Edited by€¦ · Protein Engineering Handbook 2 Volumes 2008 Hardcover ISBN: 978-3-527-31850-6 Robinson, A. S. (Ed.) Production of Membrane Proteins Strategies for Expression and

Edited by

Stefan Lutz and

Uwe T. Bornscheuer

Protein Engineering Handbook

Related Titles

Lutz, S., Bornscheuer, U. T. (Eds.)

Protein Engineering Handbook2 Volumes

2008

Hardcover

ISBN: 978-3-527-31850-6

Robinson, A. S. (Ed.)

Production of Membrane ProteinsStrategies for Expression and Isolation

2011

Hardcover

ISBN: 978-3-527-32729-4

Hughes, A. B. (Ed.)

Amino Acids, Peptides and Proteins in Organic ChemistryVolume 4 - Protection Reactions, Medicinal Chemistry, Combinatorial SynthesisSeries: Amino Acids, Peptides and Proteins in Organic Chemistry

2011

Hardcover

ISBN: 978-3-527-32103-2

Jensen, K. J. (Ed.)

Peptide and Protein Design for Biopharmaceutical Applications

2009

Hardcover

ISBN: 978-0-470-31961-1

Walsh, G. (Ed.)

Post-translational Modification of Protein Biopharmaceuticals

2009

Hardcover

ISBN: 978-3-527-32074-5

Protein Engineering Handbook

Volume 3

Edited by Stefan Lutz and Uwe T. Bornscheuer

The Editors

Prof. Dr. Stefan LutzEmory UniversityDept. of Chemistry1515 Dickey DriveAtlanta GA 30322USA

Prof. Dr. Uwe T. BornscheuerUniversity of GreifswaldInstitute of BiochemistryFelix-Hausdorff-Str. 417487 GreifswaldGermany

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.

Library of Congress Card No.: applied for

British Library Cataloguing-in-Publication DataA catalogue record for this book is available from the British Library.

Bibliographic information published by the Deutsche NationalbibliothekThe Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.

© 2013 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany

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.

Print ISBN: 978-3-527-33123-9ePDF ISBN: 978-3-527-66699-7ePub ISBN: 978-3-527-66698-0mobi ISBN: 978-3-527-66697-3

Cover Design Adam-Design, WeinheimTypesetting Toppan Best-set Premedia Limited, Hong KongPrinting and Binding

Printed inPrinted on acid-free paper

Contents

Preface XV ListofContributors XVII

1 DirigentEffectsinBiocatalysis 1 BettinaM.Nestl,BerndA.Nebel,andBernhardHauer1.1 Introduction 11.2 Dirigent Proteins 31.3 Solvents and Unconventional Reaction Media 41.3.1 Ionic Liquids 71.3.2 Microemulsions and Reversed Micelles Systems 101.4 Structure and Folding 121.5 Structured and Unstructured Domains 141.6 Isozymes, Moonlighting Proteins, and Promiscuity: Supertalented

Enzymes 191.7 Conclusions 22 Acknowledgment 23 References 23

2 ProteinEngineeringGuidedbyNaturalDiversity 29 JamesT.Kratzer,MeganF.Cole,andEricA.Gaucher2.1 Approaches 292.1.1 Ancestral Sequence Reconstruction (ASR) 302.1.2 Ancestral Mutation Method 312.1.3 Reconstructing Evolutionary Adaptive Paths (REAP) 322.2 Protocols 342.2.1 Practical Steps to Using ASR 342.2.2 Reconstructing Evolutionary Adaptive Paths: A Focused Application

of ASR 362.3 Future Directions 382.3.1 Industrial Applications 402.3.2 Biomedical 412.3.3 Drug Discovery 412.3.4 Paleobiology 42

V

VI Contents

2.3.5 Synthetic Biology 432.3.6 Experimental Validation of ASR 432.4 Conclusions 44 References 44

3 ProteinEngineeringUsingEukaryoticExpressionSystems 47 MartinaGeierandAntonGlieder3.1 Introduction 473.2 Eukaryotic Expression Systems 483.2.1 Yeast Expression Platforms 483.2.1.1 Saccharomyces cerevisiae 483.2.1.2 Pichia pastoris 513.2.1.3 Pichia angusta 543.2.1.4 Alternative Yeasts 553.2.2 Filamentous Fungi 563.2.3 Insect Cells 583.2.4 Mammalian Cell Cultures 593.2.5 Transgenic Animals and Plants 613.2.6 Cell-Free Expression Systems 613.3 Conclusions 63 References 65

4 ProteinEngineeringinMicrodroplets 73 YolandaSchaerli,BalintKintses,andFlorianHollfelder4.1 Introduction 734.2 Droplet Formats 754.2.1 “Bulk” Emulsions 754.2.1.1 Catalytic Selections Involving DNA Substrates 764.2.1.2 Using the Droplet Compartment to Form a Permanent

Genotype-Phenotype Linkage for Selections of Binders 774.2.2 Double “Bulk” Emulsions 784.2.3 Microfluidic Droplets 794.3 Perspectives 83 Acknowledgments 84 References 84

5 FoldingandDynamicsofEngineeredProteins 89 MichelleE.McCullyandValerieDaggett5.1 Introduction 895.2 Proof-of-Principle Protein Designs 905.2.1 FSD-1, a Heterogeneous Native State and Complicated Folding

Pathway 915.2.2 α3D, a Dynamic Core Leads to Fast Folding and Thermal Stability 945.2.3 Three-Helix Bundle Thermostabilized Proteins 965.2.4 Top7, a Novel Fold Topology 97

Contents VII

5.2.5 Other Rosetta Designs 1005.3 Proteins Designed for Function 1025.3.1 Ligands 1035.3.1.1 Metal-Binding Four-Helix Bundles, the Effectiveness of Negative

Design 1035.3.1.2 Peptide Binding 1055.3.2 Enzymes 1065.3.2.1 Retro-Aldol Enzyme, Accommodating a Two-Step Reaction 1065.3.2.2 Kemp Elimination Enzyme, Rigid Active Site Geometry

Promotes Catalysis 1085.4 Conclusions and Outlook 110 Acknowledgments 111 References 112

6 EngineeringProteinStability 115 CiaránÓ’Fágáin6.1 Introduction 1156.2 Power and Scope of Protein Engineering to Enhance Stability 1166.2.1 Thermal Stabilizations 1166.2.1.1 Potential Therapeutics: Rational Design with Computational

Support 1166.2.1.2 Analytical Tools: Green Fluorescent Protein and Luciferase 1286.2.1.3 “Stiffening” a Protein by Gly-to-Pro Replacement: Methyl

Parathion Hydrolase 1286.2.2 Thermal Is Not the Only Stability: Oxidative and Other Chemical

Stabilities 1296.2.2.1 Oxidative Stability 1296.2.2.2 Stabilization against Aldehydes and Solvents 1306.2.2.3 Alkaline Tolerance 1316.3 Measurement of a Protein’s Kinetic Stability 1326.3.1 Materials and General Hints 1326.3.2 Thermal Stability 1326.3.2.1 Thermal Profile 1326.3.2.2 Thermal Inactivation 1336.3.3 Measurement of Oxidative Stability 1346.3.4 Stability Analysis and Accelerated Degradation Testing 1356.3.4.1 Set-Up 1366.3.4.2 Analysis of Results 1376.4 Developments in Protein Stabilization 137 References 139

7 EnzymesfromThermophilicOrganisms 145 TamotsuKanaiandHaruyukiAtomi7.1 Introduction 1457.2 Hyperthermophiles 146

VIII Contents

7.3 Enzymes from Thermophiles and Their Reactions 1467.4 Production of Proteins from (Hyper)Thermophiles 1487.5 Protein Engineering of Thermophilic Proteins 1547.6 Cell Engineering in Hyperthermophiles 1567.7 Future Perspectives 157 References 157

8 EnzymeEngineeringbyCofactorRedesign 163 MalgorzataM.Kopacz,Frank.Hollmann,andMarcoW.Fraaije8.1 Introduction 1638.2 Natural Cofactors: Types, Occurrence, and Chemistry 1648.3 Inorganic Cofactors 1658.4 Organic Cofactors 1688.5 Redox Cofactors 1698.5.1 Nicotinamide Cofactor Engineering 1708.5.2 Heme Cofactor Engineering 1738.5.2.1 Reconstitution of Myoglobin 1748.5.2.2 Artificial Metalloproteins Based on Serum Albumins 1758.5.3 Flavin Cofactor Engineering 1768.6 Concluding Remarks 180 References 181

9 BiocatalystIdentificationbyAnaerobicHigh-ThroughputScreeningofEnzymeLibrariesandAnaerobicMicroorganisms 193

HelenS.ToogoodandNigelS.Scrutton9.1 Introduction 1939.2 Oxygen-Sensitive Biocatalysts 1949.2.1 Flavoproteins 1949.2.2 Iron-Sulfur-Containing Proteins 1959.2.3 Other Causes of Oxygen Sensitivity 1979.3 Biocatalytic Potential of Oxygen-Sensitive Enzymes and

Microorganisms 1989.3.1 Old Yellow Enzymes (OYEs) 1989.3.2 Enoate Reductases 2009.3.3 Other Enzymes 2029.3.4 Whole-Cell Anaerobic Fermentations 2029.4 Anaerobic High-Throughput Screening 2039.4.1 Semi-Anaerobic Screening Protocols 2049.4.2 Anaerobic Robotic High-Throughput Screening 2059.4.2.1 Purified Enzyme versus Whole-Cell Extracts 2079.4.2.2 Indirect Kinetic Screening versus Direct Product Determination 2089.4.3 Potential Extensions of Robotic Anaerobic High-Throughput

Screening 2099.5 Conclusions and Outlook 210 References 210

Contents IX

10 OrganometallicChemistryinProteinScaffolds 215 YvonneM.Wilson,MarcDürrenberger,andThomasR.Ward10.1 Introduction 21510.1.1 Concept 21510.1.2 Considerations for Designing an Artificial Metalloenzyme 21610.1.2.1 Organometallic Complex 21610.1.2.2 Biomolecular Scaffold 21810.1.2.3 Anchoring Strategy 21910.1.2.4 Advantages and Disadvantages of the Different Anchoring

Modes 22110.1.2.5 Spacer 22210.1.3 Other Key Developments in the Field 22310.1.4 Why Develop Artificial Metalloenzymes? 22310.2 Protocol/Practical Considerations 22610.2.1 Protein Scaffold 22610.2.1.1 Determination of Free Binding Sites 22610.2.2 Organometallic Catalyst 22810.2.2.1 Synthesis of [Cp*Ir(biot-p-L)Cl] 22910.2.2.2 N′-(4-Biotinamidophenylsulfonyl)-Ethylenediamine TFA Salt

23010.2.3 Combination of Biotinylated Metal Catalyst and Streptavidin

Host 23110.2.3.1 Binding Affinity of the Biotinylated Complex to Streptavidin 23110.2.4 Catalysis 23210.2.4.1 Catalysis Controls 23210.3 Goals 23410.3.1 Rate Acceleration 23410.3.2 High-Throughput Screening 23410.3.2.1 Considerations for Screening of Artificial Metalloenzymes 23510.3.3 Expansion of Substrate Scope 23610.3.4 Upscaling 23610.3.5 Potential Applications 23710.4 Summary 237 Acknowledgments 237 References 238

11 EngineeringProteaseSpecificity 243 PhilipN.Bryan11.1 Introduction 24311.1.1 Overview 24311.1.2 Some Basic Points 24411.1.2.1 Mechanism for a Serine Protease 24411.1.2.2 Measuring Specificity 24411.1.2.3 Binding Interactions 24511.1.3 Nature versus Researcher 247

X Contents

11.1.3.1 P1 Specificity of Chymotrypsin-like Proteases 24711.1.3.2 The S1 Site of Subtilisin 24711.1.3.3 The S4 Site of Subtilisin 25011.1.3.4 Other Subsites in Subtilisin 25011.1.3.5 Kinetic Coupling and Specificity 25111.2 Protocol and Practical Considerations 25111.2.1 Remove and Regenerate 25111.2.2 Engineering Highly Stable and Independently Folding Subtilisins 25211.2.3 Engineering of P4 Pocket to Increase Substrate Specificity 25311.2.4 Destroying the Active Site in Order to Save It 25411.2.5 Identifying a Cognate Sequence for Anion-Triggered Proteases Using

the Subtilisin Prodomain 25511.2.6 Tunable Chemistry and Specificity 25711.2.7 Purification Proteases Based on Prodomain–Subtilisin Interactions

and Triggered Catalysis 25811.2.8 Design of a Mechanism-Based Selection System 25911.2.8.1 Step 1: Ternary Complex Formation 25911.2.8.2 Step 2: Acylation 26311.2.8.3 Steps 3 and 4: Deacylation and Product Release 26511.2.9 Evolving Protease Specificity Regulated with Anion Cofactors by

Phage Display 26611.2.9.1 Construction and Testing of Subtilisin Phage 26611.2.9.2 Random Mutagenesis and Transformation 26711.2.9.3 Selection of Anions 26711.2.9.4 Evolving the Anion Site 26711.2.9.5 Catch-and-Release Phage Display 26711.2.9.6 Conclusions 26911.2.10 Evolving New Specificities at P4 26911.3 Concepts, Challenges, and Visions on Future Developments 27011.3.1 Design Challenges 27011.3.2 Challenges in Directed Evolution 27111.3.2.1 One Must Go Deep into Sequence Space 27111.3.2.2 Methods Which Maximize Substrate Binding Affinity Are Not

Productive 27211.3.2.3 The Desired Protease May Be Toxic to Cells 27211.3.3 The Quest for Restriction Proteases 27211.3.3.1 Not All Substrate Sequences Are Created Equal 27311.3.4 Final Thoughts: Gilded or Golden? 273 Acknowledgments 274 References 274

12 PolymeraseEngineering:FromPCRandSequencingtoSyntheticBiology 279

VitorB.Pinheiro,JenniferL.Ong,andPhilippHolliger12.1 Introduction 27912.2 PCR 281

Contents XI

12.3 Sequencing 28112.3.1 First-Generation Sequencing 28212.3.2 Next-Generation Sequencing Technologies 28412.4 Polymerase Engineering Strategies 28812.5 Synthetic Informational Polymers 291 References 295

13 EngineeringGlycosyltransferases 303 JohnMcArthurandGavinJ.Williams13.1 Introduction to Glycosyltransferases 30313.2 Glycosyltransferase Sequence, Structure, and Mechanism 30413.3 Examples of Glycosyltransferase Engineering 30713.3.1 Chimeragenesis and Rational Design 30713.3.2 Directed Evolution 31013.3.2.1 Fluorescence-Based Screening 31113.3.2.2 Reverse Glycosylation Reactions 31213.3.2.3 ELISA-Based Screens 31313.3.2.4 pH Indicator Assays 31413.3.2.5 Chemical Complementation 31413.3.2.6 Low-Throughput Assays 31413.4 Practical Considerations for Screening Glycosyltransferases 31513.4.1 Enzyme Expression and Choice of Expression Vector 31513.4.2 Provision of Acceptor and NDP-donor Substrate 31513.4.3 General Considerations for Microplate-Based Screens 31713.4.4 Promiscuity, Proficiency, and Specificity 31713.5 Future Directions and Outlook 318 References 319

14 ProteinEngineeringofCytochromeP450Monooxygenases 327 KatjaKoschorreck,ClemensJ.vonBühler,SebastianSchulz,

andVladaB.Urlacher14.1 Cytochrome P450 Monooxygenases 32714.1.1 Introduction 32714.1.2 Catalytic Cycle of Cytochrome P450 Monooxygenases 32814.1.3 Redox Partner Proteins 32914.2 Engineering of P450 Monooxygenases 33014.2.1 Molecular Background for P450 Engineering 33014.2.2 Altering Substrate Selectivity and Improving Enzyme Activity 33214.2.2.1 Rational and Semi-Rational Design 33214.2.2.2 Directed Evolution and Its Combination with Computational

Design 33614.2.2.3 Decoy Molecules 33814.2.3 Improving Solvent and Temperature Stability of P450

Monooxygenases 34014.2.3.1 Solvent Stability 34114.2.3.2 Thermostability 342

XII Contents

14.2.4 Improving Recombinant Expression and Solubility of P450 Monooxygenases 343

14.2.4.1 N-Terminal Modifications 34414.2.4.2 Modifications within the F-G Loop 34614.2.4.3 Improving Expression by Rational Protein Design and Directed

Evolution 34814.2.5 Engineering the Electron Transport Chain and Cofactors of

P450s 34914.2.5.1 Genetic Fusion of Proteins 34914.2.5.2 Enzymatic Fusion and Self-Assembling Oligomers 35214.3 Conclusions 354 References 355

15 ProgressandChallengesinComputationalProteinDesign 363 Yih-EnAndrewBan,DanielaRöthlisberger-Grabs,EricA.Althoff,

andAlexandreZanghellini15.1 Introduction 36315.2 The Technique of Computational Protein Design 36315.2.1 Principles of Protein Design 36315.2.2 A Brief Review of Force-Fields for CPD 36415.2.3 Optimization Algorithms for Fixed-Backbone Protein Design (P1′) 36815.3 Protein Core Redesign, Structural Alterations, and

Thermostabilization 37115.3.1 Protein Core Redesign and de novo Fold Design 37115.3.2 Computational Alteration of Protein Folds 37315.3.2.1 Loop Grafting 37415.3.2.2 de novo Loop Design 37515.3.2.3 Fold Switching 37615.3.2.4 Fold Alteration: Looking Ahead 37715.3.2.5 Computational Optimization of the Thermostability of Proteins 37715.4 Computational Enzyme Design 38015.4.1 de novo Enzyme Design 38015.4.1.1 Initial Proofs-of-Concept 38015.4.1.2 Review of Recent Developments 38215.4.2 Computational Redesign of the Substrate Specificity of Enzymes 38315.4.2.1 Fixed-Backbone and Flexible-Backbone Substrate Specificity

Switches 38315.4.2.2 Limitations and Feedback Obtained from Experimental Optimization

Attempts 38515.4.3 Frontiers in Computational Enzyme Design 38615.5 Computational Protein–Protein Interface Design 38815.5.1 Natural Protein–Protein Interfaces Redesign 38915.5.2 Two-Sided de novo Design of Protein Interfaces 39015.5.3 One-Sided de novo Design of Protein Interfaces 39215.5.4 Frontiers in Protein–Protein Interaction Design 393

Contents XIII

15.6 Computational Redesign of DNA Binding and Specificity 39415.7 Conclusions 396 References 396

16 SimulationofEnzymesinOrganicSolvents 407 TobiasKulschewskiandJürgenPleiss16.1 Enzymes in Organic Solvents 40716.2 Molecular Dynamics Simulations of Proteins and Solvents 40816.3 The Role of the Solvent 41016.4 Simulation of Protein Structure and Flexibility 41116.5 Simulation of Catalytic Activity and Enantioselectivity 41316.6 Simulation of Solvent-Induced Conformational Transitions 41416.7 Challenges 41516.8 The Future of Biocatalyst Design 416 References 417

17 EngineeringofProteinTunnels:TheKeyhole–Lock–KeyModelforCatalysisbyEnzymeswithBuriedActiveSites 421

ZbynekProkop,ArturGora,JanBrezovsky,RadkaChaloupkova,VeronikaStepankova,andJiriDamborsky

17.1 Traditional Models of Enzymatic Catalysis 42117.2 Definition of the Keyhole–Lock–Key Model 42217.3 Robustness and Applicability of the Keyhole–Lock–Key Model 42417.3.1 Enzymes with One Tunnel Connecting a Buried Active Site to the

Protein Surface 42417.3.2 Enzymes with More than One Tunnel Connecting a Buried Active Site

to the Protein Surface 43317.3.3 Enzymes with One Tunnel Between Two Distinct Active Sites 43617.4 Evolutionary and Functional Implications of the Keyhole–Lock–Key

Model 43717.5 Engineering Implications of the Keyhole–Lock–Key Model 43817.5.1 Engineering Activity 44217.5.2 Engineering Specificity 44317.5.3 Engineering Stereoselectivity 44317.5.4 Engineering Stability 44317.6 Software Tools for the Rational Engineering of Keyholes 44417.6.1 Analysis of Tunnels in a Single Protein Structure 44517.6.2 Analysis of Tunnels in the Ensemble of Protein Structures 44517.6.3 Analysis of Tunnels in the Ensemble of Protein–Ligand

Complexes 44717.7 Case Studies with Haloalkane Dehalogenases 44817.8 Conclusions 450 References 452

Index 465

XV

Almost four years have passed since the first two volumes of the Protein Engineer-ing Handbook were published. In this time, the development of novel, effective, and sustainable catalysts through enzyme engineering has continued. Moreover, innovative new strategies and improvements to existing methods have accelerated the discovery process and have yielded biocatalysts with impressive performance enhancements. Our desire to capture many of these recent advances, as well as to cover topics not included in the previous two volumes, has inspired the creation of this third volume.

An introduction to the seemingly endless opportunities and challenges facing today’s enzyme engineers is presented by Hauer and colleagues who, in Chapter 1, highlight the importance of third-party effects such as dirigent proteins and solvent environment on enzyme performance. It also emphasizes the relevance of protein folding and dynamics, a topic that is discussed in more detail in Chapter 5 by Daggett and McCully. Separately, a rigorous review of methods for assessing and enhancing protein stability through protein engineering is contributed by Ó’Fágáin in Chapter 6.

On the technology side, many new methods for tailoring proteins in the labora-tory concentrate on strategies that allow for the creation of smaller libraries with a high functional content. Protein engineering based on consensus sequence infor-mation is one such approach, facilitated by the rapid growth of gene and protein databases. In Chapter 2, Gaucher and colleagues describe new ways of utilizing and expanding upon such natural diversity for protein engineering through ancestral sequence reconstruction. At the same time, more effective experimental tech-niques for assessing largely combinatorial libraries are being developed. Beyond plate-based and cell-sorting assays, recent innovations in microfluidics have dem-onstrated the versatility of this technology and opened new avenues for scientists to screen large number of enzyme variants, as reported by Hollfelder et al. in Chapter 4. In addition to a smarter and faster library evaluation, protein engineering has expanded into more complex host systems. Eukaryotic expression systems have been improved, as discussed by Glieder and Geier in Chapter 3, while the frontier for host strains has been expanded to include extremophilic microorganisms, as described by Atomi and Kanai in Chapter 7. Separately, Scrutton and Toogood, in Chapter 9, have shown that switching to an anaerobic environment can offer

Preface

XVI Preface

significant advantages for the functional evaluation of protein libraries. Impor-tantly, protein engineering must not be limited to modifications of the polypeptide sequences, as noted by Fraaije et al. and Ward et al. in Chapters 8 and 10 which describe, respectively, the tuning of catalytic activity through the modification of native enzyme cofactors and artificial organometallics.

Rather than focusing exclusively on the methods for protein engineering, four chapters have been added on the tailoring of enzymes from a perspective of bio-catalyst category or family. Proteases were among the first enzymes to be targeted by protein engineers, due to their importance in many commercial applications. Beyond these early studies, Chapter 11 by Bryan and coworkers captures some of the more recent strategies for controlling and directing the potentially destructive power of these hydrolases. Next, in Chapter 12, Holliger and colleagues report on elegant and powerful engineering approaches for customizing DNA polymer-ases. The creation of efficient and accurate polymerases is a key element for new genome sequencing technologies and diagnostics in the twenty-first century. Simi-larly, in Chapter 13, Williams and McArthur summarize the state of the art in solving new synthetic challenges in glycobiology through the engineering of glyco-syltransferases. Finally, in Chapter 14, Urlacher and coworkers have surveyed the engineering of cytochrome P450 monooxygenases, a class of enzymes which has great significance not only to organic synthetic chemistry but also to drug discovery and metabolism.

In complementing and guiding an increasing number of experimental studies, we wish to (re-)emphasize the ever-growing importance of computational tools in protein engineering and design. By using in-silico approaches for enzyme design, Zanghellini and colleagues have reviewed, in Chapter 15, the possibilities and limitations of the versatile Rosetta algorithm in creating novel biocatalysts, while Damborsky and coworkers, in Chapter 17, have developed invaluable new algo-rithms for accessing the impact of protein engineering in or near the active site itself. A quite different – but equally important – challenge emerges from environ-mental effects on biocatalysts, and in Chapter 16 Pleiss and Kulschewski provide an impressive demonstration of the advances in computational tools to accurately predict the effects of nonaqueous solvents.

In closing, we would like to thank all members of the scientific community for their positive feedback and constructive suggestions that ultimately encouraged us to tackle the editing of this third installment of the Protein Engineering Handbook series. Beyond these initial “catalytic” events, however, we are greatly indebted to the many individuals whose invaluable contributions have helped us to assemble the book. Besides thanking all of the authors for their efforts, we would like to acknowledge our colleagues and students at Emory University and the University of Greifswald for their advice, patience, and willingness to review and proof-read many pages. Finally, our special thanks are also extended to the people at Wiley-VCH, namely Dr Frank Weinreich and Dr Heike Nöthe for their editorial assist-ance, as well as Andrea Zschäge for her help during the printing stage of the book.

Atlanta/Greifswald, October 2012 Stefan Lutz and Uwe T. Bornscheuer

XVII

Eric A. AlthoffArzeda Corporation2722 Eastlake Ave E.Suite 150Seattle, WA 98102USA

Haruyuki AtomiKyoto UniversityGraduate School of EngineeringDepartment of Synthetic Chemistry and Biological ChemistryKatsuraNishikyo-kuKyoto 615-8510Japan

and

JSTCRESTSanbanchoChiyoda-kuTokyo 102-0075Japan

Yih-En Andrew BanArzeda Corporation2722 Eastlake Ave E.Suite 150Seattle, WA 98102USA

ListofContributors

Jan BrezovskyMasaryk UniversityLoschmidt LaboratoriesDepartment of Experimental Biology and Centre for Toxic Compounds in the EnvironmentKamenice 5/A13625 00 BrnoCzech Republic

Philip N. BryanUniversity of MarylandInstitute for Bioscience and Biotechnology Researchand Department of Bioengineering9600 Gudelsky DriveRockville, MD 20850USA

Clemens J. von BühlerHeinrich-Heine University DüsseldorfInstitute of BiochemistryUniversitätsstrasse 140225 DüsseldorfGermany

XVIII ListofContributors

Radka ChaloupkovaMasaryk UniversityLoschmidt LaboratoriesDepartment of Experimental Biology and Centre for Toxic Compounds in the EnvironmentKamenice 5/A13625 00 BrnoCzech Republic

Megan F. ColeGeorgia Institute of TechnologySchool of Biology310 Ferst DriveAtlanta, GA 30332USA

Valerie DaggettUniversity of WashingtonDepartment of BioengineeringBox 355013Seattle, WA 98195-5013USA

Jiri DamborskyMasaryk UniversityLoschmidt LaboratoriesDepartment of Experimental Biology and Centre for Toxic Compounds in the EnvironmentKamenice 5/A13625 00 BrnoCzech Republic

and

St Anne’s University Hospital BrnoInternational Centre for Clinical ResearchPekarska 53656 91 BrnoCzech Republic

Marc DürrenbergerUniversity of BaselDepartment of ChemistrySpitalstr. 514056 BaselSwitzerland

Marco W. FraaijeUniversity of GroningenGroningen Biomolecular Sciences and Biotechnology InstituteLaboratory of BiochemistryNijenborgh 49747 AG GroningenThe Netherlands

Eric A. GaucherGeorgia Institute of TechnologySchool of Biology310 Ferst DriveAtlanta, GA 30332USA

and

Georgia Institute of TechnologySchool of Chemistry901 Atlantic DriveAtlanta, GA 30332USA

Martina GeierGraz University of TechnologyInstitute of Molecular BiotechnologyPetersgasse 148010 GrazAustria

Anton GliederAustrian Centre of Industrial Biotechnology (ACIB)Petersgasse 148010 GrazAustria

ListofContributors XIX

Artur GoraMasaryk UniversityLoschmidt LaboratoriesDepartment of Experimental Biology and Centre for Toxic Compounds in the EnvironmentKamenice 5/A13625 00 BrnoCzech Republic

Bernhard HauerUniversität StuttgartInstitut für Technische BiochemieAllmandring 31D-70569 StuttgartGermany

Florian HollfelderUniversity of CambridgeDepartment of Biochemistry80 Tennis Court RoadCambridge CB2 1GAUK

Philipp HolligerMedical Research Council UKLaboratory of Molecular BiologyHills RoadCambridge CB2 0QHUK

Frank. HollmannDelft University of TechnologyDepartment of BiotechnologyJulianalaan 1362628 BL DelftThe Netherlands

Tamotsu KanaiKyoto UniversityGraduate School of EngineeringDepartment of Synthetic Chemistry and Biological ChemistryKatsuraNishikyo-kuKyoto 615-8510Japan

and

JSTCRESTSanbanchoChiyoda-kuTokyo 102-0075Japan

Balint KintsesUniversity of CambridgeDepartment of Biochemistry80 Tennis Court RoadCambridge CB2 1GAUK

Malgorzata M. KopaczUniversity of GroningenGroningen Biomolecular Sciences and Biotechnology InstituteLaboratory of BiochemistryNijenborgh 49747 AG GroningenThe Netherlands

Katja KoschorreckHeinrich-Heine University DüsseldorfInstitute of BiochemistryUniversitätsstrasse 140225 DüsseldorfGermany

XX ListofContributors

James T. KratzerGeorgia Institute of TechnologySchool of Chemistry901 Atlantic DriveAtlanta, GA 30332USA

Tobias KulschewskiUniversity of StuttgartInstitute of Technical BiochemistryAllmandring 3170569 StuttgartGermany

John McArthurNorth Carolina State UniversityDepartment of ChemistryRaleigh, NC 27695-8204USA

Michelle E. McCullyUniversity of WashingtonDepartment of BioengineeringBox 355013Seattle, WA 98195-5013USA

Bernd A. NebelUniversität StuttgartInstitut für Technische BiochemieAllmandring 3170569 StuttgartGermany

Bettina M. NestlUniversität StuttgartInstitut für Technische BiochemieAllmandring 3170569 StuttgartGermany

Ciarán Ó’FágáinDublin City UniversitySchool of Biotechnology and National Centre for Sensor ResearchDublin 9Ireland

Jennifer L. OngNew England Biolabs240 County RoadIpswich, MA 019838-2723USA

Vitor B. PinheiroMedical Research Council UKLaboratory of Molecular BiologyHills RoadCambridge CB2 0QHUK

Jürgen PleissUniversity of StuttgartInstitute of Technical BiochemistryAllmandring 3170569 StuttgartGermany

Zbynek ProkopMasaryk UniversityLoschmidt LaboratoriesDepartment of Experimental Biology and Centre for Toxic Compounds in the EnvironmentKamenice 5/A13625 00 BrnoCzech Republic

Daniela Röthlisberger-GrabsArzeda Corporation2722 Eastlake Ave E.Suite 150Seattle, WA 98102USA

ListofContributors XXI

Yolanda SchaerliUniversity of CambridgeDepartment of Biochemistry80 Tennis Court RoadCambridge CB2 1GAUK

Sebastian SchulzHeinrich-Heine University DüsseldorfInstitute of BiochemistryUniversitätsstrasse 140225 DüsseldorfGermany

Nigel S. ScruttonUniversity of ManchesterManchester Institute of BiotechnologyFaculty of Life Sciences131 Princess StreetManchester M1 7DNUK

Veronika StepankovaMasaryk UniversityLoschmidt LaboratoriesDepartment of Experimental Biology and Centre for Toxic Compounds in the EnvironmentKamenice 5/A13625 00 BrnoCzech Republic

Helen S. ToogoodUniversity of ManchesterManchester Institute of BiotechnologyFaculty of Life Sciences131 Princess StreetManchester M1 7DNUK

Vlada B. UrlacherHeinrich-Heine University DüsseldorfInstitute of BiochemistryUniversitätsstrasse 140225 DüsseldorfGermany

Thomas R. WardUniversity of BaselDepartment of ChemistrySpitalstr. 514056 BaselSwitzerland

Gavin J. WilliamsNorth Carolina State UniversityDepartment of ChemistryRaleigh, NC 27695-8204USA

Yvonne M. WilsonUniversity of BaselDepartment of ChemistrySpitalstr. 514056 BaselSwitzerland

Alexandre ZanghelliniArzeda Corporation2722 Eastlake Ave E.Suite 150Seattle, WA 98102USA

1

DirigentEffectsinBiocatalysisBettinaM.Nestl,BerndA.Nebel,andBernhardHauer

1.1Introduction

Enzymes are capable of accepting a broad range of substrates and are highly selec-tive, manifested as stereoselectivity, positional selectivity, and functional group selectivity. The nature of the enzyme, the predefined selective molecular recogni-tion of the substrate molecule within its active site, guides selectivity by discrimi-nating between substrate enantiomers converting only one enantiomer. Enzyme engineering is a powerful tool and a widely accepted methodology to optimize and influence enzyme properties, such as the overall activity, selectivity, thermo- and storage stability and the stability toward organic solvents, as it encompasses both directed evolution and rational design. Besides the alteration of enzyme function, enzyme engineering is also capable of directly influencing enzyme-catalyzed reac-tions, thus controlling their product formation. A proof of concept was presented by the divergent evolution of the promiscuous sesquiterpene synthase γ-humulene synthase. The rational design of residues in the active site of the γ-humulene synthase has shown to possess an additive influence on protein function and promiscuity. By using this approach, a large number of novel specific sesquiter-pene synthases has been constructed, each producing one or a few end-products via different reaction pathways including new molecules that do not exist in Nature (Figure 1.1) [1]. This instructive example shows the capability to redesign enzyme function by single amino acid substitutions, and to direct the biocatalytic transfor-mation of diverse substrates via the same mechanism.

This chapter attempts to demonstrate that dirigent effects have been described influencing the outcome of enzyme-catalyzed reactions. It consists of several subchapters that could expand the concept of dirigent properties in biocatalysis by exploiting intrinsic effects which have a considerable impact on the catalytic event of enzyme reactions, but also which extend beyond.

First, a brief report is provided of auxiliary or dirigent proteins which play an important role in free-radical coupling in lignin biosynthesis. These naturally evolved proteins capture the oxidized free-radical substrate, providing a scaffold upon which a selective, radical coupling can occur to yield an optically active product.

1

Protein Engineering Handbook: Volume 3, First Edition. Edited by Stefan Lutz and Uwe T. Bornscheuer.© 2013 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2013 by Wiley-VCH Verlag GmbH & Co. KGaA.

2 1 DirigentEffectsinBiocatalysis

The different behaviors of enzymes in organic solvents and unconventional reaction media, and such behaviors toward enzyme catalysis with respect to activity and selectivity, are also discussed. The active site of an enzyme is determined by the presence or absence of solvent molecules which, in addition to an unconven-tional reaction medium, may influence not only the size and shape of the active site but also the dielectric constant (and consequently the pKa values) of the host side chain and electrostatic potential of the site.

In addition, attention is focused on protein structure–function relationships in order to deepen the general understanding of the mechanism, and the folding or motion of proteins of biologically active catalysts. The aim at this point is to

Figure1.1 Construction of seven specific and active γ-humulene synthases that use different reaction pathways to produce sesquiterpenes. Modified from Ref. [1].

OPP

E, E-farnesyl diphosphate

sibirene

E-β-farnesene

Z,E-α-farnesene longifolene β-bisabolene

α-ylangene

γ-humulene α-longipinene

1.2DirigentProteins 3

broaden the current understanding of protein folding, which is important for enzyme catalysis as new protein functions can be obtained from existing ones through mutations that alter the amino acid sequence and, hence, the active site architecture.

The intrinsically disordered proteins form a separate class of proteins with specific sequence compositions and functions. Although such proteins fold to form defined structures upon binding, certain parts remain disordered throughout the process. During recent years, these short disordered segments and their func-tion in ordered proteins have undergone extensive investigation and discussion. Indeed, in the case of a small number of intrinsically disordered proteins catalytic activity has been observed; specific examples for which such activity has been identified are described here.

The subject then switches to isozymes, moonlighting proteins and promiscuous enzymes, and their different biological selectivities. More comprehensive over-views on these mechanisms and on catalytic promiscuity have been produced by Hult and Berglund [2] and Bornscheuer and Kazlauskas [3], and by Stefanie Jonas and Florian Hollfelder in volume 1 of this book series. It is suggested that “super-talented” enzymes, defined as the catalysis of multiple reactions versus the cataly-sis of a single reaction with different substrates, are connected to several conditions (expression, environmental conditions, ligand concentration) and the structural flexibility of the protein.

It should be noted that this chapter does not aim to provide a comprehensive overview on the topics introduced, as detailed information on each subject is avail-able in reviews and articles cited (and in the references therein). Rather, the inten-tion is to present concepts from selected examples that can be put into practice by the reader to help to understand dirigent effects in biocatalysis, and thereby to offer some food for thought concerning the strategies required to engineer enzymes of interest.

1.2DirigentProteins

The discovery of “dirigent proteins” in the mid-1990s, and their abundance in plants, has provided the insight that proteins and enzymes must exist which are able to dictate the stereochemistry of a compound that is synthesized by other proteins. These are enzymes which bind differentially not only various phenylpro-panoid (monolignol)-derived substrates (thus guiding the outcome of their cou-pling yielding in lignin), but also several other compounds such as lignans, flavonolignans, and alkaloids [4]. Hence, specific monolignol (radical)-binding sites have been identified for such proteins, forming the biochemical basis for both regioselective and stereoselective monolignol coupling reactions.

The first dirigent protein was isolated and characterized for the stereoselective bimolecular phenoxy radical coupling in the presence of an oxidase or a one-electron oxidant in 1997 [5]. As a consequence, the term “dirigent” (from the Latin

4 1 DirigentEffectsinBiocatalysis

dirigere, which when translated means to align or to guide) was chosen to label this new, rather small class of proteins for which about 20 reports have been made to date.

One example of an in vitro reaction involving a dirigent protein was the laccase-catalyzed oxidation of coniferyl alcohol, which resulted in the formation of racemic pinoresinol, whereas in the presence of a dirigent protein isolated from Forsythia intermedia, enantiomerically pure (+)-pinoresinol was obtained (Figure 1.2) [6]. Recently, an enantiocomplementary dirigent protein from Arabidopsis thaliana was characterized which mediates the oxidative phenol coupling to the pure (−)-pinor-esinol product [7]. The specificity of the dirigent reaction described above requires two proteins to be localized near to one another, at a concentration that promotes this interaction. Whereas, binding to a specific partner represents one mechanism, another involves the interaction with many partners (multispecificity) that can be important for biological functions. Examples of multispecific, flexible enzymes include the cytochrome P450 monooxygenases, which represent a wide range of different active-site conformations that bind and transform diverse substrates [8].

Unlike dirigent proteins, with their distinct biochemical mechanism and their ability to manage the selective assembly of free radicals formed from monolignol pathways, the establishment of further, more artificial effects appeared to influ-ence the mechanism and the performance of enzyme-catalyzed reactions. This view does not rest solely on experimental evidence, which thus far is incomplete, but instead relies more on the current knowledge of biochemical processes and their reaction conditions, and on considerations of structural, mechanistic and evolutionary implications on the selectivity, specificity and activity of enzymes.

1.3SolventsandUnconventionalReactionMedia

A large number of enzymes show different behaviors in non-aqueous organic media, with effects such as changes in enzyme stability, activity and selectivity being observed by using organic solvents. “Anhydrous” in this context does not mean “no water,” but rather that the amount of water compared to the entire reac-

Figure1.2 Oxidative, enantioselective coupling of coniferyl alcohol results in enantiomerically pure (+)-pinoresinol product formation [6].

OH

OH O

OH

oxidase dirigent protein

O

O

OH

OCH3

OCH3

H3COH3COHO

1.3SolventsandUnconventionalReactionMedia 5

tion volume is low (often <5%). Furthermore, it should be noted that the effects vary significantly with the solvent used. Although, in general there is no thorough understanding of these solvent effects, the general suggestions can be divided into two categories:

• Organic solvents change the enzyme’s conformation and its flexibility, both of which are crucial for enzyme efficiency.

• The use of an organic solvent changes the solubility and desolvation of non-polar substrates and products.

The solubility behavior is often explained as a relationship between enzyme activity and the solvent hydrophobicity log P. Thus, log P is denoted as the ratio of con-centrations of a compound in the two phases of a mixture of two immiscible solvents at equilibrium. In detail, the log P value is the partitioning coefficient of a solvent in the two phase system:

log log( )P c c= solvent solvent/1 2

The use of organic solvents can help to suppress unwanted water-dependent side reactions. Solvent effects can also be explained as competitive inhibition, as dif-ferent types of solvent molecules will have different abilities to bind and block the active site of an enzyme, where there they will act as a competitive inhibitor to the substrate [9, 10]. This ability to affect the properties of enzymes is often summa-rized in the term “medium engineering” [11, 12]. The solubility of proteins in organic solvents is controlled by parameters, such as: (i) the nature of the solvent used; (ii) the assembly of the enzyme–solvent interface; and (iii) the nature of the protein. Typically, enzymes are very soluble in solvents which are very hydrophilic, polar and protic, such as dimethylsulfoxide (DMSO), ethylene glycol, and forma-mide, because they do not strip the essential water from the enzymes. A few water molecules bound to charged groups on the surface of enzyme molecules are required in order for enzymatic function to occur. These factors cause different electrostatic and hydrogen interactions at the enzyme–solvent boundary. The dif-ferent solubility is also reflected in a different activity in the presence of organic solvents [13], with higher catalytic activity being observed in organic solvent if the enzyme is lyophilized from a buffered, aqueous solution at the correct pH value for enzymatic activity [14]. In light of this, the pH of a storage buffer used prior to lyophilization will have an influence on the activity and solubility behavior. The solubility of an enzyme is increased when the pH value of the previously used buffer is far from the isoelectric point of the protein. Due to the fact that the pH concept can only be applied in aqueous systems, the so-called “pH memory effect” was relativized later as a result of the net charge of buffer ions presented in the lyophilizate. The pH memory effect can be overruled by using volatile buffers during the freeze-drying step [15]. Denaturation or inactivation can occur in cases where enzymes are completely dissolved in an anhydrous, homogeneous organic solvent system. In terms of the specificity of active enzymes in anhydrous systems, it should be mentioned that this feature is lower compared to aqueous systems. This type of system has several advantages, especially for up-scaling applications;

6 1 DirigentEffectsinBiocatalysis

notably, due to a lack of any water–organic interface there are no diffusional limita-tions for substrate and product, while substrate utilization and product formation on the enzyme surface can be controlled much more easily.

All of the different above-mentioned solvent effects can be further increased and decreased by employing different strategies that include protein engineering, the covalent attachment of amphipathic compounds, entrapment in water-in-oil microemulsions and reversed micelles, immobilization, and “solid” enzymes such as lyophilized or crosslinked enzyme aggregates.

The different effects of organic solvents on enzyme hydration and solvent binding at the active site have been investigated using molecular dynamics (MD) simulations [16]. Over a timescale of few nanoseconds, no significant structural conformation and flexibility changes could be observed on surfactant-solubilized subtilisin BPN′ testing four solvents (water, acetonitrile, tetrahydrofuran, octane). Besides the absence of any significant structural changes, the key factor that causes changes in activity, stability and selectivity is the partitioning of hydration water between the enzyme and the bulk solvent. Depending on the polarity of the bulk solvent used, the important “interface water” is removed from the enzyme to differ-ent extents. This behavior of organic solvents can be described as a “water-stripping effect” [16]. The remaining water is very important, however, for the hydration procedure of an enzyme. In short, the activity of an enzyme will increase in line with the amount of water available for the reaction in an anhydrous organic solvent. Is there too little water available, this will lead to an incomplete hydration of the enzyme, accompanied by a loss of flexibility and/or activity. The water concentra-tion can be expressed quantitatively by the thermodynamic water activity aw. This parameter was developed to account for the intensity with which water associates with various non-aqueous constituents, and is the product of the concentration of water expressed as its molar fraction xw and the water activity coefficient γw [17, 18].

a xw w w= ∗γ

This equation leads to the rule: with increasing hydrophilicity of a solvent, an increasing amount of water is necessary to achieve optimal activity. For example, a comparatively low water content is necessary when using diethyl ether as solvent, whereas a much higher content of water is required when using 2-butanol [19].

Water can be replaced with an organic solvent without affecting the enzyme in a negative manner, thus demonstrating that enzymes are able to function in a near-anhydrous organic medium. Some experimental confirmation has been pro-vided of the above rationale by the sulfoxidation of thioanisole in different organic solvents. The negative aspects of this reaction in an aqueous system have been problematic because of the poor water solubility of diaryl and alkyl sulfides, the spontaneous uncatalyzed oxidation of sulfides resulting in racemic sulfoxides, and the poor stereoselectivity of the peroxidase used (Scheme 1.1). Each of these three problems can be solved, however, by using organic solvents as the reaction media. Notably, the reaction occurred between 10- and 100-fold faster when using metha-nol as the reaction media, while the spontaneous auto-oxidation was 100- to 1000-fold slower when using ethyl acetate or acetonitrile [20, 21].