – living systems, in 293–295 – perspectives 308–309 – protein dynamics, computational techniques for 297–298 – protein motions, analysis of 298–308 – – circular dichroism 305–308 – – protein dynamics, monitoring of 302–305 – – protein structure, understanding of 300–302 – protein secondary structures 295–296 – – assignment 296–297 – – coil 296 – – α helix 296 – – poly-l-proline II 296 – – β sheets 296 – – β turns 296 chirality index 298–300 – protein dynamics, tool for monitoring 302–305 circular dichroism 305–308 cisplatin 103, 197 cleavable microspheres 132 clioquinol 8 coarse-grained model 298 cobalamin – alkylation 102–103 – enzymatic reactions, dependent 98 – gastrointestinal pathway 96 – metabolism of 98 coiled coils 308 combinatorial chemistry 167, 216 Concanavalin A (ConA) 258 conformational diseases 13 conformational movement 195 copper(III) – bioinorganic implications 33–41 – – copper enzymes 33–36 – – mononuclear monooxygenating copper-based enzymes 38–40 – – particulate methano monooxygenase (pMMO) 36–38 – – trinuclear copper models for laccase 40–41 – cuprate superconducting materials 52 – organometallic CuIII species 41–51

Index – – aliphatic C–H bond functionalizations 45–51 – – aromatic C–H bond functionalizations 45–51 – – aryl–heteroatom bond formation 44–45 – – C–C bond formation 42–44 – redox potentials 32 – stabilization of 31 copper chaperone for superoxide dismutase (CCS) 7 copper dioxygen adducts 272–275 copper ion – CuII binding, spectroscopic characterization of 15–17 – CuII -induced self-oligomerization 17–18 – oxidation states 5 copper transport (Ctr) protein 6 covalent protein modiﬁcation 61 cross-ring cleavage 150–154 cuprate superconducting materials 52 cyanocobalamin (vitamin B12) 95 cyclic peptide inhibitors 224–226 cyclooxygenases (COXs) 201 cyclosporine 197 cysteine 59 – modiﬁcation 64–70 – – alkylation 65–66 – – desulfurization 67–70 – – disulﬁdes 66–67 – oxidative elimination of 72 cytidine 5 -diphosphate (CDP) binding pocket 170 – targeting of 174–182

d DEA, See dissociative electron attachment defensin 197 dehydroalanine 71 dehydrons 20 dendrimers 118 density functional theory (DFT) 147, 279 dichlorodiphenyl trichloroethane (DDT) 197 dictionary of protein secondary structures (DSSP) 296 dicyclohexylcarbodiimide (DCC) 124 dicyclohexylurea 124 diisopropylcarbodiimide (DIC) 124 dimethylbenzimidazole 102 dinuclear type-3 copper enzymes 33–36 dinucleating ligands 276 4-diphosphocytidyl-2C-methyl-d-erythritol Kinase (IspE) 170 – active site 170–174

– CDP-binding pocket, targeting 174–182 – – design 174–176 – – hydrophobic pocket, optimization of ﬁlling of 179–181 – – possible ribose analogues 175 – – ribose analogue, optimization of 176–178 – – sulfone substituents, evaluation of inhibitors featuring 180–181 – – vector design 176 – – vector, importance of 178–179 – structure 170 dispersion polymerization 120 dissociative electron attachment (DEA) 143 – electron attachment to d-ribose 148–150 – energetic position of 146 – gas-phase dna building blocks 147–148 – LIAD, use of 159 – sugar–phosphate cleavage 159–161 – tetraacetyl-d-ribose 155–159 – transient negative anions, nature of the 154–155 disuccinimidyl glutarate (DSG) 62 disuccinimidyl suberate (DSS) 62 disulﬁde bonds 132–133 disulﬁde cleavage 133 disulﬁdes 66–67 – contraction to thioether 67 – native chemical ligation 67–69 dithiothreitol (DTT) 71, 133 DNA, radiation damage 143–162 DNA backbone – model compounds – – DEA studies on 148–161 DNA binding proteins detection 319–321 dopamine β-monooxygenase (DβM) 38 double emulsion technique 120 double strand breaks (DSBs) 144 d-ribose 148–150 – cross-ring cleavage 150–154 drug–membrane interactions 191–206 – analysis and quantiﬁcation 199–201 – biological membranes 191–195 – clinical relevance of studies 195–199 – – drug development, contribution for 195–197 – – enzymatic inhibition, controlling 198–199 – – multidrug resistance (MDR) 198 – – therapeutic and toxic effect of drugs 197 – experimental techniques 200–201 – future research directions 206 – membrane model systems 199–200

329

330

Index drug–membrane interactions (contd.) – nonsteroidal anti-inﬂammatory drugs (NSAIDs), study of 201–206 – – drug fundamental physical–chemical studies 202–203 – – membrane dynamic studies 203–205 – – membrane structural studies 203–205 – – results 205–206 – possible effects 195–199 dual functionality 124 dual functionalized microspheres 123 dual modiﬁcation 81

g

f

h

ﬁrst-generation inhibitors 182 – water solubility of 182 ﬂuid mosaic model 191 ﬂuorescence quenching 203 fmoc chemistry 121–122 – basic steps 123 fosmidomycin 169 Franck–condon transition 158 freeze-fracture electron microscopy (FFEM) 201 fumonisin B1 198

H5N1 avian inﬂuenza 230 haloarenes, cross-coupling of 45 halothane 197 haptocorrin 96 h-bonds 20 α helix 296 hemagglutinin 230 hemocyanin, 33–34 269–270 – site of 269 hemoglobin 304 – helix of 304

gain-of-function mechanism 9 gene silencing 128 – microsphere-mediated 129 giant unilamellar vesicles (GUVs) 199 global aliphatic hydroxylation 39 glucansucrases 251–256 glutathione 132 glycans 240 glyceraldehyde 3-phosphate 168 glycolipids 192 glycomics 239–259 – bioinformatics 247 – cell, tissue, and metabolic labeling e 246–247 E-DNA sensors, collisional mechanism–based – future perspectives 259 313–324 – DNA binding proteins detection 319–321 – mass spectrometry 245–246 – microarrays 246 – features 313–314 – new insights 247–259 – future perspectives 321–323 – – acceptor speciﬁcities of glycosyltransferase – signaling mechanism 315–319 R, identiﬁcation of 257–259 efﬂux pump effect 108 – – microwave-assisted glycosylation electron crystallography 6 247–251 electron paramagnetic resonance (EPR) 15, – – novel branched thiooligosaccharides, 269 chemoenzymatic synthesis of electrospray ionization mass spectrometry 251–256 (ESI-MS) 15 – state of the art 239–247 emulsion polymerization 120–121 – – carbohydrate-based drugs 239–240 enantiomers 294 – – carbohydrate synthesis 240–245 endoglycosidases 243 glycopeptides enﬂurane 197 – synthesis enhanced green ﬂuorescent protein – – microwave-assisted glycosylation (EGFP) 128 247–251 enzymatic synthesis 243–244 glycoprotein synthesis 244–245 – glycosidases 243 glycosyl phosphatidylinositol 68 – glycosyltransferases 243–244 glycosylamino acids, synthesis 248 – glycosynthases 243 glycosylation 59 epimerization 67 glycosyltransferases 243 erythropoietin 240 grazing-incidence X-ray diffraction ester bonds 130–132 (GIXD) 204 exoglycosidases 243 green ﬂuorescent protein (GFP) 128 expressed protein ligation (EPL) 244

Index heparin 240 heteronuclear single quantum coherence (HSQC) spectra 16 hexamethylphosphorus triamide (HMPTL) 67 HIF-1 226 HIF-1α 226–227 high-throughput screening (HTS) 167 HIV – lifecycle of 219 – structure of 220 HIV budding, cyclic peptide inhibitors of 224–226 homochirality 293 hot spot 215 Huisgen cycloaddition 75 human cervical cancer (HeLa) 128 human protein double minute 2 (HDM2) 217 human serum albumin (HSA) 103 Huntington’s disease 13 hydrophobic effect 193 hydroxylation 59

i iminodiacetic acid (IDA) 17 2-imino-2-methoxyethyl reagents (IME) 63 immunoglobulin antigen 304 inﬂuenza virus 230–232 inosine monophosphate (IMP) 222 interleukin-3 (IL-3) 228 – signaling pathways in asthma 229 interleukin-4 (IL-4) 228 – signaling pathways in asthma 229 intracellular sensing 126–127 intrinsic factor 95–97 iodoacetamides 65 isocyanates 62 isoprene rule 169 isoprenoids 168–169 isothiocyanates 62

j jun 217 jun activation domain-binding protein-1 (JAB1) 10 juxtanuclear compartment 12

k ketones – oxime ligation 78–79 – site-speciﬁc modiﬁcation

79

l label-free electrochemical biosensors 313 laccase, trinuclear copper models for 40–41 lamellar phase 193 large unilamellar vesicles (LUVs) 199 laser-induced acoustic desorption (LIAD) 159 latex beads 119 Lawesson’s reagent 66 lectins 246 Leloir glycosyltransferase 251 – substrate 257 LIAD, See laser-induced acoustic desorption (LIAD) lindane 197 lipoplexes 192 liposomes 117–118, 199 liquid-crystalline phase 194 loss-of-function mechanism 9 low-molecular-weight heparins (LMWHs) 240 lysine 59 – iridium-catalyzed reductive alkylation 74 – modiﬁcation 62–64 – – activated esters 62 – – IME reagents 63–64 – – isocyanates 62 – – isothiocyanates 62 – – reductive alkylation 62–63 – – strategies 63 lysosomes 10

m M.HhaI methyltransferase (M.HhaI) 319 magainin 197 malathion 197 maleimides 66 mammal proteins 305 maraviroc (UK-427 857) 219 mass spectrometry 245–246 matrix-assisted laser desorption and ionization (MALDI) 150 melanins 270 – synthesis of 270 membrane ﬂuidity 195 membrane model systems 199–200 metallothionein (MT) 7 metal-mediated protein modiﬁcation 71–80 – iridium-catalyzed reductive alkylation of lysine 74 – at natural residues 72–74 – – palladium-catalyzed allylation of tyrosine 73

331

332

Index metal-mediated protein modiﬁcation (contd.) – – reductive desulfurization of cysteine 72 – – rhodium carbenoid alkylation of tryptophan 73 – oleﬁn metathesis at s-allyl cysteine 77–78 – of unnatural residues 74–77 – – alkynes, cu-catalyzed [3 + 2] cycloaddition of 74 – – azides, cu-catalyzed [3 + 2] cycloaddition of 74 – – palladium-catalyzed carbon–carbon bond formation 76–77 methane, hydroxylation 38 methionine 71 methionine(Met)-rich motif 6 methylcobalamin 97 – dependent reactions 99 methylene blue 319 methylerythritol (ME)-binding pocket 170 methyl Gilman reagents 43 methylococcus capsulatus 36–37 – pMMO crystal structure 37 microarrays 246 microbeads 200 microsphere-mediated cellular delivery 117–135 – bioconjugation 133–135 – – streptavidin–biotin 134–135 – biodegradable microspheres 119–120 – biostable microspheres 120–121 – cleavable linkers 130–133 – – disulﬁde bonds 132–133 – – ester bonds 130–132 – delivery devices 117–119 – – cell-penetrating peptides 118 – – dendrimers 118 – – liposomes 117–118 – – nanomaterials 118–119 – noncleavable link 126–130 – – intracellular sensing 126–127 – – siRNA delivery 127–130 – solid-phase chemistry and 121–125 – – coupling agents 124–125 – – dual functionality 122–124 – – fmoc chemistry 121–122 – – preparation 121 microsphere spacers 123 model copper systems, O2 reactivity at 267–286 – biomimetic Cu complexes 272–285 – – aromatic rings, hydroxylation of 278–285 – – copper dioxygen adducts 272–275

– – ligand architecture 275–278 – – O2 binding and activation 272–285 – copper proteins involved 268–271 – – hemocyanin 269–270 – – tyrosinase 270–271 – O2 activation – – model systems and 267–268 – properties of 269 – use of 267 modiﬁed proteins, See chemical protein modiﬁcation molecular chirality 293–295 molecular dynamics 297 MOLOC 170, 176, 179 monoubiquitination 10 Monte–Carlo simulations 144 multidrug resistance (MDR) 198 multifunctionalization 123–125 m-xylyl 279 mycobacterium tuberculosis 168 myristic acid 192

n nanomaterials 118–119 nanomaterials-supported lipid bilayers (nanoSLBs) 200 native chemical ligation (NCL) 67–69, 244 – limitation of 69 neuraminic acid 240 neurodegenerative disorders – biological implications 21–23 – – phospholipids 22–23 – – redox state of cellular copper 21–22 – copper ion, role of 3–24 – – brain, metal ions in 4 – – brain copper homeostasis 5–8 – – brain copper and 8–9 – ubiquitin system, role of 3–24 – – failure of 13–15 – – metal ions, interaction with 15–20 – – protein degradation 9–12 neuronal copper homeostasis 6 N-hydroxy succinimide (NHS) esters 62 nitrosylcobalamin 102 N-methylation 178 N-methyl d-aspartate (NMDA) 7 non-Leloir glycosyltransferase 251 – substrate 257 nonmevalonate pathway 168–169 nonsteroidal anti-inﬂammatory drugs (NSAIDs) 201–206 – drug fundamental physical–chemical studies 202–203 – membrane dynamic studies 203–205

Index – membrane structural studies 203–205 normal mode analysis (NMA) 297 novel branched thiooligosaccharides 251–256 nucleobases 147–148

o oleﬁn metathesis 77–78 oleic acid 192 one-pot synthesis 241–242 optimer 242 organocopper 42 organocuprate(I) catalysis 42–44 – C–C bond formation in – – acetylene carbocupration 43 – – conjugate addition to α-enones 42–43 – – SN2 alkylations 43–44 – – SN2 alkylations 43–44 ortho-hydroxylation of 280–285 orthorhombic ZnII -ubiquitin crystals 20 oseltamivir 240 oxidation 59 oxime ligation 78–79

p palladium-catalyzed allylation 73 palmitic acid 192 parathion 197 Parkinson’s disease, causes 14 particulate methano monooxygenase (pMMO) 36–38 peptide ligation 244 peptidylglycine α-hydroxylating monooxygenase (PHM) 38 perivacuolar compartment 12 pernicious anemia 95 P-glycoprotein (PgP) 198 phospholipids 192 – monolayers 199–200 – polar groups of 192 photolysis 102 p-iodophenylalanine 76 piperidinyl ring 178 plasma membrane 191 plasmodium falciparum 168 poly-l-proline 296 poly-l-proline II 296 polymer colloid 119 polypeptidic chain 270 polyprenylation 59 polystyrene microspheres 121 polyubiquitination process 17 posttranslational modiﬁcation 59–60 Prion diseases 305

Prion protein (PrP) 7, 305 proof-of-principle approach 232 propionibacterium shermanii 95 protein degradation, ubiquitin, role of 9–12 protein–protein interactions 215–232 – biosynthesized libraries – – AICAR transformylase activity 222–224 – – high-throughput screening of 220–226 – – HIV budding, cyclic peptide inhibitors of 224–226 – chemical libraries – – high-throughput screening 216–220 – future direction 226–232 – – asthma 228–229 – – networks of inﬂuenza virus 230–232 – – tumor hypoxia response network 226–228 protoﬁbrils 13 proton-coupled electron transfer mechanism (PCET) 282 pseudomonas denitriﬁcans 95 pseudo-π sandwich 173 purine synthesis 222–223 pyrazolyl 178 pyruvate 168

r radiation damage – to DNA 143–162 – – chemical bonds, breaking of 145–147 – – DEA studies on model compounds 148–161 – – low-energy electrons role 143–148 – – nucleobases 147–148 Raman spectrum 269 reductive desulfurization 72 reductive slkylation 63 remodeling 228 resonance 145 resoruﬁn 131 reverse two-hybrid system (RTHS) 221 – heterodimeric 222 rhodium carbenoid alkylation 73 ribose moiety 101–102 ripple phase 194 RNA interference (RNAi) pathway 127 RNA-induced silencing complex (RISC) 127 root mean square deviations (RMSDs) 305 55% rule 179–180

s 26S proteasome SAXS 204

10

333

334

Index scattering methods 204 shape resonance 155 β sheets 296 Shilov system 45 sialic acid 230 SICLOPPS mechanism 221 signal transducer and activator of transcription factor 6 (STAT-6) 228 single strand breaks (SSBs) 144 small interfering RNA (siRNA) 127 small unilamellar vesicles (SUVs) 199 sodium cyanoborohydride 62 solid-phase chemistry 121–125 solid-phase synthesis 242–243 soluble methane monooxygenase (sMMO) 36 solvent polarity 18–19 Sonogashira coupling 76 sonolysis 102 spin-forbidden process 267 spray-drying 120 Staudinger ligations 79 – site-selective modiﬁcation 80 steap proteins 6 stearic acid 192 sterols 192 stoichiometric systems 47–51 streptavidin 134–135 STRIDE algorithm 297 structural analysis 184 structure-based design 167 substantia nigra 5 substrate microarrays 257–259 superoxide dismutase (SOD) 7 supported phospholipid bilayers (SPBs) 200 Suzuki–Miyaura cross-coupling 77 α-synuclein (α-syn) 9

t tartaric acid 293 TATA-box binding protein (TBP) 319 tetraacetyl-d-ribose 155–159 tetrahydrothiophenyl ring 178 tetrazine ligation 81 tetrazole-containing proteins, selective modiﬁcation of 80–81 thiooligosaccharides 254–256 toluene 248 transcobalamin I 96 trans-Golgi network (TGN) 7 translational movement 195 triazamacrocyclic ligand 50

triﬂuoroperazine 197 trinuclear copper models 40–41 tryptophan modiﬁcation, rhodium carbenoid alkylation 73 tumor hypoxia response network 226–228 β turns 296 tyr 33–34 tyrosinase 268, 270–271 – function of 270 – structure of 270 tyrosine, palladium-catalyzed allylation 73

u ubiquitin 3 – metal ions, interaction with 15–20 – – CuII-binding and solvent polarity, cooperativity between 18–19 – – CuII-induced self-oligomerization 17–18 – – polyubiquitination process, possible implications for 17 – – thermal stability 15 – neurodegenerative disorders, failure in 13–15 – phospholipids and 22–23 – protein degradation 9–12 ubiquitination 10 ubiquitin-proteasome system (UPS) 10, 11 Ullmann chemistry 41

v vanadium 102 vertical attachment energy (VAE) 146 vinyl sulfones 66 virtual screening 167 vitamin B12 95–109 – derivatives 99–108 – – β-axial position 102–108 – – b-, d-, e-cobalamin derivatives 99–101 – – molecular structure 100 – – ribose moiety, modiﬁcations on 101–102 – – structure 99 – discovery of 95 – metabolism of 97–99 – – adenosylcobalamin-dependent reactions 97–99 – – methylcobalamin-dependent reactions 99 – transport mechanism 96–97

w water-soluble inhibitors – design of 182–183

Index – – enzyme assays of inhibitors WAXS 204 Wilson disease 5

183

– water-soluble inhibitors, design of 182–183 – – enzyme assays of inhibitors 183

x

z

X-ray cocrystal structure analysis 182–184

zanamivir

240

335

Ideas in Chemistry and Molecular Sciences: Where Chemistry Meets Life

Recommend Documents