Biocatalysis for the Pharmaceutical Industry: Discovery, Development, and Manufacturing

BIOCATALYSIS FOR THE PHARMACEUTICAL INDUSTRY Discovery, Development, and Manufacturing Junhua (Alex) Tao Elevance Renew...

Author: Junhua Tao | Guo-Qiang Lin | Andreas Liese

80 downloads 2048 Views 3MB 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

BIOCATALYSIS FOR THE PHARMACEUTICAL INDUSTRY Discovery, Development, and Manufacturing Junhua (Alex) Tao Elevance Renewable Sciences, USA

Guo-Qiang Lin Shanghai Institute of Organic Chemistry, CAS, China

Andreas Liese Hamburg University of Technology, Germany

BIOCATALYSIS FOR THE PHARMACEUTICAL INDUSTRY

BIOCATALYSIS FOR THE PHARMACEUTICAL INDUSTRY Discovery, Development, and Manufacturing Junhua (Alex) Tao Elevance Renewable Sciences, USA

Guo-Qiang Lin Shanghai Institute of Organic Chemistry, CAS, China

Andreas Liese Hamburg University of Technology, Germany

Copyright # 2009

John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop, # 02-01, Singapore 129809

Visit our Home Page on www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as expressly permitted by law, without either the prior written permission of the Publisher, or authorization through payment of the appropriate photocopy fee to the Copyright Clearance Center. Requests for permission should be addressed to the Publisher, John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop, # 02-01, Singapore 129809, tel: 65-64632400, fax: 65-64646912, email: [email protected]. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. All trademarks referred to in the text of this publication are the property of their respective owners. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Other Wiley Editorial Offices John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstrasse 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons Canada Ltd, 5353 Dundas Street West, Suite 400, Toronto, ONT, M9B 6H8, Canada Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Library of Congress Cataloging-in-Publication Data Tao, Junhua. Biocatalysis for the pharmaceutical industry: discovery, development, and manufacturing/Junhua Tao, Guo-Qiang Lin, Andreas Liese. p. cm. Includes bibliographical references and index. ISBN 978-0-470-82314-9 (cloth) 1. Pharmaceutical biotechnology. 2. Enzymes–Biotechnology. I. Lin, Guo-Qiang. II. Liese, A. (Andreas) III. Title. RS380.T36 2009 2008045613 6150 .19–dc22 ISBN 978-0-470-82314-9 (HB) Typeset in 10/12pt Times by Thomson Digital, Noida, India. Printed and bound in Singapore by Markono Print Media Pte Ltd, Singapore. This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.

Contents Preface

xi

1

Enzymes and Their Synthetic Applications: An Overview 1.1 Introduction 1.2 Enzyme Families 1.3 Enzyme Discovery and Optimization 1.4 Enzyme Production 1.5 Enzymes and Synthetic Applications 1.5.1 Ketoreductases (EC 1.1.1.2) 1.5.2 Enoate Reductases or Ene Reductases (EC 1.3.1.16) 1.5.3 Oxygenases (EC. xxxx) 1.5.4 Alcohol Oxidases (EC 1.1.3.X) 1.5.5 Peroxidases (EC 1.11.1.X) 1.5.6 Halogenases (EC. xxxx) 1.5.7 Nitrilases (EC 3.5.5.1) 1.5.8 Nitrile Hydratases (EC 4.2.1.84) 1.5.9 Epoxide Hydrolases (EC 3.3.2.X) 1.5.10 !-Transaminases (EC 2.6.1.X) 1.5.11 Hydroxynitrile Lyases (EC 4.1.2.X) 1.5.12 Aldolases (EC. xxxx) 1.5.13 Glycosidases (EC. xxxx) 1.5.14 Glycosyltransferase (EC. xxxx) 1.6 Conclusions

1 1 1 2 3 3 4 4 5 6 7 8 9 9 10 10 10 12 12 14 15

2

Expression Hosts for Enzyme Discovery and Production 2.1 Introduction 2.2 How to Choose an Expression System 2.3 Prokaryotic Expression Systems 2.3.1 Posttranslational Modification in Prokaryotes 2.3.2 Escherichia coli 2.3.3 Bacilli 2.3.4 Pseudomonas fluorescens 2.3.5 Other Prokaryotic Expression Systems

21 21 22 24 25 25 26 27 28

vi

Contents

2.4

2.5 2.6

Eukaryotic Expression Systems 2.4.1 Yeasts 2.4.2 Filamentous Fungi 2.4.3 Insect/Baculovirus System 2.4.4 Mammalian Cell Cultures 2.4.5 Other Expression Systems Cell-Free Expression Systems Conclusions

29 29 32 33 34 35 36 37

3

Directed Enzyme Evolution and High-Throughput Screening 3.1 Introduction 3.2 Directed Evolution Library Creation Strategies 3.2.1 Random and Semi-Rational Mutagenesis 3.2.2 Gene Shuffling 3.3 Directed Evolution Library Screening/Selection Methods 3.3.1 In Vivo Methods: Genetic Complementation 3.3.2 In Vivo Methods: Chemical Complementation 3.3.3 In Vivo Methods: Surface Display 3.3.4 In Vitro Methods: Lysate Assay 3.3.5 In Vitro Methods: Ribosome Display 3.3.6 In Vitro Methods: In Vitro Compartmentalization 3.3.7 Equipment/Automation 3.4 Selected Industrial Examples 3.4.1 Activity 3.4.2 Thermostability 3.4.3 Substrate Specificity 3.4.4 Product Specificity 3.4.5 Enantioselectivity 3.5 Conclusions and Future Directions

45 45 46 47 48 52 52 53 53 55 55 55 56 56 56 58 58 59 59 60

4

Applications of Reaction Engineering to Industrial Biotransformations 4.1 Introduction 4.2 Metabolic Bioconversion 4.3 Enzymatic Biotransformations 4.3.1 Cofactor Regeneration 4.3.2 Racemic Mixtures 4.3.3 Equilibrium Conversion 4.3.4 By-Product Formation 4.3.5 Substrate Inhibition 4.3.6 Low Solubility 4.4 Conclusions

65 65 66 67 67 69 72 76 79 82 83

5

Chiral Synthesis of Pharmaceutical Intermediates Using Oxynitrilases 5.1 Introduction 5.2 HNL 5.2.1 The Natural Function and Distribution of HNLs

89 89 89 89

Contents

vii

5.2.2 Classification of HNLs 5.2.3 New HNLs and High-Throughput Screening 5.3. Reaction of HNLs 5.3.1 Reaction System 5.3.2 Immobilization of Enzyme 5.3.3 Continuous Reactors 5.3.4 Henry Reaction 5.4 Transformation of Cyanohydrins 5.4.1 Transformation of Hydroxyl Group 5.4.2 Transformation of Nitrile Group 5.4.3 Intramolecular Reaction 5.5 Summary

91 93 95 95 97 98 98 99 99 100 102 104

6

Expanding the Scope of Aldolases as Tools for Organic Synthesis 6.1 Directed Evolution and Rational Mutagenesis 6.2 Reaction Engineering 6.3 Broad Substrate Tolerance of Wild-Type Aldolases 6.4 Conclusions

111 111 114 116 118

7

Synthetic Applications of Ketoreductases and Alcohol Oxidases 7.1 Ketoreductases 7.1.1 Wild-Type Whole-Cell Biocatalysts 7.1.2 Recombinant Whole-Cell Biocatalysts Overexpressing Catalytic Enzymes 7.1.3 Isolated Enzyme 7.2 Alcohol Oxidases 7.2.1 Primary Alcohol Oxidases 7.2.2 Secondary Alcohol Oxidases

121 121 122

8

Applications of Nitrile Hydratases and Nitrilases 8.1 Introduction 8.2 NHase 8.2.1 New NHases 8.2.2 Applications 8.3 Nitrilase 8.3.1 New Nitrilases 8.3.2 Applications 8.4 Conclusions

153 153 153 153 154 163 163 167 178

9

Biosynthesis of Drug Metabolites 9.1 Introduction 9.2 Metabolite Synthesis Using Mammalian Bioreactors 9.2.1 Selection of In Vitro Systems 9.2.2 Reaction Condition Optimization 9.2.3 Large Scale Incubations

183 183 184 185 187 190

125 129 142 142 144

viii

Contents

9.3

9.4

9.5

9.2.4 Examples with Mammalian Bioreactors 9.2.5 In Vivo Samples Metabolite Synthesis Using Microbial Bioreactors 9.3.1 Microbial Bioreactors Used in Metabolite Structure Elucidation 9.3.2 Microbial Bioreactors Used in Synthesis of Key Metabolites 9.3.3 Strain Selection 9.3.4 Microbial Glycoside Conjugation 9.3.5 Large Scale Reactions 9.3.6 Examples for Metabolite Synthesis with a Microbial Bioreactor Recombinant Enzyme Bioreactors 9.4.1 Advantages of Using CYP Enzymes for Producing Drug Metabolites 9.4.2 Human Cytochrome Biocatalysts 9.4.3 Microbial CYP Enzymes Summary

10 Application of Whole-Cell Biotransformation in the Pharmaceutical Industry 10.1 Introduction 10.1.1 Whole-Cell Biotransformation Processes Used in Commercial Production of Pharmaceuticals 10.1.2 Application of Whole-Cell Biotransformation Process in the Synthesis of Chiral Pharmaceutical Intermediates 10.2 Disadvantages of Whole-Cell Process Compared with the Isolated Enzyme Process 10.2.1 Substrate Availability and Recovery of Products in Low Concentrations 10.2.2 Undesirable Side Reactions 10.2.3 Toxicity of Substrate and Product 10.3 Advantages of Whole-Cell Process Compared with the Isolated Enzyme Process 10.3.1 More Stable Sources than Isolated Enzymes 10.3.2 Regeneration of Cofactors and Multi-Enzymes Reactions 10.3.3 Diversity and Availability 10.3.4 Reactions with Non-Commercially Available Isolated Enzymes for Preparative Scale Synthesis 10.3.5 Cost Effectiveness and Ease of Operation 10.4 Approaches to Address the Disadvantages of Whole-Cell Biotransformation 10.4.1 Control of Substrate and Product Concentration by Absorbing Resins 10.4.2 Immobilized-Cell Technology 10.4.3 Aqueous–Organic Two-Phase System 10.4.4 Genetic Engineering Approaches 10.5 Conclusions

190 192 192 193 193 194 196 199 200 202 202 205 206 207

213 213 214 214 217 217 217 217 218 218 218 218 218 219 220 220 221 222 223 224

Contents

11 Combinatorial Biosynthesis of Pharmaceutical Natural Products 11.1 Introduction 11.2 Combinatorial Biosynthesis: The Natural Way for Structural Diversity 11.3 Examples of Combinatorial Biosynthesis of Pharmaceutical Natural Products 11.3.1 Erythromycin (Polyketide Biosynthesis) 11.3.2 Daptomycin (Nonribosomal Peptide Biosynthesis) 11.3.3 Patellamide (Ribosomal Peptide Biosynthesis) 11.4 Summary and Perspectives 12 Metabolic Engineering for the Development and Manufacturing of Pharmaceuticals 12.1 Introduction 12.2 Metabolic Engineering Tools 12.2.1 Tools for the Cellular Metabolic Network Analysis 12.2.2 Tools for Rational Genetic Modification 12.3 Metabolic Engineering for the Development and Production of Polyketide Pharmaceuticals 12.3.1 Biosynthesis of Polyketides 12.3.2 Metabolic Engineering for Improved Erythromycin Production 12.3.3 Metabolic Engineering for Overproduction of 6dEB in Heterologous Hosts 12.3.4 Metabolic Engineering of Other Polyketides 12.3.5 Development of Novel Polyketides for Drug Discovery 12.4 Metabolic Engineering for the Production of -Lactam 12.5 Metabolic Engineering for Isoprenoid Production 12.5.1 Biosynthesis Pathway of Isoprenoids 12.5.2 Metabolic Engineering for Enhancing Precursor Supply for Isoprenoid Production 12.5.3 Metabolic Engineering for Artemisinine Development and Production 12.5.4 Metabolic Engineering for Carotenoid Production 12.5.5 Metabolic Engineering for Taxol Development and Production 12.6 Conclusions 13 Multimodular Synthases and Supporting Enzymes for Chemical Production 13.1 Introduction 13.2 Background 13.2.1 Multimodular Synthase Architecture 13.2.2 Natural Product Biosynthetic Cycle 13.3 Metabolic Engineering of Megasynthases 13.3.1 Daptomycin: Metabolic Engineering by Domain Swap 13.3.2 Avermectin: Metabolic Engineering by Directed Fermentation

ix

229 229 230 234 234 236 239 240

247 247 248 248 251 253 253 253 254 256 256 257 258 259 260 261 262 262 265

273 273 274 274 276 277 278 281

x

Contents

13.4 Excised Domains for Chemical Transformations 13.4.1 Function of Individual Domains, Domain Autonomy 13.4.2 Cyclization 13.4.3 Halogenation 13.4.4 Heterocyclization/Aromatization 13.4.5 Methylation 13.4.6 Oxygenation 13.4.7 Glycosylation 13.5 Conclusions

284 284 285 288 290 292 294 295 299

14 Green Chemistry with Biocatalysis for Production of Pharmaceuticals 14.1 Introduction 14.2 Enzymatic Resolutions: Higher Yields, Less Waste 14.3 Bioreductions: Greener Ligands, Renewable Hydride Donors, No Metals 14.3.1 Enzymatic Oxidations: Clean, Highly Selective and Catalytic 14.4 C C Bond Formations: Atom Efficiency at Its Best 14.5 Summary and Outlook

305 305 309 312 314 316 318

Index

323

Preface Biocatalysis is evolving to be a transformational technology as a result of a confluence of factors, which include (1) large scale and ever increasingly cost-efficient DNA sequencing technologies; (2) exponential growth in GenBank; (3) powerful directed enzyme evolution and high-throughput screening technologies; (4) robust expression systems for enzyme production; (5) deep understanding of the logic of natural product biosynthesis; (6) industrial successes of metabolic engineering and pathway engineering. Consequently, many successful stories and a number of reviews have been reported recently in developing biocatalysis for the pharmaceutical industry, across drug discovery, development, and manufacturing. The book is dedicated to these advances, and divided into four parts: .

.

. . .

Chapters 1–4 serve as an introduction to emerging biocatalysts, modern expression hosts, state of the art of directed evolution, high-throughput screening, and bioprocess engineering for industrial applications. Chapters 5–8 are directed to emerging enzymes, which include oxynitrilases, aldolases, ketoreductases, oxidases, nitrile hydratases, and nitrilases, and their recent applications especially in synthesis of chiral drugs and intermediates. Chapters 9 and 10 focus on synthesis of drug metabolites and intermediates catalyzed by P450s or whole cells. Chapters 11–13 are devoted to combinatorial biosynthesis, metabolic engineering, and autonomous enzymes for the synthesis and development of complex medicinal molecules. Chapter 14 discusses the recent impact of biocatalysis in green chemistry and chemical development.

Our main goal is to come up with a concise but comprehensive, practical but insightful book covering the topics discussed above. We hope you enjoy reading this book. Any suggestions and comments are welcome. Junhua (Alex) Tao Elevance Renewable Sciences, USA Email: [email protected] Guo-Qiang Lin Guo-Qiang Lin Shanghai Institute of Organic Chemistry, China Email: [email protected] Andreas Liese Hamburg University of Technology, Germany Email: [email protected]

1 Enzymes and Their Synthetic Applications: An Overview Junhua (Alex) Tao1 and Jian-He Xu2 1 2

Elevance Renewable Sciences, 175 E. Crossroads Parkway, Bolingbrook, IL 60440, USA East China University of Science and Technology (ECUST) 130 Meilong Road, Shanghai, 200237, PR China

1.1 Introduction Whole-cell biocatalysis has been exploited for thousands of years; for example, in preparing barley for beer brewing. While the chemical, economic and social advantages of biocatalysis over traditional chemical approaches were recognized a long time ago, their applications for the drug industry have been largely underexplored until the recent technological breakthroughs in large-scale DNA sequencing, robust protein expression systems, metabolic engineering and directed evolution. In this chapter, emphasis will be directed to the discussion of those isolated enzymes which are uniquely suited for the synthesis of small-molecule pharmaceutical ingredients.

1.2 Enzyme Families Based on reactions they catalyze, enzymes can be broadly classified into six major categories (Table 1.1) [1]. It was estimated that about 60% of biotransformations currently rely on the use of hydrolases, followed by 20% of oxidoreductases [2]. On the other hand, some of the CC bond-forming and oxygenation enzymes catalyze reactions with very high reaction efficiency and very low waste generation, underlining the potential of emerging enzymes.

Biocatalysis for the Pharmaceutical Industry edited by Junhua Tao, Guo-Qiang Lin, and Andreas Liese 2009 John Wiley & Sons Asia (Pte) Ltd

Biocatalysis for the Pharmaceutical Industry

2

Table 1.1

Enzyme classes

Enzyme class

Examples

Reaction catalyzed

Hydrolases

lipase, protease, esterase nitrilase, nitrile hydratase glycosidase, phosphatase

hydrolysis reactions in H2O

Oxidoreductases

dehydrogenase, oxidase oxygenase, peroxidase

oxidation or reduction

Transferases

transaminase, glycosyltransferase transaldolase

transfer of a group from one molecule to another

Lyases

decarboxylase, dehydratase, deoxyribose-phosphate aldolase

nonhydrolytic bond cleavage

Isomerases

racemase, mutase

intramolecular rearrangement

Ligases

DNA ligase

bond formation requiring triphosphate

1.3 Enzyme Discovery and Optimization Traditionally, enzymes are discovered through screening of environmental samples and culture enrichment. As a result of recent technological breakthroughs in large-scale DNA sequencing and high-throughput screening, both the metagenomic approach and sequence-based discovery have drastically shortened the cycle of enzyme discovery. In the metagenomic approach, DNA was directly extracted from uncultured samples followed by cloning and expression [3]. For example, by combination of directed evolution with the metagenome approach, an a-amylase mutant with optimal activity at pH 4.5 and optimal thermostability at 105 C was discovered for starch liquefaction and EtOH production [4]. Sequence-based discovery (genome hunting) is increasingly attractive, as the public sequence bank is growing rapidly. In this approach, known sequences encoding an enzyme of interest are used to search gene databases to uncover enzymes of homologous sequences. For example, using this method, a library of deoxyribose-phosphate aldolases (DERAs) were rapidly constructed and from them a novel DERA was identified to catalyze a sequential aldol reaction of a nonnative substrate with high throughput and excellent stereoselectivity for the synthesis of statin side chains [5]. Since most synthetic applications require enzymes catalyzing nonnatural substrates, their properties often have to be improved. One way to achieve this is to optimize reaction conditions such as pH, temperature, solvents, additives, etc. [6–9]. Another way is to modulate the substrates without compromising the synthetic efficiency of the overall reaction [10]. In most cases for commercial manufacturing, however, the protein sequences have to be altered to enhance reactivity, stereoselectivity and stability. It was estimated that over 30 commercial enzymes worldwide have been engineered for industrial applications [11]. Precise prediction of which amino acids to mutate is difficult to achieve. Since the mid 1990s, directed evolution

Enzymes and Their Synthetic Applications: An Overview

3

has been demonstrated to be a powerful and robust technology to improve the desired properties [12,13]. Among them, the error-prone PCR method is probably the most popular to create random mutants by changing polymerization reaction conditions [14]. Alternatively, recombination of homologous sequences or DNA-shuffling methods can be used to introduce mutants with improved properties [15]. The major challenge in directed evolution is not generation of mutant libraries; rather, it is the availability of high-throughput assays [16]. In most cases, it requires screening of tens of thousands of mutants, which is usually tedious and time consuming. As more and more protein structures are available from the protein database, focused directed evolution or semi-rational protein design is becoming more and more popular [17]. In this approach, the three-dimensional (3-D) structure of a suboptimal enzyme is constructed by a computer algorithm from a homologous enzyme with known 3-D structure. Docking studies are then applied to search potential ‘hot spots’, which are then swapped with other amino acids by site-saturation mutagenesis. In this way, there are generally less than a few thousand mutants to be screened, significantly shortening the cycle of enzyme development. The fact that most beneficial mutations are proved to be near the active site makes this approach even more attractive [18].

1.4 Enzyme Production Although some enzymes are still extracted from animal or plant tissue, most of them are now produced from microorganisms by fermentation. Bacteria and fungi are the most popular hosts for producing industrial enzymes, due to easy handling and high productivity. They can also be readily genetically engineered to improve their performance; for example, by incorporating secretion systems to facilitate enzyme isolation and purification. Some of the most popular expression hosts are Escherichia coli, Pichia pastoris, Pseudomonas fluorescens, Aspergillus sp. and Bacillus sp. Mammalian or plant cells are used in special cases [19–21]. By regulation, the production host should have GRAS status (Generally Regarded as Safe Status). In a typical enzyme production procedure, cells containing genes encoding desired enzymes are grown in an Erlenmeyer flask. At large scale, a computer-controlled fermenter or bioreactor is required to maintain an appropriate control of pH, O2, NH3 and CO2 to maximize cell density. The cells are harvested by centrifugation in a batch or continuous fashion. Alternatively, they can be collected through membrane filtration devices. The cell membranes are then disrupted by an ultrasonicator or French press at small scale. At a scale of over 5–10 L, a homogenizer is usually used. After centrifugation to remove cell debris, the crude enzymes remain in the supernatant and can be concentrated through precipitation by adding either inorganic salts (e.g. ammonium sulfate) or organic solvents (e.g. acetone). The crude enzymes are then purified by dialysis or a variety of chromatographic methods. The dry powder is usually obtained after lyophilization under freeze-drying conditions [22,23].

1.5 Enzymes and Synthetic Applications Historically, the most popular enzymes used for chemical synthesis are lipases, esterases, proteases, acylases and amidases, among others. Recently, a number of recombinant biocatalysts have been discovered and isolated, significantly expanding the toolbox for biotransformations. In this section, the focus will be on these new enzymes.


4

1.5.1 Ketoreductases (EC 1.1.1.2) Ketoreductases (KREDs) catalyze the conversion of a wide range of ketones and some aldehydes to chiral alcohols regio- and stereo-selectively in the presence of NADH or NADPH (Figure 1.1) [24,25]. This powerful transformation has been demonstrated in a number of industrial transformations using either isolated enzymes or whole cells. The use of isolated enzymes is often preferred because of a higher volumetric productivity and the absence of side reactions. A key to its success is the availability of efficient and cost-effective cofactor regeneration methods by using a formate dehydrogenase to recycle NAD þ or a glucose dehydrogenase to recycle NADP þ (Figure 1.1) [26,27]. It shall be noted that some alcohol dehydrogenases are also able to catalyze the oxidation of alcohols to ketones or aldehydes [28].

Reduction and Cofactor Regeneration O

OH

ketoreductase R2

R1

R1 * R2 NAD

NADH CO 2 + NH3

+

HCOONH 2 formate dehydrogenase

O

OH

ketoreductase R2

R1

R1 * R2 NADP

NADPH

+

glucose

gluconate glucose dehydrogenase Product Examples: OH R1 * R2

OH

OH OR 2

R1 *

R1 *

R2 R3

O

OH

O

O

R1 *

OR 2 R3

Figure 1.1

1.5.2 Enoate Reductases or Ene Reductases (EC 1.3.1.16) Enoate reductase (ER) catalyzes NAD(P)H-dependent reduction of carbon–carbon double bonds of nonactivated enoates, as well as of a,b-unsaturated aldehydes, ketones, nitros, and nitriles (Figure 1.2). For example, the ER from Clostridium tyrobutyricum shows high stereospecificity and regioselectivity and broad substrate specificity [29]. Alkanes with up to two chiral centers can be directly produced by asymmetric reduction of electron-deficient alkene enzymes from the ‘old yellow enzyme’ family at the expense of NAD(P)H. The cofactor can be regenerated in vitro using a formate dehydrogenase or glucose dehydrogenase. Alternatively, a whole-cell system can be used to co-express ERs with redox enzymes for NAD(P)H recycling [30].


R1 R2 R3

R1

ene reductase

EWG

5

R2 * NAD(P) +

NAD(P)H

Product Examples: R1 O R2 *

* R3

R2 *

O

R1

R1 H

EWG = aldehydes, ketones, nitros, nitriles

EWG * R3

NO 2 * R3

R2 *

R1

CN R'

* R3

* *

R2

Figure 1.2

1.5.3 Oxygenases (EC. xxxx) Oxygenases catalyze direct incorporation of molecular oxygen into substrates to produce oxygenated molecules [31,32]. They are categorized as either monooxygenases (MOs) or dioxygenases, depending upon whether one or both atoms of dioxygen are inserted into a substrate. The metal-dependent MOs, such as P450s, catalyze a wide range of reactions via metal oxo species (e.g. hydroxylation of alkanes and aromatics; epoxidation of alkenes (Figure 1.3)), while flavin adenine dinucleotide (FAD)-dependent MOs are found to catalyze oxidation of heteroatoms (S, N, Se, P) and Baeyer–Villiger reactions via FAD-hydroperoxide (FAD-OOH) species (Figure 1.4).

Simplified Mechanism (P450s) N N

Fe 2+

N N

O2 NAD(P)H

N N

O R-H

N N

Fe 5+

H2 O

S iron oxo species

S

N: representing heme S: representing cysteine residual

Examples: R1 R2 R3

H

P-450s, O 2 hydroxylation

R1 R2 R3

P-450s, O 2 epoxidation

R

O R OH

H P-450s, O 2 hydroxylation

Figure 1.3

OH

R-OH


6 Simplified Mechanism (FAD-dependent MOs) H N

N

NH

N

H N

O FAD-MOs, O 2 NADPH

N H O O HO FAD-OOH

O

O R1

Baeyer-Villiger

FAD-MOs, O2 R X heteroatom oxidation X = S, N, Se, P

R1

N

N H O O H

O NH

O

O R1

R2

O

FAD-MOs, O2 R2

NH

R1

Examples:

H N

O

N

O R2

O R2

R X O

Figure 1.4

Dioxygenases usually contain a tightly bound iron atom and catalyze hydroperoxidation of allylic molecules or carboxylic acids, and dihydroxylation of aromatics (Figure 1.5) [33]. Currently, these oxygenation reactions are usually carried out in whole cells, the outcome of which is often unpredictable. The discovery of novel oxygenases and efficient hosts for protein expression remain keys to further expanding the applications of these enzymes in chemical synthesis and drug metabolism studies [34–37]. OOH

dioxygenases hydroperoxidation

R O R

R O

dioxygenases OH

hydroperoxidation

R

OH OOH

dioxygenases dihydroxylation

OH OH

Figure 1.5

1.5.4 Alcohol Oxidases (EC 1.1.3.X) Alcohol oxidases (AOs) catalyze oxidation of alcohols to aldehydes or ketones in the presence of molecular oxygen, with hydrogen peroxide being the usual by-product (Figure 1.6). Some of the most well-studied AOs are cholesterol oxidases, short-chain aliphatic alcohol oxidases, aromatic alcohol oxidases, pyranose oxidases, glycolate oxidases, glucose oxidases, galactose oxidases and nucleoside oxidases [38–40]. Cholesterol oxidases catalyze oxidation of allylic alcohol in the cholesterol scaffolds [41]. The regeneration of cofactor FAD is relatively easier, as it is tightly bound. Although these enzymes use oxygen, they can also be deactivated by oxygen or the hydrogen peroxide by-product. It is to be noted that peroxidases and cholorperoxidases can also catalyze the oxidation of alcohols using hydrogen peroxide (H2O2).

Enzymes and Their Synthetic Applications: An Overview AOs OH

R

R = alkyl, aryl

O + H 2O 2

R

O2

Examples:

O

O P2O O

HO

OH

HO

OH

OH OH

O2

HO

7

P2O = pyranose oxidase GAOX = galactose oxidase

OH OH

GAOX

O

HO

OH

HO

O OH

R

R

cholesterol oxidase O

HO

Figure 1.6

1.5.5 Peroxidases (EC 1.11.1.X) Peroxidases utilize H2O2 as the oxidant (Figure 1.7). The active site of peroxidases may involve a heme unit (horseradish peroxidase), selenium (glutathione peroxidase), vanadium (bromoperoxidase) and manganese (manganese peroxidase). These enzymes catalyze a wide range of oxidations, including hydroxylation of arenes, oligomerization of phenols and aromatic amines, epoxidation and halogenation of olefins, oxygenation of heteroatoms and reduction of hydroperoxides [42–44].

MeO

R

R

peroxidase H 2 O2

MeO

HO

OH R OMe

OH

OH peroxidase H 2 O2

OH OMe

OH

peroxidase H 2 O2

OMe n polymer R1

R 1 peroxidase H 2 O2

Figure 1.7

O


8

1.5.6 Halogenases (EC. xxxx) Halogenases catalyze regio- and stereo-selective halogenation (Figure 1.8) [45,46]. For electron-rich substrates, nature often uses flavin-dependent halogenases for chlorination, bromination or iodination via FADH-OX (X ¼ halide) as the halogenation agent (Figure 1.9). For electron-deficient molecules such as alkanes, mononuclear iron halogenases are utilized through a radical mechanism (Figure 1.9). Fluorinases adopt an SN2 nucleophilic substitution R2 R1

R2

tryptophan 7-halogenase R 1 = H, Me R 2 = amine, alkyl, nitrile

N H Cl

chlorinase

O

H 2N

Cl

R1

NaCl, FADH 2, O 2

N H

O

H 2N

ClS

S

enzyme

enzyme NH2

NH2

HO 2 C

Me

N

S+

N

N

N F

fluorinase F-

N

N

O

O

NH2

N

N

OH OH

OH OH

Figure 1.8

Cofactors and Oxidants of Halogenating Enzymes H N O2 C R N N H

O H N

O

O

Cl NH

mononuclear iron halogenase R and Cl .

O O N V OH O

vanadium haloperoxidase H 2 O2

N

FeII N

NH

O flavin-dependent halogenase O 2

HN

O H

N N

.

N FeIII N

heme-iron haloperoxidase H 2O2

Figure 1.9


9

mechanism in the presence of F to introduce a fluorine atom [47]. While it is still in its infancy, the use of enzymatic halogenation has shown great promise, especially in whole-cell systems (Figure 1.9). For example, tryptophan 7-halogenase was able to catalyze regioselective halogenation of a wide range of indole derivatives and aromatic heterocycles [48]. It is to be noted that halogenation can also be catalyzed by haloperoxidases, which often gives poor regio- and stereo-selectivity, since the activated halogen source, (e.g. hypohalous acids) is freely diffusible within and away from enzymes (Figure 1.8) [49].

1.5.7 Nitrilases (EC 3.5.5.1) Nitrilases convert nitriles to the corresponding carboxylic acids and NH3 through a cysteine residue in the active site [50]. Because of their high enantio- and regio-selectivity, nitrilases are attractive as ‘green’ catalysts for the synthesis of a variety of carboxylic acids and derivatives (Figure 1.10) [51,52]. Recently, a number of recombinant nitrilases have been cloned and characterized heterologously for synthetic applications [50,53,54]. Mechanism: R

N

NH2

nitrilase H 2O

R

O

S-enzyme OH

*R

OH

R = aryl, alkenyl, alkyl Product Examples:

*R

O

O

O OH

N

n

R1

OH

OH

* R2

Figure 1.10

1.5.8 Nitrile Hydratases (EC 4.2.1.84) Nitrile hydratase (NHase) catalyzes the hydration of nitriles to amides (Figure 1.11) and has been used for production of acrylamide and nicotinamide at large scale. NHases are roughly Mechanism: nitrile hydratase R N R M 3+ R = aryl, alkenyl, alkyl

N

M 3+

R

Product Examples: O

O *R

NH 2

N

n

O NH 2

R1

Figure 1.11

* R2

O

NH

H 2O

NH2

OH

*R

NH 2


10

classified into iron and cobalt types according to the metal involved in the active site [55,56]. Recent elucidation of the catalytic mechanism and characterization of a number of NHases have led to a wide range of applications in both biotransformation and bioremediation [53].

1.5.9 Epoxide Hydrolases (EC 3.3.2.X) Epoxide hydrolases (EHs) catalyze the hydrolysis of a wide range of epoxides (Figure 1.12). They are cofactor independent and robust for the synthesis of enantiopure epoxides, diols and their derivatives [57–59]. There are over 100 epoxide hydrolase gene sequences, and the X-ray structures are available for fungal, bacterial and mammalian epoxide hydrolases [60]. Heterologous expression in E. coli and other hosts has also been successful. In addition, several efficient high-throughput screening methods have been developed, allowing the improvement of EHs through site-directed mutagenesis and directed evolution [61,62]. EHs also play important roles in the detoxification of genotoxic compounds and drug metabolism [63]. Mechanism: epoxide hydrolase

enzyme

enzyme O

O

-

O

OH H2 O

O

H2 O

O

OH OH

O Examples: R

R1

O

O

R2

R1

R1

R1 R2

R3

O

R2 O

R3 O

R2 R4

Figure 1.12

1.5.10 v-Transaminases (EC 2.6.1.X) v-Transaminases catalyze the conversion of a ketone group to an amine in the presence of the cofactor pyridoxal-50 -phosphate (PLP), which is tightly bound to the active site. The catalysis starts with the formation of an imine between the cofactor and an amine donor (Figure 1.13) [64]. Since this is a one-step method to prepare (R)- or (S)-amines from ketones, it has huge potential (Figure 1.14) [65,66]. For example, a high-throughput biocatalytic process to (S)-methoxyisopropylamine, a moiety common to the two important chloroacetamide herbicides metolachlor and dimethenamid, has been reported by enzymatic transamination of methoxyacetone using isopropylamine as the donor [67].

1.5.11 Hydroxynitrile Lyases (EC 4.1.2.X) Hydroxynitrile lyases (HNLs or oxynitrilases) catalyze CC bond-forming reactions between an aldehyde or ketone and cyanide to form enantiopure cyanohydrins (Figure 1.15), which are versatile building blocks for the chiral synthesis of amino acids, hydroxy ketones, hydroxy acids, amines and so on [68]. Screening of natural sources has led to the discovery of both


Mechanism: HO R'

R1

CHO PO3

R"

R2

R'

HO

HO

PO3 2-

O

N

R2 N

R1 R2 (amine acceptor)

HO

NH2

PO3 2-

R1 * R2

N CHO HO

PO3 2N

Figure 1.13

Examples: O R1

R2 O

R1

NH 2

(R)-transaminases R1

PLP

R2

(R)-amines

R2 NH 2

(S)-transaminases R1

PLP

(S)-amines

R2

Other product examples: NH 2 O

NH 2 OR 2

R1 *

R1 *

OR 2 R3

O

Figure 1.14

O R

H

CN

R

CN

NH2

R *

OH

(S)-oxynitrilase H

-

CN

R

CN

R = aromatic, alkyl, heterocyclic

Figure 1.15

R'

R * O

OH

R * O OH

OH

O R

-

OH

OH

OH

(R)-oxynitrilase

R" O PO3 2-

N R1

+

H 2N

N

2-

N

NH2 (amine donor)

11

NH

R *

2

O


12

(R)- and (S)-selective HNLs. A number of recombinant HNLs have also been expressed in E. coli, Saccharomyces cerevisiae, and Pichia pastoris. Recently, protein engineering has been successfully applied to the development of a tailor-made HNL for large-scale production of specific cyanohydrins [69,70].

1.5.12 Aldolases (EC. xxxx) Aldolases catalyze asymmetric aldol reactions via either Schiff base formation (type I aldolase) or activation by Zn2 þ (type II aldolase) (Figure 1.16). The most common natural donors of aldoalses are dihydroxyacetone phosphate (DHAP), pyruvate/phosphoenolpyruvate (PEP), acetaldehyde and glycine (Figure 1.17) [71]. When acetaldehyde is used as the donor, 2-deoxyribose-5-phosphate aldolases (DERAs) are able to catalyze a sequential aldol reaction to form 2,4-didexoyhexoses [72,73]. Aldolases have been used to synthesize a variety of carbohydrates and derivatives, such as azasugars, cyclitols and densely functionalized chiral linear or cyclic molecules [74,75]. Schiff-base Mechanism O X

N

Enzyme-NH2

X

R1 OH

R

N

Enzyme R1

Enzyme R1

OH

R R1

H

aldol reaction

O

R

X

HN X

O

Enzyme

R1 X

Zn2+ chelation Mechanism O X

OH O

X R

Zn 2+ enzyme

OH

Zn 2+

X

Zn 2+ O

O O

O

H

O X

Zn 2+ O

R

H

H aldol reaction

O X

OH R

OH

Figure 1.16

1.5.13 Glycosidases (EC. xxxx) Glycosidases catalyze the hydrolysis of glycosidic linkages via an oxonium intermediate or transition state similarly to acid-catalyzed hydrolysis of glycosides under either a retention of the configuration at the anomeric center (Mechanism, Figure 1.18) or less common inversion. The oxonium is presumably stabilized by a carboxylate group such as glutamic acid, a common structural motif in the active site of glycosidases [76]. Glycosidase-catalyzed synthesis of glycosides can be achieved under either equilibrium-controlled conditions or kinetic-controlled conditions [77–79]. In the former case, the reaction is established to shift the equilibrium toward a product; for example, by adding organic solvents. In the latter case, activated glycosyl donors are


13

Examples: O -

O

O

O P O OH ODHAP O

O +

-

O2 C

H

O

O +

H acetaldehyde

H

H

* X

O Z

*

-

Y * X

pyruvate

+

Y

O Z

*

* X

O2 C

*

+

OH

H O

R

H

OH * X

OH

OH

Y

* X

*

Z

Y

* X O

Z

O

O

OH

-

Y

O

O P O O-

Z

*

X, Y, Z = H, O, N

Y Z

*

OH

HO

R = alkyl, aryl

R NH2

NH2 glycine O

O + H H acetaldehyde

R

DERAs

OH

O

R = alkyl

R OH

Figure 1.17 Mechanism: OH

OH

OH

O

HO

OH

NuH

+

O

HO

HO -O

OH

O

OR

HO

OH

Nu

HO

-

O

OH NuH = nucleophile

O

O Enzyme enzyme-bound oxonium

Enzyme Examples: OH

OH O

HO

OH OR + R'OH

β-galactosidase

O HO OH

OH

O HO

OR'

OH R = aryl, oligosaccharides, etc. OH OH

OH

O

OH

NO 2 +

OH OH O HO OCH3 OH

Figure 1.18

O

HO α-fucosidase

OCH3

O O HO OH

OH


14

used, which include di- or oligo-saccharides, aryl glycosides, glycosyl fluorides and so on [80]. Owing to greater promiscuity toward donors and wide availability, these enzymes were found to have a wide range of applications in preparation of carbohydrates and derivatives [81–83].

1.5.14 Glycosyltransferase (EC. xxxx) Glycosyltransferases (Gtfs) accept activated sugars such as uridine diphosphate (UDP) nucleotide sugars or glycosyl phosphates as monosaccharide donors under either a retention or inversion of the configuration at the anomeric center (Mechanism, Figure 1.19; only inversion scenario shown). Recently, a number of Gtfs have been discovered to be quite promiscuous, and used to synthesize many oligosaccharides, their derivatives or glycosylated natural products, which are otherwise difficult to obtain [84–87]. For large-scale applications, the main issue to be overcome is to recycle released nucleotide monophosphate (NMP) or nucleotide diphosphate (NDP), which are expensive. Several methods have been reported [88,89], and one utilizes sugar nucleotide pyrophosphorylase, which transfers the sugar moiety of a sugar phosphate to a free uridine triphosphate (UTP) to regenerate the desired UTP-sugar (Figure 1.19) [90]. The availability of a wide range of Gtfs and sugar donors also provides a general strategy to synthesize glycosylated molecules in vivo through pathway engineering and combinatorial biosynthesis. For example, novel macrocyclic polyketides have been produced by applying a promiscuous Gtf from picromycin biosynthesis, which accepts a wide range of sugar donors (Figure 1.19) [91,92].

Mechanism (inversion scenario): OH

OH

H-Nu

O HO

H

HO O

O

OH

OH O

O

HO

Nu

OH

Enzyme LG LG = leaving group: phosphate, UTP, TDP, etc. NuH = nucleophile: alcohol, sulfide, amine, etc. Examples: OH

OH O

HO

OH +

HO HO

HO OUDP

β-1,4-galactosyl transferase

O

OH NHAc

OH

OH

HO HO

O

OH

O

O HO glycosyl transferase

HO O O

HO

OH NHAc

O NHAc O

OH O

O

O HO

O sugar sugar =

O O

HO

O

Figure 1.19

NHAc OH

O

OH OH

O O NHMe 2 glucose

OH

O HO

+ Sugar

HO

O HO

OH OH Me


15

1.6 Conclusions Biocatalysis has been practiced historically mostly by whole-cell systems, which limits its applications due to low throughput and complex in vivo pathways. As a result of recent advances in genomics and high-throughput screening, more and more diverse recombinant enzymes are available in catalogs. Subsequently, a number of them have been successfully applied to the commercial production of nonnatural molecules. More recently, biocatalysis is emerging to be one of the greenest technologies for chemical synthesis [93,94]. Specifically, biocatalysis can prevent waste generation by using catalytic processes with high stereo- and region-selectivity, prevent or limit the use of hazardous organic reagents by using water as the green solvent, design processes with higher energy efficiency and safer chemistry by conducting reactions at room temperature under ambient atmosphere, and increase atom economy by avoiding extensive protection and deprotection to maximize the use of renewable feedstock designed for degradation. Enzymes can catalyze transformations which are difficult to achieve by traditional chemical methods. To truly realize the promise of this emerging technology for the pharmaceutical industry, it is essential to integrate biocatalysis into drug discovery, development and manufacturing. For drug production, the key relies on integration of enzymatic transformations with modern chemical research and development at the retrosynthetic level to deliver efficient and practical synthetic sequences with fewer synthetic steps and significantly reduced waste streams [95,96].

References [1] Tipton, K. and Boyce, S. (2000) History of the enzyme nomenclature system. Bioinformatics, 16, 34–40. [2] Straathof, A.J., Panke, S. and Schmid, A. (2002) The production of fine chemicals by biotransformations. Current Opinion in Biotechnology, 13, 548–556. [3] Langer, M., Gabor, E.M., Liebeton, K. et al. (2006) Metagenomics: an inexhaustible access to natures diversity. Biotechnology Journal, 1, 815–821. [4] Mathur, E.J., Toledo, G., Green, B.D. et al. (2005) A biodiversity-based approach to development of performance enzymes: applied metagenomics and directed evolution. Industrial Biotechnology, 1, 283–287. [5] Hu, S., Tao, J. and Xie, Z. Chemoenzymic synthesis of atorvastatin. WO2006134482. [6] Liu, Y.-Y., Xu, J.-H., Xu, Q.-G. and Hu, Y. (1999) Significant enhancement of lipase enantioselectivity toward (S)-ketoprofen ester at pH 2. Biotechnology Letters, 21, 143–146. [7] Xu, T.-W., Xu, J.-H., Yu, W. and Zhong, J.-H. (2006) Investigating pH and Cu(II) effects on lipase activity and enantioselectivity via kinetic and spectroscopic methods. Biotechnology Journal, 1, 1293–1301. [8] Liu, Y.-Y., Xu, J.-H. and Hu, Y. (2000) Enhancing effect of Tween-80 on lipase performance in enantioselective hydrolysis of ketoprofen ester. Journal of Molecular Catalysis B: Enzymatic, 10, 523–529. [9] Gong, P.-F., Xu, J.-H, Tang, Y.-F. and Wu, H.-Y. (2003) Improved catalytic performance of Bacillus megaterium epoxide hydrolase in a medium containing Tween-80. Biotechnology Progress, 19, 652–654. [10] Tao, J., Zhao, L. and Ran, N. (2007) Recent advances in developing chemoenzymatic processes for active pharmaceutical ingredients. Organic Process Research Development, 11, 259–267. [11] Eggleston, G. and Vercellotti, J.R. (eds) (2007) Industrial Applications of Enzymes on Carbohydrate-based Materials, ACS Symposium Series 6, Oxford University Press. [12] Bloom, J.D., Meyer, M.M., Meinhold, P. et al. (2005) Evolving strategies for enzyme engineering. Current Opinion in Structural Biology, 15, 447–452. [13] Reetz, M.T. (2007) Directed evolution of enzymes for asymmetric syntheses, in Asymmetric Synthesis (eds Christmann M. and Braese S.), Wiley–VCH, pp. 207–211. [14] Eggert, T., Jaeger, K.-E. and Reetz, M.T. (2004) Directed evolution of random mutagenesis: a critical evaluation, in Enzyme Functionality (ed. A. Svendsen), Marcel Dekker, Inc., New York, pp. 375–390.

16


[15] Kagami, O., Baik, S.-H. and Harayama, S. (2004) Effective DNA shuffling methods for enzyme evolution, in Enzyme Functionality (ed. A. Svendsen), Marcel Dekker, Inc., New York, pp. 425–441. [16] Arnold, F.H. and Georgiou, G. (eds) (2003) Directed Enzyme Evolution: Screening and Selection Methods, Methods in Molecular Biology, vol. 230, Humana Press, Inc., Clifton, NJ. [17] Lutz, S. and Patrick, W.M. (2004) Novel methods for directed evolution of enzymes: quality, not quantity. Current Opinion in Biotechnology, 15, 291–297. [18] Morley, K.L. and Kazlauskas, R.J. (2005) Improving enzyme properties: when are closer mutations better? Trends in Biotechnology, 23, 231–237. [19] Macauley-Patrick, S., Fazenda, M.L., McNeil, B. and Harvey, L.M. (2005) Heterologous protein production using the Pichia pastoris expression system. Yeast, 22, 249–270. [20] Sorensen, H.P. and Mortensen, K.K. (2005) Advanced genetic strategies for recombinant protein expression in Escherichia coli. Journal of Biotechnology, 115, 113–128. [21] Nohr, J., Kristiansen, K. and Krogsdam, A.-M. (2003) Expression of Recombinant Proteins: An Introduction, in Methods in Molecular Biology, vol. 232, Humana Press, Inc., Clifton, NJ, pp. 93–101. [22] Gellissen, G. (ed.) (2005) Production of Recombinant Proteins, Wiley–VCH. [23] Scopes, R.K. (ed.) (1994) Protein Purification, 3rd edn, Springer-Verlag. [24] Goldberg, K., Schroer, K., Luetz, S. and Liese, A. (2007) Biocatalytic ketone reduction – a powerful tool for the production of chiral alcohols – part I: processes with isolated enzymes. Applied Microbiology and Biotechnology, 76, 237–248. [25] Yang, W., Xu, J.-H., Xie, Y. et al. (2006) Asymmetric reduction of ketones by employing Rhodotorula sp. AS2.2241 and synthesis of the b-blocker (R)-nifenalol. Tetrahedron: Asymmetry, 17, 1769–1774. [26] Husain, M. and Husain, Q. (2008) Applications of redox mediators in the treatment of organic pollutants by using oxidoreductive enzymes: a review. Critical Reviews in Environmental Science and Technology, 38, 1–42. [27] Goldberg, K., Schroer, K., Luetz, S. and Liese, A. (2007) Biocatalytic ketone reduction – a powerful tool for the production of chiral alcohols – part II: whole-cell reductions. Applied Microbiology and Biotechnology, 76, 249–255. [28] Kroutil, W., Mang, H., Edegger, K. and Faber, K. (2004) Biocatalytic oxidation of primary and secondary alcohols. Advanced Synthesis & Catalysis, 346, 125–142. [29] Steinbacher, S., Stumpf, M., Weinkauf, S. et al. (2002) Enoate reductase family. In Flavins and Flavoproteins (2002), Proceedings of the 14th International Symposium, Cambridge, United Kingdom, July 14–18, pp. 941–949. [30] Stuermer, R., Hauer, B., Hall, M. and Faber, K. (2007) Asymmetric bioreduction of activated C:C bonds using enoate reductases from the old yellow enzyme family. Current Opinion in Chemical Biology, 11, 203–213. [31] Holland H.L. (1999) CH activation. Current Opinion in Chemical Biology, 3, 22–27. [32] Que, L. Jr. (2000) One motif–many different reactions. Nature Structural Biology, 7, 182–184. [33] Emerson, J.P., Farquhar, E.R. and Que, L. Jr. (2007) Structural ‘snapshots’ along reaction pathways of non-heme iron enzymes. Angewandte Chemie, International Edition, 46, 8553–8556. [34] Yun, C.-H., Yim, S.-K., Kim, D.-H. and Ahn, T. (2006) Functional expression of human cytochrome P450 enzymes in Escherichia coli. Current Drug Metabolism, 7, 411–429. [35] Celik, A., Flitsch, S.L. and Turner, N.J. (2005) Efficient terpene hydroxylation catalysts based upon P450 enzymes derived from actinomycetes. Organic & Biomolecular Chemistry, 3, 2930–2934. [36] Fujita, K.-I. and Kamataki, T. (2002) Genetically engineered bacterial cells co-expressing human cytochrome P450 with NADPH-cytochrome P450 reductase: prediction of metabolism and toxicity of drugs in humans. Drug Metabolism and Pharmacokinetics, 17, 1–22. [37] Uno, T., Nakao, A., Masuda, S. et al. (2006) Modification of small molecules by using cytochrome P450 expressed in Escherichia coli. Journal of Industrial Microbiology & Biotechnology, 33, 1043–1050. [38] Kroutil, W., Mang, H., Edegger, K. and Faber, K. (2004) Biocatalytic oxidation of primary and secondary alcohols. Advanced Synthesis & Catalysis, 346, 125–142. [39] Kumar, A.K. and Goswami, P. (2006) Functional characterization of alcohol oxidases from Aspergillus terreus MTCC 6324. Applied Microbiology and Biotechnology, 72, 906–911. [40] Overhage, J., Steinbuechel, A. and Priefert, H. (2006) Harnessing eugenol as a substrate for production of aromatic compounds with recombinant strains of Amycolatopsis sp. HR167. Journal of Biotechnology, 125, 369–376.


17

[41] Chen, Y.-R., Huang, H.-H., Cheng, Y.-F. et al. (2006) Expression of a cholesterol oxidase gene from Arthrobacter simplex in Escherichia coli and Pichia pastoris. Enzyme and Microbial Technology, 39, 854–860. [42] Adam, W., Lazarus, M., Saha-Moller, C.R. et al. (1999) Biotransformations with peroxidases, in Biotransformations (ed. K. Faber) Advances in Biochemical Engineering/Biotechnology, vol. 63, Springer-Verlag, pp. 73–108. [43] Costa-Ferreira, M., Soares, M.B. and Maximo, C. (2006) An overview on the use of microbial and enzymatic systems for dye biotransformation. Journal of Natural Fibers, 3, 69–80. [44] Liu, H.-L., Kong, L.-Y., Takaya, Y. and Niwa, M. (2005) Biotransformation of ferulic acid into two new dihydrotrimers by Momordica charantia peroxidase. Chemical & Pharmaceutical Bulletin, 53, 816–819. [45] Fujimori, D.G. and Walsh, C.T. (2007) Whats new in enzymatic halogenations. Current Opinion in Chemical Biology, 11, 553–560. [46] Vaillancourt, F.H., Yeh, E., Vosburg, D.A. et al. (2006) Natures inventory of halogenation catalysts: oxidative strategies predominate. Chemical Reviews (Washington, DC, United States), 106, 3364–3378. [47] OHagan, D. (2006) Recent developments on the fluorinase from Streptomyces cattleya. Journal of Fluorine Chemistry, 127, 1479–1483. [48] van Pee, K.-H. and Patallo, E.P. (2006) Flavin-dependent halogenases involved in secondary metabolism in bacteria. Applied Microbiology and Biotechnology, 70, 631–641. [49] Van Pee, K.-H., Dong, C., Flecks, S. et al. (2006) Biological halogenation has moved far beyond haloperoxidases. Advances in Applied Microbiology, 59, 127–157. [50] OReilly, C. and Turner, P.D. (2003) The nitrilase family of CN hydrolyzing enzymes – a comparative study. Journal of Applied Microbiology, 95, 1161–1174. [51] Singh, R., Sharma, R., Tewari, N. et al. (2006) Nitrilase and its application as a ‘green’ catalyst. Chemistry & Biodiversity, 3, 1279–1287. [52] He, Y.-C., Xu, J.-H., Xu, Y. et al. (2007) Biocatalytic synthesis of (R)-()-mandelic acid from racemic mandelonitrile by a newly isolated nitrilase-producer Alcaligenes sp. ECU0401 Chinese Chemical Letters 18, 677–680. [53] DiCosimo, R. (2007) Nitrilases and nitrile hydratases, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press LLC, Boca Raton, FL, pp. 1–2. [54] Banerjee, A., Sharma, R. and Banerjee, U.C. (2002) The nitrile-degrading enzymes: current status and future prospects. Applied Microbiology and Biotechnology, 60, 33–44. [55] Kobayashi, M. and Shimizu, S. (1998) Metalloenzyme nitrile hydratase: structure, regulation, and application to biotechnology. Nature Biotechnology, 16, 733–736. [56] Harrop, T.C. and Mascharak, P.K. (2004) Fe(III) and Co(III) centers with carboxamido nitrogen and modified sulfur coordination: lessons learned from nitrile hydratase. Accounts of Chemical Research, 37, 253–260. [57] Tang, Y.-F., Xu, J.-H., Ye, Q. and Schulze, B. (2001) Biocatalytic preparation of (S)-phenyl glycidyl ether using newly isolated Bacillus megaterium ECU1001. Journal of Molecular Catalysis B: Enzymatic, 13, 61–68. [58] Gong, P.-F. and Xu, J.-H. (2005) Bio-resolution of a chiral epoxide using whole cells of Bacillus megaterium ECU1001 in a biphasic system. Enzyme and Microbial Technology, 36, 252–257. [59] Xu, W., Xu, J.-H., Pan, J. et al. (2006) Enantioconvergent hydrolysis of styrene epoxides by newly discovered epoxide hydrolases in mung bean. Organic Letters, 8, 1737–1740. [60] Smit, M.S. and Labuschagne, M. (2006) Diversity of epoxide hydrolase biocatalysts. Current Organic Chemistry, 10, 1145–1161. [61] Faber, K. and Orru, V.A. (2002) Hydrolysis of epoxides, in Enzyme Catalysis in Organic Synthesis, 2nd edn, vol. 2 (eds K. Drauz and H. Waldmann), Wiley–VCH, pp. 579–608. [62] Lee, E.Y. and Shuler, M.L. (2007) Molecular engineering of epoxide hydrolase and its application to asymmetric and enantioconvergent hydrolysis. Biotechnology and Bioengineering, 98, 318–327. [63] Arand, M., Cronin, A., Adamska, M. and Oesch, F. (2005) Epoxide hydrolases: structure, function, mechanism, and assay, Methods in Enzymology, 400, 569–588. [64] Hwang, B.-Y. and Kim, B.-G. (2004) High-throughput screening method for the identification of active and enantioselective w-transaminases. Enzyme and Microbial Technology, 34, 429–436. [65] Rozzell, J.D. and Bommarius, A.S., (2002) Transaminations, in Enzyme Catalysis in Organic Synthesis, 2nd edn, vol. 2 (eds K. Drauz and H. Waldmann), Wiley–VCH, pp. 873–893. [66] Stewart, J.D. (2001) Dehydrogenases and transaminases in asymmetric synthesis. Current Opinion in Chemical Biology, 5, 120–129.

18


[67] Matcham, G., Bhatia, M., Lang, W. et al. (1999) Enzyme and reaction engineering in biocatalysis: synthesis of (S)-methoxyisopropylamine. Chimia, 53, 584–589. [68] Purkarthofer, T., Skranc, W., Schuster, C. and Griengl, H. (2007) Potential and capabilities of hydroxynitrile lyases as biocatalysts in the chemical industry. Applied Microbiology and Biotechnology, 76, 309–320. [69] Sharma, M., Sharma, N. and Bhalla, C. (2005) Hydroxynitrile lyases: at the interface of biology and chemistry. Enzyme and Microbial Technology, 37, 279–294. [70] Effenberger, F., Forster, S. and Wajant, H. (2000) Hydroxynitrile lyases in stereoselective catalysis. Current Opinion in Biotechnology, 11, 532–539. [71] Silvestri, G., Desantis, G., Mitchell, M. and Wong, C.-H. (2003) Asymmetric aldol reactions using aldolases. Topics in Stereochemistry, 23, 267–342. [72] Wong, C.-H. and Greenberg, W.A. (2007) Asymmetric synthesis using deoxyribose-5-phosphate aldolase. Asymmetric Synthesis, 217–221. [73] Fessner, W.-D. and Jennewein, S. (2007) Biotechnological applications of aldolases, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press LLC, Boca Raton, FL, pp. 363–400. [74] Fessner, W.-D. (2004) Enzyme-catalyzed aldol additions, in Modern Aldol Reactions, vol. 1 (ed. R. Mahrwald), Wiley–VCH, Weinheim, pp. 201–272. [75] Dean, S.M., Greenberg, W.A. and Wong, C.-H. (2007) Recent advances in aldolase-catalyzed asymmetric synthesis. Advanced Synthesis & Catalysis, 349, 1308–1320. [76] Faijes, M. and Planas, A. (2007) In vitro synthesis of artificial polysaccharides by glycosidases and glycosynthases. Carbohydrate Research, 342, 1581–1594. [77] Lu, W.-Y., Lin, G.-Q., Yu, H.-L. et al. (2007) Facile synthesis of alkyl b-D-glucopyranosides from D-glucose and the corresponding alcohols using fruit seed meals. Journal of Molecular Catalysis B: Enzymatic, 44, 72–77. [78] Yu, H.-L., Xu, J.-H., Lu, W.-Y. and Lin, G.-Q. (2008) Environmentally benign synthesis of natural glycosides using apple seed meal as green and robust biocatalyst. Journal of Biotechnology, 133, 469–477. [79] Yu, H.-L., Xu, J.-H., Wang, Y.-X. et al. (2008) Assembly of a three-dimensional array of glycoconjugates by combinatorial biocatalysis in nonaqueous media. Journal of Combinatorial Chemistry, 10, 79–87. [80] Rowan, A.S. and Hamilton, C.J. (2006) Recent developments in preparative enzymatic syntheses of carbohydrates. Natural Product Reports, 23, 412–443. [81] Moracci, M., Cobucci-Ponzano, B., Perugino, G. et al. (2005) Recent developments in the synthesis of oligosaccharides by hyperthermophilic glycosidases, in Handbook of Carbohydrate Engineering (ed. K.J. Yarema), CRC Press LLC, Boca Raton, FL, pp. 587–612. [82] Nishio, T., Hakamata, W., Ogawa, M. et al. (2005) Investigations of a useful a-glycosidase for the enzymatic synthesis of rare sugar oligosaccharides. Journal of Applied Glycoscience, 52, 153–160. [83] Murata, T. and Usui, T. (2006) Enzymatic synthesis of oligosaccharides and neoglycoconjugates. Bioscience, Biotechnology, and Biochemistry, 70, 1049–1059. [84] Rowan, A.S. and Hamilton, C.J. (2006) Recent developments in preparative enzymatic syntheses of carbohydrates. Natural Product Reports, 23, 412–443. [85] Brik, A. and Wong, C.-H. (2007) Sugar-assisted ligation for the synthesis of glycopeptides. Chemistry – A European Journal, 13, 5670–5675. [86] Trincone, A. and Giordano, A. (2006) Glycosyl hydrolases and glycosyltransferases in the synthesis of oligosaccharides. Current Organic Chemistry, 10, 1163–1193. [87] Rabbani, S., Schwardt, O. and Ernst, B. (2006) Glycosyltransferases: an efficient tool for the enzymatic synthesis of oligosaccharides and derivatives as well as mimetics thereof. Chimia, 60, 23–27. [88] Rupprath, C., Kopp, M., Hirtz, D. et al. (2007) An enzyme module system for in situ regeneration of deoxythymidine 50 -diphosphate (dTDP)-activated deoxy sugars. Advanced Synthesis & Catalysis, 349, 1489–1496. [89] Chen, X., Fang, J., Zhang, J. et al. (2001) Sugar nucleotide regeneration beads (superbeads): a versatile tool for the practical synthesis of oligosaccharides. Journal of the American Chemical Society, 123, 2081–2082. [90] Bulter, T. and Elling, L. (1999) Enzymatic synthesis of nucleotide sugars. Glycoconjugate Journal, 16, 147–159. [91] Blanchard, S. and Thorson, J.S. (2006) Enzymatic tools for engineering natural product glycosylation. Current Opinion in Chemical Biology, 10, 263–271.


19

[92] Borisova, S., Zhang, C., Takahashi, H. et al. (2006) Substrate specificity of the macrolide-glycosylating enzyme pair DesVII/DesVIII: opportunities, limitations, and mechanistic hypotheses. Angewandte Chemie, International Edition, 45, 2748–2753. [93] Ran, N., Zhao, T., Chen, Z. and Tao, J. (2008) Recent applications of biocatalysis in developing green chemistry for chemical synthesis at industrial scale. Green Chemistry, 10, 361–372. [94] Anastas, P.A. and Warner, J.C. (1998) Green Chemistry: Theory and Practice, Oxford University Press, New York. [95] Tucker, J.L. (2006) Green chemistry, a pharmaceutical perspective. Organic Process Research Development, 10, 315–319. [96] Tao, J., Zhao, L. and Ran, N. (2007) Recent advances in developing chemoenzymatic processes for active pharmaceutical ingredients. Organic Process Research & Development, 11, 259–267.

2 Expression Hosts for Enzyme Discovery and Production Aleksandra Andryushkova and Anton Glieder Graz University of Technology, Austria

2.1 Introduction Enzymes are useful catalysts for a broad diversity of chemical reactions that enable the synthesis of natural and unnatural highly pure pharmaceutically active compounds. However, proteins themselves can also be the pharmaceutical ingredients. The discovery of proteins being directly used as pharmaceuticals or indirectly by catalyzing the syntheses of pharmaceutically active small compounds is one issue in protein expression. But how can we obtain enough of such treasuries and also in sufficient quality for industrial or pharmaceutical applications? In the early days of biocatalysis only native enzymes that were easily available from fungal and bacterial strains were used. Enzymes such as proteases and lipases that were originally produced for other purposes, such as ingredients for washing powder or for food and feed processing, were employed in catalytic routes for the production of chiral compounds. However, although cheaply available, these enzymes often do not fulfil the needs for efficient industrial bioconversions of unnatural substrates. Mutants can be designed to adapt natural enzymes to the environment of industrial processes, and heterologous hosts provide reliable systems for their production. Efficient overexpression is also needed to get access to enzymes which in nature are only expressed in small amounts or spatially or temporarily regulated in specialized tissues. There is a broad diversity of heterologous hosts for gene expression, and there is no ‘best’ or ‘universal’ host which is suitable for the production of all possible proteins. Nevertheless, many proteins can be expressed cheaply and successfully in Escherichia coli, which is still the most preferred host for heterologous production of proteins without posttranslational modifications. Other proteins, such as those needing posttranslational modification for their correct folding or activity, still depend on alternative expression systems. In addition, hosts which are suitable


22


for high-throughput screening and robotization are usually not the optimal hosts for the production of tricky proteins requiring complicated co- and post-translational mechanisms for expression. In many cases success still cannot be predicted, and trial and error decides success in protein expression. A broad toolbox of different and simple expression systems can help to increase the success rate for protein expression within a reasonable time and often provides quicker and simpler solutions than adopting E. coli and other frequently used expression hosts for the expression of every individual new target enzyme. Also, the range and diversity of the proteins with very different applications in the pharmaceutical industry is growing geometrically. Depending on their application, the priorities in protein production are usually either in productivity or quality. In the case of enzymes for fine chemical production, high-level expression of correctly folded, stable enzymes and low enzymatic background due to intrinsic enzymes of the host are most important, while preparations of pharmaceutical proteins need to be uniform with low microheterogeneities caused by the expression system or the production process. In the case of enzyme discovery, where ‘yes or no’ answers from screenings are sufficiently satisfying, the first expression of native (i.e. unmodified) gene sequences to get active enzymes is more important than a maximal yield of protein in highest quality. Traditional hosts are often used not because they are best, but because they were the first to be available and characterized on a molecular level or the first to be approved for pharmaceutical production processes. On the one hand, such traditional and well-characterized expression hosts like E. coli get engineered to expand their capabilities for the expression of complex proteins; on the other hand, the vast variety of the established and newly arriving systems is steadily growing.

2.2 How to Choose an Expression System The first, and maybe most important, step to a successful heterologous expression is to classify the target protein due to original native host, the location of the target protein in the original natural host, the proteins structural features and to compare the target with other similar examples (Figure 2.1). The location in the natural host also gives some indications which redox conditions are preferable for the target protein. Roughly, proteins can be divided into two groups: . .

proteins which do not require posttranslational modifications and special targeting; proteins which require posttranslational modifications to be active, such as the most prominent example – N-glycosylation.

Glycosylation greatly affects folding [1], biological activity [2,3], functions [4], clearance from blood circulation [5] and antigenicity of the proteins [6]. More than 60% of the SWISS-PROT entries contain N-glycosylation sequon (Asn-Xxx-Ser/Thr). Glycosylation is one of the principal co- and post-translational modifications of the protein. There are two types of glycosylation: Nlinked (when a sugar chain is added to amide nitrogen of asparagine side chains) and O-linked (when it is attached to the side chains of serine or threonine). N-Glycans are usually added to asparagines of the Asn-Xxx-Ser/Thr consensus sequence. But there is usually no consensus sequence for O-glycosylation. In the end, the glycoprotein normally ends up in a range of different glycosylation forms, with varying activity/stability properties. The exact content of the glycans depends on the developmental state of the cell, on its health and the nutritional conditions [7].

E.coli BL21 B.megaterium B.subtilis WB800 B. B brevis 31-OK 31 OK P. fluorescens S.carnosus

Figure 2.1

E.coli Origami E.coli JM83 P. fluorescens E.coli + DsbC

Y.lipolytica A. A adeninivorans

Conditions tolerant

P.pastoris S.pombe H.polymorpha

Membrain/ Complex/ Soluble

P.pastoris YSH strains Insect cells Mammalian cultures

P.pastoris S.pombe H.polymorpha S.cerevisiae C lucknowense

P.pastoris S .pombe H.polymorpha Aspergillus sp T.reesi

Good secreetion

Human-like glycosylation

High throughput

Eukaryotic host

Very complex enzyme

Insect cells Plants Mammalian cultures lt

Membrain/ Complex/ Soluble

Requires posttranslational modifications for activity

Eukaryotic

Decision tree for the choice of a preferable expression host

E.coli Rosetta-gami E.coli BL21CodonPlus

Eukaryotic gene

Labeling E.coli B834 Anabaena sp Cell free systems

Low GC

Large scale

Complex enzyme Yeasts and filamentous fungi

Does not require posttranslational modifications for activity

Protein origin

content E.coli L.lactis Bacilli P.fluorescens Streptomyces sp

Disulfid bonds

E.coli C41 E.coli C43 L.lactis S.carnosus P.fluorescens

Low proteolysis

Toxic

B.megaterium B.subtilis C.crescentus L.lactis S.carnosus

Membrain

Bacterial host

Bacterial

Decision chart

Expression Hosts for Enzyme Discovery and Production 23

24


For the first group (i.e. intracellular soluble enzymes and proteins), which need no posttranslational modification and complex domain organization influencing protein folding, E. coli is the most preferred choice. However, for the other targets, alternative expression systems often provide a higher rate of success. The most common expression systems are presented in this chapter. The right cloning strategy can also save a lot of time and effort [8]. Recently developed strategies like the ligation-free cloning Gateway system (Invitrogen) or Creator System (ClonTech) allow a quick and efficient shuttling between different expression vectors and are potentially applicable for any host. Also, vectors that allow expression of the gene of interest in several hosts without recloning are serving the same purpose. Some bacterial plasmids can work in other bacteria; for example, staphylococcal pUB110 and pE194 also function in Bacillus subtilis [9]. Steinborn et al. [10] describe such a vector system for several yeasts, and two similar functioning vectors were constructed for expression in both insect cells and mammalian cell cultures [11,12].

2.3 Prokaryotic Expression Systems There is a great variety of available bacterial expression systems, with E. coli still being the most popular (Table 2.1) [13]. They share the advantages of simplicity in handling from the laboratory up to the production plant. Extensive knowledge about genetics exists for most of them, and they grow fast and on cheap culture media. Generally, they are also applicable in high-throughput screening [14,15]. Besides many other (mostly more simple) enzymes, highthroughput expression and screening of enzyme variants in E. coli was, for example, used for the extensive engineering of P450 BM3 for biotechnological applications, including fine chemical synthesis and production of human metabolites of drugs [16]. High concentrations of heterologous protein can be obtained even in shake flask cultures. Challenging problems in large-scale production employing high cell densities were solved by strain engineering and bioprocess optimization [17].

Table 2.1 Transformation efficiency, presented as the number of colonies/mg plasmid DNA or as a percentage of the positive transformants of all cells transformed/transfected Host system

Transformation efficiency Time from gene to protein

Prokaryotes

Eukaryotes

Cellfree systems

E. coli

B. subtilis

Pseudomonas sp.

Yeasts

Insect cells

Mammalian cells

Plants

108–1010

104–105

107 [91]

0.5–12% [95]

2–12% [96]

1% [97]

—

[89] 7–9 days

[90] 7–9 days

7–9 days

105–107 (0.1%) [92–94] 2 weeks

7–8 weeks

8–10 weeks

16 weeks

2–6 days

Expression Hosts for Enzyme Discovery and Production

25

2.3.1 Posttranslational Modification in Prokaryotes Even prokaryotes can perform many different posttranslational modifications, though through different mechanisms than eukaryotes. Cleavage of N-terminal signal sequences and acylation, or even the modification of amino acid side-chains, for example, by oxidation can be performed [18–20]. For nitrile hydratase from Rhodococcus sp., oxidation of cysteine is even essential for catalytic activity. Since prokaryotic organisms have no endoplasmic reticulum and Golgi apparatus, they were generally supposed to perform no glycosylation. Also, as most of the eukaryotic glycosyl transferases are glycoproteins themselves, their genes cannot be transferred into bacterial hosts to provide active glycosylation machinery. Nevertheless, recent advances in this field show that some bacteria (mostly Archeabacteria and Eubacteria) not only glycosylate their proteins [21] but are also even able to produce quite complex structures (e.g. human-like sugar structures) [22]. Like eukaryotes, they can produce N- and O-linked glycans, though the variety of the sugar units is much wider than in eukaryotes and the final structures differ vastly from those of eukaryotes. Nevertheless, the consequences of that can be extremely interesting, as traditional bacterial expression hosts (like E. coli) might theoretically be engineered to produce glycans of eukaryotic nature, or glycosylation could be done by a biocatalytic approach in vitro using bacterial glycosyltransferases. The first steps in this direction have already been made [23]. Their ability to add untypical sugar units can perhaps be used for economically efficient production of unusually glycosylated proteins with new interesting properties [24].

2.3.2 Escherichia coli This traditional host is the prokaryotic expression platform used most. It was the first to be used for the expression of heterologous proteins and also was the first for pharmaceutical protein production, namely Eli Lillys human insulin (1982). Many industrially and pharmaceutically relevant proteins were expressed with yields of around 5–10 g L1 media using high celldensity cultures (above 100 g dry cell weight per liter media) and applying new strategies of strain improvements, listed in a very interesting review from Choi and Keum [25]. A number of reviews exist regarding detailed information on fundamental genetic features of the host relevant for heterologous expression [26], on new or improved expression strategies [27] and on additional considerations for the production of pharmaceutical proteins [28]. On a laboratory scale, the most commonly used E. coli strains are BL21 (protease deficient) and K12 (seems to show enhanced disulfide bond formation) and some optimized derivatives [13]: for example, C41 and C43 for membrane proteins expression, Rosetta strains and BL21 CodonPlus strains with overexpressed tRNAs to deal with rare E. coli codons, or JM 83 for proteins secreted into the periplasm. Strains and plasmids are commercially available and were reviewed in several papers [13,26]. The high space–time yields are the result of a doubling time of only 30 min and its applicability for high cell-density cultures. However, it is hardly possible to excrete overexpressed proteins into cultivation media. In addition, accumulation of pyrogenic lipopolysaccharides in its outer membrane (a distinctive feature of Gram-negative bacteria) make additional purification steps necessary if pharmaceutical proteins are produced by E. coli [29]. Even if the desired protein does not need posttranslational modifications, expression can meet obstacles. Long experience in expression using E. coli resulted in a list of the most

26


frequent problems and possible ways to overcome them [27]. Many expression problems in E. coli are the result of differences in codon usage of E. coli and the original host which usually results in no expression or truncated protein. Overexpression of genes for rare tRNAs can help, but does not reduce difficulties due to extremely high or low guanine–cytosine (GC) content or unfavorable mRNA secondary structures. Codon optimization, and consequently the use of synthetic genes, can help in this respect. E. coli, like other Gram-negative bacteria, has an outer membrane which hampers excretion of proteins to the culture media. Thus, expressed proteins can remain in the cytoplasm or can be directed into the periplasm employing the N-terminus fusion of a signal peptide (e.g. OmpA, pelB, OmpF, PhoA, Tat signal peptides) [30]. Expression in the cytosol usually results in higher yields; but, due to overexpression, heterologous protein tends to form inclusion bodies – insoluble constructs which are easy to isolate from the whole-cell lysate. Mostly they are protease resistant. Expression of toxic proteins as inclusion bodies can also be an advantage, as such proteins are usually not active, thus avoiding problems which are due to cytotoxicity of the target protein. However, complicated, space- and time-demanding refolding procedures are necessary to regain the proteins activity. Often, refolding of the fully intact protein is even impossible. A broad set of engineered hosts like E. coli strains with modified cytoplasmic redox potential for disulfide bridge formation, strains overexpressing chaperones and foldases significantly contributed to a high success rate for gene expression in this well-characterized bacterium [31]. Procedures for refolding of proteins from inclusion bodies were also described [32]. Furthermore, although interesting for pharmaceutical proteins, refolding is too complicated and costly for most enzymes employed for fine chemical production. Disulfide bonds are hardly formed in cytosol of unmodified E. coli strains. Such enzymes can be expressed in E. coli ‘Origami’ strain, in cells coexpressing helper proteins, such as PDI or DsbC, or they can be directed to the more oxidative periplasm. Periplasmatic expression has even more advantages for protein expression [33]: it also simplifies downstream processing, N-terminal processing and correct folding, and can reduce proteolysis. Also, combinations like chaperone coexpression and export to the periplasm using, for example, the twin-arginine translocation (TAT) secretion system, which is able to secrete folded proteins, can be very successful [34]. Nevertheless, incomplete translocation across the inner membrane and existence still of proteolytic degradation can lower the yield significantly. Once proteins are transferred into the periplasm they can also be excreted into cultivation media (but with further reduction of the yield). This mainly happens by leakage, which can be enhanced by osmotic shock or cell wall permeabilization (by mutations or enzymatically). However, in some cases passive excretion does not work. There are also strategies with transplantation of active transport proteins for one or both of the membranes [35], but so far they are not that reliable. Thus far, many of the problems described above were solved by simply changing the host organism.

2.3.3 Bacilli The Gram-positive bacteria Bacillus subtilis, Bacillus megaterium and Bacillus brevis are also used for heterologous expression. Terpe [13] presented a table with some pharmaceutically and industrially relevant proteins expressed with this host. It is the most used prokaryotic host after


27

E. coli; for example, it is used for the production of enzymes as additives in detergent solutions. Proteins are secreted into cultivation media and are usually biologically active. B. subtilis is the Bacillus species used most, and also the best characterized host [36,37]. A problem with high proteolysis of secreted proteins was solved by constructing the sacB–sacY sucrose-inducible expression system [38] and developing the six protease-deficient strain WB600 or even the eight protease-deficient strain WB800 [39]. Yields can be up to 3 g L1 media [40]. B. megaterium, in contrast to B. subtilis, has low protease activity. It is able to grow on a wide variety of substrates. Plasmids with xylose-inducible promoter are most frequently used for high-level expression [41]. It was successfully used to express toxins and other difficult proteins where the proteins were intact after secretion into the media. Yields of about 30% of total soluble protein were obtained in some cases [42]. B. brevis is not so well studied. It also shows low extracellular protease activity, and a protease-deficient strain is available [43]. High [44] and low [45] copy number plasmids are constructed for different levels of expression. Modification of signal sequences can enhance yields of eukaryotic proteins [46]. The yields can be up to 3 g L1 media. Compared with E. coli, there are several characteristic advantages for all three Bacillus species mentioned: being Gram-positive, Bacillus has no outer membrane and so secreted proteins get transferred directly into the cultivation media, and no lipopolysaccharide contamination of the product occurs. Furthermore, Bacillus is ‘generally recognized as safe’ (GRAS) [36], its genetics are well studied and high cell-density cultivation is possible [47]. Bacillus is also an attractive alternative to E. coli not only for enzyme production, but also for enzyme discovery (e.g. by metagenomic approaches). Nevertheless, bacilli also show some disadvantages, such as the generally lower yield than with E. coli; for example, human interferon a or ScFv [13,25], lower transformation efficiencies and more complicated transformation protocols, and sometimes also instability of plasmids.

2.3.4 Pseudomonas fluorescens Owing to its special characteristics, Pseudomonas fluorescens became a serious concurrent to E. coli. It was established by DowPharma (www.dowpharma.com) to a robust expression host for non-glycosylated pharmaceutical proteins and biocatalysts which require complex folding. For a dodecameric nitrilase with a molecular weight of about 400 kDa, DowpharmaSM reported production of soluble and active enzyme with yields of 25 g L1 (more than 50% of the total cell weight), this being more than ever reported for E. coli [48]. Also, classical examples like INF-g (active fraction 4 g L1), hGH (5 g L1) and single-chain antibodies ScFvs (active fraction 3–5 g L1) were produced in soluble form, whereas E. coli made inclusion bodies. P. fluorescens is Gram negative, just as E. coli, and nonpathogenic for plants and mammals. The strain used for protein expression by DowPharma was derived from P. fluorescens biovar I strain MB101 [49]. Most of the wide range of low and high copy number plasmids are based on the classical vectors RSF1010, pBR322 and pPS10 [50]. They use no antibiotic markers, since they were developed for pharmaceutical production. P. fluorescens has well-developed mechanisms for the secretion of proteins into the periplasm which facilitates SS bond formation and proper N-terminal processing. It also allows one to reduce the formation of inclusion bodies and, thus, the additional costs caused by refolding processes. Proteolytic degradation of the expressed protein is also low, and very high

28


cell densities (final optical density measured at 600 nm OD600 > 400, about 100 g L1 dry cell weight) in the fermentation were obtained. Nevertheless, like for E. coli, for the isolation of protein from the periplasm and its further purification, intensive downstream processing is necessary.

2.3.5 Other Prokaryotic Expression Systems E. coli, Bacillus and Pseudomonas present quite a solid base for the production of nonglycosylated proteins. Nevertheless, a number of other prokaryotic expression hosts exist that are not as well established and which show features that are not present in the major expression organisms, and this could be extremely useful for ‘special case’ proteins. Although being Gram-negative, Caulobacter crescentus, in contrast to E. coli, demonstrated extremely good secretion capacity [51], which is perfect for the expression of exoenzymes or toxic proteins. It has a paracrystalline protein surface layer covering the outer membrane, and its secretion signal is used for protein direction into the media. High and low copy number plasmids were established [52]. Vaccine candidate proteins were overexpressed and tested in a mouse model with encouraging results [53]. Streptomyces strains are Gram positive [54], they have good secretion capacities and extensive fermentation knowledge has been accumulated. Mostly, they were used for the production of secondary metabolites with potent biological activities, such as antibiotics, immunosuppressors or pesticides. Constitutive [55] and inducible [56] expression is possible. Up to 40% of the total soluble cellular protein was reported in the case of inducible expression. Lactococcus lactis is also Gram positive and traditionally used in food fermentations and, thus, perfect for food-grade enzymes. The system was optimized for industrial applications [57] and even for pilot production of pharmaceutical products [57]. Lactic acid bacteria are generally regarded as safe and well established for the expression of proteins for fundamental studies. They showed advantages for membrane protein production, the expression of low GC genes, and genes coding for toxic proteins [57]. However, no protocols for high-density culture fermentation are available so far for this host. Staphylococcus carnosus is genetically highly stable, a good secretor (Gram positive) able to translocate proteins containing several hydrophobic transmembrane regions [58], and it has no extracellular proteases, which makes it suitable for production of the secreted enzymes [58]. Ralstonia eutropha, a Gram-negative bacterium that has the special feature of high-level expression without inclusion body formation, can also be regarded as a competitor to E. coli [59]. Enzymes from extremophiles, which are highly stable at high temperature, low pH or high/ low salt, are extremely interesting for biocatalysis, as harsh conditions allow one to avoid background reactions or enable ‘one-pot’ processes where the product of an enzymatic reaction is chemically modified in the same vessel without laborious purification steps. Such enzymes often need natural conditions (e.g. high temperature) or the presence of their native cofactors for correct folding. That is why there is a high interest in thermophilic organisms as hosts for gene expression, and one of them, the archaeon Sulfolobus solfataricus, is already quite well established [60]. Other Gram-positive food-grade bacteria, such as the corynebacterium Corynebacterium glutamicum, may soon become attractive hosts for heterologous protein production due to low


29

protease activity and well-known molecular biology, including systems biology [61]. Traditionally, C. glutamicum is used for the production of amino acids like glutamate (flavoring agent, 700 000 t year1) and lysine (food and feed additive, 300 000 t year1). Recent success in establishing flux to isoleucine [62] showed that it can be potentially engineered for production of other amino acids. At the moment, most of these more exotic bacteria are mainly used in expression for fundamental studies. Considering the speed of developments in biotechnology, their application for industrial or pharmaceutical protein production might become even more important very soon.

2.4 Eukaryotic Expression Systems Yeasts, filamentous fungi, insect cell cultures, plants and mammalian cell cultures are wellestablished eukaryotic hosts. All of them are able to provide complex folding and posttranslational modifications of proteins, and they enable the production of both secreted and intracellular proteins. The protein secretion machinery is very similar in all these organisms. N-terminal processing in the heterologous host is feasible, although product heterogeneity is quite frequent. Being evolutionarily significantly apart from each other, they show variations in posttranslational protein modification. For example, only mammalian cell cultures and plants are able to perform g-carboxylation – a posttranslational modification required, for example, for bone metabolism and blood coagulation proteins. Generated g-carboxyglutamate residues generated are usually involved in binding calcium and are essential for the biological activity of all known Gla-proteins [63]. Also, product glycosylation differs significantly between different eukaryotic hosts, both in quality and quantity. Some disadvantages are common for most eukaryotic expression systems: lower growth rates, lower transformation efficiency, and often more expensive cultivation media compared with bacterial systems.

2.4.1 Yeasts Yeasts are so-called lower eukaryotes and became an absolutely unique system, combining the strength of eukaryotes to produce complex posttranslational modifications and all the advantages of prokaryotes, like: (1) short time frames from development to production of active and pure protein (Table 2.1); (2) uncomplicated handling and cheap media; (3) simple purification – proteins usually can be efficiently secreted; (4) simple scale-up, including high cell-density cultures; (5) suitability for automation and high-throughput screening; (6) generally high yields – for example, more than 10 g L1 secreted protein by high cell-density cultures [64]. Unlike prokaryotes, however, they are not inclined to accumulate lipopolysaccharides; and in contrast to mammalian cell cultures, there is no risk of product contamination by oncogenes or animal viruses. They also provide the highly conserved eukaryotic secretory pathway and can also be very useful for the expression of membrane proteins for structure– function analysis [65]. The recent impressive work from Hamilton and Gerngross [66] about controlled humanization of N-glycosylation opened up very attractive perspectives for yeasts for large-scale production of eukaryotic fully humanized sialylated glycoproteins. That makes them serious

30


competitors for higher eukaryotic expression systems and offers the chance for a dramatic simplification of pharmaceutical protein production and increase in space–time yields. Yeast–bacteria shuttle plasmids are usually able to be maintained in both E. coli and yeast and can be episomal or integrated into the host genome. First, plasmid DNA is usually amplified in E. coli before yeast gets transformed. Different antibiotic resistance cassettes are available, and also an abundance of autotrophic markers was established. Traditional and well-established yeast species are Saccharomyces cerevisiae, Hansenula polymorpha, Klyveromyces lactis, Pichia pastoris and Schizosaccharomyces pombe. With every year that passes they are increasingly being used in industrial and pharmaceutical enzyme production on a large scale. Many further yeasts present interesting features (e.g. Arxula adeninivorans and Yarrowia lipolytica), but are not that widely used. Typical glycosylation by yeasts [67] is of the high mannose type (up to 10 mannose residues), or hypermannosylation when up to 200 mannose residues are added (S. cerevisiae). This does not occur in humans. Yeasts are generally inclined to hyperglycosylate secreted proteins because in this way they produce proteins for their cell wall, but this is most pronounced in S. cerevisiae [68] and less so in P. pastoris [69] and S. pombe [67]. They do not add sialic acid, which is a typical finishing glycan decoration in human cells. Glycoproteins with terminal mannose are not very suitable for therapeutic purposes as they are quickly filtered from blood through the mammalian macrophage Man-receptor system. However, glycosylation can be essential to obtain robust industrial enzymes by heterologous expression [2]. Some yeasts are able to add other monosaccharides, like galactose (S. pombe), but with unusual bonds [70], and they also tend to phosphorylate their glycans everywhere to obtain an additional charge on proteins [71]. 2.4.1.1 Saccharomyces cerevisiae Bakers yeast S. cerevisiae was used for centuries in the food industry, for decades in research of basic biological processes, and finally it was the first yeast used for expressing heterologous proteins [72]. For years it was the yeast; consequently, there are some advantages of this ‘single child’ situation: (1) it is a very well established expression system with a wide range of vectors (episomal or integrative, inducible or constitutive promoters with different strength) and strains (collection of knock-out mutants) for different kinds of expression [73]; (2) its genome was sequenced in the early days of genomics, and the molecular biology and physiology are well known; (3) there are commercially available S. cerevisiae expression kits (e.g. Invitrogens pYES2 or Clontechs pGBKT7 vectors); (4) it is well established for safe, large-scale production [74], and high-throughput technologies for strain development were established [73]. Nevertheless, compared with other yeasts, there are significant disadvantages, like low product yield and in many cases low secretion capacities. Poor plasmid stability was frequently reported and difficulties in scaling up production processes were observed. The usual transformation efficiency of about 105 colonies/mg plasmid [75] is higher than for most other eukaryotic hosts, but still quite low for production of highly representative expression or mutant libraries. Finally, commonly occurring hyperglycosylation of secreted proteins can lead to a change in the proteins properties, which is sometimes favorable but can also reduce enzymatic activity. In addition, the formation of immunogenic a-1,3-mannose linkages by yeasts is highly problematic for pharmaceutical protein production.


31

2.4.1.2 Methylotrophic Yeasts The methylotrophic yeasts P. pastoris and H. polymorpha are non-conventional yeasts which quickly became widely accepted and used due to some prominent advantages: (1) growth and production using methanol as the only carbon source is cheap; (2) their genetics are relatively well known; (3) they showed good secretion capacity; (4) are agreeable to high-throughput screening [76]; (5) in addition, high cell-density fermentations are well established, reaching up to 130 g L1 dry cell weight for P. pastoris (OD600 500); the yields are generally in the range of 0.7–2 g protein/liter media [77–79]. H. polymorpha [80] is generally referred to as safe and was widely used for pharma-protein production (cytokines, vaccines, coagulation factors) [80]. The genome sequence is known and DNA chip technologies have been established. Hansenula was developed to an expression platform which was implemented for the production of recombinant vaccines. P. pastoris was used for the production of several hundred proteins for industry and research needs [81], many of those being mammalian proteins. Recently, it was reported to produce humanized IgG in a glycoengineered strain [82]. It is normally prone to hypermannosylation of secreted proteins [67], but recently, after much engineering [83,84], a strain was developed that is able to produce uniform non-hypermannosylated glycosylation patterns and even can perform controlled, totally humanized glycosylation, including sialylation of expressed proteins [66]. This opens advantages of production of economically feasible pharmaceutical proteins on a big scale with a certain and stable biological activity, which is not possible with mammalian cultures, as they produce a profile of different glyco-variants of one protein. These strains might also be useful for the production of biocatalysts where hypermannosylation negatively influences enzyme activity or other industrially desirable features. A relatively low transformation efficiency of 103–105 colonies/mg DNA [85] and multicopy integration of expression cassettes make enzyme engineering and screening difficult. However, the first successful approaches using linear, integrative expression cassettes have been reported [86]. Besides the standard vectors for constitutive and inducible expression of genes in P. pastoris, new expression vectors and promoters with new regulatory features allowing, for example, also methanol-free inducible high-level expression were discovered or developed by mutagenesis of the mostly used alcohol oxidase promoter PAOX1 [87]. Thus, a new P. pastoris expression system became available, which is commercialized by the company VTU (AT). 2.4.1.3 Schizosaccharomyces pombe Fission yeast S. pombe is also referred to as safe – it was used for beer production in South Africa. Along with S. cerevisiae it is a very well established tool for studying basic biological processes like control of cell cycle and DNA repair processes [88], and its genome sequence is known [89]. S. pombe might be especially interesting for DNA library construction for enzyme discovery, since, compared with other yeasts [90], it is phylogenetically more closely related to higher eukaryotes. Consequently, S. pombe can often recognize leader sequences from humans [91,92] and fungi [93] and also promoters [94] from higher eukaryotes, which often are not functional in S. cerevisiae. The most interesting fact is the ability of S. pombe to recognize introns in RNA of higher eukaryotes and perform their splicing [95]. This could be extremely interesting for

32


screening genomic DNA libraries from eukaryotes. Also, its transformation efficiency is generally one order higher than S. cerevisiae and P. pastoris. This can significantly increase the library size for enzyme discovery and screening in this host. It also became one of the most important yeasts for the expression of active complex cytoplasmic membrane proteins and secreted proteins, as reviewed by Giga-Hama and Kumagai [96]. It was established for the expression of rat microsomal cytochrome P450 (CYP2C1) by Yamazaki et al. [97], human cytochrome P450 and other proteins by Aspex Asahi Glass Company (http://www.agc.co.jp/aspex/english/gijyutsu/hatsugen006.html) and for human mitochondrial CYP11B2 by Bureik et al. [92]. Since only a limited number of CYPs were functionally expressed in E. coli so far, S. pombe is an attractive host for this purpose and also for further drug metabolite production. 2.4.1.4 Other Yeasts Some yeasts are interesting because of their ability to tolerate harsh conditions, which can be relevant for industrial applications [78]. A. adeninivorans [80], for example, is highly osmoand thermo-tolerant. Y. lipolytica can grow on n-paraffins and produced high amounts of organic acids and is tolerant to organic components in cultivation media. It also provided high yields (1–2 g L1) of heterologous proteins, becoming competitive to traditional yeast hosts [98]. Kluyveromyces lactis was reported to secrete high molecular weight proteins [99], and an expression kit is available from New England Biolabs.

2.4.2 Filamentous Fungi Filamentous fungi can be cultivated in cheap media and on a large scale. Their genetic manipulation compared with higher eukaryotes is still relatively simple, and the genome sequences of the most important production hosts are known. Owing to their high capacity for protein secretion, they are well employed for commercial production of recombinant proteins [100,101]. Aspergillus species have long been used in the food industry and referred to as safe. Several companies, such as DSM, Novozymes and Genencor, use filamentous fungi such as Aspergillus niger, Aspergillus oryzae and Trichoderma reesei for large-scale production of industrial enzymes. A. niger was also reported to produce functional full-length IgG with a titer of 1 gL1. Fungi are not as well characterized as yeasts are in terms of molecular biological mechanisms, but attempts are being made to clear up the situation [102]. Their transformation efficiency is relatively low compared with yeasts (103 transformants/mg of DNA) [103], which is not enough for construction of representative DNA libraries, and parallel growth of filamentous fungi with high throughput might be difficult. Nevertheless, some success has been reported [104]. The company Dyadic and the company TNO from the Netherlands reported about possible high-throughput screening in filamentous fungi using a mutant strain showing reduced viscosity [105]. In addition, interesting industrial properties and the availability of protease-deficient strains make Chrysosporium lucknowense an interesting new fungal expression system. Glycosylation in filamentous fungi is not that well known. Mostly, it was studied for A. niger. It appears to perform both N- and O-linked glycosylation, and, as with yeasts, mostly produces long or short mannose structures. However, unlike yeasts, A. niger generally does not produce hypermannosylated proteins [106]. They also produce fungal modifications


33

which are so far not known from humans, like, for example, single added mannose, galactose, glucose residues, phosphate and sulfate groups. Also, no evidence of sialylation has been found so far. One of the possible ways to produce human-type glycosylated proteins is downstream processing. Such a process was established for T. reesei, as it is able to secrete glycoproteins which are accepted as intermediates by human glycosyltransferases in vitro and thus processed to the final product [107]. Research to humanize fungal glycosylation pathways was started [108], but has not given positives results so far. However, as in the case of P. pastoris, this also has a long way to go [83].

2.4.3 Insect/Baculovirus System Higher eukaryotic expression systems are usually used for production of small amounts of protein; for example, for analytical purposes or if a native conformation of complex enzymes has to be obtained. They can further be used to modify enzymes which were produced by lower eukaryotic or microbial hosts. Insects and insect cell cultures were widely used for expression of intracellular soluble enzymes from higher eukaryotes. About a thousand proteins have already been successfully expressed [109,110], including possible vaccines [111], proteins for gene therapy [112] and baculovirus insecticides [113]. Also, some complex enzymes used for biotransformations and metabolite production, such as cytochrome P450 enzymes [114], epoxide hydrolases [115] or phospholipase D [116], were made by employing insect cell expression systems. Expression of human cytochrome P450 enzymes in insect cells allows the production of enzymes with their authentic N-terminal end. These native-like enzymes are sometimes preferred in screenings for human drug metabolization, and commercial kits are offered by BD Biosciences. Insect cells expressing several human enzymes were even used as in vitro human liver models for drug development and evaluation [117]. Most of the viral vectors were constructed using: (1) the Autographa californica nuclear polyhedrosis virus (AcNPV), which is able to infect moth species, Spodoptera frugiperda ovarian cell lines and, in specific conditions, Drosophila cells; (2) the Bombyx mori nuclear polyhedrosis virus (BmNPV), which is able to infect silkworm cells. To broaden the range of infection of hosts, a hybrid virus was generated [118,119]. Insect cell systems represent multiple advantages compared with mammalian cell cultures: (1) they are easier to handle (Table 2.1); (2) cultivation media are usually cheaper; (3) they need only minimum safety precautions, as baculovirus is harmless for humans; (4) they provide most higher eukaryotic posttranslational modifications and heterologous eukaryotic proteins are usually obtained in their native conformation; (5) the baculovirus system is easily scalable to the bioreactor scale. However, because of the viral nature of the system, continuous fermentation for transient expression is not possible – the cells finally die. In addition, problems can occur because of differences in posttranslational modifications; for example, in glycosylation [120,121]. The degree of core mannosylation depends on culture conditions [7]. Moreover, fucosylation of proximal GlcNAc is common in both insects and plants [7], but is unusual and immunogenic for humans. However, recent success was reported where a strain was developed having normal growth rate and producing N-glycans of human nature [122] which facilitates production of higher eukaryotic proteins with proper glycosylation for pharmaceutical applications.

34


To speed up and simplify expression using insect cells, multiple improving technologies were developed: the Bac-to-Bac (Invitrogen) system allows one to shift recombinant processes from insect cells to E. coli (www.invitrogen.com). The time needed for virus preparation is thus decreased to about 3 weeks. For the expression of several genes from one vector, a bicistronic vector with two promoters (pFastBac DUAL) is available (Invitrogen). Furthermore, a system for the expression of protein complexes with many subunits was described (MultiBac), which is based on vectors, allowing an assembly of the polycistronic expression cassettes after transfection [123]. Also, special vectors allowing expression in both insect cells and mammalian cell cultures from the same vector (pMamaBac [11] and pBacMam [12]) were described, though the amount required for mammalian transfection with one of these vectors is twofold higher than for insect cells, which makes it applicable only for assessment of suitability for a certain cell culture. The FlashBAC technology (http://www.expressiontechnologies.com/flashBAC/default. asp) was introduced with a one-step procedure allowing one to avoid the plaque purification step, which greatly facilitates high-throughput production of virus with automated systems. The time needed for production of virus particles is just about 5 days. A plasmid-based transient expression system (InsectDirect system from EMD Biosciences Inc., USA; www.emdbiosciences.com) will most probably greatly facilitate parallelization and automation for insect cell cultures. It generally gives lower yields, since expression is driven by an early baculoviral promoter, but it is possible to evaluate protein activity and expression level 24 h after transfection. It is also scalable to 1 L volume. The two main disadvantages, namely the large amount of transfection agent required and the limitation in scalability, can probably be overcome in future. It can thus be concluded that, with the possibilities of parallelization and automation arising, insect cells are becoming an agreeable expression host for high-throughput screening of hardto-express and complex proteins, thus providing access to enzymes for early-stage drug development and evaluation.

2.4.4 Mammalian Cell Cultures This class of expression systems is obviously the most suitable for functional expression of foreign genes from higher eukaryotes because of the high similarity of the expression and secretion machinery in all of them. The list of successful examples is huge, containing secreted, intracellular, transmembrane, membrane-associated and other complex enzymes [124–126]. Although mostly used for biopharmaceutical production, animal cell cultures were also used for the production of enzymes for the bioconversion of small compounds. For example, enzymes that are responsible for human drug metabolization (e.g. human cytochrome P450 enzymes [127]) were expressed and are now available for human drug metabolite production and for the first studies of metabolization of new drug candidates. In addition, cell cultures are used to produce enzymes for the posttranslational modification of biopharmaceuticals in vitro [128]. In spite of several drawbacks (i.e. expensive and laborious handling procedures, low space–time yields (Table 2.1), high demand on biosafety, potential contaminations, limited applicability for continuous fermentations [129], and problems obtaining the same glycosylation profile from batch to batch [130]), mammalian cell cultures are widely used for small-scale expression and more recently even on a multi-cubic-meter scale. The system works like insect


35

cell cultures. First, the mammalian cell culture is transfected with virus particles, containing the gene of interest. They are multiplied in the first round of transfection and applied to the culture again for formation of cells producing the heterologous protein. A lot of cell lines are established, but the most commonly used are COS cells derived from African green monkey (usually used for transient expression [131]) and Chinese hamster ovary (CHO) cells (for stable expression), if the gene of interest is integrated into the genome of the host cell [132]. Other stably transfected cell lines have been established; for example, baby hamster cells, mouse L-cells, human embryonic kidney cells and different myeloma cell lines [133]. Transiently expressing cell lines give higher yields and they are easier to produce, but at the same time the chance for continuous fermentation is limited. Nevertheless, stable cell lines are able to produce recombinant enzymes with yields up to 20% of the total cell protein (adenovirus system [129,133]). Four main groups of viruses were used for stable expression: (1) Retroviridae – especially Lentiviruses, which are able to transfect both dividing and not dividing cells and compatible with ligation-free cloning (Gatewat) in ViraPower system from Invitrogen (www.invitrogen.com); (2) adenoviruses, due to double-stranded DNA allowing simple manipulation in vitro and the possibility to transfect a variety of hosts [134]; (3) vaccinia viruses, which are also suitable for different hosts and scalable up to 1000 L fermentation [135]; (4) alphaviruses showing high expression and scalability to bioreactor volumes [136], which was reported to be particularly useful for membrane protein expression [137]. For a long time, the application of mammalian cell lines in the production of proteins as tools (e.g. for assay development and screening, for structure elucidation and as antigens) was referred to as too laborious and expensive. Nevertheless, with development of large-scale transient transfection, this limitation has been significantly reduced [138]. The commercially available FreeStyle expression systems from Invitrogen (www.invitrogen.com), similar to the InsectDirect system, enable the use of plasmids instead of viral expression vectors. In contrast to several months with a classic protocol, this gives proteins within days.

2.4.5 Other Expression Systems Among other expression systems, plants or plant cell cultures [139,140] and some individual organisms [141] are interesting expression hosts. On the one hand, being incomparably slow (Table 2.1), thus requiring months from gene cloning to a produced protein, both of them have a rather hypothetical significance for enzyme discovery. On the other hand, they have a certain potential in the means of economically efficient protein production, since their cultivation, harvest and processing are usually well established and cheap (plants can be grown in fields, and animals like cows, goats or chickens do not need extensive care). Yields can be high and prolonged, and protein can be harvested in quite concentrated form. Employing a fusion to signal peptides, protein transfer to milk or eggs of animals and into plant seeds can be carried out [142–144]. Both systems mentioned are higher eukaryotic, and expressed proteins are mainly soluble, in their native folding state and, thus, showing biologic activity. Plant research groups have also reported advances in humanization of glycosylation pathways [145] for the production of therapeutic glycoproteins (e.g. antibodies) [146] and the production of industrially applied enzymes (e.g. a-amylase) [139]. However, the development times from gene to plants, including field trials, are extremely long, and seasonal variations and unforeseeable environmental influences might make economic enzyme production by transgenic plants rather unrealistic.

36


2.5 Cell-Free Expression Systems Originally, in vitro expression systems were just used for small-scale production of proteins and appeared to have significant limitations, especially the high costs and limited life-time of translation activity. But their undoubted advantages for highly controlled synthesis and the possibility to express proteins which are seriously interfering with cell physiology attracted a lot of interest; finally, most problems were partly or completely solved. Today, cell-free expression is emerging as a most interesting system, with cloning-free nature and little preference for special codon usage (good expression from high AT-rich cDNA), which provides hard to express proteins on a 5–6 mg mL1 scale and, at the same time, offering wide possibilities for labeling. All this makes cell-free expression systems highly suitable for analytical applications like high-throughput automated (http://www.cfsciences.com/), functional [147] and structural [148] proteomics, for protein microarrays [149] and all possible combinations thereof. Even upscaling to larger volumes [150] and in vitro production of biopharmaceuticals for personalized medicine has been discussed [151]. Cell-free translation is based on monomeric building blocks (amino acids), energy (ATP) and a source of information (DNA). No energy gets wasted on cell reproduction, and the absence of the cell wall allows one to avoid limitations in substrate access, transport limitations, product removal and sampling problems. Providing only one gene for translation, the target protein gets highly enriched. It was established using E. coli cell lysate [150], and the common procedure includes (1) cell growth and lysis, (2) extract preparation, (3) addition of substrates and salts, (4) template addition and (5) incubation. Owing to constant efforts, both main problems were eliminated: fine tuning of the system reduced the costs [150], and the life-time of the translation activity was significantly prolonged by the development of a continuous-flow cell-free translation method where a ‘feeding solution’ containing amino acids and energy is supplied to a translation chamber [152,153]. Some interesting improvements made the system even more attractive; for example, a pure system where the cell-free system consists exclusively of His-tagged proteins which can easily be removed from the expressed product [154]. With an automated approach, the time needed from template preparation till protein expression is assessed within 2–6 days (Table 2.1) (http:// www.cfsciences.com/). In protein engineering, cell-free translation by ribosome display or ribosome–protein fusion enabled fast enzyme breeding cycles and the employment of large variant libraries without loss of diversity due to ligation and transformation. One could also imagine employing cell-free translation to speed up engineering techniques such as alanine scanning, where each amino acid in a protein is exchanged to alanine and the protein variants are consequently analyzed. In the first examples of cell-free expression, eukaryotic proteins requiring complex folding, posttranslational modifications or multidomain proteins were often produced misfolded. The production of eukaryotic proteins often requires co-translational folding or targeting, but folding in E. coli occurs after translation, and that was limiting the applicability of cell-free translation. Supplementation with chaperones and membrane vesicles derived from endoplasmic reticulum preparations of higher eukaryotes was lightening the problem. Now, systems based on higher eukaryotic cell lysates are also used, mainly from rabbit reticulocytes, wheat embryos, yeast, tumor cells or insect cells. They can provide protein phosphorylation, adenylation, myristilation, isoprenylation and farnisylation, and even glycosylation (insect Sf-21 cells) [155,156]. The possibility of controlling these reactions is a matter of chance, since


37

the dynamic heterogeneity of complex systems of eukaryotes modifies expressed proteins to different variants to varying degrees. Most eukaryotic cell-free systems give relatively low yields. The most popular are rabbit reticulocytes and the wheat-embryo-derived system. The former is most suitable for globin synthesis (due to the specificity of reticulocytes), but for other proteins it gives very low titers (6 mg mL1 compared with some milligrams per milliliter for E. coli and wheat embryo cellfree systems) [156]. In contrast, the germ cell-free system is able to express properly folded, fully functional complex proteins (even having multiple domains) and also gives yields higher than E. coli extracts with the same costs [156]. Membrane-integrated proteins were always hard to express in cell-based systems in sufficient quantity for structural analysis. In cell-free systems, they can be produced on a milligrams per milliliter scale, which, combined with labeling with stable isotopes, is also very amenable for NMR spectroscopy [157–161]. Possible applications of in vitro expression systems also include incorporation of selenomethionine (Se-Met) into proteins for multiwavelength anomalous diffraction phasing of protein crystal structures [162]. Se-Metcontaining proteins are usually toxic for cellular systems [163]. Consequently, rational design of more efficient biocatalysts is facilitated by quick access to structural information about the enzyme.

2.6 Conclusions Expression hosts for the production of recombinant proteins are constantly being improved, gaining features which are moving them towards the mutual goal of fast and efficient bulk protein production in high quality. But, in coming to a choice of expression host, it is essential to know the key differences between them and to be able to make approximate estimations of the given time frame for each choice. The first point was discussed in previous parts of this chapter, and Table 2.1 presents a rough estimation of the necessary time from expression cassette construction to the heterologously expressed protein.

References [1] Parodi, A.J. (2000) Protein glucosylation and its role in protein folding. Annual Review of Biochemistry, 69, 69–93. [2] Guoa, M., Hang, H. and Zhua, T. (2008) Effect of glycosylation on biochemical characterization of recombinant phytase expressed in Pichia pastoris. Enzyme and Microbial Technology, 42, 340–345. [3] Weis, R., Gaisberger, R., Gruber, K. and Glieder, A. (2007) Serine scanning: a tool to prove the consequences of N-glycosylation of proteins. Journal of Biotechnology, 129 (1), 50–61. [4] Erbayraktar, S., Grasso, G., Sfacteria, A. et al. (2003) Asialoerythropoietin is a nonerythropoietic cytokine with broad neuroprotective activity in vivo. Proceedings of the National Academy of Sciences of the United States of America, 100 (11), 6741–6746. [5] Fukuda, M.N., Sasaki, H., Lopez, L. and Fukuda, M. (1989) Survival of recombinant erythropoietin in the circulation: the role of carbohydrates. Blood, 73 (1), 84–89. [6] LaTemple, D.C., Abrams, J.T., Zhang, S.Y. et al. (1999) Increased immunogenicity of tumor vaccines complexed with anti-Gal: studies in knockout mice for a1,3galactosyltransferase. Cancer Research, 59 (14), 3417–3423. [7] Zhang, F., Saarinen, M.A., Itle, L.J. et al. (2002) The effect of dissolved oxygen (DO) concentration on the glycosylation of recombinant protein produced by the insect cell-baculovirus expression system. Biotechnology and Bioengineering, 77 (2), 219–224.

38


[8] Hartley, J.L. (2006) Cloning technologies for protein expression and purification. Current Opinion in Biotechnology, 17 (4), 359–366. [9] Wu, S.C. and Wong, S.L. (1999) Development of improved pUB110-based vectors for expression and secretion studies in Bacillus subtilis. Journal of Biotechnology, 72 (1–2), 185–195. [10] Steinborn, G., B€ oer, E., Scholz, A. et al. (2006) Application of a wide-range yeast vector (CoMed) system to recombinant protein production in dimorphic Arxula adeninivorans, methylotrophic Hansenula polymorpha and other yeasts. Microbial Cell Factories, 5, 33. [11] Philipps, B., Forstner, M. and Mayr, L.M. (2005) A baculovirus expression vector system for simultaneous protein expression in insect and mammalian cells. Biotechnology Progress, 21 (3), 708–711. [12] Kost, T.A., Condreay, J.P. and Jarvis, D.L. (2005) Baculovirus as versatile vectors for protein expression in insect and mammalian cells. Nature Biotechnology, 23 (5), 567–575. [13] Terpe, K. (2006) Overview of bacterial expression systems for heterologous protein production: from molecular and biochemical fundamentals to commercial systems. Applied Microbiology and Biotechnology, 72 (2), 211–222. [14] Lin, C.T., Moore, P.A., Auberry, D.L. et al. (2006) Automated purification of recombinant proteins: combining high-throughput with high yield. Protein Expression and Purification, 47 (1), 16–24. [15] Scheich, C., Leitner, D., Sievert, V. et al. (2004) Fast identification of folded human protein domains expressed in E. coli suitable for structural analysis. BMC Structural Biology, 4, 4. [16] Fasan, R., Chen, M.M., Crook, N.C. et al. (2007) Engineered alkane-hydroxylating cytochrome P450 (BM3) exhibiting nativelike catalytic properties. Angewandte Chemie (International Edition in English), 46 (44), 8414–8418. [17] De Mey, M., Lequeux, G.J., Beauprez, J.J. et al. (2007) Comparison of different strategies to reduce acetate formation in Escherichia coli. Biotechnology Progress, 23 (5), 1053–1063. [18] Sletta, H., Aune, R., Nedal, A. et al. (2007) The presence of N-terminal secretion signal sequences leads to strong stimulation of the total expression levels of three tested medically important proteins during high-cell-density cultivations of Pichia pastoris. Applied and Environmental Microbiology, 73 (3), 906–912. [19] Okuyama, H., Yamada, K., Ikezawa, H. and Wakil, S.J. (1976) Factors affecting the acyl selectivities of acyltransferases in Escherichia coli. The Journal of Biological Chemistry, 251 (8), 2487–2492. [20] Wu, S., Fallon, R.D. and Payne, M.S. (1997) Over-production of stereoselective nitrile hydratase from Pseudomonas putida 5B in Pichia pastoris: activity requires a novel downstream protein. Applied Microbiology and Biotechnology, 48 (6), 704–708. [21] Upreti, R.K., Kumar, M. and Shankar, V. (2003) Bacterial glycoproteins: functions, biosynthesis and applications. Proteomics, 3 (4), 363–379. [22] Benz, I. and Schmidt, M.A. (2002) Never say never again: protein glycosylation in pathogenic bacteria. Molecular Microbiology, 45 (2), 267–276. [23] Wacker, M., Linton, D., Hitchen, P.G. et al. (2002) N-linked glycosylation in Campylobacter jejuni and its functional transfer into E. coli. Science, 298 (5599), 1790–1793. [24] Schaffer, C. and Messner, P. (2001) Glycobiology of surface layer proteins. Biochimie, 83 (7), 591–599. [25] Choi, D.H. and Keum, K.C. (2006) Production of recombinant proteins by high cell density culture of Pichia pastoris. Chemical Engineering Science, 61, 876–885. [26] Baneyx, F. (1999) Recombinant protein expression in Pichia pastoris. Current Opinion in Biotechnology, 10 (5), 411–421. [27] Sorensen, H.P. and Mortensen, K.K. (2005) Advanced genetic strategies for recombinant protein expression in Pichia pastoris. Journal of Biotechnology, 115 (2), 113–128. [28] Swartz, J.R. (2001) Advances in Pichia pastoris production of therapeutic proteins. Current Opinion in Biotechnology, 12 (2), 195–201. [29] Petsch, D. and Anspach, F.B. (2000) Endotoxin removal from protein solutions. Journal of Biotechnology, 76 (2–3), 97–119. [30] Choi, J.H. and Lee, S.Y. (2004) Secretory and extracellular production of recombinant proteins using Pichia pastoris. Applied Microbiology and Biotechnology, 64 (5), 625–635. [31] Sorensen, H.P. and Mortensen, K.K. (2005) Soluble expression of recombinant proteins in the cytoplasm of Pichia pastoris. Microbial Cell Factories, 4 (1), 1. [32] Misawa, S. and Kumagai, I. (1999) Refolding of therapeutic proteins produced in Pichia pastoris as inclusion bodies. Biopolymers, 51 (4), 297–307.


39

[33] Georgiou, G. and Segatori, L. (2005) Preparative expression of secreted proteins in bacteria: status report and future prospects. Current Opinion in Biotechnology, 16 (5), 538–545. [34] Angelini, S., Moreno, R., Gouffi, K. et al. (2001) Export of Thermus thermophilus alkaline phosphatase via the twin-arginine translocation pathway in Escherichia coli. FEBS Letters, 506 (2), 103–107. [35] Shokri, A., Sanden, A.M. and Larsson, G. (2003) Cell and process design for targeting of recombinant protein into the culture medium of Pichia pastoris. Applied Microbiology and Biotechnology, 60 (6), 654–664. [36] Li, W., Zhou, X. and Lu, P. (2004) Bottlenecks in the expression and secretion of heterologous proteins in Bacillus subtilis. Research in Microbiology, 155 (8), 605–610. [37] Ferreira, L.C., Ferreira, R.C. and Schumann, W. (2005) Bacillus subtilis as a tool for vaccine development: from antigen factories to delivery vectors. Anais da Academia Brasileira de Ciencias, 77 (1), 113–124. [38] Wu, X.C., Lee, W., Tran, L. and Wong, S.L. (1991) Engineering a Bacillus subtilis expression-secretion system with a strain deficient in six extracellular proteases. Journal of Bacteriology, 173 (16), 4952–4958. [39] Murashima, K., Chen, C.-L., Kosugi, A. et al. (2002) Heterologous production of Clostridium cellulovorans engB, using protease-deficient Bacillus subtilis, and preparation of active recombinant cellulosomes. Journal of Bacteriology, 184 (1), 76–81. [40] Palva, I. (1982) Molecular cloning of alpha-amylase gene from Bacillus amyloliquefaciens and its expression in B. subtilis. Gene, 19 (1), 81–87. [41] Rygus, T. and Hillen, W. (1991) Inducible high-level expression of heterologous genes in Bacillus megaterium using the regulatory elements of the xylose-utilization operon. Applied Microbiology and Biotechnology, 35 (5), 594–599. [42] England, D.F., Penfold, R.J., Delaney, S.F. and Rogers, P.L. (1997) Isolation of Bacillus megaterium mutants that produce high levels of heterologous protein, and their use to construct a highly mosquitocidal strain. Current Microbiology, 35 (2), 71–76. [43] Kajino, T., Saito, Y., Asami, O. et al. (1997) Extracellular production of an intact and biologically active human growth hormone by the Bacillus brevis system. Journal of Industrial Microbiology & Biotechnology, 19 (4), 227–231. [44] McKenzie, T., Hoshino, T., Tanaka, T. and Sueoka, N. (1986) The nucleotide sequence of pUB110: some salient features in relation to replication and its regulation. Plasmid, 15 (2), 93–103. [45] Horinouchi, S. and Weisblum, B. (1982) Nucleotide sequence and functional map of pE194, a plasmid that specifies inducible resistance to macrolide, lincosamide, and streptogramin type B antibodies. Journal of Bacteriology, 150 (2), 804–814. [46] Sagiya, Y., Yamagata, H. and Udaka, S. (1994) Direct high-level secretion into the culture medium of tuna growth hormone in biologically active form by Bacillus brevis. Applied Microbiology and Biotechnology, 42 (2–3), 358–363. [47] Nijland, R., Veening, J.W. and Kuipers, O.P. (2007) A derepression system based on the Bacillus subtilis sporulation pathway offers dynamic control of heterologous gene expression. Applied and Environmental Microbiology, 73 (7), 2390–2393. [48] Squires, C. and Talbot, H. (2004) Pseudomonas fluorescens – a robust manufacturing platform. Speciality Chemicals Magazine. 24, 25–26. [49] Landry, T.D., Chew, L., Davis, J.W. et al. (2003) Safety evaluation of an alpha-amylase enzyme preparation derived from the archaeal order Thermococcales as expressed in Pseudomonas fluorescens biovar I. Regulatory Toxicology and Pharmacology, 37 (1), 149–168. [50] Schneider, J.C., Jenings, A.F., Mun, D.M. et al. (2005) Auxotrophic markers pyrF and proC can replace antibiotic markers on protein production plasmids in high-cell-density Pseudomonas fluorescens fermentation. Biotechnology Progress, 21 (2), 343–348. [51] Bingle, W.H., Nomellini, J.F. and Smit, J. (2000) Secretion of the Caulobacter crescentus S-layer protein: further localization of the C-terminal secretion signal and its use for secretion of recombinant proteins. Journal of Bacteriology, 182 (11), 3298–3301. [52] Umelo-Njaka, E., Nomellini, J.F., Yim, H. and Smit, J. (2001) Development of small high-copy-number plasmid vectors for gene expression in Caulobacter crescentus. Plasmid, 46 (1), 37–46. [53] Umelo-Njaka, E., Nomellini, J.F., Bingle, W.H. et al. (2001) Expression and testing of Pseudomonas aeruginosa vaccine candidate proteins prepared with the Caulobacter crescentus S-layer protein expression system. Vaccine, 19 (11–12), 1406–1415. [54] Hopwood, D.A. (1999) Forty years of genetics with Streptomyces: from in vivo through in vitro to in silico. Microbiology (Reading, England), 145 (Pt 9), 2183–2202.

40


[55] Binnie, C., Cossar, J.D. and Stewart, D.I. (1997) Heterologous biopharmaceutical protein expression in Streptomyces. Trends in Biotechnology, 15 (8), 315–320. [56] Herai, S., Hashimoto, Y., Higashibata, H. et al. (2004) Hyper-inducible expression system for streptomycetes. Proceedings of the National Academy of Sciences of the United States of America, 101 (39), 14031–14035. [57] Mierau, I. and Kleerebezem, M. (2005) 10 years of the nisin-controlled gene expression system (NICE) in Lactococcus lactis. Applied Microbiology and Biotechnology, 68 (6), 705–717. [58] Hansson, M., Samuelson, P., Nguyen, T.N. and Stahl, S. (2002) General expression vectors for Staphylococcus carnosus enabled efficient production of the outer membrane protein A of Klebsiella pneumoniae. FEMS Microbiology Letters, 210 (2), 263–270. [59] Srinivasan, S., Barnard, G.C. and Gerngross, T.U. (2002) A novel high-cell-density protein expression system based on Ralstonia eutropha. Applied and Environmental Microbiology, 68 (12), 5925–5932. [60] Albers, S.V., Jonuscheit, M., Dinkelaker, S. et al. (2006) Production of recombinant and tagged proteins in the hyperthermophilic archaeon Sulfolobus solfataricus. Applied and Environmental Microbiology, 72 (1), 102–111. [61] Srivastava, P. and Deb, J.K. (2005) Gene expression systems in corynebacteria. Protein Expression and Purification, 40 (2), 221–229. [62] Morbach, S., Sahm, H. and Eggeling, L. (1996) l-Isoleucine Production with Corynebacterium glutamicum: further flux increase and limitation of export. Applied and Environmental Microbiology, 62 (12), 4345–4351. [63] Furie, B., Bouchard, B.A. and Furie, B.C. (1999) Vitamin K-dependent biosynthesis of gamma-carboxyglutamic acid. Blood, 93 (6), 1798–1808. [64] Clare, J.J., Rayment, F.B., Ballantine, S.P. et al. (1991) High-level expression of tetanus toxin fragment C in Pichia pastoris strains containing multiple tandem integrations of the gene. Bio/Technology (Nature Publishing Company), 9 (5), 455–460. [65] Bill, R.M. (2001) Yeast – a panacea for the structure–function analysis of membrane proteins? Current Genetics, 40 (3), 157–171. [66] Hamilton, S.R. and Gerngross, T.U. (2007) Glycosylation engineering in yeast: the advent of fully humanized yeast. Current Opinion in Biotechnology, 18 (5), 387–392. [67] Gemmill, T.R. and Trimble, R.B. (1999) Overview of N- and O-linked oligosaccharide structures found in various yeast species. Biochimica et Biophysica Acta, 1426 (2), 227–237. [68] Munro, S. (2001) What can yeast tell us about N-linked glycosylation in the Golgi apparatus? FEBS Letters, 498 (2–3), 223–227. [69] Trimble, R.B., Atkinson, P.H., Tschopp, J.F. et al. (1991) Structure of oligosaccharides on Saccharomyces SUC2 invertase secreted by the methylotrophic yeast Pichia pastoris. The Journal of Biological Chemistry, 266 (34), 22807–22817. [70] Ziegler, F.D., Cavanagh, J., Lubowski, C. and Trimble, R.B. (1999) Novel Schizosaccharomyces pombe Nlinked GalMan9GlcNAc isomers: role of the Golgi GMA12 galactosyltransferase in core glycan galactosylation. Glycobiology, 9 (5), 497–505. [71] Shibata, N., Kobayashi, H., Tojo, M. and Suzuki, S. (1986) Characterization of phosphomannan-protein complexes isolated from viable cells of yeast and mycelial forms of Candida albicans NIH B-792 strain by the action of Zymolyase-100T. Archives of Biochemistry and Biophysics, 251 (2), 697–708. [72] Hitzeman, R.A., Hagie, F.E., Levine, H.L. et al. (1981) Expression of a human gene for interferon in yeast. Nature, 293 (5835), 717–722. [73] Ton, V.K. and Rao, R. (2004) Functional expression of heterologous proteins in yeast: insights into Ca2 þ signaling and Ca2 þ -transporting ATPases. American Journal of Physiology. Cell Physiology, 287 (3), C580–C589. [74] Hensing, M.C., Rouwenhorst, R.J., Heijnen, J.J. et al. (1995) Physiological and technological aspects of largescale heterologous-protein production with yeasts. Antonie van Leeuwenhoek, 67 (3), 261–279. [75] Schiestl, R.H. and Gietz, R.D. (1989) High efficiency transformation of intact yeast cells using single stranded nucleic acids as a carrier. Current Genetics, 16 (5–6), 339–346. [76] Weis, R., Luiten, R., Skranc, W. et al. (2004) Reliable high-throughput screening with Pichia pastoris by limiting yeast cell death phenomena. FEMS Yeast Research, 5 (2), 179–189. [77] Cereghino, J.L. and Cregg, J.M. (2000) Heterologous protein expression in the methylotrophic yeast Pichia pastoris. FEMS Microbiology Reviews, 24 (1), 45–66.


41

[78] Domınguez, A., Ferminãn, E., Sanchez, M. et al. (1998) Non-conventional yeasts as hosts for heterologous protein production. International Microbiology, 1 (2), 131–142. [79] Mayer, A.F., Hellmuth, K., Schlieker, H. et al. (1999) An expression system matures: a highly efficient and costeffective process for phytase production by recombinant strains of Hansenula polymorpha. Biotechnology and Bioengineering, 63 (3), 373–381. [80] Gellissen, G., Kunze, G., Gaillardin, C. et al. (2005) New yeast expression platforms based on methylotrophic Hansenula polymorpha and Pichia pastoris and on dimorphic Arxula adeninivorans and Yarrowia lipolytica – a comparison. FEMS Yeast Research, 5 (11), 1079–1096. [81] Cregg, J.M., Cereghino, J.L., Shi, J. and Higgins, D.R. (2000) Recombinant protein expression in Pichia pastoris. Molecular Biotechnology, 16 (1), 23–52. [82] Li, H., Sethuraman, N., Stadheim, T.A. et al. (2006) Optimization of humanized IgGs in glycoengineered Pichia pastoris. Nature Biotechnology, 24 (2), 210–215. [83] Hamilton, S.R., Bobrowicz, P., Bobrowicz, B. et al. (2003) Production of complex human glycoproteins in yeast. Science, 301 (5637), 1244–1246. [84] Hamilton, S.R., Davidson, R.C., Sethuraman, N. et al. (2006) Humanization of yeast to produce complex terminally sialylated glycoproteins. Science, 313 (5792), 1441–1443. [85] Lin-Cereghino, J., Wong, W.W., Xiong, S. et al. (2005) Condensed protocol for competent cell preparation and transformation of the methylotrophic yeast Pichia pastoris. Biotechniques, 38 (1), 44, 46, 48. [86] Liu, J., Pscheidt, B., Avi, M. et al. (2008) Laboratory evolved biocatalysts for stereoselective syntheses of substituted benzaldehyde cyanohydrins. Chembiochem: A European Journal of Chemical Biology, 4;9 (1), 58–61. [87] Hartner, F.S. and Glieder, A. (2006) Regulation of methanol utilisation pathway genes in yeasts. Microbial Cell Factories, 5, 39. [88] Beach, D., Durkacz, B. and Nurse, P. (1982) Functionally homologous cell cycle control genes in budding and fission yeast. Nature, 300 (5894), 706–709. [89] Wood, V., Gwilliam, R., Rajandream, M.A. et al. (2002) The genome sequence of Schizosaccharomyces pombe. Nature, 415 (6874), 871–880. [90] Sankoff, D., Leduc, G., Antoine, N. et al. (1992) Gene order comparisons for phylogenetic inference: evolution of the mitochondrial genome. Proceedings of the National Academy of Sciences of the United States of America, 89 (14), 6575–6579. [91] Ikeda, S., Nikaido, K., Araki, F K. et al. (2004) Production of recombinant human lysosomal acid lipase in Schizosaccharomyces pombe: development of a fed-batch fermentation and purification process. Journal of Bioscience and Bioengineering, 98 (5), 366–373. [92] Bureik, M., Schiffler, B., Hiraoka, Y. et al. (2002) Functional expression of human mitochondrial CYP11B2 in fission yeast and identification of a new internal electron transfer protein, etp1. Biochemistry, 41 (7), 2311–2321. [93] Azam, M., Kesarwani, M., Natarajan, K. and Datta, A. (2001) A secretion signal is present in the Collybia velutipes oxalate decarboxylase gene. Biochemical and Biophysical Research Communications, 289 (4), 807–812. [94] Toyama, R. and Okayama, H. (1990) Human chorionic gonadotropin alpha and human cytomegalovirus promoters are extremely active in the fission yeast Schizosaccharomyces pombe. FEBS Letters, 268 (1), 217–221. [95] Sarmah, B., Chakraborty, N., Chakraborty, S. and Datta, A. (2002) Plant pre-mRNA splicing in fission yeast, Schizosaccharomyces pombe. Biochemical and Biophysical Research Communications, 293 (4), 1209–1216. [96] Giga-Hama, Y. and Kumagai, H. (1999) Expression system for foreign genes using the fission yeast Schizosaccharomyces pombe. Biotechnology and Applied Biochemistry, 30Pt (3), 235–244. [97] Yamazaki, S., Sato, K., Suhara, K. et al. (1993) Importance of the proline-rich region following signal-anchor sequence in the formation of correct conformation of microsomal cytochrome P-450s. Journal of Biochemistry, 114 (5), 652–657. [98] Madzak, C., Gaillardin, C. and Beckerich, J.M. (2004) Heterologous protein expression and secretion in the nonconventional yeast Yarrowia lipolytica: a review. Journal of Biotechnology, 109 (1–2), 63–81. [99] Morlino, G.B., Tizzani, L., Fleer, R. et al. (1999) Inducible amplification of gene copy number and heterologous protein production in the yeast Kluyveromyces lactis. Applied and Environmental Microbiology, 65 (11), 4808–4813. [100] Van Hartingsveldt, W., Mattern, I.E., van Zeijl, C.M. et al. (1987) Development of a homologous transformation system for Aspergillus niger based on the pyrG gene. Molecular & General Genetics, 206 (1), 71–75.

42


[101] Sims, A.H., Gent, M.E., Lanthaler, K. et al. (2005) Transcriptome analysis of recombinant protein secretion by Aspergillus nidulans and the unfolded-protein response in vivo. Applied and Environmental Microbiology, 71 (5), 2737–2747. [102] Iwashita, K. (2002) Recent studies of protein secretion by filamentous fungi. Journal of Bioscience and Bioengineering, 94 (6), 530–535. [103] Campbell, E.I., Unkles, S.E., Macro, J.A. et al. (1989) Improved transformation efficiency of Aspergillus niger using the homologous niaD gene for nitrate reductase. Current Genetics, 16 (1), 53–56. [104] Suzuki, S., Taketani, H., Kusumoto, K. and Kashiwagi, Y. (2003) High-throughput screening for isolation of enhanced green fluorescent protein expressing transformants of filamentous fungus using an image analyzer. Journal of Bioscience and Bioengineering, 96 (3), 313–316. [105] Verdoes, J.C. (2007) A dedicated vector for efficient library construction and high throughput screening in the hyphal fungus Chrysosporium lucknowense. Industrial Biotechnology, 3 (1), 48–57. [106] Maras, M., van Die, I., Contreras, R. and van den Hondel, C.A. (1999) Filamentous fungi as production organisms for glycoproteins of bio-medical interest. Glycoconjugate Journal, 16 (2), 99–107. [107] Maras, M., Saelens, X., Laroy, W. et al. (1997) In vitro conversion of the carbohydrate moiety of fungal glycoproteins to mammalian-type oligosaccharides – evidence for N-acetylglucosaminyltransferase-I-accepting glycans from Trichoderma reesei. European Journal of Biochemistry, 249 (3), 701–707. [108] Kalsner, I., Gallmetzer, A., Hatzl, C. et al. (1995) Insertion into Aspergillus nidulans of functional UDPGlcNAc: alpha 3-D-mannoside beta-1,2-N-acetylglucosaminyl-transferase I, the enzyme catalysing the first committed step from oligomannose to hybrid and complex N-glycans. Glycoconjugate Journal, 12 (3), 360–370. [109] Huynh, C.Q. and Zieler, H. (1999) Construction of modular and versatile plasmid vectors for the high-level expression of single or multiple genes in insects and insect cell lines. Journal of Molecular Biology, 288 (1), 13–20. [110] Zeng, C.Q., Wentz, M.J., Cohen, J. et al. (1996) Characterization and replicase activity of double-layered and single-layered rotavirus-like particles expressed from baculovirus recombinants. Journal of Virology, 70 (5), 2736–2742. [111] Kang, C.Y. (1997) Expression of human immunodeficiency virus genes using baculovirus expression system. Molecular Biotechnology, 8 (2), 173–187. [112] Ghosh, S., Parvez, M.K., Banerjee, K. et al. (2002) Baculovirus as mammalian cell expression vector for gene therapy: an emerging strategy. Molecular Therapy: The Journal of the American Society of Gene Therapy, 6 (1), 5–11. [113] Assenga, S.P., You, M., Shy, C.H. et al. (2006) The use of a recombinant baculovirus expressing a chitinase from the hard tick Haemaphysalis longicornis and its potential application as a bioacaricide for tick control. Parasitology Research, 98 (2), 111–118. [114] Hariparsad, N., Sane, R.S., Strom, S.C. and Desai, P.B. (2006) In vitro methods in human drug biotransformation research: implications for cancer chemotherapy. Toxicology In Vitro: An International Journal Published in Association with BIBRA, 20 (2), 135–153. [115] Chiappe, C., Leandri, E., Hammock, B.D. and Morisseau, C. (2007) Effect of ionic liquids on epoxide hydrolasecatalyzed synthesis of chiral 1,2-diols. Green Chemistry, 2007 (9), 162–168. [116] Ulbrich-Hofmann, R., Lerchner, A., Oblozinsky, M. and Bezakova, L. (2005) Phospholipase D and its application in biocatalysis. Biotechnology Letters, 27 (8), 535–544. [117] Brandon, E.F., Raap, C.D., Meijerman, I. et al. (2003) An update on in vitro test methods in human hepatic drug biotransformation research: pros and cons. Toxicology and Applied Pharmacology, 189 (3), 233–246. [118] Kondo, A. and Maeda, S. (1991) Host range expansion by recombination of the baculoviruses Bombyx mori nuclear polyhedrosis virus and Autographa californica nuclear polyhedrosis virus. Journal of Virology, 65 (7), 3625–3632. [119] Wu, X., Cao, C., Xu, Y. and Lu, X. (2004) Construction of a host range-expanded hybrid baculovirus of BmNPV and AcNPV, and knockout of cysteinase gene for more efficient expression. Science in China. Series C, Life Sciences, 47 (5), 406–415. [120] Altmann, F., Fabini, G., Ahorn, H. and Wilson, I.B. (2001) Genetic model organisms in the study of N-glycans. Biochimie, 83 (8), 703–712. [121] Brooks, S.A. (2004) Appropriate glycosylation of recombinant proteins for human use: implications of choice of expression system. Molecular Biotechnology, 28 (3), 241–255. [122] Jarvis, D.L. (2003) Developing baculovirus-insect cell expression systems for humanized recombinant glycoprotein production. Virology, 310 (1), 1–7.


43

[123] Berger, I., Fitzgerald, D.J. and Richmond, T.J. (2004) Baculovirus expression system for heterologous multiprotein complexes. Nature Biotechnology, 22 (12), 1583–1587. [124] Gentz, R., Hayes, A., Grau, N. et al. (1992) Analysis of soluble human and mouse interferon-gamma receptors expressed in eukaryotic cells. European Journal of Biochemistry, 210 (2), 545–554. [125] Martiniuk, F., Chen, A., Donnabella, V. et al. (2000) Correction of glycogen storage disease type II by enzyme replacement with a recombinant human acid maltase produced by over-expression in a CHO-DHFR(neg) cell line. Biochemical and Biophysical Research Communications, 276 (3), 917–923. [126] Park, M. and Tenner, A.J. (2003) Cell surface expression of C1qRP/CD93 is stabilized by O-glycosylation. Journal of Cellular Physiology, 196 (3), 512–522. [127] Patten, C.J., Ishizaki, H., Aoyama, T. et al. (1992) Catalytic properties of the human cytochrome P450 2E1 produced by cDNA expression in mammalian cells. Archives of Biochemistry and Biophysics, 299 (1), 163–171. [128] Prigge, S.T., Kolhekar, A.S., Eipper, B.A. et al. (1997) Amidation of bioactive peptides: the structure of peptidylglycine alpha-hydroxylating monooxygenase. Science, 278 (5341), 1300–1305. [129] Garnier, A., Coˆte, J., Nadeau, I. et al. (1994) Scale-up of the adenovirus expression system for the production of recombinant protein in human 293S cells. Cytotechnology, 15 (1–3), 145–155. [130] Yuen, C.T., Storring, P.L., Tiplady, R.J. et al. (2003) Relationships between the N-glycan structures and biological activities of recombinant human erythropoietins produced using different culture conditions and purification procedures. British Journal of Haematology, 121 (3), 511–526. [131] Edwards, C.P. and Aruffo, A. (1993) Current applications of COS cell based transient expression systems. Current Opinion in Biotechnology, 4 (5), 558–563. [132] Davis, S.J., Ward, H.A., Puklavec, M.J. et al. (1990) High level expression in Chinese hamster ovary cells of soluble forms of CD4 T lymphocyte glycoprotein including glycosylation variants. The Journal of Biological Chemistry, 265 (18), 10410–10418. [133] Geisse, S., Gram, H., Kleuser, B. and Kocher, H.P. (1996) Eukaryotic expression systems: a comparison. Protein Expression and Purification, 8 (3), 271–282. [134] Massie, B., Dionne, J., Lamarche, N. et al. (1995) Improved adenovirus vector provides herpes simplex virus ribonucleotide reductase R1 and R2 subunits very efficiently. Bio/Technology (Nature Publishing Company), 13 (6), 602–608. [135] Pialoux, G., Excler, J.L., Riviere, Y. et al. (1995) A prime-boost approach to HIV preventive vaccine using a recombinant canarypox virus expressing glycoprotein 160 (MN) followed by a recombinant glycoprotein 160 (MN/LAI). The AGIS Group, and lAgence Nationale de Recherche sur le SIDA. AIDS Research and Human Retroviruses, 11 (3), 373–381. [136] Lundstrom, K. (2003) Semliki Forest virus vectors for large-scale production of recombinant proteins. Methods in Molecular Medicine, 76, 525–543. [137] Lundstrom, K. (2003) Semliki Forest virus vectors for rapid and high-level expression of integral membrane proteins. Biochimica et Biophysica Acta, 1610 (1), 90–96. [138] Geisse, S. and Henke, M. (2005) Large-scale transient transfection of mammalian cells: a newly emerging attractive option for recombinant protein production. Journal of Structural and Functional Genomics, 6 (2–3), 165–170. [139] Streatfield, S.J. (2007) Approaches to achieve high-level heterologous protein production in plants. Plant Biotechnol J, 5 (1), 2–15. [140] Boehm, R. (2007) Bioproduction of therapeutic proteins in the 21st century and the role of plants and plant cells as production platforms. Annals of the New York Academy of Sciences, 1102, 121–134. [141] Houdebine, L.M. (2002) Animal transgenesis: recent data and perspectives. Biochimie, 84 (11), 1137–1141. [142] Van Berkel, P.H., Welling, M.M., Geerts, M. et al. (2002) Large scale production of recombinant human lactoferrin in the milk of transgenic cows. Nature Biotechnology, 20 (5), 484–487. [143] Zhu, L., van de Lavoir, M,C., Albanese, J. et al. (2005) Production of human monoclonal antibody in eggs of chimeric chickens. Nature Biotechnology, 23 (9), 1159–1169. [144] Stoger, E., Ma, J.K., Fischer, R. and Christou, P. (2005) Sowing the seeds of success: pharmaceutical proteins from plants. Current Opinion in Biotechnology, 16 (2), 167–173. [145] Misaki, R., Fujiyama, K. and Seki, T. (2006) Expression of human CMP-N-acetylneuraminic acid synthetase and CMP-sialic acid transporter in tobacco suspension-cultured cell. Biochemical and Biophysical Research Communications, 339 (4), 1184–1189.

44


[146] Stoger, E., Sack, M., Fischer, R. and Christou, P. (2002) Plantibodies: applications, advantages and bottlenecks. Current Opinion in Biotechnology, 13 (2), 161–166. [147] Sawasaki, T., Ogasawara, T., Morishita, R. and Endo, Y. (2002) A cell-free protein synthesis system for highthroughput proteomics. Proceedings of the National Academy of Sciences of the United States of America, 99 (23), 14652–14657. [148] Vinarov, D.A. and Markley, J.L. (2005) High-throughput automated platform for nuclear magnetic resonancebased structural proteomics. Expert Review of Proteomics, 2 (1), 49–55. [149] Ikeda, K., Nakazawa, H., Shimo-Oka, A. et al. (2006) Immobilization of diverse foreign proteins in viral polyhedra and potential application for protein microarrays. Proteomics, 6 (1), 54–66. [150] Swartz, J. (2006) Developing cell-free biology for industrial applications. Journal of Industrial Microbiology & Biotechnology, 33, 476–485. [151] Kanter, G., Yang, J., Voloshin, A. et al. (2007) Cell-free production of scFv fusion proteins: an efficient approach for personalized lymphoma vaccines. Blood, 109 (8), 3393–3399. [152] Spirin, A.S., Baranov, V.I., Ryabova, L.A. et al. (1988) A continuous cell-free translation system capable of producing polypeptides in high yield. Science, 242 (4882), 1162–1164. [153] Sawasaki, T., Hasegawa, Y., Tsuchimochi, M. et al. (2002) A bilayer cell-free protein synthesis system for highthroughput screening of gene products. FEBS Letters, 514 (1), 102–105. [154] Shimizu, Y., Inoue, A., Tomari, Y. et al. (2001) Cell-free translation reconstituted with purified components. Nature Biotechnology, 19 (8), 751–755. [155] Tarui, H., Murata, M., Tani, I. et al. (2001) Establishment and characterization of cell-free translation/ glycosylation in insect cell (Spodoptera frugiperda 21) extract prepared with high pressure treatment. Applied Microbiology and Biotechnology, 55 (4), 446–453. [156] Jackson, A.M., Boutell, J., Cooley, N. and He, M. (2004) Cell-free protein synthesis for proteomics. Briefings in Functional Genomics and Proteomics, 2 (4), 308–319. [157] Klammt, C., Schwarz, D., L€ohr, F. et al. (2006) Cell-free expression as an emerging technique for the large scale production of integral membrane protein. FEBS Journal, 273 (18), 4141–4153. [158] Guignard, L., Ozawa, K., Pursglove, S.E. et al. (2002) NMR analysis of in vitro-synthesized proteins without purification: a high-throughput approach. FEBS Letters, 524 (1–3), 159–162. [159] Kigawa, T., Yabuki, T., Yoshida, Y. et al. (1999) Cell-free production and stable-isotope labeling of milligram quantities of proteins. FEBS Letters, 442 (1), 15–19. [160] Morita, E.H., Shimizu, M., Ogasawara, T. et al. (2004) A novel way of amino acid-specific assignment in 1 H -15 N HSQC spectra with a wheat germ cell-free protein synthesis system. Journal of Biomolecular NMR, 30 (1) 37–45. [161] Kainosho, M., Torizawa, T, Iwashita, Y. et al. (2006) Optimal isotope labelling for NMR protein structure determinations. Nature, 440 (7080), 52–57. [162] Hendrickson, W.A., Horton, J.R. and LeMaster, D.M. (1990) Selenomethionyl proteins produced for analysis by multiwavelength anomalous diffraction (MAD): a vehicle for direct determination of three-dimensional structure. The EMBO Journal, 9 (5), 1665–1672. [163] Kigawa, T., Yamaguchi-Nunokawa, E., Kodama, K. et al. (2002) Selenomethionine incorporation into a protein by cell-free synthesis. Journal of Structural and Functional Genomics, 2 (1), 29–35.

3 Directed Enzyme Evolution and High-Throughput Screening Michael J. McLachlan,1 Ryan P. Sullivan2 and Huimin Zhao3 1

Center for Biophysics and Computational Biology, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA 2 Department of Chemical and Biomolecular Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA 3 Departments of Chemical and Biomolecular Engineering, and Chemistry, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA

3.1 Introduction Early enzyme products developed for industrial applications were produced by native hosts via fermentation and consisted of a complex mixture of secreted enzymes produced at low yields. Now, over 90% of industrial enzymes are produced recombinantly for maximal purity and productivity [1]. The integration of enzymes into a commercial process relies on the availability of those with high activity and stability under process conditions, desired substrate specificity, and high selectivity. More often than not, naturally occurring enzymes do not fulfill the requirements of these harsh industrial conditions, and optimization is necessary to obtain a suitable enzyme catalyst for production needs. This tailoring of enzymes can be accomplished through two experimental routes. The first is rational design, which targets specific residues of a protein for mutagenesis to predetermined amino acid mutations, and is only applicable when there is detailed knowledge of the relationships between the enzymes structure and mechanism/function. And while an increasing number of enzymes are being characterized, the majority do not have this depth of information readily available, as it requires considerable effort to obtain. In the absence of this information, the tailoring of an enzyme can still be accomplished through the second route: directed evolution.


46


Directed evolution is the general term applied to the combined techniques of generation of a library of protein mutants (or variants) and selection of a protein with desirable function from within that library [2]. It is an iterative Darwinian optimization process, whereby the fittest variants are selected from an ensemble of mutants [3]. Directed evolution can be used to target a number of enzymatic characteristics, including activity, substrate specificity, thermal and oxidative stability, enantioselectivity or enantiospecificity, pH optima or range, and tolerance to solvent [4]. While a typical directed evolution experiment focuses on a single enzymatic trait, there are some examples of improving several traits simultaneously. Developed primarily in academic laboratories, directed evolution practices in industry are still in their infancy and present an imposing challenge for process scientists today. Choosing the appropriate methods of library generation and screening or selection is paramount to the success of any directed evolution experiment. Library diversity can be created through either mutagenesis (random or semi-rational) or gene recombination, and which of these methods is chosen depends on many factors, such as the availabilities of homologous genes, structural knowledge, and characteristic data of the enzyme of interest. The library size created is typically very large ( > 104–6), and close evaluation of each variant is not feasible. The need for a method to find the improved ‘needle in a haystack’ enzyme becomes evident, with several strategies, including selection, enrichment, and high-throughput screening, offering ways to sift through the library clutter and find a variant with the desired enzymatic trait. However, once an evolved enzyme has been found that exhibits improved characteristics, the artificial conditions in which the selection method was carried out may result in an enzyme whose properties may not carry over to the real biocatalytic process. Therefore, the more similar a screening system is to the actual application process, the more likely it is to find an improved enzyme that will be complementary to the application. The aims of this chapter are to familiarize the reader with the various techniques used in library creation and high-throughput screening, as well as representative examples in the pharmaceutical industry and the technology used to adapt the laboratory-based techniques to industrial settings. The topics covered do not represent a complete listing of methods utilized, as more strategies for library creation and high-throughput screening are constantly introduced.

3.2 Directed Evolution Library Creation Strategies Library creation strategies generally fall into three main categories: random mutagenesis, semi-rational design, and gene shuffling (Figure 3.1). Random mutagenesis introduces mutations throughout a target gene encoding for the industrial enzyme of interest. These mutations may be in the form of point mutations (either transitions or transversions), insertions, deletions, inversions, or frame-shift mutations. Semi-rational design is a combination of random mutagenesis and site-directed mutagenesis, in which specific residue positions are rationally determined to play important roles in the enzymes function, and are subsequently randomized to all 20 amino acids. Gene shuffling involves exchanging fragments of genes with one another to create a library of chimeric progeny. In directed evolution studies, this is typically accomplished by homologous recombination for sequences with high similarity or by nonhomologous recombination for those with low similarity.

Directed Enzyme Evolution and High-Throughput Screening

47

Figure 3.1 Overview of DNA library creation strategies. Random mutagenesis introduces mutations at positions throughout the gene sequence. Semi-rational design randomizes only the specific position(s) of interest. Gene shuffling brings existing sequence diversity from different parental DNA sequences together to form a chimeric library

3.2.1 Random and Semi-Rational Mutagenesis The most commonly used method for random mutagenesis library creation is error-prone polymerase chain reaction (epPCR), which introduces mutations in a gene product by lowering the fidelity of DNA polymerase. This can be achieved through many different reaction conditions, including use of nucleotide base analogs and alkylating agents into the polymerase chain reaction (PCR) reaction mixture, but severe biases typically result from such modification. The more favored protocols utilize an Mg2þ -dependent polymerase that lacks exonuclease activity, substitute Mn2þ for Mg2þ , and extend DNA products with an uneven mixture of the four precursor deoxynucleotidetriphosphates [5]. The mutation rate can be easily adjusted by varying the concentration of Mn2þ in the reaction mixture or by changing the number of cycles of amplification. Typically, a library is created with a low mutation rate (one to five base pairs, or one to three amino acids per gene) to prevent disruption of enzymatic activity and generation of a library too large to screen comprehensively. However, where there is an efficient selection scheme or robust high-throughput screen available, higher mutation rates have been successful in gaining improved function [6,7]. epPCR is considered as a random mutagenesis technique, but the term random should be interpreted loosely. Ideally, a protein library should satisfy the requirement of having an equal probability of substitution of any of the 20 amino acids randomly into any residue position of the protein. However, the library creation is performed at the level of the gene encoding the protein; thus, an ideal distribution is improbable due to the redundancy of the genetic code. Mutations at the wobble position (third base of the codon) typically result in a neutral mutation, and the frequency of two adjacent bases being mutated is considerably low, which results in only six possible amino acids on average at each position. As a result, the DNA sequences of selected progeny with improved function are determined, and positions at which mutations have been incorporated are sometimes subjected to an additional round of screening using saturation mutagenesis. Saturation mutagenesis involves creating a library by designing

48


degenerate primers for the residue position in question and screening the resultant library to determine which of the 20 amino acids results in the most dramatic improved effect at that position. Gene site saturation mutagenesis (GSSM) takes this concept to its fullest extent by creating a library consisting of all 20 possible amino acid substitutions for each position in a protein [8,9]. Sequence saturation mutagenesis (SeSaM) [10] is an alternative four-step approach to epPCR that overcomes several barriers faced by its predecessor. First, a library is created with standard nucleotides with the presence of a-phosphothionate nucleotide, an alkaline labile analog. The incorporated analog is hydrolyzed, thus creating a library of fragments of random length. The fragments are then treated with a terminal deoxynucleotidyl tranferase (TdT) to incorporate a random number of universal bases at the 30 -termini, followed by elongation of labeled gene product to full length. A concluding PCR then replaces the universal base substitutions with standard nucleotides. The mutations incorporated into the SeSaM library are randomly distributed throughout the parent gene template and independent of the DNA polymerase, thus avoiding any potential mutational bias and resulting in the exchange of an amino acid at any position to all other 19 possible amino acids. The SeSaM protocol has also been modified to enrich the transversion rate and allow for adjustable mutational biases [11]. However, the use of different base analogs, biotinylated primers, and the additional time-consuming steps for library creation are drawbacks to this method. Insertion and deletion mutagenesis expands the reach of random mutagenesis to include alteration of the size of the gene of interest when generating a library. This process requires the use of transposable elements that insert randomly into the target and are subsequently excised, leaving behind an in-frame fragment. Pentapeptide scanning mutagenesis results in the insertion of a 15-bp fragment, but the amino acids incorporated are predetermined due to the cognate target site of the transposon and preset restriction site sequence [12]. This results in a less than optimal library with less diversity and more bias than other methods. It is particularly suited towards obtaining mutants whose loop structures dictate activity enhancements. An improved technique of the same style is random insertion/deletion, which makes it possible to delete a multitude of consecutive bases from random locations in a gene and insert either a specific or randomized sequence of arbitrary length in its place [13]. The diversity obtained from this method far outstretches that of conventional epPCR, but it comes at the cost of several additional processing steps.

3.2.2 Gene Shuffling Although random mutagenesis techniques offer a relatively simple scheme to approach a desired enzymatic trait, the frequency of gaining beneficial mutations is generally low. Typically, only one or two amino acid changes are generated in each round, with higher random mutation rates leading to either a library too large to screen successfully or to incorporation of mutations that lead to loss of function. Gene shuffling can overcome this limitation by allowing a large number of beneficial mutations from multiple genes to incorporate at a single step of the library creation process. There are a variety of shuffling methods available, which can be divided into categories of homology dependent and homology independent.


49

3.2.2.1 Homology-Dependent, Primer-Independent Assembly The pioneering technique of DNA shuffling [14] was a breakthrough for in vitro recombinant mutagenesis and diversity generation. The method is centered on taking a number of homologous parental template genes, randomly digesting them with DNase I and reassembling the pieces back together into full-length genes through a primerless PCR reaction (Figure 3.2a). In this extension process, pieces of different genes can anneal to each other at homologous regions and extend to full-length fragments through the aid of DNA polymerase. DNA shuffling can be applied to combining beneficial mutations found from random mutagenesis screening, which can reduce the processing steps required to determine which mutations screened for are actually necessary for improved function. A single gene variation of DNA shuffling involves incorporation of mutations at the extension level, screening for improved progeny, and subsequent shuffling of these mutants for further screening. In contrast, family shuffling [15] begins the process with naturally occurring homologous genes (typically > 60% identical). Because the parental templates in this process have already been subjected to natural selection, much larger numbers of mutations are tolerated, leading to a library with a broader sequence space that maintains a higher percentage of functional enzymes. These methods all have similar limitations, however, as they rely on sequence homology for recombination events to occur. Gene fragments will tend to anneal to each other only at areas of

Figure 3.2 Examples of gene shuffling methods used for DNA library creation in directed evolution. (a) Homology-dependent primer-independent DNA shuffling; (b) homology-dependent primerdependent StEP; (c) homology-independent SHIPREC

50


high sequence similarity, leaving out the possibility of crossovers in other regions with low similarity. Also, the chimeric genes typically only contain one to four crossover events, which strongly limits the accessible sequence space. This low crossover rate of DNA shuffling was addressed with a method called random chimeragenesis on transient templates (RACHITT) [16]. The parental template in this case contained uracil and was single stranded to serve as a scaffold for hybridization of second-strand fragments from homologous genes. The crossover rate was dramatically increased (around 14 per chimera product), leading to much greater diversity in the library. However, this method is more labor intensive and is subject to optimization of reaction times and temperatures to obtain full-length genes for screening. 3.2.2.2 Homology-Dependent, Primer-Dependent Assembly In contrast to the methods described above, several techniques have been developed that require end-primers or addition of oligonucleotides to encourage frequent crossover between templates. Random priming recombination [17] involves a parental template gene as a scaffold, with random primers used in place of DNase I fragments to anneal randomly on the template and extend to full-length genes. A staggered extension process (StEP, [18]) starts the extension step of shuffling only with end-primers. Instead of extending the gene to full length in one cycle, the process is carried out through very short extension and annealing steps, resulting in template switching throughout the course of full-length gene assembly, thereby producing multiple crossovers (Figure 3.2b). However, the limitation in combining close-proximity mutations is still present. Synthetic shuffling is accomplished without any full-length gene templates, but rather with synthetic degenerate oligonucleotides designed with bioinformatic information on the gene of interest [19]. The oligonucleotides contain overlapping ends that anneal and extend into full-length composite genes. This method opens up the capabilities of introducing site-directed mutations (via semi-synthetic shuffling) or codon optimization into the gene library. Biased mutation-assembling [20] takes wild-type and mutant genes with distinct beneficial mutations, defines blocks within them containing only single mutations, and amplifies the regions with small overlaps to adjacent blocks. The blocks are then reassembled by overlap extension PCR, but are combined in ratios which favor incorporation of mutations which displayed the most improvement. 3.2.2.3 In Vivo Assembly All of the gene shuffling methods discussed so far involve the application of in vitro techniques for generation of the gene library. Another possibility to create low-complexity chimeric libraries is through in vivo recombination methods. In combinatorial libraries enhanced by recombination in yeast [21], parental genes are again randomly fragmented with DNase I and reassembled in a primerless reaction. However, the annealing step is repeated at lower and lower temperatures as the reaction progresses (‘progressive hybridization’), facilitating the annealing of low-homology genes. Amplification of full-length genes is then carried out with primers designed with overlapping regions on the expression vector. The linearized vector and library of genes are then transformed into yeast to promote in vivo homologous recombination. Because there is typically more than one gene hybrid transformed into a yeast cell, recombination results in an increased diversity.


51

Heteroduplex recombination involves preparing single-stranded DNA from two different homologous genes and mixing them to form heteroduplexes. These heteroduplexes are then transformed into a host that then creates hybrid homoduplexes through in vivo mismatch repair mechanisms [22]. Another technique takes advantage of recA-mediated homologous recombination in a recBC sbcA Escherichia coli mutant [23]. In this method, the parental genes flank a linearized plasmid, and transformation along with in vivo homologous recombination results in single crossover progeny on circularized plasmids. Repeating this process produces one crossover per iteration, and can be modified to include a different parental gene per round. 3.2.2.4 Homology-Independent Assembly As stated earlier, one of the major drawbacks to DNA shuffling and techniques derived from this method is that crossovers occur only at homologous regions. If low-homology parent genes are used in these methods, then the majority of product genes tend to be parental themselves instead of hybrids [24]. In an attempt to dissociate the creation of hybrid enzymes from DNA sequence homology, a technique called incremental truncation for the creation of hybrid enzymes (ITCHY) was created [25]. In this procedure, two parent enzymes are incrementally truncated with exonuclease III under controlled conditions, and the various generated 50 -fragments and 30 -fragments are ligated back together to form a library of chimeric sequences. The resulting enzymes screened are fusions of the amino-terminal portion of one parent and the carboxy-terminal end of the other parent. However, careful optimization and control of the fragment generation is required for a successful library, and makes the method difficult and time consuming. In addition, the fusion enzymes created contain only one crossover region, are not necessarily full-length genes, and can be conjoined at places that are not at structurally related sites [26]. A modified version of this protocol, called THIO-ITCHY, includes the random incorporation of a-phosphothioate nucleotide analogs into the parent genes [27,28]. Exonuclease III activity is inhibited at sites of analog incorporation, which relieves the efforts of producing incremental truncation aliquots. In combination with epPCR, the diversity of the fusion libraries created can be further expanded. Yet another variant of the ITCHY method, fittingly called SCRATCHY, was developed through the combination of ITCHY with DNA shuffling to create multiple crossover libraries independent of homology [27,28]. Further development of this technique produced enhancedcrossover SCRATCHY, in which amplification of ITCHY hybrids is carried out in defined blocks with skewed primers (a forward primer from one parent and a reverse primer from the other). These amplified fragments are then pooled and subjected to DNA shuffling, resulting in a process that selectively enriches hybrids that contain multiple crossovers [29]. Another method that produces hybrid genes independent of homology is sequence homology independent protein recombination (SHIPREC) [30]. Parental genes are first fused together via a linker containing a unique restriction site to form heterodimers. These heterodimers are then randomly digested with DNase I in a controlled reaction. The fragments corresponding to the length of either parental gene are then isolated, blunt-end digested, and ligated to create circularized gene hybrids. These hybrids are then linearized with the unique restriction site introduced originally into the linker (Figure 3.2c). The library is still limited to one crossover event, but is also more likely to contain structurally conserved crossover events.

52


3.2.2.5 Multiple-Parent, Nonhomologous Assembly All of the previous homology-independent shuffling methods are limited by the use of only two parental genes. However, methods to recombine DNA from multiple parents do exist. In cases where gene templates are from eukaryotic sources, exon shuffling can be implemented to interchange domain-encoding exons [31]. In this way, functional domains are connected through design of oligonucleotides used to amplify permitted crossovers, and the random aspect of assembly is avoided. Random multi-recombinant PCR (RM-PCR) is a more generalized scheme of exon shuffling in which reassembly of blocks into full-length genes is possible through overlap-extension PCR, with oligonucleotides containing the crossovers, and can be used for prokaryotic and eukaryotic genes alike [32]. Degenerate homoduplex recombination is similar to synthetic shuffling, and utilizes a set of degenerate top-strand primers that are designed to encompass the diversity information of the genes to be shuffled. Gaps in between these top-strand primers are filled in using dephosphorylated bottom-strand oligonucleotides as templates [33]. Nonhomologous random recombination assembles DNase I-digested fragments that have been blunt-end polished prior to ligation. The addition of hairpin oligonucleotides that cap only the ends of the full-length genes results in preferential ligation of intermolecular pieces [34]. To address limitations with creating frame-shift mutants with premature stop codons, the addition of a chloroamphenicol acetyltransferase (CAT) fusion allowed for preferential selection against truncated or insoluble proteins [35]. There is no single library creation method that will satisfy the ideal requirements of highdiversity quality, large library size, and ultimately the feasibility of comprehensive screening of the generated library. Evaluation of each case is necessary to determine which library creation method is suitable for the projects goals.

3.3 Directed Evolution Library Screening/Selection Methods The options for analyzing a library will be guided in part by the properties of the protein being studied. Whether the target is a binding protein or an enzyme, what substrates or derivatives are available and how these are linked to metabolism will help determine the appropriate screening technique. The analysis involves finding a method that will link the desired activity with the gene variant that encodes it, and can be split into two approaches: screening and selection. In a screening method, variants in the library are assayed individually. For example, by measuring the enzymatic activity in a cell lysate, both active and inactive clones will be discovered. Alternatively, selection methods deal with the entire library at once. For example, by applying antibiotics to bacteria on an agar plate, only resistant clones will grow. A selection method is desirable, since it increases the size of the library that can be practically assayed. The methods used can also be classified as either in vivo (where intact cells are used) or in vitro (where isolated cellular components are utilized). The following will give an overview of these methods, with more details available in published reviews [36,37].

3.3.1 In Vivo Methods: Genetic Complementation The classic microbiological approach of genetic complementation can be very useful as a directed evolution selection [38,39]. The activity being investigated is intrinsically linked to


53

the growth of a microbe such as E. coli or Saccharomyces cerevisiae, as it either provides an advantage to the wild-type strain or the library can complement a particular mutant strain. By creating an E. coli strain that was unable to utilize glucose as a carbon source, an overexpression library was screened to discover genes with latent glucokinase activity [40]. When the desired reaction is not essential, creative thinking must be employed. Adding a selectable moiety may create derivatives of a natural substrate that allow for selection. Hwang et al. [41] used this approach to derive a screen for enantioselective hydrolases. By linking the antibiotic chloramphenicol to either the R- or S-enantiomer of 2-phenylbutyric acid, they showed that Exiguobacterium acetylicum could hydrolyse the R-form (as the released chloramphenicol inhibited growth) but not the S-form. However, one must be aware that activity on a derivative may not always correspond to the activity on the actual desired substrate. The degree of selection can be adjusted as needed by the use of different vectors or promoters to alter the expression level of the enzyme within the cell. Alternatively, the stringency of selection can be increased by channeling the substrate away by adding a competing pathway [42].

3.3.2 In Vivo Methods: Chemical Complementation The drawback of traditional genetic complementation approaches is that they are specific for the gene being studied, and new screens must be devised for each enzyme being investigated. Chemical complementation methods aim to be more general in their approach, by using the substrate of the targeted reaction to influence an unrelated reporter system. This allows one to use well-studied reporters, such as b-galactosidase, or the amino-acid-selectable markers of yeast if they can be linked in some manner to the substrate. This has been accomplished through the use of the yeast three-hybrid assay [43]. Here, the phenotypic readout is the blue color produced by the action of b-galactosidase on X-gal or onitrophenol-b-D-galactopyranoside. The transcription of this reporter gene is determined by the dimerization of a methotrexate-binding DNA binding domain to a dexamethasone-binding activation domain. The two components are linked by a composite molecule containing the substrate of interest, methotrexate, and dexamethasone. If the library variant cleaves the substrate, then there is no dimerization of the transcriptional components, b-galactosidase will not be produced, and there will be no blue color. By changing the reporter gene, this type of screen could be converted to a selection, as was demonstrated with a glycosynthase, where bond formation controlled the transcription of a leucine biosynthetic reporter [44]. Other approaches could use transcriptional regulators that directly bind the substrate or product of the reaction and activate the reporter gene. For instance, a mutant transcriptional activator from Pseudomonas putida, NahH, was used that can bind various benzoic acids to develop a screening/selection method to detect the action of benzaldehyde dehydrogenase [45]. A transcriptional regulator may need to be engineered to bind the desired compound before it can be used in such a manner [46].

3.3.3 In Vivo Methods: Surface Display The ability to display proteins on the surface of an organism has been exploited as a screening method, typically in conjunction with fluorescence-activated cell sorting (FACS) [37,47,48]

54


Figure 3.3 Cell surface display. Proteins displayed on the cell surface of organisms bind a fluorescent molecule. Cells are passed through a FACS machine, allowing separation into populations that do or do not bind the target

(Figure 3.3). The approach is particularly well suited for engineering binding proteins such as antibodies, although it has also been applied to enzymes. One of the first widely used surface display methods was through the use of phage [49]. By inserting a target protein into one of the coat proteins of filamentous phage (initially protein pIII, but others such as pVIII have also been used), very large libraries (1010) of variants can be expressed in E. coli and subsequently packaged into active phage. Antibodies are used to enrich the desired variants in vitro, and the phage can then reinfect bacteria in multiple rounds of selection. The selection strategy is useful for increasing the binding affinity of a protein. Stability can also be engineered through the use of proteases, which will cleave a poorly folded protein more rapidly and reduce the infectivity of the phage [50]. Another method is yeast display, which consists of fusing the engineering target to the C-terminus of the Aga2 cell surface agglutinin protein, allowing exposure to fluorescently labeled substrates in the media [51]. Bacterial display systems have also been produced using a number of different scaffold proteins, such as the outer membrane proteins OmpA or OmpX, the thioredoxin protein within the bacterial flagella, or autotransporters such as AIDA-I [37]. Proteins may express differently depending on which scaffold they are fused with, or whether bacteria or yeast are used. The use of yeast as an expression host may be beneficial if the protein requires post-translational modifications.


55

3.3.4 In Vitro Methods: Lysate Assay Perhaps the most direct way of analyzing a library is to perform the functional assay on the protein itself. Individual clones from cells transformed with the library can be grown and made to express the protein of interest. This can then be isolated and the relevant assay performed. A cell lysate will often contain enough of the protein such that further purification is not necessary for screening purposes. Using microplates increases the throughput of the screening and is useful where the progression of a reaction can be monitored visually, either by the distinct absorbance of the substrate or product or of a coupled reaction. The throughput of this type of screen is comparatively low, on the order of 104, but the approach is flexible and easily implemented. Reaction products from such assays could also be detected directly through the use of gas chromatography, high-performance liquid chromatography, or mass spectrometry.

3.3.5 In Vitro Methods: Ribosome Display Ribosome display is an in vitro display method by which one can link the protein of interest to the gene coding for it. As demonstrated with a single-chain scFv antibody, by using in vitro transcription and translation on DNA lacking a stop codon, a complex consisting of the mRNA, ribosome, and translated protein is generated [52]. The variants of interest are then selected by exposing the complex to the proteins binding partner and washing away nonbinding complexes. The bound complexes are dissociated and the mRNA molecules can be collected and converted back into DNA via reverse transcription, resulting in a pool of sequences enriched for the desired binding activity. A similar approach is used in the technique of mRNA display, but the protein of interest is linked to its mRNA by the use of a puromycin molecule [53,54]. Both techniques have the advantage of being completely in vitro, increasing the potential size of the library (to 1013) and the speed of screening by avoiding transformation steps into cells.

3.3.6 In Vitro Methods: In Vitro Compartmentalization Inspired by the natural linkage of genotype and phenotype observed with cells, Tawfik and Griffiths [55] introduced the idea of in vitro compartmentalization (IVC) by coupling transcription and translation within a water-in-oil emulsion. By forming aqueous droplets of around 2 mm in diameter, genes, their proteins, and more importently substrates and products of the protein are all contained, which allows for the selection of catalytic activities. The concept was initially demonstrated on enzymes that act on DNA, such as HaeIII methyltransferase, where the catalytic activity acted directly on the genotype. The use of antibodies and streptavidin-coated microbeads allowed attachment of a gene, its encoded protein, and biotintagged substrate which was converted to product. An antibody to the product allowed the isolation of improved variants of the enzyme phosphotriesterase through flow cytometry [56]. Extending the emulsion to a water-in-oil-in-water mixture allowed further refinement of the IVC concept. Compartmentalization of E. coli containing serum paraoxonase variants allowed the accumulation of fluorescent product to a point where it could be detected by FACS [57]. This approach was also used with in vitro transcription and translation to evolve b-galactosidase activity from the Ebg gene [58].

56


3.3.7 Equipment/Automation Directed evolution relies on the analysis of large numbers of clones to enable the discovery of rare variants with improved function. In order to analyze these large libraries, methods of screening or selection have been developed, many of which use specialized equipment or automation. These range from the use of multichannel pipettes, all the way up to robotics, depending on the level of investment [59]. Specialized robotic systems are available to perform tasks such as colony picking, cell culture, protein purification, and cell-based assays. A very versatile piece of equipment that is affordable for individual laboratories is the microplate reader. This allows multiple samples to be analyzed at once, commonly in a 96-well format, although 384- and 1536-well formats are available. Typical measurements that can be performed include UV–Vis absorbance, fluorescence, or luminescence, allowing a range of assays to be performed, such as cell growth, enzyme kinetics, enzyme stability, or enzymelinked immunosorbent assay [60–62]. Functionality can be increased by the use of liquid dispensing systems or automatic plate handling. Flow cytometry is increasingly used in high-throughput screening [63]. Here, a cell solution is pressurized into a stream and directed through various lasers, with information on forward and side scatter and on fluorescence being collected. Based on this data, the cells can be directed into separate pools. The throughput of these machines is up to 40 000 cells per second, making it feasible to screen libraries containing on the order of 109 variants.

3.4 Selected Industrial Examples One of the main caveats that come with the integration of enzymatic processes into industrial settings is that enzymes have evolved through Darwinian evolution for the purpose of survival, not for the purpose of overproducing valuable pharmaceutical products in harsh industrial settings. While an enzyme may have the desired catalytic function for a particular processing step, it most likely will not have sufficient catalytic properties, stability, specificity, or enantioselectivity when simply exchanged with a developed chemical route of synthesis. In addition, the production of the enzyme itself may be economically infeasible due to low expression or solubility. Therefore, there are now many examples of utilizing library creation and screening tools to enhance enzymes for industrial applications. A small sampling of recent examples is listed below for the improvement of various enzymatic functions, with further references for more in-depth study provided for the readers benefit.

3.4.1 Activity The activity of an enzyme is crucial to the productivity of an industrial process. Many pharmaceutically relevant enzymatic processes utilize enzymes with multiple domains that catalyze different reactions. One example is a hybrid, nonribosomal peptide synthase–polyketide synthase (NRPS-PKS) from Pantoea agglomerans [64], which produces the broadspectrum antibiotic andrimid [65,66], an acetyl-CoA carboxylase inhibitor. Based on the proposed pathway for andrimid biosynthesis, the valine-specific A domain of the AdmK protein was replaced with CytC1, a 2-aminobutyrate-incorporating A domain from the Streptomyces sp. RK95-74 cytotrienin NRPS-PKS. The CtyC1 domain was chosen due to its broad specificity profile. While andrimid was still produced with AdmK-CtyC1 hybrid, the


57

Figure 3.4 Improvement of the activity of chimeric NRPSs using directed evolution. (1) A heterologous A domain is swapped into an NRPS, typically resulting in a significant loss of synthetase activity. (2) A library of chimeric synthetase mutants is constructed in which the heterologous A domain has been diversified (for example, by error-prone PCR). (3) The library is subjected to an in vivo screen for production of the unnatural nonribosomal peptide derivative. (4) Clones showing improved production are characterized and subjected to further rounds of diversification and screening

level was reduced by 32-fold compared with the wild-type AdmK. Several rounds of errorprone PCR were performed, and the resulting mutant AdmK-CtyC1 enzyme produced 10.7fold more andrimid than the wild-type AdmK-CtyC1 and only threefold less than the wild-type AdmK. The screen was also extended to test for the increased production of andrimid derivatives produced by the new CtyC1 A domain (Figure 3.4). The results indicated that chimeric NRPSs can be functionally improved, as well as reach activity levels approaching those of the wild-type counterparts [67]. Cephalosporins are a class of antibiotic produced via the intermediate 7-aminocephalosporanic acid (7-ACA), or 7-aminodesacetoxycephalosporanic acid (7-ADCA). Directed evolution has been used to improve the activity of cephalosporin acylases to produce these intermediates from adipyl-7-ACA or cephalosporin C [68]. Using site-directed saturation mutagenesis and a selection system whereby the E. coli host is dependent on leucine liberated from derivatives of the cephalosporin side-chains, a mutant was found that increased the catalytic efficiency toward adipyl-7-ADCA by 36-fold. Oxidoreductases are a family of enzymes that catalyze a number of industrially important reactions, but they often require additional nicotinamide (NADH or NADPH) cofactors which

58


are too expensive to supply stoichiometrically. Regeneration of the reduced cofactor form was accomplished by Woodyer et al. [69] by linking an oxidoreductase reaction to phosphite dehydrogenase, an enzyme that oxidizes inert phosphite to phosphate with concomitant reduction of NAD(P) to NAD(P)H. However, the wild-type enzyme had relatively low activity. Through several rounds of random mutagenesis, the activity of the wild-type enzyme was improved by sixfold. The advantage of the wild-type enzyme was demonstrated using the model industrial bioconversion reaction of trimethylpyruvate to L-tert-leucine, an unnatural amino acid derivative.

3.4.2 Thermostability An enzymes thermostability is generally one of the more difficult characteristics to rationally improve. Beneficial mutations that lead to increased thermostability can be involved in many different mechanisms, including solvent interactions, structural support, and electrostatic balance. The majority of attempts to enhance the thermostability of an enzyme have, in some way or another, incorporated random mutagenesis techniques to generate improved enzymes. However, once improved mutants have been screened for and isolated, combining the different mutations together can be labor intensive. Through biased mutation-assembling, which simplified and reduced the workload of incorporating multiple beneficial mutations, Hamamatsu et al. [20] were able to improve the thermostability of a prolyl endopeptidase from Flavobacterium meningosepticum, resulting in a 1200-fold improvement of the enzymes half-life at 60 C. Another example involved a combination of computational data, rational design, and directed evolution. Iterative saturation mutagenesis [70] was used on a model enzyme lipase from Bacillus subtilis to improve its thermostability. Appropriate amino acid residues were chosen based on X-ray data, in particular the B factors (or B values). These factors reflect smearing of atomic electron densities with respect to their equilibrium positions as a result of thermal motion and positional disorder. Amino acids which display large B factors correspond to those which have pronounced flexibility and are targets for stabilization. Each position was subjected to saturation mutagenesis, to generate as many single position libraries as were suitable for the experiment. After screening, the improved mutants found were ranked by position on their contribution to the enzymes improved thermostability. Consecutive rounds of saturation mutagenesis were then performed, first starting with the position of highest improvement, followed by the second highest, and so on until all positions had been explored. The B. subtilis lipase was evolved through this method to two mutants (five and seven total mutations each) that had half-lives of more than 15 h at 55 C, compared with 2 min for the wild-type enzyme. The principle of additive mutational effects played an important role in this design. Also, the selection of residues was limited to having crystal structures available, although the same process can be carried out with template mutants found by other random mutagenesis techniques.

3.4.3 Substrate Specificity The cholesterol-lowering drug atorvastatin, marketed as Lipitor, is an example where biocatalysis research has been applied extensively and is in industrial use. The enzyme 2-deoxyribose5-phosphate aldolase (DERA) has been a target of directed evolution for the production of atorvastatin intermediates [8,9,71]. DeSantis and coworkers [8,9] used structure-based


59

mutagenesis in an attempt to expand the substrate specificity of the native enzyme. The enzyme variants were fully purified and assayed in vitro by monitoring the decrease of NADH at 340 nm using a spectrophotometer. One variant, S238D, showed new activity towards 3-azidopropinaldehyde to form an azido pyranose which is an intermediate in atorvastatin synthesis. Jennewein et al. [71] also studied this wild-type enzyme, creating a library with error-prone PCR and recombining positive mutants. By screening mutants with a microplate reader or with gas chromatography, they managed to increase the synthesis of the intermediate (3R,5S)-6chloro-2,4,6-trideoxyhexapyranoside by 10-fold.

3.4.4 Product Specificity Polyketides are a structurally diverse group of compounds and one of the richest sources of pharmaceuticals. The biosynthesis of the polyketide avermectin-analog doramectin (commercially sold by Pfizer as Dectomax) in a mutant Streptomyces avermitilis strain was hampered by the product specificity of one of the key intermediate enzymes, aveC, in the doramectin production pathway. Although the particular function of aveC was unknown, the occurrence of undesirable product CHC-B2 was found to be a result of its dual-product specificity. In an effort to improve the production ratio of doramectin:CHC-B2, site-specific and error-prone mutagenesis libraries were screened and resulted in several unrelated beneficial mutations that improved the production ratio up to fourfold [72]. However, much of the screened library had mutations that adversely affected the production ratio of doramectin. Continuing the efforts to improve aveC for doramectin production, a new approach was established to curb the complex and time-consuming initial screens, due in large part to the long growth periods of the S. avermitilis host strain and involved the electrospray ionization tandem mass spectrometry method to detect the resultant products. A semi-synthetic shuffling method [73] was utilized to allow for screening of the diversity of mutations generated in the initial rounds while reducing the library size to be screened. After three rounds of semi-synthetic shuffling, an improved mutant containing 10 mutations was inserted into production strains, and doramectin ratios were found to be improved more than 23-fold over the wild type [74].

3.4.5 Enantioselectivity The chiral molecule (R)-4-cyano-3-hydroxybutyric acid is another intermediate in the synthesis of atorvastatin (Figure 3.5), and its production enzymatically has been targeted by a number of groups [8,9,75]. DeSantis and coworkers [8,9] used a route involving the hydrolysis of 3-hydroxyglutaryl nitrile by a nitrilase enzyme. GSSM was used to increase the enantioselectivity under high substrate conditions. A chiral substrate was synthesized whereby one nitrile group contained the nitrogen isotope 15 N, meaning that the R- and S-products differed by one mass unit, which was detected by mass spectrometry. Screening of over 30 000 variants indicated the best mutant contained the Ala190His mutation, which increased the enantiomeric excess (ee) to 98.1% from 87.8% and had complete conversion in 15 h as opposed to 24 h when 2.25 M hydroxyglutaryl nitrile was used. Fox et al. [75] used a halohydrin dehydrogenase to produce ethyl (R)-4-cyano-3-hydroxybutyrate, a starting point for atorvastatin synthesis. The enzyme from Agrobacterium radiobacter acts on ethyl (S)-4-chloro-3-hydroxybutyrate, forming the product but initially at a low rate. A statistical model based on protein structure–activity relationships was incorporated, and 18 rounds of screening were carried

60


Figure 3.5 Directed evolution applied to enzymes for the production of atorvastatin synthesis intermediates. The substrates and products of the evolved enzymes are shown

out. The final mutant showed a 4000-fold improvement in volumetric production, with the ethyl (R)-4-cyano-3-hydroxybutyrate product being 99.5% pure and with an ee of over 99.9%. Epothilones are a class of molecules that show anticancer activity. Production of a synthetic intermediate was investigated through the action of an esterase on various sterically hindered 3-hydroxy esters [76]. No initial activity was observed, so a Pseudomonas fluorescens esterase was transformed into a mutator strain Epicurian coli and screened using an indicator in the growth plates that would produce a red color if hydrolysis occurred. An ee of 25% was achieved from a variant containing two mutations.

3.5 Conclusions and Future Directions The complexity of todays pharmaceutical compounds and an increasing awareness of the environmental impact of traditional chemical syntheses have opened the door to biocatalysis. Directed evolution is an integral tool in the development of synthetic enzymes, ensuring they are suitable for use in an industrial setting. The past success of this approach indicates that it will continue to provide many examples of safe and efficient production of chemical intermediates and medical compounds.


61

References [1] Sen, S., Venkata Dasu, V. and Mandal, B. (2007) Developments in directed evolution for improving enzyme functions. Applied Biochemistry and Biotechnology, 143, 212–223. [2] Yuan, L., Kurek, I., English, J. and Keenan, R. (2005) Laboratory-directed protein evolution. Microbiology and Molecular Biology Reviews, 69, 373–392. [3] Roodveldt, C., Aharoni, A. and Tawfik, D.S. (2005) Directed evolution of proteins for heterologous expression and stability. Current Opinion in Structural Biology, 15, 50–56. [4] Hibbert, E.G., Baganz, F., Hailes, H.C. et al. (2005) Directed evolution of biocatalytic processes. Biomolecular Engineering, 22, 11–19. [5] Leung, D.W., Chen, E. and Goeddel, D.V. (1989) A method for random mutagenesis of a defined DNA segment using a modified polymerase chain reaction. Technique, 1, 11–15. [6] Daugherty, P.S., Chen, G., Iverson, B.L. and Georgiou, G. (2000) Quantitative analysis of the effect of the mutation frequency on the affinity maturation of single chain Fv antibodies. Proceedings of the National Academy of Sciences of the United States of America, 97, 2029–2034. [7] Kunichika, K., Hashimoto, Y. and Imoto, T. (2002) Robustness of hen lysozyme monitored by random mutations. Protein Engineering, 15, 805–809. [8] DeSantis, G., Liu, J., Clark, D.P. et al. (2003) Structure-based mutagenesis approaches toward expanding the substrate specificity of D-2-deoxyribose-5-phosphate aldolase. Bioorganic and Medicinal Chemistry, 11, 43–52. [9] DeSantis, G., Wong, K., Farwell, B. et al. (2003) Creation of a productive, highly enantioselective nitrilase through gene site saturation mutagenesis (GSSM). Journal of the American Chemical Society, 125, 11476–11477. [10] Wong, T.S., Tee, K.L., Hauer, B. and Schwaneberg, U. (2004) Sequence saturation mutagenesis (SeSaM): a novel method for directed evolution. Nucleic Acids Research, 32, e26. [11] Seng Wong, T., Roccatano, D., Loakes, D. et al. (2007) Transversion-enriched sequence saturation mutagenesis (SeSaM-Tv þ ): a random mutagenesis method with consecutive nucleotide exchanges that complements the bias of error-prone PCR. Biotechnology Journal, 3, 74–82. [12] Hallet, B., Sherratt, D.J. and Hayes, F. (1997) Pentapeptide scanning mutagenesis: random insertion of a variable five amino acid cassette in a target protein. Nucleic Acids Research, 25, 1866–1867. [13] Murakami, H., Hohsaka, T. and Sisido, M. (2002) Random insertion and deletion of arbitrary number of bases for codon-based random mutation of DNAs. Nature Biotechnology, 20, 76–81. [14] Stemmer, W.P. (1994) DNA shuffling by random fragmentation and reassembly: in vitro recombination for molecular evolution. Proceedings of the National Academy of Sciences of the United States of America, 91, 10747–10751. [15] Crameri, A., Raillard, S.A., Bermudez, E. and Stemmer, W.P. (1998) DNA shuffling of a family of genes from diverse species accelerates directed evolution. Nature, 391, 288–291. [16] Coco, W.M. (2003) RACHITT: gene family shuffling by random chimeragenesis on transient templates. Methods in Molecular Biology (Clifton, NJ), 231, 111–127. [17] Shao, Z., Zhao, H., Giver, L. and Arnold, F.H. (1998) Random-priming in vitro recombination: an effective tool for directed evolution. Nucleic Acids Research, 26, 681–683. [18] Zhao, H., Giver, L., Shao, Z. et al. (1998) Molecular evolution by staggered extension process (StEP) in vitro recombination. Nature Biotechnology, 16, 258–261. [19] Ness, J.E., Kim, S., Gottman, A. et al. (2002) Synthetic shuffling expands functional protein diversity by allowing amino acids to recombine independently. Nature Biotechnology, 20, 1251–1255. [20] Hamamatsu, N., Aita, T., Nomiya, Y. et al. (2005) Biased mutation-assembling: an efficient method for rapid directed evolution through simultaneous mutation accumulation. Protein Engineering Design & Selection, 18, 265–271. [21] Abecassis, V., Pompon, D. and Truan, G. (2000) High efficiency family shuffling based on multi-step PCR and in vivo DNA recombination in yeast: statistical and functional analysis of a combinatorial library between human cytochrome P450 1A1 and 1A2. Nucleic Acids Research, 28, E88. [22] Volkov, A.A., Shao, Z. and Arnold, F.H. (1999) Recombination and chimeragenesis by in vitro heteroduplex formation and in vivo repair. Nucleic Acids Research, 27, e18. [23] Xu, S., Ju, J., Misono, H. and Ohnishi, K. (2006) Directed evolution of extradiol dioxygenase by a novel in vivo DNA shuffling. Gene, 368, 126–137.

62


[24] Joern, J.M., Meinhold, P. and Arnold, F.H. (2002) Analysis of shuffled gene libraries. Journal of Molecular Biology, 316, 643–656. [25] Ostermeier, M., Shim, J.H. and Benkovic, S.J. (1999) A combinatorial approach to hybrid enzymes independent of DNA homology. Nature Biotechnology, 17, 1205–1209. [26] Sieber, V., Martinez, C.A. and Arnold, F.H. (2001) Libraries of hybrid proteins from distantly related sequences. Nature Biotechnology, 19, 456–460. [27] Lutz, S., Ostermeier, M. and Benkovic, S.J. (2001) Rapid generation of incremental truncation libraries for protein engineering using alpha-phosphothioate nucleotides. Nucleic Acids Research, 29, E16. [28] Lutz, S., Ostermeier, M., Moore, G.L. et al. (2001) Creating multiple-crossover DNA libraries independent of sequence identity. Proceedings of the National Academy of Sciences of the United States of America, 98, 11248–11253. [29] Kawarasaki, Y., Griswold, K.E., Stevenson, J.D. et al. (2003) Enhanced crossover SCRATCHY: construction and high-throughput screening of a combinatorial library containing multiple non-homologous crossovers. Nucleic Acids Research, 31, e126. [30] Udit, A.K., Silberg, J.J. and Sieber, V. (2003) Sequence homology-independent protein recombination (SHIPREC). Methods in Molecular Biology (Clifton, NJ), 231, 153–163. [31] Kolkman, J.A. and Stemmer, W.P. (2001) Directed evolution of proteins by exon shuffling. Nature Biotechnology, 19, 423–428. [32] Tsuji, T., Onimaru, M. and Yanagawa, H. (2001) Random multi-recombinant PCR for the construction of combinatorial protein libraries. Nucleic Acids Research, 29, E97. [33] Coco, W.M., Encell, L.P., Levinson, W.E. et al. (2002) Growth factor engineering by degenerate homoduplex gene family recombination. Nature Biotechnology, 20, 1246–1250. [34] Bittker, J.A., Le, B.V. and Liu, D.R. (2002) Nucleic acid evolution and minimization by nonhomologous random recombination. Nature Biotechnology, 20, 1024–1029. [35] Bittker, J.A., Le, B.V., Liu, J.M. and Liu, D.R. (2004) Directed evolution of protein enzymes using nonhomologous random recombination. Proceedings of the National Academy of Sciences of the United States of America, 101, 7011–7016. [36] Boersma, Y.L., Droge, M.J. and Quax, W.J. (2007) Selection strategies for improved biocatalysts. FEBS Journal, 274, 2181–2195. [37] Daugherty, P.S. (2007) Protein engineering with bacterial display. Current Opinion in Structural Biology, 17, 474–480. [38] Francis, J.C. and Hansche, P.E. (1972) Directed evolution of metabolic pathways in microbial populations. I. Modification of the acid phosphatase pH optimum in S. cerevisiae. Genetics, 70, 59–73. [39] Hall, B.G. (1981) Changes in the substrate specificities of an enzyme during directed evolution of new functions. Biochemistry, 20, 4042–4049. [40] Miller, B.G. and Raines, R.T. (2004) Identifying latent enzyme activities: substrate ambiguity within modern bacterial sugar kinases. Biochemistry, 43, 6387–6392. [41] Hwang, B.Y., Oh, J.M., Kim, J. and Kim, B.G. (2006) Pro-antibiotic substrates for the identification of enantioselective hydrolases. Biotechnology Letters, 28, 1181–1185. [42] Kleeb, A.C., Edalat, M.H., Gamper, M. et al. (2007) Metabolic engineering of a genetic selection system with tunable stringency. Proceedings of the National Academy of Sciences of the United States of America, 104, 13907–13912. [43] Baker, K., Bleczinski, C., Lin, H. et al. (2002) Chemical complementation: a reaction-independent genetic assay for enzyme catalysis. Proceedings of the National Academy of Sciences of the United States of America, 99, 16537–16542. [44] Lin, H., Tao, H. and Cornish, V.W. (2004) Directed evolution of a glycosynthase via chemical complementation. Journal of the American Chemical Society, 126, 15051–15059. [45] Van Sint Fiet, S., van Beilen, J.B. and Witholt, B. (2006) Selection of biocatalysts for chemical synthesis. Proceedings of the National Academy of Sciences of the United States of America, 103, 1693–1698. [46] Mohn, W.W., Garmendia, J., Galvao, T.C. and de Lorenzo, V. (2006) Surveying biotransformations with a la carte genetic traps: translating dehydrochlorination of lindane (gamma-hexachlorocyclohexane) into lacZ-based phenotypes. Environmental Microbiology, 8, 546–555. [47] Becker, S., Schmoldt, H.U., Adams, T.M. et al. (2004) Ultra-high-throughput screening based on cell-surface display and fluorescence-activated cell sorting for the identification of novel biocatalysts. Current Opinion in Biotechnology, 15, 323–329.


63

[48] Gai, S.A. and Wittrup, K.D. (2007) Yeast surface display for protein engineering and characterization. Current Opinion in Structural Biology, 17, 467–473. [49] Smith, G.P. (1985) Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science, 228, 1315–1317. [50] Paschke, M. (2006) Phage display systems and their applications. Applied Microbiology and Biotechnology, 70, 2–11. [51] Boder, E.T. and Wittrup, K.D. (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nature Biotechnology, 15, 553–557. [52] Hanes, J. and Pluckthun, A. (1997) In vitro selection and evolution of functional proteins by using ribosome display. Proceedings of the National Academy of Sciences of the United States of America, 94, 4937–4942. [53] Nemoto, N., Miyamoto-Sato, E., Husimi, Y. and Yanagawa, H. (1997) In vitro virus: bonding of mRNA bearing puromycin at the 30 -terminal end to the C-terminal end of its encoded protein on the ribosome in vitro. FEBS Letters, 414, 405–408. [54] Roberts, R.W. and Szostak, J.W. (1997) RNA-peptide fusions for the in vitro selection of peptides and proteins. Proceedings of the National Academy of Sciences of the United States of America, 94, 12297–12302. [55] Tawfik, D.S. and Griffiths, A.D. (1998) Man-made cell-like compartments for molecular evolution. Nature Biotechnology, 16, 652–656. [56] Griffiths, A.D. and Tawfik, D.S. (2003) Directed evolution of an extremely fast phosphotriesterase by in vitro compartmentalization. The EMBO Journal, 22, 24–35. [57] Aharoni, A., Amitai, G., Bernath, K. et al. (2005) High-throughput screening of enzyme libraries: thiolactonases evolved by fluorescence-activated sorting of single cells in emulsion compartments. Chemistry & Biology, 12, 1281–1289. [58] Mastrobattista, E., Taly, V., Chanudet, E. et al. (2005) High-throughput screening of enzyme libraries: in vitro evolution of a beta-galactosidase by fluorescence-activated sorting of double emulsions. Chemistry & Biology, 12, 1291–1300. [59] Blow, N. (2008) Lab automation: tales along the road to automation. Nature Methods, 5, 109–112. [60] Aucamp, J.P., Cosme, A.M., Lye, G.J. and Dalby, P.A. (2005) High-throughput measurement of protein stability in microtiter plates. Biotechnology and Bioengineering, 89, 599–607. [61] Geddie, M.L., Rowe, L.A., Alexander, O.B. and Matsumura, I. (2004) High throughput microplate screens for directed protein evolution. Methods in Enzymology, 388, 134–145. [62] Huang, K.S., Mark, D. and Gandenberger, F.U. (2006) High-throughput measurements of biochemical responses using the plate::vision multimode 96 minilens array reader. Methods in Enzymology, 414, 589–600. [63] Ibrahim, S.F. and van den Engh, G. (2007) Flow cytometry and cell sorting. Advances in Biochemical Engineering/Biotechnology, 106, 19–39. [64] Jin, M., Fischbach, M.A. and Clardy, J. (2006) A biosynthetic gene cluster for the acetyl-CoA carboxylase inhibitor andrimid. Journal of the American Chemical Society, 128, 10660–10661. [65] Fischbach, M.A. and Walsh, C.T. (2006) Assembly-line enzymology for polyketide and nonribosomal peptide antibiotics: logic, machinery, and mechanisms. Chemical Reviews, 106, 3468–3496. [66] Pohlmann, J., Lampe, T., Shimada, M. et al. (2005) Pyrrolidinedione derivatives as antibacterial agents with a novel mode of action. Bioorganic & Medicinal Chemistry Letters, 15, 1189–1192. [67] Fischbach, M.A., Lai, J.R., Roche, E.D. et al. (2007) Directed evolution can rapidly improve the activity of chimeric assembly-line enzymes. Proceedings of the National Academy of Sciences of the United States of America, 104, 11951–11956. [68] Otten, L.G., Sio, C.F., Reis, C.R. et al. (2007) A highly active adipyl-cephalosporin acylase obtained via rational randomization. FEBS Journal, 274, 5600–5610. [69] Woodyer, R., van der Donk, W.A. and Zhao, H.M. (2006) Optimizing a biocatalyst for improved NAD(P)H regeneration: directed evolution of phosphite dehydrogenase. Combinatorial Chemistry & High Throughput Screening, 9, 237–245. [70] Reetz, M.T., Carballeira, J.D. and Vogel, A. (2006) Iterative saturation mutagenesis on the basis of B factors as a strategy for increasing protein thermostability. Angewandte Chemie (International Edition in English), 45, 7745–7751. [71] Jennewein, S., Schurmann, M., Wolberg, M. et al. (2006) Directed evolution of an industrial biocatalyst: 2-deoxyD-ribose 5-phosphate aldolase. Biotechnology Journal, 1, 537–548. [72] Stutzman-Engwall, K., Conlon, S., Fedechko, R. et al. (2003) Engineering the aveC gene to enhance the ratio of doramectin to its CHC-B2 analogue produced in Streptomyces avermitilis. Biotechnology and Bioengineering, 82, 359–369.

64


[73] Crameri, A., Stemmer, W.P.C., Minshull, J. et al. (2000) Oligonucleotide mediated nucleic acid recombination. PCT International Patent Application WO 00/42561. [74] Stutzman-Engwall, K., Conlon, S., Fedechko, R. et al. (2005) Semi-synthetic DNA shuffling of aveC leads to improved industrial scale production of doramectin by Streptomyces avermitilis. Metabolic Engineering, 7, 27–37. [75] Fox, R.J., Davis, S.C., Mundorff, E.C. et al. (2007) Improving catalytic function by ProSAR-driven enzyme evolution. Nature Biotechnology, 25, 338–344. [76] Bornscheuer, U.T., Altenbuchner, J. and Meyer, H.H. (1998) Directed evolution of an esterase for the stereoselective resolution of a key intermediate in the synthesis of epothilones. Biotechnology and Bioengineering, 58, 554–559.

4 Applications of Reaction Engineering to Industrial Biotransformations Lutz Hilterhaus and Andreas Liese

4.1 Introduction The enzyme is the elementary operational component in a bioprocess, while the spectrum of bioprocesses ranges from reactions with single purified enzymes to complex cellular, animal and plant systems [1]. The classification of the different biocatalysts can be done by differentiation into enzymatic biotransformations and metabolic bioconversions (Figure 4.1). Enzymatic biotransformations are characterized by a low number of specific reactions, whereas metabolic bioconversions need the metabolic system of living and growing cells [2]. The latter can be bacteria, yeasts, or fungi, as well as single cultivated mammalian and plant cells. In contrast to these, enzymatic biotransformations are carried out by resting cells or isolated proteins in bioreactors. Furthermore, one can distinguish between the products being formed, which are in general chemical intermediates of pharmaceuticals in enzymatic biotransformations and additionally pharmaceutical proteins in metabolic bioconversions. Biotransformations are becoming a key component in the toolbox of process chemists, with a place alongside chemocatalysis and chromatographic separations [3]. The aims of all bioprocesses, and especially of biotransformations, can be summarized as follows: . . . . .

high yield, product concentration and productivity; easy downstream processing; minimal amounts of by-products; high stability and safety of biocatalyst; optimal reaction conditions – for example, temperature, oxygen supply, shear sensitivity, foam formation, and so on.


66


Figure 4.1

Comparison of metabolic bioconversions and enzymatic biotransformations

The combination of kinetics, hydrodynamics and transport phenomena that provide the proper scale-up of bioreactors from laboratory to the industrial scale also has to be taken into account [4]. Different process solutions will be discussed within this chapter in detail, starting from concrete problem and illustrating possibilities to overcome this problem.

4.2 Metabolic Bioconversion The production of biopharmaceuticals by plants was reviewed by Fischer et al. [5] and recently by Deckor and Reski [6]. Limitations occurring within this method are exemplified for eukaryotic protein production, which are the low yields that are achieved for many proteins as products, difficulties in downstream processing and the presence of nonauthentic glycan structures on recombinant human proteins. The general approach to affect recombinant protein yields is to maximize both the efficiency of all stages of gene expression and protein stability [7]. Additionally, transport limitations have been regarded as one of the problems leading to process yield reduction in large-scale bioprocesses [8]. Although high-level expression is necessary to provide good yields in whole-cell production systems, the efficient recovery of recombinant proteins must also be optimized. Secretion systems are advantageous because no disruption of whole cells is necessary during protein recovery. Nevertheless, the recombinant proteins may be unstable in the culture medium. The use of affinity tags to facilitate the recovery of proteins is a useful strategy as long as the tag can be removed after purification to restore the native structure of the protein. Summarizing these attempts to overcome limitations, it is obvious that the yield of production of proteins is addressed by genetic engineering and screening of cells, the downstream processing is simplified by tags for purification, and the correct glycoprotein synthesis is established by an adequate selection of the organism. Besides the metabolic system of living and growing eukaryotes, the use of prokaryotes in industrial production of pharmaceutical precursors by fermentation is a common strategy [9,10]. Here, the crucial steps are the inocula preparation to reach reproducible yields and the

Applications of Reaction Engineering to Industrial Biotransformations

67

optimization and control of reaction parameters like pH, aeration, feed rate of carbon source, or process temperature. Preconditions of bioprocesses in scales up to 500 m3 are the increased hydrostatic pressure in the vessel and a limited agitation speed resulting in a limited mixing time. Furthermore, pCO2 is usually increased because of lower aeration in bigger volume bioreactors. This point might be crucial for scale-up of such bioprocesses.

4.3 Enzymatic Biotransformations Next to the metabolic bioconversions there are enzymatic biotransformations, which are characterized by a low number of fundamental well-defined reactions [11]. However, there are often inherent limitations that need to be addressed; the crucial ones are the following: . . . . . .

requirement of expensive cofactor; maximum of 50% yield in racemic resolutions; thermodynamic limitation of maximum conversion; by-product formation; substrate inhibition; low substrate solubility.

Approaches to overcome these limitations by means of reaction engineering will be discussed in the following.

4.3.1 Cofactor Regeneration In respect of designing an economic production process, the stoichiometric cofactor required in carbonyl reductions or the respective oxidation reactions needs to be minimized; that is, enabled by recycling of the cofactor. The measure for the efficiency of the recycling process is the total turnover number (TTN), which describes the moles of product synthesized in relation to the moles of cofactor needed. The different approaches in cofactor recycling were recently reviewed by Goldberg et al. [12]. For the synthesis of amphetamines, (S)-1-phenyl-2-propanol (2) is used as an intermediate. (S)-4-Phenyl-2-butanol is used as a precursor for anti-hypertensive agents and spasmolytics, as well as anti-epileptics. Both intermediates can be synthesized by applying alcohol dehydrogenase (ADH) from Rodococcus erythropolis. The majority of ADHs are dependent on the nicotinamide cofactors b-1,4-nicotinamide adenine dinucleotide (NADH) or b-1,4nicotinamide adenine dinucleotide phosphate (NADPH). These cofactors are too expensive to be used stoichiometrically (with prices of D 1300 per mole for NAD þ ) and there has been significant interest in developing efficient cofactor regeneration processes [13]. Within the process described here, the production of NADH is facilitated by means of the enzymatic method. The cofactor regeneration is carried out via the enzyme-coupled approach with formate dehydrogenase from Candida boidinii utilizing formate which is oxidized to CO2. The reactor concept consists of a continuously operated stirred-tank reactor which is coupled to two membrane modules [14,15]. This enzyme bi-membrane reactor consists of three loops: The reactant (e.g. phenylpropanone) is directly titrated into the aqueous phase (Figure 4.2). This aqueous phase is pumped in a loop through the ultrafiltration module. The hydrophilic membrane retains the enzymes and the filtrate is pumped through the second loop. This

68


Figure 4.2 Enzyme-bi-membrane-reactor: synthesis of 1-phenyl-2-propanol from 1-phenyl-2propanone applying a stirred-tank reactor, ultrafiltration module, extraction module and distillation

aqueous solution contents, products, reactants and cofactors, is passed through the lumen of the extraction module. Here, a microporous, hydrophobic hollow-fiber membrane separates the aqueous and the organic phases and the product and reactants are extracted, whereas the charged cofactors are retained in the aqueous phase. The organic solvent is recycled by continuous distillation and the product remains at the bottom of the distillation column. Next to the cofactor regeneration, this process also allows the usage of reactants of low solubility. Here, a TTN of up to 1350 and a space–time yield (STY) of 64 g L1 day1 is reached. The thermostability, long-term activity and storage half-life time of the enzyme can be enhanced by use of entrapment of the enzyme and its cofactor in polyvinyl alcohol gel beads [16,17]. A simpler approach is carried out within the production of (R)-methyl-3-(4-fluorophenyl)-2hydroxypropanoate, which is a building block for the synthesis of Rupintrivir, a rhinovirus protease inhibitor. Here, the continuously operated stirred-tank reactor is connected with only one ultrafiltration module to retain the enzymes D-lactate dehydrogenase and formate dehydrogenase [18]. STYs up to 560 g L1 day1 are reached. The first industrial application of the enzyme membrane reaction technology was published by the Degussa Company in the 1980s. This process is used in the production of L-tert-leucine and will be discussed later. An overview about the status quo of membrane reactor technology is given by Giorno and Drioli [19] and more recently by Judd [20]. Besides the enzyme-coupled approach, a second system, substrate-coupled cofactor regeneration, is becoming more and more prominent in industrial processes. Here, only one enzyme is utilized, with oxidation and reduction being carried out at the same time [21,22]. This, for example, is realized in the production of chiral b-hydroxyesters, which are widely applied as building blocks for pharmaceuticals, agrochemicals and fragrances. The cofactor regeneration makes use of the oxidation of isopropanol to yield acetone, which is done by the same enzyme at the same time catalyzing the enantioselective reduction of prochiral ketones. However, these processes most of the time suffer from thermodynamic limitations. One example is the asymmetric reduction of ethyl acetoacetate. The continuous stripping of the co-product acetone shifts the equilibrium reaction towards complete turnover [23–25]. Further, continuous reuse of the aqueous phase in a standard stirred batch reactor is possible. The product is extracted from the aqueous phase and obtained by distillation of the organic solvent. The enzyme-coupled process illustrated in Figure 4.2 requires the application of two enzymes at the same time. The substrate-coupled approach carried out by Codexis is illustrated in Figure 4.3,


69

Figure 4.3 Continuous stripping: synthesis of (R)-ethyl-3-hydroxybutyrate from ethyl acetoacetate applying a stirred-tank reactor, stripping module, extraction module and distillation

whereby only one enzyme is applied for the production of the desired compound. Therefore, this cofactor regeneration is a powerful alternative in comparison with the enzyme-coupled approach, especially in the production of (R)-ethyl-3-hydroxybutyrate, obtaining an STY of 92 g L1 day1. To produce 2-hydroxy-4-phenyl-butyric acid, which is a precursor for different ACEinhibitors, immobilized whole cells of Proteus vulgaris are used in a fixed-bed reactor [26]. The cells are immobilized in ionotropic gel-beads and small units of biomass allow the production of high amounts of the product. Instead of NADH or NADPH electron mediators like benzyl viologen or carbamoyl methyl viologen are used. The enzymes D-lactate-dehydrogenase and formate dehydrogenase in the cells are capable of reducing and oxidizing the mediators. The process has to be carried out under a pressure of 3 bar to avoid degassing of CO2. These first examples illustrate the importance of a sufficient separation of products and byproducts, whereas membranes are one possibility in pharmaceutical production to obtain this aim. Therefore, they are one key tool to obtaining better quality products and environmentally friendly processes. For a more detailed article about the state of the art of membranes in biotechnology, see Rios et al. [27]. At the same time, it can be seen that stoichiometric cofactor need is no longer a limitation for industrial biotransformations, since they can be overcome with efficient recyclization methods.

4.3.2 Racemic Mixtures In carrying out kinetic resolution, these in the standard approach are limited to 50% yield regarding the racemate. However, different approaches were developed [28] to overcome this limitation. The classical standard solution is to reracemize the unconverted enantiomer. A more advanced solution is the establishment of a dynamic kinetic resolution that has considerably expanded the synthetic scope of chemical processes. Here, the unconverted enantiomer is, in contrast to the latter method, racemized in situ. A great number of novel enzymatic methods have been developed [29]. Within this chapter, process solutions for enzymatic resolutions of racemic mixtures will be highlighted. (S)-N-(tert-Butoxycarbonyl)-3-hydroxymethylpiperidine is a building block in the synthesis of a potent tryptase inhibitor. It is produced by stereospecific esterification of the racemic alcohol with succinic anhydride [30]. The enzymatic resolution is followed up by separation of

70


Figure 4.4 Stereospecific esterification: synthesis of (R)-N-(tert-butoxycarbonyl)-3-hydroxymethylpiperidine from (R,S)-N-(tert-butoxycarbonyl)-3-hydroxymethylpiperidine applying a sequence of stirred-tank reactor, extraction module, second stirred-tank reactor and second extraction module

the unreacted (R)-alcohol by extraction with a basic NaHCO3 solution (Figure 4.4). The subsequent hydrolysis of the (S)-hemisuccinate with NaOH provides the desired (S)-alcohol. To improve the enantiomeric excess and the yield, repeated esterification of the enriched alcohol can be pursued. Therefore, this process carried out by Bristol-Myers-Squibb is the analog to the enantioselective hydrolysis shown above. The production process for (S)-phenylalanine as an intermediate in aspartame perpetuates the principle of reracemization of the nondesired enantiomer (Figure 4.5) in a hollow fiber/ liquid membrane reactor. Asymmetric hydrolysis of the racemic phenylalanine isopropylester at pH 7.5 leads to enantiopure phenylalanine applying subtilisin Carlsberg. The unconverted enantiomer is continuously extracted via a supported liquid membrane [31] that is immobilized in a microporous membrane into an aqueous solution of pH 3.5. The desired hydrolysis product is charged at high pH and cannot, therefore, be extracted into the acidic solution [32].

Figure 4.5 Enzymatic resolution: synthesis of phenylalanine from phenylalanine isopropylester applying a continuous stirred-tank reactor with continuous extraction of the unconverted enantiomer


71

Figure 4.6 Classical kinetic resolution with subsequent reracemization of unconverted enantiomer: Synthesis of pantoic acid from pantolactone applying a stirred-tank reactor, extraction module and racemization step

The continuously operated stirred-tank reactor with continuous extraction of the unconverted enantiomer yields an enantiomeric excess of 95%. Afterwards, the unconverted enantiomer is racemized and reused in the synthesis process carried out by Coca Cola. D-Pantolactone and L-pantolactone are used as chiral intermediates in chemical synthesis, whereas pantoic acid is used as a vitamin B2 complex. All can be obtained from racemic mixtures by consecutive enzymatic hydrolysis and extraction. Subsequently, the desired hydrolysed enantiomer is lactonized, extracted and crystallized (Figure 4.6). The nondesired enantiomer is reracemized and recycled into the plug-flow reactor [33,34]. Herewith, a conversion of 90–95% is reached, meaning that the resolution of racemic mixtures is an alternative to a possible chiral synthesis. The applied g-lactonase from Fusarium oxysporum in the form of resting whole cells immobilized in calcium alginate beads retains more than 90% of its initial activity even after 180 days of continuous use. The biotransformation yielding D-pantolactone in a fixed-bed reactor skips several steps here that are necessary in the chemical resolution. Hence, the illustrated process carried out by Fuji Chemical Industries Co., Ltd is an elegant way for resolution of racemic mixtures. An alternative to extraction crystallization is used to obtain a desired enantiomer after asymmetric hydrolysis by Evonik Industries. In such a way, L-amino acids for infusion solutions or as intermediates for pharmaceuticals are prepared [35,36]. For example, nonproteinogenic amino acids like L-norvaline or L-norleucine are possible products. The racemic N-acteyl-amino acid is converted by acylase 1 from Aspergillus oryzae to yield the enantiopure L-amino acid, acetic acid and the unconverted substrate (Figure 4.7). The product recovery is achieved by crystallization, benefiting from the low solubility of the product. The product mixture is filtrated by an ultrafiltration membrane and the unconverted acetyl-amino acid is reracemized in a subsequent step. The product yield is 80% and the enantiomeric excess 99.5%. Within the synthesis of pyrethroids, which are used as insecticides, (S)-4-hydroxy-3-methyl2-prop-2-ynyl-cyclopent-2-enone is needed. Starting from a racemic mixture of the esterified

72


Figure 4.7 Classical kinetic resolution: synthesis of L-methionine from N-acetyl-methionine applying an ultrafiltration-membrane reactor and crystallization step as well as racemization step

cyclopentenone, enantioselective lipase-catalyzed hydrolysis is carried out [37]. The alcohol which is formed is sulfonated in the presence of the acylated compound. The following hydrolysis of this sulfonated enantiomer takes place under inversion of the chiral center. Because of this chemical follow-up step, very high overall yields regarding the racemate are obtained, as well as an enantiomeric excess of 99.2%. Several hundred tons of L-methionine per year are produced by enzymatic conversion in an enzyme membrane reactor. An alternative approach is dynamic resolution, where the unconverted enantiomer is racemized in situ. Starting from racemic N-acetyl-amino acid, the enantioselective L-acylase is used in combination with an N-acyl-amino acid racemase to enable nearly total conversion of the substrate. Kanegafuchi Chemical Industries produce D-p-hydroxyphenyl glycine, which is a key raw material for the semisynthetic penicillins ampicillin and amoxycillin. Here, an enantioselective hydantoinase is applied to convert the hydantoin to the D-p-hydroxyphenyl glycine. The quantitative conversion of the amide hydrolysis is achieved because of the in situ racemization of the unreacted hydantoins. Under the conditions of enzymatic hydrolysis, the starting material readily racemizes. Therefore, this process enables the stereospecific preparation of various amino acids at a conversion of 100% [38].

4.3.3 Equilibrium Conversion Sometimes, the maximum conversion is limited by thermodynamics. In such cases, different approaches need to be taken to surpass the thermodynamic limitation. Standard methods that are also realized on an industrial scale are: . .

in situ subsequent chemical or enzymatic reaction; in situ product removal.

The first example demonstrates an in situ started chemical reaction sequence, modifying the reaction product in such a manner that the reverse reaction can no longer be initiated. Amino acids for infusion solutions are produced by amino group transfer reactions applying transaminases. Here, a major drawback is the equilibrium conversion of only 50%. Therefore,


73

Figure 4.8 Coupling with irreversible step: synthesis of D-amino acids from racemic mixture applying a batch process coupling the reaction to the formation of the gaseous side-product CO2

in the production by NSC Technologies, Monsanto, D-aspartate is used as the amino donor for D-amino acid transaminase [39]. The resulting product decarboxylates to pyruvic acid and CO2. Applying acetolactate synthase, the pyruvic acid is dimerized to acetolactate, which undergoes spontaneous decarboxylation to acetoin (Figure 4.8). This is inert and can be removed easily. Using amino acid racemase, the amino group donor is accessible from racemic mixtures of amino acids. The industrial process uses Escherichia coli cells in which all necessary genetic information is cloned to conduct all steps of synthesis (compare Figure 4.1). High yields of D-amino acids like alanine, phenylalanine, leucine, glutamic acid and tyrosine are achieved [40]. Next to reactions catalyzed by transaminases, hydrolase-catalyzed reactions also lead to limitations regarding the equilibrium. This problem occurs during ester synthesis, because this condensation reaction produces water. The equilibrium is shifted by high amounts of water towards the reactants; therefore, an efficient removal is necessary to reach high conversions. Here, two process setups of Unichema Chemie BV will be discussed illustrating in situ product removal [41]. The first setup is based on azeotropic distillation of the water produced

74


Figure 4.9 Azeotropic distillation: synthesis of isopropyl palmitate from palmitic acid applying a stirred-tank reactor with azeotropic distillation of the water formed

(Figure 4.9) to synthesize isopropyl myristate. This is enabled by a continuous feed of 2-propanol to the reactor, which forms an azeotropic mixture with water. This mixture is distilled, thereby removing the water produced. Afterwards, the immobilized enzyme used can be easily removed by filtration. The feasibility of this technique is also illustrated within the synthesis of glucose stearate. A mixture of ethyl methylketone and hexane as solvent is used, forming an azeotropic mixture with the water produced [42]. Alternatively, a sequence of pervaporation steps can be used to remove the reaction water (Figure 4.10). The reaction solution is pumped through a first fixed-bed reactor and afterwards a pervaporation module is used to remove the water at 80 C. The solution is subsequently cooled to 60 C and passed through a second reactor unit. After the next pervaporation step the water content is lowered to 0.2 wt%. By this process, isopropyl myristate is produced from myristyric acid. This and other esters of isopropyl alcohol are used in soaps, skin creams, lubricants and greases [43]. In comparison with low viscosity isopropanol-based ester synthesis or synthesis applying solvents, solvent-free synthesis of surfactants from highly viscous or high-melting reactants (e.g. sugar and glycerol derivatives) is not possible by the existing fixed-bed technology due to

Figure 4.10 Pervaporation: synthesis of isopropyl palmitate from palmitic acid applying a sequence of fixed-bed reactors and pervaporation modules


75

Figure 4.11 Stripping in solvent-free medium: Synthesis of myristyl myristate from myristyric acid and myristyl alcohol applying a bubble column reactor stripping the reaction water

the pressure drop. Production in conventional stirred-tank reactors is also prohibited due to the mechanical stress caused by the stirrer, which usually destroys the heterogeneous biocatalysts used and thereby reduces its half-life time significantly. Further, solvent-free esterification is limited by slow removal of the reaction water formed and, therefore, by incomplete reaction. The slow water removal is obvious within the synthesis of, for example, myristyl myristate determining the total reaction time. In a stirred-tank reactor it takes 24 h to reach a conversion of 99.6% and in a fixed-bed reactor 14 h. Therefore, a new synthesis platform (Figure 4.11) which also enables conversion of highly viscous polyols and fatty acids from renewable resources to ester-based surfactants was designed. It is used by Evonik on a pilot scale, outperforming conventional methods, such as stirred-tank or fixed-bed reactors. In contrast to the setups introduced before, conversion of > 99.6% is already obtained after 5.5 h in the bubble column reactor [44–47].

Figure 4.12 Precipitation of product: synthesis of aspartame from phenylalanine methylester applying a batch process in combination with two filtration steps

76


A more complex production process which is limited by the products is the synthesis of aspartame, which is used as a low-calorie sweetener in food and in pharmaceuticals. In contrast to the biotransformation conducted by DSM, the chemical synthesis shows one central drawback, which is the formation of the b-isomer as by-product. Since the reaction is limited by the equilibrium, the products have to be removed to achieve high yields. Therefore, an excess of phenylalanine methylester, being inert to the reaction, is added [48]. By this, the carboxylic anion of the protected aspartame forms a poorly soluble adduct, which precipitates from the reaction mixture. This precipitate is removed by filtration and subsequently the protecting groups are removed again (Figure 4.12). The nontargeted by-product L-amino acid is racemized and recycled.

4.3.4 By-Product Formation In the case of by-product formation that negatively influences the course of reaction, the same procedures can be integrated as in the case of limitations by the equilibrium (Section 4.3.3). Within the production of dinitrobenzyl by Novartis, which is a pharmaceutical intermediate, hydrogen peroxide is an undesired by-product. Here, an enzymatic step is used to remove this by-product, whereas the reaction itself is catalyzed chemically [49]. The oxidative coupling of nitrotoluene to dinitrodibenzyl is carried out in a stirred-tank reactor, where the reaction solution consists of water, an organic and a solid phase (Figure 4.13). The reaction solution, after adjusting the pH with acetic acid, is passed to a second reactor within a cascade of three reactors. Here, catalase is used to remove hydrogen peroxide. Within the last reactor a continuous flow of nitrogen is applied to dilute the oxygen formed. This process example illustrates the potential of the combination of chemical synthesis and a biotransformation for downstream processing. The preparation of fatty acids substituted within an aliphatic chain is necessary to prepare dermatological pharmaceuticals. The unsaturated acyl fatty acids are the intermediate and can be produced by Kao Corporation starting from low-cost saturated compounds applying whole

Figure 4.13 Enzymatic by-product removal: synthesis of dinitrodibenzyl from nitrotoluene applying a cascade of continuous stirred-tank reactors while degassing with nitrogen


77

Figure 4.14 Chemical adducts for by-product removal: synthesis of isopropyl-cis-D6-hexadecanoate from isopropylpalmitate applying a repetitive batch process using a sequence of stirred-tank reactor, extraction module, filtration step and chromatographic downstream processing

cells from Rhodococcus sp. KSM-B-MT66 [50]. This mutant cis-desaturates a broad variety of acyl fatty acids at central positions of the chain. The reaction is started as an oil-in-water emulsion which is inverted by addition of isopropyl hexadecanoate [51]. The aqueous phase is passed through a hydrophobic hollow-fiber module to recover the product and residual substrate into the oil phase (Figure 4.14). The product separation is performed by the urea-adduct method and chromatography on silica gel. This process example (in contrast to the enzymatic downstream processing by catalase mentioned before) illustrates the potential of the combination of biotransformation and downstream processing based on chemical adducts. Cyclodextrins serve as molecular hosts and are used in the food industry and in formulation of pharmaceuticals for capturing and retaining specific molecules. The enzymatic synthesis of cyclodextrins by Mercian Co., Ltd applying glycosyltransferases leads to a mixture of a-, b- and g-cyclic oligosaccharides. To establish an economic cyclodextrin production, the separation of the cyclodextrins from the reaction medium must be carried out [52]. This is required because the reaction medium contains many by-products and the enzyme is inhibited by a high cyclodextrin concentration. The selective adsorption of a- and b-cyclodextrin on chitosan beads with appropriate ligands enables the separation: a-cyclodextrins show specific interaction with stearic acid, b-cyclodextrins with cyclohexanepropanamide-n-caproic acid. To enable effective adsorption, the temperature is lowered to 30 C prior to passing through the adsorption column (Figure 4.15). At this temperature, almost no cyclodextrins are formed; before reentering the reactor the temperature of the solution is readjusted to 55 C by passing through a heat exchanger. The dissolved enzyme used in this process does not adsorb on the chitosan beads when the reaction solution contains more than 3 % w/v sodium chloride.

78


Figure 4.15 Selective adsorption: synthesis of a-cyclodextrin from starch applying a batch process using a sequence of stirred-tank reactor, heat exchanger modules and adsorption step

L-Aspartic

acid and D-aspartic acid are used as food additives and in infusion solutions;

D-aspartic acid is also an intermediate for the semi-synthetic penicillin apoxycillin. The process

for production of L-aspartic acid uses L-aspartic b-decarboxylase from Pseudomonas dacunhae in combination with aspartase from E. coli [53]. The two-step biotransformation is carried out by Tanabe Seiyaku Co., Ltd using immobilized cells of E. coli and P. dacunhae in two separated reactors. The aspartate-catalyzed synthesis of L-aspartic acid from fumarate is carried out at an optimal pH of 8.5, whereas L-aspartate b-decarboxylase has an optimum of pH 6.0. Within the production of L-alanine and D-aspartic acid, the by-product CO2 causes difficulties regarding the flow conditions in a fixed-bed reactor. Because the production of CO2 occurs stoichiometrically, there are difficulties in obtaining plug-flow conditions in fixed-bed reactors and a pH shift occurs. Therefore, a pressurized fixed-bed reactor with 10 bar was designed, preventing degassing of CO2 in the packed bed (Figure 4.16).

Figure 4.16 Pressurized fixed-bed reactor: synthesis of L-aspartic acid from fumarate applying a plug flow reactor followed by a crystallization step for downstream processing


79

Acryl amide is an important bulk chemical used in coagulators, soil conditioners and stock additives. The chemical synthesis has several drawbacks because the rate of acryl amide formation is lower than the formation of the by-product acrylic acid [54]. Further, the double bonds of the reactants and products cause by-product formations as well as formation of polymerization products. As a result of optimization with methods of molecular engineering, a very high activity of the biocatalyst nitrile hydratase at low temperature is yielded, enabling a successful biotransformation that is superior to the chemical route. Here, the synthesis is carried out at a low temperature of about 5 C, showing a conversion of 100%.

4.3.5 Substrate Inhibition The orally active benzodiazepine LY 300164 is produced from (3,4-methylenedioxyphenyl)-2propanol. This is synthesized by Eli Lilly applying suspended whole cells of Zygosaccharomyces rouxii in a slurry reactor [55]. The toxic limit for the substrate is 6 g L1; therefore, the substrate is adsorbed on XAD-7 resin (Figure 4.17). Even though the amount of substrate applied would result in an even higher substrate concentration, because of the equilibrium between adsorbed and free compound a residual substrate concentration of 2 g L1 results. In this way the volumetric productivity is held at a high level and at the same time the substrate inhibition is overcome. Here, the XAD-7 resin has the advantage of being nontoxic and nondenaturing. In addition, the resin can be reused for three times without loss in performance. The process consists of Rosenmund agitated filter-dryer which enables a filtration without clogging of the filter. The yeast cells from Z. rouxii are separated from the resin by filtration through a 150 mm filter screen. Here, the resin with the adsorbed product is retained, whereas the yeast cells stay in the filtrate. Afterwards, washing the resin with acetone yields the product with a yield of 96%. In recent times, a method has been developed to measure enzyme activity and to determine the key parameters of bioreduction. Excellent enantioselectivity was found [56].

Figure 4.17 Substrate adsorption: synthesis of ((S)-3,4-methylenedioxyphenyl)-2-propanol from 3,4methylenedioxyphenylacetone applying a batch process followed by a filtration step and resin extraction

80


Figure 4.18 Enzyme membrane reactor: synthesis of L-tert-leucine from trimethylpyruvic acid in an continuously operated enzyme membrane reactor with ultrafiltration followed by a crystallization step

Amino acids like L-tert-leucine, L-neopentylglycine or L-5,5-dimethyl-butyl-glycine are building blocks for drug synthesis, especially for the syntheses of antitumor or HIV protease inhibitors, and are used as templates in asymmetric synthesis [36]. They are synthesized by Evonik Industries starting from ammonia and keto acid. Here, high concentrations of the substrate trimethylpyruvate lead to inhibition of the enzyme L-leucine-dehydrogenase within the production of L-tert-leucine. Additionally, the productivity of the process is limited by chemical side reactions, because the formation of by-products is increased at higher concentrations of ammonia and keto acid. Therefore, a continuously operated process with an ultrafiltration step is applied and the product is crystallized afterwards (Figure 4.18). This enzyme membrane reactor enables low concentrations of the substrates and recycling of the enzyme [57]. To produce 2-phenylphenol and 3-phenylcatechol, which are starting materials for the synthesis of pharmaceuticals like barabtusol, taxodione or L-DOPA analogues, suspended whole cells from E. coli JM101 are used. Here, the substrates are highly toxic to the cells and have to be fed continuously into the reactor, thereby keeping the concentration below the toxic level [58]. Because catechols are readily soluble in the aqueous phase and dissolved 3-phenylcatechol polymerizes, the product is removed by continuous adsorption on the resin Amberlite XAD-4 (Figure 4.19). This process is carried out by Sigma Aldrich, where the

Figure 4.19 Continuous feed of substrate and removal of products: synthesis of 3-phenylcatechol from 2-phenylphenol applying a fed batch process with fluidized bed adsorption followed recrystallization


81

reaction mixture is continuously pumped through an external loop with a fluidized bed of the resin. The whole cells are not retained because the growth rate of the E. coli cells is unaffected in the presence of XAD-4. The product is eluted from the resin by acidic methanol followed by crystallization in n-hexane. Aliphatic epoxides are used to synthesize various chiral intermediates; in particular, there are the 2-methyl-1,2-epoxyalkanes, which are precursors for tertiary alcohols used in the synthesis of pharmaceuticals such as prostaglandins. To produce such epoxides, suspended whole cells from Nocardia coralline B276 are used whereas the alkene monooxygenase catalyzes the stereospecific epoxidation reaction of terminal und sub-terminal alkenes [59]. The biotransformation is carried out by Nippon Mining Holdings, Inc. in a classical fermentation setup (Figure 4.20). Problematic here are short-chain epoxides with a chain length of three to five carbon atoms, because they are gaseous and very toxic for the cell. This makes the product recovery complicated. In the case of such epoxides, the rate of aeration has to be raised to extract the short-chain toxic epoxides from the fermentation medium. The very low amount of epoxides in the gas phase is recovered by a special solvent extraction system.

Figure 4.20 Continuous removal of product by stripping: synthesis of epoxides from alkenes applying a batch process with resting cells or a fermenter connected to absorption, extraction and distillation

82


4.3.6 Low Solubility Ibuprofen is an important nonsteroidal anti-inflamatory drug where the in vivo activity of the (S)-enantiomer is 100 times higher than that of the (R)-enantiomer. Therefore, an enantioselective synthesis is preferable. A low pH has to be employed to minimize deactivation of the enzyme by the substrate ibuprofen methoxyethyl ester. The low pH leads to a low solubility of the substrate in water, which is the central drawback of the enzymatic synthesis. To circumvent problems of handling big volumes of water, a multiphase membrane reactor concept was realized [60]. A hollow-fiber membrane in which the lipase is immobilized is used by Pfizer Inc. for the kinetic resolution of ibuprofen [61]. The substrate ibuprofen methoxyethyl ester is hydrophobic and is dissolved in the organic phase, whereas ibuprofen is extracted to the aqueous phase within the hollow-fiber module (Figure 4.21). The stabilizing effect of the membrane to the aqueous–organic interface leads to a high surface area without dispersion of both phases. This hollow-fiber module is combined with another membrane module which is adjusted to a higher pH. In this way the product is easily separated from the unconverted substrate. This process solution allows a low ibuprofen concentration and reaches an enantiomeric excess of 96%.

Figure 4.21 Multiphase membrane reactor: synthesis of ibuprofen from ibuprofen methoxyethyl ester applying a multiphase membrane reactor in batch mode followed by extraction and distillation for downstream processing

In the food, cosmetic and pharmaceutical sector, about 40 000 tons per year of malic acid are used worldwide. The enantioselective synthesis of L-malate from fumarate is carried out in a slurry reactor [62]. Here, the reaction is conducted by Amino GmbH in a slurry of crystalline calcium fumarate and crystalline calcium malate (Figure 4.22). The precipitation of the product shifts the equilibrium towards calcium malate. The separation of the biocatalyst, which is suspended whole cells of Corynebacterium glutamicum, is performed by filtration of the slurry. The downstream processing is done by acidification with sulfuric acid, yielding L-malic acid and gypsum. After filtration, L-malic acid is purified by ion-exchange chromatography with a chemical purity of > 99%.


83

Figure 4.22 Slurry reactor: synthesis of malic acid from fumaric acid applying a batch process followed by precipitation and crystallization

4.4 Conclusions In the 20th century, production processes were developed to synthesize many pharmaceutical products from oil using chemical catalysts, high temperatures and high pressure. Some of these chemical production processes may be and are being replaced by biotechnological processes [63]. For a wide range of applications, biotransformations have potentially large benefits, like high yield and productivity, high selectivity and environmental safety of biocatalysts, minimal amounts of by-products and easy downstream processing, as well as an ambient reaction temperature [64]. These benefits result in less reaction, less separation and purification steps, and less energy consumption, as well as less emissions, finally enabling a reduction of production costs. By the application of ‘white biotechnology’ using living cells from microorganisms, plants or animals, as well as isolated enzymes, new and ‘greener’ pharmaceutical products can be obtained [65]. Today, molecular biology allows precise modifications of the genetic information and improvements in enzyme properties [66]. This allows shorter timelines for the development of processes and potentiates the production of new pharmaceutical products. Further, through the development of new biocatalysts, the number of bioprocesses will increase. Additionally, the combination of fermentation steps or biotransformation with chemical downstream processing steps leads to a more successful production of pharmaceuticals. Nevertheless, there are several hurdles to overcome. Biotransformation has to compete with a chemical production process. The production of bio-based bulk chemicals and intermediates from white biotechnology must be economically viabile [67]. This means that the biotechnological product must be cheaper to produce or of higher quality than products based on classical chemical routes. And, incidentally, switching to a novel process takes time and money. Therefore, process solutions are required which are adapted to the special needs of biocatalyzed reactions. Within this chapter, several specific tasks have been addressed, illustrating universal possibilities to overcome the limitations found in bioprocesses. Research and development in industry and in academia have enhanced and will enhance bioprocesses working at the interface of chemical process engineering and biotechnology. For

84 Table 4.1


Summary of introduced process solutions

Problem

Possible solution

Expensive stoichiometric cofactor need

Cofactor regeneration:

Limited yield by racemic resolutions

substrate coupled enzyme coupled

stripping enzyme membrane reactor

dynamic kinetic resolution

separation by extraction followed by chemical racemization separation by crystallization followed by chemical racemization in situ enzymatic or chemical racemization

reracemization of one reaction compound Thermodynamic limitation

coupling with irreversible step removal of side product removal of product coupling with chemical step

By-product formation

Process type

coupling with enzymatic step

formation of gaseous side-product azeotropic distillation pervaporation stripping filtration of precipitate

thermodynamically

catalase to remove hydrogen peroxide urea-adduct method selective adsorption pressurized fixed-bed reactor low-temperature reaction

Substrate inhibition

physical binding of substrate repetitive feed of substrate continuous feed of substrate and removal of products continuous removal of product

adsorption of substrate on resin enzyme membrane reactor fluidized bed for adsorption of product stripping

Low solubility

two-phase system removal of product and feed of substrate

multiphase membrane reactor slurry reactor

coupling with chemical step coupling with physical step

specific problems, different possibilities to overcome them have been verified; these are summarized in Table 4.1. In conclusion, optimization of a process can only be reached if molecular and reaction engineering complement each other. Furthermore, reaction engineering must be focused on downstream processing in addition to reaction progress. Starting at the reaction itself, a selective synthesis circumvents complicated downstream processing. Within this chapter, this was illustrated in the case of resolution of racemic mixtures, separation of by-product and shifting of chemical equilibrium. Figure 4.23 shows the points of application regarding reaction


Figure 4.23

85

Industrial biotransformations: points of application for reaction engineering

engineering in industrial biotransformations. The separation and purification steps have been addressed both within in classical downstream processing and integrated in reaction engineering. Adapting the reaction parameters in respect to easy downstream processing enables a reduction of production costs, which is valid for reaction technology and molecular engineering.

References [1] Liese, A., Seelbach, K. and Wandrey, C. (2006) Industrial Biotransformations, 2nd edn, VCH–Wiley, Weinheim. [2] Hilterhaus, L. and Liese, A. (2007) Building blocks. Advances in Biochemical Engineering/Biotechnology, 105, 133–173. [3] Pollard, D.J. and Woodley, J.M. (2007) Biocatalysis for pharmaceutical intermediates: the future is now. Trends in Biotechnology, 25 (2), 66–73. [4] Leib, T.M., Pereira, C.J. and Villadsen, J. (2001) Bioreactors: a chemical engineering perspective. Chemical Engineering Science, 56 (19), 5485–5497. [5] Fischer, R., Stoger, E., Schillberg, S. et al. (2004) Plant-based production of biopharmaceuticals. Current Opinion in Plant Biology, 7 (2), 152–158. [6] Deckor, E.L. and Reski, R. (2007) Moss bioreactors producing improved biopharmaceuticals. Current Opinion in Biotechnology, 18 (5), 393–398. [7] Bourgaud, F., Gravot, A., Milesi, S. and Gontier, E. (2001) Production of plant secondary metabolites: a historical perspective. Plant Science (Shannon, Ireland), 161 (5), 839–851. [8] Gogate, P.R., Beenackers, A.A.C.M. and Prandit, A.B. (2000) Multiple impeller systems with a special emphasis on bioreactors: a critical review. Biochemical Engineering Journal, 6, 109–144. [9] Hermann, T. (2003) Industrial production of amino acids by coryneform bacteria. Journal of Biotechnology, 104, 155–172. [10] Sauer, M., Porro, D., Mattanovich, D. and Branduardi, P. (2008) Microbial production of organic acids: expanding the markets. Trends in Biotechnology, 26 (2), 100–108. [11] Rasor, J.P. and Voss, E. (2001) Enzyme-catalyzed processes in pharmaceutical industry. Applied Catalysis A–General, 221 (1–2), 145–158. [12] Goldberg, K., Schroer, K., Lutz, S. and Liese, A. (2007) Biocatalytic ketone reduction – a powerful tool for the production of chiral alcohols – part I: processes with isolated enzymes. Applied Microbiology and Biotechnology, 76 (2), 237–248.

86


[13] Goldberg, K., Schroer, K., Lutz, S. and Liese, A. (2007) Biocatalytic ketone reduction – a powerful tool for the production of chiral alcohols – part II: whole-cell reductions. Applied Microbiology and Biotechnology, 76 (2), 249–255. [14] Kragl, U., Kruse, W., Hummel, W. and Wandrey, C. (1996) Enzyme engineering aspects of biocatalysis: cofactor regeneration as example. Biotechnology and Bioengineering, 52, 309–319. [15] Kragl, U. and Dwars, T. (2001) The development of new methods for the recycling of chiral catalysts. Trends in Biotechnology, 19 (11), 442–449. [16] De Temino, D.M., Hartmeier, W. and Ansorge-Schumacher, M.B. (2005) Entrapment of the alcohol dehydrogenase from Lactobacillus kefir in polyvinyl alcohol for the synthesis of chiral hydrophobic alcohols in organic solvents. Enzyme and Microbial Technology, 36 (1), 3–9. [17] Hoyos, P., Buthe, A., Ansorge-Schumacher, M.B. et al. (2008) Highly efficient one pot dynamic kinetic resolution of benzoins with entrapped Pseudomonas stutzeri lipase. Journal of Molecular Catalysis B, Enzymatic, 52–53, 133–139. [18] Tao, J.H. and McGee, K. (2004) Development of a continuous enzymatic process for the preparation of (R)-3-(4fluorophenyl)-2-hyroxy propionic acid. Organic Process Research & Development, 6, 520–524. [19] Giorno, L. and Drioli, E. (2000) Biocatalytic membrane reactors: applications and perspectives. Trends in Biotechnology, 18, 339–349. [20] Judd, S. (2008) The status of membrane bioreactor technology. Trends in Biotechnology, 26 (2), 109–116. [21] Hummel, W. (1997) New alcohol dehydrogenases for the synthesis of chiral compounds. Advances in Biochemical Engineering/Biotechnology, 58, 145–184. [22] Schroer, K., Mackfeld, U., Tana, I.A.W. et al. (2007) Continuous asymmetric ketone reduction processes with recombinant Escherichia coli. Journal of Biotechnology, 132 (4), 438–444. ¨ berwindung von thermodynamischen Limit[23] Stillger, T., B€ onitz, M., Ville la Filho, M. and Liese, A. (2002) U ierungen in substratgekoppelten Cofaktorregenerierungsverfahren. Chemie Ingenieur Technik, 74 (7), 1035–1039. [24] Stillger, T., Villela, V., Liese, A. and Wandrey, C. (2005) Verfahren und Vorrichtung zur enzymatischen Co-Faktoren-Regenerierung. DE10060602. [25] Goldberg, K., Edegger, K., Kroutil, W. and Liese, A. (2006) Overcoming the thermodynamic limitation in asymmetric hydrogen transfer reactions catalyzed by whole cells. Biotechnology and Bioengineering, 95, 192–198. [26] Schmidt, E., Ghisalba, O., Gygax, D. and Sedelmeier, G. (1992) Optimization of a process for the enzymatic production of (R)-2-hyroxy-4-phenylbutyric acid – an intermediate for inhibitors of angiotensin converting enzymes. Journal of Biotechnology, 24, 315–327. [27] Rios, G.M., Belleville, M.P. and Paolucci-Jeanjean, D. (2007) Membrane engineering in biotechnology: quo vamus? Trends in Biotechnology, 25 (6), 242–246. [28] Schnell, B., Faber, K. and Kroutil W. (2003) Enzymatic racemisation and its application to synthetic biotransformations. Advanced Synthesis and Catalysis, 345 (6–7), 653–666. [29] Pellissier, H. (2008) Recent developments in dynamic kinetic resolution. Tetrahedron, 64 (8), 1563–1601. [30] Goswami, A., Howell, J.M., Hua, E.Y. et al. (2001) Chemical and enzymatic resolution of (R,S)-N-(tertbutoxycarbonyl)-3-hydroxymethylpiperidine. Organic Process Research & Development, 5, 415–420. [31] Huang, D.S., Huang, K.L., Chen, S.P. et al. (2008) Rapid reaction-diffusion model for the enantioseparation of phenylalanine across hollow fiber supported liquid membrane. Separation Science and Technology, 43 (2), 259–272. [32] Ricks, E.E., Estrada-Vades, M.C., McLean, T.L. and Iacobucci, G.A. (1992) Highly enantioselective hydrolysis of (R,S)-phenylalanine isopropyl ester by subtilisin Carlsberg. Continuous synthesis of (S)-phenylalanine in a hollow fibre/liquid membrane reactor. Biotechnology Progress, 8, 197–203. [33] Shimizu, H., Ogawa, J., Kataoka, M. and Kobayashi, M. (1997) Screening of novel microbial enzymes for the production of biologically and chemically useful compounds, in New Enzymes for Organic Synthesis (ed. T. Scheper), Advances in Biochemical Engineering Biotechnology, vol. 58, Springer, New York, pp. 45–99. [34] Shimizu, S., Kataoka, M., Honda, K. and Sakamoto, K. (2001) Lactone-ring-cleaving enzymes of microorganisms: their diversity and applications. Journal of Biotechnology, 92 (2), 187–194. [35] Bommarius, A.S., Drauz, K., Klenk, H. and Wandrey, C. (1992) Operational stability of enzymes – acylase catalyzed resolution of N-acetyl amino acids to enantiomerically pure L-amino acids. Annals of the New York Academy of Sciences, 672, 126–136. [36] Bommarius, A.S., Scharm, M. and Drauz, K. (1998) Biocatalysis to amino acid-based chiral pharmaceuticals – examples and perspectives. Journal of Molecular Catalysis B, Enzymatic, 5, 1–11. [37] Hirohara, H. and Nishizawa, M. (1998) Biochemical synthesis of several chemical insecticide intermediates and mechanism of action of relevant enzymes. Bioscience, Biotechnology, and Biochemistry, 62, 1–9.


87

[38] Cheetham, P.S.J. (1994) Case studies in applied biocatalysis, in Applied Biocatalysis (eds J.M.S. Carbral, D. Best, L. Borros and J. Tramper), Harwood Academic Publishers, Chur, Switzerland, pp. 68–70. [39] Ager, D.J., Fotheringham, I.G., Laneman, S.A. et al. (1997) The large-scale synthesis of unnatural acids. Chimica Oggi, 15 (3/4), 11–14. [40] Ager, D.J., Li, T., Pantaleone, D.P. et al. (2001) Novel biosynthetic routes to non-proteinogenic amino acids as chiral pharmaceutical intermediates. Journal of Molecular Catalysis B, Enzymatic, 11 (4–6), 199–205. ¨ le-Fette-Wachse, 116 (6), 201–205. [41] McCrae, A.R., Roehl, E.-L. and Brand, H.M. (1990) Bio-ester – bio-esters. Seifen-O [42] Yan, Y.C., Bornscheuer, U.T. and Schmid, R.D. (2002) Efficient water removal in lipase-catalyzed esterifications using a low-boiling-point azeotrope. Biotechnology and Bioengineering, 78 (1), 31–34. [43] Del Amor Villa, E.M. and Wichmann, R. (2005) Membranes in the enzymatic synthesis of biotensides from renewable sources. Catalysis Today, 104 (2–4), 318–322. [44] Hilterhaus, L., Minow, B., M€uller, J. et al. (2008) Practical application of different enzymes immobilized on Sepabeads. Bioprocess and Biosystems Engineering, 31, 163–171. [45] Hilterhaus, L., Thum, O. and Liese, A. (2008) Reactor concept for lipase-catalyzed solvent-free conversion of highly viscous reactants forming two-phase systems. Organic Process Research & Development, 12, 618–625. [46] Thum, O., Hilterhaus, L. and Liese, A. (2008). Verfahren zur enzymatischen Herstellung von Carbons€aureestern. DE 10 2008 004726.0. [47] Thum, O., Hilterhaus, L. and Liese, A. (2008). Verfahren zur heterogenkatalysierten Herstellung von Carbons€aurederivaten. DE 10 2008 004725.2. [48] Oyama, K. (1992) The industrial production of aspartame, in Chirality in Industry (eds A.N. Collins, G.N. Sheldrake and J. Crosby), John Wiley & Sons, Ltd, New York, pp. 237–247. [49] Onken, U., Schmidt, E. and Weissenrieder, T. (1996) Enzymatic H2O2 decomposition in a three phase suspension, Ciba Geigy. International Conference on Biotechnology for Industrial Production of fine Chemicals, 93rd event of the EFB, Zermatt, Switzerland, 29.09.1996. [50] Koike, K., Takaiwa, M., Ara, K. et al. (2000) Production of isopropyl cis-6-hexadecenoate by regiospecific desaturation of isopropyl palmitate by a double mutant of a Rhodococcus strain. Bioscience, Biotechnology, and Biochemistry, 64 (2), 399–404. [51] Takeuchi, K., Koike, K. and Ito, S. (1990) Production of cis-unsaturated hydrocarbons by a strain of Rhodococcus in repeated batch culture with a phase-inversion, hollow fiber system. Journal of Biotechnology, 14, 179–186. [52] Tsuchiyama, Y., Yamamoto, K.-I., Asou, T. et al. (1991) A novel process of cyclodextrin production by use of specific adsorbents. Part I. Screening of specific adsorbents. Journal of Fermentation and Bioengineering, 71, 407–412. [53] Furui, M. and Yamashita, K. (1983) Pressurized reaction method for continuous production of L-alanine by immobilized Pseudomonas dacunhae cells. Journal of Fermentation Technology, 61, 587–591. [54] Nagasawa, T., Shimizu, H. and Yamada, H. (1993) The superiority of the third-generation catalyst, Rhodococcus rhodochrous J1 nitrile hydratase, for industrial production of acrylamide. Applied Mircobiology and Biotechnology, 40, 189–195. [55] Anderson, B.A., Hansen, M.M., Harkness, A.R. et al. (1995) Application of a practical biocatalytic reduction to an enantioselective synthesis of the 5H-2,3-benzodiazepine LY300164. Journal of the American Chemical Society, 117, 12358–12359. [56] Erdelyi, B., Szabo, A., Birincsik, L. and Hoschke A. (2004) Process development of methylenedioxyphenylacetone chiral bioreduction. Journal of Molecular Catalysis B, Enzymatic, 29 (1–6), 195–199. [57] Wichmann, R., Wandrey, C., B€uckmann, A.F. and Kula, M.-R. (1981) Continuous enzymatic transformation in an enzyme membrane reactor with simultaneous NAD(H) regeneration. Biotechnology and Bioengineering, 23, 2789–2796. [58] Held, M., Schmid, A., Kohler, H.-P. et al. (1999) An integrated process for the production of toxic catechols from toxic phenols based on a designer biocatalyst. Biotechnology and Bioengineering, 62 (6), 641–648. [59] Furuhashi, K. (1992) Biological routes to optically active epoxides, in Chirality in Industry (eds A. Collins, G.N. Sheldrake and J. Crosby), John Wiley & Sons, Ltd, New York, pp. 167–186. [60] Lopez, J.L., Wald, S.A., Matson, S.L. and Quinn, J.A. (1990) Multiphase membrane reactors for separating stereoisomers. Annals of the New York Academy of Sciences, 613, 155–166. [61] Long, W.S., Kamaruddin, A.H. and Bhatia, S. (2005) Enzyme kinetics of kinetic resolution of racemic ibuprofen ester using enzymatic membrane reactor. Chemical Engineering Science, 60 (18), 4957–4970. [62] Danneel, H.J., Busse, M. and Faurie, R. (1995) Pharmaceutical grade L-malic acid from fumaric acid – development of an integrated biotransformation and product purification process. Mededlingen van de Faculteit Landbouwwetenschappen, Rijksuniversiteit te Gent, 60 (4a), 2093–2096.

88


[63] Constable, D.J.C., Dunn, P.J., Hayler, J.D. et al. (2007) Key green chemistry research areas – a perspective from pharmaceutical manufacturers. Green Chemistry, 9 (5), 411–420. [64] Hatti-Kaul, R., Tornvall, U., Gustafsson, L. and Borjesson, P. (2007) Industrial biotechnology for the production of bio-based chemicals – a cradle-to-grave perspective. Trends in Biotechnology, 25 (3), 119–124. [65] Woodley, J.M. (2008) New opportunities for biocatalysis: making pharmaceutical processes greener. Trends in Biotechnology, 26 (6), 321–327. [66] Schmidt, M., Baumann, M., Henke, E. et al. (2004) Directed evolution of lipases and esterases. Methods in Enzymology, 388, 199–207. [67] Hermann, B.G. and Patel, M. (2007) Todays and tomorrows bio-based bulk chemicals from white biotechnology. Applied Biochemistry and Biotechnology, 136, 361–388.

5 Chiral Synthesis of Pharmaceutical Intermediates Using Oxynitrilases Wen-Ya Lu and Guo-Qiang Lin Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, Shanghai, 200032, PR China

5.1 Introduction Hydroxynitrile lyases (HNLs, or oxynitrilases, E.C.4.2.1.x) are enzymes that catalyze enantioselective cleavage and synthesis of cyanohydrins. As early as 1830, Robiquet and Boutron-Chalard isolated amygdalin and discovered its hydrolytic splitting by extraction of defatted bitter almonds. The agent was named ‘emulsin’ (b-glucosidase and HNL) by Liebig and W€ ohler in 1837. Rosenthaler first carried out condensation of HCN with carbonyl compounds to produce chiral cyanohydrin in 1909 using crude extract of bitter almond [1]. In the 1960s, Pfeil and coworkers purified the HNLs from almond and carried out the reaction in an aqueous/alcoholic buffer system [2]. It is not until the 1980s, with the introduction of waterimmiscible organic solvents, suppression of the chemical reaction occurred to a significant extent; the enantiomeric purity of the cyanohydrins was thereby considerably increased [3]. Today, some HNLs, such as HbHNL from the tropical rubber tree Hevea brasiliensis and PaHNL from the almond tree Prunus amygdalus, are already used for large-scale industrial applications.

5.2 HNL 5.2.1 The Natural Function and Distribution of HNLs Cyanogensis, the ability of organisms to liberate HCN under certain conditions, has been reported in more than the 3000 species of vascular plant taxa comprising 105 families of flowering plants, pteridophytes (ferns), and gymnosperms. Prominent among these are the plants belonging to families like Linaceae, Euphorbiaceae, Clusiaceae, Olacaceae, Rosaceae, Biocatalysis for the Pharmaceutical Industry edited by Junhua Tao, Guo-Qiang Lin, and Andreas Liese 2009 John Wiley & Sons Asia (Pte) Ltd

90


Gramineae (monocotyledons), and Filitaceae (ferns) [4,5]. In addition to plants, cyanogenesis is also reported in taxonomically diverse groups of organisms like bacteria (Chromobacterium violaceum, a few species of Pseudomonas) [6–8], fungi, lichen, and arthropods (e.g. millipedes such as Apheloria corrugata and moths such as Zygaena trifollii). Endogenous cyanidecontaining compounds occur mostly as cyanogenic glycosides (sometimes as cyanolipids) in organisms. Most cyanogenic glycosides are biogenetically derived from the amino acids phenylalanine, tyrosine, valine, isoleucine, or leucine; but the non-protein amino acid cyclopentenylglycine and probably, nicotinic acid also serve as precursors (Figure 5.1) [9]. OR

OR O

HO HO

O

OH

OH

CN

Linamarin Linustatin

OH

O

OH

R=H R=H R = glucose

HO HO

CN

3-Hydroxy-heterodendrin (2S)

Heterodendrin (2S) Epiheterodendrin (2R) leucine

CN

HO HO

OH

CN

leucine

O

CN OH

Volkenin (1R, 4R) (=Epiteraphyllin B) Tetraphyllin B (1S, 4S) (Barerin) Epivolkenin (1S, 4R) (Passicoriacin) Taraktophyllin (1R, 4S) (Epipassicoriacin)

cyclopentenylglycine

O

OH

OH

O OH

Deidaclin (2R) Tetraphyllin A (1S)

Figure 5.1

O

Proacacipetalin (2S) Epiproacacipetalin (2R)

OH O

Lotaustralin (2R) Epilotaustralin (2S) Neolinustatin (2R) isoleucine

leucine

OH

CN

OH O

CN

O

OH

valine

HO HO

O

O

O

HO HO

OH

OH

HO HO

O

R=H R = glucose

OH O

Zierin (2S ) Holocalin (2R) Zierin xyloside (2S) phenylalanine

OH

CN

R=H Dhurrin (2S) R=H Taxiphyllin (2R) R = glucose Proteacin (2S) tyrosine

HO HO

R=H R=H R = xylose

OH CN

OR O

HO HO

O

OH

Sambunigrin (2S) Neoamygdalin (2S) Epilucumin (2S) phenylalanine

O

OH

CN

OR

OR

O

HO HO

R=H R = glucose R = xylose

O

HO HO

O

OH

CN

R=H Prunasin (2R) R = glucose Amygdalin (2R) R = xylose Lucumin (2R) phenylalanine

OR O

HO HO

cyclopentenylglycine

HO HO

NC O O OH HO

OMe

N

O

Acalyphin

nicotinic acid

Cyanogenic glycosides derived from amino acids (and nicotinic acid)

91

Chiral Synthesis of Pharmaceutical Intermediates Using Oxynitrilases

5.2.2 Classification of HNLs About 200 species of plants have been examined for their HNLs activity and about 18 sources have been purified and characterized. Previously, the HNLs were divided into two groups, namely HNL I and HNL II, based on the presence or absence of flavin adenine dinucleotide (FAD) [10]. However, comparison of the amino acid sequence of various recently cloned HNLs revealed sequence homologies to other proteins; HNLs are of widely different phylogenetic origin, like subtilisin, cysteine proteases, eucaryotic serine proteases, and a/b hydrolases, and share the common structural motif of eight b sheets connected by b helices beside similar protein folds [11,12]. Therefore, HNLs were believed to be the result of both of convergent evolution and divergent evolution (Figure 5.2). ?

? XaHNL

PhaHNL SbHNL, SvHNL

HNLs from Prunoideae and Maloideae

HbHNL MeHNL

PaHNL

LuHNL

PsHNL

convergent evolution

FAD-dependent (…) Zn2+ containing oxidoreductase ADH

carboxypeptiedase (…) acetylcholine esterase

divergent evolution

α/β hydrolase super family

Figure 5.2

βαβ motif (ADP-binding domain)

Phylogeny of HNLs of higher plants

Generally, HNLs utilize an acid–base catalysis mechanism. The amino acid residues at active sites of these enzymes differ significantly, but share the common motif for cyanogenesis. 5.2.2.1 bab Motif Super Family .

HNLs that are evolved from FAD-dependent oxidoreductases: FAD-containing enzymes have been isolated exclusively from species of the Rosaceae subfamilies Prunoideae and Maloideae (e.g. Prunus serotina, Prunus lyonii, Prunus laurocerasus, Prunus capuli, Prunus amygdalus, Prunus mume, Mammea americana, Malus communis, Eriobotrya japonica). The amino acid sequence of FAD-containing HNLs showed that they are related to members of the glucose–methanol–choline oxidoreductase family (30% sequence identity) [13]. HNLs are proteins of size ranging from 250 to 664 amino acid residues, which share a

92

.


number of highly conserved regions ( > 89%). One of these regions, located in the N-terminal section, corresponds to the FAD/adenosine diphosphate (ADP)-binding domain, while another one is located in the center of the protein as a signature pattern [14]. In 2001, Kratky and coworkers [15] determined the crystal structure of a 61 kDa HNL isoenzyme from Prunus amygdalus (PaHNL1) to 1.5 A resolution. Analysis of the modeled substrate complexes supports an enzymatic mechanism that includes the flavin cofactor as a mere ‘spectator’ of the reaction and relies on general acid–base catalysis by the conserved His497. Stabilization of the negative charge of the cyanide ion is accomplished by a pronounced positive electrostatic potential at the binding site. PaHNL activity requires the FAD cofactor to be bound in its oxidized form, and calculations of the pKa of enzyme-bound HCN showed that the observed inactivation upon cofactor reduction is largely caused by the reversal of the electrostatic potential within the active site (Figure 5.3). Although the actual amino acid residues involved in the catalytic cycle are completely different in the two enzymes, a common motif for the mechanism of cyanogenesis (general acid–base catalysis plus electrostatic stabilization of the cyanide ion) becomes evident [16]. Zn2þ -containing alcohol dehydrogenase: Linum usitatassimum HNL (LuHNL) has an ADP-binding bab domain and catalytic domain containing two Zn2þ , which are not directly involved in catalysis [17].

Arg300

Arg300 Tyr

Tyr457

457

O

H H

C

N Lys361

O

H

H

S Cys326

N HN

H O

PaHNL

N

H

O

H

Figure 5.3

Lys361

H

S Cys326

N HN

His497

C

His497

Reaction mechanism of PaHNL

5.2.2.2 a/b Hydrolase Super Family .

.

HbHNL was the first enzyme whose three-dimensional crystal structure was elucidated, and this was followed by other HNLs [18,19]. HbHNL [20] and MeHNL [21] belong to the a/b hydrolase super family with a deeply buried active site within the protein and linked to the outer environment by a narrow channel flanked by apolar residues (Figure 5.4). Eukaryotic serine proteases: during catalysis in SbHNL, hydrogen bonds are formed between the substrate hydroxyl group, serine-158 and an oxygen atom of the tryptophan-270 carboxyl group. A water molecule is suspended between other carboxylate groups of tryptophan-270 and the nitrile group of the substrate. Here, the carboxylate moiety of tryptophan acts as catalytic base, causing abstraction of proton from the cyanohydrin hydroxyl group, while water at the active site transfers this proton to cyanide, resulting in carbon–carbon bond cleavage (Figure 5.5) [22].

93


Thr11 O H A

H3N

C N O H

N

O H Ser80

Thr11 O H B

Lys236 Asp

O

N H

Thr11 207

His235

N

H3N

H N

Ser80

Lys236

N H

Asp207

O O

His235

Thr11

C N

N

O H Ser80

Asp208

O

N H

O

O H

MeHNL

C

H O

O

His236

N

H N

Ser80

N H

Asp208

O O

His236

A: Reaction mechanism of HbHNL; B: Reaction mechanism of MeHNL

Trp270

Trp270 O

O

158

H O

Ser

O

H O C

O H N

H

SbHNL 158

Ser

O

O H O

Figure 5.5

H O

H

C

N

H

H

OH

OH

.

C

H O

O

O H

Figure 5.4

O

O H

HbHNL

Reaction mechanism of SbHNL

To date, noncloned HNLs from Ximenia americana (XaHNL) [23] and Phlebodium aureum (PhaHNL) [24] cannot be defined as phylogenetical groups of existing HNLs.

5.2.3 New HNLs and High-Throughput Screening Novel or improved HNLs can be found by screening plants for appropriate enzymes, by directed evolution, rational design, or by metagenomic approaches [25]. However, one major necessity for all these strategies is a simple and powerful high-throughput screening assay to identify potential novel or better performing HNLs. Because of the instability of cyanohydrins, the characterization of cyanohydrins mostly should be hydroxyl protected. In 2001, Gerrits et al. [26] investigated the influence of solvent composition on the stability of unprotected cyanohydrins and then described a method to analyze unprotected cyanohydrins (with regard to enantiomeric purity and conversion) via chiral high-performance liquid chromatography (HPLC). Hernandez et al. [27] and the groups

94


of Asano and of Hickel [28] reported several new HNLs for biocatalytic processes by screening about 200 plant extracts. Additionally, it is worth noting that an HNL from the same plant but from different tissue may perform differently [29]. The HNLs of some plants have been cloned in microbial systems, since the latter can be easily transformed and cultured in a short time for large-scale production. MeHNL was overexpressed in Escherichia coli and 40 L scale fermentation gave 40 000 IU of HNL for (S)-cyanohydrin synthesis [30,31]. The HbHNL gene has been overexpressed in Escherichia coli using a strong inducible tac promoter and the gene product formed inclusion bodies, which were insoluble and inactive [32]. The more successful heterologous expression of HbHNL was achieved in Saccharomyces cerevisiae and a methanol-inducible Pichia pastoris expression system in soluble and highly active form comprising about 60% of total protein of the cell [33]. A recombinant Pichia pastoris containing HbHNL gene was cultured using a high celldensity culture technique in a 30 L laboratory fermenter, which yielded 23 g L1HbHNL and was purified by one-step ion exchange chromatography [34]. Roche Diagnostics, Penzberg, Germany, has produced HNLs on up to a 50 000 L fermenter scale with approximately similar yields on a laboratory scale and used for the synthesis of (S)-cyanohydrin of 3-phenoxybenzaldehyde, which is an important intermediate for synthetic pyrethroid. SbHNL recombinant expression in a prokaryotic system failed, since it required complex posttranslational modification of the native enzyme. Effenberger et al. cloned a myc-His-tagged LuHNL-cDNA under control of the methanol-inducible AOX1 (alcohol oxidase) promotor of Pichia pastoris and introduced it in the SMD1168 strain; recombinant LuHNL was kinetically indistinguishable from the authentic flax enzyme [17]. With the availability of the biochemical and structural information of PaHNL, Glieder and coworkers cloned the PaHNL5 gene from Prunus amygdalus. As reported, recombinant isoenzyme PaHNL5 was very stable under acidic conditions (pH 2.6). Rational engineering of the active site led to the adaptation of PaHNL5 for the difficult stereoselective conversion of unnatural, industrially interesting substrates such as 2-chlorobenzaldehyde [35,36] and 3-phenylpropanal [37] (Figure 5.6) into corresponding (R)-cyanohydrins with high enantioselectivity. MeO

O

Cl

N S

HOOC

NH2 N H

O N

EtOOC N H

COOH

O N

COOH

H2SO4 clopidogrel

Figure 5.6

lisinoril

enalapi

Cyanohydrins are the important intermediates of some drugs

So far, HNL activity is usually determined by gas chromatography (GC) or HPLC analysis, which is not practical for high throughput (Table 5.1). Another spectrophotometric assay for activity towards benzaldehyde cyanohydrin is available, which detects the increase in absorption of the released benzaldehyde at 280 nm wavelength [38]. Although this assay is applicable for high throughput, it is restricted to aromatic substrates only and requires microtiter plates (MTPs) which are transparent in the UV. Furthermore, recently, a colony assay, based on NADH-fluorescence, has been developed to detect HNL activity towards benzaldehyde cyanohydrin [39]. Eggert and coworkers described a high-throughput screening

95

Chiral Synthesis of Pharmaceutical Intermediates Using Oxynitrilases Table 5.1 Assays useful for HNL screening Detection method

Description

Ref.

HPLC, GC Spectrophotometric NADH-fluorescence Spectrophotometric HCN sensitive HCN-sensitive Feigl–Anger test paper Microreactor

Low-throughput Aromatic substrates Colony assay Microtiter plates (MTPs), Colony assay Reaction assay

[26,27] [38] [39] [40] [41] [42]

assay in MTP format (200 mL) which allows screening in high throughput using automated pipetting workstations for HNL activity accepting a wide range of HNL substrates. This method is useful either for enzyme fingerprinting or screening of huge variant libraries generated from metagenome or directed evolution approaches [40]. In 2007, Schwab and coworkers developed a novel HCN-based high-throughput screening assay for HNL activity (Feigl–Anger test paper, prepared by impregnating Whatman paper with a chloroform solution containing copper (II) ethyl-acetoacetate and N,N-dimethylaniline). The assay is useful for detecting activity and enantioselectivity of HNLs theoretically towards any cyanohydrin substrate [41]. In 2008, van Hest and coworkers demonstrated that crude cell lysate containing an HNL can be used for the enantioselective synthesis of cyanohydrins from aldehydes in microchannels. Using a microreactor setup, two important parameters were efficiently screened consuming only minute amounts of reagents. Moreover, results from the continuous flow reaction were fully consistent with results obtained from larger batchwise processes in which a stable emulsion was formed [42]. In 2008, based on sequence similarities to the known HNLs, MeHNL and HbHNL [43], Kragl and coworkers identified a novel R-specific HNL from a noncyanogenic plant, Arabidopsis thaliana (mouse-ear cress). This is the first R-selective HNL, but contains an a/b-hydrolase fold [25].

5.3 Reaction of HNLs 5.3.1 Reaction System Enantiomeric or specific synthesis of cyanohydrin is influenced by the reaction medium, cyanide source, water content, buffer pH, enzyme, and temperature during the HNLcatalyzed reaction. To maximize the enantiomeric excess of the cyanohydrin product, care must be taken to minimize the parallel chemical (nonenzymatic) condensation and racemization of products. At low temperatures, the nonenzymatic reaction is reduced to a larger extent than the enzymatic reaction. The mass transfer rate is reduced to a smaller extent. Mass transfer limitation is required for high enantiomeric excess and determines the conversion rate. Therefore, the volumetric productivity decreases at lower temperatures. The equilibrium constant is considerably higher at low temperatures, resulting in a higher extent of conversion or a lower HCN requirement. Both the volumetric productivity and the required enzyme concentration increase by increasing the reaction temperature and aqueous-phase volume while meeting the required conversion and enantiomeric excess [44]. The influence of the reaction medium (solvent and water activity) is much more difficult to rationalize and predict [45].

96


HCN is the most preferred cyanide source in cyanohydrin synthesis. Besides HCN, several different cyanide sources, like potassium cyanide, are being used in biotransformation. Alternative methods for the safe handling of cyanides on a laboratory scale are, for instance, to use cyanide salts in solution. These solutions can be acidified and used as the aqueous layer in two-phase systems or the HCN can be extracted into the organic layer with the desired solvent for reactions in an organic phase. After the reaction, excess cyanide can readily be destroyed with iron(II) sulfate, or iron(III) chloride or bleach. Cyanide can also be presented in the form of organic cyano, such as acetone cyanohydrin [46] or cyanoformates. However, as claimed by Effenberger, some results could not be reproduced [47]. Radke and coworkers designed a novel recycle reactor to determine the interfacial activity of HNL in a diisopropyl ether–water two-phase system. In 1999, Radke and coworkers established that this enzymatic reaction was carried out by the HNL residing at the organic solvent–water interface, not in the aqueous bulk phase. HNL adsorbs at the interface and exhibits extraordinary stability. Adsorption of PaHNL at the interface is fit by the Langmuir equilibrium adsorption model with an adsorption equilibrium constant of 0.032 L mg1. For the mandelonitrile cleavage reaction at ambient temperature, PaHNL follows Michaelis– Menten kinetics at the interface with a Michaelis constant of 14.4 mM and a specific activity close that for the bulk aqueous phase [48]. The degree of enzyme deactivation for HbHNL upon exposure to organic solvents can be correlated with their respective partition coefficients (log P values). However, three unexpected results were obtained: (1) the deactivation exerted by protic solvents (e.g. methanol) is severely underestimated; (2) little deactivation by an organic solvent cannot necessarily be correlated to catalytic activity in this medium, and (3) in contrast to other enzymes, HNLs are exceptionally stable towards deactivation by dimethylformamide [49]. Lin and coworkers disclosed that, at room temperature, nonenzymatic chemical addition was still observed in a water–organic solvent biphasic reaction system, though the volume of aqueous phases was relative small. Lin developed a method of preparing an active enzyme meal that contained essential water to retain its power for catalysis and found a new catalytic reaction system by application of the prepared meal in a nonaqueous monophasic organic medium (Figure 5.7). There was no problem over a wide range of temperature (from 0–30 C) when the reactions were carried out under micro-aqueous conditions [50]. C

... ... ... H ... A ... S ... P ... ...

C

B+E

A

E

... ... . ... . ... H ... . . S . ... . .... .. ... . P .... . .... . ...

(a) system I: biphase system

(b) system II: micro-aqueousphase crude enzyme system A: Aqueous phase; B: Buffer solution; C: Organic solvent E: Enzyme; H: HCN; S: Substrate; P: Product.

Figure 5.7 system

Comparison of a water–organic solvent biphasic system with a micro-aqueous organic phase


97

As reported by Griengl and coworkers, benzaldehyde, decanal, undecanal, and dodecanal were reacted with HCN in a two-phase solvent system aqueous buffer and ionic liquids 1-ethyl3-methylimidazolium tetrafluoroborate, 1-methyl-3-propylimidazolium tetrafluoroborate, and 1-butyl-3-methyl-imidazolium tetrafluoroborate in the presence of the HNLs from Prunus amygdalus and Hevea brasiliensis. When compared with the use of organic solvents as the nonaqueous phase, the reaction rate was significantly increased and the enantioselectivity remained good [51]. Kragl and coworkers investigated using organic-solvent-free systems to overcome the thermodynamic limitations in the synthesis of optically active ketone cyanohydrins. With organic-solvent-free systems under optimized reaction conditions, conversions up to 78% with > 99.0 enantiomeric excess (ee) (S) were obtained. Finally, 5 mL of (S)-acetophenone cyanohydrin with an ee of 98.5% was synthesized using MeHNL [52].

5.3.2 Immobilization of Enzyme In this context, it is important to notice that HNLs only have in common that they all catalyze the cyanogenesis. Structurally, they are belong to different families of proteins and, therefore, are not always comparable. Purified MeHNL was crystallized by the sitting-drop vapor-diffusion method. The 10–20 mm bipyramidal crystals formed were cross-linked with glutaraldehyde and used as biocatalyst for the synthesis of optically active cyanohydrins. The cross-linked crystals were more stable than Celite-immobilized enzymes when incubated in organic solvents, especially in polar solvents. After six consecutive batch reactions in dibutyl ether, the remaining activity of the cross-linked crystals was more than 70 times higher than for the immobilized enzymes. Nevertheless, the specific activity of the cross-linked crystals per milligram protein was reduced compared with the activity of Celite-immobilized enzymes [53]. Hanefeld and coworkers developed a straightforward process for the encapsulation of HbHNL under low methanol conditions. By adding a sol, prepared by hydrolysis of tetramethoxysilane–methyltrimethoxysilane at pH 2.8 with continuous removal of methanol, to a stirred solution of the enzyme in a buffer at pH 6.5, at least 65% of the activity of the free enzyme could be recovered after the encapsulation. The aquagels were successfully used in the synthesis of (S)-cyanohydrins [54]. Vorlop et al. described a novel cross-linked and subsequently poly(vinyl alcohol)entrapped PaHNL for synthesis of (R)-cyanohydrins. These immobilized lens-shaped biocatalysts have a well-defined macroscopic size in the millimeter range, show no catalyst leaching, and can be recycled efficiently. Furthermore, this immobilization method is cheap and the entrapped (R)-oxynitrilases gave similar good results compared with those of free enzymes. The (R)-cyanohydrin was obtained in good yields and with high enantioselectivity of up to > 99% ee [55]. Sheldon and coworkers precipitated (R)-oxynitrilase from almonds with 1,2-dimethoxyethane and subsequent cross-linking using glutaraldehyde. The resulting cross-linked enzyme aggregate (CLEA) was a highly effective hydrocyanation catalyst under microaqueous conditions, which suppress the nonenzymatic background reaction. The beneficial effect of CLEA–micro-aqueous conditions on the hydrocyanation of slow-reacting aldehydes is demonstrated. The oxynitrilase CLEA was recycled 10 times without loss of activity [56].

98


5.3.3 Continuous Reactors Lin and coworkers reported a practical high-throughput continuous process for the synthesis of chiral cyanohydrins. Pretreated almond meal (or other solid raw enzyme source) was loaded in a column to form a reactor, to which was attached a supply system to deliver a solution of substrate and HCN in solvent at one end and a collecting–separating system on the other end. By controlling the flow rate, optimal conditions were achieved for the hydrocyanation of various aromatic carboxaldehydes in a ‘micro-aqueous’ medium to produce chiral cyanohydrins in high yields and high ee with high substrate/catalyst ratio (Figure 5.8) [57]. 5

2 HCN + IPE

1. column filled with defatted crude almond meal 2, 3. sunstrates container (IPE = diisopropyl ether) 4. mixer 5. constant flowing pump 6. volve 7. product container

4 HCN + substrate + IPE

1

substrate + IPE

3

6

7 product container

Figure 5.8

A schematic flow sheet of the continuous process

5.3.4 Henry Reaction Recently, HbHNL has been found to catalyze the stereoselective addition of nitroalkanes to aldehydes in an S-selective fashion, which is in agreement with the known stereopreference of this enzyme. This is the first example for a substitution of HCN by another carbon nucleophile, expanding the synthetic scope of this biocatalytic transformation. The addition of nitromethane to different aldehydes with moderate to good yields and enantioselectivity has been demonstrated (Figure 5.9) [58]. However, large amounts of enzyme are required to OH

OH HbHNL, 48h NO2

NO2

Yield 67% (4 diastereomers) ee 95% (anti/syn = 9/1)

Figure 5.9

O

NO2

HbHNL, 48h CH3NO2 Yield 63% ee 92%

HbHNL-catalyzed enantioselective addition of nitromethane to aldehydes

99


reach acceptable conversions, and it is assumed that the low catalytic activity of HbHNL in this reaction can be attributed to both the electronic and steric nature of nitroalkanes. Apart from nitromethane, HbHNL also accepts nitroethane as the nucleophile [59]. Molecular modeling studies revealed a similar binding mode for (S)-2-nitro-1-phenylethanol in the catalytic center of HbHNL as was determined experimentally for (S)-mandelonitrile, preserving all mechanistically important polar interactions with active-site residues. This implies that the mechanism for the cyanohydrin reaction applies to the nitroaldol reaction as well.

5.4 Transformation of Cyanohydrins Cyanohydrins can be transformed into variety of products using chemical reactions like (1) reaction of hydroxyl group (conversion of OH group to a good leaving group to allow nucleophilic displacement with inversion of configuration), (2) reactions of cyano group (solvolysis, Grignard reaction and hydride addition) and (3) intramolecular reactions.

5.4.1 Transformation of Hydroxyl Group Cyanohydrins reacted with acid chlorides or anhydrides and pyridine produce cyanohydrin esters, which is very important to preparation of pyrethroid such as deltamethrin and esfenvalerate (Figure 5.10). The esterification of hydroxyl (Figure 5.11) can also be validated using the Mitsunobu reaction with inversion of the configuration of aromatic and allylic cyanohydrins but retaining the configuration of aliphatic systems [60]. It should be noted that the cyanohydrins containing strongly electron-donating substituents gave extensive racemization. While N-protected amino was used as a Mitsunobu nucleophilic reagent, cyanohydrins can be transformed into N-protected 2-aminonitrile, which is the precursor of amino acid and 1,2-diamine [61]. Br Br

Cl

CN O

O

Cl

O

CN

O

O

O

O

CN O

O

Cl deltamethrin

cypermethrin

Figure 5.10

esfenvalerate

Pyrethroids

The hydroxyl group was usually protected, because cyanohydrins have tendency to racemization or even decomposition. Vinyl ethers or acetal and acid catalysts furnish acetals [62]. Trialkylsilyl chlorides and imidazole are used to give silyl ethers [63]. Commonly used protective groups are silyl ether, ester, methoxy isopropyl (MIP) ether, and tetrahydropyranyl ether. O-Protected cyanohydrins are tolerant to a wider range of cyanide/nitrile transformations and are utilized widely in the synthesis of compounds of synthetic relevance in organic chemistry. (R)-Cyanohydrins react with toluenesulfonyl chloride, methanesulfonyl chloride or 4-nitrobenzenesulfonyl chloride without loss of stereochemical purity, and the 2-sulfonyloxynitrile reacts with a variety of SN2 reactions to give a variety of products, such as 2-fluoro nitrile [64], 2-azidonitrile [65] and N-phthaloyl-protected 2-aminonitrile [66], 2-acetoxy nitrile [66], and 2-mercapto nitrile [67]. Hydrogenation of 2-sulfonyloxynitriles with LiAlH4 in good chemical yields and high ee afforded 2-monosubstituted (S)-aziridines [68].

100


O O

R

TMS

O O

R' CN

CN R 2-fluoro nitrile

CN

R CN 2-acetoxy nitrile

ry Cl, P R'CO OMe

MeO

F-

DMF, imide, R3SiCl OH

OTs

TsCl, Pyr R

R CN cyanohydrin POCl3 OMe

N3

NaOAc

R CN 2-azidonitrile

NaN3

CN

LiAlH4

KNPhth

NPhth R

NaSR'

CN

P

R

D EA

D, P Ph Boc( 3 SES) NH

CN

OAc

2O 5

N

R

Ph 3 ,P AD OH DE 'CO R

Boc

R'

F DAST

CN

R O2 S

OSiR3

OMOM R

CN

Figure 5.11

H N

OMIP R

CN

R

SR' R' = Ac, Bn R

CN

Transformation of hydroxyl group

5.4.2 Transformation of Nitrile Group The reduction of a-hydroxynitriles to yield vicinal amino alcohols is conveniently accomplished with complex metal hydrides; for example, lithium aluminum hydride or sodium borohydride [69]. However, it is still worth noting that a two-step chemo-enzymatic synthesis of (R)-2-amino-1-(2-furyl)ethanol for laboratory production was developed followed by successful up-scaling to kilogram scale using NaBH4/CF3COOH as reductant [70]. The cyanohydrins were hydrolyzed with concentrated HCl at ambient temperature in very high yield without any racemization to give 2-hydroxy carboxylic acids [71]. Likewise, cyanohydrins were solvolyzed in solution of 2 M HCl and 2 M alcohol to afford corresponding 2-hydroxy carboxylic ester [72]. The reductions of O-protected cyanohydrins with diisobutylalumium hydride (DIBAL-H) afford unstable imines (Figure 5.12). By working-up the reaction with H2SO4 it is possible to isolate a-hydroxy-aldehyde [73]; by quenching the reaction by MeOH followed by adding N-hydroxy(phenyl)methanamine one obtains a-hydroxy-aldonitrone [74]. 1,3-Dipolar cycloaddition of the chiral nitrone was also achieved [75]. The reduction of the cyano group can also be followed by transimination and reduction, affording N-substituted b-amino alcohol [76]. An alternative entry to N-substituted b-amino alcohol is the cyanohydrin esters undergoing the Ritter reaction followed by reduction [77]. Grignard reagents also add to the nitrile group of cyanohydrin trimethylsilyl ether, affording imine. By working-up the reaction with H2SO4 it is possible to isolate acyloin [78]; by quenching the reaction by MeOH followed by adding N-benzyl-hydroxylamine one obtains ahydroxy-ketonitrone [74]. The reaction of a nitrile with a Reformatsky reagent is known as the Blaise reaction and when applied to O-trimethylsilyl cyanohydrins leads to the formation of tetronic acids with high ee [79]. By working-up the Blaise reaction with ammonium chloride it is possible to isolate

101


r gB R'M HOH 1) N Bn 2)

CH2NH2

LiA or lH4 BH 3

OH COOR

R

CN

or POCl3, MIP

2 SO 4,

H

COOH

OAc

OH

R'

HOOC

R

Ph

O

Ph

N Cbz

1) TsOH, DHP 2) PhMgBr 3) NH2CH2COOMe

O

4) NaBH4 5) CbzCl 6) TsOH

NHR3

R

OH

NH2 from(S)-cyanohydrin

Figure 5.13

O-

Ar NHR'

NH2

(S)-

R

R2

AcHN R1

O

OH

OH OH

N+

NHR'

R

1) R3SiCl, imide ' 2) RMgBr 3) NaBH4/CH3OH 4) ¡- ¡1) TMSCl, imide 2) R2CH=CHCH2MgBr OH 3) NaBH4/CH3OH 4) Ac2O R CN 5) O3 6) NaOMe, then H+ 1) POCl3 2) DIDAL-H, then,CH3OH 3) BnNH2, HCN 4) (im)2CO, 5) KOH, then, HCl

OH C13H27

R'

R

NH2

NHFmoc

Bn

OH

OH R'

R

Cyanide group transformations

OH

OH

OSiR3

O

R

O

Figure 5.12

OH

CH 2 rZn B 1) + H 2) OP 1) DIBAL-H R CN 2) BnNHOH 1) 2) DIB A R' NH L-H 2, Na BH 4 OH

NHBut

R

R'

R e OM CO

-H IBAL 1) D + 2) H

HCl

OH R

R3SiCl, imide or Ac2O, Pyr

OH

MeOH, HCl

t-B uO H

R

O

R' M H + gBr

Bn OH

O O

O

-

1)

N+

R'

R

2)

R

R

OH

OSiR3 R'

OH NHCOCHCl2

COOH NHR3

MeO2S

thiamphenicol

Introduction of second chiral center

O

102


HNL HCN MeS

OH

OH

O

1) POCl3, MIP 2) DIDAL-H, then,CH3OH

CN

3) BnNH2, HCN 4) HCl,

MeS

95% yield, 99% e.e.

CN MeS

NHBnHCl

d.r. > 20:1 OH

1) MCPBA, 90%;¡¡ 2) 2N KOH, reflux, 85%; O 1) (im)2CO, TEA, 82%; 2) K2CO3, ethanol, then,1N HCl, 91%; 3) NaBH4, methanol, 85%.

O

MeS

3) Pd/C, H2, 90%; MeO2S 4) MeOH, CHCl2COOEt, TEA, 100%.

NBn OH

1) DAST, THF, 85%; 2) MCPBA, 90%; 3) 8N H2SO4, 140 °C, 71%; 4) Pd/C, H2 85%; MeO2S 5) MeOH, CHCl2COOEt, TEA, 100%

Figure 5.14

OH NHCOCHCl2

thiamphenicol OH F NHCOCHCl2

florfenicol

Syntheses of thiamphenicol and florfenicol starting from 4-methylsulfanyl-mandelonitrile

acyclic g-trimethylsilyloxy-b-amino-a,b-didehydro esters from this reaction. These compounds can subsequently be cyclized to the amino analogues of tetronic acids [80]. The transformation of the cyano group could also introduce a new chiral center under diastereoselective control (Figure 5.13). Grignard–transimination–reduction sequences have been employed in a synthesis of heterocyclic analogues of ephedrine [81]. The preferential formation of erythro-b-amino alcohols may be explained by preferential hydride attack on the less-hindered face of the intermediate imine [82], and hydrocyanation of the imine would also appear to proceed via the same type of transition state. In the case of a,b-unsaturated systems, reduction– transimination–reduction may be followed by protection of the b-amino alcohol to an oxazolidinone, ozonolysis with oxidative workup, and alkali hydrolysis to give a-hydroxyb-amino acids [83]. This method has been successfully employed in the synthesis L-threosphingosine [84]. Starting from enantiomerically pure 4-methylsulfanyl-mandelonitrile, thiamphenicol and florfenicol have been enantioselectively synthesized (Figure 5.14). The enantiomerically pure 4-methylsulfanyl-mandelonitrile was obtained by hydrocyanation reaction of 4-methylsulfanylbenzaldehyde catalyzed by (R)-hydroxynitrile lyase of Badamu (almond from Xinjiang, China) (Prunus communis L. var. dulcis Borkh), which, after an extensive screening, was found to be a highly effective bio-catalyst for this reaction [85].

5.4.3 Intramolecular Reaction Enzymatic transcyanation of v-bromoaldehydes could afford optically active v-bromocyanohydrins in high ee (95%) [86]. The v-bromocyanohydrins treated with silver perchlorate give (R)-2-cyanotetrahydrofuran or and (R)-2-cyanotetrahydropyran [86a]. The reduction of (R)-v-bromocyanohydrins affords, in one pot, piperidin-3-ol, azepan-3-ol or azocan-3-ol without racemization [86b]. Starting from hydroxyl-protected v-bromocyanohydrin, 2,3-cisdisubstituted piperidines were synthesized through a simple tandem Grignard addition– sodium borohydride reduction with complete diastereoselectivity (Figure 5.15) [86c]. 5-Hydroxypiperidin-2-one is a versatile building block for the preparation of potentially biologically active compounds. Griengl and coworkers detail an enantioselective biocatalytic

103


Ag+

n O

CN

OH Br

CN

1) BH 3

n

n R3 N Cbz

2) CBzCl

OH

n = 1, 2

OH

1) R 3MgBr

n

2) NaBH 4 3) CBzCl

Figure 5.15

R3 N Cbz

Transformation of v-bromocyanohydrin

approach towards its synthesis using HbHNL-mediated cyanohydrin formation, followed by hydrogenation. By adjusting the conditions of the latter step, 5-hydroxypiperidinone-derived (bicyclic) N,N-acetals were obtained via an unprecedented reductive amination of the nitrile group, as well as from N-alkylated 5-hydroxypiperidinone in a single step from the same cyanohydrin intermediate (Figure 5.16) [87].

OMOM

OMOM

Pd/C, H2 NH3

O

N H

O

NH2

NH2

N H

86% 42:58 OMOM

Pd/C, H2 BuNH2 MeOOC

O

CN OMOM

N Bu

97%

MOMO

Pd/C, H2 NH2

H2N

HN

MOMO N

O

HN

N

O

98% 67:33 Pd/C, H2

MOMO HN

H2N

MOMO N

O

NH2 86% 64:36

Figure 5.16

Novel reductive amination of nitriles

HN

N

O

104


The fact that a,b-unsaturated aldehydes are good substrates for the HbHNL (Hevea brasiliensis) has been exploited in the synthesis of natural products. Thus, Johnson and Griengl developed a nine-step synthesis of (S)-coriolic acid starting from oct-2-enal, as shown in Figure 5.17. The addition of HCN to the aldehyde gave the (S)-cyanohydrin with 99% ee. Esterification of the cyanohydrin followed by palladium-catalyzed [3,3]-sigmatropic rearrangement gave a,b-unsaturated nitrile with complete retention of configuration. The remainder of the synthesis involved standard protection, reduction, Wittig reaction and deprotection steps [88].

O

CN

CN

O O 1) HCN, HbHNL

O

Pd(CH3CN)2Cl2

HO

O

2) (C3H7CO)2O, Pyr, DMAP COOH 13-(S)-HODE

Figure 5.17

Synthesis of (S)-coriolic acid

Effenberger and coworkers have utilized the tolerance of methyl ketones by the recombinant enzyme to develop an alternative synthesis of tetronic acids and their amino derivatives, as shown in Figure 5.18. Treatment of O-acyl cyanohydrins with lithium disilazide resulted in base-induced ring closure to amino tetronic acid derivatives. Alternatively, the cyanohydrins could be converted to a-hydroxy esters prior to acylation, and the same base-induced cyclization then led to tetronic acid derivatives [89]. O R2

O

R2CH2COCl/Pyr

LiN(SiMe3)2 CN

R1 O R1

HCN

R1

CN

O 1) HCl, ethanol

O

2

2) R CH2COCl/Pyr

Figure 5.18

O R2

H2N

OH

(S)-MeHNL

O

R1

R1

R2

LiN(SiMe3)2

R1

COOEt HO

O

O R2

Synthesis of tetronic acid derivatives

5.5 Summary Cyanohydrins have considerable synthetic potential as chiral building blocks, especially in a wide range of pharmaceutical and agrochemical applications. The remarkable properties of the HNLs can be exploited in catalyzing stereoselective synthesis of cyanohydrins. Especially


105

during the last two decades, HNLs have been developed into valuable biocatalysts, which can be produced in recombinant form by overexpression in microbial hosts, resulting in the implementation of industrial processes utilizing these enzymes. Recently, protein engineering in combination with high-throughput screening gave rise to the development of a tailor-made HNL for large-scale manufacturing of a highly valuable cyanohydrin.

References [1] Rosenthaler, L. (1908) Durch Enzyme bewirkte asymmetrische Synthesen. Biochemische Zeitschrift, 14, 238–253 [2] (a) Becker, W., Benthin, U., Eschenhof, E. and Pfeil, E. (1963) Zur Kenntnis der Cyanhydrinsynthese. II. Reindarstellung und Eigenschaften der Oxynitrilase aus Mandeln (Prunus communis Stokes). Biochemische Zeitschrift, 337, 156–166; (b) Becker, W. and Pfeil, E. (1965) Uber das Flavinenzyme Image d-Oxynitrilase. Biochemische Zeitschrift, 346, 301–321; (c) Becker, W., Freund, H. and Pfeil, E. (1965) Stereospecific synthesis of d-hydroxy nitriles and optically active ethanolamines. Angewandte Chemie (International Edition in English), 4, 1079. [3] Effenberger, F., Ziegler, T. and Forster, S. (1987) Enzyme-catalyzed cyanohydrin. synthesis in organic solvents. Angewandte Chemie (International Edition in English), 26, 458–460. [4] Siegler, D.S. and Brinker, A.M. (1993) Characterisation of cyanogenic glycoside, cyanolipids, nitroglycosides, organic nitro compounds and nitrile glycosides from plants, in Methods of Plant Biochemistry, Alkaloids and Sulfur Compounds (eds P.M. Dey and J.B. Harborne) Academic Press, pp. 51–93. [5] Stumpf, P.K. and Conn, E.E. (1981) The Biochemistry of Plants, a Comprehensive Treatise Secondary Plant Products, 7, Academic Press, New York, pp. 479–500. [6] Knowles, C.J. and Bunch, A.W. (1986) Microbial cyanide metabolism. Advances in Microbial Physiology, 27, 73–111. [7] Bunch, A.W. and Knowles, C.J. (1980) Cyanide production and degradation during growth of the snow mold fungus. Journal of General Microbiology, 116, 9–16. [8] Knowles, C.J. (1988) Cyanide utilization and degradation by microorganisms. Ciba Foundation Symposium, 140, 3–15. [9] Lechtenberg, M. and Nahrstedt, A. (1999) Cyanogenic glycosides, in Naturally Occurring Glycosides (ed. R. Ikan) John Wiley & Sons, Ltd, Chichester, UK, pp. 147–191. [10] Sharma, M., Sharma, N.N. and Bhalla, T.C. (2005) Hydroxynitrile lyases: at the interface of biology and chemistry. Enzyme and Microbial Technology, 37, 279–294. [11] Ollis, D.L., Cheah, E., Cygler, M. et al. (1970) The alpha/beta hydrolase fold. Protein Engineering, 5, 197–211. [12] Holmquist, M. (2000) Alpha/beta-hydrolase fold enzymes: structures, functions and mechanisms. Current Protein & Peptide Science, 1, 209–235. [13] Kiess, M., Hecht, H.J. and Kalisz, H.M. (1998) Glucose oxidase from Penicillium amagasakiense: primary structure and comparison with other glucose–methanol–choline (GMC) oxidoreductases. European Journal of Biochemistry, 252, 90–99. [14] Wierenga, R.K., Terpstra, P. and Hol, W.G.J. (1986) Prediction of the occurrence of the ADP-binding bab-fold in proteins, using an amino acid sequence fingerprint. Journal of Molecular Biology, 187, 101–107. [15] Dreveny, I., Gruber, K., Glieder, A. et al. (2001) The hydroxynitrile lyase from almond: a lyase that looks like an oxidoreductase. Structure (London, England: 1993), 9, 803–815. [16] Dreveny, I., Kratky, C. and Gruber, K. (2002) The active site of hydroxynitrile lyase from Prunus amygdalus: modeling studies provide new insights into the mechanism of cyanogenesis. Protein Science: A Publication of the Protein Society, 11, 292–300. [17] (a) Trummler, K. and Wajant, H. (1997) Molecular cloning of acetone cyanohydrin lyase from flax (Linum usitatissimum). Definition of a novel class of hydroxynitrile lyases. The Journal of Biological Chemistry, 272, 4770–4774; (b) Trummler, K., Roos, J., Schwaneberg, J.R.U. et al. (1998) Expression of the Zn2þ -containing hydroxynitrile lyase from flax (Linum usitatissimum) in Pichia pastoris – utilization of the recombinant enzyme for enzymatic analysis and site-directed mutagenesis. Plant Science (Shannon, Ireland), 139, 19–27. [18] Hasslacher, M., Schall, M., Hayn, M. et al. (1997) High-level intracellular expression of hydroxynitrile lyase from the tropical rubber tree Hevea brasiliensis in microbial hosts. Protein Expression and Purification, 11, 61–71.

106


[19] Zuegg, J., Gruber, K., Gugganig, M. et al. (1999) Three-dimensional structures of enzyme–substrate complexes of the hydroxynitrile lyase from Hevea brasiliensis. Protein Science: A Publication of the Protein Society, 8, 1990–2000. [20] Wagner, U.G., Hasslacher, M., Griengl, H. et al. (1996) Mechanism of cyanogenesis: the crystal structure of hydroxynitrile lyase from Hevea brasiliensis. Structure (London, England: 1993), 4, 811–822. [21] (a) Wajant, H. and Pfizenmaier, K. (1996) Identification of potential active-site residues in the hydroxynitrile lyase from Manihot esculenta by site-directed mutagenesis. The Journal of Biological Chemistry, 271, 25830–25834; (b) Lauble, H., Miehlich, B., F€orster, S. et al. (2001) Mechanistic aspects of cyanogenesis from active-site mutant Ser80Ala of hydroxynitrile lyase from Manihot esculenta in complex with acetone cyanohydrin. Protein Science: A Publication of the Protein Society, 10, 1015–1022. [22] Lauble, H., Miehlich, B., F€orster, S. et al. (2002) Crystal structure of hydroxynitrile lyase from Sorghum bicolor in complex with the inhibitor benzoic acid: a novel cyanogenic enzyme. Biochemistry, 41, 12043–12050. [23] Kuroki, G.W. and Conn, E.E. (1989) Mandelonitrile lyase from Ximenia americana L.: stereospecificity and lack of flavin prosthetic group. Proceedings of the National Academy of Sciences of the United States of America, 86, 6978–6981. [24] Wajant, H. and Effenberger, F. (1996) Hydroxynitrile lyases of higher plants. The Journal of Biological Chemistry, 377, 611–617. [25] Andexer, J., von Langermann, J., Mell, A. et al. (2007) An R-selective hydroxynitrile lyase from Arabidopsis thaliana with an alpha/beta-hydrolase fold. Angewandte Chemie–International Edition, 46, 8679–8681. [26] Gerrits, P.J., Zumbr€agel F. and Marcus, J. (2001) Analyzing the hydrocyanation reaction: chiral HPLC and the synthesis of racemic cyanohydrins. Tetrahedron, 57, 8691–8698. [27] Hernandez, L., Luna, H., Ruız-Teran, F. and Vazquez, A. (2004) Screening for hydroxynitrile lyase activity in crude preparations of some edible plants. Journal of Molecular Catalysis B–Enzymatic, 30, 105–108. [28] (a) Asano, Y., Tamura, K., Doi, N. et al. (2005) Screening for new hydroxynitrilases from plants. Bioscience, Biotechnology, and Biochemistry, 69, 2349; (b) Hickel, A., Heinrich, G., Schwab, H. and Griengl, H. (1997) Screening for hydroxynitrile lyases in plants. Biotechnology Techniques, 11, 55–58. [29] Hernandez, L., Luna, H., Solısa, A. and Vazquez, A. (2006) Application of crude preparations of leaves from food plants for the formation of cyanohydrins with high enantiomeric excesses. Tetrahedron: Asymmetry, 17, 2813–2816. [30] Hughes, J., Decarvalho, J.P.C. and Hughes, M.A. (1994) Purification, characterization, and cloning of alphahydroxynitrile lyase from cassava (Manihot esculenta Crantz). Archives of Biochemistry and Biophysics, 311, 496–502. [31] Hughes, J., Lakey, J.H. and Hughes, M.A. (1997) Production and characterization of a plant alpha-hydroxynitrile lyase in Escherichia coli. Biotechnology and Bioengineering, 53, 332–338. [32] Hasslacher, M., Schall, M., Hayn, M. et al. (1996) Molecular cloning of the full-length cDNA of (S)hydroxynitrile lyase from Hevea brasiliensis. Functional expression in Escherichia coli and Saccharomyces cerevisiae and identification of an active site residue. The Journal of Biological Chemistry, 271, 5884–5891. [33] F€ orster, S., Roos, J., Effenberger, F. et al. (1996) The first recombinant hydroxynitrile lyase and its application in the synthesis of (S)-cyanohydrins. Angewandte Chemie (International Edition in English), 35, 437–439. [34] Hasslacher, M., Schall, M., Hayn, M. et al. (1997) High-level intracellular expression of hydroxynitrile lyase from the tropical rubber tree Hevea brasiliensis in microbial hosts. Protein Expression and Purification, 11, 61–71. [35] Glieder, A., Weis, R., Skranc, W. et al. (2003) Comprehensive step-by-step engineering of an (R)-hydroxynitrile lyase for large-scale asymmetric synthesis. Angewandte Chemie–International Edition, 42, 4815–4818. [36] Liu, Z., Pscheidt, B., Avi, M. et al. (2008) Laboratory evolved biocatalysts for stereoselective syntheses of substituted benzaldehyde cyanohydrins. Chem Bio Chem, 9, 58–61. [37] Weis, R., Gaisberger, R., Skranc, W. et al. (2005) Carving the active site of almond R-HNL for increased enantioselectivity. Angewandte Chemie–International Edition, 44, 4700–4704. [38] 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. Biotechnology and Bioengineering, 62, 20–29. [39] Reisinger, C., van Assema, F., Schuermann, M. et al. (2006) A versatile colony assay based on NADH fluorescence. Journal of Molecular Catalysis B–Enzymatic, 39, 149–155. [40] Andexer, J., Guterl, J.-K., Pohl, M. and Eggert, T. (2006) A high-throughput screening assay for hydroxynitrile lyase activity. Chemical Communications, 4201–4203. [41] Krammer, B., Rumbold, K., Tschemmernegg, M. et al. (2007) A novel screening assay for hydroxynitrile lyases suitable for high-throughput screening. Journal of Biotechnology, 129, 151–161.


107

[42] (a) Koch, K., van den Berg, R.J.F., Nieuwland P.J. et al. (2008) Enzymatic synthesis of optically pure cyanohydrins in microchannels using a crude cell lysate. Chemical Engineering Journal, 135 (suppl 1), S89–S92; (b) Koch, K., van den Berg, R.J.F., Nieuwland P.J. et al. (2008) Enzymatic enantioselective C–Cbond formation in microreactors. Biotechnology and Bioengineering, 99, 1028–1033. [43] (a) Jaeger, K.-E., Eggert, T., Eipper, A. and Reetz, M.T. (2001) Directed evolution and the creation of enantioselective biocatalysts. Applied Microbiology and Biotechnology, 55, 519–530; (b) Berglund, P. and Park, S. (2005) Strategies for altering enzyme reaction specificity for applied biocatalysis. Current Organic Chemistry, 9, 325–336; (c) Lorenz, P. and Eck, J. (2005) Metagenomics and industrial applications. Nature Reviews Microbiology, 3, 510–516. [44] Willeman, W.F., Straathof, A.J.J. and Heijnen, J.J. (2002) Reaction temperature optimization procedure for the synthesis of (R)-mandelonitrile by Prunus amygdalus hydroxynitrile lyase using a process model approach. Enzyme and Microbial Technology, 30, 200–208. [45] Persson, M., Costes, D., Wehtje, E. and Adlercreutz, P. (2002) Effects of solvent, water activity and temperature on lipase and hydroxynitrile lyase enantioselectivity. Enzyme and Microbial Technology, 30, 916–923. [46] Ognyanov, V.I., Datcheva, V.K. and Kyler, K.S. (1991) Preparation of chiral cyanohydrins by an oxynitrilasemediated transcyanation. Journal of the American Chemical Society, 113, 6992–6996. [47] (a) Effenberger, F. (1994) Synthesis and reactions of optically active cyanohydrins. Angewandte Chemie (International Edition in English), 33, 1555–1564; (b) Effenberger, F. (2000) Hydroxynitryl lipases in stereoselective synthesis, in Stereoselective Biocatalysis (ed. R.N. Patel), Marcel Dekker Inc., New York, p. 332. [48] Hickel, A., Radke, C.J. and Blanch, H.W. (1999) Hydroxynitrile lyase at the diisopropyl ether/water interface: evidence for interfacial enzyme activity. Biotechnology and Bioengineering, 65, 425–436. [49] Pogorevc, M., Stecher, H. and Faber, K. (2002) A caveat for the use of log P values for the assessment of the biocompatibility of organic solvents. Biotechnology Letters, 24, 857–860. [50] (a) Han, S., Lin, G. and Li, Z. (1998) Synthesis of (R)-cyanohydrins by crude (R)-oxynitrilase-catalyzed reactions in micro-aqueous medium. Tetrahedron: Asymmetry, 9, 1835–1838; (b) Lin, G., Han, S. and Li, Z. (1999) Enzymatic synthesis of (R)-cyanohydrins by three (R)-oxynitrilase sources in micro-aqueous organic medium. Tetrahedron, 55, 3531–3540; (c) Han, S., Chen, P., Lin, G. et al. (2001) (R)-oxynitrilase-catalyzed hydrocyanation: the first synthesis of optically active fluorinated mandelonitriles. Tetrahedron: Asymmetry, 12, 843–846; (d) Chen, P., Han, S., Lin, G. et al. (2001) A study of asymmetric hydrocyanation of heteroaryl carboxaldehydes catalyzed by (R)-oxynitrilase under micro-aqueous conditions. Tetrahedron: Asymmetry, 12, 3273–3279. [51] Gaisberger, R.P., Fechter, M.H. and Griengl, H. (2004) The first hydroxynitrile lyase catalysed cyanohydrin formation in ionic liquids. Tetrahedron: Asymmetry, 15, 2959–2963. [52] Von Langermann, J., Mell, A., Paetzold, E. et al. (2007) Hydroxynitrile lyase in organic solvent-free systems to over come thermodynamic limitations. Advanced Synthesis and Catalysis, 349, 1418–1424. [53] Costes, D., Wehtje, E. and Adlercreutz, P. (2001) Cross-linked crystals of hydroxynitrile lyase as catalyst for the synthesis of optically active cyanohydrins. Journal of Molecular Catalysis B–Enzymatic, 11, 607–612. [54] Veum, L., Hanefeld, U. and Pierre, A. (2004) The first encapsulation of hydroxynitrile lyase from Hevea brasiliensis in a sol–gel matrix. Tetrahedron, 60, 10419–10425. [55] Groger, H., Copan, E., Bathuber, A. and Vorlop, K. (2001) Asymmetric synthesis of an (R)-cyanohydrin using enzymes entrapped in lens-shaped gels. Organic Letters, 13, 1969–1971. [56] Van Langen, L.M., Selassa, R.P., van Rantwijk, F. and Sheldon, R.A. (2005) Cross-linked aggregates of (R)-oxynitrilase: a stable, recyclable biocatalyst for enantioselective hydrocyanation. Organic Letters, 7, 327–331. [57] Chen, P., Han, S., Lin, G. and Li, Z. (2002) A practical high through-put continuous process for the synthesis of chiral cyanohydrins. The Journal of Organic Chemistry, 67, 8251–8253. [58] Purkarthofer, T., Gruber, K., Gruber-Khadjawi, M.,et al. (2006) A biocatalytic Henry reaction – the hydroxynitrile lyase from Hevea brasiliensis also catalyzes nitroaldol reactions. Angewandte Chemie (International Edition in English), 45, 3454–3456. [59] Gruber-Khadjawi, M., Purkarthofer, T. Skranc, W. et al. (2007) Hydroxynitrile lyase-catalyzed enzymatic nitroaldol (Henry) reaction. Advanced Synthesis and Catalysis, 349, 1445–1450. [60] (a) V€antinnen, E. and Kanerva, L.T. (1995) Combination of the lipase-catalysed resolution with the Mitsunobu esterification in one pot. Tetrahedron: Asymmetry, 6, 1779–1786; (b) Warmerdam, E.G.J.C., Brussee, J., Kruse, C.G. and van der Gen, A. (1993) Inversion of the configuration of cyanohydrins by a Mitsunobu esterification reaction. Tetrahedron, 49, 1063–1070.

108


[61] Decicco, C.P. and Grover, P. (1997) Asymmetric amino acid synthesis: Mitsunobu reactions on chiral cyanohydrins. Synlett, 529–530. [62] Zandbergen, P., Willems, H.M.G., van der Marel, G.A. et al. (1992) 2-Phenoxypropene as protective reagent of chiral alcohols. Synthetic Communications, 22, 2781–2787. [63] Brussee, J., Loos, W.T., Kruse, C.G. and van der Gen, A. (1990) Synthesis of optically active silyl protected cyanohydrins. Tetrahedron, 46, 979–986. [64] Stelzer, U. and Effenberger, F. (1993) Preparation of (S)-fluoronitriles. Tetrahedron: Asymmetry, 4, 161–164. [65] Effenberger, F., Kremser, A. and Stelzer, U. (1996) A convenient synthesis of (S)-2-azidonitriles, (S)-2aminonitriles and (S)-1,2-diamines. Tetrahedron: Asymmetry, 7, 607. [66] Effenberger, F. and Stelzer, U. (1991) Synthesis and stereoselective reactions of (R)-a-sulfonyloxynitriles. Angewandte Chemie (International Edition in English), 30, 873–874. [67] Effenberger, F. and Gaupp, S. (1999) Stereoselective substitution of (R)-2-(sulfonyloxy)nitriles with sulfur nucleophiles. Tetrahedron: Asymmetry, 10, 1765–1776. [68] Effenberger, F. and Stelzer, U. (1995) A convenient preparation of 2-substituted (S)-aziridines. Tetrahedron: Asymmetry, 6, 283–286. [69] (a) Ziegler, T., H€ orsch, B. and Effenberger, F. (1990) A convenient route to (R)-a-hydroxy carboxylic acids and (2R)1-amino-2-alkanols from (R)-cyanohydrins. Synthesis, 575–578; (b) Monterde, M.I., Nazabadioko, S., Robolledo, F. et al. (1999) Chemoenzymatic synthesis of azacycloalkan-3-ols. Tetrahedron: Asymmetry, 10, 3449–3455. [70] Purkarthofer, T., Pabst, T., van den Broek, C. et al. (2006) Large-scale synthesis of (R)-2-amino-1-(2-furyl) ethanol via a chemoenzymatic approach. Organic Process Research & Development, 10, 618–621. [71] Effenberger, F., Hoersch, B., Weingart, F. et al. (1991) Enzyme-catalyzed synthesis or (R)-ketone-cyanohydrins and their hydrolysis to (R)-a-hydroxy-a-methyl-carboxylic acids. Tetrahedron Letters, 32, 2605–2608. [72] Warmerdam, E.G.J.C., van den Nieuwendijk, A.M.C.H., Kruse, C.G. et al. (1996) Synthesis of (R)- and (S)-2hydroxy-3-enoic acid esters. Recueil des Travaux Chimiques des Pays-Bas, 115, 20–24. [73] Hayashi, M., Yoshiga, T., Nakatani, K. et al. (1994) Reduction of a-trialkylsiloxy nitriles with diisobutylaluminium hydride (DIBAH): a facile preparation of a-trialkylsiloxy aldehydes and their derivatives. Tetrahedron, 50, 2821–2830. [74] Hulsbos, E., Marcus, J., Brussee, J. and van der Gen, A. (1997) Direct conversion of chiral cyanohydrins to chiral nitrones by transimination. Tetrahedron: Asymmetry, 8, 1061–1067. [75] Van den Nieuwendijk, A.M.C.H., Ghisaidoobe, A.B.T., Overkleeft, H.S. et al. (2004) Conversion of chiral unsaturated cyanohydrins into chiral carba- and heterocycles via ring-closing metathesis. Tetrahedron, 60, 10385–10396. [76] Zandbergen, P., van den Nieuwendijk, A.M.C.H., Brussee, J. et al. (1992) A one-pot reduction–transimition– reduction synthesis of N-substituted b-ethanolamines from cyanohydrins. Tetrahedron, 48, 3977–3982. [77] Effenberger, F. and J€ager, J. (1997) Synthesis of the adrenergic bronchodilators (R)-terbutaline and (R)salbutamol from (R)-cyanohydrins. The Journal of Organic Chemistry, 62, 3867–3873. [78] Krepski, L.R., Jensen, K.M., Heilmann, S.M. and Rasmussen, J.K. (1986) Synthesis, 301–303. [79] Duffield, J.J. and Regan, A.C. (1996) Asymmetric synthesis of tetronic acids by Blaise reaction of protected optically active cyanohydrins. Tetrahedron: Asymmetry, 7, 663–666. [80] Syed, J., F€ orster, S. and Effenberger, F. (1998) Application of the Blaise reaction: stereoselective synthesis of (4R)-tert-butyl 3-amino-4-trimethylsilyloxy-2-alkenoates from (R)-cyanohydrins. Tetrahedron: Asymmetry, 9, 805–815. [81] Effenberger, F. and Eichhorn, J. (1997) Stereoselective synthesis of thienyl and furyl analogues of ephedrine. Tetrahedron: Asymmetry, 8, 469–476. [82] Krepski, L.R., Jensen, K.M., Heilmann, S.M. and Rasmussen, J.K. (1986) A new synthesis of 2-aminoalcohols from O-trimethylsilylated cyanohydrins. Synthesis, 301–303. [83] Warmerdam, E.G.J.C., van Rijn, R.D., Brussee, J. et al. (1996) Synthesis of a-hydroxy-b-amino acids from chiral cyanohydrins. Tetrahedron: Asymmetry, 7, 1723–1732. [84] Johnson, D.V., Felfer, U. and Griengl, H. (2000) A chemoenzymatic access to D- and L-sphingosines employing hydroxynitrile lyases. Tetrahedron, 56, 781–790. [85] (a) Lin, G., Lu, W. and Chen, P. ‘Synthetic routes for the preparation of thiamphenicol and florfenical and key intermediates’. China patent, CN1743308; (b) Lin, G., Lu, W. and Chen, P. ‘Enzymatic synthesis of optically active cyanohydrins’. China patent, CN1743309. [86] (a) Menendez, E., Brieva, R., Rebolledo, F. and Gotor, V. (1995) Optically active (S)-ketone- and (R)aldehyde-cyanohydrins via an (R)-oxynitrifase-catalysed transcyanation. Chemoenzymatic syntheses of 2-


109

cyanotetrahydrofuran and 2-cyanotetrahydropyran. Journal of the Chemical Society. Chemical Communications, 10, 989–990; (b) Monterde, M.I., Nazabadioko, S., Rebolledo, F. et al. (1999) Chemoenzymatic synthesis of azacycloalkan-3-ols. Tetrahedron: Asymmetry, 10, 3449–3455; (c) Monterde, M.I., Brieva, R. and Gotor, V. (2001) Stereocontrolled chemoenzymatic synthesis of 2,3-disubstituted piperidines. Tetrahedron: Asymmetry, 12, 525–528 [87] Vink, M.K.S., Schortinghuis, C.A., Mackova-Zabelinskaja, A. et al. (2003) Novel reductive amination of nitriles: an efficient route to 5-hydroxypiperidone-derived N,N-acetals. Advanced Synthesis and Catalysis, 345, 483–487. [88] Johnson, D.V. and Griengl, H. (1997) The chemoenzymatic synthesis of (S)-13-hydroxyoctadeca-(9Z,11E)dienoic acid using the hydroxynitrile lyase from Hevea brasiliensis. Tetrahedron, 53, 617–624. [89] B€ uhler, H., Bayer, A. and Effenberger, F. (2000) Enzyme-catalyzed reactions, part 39. A convenient synthesis of optically active 5,5-disubstituted 4-amino- and 4-hydroxy-2(5H)-furanones from (S)-ketone cyanohydrins. Chemistry – A European Journal, 6, 2564–2571.

6 Expanding the Scope of Aldolases as Tools for Organic Synthesis William A. Greenberg Department of Chemistry, The Scripps Research Institute, La Jolla, CA 92037, USA

Among the many classes of enzymes that have been used as biocatalysts, the aldolases are particularly appealing because they stereoselectively catalyze carbon–carbon bond-forming reactions, which are central to the practice of the organic chemist. Over the past two decades, a wide variety of synthetic applications of aldolases have been demonstrated [1–4]. These efforts mapped out the scope and limitations of these enzymes as practical catalysts for preparative synthesis. The major limitation has been perceived to be narrow substrate specificity, which is not surprising when considering the physiological role of aldolases. They have evolved to accelerate the rate of aldol or retroaldol reactions on very specific metabolites in vivo, and modification of substrate structure or reaction conditions to suit the needs of the organic chemist generally leads to reduced catalytic efficiency. However, the advent of modern molecular biology techniques and high-throughput screening technology has allowed for the systematic evolution of biocatalysts with more useful properties, whether for broader substrate scope, enhanced catalytic efficiency, improved stability, or even entirely new activities [5–9]. Significant progress has been made in evolution of new or improved aldolase functions. In other cases, modification of substrates or reaction conditions has led to desired effects. In addition, re-evaluation of the substrate scope of wildtype enzymes has to some extent broken down the dogma that aldolases possess very narrow substrate tolerance. In this chapter, the most recent of these efforts to expand the utility of aldolases will be summarized.

6.1 Directed Evolution and Rational Mutagenesis The power of directed evolution has been demonstrated by the conversion of an aldolase into a new kind of aldolase. Wong and coworkers evolved a pyruvate-dependent sialic acid aldolase Biocatalysis for the Pharmaceutical Industry edited by Junhua Tao, Guo-Qiang Lin, and Andreas Liese Ó 2009 John Wiley & Sons Asia (Pte) Ltd


112

NHAc O

HO HO HO

HO

O

wild-type

OH

+

CO2H

sialic acid aldolase

D-ManNAc

OH CO2H

HO AcHN

O

OH

HO

Pyruvate

D-sialic acid

Directed Evolution: Error-prone PCR

OH OH OH

O

O

+

OH

HO

Mutant Aldolase

TMS

CO2H

Figure 6.1

OH OH

OH

Pyruvate

L-arabinose

OH

O

L-KDO

Alteration of substrate specificity of sialic acid aldolase by directed evolution

into an L-3-deoxymanno-2-octulosonic acid (L-KDO) aldolase, a function not found in nature [10]. The acceptor substrate preference was changed from wild-type N-acetyl-Dmannosamine to L-arabinose (Figure 6.1), while maintaining a wild-type level of catalytic efficiency. Berry and coworkers evolved a tagatose 1,6-bisphosphate aldolase into a fructose 1,6-phosphate aldolase [11]. In this case, the same substrates, dihydroxyacetone phosphate and D-glyceraldehyde-3-phosphate, were being used, but the stereochemical sense of the newly formed carbon–carbon bond was altered (Figure 6.2). Directed evolution methods, as well as rational structure-based mutagenesis approaches, have been successful in broadening the substrate tolerance of aldolases. A common goal is to

O

O 2-O PO 3

OH

OH O

+

OPO32-

WT TBP Aldolase

OPO32-

2-O PO 3

OH

OH OH

D-tagatose-1,6-bisphosphate Directed Evolution: DNA Shuffling

O

O 2-O PO 3

OH

OH O

Mutant FBP Aldolase +

OPO32OH

OPO32-

2-O PO 3

OH OH

D-fructose-1,6-bisphosphate

Figure 6.2

Alteration of product stereochemistry from the same substrates by directed evolution

Expanding the Scope of Aldolases as Tools for Organic Synthesis

OH OH O + OH

HO

CO2H

NHAc

D-ManNAc

OH H

WT sialic acid aldolase

O

HO

113

OH AcHN

Pyruvate

O

CO2H OH

OH

D-sialic acid Active site Saturation Mutagenesis

+

N O

CO2H

OH

E192N/T167G: A selective E192N/T167V/S208V: B selective

Figure 6.3

O

O

E192N Mutant, 82:18 A:B

O

OH O

N HO

O

CO2H OH +

OH

A

N HO

O

CO2H OH

OH

B

Synthesis of modified sugars with sialic acid aldolase mutants

overcome the requirement for phosphorylated substrates. While most aldolases utilize phosphorlyated substrates in vivo, obviating the need for them would significantly reduce the cost and number of steps involved in an aldolase-mediated synthesis. Wong and coworkers achieved this goal in an early application of directed evolution [12], converting Escherichia coli 2-keto-3-deoxy-6-phosphogluconate (KDPG) aldolase into a 3-deoxy-2-keto-gluconate aldolase. Structure-based mutagenesis achieved the same goal for deoxyribose-5-phosphate aldolase (DERA) [13,14]. In general, aldolases with greater binding-site plasticity, leading to a broader substrate tolerance, will be more useful as biocatalysts. In several cases, just a few amino acid mutations have been enough to achieve this goal without greatly affecting catalytic efficiency. With a single mutation site suggested by analysis of the crystal structure of the enzyme, Nelson and coworkers were able to coax a sialic acid aldolase into accepting substrates that were very different from the wild-type acceptor N-acetylmannosamine [15] (Figure 6.3). Further mutagenesis and screening led to the discovery of new variants with high diastereoselectivity, generating a pair of complementary biocatalysts [16]. Recently, Fierke and coworkers observed a similar effect from a single binding site mutation in KDPG aldolase [17]. Directed evolution has also been applied to address specific technical challenges in commercially important biotransformations. The group at DSM used it to evolve a mutant DERA that could tolerate the high concentrations of electrophilic aldehyde substrate that were required in an economically viable large-scale process for synthesis of the statin side chain of HMG-CoA reductase inhibitors [18]. Recently, Ran and coworkers reported an evolved KDPGal aldolase with improved activity in the synthesis of 3-deoxy-D-arabino-heptulosonic acid-7-phosphate (DAHP). The evolved enzyme was incorporated into a whole-cell process for synthesis of 3-dehydroshikimate, leading to production levels of almost 20 g L 1 of this highly functionalized chiral intermediate (Figure 6.4) [19,20].


114 O

O OH

+

OPO32-

WT KDPGal Aldolase

CO2H HO

OH

O

OH

O 2-O PO 3

KDPGal Directed Evolution

HO

O

O OH O

OH OPO32-

+ OH

CO2H

O

Mutant Aldolase

OH 2-O

3PO

OH DAHP

whole cell biotransformation CO2H

O

OH OH

Figure 6.4 Directed evolution of KDPGal aldolase for application to whole-cell production of 3-dehydroshikimate

6.2 Reaction Engineering In addition to the enzymes amino acid sequence, other parameters can affect the outcome of a biocatalytic process. For instance, a similar outcome in the aforementioned DERA-catalyzed statin synthesis was achieved by process improvements [21]. Using a thermostable variant of DERA (thermostability generally correlates well with tolerance to high concentrations of organic reagents or cosolvents), and fed-batch conditions, an efficient process that overcame sensitivity to high concentrations of chloroacetaldehyde was developed. Modification of a substrate with a protecting group can have profound effects on the outcome of an aldolase-catalyzed reaction. Bull and coworkers found that an inherently non-diastereoselective thermostable 2-keto-3-deoxygluconate aldolase (KDGA) became diastereoselective when glyceraldehyde acetonide was used in place of glyceraldehyde [22] (Figure 6.5). Although additional protection and deprotection steps are present, protecting groups can often assist in isolation of otherwise polar material, and the differentially protected polyols that are produced may be carried on as useful synthons for further elaboration. In the case of L-rhamnulose-1-phosphate aldolase (RhaD), we found that the problem of phosphorylated substrate requirement (dihydroxyactone phosphate (DHAP)) could be overcome by a simple change in buffer. Thus, when using borate buffer, reversible borate ester formation created a viable substrate out of dihydroxyacetone, which is not otherwise accepted by the wild-type enzyme (Figure 6.6) [23]. The process was used in a one-step synthesis of

Expanding the Scope of Aldolases as Tools for Organic Synthesis O

O

OH

HO2C

KDGA

OH

O

Pyruvate OH

115

OH

+ HO2C

OH

OH

OH

OH

4S, 5R

4R, 5R

1:1 diastereomeric ratio O

O

Pyruvate OH

KDGA

OH

OH

O

HO2C

OH

OH

+ HO2C

OH

OH

OH

4R, 5S

4S, 5S

~1:1 diastereomeric ratio

O

O

Pyruvate O

HO2C

KDGA

O

OH O O

4S, 5R >92% de

O

O

OH

Pyruvate O

HO2C

KDGA

O

O O

4S, 5S >94% de

Substrate engineering with protecting groups improves diastereoselectivity of KDGA

Figure 6.5

O

O HO

OH

O

RhaD OH OH

DHA

sodium borate pH 7.6

O

O HO

OH DHA

O

B OH

L-fructose OH

1. DHA, RhaD, borate

+

N3 OH

OH OH OH

O HO

OH

HO

+

2. Pd/C, H2, MeOH

HO

H N

HO

OH OH

L-deoxymannojirimycin

Figure 6.6

In situ borate ester formation obviates the need for phosphorylated substrates with RhaD


116

L-fructose and two-step synthesis of L-iminocyclitols. It remains to be seen how general the borate effect is. A classical approach to driving the unfavorable equilibrium of an enzymatic process is to couple it to another, irreversible enzymatic process. Griengl and coworkers have applied this concept to asymmetric synthesis of 1,2-amino alcohols with a threonine aldolase [24] (Figure 6.7). While the equilibrium in threonine aldolase reactions typically does not favor the synthetic direction, and the bond formation leads to nearly equal amounts of two diastereomers, coupling the aldolase reaction with a selective tyrosine decarboxylase leads to irreversible formation of aryl amino alcohols in reasonable enantiomeric excess via a dynamic kinetic asymmetric transformation. A one-pot, two-enzyme asymmetric synthesis of amino alcohols, including noradrenaline and octopamine, from readily available starting materials was developed [25]. O CO2H H

+

OH

OH

L-threonine aldolase

CO2H

NH2

NH2

PLP

L-tyrosine decarboxylase NH2

PLP

Figure 6.7 Enzymatic coupling to drive aldol reaction equilibrium in the synthetic direction

6.3 Broad Substrate Tolerance of Wild-Type Aldolases In addition to the efforts described above, it has recently been found that some wild-type aldolases have surprisingly broad substrate tolerances and are quite useful catalysts in their own right. The recently discovered fructose-6-phosphate aldolase (FSA) [26,27] naturally accepts dihydroxyacetone (DHA) instead of DHAP, taking care of one of the major targets of directed evolution efforts. It has been used for especially facile syntheses of iminocyclitols [28], and we have developed a one-pot protocol to these important glycosidase inhibitors [29] (Figure 6.8). Remarkably, it has been found that FSA has a broad substrate tolerance both with respect to acceptor substrates and donor substrates, contrary to conventional dogma. Thus, FSA utilizes O CbzHN

O

+

R2

HO

R1 R1=OH, H

R2=OH, H, CH3

O

+

HO

Figure 6.8

R1

2. Pd(OH)2/C, H2 excess diethylamine R4=OH, H, CH3

OH OH

R3

R4

R3 R3=H, CH2OH, CH2NHAc

2. Pd(OH)2/C, H2 excess diethylamine one-pot

1. FSA

O N3

R2

H N

1. FSA

H N

R4

HO OH

one-pot

Versatile one-pot synthesis of D-iminocyclitols with fructose-6-phosphate aldolase


OR3 OH

O

R4O

H

CO2H

R 2O R1

sialic acid aldolase

OR2 R1

OR3

R4O

pyruvate

117

O

OH

HO

OH O

HO R1 = OH, NHGc R2 = H or Gal or Glc R3 = H or Gal R4 = H or Gal, Glc, Man

OH OH O HO OH HO

O

OH

HO CO2H

HO HO

O

OH OH O O HO OH

OH

HO

HO HO CO2H

OH H N O

O

OH

O

OH

CO2H HO HO

HO

O

OH

HO

Figure 6.9 Broad acceptor substrate tolerance of sialic acid aldolase in synthesis of nonnatural disaccharides

hydroxyacetone and hydroxybutanone as donors equally well compared with DHA, with very similar kinetic parameters. Similarly, Chen and coworkers have recently discovered that E. coli sialic acid aldolase is much more tolerant of a variety of donor and acceptor substrates than was previously expected. They have used this catalytic plasticity to great effect in preparing novel sialosides. The native reaction catalyzed by sialic acid aldolase is the condensation of pyruvate donor with N-acetylmannosamine, forming the nine-carbon monosaccharide N-acetylneuraminic acid (sialic acid) from a six-carbon monosaccharide. Chen and coworkers have found that disaccharide analogs are tolerated as acceptors, leading to synthesis of unusual disaccharides containing sialic acid residues on the reducing end, some of which are illustrated in Figure 6.9 [30,31]. Furthermore, analogs of pyruvate can be accepted as donors as well. Specifically, fluoropyruvate is a viable substrate for synthesis of 3-fluorosialic acid, which can subsequently be enzymatically converted to the corresponding modified nucleotide sugars. These compounds are valuable as probes for studying the mechanism and inhibition of sialidases and sialyltransferases, enzymes that play a key role in many disease states, including influenza infection (Figure 6.10) [32]. HO

OH CO2H HO R

O HO

HO HO HO

R

F

O

O OH

+

F

wild-type CO2H

~ 1:1

sialic acid aldolase HO

R= NHAc, NHGc, OH

OH

OH CO2H HO R

O

OH F

HO

Figure 6.10 Broad donor substrate tolerance of sialic acid aldolase in synthesis of fluorinated sialic acid analogues

118


6.4 Conclusions The work described in this chapter illustrates that several approaches can successfully achieve the goal of broadening the substrate scope of aldolases. Whereas these enzymes have been perceived as being useful only in very specific applications due to their strict substrate specificity, it is becoming clear that they can in fact be versatile, practical biocatalysts that can be applied to a wider range of synthetic problems.

References [1] Machajewski, T.D. and Wong, C.-H. (2000) The catalytic asymmetric aldol reaction. Angewandte Chemie– International Edition, 39, 1352–1374. [2] Fessner, W.D. and Helaine, V. (2001) Biocatalytic synthesis of hydroxylated natural products using aldolases and related enzymes. Current Opinion in Biotechnology, 12, 574–586. [3] Samland, A.K. and Sprenger, G.A. (2006) Microbial aldolases as C C bonding enzymes – unknown treasures and new developments. Applied Microbiology and Biotechnology, 71, 253–264. [4] Dean, S.M., Greenberg, W.A. and Wong, C.-H. (2007) Recent advances in aldolase-catalyzed asymmetric synthesis. Advanced Synthesis and Catalysis, 349, 1308–1320. [5] Petrounia, I.P. and Arnold, F.H. (2000) Designed evolution of enzymatic properties. Current Opinion in Biotechnology, 11, 325–330. [6] Arnold, F.H. (2001) Combinatorial and computational challenges for biocatalyst design. Nature, 409, 253–257. [7] Powell, K.A., Ramer, S.W., del Cardayre, S.B. et al. (2001) Directed evolution and biocatalysis. Angewandte Chemie–International Edition, 40, 3948–3959. [8] Zhao, H.M., Chockalingam, K. and Chen, Z.L. (2002) Directed evolution of enzymes and pathways for industrial biocatalysis. Current Opinion in Biotechnology, 13, 104–110. [9] Jaeger, K.E. and Eggert, T. (2004) Enantioselective biocatalysis optimized by directed evolution. Current Opinion in Biotechnology, 15, 305–313. [10] Hsu, C.C., Hong, Z.Y., Wada, M. et al. (2005) Directed evolution of D-sialic acid aldolase to L-3-deoxy-manno-2octulosonic acid (L-KDO) aldolase. Proceedings of the National Academy of Sciences of the United States of America, 102, 9122–9126. [11] Williams, G.J., Domann, S., Nelson, A. and Berry, A. (2003) Modifying the stereochemistry of an enzymecatalyzed reaction by directed evolution. Proceedings of the National Academy of Sciences of the United States of America, 100, 3143–3148. [12] Fong, S., Machajewski, T.D., Mak, C.C. and Wong, C.-H. (2000) Directed evolution of D-2-keto-3-deoxy-6phosphogluconate aldolase to new variants for the efficient synthesis of D- and L-sugars. Chemistry & Biology, 7, 873–883. [13] DeSantis, G., Liu, J.J., Clark, D.P. et al. (2003) Structure-based mutagenesis approaches toward expanding the substrate specificity of D-2-deoxyribose-5-phosphate aldolase. Bioorganic and Medicinal Chemistry, 11, 43–52. [14] Liu, J.J., Hsu, C.C. and Wong, C.-H. (2004) Sequential aldol condensation catalyzed by DERA mutant Ser238Asp and a formal total synthesis of atorvastatin. Tetrahedron Letters, 45, 2439–2441. [15] Woodhall, T., Williams, G., Berry, A. and Nelson, A. (2005) Creation of a tailored aldolase for the parallel synthesis of sialic acid mimetics. Angewandte Chemie–International Edition, 44, 2109–2112. [16] Williams, G.J., Woodhall, T., Farnsworth, L.M. et al. (2006) Creation of a pair of stereochemically complementary biocatalysts. Journal of the American Chemical Society, 128, 16238–16247. [17] Cheriyan, M., Toone, E.J. and Fierke, C.A. (2007) Mutagenesis of the phosphate-binding pocket of KDPG aldolase enhances selectivity for hydrophobic substrates. Protein Science: A Publication of the Protein Society, 16, 2368–2377. [18] Jennewein, S., Schuermann, M., Wolberg, M. et al. (2006) Directed evolution of an industrial biocatalyst: 2-deoxy-D-ribose 5-phosphate aldolase. Biotechnology Journal, 1, 537–548. [19] Ran, N.Q., Draths, K.M. and Frost, J.W. (2004) Creation of a shikimate pathway variant. Journal of the American Chemical Society, 126, 6856–6857.


119

[20] Ran, N.Q. and Frost, J.W. (2007) Directed evolution of 2-keto-3-deoxy-6-phosphogalactonate aldolase to replace 3-deoxy-D-arabino-heptulosonic acid 7-phosphate synthase. Journal of the American Chemical Society, 129, 6130–6139. [21] Greenberg, W.A., Varvak, A., Hanson, S.R. et al. (2005) Development of an efficient, scalable, aldolase-catalyzed process for enantioselective synthesis of statin intermediates. Proceedings of the National Academy of Sciences of the United States of America, 101, 5788–5793. [22] Lamble, H.J., Danson, M.J., Hough, D.W. and Bull, S.D. (2005) Engineering stereocontrol into an aldolasecatalysed reaction. Chem. Commun., 124–126. [23] Sugiyama, M., Hong, Z.Y., Whalen, L.J. et al. (2006) Advanced Synthesis and Catalysis, 348, 2555–2559. [24] Steinreiber, J., Schurmann, M., Wolberg, M. et al. (2007) Overcoming thermodynamic and kinetic limitations of aldolase-catalyzed reactions by applying multienzymatic dynamic kinetic asymmetric transformations. Angewandte Chemie–International Edition, 46, 1624–1626. [25] Steinreiber, J., Schurmann, M., van Assema, F. et al. (2007) Synthesis of aromatic 1,2-amino alcohols utilizing a bienzymatic dynamic kinetic asymmetric transformation. Advanced Synthesis and Catalysis, 349, 1379–1386. [26] Schurmann, R. and Sprenger, G.A. (2001) Fructose-6-phosphate aldolase is a novel class I aldolase from Escherichia coli and is related to a novel group of bacterial transaldolases. The Journal of Biological Chemistry, 276, 11055–11061. [27] Schurmann, M., Schurmann, M. and Sprenger, G.A. (2002) Fructose 6-phosphate aldolase and 1-deoxy-Dxylulose 5-phosphate synthase from Escherichia coli as tools in enzymatic synthesis of 1-deoxysugars. Journal of Molecular Catalysis B, Enzymatic, 19, 247–252. [28] Castillo, J.A., Calveras, J., Casas, J. et al. (2006) Fructose-6-phosphate aldolase in organic synthesis: preparation of D-fagomine, N-alkylated derivatives, and preliminary biological assays. Organic Letters, 8, 6067–6070. [29] Sugiyama, M., Hong, Z.Y., Liang, P.H. et al. (2007) D-Fructose-6-phosphate aldolase-catalyzed one-pot synthesis of iminocyclitols. Journal of the American Chemical Society, 129, 14811–14817. [30] Yu, H. and Chen, X. (2006) Aldolase-catalyzed synthesis of beta-D-Gal-(1-9)-D-KDN: a novel acceptor for sialyltransferases. Organic Letters, 8, 2393–2396. [31] Huang, S.S., Yu, H. and Chen, X. (2007) Disaccharides as sialic acid aldolase substrates: synthesis of disaccharides containing a sialic acid at the reducing end. Angewandte Chemie–International Edition, 46, 2249–2253. [32] Chokhawala, H.A., Cao, H.Z., Yu, H. and Chen, X. (2007) Enzymatic synthesis of fluorinated mechanistic probes for sialidases and sialyltransferases. Journal of the American Chemical Society, 129, 10630–10631.

7 Synthetic Applications of Ketoreductases and Alcohol Oxidases Dunming Zhu1,2 and Ling Hua1,3 1

Department of Chemistry, Southern Methodist University, Dallas, TX 75275, USA Tianjin Industrial Biotechnology Research and Development Center, Chinese Academy of Sciences, PR China 3 China Research Center, Genencor International, a Danisco Division, Shanghai, PR China 2

Ketoreductases catalyze the reversible reduction of ketones and oxidation of alcohols using cofactor NADH/NADPH as the reductant or NAD þ /NADP þ as oxidant. Alcohol oxidases catalyze the oxidation of alcohols with dioxygen as the oxidant. Both categories of enzymes belong to the oxidoreductase family. In this chapter, the recent advances in the synthetic application of these two categories of enzymes are described.

7.1 Ketoreductases Enantiometrically pure alcohols are important and valuable intermediates in the synthesis of pharmaceuticals and other fine chemicals. A variety of synthetic methods have been developed to obtain optically pure alcohols. Among these methods, a straightforward approach is the reduction of prochiral ketones to chiral alcohols. In this context, varieties of chiral metal complexes have been developed as catalysts in asymmetric ketone reductions [1–3]. However, in many cases, difficulties remain in the process operation, and in obtaining sufficient enantiomeric purity and productivity [2,3]. In addition, residual metal in the products originating from the metal catalyst presents another challenge because of the ever more stringent regulatory restrictions on the level of metals allowed in pharmaceutical products [4]. An alternative to the chemical asymmetric reduction processes is biocatalytic transformation, which offers


122


advantages such as mild and environmentally benign reaction conditions, high chemo-, regioand stereo-selectivity, and void of residual metal in the products [5–7]. Therefore, great efforts have been made to develop biocatalytic ketone reduction processes in recent years [8,9]. Biocatalytic transformations can be achieved by three approaches: wild-type whole-cell microorganism, recombinant whole cell overexpressing catalytic enzymes, and isolated enzyme. The recent progress in the synthetic application of ketoreductases is described and organized according to these three approaches.

7.1.1 Wild-Type Whole-Cell Biocatalysts The major advantage of using whole cells as catalysts for the asymmetric reduction of ketones is the in vivo recycling of the redox co-enzymes (NADH or NADPH). In many cases, whole-cell biocatalysts are more stable because they offer a natural environment for the enzyme. In addition, there is no need for downstream processing and purification of enzymes. Therefore, use of whole-cell biocatalysts has still attracted considerable interest. Saccharomyces cerevisiae (bakers yeast), a readily available and cheap biocatalyst, has been the most extensively studied microorganism in biocatalytic ketone reductions [10,11]. Recently, a taxomonic screening of 421 microorganisms carried out by Sinisterra and coworkers led to the discovery of Yarrowia lipolytica CECT1240 and Pichia mexicana CECT11015, which catalyzed the reductions of a-chloroketones to give greater than 90% enantiomeric excess (ee) and higher than 85% yield for the (R)- and (S)-halohydrins respectively (Figure 7.1). These chiral halohydrins are key precursors for the synthesis of adrenergic b-blockers [12]. They also identified a fungus, Diplogelasinospora glovesii IMI 171 018, with very high activity and selectivity toward the reduction of cyclic ketones [13]. Patel and coworkers reported that microbial reduction of ethyl 1-benzyl-3-oxopiperidine-4carboxylate by Candida parapsilosis SC16 347 gave ethyl cis-(3R,4R)-1-benzyl-3R-hydroxypiperidine-4R-carboxylate as the major product in 97.4% diastereomeric excess (de) and 99.8% ee (Figure 7.2), while 99.5% de and 98.2% ee were achieved in the reduction catalyzed by Pichia methanolica SC16 415 [14]. The enantioselective reduction of a-ketoester was carried out using the cell suspension of Aureobasidium pullulans SC13 849 to produce the corresponding (R)-alcohol in 94% isolated O Cl

OH O

Cl

Ar

(R)

OH O

Cl

or

Ar

(S)

O

Ar

Figure 7.1 The reductions of a-chloroketones catalyzed by Yarrowia lipolytica CECT1240 and Pichia mexicana CECT11015 O

O

O O

N Bn

O OH

C. parapsilosis N Bn

Figure 7.2 Synthesis of ethyl cis-(3R,4R)-1-benzyl-3R-hydroxypiperidine-4R-carboxylate via microbial reduction

123

Synthetic Applications of Ketoreductases and Alcohol Oxidases O

OH O

O A. pullulans

O

O

OH

H N

F

O

O OH

Figure 7.3

Biocatalytic synthesis of the intermediate of a retinoic acid receptor gamma-specific agonist

yield and with 97% ee, which is the intermediate in the synthesis of the retinoic acid receptor gamma-specific agonist (R)-3-fluoro-4-[[hydroxy(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2naphthalenyl)acetyl]amino]benzoic acid (Figure 7.3) [15]. In a study aim to develop biocatalytic process for the synthesis of Kaneka alcohol, a potential intermediate for the synthesis of HMG-CoA reductase inhibitors, cell suspensions of Acinetobacter sp. SC13 874 was found to reduce diketo ethyl ester to give the desired syn-(3R,5S)dihydroxy ester with an ee of 99% and a de of 63% (Figure 7.4). When the tert-butyl ester was used as the starting material, a mixture of mono- and di-hydroxy esters was obtained with the dihydroxy ester showing an ee of 87% and de of 51% for the desired syn-(3R,5S)-dihydroxy ester [16]. Three different ketoreductases were purified from this strain. Reductase I only catalyzes the reduction of diketo ester to its monohydroxy products, whereas reductase II catalyzes the formation of dihydroxy products from monohydroxy substrates. A third reductase (III) catalyzes the reduction of diketo ester to syn-(3R,5S)-dihydroxy ester.

O

O

O

HO

O

R = Et, t-Bu

R

Kaneka alcohol OH O O O O

O

O O

O

(S)

OH O O

R

O

R

O

R

O

R

O

O

(R)

O

R

R O O

R = Et, t-Bu

OH O (R)

O O

OH OH O O

(S)

(R)

Figure 7.4

(R)

(R)

O

R

OH OH O O

OH OH O O

OH O (S)

(R)

(S)

O

R

OH OH O O

Microbial reduction of 3,5-diketoesters

(S)

(S)

O

R

124


Similarly, whole-cell Lactobacillus kefir DSM 20 587, which possesses two alcohol dehydrogenases for both asymmetric reduction steps, was applied in the reduction of tert-butyl 6-chloro-3,5-dioxohexanoate for asymmetric synthesis of tert-butyl-(3R,5S)-6chloro-dihydroxyhexanoate (Figure 7.5), a chiral building block for the HMG-CoA reductase inhibitor [17]. A final product concentration of 120 mM and a specific product capacity of 2.4 mmol per gram dry cell were achieved in an optimized fed-batch process. A de > 99% was obtained for (3R,5S)- and (3S,5S)-tert-butyl-6-chloro-dihydroxyhexanoate with the space– time yield being 4.7 mmol L1 h1. O

O

Cl

O O

R

Lactobacillus kefir

OH OH O Cl

(S)

(R)

O

R

Figure 7.5 Reduction of tert-butyl 6-chloro-3,5-dioxohexanoate by whole-cell Lactobacillus kefir DSM 20587

The enantio- and diastereo-selective reduction of 2,5-hexanedione has also been achieved using the resting whole cells of Lactobacillus kefir DSM 20 587 and the product (2R,5R)hexanediol was obtained with ee > 99% and de > 99% (Figure 7.6) [18]. The optimal reaction conditions were at pH 6, 30 C and with equal molar amounts of substrate and co-substrate (glucose). A continuous process for the production of (2R,5R)-hexanediol was developed with a typical time–space yield over 5 days being 64 g L1 day1 [19]. When the immobilized whole cells of Lactobacillus kefir DSM 20 587 on sodium cellulose sulfate were used as biocatalyst, the selective reduction of 2,5-hexanedione to (5R)-hydroxyhexane-2-one ( > 99% ee) was realized in the plug flow reactor (Figure 7.6) [20]. The selectivity and time–space yield for the production of (5R)-hydroxyhexane-2-one were 95% and 87 g L1 day1 respectively, and the biocatalyst remained active (68% residual activity) after 6 days of operation. (2S,5S)-Hexanediol could be biocatalytically synthesized with whole cells of Saccharomyces cerevisiae and glucose for the internal cofactor regeneration (Figure 7.6) [11]. On a preparative scale with 25 mmol of hexanedione, the substrate was completely converted to (2S,5S)-hexanediol with ee > 99% and de 96%. An isolated yield of 75% was obtained after purification. OH

O

S. cerevisiae

L. kefir OH

O

de >99%, ee >99%

Immobilized L. kefir

OH

OH de 96%, ee .99%

OH

O ee >99%

Figure 7.6

Biocatalytic synthesis of chiral hexanediol and hydroxyhexane-2-one

Lactobacillus kefir was also employed as the whole-cell biocatalyst for the asymmetric reduction of ethyl 4-chloroacetoacetate to ethyl (S)-4-chloro-3-hydroxybutanoate, the chiral

Synthetic Applications of Ketoreductases and Alcohol Oxidases

125

synthon of the number one cholesterol-lowering drug (Lipitor) [21]. The reduction was carried out using 2-propanol as co-substrate and the product (S)-alcohol was produced with a final yield of 97% and ee 99.5% within 14 h. A high space–time yield of 85.7 mmol L1 h1 and a high specific product capacity of 24 mmol per gram DCW were achieved. After the screening study, Kroutil and coworkers found that the lyophilized cells of Rhodococcus ruber DSM 44 541 catalyzed the asymmetric hydrogen transfer reduction of various ketones using 2-propanol as hydrogen donor and enantioselective oxidation of the racemic alcohol using acetone as hydrogen acceptor. Chiral sec-alcohols of opposite absolute configuration were obtained employing this microbial reduction–oxidation system and 99% ee could be reached in many cases, including alkyl-, aryl-, and hetereoaryl-substituted 2ethanones [22–25]. Recently, the application of lyophilized cells of Rhodococcus ruber DSM 44 541 has been extended to the regio- and stereo-selective reduction of diketones and oxidation of diols [26]. a,b-Unsaturated ketones (enones) were chemo-selectively reduced by employing whole lyophilized cells of Rhodococcus ruber DSM 44 541 to furnish the corresponding allylic alcohols with up to > 99% ee. No kinetic resolution of racemic ketones with a chiral carbon at the g-position of the ketone moiety was observed, since it was too distant from the reaction center to induce any significant diastereoselectivity [27].

7.1.2 Recombinant Whole-Cell Biocatalysts Overexpressing Catalytic Enzymes When wild-type whole cells are used as biocatalysts, limited amounts of active enzymes are available for the reduction, often leading to low reaction rate and low product concentration. In addition, a whole cell may contain more than one ketoreductase, frequently with opposing stereoselectivity. Therefore, not all wild-type whole-cell-mediated ketone reductions provide product chiral alcohols in high optical purity. Two strategies, one carrying out the reduction with a recombinant whole cell overexpressing a ketoreductase and another with isolated ketoreductase, have been used to solve these problems. The recent advances in the strategy of using a recombinant whole cell as biocatalyst for the synthesis of chiral alcohols are reviewed in this section and the use of isolated ketoreductase will be the topic of next section. While whole cells of bakers yeast (Saccharomyces cerevisiae) are a convenient biocatalyst for the reduction of a wide variety of carbonyl compounds, mixtures of stereoisomeric alcohols are often observed because the organism contains a large number of ketoreductase enzymes with overlapping substrate specificities but differing stereoselectivities. By using recombinant DNA techniques, three enzymes (fatty acid synthase, Fasp; aldo-keto reductase, Ypr1p; Racetoxy ketone reductase, Gre2p) from yeast were either knock out or overexpressed. The ‘second-generation’ yeast strains were created by combining gene knockout and overexpression in single strains. The genetically modified yeasts significantly improved the stereoselectivities of b-ketoester reductions [28]. Another approach in this strategy is the overexpression of a ketoreductase in a host cell line (usually Escherichia coli) that lacks the enzyme with overlapping substrate specificity [29]. Owing to the higher concentration of active ketoreductase in the recombinant cells, the internal cofactor regeneration is usually not efficient enough. Thus, the desired ketoreductase is often co-expressed with cofactor regenerating enzyme in host cells. For example, a Saccharomyces cerevisiae (bakers yeast) carbonyl reductase (SCR) gene Gre2 (YOL151w) and a glucose dehydrogenase (GDH) gene from Bacillus megaterium NBRC (formerly IFO) 15 308 were

126


co-expressed in Escherichia coli BL21(DE3) in two types of mode [30]. One is the tandem system, where the genes encoding SCR and GDH are located in the same plasmid, and the other is the two-plasmid system, where each of the SCR and GDH genes is located in separate plasmids that can coexist in one Escherichia coli cell. These recombinant Escherichia coli cells co-producing SCR and GDH were used as biocatalyst in the reduction of ketones with diverse structures, and 11 out of 20 resulting chiral alcohols had enantiomeric purities > 98% ee. When the two co-expression systems were compared in terms of the conversion of 2,4-octanedione, the two-plasmid system showed better performance, giving (S)-2-hydroxy-4-octanone with complete enantioselectivity and regioselectivity in 71% isolated yield, in which the space productivity reached 41 g L1. This synthetically useful whole-cell biocatalyst was recently employed in the synthesis of methyl (R)-o-chloromandelate, an intermediate for a platelet aggregation inhibitor named clopidogrel, with excellent enantioselectivity ( > 99% ee) and space productivity (178 g L1) (Figure 7.7) [31]. Cl

O

Cl CO2CH3

OH

SCR NADPH

Gluconic acid

Cl

CO2CH3

CO2CH3

N

S

NADP +

Clopidogrel

Glucose

GDH

Recombinant E. coli

Figure 7.7

Biocatalytic synthesis of methyl (R)-o-chloromandelate

Both the carbonyl reductase (S1) gene from Candida magnoliae and the GDH gene from Bacillus megaterium were co-expressed in Escherichia coli HB101 cells and the recombinant cells were used as biocatalyst for the asymmetric reduction of ethyl 4-chloro-3-oxobutanoate (COBE) to ethyl (S)-4-chloro-3-hydroxybutanoate (S)-CHBE) [32]. In an organic-solvent– water two-phase system, (S)-CHBE formed in the organic phase amounted to 2.58 M (430 g L1) with the molar yield being 85%. In an aqueous mono-phase system with continuous feeding of substrate COBE, (S)-CHBE accumulated up to 1.25 M (208 g g L1). In this case, the calculated turnover of CHBE to NADPH was 21 600. The optical purity of the product (S)-CHBE was 100% ee in both systems. The aqueous system used for the reduction reaction involving Escherichia coli cells carrying a plasmid containing the S1 and GDH genes as a catalyst is simple and highly advantageous for the practical synthesis of optically pure (S)-CHBE. The 4-bromo and 4-hydroxy analogues of (S)-CHBE were also synthesized using the recombinant Escherichia coli cells co-producing the carbonyl reductase (S1) from Candida magnoliae and the GDH from Bacillus megaterium in over 90% isolated yields and 99% ee (Figure 7.8) [33]. O X

O

OH O S1

CO2C2H5 NADPH

Gluconic acid

X NADP +

GDH

Recombinant E. coli

Figure 7.8

Glucose

CO2C2H5 X Yield(%) Cl 95 Br 95 OH 90

ee(%) 100 99 99

Biocatalytic synthesis of optically pure (S)-CHBE and the 4-bromo and 4-hydroxy analogues

127


The carbonyl reductase gene from Candida magnoliae was also co-expressed with the GDH gene from Bacillus subtilis in the methylotrophic yeast strain Pichia pastoris GC909. The reduction of ethyl 4-chloroacetoacetate using this recombinant strain was performed in a twophase reaction system with butyl acetate as organic solvent. Complete conversion of the substrate (350 mM) was achieved by continuous feeding of substrate and cell suspension, with an isolated yield of 91% and 95% ee for ethyl (S)-CHBE [34]. The asymmetric reduction of ethyl 4-chloro-3-oxobutanoate catalyzed by recombinant Escherichia coli cells, in which an NADPH-dependent menadione reductase gene of Candida macedoniensis AKU4588 and a GDH gene were co-expressed, afforded (S)-CHBE in 92% yield and with 92% ee. The space productivity reached 1680 mM (281 mg mL1) and the calculated turnover number of NADP þ was 12 900 [35]. The synthesis of ethyl (R)-4-chloro-3-hydroxybutanoate (R)-ECHB) from ethyl 4-chloroacetoacetate was achieved using recombinant whole cells of Escherichia coli expressing a secondary alcohol dehydrogenase of Candida parasilosis [36]. Cofactor NADH was regenerated using 2-propanol as the hydrogen donor. The space productivity of (R)-ECHB reached 36.6 g L1 ( > 99% ee, 95.2% conversion) without addition of NADH to the reaction mixture. From the screening of 119 microbial cultures for reduction of a-chloro-30 -chloroacetophenone to (S)-2-chloro-1-(30 -chlorophenyl)-ethanol, Hansenula polymorpha ATCC 58 401 and Rhodococcus globerulus ATCC 21 505 were found to have the highest enantioselectivity for producing the desired alcohol with 73.8% ee and 71.8% ee respectively. A ketoreductase from Hansenula polymorpha was identified to give the (S)-alcohol with 100% ee (Figure 7.9), indicating that the low enantioselectivity using wild-type whole cell might result from the coexistence of more than one ketoreductase in Hansenula polymorpha. The cloned ketoreductase was expressed in Escherichia coli together with a cloned glucose 6-phosphate dehydrogenase from Saccharomyces cerevisiae to allow regeneration of the NADPH required by the ketoreductase. An extract or intact cells of Escherichia coli containing the two recombinant enzymes was used to reduce 2-chloro-1-(30 -chloro-40 -fluorophenyl)-ethanone to optically pure (S)-2-chloro-1-(30 -chloro-40 -fluorophenyl)ethanol in 91% and 89% yield respectively [37]. These (S)-alcohols are the intermediates for the synthesis of some leading anticancer candidate compounds. OH

O Cl Whole-cell H. polymorpha

Cl 73.8% ee

OH Cl Enzyme H. polymorpha

Cl

Cl

Cl 100% ee

Figure 7.9 Reduction of a-chloro-30 -chloroacetophenone catalyzed by a whole-cell biocatalyst and the corresponding purified enzyme

An (S)-specific alcohol dehydrogenase gene from Rhodococcus erythropolis and GDH from Bacillus subtilis were ligated into one plasmid, which was expressed in Escherichia coli strain DSM14 459 to provide an (S)-selective whole-cell catalyst. An (R)-selective counterpart was constructed in the same manner by using Escherichia coli DSM14 459 as host organism, but in this case two separate plasmids were used, which contain

128

Biocatalysis for the Pharmaceutical Industry O

OH (R)-selective "designer cells"

X

X X = F, 87% yield, >99% ee; X = Cl, 91% yield, >99% ee

Figure 7.10

Reduction of acetophenones with ‘designer cells’

genes encoding an (R)-specific alcohol dehydrogenase from Lactobacillus kefir or a GDH from Thermoplasma acidophilum. These ‘designer cells’ could reduce a wide range of ketones at high substrate concentration and without the addition of an ‘external’ cofactor in aqueous reaction medium. To demonstrate the practical applicability, the reduction of 40 -chloroacetophenone was carried out on 10 L scale using (R)-selective ‘designer cells’ and (R)-1-(40 chlorophenyl)ethanol was obtained in 91% yield and with > 99% ee [38]. The reduction of 40 fluoroacetophenone on about 25 mmol scale using (R)-selective ‘designer cells’ produced (R)1-(40 -fluorophenyl)ethanol in 87% yield and with > 99% ee (Figure 7.10). This recombinant whole-cell biocatalyst showed somewhat lower activity toward 20 - and 30 -fluoroacetophenones [39]. 1-Bromo-2-octanone and 1-chloro-2-octanone were also reduced at high substrate concentration up to 208 g L1 using the recombinant Escherichia coli cells bearing an (R)specific alcohol dehydrogenase from Lactobacillus kefir and a GDH from Thermoplasma acidophilum as catalyst to afford the corresponding (S)-alcohols with conversion being > 95% and ee > 99%. In a gram-scale process, (S)-1-bromo-2-octanol was isolated in 85% yield and > 99% ee, and then quantitatively converted to the (S)-epoxide (Figure 7.11) [40]. O X

(R)-selective "designer cells"

OH X

X = Cl, Br, >95% conversion, >99% ee O

Figure 7.11 Biocatalytic and chemical synthesis of epoxide

The reduction of several ketones, which were transformed by the wild-type lyophilized cells of Rhodococcus ruber DSM 44 541 with moderate stereoselectivity, was reinvestigated employing lyophilized cells of Escherichia coli containing the overexpressed alcohol dehydrogenase (ADH-‘A’) from Rhodococcus ruber DSM 44 541. The recombinant whole-cell biocatalyst significantly increased the activity and enantioselectivity [41]. For example, the enantiomeric excess of (R)-2-chloro-1-phenylethanol increased from 43 to > 99%. This study clearly demonstrated the advantages of the recombinant whole cell biocatalysts over the wildtype whole cells. Besides GDH, formate dehydrogenase (FDH) was co-expressed with alcohol dehydrogenase for efficient NADH regeneration in the whole-cell bioreduction of ketones. For example, an NAD þ -dependent FDH gene from Mycobacterium vaccae N10 was co-expressed together with an alcohol dehydrogenase gene from Lactobacillus brevis in Escherichia coli cells, and the recombinant cells were used for the asymmetric reduction of methyl acetoacetate [42].

129

Synthetic Applications of Ketoreductases and Alcohol Oxidases HO O

OH OH

OH

OH CO2

OH OH MDH NADH

HO

NAD+

FDH

OH OH OH

HCO2-

Recombinant E. coli

Figure 7.12

Biocatalytic transformation of D-fructose to D-mannitol

Whole resting cells of this strain catalyzed the formation of methyl (R)-3-hydroxy butanoate from methyl acetoacetate at 40 mM concentration. The product yield was 100 mol% at a productivity of 200 mmol g1 min1 (cell dry weight). The same FDH gene was also coexpressed with a mannitol dehydrogenase gene of Leuconostoc pseudomesenteroides ATCC 12 291 in Escherichia coli cells [43]. The recombinant Escherichia coli cells were able to transform D-fructose to D-mannitol with concentration of up to 15 mM (Figure 7.12). The introduction of another gene, encoding the glucose facilitator protein of Zymomonas mobilis (GLF), allowed the cells to take up D-fructose efficiently, without simultaneous phosphorylation. Resting cells of this Escherichia coli strain (3 g L1 cell dry weight) produced D-mannitol, with product concentration reaching 216 mM in 17 h. The product D-mannitol concentration reached 362 mM (84% yield) within 8 h when the pH of the reaction mixture was controlled by addition of formic acid.

7.1.3 Isolated Enzyme The use of isolated enzymes as biocatalysts often offers some advantages over natural wholecell biocatalysts. The presence of only one enzyme in the reaction mixture prevents the occurrence of side reactions, thus enhancing the chemo-, regio- and stereo-selectivity and simplifying the downstream processing and purification of the products. The enzymes can be stored and used as normal chemical reagents and no microbiological knowledge is required. In addition, advances in molecular biotechnology have enabled the availability of isolated enzymes in large quantity. As a result, the application of isolated ketoreductases in the synthesis of optically pure chiral alcohols has been rapidly expanded in the last few years. Bakers yeast (Saccharomyces cerevisiae) is a convenient biocatalyst for the reduction of a wide variety of carbonyl compounds and contains a large number of ketoreductase enzymes. Analysis of the complete yeast genome revealed that the protein products of about 50 open reading frames might catalyze ketone reductions [44]. With the aim to avoid the mutual interference of multiple enzymes with divergent enantio- and diastereomeric preference and to discover synthetically useful biocatalysts, Stewart and coworkers cloned 19 dehydrogenases from bakers yeast (Saccharomyces cerevisiae) and expressed as fusion proteins with glutathione (S)-transferase [45]. A representative set of a- and b-ketoesters were tested as substrates (total 11) for each purified fusion protein. The stereoselectivities of b-ketoester reductions depended both on the identity of the enzyme and the substrate structure, and some reductases were identified to yield either L- or D-alcohols with high stereoselectivities. While a-keto esters were generally reduced with lower enantioselectivities, it was possible in all but one case to identify pairs of yeast reductases that delivered both alcohol antipodes in optically pure form. In preparative scale, optically pure (S)-ethyl-3-hydroxybutyrate was produced with a

130


Baker's yeast reductase

CO2C2H5

O

OH

OH CO2C2H5

OH Baker's yeast reductase

CO2C2H5

OH CO2C2H5

CO2C2H5 +

O

OH CO2C2H5

Figure 7.13

CO2C2H5

+

Baker's yeast reductase

CO2C2H5 +

OH CO2C2H5

Bioreduction of cyclic b-keto esters by bakers yeast reductases

space– time yield of 2.0 g L1 h1 and a final product titer of 16 g L1 using an Escherichia coli strain that overexpressed the yeast YOL151w protein. These yeast reductases were tested for their ability to reduce three homologous cyclic b-ketoesters. The majority of dehydrogenases reduced ethyl 2-oxo-cyclopentanecarboxylate, yielding a pair of diastereomeric alcohols with consistent (2R)-stereochemistry. Reduction of ethyl 2-oxo-cyclohexanecarboxylate afforded only cis-alcohol. Two enzymes in the collection catalyzed the reduction of ethyl 2-oxo-cycloheptanecarboxylate to yield mainly the cis(1R,2S)-alcohol (Figure 7.13). Escherichia coli cells overexpressing the YDL124w gene were used in a dynamic kinetic resolution of ethyl 2-oxo-cyclohexanecarboxylate, and the reduction product was subsequently transformed to (1R,2S)-2-methyl-1-cyclohexanol, an important chiral building block (Figure 7.14) [46]. O

O

OH O CO2C2H5

Figure 7.14

YDL124w

OH CO2C2H5

Biocatalytic and chemical synthesis of (1R,2S)-2-methyl-1-cyclohexanol

Homochiral 3-hydroxy-4-substituted b-lactams serve as precursors to the corresponding a-hydroxy-b-amino acids – key components of many biologically and therapeutically important compounds. To develop a short synthetic sequence for these targets using a biocatalytic reduction to install the desired chirality, 19 individual bakers yeast reductases were screened for their efficiency and enantioselectivity with respect to 3-oxo-4-phenyl-b-lactam. Although these reductases readily accepted 3-oxo-4-phenyl-b-lactam as a substrate, the stereoselectivity was moderate. Four of the most selective enzymes (Yjr096w, Ydl124w, Ybr149w, and Ycr107w) belong to the aldose reductase (AKR) superfamily, with cis-alcohol being the major product. The Ybr149w and Ycr107w proteins gave primarily (3S,4R)-alcohol with 94% ee and 87% ee respectively, while the Yjr096w and Ydl124w reductases predominantly produced (3R,4S)alcohol with 67% ee and 84% ee respectively. The modeled three-dimensional structures of these enzymes provide a rationale for this behavior and suggest strategies for their further improvement [47].

131


A series of a-chloro-b-ketoesters were also reduced by these individual reductases from bakers yeast (Saccharomyces cerevisiae). In nearly all cases, it was possible to produce at least two of the four possible a-chloro-b-hydroxy ester diastereomers with high optical purities. The utility of this approach was demonstrated by reducing ethyl 2-chloroacetoacetate to the corresponding syn-(2R,3S)-alcohol on a multigram scale using whole cells of an Escherichia coli strain overexpressing YOR120w reductase [48]. The aldose reductase YDL124w catalyzed the reduction of ethyl 2-chloro-3-oxo-3-phenylpropionate to produce syn-(2S,3R)-ethyl2-chloro-3-hydroxy-3-phenylpropionate as the only detectable product, whereas short-chain dehydrogenase YGL039w afforded a 9:1 mixture of syn-(2R,3S)- and anti-(2S,3S)-alcohols. The isolated syn-(2S,3R)-ethyl-2-chloro-3-hydroxy-3-phenylpropionate and the syn-(2R,3S)enantiomer were used as precursors for the antipodes of the N-benzoyl phenylisoserine Taxol side chain (Figure 7.15) [49]. After base-mediated ring closure of the chlorohydrin enantiomers, the epoxides were converted directly to the oxazoline form of the target molecules using a Ritter reaction with benzonitrile. These were hydrolyzed to the ethyl ester form of the Taxol side-chain enantiomers under acidic conditions. OH O

C6H5COHN

O

OC2H5 O

O

YDL124w 91%

OC2H5

Cl

OH

>98% de, >98% ee

OC2H5 Cl

85% YGL039w

C6H5COHN

OH O

O OC2H5

OC2H5 Cl

OH

80% de >98% de

Figure 7.15

Synthesis of the antipodes of N-benzoyl phenylisoserine Taxol side chain

Three enzymes in the bakers yeast reductase collection catalyzed the stereoselective reduction of ethyl 2-chloro-3-oxo-4-phenylbutyrate. The short-chain dehydrogenases YGL039w and YGL157w produced the (2S,3S)-chlorohydrin in 41% ee and > 98% ee respectively, while aldose reductase YDR368w afforded the (2R,3S)-diastereomer in > 98% ee and > 98% de [48]. The reduction with aldose reductase YDR368w was scaled up to 1 L and the isolated (2R,3S)-ethyl-2-chloro-3-hydroxy-4-phenylbutyrate was employed as the key intermediate for a total synthesis of the a-hydroxy-b-amino acid moiety of ()bestatin (Figure 7.16). The reduction product was cyclized to a glycidic ester that was opened in a Ritter reaction with benzonitrile, affording a trans-oxazoline, which was hydrolyzed under acidic conditions to the target molecule [50]. These studies have demonstrated the synthetic utility of these individual bakers yeast reductases. It has been reported that bioreduction of bicyclo[2.2.2]octane-2,5-dione and bicyclo[2.2.2] oct-7-ene-2,5-dione by genetically engineered yeast, which overexpressed YMR226c gene, resulted in (1R,4R,5S)-5-hydroxybicyclo[2.2.2]octan-2-one ( > 99% ee) and (1S,4S,5S)-5hydroxybicyclo[2.2.2]octan-2-one (98% ee) and in (1R,4R,5S)-5-hydroxybicyclo[2.2.2]oct7-en-2-one ( > 99% ee) and (1S,4S,5S)-5-hydroxybicyclo[2.2.2]oct-7-en-2-one (93% ee)

132


O

O

YDR368w OC2H5

OC2H5 Cl >98%de, >98% ee

Cl

H2N

H2N

O

OH

OH O

N H

O OH

CO2H OH

(-)-Bestatin

Figure 7.16

Enzymatic synthesis of the a-hydroxy-b-amino acid moiety of ()-bestatin

respectively [51,52]. This indicated that the alcohol dehydrogenase encoded by gene YMR226c might have great potential in the synthesis of chiral building blocks, stimulating the further studies on the activity and enantioselectivity of this recombinant enzyme toward the reduction of various structurally diverse ketones [53]. This bakers yeast reductase showed a broad substrate preference and effectively catalyzed enantioselective reductions of arylsubstituted acetophenones, a-chloroacetophenone derivatives, aliphatic ketones, and a- and b-ketoesters. In particular, the acetophenone derivatives, aromatic a-ketoesters, and some of a-chloroacetophenone derivatives and aliphatic ketones were reduced to the corresponding chiral alcohols with excellent enantioselectivity. The enantiopreference of this enzyme generally followed Prelogs rule for the aliphatic and aromatic ketones, while both the ester functionality and steric factor were important in determining the enzymes enantiopreference for the reduction of a- and b-ketoesters. Therefore, the alcohol dehydrogenase (YMR226c) is a versatile biocatalyst for asymmetric reductions of a variety of ketones. Two interesting yeast carbonyl reductases, one from Candida magnoliae (CMCR) [33,54] and the other from Sporobolomyces salmonicolor (SSCR) [55], were found to catalyze the reduction of ethyl 4-chloro-3-oxobutanoate to give ethyl (S)-4-chloro-3-hydroxybutanoate, a useful chiral building block. In an effort to search for carbonyl reductases with anti-Prelog enantioselectivity, the activity and enantioselectivity of CMCR and SSCR have been evaluated toward the reduction of various ketones, including a- and b-ketoesters, and their application potential in the synthesis of pharmaceutically important chiral alcohol intermediates have been explored [56–58]. The carbonyl reductase from Candida magnoliae catalyzed the enantioselective reduction of a diversity of ketones, including aliphatic and aromatic ketones and a- and b-ketoesters (Figure 7.17), to anti-Prelog configurated alcohols in excellent optical purity (99% ee or higher) [56]. The usefulness of the carbonyl reductase from Candida magnoliae as an enzyme catalyst in the synthesis of chiral alcohol intermediates has been demonstrated by carrying out the reduction of several ketones on a preparative scale [56]. The isolated yields and enantiomeric excess of the product alcohols are summarized in Table 7.1, from which it can be seen that these chiral alcohols were obtained in essentially optically pure forms in excellent yields. These chiral alcohols are important intermediates in the synthesis of pharmaceuticals and agrichemicals. For example, optically active 2-hydroxy-3-methylbutyrate is an important chiral synthon

O R

O

O

O R

O

R

O

R = CH3 CH2CH3 CH(CH3)2 CF3 CH2Cl

O

X

O

R'

R

R = CH(CH3)2 C(CH3)3

Figure 7.17

X =

R = R' = (CH2)4CH3 CH3 CH3 (CH2)5CH3 CH3 (CH2)6CH3 CH3 1-admantyl (CH2)4CH3 CH2CH3

R= H H H H H H H H Cl

H F Cl Br CH3 OCH3 C(CH3)3 CF3 H

Ketones with diverse structures

Table 7.1 The isolated yields, ee and optical rotation of the product alcohols on the preparative scale Ketones

Yield (%)

O

ee (%)

OH O O

O

87

> 99 (R)

O

91

99 (R)

89

> 99 (R)

94

> 99 (R)

92

> 99 (S)

92

> 99 (R)

95

> 99 (R)

90

> 99 (R)

O

O

OH O O

O O

OH

F

F O

F3 C

OH

F3 C

O

OH Cl

O

Cl

OH

O

O

OH

OH

134


in the preparation of a potent, selective and cell-penetrable inhibitor of caspase 3 [59]. (R)-2Hydroxy-3,3-dimethylbutyrate is a key component P3 of thrombin inhibitor identified by Merck [60]. Chiral 2-chloro-1-phenylethanol is a key synthon for the preparation of a large group of antidepressants and potential cocaine-abuse therapeutic agents [61]. Optically pure 1(40 -flurophenyl)ethanol and 1-(40 -trifluromethylphenyl)ethanol have been used to synthesize m2 muscarinic antagonists and the modulators of CCR-5 chemokine receptor for treating patients with HIV [62–64]. OH

O OC2H5

R

OH

SSCR

OC2H5

R

O

SSCR

OC2H5

R

O

O

R = Aryl ee = 43-99%

R = CH(CH3)2, C(CH3)3 ee = 99%

Figure 7.18 Reduction of a-ketoesters by a carbonyl reductase from Sporobolomyces salmonicolor (SSCR)

Similarly, the carbonyl reductase from Sporobolomyces salmonicolor (SSCR) reduced a broad range of ketones, including aliphatic and aromatic ketones and a- and b-ketoesters [57]. Among these substrates, SSCR showed highest activity toward the reduction of a-ketoesters. Aromatic a-ketoesters were reduced to (S)-a-hydroxy esters, while (R)-enantiomers were obtained from the reduction of aliphatic counterparts (Figure 7.18). This interesting observation is consistent with enzyme–substrate docking studies, which show that hydride transfer occurs at the different faces of the carbonyl group for aromatic and aliphatic a-ketoesters. It is worth noting that some bulky ketone substrates, such as 20 -methoxyacetophenone, 1-adamantyl methyl ketone, ethyl 4,4-dimethyl-3-oxopentanoate, and ethyl 3,3-dimethyl-2-oxobutanoate, were reduced to the corresponding alcohols with excellent optical purity. SSCR also effectively catalyzed the enantioselective reduction of aryl alkyl ketones having a number of diverse alkyl groups [58]. While linear alkyl chain induced an enantiopreference reversal of the reduction, the enantiopreference remained unchanged and enantioselectivity was dramatically enhanced when the alkyl group became branched (Figure 7.19). The lowest energy conformations of the aryl alkyl ketones in the enzyme active site, obtained from enzyme–substrate docking studies, were found to be consistent with these intriguing experimental results [65]. Significantly, ketones with sterically bulky alkyl groups, such as iso-propyl, tert-butyl, and OH R

R CH 3 CH 2CH3 CH(CH 3)2 C(CH 3)3 cyclo-C3H5

Figure 7.19 (SSCR)

O SSCR

ee(%) 42 28 98 98 96

OH R

SSCR

R CH2CH2CH3 CH2CH2CH2CH3 CH2CH2CH2CH2CH3 CH2CH2CH2CH2CH2CH3

R

ee(%) 88 87 34 27

Reduction of aryl alkyl ketones by a carbonyl reductase from Sporobolomyces salmonicolor

135


cyclo-propyl, were enzymatically reduced to the corresponding alcohols with greater than 96% ee. Thus, SSCR possesses an unusually broad substrate spectrum and seems to be especially useful for the reduction of ketones with sterically bulky substituents. Both the alcohol dehydrogenase (YMR226c) and the carbonyl reductase (CMCR) efficiently catalyzed the reduction of various aromatic b-ketonitriles bearing an electron-withdrawing or electron-donating group on the phenyl ring to the corresponding b-hydroxy nitriles with 95% ee or higher [66]. While the CMCR-catalyzed reduction gave the (R)-enantiomer, (S)configured b-hydroxy nitriles were obtained with the alcohol dehydrogenase (YMR226c) as the biocatalyst (Figure 7.20). In all cases, use of isolated enzyme eliminated the competing aethylation reaction, which was often observed in the reduction of aromatic b-ketonitriles using whole-cell biocatalysts. The (S)- and (R)-b-hydroxy nitriles obtained are a class of important compounds, because they can be converted to chiral 1,3-amino alcohols and b-hydroxy carboxylic acids. In the same study, (S)- and (R)-b-hydroxy nitriles were further converted to the (S)- and (R)-b-hydroxy carboxylic acids in high yields respectively using a nitrilase (bll6402) from Bradyrhizobium japonicum strain USDA110 or a nitrilase (NIT6803) from cyanobacterium Synechocystis sp. strain PCC 6803 (Figure 7.20). The coupled enzymatic reduction and hydrolysis can be carried out in a one-pot, two-step fashion without the purification of the intermediate product b-hydroxy nitriles, allowing ready access to both chiral b-hydroxy nitriles and b-hydroxy carboxylic acids of pharmaceutical importance. O

OH CN

R

Ymr226C

(S)

OH CN

R

Yield >78%, ee 95-99% O CMCR

(S)

COOH

Yield >87%

(R)

R

Yield >85%, ee 97-99%

Figure 7.20

R

OH CN

R

bll6402

OH CN

NIT6803

R

(R)

COOH

Yield >80%

Synthesis of b-hydroxy carboxylic acids with coupled enzymatic processes

After having screened several carbonyl reductases from various sources, we have found that the carbonyl reductase from Candida magnoliae (CMCR) and bakers yeast alcohol dehydrogenase (YMR226c) also catalyzed the reduction of a series of a-azidoacetophenone derivatives to afford (S)- and (R)-2-azido-1-arylethanols with excellent yield and optical purity respectively, providing an effective route to this class of important compounds (Figure 7.21). The azido alcohols readily reacted with alkynes via click chemistry to afford b-blocker analogues with a triazol functional moiety. 3,5-Dioxocarboxylates (b,d-diketo esters) unify the substructures of b-diketones and bketoesters in an overlapping manner within a single molecule, and represent a class of exceptionally demanding substrates with respect to regio- and enantio-selective reduction. Screening of several reductases in the presence of alkyl 3,5-dioxocarboxylates revealed that an NADP þ -dependent alcohol dehydrogenase of Lactobacillus brevis (LBADH) accepted these compounds as substrates [67,68]. Preparative-scale reduction of tert-butyl 3,5-dioxohexanoate

136


OH

O N3

OH N3

CMCR

X

X

Click Chemistry R

N N N

R

N N N

R

X

Yield >80%, ee >99% OH

O

OH N3

N3 Ymr226c

X

X

Click Chemistry R

X

Yield >82%, ee >99%

Figure 7.21 Synthesis of b-blocker analogues with triazol functional moiety

with LBADH gave the 3-oxo-5-hydroxy esters with > 99% ee and complete regioselectivity (Figure 7.22). When X was Cl, the (R)-enantiomer of 3-oxo-5-hydroxy ester could be prepared with 90–94%ee bythebiocatalytic reduction using bakers yeastin a biphasic system (50% yield). The 3-oxo group of 3-oxo-5-hydroxy ester could be reduced by syn- or anti-selective borohydride reduction to yield all four stereomers of the chlorinated b,d-dihydroxy hexanoate [69]. O

O

O

OH O LBADH

X

O

O

NADP+

NADPH O

X = H, 77% yield, 99.4% ee X = Cl, 72% yield, >99.5% ee X = Me, 66% yield, 98.5% ee

OH LBADH

Figure 7.22

O

X

Selective reduction of tert-butyl 3,5-dioxohexanoate with LBADH

Furthermore, tert-butyl 4-methyl-3,5-dioxohexanoate was regio- and enantio-selectively reduced via dynamic kinetic resolution using LBADH as catalyst to give an almost enantiomerically and diastereomerically pure tert-butyl (4S,5R)-5-hydroxy-4-methyl-3-oxohexanoate (Figure 7.23) [70]. O

O

O

OH O

O

LBADH O

O NADP+

NADPH O

OH

ee 99.2% syn:anti 97:3

LBADH

Figure 7.23

Selective reduction of tert-butyl 4-methyl-3,5-dioxohexanoate with LBADH

LBADH also catalyzed the asymmetric reduction of a broad variety of differently substituted acetylenic ketones, including aromatic alkynones and a number of aliphatic derivatives [71]. For example, methyl alkynones bearing an aromatic unit attached to the triple bond were reduced to the corresponding (R)-propargylic alcohols with > 99% ee. Similarly, alkylsilyl-substituted

137


OH LBADH R

R

R = aryl, heteroaryl, alkylsilyl, >99% ee R = H, 60% ee

Figure 7.24

Enzymatic reduction of alkynones

(R)-propargylic alcohols were also obtained with > 99% ee (Figure 7.24). The reduction of 3butyn-2-one by LBADH gave (R)-3-butyn-2-ol with only 60% ee, but optically pure (R)-3butyn-2-ol could be prepared by desilylation of alkylsilyl-substituted (R)-propargylic alcohols, which were obtained in the enzymatic reduction, with borax. a-Halogenated propargylic ketones were also reduced by LBADH to give the corresponding propargylic a-chlorohydrins with > 98% ee, which were chemically converted to propargylic epoxides (Figure 7.25) [72]. O

OH

X

O

X

LBADH R

R

DBU EtOH/H2O

R

X = Cl, R = Ph, TBS, TMS, ee >99%; X = Br, R = TMS, ee = 98.5%

Figure 7.25

Enzymatic reduction of a-halogenated propargylic ketones by LBADH

When n-alkyl ethynyl ketones were tested as the substrate of LBADH, the preferred stereochemistry and optical purity of the resulting propargylic alcohol were dependent upon the size of the alkyl group (Figure 7.26) [71]. The substituted acetylenic ketones including aromatic alkynones as well as a number of aliphatic derivatives were also reduced enantioselectively by NAD þ -dependent Candida parapsilosis carbonyl reductase (CPCR). Aromatic propargylic alcohols were obtained with ee values higher than 99%, but with opposite configuration of those obtained by LBADH-catalyzed reduction. For aliphatic derivatives, the ee was dependent on the size of the substituents, and the enantioselectivity was lower than that of LBADH in most cases [71]. Horse liver alcohol dehydrogenase and Thermoanaerobacter brockii alcohol dehydrogenase were used to catalyze the stereoselective reduction of 3-butyn-2one derivatives to give (R)-enantiomer of 3-butyn-2-ol derivatives with > 99% ee. The O

OH R

LBADH

R

R = CH3, 60% ee (R) C2H5, 34% ee (S) C3H7, >99% ee (S) C5H11, >99% ee (S)

Figure 7.26 Enzymatic reduction of n-alkyl ethynyl ketones by LBADH

138


OH

Cl

HLADH

O

Cl

Ph

Ph

DBU EtOH/H2O

Ph

Yield 97%, ee 99%

Figure 7.27

Synthesis of ()-(2R)-(phenylethynyl)oxirane

reduction of 1-chloro-4-phenyl-3-butyn-2-one using horse liver alcohol dehydrogenase gave ()-(2R)-1-chloro-4-phenyl-3-butyn-2-ol (97% yield, 99% ee), which was further converted into ()-(2R)-(phenylethynyl)oxirane (Figure 7.27) [72]. Lactobacillus kefir alcohol dehydrogenase (LKADH), an enzyme highly homologous to LBADH, has very broad substrate specificity and accepts a wide range of aromatic, aliphatic and cyclic ketones. This NADP þ -dependent enzyme exhibits the anti-Prelog rule and transfers the hydride from cofactor to the si-face of the carbonyl group to give (R)-configurated alcohols with high enantioselectivity [73]. Recently, the gene encoding LKADH from Lactobacillus kefir DSM 20 587 was cloned and expressed in Escherichia coli. The deduced amino acid sequence indicated a high degree of similarity to short-chain dehydrogenases. For the reduction of acetophenone, the specific activity of the homogeneous recombinant LKADH was 558 U mg1. The enzyme showed its maximum activity at 50 C, while the pH optimum was at pH 7.0. The purified LKADH exhibited broad substrate specificity. The stereoselective reduction of several tested aliphatic and aromatic ketones, as well as b-ketoesters, afforded the corresponding alcohols with > 99% ee and in the case of diketones > 99% de [74]. The regeneration of NADPH could be realized by GDH or alternatively by a substrate-coupled approach using isopropanol as hydrogen donor. LKADH-catalyzed asymmetric reduction of 2-acetylchromen-4-one was achieved with 68% yield and 90% ee (Figure 7.28). The O

O LKADH

O

O O

OH

NADP+

NADPH O

OH LKADH

Figure 7.28

LKADH-catalyzed asymmetric reduction of 2-acetylchromen-4-one

(R)-stereochemistry of the corresponding alcohol was unambiguously assigned by means of single-crystal X-ray analysis [75]. Enantioselective reduction of a,b-unsaturated ketones was achieved using LKADH as biocatalyst to produce (R)-allylic alcohol with high optical purity (Figure 7.29) [76]. The enzymatic reduction could be combined with the preparation step of a,b-unsaturated ketones (Wittig reaction) in a one-pot, two-step fashion. Similarly, chiral (R)and (S)-allylic alcohols were also prepared in good to high overall isolated yields through a two-step, one-pot chemoenzymatic process based on the palladium-catalyzed Heck reaction of aryl iodides with butenone followed by an enzymatic reduction of the resultant a,b-unsaturated

139


OH LKADH

R

R NADP+

NADPH O

OH LKADH

R = H, >99% ee; R = CH3, 95% ee; R = NO2, >99% ee

Figure 7.29

Enantioselective reduction of a,b-unsaturated ketones using LKADH

ArI

+ O Pd cat

OH

O

T-ADH

Ar

Ar

Ar +

NADP

NADPH

OH

O T-ADH

Figure 7.30

OH

LBADH NADP +

NADPH

OH

O LBADH

Enzymatic synthesis of chiral (R)- and (S)-allylic alcohols

ketones using alcohol dehydrogenases from Lactobacillus brevis and Thermoanaerobacter species respectively (Figure 7.30) [77]. The (S)-enantiomers of the corresponding allylic alcohols were synthesized via enzymatic reduction of a,b-unsaturated ketones catalyzed by an alcohol dehydrogenase from Rhodococcus sp. [76]. The NAD þ -dependent (S)-alcohol dehydrogenase (READH) gene from Rhodococcus erythropolis strain DSM 43 297 was cloned and expressed in Escherichia coli [78,79]. The recombinant enzyme showed a broad substrate range comprising aliphatic and aromatic ketones, as well as b-ketoesters. Several tested substrates were reduced to the corresponding (S)-alcohols with > 99% ee using READH coupled with FDH from Candida boidinii for in situ NADH regeneration. Among them, (S)-1-(p-chlorophenyl)ethanol ( > 99% ee) was obtained in78% isolated yield on a preparative scale. In addition, READH was stable in biphasic reaction medium containing 25% of heptane. The FDH from Candida boidinii was found to be active under the same reaction conditions. Thus, reductions of poorly water-soluble ketones such as p-chloroacetophenone in the presence of the alcohol dehydrogenase from Rhodococcus erythropolis and an FDH from Candida boidinii could be carried out at higher substrate concentration of 10–200 mM. The resulting (S)-alcohols were formed with moderate to good conversion rates, and with up to > 99% ee [78]. A series of ethynyl ketones and ethynylketoesters were reduced enantioselectively to the corresponding nonracemic propargyl alcohols using a secondary alcohol dehydrogenase from

140


Thermoanaerobacter ethanolicus (TESADH). However, the substrate-binding pocket is too small to accommodate phenyl-ring-containing ketones, such as 4-phenyl-2-butanone. Recently, Phillips and coworkers have reported that a mutant of TESADH, where tryptophan-110 was substituted by alanine (W110A TESADH), catalyzed the enantioselective asymmetric reduction of phenyl-ring-containing prochiral ketones to yield the corresponding optically active secondary alcohols in Tris buffer using 2-propanol (30%, v/v) as cosolvent and cosubstrate. This concentration of 2-propanol was critical not only to enhance the solubility of hydrophobic phenyl-ring-containing substrates in the aqueous reaction medium, but also to shift the equilibrium in the reduction direction. The resulting alcohols have the (S)-configuration, in agreement with Prelogs rule. A series of phenyl-ring-containing ketones, such as 4phenyl-2-butanone and 1-phenyl-1,3-butadione, were reduced with good to excellent yields and high enantioselectivities. On the other hand, 1-phenyl-2-propanone was reduced with lower ee. (R)-Alcohols were obtained by enantiospecific oxidation of (S)-alcohols through oxidative kinetic resolution of the racemic alcohols using W110A TESADH in Tris buffer/ acetone (90:10, v/v) [80]. The asymmetric reduction of hydrophobic ketones by xerogelimmobilized W110A TESADH in organic solvents afforded the (S)-configured alcohols in yields comparable to those achieved by using the free enzyme and, in some cases, with higher enantioselectivities [81]. The substrate specificity and enantioselectivity of an NAD þ -dependent alcohol dehydrogenase (LSADH) isolated from Leifsonia sp. S749 were studied. The enzyme catalyzed the asymmetric reductions of various prochiral ketones, including aliphatic and aromatic methyl ketones, chloromethyl ketones, and a- and b-ketoesters, to yield optically active secondary alcohols with anti-Prelog configuration. In most cases, the product alcohols were obtained with 98% ee or higher, the only exceptions being aromatic a-ketoesters and 20 ,40 -dichloroacetophenone. LSADH was able to efficiently reproduce NADH when 2-propanol was used as a hydrogen donor in the reaction mixture. Under optimized conditions using Escherichia coli cell overexpressing LSADH as catalyst, reduction of phenyl trifluoromethyl ketone to (S)-1phenyl-2,2,2-trifluoroethanol reached 99% conversion (about 570 mM in the reaction mixture, 99 g L1) [82,83]. Leifsonia alcohol dehydrogenase (LSADH) also catalyzed the reduction of short-chain ketone substrates (C3–C4) such as chloro-, hydroxyacetones and hydroxybutanones to give the corresponding alcohols with 82–99% ee. Among them, 4-hydroxy-2butanone was reduced to (R)-1,3-butanediol, a useful intermediate for pharmaceuticals, in high yield with 99% ee. These water-soluble ketone substrates were also reduced via asymmetric hydrogen-transfer catalyzed by mutated Rhodococcus phenylacetaldehyde reductase (PAR) with ee in the range 49 to 87% [84]. Enzymes from hyperthermophiles (that are, microorganisms that grow optimally above 80 C) usually display not only extreme stability at high temperature and high pressure, but also high tolerance of chemical denaturants, such as organic solvents [85]. In addition, the hyperthermophilic enzymes can be subjected to heat treatment, which is an easy-handling and effective protein purification process. These unique features of hyperthermophilic enzymes are particularly important from a practical point of view [86,87]. For example, an alcohol dehydrogenase from the hyperthermophilic archaeon Pyrococcus furiosus (PFADH) shows high resistance to thermal deactivation, which has a half-life of 130 min at 100 C [88]. Further studies have shown that PFADH possesses high tolerance of organic solvents such as dimethyl sulfoxide, isopropanol, methyl tert-butyl ether and hexane, a particularly important and useful feature for the reduction of hydrophobic ketones with low availability in pure

141


aqueous buffers [89]. The temperature-dependent studies indicated that the enzyme activity was enhanced as the reaction temperature increased. Surprisingly, the enantioselectivity did not decrease at higher temperatures, a useful feature for certain applications requiring asymmetric reduction at high temperatures. Assessment of the substrate specificity and stereoselectivity indicated PFADH catalyzed the asymmetric reduction of a variety of ketones, including aryl ketones and a- and b-ketoesters, to the corresponding chiral alcohols. An interesting observation was that aryl ketones and phenyl-substituted a- and b-ketoesters were reduced to the corresponding chiral alcohols in optically pure forms, while the substrates without phenyl groups were reduced with moderate enantioselectivity. This suggests that a phenyl group next to the carbonyl group might be required to achieve excellent enantioselectivity. The chloro group at a-carbon of the carbonyl group improved the enzyme activity, as shown in the reductions of a-chloroacetophenone and ethyl 4-chloro-3-oxo-butyrate. These observations would provide useful guidance for the future application of this unique enzyme through rational substrate design. In an effort to develop ‘easy-to-use’ ketoreductase ‘toolbox’, we have surveyed the activity and enantioselectivity of a collection of ketoreductases (KRED) from various sources toward the reduction of a variety of ketones [90,91]. These studies served as a useful guideline for developing enzymatic processes for the production of optically pure chiral alcohols. For example, several chiral chlorohydrins of pharmaceutical importance were synthesized in both enantiomeric forms using the enzymes in this ketoreductase collection (Table 7.2) [92]. Further applications of this collection and other commercially available ketoreductases can be found in a recent review [9]. In summary, ketoreductases have emerged as valuable catalysts for asymmetric ketone reductions and are preparing to enter the mainstream of synthetic chemistry of chiral alcohols. These biocatalysts are used in three forms: wild-type whole-cell microorganism, recombinant Table 7.2

Enzymatic preparation of both enantiomers of chiral chlorohydrins OH Cl

R

a-Chloro alcohols (S)-1 (H) (R)-1 (H) (S)-2 (30 -Cl) (R)-2 (30 -Cl) (S)-3 (40 -Cl) (R)-3 (40 -Cl) (S)-4 (30 ,40 -Cl2) (R)-4 (30 ,40 -Cl2) (S)-5 (40 -NO2) (R)-5 (40 -NO2) (S)-6 (40 -CH3SO2NH) (R)-6 (40 -CH3SO2NH)

KRED

Yield (%)

112 130 112 130 112 130 112 130 107 130 113 130

72 94 69 99 97 81 79 88 87 76 96 89

ee (%) > 99 > 99 > 99 > 99 > 99 > 99 > 99 > 99 > 99 > 99 > 99 > 99

142


whole cell overexpressing catalytic enzymes, and isolated enzyme. Although each form has its own advantages and limitations, the latter two are more effective from the practical point of view. In particular, isolated enzyme is more feasible for synthetic organic chemists to integrate the biotransformation into the complex chemical processes.

7.2 Alcohol Oxidases Oxidation of alcohols to carbonyl compounds is one of the most fundamental transformations in organic chemistry. In particular, selective oxidation of primary alcohols to aldehydes still presents a great challenge in synthetic organic chemistry, because the product aldehydes can be further oxidized to carboxylic acids. In nature, this transformation is achieved by alcohol dehydrogenases and alcohol oxidases. Alcohol dehydrogenases catalyze the reversible oxidation of alcohols and require a cofactor (usually NAD þ or NADP þ ). Although it has been used to catalyze the oxidation of primary and secondary alcohols to aldehydes and ketones [93], alcohol dehydrogenase has mainly been used in the asymmetric reduction of ketones, as described in the previous section. In contrast to alcohol dehydrogenases, alcohol oxidases catalyze the irreversible oxidation of alcohols, with concomitant electron transfer onto molecular oxygen to form hydrogen peroxide (H2O2) as by-product. In this case, the necessary cofactor, flavine adenine dinucleotide (FAD), is incorporated into the molecular structure of the alcohol oxidase and regenerated during the reaction or with the help of oxygen in the reaction medium. The highly reactive H2O2 deactivates enzymes, so catalases are often coupled with alcohol oxidases to avoid high concentration of H2O2 in the reaction mixture via catalytic disproportionation of H2O2 into H2O and O2. Alcohol oxidases are conveniently subdivided into primary and secondary alcohol oxidases, and their synthetic applications will be presented accordingly.

7.2.1 Primary Alcohol Oxidases FAD-dependent aliphatic alcohol oxidases from methylotropic yeasts such as Candida boidinii and Pichia pastoris catalyze the oxidation of primary alkanols to the corresponding aldehydes, with a clear preference for unbranched short-chain aliphatic primary alcohols. The widest substrate specificity is found for alcohol oxidase from the yeast Pichia pastoris. For example, numerous unbranched primary alcohols including propargyl alcohol, 2-chloroethanol, and 2cyanoethanol were effectively oxidized by the alcohol oxidase from the yeast Pichia pastoris [94]. Pichia pastoris alcohol oxidase together with the catalase of bovine liver was used to catalyze the oxidation of hexanol to hexanal in a two-phase system composed of the aqueous buffer phase and the hexanol phase, in which the alcoholic substrate itself plays the role of the organic phase. High enzymatic activity was observed with hexanol concentration between 20 and 80 mass% [95]. Enzymatic production of glyoxal from ethylene glycol has been achieved using alcohol oxidase from Pichia pastoris or glycerol oxidase from Aspergillus japonicus [96,97]. Both alcohol and glycerol oxidases oxidized ethylene glycol to glyoxal via glycolaldehyde intermediate. By carefully controlling the reaction conditions, glycolaldehyde was prepared in 92 or 97% yield using alcohol oxidase or glycerol oxidase, offering a new route to glycolaldehyde which is superior to the chemical methods in terms of conversion yields and selectivity [96]. Clark et al. have examined the oxidation of some racemic 2-methyl-1-alkanols catalyzed by alcohol oxidases from Candida boidinii, Hansenula sp., Pichia pastoris and Torulopsus methanothermo. The (S)-enantiomer was oxidized faster than the (R)-counterpart

143


and the alcohol oxidase from Candida boidinii showed highest enantioselectivity with E ¼ 4.8 for 2-methyl-1-pentanol [98]. The alcohol oxidase-catalyzed oxidation step can be coupled with an aldol condensation catalyzed by aldolase. For example, when 4-pentenol was oxidized by the alcohol oxidase from Pichia pastoris in the presence of acetone and an aldolase, the aldol product (S)-4-hydroxy-7-octen-2-one was produced in 30% yield with 70% ee (Figure 7.31) [99]. These results show that the oxidase/aldolase-catalyzed one-pot reaction is a very promising route for the production of chiral aldol products from a range of achiral primary alcohols. OH

Alcohol oxidase Catalase

Aldolase Acetone

O

Figure 7.31

OH O

Synthesis of (S)-4-hydroxy-7-octen-2-one

The side product H2O2 can serve as the oxidant in the peroxidase-catalyzed oxidation. In a bienzymatic system, the oxidation of methanol catalyzed by the alcohol oxidase from yeast Pichia pastoris was coupled with a peroxidase- (from Coprinus cinereus) catalyzed oxidation of thioanisole, producing (S)-methyl-phenyl-sulfoxide in 72% isolated yield with 75% ee (Figure 7.32) [100]. This approach avoids the inactivation of peroxidase at high H2O2 concentration.Otheroxidases,suchasglucoseoxidasefromAspergillusnigerand D-aminoacidoxidasefrom Trygonopsis variabilis, have also been used to produce H2O2 in situ for similar purpose [101,102]. CH3OH

Alcohol oxidase

HCHO _

O2

H2O2

H2O

S

O S

+

Peroxidase

Figure 7.32

Biocatalytic oxidation of thioanisole to (S)-methyl-phenyl-sulfoxide

Aryl alcohol oxidase from the ligninolytic fungus Pleurotus eryngii had a strong preference for benzylic and allylic alcohols, showing activity on phenyl-substituted benzyl, cinnamyl, naphthyl and 2,4-hexadien-1-ol [103,104]. Another aryl alcohol oxidase, vanillyl alcohol oxidase (VAO) from the ascomycete Penicillium simplicissimum catalyzed the oxidation of vanillyl alcohol and the demethylation of 4-(methoxymethyl)phenol to vanillin and 4-hydroxybenzaldehyde. In addition, VAO also catalyzed deamination of vanillyl amine to vanillin, and hydroxylation and dehydrogenation of 4-alkylphenols. For the oxidation of 4-alkylphenol, the ratio between the alcohol and alkene product depended on the length and bulkiness of the alkyl side-chain [105,106]. 4-Ethylphenol and 4-propylphenol, were mainly converted to (R)1-(40 -hydroxyphenyl) alcohols, whereas medium-chain 4-alkylphenols such as 4-butylphenol were converted to 1-(40 -hydroxyphenyl)alkenes. Besides its natural substrates, galactose oxidases efficiently catalyzed the oxidation of substrates with an accessible D-galactose moiety and some water-soluble small alcohols such as threitol, glycerol and 3-halo-1,2-propanediols to the corresponding aldehydes [99]. The transformation of glycerol to L-glyceraldehyde catalyzed by galactose oxidase was integrated into a one-pot multi-enzyme synthesis of L-fructose (Figure 7.33) [107].

144


OH

HO

GOase, pH 6.8

OH

O

OH

O

1. RhaD, pH 6.8 2. AP, pH 4.7

OH

OH

OH OH

H2O2

O2

H3PO4

O H2O3PO

Catalase

OH

HO

OH

H2O + 1/2 O2

Figure 7.33

One-pot multi-enzyme synthesis of L-fructose

In another example, the C-6 oxidation of p-nitrophenyl 2-acetamido-2-deoxy-b-D-galactopyranoside was effected using a galactose oxidase from Dactylium dendroides to afford p-nitrophenyl-2-acetamido-2-deoxy-b-D-galactohexodialdo-1,5-pyranoside (Figure 7.34) [108]. The transglycosylation reactions using the resulting aldehyde as the donor and 2-acetamido-2-deoxyD-glucopyranose as the acceptor catalyzed by the b-N-acetylhexosaminidase from Talaromyces flavus gave 37% yield of 2-acetamido-2-deoxy-b-D-galactohexodialdo-1,5-pyranosyl-(1 ! 4)-2acetamido-2-deoxy-D-glucopyranose, which was oxidized in situ to yield 2-acetamido-2-deoxyb-D-galactopyranosyluronic acid-(1 ! 4)-2-acetamido-2-deoxy-D-glucopyranose, a high-affinity ligand for two natural killer-cell activation receptors, NKR-P1A and CD69. H2O + 1/2 O2 Catalase HO HO

NO2

OH O

O

O2

H2O2

Galactose oxidase

NHAc

NO2

HO OH HO O O HO NHAc OH

HO HO

O

OH

NHAc

HO CO H 2 O O HO HO NHAc

OH O

OH NHAc

NaClO2, H2O

β-N-Acetylhexosaminidase (Talaromyces flavus) pH 5.0

HO OH HO O O HO HO NHAc

OH O

OH NHAc

Figure 7.34 Synthesis of 2-acetamido-2-deoxy-b-D-galactopyranosyluronic acid-(1 ! 4)-2-acetamido2-deoxy-D-glucopyranose

7.2.2 Secondary Alcohol Oxidases Secondary alcohol oxidases catalyze the oxidation of secondary alcohols to ketones using molecular oxygen as oxidant. A secondary alcohol oxidase from polyvinyl alcohol-degrading bacterium Pseudomonas vesicularis var. povalolyticus PH exhibited activity toward several

145


secondary alcohols, but no activity toward primary alcohols. In particular, the enzyme displayed high activity towards the long-chain secondary alcohols such as 4-heptanol, 2octanol, and 4-decanol [109]. Glycolate oxidase converts glycolic acid to glyoxylic acid and has been employed as a biocatalyst for the production of glyoxylic acid and optically active 2-hydroxy carboxylic acids [110]. For example, enantiomerically pure (R)-2-hydroxy acids (up to > 99% ee) were obtained by enantioselective oxidation of racemic acids with molecular oxygen catalyzed by the glycolate oxidase from spinach (Spinacia oleracea) [111]. This enantioselective oxidation was coupled with asymmetric reduction of 2-oxo acids with D-lactate dehydrogenase from Lactobacillus leichmannii to produce enantiomerically pure (R)-2-hydroxy acids in up to 89% yield based on the racemate (Figure 7.35) [112]. OH R

O

Glycolate oxidase

OH

OH OH

R

O

O

O

LDH

H2O2

O2

Catalase

NAD+

NADH

H2O + 1/2 O2

Figure 7.35

OH

R

CO2

HCO2H

Stereoinversion of 2-hydroxy acids using sequential oxidation and reduction

The addition of (aminomethyl)phosphonic acid into the oxidation reaction mixture of glycolic acid improved the yield of glyoxylic acid. After separation and recovery of the biocatalyst from the oxidation product mixture for reuse, the resulting solution of glyoxylic acid and (aminomethyl)phosphonic acid was subsequently hydrogenated with a palladium/ carbon catalyst to produce N-(phosphonomethyl)glycine (glyphosate), a broad-spectrum postemergent herbicide, in 86% yield with > 98% purity (Figure 7.36) [113]. H2O

Catalase

1/2O2

O +

HO

+

O-

H3N

PO32-

H2O2 O

Glycolate oxidase

O

HN

O

O2

O OH PO3H2

1. H2, Pd/C 2. HCl

N

+

+ O-

H3N

PO32-

O OPO32-

+H2O -H2O

HO HN

OPO32-

Glyphosate

Figure 7.36

Chemoenzymatic synthesis of N-(phosphonomethyl)glycine (glyphosate)

146


Compared with ketoreductases, the synthetic application of alcohol oxidases has been less explored. However, selective oxidation of primary alcohols to aldehydes is superior to the chemical methods in terms of conversion yields, selectivity, and environmental friendliness of reaction conditions. In addition, coupling of alcohol oxidase with other enzymes provides a tremendous opportunity to develop multi-enzyme processes for the production of complex molecules. Therefore, a growing impact of alcohol oxidases on synthetic organic chemistry is expected in the coming years.

References [1] Ohkuma, T. and Noyori, R. (2004) Hydrogenation of carbonyl groups, in Comprehensive Asymmetric Catalysis, Supplement I (eds. E.N. Jacobsen, A. Pfaltz and H. Yamamoto), Springer, Berlin, pp. 1–41. [2] Noyori, R. and Okhuma, T. (2001) Asymmetric catalysis by architectural and functional molecular engineering: practical chemo- and stereoselective hydrogenation of ketones. Angewandte Chemie–International Edition, 40 (1), 40–73. [3] Wills, M. and Hannedouche, J. (2002) New methodology for the asymmetric reduction of ketones. Current Opinion in Drug Discovery & Development, 5 (6), 881–891. [4] Thayer, A. (2005) Removing impurities. Chemical & Engineering News, 83 (36), 55–58. [5] Kroutil, W., Mang, H., Edegger, K. and Faber, K. (2004) Recent advances in the biocatalytic reduction of ketones and oxidation of sec-alcohols. Current Opinion in Chemical Biology, 8 (2), 120–126. [6] Nakamura, K., Yamanaka, R., Matsuda, T. and Harada, T. (2003) Recent developments in asymmetric reduction of ketones with biocatalysts. Tetrahedron: Asymmetry, 14 (18), 2659–2681. [7] Patel, R.N. (2001) Biocatalytic synthesis of intermediates for the synthesis of chiral drug substances. Current Opinion in Biotechnology, 12 (6), 587–604. [8] De Wildeman, S.M.A., Sonke, T., Schoemaker, H.E. and May, O. (2007) Biocatalytic reductions: from lab curiosity to ‘first choice’. Accounts of Chemical Research, 40 (12), 1260–1266. [9] Moore, J.C., Pollard, D.J., Kosjek, B. and Devine, P.N. (2007) Advances in the enzymatic reduction of ketones. Accounts of Chemical Research, 40 (12), 1412–1419. [10] Csuk, R. and Glaenzer, B.I. (1991) Bakers yeast mediated transformations in organic chemistry. Chemical Reviews (Washington, DC, United States), 91 (1), 49–97. [11] Bertau, M. and Burli, M. (2000) Enantioselective microbial reduction with bakers yeast on an industrial scale. Chimia, 54 (9), 503–507. [12] Martinez Lagos, F., Carballeira, J.D., Bermudez, J.L. et al. (2004) Highly stereoselective reduction of haloketones using three new yeasts: application to the synthesis of (S)-adrenergic beta-blockers related to propranolol. Tetrahedron: Asymmetry, 15 (5), 763–770. [13] Carballeira, J.D., Alvarez, E., Campillo, M.,et al. (2004) Diplogelasinospora grovesii IMI 171018, a new whole cell biocatalyst for the stereoselective reduction of ketones. Tetrahedron: Asymmetry, 15 (6), 951–962. [14] Guo, Z., Patel, B.P., Corbett, R.M. et al. (2006) Stereospecific microbial reduction of ethyl 1-benzyl-3-oxopiperidine-4-carboxylate. Tetrahedron: Asymmetry, 17 (13), 2015–2020. [15] Patel, R.N., Chu, L., Chidambaram, R. et al. (2002) Enantioselective microbial reduction of 2-oxo-2-(10 ,20 ,30 ,40 tetrahydro-10 ,10 ,40 ,40 -tetramethyl-60 -naphthalenyl)acetic acid and its ethyl ester. Tetrahedron: Asymmetry, 13 (4), 349–355. [16] Guo, Z., Chen, Y., Goswami, A. et al. (2006) Synthesis of ethyl and t-butyl (3R,5S)-dihydroxy-6-benzyloxyhexanoates via diastereo- and enantioselective microbial reduction. Tetrahedron: Asymmetry, 17 (10), 1589–1602. [17] Amidjojo, M., Franco-Lara, E., Nowak, A. et al. (2005) Asymmetric synthesis of tert-butyl (3R,5S) 6-chlorodihydroxyhexanoate with Lactobacillus kefir. Applied Microbiology and Biotechnology, 69 (1), 9–15. [18] Haberland, J., Kriegesmann, A., Wolfram, E. et al. (2002) Diastereoselective synthesis of optically active (2R,5R)-hexanediol. Applied Microbiology and Biotechnology, 58 (5), 595–599. [19] Haberland, J., Hummel, W., Daussmann, T. and Liese, A. (2002) New continuous production process for enantiopure (2R,5R)-hexanediol. Organic Process Research & Development, 6 (4), 458–462.


147

[20] Tan, A.W.I., Fischbach, M., Huebner, H. et al. (2006) Synthesis of enantiopure (5R)-hydroxyhexane-2-one with immobilised whole cells of Lactobacillus kefiri. Applied Microbiology and Biotechnology, 71 (3), 289–293. [21] Amidjojo, M. and Weuster-Botz, D. (2005) Asymmetric synthesis of the chiral synthon ethyl (S)-4-chloro-3hydroxybutanoate using Lactobacillus kefir. Tetrahedron: Asymmetry, 16 (4), 899–901. [22] Poessl, T.M., Kosjek, B., Ellmer, U. et al. (2005) Non-racemic halohydrins via biocatalytic hydrogen-transfer reduction of halo-ketones and one-pot cascade reaction to enantiopure epoxides. Advanced Synthesis and Catalysis, 347 (14), 1827–1834. [23] Stampfer, W., Edegger, K., Kosjek, B. et al. (2004) Simple biocatalytic access to enantiopure (S)-1-heteroarylethanols employing a microbial hydrogen transfer reaction. Advanced Synthesis and Catalysis, 346 (1), 57–62. [24] Stampfer, W., Kosjek, B., Faber, K. and Kroutil, W. (2003) Biocatalytic asymmetric hydrogen transfer employing Rhodococcus ruber DSM 44541. The Journal of Organic Chemistry, 68 (2), 402–406. [25] Stampfer, W., Kosjek, B., Moitzi, C. et al. (2002) Biocatalytic asymmetric hydrogen transfer. Angewandte Chemie–International Edition, 41 (6), 1014–1017. [26] Edegger, K., Stampfer, W., Seisser, B. et al. (2006) Regio- and stereoselective reduction of diketones and oxidation of diols by biocatalytic hydrogen transfer. European Journal of Organic Chemistry, (8), 1904–1909. [27] Van Deursen, R., Stampfer, W., Edegger, K. et al. (2004) Chemo- and stereo-selective biocatalytic reduction of a,b-unsaturated ketones employing a chemo-tolerant ADH from Rhodococcus ruber DSM 44541. Journal of Molecular Catalysis B–Enzymatic, 31 (4–6), 159–163. [28] Rodriguez, S., Kayser, M.M. and Stewart, J.D. (2001) Highly stereoselective reagents for b-keto ester reductions by genetic engineering of bakers yeast. Journal of the American Chemical Society, 123 (8), 1547–1555. [29] Rodriguez, S., Schroeder, K.T., Kayser, M.M. and Stewart, J.D. (2000) Asymmetric synthesis of b-hydroxy esters and alpha-alkyl-beta-hydroxy esters by recombinant Escherichia coli expressing enzymes from bakers yeast. The Journal of Organic Chemistry, 65 (8), 2586–2587. [30] Ema, T., Yagasaki, H., Okita, N. et al. (2006) Asymmetric reduction of ketones using recombinant E. coli cells that produce a versatile carbonyl reductase with high enantioselectivity and broad substrate specificity. Tetrahedron, 62 (26), 6143–6149. [31] Ema, T., Okita, N., Ide, S. and Sakai, T. (2007) Highly enantioselective and efficient synthesis of methyl (R)-ochloromandelate with recombinant E. coli: toward practical and green access to clopidogrel. Organic and Biomolecular Chemistry, 5 (8), 1175–1176. [32] Kizaki, N., Yasohara, Y., Hasegawa, J. et al. (2001) Synthesis of optically pure ethyl (S)-4-chloro-3-hydroxybutanoate by Escherichia coli transformant cells coexpressing the carbonyl reductase and glucose dehydrogenase genes. Applied Microbiology and Biotechnology, 55 (5), 590–595. [33] Yasohara, Y., Kizaki, N., Hasegawa, J. et al. (2001) Stereoselective reduction of alkyl 3-oxobutanoates by carbonyl reductase from Candida magnoliae. Tetrahedron: Asymmetry, 12 (12), 1713–1718. [34] Engelking, H., Pfaller, R., Wich, G. and Weuster-Botz, D. (2004) Stereoselective reduction of ethyl 4-chloro acetoacetate with recombinant Pichia pastoris. Tetrahedron: Asymmetry, 15 (22), 3591–3593. [35] Kataoka, M., Hoshino-Hasegawa, A., Thiwthong, R. et al. (2006) Gene cloning of an NADPH-dependent menadione reductase from Candida macedoniensis, and its application to chiral alcohol production. Enzyme and Microbial Technology, 38 (7), 944–951. [36] Yamamoto, H., Matsuyama, A. and Kobayashi, Y. (2002) Synthesis of ethyl (R)-4-chloro-3-hydroxybutanoate with recombinant Escherichia coli cells expressing (S)-specific secondary alcohol dehydrogenase. Bioscience Biotechnology and Biochemistry, 66 (2), 481–483. [37] Hanson, R.L., Goldberg, S., Goswami, A. et al. (2005) Purification and cloning of a ketoreductase used for the preparation of chiral alcohols. Advanced Synthesis and Catalysis, 347 (7–8), 1073–1080. [38] Groeger, H., Chamouleau, F., Orologas, N. et al. (2006) Enantioselective reduction of ketones with ‘designer cells’ at high substrate concentrations: highly efficient access to functionalized optically active alcohols. Angewandte Chemie–International Edition, 45 (34), 5677–5681. [39] Groeger, H., Rollmann, C., Chamouleau, F. et al. (2007) Enantioselective reduction of 4-fluoroacetophenone at high substrate concentration using a tailor-made recombinant whole-cell catalyst. Advanced Synthesis and Catalysis, 349 (4–5), 709–712. [40] Berkessel, A., Rollmann, C., Chamouleau, F. et al. (2007) Practical two-step synthesis of an enantiopure aliphatic terminal (S)-epoxide based on reduction of haloalkanones with ‘designer cells’. Advanced Synthesis and Catalysis, 349 (17–18), 2697–2704.

148


[41] Edegger, K., Gruber, C.C., Poessl, T.M. et al. (2006) Biocatalytic deuterium- and hydrogen-transfer using overexpressed ADH-‘A’: enhanced stereoselectivity and 2H-labeled chiral alcohols. Chemical Communications, (22), 2402–2404. [42] Ernst, M., Kaup, B., Mueller, M. et al. (2005) Enantioselective reduction of carbonyl compounds by whole-cell biotransformation, combining a formate dehydrogenase and a (R)-specific alcohol dehydrogenase. Applied Microbiology and Biotechnology, 66 (6), 629–634. [43] Kaup, B., Bringer-Meyer, S. and Sahm, H. (2004) Metabolic engineering of Escherichia coli: construction of an efficient biocatalyst for D-mannitol formation in a whole-cell biotransformation. Applied Microbiology and Biotechnology, 64 (3), 333–339. [44] Stewart, J.D., Rodriguez, S. and Kayser, M.M. (2001) Cloning, structure, and activity of ketone reductases from bakers yeast. Enzyme Technologies for Pharmaceutical and Biotechnological, Applications 175–207. [45] Kaluzna, I.A., Matsuda, T., Sewell, A.K. and Stewart, J.D. (2004) Systematic investigation of Saccharomyces cerevisiae enzymes catalyzing carbonyl reductions. Journal of the American Chemical Society, 126 (40), 12827–12832. [46] Padhi, S.K., Kaluzna, I.A., Buisson, D. et al. (2007) Reductions of cyclic beta-keto esters by individual Saccharomyces cerevisiae dehydrogenases and a chemo-enzymatic route to (1R,2S)-2-methyl-1-cyclohexanol. Tetrahedron: Asymmetry, 18 (18), 2133–2138. [47] Kayser, M.M., Drolet, M. and Stewart, J.D. (2005) Application of newly available bio-reducing agents to the synthesis of chiral hydroxy-beta-lactams: model for aldose reductase selectivity. Tetrahedron: Asymmetry, 16 (24), 4004–4009. [48] Kaluzna, I.A., Feske, B.D., Wittayanan, W. et al. (2005) Stereoselective, biocatalytic reductions of alpha-chlorobeta-keto esters. The Journal of Organic Chemistry, 70 (1), 342–345. [49] Feske, B.D., Kaluzna, I.A. and Stewart, J.D. (2005) Enantiodivergent, biocatalytic routes to both taxol side chain antipodes. The Journal of Organic Chemistry, 70 (23), 9654–9657. [50] Feske, B.D. and Stewart, J.D. (2005) Chemoenzymatic formal total synthesis of ()-bestatin. Tetrahedron: Asymmetry, 16 (18), 3124–3127. [51] Friberg, A., Johanson, T., Franzen, J. et al. (2006) Efficient bioreduction of bicyclo[2.2.2]octane-2,5-dione and bicyclo[2.2.2]oct-7-ene-2,5-dione by genetically engineered Saccharomyces cerevisiae. Organic and Biomolecular Chemistry, 4 (11), 2304–2312. [52] Katz, M., Frejd, T., Hahn-Haegerdal, B. and Gorwa-Grauslund, M.F. (2003) Efficient anaerobic whole cell stereoselective bioreduction with recombinant Saccharomyces cerevisiae. Biotechnology and Bioengineering, 84 (5), 573–582. [53] Yang, Y., Zhu, D., Piegat, T.J. and Hua, L. (2007) Enzymatic ketone reduction: mapping the substrate profile of a short-chain alcohol dehydrogenase (YMR226c) from Saccharomyces cerevisiae. Tetrahedron: Asymmetry, 18 (15), 1799–1803. [54] Wada, M., Kataoka, M., Kawabata, H. et al. (1998) Purification and characterization of NADPH-dependent carbonyl reductase, involved in stereoselective reduction of ethyl 4-chloro-3-oxobutanoate, from Candida magnoliae. Bioscience Biotechnology and Biochemistry, 62 (2), 280–285. [55] Kita, K., Nakase, K.-i., Yanase, H. et al. (1999) Purification and characterization of new aldehyde reductases from Sporobolomyces salmonicolor AKU4429. Journal of Molecular Catalysis B–Enzymatic, 6 (3), 305–313. [56] Zhu, D., Yang, Y. and Hua, L. (2006) Stereoselective enzymatic synthesis of chiral alcohols with the use of a carbonyl reductase from Candida magnoliae with anti-Prelog enantioselectivity. The Journal of Organic Chemistry, 71 (11), 4202–4205. [57] Zhu, D., Yang, Y., Buynak, J.D. and Hua, L. (2006) Stereoselective ketone reduction by a carbonyl reductase from Sporobolomyces salmonicolor. Substrate specificity, enantioselectivity and enzyme–substrate docking studies. Organic and Biomolecular Chemistry, 4 (14), 2690–2695. [58] Zhu, D. and Hua, L. (2006) Enantioselective enzymatic reductions of sterically bulky aryl alkyl ketones catalyzed by a NADPH-dependent carbonyl reductase. The Journal of Organic Chemistry, 71 (25), 9484–9486. [59] Mellon, C., Aspiotis, R., Black,et al. (2005) Lipophilic versus hydrogen-bonding effect in P3 on potency and selectivity of valine aspartyl ketones as caspase 3 inhibitors. Bioorganic & Medicinal Chemistry Letters, 15 (17), 3886–3890. [60] Nelson, T.D., LeBlond, C.R., Frantz, D.E. et al. (2004) Stereoselective synthesis of a potent thrombin inhibitor by a novel P2–P3 lactone ring opening. The Journal of Organic Chemistry, 69 (11), 3620–3627.


149

[61] Hsin, L.-W., Dersch, C.M., Baumann, M.H. et al. (2002) Development of long-acting dopamine transporter ligands as potential cocaine-abuse therapeutic agents: chiral hydroxyl-containing derivatives of 1-[2-[bis (4-fluorophenyl)methoxy]ethyl]-4-(3-phenylpropyl)piperazine and 1-[2-(diphenylmethoxy)ethyl]-4-(3-phenylpropyl)piperazine. Journal of Medicinal Chemistry, 45 (6), 1321–1329. [62] Harriman, G.C.B., Carson, K.G., Flynn, D.L.,et al. (2002) Preparation of functionalized indoles, benzimidazolones and related heterocycles as modulators of CCR-5 chemokine receptor and use in treating patients with HIV. WO, 2002072549. [63] McKittrick, B.A., Guo, G., Zhu, Z. and Ye, Y. (2002). Preparation of benzylpiperidines as m2 muscarinic antagonists. WO, 2002051808. [64] Tagat, J.R., Steensma, R.W., McCombie, S.W. et al. (2001) Piperazine-based CCR5 antagonists as HIV-1 inhibitors. II. Discovery of 1-[(2,4-dimethyl-3-pyridinyl)carbonyl]-4- methyl-4-[3(S)-methyl-4-[1(S)-[4(trifluoromethyl)phenyl]ethyl]-1-piperazinyl]-piperidine N1-oxide (sch-350634), an orally bioavailable, potent CCR5 antagonist. Journal of Medicinal Chemistry, 44 (21), 3343–3346. [65] Cundari, T.R., Dinescu, A., Zhu, D. and Hua, L. (2007) A molecular modeling study on the enantioselectivity of aryl alkyl ketone reductions by a NADPH-dependent carbonyl reductase. Journal of Molecular Modeling, 13 (6–7), 685–690. [66] Zhu, D., Ankati, H., Mukherjee, C. et al. (2007) Asymmetric reduction of b-ketonitriles with a recombinant carbonyl reductase and enzymatic transformation to optically pure b-hydroxy carboxylic acids. Organic Letters, 9 (13), 2561–2563. [67] Wolberg, M., Hummel, W., Wandrey, C. and Muller, M. (2000) Highly regio- and enantioselective reduction of 3,5-dioxocarboxylates. Angewandte Chemie–International Edition, 39 (23), 4306–4308. [68] Wolberg, M., Ji, A., Hummel, W. and Muller, M. (2001) Enzymatic reduction of hydrophobic beta,delta-diketo esters. Synthesis, (6), 937–942. [69] Wolberg, M., Hummel, W. and Muller, M. (2001) Biocatalytic reduction of beta,delta-diketo esters: a highly stereoselective approach to all four stereoisomers of a chlorinated beta,delta-dihydroxy hexanoate. Chemistry – A European Journal, 7 (21), 4562–4571. [70] Ji, A., Wolberg, M., Wandrey, C. et al. (2001) Dynamic kinetic resolution of tert-butyl 4-methyl-3,5dioxohexanoate through enzymatic reduction. Chemical Communications (Cambridge) (1), 57–58. [71] Schubert, T., Hummel, W., Kula, M.-R. and Muller, M. (2001) Enantioselective synthesis of both enantiomers of various propargylic alcohols by use of two oxidoreductases. European Journal of Organic Chemistry, (22), 4181–4187. [72] Schubert, T., Hummel, W. and Muller, M. (2002) Highly enantioselective preparation of multifunctionalized propargylic building blocks. Angewandte Chemie-International Edition, 41 (4), 634–637. [73] Bradshaw, C.W., Hummel, W. and Wong, C.H. (1992) Lactobacillus kefir alcohol dehydrogenase: a useful catalyst for synthesis. The Journal of Organic Chemistry, 57 (5), 1532–1536. [74] Weckbecker, A. and Hummel, W. (2006) Cloning, expression, and characterization of an (R)-specific alcohol dehydrogenase from Lactobacillus kefir. Biocatalysis and Biotransformation, 24 (5), 380–389. [75] Bisel, P., Walter, L., Nieger, M. et al. (2007) Stereochemical clarification of the enzyme-catalyzed reduction of 2-acetylchromen-4-one. Tetrahedron: Asymmetry, 18 (9), 1142–1144. [76] Krauueer, M., Hummel, W. and Groeger, H. (2007) Enantioselective one-pot two-step synthesis of hydrophobic allylic alcohols in aqueous medium through the combination of a Wittig reaction and an enzymatic ketone reduction. European Journal of Organic Chemistry, (31), 5175–5179. [77] Sgalla, S., Fabrizi, G., Cirilli, R. et al. (2007) Chiral (R)- and (S)-allylic alcohols via a one-pot chemoenzymatic synthesis. Tetrahedron: Asymmetry, 18 (23), 2791–2796. [78] Groger, H., Hummel, W., Rollmann, C. et al. (2004) Preparative asymmetric reduction of ketones in a biphasic medium with an (S)-alcohol dehydrogenase under in situ-cofactor-recycling with a formate dehydrogenase. Tetrahedron, 60 (3), 633–640. [79] Hummel, W., Abokitse, K., Drauz, K. et al. (2003) Towards a large-scale asymmetric reduction process with isolated enzymes: Expression of an (S)-alcohol dehydrogenase in E. coli and studies on the synthetic potential of this biocatalyst. Advanced Synthesis and Catalysis, 345 (1 þ 2), 153–159. [80] Musa, M.M., Ziegelmann-Fjeld, K.I., Vieille, C. et al. (2007) Asymmetric reduction and oxidation of aromatic ketones and alcohols using W110A secondary alcohol dehydrogenase from Thermoanaerobacter ethanolicus. The Journal of Organic Chemistry, 72 (1), 30–34.

150


[81] Musa, M.M., Ziegelmann-Field, K.I., Vieille, C. et al. (2007) Xerogel-encapsulated W110A secondary alcohol dehydrogenase from Thermoanaerobacter ethanolicus performs asymmetric reduction of hydrophobic ketones in organic solvents. Angewandte Chemie–International Edition, 46 (17), 3091–3094. [82] Inoue, K., Makino, Y. and Itoh, N. (2005) Production of (R)-chiral alcohols by a hydrogen-transfer bioreduction with NADH-dependent Leifsonia alcohol dehydrogenase (LSADH). Tetrahedron: Asymmetry, 16 (15), 2539–2549. [83] Inoue, K., Makino, Y., Dairi, T. and Itoh, N. (2006) Gene cloning and expression of Leifsonia alcohol dehydrogenase (LSADH) involved in asymmetric hydrogen-transfer bioreduction to produce (R)-form chiral alcohols. Bioscience Biotechnology and Biochemistry, 70 (2), 418–426. [84] Itoh, N., Nakamura, M., Inoue, K. and Makino, Y. (2007) Continuous production of chiral 1,3-butanediol using immobilized biocatalysts in a packed bed reactor: promising biocatalysis method with an asymmetric hydrogentransfer bioreduction. Applied Microbiology and Biotechnology, 75 (6), 1249–1256. [85] Vieille, C. and Zeikus, G.J. (2001) Hyperthermophilic enzymes: sources, uses, and molecular mechanisms for thermostability. Microbiology and Molecular Biology Reviews, 65 (1), 1–43. [86] Comfort, D.A., Chhabra, S.R., Conners, S.B. et al. (2004) Strategic biocatalysis with hyperthermophilic enzymes. Green Chemistry, 6 (9), 459–465. [87] De Miguel Bouzas, T., Barros-Velazquez, J. and Villa, T.G. (2006) Industrial applications of hyperthermophilic enzymes: a review. Protein & Peptide Letters, 13 (7), 645–651. [88] Machielsen, R., Uria, A.R., Kengen, S.W.M. and van der Oost, J. (2006) Production and characterization of a thermostable alcohol dehydrogenase that belongs to the aldo-keto reductase superfamily. Applied and Environmental Microbiology, 72 (1), 233–238. [89] Zhu, D., Malik, H.T. and Hua, L. (2006) Asymmetric ketone reduction by a hyperthermophilic alcohol dehydrogenase. The substrate specificity, enantioselectivity and tolerance of organic solvents. Tetrahedron: Asymmetry, 17 (21), 3010–3014. [90] Zhu, D., Rios, B.E., Rozzell, J.D. and Hua, L. (2005) Evaluation of substituent effects on activity and enantioselectivity in the enzymatic reduction of aryl ketones. Tetrahedron: Asymmetry, 16 (8), 1541–1546. [91] Zhu, D., Mukherjee, C., Rozzell, J.D. et al. (2006) A recombinant ketoreductase tool-box. Assessing the substrate selectivity and stereoselectivity toward the reduction of beta-ketoesters. Tetrahedron, 62 (5), 901–905. [92] Zhu, D., Mukherjee, C. and Hua, L. (2005) ‘Green’ synthesis of important pharmaceutical building blocks: enzymatic access to enantiomerically pure a-chloroalcohols. Tetrahedron: Asymmetry, 16 (19), 3275–3278. [93] Weenen, H., Boog, R.L.G.M. and Apeldoorn, W. (1995) Aldehydes from alcohols using alcohol oxidases and dehydrogenases. Colloques – Institut National de la Recherche Agronomique, 75 (Bioflavour 95), 375–380. [94] Dienys, G., Jarmalavicius, S., Budriene, S. et al. (2003) Alcohol oxidase from the yeast Pichia pastoris – a potential catalyst for organic synthesis. Journal of Molecular Catalysis B–Enzymatic, 21 (1–2), 47–49. [95] Karra-Chaabouni, M., Pulvin, S., Meziani, A. et al. (2003) Biooxidation of n-hexanol by alcohol oxidase and catalase in biphasic and micellar systems without solvent. Biotechnology and Bioengineering, 81 (1), 27–32. [96] Isobe, K. and Nishise, H. (1995) A new enzymic method for glycolaldehyde production from ethylene glycol. Journal of Molecular Catalysis B–Enzymatic, 1 (1), 37–43. [97] Isobe, K. and Nishise, H. (1994) Enzymic production of glyoxal from ethylene glycol using alcohol oxidase from methanol yeast. Bioscience Biotechnology and Biochemistry, 58 (1), 170–173. [98] Clark, D.S., Geresh, S. and DiCosimo, R. (1995) Enantioselective oxidation of 2-methyl-1-alkanols by alcohol oxidase from methylotrophic yeasts. Bioorganic & Medicinal Chemistry Letters, 5 (13), 1383–1388. [99] Siebum, A., van Wijk, A., Schoevaart, R. and Kieboom, T. (2006) Galactose oxidase and alcohol oxidase: scope and limitations for the enzymatic synthesis of aldehydes. Journal of Molecular Catalysis B–Enzymatic, 41 (3–4), 141–145. [100] Pezzotti, F. and Therisod, M. (2007) Enantioselective oxidation of thioanisole with an alcohol oxidase/ peroxidase bienzymatic system. Tetrahedron: Asymmetry, 18 (6), 701–704. [101] Pezzotti, F., Okrasa, K. and Therisod, M. (2005) Bienzymatic synthesis of chiral heteroaryl-methyl-sulfoxides. Tetrahedron: Asymmetry, 16 (16), 2681–2683. [102] Okrasa, K., Falcimaigne, A., Guibe-Jampel, E. and Therisod, M. (2002) Enantioselective synthesis of sulfoxides catalyzed by an oxidase–peroxidase bienzymatic system. Tetrahedron: Asymmetry, 13 (5), 519–522. [103] Guillen, F., Martinez, A.T. and Martinez, M.J. (1992) Substrate specificity and properties of the aryl-alcohol oxidase from the ligninolytic fungus Pleurotus eryngii. European Journal of Biochemistry, 209 (2), 603–611.


151

[104] Varela, E., Guillen, F., Martinez, A.T. and Martinez, M.J. (2001) Expression of Pleurotus eryngii aryl-alcohol oxidase in Aspergillus nidulans: purification and characterization of the recombinant enzyme. Biochimica et Biophysica Acta, Protein Structure and Molecular, Enzymology, 1546 (1), 107–113. [105] Van den Heuvel, R.H.H., Fraaije, M.W., Mattevi, A. et al. (2001) Vanillyl-alcohol oxidase, a tasteful biocatalyst. Journal of Molecular Catalysis B–Enzymatic, 11 (4–6), 185–188. [106] Van den Heuvel, R.H.H., Laane, C. and van Berkel, W.J.H. (2001) Exploring the biocatalytic potential of vanillyl-alcohol oxidase by site-directed mutagenesis. Advanced Synthesis and Catalysis, 343 (6–7), 515–520. [107] Franke, D., Machajewski, T., Hsu, C.-C. and Wong, C.-H. (2003) One-pot synthesis of L-fructose using coupled multienzyme systems based on rhamnulose-1-phosphate aldolase. The Journal of Organic Chemistry, 68 (17), 6828–6831. [108] Fialova, P., Namdjou, D.-J., Ettrich, R. et al. (2005) Combined application of galactose oxidase and beta-Nacetylhexosaminidase in the synthesis of complex immuno-active N-acetyl-D-galactosaminides. Advanced Synthesis and Catalysis, 347 (7 þ 8), 997–1006. [109] Kawagoshi, Y. and Fujita, M. (1997) Purification and properties of polyvinyl alcohol oxidase with broad substrate range obtained from Pseudomonas vesicularis var. povalolyticus PH. World Journal of Microbiology & Biotechnology, 13 (3), 273–277. [110] Seip, J.E., Fager, S.K., Gavagan, J.E. et al. (1993) Biocatalytic production of glyoxylic acid. The Journal of Organic Chemistry, 58 (8), 2253–2259. [111] Adam, W., Lazarus, M., Boss, B. et al. (1997) Enzymic resolution of chiral 2-hydroxy carboxylic acids by enantioselective oxidation with molecular oxygen catalyzed by the glycolate oxidase from spinach (Spinacia oleracea). The Journal of Organic Chemistry, 62 (22), 7841–7843. [112] Adam, W., Lazarus, M., Saha-Moller, C.R. and Schreier, P. (1998) Quantitative transformation of racemic 2-hydroxy acids into (R)-2-hydroxy acids by enantioselective oxidation with glycolate oxidase and subsequent reduction of 2-keto acids with D-lactate dehydrogenase. Tetrahedron: Asymmetry, 9 (2), 351–355. [113] Gavagan, J.E., Fager, S.K., Seip, J.E. et al. (1997) Chemoenzymic synthesis of N-(phosphonomethyl)glycine. The Journal of Organic Chemistry, 62 (16), 5419–5427.

8 Applications of Nitrile Hydratases and Nitrilases Grace DeSantis1 and Robert DiCosimo2 1

Biosite, An Inverness Medical Company, 9975 Summers Ridge Road, San Diego, CA 92121, USA 2 DuPont, Experimental Station, PO Box 80328, Wilmington, DE 19880-0328, USA

8.1 Introduction Existing synthetic methods and commercial processes that employ nitrile hydratases (NHases) and nitrilases continue to be improved by directed evolution of existing enzymes, or by the discovery of new enzymes with improved properties, and new applications of these catalysts have recently been described. Numerous reviews have previously been published that describe applications of NHase [1–6] and nitrilase [1,4–11], and in this review we present examples of new applications of these nitrile-utilizing catalysts from journal articles, patent applications, and issued patents that have been published in the past 2–3 years.

8.2 NHase 8.2.1 New NHases Many previously-characterized NHases have been shown to have poor temperature stability above ambient temperature. Geobacillus caldoxylosilyticus M16 [12], a thermophile that can be grown at 70 C, produces a thermostable NHase when induced during culture with valeronitrile, crotononitrile or crotonamide. The microbial NHase is reported to be active at temperatures up to 75 C. Hydration of a 1.1 wt% solution of acrylonitrile in phosphate buffer (50 mM, pH 7.7) at 20–75 C using 5 U mL1 Geobacillus caldoxylosilyticus NHase (activity measured at 10 C) produced acrylamide in quantitative yield. Similarly, 1.1 wt% solutions of adiponitrile, acetonitrile, isobutyronitrile, n-valeronitrile, n-butyronitrile, n-hexanenitrile, and benzonitrile were quantitatively converted to the corresponding amides or diamides at 30 C. Biocatalysis for the Pharmaceutical Industry edited by Junhua Tao, Guo-Qiang Lin, and Andreas Liese 2009 John Wiley & Sons Asia (Pte) Ltd

154


8.2.2 Applications 8.2.2.1 Acrylamide A Geobacillus thermoglucosidasius Q-6 NHase was expressed in Rhodococcus rhodochrous strain M33 [13]. The recovery of NHase activity after 30 min at 60 C, 70 C, and 80 C was 88%, 84%, and 32% respectively, and the optimum reaction temperature was about 60 C. There was no decrease in the microbial NHase activity of this transformant in reactions with increasing acrylonitrile concentrations up to 6% (w/v). In a batch reaction using unimmobilized cells as catalyst, an initial catalyst loading of 0.11 wt% dry cell weight (dcw) Rhodococcus rhodochrous M33/Q6 was used to convert a continuous feed of acrylonitrile (not exceeding 2% in the reaction mixture) to acrylamide at 20–25 C, where a final concentration of 52% (weight/solution) acrylamide was accumulated. Rhodococcus rhodochrous J-1 has been used for the industrial production of acrylamide from acrylonitrile, but the J-1 NHase is not very stable above 30 C. Mutant gene libraries of Rhodococcus rhodochrous J-1 and Rhodococcus rhodochrous M8 NHases have been prepared, where amino acid substitutions were made in both a and b subunits of the enzyme, including substitutions in the region around the cofactor binding domain of the a-subunit, and in the b-subunit. Mutant NHases with improved temperature stability, and improved stability in high concentrations of acrylamide produced by the hydration of 5 wt% acrylonitrile, were identified, and these mutant NHases were expressed in transformed Rhodococcus and Escherichia coli microbial host strains [14]. A Rhodococcus rhodochrous J-1 NHase with an Eb93G mutation was one of 16 NHase mutants expressed in Escherichia coli JM109 that showed a significant improvement in temperature stability at 50–60 C when compared with the NHase of the parent strain (Table 8.1). Eb93G and Nb167S mutations of the Rhodococcus rhodochrous M8 NHase had similarly improved temperature stabilities relative to the NHase of the parent strain. Additional testing of the mutant NHases for conversion of 5 wt% acrylonitrile in 30% acrylamide also demonstrated improved retention of stability or improvement in reaction rates, relative to the parent strain NHase under these reaction conditions. Table 8.1 Temperature stability of Rhodococcus rhodochrous J1 Eb93G mutant NHase compared with J1 NHase, each expressed in Escherichia coli JM109 NHase

J1 J1 Eb93G

Recovered activity (%) Untreated

50 C, 20 min

100 100

54 80

55 C, 20 min 9 97

60 C, 5 min 6 105

A process for production of high-purity acrylamide from acrylonitrile that has a significant concentration of acrolein impurity has been demonstrated, where the acrylonitrile contains 100 ppm acrolein [15]. The process utilizes microbial cell catalysts that convert acrylonitrile feeds having at least 10 ppm acrolein impurity directly to high-purity acrylamide with no detectable levels of acrolein, avoiding the need for purification of the acrylonitrile feed prior to conversion, or removal of acrolein from the product mixture. High-molecular-weight polymer with high intrinsic viscosity is produced from acrylamide free from by-products derived from acrolein. In one example, an acrylonitrile solution containing 50 ppm of acrolein was continuously fed at 25 C into a reaction mixture containing Rhodococcus

Applications of Nitrile Hydratases and Nitrilases

155

rhodochrous strain 2368, producing a solution containing about 50 wt% acrylamide with no detectable acrolein. Acrylonitrile produced industrially via propylene ammoxidation contains trace amounts of benzene. When using Pseudonocardia thermophila JCM3095 or Rhodococcus rhodochrous J-1 as microbial NHase catalyst for conversion of acrylonitrile to acrylamide, concentrations of benzene of 4 ppm produced a significant increase in the reaction rate [16]. Maintaining the concentration of HCN and oxazole at 5 ppm and 200 mM) with respect to 3-cyanopyridine. More desirable was a microbial catalyst with significantly improved temperature stability and a relatively low KM for 3-cyanopyridine. Rhodococcus sp. FZ4 [20] has a KM value of 80.5 mM for 3-cyanopyridine and a temperature optimum for NHase activity of 60 C. A comparison of temperature stability and the influence of substrate concentration on enzyme activity for these enzymes are presented in Tables 8.2 and 8.3 respectively. Corynebacterium glutamicum (CGMCC No.1464) cells immobilized in calcium alginate beads cross-linked with polyethenimine and glutaraldehyde have been employed for the production of nicotinamide from 3-cyanopyridine [21]. The reaction was run at 10–15 C,

156


Table 8.2 Comparison of the thermal stability of NHase activities of Rhodococcus sp. FZ4 and GF270 to Amycolatopsis sp. NA40 and Rhodococcus rhodochrous J1 Incubation time (min)

Relative activity (%)a

Incubation temp. ( C)

15 15 15 60 60 60 60 60 60

Rhodococcus sp. FZ4

Rhodococcus sp. GF270

Amycolatopsis sp. NA40

Rhodococcus rhodochrous J1

100 93 2 100 100 100 100 100 6

100 95 5 100 100 100 100 89 0

nd nd nd 100 95 80 32 0 0

100 80 0 nd nd nd nd nd nd

50 60 70 20 30 40 50 60 70

nd ¼ Not determined.

a

Table 8.3 Dependence of Rhodococcus sp. FZ4, Rhodococcus sp. GF270, Amycolatopsis sp. NA40 and Rhodococcus rhodochrous J1 NHase activities on 3-cyanopyridine concentration 3-Cyanopyridine (% w/v)

0 2.5 5.0 7.5 10.0

Rhodococcus sp. FZ4a

100 100 100 100 100

Relative activity (%) Rhodococcus sp. GF270a

Amycolatopsis sp. NA40b

100 100 100 100 100

100 74 56 47 16

Rhodococcus rhodochrous J1b 100 ndc 86 ndc 63

a

Incubation for 60 min. Incubation for 15 min. c Not determined. b

where an aqueous 3-cyanopyridine solution was added continuously at a gradually decreasing rate such that the total weight of 3-cyanopyridine added was 10–25 wt% of the reaction mixture; the concentration of unreacted 3-cyanopyridine at the conclusion of the reaction was 80–500 ppm. A thermally stable NHase from Comamonas testosteroni 5-MGAM-4D (ATCC 55 744) [22] was recombinantly expressed in Escherichia coli, and the resulting transformant cells immobilized in alginate beads that were subsequently chemically cross-linked with glutaraldehyde and polyethylenimine. This immobilized cell catalyst (at 0.5% dcw per reaction volume) was added to an aqueous reaction mixture containing 32 wt% 3-cyanopyridine at 25 C, and a quantitative conversion to nicotinamide was obtained. The versatility of this catalyst system was further illustrated by a systematic study of substrates, which included

157


acrylonitrile, methacrylonitrile, adiponitrile, butyronitrile, 3-hydroxyvaleronitrile, and glycolonitrile. Typically, reactions were completed in 1 h and reached 100% conversion even at nitrile concentrations up to 3 M [23]. Unimmobilized Corynebacterium propinquum (CGMCC No. 0886) cells containing a cobalt-dependent NHase were employed in either batch or continuous reactions for the production of nicotinamide from 3-cyanopyridine [24]. In the continuous process, membrane filtration separated precipitated product (5 wt%) and the microbial cell catalyst from the reaction mixture, where the catalyst was then recovered and returned to the reactor; using a continuous addition of aqueous 3-cyanpyridine to maintain substrate concentration at 20% (w/v), a final conversion of 99% was obtained. 8.2.2.4 a-Hydroxycarboxylic Acids/Amides The activities of NHases from Rhodococcus sp. Adp12 and Gordonia sp. BR-1 strains have been partially characterized [25]. In reactions that catalyze the hydration of a-hydroxynitriles such as lactonitrile or glycolonitrile, the substrate can dissociate to produce HCN and the corresponding aldehydes. HCN can inhibit and/or inactivate NHase, and it was determined that these two enzymes remain active in the presence of cyanide ion at concentrations up to 20 mM. The dependence of the NHase activity of cell-free extracts of Rhodococcus rhodochrous J1 and Gordonia sp. BR-1 on cyanide ion concentration is illustrated in Figure 8.1, demonstrating the improved cyanide stability of BR-1 NHase relative to that of J1.

relative activity (%)

120 100 80 60 40 20 0 0

5

10

15

20

25

KCN (mM)

Figure 8.1 Dependence of the NHase activity of cell-free extracts of Rhodococcus rhodochrous J1 (~) and Gordonia sp. BR-1 (&) on cyanide ion concentration

In instances where the nitrile has poor solubility in water, the addition of an organic cosolvent improved solubility and reaction rate. The recovered activity of Rhodococcus Adp12 NHase after 10 min in reaction mixtures containing 10–40% (v/v) organic co-solvent at 20 C was determined (Figure 8.2), demonstrating stability in a single-phase reaction mixture containing methanol or ethanol, or in a two-phase aqueous–organic mixtures containing ethyl acetate or n-hexane. Rhodococcus equi XL-1 has also been demonstrated to have superior stability in solutions containing up to 20 mM HCN when compared with several Rhodococcus erythropolis

158

Biocatalysis for the Pharmaceutical Industry 120

relative activity (%)

100 80 60 40 20 0 0

10

20

30

40

50

co-colvent (%)

Figure 8.2 Residual activity of Rhodococcus Adp12 NHase after 10 min in reaction mixtures containing 10–40% (v/v) methanol (&), ethanol (~), ethyl acetate (.) and n-hexane (~) at 20 C

microbial NHase catalysts [26], as shown in Table 8.4 for reactions run by first adding 0–20 mM KCN to the reaction mixture containing only enzyme and buffer. After the solution was incubated at 20 C for 30 min, the enzyme reaction was started with the addition of 3-cyanopyridine. 2-Hydroxy-4-methylthiobutyroamide (HMBAm, useful as a feed additive as a methionine substitute) was produced by suspending Rhodococcus equi XL-1 wet cells (1 g) in 25 g of phosphate buffer (0.05 M, pH 6.5) containing 2-hydroxy-4-methylthiobutyronitrile (HMTBN; 0.2 wt%) and NaCl (0.34 wt%). The resulting mixture was incubated at 30 C for 43 h with stirring, adding aliquots of HMTBN during the reaction. When the initial HMTBN concentration was 537 mM, the concentration of cyanide ion in the reaction solution was 2.33 mM. A total HMBAm concentration of 75 g L1 was produced. Table 8.4 Residual activity of Rhodococcus NHases after incubation in 1–20 mM cyanide for 30 min Strain

Rhodococcus Rhodococcus Rhodococcus Rhodococcus Rhodococcus Rhodococcus Rhodococcus

KCN (mM)

equi XL-1 erythropolis IFO12 539 erythropolis IFO12 540 erythropolis IFO12 567 erythropolis IFO12 320 erythropolis ATCC11 048 erythropolis ATCC33 278

0

1

5

10

15

20

100 100 100 100 100 100 100

100 51 92 82 21 54 61

100 13 79 62 2 30 36

98 9 39 47 1 18 42

94 5 16 16 0 10 14

89 4 13 11 0 7 11

Rhodococcus erythropolis NCIMB 11 540 has been employed as biocatalyst for the conversion of (R)- or (S)-cyanohydrins to the corresponding (R)- or (S)-a-hydroxycarboxylic acids with an optical purity of up to >99% enatiomeric excess (ee) [27–29]; the chiral cyanohydrins can separately be produced using hydroxynitrile lyase from Hevea braziliensis or from Prunus anygdalis [30]. Using the combined NHase–amidase enzyme system of the Rhodococcus erythropolis NCIMB 11 540, the chiral cyanohydrins were first hydrolyzed to the


159

chiral hydroxyamide, which was subsequently converted by the amidase into the corresponding chiral a-hydroxycarboxylic acid. This microbial biocatalyst was used to convert (R)-2chloromandelonitrile (>99% ee) to (R)-2-chloromandelic acid with a product ee of >99% (crude yield of 98%). In addition to the direct use of NHase–amidase of Rhodococcus cells, the enzymes have both been overexpressed in Escherichia coli in active form; in particular, the amidase can be expressed at levels significantly greater than found in the native Rhodococcus cells. 3-Hydroxyvaleric acid can be used as a substitute for e-caprolactone in the preparation of highly branched copolyesters [31]. Immobilization of Comamonas testosteroni 5-MGAM-4D ATCC 55 744 cells in alginate beads, followed by chemical cross-linking with glutaraldehyde and polyethylenimine, produced a catalyst with high specific activity and productivity for the conversion of 3-hydroxyvaleronitrile acid, 3-hydroxybutryonitrile, and 3-hydroxypropionitrile to the corresponding 3-hydroxycarboxylic acids in 99% yields [32,33]. In a series of 85 consecutive batch reactions with biocatalyst recycle, 670 g 3-hydroxyvaleric acid/g dcw was produced with an initial volumetric productivity of 44 g 3-HVA/(L h) and a final product concentration in each batch reaction of 118 g L1. 8.2.2.5 4-Methylthio-a-Hydroxybutryamide Isolated polynucleotide clusters from Rhodococcus opacus which encode four polypeptides possessing the activities of a NHase (a and b subunits), an auxiliary protein P15K that activates the NHase, and a cobalt transporter protein were expressed in Escherichia coli DSM 14 459 cells [34]. Methionine nitrile was added continuously to a suspension of the transformant cells (5.6% w/v of wet cells) in phosphate buffer (50 mM, pH 7.5) at 20 C, at a rate where the nitrile concentration did not exceed 15 g L1 while maintaining the pH constant at 7.5. After 320 min, the nitrile was completely converted into amide, corresponding to a final product concentration of 176 g L1. 4-Methylthio-a-hydroxybutyramide is readily hydrolyzed with calcium hydroxide, where the calcium salt of 4-methylthio-a-hydroxybutyric acid (MHA) can be directly used as a nutritional supplement in animal feed as an alternative to methionine or MHA. 8.2.2.6 Glycine High-purity glycine, useful as a food additive and as a raw material for synthesizing pharmaceuticals, agricultural chemicals, and detergents, was produced by hydrolysis of glycinonitrile using a microbial catalyst having nitrilase activity, or a combination of NHase and amidase activities [35]. By running the reaction in the absence of air or oxygen (limiting oxygen to 5 ppm or less, preferably less than 0.01 ppm using a continuous nitrogen purge) in the presence of a slight excess of ammonia, and by using a minimal concentration of added buffer to control pH, the production of organic impurities that inhibited the microbial enzyme was reduced, leading to an improvement in product purity. Microbial nitrile-hydrolyzing catalysts included Acinetobacter sp. AK226, Rhodococcus maris BP-479-9, Corynebacterium nitrilophilus ATCC 21 419, Alcaligenes faecalis IFO 13 111, Mycobacterium sp. AC777, Rhodopseudomonas spheroides ATCC 11 167 and Candida tropicalis ATCC 20 311. Acinetobacter sp. AK226 was preferred, having stable enzyme activity at temperatures up to 50 C, and producing as much as 461 g product/g dcw at a glycine production of 19 g/(g dcw h). The addition of reducing agents such as sodium sulfite, ascorbic acid, or L-cysteine resulted in a significant reduction in by-product

160


impurities and discoloration of glycine, where glycine yield and purity were as high as 100% and 99.99% respectively after recrystallization. 8.2.2.7 3,3,3-Trifluoro-2-Hydroxy-2-Methylpropionic Acid (S)-3,3,3-Trifluoro-2-hydroxy-2-methylpropionic acid (S)-2,2-HTFMPS) is an important intermediate for the preparation of therapeutic amides [36,37]. The racemic amide (S)-3,3,3trifluoro-2-hydroxy-2-methyl-propionamide ((R,S)-2,2-HTFMPA) was first prepared in quantitative yield by the hydration of 2-hydroxy-2-methyl-3,3,3-trifluoromethylpropionitrile using a mutant of Rhodococcus equi TG 328-2 that lacked amidase activity. The preparation of (R)2,2-HTFMPS and (S)-2,2-HTFMPA from the racemic amide was subsequently accomplished using Klebsiella oxytoca PRS1, Klebsiella oxytoca PRS1K17, Klebsiella planticula ID-624, Klebsiella pneumoniae ID-625, and an Escherichia coli transformant expressing the stereospecific amidohydrolase activity-derived Klebsiella oxytoca PRS1K17 [38,39]. The preparation of (S)-2,2-HTFMPS and (R)-2,2-HTFMPA from the racemic amide was performed using Pseudomonas sp. DSM 11 010, Rhodococcus opacus ID-622, Arthrobacter ramosus ID-620, and Bacillus sp. ID-621. For example, a Klebsiella oxytoca PRS1 cell suspension was used to hydrolyze 1.0 wt% (R,S)-2,2-HTFMPA in 0.05 M phosphate buffer (pH 8.0) at 40 C; after 5.5 h, (R)-2,2-HTFMPA was completely converted into the corresponding acid in 100% ee and 48% yield (Figure 8.3).

H3C OH H2N CF3 O (R,S)-2,2-HTFMPA

Klebsiella oxytoca PRS1 50 mM KH2PO4 (pH 8.0), 40 oC

H3C OH H2N CF3 O (S)-2,2-HTFMPA

+

H3C OH HO CF3 O (R )-2,2-HTFMPS, 48% yield, 100% ee

Figure 8.3 Preparation of (R)-2,2-HTFMPS and (S)-2,2-HTFMPA from racemic 3,3,3-trifluoro-2hydroxy-2-methyl-propionamide using Klebsiella oxytoca PRS1

8.2.2.8 (S )-3-(Thiophen-2-Ylthio) Butanoic Acid After all attempted chemical methods of nitrile hydrolysis failed, a suitable nitrilase catalyst was identified for the conversion of (S)-3-(thiophen-2-ylthio) butanenitrile to the corresponding acid, a building block for Mercks MK-00 507 carbonic anhydrase inhibitor, with trade name Dorzolamine (Figure 8.4). A screen of 53 strains revealed 12 that provided >75% conversion of the nitrile. Of these, one yielded the desired acid product at high conversion levels and one yielded the amide product (strain apparently has NHase but lacks amide hydrolase activity). These strains are Brevibacterium A4 and Brevibacterium R312 pYG811b (reclassified as Rhodococcus erythropolis) respectively. The recombinant Rhodococcus erythropolis strain (formerly Brevibacterium R312 pYG811b) was applied at pH 7.0, 30 C at a substrate loading of 5 mg mL1 (although up to 30 mg mL1 was shown to be tolerated) in the presence of 3% acetonitrile as co-solvent (acetonitrile was confirmed not to be a substrate of this biocatalyst) to generate (S)-3-(thiophen-2-ylthio) butanoic acid on gram scale. Under these conditions, 60% yield of product was generated after 5 days incubation. Contaminating amide

161

Applications of Nitrile Hydratases and Nitrilases H NCH2CH3

O CN

COOH nitrilase

S

S

S

S

S

S OO

S

S

SO2NH2

Dorzolamine

Figure 8.4 Nitrilase-catalyzed conversion of (S)-3-(thiophen-2-ylthio)butanenitrile to (S)-3,3,3trifluoro-2-hydroxy-2-methyl-propionamide, an intermediate in the synthesis of carbonic anhydrase inhibitor Dorzolamine

was reported at 5%. Since the stereogenic center was set earlier in the synthesis, an absolute requirement for enantioselectivity of the biocatalyst was not necessary in this example [40]. 8.2.2.9 Malonic Acid Derivatives The NHase and amidase from Rhodococcus rhodochrous IFO 15 564 was studied using a series of a,a-disubstituted malononitriles. This amidase preferentially hydrolyzes the pro (R) amide of the prochiral di-amide, which is an intermediate resulting from the nonenantiotopic NHase activity on the dinitrile substrate. This transformation was combined with a Hofmann rearrangement to generate a key precursor of (S)-methyldopa in 98.2% ee and 95% yield (Figure 8.5) [41]. O O

O CN R. Rhodochrous CN

O

O CNH2 COOH

CH2N2

O CNH2 O COOCH3 98.2% ee, 95% overall yield

O

Br2, MeONa/MeOH Hofmann Rearrangement

HO NH2 HO COOH (S)-α-methyldopa

O O

NHCOOCH3 COOCH3

Figure 8.5 Conversion of 2-(1,3-benzodioxol-5-ylmethyl)-2-methyl-propanedinitrile to (R)-a(aminocarbonyl)-a-methyl-1,3-benzodioxole-5-propanoic acid using the non-enantiotopic NHase activity and enantioselective amidase activity of Rhodococcus rhodochrous IFO 15564

8.2.2.10 Cyclopropane Carboxylic Acid Derivatives By screening 53 Rhodococcus and Pseudomonas strains, an NHase–amidase biocatalyst system was identified for the production of the 2,2-dimethylcyclopropane carboxylic acid precursor of the dehydropeptidase inhibitor Cilastatin, which is used to prolong the antibacterial effect of Imipenem. A systematic study of the most selective of these strains, Rhodococcus erythropolis ATCC25 544, revealed that maximal product formation occurs at pH 8.0 but that ee decreased above pH 7.0. In addition, significant enantioselectivity decreases were observed above 20 C. A survey of organic solvent effects identified methanol (10% v/v) as the

162


CN H3C H3C

HOOC H R. erythropolis

H CN +

pH 7, 20 C, 64 h 10% v/v methanol

H3C H3C 82% ee, 45 % yield

H3C H3C

(S)-2,2-dimethylcyclopropane carboxylic acid

Figure 8.6

Preparation of (S)-2,2-dimethylcyclopropane carboxylic acid, a precursor of Cilastatin

co-solvent providing the greatest enantioselectivity. Under the optimized conditions reported, the (S)-2,2-dimethylcyclopropane carboxylic acid was produced with an observed ee of 82% and with overall conversion yield of 45% (Figure 8.6) [42]. NHase from Rhodococcus. sp. AJ270 was isolated, purified, and applied to the enantioselective transformation of a series of cyclopropane carbonitriles. Amides with moderate ee were isolated from conversion of many of the cyclopropane substrates, to yield the amides: trans-(1R, 2R)-3-phenylcyclopropane carbonitrile (49% conv. 22.7% ee), trans-(1S, 3S)-2,2-dimethyl-3phenylcyclopropanecarbonitrile (40% conv. 84.7% ee), trans-(1R, 3R)-2,2-dibromo-3-phenylcyclopropanecarbonitrile (11.6% conv. 83.8% ee), cis-(1R, 2S)-3-phenylcyclopropanecarbonitrile (25.8% conv. 95.4% ee), and cis-(1R, 2S)-2,2-dimethyl-3-phenylcyclopropanecarbonitrile (7.9% conv. 3.2% ee) [43]. 8.2.2.11 Oxirane Carboxylic Acid Derivatives Rhodococcus sp. AJ270 was applied to the transformation of a number of racemic cis- and trans-3-aryl-2-methyloxiranecarbonitriles (Figure 8.7). In all cases, the NHase activity proceeded very rapidly and with poor enantioselectivity. In contrast, the amidase activity was strongly dependent upon substrate structure. In general, the biocatalyst displays a strong preference for the unsubstituted phenyl side chain or para-substituted phenyl side chain compared with ortho- or meta-, and this is manifest both with respect to observed conversion and rate and also observed enantioselectivity. In contrast, the biotransformations of

Ar

O CN CH3

Ar

Rhodococcus sp. AJ270 phosphate buffer, pH 7.25, 30 C

O

Ar

CONH2 CH3

2S,3R-enantioselective amidase

O

CONH2 + Ar CH3

2R, 3S-amide racemic nitrile racemic amide trans-2-methyl-3-phenyloxiranecarbonitrile Summary of selected biotransformations Ar = C 6 H5 45% yield, Ar = 4-F -C6H5 31% yield, Ar = 4-Cl-C6H5 49% yield, Ar = 3-Cl-C6H5 32% yield, Ar = 2-Cl-C6H5 40% yield, Ar = 4-Me-C6H5 31% yield, Ar = 2-Me-C6H5 44% yield, Ar = 3,4-OCH 2O-C6H5 32% yield,

O COOH

CH3 2S, 3R-acid

> 99.5 % ee > 99 % ee > 99.5 % ee > 20 % ee < 5 % ee > 99.5 % ee < 5 % ee 50 % ee

Figure 8.7 Transformation of trans-3-aryl-2-methyloxiranecarbonitriles using the combined NHase and amidase activities of Rhodococcus sp. AJ270


163

cis-2-methyl-3-phenyloxiranecarbonitrile substrates proceeded sluggishly. In addition, the authors report that, for the 2,3-dimethyl-3-phenyloxiranecarbonitrile series of substrates (either cis or trans), much lower enantioselectivities were observed. The 2R,3S-amide products thus produced may be further transformed to access a-methylated serine and isoserine derivatives [44]. In a subsequent study, the authors established that immobilization of this biocatalyst in alginate capsules permitted efficient reuse of the catalyst and tolerated 5% acetone or methanol, but not ethyl acetate [45]. 8.2.2.12 NHases for Bioremediation Three cyanide-degrading nitrilases were recently cloned and purified and their kinetic profiles were evaluated in order to better understand their applicability to cyanide bioremediation. CynD from Bacillus pumilus C1 and DyngD from Pseudomonas stutzeri exhibit fairly broad pH profiles with >50% activity retained across pH 5.2 to pH 8.0 while the CHT (NHase) from Gloeocercospora sorghi exhibited a more alkaline pH activity profile with almost all of its activity retained at pH 8.5, slightly lower thermal tolerance, and quite different metal tolerance compared with the two bacterial enzymes [46]. 2,6-Dichlorobenzonitrile (dichlobenil) is the active ingredient in herbicides Prefix G and Casoron G. In soil, dichlobenil is degraded to the persistent metabolite 2,6-dichlorobenzamide (BAM) by several common soil bacteria. BAM is soluble and readily leached into groundwater and has a low to moderate toxicity with an LD50 of 1144–2300 mg kg1 in mice. Analysis of a series of common soil bacteria, including several known to express nitrilases, did not reveal any which could degrade dichlobenil to its diacid. Variovorax sp. is known to degrade the nonhalogenated analogue benzamide. Apparently, the steric hindrance created by the orthochlorosubstituents makes this substrate unacceptable to any amidases or nitrilases expressed by common soil bacteria tested thus far [47]. Similarly, the selective herbicides, bromoxynil (3,5-dibromo-4-hydroxybenzonitrile) and ioxynil (3,5-diiodo-4-hydroxybenzonitrile) are degraded by soil bacteria to their corresponding amideproducts3,5-dibromo-4-hydroxybenzamide(BrAM)and3,5-diiodo-4-hydroxybenzamide (IAM) but are not further degraded to the corresponding acids. The identification of amidases or nitrilases able to effect these transformations, in a soil bacterium, would be of value as a bioremediation agent [48].

8.3 Nitrilase 8.3.1 New Nitrilases 8.3.1.1 Bradyrhizobium japonicum Bradyrhizobium japonicum USDA110 is a Gram-negative nitrogen-fixing microbe that expresses a nitrilase (bll6402) whose function may be to detoxify and utilize hydroxynitriles produced in the metabolism of cyanogenic glycosides. The nitrilase gene was cloned and expressed in Escherichia coli [49], and the nitrilase had high activity toward mandelonitrile, with a Vmax and Km of 44.7 U mg1 and 0.26 mM respectively. Similarly, bll6402 also provided high conversion of phenyl hydroxyl acetonitrile, but in neither case was any enantioselectivity observed [50]. Despite the apparent lack of selectivity for a-substituted nitriles, nitrilase bll6402 catalyzed the enantioselective hydrolysis of aromatic b-hydroxynitriles to give

164


Table 8.5 Enantioselective hydrolysis of b-hydroxynitriles catalyzed by nitrilase bll6402 b-Hydroxynitrile

Recovered nitrile (R)-1 HO

H

H

HO

CN

CO2H X

X

yield (%)a X ¼ 4-H X ¼ 4-F X ¼ 4-Cl X ¼ 4-CH3 X ¼ 4-OCH3 X ¼ 2-OCH3 X ¼ 3-OCH3 X ¼ 2-Cl X ¼ 2,4-Cl2

41 37 40 35 57 46 46 40 42

Eb

Product acid (S)-2

ee (%)

yield (%)a

ee (%)

53 74 53 76 66 75 67 75 37

36 38 32 40 27 36 35 32 39

48 60 65 42 90 43 91 84 59

5 9 8 5 43 5 52 27 13

a

Isolated yield. Enantiomeric ratio.

b

(S)-enriched b-hydroxycarboxylic acids with recovery of (R)-enriched b-hydroxynitriles (Table 8.5) [51,52]. It also selectively hydrolyzed some a,v-dinitriles to v-cyanocarboxylic acids [53]; for example, nitrilase bll6402 hydrolyzes 1-cyanocyclohexaneacetonitrile to 1-cyanocyclohexaneacetic acid (88% isolated yield), a precursor for the antidepressant gabapentin (Figure 8.8). This same gabapentin precursor has also been prepared using microbial nitrilase catalysts such as Acidovorax facilis 72W and Escherichia coli SS1001 (a transformant expressing the Acidovorax facilis 72W nitrilase) [54], where quantitative conversion of the dinitrile with 100% regioselectivity to the desired product was obtained. CN CN

CO2H nitrilase bll6402 100 mM KH2PO4 (pH 7.2) 30 oC; 88% yield

CN

CO2H NH2

gabapentin

Figure 8.8 Nitrilase bll6402-catalyzed hydrolysis of 1-cyanocyclohexaneacetonitrile to 1-cyanocyclohexaneacetic acid, a precursor for the antidepressant gabapentin

8.3.1.2 Exophiala oligosperma A new nitrilasewas discovered from the black fungus Exophiala oligosperma (established by 18S rRNA) by growth enrichments performed with glucose and phenylacetonitrile as the sole carbon and nitrogen sources respectively under acid conditions. The novel nitrilase could convert phenylacetonitrile to phenylacetamide. Resting cells of Exophiala oligosperma exhibit nitrilase activity in the range pH 2–9, and have an activity optimum at pH 8–9. In addition, single chloroand hyroxy-phenylacetonitrile derivatives were also converted to the corresponding acids at comparable rates. In contrast to nitrilases from the fungi Aspergillus nidulans and Fusarium


165

solanii, this nitrilase was not effective in hydrolysis of benzonitrile. Since phenylacetamide was not observed as an intermediate in the conversion of phenylacetonitrile, and since the amide substrate was hydrolyzed at a rate threefold less than the nitrile substrate, the authors concluded that the activity is a true nitrilase rather than nitrile hydrolase–amidase combination [55]. 8.3.1.3 Streptomyces sp. Thermally stable nitrilase from Streptomyces sp. MTCC 7546 was induced by benzonitrile enrichment. While discovered by induction with aromatic nitrile, the enzyme was shown to exhibit a strong preference for aliphatic nitriles, with as high as 30-fold greater activity with aliphatic substrates compared with benzonitrile. The enzyme displays optimal activity at pH 7 and 50 C [56]. 8.3.1.4 Aspergillus niger A novel nitrilase was purified from Aspergillus niger K10 cultivated on 2-cyanopyridine. It was found to be homologous to a putative nitrilase from Aspergillus fumigatus Af293. The nitrilase exhibited maximum activity at 45 C and pH 8.0 with much less activity observed at slightly acid pH. Its substrate preference was for 4-cyanopyridine, benzonitrile, 1,4-dicyanobenzene, thio-phen-2-acetonitrile, 3-chlorobenzonitrile, 3-cyanopyridine, and 4-chlorobenzonitrile. ()-2-Phenylpropionitrile was only poorly converted by this enzyme and with minimal enantioselectivity. The enzyme was shown to be multimeric (>650 kDa) and be stabilized in the presence of sorbitol and xylitol [57]. 8.3.1.5 Pyrococcus abyssi A nitrilase from the hyperthermophile Pyrococcus abyssi, which exhibits optimal growth at 100 C, was cloned and overexpressed. Characterization of this nitrilase revealed that it is operational as a dimer (rather than the more common multimeric structure for nitrilases), with optimal pH at 7.4 and optimal apparent activity at 80 C with Tm (DSC) at 112.7 C. The substrate specificity of the nitrilase is narrow and it does not accept aromatic nitriles. The nitrilase converts the dinitriles fumaronitrile and malononitrile to their corresponding mononitriles [58]. 8.3.1.6 Pseudomonas putida An enantioselective nitrilase from Pseudomonas putida isolated from soil cultured with 2 mM phenylacetonitrile was purified and characterized. This enzyme is comprised of 9–10 identical subunits each of 43 kDa. It exhibits a pH optimum at 7.0 and a temperature optimum at 40 C (T1/2 ¼ 160 min) and requires a reducing environment for activity. This nitrilase was shown to have an unusually high tolerance for acetone as co-solvent, with >50% activity retained in the presence of 30% acetone. The kinetic profile of this nitrilase reveals: KM ¼ 13.4 mM, kcat/KM ¼ 0.9 s1 mM1 for mandelonitrile, KM ¼ 3.6 mM, kcat/KM ¼ 5.2 s1 mM1 for phenylacetonitrile, and KM ¼ 5.3 mM, kcat/KM ¼ 2.5 s1 mM1 for indole 3-acetonitrile. Preliminary analysis of this enzyme with 5 mM mandelonitrile revealed formation of (R)-mandelic acid with 99.9% ee [59]. A systematic study of the substrate specificity profile of this nitrilase (Table 8.6) illustrated that arylacetonitriles, including phenylacetonitrile derivatives indole-3-acetonitrile and

166 Table 8.6


Summary of substrate specificity of Pseudomonas putida nitrilase (data is not comprehensive)

Structural class

Structure

Substrate

Aliphatic nitrile

CH3CN CH3(CH2)2CH2CN CH2¼CHCN CH2¼CH(CH2)3CN NC(CH2)CN NC(CH2)4CN

Acetonitrile Valeronitrile Acrylonitrile 6-Hexene nitrile Malononitrile Adiponitrile

2.1 10.5 4.0 8 6.8 17.2

Cyclopropane carbonitrile Cyclohexane carbonitrile

4.1 12.5

Benzonitrile

12.7

CN

2-Hydroxy benzonitrile 4-Hydroxy benzonitrile

2.3 23.1

CN

2-Cyanopyridine

Unsaturated nitriles Dinitrile

CN

Cyclic nitrile CN

Aromatic nitriles

Relative activity (%)

OH, H

Heterocyclic nitriles

136.7

N

H N

( )n n=1,2 NC S

CN

Indole 3-carbonitrile (n ¼ 1) Indole 3-acetonitrile (n ¼ 2)

15.7 320

2-Thiophene acetonitrile

136.7

Mandelonitrile

100

Phenyl glycinenitrile

192.5

Phenyl acetonitrile

342.3

4-Hydroxy phenyl acetonitrile 2-Chloro phenyl acetonitrile

435 76.1

OH

Aryl acetonitriles

CN NH2 CN

CN

CN OH, Cl

Phenyl-substituted aliphatics

CN

CN

3-Phenyl propionitrile

22.8

4-Phenyl butyronitrile

7.6

167


2-thiopheneacetonitrile, are suitable substrates. Interestingly, para substitution of the phenyl ring dramatically increased activity, such that p-chloro and p-amino phenylacetonitrile exhibited higher activity than phenylacetonitrile. However, ortho substituents decreased activity. In addition, longer chain phenyl nitriles were less preferred by the enzyme: phenylacetonitrile > 3-phenyl propionitrile > 4-phenyl butyronitrile. Substitution of phenylacetonitrile at 2-position (for example, OH, NH2, CH3) also resulted in a decrease in activity. 8.3.1.7 Alcaligenes sp. A newly isolated nitrilase producer, Alcaligenes sp. ECU0401, was isolated from soil using acetonitrile as the sole nitrogen source. (R)-Mandelic acid, the chiral building block for production of anti-obesity agents, antitumor agents, penicillins, semisynthetic cephalosporins and used as a chiral resolving agent, was produced from racemic mandelic acid at pH 7.0 at 20 mM substrate loading in 12% isolated yield and 99.9% ee using this biocatalyst. Mandelamide was not observed as a side product [60].

8.3.2 Applications 8.3.2.1 Cyanobenzoic Acids Rhodococcus sp. ATCC 39 484 converts o-phthalonitrile, isophthalonitrile, or terephthalonitrile to o-cyanobenzoic acid, m-cyanobenzoic acid or p-cyanobenzoic acid respectively as the major reaction product, while at the same time producing the corresponding cyanobenzamide and phthalic acid monoamide as unwanted by-products [61]. By-product formation was found to be dependent on an NHase–amidase hydrolysis pathway that coexists competitively with a nitrilase pathway in the microbial catalyst. Chemical mutagenesis was employed to produce a variant of the parent strain (Rhodococcus sp. SD826) lacking the by-product-forming pathway, whereby cyanobenzamide and phthalic acid monoamide production were significantly reduced. A comparison of reaction products obtained from isophthalonitrile using the parent strain and the SD826 variant is shown in Table 8.7, demonstrating a reduction in production of m-cyanobenzamide and isophthalic acid monoamide of 85% and 82% respectively when using the SD826 strain; a similar improvement in selectivity to cyanoacid was obtained with terephthalonitrile. A microbial catalyst with improved nitrilase-specific activity relative to the SD826 variant was produced by the cloning and expression of the nitrilase gene in an Escherichia coli transformant.

Table 8.7 Reduction of byproduct formation by elimination of NHase/amidase pathway in Rhodococcus sp. ATCC 39 484 Strain

ATCC39 484 SD826

m-Cyanobenzoic acid

m-Cyanobenzamide

Conc. (%)

Conversion (mol%)

Conc. (%)

5.638 5.721

98.22 99.67

0.023 0.003

Conversion (mol%) 0.40 0.06

Isophthalic acid monoamide Conc. (%) 0.087 0.016

Conversion (mol%) 1.34 0.24

168


The regioselectivity of a Rhodococcus rhodochrous nitrilase has been demonstrated for the conversion of 5-fluoro-1,3-dicyanobenzene to 5-fluoro-3-cyano-benzoic acid [62]. The nitrilase was expressed in an Escherichia coli transformant, and a cell-free extract was employed as catalyst (0.14 wt% cell-free extract) in 0.1 M sodium phosphate buffer (pH 7.2) at 25 C containing 0.18 M 5-fluoro-1,3-dicyanobenzene. After 72 h, the conversion was >98% and the reaction was stopped by addition of phosphoric acid (pH 2.4) to yield 5-fluoro-3-cyano-benzoic acid as a crystalline product (97% isolated yield). 8.3.2.2 Glycolic Acid When used for the treatment of recalcitrant melasma [63], and as a monomer in the preparation of polyglycolic acid for dissolvable sutures [64], drug-delivery materials [65,66], and gasbarrier packaging materials [67], a high-purity glycolic acid is required. A chemoenzymatic process for the production of high-purity glycolic acid has been developed (Figure 8.9), starting with the reaction of formaldehyde and hydrogen cyanide to produce glycolonitrile in >99% yield and purity [68]. The resulting aqueous glycolonitrile was subsequently converted without further purification to ammonium glycolate using a high-activity biocatalyst based on Acidovorax facilis 72W nitrilase, where protein engineering and optimized protein expression in an Escherichia coli transformant host were used to improve microbial nitrilase specific activity by 33-fold compared with the wild-type culture [69]. A biocatalyst productivity of >1000 g glycolic acid/g dcw was achieved using a glutaraldehyde/polyethylenimine crosslinked carrageenan-immobilized Escherichia coli MG1655 transformant expressing the Acidovorax facilis 72W Phe168Val nitrilase mutant, where 3.2 M ammonium glycolate was produced in consecutive batch reactions with biocatalyst recycle, or in a continuous stirredtank reactor [70–73]. Direct conversion of the unpurified ammonium glycolate product solution to high-purity (>99% pure) aqueous glycolic acid was accomplished by fixed-bed ion exchange (IEX). Glycolic acid has also been produced from glycolonitrile using microbial biocatalysts such as Acinetobacter sp. AK226 [74], Corynebacterium propinquum [75], and Brevibacterium casei (CGMCC No. 0887) [76].

NaOH HCHO + HCN > 99%

HO H2C CN

nitrilase, H2 O > 99%

HO O IEX H2C C O- NH4+

HO O H2C C OH

Figure 8.9 Chemoenzymatic process for production of high-purity glycolic acid employing Acidovorax facilis 72W Phe168Val nitrilase

8.3.2.3 2-Hydroxy-4-Methylthiobutyrate Ammonium Salt The nitrile-hydrolyzing activity of Arthrobacter spp. NSSC 104 was shown to be resistant to the suppressing effect of a-hydroxy nitriles such as lactonitrile and HMTBN, and accumulated the corresponding a-hydroxy acid ammonium salt at a high concentration [77]. HMTBN (200 mM) was added to a suspension of Arthrobacter spp. NSSC 104 cells (4% dcw) in phosphate buffer (0.1 M, pH 7.5) and mixed at 30 C; seven more additions of the same amount of HMTBN were addedat1 hintervals,then afurthereightadditionsmadeat1.5 h intervalsoveratotalreactiontime of 19 h. At completion of the reaction, the concentration of 2-hydroxy-4-methylthiobutyrate


169

ammonium salt (HMTBS) was 49 wt% (96% yield). In a series of consecutive batch reactions at 30 C with recycle of the Arthrobacterspp. NSSC 104unimmobilized cells (3.2% dcw), HMTBN was continuously fed to the cell suspension in water (no added buffer, pH maintained at 7.4–7.6 by addition of aqueous ammonia) to produce HMTBS at 36 wt% (96–97% yield) in each of 10 consecutive batch reactions. The gene coding for the nitA nitrilase from Arthrobacter sp. NSSC104 has been cloned and expressed in Escherichia coli [78]. The conversion of HMTBN to HMTBS has been performed using an immobilized microbial cell catalyst [79]. An Escherichia coli W strain transformant (BIOCAT 714) that expressed the Alcaligenes faecalis ATCC 8750 nitrilase was suspended in phosphate buffer (pH 8.0) (12% dcw final concentration), then glutaraldehyde (6 wt% solution, 0.5 wt% final concentration) and polyazetidine (Kymene 617 solution, 12.5 wt%, 2.4 wt% final concentration) were sequentially added to the cell suspension, and the resulting mixture sprayed onto 2.0 mm alumina beads. The resulting catalyst contained 25.5% by weight of dry cells, and the thickness of the coating was 330 mm. The activity of this catalyst was 0.56 kg of HMTBN converted to HMTBS per hour and per kilogram of catalyst (25 C, pH 6.6, 0.1 M HMTBN). The immobilized microbial catalyst was charged to a thermostatic column reactor maintained at 35 C and fitted with a pump via a recirculation loop, and a 95% conversion of HMTBN to HMTBS (25 wt% of product solution) was achieved. The immobilized catalyst half-life in the presence of 0.2 M HMTBN was 30 h. An electrodialysis unit and a means of concentrating the final product of the reaction have also been described [80]. 8.3.2.4 Acrylic Acid A Corynebacterium propinquum microbial cell catalyst was employed to convert acrylonitrile to ammonium acrylate, where the final concentration of product was 10–20% and the concentration of unconverted acrylonitrile was 150), the level of activity rate was inadequate. The authors report a threefold enhancement in activity without sacrifice in

175


CN ISBN

CN

CN

nitrilase

+

COOH

CN

CN

(S)-3-cyano-5-methyl hexanoic acid (R)-isobutylsuccinonitrile

NH2 COOH Pregabalin (LyricaTM)

Figure 8.14 Preparation of (S)-3-cyano-5-methyl hexanoic acid from isobutylsuccinonitrile using a regio- and stereo-selective nitrilase from AtNit1 Arabidopsis thaliana

enantioselectivity by a single mutation C236S which was identified by subjecting the cloned gene to error-prone polymerase chain reaction-based mutagenesis and screening [92]. 8.3.2.10 (R)-4-Cyano-3-Hydroxy-Butyrate A summary of the industrial-scale process development for the nitrilase-catalyzed [93] route to ethyl (R)-4-cyano-3-hydroxy-butyrate, an intermediate in the synthesis of Atorvastatin (Pfizer Lipitor) from epichlorohydrin via 3-hydroxyglutaronitrile (3-HGN) was recently reported (Figure 8.15) [94]. The reaction conditions were further optimized to operate at 3 M (330 g L1) substrate, pH 7.5 and 27 C. Under these conditions, 100% conversion and product ee of 99% was obtained in 16 h reaction time with a crude enzyme loading of 6% (based on total protein, 0.1 U mg1). It is noted that at pH < 6.0 the reaction stalled at 99% conversion) by microbial Nit338 biocatalyst in sodium phosphate buffer (10 mM, pH 7.5) at 40 C, and the isolated endo-5-norbornene-2-carboxylic acid was >99% pure.

H CN

Alcaligenes faecalis Nit338

H

10 mM NaH2PO4 (pH 7.5), 40 oC

CO2H

endo-5-norbornene2-carbonitrile

endo-5-norbornene2-carboxylic acid

Figure 8.16 Conversion of endo-5-norbornene-2-carbonitrile to the corresponding carboxylic acid using Alcaligenes faecalis Nit338

8.3.2.12 Malonic Acid Monoesters The preparation of malonic acid monoesters has been demonstrated using the microbial nitrilase activity of Corynebacterium nitrilophilus ATCC 21 419, Gordona terrae MA-1, or Rhodococcus rhodochrous ATCC 33 025 to hydrolyze methyl cyanoacetate, ethyl cyanoacetate, n-propyl cyanoacetate, isopropyl cyanoacetate, n-butyl cyanoacetate, tertbutyl cyanoacetate, 2-ethylhexyl cyanoacetate, allyl cyanoacetate, and benzyl cyanoacetate [96]. By maintaining the concentration of nitrile in a reaction mixture at 5 wt%, significant inactivation of the nitrilase activity was avoided; for example, a total of 25 g of n-propyl cyanoacetate was added in sequential 5 g portions to a 100 mL suspension of Rhodococcus rhodochrous ATCC 33 025 cells (OD630 ¼ 5.6) in 50 mM phosphate buffer (pH 7.0) over 30 h at 25 C to produce mono-n-propyl malonate in 100% yield (Figure 8.17).

177

Applications of Nitrile Hydratases and Nitrilases R. rhodochrous ATCC 33025 NCCH2CO2CH2CH2CH3

50 mM phosphate (pH 7.0), 25 oC

HO2CCH2CO2CH2CH2CH3

Figure 8.17 Production of mono-n-propyl malonate from n-propyl cyanoacetate using the microbial nitrilase activity of Rhodococcus rhodochrous ATCC 33025

8.3.2.13 Prochiral Sulfoxide Resolution A prochiral bis(cyanomethyl) sulfoxide was converted into the corresponding mono-acid with enantiomeric excesses as high as 99% using a nitrilase–NHase biocatalyst. The whole-cell biocatalyst Rhodococcus erythropolis NCIMB 11 540 and a series of commercially available nitrilases NIT-101 to NIT-107 were evaluated in this study. As outlined in Figure 8.18, the prochiral sulfoxide may be transformed into five different products (plus enantiomeric isoforms), of which, three are chiral (A, B, and C) and two achiral (D and E). Only products A, B, and E were observed with the biocatalysts employed in this investigation. Both enantiomerically enriched forms of both A and C could be obtained with one of the catalysts used. The best selectivities are as follows: (S)-A 99% ee, (R)-A 33% ee, (S)-C 66% ee, and (R)-C 99% ee, using NIT-104, NIT-103, NIT-108, and NIT-107 respectively. Each of these catalysts produced more than one product and, thus, they are not chemoselective [97]. chiral -

NC

O .. O S+ CNH2 A

-

NC

O .. S+ CN

achiral O -O .. O S+ CNH2 H2NC D

-

nitrilase or nitrile hydratase

NC

O .. O S+ COH B

O -O .. O H2NC S+ COH C

O -O .. O HOC S+ COH E

Figure 8.18 Conversion of a prochiral bis(cyanomethyl) sulfoxide into the corresponding mono-acid using a nitrilase–NHase biocatalyst

8.3.2.14 Polymer Modification Recently, nitrilases have been applied to polymer modification, specifically to the modification of polyacrylonitrile (PAN). Nearly 3 106 tons of PAN are produced per annum and used in the textile industry. However, there is a great need to improve moisture uptake, dyeability with ionic dyes, and feel of this acrylic fiber. The cyano moieties of PAN have been successfully modified to carboxylates with the commercial Cyanovacta nitrilase, thus enhancing the aforementioned properties of PAN [98]. Nitrilase action on the acrylic fabric was improved

178


by addition of 1 M sorbitol and 4% N,N-dimethylacetamide, which serve to make the fiber more plastic and, thus, increase surface accessibility. In addition, using PAN polymers as a sole carbon source, a new nitrilase was discovered from Micrococcus luteus strain BST20, which was also able to modify PAN fabric [99].

8.4 Conclusions NHases and nitrilases each continue to find use in reactions where significant improvements in substrate conversion, product yield and purity, enantioselectivity, and/or regioselectivity are obtained relative to alternative nonenzymatic reactions. The high substrate conversions and product yields and purities achieved with these enzymes often drastically reduce or eliminate the need for downstream purification steps, leading to the commercialization of more costefficient processes with less by-product formation and waste disposal requirements. The use of genetic engineering to modify protein structure, through random mutagenesis or by design, has now routinely been shown to improve the desired functionality of these enzymes. One disadvantage of using nitrilases to produce carboxylic acids remains the co-production of one equivalent of ammonia, and conversion of the initially produced ammonium carboxylate to the desired carboxylic acid often results in the production of an inorganic ammonium salt as byproduct; finding ways to recycle this by-product stream in the process, or to identify economically attractive and sustainable uses for by-product streams, remains as a challenge to adoption of nitrilase-catalyzed production of large-volume commodity chemicals.

References [1] DiCosimo, R. (2007) Nitrilases and nitrile hydratases, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press LLC, Boca Raton, FL, Chapter 1. [2] Yamada, H., Shimizu, S. and Kobayashi, M. (2001) Hydratases involved in nitrile conversion: screening, characterization and application. Chemical Record, 1, 152–161. [3] Martinkova, L. and Mylerova, V. (2003) Synthetic applications of nitrile-converting enzymes. Current Organic Chemistry, 7, 1279–1295. [4] Martinkova, L. and Kren, V. (2002) Nitrile- and amide-converting microbial enzymes: stereo-, regio- and chemoselectivity. Biocatalysis and Biotransformation, 20,73–93. [5] Banerjee, A., Sharma, R. and Banerjee, U.C. (2002) The nitrile-degrading enzymes: current status and future prospects. Applied Microbiology and Biotechnology, 60, 33–44. [6] Zhou, Z., Hashimoto, Y. and Kobayashi, M. (2005) Nitrile degradation by Rhodococcus: useful microbial metabolism for industrial productions. Actinomycetologica, 19, 18–26. [7] OReilly, C. and Turner, P.D. (2003) The nitrilase family of CN hydrolyzing enzymes – a comparative study. Journal of Applied Microbiology, 95, 1161–1174. [8] Brenner, C. (2002) Catalysis in the nitrilase superfamily. Current Opinion in Structural Biology, 12, 775–782. [9] Groger, H. (2001) Enzymatic routes to enantiomerically pure aromatic a-hydroxy carboxylic acids: a further example for the diversity of biocatalysis. Advanced Synthesis and Catalysis, 343, 547–558. [10] Wieser, M. and Nagasawa, T. (2000) Stereoselective nitrile-converting enzymes, in Stereoselective Biocatalysis (ed. R.N. Patel), Marcel Dekker, New York, Chapter 17. [11] Singh, R., Sharma, R., Tewari, N. et al. (2006) Nitrilase and its application as a ‘green catalyst. Chemistry & Biodiversity, 3, 1279–1287. [12] Tamaki, A. and Furuya, K. (2006) Jpn. Kokai Tokkyo Koho JP 2006158323 A. [13] Furuya, K., Tamaki, A., Nagasawa, S. and Suzuki, A. (2006) PCT Int. Appl. WO 2006062189 A1. [14] Watanabe, F., Ujihara, D., Sakai, M. et al. (2007) U.S. Pat. Appl. Publ. US 20070231868A1. [15] Mistry, D. and Kullar, J.S. (2006) PCT Int. Appl. WO 2006007957 A1.


179

[16] Shibamoto, H., Aikawa, T., Arii, T. et al. (2007) PCT Int. Appl. WO 2007043466 A1. [17] Hughes, J. and Armitage, Y. (2007) U.S. Pat. Appl. Publ. US 20070184535 A1. [18] Raj, J., Seth, A., Prasad, S. and Bhalla, T.C. (2007) Bioconverson of butyronitrile to butyramide using whole cells of Rhodococcus rhodochrous PA-34. Applied Microbiology and Biotechnology, 74, 535–539. [19] Robins, K.T. and Nagasawa, T. (2006) U.S. Patent 7105322 B2. [20] Robins, K.T. and Nagasawa, T. (2007) U.S. Pat. Appl. Publ. US 20070148743 A1. [21] Zou, J., Li, M. and Liu, X. (2007) Faming Zhuanli Shenqing Gongkai Shuomingshu CN 1952114 A. [22] Payne, M.S., DiCosimo, R., Gavagan, J.E. and Fallon, R.D. (2007) U.S. Patent US 7294493 B2. [23] Petrillo, K.L., Wu, S., Hann, E.C. et al. (2005) Over-expression in Escherichia coli of a thermally stable and regio-selective nitrile hydratase from Comamonas testosteroni 5-MGAM-4D. Applied Microbiology and Biotechnology, 67 (5), 664–670. [24] Shen, Y., Xue, J., Li, H. et al. (2006) Faming Zhuanli Shenqing Gongkai Shuomingshu CN 1730660. [25] Nagasawa, T., Yoshida, T. and Matsuyama, A. (2007) Jpn. Kokai Tokkyo Koho JP 2007061035 A. [26] Nagasawa, T. and Matsuyama, A. (2005) U.S. Patent US 6900037 B2. [27] Griengl, H., Osprian, I., Schoemaker, H. et al. (2006) U.S. Pat. Appl. Publ. US 2006199256 A1. [28] Griengl, H., Osprian, I., Schoemaker, H. et al. (2006) U.S. Pat. Appl. Publ. US 20060099696 A1. [29] Reisinger, C., Osprian, I., Glieder, A. et al. (2004) Enzymatic hydrolysis of cyanohydrins with recombinant nitrile hydratase and amidase from Rhodococcus erythropolis. Biotechnology Letters, 26, 1675–1680. [30] Purkarthofer, T., Skranc, W., Schuster, C. and Griengl, H. (2007) Potential and capabilities of hydroxynitrile lyases as biocatalysts in the chemical industry. Applied Microbiology and Biotechnology, 76, 309–320. [31] Bramucci, M.G., DiCosimo, R., Fallon, R.D. et al. (2006) U.S. Patent US 7138480 B2. [32] Bramucci, M.G., DiCosimo, R., Fallon, R.D. et al. (2006) U.S. Patent US 7091011 B2. [33] Hann, E.C., Sigmund, A.E., Fager, S.K. et al. (2003) Biocatalytic hydrolysis of 3-hydroxyalkanenitriles to 3-hydroxyalkanoic acids. Advanced Synthesis and Catalysis, 345, 775–782. [34] Osswald, S., Verseck, S., Deiting, U. et al. (2006) U.S. Pat. Appl. Publ. US 20060068467 A1. [35] Aoki, T., Kawakami, K. and Otsubo, K. (2005) U.S. Patent US 6916638 B2. [36] Butlin, R.J. (2003) PCT Int. Appl. WO 2003017995 A1. [37] Aicher, T.D., Anderson, R.C., Bebernitz, G.R.,et al. (1999) (R)-3,3,3-Trifluoro-2-hydroxy-2-methylpropionamides are orally active inhibitors of pyruvate dehydrogenase kinase. Journal of Medicinal Chemistry, 42, 2741–2746. [38] Brieden, W., Naughton, A., Robins, K. and Shaw, N. (2006) U.S. Patent US 7074606 B2. [39] Shaw, N.M., Naughton, A., Robins, K., et al. (2002) Selection, purification, characterisation, and cloning of a novel heat-stable stereo-specific amidase from Klebsiella oxytoca, and its application in the synthesis of enantiomerically pure (R)- and (S)-3,3,3-trifluoro-2-hydroxy-2-methylpropionic acids and (S)-3,3,3-trifluoro2-hydroxy-2-methylpropionamide. Organic Process Research & Development, 6, 497–504. [40] Gelo-Pujic, M., Marion, C., Mauger, C. et al. (2006) Biohydrolysis of (S)-3-(thiophen-2-ylthio)butanenitrile. Tetrahedron Letters, 47, 8119–8123. [41] Yokoyama, M., Kashiwagi, M., Iwasaki, M. et al. (2004) Realization of the synthesis of a,a-disubstituted carbamylacetates and cyanoacetates by either enzymatic or chemical functional group transformation, depending upon the substrate specificity of Rhodococcus amidase. Tetrahedron: Asymmetry, 15, 2817–2820. [42] Yeom, S.-J., Kim, H.-J. and Oh, D.-K. (2007) Enantioselective production of 2,2-dimethylcyclopropane carboxylic acid from 2,2-dimethylcyclopropane carbonitrile using the nitrile hydratase and amidase of Rhodococcus erythropolis ATCC 25544. Enzyme and Microbial Technology, 41, 842–848. [43] Song, L., Wang, M., Yang, X. and Qian, S. (2007) Purification and characterization of the enantioselective nitrile hydratase from Rhodococcus sp. AJ270. Biotechnology Journal, 2, 717–724. [44] Wang, M.W., Deng, G., Wang, D.-W. and Zheng, Q.Y. (2005) Nitrile biotransformations for highly enantioselective synthesis of oxiranecarboxamides with tertiary and quaternary sterocenters: efficient chemoenzymatic approaches to enantiopure a-methylated serine and isoserine derivatives. Enantioselective production of 2,2-dimethylcyclopropane carboxylic acid from 2,2-dimethylcyclopropane carbonitrile using the nitrile hydratase and amidase of Rhodococcus erythropolis ATCC 25544. The Journal of Organic Chemistry, 70, 2439–2444. [45] Guo, X.-L., Deng, G., Xu, J. and Wang, M.-X. (2006) Immobilization of Rhodococcus sp. AJ270 in alginate capsules and its application in enantioselective biotransformation of trans-2-methyl-3- phenyl-oxiranecarbonitrile and amide. Enzyme and Microbial Technology, 39, 1–5.

180


[46] Jandhyala, D.M., Willson, R.C., Sewell, B.T. and Benedik, M.J. (2005) Comparison of cyanide-degrading nitrilases. Applied Microbiology and Biotechnology, 68, 327–335. [47] Holtze, M.S., Sorensen, J., Christian, H. and Aamand, J. (2006) Transformation of the herbicide 2,6dichlorobenzonitrile to the persistent metabolite 2,6-dichlorobenzamide (BAM) by soil bacteria known to harbor nitrile hydratase or nitrilase. Biodegradation, 17, 503–510. [48] Nielsen, M.K.K., Holtze, M.S., Svensmark, B. and Juhler, R. (2007) Demonstrating formation of potentially persistent transformation products from the herbicides bromoxynil and ioxynil using liquid chromatography– tandem mass spectrometry (LC–MS/MS). Pest Management Science, 63, 141–149. [49] Zhu, D., Mukherjee, C., Biehl, E.R. and Hua, L. (2007) Discovery of a mandelonitrile hydrolase from Bradyrhizobium japonicum USDA110 by rational genome mining. Journal of Biotechnology, 129, 645–650. [50] Zhu, D., Mukherjee, C., Biehl, E.R. and Hua, L. (2007) Discovery of a mandelonitrile hydrolase from Bradyrhizobium japonicum USDA110 by rational genome mining. Journal of Biotechnology, 129, 645–650. [51] Zhu, D., Hua, L., Biehl, E.R. and Mukherjee, C. (2007) U.S. Pat. Appl. Publ. US 20070178575 A1. [52] Kamila, S., Zhu, D., Biehl, E.R. and Hua, L. (2006) Unexpected stereorecognition in nitrilase-catalyzed hydrolysis of b-hydroxy nitriles. Organic Letters, 8, 4429–4431. [53] Zhu, D., Mukherjee, C., Biehl, E.R. and Hua, L. (2007) Nitrilase-catalyzed selective hydrolysis of dinitriles and green access to the cyanocarboxylic acids of pharmaceutical importance. Advanced Synthesis and Catalysis, 349, 1667–1670. [54] Burns, M.P. and Wong, J.W. (2005) U.S. Patent Appl. US 2005009154 A1. [55] Rustler, S. and Stolz, A. (2007) Isolation and characterization of a nitrile-hydrolysing acidotolerant black yeast, Exophiala oligosperma R1. Applied Microbiology and Biotechnology, 75, 899–908. [56] Khandelwal, A.K., Nigam, V.K., Choudhury, B. et al. (2007) Optimization of nitrilase production from a new thermophilic isolate. Journal of Chemical Technology and Biotechnology, 82, 646–651. [57] Kaplan, O., Vejvoda, V., Plinhal, O.,et al. (2006) Purification and characterization of a nitrilase from Aspergillus niger K10. Applied Microbiology and Biotechnology, 73, 567–575. [58] Mueller, P., Egorova, K., Vorgias, C.E. et al. (2006) Cloning, overexpression, and characterization of a thermoactive nitrilase from the hyperthermophilic archaeon Pyrococcus abyssi. Protein Expression and Purification, 47, 672–668. [59] Banerjee, A., Kaul, P. and Banerjee, U.C. (2006) Purification and characterization of an enantioselective arylacetonitrilase from Pseudomonas putida. Archives of Microbiology, 184, 407–418. [60] He, Y.C., Xy, J.H., Xu, Y. et al. (2007) Biocatalytic synthesis of (R)-()-mandelic acid from racemic mandelonitrile by a newly isolated nitrilase -producer Alcaligenes sp. ECU0401. Chinese Chemical Letters, 677–680. [61] Aoki, H. and Kamachi, H. (2006) U.S. Patent US 7118898 B1. [62] Wells, A. and De Wildeman, S. (2007) PCT Int. Appl. WO 2007035161 A1. [63] Erbil, H., Sezer, E., Tastan, B. et al. (2007) Efficacy and safety of serial glycolic acid peels and a topical regimen in the treatment of recalcitrant melasma. The Journal of Dermatology, 34, 25–30. [64] Hayes, M.J. and Lauren, M.D. (1994) Chemical stress relaxation of polyglycolic acid suture. Journal of Applied Biomaterials, 5, 215–220. [65] Shawe, S., Buchanan, F., Harkin-Jones, E. and Farrar, D. (2006) A study on the rate of degradation of the bioabsorbable polymer polyglycolic acid (PGA). Journal of Materials Science, 41, 4832–4838. [66] Park, J.-H., Allen, M.G. and Prausnitz, M.R. (2005) Biodegradable polymer microneedles: fabrication, mechanics and transdermal drug delivery. Journal of Controlled Release, 104, 51–66. [67] Wada, M., Mochizuki, D., Takahashi, H. et al. (2005) PCT Int. Appl. WO 2005106005 A1. [68] Foo, T. and Panova, A. (2006) U.S. Patent Appl. US 20060160196 A1. [69] Wu, S., Fogiel, A.J., Petrillo, K.L. et al. (2008) Protein engineering of nitrilase for chemoenzymatic production of glycolic acid. Biotechnology and Bioengineering, 99, 717–720. [70] Panova, A., Mersinger, L.J., Liu, Q. et al. (2007) Chemoenzymatic synthesis of glycolic acid. Advanced Synthesis and Catalysis, 349, 1462–1474. [71] DiCosimo, R., Payne, M.S., Panova, A. et al. (2007) U.S. Patent US 7198927 B2. [72] DiCosimo, R., Payne, M.S., Panova, A. et al. (2007) U.S. Patent Appl. US 20070178556 A1. [73] DiCosimo, R., Panova, A., Thompson, J.S. et al. (2006) U.S. Patent Appl. US 2006247467 A1. [74] Hinago, H., Nagahara, H. and Aoki, T. (2006) PCT Int. Appl. WO 2006126626 A1. [75] Xia, C., Wu, S., Liu, H. et al. (2006) Faming Zhuanli Shenqing Gongkai Shuomingshu CN 1880464 A.


[76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86]

[87] [88]

[89] [90] [91]

[92] [93]

[94]

[95] [96] [97] [98] [99]

181

Xue, J., Luo, J., Li, H. et al. (2006) Faming Zhuanli Shenqing Gongkai Shuomingshu CN 1772912 A. Kobayashi, Y., Watabe, K., Ohira, M. and Hayakawa, K. (2007) Eur. Pat. Appl. EP 974669 B1. Watabe, K. (2003) PCT Int. Appl. WO 2003014355 A1. Favre Bulle, O. and Le Thiesse, J.C. (2007) U.S. Patent US 7247462 B2. Favre Bulle, O., Pierrard, J., David, C. et al. (2006) Eur. Patent EP 1411126 B1. Xia, C., Wu, S., Liu, H. et al. (2006) Faming Zhuanli Shenqing Gongkai Shuomingshu CN 1880463 A. Okamoto, M. and Nakagawa, T. (2006) Jpn. Kokai Tokkyo Koho JP 2006055004 A. Kodama, T. (2006) Jpn. Kokai Tokkyo Koho JP 2006225456 A. Nagasawa, T., Yoshida, T. and Matsuyama, A. (2006) Jpn. Kokai Tokkyo Koho JP 2006223246 A. Zelinski, T., Kesseler, M., Hauer, B. and Friedrich, T. (2006) U.S. Patent Appl. US 20060259999 A1. Mateo, C., Chmura, A., Rustler, S. et al. (2006) Synthesis of enantiomerically pure (S)-mandelic acid using an oxynitrilase–nitrilase bienzymatic cascade: a nitrilase surprisingly shows nitrile hydratase activity. Tetrahedron Asymmetry, 17, 320–323. Fernandes, B.C.M., Mateo, C., Kiziak, C. et al. (2006) Nitrile hydratase activity of a recombinant nitrilase. Advanced Synthesis and Catalysis, 348, 2597–2603. Rustler, S., Muller, A., Windeisen, V. et al. (2007) Conversion of mandelonitrile and phenylglycinenitrile by recombinant E. coli cells synthesizing a nitrilase from Pseudomonas fluorescens EBC191. Enzyme and Microbial Technology, 40, 598–606. Mukherjee, C., Zhu, D., Biehl, E.R. and Hua, L. (2006) Exploring the synthetic applicability of a cyanobacterium nitrilase as catalyst for nitrile hydrolysis. European Journal of Organic Chemistry, 5238–5242. Pollock, J.A., Clark, K.M., Martynowicz, B.J. et al. (2007) A mild biosynthesis of lactones via enantioselective hydrolysis of hydroxynitriles. Tetrahedron Asymmetry, 18, 1888–1892. Winkler, M., Knall, A.C., Kulterer, M.R. and Klempier, N. (2007) Nitrilases catalyze key step to conformationally constrained GABA analogous g-amino acids in high optical purity. The Journal of Organic Chemistry, 72, 7423–7426. Xie, Z., Feng, J., Garcia, E. et al. (2006) Cloning and optimization of a nitrilase for the synthesis of (3S)-3cyano-5-methyl hexanoic acid. Journal of Molecular Catalysis B: Enzymatic, 41, 75–80. DeSantis, G., Wong, K., Farwell, B. et al. (2003) Creation of a productive, highly enantioselective nitrilase through gene site saturation mutagenesis (GSSM). Journal of the American Chemical Society, 125, 11476–11477. Bergeron, S., Chaplin, D.A., Edwards, J.H. et al. (2006) Nitrilase-catalyzed desymmetrization of 3-hydroxyglutaronitrile: preparation of a statin side-chain intermediate. Organic Process Research & Development, 10, 661–665. Kesseler, M. and Hauer, B. (2007) PCT Int. Appl. WO 2007071578 A2. Ozaki, E., Enomoto, K. and Endo, T. (2006) Eur. Pat. EP 1008655 B1. Kielbasinski, P., Rachwalski, M., Mikolajczyk, M. et al. (2007) Enzyme-promoted desymmetrisation of prochiral bis(cyanomethyl) sulfoxide. Advanced Synthesis and Catalysis, 349, 1387–1392. Fischer-Colbrie, G., Matama, T., Heumann, S. et al. (2007) Surface hydrolysis of polyacrylonitrile with nitrile hydrolysing enzymes from Micrococcus luteus BST20. Journal of Biotechnology, 129, 62–68. Matama, T., Carneiro, F., Caparro´s, C. et al. (2007) Using a nitrilase for the surface modification of acrylic fibers. Biotechnology Journal, 2, 353–360.

9 Biosynthesis of Drug Metabolites Wenying Li1, David Rozzell2, Spiros Kambourakis2 and Martin Mayhew2 1

Bristol-Myers Squibb Research and Development, Department of Biotransformation, Route 206 and Province Line Road, Princeton, NJ 08540, USA 2 Codexis, Inc., 129 N. Hill Avenue, Suite 103, Pasadena, CA 91106, USA

9.1 Introduction Drugs being administrated to humans or animals are often subjected to biotransformation, which leads to formation of drug metabolites. Drug biotransformation reactions are catalyzed by metabolizing enzymes in the liver and other organs, and generally include oxidation, reduction, hydrolysis, acylation, methylation, and conjugation with glucuronic acid, sulfate, glutathione (GSH) and amino acids. It is critical to characterize drug metabolites in drug discovery and development process, as drug biotransformation not only affects elimination and exposure of drug, but also may lead to reactive metabolites that may react with proteins, DNA or small molecules and have toxicological consequences [1–3], and active metabolites that have pharmacological and/or toxicological activities. Metabolites are characterized with different emphases at different stages of the drug discovery and development process. In the discovery stage, metabolite characterization is generally limited to identification of metabolic soft-spots and reactive metabolites (sometimes via their stable products, such as GSH conjugate), which in most of cases can be achieved by liquid chromatography–tandem mass spectrometry (LC–MS/MS) analysis with minimal amounts of in vitro or in vivo samples. When structure determination of key metabolites with NMR analysis becomes necessary or in vitro activity assays for suspected active metabolites become warranted, larger quantities of a metabolite sample may need to be prepared. However, these quantities can often be purified from the same biological sample used for the initial characterization.


184


In the development stage, the metabolite sample/standard is generated for the following purposes: . . . .

structure elucidation of key human and animal metabolites; for use as an analytical standard in the determination of metabolite exposure in humans and animals by LC–MS/MS analysis; determination of pharmacological and toxicological activities of metabolites; to ensure that the metabolite will not interfere with analysis of parent drug or other metabolites due to coelution/cross-talk [4] or ex vivo conversion.

Recent US FDA guidance on safety testing of drug metabolites [5] highlights the importance of measuring major metabolites in human and toxicological species. This increased scrutiny on the role of metabolites in the evaluation of efficacy and safety will lead to increasing demand for metabolites as analytical standards. The amount of a metabolite required for different purposes varies, from sub-milligram for structure determination/in vitro activity assay, to grams for bioanalytical assay or in vivo studies, so the method for metabolite preparation (chemical or biosynthetic methods) should be selected based on metabolite structure information, amount needed, and timeline required (Table 9.1). Chemical synthesis is the preferred method for larger scale metabolite preparation, but it is often a resource-intensive exercise and certain metabolites present particularly difficult synthetic challenges. Alternatively, microbial bioreactors may be considered for preparing quantities needed for early characterization studies, usually below 50 mg. In addition, microbial bioreactors can be used to aid in metabolite structure elucidation or to substitute chemical synthesis for large-scale metabolite preparation. Owing to the rapid development in functional expression of recombinant mammalian cytochrome P450 (CYP) and uridine diphosphoglucuronosyl transferase (UGT) enzymes in other hosts, especially in Escherichia coli, recombinant mammalian enzyme bioreactors for large-scale metabolite synthesis have become reality [6]. Mutants of the CYP BM-3 (CYP102A1) of Bacillus megaterium expressed in E. coli [7] could be another set of powerful tools for metabolite synthesis. This chapter is not intended to serve as a comprehensive review in drug metabolite biosynthesis; rather, we will focus on practical considerations for metabolite synthesis at small to medium scale with three bioreactor systems: mammalian bioreactors, microbial bioreactors and recombinant enzyme bioreactors.

9.2 Metabolite Synthesis Using Mammalian Bioreactors Drug metabolites produced in mammalian systems are most relevant to drug discovery and development. Therefore, mammalian bioreactors, including microsomes, S9 fractions, hepatocytes, liver slices and animals, are the native systems for drug metabolite synthesis. Owing to limited availability and high cost, the mammalian bioreactor is not suitable for large-scale synthesis of metabolites. However, as a native system, the mammalian bioreactor is the preferred system for generating metabolites for structure elucidation, and it is useful for rapid synthesis of a limited amount of metabolites for in vitro activity evaluation. Among common in vitro metabolizing systems, liver microsomes and liver S9 fractions are used more often in metabolite synthesis than other systems. The majority of drug metabolism is mediated by CYPs [8]. Liver microsomes contain a high concentration of CYPs and other

185

Biosynthesis of Drug Metabolites

metabolizing enzymes, including UGTs and flavin-containing monooxidases (FMOs). In addition to microsomal enzymes, S9 contains cytosolic enzymes, including alcohol/aldehyde dehydrogenases and sulfotransferases (SULTs), but CYP activities in S9 are substantially lower than those in the corresponding microsomes. Liver S9 from aroclor 1254-induced rats has high CYP and UGT activities, as aroclor 1254 (a complex mixture of polychlorinated biphenyls) induction greatly increases levels of many metabolizing enzymes [9]. So, an in vitro enzyme system suitable for metabolite synthesis can often be selected from liver microsomes of humans and animals, and liver S9 of aroclor 1254-induced rats.

9.2.1 Selection of In Vitro Systems Although there are many enzymes involved in drug metabolism, the in vitro systems containing CYPs and UGTs are often used for metabolite synthesis due to the reasons discussed before. CYPs require P450 reductase, NADPH (Figure 9.1) and molecular oxygen for the activity (Equation (1) in Figure 9.2), and catalyze a variety of reactions [10], encompassing C- and N-hydroxylation, S- and N-oxygenation, N- and O-dealkylation, oxidative deamination, dehalogenation, epoxidation and dehydrogenation. Chemically unstable product resulting from CYP-mediated biotransformation may undergo rearrangement, elimination/addition of H H

O NH2

O N

HN O

N

O

N CH2 NH2 H C OH N N H C OH N H C OH O O N H2C O P O P O O O O HO

O

O

N O

NH

HO HO

O

OH NH2

O P ON O O P O- N O O

HO N

HO

N

O O O P O P O O O

O

OH HO

N

UDP

OH HO

UDPGA

OH

NADPH

FAD

Figure 9.1

Structures of FAD, NADPH and UDPGA

(1)

RH + O2 + NADPH + H

(2)

RX + O2 + NADPH + H

(3)

RX + UDPGA

+

+

UGT

CYP

ROH + H2O + NADP+

P450 reductase FMO-FAD

RX-O + H2O + NADP+

HO HO

COOH O

X

R

OH β -glucuronide

Figure 9.2

General reactions of CYP, FMO and UGT

+

UDP

OH

O

186


water, or addition of GSH/amino acids to form more complex metabolites. Some metabolites may be subjected to further oxidation by CYPs or other enzymes. FMO, a flavin adenine dinucleotide (FAD)-containing enzyme (Figure 9.1), catalyzes NADPH- and oxygen-dependent N- and S-oxygenations of a wide range of heteroatomcontaining compounds (Equation (2) in Figure 9.2) [11], Compared with CYPs, FMOs involvement in drug metabolism is less important, so experimental conditions for metabolite synthesis are generally designed to favor CYP activity [12]. However, if a metabolite of interest is mainly generated via FMOs, then modifying the reaction conditions to promote FMO activity is needed. UGT catalyzes the transfer of the glucuronic acid moiety from cofactor uridine diphosphate glucuronic acid (UDPGA) (Figure 9.1) to the O-, N-, S, and C-centers of the substrate (Equation (3) in Figure 9.2) [13]. The common glucuronide moiety-accepting centers are the hydroxyl/carboxylic group, amine and aromatic nitrogen. Screening is usually carried out with liver microsomes from humans, rats, mice, dogs and monkeys and liver S9 fraction from aroclor 1254-induced rats. The incubation is typically run with a volume of 0.2–1.0 mL in a microcentrifuge or a glass tube. Different incubation conditions are used for CYP and UGT reactions. The incubation mixture for formation of oxidative metabolites and/or GSH conjugates contains: . . . . .

sodium/potassium phosphate buffer (50–100 mM, pH 7.4); microsomes (1–2 mg protein/mL) or S9 (3–7 mg protein/mL); substrate (10–50 mM), added as solution in suitable solvent or solvent mixture; NADPH (1–2 mM); GSH (5–10 mM), for generating GSH conjugate.

The incubation mixture for formation of glucuronides contains: . . . . .

sodium/potassium phosphate buffer (50–100 mM, pH 7.4); MgCl2 (1–10 mM); microsomes (1–2 mg protein/mL) or S9 (3–7 mg protein/mL), pretreated with alamethicn (50 mg/mg protein) in ice-water bath for 5 min; substrate (10–50 mM), added as solution in suitable solvent or solvent mixture; UDPGA (2–5 mM).

To prepare glucuronide conjugates of products formed from initial oxidation (to yield a hydroxyl or carboxylic acid group), the oxidative metabolite should be prepared first and used as a substrate in the incubation. Alternatively, an incubation with parent drug, NADPH and UDPGA may produce the desired glucuronide. These incubations are often carried out at 37 C for 1–2 h. At different time points, 20–200 mL of incubation mixture is withdrawn from each incubation and mixed with equal volume of ice-cold acetonitrile by vortexing. For preparation of acyl glucuronide, ice-cold acetonitrile containing 1% of formic acid is used to minimize acyl-migration [3,14]. After centrifugation at 13 000 rpm for 5–15 min, the supernatant (10–30 mL) is analyzed by highperformance liquid chromatography (HPLC)–UV–MS. The metabolite of interest is identified based on HPLC retention time, UV spectrum and MS/MS data. Conversion yield is estimated based on UV absorption peak areas. A suitable in vitro enzyme system for scale-up is then


187

selected by considering the cost of microsomes or S9, conversion yield to the desired metabolite, secondary reactions that will consume the desired metabolite, and formation of other metabolites that may interfere with the purification.

9.2.2 Reaction Condition Optimization The reaction conditions should be optimized prior to scale-up. As the in vivo system is mainly used for initial metabolite characterization, the goal of reaction optimization is finding suitable conditions (not necessary the optimal conditions) to enable rapid synthesis of small amounts of the desired metabolite at a reasonable cost. 9.2.2.1 Buffer and pH Recombinant CYP3A4, the major CYP enzyme in liver microsomes, has been shown to have different activities toward nifedipine oxidation with different buffers (phosphate, HEPES, and Tris) [15]. HEPES yielded the highest activity when MgCl2, GSH and cytochrome b5 (b5) were included, but phosphate appeared to be the best buffer when MgCl2 was omitted. Nevertheless, phosphate or Tris buffer (50–100 mM, pH 7.4) is widely used in incubations as it supports good CYP activity, and changing to other buffer systems is generally not necessary for CYP reactions. As optimal pH for FMO activity is around pH 9 [16,17], elevated pH may help N- or S-oxygenation, which may be catalyzed by FMOs. The pH optima of UGTs vary, and sometimes are substrate dependent. Recombinant UGTs 1A1, 1A7, 1A8, 1A9 and 1A10 exhibited pH optima of 6.4, 8.5, 7.0, 7.6 and 6.4 respectively for certain substrates [18,19]. Under similar conditions, UGT1A7 maintained good activity over a pH range of 6.4–7.6 and 6.0–8.0 for two different substrates. Although pH 7.4 is generally used in UGT assay, pH 6.4 and pH 8.4 should also be tested for glucuronide synthesis. Other factors should be considered when selecting pH for the incubations. Human b-glucuronidase, which catalyzes the hydrolysis of glucuronides, exhibits a pH optimum at 5.5, and the activity toward an O-glucuronide at pH 7.4 was less than 20% of the activity at pH 5.5 [20]. High pH will suppress b-glucuronidase activity, but will promote acyl-migration of acyl-glucuronide. 9.2.2.2 Divalent Metal Cations MgCl2 (10 mM) increased the apparent Km (83 to 173 mM) and reduced the Vmax (3.4 to 2.4 min1) of triazolam 4-hydroxylation by expressed CYP3A4 [21]. However, both MgCl2 (30 mM) and CaCl2 (30 mM) significantly increased reaction rates of testosterone 6bhydroxylation (approximately threefold) and nifedipine oxidation (three- to six-fold) by human liver microsomes (HLMs) or recombinant CYP3A4 (reconstituted with b5 and GSH) [15]. It was suggested that divalent cation stimulation on the activity was related to involvement of b5 in CYP 3A4 reaction. As the divalent cation effect on CYP3A4 activity appears to be substrate dependent, the benefit of including MgCl2 (1–30 mM) in the oxidative incubation should be examined for each substrate. MgCl2 (1 and 10 mM) enhanced the glucuronidation activity of HLMs toward estradiol, acetaminophen and morphine [22]. So it is beneficial to include MgCl2 in glucuronidation reactions. However, it may not be feasible to add MgCl2 to a high pH buffer system due to

188


precipitation of Mg(OH)2. Ca2þ may have a similar effect, as CaCl2 (6 mM) was able to support glucuronidation of benzyl alcohol and 4-methyl phenol by bovine liver UGTs [23]. 9.2.2.3 Other Additives The active site of UGT is on the lumenal site of the endoplasmic reticulum (ER) membrane [24]. The ER membrane may hinder access of substrates and cofactors to the UGT active site. Portforming peptide alamethicin (50 mg/mg microsomal protein) increased rates of acetaminophen and morphine glucuronidation with HLMs by two- to three-fold, which is suggested to be due to alamethicins ability to aid substrate and cofactor crossing the ER membrane [22]. It has become a general practice to include alamethicin in UGTassays, but using alamethicin in largescale incubation might be cost prohibitive. Alternatively, microsmes can be activated by sonication or detergent (such as Brij58, Triton X-100) treatment, although they were less effective and treatment conditions will need to be optimized for each substrate [25,26]. Without pretreatment using these three methods, microsomes or S9 may still be able to provide reasonable UGT activity for glucuronide synthesis, as shown by Stevenson and Hubl [23] and the example in Section 9.2.2.2. Glycerol is widely used for protein stabilization. Albumen also has a protein stabilizing effect. Glycerol (20% v/v) or bovine serum albumen (BSA; 30 mg mL1) was able to prolong UGT activity and lead to higher glucuronidation yield [23]. BSA is shown to enhance UGT1A9 activity [27]. However, glycerol may inhibit CYP enzyme activity [28]. Therefore, while addition of glycerol (10–20% v/v) in glucuronidation is helpful, its effect on oxidative reactions is not predictable. On the other hand, in addition to a positive effect on UGTs, BSA may have a similar stabilizing effect on CYPs. D-Saccharic acid 1,4-lactone (SAL; 5–10 mM), a selective inhibitor of b-glucuronidase [29], is often included in UGT reactions for improving yield. As b-glucuronidases and UGTs have different optimal pH ranges, use of SAL in preparative-scale reactions may be avoided. 9.2.2.4 Organic Solvents For easy dispensation, a drug substrate is often dissolved in an organic solvent or solvent mixture before being added to the reaction mixture. The effects of organic solvents on human CYP and UGT enzymes are well documented [30–33]. Common organic solvents (acetonitrile, methanol and dimethyl sulfoxide (DMSO)) at final concentration of 5% were shown to inhibit CYP activities strongly in HLMs. At lower concentration, the effect of each solvent on different major CYP isoforms varied. Acetonitrile and methanol had minimal effects on CYP activity at concentrations of up to 1%, while DMSO at a concentration of 0.2% inhibited more than 50% of CYP2E1 in HLMs [30–33]. If a CYP isoform responsible for formation of the desired metabolite is known, then one will be able to select a suitable solvent based on the solvent effect on a specific CYP isform. In most cases when the CYP isoform information is not available, it is suggested that organic solvent in the preparative in vitro reaction be kept lower than 1% for acetonitrile and methanol and lower than 0.2% for DMSO. Acetonitrile, methanol and DMSO had no apparent effect on umbelliferone glucuronidation in human hepatocytes at concentration up to 2% [32]. With HLMs or expressed UGTs, inhibitory effects of organic solvents on glucuronidation of 7-hydroxy-4-trifluoromethylcoumarin (7-HFC) and estradiol generally followed the order acetonitrile > ethanol > methonal > DMSO [33]. DMSO did not inhibit estradiol-3-glucuronidation activity at a concentration up

189


to 5%, while acetonitrile (at a concentration of 1–5%) gave stronger stimulation effect of 7-HFC glucuronidation activity with expressed UGT1A1, 1A7 and 2B4. Therefore, DMSO is a preferred solvent for most UGT reactions, but acetonitrile may benefit some reactions. It should be noted that the purpose of using organic solvent is to help delivery of substrate evenly into the reaction mixture, but not to help solubilize substrate in the reaction mixture. So the amount of organic solvent should be kept to a minimum, just enough to dissolve substrate. 9.2.2.5 Enzyme, Substrate, Cofactor and Reaction Time

Remaininng%

100.0

Conversion%

In general, microsomes/S9 and cofactor (NADPH or UDPGA) are the most costly components of an incubation, but substrate, especially for early discovery compounds, is sometimes scarce. Higher enzyme concentration will lead to higher volume productivity and less amount of the cofactor to maintain the same cofactor concentration. However, the relationship of reaction rate and enzyme concentration may not be linear, so a higher enzyme concentration may yield lower enzyme productivity (amount of product per milligram enzyme used). Therefore, different protein concentration levels should be screened to obtain a good balance between volume and enzyme productivities. If a high turnover (>50%) is observed in the initial screening incubation at 10–50 mM, then a higher substrate concentration should be tested to obtain a good balance between the enzyme productivity and the conversion yield. The substrate concentration should be mainly selected based on the activity of enzymes, and may be increased until the benefit of enzyme productivity increase is offset by cost and availability of the substrate. The solubility of substrate in buffer may limit the use of high concentrations, but a low aqueous solubility compound is readily absorbed by microsomes/S9, which will help to keep the substrate in the incubation mixture even at a much higher concentration than its buffer solubility. Substrate concentration up to sub-millimole level can be used as long as no inhibitory effect is shown and a reasonable turnover is achieved.

15.0

Drug substrate

90.0 80.0 70.0 20.0 M1 M2 M3

10.0

Oxidative metabolites

5.0 0.0 0

50

100

150

200

250

300

Incubation Time (min) Figure 9.3 Reaction time course of a large scale rat liver microsomal (RLM) incubation (250 mL) at 37 C (Li, unpublished results). Conversion% was calculated based on UV absorbance peak

190


An NADPH-regenerating system [34,35] such as D-glucose-6-phosphate/glucose-6-phosphate dehydrogenase can be used to maintain the NADPH level in the reaction mixture. However, for the purpose of rapid generation of small amounts of metabolites, it may not be worth introducing another set of variables by including a NADPH-regenerating system. A typical CYP reaction length is 1–2 h, but CYP activity can survive longer at 37 C. Figure 9.3 shows that product turnover of the CYP reactions occurred over a 6 h incubation (Li, unpublished results). UGT activity can last longer than 24 h [23]. With accumulation of product, secondary reactions, such as further oxidation of product or hydrolysis of glucuronide, may become noticeable. Therefore, monitoring the reaction with HPLC–UV–MS is critical for identifying the best time to terminate the reaction.

9.2.3 Large-Scale Incubations A typical procedure for setting up a preparative incubation is as follows: the substrate solution with a minimal amount of organic solvent or solvent mixture is mixed with buffer (with necessary additives) in an Erlenmeyer flask. Microsomes or S9 is then added to the mixture, followed by addition of NADPH (or UDPGA) stock solution. The flask is incubated at 37 C in a water bath with gentle shaking. The reaction kinetics in a large-volume incubation will be different from the 0.2–1.0 mL incubation, so progress of the reaction should be monitored closely with a short HPLC–UV–MS method. Reaction is terminated by acetonitrile quenching, or by liquid–liquid extraction with waterimmiscible organic solvent, provided that the extraction efficiency and the effect of the organic solvent on product stability are tested at the small scale. Based on properties of product, the pH of the reaction mixture should be adjusted before termination to allow maximal recovery of the product. For example, acid is usually added to the acyl-glucuronide product mixture at the end of the reaction to minimize acyl migration.

9.2.4 Examples with Mammalian Bioreactors The following examples are used to illustrate how preparative-scale metabolite synthesis is typically carried out with mammalian bioreactors in our laboratory. 9.2.4.1 Synthesis of M3 of BMS-1 for Structure Determination BMS-1 is a potent dual-acting angiotensin-1 and endothelin-a receptor antagonist, but it exhibits low systemic exposure in cynomolgus monkeys after oral administration. The major biotransformation pathway in monkey microsomes was identified by HPLC–MS as hydroxylation of the dimethyl isoxazole moiety, leading to metabolite M3 [36] (Figures 9.4 and 9.5). Modification at the hydroxylation site might block the biotransformation and lead to metabolically stable compounds that would have higher in vivo exposure. In order to provide an M3 sample for identification of the hydroxylation site by NMR analysis, synthesis of M3 with cynomolgus monkey liver microsomes (CLMs) was carried out using the following procedure: 0.4 mL 20 mM BMS-1 solution in DMSO and 4 mL pooled CLMs (20 mg protein/mL) were added to a mixture of 29.6 mL deionized water, 4.4 mL 1 M potassium phosphate buffer (pH 7.4), 0.4 mL 1.0 M MgCl2 in a 100 mL Erlenmeyer flask. The flask was incubated at 37 C in a water bath for 3 min and NADPH (1.6 mL of 100 mM solution in water) was added. The flask

191

Biosynthesis of Drug Metabolites O

O N H

N

Monkey liver microsomes

O

S

O

NADPH, oxygen

O O N

O

N H

N

O O N

O S

N H

Figure 9.4

10

N H 9

M3

BMS-1

OH

Formation of M3 via biotranformation of BMS-1 with monkey liver microsomes

M3

100

UV Absorbance at 280 nm

90 80 70 60 50 40 30

BMS-1

20 10 13

14

15

16

17

18

19

20

Time (min)

Figure 9.5

HPLC–UV chromatogram of the 2 h monkey LM incubation of BMS-1

was incubated in the 37 C water bath for 2 h with occasional gentle shaking. The incubation mixture was extracted with ethyl acetate. M3 (formic acid salt, 3 mg, 59% yield) was purified from the extract with semi-prep HPLC using acetonitrile–water containing 0.1% formic acid as eluents. NMR analysis on the purified M3 indicated that the hydroxyl group was at the C-9 position. 9.2.4.2 Generating Gemfibrozil Acyl Glucuronide with Rat S9 Gemfibrozil, a fibric acid antilipemic agent, is used to treat hyperlipoproteinemia and as a second-line therapy for type IIb hypercholesterolemia. Gemfibrozil 1-O-b-glucuronide is one of the major metabolites in humans. An efficient method for preparing acyl 1-O-bglucuronide of gemfibrozil was developed, using liver S9 fraction of aroclor 1254-induced rats as a biocatalyst (Figure 9.6) [37]. A typical reaction was performed as follows: 1.1 mL 0.18 M gemfibrozil solution in acetonitrile and S9 fraction of aroclor 1254-induced SD rats

192

Biocatalysis for the Pharmaceutical Industry OH O O

HO O

liver S9 of Aroclor 1254-induced rats OH

Gemfibrozil

UDPGA

O

O

OH O

COOH

Gemfibrozil 1-O -β-glucuronide

Figure 9.6 Conversion of gemfibrazil to gemfibrazil 1-O-b-glucuronide via with liver S9 of aroclor 1254-induced rats

(10 mL, 36 mg protein/mL) were added to a 250 mL Erlenmeyer flask containing 13.9 mL deionized water, 5 mL 1 M glycineNa buffer (pH 8.6) and 10 mL 50% (v/v) glycerol. UDPGA water solution (10 mL, 40 mM) was added and the flask was incubated at 37 C in a water bath at 50 rpm. The reaction was monitored by HPLC analysis. After 2 h, conversion reached 84% and the reaction was quenched by addition of 2 mL 1 M HCl. The reaction mixture was extracted with ethyl acetate. Gemfibrozil 1-O-b-glucuronide (49 mg, 58% yield) was purified from the extract with semi-prep HPLC using acetonitrile–water containing 0.1% formic acid as eluents.

9.2.5 In Vivo Samples Metabolite can be isolated from in vivo samples of animals or humans (urine, feces, bile and plasma), when such samples are available. Often, a drug candidate that is metabolically stable in vitro is found to be extensively metabolized in vivo. Some major in vivo metabolites, such as sequential metabolites or products of non-CYP enzyme-mediated biotransformation, are not formed in vitro significantly. Therefore, in vivo samples would be a more suitable source for isolating these metabolites. In some cases, animals are dosed just for generation of metabolites. But there are disadvantages in using animals solely for metabolite isolation purposes, including: (1) animal studies are expensive; (2) substantial amounts of the parent drug are needed for dosing animals, especially for large animals such dogs or monkeys; (3) the yield of the desired metabolite may be low, although the parent drug may be extensively metabolized; (4) purification is challenging due to interference from other metabolites or endogenous compounds. However, it would be prudent to collect animal excreta samples from routine pharmacokinetic (PK), absorption, distribution, metabolism and excretion (ADME) or toxicological studies for metabolite isolation. Owing to the much higher dose ranges used in toxicological studies, excreta from these studies often contain large amount of metabolites and, thus, are excellent samples for metabolite isolation. It is acceptable to isolate metabolites from human ADME samples. In some special cases, metabolites may also be isolated from other clinical samples.

9.3 Metabolite Synthesis Using Microbial Bioreactors In the 1970s, Smith and Rosazza introduced the concept of ‘microbial models of mammalian metabolism’ [38,39]. Two reviews in this area were published in 1999 [40,41]. Since these, there have been numerous reports describing the use of microbial biotransformation in drug metabolism studies [42–54]. These reports demonstrated that microbial enzymes were able to


193

mimic mammalian CYP enzymes and often carried out the same types of oxidative biotransformation and, thus, could be employed for synthesis of mammalian metabolites.

9.3.1 Microbial Bioreactors Used in Metabolite Structure Elucidation Microbial enzymes, especially microbial CYP enzymes, catalyze oxidative biotransformtion of xenobiotics in a fashion similar to mammalian CYP enzymes. Based on the concept of microbial models of mammalian metabolism, the drug substrate is biotransformed by microbial cells to microbial products, many of which are the same as the mammalian metabolites. Microbial products that have identical LC, UV and MS properties to those of mammalian metabolites are presumably the same as the mammalian metabolites. After purification and structure determination of the microbial products, the identity of the corresponding mammalian metabolites is determined. The purified microbial products can also be used in biological activity tests. Some microbial products do not match to known mammalian metabolites, but they are useful in ‘predicting’ possible mammalian metabolites or for ruling out certain structure assignments of mammalian metabolites. Cautions need to be taken when a chiral center is created by the biotransformation. Unambiguous assignment of such a structure would require that both enantiomers/isomers be prepared and compared with the mammalian metabolite. Highly active ‘super’ strains, such as Cunninghamella sp., Beauveria bassiana and Streptomyces rimosus, are particularly useful here, as they tend to produce multiple metabolites; thus, the majority of the desired metabolites can be obtained with a minimal number of incubations. When using microbial products for mammalian metabolite identification, it is suggested to compare all the analytical data available. For example, slight differences in MS2 or MS3 spectra may indicate that the microbial products are not the same as the mammalian metabolite. Owing to matrix effects, HPLC retention time often varies from run to run, so it is good practice to spike a comparable amount of purified microbial product into the in vitro, in vivo or purified samples that contain the mammalian metabolite of interest. If the microbial metabolite and the mammalian metabolite are the same compound, then they should co-elute under different HPLC conditions, including different solvent pH, and the MS and/or UV peak area would increase accordingly. Owing to rapid development in analytical techniques, metabolite identification and structure elucidation have become possible even with trace levels of metabolites generated with in vitro or in vivo mammalian systems. However, the microbial bioreactor is still a valuable system for metabolite structure determination, especially when the metabolite of interest presents at a low level in in vitro or in vivo mammalian systems and the isolation from these matrices is hindered by the interference of other metabolites, the parent drug or endogenous compounds, or the structure determination requires appreciable amounts of samples due to structure complexity.

9.3.2 Microbial Bioreactors Used in Synthesis of Key Metabolites Considerable effort is often required to prepare sufficient quantities of key mammalian metabolites of drug candidates for biological activity evaluation or for use as analytical standards. The new regulatory guidance in drug development [5] will certainly lead to more emphasis on key human metabolite characterization. Microbial bioreactors can be used for

194


large-scale preparation of key human metabolites. Strains suitable for this type of application should have high activity as well as high selectivity toward formation of the desired metabolite. The ‘super’ strains that produce multiple microbial products may not serve the purpose well, as activity toward other products lowers the yield of the desired product and increases the difficulty of purification.

9.3.3 Strain Selection Flask fermentation for yeast, fungi and bacteria is usually carried out based on a two-stage procedure that starts with cultures maintained on suitable agar slant, followed by two-stage growth in soybean meal–glucose media (Table 9.3) [29]. The procedure offers high activity, but may not be suitable for rapid screening. A bacterial strain replication/growth system with a 96 deep-well plate has been developed [59,60], although it requires special apparatus. Table 9.1

Purpose of metabolite biosynthesis

Purpose of metabolite biosynthesis

Details

Coarse structure identification (by LC–UV–MS)

1–10 ng With little or no purification, in vitro and in vivo samples are analyzed by LC–UV–MS for metabolite identification. Types and sites of the modifications are deduced based on LC, UV and MS data, but usually exact structures are not determined. In drug discovery, coarse structure information of prominent metabolites will allow blocking the major biotransformation pathway and lead to new analogs with improved metabolic stability 10 mg–1 mg Purified metabolite is analyzed by NMR analysis for exact structure determination. In drug discovery, structure information of reactive metabolites or GSH adducts will help identifying problematic functional groups. In drug development, structure information of prominent human metabolites are important for both fulfillment of regulatory requirement and identification of major biotransformation pathways Metabolites may interfere with LC–MS 0.1–1 mg assay of parent or other metabolites by co-elution/cross-talk [4] or ex vivo conversion and lead to failure of the assay. Metabolite standard will help to prevent such assay failures

Structure elucidation (by NMR)

LC–MS assay development of parent drug or other metabolites

Typical amount needed

Preferred sample generating methods Mammalian bioreactor, in vivo human or animal samples

Mammalian bioreactor, in vivo human or animal samples

Mammalian bioreactor

(continued)

195

Biosynthesis of Drug Metabolites Table 9.1 (Continued) Purpose of metabolite biosynthesis

Details

Activity of the metabolite toward the In vitro activity therapeutic target is determined in vitro assay (for pharmawith a purified sample. Sometimes, the cological activity metabolite is subjected to a suite of and toxicity in vitro assays to access its potential evaluation) liability, such as CYP inhibition and HERG activity. In drug discovery, an apparent PK/PD disconnect may result from metabolites with similar or superior pharmacological activities to that of parent. Such active metabolites may be developed as drugs by their own rights. In drug development, prominent human circulating metabolites may be required by regulatory agencies to have biological activities evaluated The prominent human circulating Quantitative metabolites may need to be monitored LC–MS assay in human and animal studies to ensure (used as an that the metabolites have adequate analytical exposures in toxicology animal species. standard) A non-validated LC–MS assay requires a much lesser amount of metabolite standard than a validated GLP-assay In some rare cases, a prominent human In vivo study circulating metabolite that is formed at (mainly for greater than 10% of parent systemic toxicity exposure at steady state in humans evaluation) may be present at much lower levels in toxicology in animal species. It may be necessary to directly administer the metabolite to animals for further safety evaluation. Depending on animals tested, dose and study length, a large amount of the metabolite may be needed to fund the studies

Typical amount needed

Preferred sample generating methods

0.1–10 mg

Mammalian or Microbial bioreactors

5–500 mg

Microbial or recombinant enzyme bioreactors

Multi-gram

Microbial or recombinant enzyme bioreactors

We have reported a simple actinomycetes strain storage/growth/screening system with 24 deep-well plates which does not require special equipment [58]. The screening plates are prepared as follows: inoculated with frozen cultures, cultures are grown in 500 ml flask containing 100 mL of malt extract medium (Table 9.3) for 3 days at 28 C on a rotary shaker operating at 250 rpm. The resulting culture (8 mL) is transferred into a well of a Uniplate (24 wells, 10 mL, irradiated, Whatman, Clifton, NJ) to form a master plate. With a multi-channel

196


pipette, 0.1 mL of culture from each well of the master plate is transferred to the corresponding well of a Uniplate to give the screening plate. A large number of 24-well deep-well screening plates containing 20 actinomycetes strains are prepared and stored at 78 C until needed. In a routine screening, one frozen screening plate is thawed at room temperature and 1 mL of the malt extract medium is added to each well. The plate is incubated for 2 days at 28 C on a rotary shaker operating at 275 rpm and the cultures usually achieve good growth. Substrate solution in DMSO or other solvent is added to each well to a final concentration of 0.1–0.3 mM. The plate is incubated with shaking for one additional day at 28 C. Then 1 mL of methanol or acetonitrile is added to each well and the plate is shaken at room temperature for 1 h and centrifuged at 3000 rpm for 15 min. The supernatant (10 mL) is analyzed by HPLC–MS/MS under the same conditions in which the desired mammalian metabolite is detected. This system provides a simple way to screen microbial strains with a small amount (4 mg) of parent compound; the entire screening process can be completed in 4 days, including 1 day for analysis. Selection of strains for scale-up reactions is based on the yield of desired metabolites and on the number and yields of undesired metabolites. A large number of microbial strains have been employed for modifying natural products: synthetic and semi-synthetic compounds. For drug metabolite synthesis, fungus Cunninghamella strains have been heavily used, but other strains were shown to have good activities. Active strains reported for metabolite synthesis are summarized in Table 9.2. It is by no means a complete list, but could be a good staring point for strain selection. ‘Universal’ media for those strains are listed in Table 9.3, but the reader should consult with the references for more specific media and fermentation conditions.

9.3.4 Microbial Glycoside Conjugation Microbial oxidation of drug substrates occurs in a similar fashion to mammalian oxidative biotransformation. In contrast, microbial cultures rarely catalyze conjugations comparable to those in mammalian system (glucuronidation, sulfation and GSH conjugation). It is thus not surprising that microbial bioreactors are mainly used in the synthesis of oxidative metabolites. Glucuronide is the predominant glycoside formed in mammalian systems, although glucoside is occasionally detected as a mammalian drug metabolite [70,71]. Microbial glycosylation of xenobiotics occurs sporadically, and the primary products are glucosides [50,62,64,65,67] and 6-deoxyglycoside [73] (and Li, unpublished results). Microbial glucuronidation has been seen in the following reports: phenolic O-glucuronidation of aromatic hydrocarbons 1-naphthol by Cunninghamella elegans ATCC35112 [72], N-glucuronidation of MK954 by Streptomyces sp. ATCC55043 [69] and phenolic O-glucuronidation of raloxifene, again by Streptomyces sp. ATCC55043 (Figure 9.7) [43]. Obviously, microbial glucuronidation has to compete with oxidative biotransformation on the samesubstrate.Cunninghamellaelegans ATCC35112 maynotbeasuitablestrainforglucuronide synthesis owing to its high oxidative activity. Streptomyces sp. ATCC55043 is a very useful microbial strain for glucuronide synthesis, as its high glucuronidation activity seems to overshadow the oxidative activity. In addition to Streptomyces sp. ATCC55043, we found several actinomycetes strains that had high glucuronidation and glycosylation activities toward various substrates [78] (and Li, unpublished results). These strains were able to make acyl-glucuronide, phenolic O-glucuronide, aliphatic O-glucuronide and N-glucuronide. Development in this area will certainly lead to more applications of the microbial method on glucuronide synthesis.

197

Biosynthesis of Drug Metabolites Table 9.2 Active strains that have been used in metabolite synthesis Strain

Identification numbera,b

Reaction typec

References

Actinoplanes missouriensis Actinoplanes sp. Dactylosporangium variesporum Nocardia autotrophica Amycolatopsis orientalis Pseudonocardia autotrophica Saccharopolyspora erythrara Saccharopolyspora hirsuta Saccharothrix sp. Streptomyces flocculus Streptomyces fradiae Streptomyces griseus

NRRL B3342 ATCC 53771 IFO 14104 ATCC 35203 PTA 1043 ATCC 35204 B-2338 ATCC 20501 NBRC 13785 ATCC 25453 NRRL B1195 ATCC 13273 ATCC 10137 ATCC 11796 NRRL 2364 ATCC10970 (NRRL 2234) ATCC 13400 ATCC 10970 NRRL 2234 ATCC 55043 (NRRL 21489) ATCC 55293 SC 15761 AF528 ATCC 31560 AF935 AF940 ATCC 24169 LCP 63.1800 LCP 57.1569 MMP 3010 MMP 2092 SC 15850 SC 15851 SC 15837 SC 15838 SC 15839 NRRL 315 LCP 75.2296 ATCC 7159 MMP 1837 ATCC 11064 NRRL 3644 AS 3.153 AS 3.910 ATCC 8688a

O O O O O O O O O O O O O O O O O O O GA O O O O O O O O O O G O O O O O O O O, G O O O O, G R, G O

[55] [48,49,58,69] [48] [58] [58] [48,58] [48] [53] [48] [58] [66] [49,58] [58] [49] [66] [63,64] [58] [49] [66] [43,69] [69] [58] [48] [58] [48] [48] [49] [66] [66] [65] [65] [58] [58] [58] [58] [58] [66] [66] [62,64] [66] [68] [68] [47,50,51] [51] [68]

Streptomyces Streptomyces Streptomyces Streptomyces Streptomyces

griseolus platensis rimosus roseochromogenus rimosus

Streptomyces sp.

Streptomyces violascens Streptosporangium sp. Absidia pseudocylindrospora Absidia corymbifera Absidia cylindrospora Acremonium alternatum Actinomucor elegans Actinomycetes sp.

Aspergillus alliaceus Aspergillus terreus Baeuveria bassiana Circinella minor Cunninghamella bertholletiae Cunninghamella blakesleeana

(continued)

198


Table 9.2 (Continued) Strain

Identification numbera,b

Reaction typec

References

Cunninghamella echinulata

ATCC 9244 NRRL 3655 LCP 73.2203 ATCC 42616 AS 3.2004 MMP 2203 NRRL 2310 ATCC 36112 ATCC 26269 ATCC 9245 ATCC 20230 AS 3.156 AS 3.2028 ATCC 22110 NRRL 2380 2380 MRC-826 ATCC 14717 3719 ATCC 42613 (NRRL 1757) NMP 108 LCP 52.108

O O O O O O O O, G O, G O, S O O O O O O O O O O, S, G O O O O O O, S O, S O O O O O O O O O

[66,68] [66] [66] [44,47,51] [51,62] [68] [68] [44,61,62,66–68] [62,66] [42,45,46] [44,53] [47,51] [47,51] [44] [65,66] [44] [44] [66] [44] [49,62,65,66] [62] [66] [50] [66] [66] [65,66] [65] [66] [66] [66] [65] [56] [66] [49] [66] [49]

Cunninghamella elegans

Cunninghamella phaeospora Curvalaria lunata Fusarium oxysporum Fusarium moniliforme Fusarium roseum Fusarium solani Mortierella isabellina

Mortierella zonata Mucor circinelloides Mucor griseocyanus Mucor hiemalis Mucor janssenii Mucor plumbeus Mucor racemosus Mucor rouxii Penicillium janthinellum Rhizopus arrhizus Syncephalastrum racemosum Thamnostylum piriforme Verticillium theobromae a

CBS 108-16 ATCC 1207a BO NRRL 3628 CBS 110-16 ATCC 4740 BO CBS 41677 As 3.510 ATCC 11145 ATCC 18192 ATCC 8992 ATCC 12474

Strain with ATCC number could be obtained from American Type Culture Collection (Manassas, VA, USA). b Strain with NRRL number could be obtained from the National Center for Agricultural Utilization Research (Peoria, IL, USA). c Reaction types noted: oxidation (O), reduction (R), glycosylation (G), glucuronidation (GA) and sulfation (S).

199

Biosynthesis of Drug Metabolites Table 9.3 Representative media for microbial biotransformation Media name

Soybean meal–glucose [26] Malt extract [58]

Corn Steep Liquor [62]

Composition

Dextrose, 20.0 g Yeast extract, 5.0 g Soybean meal, 5.0 g NaCl, 5.0 g K2HPO4, 5.0 g Deionized water, 1 L

Corn steep liquor, 10.0 g Dextrose, 30.0 g K2HPO4, 2.0 g KH2PO4, 1.0 g NaNO3, 2.0 g KCl, 0.5 g MgSO47H2O, 0.5 g FeSO47H2O, 0.02 g Tap water, 1 L No pH adjustment 121 C, 30 min Fungi

Malt extract, 10.0 g Yeast extract, 10.0 g Peptone, 1.0 g Dextrose 20.0 g Deionized water, 1 L

pH before sterilization pH 7.0 Sterilization 121 C, 15 min Application Fungi, yeast, bacteria

pH 7.0 121 C, 20 min Bacteria

HO N Cl

N

O

OH OH

N N NH N OH

1-Naphthol Figure 9.7

MK 954

O

N

Raloxifene

Substrates of microbial glucuronidation. Site of conjugation is marked by the arrow

9.3.5 Large-Scale Reactions Initial scale-up of microbial biotransformation is conveniently run with multiple flasks without extensive reaction optimization. A typical flask fermentation is performed at 28 C, 250 rpm with 100 mL culture in a 500 mL Erlenmeyer flask, although other settings will work fine too. Three parameters need to be investigated before scale-up: the time for adding the substrate, the optimal substrate concentration and the time course of product formation. Optimization of other factors, such as medium composition and pH, growing cells versus resting cells [74], is helpful, if the timeline allows and if there is a sufficient amount of the substrate to support the screening. Substrate can be added to the cultures as a solid, a suspension, or a solution in DMSO, methanol, ethanol, acetonitrile or water. As the microbial culture generally has high tolerance toward organic solvents, there is less restriction on the choice and amount of solvent to be used for dispensing the substrate. Aqueous solubility of substrates normally will not affect compound loading, as a compound with poor aqueous solubility will likely be absorbed by the cells and still be subjected to biotransformation.

200


Biotransformation with flasks can be used to make gram quantities of a desired product, as shown for the 21-hydroxylation of epothilone B [75]. In cases when greater quantities of a metabolite are needed, microbial biotransformations can be carried out in a fermentor, which will allow better monitoring and control of fermentation conditions (such as pH, oxygen and glucose levels, etc.) for reaction optimization [76].

9.3.6 Examples for Metabolite Synthesis with a Microbial Bioreactor 9.3.6.1 Preparation of Dasatinib Metabolite M20 The M20 metabolite is a major circulating metabolite of dasatinib (SPRYCEL) in humans [77]. A large quantity of M20 was needed to serve as an analytical standard, but was not readily accessible by mammalian bioreactors or chemical synthesis. Microbial biotransformation was used to make the metabolite to support the development of dasatinib [58]. Screening for Active Strains Selection of active strains was achieved using the 20-well screening plate system discussed in Section 9.3.3. One screening plate containing 20 actinomycetes strains in malt extract medium was incubated for 2 days at 28 C and 275 rpm. Dasatinib (2 mL of a 100 mM solution in DMSO) was added to each well and the plate was incubated for one additional day at 28 C. Then 1 mL of methanol was added to each well and the plate incubated at 28 C at 150 rpm for 10 min and centrifuged at 3000 rpm for 15 min. The supernatant was analyzed by HPLC–UV–MS. LC–UV–MS data showed M20 and M24 were produced by seven strains, from which Streptomyces sp. SC15761 was selected for scale-up based on the yield of M20 (Figure 9.8). N H N

HN S N

N

HO

N

Cl

Streptomyces sp. SC15761

O

N

Dasatinib N H N

HN S N

N

HO

N N

N

Cl

S

+ O

H N

HN

OH

N HO

N

N N

Cl

O OH

M20

Figure 9.8

M24

Microbial biotransformation of dasatinib with Streptomyces sp. SC15761

Preparation of M20 by Microbial Biotransformation A 500 mL flask containing 100 mL of the malt extract medium was inoculated with 2 mL frozen stock culture of Streptomyces sp. SC15761 and incubated for 3 days at 28 C and 250 rpm. Then 1 mL of the resulting culture was added to each 500 mL flask (11 in total) containing 100 mL of the malt extract broth. The cultures were incubated at 28 C and 250 rpm for 2 days. Dasatinib (200 mL of a 48.9 mg mL1 solution in DMSO) was then added to each of the 11 flasks. The flasks were returned to the shaker and incubated for an additional 27 h at 28 C and 250 rpm. The reaction cultures were pooled and extracted twice, once with 1000 mL of ethyl acetate and once

201


with 500 mL of ethyl acetate. The combined ethyl acetate extract was evaporated to dryness in vacuo. The residue was then dissolved in 2 mL of DMSO. From half of the DMSO solution, 23 mg of M20 (formic acid salt) was obtained with semi-preparative HPLC. 9.3.6.2 Preparation of S-Ketoprofen 1-O-b-Glucuronide A frozen culture of Streptomyces sp. ATCC 55043 (2 ml) was added to a 500 mL flask containing 100 mL of the malt extract medium. The flask was incubated for 3 days at 28 C and 250 rpm. Then 1 mL of the resulting culture was used to inoculate each of five 500 mL flasks containing 100 mL of malt extract medium. The flasks were incubated at 28 C on a rotary shaker operating at 250 rpm for 1 day. (S)-(þ )-Ketoprofen (0.8 mL of 100 mM solution in DMSO) was added to each flask. The culture was incubated for an additional 26 h at 28 C and 250 rpm. At the end of the incubation, cultures from the five flasks were pooled and cells were removed by centrifugation at 3000 rpm for 15 min. After mixing with formic acid (1.2 mL), the supernatant was loaded onto a Varian MEGA BE-C18 solid-phase extraction column (60 mL, 10 g) and eluted with methanol. The methanol eluent was concentrated in vacuo with a rotary evaporator to a small volume (6 mL). (S)-Ketoprofen 1-O-b-glucuronide (48 mg, 28.0% yield) was purified from the concentrated extract with semi-prep HPLC using a Waters SunFire Prep C18 OBD column (19 mm 150 mm, S5) and acetonitrile–water containing 0.1% formic acid as eluents (Figures 9.9 and 9.10) [78]. OH Streptomyces sp. ATCC55043 CO2H

O

O

O

COOH

O

S-Ketoprofen-1-O-β -Glucuronide

S(+)-Ketoprofen

Figure 9.9

Microbial biotransformation of (S)-ketoprofen with Streptomyces sp. ATCC 55043

DAD1 A, Sig=275,10 Ref=500,80

mAU

UV absorbance at 275 nm

OH

HO O

S-Ketoprofen 1-O -β -glucuronide

350 300

S-Ketoprofen

250 200 150 100 50 0 0

2

4

6

8

10

12

14

16

18 min

Time (min)

Figure 9.10 HPLC–UV chromatogram of the 26 h microbial incubation of S-ketoprofen with Streptomyces sp. ATCC 55043

202


9.4 Recombinant Enzyme Bioreactors As a large superfamily of hemoproteins that are widely distributed throughout nature, CYP enzymes have been found in organisms ranging from archaebacteria to primates. Thousands of such enzymes have been identified, and this number is increasing almost daily, as interest in the study of CYPs has surged in recent years. An excellent searchable source for all enzymes identified to date is available in the P450 database kept by David Nelson at http://drnelson. utmem.edu/CytochromeP450.html (last access October 2008). The physiological role of CYPs is quite broad, with these enzymes playing key roles in both primary and secondary metabolism. From a chemical perspective, interest in CYP enzymes has been driven by their ability to catalyze a range of reactions, usually based on formal oxidation chemistry, including hydroxylation of carbon atoms and hetereoatoms, dealkylation, epoxidation, oxidative CC bond cleavage and formation, aromatic dehalogenation and aromatic ring coupling [5]. Figure 9.11 shows a number of the key reactions catalyzed by CYPs. Of the various known reactions, perhaps the one that has generated the most interest is the insertion of an atom from atmospheric oxygen into an unactivated CH bond. This is a challenging reaction from a chemical perspective; but, catalyzed by CYP enzymes, the reaction often shows high stereoselectivity or regioselectivity, as seen in the biosynthesis of natural products such as steroids and polyketides. Equally important is the functioning of CYP enzymes in animals in the pathways for the detoxification and elimination of xenobiotic compounds. This latter role is especially relevant to the critical process of drug metabolism [79,80].

9.4.1 Advantages of Using CYP Enzymes for Producing Drug Metabolites Producing metabolites in vitro in amounts sufficient for structural identification and preliminary toxicity and activity testing poses significant challenges. Most drugs contain between10 and 20 sites susceptible to CYP-catalyzed reactions. Selective hydroxylation or dealkylation is quite difficult in such compounds, particularly when many CH bonds have similar chemical reactivity. Although chemical synthetic methods have been traditionally employed to produce metabolites, the synthetic route often requires many steps and months of time to develop and carry out, resulting in a low yield and a high cost in both time and materials. In addition, since the structure of a metabolite is usually not known at the outset (indeed, a main reason for preparing metabolites is so that the chemical structures can be determined), a chemist must make an educated guess at the structure of a desired metabolite, synthesize that molecule, and then compare the molecule synthesized with the actual metabolite produced in vivo. Significant time can be lost when the chemist guesses wrongly and the compounds do not match. An alternative to chemical synthesis is to use human CYP enzymes to generate the desired human drug metabolites. Various means of making human P450s have been used, all with certain drawbacks [81]. The most common source is pooled HLMs, which has been described in detail previously, but these microsomal preparations contain a mixture of many different enzymes, and their cost, batch-to-batch variability in activity and restrictions on availability can limit the usefulness of HLMs for preparative synthetic work. These limitations can become particularly acute when the required amount of a pure metabolite exceeds 5–10 mg. Liver microsomes from alternative animal sources have been used as in in vitro surrogate for HLMs [82,83]. The enzymes produced from animal sources are similar to their human

203


Alkyl Hydroxylation OH

OH

O

O OH Testosterone

6-β-Hydroxytestosterone HO

O

4-Hydroxy-α -Thujone

O

O 7-Hydroxy–Thujone

α-Thujone OH

Aromatic Hydroxylation HO2C

HO2C H N

Cl

Cl

Cl

Diclofenac

OH

4-Hydroxydiclofenac

H N

Cl

Cl

H N

Cl

H N

HO

O

O

O

O Chloroxazone

6-Hydroxychloroxazone

Heteroatom Oxidation N N

Cl

N

N H

Clozapine

Figure 9.11

N+ O Cl

N

N

N H Clozapine N-Oxide

Examples of P450 reactions important in metabolite formation

204


Epoxidation O

Styrene

Styreneoxide

Dehydrogenation/Aromatization H N

H3C

CH3

H 3C

N

CH3

3A4 H3CO2C

CO2CH3 NO2

H3CO2C

Nifedipine

CO2CH3 NO2

Oxidized nifedipine

O-Dealkylation H N

H N O

O Phenacetin H N

H3CO

O

HO

4-Acetamidophenol N

H N

HO

Harmine

N

Harmol

N-Dealkylation O

O

N

NH

Dextromethorphan O OH

Propranolol

Methoxymorphinan

N H

O

NH2 OH

Desisopropyl propranolol

Figure 9.11 (continued)


205

counterparts, but not identical, sometimes resulting in different metabolic profiles or even different metabolites being generated. Additionally, although some of the same caveats apply for animal liver microsomes as for HLMs, ethical considerations, availability and cost are all improved. Ideally, metabolite synthesis would be carried out using a source of isolated, single human CYP enzymes free of other activities. The cloning and expression of human CYP genes has made such isolated CYP enzymes available, and many of the problems encountered in the use of microsomes or in vivo systems can be solved through the use of recombinant human CYPs. The biggest advantage to the use of recombinant human CYPs is the ability to generate a desired metabolite in a straightforward one-step reaction. No guesswork is required, as is the case for chemical synthesis. Since the same enzyme that is responsible for metabolism of the drug in the human liver is being used for the synthesis, the desired human metabolite will be produced. The expression of all the important human liver CYPs has been demonstrated in various host–vector systems, including bacteria, yeast and insect cells [6,84–86]. The human liver CYPs are not always highly expressed, but the levels achieved have been sufficient for metabolite profiling, small-scale screening reactions and the preparation of most metabolites at the laboratory scale. Human CYPs are multicomponent enzyme systems, requiring at a minimum the CYP enzyme component and a reductase component to be functional. The reductase requires a reduced nicotinamide cofactor, typically NADPH, and this cofactor must be regenerated to provide a steady supply of reducing equivalents for the reductase. Regeneration is accomplished with a separate substrate and enzyme. Glucose-6-phosphate and glucose-6-phosphate dehydrogenase have been widely used for this purpose. The overall complexity of the reaction mixtures and their cost have been barriers to the widespread use of recombinant human CYPs for metabolite synthesis in the past.

9.4.2 Human Cytochrome Biocatalysts As membrane-associated enzymes, human CYPs have been traditionally produced as microsomal preparations by isolating the membrane fraction using ultra-centrifugation. Recently, a product developed and offered by Codexis, Inc. has dramatically simplified the production of metabolites using recombinant human CYPs. The human cytochrome biocatalyst (HCB) is a human CYP enzyme preparation produced from a synthetic, redesigned gene and formulated with all the components necessary for a fully functional catalytic CYP system. Unlike other preparations of recombinant human cytochromes that are only offered as microsomes, HCB is available as a lyophilized powder containing the P450 enzyme, the reductase, NADPH, an enzymatic cofactor regeneration system and an optimized reaction buffer. As a lyophilized preparation the HCB is easier to ship, store and handle; and because it is fully formulated with all the necessary components for activity, the user need only add water and the drug to be metabolized. All the major human liver CYPs are available as HCBs, allowing each human CYP enzyme to be screened in parallel to determine which one is the active enzyme for metabolism of a given drug. Once identified, the most active human CYP can be scaled up and used to produce larger quantities of the desired metabolite. Some examples are illustrative of the use of the use of HCBs for metabolite synthesis. Table 9.4 shows the reaction of various HCBs for the production of metabolites. Reactions are typically complete within 4 h. In the examples shown, conversions range from 70 to 100%; in

206


Table 9.4 Metabolite production of selected drugs using human cytochrome biocatalysts available from Codexis, Inca Biocatalyst Testosterone Nifedipine Dextromethorphan Diclofenac

HCB HCB HCB HCB

3A4 3A4 2D6 2C9

Drug (mg L1)

Conversion (%)

Metabolite(s) (mg L1)

Reaction time (h)

100 69 54 30

70 70 89 100

74 48 48 32

4 3 4 1

Each reaction was performed with a CYP biocatalyst concentration of 1 mM (1000 nmol L1), in the presence of a corresponding CYP reaction mix containing reduced nicotinamide cofactor and a cofactor recycling system at 30 C, with agitation to promote oxygen transfer to the reaction solution.

a

practice, conversions are highly variable and depend on the starting drug and its rate of conversion, its solubility and product inhibition by the metabolite.

9.4.3 Microbial CYP Enzymes Although microbial cell biotransformations have been used for the preparation of metabolites as described earlier, isolated microbial CYP enzymes are not generally used for this purpose. Given the multicomponent nature of most of these enzymes and the difficulties in formulating them to be active, stable biocatalysts, little work has been carried out to establish such isolated enzyme systems for metabolite synthesis. One class of CYP enzymes, the CYP102A1 class, has been the subject of intensive research as a platform for developing practical metabolite synthesis catalysts. The distinctive feature of the CYP102A1 enzymes is that they are catalytically self-sufficient, meaning that both the CYP and reductase components are contained in a single-chain polypeptide chain. As a single protein, the problems of expression and formulation are greatly reduced. The best-studied member of this class, the CYP102A1 from Bacillus megaterium, or so-called BM3 enzyme, is both robust and highly active on its natural fatty acid substrate. The main problem with the BM3 enzyme and its analogs is the poor substrate range of the native enzyme: drug-like compounds are hydroxylated either very slowly or not at all in the presence of native BM3. Recent work by Arnold and coworkers has demonstrated that the substrate range of BM3 can be expanded by directed evolution [7,87,88]. A large number of mutants of BM3 have been produced and characterized, and many of these enzymes have been shown to be equal to or greater in activity for the production of metabolites than the human CYPs. In addition, the greater stability of the BM3 variants enables longer reaction times, higher substrate concentrations and higher yields, offering the possibility that these reactions can use a viable alternative to chemical synthesis, even for the preparation of gram quantities of metabolite. Identifying the best BM3 mutant has been facilitated by the availability of a product developed and offered by Codexis, Inc.: the Codex MicroCyp Plate. Containing 96 different BM3 mutants and offered in a microtiter plate format as preformulated enzyme systems for screening, the Codex MicroCyp Plate enables the rapid identification of the best BM3 mutant for a target substrate. Since each mutant gene has been cloned into a production host, larger quantities of any desired mutant can be produced, facilitating reaction optimization and scale-

207

Biosynthesis of Drug Metabolites Table 9.5 Metabolite distribution upon diclofenac hydroxylation with the MicroCyp Plate MCyp Variant 1 2 3 4 5 6

(5-OH) (%) 58 24 90 90 20 73

Yield (%) 40 12 20 45 31 25

up to produce larger quantities of metabolite. Table 9.5 shows a typical screening result for the hydroxylation of diclofenac. Multiple hits were identified, with enzymes producing both the 5-hydroxy and 40 -hydroxy derivatives. As a can be seen in the table, MicroCyp variant 3 produced predominantly the 5-hydroxy derivative, whereas MicroCyp variant 4 produced largely the 40 -hydroxy derivative. By selecting the appropriate variant, either compound can be generated in at least 90% purity in the crude reaction mixture. Further purification by HPLC, made easier by the relative selectivity of the enzymatic hydroxylation, affords pure compound. For scaled-up synthesis of larger amounts of metabolite, the following example is illustrative of the use of the BM3 mutant enzymes. A reaction mixture containing 25 mL MicroCyp reaction mix, 1 mM MicroCyp variant 1, and 1 mM diclofenac were placed into 250 mL baffled Erlenmeyer flasks and incubated with agitation at 30 C. The reaction was complete in less than an hour, reaching 75% conversion (0.23 g L1). The concentration of product achieved in this reaction with the BM3 mutant CYP enzyme represents an improvement of more than an order of magnitude compared with the reaction carried out with the recombinant human CYP2C9 (0.02 g L1).

9.5 Summary Metabolite biosynthesis has demonstrated its utility in drug metabolite preparation and characterization, and contributed to drug discovery and development. Although metabolite biosynthesis is a prerequisite step for metabolite structure elucidation in many cases, it is complementary to chemical synthesis in large-scale metabolite preparations. The merits for using these techniques should be determined on a case-by-case fashion. New techniques, such as recombinant enzyme and microbial glucuronidation systems, would have a great impact on the field.

Acknowledgments We thank Dr W. Griffith Humphreys and Dr Mingshe Zhu for critical comments and helpful discussion.

References [1] Zhou, S., Chan, E., Duan, W. et al. (2005) Drug bioactivation, covalent binding to target proteins and toxicity relevance. Drug Metabolism Reviews, 1, 41–213.

208


[2] Kalgutkar, A.S., Gardner, I., Obach, R.S. et al. (2005) A comprehensive listing of bioactivation pathways of organic functional groups. Current Drug Metabolism, 6, 161–225. [3] Mark, J., Bailey, M.J., Ronald, G. and Dickinson, R.G. (2003) Acyl glucuronide reactivity in perspective: biological consequences. Chemico-Biological Interactions, 145, 117–137. [4] Ackermann, B.L., Michael, J., Berna, M.J. et al. (2002) Recent advances in use of LC/MS/MS for quantitative high-throughput bioanalytical support of drug discovery. Current Topics in Medicinal Chemistry, 2, 53–66. [5] US Food Drug Adminstration (2008) Guidance for Industry, Safety Testing of Drug Metabolites. http://www.fda. gov/cder/guidance/index.htm (last access October 2008). [6] Vail, R.B., Homann, M.J., Hanna, I. and Zaks, A. (2005) Preparative synthesis of drug metabolites using human cytochrome P450s 3A4, 2C9 and 1A2 with NADPH P450 reductase expressed in Escherichia coli. Journal of Industrial Microbiology & Biotechnology, 32, 67–74. [7] Otey, C.R., Bandara, G., Lalonde, J.L. et al. (2006) Preparation of human metabolites of propranolol using laboratory-evolved bacterial cytochromes P450. Biotechnology and Bioengineering, 93, 494–499. [8] Guengerich, F.P. (1997) Cytochrome P450 enzymes, in Commprehensive Toxicology (eds, Guengerich, F.P., Sipes, I.G., McQueen, A. and Gondolfi A.J.), Pergamon, Cambridge, pp. 37–68. [9] Parkinson, A., Thomas, P.E., Ryan, D.E. et al. (1983) Differential time course of induction of rat liver microsomal cytochrome P 450 isozymes and epoxide hydrolase by Aroclor 1254. Archives of Biochemistry and Biophysics, 225, 203–215. [10] Guengerich, F.P. (2001) Common and uncommon cytochrome P450 reactions related to metabolism and chemical toxicity. Chemical Research in Toxicology, 14, 611–650. [11] Ziegler, D.M. (2002) An overview of the mechanism, substrate specificities, and structure of FMOs. Drug Metabolism Reviews, 24, 503–511. [12] Cashman, J.R. (2005) Some distinctions between flavin-containing and cytochrome P450 monooxygenases. Biochemical and Biophysical Research Communications, 338, 599–604. [13] Fisher, M.B., Paine, M.F., Strelevitz, T.J. and Wrighton, S.A. (2001) The role of hepatic and extrahepatic UDPglucuronosyltransferases in human drug metabolism. Drug Metabolism Reviews, 33, 273–297. [14] Akira, K., Uchijima, T. and Hashimot, T. (2002) Rapid internal acyl migration and protein binding of synthetic probenecid glucuronides. Chemical Research in Toxicology, 15, 765–772. [15] Yamazaki, H., Ueng, Y.F., Shimada, T. and Guengerich, F.P. (1995) Roles of divalent metal ions in oxidations catalyzed by recombinant cytochrome P450 3A4 and replacement of NADPH–cytochrome P450 reductase with other flavoproteins, ferredoxin, and oxygen surrogates. Biochemistry, 34, 8380–8389. [16] Itagaki, K., Carver, G.T. and Philpot, R.M. (1996) Expression and characterization of a modified flavin-containing monooxygenase 4 from humans. The Journal of Biological Chemistry, 271, 2012–20107. [17] Sabourin, P.J. and Hodgson, E. (1984) Characterization of the purified microsomal FAD-containing monooxygenase from mouse and pig liver. Chemico-Biological Interactions, 51, 125–139. [18] Ciotti, M. and Owens, I.S. (1996) Evidence for overlapping active sites for 17a-ethynlestradiol and bilirubin in the human major bilirubin UDPglucuronosyltransferase. Biochemistry, 35, 10119–10124. [19] Basu, N.K., Ciotti, M., Hwang, M.S. et al. (2004) Differential and special properties of the major human UGT1encoded gastrointestinal UDP-glucuronosyltransferases enhance potential to control chemical uptake. The Journal of Biological Chemistry, 279, 1429–1441. [20] Zenser, T.V., Lakshmi, V.M. and Davis, B.B. (1999) Human and Escherichia coli b-glucuronidase hydrolysis of glucuronide conjugates of benzidine and 4-aminobiphenyl, and their hydroxy metabolites. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 27, 1064–1067. [21] Schrag, M.L. and Wienkers, L.C. (2000) Topological alteration of the CYP3A4 active site by the divalent cation Mg2þ . Drug Metabolism and Disposition: The Biological Fate of Chemicals, 28, 1198–1201. [22] Fisher, M.B., Campanale, K., Ackermann, B.L. et al. (2000) In vitro glucuronidation using human liver microsomes and the pore-forming peptide alamethicin. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 28, 560–566. [23] Stevenson, D.E. and Hubl, U. (1999) Optimization of b-D-glucuronide synthesis using UDP-glucuronyl transferase Enzyme and Microbial Technology 24, 388–396. [24] Meech, R. and Mackenzie, P.I. (1997) Structure and function of uridine diphosphate glucuronosyltransferase. Clinical and Experimental Pharmacology & Physiology, 24, 907–915. [25] Soars, M.G., Ring, B.J. and Wrighton, S.A. (2003) The effect of incubation conditions on the enzyme kinetic of UDP-glucuronsyltransferases. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 31, 762–767.


209

[26] Hansen, K.T. and Stentoft, K. (1995) Characterization of benzazepine UDP-glucuronosyl-transferases in laboratory animals and man. Xenobiotica; The Fate of Foreign Compounds in Biological Systems, 25, 611–622. [27] Rowland, A., Knights, K.M., Mackenzie, P.I. and Miners, J.O. (2008) The ‘albumin effect’ and drug glucuronidation: bovine serum albumin and fatty acid free serum albumin enhance the glucuronidation of UGT1A9 substrates but not UGT1A1 and UGT1A6 activities. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 36, 1056–1062. [28] Voznesensky, A.I. and Schenkman, J.B. (1992) Inhibition of cytochrome P450 reductase by polyols has an electrostatic nature. European Journal of Biochemistry, 210, 741–746. [29] Kushinsky, S. and Chen, V.L. (1967) The inhibition of b-glucuronidase from bovine liver by 1,4-saccharolactone. Comparative Biochemistry and Physiology, 20, 535–542. [30] Hickman, D., Wang, J.-P., Wang, Y. and Unadkat, J.D. (1998) Evaluation of the selectivity of in vitro probes and suitability of organic solvents for the measurement of human cytochrome P450 monooxygenase activity. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 26, 207–215. [31] Chauret, N., Gauthier, A. and Nicoll-Griffth, D.A. (1998) Effect of common solvents on in vitro cytochrome P450 mediated metabolic activities in human liver microsomes. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 26, 1–4. [32] Easterbrook, J., Liu, C., Sakai, Y. and Li, A.P. (2001) Effects of organic solvents on the activities of cytochrome P450 isoforms, UDP-dependent glucuronyl transferase, and phenol sulfotransferase in human hepatocytes. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 29, 141–144. [33] Dehal, S.S., Gagne, P.V., Crespi, C.L. and Parren, C.J. (2002) Effect of common organic solvent on human UGT enzyme activities. 11th North American ISSX Meeting, October 27–31, 2002, Orlando, Florida, USA, Abstract #370. [34] Forti, G.C., Paolini, M., Hrelia, P. et al. (1984) NADPH-generating system: influence on microsomal monooxygenase stability during incubation for the liver microsomal assay with rat and mouse S9 fractions. Mutation Research, 129, 291–297. [35] Seelbach, K., Riebel, B., Hummel, W. et al. (1996) A novel, efficient regenerating method of NADPH using a new formate dehydrogenase. Tetrahedron Letters, 37, 1377–1380. [36] Zhang, H., Zhang, D., Li, W. et al. (2007) Reduction of site-specific CYP3A-mediated metabolism for dual angiotensin and endothelin receptor antagonists in various in vitro systems and in cynomolgus monkeys. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 35, 795–805. [37] Ogilvie, B.W., Zhang, D., Li, W. et al. (2006) Glucuronidation converts gemfibrozil to a potent, metabolismdependent inhibitor of CYP2C8: implications for drug–drug interactions. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 34, 191–197. [38] Smith, R.V. and Rosazza, J.P. (1974) Microbial models of mammalian metabolism. Aromatic hydroxylation. Archives of Biochemistry and Biophysics, 161, 551–558. [39] Smith, R.V. and Rosazza, J.P. (1975) Microbial models of mammalian metabolism. Journal of Pharmaceutical Sciences, 64, 1737–1759. [40] Abourashed, E.A., Clark, A.M. and Hufford, C.D. (1999) Microbial models of mammalian metabolism of xenobiotics: an updated review. Current Medicinal Chemistry, 6, 359–374. [41] Azerd, R. (1999) Microbial models for the drug metabolism. Advances in Biochemical Engineering/Biotechnology, 63, 169–218. [42] Moody, J.D., Freeman, J.P. and Cerniglia, C.E. (1999) Biotransformation of doxepin by Cunninghamella elegans. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 27, 1157–1164. [43] Briggs, B.S., Baker, P.J., Belvo, M.D. et al. (1999) Microbial process for the preparation of glucuronides of raloxifene. Journal of Industrial Microbiology & Biotechnology, 23, 194–197. [44] Duhart, B.T., Zhang, D., Deck, J. et al. (1999) Biotransformation of protriptyline by filamentous fungi and yeast. Xenobiotica, 7, 733–746. [45] Sutherland, J.B., Freeman, J.P., Heinze, T.M. et al. (2001) Oxidation of phenothiazine and phenoxazine by Cunninghamella elegans. Xenobiotica; The Fate of Foreign Compounds in Biological Systems, 31, 799–809. [46] Moody, J.D., Freeman, J.P., Fu, P.P. and Cerniglia, C.E. (2002) Biotransformation of mirtazapine by Cunninghamella elegans. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 30, 1274–1279. [47] Sun, L., Huang, H.-H., Liu, L. and Zhong, D.-F. (2004) Transformation of verapamil by Cunninghamella blakesleeana. Applied and Environmental Microbiology, 70, 2722–2727. [48] Ohta, K., Agematu, H., Yamada, T. et al. (2005) Production of human metabolites of cyclosporin A, AM1, AM4N and AM9, by microbial conversion. Journal of Bioscience and Bioengineering, 94, 390–395.

210


[49] Hu, S., Martinez, C.A., Tao, J. and Yazbeck, D.R. (2005) Efficient microbial preparation of capravirine metabolites M4 and M5. U.S. Pat. Appl. Publ. US 2005043363 A1, Feb. 24, 2005. [50] Xie, Z.Y., Huang, H.H. and Zhong, D.F. (2005) Biotransformation of pantoprazole by the fungus Cunninghamella blakesleeana. Xenobiotica; The Fate of Foreign Compounds in Biological Systems, 35, 467–477. [51] Huang, H., Yang, X., Li, Q. et al. (2006) Biotransformation of tolbutamide to 40 -hydroxytolbutamide by the fungus Cunninghamella blakesleeana. Applied Microbiology and Biotechnology, 72, 486–491. [52] Barbuch, R.J., Campanale, K., Hadden, C.E. et al. (2006) In vivo metabolism of [14 C]ruboxistaurin in dogs, mice, and rats following oral administration and the structure determination of its metabolites by liquid chromatography/mass spectrometry and NMR spectrometry. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 34, 213–224. [53] Zhang, D., Zhang, H., Aranibar, N. et al. (2006) Structural elucidation of human oxidative metabolites of muraglitazar: use of microbial bioreactors in the biosynthesis of metabolite standards. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 34, 267–280. [54] Zhang, D., Hanson, R., Roongta, V. et al. (2006) In vitro and in vivo metabolism of a gamma-secretase inhibitor BMS-299897 and generation of active metabolites in milligram quantities with a microbial bioreactor. Current Drug Metabolism, 7, 883–896. [55] Zmijewski, M., Gillespie, T.A., Jackson, D.A. et al. (2006) Application of biocatalysis to drug metabolism: preparation of mammalian metabolites of a biaryl-bis-sulfonamide AMPA (a-amino-3-hydroxy-5-methylisoxazole-4propionic acid) receptor potentiator using Actinoplanes missouriensis. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 34, 925–931. [56] Li, L., Liu, R., Ye, M. et al. (2006) Microbial metabolism of evodiamine by Penicillium janthinellum and its application for metabolite identification in rat urine. Enzyme and Microbial Technology, 39, 561–567. [57] Osorio-Lozada, A., Surapaneni, S., Skiles, G. and Subramanian, R. (2008) Biosynthesis of drug metabolites using microbes in hollow fiber cartridge reactors: case study of diclofenac metabolism by Actinoplanes sp. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 36, 234–240. [58] Li, W., Josephs, J.L., Skiles, G. and Humphreys, W.G. (2008) Metabolite generation via microbial biotransformation with actinomycetes: rapid screening methods and synthesis of important human metabolites of two development stage compounds, BMS-587101 and dasatinib. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 36, 721–730. [59] Duetz, W.A., Buedi, L., Hermann, R. et al. (2000) Methods for intense aeration, growth, storage, and replication of bacterial strains in microtiter plates. Applied and Environmental Microbiology, 66, 2541–2646. [60] Duetz, W.A. and Witholt, B.W. (2001) Effectiveness of orbital shaking for the aeration of suspended bacterial cultures in square-deepwell microtiter plates. Biochemical Engineering Journal, 7, 113–115. [61] Rao, G.P. and Davis, P.J. (1997) Microbial models of mammalian metabolism: biotransformation of HP 749 (besipirdine) using Cunninghamella elegans. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 25, 709–715. [62] Lacroix, I., Biton, J. and Azerad, R. (1997) Microbial biotransformation of a synthetic immunomodulating agent, HR325. Bioorganic and Medicinal Chemistry, 5, 1369–1380. [63] Cannell, R.J., Knaggs, A.R., Dawson, M.J. et al. (1995) Microbial biotransformation of the angiotensin II antagonist GR117289 by Streptomyces rimosus to identify a mammalian metabolite. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 23, 724–729. [64] Cannel, R.J.P., Rashid, T., Ismail, I.N. et al. (1997) Novel metabolites of warfarin produced by Beauveria bassiana and Streptomyces rimosus: a novel application of HPLC–NMR. Xenobiotica; The Fate of Foreign Compounds in Biological Systems, 27, 147–157. [65] Moussa, C., Houziaux, P., Danree, B. and Azerad, R. (1997) Microbial models of mammalian metabolism. Fungal metabolism of phenolic and nonphenolic p-cymene-related drugs and prodrugs. I. Metabolites of thymoxamine. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 25, 301–310. [66] Lacroix, I., Biton, J. and Azerad, R. (1999) Microbial Models of drug metabolism: microbial transformations of Trimegestone (RU27987), a 3-keto-D4,9(10)-19-norsteroid drug. Bioorganic and Medicinal Chemistry, 7, 2329–2341. [67] Hezari, M. and Davis, P.J. (1993) Microbial models of mammalian metabolism: furosemide glucoside formation using the fungus Cunninghamella elegans. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 21, 259–267.


211

[68] Yang, W. and Davis, P.J. (1992) Microbial models of mammalian metabolism: biotransformation of Nmethylcarbazole using the fungus Cunninghamella echinulata. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 20, 38–46. [69] Chen, T.S., White, S.R. and Monaghan, R.L. (1993) Microbial hydroxylation and glucuronidation of the angiotensin II (AII) receptor antagonist MK 954. The Journal of Antibiotics, 46, 131–134. [70] Kirkman, S.K., Zhang, M.-Y., Horwatt, P.M. and Scatina, J. (1998) Isolation and identification of bromfenac glucoside from rat bile. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 26, 720–723. [71] Savage, R.E., Tyler, A.N., Miao, X.-S. and Chan, T.C.K. (2008) Identification of a novel glucosylsulfate conjugate as a metabolite of ARQ 501 (b-lapachone) in mammal. Drug Metabolism and Disposition: The Biological Fate of Chemicals, DOI: 10.1124/dmd.107.018655. [72] Cerniglia, C.E., Freeman, J.P. and Mitchum, R.K. (1982) Glucuronide and sulfate conjugation in the fungal metabolism of aromatic hydrocarbons. Applied and Environmental Microbiology, 43, 1070–1075. [73] Li, W., Leet, J.E. and Lam, K.S. (2000) Nocathiacin antibiotic derivatives prepared by microbial biotransformation. PCT Int. Appl. WO 2000014100 A1, March 16, 2000. [74] Freitag, D.G., Foster, R.T., Coutts, R.T. et al. (1997) Stereoselective metabolism of rac-mexiletine by the fungus Cunninghamella echinulata yields the major human metabolites hydroxymethylmexiletine and p-hydroxymexiletine. Drug Metabolism and Disposition: The Biological Fate of Chemicals, 25, 685–692. [75] Li, W., Matson, J.A., Xiaohua, H. et al. (2004) Microbial transformation method for the preparation of an epothilone. US patent, US 6,780,620 B1, August 24, 2004. [76] Hosobuchi, M., Kurosawa, K. and Yoshikawa, H. (1993) Application of computer on monitoring and control of fermentation process: microbial conversion of ML-236B sodium to pravastatin. Biotechnology and Bioengineering, 42, 815–820. [77] Christopher, L., Cui, D., Wu, C. et al. (2008) Metabolism and disposition of dasatinib after oral administration to humans. Drug Metabolism and Disposition: The Biological Fate of Chemicals, DOI: 10.1124/dmd.107.018267. [78] Li, W., Zhao, W., Yao, M. et al. (2008) Conjugated metabolite generation via microbial biotransformation with actinomycetes: a novel biosynthesis methodology for glucuronides and glucosides. The 2nd Asian Pacific Regional ISSX Meeting, Shanghai, China. May 11–13; Abstract/poster #90. [79] Guengerich, F.P. (2004) Cytochrome P450: what have we learned and what are the future issues? Drug Metabolism Reviews, 36, 159–197. [80] Bernhardt, R. (2006) Cytochromes P450 as versatile biocatalysts. Journal of Biotechnology, 124, 128–145. [81] Rushmore, T.H., Reider, P.J., Slaughter, D. et al. (2000) Bioreactor systems in drug metabolism: synthesis of cytochrome P450-generated metabolites. Metabolic Engineering, 2, 115–125. [82] Ueng, Y.F., Yu, H.J., Lee, C.H. et al. (2005) Identification of the microsomal oxidation metabolites of rutaecarpine, a main active alkaloid of the medicinal herb Evodia rutaecarpa. Journal of Chromatography. A, 1076, 103–109. [83] Hubl, U. and Stevenson, D.E. (2001) In vitro enzymic synthesis of mammalian liver xenobiotic metabolites catalyzed by ovine liver microsomal cytochrome P450. Enzyme and Microbial Technology, 29, 306–311. [84] Parikh, A., Gillam, E.M. and Guengerich, F.P. (1997) Drug metabolism by Escherichia coli expressing human cytochromes P450. Nature Biotechnology, 15, 784–788. [85] Guengerich, F.P. (2002) Cytochrome P450 enzymes in the generation of commercial products. Nature Reviews. Drug Discovery, 1, 359–366. [86] Lamb, D.C., Waterman, M.R., Kelly, S.L. and Guengerich, F.P. (2007) Cytochromes P450 and drug discovery. Current Opinion in Biotechnology, 18, 504–512. [87] Landwehr, M., Hochrein, L., Otey, C.R. et al. (2006) Enantioselective a-hydroxylation of 2-arylacetic acid derivatives and buspirone catalyzed by engineered cytochrome P450 BM-3. Journal of the American Chemical Society, 128, 6058–6059. [88] Wong, T.S., Arnold, F.H. and Schwaneberg, U. (2004) Laboratory evolution of cytochrome P450 BM-3 monooxygenase for organic cosolvents. Biotechnology and Bioengineering, 85, 351.

10 Application of Whole-Cell Biotransformation in the Pharmaceutical Industry Kin Sing Lam Nereus Pharmaceuticals, Inc., Department of Microbiology and Industrial Fermentation, San Diego, CA 92121, USA

10.1 Introduction Biotransformation is the use of biological systems, including whole cells, organs or isolated enzymes, to catalyze the conversion of organic compounds from synthetics and natural products. There are many applications of biotransformation in the pharmaceutical industry, including synthesis of drug metabolites for metabolism studies, lead optimization, chiral resolution and the creation of libraries of derivatives from highly diverse lead compounds for structure–activity relationship (SAR) and screening studies. A few of these applications have been reviewed in this book. Studies in the past decade have witnessed a tremendous growth in the applications of biotransformation in the pharmaceutical industries [1–3]. These studies have provided complementary approaches and sometimes powerful alternatives to conventional synthetic chemical techniques, due to the following advantages and traits of biotransformation processes: . . . . . .

stereoselectivity; regioselectivity; creating new functional groups at nonactivated sites; modification of complex molecules without the need of protection/deprotection steps; cloning, overexpression and tailor-made biocatalysts; immobilization and reusability of biocatalysts;

Biocatalysis for the Pharmaceutical Industry edited by Junhua Tao, Guo-Qiang Lin, and Andreas Liese Ó 2009 John Wiley & Sons Asia (Pte) Ltd

214 . .


mild reaction conditions at ambient temperature and atmospheric pressure; waste minimization.

10.1.1 Whole-Cell Biotransformation Processes Used in Commercial Production of Pharmaceuticals Whole-cell biotransformation processes have been successfully applied for commercial production of pharmaceuticals, either as the drug substance itself or as an intermediate for the synthesis of the final drug substance. Some examples of the whole-cell biotransformation processes used by pharmaceutical industry are shown in Table 10.1. The structures of the biotransformation products are shown in Figure 10.1. The examples of the whole-cell biotransformation processes shown in Table 10.1 illustrate the aforementioned advantages of the biotransformation reaction. In addition to the stereoselectivity, mild reaction conditions and waste minimization essentially demonstrated by all 18 whole-cells processes, the regioselectivity and creation of new functional groups at nonactivated sites are demonstrated by the hydroxylation production of 11a-hydroxypro-gesterone, b-hydroxy-isobutyric acid, and b-hydroxy-n-butyric acid. Recombinant microorganisms have been engineered for the production of (R)-ethyl-4,4, 4-trifluoro-3-hydroxybutanoate, (S)-2,2-dimethyl-cyclopropanecarboxamide, L-piperidine2-carboxylic acid and L-carnitine. Immobilized-cell technology has been applied for the production of D-4-hydroxyphenyl glycine, nicotinamide and D-aspartic acid. Modification of two complex natural products, progesterone and compactin, to the corresponding 11ahydroxypro-gesterone and pravastatin has been successfully carried out in high yield without any protection steps. These findings firmly establish that whole-cell biotransformation is an important complementary tool to organic synthesis in the preparation of drug substances and useful chiral pharmaceutical intermediates.

10.1.2 Application of Whole-Cell Biotransformation Process in the Synthesis of Chiral Pharmaceutical Intermediates Fourteen of the 18 whole-cell processes shown in Table 10.1 are used in the preparation of chiral intermediates for the production of the final drug substances. Whole-cell biotransformation has been routinely used to generate a wide variety of chiral pharmaceutical intermediates [2,12,13]. This includes the use of oxido-reductases and aminotransferases in the whole cells for the synthesis of chiral alcohols, aminoalcohols, amino acids and amines. Monooxygenases in the whole cells have been used in enantioselective and regioselective hydroxylation, epoxidation and Baeyer–Villiger reactions. Dioxygenases have been used in the synthesis of chiral diols. Hydrolytic enzymes in the whole cells have been applied for the resolution of a variety of racemic compounds and in the asymmetric synthesis of enantiomerically enriched chiral compounds. Aldolases and decarboxylases have been effectively used in asymmetric synthesis by aldol and acyloin condensation reactions. The production of single enantiomers of drug intermediates is increasingly important in the pharmaceutical industry. Whole-cell biotransformation is expected to play a significant role in this rapidly growing area of industry.

Hydrolysis

Rhodococcus rhodochrous Pseudomonas fluorescens Arthrobacter sp. Comononas acidivorans and recombinant E. coli Multi-enzymatic Agrobacterium sp. pathway Lyase Saccharomyces cerevisiae Decarboxylation Pseudomonas dacunhae Lyase Erwinia herbicola

Antibacterial Treatment of pellagra Anticancer Treatment of migraine Dehydropeptide inhibitor Thyroid inhibitor Treatment for asthma Antibacterial Treatment of Parkinsonism

(R)-Ethyl-4,4-4-trifluoro-3Belfoxatone hydroxybutanoate 8 D-4-Hydroxyphenyl glycine 9 Ampicillin/ amoxycillin Nicotinamide 10 Nicotinamide (S)-Pipecolic acid 11 Incel CBZ-D-proline 12 Eletriptan (S)-2,2-Dimethyl-cycloproCilastatin panecarbox-amide 13 D-Carnitine 14 L-Carnitine

Phenylacetylcarbinol 15 D-Aspartic acid 16 3,4-Dihydroxy-Lphenylalanind (L-DOPA) 17

Ephedrine Apoxycillin L-DOPA

Reduction

Trusopt

Hydrolysis Hydrolysis Hydrolysis Hydrolysis

Reduction Recombinant Escherichia coli Bacillus brevis

Neurospora crassa

[7] [7] [7] [7]

[8]

[7]

[6]

[7]

[6]

[4] [5] [6]

[4] [4]

Ref.

Single-stage fermentation of [7] four processes Single-stage fermentation [9] Immobilized cells [10] Single-stage fermentation [11]

Immobilized cells Single-stage fermentation Single-stage fermentation One-pot, two-step processes

Two-phase water–butyl acetate Immobilized cells

Single-stage fermentation


(4S,6S)-Hydroxy-sulfone 7

Oxidation

Pseudomonas oleovorans

Treatment for osteoporosis Treatment for glaucoma Anti-depressant


Pseudomonas putida

Risedronate

Oxidation

Single-stage fermentation Single-stage fermentation Single-stage fermentation

Single-stage fermentation Single-stage fermentation

Process

Candida rugosa Streptomyces carbophilus Pseudomonas putida

Anti-diabetic

Hydroxylation Hydroxylation Oxidation

Glipicide

Rhizopus arrhius Candida rugosa

Microorganism

Carbapenem Pravastatin Acipimox

Hydroxylation Hydroxylation

Reaction type

b-Hydroxy-n-butyric acid 3 Pravastatin 4 5-Methylpyrazine-2carboxylic acid 5 5-Methylpyrazine-2carboxylic acid 5 Pyridyl-3-acetic acid 6

Indication Anti-inflammation Treatment for hypertension Antibacterial Anti-cholesterol Anti-lipolytic

11a-Hydroxypro-gesterone 1 Cortisone b-Hydroxy-isobutyric acid 2 Captopril

Final product

Commercial whole-cell biotransformation processes for the preparation of pharmaceuticals and pharmaceutical intermediates

Biotransformation Product

Table 10.1

Application of Whole-Cell Biotransformation in the Pharmaceutical Industry 215

216


O

O

H

HO

O

OH HO

O 1

2

H

6

COOH

COOH

HO

3

4

OH O N

OH

OH

OH N

N 5

H2N

S

S

O

O

6

F3C

CO2Et

O 7

8

COOH

N

COOH O

O

CONH2 HO 9

COOH

N H

N 10

11

12

O

HO OH

N

13

COOH

HOOC

OH

O

HO

NH2

COOH

14

15

16

COOH NH2

HO 17

Figure 10.1

OH

NaOOC HO

Structures of biotransformation products from Table 10.1

Application of Whole-Cell Biotransformation in the Pharmaceutical Industry

217

10.2 Disadvantages of Whole-Cell Process Compared with the Isolated Enzyme Process Although whole-cell biotransformation has been established as a valuable tool for the synthesis of pharmaceuticals and pharmaceutical intermediates, there are several inherent problems associated with this process that have hindered the advancement of this technology in the pharmaceutical industry. Attempts to clear some of these obstacles have recently allowed the use of isolated enzyme in the biotransformation process to garner greater attention than whole-cell biotransformation, since the reaction mechanism and kinetics of a single biocatalyst are simpler than the complex biochemical synthetic methods which use whole cells. In comparison with the whole-cell processes, fewer side-products are formed in enzymatic transformations, complex expensive fermentors are not required, aeration, agitation and sterility need not necessarily be maintained and the substrate is not diverted into the formation of a de novo cellular biomass. In this and the next section, comparison of application of whole cells and isolated enzymes in biotransformation processes is made. The following are the disadvantages of the whole-cell biotransformation process compared with the isolated enzyme biotransformation process.

10.2.1 Substrate Availability and Recovery of Products in Low Concentrations Whole-cell biotransformation usually occurs in a water medium with substrate concentration rarely exceeding 10 g L 1. The substrate concentration is also affected by the mass transfer limitations imposed by the cellular membrane, which further reduces the production yield, an issue not seen with the use of isolated enzymes. The low substrate availability in the reaction leads to the formation of a large volume of aqueous solution containing a relatively low concentration of product, which imposes technical difficulty in the downstream process and adds to the manufacturing cost of the product. The ability to effectively overcome the problems of substrate availability, effective space–time yield production and product recovery is very important in achieving cost-effective production with this technology. Without much improvement of the above technical issues, the application of whole-cell biotransformation is limited to high-priced chiral pharmaceutical intermediates or high value-added chemicals.

10.2.2 Undesirable Side Reactions Whole-cell biotransformations frequently showed insufficient stereoselectivities and/or undesired side reactions because of competing enzymatic activities present in the cells. These side reactions can modify the substrates and/or products. Furthermore, whole-cell biotransformations are limited due to the intrinsic need to grow biomass, which generates its own metabolites that are not related to the biotransformation reactions and, therefore, which need to be removed during the downstream process. Both the cells themselves and the unrelated metabolites produced are impurities that need to be removed after the biotransformation reaction. With isolated enzymes, there are no organism and unrelated metabolites to remove after the biotransformation processes.

10.2.3 Toxicity of Substrate and Product The biochemical system of a microorganism is more complicated than an isolated enzyme and possesses dynamic and regulatory properties. Both substrate and product of the

218


biotransformation process can exert inhibition or toxicity to the microorganism, resulting in reduced production [14].

10.3 Advantages of Whole-Cell Process Compared with the Isolated Enzyme Process In this section, the advantages of the use of whole cells over the use of isolated enzymes in biotransformation processes are presented.

10.3.1 More Stable Sources than Isolated Enzymes Compared with isolated enzymes, enzymes used in whole-cell biotransformations are often more stable due to the presence of their natural environment inside the cell. This is especially true for the enzymes involved in the oxidation and hydroxylation reactions that are labile once isolated from the cells. They are a convenient and stable source of enzymes that are often synthesized by cells in response to the presence of the substrate.

10.3.2 Regeneration of Cofactors and Multi-Enzyme Reactions In the processes that require regeneration of cofactors such as nicotinamide adenine dinucleotide phosphate (NAD(P)H) and adenosine triphosphate (ATP), whole-cell biotransformations are more advantageous than enzymatic systems [12,15]. Whole cells also have a competitive edge over the isolated enzymes in complex conversions involving multiple enzymatic reactions [14].

10.3.3 Diversity and Availability One can easily access a collection of microbial biocatalysts with broader diversity and higher activities and selectivities than the commercially available enzymes. Most academic and industrial institutions possess microbial culture collections that can be utilized for whole-cell biotransformation processes. Furthermore, one can rapidly assemble a relatively small number of nonpathogenic, often food-grade microorganisms (such as bakers yeast and bacteria), for whole-cell biotransformation applications. The microbial biocatalysts are inexpensive, readily available and renewable. The selection of the microorganisms from the culture collection for a specific biotransformation reaction is based on the internal experience of the research group and on published information. Screening of the microbial biocatalysts takes advantage of microbial diversity as much as possible by evaluating a broad range of microbial species that have been isolated from very diverse environments.

10.3.4 Reactions with Non-Commercially Available Isolated Enzymes for Preparative-Scale Synthesis 10.3.4.1 Cytochrome P450 Monooxygenases Cytochrome P450 monooxygenases (P450s) have significant potential in biotransformation applications because their ability to insert molecular oxygen regiospecifically and


219

stereoselectively into allylic positions or into unactivated C H bonds under mild conditions has no equivalent in organic synthesis [16]. The membrane-bound nature of the majority of these enzymes and typically low activities [17], together with their functional dependence on the presence of cofactors and related electron transport proteins, has ensured that preparative biocatalytic hydroxylations are most usefully performed with whole-cell catalysts [16,17]. 10.3.4.2 Aromatic Hydrocarbon Dioxygenases Aromatic hydrocarbon dioxygenases catalyze the oxidation of the aromatic-ring hydrocarbons, compounds containing only carbon and hydrogen. Representatives of this class of enzymes are toluene dioxygenase (TDO) and naphthalene dioxygenase (NDO) [18,19]. Like P450s, several properties associated with aromatic-ring-hydroxylating dioxygenases pose challenges for their application to the targeted preparation of oxidation products. These include their cofactor requirements for reduced nicotinamide cofactor(s) NAD(P)H, the multicomponent nature of the iron–sulfur-containing subunit, low specific activity and instability. Therefore, applications based on aromatic-ring-hydroxylating dioxygenases have been based largely on whole-cell biotransformations with enzymes typically overexpressed in recombinant hosts such as Escherichia coli or Pseudomonas sp. [20,21]. An excellent example of the application of aromatic hydrocarbon dioxygenase in whole-cell biotransformation is the production of indigo to levels exceeding 18 g L 1 by overexpressed NDO from Pseudomonas putida in a recombinant Escherichia coli strain [21]. 10.3.4.3 Enoate Reductase in Reduction of Triply Substituted Double Bonds Enoate reductase reduces a,b-unsaturated carboxylate ions in an NADPH-dependent reaction to saturated carboxylated anions. Useful chiral synthons can be conveniently prepared by the asymmetric reduction of a triply substituted C C bond by the action of enoate reductase, when the double bond is activated with strongly polarizing groups [22]. Enoate reductases are not commercially available as isolated enzymes; therefore, microorganisms such as bakers yeast or Clostridium sp. containing enoate reductase are used to carry out the reduction reaction. 10.3.4.4

D-Aminoacylase

for Production of D-Amino Acid

Since there is no commercially available D-aminoacylase, the production process of D-amino acids involves cloning of the D-aminoacylase and the whole cells containing the recombinant Daminoacylase are used in biotransformation of N-acetyl-D-amino acid. D-Amino acids can be generated in large quantities at low cost using whole-cell biotransformation [23].

10.3.5 Cost Effectiveness and Ease of Operation Compared with isolated enzymes, application of whole cells as biocatalysts is usually more economical since there is no protein purification process involved. Whole cells can be used directly in chemical processes, thereby greatly minimizing formulation costs. Whole cells are cheap to produce and no prior knowledge of genetic details is required. Microorganisms have adapted to the natural environment and produce both simple and complex metabolic products from their nutrient sources through complex, integrated pathways.

220


10.4 Approaches to Address the Disadvantages of Whole-Cell Biotransformation During the last decade, significant advancements in biochemistry, molecular cloning, and random and site-directed mutagenesis, directed evolution of biocatalysts, metabolic engineering and fermentation technology have led us to devise methods to circumvent the disadvantages of whole-cell biotransformation discussed in Section 10.2. The applications of these methods are summarized in this section.

10.4.1 Control of Substrate and Product Concentration by Absorbing Resins Absorbing resins have been extensively applied for enhancing the production and the recovery of natural products from microbial fermentations [24]. Resins act as ‘stabilization agents’ when added to the fermentation broths during the production phase by capturing the products and preventing product degradation and repression of their production. Addition of resin to the fermentation broths significantly reduces the volume of the fermentation broth for downstream processing, since resin can be effectively separated from the microbial cells and the liquid medium [24]. This technique has been applied to the whole-cell biotransformation processes to circumvent the problems associated with substrate availability and recovery of products in low concentrations. Listed below are a few examples of the use of resins in the whole-cell biotransformation processes. 10.4.1.1 Production of (S)-4-(3, 4-Methylene-Dioxyphenyl)-2-Propanol Compound LY 300 164, a 2,3-benzodiazepin, is a drug candidate being evaluated clinically for treatment of epilepsy and Parkinsons disease. It was prepared from (S)-4-(3,4-methylenedioxyphenyl)-2-propanol, which in turn was derived from 3,4-methylenedioxyphenylacetone by a dehydrogenase from the yeast Zygosaccharomyces rouxii via the whole-cell biotransformation process [25]. The volumetric productivity of this biotransformation process is limited by the toxicity of the substrate to the yeast cells at a concentration of 6 g L 1. By binding 80 g of the substrate to 1 L of Amberlite XAD-7 resin, the substrate concentration in the water medium was kept around 2 g L 1, significantly lower than the toxicity level. A sevenfold greater substrate concentration (40 g L 1) could be reached with 96% conversion and >99.9% enantiomeric excess (ee) by simple addition of a polymeric absorbing resin to the reaction mixture. 10.4.1.2 Reduction of Bromocinnamaldehyde 2-(R)-Benzylmorpholine is a potent appetite suppressant drug. The synthesis of 2-(R)-benzylmorpholine began with the reduction of the unsaturated bromocinnamaldehyde to the corresponding saturated (S)-bromo-alcohol by bakers yeast, with a very low ee of 63%. However, an efficient transformation can be achieved by controlling the substrate concentration with the addition of hydrophobic resin Amberlite XAD-1180 [26]. With a resin-to-substrate ratio of one and an initial substrate concentration of 5 g L 1, the saturated (S)-bromo-alcohol was recovered at nearly quantitative yield and 98.6% ee.


221

10.4.1.3 Production of Vanillin Vanillin (4-hydroxy-3-methoxybenzaldehyde) is widely used in foods, beverages, perfumes and the pharmaceuticals industries. Biotransformation of isoeugenol from essential oil to vanillin represents an economic route for the supply of vanillin, which has a limited supply due to the availability of vanilli pod plants. The conversion yield of isoeugenol to vanillin by the whole-cell biotransformation process of Bacillus fusiformis was low due to the product inhibition effect. Adding resin HD-8 to the whole-cell biotransformation eliminated the product inhibition effect, yielding 8 g L 1 of vanillin in the final reaction mixture [27]. The resin HD-8 also facilitated the separation of vanillin from the used substrate. The recovered isoeugenol can be used for the subsequent biotransformation reaction.

10.4.2 Immobilized-Cell Technology In order to extend the biocatalytic activities of the biotransformation processes and reduce the frequency of producing cell mass and undesirable side products, immobilized-cell technology has been successfully applied to the whole-cell biotransformation processes. In addition to the three commercial immobilized whole-cell biotransformation processes shown in Table 10.1, examples of immobilization of three different microorganisms for whole-cell biotransformations are shown below to demonstrate the broad application of the immobilized whole-cell biotransformation processes. 10.4.2.1 Production of Acrylamide The significance of the application of immobilized-cell technology in the production of industrially important chemicals is exemplified by the production of acrylamide by immobilized Escherichia coli cells containing nitrile hydratase. The immobilized Escherichia coli cells convert acrylonitrile to acrylamide, yielding 6000 tons of acrylamide per year by this process [28]. 10.4.2.2 Production of (5R)-Hydroxyhexane-2-One by Immobilized Cells Tan et al. [29] demonstrated the use of a plug flow reactor of immobilized Lactobacillus kefiri cells for the synthesis of the intermediate (5R)-hydroxyhexane-2-one. This immobilized-cell reactor operated at a maximum conversion yield of 100% and a selectivity of 95%. The production of (5R)-hydroxyhexane-2-one was extended to an operation time of 6 days. During this time (91 residence times), a space–time yield of 87 g L 1 day 1 and a productivity of 0:7 g g wet cell weight 1 were obtained. 10.4.2.3 Side-Chain Cleavage of b-Sitosterol b-Sitosterol is an abundant and low-cost raw material for the production of pharmaceutical steroids. 4-Androstene-3,17-dione, the precursor for the synthesis of corticosteroid hormones, can be derived from the side-chain cleavage of b-sitosterol. Immobilized cells of Mycobacterium sp. NRRL B-3805 on Celite matrix (80–120 mesh) was found to be effective in cleaving the side chain of b-sitosterol (5 g L 1) with a molar conversion yield of 70% in 50 h [30].

222


10.4.3 Aqueous–Organic Two-Phase System The issues of autotoxicity and product recovery associated with the whole-cell biotransformation processes can be addressed with the aqueous–organic two-phase system. Similar to the effect of absorbing resins, aqueous–organic two-phase biotransformations offer the advantage of controlling the concentrations of both substrate and product in the biotransformation reaction, allowing higher productivities and product concentrations [14]. The two-phase systems enable substrate transfer from the organic phase to the aqueous biocatalyst phase in a long-term regeneration and equilibrium manner to avoid substrate toxicity and inhibition [14,31]. Similarly, the product preferentially and continually partitions into the organic phase and away from the aqueous biocatalyst phase, thereby preventing product toxicity and inhibition. Furthermore, as the product accumulates in the organic solvent phase, downstream processing has already been partially implemented during the biotransformation process and, therefore, simplified the in situ product isolation and recovery [14,32]. One of the main obstacles for whole-cell microbial transformation in an organic solvent is its biocompatibility, which has led to screening for organic-solvent-tolerant microorganisms. Numerous organic-solvent-tolerant microorganisms have been found and their tolerance mechanisms have been reviewed [14,33,34]. Two-phase biotransformation systems have been successfully implemented for the production of pharmaceutically relevant metabolites. 10.4.3.1 Production of Ethyl-(S )-4-Chloro-3-Hydroxybutanoate The asymmetric reduction of ethyl-4-chloro-3-oxobutanoate to ethyl-(S)-4-chloro-3-hydroxybutanoate is efficiently carried out by Escherichia coli cells (containing carbonyl reductase) growing in a water–n-butyl acetate two-phase system. n-Butyl acetate was selected as the organic solvent to avoid substrate degradation and enzyme inhibition. Glucose and glucose dehydrogenase are used for cofactor regeneration [35]. 10.4.3.2 Production of a-Terpineol A recombinant Escherichia coli strain containing the cloned limonene hydratase gene was able to grow in a water–limonene two-phase system and converted limonene to a-terpineol [36]. Limonene, a cost-effective and readily available monoterpene, served both as the substrate and the neat solvent for the production of a-terpineol. 10.4.3.3 Production of (S )-2-Octonol (S)-2-Octonol is an intermediate for the production of several optically active pharmaceuticals, such as steroids and vitamins. The asymmetric reduction of 2-octanone to (S)-2-octonol by bakers yeast was inhibited severely by substrate and product concentration of 10 mM and 6 mM respectively. Whole-cell biotransformation of 2-octanone in a water–n-dodecane biphasic system yielded a high product concentration of 106 mM with 89% ee in 96 h [37]. 10.4.3.4 Production of (S )-2-Ethylhexan-1-ol (S)-2-Ethylhexan-1-ol is a useful building block for the pharmaceutical, food, cosmetic and chemical industries. The reduction of 2-ethylhex-2-enal to (S)-2-ethylhexan-1-ol by bakers


223

yeast can be effectively carried out in a biphasic system of water–ether and resulted in high yield (91%) and high ee value of >99% [38].

10.4.4 Genetic Engineering Approaches By developing recombinant microorganisms, whole-cell biocatalysts have been created that can compete against isolated enzymes. The overexpression of oxidoreductases and cofactor regenerating enzymes leads to sufficient specific activities of these enzymes, so that side reactions are negligible. Incomplete stereochemical purity of the products can often be corrected by inhibiting or improving the synthesis of a specific enzyme inside the cells. The permeability of the substrates into the cells can be overcome by expressing the required enzymes on the cell surface. Listed below are a few examples to demonstrate the application of genetic engineering techniques in improving the whole-cell biotransformation processes. 10.4.4.1 Increased Stereoselectivity of Reductase in Saccharomyces cerevisiae To address the lack of stereoselectivity of Saccharomyces cerevisiae in conducting the reduction of ketones to alcohols, the parent yeast strain was modified through genetic engineering to selectively increase the amount of the desirable reductase inside the yeast cells [39]. Biotransformation with the genetically modified strains using glucose or galactose as carbon source and various b-keto esters as substrates showed that the ee of the products was increased significantly. Another successful example to overcome the limitations of wild-type Saccharomyces cerevisiae strain for stereoselective reduction b-keto esters is the application of a recombinant Saccharomyces cerevisiae strain that overexpressed a stereospecific carbonyl reductase and a cofactor regeneration enzyme [40]. 10.4.4.2 Increased Efficiency of Aromatic-Ring Hydroxylating Dioxygenases Owing to the cofactor requirements and increased efforts associated with the use of purified enzyme components, biocatalytic applications employing aromatic-ring-hydroxylating dioxygenases (TDO and NDO) have been predominantly developed using whole-cell biotransformations [41,42]. These whole-cell biotransformations are typically facilitated by the inducible overexpression of multicomponent dioxygenases in recombinant host strains, such as Escherichia coli or Pseudomonas sp., which can be grown to high cell densities. These approaches have been employed for the application of oxygenases in the production of multi-kilogram quantities of chiral metabolites for the preparation of natural products, polyfunctionalized metabolites, and pharmaceutical intermediates [15]. Two representative examples from the above applications are the production of ( )-cis-(1S,2R)-indandiol by recombinant Pseudomonas putida strain overexpressing TDO [41] and the production of indigo by recombinant Escherichia coli strain overexpressing NDO [42]. ( )-cis-(1S,2R)-Indandiol is a key intermediate for the production of HIV-1 protease inhibitor Crixivan. Indigo is an important commodity chemical. 10.4.4.3 Developing a New Whole-Cell Biotransformation Process for the Synthesis of Simvastatin Simvastatin is an important cholesterol-lowering drug and is currently synthesized from the natural product lovastatin via a tedious multistep chemical synthesis. A one-step, whole-cell

224


biotransformation process for the synthesis of gram quantity simvastatin with >90% conversion yield and >98% purity using a recombinant Escherichia coli strain overexpressing acyltransferase LovD has been reported recently [43]. Further genetic modification of the Escherichia coli strain was performed by inactivating an undesirable enzyme that reduced both the substrate conversion and recovery yields by hydrolyzing the substrate [44]. The improved Escherichia coli strain has a lower substrate requirement and a faster rate of simvastatin synthesis compared with the parent recombinant strain. In the improved strain, 99% conversion of 6 g L 1 of simvastatin with higher purity (devoid of hydrolyzed product) was achieved in less than 12 h. While the economic feasibility of this process still requires further in-depth evaluation, this example nevertheless demonstrates the significance of genetic approaches in enhancing the efficiency of whole-cell biotransformation processes. 10.4.4.4 High Cell-Density Cultivation of Escherichia coli with Surface-Anchored Transglucosidase One of the disadvantages of whole-cell biotransformation is the mass transfer limitations imposed by the cell membrane, which leads to lower productivities. Development of whole-cell biocatalysts displaying the target enzyme directly on the cell surface is one of the methods of overcoming the substrate permeation barrier and to improve product yields. An application of this technique was demonstrated by the production of arbutin by recombinant Escherichia coli cells anchoring surface-displayed transglucosidase in a fed-batch process [45]. Arbutin is a hydroquinone glucoside that inhibits tyrosinase activity in melanin biosynthesis and has market potential in the cosmetics industry. The whole-cell biocatalysts showed a specific activity of 501 nkat/g cell and produced 21 g L 1 of arbutin. Other examples include using Pseudomonas syringae ice-nucleation protein as a carrier protein, and levansucrase [46], carboxymethylcellulase [47] and organophosphorus hydrolase [48] have been successfully displayed on Escherichia coli surface for use as whole-cell biocatalysts in the related applications. 10.4.4.5 Designed Biocatalysts Gr€ oger et al. [49] demonstrated the use of ‘designed biocatalysts’ for efficient asymmetric reduction of ketones (>90% yield and >99% ee) at high substrate concentrations and with no external addition of cofactor. The designed biocatalyst was prepared by cloning alcohol dehydrogenase (Lactobacillus kefir or Rhodococcus erythropolis) with high specific activity (1000 U mg 1) and glucose dehydrogenase (Thermoplasma acidophilum or Bacillus subtilis) with high cofactor regenerating activity into Escherichia coli DSM 1445 host strain. This designed biocatalyst demonstrated 95% conversion of 4-chloro-acetophenone at a concentration of 78 g L 1. This is an excellent example of a tailor-made microorganism for carrying out a specific biotransformation with extremely high efficiency.

10.5 Conclusions It is clear from the discussion in Section 10.4 that the advancements in biochemistry, molecular cloning, random and site-directed mutagenesis, directed evolution of biocatalysts, metabolic engineering and fermentation technology can circumvent the major technical problems


225

associated with the whole-cell biotransformation processes. The genetic engineering approach can be used in combination with other approaches to address the problems encountered with the whole-cell processes. The number of tailor-made microorganisms created to address specific problems of the biotransformation processes will be increased rapidly in the next few years. There are many microorganisms, such as Clostridium spp. and cyanobacteria, which show tremendous potential in biotransformation applications [50,51]. The special growth requirements of these microorganisms slow their development as useful microbial biocatalysts. However, any experienced fermentation scientists should have no problem in developing these microbial biocatalysts for whole-cell biotransformation applications. Furthermore, new microbial biocatalysts will continuously be isolated from different environments due to the tremendous microbial diversity in nature [52]. The discovery and development of new microbial biocatalysts would lead to new whole-cell biotransformation applications. Therefore, whole-cell biotransformations will likely play an increasingly important role in the synthesis of pharmaceuticals and pharmaceutical intermediates. One advantage of whole-cell biotransformation that has not been addressed adequately in this chapter is the ability to modify compounds with complex structure, such as natural products. Natural products are ideal substrates for biotransformation reactions since they are synthesized in a series of enzymatic reactions by the whole cells. The modification of natural products by biotransformation has been reviewed recently by Azerad [13] and a majority of the modifications were carried out by whole-cell biotransformations. Additional examples of modification of natural products by whole-cell biotransformations can also be found in the review article by Patel [2]. Natural products are an important source of new drugs and new drug leads [53]. The use of biotransformation, especially whole-cell biotransformation, in modification of natural products for lead optimization and generating libraries of derivatives for SAR and screening studies is important for the pharmaceutical industry.

References [1] Mahmoudian, M. (2007) A decade of biocatalysis at glaxo wellcome, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press, Boca Raton, FL, pp. 53–102. [2] Patel, R.N. (2007) Biocatalysis for synthesis for chiral pharmaceutical intermediates, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press, Boca Raton, FL, pp. 103–158. [3] Gotor-Fernandez, V., Rebolledo, F. and Gotor, V. (2007) Preparation of chiral pharmaceuticals through enzymatic acylation of alcohols and amines, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press, Boca Raton, FL, pp. 203–248. [4] Kieslich, K. (1991) Biotransformation of industrial use. Acta Biotechnologica, 11, 559–570. [5] Serizawa, N., Hosobuchi, M. and Yoshikawa, H. (1997) Biochemical and fermentation technological approaches to production of pravastastin, a HMG-CoA reductase inhibitor, in Biotechnology of Antibiotics, 2nd edn (ed. W.R. Strohl), Marcel Dekker, New York, pp. 779–805. [6] Zaks, A. and Dodds, D.R. (1997) Application of biocatalysis and biotransformations to the synthesis of pharmaceuticals. Drug Discovery Today, 2, 513–531. [7] Shaw, N.M., Robins, K.T. and Kiener, A. (2003) Lonza: 20 years of biotransformations. Advanced Synthesis and Catalysis, 345, 425–435. [8] Cheetham, P.S.J. (1994) Case studies in applied biocatalysis, in Applied Biocatalysis (eds J.M.S. Cabral, D. Best, L. Boross and J. Tramper), Harwood Academic Publishers, Chur, Switzerland, pp. 68–70. [9] Cheetham, P.S.J. (1994) Case studies in applied biocatalysis – from ideas to products, in Applied Biocatalysis (eds J.M.S. Cabral, D. Best, L. Boross and J. Tramper), Harwood Academic Publishers, Chur, Switzerland, pp. 87–89.

226


[10] Chibata, L., Tosa, T. and Shibatani, T. (1992) The industrial production of optically active compounds by immobilized biocatalysts, in Chirality in Industry (eds A.N. Collins, G.N. Sheldrake and J. Crosby), John Wiley & Sons, Ltd, New York, pp. 351–370. [11] Kumagai, H. (2000) Microbial production of amino acids in Japan. Advances in Biochemical Engineering/ Biotechnology, 69, 71–85. [12] Goldberg, K., Schroer, K., L€utz, S. and Liese, A. (2007) Biocatalytic ketone reduction – a powerful tool for the production of chiral alcohols – part II: whole-cell reductions. Applied Microbiology and Biotechnology, 76, 249–255. [13] Azerad, R. (2007) Biotransformation of natural or synthetic compounds for the generation of molecular diversity, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press Boca Raton, FL, pp. 273–297. [14] Heipieper, H.J., Neumann, G., Cornelissen, S. and Meinhardt, F. (2007) Solvent-tolerant bacteria for biotransformations in two-phase fermentation systems. Applied Microbiology and Biotechnology, 74, 961–973. [15] Parales, R.E. and Resnick, S.M. (2007) Applications of aromatic hydrocarbon dioxygenases, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press, Boca Raton, FL, pp. 299–332. [16] Schewe, H., Haup, B.-A. and Schrader, J. (2008) Improvement of P450BM-3 whole-cell biocatalysis by integrating heterologous cofactor regeneration combining glucose facilitator and dehydrogenase in E. coli. Applied Microbiology and Biotechnology, 78, 55–65. [17] Urlacher, V.B. and Eiben, S. (2006) Cytochrome P450 monooxygenases: perspectives for synthetic application. Trends in Biotechnology, 24, 324–330. [18] Gibson, D.T. and Parales, R.R. (2000) Aromatic hydrocarbon dioxygenases in environmental biotechnology. Current Opinion in Biotechnology, 11, 236–243. [19] Butler, C.S. and Mason, J.R. (1997) Structure–function analysis of the bacterial aromatic ring-hydroxylating dioxygenases. Advances in Microbial Physiology, 38, 47–84. [20] Buhler, B. and Schmid, A. (2004) Process implementation aspects for biocatalytic hydrocarbon oxyfunctionalization. Journal of Biotechnology, 113, 183–210. [21] Berry, A., Dodge, T.C., Pepsin, M. and Weyler, W. (2002) Application of metabolic engineering to improve both the production and use of biotech indigo. Journal of Industrial Microbiology & Biotechnology, 28, 127–133. [22] Stuermer, R., Hauer, B., Hall, M. and Faber, K. (2007) Asymmetric bioreduction of activated C C bonds using enoate reductases from the old yellow enzyme family. Current Opinion in Chemical Biology, 11, 203–213. [23] Taylor, S.J.C., Holt, K.E., Brown, R.C. et al. (2000) Choice of biocatalyst in the development of industrial biotransformation, in Stereoselective Biocatalysis (ed. R.N. Patel), Marcel Dekker, New York, pp. 397–413. [24] Frykman, S., Tsuruta, H., Galazzo, J. and Licari, P. (2006) Characterization of product capture resin during microbial cultivations. Journal of Industrial Microbiology & Biotechnology, 33, 445–453. [25] Vicenzi, J.T., Zmijewski, M.J., Reinhard, M.R. et al. (1997) Large-scale stereoselective enzymatic ketone reduction with in-situ product removal via polymeric adsorbent resins. Enzyme and Microbial Technology, 20, 494–499. [26] DArrigo, P., Lattanzio, M., Fantoni, P. and Servi, S. (1998) Chemo-enzymatic synthesis of the active enantiomer of the anorressant 2-benzylmorpholine. Tetrahedron Asymmetry, 9, 4021–4026. [27] Zhao, L.-Q., Sun, Z.-H., Zheng, P. and He, J.-Y. (2006) Biotransformation of isoeugenol to vanillin by Bacillus fusiformis CGMCC1347 with the addition of resin HD-8. Process Biochemistry (Barking, London, England), 41, 1673–1676. [28] Yamada, H. and Kobayashi, M. (1996) Nitrile hydratase and its application to industrial production of acrylamide. Bioscience, Biotechnology, and Biochemistry, 60, 1391–1400. [29] Tan, A.W., Fischbach, M., Huebner, H. et al. (2006) Synthesis of enantiopure (5R)-hydroxyhexane-2-one with immobilised whole cells of Lactobacillus kefiri. Applied Microbiology and Biotechnology, 71, 289–293. [30] Dias, A.C.P., Cabral, J.M.S. and Pinheiro, H.M. (1994) Sterol side-chain cleavage with immobilized Mycobacterium cells in water-immiscible organic solvents. Enzyme and Microbial Technology, 16, 708–714. [31] Cruz, A., Fernandes, P., Cabral, J.M.S. and Pinheiro, H.M. (2004) Solvent partitioning and whole-cell sitosterol bioconversion activity in aqueous–organic two-phase systems. Enzyme and Microbial Technology, 34, 342–353. [32] Malinowski, J.J. (2001) Two-phase partitioning bioreactors in fermentation technology. Biotechnology Advances, 19, 525–538. [33] Sardessai, Y.N. and Bhosle, S. (2004) Industrial potential of organic solvent tolerant bacteria. Biotechnology Progress, 20, 655–660.


227

[34] Yamashita, S., Satoi, M., Iwasa, Y. et al. (2007) Utilization of hydrophobic bacterium Rhodococcus opacus B-4 as whole-cell catalyst in anhydrous organic solvents. Applied Microbiology and Biotechnology, 74, 761–767. [35] Kizaki, N., Yasohara, Y., Hasegawa, J. et al. (2001) Synthesis of optically pure ethyl (S)-4-chloro-3-hydroxybutanoate by Escherichia coli cells coexpressing the carbonyl reductase and glucose dehydrogenase genes. Applied Microbiology and Biotechnology, 55, 590–595. [36] Savithiry, N., Cheong, T.K. and Oriel, P. (1997) Production of a-terpineol from Escherichia coli cells expressing thermostable limonene hydratase. Applied Biochemistry and Biotechnology, 63–65, 213–220. [37] Li, Y.-N., Shi, X.-A., Zong, M.-H. et al. (2007) Asymmetric reduction of 2-octanone in water/organic solvent biphasic system with bakers yeast FD-12. Enzyme and Microbial Technology, 40, 1305–1311. [38] Huang, Y., Zhang, F. and Gong, Y. (2005) A convenient approach to (S)-2-ethylhexan-1-ol mediated by bakers yeast. Tetrahedron Letters, 46, 7217–7219. [39] Rodriguez, S., Kayser, M.M. and Stewart, J.D. (2001) Highly stereospecific reagents for b-keto ester reductions by genetic engineering of bakers yeast. Journal of the American Chemical Society, 123, 1547–1555. [40] Engelking, H., Pfaller, R., Wich, G. and Weuster-Botz, D. (2006) Reaction engineering studies on b-ketoester reductions with whole cells of recombinant Saccharomyces cerevisiae. Enzyme and Microbial Technology, 38, 536–544. [41] Boyd, D.R., Sharma, N.D., Bowers, N.I. et al. (1996) Stereoselective dioxygenase-catalyzed benzylic hydroxylation at prochiral methylene groups in the chemoenzymatic synthesis of enantiopure vicinal aminoindanols. Tetrahedron Asymmetry, 7, 1559–1562. [42] Connors, N., Prevoznak, R., Chartrain, M. et al. (1997) Conversion of indene to cis-(1S,2R)-indandiol by mutants of Pseudomonas putida F1. Journal of Industrial Microbiology & Biotechnology, 18, 353–359. [43] Xie, X. and Tang, Y. (2007) Efficient synthesis of Simvastatin by use of whole-cell biocatalysis. Applied and Environmental Microbiology, 73, 2054–2060. [44] Xie, X., Wong, W.W. and Tang, Y. (2007) Improving simvastatin bioconversion in Escherichia coli by deletion of bioH. Metabolic Engineering, 9, 379–386. [45] Wu, P.-H., Nair, G.R., Chu, I.-M. and Wu, W.-T. (2008) High cell density cultivation of Escherichia coli with surface anchored transglucosidase for use as whole-cell biocatalyst for a-arbutin synthesis. Journal of Industrial Microbiology & Biotechnology, 35, 95–101. [46] Jung, H.C., Lebeault, J.M. and Pan, J.G. (1998) Surface display of Zymomonas mobilis levansucrase by using the ice-nucleation protein of Pseudomonas syringae. Nature Biotechnology, 16, 576–580. [47] Jung, H.C., Park, J.H., Park, S.H. et al. (1998) Expression of carboxymethylcellulase on the surface of Escherichia coli using Pseudomonas syringae ice nucleation protein. Enzyme and Microbial Technology, 22, 348–354. [48] Shimazu, M., Mulchandani, A. and Chen, W. (2001) Cell surface display of organophosphorus hydrolase using ice nucleation protein. Biotechnology Progress, 17, 76–80. [49] Gr€ oger, H., Chamouleau, F., Orologas, N. et al. (2006) Enantioselective reduction of ketones with ‘designer cells’ at high substrate concentrations: high efficient access to functionalized optically active alcohols. Angewandte Chemie (International Edition in English), 45, 5677–5681. [50] Havel, J. and Weuster-Botz, D. (2006) Comparative study of cyanobacteria as biocatalysts for the asymmetric synthesis of chiral building blocks. Engineering in Life Sciences, 6, 175–179. [51] Bhushan, B., Halasz, A. and Hawari, J. (2005) Biotransformation of CL-20 by a dehydrogenase enzyme from Clostridium sp. EDB2. Applied Microbiology and Biotechnology, 69, 448–455. [52] Antranikian, G., Vorgias, C.E. and Bertoldo, C. (2005) Extreme environments as a resource for microorganisms and novel biocatalysts. Advances in Biochemical Engineering/Biotechnology, 96, 219–262. [53] Lam, K.S. (2007) New aspects of natural products in drug discovery. Trends in Microbiology, 15, 279–289.

11 Combinatorial Biosynthesis of Pharmaceutical Natural Products Wen Liu and Yi Yu State Key Laboratory of Bio-Organic and Natural Product Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 354 Fenglin Rd, Shanghai, 200032, PR China

11.1 Introduction Natural products have played a significant role in the drug discovery and development process. In 1997, 2003 and 2007, upon successive analyses of the sources of new and approved drugs (for the treatment of all human diseases from 1981 and cancers from 1950) to date, the fact has been repeatedly impressed that near half of clinically used drugs are originated from natural products and their derivatives [1–3]. Compared with random millions of compounds assembled by combinatorial chemistry techniques, natural products that have been validated over long periods of evolution in nature are often structurally more complex and biologically more versatile. Many pharmaceutical moieties that reside in the architectures of natural products are structurally unusual and serve as templates for the chemical synthesis, resulting in their clinically successful mimics with biologically more specific activities to targets. Traditional natural product drug discovery approaches depend primarily on bioassayguided screening for lead identification. During the 1980s and 1990s, the pharmaceutical research into natural products had been de-emphasized by industry, mainly due to the laborious product-development cycles, small quantities of materials available from nature and lack of compatibility of natural product libraries with high-throughput screening (based on current screening models) [4]. On the other hand, given the structural complexity and the rich variety of functional groups that are characteristic of many natural products, chemical synthesis has very limited practical value to reach them, and analogs generation by chemical modification of natural products remains problematic and often requires multiple protection and deprotection steps. Recently, there has been a renewed interest in natural product


230


research due to the inefficiency of alternative drug discovery methods to deliver leading compounds into clinical trails [5]. To continue to be competitive, natural product research needs to address the suitability of the screening methods, speed up the development cycles, deal with large-scale compound supply and develop efficient synthetic routes. Combinatorial biosynthesis, which was defined as the application of various biotechnological manipulations to engineer biosynthetic pathways of natural products aiming to expand their structural diversities in natural biosynthetic machineries, offers a promising alternative way to solve the above problems. Specific structural alteration in the presence of abundant functional groups can often to be achieved, and biosynthetic pathways can be evaluated and reprogrammed at the genetic level to create many ‘hybrid’ natural leads to meet the challenge for drug discovery and development. The target molecules will be produced by a recombinant organism that is amenable for large-scale fermentation, thus lowering the production cost and reducing the environmental concern associated with conventional chemical synthesis. A few excellent reviews and book chapters have been published elsewhere describing combinatorial biosynthesis and its progress for drug discovery [6–33]. In this chapter, we first demonstrate the principle and strategy of combinatorial biosynthesis through learning from the natural way to create structural diversity, and then choose a set of pharmaceutical natural products to showcase the practices and potential utilities of combinatorial biosynthesis by application of various biotechnologies into the production and development of novel natural product-based drugs, including gene disruption/replacement, hybrid pathway construction, mutasynthesis, chemoenzymatic synthesis and heterologous expression. Owing to space limitation, we narrow the scope of this chapter mainly to bacterial natural products.

11.2 Combinatorial Biosynthesis: The Natural Way for Structural Diversity Over long periods, natural product discovery proves that many microorganisms share the ability to produce identical natural products. This is exemplified by erythromycin A, which is equally produced by Saccharopolyspora erythraea and Aeromicrobium erythreum. Sequence comparison revealed that the 55.4 kb ery gene cluster from Aeromicrobium erythreum contains 25 genes, 21 of which share high homology in both gene organization and amino acid residues to those in the ery gene cluster from Saccharopolyspora erythraea, suggesting that both biosynthetic pathways share a common evolutionary origin [34]. Bioinformatical analysis of the growing genome sequences provides further evidences at the genetic level. For example, the genome of Streptomyces ceolicolor harbors more than 20 clusters that encode the biosynthesis of secondary metabolites [35], many of which have been identified in other microorganisms. These include aromatic polyketides actinorhodin produced by Streptomyces lividans and tetrahydroxynaphthalene produced by Streptomyces griseus, siderophore desferrioxamines produced by Streptomyces pilous and Erwinia amylovora, polyunsaturated fatty acid eicosapentaenoic acid produced by Shewanella sp., and the terpenoids carotenoid and geosmin produced by Streptomyces griseus (Figure 11.1). The horizontal gene transfer (HGT) driven by genetic factors might play a major role in distribution of such abilities among microorganisms [36].

231

Combinatorial Biosynthesis of Pharmaceutical Natural Products (a) O

OH

HO2C

O

OH

O

O O

OH OH

Actinorhodin OH

CO2H O (b)

OH

O

HO N

O HO

OH

OH

NH2

Tetrahydroxynaphthalene O

NH O

O

NH

N H

OH

N OH O

Desferrioxamines (c)

CO2H Eicosapentaenoic acid

(d)

OH

Beta-carotene

Geosmin

Figure 11.1 Representative secondary metabolites produced by Streptomyces ceolicolor and other microorganisms, including aromatic polyketides actinorhodin and tetrohydroxynaphthalene (a), siderophore desferrioxamines (b), polyunsaturated fatty acid eicosapentaenoic acid (c) and terpenoids betacarotene and geosmin (d)

More often than that, in nature, various microorganisms with distinct origins or genera produce a set of natural products that are structurally related. According to the structural similarity, natural products are often grouped into many families, such as 12-, 14- and 16-membered macrolides, polyenes, polyethers, antracyclines, ansamycins, spirotetronates, thiopeptides and enediynes. Systematic characterization of the biosynthetic mechanisms of individual members in each family would facilitate understanding

232


natures wisdom to create the diversity of structures and biological activities of complex natural products. For example, the nine-membered enediynes constitute an extremely potent anticancer family that feature a characteristic unit that contains two acetylenic groups conjugated with a double bond within a nine-membered ring [23]. Members of this family include neocarzinostatin (NCS) from Streptomyces carzinostaticus, C-1027 from Streptomyces globisporus, kedarcidin from Actinomycete L585-6, maduropeptin (MDP) from Actinomadura madurea, and N1999A2 from Streptomyces sp. AJ9493. Recently, the biosynthetic gene clusters of C-1027, NCS, MDP, and the 10-membered enediyne calicheamicin (CAL, from Micromonospora echinospora) have been cloned and sequenced [37–41], revealing that the first biosynthetic step toward the enediyne core was speculated to be catalyzed by a novel type iterative type I polyketide synthase (PKS), namely enediyne PKS (PKSE), which is distinct from all known PKSs but shared among the entire enediyne family [42,43]. The PKSEs SgcE and NcsE (involved in C-1027 and NCS, respectively) are functional in heterologous hosts, and coexpression with an enediyne thioesterase (TE) produces 1,3,5,7,9,11,13-pentadecaheptaene, the first isolable compound leading to the enediyne core [44]. The PKSEs are cross-complemented and interchangeable within the nine-membered enediyne family, suggesting that the biosynthesis of the enediyne core occurs through a common polyene intermediate [41,44]. More importantly, it could be envisioned that the enediyne core can be subsequently decorated via different post-PKS tailoring enzymes and convergently assembled with different building blocks in each case, rendering structural divergence and affording the individual members in this family (Figure 11.2). Therefore, natures ways for structural diversity of natural products include: (a) tailoring of a characteristic unit that can be further combined with various structural moieties; (b) increase of structural difference caused by the broad substrate specificities of biosynthetic enzymes (featured by a group of structurally similar natural products often produced by one certain organism and by the third example (patellamide) in Section 11.3.3). To practice these strategies, natural product evolution has been driven by mutation, recombination and selection for fitness in nature over billion of years or so [6]. Under various environmental selective pressures, mutation and recombination of an original biosynthetic gene cluster obtained by HGT might set the genetic stage for diversity of structures and biological activities of natural products in different organisms. To apply this principle into the field of drug discovery and development, combinatorial biosynthesis discussed here encompasses various traditional methodologies of metabolic engineering and all aspects of molecular biological manipulations aiming to speed up the process of natural product evolution [6]. Rational engineering of the biosynthetic pathways will lead to novel analogs with desired structural modification, and recombination of various biosynthetic genes at the genetic level will construct a library of ‘hybrid’ pathways in recombinant strains, which can produce many ‘unnatural’ natural products to further expand structural diversity (Figure 11.3). Compared with the natural ways for natural product evolution, combinatorial biosynthesis accelerates the process of mutation and recombination between biosynthetic genes that normally reside in different genera, the DNAs of which may not be inter-exchangeable under natural conditions. The success of this approach depends mainly on (a) the genetic and biochemical characterization of the biosynthetic pathways of natural products and (b) the development of strategies and methods for combinatorial manipulation of biosynthetic gene clusters.

Figure 11.2 Biosynthesis of the nine-membered enediynes. Members of this family share a common biosynthetic pathway for the enediyne core intermediate. Domains are shown in circles with abbreviations (KS, ketosynthase; AT, acyltransferase; KR, ketoreductase; DH, dehydratase; TE, thioesterase; ACP, acyl carrier protein; PPT, phosphopantetheine transferase)

Combinatorial Biosynthesis of Pharmaceutical Natural Products 233

234


Figure 11.3 Principle of combinatorial biosynthesis. Genetic engineering and recombination of the biosynthetic pathways are two major strategies in combinatorial biosynthesis to expand structural diversity

11.3 Examples of Combinatorial Biosynthesis of Pharmaceutical Natural Products 11.3.1 Erythromycin (Polyketide Biosynthesis) Erythromycins are a group of important antibiotics with broad-spectrum activity against pathogenic Gram-positive bacteria [24]. Erythromycin A, the most potent and clinically used member in this family, contains a characteristic 14-membered macrolide that is decorated by two unusual deoxysugars: mycarose and desosamine. Its importance in infectious disease therapy is further highlighted by the fact that several semi-synthetic derivatives known as second- and third-generation products (e.g. azithromycin, flurithromycin and telithromycin) are commercially successful and widely used in clinics [45], continuously reviving efforts to develop new analogs with better potency and lower side effects. The biosynthesis of erythromycins could be briefly divided into two phases (Figure 11.4a) [24]. In the first constructive phase, a set of multifunctional type I PKSs, namely, 6-deoxyerythronolide B synthases (6-DEBSs) 1–3, catalyze the assembly of the polyketide backbone from one propionyl-CoA and six methylmalonyl-CoAs by sequentially decarboxylative condensations. The resulting polyketide chain then undergoes an intramolecular cyclization to give the first macrocyclic lactone intermediate, 6-deoxyerythronolide B (6-DEB). In the second phase, a series of tailoring proteins sequentially carry out elaborate modifications, including regiospecific hydroxylations, glycosylations and a methylation, on 6-DEB to afford erythromycin A finally. Since a few members in the erythromycin family, including erythromycin B, C and D, occur as the intermediates in this biosynthetic pathway, we recently developed an approach to improve the erythromycin A purity and production at the fermentation stage by systematic genetic modulation of the overexpression of tailoring genes eryK (encoding a P450 hydroxylase) and eryG (encoding an S-adenosyl methionine-dependent O-methyltransferase), aiming to simplify the downstream purification process and lower the production cost and environmental concern in industry [46].

235

Combinatorial Biosynthesis of Pharmaceutical Natural Products (A)

DEBS2

DEBS1

Load Module 1 Module 2 Module 3 KR KR ACP AT ACP KS AT ACP KS AT

S

O

S

S

Module 5

KR KR KR

KR

KS AT ACP KS AT

S

DEBS3

Module 4

O

O

O

OH

OH

O

OH

OH

ACP

KS AT

O

OH

ACP TE

S O

O

OH

O

Ethyl C-7

C-13 substituted groups 7

OH O

12 13

O

C7 C6

6

OH

2

Cyclization

OH O

OH

O

O

OH

6-Deoxyerythronolide B

OH

Hydroxylation

OH

Glycosylations

O OH

O

OH

OH OH

Desmethyl C-12

H3C

ACP KS AT

OH

OH

(B)

KR

S

S

O

Module 6 End

O HO

Methylation

OH N(CH3)2 OH HO O O O O O OCH3 O

OH

Erythromycin A

OH

C-2 substituted groups Desmethyl C-2 OH OCH3

Figure 11.4 6-DEBS 1, 2 and 3 that are modularly organized PKSs direct the biosynthesis of erythromycin A (a), and genetic engineering on genes encoding DEBSs gave rise to various 6-DEB analogs (b)

Typically, type I PKSs are multifunctional enzymes that are organized into modules, each of which minimally harbors a set of distinct active domains responsible for the catalysis of one cycle of polyketide chain elongation, including ketosynthase (KS), acyltransferase (AT) and acyl carrier protein (ACP) [24]. For the past two decades, the biosynthesis of 6-DEB has provided a paradigm for understanding the structure and function of modular PKSs that are responsible for assembling complex polyketides in a noniterative manner (the ‘one enzyme, one function’ organization). The feature known as the ‘collinearity rule’ has inspired intensive attempts at reprogramming individual enzymatic activities on 6-DEBSs to generate 6-DEB or erythromycin analogs with targeted structural modification by utilizing most of combinatorial biosynthesis strategies (Figure 11.4b), if not all [47]. To change the side chain at the C-13 position, the loading AT domain of 6-DEBS has been replaced by those from the avermectin, tylosin, oleandomycin and rapamycin PKSs for producing C-13-substituted derivatives [48– 50]; to change the methyl branches, the methylmalonyl-specific AT domains have been replaced by malonyl-specific AT domains using domain exchange [51–54], site-specific mutagenesis [55,56] or inactivation/complementation in trans [57] to create analogs lacking the corresponding methyl group, and by an ethylmalonyl or 2-methoxymalonyl-specific AT domain to generate the ethyl [58] or methoxy branch [59]. Alternations of b-keto functionalities have been achieved, including that deletion of the ketoreductase (KR) domains generated keto derivatives of 6-DEB [60,61], that inactivation of the enoylreductase (ER) domain resulted in a D-anhydro erythromycin derivative [62], and that substitution of the KR domain with a module containing dehydratase (DH) þ ER þ KR tri-domains led to a 6-DEB analog that is fully saturated at the corresponding position [61]. Control of the ring size has also been

236


accomplished by rearranging the organization of the native thioesterase (TE) domain on 6DEBSs [63–65]. Remarkably, using approaches carrying multiple genetic modifications, large combinatorial libraries of 6-DEB analogs (more than 50 members) have been constructed, including derivatives with one, two, and three altered carbon centers [61,66]. To incorporate the advantages of modern synthetic chemistry into complex polyketide functionality, precursor-directed biosynthesis, namely, mutasynthesis, has been carried out by blocking a particular step in the erythromycin pathway and supplying synthetic precursors in the form of N-acylcysteamine thioesters (SNAC, mimic important features of native substrates), resulting in several novel analogs including 14- and 16 membered lactones [67–69], and those with side chains containing halogens and reactive groups for further modification by synthetic chemistry techniques [70]. While many 6-DEB or erythromycin analogs have been synthesized in the native producer Streptomyces erythraea, or other Streptomyces species such as Streptomyces lividans and Streptomyces fradiae, for the choice of rapid engineering, extensive combinatorial biosyntheses on the erythromycin biosynthetic pathway have recently been performed more in the heterologous host Escherichia coli [71–73]. To provide sufficient quantities of precursors (i.e. the start unit propionyl-CoA and extender unit (2S)-methylmalonyl-CoA) and active 6-DEBSs (with phosphopantetheine-modified ACP domains) for polyketide assembly, the Escherichia coli host was engineered by introduction of the phosphopantetheinyl transferase gene sfp from Bacillus subtilis and genes encoding a heterodimeric propionyl-CoA carboxylase pccAB from Streptomyces coelicolor, deletion of the endogenous prpRBCD genes for propionate degradation, and overexpression of the endogenous prpE and birA genes involving in propionate production [72]. Significantly, conversion of the inactive 6-DEB to the biologically active glycosylated derivative erythromycin C was achieved by introduction of two glycosylation operons into Escherichia coli [74], demonstrating that the Escherichia coli biosynthetic machinery could be further evaluated to be a useful host for the production of novel and therapeutically relevant polyketides.

11.3.2 Daptomycin (Nonribosomal Peptide Biosynthesis) Daptomycin, produced in quantity by feeding decanoic acid into the fermentation of Streptomyces roseosporus, is a 10-membered cyclic lipopeptide antibiotic that was approved in the USA under the trade name Cubicin in 2003, and represents the first clinically used member in the acidic lipopeptide family for treatment of severe skin and skin structure infections caused by Gram-positive pathogens, including methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Staphylococcus aureus (VRSA) [75]. Structurally, daptomycin contains a peptide backbone conjugated with a straight C10-lipid side chain, consisting of 13 amino acid residues, three of which are D-configuration (D-asparagine (D-Asn2), D-alanine (D-Ala8) and D-serine (D-Ser12)) and six of which (L-ornithine (L-Orn6), L-3-methylglutamate (L-3mGlu12) and L-kynurenine (L-Kyn13), as well as D-amino acids) are nonproteinogenic amino acids. The biosynthesis of daptomycin is accomplished by a set of nonribosomal peptide synthetases (NRPSs), namely DptA, BC and D (Figure 11.5a) [76]. In general, NRPSs are modularly organized megaenzymes, and each module consists minimally of an adenylation (A) domain responsible for amino acid activation, a peptidyl carrier protein (PCP) domain for thioesterification of the activated amino acid, and a condensation (C) domain for transpeptidation between the upstream and downstream peptidyl and amino acyl thioesters

Figure 11.5 Amino acid building blocks are incorporated into daptomycin backbone successively by NRPS subunits DptA, DptBC and DptD (a). Structural diversity of daptomycin peptide core can be obtained by genetic modifications of dpt gene cluster (b). C, condensation domain; A, adenylation domain; PCP, peptidyl carrier protein; E, epimerase; TE, thioesterase domain

Combinatorial Biosynthesis of Pharmaceutical Natural Products 237

238


to elongate the growing peptide chain [77]. Mechanistically analogous to the typical type I PKSs following a collinearity rule, DptA, BC and D catalyze a fatty acid-initiated assembly of the linear tridecapeptide chain and macrolactonization between the threonine (Thr4) and Kyn13 residues to afford the 10-membered ring. The nonproteinogenic amino acids incorporated into the daptomycin biosynthesis are either as precursors or formed by additional catalytic action of NRPSs, such as epimerization (E) for D-amino acids. Since the challenging nature of its structure has limited the attempts at structural diversity by chemical synthesis, combinatorial biosynthesis provides an important alternative to produce and scale up derivatives of daptomycin with improved pharmaceutical properties [6]. The available biosynthetic gene clusters of daptomycin, and other structurally related 10-membered cyclic lipopeptides (e.g. calcium-dependent antibiotic (CDA) produced by Streptomyces ceolicolor [78], A54145 produced by Streptomyces fradiae [79], and friulimicin produced by Actinoplanes friuliensis [80]), make it practical and enrich the ‘toolbox’ of combinatorial biosynthesis to genetically engineer the NRPS genes for functionality substitution on the polypeptide skeleton. Initially, by using the corresponding dpt genes, the daptomycin productions of a series of Streptomyces roseosporus mutant strains with inactivation of the dptA, dptBC or dptD have been restored at a level comparable to that of the wild-type strain, demonstrating that a trans-complementation strategy could be tested by heterologous subunits [81,82]. Consequently, without engineering the interpeptide docking sites, substitution of dptD encoding the terminal dimodular subunit for incorporation of the last two amino acids 3mGlu12 and Kyn13 with the corresponding genes cadPS3 for CDA biosynthesis or lptD for A54145 biosynthesis resulted in production of 18 daptomycin derivatives with replacement of Kyn13 by tryptophan (Trp13), isoleucine (Ile13) or valine (Val13). Encouraged by this finding, the novel daptomycin library was further expanded by combining the NRPS subunit exchange with modifications of the cyclic peptide core at residues nonconserved among daptomycin, CDA and A54145 by module exchange in DptBC, inactivation of glutamic acid 3-methyltransferase, and natural lipid sidechain variation, generating more than 70 new lipopeptides in substantial quantities (Figure 11.5b) [83,84]. Remarkably, a few members are active against Gram-positive bacteria equivalently to daptomycin, and one compound is more potent against an Escherichia coli imp mutant that has increased outer membrane permeability. The final step in nonribosomal peptide synthesis is cleavage of a linear peptide product from the covalently attached peptidyl carrier protein (ACP) domain by TE action. Recently, TE domains excised from different NRPSs have been capable of serving as versatile enzymatic tools to catalyze the cyclization of chemically synthesized peptide thioester analogs of their natural substrates, rendering the constrained and biologically active conformations that are required for specific interaction with the cellular targets [85]. Using this strategy, a few daptomycin and CDA derivatives have been prepared by macrocyclization of peptide thioester substrates with CDATE [86]. To overcome the drawback that the yields of cyclic peptides were limited by competing hydrolysis activity of single TE, the isolated di-domain enzymes PCP TEs of daptomycin and A54145 (these TE domains together with their associated PCPs) have been employed to generate their derivatives as well as hybrid molecules of both compounds [87]. As a robust macrocyclization catalyst, the A54145 PCP TE enzyme is the first cyclase of a branched cyclic lipopeptide, catalyzing both macrolactonization and macrolactamization. By combining the advantages of organic synthesis, this chemoenzymatic method further expanded the structural diversity for the study of lipopeptide structure–activity relationship and, more importantly, holds great potential for development of novel daptomycin analogs with an improved spectrum of pharmaceutical properties.

239

Combinatorial Biosynthesis of Pharmaceutical Natural Products

11.3.3 Patellamide (Ribosomal Peptide Biosynthesis) Patellamides, originally isolated from many didemnid ascidians, are a class of potentially clinical eight-membered cyclic peptides that contain a characteristic thiazole–nonpolar amino acid–oxazoline–nonpolar amino acid sequence [88,89]. The toxicity of most patellamides is moderate, but at least one member in this family exhibits an activity to reverse multidrug resistance in human tumor cells. Recently, two independent research groups showed that the obligate cyanobacterial symbionts Prochloron spp., instead of the ascidian hosts, are the true sources for patellamide production, providing the first evidence to support the hypothesis that many bioactive compounds from marine invertebrates are really made by the symbiotic bacteria they harbor [90,91]. The pat biosynthetic gene cluster has been heterologously expressed in Escherichia coli, confirming its biosynthesis via a ribosome-directed pathway with a distinct strategy to afford the heterocycles from the mode of action of NRPSs. The peptide precursors, encoded by patE, are ribosomally synthesized as pre-propeptides and cleaved to release two different propeptides containing eight amino acid residues in each. These propeptides are then processed to form thiazole and oxazoline moieties from cycteine (Cys), Ser or Thr residues, and converted to the mature macrocyclic products eventually (Figure 11.6).

patA

patB patC

patD

patG

patE patF

patH

O

Natural

S

N H

O

CTLCCTLC

N

N O

(X) 38 GLEASVTACITFCAYDGVEPS

AYDGE O

NH N

Unnatural

O N

N

H N

O O

Ulithiacyclamide

N

NH

NH2

HN

HN

O

S

HN

S

N H

O

N

S

S

QGGRGDWP

O

S

N

H N

NH O

O O O

Patellamide C

NH O N H

O

HO H N O

N H

O

N H H N

O NH

H N

O

N

O O

NH2

Eptidemnamide

Figure 11.6 The patE gene in the patellamide biosynthetic gene cluster encodes the peptide precursor (sequence in rectangle box) of patellamide C and ulithiacyclamide. Substitution of the precursor cassette for ulithiacyclamide with an artificial cassette (sequence in ellipse region) resulted in the biosynthesis of a new unnatural peptide, eptidemnamide

240


A large family of patellamides and related compounds has been isolated from prochloroncontaining didemnid ascidians, implying that the pat pathway has extensively diversified to produce a natural combinatorial library of cyclic peptides [92]. Consistently, extensive analysis of 46 ascidians revealed a multitude of patE variants by polymerase chain reaction amplification. While the 29 identified homologs differed almost exclusively in two hypervariable cassettes encoding the propeptide regions that correspond to the mature natural products, other regions of the gene cluster, including those encoding the tailoring enzymes, as well as the genes encoding 16 rRNA (used as taxonomic markers), were found to be almost identical [93]. Despite the function caused by such a diversification of the pat pathway remaining to be determined, this finding strongly indicates a natural way to create the greatest possible structural diversity by simply mutating the peptide-encoding region, and that strategy could be used to produce new cyclic compounds in vivo [92]. To test how much the coding region could be varied, an engineered patE version that contains two propeptide cassettes has been employed to produce the native compound patellamide C and a new cyclic peptide, structurally close to the clinically used anticoagulant eptifibatide. In the second cassette encoding the latter compound biosynthesis, every single amino acid of the octapeptide sequence had been mutated by introduction of arginine (Arg) and aspartic acid (Asp) for charge increase and amino acids that are not found in various native patE products, including Trp, glycine (Gly) and Glu. Indeed, the predicted cyclic peptide, namely eptidemnamide, was detected in the fermentation broth of the recombinant Escherichia coli strain that harbors the engineered pat gene cluster, supporting the extreme flexibility of the pat machinery [93]. Theoretically, further variation of PatE by genetic engineering could enlarge this combinatorial cyclic peptide library, in which members with cyclic moieties offering structural rigidity and protection against proteolysis might not be synthesized efficiently.

11.4 Summary and Perspectives The rich functionality of natural products is a source of their great strength, providing potency and selectivity. However, the difficulties associated with chemical synthesis may limit their application as drug leads for further development. Herein, we have presented several examples of how combinatorial biosynthesis has been used to introduce structural diversity into pharmaceutical natural products, including an ‘old’ antibacterial agent (erythromycin), a newly approved drug (daptomycin), and a marine natural product with anticancer potential (patellamide). Although it is far from completed, we have tried to highlight the advantages of manipulating the complex structures and enormous potential to develop metabolic capabilities by using the principle and tools of combinatorial biosynthesis. Nature has evolved a stunning diversity of natural products through mutation and recombination processes over billions of years. Although learned from nature, in the past decade the strategies and tools of combinatorial biosynthesis have been dramatically enriched by recent advances in cloning and sequencing of biosynthetic gene clusters, by a fundamental understanding of the biosynthetic mechanisms, and by emerging recombinant DNA technologies [27]. Less costly DNA technologies have driven the expansion of sequencing projects (particularly on genome sequencing), implying that identification of new biosynthetic gene clusters will no longer be the bottleneck for natural product science [28]. Technologies in DNA synthesis and efficient recombination have significantly improved the capabilities of manipulation and


241

construction of increasingly large DNA fragments that encode natural and artificial biosynthetic pathways [33]. Although it is unlikely that genetic techniques can be developed for each organism that produces a natural product of interest, various microbial expression systems have been developed for heterologous production of natural products and their derivatives from inaccessible bacteria, fungi, plant and animal sources [11], aiming to optimize drug leads in the hosts amenable for genetic engineering and eventually providing sufficient yields prepared on a large fermentation scale for further development. No doubt, a better understanding of the fundamental chemistry, biochemistry and genetics governing the biosynthetic machinery will continue to play a critical role in developing current biosynthetic paradigms, designing new strategies for combinatorial biosynthesis, and directing the evolution of enzymes with increased or altered substrate specificity or better catalytic efficiency. Knowledge of new enzymatic activities enhances the combinatorial biosynthesis toolbox, and it is foreseen that new chemistries can be used as a means to introduce a chemical handle for further chemical modification. Since the first description was only two decades ago, combinatorial biosynthesis has advanced from a limited set of proof-of-principle experiments into a more mature scientific discipline. To reach the maximal potential of natural product structural diversity, the combination of this approach with other established and emerging technologies will ultimately provide access to a rich variety of ‘unnatural’ natural products with improved properties or new biological activities for future drug discovery and development.

Acknowledgments This work was supported in part by grants from the National Natural Science Foundation of China (20321202, 30525001, 30770035, and 90713012), the Ministry of Science and Technology of China (2006AA02Z185 and 2006AA020304), the Chinese Academy of Science (KJCX2-YW-H08 and KJCX2-YW-G015), and the Science and Technology Commission of Shanghai Municipality (04DZ14901 and 074319115-3).

References [1] Cragg, G.M., Newman, D.J. and Snader, K.M. (1997) Natural products in drug discovery and development. Journal of Natural Products, 60, 52. [2] Newman, D.J. and Cragg, G.M. (2007) Natural products as sources of new drugs over the last 25 years. Journal of Natural Products, 70, 461. [3] Newman, D.J., Cragg, G.M. and Snader, K.M. (2003) Natural products as sources of new drugs over the period 1981–2002. Journal of Natural Products, 66, 1022. [4] Butler, M.S. (2004) The role of natural product chemistry in drug discovery. Journal of Natural Products, 67, 2141. [5] Wohlleben, W., Spellig, T. and M€uller-Tiemann, B. (eds) (2005) Biocombinatorial Approaches for Drug Finding, Ernst Schering Foundation Symposium Proceedings, vol. 51, Springer, Berlin. [6] Baltz, R.H. (2006) Molecular engineering approaches to peptide, polyketide and other antibiotics. Nature Biotechnology, 24, 1533. [7] Cane, D.E., Walsh, C.T. and Khosla, C. (1998) Harnessing the biosynthetic code: combinations, permutations, and mutations. Science, 282, 63. [8] Clardy, J., Fischbach, M.A. and Walsh, C.T. (2006) New antibiotics from bacterial natural products. Nature Biotechnology, 24, 1541. [9] Fischbach, M.A. and Walsh, C.T. (2006) Directing biosynthesis. Science, 314, 603.

242


[10] Floss, H.G. (2006) Combinatorial biosynthesis – potential and problem. Journal of Biotechnology, 124, 242. [11] Galm, U. and Shen, B. (2006) Expression of biosynthetic gene clusters in heterologous hosts for natural product production and combinatorial biosynthesis. Expert Opinion on Drug Discovery, 1, 409. [12] Hutchinson, C.R., Borell, C.W., Otten, S.L. et al. (1989) A perspective on drug discovery and development through the genetic engineering of antibiotic-producing microorganisms. Journal of Medicinal Chemistry, 32, 929. [13] Hutchinson, C.R. and Fujii, I. (1995) Polyketide synthase gene manipulation: a structure–function approach in engineering novel antibiotics. Annual Review of Microbiology, 49, 201. [14] Kennedy, J. and Hutchinson, C.R. (1999) Nurturing nature: engineering new antibiotics. Nature Biotechnology, 17, 538. [15] Khosla, C. (1996) Combinatorial chemistry and biology: an opportunity for engineers. Current Opinion in Biotechnology, 7, 219. [16] Khosla, C. (1997) Harnessing the biosynthetic potential of modular polyketide synthases. Chemical Reviews, 97, 2577. [17] Khosla, C. and Keasling, J.D. (2003) Metabolic engineering for drug discovery and development. Nature Reviews. Drug Discovery, 2, 1019. [18] Kittendorf, J.D. and Sherman, D.H. (2006) Developing tools for engineering hybrid polyketide synthetic pathways. Current Opinion in Biotechnology, 17, 597. [19] Menzella, H.G. and Reeves, C.D. (2007) Combinatorial biosynthesis for drug development. Current Opinion in Microbiology, 10, 238. [20] Mootz, H.D., Schwarzer, D. and Marahiel, M.A. (2002) Ways of assembling complex natural products on modular nonribosomal peptide synthetases. Chembiochem: A European Journal of Chemical Biology, 3, 490. [21] Salas, J.A. and Mendez, C. (2007) Engineering the glycosylation of natural products in actinomycetes. Trends in Microbiology, 15, 219. [22] Shen, B. (2003) Polyketide biosynthesis beyond the type I, II and III polyketide synthase paradigms. Current Opinion in Chemical Biology, 7, 285. [23] Shen, B., Liu, W. and Nonaka, K. (2003) Enediyne natural products: biosynthesis and prospect towards engineering novel antitumor agents. Current Medicinal Chemistry, 10, 2317. [24] Staunton, J. and Weissman, K.J. (2001) Polyketide biosynthesis: a millennium review. Natural Product Reports, 18, 380. [25] Staunton, J. and Wilkinson, B. (2001) Combinatorial biosynthesis of polyketides and nonribosomal peptides. Current Opinion in Chemical Biology, 5, 159. [26] Strohl, W.R. (2001) Biochemical engineering of natural product biosynthesis pathways. Metabolic Engineering, 3, 4. [27] Thomas, M.G., Bixby, K.A. and Shen, B. (2005) Combinatorial biosynthesis of anticancer natural products, in Anticancer Agents from Natural Products (eds G.M. Cragg, D.G.I. Kingston and D.J. Newman), CRC Press, Boca Raton, FL, pp. 519–551. [28] Van Lanen, S.G. and Shen, B. (2006) Microbial genomics for the improvement of natural product discovery. Current Opinion in Microbiology, 9, 252. [29] Van Lanen, S.G. and Shen, B. (2006) Progress in combinatorial biosynthesis for drug discovery. Drug Discovery Today: Technologies, 3, 285. [30] Walsh, C.T. (2002) Combinatorial biosynthesis of antibiotics: challenges and opportunities. Chembiochem: A European Journal of Chemical Biology, 3, 125. [31] Walsh, C.T. (2004) Polyketide and nonribosomal peptide antibiotics: modularity and versatility. Science, 303, 1805. [32] Weissman, K.J. and Leadlay, P.F. (2005) Combinatorial biosynthesis of reduced polyketides. Nature Reviews Microbiology, 3, 925. [33] Wilkinson, B. and Micklefield, J. (2007) Mining and engineering natural-product biosynthetic pathways. Nature Chemical Biology, 3, 379. [34] Brikun, I.A., Reeves, A.R., Cernota, W.H. et al. (2004) The erythromycin biosynthetic gene cluster of Aeromicrobium erythreum. Journal of Industrial Microbiology & Biotechnology, 31, 335. [35] Bentley, S.D., Chater, K.F., Cerdenõ-Tarraga, A.M. et al. (2002) Complete genome sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature, 417, 141. [36] Jenke-Kodama, H., Sandmann, A., Muller, R. and Dittmann, E. (2005) Evolutionary implications of bacterial polyketide synthases. Molecular Biology and Evolution, 22, 2027.


243

[37] Ahlert, J., Shepard, E., Lomovskaya, N. et al. (2002) The calicheamicin gene cluster and its iterative type I enediyne PKS. Science, 297, 1173. [38] Liu, W., Christenson, S.D., Standage, S. and Shen, B. (2002) Biosynthesis of the enediyne antitumor antibiotic C-1027. Science, 297, 1170. [39] Liu, W., Nonaka, K., Nie, L. et al. (2005) The neocarzinostatin biosynthetic gene cluster from Streptomyces carzinostaticus ATCC 15944 involving two iterative type I polyketide synthases. Chemistry & Biology, 12, 293. [40] Liu, W. and Shen, B. (2000) Genes for production of the enediyne antitumor antibiotic C-1027 in Streptomyces globisporus are clustered with the cagA gene that encodes the C-1027 apoprotein. Antimicrobial Agents and Chemotherapy, 44, 382. [41] Van Lanen, S.G., Oh, T.J., Liu, W. et al. (2007) Characterization of the maduropeptin biosynthetic gene cluster from Actinomadura madurae ATCC 39144 supporting a unifying paradigm for enediyne biosynthesis. Journal of the American Chemical Society, 129, 13082. [42] Liu, W., Ahlert, J., Gao, Q. et al. (2003) Rapid PCR amplification of minimal enediyne polyketide synthase cassettes leads to a predictive familial classification model. Proceedings of the National Academy of Sciences of the United States of America, 100, 11959. [43] Zazopoulos, E., Huang, K., Staffa, A. et al. (2003) A genomics-guided approach for discovering and expressing cryptic metabolic pathways. Nature Biotechnology, 21, 187. [44] Zhang, J., Van Lanen, S.G., Ju, J. et al. (2008) A phosphopantetheinylating polyketide synthase producing a linear polyene to initiate enediyne antitumor antibiotic biosynthesis. Proceedings of the National Academy of Sciences of the United States of America, 105, 1460. [45] Douthwaite, S. and Champney, W.S. (2001) Structures of ketolides and macrolides determine their mode of interaction with the ribosomal target site. The Journal of Antimicrobial Chemotherapy, 48 (Suppl 2), 1. [46] Chen, Y., Deng, W., Wu, J. et al. (2008) Genetic modulation of the overexpression of tailoring genes eryK and eryG leading to the improvement of erythromycin A purity and production in Saccharopolyspora erythraea fermentation. Applied and Environmental Microbiology, 74, 1820. [47] McDaniel, R., Welch, M. and Hutchinson, C.R. (2005) Genetic approaches to polyketide antibiotics. Chemical Reviews, 105, 543. [48] Long, P.F., Wilkinson, C.J., Bisang, C.P. et al. (2002) Engineering specificity of starter unit selection by the erythromycin-producing polyketide synthase. Molecular Microbiology, 43, 1215. [49] Marsden, A.F., Wilkinson, B., Cortes, J. et al. (1998) Engineering broader specificity into an antibiotic-producing polyketide synthase. Science, 279, 199. [50] Pacey, M.S., Dirlam, J.P., Geldart, R.W. et al. (1998) Novel erythromycins from a recombinant Saccharopolyspora erythraea strain NRRL 2338 pIG1. I. Fermentation, isolation and biological activity. The Journal of Antibiotics, 51, 1029. [51] Katz, L. and McDaniel, R. (1999) Novel macrolides through genetic engineering. Medicinal Research Reviews, 19, 543. [52] Liu, L., Thamchaipenet, A., Fu, H. et al. (1997) Biosynthesis of 2-nor-6-deoxyerythronolide B by rationally designed domain substitution. Journal of the American Chemical Society, 119, 10553. [53] Petkovic, H., Lill, R.E., Sheridan, R.M. et al. (2003) A novel erythromycin, 6-desmethyl erythromycin D, made by substituting an acyltransferase domain of the erythromycin polyketide synthase. The Journal of Antibiotics, 56, 543. [54] Ruan, X., Pereda, A., Stassi, D.L. et al. (1997) Acyltransferase domain substitutions in erythromycin polyketide synthase yield novel erythromycin derivatives. Journal of Bacteriology, 179, 6416. [55] Del Vecchio, F., Petkovic, H., Kendrew, S.G. et al. (2003) Active-site residue, domain and module swaps in modular polyketide synthases. Journal of Industrial Microbiology & Biotechnology, 30, 489. [56] Reeves, C.D., Murli, S., Ashley, G.W. et al. (2001) Alteration of the substrate specificity of a modular polyketide synthase acyltransferase domain through site-specific mutations. Biochemistry, 40, 15464. [57] Kumar, P., Koppisch, A.T., Cane, D.E. and Khosla, C. (2003) Enhancing the modularity of the modular polyketide synthases: transacylation in modular polyketide synthases catalyzed by malonyl-CoA:ACP transacylase. Journal of the American Chemical Society, 125, 14307. [58] Stassi, D.L., Kakavas, S.J., Reynolds, K.A. et al. (1998) Ethyl-substituted erythromycin derivatives produced by directed metabolic engineering. Proceedings of the National Academy of Sciences of the United States of America, 95, 7305. [59] Kato, Y., Bai, L., Xue, Q. et al. (2002) Functional expression of genes involved in the biosynthesis of the novel polyketide chain extension unit, methoxymalonyl-acyl carrier protein, and engineered biosynthesis of 2-desmethyl-2-methoxy-6-deoxyerythronolide B. Journal of the American Chemical Society, 124, 5268.

244


[60] Donadio, S., Staver, M.J., McAlpine, J.B. et al. (1991) Modular organization of genes required for complex polyketide biosynthesis. Science, 252, 675. [61] McDaniel, R., Thamchaipenet, A., Gustafsson, C. et al. (1999) Multiple genetic modifications of the erythromycin polyketide synthase to produce a library of novel ‘unnatural’ natural products. Proceedings of the National Academy of Sciences of the United States of America, 96, 1846. [62] Donadio, S., McAlpine, J.B., Sheldon, P.J. et al. (1993) An erythromycin analog produced by reprogramming of polyketide synthesis. Proceedings of the National Academy of Sciences of the United States of America, 90, 7119. [63] Kao, C.M., Luo, G.L., Katz, L. et al. (1994) Journal of the American Chemical Society, 116, 11612. [64] Kao, C.M., Luo, G.L., Katz, L. et al. (1995) Engineered biosynthesis of a triketide lactone from an incomplete modular polyketide synthase. Journal of the American Chemical Society, 117, 9105. [65] Pieper, R., Gokhale, R.S., Luo, G. et al. (1997) Purification and characterization of bimodular and trimodular derivatives of the erythromycin polyketide synthase. Biochemistry, 36, 1846. [66] Xue, Q., Ashley, G., Hutchinson, C.R. and Santi, D.V. (1999) A multiplasmid approach to preparing large libraries of polyketides. Proceedings of the National Academy of Sciences of the United States of America, 96, 11740. [67] Gokhale, R.S., Hunziker, D., Cane, D.E. and Khosla, C. (1999) Mechanism and specificity of the terminal thioesterase domain from the erythromycin polyketide synthase. Chemistry & Biology, 6, 117. [68] Jacobsen, J.R., Cane, D.E. and Khosla, C. (1998) Dissecting the evolutionary relationship between 14-membered and 16-membered macrolides. Journal of the American Chemical Society, 120, 9096. [69] Jacobsen, J.R., Keatinge-Clay, A.T., Cane, D.E. and Khosla, C. (1998) Precursor-directed biosynthesis of 12-ethyl erythromycin. Bioorganic and Medicinal Chemistry, 6, 1171. [70] Hutchinson, C.R. and McDaniel, R. (2001) Combinatorial biosynthesis in microorganisms as a route to new antimicrobial, antitumor and neuroregenerative drugs. Current Opinion in Investigational Drugs, 2, 1681. [71] Kennedy, J., Murli, S. and Kealey, J.T. (2003) 6-Deoxyerythronolide B analogue production in Escherichia coli through metabolic pathway engineering. Biochemistry, 42, 14342. [72] Pfeifer, B.A., Admiraal, S.J., Gramajo, H. et al. (2001) Biosynthesis of complex polyketides in a metabolically engineered strain of E. coli. Science, 291, 1790. [73] Roberts, G.A., Staunton, J. and Leadlay, P.F. (1993) Heterologous expression in Escherichia coli of an intact multienzyme component of the erythromycin-producing polyketide synthase. European Journal of Biochemistry, 214, 305. [74] Peiru, S., Menzella, H.G., Rodriguez, E. et al. (2005) Production of the potent antibacterial polyketide erythromycin C in Escherichia coli. Applied and Environmental Microbiology, 71, 2539. [75] Baltz, R.H., Miao, V. and Wrigley, S.K. (2005) Natural products to drugs: daptomycin and related lipopeptide antibiotics. Natural Product Reports, 22, 717. [76] Miao, V., Co€effet-Le Gal M.-F. et al. (2005) Daptomycin biosynthesis in Streptomyces roseosporus: cloning and analysis of the gene cluster and revision of peptide stereochemistry. Microbiology (Reading, England), 151, 1507. [77] Sieber, S.A. and Marahiel, M.A. (2005) Molecular mechanisms underlying nonribosomal peptide synthesis: approaches to new antibiotics. Chemical Reviews, 105, 715. [78] Hojati, Z., Milne, C., Harvey, B. et al. (2002) Structure, biosynthetic origin, and engineered biosynthesis of calcium-dependent antibiotics from Streptomyces coelicolor. Chemistry & Biology, 9, 1175. [79] Miao, V., Brost, R., Chapple, J. et al. (2006) The lipopeptide antibiotic A54145 biosynthetic gene cluster from Streptomyces fradiae. Journal of Industrial Microbiology & Biotechnology, 33, 129. [80] M€ uller, C., Nolden, S., Gebhardt, P. et al. (2007) Sequencing and analysis of the biosynthetic gene cluster of the lipopeptide antibiotic friulimicin in Actinoplanes friuliensis. Antimicrobial Agents and Chemotherapy, 51, 1028. [81] Coeffet-Le Gal, M.F., Thurston, L., Rich, P. et al. (2006) Complementation of daptomycin dptA and dptD deletion mutations in trans and production of hybrid lipopeptide antibiotics. Microbiology (Reading, England), 152, 2993. [82] Miao, V., Co€effet-Le Gal, M.-F., Nguyen, K. et al. (2006) Genetic engineering in Streptomyces roseosporus to produce hybrid lipopeptide antibiotics. Chemistry & Biology, 13, 269. [83] Gu, J.Q., Nguyen, K.T., Gandhi, C. et al. (2007) Structural characterization of daptomycin analogues A21978C1–3(D-Asn11) produced by a recombinant Streptomyces roseosporus strain. Journal of Natural Products, 70, 233. [84] Nguyen, K.T., Ritz, D., Gu, J.-Q. et al. (2006) Combinatorial biosynthesis of novel antibiotics related to daptomycin. Proceedings of the National Academy of Sciences of the United States of America, 103, 17462. [85] Kopp, F. and Marahiel, M.A. (2007) Macrocyclization strategies in polyketide and nonribosomal peptide biosynthesis. Natural Product Reports, 24, 735.


245

[86] Grunewald, J., Sieber, S.A., Mahlert, C. et al. (2004) Synthesis and derivatization of daptomycin: a chemoenzymatic route to acidic lipopeptide antibiotics. Journal of the American Chemical Society, 126, 17025. [87] Kopp, F., Grunewald, J., Mahlert, C. and Marahiel, M.A. (2006) Chemoenzymatic design of acidic lipopeptide hybrids: new insights into the structure–activity relationship of daptomycin and A54145FNR Biochemistry, 45, 10474. [88] Fu, X., Do, T., Schmitz, F.J. et al. (1998) New cyclic peptides from the ascidian Lissoclinum patella. Journal of Natural Products, 61, 1547. [89] Williams, A.B. and Jacobs, R.S. (1993) A marine natural product, patellamide D, reverses multidrug resistance in a human leukemic cell line. Cancer Letters, 71, 97. [90] Long, P.F., Dunlap, W.C., Battershill, C.N. and Jaspars, M. (2005) Shotgun cloning and heterologous expression of the patellamide gene cluster as a strategy to achieving sustained metabolite production. Chembiochem: A European Journal of Chemical Biology, 6, 1760. [91] Schmidt, E.W., Nelson, J.T., Rasko, D.A. et al. (2005) Patellamide A and C biosynthesis by a microcin-like pathway in Prochloron didemni, the cyanobacterial symbiont of Lissoclinum patella. Proceedings of the National Academy of Sciences of the United States of America, 102, 7315. [92] Piel, J. (2006) Combinatorial biosynthesis in symbiotic bacteria. Nature Chemical Biology, 2, 661. [93] Donia, M.S., Hathaway, B.J., Sudek, S. et al. (2006) Natural combinatorial peptide libraries in cyanobacterial symbionts of marine ascidians. Nature Chemical Biology, 2, 729.

12 Metabolic Engineering for the Development and Manufacturing of Pharmaceuticals Dongping Lu1, Philip G. Williams2 and Guangyi Wang1 1 2

Department of Oceanography, University of Hawaii at Manoa, HI 96822, USA Department of Chemistry, University of Hawaii at Manoa, HI 96822, USA

12.1 Introduction Natural products and their derivatives account for a significant proportion of modern pharmaceuticals. For example, approximately one-third of the best-selling drugs in the world are derived from natural products [1]. In the field of anticancer drug development, 47% of the 155 small-molecule new chemical entities discovered over last six decades are natural products or their derivatives [2]. Clinically relevant natural products originate from a variety of organisms, but particularly plants, fungi, and bacteria [3]. And their extreme structural complexity makes their production by total chemical synthesis infeasible. Thus, the major approaches to the manufacture of natural products for medicinal use (including the semisynthesis products) are product isolation from their native hosts and fermentation of cells from different organisms. However, the former approach requires a large quantity of natural resources and is limited, as harvesting large quantities of biomass is rarely ecologically sustainable. Therefore, much more preferable are sustainable fermentation processes that facilitate the production of a wide variety of natural products by scalable cultivation of microorganisms, plant cells, insect cells, and mammalian cells [4]. A significant advantage of this approach is the ability to increase the production yields of drugs beyond natural levels via strain selection. Random mutagenesis of the strain is accomplished through exposure to radiation or application of chemical agents, followed by screening for colonies with a desired


248


novel phenotype. Although these procedures were effective, especially in the production of some amino acids and antibiotics such as penicillin, the resulting mutations were largely undefined, and the selection procedures were labor intensive [4,5]. Fortunately, the advent of recombinant DNA technology and the subsequent emergence of the field of metabolic engineering has brought about a viable alternative to improve the yield of fermentation-based natural products [6]. Metabolic engineering is the rational modification of specific existing metabolic pathways or the introduction of novel ones using recombinant DNA technology for the overproduction of specific compounds or the production of novel compounds [7]. Thus, to achieve a specific phenotype, metabolic engineering employs a strategy of rational strain improvement using recombinant DNA technology rather than random mutagenesis [8]. The identification of a target gene for rational design involves repetitive genetic perturbation and metabolic evaluation. Conventionally, it is simplified by analysis of the carbon flux distribution to determine the rate-limiting step in a given metabolic process. However, a typical cell catalyzes thousands of intertwined biochemical reactions. Introducing an unnatural pathway alteration often causes the cells regulatory network to divert resources to processes that optimize cellular fitness. This may lead to relatively small strain improvement despite a large increase in specific enzymatic activities [8]. Therefore, the identification of a rational genetic target for metabolic engineering requires an understanding of cellular metabolism as a systematic network. In a cell, control of metabolic processes is hierarchical, with the cell exerting an influence at the different levels simultaneously: transcription, translation, and enzyme activity. Consequently, accurately predicting the cellular response of any genetic (or environmental) perturbation is an extremely complicated procedure and as many regulatory constraints as possible should be taken into consideration [9]. In this chapter, we review metabolic engineering tools and recent progress on the production of polyketide and isoprenoid pharmaceuticals in metabolically engineered hosts.

12.2 Metabolic Engineering Tools Strain improvement via metabolic engineering often involves two basic procedures in an iterative fashion: rational genetic design and metabolic network analysis. Knowledge of the metabolic network provides the basis for subsequent rational target design, while the rational design in turn is tested by the next round of metabolic network analysis, which may identify a new potential target for further genetic modification [8].

12.2.1 Tools for the Cellular Metabolic Network Analysis 12.2.1.1 Metabolic Flux Analysis and Metabolic Control Analysis To rationally improve the strain by metabolic engineering, knowledge of a cells metabolic state is required. Metabolic flux analysis (MFA) is an analytical technique that is used to quantify the metabolic fluxes through all intracellular reactions, thereby dissecting the functional aspects of a metabolic network into greater detail and elucidating the metabolic state of the cells in vivo [10,11]. Metabolite labeling, using stable isotopes (13 C, 15 N, etc.), is used to determine metabolic flux [8,11]. Spectroscopic techniques such as nuclear magnetic resonance (NMR) or gas chromatography (GC)–mass spectrometry (MS) provide quantitative

Metabolic Engineering for the Development and Manufacturing of Pharmaceuticals

249

information via analysis of the relative isotopic abundances (enrichment patterns) before and after the stable isotope feeding experiments, which are used to determine the central metabolic fluxes during an assumed steady state [10]. MFA is a useful tool for the identification of metabolic engineering targets. For example, using the 13 C-labeled gluconate tracer method, the fluxes through the oxidative branch of the pentose phosphate pathway (PPP; the main source of cytosolic NADPH) in penicillin-G-producing and -nonproducing chemostat cultures of Penicillium chrysogenum were compared [12]. Significantly higher oxidative PPP fluxes were observed in the penicillin-G-producing cultures, suggesting that penicillin production requires an ample supply of NADPH [10,12], leading to the hypothesis that PPP was a metabolic engineering target for generating a strain with higher penicillin-G production. After measuring the fluxes through the metabolic network, it is necessary to determine the extent to which each pathway or enzyme controls the net fluxes. Metabolic control analysis (MCA) is a technique used to elucidate how flux control is distributed in a metabolic network, thereby providing the information for identification of potential targets for metabolic engineering [8]. 12.2.1.2 Transcriptome Profiling DNA microarrays are a high-throughput method for measurements of gene expression on a genome-wide basis [13]. The application of microarrays in metabolic engineering enables the identification of target genes (such as genes encoding repressors, enhancers and other regulatory factors) for rational genetic alteration, in addition to metabolic enzyme genes [8]. This transcriptome information could be obtained by comparing the transcripts profiles of strains with different productivity, different genetic architecture, and with various mutations [13]. A high proliferation rate is one of the most important properties for a cell line producing therapeutic or diagnostic compounds [14]. Using DNA microarrays, the transcripts of two HeLa cell lines, a slow- and a fast-growing one, were compared to identify genes potentially influencing cellular growth [15]. Two genes, cdkl3 and cox15, had higher expression in the fastgrowing cells than in the slow-growing cells. When the cDNA for both genes was transfected into several cell lines, the overexpression of both genes resulted in elevated rates of cellular proliferation [15]. This study demonstrates that DNA microarrays can be a powerful tool for identifying rational targets for subsequent metabolic engineering [16]. The application of DNA microarrays for metabolic engineering is not only restricted to organisms where full genomic sequences are available. For less well studied microbes, genomic DNA fragments can be used to construct microarrays. For example, genomic fragment microarrays were constructed for the Aspergillus terreus genome, and transcriptional profiles were generated from strains engineered to produce varying amounts of the medically significant natural product lovastatin. By analysis of the transcriptional data, along with the metabolic information, novel components involved in the production of ( þ )-geodin were identified [17]. 12.2.1.3 Proteome Profiling Proteome profiling allows quantitative and global examination of the changes in protein expression in different strains or cells under different conditions. Two-dimensional gel

250


electrophoresis (2DE) followed by MS are employed to separate, identify, and quantify proteins [18]. Thus, it seems that proteomics allows a closer understanding of gene function than transcriptome analysis [13]. However, compared with transcriptomics, proteome profiling has the drawbacks of relatively low throughput and requires purifying the proteins from 2DE gels prior to analysis [19]. Furthermore, transcriptomic and proteomic studies on the same subject have thus far either failed to find a correlation or only exhibited a poor correlation between the two approaches [20]. Nonetheless, it is still possible to identify protein spots showing altered expression levels, which may be helpful to provide information for rational design of metabolic engineering targets [21]. Proteomic approaches have been employed successfully. Leptin is a 16 kDa protein hormone that conveys information about the bodys energy stores to the brain, thus regulating several physiologic processes, including inflammation, angiogenesis, hematopoiesis, immune function, and reproduction [22]. With the aim of increasing leptin productivity in recombinant Escherichia coli, proteomics were performed to identify target genes for metabolic engineering [21]. The proteome profiles of E. coli before and after leptin production were examined by 2DE. The results revealed that the levels of heat shock proteins were increased, while enzymes involved in the biosynthesis of the serine family of amino acids (GlyA and CysK) were decreased upon leptin overproduction. This indicates that the overproduction of leptin in recombinant E. coli leads to an imbalance between the free amino acid stocks and the translational machinery. To overcome this bottleneck, cysK, encoding for cysteine synthase A, was coexpressed in leptin-producing cells, resulting in a fourfold increase in leptin productivity [21]. Proteins involved in glucagon-like peptide-1 (GLP-1) production have also been identified by proteomic approaches. Human GLP-1 is a 31-amino-acid peptidic hormone responsible for stimulating insulin secretion after meal intake [23]. Owing to its therapeutic potential for the treatment of type 2 diabetes [24], recombinant E. coli was constructed to produce GLP-1 [25]. The protein expression profiles of a recombinant E. coli producing GLP-1 and E. coli cells without the target gene were compared by 2DE and analyzed by MS [25]. There were 35 intracellular proteins displaying differential expression levels between the two strains, with the many up-regulated proteins involved in cell protection and sugar transport, suggesting metabolic engineering strategies for the production of clinically recombinant peptides in engineered E. coli [25]. 12.2.1.4 Metabolites Profiling Metabolomics is the comprehensive (qualitative and quantitative) analysis of the complete set of low-molecular-weight metabolites present in and around growing cells at a given time during their growth or production cycle [26]. Correspondingly, the metabolome represents the entirety of the low-molecular-weight metabolites within a biological system [27]. Commonly used analysis tools for identifying and quantifying cellular metabolites include GC–MS, capillary electrophoresis (CE)–MS, high-pressure liquid chromatography (HPLC), and liquid chromatography (LC)–tandem MS (MS/MS) [19]. Metabolite profiling information provides a direct link between the metabolites and the metabolic pathways [13]. Thus, it seems feasible to apply metabolome analysis for strain improvement by metabolic engineering. However, there are few successful examples of rational alteration of target genes identified via metabolite profiling, perhaps owing to the infancy of the field. Nevertheless, a range of pathways have been


251

monitored and analyzed by metabolic profiling, and certain potential engineering targets have been proposed [27,28]. One such effort was made on the Taxus cuspidata cells from suspension cultures induced with methyl jasmonate to produce paclitaxel, a diterpene used as an anticancer agent [29].

12.2.2 Tools for Rational Genetic Modification Once the targets for rational strain improvement have been determined, the genes can be modified using recombinant DNA technologies, with the endpoint of overexpression or inactivation of these genes in suitable host organisms, such as E. coli, yeast, or fungi. 12.2.2.1 Vector The introduction of target genes into the host and subsequent gene expression are implemented with vectors; these are usually plasmids, which can be classified as low-copy, moderate-copy or high-copy plasmids [30]. For many applications, high-copy-number plasmids are often favored, because normally a high gene dosage corresponds to high expression levels and elevated titers of the metabolites. The plasmid copy number is determined by the origin of replication (ori); thus, mutations in this region can lead to an increased number of plasmid copies. For example, pBBR1 is a broad-host-range plasmid isolated from Bordetella pertussis [31], which has been used in a variety of applications [32,33]. To increase the copy number of pBBR1, its replication control region was subjected to mutagenesis resulting in three- to sevenfold higher copy numbers than the original plasmid. To demonstrate its application for metabolic engineering, the b-carotene synthesis genes were expressed on the mutant pBBR1 plasmids in E. coli, and the engineered strain produced using the increased-copy-number vector showed enhanced b-carotene production [34]. Although a high-copy-number plasmid can drive rapid expression, it exerts a metabolic burden on the host cells, which may limit its effect on the gene expression [35]. Therefore, lowcopy-number plasmids should also be taken into consideration for metabolic engineering, especially under some special circumstances; for example, when the expressed proteins or metabolites have detrimental effects on the host cells [36]. In the following example, a lowcopy-number expression vector derived from E. coli F plasmid was compared with a highcopy-number pMB1-based plasmid for production of lycopene in E. coli [37]. When the gene encoding DXP (1-deoxy-D-xylulose 5-phosphate) synthase (dxs) was expressed from the tac promoter on high-copy and low-copy plasmids, both demonstrated enhanced lycopene production by two- to three-fold over that of the control cells. However, when isopropyl b-D-thiogalactosidase (IPTG) was applied to the cultures, the one containing high-copynumber plasmid showed substantially decreased cell growth and lycopene production, suggesting that overexpression of DXP synthase imposes a metabolic burden to the host cell. Conversely, in the strain derived from the low-copy-number plasmid, no differences in cell growth or lycopene production were observed. Moreover, when dxs was placed under the control of the arabinose-inducible promoter (PBAD) on the low-copy-number plasmid, the lycopene production was proportional to the arabinose added, while no significant change in cell growth was observed [37]. One of the concerns on plasmid selection for metabolic engineering is plasmid stability problems, which include structural stability and plasmid segregational stability [30].

252


Structural stability is the correct replication over cultivation of plasmid-containing cells without altering the base sequence, and segregational plasmid stability is the successful partitioning of plasmids on to the daughter cells. Because plasmids are always a metabolic burden for the host cells, plasmid-free cells are preferred over plasmid-bearing cells during cultivations [38]. Plasmid-free cells and plasmid structural instability can significantly erode the metabolite productivity, and several methods have been developed to minimize their negative effects [30]. An alternative to plasmids is integrating the target gene into the host chromosome. There are several different ways to integrate the target genes into the host chromosome [39,40]. A more efficient system allowing integration of linear target genes directly into the host chromosome via recombination with l-Red recombinase has been proved to be extremely powerful [41]. Using this method, five genes involved in the production of the polyketide product 6-deoxyerythronolide B (6dEB, a precursor to the antibiotic erythromycin) were integrated into the E. coli chromosome, resulting in a plasmid-free engineered strain that is able to produce 6dEB at a wider range of temperatures than the control [42].

12.2.2.2 Promoters The strength of the promoter plays a pivotal role in regulating the expression of the target gene in engineered host cells, thereby affecting the overall flux through the pathway of interest. Thus, it is essential to place the target gene under a promoter that allows tight control over the genes expression [43]. The promoters used for metabolic engineering application include inducible prompters and constitutive ones. Their utilities and effects in metabolic engineering are dependent on the engineering systems in which they work. The native promoters of several genes involved in the biosynthesis of isopentenyl pyrophosphate (IPP), the building unit of b-carotene, were replaced with a strong phage T5 promoter (PT5) using the method of PCR-mediated l-Red recombination. This resulted in the construction of an E. coli PT5-dxs PT5-ispDispF PT5-idi PT5-ispB strain with higher titers (6 mg/g dry cell weight) of b-carotene than the control strain [44]. In an example of metabolic engineering for the production of lycopene in E. coli, the dxs gene was placed on several expression vectors under the control of three different inducible promoters: the arabinose-inducible araBAD promoter PBAD and the IPTG-inducible trc and lac promoters, PTRC and PLAC respectively. The highest lycopene production was obtained when the dxs gene was expressed under the control of PBAD, which was twofold higher than that under the control of PTRC and PLAC [45]. This method is applicable to fungi as well. For example, a specific region of the HMG1 gene from Saccharomyces cerevisiae (encoding the catalytic domain (cHMG1) of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase) was expressed in Neurospora crassa, a filamentous fungus, to improve the lycopene production. The cHMG1 gene was put under the control of two different promoters: one a strong, constitutive glyceraldehyde-3-phosphate dehydrogenase (GPD) promoter and the other an inducible alcohol dehydrogenase (alcA) promoter. A comparison of the results of these two promoters revealed that the strain overexpressing cHMG1 under control of the GPD promoter increased lycopene production by sixfold relative to the wild-type strain, while the overexpression of cHMG1 under the alcA promoter increased production by threefold [46].


253

12.2.2.3 The Coexpression of Multiple Genes The application of metabolic engineering often requires manipulation of more than one gene. Therefore, coordination of multiple genes in the host is essential for the production of the desired metabolites. Solutions using genes coordination by placing them under different inducible promoters was summarized in the review by Keasling [43]. In addition, to bypass the use for multiple plasmids, the multiple genes could be manipulated to function together in a synthetic operon. For instance, to engineer E. coli for production of amorphadiene, a precursor of artemsinin, the eight genes encoding the mevalonate-dependent isoprenoid pathway from Saccharomyces cerevisiae were assembled into two operons and expressed in E. coli. The results indicted that synthetic operons were functional and capable of supplying IPP and DMAPP [47].

12.3 Metabolic Engineering for the Development and Production of Polyketide Pharmaceuticals Polyketides are a large group of structurally diverse and complex natural products built from malonyl-CoA monomers. The biosynthesis of polyketides occurs through the stepwise condensations of malonyl-CoA esters. There are about 10 000 polyketides that have thus far been identified from bacteria, fungi, and plants. Although their physiological roles in the native organisms are not clear yet, many polyketides display a variety of pharmacological bioactivities, which include antibacterial activity (tetracycline, erythromycin, rifamycin), antifungal activity (amphotericin), anticancer (doxorubicin), and immunosuppression (FK506, rapamycin) [48].

12.3.1 Biosynthesis of Polyketides The biosynthesis of polyketides (including chain initiation, elongation, and termination processes) is catalyzed by large multi-enzyme complexes called polyketide synthases (PKSs). The polyketides are synthesized from starter units such as acetyl-CoA, propionyl-CoA, and other acyl-CoA units. Extender units such as malonyl-CoA and methylmalonyl-CoA are repetitively added via a decarboxylative process to a growing carbon chain. Ultimately, the polyketide chain is released from the PKS by cleavage of the thioester, usually accompanied by chain cyclization [49]. There are at least three types of PKS. Type I PKSs catalyze the biosynthesis of macrolides such as erythromycin and rapamycin. As modular enzymes, they contain separate catalytic modules for each reaction catalyzed sequentially in the polyketide biosynthetic pathway. Type II PKSs have only a few active sites on separate polypeptides, and the active sites are used iteratively, catalyzing the biosynthesis of bacterial aromatic polyketides. Type III are fungal PKSs; they are hybrids of type I and type II PKSs [49,50].

12.3.2 Metabolic Engineering for Improved Erythromycin Production Owing to the extreme complexity of their structures, polyketides for clinical use are produced mainly by fermentation rather than chemical synthesis. However, the levels of polyketides produced by microbes from nature are not high enough for commercial purposes; thus, strain

254


improvement became an inevitable part of drug development. Extensive effort has been focused on increasing the production of erythromycin, a broad-spectrum antibiotic, which is naturally produced by the soil bacterium Sacchropolyspora erythraea. In recent years, metabolic engineering has been applied and notable improvement has been achieved in the titers. In one of these examples, reverse engineering was exploited instead of a random mutagenesisand-screening method. This approach identified a genetic basis for rational strain improvement [51]. Using this strategy, an engineered Sacchropolyspora erythraea strain was developed with improved erythromycin production [51,52]. First, a different erythromycin-producing organism, Aeromicrobium erythreum, was used in place of the commercial erythromycinproducing organism, Sacchropolyspora erythraea, because Aeromicrobium erythreum grows faster and is easier to handle than Sacchropolyspora erythraea. Then, a transposon-tagged mutagenesis approach was employed that enables the identification of a number of mutants with improved erythromycin production. DNA sequence analysis of the transposon insertion site revealed one mutant with a knockout in the mutB gene that encodes the large (beta) subunit of methylmalonyl-CoA mutase (MCM). This knocking-out mutB gene resulted in a permanent metabolic switch in the flow of methylmalonyl-CoA to a secondary branch from the primary metabolic branch, increasing erythromycin overproduction. These results provided the rationale to delete the mutB gene either by a double crossover gene replacement or by in-frame deletion, and the resulting strain exhibited improved erythromycin production in carbohydrate-based and oil-based fermentations of Sacchropolyspora erythraea [51,52]. In addition, because it is used for the industrial-scale production of erythromycin A, the genome of Sacchropolyspora erythraea has been recently sequenced. Analysis of the genome sequence will provide further insight into rational for developing strains with improved yields of antibiotics [53].

12.3.3 Metabolic Engineering for Overproduction of 6dEB in Heterologous Hosts Many naive polyketide-producing organisms are not ideal hosts for large-scale and high-level polyketide production, as they grow slowly, produce low yields of polyketides, or are difficult to be manipulated genetically [54]. These problems could be potentially overcome by heterologous production of these polyketides in other hosts. Recently this was accomplished for 6dEB, the polyketide part of erythromycin. The synthesis of 6dEB is catalyzed by a type I PKS, 6dEB synthase (6DEBS) (Figure 12.1). This modular enzyme is composed of the three polypeptides DEBS1, DEBS2, DEBS3 [55,56], which contain a total of six modules, each catalyzing one of the six elongation steps required for polyketide formation [55]. Each module consists of several domains, including acyltransferase (AT), acyl carrier protein (ACP), and b-ketoacyl carrier protein synthase (KS), required for one cycle of chain elongation. Additional catalytic domains of 6DEBS that are responsible for these subsequent reduction reactions include a keto reductase (KR), a dehydratase (DH), and an enoyl reductase (ER) [50,55,56]. In Sacchropolyspora erythraea the starter unit of 6dEB synthesis is propionyl-CoA and the extender unit is (2S)-methylmalonyl-CoA. After termination, many polyketide chains are subjected to further modifications by other enzymes, such as glycosyltransferases, acyltransferases, and oxidases [57]. Notable progress has been made on the production of 6dEB in engineered E. coli. E. coli has several advantages as a heterologous host, such as its relatively fast growth rate, the

255

Metabolic Engineering for the Development and Manufacturing of Pharmaceuticals (a)

O

SCoA (2S)- Methylmalonyl-CoA

HO2C

DEBS 1 (370 kDa)

DEBS 2 (380 kDa)

Module 1 Loading

S

SCoA

Module 5

Module 4

Module 2

AT ACP KS ATKR ACP KS AT KR ACP

O

S

DEBS 3 (332 kDa)

Module 3

AT S

O

O

HO

HO

O

KS ATKR ACP KS AT KR ACP TE

O

S

S

S

S

O

HO

Module 6 Release

KS AT ACP KS AT DHER KR ACP

O

O

HO

HO

O

HO

O

HO

HO

O

HO

HO

O

HO

HO

HO Propinoyl-CoA

(b)

HO

AT ACP KS ATKR ACP KS ATKR TE

OH

O

SCoA O O

O

OH

6dEB 1

2

O O

OH OH

Figure 12.1 The biosynthesis of 6dEB catalyzed by DEBS [58]. (From B.A. Pfeifer, S.J. Admiraal, H. Gramajo et al.‘‘Biosynthesis of complex polyketides in a metabolically engineered strain of E. coli.’’ Science 291 : 1790–1792. Washington, DC: AAAS. Ó 2001 AAAS. Reprinted with permission from AAAS.)

availability of well-established techniques for its genetic manipulation, and relatively cheap starting material. An engineered E. coli strain developed by Pfeifer et al. [58] was able to produce 6dEB from exogenous propionate. In this strain, three DEBS genes (eryA1, eryA2, eryA3) under the control of an IPTG-inducible T7 promoter in two plasmids were introduced into E. coli. Moreover, a single copy of the sfp gene from Bacillus subtilis, encoding the phosphopantetheinyl transferase needed to posttranslationally activate DEBS enzyme, was integrated in the prp operon of the chromosome, resulting in the disruption of the prp operon, which is putatively responsible for propionate catabolism in E. coli. Together with the sfp gene, the native prpE gene was also placed under control of a T7 promoter. PrpE acts on the conversion of propionate into propionyl-CoA. Thus, with the overexpression of PrpE, propionyl-CoA will be produced efficiently from the exogenous propionate at the same time as active DEBS is formed. The supply of (2S)-methylmalonyl-CoAwas ensured by introduction of propionyl-CoA carboxylase (PCC) genes pccB and accA1 from Streptomyces coelicolor as the PCC enzyme catalyzes the conversion of propionyl-CoA into (2S)-methylmalonyl-CoA. This engineered strain was able to produce 6dEB heterozygously in E. coli. With the development of a high cell-density fed-batch bioprocess, the titer of 6dEB reached 80 to 100 mg L1 [59]. Furthermore, via optimization of the medium composition and fermentation conditions, the maximum cell density was increased by twofold, yielding a final titer of 1.1 g L1 of 6dEB [60]. In another approach, Murli et al. [61] evaluated the three pathways of (2S)-methylmalonylCoA biosynthesis: (1) Streptomyces coelicolor PCC, (2) Propionibacteria shermanii MCM/

256


epimerase, and (3) Streptomyces coelicolor malonyl/methylmalonyl-CoA ligase (matB). They found the highest titer was obtained when the strain contained the PCC pathway. Saccharomyces cerevisiae might also be an ideal host for metabolic engineering of heterologous polyketide production. An engineered Saccharomyces cerevisiae strain carrying the gene encoding 6-methylsalicylic acid synthase (6-MSAS, Penicillium patulum), and the gene sfp from Bacillus subtilis, is capable of producing a simple fungal polyketide, 6-methylsalicylic acid (6-MSA) with a remarkable titer of 1.7 g L1 in unoptimized shakeflask fermentations [49]. The production of complex polyketides by Saccharomyces cerevisiae is limited by its inability to biosynthesize the required precursors. Recently, the pathways for the production of methylmalonyl-coenzyme A (CoA), a precursor for complex polyketides, by both propionyl-CoA-dependent and propionyl-CoA-independent routes were introduced into Saccharomyces cerevisiae, resulting in the heterozygous production of a triketide lactone [54].

12.3.4 Metabolic Engineering of Other Polyketides The industrial organism Streptomyces lividans was used as the native host to produce two polyketides, antibiotics actinorhodin (ACT) and undecylprodigiosin (RED). ACT is derived from the intermediate acetyl-CoA, while the tripyrrole nucleus of RED is derived from pyruvate and amino acid building blocks [62]. Polyketide production was elevated through transformation of Streptomyces lividans with multicopy plasmids carrying the pathwayspecific transcriptional activator genes for either the ACT or RED biosynthetic pathway [62,63]. By applying metabolic flux analysis to a stoichiometric model, it was found that the synthesis of both ACT and RED was related to decreased carbon flux through the pentose-phosphate pathway. The production of antibiotics could be increased via deletion of genes encoding enzymes in the pentose-phosphate pathway [64]. The lower carbon flux through the pentose-phosphate pathway may increase the glucose utilization via glycolysis, which results in the accumulation of more acetyl-CoA or pyruvate, the precursors of ACT or RED [64,65].

12.3.5 Development of Novel Polyketides for Drug Discovery The modularity of type I PKSs along with the linearity between its catalytic domains make it possible to develop hybrid PKSs by combinatorial methods for the production of unnatural polyketides [66,67]. Hybrid PKS genes are constructed by substitution of sequences encoding one or more catalytic domains from one PKS gene to another. Several novel macrolactones were produced by the plasmid-based exchange of the protein subunits of the pikromycin PKS in Streptomyces venezuelae with the corresponding subunits from tylosin (Tyl) and erythromycin (DEBS) PKSs [66]. More recently, a facile approach to designing synthetic PKS genes facilitates the interchange of modules and domains from different PKSs [68], wherein sequence conservation near domain edges was exploited after aligning the amino acid sequences of 140 modules from 14 PKSs. When reverse-translated to all possible DNA sequences, a conserved six-base-pair restriction site could often be identified. Thus, novel PKSs could be assembled from ‘building blocks’ flanked by these unique restriction sites. Novel polyketides could also be generated by blocking certain catalytic domains of PKSs to allow bypass of the early steps in polyketide biosynthesis. In one example, the Cys729 ! Ala


257

mutation in the active site was introduced into the DEBS KS1 domain that acts in the first condensation step of the 6dEB biosynthetic pathway. Exogenous addition of designed synthetic molecules to this null mutant (KS10) resulted in highly selective production of unnatural polyketides, including aromatic and ring-expanded variants of 6dEB [69]. When the DEBS loading domain and first module are removed, the utilization of some chemical precursors was largely increased over its KS10 counterpart [70]. This new finding may lower the cost of synthesis of the thioester for precursor-directed polyketide biosynthesis.

12.4 Metabolic Engineering for the Production of b-Lactam b-Lactams are a broad class of antibiotics that include penicillin derivatives, cephalosporins, monobactams, carbapenems, and clavams (b-lactamase inhibitors). The metabolic engineering of penicillin and cephalosporins production has been summarized by several good reviews [71,72], so the focus here is clavulanic acid, which has attracted interest in recent years. Although penicillin and cephalosporins have been effective in the treatment of microbial infections, resistance to these antibiotics by pathogens has become a big problem. A variety of bacteria can produce b-lactamases, which hydrolyze the b-lactam ring of penicillins and cephalosporins, resulting in the inactivation of these antibiotics [73]. Clavulanic acid is a clavam produced by Streptomyces clavuligerus. Clavams are a class of b-lactams with oxygen substituted for the sulfur atoms in lactam rings of penicillins and cephalosporins. Although clavulanic acid has poor antibiotic activity, it exhibits a potent ability to inhibit a wide range of b-lactamases [74]. Therefore, clavulanic acid is used to protect penicillins and cephalosporins from the b-lactamases of resistant pathogens. Being combined with the commercial semisynthetic penicillin amoxicillin in the product augmentin and with ticarcillin in the pharmaceutical timentin, clavulanic acid has a market of over $2 billion per year [75,76], increasing demand for a high-yielding Streptomyces clavuligerus strain for the production of this valuable product. Clavulanic acid is synthesized by the condensation of L-arginine and D-glyceraldehyde3-phosphate (G3P) as the first step [75,77] (Figure 12.2). A series of experiments revealed that the synthesis of clavulanic acid was limited by the availability of the C3 precursor, resulting from the speciess limited ability to assimilate glucose [78]. Thus, the enhancement of clavulanic acid production requires alleviation of competition from other pathways for a C3 precursor [79]. To overcome the limited G3P pool and improve clavulanic acid production, two genes (gap1 and gap2) whose protein products are distinct glyceraldehyde-3-phosphate dehydrogenases (GAPDHs) were inactivated in Streptomyces clavuligerus by targeted gene disruption. The production of clavulanic acid was consistently doubled when only gap1 was disrupted [80]. Streptomyces clavuligerus is able to produce both clavulanic acid and cephamycin. The genes encoding proteins involving clavulanic acid biosynthesis are clustered and located adjacent to the cephamycin gene cluster [81–84]. The production of both cephamycin C and clavulanic acid is positively regulated by the same transcription activation protein, CcaR [85]. The amplification of ccaR in Streptomyces clavuligerus resulted in an increased yield of clavulanic acid by about threefold [85], whereas an enhancement of 9.28-fold was obtained when ccaR was integrated into the Streptomyces clavuligerus chromosome [86].

258

Biocatalysis for the Pharmaceutical Industry OH O

OP

Glyceraldehyde-3-Phosphate +

O HO NH2

Carboxyethylarginine synthase (ceaS1 and ceaS2)

NH2 N H

NH

CO2H

NH

β-lactam synthetase (bls1 and bls2)

OH N

N H

CO2H

OH

L-Arginine

O

NH2

H N

O

Clavaminate synthase (cas1 and cas2)

NH2 N H

NH2 N

N H

O

NH

CO2H

NH

Guanidnoproclavaminic Acid

Proclavaminate amidinohydrolase (pah1 and pah2)

OH N O

Clavaminate synthase (cas1 and cas2)

O

H

N O

NH2

CO2H Proclavaminic Acid

Clavaminate synthase (cas1 and cas2) NH2

N O

CO2H

CO2H Clavaminic Acid

Dihydroclavaminic Acid

R = CO2H CH2OH CH2OCHO O CH2CHNH2CO2H

O N

O R

CO2H

5S Clavams

NH2

O

OH

N O

CO2H

Clavulanic Acids

Figure 12.2 The biosynthesis of the clavulanic acid pathway [77]. (Reproduced by permission from Macmillan Publishers Ltd: M.C.Y. Chang and J.D. Keasling. Production of isoprenoid pharmaceuticals by engineered microbes. Nature Chemical Biology 2 (12): 674–682. London: Nature Publishing Group. Ó 2006 Macmillan)

12.5 Metabolic Engineering for Isoprenoid Production Isoprenoids, also referred to as terpenoids, are a diverse class of compounds originated from two five-carbon building blocks, isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), by an array of reactions, including cyclization, hydroxylations, and acylations, which result in the production of isoprenoids with extremely complex and diversified structures [87]. As the largest group of natural products, isoprenoids have been isolated from plant, animal, and microbial species and play a range of biological roles in these organisms [88]. Isoprenoids in plants perhaps are evolved to be more diversified and able to


259

serve more essential functions, such as hormones (abscisic acid, cytokinins, gibberellins, and brassinosteroids), photosynthetic pigments (carotenoids), electron carriers (cytochrome a, quinones, chlorophylls), membrane components (steroids), and insect attractants (essential oils (b-damascenone) and flower colors (carotenoids)) [89]. Besides their essential roles in nature, isoprenoids are of commercial importance in industry. Some isoprenoids have been used as flavors, fragrances, spices, and food additives, while many are used as pharmaceuticals to treat an array of human diseases, such as cancer (Taxol), malaria (artemisinin), and HIV (coumarins). In contrast to the huge market demand, isoprenoids are present only in low abundance in their host organisms. Thus, isolation of the required isoprenoids consumes a large quantity of natural resources. Furthermore, owing to their structural complexity, total chemical synthesis is often not commercially feasible. For these reasons, metabolic engineering may provide an alternative to produce these valuable isoprenoids [88,89].

12.5.1 Biosynthesis Pathway of Isoprenoids 12.5.1.1 Biosynthesis of the IPP and DMAPP Building Blocks There are two distinct pathways for biosynthesis of the IPP and DMAPP: the mevalonate (MVA) pathway and the DXP pathway (Figure 12.3). The MVA pathway functions primarily in eukaryotes, while the DXP pathway is typically present in prokaryotes and the plastids of plants [90,91]. The first reaction in the DXP pathway is the condensation of pyruvate and D-glyceraldehyde-3-phosphate (G3P) to form DXP, which is catalyzed by DXP synthase encoded by the gene dxs [92]. In the second step, DXP is reduced to 2-C-methyl-D-erythritol4-phosphate (MEP) by DXP reductoisomerase, which is encoded by the gene dxr (ispC) in E. coli. An array of other enzymes encoded by ispD, ispE, ispF, ispG, and ispH act in subsequent sequential reactions, leading to the conversion of MEP to IPP and DMAPP, which are interconverted by the enzyme encoded by idi [93–97]. In Saccharomyces cerevisiae, the first reaction in the MVA pathway involves the production of acetoacetyl-CoA from two acetyl-CoA molecules, catalyzed by the enzyme acetoacetyl-CoA thiolase, which is encoded by the gene ERG10. Then, acetoacetyl-CoA is condensed with acetylCoA to form 3-methylglutaryl-CoA (HMG-CoA) by HMG-CoA synthase, encoded by the ERG13 gene. The reduction of HMG-CoA into mevalonate is catalyzed by HMG-CoA reductase, in Saccharomyces cerevisiae, which is encoded by two copies of gene: HMG1 and HMG2. This step has received the most extensive study. The subsequent steps are catalyzed by enzymes which are encoded by ERG12, ERG8, ERG19, and IDI1 to yield IPP and DMAPP [98]. 12.5.1.2 Subsequent Steps to Form Isoprenoids In the following phases, DMAPP primes the sequential head-to-tail condensations of IPP molecules to yield the linear polyprenyl diphosphate precursors by the action of prenyl transferases. These precursors include geranyl diphosphate (GPP, C10, monoterpenoids), farnesyl diphosphate (FPP, C15, sesquiterpenoids), and geranylgeranyl diphosphate (GGPP, C20, diterpenoids). Enzymes such as squalene synthase generate larger terpene precursors by the condensation of these polyprenyl diphosphates. Then, many diverse carbon skeletons found in terpenoid natural products are formed through the cyclizations and/or rearrangements of these linear precursors by terpene synthases (also referred as terpene cyclases). These skeletons are further modified to form the native structures with bioactivity [99] (Figure 12.3).

260


(a) O

O

AAS

2 CoA

ATP

ATP

OH

OP PMK Mevalonate-5-phosphate

Mevalonate

HMGR

Hydroxymethylglutaryl-CoA

HO2C

MK

OH

2 NADPH CoA

HMGS

Acetoacetyl-CoA

OH

OH O HO2C

CoA

Acetyl-CoA

HO2C

Acetyl-CoA

O

PMD

OH HO2C

OPP Mevalonate diphosphate

IDI OPP

OPP DMAPP

IPP

(b) O

O

O Pyruvate

OH

OH

-CO2

OH +

O

OH Glyceraldehyde -3-phosphate

OH OH 4-(Cyt-5'-diphospho)-ME

OH OH

OH

IspF

OH

CTP IspD IspH

IspH

ME-4-phosphate

OP PO IspG

OPP-cyt OH

OP

DXS/IspC

1-Deoxylulose -5-phosphate OP

ATP OPP-cyt IspE

OH

NADPH OP

OP DXS

OH OH

OPP OH HMB-4-diphosphate

Figure 12.3 (a) The mevalonate pathway (AAS, acetoacetyl-CoA synthase; HMGS, HMG-CoA synthase; HMGR, HMG-CoA reductase; MK, mevalonate kinase; PMK, phosphomevalonate kinase; PMD, mevalonate diphosphate decarboxylase). (b) The DXP pathway (ME, 2-methylerythritol; cyt, cytidine; HMB, 1-hydroxy-2-methyl-2-butenyl; IspD, ME-4-phosphate cytidyltransferase; IspE, 4-(cyt50 -diphospho)-ME kinase; IspF, ME-2,4-cyclodiphosphate synthase; IspG, HMB-4-diphosphate synthase; IspH, HMB-4-diphosphate reductase) [88]. (Reproduced with permission from M. Wang, B. Cornett, J. Nettles, D.C. Liotta, and J.P. Snyder. The oxetane ring in taxol. The Journal of Organic Chemistry Feb 25, 2000. 65(4):1059–68. Washington DC: ACS. Ó 2000 ACS)

12.5.2 Metabolic Engineering for Enhancing Precursor Supply for Isoprenoid Production The IPP monomer serves as the universal building block for the production of all isoprenoids, including artemisinine, carotenoids, and Taxol. Thus, an engineered strain with high potential for generating IPP provides a platform for production of a variety of complex isoprenoids. The presence of two IPP synthesis pathways allows two approaches for engineering such strains. One is to introduce a heterozygous pathway and the other is to alter or modify the native pathway. Both approaches have been accomplished in E. coli. Lycopene is a carotenoid with anticancer properties. To improve the production of lycopene by increasing the IPP flux in an engineered E. coli, the dxs gene was overexpressed and enhanced lycopene production was obtained [45]. In another example, the native promoters of DXP pathway genes in the E. coli chromosome were replaced with the strong bacteriophage T5 promoter (PT5), and the increase in isoprenoid precursors resulted in improved b-carotene production (with a titer of 6 mg/g dry cell weight) [44].


261

Strain improvement was also achieved through introduction of the heterologous pathway of IPP biosynthesis into E. coli. When the MVA pathway genes from Saccharomyces cerevisiae along with amorpha-4,11-diene synthase (ADS) gene were overexpressed in E. coli, high levels of amorphadiene, a precursor of artemisinin, was produced [47]. Further investigation identified a new rate-limiting step in this heterologous pathway, the reduction of 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA), which is catalyzed by HMGR. When the E. coli strain was further engineered to increase the expression of HMGR, the level of mevalonate was elevated, resulting in improved isoprenoid production [100].

12.5.3 Metabolic Engineering for Artemisinine Development and Production Artemisinin is a sesquiterpene, which is originally isolated from the aerial part of Artemisia annua (sweet wormwood, or ‘Qinghao’). Artemisinin is derived from an ancient Chinese herbal remedy and, at present, it is used as an antimalarial drug throughout the world [87]. Malaria is a vector-borne (mainly mosquito) infectious disease caused by protozoan parasites. By estimation, in 2002 only, it caused disease in 300 to 660 million people and 1 to 3 million were killed [101]. Although malaria could be treated with the quinine class of drugs, there are occurrences of drug-resistant protozoan parasite strains, which are sensitive to treatment with artemisinin. Thus, artemisinin-based therapy (ACTs) was recommended by the World Health Organization [88]. The entire biosynthesis pathway of artemisinin has not been elucidated yet. The first committed step is conversion of FPP to amorphadiene via the cyclization catalyzed by ADS [102] followed by further oxidations of amorphadiene to artemisinic acid. Artemisinic acid can be used as a precursor for semi-synthesis of artemisinin and related chemicals [88]. The production of artemisinin precursor like amorphadiene with a high titer was obtained via engineered E. coli by introduction of the exogenous mevalonate pathway from Saccharomyces cerevisiae as well as ADS gene, as described above [47]. However, generation of artemisinin faces a big challenge in the functional expression of plant cytochrome P450s. Plant P450s are membrane-bound heme-centered monooxygenases involved in a wide range of reactions, such as epoxidation, hydroxylation, dehydrogenation, isomerization, CC bond cleavage, and demethylation. P450s require cytochrome P450 reductases (CPRs) for electron transfer. P450s are one of the most important classes of enzymes acting on the biochemical transformations of terpenes in plants [103,104]. The failure of P450 expression in E. coli could be due to (1) lack of CPRs required for interaction with P450s, (2) the incompatibility of plant transmembrane protein translation with the apparatus in E. coli, and (3) the problems in protein posttranslational modification [88,105,106]. In spite of these challenges, the first example of high-level production of functionalized terpenoids in E. coli was accomplished recently [105]. Amorphadiene oxidase (AMO) oxidizes amorphadiene to artemisinic acid in three steps to produce the semi-synthetic precursor of artemisinin. Using this plant-derived P450, as well as a CPR from the same native host as AMO, by N-terminal transmembrane engineering along with codon optimization, an engineered E. coli was generated to produce high levels of artemisinic acid (>100 mg L1) [105]. The functionalized artemisinin precursor, artemisinic acid, was also produced in yeast via metabolic engineering. The amorphadiene synthase gene, along with a novel cytochrome P450 monooxygenase (CYP71AV1) gene from Artemisia annua that performs a three-step oxidation of amorpha-4,11-diene to artemisinic acid, was introduced in the engineered Saccharomyces

262


cerevisiae with a modified MVA pathway. This leads to the production of artemisinic acid with high titers up to 100 mg L1 [107].

12.5.4 Metabolic Engineering for Carotenoid Production Carotenoids are tetraterpenoids, which are synthesized as hydrocarbons (carotenes, such as lycopene, a-carotene, and b-carotene) or oxygenated derivatives (xanthophylls, such as lutein, a-cryptoxanthin and b-cryptoxanthin, zeaxanthin, canthaxanthin, and astaxanthin) by plants, fungi, algae, and bacteria [108]. The natural carotenoids are able to protect cells against reactive oxygen species damage in photosynthesis [45,109]. Besides their natural biological functions, carotenoids exhibit roles of anticancer, antioxidant, and enhancing immune responses. Thus, they are used as nutrient supplements, pharmaceuticals, and food colorants [45,109]. Because most natural carotenoids are present at very low abundance and are difficult to purify, metabolic engineering provides a powerful alternative, and various carotenoids, such as lycopene, b-carotene, canthaxanthin, zeaxanthin, torulene, neurosporaxanthin, and astaxanthin, have been successfully synthesized in non-carotenogenic microbes, such as E. coli, Saccharomyces cerevisiae, and Neurospora crassa. This research is summarized in several of reviews [65,108,110].

12.5.5 Metabolic Engineering for Taxol Development and Production Paclitaxel (Taxol as trade marked by Bristol-Myers Squibb) (Figure 12.4) is a taxane diterpenoid with a polycyclic ring [111]. As a chemotherapeutical reagent, paclitaxel is used to treat patients with lung, ovarian, and breast cancer, as well as head and neck cancer [112]. Paclitaxel exterts its effect by specifically binding to a beta-tubulin subunit of a microtubule and stabilizing the microtubules, which results in the death of the cell by disrupting the normal microtubule dynamics that are required for cell division [112]. Taxol was originally isolated from the bark of the slow-growing Pacific yew tree, Taxus brevifolia, [113], but is also produced in other species of the gymnosperm Taxus genus. Paclitaxel contains an oxetane, a C-13 phenylisoserine-derived side chain, and ester groups at C-2 and C-4, all of which were thought to contribute to the observed anticancer activity. However, further research underscored that the four-membered ring oxetane may not be necessary for Taxol bioactivity [87,114]. O

R2O

OH

O 10

R1

R2

Ph t-BuO

Ac H

R1

NH

7

O 5

Taxol Taxotere

1 2

1

2'

Ph

O

3'

OH

4

2

O

H

13

OH

O2CPh

OAc

Figure 12.4 The molecular structure of a taxane [114]. Taxotere is the semi-synthetic congener of Taxol. (Reproduced by permission from Macmillan Publishers Ltd: Susan C. Roberts. Production and engineering of terpenoids in plant cell culture. Nature Chemical Biology 3 (7): 387–395. London: Nature Publishing Group. Ó 2007 Macmillan)

263


12.5.5.1 Biosynthesis of Taxol In the first step, the universal diterpenoid precursor GGPP is cyclized into taxadiene, (taxa-4 (5),11(12)-diene), the first committed intermediate in the biosynthesis of Taxol in plants. This reaction is catalyzed by taxadiene synthase (TS), which is localized in chloroplast by the direction of its transit peptide [115]. The GGPP precursor for this enzyme is synthesized from IPP isomer that is provided by the DXP pathway, which is active in plastids as described above [116]. The subsequent biosynthetic processes (Figure 12.5) proceed through an array of reactions that include hydroxylations, acylations, oxidation and formation of the oxetane ring, conjugation of the side chain to a taxane intermediate (like 10-deacetylbaccatin III), and final modification reactions to generate paclitaxel [89]. It is estimated that the biosynthesis of paclitaxel, from the universal diterpenoid precursor GGPP, involves at least 20 distinct enzymatic steps with a similar number of taxoid intermediates [117]. Many genes encoding enzymes involved in this pathway have been cloned and characterized, such as Cytochrome P450, Taxane 5a-hydroxylase [118], Taxa-4(20),11(12)-dien-5a-ol-O-acetyltransferase [119], Taxane 2a-O-benzoyltransferase [120], 10-Deacetylbaccatin III-10-O-acetyltransferase [121], Baccatin III:3-amino-3-phenylpropanoyltransferase [122], and so on.

TS

DMAPP GGPPS + IPP

MEP Pathway

OPP

H Taxa-4(5),11(12)-diene P450s Acyltransferases

Ph

O OH

AcO

O NH

O

Ph

O OH

R1O

O

H

OH OH

OAc OH OBz Paclitaxel

HO

H OH OH

CoA

OR1

Taxane Intermediate R1 = H or Ac R2 = H or Bz O

NH2 O OH

Oxidation, Oxetane Ring Formation

CoA

NH2 α-Phenylalanine

AcO

β-Phenylalanoyl-CoA

HO

Figure 12.5

O OH

O H OAc OH OBz Baccatin III

The biosynthetic pathway for paclitaxel and related taxanes [89]

264


12.5.5.2 Metabolic Engineering of Taxol in Plant and Plant Cell Culture Currently, paclitaxel is supplied by two sources. One is via the semi-synthesis of a paclitaxel precursor, 10-deacetylbaccatin III, which is isolated from yew needles [111]. The other is by plant cell culture. Without harvesting the sparsely distributed yew trees, Taxus cell cultures enable the production of paclitaxel and its precursors by an environmentally beneficial, rapidly scalable, and cost-effective approach. For example, the raw material for paclitaxel production for Bristol-Myers Squibb, one of paclitaxels producers, mainly comes from the cultured plant cell made by Phyton, a biotechnology company (http://www.phytonbiotech.com/). Thus far, a number of Taxus cell cultures have been developed from different Taxus species [123]. To enhance paclitaxel production in Taxus cell suspension culture, different secondary metabolism elicitors were applied, including methyl jasmonate [124], fungal derivative [125], and salicylate derivative [126]. The expression profile of the paclitaxel biosynthesis genes in cell cultures of Taxus cuspidata, under the elicitation by methyl jasmonate, was characterized [111], which indicated a tight correlation between taxane accumulation and certain genes expression, providing direction for future efforts in metabolic engineering of paclitaxel biosynthesis. To date, metabolic engineering of the Taxus cell, probably the most promising approach to improve paclitaxel production, has yet to be achieved. Recently, cells of both Taxus cuspidata and Taxus chinensis in the liquid culture were successfully transformed with mediation by Agrobacterium rhizogenes, and transgenic Taxus cuspidata cell lines have been stably maintained in culture for more than 20 months [127]. The transformation method developed makes it possible for metabolic engineering of Taxus cells with key regulatory genes. Alternatively, the paclitaxel synthesis was heterozygously engineered in angiosperm plant hosts, Arabidopsis thaliana [128], and tomato fruit [129]. The Taxus baccata gene encoding enzyme TS was transformed into Arabidopsis thaliana. As motioned above, TS catalyzes the first committed step in Taxol biosynthesis, the conversion of taxadiene from GGPP. Constitutive production of TS enzyme in Arabidopsis thaliana led to the accumulation of taxadiene, concomitant growth retardation, and decreased levels of endogenous plastid isoprenoid products, such as carotenoids and chlorophylls. This suggests that the constitutive production of an active TS enzyme might switch the flux of the GGPP to other pathways. Taxadiene synthase was also expressed in a yellow-fruited tomato line that lacks the ability to utilize GGPP for carotenoids synthesis, which is normally used almost exclusively for the production of colored carotenoids in tomato fruits. In this transgenic tomato fruit, the flux of GGPP is diverged to the taxadiene synthesis pathway from the carotenoids pathway, which results in the extraction of 160 mg of highly pure taxadiene from 1 kg of freeze-dried fruit, a much higher yield than obtained by transgenic Arabidopsis thaliana [129]. 12.5.5.3 Metabolic Engineering of Taxol Production in E. coli and Yeast Although progress has been made in the production of Taxol via plant cell fermentation, this technology still has a variety of disadvantages. One of the big challenges is the variability of the product accumulation among cell lines, within cell aggregates, and over the course of the cell culture [89]. In addition, compared with microbes, plant cells grow very slowly in suspension culture. For example, for the maximum accumulation of paclitaxel and baccatin III, it takes 24 days and 28 days to culture Taxus wallichiana (Himalayan Yew) cells in shake flasks and in a


265

20 L airlift bioreactor respectively [130]. The slow growth rate of plant cells leads to a high possibility of contamination during the cell culture, thereby increasing the production cost. Thus, efforts have been made towards alternative hosts, like engineering microbes to produce Taxol. As the first step towards this goal, taxadiene, the first committed intermediate of paclitaxel biosynthesis, was produced in an engineered E. coli strain at a titer of 1.3 mg L1 [131]. Taxadiene is converted from GGPP, the common precursor of various diterpenes, by the action of TS in plants. Although E. coli itself is able to synthesize GGPP by the DXP pathway, the production is at a relatively low level [92]. Therefore, to increase the substrate supply for TS, a series of genes involved biosynthesis of GGPP were overexpressed in a single E. coli strain together with the gene encoding TS. These genes include those encoding DXP synthase (Dxs), IPP isomerase (IspH), and GGDP synthase (GGPPS) [131]. In addition, yeast cells were also engineered to reconstruct taxoids biosynthesis. Genes encoding enzymes acting on five steps leading to taxadien-5a-acetoxy-10b-ol were transformed into a single yeast stain [132]. These enzymes include GGPPS, TS, the taxadienol 5a-O-acetyl transferase, and two of them are involved in taxadiene cytochrome P450 hydroxylation. It was found that taxadiene was produced in the engineered yeast line at a titer of about 0.5 mg taxadiene/g dry weight cells. The acetylation product taxadien-5a-ol was also detected in very small amounts. However, product resulting from CYP450 hydroxylation, taxadien-5a-yl acetate, could not be measured [132].

12.6 Conclusions Metabolic engineering has been proven to be a powerful tool for the development and manufacture of pharmaceuticals that are derived from natural products, especially the secondary metabolites, such as polyketides and isoprenoids. Metabolic network analysis, along with the ‘omics’ techniques, enables the identification of the rational target genes or pathways for genetic manipulations, which result in the construction of engineered strains with improved production of interesting metabolites. Combining with other technologies, such as enzyme engineering, combinatorial biosynthesis, directed evolution, and bioprocess optimization, metabolic engineering will allow the development of more novel pharmaceuticals, as well as the manufacturing of pharmaceuticals with much lower cost.

References [1] Strohl, W.R. (2000) The role of natural products in a modern drug discovery program. Drug Discovery Today, 5, 39–41. [2] Newman, D.J. and Cragg, G.M. (2007) Natural products as sources of new drugs over the last 25 years. Journal of Natural Products, 70, 461–477. [3] Zhang, L. and Demain, A.L. (2005) Natural Products, Drug Discovery and Therapeutic Medicine, The Humana Press, Totowa, NJ. [4] Nielsen, J. (2001) Metabolic engineering. Applied Microbiology and Biotechnology, 55, 263–283. [5] Mashego, M.R., Rumbold, K., De Mey, M. et al. (2006) Microbial metabolomics: past, present and future methodologies. Biotechnology Letters, 29, 1–16. [6] Khosla, C. and Keasling, J.D. (2003) Metabolic engineering for drug discovery and development. Nature Reviews Drug Discovery, 2, 1019–1025.

266


[7] Stephanopoulos, G. (1999) Metabolic fluxes and metabolic engineering. Metabolic Engineering, 1, 1–11. [8] Raab, R.M., Tyo, K. and Stephanopoulos, G. (2005) Metabolic engineering. Advances in Biochemical Engineering/Biotechnology, 100, 1–17. [9] Vemuri, G.N. and Aristidou, A.A. (2005) Metabolic engineering in the -omics era: elucidating and modulating regulatory networks. Microbiology and Molecular Biology Reviews, 69, 197–216. [10] Iwatani, S., Yamada, Y. and Usuda, Y. (2008) Metabolic flux analysis in biotechnology processes. Biotechnology Letters, 30, 791–799. [11] Kim, H.U., Kim, T.Y. and Lee, S.Y. (2008) Metabolic flux analysis and metabolic engineering of microorganisms. Molecular Biosystems, 4, 113–120. [12] Kleijn, R.J., Liu, F., Winden, W.A.v. et al. (2007) Cytosolic NADPH metabolism in penicillin-G producing and non-producing chemostat cultures of Penicillium chrysogenum. Metabolic Engineering, 9, 112–123. [13] Bro, C. and Nielsen, J. (2004) Impact of ‘ome’ analyses on inverse metabolic engineering. Metabolic Engineering, 6, 204–211. [14] Bailey, J.E., Sburlati, A., Hatzimanikatis, V. et al. (2002) Inverse metabolic engineering: a strategy for directed genetic engineering of useful phenotypes. Biotechnology and Bioengineering, 79, 568–579. [15] Jaluria, P., Betenbaugh, M., Konstantopoulos, K. and Shiloach, J. (2007) Enhancement of cell proliferation in various mammalian cell lines by gene insertion of a cyclin-dependent kinase homolog. BMC Biotechnology, 7, 71. [16] Jaluria, P., Betenbaugh, M., Konstantopoulos, K. et al. (2006) Application of microarrays to identify and characterize genes involved in attachment dependence in HeLa cells. Metabolic Engineering, 9, 241–251. [17] Askenazi, M., Driggers, E.M., Holtzman, D.A. et al. (2003) Integrating transcriptional and metabolite profiles to direct the engineering of lovastatin-producing fungal strains. Nature Biotechnology, 21, 150–156. [18] Pennington, S.R., Wilkins, M.R., Hochstrasser, D.F. and Dunn, M.J. (1999) Proteome analysis: from protein characterization to biological function. Trends in Cell Biology, 7, 168–173. [19] Park, S.J., Lee, S.Y., Cho, J. et al. (2005) Global physiological understanding and metabolic engineering of microorganisms based on omics studies. Applied Microbiology and Biotechnology, 68, 567–579. [20] Nie, L., Wu, G., Culley, D.E. et al. (2007) Integrative analysis of transcriptomic and proteomic data: challenges. Solutions and Applications. Critical Reviews in Biotechnology, 27, 63–75. [21] Han, M.-J., Jeong, K.J., Yoo, J.-S. and Lee, S.Y. (2003) Engineering Escherichia coli for increased productivity of serine-rich proteins based on proteome profiling. Applied and Environmental Microbiology, 69, 5772–5781. [22] Bl€ uher, S. and Mantzoros, C.S. (2007) Leptin in reproduction. Current Opinion in Endocrinology, Diabetes and Obesity, 14, 458–464. [23] Nauck, M.A. and Meier, J.J. (2005) Glucagon-like peptide 1 and its derivatives in the treatment of diabetes. Regulatory Peptides, 128, 135–148. [24] Gutniak, M.K., Svartberg, J., Hellstrom, P.M. et al. (2001) Antidiabetogenic action of glucagon-like peptide-1 related to administration relative to meal intake in subjects with type 2 diabetes. Journal of Internal Medicine, 250, 81–87. [25] Lee, D.-H., Kim, S.-G., Park, Y.-C. et al. (2007) Proteome analysis of recombinant Escherichia coli producing human glucagon-like peptide-1. Analytical Tools for Proteomics, 849, 323–330. [26] Mashego, M.R., Rumbold, K., Mey, M.D. et al. (2006) Microbial metabolomics: past, present and future methodologies. Biotechnology Letters, 29, 1–16. [27] Oldiges, M., L€ utz, S., Pflug, S. et al. (2007) Metabolomics: current state and evolving methodologies and tools. Applied Microbiology and Biotechnology, 76, 495–511. [28] Oldiges, M., Kunze, M., Degenring, D. et al. (2004) Stimulation, monitoring, and analysis of pathway dynamics by metabolic profiling in the aromatic amino acid pathway. Biotechnology Progress, 20, 1623–1633. [29] Ketchum, R.E.B., Rithner, C.D., Qiu, D. et al. (2003) Taxus metabolomics: methyl jasmonate preferentially induces production of taxoids oxygenated at C-13 in Taxus x media cell cultures. Plant Metabolomics, 62, 901–909. [30] Friehs, K. (2004) Plasmid copy number and plasmid stability. Advances in Biochemical Engineering/Biotechnology, 86, 47–82. [31] Antoine, R. and Locht, C. (1992) Isolation and molecular characterization of a novel broad-host-range plasmid from Bordetella bronchiseptica with sequence similarities to plasmids from Gram-positive organisms. Molecular Microbiology, 6, 1785–1799.


267

[32] Ewering, C., Lutke-Eversloh, T., Luftmann, H. and Steinbuchel, A. (2002) Identification of novel sulfurcontaining bacterial polyesters: biosynthesis of poly(3-hydroxy-S-propyl-v-thioalkanoates) containing thioether linkages in the side chains. Microbiology (Reading, England), 148, 1397–1406. [33] Overhage, J., Steinbuchel, A. and Priefert, H. (2002) Biotransformation of eugenol to ferulic acid by a recombinant strain of Ralstonia eutropha H16. Applied and Environmental Microbiology, 68, 4315–4321. [34] Tao, L., Jackson, R.E. and Cheng, Q. (2005) Directed evolution of copy number of a broad host range plasmid for metabolic engineering. Evolutionary Engineering, 7, 10–17. [35] Nacken, V., Achstetter, T. and Degryse, E. (1996) Probing the limits of expression levels by varying promoter strength and plasmid copy number in Saccharomyces cerevisiae. Gene, 175, 253–260. [36] M€ uller, J., Dijl, J.V., Venema, G. and Bron, S. (1996) Cloning of heterologous genes specifying detrimental proteins on pUC-derived plasmids in Escherichia coli. Molecular Genetics and Genomics, 252, 207–211. [37] Jones, K.L., Kim, S.-W. and Keasling, J.D. (2000) Low-copy plasmids can perform as well as or better than highcopy plasmids for metabolic engineering of bacteria. Metabolic Engineering, 2, 328–338. [38] Yazdani, S. and Mukherjee, K. (2004) Continuous-culture studies on the stability and expression of recombinant streptokinase in Escherichia coli. Bioprocess and Biosystems Engineering, 24, 341–346. [39] Peredelchuk, M.Y. and Bennett, G.N. (1997) A method for construction of E. coli strains with multiple DNA insertions in the chromosome. Gene, 187, 231–238. [40] Boyd, D., Weiss, D.S., Chen, J.C. and Beckwith, J. (2000) Towards single-copy gene expression systems making gene cloning physiologically relevant: lambda InCh, a simple Escherichia coli plasmid–chromosome shuttle system. Journal of Bacteriology, 182, 842–847. [41] Yu, D., Ellis, H.M., Lee, E.C. et al. (2000) An efficient recombination system for chromosome engineering in Escherichia coli. Proceedings of the National Academy of Sciences, 97, 5978–5983. [42] Wang, Y. and Pfeifer, B.A. (2007) 6-Deoxyerythronolide B production through chromosomal localization of the deoxyerythronolide B synthase genes in E. coli. Metabolic Engineering, 10, 33–38. [43] Keasling, J.D. (1999) Gene-expression tools for the metabolic engineering of bacteria. Trends in Biotechnology, 17, 452–460. [44] Yuan, L.Z., Rouviere, P.E., LaRossa, R.A. and Suh, W. (2006) Chromosomal promoter replacement of the isoprenoid pathway for enhancing carotenoid production in E. coli. Metabolic Engineering, 8, 79–90. [45] Kim, S.-W. and Keasling, J.D. (2001) Metabolic engineering of the nonmevalonate isopentenyl diphosphate synthesis pathway in Escherichia coli enhances lycopene production. Biotechnology and Bioengineering, 72, 408–415. [46] Wang, G.-Y. and Keasling, J.D. (2002) Amplification of HMG-CoA reductase production enhances carotenoid accumulation in Neurospora crassa. Metabolic Engineering, 4, 193–201. [47] Martin, V.J.J., Pitera, D.J., Withers, S.T. et al. (2003) Engineering a mevalonate pathway in Escherichia coli for production of terpenoids. Nature Biotechnology, 21, 796–802. [48] McDaniel, R., Licari, P. and Khosla, C. (2001) Process development and metabolic engineering for the overproduction of natural and unnatural polyketides in Metabolic Engineering, vol. 73 (ed. l. Nielsen), Springer, Berlin, pp. 31–52. [49] Kealey, J.T., Liu, L., Santi, D.V. et al. (1998) Production of a polyketide natural product in nonpolyketideproducing prokaryotic and eukaryotic hosts. Proceedings of the National Academy of Sciences of the United States of America, 95, 505–509. [50] Khosla, C.Z.R. (1996) Generation of polyketide libraries via combinatorial biosynthesis. Trends in Biotechnology, 14, 335–341. [51] Reeves, A.R., Cernota, W.H., Brikun, I.A. et al. (2004) Engineering precursor flow for increased erythromycin production in Aeromicrobium erythreum. Metabolic Engineering, 6, 300–312. [52] Reeves, A.R., Brikun, I.A., Cernota, W.H. et al. (2006) Effects of methylmalonyl-CoA mutase gene knockouts on erythromycin production in carbohydrate-based and oil-based fermentations of Saccharopolyspora erythraea. Journal of Industrial Microbiology and Biotechnology, 33, 600–609. [53] Oliynyk, M., Samborskyy, M., Lester, J.B. et al. (2007) Complete genome sequence of the erythromycinproducing bacterium Saccharopolyspora erythraea NRRL23338. Nature Biotechnology, 25, 447–453. [54] Mutka, S.C., Bondi, S.M., Carney, J.R. et al. (2006) Metabolic pathway engineering for complex polyketide biosynthesis in Saccharomyces cerevisiae. FEMS Yeast Research, 6, 40–47. [55] Donadio, S., Staver, M.J., McAlpine, J.B. et al. (1991) Modular organization of genes required for complex polyketide biosynthesis. Science, 252, 675–679.

268


[56] Nielsen, J. (1998) The role of metabolic engineering in the production of secondary metabolites. Current Opinion in Microbiology, 1, 330–336. [57] Pohl, N.L. (2002) Nonnatural substrates for polyketide synthases and their associated modifying enzymes. Current Opinion in Chemical Biology, 6, 773–778. [58] Pfeifer, B.A., Admiraal, S.J., Gramajo, H. et al. (2001) Biosynthesis of complex polyketides in a metabolically engineered strain of E. coli. Science, 291, 1790–1792. [59] Pfeifer, B., Hu, Z., Licari, P. and Khosla, C. (2002) Process and metabolic strategies for improved production of Escherichia coli-derived 6-deoxyerythronolide B. Applied and Environmental Microbiology, 68, 3287–3292. [60] Lau, J., Tran, C., Licari, P. and Galazzo, J. (2004) Development of a high cell-density fed-batch bioprocess for the heterologous production of 6-deoxyerythronolide B in Escherichia coli. Journal of Biotechnology, 110, 95–103. [61] Murli, S., Kennedy, J., Dayem, L.C. et al. (2003) Metabolic engineering of Escherichia coli for improved 6-deoxyerythronolide B production. Journal of Industrial Microbiology and Biotechnology, 30 500–509. [62] Avignone Rossa, C., White, J., Kuiper, A. et al. (2002) Carbon flux distribution in antibiotic-producing chemostat cultures of Streptomyces lividans. Metabolic engineering, 4, 138–150. [63] Takano, E., Gramajo, H.C., Strauch, E. et al. (1992) Transcriptional regulation of the redD transcriptional activator gene accounts for growth-phase-dependent production of the antibiotic undecylprodigiosin in Streptomyces coelicolor A3(2). Molecular Microbiology, 6, 2797–2804. [64] Butler, M.J., Bruheim, P., Jovetic, S. et al. (2002) Engineering of primary carbon metabolism for improved antibiotic production in Streptomyces lividans. Applied and Environmental Microbiology, 68, 4731–4739. [65] Mijts, B.N. and Schmidt-Dannert, C. (2003) Engineering of secondary metabolite pathways. Current Opinion in Biotechnology, 14, 597–602. [66] Yoon, Y.J., Beck, B.J., Kim, B.S. et al. (2002) Generation of multiple bioactive macrolides by hybrid modular polyketide synthases in Streptomyces venezuelae. Chemistry & Biology, 9, 203–214. [67] Hutchinson, C.R. (1998) Combinatorial biosynthesis for new drug discovery. Current Opinion in Microbiology, 1, 319–329. [68] Menzella, H.G., Reid, R., Carney, J.R. et al. (2005) Combinatorial polyketide biosynthesis by de novo design and rearrangement of modular polyketide synthase genes. Nature Biotechnology, 23, 1171–1176. [69] Jacobsen, J.R., Hutchinson, C.R., Cane, D.E. and Khosla, C. (1997) Precursor-directed biosynthesis of erythromycin analogs by an engineered polyketide synthase. Science, 277, 367–369. [70] Ward, S.L., Desai, R.P., Hu, Z. et al. (2007) Precursor-directed biosynthesis of 6-deoxyerythronolide B analogues is improved by removal of the initial catalytic sites of the polyketide synthase. Journal of Industrial Microbiology and Biotechnology, 34, 9–15. [71] Thykaer, J. and Nielsen, J. (2003) Metabolic engineering of b-lactam production. Metabolic Engineering, 5, 56–69. [72] Nielsen, J. (1998) The role of metabolic engineering in the production of secondary metabolites. Current Opinion in Microbiology, 1, 330–336. [73] Spratt, B.G. (1994) Resistance of antibiotics mediated by target alterations. Science, 264, 388–393. [74] Reading, C. and Cole, M. (1977) Clavulanic acid: a beta-lactamase-inhibiting beta-lactam from Streptomyces clavuligerus. Antimicrobial Agents and Chemotherapy, 11, 852–857. [75] Demain, A.L. and Vaishnav, P. (2006) Involvement of nitrogen-containing compounds in b-lactam biosynthesis and its control. Critical Reviews in Biotechnology, 26, 67–82. [76] Elander, R.P. (2003) Industrial production of b-lactam antibiotics. Applied Microbiology & Biotechnology, 61, 385–392. [77] Tahlan, K., Park, H.U., Wong, A., Beatty, P.H. and Jensen, S.E. (2004) Two sets of paralogous genes encode the enzymes involved in the early stages of clavulanic acid and clavam metabolite biosynthesis in streptomyces clavuligenis. 48, 930–939. [78] Bushell, M.E., Kirk, S., Zhao, H.-J. and Avignone-Rossa, C.A. (2006) Manipulation of the physiology of clavulanic acid biosynthesis with the aid of metabolic flux analysis. Enzyme and Microbial Technology, 39, 149–157. [79] Ives, P.R. and Bushell, M.E. (1997) Manipulation of the physiology of clavulanic acid production in Streptomyces clavuligerus. Microbiology (Reading, England), 143, 3573–3579. [80] Li, R. and Townsend, C.A. (2006) Rational strain improvement for enhanced clavulanic acid production by genetic engineering of the glycolytic pathway in Streptomyces clavuligerus. Metabolic Engineering, 8, 240–252.


269

[81] Aidoo, K.A., Wong, A., Alexander, D.C. et al. (1994) Cloning, sequencing and disruption of a gene from Streptomyces clavuligerus involved in clavulanic acid biosynthesis. Gene, 147, 41–46. [82] Jensen, S.E., Elder, K.J., Aidoo, K.A. and Paradkar, A.S., (2000) Enzymes catalyzing the early steps of clavulanic acid biosynthesis are encoded by two sets of paralogous genes in Streptomyces clavuligerus. Antimicrobial Agents and Chemotherapy, 44, 720–726. [83] Li, R., Khaleeli, N. and Townsend, C.A. (2000) Expansion of the clavulanic acid gene cluster: identification and in vivo functional analysis of three new genes required for biosynthesis of clavulanic acid by Streptomyces clavuligerus. The Journal of Bacteriology, 182, 4087–4095. [84] Jensen, S.E., Paradkar, A.S., Mosher, R.H. et al. (2004) Five additional genes are involved in clavulanic acid biosynthesis in Streptomyces clavuligerus. Antimicrobial Agents and Chemotherapy, 48, 192–202. [85] Perez-Llarena, F.J., Liras, P., Rodriguez-Garcia, A. and Martin, J.F. (1997) A regulatory gene (ccaR) required for cephamycin and clavulanic acid production in Streptomyces clavuligerus: amplification results in overproduction of both beta-lactam compounds. The Journal of Bacteriology, 179, 2053–2059. [86] Hung, T.V., Malla, S., Park, B.C. et al. (2007) Enhancement of clavulanic acid by replicative and integrative expression of ccaR and cas2 in Streptomyces clavuligerus NRRL3585. Journal of Microbiology and Biotechnology, 17, 1538–1545. [87] Wang, G., Tang, W. and Bidigare, R.R. (2005) Terpenoids as Therapeutic Drugs and Pharmaceutical (eds L. Zhang and A.L. Demain), The Humana Press, Totowa, NJ. [88] Chang, M.C.Y. and Keasling, J.D. (2006) Production of isoprenoid pharmaceuticals by engineered microbes. Nature Chemical Biology, 2, 674–681. [89] Roberts, S.C. (2007) Production and engineering of terpenoids in plant cell culture. Nature Chemical Biology, 3, 387–395. [90] Rohmer, M., Seemann, M., Horbach, S. et al. (1996) Glyceraldehyde 3-phosphate and pyruvate as precursors of isoprenic units in an alternative non-mevalonate pathway for terpenoid biosynthesis. Journal of the American Chemical Society, 118, 2564–2566. [91] Lange, B.M., Rujan, T., Martin, W. and Croteau, R. (2000) Isoprenoid biosynthesis: the evolution of two ancient and distinct pathways across genomes. Proceedings of the National Academy of Sciences of the United States of America, 97, 13172–13177. [92] Lois, L.M., Campos, N., Putra, S.R. et al. (1998) Cloning and characterization of a gene from Escherichia coli encoding a transketolase-like enzyme that catalyzes the synthesis of D-1-deoxyxylulose 5-phosphate, a common precursor for isoprenoid, thiamin, and pyridoxol biosynthesis. Proceedings of the National Academy of Sciences of the United States of America, 95, 2105–2110. [93] Rohdich, F., Zepeck, F., Adam, P. et al. (2003) The deoxyxylulose phosphate pathway of isoprenoid biosynthesis: studies on the mechanisms of the reactions catalyzed by IspG and IspH protein. Proceedings of the National Academy of Sciences of the United States of America, 100, 1586–1591. [94] Rodrıguez-Concepcioń, M., Campos, N., Lois, L.M. et al. (2000) Genetic evidence of branching in the isoprenoid pathway for the production of isopentenyl diphosphate and dimethylallyl diphosphate in Escherichia coli. FEBS Letters, 473, 328–332. [95] Rohdich, F., Wungsintaweekul, J., Fellermeier, M. et al. (1999) Cytidine 50 -triphosphate-dependent biosynthesis of isoprenoids: YgbP protein of Escherichia coli catalyzes the formation of 4-diphosphocytidyl-2-C-methylerythritol. Proceedings of the National Academy of Sciences of the United States of America, 96, 11758–11763. [96] Luttgen, H., Rohdich, F., Herz, S. et al. (2000) Biosynthesis of terpenoids: YchB protein of Escherichia coli phosphorylates the 2-hydroxy group of 4-diphosphocytidyl-2C-methyl-D-erythritol. Proceedings of the National Academy of Sciences of the United States of America, 97, 1062–1067. [97] Herz, S., Wungsintaweekul, J., Schuhr, C.A. et al. (2000) Biosynthesis of terpenoids: YgbB protein converts 4-diphosphocytidyl-2C-methyl-D-erythritol 2-phosphate to 2C-methyl-D-erythritol 2,4-cyclodiphosphate. Proceedings of the National Academy of Sciences of the United States of America, 97, 2486–2490. [98] Maury, J., Asadollahi, M.A., Møller, K. et al. (2005) Microbial isoprenoid production: an example of green chemistry through metabolic engineering. Advances in Biochemical Engineering/Biotechnology, 100, 19–51. [99] Withers, S.T. and Keasling, J.D. (2007) Biosynthesis and engineering of isoprenoid small molecules. Applied Microbiology and Biotechnology, 73, 980–990. [100] Pitera, D.J., Paddon, C.J., Newman, J.D. and Keasling, J.D. (2007) Balancing a heterologous mevalonate pathway for improved isoprenoid production in Escherichia coli. Metabolic Engineering, 9, 193–207.

270


[101] Snow, R.W., Guerra, C.A., Noor, A.M. et al. (2005) The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature, 434, 214–217. [102] Mercke, P., Bengtsson, M., Bouwmeester, H.J. et al. (2000) Molecular cloning, expression, and characterization of amorpha-4,11-diene synthase, a key enzyme of artemisinin biosynthesis in Artemisia annua L. Archives of Biochemistry and Biophysics, 381, 173–180. [103] Sono, M., Roach, M.P., Coulter, E.D. and Dawson, J.H. (1996) Heme-containing oxygenases. Chemical Reviews, 96, 2841–2888. [104] Chapple, C. (1998) Molecular-genetic analysis of plant cytochrome P450-dependent monooxygenases. Annual Review of Plant Physiology and Plant Molecular Biology, 49, 311–343. [105] Chang, M.C.Y., Eachus, R.A., Trieu, W. et al. (2007) Engineering Escherichia coli for production of functionalized terpenoids using plant P450s. Nature Chemical Biology, 3, 274–277. [106] Leonard, E. and Koffas, M.A. (2007) Engineering of artificial plant cytochrome P450 enzymes for synthesis of isoflavones by Escherichia coli. Applied and Environmental Microbiology, 73, 7246–7251. [107] Ro, D.-K., Paradise, E.M., Ouellet, M. et al. (2006) Production of the antimalarial drug precursor artemisinic acid in engineered yeast. Nature, 440, 940–943. [108] Das, A., Yoon, S.-H., Lee, S.-H. et al. (2007) An update on microbial carotenoid production: application of recent metabolic engineering tools. Applied Microbiology and Biotechnology, 77, 505–512. [109] Demmig-Adams, B. and Adams, W.W. III (2002) Antioxidants in photosynthesis and human nutrition. Science, 298, 2149–2153. [110] Klein-Marcuschamera, D., Ajikumara, P.K. and Stephanopoulos, G. (2007) Engineering microbial cell factories for biosynthesis of isoprenoid molecules: beyond lycopene. Trends in Biotechnology, 25, 417–424. [111] Nimsa, E., Duboisb, C.P., Robertsa, S.C. and Walker, E.L. (2006) Expression profiling of genes involved in paclitaxel biosynthesis for targeted metabolic engineering. Metabolic Engineering, 8, 385–394. [112] Rowinsky, E.K. and Donehower, R.C. (1995) Paclitaxel (Taxol). The New England Journal of Medicine, 332, 1004–1014. [113] Wani, M.C., Taylor, H.L., Wall, M.E. et al. (1971) Plant antitumor agents. VI. Isolation and structure of Taxol, a novel antileukemic and antitumor agent from Taxus brevifolia. Journal of the American Chemical Society, 93, 2325–2327. [114] Wang, M., Cornett, B., Nettles, J. et al. (2000) The oxetane ring in taxol. The Journal of Organic Chemistry, 65, 1059–1068. [115] Koepp, A.E., Hezari, M., Zajicek, J. et al. (1995) Cyclization of geranylgeranyl diphosphate to taxa-4 (5),11(12)diene is the committed step of Taxol biosynthesis in Pacific yew. Journal of Biological Chemistry, 270, 8686–8690. [116] Eisenreich, W., Menhard, B., Hylands, P.J. et al. (1996) Studies on the biosynthesis of Taxol: the taxane carbon skeleton is not of mevalonoid origin. Proceedings of the National Academy of Sciences of the United States of America, 93, 6431–6436. [117] Hezari, M. and Croteau, R. (1997) Taxol biosynthesis: an update. Planta Medica, 63, 291–295. [118] Jennewein, S., Long, R.M., Williams, R.M. and Croteau, R. (2004) Cytochrome P450 taxadiene 5a-hydroxylase, a mechanistically unusual monooxygenase catalyzing the first oxygenation step of Taxol biosynthesis. Chemistry & Biology, 11, 379–387. [119] Walker, K., Schoendorf, A. and Croteau, R. (2000) Molecular cloning of a taxa-4(20),11(12)-dien-5a-ol-Oacetyl transferase cDNA from Taxus and functional expression in Escherichia coli. Archives of Biochemistry and Biophysics, 374, 371–380. [120] Walker, K. and Croteau, R. (2000) Taxol biosynthesis: molecular cloning of a benzoyl-CoA:taxane 2a-Obenzoyltransferase cDNA from Taxus and functional expression in Escherichia coli. Proceedings of the National Academy of Sciences of the United States of America, 97, 13591–13596. [121] Walker, K. and Croteau, R. (2000) Molecular cloning of a 10-deacetylbaccatin III-10-O-acetyl transferase cDNA from Taxus and functional expression in Escherichia coli. Proceedings of the National Academy of Sciences of the United States of America, 97, 583–587. [122] Walker, K., Fujisaki, S., Long, R. and Croteau, R. (2002) Molecular cloning and heterologous expression of the C-13 phenylpropanoid side chain-CoA acyltransferase that functions in Taxol biosynthesis. Proceedings of the National Academy of Sciences of the United States of America, 99, 12715–12720. [123] Frense, D. (2007) Taxanes: perspectives for biotechnological production. Applied Microbiology and Biotechnology, 73, 1233–1240.


271

[124] Yukimune, Y., Tabata, H., Higashi, Y. and Hara, Y. (1996) Methyl jasmonate-induced overproduction of paclitaxel and baccatin III in Taxus cell suspension cultures. Nature Biotechnology, 14, 1129–1132. [125] Wang, C., Wu, J. and Mei, X. (2001) Enhancement of Taxol production and excretion in Taxus chinensis cell culture by fungal elicitation and medium renewal. Applied Microbiology and Biotechnology, 55, 404–410. [126] Qian, Z., Zhao, Z., Xu, Y. et al. (2006) Taxanes: perspectives for biotechnological production. Applied Microbiology and Biotechnology, 73, 1233–1240. [127] Ketchum, R.E.B., Wherland, L. and Croteau, R.B. (2007) Stable transformation and long-term maintenance of transgenic Taxus cell suspension cultures. Plant Cell Reports, 26, 1025–1033. ´ ., Sauret-G€ueto, S., Phillips, M.A. et al. (2004) Metabolic engineering of isoprenoid biosynthesis in [128] Besumbes, O Arabidopsis for the production of taxadiene, the first committed precursor of Taxol. Biotechnology and Bioengineering, 88, 168–175. [129] Kovacs, K., Zhang, L., Linforth, R.S.T. et al. (2007) Redirection of carotenoid metabolism for the efficient production of taxadiene [taxa-4(5),11(12)-diene] in transgenic tomato fruit. Transgenic Research, 16, 121–126. [130] Navia-Osorio, A., Garden, H., Cusido´, R.M. et al. (2002) Production of paclitaxel and baccatin III in a 20-L airlift bioreactor by a cell suspension of Taxus wallichiana. Planta Medica, 68, 336–340. [131] Huang, Q., Roessner, C.A., Croteau, R. and Scott, A.I. (2001) Engineering Escherichia coli for the synthesis of taxadiene, a key intermediate in the biosynthesis of Taxol. Bioorganic & Medicinal Chemistry, 9, 2237–2242. [132] Dejong, J.M., Liu, Y., Bollon, A.P. et al. (2006) Genetic engineering of taxol biosynthetic genes in Saccharomyces cerevisiae. Biotechnology and Bioengineering, 93, 212–224.

13 Multimodular Synthases and Supporting Enzymes for Chemical Production Michael Burkart1 and Junhua (Alex) Tao2 1

Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0358, USA 2 Elevance Renewable Sciences, 175 E. Crossroads Parkway, Bolingbrook, IL 60440, USA

13.1 Introduction Natural products remain an important source of new chemical entities for drug development. The majority of natural product pharmaceuticals are produced on a commercial scale through the culturing of producer organisms for fermentative production [1]. As our understanding of natural product biosynthesis progresses through the relatively new tools of modern genetics, opportunities for medicinal and process chemists have been gradually opened. These include concepts that begin with the use of enzymes as catalysts for chemical transformations [2] and extend to the intracellular, such as the genetic manipulation of natural product producers for the discovery and biosynthesis of new drug entities [3]. In this chapter, we will introduce an exciting class of natural product biosynthetic enzymes, the modular synthases, as well as their associated enzyme partners. We will discuss the use of metabolic engineering as a tool for small-molecule discovery and development, both through directed fermentation and combinatorial biosynthesis. In addition, we will review six classes of partner enzymes involved in the modification of polyketide (PK) and nonribosomal peptide (NRP) natural products. We believe that these enzymatic transformations hold great opportunities for synthetic chemists and will serve as the foundation for a new trend in both discovery and process chemistry.


274


13.2 Background 13.2.1 Multimodular Synthase Architecture Modular PKs and NRPs are constructed from respective acyl coenzyme A (CoA) and amino acid monomers through strikingly similar mechanisms [4]. These synthases are called ‘modular’ due to the nature of their structural organization [5]. In both PK and NRP systems, several large ‘megasynthase’ enzymes can participate in the biosnthesis of a single natural product. These synthases are subcategorized into modules, each of which loads and extends the growing oligomeric metabolite by one starter unit. Modules are subcategorized into domains, each of which performs a single catalytic role in the assembly-line organization. Repeated catalytic activity is seen from module to module, but differentiation exists as the growing natural products extend along these assembly lines, depending on the presence or absence of selective domains. As seen in Figure 13.1, the biosynthesis of deoxyerythromycin B is achieved by three synthases, seven modules, and 28 domains. The possible domains present in PK modules determine the outcome of the module catalysis. The process begins by posttranslational priming (conversion of apo-synthase to a holo-synthase) by the addition of a 40 - phosphopantetheinyl group to a conserved serine residue within each carrier protein domain. Activated acyl carrier protein domains display a terminal thiol that serves as the site for attachment of monomer units for the growing acyl or peptidyl chains.

Figure 13.1 PK synthase biosynthetic architecture. The erythromycin aglycone synthase (deoxyerythronolide B synthase) is shown, made up of three modules with 28 domains

In PK synthesis (Figure 13.2) [6–10], an acyltransferase catalyzes the acylation of activated carrier proteins through acyl-CoA monomers. Next, the acyl group of one of the carrier proteins is transferred to the ketosynthase, which catalyzes a Claisen-like condensation with downstream carrier protein to yield the elongated product and release CO2. The resulting ketide may be further modified, depending upon the presence or absence of additional domains. If present, the ketoreductase reduces the ketone to an alcohol. If present, the dehydratase domain will eliminate this alcohol to the olefin, resulting in an a,b-unsaturated thioester intermediate. Finally, if an enoylreductase is present, it will reduce this olefin to the fully staturated

Multimodular Synthases and Supporting Enzymes for Chemical Production

Figure 13.2

275

The biosynthetic scheme of PK synthase modules

methylene. This domain architecture determines the structural outcome of the finished natural product at each two-carbon unit. NRP synthases (NRPSs) contain a similar modular architecture (Figure 13.3), but use a different set of possible domains [11]. Similar to PK synthases, all NRP carrier proteins are similarly post-translationally modified by PPTases. Next (Figure 13.4), the loading of peptidyl carrier proteins involves an added degree of regiospecificity [8,12–14]. The adenylation domain first selects the cognate amino acid in an ATP-dependent loading and transfer to the holo-carrier protein. The resulting upstream carrier protein domain-loaded amino acid then attacks the downstream thioester through nucleophilic substitution catalyzed by the intervening condensation domain. The dipeptide product of this coupling may then be further modified, depending upon the presence or absence of additional domains. These include domains that

Figure 13.3 NRP synthase biosynthetic architecture. The tyrocidine C is shown, made up of three modules and 32 domains

Figure 13.4

The biosynthetic scheme of NRP synthase modules

276


catalyze epimerization, methyl transfer, oxidation, reduction, or cyclization. NRP synthesis is completed by repetitions of this process followed by an ultimate hydrolysis or macrocyclization. Additionally, many NRPS modules utilize nonproteinogenic amino acids as starter units recognized by adenylation domains for incorporation into growing natural products. In most cases, these unique amino acids must be biosynthesized by the producer organism. The genes that code enzymes necessary to make these precursors are often found contiguous to the modular synthases in the organismic genome and have been the subject of extensive research within the past two decades.

13.2.2 Natural Product Biosynthetic Cycle Up until the end of the 20th century, natural product biosynthesis was limited mostly to the phenotypic study of culturable organisms and screening of mutant strains. The elucidation of reaction pathways from a chemical perspective was mostly relegated to intellectual pursuit. It is largely microbiological techniques such as strain isolation and improvement that were responsible for the introduction of antibiotics and their influence on modern health. Not until the invention of the polymerase chain reaction (PCR) in 1985 were the genes involved in natural product biosynthesis able to be sequenced, cloned, and manipulated [15]. A revolution in the biological and medicinal sciences has occurred since the advent of PCR, and the study of natural product biosynthesis has blossomed accordingly. The list of fully sequenced organisms grows rapidly each year, and individual natural product biosynthetic pathways are sequenced and elucidated at an astonishing pace with a current rate of >50 molecules per year. Within these DNA sequences lies an inherent ability to manipulate small molecules in an exquisitely selective and specific manner with enzymes coded by the genes. The promise held within these sequences is the ability to produce complex small molecules selectively in vivo. To understand how the biosynthesis of natural products lends itself to small-molecule organic synthesis, we begin with the research cycle of natural product biosynthesis. First, natural molecules with biological activity are isolated and identified. This research has been ongoing for decades and is responsible for the discovery of all bioactive natural products today. Current research mainly focuses on marine organisms and involves organism collection, natural product isolation, bioassay screening, and structure elucidation. This field is a rapidly developing and vibrant course of research indispensable to the future discovery of novel molecular structure. Next is the elucidation of biosynthetic pathways from the producer organisms. This stage entails the isolation and sequencing of the biosynthetic genes involved in the natural biosynthesis of one molecule. Usually, in order to publish sequencing information, researchers must in addition definitively demonstrate activity of one enzyme in the biosynthetic cluster, either through knockout experiments which alter molecular structure or through in vitro proof of activity. Complicating this research is the abundance of nonculturable microorganisms producing highly interesting bioactive molecules. For instance, many natural products isolated from marine organisms such as sponges are believed to be produced by unculturable symbiotic bacteria living within the microorganism [16]. On a different but similar issue, many organismic strains producing a molecule of interest are often proprietary or carefully guarded by their discoverers.


277

Once a biosynthetic pathway has been fully sequenced, the activity, order, and timing of each enzyme in its pathway must be determined. Each gene product usually corresponds to an enzymatic step in the biosynthesis, and these must be determined and demonstrated. Here, one can often draw analogies from previously studied enzymes through protein sequence similarity, or homology, and parallels with other known secondary metabolism pathways. In many cases, each enzyme is produced individually and activity studies are performed in vitro to validate a proposed secondary metabolism pathway. Alternatively, the pathway may be studied genetically by generating mutants of the producer organism in which an individual gene has been inactivated, thereby producing pathway intermediates that may be correlated to the missing enzyme. Often, combinations of these techniques lead to a complete understanding of enzyme activity. A gene sequence within a given pathway does not necessarily correspond to sequential enzyme activity, and the order of events must also be correlated to enzyme function in order to fully understand metabolite construction. Particularly important to the pathways of modular synthases is the incorporation of novel precursors, including nonproteinogenic amino acids in NRP systems [17] and unique CoA thioesters in PK and fatty acid synthases [18]. These building blocks expand the primary metabolism and offer practically unlimited variability applied to natural products. Noteworthy within this context is the contiguous placement of biosynthetic genes for novel precursors within the biosynthetic gene cluster in prokaryotes. Such placement has allowed relatively facile elucidation of biosynthetic pathways and rapid discovery of novel enzyme mechanisms to create such unique building blocks. These new pathways offer a continued expansion of the enzymatic toolbox available for chemical catalysis.

13.3 Metabolic Engineering of Megasynthases Process-scale production of natural product derivatives has, to date, been limited to two techniques: semisynthesis and directed fermentation. Semisynthesis is the technique of using traditional chemical synthesis to modify the structure of a fermented natural product [19]. These processes have been essential to produce contemporary pharmaceuticals that have benefited from the structural optimization afforded by medicinal chemistry. An alternative method to produce natural product derivatives is precursor-directed fermentation, in which alternate precursors of natural product synthases are added to fermentation cultures as a means to force in vivo catalysis to adopt these analogs within a given biosynthetic pathway [20]. This technique depends heavily upon a thorough understanding of the natural product biosynthetic pathway in question. Within the context of modular synthases, both PK and NRP pathways have been engineered to accept alternate precursors, and many such examples are empirically determined. In the case of NRP systems, precursor specificity is governed by individual adenylation domains within each module. Given the great variability found in natural NRP monomers, permissivity of adenylation domains offers the intriguing potential to modify most NRP pathways. Additionally, it has been demonstrated that condensation domains also play a selective role in the acceptance of varied substrates. PK synthases, alternatively, have significantly more regulated biosyntheses with regard to monomer choice. In most cases, elongation monomers are restricted to malonate and methylmalonate starter units. The major source of variability within these synthases lies in

278


starter modules, in which a wide variety of acyl groups are known to be utilized by diverse synthases. Here, the starter unit identity is chosen by the acyltransferase module in the loading domain, which loads acyl groups from CoA thioesters. The incorporation of novel precursors within fermentative production of PKs, however, depends upon several variables. First is organismic take up of unique carboxylic acids introduced into the fermentation media. Once inside the cell, these acids must be converted into the CoA thioester through action of acyl CoA ligases. These enzymes are universally present in primary metabolism for activation of fatty acids, as well as a variety of secondary metabolic pathways. Finally, the acyltransferase domain of the PK synthase in question must accept the novel CoA thioester as a valid precursor for acyl carrier protein loading. Below, we introduce a novel example of directed fermentation through novel precursor take-up by a type I PK synthase, avermectin synthase, for commercial production.

13.3.1 Daptomycin: Metabolic Engineering by Domain Swap There are two basic schools of thought for the practice of metabolic engineering through genetic manipulation. The first involves engineering of an entire biosynthetic cluster into an organism more suitable to laboratory manipulation. This is only necessary if the natural producer is difficult to culture and ferment or if it expresses the natural product at low levels. From a microbiological standpoint, engineering the biosynthetic enzymes in an organism with known culture conditions and promoter elements is much more desirable than the enormous effort necessary to study and manipulate each producer organism individually. Take the example of epothilone B, produced by the myxobacterium Sorangium cellulosum, which grows very slowly, doubling only once every 16 h, and produces about 20 mg L1 of the metabolite [21]. Because the cost of fermenting the natural producer was prohibitive even to produce material for clinical trials, Julien and coworkers successfully engineered the entire epothilone cluster into both Streptomyces coelicolor and Myxococcus xanthus, both of which are better-studied organisms with faster growth and known genetic switches for increased protein expression [22,23]. The second and more exciting effort from a synthetic perspective is often referred to as combinatorial biosynthesis. Here, the design and biosynthesis of novel metabolites is achieved through selective construction of new pathways drawing from different biosynthetic systems [21]. For example, any bioactive natural products are of hybrid biomolecules, including NRP/PK (hybrid megasynthase products, e.g. epothilone), NRP/carbohydrate (glycopeptides, e.g. vancomycin), and PK/carbohydrate (glycoketides, e.g. daunomycin), to name just a few. An excellent example of metabolic engineering with hybrid subunits is the recent engineering of the deoxysugar pathway of oleandromycin by Aguirrezabalaga et al. [24]. The researchers produced novel elloramycin glycoketides through manipulation of modular deoxysugar biosynthetic pathways and a ‘sugar flexible’ glycosyltransferase (Gtf). This methodology could uncover novel molecules that are hybrids of metabolites with known metabolic pathways and, perhaps, display new bioactivities. Daptomycin (Figure 13.5) is a lipopeptide antibiotic produced by Streptomyces roseosporus with potent activity against Gram-positive infections. Originally discovered by Eli Lilly, daptomycin was developed and marketed by Cubist Pharmaceuticals and in 2003 was approved for clinical use under the trade name Cubicin as a topical antibiotic against Grampositive pathogens and for Staphylococcus aureus. A 13-membered depsipeptide lactone,

Multimodular Synthases and Supporting Enzymes for Chemical Production OH

OH

O

O

O

9

NH

8

N H

O H N B

O 7

O

N H

NH2

6

HN

O

10 5

O HO C

NH 11

H N

O

HN

12

NH

4

O

O

13

O

O

HN O O

HO

D

O

3

O HO NH2

O

NH2

O NH 2

HN O

O 1

N H

A Daptomycin

HN

Modifications Position A

Position C O H3C O H2N O

O

Position B

Position D CH3

HO

H3C CH3

H2N

H3C O

H2N HN

Figure 13.5

Daptomycin and engineered daptomycin analogs

279

280


daptomycin contains three nonproteinogenic amino acids: ornithine, kynurinine (Kyn), and 3-methyl-glutamic acid (3mGlu). The cyclic peptide is biosynthesized by three large NRPS enzymes and assisted by an additional 61 genes in the operon [25]. Significant interest has been placed upon metabolic engineering of the daptomycin gene cluster in an effort to modify the identity of the natural product to optimize activity and pharmacokinetic properties. While these efforts could be carried out by synthetic chemistry in order to determine structure– activity relationships (SARs) of possible analogs, several arguments favored the use of metabolic engineering as a discovery tool. First, the nonproteinogenic amino acids Kyn and 3mGlu are synthetically difficult to obtain, so the total synthesis of these analogs is not a trivial prospect. Second, while medicinal chemistry could identify structures with improved or different SAR profiles, obtaining these analogs from fermentation or semisynthetic methods would entail significant investments of time and resources. From the biosynthetic perspective, several natural product pathways had also been identified with highly homologous regions of their NRPS pathway, such that DNA shuffling between these pathways seemed fundamentally feasible. Finally, and perhaps most importantly, the daptomycin producer organism is amenable to genetic alteration, so these secondary metabolic pathways were accessible to modification. To this end, scientists at Cubist began genetic manipulation of the daptomycin synthase to incorporate domains excised from analogous synthases produced in other streptomycetes. First was modification of the second and third NRPS subunits, DptBC and DptD [26,27]. Here modules 8, 11, 12, and 13, which extend D-Ala, D-Ser, 3mGlu, and Kyn, were fully replaced by modules from analogous biosynthetic clusters from A54145 (Streptomyces fradiae) and calcium-dependent antibiotic (CDA, Streptomyces coelicolor) using lambda Red-mediated recombination [28]. Eighteen different constructs were obtained and studied for antibiotic activity against several Gram-negative and Gram-positive bacteria, with particular attention to Staphylococcus aureus. None of the engineered compounds showed significantly improved activity compared with daptomycin, although several were remarkably similar in both activity and spectrum to the parent compound. Additionally, directed fermentation was used to select for more homogenous daptomycin analogs, and special fermentative techniques were identified to incorporate decanoic acid into the culture medium [29]. Though an elegant example of metabolic engineering, the published examples demonstrated by the Cubist team cover only a small number of the thousands of possible variations that could be made within the daptomycin biosynthetic system, and an exhaustive study has the potential to identify new drug candidates [28]. These studies in daptomycin, however, offer a unique window into the possibilities for industrial applications of metabolic engineering. Here, the divisions between discovery and process chemistry are blurred, because the discovery process provides the bacterial strains necessary for fermentative scale-up. The next boundary to cross within this context is the adaptation of semisynthetic processes to the discovery process. Here, synthetic chemistry is used to further modify a fermentative product. By using the tools demonstrated in the daptomycin example, selectively reactive amino acids may be incorporated into positions within the peptide through fermentation, then synthetic chemistry may be used to functionalize these molecules further. Such a paradigm could exponentially expand the molecular space in terms of SAR studies compared with those performed by traditional drug discovery teams today.

281


13.3.2 Avermectin: Metabolic Engineering by Directed Fermentation Avermectins are a series of eight related natural products with potent anthelmintic and insecticidal activities produced by Gram-positive, filamentous soil bacteria Streptomyces avermitlis [30]. All of these molecules share a 16-membered macrocyclic PK aglycon with the attachment of a dimeric o-methyl-a-L-oleandrose to C-13 (Figure 13.6). Avermectins differ in three substituents R1, R2 and R3. R1 can be either a hydroxyl or hydrogen, in which case there is a double bond between the C-22 and C-23. R2 can be either methyl or ethyl, and R3 can be either hydrogen or methyl. OCH3 HO H3C

OCH3 O

O

H

H H3C O

O

OH

H

H

5

CH3

25

O

O

Figure 13.6

23

H 22 O

O 13

H H3C

R1

CH3

O

H

H

R2

H CH3

CH3

OR3

Chemical structures of avermectins

The avermectin gene cluster has 85 kb and 18 open reading frames within the Streptomyces avermitilis genome of about 9000 kb [31]. Earlier labeling studies showed that avermectins are derived from seven acetate, five propionate and a single sec-butyl or isopropyl group attached to C-25 [32,33]. Based on this and the organization of the biosynthesis gene cluster, the biosynthesis of the PK aglycon of avermectin A1a (R1 ¼ H, C22–C23 a double bond, R2 ¼ Methyl and R3 ¼ methyl) was proposed to start with the loading to 2-methyl butanoyl CoA to the PK synthase [31,34] (Figure 13.7, only ACP shown). After 12 chain extensions catalyzed by the PK synthase in a modular fashion, the final linear PK was synthesized as an attachment to the ACP in module 12, which is subsequently transferred to the thioesterase (TE) domain and cyclized to form the macrocyclic aglycon. Avermectin A1a was eventually generated after ketalization, benzofuran ring formation, and disaccharide decoration (not shown). By following the same mechanism, the loading of 2-methyl propionyl CoA to the same PK synthase leads to avermectins, where R2 ¼ methyl. Avermectins and their derivatives are widely used for broad-spectrum parasite control and for the treatment of river blindness. A combination of pathway engineering and mutagenesis has led to the discovery of efficient production strains and of novel avermectin analogs [35,36]. For example, by inactivating AveD, a gene encoding a methyltransferase, mutants of Streptomyces avermitilis lack of o-methyltransferase produces only avermectins with R3 ¼ H, which are biologically more active than those with R3 ¼ methyl. Furthermore, since the substituents at C-25 are dictated by starters loaded to the PKS, pathway engineering

S

S

O

ACP

OH

O

Module 1

ACP

Loading Domain

S

OH

O

Figure 13.7

ACP

Module 2

OH

23

O

OH

OH

O

O

O

H3C

O

O H3C

H

OCH3

Proposed biosynthesis of avermectin A1a

25

22

S

ACP TE

Module 12

HO

O

O 13

O

CH3

avermectin A1a

H H 3C

OCH3

O

H

H

O

O OH

H 22 O

CH3

H

O

23

CH3

H

H CH3

25

282 Biocatalysis for the Pharmaceutical Industry

283

Multimodular Synthases and Supporting Enzymes for Chemical Production O

O

OH

OH NH2

NH2 Branched-chain amino acid transaminase O

O OH

OH

O

O Branched-chain 2-keto acid dehydrogenase complex (bkd) O

O SCoA

OH

avermectin PK synthase

Figure 13.8

De novo synthesis of 2-methyl butanoyl CoA and 2-methyl propionyl CoA

could lead to the production of novel amermectin analogs. 2-Methyl butanoyl CoA and 2-methyl propionyl CoA are synthesized from isoleucine and valine respectively by amino acid transaminases followed by branched-chain a-keto acid dehydrogenase complex (bkd complex) (Figure 13.8). Indeed, mutants of Streptomyces avermitilis lacking the bkd complex allowed the production of novel avermectins with different substituents at C-25 by feeding with exogenous carboxylic acids. Among them, avermectins B1 and B2 (Figure 13.9) with a cyclohexyl group at C-25 showed the highest efficacy. In B1, there is a double bond between C-22 and C-23, while in B2 R1 ¼ OH. The B2:B1 product ratio is 1.6:1 [35,36]. The B1 analog of avermectins has the most effective antiparasitic activity and, therefore, it is desirous to improve the ratio of B1:B2. While it is proposed that differentiation of B1 and B2 are an early event in the biosynthesis, it was shown that the dehydratase domain in module 2, which is responsible for the B1 product, does not have a role in determining the ratio of B1:B2 [37]. In contrast, it was found that the aveC gene modulates the ratio of the two products [38]. Since the mechanism is unknown, directed evolution was applied for random mutagenesis of aveC. Mutants were then screened to identify improved aveC genes, which enhanced the ratio of B1:B2, by introduction into an aveC deletion strain of Streptomyces avermitilis. A production strain with B1:B2 ¼ 1:0.4 was created when a D48E and A89T double aveC mutant, which was created by error-prone PCR, was introduced into the chromosome [39]. Furthermore, gene shuffling of improved aveC genes led to a mutant giving an excellent fermentation titer of B1:B2 ¼ 1:0.07 [40]. In summary, a novel cyclohexyl avermectin analog B1 with enhanced antiparasitic activity was discovered and produced with high selectivity and excellent fermentation titer through deciphering biosynthesis, pathway engineering, and directed evolution. The new product, doramectin, is sold commercially as Dectomax.

284

Biocatalysis for the Pharmaceutical Industry OCH3 HO H3C

OCH3 O

23

CH3

O

H 22 O

O 13

H H3C

O

O H H3C O

O

OH

H

O H

CH3 25

H

H

CH3 OH

Cyclohexyl avermectin (B1)

OCH3 HO H3C

OCH3 O

O

23

H 22 O

O 13

H H3C

HO

CH3

O

O

O

OH

H

O H

CH3

25

O H H3C

H

H

H

CH3 OH

Cyclohexyl avermectin (B2)

Figure 13.9

Chemical structures of cyclohexyl avermectin B1 and B2

13.4 Excised Domains for Chemical Transformations 13.4.1 Function of Individual Domains, Domain Autonomy The activity of PK and NRPSs is often precluded and/or followed by actions upon the natural products by modifying enzymes. There exists a first level of diversity in which the monomers for respective synthases must be created. For instance, in the case of many NRPs, noncanonical amino acids must be biosynthesized by a series of enzymes found within the biosynthetic gene cluster in order for the peptides to be available for elongation by the NRPS. A second level of molecular diversity comes into play via post-synthase modification. Examples of these activities include macrocyclization, heterocyclization, aromatization, methylation, oxidation, reduction, halogenation, and glycosylation. Finally, a third level of diversity can occur in which molecules from disparate secondary metabolic pathways may interact, such as the modification of a natural product by an isoprenoid oligomer. Here, we will cover only a small subsection of

285


this diversity, namely those enzymes responsible for the first and second classes of diversity in the form of single-enzyme, discrete modifications.

13.4.2 Cyclization In order to elicit their biological properties, many natural products are constrained by macrocyclization, which can lock an elongated polypeptide or PK into distinct conformational arrangements. The cyclization of these compounds is performed in these biosynthetic systems by TE domains usually found at the C-terminus of these megasynthases. TE domains are similar to serine hydrolases, containing an active-site catalytic triad comprised of serine, threonine, and aspartic acid. In this mechanism, TE domains accept a growing oligomer form the downstream carrier protein (or thiolation) domain, creating a serine ester. Then the active site activates a nucleophile from within the oligomer (either an alcohol or amine) to attack the enzyme–substrate ester intermediate. This results in release of a cyclic product as a lactone or lactam and reset of the TE domain active site for subsequent catalysis. For an example, the TE domain of the synthase for the cyclic decapeptide tyrocidine A exists as a canonical TE domain. Figure 13.10 demonstrates this activity upon the linear tyrocidine decapeptide [41]. The excised TE domain of TycA synthase has great versatility for in vitro synthetic applications. It has been extensively studied for the ability to cyclize peptides of varying length and identity, and even the active form of the substrate (normally a carrier protein-bound thioester) may vary widely [42]. For instance, the macrolactone ring size may vary from 18 to 42 atoms with very good catalytic efficiency (Kcat ¼ 6–30 min1 and KM ¼ 3–5 mm), and all residues may be replaced with other functionalities, including a number of nonnatural variants, including PK and alkyne functionalities [43]. In one study, four amino acids (Asn, Gln, Tyr, Val) of tyrocidine A were replaced by PK moieties, allowing CP TE thioesterase H2N DPhe Pro Phe DPhe Asn Gln Tyr Val Orn Leu

(TE)

O O

tryocidine A linear decapeptide substrate

OH H2N O

H O N

N H

O

H O N

N H

O N

H N O

N O H Ph

H N O

HN

N O H NH2

O NH2 O

O

Figure 13.10

Enzymatic macrocyclization to create novel hybrid PP/PKs

286


OH

O O 1

NH2

N H

O

+NH

O N

H N O

O

O

H 9 N

3

N H

O Ph

HN H N

N H

O

O

NH2

N H NH2

O

thioesterase (TE)

O

O Resin-bound Library

OH H2N O

H O N

N H

O

H O N

N H

O N

H N O

N O H Ph

H N O

HN

N O H NH2

O NH2 O

O

Figure 13.11

TycA TE cyclization and substrate specificity

the production of hybrid peptide/PK (PP/PK) cyclic molecules [43]. In contrast to chemical macrocyclization, no protection was required for the linear precursors. In another study, the carrier protein was replaced by an enzyme compatible solid-phase resin (PEGA), and enzyme-catalyzed cyclization was used to probe substrate specificity. This study demonstrated also that oxo-esters are tolerated as substrates for TE domains, and their preparation in library format served as an excellent tool for substrate specificity studies, as well as for preparation of cyclized peptides. Figure 13.11 shows how the TycA TE showed selectivity for only residues 1 and 9 (colored in red), and changes at all other residues were tolerated [42]. Hydrogen bonding interactions are shown in green. Several compounds made from this series were shown to demonstrate improved therapeutic indices (with respect to hemolysis) while retaining antimicrobial activity. Owing to flexibility in the substrate, the TycATE was also used to synthesize a variety of novel cyclic structures. Inclusion of a propargylated amino acid into the linear substrate allowed the synthesis of over 247 macrocyclic glycopeptides, where azido-sugars were coupled onto the cyclized alkyne via copper-catalyzed 1,3-dipolar cycloaddition [44] (Figure 13.12). TE-catalyzed cyclization is not limited to the synthesis of macrocyclic peptides by catalyzing the formation of a CN bond. These enzymes are also responsible for the cyclization of NRP depsipeptide and PK lactone. Indeed, a di-domain excised from fengycin synthase was

287

Multimodular Synthases and Supporting Enzymes for Chemical Production O NH2 O

S

N H

H N

TycA TE

O Carbohydrate Azides

Alkyne Cyclic Peptide

Oligopeptide SNAC

O

H N O

R

N

N N

Carbohydrate-triazole Cyclic Peptide OH O

HO HO

O NH O

N N

O

N Ph O

N O

HN N H H2N O

O O

H N O

Tyc4PG-14

Figure 13.12

H N

N H

O

N H

NH2

O

N H

O O

NH2

H HN N O

OH

TE-catalyzed synthesis of glycosylated Tyc A Analogs

able to catalyze the formation of a macrolactone through the formation of a CO bond (Figure 13.13) [45]. Several cyclases from PK synthases have also been characterized to be functional. For example, when a TE from picromycin synthase (PICS-TE) was fused to an erythromycin module (DEBS module 3), the resulting hybrid was able to convert a diketide and OH

O

O SNAC

HO

O

O S-CoA

PICS-TE + DEBS module 3 O

Figure 13.13

Cyclization activities of TEs from PKS

O

288


S

S OH

N

cyclase O

H2O O

OH

O

Epothilone D

Figure 13.14

OH

N HO HO O

OH

O

ring opening product

Hydrolytic activity of cyclases from epothilone D synthase

2-methylmalonyl-CoA to a triketide ketolactone (Figure 13.14) [46]. However, their in vitro activity is in general much lower than the NRPS TEs and hydrolysis tends to dominate over cyclization activity, as seen in the example of epothilone D [47]. However, their in vitro activity is often low and hydrolysis tends to dominate over cyclization activity as seen in the example of epothilone D (Figure 13.14). It should be noted that while TE domains represent the most common solution in releasing macrocyclic NRPs and PKs, other pathways are known. For instance, in the biosynthesis of cyclosporine, the cyclization is proposed to be catalyzed by the most downstream C-domain [48]. Macrocyclization can also occur under reduction of a carbonyl group mediated by a reduction domain (R-domain) as proposed in the synthesis of the macrocyclic imine nostocyclopeptide [49]. The synthetic utility of these cyclization strategies has not yet been reported.

13.4.3 Halogenation Natural products require halogens to be strategically placed to achieve the desired biological activities (Figure 13.15) [50]. For electron-rich substrates, nature often uses flavin-dependent halogenases or haloperoxidases (vanadium or heme-iron) for enzymatic chlorination, bromination or iodination (snyderol, tetraiodothyronine, Figure 13.1). For electron-deficient molecules such as alkanes, mononuclear iron halogenases are utilized (barbamide, Figure 13.15) [51]. On the other hand, fluorination undergoes an SN2 nucleophilic substitution mechanism. Halogenation by haloperoxidases often takes place with poor regio- and stereo-selectivity since the activated halogen sources (hypohalous acids) are freely diffusible within and away from enzymes. As a result, their synthetic utility is largely limited. However, both flavindependent and mononuclear iron halogenases elicit high regio- and stereo-selectivity [52,53]. While it is still in its infancy, the synthetic use of enzymatic halogenation shows great promise. FADH2 halogenases such as tryptophan 7-halogenase have been shown to catalyze regioselective halogenation of a wide range of indole derivatives and aromatic heterocycles (eq. 1 in Figure 13.16) [54]. Using the same enzyme, a library of antitumor indolocarbazoles was generated in a combinatorial fashion in vivo upon halogenation and glycosylation (not shown) (eq. 2 in Figure 13.16) [55]. Much was unknown for the halogenation for unreactive substrates until very recently, when the biosynthesis of the cyclopropyl amino acid side chain of coronatine was elucidated. This intriguing pathway, which involves g-chlorination of an enzyme-bound L-isoleucine followed by chloride displacement by the a-carbon, yields the cyclopropanated precursor

CO2H H2N

I

I

HO

O

I

Br OH

I

Synderol olefin bromination

Tetraiodothyronine phenol iodination

Cl O

Cl Cl

N

OH OMe

S

OH

N

F

R2

Examples of halogenated natural products

R2

tryptophan 7-halogenase

R1

NaCl

N H

NH2

4-Fluorothreonine SN2-fluorination

Barbamide radical chlorination

Figure 13.15

O

R1

R1 = H, Me eq. 1 R2 = amine, alkyl, nitrile

Cl N H Indoles

COOH NH2

COOH tryptophan 7-halogenase

NH2

Cl-, Br-

N H

X

N H X = Cl, Br

H N

O

O

eq. 2

R1

N H

N H

R2

R1 = Cl, Br, H R2 = Cl, Br, H Indolocarbazole aglycons

Figure 13.16

Halogenation of indoles by tryptophan 7-halogenase

290

Biocatalysis for the Pharmaceutical Industry Cl H O

H2N

chlorinase

O

O

H2N

Cl-

H 2N

S

S

S

enzyme

enzyme

enzyme O H H H N

H

CO2H

O

H

Coronatine

Figure 13.17

Chlorinase in biosynthesis of coronatine

(Figure 13.17) [56]. It is believed that a variety of halogenations proceed via this mechanism for electron-deficient molecules [57,58]. It is yet to be determined if the halogenation can take place on substrates which are not covalently linked to natural product synthases. So far, only one fluorinase has been characterized in the pathways, producing organofluorine natural products such as 4-flurothreonine [59]. The enzyme uses S-adenosyl methionine (SAM) as a substrate for fluorine displacement, where L-methionine serves as the leaving group to form 5-fluoroadenosine (Figure 13.18). This fluorinase was recently used to prepare 18 F-containing positron emission reagents [60]. NH2

NH2 N

Me S+

HO2C

O

N

NH2

N

N F N

fluorinase

O

N

N N

FOH OH

OH OH

SAM

5-deoxy-5fluoroadenosine

OH

O OH

F

NH2

4-fluorothreonine

Figure 13.18

Biosynthesis of fluorinated metabolites via 50 -deoxy-50 -fluoroadenosine

13.4.4 Heterocyclization/Aromatization Heterocycles are found in a wide variety of natural products, and the chemical nature of these moieties imparts recognition elements critical to both protein and nucleic acid targets. These moieties may be biosynthesized via either ribosomal or nonribosomal pathways and can occur either singly within a molecule or as multiple, repeating heterocyclic units within the same compound (Figure 13.19).

291


S

N

N

OH

S

OH

O

S N H

OH OH

N

O

S O

Yersiniabactin

OH

O

Epothilone D

HO S HH

O

CO2H

N

S

O O

H

N

OCH3

O H O

OH

CO2H

OH OH

Pyochelin I Monesin A

Figure 13.19

Examples of heterocyclic natural products

In NRPs and hybrid NRP–PK natural products, the heterocycles oxazole and thiazole are derived from serine and cysteine amino acids respectively. For their creation, a cyclization (or Cy) domain is responsible for nucleophilic attack of the side-chain heteroatom within a dipeptide upon the amide carbonyl joining the amino acids [61]. Once the cyclic moiety is formed, the ring may be further oxidized, to form the oxazoline/thiazoline, or reduced, to form oxazolidine/thiazolidine (Figure 13.20). For substituted oxazoles and thiazoles, such as those

R

O HX

T

N H

O

Cy domain

T

X

R

X = O, serine (T = H) X = S, cysteine (T = H) T = Me, threonine

N OH H

dehydration

O T

X

X = O, oxazoline (T = H, Me) X = S, thiazoline

N R oxidase T

X N R

reductase T

X N

O

X = O, oxazole (T = H, Me) X = S, thiazole

Figure 13.20

O

H R O X = O, oxazolidine (T = H, Me) X = S, thiazolidine

Biosynthesis of heterocycles by cyclization–dehydration

292


containing 2-methyloxazoline, the reaction mechanism is the same, but the amino acid is methylated, such as in threonine. Several current efforts are focusing on the portability of enzymatic heterocyclization. For example, novel chiral heterocyclic carboxylic acids were produced by using hybrid enzymes [62] (Figure 13.21). Stimulated by biosynthesis pathways, biomimetic heterocyclization methods have also been developed with high efficiency [63].

R H2N

COOH

XH

H2N

Isoleucine

hybrid enzyme

COOH

R = H, Me X = O, S

O

N

H2N O

Figure 13.21

N

H2N

S

OH R

O

N

H2N

OH

S

O OH

Ile-Cys thiazole

Ile-Cys thiazoline

Hybrid enzymes to prepare heterocyclic carboxylic acids

13.4.5 Methylation Methyl transferases are responsible for methylation of a nucleophile, typically using SAM as the carbon donor. They are known to accept a wide range of nucleophiles such as halides (eq. 1 in Figure 13.22) [64], amines (eq. 2 in Figure 13.22) [65], hydroxyls, and enolates. As expected, the reactivity of methyl transfer to halides follows the order of iodide, bromide, and chloride, with chloride being the poorest acceptor. Methylation of amines in nucleotides and proteins plays important roles in biological activities. methyl transferase

X-

H N

Me-X

SAM

O N H

X = Cl, Br, I

H N

eq. 1

O N H

methyl transferase

eq. 2

SAM

NH2

NH Me

Figure 13.22

Enzymatic methylation of halides and amines

293


The electrophilic potency of the positively charged SAM allows unusual acceptors, such as carboxylic acids for ester formation [65] and even electron-rich carbon atoms (eq. 2 in Figure 13.23) [66]. Remarkably, in the biosynthesis of the aminocoumarin core of the novobiocin aglycon, the methylation takes place at only one phenolic carbon and none of the three hydroxyl groups was methylated [67,68]. This indicates the capacity for high regioselectivity in these transformations as a result of the binding geometry of SAM and the substrate. Methyl transfer to electron-deficient substrates often occurs under radical mechanisms requiring methylcobalamin as the cofactor, as shown in the biosynthesis of fosfomycin, where only one of the two enantiotopic hydrogen was replaced by the methyl group (Figure 13.23) [69]. Despite the higher selectivity of enzymatic methyl transfer over chemical methylation, where toxic or hazardous reagents are often employed, such as methyl sulfonate and diazomethane, the synthetic applications of these enzymes have been largely ignored primarily as a result of high costs associated with the cofactor SAM. Recent efforts have been directed to in vivo methylation, where SAM may be regenerated inside cells. For example, methyl benzoate production was engineered in recombinant Saccharomyces cerevisiae and in vivo OMe N

S R

N

methyl transferase

O

S

MeO

SAM

OH

OMe

R= N

S

O eq. 1

R

N

S

MeO

OCH3

Myxothiazol Z OH

HO

O

OH H N O

methyl transferase O

SAM OH

OH

H N

O

O

eq. 2 HO

Novobiocin aglycon

CH3 O -O

P -O

H

H OH

Figure 13.23

methyl transferase methylcobalamin

-O

O

O Me P -O

O

H OH

-O

P

Me

-O

H

H O Fosfomycin

eq. 3

Methylation in biosynthesis of myxothiazol, novobiocin, and fosfomycin

294


methylation was accomplished through heterologous expression of Antirrhinum majus benzoic acid methyl transferase [70]. Even under suboptimal conditions yielding 1 mg L1, the direct cost of this approach is comparable to that of commercially available, naturally derived methyl benzoate. This indicates the potential for this methodology in the production of plant-derived aroma compounds, which are mostly obtained through chemical synthesis and extraction from plant materials.

13.4.6 Oxygenation Nature often uses oxygenases to produce oxygenated molecules through oxidative functionalization of precursors (for an alternative pathway using CC bond-forming enzymes, see Fessner and Helaine [71]), which catalyze direct incorporation of molecular oxygen into substrates. Oxygenases are categorized as either monooxygenases or dioxygenases, depending upon whether they insert one or both atoms of dioxygen into a substrate. Oxygenases such as cytochrome P450s and heme oxygenases primarily use metal cofactors, which in their low oxidation states can complex with dioxygen, the substrate, or both. Others use an organic cofactor such as dihydroflavin or tetrahydropterin to donate an electron to dioxygen. These enzymes catalyze a variety of oxidative reactions in natural product biosynthesis with two CH hydroxylation examples shown in Figure 13.24 [72,73]. It should be noted that CH activation by nonheme iron oxygenases, such as aromatic dioxygenases, is an important pathway in degradation of aromatics into cis-dihydrodiols, which are important chiral building blocks for chemical synthesis [74,75]. Very recently, several enzymes have been reported to promote dioxygenation without apparent reliance on cofactors in the biosynthesis of aromatic PKs (eq. 1 in Figure 13.25) [76,77] O P 450 monooxygenase

S NH2 enzyme

HO

OH

O2

O

OH NH2

S

eq. 1

NH2 enzyme

HO

HO

O O Coumarin

NH N O

COOH

N H

OH N O

COOH

Figure 13.24

NH2

nonheme iron monooxygenase O2

NH N H

NH2 Clavulanic acid

eq. 2

CH activation by P450 and nonheme monooxygenases

295


O

CH3 O

O monooxygenase H

COOH

CH3 O

O2

H

eq. 1 COOH

O O

OH OH

N H

dioxygenase

CH3

O2

Figure 13.25

O OH

O

O N H

CO

CH3

OH O N H

eq. 2 CH3

Cofactorless oxygenases

and aerobic bacterial degradation of aromatic compounds (eq. 2 in Figure 13.25) [78]. Dioxygen and/or substrate activation is likely to be performed in the catalytic site through formation of a protein radical intermediate or direct electron transfer from the substrate to molecular oxygen to a form a radical pair. It is very attractive to use cofactorless oxygenases for chemical synthesis to avoid the issue of cofactor regeneration. CH activation has perhaps the highest potential of all enzyme-catalyzed transformations for synthetic applications. At the same time, these transformations are often the most difficult processes to be carried out on a practical scale. Currently, they require whole-cell processes and the outcome is often unpredictable. The discovery of new oxygenases and efficient hosts for protein expression remain keys to further expanding the synthetic applications of enzymatic hydroxylation.

13.4.7 Glycosylation Many natural products are decorated with carbohydrates, which often endow essential biological activities to the parent molecules (Figure 13.26) [79]. For example, the antibiotic activity of erythromycin is mostly abolished when the D-desosamine is removed from the molecule. The glycosylation often occurs through O-, N-, or C-linkages to their respective aglycons, as seen in erythromycin, C-1027 and novobiocin (O-linkage), rebeccamycin (N-linkage) and simocyclinone (C-linkage). In general, enzymatic glycosylation takes place toward the end of natural product biosynthesis. For example, in the biosynthesis of the lipoglycopeptide teicoplanin, the first stage involves the assembly of a linear NRP from monomeric amino acids, which can be racemized, halogenated, and hydroxylated after being loaded to the multimodular synthase (Figure 13.27). Subsequently, the linear peptide is subject to post-modification, such as oxidative crosscoupling/cyclization, glycosylation and acylation. The completion of teicoplanin biosynthesis requires three Gtfs, and the favored pathway was determined by enzymatic kinetic studies [80,81]. The recent discovery of a variety of Gtfs has greatly facilitated the synthesis of novel glycosylated PKs and NRPs through introducing nonnative aglycons and carbohydrates. Enzymatic glycosylation eliminates the need for extensive protection and deprotection required by traditional chemical methods [82]. Remarkably, these enzymes often display great promiscuity toward nucleoside diphosphate sugars. For example, one Gtfe cloned from

296


OMe

D-desosamine O

O

N H O

N

HO

O

OH HO O

O

O

O

O

O

N

O OH

OH

O

O

OH

O erythromycin H N

O

N H

Cl

NH2

C-1027

O OH

OH

H N

O

O

N O

O

OH

O

Cl OH

O

HO

O

MeO O

OH

O OH H2N

O

rebeccamycin

O

H N

novobiocin

O O HO

O

MeO OH

O

O O

O

O

OH

4

HO

O

O

OH

O

OH

OH

Simocyclinone

Figure 13.26

Exemplary glycosylated natural products

the glycopeptide A-40926 biosynthesis gene cluster allows a wide range of sugars to be installed to the A-40926 aglycons (R, R0 ¼ OH, NH2) (Figure 13.28) [80,83]. These novel glycopeptides were further modified by a promiscuous acyl transferase to generate a lipoglycopeptide library as analogs of A-40926. By incorporating azido sugars, the glyco-moiety can be further randomized through copper-catalyzed 1,3-dipolar cycloaddition [84]. The availability of a variety of Gtfs and sugar biosynthesis genes provides the basis for a general strategy to produce glycosylated molecules in vivo [85]. For example, a Gtf from

O

OH

O

HO

HO

O

HN

NH O

O

HO

HO

O

N H

O

O

H N

OH

O

O

H

N H Cl

N

OH

HO

O

O

O

O

O

N H

OH

OH

Cl H N

OH

O

HO

O

H N

Cl

H N

HO

HN

OH

OH

O NH2

OH

O

N H

O

cyclic peptide: teicoplanin

OH

O

O

OH

H N

O

linear peptide

HO

OH

O Cl

O

N H

OH

Figure 13.27 Biosynthesis of teicoplanin and logistic glycosylation

glycosylation (red) acylation (pink)

HO HO

racemization (pink) halogenation (red) hydroxylation (blue) condensation to form peptide bond

oxidative cross-coupling/cyclization (blue)

amino acids monomers

activation loading to synthase

HO

OH

NH2

Cl

OH

Multimodular Synthases and Supporting Enzymes for Chemical Production 297

298


OH O HO O HN

O

Cl H N

O N H

O

H N

N H Cl

O

HO

O

HO O

Gtf

N H

O

O HO

HO

O

HN

HO

OH O OH

OH

OH

OH

R O

HO O

N H

HN

R′ O

Cl H N

O

OH

O

O

O N H Cl

O

HO

H N O

HO O

HN

N H

O

O HO

HO

acyl transferase

O

HO

OH O OH OH

OH O H R′′

OH

N

O HO O HN

N H

R′ O

Cl H N

O

OH

O

O

O

O

HO

H N

N H Cl

O

HO O HO

O N H

R = OH, NH2 R′ = OH, NH2 R′′ = C5-C15 alkyl chain

O

O HO

HN

OH

HO

O OH OH

OH A-40926 analogs

Figure 13.28

Enzymatic glycosylation to prepare novel glycopeptides

299

Multimodular Synthases and Supporting Enzymes for Chemical Production O

O glycosyl transferase

HO

O

O

OH

O O

HO

O

O

O

O

O

HO

OH

HO

NHAc

OH O

O

O

Sugar = HO OH

glucose

NHMe2 OH OH

O HO

sugar

OH

NHAc

O HO

Me

Figure 13.29 In vivo glycosylation to create novel glycosylated macrolides

picromycin biosynthesis was able to take a surprisingly wide range of sugars allowing in vivo production of a library of novel PKs (Figure 13.29); see Borisova et al. [86] and references cited therein. Similar work has been demonstrated for the synthesis of rebeccamycin analogs from indolocarbazole aglycons in vivo [55].

13.5 Conclusions The biosynthesis of natural products offers chemists a multitude of examples and tools for the discovery and production of diverse chemical entities. Through the development of new technologies such as metabolic engineering, the first examples of directed fermentation and combinatorial control of natural product analogs points the way to the future of natural product analog production. Further, by the study of discrete enzymes in natural product biosynthesis, we can begin to enzymatically manipulate small molecules in ways that are often difficult through traditional synthetic techniques. We are now at an exciting crossroads, where the new tools of natural product biosynthesis have an opportunity to impact small-molecule design and production directly.

References [1] Demain, A.L. and Adrio, J.L. (2008) Strain improvement for production of pharmaceuticals and other microbial metabolites by fermentation. Progress in Drug Research, 65, 251, 253–289. [2] Panke, S. and Wubbolts, M. (2005) Advances in biocatalytic synthesis of pharmaceutical intermediates. Current Opinion in Chemical Biology, 9 (2), 188–194. [3] Floss, H.G. (2006) Combinatorial biosynthesis – potential and problems. Journal of Biotechnology, 124 (1), 242–257.

300


[4] Ku, J., Mirmira, R.G., Liu, L. and Santi, D.V. (1997) Expression of a functional non-ribosomal peptide synthetase module in Escherichia coli by coexpression with a phosphopantetheinyl transferase. Chemistry & Biology, 4 (3), 203–207. [5] Liou, G.F., Lau, J., Cane, D.E. and Khosla, C. (2003) Quantitative analysis of loading and extender acyltransferases of modular polyketide synthases. Biochemistry, 42 (1), 200–207. [6] Leadlay, P.F., Staunton, J., Oliynyk, M. et al. (2001) Engineering of complex polyketide biosynthesis – insights from sequencing of the monensin biosynthetic gene cluster. Journal of Industrial Microbiology & Biotechnology, 27 (6), 360–367. [7] Kwon, H.J., Smith, W.C., Scharon, A.J. et al. (2002) CO bond formation by polyketide synthases. Science, 297 (5585), 1327–1330. [8] Doekel, S. and Marahiel, M.A. (2001) Biosynthesis of natural products on modular peptide synthetases. Metabolic Engineering, 3 (1), 64–77. [9] Carreras, C.W. and Ashley, G.W. (2000) Manipulation of polyketide biosynthesis for new drug discovery. EXS, 89, 89–108. [10] Du, L., Sanchez, C. and Shen, B. (2001) Hybrid peptide–polyketide natural products: biosynthesis and prospects toward engineering novel molecules. Metabolic Engineering, 3 (1), 78–95. [11] Cane, D.E. and Walsh, C.T. (1999) The parallel and convergent universes of polyketide synthases and nonribosomal peptide synthetases. Chemistry & Biology, 6 (12), R319–R325. [12] Mootz, H.D. and Marahiel, M.A. (1997) Biosynthetic systems for nonribosomal peptide antibiotic assembly. Current Opinion in Chemical Biology, 1 (4), 543–551. [13] Weber, T. and Marahiel, M.A. (2001) Exploring the domain structure of modular nonribosomal peptide synthetases. Structure (Camb), 9 (1), R3–R9. [14] Mootz, H.D., Schwarzer, D. and Marahiel, M.A. (2002) Ways of assembling complex natural products on modular nonribosomal peptide synthetases. Chembiochem: A European Journal of Chemical Biology, 3 (6), 490–504. [15] Hopwood, D.A. and Sherman, D.H. (1990) Molecular genetics of polyketides and its comparison to fatty acid biosynthesis. Annual Review of Genetics, 24, 37–66. [16] Wang, G. (2006) Diversity and biotechnological potential of the sponge-associated microbial consortia. Journal of Industrial Microbiology & Biotechnology, 33 (7), 545–551. [17] Walsh, C.T., Chen, H., Keating, T.A. et al. (2001) Tailoring enzymes that modify nonribosomal peptides during and after chain elongation on NRPS assembly lines. Current Opinion in Chemical Biology, 5 (5), 525–534. [18] Moore, B.S. and Hertweck, C. (2002) Biosynthesis and attachment of novel bacterial polyketide synthase starter units. Natural Product Reports, 19 (1), 70–99. [19] Hamann, M.T. (2003) Enhancing marine natural product structural diversity and bioactivity through semisynthesis and biocatalysis. Current Pharmaceutical Design, 9, (11) 879–889. [20] Thiericke, R. and Rohr, J. (1993) Biological variation of microbial metabolites by precursor-directed biosynthesis. Natural Product Reports, 10 (3), 265–289. [21] Hutchinson, C.R. and McDaniel, R. (2001) Combinatorial biosynthesis in microorganisms as a route to new antimicrobial, antitumor and neuroregenerative drugs. Current Opinion in Investigative Drugs, 2 (12), 1681–1690. [22] Tang, L., Shah, S., Chung, L. et al. (2000) Cloning and heterologous expression of the epothilone gene cluster. Science, 287 (5453), 640–642. [23] Julien, B. and Shah, S. (2002) Heterologous expression of epothilone biosynthetic genes in Myxococcus xanthus. Antimicrobial Agents and Chemotherapy, 46 (9), 2772–2778. [24] Aguirrezabalaga, I., Olano, C., Allende, N. et al. (2000) Identification and expression of genes involved in biosynthesis of L-oleandrose and its intermediate L-olivose in the oleandomycin producer. Streptomyces antibioticus. Antimicrobial Agents and Chemotherapy, 44 (5), 1266–1275. [25] Miao, V., Co€effet-Legal, M.F., Brian, P. et al. (2005) Daptomycin biosynthesis in Streptomyces roseosporus: cloning and analysis of the gene cluster and revision of peptide stereochemistry. Microbiology (Reading, England), 151 (Pt 5), 1507–1523. [26] Miao, V., Co€effet-Le Gal, M.F., Nguyen, K. et al. (2006) Genetic engineering in Streptomyces roseosporus to produce hybrid lipopeptide antibiotics. Chemistry & Biology, 13 (3), 269–276. [27] Nguyen, K.T., Kau, D., Gu, J.Q. et al. (2006) A glutamic acid 3-methyltransferase encoded by an accessory gene locus important for daptomycin biosynthesis in Streptomyces roseosporus. Molecular Microbiology, 61 (5), 1294–1307. [28] Baltz, R.H. (2006) Molecular engineering approaches to peptide, polyketide and other antibiotics. Nature Biotechnology, 24 (12), 1533–1540.


301

[29] Baltz, R.H., Miao, V. and Wrigley, S.K. (2005) Natural products to drugs: daptomycin and related lipopeptide antibiotics. Natural Product Reports, 22 (6), 717–741. [30] Burg, R.W., Miller, B.M., Baker, E.E. et al. (1979) Avermectins, new family of potent anthelmintic agents: producing organism and fermentation. Antimicrobial Agents and Chemotherapy, 15, 361–367. [31] Ikeda, H., Nonomiya, T., Usami, M. et al. (1999) Organization of the biosynthetic gene cluster for the polyketide anthelmintic macrolide avermectin in Streptomyces avermitilis. Proceedings of the National Academy of Sciences of the United States of America, 96, 9509–9514. [32] Cane, D.E., Liang, T.C., Kaplan, L.K. et al. (1983) Biosynthetic origin of the carbon skeleton and oxygen atoms of the avermectins. Journal of the American Chemical Society, 105, 4110–4112. [33] Chen, T.S., Arison, B.H., Gullo, V.P. and Inamine, E.S. (1989) Further studies on the biosynthesis of the avermectins. Journal of Industrial Microbiology, 4, 231–237. [34] MacNeil, D.J., Occi, J.L., Gewain, K.M. et al. (1992) Complex organization of the Streptomyces avermetilis genes encoding the avermectin polyketide synthase. Gene, 115, 119–125. [35] Dutton, C.J., Gibson, S.P., Goudie, A.C. et al. (1991) Novel avermectins produced by mutational biosynthesis. Journal of Antibiotics (Tokyo), 44, 357–365. [36] Hafner, E.W., Holley, B.W., Holdom, K.S. et al. (1991) Branched-chain fatty acid requirement for avermectin production by a mutant of Streptomyces avermitilis lacking branched-chain 2-oxo acid dehydrogenase activity. Journal of Antibiotics (Tokyo), 44, 349–356. [37] Stutzman-Engwall, K.J., Conlon, S.W., Fedechko, R.W. and Kaczmarek, F.S. (1997) Engineering Streptomyces avermitilis biosynthetic genes, in Developments in Industrial Microbiology, vol. 35 (eds. C.R. Hutchinson and J. McAlpine), Society for Industrial Microbiology, Fairfax, VA, pp. 7–13. [38] Ikeda, H. and Omura, S. (1995) Control of avermectin biosynthesis in Streptomyces avermitilis for the selective production of a useful component. Journal of Antibiotics (Tokyo), 48, 549–562. [39] Stutzman-Engwall, K.J., Conlon, S.W., Fedechko, R.W. et al. (2003) Engineering the aveC gene to enhance the ratio of doramectin to its CHC-B2 analogue produced in Streptomyces avermitilis. Biotechnology and Bioengineering, 82, 359–369. [40] Stutzman-Engwall, K.J., Conlon, S.W., Fedechko, R.W. et al. (2005) Semi-synthetic DNA shuffling of aveC leads to improved industrial scale production of doramectin by Streptomyces avermitilis. Metabolic Engineering, 7, 27–37. [41] Trauger, J.W., Kohli, R.M. and Walsh, C.T. (2001) Cyclization of backbone-substituted peptides catalyzed by the thioesterase domain from the tyrocidine nonribosomal peptide synthetase. Biochemistry, 40 (24), 7092–7098. [42] Kohli, R.M., Walsh, C.T. and Burkart, M.D. (2002) Biomimetic synthesis and optimization of cyclic peptide antibiotics. Nature, 418 (6898), 658–661. [43] Kohli, R.M., Burke, M.D., Tao, J. and Walsh, C.T. (2003) Chemoenzymatic route to macrocyclic hybrid peptide/ polyketide-like molecules. Journal of the American Chemical Society, 125, 7160–7161. [44] Lin, H., Thayer, D.A., Wong, C.H. and Walsh, C.T. (2004) Macrolactamization of glycosylated peptide thioesters by the thioesterase domain of tyrocidine synthetase. Chemistry & Biology, 11 (12), 1635–1642. [45] Sieber, S.A., Tao, J., Walsh, C.T. and Marahiel, M.A. (2004) Peptidyl thiophenols as substrates for nonribosomal peptide cyclases. Angewandte Chemie (International Edition in English), 43, 493–498. [46] Lu, H., Tsai, S.-C., Khosla, C. and Cane, D.E. (2002) Expression, site-directed mutagenesis, and steady state kinetic analysis of the terminal thioesterase domain of the methymycin/picromycin polyketide synthase. Biochemistry, 41, 12590–12597. [47] Boddy, C.N., Schneider, T.L., Hotta, K. et al. (2003) Epothilone C macrocyclization and hydrolysis are catalyzed by the isolated thioesterase domain of epothilone polyketide synthase. Journal of the American Chemical Society, 125, 3428–3429. [48] Weber, G., Sch€orgendorfer, K., Schneider-Scherzer, E. and Leitner, E. (1994) The peptide synthetase catalyzing cyclosporine production in Tolypocladium niveum is encoded by a giant 45.8-kilobase open reading frame. Current Genetics, 26 (2), 120–125. [49] Becker, J.E., Moore, R.E. and Moore, B.S. (2004) Cloning, sequencing, and biochemical characterization of the nostocyclopeptide biosynthetic gene cluster: molecular basis for imine macrocyclization. Gene, 325, 35–42. [50] Yarnell, A. (2006) C&E News, 84 (21), 12–18. [51] Burd, V.N. and van Pee, K.-H. (2003) Halogenating enzymes in the biosynthesis of antibiotics. Biochemistry, 68, 1132–1135. [52] Dong, C., Flecks, S., Unversucht, S. et al. (2005) Tryptophan 7-halogenase (PrnA) structure suggests a mechanism for regioselective chlorination. Science, 309, 2216–2219.

302


[53] Blasiak, L.G., Vaillancourt, F.H., Walsh, C.T. and Drennan, C.L. (2006) Crystal structure of the non-haem iron halogenase SyrB2 in syringomycin biosynthesis. Nature, 440, 368–371. [54] H€ olzer, M., Burd, W., Reibig, H.-U. and van Pee, K.-H. (2001) Substrate specificity and regioselectivity of tryptophan 7-halogenase from Pseudomonas fluorescens BL915. Advanced Synthesis & Catalysis, 343, 591–595. [55] Sanchez, C., Zhu, L., Branã, A.F. et al. (2005) Combinatorial biosynthesis of antitumor indolocarbazole compounds. Proceedings of the National Academy of Sciences of the United States of America, 102, 461–466. [56] Vaillancourt, F.H., Yeh, E., Vosburg, D.A. et al. (2005) Cryptic chlorination by a non-haem iron enzyme during cyclopropyl amino acid biosynthesis. Nature, 436, 1191–1194. [57] Galonic, P.D., Vaillancourt, F.H. and Walsh, C.T. (2006) Halogenation of unactivated carbon centers in natural product biosynthesis: trichlorination of leucine during barbamide biosynthesis. Journal of the American Chemical Society, 128, 3900–3901. [58] Vaillancourt, F.H., Vosburg, D.A. and Walsh, C.T. (2006) Dichlorination and bromination of a threonyl-S-carrier protein by the non-heme Fe(II) halogenase SyrB2. ChemBioChem, 7, 748–752. [59] OHagan, D., Schaffrath, C., Cobb, S.L. et al. (2002) Biochemistry: biosynthesis of an organofluorine molecule. Nature, 416, 279. [60] OHagan, D. (2006) Recent developments on the fluorinase from Streptomyces cattleya. Journal of Fluorine Chemistry, 127, 1479–1483. [61] Roy, R.S., Gehring, A.M., Milne, J.C. et al. (1999) Thiazole and oxazole peptides: biosynthesis and molecular machinery. Natural Product Reports, 16 (2), 249–263. [62] Duerfahrt, T., Eppelmann, K., Muller, R. and Marahiel, M.A. (2004) Rational design of a bimodular model system for the investigation of heterocyclization in nonribosomal peptide biosynthesis. Chemistry & Biology, 11, 261–271. [63] You, S.-L., Razavi, H. and Kelly, J.W. (2003) A biomimetic synthesis of thiazolines using hexaphenyl-oxodiphosphonium trifluoromethanesulfonate. Angewandte Chemie (International Edition in English), 42, 83–85. [64] Wuosmaa, A.M. and Hager, L.P. (1990) Methyl chloride transferase: a carbocation route for biosynthesis of halometabolites. Science, 249, 160–162. [65] Blackburn, G.M., Gamblin, S.J. and Wilson, J.R. (2003) Mechanism and control in biological amine methylation. Helvetica Chimica Acta, 86, 4000–4006. [66] Weinig, S., Hecht, H.-J., Mahmud, T. and Muller, R. (2003) Melithiazol biosynthesis: further insights into myxobacterial PKS/NRPS systems and evidence for a new subclass of methyl transferases. Chemistry & Biology, 10, 939–952. [67] Pacholec, M., Tao, J. and Walsh, C.T. (2005) CouO and NovO: C-methyltransferases for tailoring the aminocoumarin scaffold in coumermycin and novobiocin antibiotic biosynthesis. Biochemistry, 44, 14969–14976. [68] Tao, J., Hu, S., Pacholec, M. and Walsh, C.T. (2003) Synthesis of proposed oxidation–cyclization–methylation intermediates of the coumarin antibiotic biosynthetic pathway. Organic Letters, 5, 3233–3236. [69] Woodyer, R.D., Li, G., Zhao, H. and van der Donk, W.A. (2007) New insight into the mechanism of methyl transfer during the biosynthesis of fosfomycin. Chemical Communications, 4, 359–361. [70] Farhi, M., Dudareva, N., Masci, T. et al. (2006) Synthesis of the food flavoring methyl benzoate by genetically engineered Saccharomyces cerevisiae. Journal of Biotechnology, 122, 307–315. [71] Fessner, W.-D. and Helaine, V. (2001) Biocatalytic synthesis of hydroxylated natural products using aldolases and related enzymes. Current Opinion in Biotechnology, 12, 574–586. [72] Chen, H. and Walsh, C.T. (2001) Coumarin formation in novobiocin biosynthesis: beta-hydroxylation of the aminoacyl enzyme tyrosyl-S-NovH by a cytochrome P450 NovI. Chemistry & Biology, 8, 301–312. [73] Que, L. Jr. (2000) One motif – many different reactions. Nature Structural Biology, 7, 182–184. [74] Gibson, D.T. and Parales, R.E. (2000) Aromatic hydrocarbon dioxygenases in environmental biotechnology. Current Opinion in Biotechnology, 11, 236–243. [75] Boyd, D.R., Sharma, N.D. and Allen, C. (2000) Aromatic dioxygenases: molecular biocatalysis and applications. Current Opinion in Biotechnology, 12, 564–573. [76] Fetzner, S. (2002) Oxygenases without requirement for cofactors or metal ions. Applied Microbiology and Biotechnology, 60, 243–257. [77] Kendrew, S.G., Hopwood, D.A. and Marsh, E.N.G. (1997) Identification of a monooxygenase from Streptomyces coelicolor A3(2) involved in biosynthesis of actinorhodin: purification and characterization of the recombinant enzyme. Journal of Bacteriology, 179, 4305–4310.


303

[78] Bauer, I., Max, N., Fetzner, S. and Lingens, F. (1996) 2,4-Dioxygenases catalyzing N-heterocyclic-ring cleavage and formation of carbon monoxide. Purification and some properties of 1H-3-hydroxy-4-oxoquinaldine 2,4dioxygenase from Arthrobacter sp. R€u61a and comparison with 1H-3-hydroxy-4-oxoquinoline 2,4-dioxygenase from Pseudomonas putida 33/1. European Journal of Biochemistry, 240, 576–583. [79] Borman, S. (2006) C & E News, 84 (36), 13–22. [80] Lu, W., Leimkuhler, C., Oberth€ur, M. et al. (2004) Characterization of a regiospecific epivancosaminyl transferase GtfA and enzymatic reconstitution of the antibiotic chloroeremomycin. Proceedings of the National Academy of Sciences of the United States of America, 101, 4390–4395. [81] Howard-Jones, A.R., Kruger, R.G., Lu, W. et al. (2007) Kinetic analysis of teicoplanin glycosyltransferases and acyltransferase reveal ordered tailoring of aglycone scaffold to reconstitute mature teicoplanin. Journal of the American Chemical Society, 129, 10082–10083. [82] Griffith, B.R., Langenhan, J.M. and Thorson, J.S. (2005) ‘Sweetening’ natural products via glycorandomization. Current Opinion in Biotechnology, 16, 622–630. [83] Kruger, R.G., Lu, W., Oberthur, M. et al. (2005) Tailoring of glycopeptide scaffolds by the acyltransferases from the teicoplanin and A-40,926 biosynthetic operons. Chemistry & Biology, 12, 131–140. [84] Fu, X., Albermann, C.S. and Thorson, J.S. (2005) Diversifying vancomycin via chemoenzymatic strategies. Organic Letters, 8, 1513–1515. [85] Blanchard, S. and Thorson, J.S. (2006) Enzymatic tools for engineering natural product glycosylation. Current Opinion in Chemical Biology, 10, 263–271. [86] Borisova, S., Zhao, L., Melancon, C.E. et al. (2004) Characterization of the glycosyltransferase activity of desVII: analysis of and implications for the biosynthesis of macrolide antibiotics. Journal of the American Chemical Society, 126, 6534.

14 Green Chemistry with Biocatalysis for Production of Pharmaceuticals Oliver May DSM Pharmaceutical Products, PO Box 18, 6160 MD Geleen, The Netherlands

14.1 Introduction Chemistry is a key scientific discipline which contributes to sales of approximately US$1500 billion worldwide of which more than US$500 billion sales are generated with pharmaceuticals [1]. The strong economic value mirrors the huge impact of chemistry for our society and in our everyday life, being it for convenience, food or health reasons. It has been recognized that such an impact calls for responsibility. Responsible Care is indeed the chemical industrys global voluntary initiative started in 1985 under which companies, through their national associations, work together to continuously improve their health, safety and environmental performance, and communicate with stakeholders about their products and processes in the manufacture and supply of safe and affordable goods that bring real benefits to society (for information about the Responsible Care initiative see http:// www.responsiblecare.org). In 1987 it was recognized that sustainable development, which was defined as meeting the needs of the present generation without compromising the ability of future generations to meets their own needs, should be a guiding principle of the United Nations, governments and private institutions, organizations and enterprises [2]. The concept of ‘Green Chemistry’ was first formulated in the mid 1990s by Anastas also to address environmental challenges of chemical products and process by which they are produced [3]. Similar concepts such as ‘atom economy’ or ‘E Factor’ have been developed, all with the same guiding principles of designing environmental benign products and processes [4]. The following 12 principles of green chemistry have been formulated by Anastas and Warner [5]:


306


1. Prevention: It is better to prevent waste than to treat or clean up waste after it has been created. 2. Atom Economy: Synthetic methods should be designed to maximize the incorporation of all materials used in the process into the final product. 3. Less Hazardous Chemical Syntheses: Wherever practicable, synthetic methods should be designed to use and generate substances that possess little or no toxicity to human health and the environment. 4. Design Safer Chemicals: Chemical products should be designed to effect their desired function while minimizing their toxicity. 5. Safer Solvents and Auxiliaries: The use of auxiliary substances – solvents, separation agents, and others – should be made unnecessary wherever possible and innocuous when used. 6. Design for Energy Efficiency: Energy requirements of chemical processes should be recognized for their environmental and economic impacts and should be minimized. If possible, synthetic methods should be conducted at ambient temperature and pressure. 7. Use Renewable Feedstocks: A raw material or feedstock should be renewable rather than depleting whenever technically and economically practicable. 8. Reduce Derivatives: Unnecessary derivatization – use of blocking groups, protection/ deprotection, and temporary modification of physical/chemical processes – should be minimized or avoided if possible, because such steps require additional reagents and can generate waste. 9. Catalysis: Catalytic reagents (as selective as possible) are superior to stoichiometric reagents. 10. Design for Degradation: Chemical products should be designed so that at the end of their function they break down into innocuous degradation products and do not persist in the environment. 11. Real-Time Analysis for Pollution Prevention: Analytical methodologies need to be further developed to allow for real-time, in-process monitoring and control prior to the formation of hazardous substances. 12. Inherently Safer Chemistry for Accident Prevention: Substances and the form of a substance used in a chemical process should be chosen to minimize the potential for chemical accidents, including releases, explosions, and fires. Green is a well-recognized symbolic color for environmental aspects which is also used by political parties. However, green is also the color of money, at least of the US dollar. Importantly, by applying the principles of green chemistry, economic and environmental benefits go hand in hand, which signals a clear go (do it!), just as the green in the traffic light. The opportunity of combined economic and ecologic benefits is certainly the reason why green chemistry (or synonyms) received such huge attention over the last two decades. And today, this is even more prominent on the agenda of companies, politicians and society due to concerns such as water pollution and climate change and actual cost pressure caused by increasing raw material and energy prices, as well as increasing waste disposal costs. How does green chemistry now relate to pharmaceuticals production? A special challenge for manufacturing of pharmaceuticals is the complexity of the molecules, which requires many synthesis steps. For example, an analysis of 128 drug candidates from three major pharma companies showed that, on average, eight steps are required for the synthesis of active

Green Chemistry with Biocatalysis for Production of Pharmaceuticals

307

pharmaceutical ingredients [6]. And the complexity of pharmaceutical molecules tends to increase further. The large number of synthesis steps also explains why the E factor (kg waste/ kg product) in this sector is much higher (25–100) than, for example, the bulk chemicals business (100

Lipases

3.1.1.3

esters, alcohols, amines

Nitrilases Aminoacylases Amidases Hydantoinases Haloalcohol dehalogenases Epoxide hydrolases Proteases and peptidases

3.5.5.1 3.5.1.14 3.5.1.4 3.5.2.2 3.8.1.X

nitriles N-acetyl amino acids amides 50 -hydantoins halohydrines, epoxides

3.3.2.3 3.4.X.X

epoxides peptides, carboxylic acids, esters, amides

alcohols, carboxylic acids, alcohols alcohols, carboxylic acids, alcohols, amines carboxylic acids amino acids amino acids, peptides a-amino acids diols, halohydrines, epoxides diols, epoxides peptides, carboxylic acids, esters, amides

>400 >50 >100 >200 >50 >50 >50 >1000

a

From BRENDA database [15].

among the most frequently used enzymatic transformations [14]. A list of different enzyme classes used in resolution processes is provided in Table 14.1. For more information on hydrolases, see also Chapter 5. An excellent example for an enzymatic resolution process is reported for production of Pregabalin. This drug was approved by the US Food and Drug Administration (FDA) in 2004 against neurophatic pain associated with diabetic peripheral neuropathy and postherpetic neuralgia. Several different routes have been developed based on asymmetric hydrogenation, crystallization and biocatalytic resolutions [16–20]. The most powerful and currently applied process is based on a lipase resolution, shown in Figure 14.3. This process is one of the very few CN

EtOOC

CN

Lipolase

HOOC

COOEt

CN

H2 O

COOEt

COOEt

+ CN

NaOEt

EtOOC

H2, Ni

COOEt NH2 COOH Pregabalin

Figure 14.3

Route to Pregabalin based on a lipase-based resolution

311


examples where every chemical step is performed in water. It has received the Astra Zeneca Award for Green Chemistry and Engineering. Impressive improvements compared with the first-generation route were reported, such as elimination of 5 million gallons of organic solvents annually, including tetrahydrofuran, methanol and ethanol. In addition, the overall annual reduction in used reagents and starting materials ranges from 893 to 1.116 t based on an annual production capacity of 400–500 t. Another good example for a successful process substitution has recently been reported for production of an intermediate of Aliskiren. This drug is a first in class renin inhibitor and was approved by the US FDA in 2007 for the treatment of hypertension. A key step in the synthesis is the resolution of methyl-5-chloro-2-isopropyl-4-pentenoate, which is shown Figure 14.4. For this process, a new nonanimal-derived isoform of pig liver esterase (PharmaPLE) was developed to address potential safety issues due to its animal origin and batch-to-batch variations often found in the commercially available extract of pig liver [21]. The enzymatic process could increase the productivity by >50%, and significantly reduced the waste production by avoiding the double resolution concept implemented in the first-generation chemical process which generated a large amount of organic and inorganic waste (DSM unpublished results). Of course, kinetic resolution processes are not optimal. In order to obtain >50% molar yields, racemization and recycling loops are required which often have a negative impact on solvent and/or energy consumption, as well as on waste production. Obviously, better O Cl

O

PharmaPLE TM OMe

Cl

OH

+ O Cl

OMe

OH H N

H2N O

O

O

O .1/2

O

NH2

COOH HOOC

Aliskiren Figure 14.4 Resolution of methyl-5-chloro-2-isopropyl-4-pentenoate with PharmaPLE, a nonanimalderived isoform of pig liver esterase. The corresponding S-ester is used as intermediate for production of Aliskiren

312


solutions, also from an atom economy point of view, are dynamic kinetic resolution processes. These allow for production of optically pure compounds at theoretical 100% yield of the racemic starting compound. Dynamic kinetic resolutions require mild racemization methods that can operate under the same conditions as enzymatic reactions. Such methods can be based on racemases; for example, amide racemases, N-acetyl-amino acid racemases, mandelic acid racemases or hydantoin racemases [22–25]. Also, the use of transition metals is well described and was recently reviewed [26,27]. Some of the few examples operating at the tons scale are the hydantoinase process for production of D-p-OH-phenylglycine and the nitrilase process for production of (R)-mandelic acid [28,29]. It is interesting to note that, in both examples, racemization of the educt is very fast under the reaction conditions even without any racemization catalyst. Many other targets for dynamic kinetic resolutions, such as secondary alcohols and amines, require racemization catalysts. A limited scope of the currently available racemization systems, poor compatibility with enzymatic reactions, and/or high prices of the racemization catalyst might explain the discrepancy between the large number of examples reported in the literature and the low number of dynamic kinetic resolution processes for production of pharmaceuticals operating at the industrial scale. However, this discrepancy is expected to decline with the development of new and improved enzymatic and transition metalbased racemization methods which will further improve the efficiency of enzymatic resolution processes.

14.3 Bioreductions: Greener Ligands, Renewable Hydride Donors, No Metals The quest for more economic and greener routes has triggered a lot of developments in reductive asymmetric synthesis tools based on prochiral molecules during the past two decades. Very powerful hydrogenation tools have been developed, for which William S. Knowles and Ryoji Nyori received the 2001 Noble Prize in Chemistry [30]. The developments of enzymatic reductions received less public attention. Indeed, low productivities and high E-factors due to huge biomass-to-product ratios characterized bioreductions in the past. However, this has changed dramatically during the last 5–10 years. Enzymatic reductions based on recombinantly produced enzymes and very cheap recombinant whole-cell biocatalysts are now widely accepted as competitive technologies and applied at the industrial scale for production of a number of pharma intermediates and other fine chemicals [31,32]. See also the examples in Chapter 7. An important breakthrough was achieved by the development of highly cost-efficient cofactor regeneration systems [33,34]. The concept with the highest atom economy is based on hydrogenases, which is recycling the oxidized cofactor using dihydrogen. However, this system is still in its infancy and not yet industrially applied, maybe because of limitations with respect to specific enzyme properties (activity/stability). Also, the reported total turnover numbers (moles of product per mole of NAD(P)H) are still lower than other regeneration systems [35]. A very clean cofactor regeneration system for which very large total turnover numbers have been reported is based on formate dehydrogenases and formic acid. The only by-product is CO2, which makes product isolation very simple. In 2002, Maria-Regina Kula and Martina Pohl were awarded with the very prestigious German Future Prize for the development of this system [36].

313


One of the first industrial applications using formate dehydrogenases is reported for tonscale production of L-tertiary-leucine [37]. This compound is used as a synthon for a variety of different drugs and drug candidates, such as antitumor and HIV protease inhibitors. Compared with other alternative enzymatic resolution processes – for example, based on acylases [38] or dynamic kinetic resolution using a lipase [39] – the reductive amination process is superior in terms of atom economy, yield, and productivity. Recently, this process was further improved by the development of recombinant whole-cell biocatalyst with a glucose dehydrogenase-based cofactor regeneration system. In this process, enzyme costs are lower and addition of external cofactor could be eliminated or was greatly reduced [40]. Indeed, high enantioselectivity, mild reaction conditions, saving on reducing agents, excellent atom economy, as well as there being no need for complex ligands and metals are important characteristics of enzymatic reduction reactions. It is not surprising, therefore, that enzymatic reduction processes have gained in importance over the last 5 years and have received a lot attention in the context of green chemistry. In 2006, the Presidential Green Chemistry Challenge Award from the United States Environmental Protection Agency (USEPA) was awarded to the development of a green process for a statin intermediate. As shown in Figure 14.5, this process is based on an enzymatic reduction using an alcohol dehydrogenase coupled with a cyanidation using a halohydrin dehalogenase [41]. Reported advantages are the substitution of hydrogen bromide with a renewable and biodegradable hydride donor (glucose) and less waste production, especially due to the cleaner enzymatic cyanidation step, which also eliminates the need for high-vacuum fractional distillation used in the classical chemical process (for the 2006 Greener Reaction Conditions Award of Codexis, see http://www.epa.gov/greenchemistry/pubs/pgcc/winners/grca06.html). From the very successful developments of the alcohol dehydrogenase technology for production of secondary alcohols and enzymatic reductive amination of keto-acids for production of amino acids, it is expected that we will also soon see applications for other enzymatic redox chemistries; for example, reduction of unsaturated carbonyl compounds with

O Cl (1)

O

OH

ADH Cl

OEt NAD(P)H

gluconolactone

NAD(P)+

O

HHDH

O

O

+ HCL OEt

OEt

(3)

(2)

HHDH

HCN

glucose GDH OH NC

O OEt

(4)

Figure 14.5 Scheme for production of ethyl (R)-4-cyano-3-hydroxybutyrate (4), a key intermediate of Atorvastatin. In a first step enzymatic reduction of ethyl 4-chloroacetoacetate to the corresponding chlorohydrin (2) is performed with an alcohol dehydrogenase (ADH) and a glucose dehydrogenase (GDH) for cofactor regeneration. In a second step, simultaneous epoxidation of the chlorohydrin (2) and cyanidation of the corresponding epoxide is catalyzed by a halohydrine dehalogenase (HHDH)

314

Biocatalysis for the Pharmaceutical Industry 1. Deracemization of alcohols

2. Deracemization of amines or amino acids NH2

OH R1

R2

Figure 14.6 acids

R1 R2

R2

non-selective reduction NH

O

OH R1

R1

ADH

Oxidase

NH2

R2 R1

R1

R2

R2 enantioselective amino acid or amine oxidase

Enzymatic deracemization concepts for production of chiral alcohols, amines and amino

enone or enoate reductases [42,43]. In addition, redox cycles for deracemization of amino acids, alcohols and amines based on the schemes shown in Figure 14.6 are on the brink of becoming very efficient concepts for production of chiral intermediates [44–46].

14.3.1 Enzymatic Oxidations: Clean, Highly Selective and Catalytic Oxidations are often carried out using stoichiometric amounts of oxidants, such as permanganate, manganese dioxide or chromium reagents. Catalytic alternatives, therefore, hold great promise to substitute such processes. Examples of oxidation reactions catalyzed by enzymes are shown in Figure 14.7. Some of these are applied on the tons scale with proven cost and ecologic benefits; for example, in the production of vitamin C (alcohol oxidation), 5-methylpyrazine-2-carboxylic acid (hydroxylation and oxidation), and cephalosporin C (amine oxidation) [47–49]. For more general information about alcohol oxidations, see Chapter 7. Furthermore, highly selective enzymatic hydroxylations are applied in the production of steroids [50]. One pioneering process is the production of cortisone based on the microbial conversion of progesterone to 11-a-OH-progesterone as shown in Figure 14.8. Based on this direct hydroxylation step and another starting compound, the first-generation 30-step chemical process was replaced by a much greener 15-step chemo-microbial process resulting in much higher yields, less waste and a cost reduction of hydrocortisone by an impressive factor of more than 50 [51]! From the above-described examples, one might conclude that enzymatic oxidation is a mature and well-developed technology. This, however, is not really the case. There are still many challenges to be solved before enzymatic oxidation becomes a true platform technology. A large set of novel oxidases and oxygenases are now available [52–56]. However, robust enzymes with satisfying performance parameters, such as broad product scope, activity levels of >50 U mg1, half-life under process conditions of more than 24 h, as well as high regio- and enantio-selectivities, are still scarce. Also, many of the known monooxygenases suffer from strong product inhibitions, which results in poor volumetric productivities and low final product concentrations. Solutions such as in situ product removal (ISPR) concepts have been suggested and been successfully developed [57,58]. However, with perhaps the exception of two-phase systems, only a few of these approaches are applied at a

315

Green Chemistry with Biocatalysis for Production of Pharmaceuticals XH R1

R1

O2

R2

S

X=O, N R1

R2

R1

O2

R2

R1

R1

R2

OH R2

R1 OH

monoxygenase O2

R2

OH

dioxygenase O2

R2

O

R1

O2

R2

R2

R1

O2

monoxygenase R1

S

O

monoxygenase R2

R2 O

monoxygenase

O R1

X

oxidases

O R1

R2

Figure 14.7 Examples of enzymatic oxidation reactions

large scale. This indicates the need for further improvements to make ISPR an economically viable option or to address product inhibitions by means of enzyme engineering methods [59]. Another challenge is related to production of the enzymes, especially for membrane-bound oxygenases such as many mammalian- and plant-derived P450s. Also, some interesting oxidases require cofactors that are not formed in standard enzyme production organisms. Such demanding enzyme systems, therefore, are currently applied O

O HO Rhizopus sp.

O

O Progesterone

11-α-OH-Progesterone

Figure 14.8 Microbial hydroxylation of progesterone to 11-a-OH-progesterone by Rhizopus sp.

316


as whole-cell preparations [48] with drawbacks related to biocatalyst (production) cost, high biocatalyst-to-product ratios, and unwanted side reactions. It is expected, therefore, that solutions based on enzyme engineering (see Chapter 3), new production organisms (see Chapter 2) and reaction engineering (see Chapter 4) will have a very strong impact, especially on greener oxidations based on enzymes.

14.4 C–C Bond Formations: Atom Efficiency at Its Best Creation of complex molecules from simple and cheap precursors in an enantioselective and highly atom efficient way is the essence of modern chemistry. Given the high efficiency of the approach, it is not surprising that catalytic CC bond formation is increasingly applied in the fine chemicals industry [60]. Interestingly, one of the earliest reports of asymmetric CC coupling reactions for production of ephedrine employs a biocatalyst and dates back to the 1920s [61,62]. The process described is based on the activity of a pyruvate kinase from Saccharomyces cerevisiae catalyzing the stereoselective condensation of benzaldehyde and fermentatively produced acetaldehyde to yield (R)-1-hydroxy-1-phenylpropanone according to the scheme in Figure 14.9. More than 80 years later, this process is still applied [29], which clearly demonstrates how efficient enzymatic CC coupling reactions can be. Despite the fact that HCN is not the ‘greenest’ compound, asymmetric HCN addition to aldehydes catalyzed by hydroxynilrile lyases (EC 4.2.1.x) is the method of choice for production of a variety of optically pure cyanohydrins [63]. Excellent isolated yields (>90%), high productivities (>250 g L day1) and very high stereoselectivities (>99% enatiomeric excess (ee)) makes hydroxynitrile lyase-catalyzed HCN additions greener than many other alternative routes [64]. The impressive process economics were enabled by breakthroughs in recombinant production of acid-stable (R)- and (S)-selective enzymes, design of efficient catalysts with improved activities and enantioselectivities, as well as significant progress in reaction engineering [65,66]. Figure 14.10 shows conversions which are operated at the multi-tons scale for production of (R)-2-mandelonitrile and (R)-2-hydroxy4-phenylbutyronitrile. These compounds are used as intermediates for different launched cardiovascular drugs. In contrast to the above two examples, for which applications were developed long before the responsible biocatalyst was discovered, aldolase applications are more recently developed. Indeed, aldolases and their natural function were extensively studied between the end of the 1960s and the beginning of the 1970s. The first patents about their applications in organic synthesis appear in the 1990s [67–69] and the first ton-scale applications were reported in 1997

O

Saccharomyces cerevisiae H

OH

OH CH3 catalyst, H2

(producing acetaldehyde and pyruvate decarboxylase)

O

Benzaldehyde

CH3NH2

(R)-1-hydroxy-1-phenylpropanone

CH3 HN

CH3

(-)-ephedrine

Figure 14.9 Reaction scheme for production of ()-ephedrine based on enzymatic CC bond formation using Saccharomyces cerevisiae coupled with a chemical reductive amination

317

Green Chemistry with Biocatalysis for Production of Pharmaceuticals OH O

R-HNL

+ HCN

> 98% conversion > 99%ee ca. 250g/lh

CN Cl

Cl

OH O

R-HNL

+ HCN

CN

> 98% conversion > 99%ee ca. 200g/lh

Figure 14.10 Hydroxynitrile lyase-catalyzed reactions for production of pharma intermediates used in the synthesis of cardiovascular drugs

for production of neuraminic acid based on the route shown in Figure 14.11 [70]. The latter compound is used as intermediate for Zanamavir, an antiviral drug. Another model example of a green process based on CC coupling of very cheap and simple precursor molecules is the production of statin intermediates according to the scheme in Figure 14.12. This process was pioneered by Wong and coworkers, who reported for the first time in 1994 an asymmetric tandem aldol reaction using 2-deoxyribose-5-phosphate aldolase (DERA) and its application for production of a statin intermediate some years later [71,72]. Cheap starting compounds (acetaldehyde and chloroacetaldehyde), high isolated yields and productivities (>30 g L1 h1), high optical purities (>99.5% ee, >99.5% diastereomeric excess), use of only one single enzyme and no need for hydride donors or chiral auxiliaries are the characteristics of this process which make it economically very competitive [73]. It is HO HO HOH2C HO HO

COOH

OH O

Zanamivir

AcHN

O

HN

NHAc OH N-Acetyl-D-glucosamine

NH2

HN

Epimerase or OH HO HOH2C HO HO

NHAc O

Aldolase HO

O +

OH N-Acetyl-D-mannosamine

H3C

COOH

Pyruvate

COOH

OH O

OH

AcHN HO N-Acetyl-D-neuraminic acid

Figure 14.11 Reaction scheme of aldolase-based process for production of neuraminic acid, an intermediate of Zanamivir

318


OH

OH

O

O N

OH

NH

F Atorvastatin R

O

R

O

OH

+ R= Cl, CN

O

OH

DERA

OH R

OH

DERA O

+

O

R

OH O

Figure 14.12 Asymmetric tandem aldol reaction using 2-deoxyribose-5-phosphate aldolase (DERA) and its application for production of Atorvastatin

perhaps one of the greenest routes for production of statin intermediates, implemented at an industrial scale [74]. An important breakthrough to achieve economically viable productivities and lower enzyme costs was the development of improved enzyme variants obtained by directed evolution [75].

14.5 Summary and Outlook It is tempting to repeat predictions about the growing impact of biocatalysis in the chemicals industry made by others. But we do not need predictions. We can just look at the facts on how biocatalysis can impact our manufacturing processes. This chapter aimed at providing a snapshot about actual enzymatic processes that are currently used in production of pharmaceuticals. These facts demonstrate that green chemistry with biocatalysis is a reality for production of pharmaceuticals. For sure, using enzymes is by no means always greener. However, neglecting this option due to lack of experience or simply by ignorance should raise a red flag for those concerned with green chemistry. And this should be all who manufacture pharmaceuticals. When talking about green chemistry we are talking about responsibility. And the good news is that by applying the green chemistry principles described at the beginning of this chapter we can combine the responsibility for people, our planet and the profit we want to make. Therefore, the future of chemistry is ‘green chemistry’, and with further progress in biocatalysis the future of pharmaceutical production will be even ‘greener’.


319

References [1] RNCOS (2007) EU Chemical Industry Analysis Report, RNCOS Market Research Solutions. [2] Brundtland, C.G. (1987) Our Common Future, The World Commission on Environmental Development, Oxford University Press. [3] Anastas, P. and Williamson, T.C. (1996) Green chemistry: an overview, in Green Chemistry. Designing Chemistry for the Environment (eds P. Anastas and T.C. Williamson) ACS Symposium Series, vol. 626, 1–17. [4] Sheldon, R.A. (2007) The E factor: fifteen years on. Green Chemistry, 9, 1273–1283. [5] Anastas, P. and Warner, J.C. (1998) Green Chemistry: Theory and Practice, Oxford University Press. [6] Carey, J.S., Laffan, D., Thomson, C. and Williams, M.T. (2006) Analysis of the reactions used for the preparation of drug candidate molecules. Organic and Biomolecular Chemistry, 4 (12), 2337–2347. [7] Sheldon, R.A. (1994) Consider the environmental quotient. Chemotherapy, 24 (3), 38–47. [8] Constable, D.J.C., Dunn, P.J., Hayler, J.D. et al. (2007) Key green chemistry research areas – a perspective from pharmaceutical manufacturers. Green Chemistry, 9 (5), 411–420. [9] Griffiths, M. (2001) The Application of Biotechnology to Industrial Sustainability, OECD Publications, ISBN 92-64-19546-7. [10] Schoemaker, H.E., Mink, D. and Wubbolts, M.G. (2003) Dispelling the myths – biocatalysis in industrial synthesis. Science, 299 (5613), 1694–1698. [11] Lorenz, P. and Eck, J. (2004) Screening for novel industrial biocatalysts. Engineering in Life Sciences, 4 (6), 501–504. [12] Robertson, D.E. and Steer, B.A. (2004) Recent progress in biocatalyst discovery and optimization. Current Opinion in Chemical Biology, 8 (2), 141–149. [13] Bornscheuer, U.T. and Kazlauskas, R.J. (2006) Hydrolases in Organic Synthesis, Wiley-VCH. [14] Straathof, A.J. (2006) Quantitative analysis of industrial biotransformations, in Industrial Biotransformations, 2nd edn (eds A. Liese, K. Seelbach and C. Wandrey), Wiley-VCH Verlag GmbH, pp. 515–520. [15] Barthelmes, J., Ebeling, C., Chang, A. et al. (2007) BRENDA, AMENDA and FRENDA: the enzyme information system in 2007. Nucleic Acids Research, 35, 511–514. [16] Burk, M.J., De Koning, P.D., Grote, T.M. et al. (2003) An enantioselective synthesis of (S)-( þ )-3aminomethyl-5-methylhexanoic acid via asymmetric hydrogenation. The Journal of Organic Chemistry, 68 (14), 5731–5734. [17] Huckabee, B.K. and Sobieray, D.M. (1996) Stereoselective synthesis of (S)-3-(aminomethyl)-5-methylhexanoic acid. WO 9638405. [18] Hoekstra, M.S., Sobieray, D.M., Schwindt, M.A. et al. (1997) Chemical development of CI-1008, an enantiomerically pure anticonvulsant. Organic Process Research & Development, 1 (1), 26–38. [19] Xie, Z., Feng, J., Garcia, E. et al. (2006) Cloning and optimization of a nitrilase for the synthesis of (3S)-3-cyano5-methyl hexanoic acid. Journal of Molecular Catalysis B–Enzymatic, 41 (3–4), 75–80. [20] Hu, S., Martinez, C.A., Tao, J. et al. (2005) Preparation of pregabalin and related compounds. US2005283023. [21] Hermann, M., Kietzmann, M.U., Ivancic, M. et al. (2008) Alternative pig liver esterase (APLE) – cloning, identification and functional expression in Pichia pastoris of a versatile new biocatalyst. Journal of Biotechnology, 133 (3), 301–310. [22] Asano, Y. and Yamaguchi, S. (2005) Dynamic kinetic resolution of amino acid amide catalyzed by Daminopeptidase and a-amino-e-caprolactam racemase. Journal of the American Chemical Society, 127 (21), 7696–7697. [23] Boesten, W.H.J., Raemakers-Franken, P.C., Sonke, T. et al. (2003) Protein and cDNA sequences of a-H-a-amino acid amide racemases cloned from Ochrobactrum anthropi and Arthrobacter nicotianae, WO2003106691. [24] May, O., Verseck, S., Bommarius, A. and Drauz, K. (2002) Development of dynamic kinetic resolution processes for biocatalytic production of natural and nonnatural L-amino acids. Organic Process Research & Development, 6 (4), 452–457. [25] Schnell, B., Faber, K. and Kroutil, W. (2003) Enzymatic racemization and its application to synthetic biotransformations. Advanced Synthesis and Catalysis, 345 (6–7), 653–666. [26] Martin-Matute, B. and Baeckvall, J.-E. (2007) Dynamic kinetic resolution catalyzed by enzymes and metals. Current Opinion in Chemical Biology, 11 (2), 226–232. [27] Pellissier, H. (2008) Recent developments in dynamic kinetic resolution. Tetrahedron, 64 (8), 1563–1601. [28] Pietzsch, M. and Syldatk, C. (2002) Hydrolysis and formation of hydantoins, in Enzyme Catalysis in Organic Synthesis, 2nd edn (eds K. Drauz and H. Waldmann), Wiley-VCH Verlag GmbH, pp. 761–799.

320


[29] Breuer, M., Ditrich, K., Habicher, T. et al. (2004) Industrial methods for the production of optically active intermediates. Angewandte Chemie–International Edition, 43 (7), 788–824. [30] Knowles, W.S., Noyori, R. and Sharpless, K.B. (2002) Noble lectures. Angewandte Chemie–International Edition, 41, 1998–2022. [31] Moore, J.C., Pollard, D.J., Kosjek, B. and Devine, P.N. (2007) Advances in the enzymatic reduction of ketones. Accounts of Chemical Research, 40 (12), 1412–1419. [32] De Wildeman, S.M.A., Sonke, T., Schoemaker, H.E. and May, O. (2007) Biocatalytic reductions: from lab curiosity to ‘first choice’. Accounts of Chemical Research, 40 (12), 1260–1266. [33] Eckstein, M., Daussmann, T. and Kragl, U. (2004) Recent Developments in NAD(P)H regeneration for enzymatic reductions in one- and two-phase systems. Biocatalysis and Biotransformation, 22 (2), 89–96. [34] Van der Donk, W.A. and Zhao, H. (2003) Recent developments in pyridine nucleotide regeneration. Current Opinion in Biotechnology, 14 (4), 421–426. [35] Wichmann, R. and Vasic-Racki, D. (2005) Cofactor regeneration at the lab scale. Advances in Biochemical Engineering/Biotechnology, 92, 225–260. [36] Kula, M.R. and Pohl, M. (2002) Deutscher Zukunftspreis. http://www.fz-juelich.de/ibt/aktuell/deutscherzukunftspreis-2002. [37] Bommarius, A.S., Drauz, K., Hummel, W. et al. (1994) Some new developments in reductive amination with cofactor regeneration. Biocatalysis, 10 (1–4), 37–47. [38] Liu, S.-L., Song, Q.-X., Wei, D.-Z. et al. (2006) Preparation of optically pure tert-leucine by penicillin G acylasecatalyzed resolution. Preparative Biochemistry & Biotechnology, 36 (3), 235–241. [39] Turner, N.J., Winterman, J.R., McCague, R. et al. (1995) Synthesis of homochiral L-(S)-tert-leucine via a lipase catalyzed dynamic resolution process. Tetrahedron Letters, 36 (7), 1113–1116. [40] May, O. and Groeger, H. (2005) Designer cells produce enantiomerically pure compounds. PharmaChem, 4 (9), 6–8. [41] Davis, S.C., Grate, J.H., Gray, D.R. et al. (2004) Enzymatic processes for the production of 4-substituted 3-hydroxybutyric acid derivatives. WO2004015132. [42] Tang, Y.X. and Suga, T. (1992) Characterization of two enone reductases from Nicotiana tabacum cell cultures. Phytochemical, 31 (8), 2599–260. [43] Wanner, P. and Tressl, R. (1998) Purification and characterization of two enone reductases from Saccharomyces cerevisiae. European Journal of Biochemistry, 255 (1), 271–278. [44] Fotheringham, I., Archer, I., Carr, R. et al. (2006) Preparative deracemization of unnatural amino acids. Biochemical Society Transactions, 34 (2), 287–290. [45] Voss, C.V., Gruber, C.C. and Kroutil, W. (2008) Deracemization of secondary alcohols through a concurrent tandem biocatalytic oxidation and reduction. Angewandte Chemie–International Edition, 47 (4), 741–745. [46] Turner, N.J. and Carr, R. (2007) Biocatalytic routes to nonracemic chiral amines, in Biocatalysis in the Pharmaceutical and Biotechnology Industries (ed. R.N. Patel), CRC Press LLC, pp. 743–755. [47] Hancock, R.D. and Viola, R. (2001) The use of micro-organisms for L-ascorbic acid production: current status and future perspectives. Applied Microbiology and Biotechnology, 56 (5–6), 567–576. [48] Kiener, A. (1992) Enzymic oxidation of methyl groups in heteroarenes: a versatile method for the preparation of heteroaromatic carborylic acids. Angewandte Chemie–International Edition, 31 (6), 774–775. [49] Bayer, T. (2004) 7-Aminocephalosporanic acid – chemical versus enzymatic production process, in Asymmetric Catalysis on Industrial Scale (2004), (eds H.-U. Blaser and E. Schmidt), Wiley-VCH Verlag GmbH, 117–130. [50] Pellissier, H. and Santelli, M. (2001) Chemical and biochemical hydroxylations of steroids. A review. Organic Preparations and Procedures International, 33 (1), 1–58. [51] Kardinahl, S., Rabelt, D. and Reschke, M. (2006) Biotransformation: from vision to technology. Chemie Ingenieur Technik, 78 (3), 209–217. [52] Van Berkel, W.J.H., Kamerbeek, N.M. and Fraaije, M.W. (2006) Flavoprotein monooxygenases, a diverse class of oxidative biocatalysts. Journal of Biotechnology, 124 (4), 670–689. [53] Mihovilovic, M.D. (2006) Enzyme mediated Baeyer–Villiger oxidations. Current Organic Chemistry, 10 (11), 1265–1287. [54] Urlacher, V.B. and Eiben, S. (2006) Cytochrome P450 monooxygenases: perspectives for synthetic application. Trends in Biotechnology, 24 (7), 324–330. [55] Van Hellemond, E.W., Leferink, N.G.H., Heuts, D.P.H.M. et al. (2006) Occurrence and biocatalytic potential of carbohydrate oxidases. Advances in Applied Microbiology, 60, 17–54.


321

[56] Riva, S. (2006) Laccases: blue enzymes for green chemistry. Trends in Biotechnology, 24 (5), 219–226. [57] Alphand, V., Carrea, G., Wohlgemuth, R. et al. (2003) Towards large-scale synthetic applications of Baeyer– Villiger monooxygenases. Trends in Biotechnology, 21 (7), 318–323. [58] Prichanont, S., Leak, D.J. and Stuckey, D.C. (1998) Alkene monooxygenase-catalyzed whole cell epoxidation in a two-liquid phase system. Enzyme and Microbial Technology, 22 (6), 471–479. [59] Woodley, J.M., Bisschops, M., Straathof, A.J.J. and Ottens, M. (2008) Future directions for in-situ product removal (ISPR). Journal of Chemical Technology and Biotechnology (Oxford, Oxfordshire: 1986), 83 (2), 121–123. [60] Blaser, H.-U., Indolese, A., Naud, F. et al. (2004) Industrial R&D on catalytic CC and CN coupling reactions: a personal account on goals, approaches and results. Advanced Synthesis and Catalysis, 346 (13–15), 1583–1598. [61] Neuberg, C. and Hirsch, J. (1921) An enzyme which brings about union into carbon chains (Carboligase). Biochemische Zeitschrift, 115, 282–310. [62] Neuberg, C. and Ohle, H. (1922) Carboligase. IV. Biosynthetic carbon chain union in fermentation processes. Biochemische Zeitschrift, 128, 610–618. [63] Poechlauer, P., Skranc, W. and Wubbolts, M. (2004) The large-scale biocatalytic synthesis of enantiopure cyanohydrins, in Asymmetric Catalysis on Industrial Scale (eds H.-U. Blaser and E. Schmidt), Wiley-VCH Verlag GmbH, pp. 151–164. [64] Purkarthofer, T., Skranc, W., Schuster, C. and Griengl, H. (2007) Potential and capabilities of hydroxynitrile lyases as biocatalysts in the chemical industry. Applied Microbiology and Biotechnology, 76 (2), 309–320. [65] Glieder, A., Weis, R., Skranc, W. et al. (2003) Comprehensive step-by-step engineering of an ()-hydroxynitrile lyase for large-scale asymmetric synthesis. Angewandte Chemie–International Edition, 42 (39), 4815–4818. [66] Weis, R., Gaisberger, R., Skranc, W. et al. (2005) Carving the active site of almond R-HNL for increased enantioselectivity. Angewandte Chemie–International Edition, 44 (30), 4700–4704. [67] Brockamp, H.P., Kula, M.R. and Goetz, F. (1991) A robust microbial fructose-1,6-diphosphate aldolase; its manufacture and use in sugar synthesis. DE3940431. [68] Fessner, W.D. (1992) Enzymic manufacture of enantio- and diastereomerically pure ketoses and their 1orthophosphate esters. DE4111971. [69] Wong, C.-H. (1992) Synthesis of deoxy-L-glycero- L-galactonoulosonic acids with R configuration. US5162513. [70] Mahmoudian, M., Noble, D., Drake, C.S. et al. (1997) An efficient process for production of N-acetylneuraminic acid using N-acetylneuraminic acid aldolase. Enzyme and Microbial Technology, 20 (5), 393–400. [71] Gijsen, H.J.M. and Wong, C.-H. (1994) Unprecedented asymmetric aldol reactions with three aldehyde substrates catalyzed by 2-deoxyribose-5-phosphate aldolase. Journal of the American Chemical Society, 116 (18), 8422–8423. [72] Machajewski, T.D., Wong, C.-H. and Lerner, R.A. (2000) The catalytic asymmetric aldol reaction. Angewandte Chemie–International Edition, 39 (8), 1352–1374. [73] Greenberg, W.A., Varvak, A., Hanson, S.R. et al. (2004) Development of an efficient, scalable, aldolase-catalyzed process for enantioselective synthesis of statin intermediates. Proceedings of the National Academy of Sciences of the United States of America, 101 (16), 5788–5793. [74] Thayer, A.M. (2006) Competitors want to get a piece of lipitor. Chemical & Engineering News, 84 (33), 26–27. [75] Jennewein, S., Schuermann, M., Wolberg, M. et al. (2006) Directed evolution of an industrial biocatalyst: 2-deoxy-D-ribose 5-phosphate aldolase. Biotechnology Journal, 1 (5), 537–548.

Index absorbing resins, 220, 222 acrylamide, 9, 79, 153–155, 169, 221 alcohol dehydrogenase, 92, 127, 128, 132, 135, 137–140, 142, 252, 313 alcohol oxidase, 6, 31, 94, 121, 142–144, 146 aldolase, 2, 12, 58, 111–114, 116–118, 214, 316, 317 amides, 9, 153, 155, 157, 159–163, 165, 172 aqueous-organic two phase system, 222 asymmetric synthesis, 80, 116, 124, 214, 312 bacterial hosts, 23, 25 biocatalysis, 1, 28, 60, 305, 307, 318 biopharmaceutical, 34, 36, 66 bioreactor, 3, 33, 35, 65–67, 184, 190, 192, 193, 196, 200, 202, 265 biosynthesis, 14, 56, 183, 184, 194, 195, 202, 207, 229, 230, 232, 234, 236, 238–241, 250–259, 273, 274, 276–278, 281–283, 288–297, 299 biotransformation(s), 1, 65, 67, 183, 185, 190, 192–194, 196, 199–201, 206, 213–225 carbohydrates, 12, 14, 254, 278, 295 carbon-carbon bond formation, 111, 316 carboxylic acids, 6, 9, 100, 155, 160, 162, 172, 173, 176, 178, 278 cell-free protein expression, 36, 37 cell surface display, 54 chiral alcohol, 4, 121, 125, 126, 129, 132, 141, 214, 314 chiral intermediates, 71, 81, 113, 176, 214, 314 chiral synthesis, 10, 71, 89 cofactor regeneration, 67–69, 84, 124, 205, 222, 223, 295, 312 combinatorial biosynthesis, 14, 229, 230, 232, 234, 238, 240, 241, 273, 278

cyanogensis, 89 cyanohydrin, 10, 12, 89, 92, 94–100, 103–105, 158, 316 cytochrome p450 (CYP), 184–189, 192, 193, 195, 202, 205, 207, 218, 234, 261, 265, 294 daptomycin, 236–238, 240, 278–280 directed evolution, 1, 2, 46, 52, 56–58, 60, 111–114, 153, 176, 206, 283 DNA shuffling, 49–51, 112 domains, 37, 52, 56, 274–278, 280, 281, 283–286, 288, 291 downstream processing, 26, 28, 65, 66, 77, 78, 82–85, 122, 129 drug development, 193–195, 247, 254, 273 drug metabolite, 32, 34, 183, 184, 196, 202, 207, 213 drug production, 15 E-factor, 305, 307, 312 enantioselectivity, 46, 56, 59, 79, 94, 95, 97, 98, 126–130, 134, 137, 138, 140, 141, 143, 161–165, 170, 175, 176, 178, 313 enediyne, 231–233 enzymatic biotransformation, 65–67 enzymatic oxidation, 314, 315 enzymatic reduction, 137–139, 312, 313 enzyme discovery, 2, 22, 27, 31, 32, 35 enzyme engineering, 31, 265, 309, 315, 316 enzyme thermostability, 58 erythromycin, 230,234–236, 240, 252-254, 256, 274, 287, 295 eukaryotic hosts, 23, 29, 30 genetic engineering, 66, 78, 223, 225, 234, 235, 240, 241 genetic manipulation, 255, 265, 273, 278, 280


324 genome mining green chemistry, 305–307, 309, 311, 313, 318 heterologous expression, 22, 25, 26, 30, 94, 230, 294 hevea brasiliensis, 89, 97 high-throughput screening (HTS), 2, 10, 15, 22, 29, 31, 32, 34, 45–47, 93–95, 105, 229 hydrogen cyanide (HCN), 89, 92, 95, 96, 98, 104, 168, 172, 316 hydrolases, 1, 2, 10, 33, 53, 285, 309, 310 hydroxynitrile lyase, 10, 89, 101, 158, 316 iminocyclitols, 116 immobilization, 97, 159, 163, 213, 221 immobilized cell, 78, 156, 169, 215, 221 in situ product removal, 72, 73, 314 in vitro compartmentalization, 55 in vitro incubation method, 52–56 industrial processes, 21, 68, 105, 305–307, 311 isomerases, 2 isoprenoids, 258–261, 264, 265 ketone, 4–6, 10, 68, 74, 97, 104, 121, 122, 125, 126, 128, 132–142, 144, 223, 224, 274 ketoreductase, 4, 121–123, 125, 127, 129, 141, 146, 235 library creation, 46–49, 52, 56 library screening, 52 ligases, 2 lyases, 2, 10, 89, 316 metabolic engineering, 247–262, 264, 265, 273, 277, 278, 280, 281, 299 metabolic network, 248, 249, 265 metabolite synthesis, 184–186, 190, 192, 196, 197, 200, 205, 206 metagenomic, 2, 27, 93, 176 micro-aqueous, 96–98 microbial biotransformation, 192, 199–201 multimodular synthase, 273, 274, 295

Index

overexpression, 21, 26, 53, 105, 125, 213, 223, 234, 236, 249, 251, 252, 255 oxidation, 314–316 oxidoreductases, 1, 2, 57, 91, 223 oxynitrilase, 10, 89, 97, 172 patellamide, 232, 239, 240 pharmaceutical natural products, 229, 240 pharmaceuticals, 21, 59, 65, 68, 71, 77, 80, 83, 140, 247, 248, 253, 257, 259, 262, 265, 273, 278, 305–309, 312, 318 pharmaceuticals and pharmaceutical intermediates, 215, 217, 225 polyketide(s), 14, 59, 202, 230, 232, 234–236, 248, 252–254, 256, 257, 265, 273 posttranslational modifications, 21–25, 29, 33, 34, 36 product specificity, 59 prunus amygdalus, 89, 91, 92, 94, 97 racemic resolution, 67, 84 random mutagenesis, 46–49, 58, 178, 247, 248, 283 reaction engineering, 65, 67, 84, 85, 114, 316 recombinant enzyme, 15, 35, 104, 127, 132, 139, 184, 195, 202, 207 reductases, 129, 130–132, 135, 219, 314 reduction, 58, 66-68, 83, 100, 121–142, 159, 167, 219, 220, 254, 260, 311–314 regioselectivity, 4, 126, 136, 164, 168, 178, 202, 213, 214, 293 resolution, 67, 69–72, 82, 84, 125, 130, 136, 140, 173, 177, 213, 214, 309–313 ribosomal peptide, 239 secondary metabolites, 28, 230, 231, 265 selective oxidation, 142, 146 strain improvement, 25, 248, 250, 251, 254, 261 substrate engineering, 115 substrate inhibition, 67, 79, 84 sustainability, 307 transferases, 2, 14, 25, 33, 77, 117

nicotinamide, 9, 57, 67, 155–157, 205, 206, 214, 215, 218, 219 nitrilase, 2, 9–11, 59, 135, 153, 159–161, 163–178, 310, 312 nitrile hydratase, 2, 9, 25, 79, 153, 155, 221 nonribosomal peptide, 56, 57, 236, 238, 273, 274

uridine diphosphoglucuronosyl transferase (UGT), 184–189 white biotechnology, 83 whole cell biotransformation, 114, 213–215, 217–225

Biocatalysis for the Pharmaceutical Industry: Discovery, Development, and Manufacturing

Active Pharmaceutical Ingredients Development Manufacturing and Regulation

Systems Biology in Drug Discovery and Development (Wiley Series on Technologies for the Pharmaceutical Industry)

Biocatalysis in the Pharmaceutical and Biotechnology Industries

Biocatalysis in the Pharmaceutical and Biotechnology Industries

Pharmaceutical Suspensions: From Formulation Development to Manufacturing

Ethics and the Pharmaceutical Industry

Ethics and the Pharmaceutical Industry

Research and Development in the Chemical and Pharmaceutical Industry

Research and Development Management in the Chemical and Pharmaceutical Industry

Research and Development in the Chemical and Pharmaceutical Industry

Biocatalysis - From Discovery to Application

Biocatalysis: From Discovery to Application

Validated Cleaning Technologies for Pharmaceutical Manufacturing

Containment in the Pharmaceutical Industry (Drugs and the Pharmaceutical Sciences)

The Management of Chemical Process Development in the Pharmaceutical Industry

Pharmaceutical Data Mining: Approaches and Applications for Drug Discovery (Wiley Series on Technologies for the Pharmaceutical Industry)

The Management of Chemical Process Development in the Pharmaceutical Industry

Polymorphism in the pharmaceutical industry

Careers with the Pharmaceutical Industry

The Management of Chemical Process Development in the Pharmaceutical Industry

Integration of Pharmaceutical Discovery and Development: Case Histories

Antibiotic Discovery and Development

Pharmaceutical Manufacturing Handbook Production and Processes

Good Pharmaceutical Manufacturing Practice: Rationale and Compliance

Occupational Safety and Health Simplified for the Food Manufacturing Industry

Biocatalysis for Green Chemistry and Chemical Process Development

Pharmaceutical Manufacturing Handbook: Regulations and Quality

Modern Pharmaceutical Industry: A Primer

Pharmaceutical Biotechnology: Drug Discovery and Clinical Applications

Pharmaceutical Biotechnology: Drug Discovery and Clinical Applications

Biocatalysis for the Pharmaceutical Industry: Discovery, Development, and Manufacturing

Active Pharmaceutical Ingredients Development Manufacturing and Regulation

Systems Biology in Drug Discovery and Development (Wiley Series on Technologies for the Pharmaceutical Industry)

Biocatalysis in the Pharmaceutical and Biotechnology Industries

Biocatalysis in the Pharmaceutical and Biotechnology Industries

Pharmaceutical Suspensions: From Formulation Development to Manufacturing

Ethics and the Pharmaceutical Industry

Ethics and the Pharmaceutical Industry

Research and Development in the Chemical and Pharmaceutical Industry

Research and Development Management in the Chemical and Pharmaceutical Industry

Research and Development in the Chemical and Pharmaceutical Industry

Biocatalysis - From Discovery to Application

Biocatalysis: From Discovery to Application

Validated Cleaning Technologies for Pharmaceutical Manufacturing

Containment in the Pharmaceutical Industry (Drugs and the Pharmaceutical Sciences)

The Management of Chemical Process Development in the Pharmaceutical Industry

Pharmaceutical Data Mining: Approaches and Applications for Drug Discovery (Wiley Series on Technologies for the Pharmaceutical Industry)

The Management of Chemical Process Development in the Pharmaceutical Industry

Polymorphism in the pharmaceutical industry

Careers with the Pharmaceutical Industry

The Management of Chemical Process Development in the Pharmaceutical Industry

Integration of Pharmaceutical Discovery and Development: Case Histories

Antibiotic Discovery and Development

Pharmaceutical Manufacturing Handbook Production and Processes

Good Pharmaceutical Manufacturing Practice: Rationale and Compliance

Occupational Safety and Health Simplified for the Food Manufacturing Industry

Biocatalysis for Green Chemistry and Chemical Process Development

Pharmaceutical Manufacturing Handbook: Regulations and Quality

Modern Pharmaceutical Industry: A Primer

Pharmaceutical Biotechnology: Drug Discovery and Clinical Applications

Pharmaceutical Biotechnology: Drug Discovery and Clinical Applications

Recommend Documents